# Place across an American life: report

Introduction of ZIP Code Destiny, by Vishal Singh (NYU Stern). Generated from the chapter's data files; the reading version is at https://vishalsingh.org/zipcode/place-across-an-american-life. Table n of this report is figure n of the chapter.

## Identification

- Title: Place across an American life
- Label: The whole span
- Chapter: Introduction
- Question: How do circumstances and outcomes differ across American places from birth to old age, and how does the answer change when the comparison becomes more specific?
- Dek: Birth outcomes, schooling, income, housing, health, length of life and well-being differ across American places, many in step with neighborhood deprivation, while some follow other patterns or change when the comparison moves inside a county.
- Grain: neighborhoods
- Status: published
- Published: 2026-10-11
- Updated: not revised
- Data sources:
  - CDC WONDER natality. Low birthweight by county deprivation tenth, 2018 to 2024, as reported in Birthweight and preterm birth; years: 2018-2024; weights: births
  - County Health Rankings. Low birthweight by mother's race and county deprivation quartile, births in 2016 to 2022, as reported in Birthweight and preterm birth; years: 2024 release; weights: births
  - USALEEP life expectancy. Life expectancy at birth by tract deprivation tenth, deaths in 2010 to 2015, as reported in Place and the length of a life; years: 2010-2015; weights: ACS 2019-2023 tract population
  - HUD CHAS housing affordability. Housing cost burden of renter households by tract deprivation tenth, 2018 to 2022, as reported in Renting, affordability and eviction; years: 2018-2022; weights: renter households with a computed burden
  - Area Deprivation Index. Neighborhood Atlas Area Deprivation Index 2023, the ranking behind every comparison above; years: 2023 release (version 4.0.1; built on ACS 2019-2023); weights: as in each chapter
- Canonical URL: https://vishalsingh.org/zipcode/place-across-an-american-life
- Report: https://vishalsingh.org/zipcode/place-across-an-american-life/report
- Markdown: https://vishalsingh.org/zipcode/place-across-an-american-life/report.md
- BibTeX key: singh2026placeacrossanamericanlife

```bibtex
@misc{singh2026placeacrossanamericanlife,
  author = {Singh, Vishal},
  title = {Place across an American life},
  year = {2026},
  url = {https://vishalsingh.org/zipcode/place-across-an-american-life},
  note = {In ZIP Code Destiny. Published 11 October 2026.}
}
```

## Principal estimates

The facts the dek and the opening prose cite, in order of appearance.

| Estimand | Population | Period | Value | Interval | n | Weights | Source | Fact id |
| --- | --- | --- | ---: | --- | ---: | --- | --- | --- |
| births-weighted mean of county shares (the pooled share), bootstrap interval over counties | live births in the least deprived tenth of the named counties by county deprivation rank | births in 2018 to 2024 | 7.4% | 95% CI 7.2% to 7.7% | 57 counties | births in each county | CDC WONDER natality | lbw_d1 |
| births-weighted mean of county shares (the pooled share), bootstrap interval over counties | live births in the most deprived tenth of the named counties by county deprivation rank | births in 2018 to 2024 | 9.7% | 95% CI 9.2% to 10.3% | 58 counties | births in each county | CDC WONDER natality | lbw_d10 |
| population-weighted mean of tract life expectancy in a tenth of tracts (rank bins of the ADI; tracts with tied ranks fall in the same tenth), bootstrap interval over counties | residents of tracts in the least deprived tenth of tracts by ADI national percentile | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population | 82.3 | 95% CI 82.0 to 82.6 | 7,820 tracts | ACS 2019-2023 tract population | USALEEP life expectancy | le_d1 |
| population-weighted mean of tract life expectancy in a tenth of tracts (rank bins of the ADI; tracts with tied ranks fall in the same tenth), bootstrap interval over counties | residents of tracts in the most deprived tenth of tracts by ADI national percentile | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population | 74.0 | 95% CI 73.7 to 74.3 | 7,844 tracts | ACS 2019-2023 tract population | USALEEP life expectancy | le_d10 |
| HUD's line for a housing cost burden: housing costs above this share of household income | households in the HUD CHAS tabulation | 2018 to 2022 | 30% |   |   | none (definition) | HUD CHAS housing affordability | burden_line |
| pooled share of the tenure's households with the burden over tracts in the tenth, bootstrap over counties | renter households in tracts with an Area Deprivation Index (tenths hold near-equal numbers of tracts; tracts with equal ranks stay together), least deprived tenth | 2018 to 2022; Area Deprivation Index 2023 | 44.4% | 95% CI 41.9% to 46.7% | 8,208 tracts | households of the tenure with a computed burden | HUD CHAS housing affordability | adi_ren_d1 |
| pooled share of the tenure's households with the burden over tracts in the tenth, bootstrap over counties | renter households in tracts with an Area Deprivation Index (tenths hold near-equal numbers of tracts; tracts with equal ranks stay together), most deprived tenth | 2018 to 2022; Area Deprivation Index 2023 | 47.6% | 95% CI 46.6% to 48.4% | 8,207 tracts | households of the tenure with a computed burden | HUD CHAS housing affordability | adi_ren_d10 |

## Findings

Low birthweight was 7.4% (95% CI 7.2% to 7.7%) of births in the least deprived tenth of the counties that CDC WONDER names individually and 9.7% (95% CI 9.2% to 10.3%) in the most deprived, for births in 2018 to 2024. Life expectancy at birth was 82.3 (95% CI 82.0 to 82.6) years in the least deprived tenth of census tracts and 74.0 (95% CI 73.7 to 74.3) years in the most deprived, for deaths in 2010 to 2015.[1] The share of renter households that spend more than 30% of their income on housing was 44.4% (95% CI 41.9% to 46.7%) in the least deprived tenth of tracts and 47.6% (95% CI 46.6% to 48.4%) in the most deprived, for 2018 to 2022.

The three comparisons rank places by the Area Deprivation Index (ADI), a national percentile built from census measures of income, education, employment and housing.[2] Low birthweight and life expectancy are worse at each step toward the deprived end of the ranking. Renters' cost burden changes little along it, and among low-income renters it runs the other way. The three comparisons come from different records, places and years. *ZIP Code Destiny* follows comparisons like these from birth through old age. It asks where a pattern holds, where it fails, and how the answer changes when the comparison becomes more specific.

The title names that inquiry. An address places a person among schools, housing, jobs, services and other residents, and it also reflects who can afford to live there or chooses to. The book examines those differences without assuming that a location settles anyone's future. The word "destiny" is a proposition to test, and each chapter reports how much its evidence supports. "ZIP code" is shorthand as well. Most chapters count census tracts, counties, school districts or survey respondents, and the unit is named in every chapter.

### The questions that connect the chapters

The life course sets the order of the questions. At birth they concern the health of newborns and survival through the first year. School brings achievement, gaps between groups of students, and exclusion through discipline. The move into adulthood brings college, teen births and the adult incomes of children raised in different neighborhoods.

In the working years an address appears in the data as a home value, a mortgage application, a rent burden, a job or a supply of doctors. Some records describe the people who live in a place, and others describe the employers or institutions located there. Keeping the two apart is part of the work. A county with well-paid jobs need not have residents who hold them, and a neighborhood of expensive homes can have renters who struggle to pay.

Health, early death and well-being add further measures: illness and health behavior beside access to care, the geography of different causes of death, social connection, and people's own ratings of their lives. The later chapters take up disability, retirement income, Medicare and well-being at different ages.

(Diagram: The chapters follow outcomes from pregnancy to deaths before old age, with surroundings and well-being measured at every age)

Across these subjects the aim is to find which patterns recur and where they stop. Some outcomes change steadily with neighborhood deprivation, as birthweight and life expectancy do. Others change little with it, as renters' cost burden does, follow another ordering, or change direction when the comparison moves inside a county. Groups in the same place can also reach different outcomes. These departures show what a general account of disadvantage can explain and what it cannot.

The order does not follow one person from a birth certificate to a death certificate. The chapters draw on different populations and periods. Some sources follow birth cohorts into adulthood, some interview different people each year, and some describe a place in a single year. The life course orders the questions without turning these records into one biography.

### A view no single source provides

The chapters draw on administrative records, surveys and estimates that other researchers publish. Birth and death records describe the start and the end of life. Education data compare school districts. The Opportunity Atlas reports the adult incomes of children raised in different places, from linked records that its researchers assembled. Housing and lending records describe prices, rents, applications and eviction filings. Employment records separate jobs at workplaces from jobs held by residents. Health and well-being surveys ask about experiences that no tax return or Medicare payment records.

Each source was made for its own purpose, and their boundaries and dates do not always line up. A census tract, a school district and a county can describe different settings even when their names suggest the same place. Each chapter therefore states what its record counts, whom it covers and what comparison it allows. An eviction filing is a petition in court and not a completed eviction (Chapter 5.4, Renting, affordability and eviction (https://vishalsingh.org/zipcode/renting-affordability-and-eviction)). A college earnings figure describes the former students its source covers (Chapter 4.1, College access, price and earnings (https://vishalsingh.org/zipcode/college-access-price-and-earnings)). A neighborhood estimate of an illness can be a modeled estimate and not a count of residents who have it.

The sources are not always independent of one another. The CDC PLACES neighborhood health estimates draw on the Behavioral Risk Factor Surveillance System and on census inputs that the deprivation index also uses, so agreement between them is not an independent confirmation.[3] A change of source can also show what a map cannot, such as whether the adults who report a behavior are the same adults who report a given household income (Chapter 7.2, Health behaviors, person by person (https://vishalsingh.org/zipcode/health-behaviors-person-by-person)). The chapters keep these records apart and do not combine them into one score of a good or a bad place to live.

### How the chapters compare places

Many chapters map an outcome to show where it is high or low, and most set it against income or neighborhood deprivation. A distribution shows how widely places differ and whether a national average describes many of them. The ADI supplies the recurring ranking. The book also examines the ranking itself, what its ingredients emphasize and how it treats expensive places (Chapter 1.1, What the deprivation index measures (https://vishalsingh.org/zipcode/anatomy-of-deprivation)), and what happens when neighborhoods are combined into counties (Chapter 1.2, What a county average leaves out (https://vishalsingh.org/zipcode/deprivation-at-two-zooms)). A common measure makes the chapters comparable, and it can miss differences that matter. Across all ten tenths of tracts, the renter shares over the burden line stayed within 3.6 points of one another, for 2018 to 2022. Among renters with low incomes, at or below 80% of the area median family income, the share was higher in the least deprived tenth, 75.3% (95% CI 72.6% to 78.1%) against 62.5% (95% CI 61.6% to 63.3%). Gross rent and family income are among the census measures inside the index, so this comparison sorts a ratio of rent to income partly on its own parts (Chapter 1.1, What the deprivation index measures (https://vishalsingh.org/zipcode/anatomy-of-deprivation)).

The next comparison usually narrows the question. A national relationship can reflect the mix of regions in the country, and comparing neighborhoods inside the same county removes that source of difference (Chapter 10.1, Gradients, slopes and variance (https://vishalsingh.org/zipcode/gradients-slopes-and-variance)). Comparing groups along the same ranking shows whether a difference between them persists at every level of it. In County Health Rankings data for births in 2016 to 2022, low birthweight was 12.4% (95% CI 11.9% to 12.8%) of births to non-Hispanic Black mothers in the least deprived quarter of the counties WONDER names, against 7.6% (95% CI 7.5% to 7.7%) of births to non-Hispanic white mothers in the most deprived quarter (Chapter 2.1, Birthweight and preterm birth (https://vishalsingh.org/zipcode/birthweight-and-preterm-birth)). Survey responses from individuals answer a different problem, because a pattern across neighborhood averages need not describe the people inside the neighborhoods (ecological inference).

The chapters also show how much a finding depends on measurement and coverage. Counting every county once answers a different question from counting counties by their residents. A missing value is not a zero. An interval around an estimate describes its uncertainty under the stated method, while a band of places describes how widely the places differ. Each chapter explains these points where they change the reading of a result, and its methods section gives the full procedure.

Setting each end of a ranking against the country as a whole shows a second pattern. For most outcomes the most deprived places stand farther from the national value than the least deprived places do, and for some the least deprived places stand farther (Chapter 9.1, How far one ranking goes (https://vishalsingh.org/zipcode/how-far-one-ranking-goes)).

Table 1. The most deprived places usually stand farther from the national value than the least deprived places do

No chapter has to find the pattern of the one before it. Traffic pollution is lower in more deprived tracts across the country and higher in more deprived tracts within a county (Chapter 6.1, Air pollution (https://vishalsingh.org/zipcode/air-pollution)), a result that revises the national picture and is reported as found.

### One comparison followed in detail

Chapter 8.4, Place and the length of a life (https://vishalsingh.org/zipcode/place-and-the-length-of-a-life) takes the life expectancy contrast above through the steps that the other chapters apply to their own outcomes. The difference between the least and the most deprived tenths of tracts is 8.3 (95% CI 7.9 to 8.7) years, with an interval from resampling whole counties. Both averages are weighted by population and rest on modeled estimates for non-institutional tracts with the required data, and they describe tracts and not the lifespan of any resident. The chapter asks whether the ordering holds across the middle of the deprivation scale, what a county average leaves out, how far apart nearby neighborhoods are under a stated rule, and how much of the spread among tracts is estimation error. The deprivation measures describe 2019 to 2023, after the deaths behind the estimates, and that timing limits what the comparison can show.

The same chapter separates its description of places from the question of what moving would do to a family. It reports a randomized housing-voucher experiment in its own box, labeled with the study's design, and the book keeps every such study apart from its own comparisons. Life expectancy is one example of the method. It does not stand in for birth, school, housing or well-being, and the book does not use it to rank other parts of life.

### How the book is organized

*Measuring rich and poor* opens the book with the ranking itself. It shows what the deprivation index measures and why home values set most of it, and what a county's rank says about the tracts inside it. *An American life* then follows the life course in seven parts: birth; school; becoming an adult (college, teen births, adult income and the mental health of young adults); home, work and money (jobs, home values, lending, rent and eviction); environment, connection and well-being (pollution, civic life, partnership and how people rate their lives at different ages); health and care (behaviors, chronic conditions, the supply of care, disability, retirement income and Medicare); and early death (the causes of death before old age and the length of a life).

*Reading the patterns* sets the outcomes of the book side by side, each on the ranking its chapter uses, and asks how often they change steadily with deprivation. Its later chapters group neighborhoods by their measured characteristics and look for places whose outcomes differ from a model's prediction. A type summarizes a set of places, and a model's residual records what the model leaves out, and neither identifies a policy that works. The volume closes with research of a different kind, randomized experiments and other designs that study moving families or changing places, reported apart from the book's own comparisons. *Methods and sources* explains how the estimates are made, from gradients and slopes to the grouping of places and counts that the sources withhold. A reader can consult it beside any chapter or start there.

(Diagram: The book defines its ranking of places, follows outcomes across the life course, then reads the patterns together)

The sequence follows the life course, and the subjects are not confined to particular ages. Housing and pollution concern children as well as adults, disability can begin long before retirement, and social connection matters throughout life. The order gives a readable route through related questions and does not assign each exposure to an age.

### Reading the book

Reading in order begins with Chapter 1.1, What the deprivation index measures (https://vishalsingh.org/zipcode/anatomy-of-deprivation), on what the ranking measures. A reader who wants the life course first can start with Chapter 2.1, Birthweight and preterm birth (https://vishalsingh.org/zipcode/birthweight-and-preterm-birth), and a reader who wants the method first can start with Chapter 8.4, Place and the length of a life (https://vishalsingh.org/zipcode/place-and-the-length-of-a-life), the comparison described above. The [contents page](https://vishalsingh.org/zipcode/contents) lists every chapter with its finding. Reading paths connect chapters that ask related questions across the volumes, among them [the argument in eight chapters](https://vishalsingh.org/zipcode/path/core), [differences inside one county](https://vishalsingh.org/zipcode/path/the-hyper-local-map), [race and place](https://vishalsingh.org/zipcode/path/race-and-place), [outcomes that do not follow the gradient](https://vishalsingh.org/zipcode/path/the-exception) and [methods first](https://vishalsingh.org/zipcode/path/methods-first).

Three conventions run through every chapter. A map or a chart shows gray where a value is missing or withheld and never draws it as zero. Descriptions of places use verbs that describe, such as "is higher where" and "moves with", and causal verbs appear only in boxes that report a study with a stated design. Each estimate carries its unit, its population, its period and, where it has one, an interval. The [reading guide](https://vishalsingh.org/zipcode/reference/reading-guide) explains these conventions and the kinds of evidence behind each claim. The contents also list planned chapters, which describe questions the book has not yet answered.

### What these data cannot show

Most of the book's own analyses describe places. People differ in where they live, and places differ in who lives in them. Comparing more similar places or groups narrows an explanation without ruling out every alternative. A relationship between neighborhood conditions and an outcome does not show what would happen to a person who moved or what a policy would change. Studies with experiments or other stated designs bear on those questions, and the book reports them separately, for the populations, programs and outcomes they studied.

The numbers in this introduction come from different units, sources and years. Births are compared across counties for 2018 to 2024, and for 2016 to 2022 in the comparison by mother's race. Deaths are compared across tracts for 2010 to 2015, and housing costs across tracts for 2018 to 2022. Each is ranked on a deprivation index for 2019 to 2023. They are not measurements of the same people. Some places and groups are missing from the records, some values are withheld, some estimates come from models or from data collected years apart, and several chapters describe cohorts born decades ago and not conditions today. Each chapter states these limits because they decide which comparisons are informative and which questions stay open.

## Tables

### Table 1. The most deprived places usually stand farther from the national value than the least deprived places do

Mean at each end of the ranking each chapter uses, divided by the value for all the places of the comparison, on a log scale so that twice and half the national value sit the same distance from it

| Domain | Outcome | Least deprived end | United States | Most deprived end | Least deprived end divided by the US value | Most deprived end divided by the US value | Unit |
| --- | --- | --- | --- | --- | ---: | ---: | --- |
| Birth and school | Infant mortality | 4.0 | 5.7 | 8.8 | 0.71 | 1.55 | Infant deaths per 1,000 live births |
| Becoming an adult | Distance to a four-year college | 6.1 | 9.6 | 29.3 | 0.64 | 3.05 | Miles to the nearest four-year campus, average resident |
| Becoming an adult | Incarceration, Black children from low-income families | 3.8 | 5.2% | 6.4 | 0.73 | 1.23 | Percent incarcerated on April 1, 2010 |
| Home, work and money | Mortgage denial rate | 7.8 | 9.0% | 13.8 | 0.87 | 1.54 | Denied share of decided applications (%) |
| Health and care | Current smoking | 9.1 | 14.5% | 22.8 | 0.63 | 1.57 | Percent of adults |
| Health and care | Residents per primary care physician | 1,159 | 1,334 | 2,730 | 0.87 | 2.05 | Residents per provider |
| Early death | Homicide death rate, tracts | 1.8 | 5.4 | 14.8 | 0.32 | 2.72 | Homicide deaths per 100,000 residents a year |
| Early death | Life expectancy at birth | 82.3 | 78.8 | 74.0 | 1.04 | 0.94 | Life expectancy at birth (years) |

Note: Eight of the rows of the chapter How far one ranking goes, chosen to cover the stages of life: five where the most deprived end lies farther from the national value and three where the least deprived end does, about the share in the full set. The chapter shows all of them and counts which end lies farther.

Note: The national value pools all the places of each comparison into one bin, with the estimator and weights of its chapter, and is published there. It covers the places that the chapter ranks, so places without a rank or an estimate are left out.

Source: Figures of the chapters named in each row, data years 1978 to 2024 as each chapter states; estimates as published there: full counts, crude rates, population-weighted means, unweighted survey means or modeled estimates.

Data: https://vishalsingh.org/zipcode/report/place-across-an-american-life/figures/ends_vs_us.json

## All estimates

Every fact in facts.json (14), sorted by id.

| Fact id | Value | Unit | Estimand | Population | Period | Interval | n | Weights | Source | Note |
| --- | ---: | --- | --- | --- | --- | --- | ---: | --- | --- | --- |
| adi_li_ren_d1 | 75.3% | percent of low-income renter households with housing costs above 30% of income | low-income households (at or below 80 percent of HAMFI) with a burden over low-income households of the tenure, pooled over tracts in the tenth, bootstrap over counties | low-income renter households in tracts with an Area Deprivation Index, least deprived tenth | 2018 to 2022; Area Deprivation Index 2023 | 95% CI 72.6% to 78.1% | 7,840 tracts | low-income households of the tenure | HUD CHAS housing affordability |   |
| adi_li_ren_d10 | 62.5% | percent of low-income renter households with housing costs above 30% of income | low-income households (at or below 80 percent of HAMFI) with a burden over low-income households of the tenure, pooled over tracts in the tenth, bootstrap over counties | low-income renter households in tracts with an Area Deprivation Index, most deprived tenth | 2018 to 2022; Area Deprivation Index 2023 | 95% CI 61.6% to 63.3% | 8,194 tracts | low-income households of the tenure | HUD CHAS housing affordability |   |
| adi_ren_d1 | 44.4% | percent of renter households with housing costs above 30% of income | pooled share of the tenure's households with the burden over tracts in the tenth, bootstrap over counties | renter households in tracts with an Area Deprivation Index (tenths hold near-equal numbers of tracts; tracts with equal ranks stay together), least deprived tenth | 2018 to 2022; Area Deprivation Index 2023 | 95% CI 41.9% to 46.7% | 8,208 tracts | households of the tenure with a computed burden | HUD CHAS housing affordability |   |
| adi_ren_d10 | 47.6% | percent of renter households with housing costs above 30% of income | pooled share of the tenure's households with the burden over tracts in the tenth, bootstrap over counties | renter households in tracts with an Area Deprivation Index (tenths hold near-equal numbers of tracts; tracts with equal ranks stay together), most deprived tenth | 2018 to 2022; Area Deprivation Index 2023 | 95% CI 46.6% to 48.4% | 8,207 tracts | households of the tenure with a computed burden | HUD CHAS housing affordability |   |
| adi_ren_range | 3.6 | percentage points, largest minus smallest renter share with housing costs above 30% of income among the ten tenths | largest minus smallest of the ten tenth means (a range of estimates, no interval) | renter households in tracts with an Area Deprivation Index (tenths hold near-equal numbers of tracts; tracts with equal ranks stay together) | 2018 to 2022; Area Deprivation Index 2023 |   | 10 tenths of tracts | households of the tenure with a computed burden | HUD CHAS housing affordability |   |
| burden_line | 30% | percent of household income | HUD's line for a housing cost burden: housing costs above this share of household income | households in the HUD CHAS tabulation | 2018 to 2022 |   |   | none (definition) | HUD CHAS housing affordability |   |
| hamfi_cut | 80% | percent of HUD area median family income | the income line for a low-income household: income at or below this share of HUD's area median family income (HAMFI) | households in the HUD CHAS tabulation | 2018 to 2022 |   |   | none (definition) | HUD CHAS housing affordability |   |
| lbw_black_q1 | 12.4% | percent of live births under 2,500 grams | pooled share of births, bootstrap interval over counties | live births to non-Hispanic Black mothers in the least deprived quartile of the named counties by deprivation rank | births in 2016 to 2022 | 95% CI 11.9% to 12.8% | 133 counties | births to mothers of the group in each county | County Health Rankings |   |
| lbw_d1 | 7.4% | percent of live births under 2,500 grams | births-weighted mean of county shares (the pooled share), bootstrap interval over counties | live births in the least deprived tenth of the named counties by county deprivation rank | births in 2018 to 2024 | 95% CI 7.2% to 7.7% | 57 counties | births in each county | CDC WONDER natality |   |
| lbw_d10 | 9.7% | percent of live births under 2,500 grams | births-weighted mean of county shares (the pooled share), bootstrap interval over counties | live births in the most deprived tenth of the named counties by county deprivation rank | births in 2018 to 2024 | 95% CI 9.2% to 10.3% | 58 counties | births in each county | CDC WONDER natality |   |
| lbw_white_q4 | 7.6% | percent of live births under 2,500 grams | pooled share of births, bootstrap interval over counties | live births to non-Hispanic white mothers in the most deprived quartile of the named counties by deprivation rank | births in 2016 to 2022 | 95% CI 7.5% to 7.7% | 145 counties | births to mothers of the group in each county | County Health Rankings |   |
| le_d1 | 82.3 | years of life expectancy at birth | population-weighted mean of tract life expectancy in a tenth of tracts (rank bins of the ADI; tracts with tied ranks fall in the same tenth), bootstrap interval over counties | residents of tracts in the least deprived tenth of tracts by ADI national percentile | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population | 95% CI 82.0 to 82.6 | 7,820 tracts | ACS 2019-2023 tract population | USALEEP life expectancy |   |
| le_d10 | 74.0 | years of life expectancy at birth | population-weighted mean of tract life expectancy in a tenth of tracts (rank bins of the ADI; tracts with tied ranks fall in the same tenth), bootstrap interval over counties | residents of tracts in the most deprived tenth of tracts by ADI national percentile | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population | 95% CI 73.7 to 74.3 | 7,844 tracts | ACS 2019-2023 tract population | USALEEP life expectancy |   |
| le_decile_drop | 8.3 | years of life expectancy at birth lower in the most deprived tenth of tracts | difference between the population-weighted mean life expectancy of the least and the most deprived tenth of tracts, bootstrap interval over counties (draws shared by the two means) | residents of non-institutional tracts in the 50 states and DC with a life expectancy estimate and an ADI rank | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population | 95% CI 7.9 to 8.7 | 7,844 tracts | ACS 2019-2023 tract population | USALEEP life expectancy |   |

## Definitions

- Census tract: A small area drawn by the Census Bureau to stand in for a neighborhood. Boundaries are redrawn after each decennial census.
- Area Deprivation Index (ADI): A score that ranks neighborhoods by income, education, employment and housing conditions drawn from the American Community Survey. In this book a higher national percentile means a more deprived neighborhood.
- Modeled estimate: A figure produced by a statistical model instead of a direct count or measurement, usually for small areas where direct data are thin, borrowing strength from surveys, neighboring areas or other data. Neighboring areas can share errors, so single extremes call for caution.
- Ecological inference: Drawing conclusions about individuals from data about places or groups. A pattern across counties need not hold for the people inside them.

## Methods

This introduction computes nothing. Each number it prints is copied, with its population, period, weights and interval, from the chapter that estimates it, and the comparisons come from different sources, units and years.

### Where the numbers come from

Low birthweight by deprivation tenth covers live births in 2018 to 2024 in the counties that CDC WONDER names individually, those of 100,000 residents or more, and the comparison by mother's race uses County Health Rankings rates for births in 2016 to 2022 (Birthweight and preterm birth).

Life expectancy at birth is the USALEEP estimate for deaths in 2010 to 2015, averaged with population weights over tenths of non-institutional census tracts, with an interval from a bootstrap over counties (Place and the length of a life).

The housing cost burden is the share of renter households whose housing costs exceed the stated share of income, from HUD's tabulation of the American Community Survey for 2018 to 2022, pooled over tracts in each tenth, with low-income households defined against HUD's area median family income (Renting, affordability and eviction).

### Rankings and periods

Every comparison ranks places on the 2023 Area Deprivation Index, built on the American Community Survey for 2019 to 2023. Births are ranked by county and the other comparisons by census tract, so the tenths are different sets of places, and the outcome periods differ from one another and from the period of the index.

## What these data cannot show

Most of the book's own analyses describe places. People differ in where they live, and places differ in who lives in them. Comparing more similar places or groups narrows an explanation without ruling out every alternative. A relationship between neighborhood conditions and an outcome does not show what would happen to a person who moved or what a policy would change. Studies with experiments or other stated designs bear on those questions, and the book reports them separately, for the populations, programs and outcomes they studied.

The numbers in this introduction come from different units, sources and years. Births are compared across counties for 2018 to 2024, and for 2016 to 2022 in the comparison by mother's race. Deaths are compared across tracts for 2010 to 2015, and housing costs across tracts for 2018 to 2022. Each is ranked on a deprivation index for 2019 to 2023. They are not measurements of the same people. Some places and groups are missing from the records, some values are withheld, some estimates come from models or from data collected years apart, and several chapters describe cohorts born decades ago and not conditions today. Each chapter states these limits because they decide which comparisons are informative and which questions stay open.

## References

1. Arias, Elizabeth; Escobedo, Luis A.; Kennedy, Jiaquan; Fu, Chenxi; Cisewski, Jodi (2018). U.S. Small-area Life Expectancy Estimates Project: Methodology and Results Summary. Vital and Health Statistics, Series 2, Number 181, National Center for Health Statistics. https://www.cdc.gov/nchs/data/series/sr_02/sr02_181.pdf
2. Kind, Amy J. H.; Buckingham, William R. (2018). Making Neighborhood-Disadvantage Metrics Accessible: The Neighborhood Atlas. New England Journal of Medicine 378(26): 2456-2458. https://doi.org/10.1056/NEJMp1802313
3. Greenlund, Kurt J.; Lu, Hua; Wang, Yan; Matthews, Kevin A.; LeClercq, Jennifer M.; Lee, Benjamin; Carlson, Susan A. (2022). PLACES: Local Data for Better Health. Preventing Chronic Disease 19: 210459. https://doi.org/10.5888/pcd19.210459

## Provenance

- Pipeline version: 0.1.0
- Build script: book2/chapters/place-across-an-american-life/build.py (sha256 92ec708004c8dd0b1e7c5c0e3a8b6c814571c3f18764fd00dfe954173363de8b)
- Built on: 2026-10-11
- Site commit: c2b1ff3309cb4a10833f884da91ac0057a66ad0a

Inputs:

| File | Size (bytes) | SHA-256, first 1 MB | SHA-256, last 1 MB |
| --- | ---: | --- | --- |
| C:/github/zipcode-book/content/chapters/birthweight-and-preterm-birth/data/facts.json | 39,621 | 9332a0015b74d0cb3babb0d72c1b75129ee521b3719bb6e2d1abf87321e9308a | 9332a0015b74d0cb3babb0d72c1b75129ee521b3719bb6e2d1abf87321e9308a |
| C:/github/zipcode-book/content/chapters/how-far-one-ranking-goes/data/figures/ends_vs_us.json | 10,419 | bd0996f5b4db2ac8bf6454f650dda1df440fe65dc2455d87a8d83dca8b71ff7b | bd0996f5b4db2ac8bf6454f650dda1df440fe65dc2455d87a8d83dca8b71ff7b |
| C:/github/zipcode-book/content/chapters/place-and-the-length-of-a-life/data/facts.json | 71,140 | 9eed1ad6694ae817ab5800fd787f79ec8a01a706b2d211bccb978f17d11e6e6d | 9eed1ad6694ae817ab5800fd787f79ec8a01a706b2d211bccb978f17d11e6e6d |
| C:/github/zipcode-book/content/chapters/place-and-the-length-of-a-life/data/refs.json | 1,878 | 2d976463c36bc44fd12832d343cc0a844526c07065c1c3444516db9b007dd7f8 | 2d976463c36bc44fd12832d343cc0a844526c07065c1c3444516db9b007dd7f8 |
| C:/github/zipcode-book/content/chapters/renting-affordability-and-eviction/data/facts.json | 101,510 | e406f8a10f637011935e5a498813d757cff8c7fa2eb2a6eec5f4d034d769f947 | e406f8a10f637011935e5a498813d757cff8c7fa2eb2a6eec5f4d034d769f947 |

Outputs:

| File | SHA-256 |
| --- | --- |
| facts.json | a9a9ae04177edbdbd7ccaccdb9e6b89cbe30e9ae39668cd3d318206ca943896d |
| figures/ends_vs_us.json | ce42517e10b24102fb042a692376502bfdb0c15da47ed02a99220f98a9f61da0 |
| meta.json | 72fbc882772a4a1bdedff815dfc3c214d99d5202053bfb21501195f0368a0bce |
| refs.json | 4a48e2d2a005c834353a11990cea15a99b62de6520266085c7eae7b12282b11e |

## Machine formats

- Report (HTML): https://vishalsingh.org/zipcode/place-across-an-american-life/report
- Report (Markdown): https://vishalsingh.org/zipcode/place-across-an-american-life/report.md
- Every estimate (JSON): https://vishalsingh.org/zipcode/report/place-across-an-american-life/facts.json
- Table 1, figure "ends_vs_us" (JSON): https://vishalsingh.org/zipcode/report/place-across-an-american-life/figures/ends_vs_us.json
- Index of every report: https://vishalsingh.org/zipcode/llms.txt; all reports in one file: https://vishalsingh.org/zipcode/llms-full.txt

---

# What the deprivation index measures: report

Chapter 1.1 of ZIP Code Destiny, by Vishal Singh (NYU Stern). Generated from the chapter's data files; the reading version is at https://vishalsingh.org/zipcode/anatomy-of-deprivation. Table n of this report is figure n of the chapter.

## Identification

- Title: What the deprivation index measures
- Label: Weighing the ingredients
- Chapter: 1.1 (Part I, Rankings and scale; Measuring rich and poor)
- Question: What goes into the Area Deprivation Index, and which components weigh most in a neighborhood's score?
- Dek: The Area Deprivation Index orders 75,717 census tracts almost in reverse order of median home value, with a rank correlation of −0.988, and the home value term alone accounts for 74.3% of the variance of the score its formula produces.
- Grain: neighborhoods
- Status: published
- Published: 2026-10-09
- Updated: not revised
- Data sources:
  - Area Deprivation Index. Neighborhood Atlas ADI, 2023 release (version 4.0.1), national percentile ranks of block groups averaged to 2020 tracts; years: 2023 release (version 4.0.1; built on ACS 2019-2023); weights: none (index); tract values are unweighted means of block-group ranks
  - American Community Survey. ACS 2019-2023 five-year tract estimates: the staged tract file and a Census API pull of housing, occupation, income-distribution and schooling tables; years: 2019-2023 (5-year); weights: ACS 2019-2023 tract population
  - CDC/ATSDR Social Vulnerability Index. CDC/ATSDR Social Vulnerability Index 2022, tracts: the socioeconomic theme percentile and, for a sensitivity check, the shares of adults without a high school diploma and of households without internet (ACS 2018 to 2022); years: 2022 release (ACS 2018-2022); weights: none (index); percentile ranks of tracts
  - USALEEP life expectancy. NCHS small-area life expectancy at birth, deaths 2010 to 2015, 2020 tracts (interpolated), institutional tracts dropped; years: 2010-2015; weights: ACS 2019-2023 tract population
  - Opportunity Atlas. Mean adult household income rank of children from low-income families, 1978 to 1983 birth cohorts, 2020 tracts (interpolated); years: 1978-1983 birth cohorts (2018 release); weights: ACS 2019-2023 tract population
- Canonical URL: https://vishalsingh.org/zipcode/anatomy-of-deprivation
- Report: https://vishalsingh.org/zipcode/anatomy-of-deprivation/report
- Markdown: https://vishalsingh.org/zipcode/anatomy-of-deprivation/report.md
- BibTeX key: singh2026anatomyofdeprivation

```bibtex
@misc{singh2026anatomyofdeprivation,
  author = {Singh, Vishal},
  title = {What the deprivation index measures},
  year = {2026},
  url = {https://vishalsingh.org/zipcode/anatomy-of-deprivation},
  note = {In ZIP Code Destiny. Published 9 October 2026.}
}
```

## Principal estimates

The facts the dek and the opening prose cite, in order of appearance.

| Estimand | Population | Period | Value | Interval | n | Weights | Source | Fact id |
| --- | --- | --- | ---: | --- | ---: | --- | --- | --- |
| count of tracts in the frame | 2020 census tracts with a published ADI rank, a population and all 15 rebuilt measures | ACS 2019 to 2023 five-year estimates, ADI 2023 release | 75,717 |   |   | none (full count) | American Community Survey | n_tracts |
| population-weighted correlation of the percentile ranks of the published tract ADI and of median home value, interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of the tracts in the frame (50 states and DC; a published ADI rank, a population and all rebuilt measures) | ACS 2019 to 2023 five-year estimates, ADI 2023 release | −0.988 | 95% CI −0.989 to −0.986 | 75,717 tracts | ACS 2019-2023 tract population | Area Deprivation Index | rho_adi_homevalue |
| variance of the home value term alone (Singh's coefficients times the measures in their own units) as a share of the variance of the score, with no covariance with the other terms allocated to it; interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of the tracts in the frame (50 states and DC; a published ADI rank, a population and all rebuilt measures) | ACS 2019 to 2023 five-year estimates, ADI 2023 release | 74.3% | 95% CI 70.5% to 77.0% | 75,717 tracts | ACS 2019-2023 tract population | Area Deprivation Index | hv_own_share |
| population-weighted correlation of the percentile ranks of the published tract ADI and of the replica score in raw units, interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of the tracts in the frame (50 states and DC; a published ADI rank, a population and all rebuilt measures) | ACS 2019 to 2023 five-year estimates, ADI 2023 release | +0.991 | 95% CI 0.990 to 0.992 | 75,717 tracts | ACS 2019-2023 tract population | Area Deprivation Index | rho_unstd_adi |
| count of measures in the index; a definition | measures in Singh's deprivation index as the Neighborhood Atlas applies it (since the 2020 release it counts households without internet where Singh counted households without a telephone) | 2023 release of the ADI (ACS 2019 to 2023) | 17 |   |   | none (definition) | Area Deprivation Index | n_measures |
| count of dollar-denominated measures; a definition | measures of the index that are dollar amounts (median family income, home value, gross rent and monthly mortgage cost) | 2023 release of the ADI (ACS 2019 to 2023) | 4 |   |   | none (definition) | Area Deprivation Index | n_dollar |
| count of tracts | tracts of the frame with a poverty rate at or above the high-poverty threshold | ACS 2019 to 2023 five-year estimates, ADI 2023 release | 5,100 |   |   | none (full count) | American Community Survey | n_hp |
| count of states; a definition | states of the frame, besides the District of Columbia | ACS 2019 to 2023 five-year estimates, ADI 2023 release | 50 |   |   | none (definition) | American Community Survey | n_states |
| sum of ACS tract population | residents of the high-poverty tracts of the frame | ACS 2019 to 2023 five-year estimates, ADI 2023 release | 17.2 million |   |   | none (full count) | American Community Survey | pop_hp |
| poverty rate at or above which a tract is called high-poverty; a definition | tracts called high-poverty in this chapter | ACS 2019 to 2023 five-year estimates, ADI 2023 release | 30% |   |   | none (definition) | American Community Survey | hp_threshold |
| share of residents of high-poverty tracts who live in tracts at or below the 50th percentile on the published ADI (percentile among the tracts of the frame for the replica), interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of high-poverty tracts of the frame | ACS 2019 to 2023 five-year estimates, ADI 2023 release | 16.1% | 95% CI 11.3% to 21.4% | 5,100 tracts | ACS 2019-2023 tract population | Area Deprivation Index | sh_hp_lowadi |
| percentile cutoff; a definition | rank at or below which a tract counts as among the less deprived | ACS 2019 to 2023 five-year estimates, ADI 2023 release | 50th |   |   | none (definition) | Area Deprivation Index | rank_cut |
| share of residents of high-poverty tracts who live in tracts at or below the 50th percentile on the replica score in standard deviation units (percentile among the tracts of the frame for the replica), interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of high-poverty tracts of the frame | ACS 2019 to 2023 five-year estimates, ADI 2023 release | 0.34% | 95% CI 0.15% to 0.55% | 5,100 tracts | ACS 2019-2023 tract population | Area Deprivation Index | sh_hp_lowstd |

## Findings

Across 75,717 census tracts, the Area Deprivation Index (ADI) of the Neighborhood Atlas has a population-weighted rank correlation of −0.988 (95% CI −0.989 to −0.986) with median home value. A rank correlation compares the positions of tracts in two rankings. The sign is negative because the index rises as home values fall, and the size means that the index orders tracts almost as home prices do.[1][2] A replica built from the index's published formula and American Community Survey (ACS) estimates for 2019 to 2023 reproduces the published tract rank with a rank correlation of +0.991 (95% CI 0.990 to 0.992). In the replica's score, the home value term alone accounts for 74.3% (95% CI 70.5% to 77.0%) of the variance.

The formula multiplies each of 17 census measures by a fixed coefficient and adds the products. 4 of the measures are dollar amounts and the rest are proportions or a log ratio, so the dollar amounts dominate the sum.

The same arithmetic shapes who the index counts as less deprived. In 5,100 tracts of the 50 states and DC, with 17.2 million residents, 30% or more of people are below the poverty line. The published ADI places 16.1% (95% CI 11.3% to 21.4%) of those residents in tracts ranked at or below the 50th national percentile, the less deprived end of its ranking. Standardizing the same measures before applying the same coefficients places 0.34% (95% CI 0.15% to 0.55%) of them there.

This chapter rebuilds the index, shows where its weight goes, compares it with other rankings of the same tracts and tests how closely each ranking tracks life expectancy and upward mobility. Chapter 10.1, Gradients, slopes and variance (https://vishalsingh.org/zipcode/gradients-slopes-and-variance) covers the variance of outcomes across states, counties and tracts and the slopes of outcomes on the index. Chapter 1.2, What a county average leaves out (https://vishalsingh.org/zipcode/deprivation-at-two-zooms) repeats the index for counties, and Chapter 10.2, Families of places (https://vishalsingh.org/zipcode/families-of-places) sorts tracts into groups instead of ranking them.

### How the Neighborhood Atlas builds the index

Gopal Singh chose the census measures and estimated their coefficients as factor scores from 1990 census tracts, for a study of mortality in the United States.[3] The Neighborhood Atlas at the University of Wisconsin applies his recipe to American Community Survey data and publishes national percentile ranks of census block groups, with the lowest ranks for the least deprived and the highest for the most deprived.[4] A block group's score is the sum of the 17 measures in Table 1, each multiplied by Singh's coefficient, and the Atlas publishes the ranks of the scores and not the scores.[2][5] The ADI used here is the 2023 release, version 4.0.1, rolled up from block groups to 2020 tracts. The dataset registry records it as built on ACS five-year estimates for 2019 to 2023, the period of the ACS measures used here. Since the 2020 release the Atlas has counted households without internet where Singh counted households without a telephone.[5]

Table 1. The Atlas score sums 17 measures with fixed coefficients, and 4 of them are dollar amounts

The measures cover education, work, income and housing. 4 are dollar amounts (median family income, home value, gross rent and the monthly mortgage cost), one is a log ratio of low-income to high-income households, and the rest are proportions of people, households or homes. The Atlas sums them in these units, with proportions entered as fractions of one and not as percentages.[2] The coefficients are close in size, so the unit of a measure, more than its coefficient, sets how much the measure moves the sum. A dollar amount in the hundreds of thousands varies over a wider range than a proportion.

### Most of the measures can be rebuilt from ACS data for 2020 tracts

The staged ACS tract file and a Census API pull of housing, occupation, income and schooling tables supply 15 of the 17 measures for 2020 census tracts. The 2 missing are the share of adults with at least a high school diploma and the share of households without internet. Some rebuilt measures are substitutes. The poverty rate counts persons where Singh counted families, white-collar work is the share of workers in management, business, science and arts occupations, and the mortgage cost is the median monthly owner cost of homes with a mortgage. Income disparity is Singh's ratio of low-income to high-income households, defined in the table, logged, with a continuity constant added to each count so that a tract with no low-income household keeps a value.

The replica works with tracts, and the Atlas ranks block groups. The staged ADI is the Atlas's block-group ranks averaged over each tract's block groups, without population weights. The frame covers the 50 states and DC. Puerto Rico's tracts are left out because the SVI and USALEEP have none, although the Atlas ranks them with the states.

Of the 84,415 tracts, 82,784 have a published rank and residents, and 75,717 of those have all 15 measures. The frame holds 93.0% of the residents of tracts with a rank. Of the tracts lost, 90.0% lack a median rent, home value or mortgage cost, which the ACS leaves out for tracts with almost no renters or almost no owners. The Atlas imputes missing values from the surrounding geography, and the replica does not.[5]

Table 2 shows how the rebuilt measures move together, from the measures of advantage (income, home value, rent, mortgage cost, white-collar work and homeownership) to the measures of hardship. The dollar measures move together, with home value correlating +0.86 (95% CI 0.85 to 0.87) with the mortgage cost and +0.71 (95% CI 0.68 to 0.74) with rent. They are a loose guide to hardship. Home value correlates −0.31 (95% CI −0.36 to −0.27) with the poverty rate across tracts, a weaker link than the −0.61 (95% CI −0.62 to −0.59) of family income. A score dominated by home value therefore follows the housing market more closely than it follows poverty.

Table 2. Home value correlates −0.31 with the poverty rate across tracts, a weaker link than family income has at −0.61

### In raw units, home value and family income set the score

Write the score of tract $$i$$ as a sum over the measures $$k$$ of a coefficient $$c_k$$ times the measure $$x_{ik}$$. The Atlas uses the measures as they come. A standardized version subtracts each measure's mean $$\bar x_k$$ and divides by its standard deviation $$s_k$$:

$$
S_i=\sum_{k} c_k\,x_{ik}
\qquad\text{and}\qquad
S_i^{z}=\sum_{k} c_k\,\frac{x_{ik}-\bar x_k}{s_k}.
$$

The first is the replica of the published score. The second applies the same coefficients, with the population-weighted mean and standard deviation of each measure across tracts. A sum is dominated by the terms that vary most, and variance is measured in squared units, so a measure with values in the hundreds of thousands contributes more variance than a proportion does. Table 3 splits the variance of each score among its measures. A measure's allocated share is the covariance of its term with the score divided by the variance of the score, and each covariance between two terms is shared equally between them, so the shares add to the whole of the variance.

In the raw-units score, median home value is allocated 85.4% (95% CI 83.1% to 87.0%) and median family income 14.2% (95% CI 12.6% to 16.5%). The two terms are correlated, so the allocation matters. The home value term alone has 74.3% (95% CI 70.5% to 77.0%) of the variance, and the home value and family income terms together have 99.2% (95% CI 99.2% to 99.3%).

Rent and the mortgage cost are allocated a further 0.38% (95% CI 0.36% to 0.41%), and the measures that are not dollar amounts 0.0003% (95% CI 0.0002% to 0.0003%) together. After standardization, no measure is allocated more than 12.4% (95% CI 12.2% to 12.6%), which is the near-poverty share, and home value is allocated 5.3% (95% CI 5.0% to 5.7%).

Table 3. In raw units home value accounts for 74.3% of the variance of the score on its own and 85.4% with its covariance with income shared out, against 5.3% after standardization

The same coefficients therefore give weights that depend on the scale of the measures. Petterson showed the pattern across the block groups of the Atlas, and Hannan and colleagues found that in New York State the Atlas ADI mainly reflects home value.[1][2] The shares here describe a population-weighted frame of tracts, a different population from either study.

### The replica reproduces the published rank

Ranking the replica's score among the frame's tracts gives a population-weighted rank correlation of +0.991 (95% CI 0.990 to 0.992) with the published tract rank. The match is close and not exact, because the replica uses tract estimates where the Atlas uses block groups, omits two measures and does not impute missing values. Standardizing the measures before applying the same coefficients lowers the correlation with the published rank to +0.72 (95% CI 0.66 to 0.77). The published ADI correlates −0.988 (95% CI −0.989 to −0.986) with the rank of median home value. The raw-units replica correlates −0.987 (95% CI −0.989 to −0.986) with it and the standardized replica −0.66 (95% CI −0.71 to −0.60).

The scale of the measures moves the ranking, and the choice of coefficients barely does. The first principal component of the standardized measures lets the data choose the weights, and it ranks the tracts almost as Singh's coefficients applied to the standardized measures do, with a rank correlation of +0.997 (95% CI 0.996 to 0.998).

The staged SVI file carries the two missing measures for ACS 2018 to 2022. Adding them to the replica, with the telephone measure's coefficient on households without internet because the Atlas does not publish one, moves the rank correlation of the standardized replica with the published rank to +0.71 (95% CI 0.65 to 0.77). The share of residents of high-poverty tracts that the standardized replica places at or below the 50th percentile moves to 0.61% (95% CI 0.32% to 0.93%). The raw-units replica stays at +0.991 (95% CI 0.990 to 0.992).

### The published ADI sits apart from rankings that standardize

Table 4 compares six rankings of the same tracts: the published ADI, its raw-units replica, the standardized replica, the first principal component of the standardized measures, the socioeconomic theme of the Social Vulnerability Index (SVI) of the CDC and ATSDR, and the poverty rate.[6][7] The SVI converts each measure to a percentile among tracts before summing within a theme, so it is standardized by construction. The published ADI and its replica form one pair, and the standardized replica and the principal component form another. The principal component, the SVI theme and the poverty rate each correlate more strongly with the standardized replica than with the published ADI.

Table 4. The published ADI correlates +0.991 with its raw-units replica and +0.72 with the standardized version of the same measures

### The ADI reads expensive places as less deprived

The disagreement between the published and the standardized ranking follows the cost of housing. Table 5 groups tracts by the median home value of their county, in quartiles with equal numbers of residents. County median home values are above $450,562 in the highest-cost quartile and below $216,301 in the lowest-cost quartile. The published percentile minus the standardized percentile averages +11.1 (95% CI 10.5 to 11.6) points in the lowest-cost counties and −16.5 (95% CI −21.3 to −11.5) points in the highest-cost counties. At the same standardized percentile, tracts in the highest-cost counties have a lower published ADI than tracts in the lowest-cost counties in every tenth of the standardized ranking. The gap widens toward the most deprived tenth.

The gap matters most where people are poor. 5,100 tracts in the frame, with 17.2 million residents, have poverty rates of 30% or more. The published ADI places 16.1% (95% CI 11.3% to 21.4%) of those residents in tracts ranked at or below the 50th national percentile.

The raw-units replica places 15.8% (95% CI 11.0% to 20.7%) of them there, and the standardized replica 0.34% (95% CI 0.15% to 0.55%). Across all 6,718 tracts of the 50 states and DC that have a published rank and a poverty rate, including tracts that lack some rebuilt measures, the published share is 20.6% (95% CI 14.1% to 27.9%).

The second tab of the figure lists the 15 tracts with the largest gaps among the 5,049 tracts that have a poverty rate of 30% or more and at least 1,000 residents. 12 are in New York City.

In the first row, a tract in San Francisco where 35.9% of residents are below the poverty line, the published ADI is 1, among the least deprived in the nation, and the standardized percentile is 99.7. The median home value of that tract is at the ACS top code, which applies to 495 tracts in the frame and caps a dollar value that would otherwise be higher.

Table 5. The published ADI percentile minus the standardized percentile averages −16.5 in the highest-cost counties and +11.1 in the lowest-cost counties

One Medicare payment model, ACO REACH, uses the ADI. The model adjusts payments for each beneficiary by the decile of a score that combines the ADI with the beneficiary's status as dually eligible for Medicare and Medicaid or as receiving the low-income subsidy. For performance year 2024 the Centers for Medicare and Medicaid Services added the state ADI to the score and weighted the national ADI, the state ADI and that status equally. It said the change would better identify underserved beneficiaries in areas with a high cost of living.[8] The state ADI compares tracts within a state, and this chapter does not test how much of the dominance of home value it removes.

### The measures describe more than one kind of deprivation

The correlation matrix behind Table 2 has 3 eigenvalues above the average, the usual criterion for the number of factors worth keeping, and the first 3 principal components carry 66.0% (95% CI 65.3% to 66.9%) of the variance of the measures, the first alone 39.6% (95% CI 38.4% to 41.1%).

A factor model describes the correlations among many measures with a few unobserved factors, and a measure's loading is its correlation with a factor. A maximum-likelihood factor model of the same population-weighted correlation matrix, with 3 factors and a varimax rotation that turns the factors so that each measure loads mainly on one, sorts the measures into 3 groups (Table 6).

The first, housing money, gathers home value, mortgage cost, rent, family income and white-collar work, and holds 24.7% of the variance of the measures. The second, economic hardship, gathers the poverty rate, the near-poverty share, income disparity, unemployment and single parenthood, with homeownership on the other side, and holds 21.5%. The third gathers crowding and the share of adults who left school before ninth grade, which correlate +0.64 (95% CI 0.57 to 0.70) with each other, and holds 11.9%.

Table 6. The measures sort into housing money, economic hardship, and crowding with little schooling, which together hold 58.1% of their variance

Kolak and colleagues, using a different set of tract measures, also needed several principal components and named the first socioeconomic advantage.[9] A single ranking folds these factors into one number. In raw units the number follows the housing-money factor, as the correlation with home value shows.

### Standardized rankings track mobility more closely and life expectancy about as closely

Two outcomes that no ranking used as an input test the recipes. One is life expectancy at birth from USALEEP, for deaths in 2010 to 2015 in 71,874 tracts. The other is the mean adult household income rank of children raised in the tract by parents at the 25th income percentile, from the Opportunity Atlas, for children born in 1978 to 1983 in 75,667 tracts. Table 7 shows the rank correlation of eight rankings with each outcome. Every ranking runs from least to most deprived, so every correlation is negative, and the figure plots absolute values. Comparisons below that rest on point estimates say so, and the others use intervals from paired draws of the same counties.

For life expectancy the published ADI correlates 0.615 (95% CI 0.594 to 0.633) and the standardized replica 0.619 (95% CI 0.596 to 0.642). The difference, 0.005 (95% CI −0.015 to 0.027), has an interval that includes zero, and so does the difference between family income alone, at 0.609 (95% CI 0.587 to 0.630), and the ADI, −0.006 (95% CI −0.022 to 0.012).

The standardized replica exceeds family income alone by 0.010 (95% CI 0.002 to 0.020). The first principal component, which has the highest point estimate at 0.633 (95% CI 0.613 to 0.653), exceeds family income alone by 0.025 (95% CI 0.019 to 0.031). Both intervals exclude zero, by small margins. The raw-units replica is at 0.620 (95% CI 0.601 to 0.638), and the poverty rate alone has the lowest point estimate, 0.508 (95% CI 0.484 to 0.529).

For upward mobility the order changes. The standardized replica correlates 0.582 (95% CI 0.560 to 0.603) and the published ADI 0.483 (95% CI 0.455 to 0.511), a difference of 0.100 (95% CI 0.076 to 0.123). The raw-units replica correlates 0.492 (95% CI 0.466 to 0.520) and home value alone 0.443 (95% CI 0.415 to 0.473).

Family income alone, at 0.564 (95% CI 0.543 to 0.586), correlates more strongly than the published ADI does, by 0.082 (95% CI 0.063 to 0.101). It correlates less strongly than the standardized replica, which exceeds it by 0.018 (95% CI 0.010 to 0.026), and than the first principal component, which exceeds it by 0.023 (95% CI 0.017 to 0.029).

For mobility, the poverty rate alone, at 0.504 (95% CI 0.486 to 0.522), does not differ detectably from the ADI, by 0.021 (95% CI −0.002 to 0.044). The rankings that standardize therefore edge out family income alone for both outcomes, and the published ADI does not for either.

Table 7. The published ADI tracks life expectancy about as closely as the standardized score does but upward mobility less closely, by 0.100 in correlation

### The book keeps the published index as its main ranking

Before comparing rankings across the book, the book fixed a rule: the standardized replica would replace the published ADI as the main ranking if it changed the main gradient of at least 4 of the 16 chapters that use the index. Each of those chapters recomputed its main gradient under the published ADI, the standardized replica, median household income and the share of adults with a bachelor's degree. The standardized replica changed 1, the national traffic gradient, so the published index stays (Chapter 9.1, How far one ranking goes (https://vishalsingh.org/zipcode/how-far-one-ranking-goes)). The difference between the two recipes shown in this chapter moves the rank of many tracts and leaves most of the book's gradients in the same class.

### What these data cannot show

Every result describes tracts and rankings of tracts. A tract's rank says little about how deprived any one resident is, because the data are averages for places (ecological inference), and a poor household in a tract with high home values takes the tract's rank.

The replica is a tract-level rebuild of a block-group index. It omits two measures, substitutes for some and does not impute missing values. Its ACS tract estimates carry sampling error that the staged file does not report, largest in small tracts and for medians that rest on few homes. A tract with many students living off campus can show a high poverty rate that reflects student households. The close match to the published rank, +0.991 (95% CI 0.990 to 0.992), shows that this formula reproduces the Atlas's ranks, and it does not show that the Atlas's code is the same. The frame leaves out tracts that lack a median, and the ADI ranks some of them. It also leaves out Puerto Rico, which the Atlas ranks with the states.

The data cannot say which recipe is right. A ranking dominated by home value measures the cost of housing, a property of a market, and some uses of a deprivation index may want that. The correlations with life expectancy and mobility are one test, and the two outcomes disagree about the recipes. Both outcomes are older than the ACS measures, deaths in 2010 to 2015 and children born in 1978 to 1983, and both are modeled estimates placed on 2020 tracts by interpolation from 2010 tracts. The SVI theme and the poverty rate are not Atlas products.

A factor model describes the measures it is given. Other measures, such as violence, isolation or air quality, would give other factors, and the count of factors rests on a common cutoff. The chapter does not test the state ADI that Medicare now uses, and it says nothing about how payments change under any formula.

## Tables

### Table 1. The Atlas score sums 17 measures with fixed coefficients, and 4 of them are dollar amounts

The measures in Singh's deprivation index, the unit each enters the sum in, Singh's factor score coefficient, and whether this chapter rebuilds it

| No. | Measure | Unit in the sum | Coefficient | Rebuilt here |
| ---: | --- | --- | ---: | --- |
| 1 | Adults 25 and older with less than 9 years of education | Proportion | +0.0849 | Yes, as adults who stopped at 8th grade |
| 2 | Adults 25 and older with at least a high school diploma | Proportion | −0.0970 | No: not in the staged ACS files; the SVI 2022 file has it for ACS 2018 to 2022 (sensitivity check) |
| 3 | Employed people 16 and older in white-collar occupations | Proportion | −0.0874 | Yes, as management, business, science and arts occupations |
| 4 | Median family income | Dollars | −0.0977 | Yes |
| 5 | Income disparity: log of 100 times households under $10,000 over households of $50,000 or more | Log ratio | +0.0936 | Yes, with a continuity constant |
| 6 | Median home value | Dollars | −0.0688 | Yes |
| 7 | Median gross rent | Dollars | −0.0781 | Yes |
| 8 | Median monthly mortgage | Dollars | −0.0770 | Yes, as monthly owner costs of homes with a mortgage |
| 9 | Owner-occupied housing units | Proportion | −0.0615 | Yes |
| 10 | Civilian labor force 16 and older that is unemployed | Proportion | +0.0806 | Yes |
| 11 | Families below the poverty level | Proportion | +0.0977 | Yes, as persons below the poverty line |
| 12 | Persons below 150 percent of the poverty threshold | Proportion | +0.1037 | Yes |
| 13 | Single-parent households with children under 18 | Proportion | +0.0719 | Yes, as single-parent households |
| 14 | Households without a motor vehicle | Proportion | +0.0694 | Yes |
| 15 | Households without a telephone (the Atlas now counts households without internet) | Proportion | +0.0877 | No: not in the staged ACS files; the SVI 2022 file has households without internet for ACS 2018 to 2022 (sensitivity check) |
| 16 | Occupied housing units without complete plumbing | Proportion | +0.0510 | Yes |
| 17 | Households with more than one person per room | Proportion | +0.0556 | Yes |

Note: The Atlas applies Singh's coefficients to proportions between zero and one, to dollars, and to the log ratio, and ranks the sums; it does not publish the sums. Since the 2020 release it counts households without internet where Singh counted households without a telephone.

Note: Singh estimated the coefficients as factor scores in 1990 census tract data; they are not re-estimated here.

Source: Singh (2003) factor score coefficients for 1990 census tracts, as printed in Petterson (2023); Neighborhood Atlas change log, 2022; published coefficients, not weighted

Data: https://vishalsingh.org/zipcode/report/anatomy-of-deprivation/figures/adi_measures.json

### Table 2. Home value correlates −0.31 with the poverty rate across tracts, a weaker link than family income has at −0.61

Population-weighted Pearson correlations among the 15 rebuilt measures in their own units, 75,717 tracts, ACS 2019 to 2023

| conf_high | conf_low | measure | n | other | r |
| ---: | ---: | --- | ---: | --- | ---: |
| n/a | n/a | Family income | n/a | Family income | 1 |
| 0.729 | 0.637 | Family income | 75,717 | Home value | 0.682 |
| 0.666 | 0.609 | Family income | 75,717 | Gross rent | 0.638 |
| 0.781 | 0.696 | Family income | 75,717 | Mortgage cost | 0.739 |
| 0.796 | 0.774 | Family income | 75,717 | White-collar jobs | 0.786 |
| 0.369 | 0.284 | Family income | 75,717 | Homeownership | 0.326 |
| −0.595 | −0.618 | Family income | 75,717 | Poverty | −0.606 |
| −0.696 | −0.715 | Family income | 75,717 | Below 150% of poverty | −0.705 |
| −0.425 | −0.460 | Family income | 75,717 | Income disparity | −0.442 |
| −0.268 | −0.304 | Family income | 75,717 | Unemployment | −0.286 |
| −0.477 | −0.504 | Family income | 75,717 | Single parents | −0.491 |
| −0.034 | −0.262 | Family income | 75,717 | No vehicle | −0.145 |
| −0.217 | −0.283 | Family income | 75,717 | Crowding | −0.252 |
| −0.269 | −0.297 | Family income | 75,717 | No plumbing | −0.282 |
| −0.345 | −0.384 | Family income | 75,717 | Under 9 years of school | −0.365 |
| 0.729 | 0.637 | Home value | 75,717 | Family income | 0.682 |
| n/a | n/a | Home value | n/a | Home value | 1 |
| 0.740 | 0.684 | Home value | 75,717 | Gross rent | 0.713 |
| 0.870 | 0.847 | Home value | 75,717 | Mortgage cost | 0.858 |
| 0.593 | 0.518 | Home value | 75,717 | White-collar jobs | 0.553 |
| 0.036 | −0.120 | Home value | 75,717 | Homeownership | −0.050 |
| −0.271 | −0.358 | Home value | 75,717 | Poverty | −0.313 |
| −0.339 | −0.429 | Home value | 75,717 | Below 150% of poverty | −0.381 |
| −0.217 | −0.295 | Home value | 75,717 | Income disparity | −0.252 |
| −0.036 | −0.134 | Home value | 75,717 | Unemployment | −0.084 |
| −0.254 | −0.297 | Home value | 75,717 | Single parents | −0.274 |
| 0.262 | −0.075 | Home value | 75,717 | No vehicle | 0.110 |
| 0.153 | 0.018 | Home value | 75,717 | Crowding | 0.100 |
| −0.247 | −0.281 | Home value | 75,717 | No plumbing | −0.262 |
| −0.011 | −0.145 | Home value | 75,717 | Under 9 years of school | −0.071 |
| 0.666 | 0.609 | Gross rent | 75,717 | Family income | 0.638 |
| 0.740 | 0.684 | Gross rent | 75,717 | Home value | 0.713 |
| n/a | n/a | Gross rent | n/a | Gross rent | 1 |
| 0.758 | 0.706 | Gross rent | 75,717 | Mortgage cost | 0.734 |
| 0.551 | 0.490 | Gross rent | 75,717 | White-collar jobs | 0.522 |
| 0.113 | 0.022 | Gross rent | 75,717 | Homeownership | 0.064 |
| −0.373 | −0.408 | Gross rent | 75,717 | Poverty | −0.391 |
| −0.434 | −0.472 | Gross rent | 75,717 | Below 150% of poverty | −0.453 |
| −0.341 | −0.383 | Gross rent | 75,717 | Income disparity | −0.361 |
| −0.061 | −0.117 | Gross rent | 75,717 | Unemployment | −0.089 |
| −0.196 | −0.236 | Gross rent | 75,717 | Single parents | −0.217 |
| 0.056 | −0.165 | Gross rent | 75,717 | No vehicle | −0.040 |
| 0.090 | 0.012 | Gross rent | 75,717 | Crowding | 0.057 |
| −0.317 | −0.341 | Gross rent | 75,717 | No plumbing | −0.328 |
| −0.059 | −0.131 | Gross rent | 75,717 | Under 9 years of school | −0.092 |
| 0.781 | 0.696 | Mortgage cost | 75,717 | Family income | 0.739 |
| 0.870 | 0.847 | Mortgage cost | 75,717 | Home value | 0.858 |
| 0.758 | 0.706 | Mortgage cost | 75,717 | Gross rent | 0.734 |
| n/a | n/a | Mortgage cost | n/a | Mortgage cost | 1 |
| 0.648 | 0.571 | Mortgage cost | 75,717 | White-collar jobs | 0.609 |
| 0.052 | −0.091 | Mortgage cost | 75,717 | Homeownership | −0.024 |
| −0.337 | −0.415 | Mortgage cost | 75,717 | Poverty | −0.376 |
| −0.413 | −0.496 | Mortgage cost | 75,717 | Below 150% of poverty | −0.455 |
| −0.274 | −0.341 | Mortgage cost | 75,717 | Income disparity | −0.307 |
| −0.056 | −0.143 | Mortgage cost | 75,717 | Unemployment | −0.100 |
| −0.270 | −0.317 | Mortgage cost | 75,717 | Single parents | −0.294 |
| 0.215 | −0.054 | Mortgage cost | 75,717 | No vehicle | 0.099 |
| 0.143 | 0.009 | Mortgage cost | 75,717 | Crowding | 0.087 |
| −0.292 | −0.320 | Mortgage cost | 75,717 | No plumbing | −0.305 |
| −0.023 | −0.149 | Mortgage cost | 75,717 | Under 9 years of school | −0.082 |
| 0.796 | 0.774 | White-collar jobs | 75,717 | Family income | 0.786 |
| 0.593 | 0.518 | White-collar jobs | 75,717 | Home value | 0.553 |
| 0.551 | 0.490 | White-collar jobs | 75,717 | Gross rent | 0.522 |
| 0.648 | 0.571 | White-collar jobs | 75,717 | Mortgage cost | 0.609 |
| n/a | n/a | White-collar jobs | n/a | White-collar jobs | 1 |
| 0.224 | 0.143 | White-collar jobs | 75,717 | Homeownership | 0.184 |
| −0.485 | −0.517 | White-collar jobs | 75,717 | Poverty | −0.501 |
| −0.590 | −0.618 | White-collar jobs | 75,717 | Below 150% of poverty | −0.605 |
| −0.317 | −0.357 | White-collar jobs | 75,717 | Income disparity | −0.337 |
| −0.250 | −0.286 | White-collar jobs | 75,717 | Unemployment | −0.268 |
| −0.440 | −0.481 | White-collar jobs | 75,717 | Single parents | −0.461 |
| 0.033 | −0.184 | White-collar jobs | 75,717 | No vehicle | −0.069 |
| −0.304 | −0.384 | White-collar jobs | 75,717 | Crowding | −0.346 |
| −0.249 | −0.282 | White-collar jobs | 75,717 | No plumbing | −0.265 |
| −0.463 | −0.516 | White-collar jobs | 75,717 | Under 9 years of school | −0.490 |
| 0.369 | 0.284 | Homeownership | 75,717 | Family income | 0.326 |
| 0.036 | −0.120 | Homeownership | 75,717 | Home value | −0.050 |
| 0.113 | 0.022 | Homeownership | 75,717 | Gross rent | 0.064 |
| 0.052 | −0.091 | Homeownership | 75,717 | Mortgage cost | −0.024 |
| 0.224 | 0.143 | Homeownership | 75,717 | White-collar jobs | 0.184 |
| n/a | n/a | Homeownership | n/a | Homeownership | 1 |
| −0.488 | −0.523 | Homeownership | 75,717 | Poverty | −0.505 |
| −0.496 | −0.537 | Homeownership | 75,717 | Below 150% of poverty | −0.516 |
| −0.345 | −0.375 | Homeownership | 75,717 | Income disparity | −0.359 |
| −0.243 | −0.285 | Homeownership | 75,717 | Unemployment | −0.264 |
| −0.323 | −0.376 | Homeownership | 75,717 | Single parents | −0.349 |
| −0.516 | −0.575 | Homeownership | 75,717 | No vehicle | −0.542 |
| −0.349 | −0.433 | Homeownership | 75,717 | Crowding | −0.394 |
| 0.037 | −0.005 | Homeownership | 75,717 | No plumbing | 0.017 |
| −0.265 | −0.349 | Homeownership | 75,717 | Under 9 years of school | −0.306 |
| −0.595 | −0.618 | Poverty | 75,717 | Family income | −0.606 |
| −0.271 | −0.358 | Poverty | 75,717 | Home value | −0.313 |
| −0.373 | −0.408 | Poverty | 75,717 | Gross rent | −0.391 |
| −0.337 | −0.415 | Poverty | 75,717 | Mortgage cost | −0.376 |
| −0.485 | −0.517 | Poverty | 75,717 | White-collar jobs | −0.501 |
| −0.488 | −0.523 | Poverty | 75,717 | Homeownership | −0.505 |
| n/a | n/a | Poverty | n/a | Poverty | 1 |
| 0.914 | 0.907 | Poverty | 75,717 | Below 150% of poverty | 0.911 |
| 0.593 | 0.571 | Poverty | 75,717 | Income disparity | 0.582 |
| 0.431 | 0.392 | Poverty | 75,717 | Unemployment | 0.412 |
| 0.500 | 0.469 | Poverty | 75,717 | Single parents | 0.486 |
| 0.472 | 0.322 | Poverty | 75,717 | No vehicle | 0.381 |
| 0.331 | 0.266 | Poverty | 75,717 | Crowding | 0.297 |
| 0.323 | 0.279 | Poverty | 75,717 | No plumbing | 0.301 |
| 0.397 | 0.345 | Poverty | 75,717 | Under 9 years of school | 0.371 |
| −0.696 | −0.715 | Below 150% of poverty | 75,717 | Family income | −0.705 |
| −0.339 | −0.429 | Below 150% of poverty | 75,717 | Home value | −0.381 |
| −0.434 | −0.472 | Below 150% of poverty | 75,717 | Gross rent | −0.453 |
| −0.413 | −0.496 | Below 150% of poverty | 75,717 | Mortgage cost | −0.455 |
| −0.590 | −0.618 | Below 150% of poverty | 75,717 | White-collar jobs | −0.605 |
| −0.496 | −0.537 | Below 150% of poverty | 75,717 | Homeownership | −0.516 |
| 0.914 | 0.907 | Below 150% of poverty | 75,717 | Poverty | 0.911 |
| n/a | n/a | Below 150% of poverty | n/a | Below 150% of poverty | 1 |
| 0.583 | 0.557 | Below 150% of poverty | 75,717 | Income disparity | 0.570 |
| 0.434 | 0.396 | Below 150% of poverty | 75,717 | Unemployment | 0.415 |
| 0.541 | 0.511 | Below 150% of poverty | 75,717 | Single parents | 0.527 |
| 0.461 | 0.295 | Below 150% of poverty | 75,717 | No vehicle | 0.363 |
| 0.381 | 0.315 | Below 150% of poverty | 75,717 | Crowding | 0.348 |
| 0.343 | 0.301 | Below 150% of poverty | 75,717 | No plumbing | 0.322 |
| 0.467 | 0.417 | Below 150% of poverty | 75,717 | Under 9 years of school | 0.443 |
| −0.425 | −0.460 | Income disparity | 75,717 | Family income | −0.442 |
| −0.217 | −0.295 | Income disparity | 75,717 | Home value | −0.252 |
| −0.341 | −0.383 | Income disparity | 75,717 | Gross rent | −0.361 |
| −0.274 | −0.341 | Income disparity | 75,717 | Mortgage cost | −0.307 |
| −0.317 | −0.357 | Income disparity | 75,717 | White-collar jobs | −0.337 |
| −0.345 | −0.375 | Income disparity | 75,717 | Homeownership | −0.359 |
| 0.593 | 0.571 | Income disparity | 75,717 | Poverty | 0.582 |
| 0.583 | 0.557 | Income disparity | 75,717 | Below 150% of poverty | 0.570 |
| n/a | n/a | Income disparity | n/a | Income disparity | 1 |
| 0.280 | 0.245 | Income disparity | 75,717 | Unemployment | 0.263 |
| 0.270 | 0.238 | Income disparity | 75,717 | Single parents | 0.254 |
| 0.361 | 0.259 | Income disparity | 75,717 | No vehicle | 0.295 |
| 0.127 | 0.087 | Income disparity | 75,717 | Crowding | 0.107 |
| 0.246 | 0.216 | Income disparity | 75,717 | No plumbing | 0.231 |
| 0.194 | 0.157 | Income disparity | 75,717 | Under 9 years of school | 0.175 |
| −0.268 | −0.304 | Unemployment | 75,717 | Family income | −0.286 |
| −0.036 | −0.134 | Unemployment | 75,717 | Home value | −0.084 |
| −0.061 | −0.117 | Unemployment | 75,717 | Gross rent | −0.089 |
| −0.056 | −0.143 | Unemployment | 75,717 | Mortgage cost | −0.100 |
| −0.250 | −0.286 | Unemployment | 75,717 | White-collar jobs | −0.268 |
| −0.243 | −0.285 | Unemployment | 75,717 | Homeownership | −0.264 |
| 0.431 | 0.392 | Unemployment | 75,717 | Poverty | 0.412 |
| 0.434 | 0.396 | Unemployment | 75,717 | Below 150% of poverty | 0.415 |
| 0.280 | 0.245 | Unemployment | 75,717 | Income disparity | 0.263 |
| n/a | n/a | Unemployment | n/a | Unemployment | 1 |
| 0.308 | 0.271 | Unemployment | 75,717 | Single parents | 0.290 |
| 0.296 | 0.219 | Unemployment | 75,717 | No vehicle | 0.251 |
| 0.234 | 0.181 | Unemployment | 75,717 | Crowding | 0.209 |
| 0.185 | 0.122 | Unemployment | 75,717 | No plumbing | 0.152 |
| 0.246 | 0.197 | Unemployment | 75,717 | Under 9 years of school | 0.222 |
| −0.477 | −0.504 | Single parents | 75,717 | Family income | −0.491 |
| −0.254 | −0.297 | Single parents | 75,717 | Home value | −0.274 |
| −0.196 | −0.236 | Single parents | 75,717 | Gross rent | −0.217 |
| −0.270 | −0.317 | Single parents | 75,717 | Mortgage cost | −0.294 |
| −0.440 | −0.481 | Single parents | 75,717 | White-collar jobs | −0.461 |
| −0.323 | −0.376 | Single parents | 75,717 | Homeownership | −0.349 |
| 0.500 | 0.469 | Single parents | 75,717 | Poverty | 0.486 |
| 0.541 | 0.511 | Single parents | 75,717 | Below 150% of poverty | 0.527 |
| 0.270 | 0.238 | Single parents | 75,717 | Income disparity | 0.254 |
| 0.308 | 0.271 | Single parents | 75,717 | Unemployment | 0.290 |
| n/a | n/a | Single parents | n/a | Single parents | 1 |
| 0.242 | 0.094 | Single parents | 75,717 | No vehicle | 0.157 |
| 0.330 | 0.260 | Single parents | 75,717 | Crowding | 0.296 |
| 0.128 | 0.087 | Single parents | 75,717 | No plumbing | 0.108 |
| 0.366 | 0.315 | Single parents | 75,717 | Under 9 years of school | 0.341 |
| −0.034 | −0.262 | No vehicle | 75,717 | Family income | −0.145 |
| 0.262 | −0.075 | No vehicle | 75,717 | Home value | 0.110 |
| 0.056 | −0.165 | No vehicle | 75,717 | Gross rent | −0.040 |
| 0.215 | −0.054 | No vehicle | 75,717 | Mortgage cost | 0.099 |
| 0.033 | −0.184 | No vehicle | 75,717 | White-collar jobs | −0.069 |
| −0.516 | −0.575 | No vehicle | 75,717 | Homeownership | −0.542 |
| 0.472 | 0.322 | No vehicle | 75,717 | Poverty | 0.381 |
| 0.461 | 0.295 | No vehicle | 75,717 | Below 150% of poverty | 0.363 |
| 0.361 | 0.259 | No vehicle | 75,717 | Income disparity | 0.295 |
| 0.296 | 0.219 | No vehicle | 75,717 | Unemployment | 0.251 |
| 0.242 | 0.094 | No vehicle | 75,717 | Single parents | 0.157 |
| n/a | n/a | No vehicle | n/a | No vehicle | 1 |
| 0.311 | 0.154 | No vehicle | 75,717 | Crowding | 0.227 |
| 0.164 | 0.057 | No vehicle | 75,717 | No plumbing | 0.101 |
| 0.266 | 0.161 | No vehicle | 75,717 | Under 9 years of school | 0.211 |
| −0.217 | −0.283 | Crowding | 75,717 | Family income | −0.252 |
| 0.153 | 0.018 | Crowding | 75,717 | Home value | 0.100 |
| 0.090 | 0.012 | Crowding | 75,717 | Gross rent | 0.057 |
| 0.143 | 0.009 | Crowding | 75,717 | Mortgage cost | 0.087 |
| −0.304 | −0.384 | Crowding | 75,717 | White-collar jobs | −0.346 |
| −0.349 | −0.433 | Crowding | 75,717 | Homeownership | −0.394 |
| 0.331 | 0.266 | Crowding | 75,717 | Poverty | 0.297 |
| 0.381 | 0.315 | Crowding | 75,717 | Below 150% of poverty | 0.348 |
| 0.127 | 0.087 | Crowding | 75,717 | Income disparity | 0.107 |
| 0.234 | 0.181 | Crowding | 75,717 | Unemployment | 0.209 |
| 0.330 | 0.260 | Crowding | 75,717 | Single parents | 0.296 |
| 0.311 | 0.154 | Crowding | 75,717 | No vehicle | 0.227 |
| n/a | n/a | Crowding | n/a | Crowding | 1 |
| 0.041 | −0.021 | Crowding | 75,717 | No plumbing | 0.007 |
| 0.698 | 0.571 | Crowding | 75,717 | Under 9 years of school | 0.636 |
| −0.269 | −0.297 | No plumbing | 75,717 | Family income | −0.282 |
| −0.247 | −0.281 | No plumbing | 75,717 | Home value | −0.262 |
| −0.317 | −0.341 | No plumbing | 75,717 | Gross rent | −0.328 |
| −0.292 | −0.320 | No plumbing | 75,717 | Mortgage cost | −0.305 |
| −0.249 | −0.282 | No plumbing | 75,717 | White-collar jobs | −0.265 |
| 0.037 | −0.005 | No plumbing | 75,717 | Homeownership | 0.017 |
| 0.323 | 0.279 | No plumbing | 75,717 | Poverty | 0.301 |
| 0.343 | 0.301 | No plumbing | 75,717 | Below 150% of poverty | 0.322 |
| 0.246 | 0.216 | No plumbing | 75,717 | Income disparity | 0.231 |
| 0.185 | 0.122 | No plumbing | 75,717 | Unemployment | 0.152 |
| 0.128 | 0.087 | No plumbing | 75,717 | Single parents | 0.108 |
| 0.164 | 0.057 | No plumbing | 75,717 | No vehicle | 0.101 |
| 0.041 | −0.021 | No plumbing | 75,717 | Crowding | 0.007 |
| n/a | n/a | No plumbing | n/a | No plumbing | 1 |
| 0.100 | 0.046 | No plumbing | 75,717 | Under 9 years of school | 0.071 |
| −0.345 | −0.384 | Under 9 years of school | 75,717 | Family income | −0.365 |
| −0.011 | −0.145 | Under 9 years of school | 75,717 | Home value | −0.071 |
| −0.059 | −0.131 | Under 9 years of school | 75,717 | Gross rent | −0.092 |
| −0.023 | −0.149 | Under 9 years of school | 75,717 | Mortgage cost | −0.082 |
| −0.463 | −0.516 | Under 9 years of school | 75,717 | White-collar jobs | −0.490 |
| −0.265 | −0.349 | Under 9 years of school | 75,717 | Homeownership | −0.306 |
| 0.397 | 0.345 | Under 9 years of school | 75,717 | Poverty | 0.371 |
| 0.467 | 0.417 | Under 9 years of school | 75,717 | Below 150% of poverty | 0.443 |
| 0.194 | 0.157 | Under 9 years of school | 75,717 | Income disparity | 0.175 |
| 0.246 | 0.197 | Under 9 years of school | 75,717 | Unemployment | 0.222 |
| 0.366 | 0.315 | Under 9 years of school | 75,717 | Single parents | 0.341 |
| 0.266 | 0.161 | Under 9 years of school | 75,717 | No vehicle | 0.211 |
| 0.698 | 0.571 | Under 9 years of school | 75,717 | Crowding | 0.636 |
| 0.100 | 0.046 | Under 9 years of school | 75,717 | No plumbing | 0.071 |
| n/a | n/a | Under 9 years of school | n/a | Under 9 years of school | 1 |

Note: Cells toward the teal end of the key are positive correlations and cells toward the brown end negative ones. The diagonal is one by construction and has no interval.

Note: The correlations use the measures as the ACS reports them: dollars for the four dollar measures, proportions for the rest and the log ratio for income disparity. The staged ACS file has no margins of error.

Source: ACS 2019-2023 five-year tract estimates (staged ACS file and Census API pull of housing measures); population-weighted (ACS tract population)

Data: https://vishalsingh.org/zipcode/report/anatomy-of-deprivation/figures/input_correlations.json

### Table 3. In raw units home value accounts for 74.3% of the variance of the score on its own and 85.4% with its covariance with income shared out, against 5.3% after standardization

Share of the population-weighted variance of the replica score allocated to each measure, same coefficients, measures in raw units and in standard deviation units, 75,717 tracts

#### Table 3.1. Raw units

In raw units home value and family income carry 99.2% of the variance of the score between them

Share of the variance of the score in raw units allocated to each measure, percent

| conf_high | conf_low | measure | n | share |
| ---: | ---: | --- | ---: | ---: |
| 86.99 | 83.08 | Home value | 75,717 | 85.40 |
| 16.52 | 12.64 | Family income | 75,717 | 14.23 |
| 0.245 | 0.213 | Mortgage cost | 75,717 | 0.226 |
| 0.164 | 0.143 | Gross rent | 75,717 | 0.152 |
| 0.000 | 0.000 | Income disparity | 75,717 | 0.000 |
| 0.000 | 0.000 | White-collar jobs | 75,717 | 0.000 |
| 0.000 | 0.000 | Below 150% of poverty | 75,717 | 0.000 |
| 0.000 | 0.000 | Poverty | 75,717 | 0.000 |
| 0.000 | 0.000 | Single parents | 75,717 | 0.000 |
| 0.000 | 0.000 | Under 9 years of school | 75,717 | 0.000 |
| 0.000 | 0.000 | No plumbing | 75,717 | 0.000 |
| 0.000 | 0.000 | Unemployment | 75,717 | 0.000 |
| 0.000 | 0.000 | Homeownership | 75,717 | 0.000 |
| 0.000 | 0.000 | Crowding | 75,717 | 0.000 |
| 0.000 | 0.000 | No vehicle | 75,717 | 0.000 |

#### Table 3.2. Standardized

After standardization no measure carries more than 12.4% of the variance of the score

Share of the variance of the score in standard deviation units allocated to each measure, percent

| conf_high | conf_low | measure | n | share |
| ---: | ---: | --- | ---: | ---: |
| 5.68 | 4.96 | Home value | 75,717 | 5.30 |
| 11.43 | 11.06 | Family income | 75,717 | 11.26 |
| 6.86 | 6.14 | Mortgage cost | 75,717 | 6.50 |
| 6.64 | 6.31 | Gross rent | 75,717 | 6.47 |
| 8.02 | 7.65 | Income disparity | 75,717 | 7.84 |
| 9.23 | 8.89 | White-collar jobs | 75,717 | 9.08 |
| 12.62 | 12.17 | Below 150% of poverty | 75,717 | 12.41 |
| 11.05 | 10.67 | Poverty | 75,717 | 10.88 |
| 5.98 | 5.66 | Single parents | 75,717 | 5.83 |
| 6.26 | 5.79 | Under 9 years of school | 75,717 | 6.03 |
| 2.73 | 2.50 | No plumbing | 75,717 | 2.62 |
| 5.30 | 4.91 | Unemployment | 75,717 | 5.11 |
| 4.34 | 3.97 | Homeownership | 75,717 | 4.16 |
| 3.22 | 2.76 | Crowding | 75,717 | 3.00 |
| 4.47 | 2.71 | No vehicle | 75,717 | 3.50 |

Note: Raw units: dollars for the four dollar measures, proportions between zero and one for the rest, and the log ratio for income disparity, as the Atlas applies Singh's coefficients. Standardized: each measure minus its population-weighted mean over its population-weighted standard deviation.

Note: The two measures that are not rebuilt (high school diploma and internet access) are proportions, so in raw units they would add little to the variance of the score.

Source: ACS 2019-2023 five-year tract estimates (staged ACS file and Census API pull of housing measures); Singh (2003) coefficients; population-weighted (ACS tract population)

Data: https://vishalsingh.org/zipcode/report/anatomy-of-deprivation/figures/variance_by_measure.json

### Table 4. The published ADI correlates +0.991 with its raw-units replica and +0.72 with the standardized version of the same measures

Population-weighted rank correlations among six rankings of 2020 tracts, measures from ACS 2019 to 2023

| conf_high | conf_low | n | other | r | ranking |
| ---: | ---: | ---: | --- | ---: | --- |
| n/a | n/a | n/a | Published ADI | 1 | Published ADI |
| 0.992 | 0.990 | 75,717 | Replica, raw units | 0.991 | Published ADI |
| 0.773 | 0.659 | 75,717 | Replica, standardized | 0.720 | Published ADI |
| 0.800 | 0.703 | 75,717 | First principal component | 0.755 | Published ADI |
| 0.556 | 0.406 | 75,717 | SVI socioeconomic theme | 0.485 | Published ADI |
| 0.587 | 0.486 | 75,717 | Poverty rate | 0.539 | Published ADI |
| 0.992 | 0.990 | 75,717 | Published ADI | 0.991 | Replica, raw units |
| n/a | n/a | n/a | Replica, raw units | 1 | Replica, raw units |
| 0.789 | 0.680 | 75,717 | Replica, standardized | 0.738 | Replica, raw units |
| 0.817 | 0.723 | 75,717 | First principal component | 0.773 | Replica, raw units |
| 0.575 | 0.429 | 75,717 | SVI socioeconomic theme | 0.506 | Replica, raw units |
| 0.595 | 0.496 | 75,717 | Poverty rate | 0.548 | Replica, raw units |
| 0.773 | 0.659 | 75,717 | Published ADI | 0.720 | Replica, standardized |
| 0.789 | 0.680 | 75,717 | Replica, raw units | 0.738 | Replica, standardized |
| n/a | n/a | n/a | Replica, standardized | 1 | Replica, standardized |
| 0.998 | 0.996 | 75,717 | First principal component | 0.997 | Replica, standardized |
| 0.866 | 0.852 | 75,717 | SVI socioeconomic theme | 0.859 | Replica, standardized |
| 0.859 | 0.843 | 75,717 | Poverty rate | 0.851 | Replica, standardized |
| 0.800 | 0.703 | 75,717 | Published ADI | 0.755 | First principal component |
| 0.817 | 0.723 | 75,717 | Replica, raw units | 0.773 | First principal component |
| 0.998 | 0.996 | 75,717 | Replica, standardized | 0.997 | First principal component |
| n/a | n/a | n/a | First principal component | 1 | First principal component |
| 0.850 | 0.832 | 75,717 | SVI socioeconomic theme | 0.841 | First principal component |
| 0.853 | 0.836 | 75,717 | Poverty rate | 0.844 | First principal component |
| 0.556 | 0.406 | 75,717 | Published ADI | 0.485 | SVI socioeconomic theme |
| 0.575 | 0.429 | 75,717 | Replica, raw units | 0.506 | SVI socioeconomic theme |
| 0.866 | 0.852 | 75,717 | Replica, standardized | 0.859 | SVI socioeconomic theme |
| 0.850 | 0.832 | 75,717 | First principal component | 0.841 | SVI socioeconomic theme |
| n/a | n/a | n/a | SVI socioeconomic theme | 1 | SVI socioeconomic theme |
| 0.769 | 0.753 | 75,717 | Poverty rate | 0.761 | SVI socioeconomic theme |
| 0.587 | 0.486 | 75,717 | Published ADI | 0.539 | Poverty rate |
| 0.595 | 0.496 | 75,717 | Replica, raw units | 0.548 | Poverty rate |
| 0.859 | 0.843 | 75,717 | Replica, standardized | 0.851 | Poverty rate |
| 0.853 | 0.836 | 75,717 | First principal component | 0.844 | Poverty rate |
| 0.769 | 0.753 | 75,717 | SVI socioeconomic theme | 0.761 | Poverty rate |
| n/a | n/a | n/a | Poverty rate | 1 | Poverty rate |

Note: Each cell uses the tracts that have both rankings; pairs with the SVI socioeconomic theme use the tracts that also have an SVI value, and each cell's tooltip gives its tract count.

Note: The ranks are percentiles among those tracts, and each cell is the population-weighted Pearson correlation of the two percentile ranks (Spearman's rank correlation would count each tract once); the published ADI is the Atlas's block-group ranks averaged over each tract.

Source: Neighborhood Atlas ADI 2023; ACS 2019-2023; CDC/ATSDR Social Vulnerability Index 2022; population-weighted (ACS tract population) rank correlations

Data: https://vishalsingh.org/zipcode/report/anatomy-of-deprivation/figures/index_family.json

### Table 5. The published ADI percentile minus the standardized percentile averages −16.5 in the highest-cost counties and +11.1 in the lowest-cost counties

Tract percentile on the published ADI against the percentile of the standardized replica score, by the cost of the tract's county, and the tracts with the largest gaps

#### Table 5.1. Tracts

At the same standardized percentile, tracts in the highest-cost counties have a lower published ADI than tracts in the lowest-cost counties

Published ADI against the standardized percentile, one dot per sampled tract, colored by the median home value of the tract's county

No inline rows: the values of Table 5.1 are in https://vishalsingh.org/zipcode/data/anatomy-of-deprivation/adi_vs_standardized.json.

#### Table 5.2. Largest gaps

12 of the 15 tracts with the largest gaps are in New York City

Tracts with a high poverty rate where the published ADI percentile is furthest below the standardized percentile

| County | Residents | Poverty rate (%) | Median home value ($) | Published ADI (percentile) | Standardized score (percentile) |
| --- | ---: | ---: | ---: | ---: | ---: |
| San Francisco, CA | 1,814 | 35.9 | 2,000,001 | 1 | 99.7 |
| Kings, NY | 5,261 | 41.2 | 1,767,400 | 2 | 98.1 |
| Kings, NY | 10,085 | 43.6 | 1,316,700 | 3 | 98.7 |
| Kings, NY | 1,861 | 48.4 | 1,010,000 | 6 | 99.8 |
| Los Angeles, CA | 2,168 | 35.9 | 1,137,500 | 6 | 99.7 |
| New York, NY | 4,755 | 32.0 | 2,000,001 | 1 | 94.3 |
| Bronx, NY | 5,460 | 46.4 | 1,044,300 | 7 | 98.7 |
| Queens, NY | 3,004 | 35.5 | 1,032,300 | 6 | 97.6 |
| Kings, NY | 4,887 | 45.1 | 1,138,600 | 5 | 96.6 |
| Kings, NY | 3,344 | 33.0 | 1,133,500 | 5 | 96.3 |
| Kings, NY | 2,902 | 31.8 | 1,179,700 | 5 | 95.9 |
| Kings, NY | 4,270 | 49.3 | 1,350,500 | 3 | 93.8 |
| Madison, ID | 12,195 | 63.3 | 2,000,001 | 1 | 91.6 |
| Kings, NY | 4,085 | 30.9 | 1,308,300 | 3 | 93.6 |
| New York, NY | 10,179 | 36.7 | 1,311,400 | 4 | 94.5 |

Note: Counties are grouped by their median home value into quartiles with equal numbers of residents; the middle two quartiles are drawn together in gray. Markers are population-weighted means of the published ADI within tenths of the standardized percentile for the lowest-cost and highest-cost groups, with 95% intervals from a bootstrap over counties; cells with few tracts are left out.

Note: The dots are a random sample of the frame's tracts drawn with a fixed seed (seed and share in the methods note); the means, intervals and the table use every tract.

Note: A tract's ACS medians rest on the homes and renters in the tract, which are few in some tracts, and the staged file has no margins of error; a college-town tract can show a high poverty rate because of student households. Each row illustrates what the formula does and is not an estimate of its tract's deprivation.

Source: Neighborhood Atlas ADI 2023; ACS 2019-2023; population-weighted (ACS tract population) binned means, modeled estimates

Data: https://vishalsingh.org/zipcode/report/anatomy-of-deprivation/figures/adi_vs_standardized.json

### Table 6. The measures sort into housing money, economic hardship, and crowding with little schooling, which together hold 58.1% of their variance

Varimax-rotated loadings of the 15 standardized measures on 3 maximum-likelihood factors, 75,717 tracts, ACS 2019 to 2023

#### Table 6.1. Housing money

Home value, mortgage cost, family income, rent and white-collar work load on the housing money factor

Loadings on the housing money factor, which holds 24.7% of the variance of the measures

| loading | measure |
| ---: | --- |
| 0.930 | Mortgage cost |
| 0.900 | Home value |
| 0.750 | Gross rent |
| 0.750 | Family income |
| 0.640 | White-collar jobs |
| −0.370 | Below 150% of poverty |
| −0.290 | No plumbing |
| −0.270 | Single parents |
| −0.270 | Poverty |
| −0.250 | Income disparity |
| 0.160 | No vehicle |
| 0.100 | Crowding |
| −0.090 | Homeownership |
| −0.080 | Under 9 years of school |
| −0.060 | Unemployment |

#### Table 6.2. Economic hardship

The poverty rate, near-poverty share and income disparity load on the economic hardship factor

Loadings on the economic hardship factor, which holds 21.5% of the variance of the measures

| loading | measure |
| ---: | --- |
| 0.900 | Poverty |
| 0.860 | Below 150% of poverty |
| 0.570 | Income disparity |
| −0.560 | Homeownership |
| 0.470 | No vehicle |
| 0.410 | Unemployment |
| −0.410 | Family income |
| 0.400 | Single parents |
| −0.280 | White-collar jobs |
| 0.260 | Under 9 years of school |
| 0.260 | No plumbing |
| 0.230 | Crowding |
| −0.220 | Gross rent |
| −0.140 | Mortgage cost |
| −0.080 | Home value |

#### Table 6.3. Crowding and schooling

Crowding and the share of adults with little schooling load on the third factor

Loadings on the crowding and schooling factor, which holds 11.9% of the variance of the measures

| loading | measure |
| ---: | --- |
| 0.750 | Crowding |
| 0.720 | Under 9 years of school |
| −0.490 | White-collar jobs |
| 0.330 | Single parents |
| −0.320 | Family income |
| −0.300 | Homeownership |
| 0.260 | Below 150% of poverty |
| 0.190 | Unemployment |
| 0.150 | Poverty |
| 0.150 | No vehicle |
| 0.060 | Gross rent |
| 0.050 | Home value |
| 0.050 | Mortgage cost |
| −0.040 | No plumbing |
| 0.020 | Income disparity |

Note: Maximum-likelihood factor analysis of the population-weighted correlation matrix of the measures (the matrix in the first correlation figure), 3 factors chosen by the Kaiser criterion, varimax rotation. The fit starts from seeded random loadings and the solution does not depend on the seed. Each factor is signed so its largest loading is positive.

Note: The factors describe these measures, not deprivation in general: other measures would give other factors.

Source: ACS 2019-2023 five-year tract estimates (staged ACS file and Census API pull of housing measures); population-weighted (ACS tract population) correlation matrix

Data: https://vishalsingh.org/zipcode/report/anatomy-of-deprivation/figures/three_factors.json

### Table 7. The published ADI tracks life expectancy about as closely as the standardized score does but upward mobility less closely, by 0.100 in correlation

Absolute population-weighted rank correlation of eight rankings with tract life expectancy at birth and with the adult income rank of children from low-income families, 71,874 and 75,667 tracts

| conf_high | conf_low | estimate | n | outcome | ranking |
| ---: | ---: | ---: | ---: | --- | --- |
| 0.653 | 0.613 | 0.633 | 71,874 | Life expectancy at birth | First principal component |
| 0.607 | 0.566 | 0.587 | 75,667 | Upward mobility | First principal component |
| 0.638 | 0.601 | 0.620 | 71,874 | Life expectancy at birth | Replica, raw units |
| 0.520 | 0.466 | 0.492 | 75,667 | Upward mobility | Replica, raw units |
| 0.642 | 0.596 | 0.619 | 71,874 | Life expectancy at birth | Replica, standardized |
| 0.603 | 0.560 | 0.582 | 75,667 | Upward mobility | Replica, standardized |
| 0.633 | 0.594 | 0.615 | 71,874 | Life expectancy at birth | Published ADI |
| 0.511 | 0.455 | 0.483 | 75,667 | Upward mobility | Published ADI |
| 0.630 | 0.587 | 0.609 | 71,874 | Life expectancy at birth | Family income |
| 0.586 | 0.543 | 0.564 | 75,667 | Upward mobility | Family income |
| 0.605 | 0.563 | 0.584 | 71,874 | Life expectancy at birth | Home value |
| 0.473 | 0.415 | 0.443 | 75,667 | Upward mobility | Home value |
| 0.553 | 0.485 | 0.518 | 71,874 | Life expectancy at birth | SVI socioeconomic theme |
| 0.571 | 0.519 | 0.545 | 75,667 | Upward mobility | SVI socioeconomic theme |
| 0.529 | 0.484 | 0.508 | 71,874 | Life expectancy at birth | Poverty rate |
| 0.522 | 0.486 | 0.504 | 75,667 | Upward mobility | Poverty rate |

Note: Both outcomes were modeled and neither was an input to any ranking. Each outcome uses the tracts that have it and every ranking, and institutional tracts are dropped from the life expectancy comparison. USALEEP covers deaths in 2010 to 2015 and its 2020-tract file is interpolated from 2010 tracts; the Opportunity Atlas covers children born in 1978 to 1983 and its 2020-tract file is interpolated from 2010 tracts.

Note: Family income and home value enter as negative values so that a higher rank means more deprived. The SVI theme and the poverty rate are not Atlas products.

Source: Neighborhood Atlas ADI 2023; ACS 2019-2023; CDC/ATSDR Social Vulnerability Index 2022; NCHS USALEEP 2010-2015; Opportunity Atlas 1978-1983 birth cohorts; population-weighted (ACS tract population) rank correlations, modeled estimates

Data: https://vishalsingh.org/zipcode/report/anatomy-of-deprivation/figures/index_vs_outcomes.json

### Table 8. 27% of residents live in a tract that holds the same tenth under the ADI and under median household income, and 63% are within one tenth

Residents by the tenth of their tract under the ADI (left) and under a second ranking (right), 1 = least deprived, tracts with every ranking, ACS 2019-2023

| Second ranking | Tenth by ADI | Tenth by the second ranking | Residents | Tracts |
| --- | --- | --- | ---: | ---: |
| income | 1 | 1 | 18,315,446 | 4186 |
| income | 1 | 2 | 5,637,323 | 1393 |
| income | 1 | 3 | 2,898,232 | 767 |
| income | 1 | 4 | 1,565,006 | 426 |
| income | 1 | 5 | 900,113 | 249 |
| income | 1 | 6 | 628,942 | 183 |
| income | 1 | 7 | 467,331 | 127 |
| income | 1 | 8 | 346,310 | 103 |
| income | 1 | 9 | 266,651 | 69 |
| income | 1 | 10 | 207,164 | 48 |
| income | 2 | 1 | 10,497,183 | 2141 |
| income | 2 | 2 | 8,183,210 | 1821 |
| income | 2 | 3 | 5,392,947 | 1255 |
| income | 2 | 4 | 3,261,379 | 788 |
| income | 2 | 5 | 2,209,812 | 557 |
| income | 2 | 6 | 1,351,790 | 346 |
| income | 2 | 7 | 923,588 | 248 |
| income | 2 | 8 | 756,431 | 188 |
| income | 2 | 9 | 578,848 | 147 |
| income | 2 | 10 | 281,047 | 72 |
| income | 3 | 1 | 5,060,411 | 997 |
| income | 3 | 2 | 10,515,141 | 2138 |
| income | 3 | 3 | 6,661,254 | 1467 |
| income | 3 | 4 | 4,544,760 | 1023 |
| income | 3 | 5 | 2,787,027 | 674 |
| income | 3 | 6 | 1,669,128 | 402 |
| income | 3 | 7 | 1,240,182 | 308 |
| income | 3 | 8 | 905,287 | 221 |
| income | 3 | 9 | 515,254 | 139 |
| income | 3 | 10 | 482,219 | 120 |
| income | 4 | 1 | 1,070,919 | 213 |
| income | 4 | 2 | 7,657,129 | 1564 |
| income | 4 | 3 | 8,386,020 | 1764 |
| income | 4 | 4 | 6,156,594 | 1359 |
| income | 4 | 5 | 4,459,269 | 1013 |
| income | 4 | 6 | 2,999,465 | 726 |
| income | 4 | 7 | 1,956,070 | 496 |
| income | 4 | 8 | 1,313,808 | 333 |
| income | 4 | 9 | 611,188 | 170 |
| income | 4 | 10 | 433,965 | 119 |
| income | 5 | 1 | 114,139 | 23 |
| income | 5 | 2 | 2,642,428 | 551 |
| income | 5 | 3 | 6,878,341 | 1473 |
| income | 5 | 4 | 7,293,341 | 1599 |
| income | 5 | 5 | 5,533,208 | 1265 |
| income | 5 | 6 | 3,855,944 | 905 |
| income | 5 | 7 | 2,942,717 | 705 |
| income | 5 | 8 | 2,002,193 | 491 |
| income | 5 | 9 | 1,202,613 | 311 |
| income | 5 | 10 | 603,800 | 162 |
| income | 6 | 2 | 340,298 | 84 |
| income | 6 | 3 | 2,996,424 | 683 |
| income | 6 | 4 | 6,569,679 | 1475 |
| income | 6 | 5 | 6,558,041 | 1527 |
| income | 6 | 6 | 5,668,073 | 1331 |
| income | 6 | 7 | 4,011,481 | 974 |
| income | 6 | 8 | 3,053,568 | 749 |
| income | 6 | 9 | 1,742,416 | 451 |
| income | 6 | 10 | 841,615 | 239 |
| income | 7 | 2 | 19,011 | 7 |
| income | 7 | 3 | 486,848 | 136 |
| income | 7 | 4 | 2,816,012 | 715 |
| income | 7 | 5 | 5,999,873 | 1462 |
| income | 7 | 6 | 6,735,729 | 1629 |
| income | 7 | 7 | 5,642,336 | 1386 |
| income | 7 | 8 | 4,435,676 | 1121 |
| income | 7 | 9 | 3,154,759 | 796 |
| income | 7 | 10 | 1,418,252 | 406 |
| income | 8 | 3 | 38,964 | 13 |
| income | 8 | 4 | 530,147 | 153 |
| income | 8 | 5 | 2,410,821 | 680 |
| income | 8 | 6 | 5,313,965 | 1406 |
| income | 8 | 7 | 6,505,213 | 1678 |
| income | 8 | 8 | 5,946,938 | 1565 |
| income | 8 | 9 | 4,752,221 | 1252 |
| income | 8 | 10 | 2,383,481 | 672 |
| income | 9 | 3 | 3,424 | 2 |
| income | 9 | 4 | 58,612 | 20 |
| income | 9 | 5 | 377,726 | 122 |
| income | 9 | 6 | 1,923,162 | 586 |
| income | 9 | 7 | 4,763,259 | 1382 |
| income | 9 | 8 | 6,991,709 | 2003 |
| income | 9 | 9 | 7,650,266 | 2130 |
| income | 9 | 10 | 4,742,772 | 1373 |
| income | 10 | 4 | 7,770 | 2 |
| income | 10 | 5 | 37,866 | 11 |
| income | 10 | 6 | 127,290 | 46 |
| income | 10 | 7 | 711,444 | 251 |
| income | 10 | 8 | 2,425,047 | 790 |
| income | 10 | 9 | 6,588,751 | 2095 |
| income | 10 | 10 | 12,809,305 | 4348 |
| home | 1 | 1 | 28,324,025 | 6924 |
| home | 1 | 2 | 2,899,010 | 625 |
| home | 1 | 3 | 9,483 | 2 |
| home | 2 | 1 | 2,389,563 | 589 |
| home | 2 | 2 | 26,303,664 | 5991 |
| home | 2 | 3 | 4,706,372 | 974 |
| home | 2 | 4 | 36,636 | 9 |
| home | 3 | 1 | 184,500 | 44 |
| home | 3 | 2 | 3,798,457 | 884 |
| home | 3 | 3 | 24,304,246 | 5323 |
| home | 3 | 4 | 6,013,780 | 1220 |
| home | 3 | 5 | 79,680 | 18 |
| home | 4 | 1 | 10,082 | 2 |
| home | 4 | 2 | 233,416 | 51 |
| home | 4 | 3 | 5,014,632 | 1188 |
| home | 4 | 4 | 22,522,022 | 5000 |
| home | 4 | 5 | 7,077,374 | 1472 |
| home | 4 | 6 | 186,901 | 44 |
| home | 5 | 1 | 4,438 | 2 |
| home | 5 | 2 | 18,536 | 6 |
| home | 5 | 3 | 269,665 | 59 |
| home | 5 | 4 | 5,325,964 | 1242 |
| home | 5 | 5 | 20,157,998 | 4547 |
| home | 5 | 6 | 7,034,217 | 1562 |
| home | 5 | 7 | 243,487 | 62 |
| home | 5 | 8 | 11,966 | 4 |
| home | 5 | 9 | 2,453 | 1 |
| home | 6 | 2 | 2,520 | 1 |
| home | 6 | 3 | 44,354 | 11 |
| home | 6 | 4 | 362,898 | 85 |
| home | 6 | 5 | 5,789,457 | 1408 |
| home | 6 | 6 | 18,759,565 | 4382 |
| home | 6 | 7 | 6,572,706 | 1560 |
| home | 6 | 8 | 237,089 | 63 |
| home | 6 | 9 | 4,329 | 1 |
| home | 6 | 10 | 8,677 | 2 |
| home | 7 | 3 | 2,205 | 1 |
| home | 7 | 4 | 32,975 | 10 |
| home | 7 | 5 | 449,675 | 113 |
| home | 7 | 6 | 5,882,774 | 1475 |
| home | 7 | 7 | 18,223,536 | 4480 |
| home | 7 | 8 | 5,916,739 | 1520 |
| home | 7 | 9 | 178,859 | 52 |
| home | 7 | 10 | 21,733 | 7 |
| home | 8 | 5 | 19,231 | 5 |
| home | 8 | 6 | 272,833 | 82 |
| home | 8 | 7 | 5,411,074 | 1415 |
| home | 8 | 8 | 17,868,157 | 4688 |
| home | 8 | 9 | 4,207,057 | 1199 |
| home | 8 | 10 | 103,398 | 30 |
| home | 9 | 5 | 7,597 | 1 |
| home | 9 | 6 | 6,229 | 2 |
| home | 9 | 7 | 130,452 | 41 |
| home | 9 | 8 | 4,616,359 | 1282 |
| home | 9 | 9 | 18,741,594 | 5328 |
| home | 9 | 10 | 3,008,699 | 964 |
| home | 10 | 8 | 26,522 | 10 |
| home | 10 | 9 | 3,221,803 | 978 |
| home | 10 | 10 | 19,459,148 | 6555 |

Views: Household income (to_ranking = income); Home value (to_ranking = home).

Note: Tenths are cut among the tracts with every ranking, each tract counted once, and the bands count the residents of those tracts. Median household income and median home value rank the lowest value as most deprived.

Source: Area Deprivation Index 2023 and ACS 2019-2023 tract tables (B19013, B25077); residents from ACS 2019-2023; full counts of residents

Data: https://vishalsingh.org/zipcode/report/anatomy-of-deprivation/figures/rank_flows.json

## All estimates

Every fact in facts.json (91), sorted by id.

| Fact id | Value | Unit | Estimand | Population | Period | Interval | n | Weights | Source | Note |
| --- | ---: | --- | --- | --- | --- | --- | ---: | --- | --- | --- |
| alt_changed_std | 1 | chapters | counted main gradients that change under the standardized index | the main gradients of the chapters counted by the ranking switch rule | as in each chapter |   |   | none (count of chapters) | Area Deprivation Index |   |
| alt_n_counted | 16 | chapters | chapters whose main gradient enters the count of the rule | the main gradients of the chapters counted by the ranking switch rule | as in each chapter |   |   | none (count of chapters) | Area Deprivation Index |   |
| alt_switch_changes | 4 | chapters | rule: changed main gradients under the standardized index that switch the ranking | the main gradients of the chapters counted by the ranking switch rule | as in each chapter |   |   | none (count of chapters) | Area Deprivation Index |   |
| cty_hv_cut_high | $450,562 | dollars | 75th percentile of county median home value, counties weighted by population in the frame | counties, weighted by the population of their tracts in the frame | ACS 2019 to 2023 five-year estimates, ADI 2023 release |   |   | ACS 2019-2023 tract population | American Community Survey |   |
| cty_hv_cut_low | $216,301 | dollars | 25th percentile of county median home value, counties weighted by population in the frame | counties, weighted by the population of their tracts in the frame | ACS 2019 to 2023 five-year estimates, ADI 2023 release |   |   | ACS 2019-2023 tract population | American Community Survey |   |
| cum3 | 66.0% | percent of the variance of the measures | share of the variance of the 15 standardized measures carried by the first principal components with an eigenvalue above 1, interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of the tracts in the frame (50 states and DC; a published ADI rank, a population and all rebuilt measures) | ACS 2019 to 2023 five-year estimates, ADI 2023 release | 95% CI 65.3% to 66.9% | 75,717 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| diff_kfr_income_adi | 0.082 | difference in absolute rank correlation | absolute correlation of negative median family income with the mean adult household income rank of children raised by parents at the 25th income percentile minus that of the published ADI, on the same tracts, interval from paired draws of a bootstrap over counties (Connecticut planning regions stand in for counties) | tracts with an Opportunity Atlas rank for children from low-income families, in the frame | children born 1978 to 1983, household income measured in 2014 to 2015; rankings from ACS 2019 to 2023 | 95% CI 0.063 to 0.101 | 75,667 tracts | ACS 2019-2023 tract population | Opportunity Atlas |   |
| diff_kfr_pc1_inc | 0.023 | difference in absolute rank correlation | absolute correlation of the first principal component of the standardized measures with mean adult household income rank of children raised by parents at the 25th income percentile minus that of negative median family income, on the same tracts, interval from paired draws of a bootstrap over counties (Connecticut planning regions stand in for counties) | tracts with an Opportunity Atlas rank for children from low-income families, in the frame | children born 1978 to 1983, household income measured in 2014 to 2015; rankings from ACS 2019 to 2023 | 95% CI 0.017 to 0.029 | 75,667 tracts | ACS 2019-2023 tract population | Opportunity Atlas |   |
| diff_kfr_pov_adi | 0.021 | difference in absolute rank correlation | absolute correlation of the poverty rate with the mean adult household income rank of children raised by parents at the 25th income percentile minus that of the published ADI, on the same tracts, interval from paired draws of a bootstrap over counties (Connecticut planning regions stand in for counties) | tracts with an Opportunity Atlas rank for children from low-income families, in the frame | children born 1978 to 1983, household income measured in 2014 to 2015; rankings from ACS 2019 to 2023 | 95% CI −0.002 to 0.044 | 75,667 tracts | ACS 2019-2023 tract population | Opportunity Atlas |   |
| diff_kfr_std_adi | 0.100 | difference in absolute rank correlation | absolute correlation of the standardized replica with mean adult household income rank of children raised by parents at the 25th income percentile minus that of the published ADI, on the same tracts, interval from paired draws of a bootstrap over counties (Connecticut planning regions stand in for counties) | tracts with an Opportunity Atlas rank for children from low-income families, in the frame | children born 1978 to 1983, household income measured in 2014 to 2015; rankings from ACS 2019 to 2023 | 95% CI 0.076 to 0.123 | 75,667 tracts | ACS 2019-2023 tract population | Opportunity Atlas |   |
| diff_kfr_std_inc | 0.018 | difference in absolute rank correlation | absolute correlation of the standardized replica score with mean adult household income rank of children raised by parents at the 25th income percentile minus that of negative median family income, on the same tracts, interval from paired draws of a bootstrap over counties (Connecticut planning regions stand in for counties) | tracts with an Opportunity Atlas rank for children from low-income families, in the frame | children born 1978 to 1983, household income measured in 2014 to 2015; rankings from ACS 2019 to 2023 | 95% CI 0.010 to 0.026 | 75,667 tracts | ACS 2019-2023 tract population | Opportunity Atlas |   |
| diff_le_inc_adi | −0.006 | difference in absolute rank correlation | absolute correlation of negative median family income with life expectancy at birth minus that of the published ADI, on the same tracts, interval from paired draws of a bootstrap over counties (Connecticut planning regions stand in for counties) | tracts with a life expectancy estimate that are not institutional tracts, in the frame | life expectancy for deaths in 2010 to 2015; rankings from ACS 2019 to 2023 | 95% CI −0.022 to 0.012 | 71,874 tracts | ACS 2019-2023 tract population | USALEEP life expectancy |   |
| diff_le_pc1_inc | 0.025 | difference in absolute rank correlation | absolute correlation of the first principal component of the standardized measures with life expectancy at birth minus that of negative median family income, on the same tracts, interval from paired draws of a bootstrap over counties (Connecticut planning regions stand in for counties) | tracts with a life expectancy estimate that are not institutional tracts, in the frame | life expectancy for deaths in 2010 to 2015; rankings from ACS 2019 to 2023 | 95% CI 0.019 to 0.031 | 71,874 tracts | ACS 2019-2023 tract population | USALEEP life expectancy |   |
| diff_le_std_adi | 0.005 | difference in absolute rank correlation | absolute correlation of the standardized replica with life expectancy at birth minus that of the published ADI, on the same tracts, interval from paired draws of a bootstrap over counties (Connecticut planning regions stand in for counties) | tracts with a life expectancy estimate that are not institutional tracts, in the frame | life expectancy for deaths in 2010 to 2015; rankings from ACS 2019 to 2023 | 95% CI −0.015 to 0.027 | 71,874 tracts | ACS 2019-2023 tract population | USALEEP life expectancy |   |
| diff_le_std_inc | 0.010 | difference in absolute rank correlation | absolute correlation of the standardized replica score with life expectancy at birth minus that of negative median family income, on the same tracts, interval from paired draws of a bootstrap over counties (Connecticut planning regions stand in for counties) | tracts with a life expectancy estimate that are not institutional tracts, in the frame | life expectancy for deaths in 2010 to 2015; rankings from ACS 2019 to 2023 | 95% CI 0.002 to 0.020 | 71,874 tracts | ACS 2019-2023 tract population | USALEEP life expectancy |   |
| disp_q1 | +11.1 | percentile points, published ADI minus standardized replica | population-weighted mean of the published tract ADI minus the percentile of the standardized replica score among the tracts of the frame, interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of tracts in the lowest-cost quartile of residents' counties, by county median home value | ACS 2019 to 2023 five-year estimates, ADI 2023 release | 95% CI 10.5 to 11.6 | 22,023 tracts | ACS 2019-2023 tract population | Area Deprivation Index |   |
| disp_q4 | −16.5 | percentile points, published ADI minus standardized replica | population-weighted mean of the published tract ADI minus the percentile of the standardized replica score among the tracts of the frame, interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of tracts in the highest-cost quartile of residents' counties, by county median home value | ACS 2019 to 2023 five-year estimates, ADI 2023 release | 95% CI −21.3 to −11.5 | 17,638 tracts | ACS 2019-2023 tract population | Area Deprivation Index |   |
| evr1_std | 39.6% | percent of the variance of the measures | share of the variance of the standardized measures carried by the first principal component, interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of the tracts in the frame (50 states and DC; a published ADI rank, a population and all rebuilt measures) | ACS 2019 to 2023 five-year estimates, ADI 2023 release | 95% CI 38.4% to 41.1% | 75,717 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| fa_share_crowd | 11.9% | percent of the variance of the measures | sum of squared varimax loadings of factor 3 divided by the number of measures, from a maximum-likelihood factor model of the population-weighted correlation matrix; a description of the whole frame, so no sampling interval | residents of the tracts in the frame (50 states and DC; a published ADI rank, a population and all rebuilt measures) | ACS 2019 to 2023 five-year estimates, ADI 2023 release |   | 75,717 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| fa_share_hardship | 21.5% | percent of the variance of the measures | sum of squared varimax loadings of factor 2 divided by the number of measures, from a maximum-likelihood factor model of the population-weighted correlation matrix; a description of the whole frame, so no sampling interval | residents of the tracts in the frame (50 states and DC; a published ADI rank, a population and all rebuilt measures) | ACS 2019 to 2023 five-year estimates, ADI 2023 release |   | 75,717 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| fa_share_money | 24.7% | percent of the variance of the measures | sum of squared varimax loadings of factor 1 divided by the number of measures, from a maximum-likelihood factor model of the population-weighted correlation matrix; a description of the whole frame, so no sampling interval | residents of the tracts in the frame (50 states and DC; a published ADI rank, a population and all rebuilt measures) | ACS 2019 to 2023 five-year estimates, ADI 2023 release |   | 75,717 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| fa_total | 58.1% | percent of the variance of the measures | sum of the squared loadings of the three factors divided by the number of measures; a description of the whole frame, so no sampling interval | residents of the tracts in the frame (50 states and DC; a published ADI rank, a population and all rebuilt measures) | ACS 2019 to 2023 five-year estimates, ADI 2023 release |   | 75,717 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| flow_near_home | 99% | percent of residents | share of residents whose tract holds the same or an adjacent tenth under the ADI and under median home value (tenths of tracts) | residents of the tracts with all rankings of b2.rankings | ACS 2019 to 2023 five-year estimates, ADI 2023 release |   |   | ACS 2019-2023 tract population | American Community Survey |   |
| flow_near_income | 63% | percent of residents | share of residents whose tract holds the same or an adjacent tenth under the ADI and under median household income (tenths of tracts) | residents of the tracts with all rankings of b2.rankings | ACS 2019 to 2023 five-year estimates, ADI 2023 release |   |   | ACS 2019-2023 tract population | American Community Survey |   |
| flow_same_home | 70% | percent of residents | share of residents whose tract holds the same tenth under the ADI and under median home value (tenths of tracts) | residents of the tracts with all rankings of b2.rankings | ACS 2019 to 2023 five-year estimates, ADI 2023 release |   |   | ACS 2019-2023 tract population | American Community Survey |   |
| flow_same_income | 27% | percent of residents | share of residents whose tract holds the same tenth under the ADI and under median household income (tenths of tracts) | residents of the tracts with all rankings of b2.rankings | ACS 2019 to 2023 five-year estimates, ADI 2023 release |   |   | ACS 2019-2023 tract population | American Community Survey |   |
| hp_threshold | 30% | percent of persons below the poverty line | poverty rate at or above which a tract is called high-poverty; a definition | tracts called high-poverty in this chapter | ACS 2019 to 2023 five-year estimates, ADI 2023 release |   |   | none (definition) | American Community Survey |   |
| hv_inc_var_share | 99.2% | percent of the variance of the score | variance of the sum of the home value and family income terms (Singh's coefficients times the measures in their own units) as a share of the variance of the score, with no covariance with the other terms allocated to it; interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of the tracts in the frame (50 states and DC; a published ADI rank, a population and all rebuilt measures) | ACS 2019 to 2023 five-year estimates, ADI 2023 release | 95% CI 99.2% to 99.3% | 75,717 tracts | ACS 2019-2023 tract population | Area Deprivation Index |   |
| hv_own_share | 74.3% | percent of the variance of the score | variance of the home value term alone (Singh's coefficients times the measures in their own units) as a share of the variance of the score, with no covariance with the other terms allocated to it; interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of the tracts in the frame (50 states and DC; a published ADI rank, a population and all rebuilt measures) | ACS 2019 to 2023 five-year estimates, ADI 2023 release | 95% CI 70.5% to 77.0% | 75,717 tracts | ACS 2019-2023 tract population | Area Deprivation Index |   |
| hv_share | 85.4% | percent of the variance of the score | share of the population-weighted variance of the score (Singh's coefficients times the measures in their own units) allocated to median home value: the covariance of that measure's term with the score divided by the score's variance, so each covariance between two terms is shared equally between them and the shares add to the whole; interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of the tracts in the frame (50 states and DC; a published ADI rank, a population and all rebuilt measures) | ACS 2019 to 2023 five-year estimates, ADI 2023 release | 95% CI 83.1% to 87.0% | 75,717 tracts | ACS 2019-2023 tract population | Area Deprivation Index | An allocated share: the covariance of the home value and family income terms is shared between them. The first edition's fact of this id was home value's share of the variance of the raw inputs, a different quantity; hv_own_share gives the variance of the home value term alone. |
| hv_share_std | 5.3% | percent of the variance of the score | share of the population-weighted variance of the score (Singh's coefficients times the measures in standard deviation units) allocated to median home value: the covariance of that measure's term with the score divided by the score's variance, so each covariance between two terms is shared equally between them and the shares add to the whole; interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of the tracts in the frame (50 states and DC; a published ADI rank, a population and all rebuilt measures) | ACS 2019 to 2023 five-year estimates, ADI 2023 release | 95% CI 5.0% to 5.7% | 75,717 tracts | ACS 2019-2023 tract population | Area Deprivation Index |   |
| inc_share_raw | 14.2% | percent of the variance of the score | share of the population-weighted variance of the score (Singh's coefficients times the measures in their own units) allocated to median family income: the covariance of that measure's term with the score divided by the score's variance, so each covariance between two terms is shared equally between them and the shares add to the whole; interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of the tracts in the frame (50 states and DC; a published ADI rank, a population and all rebuilt measures) | ACS 2019 to 2023 five-year estimates, ADI 2023 release | 95% CI 12.6% to 16.5% | 75,717 tracts | ACS 2019-2023 tract population | Area Deprivation Index |   |
| lost_tenure_share | 90.0% | percent of tracts | share of those tracts that lack a median gross rent, median home value or median mortgage cost, which the ACS leaves out where a tract has almost no renters or almost no owners | tracts with a published ADI rank and a population above zero that lack at least one of the 15 measures | ACS 2019 to 2023 five-year estimates, ADI 2023 release |   | 7,067 tracts | none (full count) | American Community Survey |   |
| max_share_std | 12.4% | percent of the variance of the score | share of the population-weighted variance of the score (Singh's coefficients times the measures in standard deviation units) allocated to the measure with the largest share, share below 150 percent of the poverty line: the covariance of that measure's term with the score divided by the score's variance, so each covariance between two terms is shared equally between them and the shares add to the whole; interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of the tracts in the frame (50 states and DC; a published ADI rank, a population and all rebuilt measures) | ACS 2019 to 2023 five-year estimates, ADI 2023 release | 95% CI 12.2% to 12.6% | 75,717 tracts | ACS 2019-2023 tract population | Area Deprivation Index |   |
| min_stark_pop | 1,000 | residents | smallest tract population listed; a definition | tracts eligible for the table of the largest gaps | ACS 2019 to 2023 five-year estimates, ADI 2023 release |   |   | none (definition) | American Community Survey |   |
| n_big | 5,049 | tracts | count of eligible tracts | high-poverty tracts of the frame with at least the minimum population | ACS 2019 to 2023 five-year estimates, ADI 2023 release |   |   | none (full count) | American Community Survey |   |
| n_dollar | 4 | census measures | count of dollar-denominated measures; a definition | measures of the index that are dollar amounts (median family income, home value, gross rent and monthly mortgage cost) | 2023 release of the ADI (ACS 2019 to 2023) |   |   | none (definition) | Area Deprivation Index |   |
| n_hp | 5,100 | tracts | count of tracts | tracts of the frame with a poverty rate at or above the high-poverty threshold | ACS 2019 to 2023 five-year estimates, ADI 2023 release |   |   | none (full count) | American Community Survey |   |
| n_hp_all | 6,718 | tracts | count of tracts | tracts with a published ADI rank, a population and a poverty rate at or above the high-poverty threshold, whether or not they have all rebuilt measures | ACS 2019 to 2023 five-year estimates, ADI 2023 release |   |   | none (full count) | Area Deprivation Index |   |
| n_kaiser | 3 | factors | number of eigenvalues above 1 of the population-weighted correlation matrix of the 15 measures (the Kaiser criterion); the count is the same in every bootstrap replicate | residents of the tracts in the frame (50 states and DC; a published ADI rank, a population and all rebuilt measures) | ACS 2019 to 2023 five-year estimates, ADI 2023 release |   | 75,717 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| n_kfr | 75,667 | tracts | count of tracts in the comparison, the same for every ranking in it | tracts with an Opportunity Atlas rank for children from low-income families, in the frame | children born 1978 to 1983, household income measured in 2014 to 2015; rankings from ACS 2019 to 2023 |   |   | none (full count) | Opportunity Atlas |   |
| n_le | 71,874 | tracts | count of tracts in the comparison, the same for every ranking in it | tracts with a life expectancy estimate that are not institutional tracts, in the frame | life expectancy for deaths in 2010 to 2015; rankings from ACS 2019 to 2023 |   |   | none (full count) | USALEEP life expectancy |   |
| n_measures | 17 | census measures | count of measures in the index; a definition | measures in Singh's deprivation index as the Neighborhood Atlas applies it (since the 2020 release it counts households without internet where Singh counted households without a telephone) | 2023 release of the ADI (ACS 2019 to 2023) |   |   | none (definition) | Area Deprivation Index |   |
| n_not_rebuilt | 2 | census measures | count of measures not rebuilt; a definition | measures of the index not rebuilt here (adults with at least a high school diploma and households without internet or, in Singh's list, a telephone) | ACS 2019 to 2023 five-year estimates, ADI 2023 release |   |   | none (definition) | Area Deprivation Index |   |
| n_rebuilt | 15 | census measures | count of measures rebuilt from the staged ACS tract file and the Census API pull of housing measures; a definition | measures of the index rebuilt here for 2020 census tracts | ACS 2019 to 2023 five-year estimates, ADI 2023 release |   |   | none (definition) | American Community Survey |   |
| n_sample | 3,000 | tracts | size of a random sample of the frame's tracts drawn with a fixed seed; the statistics use every tract | tracts in the scatter plot | ACS 2019 to 2023 five-year estimates, ADI 2023 release |   | 75,717 tracts | none (sample size) | Area Deprivation Index |   |
| n_stark | 15 | tracts | number of rows in the table; a definition | tracts in the table of the largest gaps | ACS 2019 to 2023 five-year estimates, ADI 2023 release |   |   | none (definition) | Area Deprivation Index |   |
| n_stark_nyc | 12 | tracts | count of the listed tracts in the five boroughs of New York City (Kings, New York, Bronx, Queens and Richmond counties) | tracts in the table of the largest gaps | ACS 2019 to 2023 five-year estimates, ADI 2023 release |   |   | none (full count) | Area Deprivation Index |   |
| n_states | 50 | states | count of states; a definition | states of the frame, besides the District of Columbia | ACS 2019 to 2023 five-year estimates, ADI 2023 release |   |   | none (definition) | American Community Survey |   |
| n_topcoded_hv | 495 | tracts | count of tracts at the top code | tracts of the frame whose median home value is the ACS top code (2,000,001 dollars) | ACS 2019 to 2023 five-year estimates, ADI 2023 release |   |   | none (full count) | American Community Survey |   |
| n_tracts | 75,717 | tracts | count of tracts in the frame | 2020 census tracts with a published ADI rank, a population and all 15 rebuilt measures | ACS 2019 to 2023 five-year estimates, ADI 2023 release |   |   | none (full count) | American Community Survey |   |
| n_tracts_rank | 82,784 | tracts | count of tracts | 2020 census tracts with a published ADI rank and an ACS population above zero | ACS 2019 to 2023 five-year estimates, ADI 2023 release |   |   | none (full count) | Area Deprivation Index |   |
| other_share_raw | 0.0003% | percent of the variance of the score | share of the population-weighted variance of the score (Singh's coefficients times the measures in their own units) allocated to the measures that are not dollar amounts, together: the covariance of that measure's term with the score divided by the score's variance, so each covariance between two terms is shared equally between them and the shares add to the whole; interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of the tracts in the frame (50 states and DC; a published ADI rank, a population and all rebuilt measures) | ACS 2019 to 2023 five-year estimates, ADI 2023 release | 95% CI 0.0002% to 0.0003% | 75,717 tracts | ACS 2019-2023 tract population | Area Deprivation Index |   |
| pop_frame_share | 93.0% | percent of residents | share of those residents who live in tracts of the frame | residents of tracts with a published ADI rank and a population above zero | ACS 2019 to 2023 five-year estimates, ADI 2023 release |   |   | ACS 2019-2023 tract population (full count of tract populations) | American Community Survey |   |
| pop_hp | 17.2 million | residents | sum of ACS tract population | residents of the high-poverty tracts of the frame | ACS 2019 to 2023 five-year estimates, ADI 2023 release |   |   | none (full count) | American Community Survey |   |
| r_crowd_edu | +0.64 | correlation coefficient | population-weighted Pearson correlation between tract crowding and tract share of adults with under 9 years of schooling in their own units, interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of the tracts in the frame (50 states and DC; a published ADI rank, a population and all rebuilt measures) | ACS 2019 to 2023 five-year estimates, ADI 2023 release | 95% CI 0.57 to 0.70 | 75,717 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| r_hv_mortgage | +0.86 | correlation coefficient | population-weighted Pearson correlation between tract median home value and tract median monthly mortgage cost in their own units, interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of the tracts in the frame (50 states and DC; a published ADI rank, a population and all rebuilt measures) | ACS 2019 to 2023 five-year estimates, ADI 2023 release | 95% CI 0.85 to 0.87 | 75,717 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| r_hv_pov | −0.31 | correlation coefficient | population-weighted Pearson correlation between tract median home value and tract poverty rate in their own units, interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of the tracts in the frame (50 states and DC; a published ADI rank, a population and all rebuilt measures) | ACS 2019 to 2023 five-year estimates, ADI 2023 release | 95% CI −0.36 to −0.27 | 75,717 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| r_hv_rent | +0.71 | correlation coefficient | population-weighted Pearson correlation between tract median home value and tract median gross rent in their own units, interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of the tracts in the frame (50 states and DC; a published ADI rank, a population and all rebuilt measures) | ACS 2019 to 2023 five-year estimates, ADI 2023 release | 95% CI 0.68 to 0.74 | 75,717 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| r_inc_pov | −0.61 | correlation coefficient | population-weighted Pearson correlation between tract median family income and tract poverty rate in their own units, interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of the tracts in the frame (50 states and DC; a published ADI rank, a population and all rebuilt measures) | ACS 2019 to 2023 five-year estimates, ADI 2023 release | 95% CI −0.62 to −0.59 | 75,717 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| r_unstd_homevalue | −0.987 | rank correlation coefficient | population-weighted correlation of the percentile ranks of the raw-units replica score and of median home value, interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of the tracts in the frame (50 states and DC; a published ADI rank, a population and all rebuilt measures) | ACS 2019 to 2023 five-year estimates, ADI 2023 release | 95% CI −0.989 to −0.986 | 75,717 tracts | ACS 2019-2023 tract population | Area Deprivation Index |   |
| rank_cut | 50th | national percentile | percentile cutoff; a definition | rank at or below which a tract counts as among the less deprived | ACS 2019 to 2023 five-year estimates, ADI 2023 release |   |   | none (definition) | Area Deprivation Index |   |
| rho_adi_homevalue | −0.988 | rank correlation coefficient | population-weighted correlation of the percentile ranks of the published tract ADI and of median home value, interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of the tracts in the frame (50 states and DC; a published ADI rank, a population and all rebuilt measures) | ACS 2019 to 2023 five-year estimates, ADI 2023 release | 95% CI −0.989 to −0.986 | 75,717 tracts | ACS 2019-2023 tract population | Area Deprivation Index |   |
| rho_std_adi | +0.72 | rank correlation coefficient | population-weighted correlation of the percentile ranks of the published tract ADI and of the replica score in standard deviation units, interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of the tracts in the frame (50 states and DC; a published ADI rank, a population and all rebuilt measures) | ACS 2019 to 2023 five-year estimates, ADI 2023 release | 95% CI 0.66 to 0.77 | 75,717 tracts | ACS 2019-2023 tract population | Area Deprivation Index | Counting each tract once gives +0.73. |
| rho_std_adi_17 | +0.71 | rank correlation coefficient | population-weighted correlation of the percentile ranks of the published tract ADI and of the standardized replica score of all 17 measures, with adults with at least a high school diploma (coefficient -0.0970, from the SVI share without a diploma) and households without internet (the telephone measure's coefficient, 0.0877, an assumption because the Atlas does not publish one for internet) added; interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of the tracts in the frame, which all have the two SVI measures | SVI 2022 measures (ACS 2018 to 2022) added to the ACS 2019 to 2023 measures; ADI 2023 release | 95% CI 0.65 to 0.77 | 75,717 tracts | ACS 2019-2023 tract population | CDC/ATSDR Social Vulnerability Index |   |
| rho_std_homevalue | −0.66 | rank correlation coefficient | population-weighted correlation of the percentile ranks of the standardized replica score and of median home value, interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of the tracts in the frame (50 states and DC; a published ADI rank, a population and all rebuilt measures) | ACS 2019 to 2023 five-year estimates, ADI 2023 release | 95% CI −0.71 to −0.60 | 75,717 tracts | ACS 2019-2023 tract population | Area Deprivation Index |   |
| rho_unstd_adi | +0.991 | rank correlation coefficient | population-weighted correlation of the percentile ranks of the published tract ADI and of the replica score in raw units, interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of the tracts in the frame (50 states and DC; a published ADI rank, a population and all rebuilt measures) | ACS 2019 to 2023 five-year estimates, ADI 2023 release | 95% CI 0.990 to 0.992 | 75,717 tracts | ACS 2019-2023 tract population | Area Deprivation Index | Counting each tract once gives +0.992. |
| rho_unstd_adi_17 | +0.991 | rank correlation coefficient | population-weighted correlation of the percentile ranks of the published tract ADI and of the raw-units replica score of all 17 measures, with adults with at least a high school diploma (coefficient -0.0970, from the SVI share without a diploma) and households without internet (the telephone measure's coefficient, 0.0877, an assumption because the Atlas does not publish one for internet) added; interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of the tracts in the frame, which all have the two SVI measures | SVI 2022 measures (ACS 2018 to 2022) added to the ACS 2019 to 2023 measures; ADI 2023 release | 95% CI 0.990 to 0.992 | 75,717 tracts | ACS 2019-2023 tract population | CDC/ATSDR Social Vulnerability Index |   |
| rho_weights | +0.997 | rank correlation coefficient | population-weighted correlation of the percentile ranks of Singh's coefficients applied to standardized measures and of the first principal component of the same measures, interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of the tracts in the frame (50 states and DC; a published ADI rank, a population and all rebuilt measures) | ACS 2019 to 2023 five-year estimates, ADI 2023 release | 95% CI 0.996 to 0.998 | 75,717 tracts | ACS 2019-2023 tract population | Area Deprivation Index |   |
| rm_share_raw | 0.38% | percent of the variance of the score | share of the population-weighted variance of the score (Singh's coefficients times the measures in their own units) allocated to median gross rent and the monthly mortgage cost, together: the covariance of that measure's term with the score divided by the score's variance, so each covariance between two terms is shared equally between them and the shares add to the whole; interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of the tracts in the frame (50 states and DC; a published ADI rank, a population and all rebuilt measures) | ACS 2019 to 2023 five-year estimates, ADI 2023 release | 95% CI 0.36% to 0.41% | 75,717 tracts | ACS 2019-2023 tract population | Area Deprivation Index |   |
| sh_hp_lowadi | 16.1% | percent of residents | share of residents of high-poverty tracts who live in tracts at or below the 50th percentile on the published ADI (percentile among the tracts of the frame for the replica), interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of high-poverty tracts of the frame | ACS 2019 to 2023 five-year estimates, ADI 2023 release | 95% CI 11.3% to 21.4% | 5,100 tracts | ACS 2019-2023 tract population | Area Deprivation Index |   |
| sh_hp_lowadi_all | 20.6% | percent of residents | share of residents of high-poverty tracts who live in tracts at or below the 50th percentile on the published ADI, interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of high-poverty tracts with a published ADI rank, whether or not they have all rebuilt measures | ACS 2019 to 2023 five-year estimates, ADI 2023 release | 95% CI 14.1% to 27.9% | 6,718 tracts | ACS 2019-2023 tract population | Area Deprivation Index |   |
| sh_hp_lowrep | 15.8% | percent of residents | share of residents of high-poverty tracts who live in tracts at or below the 50th percentile on the replica score in raw units (percentile among the tracts of the frame for the replica), interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of high-poverty tracts of the frame | ACS 2019 to 2023 five-year estimates, ADI 2023 release | 95% CI 11.0% to 20.7% | 5,100 tracts | ACS 2019-2023 tract population | Area Deprivation Index |   |
| sh_hp_lowstd | 0.34% | percent of residents | share of residents of high-poverty tracts who live in tracts at or below the 50th percentile on the replica score in standard deviation units (percentile among the tracts of the frame for the replica), interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of high-poverty tracts of the frame | ACS 2019 to 2023 five-year estimates, ADI 2023 release | 95% CI 0.15% to 0.55% | 5,100 tracts | ACS 2019-2023 tract population | Area Deprivation Index |   |
| sh_hp_lowstd_17 | 0.61% | percent of residents | share of residents of high-poverty tracts who live in tracts at or below the 50th percentile of the standardized replica score of all 17 measures (percentile among the tracts of the frame), interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of high-poverty tracts of the frame | SVI 2022 measures (ACS 2018 to 2022) added to the ACS 2019 to 2023 measures; ADI 2023 release | 95% CI 0.32% to 0.93% | 5,100 tracts | ACS 2019-2023 tract population | CDC/ATSDR Social Vulnerability Index |   |
| tracts_all | 84,415 | tracts | count of tracts | 2020 census tracts of the 50 states and DC in the staged ADI file (Puerto Rico left out) | ACS 2019 to 2023 five-year estimates, ADI 2023 release |   |   | none (full count) | Area Deprivation Index |   |
| val_kfr_adi | 0.483 | absolute rank correlation coefficient | absolute value of the population-weighted correlation between the percentile rank of the published ADI (higher means more deprived) and the percentile rank of mean adult household income rank of children raised by parents at the 25th income percentile, interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | tracts with an Opportunity Atlas rank for children from low-income families, in the frame | children born 1978 to 1983, household income measured in 2014 to 2015; rankings from ACS 2019 to 2023 | 95% CI 0.455 to 0.511 | 75,667 tracts | ACS 2019-2023 tract population | Opportunity Atlas |   |
| val_kfr_home_value | 0.443 | absolute rank correlation coefficient | absolute value of the population-weighted correlation between the percentile rank of negative median home value (higher means more deprived) and the percentile rank of mean adult household income rank of children raised by parents at the 25th income percentile, interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | tracts with an Opportunity Atlas rank for children from low-income families, in the frame | children born 1978 to 1983, household income measured in 2014 to 2015; rankings from ACS 2019 to 2023 | 95% CI 0.415 to 0.473 | 75,667 tracts | ACS 2019-2023 tract population | Opportunity Atlas |   |
| val_kfr_idx_std | 0.582 | absolute rank correlation coefficient | absolute value of the population-weighted correlation between the percentile rank of the replica score in standard deviation units (higher means more deprived) and the percentile rank of mean adult household income rank of children raised by parents at the 25th income percentile, interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | tracts with an Opportunity Atlas rank for children from low-income families, in the frame | children born 1978 to 1983, household income measured in 2014 to 2015; rankings from ACS 2019 to 2023 | 95% CI 0.560 to 0.603 | 75,667 tracts | ACS 2019-2023 tract population | Opportunity Atlas |   |
| val_kfr_idx_unstd | 0.492 | absolute rank correlation coefficient | absolute value of the population-weighted correlation between the percentile rank of the replica score in raw units (higher means more deprived) and the percentile rank of mean adult household income rank of children raised by parents at the 25th income percentile, interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | tracts with an Opportunity Atlas rank for children from low-income families, in the frame | children born 1978 to 1983, household income measured in 2014 to 2015; rankings from ACS 2019 to 2023 | 95% CI 0.466 to 0.520 | 75,667 tracts | ACS 2019-2023 tract population | Opportunity Atlas |   |
| val_kfr_income | 0.564 | absolute rank correlation coefficient | absolute value of the population-weighted correlation between the percentile rank of negative median family income (higher means more deprived) and the percentile rank of mean adult household income rank of children raised by parents at the 25th income percentile, interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | tracts with an Opportunity Atlas rank for children from low-income families, in the frame | children born 1978 to 1983, household income measured in 2014 to 2015; rankings from ACS 2019 to 2023 | 95% CI 0.543 to 0.586 | 75,667 tracts | ACS 2019-2023 tract population | Opportunity Atlas |   |
| val_kfr_poverty | 0.504 | absolute rank correlation coefficient | absolute value of the population-weighted correlation between the percentile rank of the poverty rate (higher means more deprived) and the percentile rank of mean adult household income rank of children raised by parents at the 25th income percentile, interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | tracts with an Opportunity Atlas rank for children from low-income families, in the frame | children born 1978 to 1983, household income measured in 2014 to 2015; rankings from ACS 2019 to 2023 | 95% CI 0.486 to 0.522 | 75,667 tracts | ACS 2019-2023 tract population | Opportunity Atlas |   |
| val_le_adi | 0.615 | absolute rank correlation coefficient | absolute value of the population-weighted correlation between the percentile rank of the published ADI (higher means more deprived) and the percentile rank of life expectancy at birth, interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | tracts with a life expectancy estimate that are not institutional tracts, in the frame | life expectancy for deaths in 2010 to 2015; rankings from ACS 2019 to 2023 | 95% CI 0.594 to 0.633 | 71,874 tracts | ACS 2019-2023 tract population | USALEEP life expectancy |   |
| val_le_idx_pc1 | 0.633 | absolute rank correlation coefficient | absolute value of the population-weighted correlation between the percentile rank of the first principal component of the standardized measures (higher means more deprived) and the percentile rank of life expectancy at birth, interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | tracts with a life expectancy estimate that are not institutional tracts, in the frame | life expectancy for deaths in 2010 to 2015; rankings from ACS 2019 to 2023 | 95% CI 0.613 to 0.653 | 71,874 tracts | ACS 2019-2023 tract population | USALEEP life expectancy |   |
| val_le_idx_std | 0.619 | absolute rank correlation coefficient | absolute value of the population-weighted correlation between the percentile rank of the replica score in standard deviation units (higher means more deprived) and the percentile rank of life expectancy at birth, interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | tracts with a life expectancy estimate that are not institutional tracts, in the frame | life expectancy for deaths in 2010 to 2015; rankings from ACS 2019 to 2023 | 95% CI 0.596 to 0.642 | 71,874 tracts | ACS 2019-2023 tract population | USALEEP life expectancy |   |
| val_le_idx_unstd | 0.620 | absolute rank correlation coefficient | absolute value of the population-weighted correlation between the percentile rank of the replica score in raw units (higher means more deprived) and the percentile rank of life expectancy at birth, interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | tracts with a life expectancy estimate that are not institutional tracts, in the frame | life expectancy for deaths in 2010 to 2015; rankings from ACS 2019 to 2023 | 95% CI 0.601 to 0.638 | 71,874 tracts | ACS 2019-2023 tract population | USALEEP life expectancy |   |
| val_le_income | 0.609 | absolute rank correlation coefficient | absolute value of the population-weighted correlation between the percentile rank of negative median family income (higher means more deprived) and the percentile rank of life expectancy at birth, interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | tracts with a life expectancy estimate that are not institutional tracts, in the frame | life expectancy for deaths in 2010 to 2015; rankings from ACS 2019 to 2023 | 95% CI 0.587 to 0.630 | 71,874 tracts | ACS 2019-2023 tract population | USALEEP life expectancy |   |
| val_le_poverty | 0.508 | absolute rank correlation coefficient | absolute value of the population-weighted correlation between the percentile rank of the poverty rate (higher means more deprived) and the percentile rank of life expectancy at birth, interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | tracts with a life expectancy estimate that are not institutional tracts, in the frame | life expectancy for deaths in 2010 to 2015; rankings from ACS 2019 to 2023 | 95% CI 0.484 to 0.529 | 71,874 tracts | ACS 2019-2023 tract population | USALEEP life expectancy |   |
| worst_adi | 1 | national percentile | published ADI of one tract (mean of its block groups' ranks) | the tract in the first row of the table | ACS 2019 to 2023 five-year estimates, ADI 2023 release |   |   | none (single tract) | Area Deprivation Index |   |
| worst_poverty | 35.9% | percent of persons below the poverty line | poverty rate of one tract (ACS tract estimate, margin of error not in the staged file) | residents of the tract in the first row of the table | ACS 2019 to 2023 five-year estimates, ADI 2023 release |   |   | none (single tract) | American Community Survey |   |
| worst_std | 99.7 | percentile among the tracts of the frame | percentile of the standardized replica score of one tract | the tract in the first row of the table | ACS 2019 to 2023 five-year estimates, ADI 2023 release |   |   | none (single tract) | American Community Survey |   |

## Definitions

- Area Deprivation Index (ADI): A score that ranks neighborhoods by income, education, employment and housing conditions drawn from the American Community Survey. In this book a higher national percentile means a more deprived neighborhood.
- National percentile: A neighborhood's position in the national ordering, given as the share of all neighborhoods that fall below it. A higher percentile means a higher value than more of the rest of the country.
- Principal components: A method that combines many correlated measures into a few summary scores, ordered by how much of the variation among places each captures. The first component is the single weighted combination that captures the most.
- Life expectancy at birth: The average number of years a newborn would live if death rates at every age stayed as they were in the period measured. It summarizes the death rates of a period and is not a forecast for any child.
- USALEEP: The U.S. Small-area Life Expectancy Estimates Project, a National Center for Health Statistics release of life expectancy at birth for most census tracts. It covers deaths from 2010 to 2015 only and has not been updated.
- Ecological inference: Drawing conclusions about individuals from data about places or groups. A pattern across counties need not hold for the people inside them.
- Modeled estimate: A figure produced by a statistical model instead of a direct count or measurement, usually for small areas where direct data are thin, borrowing strength from surveys, neighboring areas or other data. Neighboring areas can share errors, so single extremes call for caution.

## Methods

This chapter rebuilds the Neighborhood Atlas Area Deprivation Index from ACS 2019-2023 tract estimates and compares the replica with the published rank, with standardized rankings and with a principal component, across 75,717 census tracts of the 50 states and DC. Fifteen of the Atlas's 17 measures are rebuilt, with substitutions for some, so the replica approximates the Atlas and is not its own file; the Puerto Rico tracts are left out. Life expectancy (deaths in 2010 to 2015) and upward mobility (children born 1978 to 1983) describe earlier periods than the ACS measures, and the ACS file carries no margins of error, so the only intervals come from resampling whole counties.

### Which tracts are in the frame

The frame is every 2020 census tract of the 50 states and DC with a published ADI rank (the Atlas's block-group ranks averaged over the tract's block groups without population weights), a population above zero in the ACS 2019-2023 tract file and all 15 rebuilt measures: 75,717 of the 84,415 tracts of the 50 states and DC in the staged ADI file (adi_atlas_tract2020_2023, the Atlas's 2023 release, version 4.0.1, whose staged documentation records ACS 2019 to 2023 as its source data).

The 981 Puerto Rico tracts are left out: the SVI and USALEEP have none, and the Atlas ranks them together with the states, so percentiles of the replicas are among the tracts of the 50 states and DC.

### How the measures are rebuilt

Fifteen of the Atlas's 17 measures are rebuilt from the staged ACS tract file (poverty, homeownership, unemployment and single-parent shares) and from a Census API pull of ACS tables (B25077, B25064, B25088, B19113, B19001, C17002, C24010, B25044, B25047, B25014 and B15003) archived under data/raw/acs/deprivation_components_2019_2023 with a manifest; b2.load has no loader for that pull, so the chapter registers the files with b2.load and reads them through load_staged.

Substitutions: persons below the poverty line for families, management, business, science and arts occupations for white-collar work, owner costs of homes with a mortgage for the monthly mortgage, adults who stopped at 8th grade for adults with under 9 years of education, and an income disparity of ln(100 x (households under $10,000 + 0.5) / (households of $50,000 or more + 0.5)), with the 0.5 added so that tracts with no household under $10,000 keep a value.

The two measures not rebuilt are adults with at least a high school diploma and households without internet. Both are proportions, and a sensitivity check adds them from the SVI 2022 file (ACS 2018 to 2022; the share without a diploma enters as one minus the share, with coefficient -0.0970, and the share without internet takes the telephone measure's coefficient, 0.0877, because the Atlas does not publish one for internet).

Proportions enter as numbers between zero and one, the convention of Petterson's Table 1 (0.41 for a block group with 41 percent below 150 percent of the poverty line); entered as percentages, home value would carry 85.4 percent and family income 14.2 percent of the variance of the raw-units score.

Explains Table 1 and Table 2.

### Outcomes and comparison indexes

Life expectancy is USALEEP for deaths in 2010 to 2015 on 2020 tracts (interpolated from 2010 tracts), institutional tracts dropped; mobility is the Opportunity Atlas rank of children of parents at the 25th income percentile (1978 to 1983 birth cohorts), interpolated to 2020 tracts.

The SVI socioeconomic theme is the 2022 CDC/ATSDR percentile rank (theme 1).

Explains Table 7.

### How scores, rankings and groups are formed

Recipes: the replica score is the sum of Singh's factor score coefficients (Singh 2003, as printed in Petterson 2023, Table 1) times the measures in their own units; the standardized score uses the same coefficients on measures minus their population-weighted means over their population-weighted standard deviations; the principal component is the first eigenvector of the population-weighted correlation matrix.

Rankings are percentiles among the tracts compared, and correlations of rankings are population-weighted Pearson correlations of those percentiles.

Shares of the score's variance are cov(c_k x_k, S) / var(S) for the measures' terms c_k x_k and the score S, computed from the weighted covariance matrix, so each covariance between two terms is shared equally between them; the own-variance shares are var(sum of terms) / var(S) with no covariance with other terms allocated.

Counties are grouped by the ACS 2019-2023 median home value of the county (the Census API county pull) into quartiles with equal numbers of residents of the frame. High-poverty tracts have a poverty rate of at least 30 percent, and the table of the largest gaps lists the 15 tracts with the lowest published-minus-standardized percentile among those with at least 1,000 residents.

Explains Table 3, Table 4 and Table 5.

### Weights and uncertainty

Weights are ACS 2019-2023 tract population.

Intervals are 95% percentile intervals from 2,000 draws of whole counties with replacement (seed 12345; Connecticut's planning regions stand in for counties, 3,126 clusters in the frame), one set of draws for all statistics; differences of two correlations on the same tracts use paired draws.

The ACS file carries no margins of error.

### Factor model and plotted sample

The factor model is a maximum-likelihood fit to the population-weighted correlation matrix (statsmodels Factor; its EM start draws random loadings, seeded with 12345, and a second seed gives the same loadings to four decimals) with a varimax rotation; it and the eigenvalue shares describe the whole frame, and the eigenvalue shares also have bootstrap intervals.

The scatter plot shows a random sample of 3,000 tracts drawn with the same seed; every statistic uses all tracts.

Explains Table 6.

## What these data cannot show

Every result describes tracts and rankings of tracts. A tract's rank says little about how deprived any one resident is, because the data are averages for places (ecological inference), and a poor household in a tract with high home values takes the tract's rank.

The replica is a tract-level rebuild of a block-group index. It omits two measures, substitutes for some and does not impute missing values. Its ACS tract estimates carry sampling error that the staged file does not report, largest in small tracts and for medians that rest on few homes. A tract with many students living off campus can show a high poverty rate that reflects student households. The close match to the published rank, +0.991 (95% CI 0.990 to 0.992), shows that this formula reproduces the Atlas's ranks, and it does not show that the Atlas's code is the same. The frame leaves out tracts that lack a median, and the ADI ranks some of them. It also leaves out Puerto Rico, which the Atlas ranks with the states.

The data cannot say which recipe is right. A ranking dominated by home value measures the cost of housing, a property of a market, and some uses of a deprivation index may want that. The correlations with life expectancy and mobility are one test, and the two outcomes disagree about the recipes. Both outcomes are older than the ACS measures, deaths in 2010 to 2015 and children born in 1978 to 1983, and both are modeled estimates placed on 2020 tracts by interpolation from 2010 tracts. The SVI theme and the poverty rate are not Atlas products.

A factor model describes the measures it is given. Other measures, such as violence, isolation or air quality, would give other factors, and the count of factors rests on a common cutoff. The chapter does not test the state ADI that Medicare now uses, and it says nothing about how payments change under any formula.

## References

1. Hannan, Edward L.; Wu, Yifeng; Cozzens, Kimberly; Anderson, Brett (2023). The Neighborhood Atlas Area Deprivation Index For Measuring Socioeconomic Status: An Overemphasis On Home Value. Health Affairs 42(5): 702-709. https://doi.org/10.1377/hlthaff.2022.01406
2. Petterson, Stephen (2023). Deciphering the Neighborhood Atlas Area Deprivation Index: the consequences of not standardizing. Health Affairs Scholar 1(5): qxad063. https://doi.org/10.1093/haschl/qxad063
3. Singh, Gopal K. (2003). Area Deprivation and Widening Inequalities in US Mortality, 1969-1998. American Journal of Public Health 93(7): 1137-1143. https://doi.org/10.2105/AJPH.93.7.1137
4. Kind, Amy J. H.; Buckingham, William R. (2018). Making Neighborhood-Disadvantage Metrics Accessible: The Neighborhood Atlas. New England Journal of Medicine 378(26): 2456-2458. https://doi.org/10.1056/NEJMp1802313
5. University of Wisconsin School of Medicine and Public Health (2026). Neighborhood Atlas: Changes between versions of the ADI. Neighborhood Atlas web site, change log with entries dated 2019 to 2026. https://www.neighborhoodatlas.medicine.wisc.edu/changelog
6. Flanagan, Barry E.; Gregory, Edward W.; Hallisey, Elaine J.; Heitgerd, Janet L.; Lewis, Brian (2011). A Social Vulnerability Index for Disaster Management. Journal of Homeland Security and Emergency Management 8(1): Article 3. https://doi.org/10.2202/1547-7355.1792
7. Agency for Toxic Substances and Disease Registry (2024). CDC/ATSDR Social Vulnerability Index 2022 Documentation. CDC/ATSDR web document, October 2024. https://www.atsdr.cdc.gov/place-health/media/pdfs/2024/10/SVI2022Documentation.pdf
8. Centers for Medicare & Medicaid Services (2024). ACO REACH Model Performance Year 2024 Model Update: Health Equity Benchmark Adjustment. CMS Innovation Center web page. https://www.cms.gov/priorities/innovation/innovation-models/reach-py24-model-perf
9. Kolak, Marynia; Bhatt, Jay; Park, Yoon Hong; Padrón, Norma A.; Molefe, Ayrin (2020). Quantification of Neighborhood-Level Social Determinants of Health in the Continental United States. JAMA Network Open 3(1): e1919928. https://doi.org/10.1001/jamanetworkopen.2019.19928

## Provenance

- Pipeline version: 0.1.0
- Build script: book2/chapters/anatomy-of-deprivation/build.py (sha256 e56e63188a4234ee2595d353350743fd6698ed5af7cf5614639f56eae21977af)
- Built on: 2026-10-07
- Site commit: c2b1ff3309cb4a10833f884da91ac0057a66ad0a

Inputs:

| File | Size (bytes) | SHA-256, first 1 MB | SHA-256, last 1 MB |
| --- | ---: | --- | --- |
| book2/chapters/anatomy-of-deprivation/analysis.py | 10,402 | ba7b9a4c847f6154baa0042aca27e876f4041ba13b46ab96a32f6205c5d88f9e | ba7b9a4c847f6154baa0042aca27e876f4041ba13b46ab96a32f6205c5d88f9e |
| book2/chapters/anatomy-of-deprivation/deprivation.py | 18,028 | 092694cfd30446c54292ca7182d4d784e95f28766c0db63fedf53f84c7bb1b7d | 092694cfd30446c54292ca7182d4d784e95f28766c0db63fedf53f84c7bb1b7d |
| data/analytic/tract_atlas.parquet | 39,659,925 | 29d5e4d943552d54aeb6ba92aadc939d9bd15bdea553c78357dbd516757be627 | 1e10d1ecf722340f23ee2212d4999a90250f85b65fa89b39e32db06d9472e22e |
| data/raw/acs/deprivation_components_2019_2023/county.parquet | 1,004,968 | c894ba6cf96378244f47cc4271a24bff087d15ba27f09e49a4ebc0a4786c6f9d | c894ba6cf96378244f47cc4271a24bff087d15ba27f09e49a4ebc0a4786c6f9d |
| data/raw/acs/deprivation_components_2019_2023/tract.parquet | 13,197,793 | 5545cde2cbe96a8b56e468d65388248a241aa6bd300d3c9e325d08d3f86b11da | 5c8ec454d066988b0f1a01aaa6ed34ebfc05d162e01cf9c6bb01f9198cbaff12 |
| data/staged/adi/adi_atlas_tract2020_2023.parquet | 798,078 | 5e409bc212ff871def780fa27b4cf8b616d9da420b2b9f6455517eddcd8c1a95 | 5e409bc212ff871def780fa27b4cf8b616d9da420b2b9f6455517eddcd8c1a95 |
| data/staged/life_expectancy/usaleep_tract2020.parquet | 1,316,169 | 9c9c7f103791b75eddba74e68a89804efdf2486b6b66bb4bc02a46ba5bf8bc23 | 27317107be2465d9fd5e83271dea0cd6e915ee419e2802bedc28747d0a41d38f |
| data/staged/mobility/oa_module1_tract2020.parquet | 16,362,438 | 26cb468a319ad2e727c5eac733b4a47b77a83781ae189e1828e8c40bc0e3ace8 | 5702bad38b81b5ee0b3d01ba9d1ece7d23328985eafba269720a0af5ef6820d4 |
| data/staged/ses/acs5_tract2020_2019_2023.parquet | 7,232,424 | 5fe6f730a35225f14d5d8abfc6b9bce58b52e545f309a7eac74f1dae74d7a31c | 7b4a2475b3852d3f91804716a04566ef8ceadacec2be9fb85f5eb642078efac6 |
| data/staged/svi/svi_tract2020_2022.parquet | 17,740,415 | cb3860d8edc2aaf1f2a02211a58f7a0b80457eb6ad9a37b5ec8abb2d5da9da10 | e371a4a696d5bed05d54ad7c5b6ca2504508805b847ec415c2322be7f66ec2b2 |
| C:/github/zipcode-book/content/chapters/how-far-one-ranking-goes/data/facts.json | 12,984 | d9eee2e8e34b82b69d1588ea09d316fcfd6359a427dca3c2b8d6df2536c0a23f | d9eee2e8e34b82b69d1588ea09d316fcfd6359a427dca3c2b8d6df2536c0a23f |

Outputs:

| File | SHA-256 |
| --- | --- |
| facts.json | b2114a13c1680aece0c1276ca8e9fd2cf217c20eee5e77af8facd95a362b80d1 |
| figures/adi_measures.json | ab6f18838d1c8dd8cd1e48ce73639126d777fda7756c6fd51f83bbbd18074b3b |
| figures/adi_vs_standardized.json | 161ffe9b9069fec243b84c72f947054491a5c8661e2407645084e027f559da70 |
| figures/index_family.json | ee2d68c3f78b7bff042ed2e6d465782fa1a945a6cf902cecb07f73f57d1de1a3 |
| figures/index_vs_outcomes.json | 52d7b57ab96a78b161e43e263e286984fe10322859745796823c82e7cfb2c535 |
| figures/input_correlations.json | a4afe85136a2cfa078d349f8cce7db5f2daef7b91e7c3df9f23ba880028aefbc |
| figures/rank_flows.json | 9ca27dc04158e18ce684d8bc77f056f16ddd6898e50ff964bc92fb9daf67cacc |
| figures/three_factors.json | dcd42d8fe5c5399674821754638a9d50d69e004001fa22473687e609547821d8 |
| figures/variance_by_measure.json | fde464d3c3b4ab444de6364a1ad79df3bc6cb519904e1c23a996f9d49fabd43b |
| meta.json | 3201ca5d32429f34d990bee47644e0ace6bda27c58af0a46c472d148d7a12217 |
| public/data/anatomy-of-deprivation/adi_vs_standardized.json | 8b18d89ffc6a5974c86000a69aa13583c31ef71965af2bbd030bc5b0b9d17707 |
| refs.json | 9971c7f179dccbee8d4de6491191e71ea45422bd12a21f1a7a3a83f1251943ed |

## Machine formats

- Report (HTML): https://vishalsingh.org/zipcode/anatomy-of-deprivation/report
- Report (Markdown): https://vishalsingh.org/zipcode/anatomy-of-deprivation/report.md
- Every estimate (JSON): https://vishalsingh.org/zipcode/report/anatomy-of-deprivation/facts.json
- Table 1, figure "adi_measures" (JSON): https://vishalsingh.org/zipcode/report/anatomy-of-deprivation/figures/adi_measures.json
- Table 2, figure "input_correlations" (JSON): https://vishalsingh.org/zipcode/report/anatomy-of-deprivation/figures/input_correlations.json
- Table 3, figure "variance_by_measure" (JSON): https://vishalsingh.org/zipcode/report/anatomy-of-deprivation/figures/variance_by_measure.json
- Table 4, figure "index_family" (JSON): https://vishalsingh.org/zipcode/report/anatomy-of-deprivation/figures/index_family.json
- Table 5, figure "adi_vs_standardized" (JSON): https://vishalsingh.org/zipcode/report/anatomy-of-deprivation/figures/adi_vs_standardized.json
- Table 6, figure "three_factors" (JSON): https://vishalsingh.org/zipcode/report/anatomy-of-deprivation/figures/three_factors.json
- Table 7, figure "index_vs_outcomes" (JSON): https://vishalsingh.org/zipcode/report/anatomy-of-deprivation/figures/index_vs_outcomes.json
- Table 8, figure "rank_flows" (JSON): https://vishalsingh.org/zipcode/report/anatomy-of-deprivation/figures/rank_flows.json
- Payload of "adi_vs_standardized" (JSON): https://vishalsingh.org/zipcode/data/anatomy-of-deprivation/adi_vs_standardized.json
- Index of every report: https://vishalsingh.org/zipcode/llms.txt; all reports in one file: https://vishalsingh.org/zipcode/llms-full.txt

---

# What a county average leaves out: report

Chapter 1.2 of ZIP Code Destiny, by Vishal Singh (NYU Stern). Generated from the chapter's data files; the reading version is at https://vishalsingh.org/zipcode/deprivation-at-two-zooms. Table n of this report is figure n of the chapter.

## Identification

- Title: What a county average leaves out
- Label: Two zooms
- Chapter: 1.2 (Part I, Rankings and scale; Measuring rich and poor)
- Question: How does a deprivation index change when it is measured for counties instead of tracts?
- Dek: A county's percentile on a deprivation index follows the average of its tracts' percentiles, with a correlation of 0.96, yet 32.7% of residents live in a tract more than 25 percentile points from their county's percentile.
- Grain: neighborhoods
- Status: published
- Published: 2026-10-09
- Updated: not revised
- Data sources:
  - American Community Survey. ACS 2019-2023 five-year estimates: the staged tract file, a Census API pull of housing, occupation, income-distribution and schooling tables for tracts, and the same pull for counties, Connecticut planning regions and Puerto Rico municipios; years: 2019-2023 (5-year); weights: ACS 2019-2023 population of the tract or county
  - Area Deprivation Index. Neighborhood Atlas ADI, 2023 release (version 4.0.1): a published rank is required for a tract to enter the tract frame, and the ADI is the comparison for the tract index; years: 2023 release (version 4.0.1; built on ACS 2019-2023); weights: none (index); tract values are unweighted means of block-group ranks
  - USALEEP life expectancy. NCHS small-area life expectancy at birth, deaths 2010 to 2015, 2020 tracts (interpolated), institutional tracts dropped; years: 2010-2015; weights: ACS 2019-2023 tract population
  - Opportunity Atlas. Mean adult household income rank of children from low-income families, 1978 to 1983 birth cohorts, 2020 tracts (interpolated); years: 1978-1983 birth cohorts (2018 release); weights: ACS 2019-2023 tract population
- Canonical URL: https://vishalsingh.org/zipcode/deprivation-at-two-zooms
- Report: https://vishalsingh.org/zipcode/deprivation-at-two-zooms/report
- Markdown: https://vishalsingh.org/zipcode/deprivation-at-two-zooms/report.md
- BibTeX key: singh2026deprivationattwozooms

```bibtex
@misc{singh2026deprivationattwozooms,
  author = {Singh, Vishal},
  title = {What a county average leaves out},
  year = {2026},
  url = {https://vishalsingh.org/zipcode/deprivation-at-two-zooms},
  note = {In ZIP Code Destiny. Published 9 October 2026.}
}
```

## Principal estimates

The facts the dek and the opening prose cite, in order of appearance.

| Estimand | Population | Period | Value | Interval | n | Weights | Source | Fact id |
| --- | --- | --- | ---: | --- | ---: | --- | --- | --- |
| population-weighted correlation across counties of the county's own percentile among residents and the population-weighted mean of its tracts' percentiles among residents, interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of the counties in the county frame (all measures from the county's own ACS estimate) | ACS 2019 to 2023 five-year estimates | 0.96 | 95% CI 0.95 to 0.97 | 3,203 counties | ACS 2019-2023 county population | American Community Survey | r_own_mean |
| share of residents whose tract percentile differs from their county's percentile by more than the first cutoff, percentiles among residents, interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of the tracts in the tract frame (a published ADI rank, a population and all measures) | ACS 2019 to 2023 five-year estimates | 32.7% | 95% CI 30.9% to 34.4% | 76,495 tracts | ACS 2019-2023 tract population | American Community Survey | share25 |
| gap between a resident's tract percentile and county percentile above which the gap is called large; a definition | residents of the tract frame | ACS 2019 to 2023 five-year estimates | 25 |   |   | none (definition) | American Community Survey | gap_cut_1 |
| median absolute difference between the percentile of a resident's tract and the percentile of the resident's county, percentiles among residents (a unit's percentile is the share of residents of its box who live in units that score lower, with half of any tied units counted), interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of the tracts in the tract frame (a published ADI rank, a population and all measures) | ACS 2019 to 2023 five-year estimates | 16.0 | 95% CI 15.2 to 17.0 | 76,495 tracts | ACS 2019-2023 tract population | American Community Survey | med_disp |
| count of tracts in the tract frame | 2020 census tracts with a published ADI rank, a population above zero and all measures | ACS 2019 to 2023 five-year estimates | 76,496 |   |   | none (full count) | American Community Survey | n_tracts |
| sum of ACS tract population over the tract frame | residents of the tracts in the tract frame (a published ADI rank, a population and all measures) | ACS 2019 to 2023 five-year estimates | 309.9 million |   |   | none (full count) | American Community Survey | pop_tracts |

## Findings

A county's percentile on a deprivation index follows the average of its tracts' percentiles, with a correlation of 0.96 (95% CI 0.95 to 0.97) across counties. The percentile of a resident's tract still differs from the percentile of the resident's county by a median of 16.0 (95% CI 15.2 to 17.0) percentile points, and 32.7% (95% CI 30.9% to 34.4%) of residents live in a tract more than 25 points away. Both percentiles rank units by their residents. The usual ranking, which counts each tract and each county once, is reported beside it where it changes the answer.

This chapter builds one index in two boxes, across 76,496 census tracts and across the counties that contain them, and compares what the two boxes say about the same 309.9 million residents. Chapter 1.1, What the deprivation index measures (https://vishalsingh.org/zipcode/anatomy-of-deprivation) takes the index apart measure by measure, Chapter 10.1, Gradients, slopes and variance (https://vishalsingh.org/zipcode/gradients-slopes-and-variance) splits the variance of tract measures between states, counties and tracts, and Chapter 10.2, Families of places (https://vishalsingh.org/zipcode/families-of-places) groups tracts into families instead of ranking them.

### One recipe in two boxes

The index is the first principal component of 14 census measures, the measures that Chapter 1.1, What the deprivation index measures (https://vishalsingh.org/zipcode/anatomy-of-deprivation) rebuilds from the American Community Survey (ACS) for 2019 to 2023, without the share of single-parent households. The measures are 4 dollar amounts (family income, home value, rent and mortgage cost), the log of the ratio of low-income to high-income households, and shares of people, households or homes for the rest. Each box standardizes its measures, fits the component and signs it so that a higher score means more deprived.

A box is the set of units that are compared. In the tract box the units are the 76,496 2020 census tracts that have a published Area Deprivation Index (ADI) rank, a population and all the measures, among them 779 tracts in Puerto Rico. In the county box they are the 3,205 of 3,222 rows of the county ACS pull (counties, Connecticut planning regions and Puerto Rico municipios) that have all the measures. Every measure in the county box is the county's own ACS estimate. A county's median family income is the median of its families and not an average of its tracts' medians, which is what an analyst gets from a county release of the survey.

The staged tract file and the county file define the single-parent share differently. In the median county outside Utah and Puerto Rico the county value is 1.52 times the resident-weighted mean of its tracts' values, so one recipe cannot use it in both boxes. Adding the tract version back to the tract index changes its ranking little, with a rank correlation of 0.996 between the two.

A score becomes a percentile when it is ranked against a reference population. The usual practice ranks units, as the Neighborhood Atlas does for block groups.[1] Counties differ so much in size that this counts places and not people. The average resident lives in a county at the 26.5 (95% CI 24.5 to 28.8) percentile among counties. This chapter ranks by residents instead, so that a unit's percentile is the share of the box's residents who live in units that score lower. Tracts hold similar numbers of people, so for tracts the two conventions differ by 2.6 percentile points on average. The ranking by units returns below wherever it changes an answer.

The Neighborhood Atlas ranks block groups and not counties, so a county index has to be built, and the index here is not the ADI. It ranks tracts with a rank correlation of +0.78 (95% CI 0.73 to 0.82) with the published tract rank, because the ADI sums the measures in their own units, as Chapter 1.1, What the deprivation index measures (https://vishalsingh.org/zipcode/anatomy-of-deprivation) shows, and so follows home value.[2][3] Applying the ADI formula to county estimates gives the same result. The formula ranks counties with a rank correlation of −0.99 (95% CI −0.99 to −0.99) with home value, negative because the formula scores low home values as more deprived, and +0.90 (95% CI 0.86 to 0.92) with the county index of this chapter. Fitting the county component by counting each county once, instead of weighting counties by residents, changes the county ranking little, with a rank correlation of +0.988 (95% CI 0.986 to 0.990) between the two.

Table 1 compares the weights. The first component carries 40.6% (95% CI 39.3% to 42.3%) of the variance of the measures in the tract box and 44.2% (95% CI 42.2% to 46.8%) in the county box, and the second component carries 18.3% (95% CI 16.4% to 20.0%) and 29.3% (95% CI 25.1% to 32.8%).

8 of the 14 weights differ by less than 0.05 between the boxes. Homeownership has a weight of −0.18 (95% CI −0.20 to −0.16) in the tract box and 0.03 (95% CI −0.06 to 0.10) in the county box. Crowding goes from 0.14 (95% CI 0.12 to 0.15) to −0.05 (95% CI −0.12 to 0.03). Both county intervals include zero.

The weight on adults with little schooling falls from 0.21 (95% CI 0.20 to 0.22) to 0.09 (95% CI 0.01 to 0.15), and the weight on income disparity rises from 0.27 (95% CI 0.26 to 0.27) to 0.34 (95% CI 0.31 to 0.36). The recipe carries over from tracts to counties, and the measures that rank tracts are not all the measures that rank counties.

Table 1. Homeownership has a weight of −0.18 in the tract index and 0.03 in the county index, and crowding goes from 0.14 to −0.05

### The county ranking follows the average of its neighborhoods

A county has two percentiles here. Its own percentile recomputes the index from the county's estimates, and the mean of its tracts' percentiles averages a rank. Set each county's own percentile against the population-weighted mean of the percentiles of its tracts. Across counties the two correlate 0.96 (95% CI 0.95 to 0.97) when counties are weighted by residents and 0.96 (95% CI 0.95 to 0.96) when each county counts once.

The tract box covers only tracts that have all the measures, so a county's tracts in the frame may hold only part of its residents. Among counties where they hold at least 90% of residents the correlation is 0.96 (95% CI 0.95 to 0.97), and 605 counties, with 25.5% of residents, fall below that cutoff. Ranking counties on the mean tract score, instead of the mean tract percentile, gives a rank correlation of 0.99 (95% CI 0.98 to 0.99) with the county's own percentile. For a question about counties as units, the county file and the tract file give nearly the same order.

The agreement is in order and not in level. The county's own percentile lies farther from the middle of the scale than the mean of its tracts' percentiles in counties that hold 96.3% (95% CI 94.1% to 98.0%) of residents, and in 96.7% (95% CI 96.0% to 97.3%) of counties counted once. Across counties, each additional point of county percentile goes with 0.57 (95% CI 0.55 to 0.59) points of average tract percentile when counties are weighted by residents and 0.68 (95% CI 0.68 to 0.69) points when each county counts once. A county at an end of the county ranking has, on average, neighborhoods that are less extreme than the county itself.

### Residents sit far from their county's percentile

Give each resident two numbers, the percentile of their tract and the percentile of their county. Table 2 plots the difference. Ranking by residents, the median absolute difference is 16.0 (95% CI 15.2 to 17.0) percentile points. 32.7% (95% CI 30.9% to 34.4%) of residents are more than 25 points from their county's percentile and 14.1% (95% CI 12.9% to 15.4%) are more than 40.

Across residents the two percentiles correlate 0.57 (95% CI 0.54 to 0.61). The sign is close to even, with 46.8% (95% CI 44.9% to 48.6%) of residents living in a tract that ranks as more deprived than their county. Among counties where the frame holds at least 90% of residents, the share more than 25 points away is 32.4% (95% CI 30.3% to 34.5%).

The ranking convention changes the picture. Ranking each tract and each county once, as is usual, the median difference is 19.4 (95% CI 18.7 to 20.2) points. 40.7% (95% CI 39.1% to 42.2%) of residents are more than 25 points away and 22.8% (95% CI 21.1% to 24.6%) are more than 40.

Under that convention 81.5% (95% CI 80.3% to 82.6%) of residents live in a tract that ranks as more deprived than their county. That lean follows from the reference population. Counted by units, the average resident lives in a county at the 26.5 percentile, below the middle of the county ranking, while the average resident's tract sits near the middle of the tract ranking.

Table 2. Counted among residents, 32.7% of people live in a tract more than 25 percentile points from their county's percentile, and 40.7% when tracts and counties are ranked as units

### A county percentile hides a wide spread of neighborhoods

Los Angeles County is the most populous county. Built from the county's own estimates, its index places it at the 29.1 percentile among residents and at the 8.0 percentile among counties. Its tracts run from percentile 2.0 at the 5th percentile of residents to percentile 94.2 at the 95th. 53.0% of its residents live in a tract more than 25 points from the county's percentile.

Table 3 draws the 30 most populous counties in the same way. In each of them the county percentile lies between the 5th and 95th percentile of its residents' tracts, and in 28 it lies farther from the middle of the scale than the average of those tracts.

Across these counties the median share of residents more than 25 points from their county's percentile is 47.1%. The counties are chosen by population and ordered by the county percentile, since the spread of a small county rests on few tracts. The line is a spread of tracts and not an interval for the county.

Table 3. Los Angeles County has a county percentile of 29.1, and its residents' tracts run from 2.0 to 94.2

Across the country, 64.6% (95% CI 60.5% to 68.7%) of the variance of the tract percentile lies between tracts of the same county, which a county percentile cannot see. The within-county share depends on the index. Chapter 10.1, Gradients, slopes and variance (https://vishalsingh.org/zipcode/gradients-slopes-and-variance) splits the published ADI and a nine-measure index the same way and gets different shares.

728 counties have at least 20 tracts in the tract frame and hold 81.8% of residents. Among them, the median range from the 5th to the 95th percentile of a county's tracts is 79.0 (95% CI 75.5 to 80.8) points when counties are weighted by residents and 67.4 (95% CI 66.0 to 68.6) points when each county counts once.

The median grows with the floor on the number of tracts. Weighted by residents it goes from 76.5 at a floor of 10 to 81.0 at a floor of 50. Counting each county once it goes from 58.5 to 76.7.

A county value that keeps some of this structure is the share of its residents who live in tracts at or above the 80th percentile, the most deprived fifth of the tract ranking. That share correlates 0.75 (95% CI 0.73 to 0.77) with the county's own percentile across counties. Counties in the middle fifth of the county ranking, with an own percentile from 40th to 60th and at least 20 tracts in the frame, number 159 and hold 17.7% of residents.

In those counties, weighted by residents, the share of residents in the most deprived tracts is 4.3% (95% CI 2.3% to 5.8%) at the 5th percentile of counties, 16.6% (95% CI 14.5% to 20.4%) at the median and 25.3% (95% CI 23.9% to 27.1%) at the 95th. Counting each county once, the three values are 0.0% (95% CI 0.0% to 2.1%), 11.6% (95% CI 9.9% to 13.2%) and 23.9% (95% CI 21.9% to 25.5%). A single county percentile does not record that spread.

### Correlations change with the box

Table 4 correlates the index percentile with four tract measures, first across tracts and then across counties. A county's value of a measure is the mean of its tracts' values weighted by residents, so that only the unit changes.

Life expectancy at birth, from USALEEP for deaths in 2010 to 2015, correlates −0.62 (95% CI −0.64 to −0.60) with the index across tracts and −0.76 (95% CI −0.79 to −0.72) across counties weighted by residents, a stronger relationship at the larger unit. Counting each county once gives −0.56 (95% CI −0.59 to −0.54), weaker than across tracts, so the strengthening depends on how counties are weighted.

Upward mobility, the Opportunity Atlas rank of children from low-income families, correlates −0.58 (95% CI −0.60 to −0.56) across tracts, −0.59 (95% CI −0.64 to −0.54) across weighted counties and −0.36 (95% CI −0.38 to −0.33) across counties counted once. The tract and weighted-county intervals overlap.

The poverty rate follows the same pattern, with +0.78 (95% CI 0.78 to 0.79), +0.75 (95% CI 0.72 to 0.78) and +0.65 (95% CI 0.63 to 0.66). Poverty is one of the measures the index is built from, so that correlation is partly built in at any unit. Homeownership is another, and its correlation with the index is −0.43 (95% CI −0.46 to −0.40) across tracts, +0.06 (95% CI −0.06 to 0.17) across weighted counties, an interval that includes zero, and −0.07 (95% CI −0.11 to −0.03) across counties counted once. The loss matches the loss of weight in Table 1.

Table 4. The index correlates −0.62 with life expectancy across tracts and −0.76 across counties, and −0.43 with homeownership across tracts and +0.06 across counties

Gehlke and Biehl reported in 1934 that the size of a correlation computed from census tract data depended on how the tracts were grouped.[4] Openshaw and Taylor regrouped one data set into many zonings and obtained a wide range of correlations from it.[5][6] The pattern is the modifiable areal unit problem, which reaches multivariate analysis as well,[7] and Chapter 10.1, Gradients, slopes and variance (https://vishalsingh.org/zipcode/gradients-slopes-and-variance) measures it for correlations among ACS measures. The results here come from one set of boundaries and show no general direction. Under weights, the correlation with life expectancy rises with the unit, those with mobility and poverty hold, and the one with homeownership disappears. A correlation across counties describes counties. Robinson showed that a correlation across areas can differ from the correlation across the people inside them,[8][9] and nothing in this chapter measures individuals.

### What these data cannot show

The data describe places. A county percentile and a tract percentile are properties of places, and neither says how deprived any resident is. A poor household in a comfortable tract takes the tract's percentile, which is the problem of ecological inference. The correlations in the last section are across places and say nothing about the people inside them.

The ACS files carry no margins of error, and the medians of small tracts rest on few households, so part of the spread of tract percentiles within a county is sampling error, and the data cannot say how much. The tract box leaves out tracts with a missing measure, mostly a missing median home value, rent or mortgage cost, as Chapter 1.1, What the deprivation index measures (https://vishalsingh.org/zipcode/anatomy-of-deprivation) finds, and tracts without a published ADI rank. The frame holds 92.4% of all residents and 92.9% of the residents of ranked tracts, and 1,691 tracts with 2.1 million residents have no rank.

The tracts outside the frame are poorer, with a poverty rate of 17.8% against 12.5% inside. The missing tracts are concentrated in some counties. The frame holds 52.5% of the residents of Bronx County and 74.2% of New York County, so for those counties and others with low coverage the average and the spread of tracts describe only the tracts that remain.

The percentile convention is a choice. This chapter ranks by residents, county files usually rank counties, and the second distribution in Table 2 shows how far the choice moves a statement about residents. The analysis uses one recipe and one set of measures. The ADI formula would rank counties by home value, as the correlations above show, and the state ADI, which ranks units within a state, is not tested. The county index is not compared with published county indexes.

Life expectancy is a modeled estimate for deaths in 2010 to 2015 and upward mobility concerns children born in 1978 to 1983, so both are older than the ACS measures. They show how an index built from recent measures lines up with older outcomes at two units and do not show which unit is correct. That depends on the question. For a program run by counties the county box is the right one, and a finding in it is a finding about counties.

## Tables

### Table 1. Homeownership has a weight of −0.18 in the tract index and 0.03 in the county index, and crowding goes from 0.14 to −0.05

Weights of the 14 standardized measures in the first principal component, tract box (76,496 tracts) and county box (3,205 counties), ACS 2019 to 2023

| box | conf_high | conf_low | measure | n | weight |
| --- | ---: | ---: | --- | ---: | ---: |
| Tract box | −0.366 | −0.379 | Family income | 76,496 | −0.374 |
| County box | −0.368 | −0.383 | Family income | 3,205 | −0.377 |
| Tract box | −0.322 | −0.339 | White-collar jobs | 76,496 | −0.331 |
| County box | −0.290 | −0.319 | White-collar jobs | 3,205 | −0.307 |
| Tract box | −0.281 | −0.312 | Mortgage cost | 76,496 | −0.297 |
| County box | −0.270 | −0.344 | Mortgage cost | 3,205 | −0.317 |
| Tract box | −0.279 | −0.300 | Gross rent | 76,496 | −0.289 |
| County box | −0.299 | −0.354 | Gross rent | 3,205 | −0.333 |
| Tract box | −0.254 | −0.288 | Home value | 76,496 | −0.271 |
| County box | −0.257 | −0.324 | Home value | 3,205 | −0.297 |
| Tract box | −0.161 | −0.198 | Homeownership | 76,496 | −0.182 |
| County box | 0.101 | −0.059 | Homeownership | 3,205 | 0.028 |
| Tract box | 0.191 | 0.077 | No vehicle | 76,496 | 0.133 |
| County box | 0.108 | −0.061 | No vehicle | 3,205 | 0.013 |
| Tract box | 0.155 | 0.120 | Crowding | 76,496 | 0.138 |
| County box | 0.034 | −0.122 | Crowding | 3,205 | −0.051 |
| Tract box | 0.184 | 0.169 | No plumbing | 76,496 | 0.177 |
| County box | 0.278 | 0.245 | No plumbing | 3,205 | 0.264 |
| Tract box | 0.195 | 0.177 | Unemployment | 76,496 | 0.186 |
| County box | 0.219 | 0.116 | Unemployment | 3,205 | 0.171 |
| Tract box | 0.217 | 0.196 | Under 9 years of school | 76,496 | 0.207 |
| County box | 0.154 | 0.012 | Under 9 years of school | 3,205 | 0.086 |
| Tract box | 0.272 | 0.259 | Income disparity | 76,496 | 0.266 |
| County box | 0.356 | 0.311 | Income disparity | 3,205 | 0.338 |
| Tract box | 0.349 | 0.338 | Poverty | 76,496 | 0.344 |
| County box | 0.358 | 0.309 | Poverty | 3,205 | 0.337 |
| Tract box | 0.376 | 0.364 | Below 150% of poverty | 76,496 | 0.371 |
| County box | 0.373 | 0.331 | Below 150% of poverty | 3,205 | 0.355 |

Note: Each box standardizes its measures with population-weighted means and standard deviations, fits the component to the population-weighted correlation matrix of its units (tracts weighted by residents in the tract box, counties weighted by residents in the county box) and signs it so that the poverty rate has a positive weight.

Note: The county box takes every measure from the county's own ACS estimate, not from an average of its tracts. Counties include Connecticut planning regions and Puerto Rico municipios.

Source: ACS 2019-2023 five-year estimates (tracts and counties); population-weighted

Data: https://vishalsingh.org/zipcode/report/deprivation-at-two-zooms/figures/weights_by_zoom.json

### Table 2. Counted among residents, 32.7% of people live in a tract more than 25 percentile points from their county's percentile, and 40.7% when tracts and counties are ranked as units

Population-weighted distribution of a resident's tract percentile minus the resident's county percentile, in bins of five percentile points, 309.9 million residents of 76,496 tracts, ACS 2019 to 2023

| conf_high | conf_low | convention | n | share | x0 | x1 |
| ---: | ---: | --- | ---: | ---: | ---: | ---: |
| 0.024 | 0 | Among residents | 5 | 0.009 | −95 | −90 |
| 0.025 | 0.003 | Among residents | 12 | 0.013 | −90 | −85 |
| 0.095 | 0.019 | Among residents | 48 | 0.053 | −85 | −80 |
| 0.136 | 0.045 | Among residents | 72 | 0.086 | −80 | −75 |
| 0.220 | 0.096 | Among residents | 119 | 0.153 | −75 | −70 |
| 0.424 | 0.196 | Among residents | 241 | 0.300 | −70 | −65 |
| 0.651 | 0.324 | Among residents | 345 | 0.465 | −65 | −60 |
| 0.841 | 0.534 | Among residents | 535 | 0.680 | −60 | −55 |
| 1.24 | 0.808 | Among residents | 750 | 1.02 | −55 | −50 |
| 1.70 | 1.16 | Among residents | 1,050 | 1.42 | −50 | −45 |
| 2.12 | 1.59 | Among residents | 1,366 | 1.85 | −45 | −40 |
| 2.77 | 2.19 | Among residents | 1,801 | 2.48 | −40 | −35 |
| 3.62 | 2.97 | Among residents | 2,451 | 3.29 | −35 | −30 |
| 4.81 | 3.93 | Among residents | 3,246 | 4.34 | −30 | −25 |
| 5.52 | 4.75 | Among residents | 3,941 | 5.13 | −25 | −20 |
| 6.41 | 5.69 | Among residents | 4,717 | 6.04 | −20 | −15 |
| 7.58 | 6.71 | Among residents | 5,618 | 7.11 | −15 | −10 |
| 9.10 | 8.06 | Among residents | 6,675 | 8.57 | −10 | −5 |
| 10.88 | 9.51 | Among residents | 7,779 | 10.17 | −5 | 0 |
| 9.69 | 8.27 | Among residents | 6,783 | 8.94 | 0 | 5 |
| 7.31 | 6.53 | Among residents | 5,316 | 6.91 | 5 | 10 |
| 6.16 | 5.57 | Among residents | 4,464 | 5.86 | 10 | 15 |
| 5.00 | 4.48 | Among residents | 3,656 | 4.74 | 15 | 20 |
| 4.12 | 3.60 | Among residents | 3,006 | 3.87 | 20 | 25 |
| 3.82 | 3.16 | Among residents | 2,664 | 3.49 | 25 | 30 |
| 3.01 | 2.48 | Among residents | 2,102 | 2.75 | 30 | 35 |
| 2.50 | 1.96 | Among residents | 1,678 | 2.24 | 35 | 40 |
| 2.15 | 1.54 | Among residents | 1,475 | 1.85 | 40 | 45 |
| 1.90 | 1.32 | Among residents | 1,214 | 1.61 | 45 | 50 |
| 1.60 | 1.01 | Among residents | 981 | 1.30 | 50 | 55 |
| 1.37 | 0.768 | Among residents | 765 | 1.04 | 55 | 60 |
| 1.16 | 0.542 | Among residents | 575 | 0.800 | 60 | 65 |
| 1.03 | 0.453 | Among residents | 521 | 0.697 | 65 | 70 |
| 0.380 | 0.208 | Among residents | 217 | 0.287 | 70 | 75 |
| 0.352 | 0.148 | Among residents | 164 | 0.240 | 75 | 80 |
| 0.192 | 0.067 | Among residents | 83 | 0.122 | 80 | 85 |
| 0.113 | 0.033 | Among residents | 47 | 0.068 | 85 | 90 |
| 0.028 | 0.005 | Among residents | 11 | 0.015 | 90 | 95 |
| 0.004 | 0 | Among residents | 2 | 0.001 | 95 | 100 |
| 0.002 | 0 | Among tracts and counties | 1 | 0.001 | −85 | −80 |
| 0.009 | 0 | Among tracts and counties | 2 | 0.003 | −80 | −75 |
| 0.042 | 0.002 | Among tracts and counties | 11 | 0.017 | −75 | −70 |
| 0.035 | 0 | Among tracts and counties | 11 | 0.013 | −70 | −65 |
| 0.078 | 0.019 | Among tracts and counties | 43 | 0.045 | −65 | −60 |
| 0.118 | 0.033 | Among tracts and counties | 62 | 0.072 | −60 | −55 |
| 0.175 | 0.066 | Among tracts and counties | 104 | 0.115 | −55 | −50 |
| 0.245 | 0.103 | Among tracts and counties | 138 | 0.166 | −50 | −45 |
| 0.337 | 0.184 | Among tracts and counties | 210 | 0.255 | −45 | −40 |
| 0.501 | 0.304 | Among tracts and counties | 310 | 0.392 | −40 | −35 |
| 0.687 | 0.410 | Among tracts and counties | 424 | 0.538 | −35 | −30 |
| 1.06 | 0.666 | Among tracts and counties | 667 | 0.845 | −30 | −25 |
| 1.37 | 0.961 | Among tracts and counties | 921 | 1.16 | −25 | −20 |
| 1.99 | 1.46 | Among tracts and counties | 1,352 | 1.72 | −20 | −15 |
| 2.81 | 2.18 | Among tracts and counties | 1,940 | 2.49 | −15 | −10 |
| 4.20 | 3.52 | Among tracts and counties | 3,023 | 3.85 | −10 | −5 |
| 7.19 | 6.48 | Among tracts and counties | 5,226 | 6.83 | −5 | 0 |
| 11.42 | 9.67 | Among tracts and counties | 7,717 | 10.46 | 0 | 5 |
| 9.77 | 8.76 | Among tracts and counties | 6,789 | 9.26 | 5 | 10 |
| 9.05 | 8.27 | Among tracts and counties | 6,399 | 8.65 | 10 | 15 |
| 8.16 | 7.40 | Among tracts and counties | 5,885 | 7.78 | 15 | 20 |
| 7.48 | 6.71 | Among tracts and counties | 5,425 | 7.09 | 20 | 25 |
| 6.37 | 5.85 | Among tracts and counties | 4,652 | 6.10 | 25 | 30 |
| 5.63 | 5.14 | Among tracts and counties | 4,208 | 5.39 | 30 | 35 |
| 4.84 | 4.38 | Among tracts and counties | 3,597 | 4.60 | 35 | 40 |
| 4.13 | 3.67 | Among tracts and counties | 3,044 | 3.89 | 40 | 45 |
| 3.70 | 3.26 | Among tracts and counties | 2,730 | 3.48 | 45 | 50 |
| 3.25 | 2.77 | Among tracts and counties | 2,405 | 3.02 | 50 | 55 |
| 2.98 | 2.42 | Among tracts and counties | 2,134 | 2.71 | 55 | 60 |
| 2.71 | 2.04 | Among tracts and counties | 1,859 | 2.38 | 60 | 65 |
| 2.22 | 1.68 | Among tracts and counties | 1,530 | 1.97 | 65 | 70 |
| 1.95 | 1.36 | Among tracts and counties | 1,302 | 1.66 | 70 | 75 |
| 1.70 | 1.10 | Among tracts and counties | 1,095 | 1.40 | 75 | 80 |
| 1.32 | 0.702 | Among tracts and counties | 753 | 0.987 | 80 | 85 |
| 0.764 | 0.303 | Among tracts and counties | 402 | 0.497 | 85 | 90 |
| 0.215 | 0.100 | Among tracts and counties | 116 | 0.156 | 90 | 95 |
| 0.023 | 0.003 | Among tracts and counties | 8 | 0.011 | 95 | 100 |

Note: Percentiles come from the index in each box. Among residents, a unit's percentile is the share of residents of its box who live in units that score lower; among tracts and counties, it is the share of units that score lower.

Note: Intervals in the tooltip are 95% intervals from a bootstrap over counties (replicates and seed in the methods note). One tract has no county in the county frame and is left out.

Source: ACS 2019-2023 five-year estimates (tracts and counties); population-weighted (bins: shares of residents)

Data: https://vishalsingh.org/zipcode/report/deprivation-at-two-zooms/figures/tract_county_gap.json

### Table 3. Los Angeles County has a county percentile of 29.1, and its residents' tracts run from 2.0 to 94.2

Percentile among residents of the county index built from the county's own estimates, beside the average and the 5th to 95th percentile range of the percentiles of its tracts, the 30 most populous counties

| conf_high | conf_low | county | measure | n | percentile |
| ---: | ---: | --- | --- | ---: | ---: |
| n/a | n/a | Santa Clara County, CA | County's own percentile | n/a | 0.500 |
| 59.50 | 0.100 | Santa Clara County, CA | Average of its tracts' percentiles | 392 | 13.50 |
| n/a | n/a | Nassau County, NY | County's own percentile | n/a | 2.10 |
| 63.50 | 1.10 | Nassau County, NY | Average of its tracts' percentiles | 217 | 16.10 |
| n/a | n/a | Alameda County, CA | County's own percentile | n/a | 2.60 |
| 80.40 | 0.300 | Alameda County, CA | Average of its tracts' percentiles | 357 | 23.40 |
| n/a | n/a | Middlesex County, MA | County's own percentile | n/a | 3.50 |
| 68.70 | 0.900 | Middlesex County, MA | Average of its tracts' percentiles | 328 | 20 |
| n/a | n/a | King County, WA | County's own percentile | n/a | 4.40 |
| 73 | 0.800 | King County, WA | Average of its tracts' percentiles | 476 | 21.40 |
| n/a | n/a | Suffolk County, NY | County's own percentile | n/a | 6.40 |
| 53.80 | 2.80 | Suffolk County, NY | Average of its tracts' percentiles | 311 | 19.60 |
| n/a | n/a | Orange County, CA | County's own percentile | n/a | 8.30 |
| 81.90 | 1.40 | Orange County, CA | Average of its tracts' percentiles | 587 | 30.30 |
| n/a | n/a | New York County, NY | County's own percentile | n/a | 9.40 |
| 93.90 | 1.70 | New York County, NY | Average of its tracts' percentiles | 217 | 31.60 |
| n/a | n/a | San Diego County, CA | County's own percentile | n/a | 13.20 |
| 86.50 | 1.10 | San Diego County, CA | Average of its tracts' percentiles | 684 | 31.80 |
| n/a | n/a | Los Angeles County, CA | County's own percentile | n/a | 29.10 |
| 94.20 | 2 | Los Angeles County, CA | Average of its tracts' percentiles | 2,226 | 46.20 |
| n/a | n/a | Queens County, NY | County's own percentile | n/a | 32.10 |
| 91.70 | 14.30 | Queens County, NY | Average of its tracts' percentiles | 612 | 49.70 |
| n/a | n/a | Sacramento County, CA | County's own percentile | n/a | 35.20 |
| 92 | 6 | Sacramento County, CA | Average of its tracts' percentiles | 341 | 46.40 |
| n/a | n/a | Palm Beach County, FL | County's own percentile | n/a | 36.30 |
| 90.90 | 4.40 | Palm Beach County, FL | Average of its tracts' percentiles | 321 | 44.50 |
| n/a | n/a | Riverside County, CA | County's own percentile | n/a | 37.20 |
| 89.90 | 6.70 | Riverside County, CA | Average of its tracts' percentiles | 491 | 45.20 |
| n/a | n/a | Maricopa County, AZ | County's own percentile | n/a | 38.40 |
| 93.10 | 6.60 | Maricopa County, AZ | Average of its tracts' percentiles | 911 | 44.70 |
| n/a | n/a | Kings County, NY | County's own percentile | n/a | 40.20 |
| 93.60 | 3.40 | Kings County, NY | Average of its tracts' percentiles | 651 | 53.30 |
| n/a | n/a | Tarrant County, TX | County's own percentile | n/a | 42.60 |
| 94.60 | 5.70 | Tarrant County, TX | Average of its tracts' percentiles | 406 | 46.60 |
| n/a | n/a | Broward County, FL | County's own percentile | n/a | 44.70 |
| 90.40 | 5 | Broward County, FL | Average of its tracts' percentiles | 381 | 47.40 |
| n/a | n/a | Orange County, FL | County's own percentile | n/a | 48.80 |
| 91.30 | 10.10 | Orange County, FL | Average of its tracts' percentiles | 244 | 49.60 |
| n/a | n/a | Hillsborough County, FL | County's own percentile | n/a | 50.60 |
| 91.10 | 8.10 | Hillsborough County, FL | Average of its tracts' percentiles | 297 | 47.70 |
| n/a | n/a | San Bernardino County, CA | County's own percentile | n/a | 52 |
| 94.50 | 8.10 | San Bernardino County, CA | Average of its tracts' percentiles | 432 | 55.10 |
| n/a | n/a | Cook County, IL | County's own percentile | n/a | 54 |
| 96 | 5 | Cook County, IL | Average of its tracts' percentiles | 1,221 | 51.50 |
| n/a | n/a | Miami-Dade County, FL | County's own percentile | n/a | 57.50 |
| 94.10 | 6.70 | Miami-Dade County, FL | Average of its tracts' percentiles | 606 | 53.10 |
| n/a | n/a | Dallas County, TX | County's own percentile | n/a | 60.40 |
| 95.70 | 4 | Dallas County, TX | Average of its tracts' percentiles | 548 | 57.10 |
| n/a | n/a | Clark County, NV | County's own percentile | n/a | 63.80 |
| 95.70 | 14.90 | Clark County, NV | Average of its tracts' percentiles | 502 | 55.10 |
| n/a | n/a | Bexar County, TX | County's own percentile | n/a | 67.20 |
| 97.40 | 11.60 | Bexar County, TX | Average of its tracts' percentiles | 352 | 57.80 |
| n/a | n/a | Harris County, TX | County's own percentile | n/a | 69.50 |
| 97.60 | 5.90 | Harris County, TX | Average of its tracts' percentiles | 909 | 56.10 |
| n/a | n/a | Philadelphia County, PA | County's own percentile | n/a | 89.60 |
| 99.10 | 14.90 | Philadelphia County, PA | Average of its tracts' percentiles | 342 | 69.30 |
| n/a | n/a | Bronx County, NY | County's own percentile | n/a | 92.90 |
| 98.50 | 20.10 | Bronx County, NY | Average of its tracts' percentiles | 188 | 72.40 |
| n/a | n/a | Wayne County, MI | County's own percentile | n/a | 93.60 |
| 99.50 | 13.80 | Wayne County, MI | Average of its tracts' percentiles | 488 | 67.50 |

Note: The line is a spread of the county's tract percentiles, weighted by residents, and not a confidence interval. The county percentile has no interval because the staged county files carry no margins of error for the ACS estimates it rests on.

Note: Tract percentiles are among residents of the tract frame. Tracts missing a home value, rent or mortgage median are outside the frame, so for counties where the frame holds less than 90% of residents (Nassau County, NY; Suffolk County, NY; New York County, NY; Kings County, NY; Miami-Dade County, FL; Dallas County, TX; Harris County, TX; Bronx County, NY; Wayne County, MI) the average and the spread describe only the tracts that remain.

Note: Counties are chosen by population, and ordered by county percentile, because the spread of a small county rests on few tracts.

Source: ACS 2019-2023 five-year estimates (tracts and counties); population-weighted (the line is a spread of tract percentiles, not an interval)

Data: https://vishalsingh.org/zipcode/report/deprivation-at-two-zooms/figures/large_counties_vs_tracts.json

### Table 4. The index correlates −0.62 with life expectancy across tracts and −0.76 across counties, and −0.43 with homeownership across tracts and +0.06 across counties

Correlation of the index percentile among residents with four tract measures, across tracts and across counties, ACS 2019 to 2023, USALEEP 2010 to 2015 and Opportunity Atlas 1978 to 1983 cohorts

| conf_high | conf_low | n | outcome | r | zoom |
| ---: | ---: | ---: | --- | ---: | --- |
| −0.600 | −0.640 | 71,874 | Life expectancy at birth (deaths 2010 to 2015) | −0.621 | Tracts |
| −0.717 | −0.792 | 3,101 | Life expectancy at birth (deaths 2010 to 2015) | −0.758 | Counties, weighted by residents |
| −0.537 | −0.587 | 3,101 | Life expectancy at birth (deaths 2010 to 2015) | −0.562 | Counties, each counted once |
| −0.555 | −0.598 | 75,731 | Upward mobility (children born 1978 to 1983) | −0.577 | Tracts |
| −0.542 | −0.640 | 3,155 | Upward mobility (children born 1978 to 1983) | −0.593 | Counties, weighted by residents |
| −0.331 | −0.383 | 3,155 | Upward mobility (children born 1978 to 1983) | −0.356 | Counties, each counted once |
| 0.788 | 0.775 | 76,496 | Poverty rate (ACS 2019 to 2023) | 0.782 | Tracts |
| 0.782 | 0.724 | 3,203 | Poverty rate (ACS 2019 to 2023) | 0.752 | Counties, weighted by residents |
| 0.665 | 0.634 | 3,203 | Poverty rate (ACS 2019 to 2023) | 0.649 | Counties, each counted once |
| −0.401 | −0.463 | 76,496 | Homeownership rate (ACS 2019 to 2023) | −0.432 | Tracts |
| 0.175 | −0.059 | 3,203 | Homeownership rate (ACS 2019 to 2023) | 0.065 | Counties, weighted by residents |
| −0.032 | −0.111 | 3,203 | Homeownership rate (ACS 2019 to 2023) | −0.071 | Counties, each counted once |

Note: A county's value of a measure is the population-weighted mean of its tracts' values, so only the unit changes between the tract dots and the county dots. Poverty and homeownership are ingredients of the index, so their correlations are partly built in; life expectancy and upward mobility are not.

Note: Life expectancy is a modeled estimate for deaths in 2010 to 2015 on 2020 tracts, interpolated from 2010 tracts, with institutional tracts dropped; upward mobility is the Opportunity Atlas rank of children from low-income families, interpolated to 2020 tracts.

Source: ACS 2019-2023; NCHS USALEEP 2010-2015; Opportunity Atlas 1978-1983 birth cohorts; population-weighted unless the row says each county is counted once

Data: https://vishalsingh.org/zipcode/report/deprivation-at-two-zooms/figures/index_correlations_by_zoom.json

## All estimates

Every fact in facts.json (106), sorted by id.

| Fact id | Value | Unit | Estimand | Population | Period | Interval | n | Weights | Source | Note |
| --- | ---: | --- | --- | --- | --- | --- | ---: | --- | --- | --- |
| above | 46.8% | percent of residents | share of residents who live in a tract with a higher percentile (more deprived) than their county, percentiles among residents, interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of the tracts in the tract frame (a published ADI rank, a population and all measures) | ACS 2019 to 2023 five-year estimates | 95% CI 44.9% to 48.6% | 76,495 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| above_unit | 81.5% | percent of residents | share of residents who live in a tract with a higher percentile (more deprived) than their county, percentiles among tracts and among counties, interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of the tracts in the tract frame (a published ADI rank, a population and all measures) | ACS 2019 to 2023 five-year estimates | 95% CI 80.3% to 82.6% | 76,495 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| big_extreme_n | 28 | counties | number of those counties whose own percentile lies farther from the 50th percentile than the mean of their tracts' percentiles | the most populous counties in the county frame | ACS 2019 to 2023 five-year estimates |   |   | none (full count) | American Community Survey |   |
| big_far_median | 47.1% | percent of residents | median over those counties of the share of a county's residents whose tract percentile differs from the county's own percentile by more than the first cutoff | residents of the most populous counties in the county frame | ACS 2019 to 2023 five-year estimates |   | 30 counties | none (each county counts once) | American Community Survey |   |
| bronx_cover | 52.5% | percent of a county's residents | share of the county's residents who live in tracts of the tract frame | residents of Bronx County | ACS 2019 to 2023 five-year estimates |   |   | ACS 2019-2023 county population (full count) | American Community Survey |   |
| close_cut | 0.05 | weight on a standardized measure | difference between the tract-box and county-box weights below which a weight is called nearly unchanged; a definition | the measures of the index | ACS 2019 to 2023 five-year estimates |   |   | none (definition) | American Community Survey |   |
| corr_person | 0.57 | correlation coefficient | population-weighted correlation across residents of the percentile of their tract and the percentile of their county, percentiles among residents, interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of the tracts in the tract frame (a published ADI rank, a population and all measures) | ACS 2019 to 2023 five-year estimates | 95% CI 0.54 to 0.61 | 76,495 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| cover_cut | 90% | percent of a county's residents | share of a county's residents who must live in tracts of the tract frame for the county to enter the restricted comparison; a definition | counties compared with the average of their tracts | ACS 2019 to 2023 five-year estimates |   |   | none (definition) | American Community Survey |   |
| evr1_county | 44.2% | percent of the variance of the measures | share of the variance of the standardized measures carried by the first principal component in the county box, population-weighted correlation matrix, interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of the counties in the county frame (all measures from the county's own ACS estimate) | ACS 2019 to 2023 five-year estimates | 95% CI 42.2% to 46.8% | 3,205 counties | ACS 2019-2023 county population | American Community Survey |   |
| evr1_tract | 40.6% | percent of the variance of the measures | share of the variance of the standardized measures carried by the first principal component in the tract box, population-weighted correlation matrix, interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of the tracts in the tract frame (a published ADI rank, a population and all measures) | ACS 2019 to 2023 five-year estimates | 95% CI 39.3% to 42.3% | 76,496 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| evr2_county | 29.3% | percent of the variance of the measures | share of the variance of the standardized measures carried by the second principal component in the county box, population-weighted correlation matrix, interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of the counties in the county frame (all measures from the county's own ACS estimate) | ACS 2019 to 2023 five-year estimates | 95% CI 25.1% to 32.8% | 3,205 counties | ACS 2019-2023 county population | American Community Survey |   |
| evr2_tract | 18.3% | percent of the variance of the measures | share of the variance of the standardized measures carried by the second principal component in the tract box, population-weighted correlation matrix, interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of the tracts in the tract frame (a published ADI rank, a population and all measures) | ACS 2019 to 2023 five-year estimates | 95% CI 16.4% to 20.0% | 76,496 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| extreme_share | 96.3% | percent of residents | share of residents who live in a county whose own percentile lies farther from the 50th percentile than the mean of its tracts' percentiles, interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of the counties in the county frame (all measures from the county's own ACS estimate) | ACS 2019 to 2023 five-year estimates | 95% CI 94.1% to 98.0% | 3,203 counties | ACS 2019-2023 county population | American Community Survey |   |
| extreme_share_unw | 96.7% | percent of counties | share of counties, each counted once, whose own percentile lies farther from the 50th percentile than the mean of its tracts' percentiles, interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of the counties in the county frame (all measures from the county's own ACS estimate) | ACS 2019 to 2023 five-year estimates | 95% CI 96.0% to 97.3% | 3,203 counties | none (each county counts once) | American Community Survey |   |
| frame_share | 92.9% | percent of residents | share of those residents who live in tracts of the tract frame | residents of tracts with a published ADI rank and a population above zero | ACS 2019 to 2023 five-year estimates |   |   | ACS 2019-2023 tract population (full count) | American Community Survey |   |
| frame_share_all | 92.4% | percent of residents | share of those residents who live in tracts of the tract frame | residents of all tracts with an ACS population above zero | ACS 2019 to 2023 five-year estimates |   |   | ACS 2019-2023 tract population (full count) | American Community Survey |   |
| gap_cut_1 | 25 | percentile points | gap between a resident's tract percentile and county percentile above which the gap is called large; a definition | residents of the tract frame | ACS 2019 to 2023 five-year estimates |   |   | none (definition) | American Community Survey |   |
| gap_cut_2 | 40 | percentile points | a second, larger gap cutoff; a definition | residents of the tract frame | ACS 2019 to 2023 five-year estimates |   |   | none (definition) | American Community Survey |   |
| high_cut | 80th | percentile among residents | tract percentile at or above which a tract belongs to the most deprived fifth of the tract ranking; a definition | tracts of the tract frame | ACS 2019 to 2023 five-year estimates |   |   | none (definition) | American Community Survey |   |
| load_crowd_county | −0.05 | weight on the standardized measure (first eigenvector, unit length) | weight of the crowding in the first principal component of the county box, signed so that the poverty rate has a positive weight, interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of the counties in the county frame (all measures from the county's own ACS estimate) | ACS 2019 to 2023 five-year estimates | 95% CI −0.12 to 0.03 | 3,205 counties | ACS 2019-2023 county population | American Community Survey |   |
| load_crowd_tract | 0.14 | weight on the standardized measure (first eigenvector, unit length) | weight of the crowding in the first principal component of the tract box, signed so that the poverty rate has a positive weight, interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of the tracts in the tract frame (a published ADI rank, a population and all measures) | ACS 2019 to 2023 five-year estimates | 95% CI 0.12 to 0.15 | 76,496 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| load_disp_county | 0.34 | weight on the standardized measure (first eigenvector, unit length) | weight of the income disparity in the first principal component of the county box, signed so that the poverty rate has a positive weight, interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of the counties in the county frame (all measures from the county's own ACS estimate) | ACS 2019 to 2023 five-year estimates | 95% CI 0.31 to 0.36 | 3,205 counties | ACS 2019-2023 county population | American Community Survey |   |
| load_disp_tract | 0.27 | weight on the standardized measure (first eigenvector, unit length) | weight of the income disparity in the first principal component of the tract box, signed so that the poverty rate has a positive weight, interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of the tracts in the tract frame (a published ADI rank, a population and all measures) | ACS 2019 to 2023 five-year estimates | 95% CI 0.26 to 0.27 | 76,496 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| load_edu_county | 0.09 | weight on the standardized measure (first eigenvector, unit length) | weight of the share of adults with under 9 years of schooling in the first principal component of the county box, signed so that the poverty rate has a positive weight, interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of the counties in the county frame (all measures from the county's own ACS estimate) | ACS 2019 to 2023 five-year estimates | 95% CI 0.01 to 0.15 | 3,205 counties | ACS 2019-2023 county population | American Community Survey |   |
| load_edu_tract | 0.21 | weight on the standardized measure (first eigenvector, unit length) | weight of the share of adults with under 9 years of schooling in the first principal component of the tract box, signed so that the poverty rate has a positive weight, interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of the tracts in the tract frame (a published ADI rank, a population and all measures) | ACS 2019 to 2023 five-year estimates | 95% CI 0.20 to 0.22 | 76,496 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| load_own_county | 0.03 | weight on the standardized measure (first eigenvector, unit length) | weight of the homeownership rate in the first principal component of the county box, signed so that the poverty rate has a positive weight, interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of the counties in the county frame (all measures from the county's own ACS estimate) | ACS 2019 to 2023 five-year estimates | 95% CI −0.06 to 0.10 | 3,205 counties | ACS 2019-2023 county population | American Community Survey |   |
| load_own_tract | −0.18 | weight on the standardized measure (first eigenvector, unit length) | weight of the homeownership rate in the first principal component of the tract box, signed so that the poverty rate has a positive weight, interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of the tracts in the tract frame (a published ADI rank, a population and all measures) | ACS 2019 to 2023 five-year estimates | 95% CI −0.20 to −0.16 | 76,496 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| maup_rc_le | −0.76 | correlation coefficient | population-weighted correlation across counties of the county index percentile among residents and the population-weighted mean of the tract values of tract life expectancy at birth, interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of counties that have tract life expectancy at birth | life expectancy for deaths in 2010 to 2015; index from ACS 2019 to 2023 | 95% CI −0.79 to −0.72 | 3,101 counties | ACS 2019-2023 county population | USALEEP life expectancy |   |
| maup_rc_mob | −0.59 | correlation coefficient | population-weighted correlation across counties of the county index percentile among residents and the population-weighted mean of the tract values of the mean adult household income rank of children raised by parents at the 25th income percentile, interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of counties that have the mean adult household income rank of children raised by parents at the 25th income percentile | children born 1978 to 1983, household income measured around 2014 to 2015; index from ACS 2019 to 2023 | 95% CI −0.64 to −0.54 | 3,155 counties | ACS 2019-2023 county population | Opportunity Atlas |   |
| maup_rc_ownhome | +0.06 | correlation coefficient | population-weighted correlation across counties of the county index percentile among residents and the population-weighted mean of the tract values of the homeownership rate, interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of counties that have the homeownership rate | ACS 2019 to 2023 five-year estimates | 95% CI −0.06 to 0.17 | 3,203 counties | ACS 2019-2023 county population | American Community Survey |   |
| maup_rc_poverty | +0.75 | correlation coefficient | population-weighted correlation across counties of the county index percentile among residents and the population-weighted mean of the tract values of the poverty rate, interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of counties that have the poverty rate | ACS 2019 to 2023 five-year estimates | 95% CI 0.72 to 0.78 | 3,203 counties | ACS 2019-2023 county population | American Community Survey |   |
| maup_rt_le | −0.62 | correlation coefficient | population-weighted correlation across tracts of the index percentile among residents and tract life expectancy at birth, interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of tracts in the tract frame that have tract life expectancy at birth | life expectancy for deaths in 2010 to 2015; index from ACS 2019 to 2023 | 95% CI −0.64 to −0.60 | 71,874 tracts | ACS 2019-2023 tract population | USALEEP life expectancy |   |
| maup_rt_mob | −0.58 | correlation coefficient | population-weighted correlation across tracts of the index percentile among residents and the mean adult household income rank of children raised by parents at the 25th income percentile, interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of tracts in the tract frame that have the mean adult household income rank of children raised by parents at the 25th income percentile | children born 1978 to 1983, household income measured around 2014 to 2015; index from ACS 2019 to 2023 | 95% CI −0.60 to −0.56 | 75,731 tracts | ACS 2019-2023 tract population | Opportunity Atlas |   |
| maup_rt_ownhome | −0.43 | correlation coefficient | population-weighted correlation across tracts of the index percentile among residents and the homeownership rate, interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of tracts in the tract frame that have the homeownership rate | ACS 2019 to 2023 five-year estimates | 95% CI −0.46 to −0.40 | 76,496 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| maup_rt_poverty | +0.78 | correlation coefficient | population-weighted correlation across tracts of the index percentile among residents and the poverty rate, interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of tracts in the tract frame that have the poverty rate | ACS 2019 to 2023 five-year estimates | 95% CI 0.78 to 0.79 | 76,496 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| maup_ru_le | −0.56 | correlation coefficient | correlation across counties, each counted once, of the county index percentile among residents and the population-weighted mean of the tract values of tract life expectancy at birth, interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | counties that have tract life expectancy at birth | life expectancy for deaths in 2010 to 2015; index from ACS 2019 to 2023 | 95% CI −0.59 to −0.54 | 3,101 counties | none (each county counts once) | USALEEP life expectancy |   |
| maup_ru_mob | −0.36 | correlation coefficient | correlation across counties, each counted once, of the county index percentile among residents and the population-weighted mean of the tract values of the mean adult household income rank of children raised by parents at the 25th income percentile, interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | counties that have the mean adult household income rank of children raised by parents at the 25th income percentile | children born 1978 to 1983, household income measured around 2014 to 2015; index from ACS 2019 to 2023 | 95% CI −0.38 to −0.33 | 3,155 counties | none (each county counts once) | Opportunity Atlas |   |
| maup_ru_ownhome | −0.07 | correlation coefficient | correlation across counties, each counted once, of the county index percentile among residents and the population-weighted mean of the tract values of the homeownership rate, interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | counties that have the homeownership rate | ACS 2019 to 2023 five-year estimates | 95% CI −0.11 to −0.03 | 3,203 counties | none (each county counts once) | American Community Survey |   |
| maup_ru_poverty | +0.65 | correlation coefficient | correlation across counties, each counted once, of the county index percentile among residents and the population-weighted mean of the tract values of the poverty rate, interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | counties that have the poverty rate | ACS 2019 to 2023 five-year estimates | 95% CI 0.63 to 0.66 | 3,203 counties | none (each county counts once) | American Community Survey |   |
| mean_cty_unit | 26.5 | percentile among counties | population-weighted mean over residents of the percentile of their county among counties, each county counted once, interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of the tracts in the tract frame (a published ADI rank, a population and all measures) | ACS 2019 to 2023 five-year estimates | 95% CI 24.5 to 28.8 | 76,495 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| med_disp | 16.0 | percentile points | median absolute difference between the percentile of a resident's tract and the percentile of the resident's county, percentiles among residents (a unit's percentile is the share of residents of its box who live in units that score lower, with half of any tied units counted), interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of the tracts in the tract frame (a published ADI rank, a population and all measures) | ACS 2019 to 2023 five-year estimates | 95% CI 15.2 to 17.0 | 76,495 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| med_disp_unit | 19.4 | percentile points | median absolute difference between the percentile of a resident's tract and the percentile of the resident's county, percentiles among tracts and among counties (a unit's percentile is the share of the units of its box that score lower, with half of any tied units counted), interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of the tracts in the tract frame (a published ADI rank, a population and all measures) | ACS 2019 to 2023 five-year estimates | 95% CI 18.7 to 20.2 | 76,495 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| mid_hi | 60th | percentile among residents | end of the range of county percentiles that makes up the middle fifth of the county ranking; a definition | counties of the county frame | ACS 2019 to 2023 five-year estimates |   |   | none (definition) | American Community Survey |   |
| mid_lo | 40th | percentile among residents | end of the range of county percentiles that makes up the middle fifth of the county ranking; a definition | counties of the county frame | ACS 2019 to 2023 five-year estimates |   |   | none (definition) | American Community Survey |   |
| mid_n | 159 | counties | count of counties | counties with an own percentile in the middle fifth and at least the spread floor of tracts in the frame | ACS 2019 to 2023 five-year estimates |   |   | none (full count) | American Community Survey |   |
| mid_pop_share | 17.7% | percent of residents | share of residents of the tract frame who live in those counties | residents of the tracts in the tract frame (a published ADI rank, a population and all measures) | ACS 2019 to 2023 five-year estimates |   |   | ACS 2019-2023 tract population (full count) | American Community Survey |   |
| n_big | 30 | counties | number of counties drawn in the dot plot, chosen by population and not by any result; a definition | the most populous counties in the county frame | ACS 2019 to 2023 five-year estimates |   |   | none (definition) | American Community Survey |   |
| n_close | 8 | measures | number of measures whose weight in the first component differs by less than the cutoff between the tract box and the county box | the measures of the index | ACS 2019 to 2023 five-year estimates |   |   | none (full count) | American Community Survey |   |
| n_counties | 3,205 | counties | count of units in the county frame | counties, Connecticut planning regions and Puerto Rico municipios with a population above zero and all measures in the county ACS pull | ACS 2019 to 2023 five-year estimates |   |   | none (full count) | American Community Survey |   |
| n_county_rows | 3,222 | counties | count of rows in the county ACS pull | counties, Connecticut planning regions and Puerto Rico municipios in the county ACS pull | ACS 2019 to 2023 five-year estimates |   |   | none (full count) | American Community Survey |   |
| n_cover_low | 605 | counties | count of counties | counties in both frames whose tracts in the tract frame hold less than the coverage cutoff of residents | ACS 2019 to 2023 five-year estimates |   |   | none (full count) | American Community Survey |   |
| n_dollar | 4 | census measures | count of dollar-denominated measures; a definition | measures in the deprivation index of this chapter that are dollar amounts (family income, home value, rent and mortgage cost) | ACS 2019 to 2023 five-year estimates |   |   | none (definition) | American Community Survey |   |
| n_measures | 14 | census measures | count of measures; the 15 measures the chapter on what the deprivation index measures rebuilds less the share of single-parent households; a definition | measures in the deprivation index of this chapter | ACS 2019 to 2023 five-year estimates |   |   | none (definition) | American Community Survey |   |
| n_pr_tracts | 779 | tracts | count of tracts | Puerto Rico tracts in the tract frame | ACS 2019 to 2023 five-year estimates |   |   | none (full count) | American Community Survey |   |
| n_tracts | 76,496 | tracts | count of tracts in the tract frame | 2020 census tracts with a published ADI rank, a population above zero and all measures | ACS 2019 to 2023 five-year estimates |   |   | none (full count) | American Community Survey |   |
| n_unranked | 1,691 | tracts | count of tracts | 2020 census tracts in the staged ADI file with no published rank | ACS 2019 to 2023 five-year estimates |   |   | none (full count) | Area Deprivation Index |   |
| ny_cover | 74.2% | percent of a county's residents | share of the county's residents who live in tracts of the tract frame | residents of New York County | ACS 2019 to 2023 five-year estimates |   |   | ACS 2019-2023 county population (full count) | American Community Survey |   |
| p_hi | 95th | percentile of residents | upper end of the range drawn for each county; a definition | residents of a county, ordered by the percentile of their tract | ACS 2019 to 2023 five-year estimates |   |   | none (definition) | American Community Survey |   |
| p_lo | 5th | percentile of residents | lower end of the range drawn for each county; a definition | residents of a county, ordered by the percentile of their tract | ACS 2019 to 2023 five-year estimates |   |   | none (definition) | American Community Survey |   |
| pop_cover_low | 25.5% | percent of residents | share of residents of counties in both frames who live in a county where the tracts of the frame hold less than the coverage cutoff of residents | residents of the counties in the county frame (all measures from the county's own ACS estimate) | ACS 2019 to 2023 five-year estimates |   |   | ACS 2019-2023 county population (full count) | American Community Survey |   |
| pop_tracts | 309.9 million | residents | sum of ACS tract population over the tract frame | residents of the tracts in the tract frame (a published ADI rank, a population and all measures) | ACS 2019 to 2023 five-year estimates |   |   | none (full count) | American Community Survey |   |
| pop_unranked | 2.1 million | residents | sum of ACS tract population | residents of tracts with no published ADI rank | ACS 2019 to 2023 five-year estimates |   |   | none (full count) | American Community Survey |   |
| pov_in | 12.5% | percent of persons below the poverty line | population-weighted mean of the tract poverty rate; a description of the whole frame, so no sampling interval | residents of the tracts in the tract frame (a published ADI rank, a population and all measures) | ACS 2019 to 2023 five-year estimates |   |   | ACS 2019-2023 tract population (full count) | American Community Survey |   |
| pov_out | 17.8% | percent of persons below the poverty line | population-weighted mean of the tract poverty rate; a description of the whole group, so no sampling interval | residents of tracts outside the tract frame that have a population and a poverty rate | ACS 2019 to 2023 five-year estimates |   |   | ACS 2019-2023 tract population (full count) | American Community Survey |   |
| r_own_mean | 0.96 | correlation coefficient | population-weighted correlation across counties of the county's own percentile among residents and the population-weighted mean of its tracts' percentiles among residents, interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of the counties in the county frame (all measures from the county's own ACS estimate) | ACS 2019 to 2023 five-year estimates | 95% CI 0.95 to 0.97 | 3,203 counties | ACS 2019-2023 county population | American Community Survey |   |
| r_own_mean_cov | 0.96 | correlation coefficient | population-weighted correlation across those counties of the county's own percentile and the mean of its tracts' percentiles, interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of counties where tracts of the frame hold at least the coverage cutoff of residents | ACS 2019 to 2023 five-year estimates | 95% CI 0.95 to 0.97 | 2,598 counties | ACS 2019-2023 county population | American Community Survey |   |
| r_own_mean_unw | 0.96 | correlation coefficient | correlation across counties, each counted once, of the county's own percentile and the mean of its tracts' percentiles, interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of the counties in the county frame (all measures from the county's own ACS estimate) | ACS 2019 to 2023 five-year estimates | 95% CI 0.95 to 0.96 | 3,203 counties | none (each county counts once) | American Community Survey |   |
| r_rank | 0.99 | rank correlation coefficient | population-weighted correlation across counties of the county's own percentile and the percentile of the population-weighted mean of its tracts' index scores (the county ranking implied by the tract file), interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of the counties in the county frame (all measures from the county's own ACS estimate) | ACS 2019 to 2023 five-year estimates | 95% CI 0.98 to 0.99 | 3,203 counties | ACS 2019-2023 county population | American Community Survey |   |
| r_share | 0.75 | correlation coefficient | population-weighted correlation across counties of the county's own percentile and the share of its residents in the frame who live in tracts at or above the high cutoff, interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of the counties in the county frame (all measures from the county's own ACS estimate) | ACS 2019 to 2023 five-year estimates | 95% CI 0.73 to 0.77 | 3,203 counties | ACS 2019-2023 county population | American Community Survey |   |
| rho_adi | +0.78 | rank correlation coefficient | population-weighted correlation of the percentile ranks of the tract index of this chapter and of the published tract ADI, interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of the tracts in the tract frame (a published ADI rank, a population and all measures) | ACS 2019 to 2023 five-year estimates | 95% CI 0.73 to 0.82 | 76,496 tracts | ACS 2019-2023 tract population | Area Deprivation Index |   |
| rho_fit_unw | +0.988 | rank correlation coefficient | correlation across counties, each counted once, of the county percentile among counties when the component is fitted with population weights and when it is fitted counting each county once, interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of the counties in the county frame (all measures from the county's own ACS estimate) | ACS 2019 to 2023 five-year estimates | 95% CI 0.986 to 0.990 | 3,205 counties | none (each county counts once) | American Community Survey |   |
| rho_formula_hv | −0.99 | rank correlation coefficient | population-weighted correlation, across counties, of the percentile ranks of the published ADI formula applied to the county's 14 measures and of median home value (negative because the formula scores low home values as more deprived), interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of the counties in the county frame (all measures from the county's own ACS estimate) | ACS 2019 to 2023 five-year estimates | 95% CI −0.99 to −0.99 | 3,205 counties | ACS 2019-2023 county population | American Community Survey |   |
| rho_formula_idx | +0.90 | rank correlation coefficient | population-weighted correlation, across counties, of the percentile ranks of the published ADI formula applied to the county's 14 measures and of the county index of this chapter, interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of the counties in the county frame (all measures from the county's own ACS estimate) | ACS 2019 to 2023 five-year estimates | 95% CI 0.86 to 0.92 | 3,205 counties | ACS 2019-2023 county population | American Community Survey |   |
| rho_single | 0.996 | rank correlation coefficient | population-weighted correlation of the percentile ranks of the tract index of this chapter and of the same index with the staged single-parent measure added as a fifteenth measure, on tracts that have the staged value, interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of the tracts in the tract frame (a published ADI rank, a population and all measures) | ACS 2019 to 2023 five-year estimates | 95% CI 0.996 to 0.996 | 76,496 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| share25 | 32.7% | percent of residents | share of residents whose tract percentile differs from their county's percentile by more than the first cutoff, percentiles among residents, interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of the tracts in the tract frame (a published ADI rank, a population and all measures) | ACS 2019 to 2023 five-year estimates | 95% CI 30.9% to 34.4% | 76,495 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| share25_cov | 32.4% | percent of residents | share of those residents whose tract percentile differs from their county's percentile by more than the first cutoff, percentiles among residents, interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of the tract frame who live in counties where the frame holds at least the coverage cutoff of residents | ACS 2019 to 2023 five-year estimates | 95% CI 30.3% to 34.5% | 58,073 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| share25_unit | 40.7% | percent of residents | share of residents whose tract percentile differs from their county's percentile by more than the first cutoff, percentiles among tracts and among counties, interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of the tracts in the tract frame (a published ADI rank, a population and all measures) | ACS 2019 to 2023 five-year estimates | 95% CI 39.1% to 42.2% | 76,495 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| share40 | 14.1% | percent of residents | share of residents whose tract percentile differs from their county's percentile by more than the second cutoff, percentiles among residents, interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of the tracts in the tract frame (a published ADI rank, a population and all measures) | ACS 2019 to 2023 five-year estimates | 95% CI 12.9% to 15.4% | 76,495 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| share40_unit | 22.8% | percent of residents | share of residents whose tract percentile differs from their county's percentile by more than the second cutoff, percentiles among tracts and among counties, interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of the tracts in the tract frame (a published ADI rank, a population and all measures) | ACS 2019 to 2023 five-year estimates | 95% CI 21.1% to 24.6% | 76,495 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| share_high_med | 16.6% | percent of residents | 50th percentile, over those counties weighted by residents of the frame, of the share of a county's residents who live in tracts at or above the high cutoff, interval from a bootstrap over those counties | residents of counties with an own percentile in the middle fifth and at least the spread floor of tracts in the frame | ACS 2019 to 2023 five-year estimates | 95% CI 14.5% to 20.4% | 159 counties | ACS 2019-2023 tract population | American Community Survey |   |
| share_high_med_unw | 11.6% | percent of residents | 50th percentile, over those counties each counted once, of the share of a county's residents who live in tracts at or above the high cutoff, interval from a bootstrap over those counties | residents of counties with an own percentile in the middle fifth and at least the spread floor of tracts in the frame | ACS 2019 to 2023 five-year estimates | 95% CI 9.9% to 13.2% | 159 counties | none (each county counts once) | American Community Survey |   |
| share_high_p5 | 4.3% | percent of residents | 5th percentile, over those counties weighted by residents of the frame, of the share of a county's residents who live in tracts at or above the high cutoff, interval from a bootstrap over those counties | residents of counties with an own percentile in the middle fifth and at least the spread floor of tracts in the frame | ACS 2019 to 2023 five-year estimates | 95% CI 2.3% to 5.8% | 159 counties | ACS 2019-2023 tract population | American Community Survey |   |
| share_high_p5_unw | 0.0% | percent of residents | 5th percentile, over those counties each counted once, of the share of a county's residents who live in tracts at or above the high cutoff, interval from a bootstrap over those counties | residents of counties with an own percentile in the middle fifth and at least the spread floor of tracts in the frame | ACS 2019 to 2023 five-year estimates | 95% CI 0.0% to 2.1% | 159 counties | none (each county counts once) | American Community Survey |   |
| share_high_p95 | 25.3% | percent of residents | 95th percentile, over those counties weighted by residents of the frame, of the share of a county's residents who live in tracts at or above the high cutoff, interval from a bootstrap over those counties | residents of counties with an own percentile in the middle fifth and at least the spread floor of tracts in the frame | ACS 2019 to 2023 five-year estimates | 95% CI 23.9% to 27.1% | 159 counties | ACS 2019-2023 tract population | American Community Survey |   |
| share_high_p95_unw | 23.9% | percent of residents | 95th percentile, over those counties each counted once, of the share of a county's residents who live in tracts at or above the high cutoff, interval from a bootstrap over those counties | residents of counties with an own percentile in the middle fifth and at least the spread floor of tracts in the frame | ACS 2019 to 2023 five-year estimates | 95% CI 21.9% to 25.5% | 159 counties | none (each county counts once) | American Community Survey |   |
| slope_mean | 0.57 | points of average tract percentile per point of county percentile | population-weighted slope across counties of the population-weighted mean of tract percentiles among residents on the county's own percentile among residents, interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of the counties in the county frame (all measures from the county's own ACS estimate) | ACS 2019 to 2023 five-year estimates | 95% CI 0.55 to 0.59 | 3,203 counties | ACS 2019-2023 county population | American Community Survey |   |
| slope_mean_unw | 0.68 | points of average tract percentile per point of county percentile | slope across counties, each counted once, of the population-weighted mean of tract percentiles among residents on the county's own percentile among residents, interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of the counties in the county frame (all measures from the county's own ACS estimate) | ACS 2019 to 2023 five-year estimates | 95% CI 0.68 to 0.69 | 3,203 counties | none (each county counts once) | American Community Survey |   |
| sp_ratio | 1.52 | ratio of county value to resident-weighted mean of tract values | median over counties of the county core file's share of single-parent families divided by the population-weighted mean of the staged tract file's share of single-parent households; a description of the two definitions, so no sampling interval | counties outside Utah and Puerto Rico with a staged tract value of the single-parent measure | ACS 2019 to 2023 five-year estimates |   | 3,095 counties | none (each county counts once) | American Community Survey |   |
| spread_floor | 20 | tracts | smallest number of tracts in the frame a county needs for its spread to enter the median; a definition | counties whose spread of tract percentiles is summarized | ACS 2019 to 2023 five-year estimates |   |   | none (definition) | American Community Survey |   |
| spread_floor_high | 50 | tracts | a higher floor on the number of tracts in the frame, used to show how the median spread moves with the floor; a definition | counties whose spread of tract percentiles is summarized | ACS 2019 to 2023 five-year estimates |   |   | none (definition) | American Community Survey |   |
| spread_floor_low | 10 | tracts | a lower floor on the number of tracts in the frame, used to show how the median spread moves with the floor; a definition | counties whose spread of tract percentiles is summarized | ACS 2019 to 2023 five-year estimates |   |   | none (definition) | American Community Survey |   |
| spread_median | 79.0 | percentile points | median over those counties, counties weighted by residents of the frame, of the range from the 5th to the 95th percentile of the county's tract percentiles (tracts weighted by residents), interval from a bootstrap over those counties | residents of counties with at least 20 tracts in the tract frame | ACS 2019 to 2023 five-year estimates | 95% CI 75.5 to 80.8 | 728 counties | ACS 2019-2023 tract population | American Community Survey |   |
| spread_median_10 | 76.5 | percentile points | median over those counties, counties weighted by residents of the frame, of the range from the 5th to the 95th percentile of the county's tract percentiles (tracts weighted by residents), interval from a bootstrap over those counties | residents of counties with at least 10 tracts in the tract frame | ACS 2019 to 2023 five-year estimates | 95% CI 72.9 to 79.8 | 1,296 counties | ACS 2019-2023 tract population | American Community Survey |   |
| spread_median_10_unw | 58.5 | percentile points | median over those counties, each counted once, of the range from the 5th to the 95th percentile of the county's tract percentiles (tracts weighted by residents), interval from a bootstrap over those counties | counties with at least 10 tracts in the tract frame | ACS 2019 to 2023 five-year estimates | 95% CI 57.4 to 59.6 | 1,296 counties | none (each county counts once) | American Community Survey |   |
| spread_median_50 | 81.0 | percentile points | median over those counties, counties weighted by residents of the frame, of the range from the 5th to the 95th percentile of the county's tract percentiles (tracts weighted by residents), interval from a bootstrap over those counties | residents of counties with at least 50 tracts in the tract frame | ACS 2019 to 2023 five-year estimates | 95% CI 79.7 to 83.6 | 320 counties | ACS 2019-2023 tract population | American Community Survey |   |
| spread_median_50_unw | 76.7 | percentile points | median over those counties, each counted once, of the range from the 5th to the 95th percentile of the county's tract percentiles (tracts weighted by residents), interval from a bootstrap over those counties | counties with at least 50 tracts in the tract frame | ACS 2019 to 2023 five-year estimates | 95% CI 74.6 to 78.6 | 320 counties | none (each county counts once) | American Community Survey |   |
| spread_median_unw | 67.4 | percentile points | median over those counties, each counted once, of the range from the 5th to the 95th percentile of the county's tract percentiles (tracts weighted by residents), interval from a bootstrap over those counties | counties with at least 20 tracts in the tract frame | ACS 2019 to 2023 five-year estimates | 95% CI 66.0 to 68.6 | 728 counties | none (each county counts once) | American Community Survey |   |
| spread_n | 728 | counties | count of counties | counties with at least the spread floor of tracts in the tract frame | ACS 2019 to 2023 five-year estimates |   |   | none (full count) | American Community Survey |   |
| spread_pop_share | 81.8% | percent of residents | share of residents of the tract frame who live in a county with at least the spread floor of tracts | residents of the tracts in the tract frame (a published ADI rank, a population and all measures) | ACS 2019 to 2023 five-year estimates |   |   | ACS 2019-2023 tract population (full count) | American Community Survey |   |
| top_cres | 29.1 | percentile among residents | percentile among residents of the county index built from the county's own ACS estimates | residents of Los Angeles County, the most populous county | ACS 2019 to 2023 five-year estimates |   |   | ACS 2019-2023 county population (single county) | American Community Survey |   |
| top_cunit | 8.0 | percentile among counties | percentile among counties, each counted once, of the same county index | residents of Los Angeles County, the most populous county | ACS 2019 to 2023 five-year estimates |   |   | none (single county) | American Community Survey |   |
| top_far25 | 53.0% | percent of residents | share of the county's residents whose tract percentile differs from the county's own percentile by more than the first cutoff | residents of Los Angeles County, the most populous county | ACS 2019 to 2023 five-year estimates |   | 2,226 tracts | ACS 2019-2023 tract population (single county) | American Community Survey |   |
| top_p5 | 2.0 | percentile among residents | 5th percentile of the county's tract percentiles, tracts weighted by residents; a description of one county's tracts | residents of Los Angeles County, the most populous county | ACS 2019 to 2023 five-year estimates |   | 2,226 tracts | ACS 2019-2023 tract population (single county) | American Community Survey |   |
| top_p95 | 94.2 | percentile among residents | 95th percentile of the county's tract percentiles, tracts weighted by residents; a description of one county's tracts | residents of Los Angeles County, the most populous county | ACS 2019 to 2023 five-year estimates |   | 2,226 tracts | ACS 2019-2023 tract population (single county) | American Community Survey |   |
| tract_conv_diff | 2.6 | percentile points | population-weighted mean absolute difference between a tract's percentile among residents and its percentile among tracts; a description of the whole frame, so no sampling interval | residents of the tracts in the tract frame (a published ADI rank, a population and all measures) | ACS 2019 to 2023 five-year estimates |   | 76,496 tracts | ACS 2019-2023 tract population (full count) | American Community Survey |   |
| within_share | 64.6% | percent of the variance of the tract percentile | share of the population-weighted variance of the tract percentile (among residents) that lies between tracts of the same county, one minus the between-county share, interval from a bootstrap over counties (Connecticut planning regions stand in for counties) | residents of the tracts in the tract frame (a published ADI rank, a population and all measures) | ACS 2019 to 2023 five-year estimates | 95% CI 60.5% to 68.7% | 3,203 counties | ACS 2019-2023 tract population | American Community Survey |   |

## Definitions

- Principal components: A method that combines many correlated measures into a few summary scores, ordered by how much of the variation among places each captures. The first component is the single weighted combination that captures the most.
- Area Deprivation Index (ADI): A score that ranks neighborhoods by income, education, employment and housing conditions drawn from the American Community Survey. In this book a higher national percentile means a more deprived neighborhood.
- USALEEP: The U.S. Small-area Life Expectancy Estimates Project, a National Center for Health Statistics release of life expectancy at birth for most census tracts. It covers deaths from 2010 to 2015 only and has not been updated.
- MAUP (modifiable areal unit problem): The same data can show different patterns depending on how the map is divided, for example into tracts, counties or states.
- Ecological inference: Drawing conclusions about individuals from data about places or groups. A pattern across counties need not hold for the people inside them.

## Methods

This chapter builds one deprivation index with one recipe in two boxes, 76,496 census tracts and 3,205 counties (with Connecticut's planning regions and Puerto Rico's municipios), and ranks each unit among the residents of its box. The index is the first principal component of 14 census measures from ACS 2019-2023 and is not the Neighborhood Atlas ADI, which ranks block groups. Tracts need a published ADI rank to enter the tract box, and the single-parent measure is left out of both boxes because the tract and county files define it differently. Life expectancy (deaths in 2010 to 2015) and upward mobility (children born 1978 to 1983) describe earlier periods than the ACS measures, and the ACS files carry no margins of error, so the only intervals come from resampling whole counties.

### Which tracts and counties are in each box

The comparison uses two boxes. The tract box holds the 76,496 2020 census tracts with a published ADI rank, an ACS 2019-2023 population above zero and all 14 measures (the 15 that the chapter on what the deprivation index measures rebuilds without the share of single-parent households), out of 85,396 tracts in the staged ADI file; they hold 92.9 percent of the residents of tracts with a rank. The county box holds the 3,205 of the 3,222 rows of the county ACS pull (counties, Connecticut's nine planning regions and Puerto Rico's municipios) that have a population above zero and all 14 measures; 17 rows lack a measure.

Tracts outside the tract frame are those without a published ADI rank or with a missing measure (mostly a missing median home value, rent or mortgage cost); Puerto Rico's tracts and municipios are inside both boxes.

### Measures and the single-parent measure

The single-parent measure is left out of both boxes because the staged tract file and the county core file define it differently (a share of households and a share of families with own children), so its county value is about 1.52 times the resident-weighted mean of its tracts' values outside Utah and Puerto Rico; with the staged tract value added as a fifteenth measure the tract index ranks tracts with a rank correlation of 0.996 with the 14-measure index.

Tract measures come from the staged ACS tract file and a Census API pull archived under data/raw/acs/deprivation_components_2019_2023; county measures come from the same pull and from the county core file, so a county's median is its own estimate and not an average of tract medians. b2.load has no loader for the pull, so the chapter registers its three files and reads them through load_staged. The measures, their substitutions and Singh's income disparity are those of the chapter on what the deprivation index measures.

The published formula is applied with Singh's coefficients (Singh 2003, as printed in Petterson 2023, Table 1) to the 14 measures in their own units; adding the single-parent term from the county core file changes the county ranking by less than a rank correlation of 0.9999.

### How the index and percentiles are computed

In each box the measures are standardized with the box's population-weighted means and standard deviations, the first principal component is the leading eigenvector of the population-weighted correlation matrix (the weights are the residents of each tract or each county), signed so that the poverty rate has a positive weight, and a unit's percentile is the share of residents of its box who live in units that score lower, counting half of any tied units.

The first edition and the Neighborhood Atlas rank units instead; that convention is reported as 'among tracts and counties'.

Explains Table 1.

### How counties and tracts are compared

Counties in both frames are compared with the population-weighted mean and the 5th and 95th percentiles of the percentiles of their tracts (weighted by residents). A county enters the restricted comparison when tracts of the tract frame hold at least 90 percent of its residents, and the spread summary needs at least 20 tracts in the frame; floors of 10 and 50 are reported beside it.

The median absolute gap is read from bins of 0.1 percentile points. The variance split is the population-weighted variance of the tract percentile within counties over its total.

Explains Table 2 and Table 3.

### Outcomes and correlations

Life expectancy is USALEEP for deaths in 2010 to 2015 on 2020 tracts (institutional tracts dropped), upward mobility is the Opportunity Atlas rank of children of parents at the 25th income percentile, and poverty and homeownership are the ACS tract rates; a county's value of each is the mean of its tracts' values weighted by residents.

Correlations are Pearson correlations of the index percentile and the measure, weighted by residents of the tract or county, or counting each county once.

Explains Table 4.

### Uncertainty

Intervals are 95% percentile intervals from 2,000 draws of whole counties with replacement (seed 12345; Connecticut's planning regions stand in for counties; in the county box each county is a draw), with percentiles held at their full-box values.

The ACS files carry no margins of error.

## What these data cannot show

The data describe places. A county percentile and a tract percentile are properties of places, and neither says how deprived any resident is. A poor household in a comfortable tract takes the tract's percentile, which is the problem of ecological inference. The correlations in the last section are across places and say nothing about the people inside them.

The ACS files carry no margins of error, and the medians of small tracts rest on few households, so part of the spread of tract percentiles within a county is sampling error, and the data cannot say how much. The tract box leaves out tracts with a missing measure, mostly a missing median home value, rent or mortgage cost, as Chapter 1.1, What the deprivation index measures (https://vishalsingh.org/zipcode/anatomy-of-deprivation) finds, and tracts without a published ADI rank. The frame holds 92.4% of all residents and 92.9% of the residents of ranked tracts, and 1,691 tracts with 2.1 million residents have no rank.

The tracts outside the frame are poorer, with a poverty rate of 17.8% against 12.5% inside. The missing tracts are concentrated in some counties. The frame holds 52.5% of the residents of Bronx County and 74.2% of New York County, so for those counties and others with low coverage the average and the spread of tracts describe only the tracts that remain.

The percentile convention is a choice. This chapter ranks by residents, county files usually rank counties, and the second distribution in Table 2 shows how far the choice moves a statement about residents. The analysis uses one recipe and one set of measures. The ADI formula would rank counties by home value, as the correlations above show, and the state ADI, which ranks units within a state, is not tested. The county index is not compared with published county indexes.

Life expectancy is a modeled estimate for deaths in 2010 to 2015 and upward mobility concerns children born in 1978 to 1983, so both are older than the ACS measures. They show how an index built from recent measures lines up with older outcomes at two units and do not show which unit is correct. That depends on the question. For a program run by counties the county box is the right one, and a finding in it is a finding about counties.

## References

1. Kind, Amy J. H.; Buckingham, William R. (2018). Making Neighborhood-Disadvantage Metrics Accessible: The Neighborhood Atlas. New England Journal of Medicine 378(26): 2456-2458. https://doi.org/10.1056/NEJMp1802313
2. Singh, Gopal K. (2003). Area Deprivation and Widening Inequalities in US Mortality, 1969-1998. American Journal of Public Health 93(7): 1137-1143. https://doi.org/10.2105/AJPH.93.7.1137
3. Petterson, Stephen (2023). Deciphering the Neighborhood Atlas Area Deprivation Index: the consequences of not standardizing. Health Affairs Scholar 1(5): qxad063. https://doi.org/10.1093/haschl/qxad063
4. Gehlke, C. E.; Biehl, Katherine (1934). Certain Effects of Grouping upon the Size of the Correlation Coefficient in Census Tract Material. Journal of the American Statistical Association 29(185A): 169-170. https://doi.org/10.1080/01621459.1934.10506247
5. Openshaw, Stan; Taylor, Peter J. (1979). A million or so correlation coefficients: three experiments on the modifiable areal unit problem. In N. Wrigley (ed.), Statistical Applications in the Spatial Sciences, London: Pion, 127-144.
6. Openshaw, Stan (1984). The Modifiable Areal Unit Problem. CATMOG 38, Norwich: Geo Books.
7. Fotheringham, A. Stewart; Wong, David W. S. (1991). The modifiable areal unit problem in multivariate statistical analysis. Environment and Planning A 23(7): 1025-1044. https://doi.org/10.1068/a231025
8. Robinson, W. S. (1950). Ecological Correlations and the Behavior of Individuals. American Sociological Review 15(3): 351-357. https://doi.org/10.2307/2087176
9. Subramanian, S. V.; Jones, Kelvyn; Kaddour, Afamia; Krieger, Nancy (2009). Revisiting Robinson: the perils of individualistic and ecologic fallacy. International Journal of Epidemiology 38(2): 342-360. https://doi.org/10.1093/ije/dyn359

## Provenance

- Pipeline version: 0.1.0
- Build script: book2/chapters/deprivation-at-two-zooms/build.py (sha256 5a4c9698560c73a0696d2cd0ca91ea3d557c6b2e4ed41d8f2de05eb6470fbb54)
- Built on: 2026-10-07
- Site commit: c2b1ff3309cb4a10833f884da91ac0057a66ad0a

Inputs:

| File | Size (bytes) | SHA-256, first 1 MB | SHA-256, last 1 MB |
| --- | ---: | --- | --- |
| book2/chapters/deprivation-at-two-zooms/analysis.py | 20,220 | 67e37af36734b4a1c3bbe42a5c6b7182278c60e1c1d40a8c88419bc7d15a2312 | 67e37af36734b4a1c3bbe42a5c6b7182278c60e1c1d40a8c88419bc7d15a2312 |
| book2/chapters/deprivation-at-two-zooms/zoom.py | 10,130 | fd5d3b7a97b1f6f0176308757883af61a7bcbac91ab0214679d0edb5b33837a9 | fd5d3b7a97b1f6f0176308757883af61a7bcbac91ab0214679d0edb5b33837a9 |
| data/analytic/tract_atlas.parquet | 39,659,925 | 29d5e4d943552d54aeb6ba92aadc939d9bd15bdea553c78357dbd516757be627 | 1e10d1ecf722340f23ee2212d4999a90250f85b65fa89b39e32db06d9472e22e |
| data/raw/acs/deprivation_components_2019_2023/county.parquet | 1,004,968 | c894ba6cf96378244f47cc4271a24bff087d15ba27f09e49a4ebc0a4786c6f9d | c894ba6cf96378244f47cc4271a24bff087d15ba27f09e49a4ebc0a4786c6f9d |
| data/raw/acs/deprivation_components_2019_2023/county_core.parquet | 498,878 | 8b177857027e73c1f404f7b62191ab78811e95ee54c4f07383e9e9e17da1a345 | 8b177857027e73c1f404f7b62191ab78811e95ee54c4f07383e9e9e17da1a345 |
| data/raw/acs/deprivation_components_2019_2023/tract.parquet | 13,197,793 | 5545cde2cbe96a8b56e468d65388248a241aa6bd300d3c9e325d08d3f86b11da | 5c8ec454d066988b0f1a01aaa6ed34ebfc05d162e01cf9c6bb01f9198cbaff12 |
| data/staged/adi/adi_atlas_tract2020_2023.parquet | 798,078 | 5e409bc212ff871def780fa27b4cf8b616d9da420b2b9f6455517eddcd8c1a95 | 5e409bc212ff871def780fa27b4cf8b616d9da420b2b9f6455517eddcd8c1a95 |
| data/staged/life_expectancy/usaleep_tract2020.parquet | 1,316,169 | 9c9c7f103791b75eddba74e68a89804efdf2486b6b66bb4bc02a46ba5bf8bc23 | 27317107be2465d9fd5e83271dea0cd6e915ee419e2802bedc28747d0a41d38f |
| data/staged/mobility/oa_module1_tract2020.parquet | 16,362,438 | 26cb468a319ad2e727c5eac733b4a47b77a83781ae189e1828e8c40bc0e3ace8 | 5702bad38b81b5ee0b3d01ba9d1ece7d23328985eafba269720a0af5ef6820d4 |
| data/staged/ses/acs5_tract2020_2019_2023.parquet | 7,232,424 | 5fe6f730a35225f14d5d8abfc6b9bce58b52e545f309a7eac74f1dae74d7a31c | 7b4a2475b3852d3f91804716a04566ef8ceadacec2be9fb85f5eb642078efac6 |
| data/staged/svi/svi_tract2020_2022.parquet | 17,740,415 | cb3860d8edc2aaf1f2a02211a58f7a0b80457eb6ad9a37b5ec8abb2d5da9da10 | e371a4a696d5bed05d54ad7c5b6ca2504508805b847ec415c2322be7f66ec2b2 |

Outputs:

| File | SHA-256 |
| --- | --- |
| facts.json | 7f3758bab217c2735b3bc82d6c289eb933ea84d5463a602691239a8809e28616 |
| figures/index_correlations_by_zoom.json | 0ff5eab542cdf383d8203bc162da8344c1d4d59ebacdcbb0fcfae0f617ba9075 |
| figures/large_counties_vs_tracts.json | 86ea6588b548146120843dfdd61ccb3d5d64fe0ed3892c927c9a9469ddb48613 |
| figures/tract_county_gap.json | e7d53f2929ac6d21808499d22c4059f906a5f7e1615795b211612ec2d7ec9eda |
| figures/weights_by_zoom.json | 121344b0910a1d45c98b15f11af2b953f8d9c7299ec9c077b121445950c5ef48 |
| meta.json | f8e02bdf438ebe9d9ae98ee7d82db74f657a13cab1a770bc08d76884298986e3 |
| refs.json | d02ff5e6d75e2ae1c46782537be57bd5b70663aca0444a9163dfdda9b45c4db3 |

## Machine formats

- Report (HTML): https://vishalsingh.org/zipcode/deprivation-at-two-zooms/report
- Report (Markdown): https://vishalsingh.org/zipcode/deprivation-at-two-zooms/report.md
- Every estimate (JSON): https://vishalsingh.org/zipcode/report/deprivation-at-two-zooms/facts.json
- Table 1, figure "weights_by_zoom" (JSON): https://vishalsingh.org/zipcode/report/deprivation-at-two-zooms/figures/weights_by_zoom.json
- Table 2, figure "tract_county_gap" (JSON): https://vishalsingh.org/zipcode/report/deprivation-at-two-zooms/figures/tract_county_gap.json
- Table 3, figure "large_counties_vs_tracts" (JSON): https://vishalsingh.org/zipcode/report/deprivation-at-two-zooms/figures/large_counties_vs_tracts.json
- Table 4, figure "index_correlations_by_zoom" (JSON): https://vishalsingh.org/zipcode/report/deprivation-at-two-zooms/figures/index_correlations_by_zoom.json
- Index of every report: https://vishalsingh.org/zipcode/llms.txt; all reports in one file: https://vishalsingh.org/zipcode/llms-full.txt

---

# Birthweight and preterm birth: report

Chapter 2.1 of ZIP Code Destiny, by Vishal Singh (NYU Stern). Generated from the chapter's data files; the reading version is at https://vishalsingh.org/zipcode/birthweight-and-preterm-birth. Table n of this report is figure n of the chapter.

## Identification

- Title: Birthweight and preterm birth
- Label: The starting line
- Chapter: 2.1 (Part II, Birth; An American life)
- Question: How do the shares of babies born at low birthweight or preterm change with county deprivation, and how much of the racial gap remains at every level of deprivation?
- Dek: Low birthweight was 7.4% of births in the least deprived tenth of counties and 9.7% in the most deprived, for births in 2018 to 2024 in counties of 100,000 or more residents.
- Grain: counties
- Status: published
- Published: 2026-10-09
- Updated: not revised
- Data sources:
  - CDC WONDER natality. CDC WONDER natality, county by birthweight and gestational age, births in 2018 to 2024; years: 2018-2024; weights: none (full counts of births); county shares weighted by births
  - Area Deprivation Index. Neighborhood Atlas Area Deprivation Index 2023 (v4.0.1), tract national rank, averaged to counties with 2019-2023 tract population; years: 2023 release (version 4.0.1; built on ACS 2019-2023); weights: 2019-2023 tract population
  - County Health Rankings. County Health Rankings 2024, low birthweight (births in 2016 to 2022), overall and for non-Hispanic Black, non-Hispanic white and Hispanic mothers; years: 2024 release; weights: births (recovered from published intervals for race groups)
  - CDC WONDER linked birth and infant death. CDC WONDER linked birth and infant death, mother's single race (any ethnicity), 2018 and 2019; years: 2018-2022; weights: none (crude rates from counts)
- Canonical URL: https://vishalsingh.org/zipcode/birthweight-and-preterm-birth
- Report: https://vishalsingh.org/zipcode/birthweight-and-preterm-birth/report
- Markdown: https://vishalsingh.org/zipcode/birthweight-and-preterm-birth/report.md
- BibTeX key: singh2026birthweightandpretermbirth

```bibtex
@misc{singh2026birthweightandpretermbirth,
  author = {Singh, Vishal},
  title = {Birthweight and preterm birth},
  year = {2026},
  url = {https://vishalsingh.org/zipcode/birthweight-and-preterm-birth},
  note = {In ZIP Code Destiny. Published 9 October 2026.}
}
```

## Principal estimates

The facts the dek and the opening prose cite, in order of appearance.

| Estimand | Population | Period | Value | Interval | n | Weights | Source | Fact id |
| --- | --- | --- | ---: | --- | ---: | --- | --- | --- |
| births-weighted mean of county shares (the pooled share), bootstrap interval over counties | live births in the least deprived tenth of the named counties by county deprivation rank | births in 2018 to 2024 | 7.4% | 95% CI 7.2% to 7.7% | 57 counties | births in each county | CDC WONDER natality | lbw_d1 |
| births-weighted mean of county shares (the pooled share), bootstrap interval over counties | live births in the most deprived tenth of the named counties by county deprivation rank | births in 2018 to 2024 | 9.7% | 95% CI 9.2% to 10.3% | 58 counties | births in each county | CDC WONDER natality | lbw_d10 |
| population size at which WONDER names a county (a definition) | counties that CDC WONDER natality names individually | births in 2018 to 2024 | 100,000 |   |   | none (definition) | CDC WONDER natality | wonder_threshold |
| birthweight below which a live birth counts as low birthweight (a definition) | live births | all years (a definition) | 2,500 |   |   | none (definition) | CDC WONDER natality | lbw_cutoff_grams |
| gestational age below which a live birth counts as preterm, obstetric estimate (a definition) | live births | all years (a definition) | 37 |   |   | none (definition) | CDC WONDER natality | preterm_cutoff_weeks |
| births-weighted mean of county shares (the pooled share), bootstrap interval over counties | live births in the least deprived tenth of the named counties by county deprivation rank | births in 2018 to 2024 | 8.8% | 95% CI 8.6% to 9.0% | 57 counties | births in each county | CDC WONDER natality | preterm_lo |
| births-weighted mean of county shares (the pooled share), bootstrap interval over counties | live births in the most deprived tenth of the named counties by county deprivation rank | births in 2018 to 2024 | 12.1% | 95% CI 11.3% to 12.7% | 58 counties | births in each county | CDC WONDER natality | preterm_hi |
| count of counties | counties that WONDER natality names individually and that have births | births in 2018 to 2024 | 578 |   |   | none (full count) | CDC WONDER natality | counties_used |
| share of births in the named counties | live births in the 50 states and DC | births in 2018 to 2024 | 79.5% |   | 25,709,900 births | none (full count) | CDC WONDER natality | births_share_covered |
| pooled share of births, from the published county rates and the births they imply, bootstrap interval over counties | live births to non-Hispanic Black mothers in the named counties where County Health Rankings publishes the rate | births in 2016 to 2022 | 14.0% | 95% CI 13.8% to 14.2% | 549 counties | births to mothers of the group in each county | County Health Rankings | lbw_black |
| pooled share of births, from the published county rates and the births they imply, bootstrap interval over counties | live births to non-Hispanic white mothers in the named counties where County Health Rankings publishes the rate | births in 2016 to 2022 | 6.7% | 95% CI 6.6% to 6.8% | 578 counties | births to mothers of the group in each county | County Health Rankings | lbw_white |
| pooled share of births, bootstrap interval over counties | live births to non-Hispanic Black mothers in the least deprived quartile of the named counties by deprivation rank | births in 2016 to 2022 | 12.4% | 95% CI 11.9% to 12.8% | 133 counties | births to mothers of the group in each county | County Health Rankings | lbw_black_q1 |
| pooled share of births, bootstrap interval over counties | live births to non-Hispanic white mothers in the most deprived quartile of the named counties by deprivation rank | births in 2016 to 2022 | 7.6% | 95% CI 7.5% to 7.7% | 145 counties | births to mothers of the group in each county | County Health Rankings | lbw_white_q4 |

## Findings

Low birthweight, a weight under 2,500 grams, was 7.4% of births in the least deprived tenth of the counties that CDC WONDER names individually and 9.7% in the most deprived, for births in 2018 to 2024. Preterm birth, before 37 weeks, was 8.8% against 12.1%. WONDER names only the larger counties, 578 of them. They hold 79.5% of births in the United States, so the pattern covers most newborns, though not those in the smaller counties that WONDER pools.

In County Health Rankings rates for births in 2016 to 2022, low birthweight was 14.0% of births to non-Hispanic Black mothers and 6.7% of births to non-Hispanic white mothers in these counties. The rate for non-Hispanic Black mothers is higher than the rate for non-Hispanic white mothers in every quartile of county deprivation. In the least deprived quartile of counties it is 12.4%, higher than the rate for non-Hispanic white mothers in the most deprived quartile, 7.6%. The gap between the two groups is therefore not a difference between richer and poorer counties alone.

Everything in this chapter describes counties and groups of births, not individual mothers. A county's deprivation rank says how disadvantaged its neighborhoods are on average, and a rate is the share of that county's births. Infant mortality, the outcome that follows these births, is the subject of Chapter 2.2, Infant mortality (https://vishalsingh.org/zipcode/infant-mortality).

### Why the first two measurements matter

Birthweight and gestational age are recorded on the birth certificate, which makes them the standard early measures of how a pregnancy went. A baby under 2,500 grams counts as low birthweight, and a baby born before 37 weeks counts as preterm. In the counties WONDER names, 8.3% of babies born in 2018 to 2024 were low birthweight and 10.1% were preterm. The two overlap, since a baby born early is usually small, though a baby can also be small at full term. Across the large counties the two county shares move closely together, with a correlation of 0.84.

Economists and epidemiologists study these measures because they predict what comes later. The literature on the fetal origins of adult health links conditions in the womb to test scores, educational attainment and income as well as health in adulthood.[1] A review by Aizer and Currie reports that health at birth predicts education, income and disability. It also reports that a mother's disadvantage is associated with worse health at birth through her health behaviors, her exposure to environmental hazards, her access to medical care and her own health.[2] A gradient in birthweight across places therefore matters beyond the delivery room, although these data cannot show how much of any later outcome a baby's birthweight accounts for.

### Where low birthweight is most common

Low birthweight is not spread evenly across the country. In County Health Rankings rates for births in 2016 to 2022, 89% of the counties in the highest tenth of rates are in the South, which holds 47% of the mapped counties. Of the same highest tenth, 64% are in Alabama, Georgia, Louisiana, Mississippi and South Carolina, which hold 14%. Pooled over births, the rate is 9.1% in the South and 7.8% in the rest of the country.

Table 1. Low birthweight is most common in the South, where 89% of the counties in the highest tenth of shares lie, against 47% of mapped counties

The map draws every county with at least 20 low-birthweight births in the period, 2,900 counties in all. The other 243 counties are gray, and the tooltip gives the reason for each: 98 have no published rate and 145 fall below the floor. County Health Rankings is a second source for this chapter. It takes its birth counts from the same national vital statistics system as WONDER, although the periods overlap only in 2018 to 2022 (WONDER 2018 to 2024, County Health Rankings 2016 to 2022). Where both cover a county they agree. Across the large counties the correlation between the two county shares is 0.984. That agreement lets the chapter use WONDER for the deprivation gradient and County Health Rankings for the map and for rates by mother's race and ethnicity, which the WONDER county files used here do not carry.

### Low birthweight and preterm birth rise with county deprivation

To place counties on one scale, each census tract carries its Area Deprivation Index national rank, which orders the country's neighborhoods by deprivation using income, education, employment and housing measures from the American Community Survey.[3] A higher rank means more deprived. A county's rank in this chapter is the average of its tracts' ranks, weighting each tract by its population. The counties in the gradient are the 578 that WONDER natality names individually, those with 100,000 or more residents. WONDER pools the births of every other county in a state into one row labeled Unidentified Counties, so smaller counties cannot be separated. The named counties hold 20.5 million births in 2018 to 2024. A tenth of counties here is a tenth of these large counties, so the most deprived tenth is relative to large counties and not to the nation.

Low birthweight rises at every step from the least deprived tenth, 7.4% (95% CI 7.2% to 7.7%), to the most deprived, 9.7% (95% CI 9.2% to 10.3%), a difference of 2.3 percentage points (95% CI 1.7 percentage points to 2.9 percentage points). Preterm birth follows the same path, from 8.8% to 12.1%, a larger difference of 3.2 percentage points (95% CI 2.5 percentage points to 3.9 percentage points) computed from the unrounded shares.

Across all 578 counties the low-birthweight share moves with the deprivation rank. It is higher by 0.38 percentage points (95% CI 0.32 percentage points to 0.44 percentage points) for each ten-point step in the rank, and the correlation is 0.52. The preterm share is higher by 0.49 percentage points for each ten-point step in the rank, with a correlation of 0.69. The intervals are wider in the most deprived tenth than in the least deprived one, because the most deprived tenth holds fewer births and its counties' shares vary more. Averaging the counties without weighting by births gives the same ordering.

Table 2. Low birthweight rises from 7.4% of births in the least deprived tenth of large counties to 9.7% in the most deprived

Restricting the gradient to large counties could change the picture, so County Health Rankings offers a check. It publishes low-birthweight rates for 3,040 counties that also have a deprivation rank. In tenths of all those counties the share is 7.6% in the least deprived tenth and 10.9% in the most deprived. The middle tenths are level, so the climb is steadier across large counties than across the whole country. The most deprived tenth of all counties is made up of small counties: its median county has 770 births in the period, against 12,055 in the least deprived tenth, and its interval is wider.

WONDER withholds any cell with fewer than 10 births, so a county's count below 2,500 grams is not always published whole. A withheld cell is treated as missing: it is left out of both the count and the denominator, which is the births in the published cells. The birthweight cut has 11,250 withheld cells in these counties. Because each holds fewer than 10 births, together they could hold at most 0.5% of births. The gestational-age cut has 8,704 withheld cells, which could hold at most 0.4% of births. A pooled share could therefore move by no more than that amount.

The withheld cells are not spread evenly. Withheld cells under 2,500 grams number 1.6 per county-year in the most deprived tenth against 1.0 in the least deprived. A withheld cell of that kind lowers a published share, because its births leave the count and the denominator together. With fewer than 10 births in each, the withheld birthweight cells could hold at most 1.3% of births in the most deprived tenth and 0.2% in the least deprived. The published difference between the tenths could therefore be too small by up to the first amount, and too large by up to the second.

### The gap by mother's race

The gradient averages over mothers. Splitting births by the mother's race and ethnicity shows a gap larger than the one between the most and least deprived counties. County Health Rankings counts mothers as non-Hispanic Black, non-Hispanic white or Hispanic of any race. In the large counties, 14.0% of births to non-Hispanic Black mothers in 2016 to 2022 were low birthweight, against 7.5% for Hispanic mothers and 6.7% for non-Hispanic white mothers. The rate for non-Hispanic Black mothers is 2.1 (95% CI 2.0 to 2.1) times the rate for non-Hispanic white mothers, a gap of 7.3 percentage points (95% CI 7.0 percentage points to 7.4 percentage points). The difference between the most and least deprived tenths of counties was 2.3 percentage points, for births in 2018 to 2024.

Infant deaths follow the same order. WONDER's linked birth and infant death records report mother's race only for 2018 and 2019, as a single race of any ethnicity. They give 10.3 deaths per 1,000 live births for births to Black mothers against 4.7 for births to white mothers, a group that includes Hispanic mothers.

Table 3. Non-Hispanic Black mothers had low-birthweight births at 14.0% against 6.7% for non-Hispanic white mothers, a ratio of 2.1

These are rates for groups of births in counties and say nothing about why a given mother's baby is born small. One study that bears on the explanation compared infants born in Illinois from 1980 to 1995. Infants of African-born Black mothers were heavier on average than infants of U.S.-born Black mothers and closer in weight to infants of U.S.-born white mothers.[4] That pattern is hard to square with an explanation based on ancestry alone. Women who migrate are a selected group, though, so the comparison cannot rule out other differences between the two groups of Black mothers.

### The gap at every level of deprivation

Counties differ in deprivation, so the race gap could shrink once counties are compared at the same level. In every quartile of county deprivation, the rate for non-Hispanic Black mothers is higher than the rate for non-Hispanic white mothers, for births in 2016 to 2022 (Table 4). The gap is 6.2 percentage points (95% CI 5.8 percentage points to 6.6 percentage points) in the least deprived quartile and 8.0 percentage points (95% CI 7.7 percentage points to 8.2 percentage points) in the most deprived, and the second is larger than the first by 1.8 percentage points (95% CI 1.3 percentage points to 2.3 percentage points). The rate for non-Hispanic Black mothers in the least deprived quartile, 12.4% (95% CI 11.9% to 12.8%), is higher than the rate for non-Hispanic white mothers in the most deprived quartile, 7.6% (95% CI 7.5% to 7.7%), by 4.8 percentage points (95% CI 4.4 percentage points to 5.2 percentage points). The second view of the figure shows the same order for infant mortality, with Black and white mothers counted by single race of any ethnicity, in counties of 250,000 residents or more, for births in 2018 and 2019.

Table 4. Low birthweight is more common for non-Hispanic Black mothers in the least deprived quartile of counties (12.4%) than for non-Hispanic white mothers in the most deprived (7.6%)

 (Note: The weathering hypothesis proposed by Geronimus holds that the health of Black women may begin to deteriorate in early adulthood as a physical consequence of cumulative socioeconomic disadvantage.[5] A gap that persists at every level of county deprivation is consistent with it, but a comparison across counties cannot separate it from other explanations.)

A county's rank averages over neighborhoods that differ a great deal. A Black mother in a county ranked among the least deprived may live in a neighborhood more deprived than the county average. A white mother in a county ranked among the most deprived may live in a less deprived one. The comparison therefore cannot show that mothers under the same neighborhood conditions have different outcomes. What it shows is that sorting counties by deprivation does not remove the gap. A comparison of Black and white children raised in the same tract is in Chapter 4.3, Race, income and mobility (https://vishalsingh.org/zipcode/race-income-and-mobility).

### Births fell between 2018 and 2024

Births in the United States fell from 3.79 million in 2018 to 3.63 million in 2024, a decline of 4.3%. The shares in this chapter are rates, so they do not move with the count, but the pattern describes a smaller number of babies each year. Where the decline is steepest is the subject of Chapter 2.4, Where births are falling (https://vishalsingh.org/zipcode/where-births-are-falling).

### Two studies of income and insurance during pregnancy

The gradient and the gap are descriptions. Two studies of policy changes give evidence on whether income support and insurance coverage during pregnancy change birthweight, and each is reported under its own design.

> Causal evidence: Income support and birth outcomes (Difference-in-differences)
>
> Hoynes, Miller and Simon studied federal expansions of the Earned Income Tax Credit by comparing births to mothers whom the tax reforms gave larger credits with births to mothers who received smaller ones, before and after each reform.[6] The comparison removes changes that affected both groups, so the study attributes the lower rate of low birthweight and the higher mean birthweight among mothers with larger credits to the added income. The estimates are largest for single mothers with little education and for births to Black mothers. The paper's candidate mechanisms are more prenatal care and less smoking, and it finds little role for changes in health insurance.

> Causal evidence: Medicaid eligibility for pregnant women (State and year fixed effects on simulated eligibility)
>
> During the 1980s, states widened Medicaid eligibility for pregnant women at different times and by different amounts. Currie and Gruber built a measure of the share of women eligible under each state's rules in each year by applying the rules to a fixed national sample of women, so the measure moves with the rules and not with a state's own population. They regressed state-by-year birth outcomes on that measure with state and year fixed effects.[7] Where eligibility expanded, infant mortality fell by more than low birthweight did, and changes aimed at specific low-income groups had larger effects than broad expansions to higher incomes, which fewer eligible women took up.

Both sets of estimates apply to the policies and years studied. They do not measure how much of the county gradient or the race gap those policies would close.

### Low birthweight keeps its gradient under the standardized index and flattens under the degree share

Table 5 ranks the 578 large counties that have all four rankings in four ways. Ranked by the ADI on this common sample, the low-birthweight share of births is 7.4% (95% CI 7.2% to 7.7%) in the least deprived tenth and 9.6% (95% CI 9.1% to 10.2%) in the most deprived, and the rank correlation between tenth and share is +0.99. Under the standardized index the two ends are 7.2% (95% CI 7.0% to 7.3%) and 9.7% (95% CI 9.0% to 10.4%), with a correlation of +0.99, and under median household income they are 7.2% (95% CI 7.0% to 7.3%) and 10.0% (95% CI 9.4% to 10.6%), with a correlation of +0.96, so neither counts as a change under the book's rule (Chapter 9.1, How far one ranking goes (https://vishalsingh.org/zipcode/how-far-one-ranking-goes)). Ranked by the bachelor's degree share, the ends are 7.6% (95% CI 7.2% to 7.9%) and 8.2% (95% CI 7.8% to 8.6%) and their intervals overlap, which the rule counts as a change, although income and degree share are checks that cannot switch the book's ranking on their own.

Table 5. Low birthweight is steadily worse toward the more deprived end under the ADI; the two ends do not differ detectably under the degree-share ranking

### What these data cannot show

The county figures describe places, and a county rate is an average over the births in it. A county ranked as least deprived can hold deprived neighborhoods, and the Area Deprivation Index understates deprivation in high-cost cities (Chapter 1.1, What the deprivation index measures (https://vishalsingh.org/zipcode/anatomy-of-deprivation)). A pattern across counties does not describe the babies in any one of them.

The WONDER figures cover only counties of 100,000 or more residents, which hold 79.5% of births. Births in smaller counties are pooled by WONDER and cannot be placed on the deprivation scale, and the check against County Health Rankings shows that the pattern is less regular across all counties than across large ones.

The sources differ in period and definition. The deprivation index draws on the American Community Survey for 2019 to 2023, while the births span 2018 to 2024. Rates by mother's race come from County Health Rankings for births in 2016 to 2022, for non-Hispanic Black, non-Hispanic white and Hispanic mothers, and they are pooled over the counties where a rate is published. Infant mortality by mother's race comes from 2018 and 2019 only, for single races of any ethnicity, so its white group includes Hispanic mothers and is not the group in the low-birthweight rates. The infant mortality county view covers only larger counties and leaves out county-years in which WONDER withholds the deaths, which are 13.6% of Black births and 5.2% of white births there. The intervals leave out those differences in source, period and group.

Finally, none of these comparisons identifies a mechanism. Deprivation and race are tied to smoking, prenatal care, chronic stress, pollution, maternal age and education, and to the quality of the nearest hospital. Birthweight is associated with all of them. The two policy studies above are the only evidence in this chapter on what changes the numbers, and they speak to income and coverage, not to the rest. A county map cannot settle the mechanism behind any one county's rate or behind the gap between groups.

## Tables

### Table 1. Low birthweight is most common in the South, where 89% of the counties in the highest tenth of shares lie, against 47% of mapped counties

Share of live births under 2,500 grams by county, County Health Rankings 2024, births in 2016 to 2022

No inline rows: the values of Table 1 are in https://vishalsingh.org/zipcode/data/birthweight-and-preterm-birth/lbw_map.json.

Note: Counties with fewer than 10 low-birthweight births in the period have no rate in County Health Rankings, and counties with fewer than 20 are drawn as no data here, the same floor the book uses for death maps. 243 counties are not mapped.

Note: The South is the Census South region (16 states and DC). County Health Rankings rates come from the same NCHS birth records as WONDER; across the large counties the two sources correlate at 0.984.

Note: Connecticut is shown as its eight legacy counties, the coding of the source. Alaska is drawn. Puerto Rico is not drawn.

Source: County Health Rankings 2024, births in 2016-2022 (NCHS natality counts); county rates, population-weighted by births

Data: https://vishalsingh.org/zipcode/report/birthweight-and-preterm-birth/figures/lbw_map.json

### Table 2. Low birthweight rises from 7.4% of births in the least deprived tenth of large counties to 9.7% in the most deprived

Share of live births under 2,500 grams (and before 37 weeks in the second tab) by county deprivation, 2018 to 2024

#### Table 2.1. Low birthweight, large counties

Across 578 large counties the low-birthweight share moves with the deprivation rank, a correlation of 0.52

Share of live births under 2,500 grams against the county deprivation rank, one dot per county, 2018 to 2024, with births-weighted decile means

| adi | births | county | lbw |
| ---: | ---: | --- | ---: |
| 27.27 | 35153 | Ada County, ID | 6.35 |
| 31.50 | 43865 | Adams County, CO | 9.34 |
| 53.02 | 6346 | Adams County, PA | 5.23 |
| 67.21 | 13354 | Aiken County, SC | 9.03 |
| 57.27 | 18344 | Alachua County, FL | 10.40 |
| 65.57 | 13473 | Alamance County, NC | 9.47 |
| 7.82 | 119533 | Alameda County, CA | 7.49 |
| 52.09 | 19652 | Albany County, NY | 8.64 |
| 18.67 | 16409 | Alexandria city, VA | 6.56 |
| 57.15 | 8760 | Allegan County, MI | 5.82 |
| 60.06 | 85018 | Allegheny County, PA | 8.94 |
| 67.85 | 36535 | Allen County, IN | 9.01 |
| 77.54 | 8561 | Allen County, OH | 8.92 |
| 37.23 | 25918 | Anchorage Borough, AK | 6.89 |
| 66.51 | 15986 | Anderson County, SC | 9.33 |
| 61.55 | 8066 | Androscoggin County, ME | 8.12 |
| 26.31 | 47568 | Anne Arundel County, MD | 7.57 |
| 41.11 | 28547 | Anoka County, MN | 6.93 |
| 25.25 | 52722 | Arapahoe County, CO | 10.10 |
| 11.96 | 17788 | Arlington County, VA | 6.18 |
| 49.71 | 11589 | Ascension Parish, LA | 9.58 |
| 80.50 | 7231 | Ashtabula County, OH | 6.37 |
| 54.06 | 19373 | Atlantic County, NJ | 7.89 |
| 50.20 | 17176 | Baldwin County, AL | 7.93 |
| 42.28 | 65464 | Baltimore County, MD | 9.32 |
| 64.94 | 50993 | Baltimore city, MD | 12.02 |
| 20.78 | 10379 | Barnstable County, MA | 5.43 |
| 54.23 | 9308 | Bartow County, GA | 6.77 |
| 55.85 | 14687 | Bay County, FL | 8.26 |
| 80.21 | 6392 | Bay County, MI | 7.58 |
| 37.13 | 13141 | Beaufort County, SC | 8.11 |
| 69.45 | 10623 | Beaver County, PA | 6.79 |
| 64.57 | 42199 | Bell County, TX | 9.12 |
| 48.22 | 26503 | Benton County, AR | 7.35 |
| 40.98 | 17730 | Benton County, WA | 6.27 |
| 19.42 | 63171 | Bergen County, NJ | 7.44 |
| 50.80 | 20807 | Berkeley County, SC | 9.31 |
| 54.82 | 10636 | Berkeley County, WV | 8.17 |
| 60.09 | 31836 | Berks County, PA | 7.92 |
| 49.99 | 6398 | Berkshire County, MA | 7.68 |
| 54.44 | 46209 | Bernalillo County, NM | 9.73 |
| 66.53 | 10686 | Berrien County, MI | 8 |
| 59.17 | 183350 | Bexar County, TX | 8.83 |
| 75.80 | 14382 | Bibb County, GA | 13.54 |
| 69.55 | 11194 | Black Hawk County, IA | 8.05 |
| 76.45 | 8120 | Blair County, PA | 6.09 |
| 53.45 | 9120 | Blount County, TN | 7.61 |
| 46 | 12633 | Bonneville County, ID | 7.81 |
| 49.97 | 11867 | Boone County, KY | 6.60 |
| 58.43 | 13758 | Boone County, MO | 8.40 |
| 66.36 | 11338 | Bossier Parish, LA | 11.42 |
| 15.94 | 17248 | Boulder County, CO | 7.77 |
| 49.43 | 32862 | Brazoria County, TX | 8.32 |
| 55.97 | 18815 | Brazos County, TX | 8.09 |
| 48.90 | 36470 | Brevard County, FL | 8.41 |
| 34.02 | 39240 | Bristol County, MA | 8.48 |
| 31.09 | 124256 | Bronx County, NY | 10.45 |
| 78.39 | 12584 | Broome County, NY | 7.98 |
| 40.97 | 145472 | Broward County, FL | 9.63 |
| 56.90 | 21131 | Brown County, WI | 7.06 |
| 48.22 | 7781 | Brunswick County, NC | 7.76 |
| 29.71 | 39164 | Bucks County, PA | 6.99 |
| 42.50 | 16802 | Buncombe County, NC | 8.30 |
| 40.10 | 31808 | Burlington County, NJ | 7.72 |
| 58.33 | 30299 | Butler County, OH | 8 |
| 50.28 | 12033 | Butler County, PA | 5.92 |
| 38.13 | 14612 | Butte County, CA | 6.20 |
| 46.20 | 18406 | Cabarrus County, NC | 8.82 |
| 37.55 | 14438 | Cache County, UT | 5.83 |
| 76.41 | 20103 | Caddo Parish, LA | 14.52 |
| 66.05 | 18828 | Calcasieu Parish, LA | 10.55 |
| 80.93 | 9096 | Calhoun County, AL | 8.77 |
| 77.35 | 10306 | Calhoun County, MI | 8.76 |
| 84.10 | 8283 | Cambria County, PA | 8.21 |
| 54.47 | 42450 | Camden County, NJ | 9.05 |
| 84.05 | 41975 | Cameron County, TX | 8.51 |
| 57.52 | 13874 | Canadian County, OK | 6.76 |
| 42.39 | 22809 | Canyon County, ID | 6.65 |
| 60.26 | 11132 | Carroll County, GA | 8.14 |
| 29.66 | 11977 | Carroll County, MD | 5.58 |
| 49.09 | 16894 | Cass County, ND | 6.99 |
| 64.85 | 11691 | Catawba County, NC | 8.15 |
| 44.54 | 8004 | Cecil County, MD | 7.08 |
| 46.22 | 7675 | Centre County, PA | 4.55 |
| 66.50 | 14484 | Champaign County, IL | 8.81 |
| 30 | 12676 | Charles County, MD | 9.27 |
| 35.04 | 34737 | Charleston County, SC | 9.36 |
| 51.12 | 7422 | Charlotte County, FL | 7.89 |
| 56.20 | 25303 | Chatham County, GA | 11.15 |
| 83.81 | 8651 | Chautauqua County, NY | 7.72 |
| 33.22 | 19655 | Cherokee County, GA | 6.77 |
| 39.92 | 19793 | Chesapeake city, VA | 9.07 |
| 26.68 | 38149 | Chester County, PA | 6.19 |
| 40.91 | 28155 | Chesterfield County, VA | 7.96 |
| 31.50 | 9595 | Chittenden County, VT | 5.44 |
| 66.62 | 7253 | Citrus County, FL | 7.34 |
| 21.10 | 26316 | Clackamas County, OR | 5.97 |
| 64.21 | 10055 | Clark County, IN | 7.84 |
| 39.24 | 176293 | Clark County, NV | 9.37 |
| 77.65 | 11089 | Clark County, OH | 8.89 |
| 26.64 | 38736 | Clark County, WA | 6.67 |
| 53.22 | 9286 | Clarke County, GA | 9.88 |
| 49.93 | 15614 | Clay County, FL | 8.42 |
| 52.12 | 20890 | Clay County, MO | 7.43 |
| 71.44 | 24539 | Clayton County, GA | 13.52 |
| 55.43 | 15577 | Clermont County, OH | 7.38 |
| 59.97 | 19200 | Cleveland County, OK | 7.63 |
| 37.38 | 61759 | Cobb County, GA | 8.82 |
| 71.06 | 8859 | Cochise County, AZ | 7.32 |
| 40.84 | 9348 | Coconino County, AZ | 6.95 |
| 30.31 | 23129 | Collier County, FL | 7.14 |
| 26.55 | 85845 | Collin County, TX | 7.47 |
| 48.03 | 11898 | Columbia County, GA | 7.21 |
| 80.83 | 6778 | Columbiana County, OH | 7.32 |
| 35.12 | 12791 | Comal County, TX | 6.84 |
| 77.51 | 11401 | Comanche County, OK | 7.63 |
| 12.71 | 81039 | Contra Costa County, CA | 7.34 |
| 47.08 | 384972 | Cook County, IL | 8.94 |
| 43.78 | 11498 | Coweta County, GA | 9.35 |
| 41.54 | 8603 | Cowlitz County, WA | 5.87 |
| 66.65 | 9031 | Craven County, NC | 8.23 |
| 30.73 | 19393 | Cumberland County, ME | 6.76 |
| 72.53 | 36672 | Cumberland County, NC | 10.35 |
| 68.20 | 12811 | Cumberland County, NJ | 9.47 |
| 53.15 | 18447 | Cumberland County, PA | 6.79 |
| 69.33 | 92117 | Cuyahoga County, OH | 10.62 |
| 35.84 | 34751 | Dakota County, MN | 6.78 |
| 51.84 | 258092 | Dallas County, TX | 8.49 |
| 37.44 | 39430 | Dane County, WI | 7.16 |
| 62.88 | 23065 | Dauphin County, PA | 9.31 |
| 68.39 | 12395 | Davidson County, NC | 9.54 |
| 38 | 70746 | Davidson County, TN | 9.12 |
| 26.64 | 34072 | Davis County, UT | 7.41 |
| 44.32 | 70179 | DeKalb County, GA | 10.70 |
| 60.71 | 7341 | DeKalb County, IL | 7.53 |
| 56.53 | 15346 | DeSoto County, MS | 9.96 |
| 83.14 | 7493 | Delaware County, IN | 7.69 |
| 29.76 | 14891 | Delaware County, OH | 6.53 |
| 47.79 | 44118 | Delaware County, PA | 8.56 |
| 31.79 | 75980 | Denton County, TX | 7.44 |
| 21.31 | 64002 | Denver County, CO | 9.97 |
| 21.34 | 13013 | Deschutes County, OR | 6.50 |
| 16.41 | 59412 | District of Columbia, DC | 10.07 |
| 70.08 | 17975 | Dona Ana County, NM | 8.19 |
| 50.03 | 13688 | Dorchester County, SC | 9.18 |
| 13.04 | 25912 | Douglas County, CO | 8.76 |
| 52.95 | 11757 | Douglas County, GA | 10.92 |
| 51.04 | 7191 | Douglas County, KS | 6.60 |
| 55.92 | 56047 | Douglas County, NE | 8.94 |
| 53.80 | 6999 | Douglas County, OR | 6.39 |
| 34.30 | 68211 | DuPage County, IL | 6.87 |
| 41.88 | 28912 | Durham County, NC | 8.83 |
| 38.30 | 18277 | Dutchess County, NY | 7.42 |
| 55.54 | 89270 | Duval County, FL | 10.39 |
| 61.83 | 38895 | East Baton Rouge Parish, LA | 12.33 |
| 63.59 | 7301 | Eaton County, MI | 7.86 |
| 66.99 | 20874 | Ector County, TX | 9.26 |
| 17.71 | 10781 | El Dorado County, CA | 6.05 |
| 32.73 | 63679 | El Paso County, CO | 9.65 |
| 74.50 | 78465 | El Paso County, TX | 9.30 |
| 67.65 | 20605 | Elkhart County, IN | 6.82 |
| 45.56 | 18052 | Ellis County, TX | 7.30 |
| 62.29 | 65100 | Erie County, NY | 8.98 |
| 74.57 | 18441 | Erie County, PA | 9.13 |
| 62.76 | 26111 | Escambia County, FL | 10.78 |
| 21.97 | 59384 | Essex County, MA | 7.26 |
| 33.03 | 69651 | Essex County, NJ | 9.26 |
| 79.30 | 8297 | Etowah County, AL | 8.03 |
| 14.29 | 89085 | Fairfax County, VA | 7.05 |
| 26.85 | 69289 | Fairfield County, CT | 7.22 |
| 53.08 | 11758 | Fairfield County, OH | 7.21 |
| 64.89 | 10394 | Faulkner County, AR | 7.65 |
| 31.83 | 6286 | Fayette County, GA | 9.46 |
| 55.11 | 26238 | Fayette County, KY | 9.23 |
| 82.19 | 8340 | Fayette County, PA | 8.56 |
| 77.03 | 11123 | Florence County, SC | 13.09 |
| 64.43 | 7111 | Fond du Lac County, WI | 6.31 |
| 22.24 | 16062 | Forsyth County, GA | 6.84 |
| 62.43 | 30894 | Forsyth County, NC | 10.40 |
| 37.63 | 64832 | Fort Bend County, TX | 8.77 |
| 62.33 | 8030 | Franklin County, MO | 7.42 |
| 55.13 | 123800 | Franklin County, OH | 9.29 |
| 58.84 | 11941 | Franklin County, PA | 6.90 |
| 27.94 | 21568 | Frederick County, MD | 6.61 |
| 46.34 | 96600 | Fresno County, CA | 7.76 |
| 36.01 | 86524 | Fulton County, GA | 11.37 |
| 50.28 | 27382 | Galveston County, TX | 8.83 |
| 62.53 | 18847 | Gaston County, NC | 10.28 |
| 73.71 | 30357 | Genesee County, MI | 10.49 |
| 45 | 20365 | Gloucester County, NJ | 7.79 |
| 61.48 | 11943 | Grayson County, TX | 7.44 |
| 68.78 | 23058 | Greene County, MO | 8.16 |
| 56.86 | 11888 | Greene County, OH | 6.86 |
| 52.19 | 45163 | Greenville County, SC | 8.48 |
| 71.55 | 11625 | Gregg County, TX | 8.94 |
| 49.21 | 13972 | Guadalupe County, TX | 7.62 |
| 62.39 | 41153 | Guilford County, NC | 10.02 |
| 41.21 | 76833 | Gwinnett County, GA | 9.25 |
| 49.03 | 18786 | Hall County, GA | 6.80 |
| 33.43 | 26854 | Hamilton County, IN | 6.49 |
| 59.80 | 72644 | Hamilton County, OH | 9.90 |
| 51.80 | 30828 | Hamilton County, TN | 8.71 |
| 54.48 | 33272 | Hampden County, MA | 8.96 |
| 37.36 | 6428 | Hampshire County, MA | 6.40 |
| 60.22 | 11058 | Hampton city, VA | 10.17 |
| 64.32 | 9474 | Hardin County, KY | 7.58 |
| 36.71 | 18296 | Harford County, MD | 7.55 |
| 63.61 | 13138 | Harnett County, NC | 8.33 |
| 55.59 | 455394 | Harris County, TX | 9.17 |
| 69.85 | 17487 | Harrison County, MS | 10.87 |
| 47.60 | 62452 | Hartford County, CT | 8.64 |
| 29.53 | 14568 | Hawaii County, HI | 8.70 |
| 40.66 | 21150 | Hays County, TX | 6.97 |
| 49.59 | 7168 | Henderson County, NC | 7.53 |
| 45.88 | 12504 | Hendricks County, IN | 6.43 |
| 36.21 | 103352 | Hennepin County, MN | 7.74 |
| 42.82 | 25990 | Henrico County, VA | 8.35 |
| 49.74 | 19461 | Henry County, GA | 12.03 |
| 60.04 | 11629 | Hernando County, FL | 8.33 |
| 83.04 | 96697 | Hidalgo County, TX | 9.29 |
| 45.98 | 119285 | Hillsborough County, FL | 9.03 |
| 35.48 | 29024 | Hillsborough County, NH | 6.91 |
| 80.51 | 19288 | Hinds County, MS | 15.39 |
| 11.87 | 78677 | Honolulu County, HI | 8.52 |
| 57.80 | 22858 | Horry County, SC | 8.36 |
| 74.54 | 9155 | Houston County, AL | 10.05 |
| 64.74 | 13805 | Houston County, GA | 9.57 |
| 19.35 | 22481 | Howard County, MD | 8.32 |
| 24.79 | 65616 | Hudson County, NJ | 8.28 |
| 36.22 | 8646 | Humboldt County, CA | 5.54 |
| 22.68 | 7266 | Hunterdon County, NJ | 4.94 |
| 57.41 | 16878 | Imperial County, CA | 6.24 |
| 49.19 | 8699 | Indian River County, FL | 7.33 |
| 68 | 20257 | Ingham County, MI | 8.71 |
| 50.75 | 13891 | Iredell County, NC | 8.29 |
| 72.38 | 11398 | Jackson County, MI | 9.02 |
| 64.74 | 60827 | Jackson County, MO | 9.83 |
| 71.13 | 10872 | Jackson County, MS | 9.68 |
| 36.24 | 14713 | Jackson County, OR | 6.98 |
| 77.71 | 10937 | Jasper County, MO | 7.44 |
| 62.60 | 56397 | Jefferson County, AL | 11.36 |
| 19.12 | 38651 | Jefferson County, CO | 9.02 |
| 60.18 | 67103 | Jefferson County, KY | 9.39 |
| 59.66 | 16932 | Jefferson County, MO | 8.20 |
| 70.75 | 12213 | Jefferson County, NY | 7.56 |
| 75.11 | 22535 | Jefferson County, TX | 10.73 |
| 60.17 | 37531 | Jefferson Parish, LA | 10.22 |
| 48.25 | 11748 | Johnson County, IA | 6.42 |
| 52.49 | 13231 | Johnson County, IN | 7.34 |
| 35.27 | 48504 | Johnson County, KS | 6.53 |
| 55.51 | 16420 | Johnson County, TX | 6.66 |
| 53.69 | 18770 | Johnston County, NC | 8.05 |
| 60.96 | 19313 | Kalamazoo County, MI | 7.94 |
| 80.09 | 11993 | Kanawha County, WV | 10.63 |
| 45.72 | 40039 | Kane County, IL | 7.43 |
| 69.71 | 8127 | Kankakee County, IL | 8.33 |
| 50.78 | 16592 | Kaufman County, TX | 7.73 |
| 42.23 | 10514 | Kendall County, IL | 6.35 |
| 64.60 | 7573 | Kennebec County, ME | 6.75 |
| 54.96 | 12018 | Kenosha County, WI | 7.20 |
| 51.03 | 14752 | Kent County, DE | 9.39 |
| 51.75 | 57233 | Kent County, MI | 8.79 |
| 39.65 | 10222 | Kent County, RI | 6.38 |
| 58.65 | 15280 | Kenton County, KY | 7.69 |
| 52.63 | 86859 | Kern County, CA | 7.65 |
| 12.92 | 163987 | King County, WA | 6.98 |
| 51.62 | 15013 | Kings County, CA | 6.52 |
| 9.96 | 243582 | Kings County, NY | 7.87 |
| 24.37 | 19887 | Kitsap County, WA | 6.59 |
| 52.49 | 37089 | Knox County, TN | 8.13 |
| 30.06 | 13287 | Kootenai County, ID | 5.04 |
| 58.76 | 7890 | La Crosse County, WI | 5.81 |
| 70.57 | 7995 | La Porte County, IN | 8.12 |
| 73.64 | 7663 | LaSalle County, IL | 6.66 |
| 70.82 | 14150 | Lackawanna County, PA | 8.38 |
| 60.48 | 23128 | Lafayette Parish, LA | 9.70 |
| 52.55 | 25103 | Lake County, FL | 8.56 |
| 42.15 | 47218 | Lake County, IL | 7.64 |
| 65.67 | 37650 | Lake County, IN | 8.95 |
| 62.50 | 14729 | Lake County, OH | 7.18 |
| 55.73 | 26011 | Lancaster County, NE | 7.04 |
| 49.56 | 48591 | Lancaster County, PA | 6.29 |
| 37.21 | 21449 | Lane County, OR | 6.59 |
| 24.18 | 21477 | Larimer County, CO | 8.72 |
| 60.56 | 10966 | Lebanon County, PA | 7.23 |
| 59.15 | 13390 | Lee County, AL | 9.61 |
| 47.61 | 51118 | Lee County, FL | 8.64 |
| 54.75 | 28087 | Lehigh County, PA | 8.38 |
| 57.97 | 19801 | Leon County, FL | 11.05 |
| 62.02 | 22719 | Lexington County, SC | 8.84 |
| 55.12 | 13283 | Licking County, OH | 7.21 |
| 63.31 | 18307 | Linn County, IA | 6.80 |
| 43.32 | 10146 | Linn County, OR | 6 |
| 43.13 | 9953 | Litchfield County, CT | 5.58 |
| 40.20 | 11994 | Livingston County, MI | 6.22 |
| 61.02 | 12929 | Livingston Parish, LA | 8.21 |
| 64.59 | 22154 | Lorain County, OH | 8.77 |
| 14.23 | 688901 | Los Angeles County, CA | 7.54 |
| 11.94 | 33757 | Loudoun County, VA | 6.65 |
| 73.96 | 11240 | Lowndes County, GA | 11.63 |
| 67.26 | 28358 | Lubbock County, TX | 8.81 |
| 74.43 | 34713 | Lucas County, OH | 9.97 |
| 75.48 | 22473 | Luzerne County, PA | 8.39 |
| 69.88 | 7892 | Lycoming County, PA | 7.39 |
| 58.72 | 61638 | Macomb County, MI | 9.18 |
| 81.61 | 8213 | Macon County, IL | 11.12 |
| 43.77 | 14714 | Madera County, CA | 6 |
| 53.80 | 30893 | Madison County, AL | 10.12 |
| 68.92 | 18745 | Madison County, IL | 8.81 |
| 78.21 | 9983 | Madison County, IN | 8.79 |
| 78.54 | 16277 | Mahoning County, OH | 10.81 |
| 46.74 | 25916 | Manatee County, FL | 8.09 |
| 64.36 | 10200 | Marathon County, WI | 6.25 |
| 36.11 | 354643 | Maricopa County, AZ | 7.64 |
| 4.89 | 15052 | Marin County, CA | 5.01 |
| 66.59 | 25152 | Marion County, FL | 8.72 |
| 67.02 | 98934 | Marion County, IN | 10 |
| 38.99 | 27904 | Marion County, OR | 6.49 |
| 40.77 | 8868 | Martin County, FL | 6.15 |
| 11.59 | 11878 | Maui County, HI | 7.34 |
| 45.48 | 21327 | McHenry County, IL | 6.44 |
| 63.42 | 11786 | McLean County, IL | 7.26 |
| 66.20 | 23344 | McLennan County, TX | 8.66 |
| 40.70 | 104482 | Mecklenburg County, NC | 9.65 |
| 49.06 | 11554 | Medina County, OH | 5.68 |
| 42.31 | 26538 | Merced County, CA | 6.40 |
| 42.50 | 28605 | Mercer County, NJ | 7.89 |
| 77.08 | 7124 | Mercer County, PA | 7.78 |
| 41.73 | 8984 | Merrimack County, NH | 5.77 |
| 42.80 | 10649 | Mesa County, CO | 8.12 |
| 63.70 | 8098 | Miami County, OH | 5.29 |
| 37.24 | 204515 | Miami-Dade County, FL | 8.08 |
| 39.08 | 9099 | Middlesex County, CT | 6.57 |
| 14.53 | 112506 | Middlesex County, MA | 7.06 |
| 29.18 | 59673 | Middlesex County, NJ | 8.11 |
| 48.70 | 21004 | Midland County, TX | 7.75 |
| 66.79 | 84871 | Milwaukee County, WI | 11.06 |
| 52.58 | 19552 | Minnehaha County, SD | 7.16 |
| 33.02 | 7492 | Missoula County, MT | 7.06 |
| 72.95 | 36155 | Mobile County, AL | 12.23 |
| 60.90 | 12258 | Mohave County, AZ | 7.01 |
| 20.24 | 42807 | Monmouth County, NJ | 6.92 |
| 53.66 | 7584 | Monroe County, IN | 7.39 |
| 62.39 | 10396 | Monroe County, MI | 7.02 |
| 65.30 | 52483 | Monroe County, NY | 9.38 |
| 55.79 | 9662 | Monroe County, PA | 8.58 |
| 16.90 | 38951 | Monterey County, CA | 6.57 |
| 74.34 | 21006 | Montgomery County, AL | 13.32 |
| 18.93 | 81398 | Montgomery County, MD | 7.43 |
| 72.95 | 43206 | Montgomery County, OH | 9.19 |
| 31.41 | 59114 | Montgomery County, PA | 7.42 |
| 56.86 | 25284 | Montgomery County, TN | 8.84 |
| 43.03 | 56753 | Montgomery County, TX | 7.44 |
| 71.62 | 10187 | Morgan County, AL | 8.41 |
| 19.35 | 33488 | Morris County, NJ | 6.06 |
| 24.43 | 52583 | Multnomah County, OR | 7.18 |
| 73.55 | 18953 | Muscogee County, GA | 12.06 |
| 73.29 | 13260 | Muskegon County, MI | 9.79 |
| 12.88 | 8402 | Napa County, CA | 5.36 |
| 13.81 | 96140 | Nassau County, NY | 8.10 |
| 74.73 | 8909 | Navajo County, AZ | 7.77 |
| 43.82 | 42669 | New Castle County, DE | 9.35 |
| 43.45 | 15175 | New Hanover County, NC | 8.74 |
| 48.10 | 59132 | New Haven County, CT | 8.37 |
| 46.99 | 17374 | New London County, CT | 7.54 |
| 8.92 | 103804 | New York County, NY | 8.22 |
| 62.74 | 17861 | Newport News city, VA | 10.06 |
| 71 | 13750 | Niagara County, NY | 8.62 |
| 14.99 | 49215 | Norfolk County, MA | 6.83 |
| 53.51 | 22047 | Norfolk city, VA | 11.02 |
| 50.03 | 19439 | Northampton County, PA | 7.77 |
| 68.22 | 30358 | Nueces County, TX | 8.96 |
| 43.65 | 88000 | Oakland County, MI | 8.11 |
| 35.14 | 69158 | Ocean County, NJ | 5.77 |
| 45.21 | 18428 | Okaloosa County, FL | 8.23 |
| 66.32 | 77198 | Oklahoma County, OK | 9.12 |
| 45.24 | 13733 | Olmsted County, MN | 6.69 |
| 72.81 | 16435 | Oneida County, NY | 8.61 |
| 67.47 | 33511 | Onondaga County, NY | 8.65 |
| 63.46 | 26615 | Onslow County, NC | 7.05 |
| 62.35 | 6746 | Ontario County, NY | 6.24 |
| 11.18 | 224357 | Orange County, CA | 6.80 |
| 44.12 | 114737 | Orange County, FL | 9.12 |
| 31.23 | 8057 | Orange County, NC | 7.90 |
| 36.70 | 37934 | Orange County, NY | 6.93 |
| 51.55 | 29500 | Orleans Parish, LA | 12.43 |
| 47.66 | 33136 | Osceola County, FL | 8.54 |
| 78.33 | 8126 | Oswego County, NY | 7.29 |
| 47.32 | 21833 | Ottawa County, MI | 6.22 |
| 74.81 | 13939 | Ouachita Parish, LA | 13.14 |
| 55.44 | 14935 | Outagamie County, WI | 6.17 |
| 38.83 | 103939 | Palm Beach County, FL | 8.85 |
| 40 | 12572 | Parker County, TX | 6.12 |
| 55.44 | 38238 | Pasco County, FL | 8.06 |
| 34.54 | 44279 | Passaic County, NJ | 9.01 |
| 46.14 | 12685 | Paulding County, GA | 8.90 |
| 53.49 | 9998 | Pennington County, SD | 6.77 |
| 68.96 | 8987 | Penobscot County, ME | 6.09 |
| 74.69 | 15375 | Peoria County, IL | 9.90 |
| 62.40 | 136032 | Philadelphia County, PA | 11.07 |
| 65.16 | 8567 | Pickens County, SC | 7.57 |
| 26.23 | 76617 | Pierce County, WA | 6.99 |
| 53.40 | 70658 | Pima County, AZ | 8.29 |
| 50.23 | 34632 | Pinal County, AZ | 7.09 |
| 48.07 | 52352 | Pinellas County, FL | 8.31 |
| 71.07 | 14425 | Pitt County, NC | 11.62 |
| 15.87 | 26154 | Placer County, CA | 5.53 |
| 23.13 | 36711 | Plymouth County, MA | 7.07 |
| 62.59 | 60308 | Polk County, FL | 9.18 |
| 55.05 | 45272 | Polk County, IA | 7.60 |
| 63.47 | 9502 | Portage County, OH | 7.17 |
| 52.82 | 11257 | Porter County, IN | 6.91 |
| 81.09 | 11015 | Potter County, TX | 9.43 |
| 33.05 | 78003 | Prince George's County, MD | 9.51 |
| 23.33 | 43996 | Prince William County, VA | 7.55 |
| 42.93 | 48401 | Providence County, RI | 8.25 |
| 58.04 | 12052 | Pueblo County, CO | 10.64 |
| 68.10 | 34780 | Pulaski County, AR | 11.78 |
| 16.43 | 175615 | Queens County, NY | 9.07 |
| 59.85 | 15175 | Racine County, WI | 8.77 |
| 46.60 | 47539 | Ramsey County, MN | 8.41 |
| 58.16 | 11721 | Randall County, TX | 7.58 |
| 75.29 | 10893 | Randolph County, NC | 8.40 |
| 61.45 | 12159 | Rankin County, MS | 10.38 |
| 74.64 | 11353 | Rapides Parish, LA | 11.97 |
| 56.87 | 10764 | Rensselaer County, NY | 7.52 |
| 78.32 | 9503 | Richland County, OH | 6.94 |
| 64.44 | 32209 | Richland County, SC | 11.42 |
| 76.90 | 19804 | Richmond County, GA | 12.49 |
| 15.94 | 35393 | Richmond County, NY | 8.02 |
| 49.79 | 20920 | Richmond city, VA | 10.97 |
| 28.37 | 192414 | Riverside County, CA | 7.22 |
| 93.12 | 12742 | Robeson County, NC | 11.58 |
| 64.47 | 12737 | Rock County, WI | 8.23 |
| 76.81 | 11018 | Rock Island County, IL | 8.57 |
| 25.76 | 19034 | Rockingham County, NH | 6.17 |
| 21.10 | 43343 | Rockland County, NY | 5.69 |
| 66.20 | 11123 | Rowan County, NC | 9.62 |
| 41.48 | 30611 | Rutherford County, TN | 8.60 |
| 27.87 | 127643 | Sacramento County, CA | 7.21 |
| 78.74 | 13891 | Saginaw County, MI | 9.77 |
| 62.29 | 9615 | Saline County, AR | 7.85 |
| 27.27 | 105826 | Salt Lake County, UT | 7.73 |
| 31.95 | 187306 | San Bernardino County, CA | 8.13 |
| 13.98 | 262290 | San Diego County, CA | 6.94 |
| 3.15 | 53480 | San Francisco County, CA | 7.21 |
| 28.64 | 68585 | San Joaquin County, CA | 7.71 |
| 75.94 | 9185 | San Juan County, NM | 7.20 |
| 12.69 | 16917 | San Luis Obispo County, CA | 5.98 |
| 3.18 | 53632 | San Mateo County, CA | 6.67 |
| 50.90 | 9297 | Sandoval County, NM | 8.48 |
| 69.76 | 14439 | Sangamon County, IL | 11.58 |
| 17.02 | 38130 | Santa Barbara County, CA | 6.56 |
| 5.05 | 137356 | Santa Clara County, CA | 7.29 |
| 8.58 | 15846 | Santa Cruz County, CA | 5.74 |
| 38.81 | 7907 | Santa Fe County, NM | 10.29 |
| 48.30 | 13795 | Santa Rosa County, FL | 7.79 |
| 38.87 | 20475 | Sarasota County, FL | 7.49 |
| 40.87 | 14113 | Saratoga County, NY | 6.19 |
| 45.82 | 16792 | Sarpy County, NE | 7.14 |
| 62.46 | 12448 | Schenectady County, NY | 9.05 |
| 79.37 | 8894 | Schuylkill County, PA | 7.65 |
| 61.41 | 13720 | Scott County, IA | 7.73 |
| 31.07 | 11747 | Scott County, MN | 6.25 |
| 77.36 | 11210 | Sebastian County, AR | 8.01 |
| 68.53 | 46179 | Sedgwick County, KS | 8.35 |
| 40.38 | 31709 | Seminole County, FL | 7.92 |
| 44.05 | 12870 | Shasta County, CA | 6.40 |
| 72.58 | 13889 | Shawnee County, KS | 8.03 |
| 62.28 | 8041 | Sheboygan County, WI | 6.60 |
| 45.87 | 16085 | Shelby County, AL | 8.56 |
| 66.30 | 86868 | Shelby County, TN | 11.97 |
| 28.34 | 9624 | Skagit County, WA | 5.18 |
| 63.08 | 21093 | Smith County, TX | 7.76 |
| 17.55 | 68121 | Snohomish County, WA | 6.71 |
| 20.56 | 34136 | Solano County, CA | 7.36 |
| 21.63 | 22160 | Somerset County, NJ | 7.02 |
| 14.25 | 30993 | Sonoma County, CA | 5.89 |
| 66.48 | 28675 | Spartanburg County, SC | 8.78 |
| 40.64 | 40755 | Spokane County, WA | 6.89 |
| 33.95 | 11116 | Spotsylvania County, VA | 7.35 |
| 44.15 | 29232 | St. Charles County, MO | 7.29 |
| 70.57 | 19164 | St. Clair County, IL | 11.12 |
| 64.39 | 10308 | St. Clair County, MI | 6.64 |
| 29.10 | 16651 | St. Johns County, FL | 6.88 |
| 71.71 | 22927 | St. Joseph County, IN | 8.93 |
| 83.17 | 6920 | St. Lawrence County, NY | 6.09 |
| 63.29 | 12515 | St. Louis County, MN | 6.79 |
| 53.57 | 73959 | St. Louis County, MO | 10.13 |
| 70.84 | 25311 | St. Louis city, MO | 12.45 |
| 50.80 | 23974 | St. Lucie County, FL | 9.01 |
| 31.67 | 9557 | St. Mary's County, MD | 6.37 |
| 50.75 | 20203 | St. Tammany Parish, LA | 8.56 |
| 24.63 | 12630 | Stafford County, VA | 7.75 |
| 33.77 | 49716 | Stanislaus County, CA | 6.54 |
| 71.61 | 27695 | Stark County, OH | 8.52 |
| 54.03 | 14064 | Stearns County, MN | 6.36 |
| 40.55 | 8060 | Strafford County, NH | 6.23 |
| 16.54 | 59170 | Suffolk County, MA | 8.20 |
| 19.74 | 108298 | Suffolk County, NY | 8.06 |

The first 500 of 578 rows; every row is in https://vishalsingh.org/zipcode/report/birthweight-and-preterm-birth/figures/lbw_by_deprivation.json.

#### Table 2.2. Preterm and low birthweight

Preterm birth is more common than low birthweight in every decile and rises more, from 8.8% to 12.1%

Births-weighted share of live births before 37 weeks and under 2,500 grams by decile of county deprivation, 2018 to 2024

| conf_high | conf_low | decile | estimate | measure | n |
| ---: | ---: | ---: | ---: | --- | ---: |
| 9.01 | 8.59 | 1 | 8.82 | Preterm birth | 4,151,203 |
| 7.68 | 7.18 | 1 | 7.43 | Low birthweight | 4,150,007 |
| 9.48 | 9.08 | 2 | 9.28 | Preterm birth | 2,647,866 |
| 8.00 | 7.32 | 2 | 7.63 | Low birthweight | 2,646,210 |
| 10.12 | 9.47 | 3 | 9.82 | Preterm birth | 2,523,259 |
| 8.36 | 7.55 | 3 | 7.96 | Low birthweight | 2,522,181 |
| 10.29 | 9.80 | 4 | 10.05 | Preterm birth | 1,825,582 |
| 8.67 | 7.84 | 4 | 8.29 | Low birthweight | 1,824,679 |
| 10.53 | 9.98 | 5 | 10.27 | Preterm birth | 1,923,159 |
| 8.65 | 7.97 | 5 | 8.37 | Low birthweight | 1,922,260 |
| 11.28 | 10.26 | 6 | 10.80 | Preterm birth | 2,196,492 |
| 8.98 | 8.29 | 6 | 8.69 | Low birthweight | 2,195,346 |
| 11.22 | 10.56 | 7 | 10.92 | Preterm birth | 1,520,420 |
| 9.50 | 8.57 | 7 | 9.05 | Low birthweight | 1,519,471 |
| 11.59 | 10.92 | 8 | 11.27 | Preterm birth | 1,469,109 |
| 9.85 | 8.89 | 8 | 9.40 | Low birthweight | 1,468,025 |
| 11.82 | 11.01 | 9 | 11.42 | Preterm birth | 1,269,566 |
| 10.18 | 9.00 | 9 | 9.63 | Low birthweight | 1,268,228 |
| 12.66 | 11.32 | 10 | 12.05 | Preterm birth | 891,594 |
| 10.26 | 9.20 | 10 | 9.68 | Low birthweight | 890,018 |

#### Table 2.3. All counties

Across all counties with a published rate the share is 7.6% in the least deprived tenth and 10.9% in the most deprived, and levels off in between

Births-weighted low-birthweight share by decile of county deprivation rank, 3,040 counties, County Health Rankings, births in 2016 to 2022

| conf_high | conf_low | decile | estimate | n |
| ---: | ---: | ---: | ---: | ---: |
| 7.77 | 7.41 | 1 | 7.58 | 10,018,514 |
| 8.44 | 8.00 | 2 | 8.24 | 4,922,587 |
| 8.81 | 8.36 | 3 | 8.62 | 3,381,491 |
| 9.50 | 8.74 | 4 | 9.14 | 2,597,848 |
| 9.65 | 8.70 | 5 | 9.19 | 1,741,523 |
| 9.56 | 8.80 | 6 | 9.17 | 1,216,083 |
| 9.29 | 8.56 | 7 | 8.89 | 846,209 |
| 9.43 | 8.95 | 8 | 9.17 | 752,806 |
| 10.23 | 9.53 | 9 | 9.88 | 447,237 |
| 11.30 | 10.56 | 10 | 10.93 | 327,359 |

Note: The counties are the 578 that WONDER natality names individually (at least 100,000 residents), which hold 79.5% of births. WONDER pools the births of every other county in each state, so those counties are not in this figure.

Note: A county's deprivation rank is the population-weighted mean of the Area Deprivation Index national ranks of its census tracts (1 is least deprived, 100 most), weighted by 2019-2023 tract population. Deciles hold equal numbers of these counties, so the most deprived tenth is relative to large counties and not to the nation.

Note: WONDER withholds cells of 1 to 9 births. A withheld cell is missing: it is left out of both the count and the denominator. The birthweight cut has 11,250 withheld cells in these counties; at 9 births each they could hold at most 0.5% of births, so a pooled share could move by at most that many percentage points.

Note: Decile means are births-weighted. Intervals are 95% intervals from a bootstrap over counties, described in the methods note. The rates pool births in 2018 to 2024.

Note: The third tab uses County Health Rankings rates for every county with a published rate, not the WONDER counts, and deciles of all counties, so its most deprived tenth is made up of small counties.

Source: CDC WONDER natality, births in 2018-2024, and Area Deprivation Index 2023; full counts of births, county shares weighted by births (population-weighted); third tab: County Health Rankings 2024, births in 2016-2022, births-weighted (population-weighted)

Data: https://vishalsingh.org/zipcode/report/birthweight-and-preterm-birth/figures/lbw_by_deprivation.json

### Table 3. Non-Hispanic Black mothers had low-birthweight births at 14.0% against 6.7% for non-Hispanic white mothers, a ratio of 2.1

First panel: share of live births under 2,500 grams in the large counties, by mother's race and ethnicity, County Health Rankings, births in 2016 to 2022. Second panel: infant deaths per 1,000 live births in the United States by mother's single race, any ethnicity, 2018 and 2019

| conf_high | conf_low | label | n | panel | race | value |
| ---: | ---: | --- | ---: | --- | --- | ---: |
| 14.19 | 13.75 | Black (non-Hispanic) | 3,249,387 | Low birthweight (% of live births) | Black | 13.98 |
| 7.70 | 7.39 | Hispanic (any race) | 5,632,060 | Low birthweight (% of live births) | Hispanic | 7.54 |
| 6.82 | 6.64 | White (non-Hispanic) | 9,614,256 | Low birthweight (% of live births) | White | 6.73 |
| 10.49 | 10.14 | Black (any ethnicity) | 1,255,629 | Infant mortality (deaths per 1,000 live births) | Black | 10.32 |
| 4.78 | 4.66 | White (any ethnicity) | 5,617,093 | Infant mortality (deaths per 1,000 live births) | White | 4.72 |

Note: Low birthweight comes from County Health Rankings county rates for the 578 large counties. County Health Rankings counts mothers as non-Hispanic Black, non-Hispanic white or Hispanic of any race. It publishes each rate with its interval but not the births behind it; the births are recovered from the interval, and each group's rate is pooled over the counties where the rate is published.

Note: Infant mortality is pooled over states, so it covers all counties, not only the large ones. WONDER's linked file counts mothers by single race of any ethnicity, so its white group includes Hispanic mothers and is not the same group as in the first panel. WONDER withholds a state's deaths for a race with 1 to 9 deaths; those cells leave both the deaths and the births.

Note: Low-birthweight intervals come from a bootstrap that resamples whole counties. Infant mortality intervals are gamma or normal intervals on the count, which treat births as independent and leave out differences between states.

Source: County Health Rankings 2024, births in 2016-2022 (rates pooled by births, population-weighted); CDC WONDER linked birth and infant death, 2018-2019, mother's race (crude rates)

Data: https://vishalsingh.org/zipcode/report/birthweight-and-preterm-birth/figures/birth_outcomes_by_race.json

### Table 4. Low birthweight is more common for non-Hispanic Black mothers in the least deprived quartile of counties (12.4%) than for non-Hispanic white mothers in the most deprived (7.6%)

Low birthweight (share of live births under 2,500 grams, births in 2016 to 2022) or infant mortality (deaths per 1,000 live births, 2018 and 2019) by quartile of county deprivation rank and mother's race

| conf_high | conf_low | counties | measure | n | quartile | race | rate |
| ---: | ---: | ---: | --- | ---: | --- | --- | ---: |
| 12.76 | 11.95 | 133 | lbw | 830,778 | Least deprived quartile | Black (non-Hispanic) | 12.37 |
| 6.34 | 6.05 | 144 | lbw | 3,597,717 | Least deprived quartile | White (non-Hispanic) | 6.18 |
| 14.02 | 13.46 | 137 | lbw | 862,783 | Second quartile | Black (non-Hispanic) | 13.76 |
| 6.75 | 6.53 | 145 | lbw | 2,307,987 | Second quartile | White (non-Hispanic) | 6.64 |
| 14.78 | 14.14 | 139 | lbw | 823,241 | Third quartile | Black (non-Hispanic) | 14.46 |
| 7.23 | 7.05 | 144 | lbw | 2,156,968 | Third quartile | White (non-Hispanic) | 7.14 |
| 15.80 | 15.25 | 140 | lbw | 732,585 | Most deprived quartile | Black (non-Hispanic) | 15.53 |
| 7.70 | 7.45 | 145 | lbw | 1,551,584 | Most deprived quartile | White (non-Hispanic) | 7.58 |
| 9.29 | 7.87 | 65 | imr | 229,878 | Least deprived quartile | Black (any ethnicity) | 8.55 |
| 4 | 3.54 | 95 | imr | 1,555,633 | Least deprived quartile | White (any ethnicity) | 3.76 |
| 10.73 | 9.51 | 43 | imr | 214,646 | Second quartile | Black (any ethnicity) | 10.15 |
| 4.74 | 4.19 | 58 | imr | 728,879 | Second quartile | White (any ethnicity) | 4.47 |
| 12.07 | 10.68 | 39 | imr | 200,671 | Third quartile | Black (any ethnicity) | 11.33 |
| 5.42 | 4.87 | 50 | imr | 602,598 | Third quartile | White (any ethnicity) | 5.12 |
| 13.24 | 11.54 | 33 | imr | 148,514 | Most deprived quartile | Black (any ethnicity) | 12.45 |
| 5.67 | 4.93 | 39 | imr | 338,122 | Most deprived quartile | White (any ethnicity) | 5.29 |

Views: Low birthweight (measure = lbw); Infant mortality (measure = imr).

Note: Counties are the 578 large counties, ranked by deprivation as in the gradient figure. The low-birthweight view pools County Health Rankings rates over the counties where each group's rate is published, for non-Hispanic Black and non-Hispanic white mothers; the counts of counties are in the data table.

Note: The infant mortality view covers counties of 250,000 residents or more, where WONDER lists deaths by mother's race, for births in 2018 and 2019, with mothers counted by single race of any ethnicity (the white group includes Hispanic mothers). WONDER withholds a county-year with 1 to 9 deaths. Those county-years are left out of both deaths and births, which removes 13.6% of Black births and 5.2% of white births in these counties, so the bars describe the county-years with at least 10 infant deaths.

Note: Intervals are 95% intervals from a bootstrap that resamples whole counties, described in the methods note, with the same draws for both groups so that gaps have their own interval.

Source: County Health Rankings 2024, births in 2016-2022, and CDC WONDER linked birth and infant death, 2018-2019; rates pooled by births (population-weighted)

Data: https://vishalsingh.org/zipcode/report/birthweight-and-preterm-birth/figures/weathering.json

### Table 5. Low birthweight is steadily worse toward the more deprived end under the ADI; the two ends do not differ detectably under the degree-share ranking

Births under 2,500 grams (% of live births) by tenth of counties ranked four ways, over the 578 counties that have all four rankings

| bin | conf_high | conf_low | estimate | n_units | ranking |
| ---: | ---: | ---: | ---: | ---: | --- |
| 1 | 7.68 | 7.18 | 7.43 | 57 | Area Deprivation Index |
| 2 | 8.00 | 7.32 | 7.63 | 58 | Area Deprivation Index |
| 3 | 8.47 | 7.63 | 8.07 | 58 | Area Deprivation Index |
| 4 | 8.52 | 7.74 | 8.15 | 58 | Area Deprivation Index |
| 5 | 8.63 | 7.99 | 8.36 | 58 | Area Deprivation Index |
| 6 | 9.01 | 8.29 | 8.73 | 57 | Area Deprivation Index |
| 7 | 9.50 | 8.59 | 9.04 | 58 | Area Deprivation Index |
| 8 | 9.58 | 8.68 | 9.15 | 58 | Area Deprivation Index |
| 9 | 10.36 | 9.28 | 9.86 | 58 | Area Deprivation Index |
| 10 | 10.17 | 9.14 | 9.63 | 58 | Area Deprivation Index |
| 1 | 7.32 | 6.96 | 7.16 | 57 | Standardized index |
| 2 | 7.46 | 7.03 | 7.24 | 58 | Standardized index |
| 3 | 8.42 | 7.41 | 7.92 | 58 | Standardized index |
| 4 | 8.62 | 7.91 | 8.24 | 58 | Standardized index |
| 5 | 8.54 | 7.75 | 8.17 | 58 | Standardized index |
| 6 | 8.85 | 7.89 | 8.28 | 57 | Standardized index |
| 7 | 9.15 | 8.54 | 8.86 | 58 | Standardized index |
| 8 | 9.22 | 8.64 | 8.93 | 58 | Standardized index |
| 9 | 10.15 | 8.78 | 9.42 | 58 | Standardized index |
| 10 | 10.37 | 8.98 | 9.73 | 58 | Standardized index |
| 1 | 7.32 | 6.99 | 7.16 | 57 | Median household income |
| 2 | 7.64 | 7.04 | 7.33 | 58 | Median household income |
| 3 | 8.43 | 7.65 | 8.01 | 58 | Median household income |
| 4 | 8.44 | 7.71 | 8.01 | 58 | Median household income |
| 5 | 8.70 | 8.03 | 8.40 | 58 | Median household income |
| 6 | 8.95 | 8.32 | 8.72 | 57 | Median household income |
| 7 | 8.87 | 8.26 | 8.56 | 58 | Median household income |
| 8 | 9.46 | 8.54 | 9.03 | 58 | Median household income |
| 9 | 10.49 | 9.55 | 10.08 | 58 | Median household income |
| 10 | 10.65 | 9.42 | 9.99 | 58 | Median household income |
| 1 | 7.95 | 7.22 | 7.55 | 57 | Bachelor's degree share |
| 2 | 8.43 | 7.66 | 8.04 | 58 | Bachelor's degree share |
| 3 | 8.41 | 7.56 | 7.95 | 58 | Bachelor's degree share |
| 4 | 8.60 | 7.79 | 8.26 | 58 | Bachelor's degree share |
| 5 | 8.93 | 8.01 | 8.46 | 58 | Bachelor's degree share |
| 6 | 9.15 | 7.99 | 8.43 | 57 | Bachelor's degree share |
| 7 | 9.42 | 8.48 | 8.96 | 58 | Bachelor's degree share |
| 8 | 9.39 | 8.56 | 9.01 | 58 | Bachelor's degree share |
| 9 | 9.66 | 8.12 | 8.85 | 58 | Bachelor's degree share |
| 10 | 8.58 | 7.76 | 8.16 | 58 | Bachelor's degree share |

Note: Least and most deprived tenth: ADI 7.4% and 9.6%, rank correlation +0.99; standardized index 7.2% and 9.7%, +0.99; median household income 7.2% and 10.0%, +0.96; bachelor's degree share 7.6% and 8.2%, +0.68.

Note: The standardized index applies the ADI's coefficients to its measures after standardizing each one, so home value and income carry less weight than in the ADI. Income and degree share are single census measures, each ranked from the highest value to the lowest.

Source: CDC WONDER natality, births in 2018-2024, and Area Deprivation Index 2023; full counts of births, county shares weighted by births (population-weighted); ADI 2023 and ACS 2019-2023 tract tables (B19013, B15003 and the 15 rebuilt ADI measures)

Data: https://vishalsingh.org/zipcode/report/birthweight-and-preterm-birth/figures/alt_rankings.json

## All estimates

Every fact in facts.json (79), sorted by id.

| Fact id | Value | Unit | Estimand | Population | Period | Interval | n | Weights | Source | Note |
| --- | ---: | --- | --- | --- | --- | --- | ---: | --- | --- | --- |
| alt_adi_least | 7.4% | Births under 2,500 grams (% of live births) | births-weighted mean of county shares (the pooled share); 95% intervals from a bootstrap over counties | the least deprived tenth of counties ranked by Area Deprivation Index, among the counties with all four rankings | births in 2018 to 2024 | 95% CI 7.2% to 7.7% | 57 counties | births in each county | CDC WONDER natality |   |
| alt_adi_most | 9.6% | Births under 2,500 grams (% of live births) | births-weighted mean of county shares (the pooled share); 95% intervals from a bootstrap over counties | the most deprived tenth of counties ranked by Area Deprivation Index, among the counties with all four rankings | births in 2018 to 2024 | 95% CI 9.1% to 10.2% | 58 counties | births in each county | CDC WONDER natality |   |
| alt_adi_rho | +0.99 | rank correlation | Spearman rank correlation between bin order and bin mean, each bin counted once | the bins of counties ranked by Area Deprivation Index | births in 2018 to 2024 |   |   | each bin counted once | CDC WONDER natality |   |
| alt_degree_least | 7.6% | Births under 2,500 grams (% of live births) | births-weighted mean of county shares (the pooled share); 95% intervals from a bootstrap over counties | the least deprived tenth of counties ranked by Bachelor's degree share, among the counties with all four rankings | births in 2018 to 2024 | 95% CI 7.2% to 7.9% | 57 counties | births in each county | CDC WONDER natality |   |
| alt_degree_most | 8.2% | Births under 2,500 grams (% of live births) | births-weighted mean of county shares (the pooled share); 95% intervals from a bootstrap over counties | the most deprived tenth of counties ranked by Bachelor's degree share, among the counties with all four rankings | births in 2018 to 2024 | 95% CI 7.8% to 8.6% | 58 counties | births in each county | CDC WONDER natality |   |
| alt_degree_rho | +0.68 | rank correlation | Spearman rank correlation between bin order and bin mean, each bin counted once | the bins of counties ranked by Bachelor's degree share | births in 2018 to 2024 |   |   | each bin counted once | CDC WONDER natality |   |
| alt_income_least | 7.2% | Births under 2,500 grams (% of live births) | births-weighted mean of county shares (the pooled share); 95% intervals from a bootstrap over counties | the least deprived tenth of counties ranked by Median household income, among the counties with all four rankings | births in 2018 to 2024 | 95% CI 7.0% to 7.3% | 57 counties | births in each county | CDC WONDER natality |   |
| alt_income_most | 10.0% | Births under 2,500 grams (% of live births) | births-weighted mean of county shares (the pooled share); 95% intervals from a bootstrap over counties | the most deprived tenth of counties ranked by Median household income, among the counties with all four rankings | births in 2018 to 2024 | 95% CI 9.4% to 10.6% | 58 counties | births in each county | CDC WONDER natality |   |
| alt_income_rho | +0.96 | rank correlation | Spearman rank correlation between bin order and bin mean, each bin counted once | the bins of counties ranked by Median household income | births in 2018 to 2024 |   |   | each bin counted once | CDC WONDER natality |   |
| alt_n_places | 578 | counties | count of counties in the common sample of the ranking comparison | counties with the outcome and all four rankings | births in 2018 to 2024 |   |   | none (full count) | CDC WONDER natality |   |
| alt_rho_cut | 0.9 | absolute rank correlation | threshold between a steady and an uneven gradient | the rule for changed gradients (docs/RANKING_SWITCH_RULE.md) | births in 2018 to 2024 |   |   | none (a fixed rule) | CDC WONDER natality |   |
| alt_rho_margin | 0.1 | absolute rank correlation | smallest move in the rank correlation that counts as a change of shape | the rule for changed gradients (docs/RANKING_SWITCH_RULE.md) | births in 2018 to 2024 |   |   | none (a fixed rule) | CDC WONDER natality |   |
| alt_std_least | 7.2% | Births under 2,500 grams (% of live births) | births-weighted mean of county shares (the pooled share); 95% intervals from a bootstrap over counties | the least deprived tenth of counties ranked by Standardized index, among the counties with all four rankings | births in 2018 to 2024 | 95% CI 7.0% to 7.3% | 57 counties | births in each county | CDC WONDER natality |   |
| alt_std_most | 9.7% | Births under 2,500 grams (% of live births) | births-weighted mean of county shares (the pooled share); 95% intervals from a bootstrap over counties | the most deprived tenth of counties ranked by Standardized index, among the counties with all four rankings | births in 2018 to 2024 | 95% CI 9.0% to 10.4% | 58 counties | births in each county | CDC WONDER natality |   |
| alt_std_rho | +0.99 | rank correlation | Spearman rank correlation between bin order and bin mean, each bin counted once | the bins of counties ranked by Standardized index | births in 2018 to 2024 |   |   | each bin counted once | CDC WONDER natality |   |
| births_2018 | 3.79 million | live births | count of live births | live births in the United States | 2018 |   |   | none (full count) | CDC WONDER natality |   |
| births_2024 | 3.63 million | live births | count of live births | live births in the United States | 2024 |   |   | none (full count) | CDC WONDER natality |   |
| births_covered | 20.5 million | live births | count of live births | live births in the counties that CDC WONDER natality names individually (counties of 100,000 residents or more) | births in 2018 to 2024 |   | 20,450,654 births | none (full count) | CDC WONDER natality |   |
| births_decline_pct | 4.3% | percent of 2018 births | fall in the count of births from 2018 to 2024 | live births in the United States | 2018 to 2024 |   |   | none (full count) | CDC WONDER natality |   |
| births_median_all_d1 | 12,055 | live births | median of the county birth counts behind the rate (CHR denominator) | the median county in the least deprived tenth of all counties with a County Health Rankings rate and an ADI | births in 2016 to 2022 |   | 304 counties | none (each county counts once) | County Health Rankings |   |
| births_median_all_d10 | 770 | live births | median of the county birth counts behind the rate (CHR denominator) | the median county in the most deprived tenth of all counties with a County Health Rankings rate and an ADI | births in 2016 to 2022 |   | 304 counties | none (each county counts once) | County Health Rankings |   |
| births_share_covered | 79.5% | percent of live births in the United States | share of births in the named counties | live births in the 50 states and DC | births in 2018 to 2024 |   | 25,709,900 births | none (full count) | CDC WONDER natality |   |
| chr_wonder_corr | 0.984 | correlation coefficient | births-weighted correlation between the county low-birthweight share in WONDER and in County Health Rankings, county-cluster bootstrap interval | the named counties | WONDER births in 2018 to 2024 and CHR births in 2016 to 2022 | 95% CI 0.980 to 0.987 | 578 counties | WONDER births in each county | County Health Rankings |   |
| counties_all_gradient | 3,040 | counties | count of counties | counties with a County Health Rankings low-birthweight rate and an ADI | births in 2016 to 2022 |   |   | none (full count) | County Health Rankings |   |
| counties_below_floor | 145 | counties | count of counties with a rate but fewer than 20 low-birthweight births, which the map leaves as no data | counties and county equivalents on the map (50 states and DC) | births in 2016 to 2022 |   |   | none (full count) | County Health Rankings |   |
| counties_mapped | 2,900 | counties | count of counties with a rate and at least 20 low-birthweight births | counties and county equivalents on the map (50 states and DC) | births in 2016 to 2022 |   |   | none (full count) | County Health Rankings |   |
| counties_no_rate | 98 | counties | count of counties for which County Health Rankings publishes no low-birthweight rate | counties and county equivalents on the map (50 states and DC) | births in 2016 to 2022 |   |   | none (full count) | County Health Rankings |   |
| counties_not_mapped | 243 | counties | count of counties with no rate or fewer than 20 low-birthweight births | counties and county equivalents on the map (50 states and DC) | births in 2016 to 2022 |   |   | none (full count) | County Health Rankings |   |
| counties_used | 578 | counties | count of counties | counties that WONDER natality names individually and that have births | births in 2018 to 2024 |   |   | none (full count) | CDC WONDER natality |   |
| deep_south_all_share | 14% | percent of counties | share of mapped counties in Alabama, Georgia, Louisiana, Mississippi and South Carolina | all mapped counties | births in 2016 to 2022 |   | 2,900 counties | none (each county counts once) | County Health Rankings |   |
| deep_south_top_share | 64% | percent of counties | share of those counties in Alabama, Georgia, Louisiana, Mississippi and South Carolina | mapped counties in the highest tenth of low-birthweight shares | births in 2016 to 2022 |   | 290 counties | none (each county counts once) | County Health Rankings |   |
| im_black | 10.3 deaths per 1,000 live births | infant deaths per 1,000 live births | crude rate from state counts, normal interval | live births to Black mothers (single race, any ethnicity) in the United States, states with a published count | births in 2018 and 2019 | 95% CI 10.10 to 10.50 | 12,953 infant deaths | none (crude rate from counts) | CDC WONDER linked birth and infant death |   |
| im_left_out_black | 13.6% | percent of live births | share of births in county-years whose deaths WONDER withholds, which the county view leaves out of deaths and births | live births to black mothers (single race, any ethnicity) in counties of 250,000 residents or more | births in 2018 and 2019 |   |   | none (full counts of births) | CDC WONDER linked birth and infant death |   |
| im_left_out_white | 5.2% | percent of live births | share of births in county-years whose deaths WONDER withholds, which the county view leaves out of deaths and births | live births to white mothers (single race, any ethnicity) in counties of 250,000 residents or more | births in 2018 and 2019 |   |   | none (full counts of births) | CDC WONDER linked birth and infant death |   |
| im_white | 4.7 | infant deaths per 1,000 live births | crude rate from state counts, normal interval | live births to white mothers (single race, any ethnicity) in the United States, states with a published count | births in 2018 and 2019 | 95% CI 4.7 to 4.8 | 26,516 infant deaths | none (crude rate from counts) | CDC WONDER linked birth and infant death |   |
| lbd_county_threshold | 250,000 | residents | population size at which WONDER names a county in the linked file (a definition) | counties that CDC WONDER linked birth and infant death names individually | births in 2018 and 2019 |   |   | none (definition) | CDC WONDER linked birth and infant death |   |
| lbw_adi_corr | 0.52 | correlation coefficient | births-weighted correlation of the county low-birthweight share with county deprivation rank, county-cluster bootstrap interval | live births in the named counties | births in 2018 to 2024 | 95% CI 0.45 to 0.58 | 578 counties | births in each county | CDC WONDER natality |   |
| lbw_all_d1 | 7.6% | percent of live births under 2,500 grams | births-weighted mean of county shares, bootstrap interval over counties | live births in the least deprived tenth of all counties with a County Health Rankings rate and an ADI | births in 2016 to 2022 | 95% CI 7.4% to 7.8% | 304 counties | births in each county (CHR denominator) | County Health Rankings |   |
| lbw_all_d10 | 10.9% | percent of live births under 2,500 grams | births-weighted mean of county shares, bootstrap interval over counties | live births in the most deprived tenth of all counties with a County Health Rankings rate and an ADI | births in 2016 to 2022 | 95% CI 10.6% to 11.3% | 304 counties | births in each county (CHR denominator) | County Health Rankings |   |
| lbw_b1_w4_diff | 4.8 percentage points | percentage points | difference of two pooled low-birthweight shares, county-cluster bootstrap interval | live births in the named counties: non-Hispanic Black mothers in the least deprived quartile minus non-Hispanic white mothers in the most deprived quartile | births in 2016 to 2022 | 95% CI 4.4 percentage points to 5.2 percentage points | 578 counties | births to mothers of the group in each county | County Health Rankings |   |
| lbw_black | 14.0% | percent of live births under 2,500 grams | pooled share of births, from the published county rates and the births they imply, bootstrap interval over counties | live births to non-Hispanic Black mothers in the named counties where County Health Rankings publishes the rate | births in 2016 to 2022 | 95% CI 13.8% to 14.2% | 549 counties | births to mothers of the group in each county | County Health Rankings |   |
| lbw_black_q1 | 12.4% | percent of live births under 2,500 grams | pooled share of births, bootstrap interval over counties | live births to non-Hispanic Black mothers in the least deprived quartile of the named counties by deprivation rank | births in 2016 to 2022 | 95% CI 11.9% to 12.8% | 133 counties | births to mothers of the group in each county | County Health Rankings |   |
| lbw_cutoff_grams | 2,500 | grams | birthweight below which a live birth counts as low birthweight (a definition) | live births | all years (a definition) |   |   | none (definition) | CDC WONDER natality |   |
| lbw_d1 | 7.4% | percent of live births under 2,500 grams | births-weighted mean of county shares (the pooled share), bootstrap interval over counties | live births in the least deprived tenth of the named counties by county deprivation rank | births in 2018 to 2024 | 95% CI 7.2% to 7.7% | 57 counties | births in each county | CDC WONDER natality |   |
| lbw_d10 | 9.7% | percent of live births under 2,500 grams | births-weighted mean of county shares (the pooled share), bootstrap interval over counties | live births in the most deprived tenth of the named counties by county deprivation rank | births in 2018 to 2024 | 95% CI 9.2% to 10.3% | 58 counties | births in each county | CDC WONDER natality |   |
| lbw_gap_pp | 7.3 percentage points | percentage points | difference of pooled low-birthweight shares, county-cluster bootstrap interval | live births in the named counties, non-Hispanic Black mothers minus non-Hispanic white mothers | births in 2016 to 2022 | 95% CI 7.0 percentage points to 7.4 percentage points | 578 counties | births to mothers of the group in each county | County Health Rankings |   |
| lbw_gap_q1 | 6.2 percentage points | percentage points | difference of pooled low-birthweight shares, county-cluster bootstrap interval | live births in the least deprived quartile of the named counties, non-Hispanic Black mothers minus non-Hispanic white mothers | births in 2016 to 2022 | 95% CI 5.8 percentage points to 6.6 percentage points | 578 counties | births to mothers of the group in each county | County Health Rankings |   |
| lbw_gap_q4 | 8.0 percentage points | percentage points | difference of pooled low-birthweight shares, county-cluster bootstrap interval | live births in the most deprived quartile of the named counties, non-Hispanic Black mothers minus non-Hispanic white mothers | births in 2016 to 2022 | 95% CI 7.7 percentage points to 8.2 percentage points | 578 counties | births to mothers of the group in each county | County Health Rankings |   |
| lbw_gap_widening | 1.8 percentage points | percentage points | difference of two Black-white gaps in pooled low-birthweight shares, county-cluster bootstrap interval | live births in the named counties, gap between non-Hispanic Black and non-Hispanic white mothers in the most deprived quartile minus the gap in the least deprived quartile | births in 2016 to 2022 | 95% CI 1.3 percentage points to 2.3 percentage points | 578 counties | births to mothers of the group in each county | County Health Rankings |   |
| lbw_hispanic | 7.5% | percent of live births under 2,500 grams | pooled share of births, from the published county rates and the births they imply, bootstrap interval over counties | live births to Hispanic (any race) mothers in the named counties where County Health Rankings publishes the rate | births in 2016 to 2022 | 95% CI 7.4% to 7.7% | 575 counties | births to mothers of the group in each county | County Health Rankings |   |
| lbw_natl | 8.3% | percent of live births under 2,500 grams | pooled share of births in published cells (events over births), Wilson interval | live births in the counties that CDC WONDER natality names individually (counties of 100,000 residents or more), births in published cells | births in 2018 to 2024 | 95% CI 8.3% to 8.3% | 20,406,425 births | none (full counts of births) | CDC WONDER natality | WONDER withholds cells of 1 to 9 births; a withheld cell is left out of both the count and the denominator. |
| lbw_preterm_corr | 0.84 | correlation coefficient | births-weighted correlation between the county low-birthweight share and the county preterm share, county-cluster bootstrap interval | live births in the named counties | births in 2018 to 2024 | 95% CI 0.81 to 0.87 | 578 counties | births in each county | CDC WONDER natality |   |
| lbw_range_pp | 2.3 percentage points | percentage points | difference of two births-weighted decile means of the low-birthweight share, bootstrap interval over counties | live births in the named counties, most deprived tenth minus least deprived tenth | births in 2018 to 2024 | 95% CI 1.7 percentage points to 2.9 percentage points | 578 counties | births in each county | CDC WONDER natality |   |
| lbw_ratio_bw | 2.1 | ratio of rates | ratio of pooled low-birthweight shares, county-cluster bootstrap interval | live births in the named counties, non-Hispanic Black mothers over non-Hispanic white mothers | births in 2016 to 2022 | 95% CI 2.0 to 2.1 | 578 counties | births to mothers of the group in each county | County Health Rankings |   |
| lbw_rest | 7.8% | percent of live births under 2,500 grams | pooled share of births, bootstrap interval over counties | live births in mapped counties outside the Census South region | births in 2016 to 2022 | 95% CI 7.7% to 8.0% | 1,529 counties | births in each county (CHR denominator) | County Health Rankings |   |
| lbw_slope_per_10 | 0.38 percentage points | percentage points of the low-birthweight share per ten points of county deprivation rank | births-weighted slope of the county low-birthweight share on county deprivation rank, county-cluster bootstrap interval | live births in the named counties | births in 2018 to 2024 | 95% CI 0.32 percentage points to 0.44 percentage points | 578 counties | births in each county | CDC WONDER natality |   |
| lbw_south | 9.1% | percent of live births under 2,500 grams | pooled share of births, bootstrap interval over counties | live births in mapped counties in the Census South region (16 states and DC) | births in 2016 to 2022 | 95% CI 9.0% to 9.3% | 1,371 counties | births in each county (CHR denominator) | County Health Rankings |   |
| lbw_wh_low_d1 | 1.0 | withheld cells per county-year | withheld birthweight cells below the low-birthweight cutoff, per county-year | county-years of the least deprived tenth of the named counties | births in 2018 to 2024 |   | 399 county-years | none (full count) | CDC WONDER natality |   |
| lbw_wh_low_d10 | 1.6 | withheld cells per county-year | withheld birthweight cells below the low-birthweight cutoff, per county-year | county-years of the most deprived tenth of the named counties | births in 2018 to 2024 |   | 406 county-years | none (full count) | CDC WONDER natality |   |
| lbw_wh_max_d1 | 0.2% | percent of live births | largest share of births the withheld birthweight cells could hold: 9 births in each withheld cell, over all births in these counties | live births in the least deprived tenth of the named counties | births in 2018 to 2024 |   |   | none (an upper limit from the cell count) | CDC WONDER natality |   |
| lbw_wh_max_d10 | 1.3% | percent of live births | largest share of births the withheld birthweight cells could hold: 9 births in each withheld cell, over all births in these counties | live births in the most deprived tenth of the named counties | births in 2018 to 2024 |   |   | none (an upper limit from the cell count) | CDC WONDER natality |   |
| lbw_white | 6.7% | percent of live births under 2,500 grams | pooled share of births, from the published county rates and the births they imply, bootstrap interval over counties | live births to non-Hispanic white mothers in the named counties where County Health Rankings publishes the rate | births in 2016 to 2022 | 95% CI 6.6% to 6.8% | 578 counties | births to mothers of the group in each county | County Health Rankings |   |
| lbw_white_q4 | 7.6% | percent of live births under 2,500 grams | pooled share of births, bootstrap interval over counties | live births to non-Hispanic white mothers in the most deprived quartile of the named counties by deprivation rank | births in 2016 to 2022 | 95% CI 7.5% to 7.7% | 145 counties | births to mothers of the group in each county | County Health Rankings |   |
| lbw_withheld_cells | 11,250 | withheld cells | count of cells with 1 to 9 births, which WONDER withholds | county-year cells of the WONDER birthweight cut in the named counties | births in 2018 to 2024 |   |   | none (full count) | CDC WONDER natality |   |
| lbw_withheld_max_share | 0.5% | percent of live births | largest share of births the withheld birthweight cells could hold: 9 births in each withheld cell, over all births in the counties | live births in the named counties | births in 2018 to 2024 |   |   | none (an upper limit from the cell count) | CDC WONDER natality |   |
| map_floor_births | 20 | low-birthweight births | fewest low-birthweight births for which a county is drawn (a floor set for this book) | counties on the map | births in 2016 to 2022 |   |   | none (definition) | County Health Rankings |   |
| preterm_adi_corr | 0.69 | correlation coefficient | births-weighted correlation of the county preterm share with county deprivation rank, county-cluster bootstrap interval | live births in the named counties | births in 2018 to 2024 | 95% CI 0.64 to 0.73 | 578 counties | births in each county | CDC WONDER natality |   |
| preterm_cutoff_weeks | 37 | completed weeks of gestation | gestational age below which a live birth counts as preterm, obstetric estimate (a definition) | live births | all years (a definition) |   |   | none (definition) | CDC WONDER natality |   |
| preterm_hi | 12.1% | percent of live births before 37 weeks | births-weighted mean of county shares (the pooled share), bootstrap interval over counties | live births in the most deprived tenth of the named counties by county deprivation rank | births in 2018 to 2024 | 95% CI 11.3% to 12.7% | 58 counties | births in each county | CDC WONDER natality |   |
| preterm_lo | 8.8% | percent of live births before 37 weeks | births-weighted mean of county shares (the pooled share), bootstrap interval over counties | live births in the least deprived tenth of the named counties by county deprivation rank | births in 2018 to 2024 | 95% CI 8.6% to 9.0% | 57 counties | births in each county | CDC WONDER natality |   |
| preterm_natl | 10.1% | percent of live births before 37 weeks | pooled share of births in published cells (events over births), Wilson interval | live births in the counties that CDC WONDER natality names individually (counties of 100,000 residents or more), births in published cells | births in 2018 to 2024 | 95% CI 10.1% to 10.1% | 20,418,250 births | none (full counts of births) | CDC WONDER natality |   |
| preterm_range_pp | 3.2 percentage points | percentage points | difference of two births-weighted decile means of the preterm share, bootstrap interval over counties | live births in the named counties, most deprived tenth minus least deprived tenth | births in 2018 to 2024 | 95% CI 2.5 percentage points to 3.9 percentage points | 578 counties | births in each county | CDC WONDER natality |   |
| preterm_slope_per_10 | 0.49 percentage points | percentage points of the preterm share per ten points of county deprivation rank | births-weighted slope of the county preterm share on county deprivation rank, county-cluster bootstrap interval | live births in the named counties | births in 2018 to 2024 | 95% CI 0.44 percentage points to 0.55 percentage points | 578 counties | births in each county | CDC WONDER natality |   |
| preterm_withheld_cells | 8,704 | withheld cells | count of cells with 1 to 9 births, which WONDER withholds | county-year cells of the WONDER gestational-age cut in the named counties | births in 2018 to 2024 |   |   | none (full count) | CDC WONDER natality |   |
| preterm_withheld_max_share | 0.4% | percent of live births | largest share of births the withheld gestational-age cells could hold: 9 births in each withheld cell, over all births in the counties | live births in the named counties | births in 2018 to 2024 |   |   | none (an upper limit from the cell count) | CDC WONDER natality |   |
| south_all_share | 47% | percent of counties | share of mapped counties in the Census South region (16 states and DC) | all mapped counties | births in 2016 to 2022 |   | 2,900 counties | none (each county counts once) | County Health Rankings |   |
| south_top_share | 89% | percent of counties | share of those counties in the Census South region (16 states and DC) | mapped counties in the highest tenth of low-birthweight shares | births in 2016 to 2022 |   | 290 counties | none (each county counts once) | County Health Rankings |   |
| wonder_suppression_floor | 10 | births | a cell with fewer births than this is withheld (a definition) | cells of the WONDER natality files | births in 2018 to 2024 |   |   | none (definition) | CDC WONDER natality |   |
| wonder_threshold | 100,000 | residents | population size at which WONDER names a county (a definition) | counties that CDC WONDER natality names individually | births in 2018 to 2024 |   |   | none (definition) | CDC WONDER natality |   |

## Definitions

- Low birthweight: A live birth weighing less than 2,500 grams, the cutoff used in vital statistics.
- Preterm birth: A birth before 37 weeks of pregnancy.
- Area Deprivation Index (ADI): A score that ranks neighborhoods by income, education, employment and housing conditions drawn from the American Community Survey. In this book a higher national percentile means a more deprived neighborhood.
- Suppression: The withholding of a number because the count behind it is too small to publish without risking the identification of a person. A suppressed value is missing, not zero.
- Weathering: The hypothesis that chronic stress from disadvantage and discrimination wears down health faster, so that the same outcomes appear at younger ages in affected groups. Arline Geronimus proposed it in studies of birth outcomes.
- EITC: The Earned Income Tax Credit, a refundable federal tax credit for low and moderate income workers. It grows with earnings up to a limit and then phases out.

## Methods

The county deprivation comparison uses births in 2018 to 2024 in the 578 counties that CDC WONDER names individually, the counties of 100,000 residents or more, which hold 79.5% of births. Smaller counties are not part of the WONDER figures. The comparisons by mother's race use other sources and birth periods: County Health Rankings rates for births in 2016 to 2022, and WONDER linked birth and infant death records for 2018 and 2019. Coverage and denominators differ across these sources, so each is described separately below.

### The county comparison uses only counties that WONDER names

Low birthweight is a live birth under 2,500 grams and preterm birth one before 37 completed weeks (obstetric estimate).

Counts come from CDC WONDER natality for 2018 to 2024, for the 578 counties that WONDER names individually (counties of 100,000 residents or more, 79.5% of births); WONDER pools every other county in a state into one Unidentified Counties row, which is not used. WONDER's 2024 natality data are final.

Explains Table 2.

### Rates by mother's race and the map use other sources and birth periods

Race-specific low birthweight comes from County Health Rankings county rates for births in 2016 to 2022, for non-Hispanic Black, non-Hispanic white and Hispanic (any race) mothers: CHR publishes each rate with a 95% interval, from which the births behind the rate are recovered, and the rates are pooled by births over the counties where a rate is published.

Infant mortality by mother's race comes from WONDER linked birth and infant death for 2018 and 2019, the only years with mother's race, with mothers counted by single race of any ethnicity: national rates pool the state files, and the county view covers counties of 250,000 residents or more, with county-years that have withheld deaths left out of both deaths and births.

The map shows County Health Rankings rates for every county with at least 20 low-birthweight births.

Explains Table 4, Table 3 and Table 1.

### How counties are ranked and grouped

A county's deprivation rank is the population-weighted mean of the Area Deprivation Index national ranks of its census tracts, weighted by 2019-2023 tract population; Connecticut tracts are assigned to the eight legacy counties.

Deciles and quartiles hold equal numbers of the 578 large counties. Decile means are births-weighted means of county shares.

Explains Table 2 and Table 4.

### How uncertainty is measured

Pooled WONDER shares carry Wilson score intervals on the counts. Decile and quartile means, pooled rates by mother's race, differences, ratios, correlations and slopes carry 95% percentile intervals from a bootstrap that resamples whole counties (2,000 draws, seed 12345); gaps between groups use the same draws for both groups.

### Withheld cells are excluded and never reconstructed

WONDER withholds cells of 1 to 9 births. A withheld cell is missing: it leaves both the count below the cutoff and the denominator, which is the births in the published cells, and nothing is derived from the county totals.

The birthweight cut has 11,250 withheld cells in these counties and the gestational-age cut 8,704; at 9 births each they could hold at most 0.5% and 0.4% of births.

### Four rankings of the same places

The comparison follows the book's ranking switch rule, fixed on October 28, 2026 before any of these gradients was computed. The main gradient of the chapter is recomputed under four rankings of the same places, with the outcome, period, weights, bins and intervals unchanged: the published ADI; a standardized index that applies the ADI's coefficients to its measures after standardizing each one across tracts; median household income; and the share of adults 25 and older with a bachelor's degree or more. The three alternatives come from ACS tract estimates for 2019 to 2023, the period of the ADI.

All four rankings use the same places, the 578 counties that have the outcome and all four rankings, so the ADI panel can differ slightly from the chapter's main figure. Tracts enter only if they have every input of the standardized index, which leaves out tracts without a median rent, home value or mortgage cost. A county takes the population-weighted mean of the ranks of its common-sample tracts, weighted by 2019-2023 tract population, with Connecticut tracts assigned to the eight legacy counties that WONDER uses.

A pattern counts as changed when the worse end switches sides, when the two ends differ under one ranking and not the other, or when the rank correlation of bin order and bin mean falls on the other side of 0.9 and moves by at least 0.1.

Explains Table 5.

## What these data cannot show

The county figures describe places, and a county rate is an average over the births in it. A county ranked as least deprived can hold deprived neighborhoods, and the Area Deprivation Index understates deprivation in high-cost cities (Chapter 1.1, What the deprivation index measures (https://vishalsingh.org/zipcode/anatomy-of-deprivation)). A pattern across counties does not describe the babies in any one of them.

The WONDER figures cover only counties of 100,000 or more residents, which hold 79.5% of births. Births in smaller counties are pooled by WONDER and cannot be placed on the deprivation scale, and the check against County Health Rankings shows that the pattern is less regular across all counties than across large ones.

The sources differ in period and definition. The deprivation index draws on the American Community Survey for 2019 to 2023, while the births span 2018 to 2024. Rates by mother's race come from County Health Rankings for births in 2016 to 2022, for non-Hispanic Black, non-Hispanic white and Hispanic mothers, and they are pooled over the counties where a rate is published. Infant mortality by mother's race comes from 2018 and 2019 only, for single races of any ethnicity, so its white group includes Hispanic mothers and is not the group in the low-birthweight rates. The infant mortality county view covers only larger counties and leaves out county-years in which WONDER withholds the deaths, which are 13.6% of Black births and 5.2% of white births there. The intervals leave out those differences in source, period and group.

Finally, none of these comparisons identifies a mechanism. Deprivation and race are tied to smoking, prenatal care, chronic stress, pollution, maternal age and education, and to the quality of the nearest hospital. Birthweight is associated with all of them. The two policy studies above are the only evidence in this chapter on what changes the numbers, and they speak to income and coverage, not to the rest. A county map cannot settle the mechanism behind any one county's rate or behind the gap between groups.

## References

1. Almond, Douglas; Currie, Janet (2011). Killing Me Softly: The Fetal Origins Hypothesis. Journal of Economic Perspectives 25(3): 153-172. https://doi.org/10.1257/jep.25.3.153
2. Aizer, Anna; Currie, Janet (2014). The Intergenerational Transmission of Inequality: Maternal Disadvantage and Health at Birth. Science 344(6186): 856-861. https://doi.org/10.1126/science.1251872
3. Kind, Amy J. H.; Buckingham, William R. (2018). Making Neighborhood-Disadvantage Metrics Accessible: The Neighborhood Atlas. New England Journal of Medicine 378(26): 2456-2458. https://doi.org/10.1056/NEJMp1802313
4. David, Richard J.; Collins, James W., Jr. (1997). Differing Birth Weight among Infants of U.S.-Born Blacks, African-Born Blacks, and U.S.-Born Whites. New England Journal of Medicine 337(17): 1209-1214. https://doi.org/10.1056/NEJM199710233371706
5. Geronimus, Arline T. (1992). The Weathering Hypothesis and the Health of African-American Women and Infants: Evidence and Speculations. Ethnicity & Disease 2(3): 207-221.
6. Hoynes, Hilary; Miller, Doug; Simon, David (2015). Income, the Earned Income Tax Credit, and Infant Health. American Economic Journal: Economic Policy 7(1): 172-211. https://doi.org/10.1257/pol.20120179
7. Currie, Janet; Gruber, Jonathan (1996). Saving Babies: The Efficacy and Cost of Recent Changes in the Medicaid Eligibility of Pregnant Women. Journal of Political Economy 104(6): 1263-1296. https://doi.org/10.1086/262059

## Provenance

- Pipeline version: 0.1.0
- Build script: book2/chapters/birthweight-and-preterm-birth/build.py (sha256 15908823ab9d370445413480d8ba74adfabfeb7a1e1ba498f748384d128e0e4e)
- Built on: 2026-10-07
- Site commit: c2b1ff3309cb4a10833f884da91ac0057a66ad0a

Inputs:

| File | Size (bytes) | SHA-256, first 1 MB | SHA-256, last 1 MB |
| --- | ---: | --- | --- |
| data/analytic/tract_atlas.parquet | 39,659,925 | 29d5e4d943552d54aeb6ba92aadc939d9bd15bdea553c78357dbd516757be627 | 1e10d1ecf722340f23ee2212d4999a90250f85b65fa89b39e32db06d9472e22e |
| data/raw/acs/deprivation_components_2019_2023/tract.parquet | 13,197,793 | 5545cde2cbe96a8b56e468d65388248a241aa6bd300d3c9e325d08d3f86b11da | 5c8ec454d066988b0f1a01aaa6ed34ebfc05d162e01cf9c6bb01f9198cbaff12 |
| data/staged/adi/adi_atlas_tract2020_2023.parquet | 798,078 | 5e409bc212ff871def780fa27b4cf8b616d9da420b2b9f6455517eddcd8c1a95 | 5e409bc212ff871def780fa27b4cf8b616d9da420b2b9f6455517eddcd8c1a95 |
| data/staged/births/wonder_natality_county_birthweight_2018_2024.parquet | 115,259 | 17a46844906a1ab03d02c98fd4e2aca0535a3ad4da186e0feaafb17a03dfa9ef | 17a46844906a1ab03d02c98fd4e2aca0535a3ad4da186e0feaafb17a03dfa9ef |
| data/staged/births/wonder_natality_county_gestational_age_2018_2024.parquet | 104,715 | 4e24d1ed2292840cbf08b8939b316ae7ecb029187bfbfc9390da95721a484dd4 | 4e24d1ed2292840cbf08b8939b316ae7ecb029187bfbfc9390da95721a484dd4 |
| data/staged/births/wonder_natality_county_total_2018_2024.parquet | 41,774 | e85d351fa72c022a05415d332185113b6badcaff1119da1d6f45c76ffe16d86d | e85d351fa72c022a05415d332185113b6badcaff1119da1d6f45c76ffe16d86d |
| data/staged/crosswalks/ct_tract2020_to_old_county.parquet | 15,842 | 961ef39f31596ef65fca2c67d5f2f63e533c5dc9b49745d3bb6437844e4b831d | 961ef39f31596ef65fca2c67d5f2f63e533c5dc9b49745d3bb6437844e4b831d |
| data/staged/health/chr_county_2024.parquet | 8,776,933 | 619795f7ecfec8aff10694b649eaee5e421dbeffbe4be47af4161b51ceefa5cd | c69357f1bcf630407bb9716cb445618838999558c019aaaf77d4681b6bf45c56 |
| data/staged/mortality/wonder_lbd_county_race_2018_2022.parquet | 50,622 | cbe4c336184a835d2d61a2b626c3f4c48455a254ea8ff2a7a6d6eb6fbca700ed | cbe4c336184a835d2d61a2b626c3f4c48455a254ea8ff2a7a6d6eb6fbca700ed |
| data/staged/mortality/wonder_lbd_county_total_2018_2022.parquet | 29,903 | 8e0f2604f2be86d6d86c2d206fa4486b2479411c88a6bb9b1595f3d945a4e5b8 | 8e0f2604f2be86d6d86c2d206fa4486b2479411c88a6bb9b1595f3d945a4e5b8 |
| data/staged/mortality/wonder_lbd_state_race_2018_2022.parquet | 16,461 | 9d17fb750f0d7d3e99d48945e1131654951deddcabfa1d520f0cbf3d717ab710 | 9d17fb750f0d7d3e99d48945e1131654951deddcabfa1d520f0cbf3d717ab710 |
| data/staged/ses/acs5_tract2020_2019_2023.parquet | 7,232,424 | 5fe6f730a35225f14d5d8abfc6b9bce58b52e545f309a7eac74f1dae74d7a31c | 7b4a2475b3852d3f91804716a04566ef8ceadacec2be9fb85f5eb642078efac6 |
| data/staged/svi/svi_tract2020_2022.parquet | 17,740,415 | cb3860d8edc2aaf1f2a02211a58f7a0b80457eb6ad9a37b5ec8abb2d5da9da10 | e371a4a696d5bed05d54ad7c5b6ca2504508805b847ec415c2322be7f66ec2b2 |

Outputs:

| File | SHA-256 |
| --- | --- |
| facts.json | 9332a0015b74d0cb3babb0d72c1b75129ee521b3719bb6e2d1abf87321e9308a |
| figures/alt_rankings.json | 2b4facd8450b8f46d33c77517e55ab89b6e87e3383c8cbbc365b108150ce1206 |
| figures/birth_outcomes_by_race.json | 31c49d004799b6689056601578601d1a566275ad0addd032f1f6ceb22e229076 |
| figures/lbw_by_deprivation.json | d98d50597d29c7dc3a12ce69ff8d5abf68a205051a62824ffb30791b0c827796 |
| figures/lbw_map.json | df63b4e72329eb8b87443c32a0fd49aeaea240377fa620c7fb8f0e104b73493b |
| figures/weathering.json | 02bc86d58e45065993f13cbe2fd4e4aec07d9cc051d2ee4e82ddf44712b50d06 |
| meta.json | a41f90bf2fceb47a0878676b734993cb0fdbd657183877356b12098162593b23 |
| public/data/birthweight-and-preterm-birth/lbw_map.json | bd8bbce4271837be5dd0fa8d06df21722f86b1ad86e0ee4e17eafd7426199912 |
| refs.json | 257338ba1458f99dd66dd41995198798df5660ed8da0bd35a2966fb23ad0e8b3 |

## Machine formats

- Report (HTML): https://vishalsingh.org/zipcode/birthweight-and-preterm-birth/report
- Report (Markdown): https://vishalsingh.org/zipcode/birthweight-and-preterm-birth/report.md
- Every estimate (JSON): https://vishalsingh.org/zipcode/report/birthweight-and-preterm-birth/facts.json
- Table 1, figure "lbw_map" (JSON): https://vishalsingh.org/zipcode/report/birthweight-and-preterm-birth/figures/lbw_map.json
- Table 2, figure "lbw_by_deprivation" (JSON): https://vishalsingh.org/zipcode/report/birthweight-and-preterm-birth/figures/lbw_by_deprivation.json
- Table 3, figure "birth_outcomes_by_race" (JSON): https://vishalsingh.org/zipcode/report/birthweight-and-preterm-birth/figures/birth_outcomes_by_race.json
- Table 4, figure "weathering" (JSON): https://vishalsingh.org/zipcode/report/birthweight-and-preterm-birth/figures/weathering.json
- Table 5, figure "alt_rankings" (JSON): https://vishalsingh.org/zipcode/report/birthweight-and-preterm-birth/figures/alt_rankings.json
- Payload of "lbw_map" (JSON): https://vishalsingh.org/zipcode/data/birthweight-and-preterm-birth/lbw_map.json
- Index of every report: https://vishalsingh.org/zipcode/llms.txt; all reports in one file: https://vishalsingh.org/zipcode/llms-full.txt

---

# Infant mortality: report

Chapter 2.2 of ZIP Code Destiny, by Vishal Singh (NYU Stern). Generated from the chapter's data files; the reading version is at https://vishalsingh.org/zipcode/infant-mortality. Table n of this report is figure n of the chapter.

## Identification

- Title: Infant mortality
- Label: The first year
- Chapter: 2.2 (Part II, Birth; An American life)
- Question: How does the infant mortality rate vary across states and levels of county deprivation, and how does it differ by race and by age at death?
- Dek: Infant mortality in 2018 to 2022 ran from 3.7 deaths per 1,000 live births in Massachusetts and New Hampshire to 8.7 in Mississippi.
- Grain: counties
- Status: published
- Published: 2026-10-09
- Updated: not revised
- Data sources:
  - CDC WONDER linked birth and infant death. CDC WONDER linked birth and infant death, state totals and race, county totals and age at death, 2018 to 2022 (mother's race 2018 and 2019 only); years: 2018-2022; weights: none (full counts; crude rates)
  - Area Deprivation Index. Neighborhood Atlas Area Deprivation Index 2023 (v4.0.1), tract national rank, averaged to counties with 2019-2023 tract population; years: 2023 release (version 4.0.1; built on ACS 2019-2023); weights: 2019-2023 tract population
  - County Health Rankings. County Health Rankings 2024, infant mortality (infant deaths per 1,000 live births, 2015 to 2021), counties with at least 20 infant deaths; years: 2024 release; weights: births (County Health Rankings denominator)
- Canonical URL: https://vishalsingh.org/zipcode/infant-mortality
- Report: https://vishalsingh.org/zipcode/infant-mortality/report
- Markdown: https://vishalsingh.org/zipcode/infant-mortality/report.md
- BibTeX key: singh2026infantmortality

```bibtex
@misc{singh2026infantmortality,
  author = {Singh, Vishal},
  title = {Infant mortality},
  year = {2026},
  url = {https://vishalsingh.org/zipcode/infant-mortality},
  note = {In ZIP Code Destiny. Published 9 October 2026.}
}
```

## Principal estimates

The facts the dek and the opening prose cite, in order of appearance.

| Estimand | Population | Period | Value | Interval | n | Weights | Source | Fact id |
| --- | --- | --- | ---: | --- | ---: | --- | --- | --- |
| crude rate from the death count, normal interval | live births in Massachusetts | 2018 to 2022 | 3.7 | 95% CI 3.5 to 3.9 | 1,260 infant deaths | none (crude rate from counts) | CDC WONDER linked birth and infant death | imr_state_low |
| the denominator of an infant mortality rate (a definition) | all live births | all years (a definition) | 1,000 |   |   | none (definition) | CDC WONDER linked birth and infant death | per_1000 |
| crude rate from the death count, normal interval | live births in Mississippi | 2018 to 2022 | 8.7 | 95% CI 8.3 to 9.2 | 1,564 infant deaths | none (crude rate from counts) | CDC WONDER linked birth and infant death | imr_state_high |
| crude rate from the death count, normal interval | live births in the 50 states and DC | 2018 to 2022 | 5.5 | 95% CI 5.5 to 5.6 | 102,503 infant deaths | none (crude rate from counts) | CDC WONDER linked birth and infant death | imr_natl |
| count of counties | named counties with a published death count in every year 2018 to 2022 | infant deaths and live births in 2018 to 2022 | 240 |   |   | none (full count) | CDC WONDER linked birth and infant death | counties_used |
| births-weighted mean of county rates (the pooled rate), bootstrap interval over counties | live births in the least deprived quarter of the counties used, by county deprivation rank | 2018 to 2022 | 3.9 | 95% CI 3.7 to 4.1 | 60 counties | births in each county | CDC WONDER linked birth and infant death | imr_q1 |
| births-weighted mean of county rates (the pooled rate), bootstrap interval over counties | live births in the most deprived quarter of the counties used, by county deprivation rank | 2018 to 2022 | 7.2 | 95% CI 6.7 to 7.6 | 60 counties | births in each county | CDC WONDER linked birth and infant death | imr_q4 |
| crude rate from state counts, normal interval | live births to Black mothers (single race, any ethnicity) in the United States, states with a published count | 2018 and 2019 | 10.3 | 95% CI 10.1 to 10.5 | 12,953 infant deaths | none (crude rate from counts) | CDC WONDER linked birth and infant death | imr_black |
| crude rate from state counts, normal interval | live births to white mothers (single race, any ethnicity) in the United States, states with a published count | 2018 and 2019 | 4.7 | 95% CI 4.7 to 4.8 | 26,516 infant deaths | none (crude rate from counts) | CDC WONDER linked birth and infant death | imr_white |

## Findings

Infant mortality in 2018 to 2022 was 8.7 deaths per 1,000 live births in Mississippi and 3.7 in Massachusetts and New Hampshire, against 5.5 for the United States. Among the 240 large counties that publish a death count in every year, the rate rises from 3.9 in the least deprived quarter of counties to 7.2 in the most deprived. Requiring a count in every year leaves out the counties whose deaths are withheld in any year. In 2018 and 2019, the only years for which the linked birth and infant death file reports the mother's race, the rate was 10.3 for infants of Black mothers and 4.7 for infants of white mothers.

These are rates for states, counties and groups of births. They describe where and among whom infants died and say nothing about why a particular infant did. The conditions that come before many infant deaths, low birthweight and preterm birth, are in Chapter 2.1, Birthweight and preterm birth (https://vishalsingh.org/zipcode/birthweight-and-preterm-birth), which also shows infant mortality for Black and white mothers by county deprivation. The fall in the number of births is in Chapter 2.4, Where births are falling (https://vishalsingh.org/zipcode/where-births-are-falling).

### What the rate counts and where the numbers come from

An infant death is the death of a baby before its first birthday. The infant mortality rate divides such deaths by live births and is reported here as deaths per 1,000. It is a crude rate, with no adjustment for the age or health of mothers. Researchers also split infant deaths by age. Neonatal deaths occur before 28 days of age, and postneonatal deaths occur from that age to the first birthday. A comparison of the United States with European countries found that the American disadvantage is concentrated in the postneonatal period and among infants of lower socioeconomic status.[1]

The numbers come from the linked birth and infant death records on CDC WONDER, which match each infant death certificate to the infant's birth certificate.[2] State files cover all 51 jurisdictions, and WONDER publishes every state total in full. They record 102,503 infant deaths among 18.5 million live births in 2018 to 2022. County files name only counties of 250,000 residents or more, pool every other county in a state into one row, and withhold any county cell with fewer than 10 deaths. A withheld cell is missing, not zero, and nothing in this chapter is derived from a total minus the cells that are published. The state file for 2021 gives a national rate of 5.44, which is the rate NCHS reports for that year.[2]

 (Note: WONDER weights infant deaths in the linked file, so a count can differ from the number of death records, and the age groups from a total, by rounding.)

### State rates vary widely and most of the highest are in the South

Mississippi has the highest state rate, 8.7 (95% CI 8.3 to 9.2). Massachusetts, at 3.7 (95% CI 3.5 to 3.9), and New Hampshire, at 3.7 (95% CI 3.2 to 4.2), have the lowest rates, and their intervals overlap. The ratio of the rate in Mississippi to the rate in Massachusetts is 2.4 (95% CI 2.2 to 2.6) (Table 1). Arkansas, Louisiana and Alabama follow Mississippi.

The Census South region holds 7 of the 10 states with the highest rates and 17 of the 51 jurisdictions. The remaining states in that group are Ohio, South Dakota and Indiana.

Pooled over births, the rate in the South is 6.3 (95% CI 5.9 to 6.9) against 5.1 (95% CI 4.6 to 5.7) in the rest of the country, a ratio of 1.2 (95% CI 1.1 to 1.4). Within the rest, the Midwest has 6.1 (95% CI 5.5 to 6.5), close to the South, while the Northeast has 4.5 (95% CI 3.9 to 5.2) and the West 4.6 (95% CI 4.3 to 5.3). Intervals resample whole states.

Of the 51 jurisdictions, 21 have an interval entirely above the national rate and 16 have one entirely below it. The others cannot be told apart from the national rate on these counts. The pooled period includes the pandemic years, but no single year sets the pooled rates. The national rate was lowest in 2020, at 5.4, and highest in 2018, at 5.7.

Table 1. Infant mortality ran from 3.7 per 1,000 in Massachusetts and New Hampshire to 8.7 in Mississippi

### Counties with high rates are concentrated in the South

WONDER names only 261 counties in its linked file, which is too few to map infant mortality. County Health Rankings, from the same NCHS vital statistics, publishes a rate for every county with at least 20 infant deaths in the period. It gives 1,206 counties a rate for infant deaths and live births in 2015 to 2021 (Table 2). That period is not WONDER's, so the two sources are compared before the map is used. Across the 240 large counties that both cover, the two sets of county rates correlate at 0.980 (95% CI 0.974 to 0.985).

Of the mapped counties in the highest tenth of rates, 85.1% are in the South, which holds 49.3% of the mapped counties. Mississippi holds 25 of those counties, more than any other state. The 1,937 gray counties had fewer than 20 infant deaths in the period, which in practice means small counties. Gray is not zero, and the tooltip gives the reason. The key runs from 3.7 to 10.6, the fifth to ninety-fifth percentile of the mapped county rates, and the end colors include the counties beyond those limits. The highest county rates rest on few deaths, so a single county's rate is imprecise and the pattern across many counties is what the map shows.

Table 2. 85.1% of the counties in the highest tenth of infant mortality rates are in the South, which holds 49.3% of the 1,206 counties that publish a rate

### Infant mortality rises with county deprivation

To place counties on one scale, each census tract carries its Area Deprivation Index national rank, which orders neighborhoods by deprivation with a higher rank meaning more deprived.[3] A county's rank is the average of its tracts' ranks weighted by population, the same measure used in Chapter 2.1, Birthweight and preterm birth (https://vishalsingh.org/zipcode/birthweight-and-preterm-birth). That chapter cuts its quartiles within the larger set of counties that WONDER natality names, and the quartiles here are cut within the counties used below. The quartile edges differ, so quartile rates in the two chapters are not directly comparable. The gradient uses the 240 of the 261 named counties that publish a death count in every year from 2018 to 2022. They hold 62.3% of births in the United States. The quarters here are quarters of these large counties, so the most deprived quarter is relative to large counties and not to the nation.

Their ranks run from 3 to 84, while 712 counties in the country rank above the highest of them. Those counties are home to 3.3% of residents, and every one has fewer than 250,000 residents, so WONDER does not name them. The most deprived large county is less deprived than the most deprived places in the nation, which is part of why the County Health Rankings check below matters.

A county-year with fewer than 10 deaths is withheld, which makes a county's published years its high years. Pooling a county over only those years would overstate its rate, so a county with any withheld year is left out whole. That removes 21 counties, which hold 2.6% of births in the named counties and account for 44 withheld county-years. Each of those holds fewer than 10 deaths, so together they could hold at most 0.6% of infant deaths in the named counties. The wider problem of what suppression leaves out of death data is the subject of Chapter 10.3, Suppressed counties in death data (https://vishalsingh.org/zipcode/suppressed-counties-in-death-data).

The rate rises across the quarters of counties, from 3.9 (95% CI 3.7 to 4.1) in the least deprived to 7.2 (95% CI 6.7 to 7.6) in the most deprived (Table 3). The difference is 3.3 (95% CI 2.8 to 3.8) deaths per 1,000, and the rate in the most deprived quarter is 1.9 (95% CI 1.7 to 2.0) times the rate in the least deprived. Across all of these counties the rate moves with the rank, with a births-weighted correlation of 0.80 (95% CI 0.73 to 0.86). The slope is 0.63 (95% CI 0.54 to 0.71) deaths per 1,000 for each 10-point step in the rank.

A second source reaches smaller counties. County Health Rankings rates for 2015 to 2021 cover 1,206 counties of every size that reach the floor of 20 deaths. In tenths of those counties the rate is 4.0 (95% CI 3.8 to 4.2) in the least deprived tenth and 8.8 (95% CI 8.1 to 9.5) in the most deprived, with a correlation of 0.76 (95% CI 0.71 to 0.80). The line is not monotone: it rises over the lower deciles and dips across the eighth and ninth before the most deprived tenth. The interval is wider in the most deprived tenth, whose counties have fewer births.

Table 3. Infant mortality rises from 3.9 per 1,000 in the least deprived quarter of large counties to 7.2 in the most deprived

#### The rise holds within states and within groups of mothers

Counties in the South have higher deprivation ranks and higher rates, so the rise could come from differences between states alone. Comparing counties inside the same state, with a fixed effect for each state, the slope is 0.62 (95% CI 0.47 to 0.84) deaths per 1,000 for each 10-point step, nearly the same as across all counties (Table 4). It is also nearly the same when only births in 2018 and 2019 are used, at 0.62 (95% CI 0.52 to 0.71), the only years for which WONDER reports the mother's race.

For births to white mothers, in the 203 counties that publish them in both years and hold 93.9% of those births, the slope is 0.36 (95% CI 0.29 to 0.44). For births to Black mothers, in the 115 counties that do and hold 82.5% of those births, it is 0.82 (95% CI 0.62 to 1.01). The white slope is smaller than the slope for all births, and the Black slope is larger than the white slope.

Counties with higher ranks also have a larger share of births to Black mothers, 11.1% of births in the least deprived quarter and 31.1% in the most deprived in 2018 and 2019. The slope for all births therefore combines a rise within each group with a shift in the mix of mothers. A slope across counties describes places. It does not say that a family moved to a less deprived county would see a different outcome.

Table 4. The rise with county deprivation holds within states and for births to white mothers and to Black mothers, at 0.63 deaths per 1,000 for each 10-point step across all births

### The Black-white gap appears in every state that publishes both groups

The linked file reports the mother's race only for 2018 and 2019, and WONDER counts mothers by single race, with Hispanic mothers included in each group. In those years the rate was 10.3 (95% CI 10.1 to 10.5) deaths per 1,000 births to Black mothers, 4.7 (95% CI 4.7 to 4.8) for white mothers and 3.9 (95% CI 3.7 to 4.1) for Asian or Pacific Islander mothers, against 5.6 for all mothers (Table 5). The Black rate is 2.2 (95% CI 2.1 to 2.2) times the white rate, a difference of 5.6 (95% CI 5.4 to 5.8) deaths per 1,000. The Black and white rates are the two printed in Chapter 2.1, Birthweight and preterm birth (https://vishalsingh.org/zipcode/birthweight-and-preterm-birth), computed from the same state files.

WONDER withholds a state-year cell with fewer than 10 deaths. The withheld cells hold 1.0% of births to Black mothers, 0.1% of births to white mothers and 6.7% of births to Asian or Pacific Islander mothers, so those rates describe almost all of the births in each group. For American Indian or Alaska Native mothers the situation differs. WONDER withholds 66 of the 102 state-year cells, and the published cells hold only 65.6% of their births. A pooled rate over those cells would describe the states with the most deaths, so none is shown.

Table 5. Infant mortality in 2018 and 2019 was 10.3 per 1,000 for Black mothers and 4.7 for white mothers, a ratio of 2.2

Infants of Black mothers died at a higher rate than infants of white mothers in each of the 36 states that publish both groups in both years, which hold 98.0% of births to Black mothers (Table 6). In none of them does the interval for one group touch the interval for the other. The ratio has a median of 2.1 across these states. It is smallest in Mississippi, at 1.7 (95% CI 1.4 to 1.9), and largest in Delaware, at 3.2 (95% CI 2.3 to 4.7).

Mississippi has the smallest ratio and also the highest white rate of the states shown, 6.7 (95% CI 5.8 to 7.5). The lowest white rate is 3.2 (95% CI 2.9 to 3.5), in New Jersey. The white rate itself therefore differs across states, and the state ranking in Table 1 is not only a difference in racial composition.

Table 6. Black mothers' infants died at a higher rate than white mothers' in all 36 states that publish both, by a median ratio of 2.1

A rate for a racial group in a state averages over mothers whose circumstances differ greatly, so these comparisons do not identify which circumstances matter. Chapter 2.1, Birthweight and preterm birth (https://vishalsingh.org/zipcode/birthweight-and-preterm-birth) reports studies of income support and insurance during pregnancy and the birth outcomes that precede these deaths.

### Age at death is published for only a few counties

WONDER's county file splits infant deaths into five age groups, from under an hour to the end of the first year. It withholds each county-year-age cell with fewer than 10 deaths, which is 4,328 of the 6,525 cells for the named counties. Only 14 counties publish every cell in every year (Table 7). They are among the largest counties and hold 23.6% of births in the named counties. The age groups cannot be summed for any other county without filling in withheld cells, so this section covers these counties and no others.

In these counties the neonatal rate is 3.6 (95% CI 3.1 to 4.1) and the postneonatal rate is 1.6 (95% CI 1.4 to 1.9). Neonatal deaths are 68.7% (95% CI 67.4% to 70.2%) of infant deaths, and 42.3% (95% CI 40.5% to 44.2%) of infant deaths occur on the first day of life.

Both rates are higher where the deprivation rank is higher. Dividing the 14 counties at the median rank, the neonatal rate in the more deprived group is 1.5 (95% CI 1.2 to 1.8) times the rate in the less deprived group, and the postneonatal rate is 1.6 (95% CI 1.3 to 2.1) times. The intervals overlap, so these counties cannot show whether the postneonatal rate rises faster with deprivation. The comparison of the United States with Europe placed the American disadvantage in the postneonatal period, among infants of lower socioeconomic status.[1] These counties have no European comparison and too few observations to test that.

Table 7. Neonatal deaths made up 68.7% of infant deaths in the 14 counties that publish every age group, and both rates are higher where deprivation is higher

### Infant mortality rises with deprivation under all four rankings

Table 8 ranks the 1,206 counties that have a County Health Rankings rate and all four rankings in four ways. Ranked by the ADI on this common sample, the rate is 4.0 (95% CI 3.8 to 4.2) deaths per 1,000 in the least deprived tenth and 8.7 (95% CI 8.0 to 9.4) in the most deprived, and the rank correlation between tenth and rate is +0.95. Under the standardized index the rate is 4.0 (95% CI 3.8 to 4.2) in the least deprived tenth and 7.0 (95% CI 6.1 to 8.3) in the most deprived, with a correlation of +0.92, and it is highest in the eighth tenth. Under median household income the two ends are 3.9 (95% CI 3.7 to 4.1) and 9.0 (95% CI 8.5 to 9.4), with a correlation of +1.00, and under the bachelor's degree share they are 4.3 (95% CI 4.1 to 4.6) and 7.8 (95% CI 7.2 to 8.5), with a correlation of +0.99, so none of the three counts as a change under the book's rule (Chapter 9.1, How far one ranking goes (https://vishalsingh.org/zipcode/how-far-one-ranking-goes)).

Table 8. Infant mortality is steadily worse toward the more deprived end under the ADI and under each of the three other rankings

### What these data cannot show

The county gradient describes the 240 largest counties, which hold 62.3% of births. WONDER pools the rest by state. The County Health Rankings check reaches smaller counties but leaves out those that do not reach 20 deaths in the period, and the counties just above that floor have the least precise rates on the map. The county rates describe counties that publish a figure, which are not a sample of all counties, and the age-at-death figure covers only 14 of them.

Sources and periods differ, and each is named where it appears. WONDER covers 2018 to 2022 for states and counties and 2018 and 2019 for mother's race, and County Health Rankings covers 2015 to 2021. WONDER's race groups are single races that include Hispanic mothers, so its white group is not the non-Hispanic white group of other sources, and Hispanic origin is not in the files used here. American Indian or Alaska Native mothers are not shown.

The Area Deprivation Index understates deprivation in high-cost cities (Chapter 1.1, What the deprivation index measures (https://vishalsingh.org/zipcode/anatomy-of-deprivation)) and draws on the American Community Survey for 2019 to 2023, while the deaths are from 2018 to 2022. A county's rank averages over tracts that differ within the county, and it describes no family's circumstances. Patterns across counties and states are patterns of places, not of individual infants, a limit known as ecological inference.

Finally, the files carry no cause of death, no gestational age or birthweight, and no information on mothers beyond race, so nothing here identifies a mechanism. The rates are crude period rates, and 2020 to 2022 include the pandemic. Intervals for state and race rates treat deaths as independent and leave out differences between places within each rate, while the county figures resample whole counties.

## Tables

### Table 1. Infant mortality ran from 3.7 per 1,000 in Massachusetts and New Hampshire to 8.7 in Mississippi

Infant deaths per 1,000, by state and DC, 2018 to 2022; the first tab maps the rates and the second ranks them with intervals

#### Table 1.1. Map of states

7 of the 10 states with the highest infant mortality rates are in the South, which has 6.3 deaths per 1,000 against 5.1 elsewhere

Infant deaths per 1,000, by state and DC, 2018 to 2022

| conf_high | conf_low | n | rate | state |
| ---: | ---: | ---: | ---: | --- |
| 6.74 | 5.35 | 48,103 | 6.05 | AK |
| 7.49 | 6.87 | 290,226 | 7.18 | AL |
| 8.03 | 7.22 | 180,269 | 7.63 | AR |
| 5.83 | 5.36 | 393,508 | 5.60 | AZ |
| 4.20 | 4.03 | 2,161,370 | 4.11 | CA |
| 5.03 | 4.54 | 312,580 | 4.79 | CO |
| 4.68 | 4.05 | 173,445 | 4.36 | CT |
| 6.69 | 5.25 | 43,900 | 5.97 | DC |
| 6.62 | 5.30 | 52,873 | 5.96 | DE |
| 6.11 | 5.82 | 1,091,908 | 5.97 | FL |
| 6.94 | 6.54 | 625,219 | 6.74 | GA |
| 5.97 | 4.95 | 80,709 | 5.46 | HI |
| 5.04 | 4.41 | 184,889 | 4.73 | IA |
| 5.38 | 4.55 | 109,817 | 4.96 | ID |
| 5.98 | 5.62 | 678,780 | 5.80 | IL |
| 7.04 | 6.53 | 400,716 | 6.79 | IN |
| 6.26 | 5.54 | 175,138 | 5.90 | KS |
| 6.15 | 5.56 | 263,188 | 5.86 | KY |
| 7.88 | 7.25 | 289,800 | 7.57 | LA |
| 3.88 | 3.48 | 342,375 | 3.68 | MA |
| 6.18 | 5.67 | 346,879 | 5.92 | MD |
| 6.33 | 5.12 | 59,728 | 5.73 | ME |
| 6.62 | 6.19 | 529,293 | 6.40 | MI |
| 4.85 | 4.38 | 325,254 | 4.61 | MN |
| 6.46 | 5.94 | 353,119 | 6.20 | MO |
| 9.17 | 8.31 | 178,940 | 8.74 | MS |
| 5.40 | 4.25 | 55,789 | 4.82 | MT |
| 6.97 | 6.55 | 596,437 | 6.76 | NC |
| 5.80 | 4.55 | 50,828 | 5.17 | ND |
| 5.95 | 5.12 | 123,488 | 5.54 | NE |
| 4.20 | 3.23 | 60,327 | 3.71 | NH |
| 4 | 3.66 | 503,152 | 3.83 | NJ |
| 5.90 | 5.03 | 110,907 | 5.46 | NM |
| 5.70 | 5.01 | 171,286 | 5.35 | NV |
| 4.36 | 4.11 | 1,075,631 | 4.23 | NY |
| 7.14 | 6.74 | 656,808 | 6.94 | OH |
| 7.13 | 6.48 | 243,308 | 6.81 | OK |
| 4.60 | 4.03 | 204,273 | 4.31 | OR |
| 5.89 | 5.52 | 663,470 | 5.70 | PA |
| 5.19 | 4.01 | 51,515 | 4.60 | RI |
| 7.24 | 6.63 | 284,416 | 6.93 | SC |
| 7.46 | 6.11 | 56,872 | 6.79 | SD |
| 6.85 | 6.35 | 403,872 | 6.60 | TN |
| 5.56 | 5.35 | 1,887,748 | 5.45 | TX |
| 5.45 | 4.87 | 232,217 | 5.16 | UT |
| 6.08 | 5.65 | 483,476 | 5.87 | VA |
| 4.94 | 3.39 | 26,626 | 4.17 | VT |
| 4.64 | 4.24 | 421,310 | 4.44 | WA |
| 6.09 | 5.55 | 309,792 | 5.82 | WI |
| 7.45 | 6.35 | 87,834 | 6.90 | WV |
| 6.61 | 4.93 | 31,541 | 5.77 | WY |

#### Table 1.2. States ranked

Mississippi's rate of 8.7 is 2.4 times Massachusetts's 3.7

Infant deaths per 1,000 with 95% intervals, by state and DC, 2018 to 2022, ranked from the highest rate

| conf_high | conf_low | n | rate | state |
| ---: | ---: | ---: | ---: | --- |
| 9.17 | 8.31 | 178,940 | 8.74 | Mississippi |
| 8.03 | 7.22 | 180,269 | 7.63 | Arkansas |
| 7.88 | 7.25 | 289,800 | 7.57 | Louisiana |
| 7.49 | 6.87 | 290,226 | 7.18 | Alabama |
| 7.14 | 6.74 | 656,808 | 6.94 | Ohio |
| 7.24 | 6.63 | 284,416 | 6.93 | South Carolina |
| 7.45 | 6.35 | 87,834 | 6.90 | West Virginia |
| 7.13 | 6.48 | 243,308 | 6.81 | Oklahoma |
| 7.46 | 6.11 | 56,872 | 6.79 | South Dakota |
| 7.04 | 6.53 | 400,716 | 6.79 | Indiana |
| 6.97 | 6.55 | 596,437 | 6.76 | North Carolina |
| 6.94 | 6.54 | 625,219 | 6.74 | Georgia |
| 6.85 | 6.35 | 403,872 | 6.60 | Tennessee |
| 6.62 | 6.19 | 529,293 | 6.40 | Michigan |
| 6.46 | 5.94 | 353,119 | 6.20 | Missouri |
| 6.74 | 5.35 | 48,103 | 6.05 | Alaska |
| 6.69 | 5.25 | 43,900 | 5.97 | District of Columbia |
| 6.11 | 5.82 | 1,091,908 | 5.97 | Florida |
| 6.62 | 5.30 | 52,873 | 5.96 | Delaware |
| 6.18 | 5.67 | 346,879 | 5.92 | Maryland |
| 6.26 | 5.54 | 175,138 | 5.90 | Kansas |
| 6.08 | 5.65 | 483,476 | 5.87 | Virginia |
| 6.15 | 5.56 | 263,188 | 5.86 | Kentucky |
| 6.09 | 5.55 | 309,792 | 5.82 | Wisconsin |
| 5.98 | 5.62 | 678,780 | 5.80 | Illinois |
| 6.61 | 4.93 | 31,541 | 5.77 | Wyoming |
| 6.33 | 5.12 | 59,728 | 5.73 | Maine |
| 5.89 | 5.52 | 663,470 | 5.70 | Pennsylvania |
| 5.83 | 5.36 | 393,508 | 5.60 | Arizona |
| 5.95 | 5.12 | 123,488 | 5.54 | Nebraska |
| 5.97 | 4.95 | 80,709 | 5.46 | Hawaii |
| 5.90 | 5.03 | 110,907 | 5.46 | New Mexico |
| 5.56 | 5.35 | 1,887,748 | 5.45 | Texas |
| 5.70 | 5.01 | 171,286 | 5.35 | Nevada |
| 5.80 | 4.55 | 50,828 | 5.17 | North Dakota |
| 5.45 | 4.87 | 232,217 | 5.16 | Utah |
| 5.38 | 4.55 | 109,817 | 4.96 | Idaho |
| 5.40 | 4.25 | 55,789 | 4.82 | Montana |
| 5.03 | 4.54 | 312,580 | 4.79 | Colorado |
| 5.04 | 4.41 | 184,889 | 4.73 | Iowa |
| 4.85 | 4.38 | 325,254 | 4.61 | Minnesota |
| 5.19 | 4.01 | 51,515 | 4.60 | Rhode Island |
| 4.64 | 4.24 | 421,310 | 4.44 | Washington |
| 4.68 | 4.05 | 173,445 | 4.36 | Connecticut |
| 4.60 | 4.03 | 204,273 | 4.31 | Oregon |
| 4.36 | 4.11 | 1,075,631 | 4.23 | New York |
| 4.94 | 3.39 | 26,626 | 4.17 | Vermont |
| 4.20 | 4.03 | 2,161,370 | 4.11 | California |
| 4 | 3.66 | 503,152 | 3.83 | New Jersey |
| 4.20 | 3.23 | 60,327 | 3.71 | New Hampshire |
| 3.88 | 3.48 | 342,375 | 3.68 | Massachusetts |

Note: Rates are infant deaths in 2018 to 2022 over live births in the same years, pooled over the five years, for the 51 jurisdictions. WONDER publishes every state total in full, so no state has a withheld cell.

Note: Intervals are 95% intervals on the death counts (gamma for small counts and normal for large ones, the NCHS convention described in the methods note); they treat deaths as independent and leave out differences between counties within a state.

Note: WONDER weights infant deaths, so a total can differ from the unweighted number of records by rounding.

Source: CDC WONDER linked birth and infant death, 2018-2022; crude rates from full counts

Data: https://vishalsingh.org/zipcode/report/infant-mortality/figures/imr_by_state.json

### Table 2. 85.1% of the counties in the highest tenth of infant mortality rates are in the South, which holds 49.3% of the 1,206 counties that publish a rate

Infant deaths per 1,000, by county, County Health Rankings 2024, infant deaths and live births in 2015 to 2021

No inline rows: the values of Table 2 are in https://vishalsingh.org/zipcode/data/infant-mortality/imr_county_map.json.

Note: County Health Rankings publishes a county's infant mortality rate only from 20 infant deaths in the period, which leaves out most small counties and is the same floor the book uses for death maps. Gray counties are not zero; they are counties where the rate cannot be published.

Note: The period is 2015 to 2021, not the 2018 to 2022 of the WONDER figures. Across the 240 large counties the two sources correlate at 0.980.

Note: The key runs from 3.7 to 10.6, the fifth to ninety-fifth percentile of the mapped county rates. Counties beyond either end take the end color, so the end colors include the tails of the distribution.

Note: Connecticut is shown as its eight legacy counties, the coding of the source. Alaska is drawn. Puerto Rico is not drawn.

Source: County Health Rankings 2024 (NCHS mortality and natality counts; window and 20-death rule as in the CHR 2024 technical documentation, see the dataset note), infant deaths and live births in 2015-2021; crude rates from full counts

Data: https://vishalsingh.org/zipcode/report/infant-mortality/figures/imr_county_map.json

### Table 3. Infant mortality rises from 3.9 per 1,000 in the least deprived quarter of large counties to 7.2 in the most deprived

Infant deaths per 1,000 by county deprivation, WONDER 2018 to 2022 for the large counties and County Health Rankings 2015 to 2021 for all counties with a rate

#### Table 3.1. Large counties, 2018 to 2022

Across 240 large counties the infant mortality rate moves with the deprivation rank, a correlation of 0.80

Infant deaths per 1,000 against the county deprivation rank, one dot per county, 2018 to 2022, with births-weighted quartile means

| adi | births | conf_high | conf_low | county | imr |
| ---: | ---: | ---: | ---: | --- | ---: |
| 27.27 | 24712 | 5.68 | 3.95 | Ada County, ID | 4.82 |
| 31.50 | 30996 | 6.27 | 4.63 | Adams County, CO | 5.45 |
| 7.82 | 87316 | 3.81 | 3.04 | Alameda County, CA | 3.42 |
| 60.06 | 61976 | 6.97 | 5.71 | Allegheny County, PA | 6.34 |
| 67.85 | 26159 | 8.09 | 6.05 | Allen County, IN | 7.07 |
| 37.23 | 18908 | 6.14 | 4.06 | Anchorage Borough, AK | 5.02 |
| 26.31 | 34328 | 5.23 | 3.80 | Anne Arundel County, MD | 4.52 |
| 41.11 | 20680 | 5.51 | 3.63 | Anoka County, MN | 4.50 |
| 25.25 | 37521 | 5.53 | 4.12 | Arapahoe County, CO | 4.82 |
| 54.06 | 13899 | 7.64 | 4.95 | Atlantic County, NJ | 6.19 |
| 42.28 | 47439 | 7.24 | 5.79 | Baltimore County, MD | 6.51 |
| 64.94 | 37283 | 9.78 | 7.87 | Baltimore city, MD | 8.82 |
| 64.57 | 30340 | 7.68 | 5.83 | Bell County, TX | 6.76 |
| 19.42 | 45255 | 3.32 | 2.34 | Bergen County, NJ | 2.83 |
| 60.09 | 22847 | 7.10 | 5.07 | Berks County, PA | 6.08 |
| 54.44 | 33665 | 5.94 | 4.40 | Bernalillo County, NM | 5.17 |
| 59.17 | 130037 | 6.54 | 5.69 | Bexar County, TX | 6.11 |
| 49.43 | 23485 | 6.16 | 4.31 | Brazoria County, TX | 5.24 |
| 48.90 | 25720 | 7.23 | 5.29 | Brevard County, FL | 6.26 |
| 34.02 | 27992 | 5.09 | 3.55 | Bristol County, MA | 4.32 |
| 31.09 | 91577 | 5.39 | 4.48 | Bronx County, NY | 4.94 |
| 40.97 | 105205 | 5.56 | 4.69 | Broward County, FL | 5.12 |
| 29.71 | 28196 | 4.12 | 2.73 | Bucks County, PA | 3.37 |
| 40.10 | 22548 | 4.61 | 2.97 | Burlington County, NJ | 3.73 |
| 58.33 | 21858 | 8.50 | 6.23 | Butler County, OH | 7.37 |
| 76.41 | 14759 | 13.47 | 9.97 | Caddo Parish, LA | 11.72 |
| 54.47 | 30660 | 6.48 | 4.80 | Camden County, NJ | 5.64 |
| 84.05 | 30306 | 6.10 | 4.46 | Cameron County, TX | 5.28 |
| 35.04 | 24786 | 7.41 | 5.42 | Charleston County, SC | 6.41 |
| 56.20 | 18040 | 8.44 | 5.97 | Chatham County, GA | 7.21 |
| 26.68 | 27190 | 4.92 | 3.39 | Chester County, PA | 4.16 |
| 40.91 | 19995 | 7.02 | 4.88 | Chesterfield County, VA | 5.95 |
| 39.24 | 128486 | 5.71 | 4.92 | Clark County, NV | 5.32 |
| 26.64 | 27580 | 4.93 | 3.41 | Clark County, WA | 4.17 |
| 71.44 | 18045 | 9.94 | 7.24 | Clayton County, GA | 8.59 |
| 59.97 | 13801 | 8.41 | 5.57 | Cleveland County, OK | 6.88 |
| 37.38 | 44285 | 6.11 | 4.73 | Cobb County, GA | 5.42 |
| 26.55 | 59079 | 4.58 | 3.55 | Collin County, TX | 4.06 |
| 12.71 | 58952 | 4.22 | 3.24 | Contra Costa County, CA | 3.73 |
| 47.08 | 282978 | 6.27 | 5.70 | Cook County, IL | 5.99 |
| 30.73 | 13937 | 6.02 | 3.66 | Cumberland County, ME | 4.74 |
| 72.53 | 26502 | 9.36 | 7.17 | Cumberland County, NC | 8.26 |
| 69.33 | 66965 | 8.62 | 7.27 | Cuyahoga County, OH | 7.94 |
| 35.84 | 25220 | 4.34 | 2.83 | Dakota County, MN | 3.53 |
| 51.84 | 184635 | 6.68 | 5.96 | Dallas County, TX | 6.32 |
| 37.44 | 28463 | 5.66 | 4.04 | Dane County, WI | 4.85 |
| 62.88 | 16619 | 7.79 | 5.33 | Dauphin County, PA | 6.56 |
| 38 | 49796 | 7.43 | 5.99 | Davidson County, TN | 6.71 |
| 26.64 | 24783 | 5.75 | 4.01 | Davis County, UT | 4.88 |
| 44.32 | 51244 | 7.61 | 6.17 | DeKalb County, GA | 6.89 |
| 47.79 | 31972 | 7.26 | 5.51 | Delaware County, PA | 6.38 |
| 31.79 | 52331 | 4.76 | 3.65 | Denton County, TX | 4.20 |
| 21.31 | 45889 | 5.36 | 4.10 | Denver County, CO | 4.73 |
| 16.41 | 43900 | 6.69 | 5.25 | District of Columbia, DC | 5.97 |
| 55.92 | 40205 | 6.94 | 5.40 | Douglas County, NE | 6.17 |
| 34.30 | 49162 | 4.39 | 3.30 | DuPage County, IL | 3.84 |
| 41.88 | 20648 | 6.90 | 4.82 | Durham County, NC | 5.86 |
| 55.54 | 63717 | 8.47 | 7.10 | Duval County, FL | 7.78 |
| 61.83 | 28385 | 9.38 | 7.25 | East Baton Rouge Parish, LA | 8.31 |
| 32.73 | 45669 | 6.25 | 4.88 | El Paso County, CO | 5.56 |
| 74.50 | 56931 | 4.90 | 3.81 | El Paso County, TX | 4.36 |
| 62.29 | 47598 | 6.99 | 5.57 | Erie County, NY | 6.28 |
| 74.57 | 13589 | 9.12 | 6.18 | Erie County, PA | 7.65 |
| 62.76 | 18763 | 9.27 | 6.72 | Escambia County, FL | 7.99 |
| 21.97 | 42043 | 4.40 | 3.22 | Essex County, MA | 3.81 |
| 33.03 | 49864 | 5.40 | 4.19 | Essex County, NJ | 4.79 |
| 14.29 | 64587 | 4.28 | 3.33 | Fairfax County, VA | 3.81 |
| 26.85 | 49089 | 4.33 | 3.24 | Fairfield County, CT | 3.79 |
| 55.11 | 18798 | 6.89 | 4.71 | Fayette County, KY | 5.80 |
| 62.43 | 22019 | 9.61 | 7.19 | Forsyth County, NC | 8.40 |
| 37.63 | 45977 | 5 | 3.79 | Fort Bend County, TX | 4.39 |
| 55.13 | 89264 | 8.11 | 6.97 | Franklin County, OH | 7.54 |
| 46.34 | 70267 | 6.50 | 5.36 | Fresno County, CA | 5.93 |
| 36.01 | 61856 | 7.68 | 6.36 | Fulton County, GA | 7.02 |
| 50.28 | 19552 | 6.29 | 4.25 | Galveston County, TX | 5.27 |
| 73.71 | 22242 | 8.41 | 6.16 | Genesee County, MI | 7.28 |
| 68.78 | 16516 | 7.31 | 4.92 | Greene County, MO | 6.12 |
| 52.19 | 31893 | 6.17 | 4.56 | Greenville County, SC | 5.36 |
| 62.39 | 29763 | 9.55 | 7.45 | Guilford County, NC | 8.50 |
| 41.21 | 54430 | 6.31 | 5.04 | Gwinnett County, GA | 5.68 |
| 33.43 | 18889 | 5.80 | 3.78 | Hamilton County, IN | 4.71 |
| 59.80 | 52488 | 8.79 | 7.25 | Hamilton County, OH | 8.02 |
| 51.80 | 21724 | 7.36 | 5.25 | Hamilton County, TN | 6.31 |
| 54.48 | 24027 | 6.25 | 4.40 | Hampden County, MA | 5.33 |
| 55.59 | 326538 | 5.90 | 5.39 | Harris County, TX | 5.65 |
| 47.60 | 44945 | 5.38 | 4.10 | Hartford County, CT | 4.74 |
| 36.21 | 75308 | 5.15 | 4.17 | Hennepin County, MN | 4.66 |
| 42.82 | 18663 | 7.23 | 4.99 | Henrico County, VA | 6.11 |
| 83.04 | 69018 | 5.14 | 4.13 | Hidalgo County, TX | 4.64 |
| 45.98 | 85289 | 7.12 | 6.03 | Hillsborough County, FL | 6.58 |
| 35.48 | 20792 | 4.90 | 3.14 | Hillsborough County, NH | 3.94 |
| 11.87 | 57712 | 6.02 | 4.82 | Honolulu County, HI | 5.42 |
| 57.80 | 15979 | 7.97 | 5.43 | Horry County, SC | 6.70 |
| 19.35 | 16518 | 6.36 | 4.11 | Howard County, MD | 5.15 |
| 24.79 | 47913 | 3 | 2.09 | Hudson County, NJ | 2.55 |
| 68 | 14800 | 8.45 | 5.74 | Ingham County, MI | 7.09 |
| 64.74 | 43660 | 8.04 | 6.44 | Jackson County, MO | 7.24 |
| 62.60 | 40646 | 9.41 | 7.62 | Jefferson County, AL | 8.51 |
| 19.12 | 27859 | 5.23 | 3.67 | Jefferson County, CO | 4.45 |
| 60.18 | 47503 | 6.76 | 5.36 | Jefferson County, KY | 6.06 |
| 75.11 | 16350 | 9.52 | 6.75 | Jefferson County, TX | 8.13 |
| 60.17 | 27429 | 7.40 | 5.50 | Jefferson Parish, LA | 6.45 |
| 35.27 | 34859 | 4.71 | 3.38 | Johnson County, KS | 4.04 |
| 60.96 | 14235 | 6.76 | 4.27 | Kalamazoo County, MI | 5.41 |
| 45.72 | 29065 | 6.06 | 4.40 | Kane County, IL | 5.23 |
| 51.75 | 41330 | 6.13 | 4.71 | Kent County, MI | 5.42 |
| 52.63 | 63025 | 6.17 | 5 | Kern County, CA | 5.59 |
| 12.92 | 118597 | 4.30 | 3.59 | King County, WA | 3.95 |
| 9.96 | 177304 | 3.65 | 3.11 | Kings County, NY | 3.38 |
| 52.49 | 26154 | 6.98 | 5.10 | Knox County, TN | 6.04 |
| 52.55 | 17380 | 6.93 | 4.63 | Lake County, FL | 5.70 |
| 42.15 | 34279 | 4.79 | 3.43 | Lake County, IL | 4.11 |
| 65.67 | 27602 | 7.86 | 5.90 | Lake County, IN | 6.88 |
| 55.73 | 18657 | 6.41 | 4.31 | Lancaster County, NE | 5.36 |
| 49.56 | 34889 | 6.56 | 4.96 | Lancaster County, PA | 5.76 |
| 37.21 | 15763 | 6.03 | 3.80 | Lane County, OR | 4.82 |
| 24.18 | 15478 | 5.57 | 3.41 | Larimer County, CO | 4.39 |
| 47.61 | 35217 | 6.44 | 4.87 | Lee County, FL | 5.65 |
| 54.75 | 20497 | 6.85 | 4.76 | Lehigh County, PA | 5.81 |
| 57.97 | 14368 | 11.13 | 7.94 | Leon County, FL | 9.54 |
| 62.02 | 16300 | 7.33 | 4.88 | Lexington County, SC | 6.01 |
| 64.59 | 16003 | 7.19 | 4.75 | Lorain County, OH | 5.87 |
| 14.23 | 507627 | 4.05 | 3.70 | Los Angeles County, CA | 3.87 |
| 11.94 | 24624 | 4.27 | 2.76 | Loudoun County, VA | 3.45 |
| 67.26 | 20115 | 6.38 | 4.36 | Lubbock County, TX | 5.37 |
| 74.43 | 25684 | 10.20 | 7.87 | Lucas County, OH | 9.03 |
| 75.48 | 16015 | 7.11 | 4.69 | Luzerne County, PA | 5.81 |
| 58.72 | 44601 | 5.68 | 4.36 | Macomb County, MI | 5.02 |
| 53.80 | 21699 | 7.07 | 5 | Madison County, AL | 6.04 |
| 46.74 | 17713 | 6.75 | 4.54 | Manatee County, FL | 5.65 |
| 36.11 | 253434 | 5.68 | 5.10 | Maricopa County, AZ | 5.39 |
| 66.59 | 17607 | 8.41 | 5.91 | Marion County, FL | 7.16 |
| 67.02 | 69959 | 8.79 | 7.45 | Marion County, IN | 8.12 |
| 38.99 | 20047 | 4.91 | 3.12 | Marion County, OR | 3.94 |
| 40.70 | 73809 | 5.69 | 4.66 | Mecklenburg County, NC | 5.18 |
| 42.31 | 19125 | 5.32 | 3.41 | Merced County, CA | 4.29 |
| 42.50 | 20404 | 5.97 | 4.03 | Mercer County, NJ | 5 |
| 37.24 | 146849 | 5.11 | 4.41 | Miami-Dade County, FL | 4.76 |
| 14.53 | 81044 | 3.07 | 2.36 | Middlesex County, MA | 2.71 |
| 29.18 | 43015 | 4.40 | 3.23 | Middlesex County, NJ | 3.81 |
| 66.79 | 61939 | 9.02 | 7.58 | Milwaukee County, WI | 8.30 |
| 72.95 | 26321 | 9.38 | 7.18 | Mobile County, AL | 8.28 |
| 20.24 | 30024 | 3.76 | 2.47 | Monmouth County, NJ | 3.06 |
| 65.30 | 38357 | 6.91 | 5.34 | Monroe County, NY | 6.13 |
| 16.90 | 28437 | 5.81 | 4.17 | Monterey County, CA | 4.99 |
| 18.93 | 59475 | 5.05 | 3.97 | Montgomery County, MD | 4.51 |
| 72.95 | 31169 | 9.12 | 7.12 | Montgomery County, OH | 8.12 |
| 31.41 | 42442 | 4.40 | 3.23 | Montgomery County, PA | 3.82 |
| 43.03 | 38709 | 5.19 | 3.85 | Montgomery County, TX | 4.52 |
| 24.43 | 38803 | 4.51 | 3.27 | Multnomah County, OR | 3.89 |
| 13.81 | 69152 | 3.29 | 2.49 | Nassau County, NY | 2.89 |
| 43.82 | 30714 | 7.17 | 5.40 | New Castle County, DE | 6.28 |
| 48.10 | 42444 | 5.94 | 4.56 | New Haven County, CT | 5.25 |
| 8.92 | 75763 | 3.10 | 2.36 | New York County, NY | 2.73 |
| 14.99 | 35515 | 3.39 | 2.27 | Norfolk County, MA | 2.79 |
| 68.22 | 21911 | 6.26 | 4.33 | Nueces County, TX | 5.29 |
| 43.65 | 63665 | 5.28 | 4.21 | Oakland County, MI | 4.74 |
| 35.14 | 47818 | 3.39 | 2.42 | Ocean County, NJ | 2.91 |
| 66.32 | 55427 | 7.28 | 5.93 | Oklahoma County, OK | 6.60 |
| 67.47 | 24676 | 6.31 | 4.47 | Onondaga County, NY | 5.39 |
| 11.18 | 163437 | 3.50 | 2.95 | Orange County, CA | 3.22 |
| 44.12 | 82031 | 6.25 | 5.21 | Orange County, FL | 5.73 |
| 36.70 | 26515 | 4.50 | 3 | Orange County, NY | 3.70 |
| 51.55 | 21909 | 8.34 | 6.09 | Orleans Parish, LA | 7.21 |
| 47.66 | 23048 | 5.38 | 3.65 | Osceola County, FL | 4.51 |
| 38.83 | 73930 | 5.19 | 4.20 | Palm Beach County, FL | 4.69 |
| 55.44 | 26164 | 6.20 | 4.43 | Pasco County, FL | 5.31 |
| 34.54 | 31863 | 4.41 | 3.06 | Passaic County, NJ | 3.73 |
| 62.40 | 99523 | 7.53 | 6.49 | Philadelphia County, PA | 7.01 |
| 26.23 | 55649 | 5.73 | 4.54 | Pierce County, WA | 5.14 |
| 53.40 | 51086 | 6.21 | 4.91 | Pima County, AZ | 5.56 |
| 50.23 | 23631 | 6.63 | 4.71 | Pinal County, AZ | 5.67 |
| 48.07 | 38298 | 6.59 | 5.06 | Pinellas County, FL | 5.82 |
| 23.13 | 26014 | 4.34 | 2.85 | Plymouth County, MA | 3.54 |
| 62.59 | 41609 | 8.18 | 6.53 | Polk County, FL | 7.35 |
| 55.05 | 32415 | 5.70 | 4.17 | Polk County, IA | 4.94 |
| 33.05 | 57070 | 7.60 | 6.24 | Prince George's County, MD | 6.92 |
| 23.33 | 31727 | 5.76 | 4.20 | Prince William County, VA | 4.98 |
| 42.93 | 34996 | 6.14 | 4.60 | Providence County, RI | 5.37 |
| 68.10 | 25254 | 10.33 | 7.97 | Pulaski County, AR | 9.15 |
| 16.43 | 127498 | 4.01 | 3.35 | Queens County, NY | 3.68 |
| 46.60 | 34908 | 6.68 | 5.07 | Ramsey County, MN | 5.87 |
| 64.44 | 23223 | 8.14 | 5.98 | Richland County, SC | 7.06 |
| 15.94 | 25520 | 4.99 | 3.40 | Richmond County, NY | 4.19 |
| 28.37 | 139031 | 4.83 | 4.12 | Riverside County, CA | 4.47 |
| 21.10 | 29748 | 3.53 | 2.28 | Rockland County, NY | 2.86 |
| 41.48 | 21292 | 7.36 | 5.23 | Rutherford County, TN | 6.29 |
| 27.87 | 92812 | 5.60 | 4.68 | Sacramento County, CA | 5.14 |
| 27.27 | 77022 | 5.92 | 4.88 | Salt Lake County, UT | 5.40 |
| 31.95 | 137256 | 5.86 | 5.07 | San Bernardino County, CA | 5.46 |
| 13.98 | 190765 | 3.87 | 3.33 | San Diego County, CA | 3.60 |
| 3.15 | 39750 | 3.17 | 2.16 | San Francisco County, CA | 2.67 |
| 28.64 | 49351 | 6.08 | 4.78 | San Joaquin County, CA | 5.43 |
| 3.18 | 39199 | 3.41 | 2.35 | San Mateo County, CA | 2.88 |
| 17.02 | 27310 | 4.86 | 3.34 | Santa Barbara County, CA | 4.10 |
| 5.05 | 100305 | 3.03 | 2.39 | Santa Clara County, CA | 2.71 |
| 68.53 | 33214 | 7.85 | 6.06 | Sedgwick County, KS | 6.95 |
| 40.38 | 22970 | 6.59 | 4.65 | Seminole County, FL | 5.62 |
| 66.30 | 63231 | 9.03 | 7.61 | Shelby County, TN | 8.32 |
| 17.55 | 48396 | 4.04 | 2.98 | Snohomish County, WA | 3.51 |
| 20.56 | 24894 | 5.20 | 3.56 | Solano County, CA | 4.38 |
| 14.25 | 22266 | 3.82 | 2.33 | Sonoma County, CA | 3.01 |
| 66.48 | 19748 | 6.44 | 4.39 | Spartanburg County, SC | 5.42 |
| 40.64 | 29031 | 5.14 | 3.61 | Spokane County, WA | 4.37 |
| 44.15 | 21156 | 5.86 | 3.97 | St. Charles County, MO | 4.92 |
| 70.57 | 14336 | 10.17 | 7.13 | St. Clair County, IL | 8.65 |
| 71.71 | 16710 | 9.18 | 6.50 | St. Joseph County, IN | 7.84 |
| 53.57 | 54322 | 6.56 | 5.26 | St. Louis County, MO | 5.91 |
| 70.84 | 18927 | 10.39 | 7.68 | St. Louis city, MO | 9.03 |
| 50.80 | 16350 | 6.43 | 4.15 | St. Lucie County, FL | 5.20 |
| 33.77 | 36279 | 5.78 | 4.31 | Stanislaus County, CA | 5.04 |
| 71.61 | 19961 | 7.58 | 5.35 | Stark County, OH | 6.46 |
| 16.54 | 42713 | 4.50 | 3.32 | Suffolk County, MA | 3.91 |
| 19.74 | 77073 | 4.32 | 3.44 | Suffolk County, NY | 3.88 |
| 66.08 | 28116 | 7.07 | 5.24 | Summit County, OH | 6.15 |
| 48.92 | 136253 | 6.25 | 5.44 | Tarrant County, TX | 5.84 |
| 30.13 | 15251 | 6.53 | 4.16 | Thurston County, WA | 5.25 |
| 27.39 | 77756 | 4.53 | 3.63 | Travis County, TX | 4.08 |
| 50.21 | 33845 | 5.81 | 4.30 | Tulare County, CA | 5.05 |
| 64.81 | 43862 | 8.17 | 6.56 | Tulsa County, OK | 7.36 |
| 28.09 | 33101 | 4.57 | 3.22 | Union County, NJ | 3.90 |
| 26.08 | 58954 | 5.36 | 4.24 | Utah County, UT | 4.80 |
| 14.46 | 43372 | 4.43 | 3.27 | Ventura County, CA | 3.85 |
| 38.95 | 26879 | 6.51 | 4.72 | Virginia Beach city, VA | 5.62 |
| 55.90 | 24368 | 6.61 | 4.72 | Volusia County, FL | 5.66 |
| 32 | 64176 | 5.88 | 4.75 | Wake County, NC | 5.31 |
| 21.83 | 30945 | 4.47 | 3.10 | Washington County, OR | 3.78 |
| 28.88 | 25572 | 6.64 | 4.78 | Washoe County, NV | 5.71 |
| 40.20 | 16820 | 6.77 | 4.46 | Washtenaw County, MI | 5.53 |
| 34.74 | 19067 | 5.69 | 3.70 | Waukesha County, WI | 4.62 |
| 72.84 | 109697 | 10.22 | 9.05 | Wayne County, MI | 9.64 |
| 73.52 | 22421 | 5.87 | 4.03 | Webb County, TX | 4.95 |
| 32.31 | 22285 | 4.77 | 3.09 | Weld County, CO | 3.86 |
| 20.19 | 49530 | 4.29 | 3.22 | Westchester County, NY | 3.76 |
| 66.04 | 14352 | 6.78 | 4.30 | Westmoreland County, PA | 5.43 |
| 43.71 | 35331 | 5.87 | 4.38 | Will County, IL | 5.12 |
| 31.02 | 35958 | 4.39 | 3.12 | Williamson County, TX | 3.75 |
| 76.98 | 16888 | 9.41 | 6.70 | Winnebago County, IL | 8.05 |
| 35.79 | 42128 | 4.72 | 3.49 | Worcester County, MA | 4.11 |
| 58.42 | 23830 | 5.62 | 3.87 | York County, PA | 4.74 |

#### Table 3.2. All counties with a rate, 2015 to 2021

Across all 1,206 counties with a published rate the rate is 4.0 in the least deprived tenth and 8.8 in the most deprived

Births-weighted infant deaths per 1,000 by decile of county deprivation rank, counties with at least 20 infant deaths, County Health Rankings, 2015 to 2021

| conf_high | conf_low | decile | estimate | n |
| ---: | ---: | ---: | ---: | ---: |
| 4.19 | 3.83 | 1 | 4.00 | 6,629,995 |
| 5.32 | 4.88 | 2 | 5.11 | 4,124,889 |
| 5.97 | 5.51 | 3 | 5.78 | 3,070,833 |
| 6.34 | 5.86 | 4 | 6.09 | 2,816,097 |
| 7.06 | 6.37 | 5 | 6.72 | 2,100,337 |
| 7.34 | 6.70 | 6 | 7.05 | 1,871,135 |
| 8.12 | 7.12 | 7 | 7.65 | 1,376,448 |
| 7.96 | 6.73 | 8 | 7.38 | 935,698 |
| 7.92 | 6.49 | 9 | 7.24 | 748,936 |
| 9.46 | 8.05 | 10 | 8.77 | 467,875 |

Note: The counties are the 240 of 261 named counties (at least 250,000 residents) with a published death count in every year; they hold 62.3% of births in the United States. The other 21 named counties have a withheld year (44 withheld county-years in all, each of 1 to 9 deaths) and are left out whole, not pooled over their published years.

Note: A county's deprivation rank is the population-weighted mean of the Area Deprivation Index national ranks of its census tracts (1 is least deprived, 100 most), weighted by 2019-2023 tract population. Quartiles hold equal numbers of counties, so the most deprived quarter is relative to large counties and not to the nation.

Note: Quartile means are births-weighted. Intervals are 95% intervals from a bootstrap over counties (replicates and seed in the methods note).

Note: The second tab uses County Health Rankings rates for counties with at least 20 infant deaths in 2015 to 2021, so it leaves out the smallest counties, and its deciles are deciles of the counties that publish.

Note: Across all the counties in the second tab the rate is 5.7 per 1,000.

Source: CDC WONDER linked birth and infant death, 2018-2022, and Area Deprivation Index 2023; crude rates from full counts, births-weighted (population-weighted); second tab: County Health Rankings 2024, 2015-2021

Data: https://vishalsingh.org/zipcode/report/infant-mortality/figures/imr_by_deprivation.json

### Table 4. The rise with county deprivation holds within states and for births to white mothers and to Black mothers, at 0.63 deaths per 1,000 for each 10-point step across all births

Change in infant deaths per 1,000 for each 10-point rise in county deprivation rank (births-weighted slope) with 95% intervals, counties of 250,000 residents or more

| conf_high | conf_low | group | label | n | slope |
| ---: | ---: | --- | --- | ---: | ---: |
| 0.715 | 0.544 | All births | All births, 2018 to 2022 (240 counties) | 240 | 0.631 |
| 0.840 | 0.467 | All births | All births, 2018 to 2022, within states (44 states) | 240 | 0.624 |
| 0.709 | 0.518 | All births | All births, 2018 and 2019 (240 counties) | 240 | 0.619 |
| 0.440 | 0.289 | White | White mothers, 2018 and 2019 (203 counties) | 203 | 0.362 |
| 1.01 | 0.617 | Black | Black mothers, 2018 and 2019 (115 counties) | 115 | 0.817 |

Note: The slope is the births-weighted regression slope of the county rate on the county deprivation rank, multiplied by ten. Intervals come from a bootstrap over counties, except the within-states row, which resamples whole states because the fixed effects pool counties within a state.

Note: Race rows use WONDER's single-race groups of any ethnicity and the 240 counties, restricted to the counties that publish the group in both 2018 and 2019: 203 counties for white mothers (93.9% of their births in the counties used) and 115 for Black mothers (82.5%).

Note: Counties with a higher deprivation rank also have a larger share of births to Black mothers, 11.1% of births in the least deprived quartile and 31.1% in the most deprived (2018 and 2019), so the slope for all births mixes the rise within each group with the shift in the mix of mothers. A county slope describes places, not individual mothers.

Source: CDC WONDER linked birth and infant death, 2018-2022 (mother's race 2018-2019), and Area Deprivation Index 2023; crude rates from full counts, births-weighted (population-weighted)

Data: https://vishalsingh.org/zipcode/report/infant-mortality/figures/imr_gradient_checks.json

### Table 5. Infant mortality in 2018 and 2019 was 10.3 per 1,000 for Black mothers and 4.7 for white mothers, a ratio of 2.2

Infant deaths per 1,000 by mother's single race, any ethnicity, United States, 2018 and 2019, with 95% intervals

| conf_high | conf_low | label | n | race | rate |
| ---: | ---: | --- | ---: | --- | ---: |
| 5.68 | 5.57 | All mothers | 7,539,252 | All | 5.63 |
| 10.49 | 10.14 | Black (any ethnicity) | 1,255,629 | Black | 10.32 |
| 4.78 | 4.66 | White (any ethnicity) | 5,617,093 | White | 4.72 |
| 4.08 | 3.74 | Asian or Pacific Islander (any ethnicity) | 524,395 | Asian | 3.91 |

Note: WONDER counts mothers by single race and includes Hispanic mothers in each group, so the white group here is not the non-Hispanic white group of County Health Rankings. Mother's race is reported in the linked file only for 2018 and 2019.

Note: A state-year cell with fewer than 10 deaths is withheld (1 to 9 deaths) and leaves both deaths and births. The withheld cells hold 1.0% of births to Black mothers, 0.1% of births to white mothers and 6.7% of births to Asian or Pacific Islander mothers.

Note: For American Indian or Alaska Native mothers WONDER withholds 66 of 102 state-year cells and the published cells hold only 65.6% of their births, so a pooled rate would describe the states with the most deaths. No rate is shown.

Note: Intervals are gamma or normal intervals on the death counts; they treat deaths as independent and leave out differences between states.

Source: CDC WONDER linked birth and infant death, 2018-2019, mother's race (crude rates from full counts)

Data: https://vishalsingh.org/zipcode/report/infant-mortality/figures/imr_by_race.json

### Table 6. Black mothers' infants died at a higher rate than white mothers' in all 36 states that publish both, by a median ratio of 2.1

Infant deaths per 1,000 by mother's single race, any ethnicity, 36 states, 2018 and 2019, with 95% intervals, ordered by the Black-white difference

| black | hi_b | hi_w | lo_b | lo_w | n_b | n_w | state | white |
| ---: | ---: | ---: | ---: | ---: | ---: | ---: | --- | ---: |
| 11.67 | 14.62 | 4.78 | 9.18 | 2.66 | 6,429 | 13,325 | Delaware | 3.60 |
| 12.73 | 13.79 | 5.02 | 11.68 | 4.35 | 44,064 | 163,307 | Michigan | 4.68 |
| 12.66 | 14.45 | 5.41 | 10.87 | 4.56 | 15,162 | 103,924 | Wisconsin | 4.98 |
| 12.29 | 16.08 | 5.35 | 9.21 | 4.05 | 4,312 | 42,773 | Nebraska | 4.70 |
| 12.97 | 13.96 | 5.86 | 11.97 | 5.22 | 50,431 | 208,964 | Ohio | 5.54 |
| 12.13 | 13.09 | 5.19 | 11.18 | 4.60 | 51,108 | 213,812 | Illinois | 4.90 |
| 11.72 | 12.83 | 5.24 | 10.61 | 4.25 | 36,444 | 73,958 | South Carolina | 4.75 |
| 12.15 | 15.23 | 5.89 | 9.56 | 4.74 | 6,173 | 62,047 | Kansas | 5.32 |
| 12.69 | 14.83 | 6.76 | 10.56 | 5.63 | 10,713 | 73,285 | Oklahoma | 6.19 |
| 11.93 | 13.07 | 6.03 | 10.78 | 5.18 | 34,960 | 121,143 | Tennessee | 5.60 |
| 11.16 | 11.97 | 5.39 | 10.34 | 4.69 | 64,533 | 157,420 | North Carolina | 5.04 |
| 10.63 | 11.25 | 4.81 | 10.01 | 4.34 | 106,859 | 317,717 | Florida | 4.58 |
| 10.41 | 13.25 | 4.92 | 8.05 | 3.90 | 6,338 | 65,737 | Iowa | 4.41 |
| 10.64 | 11.30 | 5.36 | 9.97 | 4.63 | 93,071 | 145,932 | Georgia | 5 |
| 10.60 | 11.53 | 5.29 | 9.68 | 4.68 | 47,243 | 207,572 | Pennsylvania | 4.99 |
| 10.99 | 12.06 | 6.16 | 9.92 | 5.10 | 36,953 | 76,765 | Alabama | 5.63 |
| 11.47 | 13.18 | 6.81 | 9.76 | 5.50 | 15,082 | 55,070 | Arkansas | 6.16 |
| 9.34 | 10.20 | 4.48 | 8.48 | 3.60 | 48,404 | 80,981 | Maryland | 4.04 |
| 9.43 | 11.17 | 4.49 | 7.69 | 3.80 | 11,984 | 131,533 | Washington | 4.14 |
| 9.77 | 10.36 | 5 | 9.17 | 4.65 | 105,253 | 604,692 | Texas | 4.83 |
| 10.37 | 11.66 | 5.88 | 9.07 | 5.03 | 23,828 | 116,241 | Missouri | 5.45 |
| 8.11 | 9.06 | 3.53 | 7.16 | 2.94 | 34,767 | 141,428 | New Jersey | 3.23 |
| 10.82 | 11.76 | 6.54 | 9.88 | 5.38 | 46,868 | 68,003 | Louisiana | 5.96 |
| 8.63 | 9.98 | 4.29 | 7.28 | 3.52 | 18,199 | 101,548 | Minnesota | 3.91 |
| 10.64 | 11.98 | 6.42 | 9.31 | 5.59 | 22,926 | 134,146 | Indiana | 6.01 |
| 9.06 | 13.09 | 5.77 | 6.02 | 3.61 | 3,092 | 16,093 | Rhode Island | 4.60 |
| 9.18 | 10.06 | 5.17 | 8.30 | 4.43 | 45,536 | 133,860 | Virginia | 4.80 |
| 7.96 | 8.56 | 3.81 | 7.37 | 3.38 | 87,384 | 307,616 | New York | 3.60 |
| 10.99 | 12.13 | 7.45 | 9.86 | 5.85 | 32,561 | 39,548 | Mississippi | 6.65 |
| 9.33 | 11.13 | 5.79 | 7.53 | 4.55 | 11,042 | 51,823 | Nevada | 5.17 |
| 9.20 | 10.97 | 5.47 | 7.43 | 4.70 | 11,303 | 132,097 | Arizona | 5.09 |
| 8.16 | 8.89 | 4.21 | 7.43 | 3.90 | 58,843 | 677,920 | California | 4.06 |
| 9 | 10.74 | 5.50 | 7.26 | 4.58 | 11,443 | 92,669 | Kentucky | 5.04 |
| 7.66 | 9.46 | 4.28 | 6.12 | 3.23 | 11,231 | 52,490 | Connecticut | 3.75 |
| 8.25 | 10.42 | 4.97 | 6.43 | 4.17 | 8,488 | 109,155 | Colorado | 4.57 |
| 6.83 | 7.97 | 3.90 | 5.68 | 3.18 | 20,073 | 104,559 | Massachusetts | 3.54 |

Note: Only states that publish infant deaths for both groups in both years appear; they hold 98.0% of births to Black mothers. A state-year cell of 1 to 9 deaths is withheld, and a state is shown only when no cell of either group is withheld, so states with few Black births are absent.

Note: WONDER counts mothers by single race and includes Hispanic mothers in each group. Intervals are gamma or normal intervals on the death counts and leave out differences between counties within a state.

Source: CDC WONDER linked birth and infant death, 2018-2019, mother's race (crude rates from full counts)

Data: https://vishalsingh.org/zipcode/report/infant-mortality/figures/imr_black_white_states.json

### Table 7. Neonatal deaths made up 68.7% of infant deaths in the 14 counties that publish every age group, and both rates are higher where deprivation is higher

Infant deaths per 1,000 by age at death, 14 counties of 250,000 residents or more, 2018 to 2022, ordered from the least to the most deprived county rank

| age | conf_high | conf_low | county | n | rate |
| --- | ---: | ---: | --- | ---: | ---: |
| Neonatal (under 28 days) | 2.69 | 2.21 | Orange County, CA (rank 11) | 163,437 | 2.45 |
| Postneonatal (28 to 364 days) | 0.880 | 0.610 | Orange County, CA (rank 11) | 163,437 | 0.750 |
| Neonatal (under 28 days) | 2.90 | 2.44 | San Diego County, CA (rank 14) | 190,765 | 2.67 |
| Postneonatal (28 to 364 days) | 1.06 | 0.790 | San Diego County, CA (rank 14) | 190,765 | 0.920 |
| Neonatal (under 28 days) | 2.82 | 2.53 | Los Angeles County, CA (rank 14) | 507,627 | 2.67 |
| Postneonatal (28 to 364 days) | 1.30 | 1.11 | Los Angeles County, CA (rank 14) | 507,627 | 1.20 |
| Neonatal (under 28 days) | 2.73 | 2.18 | Queens County, NY (rank 16) | 127,498 | 2.45 |
| Postneonatal (28 to 364 days) | 1.42 | 1.03 | Queens County, NY (rank 16) | 127,498 | 1.22 |
| Neonatal (under 28 days) | 4.27 | 3.61 | San Bernardino County, CA (rank 32) | 137,256 | 3.94 |
| Postneonatal (28 to 364 days) | 1.71 | 1.30 | San Bernardino County, CA (rank 32) | 137,256 | 1.51 |
| Neonatal (under 28 days) | 3.88 | 3.41 | Maricopa County, AZ (rank 36) | 253,434 | 3.65 |
| Postneonatal (28 to 364 days) | 1.92 | 1.59 | Maricopa County, AZ (rank 36) | 253,434 | 1.76 |
| Neonatal (under 28 days) | 3.56 | 2.98 | Miami-Dade County, FL (rank 37) | 146,849 | 3.27 |
| Postneonatal (28 to 364 days) | 1.68 | 1.29 | Miami-Dade County, FL (rank 37) | 146,849 | 1.48 |
| Neonatal (under 28 days) | 4.35 | 3.88 | Cook County, IL (rank 47) | 282,978 | 4.11 |
| Postneonatal (28 to 364 days) | 2.03 | 1.71 | Cook County, IL (rank 47) | 282,978 | 1.87 |
| Neonatal (under 28 days) | 4.34 | 3.66 | Tarrant County, TX (rank 49) | 136,253 | 4 |
| Postneonatal (28 to 364 days) | 2.09 | 1.63 | Tarrant County, TX (rank 49) | 136,253 | 1.86 |
| Neonatal (under 28 days) | 4.62 | 4.02 | Dallas County, TX (rank 52) | 184,635 | 4.32 |
| Postneonatal (28 to 364 days) | 2.20 | 1.79 | Dallas County, TX (rank 52) | 184,635 | 1.99 |
| Neonatal (under 28 days) | 5.66 | 4.71 | Franklin County, OH (rank 55) | 89,264 | 5.19 |
| Postneonatal (28 to 364 days) | 2.65 | 2.01 | Franklin County, OH (rank 55) | 89,264 | 2.33 |
| Neonatal (under 28 days) | 3.91 | 3.50 | Harris County, TX (rank 56) | 326,538 | 3.71 |
| Postneonatal (28 to 364 days) | 2.09 | 1.78 | Harris County, TX (rank 56) | 326,538 | 1.94 |
| Neonatal (under 28 days) | 4.66 | 3.95 | Bexar County, TX (rank 59) | 130,037 | 4.31 |
| Postneonatal (28 to 364 days) | 2.03 | 1.57 | Bexar County, TX (rank 59) | 130,037 | 1.80 |
| Neonatal (under 28 days) | 6.65 | 5.72 | Wayne County, MI (rank 73) | 109,697 | 6.18 |
| Postneonatal (28 to 364 days) | 3.79 | 3.10 | Wayne County, MI (rank 73) | 109,697 | 3.45 |

Note: WONDER withholds 4,328 of 6,525 county-year-age cells in the named counties (1 to 9 deaths each), so only 14 counties publish every age group in every year: among the largest counties, holding 23.6% of births in the named counties. Counties with any withheld cell are left out whole; no group is summed from published cells alone and nothing is derived from totals.

Note: The row label gives the county deprivation rank (1 least deprived, 100 most). WONDER weights infant deaths, so the age groups can differ from a county's total by rounding.

Note: Intervals are gamma or normal intervals on each county's death count; they treat deaths as independent.

Source: CDC WONDER linked birth and infant death, 2018-2022, by age at death; crude rates from full counts

Data: https://vishalsingh.org/zipcode/report/infant-mortality/figures/imr_age_at_death.json

### Table 8. Infant mortality is steadily worse toward the more deprived end under the ADI and under each of the three other rankings

Infant deaths per 1,000 live births by tenth of counties ranked four ways, over the 1,206 counties that have all four rankings

| bin | conf_high | conf_low | estimate | n_units | ranking |
| ---: | ---: | ---: | ---: | ---: | --- |
| 1 | 4.21 | 3.84 | 4.01 | 120 | Area Deprivation Index |
| 2 | 5.28 | 4.85 | 5.07 | 121 | Area Deprivation Index |
| 3 | 5.98 | 5.51 | 5.78 | 120 | Area Deprivation Index |
| 4 | 6.32 | 5.81 | 6.06 | 121 | Area Deprivation Index |
| 5 | 6.96 | 6.35 | 6.65 | 121 | Area Deprivation Index |
| 6 | 7.44 | 6.79 | 7.14 | 120 | Area Deprivation Index |
| 7 | 8.11 | 7.10 | 7.64 | 121 | Area Deprivation Index |
| 8 | 7.98 | 6.78 | 7.41 | 120 | Area Deprivation Index |
| 9 | 7.86 | 6.47 | 7.21 | 121 | Area Deprivation Index |
| 10 | 9.39 | 7.98 | 8.68 | 121 | Area Deprivation Index |
| 1 | 4.21 | 3.82 | 4.01 | 120 | Standardized index |
| 2 | 5.18 | 4.44 | 4.80 | 121 | Standardized index |
| 3 | 5.77 | 5.25 | 5.50 | 120 | Standardized index |
| 4 | 6.33 | 4.99 | 5.61 | 121 | Standardized index |
| 5 | 6.51 | 5.56 | 6.02 | 121 | Standardized index |
| 6 | 6.93 | 5.78 | 6.33 | 120 | Standardized index |
| 7 | 7.61 | 6.90 | 7.28 | 121 | Standardized index |
| 8 | 8.30 | 7.18 | 7.72 | 120 | Standardized index |
| 9 | 8.09 | 6.65 | 7.32 | 121 | Standardized index |
| 10 | 8.35 | 6.13 | 6.95 | 121 | Standardized index |
| 1 | 4.11 | 3.72 | 3.91 | 120 | Median household income |
| 2 | 5.27 | 4.56 | 4.88 | 121 | Median household income |
| 3 | 5.85 | 5.10 | 5.50 | 120 | Median household income |
| 4 | 6.26 | 5.78 | 6.02 | 121 | Median household income |
| 5 | 6.58 | 5.82 | 6.19 | 121 | Median household income |
| 6 | 7.39 | 6.62 | 7.02 | 120 | Median household income |
| 7 | 7.81 | 6.60 | 7.25 | 121 | Median household income |
| 8 | 7.89 | 6.76 | 7.38 | 120 | Median household income |
| 9 | 8.30 | 6.66 | 7.49 | 121 | Median household income |
| 10 | 9.44 | 8.55 | 8.99 | 121 | Median household income |
| 1 | 4.61 | 4.07 | 4.33 | 120 | Bachelor's degree share |
| 2 | 5.43 | 4.57 | 4.98 | 121 | Bachelor's degree share |
| 3 | 6.12 | 5.40 | 5.76 | 120 | Bachelor's degree share |
| 4 | 6.79 | 5.31 | 5.99 | 121 | Bachelor's degree share |
| 5 | 6.43 | 5.84 | 6.12 | 121 | Bachelor's degree share |
| 6 | 7.41 | 5.68 | 6.49 | 120 | Bachelor's degree share |
| 7 | 7.19 | 6.35 | 6.70 | 121 | Bachelor's degree share |
| 8 | 7.14 | 6.05 | 6.58 | 120 | Bachelor's degree share |
| 9 | 7.54 | 6.55 | 7.00 | 121 | Bachelor's degree share |
| 10 | 8.47 | 7.18 | 7.76 | 121 | Bachelor's degree share |

Note: Least and most deprived tenth: ADI 4.0 and 8.7, rank correlation +0.95; standardized index 4.0 and 7.0, +0.92; median household income 3.9 and 9.0, +1.00; bachelor's degree share 4.3 and 7.8, +0.99.

Note: The standardized index applies the ADI's coefficients to its measures after standardizing each one, so home value and income carry less weight than in the ADI. Income and degree share are single census measures, each ranked from the highest value to the lowest.

Source: County Health Rankings 2024, infant deaths and live births in 2015-2021, and Area Deprivation Index 2023; crude rates from full counts, births-weighted (population-weighted); ADI 2023 and ACS 2019-2023 tract tables (B19013, B15003 and the 15 rebuilt ADI measures)

Data: https://vishalsingh.org/zipcode/report/infant-mortality/figures/alt_rankings.json

## All estimates

Every fact in facts.json (111), sorted by id.

| Fact id | Value | Unit | Estimand | Population | Period | Interval | n | Weights | Source | Note |
| --- | ---: | --- | --- | --- | --- | --- | ---: | --- | --- | --- |
| adi_above_pop_share | 3.3% | percent of residents | share of residents living in counties ranked above the highest county used, 2019-2023 tract population | residents of the 50 states and DC | 2019-2023 |   |   | 2019-2023 tract population | Area Deprivation Index |   |
| adi_max_used | 84 | national percentile rank (1 least deprived, 100 most) | highest county deprivation rank among the counties used | the counties used | 2019-2023 tract population weights |   | 240 counties | 2019-2023 tract population | Area Deprivation Index |   |
| adi_min_used | 3 | national percentile rank (1 least deprived, 100 most) | lowest county deprivation rank among the counties used | the counties used | 2019-2023 tract population weights |   | 240 counties | 2019-2023 tract population | Area Deprivation Index |   |
| age_births_share | 23.6% | percent of live births | share of births in the counties shown | live births in the named counties | 2018 to 2022 |   |   | none (full count) | CDC WONDER linked birth and infant death |   |
| age_cells | 6,525 | county-year-age cells | count of cells | named counties, five age groups, 2018 to 2022 | infant deaths and live births in 2018 to 2022 |   |   | none (full count) | CDC WONDER linked birth and infant death |   |
| age_cells_withheld | 4,328 | withheld cells | count of cells with 1 to 9 infant deaths, which WONDER withholds | named counties, five age groups, 2018 to 2022 | infant deaths and live births in 2018 to 2022 |   |   | none (full count) | CDC WONDER linked birth and infant death |   |
| age_counties | 14 | counties | count of counties | named counties that publish all five age groups in every year 2018 to 2022 | infant deaths and live births in 2018 to 2022 |   |   | none (full count) | CDC WONDER linked birth and infant death |   |
| aian_cells | 102 | state-year cells | count of state-year cells | American Indian or Alaska Native mothers (single race, any ethnicity) in 50 states and DC | infant deaths and live births in 2018 and 2019 |   |   | none (full count) | CDC WONDER linked birth and infant death |   |
| aian_published_share | 65.6% | percent of live births | share of births in state-years whose deaths WONDER publishes | live births to American Indian or Alaska Native mothers (single race, any ethnicity) | 2018 and 2019 |   |   | none (full counts of births) | CDC WONDER linked birth and infant death |   |
| aian_withheld_cells | 66 | withheld state-year cells | count of state-year cells with 1 to 9 infant deaths, which WONDER withholds | American Indian or Alaska Native mothers (single race, any ethnicity) in 50 states and DC | infant deaths and live births in 2018 and 2019 |   |   | none (full count) | CDC WONDER linked birth and infant death |   |
| alt_adi_least | 4.0 | Infant deaths per 1,000 live births | births-weighted mean of county rates; 95% intervals from a bootstrap over counties | the least deprived tenth of counties ranked by Area Deprivation Index, among the counties with all four rankings | infant deaths and live births in 2015 to 2021 | 95% CI 3.8 to 4.2 | 120 counties | births in each county (County Health Rankings denominator) | County Health Rankings |   |
| alt_adi_most | 8.7 | Infant deaths per 1,000 live births | births-weighted mean of county rates; 95% intervals from a bootstrap over counties | the most deprived tenth of counties ranked by Area Deprivation Index, among the counties with all four rankings | infant deaths and live births in 2015 to 2021 | 95% CI 8.0 to 9.4 | 121 counties | births in each county (County Health Rankings denominator) | County Health Rankings |   |
| alt_adi_rho | +0.95 | rank correlation | Spearman rank correlation between bin order and bin mean, each bin counted once | the bins of counties ranked by Area Deprivation Index | infant deaths and live births in 2015 to 2021 |   |   | each bin counted once | County Health Rankings |   |
| alt_degree_least | 4.3 | Infant deaths per 1,000 live births | births-weighted mean of county rates; 95% intervals from a bootstrap over counties | the least deprived tenth of counties ranked by Bachelor's degree share, among the counties with all four rankings | infant deaths and live births in 2015 to 2021 | 95% CI 4.1 to 4.6 | 120 counties | births in each county (County Health Rankings denominator) | County Health Rankings |   |
| alt_degree_most | 7.8 | Infant deaths per 1,000 live births | births-weighted mean of county rates; 95% intervals from a bootstrap over counties | the most deprived tenth of counties ranked by Bachelor's degree share, among the counties with all four rankings | infant deaths and live births in 2015 to 2021 | 95% CI 7.2 to 8.5 | 121 counties | births in each county (County Health Rankings denominator) | County Health Rankings |   |
| alt_degree_rho | +0.99 | rank correlation | Spearman rank correlation between bin order and bin mean, each bin counted once | the bins of counties ranked by Bachelor's degree share | infant deaths and live births in 2015 to 2021 |   |   | each bin counted once | County Health Rankings |   |
| alt_income_least | 3.9 | Infant deaths per 1,000 live births | births-weighted mean of county rates; 95% intervals from a bootstrap over counties | the least deprived tenth of counties ranked by Median household income, among the counties with all four rankings | infant deaths and live births in 2015 to 2021 | 95% CI 3.7 to 4.1 | 120 counties | births in each county (County Health Rankings denominator) | County Health Rankings |   |
| alt_income_most | 9.0 | Infant deaths per 1,000 live births | births-weighted mean of county rates; 95% intervals from a bootstrap over counties | the most deprived tenth of counties ranked by Median household income, among the counties with all four rankings | infant deaths and live births in 2015 to 2021 | 95% CI 8.5 to 9.4 | 121 counties | births in each county (County Health Rankings denominator) | County Health Rankings |   |
| alt_income_rho | +1.00 | rank correlation | Spearman rank correlation between bin order and bin mean, each bin counted once | the bins of counties ranked by Median household income | infant deaths and live births in 2015 to 2021 |   |   | each bin counted once | County Health Rankings |   |
| alt_n_places | 1,206 | counties | count of counties in the common sample of the ranking comparison | counties with the outcome and all four rankings | infant deaths and live births in 2015 to 2021 |   |   | none (full count) | County Health Rankings |   |
| alt_rho_cut | 0.9 | absolute rank correlation | threshold between a steady and an uneven gradient | the rule for changed gradients (docs/RANKING_SWITCH_RULE.md) | infant deaths and live births in 2015 to 2021 |   |   | none (a fixed rule) | County Health Rankings |   |
| alt_rho_margin | 0.1 | absolute rank correlation | smallest move in the rank correlation that counts as a change of shape | the rule for changed gradients (docs/RANKING_SWITCH_RULE.md) | infant deaths and live births in 2015 to 2021 |   |   | none (a fixed rule) | County Health Rankings |   |
| alt_std_least | 4.0 | Infant deaths per 1,000 live births | births-weighted mean of county rates; 95% intervals from a bootstrap over counties | the least deprived tenth of counties ranked by Standardized index, among the counties with all four rankings | infant deaths and live births in 2015 to 2021 | 95% CI 3.8 to 4.2 | 120 counties | births in each county (County Health Rankings denominator) | County Health Rankings |   |
| alt_std_most | 7.0 | Infant deaths per 1,000 live births | births-weighted mean of county rates; 95% intervals from a bootstrap over counties | the most deprived tenth of counties ranked by Standardized index, among the counties with all four rankings | infant deaths and live births in 2015 to 2021 | 95% CI 6.1 to 8.3 | 121 counties | births in each county (County Health Rankings denominator) | County Health Rankings |   |
| alt_std_rho | +0.92 | rank correlation | Spearman rank correlation between bin order and bin mean, each bin counted once | the bins of counties ranked by Standardized index | infant deaths and live births in 2015 to 2021 |   |   | each bin counted once | County Health Rankings |   |
| asian_left_out_share | 6.7% | percent of live births | share of births in state-years whose deaths WONDER withholds, which leave both deaths and births | live births to Asian or Pacific Islander mothers (single race, any ethnicity) in the United States | 2018 and 2019 |   |   | none (full counts of births) | CDC WONDER linked birth and infant death |   |
| births_left_out_share | 2.6% | percent of live births | share of births in the counties left out because a year is withheld | live births in the named counties | 2018 to 2022 |   |   | none (full count) | CDC WONDER linked birth and infant death |   |
| births_share_used | 62.3% | percent of live births in the United States | share of births in the counties used | live births in the 50 states and DC | 2018 to 2022 |   | 18,484,949 births | none (full count) | CDC WONDER linked birth and infant death |   |
| births_total | 18.5 million | live births | count of live births | live births in the 50 states and DC | 2018 to 2022 |   | 18,484,949 births | none (full count) | CDC WONDER linked birth and infant death |   |
| black_left_out_share | 1.0% | percent of live births | share of births in state-years whose deaths WONDER withholds, which leave both deaths and births | live births to Black mothers (single race, any ethnicity) in the United States | 2018 and 2019 |   |   | none (full counts of births) | CDC WONDER linked birth and infant death |   |
| black_share_q1 | 11.1% | percent of live births | share of births to Black mothers (single race, any ethnicity) over all births, in the county-years where WONDER publishes the group's births; a county-year with no published count leaves both the numerator and the denominator | live births in the least deprived quarter of the counties used, by county deprivation rank | 2018 and 2019 |   |   | none (full counts of births) | CDC WONDER linked birth and infant death |   |
| black_share_q4 | 31.1% | percent of live births | share of births to Black mothers (single race, any ethnicity) over all births, in the county-years where WONDER publishes the group's births; a county-year with no published count leaves both the numerator and the denominator | live births in the most deprived quarter of the counties used, by county deprivation rank | 2018 and 2019 |   |   | none (full counts of births) | CDC WONDER linked birth and infant death | Webb County, TX has no published count of births to Black mothers in 2 county-years, so those county-years are left out. Counting them as zero would give 30.7% for this quarter. |
| bw_black_births_share | 98.0% | percent of live births | share of those births in the states shown | live births to Black mothers (single race, any ethnicity) in the United States | 2018 and 2019 |   |   | none (full counts of births) | CDC WONDER linked birth and infant death |   |
| bw_ratio_high | 3.2 | ratio of rates | ratio of crude rates in the state with the largest ratio, log-normal interval on the two death counts | live births in Delaware, Black mothers over white mothers (single race, any ethnicity) | 2018 and 2019 | 95% CI 2.3 to 4.7 | 123 infant deaths | none (crude rates from counts) | CDC WONDER linked birth and infant death |   |
| bw_ratio_low | 1.7 | ratio of rates | ratio of crude rates in the state with the smallest ratio, log-normal interval on the two death counts | live births in Mississippi, Black mothers over white mothers (single race, any ethnicity) | 2018 and 2019 | 95% CI 1.4 to 1.9 | 621 infant deaths | none (crude rates from counts) | CDC WONDER linked birth and infant death |   |
| bw_ratio_median | 2.1 | ratio of rates | median of the state ratios | live births in the states shown, Black mothers over white mothers (single race, any ethnicity) | 2018 and 2019 |   | 36 states | none (each state counts once) | CDC WONDER linked birth and infant death |   |
| bw_states | 36 | states | count of states | states that publish infant deaths for both Black and white mothers in both 2018 and 2019 | infant deaths and live births in 2018 and 2019 |   |   | none (full count) | CDC WONDER linked birth and infant death |   |
| chr_all_corr | 0.76 | correlation coefficient | births-weighted correlation of the county rate with county deprivation rank, county-cluster bootstrap interval | live births in counties with a County Health Rankings rate | infant deaths and live births in 2015 to 2021 | 95% CI 0.71 to 0.80 | 1,206 counties | births in each county (County Health Rankings denominator) | County Health Rankings |   |
| chr_all_d1 | 4.0 | infant deaths per 1,000 live births | births-weighted mean of county rates, bootstrap interval over counties | live births in the least deprived tenth of counties with a County Health Rankings rate | infant deaths and live births in 2015 to 2021 | 95% CI 3.8 to 4.2 | 120 counties | births in each county (County Health Rankings denominator) | County Health Rankings |   |
| chr_all_d10 | 8.8 | infant deaths per 1,000 live births | births-weighted mean of county rates, bootstrap interval over counties | live births in the most deprived tenth of counties with a County Health Rankings rate | infant deaths and live births in 2015 to 2021 | 95% CI 8.1 to 9.5 | 121 counties | births in each county (County Health Rankings denominator) | County Health Rankings |   |
| chr_counties | 1,206 | counties | count of counties with at least 20 infant deaths in the period | counties and county equivalents with a County Health Rankings infant mortality rate and an ADI | infant deaths and live births in 2015 to 2021 |   |   | none (full count) | County Health Rankings |   |
| chr_lbd_corr | 0.980 | correlation coefficient | births-weighted correlation between the county rate in WONDER and in County Health Rankings, county-cluster bootstrap interval | live births in the counties of 250,000 residents or more that publish infant deaths in every year 2018 to 2022 | WONDER 2018 to 2022 and County Health Rankings 2015 to 2021 | 95% CI 0.974 to 0.985 | 240 counties | WONDER births in each county | County Health Rankings |   |
| counties_above_adi_max | 712 | counties | count of counties whose rank is above the highest rank among the counties used | counties and county equivalents in the 50 states and DC with a deprivation rank | 2019-2023 tract population weights |   | 3,128 counties | none (full count) | Area Deprivation Index |   |
| counties_left_out | 21 | counties | count of counties left out of the county rates | named counties with a withheld death count in at least one year 2018 to 2022 | infant deaths and live births in 2018 to 2022 |   |   | none (full count) | CDC WONDER linked birth and infant death |   |
| counties_named | 261 | counties | count of counties (two identifiers with no births are left out) | counties that WONDER linked birth and infant death names individually and that have births | infant deaths and live births in 2018 to 2022 |   |   | none (full count) | CDC WONDER linked birth and infant death |   |
| counties_not_mapped | 1,937 | counties | count of counties with no County Health Rankings rate | counties and county equivalents on the map (50 states and DC) | infant deaths and live births in 2015 to 2021 |   |   | none (full count) | County Health Rankings |   |
| counties_used | 240 | counties | count of counties | named counties with a published death count in every year 2018 to 2022 | infant deaths and live births in 2018 to 2022 |   |   | none (full count) | CDC WONDER linked birth and infant death |   |
| deaths_total | 102,503 | infant deaths | count of infant deaths (WONDER weights deaths, so the count can differ from the unweighted number of records by rounding) | infants born in the 50 states and DC | 2018 to 2022 |   |   | none (full count) | CDC WONDER linked birth and infant death |   |
| first_day_share | 42.3% | percent of infant deaths | pooled share that are deaths in the first 24 hours, bootstrap interval over counties | infant deaths in the 14 counties that publish all five age groups | 2018 to 2022 | 95% CI 40.5% to 44.2% | 14 counties | infant deaths in each county | CDC WONDER linked birth and infant death |   |
| imr_2021 | 5.44 | infant deaths per 1,000 live births | crude rate from the death count, normal interval | live births in the 50 states and DC | 2021 | 95% CI 5.36 to 5.51 | 19,927 infant deaths | none (crude rate from counts) | CDC WONDER linked birth and infant death |   |
| imr_adi_corr | 0.80 | correlation coefficient | births-weighted correlation of the county rate with county deprivation rank, county-cluster bootstrap interval | live births in the counties of 250,000 residents or more that publish infant deaths in every year 2018 to 2022 | 2018 to 2022 | 95% CI 0.73 to 0.86 | 240 counties | births in each county | CDC WONDER linked birth and infant death |   |
| imr_all_1819 | 5.6 | infant deaths per 1,000 live births | crude rate from the death count, normal interval | live births in the 50 states and DC | 2018 and 2019 | 95% CI 5.6 to 5.7 | 42,421 infant deaths | none (crude rate from counts) | CDC WONDER linked birth and infant death |   |
| imr_asian | 3.9 | infant deaths per 1,000 live births | crude rate from state counts, normal interval | live births to Asian or Pacific Islander mothers (single race, any ethnicity) in the United States, states with a published count | 2018 and 2019 | 95% CI 3.7 to 4.1 | 2,051 infant deaths | none (crude rate from counts) | CDC WONDER linked birth and infant death |   |
| imr_black | 10.3 | infant deaths per 1,000 live births | crude rate from state counts, normal interval | live births to Black mothers (single race, any ethnicity) in the United States, states with a published count | 2018 and 2019 | 95% CI 10.1 to 10.5 | 12,953 infant deaths | none (crude rate from counts) | CDC WONDER linked birth and infant death |   |
| imr_midwest | 6.1 | infant deaths per 1,000 live births | pooled crude rate over the region's states (12 states), bootstrap interval over those states | live births in the Midwest Census region | 2018 to 2022 | 95% CI 5.5 to 6.5 | 12 states | births in each state | CDC WONDER linked birth and infant death |   |
| imr_natl | 5.5 | infant deaths per 1,000 live births | crude rate from the death count, normal interval | live births in the 50 states and DC | 2018 to 2022 | 95% CI 5.5 to 5.6 | 102,503 infant deaths | none (crude rate from counts) | CDC WONDER linked birth and infant death |   |
| imr_northeast | 4.5 | infant deaths per 1,000 live births | pooled crude rate over the region's states (9 states), bootstrap interval over those states | live births in the Northeast Census region | 2018 to 2022 | 95% CI 3.9 to 5.2 | 9 states | births in each state | CDC WONDER linked birth and infant death |   |
| imr_q1 | 3.9 | infant deaths per 1,000 live births | births-weighted mean of county rates (the pooled rate), bootstrap interval over counties | live births in the least deprived quarter of the counties used, by county deprivation rank | 2018 to 2022 | 95% CI 3.7 to 4.1 | 60 counties | births in each county | CDC WONDER linked birth and infant death |   |
| imr_q4 | 7.2 | infant deaths per 1,000 live births | births-weighted mean of county rates (the pooled rate), bootstrap interval over counties | live births in the most deprived quarter of the counties used, by county deprivation rank | 2018 to 2022 | 95% CI 6.7 to 7.6 | 60 counties | births in each county | CDC WONDER linked birth and infant death |   |
| imr_q_diff | 3.3 | infant deaths per 1,000 live births | difference of two pooled rates, bootstrap interval over counties with shared draws | live births in the counties used, most deprived quarter minus least deprived quarter | 2018 to 2022 | 95% CI 2.8 to 3.8 | 240 counties | births in each county | CDC WONDER linked birth and infant death |   |
| imr_q_ratio | 1.9 | ratio of rates | ratio of two pooled rates, bootstrap interval over counties with shared draws | live births in the counties used, most deprived quarter over least deprived quarter | 2018 to 2022 | 95% CI 1.7 to 2.0 | 240 counties | births in each county | CDC WONDER linked birth and infant death |   |
| imr_race_gap | 5.6 | infant deaths per 1,000 live births | difference of crude rates, normal interval on the two death counts | live births in the United States, Black mothers minus white mothers (single race, any ethnicity) | 2018 and 2019 | 95% CI 5.4 to 5.8 | 39,469 infant deaths | none (crude rates from counts) | CDC WONDER linked birth and infant death |   |
| imr_race_ratio | 2.2 | ratio of rates | ratio of crude rates, log-normal interval on the two death counts | live births in the United States, Black mothers over white mothers (single race, any ethnicity) | 2018 and 2019 | 95% CI 2.1 to 2.2 | 39,469 infant deaths | none (crude rates from counts) | CDC WONDER linked birth and infant death |   |
| imr_rest | 5.1 | infant deaths per 1,000 live births | pooled crude rate over the states of the group (34 states), bootstrap interval over those states | live births outside the Census South region | 2018 to 2022 | 95% CI 4.6 to 5.7 | 34 states | births in each state | CDC WONDER linked birth and infant death |   |
| imr_slope_1819_per_10 | 0.62 | infant deaths per 1,000 live births for each ten points of county deprivation rank | births-weighted slope of the county rate on county deprivation rank, county-cluster bootstrap interval | live births in the counties of 250,000 residents or more that publish infant deaths in every year 2018 to 2022, deaths and births in 2018 and 2019 only | 2018 and 2019 | 95% CI 0.52 to 0.71 | 240 counties | births in each county | CDC WONDER linked birth and infant death |   |
| imr_slope_black_per_10 | 0.82 | infant deaths per 1,000 live births for each ten points of county deprivation rank | births-weighted slope of the county rate on county deprivation rank, county-cluster bootstrap interval | live births to Black mothers (single race, any ethnicity) in the counties used that publish the group in both years | 2018 and 2019 | 95% CI 0.62 to 1.01 | 115 counties | births in each county | CDC WONDER linked birth and infant death |   |
| imr_slope_fe_per_10 | 0.62 | infant deaths per 1,000 live births for each ten points of county deprivation rank | births-weighted slope of the county rate on county deprivation rank with state fixed effects, bootstrap interval over states | live births in the counties of 250,000 residents or more that publish infant deaths in every year 2018 to 2022 | 2018 to 2022 | 95% CI 0.47 to 0.84 | 44 states | births in each county | CDC WONDER linked birth and infant death |   |
| imr_slope_per_10 | 0.63 | infant deaths per 1,000 live births for each ten points of county deprivation rank | births-weighted slope of the county rate on county deprivation rank, county-cluster bootstrap interval | live births in the counties of 250,000 residents or more that publish infant deaths in every year 2018 to 2022 | 2018 to 2022 | 95% CI 0.54 to 0.71 | 240 counties | births in each county | CDC WONDER linked birth and infant death |   |
| imr_slope_white_per_10 | 0.36 | infant deaths per 1,000 live births for each ten points of county deprivation rank | births-weighted slope of the county rate on county deprivation rank, county-cluster bootstrap interval | live births to white mothers (single race, any ethnicity) in the counties used that publish the group in both years | 2018 and 2019 | 95% CI 0.29 to 0.44 | 203 counties | births in each county | CDC WONDER linked birth and infant death |   |
| imr_south | 6.3 | infant deaths per 1,000 live births | pooled crude rate over the states of the group (17 states), bootstrap interval over those states | live births in the Census South region (16 states and DC) | 2018 to 2022 | 95% CI 5.9 to 6.9 | 17 states | births in each state | CDC WONDER linked birth and infant death |   |
| imr_state_high | 8.7 | infant deaths per 1,000 live births | crude rate from the death count, normal interval | live births in Mississippi | 2018 to 2022 | 95% CI 8.3 to 9.2 | 1,564 infant deaths | none (crude rate from counts) | CDC WONDER linked birth and infant death |   |
| imr_state_low | 3.7 | infant deaths per 1,000 live births | crude rate from the death count, normal interval | live births in Massachusetts | 2018 to 2022 | 95% CI 3.5 to 3.9 | 1,260 infant deaths | none (crude rate from counts) | CDC WONDER linked birth and infant death |   |
| imr_state_low_nh | 3.7 | infant deaths per 1,000 live births | crude rate from the death count, normal interval | live births in New Hampshire | 2018 to 2022 | 95% CI 3.2 to 4.2 | 224 infant deaths | none (crude rate from counts) | CDC WONDER linked birth and infant death |   |
| imr_state_ratio | 2.4 | ratio of rates | ratio of Mississippi's crude rate to Massachusetts's, log-normal interval on the two death counts | live births in Mississippi and in Massachusetts | 2018 to 2022 | 95% CI 2.2 to 2.6 | 2,824 infant deaths | none (crude rates from counts) | CDC WONDER linked birth and infant death |   |
| imr_west | 4.6 | infant deaths per 1,000 live births | pooled crude rate over the region's states (13 states), bootstrap interval over those states | live births in the West Census region | 2018 to 2022 | 95% CI 4.3 to 5.3 | 13 states | births in each state | CDC WONDER linked birth and infant death |   |
| imr_white | 4.7 | infant deaths per 1,000 live births | crude rate from state counts, normal interval | live births to white mothers (single race, any ethnicity) in the United States, states with a published count | 2018 and 2019 | 95% CI 4.7 to 4.8 | 26,516 infant deaths | none (crude rate from counts) | CDC WONDER linked birth and infant death |   |
| imr_year_high | 5.7 | infant deaths per 1,000 live births | crude rate from the death count, normal interval | live births in the 50 states and DC | 2018 | 95% CI 5.6 to 5.7 | 21,496 infant deaths | none (crude rate from counts) | CDC WONDER linked birth and infant death |   |
| imr_year_low | 5.4 | infant deaths per 1,000 live births | crude rate from the death count, normal interval | live births in the 50 states and DC | 2020 | 95% CI 5.3 to 5.5 | 19,576 infant deaths | none (crude rate from counts) | CDC WONDER linked birth and infant death |   |
| jurisdictions | 51 | jurisdictions | count of states and DC in the state files | the 50 states and the District of Columbia | infant deaths and live births in 2018 to 2022 |   |   | none (full count) | CDC WONDER linked birth and infant death |   |
| lbd_county_threshold | 250,000 | residents | population size at which WONDER names a county in the linked file (a definition) | counties that CDC WONDER linked birth and infant death names individually | infant deaths and live births in 2018 to 2022 |   |   | none (definition) | CDC WONDER linked birth and infant death |   |
| lbd_withheld_cells | 44 | withheld county-years | count of county-years with 1 to 9 infant deaths, which WONDER withholds | county-year rows of the named counties | infant deaths and live births in 2018 to 2022 |   |   | none (full count) | CDC WONDER linked birth and infant death |   |
| lbd_withheld_max_share | 0.6% | percent of infant deaths | largest share of deaths the withheld county-years could hold: 9 deaths in each withheld county-year, over published deaths plus that limit | infant deaths in the named counties | 2018 to 2022 |   |   | none (an upper limit from the cell count) | CDC WONDER linked birth and infant death |   |
| map_floor_deaths | 20 | infant deaths | fewest infant deaths in the period for which County Health Rankings publishes a county rate, which is also the book's map floor (a definition) | counties on the map | infant deaths and live births in 2015 to 2021 |   |   | none (definition) | County Health Rankings |   |
| map_key_high | 10.6 | infant deaths per 1,000 live births | ninety-fifth percentile of the mapped county rates, the high end of the map key (counties beyond it take the end color) | mapped counties | infant deaths and live births in 2015 to 2021 |   | 1,206 counties | none (each county counts once) | County Health Rankings |   |
| map_key_low | 3.7 | infant deaths per 1,000 live births | fifth percentile of the mapped county rates, the low end of the map key (counties beyond it take the end color) | mapped counties | infant deaths and live births in 2015 to 2021 |   | 1,206 counties | none (each county counts once) | County Health Rankings |   |
| map_ms_top | 25 | counties | count of those counties in Mississippi | mapped counties in the highest tenth of infant mortality rates | infant deaths and live births in 2015 to 2021 |   | 121 counties | none (each county counts once) | County Health Rankings |   |
| map_south_all_share | 49.3% | percent of counties | share of mapped counties in the Census South region (16 states and DC) | all mapped counties | infant deaths and live births in 2015 to 2021 |   | 1,206 counties | none (each county counts once) | County Health Rankings |   |
| map_south_top_share | 85.1% | percent of counties | share of those counties in the Census South region (16 states and DC) | mapped counties in the highest tenth of infant mortality rates | infant deaths and live births in 2015 to 2021 |   | 121 counties | none (each county counts once) | County Health Rankings |   |
| neo_rate | 3.6 | infant deaths per 1,000 live births | pooled neonatal (under 28 days) death rate, bootstrap interval over counties | live births in the 14 counties that publish all five age groups | 2018 to 2022 | 95% CI 3.1 to 4.1 | 14 counties | births in each county | CDC WONDER linked birth and infant death |   |
| neo_ratio_halves | 1.5 | ratio of rates | ratio of pooled neonatal death rates, bootstrap interval over counties with shared draws | live births in the 14 counties that publish all five age groups, the counties above the median deprivation rank over those below it | 2018 to 2022 | 95% CI 1.2 to 1.8 | 14 counties | births in each county | CDC WONDER linked birth and infant death |   |
| neo_share | 68.7% | percent of infant deaths | pooled share that are neonatal deaths (under 28 days), bootstrap interval over counties | infant deaths in the 14 counties that publish all five age groups | 2018 to 2022 | 95% CI 67.4% to 70.2% | 14 counties | infant deaths in each county | CDC WONDER linked birth and infant death |   |
| neonatal_cutoff_days | 28 | days of age | age below which an infant death is neonatal; from this age to the first birthday it is postneonatal (a definition) | infants | all years (a definition) |   |   | none (definition) | CDC WONDER linked birth and infant death |   |
| per_1000 | 1,000 | live births | the denominator of an infant mortality rate (a definition) | all live births | all years (a definition) |   |   | none (definition) | CDC WONDER linked birth and infant death |   |
| post_rate | 1.6 | infant deaths per 1,000 live births | pooled postneonatal (28 to 364 days) death rate, bootstrap interval over counties | live births in the 14 counties that publish all five age groups | 2018 to 2022 | 95% CI 1.4 to 1.9 | 14 counties | births in each county | CDC WONDER linked birth and infant death |   |
| post_ratio_halves | 1.6 | ratio of rates | ratio of pooled postneonatal death rates, bootstrap interval over counties with shared draws | live births in the 14 counties that publish all five age groups, the counties above the median deprivation rank over those below it | 2018 to 2022 | 95% CI 1.3 to 2.1 | 14 counties | births in each county | CDC WONDER linked birth and infant death |   |
| race_births_share_black | 82.5% | percent of live births | share of those births in counties that publish the group in both years | live births to Black mothers in the counties used | 2018 and 2019 |   |   | none (full count) | CDC WONDER linked birth and infant death |   |
| race_births_share_white | 93.9% | percent of live births | share of those births in counties that publish the group in both years | live births to white mothers in the counties used | 2018 and 2019 |   |   | none (full count) | CDC WONDER linked birth and infant death |   |
| race_counties_black | 115 | counties | count of counties | counties used that publish Black infant deaths in both 2018 and 2019 | infant deaths and live births in 2018 and 2019 |   |   | none (full count) | CDC WONDER linked birth and infant death |   |
| race_counties_white | 203 | counties | count of counties | counties used that publish white infant deaths in both 2018 and 2019 | infant deaths and live births in 2018 and 2019 |   |   | none (full count) | CDC WONDER linked birth and infant death |   |
| rank_step | 10 | points of county deprivation rank | the step in the county deprivation rank for which a slope is reported (a definition) | counties | all years (a definition) |   |   | none (definition) | Area Deprivation Index |   |
| south_in_top | 7 | states | highest-rate states in the Census South region (16 states and DC) | the ten states with the highest infant mortality rates | 2018 to 2022 |   | 10 states | none (each state counts once) | CDC WONDER linked birth and infant death |   |
| south_n | 17 | jurisdictions | jurisdictions in the Census South region (16 states and DC) | the 50 states and DC | 2018 to 2022 |   |   | none (full count) | CDC WONDER linked birth and infant death |   |
| south_rest_ratio | 1.2 | ratio of rates | ratio of pooled crude rates (17 South states over 34 other states), stratified bootstrap interval that resamples the states of each group separately | live births in the 50 states and DC, Census South region over the rest | 2018 to 2022 | 95% CI 1.1 to 1.4 | 51 states in the two groups | births in each state | CDC WONDER linked birth and infant death |   |
| states_above | 21 | states | states whose 95% interval lies above the national rate | the 50 states and DC | 2018 to 2022 |   | 51 jurisdictions | none (crude rates from counts) | CDC WONDER linked birth and infant death |   |
| states_below | 16 | states | states whose 95% interval lies below the national rate | the 50 states and DC | 2018 to 2022 |   | 51 jurisdictions | none (crude rates from counts) | CDC WONDER linked birth and infant death |   |
| states_top_n | 10 | states | the number of highest-rate states examined (a definition) | the 50 states and DC | 2018 to 2022 |   |   | none (definition) | CDC WONDER linked birth and infant death |   |
| us_imr | 5.7 | infant deaths per 1,000 live births | births-weighted mean of county rates over all the deciles, bootstrap interval over counties | live births in all counties with a County Health Rankings rate, ranked by the ADI | infant deaths and live births in 2015 to 2021 | 95% CI 5.5 to 5.9 | 1,206 counties | births in each county (County Health Rankings denominator) | County Health Rankings |   |
| white_left_out_share | 0.1% | percent of live births | share of births in state-years whose deaths WONDER withholds, which leave both deaths and births | live births to white mothers (single race, any ethnicity) in the United States | 2018 and 2019 |   |   | none (full counts of births) | CDC WONDER linked birth and infant death |   |
| white_state_high | 6.7 | infant deaths per 1,000 live births | crude rate from the death count, normal interval | live births to white mothers (single race, any ethnicity) in Mississippi | 2018 and 2019 | 95% CI 5.8 to 7.5 | 263 infant deaths | none (crude rate from counts) | CDC WONDER linked birth and infant death |   |
| white_state_low | 3.2 | infant deaths per 1,000 live births | crude rate from the death count, normal interval | live births to white mothers (single race, any ethnicity) in New Jersey | 2018 and 2019 | 95% CI 2.9 to 3.5 | 457 infant deaths | none (crude rate from counts) | CDC WONDER linked birth and infant death |   |
| wonder_suppression_floor | 10 | infant deaths | a cell with fewer deaths than this is withheld (a definition) | cells of the WONDER linked birth and infant death files | infant deaths and live births in 2018 to 2022 |   |   | none (definition) | CDC WONDER linked birth and infant death |   |

## Definitions

- Infant mortality rate: The number of babies who die before their first birthday for every thousand live births.
- Crude rate: Deaths or births divided by the population, with no adjustment for age. A place with older residents has a higher crude death rate even if health at each age is the same.
- Suppression: The withholding of a number because the count behind it is too small to publish without risking the identification of a person. A suppressed value is missing, not zero.
- Area Deprivation Index (ADI): A score that ranks neighborhoods by income, education, employment and housing conditions drawn from the American Community Survey. In this book a higher national percentile means a more deprived neighborhood.
- Ecological inference: Drawing conclusions about individuals from data about places or groups. A pattern across counties need not hold for the people inside them.

## Methods

This chapter compares infant deaths per 1,000 live births across states, counties, groups of mothers by race and ages at death, using CDC WONDER linked birth and infant death records for 2018 to 2022. The rates are crude, with no adjustment for the mother's age or other characteristics. County rates cover only counties that WONDER names (250,000 residents or more) and that publish a death count in every year, so the 21 named counties with a withheld year are left out. Mother's race is reported only for 2018 and 2019, and the county map and the all-county check use County Health Rankings rates for 2015 to 2021, a different source and period.

### What is counted and for which years

Infant mortality is the number of infant deaths (before the first birthday) per 1,000. Rates are crude: nothing is adjusted for the mother's age or other characteristics.

Counts come from CDC WONDER linked birth and infant death records, which link death certificates to birth certificates; a death is assigned to the year of death and divided by live births in the same year. WONDER weights infant deaths, so counts can differ from the number of records, and age groups from a total, by rounding.

State rates pool 2018 to 2022 over the 51 jurisdictions, whose totals have no withheld cell. The 2021 state file reproduces the national rate and the highest and lowest state in the NCHS report for that year (5.44 per 1,000 nationally).

Explains Table 1.

### Which counties are included

County rates cover the counties that WONDER names (250,000 residents or more; WONDER pools every other county in a state into one Unidentified Counties row) and use only counties with a published death count in every year 2018 to 2022: a county-year with 1 to 9 deaths is withheld, a county's published years are its high years, and pooling over them would bias the rate upward. A withheld cell is missing; nothing is derived from totals minus published cells. The 21 named counties with a withheld year are left out whole.

Explains Table 3.

### Why mother's race and age at death cover fewer places

Mother's race comes from the linked file for 2018 and 2019, the only years it is reported, as single race of any ethnicity; a state-year cell of 1 to 9 deaths is withheld and leaves both deaths and births, a group's national rate pools the published state-years, and Black and white rates by state use states that publish both groups in both years. American Indian or Alaska Native mothers are not shown because too few of their state-years are published.

Age at death uses the county file, in which the five age groups (under 1 hour, 1 to 23 hours, 1 to 6 days, 7 to 27 days, 28 to 364 days) are each withheld below 10 deaths; neonatal is the first four groups and postneonatal the last, and only counties that publish all five groups in every year are used.

Explains Table 5, Table 6 and Table 7.

### The county map and the all-county check use County Health Rankings

The county map and the all-county gradient use County Health Rankings 2024, whose infant mortality rate covers 2015 to 2021 and is published only from 20 deaths (window and rule as in the County Health Rankings 2024 technical documentation and the dataset note; as a consistency check, its denominators run about one percent above the 2016 to 2022 births behind its low birthweight measure, as 2015 to 2021 births would).

Explains Table 2 and Table 4.

### How counties are ranked and grouped

A county's deprivation rank is the population-weighted mean of the Area Deprivation Index national ranks of its census tracts, weighted by 2019-2023 tract population; Connecticut tracts are assigned to the eight legacy counties. Quartiles and deciles hold equal numbers of counties. Quartile and decile means are births-weighted means of county rates (pooled rates).

Explains Table 3.

### How uncertainty is measured

State rates carry gamma (fewer than 100 deaths) or normal (100 or more) intervals on the death count, the NCHS convention. Ratios of two crude rates carry log-normal intervals on the two death counts and differences carry normal intervals.

Means, differences, ratios, correlations and slopes across counties carry 95% percentile intervals from a bootstrap that resamples whole counties (2,000 replicates, seed 12345); the state fixed-effects slope resamples whole states, and so does each region's pooled rate, within the states of that region; a group needs at least five states, and the Northeast, with nine, has the roughest interval; the ratio of the South to the other states resamples the states of each group separately (seeds 12345 and 12346).

### Four rankings of the same places

The comparison follows the book's ranking switch rule, fixed on October 28, 2026 before any of these gradients was computed. The main gradient of the chapter is recomputed under four rankings of the same places, with the outcome, period, weights, bins and intervals unchanged: the published ADI; a standardized index that applies the ADI's coefficients to its measures after standardizing each one across tracts; median household income; and the share of adults 25 and older with a bachelor's degree or more. The three alternatives come from ACS tract estimates for 2019 to 2023, the period of the ADI.

All four rankings use the same places, the 1,206 counties that have the outcome and all four rankings, so the ADI panel can differ slightly from the chapter's main figure. Tracts enter only if they have every input of the standardized index, which leaves out tracts without a median rent, home value or mortgage cost. A county takes the population-weighted mean of the ranks of its common-sample tracts, weighted by 2019-2023 tract population, with Connecticut tracts assigned to the eight legacy counties that County Health Rankings uses.

A pattern counts as changed when the worse end switches sides, when the two ends differ under one ranking and not the other, or when the rank correlation of bin order and bin mean falls on the other side of 0.9 and moves by at least 0.1.

Explains Table 8.

## What these data cannot show

The county gradient describes the 240 largest counties, which hold 62.3% of births. WONDER pools the rest by state. The County Health Rankings check reaches smaller counties but leaves out those that do not reach 20 deaths in the period, and the counties just above that floor have the least precise rates on the map. The county rates describe counties that publish a figure, which are not a sample of all counties, and the age-at-death figure covers only 14 of them.

Sources and periods differ, and each is named where it appears. WONDER covers 2018 to 2022 for states and counties and 2018 and 2019 for mother's race, and County Health Rankings covers 2015 to 2021. WONDER's race groups are single races that include Hispanic mothers, so its white group is not the non-Hispanic white group of other sources, and Hispanic origin is not in the files used here. American Indian or Alaska Native mothers are not shown.

The Area Deprivation Index understates deprivation in high-cost cities (Chapter 1.1, What the deprivation index measures (https://vishalsingh.org/zipcode/anatomy-of-deprivation)) and draws on the American Community Survey for 2019 to 2023, while the deaths are from 2018 to 2022. A county's rank averages over tracts that differ within the county, and it describes no family's circumstances. Patterns across counties and states are patterns of places, not of individual infants, a limit known as ecological inference.

Finally, the files carry no cause of death, no gestational age or birthweight, and no information on mothers beyond race, so nothing here identifies a mechanism. The rates are crude period rates, and 2020 to 2022 include the pandemic. Intervals for state and race rates treat deaths as independent and leave out differences between places within each rate, while the county figures resample whole counties.

## References

1. Chen, Alice; Oster, Emily; Williams, Heidi (2016). Why Is Infant Mortality Higher in the United States than in Europe?. American Economic Journal: Economic Policy 8(2): 89-124. https://doi.org/10.1257/pol.20140224
2. Ely, Danielle M.; Driscoll, Anne K. (2023). Infant Mortality in the United States, 2021: Data From the Period Linked Birth/Infant Death File. National Vital Statistics Reports 72(11). https://doi.org/10.15620/cdc:131356
3. Kind, Amy J. H.; Buckingham, William R. (2018). Making Neighborhood-Disadvantage Metrics Accessible: The Neighborhood Atlas. New England Journal of Medicine 378(26): 2456-2458. https://doi.org/10.1056/NEJMp1802313

## Provenance

- Pipeline version: 0.1.0
- Build script: book2/chapters/infant-mortality/build.py (sha256 a02fbe8b137224283f0e9dd52c1a4846216f2697c9f9329890e71495d137658c)
- Built on: 2026-10-07
- Site commit: c2b1ff3309cb4a10833f884da91ac0057a66ad0a

Inputs:

| File | Size (bytes) | SHA-256, first 1 MB | SHA-256, last 1 MB |
| --- | ---: | --- | --- |
| data/analytic/tract_atlas.parquet | 39,659,925 | 29d5e4d943552d54aeb6ba92aadc939d9bd15bdea553c78357dbd516757be627 | 1e10d1ecf722340f23ee2212d4999a90250f85b65fa89b39e32db06d9472e22e |
| data/geo/dim_geo.parquet | 1,180,921 | 29d5e4d943552d54aeb6ba92aadc939d9bd15bdea553c78357dbd516757be627 | e58c8c488df54ab3187f5461f179322afdc49c9bd2315404003fe1272eea32c3 |
| data/raw/acs/deprivation_components_2019_2023/tract.parquet | 13,197,793 | 5545cde2cbe96a8b56e468d65388248a241aa6bd300d3c9e325d08d3f86b11da | 5c8ec454d066988b0f1a01aaa6ed34ebfc05d162e01cf9c6bb01f9198cbaff12 |
| data/staged/adi/adi_atlas_tract2020_2023.parquet | 798,078 | 5e409bc212ff871def780fa27b4cf8b616d9da420b2b9f6455517eddcd8c1a95 | 5e409bc212ff871def780fa27b4cf8b616d9da420b2b9f6455517eddcd8c1a95 |
| data/staged/crosswalks/ct_tract2020_to_old_county.parquet | 15,842 | 961ef39f31596ef65fca2c67d5f2f63e533c5dc9b49745d3bb6437844e4b831d | 961ef39f31596ef65fca2c67d5f2f63e533c5dc9b49745d3bb6437844e4b831d |
| data/staged/health/chr_county_2024.parquet | 8,776,933 | 619795f7ecfec8aff10694b649eaee5e421dbeffbe4be47af4161b51ceefa5cd | c69357f1bcf630407bb9716cb445618838999558c019aaaf77d4681b6bf45c56 |
| data/staged/mortality/wonder_lbd_county_age_2018_2022.parquet | 44,111 | 3c93255ccfddd6ef6a06b7c2aec1f5a66ad62f3a4a1f342a6f19f67a47bcdfe5 | 3c93255ccfddd6ef6a06b7c2aec1f5a66ad62f3a4a1f342a6f19f67a47bcdfe5 |
| data/staged/mortality/wonder_lbd_county_race_2018_2022.parquet | 50,622 | cbe4c336184a835d2d61a2b626c3f4c48455a254ea8ff2a7a6d6eb6fbca700ed | cbe4c336184a835d2d61a2b626c3f4c48455a254ea8ff2a7a6d6eb6fbca700ed |
| data/staged/mortality/wonder_lbd_county_total_2018_2022.parquet | 29,903 | 8e0f2604f2be86d6d86c2d206fa4486b2479411c88a6bb9b1595f3d945a4e5b8 | 8e0f2604f2be86d6d86c2d206fa4486b2479411c88a6bb9b1595f3d945a4e5b8 |
| data/staged/mortality/wonder_lbd_state_race_2018_2022.parquet | 16,461 | 9d17fb750f0d7d3e99d48945e1131654951deddcabfa1d520f0cbf3d717ab710 | 9d17fb750f0d7d3e99d48945e1131654951deddcabfa1d520f0cbf3d717ab710 |
| data/staged/mortality/wonder_lbd_state_total_2018_2022.parquet | 10,241 | 60c10382f900be2cd9d44fbcef5dcc9c559e857ab644097d2f546253484ac42d | 60c10382f900be2cd9d44fbcef5dcc9c559e857ab644097d2f546253484ac42d |
| data/staged/ses/acs5_tract2020_2019_2023.parquet | 7,232,424 | 5fe6f730a35225f14d5d8abfc6b9bce58b52e545f309a7eac74f1dae74d7a31c | 7b4a2475b3852d3f91804716a04566ef8ceadacec2be9fb85f5eb642078efac6 |
| data/staged/svi/svi_tract2020_2022.parquet | 17,740,415 | cb3860d8edc2aaf1f2a02211a58f7a0b80457eb6ad9a37b5ec8abb2d5da9da10 | e371a4a696d5bed05d54ad7c5b6ca2504508805b847ec415c2322be7f66ec2b2 |

Outputs:

| File | SHA-256 |
| --- | --- |
| facts.json | 692b8c0a99be5f6003eee48dd12459e7f65edc244298bf9b5c1f6044d810cfc6 |
| figures/alt_rankings.json | f65e1e27fd5688264db1374dadbdc45f9ec310c1c979dcb715318361b537b178 |
| figures/imr_age_at_death.json | 0b930545b9cbe520881610e4cf9da71c5f80788fcdea214a8a3f4b7404213339 |
| figures/imr_black_white_states.json | 21dbd5660b053825d8e216d901282e3837d8eb99f38a9d78026334a33022a045 |
| figures/imr_by_deprivation.json | ebe5d39824c8881f3b66cf5662e8da7229643c1894815e73029362570c2ff062 |
| figures/imr_by_race.json | 5f4a314c27901d8332ce34ad143de499ff41c7ecac5ee5670973aff1b23568f7 |
| figures/imr_by_state.json | cc13e7b94a90fe6a337d8a5dcb1e521a6828ad908f5e31c95263a70d1f3bcbe0 |
| figures/imr_county_map.json | e2ab821b4602e116709c5d01d6cdf5989caff366a4758d79fbddeaf4573ac622 |
| figures/imr_gradient_checks.json | 069e6db198ea6885e56ada2d237bcdce8f80558b632aedc4677e82e882d665fd |
| meta.json | d8ebd63357145818c218da552fc524ccb8054aeedbcc5b89e9acda517ac8c13c |
| public/data/infant-mortality/imr_county_map.json | 71c917b82e4dbc153c8690b5c6870faa21f8c7a998e3f36fb38057f6cbd963fa |
| refs.json | 9a6c7e9cd1e6cc5e38bee847979896206ef20d8b68fd0af97b3022c93232bdde |

## Machine formats

- Report (HTML): https://vishalsingh.org/zipcode/infant-mortality/report
- Report (Markdown): https://vishalsingh.org/zipcode/infant-mortality/report.md
- Every estimate (JSON): https://vishalsingh.org/zipcode/report/infant-mortality/facts.json
- Table 1, figure "imr_by_state" (JSON): https://vishalsingh.org/zipcode/report/infant-mortality/figures/imr_by_state.json
- Table 2, figure "imr_county_map" (JSON): https://vishalsingh.org/zipcode/report/infant-mortality/figures/imr_county_map.json
- Table 3, figure "imr_by_deprivation" (JSON): https://vishalsingh.org/zipcode/report/infant-mortality/figures/imr_by_deprivation.json
- Table 4, figure "imr_gradient_checks" (JSON): https://vishalsingh.org/zipcode/report/infant-mortality/figures/imr_gradient_checks.json
- Table 5, figure "imr_by_race" (JSON): https://vishalsingh.org/zipcode/report/infant-mortality/figures/imr_by_race.json
- Table 6, figure "imr_black_white_states" (JSON): https://vishalsingh.org/zipcode/report/infant-mortality/figures/imr_black_white_states.json
- Table 7, figure "imr_age_at_death" (JSON): https://vishalsingh.org/zipcode/report/infant-mortality/figures/imr_age_at_death.json
- Table 8, figure "alt_rankings" (JSON): https://vishalsingh.org/zipcode/report/infant-mortality/figures/alt_rankings.json
- Payload of "imr_county_map" (JSON): https://vishalsingh.org/zipcode/data/infant-mortality/imr_county_map.json
- Index of every report: https://vishalsingh.org/zipcode/llms.txt; all reports in one file: https://vishalsingh.org/zipcode/llms-full.txt

---

# Smoking in pregnancy: report

Chapter 2.3 of ZIP Code Destiny, by Vishal Singh (NYU Stern). Generated from the chapter's data files; the reading version is at https://vishalsingh.org/zipcode/smoking-in-pregnancy. Table n of this report is figure n of the chapter.

## Identification

- Title: Smoking in pregnancy
- Label: Before the first breath
- Chapter: 2.3 (Part II, Birth; An American life)
- Question: How does smoking during pregnancy vary with county income within each state, and is the slope the same everywhere?
- Dek: In large counties with a usable count, the share of births to mothers who reported smoking fell from 5.4% in 2018 to 1.9% in 2024 and stayed higher where county income was lower.
- Grain: counties
- Status: published
- Published: 2026-10-09
- Updated: not revised
- Data sources:
  - CDC WONDER natality. CDC WONDER natality, county by tobacco use during pregnancy and county totals, births in 2018 to 2024 (consolidated pulls); 2024 is final; years: 2018-2024; weights: none (full counts of births); shares pooled over counties, births-weighted or unweighted as stated
  - County Health Rankings. County Health Rankings 2024 release, median household income (Census Small Area Income and Poverty Estimates, 2022); years: 2024 release; weights: none (published county values)
  - Area Deprivation Index. Neighborhood Atlas Area Deprivation Index 2023 (v4.0.1), tract national rank, averaged to counties with 2019-2023 tract population; used as a check on the income measure; years: 2023 release (version 4.0.1; built on ACS 2019-2023); weights: 2019-2023 tract population
  - Published estimates from cited studies. Findings of two studies of state policy and maternal smoking: the abstract of Lien and Evans (2005) and Table 2 of Markowitz and coauthors (2017); years: as reported in each study; weights: not applicable
- Canonical URL: https://vishalsingh.org/zipcode/smoking-in-pregnancy
- Report: https://vishalsingh.org/zipcode/smoking-in-pregnancy/report
- Markdown: https://vishalsingh.org/zipcode/smoking-in-pregnancy/report.md
- BibTeX key: singh2026smokinginpregnancy

```bibtex
@misc{singh2026smokinginpregnancy,
  author = {Singh, Vishal},
  title = {Smoking in pregnancy},
  year = {2026},
  url = {https://vishalsingh.org/zipcode/smoking-in-pregnancy},
  note = {In ZIP Code Destiny. Published 9 October 2026.}
}
```

## Principal estimates

The facts the dek and the opening prose cite, in order of appearance.

| Estimand | Population | Period | Value | Interval | n | Weights | Source | Fact id |
| --- | --- | --- | ---: | --- | ---: | --- | --- | --- |
| pooled share of births to mothers who reported smoking; bootstrap interval over counties | live births with a known tobacco answer in the large counties that CDC WONDER names in every year, leaving out California and the counties with a withheld smoking count | 2018 | 5.4% | 95% CI 4.9% to 5.9% | 530 counties | none (full counts of births); interval from a bootstrap over counties | CDC WONDER natality | share_all_2018 |
| pooled share of births to mothers who reported smoking; bootstrap interval over counties | live births with a known tobacco answer in the large counties that CDC WONDER names in every year, leaving out California and the counties with a withheld smoking count | 2024 | 1.9% | 95% CI 1.7% to 2.1% | 530 counties | none (full counts of births); interval from a bootstrap over counties | CDC WONDER natality | share_all_2024 |
| count of complete counties | named counties outside California with a published smoking count in every year | births in 2018 to 2024 | 530 |   |   | none (full count) | CDC WONDER natality | n_complete |
| fall in the pooled share from 2018 to 2024; bootstrap interval over counties | live births with a known tobacco answer in the large counties that CDC WONDER names in every year, leaving out California and the counties with a withheld smoking count | 2018 to 2024 | 64.6% | 95% CI 63.5% to 65.7% | 530 counties | none (full counts of births); interval from a bootstrap over counties | CDC WONDER natality | decline_all |
| difference of two pooled shares; shared bootstrap draws over counties | live births with a known tobacco answer in the large counties that CDC WONDER names in every year, leaving out California and the counties with a withheld smoking count; lowest-income third minus highest-income third of counties | 2018 | 4.6 | 95% CI 3.6 to 5.6 | 530 counties | none (full counts of births); interval from a bootstrap over counties | CDC WONDER natality | gap_2018 |
| state fixed-effects slope of the smoking share (percent) on log2 county income, weighted by births with a known tobacco answer; bootstrap interval over states | live births with a known tobacco answer in the large counties that CDC WONDER names in every year, leaving out California and the counties with a withheld smoking count; slope within states | births in 2018 to 2024 | 3.9 | 95% CI 2.7 to 5.5 | 45 states | births with a known tobacco answer; interval from a bootstrap over states | CDC WONDER natality | slope_within_w |
| state fixed-effects slope of the smoking share (percent) on log2 county income, each county counts once; bootstrap interval over states | live births with a known tobacco answer in the large counties that CDC WONDER names in every year, leaving out California and the counties with a withheld smoking count; slope within states | births in 2018 to 2024 | 6.6 | 95% CI 5.1 to 8.2 | 45 states | none (each county counts once); interval from a bootstrap over states | CDC WONDER natality | slope_within_u |
| population size at which WONDER natality names a county (a definition) | the counties and cells of CDC WONDER natality | all years (a definition) | 100,000 |   |   | none (definition) | CDC WONDER natality | wonder_threshold |
| count of counties | counties that WONDER natality names in every year from 2018 to 2024 and that have births | births in 2018 to 2024 | 578 |   |   | none (full count) | CDC WONDER natality | n_named |
| share of births in the named counties | live births in the 50 states and DC | births in 2018 to 2024 | 79.5% |   | 578 counties | none (full count) | CDC WONDER natality | named_share |
| share of births in the complete counties | live births in the 50 states and DC | births in 2018 to 2024 | 66.9% |   | 530 counties | none (full count) | CDC WONDER natality | complete_share |

## Findings

Among births in the 530 large counties that CDC WONDER names and that publish a smoking count in every year, the share to mothers who reported smoking fell from 5.4% (95% CI 4.9% to 5.9%) in 2018 to 1.9% (95% CI 1.7% to 2.1%) in 2024. That is a fall of 64.6% (95% CI 63.5% to 65.7%) of the 2018 share. The share was higher in counties with lower median household income in every year, and in 2018 the lowest-income third of counties exceeded the highest-income third by 4.6 (95% CI 3.6 to 5.6) points. Across counties within a state, each doubling of county median household income goes with a smoking share 3.9 (95% CI 2.7 to 5.5) points lower when counties are weighted by their births, for births pooled over 2018 to 2024. When each county counts once it goes with a share 6.6 (95% CI 5.1 to 8.2) points lower.

WONDER names a county only if it has 100,000 residents or more, and every county number in this chapter describes a county of that size. The 578 counties named in every year hold 79.5% of the births in the United States from 2018 to 2024, and the 530 counties used here hold 66.9%. California, which does not collect the item routinely, is left out. Requiring a published smoking count in every year selects on the outcome and leaves out affluent counties, as the next section shows. The smaller counties are pooled by WONDER and cannot be placed on an income scale. Everything below describes counties and groups of births, not mothers. Smoking among adults is the subject of Chapter 7.1, Health behaviors and the supply of care (https://vishalsingh.org/zipcode/health-behaviors-and-the-supply-of-care) for neighborhoods and of Chapter 7.2, Health behaviors, person by person (https://vishalsingh.org/zipcode/health-behaviors-person-by-person) for people. The weight of the babies born to these mothers is the subject of Chapter 2.1, Birthweight and preterm birth (https://vishalsingh.org/zipcode/birthweight-and-preterm-birth).

### What the tobacco item counts

The birth certificate records whether the mother used tobacco during pregnancy, and WONDER reports the answer as Yes, No, Unknown or Not Stated, or Not Reported. This chapter writes smoking for a Yes answer, although the item is named tobacco use and the documentation that comes with the data does not say which products count. A county's smoking share is its Yes births over its births with a known answer, Yes plus No, so births with an Unknown or Not Stated answer are left out of the denominator. The Unknown cells that WONDER publishes hold 0.4% of births. WONDER withholds 1,794 of the 3,710 Unknown cells because each holds fewer than 10 births, so together they could hold at most 0.1%, the most that the left-out births could add.

Not Reported, the code for reporting areas that did not collect the item, is a published zero in all 4,046 county-years. The counties used record 17.1 million births with a known answer from 2018 to 2024, and 632,733 of them are to mothers who reported smoking.

California is left out. WONDER's documentation says that every reporting area except California routinely collects the item. The California counties report a share of 1.2% for 2018, against 5.4% in the counties kept, and below the share in every state kept. The item in California therefore cannot be compared with the item elsewhere, and the 35 California counties are in no estimate.

The smoking count is withheld whenever a county has fewer than 10 smoking births in a year, which happens in low-smoking counties. A withheld cell is missing, never zero, and nothing here is derived from a county total minus the published cells. The named counties have 68 withheld Yes cells, 33 of them in California. The other 35 sit in 13 counties, and those counties are left out of every estimate. Pooling a county over only the years in which its count is published would drop its lowest years and raise its share. A county therefore stays only if its count is published in every year from 2018 to 2024.

The withheld cells could hold at most 315 births, or 0.05% of the smoking births in the counties kept, so they change who is in the sample and not the totals. Virginia lost 4 of its 15 large counties (Arlington, Fairfax, Loudoun and Alexandria), and 9 more counties in Florida, Georgia, Illinois, New York, Tennessee and Texas are out.

The rule selects on the outcome, and the counties it removes are affluent. Of the 13 counties, 11 are above the cut for the highest-income third of the counties kept, and their median income is $109,358 against $72,077 for the counties kept. Suppose each of the 29 withheld cells in the high-income counties held one fewer birth than the withholding floor, a limit and not a value for any cell. Those cells could then add at most 0.17% to the smoking births of the highest-income third.

Adding back the 9 counties whose 2018 count is published moves the weighted within-state slope for 2018 from 4.9 to 4.8 points lower. The same check for 2024 adds only 1 county, because most of the counties left out have a withheld 2024 cell, so it says little about the later years.

### Smoking fell in every income group

Ranking the 530 counties on median household income and splitting them into equal thirds gives three groups, with the lowest-income third up to $64,868 and the middle third up to $80,017. The smoking share in the lowest-income third went from 8.2% (95% CI 7.3% to 9.2%) in 2018 to 3.3% (95% CI 2.9% to 3.7%) in 2024. In the middle third it went from 5.5% (95% CI 4.4% to 6.7%) to 1.9% (95% CI 1.5% to 2.3%). In the highest-income third it went from 3.6% (95% CI 3.2% to 4.0%) to 1.1% (95% CI 1.0% to 1.3%) (Table 1). All of the lines fall in every year and the thirds keep their order, with separate intervals in each year.

Table 1. Smoking in pregnancy fell from 8.2% to 3.3% of births in the lowest-income third of large counties and from 3.6% to 1.1% in the highest-income third

Smoking fell faster in proportion where income is higher. The lowest-income third lost 60.0% (95% CI 57.7% to 62.3%) of its 2018 share and the highest-income third lost 68.8% (95% CI 67.0% to 70.4%), and the two intervals do not overlap.

The gap in points between the lowest-income and highest-income thirds narrowed from 4.6 (95% CI 3.6 to 5.6) to 2.1 (95% CI 1.7 to 2.6), a narrowing of 2.4 (95% CI 1.8 to 3.1) points. The ratio of their shares rose from 2.3 (95% CI 1.9 to 2.7) to 2.9 (95% CI 2.4 to 3.5), a rise of 0.6 (95% CI 0.4 to 0.9). A gap in points shrinks when the level falls even if the proportional difference stays the same. Here the proportional difference grew, so the gradient flattened in one scale and steepened in the other, a pattern that a later section follows year by year.

### Where smoking in pregnancy is most common

Pooled over 2018 to 2024, the share of births to mothers who reported smoking runs from 0.8% in the county at the 5th percentile to 13.2% at the 95th (Table 2). The counties in the top tenth by share are concentrated in Pennsylvania, Ohio, Michigan, New York, Indiana and Missouri. The counties in the bottom tenth are concentrated in Texas, New Jersey, New York, Florida and Georgia, and New York has counties in both groups.

Table 2. The share of births to mothers who smoked is 0.8% in the county at the 5th percentile and 13.2% at the 95th percentile of large counties, and the highest shares cluster in Pennsylvania, Ohio, Michigan, New York, Indiana and Missouri

A county's shade is its own count, so a county with few smoking births has a less stable shade than a large one, and the tooltip gives its interval. The gray counties are not a random draw of counties. Of the 2,613 gray counties, 2,563 are below WONDER's size threshold. The California counties are gray for the reason above, and the counties with a withheld smoking count are gray because their smoking births fell below the withholding floor in some year.

### Smoking is lower where county income is higher

Median household income places each county on one scale. This chapter takes it from County Health Rankings, which reports a Census estimate for 2022, so its year differs from the births, which are from 2018 to 2024. The lowest-income tenth of counties has a median income of $54,059 and the highest-income tenth $107,649, a ratio of 2.0. The smoking share is 5.7% (95% CI 4.2% to 8.0%) in the lowest-income tenth and 1.7% (95% CI 1.4% to 2.1%) in the highest, and the intervals do not overlap (Table 3). The decline is not smooth from tenth to tenth, and counties at lower incomes differ more from one another than counties at higher incomes.

Table 3. Large counties with higher median household income have a lower share of births to mothers who smoked, from 5.7% in the lowest-income tenth to 1.7% in the highest

A second measure checks the choice of income. The Area Deprivation Index orders neighborhoods by deprivation, and a county's rank here is the average of its tracts' ranks, weighted by tract population.[1] The check is not independent, since income is among the index's inputs. Within a state, the smoking share is higher by 1.3 (95% CI 0.9 to 1.6) points for each ten-point step in a county's rank. The correlation between a county's rank and the log of its median income is −0.79 (95% CI −0.90 to −0.67), so the two measures place counties alike. That is agreement and not independent confirmation, and the chapter uses income.

### The slope within states is close to the slope across all counties

A slope across all counties can come from comparing states that differ in both income and smoking. Table 4 separates the two. The first slope is a least-squares line through all counties. The second gives each state its own intercept, a fixed-effects slope that compares counties only inside a state, and the third compares state averages. The slopes are in points of the smoking share for a doubling of county median household income.

Table 4. Within a state, a doubling of county income goes with a smoking share 3.9 points lower when counties are weighted by births, and 6.6 points lower when each county counts once

When counties are weighted by their births with a known answer, the slope across all counties is 4.1 (95% CI 2.7 to 5.9) points, the slope within states 3.9 (95% CI 2.7 to 5.5) and the slope between states 4.7 (95% CI 1.8 to 8.3). The within-state slope differs from the slope across the counties of the same 45 states, those with two or more counties, by +0.3 (95% CI −0.7 to 1.1) points, an interval that includes zero.

When each county counts once, the three are 6.2 (95% CI 4.6 to 8.0), 6.6 (95% CI 5.1 to 8.2) and 5.3 (95% CI 2.2 to 8.7). The within-state slope then differs from that comparison slope by −0.4 (95% CI −1.3 to 0.5) points, again an interval that includes zero. The between-state slope is the least precise because it rests on 49 state averages. The gap between higher-income and lower-income counties therefore appears inside states about as it appears across them.

The weights matter. The weighted within-state slope is shallower than the unweighted one by 2.8 (95% CI 1.3 to 3.9) points. The counties at or above the median in births hold 82.2% of the births. Among them the within-state slope is 3.7 (95% CI 2.6 to 5.1) points, against 8.3 (95% CI 6.3 to 10.2) among the counties below the median, a difference of 4.6 (95% CI 2.5 to 6.6) points. A weighted slope leans on large counties, where the gradient is shallower in points.

Expressed as the proportional difference in the smoking share for a 10% higher county income, the smoking share is 12.0% (95% CI 8.2% to 17.0%) lower within a state when counties are weighted and 16.7% (95% CI 14.4% to 19.0%) lower when each county counts once. The scope figure shows both weightings, the figure of state slopes has a view for each, and the other state figures are weighted by births.

### One slope for each state with enough counties

A state gets a slope of its own if it has at least 8 complete counties. That leaves 26 states and 449 counties, and the other states have too few large counties to fit a line (Table 5). Across these states the common within-state slope is 3.8 (95% CI 2.6 to 5.6) points per doubling.

Every state's estimated slope is below zero, 26 of 26. The interval lies wholly below zero in 10 of them (Colorado, Georgia, Illinois, Indiana, Massachusetts, Missouri, North Carolina, Ohio, Oregon and Virginia), and no interval lies wholly above zero (0 states). With each county counting once, the estimated slope is below zero in 26 states and the interval lies wholly below zero in 12. No interval lies wholly above zero (0 states), as the second view of the figure shows.

Table 5. In all 26 states with a slope of their own the estimated slope is below zero, and the 95% interval lies below zero in 10 when counties are weighted by births

The intervals are wide because a state's line rests on a handful of counties that span less than a doubling of income. The median state's highest-income county has 0.9 doublings more income than its lowest-income county, so a slope per doubling extends past what most states' own counties cover. The figure orders its rows by the estimate only to make it readable, and the order is not a ranking. A state's dot moves with the counties it happens to hold, the intervals overlap widely, and a state is said to differ from another only where the intervals separate. Virginia has 4 of its 15 large counties left out for a withheld count. Those counties are among the highest in income of all the large counties, so its slope describes the counties that remain.

### Whether the slope is the same everywhere

Whether the within-state slope is the same in every state depends on the weights and on the scale. Each of the 26 states with enough large counties for a slope of its own has an estimated slope below zero. The intervals are wide, though, and only 10 lie wholly below zero when counties are weighted by births. Only one of four tests of equal slopes rejects, the test in points when each county counts once (p = 0.009). It does not reject in points when counties are weighted by births (p = 0.108) or in proportion to a state's own smoking share (p = 0.664 weighted and 0.343 unweighted).

A visual impression of spread in a figure of intervals does not test equality, so the chapter tests it. The test compares the spread of the state slopes with the spread their own intervals allow, which is Cochran's Q. With a handful of counties per state the usual reference distribution for Q rejects too often. Each p-value here is therefore the share of bootstrap draws in which Q is at least as large as the observed one, with the draws generated on the assumption that every state has the same slope. The test is run in points and in proportion, with counties weighted by births and with each county counting once.

In points the answer depends on the weights. With each county counting once, p = 0.009 and the estimated standard deviation of the true state slopes is 3.0 (95% CI 1.6 to 3.6) points, with 70% of the variation in the slopes beyond what sampling allows. With counties weighted by births, p = 0.108 and the standard deviation is 1.9 (95% CI 0.0 to 2.7) points, an interval that reaches zero, with 52% of the variation beyond sampling. The standard deviation is the DerSimonian and Laird estimate used in meta-analysis.[2]

The steeper slopes also belong to states with more smoking. Across the states, the correlation between a state's pooled smoking share and its slope is −0.62 (95% CI −0.83 to −0.20), weighted by births (Table 6).

Table 6. States with a higher share of births to mothers who smoked have steeper slopes in points, a correlation of −0.62, with no detectable relation in proportion, where the correlation is −0.19

The same proportional gradient gives a larger difference in points in a state with more smoking, so a steeper slope in points is expected there even if the slope in proportion is the same. The second view of the figure tests that. Written as the proportional difference in the smoking share for a 10% higher county income, the smoking share is 11.9% (95% CI 8.0% to 17.5%) lower. The tests of equal slopes give p = 0.664 weighted and 0.343 unweighted. The correlation with the state's smoking level is −0.19 (95% CI −0.56 to 0.37), an interval that includes zero.

Of the 26 estimated slopes, 25 are below zero in proportion and 8 have an interval wholly below zero. With each county counting once the counts are 26 and 12.

The spread of the true proportional slopes is estimated at 2.1% (95% CI 0.0% to 6.5%) weighted and 2.7% (95% CI 0.0% to 7.0%) unweighted, in the units of the proportional slope. The data are therefore compatible with state slopes that differ moderately around the common value and do not show that they do. The proportional tests have little power: the upper limit of the interval for the unweighted spread sits beside a within-state slope of 16.7% lower. A rejection in points with no rejection in proportion does not show that the scales differ in heterogeneity.

### How the minimum number of counties changes the results

The floor of 8 counties is a choice. A lower floor adds states with a few counties, whose lines are unstable, and a higher floor drops states (Table 7). Moving the floor from 3 to 10 counties changes the number of states from 41 to 23. The number of states with an interval wholly below zero stays between 9 and 11, and at no floor does an interval lie wholly above zero (0 states).

Table 7. Raising the minimum number of counties from 3 to 10 leaves 9 to 11 states with an interval below zero and none above it, and the test of equal proportional slopes never rejects

The tests depend on the weights more than on the floor. In proportion the smallest p-value across all floors and both weightings is 0.319. In points, with counties weighted by births, p runs from 0.108 to 0.165. With each county counting once p is at most 0.009 at every floor above 3 counties. It is 0.264 at a floor of 3 counties, where the lines of the smallest states add noise. The conclusion that the state slopes differ in points rests on counting each county once.

### The slope in points shrank and the slope in proportion grew

The within-state slope can be measured year by year for the same counties (Table 8). In 2018 a doubling of county income went with a smoking share 4.9 (95% CI 3.5 to 7.0) points lower, and in 2024 with a share 2.4 (95% CI 1.7 to 3.4) points lower, so the slope became shallower by 2.5 (95% CI 1.7 to 3.7) points.

In proportion the slope went the other way. A 10% higher county income went with a smoking share 10.1% (95% CI 6.3% to 15.1%) lower in 2018 and 14.7% (95% CI 11.6% to 19.3%) lower in 2024, a steepening of 4.5 (95% CI 3.3 to 5.8) points.

Table 8. The within-state slope in points shrank from 4.9 to 2.4 points per doubling of county income from 2018 to 2024 while the slope in proportion grew from 10.1% to 14.7%

### What moves smoking in pregnancy

Two studies of state policy, each reported under its own design, bear on what changes smoking in pregnancy. Neither measures the county gradient in this chapter.

> Causal evidence: Cigarette taxes and smoking in pregnancy (Matched comparison states around large tax increases, with an instrumental variable for birth weight)
>
> Lien and Evans studied 4 states that adopted large cigarette tax increases and chose control groups of states by a statistical matching procedure. Smoking participation among pregnant women fell in the states with the increases relative to their matched controls.[3] The paper also uses the tax increases as an instrument for smoking in equations for birth weight, pooled across the experiments, and estimates that smoking in pregnancy raises the chance of a low-birthweight birth to 2 times its level without smoking. These statements come from the paper's abstract. The study estimates the effect of higher cigarette taxes on smoking and does not speak to why smoking differs between higher-income and lower-income counties.

> Causal evidence: State income support and maternal smoking (Difference-in-differences)
>
> Markowitz and colleagues compared births in states with and without state Earned Income Tax Credit laws from 1995 to 2013, in a multi-state, multi-year difference-in-differences analysis with state and year-by-quarter fixed effects, standard errors clustered by state and a credit measured in the year before conception.[4] For tobacco use during pregnancy, a low state credit with no refund went with a probability of smoking 1.6 points lower, a high credit with no refund with a probability 0.9 points higher that is significant only at the weakest of the usual levels, and credits with a refund with no significant change. The authors find little difference in maternal health behaviors associated with the credits and write that the policies do not appear to reduce smoking. They find small improvements in birth weight and gestation in states with credits. The estimates are for mothers with a high-school education or less and for the laws in force from 1995 to 2013.

### What these data cannot show

The comparison is between counties. A county with higher median household income has a lower smoking share, and that says nothing about the income of the mothers who smoked, which these data do not hold. The pattern is ecological and need not hold for the mothers inside the counties. The county tables staged for this book carry one stratifier each, so mother's education cannot be crossed with the tobacco item.

The item is reported at birth. The fall in the share from 5.4% to 1.9% can reflect less smoking, a change in how or whether the item is recorded, or a change in who gives birth, and these data cannot separate them. The item is named tobacco use and may include products other than cigarettes.

The counties are the 530 large counties that remain after the rules above, and they hold 66.9% of the births in the United States. California is out because its item is not collected routinely, and 13 counties are out because their smoking count is withheld in some year, a rule that selects on the outcome. Smaller counties are not in these data, so the chapter says nothing about them. The income measure is an estimate for 2022 and the births are from 2018 to 2024, so a county's income stands in for its income in earlier years. The Area Deprivation Index describes neighborhoods in 2019 to 2023.

State slopes rest on 8 counties or more, and the states with a slope hold 449 counties in all. The intervals use a t distribution and standard errors that allow for unequal variance, and they can be too narrow where one county dominates the line. The tests have little power to detect moderate differences, their p-values depend on how counties are weighted, and a test that finds no difference does not show equality. A slope per doubling of income is a statement about a difference larger than most states' own counties span. None of this identifies why smoking is less common where incomes are higher, and the studies in the box above are the only evidence here on what changes it.

## Tables

### Table 1. Smoking in pregnancy fell from 8.2% to 3.3% of births in the lowest-income third of large counties and from 3.6% to 1.1% in the highest-income third

Share of births to mothers who reported smoking, among births with a known tobacco answer, by third of the 530 large counties ranked on median household income, 2018 to 2024, with 95% intervals from a bootstrap over counties

| conf_high | conf_low | n | series | share | year |
| ---: | ---: | ---: | --- | ---: | ---: |
| 9.17 | 7.31 | 603,928 | Lowest-income third | 8.18 | 2018 |
| 8.45 | 6.75 | 599,484 | Lowest-income third | 7.56 | 2019 |
| 7.99 | 6.34 | 580,344 | Lowest-income third | 7.14 | 2020 |
| 6.75 | 5.32 | 579,539 | Lowest-income third | 6.01 | 2021 |
| 5.43 | 4.26 | 579,736 | Lowest-income third | 4.81 | 2022 |
| 4.49 | 3.51 | 567,037 | Lowest-income third | 3.97 | 2023 |
| 3.71 | 2.89 | 567,610 | Lowest-income third | 3.27 | 2024 |
| 6.73 | 4.42 | 938,985 | Middle third | 5.47 | 2018 |
| 6.17 | 4.04 | 924,476 | Middle third | 5 | 2019 |
| 5.63 | 3.70 | 890,673 | Middle third | 4.58 | 2020 |
| 4.59 | 2.99 | 899,921 | Middle third | 3.72 | 2021 |
| 3.66 | 2.37 | 906,572 | Middle third | 2.96 | 2022 |
| 2.91 | 1.89 | 890,288 | Middle third | 2.36 | 2023 |
| 2.32 | 1.52 | 894,429 | Middle third | 1.89 | 2024 |
| 4.04 | 3.22 | 968,024 | Highest-income third | 3.61 | 2018 |
| 3.60 | 2.85 | 959,133 | Highest-income third | 3.21 | 2019 |
| 3.23 | 2.56 | 932,668 | Highest-income third | 2.88 | 2020 |
| 2.58 | 2.05 | 960,611 | Highest-income third | 2.30 | 2021 |
| 1.98 | 1.57 | 962,137 | Highest-income third | 1.76 | 2022 |
| 1.65 | 1.29 | 946,374 | Highest-income third | 1.46 | 2023 |
| 1.28 | 0.990 | 961,896 | Highest-income third | 1.13 | 2024 |
| 5.91 | 4.88 | 2,510,937 | All large counties | 5.40 | 2018 |
| 5.41 | 4.45 | 2,483,093 | All large counties | 4.93 | 2019 |
| 5 | 4.09 | 2,403,685 | All large counties | 4.54 | 2020 |
| 4.08 | 3.34 | 2,440,071 | All large counties | 3.71 | 2021 |
| 3.23 | 2.63 | 2,448,445 | All large counties | 2.93 | 2022 |
| 2.64 | 2.14 | 2,403,699 | All large counties | 2.38 | 2023 |
| 2.11 | 1.72 | 2,423,935 | All large counties | 1.91 | 2024 |

Note: Large counties are the counties that CDC WONDER names individually: those with 100,000 residents or more. WONDER pools the births of every smaller county in a state into one row, so smaller counties are not in these data. Of the 578 large counties named in every year from 2018 to 2024, California (35) and 13 counties with a withheld smoking count are left out, which leaves 530.

Note: A county's smoking share is the births to mothers who reported smoking over the births with a known answer, Yes plus No. Births with an Unknown or Not Stated answer are left out of the denominator.

Note: County median household income is the County Health Rankings 2024 release, a Census Small Area Income and Poverty Estimates value for 2022. Its year differs from the births, which are from 2018 to 2024. Thirds hold equal numbers of counties.

Note: Intervals are 95% intervals from a bootstrap over counties (replicates and seed in the methods note). Births are full counts, so an interval shows how much a pooled share depends on which counties are in the group.

Source: CDC WONDER natality, tobacco use during pregnancy, births in 2018-2024; full counts of births; counties ranked on County Health Rankings 2024 median household income (Census SAIPE 2022); intervals from a bootstrap over counties

Data: https://vishalsingh.org/zipcode/report/smoking-in-pregnancy/figures/smoking_by_income_third.json

### Table 2. The share of births to mothers who smoked is 0.8% in the county at the 5th percentile and 13.2% at the 95th percentile of large counties, and the highest shares cluster in Pennsylvania, Ohio, Michigan, New York, Indiana and Missouri

Share of births to mothers who reported smoking among births with a known tobacco answer, 2018 to 2024, 530 large counties (counties of 100,000 residents or more)

No inline rows: the values of Table 2 are in https://vishalsingh.org/zipcode/data/smoking-in-pregnancy/smoking_county_map.json.

Note: Large counties are the counties that CDC WONDER names individually: those with 100,000 residents or more. WONDER pools the births of every smaller county in a state into one row, so smaller counties are not in these data. Of the 578 large counties named in every year from 2018 to 2024, California (35) and 13 counties with a withheld smoking count are left out, which leaves 530.

Note: A county's smoking share is the births to mothers who reported smoking over the births with a known answer, Yes plus No. Births with an Unknown or Not Stated answer are left out of the denominator.

Note: A county's interval is a Wilson score interval on its births, which treats births as independent; it describes the county's own count and not the spread across counties.

Note: Connecticut is shown as its eight legacy counties, the coding of the source. Alaska is drawn. Puerto Rico is not in these data and is not drawn.

Note: California's counties are gray because the source states that California does not routinely collect the item; its reported share for 2018 is 1.2%, against 5.4% in the counties kept, and lower than the share in any state kept.

Source: CDC WONDER natality, tobacco use during pregnancy, births in 2018-2024; full counts of births; county shares are not weighted

Data: https://vishalsingh.org/zipcode/report/smoking-in-pregnancy/figures/smoking_county_map.json

### Table 3. Large counties with higher median household income have a lower share of births to mothers who smoked, from 5.7% in the lowest-income tenth to 1.7% in the highest

Share of births to mothers who reported smoking, 2018 to 2024, against county median household income (SAIPE 2022, log scale), one dot per county, with pooled shares for tenths of counties

| births | county | income | share |
| ---: | --- | ---: | ---: |
| 17160 | Baldwin County, AL | 71704 | 5.15 |
| 9085 | Calhoun County, AL | 52772 | 11.61 |
| 8291 | Etowah County, AL | 55487 | 9.20 |
| 9150 | Houston County, AL | 56842 | 7.64 |
| 56274 | Jefferson County, AL | 61996 | 2.81 |
| 13383 | Lee County, AL | 59637 | 2.79 |
| 30871 | Madison County, AL | 80426 | 4.14 |
| 36143 | Mobile County, AL | 54315 | 5.40 |
| 20991 | Montgomery County, AL | 56862 | 3.02 |
| 10183 | Morgan County, AL | 62260 | 9.38 |
| 16052 | Shelby County, AL | 98263 | 2.70 |
| 17613 | Tuscaloosa County, AL | 60875 | 3.37 |
| 25397 | Anchorage Borough, AK | 99162 | 5.15 |
| 8855 | Cochise County, AZ | 55254 | 6.15 |
| 9343 | Coconino County, AZ | 64016 | 2.24 |
| 354234 | Maricopa County, AZ | 83668 | 2.41 |
| 12222 | Mohave County, AZ | 54842 | 10.94 |
| 8894 | Navajo County, AZ | 49948 | 5.73 |
| 70533 | Pima County, AZ | 64007 | 4.14 |
| 34615 | Pinal County, AZ | 76232 | 4.17 |
| 12334 | Yavapai County, AZ | 64187 | 8.72 |
| 20395 | Yuma County, AZ | 54209 | 1.73 |
| 26446 | Benton County, AR | 86374 | 4.39 |
| 10357 | Faulkner County, AR | 63429 | 7.01 |
| 34300 | Pulaski County, AR | 57057 | 4.08 |
| 9524 | Saline County, AR | 70330 | 6.72 |
| 11184 | Sebastian County, AR | 52679 | 13.26 |
| 22486 | Washington County, AR | 62100 | 4.57 |
| 43713 | Adams County, CO | 89849 | 3.90 |
| 52605 | Arapahoe County, CO | 93525 | 3.38 |
| 17177 | Boulder County, CO | 96584 | 3.05 |
| 63854 | Denver County, CO | 87619 | 2.94 |
| 25857 | Douglas County, CO | 140768 | 1.44 |
| 63657 | El Paso County, CO | 82248 | 2.95 |
| 38524 | Jefferson County, CO | 102413 | 3.57 |
| 21451 | Larimer County, CO | 88338 | 4.69 |
| 10630 | Mesa County, CO | 69578 | 7.97 |
| 11938 | Pueblo County, CO | 58314 | 8.55 |
| 32098 | Weld County, CO | 90770 | 5.16 |
| 69264 | Fairfield County, CT | 100703 | 0.800 |
| 62361 | Hartford County, CT | 79958 | 2.61 |
| 9943 | Litchfield County, CT | 84689 | 5 |
| 9096 | Middlesex County, CT | 94735 | 2.86 |
| 59057 | New Haven County, CT | 75094 | 3.15 |
| 17367 | New London County, CT | 78552 | 4.35 |
| 8040 | Tolland County, CT | 86430 | 3.40 |
| 7251 | Windham County, CT | 71581 | 7.06 |
| 14677 | Kent County, DE | 71866 | 7.02 |
| 42417 | New Castle County, DE | 85427 | 4.37 |
| 16347 | Sussex County, DE | 81188 | 7.18 |
| 58922 | District of Columbia, DC | 99897 | 1.59 |
| 18307 | Alachua County, FL | 58783 | 3.30 |
| 14566 | Bay County, FL | 65521 | 4.82 |
| 36399 | Brevard County, FL | 74899 | 5.84 |
| 145116 | Broward County, FL | 70834 | 0.440 |
| 7347 | Charlotte County, FL | 65040 | 7.51 |
| 7206 | Citrus County, FL | 52534 | 12.14 |
| 15562 | Clay County, FL | 84364 | 5.49 |
| 88927 | Duval County, FL | 70203 | 3.97 |
| 26002 | Escambia County, FL | 61634 | 7.33 |
| 11396 | Hernando County, FL | 60582 | 6.90 |
| 118938 | Hillsborough County, FL | 74091 | 2.68 |
| 8688 | Indian River County, FL | 69462 | 4.52 |
| 24964 | Lake County, FL | 67637 | 3.71 |
| 50754 | Lee County, FL | 71607 | 2.27 |
| 19699 | Leon County, FL | 61560 | 2.56 |
| 25681 | Manatee County, FL | 72176 | 3.82 |
| 25082 | Marion County, FL | 54517 | 6.04 |
| 8810 | Martin County, FL | 80678 | 2.95 |
| 203848 | Miami-Dade County, FL | 66952 | 0.330 |
| 18394 | Okaloosa County, FL | 76663 | 6.73 |
| 113897 | Orange County, FL | 72129 | 0.960 |
| 32984 | Osceola County, FL | 63782 | 0.880 |
| 103462 | Palm Beach County, FL | 76713 | 0.900 |
| 37949 | Pasco County, FL | 65933 | 6.02 |
| 52244 | Pinellas County, FL | 66427 | 5.59 |
| 60104 | Polk County, FL | 61941 | 3.99 |
| 16550 | St. Johns County, FL | 103757 | 2.76 |
| 23841 | St. Lucie County, FL | 64038 | 3.04 |
| 13753 | Santa Rosa County, FL | 82537 | 7.42 |
| 20203 | Sarasota County, FL | 78309 | 3.20 |
| 31335 | Seminole County, FL | 80296 | 1.63 |
| 34159 | Volusia County, FL | 64601 | 5.03 |
| 9290 | Bartow County, GA | 80017 | 5.87 |
| 14374 | Bibb County, GA | 48539 | 5.57 |
| 11113 | Carroll County, GA | 69924 | 6.63 |
| 25199 | Chatham County, GA | 64029 | 2.04 |
| 19635 | Cherokee County, GA | 99932 | 1.95 |
| 9263 | Clarke County, GA | 49832 | 3.57 |
| 24517 | Clayton County, GA | 57625 | 1.51 |
| 61734 | Cobb County, GA | 96798 | 0.640 |
| 11456 | Columbia County, GA | 94224 | 1.55 |
| 11488 | Coweta County, GA | 87618 | 2.15 |
| 70132 | DeKalb County, GA | 76736 | 0.550 |
| 11749 | Douglas County, GA | 79242 | 2.55 |
| 86437 | Fulton County, GA | 89798 | 0.810 |
| 76767 | Gwinnett County, GA | 83918 | 0.530 |
| 18680 | Hall County, GA | 76789 | 2.17 |
| 19450 | Henry County, GA | 82379 | 1.84 |
| 13798 | Houston County, GA | 71237 | 3.26 |
| 11235 | Lowndes County, GA | 52623 | 3.90 |
| 18898 | Muscogee County, GA | 53740 | 3.33 |
| 12678 | Paulding County, GA | 87459 | 2.07 |
| 18963 | Richmond County, GA | 50079 | 3.09 |
| 8845 | Whitfield County, GA | 61680 | 8.64 |
| 13721 | Hawaii County, HI | 71143 | 5.29 |
| 78177 | Honolulu County, HI | 96304 | 0.980 |
| 11460 | Maui County, HI | 89794 | 2.34 |
| 35094 | Ada County, ID | 87748 | 3.21 |
| 12619 | Bonneville County, ID | 74505 | 5.42 |
| 22754 | Canyon County, ID | 70594 | 5.05 |
| 13213 | Kootenai County, ID | 75583 | 5.87 |
| 14447 | Champaign County, IL | 61205 | 4.98 |
| 383990 | Cook County, IL | 76614 | 1.23 |
| 7336 | DeKalb County, IL | 71882 | 4.57 |
| 68155 | DuPage County, IL | 102383 | 0.910 |
| 40030 | Kane County, IL | 93475 | 1.52 |
| 8124 | Kankakee County, IL | 70494 | 8.03 |
| 47165 | Lake County, IL | 101621 | 1.47 |
| 7651 | LaSalle County, IL | 64711 | 11.93 |
| 21307 | McHenry County, IL | 99128 | 2.45 |
| 11768 | McLean County, IL | 71825 | 6.26 |
| 8211 | Macon County, IL | 60648 | 11.53 |
| 18669 | Madison County, IL | 67441 | 8.96 |
| 15369 | Peoria County, IL | 65421 | 9.19 |
| 11015 | Rock Island County, IL | 62652 | 7.46 |
| 19069 | St. Clair County, IL | 67185 | 8.16 |
| 14427 | Sangamon County, IL | 73684 | 11.01 |
| 9039 | Tazewell County, IL | 73285 | 9.55 |
| 48876 | Will County, IL | 97076 | 2.09 |
| 23007 | Winnebago County, IL | 62557 | 9.35 |
| 36434 | Allen County, IN | 65967 | 6.36 |
| 10024 | Clark County, IN | 68493 | 8.82 |
| 7485 | Delaware County, IN | 54059 | 14.83 |
| 20565 | Elkhart County, IN | 63777 | 6.54 |
| 26730 | Hamilton County, IN | 117068 | 1.08 |
| 12470 | Hendricks County, IN | 94760 | 3.70 |
| 13190 | Johnson County, IN | 83036 | 7.22 |
| 37599 | Lake County, IN | 66169 | 4.72 |
| 7988 | La Porte County, IN | 66336 | 14.20 |
| 9959 | Madison County, IN | 58926 | 14.33 |
| 98302 | Marion County, IN | 62776 | 5.89 |
| 7574 | Monroe County, IN | 64299 | 9.69 |
| 11244 | Porter County, IN | 84352 | 5.55 |
| 22894 | St. Joseph County, IN | 61028 | 6.36 |
| 13904 | Tippecanoe County, IN | 56088 | 7.40 |
| 14931 | Vanderburgh County, IN | 57997 | 9.65 |
| 7850 | Vigo County, IN | 52286 | 14.76 |
| 11181 | Black Hawk County, IA | 62020 | 10.06 |
| 11741 | Johnson County, IA | 74909 | 2.67 |
| 18297 | Linn County, IA | 74134 | 7.81 |
| 45236 | Polk County, IA | 77369 | 6.53 |
| 13715 | Scott County, IA | 68576 | 8.09 |
| 9940 | Woodbury County, IA | 66830 | 11.03 |
| 7172 | Douglas County, KS | 67482 | 7.11 |
| 48396 | Johnson County, KS | 99174 | 1.51 |
| 46079 | Sedgwick County, KS | 64720 | 7.31 |
| 13791 | Shawnee County, KS | 61255 | 11 |
| 16991 | Wyandotte County, KS | 58701 | 4.74 |
| 11860 | Boone County, KY | 90265 | 7.10 |
| 26174 | Fayette County, KY | 63667 | 7.25 |
| 9385 | Hardin County, KY | 61920 | 12.66 |
| 66206 | Jefferson County, KY | 64719 | 7.48 |
| 15267 | Kenton County, KY | 77284 | 11.12 |
| 12523 | Warren County, KY | 60651 | 5.40 |
| 11564 | Ascension Parish, LA | 81284 | 1.79 |
| 11284 | Bossier Parish, LA | 63812 | 4.33 |
| 19927 | Caddo Parish, LA | 47699 | 5.03 |
| 17588 | Calcasieu Parish, LA | 61797 | 3.29 |
| 38804 | East Baton Rouge Parish, LA | 61376 | 1.83 |
| 37142 | Jefferson Parish, LA | 61157 | 1.77 |
| 23015 | Lafayette Parish, LA | 58641 | 2.65 |
| 12889 | Livingston Parish, LA | 72887 | 3.41 |
| 29212 | Orleans Parish, LA | 51913 | 1.57 |
| 13883 | Ouachita Parish, LA | 47283 | 5.01 |
| 11340 | Rapides Parish, LA | 55870 | 5.91 |
| 20165 | St. Tammany Parish, LA | 78629 | 4.81 |
| 13379 | Tangipahoa Parish, LA | 54942 | 2.92 |
| 9366 | Terrebonne Parish, LA | 57376 | 4.49 |
| 8063 | Androscoggin County, ME | 62848 | 10.62 |
| 19382 | Cumberland County, ME | 88571 | 4.16 |
| 7569 | Kennebec County, ME | 63191 | 10.13 |
| 8985 | Penobscot County, ME | 60761 | 10.92 |
| 12761 | York County, ME | 81315 | 7.14 |
| 47232 | Anne Arundel County, MD | 112525 | 2.98 |
| 65127 | Baltimore County, MD | 86198 | 3.64 |
| 11842 | Carroll County, MD | 105096 | 4.52 |
| 7980 | Cecil County, MD | 84513 | 12.29 |
| 12626 | Charles County, MD | 112932 | 3.08 |
| 21532 | Frederick County, MD | 116796 | 3.27 |
| 18269 | Harford County, MD | 100334 | 5.16 |
| 22263 | Howard County, MD | 133068 | 1.07 |
| 81230 | Montgomery County, MD | 118020 | 0.680 |
| 77654 | Prince George's County, MD | 93833 | 0.930 |
| 9508 | St. Mary's County, MD | 112154 | 5.57 |
| 11317 | Washington County, MD | 69271 | 11.33 |
| 50682 | Baltimore city, MD | 54735 | 5.46 |
| 10378 | Barnstable County, MA | 89774 | 5.69 |
| 6394 | Berkshire County, MA | 71102 | 10.76 |
| 39208 | Bristol County, MA | 80702 | 5.39 |
| 59374 | Essex County, MA | 92118 | 1.94 |
| 33261 | Hampden County, MA | 64030 | 4.36 |
| 6421 | Hampshire County, MA | 81878 | 4.06 |
| 112463 | Middlesex County, MA | 118494 | 1.49 |
| 49207 | Norfolk County, MA | 115770 | 1.39 |
| 36705 | Plymouth County, MA | 102647 | 3.06 |
| 59159 | Suffolk County, MA | 84548 | 1.12 |
| 59265 | Worcester County, MA | 86078 | 4.64 |
| 8704 | Allegan County, MI | 75893 | 8.95 |
| 6344 | Bay County, MI | 56388 | 15.61 |
| 10619 | Berrien County, MI | 61790 | 10.17 |
| 10267 | Calhoun County, MI | 59652 | 13.60 |
| 7231 | Eaton County, MI | 71499 | 8.39 |
| 30202 | Genesee County, MI | 57443 | 13.24 |
| 20064 | Ingham County, MI | 60571 | 9.30 |
| 11293 | Jackson County, MI | 61299 | 14.49 |
| 19251 | Kalamazoo County, MI | 69689 | 11.28 |
| 56939 | Kent County, MI | 76923 | 2.97 |
| 11820 | Livingston County, MI | 101227 | 4.42 |
| 61113 | Macomb County, MI | 73927 | 5.40 |
| 10327 | Monroe County, MI | 72166 | 10.08 |
| 13230 | Muskegon County, MI | 58303 | 12.49 |
| 87631 | Oakland County, MI | 90801 | 2.97 |
| 21658 | Ottawa County, MI | 85968 | 3.42 |
| 13662 | Saginaw County, MI | 54064 | 12.71 |
| 10247 | St. Clair County, MI | 64894 | 14.59 |
| 22242 | Washtenaw County, MI | 80941 | 4.21 |
| 147992 | Wayne County, MI | 55928 | 5.38 |
| 28534 | Anoka County, MN | 92249 | 4.04 |
| 34738 | Dakota County, MN | 97727 | 2.62 |
| 103273 | Hennepin County, MN | 89418 | 2.57 |
| 13730 | Olmsted County, MN | 88032 | 4.29 |
| 47518 | Ramsey County, MN | 75041 | 3.96 |
| 12496 | St. Louis County, MN | 63141 | 12.86 |
| 11743 | Scott County, MN | 115687 | 2.73 |
| 14058 | Stearns County, MN | 72378 | 6.62 |
| 19167 | Washington County, MN | 107649 | 2.69 |
| 12703 | Wright County, MN | 95259 | 4.09 |
| 15318 | DeSoto County, MS | 79967 | 4.69 |
| 17326 | Harrison County, MS | 55916 | 6.20 |
| 19281 | Hinds County, MS | 47479 | 4.05 |
| 10864 | Jackson County, MS | 61515 | 7.29 |
| 12150 | Rankin County, MS | 72906 | 4.06 |
| 13590 | Boone County, MO | 63981 | 7.45 |
| 20781 | Clay County, MO | 78930 | 6.28 |
| 7934 | Franklin County, MO | 65779 | 13.12 |
| 22914 | Greene County, MO | 55751 | 11.69 |
| 60264 | Jackson County, MO | 62509 | 7.23 |
| 10728 | Jasper County, MO | 53548 | 11.36 |
| 16677 | Jefferson County, MO | 78297 | 11.52 |
| 28851 | St. Charles County, MO | 98925 | 4.23 |
| 72521 | St. Louis County, MO | 79609 | 4.59 |
| 24881 | St. Louis city, MO | 52278 | 7.78 |
| 7491 | Missoula County, MT | 68210 | 7.74 |
| 12828 | Yellowstone County, MT | 78216 | 13.21 |
| 55993 | Douglas County, NE | 74424 | 4.89 |
| 26001 | Lancaster County, NE | 68358 | 5.90 |
| 16778 | Sarpy County, NE | 98619 | 3.64 |
| 175327 | Clark County, NV | 70838 | 2.53 |
| 34800 | Washoe County, NV | 80245 | 3.30 |
| 28953 | Hillsborough County, NH | 96415 | 5.46 |
| 8961 | Merrimack County, NH | 84898 | 5.99 |
| 18894 | Rockingham County, NH | 107445 | 2.88 |
| 7901 | Strafford County, NH | 82845 | 8.10 |
| 19276 | Atlantic County, NJ | 76108 | 6.15 |
| 61993 | Bergen County, NJ | 113885 | 0.640 |
| 31600 | Burlington County, NJ | 100188 | 3.37 |
| 42040 | Camden County, NJ | 81372 | 4.21 |
| 12773 | Cumberland County, NJ | 60892 | 6.73 |
| 69450 | Essex County, NJ | 74747 | 1.39 |
| 20188 | Gloucester County, NJ | 97851 | 4.81 |
| 65455 | Hudson County, NJ | 82361 | 0.770 |
| 7241 | Hunterdon County, NJ | 137334 | 1.84 |
| 28477 | Mercer County, NJ | 94832 | 2.14 |
| 59370 | Middlesex County, NJ | 102073 | 1.24 |
| 42622 | Monmouth County, NJ | 117699 | 1.37 |
| 33345 | Morris County, NJ | 131562 | 0.700 |
| 68988 | Ocean County, NJ | 81152 | 1.57 |
| 43760 | Passaic County, NJ | 78779 | 1.30 |
| 22098 | Somerset County, NJ | 134735 | 1.27 |
| 9249 | Sussex County, NJ | 111308 | 3.80 |
| 46241 | Union County, NJ | 96483 | 0.730 |
| 7177 | Warren County, NJ | 97729 | 4.82 |
| 45975 | Bernalillo County, NM | 65034 | 3.68 |
| 17941 | Dona Ana County, NM | 52011 | 2.31 |
| 9258 | Sandoval County, NM | 82862 | 2.59 |
| 9165 | San Juan County, NM | 51898 | 5.72 |
| 7868 | Santa Fe County, NM | 72836 | 3.34 |
| 19642 | Albany County, NY | 76997 | 6.03 |
| 124230 | Bronx County, NY | 45864 | 0.610 |
| 12582 | Broome County, NY | 60533 | 12.14 |
| 8649 | Chautauqua County, NY | 55644 | 12.90 |
| 18274 | Dutchess County, NY | 90740 | 3.05 |
| 65097 | Erie County, NY | 69888 | 6.82 |
| 12213 | Jefferson County, NY | 58763 | 9.42 |
| 243533 | Kings County, NY | 73244 | 0.230 |
| 51812 | Monroe County, NY | 68169 | 5.89 |
| 96131 | Nassau County, NY | 135528 | 0.420 |
| 103761 | New York County, NY | 95514 | 0.340 |
| 13749 | Niagara County, NY | 63565 | 12.44 |
| 16428 | Oneida County, NY | 63894 | 9.73 |
| 33510 | Onondaga County, NY | 70939 | 7.68 |
| 6712 | Ontario County, NY | 76601 | 8.05 |
| 37791 | Orange County, NY | 88532 | 1.83 |
| 8126 | Oswego County, NY | 59335 | 16.21 |
| 175588 | Queens County, NY | 80180 | 0.310 |
| 10754 | Rensselaer County, NY | 80040 | 7.86 |
| 35382 | Richmond County, NY | 91582 | 0.840 |
| 6920 | St. Lawrence County, NY | 59451 | 14.29 |
| 14110 | Saratoga County, NY | 93341 | 5.07 |
| 12440 | Schenectady County, NY | 74206 | 3.62 |
| 108290 | Suffolk County, NY | 119253 | 1.47 |
| 4697 | Tompkins County, NY | 72025 | 8.11 |
| 10433 | Ulster County, NY | 79340 | 6.58 |
| 69003 | Westchester County, NY | 108037 | 0.520 |
| 13463 | Alamance County, NC | 61034 | 5.30 |
| 7779 | Brunswick County, NC | 73536 | 10.09 |
| 16798 | Buncombe County, NC | 67906 | 6.48 |
| 18401 | Cabarrus County, NC | 79649 | 3.74 |
| 11683 | Catawba County, NC | 63934 | 9.60 |
| 9020 | Craven County, NC | 63076 | 6.43 |
| 36657 | Cumberland County, NC | 58013 | 5.94 |
| 12394 | Davidson County, NC | 63590 | 10.56 |
| 28827 | Durham County, NC | 79524 | 1.67 |
| 30890 | Forsyth County, NC | 62992 | 4.78 |
| 18836 | Gaston County, NC | 65430 | 9.20 |
| 41136 | Guilford County, NC | 63822 | 4.19 |
| 13131 | Harnett County, NC | 65242 | 6.06 |
| 7163 | Henderson County, NC | 65252 | 6.91 |
| 13881 | Iredell County, NC | 73446 | 6.37 |
| 18759 | Johnston County, NC | 80612 | 1.88 |
| 104441 | Mecklenburg County, NC | 80645 | 1.24 |
| 15155 | New Hanover County, NC | 70995 | 4.47 |
| 26602 | Onslow County, NC | 62190 | 3.71 |
| 8043 | Orange County, NC | 87780 | 2.90 |
| 14419 | Pitt County, NC | 57319 | 5.62 |
| 10891 | Randolph County, NC | 58037 | 8.76 |
| 12735 | Robeson County, NC | 39931 | 12.40 |
| 11120 | Rowan County, NC | 61105 | 9.38 |
| 17327 | Union County, NC | 95079 | 2.68 |
| 90404 | Wake County, NC | 97099 | 1.31 |
| 11364 | Wayne County, NC | 55635 | 5.10 |
| 16501 | Cass County, ND | 70002 | 5.67 |
| 8560 | Allen County, OH | 60384 | 13.19 |
| 7213 | Ashtabula County, OH | 52468 | 17.37 |
| 30275 | Butler County, OH | 74841 | 9.08 |
| 11086 | Clark County, OH | 57264 | 13.77 |
| 15560 | Clermont County, OH | 79492 | 10.49 |
| 6770 | Columbiana County, OH | 54167 | 19.70 |
| 91896 | Cuyahoga County, OH | 60808 | 5.11 |
| 14885 | Delaware County, OH | 121528 | 3.07 |
| 11746 | Fairfield County, OH | 82447 | 9.12 |
| 123475 | Franklin County, OH | 69762 | 6.16 |
| 11884 | Greene County, OH | 82769 | 7.20 |
| 72529 | Hamilton County, OH | 67033 | 6.84 |
| 14711 | Lake County, OH | 76163 | 7.22 |
| 13273 | Licking County, OH | 76596 | 10.69 |
| 22139 | Lorain County, OH | 67124 | 10.23 |
| 34681 | Lucas County, OH | 57971 | 8.80 |
| 16265 | Mahoning County, OH | 54774 | 17.37 |
| 11543 | Medina County, OH | 89108 | 5.09 |
| 8098 | Miami County, OH | 72887 | 9.95 |
| 43181 | Montgomery County, OH | 62794 | 7.89 |
| 9489 | Portage County, OH | 72965 | 10.39 |
| 9476 | Richland County, OH | 54906 | 14.90 |
| 27672 | Stark County, OH | 64557 | 12 |
| 38632 | Summit County, OH | 66034 | 8.82 |
| 13986 | Trumbull County, OH | 54406 | 18.90 |
| 16762 | Warren County, OH | 104523 | 4.95 |
| 9852 | Wayne County, OH | 70121 | 8.09 |
| 8697 | Wood County, OH | 67254 | 6.17 |
| 13869 | Canadian County, OK | 81761 | 3.17 |
| 19195 | Cleveland County, OK | 70923 | 5.55 |
| 11392 | Comanche County, OK | 55600 | 6.77 |
| 77162 | Oklahoma County, OK | 61580 | 4.16 |
| 61069 | Tulsa County, OK | 63442 | 6.06 |
| 26210 | Clackamas County, OR | 96915 | 3.64 |
| 12969 | Deschutes County, OR | 84281 | 4.83 |
| 6962 | Douglas County, OR | 55837 | 14.02 |
| 14669 | Jackson County, OR | 68239 | 7.53 |
| 21179 | Lane County, OR | 64192 | 7.80 |
| 10064 | Linn County, OR | 67265 | 9.08 |
| 27793 | Marion County, OR | 70861 | 4.88 |
| 52225 | Multnomah County, OR | 79432 | 3.67 |
| 42730 | Washington County, OR | 98706 | 1.64 |
| 6311 | Adams County, PA | 76554 | 7.68 |
| 84377 | Allegheny County, PA | 71973 | 6.21 |
| 10503 | Beaver County, PA | 67135 | 11.60 |
| 31618 | Berks County, PA | 72157 | 5.17 |
| 8107 | Blair County, PA | 57884 | 14.60 |
| 38785 | Bucks County, PA | 105202 | 4.32 |
| 11971 | Butler County, PA | 81507 | 7.97 |
| 8256 | Cambria County, PA | 54612 | 15.42 |
| 7626 | Centre County, PA | 67654 | 5.76 |
| 37871 | Chester County, PA | 117326 | 2.68 |
| 18379 | Cumberland County, PA | 81295 | 5.94 |
| 22887 | Dauphin County, PA | 67813 | 7.02 |
| 43218 | Delaware County, PA | 83856 | 4.01 |
| 18412 | Erie County, PA | 56489 | 12.92 |
| 8259 | Fayette County, PA | 51494 | 19.22 |
| 11879 | Franklin County, PA | 73709 | 8.28 |
| 14059 | Lackawanna County, PA | 61168 | 12.14 |
| 48303 | Lancaster County, PA | 82434 | 4.37 |
| 10896 | Lebanon County, PA | 69490 | 6.65 |
| 27840 | Lehigh County, PA | 72590 | 5.17 |
| 22049 | Luzerne County, PA | 59358 | 10.59 |
| 7855 | Lycoming County, PA | 61729 | 14.41 |
| 7091 | Mercer County, PA | 57702 | 15.13 |
| 9632 | Monroe County, PA | 79824 | 8.13 |
| 58583 | Montgomery County, PA | 106743 | 2.96 |
| 19365 | Northampton County, PA | 75468 | 5.68 |
| 132998 | Philadelphia County, PA | 56385 | 3.77 |
| 8821 | Schuylkill County, PA | 63680 | 15.63 |
| 13560 | Washington County, PA | 70751 | 11.75 |
| 19863 | Westmoreland County, PA | 71152 | 10.76 |
| 33202 | York County, PA | 79938 | 6.98 |
| 10172 | Kent County, RI | 84230 | 4.49 |
| 47909 | Providence County, RI | 74190 | 3 |
| 5881 | Washington County, RI | 99015 | 4.06 |
| 12958 | Aiken County, SC | 67728 | 5.78 |
| 15927 | Anderson County, SC | 62708 | 9.42 |
| 13125 | Beaufort County, SC | 83687 | 2.65 |
| 20728 | Berkeley County, SC | 79259 | 4.68 |
| 34628 | Charleston County, SC | 79969 | 2.09 |
| 13622 | Dorchester County, SC | 74730 | 4.77 |
| 11111 | Florence County, SC | 52650 | 5.86 |
| 44972 | Greenville County, SC | 71833 | 4.05 |
| 22821 | Horry County, SC | 61375 | 8.01 |
| 22711 | Lexington County, SC | 73022 | 3.87 |
| 8508 | Pickens County, SC | 55333 | 8.02 |
| 32196 | Richland County, SC | 61010 | 1.17 |
| 28596 | Spartanburg County, SC | 58535 | 5.81 |
| 9377 | Sumter County, SC | 56937 | 3.55 |
| 20749 | York County, SC | 83692 | 4.07 |
| 19479 | Minnehaha County, SD | 75623 | 5.87 |
| 9955 | Pennington County, SD | 67573 | 10.60 |
| 9054 | Blount County, TN | 69713 | 11.70 |
| 70452 | Davidson County, TN | 72561 | 3.48 |
| 30527 | Hamilton County, TN | 75127 | 5.74 |
| 36932 | Knox County, TN | 70013 | 8.33 |
| 25144 | Montgomery County, TN | 67478 | 6.31 |
| 30515 | Rutherford County, TN | 77320 | 4.44 |
| 86476 | Shelby County, TN | 61452 | 3.14 |
| 10394 | Sullivan County, TN | 54361 | 15.76 |
| 15744 | Sumner County, TN | 87453 | 7.40 |
| 8668 | Washington County, TN | 64270 | 12.01 |
| 12421 | Wilson County, TN | 88462 | 6.09 |
| 42118 | Bell County, TX | 62412 | 2.56 |
| 183122 | Bexar County, TX | 65839 | 0.730 |
| 32604 | Brazoria County, TX | 86197 | 2.06 |
| 18809 | Brazos County, TX | 60355 | 1.64 |
| 85710 | Collin County, TX | 113943 | 0.490 |
| 12761 | Comal County, TX | 92979 | 1.83 |
| 257839 | Dallas County, TX | 70871 | 0.520 |
| 75897 | Denton County, TX | 102711 | 0.790 |
| 20859 | Ector County, TX | 64868 | 3.33 |
| 18034 | Ellis County, TX | 89314 | 1.79 |
| 78428 | El Paso County, TX | 53441 | 0.600 |
| 64701 | Fort Bend County, TX | 105583 | 0.370 |
| 27343 | Galveston County, TX | 75947 | 3.09 |
| 11900 | Grayson County, TX | 66141 | 4.61 |
| 11611 | Gregg County, TX | 60932 | 5.27 |
| 13957 | Guadalupe County, TX | 86043 | 1.58 |
| 454216 | Harris County, TX | 68748 | 0.640 |
| 96678 | Hidalgo County, TX | 48825 | 0.260 |
| 22497 | Jefferson County, TX | 55065 | 4.38 |
| 16371 | Johnson County, TX | 76747 | 4.19 |
| 16578 | Kaufman County, TX | 92814 | 1.74 |
| 28325 | Lubbock County, TX | 59138 | 3.22 |
| 23327 | McLennan County, TX | 63160 | 3.55 |
| 20997 | Midland County, TX | 81187 | 2.76 |
| 56694 | Montgomery County, TX | 95241 | 2.02 |
| 30350 | Nueces County, TX | 60250 | 1.62 |
| 12350 | Parker County, TX | 95277 | 3.38 |
| 11012 | Potter County, TX | 51028 | 8.70 |
| 11715 | Randall County, TX | 77066 | 5.63 |
| 21030 | Smith County, TX | 67978 | 2.18 |
| 191010 | Tarrant County, TX | 76285 | 1.45 |
| 13649 | Taylor County, TX | 61798 | 5.19 |
| 10398 | Tom Green County, TX | 60348 | 3.65 |
| 110483 | Travis County, TX | 95151 | 0.760 |
| 10856 | Wichita County, TX | 56634 | 7.51 |
| 53621 | Williamson County, TX | 101891 | 0.840 |
| 14399 | Cache County, UT | 76631 | 1.11 |
| 33922 | Davis County, UT | 103143 | 1.70 |
| 105389 | Salt Lake County, UT | 91713 | 1.94 |
| 83649 | Utah County, UT | 95085 | 0.770 |
| 15947 | Washington County, UT | 75572 | 1.52 |
| 24434 | Weber County, UT | 83465 | 3.07 |
| 9528 | Chittenden County, VT | 86579 | 5.78 |
| 28141 | Chesterfield County, VA | 100149 | 2.13 |
| 25974 | Henrico County, VA | 82187 | 2.69 |
| 43962 | Prince William County, VA | 119051 | 0.710 |
| 11100 | Spotsylvania County, VA | 96322 | 4.94 |
| 12619 | Stafford County, VA | 134456 | 2.60 |
| 19782 | Chesapeake city, VA | 87127 | 2.85 |
| 11050 | Hampton city, VA | 64086 | 4.24 |
| 17847 | Newport News city, VA | 63350 | 4.93 |
| 22033 | Norfolk city, VA | 60030 | 3.92 |
| 20909 | Richmond city, VA | 58719 | 3.95 |

The first 500 of 530 rows; every row is in https://vishalsingh.org/zipcode/report/smoking-in-pregnancy/figures/smoking_vs_income_counties.json.

Note: Large counties are the counties that CDC WONDER names individually: those with 100,000 residents or more. WONDER pools the births of every smaller county in a state into one row, so smaller counties are not in these data. Of the 578 large counties named in every year from 2018 to 2024, California (35) and 13 counties with a withheld smoking count are left out, which leaves 530.

Note: A county's smoking share is the births to mothers who reported smoking over the births with a known answer, Yes plus No. Births with an Unknown or Not Stated answer are left out of the denominator.

Note: County median household income is the County Health Rankings 2024 release, a Census Small Area Income and Poverty Estimates value for 2022. Its year differs from the births, which are from 2018 to 2024.

Note: Intervals are 95% intervals from a bootstrap over counties (replicates and seed in the methods note). Births are full counts, so an interval shows how much a pooled share depends on which counties are in the group. Tenths hold equal numbers of counties.

Source: CDC WONDER natality, tobacco use during pregnancy, births in 2018-2024; full counts of births; counties ranked on County Health Rankings 2024 median household income (Census SAIPE 2022); tenths pooled by births, intervals from a bootstrap over counties

Data: https://vishalsingh.org/zipcode/report/smoking-in-pregnancy/figures/smoking_vs_income_counties.json

### Table 4. Within a state, a doubling of county income goes with a smoking share 3.9 points lower when counties are weighted by births, and 6.6 points lower when each county counts once

Change in the share of births to mothers who smoked, in percentage points per doubling of county median household income (SAIPE 2022), across all counties, within states and between states, 2018 to 2024, with 95% intervals from a bootstrap over states

| conf_high | conf_low | estimate | n | scope | weighting |
| ---: | ---: | ---: | ---: | --- | --- |
| −2.71 | −5.91 | −4.12 | 49 | Across all counties | Weighted by births |
| −2.75 | −5.53 | −3.88 | 45 | Within states | Weighted by births |
| −1.82 | −8.33 | −4.71 | 49 | Between states | Weighted by births |
| −4.57 | −8 | −6.24 | 49 | Across all counties | Each county counts once |
| −5.11 | −8.20 | −6.63 | 45 | Within states | Each county counts once |
| −2.19 | −8.69 | −5.29 | 49 | Between states | Each county counts once |

Note: Large counties are the counties that CDC WONDER names individually: those with 100,000 residents or more. WONDER pools the births of every smaller county in a state into one row, so smaller counties are not in these data. Of the 578 large counties named in every year from 2018 to 2024, California (35) and 13 counties with a withheld smoking count are left out, which leaves 530.

Note: County median household income is the County Health Rankings 2024 release, a Census Small Area Income and Poverty Estimates value for 2022. Its year differs from the births, which are from 2018 to 2024.

Note: Intervals are 95% intervals from a bootstrap over states (replicates and seed in the methods note), which allows for counties in one state resembling each other. The between-state slope compares 49 state means, so its interval is the widest; 45 states have two or more counties and contribute to the within-state slope.

Note: The weighted slopes weight each county by its births with a known answer. The unweighted slopes count each county once.

Source: CDC WONDER natality, tobacco use during pregnancy, births in 2018-2024; full counts of births; counties ranked on County Health Rankings 2024 median household income (Census SAIPE 2022); slopes weighted by births or unweighted as labeled; intervals from a bootstrap over states

Data: https://vishalsingh.org/zipcode/report/smoking-in-pregnancy/figures/slope_scope.json

### Table 5. In all 26 states with a slope of their own the estimated slope is below zero, and the 95% interval lies below zero in 10 when counties are weighted by births

Change in the share of births to mothers who smoked per doubling of county median household income (SAIPE 2022), one dot per state with at least 8 complete large counties, 2018 to 2024, with 95% intervals; four views by scale and by weighting

| State | Scale | Counties | Slope | Lower limit of the 95% interval | Upper limit of the 95% interval | Complete counties | Counties left out for a withheld count | State share (%) |
| --- | --- | --- | ---: | ---: | ---: | ---: | ---: | ---: |
| Alabama | points | Weighted by births | −3.28 | −7.14 | 0.58 | 12 | 0 | 4.6 |
| Arizona | points | Weighted by births | −4.97 | −13.64 | 3.70 | 9 | 0 | 3.2 |
| Colorado | points | Weighted by births | −4.35 | −8.56 | −0.14 | 11 | 0 | 3.7 |
| Connecticut | points | Weighted by births | −6.42 | −16.92 | 4.08 | 8 | 0 | 2.6 |
| Florida | points | Weighted by births | −3.97 | −10.03 | 2.09 | 32 | 1 | 2.7 |
| Georgia | points | Weighted by births | −3.34 | −4.95 | −1.73 | 22 | 2 | 1.8 |
| Illinois | points | Weighted by births | −8.37 | −16.40 | −0.33 | 19 | 1 | 2.9 |
| Indiana | points | Weighted by births | −7.17 | −10.21 | −4.12 | 17 | 0 | 6.7 |
| Louisiana | points | Weighted by births | −0.81 | −6.94 | 5.31 | 14 | 0 | 3.1 |
| Maryland | points | Weighted by births | −4.04 | −9.51 | 1.43 | 13 | 0 | 3.1 |
| Massachusetts | points | Weighted by births | −4.24 | −7.52 | −0.97 | 11 | 0 | 2.8 |
| Michigan | points | Weighted by births | −7.45 | −24.40 | 9.51 | 20 | 0 | 6.6 |
| Minnesota | points | Weighted by births | −7.80 | −19.54 | 3.93 | 10 | 0 | 3.7 |
| Missouri | points | Weighted by births | −6.23 | −12.12 | −0.34 | 10 | 0 | 7.2 |
| New Jersey | points | Weighted by births | −1.86 | −4.47 | 0.75 | 19 | 0 | 1.9 |
| New York | points | Weighted by births | −2.36 | −9.13 | 4.41 | 27 | 1 | 2.3 |
| North Carolina | points | Weighted by births | −8.20 | −11.15 | −5.25 | 27 | 0 | 4.2 |
| Ohio | points | Weighted by births | −7.34 | −13.71 | −0.98 | 28 | 0 | 8.3 |
| Oregon | points | Weighted by births | −9.98 | −15.26 | −4.70 | 9 | 0 | 4.8 |
| Pennsylvania | points | Weighted by births | −5.10 | −15.68 | 5.47 | 31 | 0 | 6.4 |
| South Carolina | points | Weighted by births | −4.51 | −10.99 | 1.97 | 15 | 0 | 4.5 |
| Tennessee | points | Weighted by births | −0.91 | −28.66 | 26.85 | 11 | 1 | 5.5 |
| Texas | points | Weighted by births | −0.63 | −2.10 | 0.83 | 36 | 3 | 1.3 |
| Virginia | points | Weighted by births | −2.72 | −5.20 | −0.25 | 11 | 4 | 2.8 |
| Washington | points | Weighted by births | −5.69 | −14.20 | 2.83 | 12 | 0 | 3.7 |
| Wisconsin | points | Weighted by births | −3.37 | −20.54 | 13.80 | 15 | 0 | 5.5 |
| Alabama | points | Each county counts once | −5.47 | −11.26 | 0.33 | 12 | 0 | 4.6 |
| Arizona | points | Each county counts once | −4.59 | −13.11 | 3.92 | 9 | 0 | 3.2 |
| Colorado | points | Each county counts once | −5.89 | −9.84 | −1.94 | 11 | 0 | 3.7 |
| Connecticut | points | Each county counts once | −8.16 | −18.13 | 1.81 | 8 | 0 | 2.6 |
| Florida | points | Each county counts once | −4.49 | −10.73 | 1.75 | 32 | 1 | 2.7 |
| Georgia | points | Each county counts once | −2.96 | −4.84 | −1.07 | 22 | 2 | 1.8 |
| Illinois | points | Each county counts once | −11.40 | −14.72 | −8.09 | 19 | 1 | 2.9 |
| Indiana | points | Each county counts once | −9.50 | −13.06 | −5.93 | 17 | 0 | 6.7 |
| Louisiana | points | Each county counts once | −1.54 | −6.01 | 2.93 | 14 | 0 | 3.1 |
| Maryland | points | Each county counts once | −5.80 | −17.26 | 5.67 | 13 | 0 | 3.1 |
| Massachusetts | points | Each county counts once | −6.41 | −14.99 | 2.16 | 11 | 0 | 2.8 |
| Michigan | points | Each county counts once | −12.63 | −18.78 | −6.48 | 20 | 0 | 6.6 |
| Minnesota | points | Each county counts once | −9.52 | −22.26 | 3.22 | 10 | 0 | 3.7 |
| Missouri | points | Each county counts once | −5.82 | −12.76 | 1.12 | 10 | 0 | 7.2 |
| New Jersey | points | Each county counts once | −2.87 | −6.11 | 0.37 | 19 | 0 | 1.9 |
| New York | points | Each county counts once | −8.60 | −16.15 | −1.06 | 27 | 1 | 2.3 |
| North Carolina | points | Each county counts once | −7.86 | −10.18 | −5.54 | 27 | 0 | 4.2 |
| Ohio | points | Each county counts once | −11.15 | −15.57 | −6.73 | 28 | 0 | 8.3 |
| Oregon | points | Each county counts once | −12.13 | −20.36 | −3.90 | 9 | 0 | 4.8 |
| Pennsylvania | points | Each county counts once | −11.72 | −16.34 | −7.11 | 31 | 0 | 6.4 |
| South Carolina | points | Each county counts once | −4.71 | −9.31 | −0.10 | 15 | 0 | 4.5 |
| Tennessee | points | Each county counts once | −9.43 | −29.00 | 10.13 | 11 | 1 | 5.5 |
| Texas | points | Each county counts once | −2.73 | −5.19 | −0.26 | 36 | 3 | 1.3 |
| Virginia | points | Each county counts once | −2.09 | −4.23 | 0.05 | 11 | 4 | 2.8 |
| Washington | points | Each county counts once | −4.69 | −15.06 | 5.67 | 12 | 0 | 3.7 |
| Wisconsin | points | Each county counts once | −6.69 | −15.34 | 1.95 | 15 | 0 | 5.5 |
| Alabama | proportion | Weighted by births | −7.62 | −17.55 | 3.51 | 12 | 0 | 4.6 |
| Arizona | proportion | Weighted by births | −13.80 | −42.48 | 29.20 | 9 | 0 | 3.2 |
| Colorado | proportion | Weighted by births | −14.86 | −24.34 | −4.20 | 11 | 0 | 3.7 |
| Connecticut | proportion | Weighted by births | −37.17 | −76.01 | 64.60 | 8 | 0 | 2.6 |
| Florida | proportion | Weighted by births | −9.96 | −39.70 | 34.45 | 32 | 1 | 2.7 |
| Georgia | proportion | Weighted by births | −20.60 | −27.42 | −13.14 | 22 | 2 | 1.8 |
| Illinois | proportion | Weighted by births | −24.70 | −43.86 | 1.01 | 19 | 1 | 2.9 |
| Indiana | proportion | Weighted by births | −21.06 | −37.44 | −0.40 | 17 | 0 | 6.7 |
| Louisiana | proportion | Weighted by births | −2.35 | −25.39 | 27.80 | 14 | 0 | 3.1 |
| Maryland | proportion | Weighted by births | −17.27 | −33.50 | 2.91 | 13 | 0 | 3.1 |
| Massachusetts | proportion | Weighted by births | −17.13 | −28.02 | −4.60 | 11 | 0 | 2.8 |
| Michigan | proportion | Weighted by births | −15.68 | −37.23 | 13.26 | 20 | 0 | 6.6 |
| Minnesota | proportion | Weighted by births | −18.02 | −32.93 | 0.21 | 10 | 0 | 3.7 |
| Missouri | proportion | Weighted by births | −12.11 | −20.52 | −2.81 | 10 | 0 | 7.2 |
| New Jersey | proportion | Weighted by births | −11.04 | −23.37 | 3.28 | 19 | 0 | 1.9 |
| New York | proportion | Weighted by births | −7.29 | −27.80 | 19.05 | 27 | 1 | 2.3 |
| North Carolina | proportion | Weighted by births | −26.20 | −33.19 | −18.48 | 27 | 0 | 4.2 |
| Ohio | proportion | Weighted by births | −10.74 | −19.29 | −1.28 | 28 | 0 | 8.3 |
| Oregon | proportion | Weighted by births | −25.39 | −40.79 | −5.98 | 9 | 0 | 4.8 |
| Pennsylvania | proportion | Weighted by births | −9.71 | −27.57 | 12.55 | 31 | 0 | 6.4 |
| South Carolina | proportion | Weighted by births | −9.25 | −30.76 | 18.96 | 15 | 0 | 4.5 |
| Tennessee | proportion | Weighted by births | 5.28 | −39.93 | 84.52 | 11 | 1 | 5.5 |
| Texas | proportion | Weighted by births | −0.52 | −15.83 | 17.58 | 36 | 3 | 1.3 |
| Virginia | proportion | Weighted by births | −16.02 | −33.86 | 6.62 | 11 | 4 | 2.8 |
| Washington | proportion | Weighted by births | −21.40 | −49.98 | 23.49 | 12 | 0 | 3.7 |
| Wisconsin | proportion | Weighted by births | −10.92 | −43.04 | 39.32 | 15 | 0 | 5.5 |
| Alabama | proportion | Each county counts once | −12.06 | −21.15 | −1.91 | 12 | 0 | 4.6 |
| Arizona | proportion | Each county counts once | −10.74 | −31.45 | 16.21 | 9 | 0 | 3.2 |
| Colorado | proportion | Each county counts once | −17.41 | −21.63 | −12.96 | 11 | 0 | 3.7 |
| Connecticut | proportion | Each county counts once | −33.42 | −63.91 | 22.83 | 8 | 0 | 2.6 |
| Florida | proportion | Each county counts once | −11.39 | −26.28 | 6.51 | 32 | 1 | 2.7 |
| Georgia | proportion | Each county counts once | −15.69 | −23.22 | −7.43 | 22 | 2 | 1.8 |
| Illinois | proportion | Each county counts once | −31.15 | −37.93 | −23.62 | 19 | 1 | 2.9 |
| Indiana | proportion | Each county counts once | −21.07 | −33.11 | −6.85 | 17 | 0 | 6.7 |
| Louisiana | proportion | Each county counts once | −5.57 | −22.73 | 15.41 | 14 | 0 | 3.1 |
| Maryland | proportion | Each county counts once | −17.22 | −37.28 | 9.25 | 13 | 0 | 3.1 |
| Massachusetts | proportion | Each county counts once | −20.85 | −35.58 | −2.75 | 11 | 0 | 2.8 |
| Michigan | proportion | Each county counts once | −20.74 | −29.38 | −11.04 | 20 | 0 | 6.6 |
| Minnesota | proportion | Each county counts once | −19.90 | −33.80 | −3.08 | 10 | 0 | 3.7 |
| Missouri | proportion | Each county counts once | −10.92 | −21.19 | 0.69 | 10 | 0 | 7.2 |
| New Jersey | proportion | Each county counts once | −12.55 | −24.87 | 1.79 | 19 | 0 | 1.9 |
| New York | proportion | Each county counts once | −24.26 | −42.67 | 0.05 | 27 | 1 | 2.3 |
| North Carolina | proportion | Each county counts once | −20.97 | −28.79 | −12.29 | 27 | 0 | 4.2 |
| Ohio | proportion | Each county counts once | −15.14 | −18.88 | −11.24 | 28 | 0 | 8.3 |
| Oregon | proportion | Each county counts once | −24.96 | −36.84 | −10.85 | 9 | 0 | 4.8 |
| Pennsylvania | proportion | Each county counts once | −17.99 | −23.00 | −12.65 | 31 | 0 | 6.4 |
| South Carolina | proportion | Each county counts once | −11.04 | −23.82 | 3.89 | 15 | 0 | 4.5 |
| Tennessee | proportion | Each county counts once | −10.64 | −36.17 | 25.12 | 11 | 1 | 5.5 |
| Texas | proportion | Each county counts once | −11.50 | −25.12 | 4.60 | 36 | 3 | 1.3 |
| Virginia | proportion | Each county counts once | −11.24 | −25.06 | 5.14 | 11 | 4 | 2.8 |
| Washington | proportion | Each county counts once | −13.76 | −38.91 | 21.75 | 12 | 0 | 3.7 |
| Wisconsin | proportion | Each county counts once | −17.17 | −34.81 | 5.24 | 15 | 0 | 5.5 |

Views: Points, weighted by births (measure = points, weighting = Weighted by births); Points, each county once (measure = points, weighting = Each county counts once); Percent, weighted by births (measure = proportion, weighting = Weighted by births); Percent, each county once (measure = proportion, weighting = Each county counts once).

Note: Large counties are the counties that CDC WONDER names individually: those with 100,000 residents or more. WONDER pools the births of every smaller county in a state into one row, so smaller counties are not in these data. Of the 578 large counties named in every year from 2018 to 2024, California (35) and 13 counties with a withheld smoking count are left out, which leaves 530.

Note: County median household income is the County Health Rankings 2024 release, a Census Small Area Income and Poverty Estimates value for 2022. Its year differs from the births, which are from 2018 to 2024.

Note: A state gets a slope when it has at least 8 complete counties, which gives 26 states and 449 counties. Intervals are 95% intervals from a t distribution with the state's counties minus two degrees of freedom and heteroskedasticity-robust (HC3) standard errors, so a state with few counties has a wide interval and the interval may be too narrow where one county dominates.

Note: Virginia has 4 of its 15 large counties left out for a withheld smoking count, and counties left out in other states are listed in the data table. A state with counties left out rests on the rest.

Note: Counties differ by less than a doubling of income in a typical state: the median span between a state's lowest and highest county income is 0.9 doublings, so the slopes extend beyond what most states' own counties cover.

Source: CDC WONDER natality, tobacco use during pregnancy, births in 2018-2024; full counts of births; counties ranked on County Health Rankings 2024 median household income (Census SAIPE 2022); state slopes weighted by births with a known answer or unweighted as labeled, 95% t intervals

Data: https://vishalsingh.org/zipcode/report/smoking-in-pregnancy/figures/state_slopes.json

### Table 6. States with a higher share of births to mothers who smoked have steeper slopes in points, a correlation of −0.62, with no detectable relation in proportion, where the correlation is −0.19

Slope of the smoking share on county median household income against the state's pooled smoking share, one dot per state with a slope of its own, 2018 to 2024; slopes in percentage points per doubling of income or as a percent difference per 10% higher income

| births | level | measure | slope | state |
| ---: | ---: | --- | ---: | --- |
| 245196 | 4.57 | points | −3.28 | Alabama |
| 531425 | 3.19 | points | −4.97 | Arizona |
| 381504 | 3.68 | points | −4.35 | Colorado |
| 242379 | 2.62 | points | −6.42 | Connecticut |
| 1416167 | 2.71 | points | −3.97 | Florida |
| 565701 | 1.80 | points | −3.34 | Georgia |
| 777655 | 2.90 | points | −8.37 | Illinois |
| 359143 | 6.66 | points | −7.17 | Indiana |
| 269558 | 3.08 | points | −0.810 | Louisiana |
| 437262 | 3.07 | points | −4.04 | Maryland |
| 471835 | 2.79 | points | −4.24 | Massachusetts |
| 580836 | 6.58 | points | −7.45 | Michigan |
| 297960 | 3.72 | points | −7.80 | Minnesota |
| 279141 | 7.17 | points | −6.23 | Missouri |
| 691343 | 1.88 | points | −1.86 | New Jersey |
| 1319857 | 2.30 | points | −2.36 | New York |
| 621319 | 4.22 | points | −8.20 | North Carolina |
| 704336 | 8.32 | points | −7.34 | Ohio |
| 214801 | 4.75 | points | −9.98 | Oregon |
| 802576 | 6.36 | points | −5.10 | Pennsylvania |
| 312029 | 4.49 | points | −4.51 | South Carolina |
| 336327 | 5.53 | points | −0.910 | Tennessee |
| 2167850 | 1.27 | points | −0.630 | Texas |
| 249880 | 2.82 | points | −2.72 | Virginia |
| 491503 | 3.66 | points | −5.69 | Washington |
| 284970 | 5.53 | points | −3.37 | Wisconsin |
| 245196 | 4.57 | proportion | −7.62 | Alabama |
| 531425 | 3.19 | proportion | −13.80 | Arizona |
| 381504 | 3.68 | proportion | −14.86 | Colorado |
| 242379 | 2.62 | proportion | −37.17 | Connecticut |
| 1416167 | 2.71 | proportion | −9.96 | Florida |
| 565701 | 1.80 | proportion | −20.60 | Georgia |
| 777655 | 2.90 | proportion | −24.70 | Illinois |
| 359143 | 6.66 | proportion | −21.06 | Indiana |
| 269558 | 3.08 | proportion | −2.35 | Louisiana |
| 437262 | 3.07 | proportion | −17.27 | Maryland |
| 471835 | 2.79 | proportion | −17.13 | Massachusetts |
| 580836 | 6.58 | proportion | −15.68 | Michigan |
| 297960 | 3.72 | proportion | −18.02 | Minnesota |
| 279141 | 7.17 | proportion | −12.11 | Missouri |
| 691343 | 1.88 | proportion | −11.04 | New Jersey |
| 1319857 | 2.30 | proportion | −7.29 | New York |
| 621319 | 4.22 | proportion | −26.20 | North Carolina |
| 704336 | 8.32 | proportion | −10.74 | Ohio |
| 214801 | 4.75 | proportion | −25.39 | Oregon |
| 802576 | 6.36 | proportion | −9.71 | Pennsylvania |
| 312029 | 4.49 | proportion | −9.25 | South Carolina |
| 336327 | 5.53 | proportion | 5.28 | Tennessee |
| 2167850 | 1.27 | proportion | −0.520 | Texas |
| 249880 | 2.82 | proportion | −16.02 | Virginia |
| 491503 | 3.66 | proportion | −21.40 | Washington |
| 284970 | 5.53 | proportion | −10.92 | Wisconsin |

Views: Percentage points per doubling (measure = points); Percent per 10 percent higher income (measure = proportion).

Note: Large counties are the counties that CDC WONDER names individually: those with 100,000 residents or more. WONDER pools the births of every smaller county in a state into one row, so smaller counties are not in these data. Of the 578 large counties named in every year from 2018 to 2024, California (35) and 13 counties with a withheld smoking count are left out, which leaves 530.

Note: County median household income is the County Health Rankings 2024 release, a Census Small Area Income and Poverty Estimates value for 2022. Its year differs from the births, which are from 2018 to 2024.

Note: Dots are the same 26 states as in the previous figure, with counties weighted by births, and the state slopes carry the wide intervals shown there; the correlation interval allows for the states drawn.

Note: The state's smoking share is the births to mothers who reported smoking over births with a known answer, pooled over its complete counties.

Source: CDC WONDER natality, tobacco use during pregnancy, births in 2018-2024; full counts of births; counties ranked on County Health Rankings 2024 median household income (Census SAIPE 2022); state slopes weighted by births with a known answer; correlations weighted by births, intervals from a bootstrap over states

Data: https://vishalsingh.org/zipcode/report/smoking-in-pregnancy/figures/slope_vs_state_level.json

### Table 7. Raising the minimum number of counties from 3 to 10 leaves 9 to 11 states with an interval below zero and none above it, and the test of equal proportional slopes never rejects

States with a slope of their own, states with an interval below or above zero, and wild-bootstrap p-values for equal slopes in percentage points and in proportion, by the fewest complete counties a state needs, 2018 to 2024

| Fewest complete counties | States | Counties | States with interval below zero | States with interval above zero | p, points, weighted | p, points, each county once | p, proportion, weighted | p, proportion, each county once |
| ---: | ---: | ---: | ---: | ---: | --- | --- | --- | --- |
| 3 | 41 | 518 | 11 | 0 | 0.109 | 0.264 | 0.788 | 0.784 |
| 4 | 37 | 506 | 11 | 0 | 0.121 | 0.006 | 0.790 | 0.710 |
| 5 | 35 | 498 | 11 | 0 | 0.118 | 0.006 | 0.743 | 0.648 |
| 6 | 30 | 473 | 10 | 0 | 0.117 | 0.006 | 0.639 | 0.344 |
| 8 | 26 | 449 | 10 | 0 | 0.108 | 0.009 | 0.664 | 0.343 |
| 10 | 23 | 423 | 9 | 0 | 0.165 | 0.0005 | 0.601 | 0.319 |

Note: A p-value tests that every state in the row has the same slope. It is the share of wild-bootstrap draws (replicates and seed in the methods note) in which Cochran's Q is at least as large as observed, so the smallest value it can take is one over the number of draws plus one.

Note: Large counties are the counties that CDC WONDER names individually: those with 100,000 residents or more. WONDER pools the births of every smaller county in a state into one row, so smaller counties are not in these data. Of the 578 large counties named in every year from 2018 to 2024, California (35) and 13 counties with a withheld smoking count are left out, which leaves 530.

Note: County median household income is the County Health Rankings 2024 release, a Census Small Area Income and Poverty Estimates value for 2022. Its year differs from the births, which are from 2018 to 2024.

Source: CDC WONDER natality, tobacco use during pregnancy, births in 2018-2024; full counts of births; counties ranked on County Health Rankings 2024 median household income (Census SAIPE 2022); state slopes weighted by births or unweighted as labeled; p-values from a wild bootstrap

Data: https://vishalsingh.org/zipcode/report/smoking-in-pregnancy/figures/cutoff_sensitivity.json

### Table 8. The within-state slope in points shrank from 4.9 to 2.4 points per doubling of county income from 2018 to 2024 while the slope in proportion grew from 10.1% to 14.7%

Within-state slope of the smoking share on county median household income (SAIPE 2022), by year of birth, 2018 to 2024, in percentage points per doubling of income or as a percent difference per 10% higher income, with 95% intervals from a bootstrap over states

| conf_high | conf_low | measure | n | slope | year |
| ---: | ---: | --- | ---: | ---: | ---: |
| −3.45 | −6.98 | points | 45 | −4.89 | 2018 |
| −3.31 | −6.71 | points | 45 | −4.70 | 2019 |
| −3.25 | −6.62 | points | 45 | −4.65 | 2020 |
| −2.80 | −5.76 | points | 45 | −4.02 | 2021 |
| −2.32 | −4.78 | points | 45 | −3.31 | 2022 |
| −1.99 | −4.03 | points | 45 | −2.81 | 2023 |
| −1.70 | −3.36 | points | 45 | −2.39 | 2024 |
| −6.33 | −15.12 | proportion | 45 | −10.13 | 2018 |
| −6.88 | −15.88 | proportion | 45 | −10.87 | 2019 |
| −7.56 | −16.68 | proportion | 45 | −11.73 | 2020 |
| −7.69 | −17.47 | proportion | 45 | −12.18 | 2021 |
| −9.10 | −18.22 | proportion | 45 | −12.86 | 2022 |
| −10.07 | −18.86 | proportion | 45 | −13.82 | 2023 |
| −11.61 | −19.29 | proportion | 45 | −14.67 | 2024 |

Views: Percentage points per doubling (measure = points); Percent per 10 percent higher income (measure = proportion).

Note: Large counties are the counties that CDC WONDER names individually: those with 100,000 residents or more. WONDER pools the births of every smaller county in a state into one row, so smaller counties are not in these data. Of the 578 large counties named in every year from 2018 to 2024, California (35) and 13 counties with a withheld smoking count are left out, which leaves 530.

Note: County median household income is the County Health Rankings 2024 release, a Census Small Area Income and Poverty Estimates value for 2022. Its year differs from the births, which are from 2018 to 2024.

Note: Intervals are 95% intervals from a bootstrap over states (replicates and seed in the methods note), which allows for counties in one state resembling each other. The same counties are used in every year, and each is published in every year.

Note: The slopes are negative: a value farther below zero means a lower smoking share for a doubling of county income.

Source: CDC WONDER natality, tobacco use during pregnancy, births in 2018-2024; full counts of births; counties ranked on County Health Rankings 2024 median household income (Census SAIPE 2022); slopes weighted by births with a known answer, intervals from a bootstrap over states

Data: https://vishalsingh.org/zipcode/report/smoking-in-pregnancy/figures/slope_over_time.json

## All estimates

Every fact in facts.json (130), sorted by id.

| Fact id | Value | Unit | Estimand | Population | Period | Interval | n | Weights | Source | Note |
| --- | ---: | --- | --- | --- | --- | --- | ---: | --- | --- | --- |
| above_max | 0 | states | most states with a 95% interval above zero across the cutoffs | states with a slope of their own at any cutoff | births in 2018 to 2024 |   |   | none (full count) | CDC WONDER natality | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. |
| adi_income_corr | −0.79 | correlation coefficient | births-weighted correlation of log county median household income with the county ADI rank; bootstrap interval over states | the complete counties | income for 2022; ADI built on ACS 2019-2023 | 95% CI −0.90 to −0.67 | 530 counties | births with a known tobacco answer | Area Deprivation Index | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. |
| adi_slope | 1.3 | percentage points per ten-point step in the county Area Deprivation Index national rank | state fixed-effects slope of the smoking share on the county ADI rank, births-weighted; bootstrap interval over states | live births with a known tobacco answer in the large counties that CDC WONDER names in every year, leaving out California and the counties with a withheld smoking count; slope within states | births in 2018 to 2024 | 95% CI 0.9 to 1.6 | 45 states | births with a known tobacco answer; interval from a bootstrap over states | Area Deprivation Index | County ADI is the population-weighted mean of tract national ranks (1 least deprived, 100 most deprived) from the 2023 Area Deprivation Index with ACS 2019-2023 tract population; its period differs from the births, which are from 2018 to 2024. |
| below_max | 11 | states | most states with a 95% interval below zero across the cutoffs | states with a slope of their own at any cutoff | births in 2018 to 2024 |   |   | none (full count) | CDC WONDER natality | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. |
| below_min | 9 | states | fewest states with a 95% interval below zero across the cutoffs | states with a slope of their own at any cutoff | births in 2018 to 2024 |   |   | none (full count) | CDC WONDER natality | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. |
| big_births_share | 82.2% | percent of births with a known answer | share of births in the larger half of counties by births | live births with a known tobacco answer in the large counties that CDC WONDER names in every year, leaving out California and the counties with a withheld smoking count | births in 2018 to 2024 |   | 530 counties | none (full count) | CDC WONDER natality |   |
| big_slope | 3.7 | percentage points per doubling of county median household income | state fixed-effects slope among the larger half of counties by births with a known answer, each county counting once; bootstrap interval over states | live births with a known tobacco answer in the large counties that CDC WONDER names in every year, leaving out California and the counties with a withheld smoking count; the larger half of counties by births with a known answer | births in 2018 to 2024 | 95% CI 2.6 to 5.1 | 35 states | none (each county counts once); interval from a bootstrap over states | CDC WONDER natality | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. |
| births_known | 17.1 million | live births | count of births with a known tobacco answer (Yes plus No) | live births with a known tobacco answer in the large counties that CDC WONDER names in every year, leaving out California and the counties with a withheld smoking count | births in 2018 to 2024 |   | 530 counties | none (full count) | CDC WONDER natality |   |
| births_smoking | 632,733 | live births | count of births with a Yes answer to tobacco use during pregnancy | live births with a known tobacco answer in the large counties that CDC WONDER names in every year, leaving out California and the counties with a withheld smoking count | births in 2018 to 2024 |   | 530 counties | none (full count) | CDC WONDER natality |   |
| ca_share_2018 | 1.2% | percent of births with a known answer | pooled share over the published county cells of 2018; withheld cells leave both sums | live births in the California counties with a published Yes cell | 2018 |   |   | none (full count) | CDC WONDER natality |   |
| common_pct | 11.9% | percent difference in the smoking share for a 10 percent higher county median household income | state fixed-effects slope in proportion, births-weighted, over the states with a slope of their own; bootstrap interval over states | live births with a known tobacco answer in the large counties that CDC WONDER names in every year, leaving out California and the counties with a withheld smoking count; states with a slope of their own | births in 2018 to 2024 | 95% CI 8.0% to 17.5% | 26 states | births with a known tobacco answer; interval from a bootstrap over states | CDC WONDER natality | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. |
| common_points | 3.8 | percentage points per doubling of county median household income | state fixed-effects slope in points, births-weighted, over the states with a slope of their own; bootstrap interval over states | live births with a known tobacco answer in the large counties that CDC WONDER names in every year, leaving out California and the counties with a withheld smoking count; states with a slope of their own | births in 2018 to 2024 | 95% CI 2.6 to 5.6 | 26 states | births with a known tobacco answer; interval from a bootstrap over states | CDC WONDER natality | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. |
| complete_share | 66.9% | percent of live births in the United States | share of births in the complete counties | live births in the 50 states and DC | births in 2018 to 2024 |   | 530 counties | none (full count) | CDC WONDER natality |   |
| corr_level_prop | −0.19 | correlation coefficient | births-weighted correlation across states between the state's pooled smoking share and its slope in proportion; bootstrap interval over states | states with a slope of their own | births in 2018 to 2024 | 95% CI −0.56 to 0.37 | 26 states | births with a known tobacco answer in the state; interval from a bootstrap over states | CDC WONDER natality | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. |
| corr_level_pts | −0.62 | correlation coefficient | births-weighted correlation across states between the state's pooled smoking share and its slope in points; bootstrap interval over states | states with a slope of their own | births in 2018 to 2024 | 95% CI −0.83 to −0.20 | 26 states | births with a known tobacco answer in the state; interval from a bootstrap over states | CDC WONDER natality | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. |
| counties_k | 449 | counties | count of counties | complete counties in the states with a slope of their own | births in 2018 to 2024 |   |   | none (full count) | CDC WONDER natality | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. |
| county_p5 | 0.8% | percent of births with a known answer | 5th percentile of the county shares, one county one count | live births with a known tobacco answer in the large counties that CDC WONDER names in every year, leaving out California and the counties with a withheld smoking count | births in 2018 to 2024 |   | 530 counties | none (each county counts once) | CDC WONDER natality |   |
| county_p95 | 13.2% | percent of births with a known answer | 95th percentile of the county shares, one county one count | live births with a known tobacco answer in the large counties that CDC WONDER names in every year, leaving out California and the counties with a withheld smoking count | births in 2018 to 2024 |   | 530 counties | none (each county counts once) | CDC WONDER natality |   |
| cut_hi | 10 | counties | largest cutoff tried (a choice) | complete counties in a state | births in 2018 to 2024 |   |   | none (full count) | CDC WONDER natality |   |
| cut_lo | 3 | counties | smallest cutoff tried (a choice) | complete counties in a state | births in 2018 to 2024 |   |   | none (full count) | CDC WONDER natality |   |
| cut_low_mid | $64,868 | dollars | highest county median household income in the lowest-income third of the complete counties | the complete counties | income for 2022 |   | 530 counties | none (county values) | County Health Rankings | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. |
| cut_mid_high | $80,017 | dollars | highest county median household income in the middle third of the complete counties | the complete counties | income for 2022 |   | 530 counties | none (county values) | County Health Rankings | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. |
| dec10_income | $107,649 | dollars | median of the county median household incomes in the tenth | the highest-income tenth of the complete counties | income for 2022 |   | 53 counties | none (county values) | County Health Rankings | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. |
| dec10_share | 1.7% | percent of births with a known answer | pooled share of births to mothers who reported smoking; bootstrap interval over counties | live births with a known tobacco answer in the large counties that CDC WONDER names in every year, leaving out California and the counties with a withheld smoking count; highest-income tenth of counties by median household income | births in 2018 to 2024 | 95% CI 1.4% to 2.1% | 53 counties | none (full counts of births); interval from a bootstrap over counties | CDC WONDER natality | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. |
| dec1_income | $54,059 | dollars | median of the county median household incomes in the tenth | the lowest-income tenth of the complete counties | income for 2022 |   | 53 counties | none (county values) | County Health Rankings | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. |
| dec1_share | 5.7% | percent of births with a known answer | pooled share of births to mothers who reported smoking; bootstrap interval over counties | live births with a known tobacco answer in the large counties that CDC WONDER names in every year, leaving out California and the counties with a withheld smoking count; lowest-income tenth of counties by median household income | births in 2018 to 2024 | 95% CI 4.2% to 8.0% | 53 counties | none (full counts of births); interval from a bootstrap over counties | CDC WONDER natality | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. |
| dec_income_ratio | 2.0 | ratio of incomes | median county income in the highest-income tenth over the median in the lowest-income tenth | the lowest-income and highest-income tenths of the complete counties | income for 2022 |   | 106 counties | none (county values) | County Health Rankings | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. |
| decline_all | 64.6% | percent of the 2018 share | fall in the pooled share from 2018 to 2024; bootstrap interval over counties | live births with a known tobacco answer in the large counties that CDC WONDER names in every year, leaving out California and the counties with a withheld smoking count | 2018 to 2024 | 95% CI 63.5% to 65.7% | 530 counties | none (full counts of births); interval from a bootstrap over counties | CDC WONDER natality |   |
| decline_terc_1 | 60.0% | percent of the 2018 share | fall in the pooled share from 2018 to 2024; bootstrap interval over counties | live births with a known tobacco answer in the large counties that CDC WONDER names in every year, leaving out California and the counties with a withheld smoking count; the lowest-income third of the complete counties by median household income | 2018 to 2024 | 95% CI 57.7% to 62.3% | 176 counties | none (full counts of births); interval from a bootstrap over counties | CDC WONDER natality | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. |
| decline_terc_3 | 68.8% | percent of the 2018 share | fall in the pooled share from 2018 to 2024; bootstrap interval over counties | live births with a known tobacco answer in the large counties that CDC WONDER names in every year, leaving out California and the counties with a withheld smoking count; the highest-income third of the complete counties by median household income | 2018 to 2024 | 95% CI 67.0% to 70.4% | 177 counties | none (full counts of births); interval from a bootstrap over counties | CDC WONDER natality | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. |
| dropped_high | 11 | counties | count of those counties with a median household income above the cut for the highest-income third of the complete counties | named counties outside California with a withheld smoking count | births in 2018 to 2024 |   | 13 counties | none (full count) | CDC WONDER natality | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. |
| dropped_high_cells | 29 | county-year cells | count of withheld cells | withheld Yes cells of the counties left out that are above the cut for the highest-income third | births in 2018 to 2024 |   |   | none (full count) | CDC WONDER natality |   |
| dropped_high_max_share | 0.17% | percent of smoking births | largest share of those births that the withheld cells of the high-income counties left out could add, at 9 births each | smoking births in the highest-income third of the complete counties | births in 2018 to 2024 |   | 177 counties | none (full count) | CDC WONDER natality | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. |
| dropped_income_median | $109,358 | dollars | median household income of the counties left out for a withheld count | named counties outside California | income for 2022 |   | 13 counties | none (county values) | County Health Rankings | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. |
| dropped_max_births | 315 | live births | largest number of births those cells could hold, at 9 births each | the withheld Yes cells of the counties left out outside California | births in 2018 to 2024 |   |   | none (full count) | CDC WONDER natality |   |
| dropped_max_share | 0.05% | percent of smoking births | largest share of the smoking births in the complete counties that the withheld cells could add, at 9 births each | live births with a known tobacco answer in the large counties that CDC WONDER names in every year, leaving out California and the counties with a withheld smoking count | births in 2018 to 2024 |   | 530 counties | none (full count) | CDC WONDER natality |   |
| dropped_other_states | 9 | counties | count of counties left out for a withheld smoking count | named counties outside California and Virginia | births in 2018 to 2024 |   |   | none (full count) | CDC WONDER natality |   |
| gap_2018 | 4.6 | percentage points | difference of two pooled shares; shared bootstrap draws over counties | live births with a known tobacco answer in the large counties that CDC WONDER names in every year, leaving out California and the counties with a withheld smoking count; lowest-income third minus highest-income third of counties | 2018 | 95% CI 3.6 to 5.6 | 530 counties | none (full counts of births); interval from a bootstrap over counties | CDC WONDER natality | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. |
| gap_2024 | 2.1 | percentage points | difference of two pooled shares; shared bootstrap draws over counties | live births with a known tobacco answer in the large counties that CDC WONDER names in every year, leaving out California and the counties with a withheld smoking count; lowest-income third minus highest-income third of counties | 2024 | 95% CI 1.7 to 2.6 | 530 counties | none (full counts of births); interval from a bootstrap over counties | CDC WONDER natality | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. |
| gap_change | 2.4 | percentage points | fall in the difference of two pooled shares; shared bootstrap draws over counties | live births with a known tobacco answer in the large counties that CDC WONDER names in every year, leaving out California and the counties with a withheld smoking count; lowest-income third minus highest-income third of counties | 2018 to 2024 | 95% CI 1.8 to 3.1 | 530 counties | none (full counts of births); interval from a bootstrap over counties | CDC WONDER natality | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. |
| i2_pu | 70% | percent of the variation in state slopes beyond sampling error | I-squared of Cochran's Q for state slopes in points, each county counts once | states with a slope of their own | births in 2018 to 2024 |   | 26 states | none (each county counts once) | CDC WONDER natality | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. |
| i2_pw | 52% | percent of the variation in state slopes beyond sampling error | I-squared of Cochran's Q for state slopes in points, weighted by births with a known tobacco answer | states with a slope of their own | births in 2018 to 2024 |   | 26 states | births with a known tobacco answer | CDC WONDER natality | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. |
| income_step | 10% | percent | the step in county income that the proportional slopes use (a definition) | county median household income | all years (a definition) |   |   | none (definition) | County Health Rankings |   |
| k_floor | 8 | counties | fewest complete counties a state needs for a slope of its own (a choice) | complete counties in a state | births in 2018 to 2024 |   |   | none (full count) | CDC WONDER natality |   |
| kept_income_median | $72,077 | dollars | median of the county median household incomes of the complete counties | named counties outside California | income for 2022 |   | 530 counties | none (county values) | County Health Rankings | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. |
| lien_lbw_factor | 2 | times the chance of low birth weight | instrumental-variable estimate, pooled across the tax experiments, of the multiple by which smoking in pregnancy raises the chance of low birth weight; the abstract states that smoking 'doubles the chance' | births in the states and control states of Lien and Evans (2005) | study period of Lien and Evans (2005) |   |   | not reported in the abstract | Published estimates from cited studies |   |
| lien_states | 4 | states | number of states with large cigarette tax increases compared with matched control states in Lien and Evans (2005), as reported in the paper's abstract | states that adopted large cigarette tax increases in the study of Lien and Evans (2005) | study period of Lien and Evans (2005) |   | 4 states | none (count of states studied) | Published estimates from cited studies |   |
| map_gray | 2,613 | counties | count of counties drawn gray | counties on the book's county map | births in 2018 to 2024 |   | 2,613 counties | none (full count) | CDC WONDER natality |   |
| map_gray_off | 2,563 | counties | count of gray counties that WONDER does not name (below its size threshold) | counties on the book's county map | births in 2018 to 2024 |   | 2,563 counties | none (full count) | CDC WONDER natality |   |
| markowitz_high | 0.9 | percentage points | Markowitz and coauthors (2017), Table 2: difference-in-differences coefficient of tobacco use during pregnancy on a high state Earned Income Tax Credit without a refund, relative to states with none; the coefficient is +0.009 with a standard error of 0.005, significant at the 10 percent level only | singleton births in 1995 to 2013 to mothers aged 18 or older with a high-school education or less, in the 50 states and DC except Maryland, from birth certificates | births in 1995 to 2013 |   | 25,998,369 births | none (all births) | Published estimates from cited studies |   |
| markowitz_low | 1.6 | percentage points | Markowitz and coauthors (2017), Table 2: difference-in-differences coefficient (linear probability model, state and year-by-quarter fixed effects, standard errors clustered by state) of tobacco use during pregnancy on a low state Earned Income Tax Credit without a refund, relative to states with none; the coefficient is -0.016 with a standard error of 0.004 | singleton births in 1995 to 2013 to mothers aged 18 or older with a high-school education or less, in the 50 states and DC except Maryland, from birth certificates | births in 1995 to 2013 |   | 25,998,369 births | none (all births) | Published estimates from cited studies |   |
| n_ca | 35 | counties | count of counties left out for partial reporting of the tobacco item | California counties among the named counties | births in 2018 to 2024 |   |   | none (full count) | CDC WONDER natality |   |
| n_complete | 530 | counties | count of complete counties | named counties outside California with a published smoking count in every year | births in 2018 to 2024 |   |   | none (full count) | CDC WONDER natality |   |
| n_dropped | 13 | counties | count of counties with a withheld smoking count (fewer than 10 smoking births) in at least one year, left out of every estimate | named counties outside California | births in 2018 to 2024 |   |   | none (full count) | CDC WONDER natality |   |
| n_fe_states | 45 | states | count of states that contribute to a within-state slope | states and the District of Columbia with at least two complete counties | births in 2018 to 2024 |   |   | none (full count) | CDC WONDER natality |   |
| n_named | 578 | counties | count of counties | counties that WONDER natality names in every year from 2018 to 2024 and that have births | births in 2018 to 2024 |   |   | none (full count) | CDC WONDER natality |   |
| n_states_complete | 49 | states | count of states and the District of Columbia | the 50 states and the District of Columbia, except California, with at least one complete county | births in 2018 to 2024 |   |   | none (full count) | CDC WONDER natality |   |
| named_share | 79.5% | percent of live births in the United States | share of births in the named counties | live births in the 50 states and DC | births in 2018 to 2024 |   | 578 counties | none (full count) | CDC WONDER natality |   |
| nr_cells | 4,046 | county-years | count of county-years in which the Not Reported category is a published zero | named counties, 2018 to 2024, in the tobacco cut | births in 2018 to 2024 |   | 578 counties | none (full count) | CDC WONDER natality |   |
| p_het_pu | 0.009 | p-value | wild-bootstrap p-value of Cochran's Q for equal state slopes in percentage points, each county counts once | states with a slope of their own | births in 2018 to 2024 |   | 26 states | none (each county counts once) | CDC WONDER natality | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. |
| p_het_pw | 0.108 | p-value | wild-bootstrap p-value of Cochran's Q for equal state slopes in percentage points, weighted by births with a known tobacco answer | states with a slope of their own | births in 2018 to 2024 |   | 26 states | births with a known tobacco answer | CDC WONDER natality | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. |
| p_het_ru | 0.343 | p-value | wild-bootstrap p-value of Cochran's Q for equal state slopes in proportion (ln share), each county counts once | states with a slope of their own | births in 2018 to 2024 |   | 26 states | none (each county counts once) | CDC WONDER natality | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. |
| p_het_rw | 0.664 | p-value | wild-bootstrap p-value of Cochran's Q for equal state slopes in proportion (ln share), weighted by births with a known tobacco answer | states with a slope of their own | births in 2018 to 2024 |   | 26 states | births with a known tobacco answer | CDC WONDER natality | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. |
| p_prop_min | 0.319 | p-value | wild-bootstrap p-value of Cochran's Q for equal state slopes: smallest p-value across the cutoffs and both weightings, slopes in proportion | states with a slope of their own | births in 2018 to 2024 |   |   | see estimand | CDC WONDER natality | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. |
| p_pu_cut_lo | 0.264 | p-value | wild-bootstrap p-value of Cochran's Q for equal state slopes: p-value at the smallest cutoff, slopes in points, each county counting once | states with a slope of their own | births in 2018 to 2024 |   |   | see estimand | CDC WONDER natality | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. |
| p_pu_max | 0.009 | p-value | wild-bootstrap p-value of Cochran's Q for equal state slopes: largest p-value across the cutoffs above the smallest, slopes in points, each county counting once | states with a slope of their own | births in 2018 to 2024 |   |   | see estimand | CDC WONDER natality | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. |
| p_pw_max | 0.165 | p-value | wild-bootstrap p-value of Cochran's Q for equal state slopes: largest p-value across the cutoffs, slopes in points, weighted by births | states with a slope of their own | births in 2018 to 2024 |   |   | see estimand | CDC WONDER natality | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. |
| p_pw_min | 0.108 | p-value | wild-bootstrap p-value of Cochran's Q for equal state slopes: smallest p-value across the cutoffs, slopes in points, weighted by births | states with a slope of their own | births in 2018 to 2024 |   |   | see estimand | CDC WONDER natality | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. |
| pct_within_u | 16.7% | percent difference in the smoking share for a 10 percent higher county median household income | state fixed-effects slope of ln(smoking share) on log2 county income, each county counts once, expressed for a 10 percent higher income; bootstrap interval over states | live births with a known tobacco answer in the large counties that CDC WONDER names in every year, leaving out California and the counties with a withheld smoking count; slope within states | births in 2018 to 2024 | 95% CI 14.4% to 19.0% | 45 states | none (each county counts once); interval from a bootstrap over states | CDC WONDER natality | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. |
| pct_within_w | 12.0% | percent difference in the smoking share for a 10 percent higher county median household income | state fixed-effects slope of ln(smoking share) on log2 county income, weighted by births with a known tobacco answer, expressed for a 10 percent higher income; bootstrap interval over states | live births with a known tobacco answer in the large counties that CDC WONDER names in every year, leaving out California and the counties with a withheld smoking count; slope within states | births in 2018 to 2024 | 95% CI 8.2% to 17.0% | 45 states | births with a known tobacco answer; interval from a bootstrap over states | CDC WONDER natality | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. |
| ratio_2018 | 2.3 | ratio of shares | ratio of two pooled shares; shared bootstrap draws over counties | live births with a known tobacco answer in the large counties that CDC WONDER names in every year, leaving out California and the counties with a withheld smoking count; lowest-income third over highest-income third of counties | 2018 | 95% CI 1.9 to 2.7 | 530 counties | none (full counts of births); interval from a bootstrap over counties | CDC WONDER natality | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. |
| ratio_2024 | 2.9 | ratio of shares | ratio of two pooled shares; shared bootstrap draws over counties | live births with a known tobacco answer in the large counties that CDC WONDER names in every year, leaving out California and the counties with a withheld smoking count; lowest-income third over highest-income third of counties | 2024 | 95% CI 2.4 to 3.5 | 530 counties | none (full counts of births); interval from a bootstrap over counties | CDC WONDER natality | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. |
| ratio_change | 0.6 | ratio of shares | rise in the ratio of two pooled shares; shared bootstrap draws over counties | live births with a known tobacco answer in the large counties that CDC WONDER names in every year, leaving out California and the counties with a withheld smoking count; lowest-income third over highest-income third of counties | 2018 to 2024 | 95% CI 0.4 to 0.9 | 530 counties | none (full counts of births); interval from a bootstrap over counties | CDC WONDER natality | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. |
| sens_added_2018 | 9 | counties | count of counties added back | named counties outside California with a published Yes count in 2018 that are left out of the complete counties | 2018 |   | 9 counties | none (full count) | CDC WONDER natality |   |
| sens_added_2024 | 1 | counties | count of counties added back | named counties outside California with a published Yes count in 2024 that are left out of the complete counties | 2024 |   | 1 counties | none (full count) | CDC WONDER natality |   |
| sens_slope_2018 | 4.8 | percentage points per doubling of county median household income | state fixed-effects slope of the smoking share on log2 county income, births-weighted, adding back counties left out only for a withheld count in other years; bootstrap interval over states | live births with a known tobacco answer in the large counties that CDC WONDER names in every year, leaving out California and the counties with a withheld smoking count; counties outside California with a published Yes count in 2018, slope within states | 2018 | 95% CI 3.4 to 6.8 | 45 states | births with a known tobacco answer; interval from a bootstrap over states | CDC WONDER natality | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. |
| share_all_2018 | 5.4% | percent of births with a known answer | pooled share of births to mothers who reported smoking; bootstrap interval over counties | live births with a known tobacco answer in the large counties that CDC WONDER names in every year, leaving out California and the counties with a withheld smoking count | 2018 | 95% CI 4.9% to 5.9% | 530 counties | none (full counts of births); interval from a bootstrap over counties | CDC WONDER natality |   |
| share_all_2024 | 1.9% | percent of births with a known answer | pooled share of births to mothers who reported smoking; bootstrap interval over counties | live births with a known tobacco answer in the large counties that CDC WONDER names in every year, leaving out California and the counties with a withheld smoking count | 2024 | 95% CI 1.7% to 2.1% | 530 counties | none (full counts of births); interval from a bootstrap over counties | CDC WONDER natality |   |
| slope_between_u | 5.3 | percentage points per doubling of county median household income | least-squares slope of state means of the smoking share (percent) on log2 county income, each county counts once; bootstrap interval over states | live births with a known tobacco answer in the large counties that CDC WONDER names in every year, leaving out California and the counties with a withheld smoking count; slope between states | births in 2018 to 2024 | 95% CI 2.2 to 8.7 | 49 states | none (each county counts once); interval from a bootstrap over states | CDC WONDER natality | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. |
| slope_between_w | 4.7 | percentage points per doubling of county median household income | least-squares slope of state means of the smoking share (percent) on log2 county income, weighted by births with a known tobacco answer; bootstrap interval over states | live births with a known tobacco answer in the large counties that CDC WONDER names in every year, leaving out California and the counties with a withheld smoking count; slope between states | births in 2018 to 2024 | 95% CI 1.8 to 8.3 | 49 states | births with a known tobacco answer; interval from a bootstrap over states | CDC WONDER natality | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. |
| slope_pct_2018 | 10.1% | percent difference in the smoking share for a 10 percent higher county median household income | state fixed-effects slope of ln(smoking share) on log2 county income, births-weighted, expressed for a 10 percent higher income; bootstrap interval over states | live births with a known tobacco answer in the large counties that CDC WONDER names in every year, leaving out California and the counties with a withheld smoking count; slope within states | 2018 | 95% CI 6.3% to 15.1% | 45 states | births with a known tobacco answer; interval from a bootstrap over states | CDC WONDER natality | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. |
| slope_pct_2024 | 14.7% | percent difference in the smoking share for a 10 percent higher county median household income | state fixed-effects slope of ln(smoking share) on log2 county income, births-weighted, expressed for a 10 percent higher income; bootstrap interval over states | live births with a known tobacco answer in the large counties that CDC WONDER names in every year, leaving out California and the counties with a withheld smoking count; slope within states | 2024 | 95% CI 11.6% to 19.3% | 45 states | births with a known tobacco answer; interval from a bootstrap over states | CDC WONDER natality | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. |
| slope_pct_change | 4.5 | percentage points of the percent difference in the smoking share for a 10 percent higher county median household income | fall from 2018 to 2024 in the state fixed-effects proportional slope (a steeper slope), births-weighted; shared bootstrap draws over states | live births with a known tobacco answer in the large counties that CDC WONDER names in every year, leaving out California and the counties with a withheld smoking count; slope within states | 2018 to 2024 | 95% CI 3.3 to 5.8 | 45 states | births with a known tobacco answer; interval from a bootstrap over states | CDC WONDER natality | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. |
| slope_pooled_u | 6.2 | percentage points per doubling of county median household income | least-squares slope of the smoking share (percent) on log2 county income, each county counts once; bootstrap interval over states | live births with a known tobacco answer in the large counties that CDC WONDER names in every year, leaving out California and the counties with a withheld smoking count; slope across all counties | births in 2018 to 2024 | 95% CI 4.6 to 8.0 | 49 states | none (each county counts once); interval from a bootstrap over states | CDC WONDER natality | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. |
| slope_pooled_w | 4.1 | percentage points per doubling of county median household income | least-squares slope of the smoking share (percent) on log2 county income, weighted by births with a known tobacco answer; bootstrap interval over states | live births with a known tobacco answer in the large counties that CDC WONDER names in every year, leaving out California and the counties with a withheld smoking count; slope across all counties | births in 2018 to 2024 | 95% CI 2.7 to 5.9 | 49 states | births with a known tobacco answer; interval from a bootstrap over states | CDC WONDER natality | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. |
| slope_pts_2018 | 4.9 | percentage points per doubling of county median household income | state fixed-effects slope of the smoking share (percent) on log2 county income, births-weighted; bootstrap interval over states | live births with a known tobacco answer in the large counties that CDC WONDER names in every year, leaving out California and the counties with a withheld smoking count; slope within states | 2018 | 95% CI 3.5 to 7.0 | 45 states | births with a known tobacco answer; interval from a bootstrap over states | CDC WONDER natality | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. |
| slope_pts_2024 | 2.4 | percentage points per doubling of county median household income | state fixed-effects slope of the smoking share (percent) on log2 county income, births-weighted; bootstrap interval over states | live births with a known tobacco answer in the large counties that CDC WONDER names in every year, leaving out California and the counties with a withheld smoking count; slope within states | 2024 | 95% CI 1.7 to 3.4 | 45 states | births with a known tobacco answer; interval from a bootstrap over states | CDC WONDER natality | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. |
| slope_pts_change | 2.5 | percentage points per doubling of county median household income | rise from 2018 to 2024 in the state fixed-effects slope in points (a shallower slope), births-weighted; shared bootstrap draws over states | live births with a known tobacco answer in the large counties that CDC WONDER names in every year, leaving out California and the counties with a withheld smoking count; slope within states | 2018 to 2024 | 95% CI 1.7 to 3.7 | 45 states | births with a known tobacco answer; interval from a bootstrap over states | CDC WONDER natality | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. |
| slope_within_u | 6.6 | percentage points per doubling of county median household income | state fixed-effects slope of the smoking share (percent) on log2 county income, each county counts once; bootstrap interval over states | live births with a known tobacco answer in the large counties that CDC WONDER names in every year, leaving out California and the counties with a withheld smoking count; slope within states | births in 2018 to 2024 | 95% CI 5.1 to 8.2 | 45 states | none (each county counts once); interval from a bootstrap over states | CDC WONDER natality | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. |
| slope_within_w | 3.9 | percentage points per doubling of county median household income | state fixed-effects slope of the smoking share (percent) on log2 county income, weighted by births with a known tobacco answer; bootstrap interval over states | live births with a known tobacco answer in the large counties that CDC WONDER names in every year, leaving out California and the counties with a withheld smoking count; slope within states | births in 2018 to 2024 | 95% CI 2.7 to 5.5 | 45 states | births with a known tobacco answer; interval from a bootstrap over states | CDC WONDER natality | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. |
| small_slope | 8.3 | percentage points per doubling of county median household income | state fixed-effects slope among the smaller half of counties by births with a known answer, each county counting once; bootstrap interval over states | live births with a known tobacco answer in the large counties that CDC WONDER names in every year, leaving out California and the counties with a withheld smoking count; the smaller half of counties by births | births in 2018 to 2024 | 95% CI 6.3 to 10.2 | 42 states | none (each county counts once); interval from a bootstrap over states | CDC WONDER natality | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. |
| split_diff | 4.6 | percentage points per doubling of county median household income | within-state slope among the larger half of counties minus the slope among the smaller half (a shallower slope in the larger half), each county counting once; shared bootstrap draws over states | live births with a known tobacco answer in the large counties that CDC WONDER names in every year, leaving out California and the counties with a withheld smoking count | births in 2018 to 2024 | 95% CI 2.5 to 6.6 | 49 states | none (each county counts once); interval from a bootstrap over states | CDC WONDER natality | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. |
| states_above | 0 | states | count of states whose 95% interval lies above zero, counties weighted by births | states with a slope of their own | births in 2018 to 2024 |   |   | none (full count) | CDC WONDER natality | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. |
| states_above_u | 0 | states | count of states whose 95% interval lies above zero, each county counting once | states with a slope of their own | births in 2018 to 2024 |   |   | none (full count) | CDC WONDER natality | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. |
| states_below | 10 | states | count of states whose 95% interval lies below zero, counties weighted by births | states with a slope of their own | births in 2018 to 2024 |   |   | none (full count) | CDC WONDER natality | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. |
| states_below_prop | 8 | states | count of states whose 95% interval for the slope in proportion lies below zero, counties weighted by births | states with a slope of their own | births in 2018 to 2024 |   |   | none (full count) | CDC WONDER natality | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. |
| states_below_prop_u | 12 | states | count of states whose 95% interval for the slope in proportion lies below zero, each county counting once | states with a slope of their own | births in 2018 to 2024 |   |   | none (full count) | CDC WONDER natality | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. |
| states_below_u | 12 | states | count of states whose 95% interval lies below zero, each county counting once | states with a slope of their own | births in 2018 to 2024 |   |   | none (full count) | CDC WONDER natality | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. |
| states_cut_hi | 23 | states | count of states with a slope of their own | states at the largest cutoff | births in 2018 to 2024 |   |   | none (full count) | CDC WONDER natality | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. |
| states_cut_lo | 41 | states | count of states with a slope of their own | states at the smallest cutoff | births in 2018 to 2024 |   |   | none (full count) | CDC WONDER natality | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. |
| states_k | 26 | states | count of states with a slope of their own | states with enough complete counties for a slope of their own | births in 2018 to 2024 |   |   | none (full count) | CDC WONDER natality | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. |
| states_neg | 26 | states | count of states whose estimated slope is below zero, counties weighted by births | states with a slope of their own | births in 2018 to 2024 |   |   | none (full count) | CDC WONDER natality | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. |
| states_neg_prop | 25 | states | count of states whose estimated slope in proportion is below zero, counties weighted by births | states with a slope of their own | births in 2018 to 2024 |   |   | none (full count) | CDC WONDER natality | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. |
| states_neg_prop_u | 26 | states | count of states whose estimated slope in proportion is below zero, each county counting once | states with a slope of their own | births in 2018 to 2024 |   |   | none (full count) | CDC WONDER natality | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. |
| states_neg_u | 26 | states | count of states whose estimated slope is below zero, each county counting once | states with a slope of their own | births in 2018 to 2024 |   |   | none (full count) | CDC WONDER natality | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. |
| tau_pu | 3.0 | percentage points per doubling of county median household income | DerSimonian-Laird standard deviation of the true state slopes in points, each county counts once; percentile interval from resampling states | states with a slope of their own | births in 2018 to 2024 | 95% CI 1.6 to 3.6 | 26 states | none (each county counts once) | CDC WONDER natality | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. |
| tau_pw | 1.9 | percentage points per doubling of county median household income | DerSimonian-Laird standard deviation of the true state slopes in points, weighted by births with a known tobacco answer; percentile interval from resampling states | states with a slope of their own | births in 2018 to 2024 | 95% CI 0.0 to 2.7 | 26 states | births with a known tobacco answer | CDC WONDER natality | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. |
| tau_ru | 2.7% | percent of the smoking share for a 10 percent higher county median household income | DerSimonian-Laird standard deviation of the true state slopes of ln share, each county counts once, converted at the 10 percent step; percentile interval from resampling states | states with a slope of their own | births in 2018 to 2024 | 95% CI 0.0% to 7.0% | 26 states | none (each county counts once) | CDC WONDER natality | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. |
| tau_rw | 2.1% | percent of the smoking share for a 10 percent higher county median household income | DerSimonian-Laird standard deviation of the true state slopes of ln share, weighted by births with a known tobacco answer, converted at the 10 percent step; percentile interval from resampling states | states with a slope of their own | births in 2018 to 2024 | 95% CI 0.0% to 6.5% | 26 states | births with a known tobacco answer | CDC WONDER natality | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. |
| terc_1_2018 | 8.2% | percent of births with a known answer | pooled share of births to mothers who reported smoking in a third of counties by income; bootstrap interval over counties | live births with a known tobacco answer in the large counties that CDC WONDER names in every year, leaving out California and the counties with a withheld smoking count; the lowest-income third of the complete counties by median household income | 2018 | 95% CI 7.3% to 9.2% | 176 counties | none (full counts of births); interval from a bootstrap over counties | CDC WONDER natality | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. |
| terc_1_2024 | 3.3% | percent of births with a known answer | pooled share of births to mothers who reported smoking in a third of counties by income; bootstrap interval over counties | live births with a known tobacco answer in the large counties that CDC WONDER names in every year, leaving out California and the counties with a withheld smoking count; the lowest-income third of the complete counties by median household income | 2024 | 95% CI 2.9% to 3.7% | 176 counties | none (full counts of births); interval from a bootstrap over counties | CDC WONDER natality | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. |
| terc_2_2018 | 5.5% | percent of births with a known answer | pooled share of births to mothers who reported smoking in a third of counties by income; bootstrap interval over counties | live births with a known tobacco answer in the large counties that CDC WONDER names in every year, leaving out California and the counties with a withheld smoking count; the middle third of the complete counties by median household income | 2018 | 95% CI 4.4% to 6.7% | 177 counties | none (full counts of births); interval from a bootstrap over counties | CDC WONDER natality | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. |
| terc_2_2024 | 1.9% | percent of births with a known answer | pooled share of births to mothers who reported smoking in a third of counties by income; bootstrap interval over counties | live births with a known tobacco answer in the large counties that CDC WONDER names in every year, leaving out California and the counties with a withheld smoking count; the middle third of the complete counties by median household income | 2024 | 95% CI 1.5% to 2.3% | 177 counties | none (full counts of births); interval from a bootstrap over counties | CDC WONDER natality | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. |
| terc_3_2018 | 3.6% | percent of births with a known answer | pooled share of births to mothers who reported smoking in a third of counties by income; bootstrap interval over counties | live births with a known tobacco answer in the large counties that CDC WONDER names in every year, leaving out California and the counties with a withheld smoking count; the highest-income third of the complete counties by median household income | 2018 | 95% CI 3.2% to 4.0% | 177 counties | none (full counts of births); interval from a bootstrap over counties | CDC WONDER natality | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. |
| terc_3_2024 | 1.1% | percent of births with a known answer | pooled share of births to mothers who reported smoking in a third of counties by income; bootstrap interval over counties | live births with a known tobacco answer in the large counties that CDC WONDER names in every year, leaving out California and the counties with a withheld smoking count; the highest-income third of the complete counties by median household income | 2024 | 95% CI 1.0% to 1.3% | 177 counties | none (full counts of births); interval from a bootstrap over counties | CDC WONDER natality | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. |
| unk_cells | 3,710 | county-year cells | count of cells | Unknown or Not Stated cells of the complete counties | births in 2018 to 2024 |   |   | none (full count) | CDC WONDER natality |   |
| unk_cells_w | 1,794 | county-year cells | count of Unknown or Not Stated cells withheld for holding 1 to 9 births | Unknown or Not Stated cells of the complete counties | births in 2018 to 2024 |   |   | none (full count) | CDC WONDER natality |   |
| unk_pub_share | 0.4% | percent of live births | share of births in published Unknown or Not Stated cells | live births in the complete counties | births in 2018 to 2024 |   | 530 counties | none (full count) | CDC WONDER natality |   |
| unk_w_max_share | 0.1% | percent of live births | largest share of births the withheld Unknown or Not Stated cells could hold, at 9 births each | live births in the complete counties | births in 2018 to 2024 |   |   | none (full count) | CDC WONDER natality |   |
| va_dropped | 4 | counties | count of counties left out for a withheld smoking count | Virginia counties among the named counties | births in 2018 to 2024 |   |   | none (full count) | CDC WONDER natality |   |
| va_named | 15 | counties | count of counties | Virginia counties among the named counties | births in 2018 to 2024 |   |   | none (full count) | CDC WONDER natality |   |
| weight_diff | 2.8 | percentage points per doubling of county median household income | within-state slope weighted by births minus the within-state slope with each county counting once (a shallower slope); shared bootstrap draws over states | live births with a known tobacco answer in the large counties that CDC WONDER names in every year, leaving out California and the counties with a withheld smoking count | births in 2018 to 2024 | 95% CI 1.3 to 3.9 | 45 states | births with a known tobacco answer and none (each county counts once); interval from a bootstrap over states | CDC WONDER natality | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. |
| within_minus_pooled_u | −0.4 | percentage points per doubling of county median household income | within-state slope minus the least-squares slope across the counties of the same states with two or more counties, each county counts once; shared bootstrap draws over states | live births with a known tobacco answer in the large counties that CDC WONDER names in every year, leaving out California and the counties with a withheld smoking count | births in 2018 to 2024 | 95% CI −1.3 to 0.5 | 45 states | none (each county counts once); interval from a bootstrap over states | CDC WONDER natality | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. The comparison slope uses only the states with two or more counties, so it differs a little from the slope across all counties printed beside it. |
| within_minus_pooled_w | +0.3 | percentage points per doubling of county median household income | within-state slope minus the least-squares slope across the counties of the same states with two or more counties, weighted by births with a known tobacco answer; shared bootstrap draws over states | live births with a known tobacco answer in the large counties that CDC WONDER names in every year, leaving out California and the counties with a withheld smoking count | births in 2018 to 2024 | 95% CI −0.7 to 1.1 | 45 states | births with a known tobacco answer; interval from a bootstrap over states | CDC WONDER natality | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. The comparison slope uses only the states with two or more counties, so it differs a little from the slope across all counties printed beside it. |
| wonder_suppression_floor | 10 | births | a cell with fewer births than this is withheld (a definition) | the counties and cells of CDC WONDER natality | all years (a definition) |   |   | none (definition) | CDC WONDER natality |   |
| wonder_threshold | 100,000 | residents | population size at which WONDER natality names a county (a definition) | the counties and cells of CDC WONDER natality | all years (a definition) |   |   | none (definition) | CDC WONDER natality |   |
| x_range_median | 0.9 | doublings of county median household income | median across states of the span between the highest and the lowest county median household income, in doublings (log2) | large counties in the states with a slope of their own | income for 2022 |   | 26 states | none (each state counts once) | County Health Rankings | County median household income is the Census SAIPE estimate for 2022 as reported in the County Health Rankings 2024 release; its year differs from the births, which are from 2018 to 2024. |
| yes_w_ca | 33 | county-year cells | count of Yes cells withheld | Yes cells of the California counties | births in 2018 to 2024 |   |   | none (full count) | CDC WONDER natality |   |
| yes_w_dropped | 35 | county-year cells | count of Yes cells withheld | Yes cells of the named counties outside California that have a withheld cell | births in 2018 to 2024 |   |   | none (full count) | CDC WONDER natality |   |
| yes_w_total | 68 | county-year cells | count of Yes cells withheld for holding 1 to 9 births | Yes cells of the named counties | births in 2018 to 2024 |   |   | none (full count) | CDC WONDER natality |   |

## Definitions

- Suppression: The withholding of a number because the count behind it is too small to publish without risking the identification of a person. A suppressed value is missing, not zero.
- Area Deprivation Index (ADI): A score that ranks neighborhoods by income, education, employment and housing conditions drawn from the American Community Survey. In this book a higher national percentile means a more deprived neighborhood.
- Fixed effects: A regression adjustment that gives each group, such as each county, its own baseline. The estimate then rests on differences within groups.
- Ecological inference: Drawing conclusions about individuals from data about places or groups. A pattern across counties need not hold for the people inside them.

## Methods

This chapter compares the share of births with tobacco use reported during pregnancy across counties and states, using county median household income (a 2022 estimate) as the predictor for births in 2018 to 2024, so the outcome and the predictor come from different periods. The counties are those that CDC WONDER names (100,000 residents or more) in every year and that publish every smoking count: California is left out because it does not routinely collect the item, and counties with a withheld smoking count are left out, which removes counties where smoking in pregnancy was rarest. Slopes are shown weighted by births and with each county counting once, and the two weightings differ in size. A state's own slope rests on a handful of complete counties, so state intervals are wide.

### Which births, counties and years are used

Births are counts from CDC WONDER natality for 2018 to 2024 (2024 is final), by county of the mother's residence and by tobacco use during pregnancy, a birth-certificate item with the answers Yes, No, Unknown or Not Stated and Not Reported.

WONDER names a county only if it has 100,000 residents or more. The universe is the 578 counties named in every year, which hold 79.5% of births in the United States in 2018 to 2024. California (35 counties) is left out because WONDER states that California does not routinely collect the item, and its reported share for 2018 is 1.2% against 5.4% in the counties kept.

WONDER withholds cells of 1 to 9 births. A withheld cell is missing, and nothing is derived from a county total minus the published cells. 13 counties outside California have a withheld Yes cell in at least one year (35 cells, which could hold at most 315 births, or 0.05% of smoking births) and are left out of every estimate, so that no county is pooled over only its published years. The 530 complete counties are therefore selected on having at least 10 smoking births in every year, which removes counties where smoking in pregnancy was rarest.

Explains Table 2.

### How the smoking share and county income are defined

A county's smoking share is the Yes births over the Yes plus No births, so births with an Unknown or Not Stated answer are left out of the denominator; the Unknown cells that WONDER publishes hold 0.4% of births, and 1,794 of 3,710 Unknown cells are withheld and could hold at most 0.1%. Not Reported is a published zero in all 4,046 county-years.

County income is the County Health Rankings 2024 release of median household income, a Census Small Area Income and Poverty Estimates value for 2022; a slope is per doubling (log base 2) of income. A slope in proportion is the slope of the natural log of the share, shown as the proportional difference for a 10% higher income. The income measure is checked against the county Area Deprivation Index, the population-weighted mean of tract national ranks.

Explains Table 1 and Table 3.

### How counties are weighted and mapped

Slopes across counties, within states (state fixed effects) and between states (state means) are least-squares slopes, weighted by births with a known answer or unweighted.

Maps draw every county of the book's county map; counties that WONDER does not name, California counties and counties with a withheld count have no value.

Explains Table 4 and Table 2.

### How each state's slope is estimated

A state's own slope is a weighted least-squares line through its complete counties with HC3 standard errors and a t interval with the counties minus two degrees of freedom, for states with at least 8 complete counties, with cutoffs of 3 to 10 as a check.

Explains Table 5 and Table 7.

### How uncertainty is measured

Shares and differences of shares pool births over counties and carry 95% percentile intervals from a bootstrap that resamples whole counties (2,000 replicates, seed 12345) with the same draws for every statistic, so a difference or a ratio has its own interval.

The slopes across counties, within states and between states carry percentile intervals from a bootstrap that resamples whole states with the same replicate count and seed.

### How the test of equal state slopes is run

The test that every state has the same slope is Cochran's Q of the state slopes with their HC3 variances.

Q is too liberal with a handful of counties per state, so its p-value comes from a restricted wild bootstrap (2,000 draws, seed 12345): residuals from a model with a state intercept and one common slope are multiplied by random signs within county, each state's slope and error are refit, and Q is recomputed.

The spread of the true state slopes (tau) is the DerSimonian-Laird estimate with a percentile interval from resampling states.

Explains Table 6.

## What these data cannot show

The comparison is between counties. A county with higher median household income has a lower smoking share, and that says nothing about the income of the mothers who smoked, which these data do not hold. The pattern is ecological and need not hold for the mothers inside the counties. The county tables staged for this book carry one stratifier each, so mother's education cannot be crossed with the tobacco item.

The item is reported at birth. The fall in the share from 5.4% to 1.9% can reflect less smoking, a change in how or whether the item is recorded, or a change in who gives birth, and these data cannot separate them. The item is named tobacco use and may include products other than cigarettes.

The counties are the 530 large counties that remain after the rules above, and they hold 66.9% of the births in the United States. California is out because its item is not collected routinely, and 13 counties are out because their smoking count is withheld in some year, a rule that selects on the outcome. Smaller counties are not in these data, so the chapter says nothing about them. The income measure is an estimate for 2022 and the births are from 2018 to 2024, so a county's income stands in for its income in earlier years. The Area Deprivation Index describes neighborhoods in 2019 to 2023.

State slopes rest on 8 counties or more, and the states with a slope hold 449 counties in all. The intervals use a t distribution and standard errors that allow for unequal variance, and they can be too narrow where one county dominates the line. The tests have little power to detect moderate differences, their p-values depend on how counties are weighted, and a test that finds no difference does not show equality. A slope per doubling of income is a statement about a difference larger than most states' own counties span. None of this identifies why smoking is less common where incomes are higher, and the studies in the box above are the only evidence here on what changes it.

## References

1. Kind, Amy J. H.; Buckingham, William R. (2018). Making Neighborhood-Disadvantage Metrics Accessible: The Neighborhood Atlas. New England Journal of Medicine 378(26): 2456-2458. https://doi.org/10.1056/NEJMp1802313
2. DerSimonian, Rebecca; Laird, Nan (1986). Meta-analysis in clinical trials. Controlled Clinical Trials 7(3): 177-188. https://doi.org/10.1016/0197-2456(86)90046-2
3. Lien, Diana S.; Evans, William N. (2005). Estimating the Impact of Large Cigarette Tax Hikes: The Case of Maternal Smoking and Infant Birth Weight. Journal of Human Resources 40(2): 373-392. https://doi.org/10.3368/jhr.xl.2.373
4. Markowitz, Sara; Komro, Kelli A.; Livingston, Melvin D.; Lenhart, Otto; Wagenaar, Alexander C. (2017). Effects of state-level Earned Income Tax Credit laws in the U.S. on maternal health behaviors and infant health outcomes. Social Science & Medicine 194: 67-75. https://doi.org/10.1016/j.socscimed.2017.10.016

## Provenance

- Pipeline version: 0.1.0
- Build script: book2/chapters/smoking-in-pregnancy/build.py (sha256 a5062d35cd5331f0196777cdcd918ba2a0cba62dc96eade486c10f530f6b2e9d)
- Built on: 2026-10-09
- Site commit: c2b1ff3309cb4a10833f884da91ac0057a66ad0a

Inputs:

| File | Size (bytes) | SHA-256, first 1 MB | SHA-256, last 1 MB |
| --- | ---: | --- | --- |
| book2/chapters/smoking-in-pregnancy/sip_data.py | 24,571 | 85ab524c0eaa61ed20c058c3505a16de76596a87f6b80e4d6c430577684906bd | 85ab524c0eaa61ed20c058c3505a16de76596a87f6b80e4d6c430577684906bd |
| data/analytic/tract_atlas.parquet | 39,659,925 | 29d5e4d943552d54aeb6ba92aadc939d9bd15bdea553c78357dbd516757be627 | 1e10d1ecf722340f23ee2212d4999a90250f85b65fa89b39e32db06d9472e22e |
| data/staged/births/wonder_natality_county_tobacco_2018_2024.parquet | 62,751 | 6379d39b84b8a4e18190d709e9511be66b829bf96458cf22f6a10eb590655942 | 6379d39b84b8a4e18190d709e9511be66b829bf96458cf22f6a10eb590655942 |
| data/staged/births/wonder_natality_county_total_2018_2024.parquet | 41,774 | e85d351fa72c022a05415d332185113b6badcaff1119da1d6f45c76ffe16d86d | e85d351fa72c022a05415d332185113b6badcaff1119da1d6f45c76ffe16d86d |
| data/staged/crosswalks/ct_tract2020_to_old_county.parquet | 15,842 | 961ef39f31596ef65fca2c67d5f2f63e533c5dc9b49745d3bb6437844e4b831d | 961ef39f31596ef65fca2c67d5f2f63e533c5dc9b49745d3bb6437844e4b831d |
| data/staged/health/chr_county_2024.parquet | 8,776,933 | 619795f7ecfec8aff10694b649eaee5e421dbeffbe4be47af4161b51ceefa5cd | c69357f1bcf630407bb9716cb445618838999558c019aaaf77d4681b6bf45c56 |

Outputs:

| File | SHA-256 |
| --- | --- |
| facts.json | 128b072e108ff98bab967e8063031b21551dd5ab3bc6cf2918cdb9cba0215685 |
| figures/cutoff_sensitivity.json | 9bf7678cb6c2b7281532cc86889a6e85c4b2cbc221aaf329e5e2d378ee417bd4 |
| figures/slope_over_time.json | 22c552ec11a5911831aa56ff64592a637503074d03ad19e46c4aa9ebced39e5f |
| figures/slope_scope.json | ace5141f10c487a76f6db7bda4aeead9fd5daf91ff23b2a7915d3d4d3df4bf4b |
| figures/slope_vs_state_level.json | 6e00e027abf060e3ea68b7473beb1d5ae80bd7c574821db95822a9577dde3596 |
| figures/smoking_by_income_third.json | 916b60b8b87dd77d63ea9399f01dbeb788d0688a0c1d9b7f7ba2ebf3229d0a46 |
| figures/smoking_county_map.json | 11e1e14d276808a8a32c0e0b308a551c8da8baf4d572f719589401f87ca3fa1c |
| figures/smoking_vs_income_counties.json | 12535608e983a077c14ffa39e4fb2771c99568dd51c9d67217e0a6e4c0a1bd56 |
| figures/state_slopes.json | 6cb080958e80681716e748a636b84e1411d3130951cbef9baa0d59e00d2d3b97 |
| meta.json | 0f8a7932d763bf6b1a387af88314667d476d97ee553ec406dec3cf16289b5c32 |
| public/data/smoking-in-pregnancy/smoking_county_map.json | 4fa8bb6972d26a2b55aab377d7b36439403dd9de1682680a8138cd9ff352bf26 |
| refs.json | 9a26cfa02db10a18a31e2564582d45b961c08791d993ff5a7f9f522a74e1fc25 |

## Machine formats

- Report (HTML): https://vishalsingh.org/zipcode/smoking-in-pregnancy/report
- Report (Markdown): https://vishalsingh.org/zipcode/smoking-in-pregnancy/report.md
- Every estimate (JSON): https://vishalsingh.org/zipcode/report/smoking-in-pregnancy/facts.json
- Table 1, figure "smoking_by_income_third" (JSON): https://vishalsingh.org/zipcode/report/smoking-in-pregnancy/figures/smoking_by_income_third.json
- Table 2, figure "smoking_county_map" (JSON): https://vishalsingh.org/zipcode/report/smoking-in-pregnancy/figures/smoking_county_map.json
- Table 3, figure "smoking_vs_income_counties" (JSON): https://vishalsingh.org/zipcode/report/smoking-in-pregnancy/figures/smoking_vs_income_counties.json
- Table 4, figure "slope_scope" (JSON): https://vishalsingh.org/zipcode/report/smoking-in-pregnancy/figures/slope_scope.json
- Table 5, figure "state_slopes" (JSON): https://vishalsingh.org/zipcode/report/smoking-in-pregnancy/figures/state_slopes.json
- Table 6, figure "slope_vs_state_level" (JSON): https://vishalsingh.org/zipcode/report/smoking-in-pregnancy/figures/slope_vs_state_level.json
- Table 7, figure "cutoff_sensitivity" (JSON): https://vishalsingh.org/zipcode/report/smoking-in-pregnancy/figures/cutoff_sensitivity.json
- Table 8, figure "slope_over_time" (JSON): https://vishalsingh.org/zipcode/report/smoking-in-pregnancy/figures/slope_over_time.json
- Payload of "smoking_county_map" (JSON): https://vishalsingh.org/zipcode/data/smoking-in-pregnancy/smoking_county_map.json
- Index of every report: https://vishalsingh.org/zipcode/llms.txt; all reports in one file: https://vishalsingh.org/zipcode/llms-full.txt

---

# Where births are falling: report

Chapter 2.4 of ZIP Code Destiny, by Vishal Singh (NYU Stern). Generated from the chapter's data files; the reading version is at https://vishalsingh.org/zipcode/where-births-are-falling. Table n of this report is figure n of the chapter.

## Identification

- Title: Where births are falling
- Label: The vanishing cradle
- Chapter: 2.4 (Part II, Birth; An American life)
- Question: Which counties lost the most births between 2018 and 2024, and how did the age of mothers change in those places?
- Dek: Births in the counties that CDC WONDER identifies fell 4.6% from 2018 to 2024, by 5.6% in the least deprived quartile of counties and 7.3% in the most deprived.
- Grain: counties
- Status: published
- Published: 2026-10-09
- Updated: not revised
- Data sources:
  - CDC WONDER natality. CDC WONDER natality, county totals and county by mother's age, births in 2018 to 2024 (consolidated pulls); 2024 is final; years: 2018-2024; weights: births in each county (pooled ratios of sums); changes and shares pooled over counties
  - Area Deprivation Index. Neighborhood Atlas Area Deprivation Index 2023 (v4.0.1), tract national rank, averaged to counties with 2019-2023 tract population; years: 2023 release (version 4.0.1; built on ACS 2019-2023); weights: 2019-2023 tract population
- Canonical URL: https://vishalsingh.org/zipcode/where-births-are-falling
- Report: https://vishalsingh.org/zipcode/where-births-are-falling/report
- Markdown: https://vishalsingh.org/zipcode/where-births-are-falling/report.md
- BibTeX key: singh2026wherebirthsarefalling

```bibtex
@misc{singh2026wherebirthsarefalling,
  author = {Singh, Vishal},
  title = {Where births are falling},
  year = {2026},
  url = {https://vishalsingh.org/zipcode/where-births-are-falling},
  note = {In ZIP Code Destiny. Published 9 October 2026.}
}
```

## Principal estimates

The facts the dek and the opening prose cite, in order of appearance.

| Estimand | Population | Period | Value | Interval | n | Weights | Source | Fact id |
| --- | --- | --- | ---: | --- | ---: | --- | --- | --- |
| fall in the count of births from 2018 to 2024 | live births in the counties that CDC WONDER natality identifies in every year (counties with 100,000 or more residents in the population vintage WONDER uses) | 2018 to 2024 | 4.6% |   |   | none (full count) | CDC WONDER natality | decline_named |
| fall in the pooled count of births from 2018 to 2024, bootstrap interval over counties | live births in the least deprived quartile of the named counties by county Area Deprivation Index rank | births in 2018 and 2024 | 5.6% | 95% CI 2.8% to 8.2% | 144 counties | births in each county (a pooled ratio of sums over counties); interval from a bootstrap over counties | CDC WONDER natality | decline_q1 |
| fall in the pooled count of births from 2018 to 2024, bootstrap interval over counties | live births in the most deprived quartile of the named counties by county Area Deprivation Index rank | births in 2018 and 2024 | 7.3% | 95% CI 5.8% to 8.8% | 145 counties | births in each county (a pooled ratio of sums over counties); interval from a bootstrap over counties | CDC WONDER natality | decline_q4 |
| count of counties | counties that WONDER natality identifies in every year from 2018 to 2024 and that have births | births in 2018 to 2024 | 578 |   |   | none (full count) | CDC WONDER natality | counties_used |
| count of live births | live births in the counties that CDC WONDER natality identifies in every year (counties with 100,000 or more residents in the population vintage WONDER uses) | 2018 | 3.02 million |   |   | none (full count) | CDC WONDER natality | births_named_2018 |
| count of live births | live births in the counties that CDC WONDER natality identifies in every year (counties with 100,000 or more residents in the population vintage WONDER uses) | 2024 | 2.88 million |   |   | none (full count) | CDC WONDER natality | births_named_2024 |
| population size at which WONDER identifies a county, applied to an earlier population vintage than the 2020 census (a definition) | counties that CDC WONDER natality identifies | births in 2018 to 2024 | 100,000 |   |   | none (definition) | CDC WONDER natality | wonder_threshold |
| share of births in the named counties | live births in the 50 states and DC | 2024 | 79.4% |   |   | none (full count) | CDC WONDER natality | share_named_2024 |
| pooled change in the count of births from 2018 to 2024, bootstrap interval over counties | live births in the named counties in the Census South region | births in 2018 and 2024 | 0.0% | 95% CI −1.7% to 1.8% | 232 counties | births in each county (a pooled ratio of sums over counties); interval from a bootstrap over counties | CDC WONDER natality | region_south |
| pooled change in the count of births from 2018 to 2024, bootstrap interval over counties | live births in the named counties in the Census Midwest region | births in 2018 and 2024 | −8.4% | 95% CI −10.4% to −6.3% | 137 counties | births in each county (a pooled ratio of sums over counties); interval from a bootstrap over counties | CDC WONDER natality | region_midwest |
| pooled change in the count of births from 2018 to 2024, bootstrap interval over counties | live births in the named counties in the Census West region | births in 2018 and 2024 | −7.9% | 95% CI −10.7% to −5.0% | 99 counties | births in each county (a pooled ratio of sums over counties); interval from a bootstrap over counties | CDC WONDER natality | region_west |
| pooled change in the count of births from 2018 to 2024, bootstrap interval over counties | live births in the named counties in the Census Northeast region | births in 2018 and 2024 | −5.0% | 95% CI −7.3% to −2.5% | 110 counties | births in each county (a pooled ratio of sums over counties); interval from a bootstrap over counties | CDC WONDER natality | region_northeast |
| fall in the pooled count of births from 2018 to 2024, bootstrap interval over counties | live births in the second quartile of the named counties by county Area Deprivation Index rank | births in 2018 and 2024 | 2.6% | 95% CI −0.6% to 5.3% | 145 counties | births in each county (a pooled ratio of sums over counties); interval from a bootstrap over counties | CDC WONDER natality | decline_q2 |
| fall in the pooled count of births from 2018 to 2024, bootstrap interval over counties | live births in the third quartile of the named counties by county Area Deprivation Index rank | births in 2018 and 2024 | 3.1% | 95% CI 1.6% to 4.5% | 144 counties | births in each county (a pooled ratio of sums over counties); interval from a bootstrap over counties | CDC WONDER natality | decline_q3 |
| pooled change in the count of births from 2018 to 2024, bootstrap interval over counties | live births to mothers aged 15 to 19 in the named counties that publish the band in both years | births in 2018 and 2024 | 24.6% | 95% CI 22.7% to 26.5% | 577 counties | births in each county (a pooled ratio of sums over counties); interval from a bootstrap over counties | CDC WONDER natality | fall_1519 |
| age-of-mother range used in this chapter, from WONDER's five-year bands (a definition) | mothers | all years (a definition) | 15 to 19 |   |   | none (definition) | CDC WONDER natality | band_1519 |
| pooled change in the count of births from 2018 to 2024, bootstrap interval over counties | live births to mothers aged 40 to 44 in the named counties that publish the band in both years | births in 2018 and 2024 | 19.5% | 95% CI 17.0% to 22.3% | 578 counties | births in each county (a pooled ratio of sums over counties); interval from a bootstrap over counties | CDC WONDER natality | rise_4044 |
| age-of-mother range used in this chapter, from WONDER's five-year bands (a definition) | mothers | all years (a definition) | 40 to 44 |   |   | none (definition) | CDC WONDER natality | band_4044 |

## Findings

Births in the 578 counties that CDC WONDER identifies fell from 3.02 million in 2018 to 2.88 million in 2024, a decline of 4.6%. These counties have 100,000 or more residents in the population vintage WONDER uses, and they record 79.4% of the births in the United States in 2024.

Pooled over counties, births changed by 0.0% (95% CI −1.7% to 1.8%) in the South, −8.4% (95% CI −10.4% to −6.3%) in the Midwest, −7.9% (95% CI −10.7% to −5.0%) in the West and −5.0% (95% CI −7.3% to −2.5%) in the Northeast. Pooled births fell by 5.6% (95% CI 2.8% to 8.2%) in the least deprived quartile of these counties, by 2.6% and 3.1% in the second and third quartiles, and by 7.3% (95% CI 5.8% to 8.8%) in the most deprived. The losses do not rise in steps with deprivation, and region and deprivation overlap, so the chapter looks at them together.

The ages of mothers shifted too. Between 2018 and 2024 births fell by 24.6% among mothers aged 15 to 19 and rose by 19.5% among mothers aged 40 to 44.

Everything in this chapter is a count of births in a county or a share of those births. A birth rate divides births by the women who could have had them, and the files behind this book have no county counts of women by age, so the chapter reports counts and shares and no rates. The babies in these counties are the subject of Chapter 2.1, Birthweight and preterm birth (https://vishalsingh.org/zipcode/birthweight-and-preterm-birth), which asks how heavy they were at birth, and of Chapter 2.2, Infant mortality (https://vishalsingh.org/zipcode/infant-mortality), which follows them through the first year.

### Births have not returned to their 2018 level

Births in the named counties fell from 2018 to 2020 and rose through 2022 (Table 1). They fell again to a low in 2023, when they were 5.5% below 2018, and rose 0.9% in 2024. In 2020 they were 4.8% below 2018, and by 2024 they were still 4.6% below.

WONDER pools the births of every smaller county in a state into one row, 49 rows in all. Those rows record 0.77 million births in 2018 and 0.75 million in 2024, a decline of 3.0%, so the named counties lost a larger share of their births than the rest of the country combined. Across the whole country births fell from 3.79 million to 3.63 million, a decline of 4.3%. The count for 2024 is the one in the final report of the National Center for Health Statistics, and 2024 is final in these data.[1]

Birth rates in the United States have been falling since the Great Recession. Kearney, Levine and Pardue document that birth rates fell between 2007 and 2020 across many groups of women, and they find that the recession explains only part of the early decline.[2] This chapter starts in 2018 and describes where the recent loss fell.

Table 1. Births in the named counties fell 4.6% from 2018 to 2024 with a low in 2023, a larger fall than the 3.0% in all other counties

The 578 named counties are the counties that WONDER identifies in every year from 2018 to 2024 and that have births in every year. WONDER lists two more county identifiers in every year, Cape May County, New Jersey, and Portsmouth city, Virginia. It gives them no births because it folded them into their state's pooled row, so they are left out. Every named county appears in every year from 2018 to 2024, so no county drops out of the panel by entering or leaving WONDER's list.

The selection is by size as WONDER measured it. WONDER applies its threshold of 100,000 residents to an earlier population estimate than the 2020 Census. By 2020 Census population, 27 counties outside Connecticut that have 100,000 or more residents are still in the pooled rows, among them Sumter County, Florida, Gallatin County, Montana, and Flagler County, Florida. In the other direction, one named county, Ashtabula County, Ohio, is below the line. The staged atlas has no 2020 count for Connecticut, whose counties are all named. Counties that grew past the line after WONDER's estimate therefore sit in the pooled rows. The comparison of the 3.0% decline there with the 4.6% decline in the named counties compares two groups defined by an earlier population estimate and not by size today.

The named counties record 79.7% of births in 2018 and 79.4% in 2024, and the smallest of them recorded at least 613 births in every year. County results in this chapter describe these counties and not the country. The pooled rows lost fewer births, with the caveat that they include counties that now exceed the line.

### Births held up in the South and fell elsewhere

The Census Bureau groups states into the Northeast, the Midwest, the South and the West. The South, which has 16 states and the District of Columbia, holds 232 of the named counties, or 40.1%. Pooled over its counties, the South's births changed by 0.0% (95% CI −1.7% to 1.8%) from 2018 to 2024, an interval that includes zero. Births in the Midwest changed by −8.4% (95% CI −10.4% to −6.3%), in the West by −7.9% (95% CI −10.7% to −5.0%) and in the Northeast by −5.0% (95% CI −7.3% to −2.5%), so the intervals for those three regions lie below zero.

The South's change is higher than the Northeast's by 5.0 percentage points (95% CI 2.0 percentage points to 7.9 percentage points), than the Midwest's by 8.4 percentage points (95% CI 5.7 percentage points to 11.1 percentage points) and than the West's by 7.9 percentage points (95% CI 4.3 percentage points to 11.3 percentage points). The intervals come from a bootstrap that resamples whole counties, so they show how much a region's pooled change depends on which counties it holds.

Table 2. Of the 58 counties with the largest gains in births, 43 are in the South, which holds 40.1% of the named counties

The map (Table 2) shows the same split county by county. The top tenth of counties by change, with gains of 10.2% or more, holds 58 counties, and 43 of them are in the South. The bottom tenth, with declines of 13.6% or more, holds 58 counties, and 13 are in the South.

Counted one county at a time, 65.7% of the named counties lost births. Births were lower in 2024 than in 2018 in 380 counties, higher in 197 and the same in one.

The median county lost 3.8%, and from the 5th to the 95th percentile of counties, changes run from a decline of 16.5% to a gain of 19.9%. The pooled decline of 4.6% is larger than the median county's because the pooled figure weights each county by its births and the median counts each county once.

Table 3. The ten largest gains in births are all in the South, and the ten largest declines are spread across every Census region

Table 3 lists the ten counties with the largest gains and the ten with the largest declines. Beyond the middle of the distribution, where changes from the 5th to the 95th percentile of counties run from a decline of 16.5% to a gain of 19.9%, the extremes are a decline of 23.9% in St. Louis city, Missouri, and a gain of 67.6% in Kaufman County, Texas. All of the gains are in the South, in Texas, North Carolina and Georgia. The declines are spread across every region and include St. Louis city, Missouri, Hinds County, Mississippi, Humboldt County, California, San Francisco County and Orleans Parish, Louisiana. Every county listed had at least 987 births in 2018. Small counts move more from year to year by chance, so the order within each group is less stable than the groups themselves.

A county's births change when women have fewer children, when the number of women of childbearing age changes, or both, and these data cannot separate them. WONDER places a birth in the mother's county of residence, so a county that adds young families counts their births, and a gain in a fast-growing county may reflect arrivals as much as larger families.

### The most deprived counties lost more births than the middle quartiles

To place counties on a deprivation scale, each census tract carries its Area Deprivation Index national rank, which orders neighborhoods by deprivation, with a higher rank meaning more deprived.[3] A tract's rank is the unweighted mean of the ranks of its block groups, and a county's rank is the average of its tracts' ranks, weighted by tract population, the measure used in Chapter 2.1, Birthweight and preterm birth (https://vishalsingh.org/zipcode/birthweight-and-preterm-birth). Quartiles hold equal numbers of the named counties, so the most deprived quartile is relative to large counties and not to the nation.

By 2024 pooled births in the most deprived quartile were 7.3% (95% CI 5.8% to 8.8%) below 2018 (Table 4). All four quartiles fell to 2020. In 2021 births in the other three quartiles rose and births in the most deprived quartile did not. Its pooled change was −0.3% (95% CI −0.8% to 0.1%), an interval that includes zero, and the other quartiles rose by more, by 1.9 percentage points (95% CI 1.0 percentage points to 3.0 percentage points) in the least deprived quartile, 2.5 percentage points (95% CI 1.4 percentage points to 3.6 percentage points) in the second and 1.3 percentage points (95% CI 0.5 percentage points to 2.2 percentage points) in the third.

The second and third quartiles lost 2.6% (95% CI −0.6% to 5.3%) and 3.1% (95% CI 1.6% to 4.5%) from 2018 to 2024, and the middle quartiles together lost 2.8% (95% CI 1.1% to 4.5%). The most deprived quartile therefore lost 4.5 percentage points (95% CI 2.3 percentage points to 6.7 percentage points) more than the middle quartiles. The least deprived quartile lost 5.6% (95% CI 2.8% to 8.2%). It lost 2.7 percentage points (95% CI −0.5 percentage points to 5.9 percentage points) more than the middle quartiles and 1.7 percentage points (95% CI −1.2 percentage points to 5.0 percentage points) less than the most deprived quartile, and both intervals include zero, so the data do not place it with either.

A single slope does not summarize the pattern. Regressing each county's change on its deprivation rank gives +0.5 percentage points (95% CI −0.2 percentage points to 1.2 percentage points) per ten ranks when counties are weighted by their births and −0.7 percentage points (95% CI −1.1 percentage points to −0.3 percentage points) when each county counts once. The sign depends on the weights, and only the unweighted interval excludes zero.

Table 4. Births in 2024 were below 2018 in all four quartiles of counties, by 5.6% in the least deprived and 7.3% in the most deprived, and in 2021 the other three quartiles rose by more than the most deprived

Region and deprivation overlap. Of the counties in the most deprived quartile, 45% are in the South, against 24% of the counties in the least deprived quartile. Table 5 splits the quartiles by region. The most deprived quartile lost births in every region: 9.3% (95% CI 6.6% to 11.8%) in the Northeast, 9.2% (95% CI 6.3% to 11.6%) in the Midwest, 5.0% (95% CI 2.9% to 7.0%) in the South and 9.3% in the West. The West's most deprived quartile holds only 5 counties, all of which lost births. A bootstrap over so few counties is mechanical, so no interval is printed for it.

In the South the other three quartiles together changed by +1.1% (95% CI −0.9% to 3.4%), which is higher than the change in the South's most deprived quartile by 6.1 percentage points (95% CI 3.2 percentage points to 9.0 percentage points). The South's flat total therefore combines a loss in its most deprived counties with no clear change elsewhere.

Table 5. The most deprived quartile of counties lost births in every Census region, and in the South the other three quartiles show no clear loss

These are patterns across counties and describe groups of births, not the mothers in any one county. The index measures neighborhoods averaged to counties, and it understates deprivation in high-cost cities, so some counties in the least deprived quartile hold neighborhoods that are more deprived than their rank suggests.

### A larger share of births were to older mothers in 2024

From 2018 to 2024, births fell in every five-year band of mothers aged 15 to 29, by 24.6% (95% CI 22.7% to 26.5%) among mothers aged 15 to 19, by 16.6% (95% CI 15.0% to 18.3%) among mothers aged 20 to 24 and by 11.2% (95% CI 9.6% to 12.9%) among mothers aged 25 to 29. Births to mothers aged 30 to 34 changed by +1.0% (95% CI −0.6% to 2.6%), an interval that includes zero. Births rose by 9.2% (95% CI 7.3% to 11.0%) among mothers aged 35 to 39 and by 19.5% (95% CI 17.0% to 22.3%) among mothers aged 40 to 44 (Table 6).

As shares of births to mothers aged 15 to 44, mothers aged 15 to 24 went from 21.8% (95% CI 21.0% to 22.7%) to 18.7% (95% CI 18.0% to 19.5%), a fall of 3.1 percentage points (95% CI 2.9 percentage points to 3.3 percentage points). Mothers aged 35 to 44 went from 19.7% (95% CI 18.9% to 20.5%) to 22.9% (95% CI 22.1% to 23.8%), a rise of 3.2 percentage points (95% CI 3.1 percentage points to 3.4 percentage points). The shift continues a rise in the mean age of mothers that the National Center for Health Statistics documented for every birth order from 2000 to 2014.[4]

Table 6. Births fell for mothers aged 15 to 29, changed little for mothers aged 30 to 34 and rose for mothers aged 35 to 44 from 2018 to 2024

WONDER splits each county's births by the mother's age in nine bands. Six of them, the bands from 15 to 19 through 40 to 44, are published for nearly every county and year. Of the 24,276 county-year cells for those bands, WONDER withholds 11. Each holds fewer than 10 births, so together they hold at most 99 births, against 20.4 million in published cells.

The other three bands, mothers under 15 and mothers 45 and over, are mostly withheld, with 6,594 of 12,138 cells missing. The published cells hold 0.26% of births and the withheld ones at most 0.29%. The chapter therefore leaves all three bands out and compares mothers aged 15 to 44. A withheld cell is missing, never zero, and nothing is recovered from county totals.

These are births and not birth rates. A fall in births to mothers aged 15 to 19 can mean fewer births per teenager or fewer teenagers in the county, and the files behind this book have no county counts of women by age. A different measure of teen births, for women who grew up in each census tract, is in Chapter 4.2, Teen births (https://vishalsingh.org/zipcode/teen-births).

### The age mix differs more across counties than it changed

In 2018 the share of births to mothers aged 15 to 24 was 17.0% (95% CI 15.6% to 18.3%) in the least deprived quartile of counties and 30.9% (95% CI 29.9% to 32.1%) in the most deprived, a gap of 14.0 percentage points (95% CI 12.3 percentage points to 15.8 percentage points). In 2024 the shares were 14.5% (95% CI 13.4% to 15.5%) and 26.9% (95% CI 25.8% to 28.0%), a gap of 12.4 percentage points (95% CI 11.0 percentage points to 14.0 percentage points).

For mothers aged 35 to 44 the 2018 shares were 24.3% (95% CI 23.1% to 25.5%) in the least deprived quartile and 13.0% (95% CI 12.5% to 13.5%) in the most deprived, a gap of 11.3 percentage points (95% CI 10.0 percentage points to 12.5 percentage points). In 2024 they were 27.8% (95% CI 26.4% to 29.2%) and 15.8% (95% CI 15.1% to 16.5%), a gap of 12.0 percentage points (95% CI 10.5 percentage points to 13.5 percentage points) (Table 7).

Every quartile moved in the same direction. The share of births to mothers aged 15 to 24 fell by 2.5 percentage points (95% CI 1.9 percentage points to 3.0 percentage points) in the least deprived quartile and by 4.0 percentage points (95% CI 3.7 percentage points to 4.4 percentage points) in the most deprived. The share to mothers aged 35 to 44 rose by 3.5 percentage points (95% CI 3.1 percentage points to 3.9 percentage points) and by 2.8 percentage points (95% CI 2.5 percentage points to 3.1 percentage points). The share of young mothers fell by 1.5 percentage points (95% CI 0.9 percentage points to 2.2 percentage points) more in the most deprived quartile. The share of older mothers rose by 0.7 percentage points (95% CI 0.3 percentage points to 1.2 percentage points) more in the least deprived quartile, so the shifts differ in size. Each is much smaller than the gap between quartiles.

Table 7. In every quartile of county deprivation the share of births to older mothers rose and the share to younger mothers fell, and the differences between quartiles are larger than those changes

### Counties that lost the most births lost them among younger mothers

The counties that lost the most births lost them among mothers under 35. Sorting counties by the change in their total births shows where in the age range the loss fell (Table 8). The 144 counties in the quarter with the largest declines, falls of 8.8% or more, lost 13.5% of their births in all. Their births fell by 32.9% among mothers aged 15 to 19, by 25.1% among mothers aged 20 to 24, by 21.1% among mothers aged 25 to 29 and by 8.8% among mothers aged 30 to 34. Births to mothers aged 35 to 39 changed by −0.1%, and births to mothers aged 40 to 44 rose by 9.6%.

The 145 counties in the quarter with the largest gains, gains of 2.6% or more, added 11.2% in all. Births in those counties fell by 13.0% among mothers aged 15 to 19 and by 4.0% among mothers aged 20 to 24, and rose in every older band, from 5.8% among mothers aged 25 to 29 to 39.3% among mothers aged 40 to 44.

Table 8. Counties with the largest declines in births lost births at every age under 35, and counties with the largest gains added births at every age from 25

The groups are formed on the change in total births, so the figure splits each group's change by age and does not test an explanation. A separate question is whether a county's age mix predicts its change. Suppose every age band in a county changed at the all-county rate. The county's 2018 mix would then imply a change in births to mothers aged 15 to 44 between a fall of 8.9% and a fall of 1.3% at the 5th and 95th percentiles of counties. The observed changes run over a much wider range, from a fall of 16.6% to a gain of 19.8%.

The implied and observed changes have a correlation of −0.16 (95% CI −0.26 to −0.06) across 577 counties when counties are weighted by their births and −0.03 (95% CI −0.11 to 0.04) when each counts once. The one county that withholds an age cell is left out. A county's 2018 age mix therefore does not account for which counties lost births. The weighted correlation is small and negative, so counties whose mix implied larger falls lost slightly less.

### What these data cannot show

The chapter reports counts and shares of births and no rates. A birth rate needs the number of women of each age as its denominator, and the files staged for this book hold the total population of counties and tracts but not women by age. A fall in births to mothers aged 20 to 24 can mean fewer births per woman, fewer women in that age group, or both, and so can a gain. County population estimates by age and sex from the Census Bureau would supply the denominator for age-specific rates. Births are counted in the mother's county of residence, so the map is not a map of fertility.

County results describe the 578 counties that WONDER identifies, those with 100,000 or more residents in its population vintage. They hold 79.4% of births in 2024, and they are not a sample of the country. The pooled rows lost fewer births, although those rows include counties that now exceed the line. WONDER's county files used here have no cut by the mother's race or Hispanic origin, so the chapter cannot say whether the shift in mothers' ages differs by group.

The deprivation comparison is descriptive. The Area Deprivation Index describes neighborhoods in 2019 to 2023 and is averaged to counties, and region and deprivation overlap. The comparison of quartiles says nothing about why births fell in any county. It is an ecological pattern, and it need not hold for the people inside the counties. The comparison of 2018 with 2024 uses the endpoint years only. A county with few births moves from year to year by chance, so a county's direction of change is better established than its rank among the named counties.

## Tables

### Table 1. Births in the named counties fell 4.6% from 2018 to 2024 with a low in 2023, a larger fall than the 3.0% in all other counties

Live births by year in the 578 counties that CDC WONDER identifies (counties with 100,000 or more residents in WONDER's population vintage) and in all other counties, which WONDER pools by state, 2018 to 2024, indexed to 2018 or as counts

| measure | n | series | value | year |
| --- | ---: | --- | ---: | ---: |
| index | 3,020,813 | Named counties | 100 | 2018 |
| births | 3,020,813 | Named counties | 3020813 | 2018 |
| index | 2,985,056 | Named counties | 98.82 | 2019 |
| births | 2,985,056 | Named counties | 2985056 | 2019 |
| index | 2,875,172 | Named counties | 95.18 | 2020 |
| births | 2,875,172 | Named counties | 2875172 | 2020 |
| index | 2,912,953 | Named counties | 96.43 | 2021 |
| births | 2,912,953 | Named counties | 2912953 | 2021 |
| index | 2,920,239 | Named counties | 96.67 | 2022 |
| births | 2,920,239 | Named counties | 2920239 | 2022 |
| index | 2,855,036 | Named counties | 94.51 | 2023 |
| births | 2,855,036 | Named counties | 2855036 | 2023 |
| index | 2,881,385 | Named counties | 95.38 | 2024 |
| births | 2,881,385 | Named counties | 2881385 | 2024 |
| index | 770,899 | Other counties | 100 | 2018 |
| births | 770,899 | Other counties | 770899 | 2018 |
| index | 762,484 | Other counties | 98.91 | 2019 |
| births | 762,484 | Other counties | 762484 | 2019 |
| index | 738,475 | Other counties | 95.79 | 2020 |
| births | 738,475 | Other counties | 738475 | 2020 |
| index | 751,339 | Other counties | 97.46 | 2021 |
| births | 751,339 | Other counties | 751339 | 2021 |
| index | 747,519 | Other counties | 96.97 | 2022 |
| births | 747,519 | Other counties | 747519 | 2022 |
| index | 740,981 | Other counties | 96.12 | 2023 |
| births | 740,981 | Other counties | 740981 | 2023 |
| index | 747,549 | Other counties | 96.97 | 2024 |
| births | 747,549 | Other counties | 747549 | 2024 |

Views: Index, 2018 = 100 (measure = index); Live births (measure = births).

Note: The named counties are the 578 that CDC WONDER natality identifies in every year from 2018 to 2024: counties with 100,000 or more residents in the population vintage WONDER uses, which is earlier than the 2020 Census. WONDER pools the births of every other county in a state into one row, so those counties are not in the named counties.

Note: Births are full counts from birth certificates, so no interval is drawn. A count of births is not a birth rate: the series blends births per woman with the number of women.

Source: CDC WONDER natality, births in 2018-2024; full counts of births (no sampling error)

Data: https://vishalsingh.org/zipcode/report/where-births-are-falling/figures/births_trend.json

### Table 2. Of the 58 counties with the largest gains in births, 43 are in the South, which holds 40.1% of the named counties

Percent change in live births from 2018 to 2024 by county, for the 578 counties that CDC WONDER identifies (counties with 100,000 or more residents in WONDER's population vintage)

No inline rows: the values of Table 2 are in https://vishalsingh.org/zipcode/data/where-births-are-falling/births_change_map.json.

Note: The named counties are the 578 that CDC WONDER natality identifies in every year from 2018 to 2024: counties with 100,000 or more residents in the population vintage WONDER uses, which is earlier than the 2020 Census. WONDER pools the births of every other county in a state into one row, so those counties are not in the named counties.

Note: Births are full counts, so no interval is drawn for a county. A small county's count moves more from year to year by chance, so the shading of a county with few births is less stable than the shading of a large one.

Note: The tooltip gives the value of a county whose change is beyond the end of the key.

Note: Connecticut is shown as its eight legacy counties, the coding of the source. Alaska is drawn. Puerto Rico is not in these data and is not drawn.

Source: CDC WONDER natality, births in 2018-2024; full counts of births (no sampling error); county changes are not weighted

Data: https://vishalsingh.org/zipcode/report/where-births-are-falling/figures/births_change_map.json

### Table 3. The ten largest gains in births are all in the South, and the ten largest declines are spread across every Census region

Percent change in live births from 2018 to 2024 for the ten named counties with the largest gains and the ten with the largest declines, of the 578 named counties

| births_2018 | births_2024 | change | county | group |
| ---: | ---: | ---: | --- | --- |
| 1786 | 2993 | 67.58 | Kaufman County, TX | Largest gains |
| 1526 | 2203 | 44.36 | Comal County, TX | Largest gains |
| 2572 | 3670 | 42.69 | Hays County, TX | Largest gains |
| 6650 | 9325 | 40.23 | Williamson County, TX | Largest gains |
| 1537 | 2096 | 36.37 | Parker County, TX | Largest gains |
| 987 | 1320 | 33.74 | Brunswick County, NC | Largest gains |
| 2214 | 2914 | 31.62 | Ellis County, TX | Largest gains |
| 2299 | 2975 | 29.40 | Johnston County, NC | Largest gains |
| 1173 | 1501 | 27.96 | Bartow County, GA | Largest gains |
| 7355 | 9220 | 25.36 | Montgomery County, TX | Largest gains |
| 4127 | 3142 | −23.87 | St. Louis city, MO | Largest declines |
| 3041 | 2361 | −22.36 | Hinds County, MS | Largest declines |
| 1364 | 1062 | −22.14 | Humboldt County, CA | Largest declines |
| 3109 | 2449 | −21.23 | St. Clair County, IL | Largest declines |
| 1518 | 1204 | −20.69 | Monroe County, PA | Largest declines |
| 1506 | 1195 | −20.65 | Terrebonne Parish, LA | Largest declines |
| 8697 | 6930 | −20.32 | San Francisco County, CA | Largest declines |
| 1326 | 1072 | −19.16 | Fayette County, PA | Largest declines |
| 4656 | 3769 | −19.05 | Orleans Parish, LA | Largest declines |
| 1246 | 1010 | −18.94 | Monroe County, IN | Largest declines |

Note: The named counties are the 578 that CDC WONDER natality identifies in every year from 2018 to 2024: counties with 100,000 or more residents in the population vintage WONDER uses, which is earlier than the 2020 Census. WONDER pools the births of every other county in a state into one row, so those counties are not in the named counties.

Note: Births are full counts, so no interval is drawn. The data table lists each county's births in 2018 and 2024.

Source: CDC WONDER natality, births in 2018-2024; full counts of births (no sampling error); county changes are not weighted

Data: https://vishalsingh.org/zipcode/report/where-births-are-falling/figures/births_change_extremes.json

### Table 4. Births in 2024 were below 2018 in all four quartiles of counties, by 5.6% in the least deprived and 7.3% in the most deprived, and in 2021 the other three quartiles rose by more than the most deprived

Pooled live births in each quartile of the 578 named counties by Area Deprivation Index rank, indexed to 2018 = 100, 2018 to 2024, with 95% intervals from a bootstrap over counties

| conf_high | conf_low | index | n | quartile | year |
| ---: | ---: | ---: | ---: | --- | ---: |
| n/a | n/a | 100 | 1,224,834 | Least deprived quartile | 2018 |
| 99 | 98.12 | 98.51 | 1,206,641 | Least deprived quartile | 2019 |
| 95.57 | 93.10 | 94.28 | 1,154,721 | Least deprived quartile | 2020 |
| 97.84 | 93.85 | 95.77 | 1,173,023 | Least deprived quartile | 2021 |
| 97.77 | 93.60 | 95.59 | 1,170,851 | Least deprived quartile | 2022 |
| 95.57 | 90.73 | 93.09 | 1,140,185 | Least deprived quartile | 2023 |
| 97.19 | 91.77 | 94.43 | 1,156,626 | Least deprived quartile | 2024 |
| n/a | n/a | 100 | 710,664 | Second quartile | 2018 |
| 99.18 | 97.97 | 98.56 | 700,447 | Second quartile | 2019 |
| 96.58 | 94.40 | 95.43 | 678,194 | Second quartile | 2020 |
| 99.46 | 95.68 | 97.54 | 693,209 | Second quartile | 2021 |
| 100.9 | 95.90 | 98.47 | 699,762 | Second quartile | 2022 |
| 99.12 | 93.81 | 96.40 | 685,072 | Second quartile | 2023 |
| 100.6 | 94.70 | 97.44 | 692,445 | Second quartile | 2024 |
| n/a | n/a | 100 | 630,186 | Third quartile | 2018 |
| 99.75 | 98.96 | 99.31 | 625,862 | Third quartile | 2019 |
| 96.79 | 95.72 | 96.20 | 606,252 | Third quartile | 2020 |
| 98.35 | 96.20 | 97.14 | 612,146 | Third quartile | 2021 |
| 98.90 | 96.93 | 97.88 | 616,853 | Third quartile | 2022 |
| 97.59 | 95.05 | 96.34 | 607,127 | Third quartile | 2023 |
| 98.42 | 95.52 | 96.86 | 610,399 | Third quartile | 2024 |
| n/a | n/a | 100 | 455,129 | Most deprived quartile | 2018 |
| 99.76 | 98.90 | 99.34 | 452,106 | Most deprived quartile | 2019 |
| 96.26 | 95.35 | 95.80 | 436,005 | Most deprived quartile | 2020 |
| 96.17 | 94.80 | 95.48 | 434,575 | Most deprived quartile | 2021 |
| 96.14 | 93.99 | 95.09 | 432,773 | Most deprived quartile | 2022 |
| 94.09 | 91.59 | 92.86 | 422,652 | Most deprived quartile | 2023 |
| 94.20 | 91.23 | 92.70 | 421,915 | Most deprived quartile | 2024 |

Note: The named counties are the 578 that CDC WONDER natality identifies in every year from 2018 to 2024: counties with 100,000 or more residents in the population vintage WONDER uses, which is earlier than the 2020 Census. WONDER pools the births of every other county in a state into one row, so those counties are not in the named counties.

Note: A tract's Area Deprivation Index national rank (1 is least deprived, 100 most) is the unweighted mean of the ranks of its block groups. A county's deprivation rank is the mean of its tracts' ranks weighted by 2019-2023 tract population. Quartiles hold equal numbers of the named counties, so the most deprived quartile is relative to large counties and not to the nation. The index is a measure of neighborhoods, averaged to counties.

Note: Intervals are 95% intervals from a bootstrap over counties (replicates and seed in the methods note). Births are full counts, so an interval shows how much a pooled figure depends on which counties are in the group, not sampling error in the counts. The 2018 point is the base and has no interval.

Source: CDC WONDER natality, births in 2018-2024; full counts of births; counties grouped by Area Deprivation Index 2023 with ACS 2019-2023 tract population (population-weighted); pooled over counties by births (births-weighted); intervals from a bootstrap over counties

Data: https://vishalsingh.org/zipcode/report/where-births-are-falling/figures/births_by_deprivation.json

### Table 5. The most deprived quartile of counties lost births in every Census region, and in the South the other three quartiles show no clear loss

Pooled change in live births from 2018 to 2024 by Census region and quartile of county deprivation rank, named counties, with 95% intervals from a bootstrap over counties

| births_2018 | change | conf_high | conf_low | n_lost_births | n_units | quartile | region |
| ---: | ---: | ---: | ---: | ---: | ---: | --- | --- |
| 338526 | −4.36 | 0.120 | −7.66 | 24 | 39 | Least deprived quartile | Northeast |
| 76401 | −1.99 | −0.060 | −3.67 | 16 | 25 | Second quartile | Northeast |
| 96145 | −7.73 | −4.41 | −9.89 | 20 | 24 | Third quartile | Northeast |
| 48096 | −9.26 | −6.59 | −11.84 | 21 | 22 | Most deprived quartile | Northeast |
| 52620 | −5.68 | n/a | n/a | 6 | 9 | Least deprived quartile | Midwest |
| 158400 | −10.63 | −4.34 | −14.08 | 19 | 26 | Second quartile | Midwest |
| 156974 | −6.13 | −4.79 | −7.50 | 46 | 49 | Third quartile | Midwest |
| 189123 | −9.20 | −6.26 | −11.60 | 44 | 53 | Most deprived quartile | Midwest |
| 183903 | 2.15 | 8.71 | −4.06 | 17 | 35 | Least deprived quartile | South |
| 371925 | 1.30 | 4.59 | −1.59 | 25 | 69 | Second quartile | South |
| 335404 | 0.340 | 3.71 | −1.79 | 23 | 63 | Third quartile | South |
| 208110 | −5.02 | −2.92 | −6.95 | 43 | 65 | Most deprived quartile | South |
| 649785 | −8.37 | −4.74 | −11.55 | 46 | 61 | Least deprived quartile | West |
| 103938 | −4.51 | 1.79 | −7.96 | 17 | 25 | Second quartile | West |
| 41663 | −9.28 | n/a | n/a | 8 | 8 | Third quartile | West |
| 9800 | −9.27 | n/a | n/a | 5 | 5 | Most deprived quartile | West |

Note: The named counties are the 578 that CDC WONDER natality identifies in every year from 2018 to 2024: counties with 100,000 or more residents in the population vintage WONDER uses, which is earlier than the 2020 Census. WONDER pools the births of every other county in a state into one row, so those counties are not in the named counties.

Note: A tract's Area Deprivation Index national rank (1 is least deprived, 100 most) is the unweighted mean of the ranks of its block groups. A county's deprivation rank is the mean of its tracts' ranks weighted by 2019-2023 tract population. Quartiles hold equal numbers of the named counties, so the most deprived quartile is relative to large counties and not to the nation. The index is a measure of neighborhoods, averaged to counties.

Note: Intervals are 95% intervals from a bootstrap over counties (replicates and seed in the methods note). Births are full counts, so an interval shows how much a pooled figure depends on which counties are in the group, not sampling error in the counts. A cell with fewer than 10 counties shows no interval; in the West, the most deprived and third quartiles hold few counties, and the least deprived quartile of the Midwest holds fewer than 10, so those cells are thin.

Note: The Census South region is 16 states and DC; the Northeast, Midwest and West follow the Census Bureau's regions.

Source: CDC WONDER natality, births in 2018-2024; full counts of births; counties grouped by Area Deprivation Index 2023 with ACS 2019-2023 tract population (population-weighted); pooled over counties by births (births-weighted); intervals from a bootstrap over counties

Data: https://vishalsingh.org/zipcode/report/where-births-are-falling/figures/births_change_region_quartile.json

### Table 6. Births fell for mothers aged 15 to 29, changed little for mothers aged 30 to 34 and rose for mothers aged 35 to 44 from 2018 to 2024

Pooled change in live births by mother's age band, 2018 to 2024, in the 578 named counties, with 95% intervals from a bootstrap over counties

| age | births_2018 | births_2024 | change | conf_high | conf_low | n_units |
| --- | ---: | ---: | ---: | ---: | ---: | ---: |
| 15 to 19 | 129757 | 97772 | −24.65 | −22.72 | −26.53 | 577 |
| 20 to 24 | 527582 | 439765 | −16.65 | −14.98 | −18.31 | 578 |
| 25 to 29 | 848726 | 753795 | −11.19 | −9.57 | −12.87 | 578 |
| 30 to 34 | 910977 | 920282 | 1.02 | 2.65 | −0.620 | 578 |
| 35 to 39 | 490705 | 535630 | 9.16 | 11.05 | 7.32 | 578 |
| 40 to 44 | 103030 | 123085 | 19.47 | 22.34 | 17.02 | 578 |

Note: The named counties are the 578 that CDC WONDER natality identifies in every year from 2018 to 2024: counties with 100,000 or more residents in the population vintage WONDER uses, which is earlier than the 2020 Census. WONDER pools the births of every other county in a state into one row, so those counties are not in the named counties.

Note: Intervals are 95% intervals from a bootstrap over counties (replicates and seed in the methods note). Births are full counts, so an interval shows how much a pooled figure depends on which counties are in the group, not sampling error in the counts.

Note: WONDER splits births into nine age bands. Mothers under 15 and mothers 45 and over are left out because WONDER withholds most of those cells; the methods note gives the counts. A withheld cell is missing, never zero, and a county that withholds a band in 2018 or 2024 is left out of that band's bar; the data table gives the number of counties in each bar.

Source: CDC WONDER natality, births in 2018-2024; full counts of births by mother's age; changes pooled over counties; intervals from a bootstrap over counties

Data: https://vishalsingh.org/zipcode/report/where-births-are-falling/figures/births_by_mother_age.json

### Table 7. In every quartile of county deprivation the share of births to older mothers rose and the share to younger mothers fell, and the differences between quartiles are larger than those changes

Share of births to mothers aged 15 to 24 and to mothers aged 35 to 44, among births to mothers aged 15 to 44, by quartile of county deprivation rank, 2018 and 2024, with 95% intervals from a bootstrap over counties

| group_label | hi1 | hi2 | lo1 | lo2 | measure | n_units | y2018 | y2024 |
| --- | ---: | ---: | ---: | ---: | --- | ---: | ---: | ---: |
| All named counties | 22.67 | 19.48 | 20.99 | 17.98 | Mothers aged 15 to 24 | 578 | 21.83 | 18.73 |
| Least deprived quartile | 18.27 | 15.51 | 15.57 | 13.42 | Mothers aged 15 to 24 | 144 | 16.96 | 14.46 |
| Second quartile | 22.09 | 18.75 | 20.07 | 16.95 | Mothers aged 15 to 24 | 145 | 21.01 | 17.80 |
| Third quartile | 26.55 | 23.11 | 24.72 | 21.20 | Mothers aged 15 to 24 | 144 | 25.63 | 22.18 |
| Most deprived quartile | 32.08 | 28.03 | 29.93 | 25.83 | Mothers aged 15 to 24 | 145 | 30.94 | 26.91 |
| All named counties | 20.51 | 23.80 | 18.92 | 22.08 | Mothers aged 35 to 44 | 578 | 19.72 | 22.95 |
| Least deprived quartile | 25.46 | 29.18 | 23.05 | 26.40 | Mothers aged 35 to 44 | 144 | 24.32 | 27.84 |
| Second quartile | 20.45 | 23.86 | 18.51 | 21.75 | Mothers aged 35 to 44 | 145 | 19.53 | 22.86 |
| Third quartile | 16.41 | 19.40 | 15.17 | 18.01 | Mothers aged 35 to 44 | 144 | 15.87 | 18.76 |
| Most deprived quartile | 13.50 | 16.52 | 12.50 | 15.07 | Mothers aged 35 to 44 | 145 | 13.01 | 15.79 |

Note: The named counties are the 578 that CDC WONDER natality identifies in every year from 2018 to 2024: counties with 100,000 or more residents in the population vintage WONDER uses, which is earlier than the 2020 Census. WONDER pools the births of every other county in a state into one row, so those counties are not in the named counties.

Note: A tract's Area Deprivation Index national rank (1 is least deprived, 100 most) is the unweighted mean of the ranks of its block groups. A county's deprivation rank is the mean of its tracts' ranks weighted by 2019-2023 tract population. Quartiles hold equal numbers of the named counties, so the most deprived quartile is relative to large counties and not to the nation. The index is a measure of neighborhoods, averaged to counties.

Note: Intervals are 95% intervals from a bootstrap over counties (replicates and seed in the methods note). Births are full counts, so an interval shows how much a pooled figure depends on which counties are in the group, not sampling error in the counts. Shares are of births to mothers aged 15 to 44; mothers under 15 and mothers 45 and over are left out, and withheld cells are missing.

Source: CDC WONDER natality, births in 2018-2024; full counts of births by mother's age; counties grouped by Area Deprivation Index 2023 with ACS 2019-2023 tract population (population-weighted); pooled over counties by births (births-weighted); intervals from a bootstrap over counties

Data: https://vishalsingh.org/zipcode/report/where-births-are-falling/figures/mother_age_by_deprivation.json

### Table 8. Counties with the largest declines in births lost births at every age under 35, and counties with the largest gains added births at every age from 25

Pooled change in live births from 2018 to 2024 by mother's age band, in four equal-sized groups of the 578 named counties formed by the change in their total births

| age | age_position | change | group | n_units |
| --- | ---: | ---: | --- | ---: |
| 15 to 19 | 1 | −32.87 | Largest declines | 144 |
| 20 to 24 | 2 | −25.09 | Largest declines | 144 |
| 25 to 29 | 3 | −21.11 | Largest declines | 144 |
| 30 to 34 | 4 | −8.77 | Largest declines | 144 |
| 35 to 39 | 5 | −0.060 | Largest declines | 144 |
| 40 to 44 | 6 | 9.55 | Largest declines | 144 |
| 15 to 19 | 1 | −25.21 | Smaller declines | 145 |
| 20 to 24 | 2 | −17.85 | Smaller declines | 145 |
| 25 to 29 | 3 | −13.39 | Smaller declines | 145 |
| 30 to 34 | 4 | −0.670 | Smaller declines | 145 |
| 35 to 39 | 5 | 9.07 | Smaller declines | 145 |
| 40 to 44 | 6 | 20.34 | Smaller declines | 145 |
| 15 to 19 | 1 | −20.90 | Little change | 144 |
| 20 to 24 | 2 | −12.76 | Little change | 144 |
| 25 to 29 | 3 | −7.43 | Little change | 144 |
| 30 to 34 | 4 | 4.50 | Little change | 144 |
| 35 to 39 | 5 | 13.88 | Little change | 144 |
| 40 to 44 | 6 | 23.02 | Little change | 144 |
| 15 to 19 | 1 | −12.97 | Largest gains | 144 |
| 20 to 24 | 2 | −4.05 | Largest gains | 145 |
| 25 to 29 | 3 | 5.80 | Largest gains | 145 |
| 30 to 34 | 4 | 19.11 | Largest gains | 145 |
| 35 to 39 | 5 | 24.75 | Largest gains | 145 |
| 40 to 44 | 6 | 39.26 | Largest gains | 145 |

Note: The named counties are the 578 that CDC WONDER natality identifies in every year from 2018 to 2024: counties with 100,000 or more residents in the population vintage WONDER uses, which is earlier than the 2020 Census. WONDER pools the births of every other county in a state into one row, so those counties are not in the named counties.

Note: Counties are sorted into four groups of equal size by their percent change in total births from 2018 to 2024: the largest-decline group has falls of 8.8% or more, and the largest-gain group has gains of 2.6% or more. Births are full counts, so no interval is drawn; the groups are formed on the outcome, so an interval would not describe a draw of counties.

Note: A county that withholds a band in 2018 or 2024 is left out of that band's point.

Source: CDC WONDER natality, births in 2018-2024; full counts of births by mother's age; changes pooled over counties in each group

Data: https://vishalsingh.org/zipcode/report/where-births-are-falling/figures/age_change_by_county_change.json

## All estimates

Every fact in facts.json (155), sorted by id.

| Fact id | Value | Unit | Estimand | Population | Period | Interval | n | Weights | Source | Note |
| --- | ---: | --- | --- | --- | --- | --- | ---: | --- | --- | --- |
| age_15 | 15 | years of age | age below which WONDER's under-15 band applies; mothers under this age are left out (a definition) | mothers | all years (a definition) |   |   | none (definition) | CDC WONDER natality |   |
| age_25 | 25 | years of age | age from which births rise in the largest-gain counties (a definition of the band boundary used in the figure title) | mothers | all years (a definition) |   |   | none (definition) | CDC WONDER natality |   |
| age_35 | 35 | years of age | age below which a mother belongs to the younger groups in this chapter (a definition) | mothers | all years (a definition) |   |   | none (definition) | CDC WONDER natality |   |
| age_45 | 45 | years of age | age at which WONDER's 45-to-49 band begins; mothers of this age and older are left out (a definition) | mothers | all years (a definition) |   |   | none (definition) | CDC WONDER natality |   |
| band_1519 | 15 to 19 | years of age | age-of-mother range used in this chapter, from WONDER's five-year bands (a definition) | mothers | all years (a definition) |   |   | none (definition) | CDC WONDER natality |   |
| band_1524 | 15 to 24 | years of age | age-of-mother range used in this chapter, from WONDER's five-year bands (a definition) | mothers | all years (a definition) |   |   | none (definition) | CDC WONDER natality |   |
| band_1529 | 15 to 29 | years of age | age-of-mother range used in this chapter, from WONDER's five-year bands (a definition) | mothers | all years (a definition) |   |   | none (definition) | CDC WONDER natality |   |
| band_1544 | 15 to 44 | years of age | age-of-mother range used in this chapter, from WONDER's five-year bands (a definition) | mothers | all years (a definition) |   |   | none (definition) | CDC WONDER natality |   |
| band_2024 | 20 to 24 | years of age | age-of-mother range used in this chapter, from WONDER's five-year bands (a definition) | mothers | all years (a definition) |   |   | none (definition) | CDC WONDER natality |   |
| band_2529 | 25 to 29 | years of age | age-of-mother range used in this chapter, from WONDER's five-year bands (a definition) | mothers | all years (a definition) |   |   | none (definition) | CDC WONDER natality |   |
| band_3034 | 30 to 34 | years of age | age-of-mother range used in this chapter, from WONDER's five-year bands (a definition) | mothers | all years (a definition) |   |   | none (definition) | CDC WONDER natality |   |
| band_3539 | 35 to 39 | years of age | age-of-mother range used in this chapter, from WONDER's five-year bands (a definition) | mothers | all years (a definition) |   |   | none (definition) | CDC WONDER natality |   |
| band_3544 | 35 to 44 | years of age | age-of-mother range used in this chapter, from WONDER's five-year bands (a definition) | mothers | all years (a definition) |   |   | none (definition) | CDC WONDER natality |   |
| band_4044 | 40 to 44 | years of age | age-of-mother range used in this chapter, from WONDER's five-year bands (a definition) | mothers | all years (a definition) |   |   | none (definition) | CDC WONDER natality |   |
| births_named_2018 | 3.02 million | live births | count of live births | live births in the counties that CDC WONDER natality identifies in every year (counties with 100,000 or more residents in the population vintage WONDER uses) | 2018 |   |   | none (full count) | CDC WONDER natality |   |
| births_named_2024 | 2.88 million | live births | count of live births | live births in the counties that CDC WONDER natality identifies in every year (counties with 100,000 or more residents in the population vintage WONDER uses) | 2024 |   |   | none (full count) | CDC WONDER natality |   |
| births_pooled_2018 | 0.77 million | live births | count of live births in the pooled rows | live births in the counties that WONDER natality does not name, which it pools within each state | 2018 |   |   | none (full count) | CDC WONDER natality |   |
| births_pooled_2024 | 0.75 million | live births | count of live births in the pooled rows | live births in the counties that WONDER natality does not name, which it pools within each state | 2024 |   |   | none (full count) | CDC WONDER natality |   |
| births_published_cells | 20.4 million | live births | count of births in published cells | live births in the published age-of-mother cells of the named counties | births in 2018 to 2024 |   | 20,431,379 births | none (full count) | CDC WONDER natality |   |
| bottom_tenth_cut | 13.6% | percent of 2018 births | fall at the 10th percentile of the county changes in births | named counties | births in 2018 and 2024 |   | 578 counties | none (full count) | CDC WONDER natality |   |
| bottom_tenth_n | 58 | counties | count of counties at or below the 10th percentile of change | named counties in the bottom tenth of change in births | births in 2018 and 2024 |   |   | none (full count) | CDC WONDER natality |   |
| bottom_tenth_south | 13 | counties | count of those counties in the Census South region (16 states and DC) | named counties in the bottom tenth of change that are in the Census South region | births in 2018 and 2024 |   |   | none (full count) | CDC WONDER natality |   |
| change_2021_q4 | −0.3% | percent of 2020 births | pooled change in the count of births from 2020 to 2021, bootstrap interval over counties | live births in the most deprived quartile of the named counties by county Area Deprivation Index rank | 2020 to 2021 | 95% CI −0.8% to 0.1% | 145 counties | births in each county (a pooled ratio of sums over counties); interval from a bootstrap over counties | CDC WONDER natality |   |
| chg_3034 | +1.0% | percent of 2018 births | pooled change in the count of births from 2018 to 2024, bootstrap interval over counties | live births to mothers aged 30 to 34 in the named counties that publish the band in both years | births in 2018 and 2024 | 95% CI −0.6% to 2.6% | 578 counties | births in each county (a pooled ratio of sums over counties); interval from a bootstrap over counties | CDC WONDER natality |   |
| core_cells | 24,276 | cells | count of cells | county-year-by-age cells for mothers aged 15 to 44 in the named counties | births in 2018 to 2024 |   |   | none (full count) | CDC WONDER natality |   |
| core_withheld_cells | 11 | cells | count of cells with 1 to 9 births, which WONDER withholds | county-year-by-age cells for mothers aged 15 to 44 in the named counties | births in 2018 to 2024 |   |   | none (full count) | CDC WONDER natality |   |
| core_withheld_max_births | 99 | live births | largest number of births the withheld cells could hold: 9 births in each | withheld cells for mothers aged 15 to 44 | births in 2018 to 2024 |   |   | none (full count) | CDC WONDER natality |   |
| counties_declined | 380 | counties | count of counties with fewer births in 2024 than in 2018 | named counties | births in 2018 and 2024 |   |   | none (full count) | CDC WONDER natality |   |
| counties_rose | 197 | counties | count of counties with more births in 2024 than in 2018 | named counties | births in 2018 and 2024 |   |   | none (full count) | CDC WONDER natality |   |
| counties_unchanged | one | counties | count of counties with the same number of births in 2018 and 2024 | named counties | births in 2018 and 2024 |   |   | none (full count) | CDC WONDER natality |   |
| counties_used | 578 | counties | count of counties | counties that WONDER natality identifies in every year from 2018 to 2024 and that have births | births in 2018 to 2024 |   |   | none (full count) | CDC WONDER natality |   |
| county_median_decline | 3.8% | percent of 2018 births | median of the county changes in births, one county one count | the median named county | births in 2018 and 2024 |   | 578 counties | none (full count) | CDC WONDER natality |   |
| county_p5_decline | 16.5% | percent of 2018 births | 5th percentile of the county changes in births | named counties at the 5th percentile of change | births in 2018 and 2024 |   | 578 counties | none (full count) | CDC WONDER natality |   |
| county_p95_gain | 19.9% | percent of 2018 births | 95th percentile of the county changes in births | named counties at the 95th percentile of change | births in 2018 and 2024 |   | 578 counties | none (full count) | CDC WONDER natality |   |
| decline_2020 | 4.8% | percent of 2018 births | fall in the count of births from 2018 to 2020 | live births in the counties that CDC WONDER natality identifies in every year (counties with 100,000 or more residents in the population vintage WONDER uses) | 2018 to 2020 |   |   | none (full count) | CDC WONDER natality |   |
| decline_2023 | 5.5% | percent of 2018 births | fall in the count of births from 2018 to 2023 | live births in the counties that CDC WONDER natality identifies in every year (counties with 100,000 or more residents in the population vintage WONDER uses) | 2018 to 2023 |   |   | none (full count) | CDC WONDER natality |   |
| decline_mid | 2.8% | percent of 2018 births | fall in the pooled count of births from 2018 to 2024, bootstrap interval over counties | live births in the second and third quartiles together of the named counties by county Area Deprivation Index rank | births in 2018 and 2024 | 95% CI 1.1% to 4.5% | 289 counties | births in each county (a pooled ratio of sums over counties); interval from a bootstrap over counties | CDC WONDER natality |   |
| decline_named | 4.6% | percent of 2018 births | fall in the count of births from 2018 to 2024 | live births in the counties that CDC WONDER natality identifies in every year (counties with 100,000 or more residents in the population vintage WONDER uses) | 2018 to 2024 |   |   | none (full count) | CDC WONDER natality |   |
| decline_pooled | 3.0% | percent of 2018 births | fall in the count of births in the pooled rows from 2018 to 2024 | live births in the counties that WONDER natality does not name, which it pools within each state | 2018 to 2024 |   |   | none (full count) | CDC WONDER natality |   |
| decline_q1 | 5.6% | percent of 2018 births | fall in the pooled count of births from 2018 to 2024, bootstrap interval over counties | live births in the least deprived quartile of the named counties by county Area Deprivation Index rank | births in 2018 and 2024 | 95% CI 2.8% to 8.2% | 144 counties | births in each county (a pooled ratio of sums over counties); interval from a bootstrap over counties | CDC WONDER natality |   |
| decline_q2 | 2.6% | percent of 2018 births | fall in the pooled count of births from 2018 to 2024, bootstrap interval over counties | live births in the second quartile of the named counties by county Area Deprivation Index rank | births in 2018 and 2024 | 95% CI −0.6% to 5.3% | 145 counties | births in each county (a pooled ratio of sums over counties); interval from a bootstrap over counties | CDC WONDER natality |   |
| decline_q3 | 3.1% | percent of 2018 births | fall in the pooled count of births from 2018 to 2024, bootstrap interval over counties | live births in the third quartile of the named counties by county Area Deprivation Index rank | births in 2018 and 2024 | 95% CI 1.6% to 4.5% | 144 counties | births in each county (a pooled ratio of sums over counties); interval from a bootstrap over counties | CDC WONDER natality |   |
| decline_q4 | 7.3% | percent of 2018 births | fall in the pooled count of births from 2018 to 2024, bootstrap interval over counties | live births in the most deprived quartile of the named counties by county Area Deprivation Index rank | births in 2018 and 2024 | 95% CI 5.8% to 8.8% | 145 counties | births in each county (a pooled ratio of sums over counties); interval from a bootstrap over counties | CDC WONDER natality |   |
| decline_us | 4.3% | percent of 2018 births | fall in the count of births from 2018 to 2024 | live births in the 50 states and DC | 2018 to 2024 |   |   | none (full count) | CDC WONDER natality |   |
| empty_shells | two | counties | count of listed counties whose births WONDER folded into the state's pooled row | county identifiers that WONDER lists in every year from 2018 to 2024 but with no births (Cape May County NJ, Portsmouth city VA) | births in 2018 to 2024 |   |   | none (full count) | CDC WONDER natality |   |
| ext_min_births | 987 | live births | smallest 2018 count of births among those counties | the 20 counties in the figure of largest gains and declines | 2018 |   | 20 counties | none (full count) | CDC WONDER natality |   |
| ext_n | ten | counties | number of counties listed in each group of largest gains and largest declines (a definition) | named counties | all years (a definition) |   |   | none (definition) | CDC WONDER natality |   |
| fall_1519 | 24.6% | percent of 2018 births | pooled change in the count of births from 2018 to 2024, bootstrap interval over counties | live births to mothers aged 15 to 19 in the named counties that publish the band in both years | births in 2018 and 2024 | 95% CI 22.7% to 26.5% | 577 counties | births in each county (a pooled ratio of sums over counties); interval from a bootstrap over counties | CDC WONDER natality |   |
| fall_2024 | 16.6% | percent of 2018 births | pooled change in the count of births from 2018 to 2024, bootstrap interval over counties | live births to mothers aged 20 to 24 in the named counties that publish the band in both years | births in 2018 and 2024 | 95% CI 15.0% to 18.3% | 578 counties | births in each county (a pooled ratio of sums over counties); interval from a bootstrap over counties | CDC WONDER natality |   |
| fall_2529 | 11.2% | percent of 2018 births | pooled change in the count of births from 2018 to 2024, bootstrap interval over counties | live births to mothers aged 25 to 29 in the named counties that publish the band in both years | births in 2018 and 2024 | 95% CI 9.6% to 12.9% | 578 counties | births in each county (a pooled ratio of sums over counties); interval from a bootstrap over counties | CDC WONDER natality |   |
| fall_share_young | 3.1 percentage points | percentage points | change from 2018 to 2024 in the pooled share of births, shared bootstrap draws over counties | live births to mothers aged 15 to 44 in the counties that CDC WONDER natality names individually, births in published cells; mothers aged 15 to 24 | births in 2018 and 2024 | 95% CI 2.9 percentage points to 3.3 percentage points | 578 counties | births in each county (a pooled ratio of sums over counties); interval from a bootstrap over counties | CDC WONDER natality |   |
| fall_young_diff | 1.5 percentage points | percentage points | difference of two changes in pooled shares: fall in the share of births to mothers aged 15 to 24, most deprived quartile minus least deprived quartile, shared bootstrap draws over counties | live births to mothers aged 15 to 44 in the counties that CDC WONDER natality names individually | births in 2018 and 2024 | 95% CI 0.9 percentage points to 2.2 percentage points | 578 counties | births in each county (a pooled ratio of sums over counties); interval from a bootstrap over counties | CDC WONDER natality |   |
| fall_young_q1 | 2.5 percentage points | percentage points | change from 2018 to 2024 in the pooled share of births, shared bootstrap draws over counties | live births to mothers aged 15 to 44 in the counties that CDC WONDER natality names individually; mothers aged 15 to 24; least deprived quartile | births in 2018 and 2024 | 95% CI 1.9 percentage points to 3.0 percentage points | 144 counties | births in each county (a pooled ratio of sums over counties); interval from a bootstrap over counties | CDC WONDER natality |   |
| fall_young_q4 | 4.0 percentage points | percentage points | change from 2018 to 2024 in the pooled share of births, shared bootstrap draws over counties | live births to mothers aged 15 to 44 in the counties that CDC WONDER natality names individually; mothers aged 15 to 24; most deprived quartile | births in 2018 and 2024 | 95% CI 3.7 percentage points to 4.4 percentage points | 145 counties | births in each county (a pooled ratio of sums over counties); interval from a bootstrap over counties | CDC WONDER natality |   |
| g1_chg_3539 | −0.1% | percent of 2018 births | pooled change in the count of births from 2018 to 2024 | live births to mothers aged 35 to 39 in the quarter of named counties with the largest declines that publish the band in both years | births in 2018 and 2024 |   | 144 counties | births (a pooled ratio of totals over counties; full counts) | CDC WONDER natality |   |
| g1_fall_1519 | 32.9% | percent of 2018 births | pooled change in the count of births from 2018 to 2024 | live births to mothers aged 15 to 19 in the quarter of named counties with the largest declines that publish the band in both years | births in 2018 and 2024 |   | 144 counties | births (a pooled ratio of totals over counties; full counts) | CDC WONDER natality |   |
| g1_fall_2024 | 25.1% | percent of 2018 births | pooled change in the count of births from 2018 to 2024 | live births to mothers aged 20 to 24 in the quarter of named counties with the largest declines that publish the band in both years | births in 2018 and 2024 |   | 144 counties | births (a pooled ratio of totals over counties; full counts) | CDC WONDER natality |   |
| g1_fall_2529 | 21.1% | percent of 2018 births | pooled change in the count of births from 2018 to 2024 | live births to mothers aged 25 to 29 in the quarter of named counties with the largest declines that publish the band in both years | births in 2018 and 2024 |   | 144 counties | births (a pooled ratio of totals over counties; full counts) | CDC WONDER natality |   |
| g1_fall_3034 | 8.8% | percent of 2018 births | pooled change in the count of births from 2018 to 2024 | live births to mothers aged 30 to 34 in the quarter of named counties with the largest declines that publish the band in both years | births in 2018 and 2024 |   | 144 counties | births (a pooled ratio of totals over counties; full counts) | CDC WONDER natality |   |
| g1_rise_4044 | 9.6% | percent of 2018 births | pooled change in the count of births from 2018 to 2024 | live births to mothers aged 40 to 44 in the quarter of named counties with the largest declines that publish the band in both years | births in 2018 and 2024 |   | 144 counties | births (a pooled ratio of totals over counties; full counts) | CDC WONDER natality |   |
| g4_fall_1519 | 13.0% | percent of 2018 births | pooled change in the count of births from 2018 to 2024 | live births to mothers aged 15 to 19 in the quarter of named counties with the largest gains that publish the band in both years | births in 2018 and 2024 |   | 144 counties | births (a pooled ratio of totals over counties; full counts) | CDC WONDER natality |   |
| g4_fall_2024 | 4.0% | percent of 2018 births | pooled change in the count of births from 2018 to 2024 | live births to mothers aged 20 to 24 in the quarter of named counties with the largest gains that publish the band in both years | births in 2018 and 2024 |   | 145 counties | births (a pooled ratio of totals over counties; full counts) | CDC WONDER natality |   |
| g4_rise_2529 | 5.8% | percent of 2018 births | pooled change in the count of births from 2018 to 2024 | live births to mothers aged 25 to 29 in the quarter of named counties with the largest gains that publish the band in both years | births in 2018 and 2024 |   | 145 counties | births (a pooled ratio of totals over counties; full counts) | CDC WONDER natality |   |
| g4_rise_4044 | 39.3% | percent of 2018 births | pooled change in the count of births from 2018 to 2024 | live births to mothers aged 40 to 44 in the quarter of named counties with the largest gains that publish the band in both years | births in 2018 and 2024 |   | 145 counties | births (a pooled ratio of totals over counties; full counts) | CDC WONDER natality |   |
| gap_2021_q1 | 1.9 percentage points | percentage points | difference of two pooled changes in the count of births, shared bootstrap draws over counties | live births in the named counties, change from 2020 to 2021 in the least deprived quartile minus the change in the most deprived quartile | 2020 to 2021 | 95% CI 1.0 percentage points to 3.0 percentage points | 578 counties | births in each county (a pooled ratio of sums over counties); interval from a bootstrap over counties | CDC WONDER natality |   |
| gap_2021_q2 | 2.5 percentage points | percentage points | difference of two pooled changes in the count of births, shared bootstrap draws over counties | live births in the named counties, change from 2020 to 2021 in the second quartile minus the change in the most deprived quartile | 2020 to 2021 | 95% CI 1.4 percentage points to 3.6 percentage points | 578 counties | births in each county (a pooled ratio of sums over counties); interval from a bootstrap over counties | CDC WONDER natality |   |
| gap_2021_q3 | 1.3 percentage points | percentage points | difference of two pooled changes in the count of births, shared bootstrap draws over counties | live births in the named counties, change from 2020 to 2021 in the third quartile minus the change in the most deprived quartile | 2020 to 2021 | 95% CI 0.5 percentage points to 2.2 percentage points | 578 counties | births in each county (a pooled ratio of sums over counties); interval from a bootstrap over counties | CDC WONDER natality |   |
| gap_older_2018 | 11.3 percentage points | percentage points | difference of two pooled shares of births, shared bootstrap draws over counties | live births to mothers aged 15 to 44 in the counties that CDC WONDER natality names individually; mothers aged 35 to 44; least deprived quartile minus most deprived quartile | 2018 | 95% CI 10.0 percentage points to 12.5 percentage points | 578 counties | births in each county (a pooled ratio of sums over counties); interval from a bootstrap over counties | CDC WONDER natality |   |
| gap_older_2024 | 12.0 percentage points | percentage points | difference of two pooled shares of births, shared bootstrap draws over counties | live births to mothers aged 15 to 44 in the counties that CDC WONDER natality names individually; mothers aged 35 to 44; least deprived quartile minus most deprived quartile | 2024 | 95% CI 10.5 percentage points to 13.5 percentage points | 578 counties | births in each county (a pooled ratio of sums over counties); interval from a bootstrap over counties | CDC WONDER natality |   |
| gap_young_2018 | 14.0 percentage points | percentage points | difference of two pooled shares of births, shared bootstrap draws over counties | live births to mothers aged 15 to 44 in the counties that CDC WONDER natality names individually; mothers aged 15 to 24; most deprived quartile minus least deprived quartile | 2018 | 95% CI 12.3 percentage points to 15.8 percentage points | 578 counties | births in each county (a pooled ratio of sums over counties); interval from a bootstrap over counties | CDC WONDER natality |   |
| gap_young_2024 | 12.4 percentage points | percentage points | difference of two pooled shares of births, shared bootstrap draws over counties | live births to mothers aged 15 to 44 in the counties that CDC WONDER natality names individually; mothers aged 15 to 24; most deprived quartile minus least deprived quartile | 2024 | 95% CI 11.0 percentage points to 14.0 percentage points | 578 counties | births in each county (a pooled ratio of sums over counties); interval from a bootstrap over counties | CDC WONDER natality |   |
| group_cut_decline | 8.8% | percent of 2018 births | smallest fall among the counties in the largest-decline quarter (the group's own boundary member) | the quarter of named counties with the largest declines | births in 2018 and 2024 |   | 144 counties | none (full count) | CDC WONDER natality |   |
| group_cut_gain | 2.6% | percent of 2018 births | smallest gain among the counties in the largest-gain quarter (the group's own boundary member) | the quarter of named counties with the largest gains | births in 2018 and 2024 |   | 145 counties | none (full count) | CDC WONDER natality |   |
| group_n_largest_declines | 144 | counties | count of counties | named counties in the quarter with the largest declines in births | births in 2018 and 2024 |   |   | none (full count) | CDC WONDER natality |   |
| group_n_largest_gains | 145 | counties | count of counties | named counties in the quarter with the largest gains in births | births in 2018 and 2024 |   |   | none (full count) | CDC WONDER natality |   |
| group_total_decline | 13.5% | percent of 2018 births | fall in the pooled count of births from 2018 to 2024 | live births in the quarter of named counties with the largest declines | births in 2018 and 2024 |   |   | births (a pooled ratio of totals over counties; full counts) | CDC WONDER natality |   |
| group_total_gain | 11.2% | percent of 2018 births | rise in the pooled count of births from 2018 to 2024 | live births in the quarter of named counties with the largest gains | births in 2018 and 2024 |   |   | births (a pooled ratio of totals over counties; full counts) | CDC WONDER natality |   |
| largest_decline | 23.9% | percent of 2018 births | fall in the county's count of births from 2018 to 2024 | the named county with the largest decline (St. Louis city, MO) | births in 2018 and 2024 |   | 4,127 births | none (full count) | CDC WONDER natality |   |
| largest_gain | 67.6% | percent of 2018 births | rise in the county's count of births from 2018 to 2024 | the named county with the largest gain (Kaufman County, TX) | births in 2018 and 2024 |   | 1,786 births | none (full count) | CDC WONDER natality |   |
| map_beyond_key | 34 | counties | count of counties with a change of more than the key limit in either direction | named counties whose change is beyond the end of the map key | births in 2018 and 2024 |   | 578 counties | none (full count) | CDC WONDER natality |   |
| map_key_limit | 20 | percent of 2018 births | change in births at which the map key ends, in each direction (a display choice) | named counties on the map | births in 2018 and 2024 |   |   | none (full count) | CDC WONDER natality |   |
| min_county_births | 613 | live births | smallest annual count of births in any of the named counties | the named county-year with the fewest births | births in 2018 to 2024 |   | 4,046 county-years | none (full count) | CDC WONDER natality |   |
| named_below_threshold | one | counties | count of named counties with a 2020 Census population below the threshold | named counties outside Connecticut whose 2020 Census population is below the threshold | 2020 Census |   | 3,135 counties | none (full count) | CDC WONDER natality |   |
| pctl_5 | 5th | percentile | percentile used to describe the spread of county changes (a definition) | named counties | all years (a definition) |   |   | none (definition) | CDC WONDER natality |   |
| pctl_95 | 95th | percentile | percentile used to describe the spread of county changes (a definition) | named counties | all years (a definition) |   |   | none (definition) | CDC WONDER natality |   |
| pooled_above_threshold | 27 | counties | count of unnamed counties with a 2020 Census population of at least the threshold | counties outside Connecticut that WONDER pools although their 2020 Census population is 100,000 or more | 2020 Census |   | 3,135 counties | none (full count) | CDC WONDER natality |   |
| pooled_rows | 49 | states | count of pooled rows | states with counties below the threshold, each with one pooled Unidentified Counties row | births in 2018 to 2024 |   |   | none (full count) | CDC WONDER natality |   |
| q1_south_share | 24% | percent of counties | share in the Census South region (16 states and DC) | counties in the least deprived quartile of the named counties | births in 2018 to 2024 |   | 144 counties | none (full count) | CDC WONDER natality |   |
| q1_vs_mid_pp | 2.7 percentage points | percentage points | difference of two pooled falls in the count of births, shared bootstrap draws over counties | live births in the named counties, fall in births for the least deprived quartile minus the second and third quartiles together | births in 2018 and 2024 | 95% CI −0.5 percentage points to 5.9 percentage points | 578 counties | births in each county (a pooled ratio of sums over counties); interval from a bootstrap over counties | CDC WONDER natality |   |
| q4_midwest | 9.2% | percent of 2018 births | fall in the pooled count of births from 2018 to 2024, bootstrap interval over counties | live births in the most deprived quartile of the named counties in the Census Midwest region | births in 2018 and 2024 | 95% CI 6.3% to 11.6% | 53 counties | births in each county (a pooled ratio of sums over counties); interval from a bootstrap over counties | CDC WONDER natality |   |
| q4_northeast | 9.3% | percent of 2018 births | fall in the pooled count of births from 2018 to 2024, bootstrap interval over counties | live births in the most deprived quartile of the named counties in the Census Northeast region | births in 2018 and 2024 | 95% CI 6.6% to 11.8% | 22 counties | births in each county (a pooled ratio of sums over counties); interval from a bootstrap over counties | CDC WONDER natality |   |
| q4_south | 5.0% | percent of 2018 births | fall in the pooled count of births from 2018 to 2024, bootstrap interval over counties | live births in the most deprived quartile of the named counties in the Census South region | births in 2018 and 2024 | 95% CI 2.9% to 7.0% | 65 counties | births in each county (a pooled ratio of sums over counties); interval from a bootstrap over counties | CDC WONDER natality |   |
| q4_south_share | 45% | percent of counties | share in the Census South region (16 states and DC) | counties in the most deprived quartile of the named counties | births in 2018 to 2024 |   | 145 counties | none (full count) | CDC WONDER natality |   |
| q4_vs_mid_pp | 4.5 percentage points | percentage points | difference of two pooled falls in the count of births, shared bootstrap draws over counties | live births in the named counties, fall in births for the most deprived quartile minus the second and third quartiles together | births in 2018 and 2024 | 95% CI 2.3 percentage points to 6.7 percentage points | 578 counties | births in each county (a pooled ratio of sums over counties); interval from a bootstrap over counties | CDC WONDER natality |   |
| q4_vs_q1_pp | 1.7 percentage points | percentage points | difference of two pooled falls in the count of births, shared bootstrap draws over counties | live births in the named counties, fall in births for the most deprived quartile minus the least deprived quartile | births in 2018 and 2024 | 95% CI −1.2 percentage points to 5.0 percentage points | 578 counties | births in each county (a pooled ratio of sums over counties); interval from a bootstrap over counties | CDC WONDER natality |   |
| q4_west | 9.3% | percent of 2018 births | fall in the pooled count of births from 2018 to 2024; no interval, because the cell holds fewer than 10 counties (all of them lost births) | live births in the most deprived quartile of the named counties in the Census West region | births in 2018 and 2024 |   | 5 counties | births (a pooled ratio of totals over counties; full counts) | CDC WONDER natality |   |
| q4_west_n | 5 | counties | count of counties | counties in the most deprived quartile of the named counties in the Census West region | births in 2018 to 2024 |   |   | none (full count) | CDC WONDER natality |   |
| region_midwest | −8.4% | percent of 2018 births | pooled change in the count of births from 2018 to 2024, bootstrap interval over counties | live births in the named counties in the Census Midwest region | births in 2018 and 2024 | 95% CI −10.4% to −6.3% | 137 counties | births in each county (a pooled ratio of sums over counties); interval from a bootstrap over counties | CDC WONDER natality |   |
| region_northeast | −5.0% | percent of 2018 births | pooled change in the count of births from 2018 to 2024, bootstrap interval over counties | live births in the named counties in the Census Northeast region | births in 2018 and 2024 | 95% CI −7.3% to −2.5% | 110 counties | births in each county (a pooled ratio of sums over counties); interval from a bootstrap over counties | CDC WONDER natality |   |
| region_south | 0.0% | percent of 2018 births | pooled change in the count of births from 2018 to 2024, bootstrap interval over counties | live births in the named counties in the Census South region | births in 2018 and 2024 | 95% CI −1.7% to 1.8% | 232 counties | births in each county (a pooled ratio of sums over counties); interval from a bootstrap over counties | CDC WONDER natality |   |
| region_west | −7.9% | percent of 2018 births | pooled change in the count of births from 2018 to 2024, bootstrap interval over counties | live births in the named counties in the Census West region | births in 2018 and 2024 | 95% CI −10.7% to −5.0% | 99 counties | births in each county (a pooled ratio of sums over counties); interval from a bootstrap over counties | CDC WONDER natality |   |
| rise_2024 | 0.9% | percent of 2023 births | rise in the count of births from 2023 to 2024 | live births in the counties that CDC WONDER natality identifies in every year (counties with 100,000 or more residents in the population vintage WONDER uses) | 2023 to 2024 |   |   | none (full count) | CDC WONDER natality |   |
| rise_3539 | 9.2% | percent of 2018 births | pooled change in the count of births from 2018 to 2024, bootstrap interval over counties | live births to mothers aged 35 to 39 in the named counties that publish the band in both years | births in 2018 and 2024 | 95% CI 7.3% to 11.0% | 578 counties | births in each county (a pooled ratio of sums over counties); interval from a bootstrap over counties | CDC WONDER natality |   |
| rise_4044 | 19.5% | percent of 2018 births | pooled change in the count of births from 2018 to 2024, bootstrap interval over counties | live births to mothers aged 40 to 44 in the named counties that publish the band in both years | births in 2018 and 2024 | 95% CI 17.0% to 22.3% | 578 counties | births in each county (a pooled ratio of sums over counties); interval from a bootstrap over counties | CDC WONDER natality |   |
| rise_older_diff | 0.7 percentage points | percentage points | difference of two changes in pooled shares: rise in the share of births to mothers aged 35 to 44, least deprived quartile minus most deprived quartile, shared bootstrap draws over counties | live births to mothers aged 15 to 44 in the counties that CDC WONDER natality names individually | births in 2018 and 2024 | 95% CI 0.3 percentage points to 1.2 percentage points | 578 counties | births in each county (a pooled ratio of sums over counties); interval from a bootstrap over counties | CDC WONDER natality |   |
| rise_older_q1 | 3.5 percentage points | percentage points | change from 2018 to 2024 in the pooled share of births, shared bootstrap draws over counties | live births to mothers aged 15 to 44 in the counties that CDC WONDER natality names individually; mothers aged 35 to 44; least deprived quartile | births in 2018 and 2024 | 95% CI 3.1 percentage points to 3.9 percentage points | 144 counties | births in each county (a pooled ratio of sums over counties); interval from a bootstrap over counties | CDC WONDER natality |   |
| rise_older_q4 | 2.8 percentage points | percentage points | change from 2018 to 2024 in the pooled share of births, shared bootstrap draws over counties | live births to mothers aged 15 to 44 in the counties that CDC WONDER natality names individually; mothers aged 35 to 44; most deprived quartile | births in 2018 and 2024 | 95% CI 2.5 percentage points to 3.1 percentage points | 145 counties | births in each county (a pooled ratio of sums over counties); interval from a bootstrap over counties | CDC WONDER natality |   |
| rise_share_older | 3.2 percentage points | percentage points | change from 2018 to 2024 in the pooled share of births, shared bootstrap draws over counties | live births to mothers aged 15 to 44 in the counties that CDC WONDER natality names individually, births in published cells; mothers aged 35 to 44 | births in 2018 and 2024 | 95% CI 3.1 percentage points to 3.4 percentage points | 578 counties | births in each county (a pooled ratio of sums over counties); interval from a bootstrap over counties | CDC WONDER natality |   |
| share_declined | 65.7% | percent of named counties | share of counties with fewer births in 2024 than in 2018 | named counties | births in 2018 and 2024 |   | 578 counties | none (full count) | CDC WONDER natality |   |
| share_named_2018 | 79.7% | percent of live births in the United States | share of births in the named counties | live births in the 50 states and DC | 2018 |   |   | none (full count) | CDC WONDER natality |   |
| share_named_2024 | 79.4% | percent of live births in the United States | share of births in the named counties | live births in the 50 states and DC | 2024 |   |   | none (full count) | CDC WONDER natality |   |
| share_older_2018 | 19.7% | percent of live births to mothers aged 15 to 44 | pooled share of births in published cells, bootstrap interval over counties | live births to mothers aged 15 to 44 in the counties that CDC WONDER natality names individually, births in published cells; mothers aged 35 to 44 | 2018 | 95% CI 18.9% to 20.5% | 578 counties | births in each county (a pooled ratio of sums over counties); interval from a bootstrap over counties | CDC WONDER natality |   |
| share_older_2024 | 22.9% | percent of live births to mothers aged 15 to 44 | pooled share of births in published cells, bootstrap interval over counties | live births to mothers aged 15 to 44 in the counties that CDC WONDER natality names individually, births in published cells; mothers aged 35 to 44 | 2024 | 95% CI 22.1% to 23.8% | 578 counties | births in each county (a pooled ratio of sums over counties); interval from a bootstrap over counties | CDC WONDER natality |   |
| share_older_q1_2018 | 24.3% | percent of live births to mothers aged 15 to 44 | pooled share of births in published cells, bootstrap interval over counties | live births to mothers aged 15 to 44 in the counties that CDC WONDER natality names individually in the least deprived quartile of counties by deprivation rank, mothers aged 35 to 44 | 2018 | 95% CI 23.1% to 25.5% | 144 counties | births in each county (a pooled ratio of sums over counties); interval from a bootstrap over counties | CDC WONDER natality |   |
| share_older_q1_2024 | 27.8% | percent of live births to mothers aged 15 to 44 | pooled share of births in published cells, bootstrap interval over counties | live births to mothers aged 15 to 44 in the counties that CDC WONDER natality names individually in the least deprived quartile of counties by deprivation rank, mothers aged 35 to 44 | 2024 | 95% CI 26.4% to 29.2% | 144 counties | births in each county (a pooled ratio of sums over counties); interval from a bootstrap over counties | CDC WONDER natality |   |
| share_older_q4_2018 | 13.0% | percent of live births to mothers aged 15 to 44 | pooled share of births in published cells, bootstrap interval over counties | live births to mothers aged 15 to 44 in the counties that CDC WONDER natality names individually in the most deprived quartile of counties by deprivation rank, mothers aged 35 to 44 | 2018 | 95% CI 12.5% to 13.5% | 145 counties | births in each county (a pooled ratio of sums over counties); interval from a bootstrap over counties | CDC WONDER natality |   |
| share_older_q4_2024 | 15.8% | percent of live births to mothers aged 15 to 44 | pooled share of births in published cells, bootstrap interval over counties | live births to mothers aged 15 to 44 in the counties that CDC WONDER natality names individually in the most deprived quartile of counties by deprivation rank, mothers aged 35 to 44 | 2024 | 95% CI 15.1% to 16.5% | 145 counties | births in each county (a pooled ratio of sums over counties); interval from a bootstrap over counties | CDC WONDER natality |   |
| share_young_2018 | 21.8% | percent of live births to mothers aged 15 to 44 | pooled share of births in published cells, bootstrap interval over counties | live births to mothers aged 15 to 44 in the counties that CDC WONDER natality names individually, births in published cells; mothers aged 15 to 24 | 2018 | 95% CI 21.0% to 22.7% | 578 counties | births in each county (a pooled ratio of sums over counties); interval from a bootstrap over counties | CDC WONDER natality |   |
| share_young_2024 | 18.7% | percent of live births to mothers aged 15 to 44 | pooled share of births in published cells, bootstrap interval over counties | live births to mothers aged 15 to 44 in the counties that CDC WONDER natality names individually, births in published cells; mothers aged 15 to 24 | 2024 | 95% CI 18.0% to 19.5% | 578 counties | births in each county (a pooled ratio of sums over counties); interval from a bootstrap over counties | CDC WONDER natality |   |
| share_young_q1_2018 | 17.0% | percent of live births to mothers aged 15 to 44 | pooled share of births in published cells, bootstrap interval over counties | live births to mothers aged 15 to 44 in the counties that CDC WONDER natality names individually in the least deprived quartile of counties by deprivation rank, mothers aged 15 to 24 | 2018 | 95% CI 15.6% to 18.3% | 144 counties | births in each county (a pooled ratio of sums over counties); interval from a bootstrap over counties | CDC WONDER natality |   |
| share_young_q1_2024 | 14.5% | percent of live births to mothers aged 15 to 44 | pooled share of births in published cells, bootstrap interval over counties | live births to mothers aged 15 to 44 in the counties that CDC WONDER natality names individually in the least deprived quartile of counties by deprivation rank, mothers aged 15 to 24 | 2024 | 95% CI 13.4% to 15.5% | 144 counties | births in each county (a pooled ratio of sums over counties); interval from a bootstrap over counties | CDC WONDER natality |   |
| share_young_q4_2018 | 30.9% | percent of live births to mothers aged 15 to 44 | pooled share of births in published cells, bootstrap interval over counties | live births to mothers aged 15 to 44 in the counties that CDC WONDER natality names individually in the most deprived quartile of counties by deprivation rank, mothers aged 15 to 24 | 2018 | 95% CI 29.9% to 32.1% | 145 counties | births in each county (a pooled ratio of sums over counties); interval from a bootstrap over counties | CDC WONDER natality |   |
| share_young_q4_2024 | 26.9% | percent of live births to mothers aged 15 to 44 | pooled share of births in published cells, bootstrap interval over counties | live births to mothers aged 15 to 44 in the counties that CDC WONDER natality names individually in the most deprived quartile of counties by deprivation rank, mothers aged 15 to 24 | 2024 | 95% CI 25.8% to 28.0% | 145 counties | births in each county (a pooled ratio of sums over counties); interval from a bootstrap over counties | CDC WONDER natality |   |
| slope_u | −0.7 percentage points | percentage points of change in births per ten points of county deprivation rank | slope of the county change in births (percent) on county Area Deprivation Index rank, unweighted, one county one count, bootstrap interval over counties | named counties | births in 2018 and 2024 | 95% CI −1.1 percentage points to −0.3 percentage points | 578 counties | none (each county counts once) | CDC WONDER natality |   |
| slope_w | +0.5 percentage points | percentage points of change in births per ten points of county deprivation rank | slope of the county change in births (percent) on county Area Deprivation Index rank, weighted by 2018 births, bootstrap interval over counties | named counties | births in 2018 and 2024 | 95% CI −0.2 percentage points to 1.2 percentage points | 578 counties | 2018 births | CDC WONDER natality |   |
| south_counties | 232 | counties | count of counties | named counties in the Census South region (16 states and DC) | births in 2018 to 2024 |   |   | none (full count) | CDC WONDER natality |   |
| south_minus_midwest | 8.4 percentage points | percentage points | difference of two pooled changes in the count of births, shared bootstrap draws over counties | live births in the named counties, South minus Midwest | births in 2018 and 2024 | 95% CI 5.7 percentage points to 11.1 percentage points | 578 counties | births in each county (a pooled ratio of sums over counties); interval from a bootstrap over counties | CDC WONDER natality |   |
| south_minus_northeast | 5.0 percentage points | percentage points | difference of two pooled changes in the count of births, shared bootstrap draws over counties | live births in the named counties, South minus Northeast | births in 2018 and 2024 | 95% CI 2.0 percentage points to 7.9 percentage points | 578 counties | births in each county (a pooled ratio of sums over counties); interval from a bootstrap over counties | CDC WONDER natality |   |
| south_minus_west | 7.9 percentage points | percentage points | difference of two pooled changes in the count of births, shared bootstrap draws over counties | live births in the named counties, South minus West | births in 2018 and 2024 | 95% CI 4.3 percentage points to 11.3 percentage points | 578 counties | births in each county (a pooled ratio of sums over counties); interval from a bootstrap over counties | CDC WONDER natality |   |
| south_q123 | +1.1% | percent of 2018 births | pooled change in the count of births from 2018 to 2024, bootstrap interval over counties | live births in the three less deprived quartiles of the named counties in the Census South region | births in 2018 and 2024 | 95% CI −0.9% to 3.4% | 167 counties | births in each county (a pooled ratio of sums over counties); interval from a bootstrap over counties | CDC WONDER natality |   |
| south_q4_gap | 6.1 percentage points | percentage points | difference of two pooled changes in the count of births, shared bootstrap draws over counties | live births in the named counties in the Census South region, three less deprived quartiles minus the most deprived | births in 2018 and 2024 | 95% CI 3.2 percentage points to 9.0 percentage points | 578 counties | births in each county (a pooled ratio of sums over counties); interval from a bootstrap over counties | CDC WONDER natality |   |
| south_share | 40.1% | percent of named counties | share of named counties in the Census South region (16 states and DC) | named counties | births in 2018 to 2024 |   | 578 counties | none (full count) | CDC WONDER natality |   |
| south_states | 16 | states | states in the Census South region, besides the District of Columbia (a definition) | states | all years (a definition) |   |   | none (definition) | CDC WONDER natality |   |
| ss_actual_p5 | 16.6% | percent of 2018 births | fall in the county's births to mothers aged 15 to 44, 5th percentile across counties | named counties at the 5th percentile of change in births to mothers aged 15 to 44 | births in 2018 and 2024 |   | 577 counties | none (full count) | CDC WONDER natality |   |
| ss_actual_p95 | 19.8% | percent of 2018 births | gain in the county's births to mothers aged 15 to 44, 95th percentile across counties | named counties at the 95th percentile of change in births to mothers aged 15 to 44 | births in 2018 and 2024 |   | 577 counties | none (full count) | CDC WONDER natality |   |
| ss_corr_u | −0.03 | correlation coefficient | correlation, unweighted, one county one count, between the change in births that a county's 2018 age mix implies (each band changing at the all-county rate) and its change; county-cluster bootstrap interval | named counties that publish all six age bands in 2018 and 2024 | births in 2018 and 2024 | 95% CI −0.11 to 0.04 | 577 counties | none (each county counts once) | CDC WONDER natality |   |
| ss_corr_w | −0.16 | correlation coefficient | correlation, weighted by 2018 births to mothers aged 15 to 44, between the change in births that a county's 2018 age mix implies (each band changing at the all-county rate) and its change; county-cluster bootstrap interval | named counties that publish all six age bands in 2018 and 2024 | births in 2018 and 2024 | 95% CI −0.26 to −0.06 | 577 counties | 2018 births to mothers aged 15 to 44 | CDC WONDER natality |   |
| ss_counties | 577 | counties | count of counties | named counties that publish all six age bands in 2018 and 2024 | births in 2018 and 2024 |   |   | none (full count) | CDC WONDER natality |   |
| ss_implied_p5 | 8.9% | percent of 2018 births | fall implied by the county's 2018 age mix, 5th percentile across counties | named counties at the 5th percentile of implied change | births in 2018 and 2024 |   | 577 counties | none (full count) | CDC WONDER natality |   |
| ss_implied_p95 | 1.3% | percent of 2018 births | fall implied by the county's 2018 age mix, 95th percentile across counties | named counties at the 95th percentile of implied change | births in 2018 and 2024 |   | 577 counties | none (full count) | CDC WONDER natality |   |
| ss_left_out | one | counties | count of counties left out of the shift-share check | named counties with a withheld age cell in 2018 or 2024 | births in 2018 and 2024 |   |   | none (full count) | CDC WONDER natality |   |
| tail_cells | 12,138 | cells | count of cells | county-year-by-age cells for mothers under 15 and 45 and over in the named counties | births in 2018 to 2024 |   |   | none (full count) | CDC WONDER natality |   |
| tail_published_share | 0.26% | percent of live births | share of those births to mothers under 15 or 45 and over, in published cells | live births in the published age-of-mother cells of the named counties | births in 2018 to 2024 |   |   | none (full count) | CDC WONDER natality |   |
| tail_withheld_cells | 6,594 | cells | count of cells with 1 to 9 births, which WONDER withholds | county-year-by-age cells for mothers under 15 and 45 and over in the named counties | births in 2018 to 2024 |   |   | none (full count) | CDC WONDER natality |   |
| tail_withheld_max_share | 0.29% | percent of live births | largest share of births the withheld tail cells could hold: 9 births in each, over births in published cells | live births in the published age-of-mother cells of the named counties | births in 2018 to 2024 |   |   | none (full count) | CDC WONDER natality |   |
| thin_cell_limit | 10 | counties | fewest counties for which a bootstrap interval over counties is printed (the library's minimum number of clusters) | cells of the region-by-quartile figure | births in 2018 and 2024 |   |   | none (full count) | CDC WONDER natality |   |
| top_tenth_cut | 10.2% | percent of 2018 births | gain at the 90th percentile of the county changes in births | named counties | births in 2018 and 2024 |   | 578 counties | none (full count) | CDC WONDER natality |   |
| top_tenth_n | 58 | counties | count of counties at or above the 90th percentile of change | named counties in the top tenth of change in births | births in 2018 and 2024 |   |   | none (full count) | CDC WONDER natality |   |
| top_tenth_other_max | 5 | counties | largest count of those counties in any state other than Texas and Florida | named counties in the top tenth of change, in the state with the next most | births in 2018 and 2024 |   |   | none (full count) | CDC WONDER natality |   |
| top_tenth_south | 43 | counties | count of those counties in the Census South region (16 states and DC) | named counties in the top tenth of change that are in the Census South region | births in 2018 and 2024 |   |   | none (full count) | CDC WONDER natality |   |
| top_tenth_texas | 12 | counties | count of those counties in Texas, which equals the count in Florida (asserted in the build) | named counties in the top tenth of change that are in Texas, a count that Florida also has | births in 2018 and 2024 |   |   | none (full count) | CDC WONDER natality |   |
| us_births_2018 | 3.79 million | live births | count of live births | live births in the 50 states and DC | 2018 |   |   | none (full count) | CDC WONDER natality |   |
| us_births_2024 | 3.63 million | live births | count of live births | live births in the 50 states and DC | 2024 |   |   | none (full count) | CDC WONDER natality |   |
| wonder_suppression_floor | 10 | births | a cell with fewer births than this is withheld (a definition) | cells of the WONDER natality files | births in 2018 to 2024 |   |   | none (definition) | CDC WONDER natality |   |
| wonder_threshold | 100,000 | residents | population size at which WONDER identifies a county, applied to an earlier population vintage than the 2020 census (a definition) | counties that CDC WONDER natality identifies | births in 2018 to 2024 |   |   | none (definition) | CDC WONDER natality |   |

## Definitions

- Bootstrap interval: A range for an estimate found by recomputing it on many random resamples of the data and reading the range from the results. It shows how much the estimate would vary from sample to sample.
- Area Deprivation Index (ADI): A score that ranks neighborhoods by income, education, employment and housing conditions drawn from the American Community Survey. In this book a higher national percentile means a more deprived neighborhood.
- Population-weighted: Averaged so that each place counts in proportion to its number of residents, which lets a large county move the result more than a small one. An unweighted average counts every place equally.
- Suppression: The withholding of a number because the count behind it is too small to publish without risking the identification of a person. A suppressed value is missing, not zero.
- Ecological inference: Drawing conclusions about individuals from data about places or groups. A pattern across counties need not hold for the people inside them.

## Methods

This chapter compares births from 2018 to 2024 across the 578 counties that CDC WONDER names individually in every year, a set chosen by population size in a vintage earlier than the 2020 Census and not by outcome, and across groups of those counties by deprivation quartile and region. The figures are counts and shares of births, not birth rates, because no file used here holds women by age for counties, and births by mother's age use six of nine bands because most cells of the youngest and oldest bands are withheld. Pooled changes weight each county by its births, and counties that WONDER pools into an Unidentified Counties row have no county value. Births are full counts, so the intervals show dependence on which counties are in a group and not sampling error.

### Which births and counties are used

Births are counts from CDC WONDER natality for 2018 to 2024 (2024 is final), by county of the mother's residence. The 578 named counties are the counties that WONDER identifies in every year and that have births in every year: counties with 100,000 or more residents in the population vintage WONDER uses, which is earlier than the 2020 Census. By 2020 Census population 27 counties outside Connecticut that have 100,000 or more residents are still pooled, and one named county is below it.

WONDER pools every other county in a state into one Unidentified Counties row (49 rows), which is used only for the comparison of the named counties with the rest of the country. WONDER lists two further identifiers in every year with no births (Cape May County NJ, Portsmouth city VA), which are left out. The panel is defined by size, not by outcome, and every named county appears in all seven years. County totals have no withheld cell.

Explains Table 1 and Table 2.

### How births by mother's age and withheld cells are handled

Births by mother's age come in nine bands. WONDER withholds cells of 1 to 9 births. A withheld cell is missing: it leaves the numerator and the denominator, which is the births in the published cells, and nothing is derived from a county total minus the published cells.

The six bands from 15 to 19 through 40 to 44 have 11 withheld cells of 24,276, which could hold at most 99 births. The under-15 band and the 45-and-over bands are left out because most of their cells are withheld (6,594 of 12,138); the published tail cells hold 0.26% of births and the withheld ones at most 0.29%.

A share is the births to mothers in an age range over births in the published cells for mothers aged 15 to 44, pooled over counties. A change in a band compares the counties that publish the band in both 2018 and 2024. No rate is computed: no file used here holds women by age for counties.

The shift-share check applies the all-county change of each age band to each county's 2018 age mix and correlates the implied change with the observed change in births to mothers aged 15 to 44.

Explains Table 6, Table 7 and Table 8.

### How counties are ranked, grouped and weighted

A tract's Area Deprivation Index national rank is the unweighted mean of the ranks of its block groups. A county's deprivation rank is the mean of its tracts' ranks weighted by 2019-2023 tract population, with Connecticut tracts assigned to the eight legacy counties; quartiles hold equal numbers of the named counties. Regions are the four Census Bureau regions (the South has 16 states and DC).

Pooled changes, shares and differences are ratios of sums over counties, so they weight each county by its births.

The map draws every county of the book's county map; counties that WONDER does not name have no value.

Explains Table 4, Table 5 and Table 2.

### What the intervals do and do not show

Births are full counts, so these intervals show dependence on which counties are in a group and not sampling error. A region-by-quartile cell with fewer than 10 counties shows no interval, because a bootstrap over so few counties is mechanical. The groups of counties formed by the change in total births are defined on the outcome and carry no interval.

### How the bootstrap intervals are computed

Pooled changes, shares and differences carry 95% percentile intervals from a bootstrap that resamples whole counties (2,000 replicates, seed 12345), with the same draws for every statistic so that a difference has its own interval. Correlations and slopes use a county-cluster bootstrap.

## What these data cannot show

The chapter reports counts and shares of births and no rates. A birth rate needs the number of women of each age as its denominator, and the files staged for this book hold the total population of counties and tracts but not women by age. A fall in births to mothers aged 20 to 24 can mean fewer births per woman, fewer women in that age group, or both, and so can a gain. County population estimates by age and sex from the Census Bureau would supply the denominator for age-specific rates. Births are counted in the mother's county of residence, so the map is not a map of fertility.

County results describe the 578 counties that WONDER identifies, those with 100,000 or more residents in its population vintage. They hold 79.4% of births in 2024, and they are not a sample of the country. The pooled rows lost fewer births, although those rows include counties that now exceed the line. WONDER's county files used here have no cut by the mother's race or Hispanic origin, so the chapter cannot say whether the shift in mothers' ages differs by group.

The deprivation comparison is descriptive. The Area Deprivation Index describes neighborhoods in 2019 to 2023 and is averaged to counties, and region and deprivation overlap. The comparison of quartiles says nothing about why births fell in any county. It is an ecological pattern, and it need not hold for the people inside the counties. The comparison of 2018 with 2024 uses the endpoint years only. A county with few births moves from year to year by chance, so a county's direction of change is better established than its rank among the named counties.

## References

1. Osterman, Michelle J. K.; Hamilton, Brady E.; Martin, Joyce A.; Driscoll, Anne K.; Valenzuela, Claudia P. (2026). Births: Final Data for 2024. National Vital Statistics Reports 75(2). https://doi.org/10.15620/cdc/252440
2. Kearney, Melissa S.; Levine, Phillip B.; Pardue, Luke (2022). The Puzzle of Falling US Birth Rates since the Great Recession. Journal of Economic Perspectives 36(1): 151-176. https://doi.org/10.1257/jep.36.1.151
3. Kind, Amy J. H.; Buckingham, William R. (2018). Making Neighborhood-Disadvantage Metrics Accessible: The Neighborhood Atlas. New England Journal of Medicine 378(26): 2456-2458. https://doi.org/10.1056/NEJMp1802313
4. Mathews, T. J.; Hamilton, Brady E. (2016). Mean Age of Mothers is on the Rise: United States, 2000-2014. NCHS Data Brief No. 232. https://www.cdc.gov/nchs/products/databriefs/db232.htm

## Provenance

- Pipeline version: 0.1.0
- Build script: book2/chapters/where-births-are-falling/build.py (sha256 b47a334b398b4f7778e70d1b6a0a82dd6de30eed8a68b4e929b8ed5020155bf4)
- Built on: 2026-10-09
- Site commit: c2b1ff3309cb4a10833f884da91ac0057a66ad0a

Inputs:

| File | Size (bytes) | SHA-256, first 1 MB | SHA-256, last 1 MB |
| --- | ---: | --- | --- |
| book2/chapters/where-births-are-falling/wbf_data.py | 10,915 | c22de29942e1370129073448041e75b29b65f809f2d479a637d32d40628460f8 | c22de29942e1370129073448041e75b29b65f809f2d479a637d32d40628460f8 |
| data/analytic/tract_atlas.parquet | 39,659,925 | 29d5e4d943552d54aeb6ba92aadc939d9bd15bdea553c78357dbd516757be627 | 1e10d1ecf722340f23ee2212d4999a90250f85b65fa89b39e32db06d9472e22e |
| data/staged/births/wonder_natality_county_age_of_mother_2018_2024.parquet | 99,054 | b31e8b07c5eab8e54c94c4b57e7100327bf4dd6058d309cf128c3e4c640e44de | b31e8b07c5eab8e54c94c4b57e7100327bf4dd6058d309cf128c3e4c640e44de |
| data/staged/births/wonder_natality_county_total_2018_2024.parquet | 41,774 | e85d351fa72c022a05415d332185113b6badcaff1119da1d6f45c76ffe16d86d | e85d351fa72c022a05415d332185113b6badcaff1119da1d6f45c76ffe16d86d |
| data/staged/crosswalks/ct_tract2020_to_old_county.parquet | 15,842 | 961ef39f31596ef65fca2c67d5f2f63e533c5dc9b49745d3bb6437844e4b831d | 961ef39f31596ef65fca2c67d5f2f63e533c5dc9b49745d3bb6437844e4b831d |

Outputs:

| File | SHA-256 |
| --- | --- |
| facts.json | 77dd7eeea51784d62bbc66c924c5860689674d3033447c7880a586600453d9d5 |
| figures/age_change_by_county_change.json | 826d10c0a72167e0dbc41166ca2774d1d5c1d8c76df276e0ba5c2e42c6a52c4e |
| figures/births_by_deprivation.json | a21507843ebed1ddc8e6328285a33c0c9239ee6232858f469a4f30a9ae0e438e |
| figures/births_by_mother_age.json | 08b341775c96401ffd6ff6bf57bbe64d8290d998ed1ed855be2d5eea3e3d26a1 |
| figures/births_change_extremes.json | b625f3ef03d0e5fe62215d49d4651873e4e901ed120902c74d870f7819eeb14a |
| figures/births_change_map.json | 56281bbfe963990b15b8a6077266a4261a7dab56d91f830f4c6db01d66e89338 |
| figures/births_change_region_quartile.json | 056abf936ea3e1c4c3f5cfb66aeb3bf3434baee4c2b9d7cf2397c50dc584dd7b |
| figures/births_trend.json | 245d4e7aa4082510358d578cd02c61ee0e58cc2ed2b5d567605029bc48714694 |
| figures/mother_age_by_deprivation.json | a38297d682101f0ebc324d312b77f965d35e255e15d478ba0ea693591638de14 |
| meta.json | 78f25853ee414eae541cc5f9ab5f47a727040344922c9465602efd30b8f951ee |
| public/data/where-births-are-falling/births_change_map.json | 9751a5d7cca25f7381cc4799d7262211ddee3ea0e1182d31d8f3646e39d6238b |
| refs.json | 8ce76203656a29a39f6fa686e3886d08269b60425bc06bce02adb1cedd12496b |

## Machine formats

- Report (HTML): https://vishalsingh.org/zipcode/where-births-are-falling/report
- Report (Markdown): https://vishalsingh.org/zipcode/where-births-are-falling/report.md
- Every estimate (JSON): https://vishalsingh.org/zipcode/report/where-births-are-falling/facts.json
- Table 1, figure "births_trend" (JSON): https://vishalsingh.org/zipcode/report/where-births-are-falling/figures/births_trend.json
- Table 2, figure "births_change_map" (JSON): https://vishalsingh.org/zipcode/report/where-births-are-falling/figures/births_change_map.json
- Table 3, figure "births_change_extremes" (JSON): https://vishalsingh.org/zipcode/report/where-births-are-falling/figures/births_change_extremes.json
- Table 4, figure "births_by_deprivation" (JSON): https://vishalsingh.org/zipcode/report/where-births-are-falling/figures/births_by_deprivation.json
- Table 5, figure "births_change_region_quartile" (JSON): https://vishalsingh.org/zipcode/report/where-births-are-falling/figures/births_change_region_quartile.json
- Table 6, figure "births_by_mother_age" (JSON): https://vishalsingh.org/zipcode/report/where-births-are-falling/figures/births_by_mother_age.json
- Table 7, figure "mother_age_by_deprivation" (JSON): https://vishalsingh.org/zipcode/report/where-births-are-falling/figures/mother_age_by_deprivation.json
- Table 8, figure "age_change_by_county_change" (JSON): https://vishalsingh.org/zipcode/report/where-births-are-falling/figures/age_change_by_county_change.json
- Payload of "births_change_map" (JSON): https://vishalsingh.org/zipcode/data/where-births-are-falling/births_change_map.json
- Index of every report: https://vishalsingh.org/zipcode/llms.txt; all reports in one file: https://vishalsingh.org/zipcode/llms-full.txt

---

# Test scores, class and race: report

Chapter 3.1 of ZIP Code Destiny, by Vishal Singh (NYU Stern). Generated from the chapter's data files; the reading version is at https://vishalsingh.org/zipcode/test-scores-class-and-race. Table n of this report is figure n of the chapter.

## Identification

- Title: Test scores, class and race
- Label: The classroom staircase
- Chapter: 3.1 (Part III, School; An American life)
- Question: How far apart are average test scores across counties and districts, and how much of the gap between groups follows family income rather than race?
- Dek: Average achievement in third through eighth grade, 2009 to 2019, differs by 2.54 grade levels between the 5th and 95th percentile of counties and rises with county income in 45 of 45 states.
- Grain: counties
- Status: published
- Published: 2026-10-09
- Updated: not revised
- Data sources:
  - Stanford Education Data Archive. Stanford Education Data Archive 6.0, pooled math and reading achievement for counties and geographic school districts, grades 3 to 8, school years 2009 to 2019, for all students and by group, with gaps between groups; years: 2009-2019 (version 6.0); weights: tests counted; each analysis names its weight
  - County Health Rankings. County Health Rankings 2024: median household income (Census SAIPE, 2022), school segregation (Theil's index, NCES Common Core of Data, school year 2022 to 2023), Black-white residential segregation (ACS 2018 to 2022), non-Hispanic Black share of residents; years: 2024 release; weights: none (published county measures); analyses weight counties by SEDA tests counted
- Canonical URL: https://vishalsingh.org/zipcode/test-scores-class-and-race
- Report: https://vishalsingh.org/zipcode/test-scores-class-and-race/report
- Markdown: https://vishalsingh.org/zipcode/test-scores-class-and-race/report.md
- BibTeX key: singh2026testscoresclassandrace

```bibtex
@misc{singh2026testscoresclassandrace,
  author = {Singh, Vishal},
  title = {Test scores, class and race},
  year = {2026},
  url = {https://vishalsingh.org/zipcode/test-scores-class-and-race},
  note = {In ZIP Code Destiny. Published 9 October 2026.}
}
```

## Principal estimates

The facts the dek and the opening prose cite, in order of appearance.

| Estimand | Population | Period | Value | Interval | n | Weights | Source | Fact id |
| --- | --- | --- | ---: | --- | ---: | --- | --- | --- |
| 95th minus 5th percentile of the county average of math and reading, counties weighted by tests counted | counties with a SEDA average for both math and reading | test results for third through eighth grade, school years 2009 to 2019 pooled | 2.54 |   | 3,000 counties | math and reading tests counted for all students in the county | Stanford Education Data Archive | ach_spread |
| count of states where the county slope of average achievement on log income is positive | states with 10 or more counties that have an average and an income | test results for third through eighth grade, school years 2009 to 2019 pooled; median household income for 2022 | 45 |   |   | none (each state counts once) | Stanford Education Data Archive | slope_pos |
| count of states fitted | states with 10 or more counties that have an average and an income | test results for third through eighth grade, school years 2009 to 2019 pooled; median household income for 2022 | 45 |   |   | none (full count) | Stanford Education Data Archive | slope_states |
| count of counties | counties and county equivalents in the SEDA county file with a math and a reading estimate for all students (Virginia, DC and a few small counties have none) | test results for third through eighth grade, school years 2009 to 2019 pooled | 3,000 |   |   | none (full count) | Stanford Education Data Archive | n_counties_ach |
| 5th percentile of the county average of math and reading, counties weighted by tests counted | counties with a SEDA average for both math and reading | test results for third through eighth grade, school years 2009 to 2019 pooled | −1.27 |   | 3,000 counties | math and reading tests counted for all students in the county | Stanford Education Data Archive | ach_p5 |
| 95th percentile of the county average of math and reading, counties weighted by tests counted | counties with a SEDA average for both math and reading | test results for third through eighth grade, school years 2009 to 2019 pooled | +1.27 |   | 3,000 counties | math and reading tests counted for all students in the county | Stanford Education Data Archive | ach_p95 |
| count of counties in the income comparison | counties with a SEDA average for both math and reading and a median household income | test results for third through eighth grade, school years 2009 to 2019 pooled; median household income for 2022 | 2,999 |   |   | none (full count) | Stanford Education Data Archive | n_counties_income |
| weighted slope of the county average on the log (base 2) of median household income, bootstrap interval over states | counties with a SEDA average for both math and reading and a median household income | test results for third through eighth grade, school years 2009 to 2019 pooled; median household income for 2022 | 1.28 | 95% CI 0.94 to 1.72 | 2,999 counties | math and reading tests counted for all students in the county | Stanford Education Data Archive | income_slope |
| weighted mean of the district gap in math, bootstrap interval over states (districts in a state resemble one another) | geographic school districts with a math estimate for both groups (white minus Black) | test results for third through eighth grade, school years 2009 to 2019 pooled (math) | 0.64 | 95% CI 0.61 to 0.67 | 5,729 districts | math tests counted for the two groups in the gap | Stanford Education Data Archive | wbg_mean |
| weighted mean of the district gap in math, bootstrap interval over states (districts in a state resemble one another) | geographic school districts with a math estimate for both groups (not economically disadvantaged minus economically disadvantaged) | test results for third through eighth grade, school years 2009 to 2019 pooled (math) | 0.58 | 95% CI 0.55 to 0.61 | 11,815 districts | math tests counted for the two groups in the gap | Stanford Education Data Archive | neg_mean |

## Findings

Average achievement in third through eighth grade differs by 2.54 grade levels between the county at the 5th percentile and the county at the 95th, among 3,000 counties whose math and reading results cover the school years 2009 to 2019. Counties are weighted by the tests counted. The 5th-percentile county averages −1.27 grade levels from the national average and the 95th-percentile county +1.27.

Average achievement moves with county income. Across 2,999 counties, each doubling of median household income goes with 1.28 (95% CI 0.94 to 1.72) more grade levels.

Inside the same school districts, white students score 0.64 (95% CI 0.61 to 0.67) standard deviations above Black students in math, and students who are not economically disadvantaged score 0.58 (95% CI 0.55 to 0.61) standard deviations above those who are. Each gap is averaged over the districts that report it. Every figure describes counties or districts, not individual students.

### One scale for every state's tests

Each state gives its own tests, so raw scores cannot be compared across state lines. The Stanford Education Data Archive (SEDA) links each state's results to the National Assessment of Educational Progress, which puts them on a common scale. It publishes averages for counties and school districts, for all students and for groups of students. The averages here pool third through eighth grade and the school years 2009 to 2019. They are empirical-Bayes estimates, which pull the figure for a place with few tests toward the average of similar places. They rest on 404.9 million math and reading tests in the county files, where a test is one student's result in one subject, grade and year.

The archive reports its scale in two ways. A standard-deviation score says how many standard deviations a group's average lies from the national average for the same grade. A grade-level score restates the same quantity so that one unit is the national average gain between adjacent grades. One standard deviation equals 3.3 grade levels in the county averages and 3.2 in the math gaps, so on the grade-level scale the white–Black math gap inside districts is about 2.0 (95% CI 1.9 to 2.1) grade levels and the income gap about 1.8 (95% CI 1.7 to 1.9). County averages below are in grade levels from the national average, so a county at zero matches the nation. Group levels, gaps and their spreads are in standard deviations, which the archive recommends for research, and the two scales never share an axis.

### Where county averages are highest and lowest

Counties with the lowest averages are concentrated in the South. Of the lowest tenth of counties, 74% are there, against 43% of all counties (Table 1, with regions as the Census Bureau defines them). The highest tenth sits mostly in the Northeast and the Midwest, which hold 67% of that tenth and 42% of all counties.

The five lowest county averages are in Kusilvak and Bethel Census Areas in Alaska, Todd and Oglala Lakota Counties in South Dakota, and Sioux County in North Dakota, led by Kusilvak at −4.3 grade levels. Forsyth County, Georgia, has the highest average, +2.3. Counting each county once instead of weighting by tests puts the 5th and 95th percentiles at −1.57 and +1.01, a spread of 2.58. Both ends move down, since the lowest-scoring counties tend to have few tests and the highest-scoring ones many.

Table 1. 74% of the counties in the lowest tenth for average achievement are in the South, which holds 43% of counties

The map leaves 142 places gray. Virginia has no math estimates in the archive's sixth version, so none of its 133 counties and independent cities has an average of math and reading. The District of Columbia has no county rows, and 8 small counties are missing an estimate for math or reading. Each gray place carries its reason in the tooltip.

### Average achievement rises with county income

County averages are higher where median household income is higher (Table 2). Across counties, average achievement has a correlation of +0.59 (95% CI 0.43 to 0.76) with the log of median household income, and income accounts for 34% (95% CI 19% to 59%) of the variance in county averages, which leaves the rest to everything else that differs between counties. The lowest tenth of counties by income averages −1.12 (95% CI −1.44 to −0.75) grade levels from the national average and the highest tenth +0.41 (95% CI 0.02 to 0.87). Counties at any income level still scatter widely around the line, with a standard deviation of 0.63 (95% CI 0.49 to 0.72) grade levels, so income alone does not place a county. The income measure is the Census Bureau's modeled estimate for 2022 as County Health Rankings reports it. The test results are for 2009 to 2019, so it stands in for income during the test years.

Table 2. Average achievement is higher in counties with higher median household income, a correlation of +0.59 with log income across 2,999 counties

Counties in one state resemble each other, so the intervals here come from a bootstrap that resamples whole states, and they are wider than intervals that treat counties as independent. Counting each county once instead of weighting by tests gives a similar correlation and a slope within the interval.

### County income and achievement inside each state

A national slope could come from comparing rich states with poor ones, with no link inside any state. Table 3 fits the slope one state at a time, for the 45 states with 10 or more counties. The estimated slope is positive in 45 of 45. Its 95% interval, from robust standard errors with t intervals, lies above zero in 40. The intervals that include zero are those for Kentucky, Wyoming, Utah, New Hampshire and North Dakota.

In the median state a doubling of county income goes with 1.57 more grade levels. The slope is steepest in Alaska (3.27 (95% CI 1.66 to 4.87)) and shallowest in Kentucky (0.33 (95% CI −0.08 to 0.74)).

Table 3. The slope of county average achievement on county median income is positive in 45 of 45 states, with a 95% interval above zero in 40

The count of intervals above zero depends on how they are built. With heteroskedasticity-consistent (HC3) standard errors and normal intervals it is 36 of 45.

It also depends on the minimum number of counties. At a minimum of 8 counties, 46 states are fitted (Connecticut joins) and 41 have an interval above zero. At a minimum of 15 counties, 42 are fitted and 38 do.

New Hampshire has exactly 10 counties and is among the states whose interval includes zero. The estimated slope stays positive in every fitted state at every minimum up to 30 counties, but not at every income year. Wyoming's slope is close to zero, and with the Census Bureau's 2019 estimate of median household income in place of the 2022 one it is slightly negative.

A slope across counties compares places, not students, and a county average reflects everything that happens to children in and out of school, so a steep slope does not show that schools in poorer counties do worse. How much students learn from one grade to the next, as distinct from where they start, is the subject of Chapter 3.3, Learning growth and starting level (https://vishalsingh.org/zipcode/learning-growth; planned).

### Gaps between groups inside the same districts

SEDA reports an average for each group and the gap between two groups in the same geographic school district. Table 4 shows the distribution of those gaps in math. Each gap is averaged over the districts that report it, weighted by the tests of the two groups, so the rows cover different districts. White students score 0.64 (95% CI 0.61 to 0.67) standard deviations above Black students, about 2.0 (95% CI 1.9 to 2.1) grade levels on the archive's own grade-level gap. White students score above Black students in 99.8% of the 5,729 districts that report both.

Students who are not economically disadvantaged score 0.58 (95% CI 0.55 to 0.61) above those who are, about 1.8 (95% CI 1.7 to 1.9) grade levels. The white–Hispanic gap is 0.43 (95% CI 0.40 to 0.46) standard deviations. The white–Asian gap is −0.31 (95% CI −0.35 to −0.26), negative because Asian students score higher, as they do in 87% of the 4,353 districts that report both groups.

Table 4. Within the same districts, white students score 0.64 standard deviations above Black students in math, and the other gaps shown are smaller or negative

The income gap is reported by more districts than the white–Black gap (11,815 against 5,729). The 6,157 districts that report only the income gap have a smaller mean income gap, 0.40 (95% CI 0.38 to 0.41).

The two headline gaps are therefore compared on the 5,658 districts that report both. There the white–Black gap averages 0.64 (95% CI 0.61 to 0.67) and the income gap 0.60 (95% CI 0.57 to 0.63), a difference of 0.04 (95% CI 0.02 to 0.06) standard deviations.

The gaps vary widely between districts. Among districts weighted by tests, the white–Black gap is 0.28 standard deviations at the 5th percentile and 1.08 at the 95th, and the second view of the figure shows the same spread for the other gaps. Weighting districts by tests gives larger gaps than counting each district once, which gives 0.52 for the white–Black gap and 0.43 for the income gap, since districts with more tests tend to have larger gaps. States define economic disadvantage in their own ways, so the income gap compares groups that are defined differently from state to state.

### Where each group stands

A gap is a difference between two levels, and Table 5 shows the levels. Relative to the national average of all students, which is zero, math scores of Asian students average +0.66 standard deviations, white students +0.24, Hispanic students −0.30, Native American students −0.38 and Black students −0.53. SEDA takes its groups from state reports under federal race and ethnicity categories that are mutually exclusive, so Hispanic students of any race form their own group and the Asian, Black, white and Native American groups exclude them.

Students who are not economically disadvantaged average +0.39 and students who are −0.35. Each mean covers the counties that report the group, so the groups are measured in different places, and the figure carries the number of counties for each.

Table 5. In math, Asian students average +0.66 standard deviations from the national average and Black students −0.53

Counting each county once instead of weighting by the group's tests pulls every level toward zero. It gives +0.44 for Asian students, +0.08 for white students, −0.25 for Hispanic students, −0.23 for Native American students and −0.47 for Black students. For students who are not economically disadvantaged it gives +0.23, and for students who are, −0.28. Native American students then rank above Hispanic students, so the order of those two groups depends on the weighting. Black students have the lowest level either way.

### School segregation and the white–Black gap

Reardon, Kalogrides and Shores found, across districts and metropolitan areas, that racial gaps in test scores are larger where schools are more segregated and where racial groups differ more in family income and education.[1] County Health Rankings' 2024 release publishes a school segregation index for the 2022 to 2023 school year. It is Theil's index of how unevenly students of every racial and ethnic group are spread across a county's schools relative to the county as a whole, lowest when every school mirrors the county and highest under complete segregation. The index has a standard deviation of 0.07 across counties. Across 2,101 counties the white–Black math gap has a correlation of +0.59 (95% CI 0.51 to 0.66) with it (Table 6). Within states, a county one standard deviation higher on the index has a gap wider by 0.11 (95% CI 0.09 to 0.14) standard deviations, a regression with state fixed effects.

Table 6. The white–Black math gap is wider in counties with more school segregation, a correlation of +0.59 across 2,101 counties

Card and Rothstein studied SAT scores across metropolitan areas and found that residential segregation was consistently associated with a larger Black–white gap while school segregation had no independent association, though they could not reject that the two were equal.[2] Here, residential segregation alone moves with the gap, with a correlation of +0.25 (95% CI 0.12 to 0.37) and a within-state coefficient of 0.07 (95% CI 0.02 to 0.12). The two indices correlate at +0.46 (95% CI 0.31 to 0.56) across counties, so they are different measures.

A county regression with state fixed effects puts four measures in at once. They are school segregation, the Black–white residential index of dissimilarity for 2018 to 2022, the non-Hispanic Black share of residents and the log of median income. Across 1,849 counties, in standard deviations of the gap per standard deviation of the measure, it gives 0.11 (95% CI 0.08 to 0.13) for school segregation and −0.01 (95% CI −0.04 to 0.02) for residential segregation. In this regression school segregation keeps its association while the residential index loses its. The Black share of residents has a coefficient of 0.03 (95% CI 0.00 to 0.07) and the log of median income 0.09 (95% CI 0.06 to 0.11).

That pattern differs from Card and Rothstein's, but the studies differ in outcome (SAT takers against every tested student in third through eighth grade), in unit (metropolitan areas against counties), in the segregation measure (a Black–white index against one that covers every group) and in period. This is therefore not a test of their result. Neither study isolates segregation from the other differences between places.

### The income gap and the white–Black gap

Counties with a wide gap between students who are not and are economically disadvantaged tend to have a wide white–Black gap (Table 7). Across 2,092 counties the two correlate at +0.80 (95% CI 0.75 to 0.84), and a line through them has slope 0.99 (95% CI 0.91 to 1.08). The income gap accounts for 64% (95% CI 57% to 71%) of the variance in the race gap. That share is partly mechanical, because the two gaps are measured on overlapping groups of students and their estimation errors are correlated.

The mean income gap across these counties is 0.69 standard deviations and the mean white–Black gap 0.74. Around the line the white–Black gap still varies from county to county, with a residual spread of 0.14 (95% CI 0.13 to 0.15) standard deviations.

Table 7. Counties with a wider income gap have a wider white–Black gap, a correlation of +0.80 across 2,092 counties, and the two gaps are of similar size

SEDA does not cross race with economic disadvantage, so these data cannot say how large a white–Black gap would be among students in the same economic circumstances. A strong correlation across counties does not make either gap a version of the other. Chapter 4.3, Race, income and mobility (https://vishalsingh.org/zipcode/race-income-and-mobility) compares Black and white children from families with the same income who grew up in the same neighborhood, with adult income as the outcome.

### How much each gap varies across counties

A gap between two groups differs from one county to the next (Table 8). Across the 2,020 counties that report all four gaps, the white–Black math gap has a standard deviation of 0.24 (95% CI 0.22 to 0.26) standard deviations, the white–Hispanic gap 0.23 (95% CI 0.21 to 0.25) and the income gap 0.19 (95% CI 0.17 to 0.20).

The gap between boys and girls in reading, a yardstick whose own size is the subject of Chapter 3.2, Gender gaps in reading and math (https://vishalsingh.org/zipcode/gender-gaps-in-reading-and-math), has a spread of 0.06 (95% CI 0.05 to 0.07). The white–Black spread is 4.1 (95% CI 3.3 to 5.2) times that (3.0 (95% CI 2.5 to 3.8) counting each county once).

Table 8. Relative to its average size, the white–Black math gap varies 1.34 times as much across counties as the boys minus girls reading gap

A larger gap has more room to vary, so the comparison that matters is the spread relative to the average gap. The white–Black gap averages 3.0 (95% CI 2.8 to 3.3) times the reading gap. Relative to its average, its spread is 0.33 (95% CI 0.30 to 0.35) against 0.24 (95% CI 0.19 to 0.29) for the reading gap, a ratio of 1.34 (95% CI 1.08 to 1.76), or 1.26 (95% CI 1.04 to 1.63) counting each county once.

The white–Hispanic gap varies most relative to its average (0.43 (95% CI 0.39 to 0.48)), and the income gap (0.27 (95% CI 0.26 to 0.29)) about as much as the reading gap. In standard deviations the race and income gaps vary several times as much as the reading gap, but relative to their size the difference is modest, so these data do not show that race and class gaps depend on place much more than the gender gap does. Standard errors are smaller than the spreads, with a median of 0.03 for a county's white–Black gap and 0.01 for its boys-minus-girls reading gap.

### Two studies of Head Start and school spending

Everything above is description. Two studies use designs that can speak to whether early education and school funding change what children achieve. Each is reported under its own design, and neither estimates what it would take to close the gaps in this chapter.

> Causal evidence: Head Start funding and children's later outcomes (Regression discontinuity)
>
> In 1965 the federal Office of Economic Opportunity gave help with writing Head Start proposals to the poorest counties, those above a cutoff in the 1960 county poverty rate. Head Start funding was sharply higher above the cutoff, a difference that lasted, while other federal social spending did not jump there. Ludwig and Miller compared counties on either side of the cutoff, which should be alike in other respects.[3] Young school-age children in counties above the cutoff had lower mortality from causes that Head Start could affect, and the study found suggestive evidence of more schooling among cohorts old enough to have been exposed. Its outcomes are mortality and educational attainment, not test scores in third through eighth grade.

> Causal evidence: School finance reforms and adult outcomes (Event study with instrumental variables)
>
> Jackson, Johnson and Persico followed children born between 1955 and 1985 through 2011. They used the timing of court-ordered school finance reforms, and the type of funding formula each reform changed, as a source of variation in per-pupil spending, and compared cohorts exposed to a reform for more of their school years with cohorts exposed for fewer.[4] Higher spending over the school years raised educational attainment and adult wages and lowered adult poverty, with much larger effects for children from low-income families.

Both studies are about the policies and years they cover. Neither measures what a change in a county's income or segregation would do for its students.

### What these data cannot show

The figures describe counties and districts, and a county average blends every child in it (ecological inference). A pattern across counties need not hold for the students inside them. An average level is not a school's contribution, since it reflects families, neighborhoods, health and schools together. Reardon found that the income gap in achievement is already large when children enter kindergarten and does not widen much during the school years.[5] That fits a gap that is mostly in place at school entry, though county averages for third through eighth grade cannot show when each gap arises.

The achievement estimates are state tests placed on a national scale, pooled over a decade of cohorts. They are empirical-Bayes estimates, so small places are pulled toward the mean and the spreads here understate the spread of the true values, most for places with few tests. Virginia has no math estimates and DC no county rows. The groups come from state reports, states define economic disadvantage differently, and the data do not cross race with income.

The covariates come from other years than the tests. Median income is for 2022, the school segregation index for the 2022 to 2023 school year and the residential index for 2018 to 2022, while the tests are for 2009 to 2019, so the comparisons assume that a county's position changed little in the years between. The school segregation index covers every racial and ethnic group, not only Black and white students. The correlations and slopes are associations across places. The state fixed effects remove differences between states but not differences between counties, so none of them says what would happen to a county's students if its income or segregation changed. How schools discipline students, and what follows, is the subject of Chapter 3.4, School discipline and incarceration (https://vishalsingh.org/zipcode/school-discipline-and-incarceration).

## Tables

### Table 1. 74% of the counties in the lowest tenth for average achievement are in the South, which holds 43% of counties

Average of math and reading achievement in third through eighth grade, in grade levels above or below the national average, by county, 2009 to 2019 pooled

No inline rows: the values of Table 1 are in https://vishalsingh.org/zipcode/data/test-scores-class-and-race/scores_map.json.

Note: Virginia has no math estimates in the archive's sixth version, so its 133 counties and independent cities have no average of math and reading. The District of Columbia has no county rows, and 8 small counties have no average because the archive lists no estimate for them or no reading estimate. Puerto Rico is not drawn.

Note: Connecticut is drawn as its eight legacy counties, the codes SEDA uses. Valdez-Cordova Census Area in Alaska (split in 2019) is drawn on its two successors, Chugach and Copper River, with the one value SEDA reports.

Note: The archive reports empirical-Bayes estimates, which pull the figure for a place with few tests toward the average of similar places. Pooled results cover third through eighth grade and the school years 2009 to 2019.

Source: Stanford Education Data Archive 6.0, 2009-2019 (county averages for all students); modeled estimates, each county drawn once (unweighted)

Data: https://vishalsingh.org/zipcode/report/test-scores-class-and-race/figures/scores_map.json

### Table 2. Average achievement is higher in counties with higher median household income, a correlation of +0.59 with log income across 2,999 counties

Average of math and reading achievement (grade levels from the national average) against median household income (dollars, log scale), one dot per county; dot area is the tests counted

No inline rows: the values of Table 2 are in https://vishalsingh.org/zipcode/data/test-scores-class-and-race/scores_vs_income.json.

Note: Income is the Census Bureau's modeled estimate of county median household income for 2022 as County Health Rankings reports it. Test results are for 2009 to 2019, so a county's income stands in for its income during the test years. One county, Valdez-Cordova in Alaska, has no income in that file and is left out.

Note: Intervals resample whole states, because counties in a state resemble one another and share state tests and school funding rules. Counting each county once instead of weighting by tests gives a slope within the interval.

Note: Forsyth County, Georgia, and Kusilvak Census Area, Alaska, are labeled as the counties with the highest and lowest averages.

Source: Stanford Education Data Archive 6.0, 2009-2019, and County Health Rankings 2024 (Census SAIPE median household income for 2022); weighted by tests counted, modeled estimates

Data: https://vishalsingh.org/zipcode/report/test-scores-class-and-race/figures/scores_vs_income.json

### Table 3. The slope of county average achievement on county median income is positive in 45 of 45 states, with a 95% interval above zero in 40

Within-state slope of the county average of math and reading on county median household income (grade levels per doubling of income), states with 10 or more counties

| conf_high | conf_low | n | slope | state |
| ---: | ---: | ---: | ---: | --- |
| 0.739 | −0.083 | 120 | 0.328 | Kentucky |
| 1.01 | 0.015 | 92 | 0.514 | Nebraska |
| 1.53 | −0.472 | 23 | 0.527 | Wyoming |
| 1.63 | −0.230 | 29 | 0.699 | Utah |
| 1.83 | −0.048 | 10 | 0.891 | New Hampshire |
| 1.44 | 0.372 | 77 | 0.907 | Oklahoma |
| 1.68 | 0.215 | 33 | 0.948 | New Mexico |
| 2.70 | −0.705 | 53 | 0.996 | North Dakota |
| 1.47 | 0.535 | 55 | 1.00 | West Virginia |
| 1.50 | 0.800 | 15 | 1.15 | Arizona |
| 1.72 | 0.669 | 115 | 1.20 | Missouri |
| 1.86 | 0.541 | 44 | 1.20 | Idaho |
| 1.59 | 1.02 | 62 | 1.30 | New York |
| 2.63 | 0.082 | 17 | 1.36 | Nevada |
| 1.81 | 0.912 | 87 | 1.36 | Minnesota |
| 1.90 | 0.834 | 64 | 1.37 | Colorado |
| 1.78 | 1.02 | 252 | 1.40 | Texas |
| 1.86 | 0.955 | 16 | 1.41 | Maine |
| 2.29 | 0.546 | 95 | 1.42 | Tennessee |
| 1.89 | 0.987 | 64 | 1.44 | Louisiana |
| 1.90 | 1.21 | 99 | 1.55 | Iowa |
| 1.93 | 1.19 | 67 | 1.56 | Alabama |
| 1.93 | 1.22 | 100 | 1.57 | North Carolina |
| 2.11 | 1.16 | 67 | 1.63 | Florida |
| 2.42 | 0.856 | 14 | 1.64 | Vermont |
| 2.23 | 1.14 | 102 | 1.69 | Illinois |
| 2.19 | 1.22 | 36 | 1.71 | Oregon |
| 2.13 | 1.36 | 88 | 1.75 | Ohio |
| 2.01 | 1.49 | 58 | 1.75 | California |
| 2.18 | 1.33 | 159 | 1.76 | Georgia |
| 2.17 | 1.35 | 46 | 1.76 | South Carolina |
| 2.17 | 1.45 | 75 | 1.81 | Arkansas |
| 2.28 | 1.47 | 92 | 1.87 | Indiana |
| 3.28 | 0.578 | 56 | 1.93 | Montana |
| 2.51 | 1.36 | 39 | 1.93 | Washington |
| 2.69 | 1.32 | 105 | 2.00 | Kansas |
| 2.41 | 1.65 | 81 | 2.03 | Mississippi |
| 3.28 | 1.07 | 67 | 2.18 | Pennsylvania |
| 2.73 | 1.64 | 14 | 2.18 | Massachusetts |
| 2.92 | 1.52 | 24 | 2.22 | Maryland |
| 2.93 | 1.63 | 82 | 2.28 | Michigan |
| 3.02 | 1.96 | 21 | 2.49 | New Jersey |
| 3.71 | 1.99 | 72 | 2.85 | Wisconsin |
| 4.33 | 1.57 | 65 | 2.95 | South Dakota |
| 4.88 | 1.66 | 27 | 3.27 | Alaska |

Note: Each state is fitted on its own counties, weighted by the tests counted. Standard errors are robust to unequal variances, with t intervals. States with fewer counties than the minimum (Delaware, Hawaii, Rhode Island, Connecticut) are not fitted, and Virginia and DC have no average.

Note: With heteroskedasticity-consistent standard errors and normal intervals the interval clears zero in 36 states. Using the Census Bureau's 2019 income estimate in place of the 2022 one gives the same pattern, except that Wyoming's slope, close to zero in both, turns slightly negative.

Note: A slope across counties describes places. It does not say what a change in a county's income would do to its students.

Source: Stanford Education Data Archive 6.0, 2009-2019, and County Health Rankings 2024 (Census SAIPE median household income for 2022); weighted by tests counted, modeled estimates

Data: https://vishalsingh.org/zipcode/report/test-scores-class-and-race/figures/income_slopes_by_state.json

### Table 4. Within the same districts, white students score 0.64 standard deviations above Black students in math, and the other gaps shown are smaller or negative

Mean gap in math achievement between two groups of students in the same geographic school district, in standard deviations of the national cohort, averaged over the districts that report each gap, with the spread of district gaps as a bar

| conf_high | conf_low | gap | grade_levels_high | grade_levels_low | mean | mean_in_grade_levels | mean_unweighted | n | p25 | p5 | p50 | p75 | p95 |
| ---: | ---: | --- | ---: | ---: | ---: | ---: | ---: | ---: | ---: | ---: | ---: | ---: | ---: |
| 0.667 | 0.607 | White minus Black students | 2.12 | 1.93 | 0.636 | 2.02 | 0.525 | 5,729 | 0.475 | 0.279 | 0.617 | 0.777 | 1.07 |
| 0.605 | 0.546 | Not disadvantaged minus disadvantaged students | 1.92 | 1.73 | 0.575 | 1.83 | 0.431 | 11,815 | 0.434 | 0.252 | 0.555 | 0.717 | 0.939 |
| 0.464 | 0.401 | White minus Hispanic students | 1.47 | 1.27 | 0.432 | 1.37 | 0.328 | 8,561 | 0.272 | 0.105 | 0.403 | 0.569 | 0.833 |
| −0.258 | −0.351 | White minus Asian students | −0.820 | −1.11 | −0.312 | −0.990 | −0.288 | 4,353 | −0.489 | −0.749 | −0.333 | −0.154 | 0.199 |

Views: Interquartile range of districts (every row); 5th to 95th percentile of districts (every row).

Note: Each gap is the first group's average minus the second group's. A positive gap means the first group scores higher. Gaps use the standard-deviation scale the archive recommends for research.

Note: Counting each district once gives a smaller mean gap for the white–Black and income gaps (0.52 and 0.43) because districts with more tests have larger gaps. The archive shrinks districts with few tests toward the mean, so the spread of district gaps understates the spread of the true gaps.

Note: States define economic disadvantage in their own ways, so the income gap compares groups that are defined differently from state to state. Virginia (no math), DC and Hawaii (no districts) are absent.

Source: Stanford Education Data Archive 6.0, 2009-2019 (geographic school districts, math, standard-deviation scale); weighted by tests counted, modeled estimates

Data: https://vishalsingh.org/zipcode/report/test-scores-class-and-race/figures/gap_distribution.json

### Table 5. In math, Asian students average +0.66 standard deviations from the national average and Black students −0.53

Mean math achievement of each group of students relative to the national average of all students, in standard deviations of the national cohort, over counties that report the group

| conf_high | conf_high_unweighted | conf_low | conf_low_unweighted | group | kind | level | level_unweighted | n |
| ---: | ---: | ---: | ---: | --- | --- | ---: | ---: | ---: |
| 0.803 | 0.497 | 0.498 | 0.388 | Asian | Race and ethnicity | 0.662 | 0.445 | 1,563 |
| 0.282 | 0.121 | 0.191 | 0.032 | White | Race and ethnicity | 0.237 | 0.078 | 2,975 |
| −0.179 | −0.218 | −0.422 | −0.285 | Hispanic | Race and ethnicity | −0.301 | −0.249 | 2,640 |
| −0.296 | −0.180 | −0.472 | −0.289 | Native American | Race and ethnicity | −0.380 | −0.231 | 1,349 |
| −0.478 | −0.434 | −0.594 | −0.510 | Black | Race and ethnicity | −0.532 | −0.471 | 2,133 |
| 0.428 | 0.265 | 0.355 | 0.184 | Not economically disadvantaged | Economic status | 0.393 | 0.225 | 2,966 |
| −0.292 | −0.237 | −0.413 | −0.320 | Economically disadvantaged | Economic status | −0.352 | −0.277 | 2,994 |

Note: Each group's mean covers only the counties that report the group, and the number of counties differs by group (it is in each row's tooltip), so the groups are not measured in the same places. SEDA takes the groups from state reports. Race and ethnicity follow the federal categories, which are mutually exclusive, so Hispanic students of any race form their own group and the Asian, Black, white and Native American groups exclude them. States define economic disadvantage themselves.

Source: Stanford Education Data Archive 6.0, 2009-2019 (county averages by group, math, standard-deviation scale); weighted by each group's tests counted, modeled estimates

Data: https://vishalsingh.org/zipcode/report/test-scores-class-and-race/figures/group_levels.json

### Table 6. The white–Black math gap is wider in counties with more school segregation, a correlation of +0.59 across 2,101 counties

White minus Black math gap (standard deviations of the national cohort) against the school segregation index (Theil's index across racial and ethnic groups), one dot per county; dot area is the tests counted

No inline rows: the values of Table 6 are in https://vishalsingh.org/zipcode/data/test-scores-class-and-race/gap_vs_segregation.json.

Note: The index measures how unevenly students of every racial and ethnic group are spread across a county's schools relative to the county's composition. It is not a Black–white index, and it is for the 2022 to 2023 school year in County Health Rankings' 2024 release of NCES school data, after the test years.

Note: The gap is white minus Black, so a positive value means white students score higher.

Source: Stanford Education Data Archive 6.0, 2009-2019 (county gaps, math), and County Health Rankings 2024 (school segregation, school year 2022 to 2023, NCES Common Core of Data); weighted by tests counted, modeled estimates

Data: https://vishalsingh.org/zipcode/report/test-scores-class-and-race/figures/gap_vs_segregation.json

### Table 7. Counties with a wider income gap have a wider white–Black gap, a correlation of +0.80 across 2,092 counties, and the two gaps are of similar size

White minus Black math gap against the gap between students who are not and are economically disadvantaged, standard deviations of the national cohort, one dot per county; dot area is the tests counted

No inline rows: the values of Table 7 are in https://vishalsingh.org/zipcode/data/test-scores-class-and-race/class_gap_vs_race_gap.json.

Note: SEDA does not cross race with economic disadvantage, so the two gaps are measured on overlapping but different groups of students, and their estimation errors are correlated. The comparison describes counties and does not isolate a racial gap among students in the same economic circumstances.

Note: Counties are weighted by all-student tests counted. The standard deviation of the race gap around the line is 0.14 standard deviations.

Source: Stanford Education Data Archive 6.0, 2009-2019 (county gaps, math); weighted by tests counted, modeled estimates

Data: https://vishalsingh.org/zipcode/report/test-scores-class-and-race/figures/class_gap_vs_race_gap.json

### Table 8. Relative to its average size, the white–Black math gap varies 1.34 times as much across counties as the boys minus girls reading gap

Standard deviation across counties of each gap divided by the average gap, counties with all four gaps, weighted by the tests of the two groups in each gap

| cv | cv_high | cv_low | cv_unweighted | gap | median_se | n | sd | sd_high | sd_low | sd_unweighted |
| ---: | ---: | ---: | ---: | --- | ---: | ---: | ---: | ---: | ---: | ---: |
| 0.326 | 0.353 | 0.300 | 0.375 | White minus Black, math | 0.028 | 2,020 | 0.238 | 0.256 | 0.218 | 0.209 |
| 0.433 | 0.484 | 0.388 | 0.606 | White minus Hispanic, math | 0.025 | 2,020 | 0.229 | 0.249 | 0.207 | 0.209 |
| 0.274 | 0.291 | 0.255 | 0.288 | Not disadvantaged minus disadvantaged, math | 0.011 | 2,020 | 0.189 | 0.201 | 0.174 | 0.157 |
| 0.244 | 0.293 | 0.187 | 0.299 | Boys minus girls, reading | 0.010 | 2,020 | 0.059 | 0.069 | 0.047 | 0.069 |

Views: Relative to the average gap (every row); Standard deviation (every row).

Note: Relative spread is the weighted standard deviation across counties divided by the weighted average gap, so a larger gap has no more room to vary. Counting each county once gives a ratio of 3.0 for the standard deviations and 1.26 for the relative spreads. The median standard error of a county's white–Black gap is 0.03 standard deviations and of its boys minus girls reading gap 0.01.

Note: The archive shrinks small counties toward the mean, which lowers every spread, most for the gaps measured least precisely. The boys minus girls gap appears only as a yardstick.

Source: Stanford Education Data Archive 6.0, 2009-2019 (county gaps, math and reading, standard-deviation scale); weighted by tests counted, modeled estimates

Data: https://vishalsingh.org/zipcode/report/test-scores-class-and-race/figures/place_made_gaps.json

## All estimates

Every fact in facts.json (115), sorted by id.

| Fact id | Value | Unit | Estimand | Population | Period | Interval | n | Weights | Source | Note |
| --- | ---: | --- | --- | --- | --- | --- | ---: | --- | --- | --- |
| ach_bottom | −4.3 | grade levels from the national average (positive is above) | county average of math and reading (grade-level scale, minus the grade the pooled estimate is centered on) | the county with the lowest average of math and reading, Kusilvak Census Area, AK | test results for third through eighth grade, school years 2009 to 2019 pooled |   | 15,890 tests | none (one county) | Stanford Education Data Archive |   |
| ach_inc_d1 | −1.12 | grade levels from the national average (positive is above) | weighted mean of the county average of math and reading within the tenth (equal numbers of counties), bootstrap interval over states | counties in the lowest tenth by median household income, among counties with a SEDA average for both math and reading and a median household income | test results for third through eighth grade, school years 2009 to 2019 pooled; median household income for 2022 | 95% CI −1.44 to −0.75 | 299 counties | math and reading tests counted for all students in the county | Stanford Education Data Archive |   |
| ach_inc_d10 | +0.41 | grade levels from the national average (positive is above) | weighted mean of the county average of math and reading within the tenth (equal numbers of counties), bootstrap interval over states | counties in the highest tenth by median household income, among counties with a SEDA average for both math and reading and a median household income | test results for third through eighth grade, school years 2009 to 2019 pooled; median household income for 2022 | 95% CI 0.02 to 0.87 | 300 counties | math and reading tests counted for all students in the county | Stanford Education Data Archive |   |
| ach_p5 | −1.27 | grade levels from the national average (positive is above) | 5th percentile of the county average of math and reading, counties weighted by tests counted | counties with a SEDA average for both math and reading | test results for third through eighth grade, school years 2009 to 2019 pooled |   | 3,000 counties | math and reading tests counted for all students in the county | Stanford Education Data Archive | A spread of the county averages, not an interval. |
| ach_p5_unw | −1.57 | grade levels from the national average (positive is above) | 5th percentile of the county average of math and reading, each county counted once | counties with a SEDA average for both math and reading | test results for third through eighth grade, school years 2009 to 2019 pooled |   | 3,000 counties | none (each county counts once) | Stanford Education Data Archive |   |
| ach_p95 | +1.27 | grade levels from the national average (positive is above) | 95th percentile of the county average of math and reading, counties weighted by tests counted | counties with a SEDA average for both math and reading | test results for third through eighth grade, school years 2009 to 2019 pooled |   | 3,000 counties | math and reading tests counted for all students in the county | Stanford Education Data Archive | A spread of the county averages, not an interval. |
| ach_p95_unw | +1.01 | grade levels from the national average (positive is above) | 95th percentile of the county average of math and reading, each county counted once | counties with a SEDA average for both math and reading | test results for third through eighth grade, school years 2009 to 2019 pooled |   | 3,000 counties | none (each county counts once) | Stanford Education Data Archive |   |
| ach_spread | 2.54 | grade levels | 95th minus 5th percentile of the county average of math and reading, counties weighted by tests counted | counties with a SEDA average for both math and reading | test results for third through eighth grade, school years 2009 to 2019 pooled |   | 3,000 counties | math and reading tests counted for all students in the county | Stanford Education Data Archive | A spread of the county averages, not an interval. The first edition's span from the lowest to the highest county rested on two counties. |
| ach_spread_unw | 2.58 | grade levels | 95th minus 5th percentile of the county average of math and reading, each county counted once | counties with a SEDA average for both math and reading | test results for third through eighth grade, school years 2009 to 2019 pooled |   | 3,000 counties | none (each county counts once) | Stanford Education Data Archive |   |
| ach_top | +2.3 | grade levels from the national average (positive is above) | county average of math and reading (grade-level scale, minus the grade the pooled estimate is centered on) | the county with the highest average of math and reading, Forsyth County, GA | test results for third through eighth grade, school years 2009 to 2019 pooled |   | 433,229 tests | none (one county) | Stanford Education Data Archive |   |
| asn_level | +0.66 | standard deviations of the national cohort | mean math achievement of the group relative to the national average of all students, over counties that report the group, bootstrap interval over states | Asian students in counties that report the group | test results for third through eighth grade, school years 2009 to 2019 pooled (math) | 95% CI 0.50 to 0.80 | 1,563 counties | the group's math tests counted in each county | Stanford Education Data Archive | Race and ethnicity groups follow the federal categories that states report: they are mutually exclusive, and Hispanic students of any race form their own group. |
| asn_level_unw | +0.44 | standard deviations of the national cohort | mean math achievement of the group relative to the national average of all students, each county counted once, bootstrap interval over states | Asian students in counties that report the group | test results for third through eighth grade, school years 2009 to 2019 pooled (math) | 95% CI 0.39 to 0.50 | 1,563 counties | none (each county counts once) | Stanford Education Data Archive | Race and ethnicity groups follow the federal categories that states report: they are mutually exclusive, and Hispanic students of any race form their own group. |
| b_black | 0.03 | standard deviations of the national cohort of the white minus Black gap per standard deviation of the measure | coefficient on the Black share of residents (standardized) in one weighted regression with state fixed effects, errors clustered on state | counties with a white minus Black math gap and a school segregation index, a residential segregation index, the Black share of residents and median household income | test results for third through eighth grade, school years 2009 to 2019 pooled (math); school segregation for the 2022 to 2023 school year, residential segregation for 2018 to 2022 (County Health Rankings 2024) | 95% CI 0.00 to 0.07 | 1,849 counties | math tests counted for the two groups in the gap | Stanford Education Data Archive |   |
| b_inc | 0.09 | standard deviations of the national cohort of the white minus Black gap per standard deviation of the measure | coefficient on the log of median household income (standardized) in one weighted regression with state fixed effects, errors clustered on state | counties with a white minus Black math gap and a school segregation index, a residential segregation index, the Black share of residents and median household income | test results for third through eighth grade, school years 2009 to 2019 pooled (math); school segregation for the 2022 to 2023 school year, residential segregation for 2018 to 2022 (County Health Rankings 2024) | 95% CI 0.06 to 0.11 | 1,849 counties | math tests counted for the two groups in the gap | Stanford Education Data Archive |   |
| b_resid | −0.01 | standard deviations of the national cohort of the white minus Black gap per standard deviation of the measure | coefficient on the Black-white residential segregation index (standardized) in one weighted regression with state fixed effects, errors clustered on state | counties with a white minus Black math gap and a school segregation index, a residential segregation index, the Black share of residents and median household income | test results for third through eighth grade, school years 2009 to 2019 pooled (math); school segregation for the 2022 to 2023 school year, residential segregation for 2018 to 2022 (County Health Rankings 2024) | 95% CI −0.04 to 0.02 | 1,849 counties | math tests counted for the two groups in the gap | Stanford Education Data Archive |   |
| b_resseg | 0.07 | standard deviations of the national cohort of the white minus Black gap per standard deviation of the residential index | slope of the gap on the standardized residential index in a weighted regression with state fixed effects (no other measure), errors clustered on state | counties with a white minus Black math gap and a Black-white residential segregation index | test results for third through eighth grade, school years 2009 to 2019 pooled (math); school segregation for the 2022 to 2023 school year, residential segregation for 2018 to 2022 (County Health Rankings 2024) | 95% CI 0.02 to 0.12 | 1,859 counties | math tests counted for the two groups in the gap | Stanford Education Data Archive |   |
| b_school | 0.11 | standard deviations of the national cohort of the white minus Black gap per standard deviation of the measure | coefficient on the school segregation index (standardized) in one weighted regression with state fixed effects, errors clustered on state | counties with a white minus Black math gap and a school segregation index, a residential segregation index, the Black share of residents and median household income | test results for third through eighth grade, school years 2009 to 2019 pooled (math); school segregation for the 2022 to 2023 school year, residential segregation for 2018 to 2022 (County Health Rankings 2024) | 95% CI 0.08 to 0.13 | 1,849 counties | math tests counted for the two groups in the gap | Stanford Education Data Archive |   |
| b_seg | 0.11 | standard deviations of the national cohort of the white minus Black gap per standard deviation of the index | slope of the gap on the standardized school segregation index in a weighted regression with state fixed effects, errors clustered on state | counties with a white minus Black math gap and a school segregation index | test results for third through eighth grade, school years 2009 to 2019 pooled (math); school segregation for the 2022 to 2023 school year, residential segregation for 2018 to 2022 (County Health Rankings 2024) | 95% CI 0.09 to 0.14 | 2,101 counties | math tests counted for the two groups in the gap | Stanford Education Data Archive |   |
| blk_level | −0.53 | standard deviations of the national cohort | mean math achievement of the group relative to the national average of all students, over counties that report the group, bootstrap interval over states | Black students in counties that report the group | test results for third through eighth grade, school years 2009 to 2019 pooled (math) | 95% CI −0.59 to −0.48 | 2,133 counties | the group's math tests counted in each county | Stanford Education Data Archive | Race and ethnicity groups follow the federal categories that states report: they are mutually exclusive, and Hispanic students of any race form their own group. |
| blk_level_unw | −0.47 | standard deviations of the national cohort | mean math achievement of the group relative to the national average of all students, each county counted once, bootstrap interval over states | Black students in counties that report the group | test results for third through eighth grade, school years 2009 to 2019 pooled (math) | 95% CI −0.51 to −0.43 | 2,133 counties | none (each county counts once) | Stanford Education Data Archive | Race and ethnicity groups follow the federal categories that states report: they are mutually exclusive, and Hispanic students of any race form their own group. |
| both_diff | 0.04 | standard deviations of the national cohort, white minus Black gap minus income gap | difference between the weighted mean white minus Black gap and the weighted mean income gap over the same districts, bootstrap interval over states | geographic school districts with a math estimate for both the white minus Black gap and the income gap | test results for third through eighth grade, school years 2009 to 2019 pooled (math) | 95% CI 0.02 to 0.06 | 5,658 districts | math tests counted for the two groups in the gap | Stanford Education Data Archive |   |
| both_n | 5,658 | districts | count of districts that report both gaps | geographic school districts with a math estimate for both the white minus Black gap and the income gap | test results for third through eighth grade, school years 2009 to 2019 pooled (math) |   |   | none (full count) | Stanford Education Data Archive |   |
| class_var | 64% | percent of the variance of the county white minus Black gap | square of the weighted correlation (the share of weighted variance in the county race gap that the income gap accounts for) | counties with a math estimate for both the income gap and the white minus Black gap | test results for third through eighth grade, school years 2009 to 2019 pooled (math) | 95% CI 57% to 71% | 2,092 counties | math and reading tests counted for all students in the county | Stanford Education Data Archive |   |
| cty_neg_mean | 0.69 | standard deviations of the national cohort | weighted mean of the county income gap in math | counties with a math estimate for both the income gap and the white minus Black gap | test results for third through eighth grade, school years 2009 to 2019 pooled (math) |   | 2,092 counties | math and reading tests counted for all students in the county | Stanford Education Data Archive |   |
| cty_wbg_mean | 0.74 | standard deviations of the national cohort | weighted mean of the county white minus Black gap in math | counties with a math estimate for both the income gap and the white minus Black gap | test results for third through eighth grade, school years 2009 to 2019 pooled (math) |   | 2,092 counties | math and reading tests counted for all students in the county | Stanford Education Data Archive |   |
| cv_mfg_rla | 0.24 | ratio of the spread to the average gap | weighted standard deviation across counties of the boys minus girls reading gap divided by the absolute value of its weighted mean, bootstrap interval over states | counties with an estimate of all four gaps (white minus Black, white minus Hispanic and income gaps in math, boys minus girls gap in reading) | test results for third through eighth grade, school years 2009 to 2019 pooled (math, and reading for the boys minus girls gap) | 95% CI 0.19 to 0.29 | 2,020 counties | math tests counted for the two groups in the gap | Stanford Education Data Archive |   |
| cv_neg | 0.27 | ratio of the spread to the average gap | weighted standard deviation across counties of the income math gap divided by the absolute value of its weighted mean, bootstrap interval over states | counties with an estimate of all four gaps (white minus Black, white minus Hispanic and income gaps in math, boys minus girls gap in reading) | test results for third through eighth grade, school years 2009 to 2019 pooled (math, and reading for the boys minus girls gap) | 95% CI 0.26 to 0.29 | 2,020 counties | math tests counted for the two groups in the gap | Stanford Education Data Archive |   |
| cv_wbg | 0.33 | ratio of the spread to the average gap | weighted standard deviation across counties of the white minus Black math gap divided by the absolute value of its weighted mean, bootstrap interval over states | counties with an estimate of all four gaps (white minus Black, white minus Hispanic and income gaps in math, boys minus girls gap in reading) | test results for third through eighth grade, school years 2009 to 2019 pooled (math, and reading for the boys minus girls gap) | 95% CI 0.30 to 0.35 | 2,020 counties | math tests counted for the two groups in the gap | Stanford Education Data Archive |   |
| cv_whg | 0.43 | ratio of the spread to the average gap | weighted standard deviation across counties of the white minus Hispanic math gap divided by the absolute value of its weighted mean, bootstrap interval over states | counties with an estimate of all four gaps (white minus Black, white minus Hispanic and income gaps in math, boys minus girls gap in reading) | test results for third through eighth grade, school years 2009 to 2019 pooled (math, and reading for the boys minus girls gap) | 95% CI 0.39 to 0.48 | 2,020 counties | math tests counted for the two groups in the gap | Stanford Education Data Archive |   |
| ecd_level | −0.35 | standard deviations of the national cohort | mean math achievement of the group relative to the national average of all students, over counties that report the group, bootstrap interval over states | economically disadvantaged students in counties that report the group | test results for third through eighth grade, school years 2009 to 2019 pooled (math) | 95% CI −0.41 to −0.29 | 2,994 counties | the group's math tests counted in each county | Stanford Education Data Archive |   |
| ecd_level_unw | −0.28 | standard deviations of the national cohort | mean math achievement of the group relative to the national average of all students, each county counted once, bootstrap interval over states | economically disadvantaged students in counties that report the group | test results for third through eighth grade, school years 2009 to 2019 pooled (math) | 95% CI −0.32 to −0.24 | 2,994 counties | none (each county counts once) | Stanford Education Data Archive |   |
| gap_seg_d1 | 0.47 | standard deviations of the national cohort, white minus Black | weighted mean white minus Black math gap within the tenth (equal numbers of counties), bootstrap interval over states | counties in the least segregated tenth by school segregation index, among counties with a white minus Black math gap and a school segregation index | test results for third through eighth grade, school years 2009 to 2019 pooled (math); school segregation for the 2022 to 2023 school year, residential segregation for 2018 to 2022 (County Health Rankings 2024) | 95% CI 0.44 to 0.50 | 210 counties | math tests counted for the two groups in the gap | Stanford Education Data Archive |   |
| gap_seg_d10 | 0.90 | standard deviations of the national cohort, white minus Black | weighted mean white minus Black math gap within the tenth (equal numbers of counties), bootstrap interval over states | counties in the most segregated tenth by school segregation index, among counties with a white minus Black math gap and a school segregation index | test results for third through eighth grade, school years 2009 to 2019 pooled (math); school segregation for the 2022 to 2023 school year, residential segregation for 2018 to 2022 (County Health Rankings 2024) | 95% CI 0.86 to 0.95 | 211 counties | math tests counted for the two groups in the gap | Stanford Education Data Archive |   |
| gl_per_sd | 3.3 | grade levels per standard deviation | slope of the county average on the grade-level scale against the county average on the standard-deviation scale (math and reading averaged), weighted by tests counted | counties with a SEDA average for both math and reading | test results for third through eighth grade, school years 2009 to 2019 pooled |   | 3,000 counties | math and reading tests counted for all students in the county | Stanford Education Data Archive |   |
| gl_per_sd_gap | 3.2 | grade levels per standard deviation | slope of the archive's grade-level gap on its standard-deviation gap across the four math gaps, bootstrap over states | geographic school districts with a math estimate for the white minus Black, white minus Hispanic, white minus Asian or income gap | test results for third through eighth grade, school years 2009 to 2019 pooled (math) |   | 30,458 district gap estimates | math tests counted for the two groups in the gap | Stanford Education Data Archive |   |
| hsp_level | −0.30 | standard deviations of the national cohort | mean math achievement of the group relative to the national average of all students, over counties that report the group, bootstrap interval over states | Hispanic students in counties that report the group | test results for third through eighth grade, school years 2009 to 2019 pooled (math) | 95% CI −0.42 to −0.18 | 2,640 counties | the group's math tests counted in each county | Stanford Education Data Archive | Race and ethnicity groups follow the federal categories that states report: they are mutually exclusive, and Hispanic students of any race form their own group. |
| hsp_level_unw | −0.25 | standard deviations of the national cohort | mean math achievement of the group relative to the national average of all students, each county counted once, bootstrap interval over states | Hispanic students in counties that report the group | test results for third through eighth grade, school years 2009 to 2019 pooled (math) | 95% CI −0.29 to −0.22 | 2,640 counties | none (each county counts once) | Stanford Education Data Archive | Race and ethnicity groups follow the federal categories that states report: they are mutually exclusive, and Hispanic students of any race form their own group. |
| income_corr | +0.59 | correlation coefficient | weighted correlation of the county average of math and reading with the log of county median household income, bootstrap interval over states (counties in a state resemble one another) | counties with a SEDA average for both math and reading and a median household income | test results for third through eighth grade, school years 2009 to 2019 pooled; median household income for 2022 | 95% CI 0.43 to 0.76 | 2,999 counties | math and reading tests counted for all students in the county | Stanford Education Data Archive | Counting each county once gives a similar correlation. |
| income_r2 | 34% | percent of the variance of county average achievement | square of the weighted correlation with log income (the share of weighted variance in county averages that income accounts for) | counties with a SEDA average for both math and reading and a median household income | test results for third through eighth grade, school years 2009 to 2019 pooled; median household income for 2022 | 95% CI 19% to 59% | 2,999 counties | math and reading tests counted for all students in the county | Stanford Education Data Archive |   |
| income_resid_sd | 0.63 | grade levels | weighted standard deviation of the county average around the income line, bootstrap interval over states | counties with a SEDA average for both math and reading and a median household income | test results for third through eighth grade, school years 2009 to 2019 pooled; median household income for 2022 | 95% CI 0.49 to 0.72 | 2,999 counties | math and reading tests counted for all students in the county | Stanford Education Data Archive | The archive shrinks small counties toward the mean, which lowers this spread. |
| income_slope | 1.28 | grade levels of county average achievement per doubling of median household income | weighted slope of the county average on the log (base 2) of median household income, bootstrap interval over states | counties with a SEDA average for both math and reading and a median household income | test results for third through eighth grade, school years 2009 to 2019 pooled; median household income for 2022 | 95% CI 0.94 to 1.72 | 2,999 counties | math and reading tests counted for all students in the county | Stanford Education Data Archive | Counting each county once gives a slope within the interval. |
| min_alt_hi | 15 | counties | a higher minimum number of counties a state needs for its own slope (a sensitivity check) | states fitted one at a time | test results for third through eighth grade, school years 2009 to 2019 pooled; median household income for 2022 |   |   | none (inclusion rule) | Stanford Education Data Archive |   |
| min_alt_lo | 8 | counties | a lower minimum number of counties a state needs for its own slope (a sensitivity check) | states fitted one at a time | test results for third through eighth grade, school years 2009 to 2019 pooled; median household income for 2022 |   |   | none (inclusion rule) | Stanford Education Data Archive |   |
| min_alt_max | 30 | counties | highest minimum number of counties tried in the sensitivity check | states fitted one at a time | test results for third through eighth grade, school years 2009 to 2019 pooled; median household income for 2022 |   |   | none (inclusion rule) | Stanford Education Data Archive |   |
| min_counties_state | 10 | counties | minimum number of counties a state needs for its own slope | states fitted one at a time | test results for third through eighth grade, school years 2009 to 2019 pooled; median household income for 2022 |   |   | none (inclusion rule) | Stanford Education Data Archive |   |
| n_classrace | 2,092 | counties | count of counties in the comparison | counties with a math estimate for both the income gap and the white minus Black gap | test results for third through eighth grade, school years 2009 to 2019 pooled (math) |   |   | none (full count) | Stanford Education Data Archive |   |
| n_common | 2,020 | counties | count of counties with all four gaps | counties with an estimate of all four gaps (white minus Black, white minus Hispanic and income gaps in math, boys minus girls gap in reading) | test results for third through eighth grade, school years 2009 to 2019 pooled (math, and reading for the boys minus girls gap) |   |   | none (full count) | Stanford Education Data Archive |   |
| n_counties_ach | 3,000 | counties | count of counties | counties and county equivalents in the SEDA county file with a math and a reading estimate for all students (Virginia, DC and a few small counties have none) | test results for third through eighth grade, school years 2009 to 2019 pooled |   |   | none (full count) | Stanford Education Data Archive |   |
| n_counties_income | 2,999 | counties | count of counties in the income comparison | counties with a SEDA average for both math and reading and a median household income | test results for third through eighth grade, school years 2009 to 2019 pooled; median household income for 2022 |   |   | none (full count) | Stanford Education Data Archive |   |
| n_joint | 1,849 | counties | count of counties in the joint regression | counties with a white minus Black math gap and a school segregation index, a residential segregation index, the Black share of residents and median household income | test results for third through eighth grade, school years 2009 to 2019 pooled (math); school segregation for the 2022 to 2023 school year, residential segregation for 2018 to 2022 (County Health Rankings 2024) |   |   | none (full count) | Stanford Education Data Archive |   |
| n_map_nodata | 142 | counties | count of drawn counties with no average of math and reading | counties and county equivalents on the map (Puerto Rico not drawn) | test results for third through eighth grade, school years 2009 to 2019 pooled |   |   | none (full count) | Stanford Education Data Archive |   |
| n_map_other | 8 | counties | count of counties missing from the SEDA county file or missing a math or a reading estimate | counties on the map outside Virginia and DC with no average | test results for third through eighth grade, school years 2009 to 2019 pooled |   |   | none (full count) | Stanford Education Data Archive |   |
| n_off_scale | 73 | counties | count of counties whose average lies farther from the national average than the end of the key | counties with a SEDA average for both math and reading | test results for third through eighth grade, school years 2009 to 2019 pooled |   |   | none (full count) | Stanford Education Data Archive |   |
| n_seg | 2,101 | counties | count of counties in the segregation comparison | counties with a white minus Black math gap and a school segregation index | test results for third through eighth grade, school years 2009 to 2019 pooled (math); school segregation for the 2022 to 2023 school year, residential segregation for 2018 to 2022 (County Health Rankings 2024) |   |   | none (full count) | Stanford Education Data Archive |   |
| n_va | 133 | counties | count of Virginia places with no average of math and reading because SEDA has no Virginia math estimates | Virginia counties and independent cities on the map | test results for third through eighth grade, school years 2009 to 2019 pooled |   |   | none (full count) | Stanford Education Data Archive |   |
| nam_level | −0.38 | standard deviations of the national cohort | mean math achievement of the group relative to the national average of all students, over counties that report the group, bootstrap interval over states | Native American students in counties that report the group | test results for third through eighth grade, school years 2009 to 2019 pooled (math) | 95% CI −0.47 to −0.30 | 1,349 counties | the group's math tests counted in each county | Stanford Education Data Archive | Race and ethnicity groups follow the federal categories that states report: they are mutually exclusive, and Hispanic students of any race form their own group. |
| nam_level_unw | −0.23 | standard deviations of the national cohort | mean math achievement of the group relative to the national average of all students, each county counted once, bootstrap interval over states | Native American students in counties that report the group | test results for third through eighth grade, school years 2009 to 2019 pooled (math) | 95% CI −0.29 to −0.18 | 1,349 counties | none (each county counts once) | Stanford Education Data Archive | Race and ethnicity groups follow the federal categories that states report: they are mutually exclusive, and Hispanic students of any race form their own group. |
| nec_level | +0.39 | standard deviations of the national cohort | mean math achievement of the group relative to the national average of all students, over counties that report the group, bootstrap interval over states | students who are not economically disadvantaged in counties that report the group | test results for third through eighth grade, school years 2009 to 2019 pooled (math) | 95% CI 0.35 to 0.43 | 2,966 counties | the group's math tests counted in each county | Stanford Education Data Archive |   |
| nec_level_unw | +0.23 | standard deviations of the national cohort | mean math achievement of the group relative to the national average of all students, each county counted once, bootstrap interval over states | students who are not economically disadvantaged in counties that report the group | test results for third through eighth grade, school years 2009 to 2019 pooled (math) | 95% CI 0.18 to 0.26 | 2,966 counties | none (each county counts once) | Stanford Education Data Archive |   |
| neg_gl | 1.8 | grade levels, not disadvantaged minus disadvantaged | weighted mean district gap in math on the archive's grade-level scale (its own grade-level gap rows), bootstrap interval over states | geographic school districts with a math estimate for both groups | test results for third through eighth grade, school years 2009 to 2019 pooled (math) | 95% CI 1.7 to 1.9 | 11,815 districts | math tests counted for the two groups in the gap | Stanford Education Data Archive |   |
| neg_mean | 0.58 | standard deviations of the national cohort, not economically disadvantaged minus economically disadvantaged | weighted mean of the district gap in math, bootstrap interval over states (districts in a state resemble one another) | geographic school districts with a math estimate for both groups (not economically disadvantaged minus economically disadvantaged) | test results for third through eighth grade, school years 2009 to 2019 pooled (math) | 95% CI 0.55 to 0.61 | 11,815 districts | math tests counted for the two groups in the gap | Stanford Education Data Archive | Each gap is averaged over the districts that report it, so the gaps cover different sets of districts. |
| neg_mean_both | 0.60 | standard deviations of the national cohort, not economically disadvantaged minus economically disadvantaged | weighted mean of the district gap in math over the districts that report both gaps, bootstrap interval over states | geographic school districts with a math estimate for both the white minus Black gap and the income gap | test results for third through eighth grade, school years 2009 to 2019 pooled (math) | 95% CI 0.57 to 0.63 | 5,658 districts | math tests counted for the two groups in the gap | Stanford Education Data Archive |   |
| neg_mean_unw | 0.43 | standard deviations of the national cohort, not disadvantaged minus disadvantaged | unweighted mean of the district gap in math (each district counts once) | geographic school districts with a math estimate for both groups | test results for third through eighth grade, school years 2009 to 2019 pooled (math) |   | 11,815 districts | none (each district counts once) | Stanford Education Data Archive |   |
| neg_n | 11,815 | districts | count of districts with a gap estimate | geographic school districts with a math estimate for both groups (not economically disadvantaged minus economically disadvantaged) | test results for third through eighth grade, school years 2009 to 2019 pooled (math) |   |   | none (full count) | Stanford Education Data Archive |   |
| neg_only_mean | 0.40 | standard deviations of the national cohort, not economically disadvantaged minus economically disadvantaged | weighted mean of the district income gap in math, bootstrap interval over states | geographic school districts with an income gap but no white minus Black gap | test results for third through eighth grade, school years 2009 to 2019 pooled (math) | 95% CI 0.38 to 0.41 | 6,157 districts | math tests counted for the two groups in the gap | Stanford Education Data Archive |   |
| neg_only_n | 6,157 | districts | count of districts that report the income gap only | geographic school districts with an income gap but no white minus Black gap | test results for third through eighth grade, school years 2009 to 2019 pooled (math) |   |   | none (full count) | Stanford Education Data Archive |   |
| nm_all_share | 42% | percent of counties in the Northeast or Midwest | share of all counties with an average that are in the Northeast or the Midwest | counties with a SEDA average for both math and reading, grouped by U.S. Census Bureau region | test results for third through eighth grade, school years 2009 to 2019 pooled |   | 3,000 counties | none (each county counts once) | Stanford Education Data Archive |   |
| nm_high_share | 67% | percent of counties in the highest tenth by average achievement that are in the Northeast or Midwest | share of the highest tenth of counties in the Northeast or the Midwest | counties with a SEDA average for both math and reading, grouped by U.S. Census Bureau region | test results for third through eighth grade, school years 2009 to 2019 pooled |   | 300 counties | none (each county counts once) | Stanford Education Data Archive |   |
| r_classrace | +0.80 | correlation coefficient | weighted correlation of the county income gap with the county white minus Black gap in math, bootstrap interval over states | counties with a math estimate for both the income gap and the white minus Black gap | test results for third through eighth grade, school years 2009 to 2019 pooled (math) | 95% CI 0.75 to 0.84 | 2,092 counties | math and reading tests counted for all students in the county | Stanford Education Data Archive |   |
| r_resseg | +0.25 | correlation coefficient | weighted correlation of the county gap with the residential index of dissimilarity (0 to 100, ACS 2018 to 2022), bootstrap interval over states | counties with a white minus Black math gap and a Black-white residential segregation index | test results for third through eighth grade, school years 2009 to 2019 pooled (math); school segregation for the 2022 to 2023 school year, residential segregation for 2018 to 2022 (County Health Rankings 2024) | 95% CI 0.12 to 0.37 | 1,859 counties | math tests counted for the two groups in the gap | Stanford Education Data Archive |   |
| r_seg | +0.59 | correlation coefficient | weighted correlation of the county white minus Black math gap with the school segregation index (Theil's index across racial and ethnic groups, 0 to 1), bootstrap interval over states | counties with a white minus Black math gap and a school segregation index | test results for third through eighth grade, school years 2009 to 2019 pooled (math); school segregation for the 2022 to 2023 school year, residential segregation for 2018 to 2022 (County Health Rankings 2024) | 95% CI 0.51 to 0.66 | 2,101 counties | math tests counted for the two groups in the gap | Stanford Education Data Archive |   |
| r_sseg_rseg | +0.46 | correlation coefficient | weighted correlation of the school segregation index with the residential index, bootstrap interval over states | counties with both segregation indices and a white minus Black math gap | test results for third through eighth grade, school years 2009 to 2019 pooled (math); school segregation for the 2022 to 2023 school year, residential segregation for 2018 to 2022 (County Health Rankings 2024) | 95% CI 0.31 to 0.56 | 1,849 counties | math tests counted for the two groups in the gap | Stanford Education Data Archive |   |
| ratio_cv | 1.34 | ratio | relative spread of the white minus Black math gap divided by the relative spread of the boys minus girls reading gap, bootstrap interval over states | counties with an estimate of all four gaps (white minus Black, white minus Hispanic and income gaps in math, boys minus girls gap in reading) | test results for third through eighth grade, school years 2009 to 2019 pooled (math, and reading for the boys minus girls gap) | 95% CI 1.08 to 1.76 | 2,020 counties | math tests counted for the two groups in the gap | Stanford Education Data Archive |   |
| ratio_cv_unw | 1.26 | ratio | the same ratio with each county counted once, bootstrap interval over states | counties with an estimate of all four gaps (white minus Black, white minus Hispanic and income gaps in math, boys minus girls gap in reading) | test results for third through eighth grade, school years 2009 to 2019 pooled (math, and reading for the boys minus girls gap) | 95% CI 1.04 to 1.63 | 2,020 counties | none (each county counts once) | Stanford Education Data Archive |   |
| ratio_mean | 3.0 | ratio | average white minus Black math gap divided by the average boys minus girls reading gap (absolute values), bootstrap interval over states | counties with an estimate of all four gaps (white minus Black, white minus Hispanic and income gaps in math, boys minus girls gap in reading) | test results for third through eighth grade, school years 2009 to 2019 pooled (math, and reading for the boys minus girls gap) | 95% CI 2.8 to 3.3 | 2,020 counties | math tests counted for the two groups in the gap | Stanford Education Data Archive |   |
| ratio_sd | 4.1 | ratio | spread of the white minus Black math gap divided by the spread of the boys minus girls reading gap, bootstrap interval over states | counties with an estimate of all four gaps (white minus Black, white minus Hispanic and income gaps in math, boys minus girls gap in reading) | test results for third through eighth grade, school years 2009 to 2019 pooled (math, and reading for the boys minus girls gap) | 95% CI 3.3 to 5.2 | 2,020 counties | math tests counted for the two groups in the gap | Stanford Education Data Archive |   |
| ratio_sd_unw | 3.0 | ratio | the same ratio with each county counted once, bootstrap interval over states | counties with an estimate of all four gaps (white minus Black, white minus Hispanic and income gaps in math, boys minus girls gap in reading) | test results for third through eighth grade, school years 2009 to 2019 pooled (math, and reading for the boys minus girls gap) | 95% CI 2.5 to 3.8 | 2,020 counties | none (each county counts once) | Stanford Education Data Archive |   |
| resid_sd | 0.14 | standard deviations of the national cohort | weighted standard deviation of the county white minus Black gap around its line on the income gap, bootstrap interval over states | counties with a math estimate for both the income gap and the white minus Black gap | test results for third through eighth grade, school years 2009 to 2019 pooled (math) | 95% CI 0.13 to 0.15 | 2,092 counties | math and reading tests counted for all students in the county | Stanford Education Data Archive | The archive shrinks small counties toward the mean, which lowers this spread. |
| scale_cap | 2 | grade levels | distance from the national average at which the key ends (a display choice) | the key of the county map | test results for third through eighth grade, school years 2009 to 2019 pooled |   |   | none (display rule) | Stanford Education Data Archive |   |
| sd_mfg_rla | 0.06 | standard deviations of the national cohort | weighted standard deviation across counties of the boys minus girls reading gap estimate, bootstrap interval over states | counties with an estimate of all four gaps (white minus Black, white minus Hispanic and income gaps in math, boys minus girls gap in reading) | test results for third through eighth grade, school years 2009 to 2019 pooled (math, and reading for the boys minus girls gap) | 95% CI 0.05 to 0.07 | 2,020 counties | math tests counted for the two groups in the gap | Stanford Education Data Archive | The archive shrinks small counties toward the mean, which lowers this spread. |
| sd_neg | 0.19 | standard deviations of the national cohort | weighted standard deviation across counties of the income math gap estimate, bootstrap interval over states | counties with an estimate of all four gaps (white minus Black, white minus Hispanic and income gaps in math, boys minus girls gap in reading) | test results for third through eighth grade, school years 2009 to 2019 pooled (math, and reading for the boys minus girls gap) | 95% CI 0.17 to 0.20 | 2,020 counties | math tests counted for the two groups in the gap | Stanford Education Data Archive | The archive shrinks small counties toward the mean, which lowers this spread. |
| sd_sseg | 0.07 | Theil's index (0 to 1) | standard deviation of the school segregation index across counties (each county once) | counties with a white minus Black math gap and a school segregation index | test results for third through eighth grade, school years 2009 to 2019 pooled (math); school segregation for the 2022 to 2023 school year, residential segregation for 2018 to 2022 (County Health Rankings 2024) |   | 2,101 counties | none (each county counts once) | Stanford Education Data Archive |   |
| sd_wbg | 0.24 | standard deviations of the national cohort | weighted standard deviation across counties of the white minus Black math gap estimate, bootstrap interval over states | counties with an estimate of all four gaps (white minus Black, white minus Hispanic and income gaps in math, boys minus girls gap in reading) | test results for third through eighth grade, school years 2009 to 2019 pooled (math, and reading for the boys minus girls gap) | 95% CI 0.22 to 0.26 | 2,020 counties | math tests counted for the two groups in the gap | Stanford Education Data Archive | The archive shrinks small counties toward the mean, which lowers this spread. |
| sd_whg | 0.23 | standard deviations of the national cohort | weighted standard deviation across counties of the white minus Hispanic math gap estimate, bootstrap interval over states | counties with an estimate of all four gaps (white minus Black, white minus Hispanic and income gaps in math, boys minus girls gap in reading) | test results for third through eighth grade, school years 2009 to 2019 pooled (math, and reading for the boys minus girls gap) | 95% CI 0.21 to 0.25 | 2,020 counties | math tests counted for the two groups in the gap | Stanford Education Data Archive | The archive shrinks small counties toward the mean, which lowers this spread. |
| se_mfg_rla_med | 0.01 | standard deviations of the national cohort | median standard error of the county gap estimate (boys minus girls reading) | counties with an estimate of all four gaps (white minus Black, white minus Hispanic and income gaps in math, boys minus girls gap in reading) | test results for third through eighth grade, school years 2009 to 2019 pooled (math, and reading for the boys minus girls gap) |   | 2,020 counties | none (median over counties) | Stanford Education Data Archive |   |
| se_wbg_med | 0.03 | standard deviations of the national cohort | median standard error of the county gap estimate (white minus Black math) | counties with an estimate of all four gaps (white minus Black, white minus Hispanic and income gaps in math, boys minus girls gap in reading) | test results for third through eighth grade, school years 2009 to 2019 pooled (math, and reading for the boys minus girls gap) |   | 2,020 counties | none (median over counties) | Stanford Education Data Archive |   |
| sig_alt_hi | 38 | states | count of those states where the 95% interval for the slope lies above zero | states with 15 or more counties that have an average and an income | test results for third through eighth grade, school years 2009 to 2019 pooled; median household income for 2022 |   |   | none (each state counts once) | Stanford Education Data Archive |   |
| sig_alt_lo | 41 | states | count of those states where the 95% interval for the slope lies above zero | states with 8 or more counties that have an average and an income | test results for third through eighth grade, school years 2009 to 2019 pooled; median household income for 2022 |   |   | none (each state counts once) | Stanford Education Data Archive |   |
| slope_classrace | 0.99 | standard deviations of the white minus Black gap per standard deviation of the income gap | weighted slope of the county white minus Black gap on the county income gap, bootstrap interval over states | counties with a math estimate for both the income gap and the white minus Black gap | test results for third through eighth grade, school years 2009 to 2019 pooled (math) | 95% CI 0.91 to 1.08 | 2,092 counties | math and reading tests counted for all students in the county | Stanford Education Data Archive |   |
| slope_hi | 3.27 | grade levels of county average achievement per doubling of median household income | tests-weighted slope of the county average on log (base 2) income within the state, robust errors with t intervals (each county its own cluster) | counties in Alaska, the state with the steepest slope | test results for third through eighth grade, school years 2009 to 2019 pooled; median household income for 2022 | 95% CI 1.66 to 4.87 | 27 counties | math and reading tests counted for all students in the county | Stanford Education Data Archive |   |
| slope_lo | 0.33 | grade levels of county average achievement per doubling of median household income | tests-weighted slope of the county average on log (base 2) income within the state, robust errors with t intervals (each county its own cluster) | counties in Kentucky, the state with the shallowest slope | test results for third through eighth grade, school years 2009 to 2019 pooled; median household income for 2022 | 95% CI −0.08 to 0.74 | 120 counties | math and reading tests counted for all students in the county | Stanford Education Data Archive |   |
| slope_med | 1.57 | grade levels of county average achievement per doubling of median household income | median across states of the within-state slope (tests-weighted regression of the county average on log (base 2) income) | states with 10 or more counties that have an average and an income | test results for third through eighth grade, school years 2009 to 2019 pooled; median household income for 2022 |   | 45 states | none (each state counts once) | Stanford Education Data Archive |   |
| slope_pos | 45 | states | count of states where the county slope of average achievement on log income is positive | states with 10 or more counties that have an average and an income | test results for third through eighth grade, school years 2009 to 2019 pooled; median household income for 2022 |   |   | none (each state counts once) | Stanford Education Data Archive |   |
| slope_sig | 40 | states | count of states where the 95% interval for the slope lies above zero | states with 10 or more counties that have an average and an income | test results for third through eighth grade, school years 2009 to 2019 pooled; median household income for 2022 |   |   | none (each state counts once) | Stanford Education Data Archive |   |
| slope_sig_hc3 | 36 | states | count of states where the 95% interval from heteroskedasticity-consistent (HC3) errors with a normal approximation lies above zero | states with 10 or more counties that have an average and an income | test results for third through eighth grade, school years 2009 to 2019 pooled; median household income for 2022 |   |   | none (each state counts once) | Stanford Education Data Archive |   |
| slope_states | 45 | states | count of states fitted | states with 10 or more counties that have an average and an income | test results for third through eighth grade, school years 2009 to 2019 pooled; median household income for 2022 |   |   | none (full count) | Stanford Education Data Archive |   |
| south_all_share | 43% | percent of counties in the South | share of all counties with an average that are in the South | counties with a SEDA average for both math and reading, grouped by U.S. Census Bureau region | test results for third through eighth grade, school years 2009 to 2019 pooled |   | 3,000 counties | none (each county counts once) | Stanford Education Data Archive |   |
| south_low_share | 74% | percent of counties in the lowest tenth by average achievement that are in the South | share of the lowest tenth of counties (equal numbers of counties, ranked by the county average) in the South | counties with a SEDA average for both math and reading, grouped by U.S. Census Bureau region | test results for third through eighth grade, school years 2009 to 2019 pooled |   | 300 counties | none (each county counts once) | Stanford Education Data Archive |   |
| states_alt_hi | 42 | states | count of states fitted at that minimum | states with 15 or more counties that have an average and an income | test results for third through eighth grade, school years 2009 to 2019 pooled; median household income for 2022 |   |   | none (full count) | Stanford Education Data Archive |   |
| states_alt_lo | 46 | states | count of states fitted at that minimum | states with 8 or more counties that have an average and an income | test results for third through eighth grade, school years 2009 to 2019 pooled; median household income for 2022 |   |   | none (full count) | Stanford Education Data Archive |   |
| tests_m | 404.9 million | math and reading tests | sum of the math and the reading tests counted behind the county averages (one test is one student's result in one subject, grade and year) | counties with a SEDA average for both math and reading | test results for third through eighth grade, school years 2009 to 2019 pooled |   |   | none (full count) | Stanford Education Data Archive |   |
| wag_asian_share | 87% | percent of districts where Asian students score higher than white students | share of districts with a negative white minus Asian gap in math (each district counts once) | geographic school districts with a math estimate for both white and Asian students | test results for third through eighth grade, school years 2009 to 2019 pooled (math) |   | 4,353 districts | none (each district counts once) | Stanford Education Data Archive |   |
| wag_mean | −0.31 | standard deviations of the national cohort, white minus Asian | weighted mean of the district gap in math, bootstrap interval over states (districts in a state resemble one another) | geographic school districts with a math estimate for both groups (white minus Asian) | test results for third through eighth grade, school years 2009 to 2019 pooled (math) | 95% CI −0.35 to −0.26 | 4,353 districts | math tests counted for the two groups in the gap | Stanford Education Data Archive | Each gap is averaged over the districts that report it, so the gaps cover different sets of districts. |
| wag_n | 4,353 | districts | count of districts with a gap estimate | geographic school districts with a math estimate for both groups (white minus Asian) | test results for third through eighth grade, school years 2009 to 2019 pooled (math) |   |   | none (full count) | Stanford Education Data Archive |   |
| wbg_gl | 2.0 | grade levels, white minus black | weighted mean district gap in math on the archive's grade-level scale (its own grade-level gap rows), bootstrap interval over states | geographic school districts with a math estimate for both groups | test results for third through eighth grade, school years 2009 to 2019 pooled (math) | 95% CI 1.9 to 2.1 | 5,729 districts | math tests counted for the two groups in the gap | Stanford Education Data Archive |   |
| wbg_mean | 0.64 | standard deviations of the national cohort, white minus Black | weighted mean of the district gap in math, bootstrap interval over states (districts in a state resemble one another) | geographic school districts with a math estimate for both groups (white minus Black) | test results for third through eighth grade, school years 2009 to 2019 pooled (math) | 95% CI 0.61 to 0.67 | 5,729 districts | math tests counted for the two groups in the gap | Stanford Education Data Archive | Each gap is averaged over the districts that report it, so the gaps cover different sets of districts. |
| wbg_mean_both | 0.64 | standard deviations of the national cohort, white minus Black | weighted mean of the district gap in math over the districts that report both gaps, bootstrap interval over states | geographic school districts with a math estimate for both the white minus Black gap and the income gap | test results for third through eighth grade, school years 2009 to 2019 pooled (math) | 95% CI 0.61 to 0.67 | 5,658 districts | math tests counted for the two groups in the gap | Stanford Education Data Archive |   |
| wbg_mean_unw | 0.52 | standard deviations of the national cohort, white minus black | unweighted mean of the district gap in math (each district counts once) | geographic school districts with a math estimate for both groups | test results for third through eighth grade, school years 2009 to 2019 pooled (math) |   | 5,729 districts | none (each district counts once) | Stanford Education Data Archive |   |
| wbg_n | 5,729 | districts | count of districts with a gap estimate | geographic school districts with a math estimate for both groups (white minus Black) | test results for third through eighth grade, school years 2009 to 2019 pooled (math) |   |   | none (full count) | Stanford Education Data Archive |   |
| wbg_p5 | 0.28 | standard deviations of the national cohort, white minus Black | 5th percentile of the district gap in math, districts weighted by tests counted | geographic school districts with a math estimate for both white and Black students | test results for third through eighth grade, school years 2009 to 2019 pooled (math) |   | 5,729 districts | math tests counted for the two groups in the gap | Stanford Education Data Archive | A spread of district gaps, not an interval. |
| wbg_p95 | 1.08 | standard deviations of the national cohort, white minus Black | 95th percentile of the district gap in math, districts weighted by tests counted | geographic school districts with a math estimate for both white and Black students | test results for third through eighth grade, school years 2009 to 2019 pooled (math) |   | 5,729 districts | math tests counted for the two groups in the gap | Stanford Education Data Archive | A spread of district gaps, not an interval. |
| wbg_pos_share | 99.8% | percent of districts where white students score higher than Black students | share of districts with a positive white minus Black gap in math (each district counts once) | geographic school districts with a math estimate for both white and Black students | test results for third through eighth grade, school years 2009 to 2019 pooled (math) |   | 5,729 districts | none (each district counts once) | Stanford Education Data Archive |   |
| whg_mean | 0.43 | standard deviations of the national cohort, white minus Hispanic | weighted mean of the district gap in math, bootstrap interval over states (districts in a state resemble one another) | geographic school districts with a math estimate for both groups (white minus Hispanic) | test results for third through eighth grade, school years 2009 to 2019 pooled (math) | 95% CI 0.40 to 0.46 | 8,561 districts | math tests counted for the two groups in the gap | Stanford Education Data Archive | Each gap is averaged over the districts that report it, so the gaps cover different sets of districts. |
| wht_level | +0.24 | standard deviations of the national cohort | mean math achievement of the group relative to the national average of all students, over counties that report the group, bootstrap interval over states | white students in counties that report the group | test results for third through eighth grade, school years 2009 to 2019 pooled (math) | 95% CI 0.19 to 0.28 | 2,975 counties | the group's math tests counted in each county | Stanford Education Data Archive | Race and ethnicity groups follow the federal categories that states report: they are mutually exclusive, and Hispanic students of any race form their own group. |
| wht_level_unw | +0.08 | standard deviations of the national cohort | mean math achievement of the group relative to the national average of all students, each county counted once, bootstrap interval over states | white students in counties that report the group | test results for third through eighth grade, school years 2009 to 2019 pooled (math) | 95% CI 0.03 to 0.12 | 2,975 counties | none (each county counts once) | Stanford Education Data Archive | Race and ethnicity groups follow the federal categories that states report: they are mutually exclusive, and Hispanic students of any race form their own group. |

## Definitions

- SD units (SEDA): Test scores stated in units of the typical spread of scores around the national average, so that a score reads as the number of such spreads above or below the national average. SEDA also reports scores in grade levels, which are a separate scale.
- Bootstrap interval: A range for an estimate found by recomputing it on many random resamples of the data and reading the range from the results. It shows how much the estimate would vary from sample to sample.
- Fixed effects: A regression adjustment that gives each group, such as each county, its own baseline. The estimate then rests on differences within groups.
- Ecological inference: Drawing conclusions about individuals from data about places or groups. A pattern across counties need not hold for the people inside them.

## Methods

This chapter compares county average achievement in grades 3 to 8, pooled over the school years 2009 to 2019 from the Stanford Education Data Archive 6.0, with county income and school and residential segregation, and compares achievement gaps between groups of students. The county income and segregation measures come from County Health Rankings and refer to other years than the test scores. The archive's estimates are empirical-Bayes, which shrink small places toward the mean and lower every spread, and counties lacking either math or reading, including all of Virginia, have no average.

### Which students, grades and years the scores cover

Achievement is the Stanford Education Data Archive 6.0 pooled estimate for grades 3 to 8 over the school years 2009 to 2019, read for counties and for geographic school districts through the book's loader.

Race and ethnicity groups follow the federal categories that states report (mutually exclusive, with Hispanic students of any race in their own group); economic disadvantage is as each state defines it.

### How grade levels, standard deviations and gaps are defined

County averages in grade levels are the archive's grade-level estimate minus the grade it is centered on (5.5), averaged over math and reading, so zero is the national average; counties lacking either subject (all of Virginia, which has no math estimates, and two others) have no average, and DC has no rows.

Group levels, gaps and spreads use the standard-deviation scale, which the archive recommends for research; the two scales are never mixed on one axis. Gaps are the archive's own gap rows (first group minus second group). Estimates are empirical-Bayes, which shrink small places toward the mean.

Explains Table 5, Table 4 and Table 8.

### Where the income and segregation measures come from

Income is County Health Rankings' 2024 median household income (Census SAIPE for 2022); school segregation is its Theil's index of the spread of students of every racial and ethnic group across a county's schools (0 to 1, NCES Common Core of Data, school year 2022 to 2023 per the 2024 Technical Document); residential segregation is the Black-white index of dissimilarity (0 to 100, ACS 2018 to 2022).

Explains Table 2, Table 3, Table 6 and Table 7.

### How counties and groups are weighted

Weights: county averages by the math and reading tests counted for all students; a group's level by that group's math tests; a gap by the math tests of the two groups; the class-race comparison by all-student tests.

### Which places are drawn

Connecticut is drawn as its eight legacy counties; Puerto Rico is not drawn.

Explains Table 1.

### How uncertainty and spread are measured

Correlations, slopes and means across counties use a bootstrap that resamples whole states (2,000 draws, seed 12345), because counties in a state resemble one another; slopes within one state use robust errors with t intervals, each county its own cluster; regressions with state fixed effects cluster on state.

Spreads of county averages and of district gaps are percentiles or standard deviations, not intervals, and the archive's shrinkage lowers them.

## What these data cannot show

The figures describe counties and districts, and a county average blends every child in it (ecological inference). A pattern across counties need not hold for the students inside them. An average level is not a school's contribution, since it reflects families, neighborhoods, health and schools together. Reardon found that the income gap in achievement is already large when children enter kindergarten and does not widen much during the school years.[5] That fits a gap that is mostly in place at school entry, though county averages for third through eighth grade cannot show when each gap arises.

The achievement estimates are state tests placed on a national scale, pooled over a decade of cohorts. They are empirical-Bayes estimates, so small places are pulled toward the mean and the spreads here understate the spread of the true values, most for places with few tests. Virginia has no math estimates and DC no county rows. The groups come from state reports, states define economic disadvantage differently, and the data do not cross race with income.

The covariates come from other years than the tests. Median income is for 2022, the school segregation index for the 2022 to 2023 school year and the residential index for 2018 to 2022, while the tests are for 2009 to 2019, so the comparisons assume that a county's position changed little in the years between. The school segregation index covers every racial and ethnic group, not only Black and white students. The correlations and slopes are associations across places. The state fixed effects remove differences between states but not differences between counties, so none of them says what would happen to a county's students if its income or segregation changed. How schools discipline students, and what follows, is the subject of Chapter 3.4, School discipline and incarceration (https://vishalsingh.org/zipcode/school-discipline-and-incarceration).

## References

1. Reardon, Sean F.; Kalogrides, Demetra; Shores, Kenneth (2019). The Geography of Racial/Ethnic Test Score Gaps. American Journal of Sociology 124(4): 1164-1221. https://doi.org/10.1086/700678
2. Card, David; Rothstein, Jesse (2007). Racial Segregation and the Black-White Test Score Gap. Journal of Public Economics 91(11-12): 2158-2184. https://doi.org/10.1016/j.jpubeco.2007.03.006
3. Ludwig, Jens; Miller, Douglas L. (2007). Does Head Start Improve Children's Life Chances? Evidence from a Regression Discontinuity Design. The Quarterly Journal of Economics 122(1): 159-208. https://doi.org/10.1162/qjec.122.1.159
4. Jackson, C. Kirabo; Johnson, Rucker C.; Persico, Claudia (2016). The Effects of School Spending on Educational and Economic Outcomes: Evidence from School Finance Reforms. The Quarterly Journal of Economics 131(1): 157-218. https://doi.org/10.1093/qje/qjv036
5. Reardon, Sean F. (2011). The Widening Academic Achievement Gap Between the Rich and the Poor: New Evidence and Possible Explanations. In Greg J. Duncan and Richard J. Murnane (eds.), Whither Opportunity? Rising Inequality, Schools, and Children's Life Chances, pp. 91-116. New York: Russell Sage Foundation.

## Provenance

- Pipeline version: 0.1.0
- Build script: book2/chapters/test-scores-class-and-race/build.py (sha256 29de886272fb6330964489d56c5ef47b1057490454547685990192499c08c6ca)
- Built on: 2026-10-07
- Site commit: c2b1ff3309cb4a10833f884da91ac0057a66ad0a

Inputs:

| File | Size (bytes) | SHA-256, first 1 MB | SHA-256, last 1 MB |
| --- | ---: | --- | --- |
| book2/chapters/test-scores-class-and-race/seda.py | 33,494 | f4f0b05e4a3ad768df827cce4120f2ee17feb040d8062984836c50cced7d27e9 | f4f0b05e4a3ad768df827cce4120f2ee17feb040d8062984836c50cced7d27e9 |
| data/staged/education/seda_county_poolsub_6.0.parquet | 3,604,804 | 730134713ed3cae8983a255caa2a1be786a2d50688a949b11372f86e90065976 | 356fb555b2de0cc487cb4c37c95f29cee0497a96b4f7385fa6f10b9975bee58d |
| data/staged/education/seda_geodist_poolsub_6.0.parquet | 12,344,123 | 48de785e0dc62aacfc4608757670e6912b318c6c02baa3a716ef4c131c0eb6b5 | 5a22f706cfd0295923dd169d985d44d0da9076236e3afe38d59c5f6dd3e518a6 |
| data/staged/health/chr_county_2024.parquet | 8,776,933 | 619795f7ecfec8aff10694b649eaee5e421dbeffbe4be47af4161b51ceefa5cd | c69357f1bcf630407bb9716cb445618838999558c019aaaf77d4681b6bf45c56 |
| data/staged/income/saipe_county_2003_2023.parquet | 3,353,599 | 9c5b5e7b6921841caeeddfc1d582f638a1bd73320edf86de8a7c71b58ab44779 | 41e3bf3a4d8d9dea49709b4e790a66ae9a34d0937e8c3b219772a12c145bb760 |

Outputs:

| File | SHA-256 |
| --- | --- |
| facts.json | 161fb9daeca96d4198ccc6aa225f9d80757de196e99cc3d23690f3e41d8ec6c4 |
| figures/class_gap_vs_race_gap.json | d13790c2ba5a956bec45bf46c44672cfa20cd95a662f8d06bb4c70e5b7b726fc |
| figures/gap_distribution.json | c55ec368097035dd5dea00f0acb1b8a19c6b862533301bc386da7fb628f61fa2 |
| figures/gap_vs_segregation.json | 32a7a9f6dc1d305c44c9583ab647e0581ac8d9250ccf7a270e540e10dcde4317 |
| figures/group_levels.json | ef08d907d3eb458fd540fc03dabee103ffef635a1e3c9313138117b698bdd48c |
| figures/income_slopes_by_state.json | c2605808594a6879940ea3a9cc0e68d7699319be0b751fbda0a748cbc06c1fa4 |
| figures/place_made_gaps.json | 98ec8200b2c13bb745c111c3bdc1a907d11bbc5e692c115acad2ee988cc9dc66 |
| figures/scores_map.json | ae4443eafcf360040f178fad3cc90a0ac4f44db24626dcc948d4250840d66e6f |
| figures/scores_vs_income.json | e97f4c2aa78353616df54e8c216f5343ff611849f580dcaf0f4db8212cc96ef8 |
| meta.json | 301bd2655f4490faf63da8c885479c1b94c95de34eb01c5da4e6aa64f8b9ad4b |
| public/data/test-scores-class-and-race/class_gap_vs_race_gap.json | 1624d8ce14499a4039542fc0f90f03bf081e4748b7cc8d8908b2f87001ee1b9b |
| public/data/test-scores-class-and-race/gap_vs_segregation.json | 43a86c93850d3e0d5b00fd1074df96be590c0b3e654486228249a837e18e0e0d |
| public/data/test-scores-class-and-race/scores_map.json | f09ba9fd4341c9551b5274219b0d35c53765e42382216815d79f3eca4b9de31b |
| public/data/test-scores-class-and-race/scores_vs_income.json | 5a71d5c5e08322fbf43ce7d9060a681e5a616ff68eec3efb0bd86ac8960611a3 |
| refs.json | 8456854d14e1dfee87d5de8fe9a053a972b6ad0cbba84262ef3758f31539ebfa |

## Machine formats

- Report (HTML): https://vishalsingh.org/zipcode/test-scores-class-and-race/report
- Report (Markdown): https://vishalsingh.org/zipcode/test-scores-class-and-race/report.md
- Every estimate (JSON): https://vishalsingh.org/zipcode/report/test-scores-class-and-race/facts.json
- Table 1, figure "scores_map" (JSON): https://vishalsingh.org/zipcode/report/test-scores-class-and-race/figures/scores_map.json
- Table 2, figure "scores_vs_income" (JSON): https://vishalsingh.org/zipcode/report/test-scores-class-and-race/figures/scores_vs_income.json
- Table 3, figure "income_slopes_by_state" (JSON): https://vishalsingh.org/zipcode/report/test-scores-class-and-race/figures/income_slopes_by_state.json
- Table 4, figure "gap_distribution" (JSON): https://vishalsingh.org/zipcode/report/test-scores-class-and-race/figures/gap_distribution.json
- Table 5, figure "group_levels" (JSON): https://vishalsingh.org/zipcode/report/test-scores-class-and-race/figures/group_levels.json
- Table 6, figure "gap_vs_segregation" (JSON): https://vishalsingh.org/zipcode/report/test-scores-class-and-race/figures/gap_vs_segregation.json
- Table 7, figure "class_gap_vs_race_gap" (JSON): https://vishalsingh.org/zipcode/report/test-scores-class-and-race/figures/class_gap_vs_race_gap.json
- Table 8, figure "place_made_gaps" (JSON): https://vishalsingh.org/zipcode/report/test-scores-class-and-race/figures/place_made_gaps.json
- Payload of "scores_map" (JSON): https://vishalsingh.org/zipcode/data/test-scores-class-and-race/scores_map.json
- Payload of "scores_vs_income" (JSON): https://vishalsingh.org/zipcode/data/test-scores-class-and-race/scores_vs_income.json
- Payload of "gap_vs_segregation" (JSON): https://vishalsingh.org/zipcode/data/test-scores-class-and-race/gap_vs_segregation.json
- Payload of "class_gap_vs_race_gap" (JSON): https://vishalsingh.org/zipcode/data/test-scores-class-and-race/class_gap_vs_race_gap.json
- Index of every report: https://vishalsingh.org/zipcode/llms.txt; all reports in one file: https://vishalsingh.org/zipcode/llms-full.txt

---

# Gender gaps in reading and math: report

Chapter 3.2 of ZIP Code Destiny, by Vishal Singh (NYU Stern). Generated from the chapter's data files; the reading version is at https://vishalsingh.org/zipcode/gender-gaps-in-reading-and-math. Table n of this report is figure n of the chapter.

## Identification

- Title: Gender gaps in reading and math
- Label: Two gaps, one district
- Chapter: 3.2 (Part III, School; An American life)
- Question: Where do girls lead boys in reading and math, and does the math gap change sign with district income?
- Dek: Girls outscored boys in reading in 99.9% of school districts over 2009 to 2019, and the math gap, boys minus girls, rose from −0.20 grade levels in the lowest-income tenth of districts to +0.08 in the highest.
- Grain: institutions
- Status: published
- Published: 2026-10-09
- Updated: not revised
- Data sources:
  - Stanford Education Data Archive. Male-female test score gaps and all-student scores in reading and math for school districts and counties, third through eighth grade pooled over 2009 to 2019 (version 6.0); years: 2009-2019 (version 6.0); weights: none in the estimates (empirical-Bayes); districts weighted by their number of test scores or counted once, as each figure says
  - American Community Survey. Tract median household income and households, 2019 to 2023, apportioned to school districts with the NCES tract to school district crosswalk (2021 to 2022 school year); years: 2019-2023 (5-year); weights: households, with each tract split across districts by land area
- Canonical URL: https://vishalsingh.org/zipcode/gender-gaps-in-reading-and-math
- Report: https://vishalsingh.org/zipcode/gender-gaps-in-reading-and-math/report
- Markdown: https://vishalsingh.org/zipcode/gender-gaps-in-reading-and-math/report.md
- BibTeX key: singh2026gendergapsinreadingandmath

```bibtex
@misc{singh2026gendergapsinreadingandmath,
  author = {Singh, Vishal},
  title = {Gender gaps in reading and math},
  year = {2026},
  url = {https://vishalsingh.org/zipcode/gender-gaps-in-reading-and-math},
  note = {In ZIP Code Destiny. Published 9 October 2026.}
}
```

## Principal estimates

The facts the dek and the opening prose cite, in order of appearance.

| Estimand | Population | Period | Value | Interval | n | Weights | Source | Fact id |
| --- | --- | --- | ---: | --- | ---: | --- | --- | --- |
| share of districts (each counted once) where girls' mean reading score exceeds boys', bootstrap interval over states | school districts with a reading gap estimate | test scores of third through eighth graders, pooled over 2009 to 2019 | 99.9% | 95% CI 99.8% to 100.0% | 12,327 districts | none (each district counts once) | Stanford Education Data Archive | pct_girls_rla |
| test-weighted mean of the male minus female math gap within a tenth of districts (equal numbers of districts), bootstrap interval over states | school districts in the lowest tenth of districts by district median household income | test scores of third through eighth graders, pooled over 2009 to 2019; district income from the American Community Survey, 2019 to 2023 | −0.20 | 95% CI −0.26 to −0.15 | 1,213 districts | number of math and reading test scores in the district | Stanford Education Data Archive | mgap_inc_d1 |
| test-weighted mean of the male minus female math gap within a tenth of districts (equal numbers of districts), bootstrap interval over states | school districts in the highest tenth of districts by district median household income | test scores of third through eighth graders, pooled over 2009 to 2019; district income from the American Community Survey, 2019 to 2023 | +0.08 | 95% CI 0.04 to 0.12 | 1,214 districts | number of math and reading test scores in the district | Stanford Education Data Archive | mgap_inc_d10 |
| count of districts | school districts with a male-female reading gap estimate | test scores of third through eighth graders, pooled over 2009 to 2019 | 12,327 |   |   | none (full count) | Stanford Education Data Archive | n_rla |
| mean over districts of the male minus female reading gap, bootstrap interval over states | school districts with a reading gap estimate | test scores of third through eighth graders, pooled over 2009 to 2019 | −0.76 | 95% CI −0.82 to −0.70 | 12,327 districts | none (each district counts once) | Stanford Education Data Archive | rla_mean |
| count of districts | school districts with a male-female math gap estimate (Virginia has none, DC is absent and Hawaii has no districts) | test scores of third through eighth graders, pooled over 2009 to 2019 | 12,207 |   |   | none (full count) | Stanford Education Data Archive | n_mth |
| mean over districts of the male minus female math gap, bootstrap interval over states | school districts with a math gap estimate | test scores of third through eighth graders, pooled over 2009 to 2019 | −0.031 | 95% CI −0.060 to −0.004 | 12,207 districts | none (each district counts once) | Stanford Education Data Archive | mth_mean |
| share of districts (each counted once) where boys' mean math score exceeds girls', bootstrap interval over states | school districts with a math gap estimate | test scores of third through eighth graders, pooled over 2009 to 2019 | 43.8% | 95% CI 38.1% to 49.4% | 12,207 districts | none (each district counts once) | Stanford Education Data Archive | pct_boys_mth |

## Findings

Girls scored higher than boys in reading in 99.9% (95% CI 99.8% to 100.0%) of the 12,327 school districts with a reading estimate, and the average district's reading gap was −0.76 (95% CI −0.82 to −0.70) grade levels. Each gap in this chapter is the mean score of boys minus the mean score of girls in the same district, so a negative gap means girls scored higher. A grade level is the scale of the Stanford Education Data Archive (SEDA), on which one unit is the learning of an average school year, and the scores pool third through eighth grade over the 2009 to 2019 school years.

The math gap has no common sign. Across the 12,207 districts with a math estimate its mean is −0.031 (95% CI −0.060 to −0.004) grade levels, and boys lead in 43.8% (95% CI 38.1% to 49.4%) of districts. The gap moves with district income. It averages −0.20 (95% CI −0.26 to −0.15) grade levels in the tenth of districts with the lowest median household income and +0.08 (95% CI 0.04 to 0.12) in the tenth with the highest. District income is measured for 2019 to 2023, after the 2009 to 2019 test scores.

Girls and boys attend school in the same district, so a gap between them is measured inside one set of schools and neighborhoods. It does not come from setting a rich district beside a poor one, as a comparison of average scores across districts does. If a gap has one sign in almost every district, the sign does not depend on the district. If the sign changes with a district's income, the gap varies with place. This chapter applies both readings to reading and to math. The gaps between income groups and between racial groups are the subject of Chapter 3.1, Test scores, class and race (https://vishalsingh.org/zipcode/test-scores-class-and-race).

### How the gaps are measured

SEDA builds district estimates from state accountability tests in reading and math for third through eighth graders and places the tests from every state on one scale through the National Assessment of Educational Progress. For each district it reports the mean score of all students, of girls and of boys, and the difference between the last two. This chapter reads the gap estimates for 12,327 districts in reading and 12,207 in math, and the county estimates that SEDA builds the same way.

Weights depend on the question. A statement about districts, such as the share where girls lead, counts each district once. A statement about the average student weights each district by its number of test scores, where a test score is one student's result in one grade and one year. The text names the weight wherever it is not each district once. Intervals come from a bootstrap that resamples whole states, because the districts of one state share tests and policies and are not independent draws. The intervals for state means in the state figure are the exception, and the state section says so.

### The two gaps across school districts

Table 1 plots the district gaps. The reading distribution lies almost wholly on the girls' side of zero. The central 90% of district reading gaps, from the 5th to the 95th percentile, run from −1.18 to −0.36 grade levels. Girls scored higher in 12,313 districts and boys in 14.

Weighting each district by its number of reading test scores gives a mean of −0.79 (95% CI −0.85 to −0.73) grade levels. SEDA also reports gaps in standard deviations of the national cohort, and in those units the district means are −0.228 (95% CI −0.246 to −0.211) for reading and −0.010 (95% CI −0.019 to −0.002) for math.

A district's gap is an estimate with its own error, so a negative estimate does not always mean that girls scored higher. Reading passes this test. The 95% interval, built from the estimate and the empirical-Bayes standard error that SEDA reports, lies below zero in 98.9% (95% CI 98.3% to 99.4%) of the districts that report a standard error. Math does not. The interval lies below zero in 31.3% (95% CI 25.7% to 37.8%) of districts, above zero in 21.5% (95% CI 17.3% to 26.3%) and on both sides of zero in 47.2% (95% CI 42.2% to 51.8%). The median standard error of a district's math gap is 0.06 grade levels, which exceeds the size of the mean math gap.

The math distribution is centered close to zero, with 43.8% of districts on the boys' side. The central 90% of math gaps run from −0.35 to +0.29.

Weighting by test scores moves the mean toward girls, to −0.06 (95% CI −0.09 to −0.02) grade levels, and districts where boys lead hold 36.5% (95% CI 29.8% to 44.1%) of math test scores. The difference comes from the largest districts and not from a steady pattern with size. The 100 districts with the most math test scores average −0.09 (95% CI −0.16 to −0.05) grade levels, and boys lead in 28% (95% CI 14% to 39%) of them.

Chicago Public Schools (−0.34 (95% CI −0.35 to −0.33)) and the Philadelphia City School District (−0.33 (95% CI −0.35 to −0.32)) are among the ten largest and have gaps well toward girls. The rank correlation between a district's size and its math gap is −0.02 (95% CI −0.09 to 0.05).

Table 1. Girls scored higher than boys in reading in 99.9% of districts, while math gaps fall on both sides of zero

### The reading gap by state

The reading gap has one sign in every state and a size that differs by state (Table 2). The test-weighted mean reading gap is below zero in each of the 48 states that have both gaps and a district income. It runs from −1.28 (95% CI −1.32 to −1.25) grade levels in Arkansas to −0.46 (95% CI −0.50 to −0.42) in Iowa.

Differences between states account for 70% (95% CI 68% to 72%) of the test-weighted variance of district reading gaps and 36% (95% CI 34% to 40%) of the test-weighted variance of district math gaps. Counting each district once, the shares are 47% (95% CI 45% to 49%) for reading and 24% (95% CI 23% to 26%) for math. Reading is the larger share by either count.

The data cannot say why the size differs. States give different tests, SEDA puts them on one scale, and a difference between states in a gap can come from schools, from students or from what a test asks. What the reading comparison establishes is the sign, and that holds in every state. The intervals for the state means treat a state's districts as independent, which the bootstrap used elsewhere in this chapter does not, so they may be too narrow.

Table 2. The reading gap is below zero in all 48 states, while the math gap is above zero in 20 of them

### The math gap by district income

District income here is the mean of tract median household incomes in the district, weighted by households, from the American Community Survey for 2019 to 2023. That period is after the 2009 to 2019 test scores. Districts are the boundaries of the 2021 to 2022 school year, matched to SEDA by district code. The income sample holds the 12,136 districts with both gaps and an income, which cover 193.8 million math test scores. Tenths of districts, or deciles, hold equal numbers of districts. Incomes in the lowest tenth are at most $51,268, and incomes in the highest tenth are at least $115,294.

The math gap rises with income (Table 3). Weighted by test scores, it averages −0.20 (95% CI −0.26 to −0.15) grade levels in the lowest tenth and +0.08 (95% CI 0.04 to 0.12) in the highest, and the two intervals do not overlap. The ten decile means span 0.28 (95% CI 0.22 to 0.35) grade levels from the smallest to the largest.

Two other rankings of districts check the income ranking. Ranking districts by their average score in math and reading instead of by income gives −0.21 (95% CI −0.29 to −0.18) in the lowest tenth and +0.13 (95% CI 0.10 to 0.17) in the highest. Ranking by the poverty rate from the same survey gives +0.12 (95% CI 0.09 to 0.16) in the tenth of districts with the lowest poverty and −0.19 (95% CI −0.24 to −0.14) in the tenth with the highest. Income is the ranking used here because an average score is itself a test outcome.

Counting each district once gives −0.16 (95% CI −0.22 to −0.10) and +0.07 (95% CI 0.05 to 0.11), the same change of sign. Assigning each tract to a single district, as the book's tract file does, leaves 9,993 districts. The lowest and highest tenths then give −0.21 (95% CI −0.27 to −0.16) and +0.09 (95% CI 0.04 to 0.13), so the change of sign does not depend on how tracts are split between districts.

The mix of districts shifts with income as well. Boys lead in math in 23.7% (95% CI 15.3% to 33.1%) of the districts in the lowest tenth, and girls lead in 33.4% (95% CI 25.2% to 40.5%) of the districts in the highest. Districts of both signs appear at every income (Table 4), so income moves the balance between the two signs and does not sort districts into two kinds. Across the income sample, the test-weighted correlation between a district's math gap and its percentile rank in income is +0.40 (95% CI 0.31 to 0.48).

The reading gap follows income much less. Its decile means run from −0.79 (95% CI −0.89 to −0.71) in the lowest tenth to −0.87 (95% CI −0.97 to −0.78) in the highest, and each of the ten lies below zero. They span 0.13 (95% CI 0.06 to 0.24) grade levels, against 0.28 for the math gap. The correlation of the reading gap with income rank is −0.09 (95% CI −0.21 to 0.04), and its interval includes zero.

Table 3. The math gap rises from −0.20 grade levels in the lowest-income tenth of districts to +0.08 in the highest, while the reading gap stays below zero

Table 4. Districts at one income level differ widely in their math gap, and the decile means rise with income, a correlation of +0.40 with income rank

#### Within states

Income differs across states, and so do tests and policies, so the national pattern could come from differences between states. A regression of the math gap on a district's income rank with a separate intercept for each state gives a slope of +0.027 (95% CI 0.022 to 0.032) grade levels for each 10 points of rank, and the slope without the state intercepts is +0.025 (95% CI 0.019 to 0.031).

Of the 45 states with at least 25 districts in the sample, the slope is positive in 45 and its 95% interval lies above zero in 38. The pattern appears within states and does not come from comparing states.

### Where the math gap favors boys

Table 5 maps the math gap for 2,996 counties. Boys scored higher in 1,043 of them (34.8% (95% CI 25.9% to 44.7%)) and girls in 1,953. The reading gap is below zero in each of the 3,125 counties with a reading estimate, so a map of it would carry one color. The tooltip on each county gives both gaps.

Among the 2,938 counties that report a standard error, the 95% interval for the math gap lies above zero in 599 counties, below zero in 1,400 counties and on both sides of zero in 939.

The map shows differences between states. The test-weighted mean math gap is above zero in 20 of the 48 states. The 13 states whose intervals lie above zero are in the Plains, the Great Lakes states, the Mountain West and Oregon. The count depends on the interval, and a linearized standard error gives 14, adding Montana.

The largest state means favor boys in Iowa (+0.26 (95% CI 0.23 to 0.30)) and Utah, and the largest favor girls in Mississippi (−0.34 (95% CI −0.38 to −0.30)) and Alabama. Counties where boys scored higher cluster in the states with intervals above zero, and every county in Alabama, Arkansas, Louisiana, Mississippi and South Carolina is on the girls' side.

SEDA has no math estimates for Virginia and no rows for the District of Columbia. It has no Hawaii districts, but its county file covers Hawaii, and Hawaii's counties are drawn with data except Kalawao. In all, 146 map units are drawn as no data, and the tooltip on each gives its reason.

Table 5. Boys scored higher in math than girls in 34.8% of counties, and those counties cluster in the Plains, the Great Lakes states and the Mountain West

### What other studies find

Reardon and colleagues estimated the same gaps with SEDA data for the school years ending in 2009 through 2016.[1] The average district had no math gap and a reading gap favoring girls, and math gaps favored boys more in districts of higher socioeconomic status and in districts with larger gender disparities in adult income, education and occupation. For English language arts they did not find strong associations with those measures. With the 2009 to 2019 data and district income in place of their socioeconomic composite, this chapter finds a math gap near zero on average, a reading gap favoring girls and a math gap that rises with income. It finds no detectable association of the reading gap with income rank, and it adds that the reading gap differs by state.

Hyde and colleagues analyzed state mathematics assessments for grades two through eleven and found average differences between girls and boys close to zero.[2] Fryer and Levitt followed a national panel of children and found no mean difference in math at school entry, with girls losing ground relative to boys over the first years of school.[3] A gap pooled over third through eighth grade, as here, mixes grades at which the gap may differ.

#### Description, not a causal test

Nothing in this chapter tests why the math gap moves with district income. A district's income goes together with its schools, families, teachers and peers, and a comparison of girls and boys inside the district does not separate those from one another. The sign of the reading gap and the slope of the math gap are descriptions of districts. An explanation would need a design that changes one of these conditions for some children and not for others, and none is used here.

### What these data cannot show

The patterns describe school districts and counties. A district's income is not a family's income, and the data do not say that a girl or a boy from a poorer family scores differently from one in a richer family in the same district (ecological inference). A district gap is an average over every student the state tested in third through eighth grade, so it cannot show when in school a gap opens.

Each district's gap carries error, and the standard error of a district's math gap is typically larger than the mean math gap. The decile means, with intervals, carry the income finding, and single districts do not. SEDA's estimates are empirical-Bayes estimates, which pull the noisiest districts toward the average and narrow the spread of district gaps. If the shrinkage pulls toward a state or national mean, it could change the shares of variance that lie between states. The 95% intervals built from the empirical-Bayes standard errors describe the shrunken estimates. The scores come from state tests that differ in content and format, and a gap can reflect what a test asks as well as how girls and boys learned.

District income is built for this chapter and is not part of SEDA. It is a household-weighted mean of tract medians, split across overlapping districts by land area, so it is not the median income of a district's households, and the survey caps the highest incomes. It is measured for 2019 to 2023, after the 2009 to 2019 scores, on the district boundaries of the 2021 to 2022 school year. A district whose boundary or code changed may be matched to SEDA imperfectly or not at all. In all, 38 districts with both gaps have no income and are left out of the income figures. Districts that differ in income also differ in other ways, such as the racial composition of their students, and the chapter does not separate those.

SEDA has no math estimates for Virginia, no rows for the District of Columbia and no districts in Hawaii (Hawaii's counties are in its county file and are drawn), and Puerto Rico is not covered. The estimates describe public school students tested in 2009 to 2019 and say nothing about later years. SEDA reports two groups, male and female. Whether gaps in school scores carry into adult outcomes is the subject of Chapter 4.4, Boys and mobility (https://vishalsingh.org/zipcode/boys-and-mobility; planned), which compares boys and girls raised in the same neighborhoods.

## Tables

### Table 1. Girls scored higher than boys in reading in 99.9% of districts, while math gaps fall on both sides of zero

Distribution across school districts of the male minus female mean test score gap in grade levels, reading and math, third through eighth grade pooled over 2009 to 2019

| density | gap | subject |
| ---: | ---: | --- |
| 0 | −1.90 | Reading |
| 1 | −1.85 | Reading |
| 1 | −1.80 | Reading |
| 3 | −1.75 | Reading |
| 4 | −1.70 | Reading |
| 9 | −1.65 | Reading |
| 16 | −1.60 | Reading |
| 24 | −1.55 | Reading |
| 32 | −1.50 | Reading |
| 49 | −1.45 | Reading |
| 89 | −1.40 | Reading |
| 117 | −1.35 | Reading |
| 153 | −1.30 | Reading |
| 226 | −1.25 | Reading |
| 329 | −1.20 | Reading |
| 446 | −1.15 | Reading |
| 593 | −1.10 | Reading |
| 778 | −1.05 | Reading |
| 1035 | −1 | Reading |
| 1247 | −0.950 | Reading |
| 1375 | −0.900 | Reading |
| 1510 | −0.850 | Reading |
| 1588 | −0.800 | Reading |
| 1578 | −0.750 | Reading |
| 1525 | −0.700 | Reading |
| 1406 | −0.650 | Reading |
| 1290 | −0.600 | Reading |
| 1139 | −0.550 | Reading |
| 928 | −0.500 | Reading |
| 727 | −0.450 | Reading |
| 567 | −0.400 | Reading |
| 421 | −0.350 | Reading |
| 297 | −0.300 | Reading |
| 195 | −0.250 | Reading |
| 129 | −0.200 | Reading |
| 71 | −0.150 | Reading |
| 44 | −0.100 | Reading |
| 29 | −0.050 | Reading |
| 10 | 0 | Reading |
| 6 | 0.050 | Reading |
| 5 | 0.100 | Reading |
| 2 | 0.150 | Reading |
| 2 | 0.200 | Reading |
| 1 | 0.250 | Reading |
| 0 | 0.300 | Reading |
| 0 | 0.350 | Reading |
| 0 | 0.400 | Reading |
| 0 | 0.450 | Reading |
| 0 | 0.500 | Reading |
| 0 | 0.550 | Reading |
| 0 | 0.600 | Reading |
| 0 | 0.650 | Reading |
| 0 | 0.700 | Reading |
| 0 | 0.750 | Reading |
| 0 | 0.800 | Reading |
| 0 | 0.850 | Reading |
| 0 | 0.900 | Reading |
| 0 | 0.950 | Reading |
| 0 | 1 | Reading |
| 0 | 1.05 | Reading |
| 0 | 1.10 | Reading |
| 0 | 1.15 | Reading |
| 1 | 1.20 | Reading |
| 1 | 1.25 | Reading |
| 0 | 1.30 | Reading |
| 0 | −1.90 | Math |
| 0 | −1.85 | Math |
| 0 | −1.80 | Math |
| 0 | −1.75 | Math |
| 0 | −1.70 | Math |
| 0 | −1.65 | Math |
| 0 | −1.60 | Math |
| 0 | −1.55 | Math |
| 0 | −1.50 | Math |
| 0 | −1.45 | Math |
| 0 | −1.40 | Math |
| 0 | −1.35 | Math |
| 0 | −1.30 | Math |
| 0 | −1.25 | Math |
| 1 | −1.20 | Math |
| 0 | −1.15 | Math |
| 0 | −1.10 | Math |
| 0 | −1.05 | Math |
| 0 | −1 | Math |
| 0 | −0.950 | Math |
| 0 | −0.900 | Math |
| 1 | −0.850 | Math |
| 7 | −0.800 | Math |
| 13 | −0.750 | Math |
| 15 | −0.700 | Math |
| 24 | −0.650 | Math |
| 55 | −0.600 | Math |
| 94 | −0.550 | Math |
| 119 | −0.500 | Math |
| 185 | −0.450 | Math |
| 313 | −0.400 | Math |
| 507 | −0.350 | Math |
| 748 | −0.300 | Math |
| 1021 | −0.250 | Math |
| 1354 | −0.200 | Math |
| 1700 | −0.150 | Math |
| 1931 | −0.100 | Math |
| 2095 | −0.050 | Math |
| 2058 | 0 | Math |
| 1955 | 0.050 | Math |
| 1662 | 0.100 | Math |
| 1278 | 0.150 | Math |
| 1007 | 0.200 | Math |
| 708 | 0.250 | Math |
| 485 | 0.300 | Math |
| 274 | 0.350 | Math |
| 182 | 0.400 | Math |
| 100 | 0.450 | Math |
| 52 | 0.500 | Math |
| 21 | 0.550 | Math |
| 19 | 0.600 | Math |
| 6 | 0.650 | Math |
| 4 | 0.700 | Math |
| 1 | 0.750 | Math |
| 1 | 0.800 | Math |
| 0 | 0.850 | Math |
| 0 | 0.900 | Math |
| 0 | 0.950 | Math |
| 0 | 1 | Math |
| 1 | 1.05 | Math |
| 0 | 1.10 | Math |
| 0 | 1.15 | Math |
| 0 | 1.20 | Math |
| 0 | 1.25 | Math |
| 0 | 1.30 | Math |

Note: Each gap is the mean score of boys minus the mean score of girls in the district, in grade levels, where one grade level is the learning of an average school year. A negative gap means girls scored higher. Scores pool third through eighth grade over 2009 to 2019, and the estimates are empirical-Bayes estimates, which pull the noisiest districts toward the average.

Note: The curves are Gaussian kernel densities with the bandwidth that Silverman's rule gives for each subject. Reading includes Virginia, which has no math estimates, so the two curves rest on different sets of districts.

Source: Stanford Education Data Archive 6.0, 2009-2019; unweighted district estimates (each district counted once), modeled estimates

Data: https://vishalsingh.org/zipcode/report/gender-gaps-in-reading-and-math/figures/gap_distributions.json

### Table 2. The reading gap is below zero in all 48 states, while the math gap is above zero in 20 of them

Test-weighted mean male minus female gap in grade levels by state, reading and math, districts with both gaps and an income, third through eighth grade pooled over 2009 to 2019

| conf_high | conf_low | estimate | n | state | subject |
| ---: | ---: | ---: | ---: | --- | --- |
| 0.299 | 0.228 | 0.263 | 326 | Iowa | Math |
| −0.424 | −0.499 | −0.461 | 326 | Iowa | Reading |
| 0.219 | 0.095 | 0.157 | 41 | Utah | Math |
| −0.722 | −0.799 | −0.760 | 41 | Utah | Reading |
| 0.164 | 0.055 | 0.109 | 110 | Idaho | Math |
| −0.618 | −0.727 | −0.672 | 110 | Idaho | Reading |
| 0.144 | 0.068 | 0.106 | 323 | Minnesota | Math |
| −0.573 | −0.641 | −0.607 | 323 | Minnesota | Reading |
| 0.135 | 0.073 | 0.104 | 509 | Oklahoma | Math |
| −0.495 | −0.563 | −0.529 | 509 | Oklahoma | Reading |
| 0.132 | 0.070 | 0.101 | 513 | Michigan | Math |
| −0.692 | −0.742 | −0.717 | 513 | Michigan | Reading |
| 0.134 | 0.060 | 0.097 | 172 | Oregon | Math |
| −0.660 | −0.737 | −0.698 | 172 | Oregon | Reading |
| 0.151 | 0.020 | 0.085 | 150 | North Dakota | Math |
| −0.685 | −0.834 | −0.760 | 150 | North Dakota | Reading |
| 0.156 | 0.011 | 0.084 | 240 | Nebraska | Math |
| −0.438 | −0.572 | −0.505 | 240 | Nebraska | Reading |
| 0.108 | 0.054 | 0.081 | 288 | Indiana | Math |
| −0.901 | −0.947 | −0.924 | 288 | Indiana | Reading |
| 0.127 | 0.028 | 0.077 | 284 | Kansas | Math |
| −0.479 | −0.566 | −0.523 | 284 | Kansas | Reading |
| 0.130 | 0.006 | 0.068 | 48 | Wyoming | Math |
| −0.531 | −0.651 | −0.591 | 48 | Wyoming | Reading |
| 0.100 | 0.023 | 0.061 | 410 | Wisconsin | Math |
| −0.676 | −0.749 | −0.712 | 410 | Wisconsin | Reading |
| 0.089 | −0.017 | 0.036 | 210 | Montana | Math |
| −0.747 | −0.866 | −0.806 | 210 | Montana | Reading |
| 0.064 | −0.009 | 0.027 | 141 | New Hampshire | Math |
| −0.937 | −1.01 | −0.973 | 141 | New Hampshire | Reading |
| 0.055 | −0.008 | 0.024 | 174 | Maine | Math |
| −0.968 | −1.05 | −1.01 | 174 | Maine | Reading |
| 0.095 | −0.063 | 0.016 | 147 | South Dakota | Math |
| −0.621 | −0.774 | −0.698 | 147 | South Dakota | Reading |
| 0.057 | −0.028 | 0.014 | 158 | Connecticut | Math |
| −0.595 | −0.654 | −0.625 | 158 | Connecticut | Reading |
| 0.067 | −0.048 | 0.009 | 16 | Delaware | Math |
| −0.705 | −0.872 | −0.788 | 16 | Delaware | Reading |
| 0.069 | −0.059 | 0.005 | 166 | Colorado | Math |
| −0.899 | −0.992 | −0.945 | 166 | Colorado | Reading |
| 0.020 | −0.047 | −0.014 | 67 | Florida | Math |
| −0.558 | −0.607 | −0.583 | 67 | Florida | Reading |
| 0.004 | −0.041 | −0.019 | 607 | Ohio | Math |
| −0.696 | −0.734 | −0.715 | 607 | Ohio | Reading |
| −0.018 | −0.055 | −0.037 | 1,004 | Texas | Math |
| −0.556 | −0.590 | −0.573 | 1,004 | Texas | Reading |
| 0.006 | −0.089 | −0.041 | 499 | Pennsylvania | Math |
| −0.897 | −0.961 | −0.929 | 499 | Pennsylvania | Reading |
| 0.059 | −0.144 | −0.043 | 16 | Nevada | Math |
| −0.776 | −1.06 | −0.920 | 16 | Nevada | Reading |
| −0.015 | −0.077 | −0.046 | 492 | New Jersey | Math |
| −0.971 | −1.01 | −0.992 | 492 | New Jersey | Reading |
| −0.003 | −0.089 | −0.046 | 51 | Vermont | Math |
| −1.02 | −1.13 | −1.07 | 51 | Vermont | Reading |
| 0.056 | −0.150 | −0.047 | 652 | New York | Math |
| −0.699 | −0.899 | −0.799 | 652 | New York | Reading |
| −0.021 | −0.075 | −0.048 | 276 | Washington | Math |
| −0.916 | −0.971 | −0.943 | 276 | Washington | Reading |
| −0.016 | −0.084 | −0.050 | 512 | Missouri | Math |
| −0.776 | −0.836 | −0.806 | 512 | Missouri | Reading |
| −0.025 | −0.096 | −0.060 | 826 | California | Math |
| −0.810 | −0.858 | −0.834 | 826 | California | Reading |
| 0.014 | −0.193 | −0.090 | 47 | Alaska | Math |
| −0.739 | −0.964 | −0.851 | 47 | Alaska | Reading |
| −0.056 | −0.150 | −0.103 | 55 | West Virginia | Math |
| −1.16 | −1.26 | −1.21 | 55 | West Virginia | Reading |
| −0.046 | −0.179 | −0.113 | 35 | Rhode Island | Math |
| −1.08 | −1.24 | −1.16 | 35 | Rhode Island | Reading |
| −0.081 | −0.151 | −0.116 | 268 | Massachusetts | Math |
| −1.12 | −1.17 | −1.14 | 268 | Massachusetts | Reading |
| −0.074 | −0.167 | −0.120 | 182 | Arizona | Math |
| −0.802 | −0.867 | −0.834 | 182 | Arizona | Reading |
| −0.091 | −0.165 | −0.128 | 115 | North Carolina | Math |
| −0.480 | −0.542 | −0.511 | 115 | North Carolina | Reading |
| −0.062 | −0.197 | −0.129 | 754 | Illinois | Math |
| −0.901 | −0.994 | −0.948 | 754 | Illinois | Reading |
| −0.125 | −0.215 | −0.170 | 180 | Georgia | Math |
| −0.868 | −0.941 | −0.904 | 180 | Georgia | Reading |
| −0.131 | −0.231 | −0.181 | 140 | Tennessee | Math |
| −0.806 | −0.879 | −0.843 | 140 | Tennessee | Reading |
| −0.141 | −0.238 | −0.189 | 172 | Kentucky | Math |
| −0.737 | −0.853 | −0.795 | 172 | Kentucky | Reading |
| −0.127 | −0.265 | −0.196 | 83 | New Mexico | Math |
| −0.957 | −1.13 | −1.04 | 83 | New Mexico | Reading |
| −0.156 | −0.247 | −0.202 | 78 | South Carolina | Math |
| −0.805 | −0.878 | −0.841 | 78 | South Carolina | Reading |
| −0.173 | −0.258 | −0.215 | 69 | Louisiana | Math |
| −0.986 | −1.05 | −1.02 | 69 | Louisiana | Reading |
| −0.198 | −0.287 | −0.243 | 234 | Arkansas | Math |
| −1.25 | −1.32 | −1.28 | 234 | Arkansas | Reading |
| −0.153 | −0.350 | −0.252 | 24 | Maryland | Math |
| −1.11 | −1.21 | −1.16 | 24 | Maryland | Reading |
| −0.244 | −0.351 | −0.298 | 137 | Alabama | Math |
| −0.764 | −0.854 | −0.809 | 137 | Alabama | Reading |
| −0.301 | −0.376 | −0.339 | 135 | Mississippi | Math |
| −0.819 | −0.884 | −0.852 | 135 | Mississippi | Reading |

Note: Each gap is the mean score of boys minus the mean score of girls in the district, in grade levels, where one grade level is the learning of an average school year. A negative gap means girls scored higher. Scores pool third through eighth grade over 2009 to 2019, and the estimates are empirical-Bayes estimates, which pull the noisiest districts toward the average. Means weight a district by its number of math and reading test scores.

Note: The intervals treat the districts of a state as independent draws, so they are narrower than the intervals in the income figures, which resample whole states. Virginia (no math estimates), Hawaii (no districts) and the District of Columbia (no rows) are not shown.

Source: Stanford Education Data Archive 6.0, 2009-2019; test-weighted district estimates, modeled estimates

Data: https://vishalsingh.org/zipcode/report/gender-gaps-in-reading-and-math/figures/state_gap_means.json

### Table 3. The math gap rises from −0.20 grade levels in the lowest-income tenth of districts to +0.08 in the highest, while the reading gap stays below zero

Test-weighted mean male minus female gap in grade levels by tenth of district median household income, reading and math, third through eighth grade pooled over 2009 to 2019

| Subject | Income tenth | Median district income (dollars) | Mean gap (grade levels) | Lower limit of the 95% interval | Upper limit of the 95% interval | Districts |
| --- | ---: | ---: | ---: | ---: | ---: | ---: |
| Reading | 1 | 46,464 | −0.79 | −0.89 | −0.71 | 1,213 |
| Reading | 2 | 54,714 | −0.80 | −0.88 | −0.72 | 1,214 |
| Reading | 3 | 60,094 | −0.78 | −0.86 | −0.70 | 1,213 |
| Reading | 4 | 64,792 | −0.78 | −0.85 | −0.70 | 1,214 |
| Reading | 5 | 69,421 | −0.74 | −0.81 | −0.68 | 1,214 |
| Reading | 6 | 74,142 | −0.76 | −0.84 | −0.69 | 1,213 |
| Reading | 7 | 80,405 | −0.77 | −0.86 | −0.69 | 1,214 |
| Reading | 8 | 88,948 | −0.80 | −0.85 | −0.72 | 1,213 |
| Reading | 9 | 102,837 | −0.81 | −0.88 | −0.73 | 1,214 |
| Reading | 10 | 136,556 | −0.87 | −0.97 | −0.78 | 1,214 |
| Math | 1 | 46,464 | −0.20 | −0.26 | −0.15 | 1,213 |
| Math | 2 | 54,714 | −0.14 | −0.19 | −0.10 | 1,214 |
| Math | 3 | 60,094 | −0.14 | −0.20 | −0.09 | 1,213 |
| Math | 4 | 64,792 | −0.10 | −0.15 | −0.05 | 1,214 |
| Math | 5 | 69,421 | −0.05 | −0.09 | −0.01 | 1,214 |
| Math | 6 | 74,142 | −0.07 | −0.11 | −0.03 | 1,213 |
| Math | 7 | 80,405 | −0.06 | −0.11 | −0.02 | 1,214 |
| Math | 8 | 88,948 | −0.06 | −0.10 | 0.00 | 1,213 |
| Math | 9 | 102,837 | +0.01 | −0.03 | +0.06 | 1,214 |
| Math | 10 | 136,556 | +0.08 | +0.04 | +0.12 | 1,214 |

Note: Each gap is the mean score of boys minus the mean score of girls in the district, in grade levels, where one grade level is the learning of an average school year. A negative gap means girls scored higher. Scores pool third through eighth grade over 2009 to 2019, and the estimates are empirical-Bayes estimates, which pull the noisiest districts toward the average.

Note: District income is the mean of tract median household incomes in the district, weighted by households, from the American Community Survey for 2019 to 2023. The test scores pool 2009 to 2019, so income is measured after the scores, and districts are the boundaries of the 2021 to 2022 school year matched to SEDA by district code. It is not the median income of the district's households, and it is capped where the survey caps the highest incomes.

Note: Deciles, the tenths of districts ranked by income, hold equal numbers of districts. Each mean weights a district by its number of test scores, and the 95% intervals come from a bootstrap that resamples whole states.

Source: Stanford Education Data Archive 6.0, 2009-2019 and American Community Survey 2019-2023; test-weighted district estimates, modeled estimates

Data: https://vishalsingh.org/zipcode/report/gender-gaps-in-reading-and-math/figures/gap_by_income_decile.json

### Table 4. Districts at one income level differ widely in their math gap, and the decile means rise with income, a correlation of +0.40 with income rank

Male minus female math gap in grade levels against district median household income in dollars (log scale), one dot per district, with test-weighted decile means

No inline rows: the values of Table 4 are in https://vishalsingh.org/zipcode/data/gender-gaps-in-reading-and-math/math_gap_by_income.json.

Note: The dots are a random sample of 5,000 districts (seed 12345) from the 12,136 in the income sample; the means, intervals, correlation and slope use all of them.

Note: Each gap is the mean score of boys minus the mean score of girls in the district, in grade levels, where one grade level is the learning of an average school year. A negative gap means girls scored higher. Scores pool third through eighth grade over 2009 to 2019, and the estimates are empirical-Bayes estimates, which pull the noisiest districts toward the average.

Note: District income is the mean of tract median household incomes in the district, weighted by households, from the American Community Survey for 2019 to 2023. The test scores pool 2009 to 2019, so income is measured after the scores, and districts are the boundaries of the 2021 to 2022 school year matched to SEDA by district code. It is not the median income of the district's households, and it is capped where the survey caps the highest incomes.

Note: Deciles, the tenths of districts ranked by income, hold equal numbers of districts. Each mean weights a district by its number of test scores, and the 95% intervals come from a bootstrap that resamples whole states. Marker positions are the test-weighted mean income of each decile.

Note: The income axis is logarithmic: equal distances are equal ratios of income.

Source: Stanford Education Data Archive 6.0, 2009-2019 and American Community Survey 2019-2023; test-weighted decile means, modeled estimates

Data: https://vishalsingh.org/zipcode/report/gender-gaps-in-reading-and-math/figures/math_gap_by_income.json

### Table 5. Boys scored higher in math than girls in 34.8% of counties, and those counties cluster in the Plains, the Great Lakes states and the Mountain West

Male minus female mean math score gap in grade levels, county, third through eighth grade pooled over 2009 to 2019

No inline rows: the values of Table 5 are in https://vishalsingh.org/zipcode/data/gender-gaps-in-reading-and-math/math_gap_map.json.

Note: Each gray county has the reason for its missing value in its tooltip. SEDA has no math estimates for Virginia and no rows for the District of Columbia, and a few small counties have no male-female estimate. SEDA has no Hawaii districts, but its county file covers Hawaii, and Hawaii's counties are drawn with data except Kalawao. Puerto Rico is not in SEDA and is not drawn.

Note: The tooltip also gives the 95% interval for the math gap and for the reading gap of the county, built from SEDA's empirical-Bayes standard errors. Each gap is the mean score of boys minus the mean score of girls in the district, in grade levels, where one grade level is the learning of an average school year. A negative gap means girls scored higher. Scores pool third through eighth grade over 2009 to 2019, and the estimates are empirical-Bayes estimates, which pull the noisiest districts toward the average.

Note: Valdez-Cordova Census Area (split in 2019) is drawn as its two successors, Chugach and Copper River, with the same value. Connecticut is drawn as its eight legacy counties, which is how SEDA codes it. Alaska is drawn.

Source: Stanford Education Data Archive 6.0, 2009-2019; county estimates, modeled estimates, each county counted once

Data: https://vishalsingh.org/zipcode/report/gender-gaps-in-reading-and-math/figures/math_gap_map.json

## All estimates

Every fact in facts.json (83), sorted by id.

| Fact id | Value | Unit | Estimand | Population | Period | Interval | n | Weights | Source | Note |
| --- | ---: | --- | --- | --- | --- | --- | ---: | --- | --- | --- |
| big_mean | −0.09 | math gap in grade levels, male minus female (negative means girls scored higher) | mean of the male minus female math gap, each district counted once, bootstrap interval over states | the 100 districts with the most math test scores | test scores of third through eighth graders, pooled over 2009 to 2019 | 95% CI −0.16 to −0.05 | 100 districts | none (each district counts once) | Stanford Education Data Archive |   |
| big_n | 100 | districts | definition: the districts with the most math test scores | school districts with a math gap estimate | test scores of third through eighth graders, pooled over 2009 to 2019 |   |   | none (definition) | Stanford Education Data Archive |   |
| big_pct_boys | 28% | percent of districts where the math gap is above zero | share of districts where boys' mean math score exceeds girls', bootstrap interval over states | the 100 districts with the most math test scores | test scores of third through eighth graders, pooled over 2009 to 2019 | 95% CI 14% to 39% | 100 districts | none (each district counts once) | Stanford Education Data Archive |   |
| central_share | 90% | percent of districts between the 5th and 95th percentiles of the gap | definition: the central range of district gaps, from the 5th to the 95th percentile | school districts with a gap estimate | test scores of third through eighth graders, pooled over 2009 to 2019 |   |   | none (definition) | Stanford Education Data Archive |   |
| chicago_gap | −0.34 | math gap in grade levels, male minus female (negative means girls scored higher) | SEDA's male minus female math gap for the district; the interval uses SEDA's empirical-Bayes standard error | students in the Chicago Public Schools District 299 | test scores of third through eighth graders, pooled over 2009 to 2019 | 95% CI −0.35 to −0.33 | 1,628,988 math test scores | none (one district) | Stanford Education Data Archive |   |
| cty_boys | 1,043 | counties | count of counties where boys' mean math score exceeds girls' | counties with a male-female math gap estimate | test scores of third through eighth graders, pooled over 2009 to 2019 |   |   | none (full count) | Stanford Education Data Archive |   |
| cty_boys_clear | 599 | counties | count of counties whose 95% interval (estimate and standard error) for the math gap lies above zero | counties with a math gap estimate and a standard error | test scores of third through eighth graders, pooled over 2009 to 2019 |   |   | none (full count) | Stanford Education Data Archive |   |
| cty_girls | 1,953 | counties | count of counties where girls' mean math score exceeds boys' | counties with a male-female math gap estimate | test scores of third through eighth graders, pooled over 2009 to 2019 |   |   | none (full count) | Stanford Education Data Archive |   |
| cty_girls_clear | 1,400 | counties | count of counties whose 95% interval for the math gap lies below zero | counties with a math gap estimate and a standard error | test scores of third through eighth graders, pooled over 2009 to 2019 |   |   | none (full count) | Stanford Education Data Archive |   |
| cty_unclear | 939 | counties | count of counties whose 95% interval for the math gap includes zero | counties with a math gap estimate and a standard error | test scores of third through eighth graders, pooled over 2009 to 2019 |   |   | none (full count) | Stanford Education Data Archive |   |
| inc_d10_min | 115,294 | dollars of district median household income | lowest income in the highest tenth (equal numbers of districts in each tenth) | the highest tenth of districts by income | household income in 2023 dollars, ACS 2019 to 2023 |   | 1,214 districts | none (a boundary) | American Community Survey | A household-weighted mean of tract medians, top-coded at 250,001 dollars. |
| inc_d1_max | 51,268 | dollars of district median household income | highest income in the lowest tenth (equal numbers of districts in each tenth) | the lowest tenth of districts by income | household income in 2023 dollars, ACS 2019 to 2023 |   | 1,213 districts | none (a boundary) | American Community Survey | A household-weighted mean of tract medians, top-coded at 250,001 dollars. |
| mgap_ach_d1 | −0.21 | math gap in grade levels, male minus female (negative means girls scored higher) | test-weighted mean of the male minus female math gap within a tenth of districts ranked by average score (equal numbers of districts), bootstrap interval over states | school districts in the lowest tenth of districts by average math and reading score of all students | test scores of third through eighth graders, pooled over 2009 to 2019 | 95% CI −0.29 to −0.18 | 1,213 districts | number of math and reading test scores in the district | Stanford Education Data Archive |   |
| mgap_ach_d10 | +0.13 | math gap in grade levels, male minus female (negative means girls scored higher) | test-weighted mean of the male minus female math gap within a tenth of districts ranked by average score (equal numbers of districts), bootstrap interval over states | school districts in the highest tenth of districts by average math and reading score of all students | test scores of third through eighth graders, pooled over 2009 to 2019 | 95% CI 0.10 to 0.17 | 1,214 districts | number of math and reading test scores in the district | Stanford Education Data Archive |   |
| mgap_inc1_d1 | −0.21 | math gap in grade levels, male minus female (negative means girls scored higher) | test-weighted mean of the male minus female math gap within a tenth of districts ranked by an income built from each tract's single assigned district (equal numbers of districts), bootstrap interval over states | school districts in the lowest tenth of districts by income when each tract belongs to one district only | test scores of third through eighth graders, pooled over 2009 to 2019; district income from the American Community Survey, 2019 to 2023 | 95% CI −0.27 to −0.16 | 999 districts | number of math and reading test scores in the district | Stanford Education Data Archive |   |
| mgap_inc1_d10 | +0.09 | math gap in grade levels, male minus female (negative means girls scored higher) | test-weighted mean of the male minus female math gap within a tenth of districts ranked by an income built from each tract's single assigned district (equal numbers of districts), bootstrap interval over states | school districts in the highest tenth of districts by income when each tract belongs to one district only | test scores of third through eighth graders, pooled over 2009 to 2019; district income from the American Community Survey, 2019 to 2023 | 95% CI 0.04 to 0.13 | 1,000 districts | number of math and reading test scores in the district | Stanford Education Data Archive |   |
| mgap_inc_d1 | −0.20 | math gap in grade levels, male minus female (negative means girls scored higher) | test-weighted mean of the male minus female math gap within a tenth of districts (equal numbers of districts), bootstrap interval over states | school districts in the lowest tenth of districts by district median household income | test scores of third through eighth graders, pooled over 2009 to 2019; district income from the American Community Survey, 2019 to 2023 | 95% CI −0.26 to −0.15 | 1,213 districts | number of math and reading test scores in the district | Stanford Education Data Archive |   |
| mgap_inc_d10 | +0.08 | math gap in grade levels, male minus female (negative means girls scored higher) | test-weighted mean of the male minus female math gap within a tenth of districts (equal numbers of districts), bootstrap interval over states | school districts in the highest tenth of districts by district median household income | test scores of third through eighth graders, pooled over 2009 to 2019; district income from the American Community Survey, 2019 to 2023 | 95% CI 0.04 to 0.12 | 1,214 districts | number of math and reading test scores in the district | Stanford Education Data Archive |   |
| mgap_inc_d10_unw | +0.07 | math gap in grade levels, male minus female (negative means girls scored higher) | mean of the male minus female math gap within a tenth of districts, each district counted once, bootstrap interval over states | school districts in the highest tenth of districts by district median household income | test scores of third through eighth graders, pooled over 2009 to 2019; district income from the American Community Survey, 2019 to 2023 | 95% CI 0.05 to 0.11 | 1,214 districts | none (each district counts once) | Stanford Education Data Archive |   |
| mgap_inc_d1_unw | −0.16 | math gap in grade levels, male minus female (negative means girls scored higher) | mean of the male minus female math gap within a tenth of districts, each district counted once, bootstrap interval over states | school districts in the lowest tenth of districts by district median household income | test scores of third through eighth graders, pooled over 2009 to 2019; district income from the American Community Survey, 2019 to 2023 | 95% CI −0.22 to −0.10 | 1,213 districts | none (each district counts once) | Stanford Education Data Archive |   |
| mgap_inc_range | 0.28 | grade levels (largest minus smallest of the ten decile means of the math gap) | largest minus smallest test-weighted mean math gap across the ten income tenths, bootstrap interval over states | school districts with both gaps and a district income | test scores of third through eighth graders, pooled over 2009 to 2019; district income from the American Community Survey, 2019 to 2023 | 95% CI 0.22 to 0.35 | 12,136 districts | number of math and reading test scores in the district | Stanford Education Data Archive |   |
| mgap_pov_d1 | +0.12 | math gap in grade levels, male minus female (negative means girls scored higher) | test-weighted mean of the male minus female math gap within a tenth of districts ranked by the share of residents below the poverty line (ACS 2019 to 2023; equal numbers of districts), bootstrap interval over states | school districts in the lowest tenth of districts by poverty rate | test scores of third through eighth graders, pooled over 2009 to 2019; district income from the American Community Survey, 2019 to 2023 | 95% CI 0.09 to 0.16 | 1,213 districts | number of math and reading test scores in the district | Stanford Education Data Archive |   |
| mgap_pov_d10 | −0.19 | math gap in grade levels, male minus female (negative means girls scored higher) | test-weighted mean of the male minus female math gap within a tenth of districts ranked by the share of residents below the poverty line (ACS 2019 to 2023; equal numbers of districts), bootstrap interval over states | school districts in the highest tenth of districts by poverty rate | test scores of third through eighth graders, pooled over 2009 to 2019; district income from the American Community Survey, 2019 to 2023 | 95% CI −0.24 to −0.14 | 1,214 districts | number of math and reading test scores in the district | Stanford Education Data Archive |   |
| min_state_districts | 25 | districts | inclusion rule: a state needs at least this many districts in the income sample | states given a slope of their own | test scores of third through eighth graders, pooled over 2009 to 2019; district income from the American Community Survey, 2019 to 2023 |   |   | none (definition) | Stanford Education Data Archive |   |
| mth_inc_r | +0.40 | correlation coefficient | test-weighted correlation of the district math gap with the district's percentile rank of income, bootstrap interval over states | school districts with both gaps and a district income | test scores of third through eighth graders, pooled over 2009 to 2019; district income from the American Community Survey, 2019 to 2023 | 95% CI 0.31 to 0.48 | 12,136 districts | number of math and reading test scores in the district | Stanford Education Data Archive |   |
| mth_mean | −0.031 | math gap in grade levels, male minus female (negative means girls scored higher) | mean over districts of the male minus female math gap, bootstrap interval over states | school districts with a math gap estimate | test scores of third through eighth graders, pooled over 2009 to 2019 | 95% CI −0.060 to −0.004 | 12,207 districts | none (each district counts once) | Stanford Education Data Archive |   |
| mth_mean_sd | −0.010 | math gap in standard deviations of the national cohort, male minus female (negative means girls scored higher) | mean over districts of the male minus female math gap, bootstrap interval over states | school districts with a math gap estimate | test scores of third through eighth graders, pooled over 2009 to 2019 | 95% CI −0.019 to −0.002 | 12,207 districts | none (each district counts once) | Stanford Education Data Archive |   |
| mth_mean_w | −0.06 | math gap in grade levels, male minus female (negative means girls scored higher) | math-test-weighted mean over districts of the male minus female math gap, bootstrap interval over states | school districts with a math gap estimate | test scores of third through eighth graders, pooled over 2009 to 2019 | 95% CI −0.09 to −0.02 | 12,207 districts | number of math test scores in the district | Stanford Education Data Archive |   |
| mth_p5 | −0.35 | math gap in grade levels, male minus female (negative means girls scored higher) | 5th percentile of district gaps, each district counted once | school districts with a math gap estimate | test scores of third through eighth graders, pooled over 2009 to 2019 |   | 12,207 districts | none (each district counts once) | Stanford Education Data Archive |   |
| mth_p95 | +0.29 | math gap in grade levels, male minus female (negative means girls scored higher) | 95th percentile of district gaps, each district counted once | school districts with a math gap estimate | test scores of third through eighth graders, pooled over 2009 to 2019 |   | 12,207 districts | none (each district counts once) | Stanford Education Data Archive |   |
| mth_slope | +0.025 | grade levels of the math gap per ten points of district income rank | slope of the math gap on the district's percentile rank of income (0 to 100), test-weighted regression without state indicators, errors clustered on state | school districts with both gaps and a district income | test scores of third through eighth graders, pooled over 2009 to 2019; district income from the American Community Survey, 2019 to 2023 | 95% CI 0.019 to 0.031 | 12,136 districts | number of math and reading test scores in the district | Stanford Education Data Archive |   |
| mth_slope_fe | +0.027 | grade levels of the math gap per ten points of district income rank | slope of the math gap on the district's percentile rank of income (0 to 100), test-weighted regression with a separate intercept for each state and state indicators, errors clustered on state | school districts with both gaps and a district income | test scores of third through eighth graders, pooled over 2009 to 2019; district income from the American Community Survey, 2019 to 2023 | 95% CI 0.022 to 0.032 | 12,136 districts | number of math and reading test scores in the district | Stanford Education Data Archive |   |
| mth_state_high | +0.26 | math gap in grade levels, male minus female (negative means girls scored higher) | test-weighted mean of district gaps in the state (the interval treats districts as independent) | school districts in the state with the largest math lead for boys | test scores of third through eighth graders, pooled over 2009 to 2019; district income from the American Community Survey, 2019 to 2023 | 95% CI 0.23 to 0.30 | 326 districts | number of math and reading test scores in the district | Stanford Education Data Archive |   |
| mth_state_low | −0.34 | math gap in grade levels, male minus female (negative means girls scored higher) | test-weighted mean of district gaps in the state (the interval treats districts as independent) | school districts in the state with the largest math lead for girls | test scores of third through eighth graders, pooled over 2009 to 2019; district income from the American Community Survey, 2019 to 2023 | 95% CI −0.38 to −0.30 | 135 districts | number of math and reading test scores in the district | Stanford Education Data Archive |   |
| mth_state_share | 36% | percent of the variance of district math gaps that lies between states | R squared of district math gaps on state, weighted by test scores; interval from resampling districts | school districts with a reading gap, a math gap and a district income | test scores of third through eighth graders, pooled over 2009 to 2019; district income from the American Community Survey, 2019 to 2023 | 95% CI 34% to 40% | 12,136 districts | number of math and reading test scores in the district | Stanford Education Data Archive | Resampling districts adds a small upward bias to the share, so the interval sits slightly high. |
| mth_state_share_unw | 24% | percent of the variance of district math gaps that lies between states | R squared of district math gaps on state, weighted by districts, each counted once; interval from resampling districts | school districts with a reading gap, a math gap and a district income | test scores of third through eighth graders, pooled over 2009 to 2019; district income from the American Community Survey, 2019 to 2023 | 95% CI 23% to 26% | 12,136 districts | none (each district counts once) | Stanford Education Data Archive | Resampling districts adds a small upward bias to the share, so the interval sits slightly high. |
| n_boys_rla | 14 | districts | count of districts where boys' mean reading score exceeds girls' | school districts with a reading gap estimate | test scores of third through eighth graders, pooled over 2009 to 2019 |   |   | none (full count) | Stanford Education Data Archive |   |
| n_county | 2,996 | counties | count of counties | counties with a male-female math gap estimate | test scores of third through eighth graders, pooled over 2009 to 2019 |   |   | none (full count) | Stanford Education Data Archive |   |
| n_county_rla | 3,125 | counties | count of counties | counties with a male-female reading gap estimate | test scores of third through eighth graders, pooled over 2009 to 2019 |   |   | none (full count) | Stanford Education Data Archive |   |
| n_county_se | 2,938 | counties | count of counties | counties with a math gap estimate and a standard error | test scores of third through eighth graders, pooled over 2009 to 2019 |   |   | none (full count) | Stanford Education Data Archive |   |
| n_girls_rla | 12,313 | districts | count of districts where girls' mean reading score exceeds boys' | school districts with a reading gap estimate | test scores of third through eighth graders, pooled over 2009 to 2019 |   |   | none (full count) | Stanford Education Data Archive |   |
| n_inc | 12,136 | districts | count of districts | school districts with a reading gap, a math gap, an all-student score and a district income | test scores of third through eighth graders, pooled over 2009 to 2019; district income from the American Community Survey, 2019 to 2023 |   |   | none (full count) | Stanford Education Data Archive |   |
| n_inc1 | 9,993 | districts | count of districts | school districts in the income sample that also have an income when each tract belongs to one district only | test scores of third through eighth graders, pooled over 2009 to 2019; district income from the American Community Survey, 2019 to 2023 |   |   | none (full count) | Stanford Education Data Archive |   |
| n_map_nodata | 146 | counties | count of drawn counties with no math gap estimate | counties and county equivalents on the map (Puerto Rico is not drawn) | test scores of third through eighth graders, pooled over 2009 to 2019 |   |   | none (full count) | Stanford Education Data Archive |   |
| n_mth | 12,207 | districts | count of districts | school districts with a male-female math gap estimate (Virginia has none, DC is absent and Hawaii has no districts) | test scores of third through eighth graders, pooled over 2009 to 2019 |   |   | none (full count) | Stanford Education Data Archive |   |
| n_no_income | 38 | districts | count of districts left out of the income figures | school districts with both gaps but no district income or no all-student score | test scores of third through eighth graders, pooled over 2009 to 2019; district income from the American Community Survey, 2019 to 2023 |   |   | none (full count) | Stanford Education Data Archive |   |
| n_rla | 12,327 | districts | count of districts | school districts with a male-female reading gap estimate | test scores of third through eighth graders, pooled over 2009 to 2019 |   |   | none (full count) | Stanford Education Data Archive |   |
| n_sample | 5,000 | districts | size of a random sample drawn with a fixed seed | districts drawn at random for the scatter | test scores of third through eighth graders, pooled over 2009 to 2019; district income from the American Community Survey, 2019 to 2023 |   | 12,136 districts | none (sample size) | Stanford Education Data Archive |   |
| n_state_slopes | 45 | states | count of states given a slope | states with at least the minimum number of districts in the income sample | test scores of third through eighth graders, pooled over 2009 to 2019; district income from the American Community Survey, 2019 to 2023 |   |   | none (full count) | Stanford Education Data Archive |   |
| n_state_slopes_clear | 38 | states | count of states where the 95% interval (heteroskedasticity-robust) for that slope lies above zero | states with at least the minimum number of districts in the income sample | test scores of third through eighth graders, pooled over 2009 to 2019; district income from the American Community Survey, 2019 to 2023 |   |   | none (full count) | Stanford Education Data Archive |   |
| n_state_slopes_pos | 45 | states | count of states where the test-weighted slope of the math gap on income rank is positive | states with at least the minimum number of districts in the income sample | test scores of third through eighth graders, pooled over 2009 to 2019; district income from the American Community Survey, 2019 to 2023 |   |   | none (full count) | Stanford Education Data Archive |   |
| n_states | 48 | states | count of states | states with districts in the income sample (no Virginia, Hawaii or DC) | test scores of third through eighth graders, pooled over 2009 to 2019; district income from the American Community Survey, 2019 to 2023 |   |   | none (full count) | Stanford Education Data Archive |   |
| n_states_boys | 20 | states | count of states whose test-weighted mean math gap is above zero | states in the income sample | test scores of third through eighth graders, pooled over 2009 to 2019; district income from the American Community Survey, 2019 to 2023 |   |   | none (full count) | Stanford Education Data Archive |   |
| n_states_boys_clear | 13 | states | count of states whose 95% interval for the test-weighted mean math gap lies above zero | states in the income sample | test scores of third through eighth graders, pooled over 2009 to 2019; district income from the American Community Survey, 2019 to 2023 |   |   | none (full count) | Stanford Education Data Archive |   |
| n_states_boys_clear_lin | 14 | states | count of states whose 95% interval for the test-weighted mean math gap lies above zero when the standard error is the Taylor-linearized one for a weighted mean of independent districts | states in the income sample | test scores of third through eighth graders, pooled over 2009 to 2019; district income from the American Community Survey, 2019 to 2023 |   |   | none (full count) | Stanford Education Data Archive |   |
| pct_boys_inc_d1 | 23.7% | percent of districts where boys' mean math score exceeds girls' | share of districts (each counted once) where boys' mean math score exceeds girls', bootstrap interval over states | school districts in the lowest tenth of districts by district median household income | test scores of third through eighth graders, pooled over 2009 to 2019; district income from the American Community Survey, 2019 to 2023 | 95% CI 15.3% to 33.1% | 1,213 districts | none (each district counts once) | Stanford Education Data Archive |   |
| pct_boys_mth | 43.8% | percent of districts where the math gap is above zero | share of districts (each counted once) where boys' mean math score exceeds girls', bootstrap interval over states | school districts with a math gap estimate | test scores of third through eighth graders, pooled over 2009 to 2019 | 95% CI 38.1% to 49.4% | 12,207 districts | none (each district counts once) | Stanford Education Data Archive |   |
| pct_boys_mth_w | 36.5% | percent of math test scores in districts where the math gap is above zero | share of math test scores that are in districts where boys' mean exceeds girls', bootstrap interval over states | school districts with a math gap estimate | test scores of third through eighth graders, pooled over 2009 to 2019 | 95% CI 29.8% to 44.1% | 12,207 districts | number of math test scores in the district | Stanford Education Data Archive |   |
| pct_cty_boys | 34.8% | percent of counties where the math gap is above zero | share of counties (each counted once) where boys' mean math score exceeds girls', bootstrap interval over states | counties with a male-female math gap estimate | test scores of third through eighth graders, pooled over 2009 to 2019 | 95% CI 25.9% to 44.7% | 2,996 counties | none (each county counts once) | Stanford Education Data Archive |   |
| pct_girls_inc_d10 | 33.4% | percent of districts where girls' mean math score exceeds boys' | share of districts (each counted once) where girls' mean math score exceeds boys', bootstrap interval over states | school districts in the highest tenth of districts by district median household income | test scores of third through eighth graders, pooled over 2009 to 2019; district income from the American Community Survey, 2019 to 2023 | 95% CI 25.2% to 40.5% | 1,214 districts | none (each district counts once) | Stanford Education Data Archive |   |
| pct_girls_rla | 99.9% | percent of districts where the reading gap is below zero | share of districts (each counted once) where girls' mean reading score exceeds boys', bootstrap interval over states | school districts with a reading gap estimate | test scores of third through eighth graders, pooled over 2009 to 2019 | 95% CI 99.8% to 100.0% | 12,327 districts | none (each district counts once) | Stanford Education Data Archive |   |
| pct_mth_clear_boys | 21.5% | percent of districts whose 95% interval for the math gap lies above zero | share of districts (each counted once) whose 95% interval, built from the estimate and its standard error, lies above zero; bootstrap interval over states | school districts with a math gap estimate and a standard error | test scores of third through eighth graders, pooled over 2009 to 2019 | 95% CI 17.3% to 26.3% | 12,140 districts | none (each district counts once) | Stanford Education Data Archive |   |
| pct_mth_clear_girls | 31.3% | percent of districts whose 95% interval for the math gap lies below zero | share of districts (each counted once) whose 95% interval, built from the estimate and its standard error, lies below zero; bootstrap interval over states | school districts with a math gap estimate and a standard error | test scores of third through eighth graders, pooled over 2009 to 2019 | 95% CI 25.7% to 37.8% | 12,140 districts | none (each district counts once) | Stanford Education Data Archive |   |
| pct_mth_unclear | 47.2% | percent of districts whose 95% interval for the math gap includes zero | share of districts (each counted once) whose 95% interval, built from the estimate and its standard error, includes zero; bootstrap interval over states | school districts with a math gap estimate and a standard error | test scores of third through eighth graders, pooled over 2009 to 2019 | 95% CI 42.2% to 51.8% | 12,140 districts | none (each district counts once) | Stanford Education Data Archive |   |
| pct_rla_clear_girls | 98.9% | percent of districts whose 95% interval for the reading gap lies below zero | share of districts (each counted once) whose 95% interval, built from the estimate and its standard error, lies below zero; bootstrap interval over states | school districts with a reading gap estimate and a standard error | test scores of third through eighth graders, pooled over 2009 to 2019 | 95% CI 98.3% to 99.4% | 12,254 districts | none (each district counts once) | Stanford Education Data Archive |   |
| philly_gap | −0.33 | math gap in grade levels, male minus female (negative means girls scored higher) | SEDA's male minus female math gap for the district; the interval uses SEDA's empirical-Bayes standard error | students in the Philadelphia City SD | test scores of third through eighth graders, pooled over 2009 to 2019 | 95% CI −0.35 to −0.32 | 960,980 math test scores | none (one district) | Stanford Education Data Archive |   |
| rank_step | 10 | points of district income rank (0 to 100) | step of the income rank in which slopes are reported | school districts with both gaps and a district income | test scores of third through eighth graders, pooled over 2009 to 2019; district income from the American Community Survey, 2019 to 2023 |   |   | none (definition) | Stanford Education Data Archive |   |
| rgap_inc_d1 | −0.79 | reading gap in grade levels, male minus female (negative means girls scored higher) | test-weighted mean of the male minus female reading gap within a tenth of districts (equal numbers of districts), bootstrap interval over states | school districts in the lowest tenth of districts by district median household income | test scores of third through eighth graders, pooled over 2009 to 2019; district income from the American Community Survey, 2019 to 2023 | 95% CI −0.89 to −0.71 | 1,213 districts | number of math and reading test scores in the district | Stanford Education Data Archive |   |
| rgap_inc_d10 | −0.87 | reading gap in grade levels, male minus female (negative means girls scored higher) | test-weighted mean of the male minus female reading gap within a tenth of districts (equal numbers of districts), bootstrap interval over states | school districts in the highest tenth of districts by district median household income | test scores of third through eighth graders, pooled over 2009 to 2019; district income from the American Community Survey, 2019 to 2023 | 95% CI −0.97 to −0.78 | 1,214 districts | number of math and reading test scores in the district | Stanford Education Data Archive |   |
| rgap_inc_range | 0.13 | grade levels (largest minus smallest of the ten decile means of the reading gap) | largest minus smallest test-weighted mean reading gap across the ten income tenths, bootstrap interval over states | school districts with both gaps and a district income | test scores of third through eighth graders, pooled over 2009 to 2019; district income from the American Community Survey, 2019 to 2023 | 95% CI 0.06 to 0.24 | 12,136 districts | number of math and reading test scores in the district | Stanford Education Data Archive | The largest minus the smallest of ten noisy means is biased upward. |
| rla_inc_r | −0.09 | correlation coefficient | test-weighted correlation of the district reading gap with the district's percentile rank of income, bootstrap interval over states | school districts with both gaps and a district income | test scores of third through eighth graders, pooled over 2009 to 2019; district income from the American Community Survey, 2019 to 2023 | 95% CI −0.21 to 0.04 | 12,136 districts | number of math and reading test scores in the district | Stanford Education Data Archive |   |
| rla_mean | −0.76 | reading gap in grade levels, male minus female (negative means girls scored higher) | mean over districts of the male minus female reading gap, bootstrap interval over states | school districts with a reading gap estimate | test scores of third through eighth graders, pooled over 2009 to 2019 | 95% CI −0.82 to −0.70 | 12,327 districts | none (each district counts once) | Stanford Education Data Archive |   |
| rla_mean_sd | −0.228 | reading gap in standard deviations of the national cohort, male minus female (negative means girls scored higher) | mean over districts of the male minus female reading gap, bootstrap interval over states | school districts with a reading gap estimate | test scores of third through eighth graders, pooled over 2009 to 2019 | 95% CI −0.246 to −0.211 | 12,327 districts | none (each district counts once) | Stanford Education Data Archive |   |
| rla_mean_w | −0.79 | reading gap in grade levels, male minus female (negative means girls scored higher) | reading-test-weighted mean over districts of the male minus female reading gap, bootstrap interval over states | school districts with a reading gap estimate | test scores of third through eighth graders, pooled over 2009 to 2019 | 95% CI −0.85 to −0.73 | 12,327 districts | number of reading test scores in the district | Stanford Education Data Archive |   |
| rla_p5 | −1.18 | reading gap in grade levels, male minus female (negative means girls scored higher) | 5th percentile of district gaps, each district counted once | school districts with a reading gap estimate | test scores of third through eighth graders, pooled over 2009 to 2019 |   | 12,327 districts | none (each district counts once) | Stanford Education Data Archive |   |
| rla_p95 | −0.36 | reading gap in grade levels, male minus female (negative means girls scored higher) | 95th percentile of district gaps, each district counted once | school districts with a reading gap estimate | test scores of third through eighth graders, pooled over 2009 to 2019 |   | 12,327 districts | none (each district counts once) | Stanford Education Data Archive |   |
| rla_state_high | −0.46 | reading gap in grade levels, male minus female (negative means girls scored higher) | test-weighted mean of district gaps in the state (the interval treats districts as independent) | school districts in the state with the smallest reading lead for girls | test scores of third through eighth graders, pooled over 2009 to 2019; district income from the American Community Survey, 2019 to 2023 | 95% CI −0.50 to −0.42 | 326 districts | number of math and reading test scores in the district | Stanford Education Data Archive |   |
| rla_state_low | −1.28 | reading gap in grade levels, male minus female (negative means girls scored higher) | test-weighted mean of district gaps in the state (the interval treats districts as independent) | school districts in the state with the largest reading lead for girls | test scores of third through eighth graders, pooled over 2009 to 2019; district income from the American Community Survey, 2019 to 2023 | 95% CI −1.32 to −1.25 | 234 districts | number of math and reading test scores in the district | Stanford Education Data Archive |   |
| rla_state_share | 70% | percent of the variance of district reading gaps that lies between states | R squared of district reading gaps on state, weighted by test scores; interval from resampling districts | school districts with a reading gap, a math gap and a district income | test scores of third through eighth graders, pooled over 2009 to 2019; district income from the American Community Survey, 2019 to 2023 | 95% CI 68% to 72% | 12,136 districts | number of math and reading test scores in the district | Stanford Education Data Archive | Resampling districts adds a small upward bias to the share, so the interval sits slightly high. |
| rla_state_share_unw | 47% | percent of the variance of district reading gaps that lies between states | R squared of district reading gaps on state, weighted by districts, each counted once; interval from resampling districts | school districts with a reading gap, a math gap and a district income | test scores of third through eighth graders, pooled over 2009 to 2019; district income from the American Community Survey, 2019 to 2023 | 95% CI 45% to 49% | 12,136 districts | none (each district counts once) | Stanford Education Data Archive | Resampling districts adds a small upward bias to the share, so the interval sits slightly high. |
| se_med_mth | 0.06 | standard error of a district's math gap, grade levels | median over districts of the standard error SEDA reports for the male minus female math gap | school districts with a math gap estimate and a standard error | test scores of third through eighth graders, pooled over 2009 to 2019 |   | 12,140 districts | none (each district counts once) | Stanford Education Data Archive |   |
| size_r | −0.02 | correlation coefficient | correlation of the percentile rank of a district's math test scores with the percentile rank of its math gap (a rank correlation), each district counted once, bootstrap interval over states | school districts with a math gap estimate | test scores of third through eighth graders, pooled over 2009 to 2019 | 95% CI −0.09 to 0.05 | 12,207 districts | none (each district counts once) | Stanford Education Data Archive |   |
| tests_inc | 193.8 million | math test scores (one per student, grade and year tested) | sum of the math test scores behind the district estimates | students in the districts with both gaps and a district income | test scores of third through eighth graders, pooled over 2009 to 2019; district income from the American Community Survey, 2019 to 2023 |   | 12,136 districts | none (full count) | Stanford Education Data Archive |   |

## Definitions

- Bootstrap interval: A range for an estimate found by recomputing it on many random resamples of the data and reading the range from the results. It shows how much the estimate would vary from sample to sample.
- SD units (SEDA): Test scores stated in units of the typical spread of scores around the national average, so that a score reads as the number of such spreads above or below the national average. SEDA also reports scores in grade levels, which are a separate scale.
- Ecological inference: Drawing conclusions about individuals from data about places or groups. A pattern across counties need not hold for the people inside them.

## Methods

This chapter compares male-female test score gaps in reading and math across school districts and counties, from the Stanford Education Data Archive 6.0, pooled over third through eighth grade and the 2009 to 2019 school years. A negative gap means girls scored higher. District income is not in the archive: it is built from tract incomes of the American Community Survey for 2019 to 2023, split across districts by land area, and it is measured after the scores (2009 to 2019). Means weight a district by its number of math and reading test scores, or count each district once where the text says so.

### Which grades, years and districts the gaps cover

Gaps come from the Stanford Education Data Archive 6.0, pooled over third through eighth grade and the 2009 to 2019 school years, through the book's SEDA loader.

The build loads every SEDA row and filters to the gender gap rows, because the loader's default filter returns only headline rows and its grade-unit check rejects gap rows.

### How the gaps are signed and scaled

The gap rows are SEDA's male-female differences, which equal the male mean minus the female mean (the build checks this on both scales), so a negative gap means girls scored higher.

Figures and facts use grade levels, in which one unit is the national average learning gain over a school year; two facts use standard-deviation units and say so.

Explains Table 1 and Table 2.

### How district income is built and matched to the scores

District income is not in SEDA. It is the household-weighted mean of tract median household income (ACS 2019 to 2023, dollars, capped at 250,001), with each tract split across the districts it overlaps in proportion to its land area in each; districts are the 2021 to 2022 school-year districts of the NCES crosswalk, matched to SEDA by district code, and the income (2019 to 2023) is measured after the scores (2009 to 2019). A check assigns each tract to one district only and recomputes the income deciles.

The income sample holds the districts with both gaps, an income and an all-student score, and ranks and tenths hold equal numbers of districts.

Explains Table 3 and Table 4.

### How districts and counties are weighted and drawn

Means weight a district by its number of math and reading test scores, or count each district once where the text says so.

County estimates are SEDA's county rows, each county counted once; the Alaska code 02261 is drawn on its two successors. Missing values are never set to zero: a county without a math estimate is drawn as no data with its reason.

Explains Table 5.

### How uncertainty is measured

Intervals for decile means, shares, correlations and differences resample whole states (2,000 draws, seed 12345); intervals for state means use a normal approximation on the effective number of districts and treat a state's districts as independent (a linearized standard error is reported as a check); slopes come from weighted regressions with errors clustered on state.

The state share of variance is the R squared of a regression on state indicators, test-weighted and with each district counted once.

A district or county is counted as clearly signed when its 95% interval, built from the estimate and its standard error with the normal multiplier, lies on one side of zero.

### How the scatter and density figures are drawn

Scatter dots are a random sample drawn with the same seed. Densities use Gaussian kernels on a fixed grid.

Explains Table 1 and Table 4.

## What these data cannot show

The patterns describe school districts and counties. A district's income is not a family's income, and the data do not say that a girl or a boy from a poorer family scores differently from one in a richer family in the same district (ecological inference). A district gap is an average over every student the state tested in third through eighth grade, so it cannot show when in school a gap opens.

Each district's gap carries error, and the standard error of a district's math gap is typically larger than the mean math gap. The decile means, with intervals, carry the income finding, and single districts do not. SEDA's estimates are empirical-Bayes estimates, which pull the noisiest districts toward the average and narrow the spread of district gaps. If the shrinkage pulls toward a state or national mean, it could change the shares of variance that lie between states. The 95% intervals built from the empirical-Bayes standard errors describe the shrunken estimates. The scores come from state tests that differ in content and format, and a gap can reflect what a test asks as well as how girls and boys learned.

District income is built for this chapter and is not part of SEDA. It is a household-weighted mean of tract medians, split across overlapping districts by land area, so it is not the median income of a district's households, and the survey caps the highest incomes. It is measured for 2019 to 2023, after the 2009 to 2019 scores, on the district boundaries of the 2021 to 2022 school year. A district whose boundary or code changed may be matched to SEDA imperfectly or not at all. In all, 38 districts with both gaps have no income and are left out of the income figures. Districts that differ in income also differ in other ways, such as the racial composition of their students, and the chapter does not separate those.

SEDA has no math estimates for Virginia, no rows for the District of Columbia and no districts in Hawaii (Hawaii's counties are in its county file and are drawn), and Puerto Rico is not covered. The estimates describe public school students tested in 2009 to 2019 and say nothing about later years. SEDA reports two groups, male and female. Whether gaps in school scores carry into adult outcomes is the subject of Chapter 4.4, Boys and mobility (https://vishalsingh.org/zipcode/boys-and-mobility; planned), which compares boys and girls raised in the same neighborhoods.

## References

1. Reardon, Sean F.; Fahle, Erin M.; Kalogrides, Demetra; Podolsky, Anne; Zárate, Rosalía C. (2019). Gender Achievement Gaps in U.S. School Districts. American Educational Research Journal 56(6): 2474-2508. https://doi.org/10.3102/0002831219843824
2. Hyde, Janet S.; Lindberg, Sara M.; Linn, Marcia C.; Ellis, Amy B.; Williams, Caroline C. (2008). Gender Similarities Characterize Math Performance. Science 321(5888): 494-495. https://doi.org/10.1126/science.1160364
3. Fryer, Roland G., Jr.; Levitt, Steven D. (2010). An Empirical Analysis of the Gender Gap in Mathematics. American Economic Journal: Applied Economics 2(2): 210-240. https://doi.org/10.1257/app.2.2.210

## Provenance

- Pipeline version: 0.1.0
- Build script: book2/chapters/gender-gaps-in-reading-and-math/build.py (sha256 f03bc4b1a85f5545112e2e7510329aa697d066922bec68920f5a20ac1b989663)
- Built on: 2026-10-07
- Site commit: c2b1ff3309cb4a10833f884da91ac0057a66ad0a

Inputs:

| File | Size (bytes) | SHA-256, first 1 MB | SHA-256, last 1 MB |
| --- | ---: | --- | --- |
| book2/chapters/gender-gaps-in-reading-and-math/gaps.py | 18,838 | bb96ad502a57c7b7093967c7beb2111e2f0a951e674ff8e0c8b52dfebb0c9b7b | bb96ad502a57c7b7093967c7beb2111e2f0a951e674ff8e0c8b52dfebb0c9b7b |
| data/geo/dim_geo.parquet | 1,180,921 | 29d5e4d943552d54aeb6ba92aadc939d9bd15bdea553c78357dbd516757be627 | e58c8c488df54ab3187f5461f179322afdc49c9bd2315404003fe1272eea32c3 |
| data/staged/crosswalks/tract_to_district_2022.parquet | 3,162,231 | 06a12eab3f61d43d70bbc874beea499f78d4b3fb7c833839909a853479f3753e | 13ffd00dbaec66f9662d8268232313e57938326840cb9303dcc9e3da6eab0a12 |
| data/staged/education/seda_county_poolsub_6.0.parquet | 3,604,804 | 730134713ed3cae8983a255caa2a1be786a2d50688a949b11372f86e90065976 | 356fb555b2de0cc487cb4c37c95f29cee0497a96b4f7385fa6f10b9975bee58d |
| data/staged/education/seda_geodist_poolsub_6.0.parquet | 12,344,123 | 48de785e0dc62aacfc4608757670e6912b318c6c02baa3a716ef4c131c0eb6b5 | 5a22f706cfd0295923dd169d985d44d0da9076236e3afe38d59c5f6dd3e518a6 |
| data/staged/ses/acs5_tract2020_2019_2023.parquet | 7,232,424 | 5fe6f730a35225f14d5d8abfc6b9bce58b52e545f309a7eac74f1dae74d7a31c | 7b4a2475b3852d3f91804716a04566ef8ceadacec2be9fb85f5eb642078efac6 |

Outputs:

| File | SHA-256 |
| --- | --- |
| facts.json | fbcfab3b9afd6a3d57b4f038fcd8374da3bfa9f119d19bc43352f8ac1a612edc |
| figures/gap_by_income_decile.json | a03a2c5902a4a1d61ad9cede1dec83114f09eeed7325a98fa43fdd93d3af6087 |
| figures/gap_distributions.json | cafb0c0869e8fed221b3780aca0106b85088eb00a60abb410981f9715a838493 |
| figures/math_gap_by_income.json | fa6db9b26a4406bbe2bb944728ef380c7980df933e46c0f646a91b6e90bb7ff4 |
| figures/math_gap_map.json | cc7576e47363af473d7b8d9401654da7487158ea7a7e43dc3aa6b222a289cb29 |
| figures/state_gap_means.json | a5c88632fd30625f6b573e64160e4f59d39a3fd8aa80dd215fb7ec338087e831 |
| meta.json | ba82023f1da7e23d9d49372cda72d90e54c92432f60ee44a06feb0a637e19010 |
| public/data/gender-gaps-in-reading-and-math/math_gap_by_income.json | a5ae9c5156b17219a1a803d34ca8e18301fb80af0ed3d8bc01bf954b5a4eab93 |
| public/data/gender-gaps-in-reading-and-math/math_gap_map.json | 769ff45da54cace745a8ec788e610b37fba44ced7514b2f3e4711b6731103557 |
| refs.json | 7c37e6d73029fbec5885e382c2491bf425c7cc156bc02a9a492b1ba2d7c675b4 |

## Machine formats

- Report (HTML): https://vishalsingh.org/zipcode/gender-gaps-in-reading-and-math/report
- Report (Markdown): https://vishalsingh.org/zipcode/gender-gaps-in-reading-and-math/report.md
- Every estimate (JSON): https://vishalsingh.org/zipcode/report/gender-gaps-in-reading-and-math/facts.json
- Table 1, figure "gap_distributions" (JSON): https://vishalsingh.org/zipcode/report/gender-gaps-in-reading-and-math/figures/gap_distributions.json
- Table 2, figure "state_gap_means" (JSON): https://vishalsingh.org/zipcode/report/gender-gaps-in-reading-and-math/figures/state_gap_means.json
- Table 3, figure "gap_by_income_decile" (JSON): https://vishalsingh.org/zipcode/report/gender-gaps-in-reading-and-math/figures/gap_by_income_decile.json
- Table 4, figure "math_gap_by_income" (JSON): https://vishalsingh.org/zipcode/report/gender-gaps-in-reading-and-math/figures/math_gap_by_income.json
- Table 5, figure "math_gap_map" (JSON): https://vishalsingh.org/zipcode/report/gender-gaps-in-reading-and-math/figures/math_gap_map.json
- Payload of "math_gap_by_income" (JSON): https://vishalsingh.org/zipcode/data/gender-gaps-in-reading-and-math/math_gap_by_income.json
- Payload of "math_gap_map" (JSON): https://vishalsingh.org/zipcode/data/gender-gaps-in-reading-and-math/math_gap_map.json
- Index of every report: https://vishalsingh.org/zipcode/llms.txt; all reports in one file: https://vishalsingh.org/zipcode/llms-full.txt

---

# School discipline and incarceration: report

Chapter 3.4 of ZIP Code Destiny, by Vishal Singh (NYU Stern). Generated from the chapter's data files; the reading version is at https://vishalsingh.org/zipcode/school-discipline-and-incarceration. Table n of this report is figure n of the chapter.

## Identification

- Title: School discipline and incarceration
- Label: The crossroads
- Chapter: 3.4 (Part III, School; An American life)
- Question: How do suspension rates rise with neighborhood deprivation, and how far apart are Black and white children raised in the same tract on later incarceration?
- Dek: Out-of-school suspensions in the 2020 to 2021 school year ran from 3.1 per thousand students in the least deprived tenth of school districts to 23.6 in the most deprived.
- Grain: institutions
- Status: published
- Published: 2026-10-09
- Updated: not revised
- Data sources:
  - Civil Rights Data Collection. Civil Rights Data Collection 2020-21 school year: students suspended out of school, expelled, referred to law enforcement and arrested, and enrollment, by district, with Black and white students separately; years: 2020-21 school year; weights: none (counts reported by districts); district rates weighted by enrollment
  - Area Deprivation Index. Neighborhood Atlas Area Deprivation Index 2023 (v4.0.1), tract national rank, averaged to districts and to 2010 tracts; years: 2023 release (version 4.0.1; built on ACS 2019-2023); weights: 2019-2023 tract population within districts; land area within 2010 tracts
  - American Community Survey. ACS 2019-2023 tract population, used only to weight the tracts of a district when its deprivation rank is averaged; years: 2019-2023 (5-year); weights: 2019-2023 tract population
  - Opportunity Atlas. Tract incarceration shares of children from low-income families, children born 1978 to 1983, incarcerated on April 1, 2010, native 2010 tracts, genders pooled, by race; years: 1978-1983 birth cohorts (2018 release); weights: 2010 census tract population
  - USDA Food Access Research Atlas. 2010 census population of each 2010 tract, used only as the weight for Opportunity Atlas comparisons; years: 2019 release (population from the 2010 census); weights: none (census counts)
- Canonical URL: https://vishalsingh.org/zipcode/school-discipline-and-incarceration
- Report: https://vishalsingh.org/zipcode/school-discipline-and-incarceration/report
- Markdown: https://vishalsingh.org/zipcode/school-discipline-and-incarceration/report.md
- BibTeX key: singh2026schooldisciplineandincarceration

```bibtex
@misc{singh2026schooldisciplineandincarceration,
  author = {Singh, Vishal},
  title = {School discipline and incarceration},
  year = {2026},
  url = {https://vishalsingh.org/zipcode/school-discipline-and-incarceration},
  note = {In ZIP Code Destiny. Published 9 October 2026.}
}
```

## Principal estimates

The facts the dek and the opening prose cite, in order of appearance.

| Estimand | Population | Period | Value | Interval | n | Weights | Source | Fact id |
| --- | --- | --- | ---: | --- | ---: | --- | --- | --- |
| enrollment-weighted mean of district rates (students suspended over students enrolled), bootstrap interval over counties | students enrolled in school districts that have a deprivation rank, at least 100 students and a reported suspension count in the least deprived tenth of districts by deprivation rank | 2020-21 school year | 3.1 | 95% CI 2.4 to 4.1 | 1,025 districts | district enrollment | Civil Rights Data Collection | oss_d1 |
| enrollment-weighted mean of district rates (students suspended over students enrolled), bootstrap interval over counties | students enrolled in school districts that have a deprivation rank, at least 100 students and a reported suspension count in the most deprived tenth of districts by deprivation rank | 2020-21 school year | 23.6 | 95% CI 20.2 to 27.3 | 1,022 districts | district enrollment | Civil Rights Data Collection | oss_d10 |
| observed Black suspensions over expected suspensions at the district's white rate, pooled, bootstrap interval over states | Black students enrolled in the 5156 districts with at least 50 Black and as many white students and a reported suspension count for each group | 2020-21 school year | 2.2 | 95% CI 2.1 to 2.4 | 5,156 districts | Black enrollment | Civil Rights Data Collection | oss_ratio |
| count of districts with at least 50 Black and as many white students and a reported count for each | school districts (local education agencies) that reported to the Civil Rights Data Collection | 2020-21 school year | 5,156 |   |   | none (full count) | Civil Rights Data Collection | bw_n |
| population-weighted mean over tracts (2010 tract population as the weight), bootstrap interval over counties | Black children raised in low-income families (parents at the 25th percentile of the national income distribution), in tracts where the Atlas also estimates the other group | children born 1978 to 1983, incarcerated on April 1, 2010 | 5.2% | 95% CI 5.0% to 5.3% | 27,795 tracts | 2010 census tract population | Opportunity Atlas | inc_black |
| population-weighted mean over tracts (2010 tract population as the weight), bootstrap interval over counties | white children raised in low-income families (parents at the 25th percentile of the national income distribution), in tracts where the Atlas also estimates the other group | children born 1978 to 1983, incarcerated on April 1, 2010 | 1.9% | 95% CI 1.8% to 2.0% | 27,795 tracts | 2010 census tract population | Opportunity Atlas | inc_white |

## Findings

Out-of-school suspensions in the 2020 to 2021 school year were 3.1 per thousand students in the least deprived tenth of school districts and 23.6 in the most deprived. Inside the same districts, Black students were suspended at 2.2 (95% CI 2.1 to 2.4) times the rate of white students, a comparison drawn from 5,156 districts. The second part of the chapter moves from districts to census tracts. Among children from low-income families who grew up in tracts where the Opportunity Atlas estimates both groups, 5.2% (95% CI 5.0% to 5.3%) of Black children and 1.9% (95% CI 1.8% to 2.0%) of white children were incarcerated on Census Day 2010.

The two parts describe different children. The district records are for students enrolled in 2020 to 2021, and the Opportunity Atlas follows children born in 1978 to 1983 into adulthood. No data in the chapter link a suspended student to a later incarceration, so what the two parts share is the question of how unevenly each outcome falls across places and between groups of children.

### How a district gets a suspension rate and a deprivation rank

The Civil Rights Data Collection gathers counts from every public school district for the Department of Education, and districts report their own counts. This chapter uses the 2020 to 2021 collection, with school counts summed to the 17,821 reporting districts (local education agencies). That year was disrupted by remote and hybrid teaching, so the levels describe an unusual year and are read here for gradients and gaps, not as typical rates. Summing the schools' counts gives 651,832 students suspended out of school that year, a student counted once for each school where they were suspended. School cells coded as suppressed or not reported add nothing to a district's sum.

A suspension rate is the number of students who received at least one out-of-school suspension, with and without disabilities, divided by the students enrolled, per thousand. A student suspended more than once at a school counts once there, a student suspended at two schools counts at each, and in-school suspensions are not included. The Collection counts Hispanic students of any race in their own group, so its Black and white students are not Hispanic. Expulsions are students expelled with or without educational services, and referrals to law enforcement and school-related arrests are counts of students at each stage, summed the same way. The Collection also counts expulsions under zero-tolerance policies, which are a subset of the expulsions with or without services and are not added again.

Deprivation is a tract measure, so a district needs a rank built from its tracts. The Area Deprivation Index ranks each tract nationally from the least to the most deprived.[1] A district's rank is the mean rank of the tracts assigned to it, weighted by tract population. Only districts known to the book's geography file (dim_geo), which assigns each tract to one school district, get a rank, which excludes charter, regional and single-grade-span districts. The index uses data for 2019 to 2023, close to the school year studied.

Of the 17,821 reporting districts, 10,257 (57.6%) have a rank, at least 100 students and a reported suspension count. They enroll 38.7 million students, 78.4% of all students in reporting districts.

Another 38 districts have a rank and enough students but report no suspension count, among them New York City, Los Angeles, Chicago and Miami-Dade. Their 6.2 million students are not in any estimate here.

### Suspension rates rise with deprivation, with no clear order at the top

Ranking districts from least to most deprived and cutting them into tenths of equal numbers of districts, the suspension rate climbs steeply over the first four tenths, from 3.1 (95% CI 2.4 to 4.1) per thousand students in the least deprived tenth to 20.1 (95% CI 17.3 to 23.1) in the fourth. It rises more slowly to 26.9 (95% CI 24.1 to 29.7) in decile 8, and the most deprived tenth is at 23.6 (95% CI 20.2 to 27.3).

The highest tenth's rate minus the most deprived tenth's is +3.2 (95% CI −0.9 to 7.1) per thousand, and the interval includes zero, so the top three tenths cannot be told apart. The rate in the most deprived tenth is 7.6 (95% CI 5.5 to 10.6) times the rate in the least deprived. Across districts, weighted by enrollment, the rate and the deprivation rank have a correlation of +0.42 (95% CI 0.37 to 0.46). The interval is wider when whole states are resampled in place of counties (+0.42 (95% CI 0.31 to 0.51)).

Table 1. Out-of-school suspensions were more common in more deprived districts, at 3.1 per thousand students in the least deprived tenth of districts against 23.6 in the most deprived

20 districts have rates above the top of the figure's vertical axis and are left off the dots, though they are in every estimate.

States differ in their rates among the ranked states (see the map below), from 3.1 per thousand students in Washington (96.4% of students covered) to 43.6 in Arkansas (93.2% covered). Deprived districts are not spread evenly across states. Within a state, each ten-point step up the deprivation rank went with 1.9 (95% CI 1.3 to 2.6) more students suspended per thousand, so the gradient is not only a difference between states.

### Every stage of exclusion is more common in the most deprived third of districts

Suspension is the widest stage. Among the 10,181 districts that report all four counts, 16.3 (95% CI 15.0 to 17.5) students per thousand were suspended, 1.73 (95% CI 1.55 to 1.90) were referred to law enforcement, 0.74 (95% CI 0.65 to 0.84) were expelled and 0.26 (95% CI 0.22 to 0.30) were arrested at school.

The three deeper stages are zero in most districts. Of these districts, 61.4% report no referrals, 75.5% no expulsions and 88.1% no arrests, so their rates come from a minority of districts.

Splitting the districts into thirds by deprivation, the rate in the most deprived third was 2.6 (95% CI 2.2 to 3.0) times the rate in the least deprived third for suspensions, 2.7 (95% CI 1.9 to 3.8) times for expulsions, 1.5 (95% CI 1.2 to 1.9) times for referrals and 1.8 (95% CI 1.2 to 2.5) times for arrests (Table 2). All four intervals lie above one.

For referrals and arrests the middle third is as high as the most deprived, at 1.7 (95% CI 1.3 to 2.0) and 2.1 (95% CI 1.5 to 3.0) times the least deprived third, so for those stages the rise is not steady across the thirds. The counts at each stage are students in a district, not students followed from one stage to the next, and the stages are not a sequence for any child.

Table 2. Every stage of school exclusion was more common in the most deprived third of districts than in the least deprived third

### Black and white students in the same district

Earlier work found that Black students are suspended more often than white students. Skiba and colleagues used one year of middle-school discipline records from an urban district. They found that the gap in referrals, suspensions and expulsions remained after controlling for socioeconomic status, and that Black students were referred more often for infractions that depend on a teacher's interpretation.[2] The Government Accountability Office, analyzing national data for 2013 to 2014, found that Black students were a larger share of suspended students than of enrolled students, and that the disparity persisted regardless of a school's poverty level or type.[3]

A comparison inside one district removes the differences between districts that a state or county average mixes in. The comparison uses the 5,156 reporting districts that have at least 50 Black and as many white students and a reported count for each group. They enroll 33.3 million students and 71.3% of the Black students in the Collection.

Pooled over those districts, 30.7 (95% CI 22.7 to 39.1) Black students per thousand were suspended. Weighting each district's white rate by its Black enrollment gives the rate those Black students would have had at their own district's white rate, 13.7 (95% CI 10.1 to 17.0) per thousand. Black students were therefore suspended at 2.2 (95% CI 2.1 to 2.4) times the white rate in the same districts. The intervals resample states, because this comparison includes districts that cannot be placed in a county.

How many districts follow that pattern depends on how districts are counted. Counting each district once, the Black rate is higher than the white rate in 65.4% (95% CI 57.5% to 72.4%) of districts, lower in 19.9% (95% CI 17.3% to 22.4%) and equal in 14.7% (95% CI 9.7% to 20.7%), almost always because both are zero. Weighting each district by its Black enrollment raises the first share to 84.0% (95% CI 77.1% to 88.8%), because districts with more Black students are more likely to have a higher Black rate. A small district's rate rests on few suspensions, so the unweighted share carries more noise than the weighted one.

Table 3. The Black suspension rate was higher than the white rate in 20 of the 24 districts with the most Black students

Table 3 shows the 24 districts with the most Black students among those compared. They are chosen by Black enrollment and not by their rates, and the Black rate is higher in 20 of them. It is not higher in Detroit, Clayton County in Georgia, the District of Columbia and Henry County in Georgia. Detroit and Clayton County enroll few white students, and the District of Columbia reports rates close to zero for both groups.

### Where suspension rates are highest

The map assigns each district to the county that holds most of its population and pools the districts of a county that report a count, weighted by enrollment. A county is drawn when those districts enroll at least 500 students and hold more than 75% of the students in all the districts assigned to it. That rule leaves out counties where a large district reports no count, such as Los Angeles and Cook counties. The map draws 2,812 counties.

The other 331 are gray. Of these, 33 are served by a district that reports no suspension count, among them the counties of New York City. Another 8 have districts that report no count holding a quarter or more of their students, 240 fall below the enrollment floor, and 50 have no district with a count assigned to them.

The highest tenth of county rates starts at 48.0 per thousand students and holds 282 counties. The rule for naming states is the number of counties in that tenth. 6 states have at least 15 of them, Georgia, Mississippi, Florida, Arkansas, South Carolina and Louisiana.

Together these states hold 61.0% of the highest tenth. In each of them at least 25.3% of the drawn counties are in that tenth, against one tenth of all drawn counties by definition.

State rates need the same care, because the districts in the analysis hold only part of the students in some states. Among the 38 states where they hold at least 75% of enrolled students, the pooled rate runs from the Washington and Arkansas values above. Florida, Maryland and California are not ranked, because the districts in the analysis hold less than that share of their students. The key stops at 61 per thousand students, and 143 counties above it share the color at the top of the key. The map shows the counts that districts recorded in a disrupted year and says nothing about how children behaved. The Collection also records each school's type of instruction in 2020 to 2021 in a separate school-level file that this chapter does not use, so the map cannot separate discipline from how much teaching was in person.

Table 4. Counties in the highest tenth of suspension rates were concentrated in 6 states, which hold 61.0% of them

### Incarceration across tracts

The rest of the chapter leaves school districts for census tracts and a different generation of children. The suspension counts above describe students enrolled in 2020 to 2021. The incarceration measure below describes children born in 1978 to 1983, and no record links a student in the Collection to a child in the Atlas.

The Opportunity Atlas links tax and census records for children born in 1978 to 1983 to the census tract where they grew up.[4] This chapter uses its incarceration measure for children whose parents were at the 25th percentile of the national income distribution. The measure is the share of children who were incarcerated on Census Day in 2010, when they were between their mid-twenties and early thirties. The staged files pool boys and girls, and the Atlas privacy noise is not clipped, so some tract estimates fall below zero.

Across the 71,478 tracts that have an estimate, a deprivation rank and a recorded 2010 population, the mean incarceration share is 2.1% (95% CI 2.0% to 2.2%), weighted by tract population. It is 1.1% (95% CI 1.0% to 1.2%) in the least deprived tenth of tracts and 3.3% (95% CI 3.1% to 3.5%) in the most deprived, and the tract rank and the share have a correlation of +0.28 (95% CI 0.25 to 0.30). The first tab of Table 5 shows that the rise is gradual across most of the range and steeper in the most deprived tenth.

The Atlas is built on 2010 tracts and the deprivation index on 2020 tracts, so each 2010 tract takes the land-area-weighted mean rank of the 2020 tracts that overlap it. The index uses data for 2019 to 2023 and the children grew up in the 1980s and 1990s, so the rank describes the tracts as they are now.

Table 5. Incarceration of children from low-income families rose with tract deprivation and was higher for Black than for white children in the same tracts

### Black and white children raised in the same tract

The Atlas also reports incarceration separately for Black and white children, who are non-Hispanic in its race groups as they are in the Collection,[5] and that allows a comparison inside one tract. A tract qualifies when both estimates exist and its 2010 population is positive, the same rules as in Chapter 4.3, Race, income and mobility (https://vishalsingh.org/zipcode/race-income-and-mobility), which makes the same comparison for adult income and links back here for the incarceration result. In all, 27,795 tracts in 1,468 counties qualify, and they hold 130.5 million people, 42.3% of the 2010 population. The tracts with both estimates are those with enough children of both groups, so they are not a sample of all tracts.

In these tracts, weighted by population, 5.2% (95% CI 5.0% to 5.3%) of Black children from low-income families were incarcerated against 1.9% (95% CI 1.8% to 2.0%) of white children, a gap of 3.2 (95% CI 3.1 to 3.4) points and a ratio of 2.7 (95% CI 2.5 to 2.8). The Black share is higher than the white share in 73.8% (95% CI 72.9% to 74.7%) of these tracts.

 (Note: Each tract estimate rests on a modest number of children, and the Atlas adds noise to protect privacy. Noise of that kind spreads the Black and white estimates in both directions, so it pulls the share of tracts with a higher Black rate toward an even split, not away from one.)

The gap is present at every level of tract deprivation (the second tab of Table 5). In the least deprived tenth of these tracts, 3.8% (95% CI 3.4% to 4.2%) of Black children and 1.2% (95% CI 1.1% to 1.4%) of white children were incarcerated, a gap of 2.6 (95% CI 2.2 to 2.9) points.

In the most deprived tenth the shares were 6.4% (95% CI 6.1% to 6.7%) and 2.6% (95% CI 2.4% to 2.8%), a gap of 3.8 (95% CI 3.4 to 4.1). The gap in points is larger in the most deprived tenth.

The ratio of the Black share to the white share is smaller in the most deprived tenth, 2.5 (95% CI 2.2 to 2.7) against 3.1 (95% CI 2.6 to 3.5), a fall of 0.6 (95% CI 0.1 to 1.1). Black children in the least deprived tenth of tracts were incarcerated more often than white children in the most deprived tenth, by 1.2 (95% CI 0.7 to 1.6) points. The gap does not close when tracts are compared at the same level of deprivation.

### Evidence on whether suspension and detention change later outcomes

The comparisons above are descriptions. Two studies test whether being suspended or incarcerated changes what happens next, and each is reported under its own design.

> Causal evidence: School suspension and adult crime (School attendance boundary change (quasi-experiment) and principal turnover)
>
> Bacher-Hicks, Billings and Deming used a large and sudden redrawing of attendance boundaries in one school district, which moved many students to different schools in a single year, to compare students sent to schools with higher suspension rates with students sent to schools with lower rates. A second design used changes of principal. Students assigned to schools that suspended more were more likely to be arrested and incarcerated as adults. The paper also reports lower educational attainment, with effects concentrated among male and minority students.[6]

> Causal evidence: Juvenile incarceration and later outcomes (Instrumental variables from the random assignment of cases to judges)
>
> Aizer and Doyle studied juvenile offenders in a large urban county. Cases were assigned to judges at random, and judges differ in how often they incarcerate, so a judge's tendency to incarcerate moves which juveniles are incarcerated for reasons unrelated to the case. Juveniles who were incarcerated were less likely to finish high school and more likely to be incarcerated as adults.[7]

Both estimates apply to the settings studied, one school district and one county, and to the years of those data. Neither says why suspension rates or incarceration shares differ across the districts and tracts in this chapter.

### Incarceration of Black children rises with deprivation under all four rankings

Table 6 ranks the 25,827 tracts where Black and white children are both estimated and that have all four rankings in four ways, with each ranking carried to 2010 tracts by the same land-area crosswalk as the ADI. Ranked by the ADI on this common sample, the share of Black children from low-income families who were incarcerated is 3.7% (95% CI 3.3% to 4.1%) in the least deprived tenth and 6.3% (95% CI 6.0% to 6.7%) in the most deprived, and the rank correlation between tenth and share is +1.00. Under the standardized index the two ends are 3.9% (95% CI 3.6% to 4.2%) and 6.4% (95% CI 6.1% to 6.7%), with a correlation of +0.98. Under median household income they are 3.8% (95% CI 3.5% to 4.1%) and 6.2% (95% CI 5.9% to 6.5%), and under the bachelor's degree share 4.1% (95% CI 3.7% to 4.5%) and 6.2% (95% CI 5.9% to 6.5%), with correlations of +1.00 and +1.00, so none of the alternatives counts as a change under the book's rule (Chapter 9.1, How far one ranking goes (https://vishalsingh.org/zipcode/how-far-one-ranking-goes)).

Table 6. Incarceration of Black children from low-income families is steadily worse toward the more deprived end under the ADI and under each of the three other rankings

### What these data cannot show

A suspension rate records what districts reported, not what children did. A higher rate in a deprived district, or for Black students, is consistent with more disruptive conduct, with harsher responses to the same conduct, with both, and with differences in how districts count, and these data cannot tell them apart. The counts are for a pandemic year, are self-reported, and omit the suppressed cells. The 38 districts with no count include some of the largest in the country, and charter, regional and single-grade-span districts have no deprivation rank, so the districts here are not all of the districts. In several states, among them Florida, Maryland and California, the districts in the analysis hold less than 75% of enrolled students, so no state rate is given for them. The Area Deprivation Index understates deprivation in high-cost cities (Chapter 1.1, What the deprivation index measures (https://vishalsingh.org/zipcode/anatomy-of-deprivation)), and a district's rank averages over tracts that differ a great deal. The map assigns each district to one county although some span several.

The incarceration part has its own limits. The Atlas measures incarceration on one day, so it counts who was incarcerated on that day and not who was ever incarcerated. It pools girls and boys, because race by gender is not in the staged data. Each tract estimate is noisy, the weights are tract population and not the number of children, and a tract with both a Black and a white estimate is one with enough children of both groups. A same-tract comparison removes differences between neighborhoods but not differences in family wealth, in the block, in schools or in how police and courts treat the children. Both parts describe places and groups, not individual children, and they describe different cohorts. No child here is followed from a suspension to a later incarceration, and the districts of 2020 to 2021 are not the tracts where the children of 1978 to 1983 grew up.

## Tables

### Table 1. Out-of-school suspensions were more common in more deprived districts, at 3.1 per thousand students in the least deprived tenth of districts against 23.6 in the most deprived

Students suspended out of school at least once per 1,000 students enrolled, by the district's deprivation rank (mean Area Deprivation Index national rank of its tracts), one dot per district, 2020-21 school year

No inline rows: the values of Table 1 are in https://vishalsingh.org/zipcode/data/school-discipline-and-incarceration/suspension_dose.json.

Note: Every district with at least 100 students enrolled and a reported suspension count is plotted, out of 10,257 districts; 20 of them have a rate above the top of the vertical axis and are not drawn, but they are in every estimate.

Note: The suspension rate is the number of students who received at least one out-of-school suspension, whether once or more than once, divided by the students enrolled. Deciles hold equal numbers of districts.

Note: Districts report these counts themselves and a single district's rate carries no sampling interval; the decile means and the correlation resample counties. Resampling states instead gives a wider interval for the correlation: +0.42.

Note: The Civil Rights Data Collection for the 2020-2021 school year was disrupted by remote and hybrid teaching, and districts report their own counts, so the levels describe an unusual year. Suppressed school cells add nothing to a district's sum.

Source: Civil Rights Data Collection, 2020-2021 school year, and Area Deprivation Index, 2023; enrollment-weighted district rates, full counts reported by districts

Data: https://vishalsingh.org/zipcode/report/school-discipline-and-incarceration/figures/suspension_dose.json

### Table 2. Every stage of school exclusion was more common in the most deprived third of districts than in the least deprived third

Four stages of exclusion from out-of-school suspension to school-related arrest, districts ranked on deprivation, 2020-21 school year

#### Table 2.1. Relative to the least deprived third

Each stage was more common in the most deprived third of districts, from 1.5 times the rate in the least deprived third for law-enforcement referrals to 2.7 times for expulsions

Rate in each third of districts divided by the rate in the least deprived third, by stage of exclusion, 2020-21 school year

| conf_high | conf_low | n | ratio | stage | tercile |
| ---: | ---: | ---: | ---: | --- | --- |
| n/a | n/a | 3,393 | 1 | Out-of-school suspension | Least deprived third |
| 2.60 | 1.90 | 3,394 | 2.22 | Out-of-school suspension | Middle third |
| 3.03 | 2.23 | 3,394 | 2.60 | Out-of-school suspension | Most deprived third |
| n/a | n/a | 3,393 | 1 | Expulsion | Least deprived third |
| 3.15 | 1.62 | 3,394 | 2.24 | Expulsion | Middle third |
| 3.83 | 1.94 | 3,394 | 2.72 | Expulsion | Most deprived third |
| n/a | n/a | 3,393 | 1 | Referral to law enforcement | Least deprived third |
| 2.04 | 1.33 | 3,394 | 1.65 | Referral to law enforcement | Middle third |
| 1.91 | 1.15 | 3,394 | 1.46 | Referral to law enforcement | Most deprived third |
| n/a | n/a | 3,393 | 1 | School-related arrest | Least deprived third |
| 2.97 | 1.48 | 3,394 | 2.10 | School-related arrest | Middle third |
| 2.53 | 1.24 | 3,394 | 1.79 | School-related arrest | Most deprived third |

#### Table 2.2. Rates per 1,000 students

Suspension was the widest stage, at 16.3 per thousand students, against 1.73 for referrals to law enforcement, 0.74 for expulsions and 0.26 for arrests

Students per 1,000 enrolled at each stage, pooled over districts, by third of districts ranked on deprivation, 2020-21 school year

| Stage | Least deprived third | Middle third | Most deprived third | All districts |
| --- | ---: | ---: | ---: | ---: |
| Out-of-school suspension | 9.85 | 21.85 | 25.59 | 16.27 |
| Expulsion | 0.44 | 0.99 | 1.20 | 0.74 |
| Referral to law enforcement | 1.35 | 2.23 | 1.97 | 1.73 |
| School-related arrest | 0.18 | 0.37 | 0.32 | 0.26 |

Note: Thirds hold equal numbers of the 10,181 districts that report all four counts. Each count is students at that stage, summed over a district's schools, not students followed from one stage to the next. Expulsions are students expelled with or without educational services; the zero-tolerance expulsions are a subset of those and are not added.

Note: The Civil Rights Data Collection for the 2020-2021 school year was disrupted by remote and hybrid teaching, and districts report their own counts, so the levels describe an unusual year. Suppressed school cells add nothing to a district's sum.

Note: Intervals resample counties; the rates are full counts reported by districts and carry no sampling interval for a single district.

Source: Civil Rights Data Collection, 2020-2021 school year, and Area Deprivation Index, 2023; enrollment-weighted district rates, full counts reported by districts

Data: https://vishalsingh.org/zipcode/report/school-discipline-and-incarceration/figures/exclusion_funnel.json

### Table 3. The Black suspension rate was higher than the white rate in 20 of the 24 districts with the most Black students

Students suspended out of school at least once per 1,000 enrolled, white and Black students in the same district, the districts with the most Black students among those compared, 2020-21 school year

| black | district | enrolled_black | enrolled_white | white |
| ---: | --- | ---: | ---: | ---: |
| 20.20 | Gwinnett County, GA | 57324 | 34490 | 12.90 |
| 2.10 | DeKalb County, GA | 55373 | 10603 | 1.30 |
| 2 | Detroit Public Schools Community District, MI | 38939 | 1177 | 8.50 |
| 15.50 | Fulton County, GA | 38322 | 23278 | 7.40 |
| 1.70 | Atlanta Public Schools, GA | 36901 | 8293 | 0.600 |
| 0.500 | Clayton County, GA | 36290 | 906 | 1.10 |
| 20.30 | Cobb County, GA | 32741 | 37526 | 10.20 |
| 6 | Guilford County Schools, NC | 29379 | 20193 | 2.30 |
| 50.50 | East Baton Rouge Parish, LA | 28779 | 4625 | 20.30 |
| 0 | District of Columbia Public Schools, DC | 28480 | 8188 | 0.100 |
| 154.9 | Mobile County, AL | 26493 | 19746 | 62.50 |
| 8.30 | Columbus City School District, OH | 24713 | 10015 | 4 |
| 12.50 | Henry County, GA | 24436 | 9752 | 17.40 |
| 45.40 | Caddo Parish, LA | 22967 | 10152 | 16.40 |
| 3.90 | Cumberland County Schools, NC | 22600 | 13156 | 1.70 |
| 10.40 | Montgomery County, AL | 22309 | 2261 | 2.70 |
| 13.60 | Cincinnati Public Schools, OH | 22032 | 7849 | 3.70 |
| 27.20 | Cypress-Fairbanks ISD, TX | 22025 | 25629 | 6.80 |
| 43.30 | Richmond County, GA | 21883 | 4380 | 30.10 |
| 2.50 | Cleveland Municipal, OH | 21492 | 5006 | 1.40 |
| 13.70 | Fort Bend ISD, TX | 21133 | 11341 | 4.40 |
| 3.60 | Savannah-Chatham County, GA | 21072 | 8018 | 1.60 |
| 2.60 | Jackson Public School District, MS | 19542 | 260 | 0 |
| 16.30 | Birmingham City, AL | 19186 | 403 | 5 |

Note: The rates are full counts reported by the districts, so they carry no sampling interval; a rate in a district with few students of one group rests on few suspensions. The suspension rate is the number of students who received at least one out-of-school suspension, whether once or more than once, divided by the students enrolled.

Note: Districts are named as the Collection lists them. The Civil Rights Data Collection for the 2020-2021 school year was disrupted by remote and hybrid teaching, and districts report their own counts, so the levels describe an unusual year. Suppressed school cells add nothing to a district's sum.

Source: Civil Rights Data Collection, 2020-2021 school year; full counts reported by districts, crude rates

Data: https://vishalsingh.org/zipcode/report/school-discipline-and-incarceration/figures/suspension_black_white.json

### Table 4. Counties in the highest tenth of suspension rates were concentrated in 6 states, which hold 61.0% of them

Students suspended out of school at least once per 1,000 students enrolled, school districts assigned to the county that holds most of their population, 2020-21 school year

No inline rows: the values of Table 4 are in https://vishalsingh.org/zipcode/data/school-discipline-and-incarceration/suspension_map.json.

Note: A county is drawn when the districts assigned to it that report a count enroll at least 500 students and hold more than 75% of the students in all the districts assigned to it. Of the gray counties, 33 are served by a district that reports no suspension count (the counties of New York City are among them), 8 have districts that report no count holding a quarter or more of their students, 240 fall below the enrollment floor and 50 have no district with a count assigned to them.

Note: States where the districts in the analysis hold less than 75% of the enrolled students, among them Florida, Maryland and California, are left out of the ranking of state rates. The highest tenth of county rates starts at 48.0 per thousand students and holds 282 counties.

Note: The key stops at 61 per thousand students, the rounded 95th percentile of county rates, and 143 counties above it share the color at the top of the key.

Note: Charter, regional and single-grade-span districts that cannot be matched to a unified district are left out, so a county's rate covers the districts that can be placed. A district that spans counties counts in the county that holds most of its population. Connecticut counties are the eight legacy counties, Alaska is drawn, and Puerto Rico is not drawn.

Note: The Civil Rights Data Collection for the 2020-2021 school year was disrupted by remote and hybrid teaching, and districts report their own counts, so the levels describe an unusual year. Suppressed school cells add nothing to a district's sum. The Collection also records each school's type of instruction in a separate school-level file that this chapter does not use, so the map cannot separate discipline from how much teaching was in person.

Source: Civil Rights Data Collection, 2020-2021 school year, and Area Deprivation Index, 2023; enrollment-weighted district rates, full counts reported by districts

Data: https://vishalsingh.org/zipcode/report/school-discipline-and-incarceration/figures/suspension_map.json

### Table 5. Incarceration of children from low-income families rose with tract deprivation and was higher for Black than for white children in the same tracts

Percent incarcerated on April 1, 2010, among children from low-income families born 1978 to 1983, by the deprivation of the tract where they grew up, ranked on 2019-2023 data

#### Table 5.1. All children, by deprivation

Incarceration of children from low-income families rose with tract deprivation, from 1.1% in the least deprived tenth of tracts to 3.3% in the most deprived

Percent of children from low-income families incarcerated on April 1, 2010, by percentile of the tract's Area Deprivation Index national rank (built from 2019-2023 data), races pooled, children born 1978 to 1983

| conf_high | conf_low | estimate | n | percentile |
| ---: | ---: | ---: | ---: | ---: |
| 1.02 | 0.480 | 0.756 | 630 | 1 |
| 1.02 | 0.572 | 0.782 | 642 | 2 |
| 1.12 | 0.785 | 0.961 | 566 | 3 |
| 1.15 | 0.825 | 0.967 | 748 | 4 |
| 1.26 | 0.847 | 1.04 | 614 | 5 |
| 1.40 | 0.948 | 1.15 | 765 | 6 |
| 1.23 | 0.845 | 1.03 | 587 | 7 |
| 1.43 | 1.06 | 1.23 | 775 | 8 |
| 1.43 | 1.02 | 1.21 | 623 | 9 |
| 1.44 | 1.05 | 1.24 | 796 | 10 |
| 1.55 | 1.19 | 1.36 | 565 | 11 |
| 1.43 | 1.07 | 1.24 | 823 | 12 |
| 1.44 | 1.11 | 1.29 | 647 | 13 |
| 1.43 | 1.10 | 1.26 | 824 | 14 |
| 1.54 | 1.18 | 1.36 | 649 | 15 |
| 1.69 | 1.37 | 1.51 | 846 | 16 |
| 1.66 | 1.26 | 1.47 | 644 | 17 |
| 1.63 | 1.37 | 1.50 | 832 | 18 |
| 1.77 | 1.43 | 1.60 | 661 | 19 |
| 1.73 | 1.45 | 1.59 | 815 | 20 |
| 1.71 | 1.40 | 1.56 | 618 | 21 |
| 1.86 | 1.47 | 1.65 | 820 | 22 |
| 1.85 | 1.46 | 1.65 | 597 | 23 |
| 1.86 | 1.58 | 1.71 | 825 | 24 |
| 1.86 | 1.53 | 1.69 | 643 | 25 |
| 1.82 | 1.52 | 1.67 | 825 | 26 |
| 1.93 | 1.58 | 1.75 | 651 | 27 |
| 1.88 | 1.58 | 1.73 | 778 | 28 |
| 1.78 | 1.49 | 1.63 | 593 | 29 |
| 1.97 | 1.62 | 1.80 | 878 | 30 |
| 1.92 | 1.60 | 1.75 | 647 | 31 |
| 2.08 | 1.76 | 1.91 | 790 | 32 |
| 1.96 | 1.64 | 1.79 | 612 | 33 |
| 2.05 | 1.77 | 1.91 | 781 | 34 |
| 2.16 | 1.81 | 1.98 | 633 | 35 |
| 1.90 | 1.65 | 1.78 | 805 | 36 |
| 2.19 | 1.78 | 1.99 | 665 | 37 |
| 2.13 | 1.80 | 1.97 | 809 | 38 |
| 2.08 | 1.78 | 1.93 | 591 | 39 |
| 2.08 | 1.75 | 1.91 | 753 | 40 |
| 2.19 | 1.85 | 2.02 | 662 | 41 |
| 2.20 | 1.89 | 2.05 | 840 | 42 |
| 2.08 | 1.74 | 1.91 | 619 | 43 |
| 2.14 | 1.83 | 1.99 | 792 | 44 |
| 2.31 | 1.97 | 2.14 | 638 | 45 |
| 2.26 | 1.94 | 2.10 | 809 | 46 |
| 2.23 | 1.93 | 2.08 | 622 | 47 |
| 2.39 | 2.05 | 2.21 | 837 | 48 |
| 2.38 | 2.05 | 2.21 | 605 | 49 |
| 2.32 | 1.98 | 2.14 | 853 | 50 |
| 2.32 | 2.00 | 2.16 | 658 | 51 |
| 2.42 | 2.13 | 2.28 | 819 | 52 |
| 2.26 | 1.97 | 2.11 | 687 | 53 |
| 2.29 | 2.00 | 2.14 | 828 | 54 |
| 2.24 | 1.98 | 2.11 | 647 | 55 |
| 2.44 | 2.14 | 2.29 | 831 | 56 |
| 2.38 | 2.08 | 2.23 | 638 | 57 |
| 2.41 | 2.14 | 2.27 | 842 | 58 |
| 2.44 | 2.17 | 2.31 | 724 | 59 |
| 2.61 | 2.32 | 2.46 | 822 | 60 |
| 2.60 | 2.31 | 2.45 | 673 | 61 |
| 2.51 | 2.21 | 2.35 | 853 | 62 |
| 2.45 | 2.14 | 2.31 | 638 | 63 |
| 2.56 | 2.27 | 2.41 | 807 | 64 |
| 2.45 | 2.17 | 2.31 | 708 | 65 |
| 2.43 | 2.15 | 2.29 | 847 | 66 |
| 2.53 | 2.24 | 2.39 | 694 | 67 |
| 2.61 | 2.26 | 2.44 | 859 | 68 |
| 2.44 | 2.12 | 2.28 | 665 | 69 |
| 2.67 | 2.36 | 2.52 | 824 | 70 |
| 2.59 | 2.28 | 2.44 | 678 | 71 |
| 2.60 | 2.30 | 2.45 | 837 | 72 |
| 2.58 | 2.25 | 2.42 | 667 | 73 |
| 2.60 | 2.26 | 2.43 | 832 | 74 |
| 2.65 | 2.32 | 2.49 | 725 | 75 |
| 2.67 | 2.36 | 2.51 | 787 | 76 |
| 2.70 | 2.38 | 2.54 | 650 | 77 |
| 2.66 | 2.37 | 2.51 | 851 | 78 |
| 2.65 | 2.34 | 2.50 | 709 | 79 |
| 2.62 | 2.32 | 2.46 | 848 | 80 |
| 2.70 | 2.35 | 2.53 | 642 | 81 |
| 2.88 | 2.56 | 2.73 | 833 | 82 |
| 2.70 | 2.38 | 2.54 | 646 | 83 |
| 2.88 | 2.52 | 2.70 | 766 | 84 |
| 2.67 | 2.30 | 2.48 | 643 | 85 |
| 2.86 | 2.52 | 2.69 | 835 | 86 |
| 2.83 | 2.46 | 2.65 | 611 | 87 |
| 2.86 | 2.53 | 2.69 | 730 | 88 |
| 2.91 | 2.52 | 2.72 | 649 | 89 |
| 2.93 | 2.54 | 2.73 | 782 | 90 |
| 3.11 | 2.68 | 2.89 | 579 | 91 |
| 3.17 | 2.76 | 2.96 | 747 | 92 |
| 3.26 | 2.74 | 3.00 | 576 | 93 |
| 3.49 | 2.88 | 3.20 | 694 | 94 |
| 3.49 | 3.00 | 3.24 | 528 | 95 |
| 3.75 | 3.24 | 3.49 | 648 | 96 |
| 3.89 | 3.29 | 3.59 | 525 | 97 |
| 4.15 | 3.67 | 3.91 | 603 | 98 |
| 4.44 | 3.90 | 4.18 | 547 | 99 |
| 4.51 | 3.96 | 4.23 | 603 | 100 |

#### Table 5.2. Black and white children in the same tracts

Black children from low-income families were incarcerated more often than white children from the same tracts at every level of tract deprivation

Percent of children from low-income families incarcerated on April 1, 2010, by tenth of the tract's Area Deprivation Index national rank (built from 2019-2023 data), Black and white children in the same tracts, children born 1978 to 1983

| conf_high | conf_low | decile | estimate | group | n |
| ---: | ---: | ---: | ---: | --- | ---: |
| 4.17 | 3.40 | 1 | 3.80 | Black children | 2,742 |
| 4.80 | 4.13 | 2 | 4.46 | Black children | 2,796 |
| 5.08 | 4.51 | 3 | 4.79 | Black children | 2,767 |
| 5.30 | 4.75 | 4 | 5.02 | Black children | 2,791 |
| 5.61 | 5.08 | 5 | 5.33 | Black children | 2,756 |
| 5.76 | 5.31 | 6 | 5.52 | Black children | 2,762 |
| 5.74 | 5.32 | 7 | 5.52 | Black children | 2,761 |
| 6.03 | 5.55 | 8 | 5.79 | Black children | 2,776 |
| 6.10 | 5.59 | 9 | 5.84 | Black children | 2,770 |
| 6.68 | 6.06 | 10 | 6.37 | Black children | 2,769 |
| 1.41 | 1.09 | 1 | 1.23 | White children | 2,742 |
| 1.61 | 1.33 | 2 | 1.47 | White children | 2,796 |
| 1.75 | 1.52 | 3 | 1.64 | White children | 2,767 |
| 2.03 | 1.80 | 4 | 1.92 | White children | 2,791 |
| 2.23 | 1.87 | 5 | 2.05 | White children | 2,756 |
| 2.29 | 1.97 | 6 | 2.13 | White children | 2,762 |
| 2.39 | 2.07 | 7 | 2.23 | White children | 2,761 |
| 2.42 | 2.10 | 8 | 2.25 | White children | 2,776 |
| 2.45 | 2.17 | 9 | 2.32 | White children | 2,770 |
| 2.81 | 2.40 | 10 | 2.59 | White children | 2,769 |

Note: Children were born in 1978 to 1983 and the incarceration share is for April 1, 2010. The Opportunity Atlas adds noise to its estimates to protect privacy and the staged values are never clipped, so some tract values are below zero. Genders are pooled.

Note: Each tract is weighted by its 2010 census population, the nearest stand-in for the number of children because the Atlas file has no child counts. Counting each tract once gives similar results.

Note: The Area Deprivation Index is defined on 2020 tracts. A 2010 tract takes the land-area-weighted mean rank of the 2020 tracts that overlap it, which is an approximation where boundaries changed. Tenths hold equal numbers of tracts.

Note: Tract estimates rest on a modest number of children. A tract is in the second tab only if both the Black and the white estimate exist, so these are tracts with sizeable populations of both groups and not a sample of all tracts.

Note: Across all the tracts in the second tab the incarceration share is 5.2% for Black children and 1.9% for white children.

Source: Opportunity Atlas, 1978-1983 birth cohorts, 2010 tracts, and Area Deprivation Index, 2023; population-weighted (2010 tract population), modeled estimates

Data: https://vishalsingh.org/zipcode/report/school-discipline-and-incarceration/figures/incarceration_gradient.json

### Table 6. Incarceration of Black children from low-income families is steadily worse toward the more deprived end under the ADI and under each of the three other rankings

Percent incarcerated on April 1, 2010 by tenth of tracts ranked four ways, over the 25,827 tracts that have all four rankings

| bin | conf_high | conf_low | estimate | n_units | ranking |
| ---: | ---: | ---: | ---: | ---: | --- |
| 1 | 4.08 | 3.27 | 3.71 | 2,585 | Area Deprivation Index |
| 2 | 4.88 | 4.17 | 4.53 | 2,597 | Area Deprivation Index |
| 3 | 5.05 | 4.47 | 4.76 | 2,568 | Area Deprivation Index |
| 4 | 5.28 | 4.72 | 4.99 | 2,571 | Area Deprivation Index |
| 5 | 5.61 | 5.09 | 5.34 | 2,592 | Area Deprivation Index |
| 6 | 5.71 | 5.27 | 5.49 | 2,583 | Area Deprivation Index |
| 7 | 5.80 | 5.34 | 5.56 | 2,541 | Area Deprivation Index |
| 8 | 6.03 | 5.51 | 5.76 | 2,624 | Area Deprivation Index |
| 9 | 6.15 | 5.62 | 5.88 | 2,599 | Area Deprivation Index |
| 10 | 6.66 | 6.04 | 6.35 | 2,567 | Area Deprivation Index |
| 1 | 4.24 | 3.61 | 3.92 | 2,582 | Standardized index |
| 2 | 4.59 | 4.03 | 4.31 | 2,583 | Standardized index |
| 3 | 5.33 | 4.80 | 5.07 | 2,583 | Standardized index |
| 4 | 5.31 | 4.82 | 5.06 | 2,582 | Standardized index |
| 5 | 5.41 | 4.94 | 5.18 | 2,583 | Standardized index |
| 6 | 5.78 | 5.27 | 5.52 | 2,583 | Standardized index |
| 7 | 5.77 | 5.30 | 5.54 | 2,582 | Standardized index |
| 8 | 5.77 | 5.33 | 5.54 | 2,583 | Standardized index |
| 9 | 6.11 | 5.55 | 5.83 | 2,583 | Standardized index |
| 10 | 6.73 | 6.13 | 6.43 | 2,583 | Standardized index |
| 1 | 4.11 | 3.48 | 3.80 | 2,582 | Median household income |
| 2 | 4.89 | 4.33 | 4.60 | 2,583 | Median household income |
| 3 | 5.03 | 4.49 | 4.76 | 2,583 | Median household income |
| 4 | 5.38 | 4.84 | 5.12 | 2,582 | Median household income |
| 5 | 5.55 | 5.10 | 5.32 | 2,584 | Median household income |
| 6 | 5.63 | 5.18 | 5.39 | 2,583 | Median household income |
| 7 | 6.00 | 5.47 | 5.74 | 2,581 | Median household income |
| 8 | 6.01 | 5.52 | 5.76 | 2,583 | Median household income |
| 9 | 6.05 | 5.55 | 5.80 | 2,583 | Median household income |
| 10 | 6.46 | 5.90 | 6.17 | 2,583 | Median household income |
| 1 | 4.47 | 3.73 | 4.10 | 2,582 | Bachelor's degree share |
| 2 | 4.80 | 4.20 | 4.50 | 2,583 | Bachelor's degree share |
| 3 | 5.09 | 4.55 | 4.81 | 2,583 | Bachelor's degree share |
| 4 | 5.35 | 4.83 | 5.08 | 2,583 | Bachelor's degree share |
| 5 | 5.50 | 4.99 | 5.24 | 2,582 | Bachelor's degree share |
| 6 | 5.57 | 5.02 | 5.30 | 2,583 | Bachelor's degree share |
| 7 | 5.65 | 5.16 | 5.40 | 2,581 | Bachelor's degree share |
| 8 | 5.78 | 5.30 | 5.54 | 2,584 | Bachelor's degree share |
| 9 | 6.05 | 5.59 | 5.81 | 2,583 | Bachelor's degree share |
| 10 | 6.45 | 5.86 | 6.16 | 2,583 | Bachelor's degree share |

Note: Least and most deprived tenth: ADI 3.7% and 6.3%, rank correlation +1.00; standardized index 3.9% and 6.4%, +0.98; median household income 3.8% and 6.2%, +1.00; bachelor's degree share 4.1% and 6.2%, +1.00.

Note: The standardized index applies the ADI's coefficients to its measures after standardizing each one, so home value and income carry less weight than in the ADI. Income and degree share are single census measures, each ranked from the highest value to the lowest.

Source: Opportunity Atlas, 1978-1983 birth cohorts, 2010 tracts, and Area Deprivation Index, 2023; population-weighted (2010 tract population), modeled estimates; ADI 2023 and ACS 2019-2023 tract tables (B19013, B15003 and the 15 rebuilt ADI measures)

Data: https://vishalsingh.org/zipcode/report/school-discipline-and-incarceration/figures/alt_rankings.json

## All estimates

Every fact in facts.json (109), sorted by id.

| Fact id | Value | Unit | Estimand | Population | Period | Interval | n | Weights | Source | Note |
| --- | ---: | --- | --- | --- | --- | --- | ---: | --- | --- | --- |
| alt_adi_least | 3.7% | Percent incarcerated on April 1, 2010 | population-weighted mean of the tract incarceration share of Black children from low-income families (2010 tract population as the weight), over the tracts where Black and white children are both estimated; 95% intervals from a bootstrap over counties | the least deprived tenth of tracts ranked by Area Deprivation Index, among the tracts with all four rankings | children born 1978 to 1983, incarcerated on April 1, 2010 | 95% CI 3.3% to 4.1% | 2,585 tracts | 2010 census tract population | Opportunity Atlas |   |
| alt_adi_most | 6.3% | Percent incarcerated on April 1, 2010 | population-weighted mean of the tract incarceration share of Black children from low-income families (2010 tract population as the weight), over the tracts where Black and white children are both estimated; 95% intervals from a bootstrap over counties | the most deprived tenth of tracts ranked by Area Deprivation Index, among the tracts with all four rankings | children born 1978 to 1983, incarcerated on April 1, 2010 | 95% CI 6.0% to 6.7% | 2,567 tracts | 2010 census tract population | Opportunity Atlas |   |
| alt_adi_rho | +1.00 | rank correlation | Spearman rank correlation between bin order and bin mean, each bin counted once | the bins of tracts ranked by Area Deprivation Index | children born 1978 to 1983, incarcerated on April 1, 2010 |   |   | each bin counted once | Opportunity Atlas |   |
| alt_degree_least | 4.1% | Percent incarcerated on April 1, 2010 | population-weighted mean of the tract incarceration share of Black children from low-income families (2010 tract population as the weight), over the tracts where Black and white children are both estimated; 95% intervals from a bootstrap over counties | the least deprived tenth of tracts ranked by Bachelor's degree share, among the tracts with all four rankings | children born 1978 to 1983, incarcerated on April 1, 2010 | 95% CI 3.7% to 4.5% | 2,582 tracts | 2010 census tract population | Opportunity Atlas |   |
| alt_degree_most | 6.2% | Percent incarcerated on April 1, 2010 | population-weighted mean of the tract incarceration share of Black children from low-income families (2010 tract population as the weight), over the tracts where Black and white children are both estimated; 95% intervals from a bootstrap over counties | the most deprived tenth of tracts ranked by Bachelor's degree share, among the tracts with all four rankings | children born 1978 to 1983, incarcerated on April 1, 2010 | 95% CI 5.9% to 6.5% | 2,583 tracts | 2010 census tract population | Opportunity Atlas |   |
| alt_degree_rho | +1.00 | rank correlation | Spearman rank correlation between bin order and bin mean, each bin counted once | the bins of tracts ranked by Bachelor's degree share | children born 1978 to 1983, incarcerated on April 1, 2010 |   |   | each bin counted once | Opportunity Atlas |   |
| alt_income_least | 3.8% | Percent incarcerated on April 1, 2010 | population-weighted mean of the tract incarceration share of Black children from low-income families (2010 tract population as the weight), over the tracts where Black and white children are both estimated; 95% intervals from a bootstrap over counties | the least deprived tenth of tracts ranked by Median household income, among the tracts with all four rankings | children born 1978 to 1983, incarcerated on April 1, 2010 | 95% CI 3.5% to 4.1% | 2,582 tracts | 2010 census tract population | Opportunity Atlas |   |
| alt_income_most | 6.2% | Percent incarcerated on April 1, 2010 | population-weighted mean of the tract incarceration share of Black children from low-income families (2010 tract population as the weight), over the tracts where Black and white children are both estimated; 95% intervals from a bootstrap over counties | the most deprived tenth of tracts ranked by Median household income, among the tracts with all four rankings | children born 1978 to 1983, incarcerated on April 1, 2010 | 95% CI 5.9% to 6.5% | 2,583 tracts | 2010 census tract population | Opportunity Atlas |   |
| alt_income_rho | +1.00 | rank correlation | Spearman rank correlation between bin order and bin mean, each bin counted once | the bins of tracts ranked by Median household income | children born 1978 to 1983, incarcerated on April 1, 2010 |   |   | each bin counted once | Opportunity Atlas |   |
| alt_n_places | 25,827 | tracts | count of tracts in the common sample of the ranking comparison | tracts with the outcome and all four rankings | children born 1978 to 1983, incarcerated on April 1, 2010 |   |   | none (full count) | Opportunity Atlas |   |
| alt_rho_cut | 0.9 | absolute rank correlation | threshold between a steady and an uneven gradient | the rule for changed gradients (docs/RANKING_SWITCH_RULE.md) | children born 1978 to 1983, incarcerated on April 1, 2010 |   |   | none (a fixed rule) | Opportunity Atlas |   |
| alt_rho_margin | 0.1 | absolute rank correlation | smallest move in the rank correlation that counts as a change of shape | the rule for changed gradients (docs/RANKING_SWITCH_RULE.md) | children born 1978 to 1983, incarcerated on April 1, 2010 |   |   | none (a fixed rule) | Opportunity Atlas |   |
| alt_std_least | 3.9% | Percent incarcerated on April 1, 2010 | population-weighted mean of the tract incarceration share of Black children from low-income families (2010 tract population as the weight), over the tracts where Black and white children are both estimated; 95% intervals from a bootstrap over counties | the least deprived tenth of tracts ranked by Standardized index, among the tracts with all four rankings | children born 1978 to 1983, incarcerated on April 1, 2010 | 95% CI 3.6% to 4.2% | 2,582 tracts | 2010 census tract population | Opportunity Atlas |   |
| alt_std_most | 6.4% | Percent incarcerated on April 1, 2010 | population-weighted mean of the tract incarceration share of Black children from low-income families (2010 tract population as the weight), over the tracts where Black and white children are both estimated; 95% intervals from a bootstrap over counties | the most deprived tenth of tracts ranked by Standardized index, among the tracts with all four rankings | children born 1978 to 1983, incarcerated on April 1, 2010 | 95% CI 6.1% to 6.7% | 2,583 tracts | 2010 census tract population | Opportunity Atlas |   |
| alt_std_rho | +0.98 | rank correlation | Spearman rank correlation between bin order and bin mean, each bin counted once | the bins of tracts ranked by Standardized index | children born 1978 to 1983, incarcerated on April 1, 2010 |   |   | each bin counted once | Opportunity Atlas |   |
| arr_mid_x | 2.1 | ratio of the school-related arrest rate in the middle third of districts to the rate in the least deprived third | ratio of pooled rates, bootstrap interval over counties | students enrolled in the 10181 districts that have a deprivation rank, at least 100 students and a reported count for all four stages | 2020-21 school year | 95% CI 1.5 to 3.0 | 10,181 districts | district enrollment | Civil Rights Data Collection |   |
| arr_terc_x | 1.8 | ratio of the school-related arrest rate in the most deprived third of districts to the rate in the least deprived third | ratio of pooled rates, bootstrap interval over counties | students enrolled in the 10181 districts that have a deprivation rank, at least 100 students and a reported count for all four stages | 2020-21 school year | 95% CI 1.2 to 2.5 | 10,181 districts | district enrollment | Civil Rights Data Collection |   |
| bw_black_cover | 71.3% | percent of Black students enrolled | share of all Black enrollment that sits in the districts of the comparison | Black students in the reporting districts | 2020-21 school year |   |   | none (full count) | Civil Rights Data Collection |   |
| bw_equal_u | 14.7% | percent of districts in which the two rates are equal | share of districts, bootstrap interval over states | districts in the Black-white comparison | 2020-21 school year | 95% CI 9.7% to 20.7% | 5,156 districts | none (each district counts once) | Civil Rights Data Collection |   |
| bw_floor | 50 | students enrolled in each group | minimum Black and minimum white enrollment (a rule, not an estimate) | districts in the comparison | 2020-21 school year |   |   | none (inclusion rule) | Civil Rights Data Collection |   |
| bw_n | 5,156 | school districts | count of districts with at least 50 Black and as many white students and a reported count for each | school districts (local education agencies) that reported to the Civil Rights Data Collection | 2020-21 school year |   |   | none (full count) | Civil Rights Data Collection |   |
| bw_students | 33.3 million | students enrolled | sum of total enrollment in the districts of the comparison | all students in the districts of the comparison | 2020-21 school year |   |   | none (full count) | Civil Rights Data Collection |   |
| bw_white_higher_u | 19.9% | percent of districts in which the white rate is higher than the Black rate | share of districts, bootstrap interval over states | districts in the Black-white comparison | 2020-21 school year | 95% CI 17.3% to 22.4% | 5,156 districts | none (each district counts once) | Civil Rights Data Collection |   |
| cloud_above | 20 | school districts | districts with a rate above the top of the scatter's vertical axis, which are in every estimate but not drawn | districts in the dose | 2020-21 school year |   |   | none (full count) | Civil Rights Data Collection |   |
| cov_floor | 75% | percent of enrolled students | minimum share of students in the districts that report a count (a rule, not an estimate) | counties drawn and states ranked on the map | 2020-21 school year |   |   | none (inclusion rule) | Civil Rights Data Collection |   |
| exp_terc_x | 2.7 | ratio of the expulsion rate in the most deprived third of districts to the rate in the least deprived third | ratio of pooled rates, bootstrap interval over counties | students enrolled in the 10181 districts that have a deprivation rank, at least 100 students and a reported count for all four stages | 2020-21 school year | 95% CI 1.9 to 3.8 | 10,181 districts | district enrollment | Civil Rights Data Collection |   |
| fn_arr | 0.26 | students per 1,000 enrolled: school-related arrest | enrollment-weighted mean of district rates, bootstrap interval over counties | students enrolled in the 10181 districts that have a deprivation rank, at least 100 students and a reported count for all four stages | 2020-21 school year | 95% CI 0.22 to 0.30 | 10,181 districts | district enrollment | Civil Rights Data Collection |   |
| fn_exp | 0.74 | students per 1,000 enrolled: expulsion | enrollment-weighted mean of district rates, bootstrap interval over counties | students enrolled in the 10181 districts that have a deprivation rank, at least 100 students and a reported count for all four stages | 2020-21 school year | 95% CI 0.65 to 0.84 | 10,181 districts | district enrollment | Civil Rights Data Collection |   |
| fn_n | 10,181 | school districts | count of districts in the funnel | districts in the dose with all four counts reported | 2020-21 school year |   |   | none (full count) | Civil Rights Data Collection |   |
| fn_oss | 16.3 | students per 1,000 enrolled: out-of-school suspension | enrollment-weighted mean of district rates, bootstrap interval over counties | students enrolled in the 10181 districts that have a deprivation rank, at least 100 students and a reported count for all four stages | 2020-21 school year | 95% CI 15.0 to 17.5 | 10,181 districts | district enrollment | Civil Rights Data Collection |   |
| fn_ref | 1.73 | students per 1,000 enrolled: referral to law enforcement | enrollment-weighted mean of district rates, bootstrap interval over counties | students enrolled in the 10181 districts that have a deprivation rank, at least 100 students and a reported count for all four stages | 2020-21 school year | 95% CI 1.55 to 1.90 | 10,181 districts | district enrollment | Civil Rights Data Collection |   |
| fn_zero_arr | 88.1% | percent of districts reporting a count of zero for school-related arrest | share of districts (each counted once) whose count is zero | districts in the funnel | 2020-21 school year |   | 10,181 districts | none (each district counts once) | Civil Rights Data Collection |   |
| fn_zero_exp | 75.5% | percent of districts reporting a count of zero for expulsion | share of districts (each counted once) whose count is zero | districts in the funnel | 2020-21 school year |   | 10,181 districts | none (each district counts once) | Civil Rights Data Collection |   |
| fn_zero_ref | 61.4% | percent of districts reporting a count of zero for referral to law enforcement | share of districts (each counted once) whose count is zero | districts in the funnel | 2020-21 school year |   | 10,181 districts | none (each district counts once) | Civil Rights Data Collection |   |
| inc_black | 5.2% | percent of children incarcerated on April 1, 2010 | population-weighted mean over tracts (2010 tract population as the weight), bootstrap interval over counties | Black children raised in low-income families (parents at the 25th percentile of the national income distribution), in tracts where the Atlas also estimates the other group | children born 1978 to 1983, incarcerated on April 1, 2010 | 95% CI 5.0% to 5.3% | 27,795 tracts | 2010 census tract population | Opportunity Atlas | Counting each tract once gives a similar mean. |
| inc_black_d1 | 3.8% | percent of children incarcerated on April 1, 2010 | population-weighted mean over tracts (2010 tract population as the weight), bootstrap interval over counties; tenths hold equal numbers of tracts | Black children raised in low-income families (parents at the 25th percentile of the national income distribution), in the least deprived tenth of tracts that have both estimates and a deprivation rank | children born 1978 to 1983, incarcerated on April 1, 2010 | 95% CI 3.4% to 4.2% | 2,742 tracts | 2010 census tract population | Opportunity Atlas |   |
| inc_black_d10 | 6.4% | percent of children incarcerated on April 1, 2010 | population-weighted mean over tracts (2010 tract population as the weight), bootstrap interval over counties; tenths hold equal numbers of tracts | Black children raised in low-income families (parents at the 25th percentile of the national income distribution), in the most deprived tenth of tracts that have both estimates and a deprivation rank | children born 1978 to 1983, incarcerated on April 1, 2010 | 95% CI 6.1% to 6.7% | 2,769 tracts | 2010 census tract population | Opportunity Atlas |   |
| inc_cross_diff | 1.2 | percentage points of incarceration: Black children in the least deprived tenth of tracts minus white children in the most deprived tenth | difference of two population-weighted tenth means, bootstrap interval over counties | children raised in low-income families (parents at the 25th percentile of the national income distribution), in tracts where the Atlas estimates both Black and white children and a deprivation rank | children born 1978 to 1983, incarcerated on April 1, 2010 | 95% CI 0.7 to 1.6 | 27,690 tracts | 2010 census tract population | Opportunity Atlas |   |
| inc_d1 | 1.1% | percent of children incarcerated on April 1, 2010 | population-weighted mean over tracts (2010 tract population as the weight), bootstrap interval over counties; tenths hold equal numbers of tracts | children raised in low-income families (parents at the 25th percentile of the national income distribution), races pooled, in tracts that have an incarceration estimate and a deprivation rank, in the least deprived tenth of tracts | children born 1978 to 1983, incarcerated on April 1, 2010 | 95% CI 1.0% to 1.2% | 7,147 tracts | 2010 census tract population | Opportunity Atlas |   |
| inc_d10 | 3.3% | percent of children incarcerated on April 1, 2010 | population-weighted mean over tracts (2010 tract population as the weight), bootstrap interval over counties; tenths hold equal numbers of tracts | children raised in low-income families (parents at the 25th percentile of the national income distribution), races pooled, in tracts that have an incarceration estimate and a deprivation rank, in the most deprived tenth of tracts | children born 1978 to 1983, incarcerated on April 1, 2010 | 95% CI 3.1% to 3.5% | 7,233 tracts | 2010 census tract population | Opportunity Atlas |   |
| inc_gap | 3.2 | percentage points of incarceration, Black minus white | mean over tracts of the Black minus white difference, population-weighted mean over tracts (2010 tract population as the weight), bootstrap interval over counties | children raised in low-income families (parents at the 25th percentile of the national income distribution), in tracts where the Atlas estimates both Black and white children | children born 1978 to 1983, incarcerated on April 1, 2010 | 95% CI 3.1 to 3.4 | 27,795 tracts | 2010 census tract population | Opportunity Atlas |   |
| inc_gap_d1 | 2.6 | percentage points of incarceration, Black minus white | mean over tracts of the Black minus white difference, population-weighted mean over tracts (2010 tract population as the weight), bootstrap interval over counties | children raised in low-income families (parents at the 25th percentile of the national income distribution), in the least deprived tenth of tracts that have both estimates and a deprivation rank | children born 1978 to 1983, incarcerated on April 1, 2010 | 95% CI 2.2 to 2.9 | 2,742 tracts | 2010 census tract population | Opportunity Atlas |   |
| inc_gap_d10 | 3.8 | percentage points of incarceration, Black minus white | mean over tracts of the Black minus white difference, population-weighted mean over tracts (2010 tract population as the weight), bootstrap interval over counties | children raised in low-income families (parents at the 25th percentile of the national income distribution), in the most deprived tenth of tracts that have both estimates and a deprivation rank | children born 1978 to 1983, incarcerated on April 1, 2010 | 95% CI 3.4 to 4.1 | 2,769 tracts | 2010 census tract population | Opportunity Atlas |   |
| inc_mean | 2.1% | percent of children incarcerated on April 1, 2010 | population-weighted mean over tracts (2010 tract population as the weight), bootstrap interval over counties | children raised in low-income families (parents at the 25th percentile of the national income distribution), races pooled, in tracts that have an incarceration estimate and a deprivation rank | children born 1978 to 1983, incarcerated on April 1, 2010 | 95% CI 2.0% to 2.2% | 71,478 tracts | 2010 census tract population | Opportunity Atlas | Counting each tract once gives a similar mean. |
| inc_n_shared | 27,795 | tracts | count of tracts | 2010 census tracts with an incarceration estimate for Black and for white children and a positive 2010 population | children born 1978 to 1983, incarcerated on April 1, 2010 |   |   | none (full count) | Opportunity Atlas |   |
| inc_n_tracts | 71,478 | tracts | count of tracts | 2010 census tracts with an incarceration estimate, a deprivation rank and a positive 2010 population | children born 1978 to 1983, incarcerated on April 1, 2010 |   |   | none (full count) | Opportunity Atlas |   |
| inc_r_adi | +0.28 | correlation coefficient | population-weighted correlation of the tract incarceration share with the tract deprivation rank, bootstrap interval over counties | children raised in low-income families (parents at the 25th percentile of the national income distribution), races pooled, in tracts that have an incarceration estimate and a deprivation rank | children born 1978 to 1983, incarcerated on April 1, 2010 | 95% CI 0.25 to 0.30 | 71,478 tracts | 2010 census tract population | Opportunity Atlas |   |
| inc_ratio | 2.7 | ratio of the Black to the white incarceration share | ratio of the population-weighted means over tracts, bootstrap interval over counties | children raised in low-income families (parents at the 25th percentile of the national income distribution), in tracts where the Atlas estimates both Black and white children | children born 1978 to 1983, incarcerated on April 1, 2010 | 95% CI 2.5 to 2.8 | 27,795 tracts | 2010 census tract population | Opportunity Atlas |   |
| inc_ratio_d1 | 3.1 | ratio of the Black to the white incarceration share | ratio of the population-weighted tenth means, bootstrap interval over counties | children raised in low-income families (parents at the 25th percentile of the national income distribution), in the least deprived tenth of tracts that have both estimates and a deprivation rank | children born 1978 to 1983, incarcerated on April 1, 2010 | 95% CI 2.6 to 3.5 | 2,742 tracts | 2010 census tract population | Opportunity Atlas |   |
| inc_ratio_d10 | 2.5 | ratio of the Black to the white incarceration share | ratio of the population-weighted tenth means, bootstrap interval over counties | children raised in low-income families (parents at the 25th percentile of the national income distribution), in the most deprived tenth of tracts that have both estimates and a deprivation rank | children born 1978 to 1983, incarcerated on April 1, 2010 | 95% CI 2.2 to 2.7 | 2,769 tracts | 2010 census tract population | Opportunity Atlas |   |
| inc_ratio_fall | 0.6 | ratio of the Black to the white incarceration share in the least deprived tenth of tracts minus the ratio in the most deprived tenth | difference of two ratios of population-weighted tenth means, bootstrap interval over counties | children raised in low-income families (parents at the 25th percentile of the national income distribution), in tracts where the Atlas estimates both Black and white children and a deprivation rank | children born 1978 to 1983, incarcerated on April 1, 2010 | 95% CI 0.1 to 1.1 | 27,690 tracts | 2010 census tract population | Opportunity Atlas |   |
| inc_share_bw | 73.8% | percent of tracts in which the Black incarceration share is higher than the white share | share of tracts (each counted once), bootstrap interval over counties | children raised in low-income families (parents at the 25th percentile of the national income distribution), in tracts where the Atlas estimates both Black and white children | children born 1978 to 1983, incarcerated on April 1, 2010 | 95% CI 72.9% to 74.7% | 27,795 tracts | none (each tract counts once) | Opportunity Atlas | Weighting tracts by 2010 population gives a similar share. Privacy noise pulls a share like this toward an even split. |
| inc_shared_counties | 1,468 | counties | count of counties | counties with at least one tract in the comparison | children born 1978 to 1983, incarcerated on April 1, 2010 |   |   | none (full count) | Opportunity Atlas |   |
| inc_shared_pop | 130.5 million | people | sum of 2010 census tract population | residents of the tracts in the same-tract comparison | 2010 census |   |   | none (full count) | Opportunity Atlas |   |
| inc_shared_pop_share | 42.3% | percent of the 2010 U.S. population | population of the same-tract comparison tracts as a share of the U.S. population | residents of the United States | 2010 census |   |   | none (full count) | Opportunity Atlas |   |
| inc_white | 1.9% | percent of children incarcerated on April 1, 2010 | population-weighted mean over tracts (2010 tract population as the weight), bootstrap interval over counties | white children raised in low-income families (parents at the 25th percentile of the national income distribution), in tracts where the Atlas also estimates the other group | children born 1978 to 1983, incarcerated on April 1, 2010 | 95% CI 1.8% to 2.0% | 27,795 tracts | 2010 census tract population | Opportunity Atlas | Counting each tract once gives a similar mean. |
| inc_white_d1 | 1.2% | percent of children incarcerated on April 1, 2010 | population-weighted mean over tracts (2010 tract population as the weight), bootstrap interval over counties; tenths hold equal numbers of tracts | white children raised in low-income families (parents at the 25th percentile of the national income distribution), in the least deprived tenth of tracts that have both estimates and a deprivation rank | children born 1978 to 1983, incarcerated on April 1, 2010 | 95% CI 1.1% to 1.4% | 2,742 tracts | 2010 census tract population | Opportunity Atlas |   |
| inc_white_d10 | 2.6% | percent of children incarcerated on April 1, 2010 | population-weighted mean over tracts (2010 tract population as the weight), bootstrap interval over counties; tenths hold equal numbers of tracts | white children raised in low-income families (parents at the 25th percentile of the national income distribution), in the most deprived tenth of tracts that have both estimates and a deprivation rank | children born 1978 to 1983, incarcerated on April 1, 2010 | 95% CI 2.4% to 2.8% | 2,769 tracts | 2010 census tract population | Opportunity Atlas |   |
| map_floor | 500 | students enrolled | minimum enrollment in a county's districts for the county to be drawn (a rule, not an estimate) | counties on the map | 2020-21 school year |   |   | none (inclusion rule) | Civil Rights Data Collection |   |
| map_key_top | 61 | students suspended per 1,000 enrolled | top of the map key, the 95th percentile of county rates rounded to a whole number; higher rates take the top color | counties drawn | 2020-21 school year |   |   | none (each county counts once) | Civil Rights Data Collection |   |
| map_n_above_key | 143 | counties | count of counties with a rate above the top of the key | counties drawn | 2020-21 school year |   |   | none (full count) | Civil Rights Data Collection |   |
| map_n_below_floor | 240 | counties | count of counties whose assigned districts enroll fewer students than the floor | counties and county equivalents on the map | 2020-21 school year |   |   | none (full count) | Civil Rights Data Collection |   |
| map_n_drawn | 2,812 | counties | count of counties with a rate | counties and county equivalents on the map | 2020-21 school year |   |   | none (full count) | Civil Rights Data Collection |   |
| map_n_no_district | 50 | counties | count of counties to which no district is assigned | counties and county equivalents on the map | 2020-21 school year |   |   | none (full count) | Civil Rights Data Collection |   |
| map_n_not_drawn | 331 | counties | count of counties with no rate | counties and county equivalents on the map | 2020-21 school year |   |   | none (full count) | Civil Rights Data Collection |   |
| map_n_partial | 8 | counties | count of counties whose assigned districts that report no count hold a quarter or more of the students | counties and county equivalents on the map | 2020-21 school year |   |   | none (full count) | Civil Rights Data Collection |   |
| map_n_unreported | 33 | counties | count of counties whose assigned districts all report no suspension count | counties and county equivalents on the map | 2020-21 school year |   |   | none (full count) | Civil Rights Data Collection |   |
| map_top_cut | 48.0 | students suspended out of school at least once per 1,000 students enrolled | 90th percentile of county rates; counties at or above it are the highest tenth | counties drawn | 2020-21 school year |   | 2,812 counties | none (each county counts once) | Civil Rights Data Collection |   |
| map_top_min | 15 | counties in the highest tenth | minimum number of counties in the highest tenth for a state to be named (a rule, not an estimate) | states named on the map | 2020-21 school year |   |   | none (inclusion rule) | Civil Rights Data Collection |   |
| map_top_n | 282 | counties | count of counties at or above the 90th percentile | counties in the highest tenth of county rates | 2020-21 school year |   |   | none (full count) | Civil Rights Data Collection |   |
| map_top_states_min_share | 25.3% | percent of a state's drawn counties that are in the highest tenth of rates | the smallest, over those states, of the share of the state's drawn counties in the highest tenth | drawn counties in the states that meet the rule | 2020-21 school year |   |   | none (each county counts once) | Civil Rights Data Collection |   |
| map_top_states_n | 6 | states | count of states that meet the rule | states with at least the minimum number of counties in the highest tenth | 2020-21 school year |   |   | none (full count) | Civil Rights Data Collection |   |
| map_top_states_share | 61.0% | percent of counties in the highest tenth of rates | share of those counties that lie in the states that meet the rule | counties in the highest tenth of county rates | 2020-21 school year |   | 282 counties | none (each county counts once) | Civil Rights Data Collection |   |
| n_big_blackhi | 20 | school districts | count of districts whose Black suspension rate is higher than the white rate | the 24 districts with the most Black students among those compared | 2020-21 school year |   |   | none (full count) | Civil Rights Data Collection |   |
| n_dumbbell | 24 | school districts | count of districts drawn (chosen by Black enrollment, not by their rates) | districts with the most Black students among those compared | 2020-21 school year |   |   | none (full count) | Civil Rights Data Collection |   |
| n_lea | 17,821 | school districts | count of reporting districts | school districts (local education agencies) that reported to the Civil Rights Data Collection | 2020-21 school year |   |   | none (full count) | Civil Rights Data Collection |   |
| n_states_ranked | 38 | states | count of states ranked on state rates (DC and Puerto Rico left out) | states where the districts in the analysis hold at least the coverage floor of enrolled students | 2020-21 school year |   |   | none (full count) | Civil Rights Data Collection |   |
| oss_black | 30.7 | students suspended out of school at least once per 1,000 students enrolled | Black students suspended over Black students enrolled, pooled over districts, bootstrap interval over states | Black students enrolled in the 5156 districts with at least 50 Black and as many white students and a reported suspension count for each group | 2020-21 school year | 95% CI 22.7 to 39.1 | 5,156 districts | Black enrollment | Civil Rights Data Collection |   |
| oss_cover_districts | 57.6% | percent of reporting districts | share of reporting districts that enter the dose | school districts (local education agencies) that reported to the Civil Rights Data Collection | 2020-21 school year |   |   | none (each district counts once) | Civil Rights Data Collection |   |
| oss_cover_enrollment | 78.4% | percent of enrolled students | share of all enrollment that sits in the districts of the dose | students enrolled in reporting districts | 2020-21 school year |   |   | enrollment (full count) | Civil Rights Data Collection |   |
| oss_d1 | 3.1 | students suspended out of school at least once per 1,000 students enrolled | enrollment-weighted mean of district rates (students suspended over students enrolled), bootstrap interval over counties | students enrolled in school districts that have a deprivation rank, at least 100 students and a reported suspension count in the least deprived tenth of districts by deprivation rank | 2020-21 school year | 95% CI 2.4 to 4.1 | 1,025 districts | district enrollment | Civil Rights Data Collection |   |
| oss_d10 | 23.6 | students suspended out of school at least once per 1,000 students enrolled | enrollment-weighted mean of district rates (students suspended over students enrolled), bootstrap interval over counties | students enrolled in school districts that have a deprivation rank, at least 100 students and a reported suspension count in the most deprived tenth of districts by deprivation rank | 2020-21 school year | 95% CI 20.2 to 27.3 | 1,022 districts | district enrollment | Civil Rights Data Collection |   |
| oss_d4 | 20.1 | students suspended out of school at least once per 1,000 students enrolled | enrollment-weighted mean of district rates (students suspended over students enrolled), bootstrap interval over counties | students enrolled in school districts that have a deprivation rank, at least 100 students and a reported suspension count in the fourth most deprived tenth of districts by deprivation rank | 2020-21 school year | 95% CI 17.3 to 23.1 | 1,025 districts | district enrollment | Civil Rights Data Collection |   |
| oss_enrollment | 38.7 million | students enrolled | sum of enrollment in the districts of the dose | school districts (local education agencies) that reported to the Civil Rights Data Collection in the dose | 2020-21 school year |   |   | none (full count) | Civil Rights Data Collection |   |
| oss_floor | 100 | students enrolled | minimum enrollment for a district to enter (a rule, not an estimate) | districts in the dose, funnel and map | 2020-21 school year |   |   | none (inclusion rule) | Civil Rights Data Collection |   |
| oss_fold | 7.6 | ratio of the suspension rate in the most deprived tenth of districts to the rate in the least deprived tenth | ratio of pooled rates, bootstrap interval over counties | students enrolled in school districts that have a deprivation rank, at least 100 students and a reported suspension count | 2020-21 school year | 95% CI 5.5 to 10.6 | 10,257 districts | district enrollment | Civil Rights Data Collection |   |
| oss_missing_n | 38 | school districts | count of districts left out of the dose, funnel and map for want of a count | districts that have a deprivation rank and at least 100 students but report no suspension count | 2020-21 school year |   |   | none (full count) | Civil Rights Data Collection |   |
| oss_missing_students | 6.2 million | students enrolled | sum of enrollment in those districts | students in the districts that report no suspension count | 2020-21 school year |   |   | none (full count) | Civil Rights Data Collection |   |
| oss_n | 10,257 | school districts | count of districts in the dose, funnel and map | districts that have a deprivation rank, at least 100 students and a reported suspension count | 2020-21 school year |   | 10,257 districts | none (full count) | Civil Rights Data Collection |   |
| oss_peak | 26.9 | students suspended out of school at least once per 1,000 students enrolled | enrollment-weighted mean of district rates (students suspended over students enrolled), bootstrap interval over counties | students enrolled in school districts that have a deprivation rank, at least 100 students and a reported suspension count in the tenth of districts with the highest mean rate | 2020-21 school year | 95% CI 24.1 to 29.7 | 1,026 districts | district enrollment | Civil Rights Data Collection |   |
| oss_peak_decile | 8 | decile of the district deprivation rank (1 = least deprived, 10 = most) | decile with the highest enrollment-weighted mean rate | districts in the dose | 2020-21 school year |   |   | district enrollment | Civil Rights Data Collection |   |
| oss_peak_vs_d10 | +3.2 | students suspended per 1,000 enrolled: the highest tenth's rate minus the most deprived tenth's rate | difference of two enrollment-weighted decile means, bootstrap interval over counties | students enrolled in school districts that have a deprivation rank, at least 100 students and a reported suspension count | 2020-21 school year | 95% CI −0.9 to 7.1 | 10,257 districts | district enrollment | Civil Rights Data Collection |   |
| oss_r | +0.42 | correlation coefficient | enrollment-weighted correlation of the district suspension rate with the district deprivation rank, bootstrap interval over counties | students enrolled in school districts that have a deprivation rank, at least 100 students and a reported suspension count | 2020-21 school year | 95% CI 0.37 to 0.46 | 10,257 districts | district enrollment | Civil Rights Data Collection |   |
| oss_r_state | +0.42 | correlation coefficient | the same correlation with a bootstrap interval that resamples whole states | students enrolled in school districts that have a deprivation rank, at least 100 students and a reported suspension count | 2020-21 school year | 95% CI 0.31 to 0.51 | 10,257 districts | district enrollment | Civil Rights Data Collection |   |
| oss_ratio | 2.2 | ratio of Black students suspended to the number expected if each district's Black students were suspended at its white rate | observed Black suspensions over expected suspensions at the district's white rate, pooled, bootstrap interval over states | Black students enrolled in the 5156 districts with at least 50 Black and as many white students and a reported suspension count for each group | 2020-21 school year | 95% CI 2.1 to 2.4 | 5,156 districts | Black enrollment | Civil Rights Data Collection |   |
| oss_share_unwt | 65.4% | percent of districts in which the Black rate is higher than the white rate | share of districts, bootstrap interval over states | districts in the Black-white comparison | 2020-21 school year | 95% CI 57.5% to 72.4% | 5,156 districts | none (each district counts once) | Civil Rights Data Collection |   |
| oss_share_wt | 84.0% | percent of districts in which the Black rate is higher than the white rate | share of districts, weighted by Black enrollment, bootstrap interval over states | districts in the Black-white comparison | 2020-21 school year | 95% CI 77.1% to 88.8% | 5,156 districts | Black enrollment | Civil Rights Data Collection |   |
| oss_slope_fe | 1.9 | students suspended per 1,000 enrolled, for each ten-point step in the district deprivation rank | enrollment-weighted regression of the district rate on the deprivation rank with state fixed effects, errors clustered by state | students enrolled in school districts that have a deprivation rank, at least 100 students and a reported suspension count | 2020-21 school year | 95% CI 1.3 to 2.6 | 10,257 districts | district enrollment | Civil Rights Data Collection |   |
| oss_students_total | 651,832 | students suspended out of school at least once, counted once for each school | sum of the schools' counts over all districts; a student suspended at two schools counts at each, and school cells coded as suppressed or not reported add nothing | students in the reporting districts | 2020-21 school year |   |   | none (full count) | Civil Rights Data Collection | Districts report their own counts. The sum counts a student once for each school where they were suspended. |
| oss_terc_x | 2.6 | ratio of the suspension rate in the most deprived third of districts to the rate in the least deprived third | ratio of pooled rates, bootstrap interval over counties | students enrolled in the 10181 districts that have a deprivation rank, at least 100 students and a reported count for all four stages | 2020-21 school year | 95% CI 2.2 to 3.0 | 10,181 districts | district enrollment | Civil Rights Data Collection |   |
| oss_white_same | 13.7 | students suspended out of school at least once per 1,000 students enrolled | mean of district white rates weighted by Black enrollment, bootstrap interval over states | white students, at the white rate of each district weighted by that district's Black enrollment | 2020-21 school year | 95% CI 10.1 to 17.0 | 5,156 districts | Black enrollment | Civil Rights Data Collection |   |
| ref_mid_x | 1.7 | ratio of the referral to law enforcement rate in the middle third of districts to the rate in the least deprived third | ratio of pooled rates, bootstrap interval over counties | students enrolled in the 10181 districts that have a deprivation rank, at least 100 students and a reported count for all four stages | 2020-21 school year | 95% CI 1.3 to 2.0 | 10,181 districts | district enrollment | Civil Rights Data Collection |   |
| ref_terc_x | 1.5 | ratio of the referral to law enforcement rate in the most deprived third of districts to the rate in the least deprived third | ratio of pooled rates, bootstrap interval over counties | students enrolled in the 10181 districts that have a deprivation rank, at least 100 students and a reported count for all four stages | 2020-21 school year | 95% CI 1.2 to 1.9 | 10,181 districts | district enrollment | Civil Rights Data Collection |   |
| state_high | 43.6 | students suspended out of school at least once per 1,000 students enrolled | pooled students suspended over students enrolled in the state's districts of the dose | students enrolled in the state with the highest rate among the ranked states, in districts of the dose | 2020-21 school year |   | 218 districts | district enrollment | Civil Rights Data Collection |   |
| state_high_cov | 93.2% | percent of the state's enrolled students in the districts of the dose | enrollment of the state's districts in the dose over enrollment of all its reporting districts | students enrolled in the state's reporting districts | 2020-21 school year |   |   | none (full count) | Civil Rights Data Collection |   |
| state_low | 3.1 | students suspended out of school at least once per 1,000 students enrolled | pooled students suspended over students enrolled in the state's districts of the dose | students enrolled in the state with the lowest rate among the ranked states, in districts of the dose | 2020-21 school year |   | 225 districts | district enrollment | Civil Rights Data Collection |   |
| state_low_cov | 96.4% | percent of the state's enrolled students in the districts of the dose | enrollment of the state's districts in the dose over enrollment of all its reporting districts | students enrolled in the state's reporting districts | 2020-21 school year |   |   | none (full count) | Civil Rights Data Collection |   |
| us_inc_black | 5.2% | percent of children incarcerated on April 1, 2010 | population-weighted mean over tracts (2010 tract population as the weight), bootstrap interval over counties; all ten tenths pooled | Black children raised in low-income families (parents at the 25th percentile of the national income distribution), in all the tracts that have both estimates and a deprivation rank | children born 1978 to 1983, incarcerated on April 1, 2010 | 95% CI 5.0% to 5.3% | 27,690 tracts | 2010 census tract population | Opportunity Atlas |   |
| us_inc_white | 1.9% | percent of children incarcerated on April 1, 2010 | population-weighted mean over tracts (2010 tract population as the weight), bootstrap interval over counties; all ten tenths pooled | white children raised in low-income families (parents at the 25th percentile of the national income distribution), in all the tracts that have both estimates and a deprivation rank | children born 1978 to 1983, incarcerated on April 1, 2010 | 95% CI 1.8% to 2.0% | 27,690 tracts | 2010 census tract population | Opportunity Atlas |   |

## Definitions

- Area Deprivation Index (ADI): A score that ranks neighborhoods by income, education, employment and housing conditions drawn from the American Community Survey. In this book a higher national percentile means a more deprived neighborhood.

## Methods

This chapter compares out-of-school suspension rates of school districts in the 2020 to 2021 school year across tenths of district deprivation, Black and white students in the same districts, and the share of children from low-income families incarcerated on April 1, 2010, across census tracts. Only districts that the book's geography file assigns a deprivation rank are analyzed, so charter, regional and single-grade-span reporting districts are left out. A student suspended at two schools counts at each, and in-school suspensions are not included. The deprivation index is built from 2019-2023 data, so it describes the tracts as they are now and not as they were when these children grew up in the 1980s and 1990s.

### Which districts and school year the discipline data cover

District data are the Civil Rights Data Collection for the 2020 to 2021 school year, school counts summed to the reporting district. School cells coded as suppressed or not reported add nothing to a district's sum.

Only districts that the book's geography file (dim_geo), which assigns each tract to one school district, knows get a rank, and charter, regional and single-grade-span reporting districts do not. A district enters the dose, the funnel and the map with at least 100 students enrolled and a reported suspension count.

The Black-white comparison uses every reporting district with at least 50 Black and as many white students and a reported count for each.

Explains Table 1, Table 2, Table 3 and Table 4.

### How suspensions, expulsions and referrals are counted

A suspension rate is the number of students who received at least one out-of-school suspension at a school, counting students with and without disabilities and counting a student once per school however many times they were suspended there, divided by the students enrolled, per 1,000; a student suspended at two schools counts at each, and in-school suspensions are not included.

Expulsions are students expelled with or without educational services (zero-tolerance expulsions are a subset of those and are not added; the staged file adds them, so the build reads the two services categories from the raw school file), and referrals to law enforcement and school-related arrests are counts of students at each stage, summed the same way.

Explains Table 1 and Table 2.

### How districts are ranked, grouped and mapped

A district's deprivation rank is the mean of the Area Deprivation Index national ranks (1 least deprived, 100 most) of the tracts assigned to it, weighted by 2019-2023 tract population.

Tenths and thirds hold equal numbers of districts, and a tenth's value is the enrollment-weighted mean of its district rates, which equals the pooled count of suspended students over pooled enrollment.

The county map assigns each district to the county that holds most of its population, pools suspended students and enrolled students over the districts of a county that report a count, and draws counties whose reporting districts enroll at least 500 students and hold more than 75% of the students in all the districts assigned to the county; state rates are ranked only for states where the districts in the analysis hold at least 75% of the students in all reporting districts.

Explains Table 1, Table 3 and Table 4.

### How tract incarceration shares are measured and ranked

Incarceration is the Opportunity Atlas share of children from low-income families (parents at the 25th percentile of the national income distribution), born 1978 to 1983, who were incarcerated on April 1, 2010, on native 2010 tracts, printed as a percent; genders are pooled and the noise added for privacy is never clipped.

A tract is in a comparison when the estimates exist and its 2010 census population is positive; each tract is weighted by that population.

A 2010 tract's deprivation rank is the land-area-weighted mean rank of the overlapping 2020 tracts that have one, when they cover at least half of its land area; the index is built from 2019-2023 data, so it describes the tracts as they are now and not as they were when these children grew up in the 1980s and 1990s.

Explains Table 5.

### How uncertainty is measured

Intervals for means, ratios and correlations are 95% percentile intervals from 2,000 draws of whole counties with replacement (seed 12345), where a district's county is the one that holds most of its population; the Black-white comparison, which includes districts that cannot be placed in a county, resamples states.

### Four rankings of the same places

The comparison follows the book's ranking switch rule, fixed on October 28, 2026 before any of these gradients was computed. The main gradient of the chapter is recomputed under four rankings of the same places, with the outcome, period, weights, bins and intervals unchanged: the published ADI; a standardized index that applies the ADI's coefficients to its measures after standardizing each one across tracts; median household income; and the share of adults 25 and older with a bachelor's degree or more. The three alternatives come from ACS tract estimates for 2019 to 2023, the period of the ADI.

All four rankings use the same places, the 25,827 tracts that have the outcome and all four rankings, so the ADI panel can differ slightly from the chapter's main figure. Tracts enter only if they have every input of the standardized index, which leaves out tracts without a median rent, home value or mortgage cost. Each 2010 tract takes the land-area-weighted mean of each ranking over the common-sample 2020 tracts that overlap it, when they cover at least half of its land area, the chapter's construction of its ADI, and tenths are tenths of the tracts where Black and white children are both estimated, as in the chapter's comparison by race.

A pattern counts as changed when the worse end switches sides, when the two ends differ under one ranking and not the other, or when the rank correlation of bin order and bin mean falls on the other side of 0.9 and moves by at least 0.1.

Explains Table 6.

## What these data cannot show

A suspension rate records what districts reported, not what children did. A higher rate in a deprived district, or for Black students, is consistent with more disruptive conduct, with harsher responses to the same conduct, with both, and with differences in how districts count, and these data cannot tell them apart. The counts are for a pandemic year, are self-reported, and omit the suppressed cells. The 38 districts with no count include some of the largest in the country, and charter, regional and single-grade-span districts have no deprivation rank, so the districts here are not all of the districts. In several states, among them Florida, Maryland and California, the districts in the analysis hold less than 75% of enrolled students, so no state rate is given for them. The Area Deprivation Index understates deprivation in high-cost cities (Chapter 1.1, What the deprivation index measures (https://vishalsingh.org/zipcode/anatomy-of-deprivation)), and a district's rank averages over tracts that differ a great deal. The map assigns each district to one county although some span several.

The incarceration part has its own limits. The Atlas measures incarceration on one day, so it counts who was incarcerated on that day and not who was ever incarcerated. It pools girls and boys, because race by gender is not in the staged data. Each tract estimate is noisy, the weights are tract population and not the number of children, and a tract with both a Black and a white estimate is one with enough children of both groups. A same-tract comparison removes differences between neighborhoods but not differences in family wealth, in the block, in schools or in how police and courts treat the children. Both parts describe places and groups, not individual children, and they describe different cohorts. No child here is followed from a suspension to a later incarceration, and the districts of 2020 to 2021 are not the tracts where the children of 1978 to 1983 grew up.

## References

1. Kind, Amy J. H.; Buckingham, William R. (2018). Making Neighborhood-Disadvantage Metrics Accessible: The Neighborhood Atlas. New England Journal of Medicine 378(26): 2456-2458. https://doi.org/10.1056/NEJMp1802313
2. Skiba, Russell J.; Michael, Robert S.; Nardo, Abra Carroll; Peterson, Reece L. (2002). The Color of Discipline: Sources of Racial and Gender Disproportionality in School Punishment. The Urban Review 34(4): 317-342. https://doi.org/10.1023/A:1021320817372
3. U.S. Government Accountability Office (2018). K-12 Education: Discipline Disparities for Black Students, Boys, and Students with Disabilities. GAO-18-258. https://www.gao.gov/products/gao-18-258
4. Chetty, Raj; Friedman, John N.; Hendren, Nathaniel; Jones, Maggie R.; Porter, Sonya R. (2018). The Opportunity Atlas: Mapping the Childhood Roots of Social Mobility. NBER Working Paper 25147. https://doi.org/10.3386/w25147
5. Chetty, Raj; Hendren, Nathaniel; Jones, Maggie R.; Porter, Sonya R. (2020). Race and Economic Opportunity in the United States: An Intergenerational Perspective. The Quarterly Journal of Economics 135(2): 711-783. https://doi.org/10.1093/qje/qjz042
6. Bacher-Hicks, Andrew; Billings, Stephen B.; Deming, David J. (2024). The School-to-Prison Pipeline: Long-Run Impacts of School Suspensions on Adult Crime. American Economic Journal: Economic Policy 16(4): 165-193. https://doi.org/10.1257/pol.20230052
7. Aizer, Anna; Doyle, Joseph J., Jr. (2015). Juvenile Incarceration, Human Capital, and Future Crime: Evidence from Randomly Assigned Judges. The Quarterly Journal of Economics 130(2): 759-803. https://doi.org/10.1093/qje/qjv003

## Provenance

- Pipeline version: 0.1.0
- Build script: book2/chapters/school-discipline-and-incarceration/build.py (sha256 7809484bea45139fb76b68c6e604b0053922bdfc25da0b185be8818e4ee9229a)
- Built on: 2026-10-07
- Site commit: c2b1ff3309cb4a10833f884da91ac0057a66ad0a

Inputs:

| File | Size (bytes) | SHA-256, first 1 MB | SHA-256, last 1 MB |
| --- | ---: | --- | --- |
| book2/chapters/school-discipline-and-incarceration/incarceration.py | 8,924 | 161d58ee04846c2d6a444dd1124a4fec9895013f72118d8866f36480a28dc062 | 161d58ee04846c2d6a444dd1124a4fec9895013f72118d8866f36480a28dc062 |
| book2/chapters/school-discipline-and-incarceration/schools.py | 22,041 | 2c521a43154abc3feea0f3f51e298cc5e98efc3f2b3c073e3e893c983b30950f | 2c521a43154abc3feea0f3f51e298cc5e98efc3f2b3c073e3e893c983b30950f |
| data/analytic/tract_atlas.parquet | 39,659,925 | 29d5e4d943552d54aeb6ba92aadc939d9bd15bdea553c78357dbd516757be627 | 1e10d1ecf722340f23ee2212d4999a90250f85b65fa89b39e32db06d9472e22e |
| data/raw/acs/deprivation_components_2019_2023/tract.parquet | 13,197,793 | 5545cde2cbe96a8b56e468d65388248a241aa6bd300d3c9e325d08d3f86b11da | 5c8ec454d066988b0f1a01aaa6ed34ebfc05d162e01cf9c6bb01f9198cbaff12 |
| data/raw/crdc/2020_21/2020-21-crdc-data.zip | 81,589,466 | e5eb2e56fab576ce2300839d0a799e859b8a13604516afd405cb0adce69067c3 | ed640cf911051de526693262a2e92ae36d54b2a60eff1346ccc4913b9669b63c |
| data/staged/adi/adi_atlas_tract2020_2023.parquet | 798,078 | 5e409bc212ff871def780fa27b4cf8b616d9da420b2b9f6455517eddcd8c1a95 | 5e409bc212ff871def780fa27b4cf8b616d9da420b2b9f6455517eddcd8c1a95 |
| data/staged/crosswalks/ct_tract2020_to_old_county.parquet | 15,842 | 961ef39f31596ef65fca2c67d5f2f63e533c5dc9b49745d3bb6437844e4b831d | 961ef39f31596ef65fca2c67d5f2f63e533c5dc9b49745d3bb6437844e4b831d |
| data/staged/crosswalks/tract_2010_to_2020.parquet | 2,602,636 | f7526f0d1bac31dbc540f6d733a850c9e0ee4c7622011214c04afedfbfa1e0a5 | 9e573492f9804b571b9944a53cc59fadab4051a175952b058f77cc5a644a98ff |
| data/staged/education/crdc_lea_2020_21.parquet | 912,989 | 559ab1d6e67839c6fb99a55d77b8b135d693b2f9ff4eb748b03afd61ceb4490b | 559ab1d6e67839c6fb99a55d77b8b135d693b2f9ff4eb748b03afd61ceb4490b |
| data/staged/food_access/food_access_tract2010_2019.parquet | 37,967,838 | c36929f1809e9ac84ac5f1cb62140bd980c315287833c6f281839df4c1619c89 | 05a17b940a79393b0c2ce5c9e9248496c319cba6896b63af3a2df34fe04f63aa |
| data/staged/mobility/oa_module1_tract2010.parquet | 14,946,600 | 32583830e03567b41ad3c245ca41aa67fb9cb55a8a46e8d9ae4c77a6cc2fcd83 | 6d67fac56edd8f07346375fb0b97e3c82621d57457d89c7fe93ec6f1b758bfaf |
| data/staged/ses/acs5_tract2020_2019_2023.parquet | 7,232,424 | 5fe6f730a35225f14d5d8abfc6b9bce58b52e545f309a7eac74f1dae74d7a31c | 7b4a2475b3852d3f91804716a04566ef8ceadacec2be9fb85f5eb642078efac6 |
| data/staged/svi/svi_tract2020_2022.parquet | 17,740,415 | cb3860d8edc2aaf1f2a02211a58f7a0b80457eb6ad9a37b5ec8abb2d5da9da10 | e371a4a696d5bed05d54ad7c5b6ca2504508805b847ec415c2322be7f66ec2b2 |

Outputs:

| File | SHA-256 |
| --- | --- |
| facts.json | 88a92432b6bedeb487d217f56ccb7d2d909106e9fd65f23139263eec30bcf089 |
| figures/alt_rankings.json | aadc8b9fe818480dbeadcf72b8d936746dcbf813936dd21af727ebb3dde2fb1f |
| figures/exclusion_funnel.json | 6ecfa3e856f8a710109a2dab344170fc399d0de0de3b117fd56adc34f3f82f67 |
| figures/incarceration_gradient.json | 0d8a4e6e4eb5ac7f097793036c26d3d7c89093f9a6905319b871d1af4212228d |
| figures/suspension_black_white.json | 120c906afc28cbe244b48c034d45d9b5fbd302365283c4409cb0968df8f35655 |
| figures/suspension_dose.json | 0c78be52ad32efc31039370c257c107de7299cff3f99ba99f21a1c4f25e4b655 |
| figures/suspension_map.json | 6e76ea59f092baa9541d477186ee048813498476ccbe47e3370ed2ebe4a79b3a |
| meta.json | 622b39e32c6a1b30fc1889dd2c99e25f8369d6cdc6f350ced7cd5037964cb585 |
| public/data/school-discipline-and-incarceration/suspension_dose.json | 61ecbb7272027e016cb21b4a534d4cf2efa4b4bc0c452e3d424f06167bf23e34 |
| public/data/school-discipline-and-incarceration/suspension_map.json | e69f0ff17a725e2b4005a971e3f11d0012a2138087b170d4eaea25df190cddf1 |
| refs.json | 1713c9c8646927a2263864328f537b14189cd9e1b57bae8d26904dd2fb46b701 |

## Machine formats

- Report (HTML): https://vishalsingh.org/zipcode/school-discipline-and-incarceration/report
- Report (Markdown): https://vishalsingh.org/zipcode/school-discipline-and-incarceration/report.md
- Every estimate (JSON): https://vishalsingh.org/zipcode/report/school-discipline-and-incarceration/facts.json
- Table 1, figure "suspension_dose" (JSON): https://vishalsingh.org/zipcode/report/school-discipline-and-incarceration/figures/suspension_dose.json
- Table 2, figure "exclusion_funnel" (JSON): https://vishalsingh.org/zipcode/report/school-discipline-and-incarceration/figures/exclusion_funnel.json
- Table 3, figure "suspension_black_white" (JSON): https://vishalsingh.org/zipcode/report/school-discipline-and-incarceration/figures/suspension_black_white.json
- Table 4, figure "suspension_map" (JSON): https://vishalsingh.org/zipcode/report/school-discipline-and-incarceration/figures/suspension_map.json
- Table 5, figure "incarceration_gradient" (JSON): https://vishalsingh.org/zipcode/report/school-discipline-and-incarceration/figures/incarceration_gradient.json
- Table 6, figure "alt_rankings" (JSON): https://vishalsingh.org/zipcode/report/school-discipline-and-incarceration/figures/alt_rankings.json
- Payload of "suspension_dose" (JSON): https://vishalsingh.org/zipcode/data/school-discipline-and-incarceration/suspension_dose.json
- Payload of "suspension_map" (JSON): https://vishalsingh.org/zipcode/data/school-discipline-and-incarceration/suspension_map.json
- Index of every report: https://vishalsingh.org/zipcode/llms.txt; all reports in one file: https://vishalsingh.org/zipcode/llms-full.txt

---

# College access, price and earnings: report

Chapter 4.1 of ZIP Code Destiny, by Vishal Singh (NYU Stern). Generated from the chapter's data files; the reading version is at https://vishalsingh.org/zipcode/college-access-price-and-earnings. Table n of this report is figure n of the chapter.

## Identification

- Title: College access, price and earnings
- Label: The toll booth
- Chapter: 4.1 (Part IV, Becoming an adult; An American life)
- Question: How do college prices, earnings after college and the distance to the nearest college differ by family income and county?
- Dek: A decade after entry, the median former student who received federal aid earned $28,330 at the median for-profit college and $42,014 at the median public college, for students who entered in 2009 to 2011.
- Grain: institutions
- Status: published
- Published: 2026-10-09
- Updated: not revised
- Data sources:
  - College Scorecard. Median earnings a decade after entry (entering classes of 2009 to 2010 and 2010 to 2011, 2022 dollars), net price by family-income band, cost of attendance, Pell share (academic year 2021 to 2022), completion and debt; institutions in the 50 states and DC, one row per 6-digit OPE ID; years: May 2025 release (most recent cohorts); weights: none (each institution counts once); enrollment weights as a check
  - IPEDS. Campus locations (latitude and longitude, county) of four-year institutions, 2023 directory; years: 2023; weights: none (campuses are points)
  - Census TIGER/Line tracts. 2020 census tract internal points, from which each tract's distance to the nearest campus is measured; years: 2020; weights: none (points)
  - Area Deprivation Index. Area Deprivation Index 2023 national rank of tracts, averaged to counties; years: 2023 release (version 4.0.1; built on ACS 2019-2023); weights: 2019 to 2023 ACS tract population
  - Rural-Urban Continuum Codes. Rural-Urban Continuum Codes 2023 (metro 1 to 3, nonmetro 4 to 9); years: 2023; weights: none (classification)
  - American Community Survey. 2019 to 2023 tract population, used to weight tracts into county values; years: 2019-2023 (5-year); weights: none (population counts)
- Canonical URL: https://vishalsingh.org/zipcode/college-access-price-and-earnings
- Report: https://vishalsingh.org/zipcode/college-access-price-and-earnings/report
- Markdown: https://vishalsingh.org/zipcode/college-access-price-and-earnings/report.md
- BibTeX key: singh2026collegeaccesspriceandearnings

```bibtex
@misc{singh2026collegeaccesspriceandearnings,
  author = {Singh, Vishal},
  title = {College access, price and earnings},
  year = {2026},
  url = {https://vishalsingh.org/zipcode/college-access-price-and-earnings},
  note = {In ZIP Code Destiny. Published 9 October 2026.}
}
```

## Principal estimates

The facts the dek and the opening prose cite, in order of appearance.

| Estimand | Population | Period | Value | Interval | n | Weights | Source | Fact id |
| --- | --- | --- | ---: | --- | ---: | --- | --- | --- |
| median across institutions of each institution's median earnings of former students who received federal aid, working and not enrolled ten years after entry; bootstrap over states | for-profit colleges in the 50 states and DC with a published earnings value | median earnings in calendar years 2020 and 2021 (2022 dollars) of students who entered in 2009 to 2010 and 2010 to 2011 | $28,330 | 95% CI $26,877 to $29,972 | 1,036 institutions | none (each institution counts once) | College Scorecard | earn_forprofit |
| median across institutions of each institution's median earnings of former students who received federal aid, working and not enrolled ten years after entry; bootstrap over states | public colleges in the 50 states and DC with a published earnings value | median earnings in calendar years 2020 and 2021 (2022 dollars) of students who entered in 2009 to 2010 and 2010 to 2011 | $42,014 | 95% CI $40,606 to $43,598 | 1,716 institutions | none (each institution counts once) | College Scorecard | earn_public |
| median across institutions of each institution's median earnings of former students who received federal aid, working and not enrolled ten years after entry; bootstrap over states | private nonprofit colleges in the 50 states and DC with a published earnings value | median earnings in calendar years 2020 and 2021 (2022 dollars) of students who entered in 2009 to 2010 and 2010 to 2011 | $53,848 | 95% CI $51,671 to $55,838 | 1,211 institutions | none (each institution counts once) | College Scorecard | earn_nonprofit |

## Findings

A decade after entering, the median former student who had received federal aid earned $28,330 at the median for-profit college, $42,014 at the median public college and $53,848 at the median private nonprofit college. Each figure is the middle value across the colleges of a sector, where a college's own value is the median earnings of its former students in 2022 dollars. The students entered in 2009 to 2010 and 2010 to 2011 and were observed in 2020 and 2021.

Three features of a college matter to a family, which are what its former students earn, what it bills and whether it is within reach. This chapter takes them in that order, with the institution as the unit for earnings and price and the county as the unit for reach. The data are the College Scorecard for earnings and price and the IPEDS directory for the location of campuses. They describe colleges and counties, and a pattern across colleges is not a statement about what a college does for a student.

### For-profit colleges have the lowest median earnings

The median across the 1,036 for-profit colleges with a published value is $28,330 (95% CI $26,877 to $29,972), against $42,014 (95% CI $40,606 to $43,598) across 1,716 public colleges. The for-profit median is $13,685 (95% CI $12,287 to $15,220) below the public median (Table 1).

The private nonprofit median, across 1,211 colleges, is $53,848 (95% CI $51,671 to $55,838), above both. The intervals come from a bootstrap that redraws whole states, because the colleges of a state share appropriations, labor markets and aid rules.

The sectors differ in what they offer. For-profit colleges run more short programs, so a comparison of all colleges could reflect the mix of programs. Among four-year colleges alone the public median is $10,738 (95% CI $8,300 to $13,249) above the for-profit median, and among two-year and shorter programs it is $11,884 (95% CI $10,591 to $13,240) above. Weighting each college by its undergraduates, which counts a large college in proportion to its students, gives a gap of $10,924 (95% CI $7,972 to $12,851). The order of the sectors holds in each of these cuts.

The sectors also differ in who enrolls. The median share of undergraduates who received a Pell Grant in academic year 2021 to 2022, a federal grant for students with financial need, is 52.5% (95% CI 50.4% to 54.6%) at for-profit colleges, 30.4% (95% CI 28.3% to 32.9%) at public colleges and 32.4% (95% CI 30.1% to 34.6%) at private nonprofit colleges. Deming, Goldin and Katz compared beginning students at for-profit colleges with comparable students at other colleges in a national survey, adjusting for observed characteristics. They found lower earnings after entry and higher loan default rates among the for-profit students.[1] The Scorecard figures point the same way a decade after entry, but they do not adjust for who enrolls, so they cannot separate what a college does from whom it enrolls.

The medians describe the colleges that report. The College Scorecard covers institutions that take part in federal student aid programs. It lists 5,000 of them in the states and the District of Columbia once the 1,275 branch records that repeat a parent institution's earnings are set aside, and 3,963 of them have a published earnings value.

Earnings are withheld when the group of aided students is small or the value fails the Scorecard's checks. The share of colleges with no published value is 4.7% of public colleges, 26.5% of private nonprofit colleges and 33.2% of for-profit colleges. A withheld value is missing and not zero, and if the colleges without a value earn systematically more or less than those with one, the sector medians move.

Table 1. At the median for-profit college, former students' median earnings a decade after entry were $28,330, against $42,014 at the median public college and $53,848 at the median private nonprofit college

### Net price rises with family income but falls as a share of it

The Scorecard reports each college's average net price for academic year 2021 to 2022, in nominal dollars, in five bands of family income. The net price is the cost of attendance minus grants and scholarships, for full-time, first-time students who received federal aid. The lowest band ends at $30,000 and the highest begins above $110,000. The medians below cover 694 public and 1,157 private nonprofit four-year colleges with a published price.

At the median public four-year college the net price was $9,347 (95% CI $8,426 to $10,430) a year in the lowest band and $18,780 (95% CI $17,910 to $19,987) in the highest. At the median private nonprofit four-year college it was $17,596 (95% CI $16,937 to $18,307) and $28,710 (95% CI $26,953 to $30,525) (Table 2). The schedule is progressive in dollars, since prices rise with income at both kinds of college.

A price means more against an income. Dividing each median by an assumed family income at the midpoint of its band, $15,000 in the lowest band and $150,000 in the highest, the public net price is 62% (95% CI 56% to 70%) of the assumed income in the lowest band and 13% (95% CI 12% to 13%) in the highest.

The private nonprofit net price is 117% (95% CI 113% to 122%) and 19% (95% CI 18% to 20%). Measured this way the schedule is regressive. The fall is steep at the start. The step from the lowest band to the second band carries 73% (95% CI 71% to 74%) of the whole fall at public colleges and 72% (95% CI 71% to 73%) at private nonprofit colleges, and the share changes slowly after that.

The share is an order of magnitude and not a measured burden. The cost of attendance includes living expenses that a family bears whether or not the student enrolls, the Scorecard does not report the incomes of the families in a band, and the highest band has no upper limit. A private nonprofit net price above the assumed income in the lowest band shows the limits of the assumption as much as it shows a burden. The contrast between the ends does not depend on the highest midpoint. With the income of the highest band set at its lower limit, the smallest a family in it can have, the public share is 17% (95% CI 16% to 18%) and the private nonprofit share 26% (95% CI 25% to 28%), still far below the shares in the lowest band.

Table 2. The median net price rises with family income, but at public four-year colleges it falls from 62% of an assumed income in the lowest band to 13% in the highest

### Cost, earnings and completion rise together

Across the 2,663 colleges with a published cost of attendance, median earnings and at least 100 undergraduates, colleges that cost more have former students who earn more, with a correlation of +0.63 (95% CI 0.59 to 0.67).

The completion rate is the share of first-time, full-time students who finish within 150% of the normal time. It moves with earnings among four-year colleges, with a correlation of +0.67 (95% CI 0.62 to 0.71), and weakly among two-year and shorter programs, with a correlation of +0.12 (95% CI 0.02 to 0.21) (Table 3). The Scorecard measures completion for students who started in fall 2016 at four-year colleges and fall 2019 at shorter programs, so the two rates are reported apart and not pooled.

One corner of the plot is different. The plot divides at the median cost of attendance, $24,602 (95% CI $22,758 to $26,688), and the median earnings, $46,092 (95% CI $44,379 to $47,866). A total of 334 colleges, 12.5% (95% CI 11.3% to 13.8%) of those plotted, cost more than the median and have median earnings at or below it.

They are 56% (95% CI 48% to 62%) of the 163 plotted for-profit colleges, 20% (95% CI 18% to 23%) of the plotted private nonprofit colleges and 2% (95% CI 1% to 3%) of the plotted public colleges. The cost of attendance is published for academic-year institutions only, so many short for-profit programs are not plotted and the for-profit share describes the plotted colleges.

The colleges in that corner enroll more students from low-income families and graduate fewer of them. Their median Pell share is 47% (95% CI 42% to 51%), against 30% (95% CI 28% to 31%) in the other three quadrants. Among four-year colleges their median completion rate is 45% (95% CI 43% to 51%), against 64% (95% CI 61% to 67%) at four-year colleges that cost more than the median and earn more. These are comparisons across colleges. They do not show that a student who enrolls in one of these colleges earns less than the student would have earned elsewhere.

Table 3. Former students of costlier colleges tend to earn more, a correlation of +0.63, but 56% of plotted for-profit colleges cost more than the median and have median earnings at or below it

### High-Pell colleges differ widely in earnings

Chetty and co-authors, who linked tax records to the colleges students attended, found that parental income is as segregated across colleges as it is across neighborhoods and that elite colleges serve mostly students from high-income families. They also found that earnings differences between students from low-income and high-income families shrink considerably among students who attend the same college.[2] An earlier report from the same project named the City University of New York and the California State campuses among the colleges where students from the bottom fifth of the income distribution most often reached the top fifth.[3]

A narrower comparison here looks only at colleges that enroll many students from low-income families. The pool is the 1,282 public and private nonprofit four-year colleges with at least 1,000 undergraduates and a published Pell share, earnings and net price for the lowest income band. The 428 colleges in the top third of the pool by Pell share, a share of 34.7% or more, are the high-Pell colleges. For-profit colleges are not on the list by rule, since the sector comparison above treats them separately, and 48 for-profit four-year colleges that meet every other rule, including a net price for the lowest income band, are left out.

Median earnings among the high-Pell colleges run from $38,184 at the tenth percentile to $62,278 at the ninetieth. The 15 highest-earning have a median of $72,282 and the 15 lowest-earning a median of $31,902 (Table 4).

The two lists do not have the same Pell share. The median is 39% in the highest-earning list and 60% in the lowest-earning list, so part of the difference in earnings may reflect the starting point of the students.

Baruch College of the City University of New York is in the first list, with a Pell share of 55%, a net price of $745 for students in the lowest income band and median earnings of $75,971. Place matters for the ranking, since 8 of the 10 highest-earning high-Pell colleges are in California or New York and earnings in these data are not adjusted for local wages.

Of the 15 lowest-earning colleges, 5 are tribal colleges or religious seminaries, namely Oglala Lakota College, Navajo Technical University, Dine College, United Talmudical Seminary and Uta Mesivta of Kiryas Joel. Another 6 are historically Black colleges and universities, namely Harris-Stowe State University, Benedict College, Mississippi Valley State University, Miles College, Langston University and Central State University. The students and local labor markets of the first group differ from those of general four-year colleges, so their earnings are not comparable with those of a general four-year degree. The measure adjusts neither for local wages nor for the starting income of the students at any of them. Net price does not separate the lists, since the median net price for the lowest income band is $9,743 at the highest-earning colleges and $10,626 at the lowest-earning ones.

Table 4. Among high-Pell colleges, the 15 highest-earning have a median of $72,282 a decade after entry and the 15 lowest-earning $31,902

### Most counties have no four-year college

The IPEDS directory lists 2,508 four-year campuses in the states and the District of Columbia, after leaving out the 273 institutions that grant only graduate degrees.

They sit in 933 of the 3,143 counties. The other 2,210 counties, 70% of all, have no four-year campus inside the county line. Connecticut counts as its eight legacy counties.

A county line is a poor measure of reach, because a county without a campus can border one. The chapter measures instead, for each census tract, the straight-line distance in miles from the tract's central point to the nearest four-year campus, and gives each county the average of its tracts weighted by population. A county's value is the distance of its average resident, so a county with a campus inside it can still have a value above zero. The average resident in the country lives 9.6 miles from a four-year campus. Measured to the nearest public four-year campus instead, the average resident lives 14.0 miles away. Counting each county once, the median value is 21 miles, and the values run from 4.0 miles at the fifth percentile to 74 miles at the ninety-fifth (Table 5).

The longest distances are in Alaska. Each of the 10 counties with the longest distances is in Alaska, and the average Alaskan lives 69 miles from a four-year campus. Among states the averages are next longest in Wyoming (49 miles), Montana (26 miles) and Mississippi (24 miles). Distances in Alaska and Hawaii are straight lines over water and terrain that no road covers.

Reach also shapes who applies. Hoxby and Avery found that most high-achieving students from low-income families do not apply to selective colleges. These students are geographically dispersed, which makes them harder to reach through recruiting concentrated in a few large metropolitan areas.[4] The map measures the supply side of that problem. It says where campuses are, not where students enroll.

Table 5. The average resident in the country lives 9.6 miles from a four-year college, and the state averages are longest in Alaska, Wyoming, Montana and Mississippi

### Distance rises with county deprivation and with rurality

Each county has a rank on the Area Deprivation Index, the average of its tracts' national ranks weighted by population, with a higher rank meaning more deprived. The 3,128 counties that have a rank are sorted into tenths of equal numbers, and the 15 counties whose tracts all have a suppressed rank are left out. The distance for the average resident rises across the tenths, from 6.1 (95% CI 5.3 to 6.9) miles in the least deprived tenth to 29.3 (95% CI 24.6 to 35.0) miles in the most deprived, or 2.1 (95% CI 1.7 to 2.5) miles for each step up. The rise is not steady, and the second, third and fourth tenths are not distinguishable from one another.

The share of counties with no campus inside rises across the tenths too, from 42% (95% CI 33% to 51%) to 92% (95% CI 87% to 95%), or 5.3 percentage points (95% CI 4.5 percentage points to 6.2 percentage points) for each step up (Table 6). The intervals describe variation across states, because the tracts and campuses are enumerated and not sampled.

Across counties the correlation of the rank with the distance is +0.31 (95% CI 0.24 to 0.40) when counties count in proportion to their residents and +0.16 (95% CI 0.09 to 0.33) when each counts once. Measured to the nearest public four-year campus, the distance rises from 9.4 (95% CI 8.3 to 10.8) miles in the least deprived tenth to 36.5 (95% CI 32.4 to 41.6) miles in the most deprived.

Rurality accounts for much of the gradient. The Rural-Urban Continuum Code of the US Department of Agriculture sorts counties by the size of their metro area or, outside metro areas, by the size of the county's urban population and its adjacency to a metro area. These comparisons use all counties, including the ones without an ADI rank. The average resident of a metro county lives 7.0 (95% CI 6.3 to 7.8) miles from a four-year campus and the average resident of a nonmetro county 25.7 (95% CI 22.8 to 29.4) miles, which is 3.7 (95% CI 3.2 to 4.2) times as far.

In the group of nonmetro counties with the smallest urban populations the distance is 34.3 (95% CI 29.7 to 40.9) miles. Of these counties, 95% (95% CI 94% to 97%) have no campus inside, against 83% (95% CI 80% to 86%) of nonmetro counties together and 49% (95% CI 42% to 55%) of metro counties.

Deprivation and rurality overlap. Of the counties in the most deprived tenth, 94% (95% CI 92% to 96%) are nonmetro, against 23% (95% CI 14% to 31%) in the least deprived tenth. Within nonmetro counties the distance does not move with the deprivation rank, with a correlation of −0.04 (95% CI −0.12 to 0.05), while within metro counties it rises modestly, with a correlation of +0.24 (95% CI 0.18 to 0.32). The gradient across all counties therefore mostly reflects that deprived counties are more often rural. The ADI is a national rank that understates deprivation in high-cost areas, and a county average hides differences among its tracts.

Table 6. Distance to a four-year college rises with county deprivation, from 6.1 miles in the least deprived tenth of counties to 29.3 in the most deprived, and with rurality

> Causal evidence: What designs that can speak to cause find about admission, aid, funding and credentials (Regression discontinuity; instrumental variables; randomized resume experiment)
>
> The comparisons above describe differences across colleges and counties. Four studies use designs that can speak to cause at specific points.
>
> Zimmerman compared students just above and just below a grade-point threshold for admission to a large public university in Florida, a regression discontinuity design. Students just above the threshold were more likely to attend a university and earned more in the years that followed, and the effect applies to students near the threshold at one university.[5] Bettinger and co-authors used the grade-point and family-income cutoffs of California's Cal Grant, also a regression discontinuity design, and found that eligibility raised degree completion and, for some groups, later earnings, without raising enrollment.[6]
>
> Chakrabarti, Gorton and Lovenheim used an instrumental variables design in which each public institution's baseline reliance on state appropriations is combined with year-to-year changes in state appropriations. At four-year colleges they found that more state support lowered debt originations and adverse debt events and shortened time to degree, with little effect on other outcomes, and that the support was passed on to students as lower tuition with no change in institutional spending. At two-year colleges the gains were broader, with higher educational attainment and credit scores, more home ownership and less delinquency, and the support went to both lower tuition and educational resources.[7] Deming and co-authors sent fictitious resumes that differed in the sector and selectivity of the college to real job postings, a randomized resume experiment, and found that a business bachelor's degree from a for-profit online institution received fewer callbacks than one from a nonselective public institution.[8] The experiment measures employers' callbacks and not earnings.
>
> Each study speaks to one margin, which is admission at a threshold, eligibility for aid, state funding or a credential on a resume. None tests whether the sector gap in earnings or the schedule of net prices in this chapter would change if a student moved from one college to another.

### Distance rises with deprivation under all four rankings

Table 7 ranks the 3,125 counties that have a distance and all four rankings in four ways, with each county taking the population-weighted mean of its tracts' ranks. Ranked by the ADI on this common sample, the distance for the average resident is 6.1 (95% CI 5.3 to 7.0) miles in the least deprived tenth and 29.7 (95% CI 24.8 to 35.5) miles in the most deprived, and the rank correlation between tenth and distance is +0.99. Under the standardized index the two ends are 6.8 (95% CI 6.2 to 7.6) and 28.2 (95% CI 22.4 to 41.4) miles, with a correlation of +0.93. Under median household income they are 6.7 (95% CI 6.0 to 7.3) and 26.6 (95% CI 22.8 to 31.7) miles, and under the bachelor's degree share 5.5 (95% CI 5.0 to 6.2) and 31.0 (95% CI 25.4 to 40.5) miles, with correlations of +0.99 and +0.99, so none of the alternatives counts as a change under the book's rule (Chapter 9.1, How far one ranking goes (https://vishalsingh.org/zipcode/how-far-one-ranking-goes)).

Table 7. Distance to the nearest four-year college is steadily worse toward the more deprived end under the ADI and under each of the three other rankings

### What these data cannot show

The geography describes supply. Distance is measured in a straight line from where residents live to where campuses sit, so it ignores roads, ferries, online programs and the colleges students choose when they leave home. The Chetty and co-authors data that link the neighborhood a student grew up in to the college the student attended and the income reached are not part of this project's data, and nothing here reproduces them. A county average also hides the tracts inside it, and the gradients describe counties and not the families who live in them.

The earnings describe aided students who entered in 2009 to 2011, not all students and not those entering now. They are medians of institutions, with noise added by the Scorecard, and a larger share of private colleges than of public colleges have no published value. Branch campuses repeat their parent's earnings and debt, which is why the chapter counts institutions by OPE ID. The Pell share, cost of attendance, net price and completion rate come from the largest campus of each institution. Earnings follow local wages and the field of study as well as the college, and the sector gap mixes what colleges do with who enrolls and what programs they run.

The net price is an average over aided, first-time, full-time students in each band, at public colleges for in-state students only, and the share of income depends on assumed incomes. The lists of high-Pell colleges are rankings of named colleges, with no interval, and they include colleges whose earnings are not comparable with those of a general four-year degree. The designs in the box above identify effects at specific margins and do not show what a college does to a student in general.

## Tables

### Table 1. At the median for-profit college, former students' median earnings a decade after entry were $28,330, against $42,014 at the median public college and $53,848 at the median private nonprofit college

Distribution across colleges of the median earnings of former students who received federal aid, a decade after entry (thousands of 2022 dollars), by sector; entering classes of 2009 to 2010 and 2010 to 2011

| density | earnings | scope | sector | weighting |
| ---: | ---: | --- | --- | --- |
| 0 | 0 | All colleges | Public | Each college counted once |
| 0 | 1 | All colleges | Public | Each college counted once |
| 0 | 2 | All colleges | Public | Each college counted once |
| 0 | 3 | All colleges | Public | Each college counted once |
| 0 | 4 | All colleges | Public | Each college counted once |
| 0 | 5 | All colleges | Public | Each college counted once |
| 0 | 6 | All colleges | Public | Each college counted once |
| 0 | 7 | All colleges | Public | Each college counted once |
| 0 | 8 | All colleges | Public | Each college counted once |
| 0 | 9 | All colleges | Public | Each college counted once |
| 0 | 10 | All colleges | Public | Each college counted once |
| 1 | 11 | All colleges | Public | Each college counted once |
| 1 | 12 | All colleges | Public | Each college counted once |
| 2 | 13 | All colleges | Public | Each college counted once |
| 3 | 14 | All colleges | Public | Each college counted once |
| 4 | 15 | All colleges | Public | Each college counted once |
| 5 | 16 | All colleges | Public | Each college counted once |
| 6 | 17 | All colleges | Public | Each college counted once |
| 8 | 18 | All colleges | Public | Each college counted once |
| 10 | 19 | All colleges | Public | Each college counted once |
| 13 | 20 | All colleges | Public | Each college counted once |
| 17 | 21 | All colleges | Public | Each college counted once |
| 21 | 22 | All colleges | Public | Each college counted once |
| 27 | 23 | All colleges | Public | Each college counted once |
| 35 | 24 | All colleges | Public | Each college counted once |
| 49 | 25 | All colleges | Public | Each college counted once |
| 71 | 26 | All colleges | Public | Each college counted once |
| 104 | 27 | All colleges | Public | Each college counted once |
| 152 | 28 | All colleges | Public | Each college counted once |
| 217 | 29 | All colleges | Public | Each college counted once |
| 297 | 30 | All colleges | Public | Each college counted once |
| 389 | 31 | All colleges | Public | Each college counted once |
| 490 | 32 | All colleges | Public | Each college counted once |
| 594 | 33 | All colleges | Public | Each college counted once |
| 695 | 34 | All colleges | Public | Each college counted once |
| 788 | 35 | All colleges | Public | Each college counted once |
| 870 | 36 | All colleges | Public | Each college counted once |
| 935 | 37 | All colleges | Public | Each college counted once |
| 979 | 38 | All colleges | Public | Each college counted once |
| 1000 | 39 | All colleges | Public | Each college counted once |
| 996 | 40 | All colleges | Public | Each college counted once |
| 970 | 41 | All colleges | Public | Each college counted once |
| 925 | 42 | All colleges | Public | Each college counted once |
| 868 | 43 | All colleges | Public | Each college counted once |
| 805 | 44 | All colleges | Public | Each college counted once |
| 741 | 45 | All colleges | Public | Each college counted once |
| 681 | 46 | All colleges | Public | Each college counted once |
| 625 | 47 | All colleges | Public | Each college counted once |
| 573 | 48 | All colleges | Public | Each college counted once |
| 524 | 49 | All colleges | Public | Each college counted once |
| 477 | 50 | All colleges | Public | Each college counted once |
| 433 | 51 | All colleges | Public | Each college counted once |
| 393 | 52 | All colleges | Public | Each college counted once |
| 360 | 53 | All colleges | Public | Each college counted once |
| 333 | 54 | All colleges | Public | Each college counted once |
| 310 | 55 | All colleges | Public | Each college counted once |
| 289 | 56 | All colleges | Public | Each college counted once |
| 266 | 57 | All colleges | Public | Each college counted once |
| 242 | 58 | All colleges | Public | Each college counted once |
| 218 | 59 | All colleges | Public | Each college counted once |
| 196 | 60 | All colleges | Public | Each college counted once |
| 178 | 61 | All colleges | Public | Each college counted once |
| 162 | 62 | All colleges | Public | Each college counted once |
| 148 | 63 | All colleges | Public | Each college counted once |
| 132 | 64 | All colleges | Public | Each college counted once |
| 117 | 65 | All colleges | Public | Each college counted once |
| 102 | 66 | All colleges | Public | Each college counted once |
| 90 | 67 | All colleges | Public | Each college counted once |
| 80 | 68 | All colleges | Public | Each college counted once |
| 73 | 69 | All colleges | Public | Each college counted once |
| 68 | 70 | All colleges | Public | Each college counted once |
| 64 | 71 | All colleges | Public | Each college counted once |
| 60 | 72 | All colleges | Public | Each college counted once |
| 57 | 73 | All colleges | Public | Each college counted once |
| 53 | 74 | All colleges | Public | Each college counted once |
| 48 | 75 | All colleges | Public | Each college counted once |
| 44 | 76 | All colleges | Public | Each college counted once |
| 40 | 77 | All colleges | Public | Each college counted once |
| 36 | 78 | All colleges | Public | Each college counted once |
| 33 | 79 | All colleges | Public | Each college counted once |
| 31 | 80 | All colleges | Public | Each college counted once |
| 30 | 81 | All colleges | Public | Each college counted once |
| 28 | 82 | All colleges | Public | Each college counted once |
| 25 | 83 | All colleges | Public | Each college counted once |
| 22 | 84 | All colleges | Public | Each college counted once |
| 19 | 85 | All colleges | Public | Each college counted once |
| 17 | 86 | All colleges | Public | Each college counted once |
| 17 | 87 | All colleges | Public | Each college counted once |
| 17 | 88 | All colleges | Public | Each college counted once |
| 19 | 89 | All colleges | Public | Each college counted once |
| 20 | 90 | All colleges | Public | Each college counted once |
| 20 | 91 | All colleges | Public | Each college counted once |
| 20 | 92 | All colleges | Public | Each college counted once |
| 18 | 93 | All colleges | Public | Each college counted once |
| 16 | 94 | All colleges | Public | Each college counted once |
| 13 | 95 | All colleges | Public | Each college counted once |
| 11 | 96 | All colleges | Public | Each college counted once |
| 8 | 97 | All colleges | Public | Each college counted once |
| 7 | 98 | All colleges | Public | Each college counted once |
| 5 | 99 | All colleges | Public | Each college counted once |
| 5 | 100 | All colleges | Public | Each college counted once |
| 5 | 101 | All colleges | Public | Each college counted once |
| 4 | 102 | All colleges | Public | Each college counted once |
| 4 | 103 | All colleges | Public | Each college counted once |
| 3 | 104 | All colleges | Public | Each college counted once |
| 2 | 105 | All colleges | Public | Each college counted once |
| 2 | 106 | All colleges | Public | Each college counted once |
| 2 | 107 | All colleges | Public | Each college counted once |
| 2 | 108 | All colleges | Public | Each college counted once |
| 2 | 109 | All colleges | Public | Each college counted once |
| 2 | 110 | All colleges | Public | Each college counted once |
| 2 | 111 | All colleges | Public | Each college counted once |
| 2 | 112 | All colleges | Public | Each college counted once |
| 2 | 113 | All colleges | Public | Each college counted once |
| 2 | 114 | All colleges | Public | Each college counted once |
| 2 | 115 | All colleges | Public | Each college counted once |
| 2 | 116 | All colleges | Public | Each college counted once |
| 2 | 117 | All colleges | Public | Each college counted once |
| 1 | 118 | All colleges | Public | Each college counted once |
| 1 | 119 | All colleges | Public | Each college counted once |
| 0 | 120 | All colleges | Public | Each college counted once |
| 0 | 0 | All colleges | Private nonprofit | Each college counted once |
| 0 | 1 | All colleges | Private nonprofit | Each college counted once |
| 0 | 2 | All colleges | Private nonprofit | Each college counted once |
| 0 | 3 | All colleges | Private nonprofit | Each college counted once |
| 0 | 4 | All colleges | Private nonprofit | Each college counted once |
| 0 | 5 | All colleges | Private nonprofit | Each college counted once |
| 0 | 6 | All colleges | Private nonprofit | Each college counted once |
| 0 | 7 | All colleges | Private nonprofit | Each college counted once |
| 0 | 8 | All colleges | Private nonprofit | Each college counted once |
| 0 | 9 | All colleges | Private nonprofit | Each college counted once |
| 0 | 10 | All colleges | Private nonprofit | Each college counted once |
| 0 | 11 | All colleges | Private nonprofit | Each college counted once |
| 1 | 12 | All colleges | Private nonprofit | Each college counted once |
| 3 | 13 | All colleges | Private nonprofit | Each college counted once |
| 6 | 14 | All colleges | Private nonprofit | Each college counted once |
| 11 | 15 | All colleges | Private nonprofit | Each college counted once |
| 17 | 16 | All colleges | Private nonprofit | Each college counted once |
| 26 | 17 | All colleges | Private nonprofit | Each college counted once |
| 37 | 18 | All colleges | Private nonprofit | Each college counted once |
| 49 | 19 | All colleges | Private nonprofit | Each college counted once |
| 63 | 20 | All colleges | Private nonprofit | Each college counted once |
| 78 | 21 | All colleges | Private nonprofit | Each college counted once |
| 93 | 22 | All colleges | Private nonprofit | Each college counted once |
| 108 | 23 | All colleges | Private nonprofit | Each college counted once |
| 121 | 24 | All colleges | Private nonprofit | Each college counted once |
| 132 | 25 | All colleges | Private nonprofit | Each college counted once |
| 143 | 26 | All colleges | Private nonprofit | Each college counted once |
| 155 | 27 | All colleges | Private nonprofit | Each college counted once |
| 170 | 28 | All colleges | Private nonprofit | Each college counted once |
| 192 | 29 | All colleges | Private nonprofit | Each college counted once |
| 219 | 30 | All colleges | Private nonprofit | Each college counted once |
| 252 | 31 | All colleges | Private nonprofit | Each college counted once |
| 289 | 32 | All colleges | Private nonprofit | Each college counted once |
| 328 | 33 | All colleges | Private nonprofit | Each college counted once |
| 368 | 34 | All colleges | Private nonprofit | Each college counted once |
| 407 | 35 | All colleges | Private nonprofit | Each college counted once |
| 445 | 36 | All colleges | Private nonprofit | Each college counted once |
| 484 | 37 | All colleges | Private nonprofit | Each college counted once |
| 524 | 38 | All colleges | Private nonprofit | Each college counted once |
| 565 | 39 | All colleges | Private nonprofit | Each college counted once |
| 607 | 40 | All colleges | Private nonprofit | Each college counted once |
| 649 | 41 | All colleges | Private nonprofit | Each college counted once |
| 689 | 42 | All colleges | Private nonprofit | Each college counted once |
| 729 | 43 | All colleges | Private nonprofit | Each college counted once |
| 771 | 44 | All colleges | Private nonprofit | Each college counted once |
| 816 | 45 | All colleges | Private nonprofit | Each college counted once |
| 862 | 46 | All colleges | Private nonprofit | Each college counted once |
| 903 | 47 | All colleges | Private nonprofit | Each college counted once |
| 934 | 48 | All colleges | Private nonprofit | Each college counted once |
| 954 | 49 | All colleges | Private nonprofit | Each college counted once |
| 968 | 50 | All colleges | Private nonprofit | Each college counted once |
| 980 | 51 | All colleges | Private nonprofit | Each college counted once |
| 992 | 52 | All colleges | Private nonprofit | Each college counted once |
| 1000 | 53 | All colleges | Private nonprofit | Each college counted once |
| 999 | 54 | All colleges | Private nonprofit | Each college counted once |
| 985 | 55 | All colleges | Private nonprofit | Each college counted once |
| 960 | 56 | All colleges | Private nonprofit | Each college counted once |
| 930 | 57 | All colleges | Private nonprofit | Each college counted once |
| 896 | 58 | All colleges | Private nonprofit | Each college counted once |
| 856 | 59 | All colleges | Private nonprofit | Each college counted once |
| 808 | 60 | All colleges | Private nonprofit | Each college counted once |
| 752 | 61 | All colleges | Private nonprofit | Each college counted once |
| 690 | 62 | All colleges | Private nonprofit | Each college counted once |
| 629 | 63 | All colleges | Private nonprofit | Each college counted once |
| 572 | 64 | All colleges | Private nonprofit | Each college counted once |
| 523 | 65 | All colleges | Private nonprofit | Each college counted once |
| 484 | 66 | All colleges | Private nonprofit | Each college counted once |
| 455 | 67 | All colleges | Private nonprofit | Each college counted once |
| 434 | 68 | All colleges | Private nonprofit | Each college counted once |
| 417 | 69 | All colleges | Private nonprofit | Each college counted once |
| 402 | 70 | All colleges | Private nonprofit | Each college counted once |
| 385 | 71 | All colleges | Private nonprofit | Each college counted once |
| 366 | 72 | All colleges | Private nonprofit | Each college counted once |
| 349 | 73 | All colleges | Private nonprofit | Each college counted once |
| 333 | 74 | All colleges | Private nonprofit | Each college counted once |
| 320 | 75 | All colleges | Private nonprofit | Each college counted once |
| 306 | 76 | All colleges | Private nonprofit | Each college counted once |
| 289 | 77 | All colleges | Private nonprofit | Each college counted once |
| 268 | 78 | All colleges | Private nonprofit | Each college counted once |
| 243 | 79 | All colleges | Private nonprofit | Each college counted once |
| 219 | 80 | All colleges | Private nonprofit | Each college counted once |
| 199 | 81 | All colleges | Private nonprofit | Each college counted once |
| 184 | 82 | All colleges | Private nonprofit | Each college counted once |
| 175 | 83 | All colleges | Private nonprofit | Each college counted once |
| 169 | 84 | All colleges | Private nonprofit | Each college counted once |
| 163 | 85 | All colleges | Private nonprofit | Each college counted once |
| 157 | 86 | All colleges | Private nonprofit | Each college counted once |
| 150 | 87 | All colleges | Private nonprofit | Each college counted once |
| 143 | 88 | All colleges | Private nonprofit | Each college counted once |
| 134 | 89 | All colleges | Private nonprofit | Each college counted once |
| 122 | 90 | All colleges | Private nonprofit | Each college counted once |
| 108 | 91 | All colleges | Private nonprofit | Each college counted once |
| 92 | 92 | All colleges | Private nonprofit | Each college counted once |
| 77 | 93 | All colleges | Private nonprofit | Each college counted once |
| 62 | 94 | All colleges | Private nonprofit | Each college counted once |
| 51 | 95 | All colleges | Private nonprofit | Each college counted once |
| 45 | 96 | All colleges | Private nonprofit | Each college counted once |
| 42 | 97 | All colleges | Private nonprofit | Each college counted once |
| 44 | 98 | All colleges | Private nonprofit | Each college counted once |
| 48 | 99 | All colleges | Private nonprofit | Each college counted once |
| 54 | 100 | All colleges | Private nonprofit | Each college counted once |
| 59 | 101 | All colleges | Private nonprofit | Each college counted once |
| 61 | 102 | All colleges | Private nonprofit | Each college counted once |
| 59 | 103 | All colleges | Private nonprofit | Each college counted once |
| 54 | 104 | All colleges | Private nonprofit | Each college counted once |
| 47 | 105 | All colleges | Private nonprofit | Each college counted once |
| 39 | 106 | All colleges | Private nonprofit | Each college counted once |
| 33 | 107 | All colleges | Private nonprofit | Each college counted once |
| 29 | 108 | All colleges | Private nonprofit | Each college counted once |
| 27 | 109 | All colleges | Private nonprofit | Each college counted once |
| 25 | 110 | All colleges | Private nonprofit | Each college counted once |
| 23 | 111 | All colleges | Private nonprofit | Each college counted once |
| 20 | 112 | All colleges | Private nonprofit | Each college counted once |
| 16 | 113 | All colleges | Private nonprofit | Each college counted once |
| 12 | 114 | All colleges | Private nonprofit | Each college counted once |
| 10 | 115 | All colleges | Private nonprofit | Each college counted once |
| 8 | 116 | All colleges | Private nonprofit | Each college counted once |
| 8 | 117 | All colleges | Private nonprofit | Each college counted once |
| 8 | 118 | All colleges | Private nonprofit | Each college counted once |
| 10 | 119 | All colleges | Private nonprofit | Each college counted once |
| 12 | 120 | All colleges | Private nonprofit | Each college counted once |
| 0 | 0 | All colleges | For-profit | Each college counted once |
| 0 | 1 | All colleges | For-profit | Each college counted once |
| 0 | 2 | All colleges | For-profit | Each college counted once |
| 0 | 3 | All colleges | For-profit | Each college counted once |
| 1 | 4 | All colleges | For-profit | Each college counted once |
| 2 | 5 | All colleges | For-profit | Each college counted once |
| 4 | 6 | All colleges | For-profit | Each college counted once |
| 7 | 7 | All colleges | For-profit | Each college counted once |
| 11 | 8 | All colleges | For-profit | Each college counted once |
| 18 | 9 | All colleges | For-profit | Each college counted once |
| 29 | 10 | All colleges | For-profit | Each college counted once |
| 46 | 11 | All colleges | For-profit | Each college counted once |
| 73 | 12 | All colleges | For-profit | Each college counted once |
| 114 | 13 | All colleges | For-profit | Each college counted once |
| 170 | 14 | All colleges | For-profit | Each college counted once |
| 242 | 15 | All colleges | For-profit | Each college counted once |
| 324 | 16 | All colleges | For-profit | Each college counted once |
| 412 | 17 | All colleges | For-profit | Each college counted once |
| 502 | 18 | All colleges | For-profit | Each college counted once |
| 593 | 19 | All colleges | For-profit | Each college counted once |
| 686 | 20 | All colleges | For-profit | Each college counted once |
| 779 | 21 | All colleges | For-profit | Each college counted once |
| 866 | 22 | All colleges | For-profit | Each college counted once |
| 936 | 23 | All colleges | For-profit | Each college counted once |
| 982 | 24 | All colleges | For-profit | Each college counted once |
| 1000 | 25 | All colleges | For-profit | Each college counted once |
| 992 | 26 | All colleges | For-profit | Each college counted once |
| 963 | 27 | All colleges | For-profit | Each college counted once |
| 921 | 28 | All colleges | For-profit | Each college counted once |
| 872 | 29 | All colleges | For-profit | Each college counted once |
| 816 | 30 | All colleges | For-profit | Each college counted once |
| 756 | 31 | All colleges | For-profit | Each college counted once |
| 691 | 32 | All colleges | For-profit | Each college counted once |
| 626 | 33 | All colleges | For-profit | Each college counted once |
| 566 | 34 | All colleges | For-profit | Each college counted once |
| 516 | 35 | All colleges | For-profit | Each college counted once |
| 476 | 36 | All colleges | For-profit | Each college counted once |
| 443 | 37 | All colleges | For-profit | Each college counted once |
| 414 | 38 | All colleges | For-profit | Each college counted once |
| 388 | 39 | All colleges | For-profit | Each college counted once |
| 360 | 40 | All colleges | For-profit | Each college counted once |
| 331 | 41 | All colleges | For-profit | Each college counted once |
| 299 | 42 | All colleges | For-profit | Each college counted once |
| 265 | 43 | All colleges | For-profit | Each college counted once |
| 232 | 44 | All colleges | For-profit | Each college counted once |
| 202 | 45 | All colleges | For-profit | Each college counted once |
| 177 | 46 | All colleges | For-profit | Each college counted once |
| 156 | 47 | All colleges | For-profit | Each college counted once |
| 141 | 48 | All colleges | For-profit | Each college counted once |
| 130 | 49 | All colleges | For-profit | Each college counted once |
| 122 | 50 | All colleges | For-profit | Each college counted once |
| 114 | 51 | All colleges | For-profit | Each college counted once |
| 105 | 52 | All colleges | For-profit | Each college counted once |
| 95 | 53 | All colleges | For-profit | Each college counted once |
| 83 | 54 | All colleges | For-profit | Each college counted once |
| 72 | 55 | All colleges | For-profit | Each college counted once |
| 63 | 56 | All colleges | For-profit | Each college counted once |
| 59 | 57 | All colleges | For-profit | Each college counted once |
| 58 | 58 | All colleges | For-profit | Each college counted once |
| 58 | 59 | All colleges | For-profit | Each college counted once |
| 60 | 60 | All colleges | For-profit | Each college counted once |
| 60 | 61 | All colleges | For-profit | Each college counted once |
| 59 | 62 | All colleges | For-profit | Each college counted once |
| 56 | 63 | All colleges | For-profit | Each college counted once |
| 52 | 64 | All colleges | For-profit | Each college counted once |
| 48 | 65 | All colleges | For-profit | Each college counted once |
| 45 | 66 | All colleges | For-profit | Each college counted once |
| 43 | 67 | All colleges | For-profit | Each college counted once |
| 40 | 68 | All colleges | For-profit | Each college counted once |
| 37 | 69 | All colleges | For-profit | Each college counted once |
| 32 | 70 | All colleges | For-profit | Each college counted once |
| 27 | 71 | All colleges | For-profit | Each college counted once |
| 23 | 72 | All colleges | For-profit | Each college counted once |
| 20 | 73 | All colleges | For-profit | Each college counted once |
| 18 | 74 | All colleges | For-profit | Each college counted once |
| 18 | 75 | All colleges | For-profit | Each college counted once |
| 18 | 76 | All colleges | For-profit | Each college counted once |
| 19 | 77 | All colleges | For-profit | Each college counted once |
| 20 | 78 | All colleges | For-profit | Each college counted once |
| 22 | 79 | All colleges | For-profit | Each college counted once |
| 21 | 80 | All colleges | For-profit | Each college counted once |
| 20 | 81 | All colleges | For-profit | Each college counted once |
| 17 | 82 | All colleges | For-profit | Each college counted once |
| 14 | 83 | All colleges | For-profit | Each college counted once |
| 12 | 84 | All colleges | For-profit | Each college counted once |
| 11 | 85 | All colleges | For-profit | Each college counted once |
| 11 | 86 | All colleges | For-profit | Each college counted once |
| 10 | 87 | All colleges | For-profit | Each college counted once |
| 10 | 88 | All colleges | For-profit | Each college counted once |
| 9 | 89 | All colleges | For-profit | Each college counted once |
| 8 | 90 | All colleges | For-profit | Each college counted once |
| 7 | 91 | All colleges | For-profit | Each college counted once |
| 8 | 92 | All colleges | For-profit | Each college counted once |
| 8 | 93 | All colleges | For-profit | Each college counted once |
| 7 | 94 | All colleges | For-profit | Each college counted once |
| 7 | 95 | All colleges | For-profit | Each college counted once |
| 6 | 96 | All colleges | For-profit | Each college counted once |
| 5 | 97 | All colleges | For-profit | Each college counted once |
| 5 | 98 | All colleges | For-profit | Each college counted once |
| 5 | 99 | All colleges | For-profit | Each college counted once |
| 5 | 100 | All colleges | For-profit | Each college counted once |
| 6 | 101 | All colleges | For-profit | Each college counted once |
| 7 | 102 | All colleges | For-profit | Each college counted once |
| 9 | 103 | All colleges | For-profit | Each college counted once |
| 10 | 104 | All colleges | For-profit | Each college counted once |
| 11 | 105 | All colleges | For-profit | Each college counted once |
| 10 | 106 | All colleges | For-profit | Each college counted once |
| 9 | 107 | All colleges | For-profit | Each college counted once |
| 7 | 108 | All colleges | For-profit | Each college counted once |
| 5 | 109 | All colleges | For-profit | Each college counted once |
| 4 | 110 | All colleges | For-profit | Each college counted once |
| 3 | 111 | All colleges | For-profit | Each college counted once |
| 3 | 112 | All colleges | For-profit | Each college counted once |
| 3 | 113 | All colleges | For-profit | Each college counted once |
| 3 | 114 | All colleges | For-profit | Each college counted once |
| 3 | 115 | All colleges | For-profit | Each college counted once |
| 3 | 116 | All colleges | For-profit | Each college counted once |
| 2 | 117 | All colleges | For-profit | Each college counted once |
| 1 | 118 | All colleges | For-profit | Each college counted once |
| 1 | 119 | All colleges | For-profit | Each college counted once |
| 0 | 120 | All colleges | For-profit | Each college counted once |
| 0 | 0 | All colleges | Public | Weighted by undergraduate enrollment |
| 0 | 1 | All colleges | Public | Weighted by undergraduate enrollment |
| 0 | 2 | All colleges | Public | Weighted by undergraduate enrollment |
| 0 | 3 | All colleges | Public | Weighted by undergraduate enrollment |
| 0 | 4 | All colleges | Public | Weighted by undergraduate enrollment |
| 0 | 5 | All colleges | Public | Weighted by undergraduate enrollment |
| 0 | 6 | All colleges | Public | Weighted by undergraduate enrollment |
| 0 | 7 | All colleges | Public | Weighted by undergraduate enrollment |
| 0 | 8 | All colleges | Public | Weighted by undergraduate enrollment |
| 0 | 9 | All colleges | Public | Weighted by undergraduate enrollment |
| 0 | 10 | All colleges | Public | Weighted by undergraduate enrollment |
| 0 | 11 | All colleges | Public | Weighted by undergraduate enrollment |
| 0 | 12 | All colleges | Public | Weighted by undergraduate enrollment |
| 0 | 13 | All colleges | Public | Weighted by undergraduate enrollment |
| 0 | 14 | All colleges | Public | Weighted by undergraduate enrollment |
| 0 | 15 | All colleges | Public | Weighted by undergraduate enrollment |
| 0 | 16 | All colleges | Public | Weighted by undergraduate enrollment |
| 0 | 17 | All colleges | Public | Weighted by undergraduate enrollment |
| 1 | 18 | All colleges | Public | Weighted by undergraduate enrollment |
| 1 | 19 | All colleges | Public | Weighted by undergraduate enrollment |
| 1 | 20 | All colleges | Public | Weighted by undergraduate enrollment |
| 2 | 21 | All colleges | Public | Weighted by undergraduate enrollment |
| 2 | 22 | All colleges | Public | Weighted by undergraduate enrollment |
| 3 | 23 | All colleges | Public | Weighted by undergraduate enrollment |
| 5 | 24 | All colleges | Public | Weighted by undergraduate enrollment |
| 9 | 25 | All colleges | Public | Weighted by undergraduate enrollment |
| 15 | 26 | All colleges | Public | Weighted by undergraduate enrollment |
| 26 | 27 | All colleges | Public | Weighted by undergraduate enrollment |
| 43 | 28 | All colleges | Public | Weighted by undergraduate enrollment |
| 69 | 29 | All colleges | Public | Weighted by undergraduate enrollment |
| 104 | 30 | All colleges | Public | Weighted by undergraduate enrollment |
| 152 | 31 | All colleges | Public | Weighted by undergraduate enrollment |
| 213 | 32 | All colleges | Public | Weighted by undergraduate enrollment |
| 289 | 33 | All colleges | Public | Weighted by undergraduate enrollment |
| 380 | 34 | All colleges | Public | Weighted by undergraduate enrollment |
| 483 | 35 | All colleges | Public | Weighted by undergraduate enrollment |
| 594 | 36 | All colleges | Public | Weighted by undergraduate enrollment |
| 703 | 37 | All colleges | Public | Weighted by undergraduate enrollment |
| 806 | 38 | All colleges | Public | Weighted by undergraduate enrollment |
| 893 | 39 | All colleges | Public | Weighted by undergraduate enrollment |
| 960 | 40 | All colleges | Public | Weighted by undergraduate enrollment |
| 998 | 41 | All colleges | Public | Weighted by undergraduate enrollment |
| 1000 | 42 | All colleges | Public | Weighted by undergraduate enrollment |
| 967 | 43 | All colleges | Public | Weighted by undergraduate enrollment |
| 905 | 44 | All colleges | Public | Weighted by undergraduate enrollment |
| 830 | 45 | All colleges | Public | Weighted by undergraduate enrollment |
| 754 | 46 | All colleges | Public | Weighted by undergraduate enrollment |
| 688 | 47 | All colleges | Public | Weighted by undergraduate enrollment |
| 634 | 48 | All colleges | Public | Weighted by undergraduate enrollment |
| 593 | 49 | All colleges | Public | Weighted by undergraduate enrollment |
| 563 | 50 | All colleges | Public | Weighted by undergraduate enrollment |
| 545 | 51 | All colleges | Public | Weighted by undergraduate enrollment |
| 540 | 52 | All colleges | Public | Weighted by undergraduate enrollment |
| 546 | 53 | All colleges | Public | Weighted by undergraduate enrollment |
| 562 | 54 | All colleges | Public | Weighted by undergraduate enrollment |
| 581 | 55 | All colleges | Public | Weighted by undergraduate enrollment |
| 596 | 56 | All colleges | Public | Weighted by undergraduate enrollment |
| 602 | 57 | All colleges | Public | Weighted by undergraduate enrollment |
| 597 | 58 | All colleges | Public | Weighted by undergraduate enrollment |
| 586 | 59 | All colleges | Public | Weighted by undergraduate enrollment |
| 574 | 60 | All colleges | Public | Weighted by undergraduate enrollment |
| 562 | 61 | All colleges | Public | Weighted by undergraduate enrollment |
| 545 | 62 | All colleges | Public | Weighted by undergraduate enrollment |
| 518 | 63 | All colleges | Public | Weighted by undergraduate enrollment |
| 476 | 64 | All colleges | Public | Weighted by undergraduate enrollment |
| 424 | 65 | All colleges | Public | Weighted by undergraduate enrollment |
| 372 | 66 | All colleges | Public | Weighted by undergraduate enrollment |
| 328 | 67 | All colleges | Public | Weighted by undergraduate enrollment |
| 298 | 68 | All colleges | Public | Weighted by undergraduate enrollment |
| 279 | 69 | All colleges | Public | Weighted by undergraduate enrollment |
| 267 | 70 | All colleges | Public | Weighted by undergraduate enrollment |
| 256 | 71 | All colleges | Public | Weighted by undergraduate enrollment |
| 242 | 72 | All colleges | Public | Weighted by undergraduate enrollment |
| 223 | 73 | All colleges | Public | Weighted by undergraduate enrollment |
| 198 | 74 | All colleges | Public | Weighted by undergraduate enrollment |
| 171 | 75 | All colleges | Public | Weighted by undergraduate enrollment |
| 145 | 76 | All colleges | Public | Weighted by undergraduate enrollment |
| 125 | 77 | All colleges | Public | Weighted by undergraduate enrollment |
| 114 | 78 | All colleges | Public | Weighted by undergraduate enrollment |
| 113 | 79 | All colleges | Public | Weighted by undergraduate enrollment |
| 117 | 80 | All colleges | Public | Weighted by undergraduate enrollment |
| 120 | 81 | All colleges | Public | Weighted by undergraduate enrollment |
| 118 | 82 | All colleges | Public | Weighted by undergraduate enrollment |
| 108 | 83 | All colleges | Public | Weighted by undergraduate enrollment |
| 93 | 84 | All colleges | Public | Weighted by undergraduate enrollment |
| 74 | 85 | All colleges | Public | Weighted by undergraduate enrollment |
| 55 | 86 | All colleges | Public | Weighted by undergraduate enrollment |
| 41 | 87 | All colleges | Public | Weighted by undergraduate enrollment |
| 31 | 88 | All colleges | Public | Weighted by undergraduate enrollment |
| 27 | 89 | All colleges | Public | Weighted by undergraduate enrollment |
| 26 | 90 | All colleges | Public | Weighted by undergraduate enrollment |
| 26 | 91 | All colleges | Public | Weighted by undergraduate enrollment |
| 25 | 92 | All colleges | Public | Weighted by undergraduate enrollment |
| 22 | 93 | All colleges | Public | Weighted by undergraduate enrollment |
| 18 | 94 | All colleges | Public | Weighted by undergraduate enrollment |
| 13 | 95 | All colleges | Public | Weighted by undergraduate enrollment |
| 10 | 96 | All colleges | Public | Weighted by undergraduate enrollment |
| 7 | 97 | All colleges | Public | Weighted by undergraduate enrollment |
| 6 | 98 | All colleges | Public | Weighted by undergraduate enrollment |
| 5 | 99 | All colleges | Public | Weighted by undergraduate enrollment |
| 6 | 100 | All colleges | Public | Weighted by undergraduate enrollment |
| 7 | 101 | All colleges | Public | Weighted by undergraduate enrollment |
| 8 | 102 | All colleges | Public | Weighted by undergraduate enrollment |
| 8 | 103 | All colleges | Public | Weighted by undergraduate enrollment |
| 7 | 104 | All colleges | Public | Weighted by undergraduate enrollment |
| 5 | 105 | All colleges | Public | Weighted by undergraduate enrollment |
| 3 | 106 | All colleges | Public | Weighted by undergraduate enrollment |
| 2 | 107 | All colleges | Public | Weighted by undergraduate enrollment |
| 1 | 108 | All colleges | Public | Weighted by undergraduate enrollment |
| 0 | 109 | All colleges | Public | Weighted by undergraduate enrollment |
| 0 | 110 | All colleges | Public | Weighted by undergraduate enrollment |
| 0 | 111 | All colleges | Public | Weighted by undergraduate enrollment |
| 0 | 112 | All colleges | Public | Weighted by undergraduate enrollment |
| 0 | 113 | All colleges | Public | Weighted by undergraduate enrollment |
| 0 | 114 | All colleges | Public | Weighted by undergraduate enrollment |
| 0 | 115 | All colleges | Public | Weighted by undergraduate enrollment |
| 0 | 116 | All colleges | Public | Weighted by undergraduate enrollment |
| 0 | 117 | All colleges | Public | Weighted by undergraduate enrollment |
| 0 | 118 | All colleges | Public | Weighted by undergraduate enrollment |
| 0 | 119 | All colleges | Public | Weighted by undergraduate enrollment |
| 0 | 120 | All colleges | Public | Weighted by undergraduate enrollment |
| 0 | 0 | All colleges | Private nonprofit | Weighted by undergraduate enrollment |
| 0 | 1 | All colleges | Private nonprofit | Weighted by undergraduate enrollment |
| 0 | 2 | All colleges | Private nonprofit | Weighted by undergraduate enrollment |
| 0 | 3 | All colleges | Private nonprofit | Weighted by undergraduate enrollment |
| 0 | 4 | All colleges | Private nonprofit | Weighted by undergraduate enrollment |
| 0 | 5 | All colleges | Private nonprofit | Weighted by undergraduate enrollment |
| 0 | 6 | All colleges | Private nonprofit | Weighted by undergraduate enrollment |
| 0 | 7 | All colleges | Private nonprofit | Weighted by undergraduate enrollment |
| 0 | 8 | All colleges | Private nonprofit | Weighted by undergraduate enrollment |
| 0 | 9 | All colleges | Private nonprofit | Weighted by undergraduate enrollment |
| 0 | 10 | All colleges | Private nonprofit | Weighted by undergraduate enrollment |
| 0 | 11 | All colleges | Private nonprofit | Weighted by undergraduate enrollment |
| 0 | 12 | All colleges | Private nonprofit | Weighted by undergraduate enrollment |
| 0 | 13 | All colleges | Private nonprofit | Weighted by undergraduate enrollment |
| 1 | 14 | All colleges | Private nonprofit | Weighted by undergraduate enrollment |
| 2 | 15 | All colleges | Private nonprofit | Weighted by undergraduate enrollment |

The first 500 of 1452 rows; every row is in https://vishalsingh.org/zipcode/report/college-access-price-and-earnings/figures/earnings_by_sector.json.

Views: All colleges, each counted once (scope = All colleges, weighting = Each college counted once); All colleges, weighted by enrollment (scope = All colleges, weighting = Weighted by undergraduate enrollment); Four-year colleges (scope = Four-year colleges, weighting = Each college counted once); Two-year and shorter programs (scope = Two-year and shorter programs, weighting = Each college counted once).

Note: Earnings are the median for students who received federal aid and were working and not enrolled a decade after entry. They are in 2022 dollars, and the Scorecard adds noise to them and withholds them for small groups, so a college with no published value is missing, never zero. Densities use one smoothing bandwidth for every ridge and run from zero to a ceiling; a few colleges above the ceiling are outside the plot.

Note: Colleges are institutions identified by a 6-digit OPE ID. Branch campuses repeat their parent's earnings, so one row per OPE ID is used, and 5,000 institutions are in the file in the 50 states and DC. Earnings are published for 3,963 of them, and the share with no value is 4.7% of public, 26.5% of private nonprofit and 33.2% of for-profit colleges.

Note: The medians describe colleges and the students who received federal aid, and they mix what a college does with who enrolls and where it sits.

Source: College Scorecard, most recent cohorts, May 2025 release (earnings of the 2009-2011 entering classes, 2022 dollars); unweighted unless the view says weighted by undergraduate enrollment

Data: https://vishalsingh.org/zipcode/report/college-access-price-and-earnings/figures/earnings_by_sector.json

### Table 2. The median net price rises with family income, but at public four-year colleges it falls from 62% of an assumed income in the lowest band to 13% in the highest

Median across colleges of the average net price after grants for full-time, first-time students who received federal aid, by family-income band, public and private nonprofit four-year colleges, academic year 2021 to 2022

| band | hi | lo | measure | n | sector | value |
| ---: | ---: | ---: | --- | ---: | --- | ---: |
| 1 | 10430 | 8426 | Median net price (dollars per year) | 693 | Public four-year colleges | 9347 |
| 1 | 69.50 | 56.20 | Net price as a share of assumed income (percent) | 693 | Public four-year colleges | 62.30 |
| 2 | 11263 | 9449 | Median net price (dollars per year) | 685 | Public four-year colleges | 10114 |
| 2 | 28.90 | 24.20 | Net price as a share of assumed income (percent) | 685 | Public four-year colleges | 25.90 |
| 3 | 14074 | 12130 | Median net price (dollars per year) | 687 | Public four-year colleges | 12857 |
| 3 | 22.90 | 19.70 | Net price as a share of assumed income (percent) | 687 | Public four-year colleges | 20.90 |
| 4 | 18000 | 16056 | Median net price (dollars per year) | 670 | Public four-year colleges | 16974 |
| 4 | 19.50 | 17.40 | Net price as a share of assumed income (percent) | 670 | Public four-year colleges | 18.40 |
| 5 | 19987 | 17910 | Median net price (dollars per year) | 647 | Public four-year colleges | 18780 |
| 5 | 13.30 | 11.90 | Net price as a share of assumed income (percent) | 647 | Public four-year colleges | 12.50 |
| 1 | 18307 | 16937 | Median net price (dollars per year) | 1,120 | Private nonprofit four-year colleges | 17596 |
| 1 | 122 | 112.9 | Net price as a share of assumed income (percent) | 1,120 | Private nonprofit four-year colleges | 117.3 |
| 2 | 18742 | 17413 | Median net price (dollars per year) | 1,092 | Private nonprofit four-year colleges | 18166 |
| 2 | 48.10 | 44.60 | Net price as a share of assumed income (percent) | 1,092 | Private nonprofit four-year colleges | 46.60 |
| 3 | 21427 | 19926 | Median net price (dollars per year) | 1,082 | Private nonprofit four-year colleges | 20684 |
| 3 | 34.80 | 32.40 | Net price as a share of assumed income (percent) | 1,082 | Private nonprofit four-year colleges | 33.60 |
| 4 | 25256 | 23554 | Median net price (dollars per year) | 1,044 | Private nonprofit four-year colleges | 24314 |
| 4 | 27.30 | 25.50 | Net price as a share of assumed income (percent) | 1,044 | Private nonprofit four-year colleges | 26.30 |
| 5 | 30525 | 26953 | Median net price (dollars per year) | 976 | Private nonprofit four-year colleges | 28710 |
| 5 | 20.30 | 18 | Net price as a share of assumed income (percent) | 976 | Private nonprofit four-year colleges | 19.10 |

Views: Dollars per year (measure = Median net price (dollars per year)); Share of assumed income (measure = Net price as a share of assumed income (percent)).

Note: Net price is the cost of attendance, which includes living expenses, minus grants and scholarships, averaged over a college's full-time, first-time students who received federal aid in the band. At public colleges it covers in-state students only. A family bears the living expenses whether or not the student enrolls, so the share of income is an order of magnitude and not a measured burden on any family.

Note: The five bands are the Scorecard's, up to $30,000 of family income and over $110,000. The Scorecard does not report the income of the families in a band, so the share assumes an income at the midpoint of each band, and the highest band has no upper limit, so its share moves in proportion to the assumed value.

Note: The medians cover 694 public and 1,157 private nonprofit four-year colleges with a published net price in at least one band; a few report a net price below zero, which is kept as reported.

Source: College Scorecard, most recent cohorts, May 2025 release (net price by family income, academic year 2021-2022, nominal dollars); unweighted, each college counted once

Data: https://vishalsingh.org/zipcode/report/college-access-price-and-earnings/figures/net_price_by_income.json

### Table 3. Former students of costlier colleges tend to earn more, a correlation of +0.63, but 56% of plotted for-profit colleges cost more than the median and have median earnings at or below it

Median earnings a decade after entry (2022 dollars) against the average annual cost of attendance (academic year 2021 to 2022), one dot per college, dot area is undergraduate enrollment

No inline rows: the values of Table 3 are in https://vishalsingh.org/zipcode/data/college-access-price-and-earnings/earnings_vs_cost.json.

Note: The plot has 2,663 colleges with a published cost of attendance, median earnings and at least 100 undergraduates. The cost of attendance is published for academic-year institutions only, so many short for-profit programs are not in it, and for-profit colleges are a small part of the plot.

Note: The quadrant lines are medians and the intervals for the shares come from a bootstrap over states that redraws the medians in every replicate. Dollar axes mix academic year 2021 to 2022 costs with earnings in 2022 dollars, which are close in time.

Source: College Scorecard, most recent cohorts, May 2025 release (cost of attendance, academic year 2021-2022; earnings of the 2009-2011 entering classes); unweighted, each college counted once

Data: https://vishalsingh.org/zipcode/report/college-access-price-and-earnings/figures/earnings_vs_cost.json

### Table 4. Among high-Pell colleges, the 15 highest-earning have a median of $72,282 a decade after entry and the 15 lowest-earning $31,902

Public and private nonprofit four-year colleges in the top third by Pell share, with at least 1,000 undergraduates; median earnings of former students who received federal aid and net price for the lowest income band

#### Table 4.1. Median earnings

Most of the highest-earning high-Pell colleges are in California, New York and New Jersey

Median earnings of former students who received federal aid a decade after entry (2022 dollars), the 15 highest-earning and the 15 lowest-earning high-Pell colleges

| College | List | Median earnings (2022 dollars) | Pell share (%) | Net price, lowest income band (dollars) |
| --- | --- | ---: | ---: | ---: |
| New Jersey Institute of Technology (NJ) | Highest-earning | $84,276 | 39.0 | $11,534 |
| University of California-Irvine (CA) | Highest-earning | $80,735 | 37.3 | $6,485 |
| San Jose State University (CA) | Highest-earning | $78,988 | 35.4 | $9,743 |
| CUNY Bernard M Baruch College (NY) | Highest-earning | $75,971 | 55.3 | $745 |
| Stony Brook University (NY) | Highest-earning | $74,502 | 38.4 | $12,744 |
| University of Redlands (CA) | Highest-earning | $72,690 | 35.1 | $21,520 |
| Mount Saint Mary's University (CA) | Highest-earning | $72,379 | 55.1 | $17,327 |
| AdventHealth University (FL) | Highest-earning | $72,282 | 37.4 | $29,906 |
| California State Polytechnic University-Pomona (CA) | Highest-earning | $71,902 | 46.0 | $7,599 |
| California State University-East Bay (CA) | Highest-earning | $71,401 | 43.6 | $8,059 |
| New York Institute of Technology (NY) | Highest-earning | $70,080 | 43.7 | $16,548 |
| Farmingdale State College (NY) | Highest-earning | $69,781 | 35.8 | $3,865 |
| University of Massachusetts-Dartmouth (MA) | Highest-earning | $68,804 | 35.6 | $14,122 |
| University of Illinois Chicago (IL) | Highest-earning | $68,740 | 49.6 | $9,257 |
| San Francisco State University (CA) | Highest-earning | $68,077 | 41.4 | $9,307 |
| Laredo College (TX) | Lowest-earning | $33,934 | 38.5 | $4,877 |
| Central State University (OH) | Lowest-earning | $33,267 | 42.2 | $15,607 |
| Langston University (OK) | Lowest-earning | $33,261 | 68.4 | $10,626 |
| Atlanta Metropolitan State College (GA) | Lowest-earning | $33,252 | 53.8 | $4,415 |
| Miles College (AL) | Lowest-earning | $32,627 | 75.5 | $13,517 |
| Bryant & Stratton College-Online (NY) | Lowest-earning | $32,568 | 75.2 | $14,685 |
| Mississippi Valley State University (MS) | Lowest-earning | $31,919 | 59.6 | $11,034 |
| Benedict College (SC) | Lowest-earning | $31,902 | 63.5 | $16,807 |
| Uta Mesivta of Kiryas Joel (NY) | Lowest-earning | $31,853 | 86.2 | $4,542 |
| Harris-Stowe State University (MO) | Lowest-earning | $31,088 | 72.9 | $12,412 |
| Dine College (AZ) | Lowest-earning | $29,188 | 39.0 | $10,031 |
| American Musical and Dramatic Academy (NY) | Lowest-earning | $26,975 | 36.0 | $25,066 |
| Navajo Technical University (NM) | Lowest-earning | $26,364 | 49.0 | $3,037 |
| United Talmudical Seminary (NY) | Lowest-earning | $25,113 | 85.6 | $6,126 |
| Oglala Lakota College (SD) | Lowest-earning | $22,517 | 46.9 | $2,188 |

#### Table 4.2. Net price, lowest income band

The highest-earning and lowest-earning high-Pell colleges charge the lowest income band similar median net prices

Average net price after grants for full-time, first-time students in the lowest family-income band who received federal aid, the same colleges, academic year 2021 to 2022

| College | List | Median earnings (2022 dollars) | Pell share (%) | Net price, lowest income band (dollars) |
| --- | --- | ---: | ---: | ---: |
| New Jersey Institute of Technology (NJ) | Highest-earning | $84,276 | 39.0 | $11,534 |
| University of California-Irvine (CA) | Highest-earning | $80,735 | 37.3 | $6,485 |
| San Jose State University (CA) | Highest-earning | $78,988 | 35.4 | $9,743 |
| CUNY Bernard M Baruch College (NY) | Highest-earning | $75,971 | 55.3 | $745 |
| Stony Brook University (NY) | Highest-earning | $74,502 | 38.4 | $12,744 |
| University of Redlands (CA) | Highest-earning | $72,690 | 35.1 | $21,520 |
| Mount Saint Mary's University (CA) | Highest-earning | $72,379 | 55.1 | $17,327 |
| AdventHealth University (FL) | Highest-earning | $72,282 | 37.4 | $29,906 |
| California State Polytechnic University-Pomona (CA) | Highest-earning | $71,902 | 46.0 | $7,599 |
| California State University-East Bay (CA) | Highest-earning | $71,401 | 43.6 | $8,059 |
| New York Institute of Technology (NY) | Highest-earning | $70,080 | 43.7 | $16,548 |
| Farmingdale State College (NY) | Highest-earning | $69,781 | 35.8 | $3,865 |
| University of Massachusetts-Dartmouth (MA) | Highest-earning | $68,804 | 35.6 | $14,122 |
| University of Illinois Chicago (IL) | Highest-earning | $68,740 | 49.6 | $9,257 |
| San Francisco State University (CA) | Highest-earning | $68,077 | 41.4 | $9,307 |
| Laredo College (TX) | Lowest-earning | $33,934 | 38.5 | $4,877 |
| Central State University (OH) | Lowest-earning | $33,267 | 42.2 | $15,607 |
| Langston University (OK) | Lowest-earning | $33,261 | 68.4 | $10,626 |
| Atlanta Metropolitan State College (GA) | Lowest-earning | $33,252 | 53.8 | $4,415 |
| Miles College (AL) | Lowest-earning | $32,627 | 75.5 | $13,517 |
| Bryant & Stratton College-Online (NY) | Lowest-earning | $32,568 | 75.2 | $14,685 |
| Mississippi Valley State University (MS) | Lowest-earning | $31,919 | 59.6 | $11,034 |
| Benedict College (SC) | Lowest-earning | $31,902 | 63.5 | $16,807 |
| Uta Mesivta of Kiryas Joel (NY) | Lowest-earning | $31,853 | 86.2 | $4,542 |
| Harris-Stowe State University (MO) | Lowest-earning | $31,088 | 72.9 | $12,412 |
| Dine College (AZ) | Lowest-earning | $29,188 | 39.0 | $10,031 |
| American Musical and Dramatic Academy (NY) | Lowest-earning | $26,975 | 36.0 | $25,066 |
| Navajo Technical University (NM) | Lowest-earning | $26,364 | 49.0 | $3,037 |
| United Talmudical Seminary (NY) | Lowest-earning | $25,113 | 85.6 | $6,126 |
| Oglala Lakota College (SD) | Lowest-earning | $22,517 | 46.9 | $2,188 |

Note: For-profit colleges are not on the list. Only public and private nonprofit colleges enter the pool, so no for-profit college can appear in either list, and 48 for-profit four-year colleges that otherwise met the rules were left out. A college is high-Pell when its Pell share reaches 34.7%, the cutoff for the top third of the pool of 1,282 colleges.

Note: The lists are rankings of named colleges and not samples, so their medians carry no interval, and the Scorecard adds noise to each college's earnings. Branch campuses share their parent's earnings, so each college appears once.

Note: Some of the lowest-earning colleges are tribal colleges or religious seminaries whose graduates' earnings are not comparable with those of a general four-year degree, and several are historically Black colleges and universities; earnings also follow local wages, which are high in the New York, New Jersey and California areas, and the measure does not adjust for the students' starting income.

Source: College Scorecard, most recent cohorts, May 2025 release (Pell share and net price, academic year 2021-2022; earnings of the 2009-2011 entering classes, 2022 dollars); unweighted, each college counted once

Data: https://vishalsingh.org/zipcode/report/college-access-price-and-earnings/figures/high_pell_colleges.json

### Table 5. The average resident in the country lives 9.6 miles from a four-year college, and the state averages are longest in Alaska, Wyoming, Montana and Mississippi

Distance in miles from the average resident of each county to the nearest four-year campus (great-circle, from each census tract's internal point, weighted by tract population), campuses of 2023, residents of 2019 to 2023

No inline rows: the values of Table 5 are in https://vishalsingh.org/zipcode/data/college-access-price-and-earnings/college_distance_map.json.

Views: Any four-year campus (every row); Public four-year campus (every row).

Note: Every county on the map has a value: each has residents, and the distance is measured to a campus in any of the 50 states or DC. Puerto Rico is not drawn and its campuses and residents are not in the data. Alaska and Hawaii distances cross water, so a county on an island or in the bush has a distance that no road or ferry covers.

Note: Four-year means a campus the IPEDS directory lists at the four-year level, leaving out the 273 institutions the Scorecard classes as entirely graduate-degree granting; 2,508 campuses remain, with branch campuses counted at their own addresses. Connecticut is drawn as its eight legacy counties, with each tract and campus assigned to its legacy county.

Note: The second view measures the distance to the nearest public four-year campus. County values are full counts of tracts and campuses and carry no sampling interval; the intervals in the gradient figure treat counties as draws. Distance is a measure of supply, not of where residents enroll.

Source: IPEDS 2023 directory (campus locations), 2020 Census tract internal points, American Community Survey 2019-2023 (tract population); population-weighted, full counts

Data: https://vishalsingh.org/zipcode/report/college-access-price-and-earnings/figures/college_distance_map.json

### Table 6. Distance to a four-year college rises with county deprivation, from 6.1 miles in the least deprived tenth of counties to 29.3 in the most deprived, and with rurality

Miles from the average resident to the nearest four-year campus, and the share of counties with no four-year campus, by tenth of county Area Deprivation Index rank and by Rural-Urban Continuum Code, residents of 2019 to 2023

#### Table 6.1. Distance by deprivation

Distance to the nearest four-year campus rises across the tenths of county deprivation, from 6.1 miles to 29.3

Miles from the average resident to the nearest four-year campus, by tenth of county ADI national rank, with 95% intervals from a bootstrap over states

| conf_high | conf_low | miles | n | tenth |
| ---: | ---: | ---: | ---: | ---: |
| 6.91 | 5.27 | 6.10 | 312 | 1 |
| 10.37 | 7.33 | 8.53 | 313 | 2 |
| 12.34 | 7.70 | 9.36 | 313 | 3 |
| 11.67 | 7.94 | 9.54 | 312 | 4 |
| 20.68 | 10.77 | 14.49 | 314 | 5 |
| 18.94 | 12.67 | 15.25 | 312 | 6 |
| 21.96 | 14.73 | 17.97 | 313 | 7 |
| 24.30 | 18.35 | 21.28 | 315 | 8 |
| 29.50 | 22.45 | 25.60 | 312 | 9 |
| 35.02 | 24.64 | 29.34 | 312 | 10 |

#### Table 6.2. Counties with no campus

The share of counties with no four-year campus rises across the tenths of county deprivation, from 42% to 92%

Percent of counties with no four-year campus inside the county, by tenth of county ADI national rank, each county counted once, with 95% intervals from a bootstrap over states

| conf_high | conf_low | n | share | tenth |
| ---: | ---: | ---: | ---: | ---: |
| 50.90 | 33 | 312 | 42.30 | 1 |
| 58.90 | 46.80 | 313 | 53.40 | 2 |
| 65.50 | 49.40 | 313 | 57.50 | 3 |
| 67.30 | 56 | 312 | 61.50 | 4 |
| 78.80 | 65.30 | 314 | 72.30 | 5 |
| 79.80 | 66.70 | 312 | 73.40 | 6 |
| 81.40 | 69.10 | 313 | 75.70 | 7 |
| 88.90 | 79 | 315 | 84.40 | 8 |
| 91.90 | 86.50 | 312 | 89.40 | 9 |
| 94.90 | 87.20 | 312 | 91.70 | 10 |

#### Table 6.3. Distance by rurality

The average resident of a nonmetro county lives 3.7 times as far from a four-year campus as the average resident of a metro county

Miles from the average resident to the nearest four-year campus, by group of USDA Rural-Urban Continuum Codes (2023), with 95% intervals from a bootstrap over states

| conf_high | conf_low | group | miles | n |
| ---: | ---: | --- | ---: | ---: |
| 7.82 | 6.33 | Metro | 7.03 | 1,185 |
| 22.92 | 16.90 | Nonmetro, urban population 20,000 or more | 19.56 | 277 |
| 30.75 | 23.43 | Nonmetro, urban population 5,000 to 20,000 | 26.63 | 626 |
| 40.93 | 29.66 | Nonmetro, urban population under 5,000 | 34.28 | 1,055 |

Note: Tracts and campuses are enumerated, not sampled, so the intervals describe how much an estimate varies from state to state and not sampling error. The rural-urban groups use all counties; the tenths use the counties that have an ADI rank.

Note: 3,128 counties have an ADI rank; 15 counties whose tracts all have a suppressed rank are not in the tenths. The ADI is a national percentile with 1 least deprived, and it understates deprivation in high-cost areas.

Note: The correlation across counties between the ADI rank and the distance is +0.31 when counties count in proportion to their residents and +0.16 when each counts once, with intervals from a bootstrap over states. Connecticut counties take the Rural-Urban Continuum Code held by most of their residents, because the code is published for the planning regions that replaced the counties.

Note: Across all the counties in the tenths the average resident lives 9.6 miles from the nearest four-year campus.

Source: IPEDS 2023 directory, American Community Survey 2019-2023 (residents), Area Deprivation Index 2023, USDA Rural-Urban Continuum Codes 2023; population-weighted unless the tab gives a share of counties

Data: https://vishalsingh.org/zipcode/report/college-access-price-and-earnings/figures/college_desert_gradient.json

### Table 7. Distance to the nearest four-year college is steadily worse toward the more deprived end under the ADI and under each of the three other rankings

Miles to the nearest four-year campus, average resident by tenth of counties ranked four ways, over the 3,125 counties that have all four rankings

| bin | conf_high | conf_low | estimate | n_units | ranking |
| ---: | ---: | ---: | ---: | ---: | --- |
| 1 | 6.97 | 5.25 | 6.10 | 312 | Area Deprivation Index |
| 2 | 10.07 | 7.38 | 8.50 | 313 | Area Deprivation Index |
| 3 | 12.04 | 7.35 | 9.01 | 312 | Area Deprivation Index |
| 4 | 12.02 | 8.69 | 10.11 | 313 | Area Deprivation Index |
| 5 | 21.70 | 11.02 | 15.14 | 312 | Area Deprivation Index |
| 6 | 17.00 | 11.49 | 13.61 | 313 | Area Deprivation Index |
| 7 | 22.71 | 15.46 | 18.69 | 312 | Area Deprivation Index |
| 8 | 23.71 | 17.75 | 20.68 | 313 | Area Deprivation Index |
| 9 | 27.26 | 21.99 | 24.61 | 312 | Area Deprivation Index |
| 10 | 35.54 | 24.84 | 29.72 | 313 | Area Deprivation Index |
| 1 | 7.59 | 6.21 | 6.78 | 312 | Standardized index |
| 2 | 10.42 | 6.99 | 8.55 | 313 | Standardized index |
| 3 | 9.54 | 5.16 | 6.72 | 312 | Standardized index |
| 4 | 9.99 | 6.95 | 8.12 | 313 | Standardized index |
| 5 | 12.86 | 7.47 | 9.53 | 312 | Standardized index |
| 6 | 20.40 | 12.29 | 15.35 | 313 | Standardized index |
| 7 | 17.79 | 11.09 | 14.07 | 312 | Standardized index |
| 8 | 28.67 | 16.75 | 21.02 | 313 | Standardized index |
| 9 | 29.84 | 15.89 | 22.81 | 312 | Standardized index |
| 10 | 41.37 | 22.40 | 28.24 | 313 | Standardized index |
| 1 | 7.28 | 6.02 | 6.66 | 312 | Median household income |
| 2 | 8.20 | 5.48 | 6.56 | 313 | Median household income |
| 3 | 11.33 | 7.59 | 8.83 | 312 | Median household income |
| 4 | 12.80 | 8.18 | 9.87 | 313 | Median household income |
| 5 | 14.15 | 9.97 | 11.66 | 312 | Median household income |
| 6 | 19.99 | 10.28 | 13.95 | 313 | Median household income |
| 7 | 25.29 | 14.41 | 18.71 | 312 | Median household income |
| 8 | 25.05 | 18.32 | 21.51 | 313 | Median household income |
| 9 | 28.15 | 22.70 | 25.10 | 312 | Median household income |
| 10 | 31.68 | 22.83 | 26.64 | 313 | Median household income |
| 1 | 6.17 | 5.02 | 5.54 | 312 | Bachelor's degree share |
| 2 | 7.46 | 5.73 | 6.42 | 313 | Bachelor's degree share |
| 3 | 10.01 | 5.74 | 7.32 | 312 | Bachelor's degree share |
| 4 | 13.83 | 9.22 | 11.21 | 313 | Bachelor's degree share |
| 5 | 17.07 | 11.58 | 13.90 | 312 | Bachelor's degree share |
| 6 | 20.99 | 14.48 | 17.20 | 313 | Bachelor's degree share |
| 7 | 25.63 | 17.09 | 20.62 | 312 | Bachelor's degree share |
| 8 | 33.93 | 21.70 | 26.86 | 313 | Bachelor's degree share |
| 9 | 33.53 | 21.92 | 26.83 | 312 | Bachelor's degree share |
| 10 | 40.50 | 25.44 | 31.03 | 313 | Bachelor's degree share |

Note: Least and most deprived tenth: ADI 6.1 and 29.7, rank correlation +0.99; standardized index 6.8 and 28.2, +0.93; median household income 6.7 and 26.6, +0.99; bachelor's degree share 5.5 and 31.0, +0.99.

Note: The standardized index applies the ADI's coefficients to its measures after standardizing each one, so home value and income carry less weight than in the ADI. Income and degree share are single census measures, each ranked from the highest value to the lowest.

Source: IPEDS 2023 directory, American Community Survey 2019-2023 (residents), Area Deprivation Index 2023; resident-weighted means over counties; ADI 2023 and ACS 2019-2023 tract tables (B19013, B15003 and the 15 rebuilt ADI measures)

Data: https://vishalsingh.org/zipcode/report/college-access-price-and-earnings/figures/alt_rankings.json

## All estimates

Every fact in facts.json (133), sorted by id.

| Fact id | Value | Unit | Estimand | Population | Period | Interval | n | Weights | Source | Note |
| --- | ---: | --- | --- | --- | --- | --- | ---: | --- | --- | --- |
| alt_adi_least | 6.1 | Miles to the nearest four-year campus, average resident | resident-weighted mean of the county distance; 95% intervals from a bootstrap over states | the least deprived tenth of counties ranked by Area Deprivation Index, among the counties with all four rankings | campuses in the 2023 IPEDS directory, residents in the 2019 to 2023 ACS, county deprivation rank 2023 | 95% CI 5.3 to 7.0 | 312 counties | 2019 to 2023 ACS tract population | IPEDS |   |
| alt_adi_most | 29.7 | Miles to the nearest four-year campus, average resident | resident-weighted mean of the county distance; 95% intervals from a bootstrap over states | the most deprived tenth of counties ranked by Area Deprivation Index, among the counties with all four rankings | campuses in the 2023 IPEDS directory, residents in the 2019 to 2023 ACS, county deprivation rank 2023 | 95% CI 24.8 to 35.5 | 313 counties | 2019 to 2023 ACS tract population | IPEDS |   |
| alt_adi_rho | +0.99 | rank correlation | Spearman rank correlation between bin order and bin mean, each bin counted once | the bins of counties ranked by Area Deprivation Index | campuses in the 2023 IPEDS directory, residents in the 2019 to 2023 ACS, county deprivation rank 2023 |   |   | each bin counted once | IPEDS |   |
| alt_degree_least | 5.5 | Miles to the nearest four-year campus, average resident | resident-weighted mean of the county distance; 95% intervals from a bootstrap over states | the least deprived tenth of counties ranked by Bachelor's degree share, among the counties with all four rankings | campuses in the 2023 IPEDS directory, residents in the 2019 to 2023 ACS, county deprivation rank 2023 | 95% CI 5.0 to 6.2 | 312 counties | 2019 to 2023 ACS tract population | IPEDS |   |
| alt_degree_most | 31.0 | Miles to the nearest four-year campus, average resident | resident-weighted mean of the county distance; 95% intervals from a bootstrap over states | the most deprived tenth of counties ranked by Bachelor's degree share, among the counties with all four rankings | campuses in the 2023 IPEDS directory, residents in the 2019 to 2023 ACS, county deprivation rank 2023 | 95% CI 25.4 to 40.5 | 313 counties | 2019 to 2023 ACS tract population | IPEDS |   |
| alt_degree_rho | +0.99 | rank correlation | Spearman rank correlation between bin order and bin mean, each bin counted once | the bins of counties ranked by Bachelor's degree share | campuses in the 2023 IPEDS directory, residents in the 2019 to 2023 ACS, county deprivation rank 2023 |   |   | each bin counted once | IPEDS |   |
| alt_income_least | 6.7 | Miles to the nearest four-year campus, average resident | resident-weighted mean of the county distance; 95% intervals from a bootstrap over states | the least deprived tenth of counties ranked by Median household income, among the counties with all four rankings | campuses in the 2023 IPEDS directory, residents in the 2019 to 2023 ACS, county deprivation rank 2023 | 95% CI 6.0 to 7.3 | 312 counties | 2019 to 2023 ACS tract population | IPEDS |   |
| alt_income_most | 26.6 | Miles to the nearest four-year campus, average resident | resident-weighted mean of the county distance; 95% intervals from a bootstrap over states | the most deprived tenth of counties ranked by Median household income, among the counties with all four rankings | campuses in the 2023 IPEDS directory, residents in the 2019 to 2023 ACS, county deprivation rank 2023 | 95% CI 22.8 to 31.7 | 313 counties | 2019 to 2023 ACS tract population | IPEDS |   |
| alt_income_rho | +0.99 | rank correlation | Spearman rank correlation between bin order and bin mean, each bin counted once | the bins of counties ranked by Median household income | campuses in the 2023 IPEDS directory, residents in the 2019 to 2023 ACS, county deprivation rank 2023 |   |   | each bin counted once | IPEDS |   |
| alt_n_places | 3,125 | counties | count of counties in the common sample of the ranking comparison | counties with the outcome and all four rankings | campuses in the 2023 IPEDS directory, residents in the 2019 to 2023 ACS, county deprivation rank 2023 |   |   | none (full count) | IPEDS |   |
| alt_rho_cut | 0.9 | absolute rank correlation | threshold between a steady and an uneven gradient | the rule for changed gradients (docs/RANKING_SWITCH_RULE.md) | campuses in the 2023 IPEDS directory, residents in the 2019 to 2023 ACS, county deprivation rank 2023 |   |   | none (a fixed rule) | IPEDS |   |
| alt_rho_margin | 0.1 | absolute rank correlation | smallest move in the rank correlation that counts as a change of shape | the rule for changed gradients (docs/RANKING_SWITCH_RULE.md) | campuses in the 2023 IPEDS directory, residents in the 2019 to 2023 ACS, county deprivation rank 2023 |   |   | none (a fixed rule) | IPEDS |   |
| alt_std_least | 6.8 | Miles to the nearest four-year campus, average resident | resident-weighted mean of the county distance; 95% intervals from a bootstrap over states | the least deprived tenth of counties ranked by Standardized index, among the counties with all four rankings | campuses in the 2023 IPEDS directory, residents in the 2019 to 2023 ACS, county deprivation rank 2023 | 95% CI 6.2 to 7.6 | 312 counties | 2019 to 2023 ACS tract population | IPEDS |   |
| alt_std_most | 28.2 | Miles to the nearest four-year campus, average resident | resident-weighted mean of the county distance; 95% intervals from a bootstrap over states | the most deprived tenth of counties ranked by Standardized index, among the counties with all four rankings | campuses in the 2023 IPEDS directory, residents in the 2019 to 2023 ACS, county deprivation rank 2023 | 95% CI 22.4 to 41.4 | 313 counties | 2019 to 2023 ACS tract population | IPEDS |   |
| alt_std_rho | +0.93 | rank correlation | Spearman rank correlation between bin order and bin mean, each bin counted once | the bins of counties ranked by Standardized index | campuses in the 2023 IPEDS directory, residents in the 2019 to 2023 ACS, county deprivation rank 2023 |   |   | each bin counted once | IPEDS |   |
| baruch_earn | $75,971 | dollars per year (2022 dollars) | median earnings of one college's former students who received federal aid | Baruch College | median earnings in calendar years 2020 and 2021 (2022 dollars) of students who entered in 2009 to 2010 and 2010 to 2011 |   |   | none (one college) | College Scorecard |   |
| baruch_net | $745 | dollars per year (nominal), net price after grants | average net price of one college | Baruch College, students in the lowest income band who receive federal aid | academic year 2021 to 2022 (nominal dollars) |   |   | none (one college) | College Scorecard |   |
| baruch_pell | 55% | percent of undergraduates who received a Pell Grant | Pell share of one college | Baruch College of the City University of New York | academic year 2021 to 2022 |   |   | none (one college) | College Scorecard |   |
| bot_earn_med | $31,902 | dollars per year (2022 dollars) | median of the institution median earnings across the list | the lowest-earning high-Pell colleges in the pool | median earnings in calendar years 2020 and 2021 (2022 dollars) of students who entered in 2009 to 2010 and 2010 to 2011 |   | 15 institutions | none (each institution counts once) | College Scorecard | The list is a ranking of named colleges and not a sample, so the median carries no sampling interval. |
| bot_net_med | $10,626 | dollars per year (nominal), net price after grants | median across the list of the average net price for the lowest income band | the lowest-earning high-Pell colleges in the pool | academic year 2021 to 2022 (nominal dollars) |   | 15 institutions | none (each institution counts once) | College Scorecard |   |
| bot_pell_med | 60% | percent of undergraduates who received a Pell Grant | median of the Pell share across the list | the lowest-earning high-Pell colleges in the pool | academic year 2021 to 2022 |   | 15 institutions | none (each institution counts once) | College Scorecard |   |
| compl_ch | 64% | percent of first-time, full-time students | median across institutions of the completion rate within 150 percent of expected time; bootstrap over states | plotted four-year institutions above the median in cost and in earnings with a completion rate | completion within 150 percent of expected time, fall 2016 entering class | 95% CI 61% to 67% | 961 institutions | none (each institution counts once) | College Scorecard |   |
| compl_cl | 45% | percent of first-time, full-time students | median across institutions of the completion rate within 150 percent of expected time; bootstrap over states | plotted four-year institutions in the costly, low-earnings quadrant with a completion rate | completion within 150 percent of expected time, fall 2016 entering class | 95% CI 43% to 51% | 277 institutions | none (each institution counts once) | College Scorecard |   |
| compl_window | 150% | percent of normal time | the follow-up window in the definition of the completion rate, relative to the normal time to complete | completion rate of the College Scorecard | completion within 150 percent of expected time by students who started in fall 2016 (four-year) or fall 2019 (shorter programs) |   |   | none (definition) | College Scorecard |   |
| cost_med | $24,602 | dollars per year | median across the plotted institutions of the cost of attendance; the position of a quadrant line; bootstrap over states | institutions in the scatter | cost of attendance in academic year 2021 to 2022 (nominal dollars), earnings of the 2009 to 2011 entering classes (2022 dollars) | 95% CI $22,758 to $26,688 | 2,663 institutions | none (each institution counts once) | College Scorecard |   |
| dist_ak | 69 | miles, great-circle distance to the nearest four-year campus | resident-weighted mean over the state's counties of the county distance | residents of Alaska | campuses in the 2023 IPEDS directory, residents in the 2019 to 2023 ACS, county deprivation rank 2023 |   |   | 2019 to 2023 ACS tract population | Census TIGER/Line tracts | A full count of the state's residents and campuses, so it carries no sampling interval. |
| dist_all | 9.6 | miles, great-circle distance to the nearest four-year campus | resident-weighted mean over counties of the county's distance (each county's value is the mean over its tracts, weighted by tract population) | residents of the 50 states and DC | campuses in the 2023 IPEDS directory, residents in the 2019 to 2023 ACS, county deprivation rank 2023 |   | 3,143 counties | 2019 to 2023 ACS tract population | Census TIGER/Line tracts | A count of every tract and campus, so it carries no sampling interval. |
| dist_all_pub | 14.0 | miles, great-circle distance to the nearest public four-year campus | resident-weighted mean over counties of the county's distance (each county's value is the mean over its tracts, weighted by tract population) | residents of the 50 states and DC | campuses in the 2023 IPEDS directory, residents in the 2019 to 2023 ACS, county deprivation rank 2023 |   | 3,143 counties | 2019 to 2023 ACS tract population | Census TIGER/Line tracts | A count of every tract and campus, so it carries no sampling interval. |
| dist_d1 | 6.1 | miles, great-circle distance to the nearest four-year campus | resident-weighted mean of the county distance within a tenth of counties; bootstrap over states | residents of counties in the least deprived tenth of counties by ADI rank | campuses in the 2023 IPEDS directory, residents in the 2019 to 2023 ACS, county deprivation rank 2023 | 95% CI 5.3 to 6.9 | 312 counties | 2019 to 2023 ACS tract population | Census TIGER/Line tracts |   |
| dist_d10 | 29.3 | miles, great-circle distance to the nearest four-year campus | resident-weighted mean of the county distance within a tenth of counties; bootstrap over states | residents of counties in the most deprived tenth of counties by ADI rank | campuses in the 2023 IPEDS directory, residents in the 2019 to 2023 ACS, county deprivation rank 2023 | 95% CI 24.6 to 35.0 | 312 counties | 2019 to 2023 ACS tract population | Census TIGER/Line tracts |   |
| dist_med | 21 | miles, great-circle distance to the nearest four-year campus | median across counties of the county's resident-weighted distance | counties of the 50 states and DC | campuses in the 2023 IPEDS directory, residents in the 2019 to 2023 ACS, county deprivation rank 2023 |   | 3,143 counties | none (each county counts once) | Census TIGER/Line tracts |   |
| dist_metro | 7.0 | miles, great-circle distance to the nearest four-year campus | resident-weighted mean of the county distance within the group of counties; bootstrap over states | residents of metro counties (Rural-Urban Continuum Code 1 to 3) | campuses in the 2023 IPEDS directory, residents in the 2019 to 2023 ACS, county deprivation rank 2023 | 95% CI 6.3 to 7.8 | 1,185 counties | 2019 to 2023 ACS tract population | Census TIGER/Line tracts |   |
| dist_ms | 24 | miles, great-circle distance to the nearest four-year campus | resident-weighted mean over the state's counties of the county distance | residents of Mississippi | campuses in the 2023 IPEDS directory, residents in the 2019 to 2023 ACS, county deprivation rank 2023 |   |   | 2019 to 2023 ACS tract population | Census TIGER/Line tracts | A full count of the state's residents and campuses, so it carries no sampling interval. |
| dist_mt | 26 | miles, great-circle distance to the nearest four-year campus | resident-weighted mean over the state's counties of the county distance | residents of Montana | campuses in the 2023 IPEDS directory, residents in the 2019 to 2023 ACS, county deprivation rank 2023 |   |   | 2019 to 2023 ACS tract population | Census TIGER/Line tracts | A full count of the state's residents and campuses, so it carries no sampling interval. |
| dist_nonmetro | 25.7 | miles, great-circle distance to the nearest four-year campus | resident-weighted mean of the county distance within the group of counties; bootstrap over states | residents of nonmetro counties (Rural-Urban Continuum Code 4 to 9) | campuses in the 2023 IPEDS directory, residents in the 2019 to 2023 ACS, county deprivation rank 2023 | 95% CI 22.8 to 29.4 | 1,958 counties | 2019 to 2023 ACS tract population | Census TIGER/Line tracts |   |
| dist_p5 | 4.0 | miles, great-circle distance to the nearest four-year campus | 5th percentile across counties of the county's resident-weighted distance | counties of the 50 states and DC | campuses in the 2023 IPEDS directory, residents in the 2019 to 2023 ACS, county deprivation rank 2023 |   | 3,143 counties | none (each county counts once) | Census TIGER/Line tracts |   |
| dist_p95 | 74 | miles, great-circle distance to the nearest four-year campus | 95th percentile across counties of the county's resident-weighted distance | counties of the 50 states and DC | campuses in the 2023 IPEDS directory, residents in the 2019 to 2023 ACS, county deprivation rank 2023 |   | 3,143 counties | none (each county counts once) | Census TIGER/Line tracts |   |
| dist_pub_d1 | 9.4 | miles, great-circle distance to the nearest public four-year campus | resident-weighted mean of the county distance within a tenth of counties; bootstrap over states | residents of counties in the least deprived tenth of counties by ADI rank | campuses in the 2023 IPEDS directory, residents in the 2019 to 2023 ACS, county deprivation rank 2023 | 95% CI 8.3 to 10.8 | 312 counties | 2019 to 2023 ACS tract population | Census TIGER/Line tracts |   |
| dist_pub_d10 | 36.5 | miles, great-circle distance to the nearest public four-year campus | resident-weighted mean of the county distance within a tenth of counties; bootstrap over states | residents of counties in the most deprived tenth of counties by ADI rank | campuses in the 2023 IPEDS directory, residents in the 2019 to 2023 ACS, county deprivation rank 2023 | 95% CI 32.4 to 41.6 | 312 counties | 2019 to 2023 ACS tract population | Census TIGER/Line tracts |   |
| dist_ratio | 3.7 | ratio of miles | resident-weighted mean distance in nonmetro counties divided by the same in metro counties; bootstrap over states | residents of nonmetro counties against residents of metro counties | campuses in the 2023 IPEDS directory, residents in the 2019 to 2023 ACS, county deprivation rank 2023 | 95% CI 3.2 to 4.2 | 3,143 counties | 2019 to 2023 ACS tract population | Census TIGER/Line tracts |   |
| dist_rural4 | 34.3 | miles, great-circle distance to the nearest four-year campus | resident-weighted mean of the county distance within the group of counties; bootstrap over states | residents of nonmetro counties with an urban population under 5,000 (Rural-Urban Continuum Code 8 or 9) | campuses in the 2023 IPEDS directory, residents in the 2019 to 2023 ACS, county deprivation rank 2023 | 95% CI 29.7 to 40.9 | 1,055 counties | 2019 to 2023 ACS tract population | Census TIGER/Line tracts |   |
| dist_slope_tenth | 2.1 | miles per step up one tenth of county deprivation | slope of the county distance on the tenth (1 to 10) of county ADI rank, weighted by residents; bootstrap over states | residents of counties with an ADI rank | campuses in the 2023 IPEDS directory, residents in the 2019 to 2023 ACS, county deprivation rank 2023 | 95% CI 1.7 to 2.5 | 3,128 counties | 2019 to 2023 ACS tract population | Census TIGER/Line tracts |   |
| dist_wy | 49 | miles, great-circle distance to the nearest four-year campus | resident-weighted mean over the state's counties of the county distance | residents of Wyoming | campuses in the 2023 IPEDS directory, residents in the 2019 to 2023 ACS, county deprivation rank 2023 |   |   | 2019 to 2023 ACS tract population | Census TIGER/Line tracts | A full count of the state's residents and campuses, so it carries no sampling interval. |
| earn_forprofit | $28,330 | dollars per year (2022 dollars) | median across institutions of each institution's median earnings of former students who received federal aid, working and not enrolled ten years after entry; bootstrap over states | for-profit colleges in the 50 states and DC with a published earnings value | median earnings in calendar years 2020 and 2021 (2022 dollars) of students who entered in 2009 to 2010 and 2010 to 2011 | 95% CI $26,877 to $29,972 | 1,036 institutions | none (each institution counts once) | College Scorecard |   |
| earn_gap | $13,685 | dollars per year (2022 dollars), public minus for-profit | difference between the two sectors' medians of institution median earnings; bootstrap over states | public and for-profit colleges in the 50 states and DC with a published earnings value | median earnings in calendar years 2020 and 2021 (2022 dollars) of students who entered in 2009 to 2010 and 2010 to 2011 | 95% CI $12,287 to $15,220 | 2,752 institutions | none (each institution counts once) | College Scorecard |   |
| earn_gap_four | $10,738 | dollars per year (2022 dollars), public minus for-profit, four-year colleges only | difference between the two sectors' medians of institution median earnings; bootstrap over states | public and for-profit four-year colleges with a published earnings value | median earnings in calendar years 2020 and 2021 (2022 dollars) of students who entered in 2009 to 2010 and 2010 to 2011 | 95% CI $8,300 to $13,249 | 857 institutions | none (each institution counts once) | College Scorecard |   |
| earn_gap_short | $11,884 | dollars per year (2022 dollars), public minus for-profit, two-year and shorter programs only | difference between the two sectors' medians of institution median earnings; bootstrap over states | public and for-profit two-year and shorter colleges with a published earnings value | median earnings in calendar years 2020 and 2021 (2022 dollars) of students who entered in 2009 to 2010 and 2010 to 2011 | 95% CI $10,591 to $13,240 | 1,895 institutions | none (each institution counts once) | College Scorecard |   |
| earn_gap_w | $10,924 | dollars per year (2022 dollars), public minus for-profit, medians weighted by undergraduate enrollment | difference between the two sectors' medians of institution median earnings; bootstrap over states | public and for-profit colleges with a published earnings value and an enrollment count | median earnings in calendar years 2020 and 2021 (2022 dollars) of students who entered in 2009 to 2010 and 2010 to 2011 | 95% CI $7,972 to $12,851 | 2,744 institutions | fall 2022 undergraduate enrollment of the reporting institution | College Scorecard |   |
| earn_med | $46,092 | dollars per year | median across the plotted institutions of the median earnings; the position of a quadrant line; bootstrap over states | institutions in the scatter | cost of attendance in academic year 2021 to 2022 (nominal dollars), earnings of the 2009 to 2011 entering classes (2022 dollars) | 95% CI $44,379 to $47,866 | 2,663 institutions | none (each institution counts once) | College Scorecard |   |
| earn_nonprofit | $53,848 | dollars per year (2022 dollars) | median across institutions of each institution's median earnings of former students who received federal aid, working and not enrolled ten years after entry; bootstrap over states | private nonprofit colleges in the 50 states and DC with a published earnings value | median earnings in calendar years 2020 and 2021 (2022 dollars) of students who entered in 2009 to 2010 and 2010 to 2011 | 95% CI $51,671 to $55,838 | 1,211 institutions | none (each institution counts once) | College Scorecard |   |
| earn_public | $42,014 | dollars per year (2022 dollars) | median across institutions of each institution's median earnings of former students who received federal aid, working and not enrolled ten years after entry; bootstrap over states | public colleges in the 50 states and DC with a published earnings value | median earnings in calendar years 2020 and 2021 (2022 dollars) of students who entered in 2009 to 2010 and 2010 to 2011 | 95% CI $40,606 to $43,598 | 1,716 institutions | none (each institution counts once) | College Scorecard |   |
| edge_high | $110,000 | dollars of family income | lower limit of the highest income band | the highest of the five family-income bands in the Scorecard net price | academic year 2021 to 2022 (nominal dollars) |   |   | none (definition) | College Scorecard |   |
| edge_low | $30,000 | dollars of family income | upper limit of the lowest income band | the lowest of the five family-income bands in the Scorecard net price | academic year 2021 to 2022 (nominal dollars) |   |   | none (definition) | College Scorecard |   |
| high_p10 | $38,184 | dollars per year (2022 dollars) | 10th percentile of institution median earnings across high-Pell colleges | high-Pell colleges in the pool | median earnings in calendar years 2020 and 2021 (2022 dollars) of students who entered in 2009 to 2010 and 2010 to 2011 |   | 428 institutions | none (each institution counts once) | College Scorecard |   |
| high_p90 | $62,278 | dollars per year (2022 dollars) | 90th percentile of institution median earnings across high-Pell colleges | high-Pell colleges in the pool | median earnings in calendar years 2020 and 2021 (2022 dollars) of students who entered in 2009 to 2010 and 2010 to 2011 |   | 428 institutions | none (each institution counts once) | College Scorecard |   |
| mid_high | $150,000 | dollars of family income (assumed) | assumed family income used to express a net price as a share of income; the band has no upper limit, so this value is a choice | the highest income band | academic year 2021 to 2022 (nominal dollars) |   |   | none (assumption) | College Scorecard | An assumption of this chapter. The band is open-ended and the share for it moves in proportion to this value. |
| mid_low | $15,000 | dollars of family income (assumed) | assumed family income used to express a net price as a share of income: the midpoint of the band | the lowest income band | academic year 2021 to 2022 (nominal dollars) |   |   | none (assumption) | College Scorecard | An assumption of this chapter. The Scorecard does not report the income of the families in a band. |
| min_ugds_pool | 1,000 | undergraduates | smallest undergraduate enrollment of an institution in the pool | institutions in the pool | median earnings in calendar years 2020 and 2021 (2022 dollars) of students who entered in 2009 to 2010 and 2010 to 2011 |   |   | none (inclusion rule) | College Scorecard |   |
| min_ugds_scatter | 100 | undergraduates | smallest undergraduate enrollment of an institution in the scatter | institutions in the scatter | cost of attendance in academic year 2021 to 2022 (nominal dollars), earnings of the 2009 to 2011 entering classes (2022 dollars) |   |   | none (inclusion rule) | College Scorecard |   |
| miss_forprofit | 33.2% | percent of institutions | share of institutions with no published earnings value (privacy-suppressed or too few aided students) | for-profit colleges in the 50 states and DC | median earnings in calendar years 2020 and 2021 (2022 dollars) of students who entered in 2009 to 2010 and 2010 to 2011 |   | 1,552 institutions | none (full count) | College Scorecard |   |
| miss_nonprofit | 26.5% | percent of institutions | share of institutions with no published earnings value (privacy-suppressed or too few aided students) | private nonprofit colleges in the 50 states and DC | median earnings in calendar years 2020 and 2021 (2022 dollars) of students who entered in 2009 to 2010 and 2010 to 2011 |   | 1,648 institutions | none (full count) | College Scorecard |   |
| miss_public | 4.7% | percent of institutions | share of institutions with no published earnings value (privacy-suppressed or too few aided students) | public colleges in the 50 states and DC | median earnings in calendar years 2020 and 2021 (2022 dollars) of students who entered in 2009 to 2010 and 2010 to 2011 |   | 1,800 institutions | none (full count) | College Scorecard |   |
| n_bot_hbcu | 6 | institutions | count of colleges in the list that this chapter classes by name as historically Black colleges and universities | the lowest-earning high-Pell colleges | median earnings in calendar years 2020 and 2021 (2022 dollars) of students who entered in 2009 to 2010 and 2010 to 2011 |   |   | none (full count) | College Scorecard |   |
| n_bot_tribal_seminary | 5 | institutions | count of colleges in the list that this chapter classes by name as tribal colleges or religious seminaries | the lowest-earning high-Pell colleges | median earnings in calendar years 2020 and 2021 (2022 dollars) of students who entered in 2009 to 2010 and 2010 to 2011 |   |   | none (full count) | College Scorecard |   |
| n_branch | 1,275 | rows | count of rows dropped because the 6-digit OPE ID already has a row (the campus with the most undergraduates) whose earnings and debt the branch row repeats | Scorecard rows for branch campuses in the 50 states and DC | College Scorecard May 2025 release |   |   | none (full count) | College Scorecard |   |
| n_campuses | 2,508 | campuses | count of campuses used as destinations, branch campuses counted separately | four-year institutions in the 50 states and DC that are not entirely graduate-degree granting | campuses in the 2023 IPEDS directory, residents in the 2019 to 2023 ACS, county deprivation rank 2023 |   |   | none (full count) | IPEDS |   |
| n_cl | 334 | institutions | count of institutions whose cost of attendance is above the median and whose median earnings are at or below the median | institutions in the scatter | cost of attendance in academic year 2021 to 2022 (nominal dollars), earnings of the 2009 to 2011 entering classes (2022 dollars) |   |   | none (full count) | College Scorecard |   |
| n_counties | 3,143 | counties | count of counties drawn on the map | counties and county equivalents of the 50 states and DC (Connecticut as its eight legacy counties) | campuses in the 2023 IPEDS directory, residents in the 2019 to 2023 ACS, county deprivation rank 2023 |   |   | none (full count) | IPEDS |   |
| n_earn | 3,963 | institutions | count of institutions with median earnings ten years after entry | institutions (6-digit OPE ID reporting units) in the 50 states and DC with a published earnings value | median earnings in calendar years 2020 and 2021 (2022 dollars) of students who entered in 2009 to 2010 and 2010 to 2011 |   |   | none (full count) | College Scorecard |   |
| n_earn_forprofit | 1,036 | institutions | count of institutions | for-profit colleges in the 50 states and DC with a published earnings value | median earnings in calendar years 2020 and 2021 (2022 dollars) of students who entered in 2009 to 2010 and 2010 to 2011 |   |   | none (full count) | College Scorecard |   |
| n_earn_nonprofit | 1,211 | institutions | count of institutions | private nonprofit colleges in the 50 states and DC with a published earnings value | median earnings in calendar years 2020 and 2021 (2022 dollars) of students who entered in 2009 to 2010 and 2010 to 2011 |   |   | none (full count) | College Scorecard |   |
| n_earn_public | 1,716 | institutions | count of institutions | public colleges in the 50 states and DC with a published earnings value | median earnings in calendar years 2020 and 2021 (2022 dollars) of students who entered in 2009 to 2010 and 2010 to 2011 |   |   | none (full count) | College Scorecard |   |
| n_far_top | 10 | counties | number of counties with the longest resident-weighted distance whose state is counted | counties of the 50 states and DC | campuses in the 2023 IPEDS directory, residents in the 2019 to 2023 ACS, county deprivation rank 2023 |   |   | none (rule) | IPEDS |   |
| n_fp_dropped | 48 | institutions | count of for-profit institutions left out of the pool by the rule that for-profit colleges are not on the list | for-profit four-year colleges that meet every pool rule, including a net price for the lowest income band | median earnings in calendar years 2020 and 2021 (2022 dollars) of students who entered in 2009 to 2010 and 2010 to 2011 |   |   | none (full count) | College Scorecard |   |
| n_grad_counties | 3,128 | counties | count of counties in the deprivation gradient | counties with a county ADI rank | campuses in the 2023 IPEDS directory, residents in the 2019 to 2023 ACS, county deprivation rank 2023 |   |   | none (full count) | Area Deprivation Index |   |
| n_grad_only | 273 | institutions | count left out because the Scorecard classes their predominant degree as graduate | four-year institutions in the 50 states and DC | campuses in the 2023 IPEDS directory, residents in the 2019 to 2023 ACS, county deprivation rank 2023 |   |   | none (full count) | IPEDS |   |
| n_high_pell | 428 | institutions | count of institutions whose Pell share reaches the cutoff | institutions in the top third of the pool by Pell share | median earnings in calendar years 2020 and 2021 (2022 dollars) of students who entered in 2009 to 2010 and 2010 to 2011 |   |   | none (full count) | College Scorecard |   |
| n_inst | 5,000 | institutions | count of reporting institutions after dropping rows outside the 50 states and DC and branch rows that repeat their parent's earnings and debt | institutions (6-digit OPE ID reporting units) in the 50 states and DC | College Scorecard May 2025 release |   |   | none (full count) | College Scorecard |   |
| n_list | 15 | institutions | number of colleges in the list of highest-earning and in the list of lowest-earning high-Pell colleges | each list of high-Pell colleges | median earnings in calendar years 2020 and 2021 (2022 dollars) of students who entered in 2009 to 2010 and 2010 to 2011 |   |   | none (rule) | College Scorecard |   |
| n_netp_priv | 1,157 | institutions | count of institutions | private nonprofit four-year colleges in the 50 states and DC with a published net price in at least one band | academic year 2021 to 2022 (nominal dollars) |   |   | none (full count) | College Scorecard |   |
| n_netp_pub | 694 | institutions | count of institutions | public four-year colleges in the 50 states and DC with a published net price in at least one band | academic year 2021 to 2022 (nominal dollars) |   |   | none (full count) | College Scorecard |   |
| n_no_adi | 15 | counties | count of counties left out of the gradient because every tract has a suppressed ADI rank | counties of the 50 states and DC | campuses in the 2023 IPEDS directory, residents in the 2019 to 2023 ACS, county deprivation rank 2023 |   |   | none (full count) | Area Deprivation Index |   |
| n_no_college | 2,210 | counties | count of counties with no four-year campus inside the county | counties of the 50 states and DC | campuses in the 2023 IPEDS directory, residents in the 2019 to 2023 ACS, county deprivation rank 2023 |   |   | none (full count) | IPEDS |   |
| n_pool | 1,282 | institutions | count of institutions in the pool | public and private nonprofit four-year colleges with at least 1,000 undergraduates, a Pell share, median earnings and a net price for the lowest income band | median earnings in calendar years 2020 and 2021 (2022 dollars) of students who entered in 2009 to 2010 and 2010 to 2011 |   |   | none (full count) | College Scorecard |   |
| n_scatter | 2,663 | institutions | count of institutions in the scatter | institutions with a cost of attendance, median earnings and at least 100 undergraduates | cost of attendance in academic year 2021 to 2022 (nominal dollars), earnings of the 2009 to 2011 entering classes (2022 dollars) |   |   | none (full count) | College Scorecard |   |
| n_scatter_fp | 163 | institutions | count of institutions | for-profit colleges in the scatter | cost of attendance in academic year 2021 to 2022 (nominal dollars), earnings of the 2009 to 2011 entering classes (2022 dollars) |   |   | none (full count) | College Scorecard |   |
| n_top10 | 10 | institutions | number of colleges at the head of the list whose location is counted | the highest-earning high-Pell colleges | median earnings in calendar years 2020 and 2021 (2022 dollars) of students who entered in 2009 to 2010 and 2010 to 2011 |   |   | none (rule) | College Scorecard |   |
| n_top10_ca_ny | 8 | institutions | count of the ten highest-earning high-Pell colleges located in California or New York | the highest-earning high-Pell colleges | median earnings in calendar years 2020 and 2021 (2022 dollars) of students who entered in 2009 to 2010 and 2010 to 2011 |   |   | none (full count) | College Scorecard |   |
| n_top_class | 57 | counties | count of counties whose resident-weighted distance is in the highest class of the map key | counties of the 50 states and DC | campuses in the 2023 IPEDS directory, residents in the 2019 to 2023 ACS, county deprivation rank 2023 |   |   | none (full count) | IPEDS |   |
| n_with_college | 933 | counties | count of counties with at least one four-year campus inside the county | counties of the 50 states and DC | campuses in the 2023 IPEDS directory, residents in the 2019 to 2023 ACS, county deprivation rank 2023 |   |   | none (full count) | IPEDS |   |
| nocoll_d1 | 42% | percent of counties | share of counties with no four-year campus inside the county, each county counted once; bootstrap over states | counties in the least deprived tenth of counties by ADI rank | campuses in the 2023 IPEDS directory, residents in the 2019 to 2023 ACS, county deprivation rank 2023 | 95% CI 33% to 51% | 312 counties | none (each county counts once) | IPEDS |   |
| nocoll_d10 | 92% | percent of counties | share of counties with no four-year campus inside the county, each county counted once; bootstrap over states | counties in the most deprived tenth of counties by ADI rank | campuses in the 2023 IPEDS directory, residents in the 2019 to 2023 ACS, county deprivation rank 2023 | 95% CI 87% to 95% | 312 counties | none (each county counts once) | IPEDS |   |
| nocoll_metro | 49% | percent of counties | share of counties with no four-year campus inside the county, each county counted once; bootstrap over states | Metro counties (Rural-Urban Continuum Code 1 to 3) | campuses in the 2023 IPEDS directory, residents in the 2019 to 2023 ACS, county deprivation rank 2023 | 95% CI 42% to 55% | 1,185 counties | none (each county counts once) | Rural-Urban Continuum Codes |   |
| nocoll_nonmetro | 83% | percent of counties | share of counties with no four-year campus inside the county, each county counted once; bootstrap over states | Nonmetro counties (Rural-Urban Continuum Code 4 to 9) | campuses in the 2023 IPEDS directory, residents in the 2019 to 2023 ACS, county deprivation rank 2023 | 95% CI 80% to 86% | 1,958 counties | none (each county counts once) | Rural-Urban Continuum Codes |   |
| nocoll_rural4 | 95% | percent of counties | share of counties with no four-year campus inside the county, each county counted once; bootstrap over states | nonmetro counties with an urban population under 5,000 (Rural-Urban Continuum Code 8 or 9) | campuses in the 2023 IPEDS directory, residents in the 2019 to 2023 ACS, county deprivation rank 2023 | 95% CI 94% to 97% | 1,055 counties | none (each county counts once) | Rural-Urban Continuum Codes |   |
| nocoll_slope_tenth | 5.3 percentage points | percentage points per step up one tenth of county deprivation | slope of the indicator that a county has no four-year campus (0 or 100) on the tenth (1 to 10) of county ADI rank, each county counted once; bootstrap over states | counties with an ADI rank | campuses in the 2023 IPEDS directory, residents in the 2019 to 2023 ACS, county deprivation rank 2023 | 95% CI 4.5 percentage points to 6.2 percentage points | 3,128 counties | none (each county counts once) | IPEDS |   |
| nonmetro_d1 | 23% | percent of counties | share of counties that are nonmetro (Rural-Urban Continuum Code 4 to 9), each county counted once; bootstrap over states | counties in the least deprived tenth of counties by ADI rank | campuses in the 2023 IPEDS directory, residents in the 2019 to 2023 ACS, county deprivation rank 2023 | 95% CI 14% to 31% | 312 counties | none (each county counts once) | Rural-Urban Continuum Codes |   |
| nonmetro_d10 | 94% | percent of counties | share of counties that are nonmetro (Rural-Urban Continuum Code 4 to 9), each county counted once; bootstrap over states | counties in the most deprived tenth of counties by ADI rank | campuses in the 2023 IPEDS directory, residents in the 2019 to 2023 ACS, county deprivation rank 2023 | 95% CI 92% to 96% | 312 counties | none (each county counts once) | Rural-Urban Continuum Codes |   |
| pell_cl | 47% | percent of undergraduates who received a Pell Grant | median across institutions of the Pell share; the quadrant lines are redrawn in every bootstrap replicate over states | plotted institutions in the costly, low-earnings quadrant | academic year 2021 to 2022 | 95% CI 42% to 51% | 334 institutions | none (each institution counts once) | College Scorecard |   |
| pell_cut | 34.7% | percent of undergraduates who received a Pell Grant | two-thirds quantile of the Pell share across the pool: a college with this share or more is high-Pell | public and private nonprofit four-year colleges with at least 1,000 undergraduates, a Pell share, median earnings and a net price for the lowest income band | academic year 2021 to 2022 |   |   | none (each institution counts once) | College Scorecard |   |
| pell_forprofit | 52.5% | percent of undergraduates who received a Pell Grant | median across institutions of the share of undergraduates receiving a Pell Grant; bootstrap over states | for-profit colleges in the 50 states and DC with a Pell share | academic year 2021 to 2022 | 95% CI 50.4% to 54.6% | 1,519 institutions | none (each institution counts once) | College Scorecard |   |
| pell_nonprofit | 32.4% | percent of undergraduates who received a Pell Grant | median across institutions of the share of undergraduates receiving a Pell Grant; bootstrap over states | private nonprofit colleges in the 50 states and DC with a Pell share | academic year 2021 to 2022 | 95% CI 30.1% to 34.6% | 1,427 institutions | none (each institution counts once) | College Scorecard |   |
| pell_public | 30.4% | percent of undergraduates who received a Pell Grant | median across institutions of the share of undergraduates receiving a Pell Grant; bootstrap over states | public colleges in the 50 states and DC with a Pell share | academic year 2021 to 2022 | 95% CI 28.3% to 32.9% | 1,781 institutions | none (each institution counts once) | College Scorecard |   |
| pell_rest | 30% | percent of undergraduates who received a Pell Grant | median across institutions of the Pell share; the quadrant lines are redrawn in every bootstrap replicate over states | plotted institutions in the other three quadrants | academic year 2021 to 2022 | 95% CI 28% to 31% | 2,329 institutions | none (each institution counts once) | College Scorecard |   |
| priv_step | 72% | percent of the fall in the share of income from the lowest to the highest band | fall in the share of assumed income from the lowest band to the second band, divided by the fall from the lowest band to the highest; bootstrap over states | private nonprofit four-year colleges with a published net price | academic year 2021 to 2022 (nominal dollars) | 95% CI 71% to 73% | 1,157 institutions | none (each institution counts once) | College Scorecard |   |
| priv_t1_burden | 117% | percent of an assumed family income | median net price divided by an assumed family income at the midpoint of the band ($15,000); the interval carries the sampling variation of the median only, not the assumption | private nonprofit four-year colleges with a published net price in the band (lowest income band) | academic year 2021 to 2022 (nominal dollars) | 95% CI 113% to 122% | 1,120 institutions | none (each institution counts once) | College Scorecard | Net price includes living costs that a family bears whether or not the student enrolls; the share is an order of magnitude and not a measured share of any family's income. |
| priv_t1_net | $17,596 | dollars per year (nominal), net price after grants | median across institutions of the institution's average net price for full-time, first-time students who received federal aid in the band; bootstrap over states | private nonprofit four-year colleges with a published net price in the band (lowest income band) | academic year 2021 to 2022 (nominal dollars) | 95% CI $16,937 to $18,307 | 1,120 institutions | none (each institution counts once) | College Scorecard |   |
| priv_t5_burden | 19% | percent of an assumed family income | median net price divided by an assumed family income at the midpoint of the band ($150,000); the interval carries the sampling variation of the median only, not the assumption | private nonprofit four-year colleges with a published net price in the band (highest income band) | academic year 2021 to 2022 (nominal dollars) | 95% CI 18% to 20% | 976 institutions | none (each institution counts once) | College Scorecard | Net price includes living costs that a family bears whether or not the student enrolls; the share is an order of magnitude and not a measured share of any family's income. |
| priv_t5_edge | 26% | percent of the lower limit of the highest income band | median net price in the highest band divided by the lower limit of the band, the smallest income a family in it can have; bootstrap over states | private nonprofit four-year colleges with a published net price (highest income band) | academic year 2021 to 2022 (nominal dollars) | 95% CI 25% to 28% | 976 institutions | none (each institution counts once) | College Scorecard |   |
| priv_t5_net | $28,710 | dollars per year (nominal), net price after grants | median across institutions of the institution's average net price for full-time, first-time students who received federal aid in the band; bootstrap over states | private nonprofit four-year colleges with a published net price in the band (highest income band) | academic year 2021 to 2022 (nominal dollars) | 95% CI $26,953 to $30,525 | 976 institutions | none (each institution counts once) | College Scorecard |   |
| pub_step | 73% | percent of the fall in the share of income from the lowest to the highest band | fall in the share of assumed income from the lowest band to the second band, divided by the fall from the lowest band to the highest; bootstrap over states | public four-year colleges with a published net price | academic year 2021 to 2022 (nominal dollars) | 95% CI 71% to 74% | 694 institutions | none (each institution counts once) | College Scorecard |   |
| pub_t1_burden | 62% | percent of an assumed family income | median net price divided by an assumed family income at the midpoint of the band ($15,000); the interval carries the sampling variation of the median only, not the assumption | public four-year colleges with a published net price in the band (lowest income band) | academic year 2021 to 2022 (nominal dollars) | 95% CI 56% to 70% | 693 institutions | none (each institution counts once) | College Scorecard | Net price includes living costs that a family bears whether or not the student enrolls; the share is an order of magnitude and not a measured share of any family's income. |
| pub_t1_net | $9,347 | dollars per year (nominal), net price after grants | median across institutions of the institution's average net price for full-time, first-time students who received federal aid in the band; bootstrap over states | public four-year colleges with a published net price in the band (lowest income band) | academic year 2021 to 2022 (nominal dollars) | 95% CI $8,426 to $10,430 | 693 institutions | none (each institution counts once) | College Scorecard |   |
| pub_t5_burden | 13% | percent of an assumed family income | median net price divided by an assumed family income at the midpoint of the band ($150,000); the interval carries the sampling variation of the median only, not the assumption | public four-year colleges with a published net price in the band (highest income band) | academic year 2021 to 2022 (nominal dollars) | 95% CI 12% to 13% | 647 institutions | none (each institution counts once) | College Scorecard | Net price includes living costs that a family bears whether or not the student enrolls; the share is an order of magnitude and not a measured share of any family's income. |
| pub_t5_edge | 17% | percent of the lower limit of the highest income band | median net price in the highest band divided by the lower limit of the band, the smallest income a family in it can have; bootstrap over states | public four-year colleges with a published net price (highest income band) | academic year 2021 to 2022 (nominal dollars) | 95% CI 16% to 18% | 647 institutions | none (each institution counts once) | College Scorecard |   |
| pub_t5_net | $18,780 | dollars per year (nominal), net price after grants | median across institutions of the institution's average net price for full-time, first-time students who received federal aid in the band; bootstrap over states | public four-year colleges with a published net price in the band (highest income band) | academic year 2021 to 2022 (nominal dollars) | 95% CI $17,910 to $19,987 | 647 institutions | none (each institution counts once) | College Scorecard |   |
| r_compl_four | +0.67 | correlation coefficient | correlation across four-year colleges of the completion rate with median earnings, each institution counted once; bootstrap over states | four-year institutions in the scatter with a completion rate | cost of attendance in academic year 2021 to 2022 (nominal dollars), earnings of the 2009 to 2011 entering classes (2022 dollars); completion within 150 percent of expected time, fall 2016 entering class | 95% CI 0.62 to 0.71 | 1,783 institutions | none (each institution counts once) | College Scorecard |   |
| r_compl_short | +0.12 | correlation coefficient | correlation across shorter programs of the completion rate with median earnings, each institution counted once; bootstrap over states | two-year and shorter institutions in the scatter with a completion rate | cost of attendance in academic year 2021 to 2022 (nominal dollars), earnings of the 2009 to 2011 entering classes (2022 dollars); completion within 150 percent of expected time, fall 2019 entering class | 95% CI 0.02 to 0.21 | 868 institutions | none (each institution counts once) | College Scorecard |   |
| r_cost_earn | +0.63 | correlation coefficient | correlation across institutions of the cost of attendance with median earnings, each institution counted once; bootstrap over states | institutions in the scatter | cost of attendance in academic year 2021 to 2022 (nominal dollars), earnings of the 2009 to 2011 entering classes (2022 dollars) | 95% CI 0.59 to 0.67 | 2,663 institutions | none (each institution counts once) | College Scorecard |   |
| r_dist_adi | +0.31 | correlation coefficient | correlation across counties of the county ADI national rank with the resident-weighted distance, weighted by residents; bootstrap over states | counties with an ADI rank | campuses in the 2023 IPEDS directory, residents in the 2019 to 2023 ACS, county deprivation rank 2023 | 95% CI 0.24 to 0.40 | 3,128 counties | 2019 to 2023 ACS tract population | IPEDS |   |
| r_dist_adi_metro | +0.24 | correlation coefficient | correlation within the group of counties of the county ADI national rank with the resident-weighted distance; bootstrap over states | Metro counties with an ADI rank | campuses in the 2023 IPEDS directory, residents in the 2019 to 2023 ACS, county deprivation rank 2023 | 95% CI 0.18 to 0.32 | 1,183 counties | 2019 to 2023 ACS tract population | Rural-Urban Continuum Codes |   |
| r_dist_adi_nonmetro | −0.04 | correlation coefficient | correlation within the group of counties of the county ADI national rank with the resident-weighted distance; bootstrap over states | Nonmetro counties with an ADI rank | campuses in the 2023 IPEDS directory, residents in the 2019 to 2023 ACS, county deprivation rank 2023 | 95% CI −0.12 to 0.05 | 1,945 counties | 2019 to 2023 ACS tract population | Rural-Urban Continuum Codes |   |
| r_dist_adi_u | +0.16 | correlation coefficient | correlation across counties of the county ADI national rank with the resident-weighted distance, each county counted once; bootstrap over states | counties with an ADI rank | campuses in the 2023 IPEDS directory, residents in the 2019 to 2023 ACS, county deprivation rank 2023 | 95% CI 0.09 to 0.33 | 3,128 counties | none (each county counts once) | IPEDS |   |
| share_cl_all | 12.5% | percent of plotted institutions | share of institutions in the costly, low-earnings quadrant; bootstrap over states | institutions in the scatter | cost of attendance in academic year 2021 to 2022 (nominal dollars), earnings of the 2009 to 2011 entering classes (2022 dollars) | 95% CI 11.3% to 13.8% | 2,663 institutions | none (each institution counts once) | College Scorecard |   |
| share_cl_forprofit | 56% | percent of plotted institutions in the sector | share of the sector's plotted institutions in the costly, low-earnings quadrant; bootstrap over states | for-profit colleges in the scatter | cost of attendance in academic year 2021 to 2022 (nominal dollars), earnings of the 2009 to 2011 entering classes (2022 dollars) | 95% CI 48% to 62% | 163 institutions | none (each institution counts once) | College Scorecard |   |
| share_cl_nonprofit | 20% | percent of plotted institutions in the sector | share of the sector's plotted institutions in the costly, low-earnings quadrant; bootstrap over states | private nonprofit colleges in the scatter | cost of attendance in academic year 2021 to 2022 (nominal dollars), earnings of the 2009 to 2011 entering classes (2022 dollars) | 95% CI 18% to 23% | 1,052 institutions | none (each institution counts once) | College Scorecard |   |
| share_cl_public | 2% | percent of plotted institutions in the sector | share of the sector's plotted institutions in the costly, low-earnings quadrant; bootstrap over states | public colleges in the scatter | cost of attendance in academic year 2021 to 2022 (nominal dollars), earnings of the 2009 to 2011 entering classes (2022 dollars) | 95% CI 1% to 3% | 1,448 institutions | none (each institution counts once) | College Scorecard |   |
| share_no_college | 70% | percent of counties | share of counties with no four-year campus inside the county, each county counted once | counties of the 50 states and DC | campuses in the 2023 IPEDS directory, residents in the 2019 to 2023 ACS, county deprivation rank 2023 |   | 3,143 counties | none (each county counts once) | IPEDS | A count of every county and campus, so it carries no sampling interval. |
| top_break | 100 | miles | lower limit of the highest class of the map key | counties of the 50 states and DC | campuses in the 2023 IPEDS directory, residents in the 2019 to 2023 ACS, county deprivation rank 2023 |   |   | none (definition) | IPEDS |   |
| top_earn_med | $72,282 | dollars per year (2022 dollars) | median of the institution median earnings across the list | the highest-earning high-Pell colleges in the pool | median earnings in calendar years 2020 and 2021 (2022 dollars) of students who entered in 2009 to 2010 and 2010 to 2011 |   | 15 institutions | none (each institution counts once) | College Scorecard | The list is a ranking of named colleges and not a sample, so the median carries no sampling interval. |
| top_net_med | $9,743 | dollars per year (nominal), net price after grants | median across the list of the average net price for the lowest income band | the highest-earning high-Pell colleges in the pool | academic year 2021 to 2022 (nominal dollars) |   | 15 institutions | none (each institution counts once) | College Scorecard |   |
| top_pell_med | 39% | percent of undergraduates who received a Pell Grant | median of the Pell share across the list | the highest-earning high-Pell colleges in the pool | academic year 2021 to 2022 |   | 15 institutions | none (each institution counts once) | College Scorecard |   |
| us_college_miles | 9.6 | miles, great-circle distance to the nearest four-year campus | resident-weighted mean of the county distance over all the tenths of counties; bootstrap over states | residents of all counties ranked by the ADI | campuses in the 2023 IPEDS directory, residents in the 2019 to 2023 ACS, county deprivation rank 2023 | 95% CI 8.5 to 11.0 | 3,128 counties | 2019 to 2023 ACS tract population | Census TIGER/Line tracts |   |

## Definitions

- Bootstrap interval: A range for an estimate found by recomputing it on many random resamples of the data and reading the range from the results. It shows how much the estimate would vary from sample to sample.
- Suppression: The withholding of a number because the count behind it is too small to publish without risking the identification of a person. A suppressed value is missing, not zero.
- Census tract: A small area drawn by the Census Bureau to stand in for a neighborhood. Boundaries are redrawn after each decennial census.
- Area Deprivation Index (ADI): A score that ranks neighborhoods by income, education, employment and housing conditions drawn from the American Community Survey. In this book a higher national percentile means a more deprived neighborhood.

## Methods

The college comparisons use one row per institution (5,000 institutions in the 50 states and DC) and the median earnings of former students who received federal aid, who entered in 2009 to 2010 and 2010 to 2011 and were observed in 2020 and 2021, so they do not cover all students or those entering now. The county comparisons measure the great-circle distance in miles from each tract's Census internal point to the nearest four-year campus, averaged over a county's tracts. Institutions, tracts and campuses are enumerated and not sampled, so the intervals describe how much an estimate varies from state to state and are not sampling errors.

### Which institutions and years the college figures cover

Institutions come from the College Scorecard institution file of May 2025 (most recent cohorts), limited to the 50 states and DC. The Scorecard attaches a 6-digit OPE ID parent's earnings and debt to each of its branch campuses, so one row per OPE ID is kept (the campus with the most undergraduates, which also supplies the Pell share, cost of attendance, net price and completion rate), which leaves 5,000 institutions from 6,429 rows; 154 rows outside the states and DC and 1,275 branch rows are dropped.

Earnings are the median earnings of students who received federal aid, were working and not enrolled ten years after entry, entered in 2009 to 2010 and 2010 to 2011, and were observed in 2020 and 2021 in 2022 dollars; the Scorecard adds noise to them and withholds them for small groups. Net price, cost of attendance and Pell share are for academic year 2021 to 2022 in nominal dollars.

Residents of Puerto Rico and campuses outside the 50 states and DC are not in the data.

Explains Table 1, Table 2, Table 3 and Table 4.

### How sectors, completion rates and medians are defined

Sector is public, private nonprofit or for-profit (Scorecard control). The two completion rates (four-year colleges, shorter programs) are never pooled. A sector figure is the median across institutions with each counted once; enrollment-weighted medians (fall 2022 undergraduates) are the check.

Explains Table 1.

### How net price, the cost and earnings scatter and the high-Pell list are defined

Net price by income band covers public and private nonprofit four-year colleges and is shown as dollars and as a share of an assumed family income at the midpoint of each band (15,000, 39,000, 61,500, 92,500 and 150,000 dollars; the highest band is open-ended, so its midpoint is a choice). In the cost and earnings scatter, colleges with fewer than 100 undergraduates or no cost of attendance are left out, and the quadrant lines are medians.

The high-Pell pool is public and private nonprofit four-year colleges with at least 1,000 undergraduates, a Pell share, earnings and a net price for the lowest band; for-profit colleges are not in the pool, and a college is high-Pell when its Pell share reaches the two-thirds quantile of the pool.

Explains Table 2, Table 3 and Table 4.

### How distance to a four-year college is measured

Distance is the great-circle (haversine) distance in miles from the Census internal point of each 2020 tract to the nearest four-year campus in the IPEDS 2023 directory, leaving out 273 institutions the Scorecard classes as entirely graduate-degree granting, so 2,508 campuses remain. A county's value is the mean of its tracts' distances weighted by the 2019 to 2023 ACS tract population, and counties are the 3,143 on the map sprite (Connecticut as eight legacy counties through the staged crosswalk; each Connecticut campus goes to the legacy county of the tract that holds its point).

Explains Table 5 and Table 6.

### How counties are grouped and weighted

County deprivation is the mean of tract ADI national ranks weighted by tract population, in tenths of equal numbers of counties; the rural-urban grouping, for all counties, is the USDA Rural-Urban Continuum Code 2023 of the county, with a legacy Connecticut county taking the code held by most of its residents. Gradient means weight counties by residents and county shares count each county once.

Explains Table 6.

### How uncertainty is measured

Intervals for sector and band medians, shares and correlations across colleges are percentile intervals from 2,000 draws of whole states with replacement (seed 12345). They describe how much an estimate varies from state to state; institutions, tracts and campuses are enumerated and not sampled, so they are not sampling errors, and counts and national distances carry none.

In the county gradients, intervals redraw whole states (2,000 draws, seed 12345).

### Four rankings of the same places

The comparison follows the book's ranking switch rule, fixed on October 28, 2026 before any of these gradients was computed. The main gradient of the chapter is recomputed under four rankings of the same places, with the outcome, period, weights, bins and intervals unchanged: the published ADI; a standardized index that applies the ADI's coefficients to its measures after standardizing each one across tracts; median household income; and the share of adults 25 and older with a bachelor's degree or more. The three alternatives come from ACS tract estimates for 2019 to 2023, the period of the ADI.

All four rankings use the same places, the 3,125 counties that have the outcome and all four rankings, so the ADI panel can differ slightly from the chapter's main figure. Tracts enter only if they have every input of the standardized index, which leaves out tracts without a median rent, home value or mortgage cost. A county takes the population-weighted mean of the ranks of its common-sample tracts, with each Connecticut tract in its legacy county through the staged crosswalk, the construction of the chapter's county ADI.

A pattern counts as changed when the worse end switches sides, when the two ends differ under one ranking and not the other, or when the rank correlation of bin order and bin mean falls on the other side of 0.9 and moves by at least 0.1.

Explains Table 7.

## What these data cannot show

The geography describes supply. Distance is measured in a straight line from where residents live to where campuses sit, so it ignores roads, ferries, online programs and the colleges students choose when they leave home. The Chetty and co-authors data that link the neighborhood a student grew up in to the college the student attended and the income reached are not part of this project's data, and nothing here reproduces them. A county average also hides the tracts inside it, and the gradients describe counties and not the families who live in them.

The earnings describe aided students who entered in 2009 to 2011, not all students and not those entering now. They are medians of institutions, with noise added by the Scorecard, and a larger share of private colleges than of public colleges have no published value. Branch campuses repeat their parent's earnings and debt, which is why the chapter counts institutions by OPE ID. The Pell share, cost of attendance, net price and completion rate come from the largest campus of each institution. Earnings follow local wages and the field of study as well as the college, and the sector gap mixes what colleges do with who enrolls and what programs they run.

The net price is an average over aided, first-time, full-time students in each band, at public colleges for in-state students only, and the share of income depends on assumed incomes. The lists of high-Pell colleges are rankings of named colleges, with no interval, and they include colleges whose earnings are not comparable with those of a general four-year degree. The designs in the box above identify effects at specific margins and do not show what a college does to a student in general.

## References

1. Deming, David J.; Goldin, Claudia; Katz, Lawrence F. (2012). The For-Profit Postsecondary School Sector: Nimble Critters or Agile Predators?. Journal of Economic Perspectives 26(1): 139-164. https://doi.org/10.1257/jep.26.1.139
2. Chetty, Raj; Friedman, John N.; Saez, Emmanuel; Turner, Nicholas; Yagan, Danny (2020). Income Segregation and Intergenerational Mobility Across Colleges in the United States. The Quarterly Journal of Economics 135(3): 1567-1633. https://doi.org/10.1093/qje/qjaa005
3. Chetty, Raj; Friedman, John N.; Saez, Emmanuel; Turner, Nicholas; Yagan, Danny (2017). Mobility Report Cards: The Role of Colleges in Intergenerational Mobility. NBER Working Paper 23618. https://www.nber.org/papers/w23618
4. Hoxby, Caroline; Avery, Christopher (2013). The Missing “One-Offs”: The Hidden Supply of High-Achieving, Low-Income Students. Brookings Papers on Economic Activity (Spring 2013): 1-65. https://doi.org/10.1353/eca.2013.0000
5. Zimmerman, Seth D. (2014). The Returns to College Admission for Academically Marginal Students. Journal of Labor Economics 32(4): 711-754. https://doi.org/10.1086/676661
6. Bettinger, Eric; Gurantz, Oded; Kawano, Laura; Sacerdote, Bruce; Stevens, Michael (2019). The Long-Run Impacts of Financial Aid: Evidence from California's Cal Grant. American Economic Journal: Economic Policy 11(1): 64-94. https://doi.org/10.1257/pol.20170466
7. Chakrabarti, Rajashri; Gorton, Nicole; Lovenheim, Michael F. (2020). State Investment in Higher Education: Effects on Human Capital Formation, Student Debt, and Long-term Financial Outcomes of Students. NBER Working Paper 27885. https://doi.org/10.3386/w27885
8. Deming, David J.; Yuchtman, Noam; Abulafi, Amira; Goldin, Claudia; Katz, Lawrence F. (2016). The Value of Postsecondary Credentials in the Labor Market: An Experimental Study. American Economic Review 106(3): 778-806. https://doi.org/10.1257/aer.20141757

## Provenance

- Pipeline version: 0.1.0
- Build script: book2/chapters/college-access-price-and-earnings/build.py (sha256 43d2eb8568920689db48ea6ab32cc0303f31011833ee609f5453951b36c0c8a3)
- Built on: 2026-10-07
- Site commit: c2b1ff3309cb4a10833f884da91ac0057a66ad0a

Inputs:

| File | Size (bytes) | SHA-256, first 1 MB | SHA-256, last 1 MB |
| --- | ---: | --- | --- |
| book2/chapters/college-access-price-and-earnings/access.py | 12,763 | c41517d82831379c71a5432d939717de2c0b9f27937c3a3374de51f3f3580abd | c41517d82831379c71a5432d939717de2c0b9f27937c3a3374de51f3f3580abd |
| book2/chapters/college-access-price-and-earnings/colleges.py | 17,758 | 57ddd52c9890742cc2de4cee90f22a88b5e0ae024bb44f588a467f5b9bd9051d | 57ddd52c9890742cc2de4cee90f22a88b5e0ae024bb44f588a467f5b9bd9051d |
| data/analytic/tract_atlas.parquet | 39,659,925 | 29d5e4d943552d54aeb6ba92aadc939d9bd15bdea553c78357dbd516757be627 | 1e10d1ecf722340f23ee2212d4999a90250f85b65fa89b39e32db06d9472e22e |
| data/geo/tracts_2020.parquet | 619,858,391 | f4f05c8c7a63d1785ec2aad283de3f6fa839b6c437f13a1b96cea31d3344bf0f | 7b3811247c6b3e322954f24a4a2384af52aef6ed8688f8dbdb0407d21ef80a59 |
| data/raw/acs/deprivation_components_2019_2023/tract.parquet | 13,197,793 | 5545cde2cbe96a8b56e468d65388248a241aa6bd300d3c9e325d08d3f86b11da | 5c8ec454d066988b0f1a01aaa6ed34ebfc05d162e01cf9c6bb01f9198cbaff12 |
| data/staged/adi/adi_atlas_tract2020_2023.parquet | 798,078 | 5e409bc212ff871def780fa27b4cf8b616d9da420b2b9f6455517eddcd8c1a95 | 5e409bc212ff871def780fa27b4cf8b616d9da420b2b9f6455517eddcd8c1a95 |
| data/staged/crosswalks/ct_tract2020_to_old_county.parquet | 15,842 | 961ef39f31596ef65fca2c67d5f2f63e533c5dc9b49745d3bb6437844e4b831d | 961ef39f31596ef65fca2c67d5f2f63e533c5dc9b49745d3bb6437844e4b831d |
| data/staged/crosswalks/ers_rucc_county_2023.parquet | 67,278 | ad01d36a53664a2bfafbcbebb0f5a921f5b81122ffa80cf96333db23b41c191e | ad01d36a53664a2bfafbcbebb0f5a921f5b81122ffa80cf96333db23b41c191e |
| data/staged/education/college_scorecard_institution_2025.parquet | 976,838 | 34b52fafbdc92d73adc16be4cc9cbb7c2b14c0720c809c8d1e93f1ac5d795057 | 34b52fafbdc92d73adc16be4cc9cbb7c2b14c0720c809c8d1e93f1ac5d795057 |
| data/staged/education/ipeds_institution_2023.parquet | 442,010 | c863b182a558dffa7506917e4f8d323dbf9a6f83d7474cad2005bdecfc2edd9f | c863b182a558dffa7506917e4f8d323dbf9a6f83d7474cad2005bdecfc2edd9f |
| data/staged/ses/acs5_tract2020_2019_2023.parquet | 7,232,424 | 5fe6f730a35225f14d5d8abfc6b9bce58b52e545f309a7eac74f1dae74d7a31c | 7b4a2475b3852d3f91804716a04566ef8ceadacec2be9fb85f5eb642078efac6 |
| data/staged/svi/svi_tract2020_2022.parquet | 17,740,415 | cb3860d8edc2aaf1f2a02211a58f7a0b80457eb6ad9a37b5ec8abb2d5da9da10 | e371a4a696d5bed05d54ad7c5b6ca2504508805b847ec415c2322be7f66ec2b2 |

Outputs:

| File | SHA-256 |
| --- | --- |
| facts.json | a880df2b6f42b0dd852789f716c41a3f6a0d06c99f5404b1c9ff66f4f10896c8 |
| figures/alt_rankings.json | 565c3e97445628662450b630a7df8f411aca8527313130186ce0589f5066fbc7 |
| figures/college_desert_gradient.json | 836f2c34fa05985f2ddb99b36c36eae630814edf58392f81ec21fafc12ac0327 |
| figures/college_distance_map.json | 32e2ff161c8c5221323e8088f644a404aa6755bd925d71716d02507da247c7c5 |
| figures/earnings_by_sector.json | 50e0c558fb4efe0175844f824b1c6451318b27d5e056f71b9126e1cbaf98cac9 |
| figures/earnings_vs_cost.json | 852a1dab9c85c3de40b1c30718997814f62a89f9c6c12861d61f7cb8a69909ea |
| figures/high_pell_colleges.json | 0114c4224a3a76af515fbff3466930f42612cbd1fbf3b56ba2dba3805b9d7ed6 |
| figures/net_price_by_income.json | 85b709559293810341bd86cac52bbf1ca408005aa8946c05e96b1e031b493c31 |
| meta.json | 8086fd5b7627686ba8fc6e7e00c6852aa9f649d1290d7087a720ebdea139484e |
| public/data/college-access-price-and-earnings/college_distance_map.json | c1d07cbd24a32cf9afaddd600b83d1a926e4ae0a7254ccbe7ee0c599ebe53678 |
| public/data/college-access-price-and-earnings/earnings_vs_cost.json | 9a3d0099c7b076c3fad2d95cf888dac1a1c95d1480c5c58664a5dd07a6b7d48f |
| refs.json | b2137e65356026747b7b5c0c4ea3e9ccd9a6df23c01648abba2c5f204cea2911 |

## Machine formats

- Report (HTML): https://vishalsingh.org/zipcode/college-access-price-and-earnings/report
- Report (Markdown): https://vishalsingh.org/zipcode/college-access-price-and-earnings/report.md
- Every estimate (JSON): https://vishalsingh.org/zipcode/report/college-access-price-and-earnings/facts.json
- Table 1, figure "earnings_by_sector" (JSON): https://vishalsingh.org/zipcode/report/college-access-price-and-earnings/figures/earnings_by_sector.json
- Table 2, figure "net_price_by_income" (JSON): https://vishalsingh.org/zipcode/report/college-access-price-and-earnings/figures/net_price_by_income.json
- Table 3, figure "earnings_vs_cost" (JSON): https://vishalsingh.org/zipcode/report/college-access-price-and-earnings/figures/earnings_vs_cost.json
- Table 4, figure "high_pell_colleges" (JSON): https://vishalsingh.org/zipcode/report/college-access-price-and-earnings/figures/high_pell_colleges.json
- Table 5, figure "college_distance_map" (JSON): https://vishalsingh.org/zipcode/report/college-access-price-and-earnings/figures/college_distance_map.json
- Table 6, figure "college_desert_gradient" (JSON): https://vishalsingh.org/zipcode/report/college-access-price-and-earnings/figures/college_desert_gradient.json
- Table 7, figure "alt_rankings" (JSON): https://vishalsingh.org/zipcode/report/college-access-price-and-earnings/figures/alt_rankings.json
- Payload of "earnings_vs_cost" (JSON): https://vishalsingh.org/zipcode/data/college-access-price-and-earnings/earnings_vs_cost.json
- Payload of "college_distance_map" (JSON): https://vishalsingh.org/zipcode/data/college-access-price-and-earnings/college_distance_map.json
- Index of every report: https://vishalsingh.org/zipcode/llms.txt; all reports in one file: https://vishalsingh.org/zipcode/llms-full.txt

---

# Teen births: report

Chapter 4.2 of ZIP Code Destiny, by Vishal Singh (NYU Stern). Generated from the chapter's data files; the reading version is at https://vishalsingh.org/zipcode/teen-births. Table n of this report is figure n of the chapter.

## Identification

- Title: Teen births
- Label: A generation back
- Chapter: 4.2 (Part IV, Becoming an adult; An American life)
- Question: How do teen birth rates change with neighborhood deprivation, and how large is the gap between groups raised in the same tract?
- Dek: The Opportunity Atlas predicts a teen birth for 38.5% of women born in 1978 to 1983 to parents at the 25th income percentile and raised in tracts now in the most deprived tenth, against 12.9% in the least deprived.
- Grain: neighborhoods
- Status: published
- Published: 2026-10-09
- Updated: not revised
- Data sources:
  - Opportunity Atlas. Predicted share of women with a teen birth (parents at the 25th income percentile) by tract and race and by county, children born 1978 to 1983, native 2010 tracts; women's adult household income rank; 1990 tract poverty rate from the tract covariates file; years: 1978-1983 birth cohorts (2018 release); weights: 2010 census tract population for tract comparisons; 2010 census county population for county summaries
  - Area Deprivation Index. Neighborhood Atlas Area Deprivation Index, 2023 release, national rank, 2020 tracts averaged onto 2010 tracts by land area; years: 2023 release (version 4.0.1; built on ACS 2019-2023); weights: 2010 census tract population
  - USDA Food Access Research Atlas. 2010 census population of each 2010 tract, from the Food Access Research Atlas file; the food-access flags are not used; years: 2019 release (population from the 2010 census); weights: none (census counts); used as the weight for tract and county comparisons
  - CDC WONDER natality. Named only, to say how this chapter's measure differs from CDC WONDER counts of births to teenage mothers in the chapter on recent births; years: 2018-2024; weights: none (not used in this chapter's estimates)
  - Published estimates from cited studies. The estimate of the Colorado Family Planning Initiative reported by Lindo and Packham (2017); years: as reported in each study; weights: none (reported estimate)
- Canonical URL: https://vishalsingh.org/zipcode/teen-births
- Report: https://vishalsingh.org/zipcode/teen-births/report
- Markdown: https://vishalsingh.org/zipcode/teen-births/report.md
- BibTeX key: singh2026teenbirths

```bibtex
@misc{singh2026teenbirths,
  author = {Singh, Vishal},
  title = {Teen births},
  year = {2026},
  url = {https://vishalsingh.org/zipcode/teen-births},
  note = {In ZIP Code Destiny. Published 9 October 2026.}
}
```

## Principal estimates

The facts the dek and the opening prose cite, in order of appearance.

| Estimand | Population | Period | Value | Interval | n | Weights | Source | Fact id |
| --- | --- | --- | ---: | --- | ---: | --- | --- | --- |
| predicted share of women with a teen birth, population-weighted mean over tracts (2010 tract population as the weight), county-bootstrap interval; tenths hold near-equal numbers of tracts | all women raised by parents at the 25th percentile of the national income distribution, in the most deprived tenth of tracts by ADI rank | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 | 38.5% | 95% CI 37.8% to 39.3% | 7,226 tracts | 2010 census tract population | Opportunity Atlas | d10_pooled |
| predicted share of women with a teen birth, population-weighted mean over tracts (2010 tract population as the weight), county-bootstrap interval; tenths hold near-equal numbers of tracts | all women raised by parents at the 25th percentile of the national income distribution, in the least deprived tenth of tracts by ADI rank | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 | 12.9% | 95% CI 11.9% to 14.0% | 7,103 tracts | 2010 census tract population | Opportunity Atlas | d1_pooled |
| lower age limit in the Atlas definition of a teen birth (the mother's age at the child's birth) | women in the Opportunity Atlas teen birth measure | women born 1978 to 1983, who had teen births in the 1990s and early 2000s | 13 |   |   | none (definition) | Opportunity Atlas | age_low |
| upper age limit in the Atlas definition of a teen birth (the mother's age at the child's birth) | women in the Opportunity Atlas teen birth measure | women born 1978 to 1983, who had teen births in the 1990s and early 2000s | 19 |   |   | none (definition) | Opportunity Atlas | age_high |
| mean over tracts of the gap in predicted teen birth shares, population-weighted mean over tracts (2010 tract population as the weight), county-bootstrap interval | non-Hispanic Black women and non-Hispanic white women raised by parents at the 25th percentile of the national income distribution, in tracts where the Atlas estimates both, with an ADI rank and a recorded 2010 population | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 | 17.6 | 95% CI 17.0 to 18.2 | 20,884 tracts | 2010 census tract population | Opportunity Atlas | gap_bw |
| mean over tracts of the gap in predicted teen birth shares, population-weighted mean over tracts (2010 tract population as the weight), county-bootstrap interval | Hispanic women of any race and non-Hispanic white women raised by parents at the 25th percentile of the national income distribution, in tracts where the Atlas estimates both, with an ADI rank and a recorded 2010 population | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 | 10.9 | 95% CI 10.3 to 11.5 | 22,695 tracts | 2010 census tract population | Opportunity Atlas | gap_hw |
| mean over tracts of the gap in predicted teen birth shares, population-weighted mean over tracts (2010 tract population as the weight), county-bootstrap interval | non-Hispanic Asian women and non-Hispanic white women raised by parents at the 25th percentile of the national income distribution, in tracts where the Atlas estimates both, with an ADI rank and a recorded 2010 population | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 | 7.1 | 95% CI 6.3 to 8.0 | 7,064 tracts | 2010 census tract population | Opportunity Atlas | gap_wa |

## Findings

Among women raised by parents at the 25th percentile of the national income distribution, the Opportunity Atlas predicts that 38.5% (95% CI 37.8% to 39.3%) of those who grew up in the most deprived tenth of census tracts had a teen birth, against 12.9% (95% CI 11.9% to 14.0%) in the least deprived tenth.[1] Tracts are ranked by their deprivation in 2019 to 2023, a generation after the births. The women were born in 1978 to 1983, so their births fell in the 1990s and early 2000s. Every predicted share in this chapter is the Atlas's model prediction for that cohort, now in middle age, and not a current teen birth rate. A teen birth here means that a woman claimed, as a dependent, a child born when she was between the ages of 13 and 19.

Inside the same tract, the prediction differs by race. Where the Atlas estimates both groups, Black women's predicted share exceeds white women's by 17.6 (95% CI 17.0 to 18.2) points and Hispanic women's exceeds it by 10.9 (95% CI 10.3 to 11.5) points, while Asian women's falls below it by 7.1 (95% CI 6.3 to 8.0) points. The Black-white gap shows no detectable change from the least to the most deprived tenth of tracts, so in these data the gradient by deprivation and the gap by race run in parallel.

### How the Atlas measures a teen birth

The Opportunity Atlas links tax and census records to follow children born in 1978 to 1983 from the census tract where they grew up into adulthood.[1] Its teen birth measure is the fraction of women who grew up in a tract and ever claimed, as a dependent, a child born when they were between the ages of 13 and 19. The Atlas reports it as a predicted value for women whose parents were at a given percentile of the national income distribution, here the 25th, so it describes women from low-income families and not every woman in the tract. A birth counts only when the mother claimed the child, so births to teenage mothers whose children someone else claimed are missed.

The Atlas adds noise to each estimate to protect privacy, and its documentation puts the noise at typically less than 10% of an estimate's standard error. Tract estimates still rest on modest numbers of women, and the staged file carries no standard errors, so a single tract is a modeled estimate with an unknown error. Nothing is clipped, so 451 of the 71,646 tracts with an estimate for all women have a negative value.

Tract comparisons weight each tract by its 2010 census population, the nearest available stand-in for the number of women, because the staged file has no counts of women. That weight counts adults and children alike. Intervals come from a bootstrap that resamples whole counties, or whole states where a quantity is built from county or regional means. Of the 73,278 tracts in the file, 71,303 have an estimate for all women, a deprivation rank (described below) and a recorded 2010 population, and they hold 99.0% of the 2010 U.S. population. The Atlas defines its race groups as Hispanic women of any race and non-Hispanic white, Black and Asian women, and the chapter uses those groups.[2]

The chapter on recent births, Chapter 2.4, Where births are falling (https://vishalsingh.org/zipcode/where-births-are-falling), counts births to mothers aged 15 to 19 in each county for 2018 to 2024 and reports no rates, because the staged files hold no counts of women by age. The measure here differs in its population (women raised in a tract, not mothers living in a county), its period (births in the 1990s and early 2000s), its age band (from age 13) and its form (a predicted share of women, not a count of births). The two should not be read as one series.

### Predicted shares are higher in the South than in any other region

Across the 3,125 counties of the states and the District of Columbia that have an Atlas estimate, the population-weighted mean predicted share is 28.5% (95% CI 26.4% to 30.8%). County values run from 16.0% at the population-weighted 5th percentile to 41.4% at the 95th (Table 1).

Pooled over its counties, the South's share is 33.5% (95% CI 31.4% to 35.6%), against 29.5% (95% CI 26.3% to 31.5%) in the Midwest, 25.0% (95% CI 23.5% to 29.2%) in the West and 21.7% (95% CI 19.7% to 23.0%) in the Northeast.

The South is higher than the Midwest by 4.0 (95% CI 1.2 to 7.6) points, than the West by 8.5 (95% CI 4.1 to 11.1) and than the Northeast by 11.8 (95% CI 9.3 to 14.6). Of the counties in the highest tenth of county values, 93% are in the South, which holds 45% of all counties.

The key of the map ends at about the 5th and 95th percentiles of the county values, so 523 of the 3,126 drawn counties lie outside it and take the color of its nearest end.

The Atlas's county mean is 3.6 (95% CI 3.2 to 3.9) points above the mean of the tract estimates in the same counties, which is 25.0% (95% CI 24.3% to 25.6%) for all women when each tract is weighted by its 2010 population. One likely source is the tract weights, since total population counts adults and children alike and the Atlas's own estimates count the cohort's children. The levels from tract estimates in this chapter, including the shares in the most and least deprived tenths, may therefore be low. Differences between groups are less sensitive to the weights. Counting each tract once, which changes them most, gives a Black-white gap of 17.9 (95% CI 17.4 to 18.5) points against 17.6 with population weights. A county average also mixes neighborhoods that differ widely, so the rest of the chapter works with tracts and, at the end, with county averages of tracts.

Table 1. Predicted teen birth shares are higher in the South than in any other region, 33.5% against 21.7% in the Northeast

### The share rises with neighborhood deprivation in every tenth of tracts

Deprivation here is the Area Deprivation Index (ADI), a national rank built for census block groups from American Community Survey measures.[3] A 2020 tract's value is the unweighted mean of its block groups' ranks and describes neighborhoods in 2019 to 2023, while the Atlas follows women from 2010 tracts whose teen births took place a generation earlier. Each 2010 tract therefore takes the land-area-weighted mean of the values of the 2020 tracts that overlap it, when those with a rank cover at least 50% of its land area. That rule leaves 343 tracts with an estimate but no rank or no recorded 2010 population. Of the tracts that remain, 56.4% overlap exactly one 2020 tract, where the crosswalk adds no further averaging.

The time gap changes what the gradient means. It compares the tract a woman grew up in with that tract's conditions now, and it matches the childhood tract only to the extent that neighborhoods keep their standing. The predicted share and the ADI rank have a correlation of +0.58 (95% CI 0.55 to 0.60) across the 71,303 tracts, and the share rises by 2.5 (95% CI 2.4 to 2.7) points for each ten-rank step in the ADI.

In tenths of tracts, the mean rises at every step, from 12.9% (95% CI 11.9% to 14.0%) in the least deprived tenth to 38.5% (95% CI 37.8% to 39.3%) in the most deprived, a rise of 25.6 (95% CI 24.3 to 26.9) points (Table 2). The mean over all tracts is 25.0% (95% CI 24.3% to 25.6%), and counting each tract once gives 25.5% (95% CI 24.7% to 26.2%).

Table 2. Predicted teen birth shares rise from 12.9% in the least deprived tenth of tracts to 38.5% in the most deprived

Three checks bear on the time gap. First, the correlation with the ADI is +0.56 (95% CI 0.53 to 0.59) among tracts that overlap one 2020 tract and +0.60 (95% CI 0.57 to 0.62) among tracts that overlap several, so the averaging in the crosswalk does not produce it. Second, the predicted share also has a correlation of −0.72 (95% CI −0.74 to −0.71) with women's mean adult household income rank in the same tracts, which the Atlas measures for the same cohort and which Chapter 4.3, Race, income and mobility (https://vishalsingh.org/zipcode/race-income-and-mobility) takes up. That relation is larger in size than the one with the ADI, so teen births and the later outcomes of the same women move together more closely than teen births and present-day deprivation do. Both are patterns across places, and neither says that a teen birth lowered any woman's income.

Third, a deprivation measure from the childhood years allows a direct check. The Atlas's own tract covariates include the poverty rate from the 1990 census, when the women were in middle childhood. In the 71,295 tracts that have it, tenths of the 1990 poverty rate give a rise in the predicted share of 24.1 (95% CI 22.7 to 25.5) points from the least to the most deprived tenth. Tenths of the ADI on the same tracts give 25.6 (95% CI 24.3 to 26.9), a difference of −1.5 (95% CI −2.8 to −0.2) points.

The correlation with the poverty rate is +0.53 (95% CI 0.51 to 0.55), against +0.58 (95% CI 0.55 to 0.60) with the ADI. A measure taken when the women were children gives a gradient of nearly the same size, a little smaller than the ADI's, so the gradient does not depend on the ADI being measured later.

### The share rises across the deprivation scale for every group

The pooled estimate mixes groups, and Table 3 separates them. Each group is drawn in the tracts where the Atlas has enough women of that group to estimate. The mean predicted share over those tracts is 41.3% (95% CI 40.4% to 42.1%) for Black women, 31.0% (95% CI 29.8% to 32.3%) for Hispanic women, 20.0% (95% CI 19.3% to 20.7%) for white women and 7.4% (95% CI 6.5% to 8.7%) for Asian women.

Every group's share is higher in the most deprived tenth of tracts than in the least deprived. The rise is 21.7 (95% CI 20.7 to 22.8) points for white women, 18.1 (95% CI 16.9 to 19.9) for Black women, 20.3 (95% CI 18.1 to 22.2) for Hispanic women and 15.8 (95% CI 11.3 to 20.1) for Asian women, whose interval is wide because few tracts with an Asian estimate lie in the most deprived tenths. Black women's line is above the others at every tenth, followed by Hispanic, white and Asian women's.

Table 3. Every group's predicted teen birth share is higher in the most deprived tenth of tracts than in the least deprived

The pooled gradient is steeper than each group's own. The rise for all women exceeds the rise for white women by 3.9 (95% CI 2.3 to 5.2) points, for Hispanic women by 5.3 (95% CI 3.7 to 7.1), for Black women by 7.5 (95% CI 6.0 to 8.9) and for Asian women by 9.8 (95% CI 5.4 to 14.4).

Tracts with a Black estimate are concentrated in the more deprived tenths, with 4,292 tracts in the most deprived tenth against 1,110 in the least deprived. A tract's pooled estimate averages over its women, so a group mix that differs across the scale would change the slope of the pooled line. The data do not size how much of the pooled gradient comes from that mix.

### Black, Hispanic and Asian women raised in the same tract as white women

A comparison inside one tract removes the differences between neighborhoods that a county or ZIP code average mixes in. It needs both groups' estimates in a tract, and the Atlas makes an estimate for a group only where it can measure enough women of that group, so these tracts are not a sample of all tracts. For Black and white women there are 20,884 such tracts, which hold 99.6 million people.

In those tracts Black women's predicted share averages 40.9% (95% CI 40.1% to 41.7%) against 23.3% (95% CI 22.6% to 24.0%) for white women, a gap of 17.6 (95% CI 17.0 to 18.2) points. Counting each tract once gives 17.9 (95% CI 17.4 to 18.5), and the Black share is higher than the white share in 89.6% (95% CI 88.9% to 90.2%) of the tracts.

 (Note: Each tract estimate rests on a modest number of women, and the Atlas adds noise to protect privacy. Noise of that kind spreads the two groups' estimates in both directions, so it pulls the share of tracts with a higher Black estimate toward an even split and not away from one.)

Hispanic women's predicted share is higher than white women's by 10.9 (95% CI 10.3 to 11.5) points across 22,695 tracts that hold 114.3 million people, and it is higher in 79.1% (95% CI 77.9% to 80.2%) of them. Counting each tract once gives a gap of 11.2 (95% CI 10.6 to 11.8) points.

Asian women's share is lower than white women's by 7.1 (95% CI 6.3 to 8.0) points across 7,064 tracts that hold 37.9 million people, and it is higher than white women's in 23.9% (95% CI 21.6% to 26.0%) of them (Table 4). Each comparison uses its own set of tracts, so the three gaps are not measured on one sample. The same comparison for adult income is in Chapter 4.3, Race, income and mobility (https://vishalsingh.org/zipcode/race-income-and-mobility), and for incarceration in Chapter 3.4, School discipline and incarceration (https://vishalsingh.org/zipcode/school-discipline-and-incarceration).

Table 4. Black women's predicted teen birth share was 17.6 points above white women's in the same tracts, Hispanic women's 10.9 points above and Asian women's 7.1 points below

### The Black-white gap shows no detectable change across the deprivation scale

Sorting tracts on one noisy estimate and then plotting a difference that contains it builds a contrast into the groups, because noise in that estimate enters both the sort and the difference. The tenths in Table 5 are tenths of the ADI, which neither group's estimate contributes to, so the grouping does not sort on either one. In the least deprived tenth the Black-white gap is 18.1 (95% CI 16.2 to 19.6) points, and in the most deprived tenth it is 17.4 (95% CI 16.6 to 18.1) points. The gap in the most deprived tenth minus the gap in the least deprived is −0.7 (95% CI −2.4 to 1.2) points, and the slope of the gap on the ADI rank is −0.1 (95% CI −0.3 to 0.1) points for each ten-rank step. Both intervals include zero, so the data show no detectable change in the gap across the scale.

As a ratio the gap does change. Since the gap in points shows no detectable change while both levels rise, Black women's share is 2.5 (95% CI 2.3 to 2.9) times white women's in the least deprived tenth and 1.5 (95% CI 1.5 to 1.6) times in the most deprived.

Table 5. The Black-white gap in predicted teen birth shares shows no detectable change from the least to the most deprived tenth of tracts, −0.7 points with an interval that includes zero

For Hispanic women the gap is 11.5 (95% CI 10.4 to 12.7) points in the least deprived tenth and 10.7 (95% CI 9.7 to 11.7) in the most deprived, a change of −0.8 (95% CI −2.4 to 0.7) points that also includes zero. The Hispanic ratio falls from 2.1 (95% CI 2.0 to 2.3) to 1.3 (95% CI 1.3 to 1.4).

The Asian-white difference is not constant. It changes by −7.0 (95% CI −12.0 to −2.7) points between the two tenths, so Asian women's shortfall is wider in the most deprived tenth, though it rests on far fewer tracts there. The gradient by deprivation and the gap by race therefore appear side by side, and the gap in points shows no detectable narrowing where deprivation is higher. Black women in the least deprived tenth of tracts have a predicted share of 29.8% (95% CI 28.0% to 31.1%), slightly below the 32.0% (95% CI 31.4% to 32.7%) of white women in the most deprived tenth, whose tracts are different ones. The difference between them is −2.3 (95% CI −4.2 to −0.8) points.

### The gap in county averages is similar under every floor on tracts

A county average damps the noise in a pair of tract estimates. Among the 261 counties with at least 20 tracts that have both estimates, the Black-white gap in county means is 18.1 (95% CI 16.8 to 19.6) points, weighting each county by its population and resampling states. It is positive in every one of those counties (Table 6).

Across counties, Black women's mean share rises 0.85 (95% CI 0.70 to 1.00) points for each point of white women's share. In the tracts of those counties the slope is 0.41 (95% CI 0.35 to 0.46) across tracts and 0.26 (95% CI 0.23 to 0.29) within counties, after each county's means are subtracted from both shares. The lower tract slopes are consistent with noise in a tract's white estimate, which pulls a slope toward zero, and with differences in structure within and between counties. The data do not separate the two. The figure therefore averages tracts to counties before it compares them, and a line of tract means grouped on the white estimate would mix the two.

Table 6. Across counties, Black women's predicted teen birth share rises 0.85 points for each point of white women's, against 0.26 among tracts within the same counties

The county number depends on the floor on tracts. With every county that has a tract pair the gap is 17.6 (95% CI 16.5 to 19.0) points, and it is positive in 94.6% of those counties.

With a floor of 10 tracts (453 counties) the gap is positive in every county, and with a floor of 50 tracts it is 18.1 (95% CI 17.0 to 19.7) points.

A floor on the population of a tract moves the tract gap little. The gap is 17.6 (95% CI 17.0 to 18.2) points for tracts of at least 2,000 residents, 17.3 (95% CI 16.7 to 17.9) for 4,000 and 16.5 (95% CI 15.8 to 17.2) for 6,000.

Counting each tract once, which gives the smallest and noisiest tracts more weight, gives 17.9 (95% CI 17.4 to 18.5) points against 17.6 (95% CI 17.0 to 18.2) with population weights, so noise in small tracts does not move the gap. Under all nine settings the interval for the gap lies above zero (Table 7). The two extreme settings, tracts of at least 6,000 residents and counties with at least 50 tracts, have overlapping intervals.

Table 7. The Black-white gap in predicted teen birth shares has an interval above zero under all nine floors on tracts per county and on tract population

### Evidence on whether contraceptive access changes teen birth rates

Everything above describes shares across places and groups. Whether a change in access to contraception changes teen births is a separate question, and one study with a stated identification strategy bears on it.

> Causal evidence: Whether expanding access to long-acting contraception changes teen birth rates (Difference-in-differences)
>
> The Colorado Family Planning Initiative began in 2009 with private funding to help low-income women gain access to long-acting reversible contraception, such as intrauterine devices and implants, through the state's Title X family-planning clinics. Teen birth rates fell across the United States in the same years, so the fall in Colorado alone does not show what the program did. Lindo and Packham compared the change in teen birth rates in Colorado counties that had a Title X clinic with the change in counties that had a Title X clinic in other states, before and after the program began.[4] The design assumes that the proportional changes in the comparison counties show what would have happened in Colorado without the program.
>
> They estimate that the program reduced the teen birth rate in counties with funded clinics by 6.4% over 5 years, with the reduction concentrated in the second through fifth years and in counties with higher poverty. A state-level synthetic control analysis points the same way but lacks statistical power. The estimate is for teenagers in counties with a clinic in the years around 2009. It does not describe the women born in 1978 to 1983 whose teen births this chapter predicts, and it does not test whether access to contraception accounts for any gap between groups or places shown above.

### The share rises across the tenths of tracts under all four rankings

Table 8 ranks the 66,005 2010 tracts that have an estimate for all women and all four rankings in four ways, with each ranking carried to 2010 tracts by the same land-area crosswalk as the ADI. Ranked by the ADI on this common sample, the predicted share of women is 12.9% (95% CI 11.9% to 13.8%) in the least deprived tenth and 38.2% (95% CI 37.5% to 38.9%) in the most deprived, and the rank correlation between tenth and share is +1.00. Under the standardized index the two ends are 12.0% (95% CI 11.3% to 12.9%) and 38.8% (95% CI 37.6% to 40.0%), with a correlation of +1.00. Under median household income they are 12.7% (95% CI 11.9% to 13.6%) and 38.1% (95% CI 37.4% to 39.0%), and under the bachelor's degree share 13.3% (95% CI 12.2% to 14.6%) and 36.8% (95% CI 36.0% to 37.6%), with correlations of +1.00 and +1.00, so none of the alternatives counts as a change under the book's rule (Chapter 9.1, How far one ranking goes (https://vishalsingh.org/zipcode/how-far-one-ranking-goes)).

Table 8. The predicted teen birth share is steadily higher toward the more deprived end under the ADI and under each of the three other rankings

### What these data cannot show

The shares are model predictions for women born in 1978 to 1983 whose parents were at the 25th income percentile. They describe teen births in the 1990s and early 2000s among women from low-income families and say nothing about women from richer families or about teenagers now. A birth counts only when the mother claimed the child as a dependent. The gradient is a pattern across tracts, an ecological pattern, and it does not say that living in a deprived tract raised any woman's chance of a teen birth. The ADI describes the tracts decades after the births, and the staged files hold no deprivation measure from the childhood years.

A same-tract comparison removes differences between neighborhoods but not all differences. Within a tract, women raised at one parental income percentile can still differ in family wealth, in schools, in the block they lived on and in how institutions treated them, and the data describe the gap without explaining it. The staged file has no standard errors and no counts of women, so tracts are weighted by their total 2010 population, which may put the levels too low. The tracts with both estimates are those with enough women of both groups. Asian estimates are low and noisy, and 20% of them are below zero.

Whether a teen birth changes a woman's later income is a separate question that this chapter does not test. Kearney and Levine reviewed state and international comparisons. They argue that early births mostly mark limited economic prospects for disadvantaged girls and carry little direct economic cost, which is a reading of patterns across places.[5] Puerto Rico has Atlas estimates but is not drawn on the map. Connecticut appears under its legacy county codes in the Atlas, and it takes its ADI rank from the 2020 planning-region tracts through the crosswalk.

## Tables

### Table 1. Predicted teen birth shares are higher in the South than in any other region, 33.5% against 21.7% in the Northeast

Predicted share of women with a teen birth, percent, by county of childhood, women born 1978 to 1983 whose parents were at the 25th income percentile (teen births in the 1990s and early 2000s, not a current rate), races pooled

No inline rows: the values of Table 1 are in https://vishalsingh.org/zipcode/data/teen-births/teen_birth_county_map.json.

Note: Each gray county has the reason for its missing value in its tooltip. Puerto Rico has estimates in the Atlas but is not drawn.

Note: The county estimates are model-based and carry privacy noise, so single small counties are noisy. The Atlas lists Connecticut under its legacy counties, which are the eight drawn here, and Valdez-Cordova Census Area (split in 2019) is drawn as its two successors with the same value. Alaska is drawn.

Note: The estimates describe where women born in 1978 to 1983 grew up, not the counties today. Regional means are population-weighted over the counties of the states and the District of Columbia, with 95% intervals from a bootstrap over states.

Note: The key runs from about the 5th to the 95th percentile of county values, weighted by population, so 523 of the 3,126 drawn counties lie outside it and take the color of its nearest end.

Source: Opportunity Atlas county outcomes, 1978-1983 birth cohorts; modeled estimates, regional means population-weighted (2010 census population)

Data: https://vishalsingh.org/zipcode/report/teen-births/figures/teen_birth_county_map.json

### Table 2. Predicted teen birth shares rise from 12.9% in the least deprived tenth of tracts to 38.5% in the most deprived

Predicted share of women with a teen birth (percent) against the Area Deprivation Index national rank, one dot per 2010 tract, women born 1978 to 1983 whose parents were at the 25th income percentile (teen births in the 1990s and early 2000s, not a current rate), races pooled

No inline rows: the values of Table 2 are in https://vishalsingh.org/zipcode/data/teen-births/teen_birth_tract_cloud.json.

Note: The dots are a random sample of 5,000 tracts inside the plotted range (seed 12345); every estimate uses all 71,303 tracts. The spread of the dots includes sampling noise and privacy noise in each tract estimate.

Note: The Area Deprivation Index is built for census block groups from American Community Survey measures for 2019 to 2023, and a 2020 tract's value is the unweighted mean of its block groups' ranks. Each 2010 tract takes the land-area-weighted mean of the values of the 2020 tracts that overlap it, when those with a rank cover at least 50% of its land area. Tenths hold near-equal numbers of tracts, with tied ranks kept together.

Note: Shares are the Atlas's predicted share of women with a teen birth, a child born when the mother was between the ages of 13 and 19 whom she claimed as a dependent, for women born in 1978 to 1983 whose parents were at the 25th income percentile. White, Black and Asian women are non-Hispanic and Hispanic women may be of any race. The Atlas adds noise to its estimates to protect privacy and the staged values are never clipped, so a few fall below 0 or above 100.

Source: Opportunity Atlas tract outcomes, 2010 tracts, 1978-1983 birth cohorts, and Area Deprivation Index, 2023 release (ACS 2019-2023); population-weighted (2010 tract population), modeled estimates

Data: https://vishalsingh.org/zipcode/report/teen-births/figures/teen_birth_by_deprivation_tracts.json

### Table 3. Every group's predicted teen birth share is higher in the most deprived tenth of tracts than in the least deprived

Predicted share of women with a teen birth (percent) by tenth of the Area Deprivation Index, tracts where each group is estimated, women born 1978 to 1983 whose parents were at the 25th income percentile (teen births in the 1990s and early 2000s, not a current rate)

| conf_high | conf_low | decile | estimate | n | series |
| ---: | ---: | ---: | ---: | ---: | --- |
| 14.01 | 11.85 | 1 | 12.95 | 7,103 | All women |
| 19.34 | 16.06 | 2 | 17.73 | 7,114 | All women |
| 21.14 | 18.97 | 3 | 20.03 | 7,106 | All women |
| 23.01 | 21.62 | 4 | 22.31 | 7,130 | All women |
| 25.17 | 23.94 | 5 | 24.55 | 7,123 | All women |
| 27.15 | 25.90 | 6 | 26.54 | 7,147 | All women |
| 29.25 | 28.03 | 7 | 28.65 | 7,144 | All women |
| 31.26 | 29.92 | 8 | 30.57 | 7,164 | All women |
| 33.37 | 32.12 | 9 | 32.75 | 7,046 | All women |
| 39.27 | 37.79 | 10 | 38.54 | 7,226 | All women |
| 9.55 | 7.69 | 1 | 8.58 | 6,277 | White women |
| 13.73 | 11.95 | 2 | 12.90 | 6,120 | White women |
| 16.07 | 14.69 | 3 | 15.37 | 6,456 | White women |
| 18.76 | 17.37 | 4 | 18.05 | 6,755 | White women |
| 20.77 | 19.49 | 5 | 20.11 | 6,806 | White women |
| 22.46 | 21.29 | 6 | 21.88 | 6,805 | White women |
| 24.40 | 23.41 | 7 | 23.89 | 6,756 | White women |
| 26.14 | 25.11 | 8 | 25.61 | 6,720 | White women |
| 27.76 | 26.86 | 9 | 27.31 | 6,554 | White women |
| 30.83 | 29.81 | 10 | 30.32 | 5,725 | White women |
| 32.41 | 29.79 | 1 | 31.41 | 1,110 | Black women |
| 35.83 | 30.70 | 2 | 33.39 | 1,951 | Black women |
| 36.72 | 33.53 | 3 | 35.19 | 2,076 | Black women |
| 38.43 | 36.06 | 4 | 37.23 | 2,214 | Black women |
| 40.76 | 38.98 | 5 | 39.86 | 2,533 | Black women |
| 42.40 | 40.31 | 6 | 41.35 | 2,630 | Black women |
| 44.29 | 42.37 | 7 | 43.32 | 2,736 | Black women |
| 45.57 | 44.07 | 8 | 44.82 | 2,747 | Black women |
| 47.29 | 46.16 | 9 | 46.72 | 2,927 | Black women |
| 50.14 | 48.95 | 10 | 49.53 | 4,292 | Black women |
| 24.36 | 20.80 | 1 | 22.41 | 3,548 | Hispanic women |
| 26.47 | 23.48 | 2 | 25.25 | 3,718 | Hispanic women |
| 29.43 | 26.60 | 3 | 28.10 | 3,338 | Hispanic women |
| 32.17 | 29.36 | 4 | 30.75 | 3,032 | Hispanic women |
| 33.90 | 31.20 | 5 | 32.51 | 2,710 | Hispanic women |
| 35.45 | 32.76 | 6 | 34.12 | 2,352 | Hispanic women |
| 37.19 | 34.82 | 7 | 35.96 | 2,089 | Hispanic women |
| 39.20 | 36.94 | 8 | 38.06 | 1,818 | Hispanic women |
| 40.45 | 38.59 | 9 | 39.50 | 1,553 | Hispanic women |
| 43.69 | 41.70 | 10 | 42.69 | 1,627 | Hispanic women |
| 5.28 | 3.65 | 1 | 4.43 | 2,509 | Asian women |
| 7.15 | 5.03 | 2 | 5.99 | 1,778 | Asian women |
| 8.06 | 5.73 | 3 | 6.90 | 1,087 | Asian women |
| 10.08 | 7.03 | 4 | 8.46 | 761 | Asian women |
| 14.26 | 8.81 | 5 | 11.40 | 533 | Asian women |
| 15.51 | 9.42 | 6 | 12.18 | 393 | Asian women |
| 17.45 | 10.13 | 7 | 12.99 | 273 | Asian women |
| 19.74 | 10.30 | 8 | 13.95 | 174 | Asian women |
| 25.62 | 16.66 | 9 | 20.67 | 122 | Asian women |
| 24.51 | 15.73 | 10 | 20.21 | 88 | Asian women |

Note: Each group has its own set of tracts, the tracts where the Atlas has enough women of that group, so the lines differ in part because the places differ. Tenths are tenths of all tracts with an ADI rank, so a tenth means the same part of the deprivation scale for every group. Few tracts with an Asian estimate fall in the most deprived tenths, and their intervals are wide.

Note: The Area Deprivation Index is built for census block groups from American Community Survey measures for 2019 to 2023, and a 2020 tract's value is the unweighted mean of its block groups' ranks. Each 2010 tract takes the land-area-weighted mean of the values of the 2020 tracts that overlap it, when those with a rank cover at least 50% of its land area. Tenths hold near-equal numbers of tracts, with tied ranks kept together.

Note: Shares are the Atlas's predicted share of women with a teen birth, a child born when the mother was between the ages of 13 and 19 whom she claimed as a dependent, for women born in 1978 to 1983 whose parents were at the 25th income percentile. White, Black and Asian women are non-Hispanic and Hispanic women may be of any race. The Atlas adds noise to its estimates to protect privacy and the staged values are never clipped, so a few fall below 0 or above 100.

Source: Opportunity Atlas tract outcomes, 2010 tracts, 1978-1983 birth cohorts, and Area Deprivation Index, 2023 release (ACS 2019-2023); population-weighted (2010 tract population), modeled estimates

Data: https://vishalsingh.org/zipcode/report/teen-births/figures/teen_birth_gradient_by_group.json

### Table 4. Black women's predicted teen birth share was 17.6 points above white women's in the same tracts, Hispanic women's 10.9 points above and Asian women's 7.1 points below

Mean predicted share of women with a teen birth for each group minus the share for white women raised in the same tract, percentage points, tracts where both are estimated, women born 1978 to 1983 whose parents were at the 25th income percentile (teen births in the 1990s and early 2000s, not a current rate)

| conf_high | conf_low | estimate | group | n |
| ---: | ---: | ---: | --- | ---: |
| 18.19 | 17.01 | 17.59 | Black women | 20,884 |
| 11.50 | 10.31 | 10.92 | Hispanic women | 22,695 |
| −6.31 | −7.96 | −7.11 | Asian women | 7,064 |

Note: Each row uses its own set of tracts, the tracts where that group and white women both have an estimate, so the rows are not a single sample. The number of tracts is in each row's tooltip. Each tract estimate rests on a modest number of women and carries privacy noise; the means average many tracts and the intervals resample counties. Shares are the Atlas's predicted share of women with a teen birth, a child born when the mother was between the ages of 13 and 19 whom she claimed as a dependent, for women born in 1978 to 1983 whose parents were at the 25th income percentile. White, Black and Asian women are non-Hispanic and Hispanic women may be of any race. The Atlas adds noise to its estimates to protect privacy and the staged values are never clipped, so a few fall below 0 or above 100.

Source: Opportunity Atlas tract outcomes, 2010 tracts, 1978-1983 birth cohorts, and Area Deprivation Index, 2023 release (ACS 2019-2023); population-weighted (2010 tract population), modeled estimates

Data: https://vishalsingh.org/zipcode/report/teen-births/figures/teen_birth_gaps_same_tract.json

### Table 5. The Black-white gap in predicted teen birth shares shows no detectable change from the least to the most deprived tenth of tracts, −0.7 points with an interval that includes zero

Predicted share of women with a teen birth (percent) for white and Black women raised in the same tracts, by tenth of the Area Deprivation Index, women born 1978 to 1983 whose parents were at the 25th income percentile (teen births in the 1990s and early 2000s, not a current rate)

| Tenth of tracts by ADI | White women (%) | Black women (%) | Gap (points) | Lower limit of the 95% interval for the gap | Upper limit of the 95% interval for the gap | Tracts |
| --- | ---: | ---: | ---: | ---: | ---: | ---: |
| Tenth 1 (least deprived) | 11.7 | 29.8 | 18.1 | 16.2 | 19.6 | 696 |
| Tenth 2 | 14.9 | 32.6 | 17.7 | 16.1 | 19.2 | 1,372 |
| Tenth 3 | 16.7 | 34.2 | 17.5 | 16.3 | 18.7 | 1,634 |
| Tenth 4 | 19.4 | 36.9 | 17.5 | 16.3 | 18.6 | 1,953 |
| Tenth 5 | 21.1 | 39.5 | 18.3 | 17.4 | 19.3 | 2,288 |
| Tenth 6 | 23.0 | 41.0 | 18.0 | 16.9 | 19.3 | 2,366 |
| Tenth 7 | 25.5 | 43.0 | 17.5 | 16.7 | 18.4 | 2,443 |
| Tenth 8 | 27.2 | 44.4 | 17.2 | 16.5 | 17.9 | 2,410 |
| Tenth 9 | 29.4 | 46.5 | 17.0 | 16.4 | 17.7 | 2,570 |
| Tenth 10 (most deprived) | 32.0 | 49.4 | 17.4 | 16.6 | 18.1 | 3,152 |

Note: The tracts are those where the Atlas estimates both groups, and tenths are formed on the ADI, a measure that neither estimate contributes to, so the grouping does not sort on either noisy estimate. Few tracts in the least deprived tenths have a Black estimate. The Area Deprivation Index is built for census block groups from American Community Survey measures for 2019 to 2023, and a 2020 tract's value is the unweighted mean of its block groups' ranks. Each 2010 tract takes the land-area-weighted mean of the values of the 2020 tracts that overlap it, when those with a rank cover at least 50% of its land area. Tenths hold near-equal numbers of tracts, with tied ranks kept together.

Note: Shares are the Atlas's predicted share of women with a teen birth, a child born when the mother was between the ages of 13 and 19 whom she claimed as a dependent, for women born in 1978 to 1983 whose parents were at the 25th income percentile. White, Black and Asian women are non-Hispanic and Hispanic women may be of any race. The Atlas adds noise to its estimates to protect privacy and the staged values are never clipped, so a few fall below 0 or above 100.

Source: Opportunity Atlas tract outcomes, 2010 tracts, 1978-1983 birth cohorts, and Area Deprivation Index, 2023 release (ACS 2019-2023); population-weighted (2010 tract population), modeled estimates

Data: https://vishalsingh.org/zipcode/report/teen-births/figures/teen_birth_black_white_by_deprivation.json

### Table 6. Across counties, Black women's predicted teen birth share rises 0.85 points for each point of white women's, against 0.26 among tracts within the same counties

Mean predicted share of women with a teen birth (percent) for Black women against white women, counties with at least a minimum number of tracts where both are estimated, women born 1978 to 1983 whose parents were at the 25th income percentile (teen births in the 1990s and early 2000s, not a current rate)

| black | n | white |
| ---: | ---: | ---: |
| 46.15 | 20 | 30.63 |
| 40.49 | 102 | 22.94 |
| 41.06 | 20 | 26.32 |
| 45.02 | 44 | 26.40 |
| 46.42 | 78 | 26.99 |
| 47.88 | 46 | 25.45 |
| 50.53 | 20 | 35.96 |
| 45.79 | 32 | 24.85 |
| 27.30 | 29 | 21.66 |
| 44.50 | 159 | 28.23 |
| 35.43 | 26 | 24.18 |
| 42.55 | 75 | 23.05 |
| 33.35 | 132 | 16.09 |
| 37.31 | 74 | 17.54 |
| 45.77 | 52 | 25.11 |
| 53.95 | 44 | 27.91 |
| 37.25 | 385 | 16.12 |
| 31.84 | 28 | 19.12 |
| 38.81 | 123 | 19.37 |
| 40.99 | 161 | 20.13 |
| 41.05 | 231 | 22.79 |
| 32.24 | 173 | 15.74 |
| 39.46 | 52 | 20.60 |
| 31.13 | 60 | 17.42 |
| 35.47 | 72 | 20.29 |
| 36.01 | 66 | 19.74 |
| 40.34 | 43 | 23.57 |
| 39.39 | 62 | 24.88 |
| 31.71 | 60 | 16.42 |
| 32.68 | 65 | 16.93 |
| 36.42 | 66 | 15.75 |
| 36.21 | 20 | 22.77 |
| 45.14 | 27 | 26.23 |
| 38.53 | 78 | 23.58 |
| 47.08 | 27 | 25.17 |
| 40.28 | 34 | 18.67 |
| 39.62 | 21 | 28.46 |
| 37.33 | 53 | 23.28 |
| 28.10 | 221 | 15.23 |
| 43.61 | 133 | 25.87 |
| 44.73 | 57 | 25.71 |
| 40.62 | 151 | 24.78 |
| 38.95 | 20 | 24.17 |
| 37.99 | 28 | 26.15 |
| 35.29 | 48 | 17.81 |
| 46.57 | 29 | 26.17 |
| 33.82 | 139 | 16.23 |
| 39.16 | 23 | 24.51 |
| 34.45 | 126 | 20.97 |
| 33.43 | 20 | 21.45 |
| 32.33 | 132 | 18.17 |
| 49.60 | 50 | 21.91 |
| 49.99 | 69 | 30.88 |
| 34.66 | 29 | 21.80 |
| 34.95 | 51 | 18.72 |
| 38.25 | 38 | 23.95 |
| 47.51 | 33 | 29.74 |
| 42.70 | 42 | 30.11 |
| 47.90 | 22 | 19.28 |
| 42.60 | 45 | 29.42 |
| 36.20 | 95 | 20.96 |
| 37.53 | 83 | 17.92 |
| 45.96 | 22 | 27.45 |
| 37.80 | 70 | 17.97 |
| 33.29 | 101 | 17.40 |
| 40.21 | 25 | 26.42 |
| 42.70 | 23 | 26.18 |
| 46.90 | 20 | 28.80 |
| 41.88 | 40 | 30.43 |
| 43.76 | 40 | 29.55 |
| 47.81 | 21 | 21.12 |
| 46.73 | 286 | 15.50 |
| 38.08 | 45 | 12.72 |
| 55.83 | 41 | 29.78 |
| 49.06 | 31 | 25.64 |
| 56.68 | 20 | 37.76 |
| 45.46 | 24 | 26.33 |
| 52.94 | 28 | 28.40 |
| 43.44 | 35 | 23.96 |
| 47.51 | 25 | 31 |
| 44.21 | 45 | 20.65 |
| 60.46 | 39 | 26.72 |
| 47.07 | 35 | 30.59 |
| 47.64 | 38 | 26.48 |
| 48.05 | 138 | 28.68 |
| 49.49 | 38 | 27.54 |
| 44.74 | 24 | 25.68 |
| 50.25 | 46 | 29.15 |
| 40.02 | 25 | 26.73 |
| 46.97 | 38 | 35.44 |
| 41.04 | 39 | 28.05 |
| 43.81 | 96 | 26.78 |
| 45.23 | 22 | 28.54 |
| 46.39 | 50 | 27.14 |
| 49.44 | 24 | 30.59 |
| 43.70 | 64 | 21.41 |
| 47.46 | 67 | 24.39 |
| 43.91 | 34 | 22.75 |
| 45.24 | 64 | 20.41 |
| 49.49 | 25 | 23.97 |
| 50.47 | 20 | 30.11 |
| 41.02 | 23 | 23.60 |
| 53.53 | 20 | 24.80 |
| 39.03 | 62 | 20.67 |
| 39.15 | 100 | 17.40 |
| 33.73 | 27 | 20.50 |
| 38.25 | 21 | 24.19 |
| 30 | 33 | 13.76 |
| 23.98 | 131 | 13.06 |
| 30.62 | 142 | 22.39 |
| 49.07 | 77 | 19.06 |
| 39.48 | 28 | 22.98 |
| 17.71 | 49 | 12.76 |
| 25.21 | 40 | 15.64 |
| 34.24 | 24 | 19.61 |
| 44.81 | 38 | 24.88 |
| 41.90 | 34 | 24.58 |
| 47.57 | 26 | 23.85 |
| 51.75 | 51 | 26.84 |
| 40.92 | 27 | 17.91 |
| 34.09 | 68 | 13.42 |
| 45.57 | 24 | 24.56 |
| 35.91 | 43 | 15.04 |
| 43.17 | 131 | 23.03 |
| 42.51 | 110 | 16.44 |
| 46.09 | 47 | 21.71 |
| 47.36 | 29 | 30.23 |
| 44.57 | 38 | 30.95 |
| 45.83 | 41 | 21.96 |
| 43.70 | 22 | 29.84 |
| 46.04 | 89 | 26.15 |
| 42.22 | 91 | 22.08 |
| 52.57 | 51 | 23.70 |
| 44.70 | 50 | 26.62 |
| 47.91 | 195 | 26.58 |
| 45.27 | 26 | 20.96 |
| 24.20 | 26 | 9.26 |
| 35.04 | 55 | 17.71 |
| 39.40 | 59 | 16.57 |
| 51.33 | 22 | 25.34 |
| 31.16 | 46 | 6.49 |
| 43.82 | 31 | 16.39 |
| 35.70 | 28 | 12.80 |
| 39.86 | 39 | 15.82 |
| 26.57 | 75 | 9.31 |
| 34.66 | 33 | 16.12 |
| 37.91 | 25 | 14.58 |
| 31.76 | 53 | 8.77 |
| 32.90 | 43 | 10.90 |
| 29.52 | 21 | 21.29 |
| 46.72 | 35 | 22.33 |
| 26.25 | 80 | 4.42 |
| 41.72 | 73 | 19.85 |
| 25.16 | 55 | 9.01 |
| 27.88 | 37 | 8.96 |
| 48.34 | 36 | 20.75 |
| 37.91 | 26 | 17.09 |
| 23.31 | 57 | 7.65 |
| 40.22 | 27 | 10.32 |
| 22.94 | 22 | 7.61 |
| 36.34 | 60 | 12.13 |
| 27.48 | 72 | 10.96 |
| 40.32 | 22 | 30.82 |
| 51.58 | 22 | 26.49 |
| 39.80 | 60 | 27.58 |
| 36.32 | 43 | 18.20 |
| 42.99 | 52 | 24.26 |
| 52.23 | 32 | 34.29 |
| 39.08 | 74 | 23.76 |
| 43.86 | 22 | 25.33 |
| 40.91 | 135 | 22.11 |
| 39.03 | 22 | 21.01 |
| 38.74 | 23 | 24.25 |
| 47.45 | 28 | 22.05 |
| 46.47 | 22 | 35.09 |
| 53.96 | 20 | 29.02 |
| 37.49 | 107 | 16.90 |
| 41.96 | 24 | 25.02 |
| 41.80 | 26 | 25.73 |
| 42.49 | 136 | 18.58 |
| 42.72 | 158 | 29.84 |
| 49.28 | 111 | 26.60 |
| 49.88 | 28 | 30.80 |
| 47.77 | 60 | 28.92 |
| 49.23 | 22 | 24.70 |
| 45.20 | 61 | 27.53 |
| 45.63 | 38 | 27.95 |
| 44.90 | 54 | 23.04 |
| 43.43 | 22 | 29.30 |
| 45.08 | 118 | 26.95 |
| 44.82 | 69 | 25.26 |
| 40.48 | 41 | 20.79 |
| 42.74 | 104 | 17.44 |
| 47.77 | 22 | 25.45 |
| 43.95 | 23 | 16.94 |
| 33.72 | 22 | 18.30 |
| 42.83 | 32 | 23.16 |
| 41.01 | 43 | 16.88 |
| 52.12 | 20 | 26.16 |
| 41.27 | 20 | 23.62 |
| 33.39 | 47 | 12.61 |
| 44.78 | 117 | 18.59 |
| 32.48 | 38 | 22.29 |
| 43.07 | 28 | 26.81 |
| 48.45 | 32 | 33.85 |
| 39.49 | 34 | 28 |
| 38.03 | 58 | 19.37 |
| 38.79 | 23 | 27.03 |
| 39.41 | 29 | 24.35 |
| 45.21 | 78 | 22.28 |
| 42.83 | 33 | 24.60 |
| 37.70 | 36 | 23.75 |
| 40.15 | 61 | 20.29 |
| 49.22 | 51 | 30.54 |
| 43.68 | 21 | 27.47 |
| 48.39 | 31 | 29.91 |
| 42.36 | 108 | 28.79 |
| 45.65 | 45 | 28.07 |
| 45.23 | 42 | 27.66 |
| 47.27 | 20 | 29.81 |
| 41 | 33 | 28.39 |
| 44.09 | 33 | 31.13 |
| 45.26 | 136 | 27.07 |
| 40.68 | 49 | 28.96 |
| 37.09 | 120 | 22.18 |
| 37.15 | 27 | 28.42 |
| 51.38 | 20 | 25.20 |
| 34.29 | 50 | 16.60 |
| 44.31 | 341 | 27.80 |
| 42.70 | 39 | 25.26 |
| 32.49 | 34 | 23.04 |
| 33.41 | 53 | 19.24 |
| 36.49 | 44 | 27.01 |
| 44.89 | 23 | 30.05 |
| 38.59 | 417 | 23.46 |
| 45.03 | 39 | 29.08 |
| 49.41 | 30 | 30.55 |
| 44.64 | 31 | 31.45 |
| 40.88 | 29 | 27.61 |
| 43.27 | 193 | 26.07 |
| 41.50 | 83 | 26.44 |
| 41.40 | 22 | 30.80 |
| 35.42 | 23 | 18.48 |
| 32.41 | 56 | 20.45 |
| 24.38 | 123 | 12.39 |
| 34.92 | 43 | 19.47 |
| 32.86 | 54 | 22.53 |
| 40.44 | 21 | 26.07 |
| 28.47 | 26 | 15.87 |
| 35.37 | 37 | 23.06 |
| 39.96 | 30 | 24.43 |
| 37.68 | 38 | 25.08 |
| 43.53 | 47 | 26.48 |
| 43.70 | 28 | 23.03 |
| 40.88 | 21 | 27.28 |
| 33.59 | 80 | 21.13 |
| 32.44 | 114 | 18.64 |
| 38.83 | 75 | 25.30 |
| 51.85 | 28 | 19.82 |
| 51.01 | 99 | 24.71 |
| 58.29 | 24 | 29.58 |

Note: A county enters when at least 20 of its tracts have both estimates, and county means weight each tract by its 2010 population. In the tracts of these counties the slope is 0.41 across tracts and 0.26 within counties (each tract's shares minus its county's means), against 0.85 between county means. Noise in the white estimate is one candidate explanation for the lower tract slopes, and the data do not separate it from structure within and between counties. Shares are the Atlas's predicted share of women with a teen birth, a child born when the mother was between the ages of 13 and 19 whom she claimed as a dependent, for women born in 1978 to 1983 whose parents were at the 25th income percentile. White, Black and Asian women are non-Hispanic and Hispanic women may be of any race. The Atlas adds noise to its estimates to protect privacy and the staged values are never clipped, so a few fall below 0 or above 100.

Source: Opportunity Atlas tract outcomes, 2010 tracts, 1978-1983 birth cohorts, and Area Deprivation Index, 2023 release (ACS 2019-2023); population-weighted (2010 tract population), modeled estimates

Data: https://vishalsingh.org/zipcode/report/teen-births/figures/teen_birth_black_white_counties.json

### Table 7. The Black-white gap in predicted teen birth shares has an interval above zero under all nine floors on tracts per county and on tract population

Mean predicted share of women with a teen birth for Black women minus white women raised in the same places, percentage points, by floor, women born 1978 to 1983 whose parents were at the 25th income percentile (teen births in the 1990s and early 2000s, not a current rate)

| conf_high | conf_low | estimate | floor | kind | n |
| ---: | ---: | ---: | --- | --- | ---: |
| 18.98 | 16.54 | 17.59 | Every county with a tract pair | Minimum tracts per county | 1,318 |
| 18.98 | 16.66 | 17.74 | Counties with at least 5 tracts | Minimum tracts per county | 719 |
| 19.20 | 16.70 | 17.86 | Counties with at least 10 tracts | Minimum tracts per county | 453 |
| 19.59 | 16.84 | 18.06 | Counties with at least 20 tracts | Minimum tracts per county | 261 |
| 19.67 | 16.96 | 18.14 | Counties with at least 50 tracts | Minimum tracts per county | 102 |
| 18.19 | 17.01 | 17.59 | Every tract | Minimum residents per tract | 20,884 |
| 18.18 | 17.03 | 17.57 | Tracts with at least 2,000 residents | Minimum residents per tract | 19,836 |
| 17.94 | 16.69 | 17.27 | Tracts with at least 4,000 residents | Minimum residents per tract | 12,788 |
| 17.21 | 15.78 | 16.50 | Tracts with at least 6,000 residents | Minimum residents per tract | 4,790 |

Note: County rows weight each tract by its 2010 population within the county and each county by its summed population, with intervals from a bootstrap over states. Tract rows weight each tract by its 2010 population, with intervals from a bootstrap over counties. The number of counties or tracts is in each row's tooltip. Shares are the Atlas's predicted share of women with a teen birth, a child born when the mother was between the ages of 13 and 19 whom she claimed as a dependent, for women born in 1978 to 1983 whose parents were at the 25th income percentile. White, Black and Asian women are non-Hispanic and Hispanic women may be of any race. The Atlas adds noise to its estimates to protect privacy and the staged values are never clipped, so a few fall below 0 or above 100.

Source: Opportunity Atlas tract outcomes, 2010 tracts, 1978-1983 birth cohorts, and Area Deprivation Index, 2023 release (ACS 2019-2023); population-weighted (2010 tract population), modeled estimates

Data: https://vishalsingh.org/zipcode/report/teen-births/figures/teen_birth_gap_floors.json

### Table 8. The predicted teen birth share is steadily higher toward the more deprived end under the ADI and under each of the three other rankings

Predicted share of women with a teen birth (percent) by tenth of tracts ranked four ways, over the 66,005 tracts that have all four rankings

| bin | conf_high | conf_low | estimate | n_units | ranking |
| ---: | ---: | ---: | ---: | ---: | --- |
| 1 | 13.78 | 11.93 | 12.90 | 6,617 | Area Deprivation Index |
| 2 | 19.58 | 16.44 | 18.04 | 6,508 | Area Deprivation Index |
| 3 | 21.02 | 18.99 | 19.99 | 6,622 | Area Deprivation Index |
| 4 | 23.05 | 21.64 | 22.38 | 6,589 | Area Deprivation Index |
| 5 | 25.09 | 23.87 | 24.51 | 6,611 | Area Deprivation Index |
| 6 | 27.15 | 25.87 | 26.52 | 6,604 | Area Deprivation Index |
| 7 | 29.10 | 27.89 | 28.51 | 6,568 | Area Deprivation Index |
| 8 | 31.05 | 29.65 | 30.36 | 6,657 | Area Deprivation Index |
| 9 | 33.17 | 31.88 | 32.53 | 6,613 | Area Deprivation Index |
| 10 | 38.92 | 37.46 | 38.19 | 6,616 | Area Deprivation Index |
| 1 | 12.90 | 11.25 | 12.04 | 6,575 | Standardized index |
| 2 | 17.83 | 16.55 | 17.18 | 6,578 | Standardized index |
| 3 | 20.82 | 19.82 | 20.32 | 6,588 | Standardized index |
| 4 | 22.71 | 21.72 | 22.24 | 6,599 | Standardized index |
| 5 | 24.88 | 23.89 | 24.38 | 6,605 | Standardized index |
| 6 | 26.80 | 25.78 | 26.29 | 6,609 | Standardized index |
| 7 | 28.81 | 27.63 | 28.23 | 6,612 | Standardized index |
| 8 | 31.24 | 29.93 | 30.61 | 6,610 | Standardized index |
| 9 | 34.21 | 32.66 | 33.44 | 6,611 | Standardized index |
| 10 | 40.03 | 37.63 | 38.85 | 6,618 | Standardized index |
| 1 | 13.57 | 11.93 | 12.74 | 6,582 | Median household income |
| 2 | 18.32 | 17.00 | 17.64 | 6,585 | Median household income |
| 3 | 21.17 | 20.09 | 20.63 | 6,592 | Median household income |
| 4 | 23.42 | 22.33 | 22.88 | 6,596 | Median household income |
| 5 | 25.27 | 24.29 | 24.79 | 6,611 | Median household income |
| 6 | 26.94 | 25.94 | 26.45 | 6,606 | Median household income |
| 7 | 28.69 | 27.61 | 28.16 | 6,613 | Median household income |
| 8 | 30.79 | 29.66 | 30.24 | 6,609 | Median household income |
| 9 | 33.58 | 32.39 | 32.99 | 6,611 | Median household income |
| 10 | 38.96 | 37.35 | 38.15 | 6,600 | Median household income |
| 1 | 14.56 | 12.20 | 13.34 | 6,568 | Bachelor's degree share |
| 2 | 17.85 | 16.08 | 16.95 | 6,577 | Bachelor's degree share |
| 3 | 20.72 | 19.21 | 19.95 | 6,596 | Bachelor's degree share |
| 4 | 22.91 | 21.61 | 22.26 | 6,595 | Bachelor's degree share |
| 5 | 25.10 | 23.85 | 24.49 | 6,605 | Bachelor's degree share |
| 6 | 26.76 | 25.63 | 26.20 | 6,604 | Bachelor's degree share |
| 7 | 28.68 | 27.66 | 28.18 | 6,614 | Bachelor's degree share |
| 8 | 30.66 | 29.60 | 30.14 | 6,609 | Bachelor's degree share |
| 9 | 33.22 | 32.04 | 32.63 | 6,620 | Bachelor's degree share |
| 10 | 37.62 | 35.98 | 36.83 | 6,617 | Bachelor's degree share |

Note: Least and most deprived tenth: ADI 12.9% and 38.2%, rank correlation +1.00; standardized index 12.0% and 38.8%, +1.00; median household income 12.7% and 38.1%, +1.00; bachelor's degree share 13.3% and 36.8%, +1.00.

Note: The standardized index applies the ADI's coefficients to its measures after standardizing each one, so home value and income carry less weight than in the ADI. Income and degree share are single census measures, each ranked from the highest value to the lowest.

Source: Opportunity Atlas tract outcomes, 2010 tracts, 1978-1983 birth cohorts, and Area Deprivation Index, 2023 release (ACS 2019-2023); population-weighted (2010 tract population), modeled estimates; ADI 2023 and ACS 2019-2023 tract tables (B19013, B15003 and the 15 rebuilt ADI measures)

Data: https://vishalsingh.org/zipcode/report/teen-births/figures/alt_rankings.json

## All estimates

Every fact in facts.json (128), sorted by id.

| Fact id | Value | Unit | Estimand | Population | Period | Interval | n | Weights | Source | Note |
| --- | ---: | --- | --- | --- | --- | --- | ---: | --- | --- | --- |
| adi_cover | 50% | percent of a 2010 tract's land area | minimum land area of a 2010 tract covered by 2020 tracts that have an ADI rank for the tract to take a rank (a rule) | 2010 census tracts | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 |   |   | none (rule) | Opportunity Atlas |   |
| age_high | 19 | years of age | upper age limit in the Atlas definition of a teen birth (the mother's age at the child's birth) | women in the Opportunity Atlas teen birth measure | women born 1978 to 1983, who had teen births in the 1990s and early 2000s |   |   | none (definition) | Opportunity Atlas |   |
| age_low | 13 | years of age | lower age limit in the Atlas definition of a teen birth (the mother's age at the child's birth) | women in the Opportunity Atlas teen birth measure | women born 1978 to 1983, who had teen births in the 1990s and early 2000s |   |   | none (definition) | Opportunity Atlas |   |
| alt_adi_least | 12.9% | Predicted share of women with a teen birth (percent) | population-weighted mean of the tract predicted teen birth share for all women (2010 tract population as the weight); 95% intervals from a bootstrap over counties | the least deprived tenth of tracts ranked by Area Deprivation Index, among the tracts with all four rankings | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 | 95% CI 11.9% to 13.8% | 6,617 tracts | 2010 census tract population | Opportunity Atlas |   |
| alt_adi_most | 38.2% | Predicted share of women with a teen birth (percent) | population-weighted mean of the tract predicted teen birth share for all women (2010 tract population as the weight); 95% intervals from a bootstrap over counties | the most deprived tenth of tracts ranked by Area Deprivation Index, among the tracts with all four rankings | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 | 95% CI 37.5% to 38.9% | 6,616 tracts | 2010 census tract population | Opportunity Atlas |   |
| alt_adi_rho | +1.00 | rank correlation | Spearman rank correlation between bin order and bin mean, each bin counted once | the bins of tracts ranked by Area Deprivation Index | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 |   |   | each bin counted once | Opportunity Atlas |   |
| alt_degree_least | 13.3% | Predicted share of women with a teen birth (percent) | population-weighted mean of the tract predicted teen birth share for all women (2010 tract population as the weight); 95% intervals from a bootstrap over counties | the least deprived tenth of tracts ranked by Bachelor's degree share, among the tracts with all four rankings | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 | 95% CI 12.2% to 14.6% | 6,568 tracts | 2010 census tract population | Opportunity Atlas |   |
| alt_degree_most | 36.8% | Predicted share of women with a teen birth (percent) | population-weighted mean of the tract predicted teen birth share for all women (2010 tract population as the weight); 95% intervals from a bootstrap over counties | the most deprived tenth of tracts ranked by Bachelor's degree share, among the tracts with all four rankings | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 | 95% CI 36.0% to 37.6% | 6,617 tracts | 2010 census tract population | Opportunity Atlas |   |
| alt_degree_rho | +1.00 | rank correlation | Spearman rank correlation between bin order and bin mean, each bin counted once | the bins of tracts ranked by Bachelor's degree share | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 |   |   | each bin counted once | Opportunity Atlas |   |
| alt_income_least | 12.7% | Predicted share of women with a teen birth (percent) | population-weighted mean of the tract predicted teen birth share for all women (2010 tract population as the weight); 95% intervals from a bootstrap over counties | the least deprived tenth of tracts ranked by Median household income, among the tracts with all four rankings | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 | 95% CI 11.9% to 13.6% | 6,582 tracts | 2010 census tract population | Opportunity Atlas |   |
| alt_income_most | 38.1% | Predicted share of women with a teen birth (percent) | population-weighted mean of the tract predicted teen birth share for all women (2010 tract population as the weight); 95% intervals from a bootstrap over counties | the most deprived tenth of tracts ranked by Median household income, among the tracts with all four rankings | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 | 95% CI 37.4% to 39.0% | 6,600 tracts | 2010 census tract population | Opportunity Atlas |   |
| alt_income_rho | +1.00 | rank correlation | Spearman rank correlation between bin order and bin mean, each bin counted once | the bins of tracts ranked by Median household income | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 |   |   | each bin counted once | Opportunity Atlas |   |
| alt_n_places | 66,005 | tracts | count of tracts in the common sample of the ranking comparison | tracts with the outcome and all four rankings | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 |   |   | none (full count) | Opportunity Atlas |   |
| alt_rho_cut | 0.9 | absolute rank correlation | threshold between a steady and an uneven gradient | the rule for changed gradients (docs/RANKING_SWITCH_RULE.md) | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 |   |   | none (a fixed rule) | Opportunity Atlas |   |
| alt_rho_margin | 0.1 | absolute rank correlation | smallest move in the rank correlation that counts as a change of shape | the rule for changed gradients (docs/RANKING_SWITCH_RULE.md) | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 |   |   | none (a fixed rule) | Opportunity Atlas |   |
| alt_std_least | 12.0% | Predicted share of women with a teen birth (percent) | population-weighted mean of the tract predicted teen birth share for all women (2010 tract population as the weight); 95% intervals from a bootstrap over counties | the least deprived tenth of tracts ranked by Standardized index, among the tracts with all four rankings | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 | 95% CI 11.3% to 12.9% | 6,575 tracts | 2010 census tract population | Opportunity Atlas |   |
| alt_std_most | 38.8% | Predicted share of women with a teen birth (percent) | population-weighted mean of the tract predicted teen birth share for all women (2010 tract population as the weight); 95% intervals from a bootstrap over counties | the most deprived tenth of tracts ranked by Standardized index, among the tracts with all four rankings | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 | 95% CI 37.6% to 40.0% | 6,618 tracts | 2010 census tract population | Opportunity Atlas |   |
| alt_std_rho | +1.00 | rank correlation | Spearman rank correlation between bin order and bin mean, each bin counted once | the bins of tracts ranked by Standardized index | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 |   |   | each bin counted once | Opportunity Atlas |   |
| asian_neg_share | 20% | percent of tracts with an Asian estimate | share of those estimates that are below zero (privacy noise, never clipped) | 2010 census tracts with an Atlas estimate for Asian women | women born 1978 to 1983, who had teen births in the 1990s and early 2000s |   | 7,732 tracts | none (each tract counts once) | Opportunity Atlas |   |
| black_bw | 40.9% | percent of women with a teen birth | predicted share, population-weighted mean over tracts (2010 tract population as the weight), county-bootstrap interval | non-Hispanic Black women raised by parents at the 25th percentile of the national income distribution, in tracts where the Atlas also estimates non-Hispanic white women | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 | 95% CI 40.1% to 41.7% | 20,884 tracts | 2010 census tract population | Opportunity Atlas |   |
| corr_adi | +0.58 | correlation coefficient | population-weighted correlation of the predicted teen birth share with the ADI national rank, county-bootstrap interval | all women raised by parents at the 25th percentile of the national income distribution, in tracts with an Atlas estimate for them, an ADI rank and a recorded 2010 population | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 | 95% CI 0.55 to 0.60 | 71,303 tracts | 2010 census tract population | Opportunity Atlas |   |
| corr_adi_common | +0.58 | correlation coefficient | population-weighted correlation of the predicted teen birth share with the ADI national rank, on the tracts with a 1990 poverty rate, county-bootstrap interval | all women raised by parents at the 25th percentile of the national income distribution, in tracts with a 1990 poverty rate | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 | 95% CI 0.55 to 0.60 | 71,295 tracts | 2010 census tract population | Opportunity Atlas |   |
| corr_adult | −0.72 | correlation coefficient | population-weighted correlation of the predicted teen birth share with women's mean adult household income rank (percentile), both for the same cohort, county-bootstrap interval | tracts with an Atlas estimate of teen births and of women's adult household income rank, women raised by parents at the 25th percentile of the national income distribution | women born 1978 to 1983, who had teen births in the 1990s and early 2000s | 95% CI −0.74 to −0.71 | 71,303 tracts | 2010 census tract population | Opportunity Atlas |   |
| corr_multi | +0.60 | correlation coefficient | population-weighted correlation of the predicted teen birth share with the ADI national rank (rank averaged over the overlapping tracts by land area), county-bootstrap interval | tracts that overlap more than one 2020 tract, with an Atlas estimate for all women | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 | 95% CI 0.57 to 0.62 | 31,104 tracts | 2010 census tract population | Opportunity Atlas |   |
| corr_pov1990 | +0.53 | correlation coefficient | population-weighted correlation of the predicted teen birth share with the 1990 tract poverty rate, county-bootstrap interval | all women raised by parents at the 25th percentile of the national income distribution, in tracts with a 1990 poverty rate | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 | 95% CI 0.51 to 0.55 | 71,295 tracts | 2010 census tract population | Opportunity Atlas |   |
| corr_single | +0.56 | correlation coefficient | population-weighted correlation of the predicted teen birth share with the ADI national rank, county-bootstrap interval | tracts that overlap exactly one 2020 tract, with an Atlas estimate for all women | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 | 95% CI 0.53 to 0.59 | 40,199 tracts | 2010 census tract population | Opportunity Atlas |   |
| cross_black_d1 | 29.8% | percent of women with a teen birth | predicted share of women with a teen birth, population-weighted mean over tracts (2010 tract population as the weight), county-bootstrap interval | non-Hispanic Black women raised by parents at the 25th percentile of the national income distribution, in the least deprived tenth of the tracts where the Atlas estimates both Black and white women | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 | 95% CI 28.0% to 31.1% | 20,884 tracts | 2010 census tract population | Opportunity Atlas |   |
| cross_diff | −2.3 | percentage points of women with a teen birth, Black women in the least deprived tenth minus white women in the most deprived tenth | mean for Black women in the least deprived tenth of tracts minus mean for white women in the most deprived tenth, county-bootstrap interval from the same draws | non-Hispanic Black women and non-Hispanic white women raised by parents at the 25th percentile of the national income distribution, in tracts where the Atlas estimates both, with an ADI rank and a recorded 2010 population | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 | 95% CI −4.2 to −0.8 | 20,884 tracts | 2010 census tract population | Opportunity Atlas |   |
| cross_white_d10 | 32.0% | percent of women with a teen birth | predicted share of women with a teen birth, population-weighted mean over tracts (2010 tract population as the weight), county-bootstrap interval | non-Hispanic white women raised by parents at the 25th percentile of the national income distribution, in the most deprived tenth of the tracts where the Atlas estimates both Black and white women | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 | 95% CI 31.4% to 32.7% | 20,884 tracts | 2010 census tract population | Opportunity Atlas |   |
| cty_mean | 28.5% | percent of women with a teen birth | predicted share of women with a teen birth, population-weighted mean over counties (2010 census population as the weight), bootstrap interval over states | women raised by parents at the 25th percentile of the national income distribution, by county of childhood, in the states and the District of Columbia | women born 1978 to 1983, who had teen births in the 1990s and early 2000s | 95% CI 26.4% to 30.8% | 3,125 counties | 2010 census county population | Opportunity Atlas |   |
| cty_p5 | 16.0% | percent of women with a teen birth | population-weighted 5th percentile of the county values | women raised by parents at the 25th percentile of the national income distribution, by county of childhood, in the states and the District of Columbia | women born 1978 to 1983, who had teen births in the 1990s and early 2000s |   | 3,125 counties | 2010 census county population | Opportunity Atlas |   |
| cty_p95 | 41.4% | percent of women with a teen birth | population-weighted 95th percentile of the county values | women raised by parents at the 25th percentile of the national income distribution, by county of childhood, in the states and the District of Columbia | women born 1978 to 1983, who had teen births in the 1990s and early 2000s |   | 3,125 counties | 2010 census county population | Opportunity Atlas |   |
| d10_pooled | 38.5% | percent of women with a teen birth | predicted share of women with a teen birth, population-weighted mean over tracts (2010 tract population as the weight), county-bootstrap interval; tenths hold near-equal numbers of tracts | all women raised by parents at the 25th percentile of the national income distribution, in the most deprived tenth of tracts by ADI rank | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 | 95% CI 37.8% to 39.3% | 7,226 tracts | 2010 census tract population | Opportunity Atlas |   |
| d1_pooled | 12.9% | percent of women with a teen birth | predicted share of women with a teen birth, population-weighted mean over tracts (2010 tract population as the weight), county-bootstrap interval; tenths hold near-equal numbers of tracts | all women raised by parents at the 25th percentile of the national income distribution, in the least deprived tenth of tracts by ADI rank | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 | 95% CI 11.9% to 14.0% | 7,103 tracts | 2010 census tract population | Opportunity Atlas |   |
| floor_county | 20 | tracts | minimum number of tracts with both estimates for a county to enter the county version | counties in the county version of the comparison | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 |   |   | none (inclusion rule) | Opportunity Atlas |   |
| floor_max | 50 | tracts | largest minimum number of tracts tried | counties in the county version of the comparison | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 |   |   | none (inclusion rule) | Opportunity Atlas |   |
| floor_pos | 10 | tracts | smallest minimum number of tracts tried at which the gap is positive in every county | counties in the county version of the comparison | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 |   |   | none (inclusion rule) | Opportunity Atlas |   |
| gap_bw | 17.6 | percentage points of women with a teen birth, Black women minus white | mean over tracts of the gap in predicted teen birth shares, population-weighted mean over tracts (2010 tract population as the weight), county-bootstrap interval | non-Hispanic Black women and non-Hispanic white women raised by parents at the 25th percentile of the national income distribution, in tracts where the Atlas estimates both, with an ADI rank and a recorded 2010 population | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 | 95% CI 17.0 to 18.2 | 20,884 tracts | 2010 census tract population | Opportunity Atlas |   |
| gap_bw_change | −0.7 | percentage points of women with a teen birth, gap in the most deprived tenth minus gap in the least deprived tenth | Black-white gap in the most deprived tenth of tracts minus the gap in the least deprived tenth, county-bootstrap interval from the same draws | non-Hispanic Black women and non-Hispanic white women raised by parents at the 25th percentile of the national income distribution, in tracts where the Atlas estimates both, with an ADI rank and a recorded 2010 population | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 | 95% CI −2.4 to 1.2 | 20,884 tracts | 2010 census tract population | Opportunity Atlas |   |
| gap_bw_d1 | 18.1 | percentage points of women with a teen birth, Black minus white | mean over tracts of the gap in predicted teen birth shares within a tenth of tracts by ADI rank, population-weighted mean over tracts (2010 tract population as the weight), county-bootstrap interval | non-Hispanic Black and white women raised by parents at the 25th percentile of the national income distribution, in the least deprived tenth of the tracts where the Atlas estimates both | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 | 95% CI 16.2 to 19.6 | 696 tracts | 2010 census tract population | Opportunity Atlas |   |
| gap_bw_d10 | 17.4 | percentage points of women with a teen birth, Black minus white | mean over tracts of the gap in predicted teen birth shares within a tenth of tracts by ADI rank, population-weighted mean over tracts (2010 tract population as the weight), county-bootstrap interval | non-Hispanic Black and white women raised by parents at the 25th percentile of the national income distribution, in the most deprived tenth of the tracts where the Atlas estimates both | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 | 95% CI 16.6 to 18.1 | 3,152 tracts | 2010 census tract population | Opportunity Atlas |   |
| gap_bw_slope10 | −0.1 | percentage points of women with a teen birth per ten points of ADI national rank, Black minus white | population-weighted slope of the tract's Black-white gap on the ADI national rank, times ten, county-bootstrap interval | non-Hispanic Black women and non-Hispanic white women raised by parents at the 25th percentile of the national income distribution, in tracts where the Atlas estimates both, with an ADI rank and a recorded 2010 population | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 | 95% CI −0.3 to 0.1 | 20,884 tracts | 2010 census tract population | Opportunity Atlas |   |
| gap_bw_unw | 17.9 | percentage points of women with a teen birth, group minus white | mean over tracts of the gap in predicted teen birth shares, each tract counted once, county-bootstrap interval | non-Hispanic Black women and non-Hispanic white women raised by parents at the 25th percentile of the national income distribution, in tracts where the Atlas estimates both, with an ADI rank and a recorded 2010 population | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 | 95% CI 17.4 to 18.5 | 20,884 tracts | none (each tract counts once) | Opportunity Atlas |   |
| gap_cty_f1 | 17.6 | percentage points of women with a teen birth, Black minus white | population-weighted mean over counties of the gap in county mean predicted shares (county mean = tract mean weighted by 2010 population), bootstrap interval over states | non-Hispanic Black and white women raised by parents at the 25th percentile of the national income distribution, in counties with at least one tract where the Atlas estimates both | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 | 95% CI 16.5 to 19.0 | 1,318 counties | 2010 census population summed to the county | Opportunity Atlas |   |
| gap_cty_f20 | 18.1 | percentage points of women with a teen birth, Black minus white | population-weighted mean over counties of the gap in county mean predicted shares (county mean = tract mean weighted by 2010 population), bootstrap interval over states | non-Hispanic Black and white women raised by parents at the 25th percentile of the national income distribution, in counties with at least 20 tracts where the Atlas estimates both | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 | 95% CI 16.8 to 19.6 | 261 counties | 2010 census population summed to the county | Opportunity Atlas |   |
| gap_cty_f50 | 18.1 | percentage points of women with a teen birth, Black minus white | population-weighted mean over counties of the gap in county mean predicted shares (county mean = tract mean weighted by 2010 population), bootstrap interval over states | non-Hispanic Black and white women raised by parents at the 25th percentile of the national income distribution, in counties with at least 50 tracts where the Atlas estimates both | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 | 95% CI 17.0 to 19.7 | 102 counties | 2010 census population summed to the county | Opportunity Atlas |   |
| gap_hw | 10.9 | percentage points of women with a teen birth, Hispanic women minus white | mean over tracts of the gap in predicted teen birth shares, population-weighted mean over tracts (2010 tract population as the weight), county-bootstrap interval | Hispanic women of any race and non-Hispanic white women raised by parents at the 25th percentile of the national income distribution, in tracts where the Atlas estimates both, with an ADI rank and a recorded 2010 population | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 | 95% CI 10.3 to 11.5 | 22,695 tracts | 2010 census tract population | Opportunity Atlas |   |
| gap_hw_change | −0.8 | percentage points of women with a teen birth, gap in the most deprived tenth minus gap in the least deprived tenth | Hispanic-white gap in the most deprived tenth of tracts minus the gap in the least deprived tenth, county-bootstrap interval from the same draws | Hispanic women of any race and non-Hispanic white women raised by parents at the 25th percentile of the national income distribution, in tracts where the Atlas estimates both, with an ADI rank and a recorded 2010 population | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 | 95% CI −2.4 to 0.7 | 22,695 tracts | 2010 census tract population | Opportunity Atlas |   |
| gap_hw_d1 | 11.5 | percentage points of women with a teen birth, Hispanic minus white | mean over tracts of the gap in predicted teen birth shares within a tenth of tracts by ADI rank, population-weighted mean over tracts (2010 tract population as the weight), county-bootstrap interval | Hispanic and non-Hispanic white women raised by parents at the 25th percentile of the national income distribution, in the least deprived tenth of the tracts where the Atlas estimates both | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 | 95% CI 10.4 to 12.7 | 3,010 tracts | 2010 census tract population | Opportunity Atlas |   |
| gap_hw_d10 | 10.7 | percentage points of women with a teen birth, Hispanic minus white | mean over tracts of the gap in predicted teen birth shares within a tenth of tracts by ADI rank, population-weighted mean over tracts (2010 tract population as the weight), county-bootstrap interval | Hispanic and non-Hispanic white women raised by parents at the 25th percentile of the national income distribution, in the most deprived tenth of the tracts where the Atlas estimates both | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 | 95% CI 9.7 to 11.7 | 1,253 tracts | 2010 census tract population | Opportunity Atlas |   |
| gap_hw_unw | 11.2 | percentage points of women with a teen birth, group minus white | mean over tracts of the gap in predicted teen birth shares, each tract counted once, county-bootstrap interval | Hispanic women of any race and non-Hispanic white women raised by parents at the 25th percentile of the national income distribution, in tracts where the Atlas estimates both, with an ADI rank and a recorded 2010 population | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 | 95% CI 10.6 to 11.8 | 22,695 tracts | none (each tract counts once) | Opportunity Atlas |   |
| gap_pf_2000 | 17.6 | percentage points of women with a teen birth, Black minus white | mean over tracts of the gap, population-weighted mean over tracts (2010 tract population as the weight), county-bootstrap interval | non-Hispanic Black and white women raised by parents at the 25th percentile of the national income distribution, in tracts with at least 2000 residents where the Atlas estimates both | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 | 95% CI 17.0 to 18.2 | 19,836 tracts | 2010 census tract population | Opportunity Atlas |   |
| gap_pf_4000 | 17.3 | percentage points of women with a teen birth, Black minus white | mean over tracts of the gap, population-weighted mean over tracts (2010 tract population as the weight), county-bootstrap interval | non-Hispanic Black and white women raised by parents at the 25th percentile of the national income distribution, in tracts with at least 4000 residents where the Atlas estimates both | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 | 95% CI 16.7 to 17.9 | 12,788 tracts | 2010 census tract population | Opportunity Atlas |   |
| gap_pf_6000 | 16.5 | percentage points of women with a teen birth, Black minus white | mean over tracts of the gap, population-weighted mean over tracts (2010 tract population as the weight), county-bootstrap interval | non-Hispanic Black and white women raised by parents at the 25th percentile of the national income distribution, in tracts with at least 6000 residents where the Atlas estimates both | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 | 95% CI 15.8 to 17.2 | 4,790 tracts | 2010 census tract population | Opportunity Atlas |   |
| gap_wa | 7.1 | percentage points of women with a teen birth, white minus Asian | mean over tracts of the gap in predicted teen birth shares, population-weighted mean over tracts (2010 tract population as the weight), county-bootstrap interval | non-Hispanic Asian women and non-Hispanic white women raised by parents at the 25th percentile of the national income distribution, in tracts where the Atlas estimates both, with an ADI rank and a recorded 2010 population | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 | 95% CI 6.3 to 8.0 | 7,064 tracts | 2010 census tract population | Opportunity Atlas |   |
| gap_wa_change | −7.0 | percentage points of women with a teen birth, Asian-white difference in the most deprived tenth minus the difference in the least deprived tenth | Asian minus white difference in the most deprived tenth of tracts minus the difference in the least deprived tenth, county-bootstrap interval from the same draws | non-Hispanic Asian women and non-Hispanic white women raised by parents at the 25th percentile of the national income distribution, in tracts where the Atlas estimates both, with an ADI rank and a recorded 2010 population | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 | 95% CI −12.0 to −2.7 | 7,064 tracts | 2010 census tract population | Opportunity Atlas |   |
| level_diff | 3.6 | percentage points of women with a teen birth, Atlas county estimates minus tract estimates | population-weighted mean of the Atlas's county estimates minus the population-weighted mean of the tract estimates in the same counties, both weighted by 2010 population, bootstrap interval over states | women raised by parents at the 25th percentile of the national income distribution, in the counties that have both an Atlas county estimate and tracts in the gradient | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 | 95% CI 3.2 to 3.9 | 3,113 counties | 2010 census population | Opportunity Atlas |   |
| lp_effect | 6.4% | percent reduction in the county teen birth rate | difference-in-differences estimate of the reduction in the teen birth rate of 15- to 19-year-old women in counties with funded clinics, relative to comparison counties, reported in the abstract of Lindo and Packham (2017); copied from the paper and not computed in this build | Colorado counties with a Title X family-planning clinic whose clinics received program funding, against counties with a Title X clinic in other states | the five years after the program began in 2009 |   |   | none (reported estimate) | Published estimates from cited studies | A published estimate, not a result of this build. |
| lp_years | 5 | years after the program began | length of the post-program window over which the abstract of Lindo and Packham (2017) reports the reduction in the teen birth rate; copied from the paper and not computed in this build | Colorado counties with a Title X family-planning clinic whose clinics received program funding | the five years after the program began in 2009 |   |   | none (reported estimate) | Published estimates from cited studies | A published figure, not a result of this build. |
| mean_asian | 7.4% | percent of women with a teen birth | predicted share of women with a teen birth, population-weighted mean over tracts (2010 tract population as the weight), county-bootstrap interval | non-Hispanic Asian women raised by parents at the 25th percentile of the national income distribution, in tracts with an Atlas estimate for them, an ADI rank and a recorded 2010 population | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 | 95% CI 6.5% to 8.7% | 7,718 tracts | 2010 census tract population | Opportunity Atlas |   |
| mean_black | 41.3% | percent of women with a teen birth | predicted share of women with a teen birth, population-weighted mean over tracts (2010 tract population as the weight), county-bootstrap interval | non-Hispanic Black women raised by parents at the 25th percentile of the national income distribution, in tracts with an Atlas estimate for them, an ADI rank and a recorded 2010 population | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 | 95% CI 40.4% to 42.1% | 25,216 tracts | 2010 census tract population | Opportunity Atlas |   |
| mean_hisp | 31.0% | percent of women with a teen birth | predicted share of women with a teen birth, population-weighted mean over tracts (2010 tract population as the weight), county-bootstrap interval | Hispanic women of any race raised by parents at the 25th percentile of the national income distribution, in tracts with an Atlas estimate for them, an ADI rank and a recorded 2010 population | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 | 95% CI 29.8% to 32.3% | 25,785 tracts | 2010 census tract population | Opportunity Atlas |   |
| mean_pooled | 25.0% | percent of women with a teen birth | predicted share of women with a teen birth, population-weighted mean over tracts (2010 tract population as the weight), county-bootstrap interval | all women raised by parents at the 25th percentile of the national income distribution, in tracts with an Atlas estimate for them, an ADI rank and a recorded 2010 population | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 | 95% CI 24.3% to 25.6% | 71,303 tracts | 2010 census tract population | Opportunity Atlas |   |
| mean_pooled_unw | 25.5% | percent of women with a teen birth | predicted share of women with a teen birth, mean over tracts with each tract counted once, county-bootstrap interval | all women raised by parents at the 25th percentile of the national income distribution, in tracts with an Atlas estimate for them, an ADI rank and a recorded 2010 population | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 | 95% CI 24.7% to 26.2% | 71,303 tracts | none (each tract counts once) | Opportunity Atlas |   |
| mean_white | 20.0% | percent of women with a teen birth | predicted share of women with a teen birth, population-weighted mean over tracts (2010 tract population as the weight), county-bootstrap interval | non-Hispanic white women raised by parents at the 25th percentile of the national income distribution, in tracts with an Atlas estimate for them, an ADI rank and a recorded 2010 population | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 | 95% CI 19.3% to 20.7% | 64,974 tracts | 2010 census tract population | Opportunity Atlas |   |
| n_below_zero | 451 | tracts | count of tracts with a negative published value (privacy noise, never clipped) | tracts whose estimate for all women is below zero | women born 1978 to 1983, who had teen births in the 1990s and early 2000s |   | 71,646 tracts | none (full count) | Opportunity Atlas |   |
| n_black_d1 | 1,110 | tracts | count of tracts | tracts in the least deprived tenth with an Atlas estimate for Black women | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 |   |   | none (full count) | Opportunity Atlas |   |
| n_black_d10 | 4,292 | tracts | count of tracts | tracts in the most deprived tenth with an Atlas estimate for Black women | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 |   |   | none (full count) | Opportunity Atlas |   |
| n_cloud | 5,000 | tracts | size of a random sample drawn with a fixed seed | tracts drawn at random for the scatter plot | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 |   | 71,303 tracts | none (sample size) | Opportunity Atlas |   |
| n_cloud_outside | 16 | tracts | count of tracts outside the plotted range of the scatter plot | all women raised by parents at the 25th percentile of the national income distribution, in tracts with an Atlas estimate for them, an ADI rank and a recorded 2010 population | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 |   |   | none (full count) | Opportunity Atlas |   |
| n_cty | 3,125 | counties | count of counties | counties in the states and the District of Columbia with an Atlas estimate for all women | women born 1978 to 1983, who had teen births in the 1990s and early 2000s |   |   | none (full count) | Opportunity Atlas |   |
| n_cty_drawn | 3,126 | counties | count of drawn counties with an Atlas estimate | counties and county equivalents drawn with a value on the map | women born 1978 to 1983, who had teen births in the 1990s and early 2000s |   |   | none (full count) | Opportunity Atlas |   |
| n_cty_f10 | 453 | counties | count of counties | counties with at least 10 tracts where the Atlas estimates both Black and white women | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 |   |   | none (full count) | Opportunity Atlas |   |
| n_cty_f20 | 261 | counties | count of counties | counties with at least 20 tracts where the Atlas estimates both Black and white women | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 |   |   | none (full count) | Opportunity Atlas |   |
| n_cty_nodata | 17 | counties | count of drawn counties with no Opportunity Atlas estimate | counties and county equivalents on the map (Puerto Rico not drawn) | women born 1978 to 1983, who had teen births in the 1990s and early 2000s |   |   | none (full count) | Opportunity Atlas |   |
| n_key_saturated | 523 | counties | count of those counties whose value lies outside the range of the key (below its low end or above its high end) and takes the end color | counties drawn with a value on the map | women born 1978 to 1983, who had teen births in the 1990s and early 2000s |   |   | none (full count) | Opportunity Atlas |   |
| n_no_adi | 343 | tracts | count of tracts left out of the gradient | tracts with an estimate for all women but no ADI rank or no recorded 2010 population | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 |   |   | none (full count) | Opportunity Atlas |   |
| n_oa_tracts | 73,278 | tracts | count of tracts in the file | 2010 census tracts in the staged Opportunity Atlas file | women born 1978 to 1983, who had teen births in the 1990s and early 2000s |   |   | none (full count) | Opportunity Atlas |   |
| n_pairs_asian | 7,064 | tracts | count of tracts | non-Hispanic Asian women and non-Hispanic white women raised by parents at the 25th percentile of the national income distribution, in tracts where the Atlas estimates both, with an ADI rank and a recorded 2010 population | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 |   |   | none (full count) | Opportunity Atlas |   |
| n_pairs_black | 20,884 | tracts | count of tracts | non-Hispanic Black women and non-Hispanic white women raised by parents at the 25th percentile of the national income distribution, in tracts where the Atlas estimates both, with an ADI rank and a recorded 2010 population | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 |   |   | none (full count) | Opportunity Atlas |   |
| n_pairs_hisp | 22,695 | tracts | count of tracts | Hispanic women of any race and non-Hispanic white women raised by parents at the 25th percentile of the national income distribution, in tracts where the Atlas estimates both, with an ADI rank and a recorded 2010 population | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 |   |   | none (full count) | Opportunity Atlas |   |
| n_pooled_est | 71,646 | tracts | count of tracts with a published value | 2010 census tracts in the staged Opportunity Atlas file with an estimate for all women | women born 1978 to 1983, who had teen births in the 1990s and early 2000s |   |   | none (full count) | Opportunity Atlas |   |
| n_pov | 71,295 | tracts | count of tracts in the sensitivity check | 2010 census tracts with an Atlas estimate for all women, an ADI rank, a recorded 2010 population and a 1990 poverty rate | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 |   |   | none (full count) | Opportunity Atlas |   |
| n_tracts | 71,303 | tracts | count of tracts | 2010 census tracts with an Atlas estimate for all women, an ADI rank and a recorded 2010 population | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 |   |   | none (full count) | Opportunity Atlas |   |
| noise_share | 10% | percent of an estimate's standard error | size of the privacy noise added to each estimate, as the Atlas codebook states it (typically less than this share of the standard error) | Opportunity Atlas tract estimates | women born 1978 to 1983, who had teen births in the 1990s and early 2000s |   |   | none (documentation) | Opportunity Atlas | Stated in the Atlas documentation, not computed in this build. |
| pct_bw | 89.6% | percent of tracts in which the group's predicted share is higher than white women's | share of tracts (each tract counted once) where the group's predicted share exceeds white women's, county-bootstrap interval | non-Hispanic Black women and non-Hispanic white women raised by parents at the 25th percentile of the national income distribution, in tracts where the Atlas estimates both, with an ADI rank and a recorded 2010 population | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 | 95% CI 88.9% to 90.2% | 20,884 tracts | none (each tract counts once) | Opportunity Atlas | Noise in the tract estimates pulls this share toward an even split. |
| pct_hw | 79.1% | percent of tracts in which the group's predicted share is higher than white women's | share of tracts (each tract counted once) where the group's predicted share exceeds white women's, county-bootstrap interval | Hispanic women of any race and non-Hispanic white women raised by parents at the 25th percentile of the national income distribution, in tracts where the Atlas estimates both, with an ADI rank and a recorded 2010 population | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 | 95% CI 77.9% to 80.2% | 22,695 tracts | none (each tract counts once) | Opportunity Atlas | Noise in the tract estimates pulls this share toward an even split. |
| pct_pos_f1 | 94.6% | percent of counties | share of counties in which the county mean predicted share is higher for Black than for white women | counties with at least one tract where the Atlas estimates both Black and white women | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 |   | 1,318 counties | none (each county counts once) | Opportunity Atlas |   |
| pct_wa | 23.9% | percent of tracts in which the group's predicted share is higher than white women's | share of tracts (each tract counted once) where the group's predicted share exceeds white women's, county-bootstrap interval | non-Hispanic Asian women and non-Hispanic white women raised by parents at the 25th percentile of the national income distribution, in tracts where the Atlas estimates both, with an ADI rank and a recorded 2010 population | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 | 95% CI 21.6% to 26.0% | 7,064 tracts | none (each tract counts once) | Opportunity Atlas | Noise in the tract estimates pulls this share toward an even split. |
| pf_2000 | 2,000 | residents | minimum 2010 population of a tract (a rule) | 2010 census tracts in the comparison of Black and white women | 2010 census |   |   | none (inclusion rule) | USDA Food Access Research Atlas |   |
| pf_4000 | 4,000 | residents | minimum 2010 population of a tract (a rule) | 2010 census tracts in the comparison of Black and white women | 2010 census |   |   | none (inclusion rule) | USDA Food Access Research Atlas |   |
| pf_6000 | 6,000 | residents | minimum 2010 population of a tract (a rule) | 2010 census tracts in the comparison of Black and white women | 2010 census |   |   | none (inclusion rule) | USDA Food Access Research Atlas |   |
| pop_pairs_asian | 37.9 million | people | sum of 2010 census tract population | residents of the tracts in the comparison with non-Hispanic Asian women | 2010 census |   |   | none (full count) | USDA Food Access Research Atlas |   |
| pop_pairs_black | 99.6 million | people | sum of 2010 census tract population | residents of the tracts in the comparison with non-Hispanic Black women | 2010 census |   |   | none (full count) | USDA Food Access Research Atlas |   |
| pop_pairs_hisp | 114.3 million | people | sum of 2010 census tract population | residents of the tracts in the comparison with Hispanic women of any race | 2010 census |   |   | none (full count) | USDA Food Access Research Atlas |   |
| pop_tracts_share | 99.0% | percent of the 2010 U.S. population | 2010 population of the tracts with an estimate for all women and an ADI rank, as a share of the U.S. population | residents of the United States | 2010 census |   |   | none (full count) | USDA Food Access Research Atlas |   |
| ratio_bw_d1 | 2.5 | ratio of the group's predicted share to white women's | ratio of population-weighted mean predicted shares (group over white women), county-bootstrap interval | non-Hispanic Black women and non-Hispanic white women raised by parents at the 25th percentile of the national income distribution, in the least deprived tenth of the tracts where the Atlas estimates both | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 | 95% CI 2.3 to 2.9 | 696 tracts | 2010 census tract population | Opportunity Atlas |   |
| ratio_bw_d10 | 1.5 | ratio of the group's predicted share to white women's | ratio of population-weighted mean predicted shares (group over white women), county-bootstrap interval | non-Hispanic Black women and non-Hispanic white women raised by parents at the 25th percentile of the national income distribution, in the most deprived tenth of the tracts where the Atlas estimates both | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 | 95% CI 1.5 to 1.6 | 3,152 tracts | 2010 census tract population | Opportunity Atlas |   |
| ratio_hw_d1 | 2.1 | ratio of the group's predicted share to white women's | ratio of population-weighted mean predicted shares (group over white women), county-bootstrap interval | Hispanic women of any race and non-Hispanic white women raised by parents at the 25th percentile of the national income distribution, in the least deprived tenth of the tracts where the Atlas estimates both | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 | 95% CI 2.0 to 2.3 | 3,010 tracts | 2010 census tract population | Opportunity Atlas |   |
| ratio_hw_d10 | 1.3 | ratio of the group's predicted share to white women's | ratio of population-weighted mean predicted shares (group over white women), county-bootstrap interval | Hispanic women of any race and non-Hispanic white women raised by parents at the 25th percentile of the national income distribution, in the most deprived tenth of the tracts where the Atlas estimates both | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 | 95% CI 1.3 to 1.4 | 1,253 tracts | 2010 census tract population | Opportunity Atlas |   |
| region_midwest | 29.5% | percent of women with a teen birth | predicted share of women with a teen birth, population-weighted mean over counties (2010 census population as the weight), bootstrap interval over states | women raised by parents at the 25th percentile of the national income distribution, by county of childhood, in counties of the Census Bureau's Midwest region | women born 1978 to 1983, who had teen births in the 1990s and early 2000s | 95% CI 26.3% to 31.5% | 1,052 counties | 2010 census county population | Opportunity Atlas |   |
| region_northeast | 21.7% | percent of women with a teen birth | predicted share of women with a teen birth, population-weighted mean over counties (2010 census population as the weight), bootstrap interval over states | women raised by parents at the 25th percentile of the national income distribution, by county of childhood, in counties of the Census Bureau's Northeast region | women born 1978 to 1983, who had teen births in the 1990s and early 2000s | 95% CI 19.7% to 23.0% | 217 counties | 2010 census county population | Opportunity Atlas |   |
| region_south | 33.5% | percent of women with a teen birth | predicted share of women with a teen birth, population-weighted mean over counties (2010 census population as the weight), bootstrap interval over states | women raised by parents at the 25th percentile of the national income distribution, by county of childhood, in counties of the Census Bureau's South region | women born 1978 to 1983, who had teen births in the 1990s and early 2000s | 95% CI 31.4% to 35.6% | 1,419 counties | 2010 census county population | Opportunity Atlas |   |
| region_west | 25.0% | percent of women with a teen birth | predicted share of women with a teen birth, population-weighted mean over counties (2010 census population as the weight), bootstrap interval over states | women raised by parents at the 25th percentile of the national income distribution, by county of childhood, in counties of the Census Bureau's West region | women born 1978 to 1983, who had teen births in the 1990s and early 2000s | 95% CI 23.5% to 29.2% | 437 counties | 2010 census county population | Opportunity Atlas |   |
| rise_adi_common | 25.6 | percentage points of women with a teen birth, most minus least deprived tenth | mean in the most deprived tenth of tracts minus mean in the least deprived tenth, tenths of the ADI national rank formed among these tracts, population-weighted mean over tracts (2010 tract population as the weight), county-bootstrap interval | all women raised by parents at the 25th percentile of the national income distribution, in tracts with a 1990 poverty rate | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 | 95% CI 24.3 to 26.9 | 71,295 tracts | 2010 census tract population | Opportunity Atlas |   |
| rise_asian | 15.8 | percentage points of women with a teen birth, most minus least deprived tenth | mean in the most deprived tenth of tracts minus mean in the least deprived tenth, population-weighted mean over tracts (2010 tract population as the weight), county-bootstrap interval | non-Hispanic Asian women raised by parents at the 25th percentile of the national income distribution, in tracts with an Atlas estimate for them, an ADI rank and a recorded 2010 population | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 | 95% CI 11.3 to 20.1 | 7,718 tracts | 2010 census tract population | Opportunity Atlas |   |
| rise_black | 18.1 | percentage points of women with a teen birth, most minus least deprived tenth | mean in the most deprived tenth of tracts minus mean in the least deprived tenth, population-weighted mean over tracts (2010 tract population as the weight), county-bootstrap interval | non-Hispanic Black women raised by parents at the 25th percentile of the national income distribution, in tracts with an Atlas estimate for them, an ADI rank and a recorded 2010 population | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 | 95% CI 16.9 to 19.9 | 25,216 tracts | 2010 census tract population | Opportunity Atlas |   |
| rise_diff_asian | 9.8 | percentage points of women with a teen birth, rise for all women minus rise for non-Hispanic Asian women | rise from the least to the most deprived tenth of tracts for all women, minus the same rise for the group (each in its own tracts), same county draws | tracts with an Atlas estimate for all women and, separately, for non-Hispanic Asian women, each with an ADI rank and a recorded 2010 population | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 | 95% CI 5.4 to 14.4 | 71,303 tracts | 2010 census tract population | Opportunity Atlas |   |
| rise_diff_black | 7.5 | percentage points of women with a teen birth, rise for all women minus rise for non-Hispanic Black women | rise from the least to the most deprived tenth of tracts for all women, minus the same rise for the group (each in its own tracts), same county draws | tracts with an Atlas estimate for all women and, separately, for non-Hispanic Black women, each with an ADI rank and a recorded 2010 population | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 | 95% CI 6.0 to 8.9 | 71,303 tracts | 2010 census tract population | Opportunity Atlas |   |
| rise_diff_hisp | 5.3 | percentage points of women with a teen birth, rise for all women minus rise for Hispanic women of any race | rise from the least to the most deprived tenth of tracts for all women, minus the same rise for the group (each in its own tracts), same county draws | tracts with an Atlas estimate for all women and, separately, for Hispanic women of any race, each with an ADI rank and a recorded 2010 population | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 | 95% CI 3.7 to 7.1 | 71,303 tracts | 2010 census tract population | Opportunity Atlas |   |
| rise_diff_white | 3.9 | percentage points of women with a teen birth, rise for all women minus rise for non-Hispanic white women | rise from the least to the most deprived tenth of tracts for all women, minus the same rise for the group (each in its own tracts), same county draws | tracts with an Atlas estimate for all women and, separately, for non-Hispanic white women, each with an ADI rank and a recorded 2010 population | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 | 95% CI 2.3 to 5.2 | 71,303 tracts | 2010 census tract population | Opportunity Atlas |   |
| rise_hisp | 20.3 | percentage points of women with a teen birth, most minus least deprived tenth | mean in the most deprived tenth of tracts minus mean in the least deprived tenth, population-weighted mean over tracts (2010 tract population as the weight), county-bootstrap interval | Hispanic women of any race raised by parents at the 25th percentile of the national income distribution, in tracts with an Atlas estimate for them, an ADI rank and a recorded 2010 population | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 | 95% CI 18.1 to 22.2 | 25,785 tracts | 2010 census tract population | Opportunity Atlas |   |
| rise_pooled | 25.6 | percentage points of women with a teen birth, most minus least deprived tenth | mean in the most deprived tenth of tracts minus mean in the least deprived tenth, population-weighted mean over tracts (2010 tract population as the weight), county-bootstrap interval | all women raised by parents at the 25th percentile of the national income distribution, in tracts with an Atlas estimate for them, an ADI rank and a recorded 2010 population | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 | 95% CI 24.3 to 26.9 | 71,303 tracts | 2010 census tract population | Opportunity Atlas |   |
| rise_pov1990 | 24.1 | percentage points of women with a teen birth, most minus least deprived tenth | mean in the most deprived tenth of tracts minus mean in the least deprived tenth, tenths of the 1990 tract poverty rate formed among these tracts, population-weighted mean over tracts (2010 tract population as the weight), county-bootstrap interval | all women raised by parents at the 25th percentile of the national income distribution, in tracts with a 1990 poverty rate | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 | 95% CI 22.7 to 25.5 | 71,295 tracts | 2010 census tract population | Opportunity Atlas |   |
| rise_pov_diff | −1.5 | percentage points of women with a teen birth, rise by tenths of 1990 poverty minus rise by tenths of the ADI | rise across tenths of the 1990 tract poverty rate minus rise across tenths of the ADI rank, on the same tracts, county-bootstrap interval from the same draws | all women raised by parents at the 25th percentile of the national income distribution, in tracts with a 1990 poverty rate | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 | 95% CI −2.8 to −0.2 | 71,295 tracts | 2010 census tract population | Opportunity Atlas |   |
| rise_white | 21.7 | percentage points of women with a teen birth, most minus least deprived tenth | mean in the most deprived tenth of tracts minus mean in the least deprived tenth, population-weighted mean over tracts (2010 tract population as the weight), county-bootstrap interval | non-Hispanic white women raised by parents at the 25th percentile of the national income distribution, in tracts with an Atlas estimate for them, an ADI rank and a recorded 2010 population | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 | 95% CI 20.7 to 22.8 | 64,974 tracts | 2010 census tract population | Opportunity Atlas |   |
| share_single | 56.4% | percent of tracts | share of 2010 tracts that overlap exactly one 2020 tract, where the crosswalk gives the ADI rank without averaging | all women raised by parents at the 25th percentile of the national income distribution, in tracts with an Atlas estimate for them, an ADI rank and a recorded 2010 population | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 |   | 71,303 tracts | none (each tract counts once) | Opportunity Atlas |   |
| slope_adi10 | 2.5 | percentage points of women with a teen birth per ten points of ADI national rank | population-weighted slope of the predicted teen birth share on the ADI national rank, times ten, county-bootstrap interval | all women raised by parents at the 25th percentile of the national income distribution, in tracts with an Atlas estimate for them, an ADI rank and a recorded 2010 population | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 | 95% CI 2.4 to 2.7 | 71,303 tracts | 2010 census tract population | Opportunity Atlas |   |
| slope_cty | 0.85 | percentage points of Black women's share per point of white women's share | slope of the county mean Black share on the county mean white share, weighted by 2010 population summed to the county, bootstrap interval over states | non-Hispanic Black and white women raised by parents at the 25th percentile of the national income distribution, in counties with at least 20 tracts where the Atlas estimates both | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 | 95% CI 0.70 to 1.00 | 261 counties | 2010 census population summed to the county | Opportunity Atlas |   |
| slope_tract | 0.41 | percentage points of Black women's share per point of white women's share | slope of the tract's Black share on its white share over these tracts, population-weighted, county-bootstrap interval | non-Hispanic Black and white women raised by parents at the 25th percentile of the national income distribution, in the tracts of the counties with at least 20 tracts where the Atlas estimates both | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 | 95% CI 0.35 to 0.46 | 15,239 tracts | 2010 census tract population | Opportunity Atlas |   |
| slope_within | 0.26 | percentage points of Black women's share per point of white women's share | slope of the tract's Black share on its white share after subtracting each county's population-weighted mean from both (differences between counties removed), population-weighted, county-bootstrap interval | non-Hispanic Black and white women raised by parents at the 25th percentile of the national income distribution, in the tracts of the counties with at least 20 tracts where the Atlas estimates both | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 | 95% CI 0.23 to 0.29 | 15,239 tracts | 2010 census tract population | Opportunity Atlas |   |
| south_all_share | 45% | percent of counties | share of those counties that are in the South region | counties in the states and DC with an estimate | women born 1978 to 1983, who had teen births in the 1990s and early 2000s |   | 3,125 counties | none (each county counts once) | Opportunity Atlas |   |
| south_minus_mw | 4.0 | percentage points of women with a teen birth, South minus Midwest | South region mean minus Midwest region mean of the predicted share, bootstrap interval over states from the same draws | women raised by parents at the 25th percentile of the national income distribution, by county of childhood, in the states and the District of Columbia | women born 1978 to 1983, who had teen births in the 1990s and early 2000s | 95% CI 1.2 to 7.6 | 3,125 counties | 2010 census county population | Opportunity Atlas |   |
| south_minus_ne | 11.8 | percentage points of women with a teen birth, South minus Northeast | South region mean minus Northeast region mean of the predicted share, bootstrap interval over states from the same draws | women raised by parents at the 25th percentile of the national income distribution, by county of childhood, in the states and the District of Columbia | women born 1978 to 1983, who had teen births in the 1990s and early 2000s | 95% CI 9.3 to 14.6 | 3,125 counties | 2010 census county population | Opportunity Atlas |   |
| south_minus_w | 8.5 | percentage points of women with a teen birth, South minus West | South region mean minus West region mean of the predicted share, bootstrap interval over states from the same draws | women raised by parents at the 25th percentile of the national income distribution, by county of childhood, in the states and the District of Columbia | women born 1978 to 1983, who had teen births in the 1990s and early 2000s | 95% CI 4.1 to 11.1 | 3,125 counties | 2010 census county population | Opportunity Atlas |   |
| south_top_share | 93% | percent of counties | share of those counties that are in the South region | counties in the highest tenth of county values (states and DC) | women born 1978 to 1983, who had teen births in the 1990s and early 2000s |   | 313 counties | none (each county counts once) | Opportunity Atlas |   |
| white_bw | 23.3% | percent of women with a teen birth | predicted share, population-weighted mean over tracts (2010 tract population as the weight), county-bootstrap interval | non-Hispanic white women raised by parents at the 25th percentile of the national income distribution, in tracts where the Atlas also estimates non-Hispanic Black women | women born 1978 to 1983, who had teen births in the 1990s and early 2000s; deprivation rank from the American Community Survey, 2019 to 2023 | 95% CI 22.6% to 24.0% | 20,884 tracts | 2010 census tract population | Opportunity Atlas |   |
| wonder_low | 15 | years of age | lower age limit of the teen band in the WONDER natality tables used in the chapter on recent births | births in CDC WONDER natality tables for the youngest mother-age band that most counties publish | 2018 to 2024 |   |   | none (definition) | CDC WONDER natality |   |

## Definitions

- Census tract: A small area drawn by the Census Bureau to stand in for a neighborhood. Boundaries are redrawn after each decennial census.
- Modeled estimate: A figure produced by a statistical model instead of a direct count or measurement, usually for small areas where direct data are thin, borrowing strength from surveys, neighboring areas or other data. Neighboring areas can share errors, so single extremes call for caution.
- Bootstrap interval: A range for an estimate found by recomputing it on many random resamples of the data and reading the range from the results. It shows how much the estimate would vary from sample to sample.
- Area Deprivation Index (ADI): A score that ranks neighborhoods by income, education, employment and housing conditions drawn from the American Community Survey. In this book a higher national percentile means a more deprived neighborhood.
- Ecological inference: Drawing conclusions about individuals from data about places or groups. A pattern across counties need not hold for the people inside them.

## Methods

The teen birth measure is the Opportunity Atlas's predicted share of women who ever claimed, as a dependent, a child born when they were between the ages of 13 and 19, for women born in 1978 to 1983 whose parents were at the 25th percentile of the national income distribution, so it is a model prediction that describes teen births in the 1990s and early 2000s. Neighborhoods are ranked by the Area Deprivation Index for 2019 to 2023, decades after the births. Tracts are weighted by their 2010 census population because the Atlas has no counts of women, and the Atlas adds privacy noise to every estimate.

### What the teen birth measure is and which tracts it covers

The teen birth measure is the Opportunity Atlas's predicted share of women who ever claimed, as a dependent, a child born when they were between the ages of 13 and 19, for women born in 1978 to 1983 whose parents were at the 25th percentile of the national income distribution (Atlas codebook for Table 4, variable teenbrth).

The staged tract file has one value for all women and one for each of white, Black, Hispanic and Asian women, on native 2010 tracts (73,278 tracts in the file). The Atlas adds privacy noise to every estimate and the values are never clipped. A tract enters a comparison when the estimates exist, its 2010 population is positive and it has a deprivation rank.

Explains Table 2, Table 3 and Table 4.

### How deprivation and its tenths are defined

Deprivation is the Area Deprivation Index national rank, a block-group index (ACS 2019 to 2023) whose 2020-tract value is the unweighted mean of its block groups' ranks; a 2010 tract takes the land-area-weighted mean of the values of the 2020 tracts that overlap it, when those with a rank cover at least half of its land area.

Tenths are tenths of all 71,642 tracts in the file with a rank and a positive population (near-equal numbers; tied ranks stay together), whether or not they have an Atlas estimate, so a tenth means the same part of the scale for every group. A sensitivity check ranks tracts by the 1990 census poverty rate in the Opportunity Atlas tract covariates file (a raw file read once in the build) in place of the ADI.

Explains Table 2, Table 3 and Table 5.

### How tracts are weighted and which tracts each comparison uses

Each tract is weighted by its 2010 census population, read from the USDA Food Access Research Atlas file (the Atlas has no counts of women); counting each tract once gives similar results.

Sentences about comparisons between groups use the tracts where both groups are estimated.

Explains Table 2, Table 3, Table 4 and Table 5.

### How the county comparisons, regions and map are built

In the county version of the comparison a county enters when at least 20 of its tracts have both estimates, and the floors on tracts per county (1, 5, 10, 20 and 50) and on tract population (2,000, 4,000 and 6,000 residents) are tried in the last figure.

Regions are the Census Bureau's four regions. The county map uses the Atlas's own county estimates for all women.

Explains Table 6, Table 7 and Table 1.

### How uncertainty is measured

Intervals are 95% percentile intervals from 2,000 draws of whole counties with replacement (seed 12345), computed by one routine for means, shares, slopes, correlations and differences; quantities built from county means, and the regional means and differences, resample states.

### Four rankings of the same places

The comparison follows the book's ranking switch rule, fixed on October 28, 2026 before any of these gradients was computed. The main gradient of the chapter is recomputed under four rankings of the same places, with the outcome, period, weights, bins and intervals unchanged: the published ADI; a standardized index that applies the ADI's coefficients to its measures after standardizing each one across tracts; median household income; and the share of adults 25 and older with a bachelor's degree or more. The three alternatives come from ACS tract estimates for 2019 to 2023, the period of the ADI.

All four rankings use the same places, the 66,005 tracts that have the outcome and all four rankings, so the ADI panel can differ slightly from the chapter's main figure. Tracts enter only if they have every input of the standardized index, which leaves out tracts without a median rent, home value or mortgage cost. Each 2010 tract takes the land-area-weighted mean of each ranking over the common-sample 2020 tracts that overlap it, when they cover at least half of its land area, the chapter's construction of its ADI, and tenths are tenths of the 2010 tracts with all four rankings and a positive population.

A pattern counts as changed when the worse end switches sides, when the two ends differ under one ranking and not the other, or when the rank correlation of bin order and bin mean falls on the other side of 0.9 and moves by at least 0.1.

Explains Table 8.

## What these data cannot show

The shares are model predictions for women born in 1978 to 1983 whose parents were at the 25th income percentile. They describe teen births in the 1990s and early 2000s among women from low-income families and say nothing about women from richer families or about teenagers now. A birth counts only when the mother claimed the child as a dependent. The gradient is a pattern across tracts, an ecological pattern, and it does not say that living in a deprived tract raised any woman's chance of a teen birth. The ADI describes the tracts decades after the births, and the staged files hold no deprivation measure from the childhood years.

A same-tract comparison removes differences between neighborhoods but not all differences. Within a tract, women raised at one parental income percentile can still differ in family wealth, in schools, in the block they lived on and in how institutions treated them, and the data describe the gap without explaining it. The staged file has no standard errors and no counts of women, so tracts are weighted by their total 2010 population, which may put the levels too low. The tracts with both estimates are those with enough women of both groups. Asian estimates are low and noisy, and 20% of them are below zero.

Whether a teen birth changes a woman's later income is a separate question that this chapter does not test. Kearney and Levine reviewed state and international comparisons. They argue that early births mostly mark limited economic prospects for disadvantaged girls and carry little direct economic cost, which is a reading of patterns across places.[5] Puerto Rico has Atlas estimates but is not drawn on the map. Connecticut appears under its legacy county codes in the Atlas, and it takes its ADI rank from the 2020 planning-region tracts through the crosswalk.

## References

1. Chetty, Raj; Friedman, John N.; Hendren, Nathaniel; Jones, Maggie R.; Porter, Sonya R. (2018). The Opportunity Atlas: Mapping the Childhood Roots of Social Mobility. NBER Working Paper 25147. https://doi.org/10.3386/w25147
2. Chetty, Raj; Hendren, Nathaniel; Jones, Maggie R.; Porter, Sonya R. (2020). Race and Economic Opportunity in the United States: An Intergenerational Perspective. The Quarterly Journal of Economics 135(2): 711-783. https://doi.org/10.1093/qje/qjz042
3. Kind, Amy J. H.; Buckingham, William R. (2018). Making Neighborhood-Disadvantage Metrics Accessible: The Neighborhood Atlas. New England Journal of Medicine 378(26): 2456-2458. https://doi.org/10.1056/NEJMp1802313
4. Lindo, Jason M.; Packham, Analisa (2017). How Much Can Expanding Access to Long-Acting Reversible Contraceptives Reduce Teen Birth Rates?. American Economic Journal: Economic Policy 9(3): 348-376. https://doi.org/10.1257/pol.20160039
5. Kearney, Melissa S.; Levine, Phillip B. (2012). Why Is the Teen Birth Rate in the United States So High and Why Does It Matter?. Journal of Economic Perspectives 26(2): 141-163. https://doi.org/10.1257/jep.26.2.141

## Provenance

- Pipeline version: 0.1.0
- Build script: book2/chapters/teen-births/build.py (sha256 4b90bc9f7e275ccdb11c1c2b5a1cc0008e14bac5341973cb7f15bf02d3da4d7b)
- Built on: 2026-10-09
- Site commit: c2b1ff3309cb4a10833f884da91ac0057a66ad0a

Inputs:

| File | Size (bytes) | SHA-256, first 1 MB | SHA-256, last 1 MB |
| --- | ---: | --- | --- |
| book2/chapters/teen-births/counties.py | 5,643 | a2a44c720be89f3468ab2764f7aa5d2e93c0e20f5029d463200d98560ceb5fe4 | a2a44c720be89f3468ab2764f7aa5d2e93c0e20f5029d463200d98560ceb5fe4 |
| book2/chapters/teen-births/tracts.py | 18,878 | 1b0162cdc75c20811103c3d872ca7851bd485bcc7d170e8ff3a503ab06b4ab61 | 1b0162cdc75c20811103c3d872ca7851bd485bcc7d170e8ff3a503ab06b4ab61 |
| data/analytic/tract_atlas.parquet | 39,659,925 | 29d5e4d943552d54aeb6ba92aadc939d9bd15bdea553c78357dbd516757be627 | 1e10d1ecf722340f23ee2212d4999a90250f85b65fa89b39e32db06d9472e22e |
| data/raw/acs/deprivation_components_2019_2023/tract.parquet | 13,197,793 | 5545cde2cbe96a8b56e468d65388248a241aa6bd300d3c9e325d08d3f86b11da | 5c8ec454d066988b0f1a01aaa6ed34ebfc05d162e01cf9c6bb01f9198cbaff12 |
| data/raw/opportunity_atlus/tract_covariates.csv | 24,907,347 | ed9f5cff16e2fa96cf7cf8c775094b21fa83c5e87ffca5be740889f946096fa9 | 09330f908e2e484b1b85b060727015d0a1106721479ab42dcba9a899c6d5d09f |
| data/staged/adi/adi_atlas_tract2020_2023.parquet | 798,078 | 5e409bc212ff871def780fa27b4cf8b616d9da420b2b9f6455517eddcd8c1a95 | 5e409bc212ff871def780fa27b4cf8b616d9da420b2b9f6455517eddcd8c1a95 |
| data/staged/crosswalks/tract_2010_to_2020.parquet | 2,602,636 | f7526f0d1bac31dbc540f6d733a850c9e0ee4c7622011214c04afedfbfa1e0a5 | 9e573492f9804b571b9944a53cc59fadab4051a175952b058f77cc5a644a98ff |
| data/staged/food_access/food_access_tract2010_2019.parquet | 37,967,838 | c36929f1809e9ac84ac5f1cb62140bd980c315287833c6f281839df4c1619c89 | 05a17b940a79393b0c2ce5c9e9248496c319cba6896b63af3a2df34fe04f63aa |
| data/staged/mobility/oa_module1_tract2010.parquet | 14,946,600 | 32583830e03567b41ad3c245ca41aa67fb9cb55a8a46e8d9ae4c77a6cc2fcd83 | 6d67fac56edd8f07346375fb0b97e3c82621d57457d89c7fe93ec6f1b758bfaf |
| data/staged/mobility/oa_outcomes_county.parquet | 721,343 | fd7bc55294fbd3f7a646f27083500c1f12415775db7a3607878b522d277f8140 | fd7bc55294fbd3f7a646f27083500c1f12415775db7a3607878b522d277f8140 |
| data/staged/ses/acs5_tract2020_2019_2023.parquet | 7,232,424 | 5fe6f730a35225f14d5d8abfc6b9bce58b52e545f309a7eac74f1dae74d7a31c | 7b4a2475b3852d3f91804716a04566ef8ceadacec2be9fb85f5eb642078efac6 |
| data/staged/svi/svi_tract2020_2022.parquet | 17,740,415 | cb3860d8edc2aaf1f2a02211a58f7a0b80457eb6ad9a37b5ec8abb2d5da9da10 | e371a4a696d5bed05d54ad7c5b6ca2504508805b847ec415c2322be7f66ec2b2 |

Outputs:

| File | SHA-256 |
| --- | --- |
| facts.json | 58432d8c5da5f66f3d277a165542e5ff751f6f1d34af62ba767f6d81a2d88c12 |
| figures/alt_rankings.json | 0a0cf476dc955f9ffece13e9d3fd238082a9c44fd2ec2d441ba0800c394cd651 |
| figures/teen_birth_black_white_by_deprivation.json | cc14ff3eda9ca3896dabeff6196305c7367d7cd52e95a33adf5578784a1c602e |
| figures/teen_birth_black_white_counties.json | dad8a08d472017e2bdf630d5e7e1df916ac7f00133bea348a0b3e4823af649c1 |
| figures/teen_birth_by_deprivation_tracts.json | bc9c3f68f15ce4f0a78770d0e8cf7d9f94ca648adbe42d20fe9bb7fbae6cdd78 |
| figures/teen_birth_county_map.json | e3b79939cdaa42e21e1f220cd8b9949f6b90116bf32ffa28740e0d966dae3d90 |
| figures/teen_birth_gap_floors.json | f431cb59d60f3ea65c610f5e18f5c4fc6b0fdd49ce65b78c9b236c20b12cbc69 |
| figures/teen_birth_gaps_same_tract.json | 9d50658745f83b0a647efbf230b0d6583c021beab92086152bf2b5b2acddec2b |
| figures/teen_birth_gradient_by_group.json | e3cbc028fa34fcb169d79f9824bcc81d831db883d748b648dfc29d8d6258a1c0 |
| meta.json | 1a45292e0de3e82d05872635eeb2ec211e77263df31f9f32fb0bb0b714542ce4 |
| public/data/teen-births/teen_birth_county_map.json | a50245ff82371abb91f6a498d0b4a9d1403b32bef5228ef8df574877d62e049a |
| public/data/teen-births/teen_birth_tract_cloud.json | cc639216a549ddec6809920c8490706e0505af9035c417a267ee22d861618a73 |
| refs.json | be60b432fb8b51ac3013c21518fb5e3d84fb7f21f7310e5f143e425c5808412d |

## Machine formats

- Report (HTML): https://vishalsingh.org/zipcode/teen-births/report
- Report (Markdown): https://vishalsingh.org/zipcode/teen-births/report.md
- Every estimate (JSON): https://vishalsingh.org/zipcode/report/teen-births/facts.json
- Table 1, figure "teen_birth_county_map" (JSON): https://vishalsingh.org/zipcode/report/teen-births/figures/teen_birth_county_map.json
- Table 2, figure "teen_birth_by_deprivation_tracts" (JSON): https://vishalsingh.org/zipcode/report/teen-births/figures/teen_birth_by_deprivation_tracts.json
- Table 3, figure "teen_birth_gradient_by_group" (JSON): https://vishalsingh.org/zipcode/report/teen-births/figures/teen_birth_gradient_by_group.json
- Table 4, figure "teen_birth_gaps_same_tract" (JSON): https://vishalsingh.org/zipcode/report/teen-births/figures/teen_birth_gaps_same_tract.json
- Table 5, figure "teen_birth_black_white_by_deprivation" (JSON): https://vishalsingh.org/zipcode/report/teen-births/figures/teen_birth_black_white_by_deprivation.json
- Table 6, figure "teen_birth_black_white_counties" (JSON): https://vishalsingh.org/zipcode/report/teen-births/figures/teen_birth_black_white_counties.json
- Table 7, figure "teen_birth_gap_floors" (JSON): https://vishalsingh.org/zipcode/report/teen-births/figures/teen_birth_gap_floors.json
- Table 8, figure "alt_rankings" (JSON): https://vishalsingh.org/zipcode/report/teen-births/figures/alt_rankings.json
- Payload of "teen_birth_county_map" (JSON): https://vishalsingh.org/zipcode/data/teen-births/teen_birth_county_map.json
- Payload of "teen_birth_by_deprivation_tracts" (JSON): https://vishalsingh.org/zipcode/data/teen-births/teen_birth_tract_cloud.json
- Index of every report: https://vishalsingh.org/zipcode/llms.txt; all reports in one file: https://vishalsingh.org/zipcode/llms-full.txt

---

# Race, income and mobility: report

Chapter 4.3 of ZIP Code Destiny, by Vishal Singh (NYU Stern). Generated from the chapter's data files; the reading version is at https://vishalsingh.org/zipcode/race-income-and-mobility. Table n of this report is figure n of the chapter.

## Identification

- Title: Race, income and mobility
- Label: Two ladders, one street
- Chapter: 4.3 (Part IV, Becoming an adult; An American life)
- Question: How much does a child's adult income depend on the tract where the child grew up, on parental income and on race, and what predicts upward mobility across counties?
- Dek: White children from low-income families born in 1978 to 1983 reached a higher adult income rank than Black children raised in the same tract in 92.5% of the 30,997 tracts where both are estimated.
- Grain: neighborhoods
- Status: published
- Published: 2026-10-09
- Updated: not revised
- Data sources:
  - Opportunity Atlas. Tract and county outcomes of children born 1978 to 1983, native 2010 tracts, ranks as percentiles; income measured in 2014 to 2015; years: 1978-1983 birth cohorts (2018 release); weights: 2010 census tract population for tract comparisons; county population for county summaries
  - Social Capital Atlas. Economic connectedness and network measures, county and ZIP code, 2022; years: 2022 (friendship snapshot); weights: 2018 population (Social Capital Atlas)
  - County Health Rankings. County Health Rankings 2024: life expectancy by race (deaths 2019 to 2021), premature death by race, Black-white residential segregation; years: 2024 release; weights: county total population; each county once where stated
  - American Community Survey. ACS 2019 to 2023 tract file, aggregated to counties for the connectedness regression; years: 2019-2023 (5-year); weights: 2019-2023 tract population
  - USDA Food Access Research Atlas. 2010 census population of each 2010 tract, from the Food Access Research Atlas file; the food-access flags are not used; years: 2019 release (population from the 2010 census); weights: none (census counts); used as the weight for tract comparisons
- Canonical URL: https://vishalsingh.org/zipcode/race-income-and-mobility
- Report: https://vishalsingh.org/zipcode/race-income-and-mobility/report
- Markdown: https://vishalsingh.org/zipcode/race-income-and-mobility/report.md
- BibTeX key: singh2026raceincomeandmobility

```bibtex
@misc{singh2026raceincomeandmobility,
  author = {Singh, Vishal},
  title = {Race, income and mobility},
  year = {2026},
  url = {https://vishalsingh.org/zipcode/race-income-and-mobility},
  note = {In ZIP Code Destiny. Published 9 October 2026.}
}
```

## Principal estimates

The facts the dek and the opening prose cite, in order of appearance.

| Estimand | Population | Period | Value | Interval | n | Weights | Source | Fact id |
| --- | --- | --- | ---: | --- | ---: | --- | --- | --- |
| share of tracts (each tract counted once) where white children reach a higher mean adult rank than Black children, county-bootstrap interval | children raised in low-income families (parents at the 25th percentile of the national income distribution) in tracts where the Atlas estimates both white and Black children | children born 1978 to 1983, household income measured in 2014 to 2015 | 92.5% | 95% CI 91.9% to 93.0% | 30,997 tracts | none (each tract counts once) | Opportunity Atlas | pct_below |
| count of tracts | 2010 census tracts with a mean adult income rank for white and for Black children from low-income families and a recorded 2010 population | children born 1978 to 1983, household income measured in 2014 to 2015 | 30,997 |   |   | none (full count) | Opportunity Atlas | n_tracts_lad |
| mean adult household income rank, population-weighted mean over tracts (2010 tract population as the weight), county-bootstrap interval | white children raised in low-income families (parents at the 25th percentile of the national income distribution), in tracts where the Atlas also estimates Black children | children born 1978 to 1983, household income measured in 2014 to 2015 | 44.9 | 95% CI 44.5 to 45.3 | 30,997 tracts | 2010 census tract population | Opportunity Atlas | kfr_white |
| mean adult household income rank, population-weighted mean over tracts (2010 tract population as the weight), county-bootstrap interval | Black children raised in low-income families (parents at the 25th percentile of the national income distribution), in tracts where the Atlas also estimates white children | children born 1978 to 1983, household income measured in 2014 to 2015 | 34.6 | 95% CI 34.3 to 34.9 | 30,997 tracts | 2010 census tract population | Opportunity Atlas | kfr_black |
| mean over tracts of the white minus Black gap in adult income rank, population-weighted mean over tracts (2010 tract population as the weight), county-bootstrap interval | children raised in low-income families (parents at the 25th percentile of the national income distribution) in tracts where the Atlas estimates both white and Black children | children born 1978 to 1983, household income measured in 2014 to 2015 | 10.3 | 95% CI 10.0 to 10.6 | 30,997 tracts | 2010 census tract population | Opportunity Atlas | kfr_gap |

## Findings

White children from low-income families reached a higher adult income rank than Black children from low-income families raised in the same census tract in 92.5% of the 30,997 tracts where the Opportunity Atlas estimates both groups.[1] The mean rank, weighted by tract population, was 44.9 for white children and 34.6 for Black children, a gap of 10.3 (95% CI 10.0 to 10.6) percentile points. Both groups were born in 1978 to 1983 to parents at the 25th percentile of the national income distribution, and the rank is the position they reached in the national distribution of household income in their thirties.

Comparing two groups inside one tract removes the differences between neighborhoods that a county or ZIP code average mixes in. The chapter applies the same logic to parental income and to life expectancy within counties, and it asks what predicts upward mobility across counties. Every Atlas figure describes children who grew up in the 1980s and 1990s, and none describes the same tracts today.

### Upward mobility across counties

The Atlas follows the children of each tract from childhood into adult tax and census records, so it can map how far children from a given parental income level rise.[1] For parents at the 25th percentile, the mean adult rank in a county averages 41.3 across counties weighted by population. It runs from 34.9 in the county at the fifth percentile to 48.4 in the county at the ninety-fifth.

Among the states, means are highest in North Dakota (53.3) and lowest in South Carolina (35.9). The highest state means lie in the Plains and the Mountain West, and the lowest lie in the Southeast (Table 1). The District of Columbia, a single county that is left out of that ranking, has a lower value than any state, 34.9. The map averages over every child in a county, and the next section separates two groups of them.

Table 1. Children from low-income families reach the highest adult income ranks in the Plains and Mountain West and the lowest in the Southeast

### Black and white children from the same tract

Because the Atlas reports outcomes by race, it can place children of two groups in one tract, with parents at the same income percentile, and compare them. Table 2 plots one dot per tract. If a tract gave children of both groups the same outcome, its dot would sit on the diagonal. The dot sits below the diagonal in 92.5% of the tracts. A tract has both estimates only where the Atlas can measure enough children of each group, so these tracts are those with sizeable Black and white populations and not a sample of all tracts. They hold 144.8 million people, 46.9% of the 2010 population. The Atlas keeps Hispanic children, of any race, apart from non-Hispanic white, Black and Asian children, and the white, Black and Asian children in this chapter are non-Hispanic.[2]

The gap is wider where white children do best, though by less than a tract-by-tract comparison suggests. Averaging the tracts within each county damps the noise in the tract estimates, and 336 counties have at least 20 tracts with both estimates. In the lowest-ranked third of these counties by white rank, white children ranked 8.4 (95% CI 8.1 to 8.7) points above Black children, and in the highest-ranked third they ranked 11.6 (95% CI 10.8 to 12.1) points above. Across the counties, each point of difference in the white rank goes with a difference of 0.60 (95% CI 0.49 to 0.71) points in the Black rank. Where white children rank higher, Black children rank higher too, by a smaller amount.

Sorting tracts on the white rank itself gives a larger contrast, 5.6 points in the lowest-ranked third and 14.8 in the highest, and a slope of 0.33. Noise in the white estimate both sorts the tracts and enters the white minus Black difference, so the tract figures overstate the contrast.

Table 2. White children from low-income families reached a higher adult income rank than Black children from the same tract in 92.5% of 30,997 tracts

 (Note: Each tract's estimate for a group rests on a modest number of children, and noise is added to protect privacy. Noise of that kind spreads dots in both directions around the diagonal. It would pull the share of tracts below the diagonal toward an even split, not away from one.)

The comparison is not limited to two groups. In tracts where white and Hispanic children are both estimated, Hispanic children ranked 2.6 (95% CI 2.3 to 2.9) points below white children. In tracts where white and Asian children are both estimated, Asian children ranked 9.4 (95% CI 8.9 to 9.8) points above them (Table 3). Each comparison uses its own set of tracts, so the three gaps are not measured on one set of tracts. The Atlas also measures incarceration on Census Day 2010, and Chapter 3.4, School discipline and incarceration (https://vishalsingh.org/zipcode/school-discipline-and-incarceration) reports the same-tract comparison for that outcome.

Table 3. Black children ranked 10.3 points below white children in the same tracts, Hispanic children 2.6 points below and Asian children 9.4 points above

### The advantage of a higher-income family, tract by tract

The Atlas also reports adult ranks for children whose parents were at the 75th percentile, with races pooled. The difference between that rank and the rank of children at the 25th percentile is a tract-level measure of the advantage of a higher-income family, which this chapter calls the premium. Across 71,924 tracts the mean rank was 43.2 for children at the 25th percentile and 56.8 for children at the 75th, a premium of 13.6 (95% CI 13.2 to 13.9) points.

The premium varies across tracts, from 5.8 at the fifth percentile of tracts to 21.3 at the ninety-fifth, and sampling noise in the tract estimates widens that range. The premium is smaller where children from low-income families do better. Across tracts, the rank of children at the 75th percentile rises 0.69 (95% CI 0.66 to 0.72) points for each point of the rank of children at the 25th, a line flatter than the diagonal in Table 4. The two ladders sit closer together in tracts where the lower ladder is higher.

Sampling noise in the rank of children at the 25th percentile pulls a tract slope toward zero, so the slope is also estimated on county averages. It is 0.70 (95% CI 0.60 to 0.82) for counties with at least 10 tracts and 0.81 (95% CI 0.69 to 0.91) for counties with at least 100 tracts, higher than the tract slope and still flatter than the diagonal.

Table 4. Children raised at the 75th income percentile reach adult ranks 13.6 points above those raised at the 25th in the same tract, a difference that is smaller where the latter do better

Among the 148 counties with at least 100 tracts, the premium is smallest in the Bronx (8.0). The 8 smallest premiums are all in New York City or California, among them San Francisco (8.9).

The premium is largest in Hamilton County, Ohio (17.3), and Luzerne County, Pennsylvania (16.9). A small premium is not a sign of a better place for children from low-income families. In the Bronx, children at the 25th percentile reached a mean rank of 39.9 and children at the 75th percentile reached 47.9, both below the mean for all tracts. In San Francisco the ranks were 50.3 and 59.2, both above it.

### Friendships across incomes and upward mobility

Across counties, upward mobility moves with who a place's residents befriend. The Social Capital Atlas builds economic connectedness from the friendships of Facebook users. It compares the share of high-status friends among people with below-median status to the share expected if friendships ignored status.[3] Across 2,942 counties, economic connectedness correlates +0.69 (95% CI 0.64 to 0.74) with the mean adult rank of children from low-income families, weighted by population.

That correlation is larger than the correlation of any civic or cohesion measure with mobility (Table 5). The density of civic organizations correlates +0.06 (95% CI −0.01 to 0.14) with it, the volunteering rate +0.19 (95% CI 0.12 to 0.25) and the network support ratio −0.25 (95% CI −0.33 to −0.17). Civic participation is the subject of Chapter 6.2, Civic life, loneliness and partnership (https://vishalsingh.org/zipcode/civic-life-loneliness-and-partnership).

Table 5. Across counties, economic connectedness correlates +0.69 with upward mobility, more than any civic or cohesion measure

Connectedness also moves with income. Across 18,896 ZIP code areas (ZCTAs), economic connectedness correlates +0.82 (95% CI 0.81 to 0.83) with the area's rank in median household income. Its mean is 0.57 (95% CI 0.56 to 0.58) in the lowest tenth of areas by median income and 1.19 (95% CI 1.16 to 1.21) in the highest tenth. Because connectedness is higher where incomes are higher, the county correlation could partly reflect income.

A weighted regression puts the log of median household income, the Black share of residents, the share with a college degree and the poverty rate beside connectedness. In it, the standardized coefficient on economic connectedness is 0.56 (95% CI 0.42 to 0.70) and the coefficient on the log of income is 0.46 (95% CI 0.26 to 0.67). The difference between them is 0.10 (95% CI −0.16 to 0.36), and its interval includes zero, so the data do not rank the two. The coefficient on connectedness stays positive with income in the model.

Economic connectedness is measured in 2022, long after the 1978 to 1983 cohorts grew up. The correlation links a recent network measure to an earlier cohort, and the data cannot show that friendships across incomes change a child's later income.

### Life expectancy of Black and white residents of the same county

The comparison inside one place extends to adult health. In County Health Rankings, 1,265 counties publish a life expectancy for both non-Hispanic white and non-Hispanic Black residents, for deaths in 2019 to 2021.[4] White life expectancy exceeds Black by 4.4 (95% CI 4.0 to 4.8) years when counties count in proportion to their population, and by 3.5 (95% CI 3.3 to 3.7) years when each county counts once.

The population-weighted means are 78.1 years for white residents and 73.7 years for Black residents. White life expectancy is the higher in 89% of these counties, and the counties hold 98% of the non-Hispanic Black residents of all counties in the file.

Table 6 shows the 20 most populous counties in that sample. The gap is positive in each, from 1.1 (95% CI 0.7 to 1.5) years in Broward County, Florida, to 9.9 (95% CI 9.7 to 10.2) years in Cook County, Illinois. Among all 1,265 counties, the widest gap is in the District of Columbia, at 17.7 (95% CI 17.0 to 18.3) years.

Table 6. In each of the 20 most populous counties, white residents live longer than Black residents, by 1.1 to 9.9 years

In 140 counties the estimate for Black residents is the higher. County Health Rankings publishes an interval for each estimate. The two intervals separate in that direction in only 9 counties, against 677 counties where the interval for white residents lies entirely above the interval for Black residents. The counties with a higher Black estimate have a median population of 59,039, below the 82,911 median of the sample, which is consistent with noisier estimates in smaller counties.

The gap extends to premature mortality, which County Health Rankings measures as age-adjusted deaths of residents under age 75 per hundred thousand residents. Across 1,242 of these counties that report both rates, the population-weighted mean for Black residents is 1.5 (95% CI 1.5 to 1.6) times the mean for white residents. This is a death rate and not the years of potential life lost. Infant mortality by race is covered in Chapter 2.2, Infant mortality (https://vishalsingh.org/zipcode/infant-mortality).

The gap is wider where residents are more segregated. The Black-white index of dissimilarity gives the share of either group that would have to move for the two to be spread evenly across tracts. Its population-weighted correlation with the gap is +0.30 (95% CI 0.14 to 0.45) across 1,260 counties. Within states, each additional point of the index goes with 0.06 (95% CI 0.00 to 0.12) more years of gap (state fixed effects), and the lower end of that interval is close to zero.

### Race, place and their interaction

Life expectancy differs across counties and between the two groups, and a decomposition says how much of the variation each source holds. The cells are the county-by-race life expectancies, a white and a Black estimate for each of the 1,265 counties. The variance across the cells splits into three parts. Place is the spread of the county averages, race is the average white-Black gap, and the interaction is the spread of the gap across counties. Weighted by county population, place accounts for 53% (95% CI 48% to 58%) of the variance, race for 31% (95% CI 26% to 37%) and the interaction for 16% (95% CI 13% to 19%).

When each county counts once, the shares are 64% (95% CI 62% to 66%) for place, 19% (95% CI 17% to 22%) for race and 17% (95% CI 15% to 19%) for the interaction. The race share is larger when large counties carry more weight, because the average gap is larger there, so the chapter reports both.

Table 7. Place accounts for 53% of the variance in county life expectancy by race, race for 31% and their interaction for 16%

Weighted by county population, the 90th percentile of Black life expectancy across counties is 78.2 years, above the 10th percentile of white life expectancy, 74.4 years. The two groups' distributions across places overlap, and the white estimate is still the higher in 89% of counties.

> Causal evidence: Whether place changes a child's adult outcomes (Randomized housing-voucher experiment; movers design)
>
> The comparisons above describe differences across places and groups. Two designs test whether place changes a child's own outcome.
>
> Moving to Opportunity was a randomized housing-voucher experiment. Families in high-poverty public housing were randomly assigned to be offered a voucher that could be used only in a low-poverty neighborhood, to be offered an unrestricted voucher, or to stay in public housing. Chetty, Hendren and Katz followed the children into adulthood and found that those who were young when their families moved earned more and attended college more often as adults, while the estimated effects were negative for children who were teenagers at the move.[5]
>
> Chetty and Hendren used a movers design. They compared children whose families moved between counties and commuting zones when the children were at different ages, including siblings in the same family. Because the age at the move varies while the family stays the same, the design isolates the years a child spent in the destination. Children's adult outcomes moved toward those of children raised in the destination in proportion to the years spent there.[6] A companion paper applies the design county by county,[7] and the Opportunity Atlas authors combine the experimental and movers estimates and report that 60% of the variation in outcomes across neighborhoods is driven by causal effects of the neighborhoods.[1]
>
> Neither design tests whether the Black-white gap inside a tract would close if the tract changed. Both estimate how much a move changes a child's own outcome, and neither compares the gap before and after.

### What these data cannot show

The patterns here describe tracts, ZIP code areas and counties. The average life expectancy of Black residents of Cook County says little about how long any one Black resident of Cook County will live, because the data are averages for places and groups (ecological inference).

A same-tract comparison removes the differences between neighborhoods but not all differences. Within a tract, families at one income percentile can still differ in wealth, in the exact block, in schooling and in how institutions treat them. The gap is a residual, and the data describe it without explaining it. Chetty, Hendren, Jones and Porter find that the gap in individual adult income between Black and white children raised at the same parental income is concentrated among boys and men. They find it smaller in tracts with lower poverty and a higher share of Black fathers present, a comparison across places and not an experiment.[2] The ranks here pool boys and girls and measure household income, which includes a spouse's income, because race by gender is not available in the staged data (see Chapter 4.4, Boys and mobility (https://vishalsingh.org/zipcode/boys-and-mobility; planned)).

The Atlas estimates are for children born in 1978 to 1983 and observed in 2014 to 2015. Each tract estimate rests on a modest number of children and carries privacy noise, which is not clipped. Tracts weigh in by their 2010 population because the file has no child counts, and the tracts with both estimates are those with enough children of both groups.

The life-expectancy estimates are computed from death counts that are small in small counties, so the share of counties with a reversed gap and the shares in the decomposition include sampling noise. County Health Rankings counts Black and white residents who are not Hispanic, and the index of dissimilarity is measured for 2018 to 2022. The connectedness measures come from active Facebook users and describe places in 2022, and the regression cannot separate connectedness from income more finely than its interval allows. Puerto Rico has Atlas estimates but is not drawn on the maps, and Connecticut is missing from the regression because its tract codes differ across the files.

## Tables

### Table 1. Children from low-income families reach the highest adult income ranks in the Plains and Mountain West and the lowest in the Southeast

Mean adult household income rank (percentile) of children born 1978 to 1983 to parents at the 25th income percentile, by county of childhood

No inline rows: the values of Table 1 are in https://vishalsingh.org/zipcode/data/race-income-and-mobility/mobility_map.json.

Note: Each gray county has the reason for its missing value in its tooltip. Puerto Rico has estimates in the Atlas but is not drawn.

Note: The county estimates are model-based and carry privacy noise, so single small counties are noisy. The Atlas lists Connecticut under its legacy counties, which are the eight drawn here, and Valdez-Cordova Census Area (split in 2019) is drawn as its two successors with the same value.

Note: Alaska is drawn. The Atlas estimates cover 1978 to 1983 birth cohorts and describe where children grew up, not the counties today.

Source: Opportunity Atlas county outcomes, 1978-1983 birth cohorts; modeled estimates, not weighted

Data: https://vishalsingh.org/zipcode/report/race-income-and-mobility/figures/mobility_map.json

### Table 2. White children from low-income families reached a higher adult income rank than Black children from the same tract in 92.5% of 30,997 tracts

Adult household income rank (percentile) of Black children against white children raised in the same census tract by parents at the 25th income percentile, one dot per tract

No inline rows: the values of Table 2 are in https://vishalsingh.org/zipcode/data/race-income-and-mobility/same_tract_black_white.json.

Note: The dots are a random sample of 5,000 tracts inside the plotted range (seed 12345); every estimate in the text uses all 30,997 tracts. Tracts have a Black and a white estimate only where the Atlas has enough children of both groups, so they are not a sample of all tracts.

Note: The markers average the tracts of each county that has at least 20 tracts with both estimates, because sorting tracts on the noisy white rank would flatten a line of tract means and widen the apparent contrast. Each tract is weighted by its 2010 census population, the nearest available stand-in for the number of children. Ranks are percentiles of the national distribution of household income for children born in 1978 to 1983, measured in 2014 to 2015. The Opportunity Atlas adds noise to its estimates to protect privacy and the staged values are never clipped, so a few fall outside 0 to 100.

Source: Opportunity Atlas tract outcomes, 2010 tracts, 1978-1983 birth cohorts; population-weighted (2010 tract population), modeled estimates

Data: https://vishalsingh.org/zipcode/report/race-income-and-mobility/figures/same_tract_black_white.json

### Table 3. Black children ranked 10.3 points below white children in the same tracts, Hispanic children 2.6 points below and Asian children 9.4 points above

Mean adult household income rank of each group minus that of white children raised in the same tract by parents at the 25th income percentile, tracts where both are estimated

| conf_high | conf_low | estimate | group | n |
| ---: | ---: | ---: | --- | ---: |
| −10.03 | −10.60 | −10.31 | Black children | 30,997 |
| −2.32 | −2.92 | −2.61 | Hispanic children | 35,579 |
| 9.83 | 8.89 | 9.37 | Asian children | 15,078 |

Note: Each row uses its own set of tracts, the tracts where that group and white children both have an estimate, so the rows are not a single sample. The number of tracts is in each row's tooltip. Ranks are percentiles of the national distribution of household income for children born in 1978 to 1983, measured in 2014 to 2015. The Opportunity Atlas adds noise to its estimates to protect privacy and the staged values are never clipped, so a few fall outside 0 to 100.

Source: Opportunity Atlas tract outcomes, 2010 tracts, 1978-1983 birth cohorts; population-weighted (2010 tract population), modeled estimates

Data: https://vishalsingh.org/zipcode/report/race-income-and-mobility/figures/group_gaps_same_tract.json

### Table 4. Children raised at the 75th income percentile reach adult ranks 13.6 points above those raised at the 25th in the same tract, a difference that is smaller where the latter do better

Adult household income rank (percentile) of children raised at the 75th income percentile against those raised at the 25th, one dot per tract, races pooled

No inline rows: the values of Table 4 are in https://vishalsingh.org/zipcode/data/race-income-and-mobility/born_rich_premium.json.

Note: The dots are a random sample of 5,000 tracts inside the plotted range (seed 12345); the means, slope and intervals use all 71,924 tracts with both estimates.

Note: The fitted line is the population-weighted least-squares line through the tracts, with slope 0.69. Sampling noise in the 25th-percentile rank pulls a tract slope toward zero. Averaging tracts to counties damps that noise: the slope is 0.70 for counties with at least 10 tracts and 0.81 for counties with at least 100, and all of them are flatter than the diagonal. Ranks are percentiles of the national distribution of household income for children born in 1978 to 1983, measured in 2014 to 2015. The Opportunity Atlas adds noise to its estimates to protect privacy and the staged values are never clipped, so a few fall outside 0 to 100.

Source: Opportunity Atlas tract outcomes, 2010 tracts, 1978-1983 birth cohorts; population-weighted (2010 tract population), modeled estimates

Data: https://vishalsingh.org/zipcode/report/race-income-and-mobility/figures/born_rich_premium.json

### Table 5. Across counties, economic connectedness correlates +0.69 with upward mobility, more than any civic or cohesion measure

Population-weighted correlation of each Social Capital Atlas county measure with the mean adult income rank of children from low-income families

| conf_high | conf_low | estimate | group | measure | n |
| ---: | ---: | ---: | --- | --- | ---: |
| 0.741 | 0.644 | 0.693 | Contact across incomes | Economic connectedness | 2,942 |
| 0.734 | 0.650 | 0.695 | Contact across incomes | Exposure to high-income friends | 2,939 |
| −0.076 | −0.316 | −0.192 | Contact across incomes | Friending bias | 2,939 |
| 0.117 | −0.012 | 0.052 | Cohesion | Network clustering | 2,983 |
| −0.168 | −0.332 | −0.254 | Cohesion | Support ratio | 2,983 |
| 0.253 | 0.121 | 0.187 | Civic life | Volunteering rate | 2,983 |
| 0.138 | −0.011 | 0.062 | Civic life | Civic organizations | 2,983 |

Note: Economic connectedness is built from Facebook friendships among active users in 2022, while the mobility ranks describe children who grew up in the 1980s and 1990s, so the correlation links a recent network measure to an earlier cohort.

Note: Intervals resample whole counties. Counties in one state tend to resemble each other, and resampling whole states gives a wider interval for every measure.

Note: Counties with a missing measure drop out of that row only, so the number of counties differs by row.

Source: Social Capital Atlas county measures, 2022, and Opportunity Atlas county outcomes, 1978-1983 birth cohorts; population-weighted (2018 county population)

Data: https://vishalsingh.org/zipcode/report/race-income-and-mobility/figures/connectedness_horse_race.json

### Table 6. In each of the 20 most populous counties, white residents live longer than Black residents, by 1.1 to 9.9 years

Life expectancy at birth in years, non-Hispanic Black and non-Hispanic white residents, for the most populous counties with both estimates, deaths in 2019 to 2021

| county | hi_black | hi_white | le_black | le_white | lo_black | lo_white | population |
| --- | ---: | ---: | ---: | ---: | ---: | ---: | ---: |
| Cook County, IL | 70.41 | 80.28 | 70.23 | 80.17 | 70.04 | 80.06 | 5109292 |
| Alameda County, CA | 73.56 | 81.60 | 73.07 | 81.37 | 72.59 | 81.15 | 1628997 |
| Wayne County, MI | 69.01 | 76.61 | 68.77 | 76.43 | 68.53 | 76.25 | 1757043 |
| Los Angeles County, CA | 73.40 | 79.93 | 73.19 | 79.84 | 72.97 | 79.74 | 9721138 |
| Maricopa County, AZ | 73.51 | 79.17 | 73.08 | 79.06 | 72.65 | 78.95 | 4551524 |
| King County, WA | 76.03 | 81.56 | 75.48 | 81.41 | 74.92 | 81.26 | 2266789 |
| Miami-Dade County, FL | 73.69 | 78.80 | 73.37 | 78.53 | 73.06 | 78.25 | 2673837 |
| Harris County, TX | 72.89 | 77.81 | 72.68 | 77.67 | 72.47 | 77.52 | 4780913 |
| San Diego County, CA | 76.09 | 80.64 | 75.56 | 80.51 | 75.04 | 80.37 | 3276208 |
| Dallas County, TX | 72.92 | 77.72 | 72.67 | 77.53 | 72.42 | 77.34 | 2600840 |
| Kings County, NY | 77.81 | 82.14 | 77.57 | 81.97 | 77.33 | 81.79 | 2590516 |
| Clark County, NV | 72.18 | 76.33 | 71.80 | 76.15 | 71.42 | 75.98 | 2322985 |
| Santa Clara County, CA | 79.30 | 82.61 | 78.26 | 82.40 | 77.21 | 82.18 | 1870945 |
| San Bernardino County, CA | 71.37 | 74.84 | 70.92 | 74.61 | 70.46 | 74.39 | 2193656 |
| Riverside County, CA | 74.90 | 77.92 | 74.39 | 77.72 | 73.88 | 77.53 | 2473902 |
| Tarrant County, TX | 74.29 | 77.28 | 73.95 | 77.11 | 73.61 | 76.94 | 2154595 |
| Orange County, CA | 78.78 | 81.01 | 77.87 | 80.86 | 76.96 | 80.72 | 3151184 |
| Queens County, NY | 77.92 | 80.61 | 77.62 | 80.39 | 77.31 | 80.17 | 2278029 |
| Bexar County, TX | 74.40 | 76.82 | 73.89 | 76.59 | 73.38 | 76.36 | 2059530 |
| Broward County, FL | 77.48 | 78.50 | 77.18 | 78.28 | 76.88 | 78.05 | 1947026 |

Note: County Health Rankings computes life expectancy from NCHS death counts for racial and ethnic groups that do not overlap, so the Black and white groups here are non-Hispanic and exclude Hispanic residents of any race. Counties with fewer deaths have wider intervals, and the intervals in this figure are County Health Rankings' own.

Note: The counties are the most populous ones among those with both estimates, which is the sample behind the headline gap.

Source: County Health Rankings 2024 (NCHS deaths 2019-2021); full counts of deaths, county life expectancy by race

Data: https://vishalsingh.org/zipcode/report/race-income-and-mobility/figures/le_black_white_counties.json

### Table 7. Place accounts for 53% of the variance in county life expectancy by race, race for 31% and their interaction for 16%

Shares of the variance of county-by-race life expectancy (a non-Hispanic white and a non-Hispanic Black estimate for each county), percent, deaths in 2019 to 2021

| Weighting | Component | Share of variance (%) | Lower limit of the 95% interval | Upper limit of the 95% interval |
| --- | --- | ---: | ---: | ---: |
| Weighted by county population | Place | 53.0 | 47.5 | 58.5 |
| Weighted by county population | Race | 31.3 | 25.7 | 37.1 |
| Weighted by county population | Race-by-place interaction | 15.7 | 13.0 | 18.8 |
| Each county counted once | Place | 64.1 | 61.8 | 66.5 |
| Each county counted once | Race | 19.1 | 16.7 | 21.7 |
| Each county counted once | Race-by-place interaction | 16.8 | 15.0 | 18.6 |

Note: Shares are exact for the county estimates in the file and sum to 100 in each bar. The 95% intervals in the table come from a bootstrap over counties.

Note: The estimates carry sampling noise that is largest in small counties; the noise sits inside the place and interaction shares and is not removed.

Note: Both groups of a county get the same weight, which keeps the split exact: the interaction equals the variance of the county gaps divided by four.

Source: County Health Rankings 2024 (NCHS deaths 2019-2021); full counts of deaths, weighted by county population in the first bar and unweighted in the second

Data: https://vishalsingh.org/zipcode/report/race-income-and-mobility/figures/le_variance_race_place.json

## All estimates

Every fact in facts.json (91), sorted by id.

| Fact id | Value | Unit | Estimand | Population | Period | Interval | n | Weights | Source | Note |
| --- | ---: | --- | --- | --- | --- | --- | ---: | --- | --- | --- |
| best_black | 78.2 | years of life expectancy at birth | 90th percentile of county Black life expectancy, weighted by county population | Black residents of the counties with both a white and a Black life expectancy in County Health Rankings | deaths in 2019 to 2021 (County Health Rankings 2024 release) |   | 1,265 counties | county total population | County Health Rankings |   |
| beta_diff | 0.10 | standard deviations of the county mobility rank per standard deviation of the measure | standardized coefficient on economic connectedness minus the coefficient on log median income, same regression, errors clustered on state | counties with a Social Capital Atlas measure and an Opportunity Atlas mobility rank and ACS controls (Connecticut left out) | friendships among Facebook users in 2022; mobility for children born 1978 to 1983; ACS 2019 to 2023 | 95% CI −0.16 to 0.36 | 2,933 counties | 2018 county population (Social Capital Atlas) | Social Capital Atlas |   |
| black_share_covered | 98% | percent of non-Hispanic Black residents | share of non-Hispanic Black residents living in counties with both race-specific estimates | residents of the counties in County Health Rankings | County Health Rankings 2024 release (population estimates for 2022) |   | 3,143 counties | county population times its non-Hispanic Black share | County Health Rankings |   |
| bronx_p25 | 39.9 | percentile points of national household income rank in adulthood | population-weighted mean adult household income rank over the county's tracts | children raised at the 25th percentile of the national income distribution in Bronx County | children born 1978 to 1983, household income measured in 2014 to 2015 |   | 328 tracts | 2010 census tract population | Opportunity Atlas |   |
| bronx_p75 | 47.9 | percentile points of national household income rank in adulthood | population-weighted mean adult household income rank over the county's tracts | children raised at the 75th percentile of the national income distribution in Bronx County | children born 1978 to 1983, household income measured in 2014 to 2015 |   | 328 tracts | 2010 census tract population | Opportunity Atlas |   |
| civic_r | +0.06 | correlation coefficient | population-weighted correlation of the county civic organizations with the county mobility rank, bootstrap interval over counties | counties with a Social Capital Atlas measure and an Opportunity Atlas mobility rank | friendships among Facebook users in 2022; mobility for children born 1978 to 1983 | 95% CI −0.01 to 0.14 | 2,983 counties | 2018 county population (Social Capital Atlas) | Social Capital Atlas |   |
| conv_slope | 0.69 | percentile points of the 75th-percentile family's rank per point of the 25th-percentile family's rank | slope of the tract's 75th-percentile rank on its 25th-percentile rank, population-weighted, county-bootstrap interval | children raised in families at the 25th and the 75th percentile of the national income distribution, in tracts where the Atlas estimates both | children born 1978 to 1983, household income measured in 2014 to 2015 | 95% CI 0.66 to 0.72 | 71,924 tracts | 2010 census tract population | Opportunity Atlas |   |
| conv_slope_county | 0.70 | percentile points of the 75th-percentile family's rank per point of the 25th-percentile family's rank | slope of the county mean 75th-percentile rank on the county mean 25th-percentile rank, weighted by 2010 population, state-bootstrap interval | counties with at least 10 tracts in the premium sample | children born 1978 to 1983, household income measured in 2014 to 2015 | 95% CI 0.60 to 0.82 | 1,172 counties | 2010 census tract population | Opportunity Atlas | Averaging tracts within counties damps sampling noise in the tract estimates; the slope rises with the minimum number of tracts. |
| conv_slope_county_big | 0.81 | percentile points of the 75th-percentile family's rank per point of the 25th-percentile family's rank | slope of the county mean 75th-percentile rank on the county mean 25th-percentile rank, weighted by 2010 population, state-bootstrap interval | counties with at least 100 tracts in the premium sample | children born 1978 to 1983, household income measured in 2014 to 2015 | 95% CI 0.69 to 0.91 | 148 counties | 2010 census tract population | Opportunity Atlas |   |
| cook_gap | 9.9 | years of life expectancy at birth, white minus Black | white minus Black county life expectancy; the interval combines County Health Rankings' two 95% intervals assuming independence | residents of Cook County, Illinois | deaths in 2019 to 2021 (County Health Rankings 2024 release) | 95% CI 9.7 to 10.2 |   | none (one county) | County Health Rankings |   |
| dc_gap | 17.7 | years of life expectancy at birth, white minus Black | white minus Black county life expectancy; the interval combines County Health Rankings' two 95% intervals assuming independence | residents of the District of Columbia | deaths in 2019 to 2021 (County Health Rankings 2024 release) | 95% CI 17.0 to 18.3 |   | none (one county) | County Health Rankings |   |
| ec_beta | 0.56 | standard deviations of the county mobility rank per standard deviation of the measure | standardized coefficient on economic connectedness in a weighted regression with controls for log median income, percent Black, percent with a college degree and percent below poverty, errors clustered on state | counties with a Social Capital Atlas measure and an Opportunity Atlas mobility rank and ACS controls (Connecticut left out) | friendships among Facebook users in 2022; mobility for children born 1978 to 1983; ACS 2019 to 2023 | 95% CI 0.42 to 0.70 | 2,933 counties | 2018 county population (Social Capital Atlas) | Social Capital Atlas |   |
| ec_n | 2,942 | counties | count of counties with economic connectedness and a mobility rank | counties with a Social Capital Atlas measure and an Opportunity Atlas mobility rank | friendships among Facebook users in 2022; mobility for children born 1978 to 1983 |   |   | none (full count) | Social Capital Atlas |   |
| ec_poor | 0.57 | economic connectedness (ratio) | population-weighted mean of ZIP economic connectedness, bootstrap interval over counties | ZIP code areas (ZCTAs) in the lowest tenth of median household income | friendships among Facebook users in 2022; income ACS 2019 to 2023 | 95% CI 0.56 to 0.58 | 1,889 ZIP code areas | 2018 ZIP population (Social Capital Atlas) | Social Capital Atlas |   |
| ec_r | +0.69 | correlation coefficient | population-weighted correlation of the county economic connectedness with the county mobility rank, bootstrap interval over counties | counties with a Social Capital Atlas measure and an Opportunity Atlas mobility rank | friendships among Facebook users in 2022; mobility for children born 1978 to 1983 | 95% CI 0.64 to 0.74 | 2,942 counties | 2018 county population (Social Capital Atlas) | Social Capital Atlas |   |
| ec_rich | 1.19 | economic connectedness (ratio) | population-weighted mean of ZIP economic connectedness, bootstrap interval over counties | ZIP code areas (ZCTAs) in the highest tenth of median household income | friendships among Facebook users in 2022; income ACS 2019 to 2023 | 95% CI 1.16 to 1.21 | 1,890 ZIP code areas | 2018 ZIP population (Social Capital Atlas) | Social Capital Atlas |   |
| gap_high_third | 11.6 | percentile points of national household income rank in adulthood, white minus Black | mean of the county white minus Black gap in adult income rank within a third of counties (equal numbers of counties, ranked by the county mean white rank), weighted by 2010 population summed to the county, state-bootstrap interval | children raised in low-income families (parents at the 25th percentile of the national income distribution) in counties with at least 20 tracts that have both estimates, in the third of those counties with the highest mean rank for white children | children born 1978 to 1983, household income measured in 2014 to 2015 | 95% CI 10.8 to 12.1 | 112 counties | 2010 census tract population | Opportunity Atlas |   |
| gap_high_third_tract | 14.8 | percentile points of national household income rank in adulthood, white minus Black | mean white minus Black gap in adult income rank within a third of tracts (equal numbers of tracts, ranked by the white rank), population-weighted mean over tracts (2010 tract population as the weight), county-bootstrap interval | children raised in low-income families (parents at the 25th percentile of the national income distribution) in tracts where the Atlas estimates both white and Black children in the third of tracts with the highest mean rank for white children | children born 1978 to 1983, household income measured in 2014 to 2015 | 95% CI 14.4 to 15.2 | 10,333 tracts | 2010 census tract population | Opportunity Atlas | Sorting on the noisy white rank itself builds part of this contrast into the thirds. |
| gap_hisp | 2.6 | percentile points of national household income rank in adulthood, white minus Hispanic | mean over tracts of the white minus Hispanic gap in adult income rank, population-weighted mean over tracts (2010 tract population as the weight), county-bootstrap interval | children raised in low-income families (parents at the 25th percentile of the national income distribution) in tracts where the Atlas estimates both white and Hispanic children | children born 1978 to 1983, household income measured in 2014 to 2015 | 95% CI 2.3 to 2.9 | 35,579 tracts | 2010 census tract population | Opportunity Atlas |   |
| gap_low_third | 8.4 | percentile points of national household income rank in adulthood, white minus Black | mean of the county white minus Black gap in adult income rank within a third of counties (equal numbers of counties, ranked by the county mean white rank), weighted by 2010 population summed to the county, state-bootstrap interval | children raised in low-income families (parents at the 25th percentile of the national income distribution) in counties with at least 20 tracts that have both estimates, in the third of those counties with the lowest mean rank for white children | children born 1978 to 1983, household income measured in 2014 to 2015 | 95% CI 8.1 to 8.7 | 112 counties | 2010 census tract population | Opportunity Atlas |   |
| gap_low_third_tract | 5.6 | percentile points of national household income rank in adulthood, white minus Black | mean white minus Black gap in adult income rank within a third of tracts (equal numbers of tracts, ranked by the white rank), population-weighted mean over tracts (2010 tract population as the weight), county-bootstrap interval | children raised in low-income families (parents at the 25th percentile of the national income distribution) in tracts where the Atlas estimates both white and Black children in the third of tracts with the lowest mean rank for white children | children born 1978 to 1983, household income measured in 2014 to 2015 | 95% CI 5.4 to 5.8 | 10,332 tracts | 2010 census tract population | Opportunity Atlas | Sorting on the noisy white rank itself builds part of this contrast into the thirds. |
| gap_min_big | 1.1 | years of life expectancy at birth, white minus Black | white minus Black county life expectancy; the interval combines County Health Rankings' two 95% intervals assuming independence | residents of Broward County, FL, the county with the smallest gap among the most populous counties | deaths in 2019 to 2021 (County Health Rankings 2024 release) | 95% CI 0.7 to 1.5 |   | none (one county) | County Health Rankings |   |
| inc_beta | 0.46 | standard deviations of the county mobility rank per standard deviation of the measure | standardized coefficient on the log of median household income in a weighted regression with controls for log median income, percent Black, percent with a college degree and percent below poverty, errors clustered on state | counties with a Social Capital Atlas measure and an Opportunity Atlas mobility rank and ACS controls (Connecticut left out) | friendships among Facebook users in 2022; mobility for children born 1978 to 1983; ACS 2019 to 2023 | 95% CI 0.26 to 0.67 | 2,933 counties | 2018 county population (Social Capital Atlas) | Social Capital Atlas |   |
| kfr_black | 34.6 | percentile points of national household income rank in adulthood | mean adult household income rank, population-weighted mean over tracts (2010 tract population as the weight), county-bootstrap interval | Black children raised in low-income families (parents at the 25th percentile of the national income distribution), in tracts where the Atlas also estimates white children | children born 1978 to 1983, household income measured in 2014 to 2015 | 95% CI 34.3 to 34.9 | 30,997 tracts | 2010 census tract population | Opportunity Atlas | Counting each tract once gives a similar mean. |
| kfr_dc | 34.9 | percentile points of national household income rank in adulthood | county value from the Opportunity Atlas | children raised in low-income families in the District of Columbia, a single county that is left out of the ranking of states | children born 1978 to 1983, household income measured in 2014 to 2015 |   | 1 counties | none (one county) | Opportunity Atlas |   |
| kfr_gap | 10.3 | percentile points of national household income rank in adulthood, white minus Black | mean over tracts of the white minus Black gap in adult income rank, population-weighted mean over tracts (2010 tract population as the weight), county-bootstrap interval | children raised in low-income families (parents at the 25th percentile of the national income distribution) in tracts where the Atlas estimates both white and Black children | children born 1978 to 1983, household income measured in 2014 to 2015 | 95% CI 10.0 to 10.6 | 30,997 tracts | 2010 census tract population | Opportunity Atlas | Counting each tract once gives a similar gap. |
| kfr_mean | 41.3 | percentile points of national household income rank in adulthood | population-weighted mean of the county values | counties with an estimate for children raised at the 25th percentile of the national income distribution | children born 1978 to 1983, household income measured in 2014 to 2015 |   | 3,136 counties | county population in County Health Rankings 2024 | Opportunity Atlas |   |
| kfr_p5 | 34.9 | percentile points of national household income rank in adulthood | population-weighted 5th percentile of the county values | counties with an estimate for children raised at the 25th percentile of the national income distribution | children born 1978 to 1983, household income measured in 2014 to 2015 |   | 3,136 counties | county population in County Health Rankings 2024 | Opportunity Atlas |   |
| kfr_p95 | 48.4 | percentile points of national household income rank in adulthood | population-weighted 95th percentile of the county values | counties with an estimate for children raised at the 25th percentile of the national income distribution | children born 1978 to 1983, household income measured in 2014 to 2015 |   | 3,136 counties | county population in County Health Rankings 2024 | Opportunity Atlas |   |
| kfr_state_high | 53.3 | percentile points of national household income rank in adulthood | state mean of the county values, weighted by county population | counties with an estimate for children raised at the 25th percentile of the national income distribution in the state with the highest mean, among the 50 states (DC left out) | children born 1978 to 1983, household income measured in 2014 to 2015 |   | 53 counties | county population in County Health Rankings 2024 | Opportunity Atlas |   |
| kfr_state_low | 35.9 | percentile points of national household income rank in adulthood | state mean of the county values, weighted by county population | counties with an estimate for children raised at the 25th percentile of the national income distribution in the state with the lowest mean, among the 50 states (DC left out) | children born 1978 to 1983, household income measured in 2014 to 2015 |   | 46 counties | county population in County Health Rankings 2024 | Opportunity Atlas |   |
| kfr_white | 44.9 | percentile points of national household income rank in adulthood | mean adult household income rank, population-weighted mean over tracts (2010 tract population as the weight), county-bootstrap interval | white children raised in low-income families (parents at the 25th percentile of the national income distribution), in tracts where the Atlas also estimates Black children | children born 1978 to 1983, household income measured in 2014 to 2015 | 95% CI 44.5 to 45.3 | 30,997 tracts | 2010 census tract population | Opportunity Atlas | Counting each tract once gives a similar mean. |
| le_b | 73.7 | years of life expectancy at birth | mean of county life expectancy, weighted by county population, bootstrap interval over counties | Black residents of the counties with both a white and a Black life expectancy in County Health Rankings | deaths in 2019 to 2021 (County Health Rankings 2024 release) | 95% CI 73.4 to 74.1 | 1,265 counties | county total population | County Health Rankings |   |
| le_gap | 4.4 | years of life expectancy at birth, non-Hispanic white minus non-Hispanic Black | mean of the county non-Hispanic white minus non-Hispanic Black gap in life expectancy, weighted by county population, bootstrap interval over counties | counties with both a non-Hispanic white and a non-Hispanic Black life expectancy in County Health Rankings | deaths in 2019 to 2021 (County Health Rankings 2024 release) | 95% CI 4.0 to 4.8 | 1,265 counties | county total population | County Health Rankings | County Health Rankings publishes a race-specific life expectancy only where deaths are numerous enough. These 1,265 counties hold 98 percent of the non-Hispanic Black residents of all counties in the file. Interval: bootstrap over counties. |
| le_gap_unweighted | 3.5 | years of life expectancy at birth, non-Hispanic white minus non-Hispanic Black | mean of the county non-Hispanic white minus non-Hispanic Black gap in life expectancy, each county counted once, bootstrap interval over counties | counties with both a non-Hispanic white and a non-Hispanic Black life expectancy in County Health Rankings | deaths in 2019 to 2021 (County Health Rankings 2024 release) | 95% CI 3.3 to 3.7 | 1,265 counties | none (each county counts once) | County Health Rankings | County Health Rankings publishes a race-specific life expectancy only where deaths are numerous enough. These 1,265 counties hold 98 percent of the non-Hispanic Black residents of all counties in the file. Interval: bootstrap over counties. |
| le_w | 78.1 | years of life expectancy at birth | mean of county life expectancy, weighted by county population, bootstrap interval over counties | white residents of the counties with both a white and a Black life expectancy in County Health Rankings | deaths in 2019 to 2021 (County Health Rankings 2024 release) | 95% CI 77.8 to 78.4 | 1,265 counties | county total population | County Health Rankings |   |
| lead_asian | 9.4 | percentile points of national household income rank in adulthood, Asian minus white | mean over tracts of the Asian minus white gap in adult income rank, population-weighted mean over tracts (2010 tract population as the weight), county-bootstrap interval | children raised in low-income families (parents at the 25th percentile of the national income distribution) in tracts where the Atlas estimates both white and Asian children | children born 1978 to 1983, household income measured in 2014 to 2015 | 95% CI 8.9 to 9.8 | 15,078 tracts | 2010 census tract population | Opportunity Atlas |   |
| m25 | 43.2 | percentile points of national household income rank in adulthood | mean adult household income rank, population-weighted mean over tracts (2010 tract population as the weight), county-bootstrap interval | children raised at the 25th percentile of the national income distribution, races pooled | children born 1978 to 1983, household income measured in 2014 to 2015 | 95% CI 42.9 to 43.5 | 71,924 tracts | 2010 census tract population | Opportunity Atlas |   |
| m75 | 56.8 | percentile points of national household income rank in adulthood | mean adult household income rank, population-weighted mean over tracts (2010 tract population as the weight), county-bootstrap interval | children raised at the 75th percentile of the national income distribution, races pooled | children born 1978 to 1983, household income measured in 2014 to 2015 | 95% CI 56.4 to 57.2 | 71,924 tracts | 2010 census tract population | Opportunity Atlas |   |
| median_pop_all | 82,911 | people | median of county total population | residents of all the counties in the sample | county population in County Health Rankings 2024 |   | 1,265 counties | none (median over counties) | County Health Rankings |   |
| min_tracts_county | 100 | tracts | minimum number of tracts with both estimates for a county to be ranked | counties ranked on the premium | children born 1978 to 1983, household income measured in 2014 to 2015 |   |   | none (inclusion rule) | Opportunity Atlas |   |
| min_tracts_gap | 20 | tracts | minimum number of tracts with both estimates for a county to enter the county version | counties in the county version of the comparison | children born 1978 to 1983, household income measured in 2014 to 2015 |   |   | none (inclusion rule) | Opportunity Atlas |   |
| min_tracts_slope | 10 | tracts | minimum number of tracts with both estimates for a county to enter the slope | counties in the county-level premium slope | children born 1978 to 1983, household income measured in 2014 to 2015 |   |   | none (inclusion rule) | Opportunity Atlas |   |
| n_clear_gap | 677 | counties | count of counties where the 95% interval for white life expectancy lies entirely above the interval for Black life expectancy | counties with both a white and a Black life expectancy in County Health Rankings | deaths in 2019 to 2021 (County Health Rankings 2024 release) |   |   | none (full count) | County Health Rankings |   |
| n_clear_reversal | 9 | counties | count of counties where the 95% interval for Black life expectancy lies entirely above the interval for white life expectancy | counties with both a white and a Black life expectancy in County Health Rankings | deaths in 2019 to 2021 (County Health Rankings 2024 release) |   |   | none (full count) | County Health Rankings |   |
| n_counties_nodata | 5 | counties | count of drawn counties with no Opportunity Atlas estimate | counties and county equivalents on the map (Puerto Rico not drawn) | children born 1978 to 1983, household income measured in 2014 to 2015 |   |   | none (full count) | Opportunity Atlas |   |
| n_dumbbell | 20 | counties | number of counties drawn in the dumbbell figure | most populous counties in the life-expectancy sample | deaths in 2019 to 2021 (County Health Rankings 2024 release) |   |   | none (rule) | County Health Rankings |   |
| n_gap_counties | 336 | counties | count of counties in the county version of the comparison | children raised in low-income families (parents at the 25th percentile of the national income distribution) in counties with at least 20 tracts that have both estimates | children born 1978 to 1983, household income measured in 2014 to 2015 |   |   | none (full count) | Opportunity Atlas |   |
| n_le_cty | 1,265 | counties | count of counties with both estimates | counties with both a non-Hispanic white and a non-Hispanic Black life expectancy in County Health Rankings | deaths in 2019 to 2021 (County Health Rankings 2024 release) |   |   | none (full count) | County Health Rankings |   |
| n_outside_lad | 35 | tracts | count of tracts outside the plotted range of the same-tract scatter | children raised in low-income families (parents at the 25th percentile of the national income distribution) in tracts where the Atlas estimates both white and Black children | children born 1978 to 1983, household income measured in 2014 to 2015 |   |   | none (full count) | Opportunity Atlas |   |
| n_outside_prem | 15 | tracts | count of tracts outside the plotted range of the premium scatter | children raised in families at the 25th and the 75th percentile of the national income distribution, in tracts where the Atlas estimates both | children born 1978 to 1983, household income measured in 2014 to 2015 |   |   | none (full count) | Opportunity Atlas |   |
| n_pm_cty | 1,242 | counties | count of counties in the premature death comparison | counties with both a white and a Black life expectancy in County Health Rankings that also report both premature death rates | deaths before age 75 in 2019 to 2021 (County Health Rankings 2024 release) |   |   | none (full count) | County Health Rankings |   |
| n_prem_counties | 148 | counties | count of counties with the minimum number of tracts | counties with enough tracts to rank on the premium | children born 1978 to 1983, household income measured in 2014 to 2015 |   |   | none (full count) | Opportunity Atlas |   |
| n_prem_lowest | 8 | counties | number of counties with the smallest premiums examined for their location | counties ranked on the premium | children born 1978 to 1983, household income measured in 2014 to 2015 |   |   | none (rule) | Opportunity Atlas |   |
| n_prem_tracts | 71,924 | tracts | count of tracts | 2010 census tracts with a mean adult income rank for both parental income levels and a recorded 2010 population | children born 1978 to 1983, household income measured in 2014 to 2015 |   |   | none (full count) | Opportunity Atlas |   |
| n_reversal | 140 | counties | count of counties where Black life expectancy is higher than white | counties with both a white and a Black life expectancy in County Health Rankings | deaths in 2019 to 2021 (County Health Rankings 2024 release) |   |   | none (full count) | County Health Rankings |   |
| n_sample | 5,000 | tracts | size of a random sample drawn with a fixed seed | tracts drawn at random for each scatter plot | children born 1978 to 1983, household income measured in 2014 to 2015 |   | 30,997 tracts | none (sample size) | Opportunity Atlas |   |
| n_tracts_lad | 30,997 | tracts | count of tracts | 2010 census tracts with a mean adult income rank for white and for Black children from low-income families and a recorded 2010 population | children born 1978 to 1983, household income measured in 2014 to 2015 |   |   | none (full count) | Opportunity Atlas |   |
| n_zip | 18,896 | ZIP code areas | count of ZIP codes (ZCTAs) in the comparison | ZIP code areas (ZCTAs) with an economic connectedness value and a median household income | friendships among Facebook users in 2022 |   |   | none (full count) | Social Capital Atlas |   |
| oa_causal_share | 60% | percent of the variation in adult outcomes across neighborhoods | share of the variation across neighborhoods that experimental and quasi-experimental estimates attribute to causal effects of neighborhoods; reported in the abstract of Chetty, Friedman, Hendren, Jones and Porter (2018), copied from it and not computed in this build | census tracts of the United States, children born 1978 to 1983 | children born 1978 to 1983, household income measured in 2014 to 2015 |   |   | none (reported estimate) | Opportunity Atlas | A published estimate, not a result of this build. |
| pct_below | 92.5% | percent of tracts in which the white children's mean rank is higher | share of tracts (each tract counted once) where white children reach a higher mean adult rank than Black children, county-bootstrap interval | children raised in low-income families (parents at the 25th percentile of the national income distribution) in tracts where the Atlas estimates both white and Black children | children born 1978 to 1983, household income measured in 2014 to 2015 | 95% CI 91.9% to 93.0% | 30,997 tracts | none (each tract counts once) | Opportunity Atlas | Weighting tracts by 2010 population gives a similar share. |
| pct_inter | 16% | percent of the variance of county-by-race life expectancy | share of the total variance attributed to the race-by-place interaction (the spread of the gap across counties), weighted by county population, bootstrap interval over counties | county-by-race life expectancy (a white and a Black estimate for each county) in the counties with both a white and a Black life expectancy in County Health Rankings | deaths in 2019 to 2021 (County Health Rankings 2024 release) | 95% CI 13% to 19% | 1,265 counties | county total population | County Health Rankings |   |
| pct_inter_unw | 17% | percent of the variance of county-by-race life expectancy | share of the total variance attributed to the race-by-place interaction (the spread of the gap across counties), each county counted once, bootstrap interval over counties | county-by-race life expectancy (a white and a Black estimate for each county) in the counties with both a white and a Black life expectancy in County Health Rankings | deaths in 2019 to 2021 (County Health Rankings 2024 release) | 95% CI 15% to 19% | 1,265 counties | none (each county counts once) | County Health Rankings |   |
| pct_place | 53% | percent of the variance of county-by-race life expectancy | share of the total variance attributed to place (the spread of county averages), weighted by county population, bootstrap interval over counties | county-by-race life expectancy (a white and a Black estimate for each county) in the counties with both a white and a Black life expectancy in County Health Rankings | deaths in 2019 to 2021 (County Health Rankings 2024 release) | 95% CI 48% to 58% | 1,265 counties | county total population | County Health Rankings |   |
| pct_place_unw | 64% | percent of the variance of county-by-race life expectancy | share of the total variance attributed to place (the spread of county averages), each county counted once, bootstrap interval over counties | county-by-race life expectancy (a white and a Black estimate for each county) in the counties with both a white and a Black life expectancy in County Health Rankings | deaths in 2019 to 2021 (County Health Rankings 2024 release) | 95% CI 62% to 66% | 1,265 counties | none (each county counts once) | County Health Rankings |   |
| pct_race | 31% | percent of the variance of county-by-race life expectancy | share of the total variance attributed to race (the average white-Black gap), weighted by county population, bootstrap interval over counties | county-by-race life expectancy (a white and a Black estimate for each county) in the counties with both a white and a Black life expectancy in County Health Rankings | deaths in 2019 to 2021 (County Health Rankings 2024 release) | 95% CI 26% to 37% | 1,265 counties | county total population | County Health Rankings |   |
| pct_race_unw | 19% | percent of the variance of county-by-race life expectancy | share of the total variance attributed to race (the average white-Black gap), each county counted once, bootstrap interval over counties | county-by-race life expectancy (a white and a Black estimate for each county) in the counties with both a white and a Black life expectancy in County Health Rankings | deaths in 2019 to 2021 (County Health Rankings 2024 release) | 95% CI 17% to 22% | 1,265 counties | none (each county counts once) | County Health Rankings |   |
| pct_white_higher | 89% | percent of counties | share of counties where non-Hispanic white life expectancy exceeds non-Hispanic Black life expectancy | counties with both a non-Hispanic white and a non-Hispanic Black life expectancy in County Health Rankings | deaths in 2019 to 2021 (County Health Rankings 2024 release) |   | 1,265 counties | none (each county counts once) | County Health Rankings |   |
| pm_ratio | 1.5 | ratio of the non-Hispanic Black to the non-Hispanic white premature age-adjusted mortality rate (deaths under age 75 per 100,000) | ratio of two county-population-weighted means of the age-adjusted premature death rate per 100,000, bootstrap interval over counties | counties with both a white and a Black life expectancy in County Health Rankings that also report both premature death rates | deaths before age 75 in 2019 to 2021 (County Health Rankings 2024 release) | 95% CI 1.5 to 1.6 | 1,242 counties | county total population | County Health Rankings |   |
| pop_lad | 144.8 million | people | sum of 2010 census tract population | residents of the tracts in the same-tract comparison | 2010 census |   |   | none (full count) | Opportunity Atlas |   |
| pop_lad_share | 46.9% | percent of the 2010 U.S. population | population of the same-tract comparison tracts as a share of the U.S. population | residents of the United States | 2010 census |   |   | none (full count) | Opportunity Atlas |   |
| prem | 13.6 | percentile points of national household income rank in adulthood, 75th-percentile parents minus 25th-percentile parents | mean over tracts of the 75th minus the 25th percentile family rank (the born-rich premium), population-weighted mean over tracts (2010 tract population as the weight), county-bootstrap interval | children raised in families at the 25th and the 75th percentile of the national income distribution, in tracts where the Atlas estimates both | children born 1978 to 1983, household income measured in 2014 to 2015 | 95% CI 13.2 to 13.9 | 71,924 tracts | 2010 census tract population | Opportunity Atlas |   |
| prem_bot_val | 8.0 | percentile points of national household income rank in adulthood, 75th-percentile parents minus 25th-percentile parents | population-weighted mean of the tract premium within the county | the county with the smallest premium, among counties with at least 100 tracts | children born 1978 to 1983, household income measured in 2014 to 2015 |   | 328 tracts | 2010 census tract population | Opportunity Atlas |   |
| prem_luzerne | 16.9 | percentile points of national household income rank in adulthood, 75th-percentile parents minus 25th-percentile parents | population-weighted mean of the tract premium within the county | Luzerne County, Pennsylvania, among counties with at least 100 tracts | children born 1978 to 1983, household income measured in 2014 to 2015 |   | 103 tracts | 2010 census tract population | Opportunity Atlas |   |
| prem_p5 | 5.8 | percentile points of national household income rank in adulthood, 75th-percentile parents minus 25th-percentile parents | population-weighted 5th percentile of the tract premium | children raised in families at the 25th and the 75th percentile of the national income distribution, in tracts where the Atlas estimates both | children born 1978 to 1983, household income measured in 2014 to 2015 |   | 71,924 tracts | 2010 census tract population | Opportunity Atlas | Tract estimates carry sampling and privacy noise, which widens this spread. |
| prem_p95 | 21.3 | percentile points of national household income rank in adulthood, 75th-percentile parents minus 25th-percentile parents | population-weighted 95th percentile of the tract premium | children raised in families at the 25th and the 75th percentile of the national income distribution, in tracts where the Atlas estimates both | children born 1978 to 1983, household income measured in 2014 to 2015 |   | 71,924 tracts | 2010 census tract population | Opportunity Atlas | Tract estimates carry sampling and privacy noise, which widens this spread. |
| prem_sf | 8.9 | percentile points of national household income rank in adulthood, 75th-percentile parents minus 25th-percentile parents | population-weighted mean of the tract premium within the county | San Francisco County, California, among counties with at least 100 tracts | children born 1978 to 1983, household income measured in 2014 to 2015 |   | 193 tracts | 2010 census tract population | Opportunity Atlas |   |
| prem_top_val | 17.3 | percentile points of national household income rank in adulthood, 75th-percentile parents minus 25th-percentile parents | population-weighted mean of the tract premium within the county | the county with the largest premium, among counties with at least 100 tracts | children born 1978 to 1983, household income measured in 2014 to 2015 |   | 222 tracts | 2010 census tract population | Opportunity Atlas |   |
| premature_age | 75 | years of age | age limit in the definition of premature death (deaths before this age) | deaths counted as premature in County Health Rankings | deaths in 2019 to 2021 (County Health Rankings 2024 release) |   |   | none (definition) | County Health Rankings |   |
| rev_median_pop | 59,039 | people | median of county total population | residents of the counties where the Black estimate is the higher | county population in County Health Rankings 2024 |   | 140 counties | none (median over counties) | County Health Rankings |   |
| seg_beta_fe | 0.06 | years of life-expectancy gap per point of the dissimilarity index | slope of the gap on the index in a county-population-weighted regression with state fixed effects, errors clustered on state | counties with both a white and a Black life expectancy in County Health Rankings that also report the Black-white residential segregation index | index from ACS 2018 to 2022; deaths in 2019 to 2021 | 95% CI 0.00 to 0.12 | 1,260 counties | county total population | County Health Rankings |   |
| seg_n | 1,260 | counties | count of counties in the segregation comparison | counties with both a white and a Black life expectancy in County Health Rankings that also report the Black-white residential segregation index | index from ACS 2018 to 2022; deaths in 2019 to 2021 |   |   | none (full count) | County Health Rankings |   |
| seg_r | +0.30 | correlation coefficient | population-weighted correlation of the white minus Black life-expectancy gap with the index of dissimilarity, bootstrap interval over counties | counties with both a white and a Black life expectancy in County Health Rankings that also report the Black-white residential segregation index | index from ACS 2018 to 2022; deaths in 2019 to 2021 | 95% CI 0.14 to 0.45 | 1,260 counties | county total population | County Health Rankings |   |
| sf_p25 | 50.3 | percentile points of national household income rank in adulthood | population-weighted mean adult household income rank over the county's tracts | children raised at the 25th percentile of the national income distribution in San Francisco County | children born 1978 to 1983, household income measured in 2014 to 2015 |   | 193 tracts | 2010 census tract population | Opportunity Atlas |   |
| sf_p75 | 59.2 | percentile points of national household income rank in adulthood | population-weighted mean adult household income rank over the county's tracts | children raised at the 75th percentile of the national income distribution in San Francisco County | children born 1978 to 1983, household income measured in 2014 to 2015 |   | 193 tracts | 2010 census tract population | Opportunity Atlas |   |
| slope_bw | 0.60 | percentile points of Black children's rank per point of white children's rank | slope of the county mean Black rank on the county mean white rank, weighted by 2010 population summed to the county, state-bootstrap interval | children raised in low-income families (parents at the 25th percentile of the national income distribution) in counties with at least 20 tracts that have both estimates | children born 1978 to 1983, household income measured in 2014 to 2015 | 95% CI 0.49 to 0.71 | 336 counties | 2010 census tract population | Opportunity Atlas |   |
| slope_bw_tract | 0.33 | percentile points of Black children's rank per point of white children's rank | slope of the tract's Black rank on its white rank, population-weighted, county-bootstrap interval | children raised in low-income families (parents at the 25th percentile of the national income distribution) in tracts where the Atlas estimates both white and Black children | children born 1978 to 1983, household income measured in 2014 to 2015 | 95% CI 0.30 to 0.36 | 30,997 tracts | 2010 census tract population | Opportunity Atlas | Sampling noise in the white rank pulls this slope toward zero by an amount the staged data cannot size; the county slope averages that noise away. |
| support_r | −0.25 | correlation coefficient | population-weighted correlation of the county support ratio with the county mobility rank, bootstrap interval over counties | counties with a Social Capital Atlas measure and an Opportunity Atlas mobility rank | friendships among Facebook users in 2022; mobility for children born 1978 to 1983 | 95% CI −0.33 to −0.17 | 2,983 counties | 2018 county population (Social Capital Atlas) | Social Capital Atlas |   |
| vol_r | +0.19 | correlation coefficient | population-weighted correlation of the county volunteering rate with the county mobility rank, bootstrap interval over counties | counties with a Social Capital Atlas measure and an Opportunity Atlas mobility rank | friendships among Facebook users in 2022; mobility for children born 1978 to 1983 | 95% CI 0.12 to 0.25 | 2,983 counties | 2018 county population (Social Capital Atlas) | Social Capital Atlas |   |
| worst_white | 74.4 | years of life expectancy at birth | 10th percentile of county white life expectancy, weighted by county population | white residents of the counties with both a white and a Black life expectancy in County Health Rankings | deaths in 2019 to 2021 (County Health Rankings 2024 release) |   | 1,265 counties | county total population | County Health Rankings |   |
| zip_r | +0.82 | correlation coefficient | population-weighted correlation of ZIP economic connectedness with the ZIP's percentile rank of median household income, bootstrap interval over counties | ZIP code areas (ZCTAs) with an economic connectedness value and a median household income | friendships among Facebook users in 2022; income ACS 2019 to 2023 | 95% CI 0.81 to 0.83 | 18,896 ZIP code areas | 2018 ZIP population (Social Capital Atlas) | Social Capital Atlas |   |

## Definitions

- Income rank (kfr): The position that adult children reach in the national distribution of household income, from the lowest rank to the highest, averaged over children who grew up in a place with parents at a stated income level. The Opportunity Atlas labels this measure kfr.
- Census tract: A small area drawn by the Census Bureau to stand in for a neighborhood. Boundaries are redrawn after each decennial census.
- Economic connectedness: A measure of friendship across class lines, based on the share of high-status friends among people below the median in socioeconomic status. It is scaled so that friendships mixing in proportion to the population give the reference value.
- ZCTA: ZIP Code Tabulation Area, the Census Bureau's approximation of a ZIP code as a set of census blocks. It lets census data be reported by ZIP code, though some ZIP codes have no matching ZCTA.
- Population-weighted: Averaged so that each place counts in proportion to its number of residents, which lets a large county move the result more than a small one. An unweighted average counts every place equally.
- Fixed effects: A regression adjustment that gives each group, such as each county, its own baseline. The estimate then rests on differences within groups.
- Ecological inference: Drawing conclusions about individuals from data about places or groups. A pattern across counties need not hold for the people inside them.

## Methods

The tract comparisons use Opportunity Atlas estimates on native 2010 tracts for children born in 1978 to 1983 and observed in 2014 to 2015, and a tract enters a comparison only when both estimates exist. Atlas privacy noise is not clipped, and tracts are weighted by their 2010 census population because the Atlas has no child counts. The county comparisons of life expectancy by race use County Health Rankings estimates for deaths in 2019 to 2021, for counties with both a non-Hispanic white and a non-Hispanic Black value, so the mobility outcomes and the life-expectancy outcomes refer to different periods. The connectedness regression drops Connecticut.

### Which tracts and children the Atlas comparisons cover

Tract comparisons use the Opportunity Atlas on native 2010 tracts. Ranks are percentiles of the national household income distribution for children born in 1978 to 1983, observed in 2014 to 2015; race-specific ranks are staged for parents at the 25th percentile with genders pooled, and the 75th-percentile rank only with races pooled. Atlas privacy noise is not clipped.

A tract enters a comparison when both estimates exist and its 2010 population is positive.

Explains Table 1, Table 2, Table 3 and Table 4.

### How life expectancy by race and the variance split are defined

Life expectancy by race is the County Health Rankings 2024 estimate, computed from NCHS death counts for deaths in 2019 to 2021, for counties with both a non-Hispanic white and a non-Hispanic Black value; the white-Black gap within counties, weighted and unweighted, comes from the book's shared routine.

The variance split treats the white and Black life expectancy of each county as two cells, gives both the same weight (county population, or one), and divides the weighted variance of the cells into the variance of county averages (place), the squared difference between the group means (race) and the remainder, which equals the weighted variance of the county gaps divided by four (interaction); the shares sum to 100.

Explains Table 6 and Table 7.

### How tracts and counties are weighted and joined

Each tract is weighted by its 2010 census population, read from the USDA Food Access Research Atlas file, which carries the 2010 census count of every 2010 tract (the Atlas has no child counts); counting each tract once gives similar results, and the share of tracts in which white children rank higher counts each tract once.

County mobility ranks, the Social Capital Atlas measures and ACS controls are joined on county codes; correlations are weighted by the Atlas's 2018 population, and the regression is a weighted fit of standardized variables with errors clustered on state (Connecticut drops out because the ACS tract file uses planning-region codes).

Explains Table 2, Table 3, Table 4 and Table 5.

### How the gap across the white rank is estimated

The contrast of the white-Black gap across the white rank is estimated at county grain, on counties with at least 20 tracts that have both estimates, because sorting tracts on a noisy rank builds the contrast in; the county slopes of the premium use counties with at least 10 and at least 100 tracts, with state-bootstrap intervals.

Explains Table 2 and Table 4.

### How uncertainty is measured

Intervals are 95% percentile intervals from 2,000 draws of whole counties with replacement (seed 12345), computed by the same routine for means, shares, slopes and correlations; state-level quantities use states as the resampling unit. Scatter plots show a random sample of tracts inside the plotted range drawn with the same seed; every estimate uses all tracts.

Intervals on the shares resample counties. Counties with a missing estimate are not drawn and are not treated as zero.

## What these data cannot show

The patterns here describe tracts, ZIP code areas and counties. The average life expectancy of Black residents of Cook County says little about how long any one Black resident of Cook County will live, because the data are averages for places and groups (ecological inference).

A same-tract comparison removes the differences between neighborhoods but not all differences. Within a tract, families at one income percentile can still differ in wealth, in the exact block, in schooling and in how institutions treat them. The gap is a residual, and the data describe it without explaining it. Chetty, Hendren, Jones and Porter find that the gap in individual adult income between Black and white children raised at the same parental income is concentrated among boys and men. They find it smaller in tracts with lower poverty and a higher share of Black fathers present, a comparison across places and not an experiment.[2] The ranks here pool boys and girls and measure household income, which includes a spouse's income, because race by gender is not available in the staged data (see Chapter 4.4, Boys and mobility (https://vishalsingh.org/zipcode/boys-and-mobility; planned)).

The Atlas estimates are for children born in 1978 to 1983 and observed in 2014 to 2015. Each tract estimate rests on a modest number of children and carries privacy noise, which is not clipped. Tracts weigh in by their 2010 population because the file has no child counts, and the tracts with both estimates are those with enough children of both groups.

The life-expectancy estimates are computed from death counts that are small in small counties, so the share of counties with a reversed gap and the shares in the decomposition include sampling noise. County Health Rankings counts Black and white residents who are not Hispanic, and the index of dissimilarity is measured for 2018 to 2022. The connectedness measures come from active Facebook users and describe places in 2022, and the regression cannot separate connectedness from income more finely than its interval allows. Puerto Rico has Atlas estimates but is not drawn on the maps, and Connecticut is missing from the regression because its tract codes differ across the files.

## References

1. Chetty, Raj; Friedman, John N.; Hendren, Nathaniel; Jones, Maggie R.; Porter, Sonya R. (2018). The Opportunity Atlas: Mapping the Childhood Roots of Social Mobility. NBER Working Paper 25147. https://doi.org/10.3386/w25147
2. Chetty, Raj; Hendren, Nathaniel; Jones, Maggie R.; Porter, Sonya R. (2020). Race and Economic Opportunity in the United States: An Intergenerational Perspective. The Quarterly Journal of Economics 135(2): 711-783. https://doi.org/10.1093/qje/qjz042
3. Chetty, Raj; Jackson, Matthew O.; Kuchler, Theresa; Stroebel, Johannes; et al. (2022). Social capital I: measurement and associations with economic mobility. Nature 608: 108-121. https://doi.org/10.1038/s41586-022-04996-4
4. University of Wisconsin Population Health Institute (2024). County Health Rankings & Roadmaps Technical Documentation, Version 2. University of Wisconsin Population Health Institute (updated March 18, 2024). https://www.countyhealthrankings.org/sites/default/files/media/document/2024%20CHRR%20Technical%20Document_1.pdf
5. Chetty, Raj; Hendren, Nathaniel; Katz, Lawrence F. (2016). The Effects of Exposure to Better Neighborhoods on Children: New Evidence from the Moving to Opportunity Experiment. American Economic Review 106(4): 855-902. https://doi.org/10.1257/aer.20150572
6. Chetty, Raj; Hendren, Nathaniel (2018). The Impacts of Neighborhoods on Intergenerational Mobility I: Childhood Exposure Effects. The Quarterly Journal of Economics 133(3): 1107-1162. https://doi.org/10.1093/qje/qjy007
7. Chetty, Raj; Hendren, Nathaniel (2018). The Impacts of Neighborhoods on Intergenerational Mobility II: County-Level Estimates. The Quarterly Journal of Economics 133(3): 1163-1228. https://doi.org/10.1093/qje/qjy006

## Provenance

- Pipeline version: 0.1.0
- Build script: book2/chapters/race-income-and-mobility/build.py (sha256 a8ec15d89c0374c9cc3e97a83cf5b548ed926ec782cbeac2d38b7cd537e1b4c4)
- Built on: 2026-10-07
- Site commit: c2b1ff3309cb4a10833f884da91ac0057a66ad0a

Inputs:

| File | Size (bytes) | SHA-256, first 1 MB | SHA-256, last 1 MB |
| --- | ---: | --- | --- |
| book2/chapters/race-income-and-mobility/places.py | 11,415 | 01c6e17503dcb1c97b10c655aafe77a7e9bd50e52de96a5bdc4ef5a0b8cb9879 | 01c6e17503dcb1c97b10c655aafe77a7e9bd50e52de96a5bdc4ef5a0b8cb9879 |
| book2/chapters/race-income-and-mobility/racegap.py | 8,577 | 1a4d09aad9a1b031c1e8eca451e5f7277d86261fc508ab91cd853f3a529c95b5 | 1a4d09aad9a1b031c1e8eca451e5f7277d86261fc508ab91cd853f3a529c95b5 |
| book2/chapters/race-income-and-mobility/tracts.py | 12,656 | 510ec8ede8a4374f1efd405be50d933f724f76e96864375865bc9bbae8a6765b | 510ec8ede8a4374f1efd405be50d933f724f76e96864375865bc9bbae8a6765b |
| data/analytic/zcta_atlas.parquet | 674,348,737 | 0abfad485b74b74eb956fdc57c88e504abe10d601a2e99a3f16467469826578c | 0e37a4ce00d8569d49dccd0b2dfdc2c8435934f9c8a606bbca2ca1dad2c9effe |
| data/staged/food_access/food_access_tract2010_2019.parquet | 37,967,838 | c36929f1809e9ac84ac5f1cb62140bd980c315287833c6f281839df4c1619c89 | 05a17b940a79393b0c2ce5c9e9248496c319cba6896b63af3a2df34fe04f63aa |
| data/staged/health/chr_county_2024.parquet | 8,776,933 | 619795f7ecfec8aff10694b649eaee5e421dbeffbe4be47af4161b51ceefa5cd | c69357f1bcf630407bb9716cb445618838999558c019aaaf77d4681b6bf45c56 |
| data/staged/mobility/oa_module1_tract2010.parquet | 14,946,600 | 32583830e03567b41ad3c245ca41aa67fb9cb55a8a46e8d9ae4c77a6cc2fcd83 | 6d67fac56edd8f07346375fb0b97e3c82621d57457d89c7fe93ec6f1b758bfaf |
| data/staged/mobility/oa_outcomes_county.parquet | 721,343 | fd7bc55294fbd3f7a646f27083500c1f12415775db7a3607878b522d277f8140 | fd7bc55294fbd3f7a646f27083500c1f12415775db7a3607878b522d277f8140 |
| data/staged/ses/acs5_tract2020_2019_2023.parquet | 7,232,424 | 5fe6f730a35225f14d5d8abfc6b9bce58b52e545f309a7eac74f1dae74d7a31c | 7b4a2475b3852d3f91804716a04566ef8ceadacec2be9fb85f5eb642078efac6 |
| data/staged/social_capital/sc_atlas_county_2022.parquet | 655,734 | 9ee5596513748a9b62c3e5fc559ad41b804696702c9f92848be04a05222f2f46 | 9ee5596513748a9b62c3e5fc559ad41b804696702c9f92848be04a05222f2f46 |
| data/staged/social_capital/sc_atlas_zip_2022.parquet | 3,110,340 | 7685825cc96df0d404153c288fc722577ad9ee08322eca981a2832f69369856a | b856fceb0423edeaefb67ea1812dc2d45d35591d792e3aba1bcc7e92c01af828 |

Outputs:

| File | SHA-256 |
| --- | --- |
| facts.json | 78f6e8841f063c454c4a54f624a06a916d204af05d3a711bcd457b0043f68ca0 |
| figures/born_rich_premium.json | 0f434894652c0e7c717f4c43a5316c4ff16915bdf9588716cf72e704c8815111 |
| figures/connectedness_horse_race.json | e19bb1d0134a2fc27148981433520f4d91e96921c71bd8f6ee0722eabc03697b |
| figures/group_gaps_same_tract.json | 9d3bda02a4d76da0881bad5df30660ed4404da03e405eca3b3c752cefc392658 |
| figures/le_black_white_counties.json | 075b15deec50eddbbd75b6ea1fe4f21556f35da33abbc1585d62ce87b5221250 |
| figures/le_variance_race_place.json | 5a78909d3acf11835314ac9ca7605ac8d30b15e19bf63867ac0be3d2f7934af0 |
| figures/mobility_map.json | 3fad1155815b20a87c60b548fb4eef49f59d400b0b630bf87f443957acedd741 |
| figures/same_tract_black_white.json | 4ab08361576c2d94c21c91580e3e23d40f3fcd651fe1ccba6d9e154c7283c3df |
| meta.json | 9963563b1cb8d9902e964ccf8190715b67348f8b988f2576656541ec774f06cf |
| public/data/race-income-and-mobility/born_rich_premium.json | 62760e10bc85a68a5a27bd8c8ef8769e92d1bc68ded23a9cc4314c63b05d458a |
| public/data/race-income-and-mobility/mobility_map.json | a040e912e07617cff3bc997fee34ca7310dd83ab7d61ddfb827ad054ce433de7 |
| public/data/race-income-and-mobility/same_tract_black_white.json | 80edb0b40b7c97260c29607c2a4c916ec42826021b7c455b6f474858f2f4e423 |
| refs.json | 52c81dac44ace635a783210962a409d0d152a5168429e5b8da06018c4ce74a7d |

## Machine formats

- Report (HTML): https://vishalsingh.org/zipcode/race-income-and-mobility/report
- Report (Markdown): https://vishalsingh.org/zipcode/race-income-and-mobility/report.md
- Every estimate (JSON): https://vishalsingh.org/zipcode/report/race-income-and-mobility/facts.json
- Table 1, figure "mobility_map" (JSON): https://vishalsingh.org/zipcode/report/race-income-and-mobility/figures/mobility_map.json
- Table 2, figure "same_tract_black_white" (JSON): https://vishalsingh.org/zipcode/report/race-income-and-mobility/figures/same_tract_black_white.json
- Table 3, figure "group_gaps_same_tract" (JSON): https://vishalsingh.org/zipcode/report/race-income-and-mobility/figures/group_gaps_same_tract.json
- Table 4, figure "born_rich_premium" (JSON): https://vishalsingh.org/zipcode/report/race-income-and-mobility/figures/born_rich_premium.json
- Table 5, figure "connectedness_horse_race" (JSON): https://vishalsingh.org/zipcode/report/race-income-and-mobility/figures/connectedness_horse_race.json
- Table 6, figure "le_black_white_counties" (JSON): https://vishalsingh.org/zipcode/report/race-income-and-mobility/figures/le_black_white_counties.json
- Table 7, figure "le_variance_race_place" (JSON): https://vishalsingh.org/zipcode/report/race-income-and-mobility/figures/le_variance_race_place.json
- Payload of "mobility_map" (JSON): https://vishalsingh.org/zipcode/data/race-income-and-mobility/mobility_map.json
- Payload of "same_tract_black_white" (JSON): https://vishalsingh.org/zipcode/data/race-income-and-mobility/same_tract_black_white.json
- Payload of "born_rich_premium" (JSON): https://vishalsingh.org/zipcode/data/race-income-and-mobility/born_rich_premium.json
- Index of every report: https://vishalsingh.org/zipcode/llms.txt; all reports in one file: https://vishalsingh.org/zipcode/llms-full.txt

---

# Mental distress in young adults: report

Chapter 4.5 of ZIP Code Destiny, by Vishal Singh (NYU Stern). Generated from the chapter's data files; the reading version is at https://vishalsingh.org/zipcode/mental-distress-in-young-adults. Table n of this report is figure n of the chapter.

## Identification

- Title: Mental distress in young adults
- Label: The young and the anxious
- Chapter: 4.5 (Part IV, Becoming an adult; An American life)
- Question: How has frequent mental distress among young adults changed since 2014, and does it differ by income and place?
- Dek: Frequent mental distress among adults 18 to 24 rose from 12.3% in 2014 to 25.3% in 2022 and was 21.8% in 2024, in a survey that draws new adults each year.
- Grain: individuals
- Status: published
- Published: 2026-10-09
- Updated: not revised
- Data sources:
  - BRFSS. BRFSS public-use microdata, 2014-2024: days of poor mental health (core item, all years), ever told of a depressive disorder (core item, all years), age, sex, household income, education and metro status; adults in the 50 states and DC; years: 2014-2024; weights: design-based: final weight, stratum and primary sampling unit
- Canonical URL: https://vishalsingh.org/zipcode/mental-distress-in-young-adults
- Report: https://vishalsingh.org/zipcode/mental-distress-in-young-adults/report
- Markdown: https://vishalsingh.org/zipcode/mental-distress-in-young-adults/report.md
- BibTeX key: singh2026mentaldistressinyoungadults

```bibtex
@misc{singh2026mentaldistressinyoungadults,
  author = {Singh, Vishal},
  title = {Mental distress in young adults},
  year = {2026},
  url = {https://vishalsingh.org/zipcode/mental-distress-in-young-adults},
  note = {In ZIP Code Destiny. Published 9 October 2026.}
}
```

## Principal estimates

The facts the dek and the opening prose cite, in order of appearance.

| Estimand | Population | Period | Value | Interval | n | Weights | Source | Fact id |
| --- | --- | --- | ---: | --- | ---: | --- | --- | --- |
| definition used in this chapter, not an estimate | an age group of respondents defined in this chapter | 2014 to 2024 | 18 to 24 |   |   | none (definition) | BRFSS | band_young |
| design-based weighted share of respondents, logit interval from Taylor linearization | adults 18 to 24 in the 50 states and DC | 2014 | 12.3% | 95% CI 11.6% to 13.0% | 23,152 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS | fmd_young_2014 |
| design-based weighted share of respondents, logit interval from Taylor linearization | adults 18 to 24 in the 50 states and DC | 2022 | 25.3% | 95% CI 24.4% to 26.2% | 25,815 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS | fmd_young_2022 |
| design-based weighted share of respondents, logit interval from Taylor linearization | adults 18 to 24 in the 50 states and DC | 2024 | 21.8% | 95% CI 20.9% to 22.7% | 28,711 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS | fmd_young_2024 |
| definition used in this chapter, not an estimate | cutoff for frequent mental distress | 2014 to 2024 | 14 |   |   | none (definition) | BRFSS | fmd_days |
| definition used in this chapter, not an estimate | the BRFSS mental health question | 2014 to 2024 | 30 |   |   | none (definition) | BRFSS | recall_days |
| definition used in this chapter, not an estimate | an age group of respondents defined in this chapter | 2014 to 2024 | 50 to 64 |   |   | none (definition) | BRFSS | band_mid |
| design-based difference in weighted shares between two groups of one survey, interval from Taylor linearization, adults 18 to 24 minus adults 50 to 64 | adults 18 to 24 and adults 50 to 64 in the 50 states and DC | 2014 | −0.2 | 95% CI −1.0 to 0.5 | 164,132 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS | gap_ym_2014 |
| design-based difference in weighted shares between two groups of one survey, interval from Taylor linearization, adults 18 to 24 minus adults 50 to 64 | adults 18 to 24 and adults 50 to 64 in the 50 states and DC | 2022 | +11.6 | 95% CI 10.6 to 12.6 | 136,149 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS | gap_ym_2022 |
| design-based difference in weighted shares between two groups of one survey, interval from Taylor linearization, adults 18 to 24 minus adults 50 to 64 | adults 18 to 24 and adults 50 to 64 in the 50 states and DC | 2024 | +7.4 | 95% CI 6.4 to 8.4 | 134,822 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS | gap_ym_2024 |

## Findings

The share of adults 18 to 24 who reported frequent mental distress rose from 12.3% (95% CI 11.6% to 13.0%) in 2014 to 25.3% (95% CI 24.4% to 26.2%) in 2022 in the Behavioral Risk Factor Surveillance System (BRFSS), an annual telephone survey run by the CDC with state health departments. Frequent mental distress means 14 or more of the past 30 days on which the respondent's mental health was not good. In 2014 the rate for these adults differed from that for adults 50 to 64 by −0.2 (95% CI −1.0 to 0.5) points, an interval that includes zero. In 2022 the difference was +11.6 (95% CI 10.6 to 12.6) points. By 2024 the rate for adults 18 to 24 had fallen to 21.8% (95% CI 20.9% to 22.7%), and the difference from adults 50 to 64 was +7.4 (95% CI 6.4 to 8.4) points.

The chapter describes respondents, and every rate is an estimate for adults living in a state or the District of Columbia. The BRFSS draws a new sample each year, so it is a repeated cross-section and not a panel. The youngest adults of 2022 are not the people interviewed in 2014 at an earlier age. Rates of frequent mental distress for places, which come from modeled estimates for census tracts, belong to Chapter 7.1, Health behaviors and the supply of care (https://vishalsingh.org/zipcode/health-behaviors-and-the-supply-of-care). How people rate their lives and describe their days is in Chapter 6.3, Life evaluation and daily feelings (https://vishalsingh.org/zipcode/life-evaluation-and-daily-feelings), and how well-being changes over the life span is in Chapter 6.4, Well-being across the life span (https://vishalsingh.org/zipcode/well-being-across-the-life-span).

### What the survey asks and how it is weighted

Respondents are asked how many of the past 30 days their mental health, which includes stress, depression and problems with emotions, was not good.[1] An answer of none counts as zero days, and a respondent who does not know or refuses is left out (1.8% of respondents). The cutoff of 14 or more days is the CDC's definition of frequent mental distress.[2] The measure is what a person says about the past month. It is not a diagnosis, and a rise in it can reflect more distress, more willingness to describe it, or both. A second item asks whether a doctor, nurse or other health professional has ever told the respondent of a depressive disorder.[1] The file has answers to both items in every year from 2014 to 2024.

The file holds 4,874,919 respondent records for 2014 to 2024, 4,790,742 of them in a state or the District of Columbia. Guam, Puerto Rico and the Virgin Islands are left out, as in Chapter 7.2, Health behaviors, person by person (https://vishalsingh.org/zipcode/health-behaviors-person-by-person), so that the two BRFSS chapters describe the same adults. 4,626,736 respondents have both a valid answer to the distress question and an age group.

Every estimate is design-based. It uses the survey weights (the final weight), the strata and the sampling units of the public file, with variances from Taylor linearization,[3] and a subgroup is treated as part of the whole sample so that no cell loses the design it belongs to. Intervals are 95% intervals. The weights spread the sample unevenly. The 79,822 respondents 18 to 24 in 2022 to 2024 carry the information of 24,390 respondents with equal weights, and the intervals reflect that loss. A difference between two groups in one period accounts for the groups sharing one design. A difference between two periods combines two independent estimates, because each survey year is drawn separately.

Not every state is present in every year. In this file New Jersey is missing in 2019, Florida in 2021, and Kentucky and Pennsylvania in 2023, and CDC reports that Tennessee could not collect enough data to be included in the 2024 public file.[4] Limiting every estimate to the 46 jurisdictions (states and the District of Columbia) present in all years moves no age-by-year estimate by more than 0.5 points, so the national series is not an artifact of which states are missing. The BRFSS has used raking and included cell phone respondents since 2011, and every year in this file follows that method, so the series does not cross the 2011 change in weighting.[4]

### Frequent distress was level across adult ages in 2014 and now falls with age

Table 1 plots the rate for six age groups. In 2014 the four younger groups lay within 0.3 points of one another, from adults 18 to 24 to adults 50 to 64, while adults 65 to 74 and 75 and older were lower. By 2024 the four were spread out, from 21.8% for adults 18 to 24 to 14.4% for adults 50 to 64.

Table 1. Frequent mental distress among adults 18 to 24 rose from 12.3% in 2014 to 25.3% in 2022, while the rate for adults 50 to 64 moved from 12.5% to 13.8%

Every group's rate was higher in 2024 than in 2014, and the rise was larger for the younger groups. It was 9.5 (95% CI 8.4 to 10.6) points for adults 18 to 24 and 7.7 (95% CI 6.8 to 8.6) for adults 25 to 34, against 1.9 (95% CI 1.3 to 2.5) for adults 50 to 64. For adults 75 and older it was 1.0 (95% CI 0.3 to 1.6), from 6.5% (95% CI 6.1% to 7.0%) to 7.5% (95% CI 7.1% to 8.0%).

Taken together, adults went from 11.4% (95% CI 11.2% to 11.6%) in 2014 to 15.8% (95% CI 15.5% to 16.0%) in 2022 and 15.6% (95% CI 15.3% to 15.8%) in 2024.

The path for the youngest group is not smooth. Its rate peaked in 2022 and was 3.5 (95% CI 2.3 to 4.8) points lower in 2024 than in 2022, while the rate for adults 35 to 49 was 0.9 (95% CI 0.2 to 1.6) points higher. In 2020, the first pandemic year, the rate for adults 18 to 24 was 1.5 (95% CI 0.2 to 2.8) points lower than in 2019, and no other age group has a lower rate in 2020 by an amount the survey can distinguish from zero. The data do not say why 2020 differs. Interviews that year took place under unusual conditions, and a fall in a self-reported measure could come from who answered as well as from how respondents felt.

Table 2 tests the comparison directly, as the difference between the rate for adults 18 to 24 and the rate for adults 50 to 64 in each year. The rate for the older group went from 12.5% in 2014 to 13.8% in 2022.

The difference was −0.2 (95% CI −1.0 to 0.5) points in 2014. The interval first lies above zero in 2015, at +0.9 (95% CI 0.1 to 1.7) points, where it only just clears zero, and it stays above zero in every later year. The difference reached +11.6 (95% CI 10.6 to 12.6) points in 2022 and narrowed to +7.4 (95% CI 6.4 to 8.4) in 2024.

Table 2. The difference in frequent distress between adults 18 to 24 and adults 50 to 64 went from −0.2 points in 2014 to +11.6 points in 2022

Table 3 shows the same change as a profile across age. In 2014 to 2016 the rate in every five-year band from the youngest through the late fifties lay between 11.7% and 13.3%, and it fell only in the sixties, to 6.4% (95% CI 6.1% to 6.7%) in the oldest band.

In 2022 to 2024 the rate was 23.4% (95% CI 22.9% to 23.9%) in the youngest band, against 13.3% (95% CI 12.9% to 13.7%) in 2014 to 2016. It then falls with age, with a flat stretch through the fifties near 14.7% (95% CI 14.2% to 15.2%), to 7.1% (95% CI 6.7% to 7.5%) in the oldest band, a small distance above that band's earlier level. The profile for 2017 to 2019 lies between the other two in every band.

Table 3. Frequent distress was close to flat from the youngest band to the late fifties in 2014 to 2016 and falls with age in 2022 to 2024

Two other studies bear on this. Blanchflower, Bryson and Xu report, from repeated cross-sections for the United States and the United Kingdom and from a survey of many countries, that ill-being used to peak in midlife and now declines steadily with age. Their US analysis uses the BRFSS itself, so for the United States it repeats this chapter's source and is not an independent check.[5] Twenge and colleagues used the National Survey on Drug Use and Health, a different survey and measure, and found that the share of young adults with a major depressive episode in the past year rose between 2009 and 2017. Their models attributed the adult trends mainly to birth cohort, and the study does not describe an age profile like the one above.[6] A repeated cross-section like the BRFSS cannot separate age, survey year and birth cohort without extra assumptions, and this chapter makes none. It reports how the age profile of one self-reported measure looked in each period.

### Women report more frequent distress than men at every age

Table 4 shows the rate for women minus the rate for men in two periods. The difference lies above zero in every age group in both periods. The difference rose most among adults 18 to 24. In 2014 to 2016 women in that group reported 16.2% and men 10.6%, a difference of 5.5 (95% CI 4.7 to 6.3) points.

In 2022 to 2024 the rates for adults in that group were 29.0% and 18.1%, a difference of 10.8 (95% CI 9.8 to 11.9) points, so the difference itself rose by 5.3 (95% CI 4.0 to 6.7) points.

Among adults 25 to 34 the difference rose by 2.7 (95% CI 1.7 to 3.7) points. Among adults 50 to 64 it went from 4.2 (95% CI 3.8 to 4.6) to 5.2 (95% CI 4.7 to 5.8), a rise of 1.0 (95% CI 0.4 to 1.7) points.

Table 4. Women reported frequent distress more often than men in every age group, and the gap among adults 18 to 24 rose from 5.5 to 10.8 points

The survey's sex item changed over these years. In the early questionnaires sex could be filled in at screening or recorded by the interviewer, later questionnaires ask whether the respondent is male or female at screening, and the latest add a question on sex at birth. The chapter calls the groups women and men as the survey codes them, so the change in a gap between 2014 to 2016 and 2022 to 2024 compares groups defined slightly differently.

The distress item records what respondents say, and the data do not show whether the difference between women and men reflects different experience, different willingness to report, or both. They show that the difference exists at every age and that it grew fastest where the overall rise was largest.

### The rise among the youngest adults reached every sex, income and education group

Among adults 18 to 24 as a whole, the rate went from 13.3% (95% CI 12.9% to 13.7%) in 2014 to 2016 to 23.4% (95% CI 22.9% to 23.9%) in 2022 to 2024, a rise of 10.1 (95% CI 9.4 to 10.8) points. Table 5 splits that rise by sex, household income, education and metro status. The rise has an interval above zero in each group defined by sex, household income or education.

Table 5. Frequent distress among adults 18 to 24 rose from 13.3% to 23.4%, and rose in every group by sex, household income and education

Women rose by 12.8 (95% CI 11.8 to 13.9) points and men by 7.5 (95% CI 6.7 to 8.3) points, and the rise for women exceeded the rise for men by 5.3 (95% CI 4.0 to 6.7) points.

By household income the second bracket rose by 15.1 (95% CI 13.5 to 16.8) points and the lowest by 8.8 (95% CI 6.8 to 10.9), and the second bracket's rise exceeded the lowest bracket's by 6.3 (95% CI 3.7 to 9.0) points. The top bracket, $75,000 or more, rose by 9.5 (95% CI 8.0 to 11.0) points, and its rise differs from the lowest bracket's by 0.7 (95% CI −1.8 to 3.2) points, an interval that includes zero. The differences among the other brackets are not tested here.

In 2022 to 2024, young adults in households at the top bracket reported 21.1% (95% CI 20.0% to 22.2%), a higher rate than young adults in households below $20,000 had reported in 2014 to 2016, 16.9% (95% CI 16.0% to 17.9%). Household income counts everyone in the household, so for an adult living with parents it describes the parents' means more than the respondent's own.

By education, college graduates went from 8.9% (95% CI 8.3% to 9.7%) to 18.6% (95% CI 17.6% to 19.6%), a rise of 9.7 (95% CI 8.4 to 10.9) points. Respondents who had not finished high school went from 17.3% (95% CI 15.9% to 18.8%) to 23.8% (95% CI 22.0% to 25.7%), a rise of 6.5 (95% CI 4.2 to 8.9) points. Education is the highest level completed, and many adults 18 to 24 are still in school, so these groups describe where respondents are in their schooling as much as their attainment.

The survey cannot say much about metro status. The BRFSS codes it for only part of the respondents, and the share falls from 38.7% of adults in 2014 to 2016 to 11.4% in 2022 to 2024.

The recent metro-status groups of adults 18 to 24 hold 103 to 384 respondents each, and the intervals for their rises are between 12.6 and 31.9 points wide. The interval lies above zero for one of the four.

The respondents with a metro code had a rate of 20.6% (95% CI 16.2% to 25.9%) in 2022 to 2024, and the interval includes the 23.4% for all adults 18 to 24. The data therefore show no clear difference between them and the rest, though the interval is too wide to rule one out. The data support no ranking of center cities, suburbs and areas outside metro areas, and they do not rule out differences among them.

> Causal evidence: Social media and the mental health of college students (Staggered difference-in-differences across colleges (natural experiment))
>
> The BRFSS comparisons above have no before and after for any respondent and no comparison group that changed for a known reason, so they cannot say why distress rose. One study that does have such a design is by Braghieri, Levy and Makarin. Facebook was introduced at US colleges in stages between 2004 and 2006, so students at colleges that received it earlier can be compared before and after with students at colleges that received it later. Using survey data on student mental health, the authors found that the introduction of Facebook at a college worsened student mental health and raised the likelihood that students reported poor mental health impairing their academic performance.[7]
>
> The result applies to college students in the first years of one platform. It does not cover the platforms, the age groups, the years or the pandemic that mark the 2014 to 2024 changes above, and it does not test the BRFSS rise.

### Household income separates middle-aged adults more than the youngest adults

Table 6 crosses age group with household income in 2022 to 2024. Among adults 50 to 64, 31.7% (95% CI 30.5% to 32.9%) in the lowest income bracket reported frequent distress against 8.0% (95% CI 7.7% to 8.3%) in the top bracket, a difference of 23.7 (95% CI 22.5 to 24.9) points. Among adults 18 to 24 the rates were 25.7% (95% CI 24.0% to 27.6%) and 21.1% (95% CI 20.0% to 22.2%), a difference of 4.7 (95% CI 2.6 to 6.8) points.

The highest cell in the grid, 32.9% (95% CI 31.3% to 34.5%), is for adults 35 to 49 in the lowest bracket, and its interval overlaps that of adults 50 to 64 in the same bracket, 31.7% (95% CI 30.5% to 32.9%). The cell for the youngest adults in that bracket is lower than both.

Table 6. Frequent distress among adults 18 to 24 was 25.7% in the lowest household income bracket and 21.1% in the top, against 31.7% and 8.0% among adults 50 to 64

The age pattern therefore depends on income. In the top bracket, adults 18 to 24 exceeded adults 50 to 64 by +13.0 (95% CI 11.9 to 14.2) points, and they were also higher in the three brackets between the top and the lowest. In the lowest bracket the sign reversed, at −6.0 (95% CI −8.1 to −3.8) points, so the youngest adults reported less frequent distress than adults 50 to 64 in households of the same bracket.

The difference between the lowest and the top bracket rises across the younger age groups, from 4.7 points for adults 18 to 24 to 11.5 (95% CI 9.7 to 13.2) for adults 25 to 34 and 21.0 (95% CI 19.4 to 22.6) for adults 35 to 49. It is similar for adults 50 to 64 and smaller among adults 65 to 74 and 75 and older.

Across all ages, the lowest bracket reported 26.9% (95% CI 26.3% to 27.6%) and the top bracket 11.1% (95% CI 10.9% to 11.3%). For a respondent who lives with parents, household income describes the parents' means.

In 2022 to 2024 household income is missing for 22.1% of adults and 32.8% of adults 18 to 24. Adults who did not report it had a rate of 14.9% (95% CI 14.6% to 15.2%) in those years, against 15.8% (95% CI 15.7% to 16.0%) for those who did, so leaving them out does not hide a group with higher rates. The brackets are in current dollars, so rising prices move households across them over the period.

### Current distress and a depression diagnosis

A rise in a self-reported item may partly reflect how people answer, which a more concrete item can help to check. Table 7 sets frequent distress beside whether a health professional has ever told the respondent of a depressive disorder. The diagnosis share is higher than the distress share at every age in both periods. In 2022 to 2024 the diagnosis share for adults 18 to 24 was 25.6% (95% CI 25.1% to 26.1%) against 23.4% with frequent distress.

Among respondents who answered both items the difference was 2.4 (95% CI 1.8 to 3.0) points, a little different from the difference of the two shares because each share covers everyone who answered that item. For adults 18 to 24 the difference was 2.2 (95% CI 1.7 to 2.7) points in 2014 to 2016. In 2022 to 2024 it was 2.8 (95% CI 2.3 to 3.3) points in the oldest band and larger in between, for example 7.6 (95% CI 7.1 to 8.1) points at ages 65 to 69, so the two measures are closest at the youngest and oldest ages.

Table 7. In 2022 to 2024 the share of adults 18 to 24 ever told of depression was 25.6%, close to the 23.4% with frequent distress now

The diagnosis share for the youngest adults rose by 10.1 (95% CI 9.4 to 10.7) points between 2014 to 2016 and 2022 to 2024, close to the 10.1 points by which frequent distress rose. The shape across age changed as well. In 2014 to 2016 the diagnosis share rose with age from 15.5% (95% CI 15.1% to 16.0%) among the youngest adults to 21.2% (95% CI 20.8% to 21.6%) at ages 55 to 59, as an accumulating count of lifetime diagnoses would.

In 2022 to 2024 it was 25.6% among the youngest and 20.7% (95% CI 20.2% to 21.2%) at ages 55 to 59, so the youngest adults now have a diagnosis share among the highest of any age band. The difference from adults 25 to 29 is +0.3 (95% CI −0.5 to 1.1) points and from adults 30 to 34 +0.7 (95% CI −0.1 to 1.5) points, neither distinguishable from zero. The youngest adults exceed every band from the late thirties upward.

A diagnosis depends on contact with care, which differs by age, insurance and period, so the second measure is not a cleaner gauge of illness. It does show that the change is not confined to one survey question, because a different item with a different time frame moves in the same direction for the same adults.

### What these data cannot show

The BRFSS is a repeated cross-section. The youngest adults in 2022 are a different set of people from those in 2014, and the data cannot separate age, survey year and birth cohort. They also cannot say whether a fall from 2022 to 2024 will continue.

Frequent mental distress is self-reported. It reflects how respondents experience the past month and how willing they are to say so, and willingness may have changed more among the young than among the old. The BRFSS reaches adults who live in households and answer a telephone call, so it misses people in hospitals, prisons and other institutions and people without a stable phone, and nonresponse can change over time in ways the weights may not fully correct. The state of residence is known and no smaller area is, so nothing here describes a neighborhood or a county.

Several comparisons rest on thin or imperfect measures. Household income is missing for a large part of the youngest adults and describes a household, not the respondent. Education among adults 18 to 24 is unfinished. Metro status is coded for a shrinking part of the sample, and the sex item changed wording over the period. The 2020 and 2021 interviews took place during the pandemic and are shown by year but left out of the three periods. Some states are absent in some years. Intervals describe sampling variation under the survey design and do not carry any bias from nonresponse or measurement.

The chapter makes no causal claim outside the study in the box above. Differences between groups, such as the difference between women and men or between income brackets, are patterns in who reports distress, and the data do not identify what produces them.

## Tables

### Table 1. Frequent mental distress among adults 18 to 24 rose from 12.3% in 2014 to 25.3% in 2022, while the rate for adults 50 to 64 moved from 12.5% to 13.8%

Adults reporting 14 or more of the past 30 days with mental health not good, percent, by age group and survey year, 50 states and DC, 2014 to 2024

| age_group | conf_high | conf_low | n | rate | year |
| --- | ---: | ---: | ---: | ---: | ---: |
| 18 to 24 | 13.02 | 11.63 | 23,152 | 12.31 | 2014 |
| 25 to 34 | 13.14 | 12.02 | 41,901 | 12.57 | 2014 |
| 35 to 49 | 12.73 | 11.87 | 83,788 | 12.29 | 2014 |
| 50 to 64 | 12.89 | 12.19 | 140,980 | 12.53 | 2014 |
| 65 to 74 | 8.46 | 7.66 | 87,891 | 8.05 | 2014 |
| 75 and older | 6.97 | 6.14 | 65,066 | 6.54 | 2014 |
| 18 to 24 | 13.94 | 12.50 | 23,237 | 13.20 | 2015 |
| 25 to 34 | 12.85 | 11.76 | 41,226 | 12.29 | 2015 |
| 35 to 49 | 12.83 | 12.01 | 78,275 | 12.41 | 2015 |
| 50 to 64 | 12.69 | 11.99 | 131,942 | 12.34 | 2015 |
| 65 to 74 | 8.29 | 7.53 | 85,003 | 7.90 | 2015 |
| 75 and older | 7.16 | 6.28 | 62,287 | 6.71 | 2015 |
| 18 to 24 | 15.19 | 13.78 | 25,698 | 14.47 | 2016 |
| 25 to 34 | 13.36 | 12.34 | 46,871 | 12.84 | 2016 |
| 35 to 49 | 12.63 | 11.81 | 84,600 | 12.21 | 2016 |
| 50 to 64 | 12.91 | 12.18 | 141,807 | 12.54 | 2016 |
| 65 to 74 | 8.24 | 7.50 | 96,271 | 7.86 | 2016 |
| 75 and older | 7.03 | 6.11 | 68,252 | 6.56 | 2016 |
| 18 to 24 | 17.36 | 15.73 | 25,372 | 16.53 | 2017 |
| 25 to 34 | 14.13 | 13.06 | 45,868 | 13.59 | 2017 |
| 35 to 49 | 13.43 | 12.50 | 79,509 | 12.96 | 2017 |
| 50 to 64 | 13.20 | 12.40 | 128,729 | 12.79 | 2017 |
| 65 to 74 | 8.99 | 8.11 | 88,940 | 8.54 | 2017 |
| 75 and older | 7.44 | 6.35 | 62,621 | 6.88 | 2017 |
| 18 to 24 | 18.69 | 17.11 | 25,142 | 17.89 | 2018 |
| 25 to 34 | 15.75 | 14.59 | 45,106 | 15.16 | 2018 |
| 35 to 49 | 13.06 | 12.20 | 78,100 | 12.62 | 2018 |
| 50 to 64 | 13.19 | 12.39 | 120,693 | 12.78 | 2018 |
| 65 to 74 | 9.04 | 8.20 | 85,844 | 8.61 | 2018 |
| 75 and older | 7.31 | 6.28 | 61,123 | 6.78 | 2018 |
| 18 to 24 | 21.36 | 19.62 | 24,005 | 20.47 | 2019 |
| 25 to 34 | 17.05 | 15.87 | 41,979 | 16.45 | 2019 |
| 35 to 49 | 14.12 | 13.20 | 71,943 | 13.65 | 2019 |
| 50 to 64 | 13.43 | 12.63 | 110,630 | 13.03 | 2019 |
| 65 to 74 | 9.57 | 8.67 | 83,604 | 9.11 | 2019 |
| 75 and older | 7.15 | 6.32 | 62,675 | 6.72 | 2019 |
| 18 to 24 | 19.92 | 18.08 | 24,611 | 18.98 | 2020 |
| 25 to 34 | 17.37 | 16.05 | 42,467 | 16.70 | 2020 |
| 35 to 49 | 14.85 | 13.80 | 74,669 | 14.31 | 2020 |
| 50 to 64 | 13.29 | 12.29 | 104,594 | 12.78 | 2020 |
| 65 to 74 | 9.27 | 8.29 | 76,936 | 8.77 | 2020 |
| 75 and older | 7.85 | 6.55 | 55,856 | 7.17 | 2020 |
| 18 to 24 | 23.54 | 21.71 | 25,110 | 22.61 | 2021 |
| 25 to 34 | 18.98 | 17.69 | 45,683 | 18.33 | 2021 |
| 35 to 49 | 15.88 | 14.82 | 83,630 | 15.34 | 2021 |
| 50 to 64 | 13.91 | 13.05 | 112,559 | 13.47 | 2021 |
| 65 to 74 | 9.48 | 8.46 | 85,976 | 8.96 | 2021 |
| 75 and older | 8.55 | 7.39 | 61,800 | 7.95 | 2021 |
| 18 to 24 | 26.23 | 24.42 | 25,815 | 25.32 | 2022 |
| 25 to 34 | 20.91 | 19.60 | 45,803 | 20.25 | 2022 |
| 35 to 49 | 16.77 | 15.83 | 83,622 | 16.30 | 2022 |
| 50 to 64 | 14.19 | 13.33 | 110,334 | 13.75 | 2022 |
| 65 to 74 | 10.66 | 9.69 | 87,158 | 10.17 | 2022 |
| 75 and older | 8.75 | 7.66 | 65,683 | 8.19 | 2022 |
| 18 to 24 | 24.19 | 22.32 | 25,296 | 23.24 | 2023 |
| 25 to 34 | 21.39 | 19.98 | 44,320 | 20.67 | 2023 |
| 35 to 49 | 17.38 | 16.27 | 79,779 | 16.82 | 2023 |
| 50 to 64 | 14.30 | 13.37 | 103,393 | 13.83 | 2023 |
| 65 to 74 | 9.99 | 8.98 | 86,587 | 9.47 | 2023 |
| 75 and older | 8.17 | 7.24 | 70,449 | 7.69 | 2023 |
| 18 to 24 | 22.65 | 20.95 | 28,711 | 21.79 | 2024 |
| 25 to 34 | 21 | 19.62 | 48,280 | 20.30 | 2024 |
| 35 to 49 | 17.76 | 16.73 | 85,851 | 17.24 | 2024 |
| 50 to 64 | 14.89 | 13.94 | 106,111 | 14.41 | 2024 |
| 65 to 74 | 10.05 | 9.15 | 89,682 | 9.59 | 2024 |
| 75 and older | 8.01 | 7.05 | 75,841 | 7.52 | 2024 |

Note: Respondents are adults in the 50 states and the District of Columbia. Each survey year is a new sample, not the same people followed over time, and every interval is design-based (final weight, stratum and primary sampling unit).

Note: Some states are missing in some years: New Jersey in 2019, Florida in 2021, Kentucky and Pennsylvania in 2023 and Tennessee in 2024. Restricting every estimate to the 46 jurisdictions present in every year moves no age-by-year estimate by more than 0.5 points.

Note: Interviews in 2020 took place during the first year of the pandemic.

Source: CDC BRFSS, 2014-2024, adults in the 50 states and DC; weighted (BRFSS final weight), design-based intervals

Data: https://vishalsingh.org/zipcode/report/mental-distress-in-young-adults/figures/fmd_by_year_age.json

### Table 2. The difference in frequent distress between adults 18 to 24 and adults 50 to 64 went from −0.2 points in 2014 to +11.6 points in 2022

Rate for adults 18 to 24 minus rate for adults 50 to 64, percentage points, design-based difference by survey year, 50 states and DC, 2014 to 2024

| conf_high | conf_low | gap | n | year |
| ---: | ---: | ---: | ---: | ---: |
| 0.550 | −1 | −0.230 | 164,132 | 2014 |
| 1.67 | 0.070 | 0.870 | 155,179 | 2015 |
| 2.73 | 1.14 | 1.93 | 167,505 | 2016 |
| 4.65 | 2.83 | 3.74 | 154,101 | 2017 |
| 5.99 | 4.22 | 5.11 | 145,835 | 2018 |
| 8.40 | 6.49 | 7.45 | 134,635 | 2019 |
| 7.25 | 5.15 | 6.20 | 129,205 | 2020 |
| 10.15 | 8.13 | 9.14 | 137,669 | 2021 |
| 12.56 | 10.56 | 11.56 | 136,149 | 2022 |
| 10.45 | 8.36 | 9.41 | 128,689 | 2023 |
| 8.36 | 6.41 | 7.38 | 134,822 | 2024 |

Source: CDC BRFSS, 2014-2024, adults in the 50 states and DC; weighted (BRFSS final weight), design-based intervals

Data: https://vishalsingh.org/zipcode/report/mental-distress-in-young-adults/figures/young_minus_midlife.json

### Table 3. Frequent distress was close to flat from the youngest band to the late fifties in 2014 to 2016 and falls with age in 2022 to 2024

Adults reporting 14 or more of the past 30 days with mental health not good, percent, by five-year age band, three periods of three survey years, 50 states and DC

| age_band | age_lower | conf_high | conf_low | n | period_label | rate |
| --- | ---: | ---: | ---: | ---: | --- | ---: |
| 18 to 24 | 18 | 13.74 | 12.93 | 72,087 | 2014 to 2016 | 13.33 |
| 25 to 29 | 25 | 13.32 | 12.40 | 60,352 | 2014 to 2016 | 12.85 |
| 30 to 34 | 30 | 12.73 | 11.90 | 69,646 | 2014 to 2016 | 12.31 |
| 35 to 39 | 35 | 12.13 | 11.31 | 74,688 | 2014 to 2016 | 11.71 |
| 40 to 44 | 40 | 12.82 | 11.98 | 79,127 | 2014 to 2016 | 12.40 |
| 45 to 49 | 45 | 13.26 | 12.42 | 92,848 | 2014 to 2016 | 12.83 |
| 50 to 54 | 50 | 13.46 | 12.75 | 121,236 | 2014 to 2016 | 13.10 |
| 55 to 59 | 55 | 13.27 | 12.57 | 141,051 | 2014 to 2016 | 12.92 |
| 60 to 64 | 60 | 11.63 | 10.93 | 152,442 | 2014 to 2016 | 11.28 |
| 65 to 69 | 65 | 8.84 | 8.23 | 151,105 | 2014 to 2016 | 8.53 |
| 70 to 74 | 70 | 7.48 | 6.84 | 118,060 | 2014 to 2016 | 7.15 |
| 75 to 79 | 75 | 7.24 | 6.46 | 85,601 | 2014 to 2016 | 6.84 |
| 80 and older | 80 | 6.72 | 6.07 | 110,004 | 2014 to 2016 | 6.39 |
| 18 to 24 | 18 | 23.95 | 22.91 | 79,822 | 2022 to 2024 | 23.43 |
| 25 to 29 | 25 | 21.84 | 20.66 | 64,452 | 2022 to 2024 | 21.24 |
| 30 to 34 | 30 | 20.25 | 19.18 | 73,951 | 2022 to 2024 | 19.71 |
| 35 to 39 | 35 | 18.24 | 17.23 | 81,594 | 2022 to 2024 | 17.73 |
| 40 to 44 | 40 | 16.99 | 16.02 | 86,224 | 2022 to 2024 | 16.50 |
| 45 to 49 | 45 | 16.64 | 15.50 | 81,434 | 2022 to 2024 | 16.06 |
| 50 to 54 | 50 | 15.19 | 14.23 | 92,862 | 2022 to 2024 | 14.71 |
| 55 to 59 | 55 | 15.20 | 14.23 | 101,915 | 2022 to 2024 | 14.71 |
| 60 to 64 | 60 | 13.16 | 12.35 | 125,061 | 2022 to 2024 | 12.75 |
| 65 to 69 | 65 | 10.44 | 9.68 | 136,072 | 2022 to 2024 | 10.05 |
| 70 to 74 | 70 | 9.82 | 9.01 | 127,355 | 2022 to 2024 | 9.41 |
| 75 to 79 | 75 | 9.04 | 8.22 | 100,143 | 2022 to 2024 | 8.62 |
| 80 and older | 80 | 7.47 | 6.66 | 111,830 | 2022 to 2024 | 7.05 |
| 18 to 24 | 18 | 18.75 | 17.80 | 74,519 | 2017 to 2019 | 18.27 |
| 25 to 29 | 25 | 16.28 | 15.31 | 63,191 | 2017 to 2019 | 15.79 |
| 30 to 34 | 30 | 14.88 | 13.98 | 69,762 | 2017 to 2019 | 14.43 |
| 35 to 39 | 35 | 13.63 | 12.75 | 74,577 | 2017 to 2019 | 13.19 |
| 40 to 44 | 40 | 13.31 | 12.38 | 72,712 | 2017 to 2019 | 12.84 |
| 45 to 49 | 45 | 13.67 | 12.77 | 82,263 | 2017 to 2019 | 13.21 |
| 50 to 54 | 50 | 13.58 | 12.76 | 100,551 | 2017 to 2019 | 13.17 |
| 55 to 59 | 55 | 14.03 | 13.21 | 122,243 | 2017 to 2019 | 13.61 |
| 60 to 64 | 60 | 12.25 | 11.46 | 137,258 | 2017 to 2019 | 11.85 |
| 65 to 69 | 65 | 9.81 | 9.09 | 138,341 | 2017 to 2019 | 9.44 |
| 70 to 74 | 70 | 8.30 | 7.60 | 120,047 | 2017 to 2019 | 7.94 |
| 75 to 79 | 75 | 7.33 | 6.57 | 84,547 | 2017 to 2019 | 6.94 |
| 80 and older | 80 | 7.09 | 6.26 | 101,872 | 2017 to 2019 | 6.66 |

Note: Each band is plotted at its lowest age: 18 stands for ages 18 to 24, 25 for 25 to 29, and so on, and 80 for 80 and older.

Source: CDC BRFSS, 2014-2019 and 2022-2024, adults in the 50 states and DC; weighted (BRFSS final weight), design-based intervals

Data: https://vishalsingh.org/zipcode/report/mental-distress-in-young-adults/figures/fmd_age_profile.json

### Table 4. Women reported frequent distress more often than men in every age group, and the gap among adults 18 to 24 rose from 5.5 to 10.8 points

Rate for women minus rate for men, percentage points, by age group, 2014 to 2016 and 2022 to 2024, 50 states and DC

| Age group | Women, 2014 to 2016 (%) | Men, 2014 to 2016 (%) | Gap, 2014 to 2016 (points) | Women, 2022 to 2024 (%) | Men, 2022 to 2024 (%) | Gap, 2022 to 2024 (points) | Change in the gap (points) | Change, lower limit | Change, upper limit | Respondents, 2014 to 2016 | Respondents, 2022 to 2024 |
| --- | ---: | ---: | ---: | ---: | ---: | ---: | ---: | ---: | ---: | ---: | ---: |
| 18 to 24 | 16.2 | 10.6 | 5.5 | 29.0 | 18.1 | 10.8 | 5.3 | 4.0 | 6.7 | 72,079 | 79,822 |
| 25 to 34 | 14.3 | 10.9 | 3.4 | 23.5 | 17.4 | 6.1 | 2.7 | 1.7 | 3.7 | 129,987 | 138,403 |
| 35 to 49 | 14.4 | 10.2 | 4.3 | 19.3 | 14.2 | 5.1 | 0.8 | 0.1 | 1.6 | 246,657 | 249,252 |
| 50 to 64 | 14.5 | 10.3 | 4.2 | 16.6 | 11.3 | 5.2 | 1.0 | 0.4 | 1.7 | 414,725 | 319,838 |
| 65 to 74 | 9.2 | 6.6 | 2.6 | 11.5 | 7.7 | 3.8 | 1.2 | 0.5 | 1.9 | 269,158 | 263,427 |
| 75 and older | 7.3 | 5.6 | 1.7 | 8.7 | 6.5 | 2.3 | 0.5 | −0.2 | 1.3 | 195,599 | 211,973 |

Note: The rates behind each gap, and the change in the gap with its interval, are in the data table. The two periods are different survey years, so the change in the gap combines two independent estimates.

Note: The survey's sex item changed over the period: in the early questionnaires sex could be filled in at screening or recorded by the interviewer, later questionnaires ask whether the respondent is male or female at screening, and the latest add a question on sex at birth. The groups are labeled women and men as the survey codes them, so the change in a gap compares groups defined slightly differently.

Source: CDC BRFSS, 2014-2016 and 2022-2024, adults in the 50 states and DC; weighted (BRFSS final weight), design-based intervals

Data: https://vishalsingh.org/zipcode/report/mental-distress-in-young-adults/figures/sex_gap_by_age.json

### Table 5. Frequent distress among adults 18 to 24 rose from 13.3% to 23.4%, and rose in every group by sex, household income and education

Adults 18 to 24 reporting 14 or more of the past 30 days with mental health not good, percent, 2014 to 2016 and 2022 to 2024, by group, 50 states and DC

| Dimension | Group | 2014 to 2016 (%) | 2022 to 2024 (%) | Rise (points) | Rise, lower limit | Rise, upper limit | Respondents, 2014 to 2016 | Respondents, 2022 to 2024 |
| --- | --- | ---: | ---: | ---: | ---: | ---: | ---: | ---: |
| All | All adults 18 to 24 | 13.3 | 23.4 | 10.1 | 9.4 | 10.8 | 72,087 | 79,822 |
| Sex | Men | 10.7 | 18.1 | 7.5 | 6.7 | 8.3 | 37,580 | 45,502 |
| Sex | Women | 16.2 | 29.0 | 12.8 | 11.8 | 13.9 | 34,499 | 34,320 |
| Household income | Under $20,000 | 16.9 | 25.7 | 8.8 | 6.8 | 10.9 | 13,932 | 6,442 |
| Household income | $20,000 to $34,999 | 13.2 | 28.4 | 15.1 | 13.5 | 16.8 | 13,656 | 12,185 |
| Household income | $35,000 to $49,999 | 11.7 | 24.4 | 12.7 | 10.8 | 14.6 | 7,745 | 9,476 |
| Household income | $50,000 to $74,999 | 11.9 | 21.9 | 10.0 | 8.0 | 11.9 | 6,523 | 9,518 |
| Household income | $75,000 or more | 11.6 | 21.1 | 9.5 | 8.0 | 11.0 | 11,152 | 17,893 |
| Education | Did not finish high school | 17.3 | 23.8 | 6.5 | 4.2 | 8.9 | 5,549 | 5,785 |
| Education | High school graduate | 13.4 | 24.1 | 10.6 | 9.6 | 11.7 | 26,802 | 34,737 |
| Education | Some college or technical school | 13.3 | 24.4 | 11.1 | 10.0 | 12.3 | 27,183 | 23,430 |
| Education | College graduate | 8.9 | 18.6 | 9.7 | 8.4 | 10.9 | 12,255 | 15,515 |
| Metro status (respondents with a code) | Center city of a metro area | 12.1 | 16.3 | 4.2 | −2.1 | 10.5 | 4,713 | 295 |
| Metro status (respondents with a code) | Outside the center city, in its county | 15.1 | 25.6 | 10.5 | −5.4 | 26.5 | 2,311 | 103 |
| Metro status (respondents with a code) | Suburban county of a metro area | 12.3 | 19.6 | 7.2 | −2.5 | 16.9 | 1,592 | 160 |
| Metro status (respondents with a code) | Not in a metro area | 13.4 | 22.9 | 9.5 | 1.2 | 17.8 | 3,958 | 384 |

Note: Respondents are adults in the 50 states and the District of Columbia. Each survey year is a new sample, not the same people followed over time, and every interval is design-based (final weight, stratum and primary sampling unit).

Note: Household income brackets are in current dollars and are not adjusted for prices. Education is the highest level completed, so adults 18 to 24 are at different stages of schooling. Household income and education are missing for some respondents, who are left out of those rows.

Note: Metro status is coded for only part of the respondents, and the share falls over time: 38.7% of adults in 2014 to 2016 and 11.4% in 2022 to 2024. The recent metro-status rows rest on 103 to 384 respondents each.

Source: CDC BRFSS, 2014-2016 and 2022-2024, adults in the 50 states and DC; weighted (BRFSS final weight), design-based intervals

Data: https://vishalsingh.org/zipcode/report/mental-distress-in-young-adults/figures/young_rise_by_group.json

### Table 6. Frequent distress among adults 18 to 24 was 25.7% in the lowest household income bracket and 21.1% in the top, against 31.7% and 8.0% among adults 50 to 64

Adults reporting 14 or more of the past 30 days with mental health not good, percent, by age group and annual household income, 2022 to 2024, 50 states and DC

| age6 | conf_high | conf_low | inc5 | n | rate |
| --- | ---: | ---: | --- | ---: | ---: |
| 18 to 24 | 29.81 | 27 | $20,000 to $34,999 | 12,185 | 28.38 |
| 18 to 24 | 25.88 | 22.88 | $35,000 to $49,999 | 9,476 | 24.35 |
| 18 to 24 | 23.37 | 20.46 | $50,000 to $74,999 | 9,518 | 21.88 |
| 18 to 24 | 22.17 | 19.99 | $75,000 or more | 17,893 | 21.06 |
| 18 to 24 | 27.58 | 23.98 | Under $20,000 | 6,442 | 25.74 |
| 25 to 34 | 26.30 | 24.02 | $20,000 to $34,999 | 18,236 | 25.14 |
| 25 to 34 | 25.85 | 23.44 | $35,000 to $49,999 | 16,326 | 24.62 |
| 25 to 34 | 23.53 | 21.29 | $50,000 to $74,999 | 21,228 | 22.39 |
| 25 to 34 | 16.19 | 15.01 | $75,000 or more | 51,332 | 15.59 |
| 25 to 34 | 28.73 | 25.44 | Under $20,000 | 9,188 | 27.05 |
| 35 to 49 | 23.57 | 21.66 | $20,000 to $34,999 | 26,204 | 22.60 |
| 35 to 49 | 21.89 | 19.80 | $35,000 to $49,999 | 21,544 | 20.83 |
| 35 to 49 | 19.39 | 17.64 | $50,000 to $74,999 | 29,848 | 18.50 |
| 35 to 49 | 12.30 | 11.56 | $75,000 or more | 121,548 | 11.92 |
| 35 to 49 | 34.52 | 31.35 | Under $20,000 | 16,152 | 32.91 |
| 50 to 64 | 23.03 | 21.14 | $20,000 to $34,999 | 36,505 | 22.07 |
| 50 to 64 | 16.73 | 14.87 | $35,000 to $49,999 | 27,226 | 15.78 |
| 50 to 64 | 14.06 | 12.52 | $50,000 to $74,999 | 39,413 | 13.27 |
| 50 to 64 | 8.32 | 7.71 | $75,000 or more | 137,852 | 8.01 |
| 50 to 64 | 32.89 | 30.54 | Under $20,000 | 27,662 | 31.71 |
| 65 to 74 | 14.15 | 12.66 | $20,000 to $34,999 | 39,906 | 13.39 |
| 65 to 74 | 11.76 | 10.13 | $35,000 to $49,999 | 32,992 | 10.92 |
| 65 to 74 | 8.49 | 7.33 | $50,000 to $74,999 | 41,346 | 7.89 |
| 65 to 74 | 5.79 | 4.97 | $75,000 or more | 76,414 | 5.36 |
| 65 to 74 | 19.73 | 17.02 | Under $20,000 | 19,938 | 18.33 |
| 75 and older | 10.84 | 9.24 | $20,000 to $34,999 | 37,533 | 10.01 |
| 75 and older | 7.80 | 6.50 | $35,000 to $49,999 | 29,842 | 7.12 |
| 75 and older | 7.32 | 6.12 | $50,000 to $74,999 | 30,253 | 6.69 |
| 75 and older | 5.92 | 4.88 | $75,000 or more | 44,963 | 5.37 |
| 75 and older | 15 | 12.16 | Under $20,000 | 13,098 | 13.52 |

Note: In 2022 to 2024 household income is missing for 22.1% of adults and 32.8% of adults 18 to 24; brackets are in current dollars.

Source: CDC BRFSS, 2022-2024, adults in the 50 states and DC; weighted (BRFSS final weight), design-based intervals

Data: https://vishalsingh.org/zipcode/report/mental-distress-in-young-adults/figures/fmd_age_by_income.json

### Table 7. In 2022 to 2024 the share of adults 18 to 24 ever told of depression was 25.6%, close to the 23.4% with frequent distress now

Adults with frequent mental distress and adults ever told by a health professional of a depressive disorder, percent, by five-year age band, 50 states and DC

| age_band | age_lower | conf_high | conf_low | measure | n | period_label | rate |
| --- | ---: | ---: | ---: | --- | ---: | --- | ---: |
| 18 to 24 | 18 | 13.74 | 12.93 | Frequent mental distress now | 72,087 | 2014 to 2016 | 13.33 |
| 18 to 24 | 18 | 15.96 | 15.13 | Ever told of a depressive disorder | 72,503 | 2014 to 2016 | 15.54 |
| 25 to 29 | 25 | 13.32 | 12.40 | Frequent mental distress now | 60,352 | 2014 to 2016 | 12.85 |
| 25 to 29 | 25 | 17.21 | 16.23 | Ever told of a depressive disorder | 60,803 | 2014 to 2016 | 16.72 |
| 30 to 34 | 30 | 12.73 | 11.90 | Frequent mental distress now | 69,646 | 2014 to 2016 | 12.31 |
| 30 to 34 | 30 | 17.95 | 17.03 | Ever told of a depressive disorder | 70,162 | 2014 to 2016 | 17.49 |
| 35 to 39 | 35 | 12.13 | 11.31 | Frequent mental distress now | 74,688 | 2014 to 2016 | 11.71 |
| 35 to 39 | 35 | 17.47 | 16.56 | Ever told of a depressive disorder | 75,282 | 2014 to 2016 | 17.01 |
| 40 to 44 | 40 | 12.82 | 11.98 | Frequent mental distress now | 79,127 | 2014 to 2016 | 12.40 |
| 40 to 44 | 40 | 18.27 | 17.35 | Ever told of a depressive disorder | 79,754 | 2014 to 2016 | 17.81 |
| 45 to 49 | 45 | 13.26 | 12.42 | Frequent mental distress now | 92,848 | 2014 to 2016 | 12.83 |
| 45 to 49 | 45 | 19.12 | 18.22 | Ever told of a depressive disorder | 93,627 | 2014 to 2016 | 18.67 |
| 50 to 54 | 50 | 13.46 | 12.75 | Frequent mental distress now | 121,236 | 2014 to 2016 | 13.10 |
| 50 to 54 | 50 | 20.13 | 19.30 | Ever told of a depressive disorder | 122,414 | 2014 to 2016 | 19.71 |
| 55 to 59 | 55 | 13.27 | 12.57 | Frequent mental distress now | 141,051 | 2014 to 2016 | 12.92 |
| 55 to 59 | 55 | 21.61 | 20.78 | Ever told of a depressive disorder | 142,659 | 2014 to 2016 | 21.19 |
| 60 to 64 | 60 | 11.63 | 10.93 | Frequent mental distress now | 152,442 | 2014 to 2016 | 11.28 |
| 60 to 64 | 60 | 20.78 | 19.99 | Ever told of a depressive disorder | 154,204 | 2014 to 2016 | 20.38 |
| 65 to 69 | 65 | 8.84 | 8.23 | Frequent mental distress now | 151,105 | 2014 to 2016 | 8.53 |
| 65 to 69 | 65 | 18.54 | 17.75 | Ever told of a depressive disorder | 152,844 | 2014 to 2016 | 18.14 |
| 70 to 74 | 70 | 7.48 | 6.84 | Frequent mental distress now | 118,060 | 2014 to 2016 | 7.15 |
| 70 to 74 | 70 | 15.42 | 14.61 | Ever told of a depressive disorder | 119,817 | 2014 to 2016 | 15.01 |
| 75 to 79 | 75 | 7.24 | 6.46 | Frequent mental distress now | 85,601 | 2014 to 2016 | 6.84 |
| 75 to 79 | 75 | 13.09 | 12.17 | Ever told of a depressive disorder | 87,307 | 2014 to 2016 | 12.62 |
| 80 and older | 80 | 6.72 | 6.07 | Frequent mental distress now | 110,004 | 2014 to 2016 | 6.39 |
| 80 and older | 80 | 10.73 | 9.96 | Ever told of a depressive disorder | 113,372 | 2014 to 2016 | 10.34 |
| 18 to 24 | 18 | 23.95 | 22.91 | Frequent mental distress now | 79,822 | 2022 to 2024 | 23.43 |
| 18 to 24 | 18 | 26.13 | 25.07 | Ever told of a depressive disorder | 80,455 | 2022 to 2024 | 25.60 |
| 25 to 29 | 25 | 21.84 | 20.66 | Frequent mental distress now | 64,452 | 2022 to 2024 | 21.24 |
| 25 to 29 | 25 | 25.95 | 24.73 | Ever told of a depressive disorder | 65,019 | 2022 to 2024 | 25.34 |
| 30 to 34 | 30 | 20.25 | 19.18 | Frequent mental distress now | 73,951 | 2022 to 2024 | 19.71 |
| 30 to 34 | 30 | 25.48 | 24.36 | Ever told of a depressive disorder | 74,693 | 2022 to 2024 | 24.92 |
| 35 to 39 | 35 | 18.24 | 17.23 | Frequent mental distress now | 81,594 | 2022 to 2024 | 17.73 |
| 35 to 39 | 35 | 23.36 | 22.28 | Ever told of a depressive disorder | 82,379 | 2022 to 2024 | 22.81 |
| 40 to 44 | 40 | 16.99 | 16.02 | Frequent mental distress now | 86,224 | 2022 to 2024 | 16.50 |
| 40 to 44 | 40 | 21.83 | 20.81 | Ever told of a depressive disorder | 87,077 | 2022 to 2024 | 21.32 |
| 45 to 49 | 45 | 16.64 | 15.50 | Frequent mental distress now | 81,434 | 2022 to 2024 | 16.06 |
| 45 to 49 | 45 | 21.70 | 20.52 | Ever told of a depressive disorder | 82,305 | 2022 to 2024 | 21.11 |
| 50 to 54 | 50 | 15.19 | 14.23 | Frequent mental distress now | 92,862 | 2022 to 2024 | 14.71 |
| 50 to 54 | 50 | 20.82 | 19.77 | Ever told of a depressive disorder | 93,866 | 2022 to 2024 | 20.29 |
| 55 to 59 | 55 | 15.20 | 14.23 | Frequent mental distress now | 101,915 | 2022 to 2024 | 14.71 |
| 55 to 59 | 55 | 21.23 | 20.18 | Ever told of a depressive disorder | 103,090 | 2022 to 2024 | 20.70 |
| 60 to 64 | 60 | 13.16 | 12.35 | Frequent mental distress now | 125,061 | 2022 to 2024 | 12.75 |
| 60 to 64 | 60 | 20.15 | 19.22 | Ever told of a depressive disorder | 126,705 | 2022 to 2024 | 19.68 |
| 65 to 69 | 65 | 10.44 | 9.68 | Frequent mental distress now | 136,072 | 2022 to 2024 | 10.05 |
| 65 to 69 | 65 | 18.27 | 17.35 | Ever told of a depressive disorder | 137,909 | 2022 to 2024 | 17.80 |
| 70 to 74 | 70 | 9.82 | 9.01 | Frequent mental distress now | 127,355 | 2022 to 2024 | 9.41 |
| 70 to 74 | 70 | 17.11 | 16.14 | Ever told of a depressive disorder | 129,175 | 2022 to 2024 | 16.62 |
| 75 to 79 | 75 | 9.04 | 8.22 | Frequent mental distress now | 100,143 | 2022 to 2024 | 8.62 |
| 75 to 79 | 75 | 14.81 | 13.86 | Ever told of a depressive disorder | 101,808 | 2022 to 2024 | 14.33 |
| 80 and older | 80 | 7.47 | 6.66 | Frequent mental distress now | 111,830 | 2022 to 2024 | 7.05 |
| 80 and older | 80 | 10.50 | 9.59 | Ever told of a depressive disorder | 114,825 | 2022 to 2024 | 10.04 |

Views: 2022 to 2024 (period_label = 2022 to 2024); 2014 to 2016 (period_label = 2014 to 2016).

Note: The diagnosis question asks whether a doctor, nurse or other health professional ever told the respondent of a depressive disorder, so it depends on contact with care as well as on symptoms.

Note: The lines plot the share answering each item. The difference between them quoted in the text is a paired difference among respondents who answered both items, so it can differ slightly from the difference of the two plotted shares.

Source: CDC BRFSS, 2014-2016 and 2022-2024, adults in the 50 states and DC; weighted (BRFSS final weight), design-based intervals

Data: https://vishalsingh.org/zipcode/report/mental-distress-in-young-adults/figures/fmd_vs_diagnosis.json

## All estimates

Every fact in facts.json (116), sorted by id.

| Fact id | Value | Unit | Estimand | Population | Period | Interval | n | Weights | Source | Note |
| --- | ---: | --- | --- | --- | --- | --- | ---: | --- | --- | --- |
| band_2529 | 25 to 29 | years of age | definition used in this chapter, not an estimate | an age group of respondents defined in this chapter | 2014 to 2024 |   |   | none (definition) | BRFSS |   |
| band_2534 | 25 to 34 | years of age | definition used in this chapter, not an estimate | an age group of respondents defined in this chapter | 2014 to 2024 |   |   | none (definition) | BRFSS |   |
| band_3034 | 30 to 34 | years of age | definition used in this chapter, not an estimate | an age group of respondents defined in this chapter | 2014 to 2024 |   |   | none (definition) | BRFSS |   |
| band_3549 | 35 to 49 | years of age | definition used in this chapter, not an estimate | an age group of respondents defined in this chapter | 2014 to 2024 |   |   | none (definition) | BRFSS |   |
| band_5559 | 55 to 59 | years of age | definition used in this chapter, not an estimate | an age group of respondents defined in this chapter | 2014 to 2024 |   |   | none (definition) | BRFSS |   |
| band_6569 | 65 to 69 | years of age | definition used in this chapter, not an estimate | an age group of respondents defined in this chapter | 2014 to 2024 |   |   | none (definition) | BRFSS |   |
| band_6574 | 65 to 74 | years of age | definition used in this chapter, not an estimate | an age group of respondents defined in this chapter | 2014 to 2024 |   |   | none (definition) | BRFSS |   |
| band_mid | 50 to 64 | years of age | definition used in this chapter, not an estimate | an age group of respondents defined in this chapter | 2014 to 2024 |   |   | none (definition) | BRFSS |   |
| band_old | 75 and older | years of age | definition used in this chapter, not an estimate | an age group of respondents defined in this chapter | 2014 to 2024 |   |   | none (definition) | BRFSS |   |
| band_young | 18 to 24 | years of age | definition used in this chapter, not an estimate | an age group of respondents defined in this chapter | 2014 to 2024 |   |   | none (definition) | BRFSS |   |
| chg_2534_14_24 | 7.7 | percentage points | design-based difference between two survey periods, estimated from independent samples (separate survey years), interval from Taylor linearization, 2024 minus 2014 | adults 25 to 34 in the 50 states and DC | 2024 compared with 2014 | 95% CI 6.8 to 8.6 | 90,181 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| chg_3549_22_24 | 0.9 | percentage points | design-based difference between two survey periods, estimated from independent samples (separate survey years), interval from Taylor linearization, 2024 minus 2022 | adults 35 to 49 in the 50 states and DC | 2024 compared with 2022 | 95% CI 0.2 to 1.6 | 169,473 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| chg_mid_14_24 | 1.9 | percentage points | design-based difference between two survey periods, estimated from independent samples (separate survey years), interval from Taylor linearization, 2024 minus 2014 | adults 50 to 64 in the 50 states and DC | 2024 compared with 2014 | 95% CI 1.3 to 2.5 | 247,091 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| chg_old_14_24 | 1.0 | percentage points | design-based difference between two survey periods, estimated from independent samples (separate survey years), interval from Taylor linearization, 2024 minus 2014 | adults 75 and older in the 50 states and DC | 2024 compared with 2014 | 95% CI 0.3 to 1.6 | 140,907 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| chg_young_14_24 | 9.5 | percentage points | design-based difference between two survey periods, estimated from independent samples (separate survey years), interval from Taylor linearization, 2024 minus 2014 | adults 18 to 24 in the 50 states and DC | 2024 compared with 2014 | 95% CI 8.4 to 10.6 | 51,863 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| const_max_diff | 0.5 | percentage points | largest absolute difference between the estimate on the states present in every year and the estimate on all states | the age-by-year estimates of frequent distress | 2014 to 2024 |   |   | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| const_states | 46 | states and the District of Columbia | count of states with respondents in all 11 survey years | jurisdictions with respondents in every survey year | 2014 to 2024 |   | 46 states | none (count) | BRFSS |   |
| dep_5559_early | 21.2% | percent of respondents ever told by a health professional that they have a depressive disorder | design-based weighted share of respondents, logit interval from Taylor linearization | adults 55 to 59 in the 50 states and DC | 2014 to 2016 | 95% CI 20.8% to 21.6% | 142,659 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| dep_5559_late | 20.7% | percent of respondents ever told by a health professional that they have a depressive disorder | design-based weighted share of respondents, logit interval from Taylor linearization | adults 55 to 59 in the 50 states and DC | 2022 to 2024 | 95% CI 20.2% to 21.2% | 103,090 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| dep_gap_vs_2529 | +0.3 | percentage points | design-based difference in weighted shares between two groups of one survey, interval from Taylor linearization, adults 18 to 24 minus adults 25 to 29 | adults 18 to 24 and adults 25 to 29 in the 50 states and DC | 2022 to 2024 | 95% CI −0.5 to 1.1 | 145,474 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| dep_gap_vs_3034 | +0.7 | percentage points | design-based difference in weighted shares between two groups of one survey, interval from Taylor linearization, adults 18 to 24 minus adults 30 to 34 | adults 18 to 24 and adults 30 to 34 in the 50 states and DC | 2022 to 2024 | 95% CI −0.1 to 1.5 | 155,148 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| dep_young_early | 15.5% | percent of respondents ever told by a health professional that they have a depressive disorder | design-based weighted share of respondents, logit interval from Taylor linearization | adults 18 to 24 in the 50 states and DC | 2014 to 2016 | 95% CI 15.1% to 16.0% | 72,503 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| dep_young_late | 25.6% | percent of respondents ever told by a health professional that they have a depressive disorder | design-based weighted share of respondents, logit interval from Taylor linearization | adults 18 to 24 in the 50 states and DC | 2022 to 2024 | 95% CI 25.1% to 26.1% | 80,455 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| did_inc_2_vs_low | 6.3 | percentage points | difference between the rises of two groups: the gap between the groups in each period, combined across the two independent periods | adults 18 to 24 in the second and lowest household income brackets | 2022 to 2024 compared with 2014 to 2016 | 95% CI 3.7 to 9.0 | 46,215 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| did_inc_top_vs_low | 0.7 | percentage points | difference between the rises of two groups: the gap between the groups in each period, combined across the two independent periods | adults 18 to 24 in the top and lowest household income brackets | 2022 to 2024 compared with 2014 to 2016 | 95% CI −1.8 to 3.2 | 49,419 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| dip_young_2020 | 1.5 | percentage points | design-based difference between two survey periods, estimated from independent samples (separate survey years), interval from Taylor linearization, 2019 minus 2020 | adults 18 to 24 in the 50 states and DC | 2019 compared with 2020 | 95% CI 0.2 to 2.8 | 48,616 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| drop_young_22_24 | 3.5 | percentage points | design-based difference between two survey periods, estimated from independent samples (separate survey years), interval from Taylor linearization, 2022 minus 2024 | adults 18 to 24 in the 50 states and DC | 2022 compared with 2024 | 95% CI 2.3 to 4.8 | 54,526 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| dx_gap_6569_late | 7.6 | percentage points | design-based mean over respondents who answered both items of each respondent's diagnosis indicator minus distress indicator, interval from Taylor linearization | adults 65 to 69 who answered both the distress and the diagnosis questions, in the 50 states and DC | 2022 to 2024 | 95% CI 7.1 to 8.1 | 135,563 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| dx_gap_80_late | 2.8 | percentage points | design-based mean over respondents who answered both items of each respondent's diagnosis indicator minus distress indicator, interval from Taylor linearization | adults 80 and older who answered both the distress and the diagnosis questions, in the 50 states and DC | 2022 to 2024 | 95% CI 2.3 to 3.3 | 111,277 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| dx_gap_young_early | 2.2 | percentage points | design-based mean over respondents who answered both items of each respondent's diagnosis indicator minus distress indicator, interval from Taylor linearization | adults 18 to 24 who answered both the distress and the diagnosis questions, in the 50 states and DC | 2014 to 2016 | 95% CI 1.7 to 2.7 | 71,627 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| dx_gap_young_late | 2.4 | percentage points | design-based mean over respondents who answered both items of each respondent's diagnosis indicator minus distress indicator, interval from Taylor linearization | adults 18 to 24 who answered both the distress and the diagnosis questions, in the 50 states and DC | 2022 to 2024 | 95% CI 1.8 to 3.0 | 79,000 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| fmd_all_2014 | 11.4% | percent of respondents who reported frequent mental distress (14 or more of the past 30 days with mental health not good) | design-based weighted share of respondents, logit interval from Taylor linearization | adults in the 50 states and the District of Columbia | 2014 | 95% CI 11.2% to 11.6% | 448,268 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| fmd_all_2022 | 15.8% | percent of respondents who reported frequent mental distress (14 or more of the past 30 days with mental health not good) | design-based weighted share of respondents, logit interval from Taylor linearization | adults in the 50 states and the District of Columbia | 2022 | 95% CI 15.5% to 16.0% | 426,923 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| fmd_all_2024 | 15.6% | percent of respondents who reported frequent mental distress (14 or more of the past 30 days with mental health not good) | design-based weighted share of respondents, logit interval from Taylor linearization | adults in the 50 states and the District of Columbia | 2024 | 95% CI 15.3% to 15.8% | 442,292 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| fmd_days | 14 | days of poor mental health in the past 30 | definition used in this chapter, not an estimate | cutoff for frequent mental distress | 2014 to 2024 |   |   | none (definition) | BRFSS |   |
| fmd_inc_missing | 14.9% | percent of respondents who reported frequent mental distress (14 or more of the past 30 days with mental health not good) | design-based weighted share of respondents, logit interval from Taylor linearization | adults who did not report household income | 2022 to 2024 | 95% CI 14.6% to 15.2% | 255,353 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| fmd_inc_reported | 15.8% | percent of respondents who reported frequent mental distress (14 or more of the past 30 days with mental health not good) | design-based weighted share of respondents, logit interval from Taylor linearization | adults who reported household income | 2022 to 2024 | 95% CI 15.7% to 16.0% | 1,030,978 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| fmd_mid_2014 | 12.5% | percent of respondents who reported frequent mental distress (14 or more of the past 30 days with mental health not good) | design-based weighted share of respondents, logit interval from Taylor linearization | adults 50 to 64 in the 50 states and DC | 2014 | 95% CI 12.2% to 12.9% | 140,980 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| fmd_mid_2022 | 13.8% | percent of respondents who reported frequent mental distress (14 or more of the past 30 days with mental health not good) | design-based weighted share of respondents, logit interval from Taylor linearization | adults 50 to 64 in the 50 states and DC | 2022 | 95% CI 13.3% to 14.2% | 110,334 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| fmd_mid_2024 | 14.4% | percent of respondents who reported frequent mental distress (14 or more of the past 30 days with mental health not good) | design-based weighted share of respondents, logit interval from Taylor linearization | adults 50 to 64 in the 50 states and DC | 2024 | 95% CI 13.9% to 14.9% | 106,111 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| fmd_old_2014 | 6.5% | percent of respondents who reported frequent mental distress (14 or more of the past 30 days with mental health not good) | design-based weighted share of respondents, logit interval from Taylor linearization | adults 75 and older in the 50 states and DC | 2014 | 95% CI 6.1% to 7.0% | 65,066 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| fmd_old_2024 | 7.5% | percent of respondents who reported frequent mental distress (14 or more of the past 30 days with mental health not good) | design-based weighted share of respondents, logit interval from Taylor linearization | adults 75 and older in the 50 states and DC | 2024 | 95% CI 7.1% to 8.0% | 75,841 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| fmd_young_2014 | 12.3% | percent of respondents who reported frequent mental distress (14 or more of the past 30 days with mental health not good) | design-based weighted share of respondents, logit interval from Taylor linearization | adults 18 to 24 in the 50 states and DC | 2014 | 95% CI 11.6% to 13.0% | 23,152 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| fmd_young_2022 | 25.3% | percent of respondents who reported frequent mental distress (14 or more of the past 30 days with mental health not good) | design-based weighted share of respondents, logit interval from Taylor linearization | adults 18 to 24 in the 50 states and DC | 2022 | 95% CI 24.4% to 26.2% | 25,815 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| fmd_young_2024 | 21.8% | percent of respondents who reported frequent mental distress (14 or more of the past 30 days with mental health not good) | design-based weighted share of respondents, logit interval from Taylor linearization | adults 18 to 24 in the 50 states and DC | 2024 | 95% CI 20.9% to 22.7% | 28,711 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| gap_ym_2014 | −0.2 | percentage points | design-based difference in weighted shares between two groups of one survey, interval from Taylor linearization, adults 18 to 24 minus adults 50 to 64 | adults 18 to 24 and adults 50 to 64 in the 50 states and DC | 2014 | 95% CI −1.0 to 0.5 | 164,132 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| gap_ym_2015 | +0.9 | percentage points | design-based difference in weighted shares between two groups of one survey, interval from Taylor linearization, adults 18 to 24 minus adults 50 to 64 | adults 18 to 24 and adults 50 to 64 in the 50 states and DC | 2015 | 95% CI 0.1 to 1.7 | 155,179 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| gap_ym_2022 | +11.6 | percentage points | design-based difference in weighted shares between two groups of one survey, interval from Taylor linearization, adults 18 to 24 minus adults 50 to 64 | adults 18 to 24 and adults 50 to 64 in the 50 states and DC | 2022 | 95% CI 10.6 to 12.6 | 136,149 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| gap_ym_2024 | +7.4 | percentage points | design-based difference in weighted shares between two groups of one survey, interval from Taylor linearization, adults 18 to 24 minus adults 50 to 64 | adults 18 to 24 and adults 50 to 64 in the 50 states and DC | 2024 | 95% CI 6.4 to 8.4 | 134,822 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| heat_age_low | −6.0 | percentage points | design-based difference in weighted shares between two groups of one survey, interval from Taylor linearization, adults 18 to 24 minus adults 50 to 64 | adults 18 to 24 and 50 to 64 in the lowest household income bracket | 2022 to 2024 | 95% CI −8.1 to −3.8 | 34,104 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| heat_age_top | +13.0 | percentage points | design-based difference in weighted shares between two groups of one survey, interval from Taylor linearization, adults 18 to 24 minus adults 50 to 64 | adults 18 to 24 and 50 to 64 in the top household income bracket | 2022 to 2024 | 95% CI 11.9 to 14.2 | 155,745 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| heat_grad_2534 | 11.5 | percentage points | design-based difference in weighted shares between two groups of one survey, interval from Taylor linearization, lowest bracket minus top bracket | adults 25 to 34 in the lowest and top household income brackets | 2022 to 2024 | 95% CI 9.7 to 13.2 | 60,520 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| heat_grad_3549 | 21.0 | percentage points | design-based difference in weighted shares between two groups of one survey, interval from Taylor linearization, lowest bracket minus top bracket | adults 35 to 49 in the lowest and top household income brackets | 2022 to 2024 | 95% CI 19.4 to 22.6 | 137,700 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| heat_grad_mid | 23.7 | percentage points | design-based difference in weighted shares between two groups of one survey, interval from Taylor linearization, lowest bracket minus top bracket | adults 50 to 64 in the lowest and top household income brackets | 2022 to 2024 | 95% CI 22.5 to 24.9 | 165,514 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| heat_grad_young | 4.7 | percentage points | design-based difference in weighted shares between two groups of one survey, interval from Taylor linearization, lowest bracket minus top bracket | adults 18 to 24 in the lowest and top household income brackets | 2022 to 2024 | 95% CI 2.6 to 6.8 | 24,335 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| heat_max | 32.9% | percent of respondents who reported frequent mental distress (14 or more of the past 30 days with mental health not good) | design-based weighted share of respondents, logit interval from Taylor linearization | adults 35 to 49 in the lowest household income bracket | 2022 to 2024 | 95% CI 31.3% to 34.5% | 16,152 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| heat_mid_low | 31.7% | percent of respondents who reported frequent mental distress (14 or more of the past 30 days with mental health not good) | design-based weighted share of respondents, logit interval from Taylor linearization | adults 50 to 64 in the lowest household income bracket | 2022 to 2024 | 95% CI 30.5% to 32.9% | 27,662 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| heat_mid_top | 8.0% | percent of respondents who reported frequent mental distress (14 or more of the past 30 days with mental health not good) | design-based weighted share of respondents, logit interval from Taylor linearization | adults 50 to 64 in the top household income bracket | 2022 to 2024 | 95% CI 7.7% to 8.3% | 137,852 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| heat_young_low | 25.7% | percent of respondents who reported frequent mental distress (14 or more of the past 30 days with mental health not good) | design-based weighted share of respondents, logit interval from Taylor linearization | adults 18 to 24 in the lowest household income bracket | 2022 to 2024 | 95% CI 24.0% to 27.6% | 6,442 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| heat_young_top | 21.1% | percent of respondents who reported frequent mental distress (14 or more of the past 30 days with mental health not good) | design-based weighted share of respondents, logit interval from Taylor linearization | adults 18 to 24 in the top household income bracket | 2022 to 2024 | 95% CI 20.0% to 22.2% | 17,893 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| inc_low_cut | $20,000 | dollars of annual household income | definition used in this chapter, not an estimate | lowest household income bracket in the survey recode | 2014 to 2024 |   |   | none (definition) | BRFSS |   |
| inc_top_cut | $75,000 | dollars of annual household income | definition used in this chapter, not an estimate | highest household income bracket in the survey recode | 2014 to 2024 |   |   | none (definition) | BRFSS |   |
| income_all_low | 26.9% | percent of respondents who reported frequent mental distress (14 or more of the past 30 days with mental health not good) | design-based weighted share of respondents, logit interval from Taylor linearization | adults in the lowest household income bracket | 2022 to 2024 | 95% CI 26.3% to 27.6% | 93,274 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| income_all_top | 11.1% | percent of respondents who reported frequent mental distress (14 or more of the past 30 days with mental health not good) | design-based weighted share of respondents, logit interval from Taylor linearization | adults in the top household income bracket | 2022 to 2024 | 95% CI 10.9% to 11.3% | 453,847 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| income_missing_all | 22.1% | percent of adults (weighted) with no valid household income | weighted share of respondents whose income is don't know, refused or missing | adults in the 50 states and the District of Columbia | 2022 to 2024 |   |   | BRFSS final weight | BRFSS |   |
| income_missing_young | 32.8% | percent of adults (weighted) with no valid household income | weighted share of respondents whose income is don't know, refused or missing | adults 18 to 24 in the 50 states and DC | 2022 to 2024 |   |   | BRFSS final weight | BRFSS |   |
| item_nonresponse | 1.8% | percent of respondents with no valid answer to the mental health question (don't know, refused or not asked) | unweighted share of rows | respondents in the 50 states and DC | 2014 to 2024 |   | 4,790,742 respondents | none (unweighted share) | BRFSS |   |
| men_young_early | 10.6% | percent of respondents who reported frequent mental distress (14 or more of the past 30 days with mental health not good) | design-based weighted share of respondents, logit interval from Taylor linearization | men 18 to 24 in the 50 states and DC | 2014 to 2016 | 95% CI 10.2% to 11.2% | 37,580 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| men_young_late | 18.1% | percent of respondents who reported frequent mental distress (14 or more of the past 30 days with mental health not good) | design-based weighted share of respondents, logit interval from Taylor linearization | men 18 to 24 in the 50 states and DC | 2022 to 2024 | 95% CI 17.5% to 18.8% | 45,502 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| metro_cov_early | 38.7% | percent of adults (weighted) with a metro status code | weighted share of respondents whose metro status is coded | adults in the 50 states and the District of Columbia | 2014 to 2016 |   |   | BRFSS final weight | BRFSS |   |
| metro_cov_late | 11.4% | percent of adults (weighted) with a metro status code | weighted share of respondents whose metro status is coded | adults in the 50 states and the District of Columbia | 2022 to 2024 |   |   | BRFSS final weight | BRFSS |   |
| metro_n_max | 384 | respondents | count of respondent rows | the largest metro-status group of adults 18 to 24 | 2022 to 2024 |   | 384 respondents | none (unweighted count) | BRFSS |   |
| metro_n_min | 103 | respondents | count of respondent rows | the smallest metro-status group of adults 18 to 24 | 2022 to 2024 |   | 103 respondents | none (unweighted count) | BRFSS |   |
| metro_n_positive | one | metro-status groups | count of groups whose interval for the rise lies above zero | the four metro-status groups of adults 18 to 24 | 2022 to 2024 compared with 2014 to 2016 |   |   | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| metro_width_max | 31.9 | percentage points | widest interval width among the four rises | the four metro-status groups of adults 18 to 24 | 2022 to 2024 compared with 2014 to 2016 |   |   | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| metro_width_min | 12.6 | percentage points | narrowest interval width among the four rises | the four metro-status groups of adults 18 to 24 | 2022 to 2024 compared with 2014 to 2016 |   |   | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| n_eff_young_late | 24,390 | respondents (effective sample size from the weights) | Kish effective sample size, (sum of weights) squared over the sum of squared weights | respondents 18 to 24 with an answer to the distress question | 2022 to 2024 |   |   | BRFSS final weight | BRFSS |   |
| n_file | 4,874,919 | respondents | count of respondent rows | respondents in the staged file, all jurisdictions | 2014 to 2024 |   | 4,874,919 respondents | none (unweighted count) | BRFSS |   |
| n_states | 4,790,742 | respondents | count of respondent rows | respondents in the 50 states and the District of Columbia | 2014 to 2024 |   | 4,790,742 respondents | none (unweighted count) | BRFSS |   |
| n_valid_age | 4,626,736 | respondents | count of respondent rows | respondents in the 50 states and DC with a valid answer to the distress question and an age group | 2014 to 2024 |   | 4,626,736 respondents | none (unweighted count) | BRFSS |   |
| n_young_late | 79,822 | respondents | count of respondent rows | respondents 18 to 24 with an answer to the distress question | 2022 to 2024 |   | 79,822 respondents | none (unweighted count) | BRFSS |   |
| prof_early_80 | 6.4% | percent of respondents who reported frequent mental distress (14 or more of the past 30 days with mental health not good) | design-based weighted share of respondents, logit interval from Taylor linearization | adults 80 and older in the 50 states and DC | 2014 to 2016 | 95% CI 6.1% to 6.7% | 110,004 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| prof_early_max | 13.3% | percent of respondents who reported frequent mental distress (14 or more of the past 30 days with mental health not good) | highest of eight design-based band estimates | the eight five-year age bands from 18 to 59 | 2014 to 2016 |   |   | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| prof_early_min | 11.7% | percent of respondents who reported frequent mental distress (14 or more of the past 30 days with mental health not good) | lowest of eight design-based band estimates | the eight five-year age bands from 18 to 59 | 2014 to 2016 |   |   | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| prof_late_55 | 14.7% | percent of respondents who reported frequent mental distress (14 or more of the past 30 days with mental health not good) | design-based weighted share of respondents, logit interval from Taylor linearization | adults 55 to 59 in the 50 states and DC | 2022 to 2024 | 95% CI 14.2% to 15.2% | 101,915 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| prof_late_80 | 7.1% | percent of respondents who reported frequent mental distress (14 or more of the past 30 days with mental health not good) | design-based weighted share of respondents, logit interval from Taylor linearization | adults 80 and older in the 50 states and DC | 2022 to 2024 | 95% CI 6.7% to 7.5% | 111,830 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| recall_days | 30 | days in the recall period of the question | definition used in this chapter, not an estimate | the BRFSS mental health question | 2014 to 2024 |   |   | none (definition) | BRFSS |   |
| rise_dep_young | 10.1 | percentage points | design-based difference between two survey periods, estimated from independent samples (separate survey years), interval from Taylor linearization, late period minus early period | adults 18 to 24 in the 50 states and DC | 2022 to 2024 compared with 2014 to 2016 | 95% CI 9.4 to 10.7 | 152,958 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| rise_edu_college | 9.7 | percentage points | design-based difference between two survey periods, estimated from independent samples (separate survey years), interval from Taylor linearization, late period minus early period | adults 18 to 24 whose highest education is: college graduate | 2022 to 2024 compared with 2014 to 2016 | 95% CI 8.4 to 10.9 | 27,770 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| rise_edu_nohs | 6.5 | percentage points | design-based difference between two survey periods, estimated from independent samples (separate survey years), interval from Taylor linearization, late period minus early period | adults 18 to 24 whose highest education is: did not finish high school | 2022 to 2024 compared with 2014 to 2016 | 95% CI 4.2 to 8.9 | 11,334 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| rise_inc_2 | 15.1 | percentage points | design-based difference between two survey periods, estimated from independent samples (separate survey years), interval from Taylor linearization, late period minus early period | adults 18 to 24 in households in the second income bracket | 2022 to 2024 compared with 2014 to 2016 | 95% CI 13.5 to 16.8 | 25,841 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| rise_inc_low | 8.8 | percentage points | design-based difference between two survey periods, estimated from independent samples (separate survey years), interval from Taylor linearization, late period minus early period | adults 18 to 24 in households below the lowest income cutoff | 2022 to 2024 compared with 2014 to 2016 | 95% CI 6.8 to 10.9 | 20,374 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| rise_inc_top | 9.5 | percentage points | design-based difference between two survey periods, estimated from independent samples (separate survey years), interval from Taylor linearization, late period minus early period | adults 18 to 24 in households at or above the top income cutoff | 2022 to 2024 compared with 2014 to 2016 | 95% CI 8.0 to 11.0 | 29,045 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| rise_men | 7.5 | percentage points | design-based difference between two survey periods, estimated from independent samples (separate survey years), interval from Taylor linearization, late period minus early period | men 18 to 24 in the 50 states and DC | 2022 to 2024 compared with 2014 to 2016 | 95% CI 6.7 to 8.3 | 83,082 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| rise_women | 12.8 | percentage points | design-based difference between two survey periods, estimated from independent samples (separate survey years), interval from Taylor linearization, late period minus early period | women 18 to 24 in the 50 states and DC | 2022 to 2024 compared with 2014 to 2016 | 95% CI 11.8 to 13.9 | 68,819 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| rise_young_all | 10.1 | percentage points | design-based difference between two survey periods, estimated from independent samples (separate survey years), interval from Taylor linearization, late period minus early period | adults 18 to 24 in the 50 states and DC | 2022 to 2024 compared with 2014 to 2016 | 95% CI 9.4 to 10.8 | 151,909 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| sex_gap_chg_2534 | 2.7 | percentage points | change in the women-minus-men gap between two independent periods, standard errors combined in quadrature | women and men 25 to 34 in the 50 states and DC | 2022 to 2024 compared with 2014 to 2016 | 95% CI 1.7 to 3.7 | 268,390 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| sex_gap_chg_mid | 1.0 | percentage points | change in the women-minus-men gap between two independent periods, standard errors combined in quadrature | women and men 50 to 64 in the 50 states and DC | 2022 to 2024 compared with 2014 to 2016 | 95% CI 0.4 to 1.7 | 734,563 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| sex_gap_chg_young | 5.3 | percentage points | change in the women-minus-men gap between two independent periods, standard errors combined in quadrature | women and men 18 to 24 in the 50 states and DC | 2022 to 2024 compared with 2014 to 2016 | 95% CI 4.0 to 6.7 | 151,901 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| sex_gap_mid_early | 4.2 | percentage points | design-based difference in weighted shares between two groups of one survey, interval from Taylor linearization, women minus men | women and men 50 to 64 in the 50 states and DC | 2014 to 2016 | 95% CI 3.8 to 4.6 | 414,725 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| sex_gap_mid_late | 5.2 | percentage points | design-based difference in weighted shares between two groups of one survey, interval from Taylor linearization, women minus men | women and men 50 to 64 in the 50 states and DC | 2022 to 2024 | 95% CI 4.7 to 5.8 | 319,838 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| sex_gap_young_early | 5.5 | percentage points | design-based difference in weighted shares between two groups of one survey, interval from Taylor linearization, women minus men | women and men 18 to 24 in the 50 states and DC | 2014 to 2016 | 95% CI 4.7 to 6.3 | 72,079 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| sex_gap_young_late | 10.8 | percentage points | design-based difference in weighted shares between two groups of one survey, interval from Taylor linearization, women minus men | women and men 18 to 24 in the 50 states and DC | 2022 to 2024 | 95% CI 9.8 to 11.9 | 79,822 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| spread_2014 | 0.3 | percentage points | highest minus lowest of four design-based group estimates | the four age groups from 18 to 64 | 2014 |   |   | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| women_young_early | 16.2% | percent of respondents who reported frequent mental distress (14 or more of the past 30 days with mental health not good) | design-based weighted share of respondents, logit interval from Taylor linearization | women 18 to 24 in the 50 states and DC | 2014 to 2016 | 95% CI 15.5% to 16.8% | 34,499 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| women_young_late | 29.0% | percent of respondents who reported frequent mental distress (14 or more of the past 30 days with mental health not good) | design-based weighted share of respondents, logit interval from Taylor linearization | women 18 to 24 in the 50 states and DC | 2022 to 2024 | 95% CI 28.1% to 29.8% | 34,320 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| year_cross | 2015 | calendar year | first survey year in which the interval for the difference lies above zero | adults 18 to 24 compared with adults 50 to 64 | 2014 to 2024 |   |   | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| young_early_all | 13.3% | percent of respondents who reported frequent mental distress (14 or more of the past 30 days with mental health not good) | design-based weighted share of respondents, logit interval from Taylor linearization | adults 18 to 24 in the 50 states and DC | 2014 to 2016 | 95% CI 12.9% to 13.7% | 72,087 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| young_edu_college_early | 8.9% | percent of respondents who reported frequent mental distress (14 or more of the past 30 days with mental health not good) | design-based weighted share of respondents, logit interval from Taylor linearization | adults 18 to 24 whose highest education is: college graduate | 2014 to 2016 | 95% CI 8.3% to 9.7% | 12,255 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| young_edu_college_late | 18.6% | percent of respondents who reported frequent mental distress (14 or more of the past 30 days with mental health not good) | design-based weighted share of respondents, logit interval from Taylor linearization | adults 18 to 24 whose highest education is: college graduate | 2022 to 2024 | 95% CI 17.6% to 19.6% | 15,515 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| young_edu_nohs_early | 17.3% | percent of respondents who reported frequent mental distress (14 or more of the past 30 days with mental health not good) | design-based weighted share of respondents, logit interval from Taylor linearization | adults 18 to 24 whose highest education is: did not finish high school | 2014 to 2016 | 95% CI 15.9% to 18.8% | 5,549 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| young_edu_nohs_late | 23.8% | percent of respondents who reported frequent mental distress (14 or more of the past 30 days with mental health not good) | design-based weighted share of respondents, logit interval from Taylor linearization | adults 18 to 24 whose highest education is: did not finish high school | 2022 to 2024 | 95% CI 22.0% to 25.7% | 5,785 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| young_inc_low_early | 16.9% | percent of respondents who reported frequent mental distress (14 or more of the past 30 days with mental health not good) | design-based weighted share of respondents, logit interval from Taylor linearization | adults 18 to 24 in households below the lowest income cutoff | 2014 to 2016 | 95% CI 16.0% to 17.9% | 13,932 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| young_inc_top_late | 21.1% | percent of respondents who reported frequent mental distress (14 or more of the past 30 days with mental health not good) | design-based weighted share of respondents, logit interval from Taylor linearization | adults 18 to 24 in households at or above the top income cutoff | 2022 to 2024 | 95% CI 20.0% to 22.2% | 17,893 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| young_late_all | 23.4% | percent of respondents who reported frequent mental distress (14 or more of the past 30 days with mental health not good) | design-based weighted share of respondents, logit interval from Taylor linearization | adults 18 to 24 in the 50 states and DC | 2022 to 2024 | 95% CI 22.9% to 23.9% | 79,822 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |
| young_late_coded | 20.6% | percent of respondents who reported frequent mental distress (14 or more of the past 30 days with mental health not good) | design-based weighted share of respondents, logit interval from Taylor linearization | adults 18 to 24 with a metro status code | 2022 to 2024 | 95% CI 16.2% to 25.9% | 942 respondents | BRFSS final weight with stratum and primary sampling unit (design-based) | BRFSS |   |

## Definitions

- Survey weights: Numbers attached to survey respondents that say how many people in the population each one represents, so that results describe the whole population and not only the people who answered.

## Methods

These estimates compare adults in the 2014 to 2024 BRFSS public-use files on frequent mental distress and on ever having been told of a depressive disorder, by age, sex, household income, education and metro status, in three periods: 2014 to 2016, 2017 to 2019 and 2022 to 2024. Every estimate is design-based, using the final weight, the stratum and the primary sampling unit, and the survey is a repeated cross-section: each year draws new respondents and nobody is followed. Some state-years have no respondents in the public files, metro status is missing for most respondents in recent years, and the sex item changed over the period, so comparisons of women and men between 2014 to 2016 and 2022 to 2024 compare groups defined slightly differently.

### Who is in the sample and which years are covered

Respondents are adults in the 50 states and the District of Columbia in the 2014 to 2024 BRFSS public-use files; Guam, Puerto Rico and the Virgin Islands are excluded, as in the chapter Health behaviors, person by person. Both questions were asked in every year.

The three periods are 2014 to 2016, 2017 to 2019 and 2022 to 2024, which leaves out the pandemic years 2020 and 2021. Every cell has at least 30 respondents.

New Jersey in 2019, Florida in 2021, Kentucky and Pennsylvania in 2023 and Tennessee in 2024 have no respondents in the public files; estimates limited to the states present in every year are within the margin stated in the text.

The survey is a repeated cross-section: each year draws new respondents and nobody is followed.

Explains Table 1.

### How the measures and groups are defined

Frequent mental distress is 14 or more of the past 30 days with mental health not good (88 means none and counts as 0 days; 77 and 99 are missing). A depression diagnosis is a yes to having ever been told by a doctor, nurse or other health professional of a depressive disorder (7 and 9 are missing).

Age groups are built from the five-year BRFSS age categories; household income is collapsed into five brackets (under $20,000, $20,000 to $34,999, $35,000 to $49,999, $50,000 to $74,999 and $75,000 or more, in current dollars) and education into four; metro status is the BRFSS code (center city, outside the center city in its county, suburban county, not in a metro area), which is missing for most respondents in recent years.

The share of adults with no valid household income is for 2022 to 2024.

The difference between the diagnosis share and the distress share is a paired difference among respondents who answered both items.

Explains Table 7 and Table 6.

### How the sex item changed

The sex item changed over the period: in the early questionnaires sex could be filled in at screening or recorded by the interviewer, later questionnaires ask whether the respondent is male or female at screening, and the latest add a question on sex at birth; the chapter's women and men are the survey's sex variable, coded male and female, so comparisons between 2014 to 2016 and 2022 to 2024 compare groups defined slightly differently.

Explains Table 4.

### How the survey weights and design enter the estimates

Every estimate is design-based: the final weight, the stratum and the primary sampling unit (each respondent is its own unit in the public file), with strata specific to a state and year and strata that hold one unit merged with the next stratum in the same state and year.

A subgroup is a domain of the whole design, so every cell keeps the strata and units of the full survey.

BRFSS has used raking and included cell phone respondents since 2011, so all years in the file use one weighting method.

### How uncertainty and differences are computed

Variances come from Taylor linearization and intervals are 95% intervals on the design degrees of freedom, logit intervals for shares and normal intervals for differences.

A difference between two groups in one period uses a linearized difference with a shared design; a difference between two periods combines two independent estimates, because survey years are drawn separately, and the change in a gap is the difference of two such differences.

Explains Table 2 and Table 5.

## What these data cannot show

The BRFSS is a repeated cross-section. The youngest adults in 2022 are a different set of people from those in 2014, and the data cannot separate age, survey year and birth cohort. They also cannot say whether a fall from 2022 to 2024 will continue.

Frequent mental distress is self-reported. It reflects how respondents experience the past month and how willing they are to say so, and willingness may have changed more among the young than among the old. The BRFSS reaches adults who live in households and answer a telephone call, so it misses people in hospitals, prisons and other institutions and people without a stable phone, and nonresponse can change over time in ways the weights may not fully correct. The state of residence is known and no smaller area is, so nothing here describes a neighborhood or a county.

Several comparisons rest on thin or imperfect measures. Household income is missing for a large part of the youngest adults and describes a household, not the respondent. Education among adults 18 to 24 is unfinished. Metro status is coded for a shrinking part of the sample, and the sex item changed wording over the period. The 2020 and 2021 interviews took place during the pandemic and are shown by year but left out of the three periods. Some states are absent in some years. Intervals describe sampling variation under the survey design and do not carry any bias from nonresponse or measurement.

The chapter makes no causal claim outside the study in the box above. Differences between groups, such as the difference between women and men or between income brackets, are patterns in who reports distress, and the data do not identify what produces them.

## References

1. Centers for Disease Control and Prevention, Disability and Health Data System (2025). Data Guides: Mental and Emotional Health. CDC Disability and Health Data System web page, last reviewed April 3, 2025. https://www.cdc.gov/dhds/data-guides/mental-and-emotional-health.html
2. Zahran, H. S.; Kobau, R.; Moriarty, D. G.; Zack, M. M.; Giles, W. H.; Lando, J. (2004). Self-reported frequent mental distress among adults, United States, 1993-2001. Morbidity and Mortality Weekly Report 53(41): 963-966. https://www.cdc.gov/mmwr/preview/mmwrhtml/mm5341a1.htm
3. Binder, David A. (1983). On the variances of asymptotically normal estimators from complex surveys. International Statistical Review 51(3): 279-292. https://doi.org/10.2307/1402588
4. Centers for Disease Control and Prevention (2025). 2024 BRFSS Survey Data and Documentation. Behavioral Risk Factor Surveillance System, CDC web page, last reviewed September 17, 2025. https://www.cdc.gov/brfss/annual_data/annual_2024.html
5. Blanchflower, David G.; Bryson, Alex; Xu, Xiaowei (2025). The declining mental health of the young and the global disappearance of the unhappiness hump shape in age. PLOS ONE 20(8): e0327858. https://doi.org/10.1371/journal.pone.0327858
6. Twenge, Jean M.; Cooper, A. Bell; Joiner, Thomas E.; Duffy, Mary E.; Binau, Sarah G. (2019). Age, period, and cohort trends in mood disorder indicators and suicide-related outcomes in a nationally representative dataset, 2005-2017. Journal of Abnormal Psychology 128(3): 185-199. https://doi.org/10.1037/abn0000410
7. Braghieri, Luca; Levy, Ro'ee; Makarin, Alexey (2022). Social media and mental health. American Economic Review 112(11): 3660-3693. https://doi.org/10.1257/aer.20211218

## Provenance

- Pipeline version: 0.1.0
- Build script: book2/chapters/mental-distress-in-young-adults/build.py (sha256 f02f44aa6c555f8239b835276a342fd70113c3c2fef29ca788d0b1445725b292)
- Built on: 2026-10-07
- Site commit: c2b1ff3309cb4a10833f884da91ac0057a66ad0a

Inputs:

| File | Size (bytes) | SHA-256, first 1 MB | SHA-256, last 1 MB |
| --- | ---: | --- | --- |
| book2/chapters/mental-distress-in-young-adults/mental_data.py | 11,416 | 4f8615a02246abc050849ee8b67fb02a86778b5e6f4eec851007a95a9cd34016 | 4f8615a02246abc050849ee8b67fb02a86778b5e6f4eec851007a95a9cd34016 |
| data/raw/brfss/brfss-micro-meta.json | 88,972 | 6dc744f4f0537ba682baa1d594e3d8d1f993cd0a70eb1928d7683cb23d14b30d | 6dc744f4f0537ba682baa1d594e3d8d1f993cd0a70eb1928d7683cb23d14b30d |
| data/raw/brfss/brfss-micro.parquet | 215,585,473 | 5cb7fcfe4caeaf86869cf7b50ad5b67114d491512c29004891215967b137b4c3 | 1ee9a472beb0fdb50a7b1263ed6e0f1e86040566c3bff006a7d1eaa0e391d0fa |

Outputs:

| File | SHA-256 |
| --- | --- |
| facts.json | 2d83e0c39ac35fbe9835817da08ecd7beccaf7e8b92ea9186555fb38c7ea0cf6 |
| figures/fmd_age_by_income.json | a667540d59b049480d15bcdd895381e87b80afa04592977b9bf14df24f54a336 |
| figures/fmd_age_profile.json | ff66c06d1d1f1c9c41062efe474d7089dccaa7fda15e07393ffac17574134bb1 |
| figures/fmd_by_year_age.json | 0ba165e0766c3e50efda784c26a749af131a42d9552170deb7dc6f4960404bb6 |
| figures/fmd_vs_diagnosis.json | b9e7ae678cf7a1d32b65bcf10cdd3dbda282c62c1921f5308cf607bb5a6d545b |
| figures/sex_gap_by_age.json | 6d446a231fe4ef3d8231633fb099bcfdc415985dedbef5efaba93969c6e97220 |
| figures/young_minus_midlife.json | 17a12e90f7ac8ad647c37e01e0c21ba5361f263dd2c026f07720d47e1419002c |
| figures/young_rise_by_group.json | b4d749ffce24a26a1aaaa2dc1c5520a94448346f5b89734f51f19ab32ed80a6f |
| meta.json | dd7b876f051d12ffb933e672899d21313300870f19b9cbda4faf042e5615367c |
| refs.json | 9d2faaa062d750052a7195717554fcdb1e22bab37da017f4fcbbad0d990b7788 |

## Machine formats

- Report (HTML): https://vishalsingh.org/zipcode/mental-distress-in-young-adults/report
- Report (Markdown): https://vishalsingh.org/zipcode/mental-distress-in-young-adults/report.md
- Every estimate (JSON): https://vishalsingh.org/zipcode/report/mental-distress-in-young-adults/facts.json
- Table 1, figure "fmd_by_year_age" (JSON): https://vishalsingh.org/zipcode/report/mental-distress-in-young-adults/figures/fmd_by_year_age.json
- Table 2, figure "young_minus_midlife" (JSON): https://vishalsingh.org/zipcode/report/mental-distress-in-young-adults/figures/young_minus_midlife.json
- Table 3, figure "fmd_age_profile" (JSON): https://vishalsingh.org/zipcode/report/mental-distress-in-young-adults/figures/fmd_age_profile.json
- Table 4, figure "sex_gap_by_age" (JSON): https://vishalsingh.org/zipcode/report/mental-distress-in-young-adults/figures/sex_gap_by_age.json
- Table 5, figure "young_rise_by_group" (JSON): https://vishalsingh.org/zipcode/report/mental-distress-in-young-adults/figures/young_rise_by_group.json
- Table 6, figure "fmd_age_by_income" (JSON): https://vishalsingh.org/zipcode/report/mental-distress-in-young-adults/figures/fmd_age_by_income.json
- Table 7, figure "fmd_vs_diagnosis" (JSON): https://vishalsingh.org/zipcode/report/mental-distress-in-young-adults/figures/fmd_vs_diagnosis.json
- Index of every report: https://vishalsingh.org/zipcode/llms.txt; all reports in one file: https://vishalsingh.org/zipcode/llms-full.txt

---

# Work, wages and access to jobs: report

Chapter 5.1 of ZIP Code Destiny, by Vishal Singh (NYU Stern). Generated from the chapter's data files; the reading version is at https://vishalsingh.org/zipcode/work-wages-and-access-to-jobs. Table n of this report is figure n of the chapter.

## Identification

- Title: Work, wages and access to jobs
- Label: Jobs here, workers there
- Chapter: 5.1 (Part V, Home, work and money; An American life)
- Question: How do employment, wages and access to jobs differ across counties and neighborhoods, and how far do residents' earnings differ from the wages that employers in their county pay?
- Dek: In 2019, in the fifth of counties with the fewest local jobs per job held by residents, 45.4% of residents' jobs were in the top earnings band, against 39.5% of the jobs located in those counties.
- Grain: counties
- Status: published
- Published: 2026-10-09
- Updated: not revised
- Data sources:
  - BLS Local Area Unemployment Statistics. County annual-average unemployment, 2000 to 2024 (map: 2023); residents by county of residence; years: 1990-2024; weights: labor force of each county for pooled rates; each county once for distributions
  - BLS Quarterly Census of Employment and Wages. County total-covered employment and wages, 2019 and 2023; jobs by county of the workplace; years: 1990-2024; weights: none (full counts of covered payroll); jobs in each county for pooled pay
  - LEHD Origin-Destination Employment Statistics. Residence, workplace and origin-destination tables, 2010, 2015 and 2019 (Alaska and Massachusetts left out); years: 2002-2022; weights: jobs in each table; bootstrap over states for counties and over counties for tracts
  - American Community Survey. 2019-2023 tract unemployment, labor force participation, commute and work-from-home shares; years: 2019-2023 (5-year); weights: 2019-2023 tract population
  - Area Deprivation Index. Area Deprivation Index 2023, national rank by tract; years: 2023 release (version 4.0.1; built on ACS 2019-2023); weights: none (index); tract values are unweighted means of block-group ranks
  - Rural-Urban Continuum Codes. 2023 Rural-Urban Continuum Codes for county types and 2020 county population for correlation weights; years: 2023; weights: 2020 county population
- Canonical URL: https://vishalsingh.org/zipcode/work-wages-and-access-to-jobs
- Report: https://vishalsingh.org/zipcode/work-wages-and-access-to-jobs/report
- Markdown: https://vishalsingh.org/zipcode/work-wages-and-access-to-jobs/report.md
- BibTeX key: singh2026workwagesandaccesstojobs

```bibtex
@misc{singh2026workwagesandaccesstojobs,
  author = {Singh, Vishal},
  title = {Work, wages and access to jobs},
  year = {2026},
  url = {https://vishalsingh.org/zipcode/work-wages-and-access-to-jobs},
  note = {In ZIP Code Destiny. Published 9 October 2026.}
}
```

## Principal estimates

The facts the dek and the opening prose cite, in order of appearance.

| Estimand | Population | Period | Value | Interval | n | Weights | Source | Fact id |
| --- | --- | --- | ---: | --- | ---: | --- | --- | --- |
| jobs in the top earnings band summed over the counties of the group, divided by jobs summed over the same counties, bootstrap interval over states | jobs held by residents of the lowest fifth of counties by jobs located per job held by residents (Alaska and Massachusetts left out) | 2019 | 45.4% | 95% CI 43.4% to 47.3% | 619 counties | jobs in each table | LEHD Origin-Destination Employment Statistics | rs_lowest_fifth |
| jobs in the top earnings band summed over the counties of the group, divided by jobs summed over the same counties, bootstrap interval over states | jobs located in the lowest fifth of counties by jobs located per job held by residents (Alaska and Massachusetts left out) | 2019 | 39.5% | 95% CI 36.7% to 42.1% | 619 counties | jobs in each table | LEHD Origin-Destination Employment Statistics | ws_lowest_fifth |
| jobs located outside the home county summed over counties, divided by jobs held by residents summed over the same counties, bootstrap interval over states | jobs held by residents of counties (Alaska and Massachusetts left out) | 2019 | 45.4% | 95% CI 41.8% to 49.1% | 3,099 counties | jobs held by residents of each county | LEHD Origin-Destination Employment Statistics | out_all_2019 |

## Findings

In 2019, in the fifth of counties with the fewest jobs located per job held by residents, 45.4% of the jobs held by residents were in the top earnings band, against 39.5% of the jobs located there. The counts are from the LEHD Origin-Destination Employment Statistics (LODES), and Alaska and Massachusetts are left out of every LODES figure. In the same year 45.4% of the jobs placed at residents' homes were located in a different county.

The jobs a county's residents hold are therefore not the jobs located in the county, and this chapter keeps the two apart. Unemployment is counted among the people who live in a county, pay is counted for the jobs located there, and the earnings that residents bring home come from jobs wherever they are.

The data are counts for counties and tracts, so every claim describes places and not the workers inside them (ecological inference). The chapter shows who is counted in each source before it compares them, and it names the group, residents or jobs located, with every number.

### What each source counts

Local Area Unemployment Statistics (LAUS) counts people by where they live. Its county figures are annual averages of monthly estimates that the Bureau of Labor Statistics builds from payroll counts, unemployment insurance claims and survey data and controls to state totals from a model.[1] The county file carries no margin of error, and the estimates for the smallest counties are the least certain. The Quarterly Census of Employment and Wages (QCEW) counts jobs by where they are located, from the wage reports of employers covered by state unemployment insurance. It leaves out the self-employed and some farm and domestic workers, counts part-time jobs as jobs, and includes bonuses in wages.[2]

LODES places each job in three tables: at the worker's home, at the workplace, and as a pair of home county and work county.[3] LODES counts jobs, not people, so a worker with two jobs is counted once for each job. It adds noise to protect confidentiality, and it reports earnings only in three fixed bands of monthly earnings in current dollars.[4] The lowest band ends at $1,250 a month, and in this chapter a job is in the top band when it pays more than $3,333 a month. The cut does not move with prices, so shares of jobs in a band are compared within one year only.

The American Community Survey (ACS) supplies unemployment, labor force participation, commuting and work-from-home measures for residents of 2020 tracts. The Area Deprivation Index (ADI) ranks those tracts from least to most deprived.[5] The ADI is built partly from tract unemployment, income, the white-collar share of employed people and the share of households without a vehicle. A gradient in unemployment, earnings, commuting or working from home against it is therefore partly mechanical, and each tract figure below says so.

### The order of county types by unemployment reversed in 2020

Across the 3,143 counties and county equivalents in the LAUS file, the pooled unemployment rate of residents in 2023 was 3.6% (95% CI 3.3% to 3.9%), each county weighted by its labor force. Counted once each, counties ran from 2.1% at the fifth percentile to 5.6% at the ninety-fifth, around a median of 3.4% (Table 1). The map shows where residents live, wherever they work.

A rate of 10% or more was published for 6 counties. The two highest published estimates were for Imperial County, California, at 17.6%, and the Kusilvak Census Area, Alaska, at 14.8%. Estimates for small counties lean hardest on the model. In 2023, 296 counties had a labor force under 2,000, and in a county that small a high rate can rest on a few hundred people.

Table 1. Unemployment among residents was 3.6% across counties in 2023, differed little between county types, and reached 10% or more in 6 counties

Unemployment differs less between county types than it moves over time. Grouping counties by the 2023 Rural-Urban Continuum Code gives three types: metropolitan counties, nonmetropolitan counties with larger towns (an urban population of 5,000 or more), and nonmetropolitan counties with smaller towns or none (Table 2). The metropolitan rate was below that of the smaller-town counties in each of the 20 years from 2000 to 2019, and the interval for the difference lies entirely below zero in 12 of them.

In 2019 the pooled rates were 3.6% (95% CI 3.4% to 3.8%) in metropolitan counties, 4.1% (95% CI 3.8% to 4.3%) in the larger-town counties and 4.3% (95% CI 3.9% to 4.7%) in the smaller-town counties. The metropolitan rate was 0.5 (95% CI 0.3 to 0.7) points below that of the larger-town counties and 0.7 (95% CI 0.4 to 1.0) points below that of the smaller-town counties.

The order reversed in 2020, when the pooled rate of all counties rose from 3.7% (95% CI 3.5% to 3.8%) in 2019 to 8.1% (95% CI 7.4% to 8.7%). Metropolitan counties then had a rate of 8.2% (95% CI 7.5% to 8.8%). That was 0.9 (95% CI 0.4 to 1.5) points above the 7.3% (95% CI 6.9% to 7.6%) of the larger-town counties and 1.5 (95% CI 0.7 to 2.2) points above the 6.7% (95% CI 6.2% to 7.3%) of the smaller-town counties.

By 2023 the three pooled rates were 3.6%, 3.7% and 3.8%. The interval for the metropolitan minus smaller-town difference, −0.1 (95% CI −0.5 to 0.2) points, includes zero. The data describe the rates and do not say why the order reversed.

The intervals come from resampling states, so they show how much each pooled rate depends on which counties and states are in a group. They do not capture the model error of the county estimates, which the county file does not carry. No annual average is published for 7 Louisiana parishes in 2005 and 2006 (14 county-years), and the pooled rates of those years leave them out.

Table 2. Metropolitan counties went from a lower unemployment rate than nonmetropolitan counties with smaller towns in 2019 (3.6% against 4.3%) to a higher one in 2020 (8.2% against 6.7%)

### What employers in a county pay

QCEW reports total wages and average employment for the jobs located in each county. In 2023, 3,141 counties had a published average annual pay per job, and together they held 148.0 million covered jobs. Counted once each, counties ran from $39,160 at the fifth percentile to $72,089 at the ninety-fifth, around a median of $49,668 (Table 3).

Weighted by jobs, the mean was $71,075. The mean sits near the metropolitan figure because metropolitan counties hold 88.9% of the covered jobs. Jobs located in metropolitan counties paid $73,585 on average, against $51,618 in nonmetropolitan counties with larger towns and $48,786 in those with smaller towns or none.

Santa Clara County and San Mateo County, California, had the two highest county averages, $174,694 and $174,319. Worth County, Missouri, with 385 jobs, had the lowest, $29,704.

Every one of these figures is the pay of jobs located in a county. It is not the earnings of the people who live there, who may work in another county, and the next section measures how far the two differ. The income of residents, as tax returns report it, is in Chapter 5.2, Home value, income and redlining (https://vishalsingh.org/zipcode/home-value-income-and-redlining).

The Bureau withholds the totals of 1 county in 2023, Eureka County, Nevada, to protect an employer's confidentiality, and the map draws it gray. The county-by-sector detail is withheld more often. In 2023 the Bureau withheld 25% of the cells for private employers, 32% for local governments and 29% for state governments. A withheld cell is missing and not zero, and the sector cells cannot be summed to rebuild a county's private pay, so this chapter uses the county totals only.

Table 3. Jobs located in metropolitan counties paid $73,585 a year on average in 2023, against $48,786 in nonmetropolitan counties with smaller towns or none

### Residents and workplaces are different groups of jobs

Commuting across county lines is common. Of the 142.0 million jobs in the 2019 LODES flows, 45.4% (95% CI 41.8% to 49.1%) were held by residents of one county and located in another. A survey of workers gives a lower share, 28% of workers traveling to a different county in the American Community Survey for 2009 to 2013.[6] LODES counts every job, second jobs included, and assigns the workplaces of employers with several sites by a model, so the two levels are not comparable.

The share rises as counties get smaller (Table 4). It was 44.1% (95% CI 40.3% to 48.3%) for residents of metropolitan counties, 51.5% (95% CI 49.1% to 53.4%) for the nonmetropolitan counties with larger towns and 62.4% (95% CI 59.5% to 64.8%) for those with smaller towns or none. The larger-town counties exceed the metropolitan counties by 7.4 (95% CI 2.7 to 10.9) points. The smaller-town counties exceed the larger-town counties by 11.0 (95% CI 9.0 to 12.9) points and the metropolitan counties by 18.4 (95% CI 13.7 to 22.0) points.

The share rose between 2010 and 2019 in each type, over the same states in every year. Across all counties it went from 44.1% to 45.4%, a rise of 1.3 (95% CI 0.9 to 1.6) points. In metropolitan counties it went from 43.0% to 44.1%, a rise of 1.1 (95% CI 0.8 to 1.5) points. In larger-town counties it went from 48.6% to 51.5%, a rise of 2.9 (95% CI 2.4 to 3.3) points, and in smaller-town counties from 59.5% to 62.4%, a rise of 3.0 (95% CI 2.5 to 3.4) points.

Table 4. In 2019, 62.4% of the jobs held by residents of nonmetropolitan counties with smaller towns or none were in another county, against 44.1% for residents of metropolitan counties

Commuting separates the jobs located in a county from the jobs its residents hold, and the two differ in what they pay. LODES reports both in the same earnings bands, so the share of jobs in the top band can be compared directly. The 3,099 counties are ranked by jobs located per job held by residents and cut into fifths. In the fifth with the fewest jobs located per resident job (at most 0.60), 45.4% (95% CI 43.4% to 47.3%) of jobs held by residents were in the top band, against 39.5% (95% CI 36.7% to 42.1%) of jobs located there (Table 5). The difference is 5.9 (95% CI 3.2 to 7.8) points.

The order reversed in the fifth with the most jobs per resident job (at least 1.01). There 47.8% (95% CI 46.0% to 49.3%) of residents' jobs were in the top band against 49.8% (95% CI 47.8% to 51.5%) of jobs located, a difference of 2.0 (95% CI 1.2 to 2.9) points in the other direction.

Table 5. In the fifth of counties with the fewest jobs per resident job, 45.4% of jobs held by residents were in the top earnings band against 39.5% of the jobs located there

The lowest and highest fifths hold different shares of the jobs and different kinds of county. The fifth with the fewest jobs located per resident job holds 6.3% of the jobs held by residents and 3.2% of the jobs located. Of its counties, 47.7% are metropolitan and 43.9% are nonmetropolitan with smaller towns or none.

The fifth with the most jobs per resident job holds 46.0% and 56.5%, and 52.3% of its counties are metropolitan. Moving the county at a cut to the next fifth changes any of these shares by up to 0.1 points.

Jobs located vary more across these fifths than jobs held by residents do. From the lowest to the highest fifth the top-band share of jobs located rises by 10.3 (95% CI 8.5 to 12.5) points and the share among jobs held by residents by 2.4 (95% CI 0.5 to 4.5) points, a difference of 7.9 (95% CI 5.7 to 9.7) points. The jobs that residents hold differ less across places than the jobs located in those places.

County by county, the top-band share of jobs held by residents exceeded that of jobs located in 66.0% (95% CI 63.8% to 68.3%) of the 2,877 counties with at least 1,000 jobs in each table (Table 6). Noise in the small-county shares pulls dots apart in both directions, so the share above the diagonal is the figure to read, not the slope of the cloud.

Table 6. In 66.0% of counties the jobs held by residents were more often in the top earnings band than the jobs located in the county

QCEW pay describes jobs located, and it tracks the workplace measure more closely than the residence measure. Across the counties with a published average pay in 2019, weighted by 2020 population, pay correlates +0.87 (95% CI 0.86 to 0.92) with the top-band share of jobs located and +0.73 (95% CI 0.69 to 0.77) with the top-band share of jobs held by residents. The difference is +0.15 (95% CI 0.11 to 0.21). That is consistent with the pay in a county describing its jobs and not its residents' earnings. Part of the contrast may come from a shared source, since QCEW and the LODES workplace table both draw on state unemployment insurance wage records. The data give residents' earnings only as three bands, so they cannot say how many dollars residents bring home. How earnings and income move across generations is the subject of Chapter 4.3, Race, income and mobility (https://vishalsingh.org/zipcode/race-income-and-mobility).

### Residents and jobs in neighborhoods of different deprivation

The ACS and the residence table of LODES describe the same tracts. The ADI orders every tract in the country, so the tenths and fifths below are national groups of tracts. Table 7 compares the least deprived and the most deprived tenth of the 81,038 tracts with a rank. The ACS measures cover people aged 16 and over.

Unemployment among residents was 4.9% (95% CI 4.3% to 5.4%) in the least deprived tenth and 8.6% (95% CI 8.2% to 9.0%) in the most deprived. Labor force participation was 66.1% (95% CI 65.4% to 66.7%) and 54.8% (95% CI 54.3% to 55.3%). The share of workers who work from home was 23.4% (95% CI 22.0% to 25.3%) and 5.9% (95% CI 5.7% to 6.2%).

Among commuting workers, those who do not work from home, the share with a commute of an hour or more was 12.0% (95% CI 10.1% to 14.3%) in the least deprived tenth and 7.3% (95% CI 6.9% to 7.7%) in the most deprived. Long commutes are more common where deprivation is lower. Among jobs held by residents, 60.1% (95% CI 57.1% to 63.3%) were in the top band in the least deprived tenth and 30.4% (95% CI 29.9% to 30.9%) in the most deprived. The index includes several of these measures, as the first section notes, so every gradient here overlaps with it and is partly mechanical.

Table 7. In the most deprived tenth of tracts, 30.4% of jobs held by residents were in the top earnings band against 60.1% in the least deprived tenth, and unemployment was 8.6% against 4.9%

The jobs located in a tract can differ from the jobs its residents hold, and the difference depends on the county (Table 8). In metropolitan counties the top-band share of jobs held by residents fell from 57.5% (95% CI 54.5% to 60.5%) in the least deprived fifth of tracts to 32.6% (95% CI 32.1% to 33.2%) in the most deprived. The share of jobs located in the same tracts fell from 53.3% (95% CI 50.0% to 56.5%) to 43.7% (95% CI 42.6% to 44.6%).

In the most deprived fifth, the gap between the top-band share of jobs located and that of jobs held by residents was +11.0 (95% CI 10.1 to 11.9) points in metropolitan counties and +1.9 (95% CI 1.4 to 2.4) points in nonmetropolitan counties with larger towns. In those with smaller towns or none it was −0.2 (95% CI −0.8 to 0.3) points, an interval that includes zero. Because the index includes tract income and employment, the fall in the share of jobs held by residents across the fifths is partly mechanical. The staged data have no tract-to-tract flows, so they cannot say whether the better-paid jobs in deprived metropolitan tracts are held by the tracts' residents or by people who commute in.

Table 8. In the most deprived fifth of tracts in metropolitan counties, 43.7% of jobs located there were in the top earnings band against 32.6% of jobs held by the tracts' residents

Whether residents of poor neighborhoods can reach the jobs that match their skills has a long history in economics under the name of spatial mismatch. The idea began with a study of housing segregation and the suburbanization of employment.[7] This chapter's tract figures show where jobs and workers are by tract. They do not measure travel time, transit or the matching of skills to openings, and they cannot test that hypothesis. The earnings that college graduates reach are in Chapter 4.1, College access, price and earnings (https://vishalsingh.org/zipcode/college-access-price-and-earnings).

### What these data cannot show

All the patterns here are descriptions of counties and tracts. The pooled rate of a county type, the share of jobs in the top band for a group of counties and the gradient across groups of tracts say nothing certain about any worker.

Unemployment and pay come from different groups. LAUS and the ACS count residents, QCEW counts jobs located in a county, and LODES counts jobs at either end of the commute, so no figure here gives the pay of a county's residents in dollars. LODES shows only three bands. A fixed band in current dollars holds more jobs each year as prices and pay rise, so the shares apply to 2019 and say nothing about change over time. The same dollar cut also buys different amounts in different places, so top-band shares are not comparisons of purchasing power. Counting jobs and not people means a worker with two jobs is counted once for each, and noise added for confidentiality makes the shares of small counties and tracts imprecise.

The county unemployment estimates are modeled, and the county file carries no margin of error. The intervals here reflect which places are in each group, and a high rate in a county with a small labor force can rest on few people. Connecticut is drawn as its legacy counties, and the unemployment values there are spread from planning regions by population, so they are approximate. Puerto Rico's 78 municipios are in the LAUS file and are not drawn. The ACS tract file has no margins of error. A cell of county type by fifth of tracts is drawn only with at least 30 tracts, which leaves out the least deprived fifth of the nonmetropolitan counties with smaller towns or none. The ADI averages block-group ranks without population weights, leaves 1,580 tracts without a rank, and understates deprivation in high-cost cities. Jobs in tracts with no rank are left out of the tract figures: 3.1% of jobs located and 0.3% of jobs held by residents. A tract has no rank when every block group in it is suppressed, for example for group quarters or small populations.

LODES is incomplete for some state-years, and the residence table is almost empty where a state has no workplace table or only a stub. Alaska has no workplace file for 2019, and its residence table holds 3,090 jobs for that year against 308,964 in 2015. The Massachusetts workplace file for 2010 is a stub of 2 tracts and 17 jobs, and its residence table holds 151,718 jobs against 3,330,566 in 2015. A stub reads as a complete file unless the jobs are compared, so the build compares workplace jobs with residence jobs. Alaska and Massachusetts are left out of every LODES figure, so every year of the commuting trend covers the same states. The 2022 files lack Alaska and Michigan, so 2019 is the year used for residents against workplaces. Puerto Rico is not drawn and is not in the LODES tables used here.

The comparison of residents with workplaces is for counties and tracts. It cannot say who holds the better-paid jobs in a deprived neighborhood, whether commuting costs or hiring practices keep residents from them, or how many workers would take them if they could.

## Tables

### Table 1. Unemployment among residents was 3.6% across counties in 2023, differed little between county types, and reached 10% or more in 6 counties

Unemployment rate, percent of the civilian labor force, annual average 2023, residents of each county, modeled estimates

No inline rows: the values of Table 1 are in https://vishalsingh.org/zipcode/data/work-wages-and-access-to-jobs/unemployment_map.json.

Note: LAUS county estimates are built from payroll counts, unemployment insurance claims and survey data and are controlled to state totals from a model; the county file carries no margin of error. 296 counties have a labor force below 2,000, and their estimates are the least certain.

Note: LAUS codes Connecticut by planning region. The map draws the eight legacy counties, with the counts of unemployed residents and the labor force of each planning region spread over them in proportion to 2020 population, so the Connecticut values are approximate. Puerto Rico is not drawn; its 78 municipios are in the file.

Note: The tooltip gives the labor force of each county.

Source: Bureau of Labor Statistics, Local Area Unemployment Statistics, county annual averages, 2023; modeled estimates, not weighted

Data: https://vishalsingh.org/zipcode/report/work-wages-and-access-to-jobs/figures/unemployment_map.json

### Table 2. Metropolitan counties went from a lower unemployment rate than nonmetropolitan counties with smaller towns in 2019 (3.6% against 4.3%) to a higher one in 2020 (8.2% against 6.7%)

Unemployment rate, percent of the civilian labor force, pooled over the counties of each type, annual averages 2000 to 2024, residents of each county, modeled estimates

| Year | County type | Pooled rate (percent) | Lower limit of the 95% interval | Upper limit of the 95% interval | Counties |
| ---: | --- | ---: | ---: | ---: | ---: |
| 2000 | All counties | 3.99 | 3.67 | 4.27 | 3,140 |
| 2000 | Metropolitan | 3.89 | 3.51 | 4.20 | 1,185 |
| 2000 | Nonmetropolitan, larger towns | 4.52 | 4.22 | 4.80 | 903 |
| 2000 | Nonmetropolitan, smaller towns or none | 4.81 | 4.43 | 5.23 | 1,052 |
| 2001 | All counties | 4.74 | 4.47 | 4.96 | 3,140 |
| 2001 | Metropolitan | 4.63 | 4.31 | 4.89 | 1,185 |
| 2001 | Nonmetropolitan, larger towns | 5.28 | 4.96 | 5.62 | 903 |
| 2001 | Nonmetropolitan, smaller towns or none | 5.54 | 5.09 | 6.00 | 1,052 |
| 2002 | All counties | 5.79 | 5.45 | 6.06 | 3,140 |
| 2002 | Metropolitan | 5.75 | 5.34 | 6.06 | 1,185 |
| 2002 | Nonmetropolitan, larger towns | 6.00 | 5.65 | 6.32 | 903 |
| 2002 | Nonmetropolitan, smaller towns or none | 6.16 | 5.70 | 6.65 | 1,052 |
| 2003 | All counties | 5.99 | 5.61 | 6.27 | 3,140 |
| 2003 | Metropolitan | 5.93 | 5.51 | 6.24 | 1,185 |
| 2003 | Nonmetropolitan, larger towns | 6.29 | 5.95 | 6.60 | 903 |
| 2003 | Nonmetropolitan, smaller towns or none | 6.42 | 5.96 | 6.87 | 1,052 |
| 2004 | All counties | 5.53 | 5.19 | 5.78 | 3,140 |
| 2004 | Metropolitan | 5.45 | 5.09 | 5.72 | 1,185 |
| 2004 | Nonmetropolitan, larger towns | 5.94 | 5.63 | 6.24 | 903 |
| 2004 | Nonmetropolitan, smaller towns or none | 6.03 | 5.61 | 6.47 | 1,052 |
| 2005 | All counties | 5.10 | 4.81 | 5.35 | 3,133 |
| 2005 | Metropolitan | 5.00 | 4.69 | 5.25 | 1,178 |
| 2005 | Nonmetropolitan, larger towns | 5.63 | 5.28 | 5.93 | 903 |
| 2005 | Nonmetropolitan, smaller towns or none | 5.83 | 5.38 | 6.29 | 1,052 |
| 2006 | All counties | 4.62 | 4.34 | 4.88 | 3,133 |
| 2006 | Metropolitan | 4.53 | 4.22 | 4.77 | 1,178 |
| 2006 | Nonmetropolitan, larger towns | 5.13 | 4.79 | 5.44 | 903 |
| 2006 | Nonmetropolitan, smaller towns or none | 5.35 | 4.89 | 5.83 | 1,052 |
| 2007 | All counties | 4.62 | 4.32 | 4.90 | 3,140 |
| 2007 | Metropolitan | 4.53 | 4.21 | 4.82 | 1,185 |
| 2007 | Nonmetropolitan, larger towns | 5.09 | 4.74 | 5.43 | 903 |
| 2007 | Nonmetropolitan, smaller towns or none | 5.34 | 4.84 | 5.84 | 1,052 |
| 2008 | All counties | 5.78 | 5.34 | 6.22 | 3,140 |
| 2008 | Metropolitan | 5.72 | 5.24 | 6.19 | 1,185 |
| 2008 | Nonmetropolitan, larger towns | 6.08 | 5.62 | 6.51 | 903 |
| 2008 | Nonmetropolitan, smaller towns or none | 6.26 | 5.65 | 6.87 | 1,052 |
| 2009 | All counties | 9.25 | 8.54 | 9.96 | 3,140 |
| 2009 | Metropolitan | 9.18 | 8.42 | 9.92 | 1,185 |
| 2009 | Nonmetropolitan, larger towns | 9.66 | 8.91 | 10.36 | 903 |
| 2009 | Nonmetropolitan, smaller towns or none | 9.62 | 8.65 | 10.55 | 1,052 |
| 2010 | All counties | 9.63 | 8.87 | 10.38 | 3,142 |
| 2010 | Metropolitan | 9.60 | 8.77 | 10.40 | 1,185 |
| 2010 | Nonmetropolitan, larger towns | 9.84 | 9.19 | 10.46 | 903 |
| 2010 | Nonmetropolitan, smaller towns or none | 9.79 | 8.83 | 10.69 | 1,054 |
| 2011 | All counties | 8.95 | 8.24 | 9.68 | 3,142 |
| 2011 | Metropolitan | 8.93 | 8.15 | 9.71 | 1,185 |
| 2011 | Nonmetropolitan, larger towns | 9.08 | 8.48 | 9.64 | 903 |
| 2011 | Nonmetropolitan, smaller towns or none | 9.08 | 8.23 | 9.88 | 1,054 |
| 2012 | All counties | 8.07 | 7.42 | 8.70 | 3,142 |
| 2012 | Metropolitan | 8.05 | 7.33 | 8.72 | 1,185 |
| 2012 | Nonmetropolitan, larger towns | 8.20 | 7.65 | 8.73 | 903 |
| 2012 | Nonmetropolitan, smaller towns or none | 8.19 | 7.39 | 8.95 | 1,054 |
| 2013 | All counties | 7.38 | 6.87 | 7.86 | 3,142 |
| 2013 | Metropolitan | 7.33 | 6.79 | 7.86 | 1,185 |
| 2013 | Nonmetropolitan, larger towns | 7.65 | 7.17 | 8.07 | 903 |
| 2013 | Nonmetropolitan, smaller towns or none | 7.77 | 7.06 | 8.47 | 1,054 |
| 2014 | All counties | 6.17 | 5.76 | 6.58 | 3,142 |
| 2014 | Metropolitan | 6.13 | 5.69 | 6.55 | 1,185 |
| 2014 | Nonmetropolitan, larger towns | 6.37 | 5.99 | 6.74 | 903 |
| 2014 | Nonmetropolitan, smaller towns or none | 6.59 | 5.99 | 7.20 | 1,054 |
| 2015 | All counties | 5.28 | 4.97 | 5.57 | 3,142 |
| 2015 | Metropolitan | 5.22 | 4.88 | 5.53 | 1,185 |
| 2015 | Nonmetropolitan, larger towns | 5.61 | 5.28 | 5.94 | 903 |
| 2015 | Nonmetropolitan, smaller towns or none | 5.85 | 5.36 | 6.35 | 1,054 |
| 2016 | All counties | 4.87 | 4.62 | 5.08 | 3,142 |
| 2016 | Metropolitan | 4.79 | 4.51 | 5.02 | 1,185 |
| 2016 | Nonmetropolitan, larger towns | 5.36 | 5.03 | 5.64 | 903 |
| 2016 | Nonmetropolitan, smaller towns or none | 5.67 | 5.25 | 6.09 | 1,054 |
| 2017 | All counties | 4.36 | 4.13 | 4.53 | 3,142 |
| 2017 | Metropolitan | 4.30 | 4.07 | 4.48 | 1,185 |
| 2017 | Nonmetropolitan, larger towns | 4.69 | 4.38 | 4.96 | 903 |
| 2017 | Nonmetropolitan, smaller towns or none | 4.92 | 4.54 | 5.30 | 1,054 |
| 2018 | All counties | 3.90 | 3.71 | 4.05 | 3,142 |
| 2018 | Metropolitan | 3.84 | 3.65 | 3.99 | 1,185 |
| 2018 | Nonmetropolitan, larger towns | 4.21 | 3.95 | 4.45 | 903 |
| 2018 | Nonmetropolitan, smaller towns or none | 4.39 | 4.05 | 4.75 | 1,054 |
| 2019 | All counties | 3.67 | 3.47 | 3.84 | 3,142 |
| 2019 | Metropolitan | 3.60 | 3.39 | 3.78 | 1,185 |
| 2019 | Nonmetropolitan, larger towns | 4.06 | 3.81 | 4.31 | 903 |
| 2019 | Nonmetropolitan, smaller towns or none | 4.29 | 3.94 | 4.66 | 1,054 |
| 2020 | All counties | 8.07 | 7.42 | 8.66 | 3,143 |
| 2020 | Metropolitan | 8.21 | 7.51 | 8.81 | 1,185 |
| 2020 | Nonmetropolitan, larger towns | 7.26 | 6.86 | 7.63 | 903 |
| 2020 | Nonmetropolitan, smaller towns or none | 6.74 | 6.19 | 7.28 | 1,055 |
| 2021 | All counties | 5.36 | 4.78 | 5.88 | 3,143 |
| 2021 | Metropolitan | 5.43 | 4.80 | 5.98 | 1,185 |
| 2021 | Nonmetropolitan, larger towns | 4.88 | 4.52 | 5.24 | 903 |
| 2021 | Nonmetropolitan, smaller towns or none | 4.69 | 4.33 | 5.07 | 1,055 |
| 2022 | All counties | 3.65 | 3.39 | 3.87 | 3,143 |
| 2022 | Metropolitan | 3.64 | 3.35 | 3.87 | 1,185 |
| 2022 | Nonmetropolitan, larger towns | 3.73 | 3.47 | 3.97 | 903 |
| 2022 | Nonmetropolitan, smaller towns or none | 3.75 | 3.46 | 4.06 | 1,055 |
| 2023 | All counties | 3.64 | 3.33 | 3.92 | 3,143 |
| 2023 | Metropolitan | 3.63 | 3.28 | 3.94 | 1,185 |
| 2023 | Nonmetropolitan, larger towns | 3.66 | 3.43 | 3.88 | 903 |
| 2023 | Nonmetropolitan, smaller towns or none | 3.75 | 3.48 | 4.03 | 1,055 |
| 2024 | All counties | 4.02 | 3.69 | 4.35 | 3,143 |
| 2024 | Metropolitan | 4.03 | 3.67 | 4.38 | 1,185 |
| 2024 | Nonmetropolitan, larger towns | 3.98 | 3.75 | 4.22 | 903 |
| 2024 | Nonmetropolitan, smaller towns or none | 4.07 | 3.77 | 4.40 | 1,055 |

Note: A county type follows the 2023 Rural-Urban Continuum Code of the county: codes 1 to 3 are metropolitan, 4 to 7 are nonmetropolitan with an urban population of 5,000 or more, and 8 and 9 are nonmetropolitan with an urban population under 5,000. One classification is applied to every year.

Note: The intervals reflect which counties and states are in each group, not the model error of the county estimates, which LAUS does not publish. The tooltips give the interval for every line.

Note: No annual average is published for 7 Louisiana parishes in 2005 and 2006, which leaves them out of the pooled rates of those two years. Puerto Rico is not included.

Source: Bureau of Labor Statistics, Local Area Unemployment Statistics, county annual averages, 2000-2024; modeled estimates, pooled with labor-force weights; county types from the USDA 2023 Rural-Urban Continuum Codes

Data: https://vishalsingh.org/zipcode/report/work-wages-and-access-to-jobs/figures/unemployment_by_county_type.json

### Table 3. Jobs located in metropolitan counties paid $73,585 a year on average in 2023, against $48,786 in nonmetropolitan counties with smaller towns or none

Average annual pay per job, current dollars, for jobs located in each county and covered by unemployment insurance, 2023, jobs at the workplace

No inline rows: the values of Table 3 are in https://vishalsingh.org/zipcode/data/work-wages-and-access-to-jobs/workplace_pay_map.json.

Note: Pay is total wages divided by average employment for jobs covered by state unemployment insurance. It mixes full-time and part-time jobs, includes bonuses, and leaves out the self-employed and some farm and domestic workers.

Note: Of the 2 gray counties, one has totals withheld by the Bureau to protect an employer's confidentiality and one is not in the file. A withheld cell is missing, never zero.

Note: QCEW codes Connecticut by its legacy counties through 2023, the counties drawn here. Puerto Rico is not drawn. Dollars are current, so this map compares counties within 2023 and not across years.

Note: Totals are full counts from employer reports, so they carry no sampling interval.

Source: Bureau of Labor Statistics, Quarterly Census of Employment and Wages, county annual averages, 2023; full counts of covered payroll

Data: https://vishalsingh.org/zipcode/report/work-wages-and-access-to-jobs/figures/workplace_pay_map.json

### Table 4. In 2019, 62.4% of the jobs held by residents of nonmetropolitan counties with smaller towns or none were in another county, against 44.1% for residents of metropolitan counties

Percent of jobs held by a county's residents that are located in a different county, pooled over the counties of each type, 2010, 2015 and 2019, jobs counted by home county

| conf_high | conf_low | estimate | jobs | n_counties | series | year |
| ---: | ---: | ---: | ---: | ---: | --- | ---: |
| 46.89 | 39.37 | 42.95 | 105160058 | 1,171 | Metropolitan | 2010 |
| 50.45 | 46.30 | 48.58 | 13636739 | 894 | Nonmetropolitan, larger towns | 2010 |
| 61.81 | 56.61 | 59.46 | 4091694 | 1,034 | Nonmetropolitan, smaller towns or none | 2010 |
| 47.03 | 39.24 | 42.87 | 115867165 | 1,171 | Metropolitan | 2015 |
| 51.16 | 46.90 | 49.24 | 14121233 | 894 | Nonmetropolitan, larger towns | 2015 |
| 62.95 | 57.67 | 60.51 | 4083810 | 1,034 | Nonmetropolitan, smaller towns or none | 2015 |
| 48.26 | 40.33 | 44.09 | 123293844 | 1,171 | Metropolitan | 2019 |
| 53.45 | 49.13 | 51.46 | 14490434 | 894 | Nonmetropolitan, larger towns | 2019 |
| 64.84 | 59.55 | 62.44 | 4179253 | 1,034 | Nonmetropolitan, smaller towns or none | 2019 |

Note: Alaska and Massachusetts are left out of every year, so every year covers the same states. Alaska has no workplace file for 2019 and its residence table holds 3,090 jobs, against 308,964 in 2015. The Massachusetts workplace file for 2010 is a stub of 2 tracts and 17 jobs, and its residence table holds 151,718 jobs, against 3,330,566 in 2015. Puerto Rico is not in the LODES tables used here.

Note: LODES counts every job, second jobs included, and assigns the workplaces of employers with several sites by a model. A survey of workers gives a lower share (28% of workers traveled to a different county in the 2009-2013 ACS), so the two levels are not comparable.

Source: LEHD LODES origin-destination tables, 2010, 2015 and 2019 (Alaska and Massachusetts left out); jobs-weighted counts with confidentiality noise; county types from the USDA 2023 Rural-Urban Continuum Codes

Data: https://vishalsingh.org/zipcode/report/work-wages-and-access-to-jobs/figures/commuting_out_of_county.json

### Table 5. In the fifth of counties with the fewest jobs per resident job, 45.4% of jobs held by residents were in the top earnings band against 39.5% of the jobs located there

Percent of jobs paying more than $3,333 a month, jobs held by a county's residents against jobs located in the county, fifths of counties ranked by jobs located per job held by residents, 2019

| Fifth of counties | Counties | Jobs located per job held by residents, lowest | Jobs located per job held by residents, highest | Top-band share, jobs located (percent) | Top-band share, jobs held by residents (percent) | Jobs held by residents | Jobs located |
| --- | ---: | ---: | ---: | ---: | ---: | ---: | ---: |
| Fewest jobs per resident job | 619 | 0.14 | 0.60 | 39.5 | 45.4 | 8,939,569 | 4,609,391 |
| Second fifth | 620 | 0.60 | 0.74 | 41.6 | 46.4 | 15,095,852 | 10,201,971 |
| Middle fifth | 620 | 0.74 | 0.86 | 43.0 | 46.7 | 20,101,609 | 16,046,376 |
| Fourth fifth | 620 | 0.86 | 1.01 | 45.3 | 46.9 | 32,659,196 | 30,931,375 |
| Most jobs per resident job | 620 | 1.01 | 11.24 | 49.8 | 47.8 | 65,464,789 | 80,367,572 |

Note: The same jobs are counted in both tables, once at the home and once at the workplace, so the two shares agree when every county is pooled; they differ in a group of counties because of commuting across the group's boundary.

Note: Counties are ranked on jobs located per job held by residents, a count ratio that does not use earnings, and the fifths hold near-equal numbers of counties, not equal numbers of jobs. The lowest fifth holds 6.3% of jobs held by residents and 3.2% of jobs located, and the highest fifth holds 46.0% and 56.5%. Moving the county at a cut to the next fifth changes any of these shares by up to 0.1 points. The ranges of the ratio are in the table. Earnings bands are fixed in current dollars, so shares are compared within 2019 only.

Note: Alaska (no 2019 workplace file) and Massachusetts (a stub 2010 workplace file) are left out of every LODES figure, so the same states are covered in every year.

Source: LEHD LODES residence and workplace tables, 2019 (Alaska and Massachusetts left out); jobs-weighted counts with confidentiality noise, earnings bands in current dollars

Data: https://vishalsingh.org/zipcode/report/work-wages-and-access-to-jobs/figures/top_band_by_jobs_per_resident.json

### Table 6. In 66.0% of counties the jobs held by residents were more often in the top earnings band than the jobs located in the county

Percent of jobs paying more than $3,333 a month, jobs held by a county's residents against jobs located in the county, one dot per county, 2019

No inline rows: the values of Table 6 are in https://vishalsingh.org/zipcode/data/work-wages-and-access-to-jobs/top_band_county_scatter.json.

Note: Counties with fewer than 1,000 jobs in either table are not drawn because noise added for confidentiality makes their shares unreliable; they are in every other figure of the chapter. Alaska and Massachusetts are left out.

Note: Both axes use the same earnings band in current dollars for the same year, so the comparison is between populations (jobs at the home and jobs at the workplace) and not between bands. The tooltip gives the county and the jobs held by residents. Noise in county shares pulls dots apart in both directions, so the share above the diagonal is the figure to read and not the slope of the cloud.

Source: LEHD LODES residence and workplace tables, 2019 (Alaska and Massachusetts left out); full counts of jobs with confidentiality noise, earnings bands in current dollars

Data: https://vishalsingh.org/zipcode/report/work-wages-and-access-to-jobs/figures/top_band_county_scatter.json

### Table 7. In the most deprived tenth of tracts, 30.4% of jobs held by residents were in the top earnings band against 60.1% in the least deprived tenth, and unemployment was 8.6% against 4.9%

Five measures of the residents of tracts in the least and the most deprived tenth by Area Deprivation Index rank, percent; the index is built partly from income, employment and vehicle access, so all five gradients are partly mechanical

| least | least_hi | least_lo | measure | most | most_hi | most_lo | tracts_least | tracts_most |
| ---: | ---: | ---: | --- | ---: | ---: | ---: | ---: | ---: |
| 4.92 | 5.41 | 4.35 | Unemployment rate (ACS 2019-2023) | 8.65 | 9.05 | 8.25 | 8021 | 8103 |
| 66.12 | 66.72 | 65.45 | Labor force participation (ACS 2019-2023) | 54.82 | 55.27 | 54.35 | 8021 | 8103 |
| 12.02 | 14.28 | 10.11 | Commuting workers with a commute of an hour or more (ACS 2019-2023) | 7.29 | 7.73 | 6.88 | 8021 | 8103 |
| 23.44 | 25.27 | 21.97 | Workers who work from home (ACS 2019-2023) | 5.94 | 6.19 | 5.69 | 8021 | 8103 |
| 60.07 | 63.33 | 57.06 | Jobs held by residents in the top earnings band (LODES 2019) | 30.41 | 30.94 | 29.93 | 8014 | 8103 |

Note: The Area Deprivation Index (rank 1 least deprived, 100 most deprived) averages block-group ranks to tracts without population weights and understates deprivation in high-cost cities. Tenths hold near-equal numbers of the 81,038 tracts with a rank (tied ranks stay together); 1,580 tracts have none.

Note: The ACS rows describe residents: the unemployment rate is a share of the civilian labor force aged 16 and over, participation a share of people aged 16 and over, the commute row a share of workers who commute (workers who do not work from home) and the work-from-home row a share of all workers. The staged ACS file has no margins of error, so the intervals show variation across counties and not the precision of single tracts.

Note: The last row counts jobs held by residents, which LODES places at the worker's home tract. Alaska, Massachusetts and Puerto Rico are left out of the whole figure so that one set of tracts is compared; Connecticut tracts are matched through the tract crosswalk.

Source: American Community Survey 2019-2023 tracts, population-weighted, no margins of error; LEHD LODES residence table 2019, jobs-weighted; Area Deprivation Index 2023, national rank

Data: https://vishalsingh.org/zipcode/report/work-wages-and-access-to-jobs/figures/tract_measures_by_adi.json

### Table 8. In the most deprived fifth of tracts in metropolitan counties, 43.7% of jobs located there were in the top earnings band against 32.6% of jobs held by the tracts' residents

Percent of jobs paying more than $3,333 a month, jobs held by residents of a tract against jobs located in it, by fifth of tracts by Area Deprivation Index rank within each county type, 2019

| conf_high | conf_low | estimate | fifth | group | kind | n_tracts |
| ---: | ---: | ---: | ---: | --- | --- | ---: |
| 60.46 | 54.48 | 57.50 | 1 | Metropolitan counties | Jobs held by residents of the tract | 15,826 |
| 56.54 | 49.97 | 53.25 | 1 | Metropolitan counties | Jobs located in the tract | 15,826 |
| 53.33 | 51.12 | 52.25 | 2 | Metropolitan counties | Jobs held by residents of the tract | 15,350 |
| 50.22 | 47.63 | 48.88 | 2 | Metropolitan counties | Jobs located in the tract | 15,350 |
| 47.57 | 45.97 | 46.76 | 3 | Metropolitan counties | Jobs held by residents of the tract | 14,289 |
| 45.88 | 44.03 | 44.94 | 3 | Metropolitan counties | Jobs located in the tract | 14,289 |
| 41.50 | 40.41 | 40.94 | 4 | Metropolitan counties | Jobs held by residents of the tract | 12,062 |
| 44.35 | 42.48 | 43.44 | 4 | Metropolitan counties | Jobs located in the tract | 12,062 |
| 33.23 | 32.06 | 32.61 | 5 | Metropolitan counties | Jobs held by residents of the tract | 9,927 |
| 44.59 | 42.64 | 43.65 | 5 | Metropolitan counties | Jobs located in the tract | 9,927 |
| 44.87 | 41.62 | 43.26 | 1 | Nonmetropolitan counties with larger towns | Jobs held by residents of the tract | 281 |
| 41.93 | 36.76 | 39.27 | 1 | Nonmetropolitan counties with larger towns | Jobs located in the tract | 281 |
| 46.84 | 43.78 | 45.27 | 2 | Nonmetropolitan counties with larger towns | Jobs held by residents of the tract | 701 |
| 43.21 | 37.96 | 40.49 | 2 | Nonmetropolitan counties with larger towns | Jobs located in the tract | 701 |
| 45.03 | 43.60 | 44.31 | 3 | Nonmetropolitan counties with larger towns | Jobs held by residents of the tract | 1,626 |
| 42.12 | 39.17 | 40.59 | 3 | Nonmetropolitan counties with larger towns | Jobs located in the tract | 1,626 |
| 42.01 | 41.22 | 41.62 | 4 | Nonmetropolitan counties with larger towns | Jobs held by residents of the tract | 3,183 |
| 40.59 | 39.09 | 39.84 | 4 | Nonmetropolitan counties with larger towns | Jobs located in the tract | 3,183 |
| 35.71 | 34.90 | 35.31 | 5 | Nonmetropolitan counties with larger towns | Jobs held by residents of the tract | 4,161 |
| 37.81 | 36.61 | 37.21 | 5 | Nonmetropolitan counties with larger towns | Jobs located in the tract | 4,161 |
| 43.36 | 38.56 | 40.92 | 2 | Nonmetropolitan counties with smaller towns or none | Jobs held by residents of the tract | 84 |
| 39.39 | 31.68 | 35.51 | 2 | Nonmetropolitan counties with smaller towns or none | Jobs located in the tract | 84 |
| 42 | 40.01 | 41.03 | 3 | Nonmetropolitan counties with smaller towns or none | Jobs held by residents of the tract | 376 |
| 40.68 | 36.57 | 38.61 | 3 | Nonmetropolitan counties with smaller towns or none | Jobs located in the tract | 376 |
| 40.37 | 39.34 | 39.85 | 4 | Nonmetropolitan counties with smaller towns or none | Jobs held by residents of the tract | 1,068 |
| 40.06 | 37.85 | 38.96 | 4 | Nonmetropolitan counties with smaller towns or none | Jobs located in the tract | 1,068 |
| 35.85 | 35.08 | 35.47 | 5 | Nonmetropolitan counties with smaller towns or none | Jobs held by residents of the tract | 2,077 |
| 35.92 | 34.51 | 35.23 | 5 | Nonmetropolitan counties with smaller towns or none | Jobs located in the tract | 2,077 |

Note: Fifths hold near-equal numbers of tracts nationally (tied ranks stay together), so a county type has few tracts in the fifths it rarely contains. A cell with fewer than 30 tracts is not drawn, which leaves out the least deprived fifth of nonmetropolitan counties with smaller towns or none.

Note: Jobs located are placed by LODES at the workplace tract and jobs held by residents at the home tract, so the two lines describe different groups of people in the same tracts. 3.1% of jobs located and 0.3% of jobs held by residents are in tracts with no index rank and are not in the figure.

Note: Earnings bands are fixed in current dollars, so shares are compared within 2019 only. County types follow the county of the tract.

Source: LEHD LODES residence and workplace tables, 2019 (Alaska and Massachusetts left out), jobs-weighted; Area Deprivation Index 2023, national rank; county types from the USDA 2023 Rural-Urban Continuum Codes

Data: https://vishalsingh.org/zipcode/report/work-wages-and-access-to-jobs/figures/tract_top_band_by_county_type.json

## All estimates

Every fact in facts.json (127), sorted by id.

| Fact id | Value | Unit | Estimand | Population | Period | Interval | n | Weights | Source | Note |
| --- | ---: | --- | --- | --- | --- | --- | ---: | --- | --- | --- |
| acs_age | 16 | years of age | lowest age counted in the labor force measures (a definition) | people counted in the ACS labor force measures | 2019-2023 |   |   | none (definition) | American Community Survey |   |
| acs_out_share | 28% | percent of workers who travel to a county other than the one where they live | share of workers who travel to a different county for work, as reported by Census Bureau staff for the 2009-2013 county-to-county commuting flows (McKenzie and Burd, 2015); quoted, not recomputed | workers in the United States | ACS 2009-2013 |   |   | as reported (ACS survey weights) | Published estimates from cited studies | A published estimate, not a result of this build. |
| ak_rac_2015 | 308,964 | jobs | sum of the Alaska residence table | jobs held by residents of Alaska in the LODES residence table | 2015 |   |   | none (full count, with confidentiality noise) | LEHD Origin-Destination Employment Statistics |   |
| ak_rac_2019 | 3,090 | jobs | sum of the Alaska residence table; the 2019 workplace file for Alaska is absent, so only jobs outside Alaska are counted | jobs held by residents of Alaska in the LODES residence table | 2019 |   |   | none (full count, with confidentiality noise) | LEHD Origin-Destination Employment Statistics |   |
| band_cut_high | $3,333 | dollars of monthly earnings from one job, current dollars | lower limit of the top of the three LODES earnings bands: a job is in the top band when it pays more than this (a definition) | jobs in the LEHD LODES tables | fixed in every LODES year |   |   | none (definition) | LEHD Origin-Destination Employment Statistics |   |
| band_cut_low | $1,250 | dollars of monthly earnings from one job, current dollars | upper limit of the lowest of the three LODES earnings bands (a definition) | jobs in the LEHD LODES tables | fixed in every LODES year |   |   | none (definition) | LEHD Origin-Destination Employment Statistics |   |
| c60_least | 12.0% | percent of commuting workers (workers who do not work from home) whose commute takes an hour or more | population-weighted mean of tract values (2019-2023 ACS tract population as the weight), county-bootstrap interval | commuting workers living in tracts outside Alaska, Massachusetts and Puerto Rico in the least deprived tenth of tracts by Area Deprivation Index rank | ACS 2019-2023 | 95% CI 10.1% to 14.3% | 8,021 tracts | 2019-2023 tract population | American Community Survey |   |
| c60_most | 7.3% | percent of commuting workers (workers who do not work from home) whose commute takes an hour or more | population-weighted mean of tract values (2019-2023 ACS tract population as the weight), county-bootstrap interval | commuting workers living in tracts outside Alaska, Massachusetts and Puerto Rico in the most deprived tenth of tracts by Area Deprivation Index rank | ACS 2019-2023 | 95% CI 6.9% to 7.7% | 8,103 tracts | 2019-2023 tract population | American Community Survey |   |
| corr_diff | +0.15 | difference of two correlation coefficients | correlation with the workplace share minus correlation with the residence share, from one set of state draws | counties with a published QCEW average pay and at least the minimum number of jobs in both LODES tables (Alaska and Massachusetts left out) | 2019 | 95% CI 0.11 to 0.21 | 2,877 counties | 2020 county population | BLS Quarterly Census of Employment and Wages |   |
| corr_pay_resid | +0.73 | correlation coefficient | population-weighted correlation across counties of QCEW average pay with the top-band share of jobs held by the county's residents (LODES residence table), bootstrap interval over states | counties with a published QCEW average pay and at least the minimum number of jobs in both LODES tables (Alaska and Massachusetts left out) | 2019 | 95% CI 0.69 to 0.77 | 2,877 counties | 2020 county population | BLS Quarterly Census of Employment and Wages |   |
| corr_pay_work | +0.87 | correlation coefficient | population-weighted correlation across counties of QCEW average pay with the top-band share of jobs located in the county (LODES workplace table), bootstrap interval over states | counties with a published QCEW average pay and at least the minimum number of jobs in both LODES tables (Alaska and Massachusetts left out) | 2019 | 95% CI 0.86 to 0.92 | 2,877 counties | 2020 county population | BLS Quarterly Census of Employment and Wages |   |
| cut_shift | 0.1 | percentage points | largest change in any fifth's share of jobs or top-band share when the county at a cut between fifths moves to the next fifth | job shares and top-band shares of the fifths of counties (Alaska and Massachusetts left out) | 2019 |   |   | none (full count) | LEHD Origin-Destination Employment Statistics |   |
| ds_highest_fifth | 2.0 | percentage points of the top-band share, absolute difference | top-band share of jobs held by residents minus that of jobs located, from one set of state draws, absolute value, bootstrap interval over states | jobs in the highest fifth of counties by jobs located per job held by residents (Alaska and Massachusetts left out) | 2019 | 95% CI 1.2 to 2.9 | 620 counties | jobs in each table | LEHD Origin-Destination Employment Statistics |   |
| ds_lowest_fifth | 5.9 | percentage points of the top-band share, absolute difference | top-band share of jobs held by residents minus that of jobs located, from one set of state draws, absolute value, bootstrap interval over states | jobs in the lowest fifth of counties by jobs located per job held by residents (Alaska and Massachusetts left out) | 2019 | 95% CI 3.2 to 7.8 | 619 counties | jobs in each table | LEHD Origin-Destination Employment Statistics |   |
| gap_2019_small | 0.7 | percentage points of the unemployment rate | difference of two pooled rates (nonmetropolitan counties with smaller towns or none minus metropolitan counties) from one set of state draws, bootstrap interval over states | residents of counties of the two types (Puerto Rico left out) | annual average, 2019 | 95% CI 0.4 to 1.0 | 3,142 counties | labor force of each county | BLS Local Area Unemployment Statistics |   |
| gap_2019_town | 0.5 | percentage points of the unemployment rate | difference of two pooled rates (nonmetropolitan counties with larger towns minus metropolitan counties) from one set of state draws, bootstrap interval over states | residents of counties of the two types (Puerto Rico left out) | annual average, 2019 | 95% CI 0.3 to 0.7 | 3,142 counties | labor force of each county | BLS Local Area Unemployment Statistics |   |
| gap_2020_small | 1.5 | percentage points of the unemployment rate | difference of two pooled rates (metropolitan counties minus nonmetropolitan counties with smaller towns or none) from one set of state draws, bootstrap interval over states | residents of counties of the two types (Puerto Rico left out) | annual average, 2020 | 95% CI 0.7 to 2.2 | 3,143 counties | labor force of each county | BLS Local Area Unemployment Statistics |   |
| gap_2020_town | 0.9 | percentage points of the unemployment rate | difference of two pooled rates (metropolitan counties minus nonmetropolitan counties with larger towns) from one set of state draws, bootstrap interval over states | residents of counties of the two types (Puerto Rico left out) | annual average, 2020 | 95% CI 0.4 to 1.5 | 3,143 counties | labor force of each county | BLS Local Area Unemployment Statistics |   |
| gap_2023_small | −0.1 | percentage points of the unemployment rate | difference of two pooled rates (metropolitan counties minus nonmetropolitan counties with smaller towns or none) from one set of state draws, bootstrap interval over states | residents of counties of the two types (Puerto Rico left out) | annual average, 2023 | 95% CI −0.5 to 0.2 | 3,143 counties | labor force of each county | BLS Local Area Unemployment Statistics |   |
| jobs_covered | 148.0 million | jobs | sum of average annual employment over the counties with a published value | jobs located in counties, covered by state unemployment insurance (Puerto Rico and statewide residual rows left out) | annual average, 2023 |   |   | none (full count) | BLS Quarterly Census of Employment and Wages |   |
| jobs_no_adi | 3.1% | percent of jobs located | jobs located in tracts with no ADI rank, as a share of all jobs located | jobs located in tracts outside Alaska and Massachusetts | LODES 2019 |   |   | none (full count, with confidentiality noise) | LEHD Origin-Destination Employment Statistics |   |
| jobs_od_2019 | 142.0 million | jobs | sum of the origin-destination flows (all jobs) | jobs held by residents of counties (Alaska and Massachusetts left out) | 2019 |   |   | none (full count, with confidentiality noise) | LEHD Origin-Destination Employment Statistics |   |
| jobs_share_metro | 88.9% | percent of covered jobs | jobs located in metropolitan counties as a share of jobs located in all counties | jobs located in counties, covered by state unemployment insurance (Puerto Rico and statewide residual rows left out) | annual average, 2023 |   |   | none (full count) | BLS Quarterly Census of Employment and Wages |   |
| lf_floor | 2,000 | persons in the labor force | labor force below which a county is counted as small (a rule) | counties and county equivalents | annual average, 2023 |   |   | none (definition) | BLS Local Area Unemployment Statistics |   |
| lfp_least | 66.1% | percent of people of working age in the civilian labor force | population-weighted mean of tract values (2019-2023 ACS tract population as the weight), county-bootstrap interval | residents of tracts outside Alaska, Massachusetts and Puerto Rico in the least deprived tenth of tracts by Area Deprivation Index rank | ACS 2019-2023 | 95% CI 65.4% to 66.7% | 8,021 tracts | 2019-2023 tract population | American Community Survey |   |
| lfp_most | 54.8% | percent of people of working age in the civilian labor force | population-weighted mean of tract values (2019-2023 ACS tract population as the weight), county-bootstrap interval | residents of tracts outside Alaska, Massachusetts and Puerto Rico in the most deprived tenth of tracts by Area Deprivation Index rank | ACS 2019-2023 | 95% CI 54.3% to 55.3% | 8,103 tracts | 2019-2023 tract population | American Community Survey |   |
| ma_rac_2010 | 151,718 | jobs | sum of the Massachusetts residence table, which counts only jobs located in states that have a workplace table | jobs held by residents of Massachusetts in the LODES residence table | 2010 |   |   | none (full count, with confidentiality noise) | LEHD Origin-Destination Employment Statistics |   |
| ma_rac_2015 | 3,330,566 | jobs | sum of the Massachusetts residence table | jobs held by residents of Massachusetts in the LODES residence table | 2015 |   |   | none (full count, with confidentiality noise) | LEHD Origin-Destination Employment Statistics |   |
| ma_wac_2010 | 17 | jobs | sum of the Massachusetts workplace table, a stub file | jobs located in Massachusetts in the LODES workplace table | 2010 |   |   | none (full count, with confidentiality noise) | LEHD Origin-Destination Employment Statistics |   |
| ma_wac_tracts_2010 | 2 | tracts | count of tracts in the stub file | tracts in the Massachusetts workplace table | 2010 |   |   | none (full count) | LEHD Origin-Destination Employment Statistics |   |
| min_jobs_cut | 1,000 | jobs | minimum number of jobs in each of the two tables for a county to enter the scatter plot and the pay correlations (a rule) | counties in the LODES residence and workplace tables | LODES 2019 |   |   | none (definition) | LEHD Origin-Destination Employment Statistics |   |
| min_tracts_cell | 30 | tracts | minimum number of tracts for a cell to be drawn (a rule) | cells of county type by fifth of tracts by ADI | LODES 2019 |   |   | none (definition) | LEHD Origin-Destination Employment Statistics |   |
| mix_highest_metro | 52.3% | percent of the counties in the fifth | share of the fifth's counties of this county type (full count) | counties in the highest fifth of counties by jobs located per job held by residents (Alaska and Massachusetts left out), of the type metropolitan counties | 2019 (types from 2023 codes) |   | 620 counties | none (each county counts once) | Rural-Urban Continuum Codes |   |
| mix_lowest_metro | 47.7% | percent of the counties in the fifth | share of the fifth's counties of this county type (full count) | counties in the lowest fifth of counties by jobs located per job held by residents (Alaska and Massachusetts left out), of the type metropolitan counties | 2019 (types from 2023 codes) |   | 619 counties | none (each county counts once) | Rural-Urban Continuum Codes |   |
| mix_lowest_small | 43.9% | percent of the counties in the fifth | share of the fifth's counties of this county type (full count) | counties in the lowest fifth of counties by jobs located per job held by residents (Alaska and Massachusetts left out), of the type nonmetropolitan counties with smaller towns or none | 2019 (types from 2023 codes) |   | 619 counties | none (each county counts once) | Rural-Urban Continuum Codes |   |
| n_counties_lodes | 3,099 | counties | count of counties in the comparison | counties in both LODES tables (Alaska and Massachusetts left out) | 2019 |   |   | none (full count) | LEHD Origin-Destination Employment Statistics |   |
| n_laus_counties | 3,143 | counties | count of counties and county equivalents | counties and county equivalents in the LAUS file (Puerto Rico left out) | annual average, 2023 |   |   | none (full count) | BLS Local Area Unemployment Statistics |   |
| n_laus_pr | 78 | municipios | count of Puerto Rico units that are in the file and not drawn | Puerto Rico municipios in the LAUS file | annual average, 2023 |   |   | none (full count) | BLS Local Area Unemployment Statistics |   |
| n_missing_parishes | 7 | parishes | count of parishes with a missing annual average | Louisiana parishes with no published annual average | 2005 and 2006 |   |   | none (full count) | BLS Local Area Unemployment Statistics |   |
| n_missing_ur | 14 | county-years | county-years with no published annual average (footnote N); they leave the pooled rates of those years | counties in the LAUS file, 2000-2024 | 2005 and 2006 |   |   | none (full count) | BLS Local Area Unemployment Statistics |   |
| n_nodata_ur | 1 | counties | count of drawn counties that are not in the LAUS file | counties on the map | annual average, 2023 |   |   | none (full count) | BLS Local Area Unemployment Statistics |   |
| n_nopay | 2 | counties | count of drawn counties with no value: one withheld and one absent from the file | counties on the map | annual average, 2023 |   |   | none (full count) | BLS Quarterly Census of Employment and Wages |   |
| n_pay_counties | 3,141 | counties | count of counties with a published value | counties with a published average pay | annual average, 2023 |   |   | none (full count) | BLS Quarterly Census of Employment and Wages |   |
| n_pre_apart | 12 | years | years in which the 95% interval of that difference lies entirely below zero | years from 2000 to 2019 | 2000-2019 |   |   | none (full count) | BLS Local Area Unemployment Statistics |   |
| n_pre_years | 20 | years | count of years in the comparison of metropolitan counties with nonmetropolitan counties with smaller towns | years from 2000 to 2019 | 2000-2019 |   |   | none (full count) | BLS Local Area Unemployment Statistics |   |
| n_scatter | 2,877 | counties | count of counties in the scatter plot | counties with at least the minimum number of jobs in both tables (Alaska and Massachusetts left out) | 2019 |   |   | none (full count) | LEHD Origin-Destination Employment Statistics |   |
| n_small_lf | 296 | counties | count of counties with a labor force below the small-county cut | counties and county equivalents in the LAUS file (Puerto Rico left out) | annual average, 2023 |   |   | none (full count) | BLS Local Area Unemployment Statistics |   |
| n_tracts_adi | 81,038 | tracts | count of tracts with a rank | tracts outside Alaska, Massachusetts and Puerto Rico with an ADI rank | ADI 2023, ACS 2019-2023 |   |   | none (full count) | Area Deprivation Index |   |
| n_tracts_no_adi | 1,580 | tracts | count of tracts with no rank (all block groups suppressed) | tracts outside Alaska, Massachusetts and Puerto Rico with no ADI rank | ADI 2023, ACS 2019-2023 |   |   | none (full count) | Area Deprivation Index |   |
| n_ur_high | 6 | counties | count of counties with a rate at or above the high-rate cut | counties and county equivalents in the LAUS file (Puerto Rico left out) | annual average, 2023 |   |   | none (full count) | BLS Local Area Unemployment Statistics |   |
| n_withheld | 1 | counties | count of counties whose total wages and employment the Bureau withholds | counties in the QCEW file | annual average, 2023 |   |   | none (full count) | BLS Quarterly Census of Employment and Wages |   |
| out_all_2010 | 44.1% | percent of jobs held by a county's residents that are located in a different county | jobs located outside the home county summed over counties, divided by jobs held by residents summed over the same counties, bootstrap interval over states | jobs held by residents of counties (Alaska and Massachusetts left out) | 2010 | 95% CI 40.8% to 47.6% | 3,099 counties | jobs held by residents of each county | LEHD Origin-Destination Employment Statistics |   |
| out_all_2019 | 45.4% | percent of jobs held by a county's residents that are located in a different county | jobs located outside the home county summed over counties, divided by jobs held by residents summed over the same counties, bootstrap interval over states | jobs held by residents of counties (Alaska and Massachusetts left out) | 2019 | 95% CI 41.8% to 49.1% | 3,099 counties | jobs held by residents of each county | LEHD Origin-Destination Employment Statistics |   |
| out_chg_all | 1.3 | percentage points of the out-of-county share | out-of-county share in 2019 minus the share in 2010, from one set of state draws, bootstrap interval over states | jobs held by residents of counties (Alaska and Massachusetts left out) | 2019 compared with 2010 | 95% CI 0.9 to 1.6 | 3,099 counties | jobs held by residents of each county | LEHD Origin-Destination Employment Statistics |   |
| out_chg_metro | 1.1 | percentage points of the out-of-county share | out-of-county share in 2019 minus the share in 2010, from one set of state draws, bootstrap interval over states | jobs held by residents of metropolitan counties (Alaska and Massachusetts left out) | 2019 compared with 2010 | 95% CI 0.8 to 1.5 | 1,171 counties | jobs held by residents of each county | LEHD Origin-Destination Employment Statistics |   |
| out_chg_small | 3.0 | percentage points of the out-of-county share | out-of-county share in 2019 minus the share in 2010, from one set of state draws, bootstrap interval over states | jobs held by residents of nonmetropolitan counties with smaller towns or none (Alaska and Massachusetts left out) | 2019 compared with 2010 | 95% CI 2.5 to 3.4 | 1,034 counties | jobs held by residents of each county | LEHD Origin-Destination Employment Statistics |   |
| out_chg_town | 2.9 | percentage points of the out-of-county share | out-of-county share in 2019 minus the share in 2010, from one set of state draws, bootstrap interval over states | jobs held by residents of nonmetropolitan counties with larger towns (Alaska and Massachusetts left out) | 2019 compared with 2010 | 95% CI 2.4 to 3.3 | 894 counties | jobs held by residents of each county | LEHD Origin-Destination Employment Statistics |   |
| out_gap_small_metro | 18.4 | percentage points of the out-of-county share | out-of-county share of nonmetropolitan counties with smaller towns or none minus that of metropolitan counties, from one set of state draws, bootstrap interval over states | jobs held by residents of the two types of counties (Alaska and Massachusetts left out) | 2019 | 95% CI 13.7 to 22.0 | 3,099 counties | jobs held by residents of each county | LEHD Origin-Destination Employment Statistics |   |
| out_gap_small_town | 11.0 | percentage points of the out-of-county share | out-of-county share of nonmetropolitan counties with smaller towns or none minus that of nonmetropolitan counties with larger towns, from one set of state draws, bootstrap interval over states | jobs held by residents of the two types of counties (Alaska and Massachusetts left out) | 2019 | 95% CI 9.0 to 12.9 | 3,099 counties | jobs held by residents of each county | LEHD Origin-Destination Employment Statistics |   |
| out_gap_town_metro | 7.4 | percentage points of the out-of-county share | out-of-county share of nonmetropolitan counties with larger towns minus that of metropolitan counties, from one set of state draws, bootstrap interval over states | jobs held by residents of the two types of counties (Alaska and Massachusetts left out) | 2019 | 95% CI 2.7 to 10.9 | 3,099 counties | jobs held by residents of each county | LEHD Origin-Destination Employment Statistics |   |
| out_metro_2010 | 43.0% | percent of jobs held by a county's residents that are located in a different county | jobs located outside the home county summed over counties, divided by jobs held by residents summed over the same counties, bootstrap interval over states | jobs held by residents of metropolitan counties (Alaska and Massachusetts left out) | 2010 | 95% CI 39.4% to 46.9% | 1,171 counties | jobs held by residents of each county | LEHD Origin-Destination Employment Statistics |   |
| out_metro_2019 | 44.1% | percent of jobs held by a county's residents that are located in a different county | jobs located outside the home county summed over counties, divided by jobs held by residents summed over the same counties, bootstrap interval over states | jobs held by residents of metropolitan counties (Alaska and Massachusetts left out) | 2019 | 95% CI 40.3% to 48.3% | 1,171 counties | jobs held by residents of each county | LEHD Origin-Destination Employment Statistics |   |
| out_small_2010 | 59.5% | percent of jobs held by a county's residents that are located in a different county | jobs located outside the home county summed over counties, divided by jobs held by residents summed over the same counties, bootstrap interval over states | jobs held by residents of nonmetropolitan counties with smaller towns or none (Alaska and Massachusetts left out) | 2010 | 95% CI 56.6% to 61.8% | 1,034 counties | jobs held by residents of each county | LEHD Origin-Destination Employment Statistics |   |
| out_small_2019 | 62.4% | percent of jobs held by a county's residents that are located in a different county | jobs located outside the home county summed over counties, divided by jobs held by residents summed over the same counties, bootstrap interval over states | jobs held by residents of nonmetropolitan counties with smaller towns or none (Alaska and Massachusetts left out) | 2019 | 95% CI 59.5% to 64.8% | 1,034 counties | jobs held by residents of each county | LEHD Origin-Destination Employment Statistics |   |
| out_town_2010 | 48.6% | percent of jobs held by a county's residents that are located in a different county | jobs located outside the home county summed over counties, divided by jobs held by residents summed over the same counties, bootstrap interval over states | jobs held by residents of nonmetropolitan counties with larger towns (Alaska and Massachusetts left out) | 2010 | 95% CI 46.3% to 50.5% | 894 counties | jobs held by residents of each county | LEHD Origin-Destination Employment Statistics |   |
| out_town_2019 | 51.5% | percent of jobs held by a county's residents that are located in a different county | jobs located outside the home county summed over counties, divided by jobs held by residents summed over the same counties, bootstrap interval over states | jobs held by residents of nonmetropolitan counties with larger towns (Alaska and Massachusetts left out) | 2019 | 95% CI 49.1% to 53.4% | 894 counties | jobs held by residents of each county | LEHD Origin-Destination Employment Statistics |   |
| pay_max | $174,694 | dollars of average annual pay per job, current dollars | county total wages divided by average employment (the highest county average) | jobs located in Santa Clara County, CA | annual average, 2023 |   | 1,119,710 jobs | none (one county) | BLS Quarterly Census of Employment and Wages |   |
| pay_mean | $71,075 | dollars of average annual pay per job, current dollars | total wages divided by average employment, summed over counties (full count, no sampling error) | jobs located in counties, covered by state unemployment insurance (Puerto Rico and statewide residual rows left out) | annual average, 2023 |   | 3,141 counties | jobs located in each county | BLS Quarterly Census of Employment and Wages |   |
| pay_median | $49,668 | dollars of average annual pay per job, current dollars | 50th percentile of county average pay, each county counted once | jobs located in counties, covered by state unemployment insurance (Puerto Rico and statewide residual rows left out) | annual average, 2023 |   | 3,141 counties | none (each county counts once) | BLS Quarterly Census of Employment and Wages |   |
| pay_metro | $73,585 | dollars of average annual pay per job, current dollars | total wages divided by average employment, summed over the counties of the type (full count, no sampling error) | jobs located in metropolitan counties, covered by state unemployment insurance | annual average, 2023 |   | 1,184 counties | jobs located in each county | BLS Quarterly Census of Employment and Wages |   |
| pay_min | $29,704 | dollars of average annual pay per job, current dollars | county total wages divided by average employment (the lowest county average) | jobs located in Worth County, MO | annual average, 2023 |   | 385 jobs | none (one county) | BLS Quarterly Census of Employment and Wages |   |
| pay_min_jobs | 385 | jobs | average annual employment of the county | jobs located in Worth County, MO | annual average, 2023 |   |   | none (one county) | BLS Quarterly Census of Employment and Wages |   |
| pay_p5 | $39,160 | dollars of average annual pay per job, current dollars | fifth percentile of county average pay, each county counted once | jobs located in counties, covered by state unemployment insurance (Puerto Rico and statewide residual rows left out) | annual average, 2023 |   | 3,141 counties | none (each county counts once) | BLS Quarterly Census of Employment and Wages |   |
| pay_p95 | $72,089 | dollars of average annual pay per job, current dollars | ninety-fifth percentile of county average pay, each county counted once | jobs located in counties, covered by state unemployment insurance (Puerto Rico and statewide residual rows left out) | annual average, 2023 |   | 3,141 counties | none (each county counts once) | BLS Quarterly Census of Employment and Wages |   |
| pay_second | $174,319 | dollars of average annual pay per job, current dollars | county total wages divided by average employment (the second highest county average) | jobs located in San Mateo County, CA | annual average, 2023 |   | 420,679 jobs | none (one county) | BLS Quarterly Census of Employment and Wages |   |
| pay_small | $48,786 | dollars of average annual pay per job, current dollars | total wages divided by average employment, summed over the counties of the type (full count, no sampling error) | jobs located in nonmetropolitan counties with smaller towns or none, covered by state unemployment insurance | annual average, 2023 |   | 1,054 counties | jobs located in each county | BLS Quarterly Census of Employment and Wages |   |
| pay_town | $51,618 | dollars of average annual pay per job, current dollars | total wages divided by average employment, summed over the counties of the type (full count, no sampling error) | jobs located in nonmetropolitan counties with larger towns, covered by state unemployment insurance | annual average, 2023 |   | 903 counties | jobs located in each county | BLS Quarterly Census of Employment and Wages |   |
| pct_resid_higher | 66.0% | percent of counties | share of counties (each counted once) in which the top-band share of jobs held by residents exceeds that of jobs located, bootstrap interval over states | counties with at least the minimum number of jobs in both tables (Alaska and Massachusetts left out) | 2019 | 95% CI 63.8% to 68.3% | 2,877 counties | none (each county counts once) | LEHD Origin-Destination Employment Statistics |   |
| ratio_cut_high | 1.01 | jobs located per job held by residents | smallest ratio in the highest fifth of counties | counties in both LODES tables (Alaska and Massachusetts left out) | 2019 |   |   | none (cut point) | LEHD Origin-Destination Employment Statistics |   |
| ratio_cut_low | 0.60 | jobs located per job held by residents | largest ratio in the lowest fifth of counties | counties in both LODES tables (Alaska and Massachusetts left out) | 2019 |   |   | none (cut point) | LEHD Origin-Destination Employment Statistics |   |
| res_no_adi | 0.3% | percent of jobs held by residents | jobs held by residents of tracts with no ADI rank, as a share of all such jobs | jobs held by residents of tracts outside Alaska and Massachusetts | LODES 2019 |   |   | none (full count, with confidentiality noise) | LEHD Origin-Destination Employment Statistics |   |
| rs_highest_fifth | 47.8% | percent of jobs paying more than the top-band cut (monthly earnings from the job) | jobs in the top earnings band summed over the counties of the group, divided by jobs summed over the same counties, bootstrap interval over states | jobs held by residents of the highest fifth of counties by jobs located per job held by residents (Alaska and Massachusetts left out) | 2019 | 95% CI 46.0% to 49.3% | 620 counties | jobs in each table | LEHD Origin-Destination Employment Statistics |   |
| rs_lowest_fifth | 45.4% | percent of jobs paying more than the top-band cut (monthly earnings from the job) | jobs in the top earnings band summed over the counties of the group, divided by jobs summed over the same counties, bootstrap interval over states | jobs held by residents of the lowest fifth of counties by jobs located per job held by residents (Alaska and Massachusetts left out) | 2019 | 95% CI 43.4% to 47.3% | 619 counties | jobs in each table | LEHD Origin-Destination Employment Statistics |   |
| sect_local | 32% | percent of county-by-sector cells withheld | cells whose employment and wages the Bureau withholds, as a share of all cells of the ownership | county-by-industry-sector cells of local governments in the QCEW file (Puerto Rico and statewide residual rows left out) | 2023 |   | 20,766 county-sector cells | none (full count) | BLS Quarterly Census of Employment and Wages |   |
| sect_private | 25% | percent of county-by-sector cells withheld | cells whose employment and wages the Bureau withholds, as a share of all cells of the ownership | county-by-industry-sector cells of private employers in the QCEW file (Puerto Rico and statewide residual rows left out) | 2023 |   | 58,534 county-sector cells | none (full count) | BLS Quarterly Census of Employment and Wages |   |
| sect_state | 29% | percent of county-by-sector cells withheld | cells whose employment and wages the Bureau withholds, as a share of all cells of the ownership | county-by-industry-sector cells of state governments in the QCEW file (Puerto Rico and statewide residual rows left out) | 2023 |   | 10,232 county-sector cells | none (full count) | BLS Quarterly Census of Employment and Wages |   |
| share_resid_highest | 46.0% | percent of all jobs held by residents in the counties compared | jobs of the fifth as a share of all such jobs in the counties compared (full count) | jobs held by residents in the highest fifth of counties by jobs located per job held by residents (Alaska and Massachusetts left out) | 2019 |   | 620 counties | none (full count, with confidentiality noise) | LEHD Origin-Destination Employment Statistics |   |
| share_resid_lowest | 6.3% | percent of all jobs held by residents in the counties compared | jobs of the fifth as a share of all such jobs in the counties compared (full count) | jobs held by residents in the lowest fifth of counties by jobs located per job held by residents (Alaska and Massachusetts left out) | 2019 |   | 619 counties | none (full count, with confidentiality noise) | LEHD Origin-Destination Employment Statistics |   |
| share_work_highest | 56.5% | percent of all jobs located in the counties compared | jobs of the fifth as a share of all such jobs in the counties compared (full count) | jobs located in the highest fifth of counties by jobs located per job held by residents (Alaska and Massachusetts left out) | 2019 |   | 620 counties | none (full count, with confidentiality noise) | LEHD Origin-Destination Employment Statistics |   |
| share_work_lowest | 3.2% | percent of all jobs located in the counties compared | jobs of the fifth as a share of all such jobs in the counties compared (full count) | jobs located in the lowest fifth of counties by jobs located per job held by residents (Alaska and Massachusetts left out) | 2019 |   | 619 counties | none (full count, with confidentiality noise) | LEHD Origin-Destination Employment Statistics |   |
| spread_diff | 7.9 | percentage points of the top-band share | spread across fifths of jobs located minus spread across fifths of jobs held by residents, from one set of state draws | jobs in both tables, highest and lowest fifth of counties by jobs located per job held by residents (Alaska and Massachusetts left out) | 2019 | 95% CI 5.7 to 9.7 | 3,099 counties | jobs in each table | LEHD Origin-Destination Employment Statistics |   |
| spread_resid | 2.4 | percentage points of the top-band share | top-band share in the highest fifth minus that in the lowest fifth, bootstrap interval over states | jobs held by residents in the highest and the lowest fifth of counties by jobs located per job held by residents (Alaska and Massachusetts left out) | 2019 | 95% CI 0.5 to 4.5 | 3,099 counties | jobs in each table | LEHD Origin-Destination Employment Statistics |   |
| spread_work | 10.3 | percentage points of the top-band share | top-band share in the highest fifth minus that in the lowest fifth, bootstrap interval over states | jobs located in the highest and the lowest fifth of counties by jobs located per job held by residents (Alaska and Massachusetts left out) | 2019 | 95% CI 8.5 to 12.5 | 3,099 counties | jobs in each table | LEHD Origin-Destination Employment Statistics |   |
| tb_gap_metro_most | +11.0 | percentage points of the top-band share, jobs located minus jobs held by residents | top-band share of jobs located minus that of jobs held by residents of the same tracts, from one set of county draws, bootstrap interval over counties | jobs in the most deprived fifth of tracts by ADI rank in metropolitan counties (outside Alaska and Massachusetts) | LODES 2019 | 95% CI 10.1 to 11.9 | 9,927 tracts | jobs in each table | LEHD Origin-Destination Employment Statistics |   |
| tb_gap_small_most | −0.2 | percentage points of the top-band share, jobs located minus jobs held by residents | top-band share of jobs located minus that of jobs held by residents of the same tracts, from one set of county draws, bootstrap interval over counties | jobs in the most deprived fifth of tracts by ADI rank in nonmetropolitan counties with smaller towns or none (outside Alaska and Massachusetts) | LODES 2019 | 95% CI −0.8 to 0.3 | 2,077 tracts | jobs in each table | LEHD Origin-Destination Employment Statistics |   |
| tb_gap_town_most | +1.9 | percentage points of the top-band share, jobs located minus jobs held by residents | top-band share of jobs located minus that of jobs held by residents of the same tracts, from one set of county draws, bootstrap interval over counties | jobs in the most deprived fifth of tracts by ADI rank in nonmetropolitan counties with larger towns (outside Alaska and Massachusetts) | LODES 2019 | 95% CI 1.4 to 2.4 | 4,161 tracts | jobs in each table | LEHD Origin-Destination Employment Statistics |   |
| tb_metro_job_least | 53.3% | percent of jobs paying more than the top-band cut (monthly earnings from the job) | jobs in the top earnings band summed over the tracts, divided by jobs summed over the same tracts, county-bootstrap interval | jobs located of metropolitan counties in the least deprived fifth of tracts by Area Deprivation Index rank (outside Alaska and Massachusetts) | LODES 2019 | 95% CI 50.0% to 56.5% | 15,826 tracts | jobs in each table | LEHD Origin-Destination Employment Statistics |   |
| tb_metro_job_most | 43.7% | percent of jobs paying more than the top-band cut (monthly earnings from the job) | jobs in the top earnings band summed over the tracts, divided by jobs summed over the same tracts, county-bootstrap interval | jobs located of metropolitan counties in the most deprived fifth of tracts by Area Deprivation Index rank (outside Alaska and Massachusetts) | LODES 2019 | 95% CI 42.6% to 44.6% | 9,927 tracts | jobs in each table | LEHD Origin-Destination Employment Statistics |   |
| tb_metro_res_least | 57.5% | percent of jobs paying more than the top-band cut (monthly earnings from the job) | jobs in the top earnings band summed over the tracts, divided by jobs summed over the same tracts, county-bootstrap interval | jobs held by residents of metropolitan counties in the least deprived fifth of tracts by Area Deprivation Index rank (outside Alaska and Massachusetts) | LODES 2019 | 95% CI 54.5% to 60.5% | 15,826 tracts | jobs in each table | LEHD Origin-Destination Employment Statistics |   |
| tb_metro_res_most | 32.6% | percent of jobs paying more than the top-band cut (monthly earnings from the job) | jobs in the top earnings band summed over the tracts, divided by jobs summed over the same tracts, county-bootstrap interval | jobs held by residents of metropolitan counties in the most deprived fifth of tracts by Area Deprivation Index rank (outside Alaska and Massachusetts) | LODES 2019 | 95% CI 32.1% to 33.2% | 9,927 tracts | jobs in each table | LEHD Origin-Destination Employment Statistics |   |
| top_res_least | 60.1% | percent of jobs paying more than the top-band cut (monthly earnings from the job) | jobs in the top earnings band summed over the tracts, divided by jobs summed over the same tracts, county-bootstrap interval | jobs held by residents of tracts outside Alaska and Massachusetts in the least deprived tenth of tracts by Area Deprivation Index rank | LODES 2019 | 95% CI 57.1% to 63.3% | 8,014 tracts | jobs held by residents of each tract | LEHD Origin-Destination Employment Statistics |   |
| top_res_most | 30.4% | percent of jobs paying more than the top-band cut (monthly earnings from the job) | jobs in the top earnings band summed over the tracts, divided by jobs summed over the same tracts, county-bootstrap interval | jobs held by residents of tracts outside Alaska and Massachusetts in the most deprived tenth of tracts by Area Deprivation Index rank | LODES 2019 | 95% CI 29.9% to 30.9% | 8,103 tracts | jobs held by residents of each tract | LEHD Origin-Destination Employment Statistics |   |
| town_cut | 5,000 | people in urban areas | urban population that separates the two groups of nonmetropolitan counties (a definition) | counties and county equivalents | 2023 Rural-Urban Continuum Code |   |   | none (definition) | Rural-Urban Continuum Codes |   |
| ur_all_2019 | 3.7% | percent of the civilian labor force that is unemployed | unemployed residents summed over the counties, divided by the labor force summed over the same counties, bootstrap interval over states | residents of counties and county equivalents in the LAUS file (Puerto Rico left out) | annual average, 2019 | 95% CI 3.5% to 3.8% | 3,142 counties | labor force of each county | BLS Local Area Unemployment Statistics |   |
| ur_all_2020 | 8.1% | percent of the civilian labor force that is unemployed | unemployed residents summed over the counties, divided by the labor force summed over the same counties, bootstrap interval over states | residents of counties and county equivalents in the LAUS file (Puerto Rico left out) | annual average, 2020 | 95% CI 7.4% to 8.7% | 3,143 counties | labor force of each county | BLS Local Area Unemployment Statistics |   |
| ur_cut_high | 10% | percent of the civilian labor force | rate at or above which a county is counted in the high group of the map (a rule) | counties and county equivalents | annual average, 2023 |   |   | none (definition) | BLS Local Area Unemployment Statistics |   |
| ur_max | 17.6% | percent of the civilian labor force that is unemployed | published county estimate (modeled; the county file carries no margin of error) | residents of Imperial County, CA | annual average, 2023 |   | 71,647 persons in the labor force | none (one county) | BLS Local Area Unemployment Statistics |   |
| ur_median_2023 | 3.4% | percent of the civilian labor force that is unemployed | 50th percentile of county rates, each county counted once | residents of counties and county equivalents in the LAUS file (Puerto Rico left out) | annual average, 2023 |   | 3,143 counties | none (each county counts once) | BLS Local Area Unemployment Statistics |   |
| ur_metro_2019 | 3.6% | percent of the civilian labor force that is unemployed | unemployed residents summed over the counties, divided by the labor force summed over the same counties, bootstrap interval over states | residents of metropolitan counties (Puerto Rico left out) | annual average, 2019 | 95% CI 3.4% to 3.8% | 1,185 counties | labor force of each county | BLS Local Area Unemployment Statistics |   |
| ur_metro_2020 | 8.2% | percent of the civilian labor force that is unemployed | unemployed residents summed over the counties, divided by the labor force summed over the same counties, bootstrap interval over states | residents of metropolitan counties (Puerto Rico left out) | annual average, 2020 | 95% CI 7.5% to 8.8% | 1,185 counties | labor force of each county | BLS Local Area Unemployment Statistics |   |
| ur_metro_2023 | 3.6% | percent of the civilian labor force that is unemployed | unemployed residents summed over the counties, divided by the labor force summed over the same counties, bootstrap interval over states | residents of metropolitan counties (Puerto Rico left out) | annual average, 2023 | 95% CI 3.3% to 3.9% | 1,185 counties | labor force of each county | BLS Local Area Unemployment Statistics |   |
| ur_p5_2023 | 2.1% | percent of the civilian labor force that is unemployed | fifth percentile of county rates, each county counted once | residents of counties and county equivalents in the LAUS file (Puerto Rico left out) | annual average, 2023 |   | 3,143 counties | none (each county counts once) | BLS Local Area Unemployment Statistics |   |
| ur_p95_2023 | 5.6% | percent of the civilian labor force that is unemployed | ninety-fifth percentile of county rates, each county counted once | residents of counties and county equivalents in the LAUS file (Puerto Rico left out) | annual average, 2023 |   | 3,143 counties | none (each county counts once) | BLS Local Area Unemployment Statistics |   |
| ur_pooled_2023 | 3.6% | percent of the civilian labor force that is unemployed | unemployed residents summed over the counties, divided by the labor force summed over the same counties, bootstrap interval over states | residents of counties and county equivalents in the LAUS file (Puerto Rico left out) | annual average, 2023 | 95% CI 3.3% to 3.9% | 3,143 counties | labor force of each county | BLS Local Area Unemployment Statistics |   |
| ur_second | 14.8% | percent of the civilian labor force that is unemployed | published county estimate (modeled; the county file carries no margin of error) | residents of Kusilvak Census Area, AK | annual average, 2023 |   | 2,326 persons in the labor force | none (one county) | BLS Local Area Unemployment Statistics |   |
| ur_small_2019 | 4.3% | percent of the civilian labor force that is unemployed | unemployed residents summed over the counties, divided by the labor force summed over the same counties, bootstrap interval over states | residents of nonmetropolitan counties with smaller towns or none (Puerto Rico left out) | annual average, 2019 | 95% CI 3.9% to 4.7% | 1,054 counties | labor force of each county | BLS Local Area Unemployment Statistics |   |
| ur_small_2020 | 6.7% | percent of the civilian labor force that is unemployed | unemployed residents summed over the counties, divided by the labor force summed over the same counties, bootstrap interval over states | residents of nonmetropolitan counties with smaller towns or none (Puerto Rico left out) | annual average, 2020 | 95% CI 6.2% to 7.3% | 1,055 counties | labor force of each county | BLS Local Area Unemployment Statistics |   |
| ur_small_2023 | 3.8% | percent of the civilian labor force that is unemployed | unemployed residents summed over the counties, divided by the labor force summed over the same counties, bootstrap interval over states | residents of nonmetropolitan counties with smaller towns or none (Puerto Rico left out) | annual average, 2023 | 95% CI 3.5% to 4.0% | 1,055 counties | labor force of each county | BLS Local Area Unemployment Statistics |   |
| ur_town_2019 | 4.1% | percent of the civilian labor force that is unemployed | unemployed residents summed over the counties, divided by the labor force summed over the same counties, bootstrap interval over states | residents of nonmetropolitan counties with larger towns (Puerto Rico left out) | annual average, 2019 | 95% CI 3.8% to 4.3% | 903 counties | labor force of each county | BLS Local Area Unemployment Statistics |   |
| ur_town_2020 | 7.3% | percent of the civilian labor force that is unemployed | unemployed residents summed over the counties, divided by the labor force summed over the same counties, bootstrap interval over states | residents of nonmetropolitan counties with larger towns (Puerto Rico left out) | annual average, 2020 | 95% CI 6.9% to 7.6% | 903 counties | labor force of each county | BLS Local Area Unemployment Statistics |   |
| ur_town_2023 | 3.7% | percent of the civilian labor force that is unemployed | unemployed residents summed over the counties, divided by the labor force summed over the same counties, bootstrap interval over states | residents of nonmetropolitan counties with larger towns (Puerto Rico left out) | annual average, 2023 | 95% CI 3.4% to 3.9% | 903 counties | labor force of each county | BLS Local Area Unemployment Statistics |   |
| ur_tract_least | 4.9% | percent of the civilian labor force that is unemployed | population-weighted mean of tract values (2019-2023 ACS tract population as the weight), county-bootstrap interval | residents of tracts outside Alaska, Massachusetts and Puerto Rico in the least deprived tenth of tracts by Area Deprivation Index rank | ACS 2019-2023 | 95% CI 4.3% to 5.4% | 8,021 tracts | 2019-2023 tract population | American Community Survey |   |
| ur_tract_most | 8.6% | percent of the civilian labor force that is unemployed | population-weighted mean of tract values (2019-2023 ACS tract population as the weight), county-bootstrap interval | residents of tracts outside Alaska, Massachusetts and Puerto Rico in the most deprived tenth of tracts by Area Deprivation Index rank | ACS 2019-2023 | 95% CI 8.2% to 9.0% | 8,103 tracts | 2019-2023 tract population | American Community Survey |   |
| wfh_least | 23.4% | percent of workers who work from home | population-weighted mean of tract values (2019-2023 ACS tract population as the weight), county-bootstrap interval | workers living in tracts outside Alaska, Massachusetts and Puerto Rico in the least deprived tenth of tracts by Area Deprivation Index rank | ACS 2019-2023 | 95% CI 22.0% to 25.3% | 8,021 tracts | 2019-2023 tract population | American Community Survey |   |
| wfh_most | 5.9% | percent of workers who work from home | population-weighted mean of tract values (2019-2023 ACS tract population as the weight), county-bootstrap interval | workers living in tracts outside Alaska, Massachusetts and Puerto Rico in the most deprived tenth of tracts by Area Deprivation Index rank | ACS 2019-2023 | 95% CI 5.7% to 6.2% | 8,103 tracts | 2019-2023 tract population | American Community Survey |   |
| ws_highest_fifth | 49.8% | percent of jobs paying more than the top-band cut (monthly earnings from the job) | jobs in the top earnings band summed over the counties of the group, divided by jobs summed over the same counties, bootstrap interval over states | jobs located in the highest fifth of counties by jobs located per job held by residents (Alaska and Massachusetts left out) | 2019 | 95% CI 47.8% to 51.5% | 620 counties | jobs in each table | LEHD Origin-Destination Employment Statistics |   |
| ws_lowest_fifth | 39.5% | percent of jobs paying more than the top-band cut (monthly earnings from the job) | jobs in the top earnings band summed over the counties of the group, divided by jobs summed over the same counties, bootstrap interval over states | jobs located in the lowest fifth of counties by jobs located per job held by residents (Alaska and Massachusetts left out) | 2019 | 95% CI 36.7% to 42.1% | 619 counties | jobs in each table | LEHD Origin-Destination Employment Statistics |   |

## Definitions

- Ecological inference: Drawing conclusions about individuals from data about places or groups. A pattern across counties need not hold for the people inside them.
- Census tract: A small area drawn by the Census Bureau to stand in for a neighborhood. Boundaries are redrawn after each decennial census.
- Area Deprivation Index (ADI): A score that ranks neighborhoods by income, education, employment and housing conditions drawn from the American Community Survey. In this book a higher national percentile means a more deprived neighborhood.

## Methods

This chapter compares where residents live with where jobs are, using residents (LAUS, the ACS and the LODES residence table), workplaces (QCEW and the LODES workplace table) and the LODES flows between home county and work county. LODES counts jobs, not people, and its earnings come in three fixed bands in current dollars, so shares are compared within one year. Alaska and Massachusetts are left out of every LODES figure because a workplace file is absent or a stub, and counties enter the scatter and correlations only with enough jobs in each table.

### Residents, workplaces and flows

Three populations are kept apart. Residents are counted where they live (LAUS, the ACS, and the LODES residence table, which places a job at the worker's home). Workplaces are counted where the job is (QCEW and the LODES workplace table). Flows are the LODES origin-destination pairs of home county and work county.

LODES counts jobs, not people: its job is a worker-employer match with positive earnings in the second quarter and the quarter before, and a worker with two jobs is counted twice.

LODES earnings come in three fixed bands of monthly earnings in current dollars (up to $1,250, $1,251 to $3,333, above $3,333); 'top band' means above $3,333. Because the bands do not move with prices, shares are compared within one year.

Explains Table 4.

### Which states and years LODES covers

LODES scope: a state-year is incomplete when its workplace table is absent or holds under half of its residence-table jobs (found from the jobs, since a stub file counts as a file).

Alaska (no 2019 workplace file) and Massachusetts (a stub workplace file for 2010, with 17 jobs) are left out of every LODES figure, so 2010, 2015 and 2019 cover the same states; 2019 is used for residents against workplaces, and 2022 is not used because Alaska and Michigan lack workplace files.

### County types and how rates and pay are measured

County types follow the 2023 USDA Rural-Urban Continuum Code of the county: 1 to 3 metropolitan, 4 to 7 nonmetropolitan with an urban population of 5,000 or more, 8 and 9 nonmetropolitan with an urban population under 5,000. Connecticut's legacy counties (used by QCEW to 2023 and by LODES) take the type of the planning region that holds most of their population; the retired Alaska codes of LAUS take the type of their successors.

LAUS pooled rates are unemployed residents over labor force summed over counties, with Puerto Rico left out; county distributions count each county once. The 7 Louisiana parishes without an annual average in 2005 and 2006 leave those two years.

QCEW pay is total wages over average employment; a county whose totals the Bureau withholds is missing, never zero, and is left out.

Explains Table 2, Table 1 and Table 3.

### Weights, fifths and tenths

County correlations are weighted by 2020 county population (ERS file). Tract rows weight ACS measures by tract population and LODES shares by jobs.

County fifths hold near-equal numbers of counties (they differ by at most one), tract fifths and tenths near-equal numbers of tracts (tied ADI ranks stay together). Cuts: counties enter the scatter and correlations with at least 1,000 jobs in each table; a tract cell is drawn with at least 30 tracts.

Explains Table 5, Table 6, Table 7 and Table 8.

### How uncertainty is measured

Intervals are 95% percentile intervals from 2,000 draws of whole units with replacement (seed 12345): states for county figures, because counties in a state share policies and labor markets, and counties for tract figures.

A difference between two pooled estimates uses one set of draws.

## What these data cannot show

All the patterns here are descriptions of counties and tracts. The pooled rate of a county type, the share of jobs in the top band for a group of counties and the gradient across groups of tracts say nothing certain about any worker.

Unemployment and pay come from different groups. LAUS and the ACS count residents, QCEW counts jobs located in a county, and LODES counts jobs at either end of the commute, so no figure here gives the pay of a county's residents in dollars. LODES shows only three bands. A fixed band in current dollars holds more jobs each year as prices and pay rise, so the shares apply to 2019 and say nothing about change over time. The same dollar cut also buys different amounts in different places, so top-band shares are not comparisons of purchasing power. Counting jobs and not people means a worker with two jobs is counted once for each, and noise added for confidentiality makes the shares of small counties and tracts imprecise.

The county unemployment estimates are modeled, and the county file carries no margin of error. The intervals here reflect which places are in each group, and a high rate in a county with a small labor force can rest on few people. Connecticut is drawn as its legacy counties, and the unemployment values there are spread from planning regions by population, so they are approximate. Puerto Rico's 78 municipios are in the LAUS file and are not drawn. The ACS tract file has no margins of error. A cell of county type by fifth of tracts is drawn only with at least 30 tracts, which leaves out the least deprived fifth of the nonmetropolitan counties with smaller towns or none. The ADI averages block-group ranks without population weights, leaves 1,580 tracts without a rank, and understates deprivation in high-cost cities. Jobs in tracts with no rank are left out of the tract figures: 3.1% of jobs located and 0.3% of jobs held by residents. A tract has no rank when every block group in it is suppressed, for example for group quarters or small populations.

LODES is incomplete for some state-years, and the residence table is almost empty where a state has no workplace table or only a stub. Alaska has no workplace file for 2019, and its residence table holds 3,090 jobs for that year against 308,964 in 2015. The Massachusetts workplace file for 2010 is a stub of 2 tracts and 17 jobs, and its residence table holds 151,718 jobs against 3,330,566 in 2015. A stub reads as a complete file unless the jobs are compared, so the build compares workplace jobs with residence jobs. Alaska and Massachusetts are left out of every LODES figure, so every year of the commuting trend covers the same states. The 2022 files lack Alaska and Michigan, so 2019 is the year used for residents against workplaces. Puerto Rico is not drawn and is not in the LODES tables used here.

The comparison of residents with workplaces is for counties and tracts. It cannot say who holds the better-paid jobs in a deprived neighborhood, whether commuting costs or hiring practices keep residents from them, or how many workers would take them if they could.

## References

1. U.S. Bureau of Labor Statistics (2018). Local Area Unemployment Statistics: Calculation. BLS Handbook of Methods (page last modified January 9, 2018). https://www.bls.gov/opub/hom/lau/calculation.htm
2. U.S. Bureau of Labor Statistics (2026). Quarterly Census of Employment and Wages: Concepts. BLS Handbook of Methods (page last modified August 31, 2026). https://www.bls.gov/opub/hom/cew/concepts.htm
3. U.S. Census Bureau, Longitudinal Employer-Household Dynamics Program (2025). LEHD Origin-Destination Employment Statistics (LODES) Dataset Structure, Format Version 8.4. U.S. Census Bureau technical documentation (revised December 3, 2025). https://lehd.ces.census.gov/data/lodes/LODES8/LODESTechDoc8.4.pdf
4. U.S. Census Bureau, Longitudinal Employer-Household Dynamics Program (2025). OnTheMap: Data Overview (LODES Version 8). U.S. Census Bureau technical documentation (revised December 2, 2025). https://lehd.ces.census.gov/doc/help/onthemap/OnTheMapDataOverview.pdf
5. Kind, Amy J. H.; Buckingham, William R. (2018). Making Neighborhood-Disadvantage Metrics Accessible: The Neighborhood Atlas. New England Journal of Medicine 378(26): 2456-2458. https://doi.org/10.1056/NEJMp1802313
6. McKenzie, Brian; Burd, Charlynn (2015). Where Do All Those Commuters Come From? Census Bureau Releases New Information on County-to-County Commuting Flows. U.S. Census Bureau, Random Samplings blog (August 13, 2015). https://www.census.gov/newsroom/blogs/random-samplings/2015/08/where-do-all-those-commuters-come-from-census-bureau-releases-new-information-on-county-to-county-commuting-flows.html
7. Kain, John F. (1968). Housing Segregation, Negro Employment, and Metropolitan Decentralization. The Quarterly Journal of Economics 82(2): 175-197. https://doi.org/10.2307/1885893

## Provenance

- Pipeline version: 0.1.0
- Build script: book2/chapters/work-wages-and-access-to-jobs/build.py (sha256 f00cafe2fed8abc3bed1fc6041d17b3d0b42918528b3d6c3238b75014c999a13)
- Built on: 2026-10-09
- Site commit: c2b1ff3309cb4a10833f884da91ac0057a66ad0a

Inputs:

| File | Size (bytes) | SHA-256, first 1 MB | SHA-256, last 1 MB |
| --- | ---: | --- | --- |
| book2/chapters/work-wages-and-access-to-jobs/analysis.py | 27,069 | dd0dd2d305b58c166b0fcfbc42df729b709db23ab7b52799634511e79e7b0879 | dd0dd2d305b58c166b0fcfbc42df729b709db23ab7b52799634511e79e7b0879 |
| book2/chapters/work-wages-and-access-to-jobs/loaders.py | 17,075 | 4cf5e816ea25e48714e6059fcf69151891b70612395326a196814d787506f667 | 4cf5e816ea25e48714e6059fcf69151891b70612395326a196814d787506f667 |
| data/analytic/tract_atlas.parquet | 39,659,925 | 29d5e4d943552d54aeb6ba92aadc939d9bd15bdea553c78357dbd516757be627 | 1e10d1ecf722340f23ee2212d4999a90250f85b65fa89b39e32db06d9472e22e |
| data/staged/crosswalks/ct_old_county_to_planning_region.parquet | 5,797 | 5fb0d280ff6c9a9f160af7f03934db4682efc3c5fda11ae27b7f614d6ad664bb | 5fb0d280ff6c9a9f160af7f03934db4682efc3c5fda11ae27b7f614d6ad664bb |
| data/staged/crosswalks/ct_tract2020_to_old_county.parquet | 15,842 | 961ef39f31596ef65fca2c67d5f2f63e533c5dc9b49745d3bb6437844e4b831d | 961ef39f31596ef65fca2c67d5f2f63e533c5dc9b49745d3bb6437844e4b831d |
| data/staged/crosswalks/ers_rucc_county_2023.parquet | 67,278 | ad01d36a53664a2bfafbcbebb0f5a921f5b81122ffa80cf96333db23b41c191e | ad01d36a53664a2bfafbcbebb0f5a921f5b81122ffa80cf96333db23b41c191e |
| data/staged/employment/bls_laus_county_annual_1990_2024.parquet | 1,355,480 | 5e2e296cc88984e2a157354b38f72e229d8c4ae74f589276d868ad8f14ce44d0 | 334ac972a9bf6b95a0d40e17bb9c695fa175310f2851152da935c59647c77d05 |
| data/staged/employment/bls_qcew_county_annual_1990_2024.parquet | 52,543,256 | 6fba69010f08637b38b2389e191170157a54601021b01a0692c605568fcf4ce1 | 970774e9a6fc43395a122dc270d1d49544056d856c855d23c724724e01b2dde4 |
| data/staged/employment/lodes_od_county_2002_2022.parquet | 36,476,635 | 674aafe40c1b660871291bd02b3ba21e8f0eac1c242b10726e7d51da70941aed | 49d7822b57a1a3bad50d178502a1677e5da5be9dabd6cc0bc538299c9fc98f5f |
| data/staged/employment/lodes_rac_tract2020_2002_2022.parquet | 15,110,704 | 256550e6cc299be7d1646dc9b392f528f947d98f8db5bf5693db11b351d851d5 | 4eafb002800801098de0a90deee8d4f9193a3c3606d702a68fc71f26a10ed2d7 |
| data/staged/employment/lodes_wac_tract2020_2002_2022.parquet | 15,493,609 | 6e3259f5450ba7ce2bf00f801fc7ff956f58911d962fd4da4371f4e6fbdc2ddd | 9ee1d087319d331bd898879656740e0edd6bb05784c454362c929c70faef9220 |

Outputs:

| File | SHA-256 |
| --- | --- |
| facts.json | b01160ee33ee4372a0cbc2dc3dc12c8465e8ace69f89f8e9feba2af810bbac6c |
| figures/commuting_out_of_county.json | 683c4d5b93aa6c11b37b3545ef09e14e5c10587728ef42e1f6ce39c3fb48266e |
| figures/top_band_by_jobs_per_resident.json | fda23651a58c298548ae75c9c35f1c795f75727d60c9f71f8b9c4cf17bb1186a |
| figures/top_band_county_scatter.json | bf062a2860940c5671c5246de9835668a91036ab52b6c31d3ffb40503cdfa916 |
| figures/tract_measures_by_adi.json | 8abafde63d94e7e29b72320f5ebf2d1eaefc2bd3e2e8dce3101a2c97534f09d9 |
| figures/tract_top_band_by_county_type.json | 415f0fb007c61369c5245eb50043f20867b85b34f9c4c5bc9bead0a3d0626f4c |
| figures/unemployment_by_county_type.json | d9250bf1f40af7ecd1a42f0b13953847121c63093481153712212a8709f951b4 |
| figures/unemployment_map.json | 80468b06f812f8c4dfcfc20edf92719dac336610e429c75499d9e8d577326fed |
| figures/workplace_pay_map.json | 084e4bd75821b64516a1737e25f733b7917a4eb085445269c53475625ea2f3c7 |
| meta.json | cbe4425280fbe28b4a3c30dcf71263ffdd79ce91c54a09618ff4777859f24b4a |
| public/data/work-wages-and-access-to-jobs/top_band_county_scatter.json | b1bd58ca09f35430d4c488fb8a3a4eec982cdb450dd12243a809141afe92fe41 |
| public/data/work-wages-and-access-to-jobs/unemployment_map.json | 0fa4b43c602e0e8b2d5864f3fc623153267425458b18c3db3de5eca4ed1c4f11 |
| public/data/work-wages-and-access-to-jobs/workplace_pay_map.json | 9c4aa13c9cf120169ba7d20856beabfbf0ed90adaf5857b9415e2d43a8137202 |
| refs.json | 94c0fb63d39904d544ccb6a362c95fd2cabcfbe5946e751cd9e095ec12ac1704 |

## Machine formats

- Report (HTML): https://vishalsingh.org/zipcode/work-wages-and-access-to-jobs/report
- Report (Markdown): https://vishalsingh.org/zipcode/work-wages-and-access-to-jobs/report.md
- Every estimate (JSON): https://vishalsingh.org/zipcode/report/work-wages-and-access-to-jobs/facts.json
- Table 1, figure "unemployment_map" (JSON): https://vishalsingh.org/zipcode/report/work-wages-and-access-to-jobs/figures/unemployment_map.json
- Table 2, figure "unemployment_by_county_type" (JSON): https://vishalsingh.org/zipcode/report/work-wages-and-access-to-jobs/figures/unemployment_by_county_type.json
- Table 3, figure "workplace_pay_map" (JSON): https://vishalsingh.org/zipcode/report/work-wages-and-access-to-jobs/figures/workplace_pay_map.json
- Table 4, figure "commuting_out_of_county" (JSON): https://vishalsingh.org/zipcode/report/work-wages-and-access-to-jobs/figures/commuting_out_of_county.json
- Table 5, figure "top_band_by_jobs_per_resident" (JSON): https://vishalsingh.org/zipcode/report/work-wages-and-access-to-jobs/figures/top_band_by_jobs_per_resident.json
- Table 6, figure "top_band_county_scatter" (JSON): https://vishalsingh.org/zipcode/report/work-wages-and-access-to-jobs/figures/top_band_county_scatter.json
- Table 7, figure "tract_measures_by_adi" (JSON): https://vishalsingh.org/zipcode/report/work-wages-and-access-to-jobs/figures/tract_measures_by_adi.json
- Table 8, figure "tract_top_band_by_county_type" (JSON): https://vishalsingh.org/zipcode/report/work-wages-and-access-to-jobs/figures/tract_top_band_by_county_type.json
- Payload of "unemployment_map" (JSON): https://vishalsingh.org/zipcode/data/work-wages-and-access-to-jobs/unemployment_map.json
- Payload of "workplace_pay_map" (JSON): https://vishalsingh.org/zipcode/data/work-wages-and-access-to-jobs/workplace_pay_map.json
- Payload of "top_band_county_scatter" (JSON): https://vishalsingh.org/zipcode/data/work-wages-and-access-to-jobs/top_band_county_scatter.json
- Index of every report: https://vishalsingh.org/zipcode/llms.txt; all reports in one file: https://vishalsingh.org/zipcode/llms-full.txt

---

# Home value, income and redlining: report

Chapter 5.2 of ZIP Code Destiny, by Vishal Singh (NYU Stern). Generated from the chapter's data files; the reading version is at https://vishalsingh.org/zipcode/home-value-income-and-redlining. Table n of this report is figure n of the chapter.

## Identification

- Title: Home value, income and redlining
- Label: The deed and the ledger
- Chapter: 5.2 (Part V, Home, work and money; An American life)
- Question: How much longer do people live where homes are worth more, and how do the 1930s redlining grades line up with outcomes today?
- Dek: Across census tracts, each counted once, life expectancy at birth for deaths in 2010 to 2015 is 2.2 years higher for each doubling of the median home value for 2019 to 2023.
- Grain: neighborhoods
- Status: published
- Published: 2026-10-09
- Updated: 2026-10-28
- Data sources:
  - USALEEP life expectancy. Life expectancy at birth by tract, deaths in 2010 to 2015, area-weighted onto 2020 tracts; years: 2010-2015; weights: ACS 2019-2023 tract population; each tract once for the unweighted slope
  - American Community Survey. ACS 2019 to 2023: median home value by tract, tract population, homeownership, household income and renter households; years: 2019-2023 (5-year); weights: 2019-2023 tract population
  - Census TIGER/Line tracts. Census 2020 county file: the core-based statistical area code of each county (CBSAFP), which assigns tracts to metros; years: 2020; weights: none (assignment of counties to metros)
  - IRS Statistics of Income. IRS Statistics of Income county file, tax year 2022: returns, returns claiming the earned income tax credit and returns by adjusted gross income bracket; years: tax year 2022; weights: none (full counts of returns)
  - Area Deprivation Index. Area Deprivation Index 2023, national rank, tracts averaged to counties; years: 2023 release (version 4.0.1; built on ACS 2019-2023); weights: ACS 2019-2023 tract population
  - Opportunity Atlas. Adult household income rank of children born 1978 to 1983 to parents at the 25th income percentile, county and 2020-tract files; years: 1978-1983 birth cohorts (2018 release); weights: county population; tract population
  - Mapping Inequality (HOLC maps). Mapping Inequality HOLC residential security maps, 1935 to 1940, graded polygons overlaid on 2020 tracts; years: 1930s-1940s maps (2023 digitization); weights: ACS 2019-2023 tract population
- Canonical URL: https://vishalsingh.org/zipcode/home-value-income-and-redlining
- Report: https://vishalsingh.org/zipcode/home-value-income-and-redlining/report
- Markdown: https://vishalsingh.org/zipcode/home-value-income-and-redlining/report.md
- BibTeX key: singh2026homevalueincomeandredlining

```bibtex
@misc{singh2026homevalueincomeandredlining,
  author = {Singh, Vishal},
  title = {Home value, income and redlining},
  year = {2026},
  url = {https://vishalsingh.org/zipcode/home-value-income-and-redlining},
  note = {In ZIP Code Destiny. Published 9 October 2026; updated 28 October 2026.}
}
```

## Principal estimates

The facts the dek and the opening prose cite, in order of appearance.

| Estimand | Population | Period | Value | Interval | n | Weights | Source | Fact id |
| --- | --- | --- | ---: | --- | ---: | --- | --- | --- |
| slope of tract life expectancy on log2 tract median home value, each tract counted once, county-cluster bootstrap interval | non-institutional 2020 census tracts with a life expectancy estimate and an ACS median home value above $10,000 (50 states and DC) | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); home values ACS 2019-2023 (tract medians) | 2.2 | 95% CI 2.1 to 2.3 | 76,702 tracts | none (each tract counts once) | USALEEP life expectancy | price_doubling_yrs |
| count of tracts in the regression | non-institutional 2020 census tracts with a life expectancy estimate and an ACS median home value above $10,000 (50 states and DC) | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); home values ACS 2019-2023 (tract medians) | 76,702 |   |   | none (full count) | USALEEP life expectancy | price_n_tracts |

## Findings

Life expectancy at birth is 2.2 (95% CI 2.1 to 2.3) years higher for each doubling of the median home value across 76,702 census tracts when each tract counts once. Life expectancy is the USALEEP estimate for deaths in 2010 to 2015, and each tract's median home value is from the American Community Survey for 2019 to 2023. The slope describes places. It does not say that a household that moved to a costlier area would live longer.

The chapter then turns to two other records that follow the address. One is the share of tax returns that claim the earned income tax credit or report a high income. The other is the set of grades that the Home Owners' Loan Corporation gave neighborhoods in the 1930s. Every comparison is between places (census tracts and counties), and none describes individual households. Eviction filings and the housing cost burden of renters are in Chapter 5.4, Renting, affordability and eviction (https://vishalsingh.org/zipcode/renting-affordability-and-eviction).

### Home values rise with life expectancy

Each census tract is paired with its own median home value, with institutional tracts such as prisons and nursing facilities left out. An earlier version of this chapter compared ZIP code areas and used a ZIP-level median home value of unstated, older vintage, which it labeled as the 2019 to 2023 survey. This version uses the tract values from that survey. Across the tracts, life expectancy has a correlation of +0.59 (95% CI 0.56 to 0.61) with the logarithm of the home value and +0.50 (95% CI 0.48 to 0.52) with the home value in dollars. It moves more closely with the logarithm, so the slope is stated per doubling of the home value. When tracts count in proportion to their population, the slope is 2.2 (95% CI 2.1 to 2.3) years per doubling, nearly the same.

Ranking tracts by home value and averaging life expectancy within twentieths of the tracts, weighted by population, gives 73.5 (95% CI 73.1 to 73.9) years in the lowest twentieth and 82.8 (95% CI 82.5 to 83.2) years in the highest. The difference is 9.3 (95% CI 8.8 to 9.8) years. Economists call the price of a bundled feature a hedonic price. The price of a house is the sum of what buyers pay for each part of the address, such as bedrooms, schools and the commute.[1] Life expectancy is not a feature that a buyer can see or purchase, so the slope is not a price that anyone pays. It records that homes cost more where the people nearby live longer, and the data do not separate what an area provides from who can afford to buy there.

Inverting the slope within a metro gives a local exchange rate. Each tract counts in the metro of its county, and the sample has 39 metros with at least 250 tracts and 1.5 million residents. In each, the median home value of its tracts is divided by the within-metro slope of life expectancy on the logarithm of home value. The result is the housing dollars that separate a pair of tracts one year of life apart near the metro median.

A year costs $33 thousand (95% CI $30 thousand to $38 thousand) in Cincinnati, the metro with the lowest point estimate, and $658 thousand (95% CI $482 thousand to $947 thousand) in the San Jose metro, the highest (Table 1). The ratio between those two estimates is 19.6 (95% CI 14.1 to 28.7). That ratio compares the extremes of 39 noisy estimates, and its interval holds those metros fixed, so it understates how far apart the extremes are.

The point estimate is above $100 thousand in 16 metros, every California metro in the comparison among them, and the intervals for Sacramento, Portland and Las Vegas include that amount. Detroit has the largest within-metro correlation between life expectancy and the logarithm of home value, +0.78 (95% CI 0.75 to 0.80), and a year costs $44 thousand (95% CI $42 thousand to $47 thousand) there. The interval for Cleveland, the next largest, overlaps Detroit's, so the order is by point estimate.

Table 1. Life expectancy is 2.2 years higher for each doubling of tract home value, and a year of life costs $33 thousand in housing dollars in Cincinnati and $658 thousand in San Jose

The exchange rate follows the price level more than the strength of the link. Across the metros, dollars per year of life correlate +0.89 (95% CI 0.82 to 0.96) with the metro's median tract home value, so a year is cheap where homes are cheap and dear where homes are dear. Part of that correlation is built in, because dollars per year is the median home value divided by the slope. The figures say more about what housing costs than about how steeply longevity rises with value. Read them as orders of magnitude, because the tracts of one metro resemble one another and the interval for a single metro is narrower than the uncertainty.

### Income brackets change shape with deprivation

Whether a household can buy where homes are dear depends on its income. The income tax returns filed in a county describe the incomes of everyone who files there, and they do not identify who bought a home. A median summarizes a distribution of incomes with one number, and the returns describe it by bracket. The comparison covers 3,129 counties, with Connecticut's planning regions standing in for its counties, ranked by the Area Deprivation Index. Tenths hold equal numbers of counties, not equal numbers of people. The least deprived tenth holds 130.1 million people and the most deprived tenth 3.8 million.

The share of individual income tax returns for tax year 2022 that claim the earned income tax credit rises from 12.0% (95% CI 10.4% to 13.2%) in the least deprived tenth of counties to 28.9% (95% CI 26.4% to 31.2%) in the most deprived tenth (Table 2). The share of returns with an adjusted gross income of $200,000 or more falls from 11.1% (95% CI 10.3% to 11.9%) to 1.9% (95% CI 1.7% to 2.1%).

Nationally, 14.9% of returns claim the credit and 7.8% are in the top bracket. Across counties weighted by population, the EITC share correlates +0.57 (95% CI 0.43 to 0.72) with deprivation and the top-bracket share correlates −0.74 (95% CI −0.82 to −0.69). The ranking and the shares overlap in what they measure, because the Area Deprivation Index is built partly from income and poverty, so some of the gradient is built in.

The brackets change shape across the range and not only in their average. In the least deprived fifth of counties, 26.7% of returns report an adjusted gross income under $25,000 and 10.0% report $200,000 or more. In the most deprived fifth the shares are 37.6% and 2.3%. The lowest income group is the largest of the five groups in every fifth of counties, and the data do not say who is in it.

Table 2. The share of tax returns that claim the EITC rises from 12.0% to 28.9% across tenths of counties by deprivation, while the share with adjusted gross income of $200,000 or more falls from 11.1% to 1.9%

Both shares exceed the national shares in only 15 of the 447 counties that have at least 150,000 residents, among them Fulton County and DeKalb County in Georgia and Essex County in New Jersey. In most of the other large counties one tail is heavier than the nation's and the other lighter.

The shares also line up with the mobility of children from low-income families, measured by the income rank that Chapter 4.3, Race, income and mobility (https://vishalsingh.org/zipcode/race-income-and-mobility) defines. In a county-population-weighted regression of standardized variables, the top-bracket share alone has a coefficient of 0.41 (95% CI 0.33 to 0.50). With both shares and the deprivation index in the model, the coefficient on the top-bracket share is −0.03 (95% CI −0.27 to 0.20), an interval that includes zero. The coefficient on the EITC share is −0.49 (95% CI −0.75 to −0.23), which stays away from zero. The two shares correlate −0.65 (95% CI −0.70 to −0.60) across counties, so the regression cannot split their common signal cleanly. All of this is association across counties.

### Grades drawn in the 1930s and neighborhoods today

Between 1935 and 1940 the Home Owners' Loan Corporation drew residential security maps for cities across the country and graded neighborhoods from A, labeled best, to D, labeled hazardous.[2] The area descriptions that came with the maps show that the race and national origin of residents were among the factors in a grade, next to the housing stock and incomes.[3] The maps have long been out of use, and the grades still line up with today's neighborhoods.

Overlaying the graded polygons on 2020 census tracts gives 10,952 tracts in 222 HOLC cities and city areas where graded polygons cover at least 50% of the tract. Each tract takes the grade that covers most of its area.

Life expectancy at birth, for deaths in 2010 to 2015, averages 80.9 (95% CI 80.1 to 81.6) years in grade A tracts and 76.8 (95% CI 75.5 to 77.8) years in grade D tracts, a gap of 4.1 (95% CI 3.3 to 5.1) years with tracts counted in proportion to their population (Table 3). The gap is 3.6 (95% CI 2.6 to 4.9) years when graded polygons must cover at least 90% of a tract, so it moves with the matching rule and stays positive.

Grade A tracts are in cities whose graded tracts have a lower average life expectancy than the cities of grade D tracts. Comparing tracts within the same city, with a separate baseline for each city (fixed effects), widens the gap to 5.8 (95% CI 5.3 to 6.4) years.

Other measures step in the same direction. The adult income rank of children from low-income families averages 46.3 (95% CI 45.3 to 47.1) in grade A tracts and 37.9 (95% CI 36.3 to 39.2) in grade D tracts, in percentile points of the national household income distribution. The rank is from the Opportunity Atlas and describes children born in 1978 to 1983.

Homeownership is 67.8% (95% CI 65.0% to 70.6%) of households in grade A tracts and 35.3% (95% CI 31.7% to 39.7%) in grade D tracts. The mean of tract median household incomes is $127 thousand (95% CI $115 thousand to $138 thousand) and $71 thousand (95% CI $63 thousand to $78 thousand) in the two grades. These comparisons describe neighborhoods today, and the two grades differ in housing, income and population from before the maps were drawn.

Table 3. Tracts graded A on the 1930s HOLC maps have a life expectancy of 80.9 years, against 76.8 years in tracts graded D, for deaths in 2010 to 2015

 (Note: The Area Deprivation Index is close to the same value in grade B, C and D tracts (41, 42 and 41) while life expectancy falls from grade B to grade D. Several of the census measures behind the index are home values, rents and mortgage payments, so places with high housing costs read as less deprived (Chapter 1.1, What the deprivation index measures (https://vishalsingh.org/zipcode/anatomy-of-deprivation)). These data do not test whether that explains the flat pattern.)

The grades shift the middle of the distribution and do not sort tracts into separate groups. In 75% of random pairs of one grade A tract and one grade D tract, drawn in proportion to population, the grade A tract has the higher life expectancy. In the remaining pairs the grade D tract does. The interquartile ranges of neighboring grades overlap, and so do those of grades A and D.

Within cities the gap has the same sign wherever it can be measured. Among the 27 cities with at least 5 grade A and 10 grade D tracts that have a life expectancy, the gap is positive in every one. Its median is 5.7 years, and it ranges from 3.4 to 10.1 years between the 5th and 95th percentiles of cities. A city's grade A mean rests on few tracts, so single cities are noisy.

> Causal evidence: Whether the HOLC maps changed the neighborhoods they graded (Boundary design with propensity-score weighting; comparison of cities around a population cutoff)
>
> The comparisons above cannot separate the effect of the maps from the neighborhoods the maps graded, because surveyors graded areas on housing, incomes and race. Aaronson, Hartley and Mazumder study the maps with decennial census data.[3] They compare changes over time in outcomes between the two sides of a grade border, within a few blocks, using census data from before the maps as well as after. Because borders already differed before the maps, they weight comparison borders by propensity scores so that the earlier patterns in race and housing match, and they separately use borders that were unlikely to have been drawn along neighborhood differences. A third analysis compares cities just below and just above the population cutoff that decided whether a city was mapped.
>
> They find that the maps reduced homeownership rates, house values and rents and increased racial segregation in later decades, and the comparison of cities around the cutoff gives broadly similar results. The paper does not study life expectancy or children's incomes, and its border estimates are local to areas beside a border.

### What these data cannot show

Every comparison in this chapter describes places. A county or a tract is not a person, and a pattern across places need not hold for the people inside them (ecological inference). The residents of a tract differ from movers in income, health and age, and the data cannot separate selection from what an area provides.

The price of a year joins life expectancy for deaths in 2010 to 2015 to home values for 2019 to 2023, so the two describe different years. The life expectancy estimates are modeled for small areas and area-weighted from 2010 to 2020 tracts. A tract median home value rests on few homes and the tract file carries no margins of error, so part of the spread across tracts is sampling error. The dollar figures divide a price level by a slope that is itself estimated with error.

The IRS file covers people who file a return, so the shares are shares of returns and leave out people who do not file. The EITC share depends on eligibility and on whether eligible households claim the credit, and the top-bracket share is a share of returns, not of people. Connecticut is absent from the mobility regressions because its planning-region codes do not match the Atlas's counties.

The grades followed the neighborhoods' housing, incomes and populations in the 1930s, so the staircase today combines what the neighborhoods were with what happened after. The boundary study in the box above finds that the maps themselves mattered for housing outcomes and segregation next to borders, and it does not show that the grades explain today's differences in life expectancy. Tracts are matched to grades by area, and the Atlas ranks are interpolated by area from 2010 tracts. The maps cover the cities that HOLC surveyed, so a city that is absent was not surveyed, which does not mean it had no redlining.

## Tables

### Table 1. Life expectancy is 2.2 years higher for each doubling of tract home value, and a year of life costs $33 thousand in housing dollars in Cincinnati and $658 thousand in San Jose

Life expectancy at birth (deaths in 2010 to 2015) against median home value (ACS 2019 to 2023) across census tracts, and the housing dollars per year of life in each large metro

#### Table 1.1. Census tracts

Life expectancy is higher in tracts with higher home values, a correlation of +0.59 with the log of home value

Life expectancy at birth in years against median home value in dollars on a log scale, one dot per census tract (random sample), with population-weighted means for twentieths of tracts

No inline rows: the values of Table 1.1 are in https://vishalsingh.org/zipcode/data/home-value-income-and-redlining/home_value_scatter.json.

#### Table 1.2. Dollars per year of life

A year of life costs between $33 thousand and $658 thousand in housing dollars at the lowest and highest point estimates among large metros, a ratio of 19.6

Median tract home value divided by the within-metro slope of life expectancy on the log of home value, thousands of dollars per year of life, metros with enough tracts

| conf_high | conf_low | correlation | dollars_k | metro | n |
| ---: | ---: | ---: | ---: | --- | ---: |
| 37.55 | 30.32 | 0.713 | 33.50 | Cincinnati | 514 |
| 39.58 | 32.99 | 0.766 | 36.05 | Cleveland | 581 |
| 41.56 | 32.15 | 0.700 | 36.46 | Indianapolis | 451 |
| 41.44 | 34.14 | 0.743 | 37.60 | Columbus, OH | 466 |
| 44.86 | 36.61 | 0.674 | 40.32 | Pittsburgh | 672 |
| 46.80 | 38.91 | 0.759 | 42.56 | St. Louis | 663 |
| 47.92 | 39.11 | 0.755 | 43.39 | Kansas City | 552 |
| 50.63 | 38.75 | 0.761 | 44 | Jacksonville | 318 |
| 47.10 | 42.07 | 0.778 | 44.42 | Detroit | 1,214 |
| 57.66 | 40.14 | 0.653 | 47.65 | Virginia Beach-Norfolk | 428 |
| 60.15 | 51.65 | 0.743 | 55.92 | Baltimore | 660 |
| 64.51 | 51.84 | 0.673 | 57.89 | Charlotte | 594 |
| 62.47 | 54.77 | 0.608 | 58.45 | Chicago | 2,159 |
| 68.50 | 52.92 | 0.610 | 59.66 | San Antonio | 451 |
| 64.08 | 56.52 | 0.695 | 59.96 | Philadelphia | 1,438 |
| 71.95 | 53.88 | 0.504 | 61.95 | Minneapolis-St. Paul | 845 |
| 73.11 | 62.19 | 0.617 | 67.43 | Atlanta | 1,336 |
| 82.12 | 56.15 | 0.521 | 67.66 | Providence | 354 |
| 77.71 | 64.76 | 0.545 | 70.50 | Houston | 1,382 |
| 88.03 | 62.90 | 0.530 | 73.22 | Nashville | 383 |
| 87.29 | 73.01 | 0.615 | 79.85 | Dallas-Fort Worth | 1,449 |
| 97.33 | 71.40 | 0.491 | 83.40 | Tampa | 682 |
| 131.2 | 76.66 | 0.416 | 98.47 | Orlando | 449 |
| 118.5 | 88.09 | 0.547 | 101.6 | Sacramento | 515 |
| 126.7 | 88.37 | 0.464 | 105.1 | Portland, OR | 533 |
| 142.3 | 102.5 | 0.495 | 119.7 | Phoenix | 932 |
| 168.6 | 94.07 | 0.400 | 125.9 | Las Vegas | 467 |
| 148.0 | 115.0 | 0.391 | 130.2 | Miami | 1,347 |
| 172.8 | 112.9 | 0.492 | 137.8 | Riverside-San Bernardino | 929 |
| 167.8 | 116.5 | 0.446 | 139.9 | Denver | 623 |
| 165.8 | 121.6 | 0.605 | 140.4 | Seattle | 823 |
| 176.6 | 134.1 | 0.383 | 153.6 | Washington, D.C. | 1,324 |
| 190.0 | 148.7 | 0.461 | 166.4 | Boston | 1,032 |
| 213.3 | 142.1 | 0.405 | 171.8 | Austin | 432 |
| 242.4 | 205.2 | 0.449 | 222.4 | Los Angeles | 2,829 |
| 279.9 | 186.0 | 0.450 | 227.5 | San Diego | 667 |
| 274.3 | 227.1 | 0.556 | 247.9 | San Francisco-Oakland | 1,024 |
| 357.7 | 275.3 | 0.288 | 311.1 | New York | 4,281 |
| 947.0 | 481.6 | 0.322 | 657.8 | San Jose | 400 |

Note: The sample is 76,702 census tracts that have a life expectancy estimate and a median home value, with institutional tracts such as prisons and nursing facilities left out. Each tract carries its own ACS median home value, which the Census Bureau top-codes at about two million dollars. An earlier version of this figure used ZIP code areas and a ZIP-level home value of unstated, older vintage.

Note: The dots are drawn at random from the tracts inside the plotted range with seed 12345. Every estimate in the chapter uses all tracts.

Note: Twentieths hold equal numbers of tracts. Intervals on the markers resample whole counties; intervals in the second tab resample the tracts of one metro and treat them as independent, so they are narrower than they should be. Each tract takes the metro of its county, and a metro enters with at least 250 tracts and 1.5 million residents in the sample.

Note: Life expectancy is the estimate for deaths in 2010 to 2015 and home values are for 2019 to 2023, so the two describe different years.

Source: USALEEP life expectancy 2010-2015 (non-institutional tracts, area-weighted onto 2020 tracts) and ACS 2019-2023 tract median home value; population-weighted twentieths, unweighted fitted line, modeled estimates

Data: https://vishalsingh.org/zipcode/report/home-value-income-and-redlining/figures/home_value_vs_le.json

### Table 2. The share of tax returns that claim the EITC rises from 12.0% to 28.9% across tenths of counties by deprivation, while the share with adjusted gross income of $200,000 or more falls from 11.1% to 1.9%

Shares of individual income tax returns by tenth of counties on the Area Deprivation Index, and the distribution of returns across five income groups by fifth of counties, tax year 2022

#### Table 2.1. By deprivation tenth

Across tenths of counties, the EITC share rises with deprivation and the top-bracket share falls

Percent of individual income tax returns, county-population-weighted mean, tax year 2022, by tenth of counties on the Area Deprivation Index (1 = least deprived)

| conf_high | conf_low | counties | decile | series | share |
| ---: | ---: | ---: | ---: | --- | ---: |
| 13.19 | 10.39 | 312 | 1 | Returns claiming the EITC | 12.03 |
| 16.07 | 12.76 | 313 | 2 | Returns claiming the EITC | 14.75 |
| 17.66 | 14.05 | 313 | 3 | Returns claiming the EITC | 15.93 |
| 18.40 | 15.68 | 313 | 4 | Returns claiming the EITC | 17.01 |
| 20.10 | 16.71 | 313 | 5 | Returns claiming the EITC | 18.34 |
| 22.56 | 17.61 | 313 | 6 | Returns claiming the EITC | 20.18 |
| 21.23 | 17.57 | 313 | 7 | Returns claiming the EITC | 19.32 |
| 26.24 | 18.63 | 315 | 8 | Returns claiming the EITC | 22.40 |
| 26.09 | 21.03 | 312 | 9 | Returns claiming the EITC | 23.69 |
| 31.16 | 26.41 | 312 | 10 | Returns claiming the EITC | 28.93 |
| 11.87 | 10.34 | 312 | 1 | Returns with adjusted gross income of $200,000 or more | 11.06 |
| 8.08 | 6.73 | 313 | 2 | Returns with adjusted gross income of $200,000 or more | 7.46 |
| 6.31 | 5.30 | 313 | 3 | Returns with adjusted gross income of $200,000 or more | 5.91 |
| 5.24 | 4.55 | 313 | 4 | Returns with adjusted gross income of $200,000 or more | 4.88 |
| 4.38 | 3.66 | 313 | 5 | Returns with adjusted gross income of $200,000 or more | 4.06 |
| 3.69 | 3.20 | 313 | 6 | Returns with adjusted gross income of $200,000 or more | 3.43 |
| 3.20 | 2.80 | 313 | 7 | Returns with adjusted gross income of $200,000 or more | 3 |
| 2.94 | 2.50 | 315 | 8 | Returns with adjusted gross income of $200,000 or more | 2.69 |
| 2.64 | 2.24 | 312 | 9 | Returns with adjusted gross income of $200,000 or more | 2.43 |
| 2.13 | 1.73 | 312 | 10 | Returns with adjusted gross income of $200,000 or more | 1.92 |

#### Table 2.2. Shape of the brackets

The lowest income group makes up a larger share of returns and the highest group a smaller share toward the most deprived fifth of counties

Share of individual income tax returns by adjusted-gross-income group, percent, returns pooled over the counties in each fifth by Area Deprivation Index, tax year 2022

| bracket | counties | place | returns | share |
| --- | ---: | --- | ---: | ---: |
| Under $25,000 | 3,143 | United States | 158707990 | 28.63 |
| $25,000 to $49,999 | 3,143 | United States | 158707990 | 22.94 |
| $50,000 to $99,999 | 3,143 | United States | 158707990 | 24.39 |
| $100,000 to $199,999 | 3,143 | United States | 158707990 | 16.26 |
| $200,000 or more | 3,143 | United States | 158707990 | 7.78 |
| Under $25,000 | 625 | Least deprived fifth of counties | 95107340 | 26.66 |
| $25,000 to $49,999 | 625 | Least deprived fifth of counties | 95107340 | 21.38 |
| $50,000 to $99,999 | 625 | Least deprived fifth of counties | 95107340 | 24.31 |
| $100,000 to $199,999 | 625 | Least deprived fifth of counties | 95107340 | 17.69 |
| $200,000 or more | 625 | Least deprived fifth of counties | 95107340 | 9.97 |
| Under $25,000 | 626 | Second fifth | 34986790 | 29.81 |
| $25,000 to $49,999 | 626 | Second fifth | 34986790 | 24.65 |
| $50,000 to $99,999 | 626 | Second fifth | 34986790 | 24.85 |
| $100,000 to $199,999 | 626 | Second fifth | 34986790 | 15.22 |
| $200,000 or more | 626 | Second fifth | 34986790 | 5.46 |
| Under $25,000 | 626 | Middle fifth | 16064920 | 32.58 |
| $25,000 to $49,999 | 626 | Middle fifth | 16064920 | 25.65 |
| $50,000 to $99,999 | 626 | Middle fifth | 16064920 | 24.40 |
| $100,000 to $199,999 | 626 | Middle fifth | 16064920 | 13.55 |
| $200,000 or more | 626 | Middle fifth | 16064920 | 3.81 |
| Under $25,000 | 628 | Fourth fifth | 8727300 | 34.25 |
| $25,000 to $49,999 | 628 | Fourth fifth | 8727300 | 26.38 |
| $50,000 to $99,999 | 628 | Fourth fifth | 8727300 | 24.25 |
| $100,000 to $199,999 | 628 | Fourth fifth | 8727300 | 12.25 |
| $200,000 or more | 628 | Fourth fifth | 8727300 | 2.87 |
| Under $25,000 | 624 | Most deprived fifth of counties | 3811920 | 37.64 |
| $25,000 to $49,999 | 624 | Most deprived fifth of counties | 3811920 | 26.99 |
| $50,000 to $99,999 | 624 | Most deprived fifth of counties | 3811920 | 22.54 |
| $100,000 to $199,999 | 624 | Most deprived fifth of counties | 3811920 | 10.58 |
| $200,000 or more | 624 | Most deprived fifth of counties | 3811920 | 2.25 |

Note: Tenths and fifths hold equal numbers of counties, not equal numbers of people, so the most deprived tenth holds far fewer returns than the least deprived. The Area Deprivation Index is built from income, poverty, education and housing measures, so a gradient of income-based shares against it is partly built in.

Note: The IRS file covers people who file a return. The EITC share is the share of returns that claim the credit, which depends on eligibility and on filing, and the top-bracket share is the share of returns with adjusted gross income of $200,000 or more. Both are counts of returns with no sampling error; the intervals in the first tab come from differences between states.

Note: 14.9% of all returns claim the EITC and 7.8% are in the top bracket. Counties with no Area Deprivation Index are left out; Puerto Rico is not in the file.

Source: IRS Statistics of Income county file, tax year 2022, and Area Deprivation Index 2023; population-weighted county means, full counts of returns

Data: https://vishalsingh.org/zipcode/report/home-value-income-and-redlining/figures/eitc_vs_top_bracket.json

### Table 3. Tracts graded A on the 1930s HOLC maps have a life expectancy of 80.9 years, against 76.8 years in tracts graded D, for deaths in 2010 to 2015

Tract outcomes by 1930s HOLC grade: means of five measures, the spread of life expectancy, and the gap between grade A and grade D tracts city by city

#### Table 3.1. Means by grade

Life expectancy, adult income rank, homeownership and household income are lower in each step from grade A to grade D

Population-weighted mean of tract values by HOLC grade, each panel in its own units

| conf_high | conf_low | grade | label | mean | n | outcome |
| ---: | ---: | --- | --- | ---: | ---: | --- |
| 81.57 | 80.13 | A | Grade A | 80.93 | 629 | Life expectancy at birth (years) |
| 79.81 | 78.30 | B | Grade B | 79.10 | 2,133 | Life expectancy at birth (years) |
| 78.74 | 76.88 | C | Grade C | 77.95 | 4,710 | Life expectancy at birth (years) |
| 77.79 | 75.50 | D | Grade D | 76.84 | 2,671 | Life expectancy at birth (years) |
| 47.15 | 45.26 | A | Grade A | 46.26 | 672 | Adult income rank of children from low-income families (percentile) |
| 44.14 | 41.99 | B | Grade B | 43.11 | 2,245 | Adult income rank of children from low-income families (percentile) |
| 42.56 | 39.57 | C | Grade C | 41.20 | 5,051 | Adult income rank of children from low-income families (percentile) |
| 39.15 | 36.27 | D | Grade D | 37.93 | 2,974 | Adult income rank of children from low-income families (percentile) |
| 70.58 | 64.96 | A | Grade A | 67.81 | 672 | Homeownership rate (% of households) |
| 56.70 | 48.44 | B | Grade B | 52.55 | 2,246 | Homeownership rate (% of households) |
| 47.75 | 39.33 | C | Grade C | 43.45 | 5,052 | Homeownership rate (% of households) |
| 39.73 | 31.66 | D | Grade D | 35.32 | 2,973 | Homeownership rate (% of households) |
| 138.4 | 114.9 | A | Grade A | 127.4 | 669 | Median household income (thousands of dollars) |
| 96.19 | 82.24 | B | Grade B | 89.54 | 2,236 | Median household income (thousands of dollars) |
| 78.97 | 68.80 | C | Grade C | 74.16 | 5,015 | Median household income (thousands of dollars) |
| 78.02 | 62.77 | D | Grade D | 71.12 | 2,920 | Median household income (thousands of dollars) |
| 38.43 | 25.53 | A | Grade A | 31.55 | 670 | Area Deprivation Index (national percentile, 100 = most deprived) |
| 49.21 | 33.19 | B | Grade B | 40.78 | 2,241 | Area Deprivation Index (national percentile, 100 = most deprived) |
| 52.72 | 33.35 | C | Grade C | 41.77 | 5,040 | Area Deprivation Index (national percentile, 100 = most deprived) |
| 52.98 | 31.56 | D | Grade D | 40.64 | 2,953 | Area Deprivation Index (national percentile, 100 = most deprived) |

#### Table 3.2. Spread of life expectancy

The distributions of tract life expectancy overlap heavily across grades, and the grade shifts the middle

Population-weighted mean of tract life expectancy at birth with the interquartile range (the 25th to the 75th percentile of tracts, weighted by tract population) by HOLC grade, deaths in 2010 to 2015

| grade | label | mean | n | q25 | q75 |
| --- | --- | ---: | ---: | ---: | ---: |
| A | Grade A | 80.93 | 629 | 78.60 | 83.60 |
| B | Grade B | 79.10 | 2,133 | 76.38 | 82 |
| C | Grade C | 77.95 | 4,710 | 75.30 | 81.10 |
| D | Grade D | 76.84 | 2,671 | 73.60 | 80.20 |

#### Table 3.3. City by city

In each of 27 HOLC cities life expectancy is higher in grade A tracts than in grade D tracts, with a median gap of 5.7 years

Population-weighted mean life expectancy at birth of grade D and grade A tracts in each city with enough tracts, deaths in 2010 to 2015

| city | hi_a | hi_d | le_a | le_d | lo_a | lo_d | tracts_a | tracts_d |
| --- | ---: | ---: | ---: | ---: | ---: | ---: | ---: | ---: |
| San Diego, CA | 84.12 | 79.86 | 82.15 | 79.15 | 79.68 | 78.46 | 5 | 37 |
| Houston, TX | 81.99 | 79.72 | 80.90 | 77.65 | 79.63 | 74.83 | 21 | 14 |
| Wichita, KS | 78.66 | 73.57 | 76.45 | 72.54 | 74.46 | 71.42 | 5 | 19 |
| Passaic County, NJ | 83.37 | 79 | 81.39 | 77.42 | 79.20 | 75.69 | 5 | 19 |
| Los Angeles, CA | 84.44 | 79.87 | 83.62 | 79.46 | 82.75 | 79.10 | 48 | 269 |
| Detroit, MI | 79.75 | 73.60 | 77.47 | 72.96 | 75.08 | 72.32 | 20 | 91 |
| San Francisco, CA | 87.22 | 81.58 | 85.44 | 80.91 | 83.56 | 80.26 | 6 | 54 |
| Manhattan, NY | 87.31 | 82.51 | 86.18 | 81.65 | 84.87 | 80.77 | 13 | 114 |
| San Antonio, TX | 80.90 | 75.36 | 79.21 | 74.41 | 77.89 | 73.64 | 5 | 22 |
| Minneapolis, MN | 83.54 | 78.26 | 81.71 | 76.85 | 79.56 | 75.36 | 8 | 15 |
| Union Co., NJ | 83.58 | 78.73 | 82.17 | 77.27 | 80.60 | 75.86 | 9 | 15 |
| Essex Co., NJ | 83.59 | 78.76 | 82.33 | 76.96 | 81.06 | 75.12 | 22 | 35 |
| Fort Worth, TX | 82.73 | 76.16 | 80.32 | 74.87 | 78.45 | 73.44 | 5 | 15 |
| Philadelphia, PA | 80.99 | 74.66 | 79.57 | 73.89 | 78.25 | 73.17 | 26 | 127 |
| Baltimore, MD | 81.72 | 74.17 | 78.62 | 72.57 | 75.50 | 70.88 | 9 | 36 |
| Miami, FL | 83.53 | 77.45 | 82.39 | 75.90 | 81.27 | 74.47 | 8 | 30 |
| St. Louis, MO | 82.03 | 74.84 | 80.36 | 73.73 | 78.71 | 72.65 | 17 | 14 |
| Des Moines, IA | 80.64 | 74.71 | 79.69 | 72.98 | 78.70 | 71.24 | 7 | 12 |
| Chicago, IL | 84.67 | 77.10 | 83.34 | 76.49 | 81.94 | 75.85 | 13 | 210 |
| Columbus, OH | 81 | 74.48 | 78.88 | 71.71 | 76.40 | 68.86 | 10 | 10 |
| Oakland, CA | 86.88 | 78.84 | 85.20 | 77.84 | 83.69 | 76.94 | 11 | 39 |
| Cleveland, OH | 81.62 | 72.58 | 80.18 | 71.71 | 78.76 | 70.89 | 23 | 43 |
| Pittsburgh, PA | 83.27 | 74.53 | 82 | 73.09 | 80.80 | 71.64 | 7 | 31 |
| Milwaukee Co., WI | 86.01 | 75.80 | 84.66 | 74.79 | 82.76 | 73.77 | 5 | 37 |
| Indianapolis, IN | 82.38 | 72.31 | 81.28 | 71.33 | 80.19 | 70.37 | 6 | 36 |
| Louisville, KY | 82.54 | 71.70 | 80.54 | 70.34 | 78.27 | 68.59 | 6 | 14 |
| Nashville, TN | 83.54 | 71.51 | 82.20 | 70.22 | 80.79 | 69.07 | 6 | 20 |

Note: A tract enters when graded polygons (A to D) cover at least 50% of its area and takes the grade that covers the most of it. The comparison is descriptive: the grades followed the neighborhoods' housing, incomes and the race and national origin of residents, so differences today do not isolate the effect of the maps.

Note: The tracts are in 222 HOLC cities and city areas, matched by overlay. With a stricter rule that graded polygons cover at least 90% of a tract the life-expectancy gap between grades A and D is 3.6 years.

Note: Life expectancy is USALEEP for deaths in 2010 to 2015, area-weighted onto 2020 tracts, with institutional tracts left out. The adult income rank is the Opportunity Atlas rank of children born 1978 to 1983 to parents at the 25th income percentile, interpolated from 2010 to 2020 tracts by area. Tract values are weighted by ACS 2019-2023 population.

Source: Mapping Inequality HOLC maps (1935-1940) overlaid on 2020 tracts; USALEEP 2010-2015, Opportunity Atlas (1978-1983 birth cohorts), ACS 2019-2023 and ADI 2023; population-weighted, modeled estimates

Data: https://vishalsingh.org/zipcode/report/home-value-income-and-redlining/figures/holc_grade_outcomes.json

## All estimates

Every fact in facts.json (70), sorted by id.

| Fact id | Value | Unit | Estimand | Population | Period | Interval | n | Weights | Source | Note |
| --- | ---: | --- | --- | --- | --- | --- | ---: | --- | --- | --- |
| barbell_min_pop | 150,000 | residents | minimum county population for the comparison with the national shares | counties compared with the national shares | tax year 2022 |   |   | none (inclusion rule) | IRS Statistics of Income |   |
| beta_eitc | −0.49 | standard deviations of the county mobility rank per standard deviation of the EITC share | standardized coefficient on the EITC share with the other share and the Area Deprivation Index in the model; coefficient in a county-population-weighted regression of standardized variables, errors clustered on state | counties with an income-tax shape, an Area Deprivation Index and an Opportunity Atlas rank (Connecticut left out) | tax year 2022; mobility for children born 1978 to 1983 | 95% CI −0.75 to −0.23 | 3,116 counties | county population (ACS 2019-2023 tract population summed to the county) | IRS Statistics of Income |   |
| beta_top | −0.03 | standard deviations of the county mobility rank per standard deviation of the top-bracket share | standardized coefficient on the top-bracket share with the other share and the Area Deprivation Index in the model; coefficient in a county-population-weighted regression of standardized variables, errors clustered on state | counties with an income-tax shape, an Area Deprivation Index and an Opportunity Atlas rank (Connecticut left out) | tax year 2022; mobility for children born 1978 to 1983 | 95% CI −0.27 to 0.20 | 3,116 counties | county population (ACS 2019-2023 tract population summed to the county) | IRS Statistics of Income |   |
| bracket_low_line | $25,000 | dollars of adjusted gross income | upper limit of the lowest of the five income groups shown (returns with adjusted gross income under this amount; agi_stub 1 to 3) | individual income tax returns | tax year 2022 |   |   | none (definition) | IRS Statistics of Income |   |
| eitc_hi | 28.9% | percent of individual returns that claim the earned income tax credit | county-population-weighted mean of the county share of returns, interval from a bootstrap over states | returns in the most deprived tenth of counties by Area Deprivation Index (tenths hold equal numbers of counties) | tax year 2022 | 95% CI 26.4% to 31.2% | 312 counties | county population (ACS 2019-2023 tract population summed to the county) | IRS Statistics of Income |   |
| eitc_lo | 12.0% | percent of individual returns that claim the earned income tax credit | county-population-weighted mean of the county share of returns, interval from a bootstrap over states | returns in the least deprived tenth of counties by Area Deprivation Index (tenths hold equal numbers of counties) | tax year 2022 | 95% CI 10.4% to 13.2% | 312 counties | county population (ACS 2019-2023 tract population summed to the county) | IRS Statistics of Income |   |
| eitc_nat | 14.9% | percent of individual returns that claim the earned income tax credit | returns claiming the credit divided by all returns, all counties pooled | all individual income tax returns in the IRS county file | tax year 2022 |   | 158,707,990 returns | none (full count of returns) | IRS Statistics of Income |   |
| eitc_top_r | −0.65 | correlation coefficient | county-population-weighted correlation of the county EITC share with the county top-bracket share, bootstrap interval over states | counties in the gradient | tax year 2022 | 95% CI −0.70 to −0.60 | 3,129 counties | county population (ACS 2019-2023 tract population summed to the county) | IRS Statistics of Income |   |
| holc_adi_b | 41 | national percentile (100 = most deprived) | Area Deprivation Index 2023 national rank, population-weighted mean over tracts of the tract value, grade assigned by the largest area overlap, bootstrap interval over counties | residents of tracts at least 50 percent covered by graded HOLC polygons that are graded B and have an Area Deprivation Index value | maps drawn 1935 to 1940; outcomes: life expectancy for deaths in 2010 to 2015, income and homeownership 2019 to 2023, mobility for children born 1978 to 1983 | 95% CI 33 to 49 | 2,241 tracts | ACS 2019-2023 tract population | Mapping Inequality (HOLC maps) |   |
| holc_adi_c | 42 | national percentile (100 = most deprived) | Area Deprivation Index 2023 national rank, population-weighted mean over tracts of the tract value, grade assigned by the largest area overlap, bootstrap interval over counties | residents of tracts at least 50 percent covered by graded HOLC polygons that are graded C and have an Area Deprivation Index value | maps drawn 1935 to 1940; outcomes: life expectancy for deaths in 2010 to 2015, income and homeownership 2019 to 2023, mobility for children born 1978 to 1983 | 95% CI 33 to 53 | 5,040 tracts | ACS 2019-2023 tract population | Mapping Inequality (HOLC maps) |   |
| holc_adi_d | 41 | national percentile (100 = most deprived) | Area Deprivation Index 2023 national rank, population-weighted mean over tracts of the tract value, grade assigned by the largest area overlap, bootstrap interval over counties | residents of tracts at least 50 percent covered by graded HOLC polygons that are graded D and have an Area Deprivation Index value | maps drawn 1935 to 1940; outcomes: life expectancy for deaths in 2010 to 2015, income and homeownership 2019 to 2023, mobility for children born 1978 to 1983 | 95% CI 32 to 53 | 2,953 tracts | ACS 2019-2023 tract population | Mapping Inequality (HOLC maps) |   |
| holc_city_median | 5.7 | years of life expectancy at birth, grade A minus grade D | median over cities of the city's gap between the population-weighted mean life expectancy of its grade A tracts and of its grade D tracts | HOLC cities in the city-by-city comparison | maps drawn 1935 to 1940; outcomes: life expectancy for deaths in 2010 to 2015, income and homeownership 2019 to 2023, mobility for children born 1978 to 1983 |   | 27 cities | none (each city counts once) | Mapping Inequality (HOLC maps) |   |
| holc_city_n | 27 | HOLC cities and city areas | count of cities in the city-by-city comparison | HOLC cities with enough grade A and grade D tracts | maps drawn 1935 to 1940; outcomes: life expectancy for deaths in 2010 to 2015, income and homeownership 2019 to 2023, mobility for children born 1978 to 1983 |   |   | none (full count) | Mapping Inequality (HOLC maps) |   |
| holc_city_p5 | 3.4 | years of life expectancy at birth, grade A minus grade D | 5th percentile over cities of the city's gap between the population-weighted mean life expectancy of its grade A tracts and of its grade D tracts | HOLC cities in the city-by-city comparison | maps drawn 1935 to 1940; outcomes: life expectancy for deaths in 2010 to 2015, income and homeownership 2019 to 2023, mobility for children born 1978 to 1983 |   | 27 cities | none (each city counts once) | Mapping Inequality (HOLC maps) |   |
| holc_city_p95 | 10.1 | years of life expectancy at birth, grade A minus grade D | 95th percentile over cities of the city's gap between the population-weighted mean life expectancy of its grade A tracts and of its grade D tracts | HOLC cities in the city-by-city comparison | maps drawn 1935 to 1940; outcomes: life expectancy for deaths in 2010 to 2015, income and homeownership 2019 to 2023, mobility for children born 1978 to 1983 |   | 27 cities | none (each city counts once) | Mapping Inequality (HOLC maps) |   |
| holc_cover_min | 50% | percent of a tract's area | minimum share of a tract covered by graded polygons for the tract to enter | tracts in the comparison | maps drawn 1935 to 1940; outcomes: life expectancy for deaths in 2010 to 2015, income and homeownership 2019 to 2023, mobility for children born 1978 to 1983 |   |   | none (inclusion rule) | Mapping Inequality (HOLC maps) |   |
| holc_cover_strict | 90% | percent of a tract's area | stricter minimum share of a tract covered by graded polygons | tracts in the stricter comparison | maps drawn 1935 to 1940; outcomes: life expectancy for deaths in 2010 to 2015, income and homeownership 2019 to 2023, mobility for children born 1978 to 1983 |   |   | none (inclusion rule) | Mapping Inequality (HOLC maps) |   |
| holc_inc_a | $127 thousand | thousands of dollars | ACS 2019-2023 tract median household income, population-weighted mean over tracts of the tract value, grade assigned by the largest area overlap, bootstrap interval over counties; a mean of tract medians, not a median of households | residents of tracts at least 50 percent covered by graded HOLC polygons that are graded A and have a median household income | maps drawn 1935 to 1940; outcomes: life expectancy for deaths in 2010 to 2015, income and homeownership 2019 to 2023, mobility for children born 1978 to 1983 | 95% CI $115 thousand to $138 thousand | 669 tracts | ACS 2019-2023 tract population | Mapping Inequality (HOLC maps) |   |
| holc_inc_d | $71 thousand | thousands of dollars | ACS 2019-2023 tract median household income, population-weighted mean over tracts of the tract value, grade assigned by the largest area overlap, bootstrap interval over counties; a mean of tract medians, not a median of households | residents of tracts at least 50 percent covered by graded HOLC polygons that are graded D and have a median household income | maps drawn 1935 to 1940; outcomes: life expectancy for deaths in 2010 to 2015, income and homeownership 2019 to 2023, mobility for children born 1978 to 1983 | 95% CI $63 thousand to $78 thousand | 2,920 tracts | ACS 2019-2023 tract population | Mapping Inequality (HOLC maps) |   |
| holc_kfr_a | 46.3 | percentile points of national household income rank | mean adult household income rank (Opportunity Atlas, children born 1978 to 1983 to parents at the 25th income percentile), population-weighted mean over tracts of the tract value, grade assigned by the largest area overlap, bootstrap interval over counties | residents of tracts at least 50 percent covered by graded HOLC polygons that are graded A and have an adult income rank of children from low-income families | maps drawn 1935 to 1940; outcomes: life expectancy for deaths in 2010 to 2015, income and homeownership 2019 to 2023, mobility for children born 1978 to 1983 | 95% CI 45.3 to 47.1 | 672 tracts | ACS 2019-2023 tract population | Mapping Inequality (HOLC maps) |   |
| holc_kfr_d | 37.9 | percentile points of national household income rank | mean adult household income rank (Opportunity Atlas, children born 1978 to 1983 to parents at the 25th income percentile), population-weighted mean over tracts of the tract value, grade assigned by the largest area overlap, bootstrap interval over counties | residents of tracts at least 50 percent covered by graded HOLC polygons that are graded D and have an adult income rank of children from low-income families | maps drawn 1935 to 1940; outcomes: life expectancy for deaths in 2010 to 2015, income and homeownership 2019 to 2023, mobility for children born 1978 to 1983 | 95% CI 36.3 to 39.2 | 2,974 tracts | ACS 2019-2023 tract population | Mapping Inequality (HOLC maps) |   |
| holc_le_a | 80.9 | years of life expectancy at birth | population-weighted mean over tracts of the tract value, grade assigned by the largest area overlap, bootstrap interval over counties | residents of tracts at least 50 percent covered by graded HOLC polygons that are graded A | maps drawn 1935 to 1940; outcomes: life expectancy for deaths in 2010 to 2015, income and homeownership 2019 to 2023, mobility for children born 1978 to 1983 | 95% CI 80.1 to 81.6 | 629 tracts | ACS 2019-2023 tract population | Mapping Inequality (HOLC maps) |   |
| holc_le_d | 76.8 | years of life expectancy at birth | population-weighted mean over tracts of the tract value, grade assigned by the largest area overlap, bootstrap interval over counties | residents of tracts at least 50 percent covered by graded HOLC polygons that are graded D | maps drawn 1935 to 1940; outcomes: life expectancy for deaths in 2010 to 2015, income and homeownership 2019 to 2023, mobility for children born 1978 to 1983 | 95% CI 75.5 to 77.8 | 2,671 tracts | ACS 2019-2023 tract population | Mapping Inequality (HOLC maps) |   |
| holc_le_gap | 4.1 | years of life expectancy at birth, grade A minus grade D | difference between the population-weighted mean over grade A tracts and over grade D tracts, bootstrap interval over counties | residents of tracts at least 50 percent covered by graded HOLC polygons | maps drawn 1935 to 1940; outcomes: life expectancy for deaths in 2010 to 2015, income and homeownership 2019 to 2023, mobility for children born 1978 to 1983 | 95% CI 3.3 to 5.1 | 3,300 tracts | ACS 2019-2023 tract population | Mapping Inequality (HOLC maps) |   |
| holc_le_gap_90 | 3.6 | years of life expectancy at birth, grade A minus grade D | difference between the population-weighted mean over grade A tracts and over grade D tracts under the stricter coverage rule, bootstrap interval over counties | residents of tracts at least 90 percent covered by graded HOLC polygons | maps drawn 1935 to 1940; outcomes: life expectancy for deaths in 2010 to 2015, income and homeownership 2019 to 2023, mobility for children born 1978 to 1983 | 95% CI 2.6 to 4.9 | 1,864 tracts | ACS 2019-2023 tract population | Mapping Inequality (HOLC maps) |   |
| holc_le_gap_fe | 5.8 | years of life expectancy at birth, grade A minus grade D, within the same city | coefficient of grade D relative to grade A (sign reversed) in a population-weighted regression of tract life expectancy on grade with a separate baseline for each city, errors clustered on county | residents of tracts at least 50 percent covered by graded HOLC polygons and a life expectancy, in 222 HOLC cities | maps drawn 1935 to 1940; outcomes: life expectancy for deaths in 2010 to 2015, income and homeownership 2019 to 2023, mobility for children born 1978 to 1983 | 95% CI 5.3 to 6.4 | 10,143 tracts | ACS 2019-2023 tract population | Mapping Inequality (HOLC maps) |   |
| holc_min_a | 5 | tracts | minimum number of grade A tracts with a life expectancy for a city to enter | HOLC cities in the city-by-city comparison | maps drawn 1935 to 1940; outcomes: life expectancy for deaths in 2010 to 2015, income and homeownership 2019 to 2023, mobility for children born 1978 to 1983 |   |   | none (inclusion rule) | Mapping Inequality (HOLC maps) |   |
| holc_min_d | 10 | tracts | minimum number of grade D tracts with a life expectancy for a city to enter | HOLC cities in the city-by-city comparison | maps drawn 1935 to 1940; outcomes: life expectancy for deaths in 2010 to 2015, income and homeownership 2019 to 2023, mobility for children born 1978 to 1983 |   |   | none (inclusion rule) | Mapping Inequality (HOLC maps) |   |
| holc_n_cities | 222 | HOLC cities and city areas | count of distinct (city, state) names of the HOLC polygon covering the largest part of a tract | tracts in the comparison | maps drawn 1935 to 1940; outcomes: life expectancy for deaths in 2010 to 2015, income and homeownership 2019 to 2023, mobility for children born 1978 to 1983 |   |   | none (full count) | Mapping Inequality (HOLC maps) |   |
| holc_n_tracts | 10,952 | tracts | count of tracts in the comparison | 2020 census tracts with a population and at least half of their area under graded HOLC polygons | maps drawn 1935 to 1940; outcomes: life expectancy for deaths in 2010 to 2015, income and homeownership 2019 to 2023, mobility for children born 1978 to 1983 |   |   | none (full count) | Mapping Inequality (HOLC maps) |   |
| holc_own_a | 67.8% | percent of households | ACS 2019-2023 homeownership rate, population-weighted mean over tracts of the tract value, grade assigned by the largest area overlap, bootstrap interval over counties | residents of tracts at least 50 percent covered by graded HOLC polygons that are graded A and have a homeownership rate | maps drawn 1935 to 1940; outcomes: life expectancy for deaths in 2010 to 2015, income and homeownership 2019 to 2023, mobility for children born 1978 to 1983 | 95% CI 65.0% to 70.6% | 672 tracts | ACS 2019-2023 tract population | Mapping Inequality (HOLC maps) |   |
| holc_own_d | 35.3% | percent of households | ACS 2019-2023 homeownership rate, population-weighted mean over tracts of the tract value, grade assigned by the largest area overlap, bootstrap interval over counties | residents of tracts at least 50 percent covered by graded HOLC polygons that are graded D and have a homeownership rate | maps drawn 1935 to 1940; outcomes: life expectancy for deaths in 2010 to 2015, income and homeownership 2019 to 2023, mobility for children born 1978 to 1983 | 95% CI 31.7% to 39.7% | 2,973 tracts | ACS 2019-2023 tract population | Mapping Inequality (HOLC maps) |   |
| holc_pair_share | 75% | percent of random pairs | probability that the grade A tract of a random pair has the higher life expectancy | pairs of one grade A tract and one grade D tract with a life expectancy, each drawn in proportion to its population | maps drawn 1935 to 1940; outcomes: life expectancy for deaths in 2010 to 2015, income and homeownership 2019 to 2023, mobility for children born 1978 to 1983 |   | 3,300 tracts | ACS 2019-2023 tract population | Mapping Inequality (HOLC maps) |   |
| n_barbell_pool | 447 | counties | count of counties with at least the minimum population | counties with returns and an Area Deprivation Index | tax year 2022 |   |   | none (full count) | IRS Statistics of Income |   |
| n_both_tails | 15 | counties | count of counties whose EITC share and top-bracket share both exceed the national shares | counties with at least 150,000 residents (ACS 2019-2023 tract population) | tax year 2022 |   | 447 counties | none (full count) | IRS Statistics of Income |   |
| n_counties | 3,129 | counties and Connecticut planning regions | count of counties in the gradient | counties (Connecticut's planning regions stand in for its counties) in the IRS Statistics of Income county file with returns and an Area Deprivation Index | tax year 2022 |   |   | none (full count) | IRS Statistics of Income |   |
| pop_d1 | 130.1 million | people | sum of the ACS 2019-2023 tract population over the counties of the tenth | residents of counties in the least deprived tenth of counties by Area Deprivation Index | ACS 2019-2023 |   | 312 counties | none (full count) | American Community Survey |   |
| pop_d10 | 3.8 million | people | sum of the ACS 2019-2023 tract population over the counties of the tenth | residents of counties in the most deprived tenth of counties by Area Deprivation Index | ACS 2019-2023 |   | 312 counties | none (full count) | American Community Survey |   |
| price_cheap_usd | $33 thousand | thousands of dollars of median home value per year of life expectancy | median of the metro's tract median home values divided by the slope of tract life expectancy on the natural log of home value across the metro's tracts (each tract once), interval from a bootstrap over tracts within the metro | tracts of the metro in the canonical tract sample (each tract takes the core-based statistical area of its county), Cincinnati metro | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); home values ACS 2019-2023 (tract medians) | 95% CI $30 thousand to $38 thousand | 514 tracts | none (each tract counts once) | USALEEP life expectancy |   |
| price_dear_usd | $658 thousand | thousands of dollars of median home value per year of life expectancy | median of the metro's tract median home values divided by the slope of tract life expectancy on the natural log of home value across the metro's tracts (each tract once), interval from a bootstrap over tracts within the metro | tracts of the metro in the canonical tract sample (each tract takes the core-based statistical area of its county), San Jose metro | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); home values ACS 2019-2023 (tract medians) | 95% CI $482 thousand to $947 thousand | 400 tracts | none (each tract counts once) | USALEEP life expectancy |   |
| price_doubling_weighted | 2.2 | years of life expectancy at birth per doubling of median home value | slope of tract life expectancy on log2 tract median home value, weighted by tract population, county-cluster bootstrap interval | non-institutional 2020 census tracts with a life expectancy estimate and an ACS median home value above $10,000 (50 states and DC) | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); home values ACS 2019-2023 (tract medians) | 95% CI 2.1 to 2.3 | 76,702 tracts | ACS 2019-2023 tract population | USALEEP life expectancy | Tract life expectancy is paired with the tract's own ACS 2019-2023 median home value (top-coded at $2,000,001). Connecticut's nine planning regions stand in for its counties. Interval: bootstrap over counties. |
| price_doubling_yrs | 2.2 | years of life expectancy at birth per doubling of median home value | slope of tract life expectancy on log2 tract median home value, each tract counted once, county-cluster bootstrap interval | non-institutional 2020 census tracts with a life expectancy estimate and an ACS median home value above $10,000 (50 states and DC) | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); home values ACS 2019-2023 (tract medians) | 95% CI 2.1 to 2.3 | 76,702 tracts | none (each tract counts once) | USALEEP life expectancy | Tract life expectancy is paired with the tract's own ACS 2019-2023 median home value (top-coded at $2,000,001). Connecticut's nine planning regions stand in for its counties. Interval: bootstrap over counties. |
| price_le_hi | 82.8 | years of life expectancy at birth | population-weighted mean of tract life expectancy in a twentieth of tracts ranked by median home value (equal numbers of tracts), county-cluster bootstrap interval | residents of tracts in the highest twentieth of tracts by median home value | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); home values ACS 2019-2023 (tract medians) | 95% CI 82.5 to 83.2 | 3,836 tracts | ACS 2019-2023 tract population | USALEEP life expectancy |   |
| price_le_lo | 73.5 | years of life expectancy at birth | population-weighted mean of tract life expectancy in a twentieth of tracts ranked by median home value (equal numbers of tracts), county-cluster bootstrap interval | residents of tracts in the lowest twentieth of tracts by median home value | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); home values ACS 2019-2023 (tract medians) | 95% CI 73.1 to 73.9 | 3,835 tracts | ACS 2019-2023 tract population | USALEEP life expectancy |   |
| price_le_span | 9.3 | years of life expectancy at birth, highest minus lowest twentieth | difference between the population-weighted mean life expectancy of the highest and the lowest twentieth of tracts by median home value, county-cluster bootstrap interval | residents of tracts in the highest and the lowest twentieth of tracts by median home value | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); home values ACS 2019-2023 (tract medians) | 95% CI 8.8 to 9.8 | 7,671 tracts | ACS 2019-2023 tract population | USALEEP life expectancy |   |
| price_maxr_r | +0.78 | correlation coefficient | correlation of tract life expectancy with the natural log of median home value within the metro, bootstrap interval over tracts | tracts of the metro in the canonical tract sample (each tract takes the core-based statistical area of its county), Detroit metro | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); home values ACS 2019-2023 (tract medians) | 95% CI 0.75 to 0.80 | 1,214 tracts | none (each tract counts once) | USALEEP life expectancy |   |
| price_maxr_usd | $44 thousand | thousands of dollars of median home value per year of life expectancy | median of the metro's tract median home values divided by the slope of tract life expectancy on the natural log of home value across the metro's tracts (each tract once), interval from a bootstrap over tracts within the metro | tracts of the metro in the canonical tract sample (each tract takes the core-based statistical area of its county), Detroit metro | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); home values ACS 2019-2023 (tract medians) | 95% CI $42 thousand to $47 thousand | 1,214 tracts | none (each tract counts once) | USALEEP life expectancy |   |
| price_metro_level_r | +0.89 | correlation coefficient | correlation across metros of dollars per year of life with the metro's median tract home value, bootstrap interval over metros | metros in the price-of-a-year comparison | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); home values ACS 2019-2023 (tract medians) | 95% CI 0.82 to 0.96 | 39 metros | none (each metro counts once) | USALEEP life expectancy |   |
| price_metro_min_pop | 1.5 million | residents | minimum population of the metro's tracts in the sample for a metro to enter | metros in the price-of-a-year comparison | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); home values ACS 2019-2023 (tract medians) |   |   | none (inclusion rule) | USALEEP life expectancy |   |
| price_metro_min_tracts | 250 | tracts | minimum number of tracts in the sample for a metro to enter | metros in the price-of-a-year comparison | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); home values ACS 2019-2023 (tract medians) |   |   | none (inclusion rule) | USALEEP life expectancy |   |
| price_n_metros | 39 | metros | count of metros with the minimum number of tracts and residents and a positive slope | core-based statistical areas with a tract in the canonical tract sample (929 areas) | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); home values ACS 2019-2023 (tract medians) |   |   | none (full count) | USALEEP life expectancy |   |
| price_n_outside | 21 | tracts | count of tracts outside the plotted rectangle (life expectancy 60 to 92 years, home value 10 thousand to 2.1 million dollars) | non-institutional 2020 census tracts with a life expectancy estimate and an ACS median home value above $10,000 (50 states and DC) | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); home values ACS 2019-2023 (tract medians) |   |   | none (full count) | USALEEP life expectancy |   |
| price_n_sample | 5,000 | tracts | size of a random sample drawn with a fixed seed for the scatter plot | non-institutional 2020 census tracts with a life expectancy estimate and an ACS median home value above $10,000 (50 states and DC) | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); home values ACS 2019-2023 (tract medians) |   | 76,702 tracts | none (sample size) | USALEEP life expectancy |   |
| price_n_six_figure | 16 | metros | count of metros where the point estimate of dollars per year of life exceeds the threshold (the interval of some of them includes the threshold) | metros in the price-of-a-year comparison | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); home values ACS 2019-2023 (tract medians) |   |   | none (full count) | USALEEP life expectancy |   |
| price_n_tracts | 76,702 | tracts | count of tracts in the regression | non-institutional 2020 census tracts with a life expectancy estimate and an ACS median home value above $10,000 (50 states and DC) | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); home values ACS 2019-2023 (tract medians) |   |   | none (full count) | USALEEP life expectancy |   |
| price_r | +0.59 | correlation coefficient | correlation of tract life expectancy with log2 tract median home value, county-cluster bootstrap interval | non-institutional 2020 census tracts with a life expectancy estimate and an ACS median home value above $10,000 (50 states and DC) | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); home values ACS 2019-2023 (tract medians) | 95% CI 0.56 to 0.61 | 76,702 tracts | none (each tract counts once) | USALEEP life expectancy | Tract life expectancy is paired with the tract's own ACS 2019-2023 median home value (top-coded at $2,000,001). Connecticut's nine planning regions stand in for its counties. Interval: bootstrap over counties. |
| price_r_linear | +0.50 | correlation coefficient | correlation of tract life expectancy with the tract median home value in dollars (not logged), each tract counted once, county-cluster bootstrap interval | non-institutional 2020 census tracts with a life expectancy estimate and an ACS median home value above $10,000 (50 states and DC) | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); home values ACS 2019-2023 (tract medians) | 95% CI 0.48 to 0.52 | 76,702 tracts | none (each tract counts once) | USALEEP life expectancy |   |
| price_ratio | 19.6 | ratio of dollars per year of life | dollars per year of life at the highest point estimate among the metros divided by that at the lowest point estimate; the interval comes from paired bootstrap replicates that hold those metros fixed, so it understates how far apart the extremes of 39 noisy estimates are | the metros with the highest and the lowest point estimates of dollars per year of life among the 39 metros that pass the size rule | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); home values ACS 2019-2023 (tract medians) | 95% CI 14.1 to 28.7 |   | none (each tract counts once) | USALEEP life expectancy |   |
| price_threshold | $100 thousand | thousands of dollars of median home value per year of life | threshold that defines a six-figure year of life | metros in the price-of-a-year comparison | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); home values ACS 2019-2023 (tract medians) |   |   | none (definition) | USALEEP life expectancy |   |
| r_eitc_adi | +0.57 | correlation coefficient | county-population-weighted correlation of the county EITC share with the Area Deprivation Index national rank, bootstrap interval over states | counties in the gradient | tax year 2022; Area Deprivation Index 2023 | 95% CI 0.43 to 0.72 | 3,129 counties | county population (ACS 2019-2023 tract population summed to the county) | IRS Statistics of Income |   |
| r_top_adi | −0.74 | correlation coefficient | county-population-weighted correlation of the county top-bracket share with the Area Deprivation Index national rank, bootstrap interval over states | counties in the gradient | tax year 2022; Area Deprivation Index 2023 | 95% CI −0.82 to −0.69 | 3,129 counties | county population (ACS 2019-2023 tract population summed to the county) | IRS Statistics of Income |   |
| shape_top_q1 | 10.0% | percent of returns with adjusted gross income $200,000 or more | share of all returns in the counties of the group (pooled, so weighted by returns) | returns in the least deprived fifth of counties (fifths of counties by Area Deprivation Index hold equal numbers of counties) | tax year 2022 |   | 625 counties | none (full count of returns) | IRS Statistics of Income |   |
| shape_top_q5 | 2.3% | percent of returns with adjusted gross income $200,000 or more | share of all returns in the counties of the group (pooled, so weighted by returns) | returns in the most deprived fifth of counties (fifths of counties by Area Deprivation Index hold equal numbers of counties) | tax year 2022 |   | 624 counties | none (full count of returns) | IRS Statistics of Income |   |
| shape_under25_q1 | 26.7% | percent of returns with adjusted gross income under $25,000 | share of all returns in the counties of the group (pooled, so weighted by returns) | returns in the least deprived fifth of counties (fifths of counties by Area Deprivation Index hold equal numbers of counties) | tax year 2022 |   | 625 counties | none (full count of returns) | IRS Statistics of Income |   |
| shape_under25_q5 | 37.6% | percent of returns with adjusted gross income under $25,000 | share of all returns in the counties of the group (pooled, so weighted by returns) | returns in the most deprived fifth of counties (fifths of counties by Area Deprivation Index hold equal numbers of counties) | tax year 2022 |   | 624 counties | none (full count of returns) | IRS Statistics of Income |   |
| top_bracket_line | $200,000 | dollars of adjusted gross income | lower limit of the top adjusted-gross-income bracket (agi_stub 8) of the IRS county file | individual income tax returns | tax year 2022 |   |   | none (definition) | IRS Statistics of Income |   |
| top_hi | 1.9% | percent of individual returns that report adjusted gross income of $200,000 or more | county-population-weighted mean of the county share of returns, interval from a bootstrap over states | returns in the most deprived tenth of counties by Area Deprivation Index (tenths hold equal numbers of counties) | tax year 2022 | 95% CI 1.7% to 2.1% | 312 counties | county population (ACS 2019-2023 tract population summed to the county) | IRS Statistics of Income |   |
| top_lo | 11.1% | percent of individual returns that report adjusted gross income of $200,000 or more | county-population-weighted mean of the county share of returns, interval from a bootstrap over states | returns in the least deprived tenth of counties by Area Deprivation Index (tenths hold equal numbers of counties) | tax year 2022 | 95% CI 10.3% to 11.9% | 312 counties | county population (ACS 2019-2023 tract population summed to the county) | IRS Statistics of Income |   |
| top_nat | 7.8% | percent of individual returns that report adjusted gross income of $200,000 or more | returns in the top bracket divided by all returns, all counties pooled | all individual income tax returns in the IRS county file | tax year 2022 |   | 158,707,990 returns | none (full count of returns) | IRS Statistics of Income |   |
| wls_top_beta | 0.41 | standard deviations of the county mobility rank per standard deviation of the top-bracket share | standardized coefficient on the top-bracket share alone; coefficient in a county-population-weighted regression of standardized variables, errors clustered on state | counties with an income-tax shape, an Area Deprivation Index and an Opportunity Atlas rank (Connecticut left out) | tax year 2022; mobility for children born 1978 to 1983 | 95% CI 0.33 to 0.50 | 3,116 counties | county population (ACS 2019-2023 tract population summed to the county) | IRS Statistics of Income |   |

## Definitions

- USALEEP: The U.S. Small-area Life Expectancy Estimates Project, a National Center for Health Statistics release of life expectancy at birth for most census tracts. It covers deaths from 2010 to 2015 only and has not been updated.
- EITC: The Earned Income Tax Credit, a refundable federal tax credit for low and moderate income workers. It grows with earnings up to a limit and then phases out.
- HOLC grade: A letter from A to D that the Home Owners' Loan Corporation gave neighborhoods on its 1930s maps, with A labeled best and D hazardous. The maps are known as redlining maps.
- Area Deprivation Index (ADI): A score that ranks neighborhoods by income, education, employment and housing conditions drawn from the American Community Survey. In this book a higher national percentile means a more deprived neighborhood.
- Income rank (kfr): The position that adult children reach in the national distribution of household income, from the lowest rank to the highest, averaged over children who grew up in a place with parents at a stated income level. The Opportunity Atlas labels this measure kfr.
- Fixed effects: A regression adjustment that gives each group, such as each county, its own baseline. The estimate then rests on differences within groups.
- Interquartile range: The distance between the first quartile and the third quartile of a set of values. It describes the spread of the middle of a distribution and is not moved by extreme values.
- Ecological inference: Drawing conclusions about individuals from data about places or groups. A pattern across counties need not hold for the people inside them.

## Methods

This chapter compares places (census tracts and counties), not households, and each comparison has its own sample, weights and interval method, set out below. Life expectancy is for deaths in 2010 to 2015, while home values and the tract measures are from the ACS 2019-2023, so the outcome and the predictor come from different periods. The metro exchange rate covers only metros with many tracts and residents in the sample, the HOLC comparison covers only tracts in graded cities, and the Area Deprivation Index that ranks counties is built partly from income, so some of the income gradient is built in.

### Tracts, life expectancy and home value

Life expectancy is the USALEEP estimate for deaths in 2010 to 2015 on 2020 tracts (area-weighted from 2010 tracts), with institutional tracts, where at least 25 percent of residents live in group quarters, left out.

Each tract is paired with its own ACS 2019-2023 median home value, which the Census Bureau top-codes at $2,000,001; tracts with a value at or below $10,000 are left out, and tracts are weighted by their ACS 2019-2023 population.

Until 2026-10-28 this comparison used ZIP Code Tabulation Areas, each given the population-weighted mean of the tracts whose largest-overlap ZCTA it was, and a ZCTA median home value from a purchased ZIP file whose vintage is not documented. That value's top code of $1,000,001 and the file's population total point to about 2010 to 2014, and the chapter had labeled it ACS 2019-2023.

Twentieths of tracts hold equal numbers of tracts.

Explains Table 1.

### Years per doubling and the metro exchange rate

Years per doubling of home value is the slope of tract life expectancy on the base-2 log of home value, with each tract counted once and weighted by population, from the book's shared routine, with intervals from a bootstrap over counties.

Tracts carry no metro code, so each tract takes the core-based statistical area of its county from the Census 2020 county file; core-based statistical areas are built from whole counties, so this assigns codes and moves no values.

The metro exchange rate is the median of the tracts' median home values divided by the within-metro slope of life expectancy on the natural log of home value, for metros with at least 250 tracts and 1.5 million residents in the sample, with intervals from a bootstrap over tracts within the metro.

Explains Table 1.

### Income shapes and the county ranking

The IRS county file for tax year 2022 gives returns in eight adjusted-gross-income brackets; the EITC share is returns claiming the credit divided by all returns, and the top-bracket share is returns with adjusted gross income of $200,000 or more divided by all returns.

Counties are ranked on the Area Deprivation Index (tract ranks averaged with ACS 2019-2023 population weights); tenths and fifths hold equal numbers of counties. Means and correlations are weighted by county population, with intervals from a bootstrap over states.

Mobility regressions use weighted z-scores of the Opportunity Atlas county rank, with errors clustered on state; Connecticut drops out because its planning-region codes do not match the Atlas's counties.

Explains Table 2.

### HOLC grades and tract outcomes

Graded polygons (A to D) are intersected with 2020 tracts in an equal-area projection; a tract with at least 50 percent of its area graded takes the grade covering most of it, and the city of the polygon covering most of it.

Tract outcomes are weighted by ACS 2019-2023 population, with intervals from a bootstrap over counties; the within-city gap uses a weighted regression with a separate baseline for each city.

Explains Table 3.

### Bootstrap settings and city comparisons

The city comparison resamples tracts within a city and grade and treats them as independent.

Bootstrap intervals use 2,000 replicates and seed 12345.

## What these data cannot show

Every comparison in this chapter describes places. A county or a tract is not a person, and a pattern across places need not hold for the people inside them (ecological inference). The residents of a tract differ from movers in income, health and age, and the data cannot separate selection from what an area provides.

The price of a year joins life expectancy for deaths in 2010 to 2015 to home values for 2019 to 2023, so the two describe different years. The life expectancy estimates are modeled for small areas and area-weighted from 2010 to 2020 tracts. A tract median home value rests on few homes and the tract file carries no margins of error, so part of the spread across tracts is sampling error. The dollar figures divide a price level by a slope that is itself estimated with error.

The IRS file covers people who file a return, so the shares are shares of returns and leave out people who do not file. The EITC share depends on eligibility and on whether eligible households claim the credit, and the top-bracket share is a share of returns, not of people. Connecticut is absent from the mobility regressions because its planning-region codes do not match the Atlas's counties.

The grades followed the neighborhoods' housing, incomes and populations in the 1930s, so the staircase today combines what the neighborhoods were with what happened after. The boundary study in the box above finds that the maps themselves mattered for housing outcomes and segregation next to borders, and it does not show that the grades explain today's differences in life expectancy. Tracts are matched to grades by area, and the Atlas ranks are interpolated by area from 2010 tracts. The maps cover the cities that HOLC surveyed, so a city that is absent was not surveyed, which does not mean it had no redlining.

## References

1. Rosen, Sherwin (1974). Hedonic Prices and Implicit Markets: Product Differentiation in Pure Competition. Journal of Political Economy 82(1): 34-55. https://doi.org/10.1086/260169
2. Nelson, Robert K.; Winling, LaDale; Marciano, Richard; Connolly, Nathan; et al. (2023). Mapping Inequality: Redlining in New Deal America. American Panorama, ed. Robert K. Nelson and Edward L. Ayers, Digital Scholarship Lab, University of Richmond. https://dsl.richmond.edu/panorama/redlining/
3. Aaronson, Daniel; Hartley, Daniel; Mazumder, Bhashkar (2021). The Effects of the 1930s HOLC "Redlining" Maps. American Economic Journal: Economic Policy 13(4): 355-392. https://doi.org/10.1257/pol.20190414

## Provenance

- Pipeline version: 0.1.0
- Build script: book2/chapters/home-value-income-and-redlining/build.py (sha256 b185acf6f8ec0e994ed1b6d0bf6944d7455fe68c4b6742ea71d4ac0e4bf719e3)
- Built on: 2026-10-28
- Site commit: c2b1ff3309cb4a10833f884da91ac0057a66ad0a

Inputs:

| File | Size (bytes) | SHA-256, first 1 MB | SHA-256, last 1 MB |
| --- | ---: | --- | --- |
| book2/chapters/home-value-income-and-redlining/holc.py | 10,359 | db4be530f82d95b32080517d92937e7574e7866571ece2b6deef7b35c3d3e657 | db4be530f82d95b32080517d92937e7574e7866571ece2b6deef7b35c3d3e657 |
| book2/chapters/home-value-income-and-redlining/income.py | 7,746 | e76c7fb52f42071e405fef69eb0989088e59f921058cd36873857e74f931b425 | e76c7fb52f42071e405fef69eb0989088e59f921058cd36873857e74f931b425 |
| book2/chapters/home-value-income-and-redlining/price.py | 10,564 | 6b34ac5a52ab9b1f9a8dc4b9691e9193a3f90bb43d27fdb1d3c993e4730ccf51 | 6b34ac5a52ab9b1f9a8dc4b9691e9193a3f90bb43d27fdb1d3c993e4730ccf51 |
| data/analytic/tract_atlas.parquet | 39,659,925 | 29d5e4d943552d54aeb6ba92aadc939d9bd15bdea553c78357dbd516757be627 | 1e10d1ecf722340f23ee2212d4999a90250f85b65fa89b39e32db06d9472e22e |
| data/geo/counties_2020.parquet | 123,034,155 | 2fce837cdc2db1f2c229f2ea453c6793338a462770307e2c64e8d1fd713ec663 | dae928bf1042f3d174d2d64d6e3ec923fa3daa40aee1bb4feb104aac2fe3c04a |
| data/geo/dim_geo.parquet | 1,180,921 | 29d5e4d943552d54aeb6ba92aadc939d9bd15bdea553c78357dbd516757be627 | e58c8c488df54ab3187f5461f179322afdc49c9bd2315404003fe1272eea32c3 |
| data/geo/holc_areas_2023.parquet | 4,084,019 | 6d293647278dc82053c0ab89a192836f0763e73aff77785f3f06ad4214102600 | 37dc09e2ca32f1f9a001d07433b85faf4bd0a31ea584c16b62654b6556084be4 |
| data/geo/tracts_2020.parquet | 619,858,391 | f4f05c8c7a63d1785ec2aad283de3f6fa839b6c437f13a1b96cea31d3344bf0f | 7b3811247c6b3e322954f24a4a2384af52aef6ed8688f8dbdb0407d21ef80a59 |
| data/raw/acs/deprivation_components_2019_2023/tract.parquet | 13,197,793 | 5545cde2cbe96a8b56e468d65388248a241aa6bd300d3c9e325d08d3f86b11da | 5c8ec454d066988b0f1a01aaa6ed34ebfc05d162e01cf9c6bb01f9198cbaff12 |
| data/staged/adi/adi_atlas_tract2020_2023.parquet | 798,078 | 5e409bc212ff871def780fa27b4cf8b616d9da420b2b9f6455517eddcd8c1a95 | 5e409bc212ff871def780fa27b4cf8b616d9da420b2b9f6455517eddcd8c1a95 |
| data/staged/health/chr_county_2024.parquet | 8,776,933 | 619795f7ecfec8aff10694b649eaee5e421dbeffbe4be47af4161b51ceefa5cd | c69357f1bcf630407bb9716cb445618838999558c019aaaf77d4681b6bf45c56 |
| data/staged/income/irs_soi_county_2022.parquet | 7,590,532 | 827be995a4744d9198cf6da5afa6c099fe5ae7da5e7e6fc6b2eaa475b17e65f1 | 2e341a6e8456554ed604a4d357031fc08815a16f5bb7b740db85b583aa56b5f1 |
| data/staged/life_expectancy/usaleep_tract2020.parquet | 1,316,169 | 9c9c7f103791b75eddba74e68a89804efdf2486b6b66bb4bc02a46ba5bf8bc23 | 27317107be2465d9fd5e83271dea0cd6e915ee419e2802bedc28747d0a41d38f |
| data/staged/mobility/oa_module1_tract2020.parquet | 16,362,438 | 26cb468a319ad2e727c5eac733b4a47b77a83781ae189e1828e8c40bc0e3ace8 | 5702bad38b81b5ee0b3d01ba9d1ece7d23328985eafba269720a0af5ef6820d4 |
| data/staged/mobility/oa_outcomes_county.parquet | 721,343 | fd7bc55294fbd3f7a646f27083500c1f12415775db7a3607878b522d277f8140 | fd7bc55294fbd3f7a646f27083500c1f12415775db7a3607878b522d277f8140 |
| data/staged/ses/acs5_tract2020_2019_2023.parquet | 7,232,424 | 5fe6f730a35225f14d5d8abfc6b9bce58b52e545f309a7eac74f1dae74d7a31c | 7b4a2475b3852d3f91804716a04566ef8ceadacec2be9fb85f5eb642078efac6 |
| data/staged/svi/svi_tract2020_2022.parquet | 17,740,415 | cb3860d8edc2aaf1f2a02211a58f7a0b80457eb6ad9a37b5ec8abb2d5da9da10 | e371a4a696d5bed05d54ad7c5b6ca2504508805b847ec415c2322be7f66ec2b2 |

Outputs:

| File | SHA-256 |
| --- | --- |
| facts.json | 645c444c9d73b77edbf4beec61af11c593246bb62108e73fcb39246cd02dc514 |
| figures/eitc_vs_top_bracket.json | 5f2d3a2175eef08ee29e9ecb47adf6dcefa80ab64945c2d6c0534ad2bee3e4e9 |
| figures/holc_grade_outcomes.json | ed71dcc8ae93a24982b4f1b513aa6b9bcef0c4bf0ca8a01149f62b28a7d3217d |
| figures/home_value_vs_le.json | a15dc362d0f26e32de82fffd0fafd3ddc10eff442a57b3dc150d7d696b3075c8 |
| meta.json | 4356ada59db3ade078370dbf85fcaab9e13d5aed93bd6e1f75290b151cf570a8 |
| public/data/home-value-income-and-redlining/home_value_scatter.json | ac50c7520da430d4362df1794bde7c8165145d1e1736699d76b59393402e5e8d |
| refs.json | 9f9f46a7d169aa82999a0f1f0ba2c13ae4e0a3d20f7f47da885b1f6a2c1c55bf |

## Machine formats

- Report (HTML): https://vishalsingh.org/zipcode/home-value-income-and-redlining/report
- Report (Markdown): https://vishalsingh.org/zipcode/home-value-income-and-redlining/report.md
- Every estimate (JSON): https://vishalsingh.org/zipcode/report/home-value-income-and-redlining/facts.json
- Table 1, figure "home_value_vs_le" (JSON): https://vishalsingh.org/zipcode/report/home-value-income-and-redlining/figures/home_value_vs_le.json
- Table 2, figure "eitc_vs_top_bracket" (JSON): https://vishalsingh.org/zipcode/report/home-value-income-and-redlining/figures/eitc_vs_top_bracket.json
- Table 3, figure "holc_grade_outcomes" (JSON): https://vishalsingh.org/zipcode/report/home-value-income-and-redlining/figures/holc_grade_outcomes.json
- Payload of "home_value_vs_le" (JSON): https://vishalsingh.org/zipcode/data/home-value-income-and-redlining/home_value_scatter.json
- Index of every report: https://vishalsingh.org/zipcode/llms.txt; all reports in one file: https://vishalsingh.org/zipcode/llms-full.txt

---

# Mortgage denials: report

Chapter 5.3 of ZIP Code Destiny, by Vishal Singh (NYU Stern). Generated from the chapter's data files; the reading version is at https://vishalsingh.org/zipcode/mortgage-denials. Table n of this report is figure n of the chapter.

## Identification

- Title: Mortgage denials
- Label: The application and the answer
- Chapter: 5.3 (Part V, Home, work and money; An American life)
- Question: How do mortgage denial rates differ by tract deprivation, applicant race and the historical redlining grade?
- Dek: Lenders denied 13.8% of the home-purchase applications they decided in 2024 in the most deprived tenth of census tracts and 7.8% in the least deprived tenth.
- Grain: neighborhoods
- Status: published
- Published: 2026-10-09
- Updated: not revised
- Data sources:
  - HMDA mortgage lending. Home Mortgage Disclosure Act Loan/Application Register, 2024, first-lien site-built one-to-four-unit principal-residence applications aggregated to 2020 tracts by loan purpose and applicant group; cells under 10 applications withheld; years: 2024; weights: none (counts of decided applications, pooled over tracts)
  - Area Deprivation Index. Area Deprivation Index 2023, national rank by tract; years: 2023 release (version 4.0.1; built on ACS 2019-2023); weights: none (tenths hold near-equal numbers of tracts)
  - Mapping Inequality (HOLC maps). Mapping Inequality HOLC residential security maps, 1935 to 1940, graded polygons overlaid on 2020 tracts; years: 1930s-1940s maps (2023 digitization); weights: decided applications
  - Published estimates from cited studies. One comparison quoted from the abstract of Hanson, Hawley, Martin and Liu (2016); years: as reported in each study; weights: as reported in the paper
- Canonical URL: https://vishalsingh.org/zipcode/mortgage-denials
- Report: https://vishalsingh.org/zipcode/mortgage-denials/report
- Markdown: https://vishalsingh.org/zipcode/mortgage-denials/report.md
- BibTeX key: singh2026mortgagedenials

```bibtex
@misc{singh2026mortgagedenials,
  author = {Singh, Vishal},
  title = {Mortgage denials},
  year = {2026},
  url = {https://vishalsingh.org/zipcode/mortgage-denials},
  note = {In ZIP Code Destiny. Published 9 October 2026.}
}
```

## Principal estimates

The facts the dek and the opening prose cite, in order of appearance.

| Estimand | Population | Period | Value | Interval | n | Weights | Source | Fact id |
| --- | --- | --- | ---: | --- | ---: | --- | --- | --- |
| denied applications divided by applications that were originated or denied, summed over the tracts (a pooled rate), county-cluster bootstrap interval | home-purchase applications decided in 2024 in the most deprived tenth of tracts by Area Deprivation Index national rank (tenths hold equal numbers of tracts) | applications decided in 2024 | 13.8% | 95% CI 13.3% to 14.3% | 7,591 tracts | none (counts of decided applications, pooled over tracts) | HMDA mortgage lending | rate_d10 |
| denied applications divided by applications that were originated or denied, summed over the tracts (a pooled rate), county-cluster bootstrap interval | home-purchase applications decided in 2024 in the least deprived tenth of tracts by Area Deprivation Index national rank (tenths hold equal numbers of tracts) | applications decided in 2024 | 7.8% | 95% CI 7.2% to 8.4% | 7,584 tracts | none (counts of decided applications, pooled over tracts) | HMDA mortgage lending | rate_d1 |
| count of tracts where the cells for non-Hispanic Black and non-Hispanic white applicants are both published | tracts of the 50 states and DC | 2024 | 5,657 |   |   | none (full count) | HMDA mortgage lending | n_shared_black |
| tracts where the cells for non-Hispanic Black and non-Hispanic white applicants are both published, divided by tracts with a published all-applicant cell | home-purchase applications that a covered lender originated or denied in 2024, first-lien loans on site-built homes with up to four units that applicants intend as a principal residence, in tracts of the 50 states and DC with a published cell | applications decided in 2024 | 7.4% |   | 76,072 tracts | none (each tract counts once) | HMDA mortgage lending | share_shared_black |
| denied applications divided by applications that were originated or denied, summed over the tracts (a pooled rate), county-cluster bootstrap interval | non-Hispanic Black applicants whose home-purchase application was decided in 2024, in the tracts where the cells for non-Hispanic Black and non-Hispanic white applicants are both published | applications decided in 2024 | 16.2% | 95% CI 15.7% to 16.8% | 5,657 tracts | none (counts of decided applications, pooled over tracts) | HMDA mortgage lending | same_black |
| denied applications divided by applications that were originated or denied, summed over the tracts (a pooled rate), county-cluster bootstrap interval | non-Hispanic white applicants whose home-purchase application was decided in 2024, in the tracts where the cells for non-Hispanic Black and non-Hispanic white applicants are both published | applications decided in 2024 | 6.2% | 95% CI 6.0% to 6.5% | 5,657 tracts | none (counts of decided applications, pooled over tracts) | HMDA mortgage lending | same_white_vs_black |

## Findings

Lenders denied 13.8% (95% CI 13.3% to 14.3%) of the home-purchase applications that they decided in 2024 in the most deprived tenth of census tracts and 7.8% (95% CI 7.2% to 8.4%) in the least deprived tenth. In the 5,657 tracts that have published rates for both groups, which are 7.4% of all tracts, the denial rate was 16.2% (95% CI 15.7% to 16.8%) for non-Hispanic Black applicants and 6.2% (95% CI 6.0% to 6.5%) for non-Hispanic white applicants.

The data are tract totals from the Home Mortgage Disclosure Act, which requires lenders above size thresholds to report every application and what became of it. A denial rate is a share of applications that reached a decision. It leaves out people who never applied, applications that were withdrawn, and everything a lender weighs about the applicant: credit history, debts, savings and the size of the loan against the value of the home. The chapter describes where and for whom applications ended in a denial. It does not measure whether a lender weighed an applicant's race in a decision, and the gaps below are not estimates of discrimination.

The chapter cuts the data three ways: tracts ranked by deprivation, applicant groups within tracts, and the grades that the Home Owners' Loan Corporation gave neighborhoods in the 1930s. Chapter 5.2, Home value, income and redlining (https://vishalsingh.org/zipcode/home-value-income-and-redlining) compares those grades with life expectancy and home values, and Chapter 5.1, Work, wages and access to jobs (https://vishalsingh.org/zipcode/work-wages-and-access-to-jobs) covers the earnings that stand behind an application.

### What a denial rate counts

The file holds first-lien loans on site-built homes with up to four units that applicants intend as a principal residence, reported for 2024 by lenders above the HMDA reporting thresholds. It splits each tract by loan purpose. Home-purchase applications were denied 9.0% of the time, refinance applications 20.7% and cash-out refinance applications 35.8% (Table 1). The chapter uses home-purchase applications throughout. The rates differ by purpose, so a mix of purposes would let the mix of loans move a place's rate. Only the home-purchase file has enough applications by group. In it, 8,232 tracts have a published rate for non-Hispanic Black applicants, against 1,155 for refinance loans and 2,938 for cash-out loans.

Table 1. Decided home-purchase applications were denied 9.0% of the time, against 20.7% for refinance and 35.8% for cash-out refinance loans

The rate is denied applications divided by decided applications, those that the lender originated or denied. Applications that the applicant withdrew or left incomplete, loans approved but not accepted, and loans bought from another lender are outside it. A borrower who was discouraged before applying, or steered to a different loan, never enters it. Every rate here is pooled, with denied and decided counts summed over tracts. Pooled over 76,072 tracts in every state and DC, lenders denied 278,451 of the 3,097,390 home-purchase applications that they decided in 2024, a rate of 9.0% (95% CI 8.7% to 9.3%). The median tract decides only 31 applications, so a single tract's rate is noisy. Intervals are 95% intervals from a bootstrap that resamples whole counties.

When the file was staged, any cell with fewer than 10 applications, whatever their outcome and purchased loans included, was suppressed. That is a rule of this book and not an HMDA flag. A suppressed cell is missing, never zero, and leaves both the denied and the decided counts. On that rule 5,999 tracts have no all-applicant home-purchase cell, each holding at most 9 applications. A cell for one group is withheld more often, because the group itself needs 10 applications in the tract. Each group's rate therefore comes from its own set of tracts.

The floor counts applications of every outcome, so a published cell can rest on few decided applications. Of the 76,072 tracts, 4,865 decided fewer than 10. Leaving out tracts below 30 decided applications lowers the overall rate from 9.0% to 8.5% (95% CI 8.2% to 8.8%).

Puerto Rico's 637 tracts with a published cell are set aside. Of those with a deprivation rank, 69% sit where the most deprived tenth of the other tracts begins, and the pooled rate there is 19.1% (95% CI 18.3% to 20.0%) on 11,409 decided applications. Including them would move the most deprived tenth from 13.8% to 14.1% and the overall rate by 0.04 points.

HMDA records an applicant's ethnicity and race as the applicant reports them, with fallback rules when the applicant does not answer. The tract file keeps four groups: applicants who are white and not Hispanic or Latino, Black or African American and not Hispanic or Latino, Hispanic or Latino of any race, and Asian and not Hispanic or Latino. The chapter calls them non-Hispanic white, non-Hispanic Black, Hispanic and non-Hispanic Asian applicants. Applicants who are American Indian or Alaska Native, Native Hawaiian or Pacific Islander, of two or more races, joint, or of unavailable race count in the all-applicant rate only. The four groups therefore do not add up to it. Loan type (conventional, FHA, VA or USDA) is not in the tract file.

### Denial rates rise with deprivation after the third tenth

The 75,978 tracts that have a deprivation rank are ranked by the Area Deprivation Index national rank and cut into tenths with near-equal numbers of tracts (tracts tied on the rank stay in one tenth). The 94 tracts with no rank are left out of this and the next two sections. The denial rate is 7.8% (95% CI 7.2% to 8.4%) in the least deprived tenth, 7.7% in the second and 7.6% in the third. The differences among those three tenths are within sampling error.

After the third tenth the rate rises at every step, to 13.8% (95% CI 13.3% to 14.3%) in the most deprived tenth (Table 2). That is 6.0 (95% CI 5.2 to 6.8) points above the least deprived tenth and 6.2 (95% CI 5.6 to 6.8) above the third. The last step, from the ninth tenth to the tenth, is the largest.

The rank is built partly from income and housing costs, so the curve mixes where an applicant lives with how much money applicants there have. The file reports both: the median applicant income and the median loan amount of each tract. Averaged over the tracts of a tenth and weighted by decided applications, they are $64 thousand (95% CI $63 thousand to $65 thousand) and $162 thousand (95% CI $158 thousand to $167 thousand) in the most deprived tenth, against $279 thousand (95% CI $266 thousand to $292 thousand) and $827 thousand (95% CI $774 thousand to $879 thousand) in the least deprived. The least deprived tenth holds the largest loans. The tract file does not say whether loan size, income or credit history sits behind a denial, so the curve describes where applications are denied and not why.

Table 2. The denial rate was 7.8% in the least deprived tenth of tracts and 13.8% in the most deprived, and it is level across the three least deprived tenths

Single tracts scatter around the tenth means. One denial moves the rate of a tract that decides few applications by several points, and the dots in the figure come only from tracts that decided at least 30.

### Black and Hispanic applicants are denied more often than white applicants

Over the tracts where each group's cell is published, non-Hispanic white applicants were denied 6.5% (95% CI 6.3% to 6.6%) of the time, non-Hispanic Asian applicants 8.9% (95% CI 8.4% to 9.4%), Hispanic applicants 13.1% (95% CI 12.4% to 13.9%) and non-Hispanic Black applicants 17.0% (95% CI 16.4% to 17.7%). Each rate rests on a different set of tracts, so that comparison mixes places with groups. Within tenths of tract deprivation, the Black and Hispanic rates are above the white rate in every tenth drawn, and so is the Asian rate in the first eight (Table 3).

The white curve is U-shaped. It is 6.7% (95% CI 6.0% to 7.5%) in the least deprived tenth, 5.4% in the third and 10.4% in the most deprived.

The other three groups rise with deprivation. The Black rate goes from 14.6% in the third tenth to 20.3% in the most deprived, and the Hispanic rate from 11.8% to 14.2%. The Asian rate goes from 7.8% in the least deprived tenth to 12.9% in the eighth.

Table 3. In every tenth of tracts with enough applications, Black and Hispanic applicants were denied more often than white applicants

A point is drawn when it rests on at least 100 tracts. That drops the Black rate in the least deprived tenth and the Asian rate in the ninth and tenth, where few tracts have 10 or more applications from the group.

### The gap within a tract shows no trend with deprivation or applicant income

Comparing groups across tracts mixes who lives where with how applications fare. A narrower comparison keeps only tracts where both groups have a published cell and sets each group's pooled rate in those tracts side by side. In the 5,657 tracts that hold both Black and white cells, Black applicants were denied 16.2% (95% CI 15.7% to 16.8%) of the time and white applicants 6.2% (95% CI 6.0% to 6.5%) (Table 4). The gap is 10.0 (95% CI 9.5 to 10.4) points, or 2.6 (95% CI 2.5 to 2.7) times the white rate.

Hispanic applicants were denied 12.8% (95% CI 12.2% to 13.5%) of the time against 6.5% (95% CI 6.2% to 6.8%) for white applicants in the same 10,189 tracts, a gap of 6.4 (95% CI 5.9 to 6.9) points or 2.0 (95% CI 1.9 to 2.1) times. For Asian applicants the rates were 8.9% (95% CI 8.5% to 9.4%) and 5.7% (95% CI 5.4% to 6.0%) in 5,490 tracts, a gap of 3.2 (95% CI 2.8 to 3.6) points or 1.6 (95% CI 1.5 to 1.7) times. The Black gap is larger than the Hispanic gap, which is larger than the Asian gap.

Over each group's own tracts the Black gap is 10.5 (95% CI 9.9 to 11.2) points, so holding the tract fixed changes little. The tracts that hold both cells are a small and selected set, and 27% of their Black cells rest on fewer than 10 decided applications. Requiring at least 10 decided applications from both groups leaves the Black gap at 10.0 (95% CI 9.5 to 10.6) points. Requiring at least 30 leaves it at 10.8 (95% CI 10.1 to 11.6) points.

Table 4. In the tracts that hold both groups, Black applicants were denied 16.2% of the time and white applicants 6.2%

The gap shows no trend with the deprivation of the tract or with the typical applicant income in it. Across tenths of the deprivation index, the same-tract gap between Black and white applicants runs from 8.6 to 10.4 points, and its interval is above zero in every tenth. Its slope is 0.00 (95% CI −0.22 to 0.21) points per tenth. Single tenths differ from one another, so the claim is about the trend and not about every tenth. The ratio of the Black rate to the white rate is lower in the most deprived tenth, 2.0 (95% CI 1.8 to 2.3), than in the fifth tenth, 2.8 (95% CI 2.6 to 3.0). The white rate is higher in the most deprived tenth, so the same gap in points is a smaller multiple.

Fifths of tracts by the median income reported on their applications run from $74 thousand or less to $125 thousand or more. Across them the gap runs from 9.1 to 10.3 points, with a slope of −0.06 (95% CI −0.33 to 0.20) points per fifth (Table 5).

Table 5. The same-tract gap between Black and white applicants stays between 8.6 and 10.4 percentage points across tenths of tract deprivation

A tract holds the neighborhood fixed and leaves the applicant free to differ. Applicants in one tract can differ in credit history, debt relative to income, the down payment, the loan type and the lender they chose, and none of these is in the tract file. A study of Boston lenders, published in 1996, added fields for credit history, debt burdens and loan-to-value ratios to the application records. With those fields, the difference in denial rates between minority and white applicants narrowed and did not disappear.[1] The tract file cannot repeat that exercise. Its gaps can be larger or smaller than the gaps among applicants with the same finances, and the data do not say which.

> Causal evidence: Whether loan officers respond differently to names that signal race (Matched-pair email correspondence experiment)
>
> Hanson, Hawley, Martin and Liu test the step that comes before an application.[2] They emailed licensed mortgage loan originators with a first inquiry about a loan. The inquiries varied the sender's name, chosen to signal a Black or a white sender, and the credit score the sender reported. Because the researchers set the name and the reported score in each inquiry, a difference in how originators replied can be attributed to them. Originators answered inquiries with Black-signaling names less often, and they gave those senders less loan information and sent fewer follow-up messages. The authors compare the size of the name effect on response with the effect of a credit score 71 points lower.
>
> The experiment measures replies to inquiries and ends before any application, so it observes no denial. Its estimates are for the originators and inquiries in the sample.

A person whose inquiry goes unanswered, or is answered with less information, may never apply, and does not appear in a denial rate. That is one way the applications in this chapter differ from the people who looked for a loan.

### Grades drawn in the 1930s and denial rates in 2024

Between 1935 and 1940 the Home Owners' Loan Corporation graded neighborhoods in the cities it surveyed from A, labeled best, to D, labeled hazardous.[3] Chapter 5.2, Home value, income and redlining (https://vishalsingh.org/zipcode/home-value-income-and-redlining) describes the maps and sets the grades beside life expectancy and home values. Here each grade is matched to 2020 tracts. A tract takes the grade that covers most of its area when graded polygons cover at least 50% of it. That gives 10,967 graded tracts in the country, of which 8,748 have a published home-purchase cell and a deprivation rank, in 221 city and state pairs. They are the cities that HOLC surveyed and not a sample of the country.

Graded tracts hold 7.3% of the decided applications in this chapter. Their pooled denial rate is 10.3% (95% CI 9.5% to 11.1%), against 8.9% (95% CI 8.6% to 9.2%) in all other tracts with a deprivation rank. Among graded tracts, denial rates rise from 7.9% (95% CI 7.0% to 9.0%) in grade A tracts to 9.4% in B and 10.9% in C, and differ little between C and D, which has 11.3% (95% CI 10.2% to 12.5%).

The gap between D and A is 3.4 (95% CI 2.1 to 4.7) points (Table 6), and the step from C to D is 0.4 (95% CI −0.6 to 1.5). Comparing tracts within the same city leaves the gap at 3.5 (95% CI 2.1 to 4.8). Requiring graded polygons to cover at least 90% of a tract leaves it at 2.8 (95% CI 1.0 to 4.6).

Comparing tracts of the same city and the same tenth of the deprivation index removes the gap. That comparison gives each of 1,008 city and tenth cells its own baseline. Grade D tracts differ from grade A tracts by 0.0 (95% CI −1.1 to 1.1) points in that comparison, and grade B and C tracts by −0.2 (95% CI −0.9 to 0.5) and −0.4 (95% CI −1.3 to 0.6), all with intervals around zero. In 2024 the differences across grades line up with the deprivation of the tracts today. The comparison does not say whether the maps had a part in how those tracts came to be deprived, and the boundary study in Chapter 5.2, Home value, income and redlining (https://vishalsingh.org/zipcode/home-value-income-and-redlining) addresses that question for housing outcomes.

Whether the grade gradient differs by applicant group cannot be told from these data. Grade D minus grade A is 1.4 (95% CI 0.2 to 2.8) points for white applicants, 0.1 (95% CI −5.5 to 6.0) for Black applicants and −1.3 (95% CI −7.7 to 4.4) for Hispanic applicants, each over the tracts where the group's cell is published. The Black and Hispanic intervals hold zero and also the all-applicant gap, and the white gap minus the Black gap is 1.3 (95% CI −5.0 to 7.4) points. Grade A tracts have few cells for these groups.

The mix of applicants also changes with the grade. Published Black and Hispanic cells account for 7.8% (95% CI 5.0% to 11.0%) of decided applications in grade A tracts and 15.0% (95% CI 11.2% to 18.9%) in grade D tracts. Those shares are lower bounds, because withheld cells are not counted. Both groups have higher denial rates, so part of the all-applicant gradient reflects who applies in each kind of tract.

Table 6. Denial rates rise from 7.9% in grade A tracts to 11.3% in grade D tracts, and no grade differs detectably from grade A among tracts of the same city and the same tenth of the deprivation index

The Cook County map shows the same data at tract scale (Table 7). Of its 1,331 tracts, 914 decided at least 20 home-purchase applications and are shaded, and 417 are gray, each with its reason in the tooltip. The Chicago map graded 759 of the county's tracts, all of them in the city, and the ungraded tracts include parts of the county that the map did not reach.

Among the shaded tracts, 53.0% of those in the top fifth of denial rates and 57.7% of those in the bottom fifth were graded, a difference of −4.7 (95% CI −14.8 to 5.6) points. That compares graded with ungraded tracts, which is not the comparison of grade D with grade A. The interval is too wide for these data to say whether a tract's place in the graded part of the county goes with a higher or a lower denial rate.

Within the graded part, the pooled rate is 5.4% (95% CI 2.5% to 8.9%) in grade A tracts, which rest on 11 tracts. It is 12.3% (95% CI 10.1% to 14.8%) in B, 11.0% (95% CI 10.0% to 12.0%) in C and 9.8% (95% CI 8.4% to 11.3%) in D, against 11.4% (95% CI 10.8% to 12.1%) in tracts with no grade.

Table 7. In Cook County, 53.0% of the tracts in the top fifth of denial rates and 57.7% of those in the bottom fifth lie in the part of the county that the 1930s map graded, a difference these data cannot tell from zero

### The denial rate rises with deprivation under all four rankings

The comparison uses the 71,169 tracts that have a published home-purchase cell and all four rankings, cut into tenths under each ranking (Table 8). Ranked by the Area Deprivation Index on that sample, the denial rate among decided applications runs from 7.8% (95% CI 7.2% to 8.5%) in the least deprived tenth to 13.7% (95% CI 13.3% to 14.3%) in the most deprived, and the rank correlation between tenth and rate is +0.95. Under the standardized index it runs from 7.0% (95% CI 6.6% to 7.4%) to 14.8% (95% CI 14.1% to 15.5%) with a rank correlation of +1.00, under median household income from 7.0% (95% CI 6.6% to 7.4%) to 13.8% (95% CI 13.2% to 14.5%) and under the bachelor's degree share from 6.8% (95% CI 6.4% to 7.4%) to 13.2% (95% CI 12.8% to 13.8%). All four rankings put the higher rate at the more deprived end with a steady rise, so none of the three alternatives counts as a change under the rule in Chapter 9.1, How far one ranking goes (https://vishalsingh.org/zipcode/how-far-one-ranking-goes).

Table 8. The mortgage denial rate is steadily worse toward the more deprived end under the ADI and under each of the three other rankings

### What these data cannot show

Applications are not the population. A denial rate is a share of the applications that reached a decision, and who applies differs across tracts and groups. People who expect a denial, who are steered to another loan or who cannot get an answer to an inquiry do not appear, and withdrawn applications leave the rate. The file covers one year and the lenders above the reporting thresholds.

The tract totals hold no credit score, debt-to-income ratio, loan-to-value ratio, loan type or lender for any applicant. A gap between groups, even inside one tract, is therefore not evidence of discrimination by itself and not a measure of how much treatment differs. The application-level HMDA records carry some of those fields and up to four denial reasons for each denied application. Lenders report those reasons themselves, they are not audited, and the tract aggregate does not keep them.

The groups are reported groups. The ethnicity and race in HMDA come from the applicant where the applicant answers and from fallback rules where not, and the four groups leave out other and unreported applicants. A group's cell exists only in tracts with 10 or more of its applications, so group comparisons cover the tracts with more applicants of that group and not a random sample of tracts.

The intervals resample counties. Applications at one lender share its rules and cross county lines, so the intervals do not capture correlation within a lender and may be narrower than the uncertainty. The tenths and grades describe tracts and the applications reported in them. They are not the experience of one borrower, and a pattern across places need not hold for the people inside them (ecological inference). The deprivation index understates deprivation in high-cost cities and is built partly from income and housing costs, which are also what a lender weighs.

The grades followed the housing, incomes and populations of the 1930s, so a tract's grade stands for a mix of what the neighborhood was and what happened after. The adjusted contrast compares tracts of the same city and the same tenth of the index and cannot say whether the maps shaped today's deprivation. A city that HOLC did not survey is absent from the comparison, which does not mean it had no redlining. Tooltip intervals on the Cook County map treat applications as independent and are narrower than a bootstrap over tracts would give.

## Tables

### Table 1. Decided home-purchase applications were denied 9.0% of the time, against 20.7% for refinance and 35.8% for cash-out refinance loans

Denied share of decided applications, tracts of the 50 states and DC with a published cell, 2024, by loan purpose

| Loan purpose | Tracts with a published cell | Decided applications | Denial rate (%) | Lower limit of the 95% interval (%) | Upper limit of the 95% interval (%) | Tracts with a published rate for Black applicants |
| --- | ---: | ---: | ---: | ---: | ---: | ---: |
| Home purchase | 76,072 | 3,097,390 | 9.0 | 8.7 | 9.3 | 8,232 |
| Refinance | 36,779 | 440,105 | 20.7 | 20.2 | 21.2 | 1,155 |
| Cash-out refinance | 46,525 | 616,812 | 35.8 | 35.3 | 36.4 | 2,938 |

Note: Each row counts tracts whose all-applicant cell for the purpose has 10 or more applications, a withholding rule of this book's staging step that counts applications of every action taken. Cells under that floor are missing, never zero, and leave both the denied and the decided counts. Tracts differ across rows because the floor binds more often for refinance and cash-out loans.

Note: The file covers first-lien loans on site-built homes with up to four units that applicants intend as a principal residence, from lenders above the HMDA reporting thresholds. Loan type (conventional, FHA, VA or USDA) is not in the tract file. Puerto Rico is left out.

Source: HMDA Loan/Application Register, 2024, tract totals of decided applications; Area Deprivation Index, 2023; full counts, pooled rates

Data: https://vishalsingh.org/zipcode/report/mortgage-denials/figures/denial_by_loan_purpose.json

### Table 2. The denial rate was 7.8% in the least deprived tenth of tracts and 13.8% in the most deprived, and it is level across the three least deprived tenths

Denied share of decided home-purchase applications, 2024, against the tract's Area Deprivation Index national rank: sampled tracts with at least 30 decided applications as dots and pooled rates for tenths of tracts as the line

No inline rows: the values of Table 2 are in https://vishalsingh.org/zipcode/data/mortgage-denials/denial_vs_deprivation_tracts.json.

Note: Dots are drawn from the tracts that decided at least 30 applications, with a fixed seed (stated in the methods note). The pooled rates use every tract with a published cell and a rank, 75,978 tracts, in tenths that hold near-equal numbers of tracts. A tract with no rank is left out (94 tracts).

Note: Across all the ranked tracts the pooled denial rate is 9.0%.

Note: The Area Deprivation Index is a national percentile, 1 least and 100 most deprived, and it understates deprivation in high-cost cities. Rates are not adjusted for the applicant's credit, debts, loan size or loan type, which the tract file does not hold.

Source: HMDA Loan/Application Register, 2024, tract totals of decided applications; Area Deprivation Index, 2023; full counts, pooled rates

Data: https://vishalsingh.org/zipcode/report/mortgage-denials/figures/denial_vs_deprivation.json

### Table 3. In every tenth of tracts with enough applications, Black and Hispanic applicants were denied more often than white applicants

Denied share of decided home-purchase applications, 2024, by applicant group and tenth of tracts on the Area Deprivation Index (1 = least deprived), points on at least 100 tracts

| conf_high | conf_low | estimate | series | tenth | tracts |
| ---: | ---: | ---: | --- | ---: | ---: |
| 7.47 | 6.01 | 6.72 | Non-Hispanic white | 1 | 4805 |
| 5.85 | 5.24 | 5.53 | Non-Hispanic white | 2 | 5356 |
| 5.65 | 5.17 | 5.41 | Non-Hispanic white | 3 | 5868 |
| 5.79 | 5.34 | 5.56 | Non-Hispanic white | 4 | 6305 |
| 5.86 | 5.42 | 5.64 | Non-Hispanic white | 5 | 6150 |
| 6.51 | 6.03 | 6.28 | Non-Hispanic white | 6 | 6079 |
| 6.90 | 6.49 | 6.69 | Non-Hispanic white | 7 | 6079 |
| 7.50 | 7.08 | 7.29 | Non-Hispanic white | 8 | 5995 |
| 8.55 | 8.08 | 8.31 | Non-Hispanic white | 9 | 5763 |
| 10.77 | 10.10 | 10.43 | Non-Hispanic white | 10 | 4227 |
| 16.80 | 13.85 | 15.46 | Non-Hispanic Black | 2 | 439 |
| 15.56 | 13.77 | 14.64 | Non-Hispanic Black | 3 | 633 |
| 16.86 | 14.74 | 15.80 | Non-Hispanic Black | 4 | 910 |
| 17.34 | 15.54 | 16.45 | Non-Hispanic Black | 5 | 1022 |
| 18.32 | 15.95 | 17.18 | Non-Hispanic Black | 6 | 1099 |
| 18.03 | 15.91 | 16.97 | Non-Hispanic Black | 7 | 1075 |
| 18.60 | 16.31 | 17.44 | Non-Hispanic Black | 8 | 973 |
| 19.54 | 17.20 | 18.38 | Non-Hispanic Black | 9 | 902 |
| 21.58 | 19.02 | 20.30 | Non-Hispanic Black | 10 | 1081 |
| 14.04 | 10.71 | 12.35 | Hispanic (any race) | 1 | 389 |
| 13.35 | 10.85 | 12.01 | Hispanic (any race) | 2 | 1376 |
| 12.59 | 11.07 | 11.85 | Hispanic (any race) | 3 | 1847 |
| 13.28 | 11.39 | 12.31 | Hispanic (any race) | 4 | 2307 |
| 14.32 | 12.51 | 13.46 | Hispanic (any race) | 5 | 2242 |
| 14.49 | 12.39 | 13.43 | Hispanic (any race) | 6 | 2143 |
| 14.89 | 12.78 | 13.83 | Hispanic (any race) | 7 | 1874 |
| 15.46 | 13.05 | 14.21 | Hispanic (any race) | 8 | 1536 |
| 15.24 | 12.83 | 14 | Hispanic (any race) | 9 | 1285 |
| 15.74 | 12.86 | 14.18 | Hispanic (any race) | 10 | 1045 |
| 8.74 | 7.25 | 7.84 | Non-Hispanic Asian | 1 | 2241 |
| 9.19 | 7.98 | 8.51 | Non-Hispanic Asian | 2 | 1661 |
| 9.69 | 8.02 | 8.85 | Non-Hispanic Asian | 3 | 1243 |
| 10.36 | 8.83 | 9.55 | Non-Hispanic Asian | 4 | 938 |
| 11.42 | 9.05 | 10.17 | Non-Hispanic Asian | 5 | 626 |
| 11.64 | 9.10 | 10.40 | Non-Hispanic Asian | 6 | 394 |
| 14.02 | 10.09 | 11.92 | Non-Hispanic Asian | 7 | 237 |
| 14.47 | 11.01 | 12.85 | Non-Hispanic Asian | 8 | 103 |

Note: Groups are the applicant's reported ethnicity and race as HMDA derives them: non-Hispanic white, non-Hispanic Black, Hispanic of any race and non-Hispanic Asian. Other and unreported applicants count in the all-applicant rate only, so the four groups do not add to it.

Note: A group's cell is published only in tracts with 10 or more of its applications, so each line covers a different set of tracts, and the sets for Black, Hispanic and Asian applicants are smaller than the set for white applicants. A point is drawn when it rests on at least 100 tracts; the Black line starts at the second tenth and the Asian line stops at the eighth.

Source: HMDA Loan/Application Register, 2024, tract totals of decided applications; Area Deprivation Index, 2023; full counts, pooled rates

Data: https://vishalsingh.org/zipcode/report/mortgage-denials/figures/denial_by_group_and_deprivation.json

### Table 4. In the tracts that hold both groups, Black applicants were denied 16.2% of the time and white applicants 6.2%

Denied share of decided home-purchase applications, 2024, for each applicant group and for non-Hispanic white applicants in the same tracts, tracts where both cells are published

| group | hi | lo | rate | tracts | white | white_hi | white_lo |
| --- | ---: | ---: | ---: | ---: | ---: | ---: | ---: |
| Non-Hispanic Black applicants | 16.77 | 15.67 | 16.20 | 5657 | 6.23 | 6.50 | 5.96 |
| Hispanic (any race) applicants | 13.53 | 12.17 | 12.83 | 10189 | 6.47 | 6.77 | 6.19 |
| Non-Hispanic Asian applicants | 9.37 | 8.50 | 8.91 | 5490 | 5.70 | 5.99 | 5.42 |

Note: The same tract holds the neighborhood fixed, not the applicant. The tract file has no credit score, debt-to-income ratio, loan-to-value ratio, loan type or lender for each applicant, so a gap here is not evidence of discrimination by itself and the data cannot say how much of it these items account for.

Note: The tracts that hold both cells are 7.4% of all tracts for Black applicants. The tracts differ by row (5,657 for Black applicants, 10,189 for Hispanic applicants and 5,490 for Asian applicants), because a cell exists only where a group has 10 or more applications. Over each group's own tracts the Black gap is 10.5 points.

Source: HMDA Loan/Application Register, 2024, tract totals of decided applications; Area Deprivation Index, 2023; full counts, pooled rates

Data: https://vishalsingh.org/zipcode/report/mortgage-denials/figures/same_tract_gaps.json

### Table 5. The same-tract gap between Black and white applicants stays between 8.6 and 10.4 percentage points across tenths of tract deprivation

Black minus white denied share of decided home-purchase applications, 2024, in tracts where both cells are published, by tenth of tracts on the deprivation index and by fifth of tracts on median applicant income

#### Table 5.1. By tract deprivation

The Black minus white gap is positive in every tenth of tract deprivation

Percentage points of denial rate, Black minus white applicants in the same tracts, by tenth of tracts on the Area Deprivation Index (1 = least deprived)

| conf_high | conf_low | gap | tenth | tracts |
| ---: | ---: | ---: | ---: | ---: |
| 13.59 | 7.64 | 10.31 | 1 | 77 |
| 11.12 | 7.84 | 9.57 | 2 | 335 |
| 9.56 | 7.69 | 8.61 | 3 | 514 |
| 10.79 | 8.83 | 9.81 | 4 | 725 |
| 11.14 | 9.46 | 10.33 | 5 | 809 |
| 11.61 | 9.34 | 10.45 | 6 | 849 |
| 10.36 | 8.55 | 9.44 | 7 | 776 |
| 10.85 | 8.70 | 9.72 | 8 | 628 |
| 11.39 | 9.06 | 10.19 | 9 | 547 |
| 10.72 | 7.92 | 9.32 | 10 | 395 |

#### Table 5.2. By tract applicant income

The Black minus white gap is positive in every fifth of tracts by median applicant income

Percentage points of denial rate, Black minus white applicants in the same tracts, by fifth of tracts on the median income reported by applicants (1 = lowest)

| conf_high | conf_low | fifth | gap | tracts |
| ---: | ---: | ---: | ---: | ---: |
| 10.06 | 8.46 | 1 | 9.23 | 1124 |
| 11.06 | 9.51 | 2 | 10.30 | 1149 |
| 11.11 | 9.30 | 3 | 10.15 | 1103 |
| 10.68 | 9.19 | 4 | 9.96 | 1146 |
| 9.99 | 8.13 | 5 | 9.12 | 1135 |

Note: The tenths hold near-equal numbers of the tracts with a rank (tracts tied on the rank stay together); the fifths hold near-equal numbers of the 5,657 tracts that hold both cells and a median applicant income, so a fifth groups tracts by the typical income of their applicants and not applicants by their own income. The median income is the median of the incomes reported on the tract's applications.

Note: The ratio of the Black to the white rate is 2.8 in the fifth tenth and 2.0 in the most deprived tenth.

Source: HMDA Loan/Application Register, 2024, tract totals of decided applications; Area Deprivation Index, 2023; full counts, pooled rates

Data: https://vishalsingh.org/zipcode/report/mortgage-denials/figures/gap_by_deprivation_and_income.json

### Table 6. Denial rates rise from 7.9% in grade A tracts to 11.3% in grade D tracts, and no grade differs detectably from grade A among tracts of the same city and the same tenth of the deprivation index

Denied share of decided home-purchase applications, 2024, in tracts graded on the 1930s HOLC maps: by grade, as a difference from grade A under three adjustments, and by applicant group

#### Table 6.1. By grade

The denial rate rises from grade A to grade C and differs little between grades C and D

Denied share of decided home-purchase applications, 2024, pooled over tracts of each HOLC grade

| conf_high | conf_low | estimate | grade | label | tracts |
| ---: | ---: | ---: | --- | --- | ---: |
| 8.96 | 6.98 | 7.91 | A | Grade A | 650 |
| 10.27 | 8.47 | 9.37 | B | Grade B | 2003 |
| 11.81 | 10.01 | 10.91 | C | Grade C | 4008 |
| 12.50 | 10.21 | 11.32 | D | Grade D | 2087 |

#### Table 6.2. Within city and deprivation

Within a city and a tenth of the deprivation index, no grade differs detectably from grade A

Denial rate of tracts of each grade minus that of grade A tracts, percentage points, from weighted regressions with errors clustered by county

| comparison | conf_high | conf_low | estimate | specification | tracts |
| --- | ---: | ---: | ---: | --- | ---: |
| Grade B against grade A | 2.41 | 0.520 | 1.47 | No adjustment | 8748 |
| Grade C against grade A | 4.10 | 1.91 | 3.01 | No adjustment | 8748 |
| Grade D against grade A | 4.74 | 2.09 | 3.42 | No adjustment | 8748 |
| Grade B against grade A | 2.59 | 1.01 | 1.80 | Same city | 8748 |
| Grade C against grade A | 3.97 | 1.99 | 2.98 | Same city | 8748 |
| Grade D against grade A | 4.83 | 2.15 | 3.49 | Same city | 8748 |
| Grade B against grade A | 0.480 | −0.910 | −0.210 | Same city and tenth of the deprivation index | 8748 |
| Grade C against grade A | 0.620 | −1.35 | −0.360 | Same city and tenth of the deprivation index | 8748 |
| Grade D against grade A | 1.08 | −1.06 | 0.010 | Same city and tenth of the deprivation index | 8748 |

#### Table 6.3. By applicant group

The grade D minus grade A difference is 1.4 points for white applicants, and the intervals for Black and Hispanic applicants are too wide to say whether it differs

Denied share of decided home-purchase applications, 2024, by applicant group and HOLC grade of the tract, points on at least 30 tracts

| conf_high | conf_low | estimate | grade | group | label | tracts |
| ---: | ---: | ---: | --- | --- | --- | ---: |
| 5.94 | 4.54 | 5.17 | A | Non-Hispanic white | Grade A | 517 |
| 6.39 | 5.12 | 5.71 | B | Non-Hispanic white | Grade B | 1250 |
| 6.79 | 5.66 | 6.18 | C | Non-Hispanic white | Grade C | 1912 |
| 7.91 | 5.47 | 6.52 | D | Non-Hispanic white | Grade D | 848 |
| 25.15 | 15.20 | 20.41 | A | Non-Hispanic Black | Grade A | 63 |
| 20.99 | 15.16 | 17.88 | B | Non-Hispanic Black | Grade B | 301 |
| 21.40 | 17.73 | 19.60 | C | Non-Hispanic Black | Grade C | 494 |
| 23.35 | 17.39 | 20.49 | D | Non-Hispanic Black | Grade D | 252 |
| 23.31 | 11.30 | 16.97 | A | Hispanic (any race) | Grade A | 60 |
| 16.20 | 12.54 | 14.17 | B | Hispanic (any race) | Grade B | 322 |
| 15.70 | 13.78 | 14.90 | C | Hispanic (any race) | Grade C | 792 |
| 17.20 | 14.08 | 15.72 | D | Hispanic (any race) | Grade D | 331 |
| 9.94 | 4.33 | 6.80 | A | Non-Hispanic Asian | Grade A | 37 |
| 12.50 | 7.66 | 10.10 | B | Non-Hispanic Asian | Grade B | 159 |
| 14.48 | 9.04 | 11.98 | C | Non-Hispanic Asian | Grade C | 313 |
| 14.19 | 8.04 | 11.17 | D | Non-Hispanic Asian | Grade D | 168 |

Note: A tract takes a grade when graded polygons cover at least 50% of its area, and the grade that covers the most of it. The tracts are the 8,748 graded tracts with a published cell and a deprivation rank, in 221 city and state pairs. With a stricter rule, graded polygons covering at least 90% of a tract, the gap between grades D and A is 2.8 points. The comparison is descriptive: the grades followed the housing, incomes and populations of the 1930s, and the HOLC cities are not a national sample.

Note: The adjusted differences come from weighted regressions of the tract denial rate on grade, weighted by decided applications, with errors clustered by county, on the 8,748 graded tracts. The third adjustment gives each of 1,008 city and tenth cells its own baseline, so a grade is compared with another grade only inside a cell that holds tracts of both.

Source: HMDA Loan/Application Register, 2024, tract totals of decided applications; Mapping Inequality HOLC maps (1935-1940) overlaid on 2020 tracts; Area Deprivation Index, 2023; full counts, pooled rates

Data: https://vishalsingh.org/zipcode/report/mortgage-denials/figures/denial_by_holc_grade.json

### Table 7. In Cook County, 53.0% of the tracts in the top fifth of denial rates and 57.7% of those in the bottom fifth lie in the part of the county that the 1930s map graded, a difference these data cannot tell from zero

Denied share of decided home-purchase applications, 2024, in Cook County tracts with at least 20 decided applications, and the HOLC grade covering most of each tract

No inline rows: the values of Table 7 are in https://vishalsingh.org/zipcode/data/mortgage-denials/cook_denial_map.json.

Views: Denial rate (every row); 1930s HOLC grade (every row).

Note: 914 of the 1,331 tracts have a rate and 417 do not, because the tract has fewer than 20 decided applications, fewer than 10 applications in all (withheld), or no row in the file. 759 tracts have a grade, all of them in Chicago. The rate in the tooltip carries a 95% interval from the Wilson formula, which treats applications as independent, so it is narrower than a bootstrap over tracts would give.

Note: The share graded in the top fifth minus the share in the bottom fifth is −4.7 percentage points, with an interval from a bootstrap over tracts that includes zero. Pooled over the shaded tracts, the denial rate is 5.4% in grade A tracts, 12.3% in B, 11.0% in C, 9.8% in D and 11.4% in tracts with no grade, with intervals from a bootstrap over tracts.

Source: HMDA Loan/Application Register, 2024, tract totals of decided applications; Mapping Inequality HOLC maps (1935-1940) overlaid on 2020 tracts; full counts, pooled rates

Data: https://vishalsingh.org/zipcode/report/mortgage-denials/figures/cook_denial_map.json

### Table 8. The mortgage denial rate is steadily worse toward the more deprived end under the ADI and under each of the three other rankings

Denied share of decided applications (%) by tenth of tracts ranked four ways, over the 71,169 tracts that have all four rankings

| bin | conf_high | conf_low | estimate | n_units | ranking |
| ---: | ---: | ---: | ---: | ---: | --- |
| 1 | 8.47 | 7.17 | 7.83 | 7,181 | Area Deprivation Index |
| 2 | 8.24 | 7.25 | 7.75 | 7,109 | Area Deprivation Index |
| 3 | 7.97 | 7.22 | 7.60 | 7,041 | Area Deprivation Index |
| 4 | 8.73 | 7.92 | 8.32 | 7,292 | Area Deprivation Index |
| 5 | 9.03 | 8.13 | 8.57 | 7,035 | Area Deprivation Index |
| 6 | 9.62 | 8.69 | 9.15 | 7,149 | Area Deprivation Index |
| 7 | 10.02 | 9.14 | 9.57 | 7,164 | Area Deprivation Index |
| 8 | 10.47 | 9.59 | 10.01 | 6,993 | Area Deprivation Index |
| 9 | 11.31 | 10.49 | 10.88 | 7,045 | Area Deprivation Index |
| 10 | 14.26 | 13.29 | 13.75 | 7,160 | Area Deprivation Index |
| 1 | 7.41 | 6.62 | 7.01 | 7,116 | Standardized index |
| 2 | 7.34 | 6.76 | 7.04 | 7,117 | Standardized index |
| 3 | 7.98 | 7.39 | 7.67 | 7,117 | Standardized index |
| 4 | 8.57 | 7.89 | 8.23 | 7,117 | Standardized index |
| 5 | 9.08 | 8.39 | 8.73 | 7,117 | Standardized index |
| 6 | 9.53 | 8.87 | 9.19 | 7,117 | Standardized index |
| 7 | 10.73 | 9.86 | 10.28 | 7,117 | Standardized index |
| 8 | 11.42 | 10.54 | 10.98 | 7,117 | Standardized index |
| 9 | 12.50 | 11.60 | 12.03 | 7,117 | Standardized index |
| 10 | 15.45 | 14.15 | 14.78 | 7,117 | Standardized index |
| 1 | 7.39 | 6.64 | 7.01 | 7,116 | Median household income |
| 2 | 7.62 | 7.02 | 7.33 | 7,118 | Median household income |
| 3 | 8.12 | 7.49 | 7.80 | 7,116 | Median household income |
| 4 | 8.71 | 8.06 | 8.38 | 7,117 | Median household income |
| 5 | 9.35 | 8.59 | 8.96 | 7,117 | Median household income |
| 6 | 9.68 | 8.92 | 9.29 | 7,117 | Median household income |
| 7 | 10.35 | 9.57 | 9.95 | 7,116 | Median household income |
| 8 | 11.18 | 10.29 | 10.73 | 7,121 | Median household income |
| 9 | 12.29 | 11.23 | 11.74 | 7,114 | Median household income |
| 10 | 14.48 | 13.23 | 13.80 | 7,117 | Median household income |
| 1 | 7.36 | 6.41 | 6.84 | 7,116 | Bachelor's degree share |
| 2 | 7.57 | 6.83 | 7.18 | 7,116 | Bachelor's degree share |
| 3 | 8.05 | 7.31 | 7.68 | 7,118 | Bachelor's degree share |
| 4 | 8.53 | 7.88 | 8.20 | 7,114 | Bachelor's degree share |
| 5 | 9.13 | 8.45 | 8.78 | 7,122 | Bachelor's degree share |
| 6 | 9.67 | 8.84 | 9.24 | 7,115 | Bachelor's degree share |
| 7 | 10.31 | 9.52 | 9.90 | 7,123 | Bachelor's degree share |
| 8 | 11.10 | 10.37 | 10.73 | 7,103 | Bachelor's degree share |
| 9 | 11.84 | 11.07 | 11.44 | 7,123 | Bachelor's degree share |
| 10 | 13.75 | 12.75 | 13.23 | 7,119 | Bachelor's degree share |

Note: Least and most deprived tenth: ADI 7.8% and 13.7%, rank correlation +0.95; standardized index 7.0% and 14.8%, +1.00; median household income 7.0% and 13.8%, +1.00; bachelor's degree share 6.8% and 13.2%, +1.00.

Note: The standardized index applies the ADI's coefficients to its measures after standardizing each one, so home value and income carry less weight than in the ADI. Income and degree share are single census measures, each ranked from the highest value to the lowest.

Source: HMDA Loan/Application Register, 2024, tract totals of decided applications; Area Deprivation Index, 2023; full counts, pooled rates; ADI 2023 and ACS 2019-2023 tract tables (B19013, B15003 and the 15 rebuilt ADI measures)

Data: https://vishalsingh.org/zipcode/report/mortgage-denials/figures/alt_rankings.json

## All estimates

Every fact in facts.json (138), sorted by id.

| Fact id | Value | Unit | Estimand | Population | Period | Interval | n | Weights | Source | Note |
| --- | ---: | --- | --- | --- | --- | --- | ---: | --- | --- | --- |
| alt_adi_least | 7.8% | Denied share of decided applications (%) | pooled denial rate (denied applications divided by applications originated or denied, summed over the tracts); 95% intervals from a bootstrap over counties | the least deprived tenth of tracts ranked by Area Deprivation Index, among the tracts with all four rankings | applications decided in 2024 | 95% CI 7.2% to 8.5% | 7,181 tracts | none (counts of decided applications, pooled over tracts) | HMDA mortgage lending |   |
| alt_adi_most | 13.7% | Denied share of decided applications (%) | pooled denial rate (denied applications divided by applications originated or denied, summed over the tracts); 95% intervals from a bootstrap over counties | the most deprived tenth of tracts ranked by Area Deprivation Index, among the tracts with all four rankings | applications decided in 2024 | 95% CI 13.3% to 14.3% | 7,160 tracts | none (counts of decided applications, pooled over tracts) | HMDA mortgage lending |   |
| alt_adi_rho | +0.95 | rank correlation | Spearman rank correlation between bin order and bin mean, each bin counted once | the bins of tracts ranked by Area Deprivation Index | applications decided in 2024 |   |   | each bin counted once | HMDA mortgage lending |   |
| alt_degree_least | 6.8% | Denied share of decided applications (%) | pooled denial rate (denied applications divided by applications originated or denied, summed over the tracts); 95% intervals from a bootstrap over counties | the least deprived tenth of tracts ranked by Bachelor's degree share, among the tracts with all four rankings | applications decided in 2024 | 95% CI 6.4% to 7.4% | 7,116 tracts | none (counts of decided applications, pooled over tracts) | HMDA mortgage lending |   |
| alt_degree_most | 13.2% | Denied share of decided applications (%) | pooled denial rate (denied applications divided by applications originated or denied, summed over the tracts); 95% intervals from a bootstrap over counties | the most deprived tenth of tracts ranked by Bachelor's degree share, among the tracts with all four rankings | applications decided in 2024 | 95% CI 12.8% to 13.8% | 7,119 tracts | none (counts of decided applications, pooled over tracts) | HMDA mortgage lending |   |
| alt_degree_rho | +1.00 | rank correlation | Spearman rank correlation between bin order and bin mean, each bin counted once | the bins of tracts ranked by Bachelor's degree share | applications decided in 2024 |   |   | each bin counted once | HMDA mortgage lending |   |
| alt_income_least | 7.0% | Denied share of decided applications (%) | pooled denial rate (denied applications divided by applications originated or denied, summed over the tracts); 95% intervals from a bootstrap over counties | the least deprived tenth of tracts ranked by Median household income, among the tracts with all four rankings | applications decided in 2024 | 95% CI 6.6% to 7.4% | 7,116 tracts | none (counts of decided applications, pooled over tracts) | HMDA mortgage lending |   |
| alt_income_most | 13.8% | Denied share of decided applications (%) | pooled denial rate (denied applications divided by applications originated or denied, summed over the tracts); 95% intervals from a bootstrap over counties | the most deprived tenth of tracts ranked by Median household income, among the tracts with all four rankings | applications decided in 2024 | 95% CI 13.2% to 14.5% | 7,117 tracts | none (counts of decided applications, pooled over tracts) | HMDA mortgage lending |   |
| alt_income_rho | +1.00 | rank correlation | Spearman rank correlation between bin order and bin mean, each bin counted once | the bins of tracts ranked by Median household income | applications decided in 2024 |   |   | each bin counted once | HMDA mortgage lending |   |
| alt_n_places | 71,169 | tracts | count of tracts in the common sample of the ranking comparison | tracts with the outcome and all four rankings | applications decided in 2024 |   |   | none (full count) | HMDA mortgage lending |   |
| alt_rho_cut | 0.9 | absolute rank correlation | threshold between a steady and an uneven gradient | the rule for changed gradients (docs/RANKING_SWITCH_RULE.md) | applications decided in 2024 |   |   | none (a fixed rule) | HMDA mortgage lending |   |
| alt_rho_margin | 0.1 | absolute rank correlation | smallest move in the rank correlation that counts as a change of shape | the rule for changed gradients (docs/RANKING_SWITCH_RULE.md) | applications decided in 2024 |   |   | none (a fixed rule) | HMDA mortgage lending |   |
| alt_std_least | 7.0% | Denied share of decided applications (%) | pooled denial rate (denied applications divided by applications originated or denied, summed over the tracts); 95% intervals from a bootstrap over counties | the least deprived tenth of tracts ranked by Standardized index, among the tracts with all four rankings | applications decided in 2024 | 95% CI 6.6% to 7.4% | 7,116 tracts | none (counts of decided applications, pooled over tracts) | HMDA mortgage lending |   |
| alt_std_most | 14.8% | Denied share of decided applications (%) | pooled denial rate (denied applications divided by applications originated or denied, summed over the tracts); 95% intervals from a bootstrap over counties | the most deprived tenth of tracts ranked by Standardized index, among the tracts with all four rankings | applications decided in 2024 | 95% CI 14.1% to 15.5% | 7,117 tracts | none (counts of decided applications, pooled over tracts) | HMDA mortgage lending |   |
| alt_std_rho | +1.00 | rank correlation | Spearman rank correlation between bin order and bin mean, each bin counted once | the bins of tracts ranked by Standardized index | applications decided in 2024 |   |   | each bin counted once | HMDA mortgage lending |   |
| cloud_floor | 30 | decided applications | the number of decided applications a tract needs to be a dot in the scatter (a rule of this chapter) | a tract in the scatter of tract rates | 2024 |   |   | none (definition) | HMDA mortgage lending |   |
| cook_bottom_graded | 57.7% | percent of tracts that are graded | share of the tracts that have a HOLC grade | Cook County tracts in the bottom fifth of shaded tracts by denial rate | applications decided in 2024 |   | 182 tracts | none (each tract counts once) | HMDA mortgage lending |   |
| cook_floor | 20 | decided applications | the number of decided applications a tract needs to be shaded (a rule of this chapter) | a Cook County tract on the map | applications decided in 2024 |   |   | none (definition) | HMDA mortgage lending |   |
| cook_graded_diff | −4.7 | percentage points of tracts graded | share graded in the top fifth minus the share graded in the bottom fifth, bootstrap over tracts | Cook County tracts in the top fifth minus the bottom fifth of shaded tracts by denial rate | applications decided in 2024 | 95% CI −14.8 to 5.6 | 365 tracts | none (each tract counts once) | HMDA mortgage lending |   |
| cook_n_graded | 759 | tracts | count of Cook County tracts with a grade (all in Chicago) | Cook County tracts covered at least half by graded HOLC polygons | HOLC maps drawn 1935 to 1940 |   |   | none (full count) | Mapping Inequality (HOLC maps) |   |
| cook_n_gray | 417 | tracts | count of tracts drawn as no data on the denial-rate view | Cook County tracts without a shaded denial rate | applications decided in 2024 |   |   | none (full count) | HMDA mortgage lending |   |
| cook_n_shaded | 914 | tracts | count of tracts shaded on the map | Cook County tracts with at least 20 decided home-purchase applications | applications decided in 2024 |   |   | none (full count) | HMDA mortgage lending |   |
| cook_n_tracts | 1,331 | tracts | count of tracts on the Cook County map | census tracts of Cook County, Illinois | 2020 tracts |   |   | none (full count) | Census TIGER/Line tracts |   |
| cook_rate_a | 5.4% | percent of decided applications denied | denied applications divided by applications that were originated or denied, summed over the tracts (a pooled rate), tract-cluster bootstrap interval | home-purchase applications decided in 2024 in shaded Cook County tracts graded A on the 1930s Chicago map | applications decided in 2024 | 95% CI 2.5% to 8.9% | 11 tracts | none (counts of decided applications, pooled over tracts) | HMDA mortgage lending |   |
| cook_rate_b | 12.3% | percent of decided applications denied | denied applications divided by applications that were originated or denied, summed over the tracts (a pooled rate), tract-cluster bootstrap interval | home-purchase applications decided in 2024 in shaded Cook County tracts graded B on the 1930s Chicago map | applications decided in 2024 | 95% CI 10.1% to 14.8% | 61 tracts | none (counts of decided applications, pooled over tracts) | HMDA mortgage lending |   |
| cook_rate_c | 11.0% | percent of decided applications denied | denied applications divided by applications that were originated or denied, summed over the tracts (a pooled rate), tract-cluster bootstrap interval | home-purchase applications decided in 2024 in shaded Cook County tracts graded C on the 1930s Chicago map | applications decided in 2024 | 95% CI 10.0% to 12.0% | 256 tracts | none (counts of decided applications, pooled over tracts) | HMDA mortgage lending |   |
| cook_rate_d | 9.8% | percent of decided applications denied | denied applications divided by applications that were originated or denied, summed over the tracts (a pooled rate), tract-cluster bootstrap interval | home-purchase applications decided in 2024 in shaded Cook County tracts graded D on the 1930s Chicago map | applications decided in 2024 | 95% CI 8.4% to 11.3% | 107 tracts | none (counts of decided applications, pooled over tracts) | HMDA mortgage lending |   |
| cook_rate_none | 11.4% | percent of decided applications denied | denied applications divided by applications that were originated or denied, summed over the tracts (a pooled rate), tract-cluster bootstrap interval | home-purchase applications decided in 2024 in shaded Cook County tracts with no grade (outside the area the Chicago map graded, or covered less than halfway) | applications decided in 2024 | 95% CI 10.8% to 12.1% | 479 tracts | none (counts of decided applications, pooled over tracts) | HMDA mortgage lending |   |
| cook_top_graded | 53.0% | percent of tracts that are graded | share of the tracts that have a HOLC grade | Cook County tracts in the top fifth of shaded tracts by denial rate | applications decided in 2024 |   | 183 tracts | none (each tract counts once) | HMDA mortgage lending |   |
| cook_tracts_a | 11 | tracts | count of tracts | shaded Cook County tracts graded A on the 1930s Chicago map | applications decided in 2024 |   |   | none (full count) | HMDA mortgage lending |   |
| gap_asian | 3.2 | percentage points of denial rate | pooled denial rate of the group minus that of non-Hispanic white applicants over the same tracts, county-cluster bootstrap interval | non-Hispanic Asian applicants minus non-Hispanic white applicants, applicants whose home-purchase application was decided in 2024, in the tracts where the cells for non-Hispanic Asian and non-Hispanic white applicants are both published | applications decided in 2024 | 95% CI 2.8 to 3.6 | 5,490 tracts | none (counts of decided applications, pooled over tracts) | HMDA mortgage lending |   |
| gap_black | 10.0 | percentage points of denial rate | pooled denial rate of the group minus that of non-Hispanic white applicants over the same tracts, county-cluster bootstrap interval | non-Hispanic Black applicants minus non-Hispanic white applicants, applicants whose home-purchase application was decided in 2024, in the tracts where the cells for non-Hispanic Black and non-Hispanic white applicants are both published | applications decided in 2024 | 95% CI 9.5 to 10.4 | 5,657 tracts | none (counts of decided applications, pooled over tracts) | HMDA mortgage lending |   |
| gap_black_ge10 | 10.0 | percentage points of denial rate | pooled denial rate of Black applicants minus that of white applicants over the same tracts, county-cluster bootstrap interval | non-Hispanic Black minus non-Hispanic white applicants decided in 2024, in tracts where both groups have at least 10 decided applications | applications decided in 2024 | 95% CI 9.5 to 10.6 | 3,848 tracts | none (counts of decided applications, pooled over tracts) | HMDA mortgage lending |   |
| gap_black_ge30 | 10.8 | percentage points of denial rate | pooled denial rate of Black applicants minus that of white applicants over the same tracts, county-cluster bootstrap interval | non-Hispanic Black minus non-Hispanic white applicants decided in 2024, in tracts where both groups have at least 30 decided applications | applications decided in 2024 | 95% CI 10.1 to 11.6 | 521 tracts | none (counts of decided applications, pooled over tracts) | HMDA mortgage lending |   |
| gap_black_own | 10.5 | percentage points of denial rate | pooled denial rate of non-Hispanic Black applicants minus that of non-Hispanic white applicants, each over its own tracts, county-cluster bootstrap interval | non-Hispanic Black applicants minus non-Hispanic white applicants whose home-purchase application was decided in 2024, each group over the tracts where its own cell is published | applications decided in 2024 | 95% CI 9.9 to 11.2 | 8,232 tracts | none (counts of decided applications, pooled over tracts) | HMDA mortgage lending |   |
| gap_d10_d1 | 6.0 | percentage points of denial rate | pooled denial rate of the most deprived tenth minus that of the comparison tenth, county-cluster bootstrap interval | home-purchase applications decided in 2024 in the most deprived tenth of tracts minus those in the least deprived tenth | applications decided in 2024 | 95% CI 5.2 to 6.8 | 3,027 counties | none (counts of decided applications, pooled over tracts) | HMDA mortgage lending |   |
| gap_d10_d3 | 6.2 | percentage points of denial rate | pooled denial rate of the most deprived tenth minus that of the comparison tenth, county-cluster bootstrap interval | home-purchase applications decided in 2024 in the most deprived tenth of tracts minus those in the third least deprived tenth | applications decided in 2024 | 95% CI 5.6 to 6.8 | 3,027 counties | none (counts of decided applications, pooled over tracts) | HMDA mortgage lending |   |
| gap_dec_max | 10.4 | percentage points of denial rate | largest over the ten tenths of the same-tract gap in pooled denial rates | non-Hispanic Black minus non-Hispanic white applicants decided in 2024, in the tracts of one tenth where both cells are published | applications decided in 2024 |   | 10 tenths of tracts | none (counts of decided applications, pooled over tracts) | HMDA mortgage lending |   |
| gap_dec_min | 8.6 | percentage points of denial rate | smallest over the ten tenths of the same-tract gap in pooled denial rates | non-Hispanic Black minus non-Hispanic white applicants decided in 2024, in the tracts of one tenth where both cells are published | applications decided in 2024 |   | 10 tenths of tracts | none (counts of decided applications, pooled over tracts) | HMDA mortgage lending |   |
| gap_hispanic | 6.4 | percentage points of denial rate | pooled denial rate of the group minus that of non-Hispanic white applicants over the same tracts, county-cluster bootstrap interval | Hispanic (any race) applicants minus non-Hispanic white applicants, applicants whose home-purchase application was decided in 2024, in the tracts where the cells for Hispanic (any race) and non-Hispanic white applicants are both published | applications decided in 2024 | 95% CI 5.9 to 6.9 | 10,189 tracts | none (counts of decided applications, pooled over tracts) | HMDA mortgage lending |   |
| gap_inc_max | 10.3 | percentage points of denial rate | largest over the five fifths of the same-tract gap in pooled denial rates | non-Hispanic Black minus non-Hispanic white applicants decided in 2024, in the tracts of one fifth by median applicant income where both cells are published | applications decided in 2024 |   | 5 fifths of tracts | none (counts of decided applications, pooled over tracts) | HMDA mortgage lending |   |
| gap_inc_min | 9.1 | percentage points of denial rate | smallest over the five fifths of the same-tract gap in pooled denial rates | non-Hispanic Black minus non-Hispanic white applicants decided in 2024, in the tracts of one fifth by median applicant income where both cells are published | applications decided in 2024 |   | 5 fifths of tracts | none (counts of decided applications, pooled over tracts) | HMDA mortgage lending |   |
| gap_slope_fifth | −0.06 | percentage points of denial rate per fifth | least-squares slope of the five same-tract gaps on the fifth number, recomputed in every county bootstrap draw | non-Hispanic Black minus non-Hispanic white applicants decided in 2024, in the tracts of each fifth by median applicant income where both cells are published | applications decided in 2024 | 95% CI −0.33 to 0.20 | 5 fifths of tracts | none (counts of decided applications, pooled over tracts) | HMDA mortgage lending |   |
| gap_slope_tenth | 0.00 | percentage points of denial rate per tenth | least-squares slope of the ten same-tract gaps on the tenth number, recomputed in every county bootstrap draw | non-Hispanic Black minus non-Hispanic white applicants decided in 2024, in the tracts of each tenth where both cells are published | applications decided in 2024 | 95% CI −0.22 to 0.21 | 10 tenths of tracts | none (counts of decided applications, pooled over tracts) | HMDA mortgage lending |   |
| hanson_credit_points | 71 | points of credit score | the authors' comparison of the effect of a Black-signaling name on the originator's response with the effect of a lower stated credit score (Hanson, Hawley, Martin and Liu 2016; email correspondence experiment); copied from the paper's abstract and not computed in this build | licensed mortgage loan originators who received email inquiries from prospective borrowers with names that signal a Black or a white sender | as reported in the paper |   |   | as in the paper | Published estimates from cited studies | A published estimate, not a result of this build. |
| holc_bh_a | 7.8% | percent of decided applications | decided applications in the published Black and Hispanic cells divided by decided applications of all applicants, summed over the tracts, county-cluster bootstrap interval | home-purchase applications decided in 2024 in tracts graded A, the share from published non-Hispanic Black and Hispanic cells (a lower bound, because withheld cells are not counted) | applications decided in 2024 | 95% CI 5.0% to 11.0% | 650 tracts | none (counts of decided applications, pooled over tracts) | HMDA mortgage lending |   |
| holc_bh_d | 15.0% | percent of decided applications | decided applications in the published Black and Hispanic cells divided by decided applications of all applicants, summed over the tracts, county-cluster bootstrap interval | home-purchase applications decided in 2024 in tracts graded D, the share from published non-Hispanic Black and Hispanic cells (a lower bound, because withheld cells are not counted) | applications decided in 2024 | 95% CI 11.2% to 18.9% | 2,087 tracts | none (counts of decided applications, pooled over tracts) | HMDA mortgage lending |   |
| holc_cover_min | 50% | percent of a tract's area | the share of a tract that graded polygons must cover for the tract to take a grade (a rule of this chapter, the same as the chapter on home value, income and redlining) | a tract matched to HOLC polygons | HOLC maps drawn 1935 to 1940 |   |   | none (definition) | Mapping Inequality (HOLC maps) |   |
| holc_cover_strict | 90% | percent of a tract's area | the stricter share of a tract that graded polygons must cover, used to show how the grade D minus grade A gap moves with the matching rule | a tract matched to HOLC polygons | HOLC maps drawn 1935 to 1940 |   |   | none (definition) | Mapping Inequality (HOLC maps) |   |
| holc_dgap_black | 0.1 | percentage points of denial rate | pooled denial rate in grade D tracts minus that in grade A tracts, county-cluster bootstrap interval | non-Hispanic Black applicants whose home-purchase application was decided in 2024, in tracts graded D minus those in tracts graded A, each over the tracts where the cell for the group is published | applications decided in 2024 | 95% CI −5.5 to 6.0 | 315 tracts | none (counts of decided applications, pooled over tracts) | HMDA mortgage lending |   |
| holc_dgap_hispanic | −1.3 | percentage points of denial rate | pooled denial rate in grade D tracts minus that in grade A tracts, county-cluster bootstrap interval | Hispanic (any race) applicants whose home-purchase application was decided in 2024, in tracts graded D minus those in tracts graded A, each over the tracts where the cell for the group is published | applications decided in 2024 | 95% CI −7.7 to 4.4 | 391 tracts | none (counts of decided applications, pooled over tracts) | HMDA mortgage lending |   |
| holc_dgap_wb | 1.3 | percentage points of denial rate | difference between the two grade D minus grade A pooled-rate gaps, county-cluster bootstrap interval | non-Hispanic white minus non-Hispanic Black applicants decided in 2024: the grade D minus grade A gap of white applicants minus that of Black applicants | applications decided in 2024 | 95% CI −5.0 to 7.4 | 315 tracts | none (counts of decided applications, pooled over tracts) | HMDA mortgage lending |   |
| holc_dgap_white | 1.4 | percentage points of denial rate | pooled denial rate in grade D tracts minus that in grade A tracts, county-cluster bootstrap interval | non-Hispanic white applicants whose home-purchase application was decided in 2024, in tracts graded D minus those in tracts graded A, each over the tracts where the cell for the group is published | applications decided in 2024 | 95% CI 0.2 to 2.8 | 1,365 tracts | none (counts of decided applications, pooled over tracts) | HMDA mortgage lending |   |
| holc_gap_b_adi | −0.2 | percentage points of denial rate | coefficient on grade B in a regression of the tract denial rate on grade with a separate baseline for each city and tenth cell, weighted by decided applications, errors clustered by county | home-purchase applications decided in 2024 in tracts graded B minus those in tracts graded A, among tracts of the same city and the same tenth of the Area Deprivation Index | applications decided in 2024 | 95% CI −0.9 to 0.5 | 8,748 tracts | decided applications | HMDA mortgage lending |   |
| holc_gap_c_adi | −0.4 | percentage points of denial rate | coefficient on grade C in a regression of the tract denial rate on grade with a separate baseline for each city and tenth cell, weighted by decided applications, errors clustered by county | home-purchase applications decided in 2024 in tracts graded C minus those in tracts graded A, among tracts of the same city and the same tenth of the Area Deprivation Index | applications decided in 2024 | 95% CI −1.3 to 0.6 | 8,748 tracts | decided applications | HMDA mortgage lending |   |
| holc_gap_d_adi | 0.0 | percentage points of denial rate | coefficient on grade D in a regression of the tract denial rate on grade (A is the base) with a separate baseline for each city and tenth cell, weighted by decided applications, errors clustered by county | home-purchase applications decided in 2024 in tracts graded D minus those in tracts graded A, among tracts of the same city and the same tenth of the Area Deprivation Index (a separate baseline for each city and tenth cell) | applications decided in 2024 | 95% CI −1.1 to 1.1 | 8,748 tracts | decided applications | HMDA mortgage lending |   |
| holc_gap_d_city | 3.5 | percentage points of denial rate | coefficient on grade D in a regression of the tract denial rate on grade (A is the base) with a separate baseline for each city, weighted by decided applications, errors clustered by county | home-purchase applications decided in 2024 in tracts graded D minus those in tracts graded A, among tracts of the same city | applications decided in 2024 | 95% CI 2.1 to 4.8 | 8,748 tracts | decided applications | HMDA mortgage lending |   |
| holc_gap_d_raw | 3.4 | percentage points of denial rate | coefficient on grade D in a regression of the tract denial rate on grade (A is the base), weighted by decided applications, errors clustered by county | home-purchase applications decided in 2024 in tracts graded D minus those in tracts graded A, with no adjustment | applications decided in 2024 | 95% CI 2.1 to 4.7 | 8,748 tracts | decided applications | HMDA mortgage lending |   |
| holc_gap_d_strict | 2.8 | percentage points of denial rate | pooled denial rate of grade D tracts minus that of grade A tracts, county-cluster bootstrap interval | home-purchase applications decided in 2024 in tracts graded D minus those in tracts graded A, among tracts at least 90 percent covered by graded polygons | applications decided in 2024 | 95% CI 1.0 to 4.6 | 4,736 tracts | none (counts of decided applications, pooled over tracts) | HMDA mortgage lending |   |
| holc_gap_dc | 0.4 | percentage points of denial rate | pooled denial rate of grade D tracts minus that of grade C tracts, county-cluster bootstrap interval | home-purchase applications decided in 2024 in tracts graded D minus those in tracts graded C | applications decided in 2024 | 95% CI −0.6 to 1.5 | 6,095 tracts | none (counts of decided applications, pooled over tracts) | HMDA mortgage lending |   |
| holc_min_group_tracts | 30 | tracts | the number of tracts a grade and group point needs to be drawn (a rule of this chapter) | a grade and applicant group point in the HOLC comparison | 2024 |   |   | none (definition) | HMDA mortgage lending |   |
| holc_n_cells | 1,008 | city and tenth cells | count of distinct (city, tenth of the deprivation index) cells that carry a baseline in the third adjustment | graded tracts with a published home-purchase cell and a deprivation rank | 2024 |   |   | none (full count) | HMDA mortgage lending |   |
| holc_n_cities | 221 | city and state pairs | count of distinct (city, state) pairs of the polygon covering most of each tract | graded tracts with a published home-purchase cell | HOLC maps drawn 1935 to 1940 |   |   | none (full count) | Mapping Inequality (HOLC maps) |   |
| holc_n_graded | 10,967 | tracts | count of tracts whose area is at least half covered by polygons graded A to D | 2020 census tracts covered at least half by graded HOLC polygons | HOLC maps drawn 1935 to 1940 |   |   | none (full count) | Mapping Inequality (HOLC maps) |   |
| holc_n_published | 8,748 | tracts | count of graded tracts in the HMDA analysis | graded tracts of the 50 states and DC with a published home-purchase cell and a deprivation rank | 2024 |   |   | none (full count) | HMDA mortgage lending |   |
| holc_rate_a | 7.9% | percent of decided applications denied | denied applications divided by applications that were originated or denied, summed over the tracts (a pooled rate), county-cluster bootstrap interval | home-purchase applications decided in 2024 in tracts graded A on the 1930s HOLC maps | applications decided in 2024 | 95% CI 7.0% to 9.0% | 650 tracts | none (counts of decided applications, pooled over tracts) | HMDA mortgage lending |   |
| holc_rate_b | 9.4% | percent of decided applications denied | denied applications divided by applications that were originated or denied, summed over the tracts (a pooled rate), county-cluster bootstrap interval | home-purchase applications decided in 2024 in tracts graded B on the 1930s HOLC maps | applications decided in 2024 | 95% CI 8.5% to 10.3% | 2,003 tracts | none (counts of decided applications, pooled over tracts) | HMDA mortgage lending |   |
| holc_rate_c | 10.9% | percent of decided applications denied | denied applications divided by applications that were originated or denied, summed over the tracts (a pooled rate), county-cluster bootstrap interval | home-purchase applications decided in 2024 in tracts graded C on the 1930s HOLC maps | applications decided in 2024 | 95% CI 10.0% to 11.8% | 4,008 tracts | none (counts of decided applications, pooled over tracts) | HMDA mortgage lending |   |
| holc_rate_d | 11.3% | percent of decided applications denied | denied applications divided by applications that were originated or denied, summed over the tracts (a pooled rate), county-cluster bootstrap interval | home-purchase applications decided in 2024 in tracts graded D on the 1930s HOLC maps | applications decided in 2024 | 95% CI 10.2% to 12.5% | 2,087 tracts | none (counts of decided applications, pooled over tracts) | HMDA mortgage lending |   |
| holc_rate_graded | 10.3% | percent of decided applications denied | denied applications divided by applications that were originated or denied, summed over the tracts (a pooled rate), county-cluster bootstrap interval | home-purchase applications decided in 2024 in graded tracts | applications decided in 2024 | 95% CI 9.5% to 11.1% | 8,748 tracts | none (counts of decided applications, pooled over tracts) | HMDA mortgage lending |   |
| holc_rate_other | 8.9% | percent of decided applications denied | denied applications divided by applications that were originated or denied, summed over the tracts (a pooled rate), county-cluster bootstrap interval | home-purchase applications decided in 2024 in tracts that are not graded | applications decided in 2024 | 95% CI 8.6% to 9.2% | 67,230 tracts | none (counts of decided applications, pooled over tracts) | HMDA mortgage lending |   |
| holc_share_decided | 7.3% | percent of decided applications | decided applications in graded tracts divided by decided applications in all tracts of the analysis | home-purchase applications that a covered lender originated or denied in 2024, first-lien loans on site-built homes with up to four units that applicants intend as a principal residence, in tracts of the 50 states and DC with a published cell | applications decided in 2024 |   |   | none (full counts) | HMDA mortgage lending |   |
| income_d1 | $279 thousand | thousands of dollars, median applicant income in the tract | mean over tracts of the tract's median income reported on the applications, weighted by decided applications, county-cluster bootstrap interval | tracts in the least deprived tenth with a published home-purchase cell | applications in 2024 | 95% CI $266 thousand to $292 thousand | 7,584 tracts | decided applications | HMDA mortgage lending |   |
| income_d10 | $64 thousand | thousands of dollars, median applicant income in the tract | mean over tracts of the tract's median income reported on the applications, weighted by decided applications, county-cluster bootstrap interval | tracts in the most deprived tenth with a published home-purchase cell | applications in 2024 | 95% CI $63 thousand to $65 thousand | 7,591 tracts | decided applications | HMDA mortgage lending |   |
| income_fifth_hi | $125 thousand | thousands of dollars, median applicant income in the tract | lowest tract median applicant income in the highest fifth | tracts in the highest fifth by median applicant income among tracts where both Black and white cells are published | applications in 2024 |   | 1,135 tracts | none (a boundary of the fifth) | HMDA mortgage lending |   |
| income_fifth_lo | $74 thousand | thousands of dollars, median applicant income in the tract | highest tract median applicant income in the lowest fifth | tracts in the lowest fifth by median applicant income among tracts where both Black and white cells are published | applications in 2024 |   | 1,124 tracts | none (a boundary of the fifth) | HMDA mortgage lending |   |
| loan_d1 | $827 thousand | thousands of dollars, median loan amount in the tract | mean over tracts of the tract's median loan amount (HMDA rounds to the nearest $10,000), weighted by decided applications, county-cluster bootstrap interval | tracts in the least deprived tenth with a published home-purchase cell | applications in 2024 | 95% CI $774 thousand to $879 thousand | 7,584 tracts | decided applications | HMDA mortgage lending |   |
| loan_d10 | $162 thousand | thousands of dollars, median loan amount in the tract | mean over tracts of the tract's median loan amount (HMDA rounds to the nearest $10,000), weighted by decided applications, county-cluster bootstrap interval | tracts in the most deprived tenth with a published home-purchase cell | applications in 2024 | 95% CI $158 thousand to $167 thousand | 7,591 tracts | decided applications | HMDA mortgage lending |   |
| median_decided | 31 | decided applications per tract | median over tracts of the originated plus denied applications | home-purchase applications that a covered lender originated or denied in 2024, first-lien loans on site-built homes with up to four units that applicants intend as a principal residence, in tracts of the 50 states and DC with a published cell | applications decided in 2024 |   | 76,072 tracts | none (each tract counts once) | HMDA mortgage lending |   |
| min_point_tracts | 100 | tracts | the number of tracts a point needs to be drawn (a rule of this chapter) | a point on the lines by applicant group | 2024 |   |   | none (definition) | HMDA mortgage lending |   |
| n_cloud | 2,000 | tracts | size of a random sample drawn with a fixed seed | tracts with at least 30 decided home-purchase applications | 2024 |   |   | none (sample size) | HMDA mortgage lending |   |
| n_cloud_pool | 39,697 | tracts | count of tracts the sample is drawn from | tracts with a deprivation rank and at least 30 decided home-purchase applications | 2024 |   |   | none (full count) | HMDA mortgage lending |   |
| n_decided | 3,097,390 | applications | originated plus denied applications summed over the tracts | home-purchase applications that a covered lender originated or denied in 2024, first-lien loans on site-built homes with up to four units that applicants intend as a principal residence, in tracts of the 50 states and DC with a published cell | applications decided in 2024 |   |   | none (full count) | HMDA mortgage lending |   |
| n_denied | 278,451 | applications | denied applications summed over the tracts | home-purchase applications that a covered lender originated or denied in 2024, first-lien loans on site-built homes with up to four units that applicants intend as a principal residence, in tracts of the 50 states and DC with a published cell | applications decided in 2024 |   |   | none (full count) | HMDA mortgage lending |   |
| n_lt10 | 4,865 | tracts | count of tracts with fewer than 10 decided applications (the staged floor counts applications of every action taken) | tracts of the 50 states and DC with a published home-purchase cell | applications decided in 2024 |   |   | none (full count) | HMDA mortgage lending |   |
| n_no_adi | 94 | tracts | count of tracts with no deprivation rank, left out of the figures that rank tracts | tracts with a published home-purchase cell | 2024 |   |   | none (full count) | Area Deprivation Index |   |
| n_pr_published | 637 | tracts | count of Puerto Rico tracts set aside | Puerto Rico tracts with a published home-purchase cell for all applicants | 2024 |   |   | none (full count) | HMDA mortgage lending |   |
| n_shared_asian | 5,490 | tracts | count of tracts where the cells for non-Hispanic Asian and non-Hispanic white applicants are both published | tracts of the 50 states and DC | 2024 |   |   | none (full count) | HMDA mortgage lending |   |
| n_shared_black | 5,657 | tracts | count of tracts where the cells for non-Hispanic Black and non-Hispanic white applicants are both published | tracts of the 50 states and DC | 2024 |   |   | none (full count) | HMDA mortgage lending |   |
| n_shared_hispanic | 10,189 | tracts | count of tracts where the cells for Hispanic (any race) and non-Hispanic white applicants are both published | tracts of the 50 states and DC | 2024 |   |   | none (full count) | HMDA mortgage lending |   |
| n_tracts | 76,072 | tracts | count of tracts whose all-applicant home-purchase cell has 10 or more applications and a decided application | tracts of the 50 states and DC with a published home-purchase cell for all applicants | 2024 |   |   | none (full count) | HMDA mortgage lending |   |
| n_tracts_adi | 75,978 | tracts | count of tracts ranked into tenths | tracts with a published home-purchase cell and an Area Deprivation Index rank | 2024 |   |   | none (full count) | HMDA mortgage lending |   |
| n_withheld | 5,999 | tracts | count of tracts whose all-applicant cell has fewer than 10 applications and is set to missing when the file is staged | tracts of the 50 states and DC with a home-purchase row in the HMDA file | 2024 |   |   | none (full count) | HMDA mortgage lending |   |
| pr_decided | 11,409 | applications | originated plus denied applications summed over the tracts | home-purchase applications decided in 2024 in Puerto Rico tracts with a published cell | applications decided in 2024 |   |   | none (full count) | HMDA mortgage lending |   |
| pr_rate | 19.1% | percent of decided applications denied | denied applications divided by applications that were originated or denied, summed over the tracts (a pooled rate), municipio-cluster bootstrap interval | home-purchase applications decided in 2024 in Puerto Rico tracts with a published cell, first-lien loans on site-built homes with up to four units that applicants intend as a principal residence | applications decided in 2024 | 95% CI 18.3% to 20.0% | 637 tracts | none (counts of decided applications, pooled over tracts) | HMDA mortgage lending |   |
| pr_shift_all | 0.04 | percentage points of denial rate | pooled denial rate with Puerto Rico tracts added minus the rate without them | home-purchase applications decided in 2024, all tracts with a published cell, with Puerto Rico minus without | applications decided in 2024 |   |   | none (counts of decided applications, pooled over tracts) | HMDA mortgage lending |   |
| pr_top_share | 69% | percent of ranked Puerto Rico tracts | share at or above the national rank where the most deprived tenth of the 50-state and DC sample starts | Puerto Rico tracts with a published cell and a deprivation rank | ADI 2023 |   |   | none (each tract counts once) | Area Deprivation Index |   |
| rate_all | 9.0% | percent of decided applications denied | denied applications divided by applications that were originated or denied, summed over the tracts (a pooled rate), county-cluster bootstrap interval | home-purchase applications that a covered lender originated or denied in 2024, first-lien loans on site-built homes with up to four units that applicants intend as a principal residence, in tracts of the 50 states and DC with a published cell | applications decided in 2024 | 95% CI 8.7% to 9.3% | 76,072 tracts | none (counts of decided applications, pooled over tracts) | HMDA mortgage lending |   |
| rate_all_ge30 | 8.5% | percent of decided applications denied | denied applications divided by applications that were originated or denied, summed over the tracts (a pooled rate), county-cluster bootstrap interval | home-purchase applications that a covered lender originated or denied in 2024, first-lien loans on site-built homes with up to four units that applicants intend as a principal residence, in tracts of the 50 states and DC with a published cell, tracts with at least 30 decided applications | applications decided in 2024 | 95% CI 8.2% to 8.8% | 39,714 tracts | none (counts of decided applications, pooled over tracts) | HMDA mortgage lending |   |
| rate_asian | 8.9% | percent of decided applications denied | denied applications divided by applications that were originated or denied, summed over the tracts (a pooled rate), county-cluster bootstrap interval | non-Hispanic Asian applicants whose home-purchase application was decided in 2024, in tracts where the cell for the group is published (10 or more applications) | applications decided in 2024 | 95% CI 8.4% to 9.4% | 7,532 tracts | none (counts of decided applications, pooled over tracts) | HMDA mortgage lending |   |
| rate_asian_d1 | 7.8% | percent of decided applications denied | denied applications divided by applications that were originated or denied, summed over the tracts (a pooled rate), county-cluster bootstrap interval | non-Hispanic Asian applicants whose home-purchase application was decided in 2024, in the tract tenth numbered 1 by Area Deprivation Index (1 = least deprived) where the cell is published | applications decided in 2024 | 95% CI 7.3% to 8.7% | 2,241 tracts | none (counts of decided applications, pooled over tracts) | HMDA mortgage lending |   |
| rate_asian_d8 | 12.9% | percent of decided applications denied | denied applications divided by applications that were originated or denied, summed over the tracts (a pooled rate), county-cluster bootstrap interval | non-Hispanic Asian applicants whose home-purchase application was decided in 2024, in the tract tenth numbered 8 by Area Deprivation Index (1 = least deprived) where the cell is published | applications decided in 2024 | 95% CI 11.0% to 14.5% | 103 tracts | none (counts of decided applications, pooled over tracts) | HMDA mortgage lending |   |
| rate_black | 17.0% | percent of decided applications denied | denied applications divided by applications that were originated or denied, summed over the tracts (a pooled rate), county-cluster bootstrap interval | non-Hispanic Black applicants whose home-purchase application was decided in 2024, in tracts where the cell for the group is published (10 or more applications) | applications decided in 2024 | 95% CI 16.4% to 17.7% | 8,232 tracts | none (counts of decided applications, pooled over tracts) | HMDA mortgage lending |   |
| rate_black_d10 | 20.3% | percent of decided applications denied | denied applications divided by applications that were originated or denied, summed over the tracts (a pooled rate), county-cluster bootstrap interval | non-Hispanic Black applicants whose home-purchase application was decided in 2024, in the tract tenth numbered 10 by Area Deprivation Index (1 = least deprived) where the cell is published | applications decided in 2024 | 95% CI 19.0% to 21.6% | 1,081 tracts | none (counts of decided applications, pooled over tracts) | HMDA mortgage lending |   |
| rate_black_d3 | 14.6% | percent of decided applications denied | denied applications divided by applications that were originated or denied, summed over the tracts (a pooled rate), county-cluster bootstrap interval | non-Hispanic Black applicants whose home-purchase application was decided in 2024, in the tract tenth numbered 3 by Area Deprivation Index (1 = least deprived) where the cell is published | applications decided in 2024 | 95% CI 13.8% to 15.6% | 633 tracts | none (counts of decided applications, pooled over tracts) | HMDA mortgage lending |   |
| rate_cashout | 35.8% | percent of decided applications denied | denied applications divided by applications that were originated or denied, summed over the tracts (a pooled rate), county-cluster bootstrap interval | cash-out refinance applications that a covered lender originated or denied in 2024, first-lien loans on site-built homes with up to four units that applicants intend as a principal residence, in tracts of the 50 states and DC with a published cell | applications decided in 2024 | 95% CI 35.3% to 36.4% | 46,525 tracts | none (counts of decided applications, pooled over tracts) | HMDA mortgage lending |   |
| rate_d1 | 7.8% | percent of decided applications denied | denied applications divided by applications that were originated or denied, summed over the tracts (a pooled rate), county-cluster bootstrap interval | home-purchase applications decided in 2024 in the least deprived tenth of tracts by Area Deprivation Index national rank (tenths hold equal numbers of tracts) | applications decided in 2024 | 95% CI 7.2% to 8.4% | 7,584 tracts | none (counts of decided applications, pooled over tracts) | HMDA mortgage lending |   |
| rate_d10 | 13.8% | percent of decided applications denied | denied applications divided by applications that were originated or denied, summed over the tracts (a pooled rate), county-cluster bootstrap interval | home-purchase applications decided in 2024 in the most deprived tenth of tracts by Area Deprivation Index national rank (tenths hold equal numbers of tracts) | applications decided in 2024 | 95% CI 13.3% to 14.3% | 7,591 tracts | none (counts of decided applications, pooled over tracts) | HMDA mortgage lending |   |
| rate_d10_with_pr | 14.1% | percent of decided applications denied | denied applications divided by originated or denied applications, summed over those tracts (a sensitivity, not a chapter estimate) | home-purchase applications decided in 2024 in the most deprived tenth of the 50-state and DC tracts plus Puerto Rico tracts at or above the rank where that tenth starts | applications decided in 2024 |   |   | none (counts of decided applications, pooled over tracts) | HMDA mortgage lending |   |
| rate_d2 | 7.7% | percent of decided applications denied | denied applications divided by applications that were originated or denied, summed over the tracts (a pooled rate), county-cluster bootstrap interval | home-purchase applications decided in 2024 in the second least deprived tenth of tracts by Area Deprivation Index national rank (tenths hold equal numbers of tracts) | applications decided in 2024 | 95% CI 7.2% to 8.2% | 7,686 tracts | none (counts of decided applications, pooled over tracts) | HMDA mortgage lending |   |
| rate_d3 | 7.6% | percent of decided applications denied | denied applications divided by applications that were originated or denied, summed over the tracts (a pooled rate), county-cluster bootstrap interval | home-purchase applications decided in 2024 in the third least deprived tenth of tracts by Area Deprivation Index national rank (tenths hold equal numbers of tracts) | applications decided in 2024 | 95% CI 7.3% to 8.0% | 7,429 tracts | none (counts of decided applications, pooled over tracts) | HMDA mortgage lending |   |
| rate_hispanic | 13.1% | percent of decided applications denied | denied applications divided by applications that were originated or denied, summed over the tracts (a pooled rate), county-cluster bootstrap interval | Hispanic (any race) applicants whose home-purchase application was decided in 2024, in tracts where the cell for the group is published (10 or more applications) | applications decided in 2024 | 95% CI 12.4% to 13.9% | 16,060 tracts | none (counts of decided applications, pooled over tracts) | HMDA mortgage lending |   |
| rate_hispanic_d10 | 14.2% | percent of decided applications denied | denied applications divided by applications that were originated or denied, summed over the tracts (a pooled rate), county-cluster bootstrap interval | Hispanic (any race) applicants whose home-purchase application was decided in 2024, in the tract tenth numbered 10 by Area Deprivation Index (1 = least deprived) where the cell is published | applications decided in 2024 | 95% CI 12.9% to 15.7% | 1,045 tracts | none (counts of decided applications, pooled over tracts) | HMDA mortgage lending |   |
| rate_hispanic_d3 | 11.8% | percent of decided applications denied | denied applications divided by applications that were originated or denied, summed over the tracts (a pooled rate), county-cluster bootstrap interval | Hispanic (any race) applicants whose home-purchase application was decided in 2024, in the tract tenth numbered 3 by Area Deprivation Index (1 = least deprived) where the cell is published | applications decided in 2024 | 95% CI 11.1% to 12.6% | 1,847 tracts | none (counts of decided applications, pooled over tracts) | HMDA mortgage lending |   |
| rate_refi | 20.7% | percent of decided applications denied | denied applications divided by applications that were originated or denied, summed over the tracts (a pooled rate), county-cluster bootstrap interval | refinance applications that a covered lender originated or denied in 2024, first-lien loans on site-built homes with up to four units that applicants intend as a principal residence, in tracts of the 50 states and DC with a published cell | applications decided in 2024 | 95% CI 20.2% to 21.2% | 36,779 tracts | none (counts of decided applications, pooled over tracts) | HMDA mortgage lending |   |
| rate_white | 6.5% | percent of decided applications denied | denied applications divided by applications that were originated or denied, summed over the tracts (a pooled rate), county-cluster bootstrap interval | non-Hispanic white applicants whose home-purchase application was decided in 2024, in tracts where the cell for the group is published (10 or more applications) | applications decided in 2024 | 95% CI 6.3% to 6.6% | 56,664 tracts | none (counts of decided applications, pooled over tracts) | HMDA mortgage lending |   |
| rate_white_d1 | 6.7% | percent of decided applications denied | denied applications divided by applications that were originated or denied, summed over the tracts (a pooled rate), county-cluster bootstrap interval | non-Hispanic white applicants whose home-purchase application was decided in 2024, in the tract tenth numbered 1 by Area Deprivation Index (1 = least deprived) where the cell is published | applications decided in 2024 | 95% CI 6.0% to 7.5% | 4,805 tracts | none (counts of decided applications, pooled over tracts) | HMDA mortgage lending |   |
| rate_white_d10 | 10.4% | percent of decided applications denied | denied applications divided by applications that were originated or denied, summed over the tracts (a pooled rate), county-cluster bootstrap interval | non-Hispanic white applicants whose home-purchase application was decided in 2024, in the tract tenth numbered 10 by Area Deprivation Index (1 = least deprived) where the cell is published | applications decided in 2024 | 95% CI 10.1% to 10.8% | 4,227 tracts | none (counts of decided applications, pooled over tracts) | HMDA mortgage lending |   |
| rate_white_d3 | 5.4% | percent of decided applications denied | denied applications divided by applications that were originated or denied, summed over the tracts (a pooled rate), county-cluster bootstrap interval | non-Hispanic white applicants whose home-purchase application was decided in 2024, in the tract tenth numbered 3 by Area Deprivation Index (1 = least deprived) where the cell is published | applications decided in 2024 | 95% CI 5.2% to 5.6% | 5,868 tracts | none (counts of decided applications, pooled over tracts) | HMDA mortgage lending |   |
| ratio_asian | 1.6 | ratio of denial rates | pooled denial rate of the group divided by that of non-Hispanic white applicants over the same tracts, county-cluster bootstrap interval | non-Hispanic Asian applicants over non-Hispanic white applicants, applicants whose home-purchase application was decided in 2024, in the tracts where the cells for non-Hispanic Asian and non-Hispanic white applicants are both published | applications decided in 2024 | 95% CI 1.5 to 1.7 | 5,490 tracts | none (counts of decided applications, pooled over tracts) | HMDA mortgage lending |   |
| ratio_black | 2.6 | ratio of denial rates | pooled denial rate of the group divided by that of non-Hispanic white applicants over the same tracts, county-cluster bootstrap interval | non-Hispanic Black applicants over non-Hispanic white applicants, applicants whose home-purchase application was decided in 2024, in the tracts where the cells for non-Hispanic Black and non-Hispanic white applicants are both published | applications decided in 2024 | 95% CI 2.5 to 2.7 | 5,657 tracts | none (counts of decided applications, pooled over tracts) | HMDA mortgage lending |   |
| ratio_d10 | 2.0 | ratio of denial rates | pooled denial rate of Black applicants divided by that of white applicants over the same tracts, county-cluster bootstrap interval | non-Hispanic Black over non-Hispanic white applicants decided in 2024, in the most deprived tenth of tracts where both cells are published | applications decided in 2024 | 95% CI 1.8 to 2.3 | 395 tracts | none (counts of decided applications, pooled over tracts) | HMDA mortgage lending |   |
| ratio_d5 | 2.8 | ratio of denial rates | pooled denial rate of Black applicants divided by that of white applicants over the same tracts, county-cluster bootstrap interval | non-Hispanic Black over non-Hispanic white applicants decided in 2024, in the fifth tenth of tracts by Area Deprivation Index where both cells are published | applications decided in 2024 | 95% CI 2.6 to 3.0 | 809 tracts | none (counts of decided applications, pooled over tracts) | HMDA mortgage lending |   |
| ratio_hispanic | 2.0 | ratio of denial rates | pooled denial rate of the group divided by that of non-Hispanic white applicants over the same tracts, county-cluster bootstrap interval | Hispanic (any race) applicants over non-Hispanic white applicants, applicants whose home-purchase application was decided in 2024, in the tracts where the cells for Hispanic (any race) and non-Hispanic white applicants are both published | applications decided in 2024 | 95% CI 1.9 to 2.1 | 10,189 tracts | none (counts of decided applications, pooled over tracts) | HMDA mortgage lending |   |
| same_asian | 8.9% | percent of decided applications denied | denied applications divided by applications that were originated or denied, summed over the tracts (a pooled rate), county-cluster bootstrap interval | non-Hispanic Asian applicants whose home-purchase application was decided in 2024, in the tracts where the cells for non-Hispanic Asian and non-Hispanic white applicants are both published | applications decided in 2024 | 95% CI 8.5% to 9.4% | 5,490 tracts | none (counts of decided applications, pooled over tracts) | HMDA mortgage lending |   |
| same_black | 16.2% | percent of decided applications denied | denied applications divided by applications that were originated or denied, summed over the tracts (a pooled rate), county-cluster bootstrap interval | non-Hispanic Black applicants whose home-purchase application was decided in 2024, in the tracts where the cells for non-Hispanic Black and non-Hispanic white applicants are both published | applications decided in 2024 | 95% CI 15.7% to 16.8% | 5,657 tracts | none (counts of decided applications, pooled over tracts) | HMDA mortgage lending |   |
| same_hispanic | 12.8% | percent of decided applications denied | denied applications divided by applications that were originated or denied, summed over the tracts (a pooled rate), county-cluster bootstrap interval | Hispanic (any race) applicants whose home-purchase application was decided in 2024, in the tracts where the cells for Hispanic (any race) and non-Hispanic white applicants are both published | applications decided in 2024 | 95% CI 12.2% to 13.5% | 10,189 tracts | none (counts of decided applications, pooled over tracts) | HMDA mortgage lending |   |
| same_white_vs_asian | 5.7% | percent of decided applications denied | denied applications divided by applications that were originated or denied, summed over the tracts (a pooled rate), county-cluster bootstrap interval | non-Hispanic white applicants whose home-purchase application was decided in 2024, in the tracts where the cells for non-Hispanic Asian and non-Hispanic white applicants are both published | applications decided in 2024 | 95% CI 5.4% to 6.0% | 5,490 tracts | none (counts of decided applications, pooled over tracts) | HMDA mortgage lending |   |
| same_white_vs_black | 6.2% | percent of decided applications denied | denied applications divided by applications that were originated or denied, summed over the tracts (a pooled rate), county-cluster bootstrap interval | non-Hispanic white applicants whose home-purchase application was decided in 2024, in the tracts where the cells for non-Hispanic Black and non-Hispanic white applicants are both published | applications decided in 2024 | 95% CI 6.0% to 6.5% | 5,657 tracts | none (counts of decided applications, pooled over tracts) | HMDA mortgage lending |   |
| same_white_vs_hispanic | 6.5% | percent of decided applications denied | denied applications divided by applications that were originated or denied, summed over the tracts (a pooled rate), county-cluster bootstrap interval | non-Hispanic white applicants whose home-purchase application was decided in 2024, in the tracts where the cells for Hispanic (any race) and non-Hispanic white applicants are both published | applications decided in 2024 | 95% CI 6.2% to 6.8% | 10,189 tracts | none (counts of decided applications, pooled over tracts) | HMDA mortgage lending |   |
| share_black_lt10 | 27% | percent of tracts that hold both cells | share of those tracts with fewer than 10 decided applications from non-Hispanic Black applicants | tracts where the cells for non-Hispanic Black and non-Hispanic white applicants are both published | applications decided in 2024 |   | 5,657 tracts | none (each tract counts once) | HMDA mortgage lending |   |
| share_shared_black | 7.4% | percent of tracts | tracts where the cells for non-Hispanic Black and non-Hispanic white applicants are both published, divided by tracts with a published all-applicant cell | home-purchase applications that a covered lender originated or denied in 2024, first-lien loans on site-built homes with up to four units that applicants intend as a principal residence, in tracts of the 50 states and DC with a published cell | applications decided in 2024 |   | 76,072 tracts | none (each tract counts once) | HMDA mortgage lending |   |
| suppress_floor | 10 | applications | the number of applications, of every action taken, below which the staging step withholds the cell (a rule of this book, not an HMDA flag) | a tract and loan purpose cell of the staged HMDA file | 2024 |   |   | none (definition) | HMDA mortgage lending |   |
| suppress_max | 9 | applications | the largest number of applications a withheld cell can hold (one below the floor) | a withheld tract and loan purpose cell of the staged HMDA file | 2024 |   |   | none (definition) | HMDA mortgage lending |   |
| tracts_black_cashout | 2,938 | tracts | count of tracts with a published cash-out refinance cell for non-Hispanic Black applicants (10 or more applications) | tracts of the 50 states and DC | 2024 |   |   | none (full count) | HMDA mortgage lending |   |
| tracts_black_purchase | 8,232 | tracts | count of tracts with a published home-purchase cell for non-Hispanic Black applicants (10 or more applications) | tracts of the 50 states and DC | 2024 |   |   | none (full count) | HMDA mortgage lending |   |
| tracts_black_refi | 1,155 | tracts | count of tracts with a published refinance cell for non-Hispanic Black applicants (10 or more applications) | tracts of the 50 states and DC | 2024 |   |   | none (full count) | HMDA mortgage lending |   |
| us_denial | 9.0% | percent of decided applications denied | denied applications divided by applications that were originated or denied, summed over the tracts (a pooled rate), county-cluster bootstrap interval | home-purchase applications decided in 2024 in all tracts with an Area Deprivation Index national rank (all ten tenths pooled) | applications decided in 2024 | 95% CI 8.7% to 9.3% | 75,978 tracts | none (counts of decided applications, pooled over tracts) | HMDA mortgage lending |   |

## Definitions

- Census tract: A small area drawn by the Census Bureau to stand in for a neighborhood. Boundaries are redrawn after each decennial census.
- Bootstrap interval: A range for an estimate found by recomputing it on many random resamples of the data and reading the range from the results. It shows how much the estimate would vary from sample to sample.
- Suppression: The withholding of a number because the count behind it is too small to publish without risking the identification of a person. A suppressed value is missing, not zero.
- Area Deprivation Index (ADI): A score that ranks neighborhoods by income, education, employment and housing conditions drawn from the American Community Survey. In this book a higher national percentile means a more deprived neighborhood.
- HOLC grade: A letter from A to D that the Home Owners' Loan Corporation gave neighborhoods on its 1930s maps, with A labeled best and D hazardous. The maps are known as redlining maps.
- Ecological inference: Drawing conclusions about individuals from data about places or groups. A pattern across counties need not hold for the people inside them.

## Methods

This chapter compares denial rates on home-purchase applications decided in 2024 across census tracts, for all applicants and for four applicant groups. The file covers lenders above the HMDA reporting thresholds and has no loan type; withdrawn and incomplete applications are not in the rate, and a tract, purpose and group cell with fewer than 10 applications is missing, so a comparison between groups covers only tracts where both cells are published. The HOLC grades come from maps drawn in the 1930s and are compared with 2024 lending, and the Area Deprivation Index that ranks tracts is built partly from income and housing costs.

### Which applications are included

The staged HMDA file holds 2024 first-lien applications on site-built homes with up to four units that applicants intend as a principal residence, from lenders above the HMDA reporting thresholds, aggregated to 2020 census tracts by loan purpose (home purchase, refinance, cash-out refinance) and by applicant group (all, and the applicant's reported non-Hispanic white, non-Hispanic Black, Hispanic of any race and non-Hispanic Asian).

The chapter uses home-purchase applications in the 50 states and DC. Puerto Rico is left out. Loan type is not in the file.

Explains Table 1.

### How the denial rate is defined

The denial rate is denied applications divided by applications that were originated or denied (decided), as a percent, summed over tracts: a pooled rate, never an average of tract rates.

Applications that were withdrawn or left incomplete, loans approved but not accepted and purchased loans are not in the rate.

### Cells with few applications are missing

When the file was staged, a tract, purpose and group cell with fewer than 10 applications (of every action taken) was set to missing; a missing cell leaves both the numerator and the denominator and is never imputed or back-calculated.

### How tracts are grouped

Tenths hold near-equal numbers of tracts ranked by the Area Deprivation Index national rank (tracts tied on the rank stay together). The fifths of tract applicant income hold near-equal numbers of tracts that have published cells for Black and white applicants.

Explains Table 2 and Table 5.

### HOLC grades and the Cook County map

Graded polygons (A to D) are intersected with 2020 tracts in an equal-area projection; a tract with at least 50 percent of its area graded takes the grade covering most of it and the city of the polygon covering most of it (the rule of the chapter on home value, income and redlining); the grade D minus grade A gap is also shown for a stricter rule of 90 percent coverage.

Grade comparisons are weighted regressions of the tract denial rate on grade, weighted by decided applications, with errors clustered by county, with no baseline, with a separate baseline for each city, and with a separate baseline for each city and tenth cell.

On the Cook County map, tracts with at least 20 decided applications are shaded; the tooltip interval is a Wilson interval on the counts.

Explains Table 6 and Table 7.

### Intervals, trends and the tract sample

Intervals are percentile intervals from a bootstrap that resamples whole counties with replacement (2,000 replicates, seed 12345, the same draws for every estimate so a difference gets one interval).

The trend of the same-tract gap is the least-squares slope of the gaps of the tenths (or fifths) on their number, recomputed in every bootstrap draw.

The scatter shows a random sample of 2,000 tracts with at least 30 decided applications, drawn with seed 12345.

Explains Table 4 and Table 2.

### Four rankings of the same places

The comparison follows the book's ranking switch rule, fixed on October 28, 2026 before any of these gradients was computed. The main gradient of the chapter is recomputed under four rankings of the same places, with the outcome, period, weights, bins and intervals unchanged: the published ADI; a standardized index that applies the ADI's coefficients to its measures after standardizing each one across tracts; median household income; and the share of adults 25 and older with a bachelor's degree or more. The three alternatives come from ACS tract estimates for 2019 to 2023, the period of the ADI.

All four rankings use the same places, the 71,169 tracts that have the outcome and all four rankings, so the ADI panel can differ slightly from the chapter's main figure. Tracts enter only if they have every input of the standardized index, which leaves out tracts without a median rent, home value or mortgage cost.

A pattern counts as changed when the worse end switches sides, when the two ends differ under one ranking and not the other, or when the rank correlation of bin order and bin mean falls on the other side of 0.9 and moves by at least 0.1.

Explains Table 8.

## What these data cannot show

Applications are not the population. A denial rate is a share of the applications that reached a decision, and who applies differs across tracts and groups. People who expect a denial, who are steered to another loan or who cannot get an answer to an inquiry do not appear, and withdrawn applications leave the rate. The file covers one year and the lenders above the reporting thresholds.

The tract totals hold no credit score, debt-to-income ratio, loan-to-value ratio, loan type or lender for any applicant. A gap between groups, even inside one tract, is therefore not evidence of discrimination by itself and not a measure of how much treatment differs. The application-level HMDA records carry some of those fields and up to four denial reasons for each denied application. Lenders report those reasons themselves, they are not audited, and the tract aggregate does not keep them.

The groups are reported groups. The ethnicity and race in HMDA come from the applicant where the applicant answers and from fallback rules where not, and the four groups leave out other and unreported applicants. A group's cell exists only in tracts with 10 or more of its applications, so group comparisons cover the tracts with more applicants of that group and not a random sample of tracts.

The intervals resample counties. Applications at one lender share its rules and cross county lines, so the intervals do not capture correlation within a lender and may be narrower than the uncertainty. The tenths and grades describe tracts and the applications reported in them. They are not the experience of one borrower, and a pattern across places need not hold for the people inside them (ecological inference). The deprivation index understates deprivation in high-cost cities and is built partly from income and housing costs, which are also what a lender weighs.

The grades followed the housing, incomes and populations of the 1930s, so a tract's grade stands for a mix of what the neighborhood was and what happened after. The adjusted contrast compares tracts of the same city and the same tenth of the index and cannot say whether the maps shaped today's deprivation. A city that HOLC did not survey is absent from the comparison, which does not mean it had no redlining. Tooltip intervals on the Cook County map treat applications as independent and are narrower than a bootstrap over tracts would give.

## References

1. Munnell, Alicia H.; Tootell, Geoffrey M. B.; Browne, Lynn E.; McEneaney, James (1996). Mortgage Lending in Boston: Interpreting HMDA Data. American Economic Review 86(1): 25-53.
2. Hanson, Andrew; Hawley, Zackary; Martin, Hal; Liu, Bo (2016). Discrimination in Mortgage Lending: Evidence from a Correspondence Experiment. Journal of Urban Economics 92: 48-65. https://doi.org/10.1016/j.jue.2015.12.004
3. Nelson, Robert K.; Winling, LaDale; Marciano, Richard; Connolly, Nathan; et al. (2023). Mapping Inequality: Redlining in New Deal America. American Panorama, ed. Robert K. Nelson and Edward L. Ayers, Digital Scholarship Lab, University of Richmond. https://dsl.richmond.edu/panorama/redlining/

## Provenance

- Pipeline version: 0.1.0
- Build script: book2/chapters/mortgage-denials/build.py (sha256 814f4cc85cbefae10bb7bd60cd4c66a5101602b66203772be914f2c965ecad4c)
- Built on: 2026-10-09
- Site commit: c2b1ff3309cb4a10833f884da91ac0057a66ad0a

Inputs:

| File | Size (bytes) | SHA-256, first 1 MB | SHA-256, last 1 MB |
| --- | ---: | --- | --- |
| book2/chapters/mortgage-denials/hmda.py | 13,482 | 4d7e4a2c3720decb65cc28555be2bd7095512d5e30dce0ff407d07f9dcf37b42 | 4d7e4a2c3720decb65cc28555be2bd7095512d5e30dce0ff407d07f9dcf37b42 |
| book2/chapters/mortgage-denials/holc_tracts.py | 2,291 | 1b366aa0e9a51cfb681e206aa807e438133ed778a1cdf6b787e3ffcb178a014c | 1b366aa0e9a51cfb681e206aa807e438133ed778a1cdf6b787e3ffcb178a014c |
| book2/chapters/mortgage-denials/redlining.py | 9,307 | 28a3d08f1c6ac598be472b24f7a0ed7511a41240bfddbf902aceb16bf3316595 | 28a3d08f1c6ac598be472b24f7a0ed7511a41240bfddbf902aceb16bf3316595 |
| data/analytic/tract_atlas.parquet | 39,659,925 | 29d5e4d943552d54aeb6ba92aadc939d9bd15bdea553c78357dbd516757be627 | 1e10d1ecf722340f23ee2212d4999a90250f85b65fa89b39e32db06d9472e22e |
| data/geo/dim_geo.parquet | 1,180,921 | 29d5e4d943552d54aeb6ba92aadc939d9bd15bdea553c78357dbd516757be627 | e58c8c488df54ab3187f5461f179322afdc49c9bd2315404003fe1272eea32c3 |
| data/geo/holc_areas_2023.parquet | 4,084,019 | 6d293647278dc82053c0ab89a192836f0763e73aff77785f3f06ad4214102600 | 37dc09e2ca32f1f9a001d07433b85faf4bd0a31ea584c16b62654b6556084be4 |
| data/geo/tracts_2020.parquet | 619,858,391 | f4f05c8c7a63d1785ec2aad283de3f6fa839b6c437f13a1b96cea31d3344bf0f | 7b3811247c6b3e322954f24a4a2384af52aef6ed8688f8dbdb0407d21ef80a59 |
| data/raw/acs/deprivation_components_2019_2023/tract.parquet | 13,197,793 | 5545cde2cbe96a8b56e468d65388248a241aa6bd300d3c9e325d08d3f86b11da | 5c8ec454d066988b0f1a01aaa6ed34ebfc05d162e01cf9c6bb01f9198cbaff12 |
| data/staged/adi/adi_atlas_tract2020_2023.parquet | 798,078 | 5e409bc212ff871def780fa27b4cf8b616d9da420b2b9f6455517eddcd8c1a95 | 5e409bc212ff871def780fa27b4cf8b616d9da420b2b9f6455517eddcd8c1a95 |
| data/staged/housing/hmda_tract_2024.parquet | 3,512,343 | b1d15b4f05f469ff89ebdfef102f048e35c92e48b035ccc53d89ec44876cecf9 | 13a15bb3dd0d47f2b9273757967d51a6a624259c0603c7a82fd7b4fba28d856f |
| data/staged/ses/acs5_tract2020_2019_2023.parquet | 7,232,424 | 5fe6f730a35225f14d5d8abfc6b9bce58b52e545f309a7eac74f1dae74d7a31c | 7b4a2475b3852d3f91804716a04566ef8ceadacec2be9fb85f5eb642078efac6 |
| data/staged/svi/svi_tract2020_2022.parquet | 17,740,415 | cb3860d8edc2aaf1f2a02211a58f7a0b80457eb6ad9a37b5ec8abb2d5da9da10 | e371a4a696d5bed05d54ad7c5b6ca2504508805b847ec415c2322be7f66ec2b2 |

Outputs:

| File | SHA-256 |
| --- | --- |
| facts.json | 753a34074702cd60ad13949eee9ff31cd4f2a6b4b4433b30561ff0352bff0f01 |
| figures/alt_rankings.json | 773da62209fd79716ba73b2fab4d60c0e69d89cc89b3e144c91194c4f7af73ec |
| figures/cook_denial_map.json | 227f66edcb44254183f05ac1eb7e4b9f09acb416cfc401d10d9799d727af55ce |
| figures/denial_by_group_and_deprivation.json | 22b4bfd20471b4129942b55ba314102d4fbaa1780069fd2da556b550ec1f32c0 |
| figures/denial_by_holc_grade.json | 4a4de27f7011fc020d03e7c18b443dab14e2bfbc8bd496154d83cfe9bce91b97 |
| figures/denial_by_loan_purpose.json | fc4e94ab111073a6835fe9050f85c88980492e1fca37097fedfdd3d9c79216db |
| figures/denial_vs_deprivation.json | 1bea4e523acf3849a3eb019195ce80b6b66a41af203a003863112da81f25fec0 |
| figures/gap_by_deprivation_and_income.json | 11a3d128277615ae2291bfb4722f3c5934a1fccafe9018000c6c90a6dd101c17 |
| figures/same_tract_gaps.json | 16d7c67b84cb0a3164d1b098b88d1c60ffa23c8fc086316331266a471dac4379 |
| meta.json | 6eeaf4dc3073b8de57a00fa35f6e2d46a24c1133219f34bfb5d685ef257d99f0 |
| public/data/mortgage-denials/cook_denial_map.json | 467118e3283844e371fbfa09dda1644b830e16d7e9353fd25d9b3ce068ec638c |
| public/data/mortgage-denials/denial_vs_deprivation_tracts.json | 66c0543541b745ce9ac42c3b19fc29de5cb4c02a27cbd9b768c909f195bc3c4d |
| refs.json | 49de2f8eefdbb181acdeede03c4e8edd6cb7ef3fab1fee54f9b540cd3c046406 |

## Machine formats

- Report (HTML): https://vishalsingh.org/zipcode/mortgage-denials/report
- Report (Markdown): https://vishalsingh.org/zipcode/mortgage-denials/report.md
- Every estimate (JSON): https://vishalsingh.org/zipcode/report/mortgage-denials/facts.json
- Table 1, figure "denial_by_loan_purpose" (JSON): https://vishalsingh.org/zipcode/report/mortgage-denials/figures/denial_by_loan_purpose.json
- Table 2, figure "denial_vs_deprivation" (JSON): https://vishalsingh.org/zipcode/report/mortgage-denials/figures/denial_vs_deprivation.json
- Table 3, figure "denial_by_group_and_deprivation" (JSON): https://vishalsingh.org/zipcode/report/mortgage-denials/figures/denial_by_group_and_deprivation.json
- Table 4, figure "same_tract_gaps" (JSON): https://vishalsingh.org/zipcode/report/mortgage-denials/figures/same_tract_gaps.json
- Table 5, figure "gap_by_deprivation_and_income" (JSON): https://vishalsingh.org/zipcode/report/mortgage-denials/figures/gap_by_deprivation_and_income.json
- Table 6, figure "denial_by_holc_grade" (JSON): https://vishalsingh.org/zipcode/report/mortgage-denials/figures/denial_by_holc_grade.json
- Table 7, figure "cook_denial_map" (JSON): https://vishalsingh.org/zipcode/report/mortgage-denials/figures/cook_denial_map.json
- Table 8, figure "alt_rankings" (JSON): https://vishalsingh.org/zipcode/report/mortgage-denials/figures/alt_rankings.json
- Payload of "denial_vs_deprivation" (JSON): https://vishalsingh.org/zipcode/data/mortgage-denials/denial_vs_deprivation_tracts.json
- Payload of "cook_denial_map" (JSON): https://vishalsingh.org/zipcode/data/mortgage-denials/cook_denial_map.json
- Index of every report: https://vishalsingh.org/zipcode/llms.txt; all reports in one file: https://vishalsingh.org/zipcode/llms-full.txt

---

# Renting, affordability and eviction: report

Chapter 5.4 of ZIP Code Destiny, by Vishal Singh (NYU Stern). Generated from the chapter's data files; the reading version is at https://vishalsingh.org/zipcode/renting-affordability-and-eviction. Table n of this report is figure n of the chapter.

## Identification

- Title: Renting, affordability and eviction
- Label: The first of the month
- Chapter: 5.4 (Part V, Home, work and money; An American life)
- Question: Where do renters spend the largest share of their income on housing, how does that burden differ from owners and across neighborhoods, and how did eviction filings change in the places where they are tracked?
- Dek: 45.6% of renter households with a computed burden spent more than 30% of their income on housing in 2018 to 2022, against 20.8% of owner households, a gap detectable only among low-income households.
- Grain: neighborhoods
- Status: published
- Published: 2026-10-09
- Updated: not revised
- Data sources:
  - HUD CHAS housing affordability. HUD Comprehensive Housing Affordability Strategy, American Community Survey 2018-2022 special tabulation: households by tenure, income relative to HUD's area median family income and housing cost burden, tracts and counties; years: 2018-2022; weights: none (pooled household counts)
  - HUD Fair Market Rents. HUD Fair Market Rents, fiscal year 2025: the two-bedroom rent standard for each FMR area; years: fiscal year 2025 (October 2024 to September 2025); weights: renter households with a computed burden (CHAS)
  - American Community Survey. ACS 2019-2023 county median household income; years: 2019-2023 (5-year); weights: household counts (CHAS) for the area mean
  - Area Deprivation Index. Area Deprivation Index 2023, national rank, tracts ranked into tenths; years: 2023 release (version 4.0.1; built on ACS 2019-2023); weights: households of the tenure with a computed burden
  - Eviction Tracking System. Eviction Tracking System, monthly eviction filings by tract and a pre-pandemic baseline, January 2020 to December 2025; years: 2020-2026 (monthly); weights: none (full counts of filings)
- Canonical URL: https://vishalsingh.org/zipcode/renting-affordability-and-eviction
- Report: https://vishalsingh.org/zipcode/renting-affordability-and-eviction/report
- Markdown: https://vishalsingh.org/zipcode/renting-affordability-and-eviction/report.md
- BibTeX key: singh2026rentingaffordabilityandeviction

```bibtex
@misc{singh2026rentingaffordabilityandeviction,
  author = {Singh, Vishal},
  title = {Renting, affordability and eviction},
  year = {2026},
  url = {https://vishalsingh.org/zipcode/renting-affordability-and-eviction},
  note = {In ZIP Code Destiny. Published 9 October 2026.}
}
```

## Principal estimates

The facts the dek and the opening prose cite, in order of appearance.

| Estimand | Population | Period | Value | Interval | n | Weights | Source | Fact id |
| --- | --- | --- | ---: | --- | ---: | --- | --- | --- |
| ratio of sums over tracts: households with the burden over households whose burden HUD computes, pooled over the tracts of the 50 states and DC; interval from a bootstrap over counties (replicates and seed in the methods note). The share carries no margin of error of its own: HUD rounds every cell independently; the count includes the households with a severe burden | renter households in the tracts of the 50 states and DC with a computed burden | 2018 to 2022 | 45.6% | 95% CI 44.7% to 46.5% | 84,410 tracts | none (pooled household counts) | HUD CHAS housing affordability | bur_renter |
| HUD's line for a housing cost burden: housing costs above this share of household income | households in the HUD CHAS tabulation | 2018 to 2022 | 30% |   |   | none (definition) | HUD CHAS housing affordability | burden_line |
| ratio of sums over tracts: households with the burden over households whose burden HUD computes, pooled over the tracts of the 50 states and DC; interval from a bootstrap over counties (replicates and seed in the methods note). The share carries no margin of error of its own: HUD rounds every cell independently; the count includes the households with a severe burden | owner households in the tracts of the 50 states and DC with a computed burden | 2018 to 2022 | 20.8% | 95% CI 20.1% to 21.5% | 84,410 tracts | none (pooled household counts) | HUD CHAS housing affordability | bur_owner |
| the income line between the lowest and the second of HUD's five income bands (income relative to HUD's area median family income, HAMFI) | households in the HUD CHAS tabulation | 2018 to 2022 | 30% |   |   | none (definition) | HUD CHAS housing affordability | hamfi_30 |
| the income line for a low-income household: income at or below this share of HUD's area median family income (HAMFI) | households in the HUD CHAS tabulation | 2018 to 2022 | 80% |   |   | none (definition) | HUD CHAS housing affordability | hamfi_cut |
| pooled renter share with a burden minus the pooled owner share with a burden; bootstrap over counties | households in the tracts of the 50 states and DC (Puerto Rico left out) | 2018 to 2022 | 24.8 | 95% CI 24.2 to 25.3 | 84,410 tracts | none (pooled household counts) | HUD CHAS housing affordability | bur_gap |
| the renter burden at the owners' mix of five income bands minus the owner burden (a reweighting of the renters' rates, not a causal decomposition); bootstrap over counties. From the raw HUD CHAS county Table 8 | households with a computed burden in the counties of the 50 states and DC (Connecticut by planning region) | 2018 to 2022 | 3.8 | 95% CI 3.3 to 4.2 | 3,144 counties | none (pooled household counts) | HUD CHAS housing affordability | bur_gap_std |

## Findings

Of the renter households with a computed burden in the states and DC, 45.6% (95% CI 44.7% to 46.5%) spent more than 30% of their income on housing in 2018 to 2022, against 20.8% (95% CI 20.1% to 21.5%) of owner households. The gap is widest at incomes between 30% and 80% of the area median family income and is not detectable above 80%. Giving renters the owners' mix of incomes would cut it from 24.8 (95% CI 24.2 to 25.3) to 3.8 (95% CI 3.3 to 4.2) points. Across census tracts ranked by the Area Deprivation Index, the share of renters with a burden has no linear trend nationally, and within a single county it is higher in the more deprived tracts.

The chapter reads the burden by tenure, county and tract. It sets the burden beside the Fair Market Rents that the Department of Housing and Urban Development (HUD) publishes for its housing voucher program. It then follows eviction filings in the cities where the Eviction Lab tracks them. Every comparison is between places or between groups of households, and none follows a household. Home values, tax-return incomes and the redlining grades are in Chapter 5.2, Home value, income and redlining (https://vishalsingh.org/zipcode/home-value-income-and-redlining), mortgage denials in Chapter 5.3, Mortgage denials (https://vishalsingh.org/zipcode/mortgage-denials) and the earnings of workers in Chapter 5.1, Work, wages and access to jobs (https://vishalsingh.org/zipcode/work-wages-and-access-to-jobs).

### Renters are burdened more often than owners, mostly at middle incomes

HUD's Comprehensive Housing Affordability Strategy (CHAS) files tabulate the American Community Survey for 2018 to 2022 by tenure, income and housing cost. A household has a cost burden when its housing costs exceed 30% of its income, a moderate burden when they run from 30% to 50% and a severe burden when they exceed 50%. Costs are rent and utilities for renters, and owner costs such as mortgage payments, taxes, insurance and utilities for owners.

The shares below divide by the households whose burden HUD computes, which leaves out households with no or negative income. That base is 43.2 million renter and 80.8 million owner households in the tracts of the states and DC. The 1.0 million renter and 0.7 million owner households without a computed burden are all in the lowest income band. Renters are 35.2% (95% CI 33.7% to 36.8%) of these households. The file's 84,410 tracts include 1,051 with no households, which add nothing to the pooled shares and drop out of the tenth means below.

A moderate burden affects 22.4% (95% CI 22.1% to 22.7%) of renter households and 12.0% (95% CI 11.7% to 12.3%) of owner households, and a severe burden 23.2% (95% CI 22.6% to 23.8%) and 8.8% (95% CI 8.5% to 9.2%) (Table 1). The renter share with a burden is 2.2 (95% CI 2.1 to 2.2) times the owner share, a gap of 24.8 (95% CI 24.2 to 25.3) points.

In all, 19.7 million renter households spent more than 30% of their income on housing. Of these, 10.0 million spent more than 50%.

Table 1. 45.6% of renter households pay more than 30% of their income for housing, against 20.8% of owner households

Income separates households more than tenure does. HUD measures income against its area median family income (HAMFI), a benchmark it sets for each area that differs from the Census median household income, and sorts households into five bands. The share with a burden runs from 83.9% (95% CI 83.3% to 84.6%) of renters at or below 30% of HAMFI to 5.3% (95% CI 4.8% to 5.9%) of renters in the highest band, and from 82.8% (95% CI 82.2% to 83.4%) to 5.5% (95% CI 5.1% to 5.9%) for owners (second tab of Table 1).

The renter share minus the owner share is +1.2 (95% CI 0.3 to 1.9) points in the lowest band. It is +24.4 (95% CI 23.4 to 25.3) points between 30% and 50% of HAMFI and +12.0 (95% CI 10.8 to 13.2) points between 50% and 80%. In the two bands above 80% it is −0.5 (95% CI −1.4 to 0.4) and −0.2 (95% CI −0.6 to 0.2) points, and the intervals include zero. The gap in the totals therefore reflects where renters sit in the income range.

Of renter households with a computed burden, 60.6% (95% CI 59.8% to 61.4%) have incomes at or below 80% of HAMFI, against 30.2% (95% CI 29.8% to 30.7%) of owner households. Among these low-income households, 69.0% (95% CI 68.1% to 69.9%) of renters and 50.7% (95% CI 49.8% to 51.7%) of owners have a burden. Low-income households are 91.8% (95% CI 91.2% to 92.4%) of burdened renter households and 73.8% (95% CI 72.8% to 74.7%) of burdened owner households.

Giving renters the owners' mix across the five bands would bring the renter share with a burden to 24.6% (95% CI 23.7% to 25.4%), which leaves a gap of 3.8 (95% CI 3.3 to 4.2) points against owners. That reweighting rearranges the shares and does not say what would happen to any household. HUD rounds every count and publishes margins of error for the counts but not for the shares, so the intervals in this chapter come from resampling counties and leave out the survey's own sampling error.

### Renter burden by county

Of the 3,143 counties, 2,753 have at least 500 renter and 500 owner households with a computed burden, and they hold 99.8% of renter households.

Counting each county once, the share of renters with a burden has a median of 37.9% and runs from 23.0% to 50.9% between the 5th and 95th percentiles of counties (Table 2). The pooled national share of 45.6% counts each county in proportion to its renter households.

Table 2. Between the 5th and 95th percentiles of counties, the share of renter households paying more than 30% of income runs from 23.0% to 50.9%, with a median county at 37.9%

In 175 of these counties the share of renter households with a burden is above 50%. Owners have a burden as often as renters, or more often, in only 31 counties.

Pooled over its counties, the renter share with a burden is 53.3% (95% CI 50.4% to 55.4%) in Florida and 50.7% (95% CI 47.3% to 52.4%) in California, against 33.9% (95% CI 30.6% to 35.8%) in South Dakota. The map shows where renters spend a large share of their income and not whether rents or incomes account for it.

The household floor matters for the tails, since a share in a county near it rests on a few hundred renter households and is uncertain by several points. With a floor of 100 households the 5th percentile is 19.2%, and owners have a burden as often as renters in 106 counties. With a floor of 1,000 the 5th percentile is 25.9% and the count is 11.

### Renter burden has no national trend by deprivation and rises within counties

Ranking the 82,786 tracts that have an Area Deprivation Index into tenths, the share of renters with a burden is 44.4% (95% CI 41.9% to 46.7%) in the least deprived tenth and 47.6% (95% CI 46.6% to 48.4%) in the most deprived (Table 3). The means for the ten tenths stay within 3.6 points of each other. The slope across the nation is 0.0 (95% CI −0.3 to 0.3) points for each 10 percentile points of the index for renters, an interval that includes zero.

For owners the slope is −0.8 (95% CI −1.0 to −0.5) points. Owner burden falls from 25.5% (95% CI 23.7% to 27.2%) in the least deprived tenth to 18.3% (95% CI 17.9% to 18.7%) in the ninth and is 19.8% (95% CI 19.3% to 20.3%) in the most deprived. The 1,624 tracts with no index are left out.

Table 3. Pooled across the nation, renter burden is 44.4% in the least deprived tenth of tracts and 47.6% in the most deprived, but within a county it rises by 2.3 points per 10 percentile points of the index

Among low-income households alone, the share with a burden is lower where the index shows more deprivation. For renters it falls from 75.3% (95% CI 72.6% to 78.1%) in the least deprived tenth to 62.5% (95% CI 61.6% to 63.3%) in the most deprived, and for owners from 65.9% (95% CI 64.0% to 68.0%) to 37.6% (95% CI 37.0% to 38.1%).

The index is built partly from median home value, median gross rent and the monthly mortgage cost, so it ranks tracts with costly housing as less deprived (Chapter 1.1, What the deprivation index measures (https://vishalsingh.org/zipcode/anatomy-of-deprivation) shows how far home value alone orders it). Costly housing also means a larger burden at a given income, so a gradient of burden against the index is partly mechanical, and the index understates deprivation in high-cost cities. A tract share is also noisy where a tract has few renter households, which is why each tenth pools the households behind its tracts.

Comparing each tract with its own county removes the differences between counties. Within a county, the renter share with a burden is higher by 2.3 (95% CI 2.1 to 2.4) points for each 10 percentile points of the index, and the owner share by 0.9 (95% CI 0.9 to 1.0). In the most deprived tenth a tract's renter share differs from its county's level by +5.0 (95% CI 4.5 to 5.5) points, and in the least deprived tenth by −3.1 (95% CI −4.0 to −2.3) points. For owners the differences are +2.0 (95% CI 1.7 to 2.3) and −1.2 (95% CI −1.6 to −0.9).

Between counties the slope runs the other way, at −1.1 (95% CI −1.4 to −0.8) points per 10 percentile points for renters and −1.6 (95% CI −1.9 to −1.4) for owners. The two patterns offset each other in the national line, which runs from 44.4% (95% CI 41.9% to 46.7%) of renters with a burden in the least deprived tenth to 47.6% (95% CI 46.6% to 48.4%) in the most deprived. The data show the pattern and not its reason. Rents and incomes both differ between counties, and the index counts both.

### Renter burden rises with the rent standard's share of area income

HUD publishes a Fair Market Rent each fiscal year for each of its areas, a monthly rent standard by bedroom count for its housing voucher program. The file used here is the release for fiscal year 2025, October 2024 to September 2025. It has one value for each area, so every county in a metropolitan area shares it, and a difference between two of them is not a difference in rent. This section therefore compares areas made of whole counties: 2,139 areas with 2,696 counties and 94.4% of the nation's renter households. New England is left out because the file lists towns there, which a county cannot be matched to.

Weighting areas by their renter households, the two-bedroom rent standard is $1,703 a month at the median and runs from $933 to $2,684 between the 5th and 95th percentiles of areas. A year of it is 23.7% of the area's median household income at the median, and the share runs from 17.6% to 35.9% between the 5th and 95th percentiles. Counting each area once, the median share is 19.9%. The income is the household-weighted mean of the counties' ACS medians for 2019 to 2023 and covers all households, not only renters, so the share is not the share of any renter's income.

Across areas the renter share with a burden moves with the rent share. Weighting areas by their renter households, the correlation is +0.70 (95% CI 0.51 to 0.79) and each point of rent share goes with 0.72 (95% CI 0.52 to 0.93) points of renter burden (Table 4). The renter share is 33.9% (95% CI 32.1% to 35.8%) in the tenth of areas with the lowest rent share and 49.9% (95% CI 45.9% to 51.4%) in the tenth with the highest. Both measures are a cost over an income, so they move together partly by construction. Counting each area once, the correlation is +0.36 (95% CI 0.25 to 0.47) and the slope 0.71 (95% CI 0.49 to 0.94).

Table 4. Renter burden is higher in areas where the two-bedroom Fair Market Rent takes a larger share of the median household income, a renter-weighted correlation of +0.70 across 2,139 areas

Areas with higher nominal incomes do not have lower renter burden. Ranked by median household income into fifths, the areas have a median household income that averages $45,717 in the lowest fifth and $91,563 in the highest. The share of renters with a burden is 39.1% (95% CI 37.6% to 40.9%) in the lowest fifth and 46.9% (95% CI 44.8% to 48.4%) in the highest.

The rent standard takes 21.5% (95% CI 20.5% to 22.4%) of income in the middle fifth, against 24.4% (95% CI 23.4% to 25.3%) in the lowest fifth and 26.1% (95% CI 23.0% to 28.3%) in the highest. A comparison of places at the same dollar income would miss that difference in the price of housing. The rent standard is for fiscal year 2025, later than the burden and the incomes. The comparison therefore sets places against one another at one set of prices and is not a statement about a single year.

### Eviction filings against the pre-pandemic baseline

The Eviction Lab's Eviction Tracking System counts eviction court filings by month and census tract in 35 cities and metro areas, which it calls sites. Filings are landlord petitions in court, not completed evictions, and the sites are the places the Eviction Lab tracks, not the nation. The tracts in the index below hold 14.9% of the nation's renter households. Of the households in them, 41.9% (95% CI 39.4% to 43.8%) rent, against 34.2% (95% CI 32.6% to 36.0%) elsewhere. Of their renters, 48.5% (95% CI 46.8% to 50.4%) have a burden, against 45.1% (95% CI 44.1% to 46.2%) elsewhere, so no filing count or index in this section describes the rest of the country.

The index divides a tract's filings in a month by the filings that the Eviction Lab records for the same calendar month before the pandemic (the baseline), summed over tracts. An index of 1.00 means filings at the baseline. The panel is the 10,243 tracts of 32 sites that have a baseline and an observation in each of the 72 months from January 2020 to December 2025. Atlanta and the Other Oregon Counties site have no baseline, and Greenville, South Carolina, has no tract observed in every month, so the three are left out.

The pooled index was 1.01 in January 2020, fell to 0.09 in April 2020, which is 91% below the baseline, and stayed below the baseline until January 2022 (Table 5). It first rose above the baseline in February 2022, reached 1.37 in March 2023, which is 37% above the baseline, and averaged 1.17 over 2023. The shaded band marks the federal moratoria, from the CARES Act in March 2020 to the Supreme Court's ruling against the CDC order in August 2021.[1] Federal protection was not continuous, and states and cities set their own rules, so the band is a guide to timing and not a measure of coverage.

Table 5. Eviction filings fell to 0.09 times their pre-pandemic level in April 2020, stayed below it until early 2022 and peaked at 1.37 times that level in March 2023

The pooled series hides differences between neighborhoods. The Eviction Lab classifies each tract by the racial or ethnic group that makes up a majority of its residents in the American Community Survey for 2015 to 2019. The classification says nothing about the race of any tenant. Black and white mean non-Hispanic, and a fourth group holds tracts where another group or no group is a majority. 236 panel tracts with no classification are left out of the comparison by group. In 2023 the index was 0.98 in majority-Black tracts, 1.11 in majority-Hispanic tracts and 1.28 in majority-white tracts (Table 6).

Since March 2021 the line for majority-Black tracts is at or near the lowest of the four groups in every month, and from 2024 the line for majority-white tracts is at or near the highest. Within the 23 sites that have at least 5 majority-Black and 5 majority-white tracts, the 2023 index is lower in majority-Black tracts than in majority-white tracts in 21 of them.

The index compares each group with its own past, and the pasts differ. Before the pandemic, filings ran at 12.7 per hundred renter households per year in majority-Black tracts and 4.2 in majority-white tracts, a ratio of 3.0. In 2023 the rates were 12.5 and 5.4. Filings in majority-Black tracts returned to about their baseline, which was already high, and filings in majority-white tracts rose above a lower baseline. Hepburn, Louis and Desmond found the highest rates of eviction filing among Black renters in court records for 2012 to 2016.[2]

Table 6. Eviction filings in majority-Black tracts ran at 0.98 times their baseline in 2023 and in majority-white tracts at 1.28 times, from baseline rates per renter household that differed by a factor of 3.0

Ranking the panel tracts that have an Area Deprivation Index by deprivation, the index over 2022 to 2025 is 1.48 in the least deprived tenth and 0.89 in the most deprived tenth. That is the reverse of the order that a gradient in distress would suggest. The least deprived tenth had a baseline of 11 filings per tract per year and the most deprived tenth 88. A small number of added filings therefore moves a low-baseline tract's index by a large ratio. The index shows where filings rose against the past, not where households face the most eviction.

Across the 32 sites in the panel, the index for 2023 to 2025 has a median of 1.11 and runs from 0.66 to 1.48 between the 5th and 95th percentiles of sites (Table 7). Portland is highest, at 1.84, and Philadelphia is among the lowest, at 0.62. These data do not say why. Local rules, court practice, rental assistance and the way filings are recorded differ by site.

Table 7. From 2023 to 2025 filings ran from 0.62 times the baseline in Philadelphia to 1.84 times in Portland, with a median site at 1.11 times

Matthew Desmond argued in *Evicted* that eviction is a cause of poverty and not only a condition of it.[3] Counts of filings cannot test that claim, since they count petitions and not what follows them, and the box below reports a study that tests a version of it.

> Causal evidence: Whether an eviction order changes a tenant's later outcomes (Instrumental variables with random assignment of cases to judges)
>
> Collinson and coauthors test a version of that claim with linked court and administrative records for Cook County, Illinois, and New York City.[4] Cases in these courts are assigned to judges at random, and judges differ in how often they order an eviction, so the judge a tenant draws shifts the chance of an eviction order independently of the tenant. Using that variation, the authors find that an eviction order increases homelessness and hospital visits and lowers earnings, durable goods consumption and access to credit in the years just after the order, with larger effects for female and Black tenants.
>
> The estimates are for tenants whose outcomes depend on the judge they draw, in two courts, and they concern a court order. They do not measure the effect of the pandemic moratoria or of the monthly filing pattern in this section.

### Renter burden rises steadily with deprivation under the standardized index and income

Over the 75,587 tracts that have renter households with a computed burden and all four rankings, the Area Deprivation Index puts the share of renter households over the burden line at 44.0% (95% CI 41.4% to 46.2%) in the least deprived tenth and 47.1% (95% CI 46.3% to 48.0%) in the most deprived, with a rank correlation between tenth and share of −0.30 (Table 8). Under the standardized index the share rises steadily, from 38.6% (95% CI 37.0% to 40.3%) to 53.5% (95% CI 52.7% to 54.3%) with a rank correlation of +0.99, and under median household income from 37.7% (95% CI 36.4% to 39.1%) to 53.4% (95% CI 52.6% to 54.1%). Under the bachelor's degree share it runs from 39.6% (95% CI 38.5% to 40.9%) to 49.3% (95% CI 48.0% to 50.6%) with a rank correlation of +0.79, uneven like the line under the deprivation index. By the rule in Chapter 9.1, How far one ranking goes (https://vishalsingh.org/zipcode/how-far-one-ranking-goes) the standardized index and income would each change the shape of this gradient, but rent and income are inputs to the deprivation index and to the standardized index (Chapter 1.1, What the deprivation index measures (https://vishalsingh.org/zipcode/anatomy-of-deprivation)), so the comparison is reported and left out of the book's count.

Table 8. The renter share over the cost-burden line is worse at the more deprived end but uneven along the way under the ADI; it is steadily worse toward the more deprived end under the standardized index and the income ranking

### What these data cannot show

Every comparison in this chapter describes places or groups of households. A county, a tract, an area in the Fair Market Rent file or a site in the Eviction Tracking System is not a person. A pattern across places need not hold for the people inside them (ecological inference). The renters of a tract differ from people who move in and out of it, and the data cannot separate who lives there from what the place provides.

The burden compares housing costs with income at one time. It does not say whether a household chose costly housing or had no cheaper option, and it does not measure the condition of the home, crowding or arrears. The shares come from a survey tabulation that HUD rounds cell by cell, tract shares are noisy, and the intervals leave out the survey's sampling error. HAMFI is HUD's benchmark and differs from the median household income used elsewhere in the book. The deprivation tenths use an index that counts housing costs among its inputs, so they cannot say whether deprivation or price levels sit behind a gradient.

The rent standard is one release for one fiscal year, set for an area and not for a county, and the incomes are the incomes of all households. The comparison holds across areas and not within a metropolitan area, and New England is not in it.

Filings are landlord petitions, not evictions or displaced households, and one household can be the subject of several filings. The index compares each tract with its own baseline, so it does not say where eviction is most common. Moratoria, emergency rental assistance, court closures and changes in local law all moved filings in these years, and the data do not separate them. The renter households behind the filing rates are from 2019 to 2023 and the filings from 2020 to 2025. The counts are full counts for the participating sites, so the series carry no sampling interval, and the sites are not a sample of the nation.

## Tables

### Table 1. 45.6% of renter households pay more than 30% of their income for housing, against 20.8% of owner households

Housing cost burden by tenure as HUD defines it, and by income relative to HUD's area median family income, 50 states and DC, 2018 to 2022

#### Table 1.1. By tenure

Of renter households, 23.2% pay more than 50% of their income for housing, against 8.8% of owner households

Percent of households by tenure with housing costs of 30% to 50% of income (moderate burden) and above 50% (severe burden), 50 states and DC, 2018 to 2022

| level | share | tenure |
| --- | ---: | --- |
| Moderate burden | 11.98 | Owner households |
| Severe burden | 8.82 | Owner households |
| Moderate burden | 22.39 | Renter households |
| Severe burden | 23.17 | Renter households |

#### Table 1.2. By income and tenure

The renter share with a burden is +24.4 points above the owner share at incomes of 30% to 50% of the area median and +12.0 points above it at 50% to 80%, and −0.5 and −0.2 points above it in the two bands over 80%

Percent of households with a computed burden paying more than 30% of income, by tenure and by income relative to HUD's area median family income (HAMFI), 2018 to 2022

| conf_high | conf_low | households | income | share | tenure |
| ---: | ---: | ---: | --- | ---: | --- |
| 83.36 | 82.24 | 5567551 | At or below 30% of HAMFI | 82.79 | Owner households |
| 84.62 | 83.26 | 9549471 | At or below 30% of HAMFI | 83.94 | Renter households |
| 54.06 | 52.15 | 6958005 | Above 30% to 50% | 53.11 | Owner households |
| 78.39 | 76.49 | 7497295 | Above 30% to 50% | 77.48 | Renter households |
| 35.51 | 33.46 | 11919509 | Above 50% to 80% | 34.49 | Owner households |
| 47.96 | 44.82 | 9118643 | Above 50% to 80% | 46.49 | Renter households |
| 22.12 | 20.07 | 8344114 | Above 80% to 100% | 21.11 | Owner households |
| 21.98 | 19.15 | 4689985 | Above 80% to 100% | 20.65 | Renter households |
| 5.94 | 5.13 | 48039984 | Above 100% | 5.53 | Owner households |
| 5.87 | 4.75 | 12336798 | Above 100% | 5.31 | Renter households |

Note: HUD counts a household as burdened when its housing costs exceed 30% of its income, and as severely burdened above 50%. Costs are rent and utilities for renters, and mortgage payments, taxes, insurance and utilities for owners. The moderate segment is the burdened share minus the severe share.

Note: The five income bands are HUD's: at or below 30% of its area median family income (HAMFI), a benchmark HUD sets for each area that differs from the Census median household income, then up to 50%, up to 80%, up to the area median and above it. Low-income households are the three lowest bands. Every share divides by households with a computed burden, which leaves out 1.0 million renter and 0.7 million owner households with no or negative income, all in the lowest band; the staged HUD percentage for low-income burdened households divides by all households and is not used.

Note: The second tab reads the raw county table of HUD's Table 8 (Connecticut by planning region), which is the only source of the five bands, and its totals differ from the tract totals in the first tab by HUD's rounding (a few hundredths of a point). In the two highest bands the renter and owner shares are about equal, so the gap in the first tab reflects where renters and owners sit in the income range.

Note: The shares are ratios of survey-based household counts that HUD rounds cell by cell; HUD publishes margins of error for the counts and none for the shares. The intervals in the second tab come from a bootstrap over counties (replicates and seed in the methods note) and do not include the survey's sampling error. The stacked bars in the first tab carry no interval.

Note: Puerto Rico is left out of every figure in this chapter.

Source: HUD CHAS (American Community Survey 2018-2022 special tabulation); shares pooled over tracts and weighted by households, survey-based estimates with no margin of error published for the shares

Data: https://vishalsingh.org/zipcode/report/renting-affordability-and-eviction/figures/burden_by_tenure.json

### Table 2. Between the 5th and 95th percentiles of counties, the share of renter households paying more than 30% of income runs from 23.0% to 50.9%, with a median county at 37.9%

Percent of renter households with housing costs above 30% of income, counties of the 50 states and DC, 2018 to 2022

No inline rows: the values of Table 2 are in https://vishalsingh.org/zipcode/data/renting-affordability-and-eviction/renter_burden_map.json.

Note: A county is drawn when it has at least 500 renter households and at least 500 owner households with a computed burden; 2,753 of 3,143 counties pass, and they hold 99.8% of the nation's renter households. A share in a county near the floor rests on a few hundred households and is uncertain by several points; HUD rounds every count and publishes no margin of error for the share.

Note: Connecticut is drawn as its eight former counties, summed from the tract file because the county file is coded by planning region. Alaska is drawn. Puerto Rico is not drawn.

Source: HUD CHAS (American Community Survey 2018-2022 special tabulation), county file; shares of renter households pooled within counties and weighted by households, survey-based estimates with no margin of error published for the shares

Data: https://vishalsingh.org/zipcode/report/renting-affordability-and-eviction/figures/renter_burden_map.json

### Table 3. Pooled across the nation, renter burden is 44.4% in the least deprived tenth of tracts and 47.6% in the most deprived, but within a county it rises by 2.3 points per 10 percentile points of the index

Share of households paying more than 30% of income by tenth of tracts on the Area Deprivation Index, for all households, for low-income households and relative to the county, 2018 to 2022

#### Table 3.1. All households

Renter burden is 44.4% in the least deprived tenth of tracts and 47.6% in the most deprived, while owner burden is 25.5% in the least deprived tenth and 18.3% in the ninth

Percent of households paying more than 30% of income by tenth of tracts on the Area Deprivation Index, pooled over tracts, 50 states and DC, 2018 to 2022

| conf_high | conf_low | decile | series | share | tracts |
| ---: | ---: | ---: | --- | ---: | ---: |
| 46.68 | 41.91 | 1 | Renter households | 44.45 | 8208 |
| 50.08 | 45 | 2 | Renter households | 47.68 | 8074 |
| 48.50 | 44.76 | 3 | Renter households | 46.65 | 8454 |
| 46.71 | 44.48 | 4 | Renter households | 45.61 | 8241 |
| 46.16 | 44.14 | 5 | Renter households | 45.17 | 8179 |
| 46.02 | 44.16 | 6 | Renter households | 45.12 | 8304 |
| 45.57 | 43.59 | 7 | Renter households | 44.64 | 8307 |
| 45.27 | 43.51 | 8 | Renter households | 44.42 | 8234 |
| 44.95 | 43.26 | 9 | Renter households | 44.13 | 8350 |
| 48.42 | 46.63 | 10 | Renter households | 47.58 | 8207 |
| 27.22 | 23.74 | 1 | Owner households | 25.55 | 8189 |
| 25.42 | 22.31 | 2 | Owner households | 23.87 | 8070 |
| 22.64 | 20.98 | 3 | Owner households | 21.79 | 8460 |
| 21.70 | 20.36 | 4 | Owner households | 21.03 | 8229 |
| 20.85 | 19.59 | 5 | Owner households | 20.20 | 8147 |
| 20 | 18.90 | 6 | Owner households | 19.45 | 8249 |
| 19.56 | 18.45 | 7 | Owner households | 18.97 | 8249 |
| 18.89 | 17.95 | 8 | Owner households | 18.42 | 8193 |
| 18.74 | 17.90 | 9 | Owner households | 18.31 | 8329 |
| 20.28 | 19.25 | 10 | Owner households | 19.78 | 8212 |

#### Table 3.2. Low-income households

Among low-income households, the share paying more than 30% of income falls from 75.3% to 62.5% for renters across tenths of tracts, and from 65.9% to 37.6% for owners

Percent of low-income households (income at or below 80% of the area median family income) paying more than 30% of income, by tenth of tracts on the Area Deprivation Index, 2018 to 2022

| conf_high | conf_low | decile | series | share | tracts |
| ---: | ---: | ---: | --- | ---: | ---: |
| 78.06 | 72.60 | 1 | Renter households | 75.33 | 7840 |
| 76.22 | 74.15 | 2 | Renter households | 75.05 | 7803 |
| 75.27 | 72.86 | 3 | Renter households | 74.01 | 8208 |
| 74.23 | 72.10 | 4 | Renter households | 73.20 | 8070 |
| 72.19 | 69.90 | 5 | Renter households | 71.11 | 8094 |
| 70.06 | 67.89 | 6 | Renter households | 68.99 | 8246 |
| 67.66 | 65.52 | 7 | Renter households | 66.62 | 8273 |
| 65.44 | 63.63 | 8 | Renter households | 64.56 | 8219 |
| 63.27 | 61.66 | 9 | Renter households | 62.51 | 8337 |
| 63.27 | 61.58 | 10 | Renter households | 62.46 | 8194 |
| 68.03 | 64.01 | 1 | Owner households | 65.94 | 8103 |
| 65.78 | 62.52 | 2 | Owner households | 64.09 | 8004 |
| 61.83 | 59.86 | 3 | Owner households | 60.84 | 8387 |
| 58.36 | 56.58 | 4 | Owner households | 57.47 | 8178 |
| 54.34 | 52.58 | 5 | Owner households | 53.46 | 8112 |
| 50.61 | 48.99 | 6 | Owner households | 49.79 | 8204 |
| 47.31 | 45.57 | 7 | Owner households | 46.39 | 8209 |
| 44.23 | 42.86 | 8 | Owner households | 43.56 | 8160 |
| 41.13 | 39.97 | 9 | Owner households | 40.53 | 8318 |
| 38.12 | 37.02 | 10 | Owner households | 37.60 | 8206 |

#### Table 3.3. Within the same county

Within a county, the renter share with a burden differs from the county's own level by −3.1 points in the least deprived tenth of tracts and by +5.0 points in the most deprived

Tract share paying more than 30% of income minus the county's household-weighted share across its tracts, in percentage points, by tenth of tracts on the Area Deprivation Index, 2018 to 2022

| conf_high | conf_low | decile | series | share | tracts |
| ---: | ---: | ---: | --- | ---: | ---: |
| −2.35 | −3.97 | 1 | Renter households | −3.06 | 8208 |
| −0.040 | −2.45 | 2 | Renter households | −1.07 | 8074 |
| −0.430 | −2.25 | 3 | Renter households | −1.33 | 8454 |
| −0.580 | −1.84 | 4 | Renter households | −1.21 | 8241 |
| −0.350 | −1.35 | 5 | Renter households | −0.840 | 8179 |
| 0.350 | −0.640 | 6 | Renter households | −0.130 | 8304 |
| 0.840 | −0.210 | 7 | Renter households | 0.330 | 8307 |
| 1.67 | 0.730 | 8 | Renter households | 1.19 | 8234 |
| 2.45 | 1.47 | 9 | Renter households | 1.96 | 8350 |
| 5.46 | 4.54 | 10 | Renter households | 5.01 | 8207 |
| −0.890 | −1.58 | 1 | Owner households | −1.19 | 8189 |
| −0.650 | −1.32 | 2 | Owner households | −0.970 | 8070 |
| −0.500 | −1.03 | 3 | Owner households | −0.780 | 8460 |
| −0.020 | −0.520 | 4 | Owner households | −0.280 | 8229 |
| 0.140 | −0.270 | 5 | Owner households | −0.070 | 8147 |
| 0.340 | −0.060 | 6 | Owner households | 0.140 | 8249 |
| 0.690 | 0.270 | 7 | Owner households | 0.470 | 8249 |
| 0.860 | 0.460 | 8 | Owner households | 0.660 | 8193 |
| 1.18 | 0.810 | 9 | Owner households | 1 | 8329 |
| 2.33 | 1.72 | 10 | Owner households | 2.03 | 8212 |

Note: The Area Deprivation Index is built partly from median home value, median gross rent and median monthly mortgage cost, so it ranks tracts with costly housing as less deprived and a burden gradient against it is partly mechanical. The index also understates deprivation in high-cost cities. Tenths hold near-equal numbers of tracts (tracts with equal ranks stay in one tenth), not equal numbers of households.

Note: 82,786 tracts have an index and 1,624 do not and are left out. Each point weights tracts by the households behind the share, and a tract share rests on a few households in a small tract.

Note: Across all the tracts with an index, 45.5% of renter households and 20.8% of owner households pay more than 30% of income, and 69.0% of low-income renter households do.

Note: In the third tab the county level is the household-weighted mean of the tract shares of the tenure across the county's tracts that have an index, so a county with one such tract adds no deviation. The slope is the within-county slope of the tract share on the index national rank with a separate baseline for each county, per 10 percentile points.

Source: HUD CHAS (American Community Survey 2018-2022 special tabulation) and Area Deprivation Index 2023 (built on ACS 2019-2023); shares pooled over tracts and weighted by households, bootstrap over counties, survey-based estimates with no margin of error published for the shares

Data: https://vishalsingh.org/zipcode/report/renting-affordability-and-eviction/figures/burden_by_deprivation.json

### Table 4. Renter burden is higher in areas where the two-bedroom Fair Market Rent takes a larger share of the median household income, a renter-weighted correlation of +0.70 across 2,139 areas

Renter households paying more than 30% of income (CHAS 2018-2022) against a year of the two-bedroom Fair Market Rent (fiscal year 2025) as a percent of the area's median household income (ACS 2019-2023), HUD FMR areas

#### Table 4.1. Rent share against renter burden

Renter burden is 33.9% in the lowest tenth of areas by the rent standard's share of median household income and 49.9% in the highest

Renter households paying more than 30% of income against a year of the two-bedroom Fair Market Rent as a percent of median household income, one dot per FMR area, with renter-weighted means for tenths of areas

No inline rows: the values of Table 4.1 are in https://vishalsingh.org/zipcode/data/renting-affordability-and-eviction/fmr_scatter.json.

#### Table 4.2. By area income

Across fifths of areas by median household income, renter burden is 39.1% in the lowest fifth and 46.9% in the highest, while the rent standard's share of income is lower in the second and middle fifths than in the lowest and highest

Renter-weighted means for fifths of FMR areas ranked by median household income: the rent standard as a percent of median household income and the percent of renter households paying more than 30% of income

| areas | conf_high | conf_low | fifth | measure | value |
| ---: | ---: | ---: | --- | --- | ---: |
| 427 | 25.28 | 23.39 | Lowest income fifth | Rent standard as a percent of median household income | 24.38 |
| 428 | 22.41 | 20.87 | Second fifth | Rent standard as a percent of median household income | 21.63 |
| 428 | 22.43 | 20.54 | Middle fifth | Rent standard as a percent of median household income | 21.48 |
| 428 | 26.94 | 20.44 | Fourth fifth | Rent standard as a percent of median household income | 23.42 |
| 428 | 28.31 | 23.02 | Highest income fifth | Rent standard as a percent of median household income | 26.11 |
| 427 | 40.91 | 37.58 | Lowest income fifth | Renter households paying more than 30% of income (%) | 39.12 |
| 428 | 42.20 | 39.58 | Second fifth | Renter households paying more than 30% of income (%) | 40.95 |
| 428 | 43.96 | 40.91 | Middle fifth | Renter households paying more than 30% of income (%) | 42.53 |
| 428 | 47.69 | 42.06 | Fourth fifth | Renter households paying more than 30% of income (%) | 44.81 |
| 428 | 48.37 | 44.83 | Highest income fifth | Renter households paying more than 30% of income (%) | 46.86 |

Note: Fair Market Rents are monthly rent standards for the housing voucher program. HUD sets one value for each FMR area and applies it to every county in the area, so counties of one metropolitan area share a value and this chapter compares areas, not counties. The file is one release, for fiscal year 2025; the incomes are ACS 2019-2023 in 2023 dollars and the burden is CHAS 2018-2022, so the three describe different years and the level of the rent share is not exact.

Note: The area income is the household-weighted mean of the median household incomes of its counties, a mean of medians and not the median of the area's households. It covers all households, not renters, so the rent share of the income of renters is higher than the share plotted.

Note: An area enters when its counties are whole in the file and hold at least 500 renter households with a computed burden. New England areas are towns that counties cannot be matched to and are left out, which covers Connecticut, Maine, Massachusetts, New Hampshire, Rhode Island and Vermont; the 2,139 areas hold 94.4% of the nation's renter households.

Note: The scatter is every area (no sample); the fitted line is the renter-weighted least-squares line, with a slope of 0.72 points of renter burden per point of rent share. Fifths and tenths hold near-equal numbers of areas, not equal numbers of renters.

Source: HUD Fair Market Rents, fiscal year 2025, with ACS 2019-2023 county median household incomes and HUD CHAS 2018-2022 renter burden; renter-weighted, one rent standard per FMR area

Data: https://vishalsingh.org/zipcode/report/renting-affordability-and-eviction/figures/rent_standard_vs_burden.json

### Table 5. Eviction filings fell to 0.09 times their pre-pandemic level in April 2020, stayed below it until early 2022 and peaked at 1.37 times that level in March 2023

Monthly eviction filings divided by the pre-pandemic baseline for the same calendar month, summed over tracts in cities and metro areas the Eviction Lab tracks, January 2020 to December 2025

| baseline | filings | index | month | x |
| ---: | ---: | ---: | --- | ---: |
| 42272.4 | 42872 | 1.01 | 2020-01 | 2020 |
| 34769.3 | 37665 | 1.08 | 2020-02 | 2020.1 |
| 31946.6 | 20221 | 0.633 | 2020-03 | 2020.2 |
| 34266.3 | 2977 | 0.087 | 2020-04 | 2020.3 |
| 38876.2 | 5210 | 0.134 | 2020-05 | 2020.3 |
| 42054.2 | 11163 | 0.265 | 2020-06 | 2020.4 |
| 42511.8 | 11588 | 0.273 | 2020-07 | 2020.5 |
| 43972.6 | 18092 | 0.411 | 2020-08 | 2020.6 |
| 41377 | 20824 | 0.503 | 2020-09 | 2020.7 |
| 41585.2 | 23256 | 0.559 | 2020-10 | 2020.8 |
| 36142.7 | 23371 | 0.647 | 2020-11 | 2020.8 |
| 36183.2 | 24355 | 0.673 | 2020-12 | 2020.9 |
| 42272.4 | 21712 | 0.514 | 2021-01 | 2021 |
| 34551.6 | 18147 | 0.525 | 2021-02 | 2021.1 |
| 31946.6 | 23346 | 0.731 | 2021-03 | 2021.2 |
| 34266.3 | 18309 | 0.534 | 2021-04 | 2021.3 |
| 38876.2 | 18073 | 0.465 | 2021-05 | 2021.3 |
| 42054.2 | 23690 | 0.563 | 2021-06 | 2021.4 |
| 42511.8 | 24160 | 0.568 | 2021-07 | 2021.5 |
| 43972.6 | 25512 | 0.580 | 2021-08 | 2021.6 |
| 41377 | 30486 | 0.737 | 2021-09 | 2021.7 |
| 41585.2 | 32950 | 0.792 | 2021-10 | 2021.8 |
| 36142.7 | 30066 | 0.832 | 2021-11 | 2021.8 |
| 36183.2 | 30786 | 0.851 | 2021-12 | 2021.9 |
| 42272.4 | 37251 | 0.881 | 2022-01 | 2022 |
| 34551.6 | 37192 | 1.08 | 2022-02 | 2022.1 |
| 31946.6 | 40492 | 1.27 | 2022-03 | 2022.2 |
| 34266.3 | 36033 | 1.05 | 2022-04 | 2022.3 |
| 38876.2 | 38922 | 1.00 | 2022-05 | 2022.3 |
| 42054.2 | 43885 | 1.04 | 2022-06 | 2022.4 |
| 42511.8 | 45073 | 1.06 | 2022-07 | 2022.5 |
| 43972.6 | 50594 | 1.15 | 2022-08 | 2022.6 |
| 41377 | 46815 | 1.13 | 2022-09 | 2022.7 |
| 41585.2 | 46443 | 1.12 | 2022-10 | 2022.8 |
| 36142.7 | 42892 | 1.19 | 2022-11 | 2022.8 |
| 36183.2 | 44738 | 1.24 | 2022-12 | 2022.9 |
| 42272.4 | 49002 | 1.16 | 2023-01 | 2023 |
| 34551.6 | 43487 | 1.26 | 2023-02 | 2023.1 |
| 31946.6 | 43843 | 1.37 | 2023-03 | 2023.2 |
| 34266.3 | 38975 | 1.14 | 2023-04 | 2023.3 |
| 38876.2 | 46373 | 1.19 | 2023-05 | 2023.3 |
| 42054.2 | 47463 | 1.13 | 2023-06 | 2023.4 |
| 42511.8 | 45985 | 1.08 | 2023-07 | 2023.5 |
| 43972.6 | 51021 | 1.16 | 2023-08 | 2023.6 |
| 41377 | 45813 | 1.11 | 2023-09 | 2023.7 |
| 41585.2 | 49261 | 1.19 | 2023-10 | 2023.8 |
| 36142.7 | 42908 | 1.19 | 2023-11 | 2023.8 |
| 36183.2 | 41475 | 1.15 | 2023-12 | 2023.9 |
| 42272.4 | 48898 | 1.16 | 2024-01 | 2024 |
| 34769.3 | 45467 | 1.31 | 2024-02 | 2024.1 |
| 31946.6 | 37506 | 1.17 | 2024-03 | 2024.2 |
| 34266.3 | 41059 | 1.20 | 2024-04 | 2024.3 |
| 38876.2 | 42532 | 1.09 | 2024-05 | 2024.3 |
| 42054.2 | 43694 | 1.04 | 2024-06 | 2024.4 |
| 42511.8 | 48633 | 1.14 | 2024-07 | 2024.5 |
| 43972.6 | 46965 | 1.07 | 2024-08 | 2024.6 |
| 41377 | 44394 | 1.07 | 2024-09 | 2024.7 |
| 41585.2 | 42052 | 1.01 | 2024-10 | 2024.8 |
| 36142.7 | 39671 | 1.10 | 2024-11 | 2024.8 |
| 36183.2 | 42537 | 1.18 | 2024-12 | 2024.9 |
| 42272.4 | 46910 | 1.11 | 2025-01 | 2025 |
| 34551.6 | 41394 | 1.20 | 2025-02 | 2025.1 |
| 31946.6 | 38220 | 1.20 | 2025-03 | 2025.2 |
| 34266.3 | 38359 | 1.12 | 2025-04 | 2025.3 |
| 38876.2 | 42221 | 1.09 | 2025-05 | 2025.3 |
| 42054.2 | 42601 | 1.01 | 2025-06 | 2025.4 |
| 42511.8 | 47242 | 1.11 | 2025-07 | 2025.5 |
| 43972.6 | 44216 | 1.01 | 2025-08 | 2025.6 |
| 41377 | 44694 | 1.08 | 2025-09 | 2025.7 |
| 41585.2 | 45124 | 1.08 | 2025-10 | 2025.8 |
| 36142.7 | 36607 | 1.01 | 2025-11 | 2025.8 |
| 36183.2 | 40997 | 1.13 | 2025-12 | 2025.9 |

Note: The index is a ratio of sums: filings summed over the tract-months of the panel divided by the baseline summed over the same tract-months. 1 means filings equal the baseline for the same calendar month before the pandemic. Filings are landlord petitions, not completed evictions.

Note: The panel is the 10,243 tracts of 32 sites that have a baseline and an observation in each of the 72 months. Atlanta and Other Oregon Counties have no baseline, and Greenville, South Carolina, has no tract observed in every month, so the three are left out; the file has 35 sites in all. The months of 2026 are incomplete and are not shown.

Note: The band runs from March 2020 to August 2021. Federal protection was not continuous: the CARES Act moratorium ended in the summer of 2020 and the CDC order began in September 2020, and states and cities set their own rules. The counts are full counts of filings in the participating sites, so the line has no sampling interval.

Source: Eviction Lab Eviction Tracking System, 2020-2025, balanced panel of tracts with a baseline; full counts of court filings, not weighted

Data: https://vishalsingh.org/zipcode/report/renting-affordability-and-eviction/figures/eviction_filing_index.json

### Table 6. Eviction filings in majority-Black tracts ran at 0.98 times their baseline in 2023 and in majority-white tracts at 1.28 times, from baseline rates per renter household that differed by a factor of 3.0

Eviction filings against the pre-pandemic baseline by the racial majority of the tract, and filings per 100 renter households per year before the pandemic and in 2023

#### Table 6.1. Against the baseline

Since March 2021 the filing index is at or near its lowest in majority-Black tracts in every month, and from 2024 it is at or near its highest in majority-white tracts

Monthly filings divided by each group's pre-pandemic baseline, by racial majority of the tract, January 2020 to December 2025

| group | index | month | tracts | x |
| --- | ---: | --- | ---: | ---: |
| Majority-Black tracts | 0.964 | 2020-01 | 1342 | 2020 |
| Majority-Black tracts | 1.02 | 2020-02 | 1342 | 2020.1 |
| Majority-Black tracts | 0.566 | 2020-03 | 1342 | 2020.2 |
| Majority-Black tracts | 0.074 | 2020-04 | 1342 | 2020.3 |
| Majority-Black tracts | 0.137 | 2020-05 | 1342 | 2020.3 |
| Majority-Black tracts | 0.296 | 2020-06 | 1342 | 2020.4 |
| Majority-Black tracts | 0.306 | 2020-07 | 1342 | 2020.5 |
| Majority-Black tracts | 0.405 | 2020-08 | 1342 | 2020.6 |
| Majority-Black tracts | 0.515 | 2020-09 | 1342 | 2020.7 |
| Majority-Black tracts | 0.521 | 2020-10 | 1342 | 2020.8 |
| Majority-Black tracts | 0.551 | 2020-11 | 1342 | 2020.8 |
| Majority-Black tracts | 0.661 | 2020-12 | 1342 | 2020.9 |
| Majority-Black tracts | 0.477 | 2021-01 | 1342 | 2021 |
| Majority-Black tracts | 0.527 | 2021-02 | 1342 | 2021.1 |
| Majority-Black tracts | 0.689 | 2021-03 | 1342 | 2021.2 |
| Majority-Black tracts | 0.498 | 2021-04 | 1342 | 2021.3 |
| Majority-Black tracts | 0.414 | 2021-05 | 1342 | 2021.3 |
| Majority-Black tracts | 0.467 | 2021-06 | 1342 | 2021.4 |
| Majority-Black tracts | 0.472 | 2021-07 | 1342 | 2021.5 |
| Majority-Black tracts | 0.497 | 2021-08 | 1342 | 2021.6 |
| Majority-Black tracts | 0.631 | 2021-09 | 1342 | 2021.7 |
| Majority-Black tracts | 0.712 | 2021-10 | 1342 | 2021.8 |
| Majority-Black tracts | 0.755 | 2021-11 | 1342 | 2021.8 |
| Majority-Black tracts | 0.791 | 2021-12 | 1342 | 2021.9 |
| Majority-Black tracts | 0.801 | 2022-01 | 1342 | 2022 |
| Majority-Black tracts | 0.932 | 2022-02 | 1342 | 2022.1 |
| Majority-Black tracts | 1.10 | 2022-03 | 1342 | 2022.2 |
| Majority-Black tracts | 0.908 | 2022-04 | 1342 | 2022.3 |
| Majority-Black tracts | 0.864 | 2022-05 | 1342 | 2022.3 |
| Majority-Black tracts | 0.876 | 2022-06 | 1342 | 2022.4 |
| Majority-Black tracts | 0.891 | 2022-07 | 1342 | 2022.5 |
| Majority-Black tracts | 0.969 | 2022-08 | 1342 | 2022.6 |
| Majority-Black tracts | 0.951 | 2022-09 | 1342 | 2022.7 |
| Majority-Black tracts | 0.939 | 2022-10 | 1342 | 2022.8 |
| Majority-Black tracts | 1.04 | 2022-11 | 1342 | 2022.8 |
| Majority-Black tracts | 1.05 | 2022-12 | 1342 | 2022.9 |
| Majority-Black tracts | 0.992 | 2023-01 | 1342 | 2023 |
| Majority-Black tracts | 1.09 | 2023-02 | 1342 | 2023.1 |
| Majority-Black tracts | 1.17 | 2023-03 | 1342 | 2023.2 |
| Majority-Black tracts | 0.973 | 2023-04 | 1342 | 2023.3 |
| Majority-Black tracts | 0.964 | 2023-05 | 1342 | 2023.3 |
| Majority-Black tracts | 0.953 | 2023-06 | 1342 | 2023.4 |
| Majority-Black tracts | 0.871 | 2023-07 | 1342 | 2023.5 |
| Majority-Black tracts | 0.974 | 2023-08 | 1342 | 2023.6 |
| Majority-Black tracts | 0.916 | 2023-09 | 1342 | 2023.7 |
| Majority-Black tracts | 0.999 | 2023-10 | 1342 | 2023.8 |
| Majority-Black tracts | 1.01 | 2023-11 | 1342 | 2023.8 |
| Majority-Black tracts | 0.965 | 2023-12 | 1342 | 2023.9 |
| Majority-Black tracts | 0.972 | 2024-01 | 1342 | 2024 |
| Majority-Black tracts | 1.13 | 2024-02 | 1342 | 2024.1 |
| Majority-Black tracts | 0.920 | 2024-03 | 1342 | 2024.2 |
| Majority-Black tracts | 1.01 | 2024-04 | 1342 | 2024.3 |
| Majority-Black tracts | 0.906 | 2024-05 | 1342 | 2024.3 |
| Majority-Black tracts | 0.880 | 2024-06 | 1342 | 2024.4 |
| Majority-Black tracts | 0.962 | 2024-07 | 1342 | 2024.5 |
| Majority-Black tracts | 0.868 | 2024-08 | 1342 | 2024.6 |
| Majority-Black tracts | 0.841 | 2024-09 | 1342 | 2024.7 |
| Majority-Black tracts | 0.833 | 2024-10 | 1342 | 2024.8 |
| Majority-Black tracts | 0.919 | 2024-11 | 1342 | 2024.8 |
| Majority-Black tracts | 0.933 | 2024-12 | 1342 | 2024.9 |
| Majority-Black tracts | 0.903 | 2025-01 | 1342 | 2025 |
| Majority-Black tracts | 0.987 | 2025-02 | 1342 | 2025.1 |
| Majority-Black tracts | 0.992 | 2025-03 | 1342 | 2025.2 |
| Majority-Black tracts | 0.910 | 2025-04 | 1342 | 2025.3 |
| Majority-Black tracts | 0.875 | 2025-05 | 1342 | 2025.3 |
| Majority-Black tracts | 0.853 | 2025-06 | 1342 | 2025.4 |
| Majority-Black tracts | 0.958 | 2025-07 | 1342 | 2025.5 |
| Majority-Black tracts | 0.834 | 2025-08 | 1342 | 2025.6 |
| Majority-Black tracts | 0.883 | 2025-09 | 1342 | 2025.7 |
| Majority-Black tracts | 0.902 | 2025-10 | 1342 | 2025.8 |
| Majority-Black tracts | 0.798 | 2025-11 | 1342 | 2025.8 |
| Majority-Black tracts | 0.954 | 2025-12 | 1342 | 2025.9 |
| Majority-Hispanic tracts | 0.987 | 2020-01 | 1591 | 2020 |
| Majority-Hispanic tracts | 1.10 | 2020-02 | 1591 | 2020.1 |
| Majority-Hispanic tracts | 0.662 | 2020-03 | 1591 | 2020.2 |
| Majority-Hispanic tracts | 0.118 | 2020-04 | 1591 | 2020.3 |
| Majority-Hispanic tracts | 0.171 | 2020-05 | 1591 | 2020.3 |
| Majority-Hispanic tracts | 0.310 | 2020-06 | 1591 | 2020.4 |
| Majority-Hispanic tracts | 0.292 | 2020-07 | 1591 | 2020.5 |
| Majority-Hispanic tracts | 0.431 | 2020-08 | 1591 | 2020.6 |
| Majority-Hispanic tracts | 0.478 | 2020-09 | 1591 | 2020.7 |
| Majority-Hispanic tracts | 0.517 | 2020-10 | 1591 | 2020.8 |
| Majority-Hispanic tracts | 0.628 | 2020-11 | 1591 | 2020.8 |
| Majority-Hispanic tracts | 0.574 | 2020-12 | 1591 | 2020.9 |
| Majority-Hispanic tracts | 0.492 | 2021-01 | 1591 | 2021 |
| Majority-Hispanic tracts | 0.471 | 2021-02 | 1591 | 2021.1 |
| Majority-Hispanic tracts | 0.700 | 2021-03 | 1591 | 2021.2 |
| Majority-Hispanic tracts | 0.580 | 2021-04 | 1591 | 2021.3 |
| Majority-Hispanic tracts | 0.492 | 2021-05 | 1591 | 2021.3 |
| Majority-Hispanic tracts | 0.579 | 2021-06 | 1591 | 2021.4 |
| Majority-Hispanic tracts | 0.593 | 2021-07 | 1591 | 2021.5 |
| Majority-Hispanic tracts | 0.597 | 2021-08 | 1591 | 2021.6 |
| Majority-Hispanic tracts | 0.762 | 2021-09 | 1591 | 2021.7 |
| Majority-Hispanic tracts | 0.816 | 2021-10 | 1591 | 2021.8 |
| Majority-Hispanic tracts | 0.833 | 2021-11 | 1591 | 2021.8 |
| Majority-Hispanic tracts | 0.880 | 2021-12 | 1591 | 2021.9 |
| Majority-Hispanic tracts | 0.885 | 2022-01 | 1591 | 2022 |
| Majority-Hispanic tracts | 1.16 | 2022-02 | 1591 | 2022.1 |
| Majority-Hispanic tracts | 1.40 | 2022-03 | 1591 | 2022.2 |
| Majority-Hispanic tracts | 1.12 | 2022-04 | 1591 | 2022.3 |
| Majority-Hispanic tracts | 1.01 | 2022-05 | 1591 | 2022.3 |
| Majority-Hispanic tracts | 1.01 | 2022-06 | 1591 | 2022.4 |
| Majority-Hispanic tracts | 1.10 | 2022-07 | 1591 | 2022.5 |
| Majority-Hispanic tracts | 1.10 | 2022-08 | 1591 | 2022.6 |
| Majority-Hispanic tracts | 1.18 | 2022-09 | 1591 | 2022.7 |
| Majority-Hispanic tracts | 1.07 | 2022-10 | 1591 | 2022.8 |
| Majority-Hispanic tracts | 1.13 | 2022-11 | 1591 | 2022.8 |
| Majority-Hispanic tracts | 1.22 | 2022-12 | 1591 | 2022.9 |
| Majority-Hispanic tracts | 1.08 | 2023-01 | 1591 | 2023 |
| Majority-Hispanic tracts | 1.19 | 2023-02 | 1591 | 2023.1 |
| Majority-Hispanic tracts | 1.35 | 2023-03 | 1591 | 2023.2 |
| Majority-Hispanic tracts | 1.17 | 2023-04 | 1591 | 2023.3 |
| Majority-Hispanic tracts | 1.17 | 2023-05 | 1591 | 2023.3 |
| Majority-Hispanic tracts | 1.01 | 2023-06 | 1591 | 2023.4 |
| Majority-Hispanic tracts | 1.06 | 2023-07 | 1591 | 2023.5 |
| Majority-Hispanic tracts | 1.09 | 2023-08 | 1591 | 2023.6 |
| Majority-Hispanic tracts | 1.05 | 2023-09 | 1591 | 2023.7 |
| Majority-Hispanic tracts | 1.07 | 2023-10 | 1591 | 2023.8 |
| Majority-Hispanic tracts | 1.08 | 2023-11 | 1591 | 2023.8 |
| Majority-Hispanic tracts | 1.05 | 2023-12 | 1591 | 2023.9 |
| Majority-Hispanic tracts | 1.06 | 2024-01 | 1591 | 2024 |
| Majority-Hispanic tracts | 1.21 | 2024-02 | 1591 | 2024.1 |
| Majority-Hispanic tracts | 1.09 | 2024-03 | 1591 | 2024.2 |
| Majority-Hispanic tracts | 1.13 | 2024-04 | 1591 | 2024.3 |
| Majority-Hispanic tracts | 1.04 | 2024-05 | 1591 | 2024.3 |
| Majority-Hispanic tracts | 0.916 | 2024-06 | 1591 | 2024.4 |
| Majority-Hispanic tracts | 1.07 | 2024-07 | 1591 | 2024.5 |
| Majority-Hispanic tracts | 0.987 | 2024-08 | 1591 | 2024.6 |
| Majority-Hispanic tracts | 1.05 | 2024-09 | 1591 | 2024.7 |
| Majority-Hispanic tracts | 0.963 | 2024-10 | 1591 | 2024.8 |
| Majority-Hispanic tracts | 0.987 | 2024-11 | 1591 | 2024.8 |
| Majority-Hispanic tracts | 1.04 | 2024-12 | 1591 | 2024.9 |
| Majority-Hispanic tracts | 0.993 | 2025-01 | 1591 | 2025 |
| Majority-Hispanic tracts | 1.06 | 2025-02 | 1591 | 2025.1 |
| Majority-Hispanic tracts | 1.12 | 2025-03 | 1591 | 2025.2 |
| Majority-Hispanic tracts | 0.993 | 2025-04 | 1591 | 2025.3 |
| Majority-Hispanic tracts | 1.01 | 2025-05 | 1591 | 2025.3 |
| Majority-Hispanic tracts | 0.865 | 2025-06 | 1591 | 2025.4 |
| Majority-Hispanic tracts | 0.988 | 2025-07 | 1591 | 2025.5 |
| Majority-Hispanic tracts | 0.886 | 2025-08 | 1591 | 2025.6 |
| Majority-Hispanic tracts | 1.00 | 2025-09 | 1591 | 2025.7 |
| Majority-Hispanic tracts | 0.937 | 2025-10 | 1591 | 2025.8 |
| Majority-Hispanic tracts | 0.898 | 2025-11 | 1591 | 2025.8 |
| Majority-Hispanic tracts | 0.997 | 2025-12 | 1591 | 2025.9 |
| No majority or other group | 1.06 | 2020-01 | 2155 | 2020 |
| No majority or other group | 1.14 | 2020-02 | 2155 | 2020.1 |
| No majority or other group | 0.684 | 2020-03 | 2155 | 2020.2 |
| No majority or other group | 0.070 | 2020-04 | 2155 | 2020.3 |
| No majority or other group | 0.112 | 2020-05 | 2155 | 2020.3 |
| No majority or other group | 0.231 | 2020-06 | 2155 | 2020.4 |
| No majority or other group | 0.227 | 2020-07 | 2155 | 2020.5 |
| No majority or other group | 0.383 | 2020-08 | 2155 | 2020.6 |
| No majority or other group | 0.481 | 2020-09 | 2155 | 2020.7 |
| No majority or other group | 0.589 | 2020-10 | 2155 | 2020.8 |
| No majority or other group | 0.735 | 2020-11 | 2155 | 2020.8 |
| No majority or other group | 0.702 | 2020-12 | 2155 | 2020.9 |
| No majority or other group | 0.542 | 2021-01 | 2155 | 2021 |
| No majority or other group | 0.546 | 2021-02 | 2155 | 2021.1 |
| No majority or other group | 0.777 | 2021-03 | 2155 | 2021.2 |
| No majority or other group | 0.541 | 2021-04 | 2155 | 2021.3 |
| No majority or other group | 0.469 | 2021-05 | 2155 | 2021.3 |
| No majority or other group | 0.605 | 2021-06 | 2155 | 2021.4 |
| No majority or other group | 0.605 | 2021-07 | 2155 | 2021.5 |
| No majority or other group | 0.599 | 2021-08 | 2155 | 2021.6 |
| No majority or other group | 0.787 | 2021-09 | 2155 | 2021.7 |
| No majority or other group | 0.830 | 2021-10 | 2155 | 2021.8 |
| No majority or other group | 0.873 | 2021-11 | 2155 | 2021.8 |
| No majority or other group | 0.862 | 2021-12 | 2155 | 2021.9 |
| No majority or other group | 0.934 | 2022-01 | 2155 | 2022 |
| No majority or other group | 1.19 | 2022-02 | 2155 | 2022.1 |
| No majority or other group | 1.43 | 2022-03 | 2155 | 2022.2 |
| No majority or other group | 1.12 | 2022-04 | 2155 | 2022.3 |
| No majority or other group | 1.09 | 2022-05 | 2155 | 2022.3 |
| No majority or other group | 1.16 | 2022-06 | 2155 | 2022.4 |
| No majority or other group | 1.20 | 2022-07 | 2155 | 2022.5 |
| No majority or other group | 1.29 | 2022-08 | 2155 | 2022.6 |
| No majority or other group | 1.21 | 2022-09 | 2155 | 2022.7 |
| No majority or other group | 1.19 | 2022-10 | 2155 | 2022.8 |
| No majority or other group | 1.27 | 2022-11 | 2155 | 2022.8 |
| No majority or other group | 1.32 | 2022-12 | 2155 | 2022.9 |
| No majority or other group | 1.26 | 2023-01 | 2155 | 2023 |
| No majority or other group | 1.37 | 2023-02 | 2155 | 2023.1 |
| No majority or other group | 1.52 | 2023-03 | 2155 | 2023.2 |
| No majority or other group | 1.25 | 2023-04 | 2155 | 2023.3 |
| No majority or other group | 1.33 | 2023-05 | 2155 | 2023.3 |
| No majority or other group | 1.21 | 2023-06 | 2155 | 2023.4 |
| No majority or other group | 1.21 | 2023-07 | 2155 | 2023.5 |
| No majority or other group | 1.25 | 2023-08 | 2155 | 2023.6 |
| No majority or other group | 1.21 | 2023-09 | 2155 | 2023.7 |
| No majority or other group | 1.27 | 2023-10 | 2155 | 2023.8 |
| No majority or other group | 1.25 | 2023-11 | 2155 | 2023.8 |
| No majority or other group | 1.18 | 2023-12 | 2155 | 2023.9 |
| No majority or other group | 1.24 | 2024-01 | 2155 | 2024 |
| No majority or other group | 1.34 | 2024-02 | 2155 | 2024.1 |
| No majority or other group | 1.30 | 2024-03 | 2155 | 2024.2 |
| No majority or other group | 1.25 | 2024-04 | 2155 | 2024.3 |
| No majority or other group | 1.17 | 2024-05 | 2155 | 2024.3 |
| No majority or other group | 1.11 | 2024-06 | 2155 | 2024.4 |
| No majority or other group | 1.21 | 2024-07 | 2155 | 2024.5 |
| No majority or other group | 1.13 | 2024-08 | 2155 | 2024.6 |
| No majority or other group | 1.15 | 2024-09 | 2155 | 2024.7 |
| No majority or other group | 1.03 | 2024-10 | 2155 | 2024.8 |
| No majority or other group | 1.13 | 2024-11 | 2155 | 2024.8 |
| No majority or other group | 1.21 | 2024-12 | 2155 | 2024.9 |
| No majority or other group | 1.17 | 2025-01 | 2155 | 2025 |
| No majority or other group | 1.27 | 2025-02 | 2155 | 2025.1 |
| No majority or other group | 1.29 | 2025-03 | 2155 | 2025.2 |
| No majority or other group | 1.16 | 2025-04 | 2155 | 2025.3 |
| No majority or other group | 1.14 | 2025-05 | 2155 | 2025.3 |
| No majority or other group | 1.06 | 2025-06 | 2155 | 2025.4 |
| No majority or other group | 1.12 | 2025-07 | 2155 | 2025.5 |
| No majority or other group | 1.04 | 2025-08 | 2155 | 2025.6 |
| No majority or other group | 1.11 | 2025-09 | 2155 | 2025.7 |
| No majority or other group | 1.13 | 2025-10 | 2155 | 2025.8 |
| No majority or other group | 1.06 | 2025-11 | 2155 | 2025.8 |
| No majority or other group | 1.12 | 2025-12 | 2155 | 2025.9 |
| Majority-white tracts | 1.01 | 2020-01 | 4919 | 2020 |
| Majority-white tracts | 1.06 | 2020-02 | 4919 | 2020.1 |
| Majority-white tracts | 0.618 | 2020-03 | 4919 | 2020.2 |
| Majority-white tracts | 0.095 | 2020-04 | 4919 | 2020.3 |
| Majority-white tracts | 0.136 | 2020-05 | 4919 | 2020.3 |
| Majority-white tracts | 0.252 | 2020-06 | 4919 | 2020.4 |
| Majority-white tracts | 0.270 | 2020-07 | 4919 | 2020.5 |
| Majority-white tracts | 0.428 | 2020-08 | 4919 | 2020.6 |
| Majority-white tracts | 0.532 | 2020-09 | 4919 | 2020.7 |
| Majority-white tracts | 0.579 | 2020-10 | 4919 | 2020.8 |
| Majority-white tracts | 0.640 | 2020-11 | 4919 | 2020.8 |
| Majority-white tracts | 0.687 | 2020-12 | 4919 | 2020.9 |
| Majority-white tracts | 0.526 | 2021-01 | 4919 | 2021 |
| Majority-white tracts | 0.531 | 2021-02 | 4919 | 2021.1 |
| Majority-white tracts | 0.739 | 2021-03 | 4919 | 2021.2 |
| Majority-white tracts | 0.550 | 2021-04 | 4919 | 2021.3 |
| Majority-white tracts | 0.497 | 2021-05 | 4919 | 2021.3 |
| Majority-white tracts | 0.604 | 2021-06 | 4919 | 2021.4 |
| Majority-white tracts | 0.612 | 2021-07 | 4919 | 2021.5 |
| Majority-white tracts | 0.643 | 2021-08 | 4919 | 2021.6 |
| Majority-white tracts | 0.781 | 2021-09 | 4919 | 2021.7 |
| Majority-white tracts | 0.817 | 2021-10 | 4919 | 2021.8 |
| Majority-white tracts | 0.862 | 2021-11 | 4919 | 2021.8 |
| Majority-white tracts | 0.871 | 2021-12 | 4919 | 2021.9 |
| Majority-white tracts | 0.892 | 2022-01 | 4919 | 2022 |
| Majority-white tracts | 1.04 | 2022-02 | 4919 | 2022.1 |
| Majority-white tracts | 1.19 | 2022-03 | 4919 | 2022.2 |
| Majority-white tracts | 1.09 | 2022-04 | 4919 | 2022.3 |
| Majority-white tracts | 1.04 | 2022-05 | 4919 | 2022.3 |
| Majority-white tracts | 1.11 | 2022-06 | 4919 | 2022.4 |
| Majority-white tracts | 1.06 | 2022-07 | 4919 | 2022.5 |
| Majority-white tracts | 1.20 | 2022-08 | 4919 | 2022.6 |
| Majority-white tracts | 1.21 | 2022-09 | 4919 | 2022.7 |
| Majority-white tracts | 1.23 | 2022-10 | 4919 | 2022.8 |
| Majority-white tracts | 1.26 | 2022-11 | 4919 | 2022.8 |
| Majority-white tracts | 1.30 | 2022-12 | 4919 | 2022.9 |
| Majority-white tracts | 1.26 | 2023-01 | 4919 | 2023 |
| Majority-white tracts | 1.31 | 2023-02 | 4919 | 2023.1 |
| Majority-white tracts | 1.40 | 2023-03 | 4919 | 2023.2 |
| Majority-white tracts | 1.14 | 2023-04 | 4919 | 2023.3 |
| Majority-white tracts | 1.29 | 2023-05 | 4919 | 2023.3 |
| Majority-white tracts | 1.28 | 2023-06 | 4919 | 2023.4 |
| Majority-white tracts | 1.18 | 2023-07 | 4919 | 2023.5 |
| Majority-white tracts | 1.29 | 2023-08 | 4919 | 2023.6 |
| Majority-white tracts | 1.22 | 2023-09 | 4919 | 2023.7 |
| Majority-white tracts | 1.33 | 2023-10 | 4919 | 2023.8 |
| Majority-white tracts | 1.33 | 2023-11 | 4919 | 2023.8 |
| Majority-white tracts | 1.32 | 2023-12 | 4919 | 2023.9 |
| Majority-white tracts | 1.30 | 2024-01 | 4919 | 2024 |
| Majority-white tracts | 1.50 | 2024-02 | 4919 | 2024.1 |
| Majority-white tracts | 1.33 | 2024-03 | 4919 | 2024.2 |
| Majority-white tracts | 1.37 | 2024-04 | 4919 | 2024.3 |
| Majority-white tracts | 1.23 | 2024-05 | 4919 | 2024.3 |
| Majority-white tracts | 1.22 | 2024-06 | 4919 | 2024.4 |
| Majority-white tracts | 1.30 | 2024-07 | 4919 | 2024.5 |
| Majority-white tracts | 1.25 | 2024-08 | 4919 | 2024.6 |
| Majority-white tracts | 1.24 | 2024-09 | 4919 | 2024.7 |
| Majority-white tracts | 1.20 | 2024-10 | 4919 | 2024.8 |
| Majority-white tracts | 1.29 | 2024-11 | 4919 | 2024.8 |
| Majority-white tracts | 1.43 | 2024-12 | 4919 | 2024.9 |
| Majority-white tracts | 1.31 | 2025-01 | 4919 | 2025 |
| Majority-white tracts | 1.40 | 2025-02 | 4919 | 2025.1 |
| Majority-white tracts | 1.34 | 2025-03 | 4919 | 2025.2 |
| Majority-white tracts | 1.35 | 2025-04 | 4919 | 2025.3 |
| Majority-white tracts | 1.29 | 2025-05 | 4919 | 2025.3 |
| Majority-white tracts | 1.23 | 2025-06 | 4919 | 2025.4 |
| Majority-white tracts | 1.32 | 2025-07 | 4919 | 2025.5 |
| Majority-white tracts | 1.22 | 2025-08 | 4919 | 2025.6 |
| Majority-white tracts | 1.29 | 2025-09 | 4919 | 2025.7 |
| Majority-white tracts | 1.31 | 2025-10 | 4919 | 2025.8 |
| Majority-white tracts | 1.23 | 2025-11 | 4919 | 2025.8 |
| Majority-white tracts | 1.38 | 2025-12 | 4919 | 2025.9 |

#### Table 6.2. Filings per renter household

Majority-Black tracts have the highest filing rate per renter household both before the pandemic and in 2023

Eviction filings per 100 renter households per year, by racial majority of the tract, an average year before the pandemic and calendar year 2023

| group | period | rate | tracts |
| --- | --- | ---: | ---: |
| Majority-Black tracts | Average year before the pandemic | 12.72 | 1336 |
| Majority-Black tracts | Calendar year 2023 | 12.50 | 1336 |
| Majority-white tracts | Average year before the pandemic | 4.25 | 4903 |
| Majority-white tracts | Calendar year 2023 | 5.42 | 4903 |
| Majority-Hispanic tracts | Average year before the pandemic | 6.41 | 1588 |
| Majority-Hispanic tracts | Calendar year 2023 | 7.09 | 1588 |
| No majority or other group | Average year before the pandemic | 8.43 | 2060 |
| No majority or other group | Calendar year 2023 | 10.71 | 2060 |

Note: Race is the Eviction Lab's tract classification by the group that makes up a majority of residents in the ACS for 2015 to 2019 (Black and White are non-Hispanic, Hispanic is of any race; the fourth group is another majority or none) and is missing for 236 panel tracts, which are left out of this figure. The index uses each group's own baseline, so it measures change from that group's past and does not compare the groups' filing rates.

Note: Filing rates divide filings in the group's tracts by renter households (ACS 2019-2023 households times one minus the homeownership rate), so the numerator and the denominator describe different years. Filings can repeat at one address, so a rate of filings per household is not a share of households.

Note: Hepburn, Louis and Desmond report the highest filing rates among Black renters in court records for 2012 to 2016, which is consistent with the baseline rates in the second tab.

Source: Eviction Lab Eviction Tracking System, 2020-2025, and ACS 2019-2023 renter households; full counts of court filings, not weighted

Data: https://vishalsingh.org/zipcode/report/renting-affordability-and-eviction/figures/eviction_by_race.json

### Table 7. From 2023 to 2025 filings ran from 0.62 times the baseline in Philadelphia to 1.84 times in Portland, with a median site at 1.11 times

Eviction filings divided by the pre-pandemic baseline, January 2023 to December 2025, by site in the balanced panel of the Eviction Tracking System

| filings_per_year | index | site | tracts |
| ---: | ---: | --- | ---: |
| 13046 | 0.619 | Philadelphia, PA | 408 |
| 5381 | 0.649 | Cleveland, OH | 160 |
| 7559 | 0.673 | Wilmington, DE | 145 |
| 9703 | 0.674 | Charleston, SC | 99 |
| 8672 | 0.800 | Albuquerque, NM | 176 |
| 10719 | 0.830 | Milwaukee, WI | 302 |
| 24977 | 0.866 | Indianapolis, IN | 253 |
| 4366 | 0.891 | New Orleans, LA | 183 |
| 5632 | 0.893 | Providence, RI | 145 |
| 2730 | 0.928 | South Bend, IN | 82 |
| 8094 | 0.954 | Palm Beach | 373 |
| 3885 | 0.975 | Bridgeport, CT | 227 |
| 2817 | 0.983 | Southwest Oregon, OR | 158 |
| 12838 | 1.00 | Cincinnati, OH | 226 |
| 6269 | 1.05 | Hartford, CT | 235 |
| 18730 | 1.07 | Tampa, FL | 610 |
| 16885 | 1.15 | St Louis, MO | 340 |
| 10646 | 1.18 | Kansas City, MO | 227 |
| 48055 | 1.19 | Dallas, TX | 645 |
| 14316 | 1.20 | Jacksonville, FL | 219 |
| 22161 | 1.21 | Columbus, OH | 328 |
| 45697 | 1.23 | Fort Worth, TX | 642 |
| 16262 | 1.26 | Miami | 707 |
| 3152 | 1.27 | Tacoma, WA | 193 |
| 78233 | 1.34 | Houston, TX | 1218 |
| 49301 | 1.35 | Las Vegas, NV | 535 |
| 29057 | 1.37 | Memphis, TN | 249 |
| 2157 | 1.38 | Gainesville, FL | 58 |
| 14062 | 1.38 | Nashville, TN | 174 |
| 2402 | 1.39 | Eugene, OR | 93 |
| 12570 | 1.59 | Minneapolis-Saint Paul, MN | 472 |
| 15492 | 1.84 | Portland, OR | 361 |

Note: The index compares a site with its own past, so it does not rank sites by how many households face eviction. Sites differ in local rules, court practice, rental assistance and how filings are recorded, and these data do not separate those.

Note: The 32 sites are the participating cities and metro areas with a baseline and at least one tract observed in every month (Atlanta, Other Oregon Counties and Greenville, South Carolina, are not). The counts are full counts of filings and carry no sampling interval.

Source: Eviction Lab Eviction Tracking System, 2023-2025, balanced panel of tracts with a baseline; full counts of court filings, not weighted

Data: https://vishalsingh.org/zipcode/report/renting-affordability-and-eviction/figures/eviction_by_city.json

### Table 8. The renter share over the cost-burden line is worse at the more deprived end but uneven along the way under the ADI; it is steadily worse toward the more deprived end under the standardized index and the income ranking

Share paying more than 30% of income (%) by tenth of tracts ranked four ways, over the 75,587 tracts that have all four rankings

| bin | conf_high | conf_low | estimate | n_units | ranking |
| ---: | ---: | ---: | ---: | ---: | --- |
| 1 | 46.19 | 41.35 | 43.97 | 7,550 | Area Deprivation Index |
| 2 | 49.62 | 45.03 | 47.43 | 7,560 | Area Deprivation Index |
| 3 | 47.36 | 44.33 | 45.79 | 7,486 | Area Deprivation Index |
| 4 | 46.30 | 44.30 | 45.31 | 7,757 | Area Deprivation Index |
| 5 | 46.14 | 44.11 | 45.08 | 7,485 | Area Deprivation Index |
| 6 | 45.43 | 43.56 | 44.47 | 7,513 | Area Deprivation Index |
| 7 | 45.11 | 43.33 | 44.24 | 7,658 | Area Deprivation Index |
| 8 | 44.67 | 42.96 | 43.84 | 7,418 | Area Deprivation Index |
| 9 | 44.51 | 42.88 | 43.69 | 7,617 | Area Deprivation Index |
| 10 | 48.00 | 46.26 | 47.12 | 7,543 | Area Deprivation Index |
| 1 | 40.25 | 37.02 | 38.63 | 7,553 | Standardized index |
| 2 | 42.15 | 39.85 | 40.93 | 7,560 | Standardized index |
| 3 | 42.48 | 40.72 | 41.56 | 7,558 | Standardized index |
| 4 | 43.31 | 41.57 | 42.42 | 7,560 | Standardized index |
| 5 | 43.18 | 41.38 | 42.28 | 7,560 | Standardized index |
| 6 | 44.53 | 42.68 | 43.56 | 7,559 | Standardized index |
| 7 | 45.67 | 43.90 | 44.78 | 7,560 | Standardized index |
| 8 | 47.56 | 45.54 | 46.56 | 7,557 | Standardized index |
| 9 | 49.89 | 48.16 | 49.03 | 7,560 | Standardized index |
| 10 | 54.30 | 52.69 | 53.51 | 7,560 | Standardized index |
| 1 | 39.11 | 36.41 | 37.69 | 7,555 | Median household income |
| 2 | 41.93 | 39.76 | 40.79 | 7,557 | Median household income |
| 3 | 43.34 | 40.95 | 42.13 | 7,559 | Median household income |
| 4 | 44.13 | 41.67 | 42.85 | 7,560 | Median household income |
| 5 | 44.36 | 42.13 | 43.26 | 7,559 | Median household income |
| 6 | 44.89 | 42.54 | 43.67 | 7,560 | Median household income |
| 7 | 45.99 | 43.84 | 44.90 | 7,554 | Median household income |
| 8 | 47.91 | 45.65 | 46.79 | 7,564 | Median household income |
| 9 | 49.65 | 47.59 | 48.59 | 7,560 | Median household income |
| 10 | 54.09 | 52.62 | 53.36 | 7,559 | Median household income |
| 1 | 40.92 | 38.48 | 39.58 | 7,554 | Bachelor's degree share |
| 2 | 44.44 | 42.86 | 43.65 | 7,556 | Bachelor's degree share |
| 3 | 46.05 | 44.25 | 45.13 | 7,564 | Bachelor's degree share |
| 4 | 46.54 | 44.93 | 45.71 | 7,560 | Bachelor's degree share |
| 5 | 46.67 | 44.96 | 45.83 | 7,553 | Bachelor's degree share |
| 6 | 46.80 | 44.84 | 45.80 | 7,558 | Bachelor's degree share |
| 7 | 46.32 | 44.35 | 45.33 | 7,561 | Bachelor's degree share |
| 8 | 46.28 | 44.30 | 45.30 | 7,569 | Bachelor's degree share |
| 9 | 47.49 | 45.32 | 46.43 | 7,548 | Bachelor's degree share |
| 10 | 50.56 | 48.03 | 49.33 | 7,564 | Bachelor's degree share |

Note: Least and most deprived tenth: ADI 44.0% and 47.1%, rank correlation −0.30; standardized index 38.6% and 53.5%, +0.99; median household income 37.7% and 53.4%, +1.00; bachelor's degree share 39.6% and 49.3%, +0.79.

Note: The standardized index applies the ADI's coefficients to its measures after standardizing each one, so home value and income carry less weight than in the ADI. Income and degree share are single census measures, each ranked from the highest value to the lowest.

Note: Rent and income are inputs to the ADI and to the standardized index, so this comparison is left out of the book's count of changed gradients.

Source: HUD CHAS (American Community Survey 2018-2022 special tabulation) and Area Deprivation Index 2023 (built on ACS 2019-2023); shares pooled over tracts and weighted by households, bootstrap over counties, survey-based estimates with no margin of error published for the shares; ADI 2023 and ACS 2019-2023 tract tables (B19013, B15003 and the 15 rebuilt ADI measures)

Data: https://vishalsingh.org/zipcode/report/renting-affordability-and-eviction/figures/alt_rankings.json

## All estimates

Every fact in facts.json (160), sorted by id.

| Fact id | Value | Unit | Estimand | Population | Period | Interval | n | Weights | Source | Note |
| --- | ---: | --- | --- | --- | --- | --- | ---: | --- | --- | --- |
| adi_li_own_d1 | 65.9% | percent of low-income owner households with housing costs above 30% of income | low-income households (at or below 80 percent of HAMFI) with a burden over low-income households of the tenure, pooled over tracts in the tenth, bootstrap over counties | low-income owner households in tracts with an Area Deprivation Index, least deprived tenth | 2018 to 2022; Area Deprivation Index 2023 | 95% CI 64.0% to 68.0% | 8,103 tracts | low-income households of the tenure | HUD CHAS housing affordability |   |
| adi_li_own_d10 | 37.6% | percent of low-income owner households with housing costs above 30% of income | low-income households (at or below 80 percent of HAMFI) with a burden over low-income households of the tenure, pooled over tracts in the tenth, bootstrap over counties | low-income owner households in tracts with an Area Deprivation Index, most deprived tenth | 2018 to 2022; Area Deprivation Index 2023 | 95% CI 37.0% to 38.1% | 8,206 tracts | low-income households of the tenure | HUD CHAS housing affordability |   |
| adi_li_ren_d1 | 75.3% | percent of low-income renter households with housing costs above 30% of income | low-income households (at or below 80 percent of HAMFI) with a burden over low-income households of the tenure, pooled over tracts in the tenth, bootstrap over counties | low-income renter households in tracts with an Area Deprivation Index, least deprived tenth | 2018 to 2022; Area Deprivation Index 2023 | 95% CI 72.6% to 78.1% | 7,840 tracts | low-income households of the tenure | HUD CHAS housing affordability |   |
| adi_li_ren_d10 | 62.5% | percent of low-income renter households with housing costs above 30% of income | low-income households (at or below 80 percent of HAMFI) with a burden over low-income households of the tenure, pooled over tracts in the tenth, bootstrap over counties | low-income renter households in tracts with an Area Deprivation Index, most deprived tenth | 2018 to 2022; Area Deprivation Index 2023 | 95% CI 61.6% to 63.3% | 8,194 tracts | low-income households of the tenure | HUD CHAS housing affordability |   |
| adi_own_d1 | 25.5% | percent of owner households with housing costs above 30% of income | pooled share of the tenure's households with the burden over tracts in the tenth, bootstrap over counties | owner households in tracts with an Area Deprivation Index (tenths hold near-equal numbers of tracts; tracts with equal ranks stay together), least deprived tenth | 2018 to 2022; Area Deprivation Index 2023 | 95% CI 23.7% to 27.2% | 8,189 tracts | households of the tenure with a computed burden | HUD CHAS housing affordability |   |
| adi_own_d10 | 19.8% | percent of owner households with housing costs above 30% of income | pooled share of the tenure's households with the burden over tracts in the tenth, bootstrap over counties | owner households in tracts with an Area Deprivation Index (tenths hold near-equal numbers of tracts; tracts with equal ranks stay together), most deprived tenth | 2018 to 2022; Area Deprivation Index 2023 | 95% CI 19.3% to 20.3% | 8,212 tracts | households of the tenure with a computed burden | HUD CHAS housing affordability |   |
| adi_own_d9 | 18.3% | percent of owner households with housing costs above 30% of income | pooled share of the tenure's households with the burden over tracts in the tenth, bootstrap over counties | owner households in tracts with an Area Deprivation Index (tenths hold near-equal numbers of tracts; tracts with equal ranks stay together), ninth tenth | 2018 to 2022; Area Deprivation Index 2023 | 95% CI 17.9% to 18.7% | 8,329 tracts | households of the tenure with a computed burden | HUD CHAS housing affordability |   |
| adi_ren_d1 | 44.4% | percent of renter households with housing costs above 30% of income | pooled share of the tenure's households with the burden over tracts in the tenth, bootstrap over counties | renter households in tracts with an Area Deprivation Index (tenths hold near-equal numbers of tracts; tracts with equal ranks stay together), least deprived tenth | 2018 to 2022; Area Deprivation Index 2023 | 95% CI 41.9% to 46.7% | 8,208 tracts | households of the tenure with a computed burden | HUD CHAS housing affordability |   |
| adi_ren_d10 | 47.6% | percent of renter households with housing costs above 30% of income | pooled share of the tenure's households with the burden over tracts in the tenth, bootstrap over counties | renter households in tracts with an Area Deprivation Index (tenths hold near-equal numbers of tracts; tracts with equal ranks stay together), most deprived tenth | 2018 to 2022; Area Deprivation Index 2023 | 95% CI 46.6% to 48.4% | 8,207 tracts | households of the tenure with a computed burden | HUD CHAS housing affordability |   |
| adi_ren_range | 3.6 | percentage points, largest minus smallest renter share with housing costs above 30% of income among the ten tenths | largest minus smallest of the ten tenth means (a range of estimates, no interval) | renter households in tracts with an Area Deprivation Index (tenths hold near-equal numbers of tracts; tracts with equal ranks stay together) | 2018 to 2022; Area Deprivation Index 2023 |   | 10 tenths of tracts | households of the tenure with a computed burden | HUD CHAS housing affordability |   |
| adi_step | 10 | percentile points of the Area Deprivation Index national rank | the step in the national percentile rank for which a slope is stated | tracts with an Area Deprivation Index | Area Deprivation Index 2023 |   |   | none (definition) | Area Deprivation Index |   |
| alt_adi_least | 44.0% | Share paying more than 30% of income (%) | pooled share of renter households with a computed burden whose housing costs are above the burden line; 95% intervals from a bootstrap over counties | the least deprived tenth of tracts ranked by Area Deprivation Index, among the tracts with all four rankings | 2018 to 2022 | 95% CI 41.4% to 46.2% | 7,550 tracts | none (pooled household counts) | HUD CHAS housing affordability |   |
| alt_adi_most | 47.1% | Share paying more than 30% of income (%) | pooled share of renter households with a computed burden whose housing costs are above the burden line; 95% intervals from a bootstrap over counties | the most deprived tenth of tracts ranked by Area Deprivation Index, among the tracts with all four rankings | 2018 to 2022 | 95% CI 46.3% to 48.0% | 7,543 tracts | none (pooled household counts) | HUD CHAS housing affordability |   |
| alt_adi_rho | −0.30 | rank correlation | Spearman rank correlation between bin order and bin mean, each bin counted once | the bins of tracts ranked by Area Deprivation Index | 2018 to 2022 |   |   | each bin counted once | HUD CHAS housing affordability |   |
| alt_degree_least | 39.6% | Share paying more than 30% of income (%) | pooled share of renter households with a computed burden whose housing costs are above the burden line; 95% intervals from a bootstrap over counties | the least deprived tenth of tracts ranked by Bachelor's degree share, among the tracts with all four rankings | 2018 to 2022 | 95% CI 38.5% to 40.9% | 7,554 tracts | none (pooled household counts) | HUD CHAS housing affordability |   |
| alt_degree_most | 49.3% | Share paying more than 30% of income (%) | pooled share of renter households with a computed burden whose housing costs are above the burden line; 95% intervals from a bootstrap over counties | the most deprived tenth of tracts ranked by Bachelor's degree share, among the tracts with all four rankings | 2018 to 2022 | 95% CI 48.0% to 50.6% | 7,564 tracts | none (pooled household counts) | HUD CHAS housing affordability |   |
| alt_degree_rho | +0.79 | rank correlation | Spearman rank correlation between bin order and bin mean, each bin counted once | the bins of tracts ranked by Bachelor's degree share | 2018 to 2022 |   |   | each bin counted once | HUD CHAS housing affordability |   |
| alt_income_least | 37.7% | Share paying more than 30% of income (%) | pooled share of renter households with a computed burden whose housing costs are above the burden line; 95% intervals from a bootstrap over counties | the least deprived tenth of tracts ranked by Median household income, among the tracts with all four rankings | 2018 to 2022 | 95% CI 36.4% to 39.1% | 7,555 tracts | none (pooled household counts) | HUD CHAS housing affordability |   |
| alt_income_most | 53.4% | Share paying more than 30% of income (%) | pooled share of renter households with a computed burden whose housing costs are above the burden line; 95% intervals from a bootstrap over counties | the most deprived tenth of tracts ranked by Median household income, among the tracts with all four rankings | 2018 to 2022 | 95% CI 52.6% to 54.1% | 7,559 tracts | none (pooled household counts) | HUD CHAS housing affordability |   |
| alt_income_rho | +1.00 | rank correlation | Spearman rank correlation between bin order and bin mean, each bin counted once | the bins of tracts ranked by Median household income | 2018 to 2022 |   |   | each bin counted once | HUD CHAS housing affordability |   |
| alt_n_places | 75,587 | tracts | count of tracts in the common sample of the ranking comparison | tracts with the outcome and all four rankings | 2018 to 2022 |   |   | none (full count) | HUD CHAS housing affordability |   |
| alt_rho_cut | 0.9 | absolute rank correlation | threshold between a steady and an uneven gradient | the rule for changed gradients (docs/RANKING_SWITCH_RULE.md) | 2018 to 2022 |   |   | none (a fixed rule) | HUD CHAS housing affordability |   |
| alt_rho_margin | 0.1 | absolute rank correlation | smallest move in the rank correlation that counts as a change of shape | the rule for changed gradients (docs/RANKING_SWITCH_RULE.md) | 2018 to 2022 |   |   | none (a fixed rule) | HUD CHAS housing affordability |   |
| alt_std_least | 38.6% | Share paying more than 30% of income (%) | pooled share of renter households with a computed burden whose housing costs are above the burden line; 95% intervals from a bootstrap over counties | the least deprived tenth of tracts ranked by Standardized index, among the tracts with all four rankings | 2018 to 2022 | 95% CI 37.0% to 40.3% | 7,553 tracts | none (pooled household counts) | HUD CHAS housing affordability |   |
| alt_std_most | 53.5% | Share paying more than 30% of income (%) | pooled share of renter households with a computed burden whose housing costs are above the burden line; 95% intervals from a bootstrap over counties | the most deprived tenth of tracts ranked by Standardized index, among the tracts with all four rankings | 2018 to 2022 | 95% CI 52.7% to 54.3% | 7,560 tracts | none (pooled household counts) | HUD CHAS housing affordability |   |
| alt_std_rho | +0.99 | rank correlation | Spearman rank correlation between bin order and bin mean, each bin counted once | the bins of tracts ranked by Standardized index | 2018 to 2022 |   |   | each bin counted once | HUD CHAS housing affordability |   |
| band_gap_1 | +1.2 | percentage points, renter share minus owner share with housing costs above 30% of income, households at or below 30 percent of HAMFI | ratio of sums over counties: renter households in the income band minus owner households in the income band with a burden above 30 percent over households in the band whose burden HUD computes; bootstrap over counties. From the raw HUD CHAS county Table 8 | households at or below 30 percent of HAMFI with a computed burden | 2018 to 2022 | 95% CI 0.3 to 1.9 | 3,144 counties | none (pooled household counts) | HUD CHAS housing affordability |   |
| band_gap_2 | +24.4 | percentage points, renter share minus owner share with housing costs above 30% of income, households above 30 and up to 50 percent of HAMFI | ratio of sums over counties: renter households in the income band minus owner households in the income band with a burden above 30 percent over households in the band whose burden HUD computes; bootstrap over counties. From the raw HUD CHAS county Table 8 | households above 30 and up to 50 percent of HAMFI with a computed burden | 2018 to 2022 | 95% CI 23.4 to 25.3 | 3,144 counties | none (pooled household counts) | HUD CHAS housing affordability |   |
| band_gap_3 | +12.0 | percentage points, renter share minus owner share with housing costs above 30% of income, households above 50 and up to 80 percent of HAMFI | ratio of sums over counties: renter households in the income band minus owner households in the income band with a burden above 30 percent over households in the band whose burden HUD computes; bootstrap over counties. From the raw HUD CHAS county Table 8 | households above 50 and up to 80 percent of HAMFI with a computed burden | 2018 to 2022 | 95% CI 10.8 to 13.2 | 3,144 counties | none (pooled household counts) | HUD CHAS housing affordability |   |
| band_gap_4 | −0.5 | percentage points, renter share minus owner share with housing costs above 30% of income, households above 80 and up to 100 percent of HAMFI | ratio of sums over counties: renter households in the income band minus owner households in the income band with a burden above 30 percent over households in the band whose burden HUD computes; bootstrap over counties. From the raw HUD CHAS county Table 8 | households above 80 and up to 100 percent of HAMFI with a computed burden | 2018 to 2022 | 95% CI −1.4 to 0.4 | 3,144 counties | none (pooled household counts) | HUD CHAS housing affordability |   |
| band_gap_5 | −0.2 | percentage points, renter share minus owner share with housing costs above 30% of income, households above 100 percent of HAMFI | ratio of sums over counties: renter households in the income band minus owner households in the income band with a burden above 30 percent over households in the band whose burden HUD computes; bootstrap over counties. From the raw HUD CHAS county Table 8 | households above 100 percent of HAMFI with a computed burden | 2018 to 2022 | 95% CI −0.6 to 0.2 | 3,144 counties | none (pooled household counts) | HUD CHAS housing affordability |   |
| band_owner_1 | 82.8% | percent of owner households with housing costs above 30% of income | ratio of sums over counties: households in the income band with a burden above 30 percent over households in the band whose burden HUD computes; bootstrap over counties. From the raw HUD CHAS county Table 8 | owner households at or below 30 percent of HAMFI with a computed burden | 2018 to 2022 | 95% CI 82.2% to 83.4% | 3,144 counties | none (pooled household counts) | HUD CHAS housing affordability |   |
| band_owner_5 | 5.5% | percent of owner households with housing costs above 30% of income | ratio of sums over counties: households in the income band with a burden above 30 percent over households in the band whose burden HUD computes; bootstrap over counties. From the raw HUD CHAS county Table 8 | owner households above 100 percent of HAMFI with a computed burden | 2018 to 2022 | 95% CI 5.1% to 5.9% | 3,144 counties | none (pooled household counts) | HUD CHAS housing affordability |   |
| band_renter_1 | 83.9% | percent of renter households with housing costs above 30% of income | ratio of sums over counties: households in the income band with a burden above 30 percent over households in the band whose burden HUD computes; bootstrap over counties. From the raw HUD CHAS county Table 8 | renter households at or below 30 percent of HAMFI with a computed burden | 2018 to 2022 | 95% CI 83.3% to 84.6% | 3,144 counties | none (pooled household counts) | HUD CHAS housing affordability |   |
| band_renter_5 | 5.3% | percent of renter households with housing costs above 30% of income | ratio of sums over counties: households in the income band with a burden above 30 percent over households in the band whose burden HUD computes; bootstrap over counties. From the raw HUD CHAS county Table 8 | renter households above 100 percent of HAMFI with a computed burden | 2018 to 2022 | 95% CI 4.8% to 5.9% | 3,144 counties | none (pooled household counts) | HUD CHAS housing affordability |   |
| bur_gap | 24.8 | percentage points, renter share minus owner share with housing costs above 30% of income | pooled renter share with a burden minus the pooled owner share with a burden; bootstrap over counties | households in the tracts of the 50 states and DC (Puerto Rico left out) | 2018 to 2022 | 95% CI 24.2 to 25.3 | 84,410 tracts | none (pooled household counts) | HUD CHAS housing affordability |   |
| bur_gap_std | 3.8 | percentage points, standardized renter share minus owner share with housing costs above 30% of income | the renter burden at the owners' mix of five income bands minus the owner burden (a reweighting of the renters' rates, not a causal decomposition); bootstrap over counties. From the raw HUD CHAS county Table 8 | households with a computed burden in the counties of the 50 states and DC (Connecticut by planning region) | 2018 to 2022 | 95% CI 3.3 to 4.2 | 3,144 counties | none (pooled household counts) | HUD CHAS housing affordability |   |
| bur_owner | 20.8% | percent of owner households with housing costs above 30% of income | ratio of sums over tracts: households with the burden over households whose burden HUD computes, pooled over the tracts of the 50 states and DC; interval from a bootstrap over counties (replicates and seed in the methods note). The share carries no margin of error of its own: HUD rounds every cell independently; the count includes the households with a severe burden | owner households in the tracts of the 50 states and DC with a computed burden | 2018 to 2022 | 95% CI 20.1% to 21.5% | 84,410 tracts | none (pooled household counts) | HUD CHAS housing affordability |   |
| bur_ratio | 2.2 | ratio of the renter share to the owner share with housing costs above 30% of income | pooled renter share with a burden over the pooled owner share with a burden; bootstrap over counties | households in the tracts of the 50 states and DC (Puerto Rico left out) | 2018 to 2022 | 95% CI 2.1 to 2.2 | 84,410 tracts | none (pooled household counts) | HUD CHAS housing affordability |   |
| bur_renter | 45.6% | percent of renter households with housing costs above 30% of income | ratio of sums over tracts: households with the burden over households whose burden HUD computes, pooled over the tracts of the 50 states and DC; interval from a bootstrap over counties (replicates and seed in the methods note). The share carries no margin of error of its own: HUD rounds every cell independently; the count includes the households with a severe burden | renter households in the tracts of the 50 states and DC with a computed burden | 2018 to 2022 | 95% CI 44.7% to 46.5% | 84,410 tracts | none (pooled household counts) | HUD CHAS housing affordability |   |
| bur_renter_std | 24.6% | percent of renter households with housing costs above 30% of income, at the owners' mix of five income bands | the sum over the five income bands of the owners' share of households with a computed burden in the band times the renters' burden rate in the band (direct standardization); bootstrap over counties. From the raw HUD CHAS county Table 8 | households with a computed burden in the counties of the 50 states and DC (Connecticut by planning region) | 2018 to 2022 | 95% CI 23.7% to 25.4% | 3,144 counties | none (pooled household counts) | HUD CHAS housing affordability |   |
| burden_line | 30% | percent of household income | HUD's line for a housing cost burden: housing costs above this share of household income | households in the HUD CHAS tabulation | 2018 to 2022 |   |   | none (definition) | HUD CHAS housing affordability |   |
| cty_floor | 500 | households with a computed burden | minimum number of renter households and of owner households with a computed burden for a county to enter the county comparisons and the map | counties in the county comparisons | 2018 to 2022 |   |   | none (inclusion rule) | HUD CHAS housing affordability |   |
| cty_floor_hi | 1,000 | households with a computed burden | an alternative minimum number of renter households and of owner households with a computed burden | counties in the sensitivity of the county comparisons to the household floor | 2018 to 2022 |   |   | none (inclusion rule) | HUD CHAS housing affordability |   |
| cty_floor_lo | 100 | households with a computed burden | an alternative minimum number of renter households and of owner households with a computed burden | counties in the sensitivity of the county comparisons to the household floor | 2018 to 2022 |   |   | none (inclusion rule) | HUD CHAS housing affordability |   |
| cty_median | 37.9% | percent of renter households with housing costs above 30% of income | median over counties of the county's renter share with a burden (each county counts once) | counties of the 50 states and DC (Connecticut as its eight former counties) with at least 500 renter and 500 owner households with a computed burden | 2018 to 2022 |   | 2,753 counties | none (each county counts once) | HUD CHAS housing affordability |   |
| cty_ok_renters | 99.8% | percent of renter households with a computed burden | renter households with a computed burden in the counties that pass the floor over those in all counties | renter households in the counties of the 50 states and DC | 2018 to 2022 |   | 2,753 counties | none (survey-weighted household counts) | HUD CHAS housing affordability |   |
| cty_p5 | 23.0% | percent of renter households with housing costs above 30% of income | 5th percentile over counties of the county's renter share with a burden (each county counts once) | counties of the 50 states and DC (Connecticut as its eight former counties) with at least 500 renter and 500 owner households with a computed burden | 2018 to 2022 |   | 2,753 counties | none (each county counts once) | HUD CHAS housing affordability |   |
| cty_p5_hi | 25.9% | percent of renter households with housing costs above 30% of income | 5th percentile over counties of the county's renter share with a burden under the alternative household floor | counties of the 50 states and DC (Connecticut as its eight former counties) with at least 1000 renter and 1000 owner households with a computed burden | 2018 to 2022 |   | 2,323 counties | none (each county counts once) | HUD CHAS housing affordability |   |
| cty_p5_lo | 19.2% | percent of renter households with housing costs above 30% of income | 5th percentile over counties of the county's renter share with a burden under the alternative household floor | counties of the 50 states and DC (Connecticut as its eight former counties) with at least 100 renter and 100 owner households with a computed burden | 2018 to 2022 |   | 3,103 counties | none (each county counts once) | HUD CHAS housing affordability |   |
| cty_p95 | 50.9% | percent of renter households with housing costs above 30% of income | 95th percentile over counties of the county's renter share with a burden (each county counts once) | counties of the 50 states and DC (Connecticut as its eight former counties) with at least 500 renter and 500 owner households with a computed burden | 2018 to 2022 |   | 2,753 counties | none (each county counts once) | HUD CHAS housing affordability |   |
| dev_own_d1 | −1.2 | percentage points of the owner share with housing costs above 30% of income, tract minus its county | weighted mean over the tenth's tracts of the tract's share minus the county's weighted mean share across its tracts with an index; bootstrap over counties | owner households in tracts with an Area Deprivation Index (tenths hold near-equal numbers of tracts; tracts with equal ranks stay together), least deprived tenth | 2018 to 2022; Area Deprivation Index 2023 | 95% CI −1.6 to −0.9 | 8,189 tracts | households of the tenure with a computed burden | HUD CHAS housing affordability |   |
| dev_own_d10 | +2.0 | percentage points of the owner share with housing costs above 30% of income, tract minus its county | weighted mean over the tenth's tracts of the tract's share minus the county's weighted mean share across its tracts with an index; bootstrap over counties | owner households in tracts with an Area Deprivation Index (tenths hold near-equal numbers of tracts; tracts with equal ranks stay together), most deprived tenth | 2018 to 2022; Area Deprivation Index 2023 | 95% CI 1.7 to 2.3 | 8,212 tracts | households of the tenure with a computed burden | HUD CHAS housing affordability |   |
| dev_ren_d1 | −3.1 | percentage points of the renter share with housing costs above 30% of income, tract minus its county | weighted mean over the tenth's tracts of the tract's share minus the county's weighted mean share across its tracts with an index; bootstrap over counties | renter households in tracts with an Area Deprivation Index (tenths hold near-equal numbers of tracts; tracts with equal ranks stay together), least deprived tenth | 2018 to 2022; Area Deprivation Index 2023 | 95% CI −4.0 to −2.3 | 8,208 tracts | households of the tenure with a computed burden | HUD CHAS housing affordability |   |
| dev_ren_d10 | +5.0 | percentage points of the renter share with housing costs above 30% of income, tract minus its county | weighted mean over the tenth's tracts of the tract's share minus the county's weighted mean share across its tracts with an index; bootstrap over counties | renter households in tracts with an Area Deprivation Index (tenths hold near-equal numbers of tracts; tracts with equal ranks stay together), most deprived tenth | 2018 to 2022; Area Deprivation Index 2023 | 95% CI 4.5 to 5.5 | 8,207 tracts | households of the tenure with a computed burden | HUD CHAS housing affordability |   |
| ev_bur_panel | 48.5% | percent of renter households with housing costs above 30% of income | ratio of sums over tracts: households with the burden over households whose burden HUD computes, pooled over the tracts of the 50 states and DC; interval from a bootstrap over counties (replicates and seed in the methods note). The share carries no margin of error of its own: HUD rounds every cell independently | tracts of the eviction filing index (10,242 of them have a CHAS record) | 2018 to 2022 | 95% CI 46.8% to 50.4% | 10,242 tracts | none (pooled household counts) | HUD CHAS housing affordability |   |
| ev_bur_rest | 45.1% | percent of renter households with housing costs above 30% of income | ratio of sums over tracts: households with the burden over households whose burden HUD computes, pooled over the tracts of the 50 states and DC; interval from a bootstrap over counties (replicates and seed in the methods note). The share carries no margin of error of its own: HUD rounds every cell independently | tracts of the 50 states and DC outside the filing index | 2018 to 2022 | 95% CI 44.1% to 46.2% | 74,168 tracts | none (pooled household counts) | HUD CHAS housing affordability |   |
| ev_rent_panel | 41.9% | percent of occupied housing units that are rented | renter households over renter and owner households, pooled over tracts; bootstrap over counties | tracts of the eviction filing index (10,242 of them have a CHAS record) | 2018 to 2022 | 95% CI 39.4% to 43.8% | 10,242 tracts | none (pooled household counts) | HUD CHAS housing affordability |   |
| ev_rent_rest | 34.2% | percent of occupied housing units that are rented | renter households over renter and owner households, pooled over tracts; bootstrap over counties | tracts of the 50 states and DC outside the filing index | 2018 to 2022 | 95% CI 32.6% to 36.0% | 74,168 tracts | none (pooled household counts) | HUD CHAS housing affordability |   |
| ev_share_renters | 14.9% | percent of the nation's renter households | renter households in the tracts of the filing index over renter households in all tracts of the 50 states and DC (the index tracts are those of the Eviction Lab sites, matched on the 2020 tract code) | renter households of the 50 states and DC | 2018 to 2022 |   | 10,242 tracts | none (survey-weighted household counts) | HUD CHAS housing affordability |   |
| fil_adi_d1 | 1.48 | ratio of filings to the same-month baseline (1 = baseline) | filings divided by the pre-pandemic baseline, summed over the tract-months of the group (a ratio of sums) | tracts in the least deprived tenth of panel tracts by Area Deprivation Index (tenths hold equal numbers of tracts) | January 2022 to December 2025 |   | 1,007 tracts | none (full counts of filings) | Eviction Tracking System |   |
| fil_adi_d10 | 0.89 | ratio of filings to the same-month baseline (1 = baseline) | filings divided by the pre-pandemic baseline, summed over the tract-months of the group (a ratio of sums) | tracts in the most deprived tenth of panel tracts by Area Deprivation Index (tenths hold equal numbers of tracts) | January 2022 to December 2025 |   | 1,008 tracts | none (full counts of filings) | Eviction Tracking System |   |
| fil_base_d1 | 11 | baseline eviction filings per tract per year | sum of the pre-pandemic baseline over the tract-months of the group divided by the number of tracts and by the years in the window | tracts in the least deprived tenth of panel tracts by Area Deprivation Index | January 2022 to December 2025 |   | 1,007 tracts | none (full counts of filings) | Eviction Tracking System |   |
| fil_base_d10 | 88 | baseline eviction filings per tract per year | sum of the pre-pandemic baseline over the tract-months of the group divided by the number of tracts and by the years in the window | tracts in the most deprived tenth of panel tracts by Area Deprivation Index | January 2022 to December 2025 |   | 1,008 tracts | none (full counts of filings) | Eviction Tracking System |   |
| fil_black_2023 | 0.98 | ratio of filings to the same-month baseline (1 = baseline) | filings divided by the pre-pandemic baseline, summed over the tract-months of the group (a ratio of sums), all tract-months of 2023 (race is the tract's majority racial or ethnic group of residents in the ACS for 2015 to 2019 as the Eviction Lab classifies it, not the race of any tenant) | tracts classified majority Black by the Eviction Lab, in the balanced panel with a baseline | calendar year 2023 |   | 1,342 tracts | none (full counts of filings) | Eviction Tracking System |   |
| fil_city_median | 1.11 | ratio of filings to the same-month baseline (1 = baseline) | median over sites of the site's index | tracts of the sites in the balanced panel that have a pre-pandemic baseline | January 2023 to December 2025 |   | 32 cities and metro areas | none (each site counts once) | Eviction Tracking System |   |
| fil_city_p5 | 0.66 | ratio of filings to the same-month baseline (1 = baseline) | 5th percentile over sites of the site's index | tracts of the sites in the balanced panel that have a pre-pandemic baseline | January 2023 to December 2025 |   | 32 cities and metro areas | none (each site counts once) | Eviction Tracking System |   |
| fil_city_p95 | 1.48 | ratio of filings to the same-month baseline (1 = baseline) | 95th percentile over sites of the site's index | tracts of the sites in the balanced panel that have a pre-pandemic baseline | January 2023 to December 2025 |   | 32 cities and metro areas | none (each site counts once) | Eviction Tracking System |   |
| fil_cool_idx | 0.62 | ratio of filings to the same-month baseline (1 = baseline) | filings divided by the pre-pandemic baseline, summed over the tract-months of the group (a ratio of sums) | tracts of the site with the lowest index, in the balanced panel with a baseline | January 2023 to December 2025 |   | 408 tracts | none (full counts of filings) | Eviction Tracking System |   |
| fil_hisp_2023 | 1.11 | ratio of filings to the same-month baseline (1 = baseline) | filings divided by the pre-pandemic baseline, summed over the tract-months of the group (a ratio of sums), all tract-months of 2023 (race is the tract's majority racial or ethnic group of residents in the ACS for 2015 to 2019 as the Eviction Lab classifies it, not the race of any tenant) | tracts classified majority Hispanic by the Eviction Lab, in the balanced panel with a baseline | calendar year 2023 |   | 1,591 tracts | none (full counts of filings) | Eviction Tracking System |   |
| fil_hot_idx | 1.84 | ratio of filings to the same-month baseline (1 = baseline) | filings divided by the pre-pandemic baseline, summed over the tract-months of the group (a ratio of sums) | tracts of the site with the highest index, in the balanced panel with a baseline | January 2023 to December 2025 |   | 361 tracts | none (full counts of filings) | Eviction Tracking System |   |
| fil_jan2020 | 1.01 | ratio of filings to the same-month baseline (1 = baseline) | filings divided by the pre-pandemic baseline, summed over the tract-months of the group (a ratio of sums), January 2020 | tracts of the sites in the balanced panel that have a pre-pandemic baseline | 2020-01 |   | 10,243 tracts | none (full counts of filings) | Eviction Tracking System |   |
| fil_n_cities | 35 | cities and metro areas | count of distinct sites in the staged file | sites in the Eviction Tracking System file | 2020-01 to 2026-06 |   |   | none (full count) | Eviction Tracking System |   |
| fil_n_cities_idx | 32 | cities and metro areas | count of sites with an index over the post-recovery years | tracts of the sites in the balanced panel that have a pre-pandemic baseline | January 2023 to December 2025 |   |   | none (full count) | Eviction Tracking System |   |
| fil_n_months | 72 | months | number of months from January 2020 to December 2025 | the window of the index | January 2020 to December 2025 |   |   | none (definition) | Eviction Tracking System |   |
| fil_n_sites | 32 | cities and metro areas | count of sites in the index | sites with a pre-pandemic baseline and tracts observed in every month | January 2020 to December 2025 |   |   | none (full count) | Eviction Tracking System |   |
| fil_n_tracts | 10,243 | tracts | count of tracts with a baseline and an observation in every month of the window | tracts of the sites in the balanced panel that have a pre-pandemic baseline | January 2020 to December 2025 |   |   | none (full count) | Eviction Tracking System |   |
| fil_one | 1.00 | ratio of filings to the same-month baseline | the index value at which filings equal the pre-pandemic baseline | tract-months of the panel | January 2020 to December 2025 |   |   | none (definition) | Eviction Tracking System |   |
| fil_overshoot_all | 1.17 | ratio of filings to the same-month baseline (1 = baseline) | filings divided by the pre-pandemic baseline, summed over the tract-months of the group (a ratio of sums), all tract-months of 2023 | tracts of the sites in the balanced panel that have a pre-pandemic baseline | calendar year 2023 |   | 10,243 tracts | none (full counts of filings) | Eviction Tracking System |   |
| fil_peak | 1.37 | ratio of filings to the same-month baseline (1 = baseline) | filings divided by the pre-pandemic baseline, summed over the tract-months of the group (a ratio of sums), the month of highest filings | tracts of the sites in the balanced panel that have a pre-pandemic baseline | 2023-03 |   | 10,243 tracts | none (full counts of filings) | Eviction Tracking System |   |
| fil_peak_above | 37% | percent above the baseline | the index in the month of highest filings minus one, times 100 | tracts of the sites in the balanced panel that have a pre-pandemic baseline | 2023-03 |   |   | none (full counts of filings) | Eviction Tracking System |   |
| fil_rate_black_2023 | 12.5 | eviction filings per 100 renter households per year | filings summed over the tracts of the group divided by the renter households of those tracts (ACS 2019-2023 households times one minus the homeownership rate), per 100 | tracts classified majority Black by the Eviction Lab, in the balanced panel with a baseline and renter households | calendar year 2023 |   | 1,336 tracts | none (full counts of filings) | Eviction Tracking System |   |
| fil_rate_black_base | 12.7 | eviction filings per 100 renter households per year | filings summed over the tracts of the group divided by the renter households of those tracts (ACS 2019-2023 households times one minus the homeownership rate), per 100 | tracts classified majority Black by the Eviction Lab, in the balanced panel with a baseline and renter households | an average year before the pandemic (baseline) |   | 1,336 tracts | none (full counts of filings) | Eviction Tracking System |   |
| fil_rate_ratio_base | 3.0 | ratio of filings per 100 renter households | baseline filing rate per 100 renter households in majority-Black tracts divided by that in majority-white tracts | tracts classified majority Black and majority white by the Eviction Lab, in the balanced panel with a baseline and renter households | an average year before the pandemic (baseline) |   |   | none (full counts of filings) | Eviction Tracking System |   |
| fil_rate_white_2023 | 5.4 | eviction filings per 100 renter households per year | filings summed over the tracts of the group divided by the renter households of those tracts (ACS 2019-2023 households times one minus the homeownership rate), per 100 | tracts classified majority white by the Eviction Lab, in the balanced panel with a baseline and renter households | calendar year 2023 |   | 4,903 tracts | none (full counts of filings) | Eviction Tracking System |   |
| fil_rate_white_base | 4.2 | eviction filings per 100 renter households per year | filings summed over the tracts of the group divided by the renter households of those tracts (ACS 2019-2023 households times one minus the homeownership rate), per 100 | tracts classified majority white by the Eviction Lab, in the balanced panel with a baseline and renter households | an average year before the pandemic (baseline) |   | 4,903 tracts | none (full counts of filings) | Eviction Tracking System |   |
| fil_site_black_lower | 21 | cities and metro areas where the 2023 index is lower in majority-Black tracts than in majority-white tracts | count of sites where filings divided by the pre-pandemic baseline, summed over the site's majority-Black tracts and 2023 months, are below the same ratio for its majority-white tracts (point estimates; the counts are full counts) | sites of the balanced panel with at least 5 majority-Black and 5 majority-white tracts | calendar year 2023 |   | 23 cities and metro areas | none (full counts of filings) | Eviction Tracking System |   |
| fil_site_min | 5 | tracts | minimum number of majority-Black tracts and of majority-white tracts in a site's panel for the site to enter the comparison | sites in the within-site comparison of racial-majority groups | calendar year 2023 |   |   | none (inclusion rule) | Eviction Tracking System |   |
| fil_site_n | 23 | cities and metro areas | count of sites in the within-site comparison | sites of the balanced panel with at least 5 majority-Black and 5 majority-white tracts | calendar year 2023 |   |   | none (full count) | Eviction Tracking System |   |
| fil_trough | 0.09 | ratio of filings to the same-month baseline (1 = baseline) | filings divided by the pre-pandemic baseline, summed over the tract-months of the group (a ratio of sums), the month of lowest filings | tracts of the sites in the balanced panel that have a pre-pandemic baseline | 2020-04 |   | 10,243 tracts | none (full counts of filings) | Eviction Tracking System |   |
| fil_trough_below | 91% | percent below the baseline | one minus the index in the month of lowest filings, times 100 | tracts of the sites in the balanced panel that have a pre-pandemic baseline | 2020-04 |   |   | none (full counts of filings) | Eviction Tracking System |   |
| fil_unclassified | 236 | tracts | count of panel tracts with no racial-majority classification from the Eviction Lab | tracts of the sites in the balanced panel that have a pre-pandemic baseline | January 2020 to December 2025 |   |   | none (full count) | Eviction Tracking System |   |
| fil_white_2023 | 1.28 | ratio of filings to the same-month baseline (1 = baseline) | filings divided by the pre-pandemic baseline, summed over the tract-months of the group (a ratio of sums), all tract-months of 2023 (race is the tract's majority racial or ethnic group of residents in the ACS for 2015 to 2019 as the Eviction Lab classifies it, not the race of any tenant) | tracts classified majority white by the Eviction Lab, in the balanced panel with a baseline | calendar year 2023 |   | 4,919 tracts | none (full counts of filings) | Eviction Tracking System |   |
| fmr_covered | 94.4% | percent of renter households with a computed burden | renter households with a computed burden in the areas compared over those in all counties of the 50 states and DC | renter households of the 50 states and DC | rent standard for fiscal year 2025 (October 2024 to September 2025); incomes ACS 2019-2023; renter burden CHAS 2018-2022 |   | 2,139 FMR areas | none (survey-weighted household counts) | HUD Fair Market Rents |   |
| fmr_median_usd | $1,703 | dollars per month, two-bedroom Fair Market Rent | median over areas of the two-bedroom Fair Market Rent, renter-weighted | HUD FMR areas made of whole counties with at least 500 renter households with a computed burden | rent standard for fiscal year 2025 (October 2024 to September 2025); incomes ACS 2019-2023; renter burden CHAS 2018-2022 |   | 2,139 FMR areas | renter households with a computed burden (CHAS 2018-2022) | HUD Fair Market Rents |   |
| fmr_n_areas | 2,139 | HUD FMR areas | count of FMR areas that consist of whole counties, have an income and pass the renter-household floor | HUD FMR areas made of whole counties with at least 500 renter households with a computed burden | rent standard for fiscal year 2025 (October 2024 to September 2025); incomes ACS 2019-2023; renter burden CHAS 2018-2022 |   | 2,696 counties | none (full count) | HUD Fair Market Rents |   |
| fmr_n_counties | 2,696 | counties | count of counties in the areas compared; counties in New England, where the file lists towns, and two other split counties are left out | counties of the 50 states and DC in the FMR areas compared | rent standard for fiscal year 2025 (October 2024 to September 2025); incomes ACS 2019-2023; renter burden CHAS 2018-2022 |   |   | none (full count) | HUD Fair Market Rents |   |
| fmr_p5_usd | $933 | dollars per month, two-bedroom Fair Market Rent | 5th percentile over areas of the two-bedroom Fair Market Rent, renter-weighted | HUD FMR areas made of whole counties with at least 500 renter households with a computed burden | rent standard for fiscal year 2025 (October 2024 to September 2025); incomes ACS 2019-2023; renter burden CHAS 2018-2022 |   | 2,139 FMR areas | renter households with a computed burden (CHAS 2018-2022) | HUD Fair Market Rents |   |
| fmr_p95_usd | $2,684 | dollars per month, two-bedroom Fair Market Rent | 95th percentile over areas of the two-bedroom Fair Market Rent, renter-weighted | HUD FMR areas made of whole counties with at least 500 renter households with a computed burden | rent standard for fiscal year 2025 (October 2024 to September 2025); incomes ACS 2019-2023; renter burden CHAS 2018-2022 |   | 2,139 FMR areas | renter households with a computed burden (CHAS 2018-2022) | HUD Fair Market Rents |   |
| fmr_r | +0.70 | correlation coefficient | renter-weighted correlation across areas of the rent share of income with the renter share above 30 percent of income (CHAS), bootstrap over states (the area's state is that of its county with the most renters) | HUD FMR areas made of whole counties with at least 500 renter households with a computed burden | rent standard for fiscal year 2025 (October 2024 to September 2025); incomes ACS 2019-2023; renter burden CHAS 2018-2022 | 95% CI 0.51 to 0.79 | 2,139 FMR areas | renter households with a computed burden (CHAS 2018-2022) | HUD Fair Market Rents |   |
| fmr_r_unw | +0.36 | correlation coefficient | correlation across areas of the rent share of income with the renter share above 30 percent of income (CHAS), each area counted once, bootstrap over states | HUD FMR areas made of whole counties with at least 500 renter households with a computed burden | rent standard for fiscal year 2025 (October 2024 to September 2025); incomes ACS 2019-2023; renter burden CHAS 2018-2022 | 95% CI 0.25 to 0.47 | 2,139 FMR areas | none (each area counts once) | HUD Fair Market Rents |   |
| fmr_share_median | 23.7% | percent of the area's median household income | median over areas of the ratio of a year of the area's two-bedroom Fair Market Rent to the area's median household income (household-weighted mean of its counties' ACS medians), as a percent; renter-weighted | HUD FMR areas made of whole counties with at least 500 renter households with a computed burden | rent standard for fiscal year 2025 (October 2024 to September 2025); incomes ACS 2019-2023; renter burden CHAS 2018-2022 |   | 2,139 FMR areas | renter households with a computed burden (CHAS 2018-2022) | HUD Fair Market Rents |   |
| fmr_share_median_unw | 19.9% | percent of the area's median household income | median over areas of the ratio of a year of the two-bedroom Fair Market Rent to the area's median household income, each area counted once | HUD FMR areas made of whole counties with at least 500 renter households with a computed burden | rent standard for fiscal year 2025 (October 2024 to September 2025); incomes ACS 2019-2023; renter burden CHAS 2018-2022 |   | 2,139 FMR areas | none (each area counts once) | HUD Fair Market Rents |   |
| fmr_share_p5 | 17.6% | percent of the area's median household income | 5th percentile over areas of the ratio of a year of the area's two-bedroom Fair Market Rent to the area's median household income (household-weighted mean of its counties' ACS medians), as a percent; renter-weighted | HUD FMR areas made of whole counties with at least 500 renter households with a computed burden | rent standard for fiscal year 2025 (October 2024 to September 2025); incomes ACS 2019-2023; renter burden CHAS 2018-2022 |   | 2,139 FMR areas | renter households with a computed burden (CHAS 2018-2022) | HUD Fair Market Rents |   |
| fmr_share_p95 | 35.9% | percent of the area's median household income | 95th percentile over areas of the ratio of a year of the area's two-bedroom Fair Market Rent to the area's median household income (household-weighted mean of its counties' ACS medians), as a percent; renter-weighted | HUD FMR areas made of whole counties with at least 500 renter households with a computed burden | rent standard for fiscal year 2025 (October 2024 to September 2025); incomes ACS 2019-2023; renter burden CHAS 2018-2022 |   | 2,139 FMR areas | renter households with a computed burden (CHAS 2018-2022) | HUD Fair Market Rents |   |
| fmr_slope | 0.72 | percentage points of the renter share above 30% of income per percentage point of the rent share of income | renter-weighted slope across areas of the renter share on the rent share, bootstrap over states | HUD FMR areas made of whole counties with at least 500 renter households with a computed burden | rent standard for fiscal year 2025 (October 2024 to September 2025); incomes ACS 2019-2023; renter burden CHAS 2018-2022 | 95% CI 0.52 to 0.93 | 2,139 FMR areas | renter households with a computed burden (CHAS 2018-2022) | HUD Fair Market Rents |   |
| fmr_slope_unw | 0.71 | percentage points of the renter share above 30% of income per percentage point of the rent share of income | slope across areas of the renter share on the rent share, each area counted once, bootstrap over states | HUD FMR areas made of whole counties with at least 500 renter households with a computed burden | rent standard for fiscal year 2025 (October 2024 to September 2025); incomes ACS 2019-2023; renter burden CHAS 2018-2022 | 95% CI 0.49 to 0.94 | 2,139 FMR areas | none (each area counts once) | HUD Fair Market Rents |   |
| fq_burden_1 | 39.1% | percent of renter households with housing costs above 30% of income | renter households with a burden over renter households with a computed burden, pooled over the fifth's areas; bootstrap over states | HUD FMR areas made of whole counties with at least 500 renter households with a computed burden, lowest fifth of areas by median household income (fifths hold near-equal numbers of areas) | rent standard for fiscal year 2025 (October 2024 to September 2025); incomes ACS 2019-2023; renter burden CHAS 2018-2022 | 95% CI 37.6% to 40.9% | 427 FMR areas | renter households with a computed burden (CHAS 2018-2022) | HUD CHAS housing affordability |   |
| fq_burden_5 | 46.9% | percent of renter households with housing costs above 30% of income | renter households with a burden over renter households with a computed burden, pooled over the fifth's areas; bootstrap over states | HUD FMR areas made of whole counties with at least 500 renter households with a computed burden, highest fifth of areas by median household income (fifths hold near-equal numbers of areas) | rent standard for fiscal year 2025 (October 2024 to September 2025); incomes ACS 2019-2023; renter burden CHAS 2018-2022 | 95% CI 44.8% to 48.4% | 428 FMR areas | renter households with a computed burden (CHAS 2018-2022) | HUD CHAS housing affordability |   |
| fq_inc_1 | $45,717 | dollars of median household income | renter-weighted mean over the fifth's areas of the area's median household income (household-weighted mean of its counties' ACS medians); bootstrap over states | HUD FMR areas made of whole counties with at least 500 renter households with a computed burden, lowest fifth of areas by median household income (fifths hold near-equal numbers of areas) | ACS 2019-2023 (2023 dollars) | 95% CI $44,955 to $46,628 | 427 FMR areas | renter households with a computed burden (CHAS 2018-2022) | American Community Survey |   |
| fq_inc_5 | $91,563 | dollars of median household income | renter-weighted mean over the fifth's areas of the area's median household income (household-weighted mean of its counties' ACS medians); bootstrap over states | HUD FMR areas made of whole counties with at least 500 renter households with a computed burden, highest fifth of areas by median household income (fifths hold near-equal numbers of areas) | ACS 2019-2023 (2023 dollars) | 95% CI $86,017 to $96,668 | 428 FMR areas | renter households with a computed burden (CHAS 2018-2022) | American Community Survey |   |
| fq_share_1 | 24.4% | percent of the area's median household income | renter-weighted mean over the fifth's areas of the ratio of a year of the area's two-bedroom Fair Market Rent to the area's median household income (household-weighted mean of its counties' ACS medians), as a percent; bootstrap over states | HUD FMR areas made of whole counties with at least 500 renter households with a computed burden, lowest fifth of areas by median household income (fifths hold near-equal numbers of areas) | rent standard for fiscal year 2025 (October 2024 to September 2025); incomes ACS 2019-2023; renter burden CHAS 2018-2022 | 95% CI 23.4% to 25.3% | 427 FMR areas | renter households with a computed burden (CHAS 2018-2022) | HUD Fair Market Rents |   |
| fq_share_3 | 21.5% | percent of the area's median household income | renter-weighted mean over the fifth's areas of the ratio of a year of the area's two-bedroom Fair Market Rent to the area's median household income (household-weighted mean of its counties' ACS medians), as a percent; bootstrap over states | HUD FMR areas made of whole counties with at least 500 renter households with a computed burden, middle fifth of areas by median household income (fifths hold near-equal numbers of areas) | rent standard for fiscal year 2025 (October 2024 to September 2025); incomes ACS 2019-2023; renter burden CHAS 2018-2022 | 95% CI 20.5% to 22.4% | 428 FMR areas | renter households with a computed burden (CHAS 2018-2022) | HUD Fair Market Rents |   |
| fq_share_5 | 26.1% | percent of the area's median household income | renter-weighted mean over the fifth's areas of the ratio of a year of the area's two-bedroom Fair Market Rent to the area's median household income (household-weighted mean of its counties' ACS medians), as a percent; bootstrap over states | HUD FMR areas made of whole counties with at least 500 renter households with a computed burden, highest fifth of areas by median household income (fifths hold near-equal numbers of areas) | rent standard for fiscal year 2025 (October 2024 to September 2025); incomes ACS 2019-2023; renter burden CHAS 2018-2022 | 95% CI 23.0% to 28.3% | 428 FMR areas | renter households with a computed burden (CHAS 2018-2022) | HUD Fair Market Rents |   |
| ft_burden_1 | 33.9% | percent of renter households with housing costs above 30% of income | renter-weighted mean over the tenth's areas of the area's renter share above 30 percent of income; bootstrap over states | HUD FMR areas made of whole counties with at least 500 renter households with a computed burden, lowest tenth of areas by the rent share of income (tenths hold near-equal numbers of areas) | rent standard for fiscal year 2025 (October 2024 to September 2025); incomes ACS 2019-2023; renter burden CHAS 2018-2022 | 95% CI 32.1% to 35.8% | 213 FMR areas | renter households with a computed burden (CHAS 2018-2022) | HUD CHAS housing affordability |   |
| ft_burden_10 | 49.9% | percent of renter households with housing costs above 30% of income | renter-weighted mean over the tenth's areas of the area's renter share above 30 percent of income; bootstrap over states | HUD FMR areas made of whole counties with at least 500 renter households with a computed burden, highest tenth of areas by the rent share of income (tenths hold near-equal numbers of areas) | rent standard for fiscal year 2025 (October 2024 to September 2025); incomes ACS 2019-2023; renter burden CHAS 2018-2022 | 95% CI 45.9% to 51.4% | 214 FMR areas | renter households with a computed burden (CHAS 2018-2022) | HUD CHAS housing affordability |   |
| hamfi_30 | 30% | percent of HUD area median family income | the income line between the lowest and the second of HUD's five income bands (income relative to HUD's area median family income, HAMFI) | households in the HUD CHAS tabulation | 2018 to 2022 |   |   | none (definition) | HUD CHAS housing affordability |   |
| hamfi_50 | 50% | percent of HUD area median family income | the income line between the second and the third of HUD's five income bands (income relative to HUD's area median family income, HAMFI) | households in the HUD CHAS tabulation | 2018 to 2022 |   |   | none (definition) | HUD CHAS housing affordability |   |
| hamfi_cut | 80% | percent of HUD area median family income | the income line for a low-income household: income at or below this share of HUD's area median family income (HAMFI) | households in the HUD CHAS tabulation | 2018 to 2022 |   |   | none (definition) | HUD CHAS housing affordability |   |
| li_bur_owner | 50.7% | percent of low-income owner households with a computed burden that have housing costs above 30% of income | low-income households (at or below 80 percent of HAMFI) with a burden above 30 percent, over low-income households with a computed burden (low-income households less the households whose burden is not computed; the staged HUD percentage divides by all households of the tenure and is not used); pooled over tracts, bootstrap over counties | low-income owner households with a computed burden in the tracts of the 50 states and DC | 2018 to 2022 | 95% CI 49.8% to 51.7% | 84,410 tracts | none (pooled household counts) | HUD CHAS housing affordability |   |
| li_bur_renter | 69.0% | percent of low-income renter households with a computed burden that have housing costs above 30% of income | low-income households (at or below 80 percent of HAMFI) with a burden above 30 percent, over low-income households with a computed burden (low-income households less the households whose burden is not computed; the staged HUD percentage divides by all households of the tenure and is not used); pooled over tracts, bootstrap over counties | low-income renter households with a computed burden in the tracts of the 50 states and DC | 2018 to 2022 | 95% CI 68.1% to 69.9% | 84,410 tracts | none (pooled household counts) | HUD CHAS housing affordability |   |
| li_of_bur_owner | 73.8% | percent of owner households with a burden above 30% that are low-income | low-income households with a burden above 30 percent over all households of the tenure with a burden above 30 percent; pooled over tracts, bootstrap over counties | owner households with housing costs above 30 percent of income | 2018 to 2022 | 95% CI 72.8% to 74.7% | 84,410 tracts | none (pooled household counts) | HUD CHAS housing affordability |   |
| li_of_bur_renter | 91.8% | percent of renter households with a burden above 30% that are low-income | low-income households with a burden above 30 percent over all households of the tenure with a burden above 30 percent; pooled over tracts, bootstrap over counties | renter households with housing costs above 30 percent of income | 2018 to 2022 | 95% CI 91.2% to 92.4% | 84,410 tracts | none (pooled household counts) | HUD CHAS housing affordability |   |
| li_share_owner | 30.2% | percent of owner households with a computed burden that have income at or below 80% of HAMFI | low-income households with a computed burden (income at or below 80 percent of HUD's area median family income, HAMFI, less the households whose burden is not computed) over all households of the tenure with a computed burden; pooled over tracts, bootstrap over counties. HAMFI is HUD's benchmark and differs from the Census median household income | owner households in the tracts of the 50 states and DC with a computed burden | 2018 to 2022 | 95% CI 29.8% to 30.7% | 84,410 tracts | none (pooled household counts) | HUD CHAS housing affordability |   |
| li_share_renter | 60.6% | percent of renter households with a computed burden that have income at or below 80% of HAMFI | low-income households with a computed burden (income at or below 80 percent of HUD's area median family income, HAMFI, less the households whose burden is not computed) over all households of the tenure with a computed burden; pooled over tracts, bootstrap over counties. HAMFI is HUD's benchmark and differs from the Census median household income | renter households in the tracts of the 50 states and DC with a computed burden | 2018 to 2022 | 95% CI 59.8% to 61.4% | 84,410 tracts | none (pooled household counts) | HUD CHAS housing affordability |   |
| majority_line | 50% | percent of renter households | the share of a county's renter households with a burden above which a majority of them have one | renter households in a county | 2018 to 2022 |   |   | none (definition) | HUD CHAS housing affordability |   |
| mod_owner | 12.0% | percent of owner households with housing costs of 30% to 50% of income | households above the burden line minus households above the severe line, over households whose burden HUD computes, pooled over tracts; interval from a bootstrap over counties | owner households in the tracts of the 50 states and DC with a computed burden | 2018 to 2022 | 95% CI 11.7% to 12.3% | 84,410 tracts | none (pooled household counts) | HUD CHAS housing affordability |   |
| mod_renter | 22.4% | percent of renter households with housing costs of 30% to 50% of income | households above the burden line minus households above the severe line, over households whose burden HUD computes, pooled over tracts; interval from a bootstrap over counties | renter households in the tracts of the 50 states and DC with a computed burden | 2018 to 2022 | 95% CI 22.1% to 22.7% | 84,410 tracts | none (pooled household counts) | HUD CHAS housing affordability |   |
| n_burdened_renters | 19.7 million | renter households with housing costs above 30% of income | sum over tracts of renter households with a burden above 30 percent | households in the tracts of the 50 states and DC (Puerto Rico left out) | 2018 to 2022 |   | 84,410 tracts | none (survey-weighted household counts) | HUD CHAS housing affordability |   |
| n_counties | 3,143 | counties and county equivalents | count of counties in the county frame: the county file for every state but Connecticut and the tract file summed over former counties for Connecticut | counties of the 50 states and DC (Connecticut as its eight former counties) in the HUD CHAS files | 2018 to 2022 |   |   | none (full count) | HUD CHAS housing affordability |   |
| n_cty_half | 175 | counties where more than half of renter households have housing costs above 30% of income | count of counties with a renter share above 50 percent | counties of the 50 states and DC (Connecticut as its eight former counties) with at least 500 renter and 500 owner households with a computed burden | 2018 to 2022 |   | 2,753 counties | none (full count) | HUD CHAS housing affordability |   |
| n_cty_ok | 2,753 | counties | count of counties that pass the household floor for both tenures | counties of the 50 states and DC (Connecticut as its eight former counties) with at least 500 renter and 500 owner households with a computed burden | 2018 to 2022 |   | 3,143 counties | none (full count) | HUD CHAS housing affordability |   |
| n_cty_owner_ge | 31 | counties where the owner share with a burden is at least the renter share | count of counties where the owner share above 30 percent of income is at least the renter share | counties of the 50 states and DC (Connecticut as its eight former counties) with at least 500 renter and 500 owner households with a computed burden | 2018 to 2022 |   | 2,753 counties | none (full count) | HUD CHAS housing affordability |   |
| n_cty_owner_ge_hi | 11 | counties where the owner share with a burden is at least the renter share | count of counties where the owner share above 30 percent of income is at least the renter share under the alternative household floor | counties of the 50 states and DC (Connecticut as its eight former counties) with at least 1000 renter and 1000 owner households with a computed burden | 2018 to 2022 |   | 2,323 counties | none (full count) | HUD CHAS housing affordability |   |
| n_cty_owner_ge_lo | 106 | counties where the owner share with a burden is at least the renter share | count of counties where the owner share above 30 percent of income is at least the renter share under the alternative household floor | counties of the 50 states and DC (Connecticut as its eight former counties) with at least 100 renter and 100 owner households with a computed burden | 2018 to 2022 |   | 3,103 counties | none (full count) | HUD CHAS housing affordability |   |
| n_nc_owner | 0.7 million | owner households whose burden HUD does not compute (no or negative income) | sum over tracts of the tenure's households less those with a computed burden; the raw county table puts all of them in the lowest income band | owner households in the tracts of the 50 states and DC | 2018 to 2022 |   | 84,410 tracts | none (survey-weighted household counts) | HUD CHAS housing affordability |   |
| n_nc_renter | 1.0 million | renter households whose burden HUD does not compute (no or negative income) | sum over tracts of the tenure's households less those with a computed burden; the raw county table puts all of them in the lowest income band | renter households in the tracts of the 50 states and DC | 2018 to 2022 |   | 84,410 tracts | none (survey-weighted household counts) | HUD CHAS housing affordability |   |
| n_owner_hh | 80.8 million | owner households with a computed burden | sum over tracts of the households whose cost burden HUD computes (households with no or negative income are left out) | owner households in the tracts of the 50 states and DC with a computed burden | 2018 to 2022 |   | 84,410 tracts | none (survey-weighted household counts) | HUD CHAS housing affordability |   |
| n_renter_hh | 43.2 million | renter households with a computed burden | sum over tracts of the households whose cost burden HUD computes (households with no or negative income are left out) | renter households in the tracts of the 50 states and DC with a computed burden | 2018 to 2022 |   | 84,410 tracts | none (survey-weighted household counts) | HUD CHAS housing affordability |   |
| n_severe_renters | 10.0 million | renter households with housing costs above 50% of income | sum over tracts of renter households with a severe burden | households in the tracts of the 50 states and DC (Puerto Rico left out) | 2018 to 2022 |   | 84,410 tracts | none (survey-weighted household counts) | HUD CHAS housing affordability |   |
| n_tracts_adi | 82,786 | tracts | count of tracts with an Area Deprivation Index national rank, the tracts that the ADI comparisons rank into tenths | tracts of the 50 states and DC in the HUD CHAS tract file | 2018 to 2022; Area Deprivation Index 2023 |   | 84,410 tracts | none (full count) | Area Deprivation Index |   |
| n_tracts_chas | 84,410 | tracts | count of tracts in the tract file after leaving out Puerto Rico and five Connecticut water-only tracts with no households | tracts of the 50 states and DC in the HUD CHAS tract file (Puerto Rico left out) | 2018 to 2022 (American Community Survey five-year special tabulation by HUD) |   |   | none (full count) | HUD CHAS housing affordability |   |
| n_tracts_empty | 1,051 | tracts with no households | count of tracts with no owner and no renter households; they add nothing to a pooled share and drop out of the tenth means | tracts of the 50 states and DC in the HUD CHAS tract file (Puerto Rico left out) | 2018 to 2022 (American Community Survey five-year special tabulation by HUD) |   | 84,410 tracts | none (full count) | HUD CHAS housing affordability |   |
| n_tracts_no_adi | 1,624 | tracts | count of tracts with no Area Deprivation Index national rank, left out of the ADI comparisons | tracts of the 50 states and DC in the HUD CHAS tract file | 2018 to 2022; Area Deprivation Index 2023 |   | 84,410 tracts | none (full count) | Area Deprivation Index |   |
| renter_hh_share | 35.2% | percent of occupied housing units that are rented | renter households over renter and owner households, pooled over tracts; bootstrap over counties | households in the tracts of the 50 states and DC (Puerto Rico left out) | 2018 to 2022 | 95% CI 33.7% to 36.8% | 84,410 tracts | none (pooled household counts) | HUD CHAS housing affordability |   |
| sev_owner | 8.8% | percent of owner households with housing costs above 50% of income | ratio of sums over tracts: households with the burden over households whose burden HUD computes, pooled over the tracts of the 50 states and DC; interval from a bootstrap over counties (replicates and seed in the methods note). The share carries no margin of error of its own: HUD rounds every cell independently | owner households in the tracts of the 50 states and DC with a computed burden | 2018 to 2022 | 95% CI 8.5% to 9.2% | 84,410 tracts | none (pooled household counts) | HUD CHAS housing affordability |   |
| sev_renter | 23.2% | percent of renter households with housing costs above 50% of income | ratio of sums over tracts: households with the burden over households whose burden HUD computes, pooled over the tracts of the 50 states and DC; interval from a bootstrap over counties (replicates and seed in the methods note). The share carries no margin of error of its own: HUD rounds every cell independently | renter households in the tracts of the 50 states and DC with a computed burden | 2018 to 2022 | 95% CI 22.6% to 23.8% | 84,410 tracts | none (pooled household counts) | HUD CHAS housing affordability |   |
| severe_line | 50% | percent of household income | HUD's line for a severe housing cost burden: housing costs above this share of household income | households in the HUD CHAS tabulation | 2018 to 2022 |   |   | none (definition) | HUD CHAS housing affordability |   |
| slope_btw_own | −1.6 | percentage points of the owner share with housing costs above 30% of income per 10 percentile points of the Area Deprivation Index, between counties | household-weighted slope across counties of the county's pooled-tract mean share on the county's mean index rank, bootstrap over counties | counties with owner households in tracts that have an Area Deprivation Index | 2018 to 2022; Area Deprivation Index 2023 | 95% CI −1.9 to −1.4 | 3,128 counties | households of the tenure with a computed burden | HUD CHAS housing affordability |   |
| slope_btw_ren | −1.1 | percentage points of the renter share with housing costs above 30% of income per 10 percentile points of the Area Deprivation Index, between counties | household-weighted slope across counties of the county's pooled-tract mean share on the county's mean index rank, bootstrap over counties | counties with renter households in tracts that have an Area Deprivation Index | 2018 to 2022; Area Deprivation Index 2023 | 95% CI −1.4 to −0.8 | 3,128 counties | households of the tenure with a computed burden | HUD CHAS housing affordability |   |
| slope_pool_own | −0.8 | percentage points of the owner share with housing costs above 30% of income per 10 percentile points of the Area Deprivation Index, across the nation | household-weighted slope of the tract share on the ADI national rank, no county baseline, bootstrap over counties | owner households in tracts with an Area Deprivation Index and a computed burden | 2018 to 2022; Area Deprivation Index 2023 | 95% CI −1.0 to −0.5 | 82,327 tracts | households of the tenure with a computed burden | HUD CHAS housing affordability |   |
| slope_pool_ren | 0.0 | percentage points of the renter share with housing costs above 30% of income per 10 percentile points of the Area Deprivation Index, across the nation | household-weighted slope of the tract share on the ADI national rank, no county baseline, bootstrap over counties | renter households in tracts with an Area Deprivation Index and a computed burden | 2018 to 2022; Area Deprivation Index 2023 | 95% CI −0.3 to 0.3 | 82,558 tracts | households of the tenure with a computed burden | HUD CHAS housing affordability |   |
| slope_wc_own | 0.9 | percentage points of the owner share with housing costs above 30% of income per 10 percentile points of the Area Deprivation Index, within the same county | household-weighted slope of the tract share on the ADI national rank after subtracting each county's weighted mean from both (a county baseline for each county), bootstrap over counties | owner households in tracts with an Area Deprivation Index and a computed burden | 2018 to 2022; Area Deprivation Index 2023 | 95% CI 0.9 to 1.0 | 82,327 tracts | households of the tenure with a computed burden | HUD CHAS housing affordability |   |
| slope_wc_ren | 2.3 | percentage points of the renter share with housing costs above 30% of income per 10 percentile points of the Area Deprivation Index, within the same county | household-weighted slope of the tract share on the ADI national rank after subtracting each county's weighted mean from both (a county baseline for each county), bootstrap over counties | renter households in tracts with an Area Deprivation Index and a computed burden | 2018 to 2022; Area Deprivation Index 2023 | 95% CI 2.1 to 2.4 | 82,558 tracts | households of the tenure with a computed burden | HUD CHAS housing affordability |   |
| st_ca | 50.7% | percent of renter households with housing costs above 30% of income | ratio of sums over tracts: households with the burden over households whose burden HUD computes, pooled over the tracts of California; interval from a bootstrap over counties (replicates and seed in the methods note). The share carries no margin of error of its own: HUD rounds every cell independently | renter households in California with a computed burden | 2018 to 2022 | 95% CI 47.3% to 52.4% | 9,129 tracts | none (pooled household counts) | HUD CHAS housing affordability |   |
| st_fl | 53.3% | percent of renter households with housing costs above 30% of income | ratio of sums over tracts: households with the burden over households whose burden HUD computes, pooled over the tracts of Florida; interval from a bootstrap over counties (replicates and seed in the methods note). The share carries no margin of error of its own: HUD rounds every cell independently | renter households in Florida with a computed burden | 2018 to 2022 | 95% CI 50.4% to 55.4% | 5,160 tracts | none (pooled household counts) | HUD CHAS housing affordability |   |
| st_sd | 33.9% | percent of renter households with housing costs above 30% of income | ratio of sums over tracts: households with the burden over households whose burden HUD computes, pooled over the tracts of South Dakota; interval from a bootstrap over counties (replicates and seed in the methods note). The share carries no margin of error of its own: HUD rounds every cell independently | renter households in South Dakota with a computed burden | 2018 to 2022 | 95% CI 30.6% to 35.8% | 242 tracts | none (pooled household counts) | HUD CHAS housing affordability |   |
| us_burden_li_renter | 69.0% | percent of low-income renter households with housing costs above 30% of income | low-income households (at or below 80 percent of HAMFI) with a burden over low-income households of the tenure, pooled over tracts in all ten tenths, bootstrap over counties | low-income renter households in all the tracts with an Area Deprivation Index (all ten tenths pooled) | 2018 to 2022; Area Deprivation Index 2023 | 95% CI 68.0% to 69.9% | 81,284 tracts | low-income households of the tenure | HUD CHAS housing affordability |   |
| us_burden_owner | 20.8% | percent of owner households with housing costs above 30% of income | pooled share of the tenure's households with the burden over tracts in all ten tenths, bootstrap over counties | owner households in all the tracts with an Area Deprivation Index (all ten tenths pooled) | 2018 to 2022; Area Deprivation Index 2023 | 95% CI 20.1% to 21.5% | 82,327 tracts | households of the tenure with a computed burden | HUD CHAS housing affordability |   |
| us_burden_renter | 45.5% | percent of renter households with housing costs above 30% of income | pooled share of the tenure's households with the burden over tracts in all ten tenths, bootstrap over counties | renter households in all the tracts with an Area Deprivation Index (all ten tenths pooled) | 2018 to 2022; Area Deprivation Index 2023 | 95% CI 44.7% to 46.4% | 82,558 tracts | households of the tenure with a computed burden | HUD CHAS housing affordability |   |

## Definitions

- Area Deprivation Index (ADI): A score that ranks neighborhoods by income, education, employment and housing conditions drawn from the American Community Survey. In this book a higher national percentile means a more deprived neighborhood.
- Ecological inference: Drawing conclusions about individuals from data about places or groups. A pattern across counties need not hold for the people inside them.

## Methods

This chapter compares the housing cost burden of renter and owner households in HUD's tabulation of the American Community Survey for 2018 to 2022, across income bands, tenths of tracts by deprivation, counties and Fair Market Rent areas, and then eviction filings by tract and by site from January 2020 to December 2025. The two parts cover different years and different tracts: the burden figures cover the 50 states and DC, with counties entering at 500 renter and 500 owner households with a computed burden, and the eviction index covers only tracts observed in all 72 months that have a baseline in every month. The burden intervals come from a bootstrap and do not include the sampling error of the survey, which HUD does not publish for shares, and HUD rounds every cell independently. The eviction series are full counts of filings and have no sampling interval.

### Which places and households are covered

The 50 states and DC are covered, with Puerto Rico left out.

The county frame takes the county file for every state but Connecticut and sums the tract file over Connecticut's eight former counties; a county enters with at least 500 renter and 500 owner households with a computed burden.

Explains Table 1 and Table 2.

### How cost burden and income bands are defined

HUD CHAS counts households by tenure, by income relative to HUD's area median family income (HAMFI) and by the share of income spent on housing: a burden above 30 percent of income, and a severe burden above 50 percent.

Every share is a ratio of summed household counts (never an average of the published tract percentages) over households whose burden HUD computes; low-income households are those at or below 80 percent of HAMFI, and the low-income shares divide by the low-income households with a computed burden. The staged HUD percentage for low-income burdened households divides by all households and is not used.

The five income bands (30, 50, 80 and 100 percent of HAMFI) come from the raw HUD CHAS county Table 8, read once, and the income-mix standardization gives renters the owners' share of households in each band.

Explains Table 1.

### How tracts and counties are grouped and weighted

Tenths of tracts hold near-equal numbers of tracts (tracts with equal ranks stay together) among those with an Area Deprivation Index national rank.

Tract shares are weighted by the households behind them; the within-county deviation subtracts each county's household-weighted mean from the tract share and from the index rank.

Explains Table 3.

### How rent standards are matched to burden and income

An FMR area enters when all of its rows in the file are whole counties; the two-bedroom rent standard is the area's single value.

The area income is the household-weighted mean of the counties' ACS 2019-2023 median household incomes, and the renter burden is pooled over the area's counties.

Explains Table 4.

### How the eviction filing index is built

The index is the sum of filings divided by the sum of the baseline over the tract-months of a group, for the tracts observed in all 72 months from January 2020 to December 2025 that have a baseline in every month.

Racial majority is the Eviction Lab's tract classification by the group that makes up a majority of residents in the ACS for 2015 to 2019. Filing rates per renter household divide filings by ACS 2019-2023 renter households.

Explains Table 5, Table 6 and Table 7.

### What the intervals include and leave out

HUD rounds every cell independently and publishes no margin of error for the shares.

Intervals come from a bootstrap over counties (tracts) or over states (FMR areas) with 2,000 replicates and seed 12345; they do not include the sampling error of the survey, which HUD does not publish for shares.

The eviction series are full counts and have no sampling interval.

### Four rankings of the same places

The comparison follows the book's ranking switch rule, fixed on October 28, 2026 before any of these gradients was computed. The main gradient of the chapter is recomputed under four rankings of the same places, with the outcome, period, weights, bins and intervals unchanged: the published ADI; a standardized index that applies the ADI's coefficients to its measures after standardizing each one across tracts; median household income; and the share of adults 25 and older with a bachelor's degree or more. The three alternatives come from ACS tract estimates for 2019 to 2023, the period of the ADI.

All four rankings use the same places, the 75,587 tracts that have the outcome and all four rankings, so the ADI panel can differ slightly from the chapter's main figure. Tracts enter only if they have every input of the standardized index, which leaves out tracts without a median rent, home value or mortgage cost.

A pattern counts as changed when the worse end switches sides, when the two ends differ under one ranking and not the other, or when the rank correlation of bin order and bin mean falls on the other side of 0.9 and moves by at least 0.1.

Explains Table 8.

## What these data cannot show

Every comparison in this chapter describes places or groups of households. A county, a tract, an area in the Fair Market Rent file or a site in the Eviction Tracking System is not a person. A pattern across places need not hold for the people inside them (ecological inference). The renters of a tract differ from people who move in and out of it, and the data cannot separate who lives there from what the place provides.

The burden compares housing costs with income at one time. It does not say whether a household chose costly housing or had no cheaper option, and it does not measure the condition of the home, crowding or arrears. The shares come from a survey tabulation that HUD rounds cell by cell, tract shares are noisy, and the intervals leave out the survey's sampling error. HAMFI is HUD's benchmark and differs from the median household income used elsewhere in the book. The deprivation tenths use an index that counts housing costs among its inputs, so they cannot say whether deprivation or price levels sit behind a gradient.

The rent standard is one release for one fiscal year, set for an area and not for a county, and the incomes are the incomes of all households. The comparison holds across areas and not within a metropolitan area, and New England is not in it.

Filings are landlord petitions, not evictions or displaced households, and one household can be the subject of several filings. The index compares each tract with its own baseline, so it does not say where eviction is most common. Moratoria, emergency rental assistance, court closures and changes in local law all moved filings in these years, and the data do not separate them. The renter households behind the filing rates are from 2019 to 2023 and the filings from 2020 to 2025. The counts are full counts for the participating sites, so the series carry no sampling interval, and the sites are not a sample of the nation.

## References

1. McCarty, Maggie; Carpenter, David H.; Perl, Libby (2021). The CDC's Federal Eviction Moratorium. Congressional Research Service, Insight IN11673 (updated August 31, 2021). https://www.everycrsreport.com/reports/IN11673.html
2. Hepburn, Peter; Louis, Renee; Desmond, Matthew (2020). Racial and Gender Disparities among Evicted Americans. Sociological Science 7: 649-662. https://doi.org/10.15195/v7.a27
3. Desmond, Matthew (2016). Evicted: Poverty and Profit in the American City. Crown.
4. Collinson, Robert; Humphries, John Eric; Mader, Nicholas; Reed, Davin; Tannenbaum, Daniel; van Dijk, Winnie (2024). Eviction and Poverty in American Cities. The Quarterly Journal of Economics 139(1): 57-120. https://doi.org/10.1093/qje/qjad042

## Provenance

- Pipeline version: 0.1.0
- Build script: book2/chapters/renting-affordability-and-eviction/build.py (sha256 4717a7206ea664bca6d253be8b22a05c907460347dafe0d2995f1664f7c6e456)
- Built on: 2026-10-09
- Site commit: c2b1ff3309cb4a10833f884da91ac0057a66ad0a

Inputs:

| File | Size (bytes) | SHA-256, first 1 MB | SHA-256, last 1 MB |
| --- | ---: | --- | --- |
| book2/chapters/renting-affordability-and-eviction/bands.py | 7,407 | 67c6908f103a5ef1a12817c41993b856281d82b58fd4838be5dc336335d1a8e6 | 67c6908f103a5ef1a12817c41993b856281d82b58fd4838be5dc336335d1a8e6 |
| book2/chapters/renting-affordability-and-eviction/burden.py | 15,084 | a4d6074ef6582fc7da11263170a40367e1204f678fc175ffd1295e72d077a578 | a4d6074ef6582fc7da11263170a40367e1204f678fc175ffd1295e72d077a578 |
| book2/chapters/renting-affordability-and-eviction/eviction.py | 10,864 | 1b9b0251f5e280b0c5548f1404beb38a4de73f8a5032e9485545d7284fb8d98c | 1b9b0251f5e280b0c5548f1404beb38a4de73f8a5032e9485545d7284fb8d98c |
| book2/chapters/renting-affordability-and-eviction/rent.py | 6,277 | 12fc2c7cba84fb452dd584fd19a863685cc5e67fc5581e4ad3d44a86efec9e55 | 12fc2c7cba84fb452dd584fd19a863685cc5e67fc5581e4ad3d44a86efec9e55 |
| data/analytic/tract_atlas.parquet | 39,659,925 | 29d5e4d943552d54aeb6ba92aadc939d9bd15bdea553c78357dbd516757be627 | 1e10d1ecf722340f23ee2212d4999a90250f85b65fa89b39e32db06d9472e22e |
| data/raw/acs/deprivation_components_2019_2023/county_core.parquet | 498,878 | 8b177857027e73c1f404f7b62191ab78811e95ee54c4f07383e9e9e17da1a345 | 8b177857027e73c1f404f7b62191ab78811e95ee54c4f07383e9e9e17da1a345 |
| data/raw/acs/deprivation_components_2019_2023/tract.parquet | 13,197,793 | 5545cde2cbe96a8b56e468d65388248a241aa6bd300d3c9e325d08d3f86b11da | 5c8ec454d066988b0f1a01aaa6ed34ebfc05d162e01cf9c6bb01f9198cbaff12 |
| data/raw/hud_chas/2018_2022/2018thru2022-050-csv.zip | 17,640,663 | 4053f34d426f5115eb0cd2df6b491628a2a154aec799a1e60d8f50fbaca2523e | f049e8659639476700aabc356a797b1ca70d8019072447ac6879e080ee543b0c |
| data/staged/adi/adi_atlas_tract2020_2023.parquet | 798,078 | 5e409bc212ff871def780fa27b4cf8b616d9da420b2b9f6455517eddcd8c1a95 | 5e409bc212ff871def780fa27b4cf8b616d9da420b2b9f6455517eddcd8c1a95 |
| data/staged/crosswalks/ct_tract2020_to_old_county.parquet | 15,842 | 961ef39f31596ef65fca2c67d5f2f63e533c5dc9b49745d3bb6437844e4b831d | 961ef39f31596ef65fca2c67d5f2f63e533c5dc9b49745d3bb6437844e4b831d |
| data/staged/housing/eviction_ets_tract_2020_2021.parquet | 1,286,128 | d7fb0e3947b961ca3cdc4e95ed0136072bd2198251d7bcd2b0bd587d5ed813af | fb1da4a871fcaa271c237fa4eb2a6cb894a524495db451a6105cdd93a755225e |
| data/staged/housing/hud_chas_county_2018_2022.parquet | 985,759 | aed3c7f19aca2598debda31c43492eb8c20e66bd71ea05cd8a204ace16f70eea | aed3c7f19aca2598debda31c43492eb8c20e66bd71ea05cd8a204ace16f70eea |
| data/staged/housing/hud_chas_tract2020_2018_2022.parquet | 11,160,086 | 8435624fc4e027cb64a4130f1b3484e6f380f2fa4ce0940fc3b9b7d27729fe37 | fde5dbb56ef312620e24275622d0ea8ba38093826433a1fe15edcfdeeea4f632 |
| data/staged/housing/hud_fmr_area_fy2025.parquet | 220,184 | e8acd06036b1bf0c46d350385c9cf08de423cd7145ead42ac2b6bf9367361b7d | e8acd06036b1bf0c46d350385c9cf08de423cd7145ead42ac2b6bf9367361b7d |
| data/staged/ses/acs5_tract2020_2019_2023.parquet | 7,232,424 | 5fe6f730a35225f14d5d8abfc6b9bce58b52e545f309a7eac74f1dae74d7a31c | 7b4a2475b3852d3f91804716a04566ef8ceadacec2be9fb85f5eb642078efac6 |
| data/staged/svi/svi_tract2020_2022.parquet | 17,740,415 | cb3860d8edc2aaf1f2a02211a58f7a0b80457eb6ad9a37b5ec8abb2d5da9da10 | e371a4a696d5bed05d54ad7c5b6ca2504508805b847ec415c2322be7f66ec2b2 |

Outputs:

| File | SHA-256 |
| --- | --- |
| facts.json | e406f8a10f637011935e5a498813d757cff8c7fa2eb2a6eec5f4d034d769f947 |
| figures/alt_rankings.json | ad5422a3e2801029f9a69d147554ac6c36d9a6fa9064f8ac23a0418bf8a2dcd4 |
| figures/burden_by_deprivation.json | a0f766e578479473605b0a240a240514814de8fe7aab2168bc7207dd044e2eee |
| figures/burden_by_tenure.json | a5d46eba16400056e5552ce37f3af4e346859ceab44b8e71415c05c38b79739b |
| figures/eviction_by_city.json | ac6dee2429bbe13f51db9672e39b47951944bd5a3dfae5d8b16c848f80bd86d3 |
| figures/eviction_by_race.json | 39480f8c3740b17173686d96f40a4a7ca10afbc6157e65cdb531a632f5df61fc |
| figures/eviction_filing_index.json | 6740df1888b777f54552d055596ded434d6cd6cbb8dfd6e583a8bcd4bc3b913d |
| figures/rent_standard_vs_burden.json | 7a4662dfacf266f30ec3d9fb75c844b95a8fee5b6e8b296927f65da78c0a22e1 |
| figures/renter_burden_map.json | 24a7f6d9409a64c9643639e89f382738501187daa0d762e7ff045b42448b8696 |
| meta.json | c5d8a211321f56303e9cec50eecb1f97d9aac8ed98f5a1f7cd4abbaa91bc32e4 |
| public/data/renting-affordability-and-eviction/fmr_scatter.json | b3c9d3d1222c54a96973a0026b091757fa86e646f547daf67c4d7b56b8b8befd |
| public/data/renting-affordability-and-eviction/renter_burden_map.json | 6ab626bb2f8043eca30ee59646103a51ff8750102535c15eee25aa67386abc37 |
| refs.json | 931fb8050c701301f487f4643e6931c0e68f23643434bd6613eb1ff5445cd0ca |

## Machine formats

- Report (HTML): https://vishalsingh.org/zipcode/renting-affordability-and-eviction/report
- Report (Markdown): https://vishalsingh.org/zipcode/renting-affordability-and-eviction/report.md
- Every estimate (JSON): https://vishalsingh.org/zipcode/report/renting-affordability-and-eviction/facts.json
- Table 1, figure "burden_by_tenure" (JSON): https://vishalsingh.org/zipcode/report/renting-affordability-and-eviction/figures/burden_by_tenure.json
- Table 2, figure "renter_burden_map" (JSON): https://vishalsingh.org/zipcode/report/renting-affordability-and-eviction/figures/renter_burden_map.json
- Table 3, figure "burden_by_deprivation" (JSON): https://vishalsingh.org/zipcode/report/renting-affordability-and-eviction/figures/burden_by_deprivation.json
- Table 4, figure "rent_standard_vs_burden" (JSON): https://vishalsingh.org/zipcode/report/renting-affordability-and-eviction/figures/rent_standard_vs_burden.json
- Table 5, figure "eviction_filing_index" (JSON): https://vishalsingh.org/zipcode/report/renting-affordability-and-eviction/figures/eviction_filing_index.json
- Table 6, figure "eviction_by_race" (JSON): https://vishalsingh.org/zipcode/report/renting-affordability-and-eviction/figures/eviction_by_race.json
- Table 7, figure "eviction_by_city" (JSON): https://vishalsingh.org/zipcode/report/renting-affordability-and-eviction/figures/eviction_by_city.json
- Table 8, figure "alt_rankings" (JSON): https://vishalsingh.org/zipcode/report/renting-affordability-and-eviction/figures/alt_rankings.json
- Payload of "renter_burden_map" (JSON): https://vishalsingh.org/zipcode/data/renting-affordability-and-eviction/renter_burden_map.json
- Payload of "rent_standard_vs_burden" (JSON): https://vishalsingh.org/zipcode/data/renting-affordability-and-eviction/fmr_scatter.json
- Index of every report: https://vishalsingh.org/zipcode/llms.txt; all reports in one file: https://vishalsingh.org/zipcode/llms-full.txt

---

# Air pollution: report

Chapter 6.1 of ZIP Code Destiny, by Vishal Singh (NYU Stern). Generated from the chapter's data files; the reading version is at https://vishalsingh.org/zipcode/air-pollution. Table n of this report is figure n of the chapter.

## Identification

- Title: Air pollution
- Label: The air you're assigned
- Chapter: 6.1 (Part VI, Environment, connection and well-being; An American life)
- Question: Who breathes the most polluted air, by neighborhood deprivation, race and historical redlining grade?
- Dek: The national traffic-pollution percentile of census tracts, modeled for 2020, has a correlation of −0.41 with the Area Deprivation Index across tracts but +0.19 within counties.
- Grain: neighborhoods
- Status: published
- Published: 2026-10-09
- Updated: not revised
- Data sources:
  - EPA EJScreen. EPA EJScreen 2024, tract file (Zenodo archive, release 2.32): national percentiles of 13 indicators on 2020 tracts; air quality modeled for 2020, toxic releases for 2021; years: 2024 (indicators); ACS 2018-2022 (demographics); weights: none (modeled indicators); tract population weights from the ACS
  - Area Deprivation Index. Neighborhood Atlas Area Deprivation Index 2023, national rank, tract values; years: 2023 release (version 4.0.1; built on ACS 2019-2023); weights: ACS 2019-2023 tract population
  - American Community Survey. ACS 2019-2023 tract population, race-alone and Hispanic or Latino shares, poverty rate; years: 2019-2023 (5-year); weights: Census Bureau survey weights applied upstream
  - CDC PLACES. CDC PLACES 2024 release, tract estimates of current asthma and COPD among adults (BRFSS 2022); years: 2024 release (tracts); 2025 release (ZCTAs, 2023 data year); weights: none (model-based estimates); ACS tract population
  - Mapping Inequality (HOLC maps). Mapping Inequality digitized HOLC maps (1935 to 1940), grades A to D, matched to 2020 tracts by area; years: 1930s-1940s maps (2023 digitization); weights: none (map polygons)
- Canonical URL: https://vishalsingh.org/zipcode/air-pollution
- Report: https://vishalsingh.org/zipcode/air-pollution/report
- Markdown: https://vishalsingh.org/zipcode/air-pollution/report.md
- BibTeX key: singh2026airpollution

```bibtex
@misc{singh2026airpollution,
  author = {Singh, Vishal},
  title = {Air pollution},
  year = {2026},
  url = {https://vishalsingh.org/zipcode/air-pollution},
  note = {In ZIP Code Destiny. Published 9 October 2026.}
}
```

## Principal estimates

The facts the dek and the opening prose cite, in order of appearance.

| Estimand | Population | Period | Value | Interval | n | Weights | Source | Fact id |
| --- | --- | --- | ---: | --- | ---: | --- | --- | --- |
| population-weighted correlation of the tract Traffic proximity national percentile with the ADI national rank, county-cluster bootstrap interval | residents of tracts with a Traffic proximity percentile and an ADI rank | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | −0.41 | 95% CI −0.49 to −0.33 | 82,198 tracts | ACS 2019-2023 tract population | EPA EJScreen | ptraf_adi |
| population-weighted correlation of the tract Traffic proximity national percentile with the ADI national rank, after removing each county's population-weighted mean from both (counties with at least 5 tracts), county-cluster bootstrap interval | residents of tracts in counties with at least 5 tracts that have a Traffic proximity percentile and the ADI national rank | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | +0.19 | 95% CI 0.15 to 0.22 | 79,730 tracts | ACS 2019-2023 tract population | EPA EJScreen | ptraf_adi_within |
| population-weighted mean of the tract Traffic proximity national percentile in an ADI decile (deciles hold equal numbers of tracts), county-cluster bootstrap interval | residents of tracts in the least deprived of the ten ADI deciles (decile 1) | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 75 | 95% CI 68 to 79 | 8,149 tracts | ACS 2019-2023 tract population | EPA EJScreen | ptraf_adi_d1 |
| population-weighted mean of the tract Traffic proximity national percentile in an ADI decile (deciles hold equal numbers of tracts), county-cluster bootstrap interval | residents of tracts in the most deprived of the ten ADI deciles (decile 10) | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 39 | 95% CI 35 to 42 | 8,203 tracts | ACS 2019-2023 tract population | EPA EJScreen | ptraf_adi_d10 |
| population-weighted correlation of the tract Traffic proximity national percentile with the ADI national rank, metropolitan tracts only, county-cluster bootstrap interval | residents of metropolitan tracts with a Traffic proximity percentile and the ADI national rank | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | −0.31 | 95% CI −0.40 to −0.21 | 66,593 tracts | ACS 2019-2023 tract population | EPA EJScreen | ptraf_adi_metro |
| population-weighted correlation of the tract Traffic proximity national percentile with the tract poverty rate, county-cluster bootstrap interval | residents of tracts with a Traffic proximity percentile and the tract poverty rate | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | +0.07 | 95% CI 0.04 to 0.11 | 82,818 tracts | ACS 2019-2023 tract population | EPA EJScreen | ptraf_pov |
| population-weighted correlation of the tract NO₂ national percentile with the tract's Hispanic or Latino share, county-cluster bootstrap interval | residents of tracts with a NO₂ percentile and a race and ethnicity share | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | +0.40 | 95% CI 0.35 to 0.46 | 83,003 tracts | ACS 2019-2023 tract population | EPA EJScreen | no2_hisp |
| population-weighted correlation of the tract NO₂ national percentile with the tract's Hispanic or Latino share after removing each county's population-weighted mean from both, county-cluster bootstrap interval | residents of tracts in counties with at least 5 tracts that have a NO₂ percentile and a race and ethnicity share | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | +0.23 | 95% CI 0.21 to 0.26 | 80,548 tracts | ACS 2019-2023 tract population | EPA EJScreen | no2_hisp_within |

## Findings

Traffic pollution ranks at a mean national percentile of 75 in the tenth of census tracts that are least deprived by the Area Deprivation Index (ADI) and 39 in the tenth that are most deprived. Across tracts, the correlation between the traffic percentile and the ADI rank is −0.41 (95% CI −0.49 to −0.33). Within a county the correlation is +0.19 (95% CI 0.15 to 0.22), so compared only with the other tracts of its county, a more deprived tract tends to rank higher on traffic. Among metropolitan tracts alone the correlation is still negative (−0.31 (95% CI −0.40 to −0.21)), and with the tract poverty rate in place of the ADI it is +0.07 (95% CI 0.04 to 0.11) across all tracts. For NO₂, diesel PM and traffic, the percentiles rise with the Hispanic share and with the Black share of a tract, across tracts and within counties. The correlation of the NO₂ percentile with Hispanic share is +0.40 (95% CI 0.35 to 0.46) across tracts and +0.23 (95% CI 0.21 to 0.26) within counties.

A percentile ranks a tract against other tracts and does not give a concentration. Every statement in this chapter is about tracts or about the residents of groups of tracts. None describes what an individual breathes, because people leave their tract during the day and the indicators are modeled estimates for particular years.

### What the pollution indicators measure

The indicators come from the Environmental Protection Agency's EJScreen. EPA removed the tool from its website in 2025, and the chapter uses the archived copy of the final 2024 release kept on Zenodo.[1] The tract file is on 2020 census tracts, as are the ADI, the American Community Survey (ACS) estimates for 2019 to 2023 and the CDC PLACES file used later, so the files join on the tract identifier with no crosswalk. Connecticut tracts carry the planning-region codes in all of them.

EPA's technical documentation dates each air indicator.[1] Fine particulate matter (PM₂.₅, in micrograms per cubic meter) and ozone (the annual mean of the ten highest daily maximum eight-hour readings, in parts per billion) are for 2020 and combine monitor readings with an air-quality model. NO₂ (parts per billion) is a 2020 estimate built from satellite data and a regression model. Diesel particulate matter (micrograms per cubic meter) is EPA's 2020 estimate from its air toxics screening. Toxic releases to air are a toxicity-weighted concentration modeled from 2021 reports, and traffic proximity is an index of 2020 traffic counts on nearby major roads, weighted by distance. The release has no air-toxics cancer-risk index. Besides the 6 air indicators, the file carries 4 indicators of proximity and discharge (Superfund sites, risk-management facilities, hazardous-waste facilities and wastewater) and 3 of old housing and infrastructure (housing built before 1960 as a proxy for lead paint, underground storage tanks and drinking-water violations), 13 in all. Each is a modeled or calculated value and none is a monitor reading at the tract.

Each indicator also comes as a national percentile. EPA's documentation describes percentiles ranked across block groups, but the percentiles in this file rank tracts, so a percentile is the share of tracts with a lower value. A percentile compares a tract with other tracts and says nothing about the level of a pollutant. The first and third quartiles of tract PM₂.₅ are 7.3 and 9.0 micrograms per cubic meter, so small differences in concentration separate tracts by many percentile points.

EJScreen has no PM₂.₅ or ozone estimate for the 1,524 tracts with residents in Alaska, Hawaii and Puerto Rico. A missing value is not a value of zero, so the analysis leaves those tracts out and every comparison covers the same places. It uses the 83,008 tracts of the contiguous states and the District of Columbia that have a positive ACS population, and they hold 330.1 million residents. 810 of those tracts have no ADI rank, and comparisons with the ADI leave them out, so 82,198 tracts have both a traffic percentile and an ADI rank. Means and correlations are population-weighted, with intervals from a bootstrap that redraws whole counties.

### Pollution percentiles fall as deprivation rises

Table 1 sorts tracts into tenths by the ADI national rank, with the first tenth the least deprived, and averages the pollution percentile of the tracts in each tenth. Every line falls from the first tenth to the ninth and edges up in the tenth. For traffic the means are 75 in the first tenth, 35 (95% CI 32 to 38) in the ninth and 39 in the tenth.

Across the tracts with both values the correlation with the ADI rank is −0.41 (95% CI −0.49 to −0.33) for traffic, −0.38 (95% CI −0.45 to −0.30) for diesel PM, −0.18 (95% CI −0.27 to −0.10) for NO₂ and −0.18 (95% CI −0.29 to −0.06) for PM₂.₅. Higher deprivation goes with lower pollution percentiles across the country, most strongly for traffic and diesel PM.

Table 1. Traffic pollution ranks at a mean national percentile of 75 in the least deprived tenth of tracts and 39 in the most deprived

A gap in percentiles is a gap in rank. For PM₂.₅ the mean percentiles are 64 (95% CI 52 to 73) in the first tenth, 44 (95% CI 41 to 47) in the ninth and 48 (95% CI 44 to 51) in the tenth. The mean PM₂.₅ concentration is 9.6 (95% CI 8.7 to 10.4) micrograms per cubic meter in the least deprived tenth and 8.1 (95% CI 7.9 to 8.2) in the most deprived. The intervals describe which places are in the file and leave out the error in EPA's models.

### Metropolitan tracts keep the fall, and the sign reverses within counties

Among metropolitan tracts the fall persists for traffic and diesel PM. One explanation is that the most deprived tracts are rural and that rural air ranks cleaner. Table 2 tests it in three tabs. The share of residents living in small-town and rural tracts, as the USDA classifies tracts, rises from 0.6% (95% CI 0.4% to 1.1%) in the least deprived tenth to 20.3% (95% CI 18.4% to 22.3%) in the ninth and eases to 18.5% (95% CI 16.2% to 21.0%) in the tenth, and it is 6.9% of residents overall. Those tracts hold a minority of residents in every tenth.

Among metropolitan tracts alone the traffic percentile still falls from 76 (95% CI 69 to 80) in the first tenth to 48 (95% CI 45 to 50) in the ninth. Diesel PM falls from 71 (95% CI 64 to 76) to 47 (95% CI 44 to 50).

The correlations of traffic and diesel PM with the ADI rank are −0.31 (95% CI −0.40 to −0.21) and −0.28 (95% CI −0.36 to −0.19) among metropolitan tracts, against −0.41 and −0.38 in all tracts. For NO₂ and PM₂.₅ the metropolitan correlations are −0.09 (95% CI −0.17 to 0.00) and −0.10 (95% CI −0.22 to 0.03), small and with intervals that include zero, and most of the fall in the metropolitan lines comes across the first three tenths. Leaving out the small-town, rural and micropolitan tracts changes the correlation less for traffic and diesel PM than for NO₂ and PM₂.₅, so rural tracts account for more of the national fall in the NO₂ and PM₂.₅ percentiles than of the traffic and diesel PM falls.

Table 2. The fall in traffic and diesel PM percentiles with deprivation persists among metropolitan tracts, and the sign reverses when tracts are compared within a county

The sign also depends on how deprivation is measured. The ADI descends from an index of census measures that includes median home value and median gross rent,[2] so places with high housing costs read as less deprived, a limit that Chapter 1.1, What the deprivation index measures (https://vishalsingh.org/zipcode/anatomy-of-deprivation) takes up. These data do not isolate that component, but the tract poverty rate does not show the negative sign. Across all tracts it correlates +0.07 (95% CI 0.04 to 0.11) with the traffic percentile and +0.18 (95% CI 0.15 to 0.21) with the NO₂ percentile.

It depends as well on which tracts are compared. Across all tracts, the comparison mixes places: a deprived tract in one county is compared with a less deprived tract in another county, which can differ in traffic, industry and housing costs. Within a county, each tract is compared only with the other tracts of its county, by removing the county's population-weighted mean from both variables. In that comparison, over the 79,730 tracts of counties that have enough tracts to compare, the ADI rank correlates +0.19 (95% CI 0.15 to 0.22) with the traffic percentile, +0.19 (95% CI 0.15 to 0.23) with diesel PM and +0.28 (95% CI 0.25 to 0.31) with NO₂. The poverty rate correlates +0.27 (95% CI 0.24 to 0.29) with the traffic percentile. Within a county, more deprived tracts have higher pollution percentiles.

Across counties the relation runs the other way. The correlation between a county's mean traffic percentile and its mean ADI rank is −0.66 (95% CI −0.73 to −0.58) across 3,095 counties, each weighted by its residents. The fall across the country reflects differences between counties.

### Pollution percentiles rise with the Hispanic and Black share of a tract

Early environmental-justice research documented waste facilities sited in Black communities in the South.[3] The same indicators that fall with the ADI rise with the racial and ethnic composition of a tract. Table 3 repeats the tenths design with the share of residents who are Hispanic or Latino, of any race, and the share who are Black or African American alone, of any ethnicity. The groups overlap, because a Hispanic resident can be Black, and they are 19.0% and 12.1% of residents.

The NO₂ percentile averages 33 (95% CI 30 to 36) in the tenth of tracts with the lowest Hispanic share and 73 (95% CI 68 to 76) in the tenth with the highest, and it averages 38 (95% CI 35 to 41) and 58 (95% CI 53 to 63) across Black-share tenths. Tracts with the same share go to the same tenth, so the lowest Black-share tenth holds all 8,815 tracts with no Black alone residents and the second holds correspondingly fewer.

Table 3. Pollution percentiles rise with the Hispanic share of a tract, from 33 to 73 for NO₂ across deciles, and with its Black share up to the fifth decile and in the tenth

In every panel the Hispanic-share lines rise from the second tenth to the tenth, and the Black-share lines rise through the fifth tenth, change little to the ninth and rise again in the tenth for traffic, diesel PM and NO₂. Across tracts, Hispanic share correlates +0.40 (95% CI 0.35 to 0.46) with the NO₂ percentile, +0.34 (95% CI 0.26 to 0.41) with diesel PM, +0.33 (95% CI 0.27 to 0.40) with traffic and +0.32 (95% CI 0.22 to 0.40) with PM₂.₅.

Black share correlates +0.14 (95% CI 0.10 to 0.19) with NO₂, +0.18 (95% CI 0.13 to 0.23) with diesel PM and +0.17 (95% CI 0.12 to 0.22) with traffic, and +0.06 (95% CI 0.00 to 0.12) with PM₂.₅, which is close to zero. For NO₂, diesel PM and traffic the correlations with both shares stay positive when states are redrawn instead of counties.

The correlations persist within counties. There the NO₂ percentile correlates +0.23 (95% CI 0.21 to 0.26) with Hispanic share and +0.17 (95% CI 0.14 to 0.21) with Black share, and the traffic percentile +0.19 (95% CI 0.16 to 0.21) and +0.16 (95% CI 0.14 to 0.19).

To see whether race and deprivation each go with pollution on their own, a joint model puts them together. In a weighted regression of the traffic percentile on the ADI rank, Hispanic share and Black share, each variable measured from its county mean and scaled to one standard deviation, the coefficients are +0.06 (95% CI 0.02 to 0.10) for the ADI rank, +0.17 (95% CI 0.14 to 0.21) for Hispanic share and +0.15 (95% CI 0.12 to 0.19) for Black share.

For NO₂ they are +0.16 (95% CI 0.13 to 0.19), +0.18 (95% CI 0.15 to 0.21) and +0.12 (95% CI 0.08 to 0.16). All are positive, and the ADI rank is the weakest for traffic.

With the poverty rate in place of the ADI rank, the traffic coefficients are +0.19 (95% CI 0.17 to 0.21) for poverty, +0.14 (95% CI 0.10 to 0.17) for Hispanic share and +0.10 (95% CI 0.07 to 0.14) for Black share. Within a county, poverty and the racial and ethnic composition of a tract each go with higher pollution percentiles, and these data do not rank them.

Weighting tracts by the residents of a group describes where the group lives. The tract of the average Hispanic resident has an NO₂ percentile of 63 (95% CI 59 to 67), that of the average Black alone resident 56 (95% CI 52 to 60), and that of the average resident 50 (95% CI 47 to 53). These are averages over tracts of residence and not measurements of individuals. A national model that attributes PM₂.₅ exposure to the consumption of goods and services finds that Black and Hispanic residents are exposed to more particulate matter than their own consumption accounts for, and non-Hispanic white residents to less.[4]

### Hazards stack in the same tracts

Single indicators understate a tract's burden when hazards coincide. Count, for each tract, how many of the 5 hazards PM₂.₅, diesel PM, NO₂, traffic proximity and toxic releases to air have a national percentile of at least 75. Table 4 shows the mean count by tenth for 82,993 tracts.

The count rises from 0.8 (95% CI 0.6 to 1.0) in the lowest Black-share tenth to 1.7 (95% CI 1.4 to 2.0) in the highest, and from 0.6 (95% CI 0.5 to 0.7) to 2.3 (95% CI 1.8 to 2.8) across Hispanic-share tenths. The Black-share line is flat from the sixth tenth to the ninth, so the highest tenth stands apart.

Across ADI tenths the count falls from 2.1 (95% CI 1.7 to 2.5) to 1.0 (95% CI 0.8 to 1.2). Ozone is not counted, because it forms in the air from other pollutants and does not mark a local source as the five counted hazards do. Adding it as a sixth hazard gives the same ordering of the three lines.

Table 4. Residents of tracts in the highest Black-share decile live with 1.7 stacked air hazards on average, against 0.8 in the lowest, and the score falls across deprivation deciles

The second view shows the share of residents living in tracts with 3 or more stacked hazards. It is 10% (95% CI 6% to 15%) in the lowest Black-share tenth and 32% (95% CI 23% to 40%) in the highest, and 6% (95% CI 4% to 10%) and 49% (95% CI 35% to 61%) across Hispanic-share tenths.

### Two components summarize the indicators

The indicators overlap, and principal components summarize them. The analysis standardizes each of the 13 national percentiles to mean zero and standard deviation one across the 80,754 tracts that have all of them, counts every tract once, and finds the weighted combinations of the indicators that capture the most variance. The first component holds 38.6% (95% CI 36.2% to 41.7%) of the variance and the second 11.7% (95% CI 10.5% to 13.1%). A population-weighted version gives almost the same components, with 37.1% in the first.

The first component has positive weights on every indicator except drinking-water violations, and the largest are on diesel PM (+0.40 (95% CI 0.38 to 0.41)), traffic proximity (+0.40 (95% CI 0.39 to 0.41)), hazardous-waste facilities (+0.39 (95% CI 0.37 to 0.40)) and NO₂ (+0.38 (95% CI 0.36 to 0.39)). It marks tracts that are dense and close to roads and facilities, and the chapter calls it the urban-industrial component.

The second weights underground storage tanks (+0.55 (95% CI 0.33 to 0.66)), lead paint (+0.45 (95% CI 0.33 to 0.56)) and drinking-water violations (+0.32 (95% CI 0.19 to 0.43)) positively and PM₂.₅ (−0.47 (95% CI −0.55 to −0.38)) and ozone (−0.31 (95% CI −0.38 to −0.16)) negatively, so it separates tracts with old housing and infrastructure from tracts with high PM₂.₅ and ozone (Table 5). The sign of a component is arbitrary, and the names describe the weights without testing them.

Table 5. The first component weights diesel PM, traffic, hazardous-waste facilities and NO₂ most, and the second weights storage tanks, lead paint and drinking-water violations against PM₂.₅ and ozone

The deprivation index relates differently to the two kinds of indicator. Storage tanks correlate +0.25 (95% CI 0.13 to 0.36) with the ADI rank and lead paint +0.16 (95% CI 0.06 to 0.26), against negative correlations for most of the air indicators.

### The urban-industrial component is highest in a few dense counties

Table 6 averages the tract scores on the first component within each county, weighting tracts by population. Cook County, Illinois, has the highest mean (+1.69, in standard deviations of the tract score), followed by Hudson County, New Jersey (+1.67), and New York County (+1.58). The independent city of St. Louis (+1.50) and the Bronx (+1.45) come next.

Los Angeles County has rank 6 among the mapped counties. The median county scores −1.2, so most counties sit below the tract average. The map covers 3,094 counties. The 49 gray counties are in Alaska and Hawaii, where EJScreen has no PM₂.₅ or ozone estimate, or are counties where fewer than 50% of residents live in tracts with all indicators. Puerto Rico is not drawn.

Table 6. The urban-industrial component is highest in Cook County, Illinois, and in dense counties around New York, and below the tract average in most counties

### Most air indicators correlate negatively with deprivation and positively with Hispanic and Black share

Table 7 lists the correlation of each of the 13 indicators with the ADI rank, Hispanic share and Black share. It shades a cell by the wider of two 95% intervals, one from a bootstrap over counties and one from a bootstrap over states, and the counts below require both to exclude zero. Of the 6 air indicators, 4 correlate negatively with the ADI rank: ozone, diesel PM, NO₂ and traffic. The PM₂.₅ correlation is negative under the county bootstrap and its interval includes zero when states are redrawn, and the correlation for toxic releases to air cannot be told from zero.

5 correlate positively with Hispanic share, all but toxic releases, and 4 with Black share: diesel PM, NO₂, traffic and toxic releases. The PM₂.₅ and ozone cells for Black share include zero. Of the 3 old-housing and infrastructure indicators, 1 is clearly positive with the ADI rank, lead paint. Storage tanks are positive under the county bootstrap and include zero when states are redrawn, and the interval for drinking-water violations includes zero.

Table 7. Pollution percentiles correlate negatively with deprivation for 4 of 6 air indicators and positively with Hispanic share for 5

### Modeled asthma does not follow the urban-industrial component

If the urban-industrial component tracked illness, modeled asthma would rise with it. CDC PLACES estimates the prevalence of current asthma among adults for every tract from a model fitted to the 2022 Behavioral Risk Factor Surveillance System, with age, sex, race and ethnicity, education and county poverty as inputs.[5] A tract's value therefore reflects who lives there and where, and a correlation of PLACES values with the ADI or with race shares is partly built into the model. The tract file has no intervals, and Chapter 7.1, Health behaviors and the supply of care (https://vishalsingh.org/zipcode/health-behaviors-and-the-supply-of-care) discusses the same limit for other PLACES measures.

Table 8 ranks tracts in twentieths on each of three measures. Modeled asthma is highest in the lowest twentieth on the urban-industrial component (11.2% (95% CI 11.1% to 11.3%)) and lowest in the highest (9.9% (95% CI 9.7% to 10.5%)). It rises from 9.7% (95% CI 9.5% to 9.9%) to 11.0% (95% CI 10.6% to 11.5%) across lead-paint twentieths, and from 9.0% (95% CI 8.9% to 9.3%) to 12.7% (95% CI 12.4% to 12.9%) across deprivation twentieths.

Across tracts the urban-industrial component correlates −0.10 (95% CI −0.18 to 0.00) with modeled asthma and −0.37 (95% CI −0.42 to −0.32) with modeled COPD, the lead-paint percentile correlates +0.27 (95% CI 0.20 to 0.34) with asthma, and the ADI rank +0.53 (95% CI 0.48 to 0.58). The deprivation line is the one most closely tied to the model's inputs, and the flat-to-falling urban-industrial line says nothing about illness itself, because the values are modeled.

Table 8. Modeled asthma prevalence is highest in the lowest twentieth of tracts on the urban-industrial component (11.2%) and rises with the lead-paint percentile and with deprivation

### Tracts graded D in the 1930s have higher traffic, NO₂ and diesel percentiles

Between 1935 and 1940 the Home Owners' Loan Corporation drew residential security maps for cities and graded neighborhoods from A, labeled best, to D, labeled hazardous.[6] The grades in general are the subject of Chapter 5.2, Home value, income and redlining (https://vishalsingh.org/zipcode/home-value-income-and-redlining), and this chapter uses them to look at air only. Each tract takes the grade that covers most of its area, among tracts where graded polygons cover at least 50% of the tract. 10,950 tracts qualify, holding 37.9 million residents in 215 counties.

Pooled across those counties, the traffic percentile steps up from 78 (95% CI 75 to 80) in grade A tracts to 86 (95% CI 83 to 89) in grade D tracts. NO₂ goes from 73 (95% CI 70 to 76) to 86 (95% CI 83 to 89), and diesel PM from 73 (95% CI 69 to 77) to 86 (95% CI 81 to 89).

The stacked-hazard score goes from 2.3 (95% CI 2.0 to 2.5) to 3.2 (95% CI 2.8 to 3.5). PM₂.₅ moves less, from 56 (95% CI 49 to 63) to 60 (95% CI 51 to 69), and the two pooled intervals overlap. Pooled gaps mix the differences between grades with the differences between the cities that were graded.

Comparing tracts within the same county separates the two. In the 199 counties that hold more than one grade, grade D tracts rank 3.9 (95% CI 2.7 to 5.1) points above grade A tracts on the traffic percentile, 9.7 (95% CI 7.2 to 12.2) on NO₂, 5.1 (95% CI 3.9 to 6.5) on diesel PM and 1.1 (95% CI 0.2 to 2.0) on PM₂.₅, and their stacked-hazard score is 0.6 (95% CI 0.4 to 0.7) higher (Table 9). Each within-county gap is smaller than the pooled gap, and the PM₂.₅ gap is the smallest with an interval that excludes zero. A comparison of the same grades across U.S. cities with modeled NO₂ and PM₂.₅ for 2010 also finds higher NO₂ in grade D than in grade A neighborhoods, and a smaller difference for PM₂.₅.[7] The graded tracts are dense urban cores, so even grade A tracts rank above the national middle for traffic and NO₂.

Table 9. Within the same county, tracts graded D in the 1930s rank 3.9 points above grade A tracts on the traffic percentile and 9.7 points on NO₂, and the stacked-hazard score is 0.6 higher

> Causal evidence: Whether redlining and traffic shape the air a neighborhood receives (Boundary design; difference-in-differences)
>
> The comparisons above describe tracts and do not test whether a grade changed the air. Two studies use designs that bear on parts of the question, and a third, in the next box, follows hazardous-waste sites over time.
>
> Salazar-Miranda and coauthors compared neighborhoods on either side of HOLC boundaries within cities, a boundary design. The lower-graded side has higher current and projected heat and flood risk, less tree canopy and less permeable ground.[8] Their outcomes are climate risks and not air pollution, so they do not test whether the grade changed air quality.
>
> Currie and Walker compared births to mothers living near toll plazas with births to mothers living farther from them, before and after E-ZPass reduced congestion at the plazas, a difference-in-differences design. Prematurity and low birth weight fell among mothers near the plazas relative to mothers farther away, which suggests that traffic congestion near a road harms infants.[9] It concerns plazas that adopted electronic tolling and does not say how much of a neighborhood's air pollution comes from a 1930s grade.

> Study evidence: Siting of hazardous-waste facilities and neighborhood demographics (Longitudinal comparison of siting)
>
> Mohai and Saha followed commercial hazardous-waste facilities sited between 1966 and 1995 and compared the neighborhoods around them at the time of siting and in later census years. The neighborhoods already had disproportionate shares of people of color and low-income residents when the facilities were sited, and demographic change had begun before siting.[10] The design is a longitudinal observational comparison. It shows which came first and does not identify an effect.

### Under the standardized index the ends of the traffic gradient no longer differ

Over the 75,058 tracts that have a traffic percentile and all four rankings, the mean traffic percentile under the ADI falls from 75 (95% CI 69 to 80) in the least deprived tenth to 38 (95% CI 35 to 41) in the most deprived, with a rank correlation between tenth and mean of −0.96 (Table 10). Under the standardized index the line falls into the middle tenths and climbs back at the deprived end, so the two ends, 62 (95% CI 58 to 65) and 59 (95% CI 55 to 64), have overlapping intervals and the rank correlation is −0.31. Under median household income the mean falls from 62 (95% CI 58 to 66) to 50 (95% CI 46 to 53) and rises again toward the deprived end (rank correlation −0.70), and under the bachelor's degree share it falls from 69 (95% CI 65 to 72) to 48 (95% CI 42 to 53) (−0.87). Under the rule in Chapter 9.1, How far one ranking goes (https://vishalsingh.org/zipcode/how-far-one-ranking-goes) the standardized index changes this gradient, since its ends do not differ, and income changes its shape, while the degree share leaves it unchanged, with a rank correlation that moves by less than 0.1.

Table 10. Traffic proximity is steadily better toward the more deprived end under the ADI; the two ends do not differ detectably under the standardized index, and it is better at the more deprived end but uneven along the way under the income ranking

### What these data cannot show

Percentiles rank tracts and do not give a level of exposure, and a percentile cannot say whether a concentration meets a health guideline. The indicators are EPA model output for 2020 and 2021 with their own error, assigned to tracts from finer geography. The negative link between deprivation and pollution percentiles depends on the measure of deprivation and on the comparison, and each comparison answers a different question. The ADI includes home values and rents, the poverty rate does not, and a within-county comparison holds the metropolitan area fixed while it discards differences between areas.

Everything here is ecological. The shares of Hispanic and Black residents are ACS estimates for 2019 to 2023 without margins of error, the two groups overlap, and a tract average does not describe any one of its residents. Counties in one state share traffic, industry and climate, so the county bootstrap gives intervals narrower than one that redraws whole states. With states redrawn the correlation of the traffic percentile with the ADI rank is −0.41 (95% CI −0.52 to −0.25), which still excludes zero, and that of the PM₂.₅ percentile is −0.18 (95% CI −0.38 to 0.10), which includes it.

The principal components are a description. They use national percentiles, which discard the differences in concentration, and they cover only the tracts with all indicators, so Alaska, Hawaii and Puerto Rico are absent. County scores are means of modeled tract values. PLACES values are modeled from survey responses and demographic inputs, they are crude prevalence among adults, and asthma and COPD depend on age and smoking in ways the tract file does not separate. The HOLC comparison is among graded urban cores, matches a 1930s polygon to a 2020 tract by area, and describes the residents of today. It cannot separate what the grades did from the features of the neighborhoods that surveyors graded, and the box above shows what the available designs do and do not cover.

## Tables

### Table 1. Traffic pollution ranks at a mean national percentile of 75 in the least deprived tenth of tracts and 39 in the most deprived

Mean national percentile of four EJScreen air indicators by decile of the Area Deprivation Index national rank, population-weighted, census tracts

| conf_high | conf_low | decile | estimate | hazard | n_units |
| ---: | ---: | ---: | ---: | --- | ---: |
| 79.18 | 68.05 | 1 | 74.70 | Traffic proximity | 8,149 |
| 71.29 | 58.96 | 2 | 65.23 | Traffic proximity | 8,293 |
| 60.23 | 53.49 | 3 | 56.92 | Traffic proximity | 8,205 |
| 53.77 | 48.87 | 4 | 51.45 | Traffic proximity | 8,184 |
| 50.34 | 44.93 | 5 | 47.72 | Traffic proximity | 8,416 |
| 47.13 | 41.66 | 6 | 44.43 | Traffic proximity | 8,052 |
| 44.32 | 38.17 | 7 | 41.19 | Traffic proximity | 8,161 |
| 40.55 | 35.39 | 8 | 38.01 | Traffic proximity | 8,288 |
| 37.53 | 32.20 | 9 | 34.92 | Traffic proximity | 8,247 |
| 42.13 | 35.42 | 10 | 38.99 | Traffic proximity | 8,203 |
| 75.36 | 62.13 | 1 | 69.69 | Diesel PM | 8,149 |
| 71.11 | 59.70 | 2 | 65.61 | Diesel PM | 8,293 |
| 61.41 | 54.39 | 3 | 58.14 | Diesel PM | 8,205 |
| 56.45 | 50.53 | 4 | 53.64 | Diesel PM | 8,184 |
| 52.62 | 46.70 | 5 | 49.71 | Diesel PM | 8,416 |
| 49.53 | 43.50 | 6 | 46.55 | Diesel PM | 8,052 |
| 46.27 | 39.94 | 7 | 43.07 | Diesel PM | 8,161 |
| 42.12 | 36.95 | 8 | 39.46 | Diesel PM | 8,288 |
| 38.91 | 33.80 | 9 | 36.35 | Diesel PM | 8,247 |
| 42 | 35.94 | 10 | 39.13 | Diesel PM | 8,203 |
| 67.58 | 54.21 | 1 | 61.37 | NO₂ | 8,146 |
| 63.92 | 51.85 | 2 | 58.08 | NO₂ | 8,292 |
| 56.05 | 48.39 | 3 | 52.30 | NO₂ | 8,204 |
| 53.46 | 47.20 | 4 | 50.45 | NO₂ | 8,184 |
| 51.82 | 45.39 | 5 | 48.71 | NO₂ | 8,416 |
| 50.96 | 44.17 | 6 | 47.59 | NO₂ | 8,052 |
| 49.47 | 42.09 | 7 | 45.71 | NO₂ | 8,161 |
| 46.95 | 40.92 | 8 | 43.89 | NO₂ | 8,288 |
| 45.25 | 39.68 | 9 | 42.52 | NO₂ | 8,247 |
| 50.71 | 44.27 | 10 | 47.64 | NO₂ | 8,203 |
| 73.24 | 52.34 | 1 | 64.27 | PM₂.₅ | 8,149 |
| 65.84 | 47.61 | 2 | 56.82 | PM₂.₅ | 8,293 |
| 56.99 | 47.69 | 3 | 52.29 | PM₂.₅ | 8,205 |
| 54.11 | 47.20 | 4 | 50.75 | PM₂.₅ | 8,184 |
| 51.99 | 46.09 | 5 | 49.21 | PM₂.₅ | 8,416 |
| 51.28 | 44.87 | 6 | 48.16 | PM₂.₅ | 8,052 |
| 50.70 | 44.20 | 7 | 47.40 | PM₂.₅ | 8,161 |
| 47.38 | 41.64 | 8 | 44.47 | PM₂.₅ | 8,288 |
| 46.62 | 41.50 | 9 | 44.06 | PM₂.₅ | 8,247 |
| 50.97 | 44.29 | 10 | 47.76 | PM₂.₅ | 8,203 |

Note: National percentiles rank modeled values, so a gap in percentiles is not a gap in concentrations: the mean PM₂.₅ concentration is 9.6 micrograms per cubic meter in the least deprived decile and 8.1 in the most deprived.

Note: Each point rests on the tracts of a decile that have a value for the indicator, and the tooltip gives the count. The 1,524 tracts with residents in Alaska, Hawaii and Puerto Rico have no PM₂.₅ or ozone estimate and are outside the analysis, and 810 tracts in it have no ADI rank.

Note: The Area Deprivation Index descends from an index that includes median home value and median gross rent, so it rates dense, high-cost metropolitan tracts as less deprived than their incomes suggest.

Source: EPA EJScreen 2024, Area Deprivation Index 2023 and ACS 2019-2023; population-weighted means, modeled estimates

Data: https://vishalsingh.org/zipcode/report/air-pollution/figures/air_vs_deprivation.json

### Table 2. The fall in traffic and diesel PM percentiles with deprivation persists among metropolitan tracts, and the sign reverses when tracts are compared within a county

Three views of the link between the Area Deprivation Index and national pollution percentiles: rural share, metropolitan tracts only, and correlations by comparison

#### Table 2.1. Rural share

The share of residents in small-town and rural tracts rises from 0.6% in the least deprived decile to 20.3% in the ninth and 18.5% in the tenth

Share of residents living in small-town and rural tracts by decile of the Area Deprivation Index national rank, population-weighted

| conf_high | conf_low | decile | n_units | rural |
| ---: | ---: | --- | ---: | ---: |
| 1.10 | 0.400 | 1 | 8,077 | 0.600 |
| 1.40 | 0.700 | 2 | 8,222 | 1 |
| 1.70 | 1 | 3 | 8,064 | 1.30 |
| 2.80 | 1.80 | 4 | 8,050 | 2.20 |
| 4.20 | 2.90 | 5 | 8,271 | 3.50 |
| 6.90 | 5 | 6 | 7,946 | 5.90 |
| 10.30 | 7.80 | 7 | 8,077 | 8.90 |
| 15.50 | 12.50 | 8 | 8,207 | 13.90 |
| 22.30 | 18.40 | 9 | 8,214 | 20.30 |
| 21 | 16.20 | 10 | 8,194 | 18.50 |

#### Table 2.2. Metropolitan tracts only

Among metropolitan tracts alone the traffic percentile still falls from 76 to 48 and diesel PM from 71 to 47, while the correlations for NO₂ and PM₂.₅ shrink to −0.09 and −0.10

Mean national percentile of four air indicators by decile of the Area Deprivation Index national rank, all tracts and metropolitan tracts only, population-weighted

| conf_high | conf_low | decile | estimate | hazard | n_units | series |
| ---: | ---: | ---: | ---: | --- | ---: | --- |
| 79.18 | 68.05 | 1 | 74.70 | Traffic proximity | 8,149 | All tracts |
| 71.29 | 58.96 | 2 | 65.23 | Traffic proximity | 8,293 | All tracts |
| 60.23 | 53.49 | 3 | 56.92 | Traffic proximity | 8,205 | All tracts |
| 53.77 | 48.87 | 4 | 51.45 | Traffic proximity | 8,184 | All tracts |
| 50.34 | 44.93 | 5 | 47.72 | Traffic proximity | 8,416 | All tracts |
| 47.13 | 41.66 | 6 | 44.43 | Traffic proximity | 8,052 | All tracts |
| 44.32 | 38.17 | 7 | 41.19 | Traffic proximity | 8,161 | All tracts |
| 40.55 | 35.39 | 8 | 38.01 | Traffic proximity | 8,288 | All tracts |
| 37.53 | 32.20 | 9 | 34.92 | Traffic proximity | 8,247 | All tracts |
| 42.13 | 35.42 | 10 | 38.99 | Traffic proximity | 8,203 | All tracts |
| 79.79 | 69.40 | 1 | 75.65 | Traffic proximity | 7,879 | Metropolitan tracts |
| 71.89 | 60.29 | 2 | 66.37 | Traffic proximity | 7,945 | Metropolitan tracts |
| 61.48 | 54.76 | 3 | 58.28 | Traffic proximity | 7,676 | Metropolitan tracts |
| 55.86 | 50.82 | 4 | 53.44 | Traffic proximity | 7,427 | Metropolitan tracts |
| 53.17 | 47.75 | 5 | 50.40 | Traffic proximity | 7,330 | Metropolitan tracts |
| 51.12 | 45.58 | 6 | 48.42 | Traffic proximity | 6,665 | Metropolitan tracts |
| 50.69 | 44.36 | 7 | 47.60 | Traffic proximity | 6,166 | Metropolitan tracts |
| 49.78 | 44.36 | 8 | 47.14 | Traffic proximity | 5,549 | Metropolitan tracts |
| 50.34 | 44.52 | 9 | 47.55 | Traffic proximity | 4,817 | Metropolitan tracts |
| 56.28 | 49.08 | 10 | 52.93 | Traffic proximity | 5,139 | Metropolitan tracts |
| 75.36 | 62.13 | 1 | 69.69 | Diesel PM | 8,149 | All tracts |
| 71.11 | 59.70 | 2 | 65.61 | Diesel PM | 8,293 | All tracts |
| 61.41 | 54.39 | 3 | 58.14 | Diesel PM | 8,205 | All tracts |
| 56.45 | 50.53 | 4 | 53.64 | Diesel PM | 8,184 | All tracts |
| 52.62 | 46.70 | 5 | 49.71 | Diesel PM | 8,416 | All tracts |
| 49.53 | 43.50 | 6 | 46.55 | Diesel PM | 8,052 | All tracts |
| 46.27 | 39.94 | 7 | 43.07 | Diesel PM | 8,161 | All tracts |
| 42.12 | 36.95 | 8 | 39.46 | Diesel PM | 8,288 | All tracts |
| 38.91 | 33.80 | 9 | 36.35 | Diesel PM | 8,247 | All tracts |
| 42 | 35.94 | 10 | 39.13 | Diesel PM | 8,203 | All tracts |
| 76.27 | 63.75 | 1 | 70.76 | Diesel PM | 7,879 | Metropolitan tracts |
| 72.08 | 61.58 | 2 | 67.05 | Diesel PM | 7,945 | Metropolitan tracts |
| 63.41 | 56.48 | 3 | 60.05 | Diesel PM | 7,676 | Metropolitan tracts |
| 59.21 | 53.22 | 4 | 56.28 | Diesel PM | 7,427 | Metropolitan tracts |
| 56.03 | 50.15 | 5 | 53.15 | Diesel PM | 7,330 | Metropolitan tracts |
| 54.05 | 48.05 | 6 | 51.04 | Diesel PM | 6,665 | Metropolitan tracts |
| 52.79 | 46.25 | 7 | 49.48 | Diesel PM | 6,166 | Metropolitan tracts |
| 50.75 | 45.10 | 8 | 47.96 | Diesel PM | 5,549 | Metropolitan tracts |
| 50.21 | 44.25 | 9 | 47.41 | Diesel PM | 4,817 | Metropolitan tracts |
| 54.37 | 46.45 | 10 | 50.87 | Diesel PM | 5,139 | Metropolitan tracts |
| 67.58 | 54.21 | 1 | 61.37 | NO₂ | 8,146 | All tracts |
| 63.92 | 51.85 | 2 | 58.08 | NO₂ | 8,292 | All tracts |
| 56.05 | 48.39 | 3 | 52.30 | NO₂ | 8,204 | All tracts |
| 53.46 | 47.20 | 4 | 50.45 | NO₂ | 8,184 | All tracts |
| 51.82 | 45.39 | 5 | 48.71 | NO₂ | 8,416 | All tracts |
| 50.96 | 44.17 | 6 | 47.59 | NO₂ | 8,052 | All tracts |
| 49.47 | 42.09 | 7 | 45.71 | NO₂ | 8,161 | All tracts |
| 46.95 | 40.92 | 8 | 43.89 | NO₂ | 8,288 | All tracts |
| 45.25 | 39.68 | 9 | 42.52 | NO₂ | 8,247 | All tracts |
| 50.71 | 44.27 | 10 | 47.64 | NO₂ | 8,203 | All tracts |
| 67.98 | 55.13 | 1 | 62.30 | NO₂ | 7,876 | Metropolitan tracts |
| 64.62 | 53.25 | 2 | 59.20 | NO₂ | 7,944 | Metropolitan tracts |
| 57.75 | 49.76 | 3 | 53.86 | NO₂ | 7,675 | Metropolitan tracts |
| 55.46 | 49.30 | 4 | 52.47 | NO₂ | 7,427 | Metropolitan tracts |
| 54.41 | 47.81 | 5 | 51.10 | NO₂ | 7,330 | Metropolitan tracts |
| 54.38 | 47.18 | 6 | 50.78 | NO₂ | 6,665 | Metropolitan tracts |
| 54.58 | 46.54 | 7 | 50.48 | NO₂ | 6,166 | Metropolitan tracts |
| 54.22 | 47.16 | 8 | 50.68 | NO₂ | 5,549 | Metropolitan tracts |
| 55.82 | 48.80 | 9 | 52.48 | NO₂ | 4,817 | Metropolitan tracts |
| 62.79 | 55.61 | 10 | 59.35 | NO₂ | 5,139 | Metropolitan tracts |
| 73.24 | 52.34 | 1 | 64.27 | PM₂.₅ | 8,149 | All tracts |
| 65.84 | 47.61 | 2 | 56.82 | PM₂.₅ | 8,293 | All tracts |
| 56.99 | 47.69 | 3 | 52.29 | PM₂.₅ | 8,205 | All tracts |
| 54.11 | 47.20 | 4 | 50.75 | PM₂.₅ | 8,184 | All tracts |
| 51.99 | 46.09 | 5 | 49.21 | PM₂.₅ | 8,416 | All tracts |
| 51.28 | 44.87 | 6 | 48.16 | PM₂.₅ | 8,052 | All tracts |
| 50.70 | 44.20 | 7 | 47.40 | PM₂.₅ | 8,161 | All tracts |
| 47.38 | 41.64 | 8 | 44.47 | PM₂.₅ | 8,288 | All tracts |
| 46.62 | 41.50 | 9 | 44.06 | PM₂.₅ | 8,247 | All tracts |
| 50.97 | 44.29 | 10 | 47.76 | PM₂.₅ | 8,203 | All tracts |
| 73.63 | 53.27 | 1 | 65.02 | PM₂.₅ | 7,879 | Metropolitan tracts |
| 66.11 | 48.04 | 2 | 57.46 | PM₂.₅ | 7,945 | Metropolitan tracts |
| 57.63 | 48.12 | 3 | 52.91 | PM₂.₅ | 7,676 | Metropolitan tracts |
| 55.43 | 48.45 | 4 | 52.03 | PM₂.₅ | 7,427 | Metropolitan tracts |
| 54.20 | 48.03 | 5 | 51.20 | PM₂.₅ | 7,330 | Metropolitan tracts |
| 54.48 | 48.07 | 6 | 51.20 | PM₂.₅ | 6,665 | Metropolitan tracts |
| 55.59 | 48.80 | 7 | 52.08 | PM₂.₅ | 6,166 | Metropolitan tracts |
| 54.13 | 47.55 | 8 | 50.83 | PM₂.₅ | 5,549 | Metropolitan tracts |
| 55.03 | 48.33 | 9 | 51.78 | PM₂.₅ | 4,817 | Metropolitan tracts |
| 59.38 | 50.89 | 10 | 55.33 | PM₂.₅ | 5,139 | Metropolitan tracts |

#### Table 2.3. Correlations by comparison

The correlation of the traffic percentile with the ADI rank is −0.41 across the tracts that have both and +0.19 within counties

Population-weighted correlation of four national pollution percentiles with the ADI national rank and with the tract poverty rate, in three comparisons

| comparison | conf_high | conf_low | hazard | measure | n | r |
| --- | ---: | ---: | --- | --- | ---: | ---: |
| All tracts | −0.331 | −0.492 | Traffic proximity | Area Deprivation Index rank | 82,198 | −0.413 |
| All tracts | 0.112 | 0.037 | Traffic proximity | Poverty rate | 82,818 | 0.075 |
| Metropolitan tracts | −0.208 | −0.398 | Traffic proximity | Area Deprivation Index rank | 66,593 | −0.306 |
| Metropolitan tracts | 0.200 | 0.123 | Traffic proximity | Poverty rate | 67,063 | 0.162 |
| Within counties | 0.225 | 0.149 | Traffic proximity | Area Deprivation Index rank | 79,730 | 0.188 |
| Within counties | 0.288 | 0.244 | Traffic proximity | Poverty rate | 80,353 | 0.267 |
| All tracts | −0.301 | −0.452 | Diesel PM | Area Deprivation Index rank | 82,198 | −0.377 |
| All tracts | 0.098 | 0.025 | Diesel PM | Poverty rate | 82,818 | 0.064 |
| Metropolitan tracts | −0.192 | −0.364 | Diesel PM | Area Deprivation Index rank | 66,593 | −0.280 |
| Metropolitan tracts | 0.180 | 0.098 | Diesel PM | Poverty rate | 67,063 | 0.141 |
| Within counties | 0.226 | 0.147 | Diesel PM | Area Deprivation Index rank | 79,730 | 0.188 |
| Within counties | 0.272 | 0.226 | Diesel PM | Poverty rate | 80,353 | 0.250 |
| All tracts | −0.103 | −0.266 | NO₂ | Area Deprivation Index rank | 82,193 | −0.182 |
| All tracts | 0.209 | 0.145 | NO₂ | Poverty rate | 82,816 | 0.178 |
| Metropolitan tracts | 0.004 | −0.173 | NO₂ | Area Deprivation Index rank | 66,588 | −0.085 |
| Metropolitan tracts | 0.277 | 0.215 | NO₂ | Poverty rate | 67,060 | 0.247 |
| Within counties | 0.305 | 0.252 | NO₂ | Area Deprivation Index rank | 79,725 | 0.279 |
| Within counties | 0.340 | 0.301 | NO₂ | Poverty rate | 80,350 | 0.321 |
| All tracts | −0.060 | −0.289 | PM₂.₅ | Area Deprivation Index rank | 82,198 | −0.176 |
| All tracts | 0.072 | 0.003 | PM₂.₅ | Poverty rate | 82,821 | 0.038 |
| Metropolitan tracts | 0.029 | −0.222 | PM₂.₅ | Area Deprivation Index rank | 66,593 | −0.102 |
| Metropolitan tracts | 0.103 | 0.025 | PM₂.₅ | Poverty rate | 67,065 | 0.064 |
| Within counties | 0.112 | 0.008 | PM₂.₅ | Area Deprivation Index rank | 79,730 | 0.063 |
| Within counties | 0.105 | 0.035 | PM₂.₅ | Poverty rate | 80,355 | 0.073 |

Source: EPA EJScreen 2024, Area Deprivation Index 2023 and ACS 2019-2023; population-weighted means, modeled estimates

Data: https://vishalsingh.org/zipcode/report/air-pollution/figures/rural_share_and_pm25.json

### Table 3. Pollution percentiles rise with the Hispanic share of a tract, from 33 to 73 for NO₂ across deciles, and with its Black share up to the fifth decile and in the tenth

Mean national percentile of four EJScreen air indicators by decile of the tract's Hispanic or Latino share and of its Black alone share, population-weighted

| conf_high | conf_low | decile | estimate | group | hazard | n_units |
| ---: | ---: | ---: | ---: | --- | --- | ---: |
| 33.11 | 27.53 | 1 | 30.28 | Hispanic or Latino share | Traffic proximity | 8,295 |
| 37.41 | 32.83 | 2 | 35.08 | Hispanic or Latino share | Traffic proximity | 8,301 |
| 42.07 | 37.43 | 3 | 39.80 | Hispanic or Latino share | Traffic proximity | 8,301 |
| 46.83 | 42.22 | 4 | 44.55 | Hispanic or Latino share | Traffic proximity | 8,301 |
| 51.21 | 46.17 | 5 | 48.74 | Hispanic or Latino share | Traffic proximity | 8,300 |
| 54.44 | 48.92 | 6 | 51.80 | Hispanic or Latino share | Traffic proximity | 8,300 |
| 58.25 | 51.88 | 7 | 55.26 | Hispanic or Latino share | Traffic proximity | 8,301 |
| 61.57 | 54.47 | 8 | 58.22 | Hispanic or Latino share | Traffic proximity | 8,301 |
| 65.67 | 58.11 | 9 | 62.12 | Hispanic or Latino share | Traffic proximity | 8,298 |
| 73.75 | 61.15 | 10 | 67.72 | Hispanic or Latino share | Traffic proximity | 8,300 |
| 38.13 | 30.82 | 1 | 34.38 | Black alone share | Traffic proximity | 8,811 |
| 41.66 | 33.58 | 2 | 37.52 | Black alone share | Traffic proximity | 7,786 |
| 49.06 | 40.54 | 3 | 44.80 | Black alone share | Traffic proximity | 8,301 |
| 54.03 | 46.09 | 4 | 50.15 | Black alone share | Traffic proximity | 8,301 |
| 57.57 | 50.14 | 5 | 54.04 | Black alone share | Traffic proximity | 8,301 |
| 59.11 | 51.61 | 6 | 55.41 | Black alone share | Traffic proximity | 8,300 |
| 57.72 | 51.71 | 7 | 54.72 | Black alone share | Traffic proximity | 8,301 |
| 57.64 | 51.82 | 8 | 54.74 | Black alone share | Traffic proximity | 8,299 |
| 57.05 | 51.07 | 9 | 54.11 | Black alone share | Traffic proximity | 8,300 |
| 63.65 | 56.11 | 10 | 60.09 | Black alone share | Traffic proximity | 8,298 |
| 34.27 | 29.53 | 1 | 31.90 | Hispanic or Latino share | Diesel PM | 8,295 |
| 37.78 | 33.56 | 2 | 35.65 | Hispanic or Latino share | Diesel PM | 8,301 |
| 41.92 | 37.69 | 3 | 39.83 | Hispanic or Latino share | Diesel PM | 8,301 |
| 46.83 | 42.35 | 4 | 44.62 | Hispanic or Latino share | Diesel PM | 8,301 |
| 51.98 | 46.55 | 5 | 49.36 | Hispanic or Latino share | Diesel PM | 8,300 |
| 55.05 | 49.39 | 6 | 52.40 | Hispanic or Latino share | Diesel PM | 8,300 |
| 58.95 | 52.77 | 7 | 56.06 | Hispanic or Latino share | Diesel PM | 8,301 |
| 62.36 | 55.56 | 8 | 59.19 | Hispanic or Latino share | Diesel PM | 8,301 |
| 66.91 | 59.68 | 9 | 63.61 | Hispanic or Latino share | Diesel PM | 8,298 |
| 74.51 | 61.29 | 10 | 68.86 | Hispanic or Latino share | Diesel PM | 8,300 |
| 38.10 | 30.66 | 1 | 34.37 | Black alone share | Diesel PM | 8,811 |
| 41.66 | 34.07 | 2 | 37.81 | Black alone share | Diesel PM | 7,786 |
| 48.49 | 40.80 | 3 | 44.63 | Black alone share | Diesel PM | 8,301 |
| 53.41 | 45.86 | 4 | 49.76 | Black alone share | Diesel PM | 8,301 |
| 57.92 | 50.84 | 5 | 54.49 | Black alone share | Diesel PM | 8,301 |
| 60.10 | 52.85 | 6 | 56.63 | Black alone share | Diesel PM | 8,300 |
| 59.47 | 53.58 | 7 | 56.62 | Black alone share | Diesel PM | 8,301 |
| 59.25 | 53.41 | 8 | 56.45 | Black alone share | Diesel PM | 8,299 |
| 58.75 | 52.52 | 9 | 55.78 | Black alone share | Diesel PM | 8,300 |
| 65.14 | 56.84 | 10 | 61.15 | Black alone share | Diesel PM | 8,298 |
| 35.63 | 29.91 | 1 | 32.64 | Hispanic or Latino share | NO₂ | 8,300 |
| 37.85 | 33.41 | 2 | 35.66 | Hispanic or Latino share | NO₂ | 8,300 |
| 40.50 | 36.41 | 3 | 38.47 | Hispanic or Latino share | NO₂ | 8,299 |
| 44.55 | 40.42 | 4 | 42.50 | Hispanic or Latino share | NO₂ | 8,300 |
| 48.59 | 43.65 | 5 | 46.07 | Hispanic or Latino share | NO₂ | 8,301 |
| 52 | 46.50 | 6 | 49.30 | Hispanic or Latino share | NO₂ | 8,300 |
| 56.32 | 50.50 | 7 | 53.53 | Hispanic or Latino share | NO₂ | 8,301 |
| 61.67 | 55.20 | 8 | 58.61 | Hispanic or Latino share | NO₂ | 8,301 |
| 66.71 | 60.36 | 9 | 63.74 | Hispanic or Latino share | NO₂ | 8,301 |
| 76.39 | 67.95 | 10 | 72.62 | Hispanic or Latino share | NO₂ | 8,300 |
| 40.85 | 34.68 | 1 | 37.81 | Black alone share | NO₂ | 8,813 |
| 42.45 | 36.05 | 2 | 39.17 | Black alone share | NO₂ | 7,786 |
| 48.13 | 41.45 | 3 | 44.78 | Black alone share | NO₂ | 8,300 |
| 52.13 | 45.61 | 4 | 48.97 | Black alone share | NO₂ | 8,299 |
| 56.35 | 50.06 | 5 | 53.35 | Black alone share | NO₂ | 8,301 |
| 57.99 | 51.28 | 6 | 54.73 | Black alone share | NO₂ | 8,300 |
| 57.40 | 51.41 | 7 | 54.50 | Black alone share | NO₂ | 8,301 |
| 57.79 | 51.05 | 8 | 54.50 | Black alone share | NO₂ | 8,301 |
| 57.51 | 49.72 | 9 | 53.74 | Black alone share | NO₂ | 8,301 |
| 62.87 | 53.48 | 10 | 58.35 | Black alone share | NO₂ | 8,301 |
| 42.72 | 37.18 | 1 | 39.90 | Hispanic or Latino share | PM₂.₅ | 8,300 |
| 41.66 | 37.05 | 2 | 39.37 | Hispanic or Latino share | PM₂.₅ | 8,301 |
| 43.40 | 39.08 | 3 | 41.23 | Hispanic or Latino share | PM₂.₅ | 8,301 |
| 46.77 | 41.99 | 4 | 44.43 | Hispanic or Latino share | PM₂.₅ | 8,301 |
| 49.67 | 44.33 | 5 | 47.04 | Hispanic or Latino share | PM₂.₅ | 8,301 |
| 52.85 | 46.43 | 6 | 49.59 | Hispanic or Latino share | PM₂.₅ | 8,300 |
| 56.43 | 48.56 | 7 | 52.56 | Hispanic or Latino share | PM₂.₅ | 8,301 |
| 60.75 | 51.13 | 8 | 56.22 | Hispanic or Latino share | PM₂.₅ | 8,301 |
| 67.09 | 56.02 | 9 | 62.01 | Hispanic or Latino share | PM₂.₅ | 8,301 |
| 76.80 | 60.46 | 10 | 69.44 | Hispanic or Latino share | PM₂.₅ | 8,301 |
| 49.02 | 40.71 | 1 | 44.78 | Black alone share | PM₂.₅ | 8,815 |
| 48.02 | 40.19 | 2 | 44 | Black alone share | PM₂.₅ | 7,786 |
| 52.73 | 43.89 | 3 | 48.27 | Black alone share | PM₂.₅ | 8,301 |
| 55.36 | 46.28 | 4 | 50.74 | Black alone share | PM₂.₅ | 8,301 |
| 57.32 | 48.36 | 5 | 52.76 | Black alone share | PM₂.₅ | 8,301 |
| 58.38 | 49.06 | 6 | 53.58 | Black alone share | PM₂.₅ | 8,300 |
| 57.34 | 49.22 | 7 | 53.36 | Black alone share | PM₂.₅ | 8,301 |
| 57.32 | 49.17 | 8 | 53.42 | Black alone share | PM₂.₅ | 8,301 |
| 55.94 | 48.20 | 9 | 52.14 | Black alone share | PM₂.₅ | 8,301 |
| 58.29 | 48.65 | 10 | 53.65 | Black alone share | PM₂.₅ | 8,301 |

Note: Deciles hold nearly equal numbers of tracts, and tracts with the same share go to the same decile. The lowest Black-share decile therefore holds all 8,815 tracts with no Black alone residents, and the second holds fewer than a tenth of the tracts.

Note: Hispanic or Latino residents can be of any race and Black alone residents can be of any ethnicity, so the two groups overlap. Both shares are ACS 2019-2023 tract estimates.

Note: Weighting each tract by its number of Hispanic or Latino residents instead, the mean NO₂ percentile is 63; weighting by Black alone residents it is 56, and weighting by all residents it is 50. These are averages over tracts of residence and not measurements of individuals.

Note: The analysis covers the contiguous states and the District of Columbia, because EJScreen has no PM₂.₅ or ozone estimate for the 1,524 tracts in Alaska, Hawaii and Puerto Rico.

Source: EPA EJScreen 2024 and ACS 2019-2023; population-weighted means, modeled estimates

Data: https://vishalsingh.org/zipcode/report/air-pollution/figures/air_by_race_deciles.json

### Table 4. Residents of tracts in the highest Black-share decile live with 1.7 stacked air hazards on average, against 0.8 in the lowest, and the score falls across deprivation deciles

Mean number of five air hazards in the top quarter of tracts, by decile of tract Black alone share, Hispanic or Latino share and ADI national rank, population-weighted, up to 82,993 tracts

| conf_high | conf_low | decile | estimate | measure | n_units | sorted_by |
| ---: | ---: | ---: | ---: | --- | ---: | --- |
| 0.990 | 0.640 | 1 | 0.800 | score | 8,809 | Black alone share |
| 15.06 | 6.03 | 1 | 10.17 | heavy | 8,809 | Black alone share |
| 1.04 | 0.650 | 2 | 0.840 | score | 7,786 | Black alone share |
| 17.05 | 6.94 | 2 | 11.71 | heavy | 7,786 | Black alone share |
| 1.27 | 0.830 | 3 | 1.04 | score | 8,300 | Black alone share |
| 22.10 | 10.42 | 3 | 16.08 | heavy | 8,300 | Black alone share |
| 1.43 | 0.980 | 4 | 1.20 | score | 8,299 | Black alone share |
| 25.58 | 13.37 | 4 | 19.37 | heavy | 8,299 | Black alone share |
| 1.58 | 1.12 | 5 | 1.35 | score | 8,301 | Black alone share |
| 28.92 | 16.40 | 5 | 22.71 | heavy | 8,301 | Black alone share |
| 1.68 | 1.17 | 6 | 1.43 | score | 8,300 | Black alone share |
| 31.73 | 18.17 | 6 | 25.08 | heavy | 8,300 | Black alone share |
| 1.57 | 1.16 | 7 | 1.36 | score | 8,301 | Black alone share |
| 28.24 | 17.10 | 7 | 22.71 | heavy | 8,301 | Black alone share |
| 1.60 | 1.16 | 8 | 1.38 | score | 8,299 | Black alone share |
| 28.79 | 17.03 | 8 | 22.94 | heavy | 8,299 | Black alone share |
| 1.59 | 1.16 | 9 | 1.37 | score | 8,300 | Black alone share |
| 30.34 | 17.82 | 9 | 24.06 | heavy | 8,300 | Black alone share |
| 2 | 1.40 | 10 | 1.70 | score | 8,298 | Black alone share |
| 40.20 | 23.19 | 10 | 31.97 | heavy | 8,298 | Black alone share |
| 0.750 | 0.470 | 1 | 0.600 | score | 8,295 | Hispanic or Latino share |
| 9.79 | 3.68 | 1 | 6.44 | heavy | 8,295 | Hispanic or Latino share |
| 0.720 | 0.530 | 2 | 0.620 | score | 8,300 | Hispanic or Latino share |
| 8.77 | 4.74 | 2 | 6.68 | heavy | 8,300 | Hispanic or Latino share |
| 0.850 | 0.650 | 3 | 0.750 | score | 8,299 | Hispanic or Latino share |
| 10.63 | 6 | 3 | 8.29 | heavy | 8,299 | Hispanic or Latino share |
| 1.01 | 0.780 | 4 | 0.890 | score | 8,300 | Hispanic or Latino share |
| 14.68 | 8.62 | 4 | 11.60 | heavy | 8,300 | Hispanic or Latino share |
| 1.21 | 0.930 | 5 | 1.07 | score | 8,300 | Hispanic or Latino share |
| 18.85 | 10.90 | 5 | 14.90 | heavy | 8,300 | Hispanic or Latino share |
| 1.34 | 1 | 6 | 1.17 | score | 8,300 | Hispanic or Latino share |
| 22.30 | 12.90 | 6 | 17.60 | heavy | 8,300 | Hispanic or Latino share |
| 1.53 | 1.14 | 7 | 1.34 | score | 8,301 | Hispanic or Latino share |
| 27.59 | 16.80 | 7 | 22.27 | heavy | 8,301 | Hispanic or Latino share |
| 1.78 | 1.30 | 8 | 1.54 | score | 8,301 | Hispanic or Latino share |
| 34.26 | 20.32 | 8 | 27.64 | heavy | 8,301 | Hispanic or Latino share |
| 2.07 | 1.55 | 9 | 1.82 | score | 8,298 | Hispanic or Latino share |
| 42.31 | 27.17 | 9 | 35.04 | heavy | 8,298 | Hispanic or Latino share |
| 2.79 | 1.84 | 10 | 2.34 | score | 8,299 | Hispanic or Latino share |
| 60.69 | 34.80 | 10 | 48.84 | heavy | 8,299 | Hispanic or Latino share |
| 2.49 | 1.67 | 1 | 2.13 | score | 8,146 | Area Deprivation Index rank |
| 56.35 | 29.54 | 1 | 44.46 | heavy | 8,146 | Area Deprivation Index rank |
| 2.26 | 1.34 | 2 | 1.79 | score | 8,292 | Area Deprivation Index rank |
| 47.03 | 22.05 | 2 | 34.58 | heavy | 8,292 | Area Deprivation Index rank |
| 1.63 | 1.16 | 3 | 1.38 | score | 8,204 | Area Deprivation Index rank |
| 30.43 | 17.64 | 3 | 24.07 | heavy | 8,204 | Area Deprivation Index rank |
| 1.39 | 1.03 | 4 | 1.21 | score | 8,184 | Area Deprivation Index rank |
| 23.21 | 13.37 | 4 | 18.35 | heavy | 8,184 | Area Deprivation Index rank |
| 1.30 | 0.920 | 5 | 1.11 | score | 8,416 | Area Deprivation Index rank |
| 21.04 | 10.84 | 5 | 15.92 | heavy | 8,416 | Area Deprivation Index rank |
| 1.24 | 0.850 | 6 | 1.04 | score | 8,052 | Area Deprivation Index rank |
| 20.28 | 9.73 | 6 | 14.88 | heavy | 8,052 | Area Deprivation Index rank |
| 1.22 | 0.780 | 7 | 0.980 | score | 8,161 | Area Deprivation Index rank |
| 19.91 | 8.44 | 7 | 13.88 | heavy | 8,161 | Area Deprivation Index rank |
| 1.04 | 0.710 | 8 | 0.870 | score | 8,288 | Area Deprivation Index rank |
| 16.11 | 7.58 | 8 | 11.47 | heavy | 8,288 | Area Deprivation Index rank |
| 0.970 | 0.660 | 9 | 0.800 | score | 8,247 | Area Deprivation Index rank |
| 14.70 | 6.77 | 9 | 10.40 | heavy | 8,247 | Area Deprivation Index rank |
| 1.17 | 0.790 | 10 | 0.980 | score | 8,203 | Area Deprivation Index rank |
| 19.56 | 9.71 | 10 | 14.49 | heavy | 8,203 | Area Deprivation Index rank |

Views: Mean number of stacked hazards (measure = score); Share of residents with three or more (measure = heavy).

Note: Deciles hold nearly equal numbers of tracts, tracts with the same value go to the same decile, and the lowest Black-share decile holds all 8,815 tracts with no Black alone residents. Bands are 95% intervals from a bootstrap over counties. The lines are not monotone in every step: the Black-share line is level from the sixth to the ninth decile and the deprivation line rises slightly in the tenth.

Note: Ozone is not counted. It forms in the air from other pollutants and does not mark a local source, as the other five hazards do. A count of six that adds ozone has the same ordering across the three lines.

Note: The analysis covers the contiguous states and the District of Columbia, because EJScreen has no PM₂.₅ or ozone estimate for the 1,524 tracts in Alaska, Hawaii and Puerto Rico.

Note: Hispanic or Latino residents can be of any race and Black alone residents can be of any ethnicity, so the groups overlap.

Source: EPA EJScreen 2024 and ACS 2019-2023; population-weighted means, modeled estimates

Data: https://vishalsingh.org/zipcode/report/air-pollution/figures/stacked_exposure.json

### Table 5. The first component weights diesel PM, traffic, hazardous-waste facilities and NO₂ most, and the second weights storage tanks, lead paint and drinking-water violations against PM₂.₅ and ozone

Weights of the 13 standardized EJScreen percentiles on the first two principal components, 80,754 tracts, each tract counted once

| group | indicator | pc1 | pc1_high | pc1_low | pc2 | pc2_high | pc2_low |
| --- | --- | ---: | ---: | ---: | ---: | ---: | ---: |
| Air | PM2.5 | 0.220 | 0.260 | 0.170 | −0.470 | −0.380 | −0.550 |
| Air | Ozone | 0.250 | 0.280 | 0.220 | −0.310 | −0.160 | −0.380 |
| Air | Diesel PM | 0.400 | 0.410 | 0.380 | −0.020 | 0.050 | −0.080 |
| Air | Nitrogen dioxide | 0.380 | 0.390 | 0.360 | 0.040 | 0.110 | −0.030 |
| Air | Traffic proximity | 0.400 | 0.410 | 0.390 | 0.020 | 0.080 | −0.050 |
| Air | Toxic releases to air | 0.230 | 0.260 | 0.200 | 0.090 | 0.280 | −0.200 |
| Proximity and discharge | Superfund proximity | 0.230 | 0.260 | 0.210 | 0.220 | 0.300 | 0.130 |
| Proximity and discharge | Risk-management facilities | 0.260 | 0.280 | 0.210 | −0.080 | 0.110 | −0.330 |
| Proximity and discharge | Hazardous-waste facilities | 0.390 | 0.400 | 0.370 | 0 | 0.070 | −0.060 |
| Proximity and discharge | Wastewater discharge | 0.220 | 0.250 | 0.190 | −0.040 | 0.060 | −0.150 |
| Old housing and infrastructure | Lead paint (housing built before 1960) | 0.140 | 0.170 | 0.100 | 0.450 | 0.560 | 0.330 |
| Old housing and infrastructure | Underground storage tanks | 0.150 | 0.260 | 0.030 | 0.550 | 0.660 | 0.330 |
| Old housing and infrastructure | Drinking-water violations | −0.080 | −0.020 | −0.120 | 0.320 | 0.430 | 0.190 |

Note: Weights are the entries of unit-length eigenvectors of the correlation matrix, so they are the weights on standardized percentiles. A population-weighted version gives almost the same components: the first holds 37.1% of the variance.

Note: The first component is labeled by its largest weights and the second by its largest positive weights; the labels are descriptions, not findings.

Source: EPA EJScreen 2024, national percentiles of the 13 indicators; principal components of the correlation matrix, each tract once, modeled estimates

Data: https://vishalsingh.org/zipcode/report/air-pollution/figures/pca_loadings.json

### Table 6. The urban-industrial component is highest in Cook County, Illinois, and in dense counties around New York, and below the tract average in most counties

Population-weighted county mean of the tract score on the first principal component of 13 EJScreen percentiles, in standard deviations of the tract score

No inline rows: the values of Table 6 are in https://vishalsingh.org/zipcode/data/air-pollution/pc1_map.json.

Note: A county is mapped when at least 50% of its residents live in tracts with all 13 indicators. Gray counties have no score: Alaska and Hawaii (no PM₂.₅ or ozone estimate), and a few counties where fewer tracts have all of them. Puerto Rico has no PM₂.₅ or ozone estimate and is not drawn.

Note: Connecticut is drawn as its eight legacy counties, with tracts assigned to them by the staged crosswalk. Scores are county means of modeled tract values and carry no sampling interval.

Note: The key runs from the first to the 99th percentile of the county scores, and counties outside that range take the colors of its two ends; the tooltip gives the value.

Source: EPA EJScreen 2024, first principal component of the 13 national percentiles; population-weighted county means, modeled estimates

Data: https://vishalsingh.org/zipcode/report/air-pollution/figures/pc1_map.json

### Table 7. Pollution percentiles correlate negatively with deprivation for 4 of 6 air indicators and positively with Hispanic share for 5

Population-weighted correlation of 13 EJScreen national percentiles with the ADI national rank, Hispanic or Latino share and Black alone share, across tracts

| against | conf_high | conf_low | county_high | county_low | group | indicator | n | r | state_high | state_low |
| --- | ---: | ---: | ---: | ---: | --- | --- | ---: | ---: | ---: | ---: |
| Area Deprivation Index | 0.101 | −0.378 | −0.060 | −0.289 | Air | PM2.5 | 82,198 | −0.176 | 0.101 | −0.378 |
| Hispanic or Latino share | 0.422 | 0.069 | 0.396 | 0.217 | Air | PM2.5 | 83,008 | 0.315 | 0.422 | 0.069 |
| Black alone share | 0.240 | −0.075 | 0.118 | 0 | Air | PM2.5 | 83,008 | 0.060 | 0.240 | −0.075 |
| Area Deprivation Index | −0.044 | −0.361 | −0.132 | −0.352 | Air | Ozone | 82,198 | −0.244 | −0.044 | −0.361 |
| Hispanic or Latino share | 0.445 | 0.172 | 0.418 | 0.236 | Air | Ozone | 83,008 | 0.336 | 0.445 | 0.172 |
| Black alone share | 0.063 | −0.167 | −0.010 | −0.122 | Air | Ozone | 83,008 | −0.066 | 0.063 | −0.167 |
| Area Deprivation Index | −0.254 | −0.455 | −0.301 | −0.452 | Air | Diesel PM | 82,198 | −0.377 | −0.254 | −0.455 |
| Hispanic or Latino share | 0.457 | 0.173 | 0.409 | 0.258 | Air | Diesel PM | 82,998 | 0.340 | 0.457 | 0.173 |
| Black alone share | 0.277 | 0.087 | 0.227 | 0.131 | Air | Diesel PM | 82,998 | 0.179 | 0.277 | 0.087 |
| Area Deprivation Index | −0.047 | −0.286 | −0.103 | −0.266 | Air | Nitrogen dioxide | 82,193 | −0.182 | −0.047 | −0.286 |
| Hispanic or Latino share | 0.455 | 0.340 | 0.455 | 0.345 | Air | Nitrogen dioxide | 83,003 | 0.402 | 0.447 | 0.340 |
| Black alone share | 0.245 | 0.058 | 0.189 | 0.098 | Air | Nitrogen dioxide | 83,003 | 0.145 | 0.245 | 0.058 |
| Area Deprivation Index | −0.247 | −0.516 | −0.331 | −0.492 | Air | Traffic proximity | 82,198 | −0.413 | −0.247 | −0.516 |
| Hispanic or Latino share | 0.419 | 0.219 | 0.401 | 0.267 | Air | Traffic proximity | 82,998 | 0.333 | 0.419 | 0.219 |
| Black alone share | 0.289 | 0.057 | 0.220 | 0.116 | Air | Traffic proximity | 82,998 | 0.167 | 0.289 | 0.057 |
| Area Deprivation Index | 0.147 | −0.063 | 0.130 | −0.001 | Air | Toxic releases to air | 82,198 | 0.064 | 0.147 | −0.063 |
| Hispanic or Latino share | 0.087 | −0.135 | 0.054 | −0.135 | Air | Toxic releases to air | 83,008 | −0.037 | 0.087 | −0.109 |
| Black alone share | 0.221 | 0.101 | 0.202 | 0.115 | Air | Toxic releases to air | 83,008 | 0.162 | 0.221 | 0.101 |
| Area Deprivation Index | −0.102 | −0.302 | −0.141 | −0.273 | Proximity and discharge | Superfund proximity | 82,198 | −0.210 | −0.102 | −0.302 |
| Hispanic or Latino share | 0.213 | 0.048 | 0.200 | 0.062 | Proximity and discharge | Superfund proximity | 82,998 | 0.134 | 0.213 | 0.048 |
| Black alone share | 0.152 | 0.006 | 0.123 | 0.027 | Proximity and discharge | Superfund proximity | 82,998 | 0.077 | 0.152 | 0.006 |
| Area Deprivation Index | 0.219 | 0.001 | 0.177 | 0.016 | Proximity and discharge | Risk-management facilities | 82,198 | 0.100 | 0.219 | 0.001 |
| Hispanic or Latino share | 0.324 | 0.103 | 0.315 | 0.174 | Proximity and discharge | Risk-management facilities | 82,998 | 0.248 | 0.324 | 0.103 |
| Black alone share | 0.200 | 0.090 | 0.185 | 0.092 | Proximity and discharge | Risk-management facilities | 82,998 | 0.142 | 0.200 | 0.090 |
| Area Deprivation Index | −0.154 | −0.497 | −0.266 | −0.450 | Proximity and discharge | Hazardous-waste facilities | 82,198 | −0.363 | −0.154 | −0.497 |
| Hispanic or Latino share | 0.405 | 0.128 | 0.358 | 0.192 | Proximity and discharge | Hazardous-waste facilities | 82,998 | 0.274 | 0.405 | 0.128 |
| Black alone share | 0.259 | 0.015 | 0.185 | 0.079 | Proximity and discharge | Hazardous-waste facilities | 82,998 | 0.131 | 0.259 | 0.015 |
| Area Deprivation Index | 0.035 | −0.172 | −0.014 | −0.170 | Proximity and discharge | Wastewater discharge | 82,198 | −0.092 | 0.035 | −0.172 |
| Hispanic or Latino share | 0.207 | −0.070 | 0.164 | −0.036 | Proximity and discharge | Wastewater discharge | 82,998 | 0.073 | 0.207 | −0.070 |
| Black alone share | 0.126 | 0.007 | 0.112 | 0.011 | Proximity and discharge | Wastewater discharge | 82,998 | 0.061 | 0.126 | 0.007 |
| Area Deprivation Index | 0.342 | 0.040 | 0.258 | 0.060 | Old housing and infrastructure | Lead paint (housing built before 1960) | 82,198 | 0.160 | 0.342 | 0.040 |
| Hispanic or Latino share | 0.131 | −0.087 | 0.107 | −0.058 | Old housing and infrastructure | Lead paint (housing built before 1960) | 83,008 | 0.019 | 0.131 | −0.087 |
| Black alone share | 0.126 | 0.002 | 0.113 | 0.007 | Old housing and infrastructure | Lead paint (housing built before 1960) | 83,008 | 0.062 | 0.126 | 0.002 |
| Area Deprivation Index | 0.461 | −0.039 | 0.365 | 0.126 | Old housing and infrastructure | Underground storage tanks | 82,198 | 0.253 | 0.461 | −0.039 |
| Hispanic or Latino share | 0.320 | −0.294 | 0.104 | −0.177 | Old housing and infrastructure | Underground storage tanks | 83,008 | −0.037 | 0.320 | −0.294 |
| Black alone share | 0.317 | 0.216 | 0.298 | 0.243 | Old housing and infrastructure | Underground storage tanks | 83,008 | 0.271 | 0.317 | 0.216 |
| Area Deprivation Index | 0.209 | −0.109 | 0.142 | −0.024 | Old housing and infrastructure | Drinking-water violations | 80,047 | 0.063 | 0.209 | −0.109 |
| Hispanic or Latino share | 0.166 | −0.142 | 0.077 | −0.098 | Old housing and infrastructure | Drinking-water violations | 80,758 | −0.017 | 0.166 | −0.142 |
| Black alone share | 0.062 | −0.107 | 0.027 | −0.065 | Old housing and infrastructure | Drinking-water violations | 80,758 | −0.020 | 0.062 | −0.107 |

Note: Indicators are tract-level national percentiles; several are modeled estimates. Correlations use all tracts that have both values, so the count varies by cell. The data table gives the county and state intervals next to the wider one that the figure uses.

Note: Hispanic or Latino residents can be of any race and Black alone residents can be of any ethnicity, so the two columns overlap.

Source: EPA EJScreen 2024, Area Deprivation Index 2023 and ACS 2019-2023; population-weighted correlations, modeled estimates

Data: https://vishalsingh.org/zipcode/report/air-pollution/figures/air_correlation_strip.json

### Table 8. Modeled asthma prevalence is highest in the lowest twentieth of tracts on the urban-industrial component (11.2%) and rises with the lead-paint percentile and with deprivation

Modeled prevalence of current asthma among adults by twentieth of tracts on three measures, population-weighted, 80,754 tracts with all 13 percentiles

| conf_high | conf_low | estimate | n_units | ranked_on | twentieth |
| ---: | ---: | ---: | ---: | --- | ---: |
| 11.28 | 11.08 | 11.17 | 4,033 | Urban-industrial component | 1 |
| 10.93 | 10.79 | 10.86 | 4,034 | Urban-industrial component | 2 |
| 10.80 | 10.66 | 10.73 | 4,034 | Urban-industrial component | 3 |
| 10.71 | 10.56 | 10.64 | 4,033 | Urban-industrial component | 4 |
| 10.64 | 10.47 | 10.56 | 4,034 | Urban-industrial component | 5 |
| 10.62 | 10.45 | 10.53 | 4,034 | Urban-industrial component | 6 |
| 10.57 | 10.38 | 10.47 | 4,033 | Urban-industrial component | 7 |
| 10.50 | 10.29 | 10.40 | 4,034 | Urban-industrial component | 8 |
| 10.52 | 10.29 | 10.41 | 4,034 | Urban-industrial component | 9 |
| 10.49 | 10.27 | 10.38 | 4,034 | Urban-industrial component | 10 |
| 10.50 | 10.24 | 10.37 | 4,033 | Urban-industrial component | 11 |
| 10.54 | 10.24 | 10.39 | 4,034 | Urban-industrial component | 12 |
| 10.60 | 10.24 | 10.41 | 4,034 | Urban-industrial component | 13 |
| 10.60 | 10.26 | 10.42 | 4,033 | Urban-industrial component | 14 |
| 10.71 | 10.32 | 10.50 | 4,034 | Urban-industrial component | 15 |
| 10.85 | 10.36 | 10.59 | 4,034 | Urban-industrial component | 16 |
| 11.04 | 10.45 | 10.73 | 4,033 | Urban-industrial component | 17 |
| 11.03 | 10.24 | 10.61 | 4,034 | Urban-industrial component | 18 |
| 11.18 | 10.05 | 10.53 | 4,034 | Urban-industrial component | 19 |
| 10.47 | 9.68 | 9.87 | 4,034 | Urban-industrial component | 20 |
| 9.89 | 9.55 | 9.72 | 4,736 | Lead-paint percentile | 1 |
| 9.88 | 9.61 | 9.74 | 3,292 | Lead-paint percentile | 2 |
| 10.01 | 9.76 | 9.88 | 4,082 | Lead-paint percentile | 3 |
| 10.17 | 9.93 | 10.05 | 4,012 | Lead-paint percentile | 4 |
| 10.33 | 10.11 | 10.22 | 4,027 | Lead-paint percentile | 5 |
| 10.47 | 10.27 | 10.38 | 3,987 | Lead-paint percentile | 6 |
| 10.66 | 10.46 | 10.56 | 3,953 | Lead-paint percentile | 7 |
| 10.75 | 10.55 | 10.65 | 3,976 | Lead-paint percentile | 8 |
| 10.81 | 10.60 | 10.71 | 4,014 | Lead-paint percentile | 9 |
| 10.86 | 10.68 | 10.77 | 4,007 | Lead-paint percentile | 10 |
| 10.93 | 10.71 | 10.82 | 4,043 | Lead-paint percentile | 11 |
| 10.96 | 10.72 | 10.84 | 4,046 | Lead-paint percentile | 12 |
| 10.93 | 10.66 | 10.80 | 4,076 | Lead-paint percentile | 13 |
| 10.95 | 10.67 | 10.81 | 4,108 | Lead-paint percentile | 14 |
| 10.96 | 10.59 | 10.77 | 4,123 | Lead-paint percentile | 15 |
| 11.01 | 10.63 | 10.81 | 4,180 | Lead-paint percentile | 16 |
| 11.05 | 10.58 | 10.80 | 4,203 | Lead-paint percentile | 17 |
| 11.18 | 10.61 | 10.88 | 3,379 | Lead-paint percentile | 18 |
| 11.26 | 10.60 | 10.90 | 4,214 | Lead-paint percentile | 19 |
| 11.47 | 10.59 | 10.99 | 4,216 | Lead-paint percentile | 20 |
| 9.29 | 8.87 | 9.05 | 4,164 | ADI national rank | 1 |
| 9.66 | 9.24 | 9.43 | 3,873 | ADI national rank | 2 |
| 9.88 | 9.57 | 9.72 | 3,982 | ADI national rank | 3 |
| 10.08 | 9.77 | 9.93 | 4,135 | ADI national rank | 4 |
| 10.24 | 9.91 | 10.07 | 3,814 | ADI national rank | 5 |
| 10.40 | 9.99 | 10.18 | 4,039 | ADI national rank | 6 |
| 10.35 | 10.04 | 10.20 | 3,933 | ADI national rank | 7 |
| 10.42 | 10.15 | 10.29 | 4,053 | ADI national rank | 8 |
| 10.49 | 10.25 | 10.37 | 4,132 | ADI national rank | 9 |
| 10.58 | 10.33 | 10.46 | 3,708 | ADI national rank | 10 |
| 10.65 | 10.42 | 10.53 | 4,205 | ADI national rank | 11 |
| 10.78 | 10.55 | 10.67 | 3,968 | ADI national rank | 12 |
| 10.83 | 10.61 | 10.72 | 4,078 | ADI national rank | 13 |
| 10.97 | 10.79 | 10.87 | 3,828 | ADI national rank | 14 |
| 11.12 | 10.92 | 11.02 | 4,143 | ADI national rank | 15 |
| 11.22 | 11.03 | 11.12 | 3,897 | ADI national rank | 16 |
| 11.36 | 11.17 | 11.27 | 4,080 | ADI national rank | 17 |
| 11.58 | 11.37 | 11.47 | 4,042 | ADI national rank | 18 |
| 11.92 | 11.65 | 11.78 | 4,017 | ADI national rank | 19 |
| 12.92 | 12.44 | 12.68 | 3,947 | ADI national rank | 20 |

Note: PLACES estimates for tracts come from a model fitted to survey responses, with age, sex, race and ethnicity, education and county poverty as inputs, applied to each tract's population; they are not counts of cases. The lead-paint indicator and the ADI are built from the same Census survey as the population inputs, which makes their lines partly mechanical.

Note: Each twentieth holds equal numbers of tracts on its own measure, so the three lines do not describe the same tracts. The tract file carries no intervals for PLACES values.

Source: CDC PLACES 2024 release (modeled estimates from BRFSS 2022), EPA EJScreen 2024 and ADI 2023; population-weighted means

Data: https://vishalsingh.org/zipcode/report/air-pollution/figures/asthma_vs_exposure.json

### Table 9. Within the same county, tracts graded D in the 1930s rank 3.9 points above grade A tracts on the traffic percentile and 9.7 points on NO₂, and the stacked-hazard score is 0.6 higher

Difference from grade A in the same county in the mean national percentile of four air indicators and in the mean stacked-hazard score, by dominant HOLC grade of the tract, population-weighted

| conf_high | conf_low | contrast | estimate | grade | label | measure | n_units |
| ---: | ---: | --- | ---: | --- | --- | --- | ---: |
| n/a | n/a | within | 0 | A | Grade A | Traffic proximity (percentile) | 672 |
| 1.78 | −0.180 | within | 0.810 | B | Grade B | Traffic proximity (percentile) | 2,246 |
| 3.15 | 0.860 | within | 2.06 | C | Grade C | Traffic proximity (percentile) | 5,054 |
| 5.12 | 2.67 | within | 3.91 | D | Grade D | Traffic proximity (percentile) | 2,978 |
| 80.28 | 74.52 | pooled | 77.55 | A | Grade A | Traffic proximity (percentile) | 672 |
| 84.21 | 78.81 | pooled | 81.74 | B | Grade B | Traffic proximity (percentile) | 2,246 |
| 87.02 | 80.11 | pooled | 84.16 | C | Grade C | Traffic proximity (percentile) | 5,054 |
| 88.62 | 82.79 | pooled | 86.34 | D | Grade D | Traffic proximity (percentile) | 2,978 |
| n/a | n/a | within | 0 | A | Grade A | NO₂ (percentile) | 672 |
| 6.41 | 2.47 | within | 4.46 | B | Grade B | NO₂ (percentile) | 2,246 |
| 8.76 | 4.48 | within | 6.67 | C | Grade C | NO₂ (percentile) | 5,054 |
| 12.20 | 7.23 | within | 9.74 | D | Grade D | NO₂ (percentile) | 2,978 |
| 76.36 | 69.73 | pooled | 73.03 | A | Grade A | NO₂ (percentile) | 672 |
| 81.99 | 74.96 | pooled | 78.72 | B | Grade B | NO₂ (percentile) | 2,246 |
| 85.72 | 79.03 | pooled | 82.73 | C | Grade C | NO₂ (percentile) | 5,054 |
| 89.21 | 82.52 | pooled | 86.36 | D | Grade D | NO₂ (percentile) | 2,978 |
| n/a | n/a | within | 0 | A | Grade A | Diesel PM (percentile) | 672 |
| 2.19 | 0.240 | within | 1.24 | B | Grade B | Diesel PM (percentile) | 2,246 |
| 3.57 | 1.71 | within | 2.70 | C | Grade C | Diesel PM (percentile) | 5,054 |
| 6.45 | 3.92 | within | 5.14 | D | Grade D | Diesel PM (percentile) | 2,978 |
| 76.57 | 68.93 | pooled | 73.05 | A | Grade A | Diesel PM (percentile) | 672 |
| 81.66 | 72.33 | pooled | 77.41 | B | Grade B | Diesel PM (percentile) | 2,246 |
| 84.97 | 75.26 | pooled | 81.17 | C | Grade C | Diesel PM (percentile) | 5,054 |
| 89.44 | 80.50 | pooled | 86.02 | D | Grade D | Diesel PM (percentile) | 2,978 |
| n/a | n/a | within | 0 | A | Grade A | PM₂.₅ (percentile) | 672 |
| 0.880 | −0.010 | within | 0.420 | B | Grade B | PM₂.₅ (percentile) | 2,246 |
| 1.28 | 0.030 | within | 0.670 | C | Grade C | PM₂.₅ (percentile) | 5,054 |
| 2.02 | 0.250 | within | 1.13 | D | Grade D | PM₂.₅ (percentile) | 2,978 |
| 62.68 | 49.05 | pooled | 55.85 | A | Grade A | PM₂.₅ (percentile) | 672 |
| 61.24 | 47.08 | pooled | 54.02 | B | Grade B | PM₂.₅ (percentile) | 2,246 |
| 68.58 | 46.88 | pooled | 58.11 | C | Grade C | PM₂.₅ (percentile) | 5,054 |
| 69.39 | 51.02 | pooled | 60.02 | D | Grade D | PM₂.₅ (percentile) | 2,978 |
| n/a | n/a | within | 0 | A | Grade A | Stacked air hazards (0 to 5) | 672 |
| 0.340 | 0.050 | within | 0.200 | B | Grade B | Stacked air hazards (0 to 5) | 2,246 |
| 0.500 | 0.240 | within | 0.380 | C | Grade C | Stacked air hazards (0 to 5) | 5,054 |
| 0.720 | 0.420 | within | 0.580 | D | Grade D | Stacked air hazards (0 to 5) | 2,978 |
| 2.52 | 2.04 | pooled | 2.29 | A | Grade A | Stacked air hazards (0 to 5) | 672 |
| 2.87 | 2.29 | pooled | 2.60 | B | Grade B | Stacked air hazards (0 to 5) | 2,246 |
| 3.35 | 2.57 | pooled | 3 | C | Grade C | Stacked air hazards (0 to 5) | 5,054 |
| 3.50 | 2.85 | pooled | 3.19 | D | Grade D | Stacked air hazards (0 to 5) | 2,978 |

Views: Within the same county (contrast = within); All graded tracts, pooled (contrast = pooled).

Note: Differences in the first view come from a population-weighted regression of the tract value on grade with county fixed effects, so only counties that hold more than one grade inform them. Grade A is the baseline and has no interval. The pooled view compares tracts across 215 counties.

Note: The HOLC maps cover the dense urban cores of cities, so even grade A tracts sit above the national median for traffic and NO₂. The comparison is among graded tracts and does not compare graded with ungraded places.

Note: A tract graded D today holds residents of today and not those of the 1930s. Grades A to D follow the original map labels, and areas graded E or F or left blank are set aside. Other outcomes by grade, including the Area Deprivation Index, are in the chapter on home value, income and redlining.

Source: Mapping Inequality HOLC maps (1935-1940) matched to 2020 tracts, EPA EJScreen 2024 and ACS 2019-2023; population-weighted means, modeled estimates

Data: https://vishalsingh.org/zipcode/report/air-pollution/figures/air_by_holc_grade.json

### Table 10. Traffic proximity is steadily better toward the more deprived end under the ADI; the two ends do not differ detectably under the standardized index, and it is better at the more deprived end but uneven along the way under the income ranking

Mean national percentile of the tract (0 to 100) by tenth of tracts ranked four ways, over the 75,058 tracts that have all four rankings

| bin | conf_high | conf_low | estimate | n_units | ranking |
| ---: | ---: | ---: | ---: | ---: | --- |
| 1 | 79.56 | 68.90 | 75.19 | 7,443 | Area Deprivation Index |
| 2 | 71.26 | 59.32 | 65.46 | 7,502 | Area Deprivation Index |
| 3 | 59.60 | 53.30 | 56.51 | 7,581 | Area Deprivation Index |
| 4 | 53.59 | 48.52 | 51.12 | 7,540 | Area Deprivation Index |
| 5 | 50.45 | 44.98 | 47.82 | 7,439 | Area Deprivation Index |
| 6 | 46.62 | 41.19 | 43.85 | 7,666 | Area Deprivation Index |
| 7 | 43.38 | 37.87 | 40.60 | 7,364 | Area Deprivation Index |
| 8 | 39.90 | 34.91 | 37.48 | 7,543 | Area Deprivation Index |
| 9 | 36.96 | 31.87 | 34.46 | 7,451 | Area Deprivation Index |
| 10 | 41.45 | 35.04 | 38.37 | 7,529 | Area Deprivation Index |
| 1 | 65.41 | 57.70 | 61.82 | 7,505 | Standardized index |
| 2 | 57.72 | 51.44 | 54.67 | 7,506 | Standardized index |
| 3 | 52.94 | 47.38 | 50.22 | 7,506 | Standardized index |
| 4 | 50.03 | 44.40 | 47.29 | 7,506 | Standardized index |
| 5 | 47.16 | 41.13 | 44.12 | 7,506 | Standardized index |
| 6 | 46.44 | 40.16 | 43.32 | 7,505 | Standardized index |
| 7 | 46.71 | 39.55 | 43.22 | 7,506 | Standardized index |
| 8 | 48.60 | 40.85 | 44.75 | 7,506 | Standardized index |
| 9 | 53.91 | 45.99 | 50.05 | 7,506 | Standardized index |
| 10 | 63.52 | 54.66 | 59.23 | 7,506 | Standardized index |
| 1 | 65.51 | 57.85 | 61.94 | 7,507 | Median household income |
| 2 | 58.96 | 51.85 | 55.48 | 7,506 | Median household income |
| 3 | 56.86 | 49.16 | 53.02 | 7,504 | Median household income |
| 4 | 53.49 | 45.72 | 49.64 | 7,525 | Median household income |
| 5 | 51.26 | 43.99 | 47.69 | 7,487 | Median household income |
| 6 | 48.65 | 41.47 | 45.08 | 7,505 | Median household income |
| 7 | 46.60 | 40.03 | 43.36 | 7,506 | Median household income |
| 8 | 47.91 | 41.10 | 44.47 | 7,505 | Median household income |
| 9 | 48.18 | 41.77 | 45.03 | 7,507 | Median household income |
| 10 | 52.84 | 46.36 | 49.77 | 7,506 | Median household income |
| 1 | 71.73 | 64.88 | 68.59 | 7,505 | Bachelor's degree share |
| 2 | 61.91 | 55.58 | 58.83 | 7,505 | Bachelor's degree share |
| 3 | 57.49 | 51.88 | 54.71 | 7,509 | Bachelor's degree share |
| 4 | 54.17 | 48.72 | 51.49 | 7,500 | Bachelor's degree share |
| 5 | 51.50 | 45.69 | 48.62 | 7,507 | Bachelor's degree share |
| 6 | 48.45 | 42.33 | 45.49 | 7,507 | Bachelor's degree share |
| 7 | 44.50 | 38.05 | 41.30 | 7,510 | Bachelor's degree share |
| 8 | 43.52 | 36.17 | 39.78 | 7,512 | Bachelor's degree share |
| 9 | 44.05 | 35.82 | 39.86 | 7,494 | Bachelor's degree share |
| 10 | 52.94 | 42.31 | 47.64 | 7,509 | Bachelor's degree share |

Note: Least and most deprived tenth: ADI 75 and 38, rank correlation −0.96; standardized index 62 and 59, −0.31; median household income 62 and 50, −0.70; bachelor's degree share 69 and 48, −0.87.

Note: The standardized index applies the ADI's coefficients to its measures after standardizing each one, so home value and income carry less weight than in the ADI. Income and degree share are single census measures, each ranked from the highest value to the lowest.

Source: EPA EJScreen 2024, Area Deprivation Index 2023 and ACS 2019-2023; population-weighted means, modeled estimates; ADI 2023 and ACS 2019-2023 tract tables (B19013, B15003 and the 15 rebuilt ADI measures)

Data: https://vishalsingh.org/zipcode/report/air-pollution/figures/alt_rankings.json

## All estimates

Every fact in facts.json (165), sorted by id.

| Fact id | Value | Unit | Estimand | Population | Period | Interval | n | Weights | Source | Note |
| --- | ---: | --- | --- | --- | --- | --- | ---: | --- | --- | --- |
| adi_asthma | +0.53 | correlation coefficient | population-weighted correlation of the tract's modeled prevalence of current asthma among adults with the ADI national rank, county-cluster bootstrap interval | residents of tracts in the PCA with a PLACES estimate for current asthma | PLACES 2024 release (BRFSS 2022); EJScreen 2024; ACS 2019-2023 | 95% CI 0.48 to 0.58 | 80,038 tracts | ACS 2019-2023 tract population | CDC PLACES | PLACES prevalence is a model-based estimate from survey responses, so it is not a measured rate. |
| alt_adi_least | 75 | Mean national percentile of the tract (0 to 100) | population-weighted mean of the tract's traffic-proximity national percentile; 95% intervals from a bootstrap over counties | the least deprived tenth of tracts ranked by Area Deprivation Index, among the tracts with all four rankings | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 69 to 80 | 7,443 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| alt_adi_most | 38 | Mean national percentile of the tract (0 to 100) | population-weighted mean of the tract's traffic-proximity national percentile; 95% intervals from a bootstrap over counties | the most deprived tenth of tracts ranked by Area Deprivation Index, among the tracts with all four rankings | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 35 to 41 | 7,529 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| alt_adi_rho | −0.96 | rank correlation | Spearman rank correlation between bin order and bin mean, each bin counted once | the bins of tracts ranked by Area Deprivation Index | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 |   |   | each bin counted once | EPA EJScreen |   |
| alt_degree_least | 69 | Mean national percentile of the tract (0 to 100) | population-weighted mean of the tract's traffic-proximity national percentile; 95% intervals from a bootstrap over counties | the least deprived tenth of tracts ranked by Bachelor's degree share, among the tracts with all four rankings | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 65 to 72 | 7,505 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| alt_degree_most | 48 | Mean national percentile of the tract (0 to 100) | population-weighted mean of the tract's traffic-proximity national percentile; 95% intervals from a bootstrap over counties | the most deprived tenth of tracts ranked by Bachelor's degree share, among the tracts with all four rankings | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 42 to 53 | 7,509 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| alt_degree_rho | −0.87 | rank correlation | Spearman rank correlation between bin order and bin mean, each bin counted once | the bins of tracts ranked by Bachelor's degree share | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 |   |   | each bin counted once | EPA EJScreen |   |
| alt_income_least | 62 | Mean national percentile of the tract (0 to 100) | population-weighted mean of the tract's traffic-proximity national percentile; 95% intervals from a bootstrap over counties | the least deprived tenth of tracts ranked by Median household income, among the tracts with all four rankings | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 58 to 66 | 7,507 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| alt_income_most | 50 | Mean national percentile of the tract (0 to 100) | population-weighted mean of the tract's traffic-proximity national percentile; 95% intervals from a bootstrap over counties | the most deprived tenth of tracts ranked by Median household income, among the tracts with all four rankings | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 46 to 53 | 7,506 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| alt_income_rho | −0.70 | rank correlation | Spearman rank correlation between bin order and bin mean, each bin counted once | the bins of tracts ranked by Median household income | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 |   |   | each bin counted once | EPA EJScreen |   |
| alt_n_places | 75,058 | tracts | count of tracts in the common sample of the ranking comparison | tracts with the outcome and all four rankings | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 |   |   | none (full count) | EPA EJScreen |   |
| alt_rho_cut | 0.9 | absolute rank correlation | threshold between a steady and an uneven gradient | the rule for changed gradients (docs/RANKING_SWITCH_RULE.md) | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 |   |   | none (a fixed rule) | EPA EJScreen |   |
| alt_rho_margin | 0.1 | absolute rank correlation | smallest move in the rank correlation that counts as a change of shape | the rule for changed gradients (docs/RANKING_SWITCH_RULE.md) | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 |   |   | none (a fixed rule) | EPA EJScreen |   |
| alt_std_least | 62 | Mean national percentile of the tract (0 to 100) | population-weighted mean of the tract's traffic-proximity national percentile; 95% intervals from a bootstrap over counties | the least deprived tenth of tracts ranked by Standardized index, among the tracts with all four rankings | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 58 to 65 | 7,505 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| alt_std_most | 59 | Mean national percentile of the tract (0 to 100) | population-weighted mean of the tract's traffic-proximity national percentile; 95% intervals from a bootstrap over counties | the most deprived tenth of tracts ranked by Standardized index, among the tracts with all four rankings | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 55 to 64 | 7,506 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| alt_std_rho | −0.31 | rank correlation | Spearman rank correlation between bin order and bin mean, each bin counted once | the bins of tracts ranked by Standardized index | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 |   |   | each bin counted once | EPA EJScreen |   |
| asth_adi_t1 | 9.0% | percent of adults with current asthma (modeled) | population-weighted mean of the tract's modeled crude prevalence of current asthma among adults, within a twentieth of tracts, county-cluster bootstrap interval | residents of tracts in the least deprived twentieth of tracts on the ADI national rank | PLACES 2024 release (BRFSS 2022) | 95% CI 8.9% to 9.3% | 4,164 tracts | ACS 2019-2023 tract population | CDC PLACES |   |
| asth_adi_t20 | 12.7% | percent of adults with current asthma (modeled) | population-weighted mean of the tract's modeled crude prevalence of current asthma among adults, within a twentieth of tracts, county-cluster bootstrap interval | residents of tracts in the most deprived twentieth of tracts on the ADI national rank | PLACES 2024 release (BRFSS 2022) | 95% CI 12.4% to 12.9% | 3,947 tracts | ACS 2019-2023 tract population | CDC PLACES |   |
| asth_ldp_t1 | 9.7% | percent of adults with current asthma (modeled) | population-weighted mean of the tract's modeled crude prevalence of current asthma among adults, within a twentieth of tracts, county-cluster bootstrap interval | residents of tracts in the lowest twentieth of tracts on the lead-paint percentile | PLACES 2024 release (BRFSS 2022) | 95% CI 9.5% to 9.9% | 4,736 tracts | ACS 2019-2023 tract population | CDC PLACES |   |
| asth_ldp_t20 | 11.0% | percent of adults with current asthma (modeled) | population-weighted mean of the tract's modeled crude prevalence of current asthma among adults, within a twentieth of tracts, county-cluster bootstrap interval | residents of tracts in the highest twentieth of tracts on the lead-paint percentile | PLACES 2024 release (BRFSS 2022) | 95% CI 10.6% to 11.5% | 4,216 tracts | ACS 2019-2023 tract population | CDC PLACES |   |
| asth_pc1_t1 | 11.2% | percent of adults with current asthma (modeled) | population-weighted mean of the tract's modeled crude prevalence of current asthma among adults, within a twentieth of tracts, county-cluster bootstrap interval | residents of tracts in the lowest twentieth of tracts on the urban-industrial component | PLACES 2024 release (BRFSS 2022) | 95% CI 11.1% to 11.3% | 4,033 tracts | ACS 2019-2023 tract population | CDC PLACES |   |
| asth_pc1_t20 | 9.9% | percent of adults with current asthma (modeled) | population-weighted mean of the tract's modeled crude prevalence of current asthma among adults, within a twentieth of tracts, county-cluster bootstrap interval | residents of tracts in the highest twentieth of tracts on the urban-industrial component | PLACES 2024 release (BRFSS 2022) | 95% CI 9.7% to 10.5% | 4,034 tracts | ACS 2019-2023 tract population | CDC PLACES |   |
| black_share_all | 12.1% | percent of residents | population-weighted mean of the tract share of residents who are Black or African American alone (of any ethnicity) | residents of the tracts in the analysis | ACS 2019-2023 |   | 83,008 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| corridor_bronx | +1.45 | standard deviations of the tract-level component 1 score | population-weighted county mean of the tract component 1 score (scaled to unit variance across tracts) | residents of Bronx County, New York | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 |   | 346 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| corridor_hudson | +1.67 | standard deviations of the tract-level component 1 score | population-weighted county mean of the tract component 1 score (scaled to unit variance across tracts) | residents of Hudson County, New Jersey | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 |   | 182 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| corridor_la_rank | 6 | rank among mapped counties (1 = highest mean score) | rank of Los Angeles County by population-weighted county mean of the tract component 1 score | mapped counties | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 |   |   | none (rank) | EPA EJScreen |   |
| corridor_newyork | +1.58 | standard deviations of the tract-level component 1 score | population-weighted county mean of the tract component 1 score (scaled to unit variance across tracts) | residents of New York County, New York | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 |   | 300 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| corridor_q50 | −1.2 | standard deviations of the tract-level component 1 score | median over mapped counties of the population-weighted county mean of the tract component 1 score (each county counts once) | mapped counties | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 |   | 3,094 counties | none (each county counts once) | EPA EJScreen |   |
| corridor_stlouis | +1.50 | standard deviations of the tract-level component 1 score | population-weighted county mean of the tract component 1 score (scaled to unit variance across tracts) | residents of the independent city of St. Louis, Missouri | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 |   | 104 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| corridor_top | +1.69 | standard deviations of the tract-level component 1 score | population-weighted county mean of the tract component 1 score (scaled to unit variance across tracts) | residents of the county with the highest mean score | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 |   | 1,318 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| dslpm_adi | −0.38 | correlation coefficient | population-weighted correlation of the tract Diesel PM national percentile with the ADI national rank, county-cluster bootstrap interval | residents of tracts with a Diesel PM percentile and an ADI rank | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI −0.45 to −0.30 | 82,198 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| dslpm_adi_metro | −0.28 | correlation coefficient | population-weighted correlation of the tract Diesel PM national percentile with the ADI national rank, metropolitan tracts only, county-cluster bootstrap interval | residents of metropolitan tracts with a Diesel PM percentile and the ADI national rank | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI −0.36 to −0.19 | 66,593 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| dslpm_adi_within | +0.19 | correlation coefficient | population-weighted correlation of the tract Diesel PM national percentile with the ADI national rank, after removing each county's population-weighted mean from both (counties with at least 5 tracts), county-cluster bootstrap interval | residents of tracts in counties with at least 5 tracts that have a Diesel PM percentile and the ADI national rank | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 0.15 to 0.23 | 79,730 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| dslpm_black | +0.18 | correlation coefficient | population-weighted correlation of the tract Diesel PM national percentile with the tract's Black alone share, county-cluster bootstrap interval | residents of tracts with a Diesel PM percentile and a race and ethnicity share | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 0.13 to 0.23 | 82,998 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| dslpm_hisp | +0.34 | correlation coefficient | population-weighted correlation of the tract Diesel PM national percentile with the tract's Hispanic or Latino share, county-cluster bootstrap interval | residents of tracts with a Diesel PM percentile and a race and ethnicity share | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 0.26 to 0.41 | 82,998 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| dslpm_metro_d1 | 71 | national percentile of the tract (0 to 100) | population-weighted mean of the tract Diesel PM national percentile, metropolitan tracts only, ADI deciles of all tracts, county-cluster bootstrap interval | residents of metropolitan tracts (RUCA codes 1 to 3) in ADI decile 1 of all tracts | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 64 to 76 | 7,879 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| dslpm_metro_d9 | 47 | national percentile of the tract (0 to 100) | population-weighted mean of the tract Diesel PM national percentile, metropolitan tracts only, ADI deciles of all tracts, county-cluster bootstrap interval | residents of metropolitan tracts (RUCA codes 1 to 3) in ADI decile 9 of all tracts | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 44 to 50 | 4,817 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| heavy_black_hi | 32% | percent of residents living in tracts with three or more stacked hazards | population-weighted share of residents in tracts with three or more of the five air hazards in the top quarter, county-cluster bootstrap interval | residents of tracts in decile 10 of the Black alone share | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 23% to 40% | 8,298 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| heavy_black_lo | 10% | percent of residents living in tracts with three or more stacked hazards | population-weighted share of residents in tracts with three or more of the five air hazards in the top quarter, county-cluster bootstrap interval | residents of tracts in decile 1 of the Black alone share | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 6% to 15% | 8,809 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| heavy_hisp_hi | 49% | percent of residents living in tracts with three or more stacked hazards | population-weighted share of residents in tracts with three or more of the five air hazards in the top quarter, county-cluster bootstrap interval | residents of tracts in decile 10 of the Hispanic or Latino share | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 35% to 61% | 8,299 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| heavy_hisp_lo | 6% | percent of residents living in tracts with three or more stacked hazards | population-weighted share of residents in tracts with three or more of the five air hazards in the top quarter, county-cluster bootstrap interval | residents of tracts in decile 1 of the Hispanic or Latino share | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 4% to 10% | 8,295 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| heavy_min | 3 | air hazards | a tract is heavily stacked when this many of the five hazards or more are in the top quarter; a definition | tracts | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 |   |   | none (definition) | EPA EJScreen |   |
| hisp_share_all | 19.0% | percent of residents | population-weighted mean of the tract share of residents who are Hispanic or Latino (of any race) | residents of the tracts in the analysis | ACS 2019-2023 |   | 83,008 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| holc_cover_min | 50% | percent of the tract's area | a tract is graded when polygons graded A to D cover at least this share of its area; a definition | 2020 tracts | HOLC maps of 1935 to 1940 |   |   | none (definition) | Mapping Inequality (HOLC maps) |   |
| ldpnt_adi | +0.16 | correlation coefficient | population-weighted correlation of the tract Lead paint (housing built before 1960) national percentile with the ADI national rank, county-cluster bootstrap interval | residents of tracts with a Lead paint (housing built before 1960) percentile and an ADI rank | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 0.06 to 0.26 | 82,198 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| ldpnt_asthma | +0.27 | correlation coefficient | population-weighted correlation of the tract's modeled prevalence of current asthma among adults with the lead-paint percentile, county-cluster bootstrap interval | residents of tracts in the PCA with a PLACES estimate for current asthma | PLACES 2024 release (BRFSS 2022); EJScreen 2024; ACS 2019-2023 | 95% CI 0.20 to 0.34 | 80,674 tracts | ACS 2019-2023 tract population | CDC PLACES | PLACES prevalence is a model-based estimate from survey responses, so it is not a measured rate. |
| map_cover_min | 50% | percent of the county's residents | a county is mapped when residents of tracts with all 13 indicators are at least this share of its residents; a definition | counties | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 |   |   | none (definition) | EPA EJScreen |   |
| n_adi_missing | 810 | tracts | count of tracts with a missing ADI national rank | tracts in the analysis with no ADI rank (every block group suppressed) | ADI 2023 |   |   | none (full count) | Area Deprivation Index |   |
| n_air_adi_neg | 4 | indicators | count of air indicators whose correlation with the ADI national rank is negative with an interval that excludes zero under the county bootstrap and under the bootstrap that redraws states | EJScreen indicators in the group: the 6 air indicators | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 |   |   | none (count of indicators) | EPA EJScreen |   |
| n_air_black_pos | 4 | indicators | count of air indicators whose correlation with the Black alone share is positive with an interval that excludes zero under the county bootstrap and under the bootstrap that redraws states | EJScreen indicators in the group: the 6 air indicators | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 |   |   | none (count of indicators) | EPA EJScreen |   |
| n_air_hisp_pos | 5 | indicators | count of air indicators whose correlation with the Hispanic or Latino share is positive with an interval that excludes zero under the county bootstrap and under the bootstrap that redraws states | EJScreen indicators in the group: the 6 air indicators | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 |   |   | none (count of indicators) | EPA EJScreen |   |
| n_air_ind | 6 | indicators | number of air indicators; a definition | EJScreen air indicators (PM₂.₅, ozone, diesel PM, NO₂, traffic proximity, toxic releases to air) | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 |   |   | none (definition) | EPA EJScreen |   |
| n_between_counties | 3,095 | counties | count of counties in the county-level correlation | counties with at least one tract that has a traffic-proximity percentile and an ADI rank | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 |   |   | none (full count) | EPA EJScreen |   |
| n_black_zero | 8,815 | tracts | count of tracts with a Black alone share of zero | tracts in the analysis with no Black alone residents in the ACS 2019-2023 estimate | ACS 2019-2023 |   |   | none (full count) | American Community Survey |   |
| n_indicators | 13 | indicators | number of national-percentile indicators entering the correlation strip and the PCA | EJScreen environmental indicators in the staged extract | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 |   |   | none (definition) | EPA EJScreen |   |
| n_map_counties | 3,094 | counties | count of mapped counties | counties and county equivalents on the map sprite with a component 1 score | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 |   |   | none (full count) | EPA EJScreen |   |
| n_map_nodata | 49 | counties | count of map units with no component 1 score | counties and county equivalents on the map sprite without a score | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 |   |   | none (full count) | EPA EJScreen |   |
| n_no_pm25 | 1,524 | tracts | count of tracts with a missing PM₂.₅ percentile | tracts with a positive population and no PM₂.₅ or ozone value (all in Alaska, Hawaii and Puerto Rico), left out of the analysis | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 |   |   | none (full count) | EPA EJScreen | Missing is not zero: EPA's fused air-quality surfaces cover the contiguous states and the District of Columbia. |
| n_old_adi_pos | 1 | indicators | count of old-housing and infrastructure indicators whose correlation with the ADI national rank is positive with an interval that excludes zero under the county bootstrap and under the bootstrap that redraws states | EJScreen indicators in the group: the 3 old-housing and infrastructure indicators | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 |   |   | none (count of indicators) | EPA EJScreen |   |
| n_old_ind | 3 | indicators | number of old-housing and infrastructure indicators; a definition | EJScreen old-housing and infrastructure indicators (lead paint, storage tanks, drinking water) | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 |   |   | none (definition) | EPA EJScreen |   |
| n_pca | 80,754 | tracts | count of tracts in the PCA | tracts in the analysis with all 13 percentiles (drinking-water violations are missing for some tracts) | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 |   |   | none (each tract counts once) | EPA EJScreen |   |
| n_prox_ind | 4 | indicators | number of proximity and discharge indicators; a definition | EJScreen proximity and discharge indicators (Superfund sites, risk-management facilities, hazardous-waste facilities, wastewater discharge) | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 |   |   | none (definition) | EPA EJScreen |   |
| n_ptraf_adi | 82,198 | tracts | count of tracts in the correlation of the traffic percentile with the ADI rank | tracts with a traffic-proximity percentile and an ADI rank (tracts without an ADI rank are left out) | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 |   |   | none (full count) | EPA EJScreen |   |
| n_ptraf_adi_within | 79,730 | tracts | count of tracts in the within-county correlation of the traffic percentile with the ADI rank | tracts with a traffic-proximity percentile and an ADI rank in counties with at least 5 such tracts | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 |   |   | none (full count) | EPA EJScreen |   |
| n_stack_tracts | 82,993 | tracts | count of tracts with PM₂.₅, diesel PM, NO₂, traffic and toxic-release percentiles | tracts in the analysis with all five air percentiles | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 |   |   | none (full count) | EPA EJScreen |   |
| n_tracts | 83,008 | tracts | count of tracts in the analysis | 2020 census tracts in the contiguous states and the District of Columbia with an EJScreen record and a positive 2019-2023 population | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 |   |   | none (full count) | EPA EJScreen | Statistics about one indicator use the tracts that have a value for it; each correlation lists its own count. |
| no2_adi | −0.18 | correlation coefficient | population-weighted correlation of the tract NO₂ national percentile with the ADI national rank, county-cluster bootstrap interval | residents of tracts with a NO₂ percentile and an ADI rank | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI −0.27 to −0.10 | 82,193 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| no2_adi_metro | −0.09 | correlation coefficient | population-weighted correlation of the tract NO₂ national percentile with the ADI national rank, metropolitan tracts only, county-cluster bootstrap interval | residents of metropolitan tracts with a NO₂ percentile and the ADI national rank | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI −0.17 to 0.00 | 66,588 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| no2_adi_within | +0.28 | correlation coefficient | population-weighted correlation of the tract NO₂ national percentile with the ADI national rank, after removing each county's population-weighted mean from both (counties with at least 5 tracts), county-cluster bootstrap interval | residents of tracts in counties with at least 5 tracts that have a NO₂ percentile and the ADI national rank | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 0.25 to 0.31 | 79,725 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| no2_black | +0.14 | correlation coefficient | population-weighted correlation of the tract NO₂ national percentile with the tract's Black alone share, county-cluster bootstrap interval | residents of tracts with a NO₂ percentile and a race and ethnicity share | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 0.10 to 0.19 | 83,003 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| no2_black_d1 | 38 | national percentile of the tract (0 to 100) | population-weighted mean of the tract NO₂ national percentile within a decile of tracts by Black alone share, county-cluster bootstrap interval | residents of tracts in decile 1 of the Black alone share of residents (deciles hold equal numbers of tracts) | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 35 to 41 | 8,813 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| no2_black_d10 | 58 | national percentile of the tract (0 to 100) | population-weighted mean of the tract NO₂ national percentile within a decile of tracts by Black alone share, county-cluster bootstrap interval | residents of tracts in decile 10 of the Black alone share of residents (deciles hold equal numbers of tracts) | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 53 to 63 | 8,301 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| no2_black_within | +0.17 | correlation coefficient | population-weighted correlation of the tract NO₂ national percentile with the tract's Black alone share after removing each county's population-weighted mean from both, county-cluster bootstrap interval | residents of tracts in counties with at least 5 tracts that have a NO₂ percentile and a race and ethnicity share | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 0.14 to 0.21 | 80,548 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| no2_hisp | +0.40 | correlation coefficient | population-weighted correlation of the tract NO₂ national percentile with the tract's Hispanic or Latino share, county-cluster bootstrap interval | residents of tracts with a NO₂ percentile and a race and ethnicity share | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 0.35 to 0.46 | 83,003 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| no2_hisp_d1 | 33 | national percentile of the tract (0 to 100) | population-weighted mean of the tract NO₂ national percentile within a decile of tracts by Hispanic or Latino share, county-cluster bootstrap interval | residents of tracts in decile 1 of the Hispanic or Latino share of residents (deciles hold equal numbers of tracts) | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 30 to 36 | 8,300 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| no2_hisp_d10 | 73 | national percentile of the tract (0 to 100) | population-weighted mean of the tract NO₂ national percentile within a decile of tracts by Hispanic or Latino share, county-cluster bootstrap interval | residents of tracts in decile 10 of the Hispanic or Latino share of residents (deciles hold equal numbers of tracts) | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 68 to 76 | 8,300 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| no2_hisp_within | +0.23 | correlation coefficient | population-weighted correlation of the tract NO₂ national percentile with the tract's Hispanic or Latino share after removing each county's population-weighted mean from both, county-cluster bootstrap interval | residents of tracts in counties with at least 5 tracts that have a NO₂ percentile and a race and ethnicity share | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 0.21 to 0.26 | 80,548 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| no2_joint_adi | +0.16 | within-county standard deviations of the NO₂ percentile per within-county standard deviation of the predictor | coefficient on the ADI national rank in a population-weighted regression of the tract NO₂ national percentile on the ADI national rank, the Hispanic or Latino share, the Black alone share, with each county's population-weighted mean removed from every variable and each variable scaled to a standard deviation of one, errors clustered on county | residents of tracts in counties with at least 5 tracts that have every variable in the model | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 0.13 to 0.19 | 79,725 tracts | ACS 2019-2023 tract population | EPA EJScreen | A descriptive association within counties, not a causal estimate. |
| no2_joint_black | +0.12 | within-county standard deviations of the NO₂ percentile per within-county standard deviation of the predictor | coefficient on the Black alone share in a population-weighted regression of the tract NO₂ national percentile on the ADI national rank, the Hispanic or Latino share, the Black alone share, with each county's population-weighted mean removed from every variable and each variable scaled to a standard deviation of one, errors clustered on county | residents of tracts in counties with at least 5 tracts that have every variable in the model | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 0.08 to 0.16 | 79,725 tracts | ACS 2019-2023 tract population | EPA EJScreen | A descriptive association within counties, not a causal estimate. |
| no2_joint_hisp | +0.18 | within-county standard deviations of the NO₂ percentile per within-county standard deviation of the predictor | coefficient on the Hispanic or Latino share in a population-weighted regression of the tract NO₂ national percentile on the ADI national rank, the Hispanic or Latino share, the Black alone share, with each county's population-weighted mean removed from every variable and each variable scaled to a standard deviation of one, errors clustered on county | residents of tracts in counties with at least 5 tracts that have every variable in the model | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 0.15 to 0.21 | 79,725 tracts | ACS 2019-2023 tract population | EPA EJScreen | A descriptive association within counties, not a causal estimate. |
| no2_pov | +0.18 | correlation coefficient | population-weighted correlation of the tract NO₂ national percentile with the tract poverty rate, county-cluster bootstrap interval | residents of tracts with a NO₂ percentile and the tract poverty rate | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 0.15 to 0.21 | 82,816 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| no2_res_all | 50 | national percentile of the tract (0 to 100) | mean over tracts of the NO₂ national percentile, each tract weighted by the number of residents in the group (tract grain, not an individual's exposure), county-cluster bootstrap interval | all residents of the tracts in the analysis | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 47 to 53 | 83,003 tracts | ACS 2019-2023 tract count of residents in the group | EPA EJScreen |   |
| no2_res_black | 56 | national percentile of the tract (0 to 100) | mean over tracts of the NO₂ national percentile, each tract weighted by the number of residents in the group (tract grain, not an individual's exposure), county-cluster bootstrap interval | Black or African American residents who report that race alone (of any ethnicity) of the tracts in the analysis | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 52 to 60 | 74,190 tracts | ACS 2019-2023 tract count of residents in the group | EPA EJScreen |   |
| no2_res_hisp | 63 | national percentile of the tract (0 to 100) | mean over tracts of the NO₂ national percentile, each tract weighted by the number of residents in the group (tract grain, not an individual's exposure), county-cluster bootstrap interval | Hispanic or Latino residents (of any race) of the tracts in the analysis | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 59 to 67 | 80,460 tracts | ACS 2019-2023 tract count of residents in the group | EPA EJScreen |   |
| pc1_asthma | −0.10 | correlation coefficient | population-weighted correlation of the tract's modeled prevalence of current asthma among adults with the urban-industrial component (PC1), county-cluster bootstrap interval | residents of tracts in the PCA with a PLACES estimate for current asthma | PLACES 2024 release (BRFSS 2022); EJScreen 2024; ACS 2019-2023 | 95% CI −0.18 to 0.00 | 80,674 tracts | ACS 2019-2023 tract population | CDC PLACES | PLACES prevalence is a model-based estimate from survey responses, so it is not a measured rate. |
| pc1_copd | −0.37 | correlation coefficient | population-weighted correlation of the tract's modeled prevalence of COPD among adults with the urban-industrial component (PC1), county-cluster bootstrap interval | residents of tracts in the PCA with a PLACES estimate for COPD | PLACES 2024 release (BRFSS 2022); EJScreen 2024; ACS 2019-2023 | 95% CI −0.42 to −0.32 | 80,674 tracts | ACS 2019-2023 tract population | CDC PLACES | PLACES prevalence is a model-based estimate from survey responses, so it is not a measured rate. |
| pc1_diesel | +0.40 | weight on the standardized Diesel PM percentile in component 1 (unit-length eigenvector) | eigenvector entry of component 1 (signs fixed so that the traffic weight on component 1 and the storage-tank weight on component 2 are positive), county-cluster bootstrap interval | tracts with all 13 EJScreen percentiles | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 0.38 to 0.41 | 80,754 tracts | none (each tract counts once) | EPA EJScreen |   |
| pc1_hazwaste | +0.39 | weight on the standardized Hazardous-waste facilities percentile in component 1 (unit-length eigenvector) | eigenvector entry of component 1 (signs fixed so that the traffic weight on component 1 and the storage-tank weight on component 2 are positive), county-cluster bootstrap interval | tracts with all 13 EJScreen percentiles | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 0.37 to 0.40 | 80,754 tracts | none (each tract counts once) | EPA EJScreen |   |
| pc1_no2 | +0.38 | weight on the standardized Nitrogen dioxide percentile in component 1 (unit-length eigenvector) | eigenvector entry of component 1 (signs fixed so that the traffic weight on component 1 and the storage-tank weight on component 2 are positive), county-cluster bootstrap interval | tracts with all 13 EJScreen percentiles | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 0.36 to 0.39 | 80,754 tracts | none (each tract counts once) | EPA EJScreen |   |
| pc1_traffic | +0.40 | weight on the standardized Traffic proximity percentile in component 1 (unit-length eigenvector) | eigenvector entry of component 1 (signs fixed so that the traffic weight on component 1 and the storage-tank weight on component 2 are positive), county-cluster bootstrap interval | tracts with all 13 EJScreen percentiles | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 0.39 to 0.41 | 80,754 tracts | none (each tract counts once) | EPA EJScreen |   |
| pc1_var | 38.6% | percent of the variance of the 13 standardized percentiles | eigenvalue share of component 1 of the correlation matrix of the 13 percentiles, each tract counted once, county-cluster bootstrap interval | tracts with all 13 EJScreen percentiles | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 36.2% to 41.7% | 80,754 tracts | none (each tract counts once) | EPA EJScreen |   |
| pc1_var_w | 37.1% | percent of the variance of the 13 standardized percentiles | eigenvalue share of component 1 when the standardization and the correlation matrix use population weights | tracts with all 13 EJScreen percentiles, each weighted by its residents | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 |   | 80,754 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| pc2_lead | +0.45 | weight on the standardized Lead paint (housing built before 1960) percentile in component 2 (unit-length eigenvector) | eigenvector entry of component 2 (signs fixed so that the traffic weight on component 1 and the storage-tank weight on component 2 are positive), county-cluster bootstrap interval | tracts with all 13 EJScreen percentiles | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 0.33 to 0.56 | 80,754 tracts | none (each tract counts once) | EPA EJScreen |   |
| pc2_ozone | −0.31 | weight on the standardized Ozone percentile in component 2 (unit-length eigenvector) | eigenvector entry of component 2 (signs fixed so that the traffic weight on component 1 and the storage-tank weight on component 2 are positive), county-cluster bootstrap interval | tracts with all 13 EJScreen percentiles | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI −0.38 to −0.16 | 80,754 tracts | none (each tract counts once) | EPA EJScreen |   |
| pc2_pm25 | −0.47 | weight on the standardized PM2.5 percentile in component 2 (unit-length eigenvector) | eigenvector entry of component 2 (signs fixed so that the traffic weight on component 1 and the storage-tank weight on component 2 are positive), county-cluster bootstrap interval | tracts with all 13 EJScreen percentiles | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI −0.55 to −0.38 | 80,754 tracts | none (each tract counts once) | EPA EJScreen |   |
| pc2_tanks | +0.55 | weight on the standardized Underground storage tanks percentile in component 2 (unit-length eigenvector) | eigenvector entry of component 2 (signs fixed so that the traffic weight on component 1 and the storage-tank weight on component 2 are positive), county-cluster bootstrap interval | tracts with all 13 EJScreen percentiles | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 0.33 to 0.66 | 80,754 tracts | none (each tract counts once) | EPA EJScreen |   |
| pc2_var | 11.7% | percent of the variance of the 13 standardized percentiles | eigenvalue share of component 2 of the correlation matrix of the 13 percentiles, each tract counted once, county-cluster bootstrap interval | tracts with all 13 EJScreen percentiles | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 10.5% to 13.1% | 80,754 tracts | none (each tract counts once) | EPA EJScreen |   |
| pc2_water | +0.32 | weight on the standardized Drinking-water violations percentile in component 2 (unit-length eigenvector) | eigenvector entry of component 2 (signs fixed so that the traffic weight on component 1 and the storage-tank weight on component 2 are positive), county-cluster bootstrap interval | tracts with all 13 EJScreen percentiles | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 0.19 to 0.43 | 80,754 tracts | none (each tract counts once) | EPA EJScreen |   |
| pm25_adi | −0.18 | correlation coefficient | population-weighted correlation of the tract PM₂.₅ national percentile with the ADI national rank, county-cluster bootstrap interval | residents of tracts with a PM₂.₅ percentile and an ADI rank | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI −0.29 to −0.06 | 82,198 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| pm25_adi_d1 | 64 | national percentile of the tract (0 to 100) | population-weighted mean of the tract PM₂.₅ national percentile in an ADI decile (deciles hold equal numbers of tracts), county-cluster bootstrap interval | residents of tracts in the least deprived of the ten ADI deciles (decile 1) | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 52 to 73 | 8,149 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| pm25_adi_d10 | 48 | national percentile of the tract (0 to 100) | population-weighted mean of the tract PM₂.₅ national percentile in an ADI decile (deciles hold equal numbers of tracts), county-cluster bootstrap interval | residents of tracts in the most deprived of the ten ADI deciles (decile 10) | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 44 to 51 | 8,203 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| pm25_adi_d9 | 44 | national percentile of the tract (0 to 100) | population-weighted mean of the tract PM₂.₅ national percentile in an ADI decile (deciles hold equal numbers of tracts), county-cluster bootstrap interval | residents of tracts in the ninth of the ten ADI deciles (decile 9) | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 41 to 47 | 8,247 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| pm25_adi_metro | −0.10 | correlation coefficient | population-weighted correlation of the tract PM₂.₅ national percentile with the ADI national rank, metropolitan tracts only, county-cluster bootstrap interval | residents of metropolitan tracts with a PM₂.₅ percentile and the ADI national rank | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI −0.22 to 0.03 | 66,593 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| pm25_adi_state | −0.18 | correlation coefficient | population-weighted correlation of the tract PM₂.₅ national percentile with the ADI national rank, bootstrap interval from redrawing whole states | residents of tracts with a PM₂.₅ percentile and the ADI national rank | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI −0.38 to 0.10 | 82,198 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| pm25_black | +0.06 | correlation coefficient | population-weighted correlation of the tract PM₂.₅ national percentile with the tract's Black alone share, county-cluster bootstrap interval | residents of tracts with a PM₂.₅ percentile and a race and ethnicity share | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 0.00 to 0.12 | 83,008 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| pm25_hisp | +0.32 | correlation coefficient | population-weighted correlation of the tract PM₂.₅ national percentile with the tract's Hispanic or Latino share, county-cluster bootstrap interval | residents of tracts with a PM₂.₅ percentile and a race and ethnicity share | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 0.22 to 0.40 | 83,008 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| pm25_q25 | 7.3 | micrograms of PM₂.₅ per cubic meter, annual average | 25th percentile of the tract concentrations (each tract counts once) | tracts with a modeled PM₂.₅ value | air quality modeled for 2020 |   | 83,008 tracts | none (each tract counts once) | EPA EJScreen |   |
| pm25_q75 | 9.0 | micrograms of PM₂.₅ per cubic meter, annual average | 75th percentile of the tract concentrations (each tract counts once) | tracts with a modeled PM₂.₅ value | air quality modeled for 2020 |   | 83,008 tracts | none (each tract counts once) | EPA EJScreen |   |
| pm25_ug_d1 | 9.6 | micrograms of PM₂.₅ per cubic meter, annual average | population-weighted mean of the tract's modeled PM₂.₅ concentration, county-cluster bootstrap interval | residents of tracts in the least deprived ADI decile | air quality modeled for 2020; ADI 2023 | 95% CI 8.7 to 10.4 | 8,149 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| pm25_ug_d10 | 8.1 | micrograms of PM₂.₅ per cubic meter, annual average | population-weighted mean of the tract's modeled PM₂.₅ concentration, county-cluster bootstrap interval | residents of tracts in the most deprived ADI decile | air quality modeled for 2020; ADI 2023 | 95% CI 7.9 to 8.2 | 8,203 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| pop_frame | 330.1 million | people | sum of tract population | residents of the tracts in the analysis | ACS 2019-2023 |   |   | none (full count) | American Community Survey |   |
| ptraf_adi | −0.41 | correlation coefficient | population-weighted correlation of the tract Traffic proximity national percentile with the ADI national rank, county-cluster bootstrap interval | residents of tracts with a Traffic proximity percentile and an ADI rank | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI −0.49 to −0.33 | 82,198 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| ptraf_adi_between | −0.66 | correlation coefficient | correlation across counties of the county's population-weighted mean traffic percentile with its population-weighted mean ADI rank, each county weighted by its residents, bootstrap interval over counties | residents of counties, one point per county with at least one tract that has a traffic-proximity percentile and an ADI rank | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI −0.73 to −0.58 | 3,095 counties | ACS 2019-2023 county population (sum of tract population) | EPA EJScreen |   |
| ptraf_adi_d1 | 75 | national percentile of the tract (0 to 100) | population-weighted mean of the tract Traffic proximity national percentile in an ADI decile (deciles hold equal numbers of tracts), county-cluster bootstrap interval | residents of tracts in the least deprived of the ten ADI deciles (decile 1) | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 68 to 79 | 8,149 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| ptraf_adi_d10 | 39 | national percentile of the tract (0 to 100) | population-weighted mean of the tract Traffic proximity national percentile in an ADI decile (deciles hold equal numbers of tracts), county-cluster bootstrap interval | residents of tracts in the most deprived of the ten ADI deciles (decile 10) | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 35 to 42 | 8,203 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| ptraf_adi_d9 | 35 | national percentile of the tract (0 to 100) | population-weighted mean of the tract Traffic proximity national percentile in an ADI decile (deciles hold equal numbers of tracts), county-cluster bootstrap interval | residents of tracts in the ninth of the ten ADI deciles (decile 9) | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 32 to 38 | 8,247 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| ptraf_adi_metro | −0.31 | correlation coefficient | population-weighted correlation of the tract Traffic proximity national percentile with the ADI national rank, metropolitan tracts only, county-cluster bootstrap interval | residents of metropolitan tracts with a Traffic proximity percentile and the ADI national rank | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI −0.40 to −0.21 | 66,593 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| ptraf_adi_state | −0.41 | correlation coefficient | population-weighted correlation of the tract Traffic proximity national percentile with the ADI national rank, bootstrap interval from redrawing whole states | residents of tracts with a Traffic proximity percentile and the ADI national rank | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI −0.52 to −0.25 | 82,198 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| ptraf_adi_within | +0.19 | correlation coefficient | population-weighted correlation of the tract Traffic proximity national percentile with the ADI national rank, after removing each county's population-weighted mean from both (counties with at least 5 tracts), county-cluster bootstrap interval | residents of tracts in counties with at least 5 tracts that have a Traffic proximity percentile and the ADI national rank | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 0.15 to 0.22 | 79,730 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| ptraf_black | +0.17 | correlation coefficient | population-weighted correlation of the tract Traffic proximity national percentile with the tract's Black alone share, county-cluster bootstrap interval | residents of tracts with a Traffic proximity percentile and a race and ethnicity share | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 0.12 to 0.22 | 82,998 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| ptraf_black_within | +0.16 | correlation coefficient | population-weighted correlation of the tract Traffic proximity national percentile with the tract's Black alone share after removing each county's population-weighted mean from both, county-cluster bootstrap interval | residents of tracts in counties with at least 5 tracts that have a Traffic proximity percentile and a race and ethnicity share | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 0.14 to 0.19 | 80,544 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| ptraf_hisp | +0.33 | correlation coefficient | population-weighted correlation of the tract Traffic proximity national percentile with the tract's Hispanic or Latino share, county-cluster bootstrap interval | residents of tracts with a Traffic proximity percentile and a race and ethnicity share | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 0.27 to 0.40 | 82,998 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| ptraf_hisp_within | +0.19 | correlation coefficient | population-weighted correlation of the tract Traffic proximity national percentile with the tract's Hispanic or Latino share after removing each county's population-weighted mean from both, county-cluster bootstrap interval | residents of tracts in counties with at least 5 tracts that have a Traffic proximity percentile and a race and ethnicity share | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 0.16 to 0.21 | 80,544 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| ptraf_joint_adi | +0.06 | within-county standard deviations of the Traffic proximity percentile per within-county standard deviation of the predictor | coefficient on the ADI national rank in a population-weighted regression of the tract Traffic proximity national percentile on the ADI national rank, the Hispanic or Latino share, the Black alone share, with each county's population-weighted mean removed from every variable and each variable scaled to a standard deviation of one, errors clustered on county | residents of tracts in counties with at least 5 tracts that have every variable in the model | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 0.02 to 0.10 | 79,730 tracts | ACS 2019-2023 tract population | EPA EJScreen | A descriptive association within counties, not a causal estimate. |
| ptraf_joint_black | +0.15 | within-county standard deviations of the Traffic proximity percentile per within-county standard deviation of the predictor | coefficient on the Black alone share in a population-weighted regression of the tract Traffic proximity national percentile on the ADI national rank, the Hispanic or Latino share, the Black alone share, with each county's population-weighted mean removed from every variable and each variable scaled to a standard deviation of one, errors clustered on county | residents of tracts in counties with at least 5 tracts that have every variable in the model | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 0.12 to 0.19 | 79,730 tracts | ACS 2019-2023 tract population | EPA EJScreen | A descriptive association within counties, not a causal estimate. |
| ptraf_joint_hisp | +0.17 | within-county standard deviations of the Traffic proximity percentile per within-county standard deviation of the predictor | coefficient on the Hispanic or Latino share in a population-weighted regression of the tract Traffic proximity national percentile on the ADI national rank, the Hispanic or Latino share, the Black alone share, with each county's population-weighted mean removed from every variable and each variable scaled to a standard deviation of one, errors clustered on county | residents of tracts in counties with at least 5 tracts that have every variable in the model | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 0.14 to 0.21 | 79,730 tracts | ACS 2019-2023 tract population | EPA EJScreen | A descriptive association within counties, not a causal estimate. |
| ptraf_jointb_black | +0.10 | within-county standard deviations of the Traffic proximity percentile per within-county standard deviation of the predictor | coefficient on the Black alone share in a population-weighted regression of the tract Traffic proximity national percentile on the tract poverty rate, the Hispanic or Latino share, the Black alone share, with each county's population-weighted mean removed from every variable and each variable scaled to a standard deviation of one, errors clustered on county | residents of tracts in counties with at least 5 tracts that have every variable in the model | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 0.07 to 0.14 | 80,353 tracts | ACS 2019-2023 tract population | EPA EJScreen | A descriptive association within counties, not a causal estimate. |
| ptraf_jointb_hisp | +0.14 | within-county standard deviations of the Traffic proximity percentile per within-county standard deviation of the predictor | coefficient on the Hispanic or Latino share in a population-weighted regression of the tract Traffic proximity national percentile on the tract poverty rate, the Hispanic or Latino share, the Black alone share, with each county's population-weighted mean removed from every variable and each variable scaled to a standard deviation of one, errors clustered on county | residents of tracts in counties with at least 5 tracts that have every variable in the model | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 0.10 to 0.17 | 80,353 tracts | ACS 2019-2023 tract population | EPA EJScreen | A descriptive association within counties, not a causal estimate. |
| ptraf_jointb_pov | +0.19 | within-county standard deviations of the Traffic proximity percentile per within-county standard deviation of the predictor | coefficient on the tract poverty rate in a population-weighted regression of the tract Traffic proximity national percentile on the tract poverty rate, the Hispanic or Latino share, the Black alone share, with each county's population-weighted mean removed from every variable and each variable scaled to a standard deviation of one, errors clustered on county | residents of tracts in counties with at least 5 tracts that have every variable in the model | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 0.17 to 0.21 | 80,353 tracts | ACS 2019-2023 tract population | EPA EJScreen | A descriptive association within counties, not a causal estimate. |
| ptraf_metro_d1 | 76 | national percentile of the tract (0 to 100) | population-weighted mean of the tract Traffic proximity national percentile, metropolitan tracts only, ADI deciles of all tracts, county-cluster bootstrap interval | residents of metropolitan tracts (RUCA codes 1 to 3) in ADI decile 1 of all tracts | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 69 to 80 | 7,879 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| ptraf_metro_d9 | 48 | national percentile of the tract (0 to 100) | population-weighted mean of the tract Traffic proximity national percentile, metropolitan tracts only, ADI deciles of all tracts, county-cluster bootstrap interval | residents of metropolitan tracts (RUCA codes 1 to 3) in ADI decile 9 of all tracts | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 45 to 50 | 4,817 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| ptraf_pov | +0.07 | correlation coefficient | population-weighted correlation of the tract Traffic proximity national percentile with the tract poverty rate, county-cluster bootstrap interval | residents of tracts with a Traffic proximity percentile and the tract poverty rate | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 0.04 to 0.11 | 82,818 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| ptraf_pov_within | +0.27 | correlation coefficient | population-weighted correlation of the tract Traffic proximity national percentile with the tract poverty rate, after removing each county's population-weighted mean from both (counties with at least 5 tracts), county-cluster bootstrap interval | residents of tracts in counties with at least 5 tracts that have a Traffic proximity percentile and the tract poverty rate | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 0.24 to 0.29 | 80,353 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| redline_counties | 215 | counties | count of counties | counties with at least one graded tract in the analysis | HOLC maps of 1935 to 1940 |   |   | none (full count) | Mapping Inequality (HOLC maps) |   |
| redline_dslpm_a | 73 | national percentile of the tract (0 to 100) | mean of the tract diesel PM national percentile, population-weighted, county-cluster bootstrap interval | residents of tracts whose dominant 1930s HOLC grade is A | HOLC maps of 1935 to 1940; air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 69 to 77 | 672 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| redline_dslpm_d | 86 | national percentile of the tract (0 to 100) | mean of the tract diesel PM national percentile, population-weighted, county-cluster bootstrap interval | residents of tracts whose dominant 1930s HOLC grade is D | HOLC maps of 1935 to 1940; air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 81 to 89 | 2,978 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| redline_fe_counties | 199 | counties | count of counties that inform a within-county contrast between grades | counties with graded tracts of at least two different grades | HOLC maps of 1935 to 1940 |   |   | none (full count) | Mapping Inequality (HOLC maps) |   |
| redline_fe_dslpm | 5.1 | national percentile points, grade D minus grade A | coefficient on grade D, against grade A, in a population-weighted regression of the tract diesel PM national percentile on grade dummies with county fixed effects, bootstrap interval over counties | residents of graded tracts in counties that hold more than one grade | HOLC maps of 1935 to 1940; air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 3.9 to 6.5 | 10,950 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| redline_fe_no2 | 9.7 | national percentile points, grade D minus grade A | coefficient on grade D, against grade A, in a population-weighted regression of the tract NO₂ national percentile on grade dummies with county fixed effects, bootstrap interval over counties | residents of graded tracts in counties that hold more than one grade | HOLC maps of 1935 to 1940; air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 7.2 to 12.2 | 10,950 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| redline_fe_pm25 | 1.1 | national percentile points, grade D minus grade A | coefficient on grade D, against grade A, in a population-weighted regression of the tract PM₂.₅ national percentile on grade dummies with county fixed effects, bootstrap interval over counties | residents of graded tracts in counties that hold more than one grade | HOLC maps of 1935 to 1940; air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 0.2 to 2.0 | 10,950 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| redline_fe_ptraf | 3.9 | national percentile points, grade D minus grade A | coefficient on grade D, against grade A, in a population-weighted regression of the tract traffic national percentile on grade dummies with county fixed effects, bootstrap interval over counties | residents of graded tracts in counties that hold more than one grade | HOLC maps of 1935 to 1940; air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 2.7 to 5.1 | 10,950 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| redline_fe_stack | 0.6 | air hazards in the top quarter of tracts (0 to 5), grade D minus grade A | coefficient on grade D, against grade A, in a population-weighted regression of the tract stacked-hazard score on grade dummies with county fixed effects, bootstrap interval over counties | residents of graded tracts in counties that hold more than one grade | HOLC maps of 1935 to 1940; air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 0.4 to 0.7 | 10,950 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| redline_n | 10,950 | tracts | count of tracts with a dominant grade | 2020 tracts in the analysis at least half covered by HOLC polygons graded A to D | HOLC maps of 1935 to 1940; EJScreen 2024 |   |   | none (full count) | Mapping Inequality (HOLC maps) |   |
| redline_no2_a | 73 | national percentile of the tract (0 to 100) | mean of the tract NO₂ national percentile, population-weighted, county-cluster bootstrap interval | residents of tracts whose dominant 1930s HOLC grade is A | HOLC maps of 1935 to 1940; air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 70 to 76 | 672 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| redline_no2_d | 86 | national percentile of the tract (0 to 100) | mean of the tract NO₂ national percentile, population-weighted, county-cluster bootstrap interval | residents of tracts whose dominant 1930s HOLC grade is D | HOLC maps of 1935 to 1940; air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 83 to 89 | 2,978 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| redline_pm25_a | 56 | national percentile of the tract (0 to 100) | mean of the tract PM₂.₅ national percentile, population-weighted, county-cluster bootstrap interval | residents of tracts whose dominant 1930s HOLC grade is A | HOLC maps of 1935 to 1940; air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 49 to 63 | 672 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| redline_pm25_d | 60 | national percentile of the tract (0 to 100) | mean of the tract PM₂.₅ national percentile, population-weighted, county-cluster bootstrap interval | residents of tracts whose dominant 1930s HOLC grade is D | HOLC maps of 1935 to 1940; air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 51 to 69 | 2,978 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| redline_pop | 37.9 million | people | sum of tract population | residents of the graded tracts | ACS 2019-2023 |   |   | none (full count) | American Community Survey |   |
| redline_ptraf_a | 78 | national percentile of the tract (0 to 100) | mean of the tract traffic national percentile, population-weighted, county-cluster bootstrap interval | residents of tracts whose dominant 1930s HOLC grade is A | HOLC maps of 1935 to 1940; air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 75 to 80 | 672 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| redline_ptraf_d | 86 | national percentile of the tract (0 to 100) | mean of the tract traffic national percentile, population-weighted, county-cluster bootstrap interval | residents of tracts whose dominant 1930s HOLC grade is D | HOLC maps of 1935 to 1940; air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 83 to 89 | 2,978 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| redline_stack_a | 2.3 | air hazards in the top quarter of tracts (0 to 5) | mean of the stacked-exposure score, population-weighted, county-cluster bootstrap interval | residents of tracts whose dominant 1930s HOLC grade is A | HOLC maps of 1935 to 1940; air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 2.0 to 2.5 | 672 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| redline_stack_d | 3.2 | air hazards in the top quarter of tracts (0 to 5) | mean of the stacked-exposure score, population-weighted, county-cluster bootstrap interval | residents of tracts whose dominant 1930s HOLC grade is D | HOLC maps of 1935 to 1940; air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 2.8 to 3.5 | 2,978 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| rural_all | 6.9% | percent of residents living in small-town and rural tracts | population-weighted share of residents in tracts coded 7 to 10 in the USDA Rural-Urban Commuting Area scheme | residents of all tracts with a RUCA code | RUCA 2010 codes on 2020 tracts; ACS 2019-2023 |   | 82,124 tracts | ACS 2019-2023 tract population | Area Deprivation Index |   |
| rural_hi | 18.5% | percent of residents living in small-town and rural tracts | population-weighted share of residents in tracts coded 7 to 10 in the USDA Rural-Urban Commuting Area scheme, county-cluster bootstrap interval | residents of tracts in ADI decile 10 with a RUCA code (Connecticut has none) | RUCA 2010 codes on 2020 tracts; ADI 2023; ACS 2019-2023 | 95% CI 16.2% to 21.0% | 8,194 tracts | ACS 2019-2023 tract population | Area Deprivation Index |   |
| rural_lo | 0.6% | percent of residents living in small-town and rural tracts | population-weighted share of residents in tracts coded 7 to 10 in the USDA Rural-Urban Commuting Area scheme, county-cluster bootstrap interval | residents of tracts in ADI decile 1 with a RUCA code (Connecticut has none) | RUCA 2010 codes on 2020 tracts; ADI 2023; ACS 2019-2023 | 95% CI 0.4% to 1.1% | 8,077 tracts | ACS 2019-2023 tract population | Area Deprivation Index |   |
| rural_peak | 20.3% | percent of residents living in small-town and rural tracts | population-weighted share of residents in tracts coded 7 to 10 in the USDA Rural-Urban Commuting Area scheme, county-cluster bootstrap interval | residents of tracts in ADI decile 9 with a RUCA code (Connecticut has none) | RUCA 2010 codes on 2020 tracts; ADI 2023; ACS 2019-2023 | 95% CI 18.4% to 22.3% | 8,214 tracts | ACS 2019-2023 tract population | Area Deprivation Index |   |
| stack_adi_hi | 1.0 | air hazards in the top quarter of tracts (0 to 5) | population-weighted mean of the stacked-exposure score within a decile of tracts, county-cluster bootstrap interval | residents of tracts in decile 10 of the ADI rank | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 0.8 to 1.2 | 8,203 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| stack_adi_lo | 2.1 | air hazards in the top quarter of tracts (0 to 5) | population-weighted mean of the stacked-exposure score within a decile of tracts, county-cluster bootstrap interval | residents of tracts in decile 1 of the ADI rank | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 1.7 to 2.5 | 8,146 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| stack_black_hi | 1.7 | air hazards in the top quarter of tracts (0 to 5) | population-weighted mean of the stacked-exposure score within a decile of tracts, county-cluster bootstrap interval | residents of tracts in decile 10 of the Black alone share | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 1.4 to 2.0 | 8,298 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| stack_black_lo | 0.8 | air hazards in the top quarter of tracts (0 to 5) | population-weighted mean of the stacked-exposure score within a decile of tracts, county-cluster bootstrap interval | residents of tracts in decile 1 of the Black alone share | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 0.6 to 1.0 | 8,809 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| stack_cut | 75 | national percentile | a hazard counts as stacked when the tract's national percentile is at or above this cutoff (the top quarter of tracts); a definition | tracts | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 |   |   | none (definition) | EPA EJScreen |   |
| stack_hisp_hi | 2.3 | air hazards in the top quarter of tracts (0 to 5) | population-weighted mean of the stacked-exposure score within a decile of tracts, county-cluster bootstrap interval | residents of tracts in decile 10 of the Hispanic or Latino share | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 1.8 to 2.8 | 8,299 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| stack_hisp_lo | 0.6 | air hazards in the top quarter of tracts (0 to 5) | population-weighted mean of the stacked-exposure score within a decile of tracts, county-cluster bootstrap interval | residents of tracts in decile 1 of the Hispanic or Latino share | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 0.5 to 0.7 | 8,295 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| stack_n_haz | 5 | air hazards | number of air hazards in the stacked score; a definition | PM₂.₅, diesel PM, NO₂, traffic proximity and toxic releases to air | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 |   |   | none (definition) | EPA EJScreen |   |
| ust_adi | +0.25 | correlation coefficient | population-weighted correlation of the tract Underground storage tanks national percentile with the ADI national rank, county-cluster bootstrap interval | residents of tracts with a Underground storage tanks percentile and an ADI rank | air indicators modeled for 2020 (toxic releases for 2021), population and deprivation 2019-2023 | 95% CI 0.13 to 0.36 | 82,198 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |

## Definitions

- National percentile: A neighborhood's position in the national ordering, given as the share of all neighborhoods that fall below it. A higher percentile means a higher value than more of the rest of the country.
- Area Deprivation Index (ADI): A score that ranks neighborhoods by income, education, employment and housing conditions drawn from the American Community Survey. In this book a higher national percentile means a more deprived neighborhood.
- Census tract: A small area drawn by the Census Bureau to stand in for a neighborhood. Boundaries are redrawn after each decennial census.
- Modeled estimate: A figure produced by a statistical model instead of a direct count or measurement, usually for small areas where direct data are thin, borrowing strength from surveys, neighboring areas or other data. Neighboring areas can share errors, so single extremes call for caution.
- Population-weighted: Averaged so that each place counts in proportion to its number of residents, which lets a large county move the result more than a small one. An unweighted average counts every place equally.
- Bootstrap interval: A range for an estimate found by recomputing it on many random resamples of the data and reading the range from the results. It shows how much the estimate would vary from sample to sample.
- Principal components: A method that combines many correlated measures into a few summary scores, ordered by how much of the variation among places each captures. The first component is the single weighted combination that captures the most.
- HOLC grade: A letter from A to D that the Home Owners' Loan Corporation gave neighborhoods on its 1930s maps, with A labeled best and D hazardous. The maps are known as redlining maps.
- Ecological inference: Drawing conclusions about individuals from data about places or groups. A pattern across counties need not hold for the people inside them.

## Methods

This chapter compares census tracts on the national percentiles of EPA's EJScreen 2024 air indicators, set against the Area Deprivation Index, race and ethnicity, rural or metropolitan status, PLACES asthma and COPD, and the grades of the 1930s HOLC maps. The indicators are EPA model output for 2020 and 2021, not monitor readings, and each percentile is the share of tracts with a lower value. The tracts in Alaska, Hawaii and Puerto Rico that lack PM2.5 and ozone are left out, and the HOLC comparisons cover only tracts that the digitized maps grade. Estimates are modeled and carry no sampling error beyond the bootstrap intervals, which describe which places are in the file.

### Where the air indicators come from

Air indicators are the national percentiles in EPA's EJScreen 2024 tract file, as archived on Zenodo (release 2.32, after EPA removed the tool in 2025). EPA's documentation describes percentiles ranked across block groups; the percentiles in this file rank tracts (their unweighted mean is close to 50), so a percentile is the share of tracts with a lower value.

EPA's documentation dates PM2.5, ozone, NO2 and diesel PM to 2020, toxic releases to air to 2021 and traffic to 2020 traffic counts; PM2.5 is in micrograms per cubic meter, ozone and NO2 in parts per billion, diesel PM in micrograms per cubic meter, and traffic and toxic releases are indices. The indicators are EPA model output, not monitor readings.

Explains Table 1.

### Which tracts are included and how they are weighted

The tract file is on 2020 tracts, like the tract atlas and PLACES, and joins on the 11-character GEOID without a crosswalk (Connecticut carries planning-region codes in all of them); the 686 territory rows of EJScreen are not on the spine and drop out.

PM2.5 and ozone are missing for all tracts in Alaska, Hawaii and Puerto Rico (the tracts with residents in those places are left out of the analysis), and diesel PM is missing for Alaska.

A tract is kept when its ACS 2019-2023 population is positive, and every mean and correlation weights tracts by that population.

### How the groups and exposure measures are defined

Deciles and twentieths hold nearly equal numbers of tracts, and tracts tied on the sorting variable share a bin, so the lowest Black-share decile holds all tracts with a share of zero.

Race and ethnicity shares come from the ACS tract file: Hispanic or Latino of any race, and Black or African American alone of any ethnicity, so the groups overlap.

The stacked score counts how many of PM2.5, diesel PM, NO2, traffic proximity and toxic releases to air have a national percentile of 75 or more. Ozone is left out because it forms in the air from other pollutants and does not mark a local source; a count of six that adds it keeps the same ordering of the three sorting variables.

Rural tracts are USDA RUCA codes 7 to 10 and metropolitan tracts codes 1 to 3, from the staged geography file (2010 revised codes mapped to 2020 tracts by largest land overlap); Connecticut has no code.

PLACES values are model-based estimates of crude prevalence among adults.

Explains Table 3, Table 4, Table 2 and Table 8.

### How redlining grades are matched to tracts

HOLC polygons graded A to D are intersected with 2020 tracts in an equal-area projection; a tract takes the grade covering most of its area when graded polygons cover at least half of it.

The pooled grade means compare tracts across counties; the within-county contrasts come from a population-weighted regression of the tract value on grade with county fixed effects.

Explains Table 9.

### How intervals and correlations are computed

Intervals for means and correlations are 95% percentile intervals from 2,000 draws of whole counties with replacement (seed 12345); Connecticut counts as its nine planning regions.

Three kinds of interval differ. The coefficients of the within-county joint models are weighted least squares with errors clustered on county and t intervals. Two correlations in the limits section (traffic and PM2.5 against the ADI rank) and the old-housing, air and race counts of the correlation strip redraw whole states instead, with the same number of draws and seed; the strip shades a cell by the wider of its county and state intervals, and a count requires both intervals to exclude zero. The county-level correlation of mean traffic percentile with mean ADI rank bootstraps counties, and the grade contrasts within counties bootstrap counties as well.

Correlations are population-weighted Pearson correlations. Within-county correlations subtract each county's population-weighted mean from both variables and use counties with at least 5 tracts that have both values.

Estimates are modeled and carry no sampling error beyond these bootstraps, which describe which places are in the file.

Explains Table 7.

### How the principal components and the map are built

The PCA standardizes the 13 percentiles to mean zero and standard deviation one across the 80,754 tracts that have all 13, counts each tract once, and takes the eigenvectors of the correlation matrix; the intervals redraw whole counties, rebuild the matrix and match eigenvectors to the full-sample components. County scores are the population-weighted means of the tract scores on the first component; a county is mapped when at least 50% of its residents live in tracts with all 13 indicators.

Puerto Rico is not drawn on maps.

Explains Table 5 and Table 6.

### Four rankings of the same places

The comparison follows the book's ranking switch rule, fixed on October 28, 2026 before any of these gradients was computed. The main gradient of the chapter is recomputed under four rankings of the same places, with the outcome, period, weights, bins and intervals unchanged: the published ADI; a standardized index that applies the ADI's coefficients to its measures after standardizing each one across tracts; median household income; and the share of adults 25 and older with a bachelor's degree or more. The three alternatives come from ACS tract estimates for 2019 to 2023, the period of the ADI.

All four rankings use the same places, the 75,058 tracts that have the outcome and all four rankings, so the ADI panel can differ slightly from the chapter's main figure. Tracts enter only if they have every input of the standardized index, which leaves out tracts without a median rent, home value or mortgage cost.

A pattern counts as changed when the worse end switches sides, when the two ends differ under one ranking and not the other, or when the rank correlation of bin order and bin mean falls on the other side of 0.9 and moves by at least 0.1.

Explains Table 10.

## What these data cannot show

Percentiles rank tracts and do not give a level of exposure, and a percentile cannot say whether a concentration meets a health guideline. The indicators are EPA model output for 2020 and 2021 with their own error, assigned to tracts from finer geography. The negative link between deprivation and pollution percentiles depends on the measure of deprivation and on the comparison, and each comparison answers a different question. The ADI includes home values and rents, the poverty rate does not, and a within-county comparison holds the metropolitan area fixed while it discards differences between areas.

Everything here is ecological. The shares of Hispanic and Black residents are ACS estimates for 2019 to 2023 without margins of error, the two groups overlap, and a tract average does not describe any one of its residents. Counties in one state share traffic, industry and climate, so the county bootstrap gives intervals narrower than one that redraws whole states. With states redrawn the correlation of the traffic percentile with the ADI rank is −0.41 (95% CI −0.52 to −0.25), which still excludes zero, and that of the PM₂.₅ percentile is −0.18 (95% CI −0.38 to 0.10), which includes it.

The principal components are a description. They use national percentiles, which discard the differences in concentration, and they cover only the tracts with all indicators, so Alaska, Hawaii and Puerto Rico are absent. County scores are means of modeled tract values. PLACES values are modeled from survey responses and demographic inputs, they are crude prevalence among adults, and asthma and COPD depend on age and smoking in ways the tract file does not separate. The HOLC comparison is among graded urban cores, matches a 1930s polygon to a 2020 tract by area, and describes the residents of today. It cannot separate what the grades did from the features of the neighborhoods that surveyors graded, and the box above shows what the available designs do and do not cover.

## References

1. Fitzgerald, William; Gehrke, Gretchen (Environmental Data & Governance Initiative) (2025). EPA Environmental Justice Screening Tool (EJScreen) data, 2015-2024. Zenodo, version 1, January 29, 2025; the record includes EPA's EJScreen Technical Documentation, version 2.3, revised July 31, 2024. https://doi.org/10.5281/zenodo.14767363
2. Singh, Gopal K. (2003). Area deprivation and widening inequalities in US mortality, 1969-1998. American Journal of Public Health 93(7): 1137-1143. https://doi.org/10.2105/ajph.93.7.1137
3. Bullard, Robert D. (1990). Dumping in Dixie: Race, Class, and Environmental Quality. Westview Press, Boulder, CO.
4. Tessum, Christopher W.; Apte, Joshua S.; Goodkind, Andrew L.; Muller, Nicholas Z.; Mullins, Kimberley A.; Paolella, David A.; Polasky, Stephen; Springer, Nathaniel P.; Thakrar, Sumil K.; Marshall, Julian D.; Hill, Jason D. (2019). Inequity in consumption of goods and services adds to racial-ethnic disparities in air pollution exposure. Proceedings of the National Academy of Sciences 116(13): 6001-6006. https://doi.org/10.1073/pnas.1818859116
5. Greenlund, Kurt J.; Lu, Hua; Wang, Yan; Matthews, Kevin A.; LeClercq, Jennifer M.; Lee, Benjamin; Carlson, Susan A. (2022). PLACES: Local Data for Better Health. Preventing Chronic Disease 19: 210459. https://doi.org/10.5888/pcd19.210459
6. Nelson, Robert K.; Winling, LaDale; Marciano, Richard; Connolly, Nathan; et al. (2023). Mapping Inequality: Redlining in New Deal America. American Panorama, ed. Robert K. Nelson and Edward L. Ayers, Digital Scholarship Lab, University of Richmond. https://dsl.richmond.edu/panorama/redlining/
7. Lane, Haley M.; Morello-Frosch, Rachel; Marshall, Julian D.; Apte, Joshua S. (2022). Historical Redlining Is Associated with Present-Day Air Pollution Disparities in U.S. Cities. Environmental Science & Technology Letters 9(4): 345-350. https://doi.org/10.1021/acs.estlett.1c01012
8. Salazar-Miranda, Arianna; Conzelmann, Claire; Phan, Toan; Hoffman, Jeremy (2024). Long-term effects of redlining on climate risk exposure. Nature Cities 1(6): 436-444. https://doi.org/10.1038/s44284-024-00076-y
9. Currie, Janet; Walker, Reed (2011). Traffic Congestion and Infant Health: Evidence from E-ZPass. American Economic Journal: Applied Economics 3(1): 65-90. https://doi.org/10.1257/app.3.1.65
10. Mohai, Paul; Saha, Robin (2015). Which came first, people or pollution? Assessing the disparate siting and post-siting demographic change hypotheses of environmental injustice. Environmental Research Letters 10(11): 115008. https://doi.org/10.1088/1748-9326/10/11/115008

## Provenance

- Pipeline version: 0.1.0
- Build script: book2/chapters/air-pollution/build.py (sha256 0b8756b035c9a7599cc06a5e52ee10ef3d07b0c772eecd492658bb8cbeea334e)
- Built on: 2026-10-07
- Site commit: c2b1ff3309cb4a10833f884da91ac0057a66ad0a

Inputs:

| File | Size (bytes) | SHA-256, first 1 MB | SHA-256, last 1 MB |
| --- | ---: | --- | --- |
| book2/chapters/air-pollution/air_data.py | 18,238 | b921a6825e8e5dc97ee09be50cf63985b022d77a81d9f4bbb70486c2dd4a8810 | b921a6825e8e5dc97ee09be50cf63985b022d77a81d9f4bbb70486c2dd4a8810 |
| data/analytic/tract_atlas.parquet | 39,659,925 | 29d5e4d943552d54aeb6ba92aadc939d9bd15bdea553c78357dbd516757be627 | 1e10d1ecf722340f23ee2212d4999a90250f85b65fa89b39e32db06d9472e22e |
| data/geo/dim_geo.parquet | 1,180,921 | 29d5e4d943552d54aeb6ba92aadc939d9bd15bdea553c78357dbd516757be627 | e58c8c488df54ab3187f5461f179322afdc49c9bd2315404003fe1272eea32c3 |
| data/geo/holc_areas_2023.parquet | 4,084,019 | 6d293647278dc82053c0ab89a192836f0763e73aff77785f3f06ad4214102600 | 37dc09e2ca32f1f9a001d07433b85faf4bd0a31ea584c16b62654b6556084be4 |
| data/geo/tracts_2020.parquet | 619,858,391 | f4f05c8c7a63d1785ec2aad283de3f6fa839b6c437f13a1b96cea31d3344bf0f | 7b3811247c6b3e322954f24a4a2384af52aef6ed8688f8dbdb0407d21ef80a59 |
| data/raw/acs/deprivation_components_2019_2023/tract.parquet | 13,197,793 | 5545cde2cbe96a8b56e468d65388248a241aa6bd300d3c9e325d08d3f86b11da | 5c8ec454d066988b0f1a01aaa6ed34ebfc05d162e01cf9c6bb01f9198cbaff12 |
| data/staged/adi/adi_atlas_tract2020_2023.parquet | 798,078 | 5e409bc212ff871def780fa27b4cf8b616d9da420b2b9f6455517eddcd8c1a95 | 5e409bc212ff871def780fa27b4cf8b616d9da420b2b9f6455517eddcd8c1a95 |
| data/staged/crosswalks/ct_tract2020_to_old_county.parquet | 15,842 | 961ef39f31596ef65fca2c67d5f2f63e533c5dc9b49745d3bb6437844e4b831d | 961ef39f31596ef65fca2c67d5f2f63e533c5dc9b49745d3bb6437844e4b831d |
| data/staged/environment/ejscreen_tract2020_2024.parquet | 10,891,402 | 8b728fb8c87df1d639a9652a31e22cc07d51aad31f3390a8ff6e82e5cebb6a86 | ea8df449172c133db5eb83c0c0df93aaaf8b781b373b75a0056fbde1b1d32824 |
| data/staged/health/places_tract2020_latest.parquet | 5,919,159 | af4c3b0d1205b7d84b81676a9529ef73323b2dbeb9d54e235f5efeebfad83177 | a2f19029ee2ef032be87e0d5288d88d489afd52379279833f178baa021df3dbf |
| data/staged/ses/acs5_tract2020_2019_2023.parquet | 7,232,424 | 5fe6f730a35225f14d5d8abfc6b9bce58b52e545f309a7eac74f1dae74d7a31c | 7b4a2475b3852d3f91804716a04566ef8ceadacec2be9fb85f5eb642078efac6 |
| data/staged/svi/svi_tract2020_2022.parquet | 17,740,415 | cb3860d8edc2aaf1f2a02211a58f7a0b80457eb6ad9a37b5ec8abb2d5da9da10 | e371a4a696d5bed05d54ad7c5b6ca2504508805b847ec415c2322be7f66ec2b2 |

Outputs:

| File | SHA-256 |
| --- | --- |
| facts.json | 08d2673946006133096b425bfb53d0134f0fa5d125180516c34ade93d4e8693a |
| figures/air_by_holc_grade.json | 8deca0cce4090c781c707ca8faf4ed46347be2749e2af8552d56f4be245de821 |
| figures/air_by_race_deciles.json | fe6c97d0faa05d3b4a776c10908210090bf1176814b3016d1f7817d8a25e6e2d |
| figures/air_correlation_strip.json | da717b75ad02951025e46984c4a4d255d4830ef3854c6074dd75c1d1d512bbfc |
| figures/air_vs_deprivation.json | f39920e02e942de8030bc8e4d33c42062e9a7775403ac2ca2f7adc7647c2dcb9 |
| figures/alt_rankings.json | fb9b7c1a5a852c52c286a80d9fd5a275f943fc01e710b5e3f23814aacf966268 |
| figures/asthma_vs_exposure.json | df2ab99b05283c073498dc67389978013e9c83033b25f65a202d3b2ca23b881c |
| figures/pc1_map.json | 4a6d9a3c3d9cb5ea01905f9585991a5068941517ac354c37262f347c535cc51a |
| figures/pca_loadings.json | 775a3e114afad55a5698d5e6cdfd613f340c4d7f734d170aa739cdda3d88fd10 |
| figures/rural_share_and_pm25.json | 9d07dee60f3c20cb84cfe3e9e804c417c15bc921328be265e4532630b69f42e2 |
| figures/stacked_exposure.json | 740c00d123abd178ee61cbc69af4fe6e164aa4c8bf4a4d71ebe237c15c221dda |
| meta.json | 369d449df1b378856e56f3c593f938cb3e46c815420d8544c8e2237d4aed43e1 |
| public/data/air-pollution/pc1_map.json | 42a309c754c9455e1a4180b45a6bba4e3bb9bd58e1fd138707d85e4663c339e0 |
| refs.json | 2af03780c82338c01d793ceef467a7e0d8296a83ad69eb166349300dcb22ee0a |

## Machine formats

- Report (HTML): https://vishalsingh.org/zipcode/air-pollution/report
- Report (Markdown): https://vishalsingh.org/zipcode/air-pollution/report.md
- Every estimate (JSON): https://vishalsingh.org/zipcode/report/air-pollution/facts.json
- Table 1, figure "air_vs_deprivation" (JSON): https://vishalsingh.org/zipcode/report/air-pollution/figures/air_vs_deprivation.json
- Table 2, figure "rural_share_and_pm25" (JSON): https://vishalsingh.org/zipcode/report/air-pollution/figures/rural_share_and_pm25.json
- Table 3, figure "air_by_race_deciles" (JSON): https://vishalsingh.org/zipcode/report/air-pollution/figures/air_by_race_deciles.json
- Table 4, figure "stacked_exposure" (JSON): https://vishalsingh.org/zipcode/report/air-pollution/figures/stacked_exposure.json
- Table 5, figure "pca_loadings" (JSON): https://vishalsingh.org/zipcode/report/air-pollution/figures/pca_loadings.json
- Table 6, figure "pc1_map" (JSON): https://vishalsingh.org/zipcode/report/air-pollution/figures/pc1_map.json
- Table 7, figure "air_correlation_strip" (JSON): https://vishalsingh.org/zipcode/report/air-pollution/figures/air_correlation_strip.json
- Table 8, figure "asthma_vs_exposure" (JSON): https://vishalsingh.org/zipcode/report/air-pollution/figures/asthma_vs_exposure.json
- Table 9, figure "air_by_holc_grade" (JSON): https://vishalsingh.org/zipcode/report/air-pollution/figures/air_by_holc_grade.json
- Table 10, figure "alt_rankings" (JSON): https://vishalsingh.org/zipcode/report/air-pollution/figures/alt_rankings.json
- Payload of "pc1_map" (JSON): https://vishalsingh.org/zipcode/data/air-pollution/pc1_map.json
- Index of every report: https://vishalsingh.org/zipcode/llms.txt; all reports in one file: https://vishalsingh.org/zipcode/llms-full.txt

---

# Civic life, loneliness and partnership: report

Chapter 6.2 of ZIP Code Destiny, by Vishal Singh (NYU Stern). Generated from the chapter's data files; the reading version is at https://vishalsingh.org/zipcode/civic-life-loneliness-and-partnership. Table n of this report is figure n of the chapter.

## Identification

- Title: Civic life, loneliness and partnership
- Label: Alone together
- Chapter: 6.2 (Part VI, Environment, connection and well-being; An American life)
- Question: How do civic participation, loneliness and living with a partner vary across places, and how do they relate to well-being?
- Dek: Voter turnout in the 2020 presidential election was 72.1% of adult citizens in the least deprived tenth of counties and 56.4% in the most deprived, and census self-response was 21.0 points lower across the same tenths.
- Grain: counties
- Status: published
- Published: 2026-10-09
- Updated: not revised
- Data sources:
  - County Health Rankings. County Health Rankings 2024: voter turnout (2020) and census participation (2020 census self-response), membership associations (County Business Patterns, 2021); years: 2024 release; weights: 2019-2023 county population for decile means; 2018 county population for correlations
  - Area Deprivation Index. Area Deprivation Index 2023, national rank, tract ranks averaged to counties; years: 2023 release (version 4.0.1; built on ACS 2019-2023); weights: 2019-2023 tract population
  - American Community Survey. ACS 2019-2023 tract file: tract population as the weight for county deprivation, and county controls (income, percent Black, poverty) for the regressions; years: 2019-2023 (5-year); weights: 2019-2023 tract population
  - Opportunity Atlas. County outcomes of children born 1978 to 1983 to parents at the 25th percentile (mean adult household income rank, as a percentile); years: 1978-1983 birth cohorts (2018 release); weights: 2018 county population (Social Capital Atlas)
  - Social Capital Atlas. Economic connectedness, civic groups per capita and volunteering group membership by county, 2022; years: 2022 (friendship snapshot); weights: 2018 county population (Social Capital Atlas)
  - JEC Social Capital Index. JEC Social Capital Index and subindices by county, 2018 release with components from 2010 to 2016, and religious congregations per 1,000 (2010); years: 2018 release (components from about 2010-2016); weights: 2018 county population (Social Capital Atlas)
  - Gallup U.S. Daily Poll. Gallup U.S. Daily tracking poll, 2009-2017 (the 2008 interviews are left out because of a question-order artifact in the ladder): Cantril ladder score and marital status; aggregates only; years: 2008-2017; weights: none (unweighted)
  - CDC PLACES. CDC PLACES 2025 release, ZCTA estimates of loneliness and depression (BRFSS 2023); ZCTA characteristics are ACS 2019-2023 five-year estimates for each ZCTA, from a Census API pull; years: 2024 release (tracts); 2025 release (ZCTAs, 2023 data year); weights: 2020 census adult population of the ZCTA
- Canonical URL: https://vishalsingh.org/zipcode/civic-life-loneliness-and-partnership
- Report: https://vishalsingh.org/zipcode/civic-life-loneliness-and-partnership/report
- Markdown: https://vishalsingh.org/zipcode/civic-life-loneliness-and-partnership/report.md
- BibTeX key: singh2026civiclifelonelinessandpartnership

```bibtex
@misc{singh2026civiclifelonelinessandpartnership,
  author = {Singh, Vishal},
  title = {Civic life, loneliness and partnership},
  year = {2026},
  url = {https://vishalsingh.org/zipcode/civic-life-loneliness-and-partnership},
  note = {In ZIP Code Destiny. Published 9 October 2026.}
}
```

## Principal estimates

The facts the dek and the opening prose cite, in order of appearance.

| Estimand | Population | Period | Value | Interval | n | Weights | Source | Fact id |
| --- | --- | --- | ---: | --- | ---: | --- | --- | --- |
| mean of county voter turnout in a tenth of counties (equal numbers of counties, ranked by county Area Deprivation Index; county ADI is the mean of the county's tract Area Deprivation Index national ranks, weighted by 2019-2023 tract population), weighted by county population, bootstrap interval over counties | citizens aged 18 and over in counties in the least deprived tenth of counties by Area Deprivation Index | 2020 presidential election | 72.1% | 95% CI 70.6% to 73.8% | 307 counties | 2019-2023 county population | County Health Rankings | voter_d1 |
| mean of county voter turnout in a tenth of counties (equal numbers of counties, ranked by county Area Deprivation Index; county ADI is the mean of the county's tract Area Deprivation Index national ranks, weighted by 2019-2023 tract population), weighted by county population, bootstrap interval over counties | citizens aged 18 and over in counties in the most deprived tenth of counties by Area Deprivation Index | 2020 presidential election | 56.4% | 95% CI 55.4% to 57.6% | 310 counties | 2019-2023 county population | County Health Rankings | voter_d10 |
| difference between the population-weighted means of census self-response in the least and the most deprived tenth of counties, bootstrap interval over counties with shared draws | households in counties in the two extreme tenths of counties by Area Deprivation Index | 2020 census | 21.0 | 95% CI 19.0 to 22.9 | 624 counties | 2019-2023 county population | County Health Rankings | census_gap |
| mean of county census self-response in a tenth of counties (equal numbers of counties, ranked by county Area Deprivation Index; county ADI is the mean of the county's tract Area Deprivation Index national ranks, weighted by 2019-2023 tract population), weighted by county population, bootstrap interval over counties | households in counties in the least deprived tenth of counties by Area Deprivation Index | 2020 census | 68.6% | 95% CI 67.4% to 69.9% | 312 counties | 2019-2023 county population | County Health Rankings | census_d1 |
| mean of county census self-response in a tenth of counties (equal numbers of counties, ranked by county Area Deprivation Index; county ADI is the mean of the county's tract Area Deprivation Index national ranks, weighted by 2019-2023 tract population), weighted by county population, bootstrap interval over counties | households in counties in the most deprived tenth of counties by Area Deprivation Index | 2020 census | 47.6% | 95% CI 46.1% to 49.1% | 312 counties | 2019-2023 county population | County Health Rankings | census_d10 |
| unweighted difference between the mean ladder score of adults who are married or in a domestic partnership and the mean of all others, interval clustered by county | adults interviewed with a ladder score, a marital status and a county | interviews in 2009 to 2017 | 0.50 | 95% CI 0.49 to 0.51 | 2,189,201 respondents | none (unweighted) | Gallup U.S. Daily Poll | partner_premium |
| correlation across ZCTAs of PLACES modeled loneliness with the married share of residents 15 and older, weighted by adult population, bootstrap interval over counties | adults in ZCTAs with a loneliness estimate and at least 500 residents | BRFSS 2023 (PLACES 2025 release) | −0.65 | 95% CI −0.69 to −0.62 | 19,681 ZCTAs | 2020 census adult population of the ZCTA | CDC PLACES | lonely_married_r |

## Findings

Voter turnout in the 2020 presidential election was 72.1% among adult citizens in the tenth of counties with the lowest Area Deprivation Index scores and 56.4% in the tenth with the highest. The share of households that answered the 2020 census on their own fell from 68.6% to 47.6% across the same ten groups.[1] The point estimates of both measures fall at every step from the least to the most deprived tenth, and neither keeps a link to upward mobility once county income, racial composition and poverty are held fixed.

Gallup respondents who were married or in a domestic partnership rated their lives 0.50 (95% CI 0.49 to 0.51) ladder points higher than other adults, and the difference was no smaller in counties with more social capital. Across ZIP Code Tabulation Areas, modeled loneliness is lower where more residents are married, a correlation of −0.65 (95% CI −0.69 to −0.62). The model that produces those estimates already uses age, so part of that gradient is built in.

The results describe three different units. County results describe places, the Gallup results describe people in unweighted interviews from 2009 to 2017, and the loneliness estimates describe ZCTAs, the Census Bureau's approximation of ZIP codes. A pattern in one unit is not a finding about another, and the chapter keeps them apart.

### Participation falls with county deprivation

Counties are ranked on the Area Deprivation Index, where each county takes the mean national percentile rank of its tracts weighted by tract population, and are cut into ten groups with equal numbers of the 3,128 counties that have a rank. County Health Rankings supplies two measures of showing up. Voter turnout is the share of adult citizens who voted in the 2020 presidential election, and census self-response is the share of all households that answered the 2020 census online, by mail or by phone without a visit from a census taker.[1] The point estimates of both are lower at every step from the least to the most deprived tenth (Table 1). Between the two extreme tenths, turnout is 15.6 (95% CI 13.9 to 17.7) points lower and self-response 21.0 (95% CI 19.0 to 22.9) points lower, so the self-response gradient is the steeper of the two.

Table 1. Voter turnout and census self-response fall with county deprivation, and across counties they correlate less with upward mobility than family structure and friendship across incomes do

These are means weighted by county population. Counting each county once puts turnout at 76.9% in the least deprived tenth and 58.8% in the most deprived, and self-response at 65.0% and 46.3%. In every tenth the unweighted turnout is higher than the weighted one and the unweighted self-response lower, so small counties vote more and respond less on their own than the counties where most people live. The gradient points the same way under either weighting.

Neither measure is a pure gauge of civic willingness. Turnout counts only citizens, depends on state voting laws, and leaves out people who have lost their voting rights.[1] Self-response depends on how a place was counted. Census takers counted households in person instead of inviting them to respond in remote Alaska, in remote parts of northern Maine and southeast Alaska, and optionally in tribal areas, so a low rate there reflects the operation as well as the residents.[2]

Table 2 maps self-response by county. With counties weighted by population, the mean is 66.0%, the fifth percentile 51.8% and the ninety-fifth percentile 77.5%. State averages are highest in Minnesota, Wisconsin and Michigan and lowest in Alaska, West Virginia and Maine, and the four lowest county values are all in Alaska.

Table 2. Census self-response in 2020 was highest in Minnesota, Wisconsin and Michigan and lowest in Alaska, West Virginia and Maine

### Voting tracks mobility less than family structure and friendship do

The second tab of Table 1 asks which county measures of civic life go with the upward mobility of children from low-income families. Mobility is the mean adult income rank of children born in 1978 to 1983 to parents at the 25th percentile of the national income distribution, which the Opportunity Atlas reports by county (Chapter 4.3, Race, income and mobility (https://vishalsingh.org/zipcode/race-income-and-mobility) describes it). Each measure is correlated with mobility across counties weighted by their 2018 population, with intervals from a bootstrap over counties.

Voter turnout correlates +0.35 (95% CI 0.28 to 0.42) with mobility and census self-response +0.32 (95% CI 0.26 to 0.37). The organization measures sit lower. Civic groups per capita, from the Social Capital Atlas, correlate +0.06 (95% CI −0.01 to 0.14), membership associations −0.05 (95% CI −0.12 to 0.02) and religious congregations −0.10 (95% CI −0.15 to −0.04), while membership in volunteering groups reaches +0.19 (95% CI 0.12 to 0.25). Putnam's account of declining social capital gave associations a central place,[3] yet across counties the density of associations says little about where children from low-income families rise, and congregations run the other way. The measures come from different sources and years (Facebook groups, business records and a 2010 religion census), which limits what a gap between two of them can show.

The Joint Economic Committee's county Social Capital Index has four parts. Family unity uses births to unmarried mothers, children in single-parent families and the share of women in midlife who are married. Community health uses nonprofits, congregations and a state-level measure of civic engagement. Institutional health uses presidential voting, census mail response and state-level confidence in institutions. Collective efficacy is measured at the county level by violent-crime rates.[4]

Family unity has the highest correlation with mobility of any measure, +0.71 (95% CI 0.67 to 0.74). Economic connectedness, the measure of friendships across income lines that Chapter 4.3, Race, income and mobility (https://vishalsingh.org/zipcode/race-income-and-mobility) examines,[5] is +0.69 (95% CI 0.64 to 0.74), with an overlapping interval. The overall index is +0.61 (95% CI 0.53 to 0.68), collective efficacy (low violent crime) +0.51 (95% CI 0.43 to 0.58), institutional health +0.24 (95% CI 0.16 to 0.33) and community health +0.19 (95% CI 0.10 to 0.28). Of the four parts, the two that describe families and safety sit well above the two that describe participation and associations.

 (Note: Collective efficacy in the county index is a violent-crime rate, so its correlation with mobility is a statement about crime and not about residents acting together.)

Counties with high turnout also differ in income, racial composition and poverty, so Table 3 enters those characteristics. With the four civic measures alone in the model, turnout has a standardized coefficient of 0.24 (95% CI 0.06 to 0.42) on mobility. With the log of median household income, the Black share of residents and the poverty rate added, it falls to 0.00 (95% CI −0.15 to 0.16) and the coefficient on census self-response is −0.08 (95% CI −0.21 to 0.05).

Once economic connectedness is added, its own coefficient is 0.63 (95% CI 0.47 to 0.79), the largest in the model, while the coefficients on census self-response and volunteering become negative, −0.15 (95% CI −0.26 to −0.05) and −0.16 (95% CI −0.22 to −0.09). These are partial associations among measures that move together, and the change of sign says that each coefficient depends on which others share the model.

Table 3. Once income, race and poverty are in the model, voter turnout has a coefficient of 0.00 on county mobility, and economic connectedness has 0.63

The measures also have different dates. Mobility is for children born in 1978 to 1983, the participation measures are for 2020, the Facebook measures for 2022 and the Joint Economic Committee components for 2010 to 2016. Using recent civic life to describe where children rose decades ago assumes that county differences persist, which these data do not test.

### The partner premium in the ladder score

The Gallup U.S. Daily survey has 2,189,201 interviews from 2009 to 2017, in 3,140 counties, with both a marital status and a score on the Cantril ladder, which asks people to place their life on a ladder from the worst possible life to the best possible one. The item was asked of nearly all respondents in 2010 to 2012, of 85% in 2009 and of 49% from 2013 to 2017, so 61% of these interviews are from 2009 to 2012. Every Gallup figure here is unweighted, because the extract has no survey weights. Intervals are clustered by the respondent's county, so they describe sampling variation in these interviews and not national prevalence.

Adults who were married or in a domestic partnership placed their lives at 7.28 (95% CI 7.27 to 7.29) on average, and adults who were never married, separated, divorced or widowed placed theirs at 6.78 (95% CI 6.77 to 6.79). The never married (6.77) and the separated, divorced or widowed (6.78) score the same, so the chapter compares the partnered with everyone else. The difference, 0.50 (95% CI 0.49 to 0.51) ladder points, is the partner premium.

The 2008 interviews are left out. In 2008 Gallup asked political questions immediately before the ladder, which lowers the ladder score, and it removed them from part of the sample in January 2009 and added a transition question in April 2009.[6] Including 2008 would put the group means at 7.22 and 6.73 and leave the premium at 0.50 (95% CI 0.49 to 0.51). Leaving out 2009 as well gives 0.50 (95% CI 0.49 to 0.51).

The premium is not an artifact of where partnered people live. Comparing each respondent only with others in the same county (county fixed effects) gives 0.50 (95% CI 0.49 to 0.50), the same as the national difference. It is larger once age and sex are taken into account, 0.64 (95% CI 0.63 to 0.65) within county and year with a fixed effect for each year of age and for sex, so the unadjusted difference is lowered by the age mix of the two groups. A fixed effect for each year of age is used because the ladder is U-shaped in age and the premium varies with it, and a single linear age term gives a lower premium. Among the interviews that report household income and education, adding both lowers the premium to 0.31 (95% CI 0.30 to 0.32). Household income includes a partner's earnings, so partnered adults are more likely to live in households with higher incomes, and this model compares adults within the same income and education group.

By age band, with the same adjustment inside each band, the premium is smallest in the youngest band (0.15 (95% CI 0.12 to 0.18)) and next smallest in the oldest (0.35 (95% CI 0.33 to 0.37)), and it is largest in the middle of adulthood, 0.80 (95% CI 0.78 to 0.82) at its highest.

Table 4. The partner premium is 0.50 ladder points overall and 0.50 within counties, and runs from 0.15 in the youngest age band to 0.80 in its largest band

> Study evidence: Whether partnership moves with the ladder score (Longitudinal panel following the same people over time)
>
> The interviews compare different people, and happier people may be more likely to marry and to stay married. Two studies follow the same people over time in the German Socio-Economic Panel. Stutzer and Frey compared the life satisfaction of single people who later married with that of people who stayed single, and found that happier singles were more likely to marry, so part of a cross-sectional difference reflects who marries.[7] Lucas and colleagues followed a large panel through changes in marital status and found that, on average, life satisfaction reacted to marriage and then moved back toward its earlier level, with large differences between individuals.[8]
>
> Neither study randomizes marriage. Lucas and colleagues compare people with themselves over time, which removes differences that do not change. Stutzer and Frey also compare people who later married with people who stayed single, which is a comparison between persons before marriage. Neither speaks to the Gallup interviews, which have no before and after.

### The premium does not shrink where county social capital is higher

If community could stand in for a partner, the premium would be smaller in counties with more of it. Table 5 sorts interviews into ten groups by the Joint Economic Committee's Social Capital Index of the respondent's county, a standardized score from the 2018 release whose components date from 2010 to 2016, and compares the partnered with everyone else in each group. The premium was 0.45 (95% CI 0.43 to 0.48) ladder points in the lowest group and 0.49 (95% CI 0.47 to 0.51) in the highest, with a largest value of 0.53 (95% CI 0.51 to 0.55) in one of the upper groups, so it did not shrink. The point estimates are smaller in the three lowest groups than in the other seven, although some intervals overlap.

The mean ladder score of the partnered is 7.25 (95% CI 7.21 to 7.30) in the lowest group and 7.35 (95% CI 7.33 to 7.37) in the highest, and that of the others 6.80 (95% CI 6.75 to 6.84) and 6.86 (95% CI 6.83 to 6.89).

Table 5. The partner premium does not shrink where county social capital is higher: 0.45 ladder points in the lowest decile and 0.49 in the highest

A regression with an interaction between partnership and the county index puts the premium 0.023 (95% CI 0.012 to 0.034) ladder points higher for each standard deviation of the index. The same coefficient has a positive point estimate for the index's four parts and for the Social Capital Atlas measures of civic groups and volunteering, from 0.007 (95% CI −0.005 to 0.019) to 0.035 (95% CI 0.019 to 0.051). These intervals cluster on the respondent's state, because the community-health and institutional-health subindexes include state-level inputs. Of the seven intervals, 3 include zero, those for community health, institutional health and civic groups. The data support "the premium does not shrink" and do not establish that it grows. The index includes family structure, which moves with the share of partnered adults in a county, and the Atlas measures do not, so they serve as a check on that overlap.

The comparison says that the premium is not smaller in places with more social capital by these measures. It does not say that community adds to a partner's value, because counties differ in age, income and other ways that this comparison does not hold fixed, and the county measures describe places while the ladder scores describe people.

### Modeled loneliness follows who lives in a ZCTA

CDC PLACES publishes modeled estimates of the share of adults who say they feel lonely always, usually or sometimes for each ZCTA. The measure comes from an optional module of the 2023 Behavioral Risk Factor Surveillance System that only some states asked. The estimates cover 39 states, those in which at least 90% of ZCTAs have an estimate, and ZCTAs that hold 75% of the residents of ZCTAs in the PLACES file.[9] The values are modeled estimates: a multilevel regression predicts a respondent's answer from age, sex, race and ethnicity and education, with the county poverty rate and state and county effects, and the prediction is applied to the 2020 census population of each ZCTA.[10][11] The county input is the share of adults below 150% of the federal poverty level. Marital status and population density are not inputs. A ZCTA estimate therefore reports who lives there, as the survey answers by age, sex, race and education and county poverty imply. A gradient against a characteristic that moves with those inputs, including age and the poverty rate, is partly built in.

Across the 19,684 ZCTAs with an estimate and at least 500 residents, weighted by adult population, modeled loneliness averages 34.4%; 68 of those ZCTAs are assigned to other states. It is lower where more adults are married, a correlation of −0.65 (95% CI −0.69 to −0.62).

Modeled loneliness is also lower where the population is older, −0.53 (95% CI −0.57 to −0.49) with median age and −0.48 (95% CI −0.51 to −0.45) with the share of older residents (Table 6). It is higher where the poverty rate is higher, +0.56 (95% CI 0.51 to 0.62), which is partly built in through the county poverty input. It is higher where population density is higher, +0.40 (95% CI 0.30 to 0.47) in logs, and lower where median household income is higher, −0.38 (95% CI −0.47 to −0.31) in logs.

Counting each ZCTA once gives −0.63 (95% CI −0.65 to −0.61) for the married share and +0.50 (95% CI 0.47 to 0.53) for poverty. The residents floor matters for the unweighted figures. With no floor, the weighted correlation with the married share is −0.65 (95% CI −0.69 to −0.62) and the unweighted one −0.54 (95% CI −0.57 to −0.52).

The Area Deprivation Index is the exception. Its correlation is +0.02 (95% CI −0.13 to 0.20) when weighted and +0.16 (95% CI 0.03 to 0.27) when each ZCTA counts once, so the book's deprivation measure says little about modeled loneliness, although the poverty rate says a good deal. The ZCTA values of the index are an area-weighted approximation, and the index understates deprivation in high-cost cities.

Table 6. Modeled loneliness is lower in ZCTAs with more married adults, a correlation of −0.65, and higher where poverty is higher, +0.56

Sorted into ten groups by married share, ZCTAs run from a mean of 28.2% to 67.4%, and modeled loneliness falls from 38.8% (95% CI 38.2% to 39.4%) to 29.9% (95% CI 29.6% to 30.3%).

Sorted by median age, from 30.2 years to 57.8 years, it falls from 37.8% (95% CI 37.1% to 38.6%) to 29.6% (95% CI 29.1% to 30.0%) (Table 7). The two gradients overlap, because ZCTAs with more married adults are older, and age is an input to the model.

Table 7. Modeled loneliness is higher in ZCTAs with fewer married adults and a younger population

Density shows that the model does not produce one gradient for every outcome. Modeled loneliness is 32.8% (95% CI 32.6% to 33.0%) in ZCTAs with fewer than 100 people per square mile and 37.8% (95% CI 36.6% to 38.9%) in those with 5,000 or more.

Across the same density groups, modeled depression falls from 23.6% (95% CI 23.4% to 23.7%) to 19.7% (95% CI 18.9% to 20.7%) (Table 8). Depression here is the share of adults ever told by a health professional that they have a depressive disorder, which depends in part on access to care, and both measures come from the same model inputs.

Table 8. Modeled loneliness rises from 32.8% to 37.8% across density classes of ZCTAs while modeled depression falls from 23.6% to 19.7%

A regression of ZCTA loneliness on median age, the share of older residents, race and ethnicity shares and the share with a bachelor's degree, weighted by adults, shows how much of the pattern lines up with composition. It has an R-squared of 65% (95% CI 62% to 69%), and adding the married share raises it by 5.2 (95% CI 3.7 to 6.8) points of R-squared. Most of the variation in the estimates lines up with the characteristics the model works from, and the married share adds little beyond them.

The ladder scores in the Gallup survey and the modeled loneliness in PLACES are different measures at different grains. The first is how a person rates their life, estimated from interviews, and the second is the modeled share of a ZCTA's adults who feel lonely. Partnered adults rate their lives higher and ZCTAs with fewer married adults have higher modeled loneliness, but the first statement is about people and the second about places built from a model, and one does not stand in for the other. Chapter 6.3, Life evaluation and daily feelings (https://vishalsingh.org/zipcode/life-evaluation-and-daily-feelings) covers the ladder and daily feelings in full.

### Voter turnout falls with deprivation under all four rankings

Table 9 ranks the 3,096 counties that have a 2020 turnout and all four rankings in four ways, with the tenths cut within that common sample. Ranked by the ADI, turnout among adult citizens is 72.0% (95% CI 70.4% to 73.6%) in the least deprived tenth and 56.5% (95% CI 55.4% to 57.7%) in the most deprived, and the rank correlation between tenth and turnout is −1.00. Under the standardized index the two ends are 75.9% (95% CI 74.9% to 77.0%) and 55.7% (95% CI 54.3% to 57.3%), with a correlation of −0.98. Under median household income they are 73.7% (95% CI 72.1% to 75.2%) and 57.9% (95% CI 56.9% to 58.8%), with a correlation of −0.99, and under the bachelor's degree share they are 73.8% (95% CI 72.5% to 75.2%) and 56.4% (95% CI 55.4% to 57.3%), with a correlation of −0.98, so none of the three counts as a change under the book's rule (Chapter 9.1, How far one ranking goes (https://vishalsingh.org/zipcode/how-far-one-ranking-goes)).

Table 9. Voter turnout is steadily worse toward the more deprived end under the ADI and under each of the three other rankings

### What these data cannot show

The county and ZCTA patterns are ecological. A county with low turnout need not be one where the same people stayed home and had fewer chances to rise, and a ZCTA with high modeled loneliness need not hold lonelier people than its neighbors. Deaths are outside this chapter (Chapter 8.2, Overdose, alcohol and suicide (https://vishalsingh.org/zipcode/overdose-alcohol-and-suicide) maps overdose, alcohol and suicide deaths).

The county measures are dated 2010 to 2022 and mobility refers to children born in 1978 to 1983, so the comparisons assume that county differences persist. Census self-response reflects the operation as well as the residents in places counted in person, and turnout is limited to citizens and shaped by state law. The Joint Economic Committee index is a compiled snapshot of inputs from different years, and the Atlas measures describe active Facebook users in 2022, not a census of ties. The regressions are partial associations among correlated measures, with errors clustered on state.

The Gallup partner premium is a comparison between people at one time, in a repeated cross-section and not a panel. Selection into partnership is the main limit, and the panel studies above find evidence of it. Partnered means married or in a domestic partnership, so adults who live with a partner and call themselves single are in the comparison group. County comes from the respondent's ZIP code and is approximate near county lines. The extract is unweighted, so the levels can differ from the weighted figures Gallup publishes, and the evaluative ladder is the only Gallup measure used here.

The PLACES loneliness estimates are modeled, exist only for the states that asked the question, and do not include marital status as an input. The ZCTA characteristics are American Community Survey estimates for 2019 to 2023 for each ZCTA, without margins of error, so small ZCTAs carry sampling error that the intervals do not show. The married share counts residents aged fifteen and older who are married and not separated. An earlier version took these characteristics from a table of undocumented year. ZCTAs are not ZIP codes. The intervals in the figures describe the bootstrap over counties and do not carry the model's own uncertainty for each ZCTA. Puerto Rico is not in County Health Rankings and is not drawn, and Connecticut appears under its legacy counties in the county files and drops out of the regressions that need the ACS tract file, which uses planning-region codes.

## Tables

### Table 1. Voter turnout and census self-response fall with county deprivation, and across counties they correlate less with upward mobility than family structure and friendship across incomes do

County Health Rankings 2024 participation measures by decile of county deprivation, and county measures of civic life against the adult income rank of children born 1978 to 1983 to parents at the 25th percentile

#### Table 1.1. Participation by deprivation

Voter turnout fell from 72.1% to 56.4% and census self-response from 68.6% to 47.6% across deciles of county deprivation

Population-weighted mean voter turnout (citizens aged 18 and over, 2020 election) and census self-response (households, 2020 census) by decile of county deprivation

| conf_high | conf_low | decile | estimate | measure | n_units |
| ---: | ---: | ---: | ---: | --- | ---: |
| 73.80 | 70.58 | 1 | 72.06 | Voter turnout, 2020 | 307 |
| 71 | 67.93 | 2 | 69.33 | Voter turnout, 2020 | 304 |
| 68.58 | 64.62 | 3 | 66.59 | Voter turnout, 2020 | 309 |
| 65.57 | 62.07 | 4 | 63.95 | Voter turnout, 2020 | 309 |
| 64.15 | 61.43 | 5 | 62.84 | Voter turnout, 2020 | 310 |
| 62.08 | 59.10 | 6 | 60.63 | Voter turnout, 2020 | 310 |
| 60.94 | 58.84 | 7 | 59.85 | Voter turnout, 2020 | 313 |
| 60.52 | 56.90 | 8 | 58.66 | Voter turnout, 2020 | 315 |
| 59.01 | 57.18 | 9 | 58.06 | Voter turnout, 2020 | 311 |
| 57.59 | 55.37 | 10 | 56.43 | Voter turnout, 2020 | 310 |
| 69.94 | 67.39 | 1 | 68.60 | Census self-response, 2020 | 312 |
| 68.15 | 65.91 | 2 | 66.97 | Census self-response, 2020 | 312 |
| 68.04 | 65 | 3 | 66.45 | Census self-response, 2020 | 312 |
| 66.25 | 63.61 | 4 | 64.94 | Census self-response, 2020 | 312 |
| 64.80 | 62.56 | 5 | 63.80 | Census self-response, 2020 | 314 |
| 62.88 | 60.24 | 6 | 61.63 | Census self-response, 2020 | 311 |
| 61.85 | 59.72 | 7 | 60.76 | Census self-response, 2020 | 313 |
| 58.76 | 55.75 | 8 | 57.15 | Census self-response, 2020 | 315 |
| 55.21 | 53.18 | 9 | 54.20 | Census self-response, 2020 | 312 |
| 49.12 | 46.11 | 10 | 47.63 | Census self-response, 2020 | 312 |

#### Table 1.2. Correlation with mobility

Across counties, family unity and economic connectedness correlate about +0.71 and +0.69 with upward mobility, voter turnout +0.35 and civic groups +0.06

Population-weighted correlation of each county measure with the mean adult income rank of children from low-income families, county measures dated 2010 to 2022

| conf_high | conf_low | estimate | group | measure | n_units |
| ---: | ---: | ---: | --- | --- | ---: |
| 0.744 | 0.666 | 0.706 | JEC index and its parts | JEC family unity | 2,913 |
| 0.741 | 0.644 | 0.693 | Contact across incomes | Economic connectedness | 2,942 |
| 0.676 | 0.535 | 0.608 | JEC index and its parts | JEC Social Capital Index | 2,889 |
| 0.584 | 0.431 | 0.514 | JEC index and its parts | JEC collective efficacy (low violent crime) | 2,885 |
| 0.417 | 0.283 | 0.354 | Voting and census response | Voter turnout, 2020 | 2,964 |
| 0.368 | 0.258 | 0.316 | Voting and census response | Census self-response, 2020 | 2,982 |
| 0.327 | 0.163 | 0.244 | JEC index and its parts | JEC institutional health | 2,959 |
| 0.253 | 0.121 | 0.187 | Organizations and membership | Volunteering group membership | 2,983 |
| 0.278 | 0.096 | 0.186 | JEC index and its parts | JEC community health | 2,983 |
| 0.138 | −0.011 | 0.062 | Organizations and membership | Civic groups per capita | 2,983 |
| 0.020 | −0.122 | −0.052 | Organizations and membership | Membership associations per 10,000 residents | 2,982 |
| −0.036 | −0.155 | −0.097 | Organizations and membership | Religious congregations per 1,000 residents | 2,983 |

Note: Deciles hold equal numbers of counties. County deprivation is the mean of the tract Area Deprivation Index national ranks (1 = least deprived), weighted by tract population. Turnout counts only citizens and depends on state voting laws; self-response is low where census takers counted households in person.

Note: Across all the counties in the first tab, population-weighted voter turnout is 67.8% and census self-response is 66.0%.

Note: Intervals resample whole counties. Resampling whole states gives wider intervals for every point, and the decile intervals still separate the first from the last decile.

Note: Counties with a missing measure drop out of that point only, so the number of counties differs by measure. The JEC collective-efficacy subindex is measured at the county level by violent-crime rates, and the JEC institutional-health and community-health subindexes include state-level inputs.

Source: County Health Rankings 2024 (voter turnout and census participation, 2020), Area Deprivation Index 2023 and ACS 2019-2023 county population, population-weighted; second tab adds Social Capital Atlas 2022, JEC Social Capital Index 2018 release and Opportunity Atlas county outcomes (1978-1983 birth cohorts), population-weighted by 2018 county population

Data: https://vishalsingh.org/zipcode/report/civic-life-loneliness-and-partnership/figures/civic_gradient.json

### Table 2. Census self-response in 2020 was highest in Minnesota, Wisconsin and Michigan and lowest in Alaska, West Virginia and Maine

Percent of households that answered the 2020 census online, by mail or by phone without a visit from a census taker, by county

No inline rows: the values of Table 2 are in https://vishalsingh.org/zipcode/data/civic-life-loneliness-and-partnership/census_response_map.json.

Note: Each gray county has its reason in the tooltip. Puerto Rico is not in County Health Rankings and is not drawn. Connecticut appears as its eight legacy counties.

Note: Census takers counted households in person instead of inviting them to respond in remote Alaska, in remote parts of northern Maine and southeast Alaska, and optionally in tribal areas, so a low self-response rate there reflects how the count was run as well as the residents.

Note: The tooltip also gives voter turnout in the 2020 presidential election, which depends on state voting laws, counts only citizens and divides ballots by an ACS estimate of adult citizens.

Source: County Health Rankings 2024 release, census participation (2020 census; full count of households) and voter turnout (2020; ballots over an ACS 5-year estimate of adult citizens)

Data: https://vishalsingh.org/zipcode/report/civic-life-loneliness-and-partnership/figures/census_response_map.json

### Table 3. Once income, race and poverty are in the model, voter turnout has a coefficient of 0.00 on county mobility, and economic connectedness has 0.63

Standardized coefficients on the adult income rank of children from low-income families (born 1978 to 1983) in three county regressions, weighted by 2018 county population

| conf_high | conf_low | estimate | model | n_units | term |
| ---: | ---: | ---: | --- | ---: | --- |
| 0.420 | 0.058 | 0.239 | Civic measures only | 2,915 | Voter turnout |
| 0.327 | −0.007 | 0.160 | Civic measures only | 2,915 | Census self-response |
| 0.127 | −0.150 | −0.012 | Civic measures only | 2,915 | Civic groups per capita |
| 0.253 | −0.016 | 0.118 | Civic measures only | 2,915 | Volunteering group membership |
| 0.161 | −0.152 | 0.004 | Plus income, percent Black and poverty | 2,915 | Voter turnout |
| 0.047 | −0.212 | −0.083 | Plus income, percent Black and poverty | 2,915 | Census self-response |
| 0.160 | −0.103 | 0.028 | Plus income, percent Black and poverty | 2,915 | Civic groups per capita |
| 0.091 | −0.061 | 0.015 | Plus income, percent Black and poverty | 2,915 | Volunteering group membership |
| 0.593 | 0.238 | 0.415 | Plus income, percent Black and poverty | 2,915 | Log median household income |
| −0.425 | −0.608 | −0.517 | Plus income, percent Black and poverty | 2,915 | Percent Black |
| 0.076 | −0.158 | −0.041 | Plus income, percent Black and poverty | 2,915 | Poverty rate |
| 0.101 | −0.121 | −0.010 | Plus economic connectedness | 2,915 | Voter turnout |
| −0.053 | −0.256 | −0.155 | Plus economic connectedness | 2,915 | Census self-response |
| 0.049 | −0.146 | −0.049 | Plus economic connectedness | 2,915 | Civic groups per capita |
| −0.094 | −0.220 | −0.157 | Plus economic connectedness | 2,915 | Volunteering group membership |
| 0.790 | 0.467 | 0.628 | Plus economic connectedness | 2,915 | Economic connectedness |
| 0.323 | −0.053 | 0.135 | Plus economic connectedness | 2,915 | Log median household income |
| −0.339 | −0.480 | −0.409 | Plus economic connectedness | 2,915 | Percent Black |
| 0.145 | −0.074 | 0.036 | Plus economic connectedness | 2,915 | Poverty rate |

Note: Every variable is a z-score using the Atlas's 2018 county population as the weight. The controls are the log of median household income, the percent Black and the percent below poverty from the ACS 2019-2023 tract file, averaged to counties; Connecticut drops out because the ACS tract file uses planning-region codes.

Note: Intervals use t distributions on the state clusters. These are partial associations among correlated county measures, and a coefficient can change sign when its neighbors change.

Source: County Health Rankings 2024, Social Capital Atlas 2022, Opportunity Atlas county outcomes (1978-1983 birth cohorts) and ACS 2019-2023 county aggregates; population-weighted by 2018 county population

Data: https://vishalsingh.org/zipcode/report/civic-life-loneliness-and-partnership/figures/participation_and_connectedness.json

### Table 4. The partner premium is 0.50 ladder points overall and 0.50 within counties, and runs from 0.15 in the youngest age band to 0.80 in its largest band

Difference in mean Cantril ladder score between adults who are married or in a domestic partnership and all others, under successive adjustments and by age band, Gallup U.S. Daily 2009 to 2017

| conf_high | conf_low | estimate | group | label | n |
| ---: | ---: | ---: | --- | --- | ---: |
| 0.506 | 0.489 | 0.498 | All adults | Difference of means, no controls | 2,189,201 |
| 0.504 | 0.488 | 0.496 | All adults | Within county | 2,189,199 |
| 0.650 | 0.630 | 0.640 | All adults | Within county and year, adjusted for single-year age and sex | 2,163,354 |
| 0.318 | 0.301 | 0.310 | All adults | Same, adding household income and education | 1,786,303 |
| 0.183 | 0.121 | 0.152 | By age band, adjusted for single-year age and sex | Ages 18 to 24 | 147,891 |
| 0.632 | 0.586 | 0.609 | By age band, adjusted for single-year age and sex | Ages 25 to 34 | 226,935 |
| 0.809 | 0.762 | 0.785 | By age band, adjusted for single-year age and sex | Ages 35 to 44 | 268,006 |
| 0.817 | 0.779 | 0.798 | By age band, adjusted for single-year age and sex | Ages 45 to 54 | 377,207 |
| 0.787 | 0.758 | 0.772 | By age band, adjusted for single-year age and sex | Ages 55 to 64 | 461,305 |
| 0.609 | 0.579 | 0.594 | By age band, adjusted for single-year age and sex | Ages 65 to 74 | 394,603 |
| 0.368 | 0.333 | 0.350 | By age band, adjusted for single-year age and sex | Ages 75 and over | 286,964 |

Note: The last all-adults row uses the interviews that report household income and education, a smaller sample than the other rows. Household income includes a partner's earnings. The rows are statistical adjustments, and none identifies an effect of partnership; the age-band rows are separate regressions with county, year, single-year age and sex fixed effects. A linear age term fits poorly because the ladder is U-shaped in age, and it gives a lower premium.

Note: Partnered means married or in a domestic partnership, which is how Gallup's three-way marital status counts domestic partners. Adults who live with a partner and call themselves single are in the other group.

Note: Gallup's extract has no survey weights, so every estimate is unweighted. County comes from the respondent's ZIP code and is approximate near county lines.

Source: Gallup U.S. Daily, 2009-2017; unweighted, intervals clustered by county

Data: https://vishalsingh.org/zipcode/report/civic-life-loneliness-and-partnership/figures/partner_premium_specifications.json

### Table 5. The partner premium does not shrink where county social capital is higher: 0.45 ladder points in the lowest decile and 0.49 in the highest

Cantril ladder scores of partnered and other adults by decile of the JEC Social Capital Index of the respondent's county, and the change in the premium per standard deviation of seven county measures

#### Table 5.1. Ladder scores

Ladder scores differ little between the lowest and highest decile of county social capital, 7.25 and 7.35 among partnered adults and 6.80 and 6.86 among others

Unweighted mean Cantril ladder score by decile of the JEC Social Capital Index of the respondent's county, Gallup U.S. Daily interviews 2009 to 2017

| conf_high | conf_low | decile | estimate | group | n |
| ---: | ---: | ---: | ---: | --- | ---: |
| 7.30 | 7.21 | 1 | 7.25 | Married or domestic partner | 112,277 |
| 6.84 | 6.75 | 1 | 6.80 | Single, separated, divorced or widowed | 106,246 |
| 7.30 | 7.23 | 2 | 7.26 | Married or domestic partner | 119,477 |
| 6.85 | 6.77 | 2 | 6.81 | Single, separated, divorced or widowed | 97,396 |
| 7.30 | 7.24 | 3 | 7.27 | Married or domestic partner | 117,601 |
| 6.85 | 6.78 | 3 | 6.81 | Single, separated, divorced or widowed | 101,443 |
| 7.28 | 7.22 | 4 | 7.25 | Married or domestic partner | 125,048 |
| 6.77 | 6.71 | 4 | 6.74 | Single, separated, divorced or widowed | 92,165 |
| 7.29 | 7.23 | 5 | 7.26 | Married or domestic partner | 126,801 |
| 6.78 | 6.71 | 5 | 6.74 | Single, separated, divorced or widowed | 91,539 |
| 7.28 | 7.22 | 6 | 7.25 | Married or domestic partner | 131,967 |
| 6.79 | 6.70 | 6 | 6.75 | Single, separated, divorced or widowed | 90,649 |
| 7.29 | 7.23 | 7 | 7.26 | Married or domestic partner | 129,786 |
| 6.77 | 6.70 | 7 | 6.74 | Single, separated, divorced or widowed | 83,949 |
| 7.30 | 7.25 | 8 | 7.28 | Married or domestic partner | 134,032 |
| 6.79 | 6.73 | 8 | 6.76 | Single, separated, divorced or widowed | 81,701 |
| 7.35 | 7.29 | 9 | 7.32 | Married or domestic partner | 135,115 |
| 6.83 | 6.76 | 9 | 6.79 | Single, separated, divorced or widowed | 82,489 |
| 7.37 | 7.33 | 10 | 7.35 | Married or domestic partner | 141,007 |
| 6.89 | 6.83 | 10 | 6.86 | Single, separated, divorced or widowed | 79,202 |

#### Table 5.2. Partner premium

The partner premium is 0.45 ladder points in the lowest decile of county social capital and 0.49 in the highest

Unweighted difference in mean ladder score, partnered minus other adults, by decile of the JEC Social Capital Index of the respondent's county, 2009 to 2017

| conf_high | conf_low | decile | estimate | n |
| ---: | ---: | ---: | ---: | ---: |
| 0.479 | 0.427 | 1 | 0.453 | 218,523 |
| 0.486 | 0.423 | 2 | 0.455 | 216,873 |
| 0.495 | 0.431 | 3 | 0.463 | 219,044 |
| 0.534 | 0.492 | 4 | 0.513 | 217,213 |
| 0.536 | 0.495 | 5 | 0.515 | 218,340 |
| 0.525 | 0.483 | 6 | 0.504 | 222,616 |
| 0.546 | 0.507 | 7 | 0.527 | 213,735 |
| 0.536 | 0.492 | 8 | 0.514 | 215,733 |
| 0.548 | 0.504 | 9 | 0.526 | 217,604 |
| 0.511 | 0.472 | 10 | 0.491 | 220,209 |

#### Table 5.3. Change per standard deviation

The partner premium is 0.007 to 0.035 ladder points larger for each standard deviation of the seven county community measures

Change in the partner premium (ladder points) per standard deviation of each county measure, interaction coefficients from regressions of the ladder score, 2009 to 2017

| conf_high | conf_low | estimate | measure | n | source |
| ---: | ---: | ---: | --- | ---: | --- |
| 0.034 | 0.012 | 0.023 | JEC Social Capital Index | 2,179,890 | JEC Social Capital Project |
| 0.039 | 0.018 | 0.028 | JEC family unity | 2,182,603 | JEC Social Capital Project |
| 0.033 | −0.010 | 0.012 | JEC community health | 2,188,939 | JEC Social Capital Project |
| 0.019 | −0.005 | 0.007 | JEC institutional health | 2,186,233 | JEC Social Capital Project |
| 0.021 | 0.007 | 0.014 | JEC collective efficacy (low violent crime) | 2,174,228 | JEC Social Capital Project |
| 0.039 | −0.002 | 0.018 | Civic groups per capita | 2,185,410 | Social Capital Atlas |
| 0.051 | 0.019 | 0.035 | Volunteering group membership | 2,185,410 | Social Capital Atlas |

Note: Gallup's extract has no survey weights, so every estimate is unweighted and levels can differ from the weighted figures Gallup publishes. County comes from the respondent's ZIP code and is approximate near county lines. The ladder is the evaluative measure, and daily feelings are not used here.

Note: Deciles hold about a tenth of interviews each, and a county sits in one decile. The JEC index combines family unity, community health, institutional health and collective efficacy, and family unity includes the share of women who are married, so the Atlas measures serve as a check that does not depend on marriage.

Note: The premium in the second tab is largest in one of the upper deciles; the highest of the ten estimates is selected for being the highest.

Note: The third tab clusters errors on the respondent's state, because the JEC community-health and institutional-health subindexes include state-level inputs; the first two tabs cluster on county.

Source: Gallup U.S. Daily, 2009-2017, unweighted, intervals clustered by county (by state in the third tab); JEC Social Capital Index 2018 release (components 2010-2016) and Social Capital Atlas 2022 joined on the county of the interview

Data: https://vishalsingh.org/zipcode/report/civic-life-loneliness-and-partnership/figures/partner_premium_by_social_capital.json

### Table 6. Modeled loneliness is lower in ZCTAs with more married adults, a correlation of −0.65, and higher where poverty is higher, +0.56

Correlation of CDC PLACES modeled loneliness (percent of adults, BRFSS 2023) with seven ZCTA characteristics, ZCTAs with an estimate and at least 500 residents

| conf_high | conf_low | estimate | group | label | n_units |
| ---: | ---: | ---: | --- | --- | ---: |
| −0.623 | −0.695 | −0.653 | Weighted by adult population | Married share, residents 15 and older | 19,681 |
| −0.610 | −0.649 | −0.630 | Each ZCTA counted once | Married share, residents 15 and older | 19,681 |
| −0.495 | −0.573 | −0.530 | Weighted by adult population | Median age | 19,679 |
| −0.468 | −0.512 | −0.490 | Each ZCTA counted once | Median age | 19,679 |
| −0.446 | −0.509 | −0.476 | Weighted by adult population | Share aged 65 and over | 19,681 |
| −0.409 | −0.459 | −0.435 | Each ZCTA counted once | Share aged 65 and over | 19,681 |
| 0.474 | 0.303 | 0.396 | Weighted by adult population | Population density (log) | 19,681 |
| 0.410 | 0.271 | 0.341 | Each ZCTA counted once | Population density (log) | 19,681 |
| −0.312 | −0.474 | −0.384 | Weighted by adult population | Median household income (log) | 19,335 |
| −0.303 | −0.425 | −0.365 | Each ZCTA counted once | Median household income (log) | 19,335 |
| 0.616 | 0.514 | 0.557 | Weighted by adult population | Poverty rate | 19,588 |
| 0.527 | 0.468 | 0.496 | Each ZCTA counted once | Poverty rate | 19,588 |
| 0.201 | −0.128 | 0.025 | Weighted by adult population | Area Deprivation Index | 19,622 |
| 0.266 | 0.034 | 0.156 | Each ZCTA counted once | Area Deprivation Index | 19,622 |

Note: PLACES estimates come from a model fed by age, sex, race and ethnicity, education and the county share of adults below a multiple of the poverty line, so a gradient against characteristics that move with those inputs, including age, poverty and income, is partly built in. Marital status and population density are not inputs.

Note: The ZCTA characteristics are ACS 2019-2023 five-year estimates for each ZCTA; the married share counts residents 15 and older who are now married, except separated. ZCTAs are not ZIP codes.

Note: The Area Deprivation Index at ZCTA grain is an area-weighted approximation and understates deprivation in high-cost cities.

Source: CDC PLACES 2025 release (BRFSS 2023), ZCTA modeled estimates, with ACS 2019-2023 ZCTA characteristics and the Area Deprivation Index 2023; modeled estimates, population-weighted by adults

Data: https://vishalsingh.org/zipcode/report/civic-life-loneliness-and-partnership/figures/loneliness_correlates.json

### Table 7. Modeled loneliness is higher in ZCTAs with fewer married adults and a younger population

CDC PLACES modeled loneliness (percent of adults, BRFSS 2023) by decile of ZCTA married share and of ZCTA median age, population-weighted

#### Table 7.1. By married share

Modeled loneliness falls from 38.8% in the tenth of ZCTAs with the lowest married share to 29.9% in the tenth with the highest

Population-weighted mean of CDC PLACES modeled loneliness by decile of ZCTA married share, ZCTAs with an estimate and at least 500 residents

| conf_high | conf_low | decile | estimate | n_units | x |
| ---: | ---: | ---: | ---: | ---: | ---: |
| 39.44 | 38.15 | 1 | 38.79 | 1,968 | 28.22 |
| 37.62 | 35.75 | 2 | 36.60 | 1,968 | 39.29 |
| 36.04 | 34.55 | 3 | 35.25 | 1,968 | 44.47 |
| 34.75 | 33.77 | 4 | 34.25 | 1,968 | 48.06 |
| 33.84 | 32.92 | 5 | 33.38 | 1,968 | 50.94 |
| 33.04 | 32.23 | 6 | 32.62 | 1,968 | 53.65 |
| 32.60 | 31.78 | 7 | 32.19 | 1,968 | 56.11 |
| 31.80 | 31.04 | 8 | 31.39 | 1,968 | 58.77 |
| 31.41 | 30.65 | 9 | 31.02 | 1,968 | 62.15 |
| 30.30 | 29.59 | 10 | 29.93 | 1,969 | 67.45 |

#### Table 7.2. By median age

Modeled loneliness falls from 37.8% in the tenth of ZCTAs with the lowest median age to 29.6% in the tenth with the highest

Population-weighted mean of CDC PLACES modeled loneliness by decile of ZCTA median age, ZCTAs with an estimate and at least 500 residents

| conf_high | conf_low | decile | estimate | n_units | x |
| ---: | ---: | ---: | ---: | ---: | ---: |
| 38.63 | 37.13 | 1 | 37.84 | 1,966 | 30.23 |
| 37.21 | 35.64 | 2 | 36.39 | 1,926 | 35.06 |
| 35.86 | 34.37 | 3 | 35.09 | 2,009 | 37.35 |
| 34.72 | 33.58 | 4 | 34.13 | 2,011 | 39.19 |
| 33.86 | 32.84 | 5 | 33.33 | 1,904 | 40.78 |
| 33.28 | 32.37 | 6 | 32.82 | 1,960 | 42.37 |
| 33.17 | 31.91 | 7 | 32.48 | 1,963 | 44.15 |
| 32.23 | 31.19 | 8 | 31.68 | 2,004 | 46.44 |
| 31.53 | 30.81 | 9 | 31.16 | 1,969 | 49.93 |
| 30.01 | 29.13 | 10 | 29.56 | 1,967 | 57.80 |

Note: PLACES uses age as an input to the model, so the age gradient is partly built in, and the married-share gradient moves with age. The estimates describe ZCTAs, not the people in them.

Note: Deciles hold equal numbers of ZCTAs. The ZCTA characteristics are ACS 2019-2023 five-year estimates; the married share counts residents 15 and older. ZCTAs are not ZIP codes.

Source: CDC PLACES 2025 release (BRFSS 2023), ZCTA modeled estimates, and ACS 2019-2023 ZCTA characteristics; modeled estimates, population-weighted by adults

Data: https://vishalsingh.org/zipcode/report/civic-life-loneliness-and-partnership/figures/loneliness_by_marriage_and_age.json

### Table 8. Modeled loneliness rises from 32.8% to 37.8% across density classes of ZCTAs while modeled depression falls from 23.6% to 19.7%

Population-weighted mean of two CDC PLACES modeled measures (percent of adults, BRFSS 2023) by population density of the ZCTA, ZCTAs with both estimates and at least 500 residents

| conf_high | conf_low | estimate | label | measure | n_units |
| ---: | ---: | ---: | --- | --- | ---: |
| 32.97 | 32.65 | 32.81 | Under 100 people per square mile | Feeling lonely | 9,893 |
| 23.70 | 23.39 | 23.55 | Under 100 people per square mile | Depression diagnosis | 9,893 |
| 33.38 | 32.84 | 33.11 | 100 to 999 people per square mile | Feeling lonely | 5,023 |
| 23.57 | 23.08 | 23.32 | 100 to 999 people per square mile | Depression diagnosis | 5,023 |
| 34.44 | 33.44 | 33.94 | 1,000 to 4,999 people per square mile | Feeling lonely | 3,262 |
| 22.02 | 20.93 | 21.47 | 1,000 to 4,999 people per square mile | Depression diagnosis | 3,262 |
| 38.90 | 36.56 | 37.75 | 5,000 and over people per square mile | Feeling lonely | 1,504 |
| 20.66 | 18.89 | 19.65 | 5,000 and over people per square mile | Depression diagnosis | 1,504 |

Note: Feeling lonely is the share of adults who say they feel lonely always, usually or sometimes. Depression is the share ever told by a health professional that they have a depressive disorder, which depends on access to diagnosis.

Note: Both are modeled from the same inputs. Density classes use ACS 2019-2023 residents per square mile of 2020 land area. Lower cuts are inclusive.

Source: CDC PLACES 2025 release (BRFSS 2023), ZCTA modeled estimates, and ZCTA population density from ACS 2019-2023 residents and 2020 TIGER land area; modeled estimates, population-weighted by adults

Data: https://vishalsingh.org/zipcode/report/civic-life-loneliness-and-partnership/figures/loneliness_by_density.json

### Table 9. Voter turnout is steadily worse toward the more deprived end under the ADI and under each of the three other rankings

Voter turnout, 2020 (% of citizens aged 18 and over) by tenth of counties ranked four ways, over the 3,096 counties that have all four rankings

| bin | conf_high | conf_low | estimate | n_units | ranking |
| ---: | ---: | ---: | ---: | ---: | --- |
| 1 | 73.60 | 70.38 | 71.96 | 309 | Area Deprivation Index |
| 2 | 71.09 | 67.92 | 69.40 | 310 | Area Deprivation Index |
| 3 | 68.75 | 64.92 | 66.81 | 309 | Area Deprivation Index |
| 4 | 65.86 | 62.24 | 64.04 | 310 | Area Deprivation Index |
| 5 | 62.99 | 60.64 | 61.82 | 310 | Area Deprivation Index |
| 6 | 63.29 | 59.41 | 61.42 | 309 | Area Deprivation Index |
| 7 | 61.47 | 59.53 | 60.54 | 310 | Area Deprivation Index |
| 8 | 60.51 | 56.80 | 58.52 | 309 | Area Deprivation Index |
| 9 | 58.84 | 56.98 | 57.88 | 310 | Area Deprivation Index |
| 10 | 57.69 | 55.38 | 56.49 | 310 | Area Deprivation Index |
| 1 | 77.02 | 74.89 | 75.95 | 309 | Standardized index |
| 2 | 71.34 | 68.61 | 69.92 | 310 | Standardized index |
| 3 | 69.73 | 67.56 | 68.47 | 309 | Standardized index |
| 4 | 65.50 | 61.85 | 63.55 | 310 | Standardized index |
| 5 | 63.63 | 60.43 | 61.90 | 310 | Standardized index |
| 6 | 64.11 | 61.26 | 62.73 | 309 | Standardized index |
| 7 | 61.02 | 58.32 | 59.69 | 310 | Standardized index |
| 8 | 59.30 | 56.88 | 58.08 | 309 | Standardized index |
| 9 | 57.18 | 53.81 | 55.34 | 310 | Standardized index |
| 10 | 57.26 | 54.34 | 55.70 | 310 | Standardized index |
| 1 | 75.22 | 72.13 | 73.67 | 309 | Median household income |
| 2 | 70.05 | 66.50 | 68.27 | 310 | Median household income |
| 3 | 68.92 | 64.64 | 66.61 | 309 | Median household income |
| 4 | 66.11 | 62.22 | 64.25 | 310 | Median household income |
| 5 | 64.40 | 60.66 | 62.50 | 310 | Median household income |
| 6 | 63.66 | 59.94 | 61.88 | 309 | Median household income |
| 7 | 61.47 | 59.33 | 60.38 | 310 | Median household income |
| 8 | 60.17 | 56.69 | 58.42 | 309 | Median household income |
| 9 | 59.50 | 57.42 | 58.47 | 310 | Median household income |
| 10 | 58.82 | 56.89 | 57.86 | 310 | Median household income |
| 1 | 75.21 | 72.46 | 73.83 | 309 | Bachelor's degree share |
| 2 | 70.22 | 67.65 | 68.97 | 310 | Bachelor's degree share |
| 3 | 66.89 | 63.66 | 65.30 | 309 | Bachelor's degree share |
| 4 | 65.51 | 61.31 | 63.47 | 310 | Bachelor's degree share |
| 5 | 65.46 | 62.93 | 64.19 | 310 | Bachelor's degree share |
| 6 | 63.74 | 60.36 | 61.97 | 309 | Bachelor's degree share |
| 7 | 63.36 | 60.99 | 62.13 | 310 | Bachelor's degree share |
| 8 | 61.38 | 58.95 | 60.19 | 309 | Bachelor's degree share |
| 9 | 59.86 | 57.83 | 58.87 | 310 | Bachelor's degree share |
| 10 | 57.34 | 55.40 | 56.36 | 310 | Bachelor's degree share |

Note: Least and most deprived tenth: ADI 72.0% and 56.5%, rank correlation −1.00; standardized index 75.9% and 55.7%, −0.98; median household income 73.7% and 57.9%, −0.99; bachelor's degree share 73.8% and 56.4%, −0.98.

Note: The standardized index applies the ADI's coefficients to its measures after standardizing each one, so home value and income carry less weight than in the ADI. Income and degree share are single census measures, each ranked from the highest value to the lowest.

Source: County Health Rankings 2024 (voter turnout, 2020 presidential election), Area Deprivation Index 2023 and ACS 2019-2023 county population, population-weighted; ADI 2023 and ACS 2019-2023 tract tables (B19013, B15003 and the 15 rebuilt ADI measures)

Data: https://vishalsingh.org/zipcode/report/civic-life-loneliness-and-partnership/figures/alt_rankings.json

## All estimates

Every fact in facts.json (117), sorted by id.

| Fact id | Value | Unit | Estimand | Population | Period | Interval | n | Weights | Source | Note |
| --- | ---: | --- | --- | --- | --- | --- | ---: | --- | --- | --- |
| alt_adi_least | 72.0% | Voter turnout, 2020 (% of citizens aged 18 and over) | population-weighted mean of county voter turnout; 95% intervals from a bootstrap over counties | the least deprived tenth of counties ranked by Area Deprivation Index, among the counties with all four rankings | 2020 presidential election | 95% CI 70.4% to 73.6% | 309 counties | 2019-2023 county population | County Health Rankings |   |
| alt_adi_most | 56.5% | Voter turnout, 2020 (% of citizens aged 18 and over) | population-weighted mean of county voter turnout; 95% intervals from a bootstrap over counties | the most deprived tenth of counties ranked by Area Deprivation Index, among the counties with all four rankings | 2020 presidential election | 95% CI 55.4% to 57.7% | 310 counties | 2019-2023 county population | County Health Rankings |   |
| alt_adi_rho | −1.00 | rank correlation | Spearman rank correlation between bin order and bin mean, each bin counted once | the bins of counties ranked by Area Deprivation Index | 2020 presidential election |   |   | each bin counted once | County Health Rankings |   |
| alt_degree_least | 73.8% | Voter turnout, 2020 (% of citizens aged 18 and over) | population-weighted mean of county voter turnout; 95% intervals from a bootstrap over counties | the least deprived tenth of counties ranked by Bachelor's degree share, among the counties with all four rankings | 2020 presidential election | 95% CI 72.5% to 75.2% | 309 counties | 2019-2023 county population | County Health Rankings |   |
| alt_degree_most | 56.4% | Voter turnout, 2020 (% of citizens aged 18 and over) | population-weighted mean of county voter turnout; 95% intervals from a bootstrap over counties | the most deprived tenth of counties ranked by Bachelor's degree share, among the counties with all four rankings | 2020 presidential election | 95% CI 55.4% to 57.3% | 310 counties | 2019-2023 county population | County Health Rankings |   |
| alt_degree_rho | −0.98 | rank correlation | Spearman rank correlation between bin order and bin mean, each bin counted once | the bins of counties ranked by Bachelor's degree share | 2020 presidential election |   |   | each bin counted once | County Health Rankings |   |
| alt_income_least | 73.7% | Voter turnout, 2020 (% of citizens aged 18 and over) | population-weighted mean of county voter turnout; 95% intervals from a bootstrap over counties | the least deprived tenth of counties ranked by Median household income, among the counties with all four rankings | 2020 presidential election | 95% CI 72.1% to 75.2% | 309 counties | 2019-2023 county population | County Health Rankings |   |
| alt_income_most | 57.9% | Voter turnout, 2020 (% of citizens aged 18 and over) | population-weighted mean of county voter turnout; 95% intervals from a bootstrap over counties | the most deprived tenth of counties ranked by Median household income, among the counties with all four rankings | 2020 presidential election | 95% CI 56.9% to 58.8% | 310 counties | 2019-2023 county population | County Health Rankings |   |
| alt_income_rho | −0.99 | rank correlation | Spearman rank correlation between bin order and bin mean, each bin counted once | the bins of counties ranked by Median household income | 2020 presidential election |   |   | each bin counted once | County Health Rankings |   |
| alt_n_places | 3,096 | counties | count of counties in the common sample of the ranking comparison | counties with the outcome and all four rankings | 2020 presidential election |   |   | none (full count) | County Health Rankings |   |
| alt_rho_cut | 0.9 | absolute rank correlation | threshold between a steady and an uneven gradient | the rule for changed gradients (docs/RANKING_SWITCH_RULE.md) | 2020 presidential election |   |   | none (a fixed rule) | County Health Rankings |   |
| alt_rho_margin | 0.1 | absolute rank correlation | smallest move in the rank correlation that counts as a change of shape | the rule for changed gradients (docs/RANKING_SWITCH_RULE.md) | 2020 presidential election |   |   | none (a fixed rule) | County Health Rankings |   |
| alt_std_least | 75.9% | Voter turnout, 2020 (% of citizens aged 18 and over) | population-weighted mean of county voter turnout; 95% intervals from a bootstrap over counties | the least deprived tenth of counties ranked by Standardized index, among the counties with all four rankings | 2020 presidential election | 95% CI 74.9% to 77.0% | 309 counties | 2019-2023 county population | County Health Rankings |   |
| alt_std_most | 55.7% | Voter turnout, 2020 (% of citizens aged 18 and over) | population-weighted mean of county voter turnout; 95% intervals from a bootstrap over counties | the most deprived tenth of counties ranked by Standardized index, among the counties with all four rankings | 2020 presidential election | 95% CI 54.3% to 57.3% | 310 counties | 2019-2023 county population | County Health Rankings |   |
| alt_std_rho | −0.98 | rank correlation | Spearman rank correlation between bin order and bin mean, each bin counted once | the bins of counties ranked by Standardized index | 2020 presidential election |   |   | each bin counted once | County Health Rankings |   |
| assoc_mobility_r | −0.05 | correlation coefficient | population-weighted correlation across counties of the measure with the mean adult household income rank, bootstrap interval over counties | counties with the measure and a mobility estimate (children born 1978 to 1983 to parents at the 25th income percentile) | county measure of membership associations in 2021 against mobility of children born 1978 to 1983 | 95% CI −0.12 to 0.02 | 2,982 counties | 2018 county population (Social Capital Atlas) | County Health Rankings |   |
| ce_mobility_r | +0.51 | correlation coefficient | population-weighted correlation across counties of the measure with the mean adult household income rank, bootstrap interval over counties | counties with the measure and a mobility estimate (children born 1978 to 1983 to parents at the 25th income percentile) | county measure of JEC collective efficacy (violent crime, FBI reports 2008 to 2014) against mobility of children born 1978 to 1983 | 95% CI 0.43 to 0.58 | 2,885 counties | 2018 county population (Social Capital Atlas) | JEC Social Capital Index |   |
| census_beta_b | −0.08 | standard deviations of county mobility per standard deviation of the measure | standardized coefficient on census self-response in model B (plus income, percent black and poverty) of county mobility, weighted least squares, errors clustered on state | counties with every measure in the model and a mobility estimate (Connecticut is absent) | county measure of census self-response against mobility of children born 1978 to 1983 | 95% CI −0.21 to 0.05 | 2,915 counties | 2018 county population (Social Capital Atlas) | County Health Rankings |   |
| census_beta_c | −0.15 | standard deviations of county mobility per standard deviation of the measure | standardized coefficient on census self-response in model C (plus economic connectedness) of county mobility, weighted least squares, errors clustered on state | counties with every measure in the model and a mobility estimate (Connecticut is absent) | county measure of census self-response against mobility of children born 1978 to 1983 | 95% CI −0.26 to −0.05 | 2,915 counties | 2018 county population (Social Capital Atlas) | County Health Rankings |   |
| census_d1 | 68.6% | percent of households that answered the census without a visit from a census taker | mean of county census self-response in a tenth of counties (equal numbers of counties, ranked by county Area Deprivation Index; county ADI is the mean of the county's tract Area Deprivation Index national ranks, weighted by 2019-2023 tract population), weighted by county population, bootstrap interval over counties | households in counties in the least deprived tenth of counties by Area Deprivation Index | 2020 census | 95% CI 67.4% to 69.9% | 312 counties | 2019-2023 county population | County Health Rankings |   |
| census_d10 | 47.6% | percent of households that answered the census without a visit from a census taker | mean of county census self-response in a tenth of counties (equal numbers of counties, ranked by county Area Deprivation Index; county ADI is the mean of the county's tract Area Deprivation Index national ranks, weighted by 2019-2023 tract population), weighted by county population, bootstrap interval over counties | households in counties in the most deprived tenth of counties by Area Deprivation Index | 2020 census | 95% CI 46.1% to 49.1% | 312 counties | 2019-2023 county population | County Health Rankings |   |
| census_d10_unw | 46.3% | percent of households that answered the census without a visit from a census taker | unweighted mean of county census self-response in a tenth of counties (each county counted once), bootstrap interval over counties | households in counties in the most deprived tenth of counties by Area Deprivation Index | 2020 census | 95% CI 45.2% to 47.3% | 312 counties | none (each county counts once) | County Health Rankings |   |
| census_d1_unw | 65.0% | percent of households that answered the census without a visit from a census taker | unweighted mean of county census self-response in a tenth of counties (each county counted once), bootstrap interval over counties | households in counties in the least deprived tenth of counties by Area Deprivation Index | 2020 census | 95% CI 63.7% to 66.3% | 312 counties | none (each county counts once) | County Health Rankings |   |
| census_gap | 21.0 | percentage points, least deprived tenth minus most deprived tenth of counties | difference between the population-weighted means of census self-response in the least and the most deprived tenth of counties, bootstrap interval over counties with shared draws | households in counties in the two extreme tenths of counties by Area Deprivation Index | 2020 census | 95% CI 19.0 to 22.9 | 624 counties | 2019-2023 county population | County Health Rankings |   |
| census_mean | 66.0% | percent of households that answered the census without a visit from a census taker | mean of county self-response rates weighted by county population | households in counties with a published rate | 2020 census |   | 3,140 counties | county population (County Health Rankings) | County Health Rankings |   |
| census_mobility_r | +0.32 | correlation coefficient | population-weighted correlation across counties of the measure with the mean adult household income rank, bootstrap interval over counties | counties with the measure and a mobility estimate (children born 1978 to 1983 to parents at the 25th income percentile) | county measure of census self-response in 2020 against mobility of children born 1978 to 1983 | 95% CI 0.26 to 0.37 | 2,982 counties | 2018 county population (Social Capital Atlas) | County Health Rankings |   |
| census_nodata | 3 | counties | count of map units with no published self-response rate | counties and county equivalents on the map | 2020 census |   |   | none (full count) | County Health Rankings |   |
| census_p5 | 51.8% | percent of households that answered the census without a visit from a census taker | 5th percentile of county self-response rates, counties weighted by population | households in counties with a published rate | 2020 census |   | 3,140 counties | county population (County Health Rankings) | County Health Rankings |   |
| census_p95 | 77.5% | percent of households that answered the census without a visit from a census taker | 95th percentile of county self-response rates, counties weighted by population | households in counties with a published rate | 2020 census |   | 3,140 counties | county population (County Health Rankings) | County Health Rankings |   |
| civorg_mobility_r | +0.06 | correlation coefficient | population-weighted correlation across counties of the measure with the mean adult household income rank, bootstrap interval over counties | counties with the measure and a mobility estimate (children born 1978 to 1983 to parents at the 25th income percentile) | county measure of civic groups in 2022 against mobility of children born 1978 to 1983 | 95% CI −0.01 to 0.14 | 2,983 counties | 2018 county population (Social Capital Atlas) | Social Capital Atlas |   |
| commhealth_mobility_r | +0.19 | correlation coefficient | population-weighted correlation across counties of the measure with the mean adult household income rank, bootstrap interval over counties | counties with the measure and a mobility estimate (children born 1978 to 1983 to parents at the 25th income percentile) | county measure of JEC community health, components from 2010 to 2015 against mobility of children born 1978 to 1983 | 95% CI 0.10 to 0.28 | 2,983 counties | 2018 county population (Social Capital Atlas) | JEC Social Capital Index |   |
| comp_gap_hi | 0.49 | ladder points, partnered minus not partnered | unweighted difference in mean ladder score between partnered and other adults within a tenth of interviews ranked by the JEC Social Capital Index, interval clustered by county | adults in counties in the highest tenth of interviews by the JEC Social Capital Index of the respondent's county | interviews in 2009 to 2017 | 95% CI 0.47 to 0.51 | 220,209 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| comp_gap_lo | 0.45 | ladder points, partnered minus not partnered | unweighted difference in mean ladder score between partnered and other adults within a tenth of interviews ranked by the JEC Social Capital Index, interval clustered by county | adults in counties in the lowest tenth of interviews by the JEC Social Capital Index of the respondent's county | interviews in 2009 to 2017 | 95% CI 0.43 to 0.48 | 218,523 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| comp_gap_peak | 0.53 | ladder points, partnered minus not partnered | the largest of the ten decile premiums, which is also the one most likely to be too high (selected for being largest) | adults in counties in the tenth of interviews with the largest premium | interviews in 2009 to 2017 | 95% CI 0.51 to 0.55 | 213,735 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| comp_other_hi | 6.86 | Cantril ladder score (0 worst possible life, 10 best possible life) | unweighted mean of the 0 to 10 ladder score, interval clustered by county | adults who are single, separated, divorced or widowed, in the highest tenth of interviews by the JEC Social Capital Index of the respondent's county | interviews in 2009 to 2017 | 95% CI 6.83 to 6.89 | 79,202 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| comp_other_lo | 6.80 | Cantril ladder score (0 worst possible life, 10 best possible life) | unweighted mean of the 0 to 10 ladder score, interval clustered by county | adults who are single, separated, divorced or widowed, in the lowest tenth of interviews by the JEC Social Capital Index of the respondent's county | interviews in 2009 to 2017 | 95% CI 6.75 to 6.84 | 106,246 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| comp_partnered_hi | 7.35 | Cantril ladder score (0 worst possible life, 10 best possible life) | unweighted mean of the 0 to 10 ladder score, interval clustered by county | adults who are married or in a domestic partnership, in the highest tenth of interviews by the JEC Social Capital Index of the respondent's county | interviews in 2009 to 2017 | 95% CI 7.33 to 7.37 | 141,007 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| comp_partnered_lo | 7.25 | Cantril ladder score (0 worst possible life, 10 best possible life) | unweighted mean of the 0 to 10 ladder score, interval clustered by county | adults who are married or in a domestic partnership, in the lowest tenth of interviews by the JEC Social Capital Index of the respondent's county | interviews in 2009 to 2017 | 95% CI 7.21 to 7.30 | 112,277 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| dens_cut_high | 5,000 | people per square mile | the higher cut between density classes (a definition of the groups) | ZCTAs in the density comparison | ACS 2019-2023 (5-year), ZCTA estimates |   |   | none (definition) | CDC PLACES |   |
| dens_cut_low | 100 | people per square mile | the lower cut between density classes (a definition of the groups) | ZCTAs in the density comparison | ACS 2019-2023 (5-year), ZCTA estimates |   |   | none (definition) | CDC PLACES |   |
| dep_dens_high | 19.7% | percent of adults ever told by a health professional that they have a depressive disorder | mean of ZCTA modeled prevalence of depression weighted by adult population, bootstrap interval over counties; ZCTAs with both measures | adults in ZCTAs with a population density at or above the higher density cut | BRFSS 2023 (PLACES 2025 release) | 95% CI 18.9% to 20.7% | 1,504 ZCTAs | 2020 census adult population of the ZCTA | CDC PLACES |   |
| dep_dens_low | 23.6% | percent of adults ever told by a health professional that they have a depressive disorder | mean of ZCTA modeled prevalence of depression weighted by adult population, bootstrap interval over counties; ZCTAs with both measures | adults in ZCTAs with a population density under the lower density cut | BRFSS 2023 (PLACES 2025 release) | 95% CI 23.4% to 23.7% | 9,893 ZCTAs | 2020 census adult population of the ZCTA | CDC PLACES |   |
| ec_beta_c | 0.63 | standard deviations of county mobility per standard deviation of the measure | standardized coefficient on economic connectedness in model C (plus economic connectedness) of county mobility, weighted least squares, errors clustered on state | counties with every measure in the model and a mobility estimate (Connecticut is absent) | county measure of economic connectedness against mobility of children born 1978 to 1983 | 95% CI 0.47 to 0.79 | 2,915 counties | 2018 county population (Social Capital Atlas) | Social Capital Atlas |   |
| ec_mobility_r | +0.69 | correlation coefficient | population-weighted correlation across counties of the measure with the mean adult household income rank, bootstrap interval over counties | counties with the measure and a mobility estimate (children born 1978 to 1983 to parents at the 25th income percentile) | county measure of economic connectedness in 2022 against mobility of children born 1978 to 1983 | 95% CI 0.64 to 0.74 | 2,942 counties | 2018 county population (Social Capital Atlas) | Social Capital Atlas |   |
| famunity_mobility_r | +0.71 | correlation coefficient | population-weighted correlation across counties of the measure with the mean adult household income rank, bootstrap interval over counties | counties with the measure and a mobility estimate (children born 1978 to 1983 to parents at the 25th income percentile) | county measure of JEC family unity, components from 2012 to 2016 against mobility of children born 1978 to 1983 | 95% CI 0.67 to 0.74 | 2,913 counties | 2018 county population (Social Capital Atlas) | JEC Social Capital Index |   |
| insthealth_mobility_r | +0.24 | correlation coefficient | population-weighted correlation across counties of the measure with the mean adult household income rank, bootstrap interval over counties | counties with the measure and a mobility estimate (children born 1978 to 1983 to parents at the 25th income percentile) | county measure of JEC institutional health, components from 2010 to 2016 against mobility of children born 1978 to 1983 | 95% CI 0.16 to 0.33 | 2,959 counties | 2018 county population (Social Capital Atlas) | JEC Social Capital Index |   |
| inter_max | 0.035 | ladder points of premium per standard deviation of the county measure | the largest of the seven interaction coefficients (Volunteering group membership) | adults interviewed in counties with the measure | interviews in 2009 to 2017 | 95% CI 0.019 to 0.051 | 2,185,410 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| inter_min | 0.007 | ladder points of premium per standard deviation of the county measure | the smallest of the seven interaction coefficients (JEC institutional health) | adults interviewed in counties with the measure | interviews in 2009 to 2017 | 95% CI −0.005 to 0.019 | 2,186,233 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| inter_n_zero | 3 | of the seven interaction coefficients | count of the seven 95% intervals (clustered by state) that include zero | adults interviewed in counties with each of the seven county measures | interviews in 2009 to 2017 |   |   | none (unweighted) | Gallup U.S. Daily Poll |   |
| inter_sci | 0.023 | ladder points of premium per standard deviation of the JEC Social Capital Index | coefficient on the interaction of partnership with the county measure (standardized across counties) in a regression of the ladder score, unweighted, errors clustered on the respondent's state because the JEC community-health and institutional-health subindexes include state-level inputs | adults interviewed in counties with a JEC index | interviews in 2009 to 2017 | 95% CI 0.012 to 0.034 | 2,179,890 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| ladder_never_married | 6.77 | Cantril ladder score (0 worst possible life, 10 best possible life) | unweighted mean of the 0 to 10 ladder score, interval clustered by the respondent's county | adults who are never married | interviews in 2009 to 2017 | 95% CI 6.76 to 6.79 | 389,678 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| ladder_partnered | 7.28 | Cantril ladder score (0 worst possible life, 10 best possible life) | unweighted mean of the 0 to 10 ladder score, interval clustered by the respondent's county | adults who are married or in a domestic partnership | interviews in 2009 to 2017 | 95% CI 7.27 to 7.29 | 1,278,658 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| ladder_partnered_with2008 | 7.22 | Cantril ladder score (0 worst possible life, 10 best possible life) | unweighted mean of the 0 to 10 ladder score including the 2008 interviews, interval clustered by the respondent's county | adults who are married or in a domestic partnership | interviews in 2008 to 2017 | 95% CI 7.21 to 7.24 | 1,489,101 respondents | none (unweighted) | Gallup U.S. Daily Poll | The 2008 ladder is lowered by a question-order artifact, so this comparison is not the chapter's estimate. |
| ladder_sdw | 6.78 | Cantril ladder score (0 worst possible life, 10 best possible life) | unweighted mean of the 0 to 10 ladder score, interval clustered by the respondent's county | adults who are separated, divorced or widowed | interviews in 2009 to 2017 | 95% CI 6.77 to 6.80 | 520,865 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| ladder_share_2009 | 85% | percent of the year's interviews that carry a ladder score | share of the year's interviews with a ladder score (unweighted count, before any county or marital-status filter) | all Gallup interviews in 2009 | 2009 |   | 353,849 respondents | none (unweighted count) | Gallup U.S. Daily Poll |   |
| ladder_share_late | 49% | percent of interviews that carry a ladder score | share of interviews with a ladder score (unweighted count, before any county or marital-status filter) | all Gallup interviews from 2013 to 2017 | 2013 to 2017 |   | 1,761,191 respondents | none (unweighted count) | Gallup U.S. Daily Poll |   |
| ladder_unpartnered | 6.78 | Cantril ladder score (0 worst possible life, 10 best possible life) | unweighted mean of the 0 to 10 ladder score, interval clustered by the respondent's county | adults who are never married, separated, divorced or widowed | interviews in 2009 to 2017 | 95% CI 6.77 to 6.79 | 910,543 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| ladder_unpartnered_with2008 | 6.73 | Cantril ladder score (0 worst possible life, 10 best possible life) | unweighted mean of the 0 to 10 ladder score including the 2008 interviews, interval clustered by the respondent's county | adults who are never married, separated, divorced or widowed | interviews in 2008 to 2017 | 95% CI 6.72 to 6.74 | 1,050,196 respondents | none (unweighted) | Gallup U.S. Daily Poll | The 2008 ladder is lowered by a question-order artifact, so this comparison is not the chapter's estimate. |
| lonely_65_r | −0.48 | correlation coefficient | correlation across ZCTAs of PLACES modeled loneliness with the share of residents aged 65 and over, weighted by adult population, bootstrap interval over counties | adults in ZCTAs with a loneliness estimate and at least 500 residents | BRFSS 2023 (PLACES 2025 release) | 95% CI −0.51 to −0.45 | 19,681 ZCTAs | 2020 census adult population of the ZCTA | CDC PLACES | PLACES estimates come from a model; the ZCTA covariate is an ACS 2019-2023 five-year estimate for the ZCTA. |
| lonely_adi_r | +0.02 | correlation coefficient | correlation across ZCTAs of PLACES modeled loneliness with the Area Deprivation Index national rank, weighted by adult population, bootstrap interval over counties | adults in ZCTAs with a loneliness estimate and at least 500 residents | BRFSS 2023 (PLACES 2025 release) | 95% CI −0.13 to 0.20 | 19,622 ZCTAs | 2020 census adult population of the ZCTA | CDC PLACES | PLACES estimates come from a model; the ZCTA covariate is an ACS 2019-2023 five-year estimate for the ZCTA. |
| lonely_adi_r_unw | +0.16 | correlation coefficient | correlation across ZCTAs of PLACES modeled loneliness with the Area Deprivation Index national rank, each ZCTA counted once, bootstrap interval over counties | adults in ZCTAs with a loneliness estimate and at least 500 residents | BRFSS 2023 (PLACES 2025 release) | 95% CI 0.03 to 0.27 | 19,622 ZCTAs | none (each ZCTA counts once) | CDC PLACES |   |
| lonely_age_d1 | 37.8% | percent of adults who feel lonely always, usually or sometimes | mean of ZCTA modeled prevalence in a tenth of ZCTAs (equal numbers, ranked by median age) weighted by adult population, bootstrap interval over counties | adults in the lowest tenth of ZCTAs by median age | BRFSS 2023 (PLACES 2025 release) | 95% CI 37.1% to 38.6% | 1,966 ZCTAs | 2020 census adult population of the ZCTA | CDC PLACES |   |
| lonely_age_d10 | 29.6% | percent of adults who feel lonely always, usually or sometimes | mean of ZCTA modeled prevalence in a tenth of ZCTAs (equal numbers, ranked by median age) weighted by adult population, bootstrap interval over counties | adults in the highest tenth of ZCTAs by median age | BRFSS 2023 (PLACES 2025 release) | 95% CI 29.1% to 30.0% | 1,967 ZCTAs | 2020 census adult population of the ZCTA | CDC PLACES |   |
| lonely_age_r | −0.53 | correlation coefficient | correlation across ZCTAs of PLACES modeled loneliness with the median age, weighted by adult population, bootstrap interval over counties | adults in ZCTAs with a loneliness estimate and at least 500 residents | BRFSS 2023 (PLACES 2025 release) | 95% CI −0.57 to −0.49 | 19,679 ZCTAs | 2020 census adult population of the ZCTA | CDC PLACES | PLACES estimates come from a model; the ZCTA covariate is an ACS 2019-2023 five-year estimate for the ZCTA. |
| lonely_dens_high | 37.8% | percent of adults who feel lonely always, usually or sometimes | mean of ZCTA modeled prevalence of feeling lonely weighted by adult population, bootstrap interval over counties; ZCTAs with both measures | adults in ZCTAs with a population density at or above the higher density cut | BRFSS 2023 (PLACES 2025 release) | 95% CI 36.6% to 38.9% | 1,504 ZCTAs | 2020 census adult population of the ZCTA | CDC PLACES |   |
| lonely_dens_low | 32.8% | percent of adults who feel lonely always, usually or sometimes | mean of ZCTA modeled prevalence of feeling lonely weighted by adult population, bootstrap interval over counties; ZCTAs with both measures | adults in ZCTAs with a population density under the lower density cut | BRFSS 2023 (PLACES 2025 release) | 95% CI 32.6% to 33.0% | 9,893 ZCTAs | 2020 census adult population of the ZCTA | CDC PLACES |   |
| lonely_density_r | +0.40 | correlation coefficient | correlation across ZCTAs of PLACES modeled loneliness with the log of population density, weighted by adult population, bootstrap interval over counties | adults in ZCTAs with a loneliness estimate and at least 500 residents | BRFSS 2023 (PLACES 2025 release) | 95% CI 0.30 to 0.47 | 19,681 ZCTAs | 2020 census adult population of the ZCTA | CDC PLACES | PLACES estimates come from a model; the ZCTA covariate is an ACS 2019-2023 five-year estimate for the ZCTA. |
| lonely_income_r | −0.38 | correlation coefficient | correlation across ZCTAs of PLACES modeled loneliness with the log of median household income, weighted by adult population, bootstrap interval over counties | adults in ZCTAs with a loneliness estimate and at least 500 residents | BRFSS 2023 (PLACES 2025 release) | 95% CI −0.47 to −0.31 | 19,335 ZCTAs | 2020 census adult population of the ZCTA | CDC PLACES | PLACES estimates come from a model; the ZCTA covariate is an ACS 2019-2023 five-year estimate for the ZCTA. |
| lonely_married_d1 | 38.8% | percent of adults who feel lonely always, usually or sometimes | mean of ZCTA modeled prevalence in a tenth of ZCTAs (equal numbers, ranked by married share of residents 15 and older) weighted by adult population, bootstrap interval over counties | adults in the lowest tenth of ZCTAs by married share of residents 15 and older | BRFSS 2023 (PLACES 2025 release) | 95% CI 38.2% to 39.4% | 1,968 ZCTAs | 2020 census adult population of the ZCTA | CDC PLACES |   |
| lonely_married_d10 | 29.9% | percent of adults who feel lonely always, usually or sometimes | mean of ZCTA modeled prevalence in a tenth of ZCTAs (equal numbers, ranked by married share of residents 15 and older) weighted by adult population, bootstrap interval over counties | adults in the highest tenth of ZCTAs by married share of residents 15 and older | BRFSS 2023 (PLACES 2025 release) | 95% CI 29.6% to 30.3% | 1,969 ZCTAs | 2020 census adult population of the ZCTA | CDC PLACES |   |
| lonely_married_r | −0.65 | correlation coefficient | correlation across ZCTAs of PLACES modeled loneliness with the married share of residents 15 and older, weighted by adult population, bootstrap interval over counties | adults in ZCTAs with a loneliness estimate and at least 500 residents | BRFSS 2023 (PLACES 2025 release) | 95% CI −0.69 to −0.62 | 19,681 ZCTAs | 2020 census adult population of the ZCTA | CDC PLACES | PLACES estimates come from a model; the ZCTA covariate is an ACS 2019-2023 five-year estimate for the ZCTA. |
| lonely_married_r_nofloor | −0.65 | correlation coefficient | correlation across ZCTAs of PLACES modeled loneliness with the married share of residents 15 and older, weighted by adult population, bootstrap interval over counties (no residents floor) | adults in ZCTAs of the states with an estimate, with no minimum number of residents | BRFSS 2023 (PLACES 2025 release) | 95% CI −0.69 to −0.62 | 23,770 ZCTAs | 2020 census adult population of the ZCTA | CDC PLACES |   |
| lonely_married_r_unw | −0.63 | correlation coefficient | correlation across ZCTAs of PLACES modeled loneliness with the married share of residents 15 and older, each ZCTA counted once, bootstrap interval over counties | adults in ZCTAs with a loneliness estimate and at least 500 residents | BRFSS 2023 (PLACES 2025 release) | 95% CI −0.65 to −0.61 | 19,681 ZCTAs | none (each ZCTA counts once) | CDC PLACES |   |
| lonely_married_r_unw_nofloor | −0.54 | correlation coefficient | correlation across ZCTAs of PLACES modeled loneliness with the married share of residents 15 and older, each ZCTA counted once, bootstrap interval over counties (no residents floor) | adults in ZCTAs of the states with an estimate, with no minimum number of residents | BRFSS 2023 (PLACES 2025 release) | 95% CI −0.57 to −0.52 | 23,770 ZCTAs | none (each ZCTA counts once) | CDC PLACES |   |
| lonely_mean | 34.4% | percent of adults who feel lonely always, usually or sometimes | mean of ZCTA modeled prevalence weighted by adult population | adults in ZCTAs with a loneliness estimate and at least 500 residents | BRFSS 2023 (PLACES 2025 release) |   | 19,684 ZCTAs | 2020 census adult population of the ZCTA | CDC PLACES |   |
| lonely_poverty_r | +0.56 | correlation coefficient | correlation across ZCTAs of PLACES modeled loneliness with the poverty rate, weighted by adult population, bootstrap interval over counties | adults in ZCTAs with a loneliness estimate and at least 500 residents | BRFSS 2023 (PLACES 2025 release) | 95% CI 0.51 to 0.62 | 19,588 ZCTAs | 2020 census adult population of the ZCTA | CDC PLACES | PLACES estimates come from a model; the ZCTA covariate is an ACS 2019-2023 five-year estimate for the ZCTA. |
| lonely_poverty_r_unw | +0.50 | correlation coefficient | correlation across ZCTAs of PLACES modeled loneliness with the poverty rate, each ZCTA counted once, bootstrap interval over counties | adults in ZCTAs with a loneliness estimate and at least 500 residents | BRFSS 2023 (PLACES 2025 release) | 95% CI 0.47 to 0.53 | 19,588 ZCTAs | none (each ZCTA counts once) | CDC PLACES |   |
| married_share_d1 | 28.2% | percent of residents 15 and older who are now married, except separated | mean of the ZCTA married share of residents 15 and older in the tenth, weighted by adult population | residents 15 and older of the lowest tenth of ZCTAs by married share of residents 15 and older | ACS 2019-2023 (5-year), ZCTA estimates |   | 1,968 ZCTAs | 2020 census adult population of the ZCTA | CDC PLACES |   |
| married_share_d10 | 67.4% | percent of residents 15 and older who are now married, except separated | mean of the ZCTA married share of residents 15 and older in the tenth, weighted by adult population | residents 15 and older of the highest tenth of ZCTAs by married share of residents 15 and older | ACS 2019-2023 (5-year), ZCTA estimates |   | 1,969 ZCTAs | 2020 census adult population of the ZCTA | CDC PLACES |   |
| median_age_d1 | 30.2 | years of age | mean of the ZCTA median age in the tenth, weighted by adult population | residents of the lowest tenth of ZCTAs by median age | ACS 2019-2023 (5-year), ZCTA estimates |   | 1,966 ZCTAs | 2020 census adult population of the ZCTA | CDC PLACES |   |
| median_age_d10 | 57.8 | years of age | mean of the ZCTA median age in the tenth, weighted by adult population | residents of the highest tenth of ZCTAs by median age | ACS 2019-2023 (5-year), ZCTA estimates |   | 1,967 ZCTAs | 2020 census adult population of the ZCTA | CDC PLACES |   |
| n_counties_adi | 3,128 | counties | count of counties ranked into tenths | counties and county equivalents with an Area Deprivation Index and a population (Puerto Rico excluded) | ADI 2023, population 2019-2023 |   |   | none (full count) | Area Deprivation Index |   |
| n_gallup | 2,189,201 | interviews | count of interviews with a ladder score, a marital status and a county | adults interviewed with a ladder score, a marital status and a county | interviews in 2009 to 2017 |   |   | none (unweighted count) | Gallup U.S. Daily Poll |   |
| n_gallup_counties | 3,140 | counties | count of distinct counties assigned to respondents (county is derived from the ZIP code) | counties of the respondents' ZIP codes in the analysis | interviews in 2009 to 2017 |   |   | none (unweighted count) | Gallup U.S. Daily Poll |   |
| n_states_lonely | 39 | states | count of states with an estimate | states in which at least 90 percent of ZCTAs have a PLACES loneliness estimate | BRFSS 2023 (PLACES 2025 release) |   |   | none (full count) | CDC PLACES |   |
| n_zcta_lonely | 19,684 | ZCTAs | count of ZCTAs in the analysis | ZCTAs with a loneliness estimate, a match in the ZCTA atlas and at least 500 residents | BRFSS 2023 (PLACES 2025 release) |   |   | none (full count) | CDC PLACES |   |
| n_zcta_outside | 68 | ZCTAs | count of analysis ZCTAs whose primary state falls under the 90 percent rule | ZCTAs in the analysis assigned to a state with an estimate in fewer than 90 percent of its ZCTAs | BRFSS 2023 (PLACES 2025 release) |   |   | none (full count) | CDC PLACES |   |
| partner_premium | 0.50 | ladder points, partnered minus not partnered | unweighted difference between the mean ladder score of adults who are married or in a domestic partnership and the mean of all others, interval clustered by county | adults interviewed with a ladder score, a marital status and a county | interviews in 2009 to 2017 | 95% CI 0.49 to 0.51 | 2,189,201 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| partner_premium_from2010 | 0.50 | ladder points, partnered minus not partnered | unweighted difference between the mean ladder score of adults who are married or in a domestic partnership and the mean of all others, leaving out 2009 as well, interval clustered by county | adults interviewed with a ladder score, a marital status and a county, from 2010 on | interviews in 2010 to 2017 | 95% CI 0.49 to 0.51 | 1,890,860 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| partner_premium_with2008 | 0.50 | ladder points, partnered minus not partnered | unweighted difference between the mean ladder score of adults who are married or in a domestic partnership and the mean of all others, including the 2008 interviews, interval clustered by county | adults interviewed with a ladder score, a marital status and a county, including 2008 | interviews in 2008 to 2017 | 95% CI 0.49 to 0.51 | 2,539,297 respondents | none (unweighted) | Gallup U.S. Daily Poll | The 2008 ladder is lowered by a question-order artifact, so this comparison is not the chapter's estimate. |
| partner_premium_within | 0.50 | ladder points, partnered minus not partnered | coefficient on being married or in a domestic partnership in a regression of the ladder score on county fixed effects, unweighted, errors clustered on the respondent's county | adults interviewed with a ladder score, a marital status and a county | interviews in 2009 to 2017 | 95% CI 0.49 to 0.50 | 2,189,199 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| places_poverty_line | 150% | percent of the federal poverty level | the county input of the PLACES model is the share of adults below this multiple of the poverty line (a definition) | adults in the county input of the PLACES model | PLACES 2025 release |   |   | none (definition) | CDC PLACES |   |
| premium_adj_age_sex | 0.64 | ladder points, partnered minus not partnered | coefficient on being married or in a domestic partnership in a regression of the ladder score on county, year, single-year age and sex fixed effects, unweighted, errors clustered on the respondent's county | adults interviewed with a ladder score, a marital status, an age and a sex | interviews in 2009 to 2017 | 95% CI 0.63 to 0.65 | 2,163,354 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| premium_adj_full | 0.31 | ladder points, partnered minus not partnered | coefficient on being married or in a domestic partnership in a regression of the ladder score on county, year, single-year age, sex, household income and education fixed effects, unweighted, errors clustered on the respondent's county | adults interviewed with a ladder score, a marital status, an age, a sex, a household income bracket and an education level | interviews in 2009 to 2017 | 95% CI 0.30 to 0.32 | 1,786,303 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| premium_age_old | 0.35 | ladder points, partnered minus not partnered | coefficient on being married or in a domestic partnership in a regression of the ladder score on county, year, single-year age and sex fixed effects, estimated separately in the age band, unweighted, errors clustered on the respondent's county | adults aged 75 and over with a ladder score, a marital status, an age and a sex | interviews in 2009 to 2017 | 95% CI 0.33 to 0.37 | 286,964 respondents | none (unweighted) | Gallup U.S. Daily Poll | Estimated within the age band, with single-year age and sex fixed effects. |
| premium_age_peak | 0.80 | ladder points, partnered minus not partnered | coefficient on being married or in a domestic partnership in a regression of the ladder score on county, year, single-year age and sex fixed effects, estimated separately in the age band, unweighted, errors clustered on the respondent's county | adults aged 45 to 54 with a ladder score, a marital status, an age and a sex | interviews in 2009 to 2017 | 95% CI 0.78 to 0.82 | 377,207 respondents | none (unweighted) | Gallup U.S. Daily Poll | Estimated within the age band, with single-year age and sex fixed effects. |
| premium_age_young | 0.15 | ladder points, partnered minus not partnered | coefficient on being married or in a domestic partnership in a regression of the ladder score on county, year, single-year age and sex fixed effects, estimated separately in the age band, unweighted, errors clustered on the respondent's county | adults aged 18 to 24 with a ladder score, a marital status, an age and a sex | interviews in 2009 to 2017 | 95% CI 0.12 to 0.18 | 147,891 respondents | none (unweighted) | Gallup U.S. Daily Poll | Estimated within the age band, with single-year age and sex fixed effects. |
| r2_composition | 65% | percent of the weighted variance in ZCTA modeled loneliness | R-squared of a regression of ZCTA modeled loneliness on median age, share aged 65 and over, race and ethnicity shares and share with a bachelor's degree or more, weighted by adult population, bootstrap interval over counties | adults in ZCTAs with a loneliness estimate and at least 500 residents | BRFSS 2023 (PLACES 2025 release) | 95% CI 62% to 69% | 19,659 ZCTAs | 2020 census adult population of the ZCTA | CDC PLACES |   |
| r2_gain_married | 5.2 | percentage points of R-squared | increase in R-squared from adding the married share to the composition regression, bootstrap interval over counties | adults in ZCTAs with a loneliness estimate and at least 500 residents | BRFSS 2023 (PLACES 2025 release) | 95% CI 3.7 to 6.8 | 19,659 ZCTAs | 2020 census adult population of the ZCTA | CDC PLACES |   |
| relig_mobility_r | −0.10 | correlation coefficient | population-weighted correlation across counties of the measure with the mean adult household income rank, bootstrap interval over counties | counties with the measure and a mobility estimate (children born 1978 to 1983 to parents at the 25th income percentile) | county measure of religious congregations in 2010 against mobility of children born 1978 to 1983 | 95% CI −0.15 to −0.04 | 2,983 counties | 2018 county population (Social Capital Atlas) | JEC Social Capital Index |   |
| sci_mobility_r | +0.61 | correlation coefficient | population-weighted correlation across counties of the measure with the mean adult household income rank, bootstrap interval over counties | counties with the measure and a mobility estimate (children born 1978 to 1983 to parents at the 25th income percentile) | county measure of the JEC index, 2018 release with components from 2010 to 2016 against mobility of children born 1978 to 1983 | 95% CI 0.53 to 0.68 | 2,889 counties | 2018 county population (Social Capital Atlas) | JEC Social Capital Index |   |
| share_interviews_early | 61% | percent of the interviews in the analysis | share of the analysis interviews that were conducted in 2009 to 2012 (unweighted count) | adults interviewed with a ladder score, a marital status and a county | interviews in 2009 to 2017 |   | 2,189,201 respondents | none (unweighted count) | Gallup U.S. Daily Poll |   |
| share_pop_lonely | 75% | percent of the residents of ZCTAs in PLACES | residents of ZCTAs with a loneliness estimate as a share of residents of all ZCTAs in the file | residents of ZCTAs in the PLACES ZCTA file | 2020 census population |   |   | none (full count) | CDC PLACES |   |
| state_coverage_cut | 90% | percent of a state's ZCTAs | minimum share of a state's ZCTAs with a loneliness estimate for the state to count as covered (a definition) | states in the PLACES ZCTA file | BRFSS 2023 (PLACES 2025 release) |   |   | none (definition) | CDC PLACES |   |
| us_census_response | 66.0% | percent of households that answered the census without a visit from a census taker | mean of county census self-response over all the counties ranked into tenths (county ADI is the mean of the county's tract Area Deprivation Index national ranks, weighted by 2019-2023 tract population), weighted by county population, bootstrap interval over counties | households in all the counties ranked into tenths by Area Deprivation Index (all ten tenths pooled) | 2020 census | 95% CI 65.5% to 66.6% | 3,125 counties | 2019-2023 county population | County Health Rankings |   |
| us_turnout | 67.8% | percent of citizens aged 18 and over who voted | mean of county voter turnout over all the counties ranked into tenths (county ADI is the mean of the county's tract Area Deprivation Index national ranks, weighted by 2019-2023 tract population), weighted by county population, bootstrap interval over counties | citizens aged 18 and over in all the counties ranked into tenths by Area Deprivation Index (all ten tenths pooled) | 2020 presidential election | 95% CI 67.1% to 68.6% | 3,098 counties | 2019-2023 county population | County Health Rankings |   |
| vol_beta_c | −0.16 | standard deviations of county mobility per standard deviation of the measure | standardized coefficient on volunteering group membership in model C (plus economic connectedness) of county mobility, weighted least squares, errors clustered on state | counties with every measure in the model and a mobility estimate (Connecticut is absent) | county measure of volunteering group membership against mobility of children born 1978 to 1983 | 95% CI −0.22 to −0.09 | 2,915 counties | 2018 county population (Social Capital Atlas) | Social Capital Atlas |   |
| vol_mobility_r | +0.19 | correlation coefficient | population-weighted correlation across counties of the measure with the mean adult household income rank, bootstrap interval over counties | counties with the measure and a mobility estimate (children born 1978 to 1983 to parents at the 25th income percentile) | county measure of volunteering group membership in 2022 against mobility of children born 1978 to 1983 | 95% CI 0.12 to 0.25 | 2,983 counties | 2018 county population (Social Capital Atlas) | Social Capital Atlas |   |
| voter_beta_a | 0.24 | standard deviations of county mobility per standard deviation of the measure | standardized coefficient on voter turnout in model A (civic measures only) of county mobility, weighted least squares, errors clustered on state | counties with every measure in the model and a mobility estimate (Connecticut is absent) | county measure of voter turnout against mobility of children born 1978 to 1983 | 95% CI 0.06 to 0.42 | 2,915 counties | 2018 county population (Social Capital Atlas) | County Health Rankings |   |
| voter_beta_b | 0.00 | standard deviations of county mobility per standard deviation of the measure | standardized coefficient on voter turnout in model B (plus income, percent black and poverty) of county mobility, weighted least squares, errors clustered on state | counties with every measure in the model and a mobility estimate (Connecticut is absent) | county measure of voter turnout against mobility of children born 1978 to 1983 | 95% CI −0.15 to 0.16 | 2,915 counties | 2018 county population (Social Capital Atlas) | County Health Rankings |   |
| voter_d1 | 72.1% | percent of citizens aged 18 and over who voted | mean of county voter turnout in a tenth of counties (equal numbers of counties, ranked by county Area Deprivation Index; county ADI is the mean of the county's tract Area Deprivation Index national ranks, weighted by 2019-2023 tract population), weighted by county population, bootstrap interval over counties | citizens aged 18 and over in counties in the least deprived tenth of counties by Area Deprivation Index | 2020 presidential election | 95% CI 70.6% to 73.8% | 307 counties | 2019-2023 county population | County Health Rankings |   |
| voter_d10 | 56.4% | percent of citizens aged 18 and over who voted | mean of county voter turnout in a tenth of counties (equal numbers of counties, ranked by county Area Deprivation Index; county ADI is the mean of the county's tract Area Deprivation Index national ranks, weighted by 2019-2023 tract population), weighted by county population, bootstrap interval over counties | citizens aged 18 and over in counties in the most deprived tenth of counties by Area Deprivation Index | 2020 presidential election | 95% CI 55.4% to 57.6% | 310 counties | 2019-2023 county population | County Health Rankings |   |
| voter_d10_unw | 58.8% | percent of citizens aged 18 and over who voted | unweighted mean of county voter turnout in a tenth of counties (each county counted once), bootstrap interval over counties | citizens aged 18 and over in counties in the most deprived tenth of counties by Area Deprivation Index | 2020 presidential election | 95% CI 57.7% to 59.9% | 310 counties | none (each county counts once) | County Health Rankings |   |
| voter_d1_unw | 76.9% | percent of citizens aged 18 and over who voted | unweighted mean of county voter turnout in a tenth of counties (each county counted once), bootstrap interval over counties | citizens aged 18 and over in counties in the least deprived tenth of counties by Area Deprivation Index | 2020 presidential election | 95% CI 76.0% to 77.9% | 307 counties | none (each county counts once) | County Health Rankings |   |
| voter_gap | 15.6 | percentage points, least deprived tenth minus most deprived tenth of counties | difference between the population-weighted means of voter turnout in the least and the most deprived tenth of counties, bootstrap interval over counties with shared draws | citizens aged 18 and over in counties in the two extreme tenths of counties by Area Deprivation Index | 2020 presidential election | 95% CI 13.9 to 17.7 | 617 counties | 2019-2023 county population | County Health Rankings |   |
| voter_mobility_r | +0.35 | correlation coefficient | population-weighted correlation across counties of the measure with the mean adult household income rank, bootstrap interval over counties | counties with the measure and a mobility estimate (children born 1978 to 1983 to parents at the 25th income percentile) | county measure of voter turnout in 2020 against mobility of children born 1978 to 1983 | 95% CI 0.28 to 0.42 | 2,964 counties | 2018 county population (Social Capital Atlas) | County Health Rankings |   |
| zcta_floor | 500 | residents | minimum number of residents for a ZCTA to enter (a definition of the sample) | ZCTAs in the loneliness analysis | 2020 census |   |   | none (definition) | CDC PLACES |   |

## Definitions

- Area Deprivation Index (ADI): A score that ranks neighborhoods by income, education, employment and housing conditions drawn from the American Community Survey. In this book a higher national percentile means a more deprived neighborhood.
- ZCTA: ZIP Code Tabulation Area, the Census Bureau's approximation of a ZIP code as a set of census blocks. It lets census data be reported by ZIP code, though some ZIP codes have no matching ZCTA.
- Income rank (kfr): The position that adult children reach in the national distribution of household income, from the lowest rank to the highest, averaged over children who grew up in a place with parents at a stated income level. The Opportunity Atlas labels this measure kfr.
- Bootstrap interval: A range for an estimate found by recomputing it on many random resamples of the data and reading the range from the results. It shows how much the estimate would vary from sample to sample.
- Economic connectedness: A measure of friendship across class lines, based on the share of high-status friends among people below the median in socioeconomic status. It is scaled so that friendships mixing in proportion to the population give the reference value.
- Cantril ladder: A survey question that asks people to place their life on a ladder whose bottom step is the worst possible life and whose top step is the best possible. It measures how people evaluate their life as a whole, not how they felt yesterday.
- Fixed effects: A regression adjustment that gives each group, such as each county, its own baseline. The estimate then rests on differences within groups.
- Modeled estimate: A figure produced by a statistical model instead of a direct count or measurement, usually for small areas where direct data are thin, borrowing strength from surveys, neighboring areas or other data. Neighboring areas can share errors, so single extremes call for caution.
- Ecological inference: Drawing conclusions about individuals from data about places or groups. A pattern across counties need not hold for the people inside them.

## Methods

This chapter compares voter turnout and census self-response across ten groups of counties ordered by deprivation, the Gallup life-evaluation score of partnered and unpartnered adults, and modeled loneliness across ZCTAs. All Gallup estimates are unweighted, the 2008 interviews are left out, and the ladder was asked of different shares of respondents in different years, so the pooled figures weight 2009 to 2012 most. Loneliness and depression are modeled estimates, and the ZCTA characteristics are ACS 2019-2023 five-year estimates without margins of error. Gallup respondent rows are read inside Box and aggregated; nothing at respondent level is published.

### County deprivation, civic measures and intervals

County deprivation is the mean of the Area Deprivation Index 2023 national ranks of a county's tracts, weighted by 2019-2023 tract population (Connecticut tracts move to the eight legacy counties; Puerto Rico is excluded), and counties are cut into ten groups of equal numbers of counties.

Decile means of voter turnout and census self-response (County Health Rankings 2024) are weighted by county population, with 95% percentile intervals from 2,000 draws of whole counties with replacement (seed 12345); the same routine gives the correlations across counties, which are weighted by the Social Capital Atlas 2018 population as in the chapter on race, income and mobility, and the differences between deciles with shared draws.

The three county regressions use z-scores with the same weight, and errors clustered on state with t intervals.

Explains Table 2, Table 1 and Table 3.

### The Gallup sample is unweighted and its years differ by question coverage

Gallup respondent rows are read inside Box and aggregated; nothing at respondent level is published. All Gallup estimates are unweighted.

The 2008 interviews are left out because Gallup asked political questions immediately before the ladder in 2008, which lowers the ladder score (Deaton 2012); including them leaves the premium at 0.50. The ladder was asked of nearly all respondents in 2010 to 2012, of 85% in 2009 and of 49% from 2013 to 2017, so 61% of the analysis interviews are from 2009 to 2012 and the pooled figures weight those years most.

Partnered means married or in a domestic partnership.

Explains Table 4.

### Adjusted comparisons and their standard errors

Group means and differences have errors clustered on the respondent's county (derived from the ZIP code), and the fixed-effect models are linear regressions with county, year, single-year age and sex fixed effects (and household income and education where stated) and errors clustered on county, with t intervals; a fixed effect for each year of age is used because the ladder is U-shaped in age and a linear age term fits poorly.

Explains Table 4.

### County social capital and the interaction models

The county Social Capital Index of the respondent's county is joined in Box on the county code and respondents are cut into ten groups of about equal numbers of interviews.

The interaction models cluster on the respondent's state, because the JEC community-health and institutional-health subindexes include state-level inputs.

Explains Table 5.

### Modeled loneliness and depression across ZCTAs

Loneliness and depression are PLACES 2025 release ZCTA estimates (BRFSS 2023), modeled by multilevel regression and poststratification; ZCTAs enter with at least 500 residents, and correlations, decile means and the R-squared are weighted by the ZCTA's adult population and use bootstrap intervals over counties with the same draws.

The R-squared and its gain come from the sums of the weighted regression within each county, resampled with the same draws.

ZCTA characteristics are ACS 2019-2023 five-year estimates for each ZCTA (the ZCTA's own estimate, not a sum of tracts), pulled from the Census API: the married share is now married except separated, of residents 15 and older; race shares are race alone; education is the share of adults 25 and older with a bachelor's degree or more; density is residents per square mile of 2020 TIGER land area. An earlier version used ACS-based values of undocumented year from the project's ZCTA atlas.

Explains Table 8, Table 7 and Table 6.

### Four rankings of the same places

The comparison follows the book's ranking switch rule, fixed on October 28, 2026 before any of these gradients was computed. The main gradient of the chapter is recomputed under four rankings of the same places, with the outcome, period, weights, bins and intervals unchanged: the published ADI; a standardized index that applies the ADI's coefficients to its measures after standardizing each one across tracts; median household income; and the share of adults 25 and older with a bachelor's degree or more. The three alternatives come from ACS tract estimates for 2019 to 2023, the period of the ADI.

All four rankings use the same places, the 3,096 counties that have the outcome and all four rankings, so the ADI panel can differ slightly from the chapter's main figure. Tracts enter only if they have every input of the standardized index, which leaves out tracts without a median rent, home value or mortgage cost. A county takes the population-weighted mean of the ranks of its common-sample tracts, weighted by 2019-2023 tract population, with Connecticut tracts assigned to the eight legacy counties; the decile means keep the chapter's weight, the county population of all its tracts, and the deciles are cut within the common sample.

A pattern counts as changed when the worse end switches sides, when the two ends differ under one ranking and not the other, or when the rank correlation of bin order and bin mean falls on the other side of 0.9 and moves by at least 0.1.

Explains Table 9.

## What these data cannot show

The county and ZCTA patterns are ecological. A county with low turnout need not be one where the same people stayed home and had fewer chances to rise, and a ZCTA with high modeled loneliness need not hold lonelier people than its neighbors. Deaths are outside this chapter (Chapter 8.2, Overdose, alcohol and suicide (https://vishalsingh.org/zipcode/overdose-alcohol-and-suicide) maps overdose, alcohol and suicide deaths).

The county measures are dated 2010 to 2022 and mobility refers to children born in 1978 to 1983, so the comparisons assume that county differences persist. Census self-response reflects the operation as well as the residents in places counted in person, and turnout is limited to citizens and shaped by state law. The Joint Economic Committee index is a compiled snapshot of inputs from different years, and the Atlas measures describe active Facebook users in 2022, not a census of ties. The regressions are partial associations among correlated measures, with errors clustered on state.

The Gallup partner premium is a comparison between people at one time, in a repeated cross-section and not a panel. Selection into partnership is the main limit, and the panel studies above find evidence of it. Partnered means married or in a domestic partnership, so adults who live with a partner and call themselves single are in the comparison group. County comes from the respondent's ZIP code and is approximate near county lines. The extract is unweighted, so the levels can differ from the weighted figures Gallup publishes, and the evaluative ladder is the only Gallup measure used here.

The PLACES loneliness estimates are modeled, exist only for the states that asked the question, and do not include marital status as an input. The ZCTA characteristics are American Community Survey estimates for 2019 to 2023 for each ZCTA, without margins of error, so small ZCTAs carry sampling error that the intervals do not show. The married share counts residents aged fifteen and older who are married and not separated. An earlier version took these characteristics from a table of undocumented year. ZCTAs are not ZIP codes. The intervals in the figures describe the bootstrap over counties and do not carry the model's own uncertainty for each ZCTA. Puerto Rico is not in County Health Rankings and is not drawn, and Connecticut appears under its legacy counties in the county files and drops out of the regressions that need the ACS tract file, which uses planning-region codes.

## References

1. University of Wisconsin Population Health Institute (2024). County Health Rankings & Roadmaps Technical Documentation, Version 2. University of Wisconsin Population Health Institute (updated March 18, 2024). https://www.countyhealthrankings.org/sites/default/files/media/document/2024%20CHRR%20Technical%20Document_1.pdf
2. U.S. Census Bureau (2024). 2020 Census Operational Timeline and Adjustments Due to COVID-19. U.S. Census Bureau web page, last revised September 16, 2024. https://www.census.gov/programs-surveys/decennial-census/decade/2020/planning-management/operational-adjustments.html
3. Putnam, Robert D. (2000). Bowling Alone: The Collapse and Revival of American Community. Simon & Schuster, New York.
4. U.S. Congress Joint Economic Committee, Social Capital Project (2018). The Geography of Social Capital in America. Social Capital Project Report No. 1-18, Joint Economic Committee, U.S. Congress. https://www.jec.senate.gov/public/index.cfm/republicans/2018/4/the-geography-of-social-capital-in-america
5. Chetty, Raj; Jackson, Matthew O.; Kuchler, Theresa; Stroebel, Johannes; et al. (2022). Social capital I: measurement and associations with economic mobility. Nature 608: 108-121. https://doi.org/10.1038/s41586-022-04996-4
6. Deaton, Angus (2012). The financial crisis and the well-being of Americans: 2011 OEP Hicks Lecture. Oxford Economic Papers 64(1): 1-26. https://doi.org/10.1093/oep/gpr051
7. Stutzer, Alois; Frey, Bruno S. (2006). Does marriage make people happy, or do happy people get married?. The Journal of Socio-Economics 35(2): 326-347. https://doi.org/10.1016/j.socec.2005.11.043
8. Lucas, Richard E.; Clark, Andrew E.; Georgellis, Yannis; Diener, Ed (2003). Reexamining adaptation and the set point model of happiness: Reactions to changes in marital status. Journal of Personality and Social Psychology 84(3): 527-539. https://doi.org/10.1037/0022-3514.84.3.527
9. Centers for Disease Control and Prevention (2025). PLACES Measure Definitions: Health-Related Social Needs. CDC PLACES: Local Data for Better Health, web page updated December 7, 2025. https://www.cdc.gov/places/measure-definitions/health-related-social-needs.html
10. Centers for Disease Control and Prevention (2025). PLACES Methodology. CDC PLACES: Local Data for Better Health, web page last reviewed December 7, 2025. https://www.cdc.gov/places/methodology/index.html
11. Zhang, Xingyou; Holt, James B.; Lu, Hua; Wheaton, Anne G.; Ford, Earl S.; Greenlund, Kurt J.; Croft, Janet B. (2014). Multilevel regression and poststratification for small-area estimation of population health outcomes: a case study of chronic obstructive pulmonary disease prevalence using the Behavioral Risk Factor Surveillance System. American Journal of Epidemiology 179(8): 1025-1033. https://doi.org/10.1093/aje/kwu018

## Provenance

- Pipeline version: 0.1.0
- Build script: book2/chapters/civic-life-loneliness-and-partnership/build.py (sha256 a980485db4acdff837ba27a3a56ad644427977a36d02b22e04fa916e4d411758)
- Built on: 2026-10-07
- Site commit: c2b1ff3309cb4a10833f884da91ac0057a66ad0a

Inputs:

| File | Size (bytes) | SHA-256, first 1 MB | SHA-256, last 1 MB |
| --- | ---: | --- | --- |
| book2/chapters/civic-life-loneliness-and-partnership/civic_data.py | 16,647 | d19ee2c18a2b9a7ab4a8f41150a0b0a66c562c463daf3a45f8116df7c329cedf | d19ee2c18a2b9a7ab4a8f41150a0b0a66c562c463daf3a45f8116df7c329cedf |
| book2/chapters/civic-life-loneliness-and-partnership/lonely_data.py | 11,042 | bdb0b9d8ff4441dc5d344112b48678e42b021e0503beb61ec7c72dd2e1522dae | bdb0b9d8ff4441dc5d344112b48678e42b021e0503beb61ec7c72dd2e1522dae |
| book2/chapters/civic-life-loneliness-and-partnership/partner_data.py | 12,972 | 56cba90356e8b6e25ea0bf73f9ff1312322f7c1a7a0d58d74b3f8bf447552bb3 | 56cba90356e8b6e25ea0bf73f9ff1312322f7c1a7a0d58d74b3f8bf447552bb3 |
| data/analytic/tract_atlas.parquet | 39,659,925 | 29d5e4d943552d54aeb6ba92aadc939d9bd15bdea553c78357dbd516757be627 | 1e10d1ecf722340f23ee2212d4999a90250f85b65fa89b39e32db06d9472e22e |
| data/analytic/zcta_atlas.parquet | 674,348,737 | 0abfad485b74b74eb956fdc57c88e504abe10d601a2e99a3f16467469826578c | 0e37a4ce00d8569d49dccd0b2dfdc2c8435934f9c8a606bbca2ca1dad2c9effe |
| data/geo/zctas_2020.parquet | 795,216,625 | f0928267645eb4deedf7037797ce412fd13f7ceb083db8d7dac0b8b9dab8fca0 | 13af30cbf46cbb47798c9ae53ff4b4f3685b8f39dea763d8721a7082dab1f7d9 |
| data/raw/acs/deprivation_components_2019_2023/tract.parquet | 13,197,793 | 5545cde2cbe96a8b56e468d65388248a241aa6bd300d3c9e325d08d3f86b11da | 5c8ec454d066988b0f1a01aaa6ed34ebfc05d162e01cf9c6bb01f9198cbaff12 |
| data/raw/acs/zcta_controls_2019_2023/zcta.parquet | 5,262,344 | 291be093b86c45b966601918649e0c30e34dba24ab5a501398659d6d3a0457a8 | 961d494333cc7df244d280317fd111c5743822c5bb18b07ca0d84cc331e020a4 |
| data/raw/gallup/gallup_us.parquet | 209,530,597 | b57c719f2c1d4749e43b1a70ca6dbd97f0480a0b538983091eea7eadfec40551 | b8167235969e50da6143a0e8d3e74ffb8f7c231743acc9fb2a78130a602af6f5 |
| data/staged/adi/adi_atlas_tract2020_2023.parquet | 798,078 | 5e409bc212ff871def780fa27b4cf8b616d9da420b2b9f6455517eddcd8c1a95 | 5e409bc212ff871def780fa27b4cf8b616d9da420b2b9f6455517eddcd8c1a95 |
| data/staged/crosswalks/ct_tract2020_to_old_county.parquet | 15,842 | 961ef39f31596ef65fca2c67d5f2f63e533c5dc9b49745d3bb6437844e4b831d | 961ef39f31596ef65fca2c67d5f2f63e533c5dc9b49745d3bb6437844e4b831d |
| data/staged/health/chr_county_2024.parquet | 8,776,933 | 619795f7ecfec8aff10694b649eaee5e421dbeffbe4be47af4161b51ceefa5cd | c69357f1bcf630407bb9716cb445618838999558c019aaaf77d4681b6bf45c56 |
| data/staged/health/places_zcta_latest.parquet | 5,315,556 | 1b38963b19eebcdd7f803cdef49d3586aee3f1c32d22bd67063ae823c6a604e4 | 9a1f9bd34ba5e645312981527a7d51bb100a0684f41cff37108b189294496f1d |
| data/staged/mobility/oa_outcomes_county.parquet | 721,343 | fd7bc55294fbd3f7a646f27083500c1f12415775db7a3607878b522d277f8140 | fd7bc55294fbd3f7a646f27083500c1f12415775db7a3607878b522d277f8140 |
| data/staged/ses/acs5_tract2020_2019_2023.parquet | 7,232,424 | 5fe6f730a35225f14d5d8abfc6b9bce58b52e545f309a7eac74f1dae74d7a31c | 7b4a2475b3852d3f91804716a04566ef8ceadacec2be9fb85f5eb642078efac6 |
| data/staged/social_capital/jec_social_capital_county.parquet | 460,297 | 1f042e817f25c79bc96cf0a256133c4068e655bd10caff55c2ae0dc3c8f7df22 | 1f042e817f25c79bc96cf0a256133c4068e655bd10caff55c2ae0dc3c8f7df22 |
| data/staged/social_capital/sc_atlas_county_2022.parquet | 655,734 | 9ee5596513748a9b62c3e5fc559ad41b804696702c9f92848be04a05222f2f46 | 9ee5596513748a9b62c3e5fc559ad41b804696702c9f92848be04a05222f2f46 |
| data/staged/svi/svi_tract2020_2022.parquet | 17,740,415 | cb3860d8edc2aaf1f2a02211a58f7a0b80457eb6ad9a37b5ec8abb2d5da9da10 | e371a4a696d5bed05d54ad7c5b6ca2504508805b847ec415c2322be7f66ec2b2 |

Outputs:

| File | SHA-256 |
| --- | --- |
| facts.json | 885b22dae15bf5507523280f10d2ec8ad50a4009d4607d9ea4e70c0c6fa8d1c6 |
| figures/alt_rankings.json | ee02e7e5c4deefeb5d2b37f55efc742af66468028d2cd7818d70a22f684b901b |
| figures/census_response_map.json | 8948582cb33833025e00bd2109e0d482ae6eb20e433aa52de952931dccf8ae89 |
| figures/civic_gradient.json | 16152938d8fc8a1900bb13ce35aeea159621130649fbd0567a20ee5cb1ee1795 |
| figures/loneliness_by_density.json | 9a3bdba30478fb6295b963b131db9173b7315949ffa9579c4d8833382f1f2590 |
| figures/loneliness_by_marriage_and_age.json | 34ebe0e3760a72788611685ed79243c9ea5424a4bcdf2e948d07573222765d95 |
| figures/loneliness_correlates.json | 4971cfa6ed3b22af3676db6031fb96df4d3ff4c8aa48a6581a914b24664530a4 |
| figures/participation_and_connectedness.json | 155995f053e9bc7077764ab3c11a60021d4304c180dbe930174d0b8ec9038069 |
| figures/partner_premium_by_social_capital.json | 18532469287af81fa09ffe4b5e5291e632ebeb72c53ee892131e6602efdd81f2 |
| figures/partner_premium_specifications.json | 4a0caa49b79969551405bd0f4be5b5f9983ab86dfa415aa137a61f34c424fc98 |
| meta.json | 591e8a5b81bbb3a06d9ad8abcb604ec32c6831dfaaef4df0eb5781b0f3e8eb27 |
| public/data/civic-life-loneliness-and-partnership/census_response_map.json | 1358ea4fe8d258685a4c30c82c573120bed317114f6b9a505c1f5b07be6d4301 |
| refs.json | 3ad232faa21f0634b62742682e7b21bf651123b771d98a12ecb44c652213a0a3 |

## Machine formats

- Report (HTML): https://vishalsingh.org/zipcode/civic-life-loneliness-and-partnership/report
- Report (Markdown): https://vishalsingh.org/zipcode/civic-life-loneliness-and-partnership/report.md
- Every estimate (JSON): https://vishalsingh.org/zipcode/report/civic-life-loneliness-and-partnership/facts.json
- Table 1, figure "civic_gradient" (JSON): https://vishalsingh.org/zipcode/report/civic-life-loneliness-and-partnership/figures/civic_gradient.json
- Table 2, figure "census_response_map" (JSON): https://vishalsingh.org/zipcode/report/civic-life-loneliness-and-partnership/figures/census_response_map.json
- Table 3, figure "participation_and_connectedness" (JSON): https://vishalsingh.org/zipcode/report/civic-life-loneliness-and-partnership/figures/participation_and_connectedness.json
- Table 4, figure "partner_premium_specifications" (JSON): https://vishalsingh.org/zipcode/report/civic-life-loneliness-and-partnership/figures/partner_premium_specifications.json
- Table 5, figure "partner_premium_by_social_capital" (JSON): https://vishalsingh.org/zipcode/report/civic-life-loneliness-and-partnership/figures/partner_premium_by_social_capital.json
- Table 6, figure "loneliness_correlates" (JSON): https://vishalsingh.org/zipcode/report/civic-life-loneliness-and-partnership/figures/loneliness_correlates.json
- Table 7, figure "loneliness_by_marriage_and_age" (JSON): https://vishalsingh.org/zipcode/report/civic-life-loneliness-and-partnership/figures/loneliness_by_marriage_and_age.json
- Table 8, figure "loneliness_by_density" (JSON): https://vishalsingh.org/zipcode/report/civic-life-loneliness-and-partnership/figures/loneliness_by_density.json
- Table 9, figure "alt_rankings" (JSON): https://vishalsingh.org/zipcode/report/civic-life-loneliness-and-partnership/figures/alt_rankings.json
- Payload of "census_response_map" (JSON): https://vishalsingh.org/zipcode/data/civic-life-loneliness-and-partnership/census_response_map.json
- Index of every report: https://vishalsingh.org/zipcode/llms.txt; all reports in one file: https://vishalsingh.org/zipcode/llms-full.txt

---

# Life evaluation and daily feelings: report

Chapter 6.3 of ZIP Code Destiny, by Vishal Singh (NYU Stern). Generated from the chapter's data files; the reading version is at https://vishalsingh.org/zipcode/life-evaluation-and-daily-feelings. Table n of this report is figure n of the chapter.

## Identification

- Title: Life evaluation and daily feelings
- Label: The ladder and the day
- Chapter: 6.3 (Part VI, Environment, connection and well-being; An American life)
- Question: How do people rate their lives and describe their days in richer and poorer places, and do the two measures move together?
- Dek: Gallup respondents in the most deprived tenth of counties rate their lives 0.24 ladder points lower than those in the least deprived tenth, a gap near zero or below at the same household income, characteristics and county rent.
- Grain: individuals
- Status: published
- Published: 2026-10-09
- Updated: not revised
- Data sources:
  - Gallup U.S. Daily Poll. Gallup U.S. Daily tracking poll: Cantril ladder now and in five years (2009-2017) and seven feelings yesterday (2008-2017), household income, age, sex, race and education; aggregates only; years: 2008-2017; weights: none (unweighted)
  - Area Deprivation Index. Area Deprivation Index 2023, national rank: tract ranks averaged to counties, and ZCTA ranks joined on the respondent's ZIP code; years: 2023 release (version 4.0.1; built on ACS 2019-2023); weights: 2019-2023 tract population for counties
  - American Community Survey. ACS 2019-2023 tract population, the weight for county deprivation; years: 2019-2023 (5-year); weights: 2019-2023 tract population
  - JEC Social Capital Index. County rate of deaths of despair for 2015 (CDC multiple cause of death), as compiled by the Joint Economic Committee for 1,341 larger counties: a cross-check series whose population base and age adjustment are not documented; years: 2018 release (components from about 2010-2016); weights: none (county rate)
  - CDC WONDER underlying cause of death. CDC WONDER underlying cause of death: county drug, alcohol and suicide counts and populations, 2018-2023, complete counties only, to check the JEC rate; years: 2018-2024; weights: none (crude rates from counts)
  - USALEEP life expectancy. Life expectancy at birth, 2010-2015 deaths, 2020 tracts: the canonical variance split among states, counties and tracts, quoted from the chapter that owns it; years: 2010-2015; weights: 2019-2023 tract population
  - Published estimates from cited studies. One estimate from the Moving to Opportunity experiment (Ludwig and coauthors 2012) and two income thresholds from Kahneman and Deaton (2010) and Killingsworth and coauthors (2023), quoted beside their citations; years: as reported in each study; weights: as in the papers
  - HUD Fair Market Rents. HUD Fair Market Rents, fiscal year 2025, two-bedroom, by FMR area averaged to counties: a stand-in for relative price levels in the interview years; years: fiscal year 2025 (October 2024 to September 2025); weights: none (one rent per FMR area)
- Canonical URL: https://vishalsingh.org/zipcode/life-evaluation-and-daily-feelings
- Report: https://vishalsingh.org/zipcode/life-evaluation-and-daily-feelings/report
- Markdown: https://vishalsingh.org/zipcode/life-evaluation-and-daily-feelings/report.md
- BibTeX key: singh2026lifeevaluationanddailyfeelings

```bibtex
@misc{singh2026lifeevaluationanddailyfeelings,
  author = {Singh, Vishal},
  title = {Life evaluation and daily feelings},
  year = {2026},
  url = {https://vishalsingh.org/zipcode/life-evaluation-and-daily-feelings},
  note = {In ZIP Code Destiny. Published 9 October 2026.}
}
```

## Principal estimates

The facts the dek and the opening prose cite, in order of appearance.

| Estimand | Population | Period | Value | Interval | n | Weights | Source | Fact id |
| --- | --- | --- | ---: | --- | ---: | --- | --- | --- |
| difference between the unweighted mean ladder scores of the least and the most deprived tenth (county ADI is the mean of its tracts' ADI 2023 national ranks weighted by 2019-2023 tract population), interval clustered by the respondent's county | respondents in the two extreme tenths of interviews, ranked by the Area Deprivation Index of the respondent's county | interviews in 2009 to 2017 | 0.24 | 95% CI 0.18 to 0.31 | 442,036 respondents | none (unweighted) | Gallup U.S. Daily Poll | ladder_drop_county |
| gap between the extreme tenths after comparing respondents within cells of five-year age band (age capped at 85), sex, race, three-level education and household income bracket, interval clustered by the respondent's county | respondents with age, sex, race, education, household income and a county rent in the two extreme county tenths | interviews in 2009 to 2017 | 0.13 | 95% CI 0.10 to 0.16 | 350,199 respondents | none (unweighted) | Gallup U.S. Daily Poll | ladder_gap_adj_county |
| gap between the extreme tenths after comparing respondents within cells of five-year age band (age capped at 85), sex, race, three-level education and household income bracket, with a quintile of county rent also in the cells, interval clustered by the respondent's county | respondents with age, sex, race, education, household income and a county rent in the two extreme county tenths | interviews in 2009 to 2017 | −0.02 | 95% CI −0.05 to 0.00 | 350,199 respondents | none (unweighted) | HUD Fair Market Rents | ladder_gap_rentq |
| difference between the unweighted shares reporting the feeling a lot of the day yesterday in the most and the least deprived tenth of interviews, ranked by county ADI, interval clustered by the respondent's county | respondents with an answer to the physical pain item in the two extreme county tenths | interviews in the years the item was asked | 8.8 | 95% CI 7.9 to 9.7 | 526,330 respondents | none (unweighted) | Gallup U.S. Daily Poll | gap_physical_pain |
| the same difference after comparing respondents within cells of five-year age band (age capped at 85), sex, race, three-level education and household income bracket, interval clustered by the respondent's county | respondents with age, sex, race, education, household income and a county rent and an answer to the physical pain item in the two extreme county tenths | interviews in the years the item was asked | 1.8 | 95% CI 1.4 to 2.3 | 412,916 respondents | none (unweighted) | Gallup U.S. Daily Poll | gap_adj_physical_pain |
| the pain gap after comparing respondents within cells of five-year age band (age capped at 85), sex, race, three-level education and household income bracket, with a quintile of county rent also in the cells, interval clustered by the respondent's county | respondents with age, sex, race, education, household income, a county rent and an answer to the pain item in the two extreme county tenths | interviews 2008 to 2017 | 1.2 | 95% CI 0.8 to 1.7 | 412,916 respondents | none (unweighted) | HUD Fair Market Rents | pain_gap_rentq |
| correlation across counties between the county mean of the item and the JEC county rate of deaths of despair (CDC multiple cause of death, 2015), counties weighted by their respondents with the item, bootstrap over counties | counties with a deaths-of-despair rate and at least 100 respondents with each item | interviews 2008 to 2017; deaths of despair 2015 | +0.65 | 95% CI 0.61 to 0.69 | 1,337 counties | respondents with the item in each county | Gallup U.S. Daily Poll | r_des_pain |
| count of interviews in the extract (unweighted) | all interviews in the Gallup U.S. Daily extract | 2008 to 2017 | 3,530,270 |   |   | none (unweighted count) | Gallup U.S. Daily Poll | n_interviews_all |

## Findings

Gallup respondents in the most deprived tenth of counties rated their lives 0.24 (95% CI 0.18 to 0.31) ladder points lower than respondents in the least deprived tenth. At the same nominal household income, and the same age, sex, race and education, the order reverses, and the most deprived tenth scores 0.13 (95% CI 0.10 to 0.16) points higher. Allowing for local rent levels removes that difference. With a quintile of county rent added to the comparison it becomes −0.02 (95% CI −0.05 to 0.00) points, so these data cannot separate a place difference from a price difference.

Daily feelings behave differently from the ladder. Physical pain was more common in the most deprived tenth of counties by 8.8 (95% CI 7.9 to 9.7) points, by 1.8 (95% CI 1.4 to 2.3) points at the same characteristics and nominal income, and by 1.2 (95% CI 0.8 to 1.7) points with county rent also allowed for. Stress, in contrast, was less common there. Across larger counties, weighting each county by its respondents, the share reporting pain correlates +0.65 (95% CI 0.61 to 0.69) with the county rate of deaths of despair. That is the highest of the seven feelings under every weighting and comparison this chapter tries.

The data are the Gallup U.S. Daily survey: 3,530,270 interviews from 2008 to 2017, one row per interview, with new respondents every year and no one followed over time. The extract has no survey weights, so every figure is unweighted and levels can differ from the weighted figures Gallup publishes. The unit of this chapter is the person. The deprivation measures describe the county or ZIP area where the person lives, so a comparison between tenths of places is partly a comparison between kinds of people. Every interval is a 95% interval clustered by the respondent's county.

### Two kinds of question about a life

The Cantril ladder asks respondents to place their life on a ladder whose bottom rung is the worst possible life and whose top rung is the best possible one.[1] It records a judgment of a life as a whole, which is why this chapter calls it life evaluation. The extract holds that ladder and a second one for where the respondent expects to stand in the future, which this chapter calls the expected ladder. Gallup also asked whether respondents experienced each of seven feelings during a lot of the day yesterday: enjoyment, happiness, worry, stress, sadness, anger and physical pain. Those answers record a day, not a life, and the chapter calls them daily feelings. The two kinds of measure are analyzed side by side and never combined into one index.

Not every item was asked in every year (Table 1). Both ladder items are used for 2009 to 2017. In 2008 political questions preceded the ladder, which makes life evaluation run low, and Gallup removed them for part of the sample in early January 2009 and added a transition question in April 2009.[2] The feelings are not affected and start in 2008. Enjoyment, worry, stress and pain run to 2017, happiness and sadness stop after 2016, and anger was asked from 2008 to 2013 and again in 2016. The ladder reached 85% of the interviews in 2009 and 49% of those from 2013 to 2017, when each item went to a subsample, so the pooled ladder figures weight 2009 to 2012 most. The ladder is present in 2,217,485 interviews that also have a county.

Table 1. The ladder runs from 2009 to 2017 and the feelings from 2008, but happiness and sadness stop after 2016 and anger was not asked in 2014, 2015 or 2017

Place is attached to each respondent before any aggregation, and only aggregates are published. The county comes from the respondent's ZIP code and is approximate near county lines. 10,731 of the 3,530,270 interviews have no county and are dropped, which leaves 3,519,539 interviews in 3,142 counties. County deprivation is the mean of the Area Deprivation Index national ranks of the county's tracts, weighted by tract population, where a higher rank means more deprived. 784 interviews are in counties with no index.

ZIP-area deprivation is the index for the ZCTA that carries the ZIP code's number. A ZIP code is not a ZCTA and Connecticut has no ZCTAs in the project files, so this measure covers 97.6% of the interviews. Respondents are cut into tenths of each deprivation measure, so each tenth holds about a tenth of the interviews. Household income is the respondent's reported bracket, in nominal dollars, available for 81% of the interviews.

### The ladder falls with deprivation and reverses only at the same nominal income

Across tenths of county deprivation the mean ladder score falls from 7.17 (95% CI 7.11 to 7.23) in the least deprived tenth to 6.93 (95% CI 6.91 to 6.94) in the most deprived, and the point estimates fall from each tenth to the next (Table 2). The tenths are tenths of interviews, not of counties. The least deprived holds interviews from 45 counties and the most deprived from 1,205.

The ZIP-area measure gives a longer fall, from 7.29 (95% CI 7.24 to 7.34) to 6.87 (95% CI 6.85 to 6.88), a drop of 0.42 (95% CI 0.37 to 0.48) points against 0.24 (95% CI 0.18 to 0.31) for counties. A finer unit of place separates respondents further. Chapter 8.4, Place and the length of a life (https://vishalsingh.org/zipcode/place-and-the-length-of-a-life) draws the corresponding gradient in life expectancy for tracts.

Table 2. The ladder score is 0.24 points lower in the most deprived tenth of counties, 0.13 points higher at the same nominal household income and other characteristics, and −0.02 once county rent levels also enter

The tenths differ in who lives in them. The second and third tabs of the figure compare respondents only with others in the same cell of five-year age band, sex, race, education and household income bracket, using one sample at every step, the 1,759,964 interviews that report every one of those characteristics and a county rent. Education has three levels and ages above the oldest band share it. Comparing within cells of age and sex leaves the county gap at −0.30 (95% CI −0.36 to −0.23) points, close to its starting value of −0.27 (95% CI −0.34 to −0.21) on the same interviews. Adding education narrows it to −0.11 (95% CI −0.15 to −0.07), adding race and ethnicity changes it little, and adding household income turns it to +0.13 (95% CI 0.10 to 0.16).

In that comparison the adjusted mean ladder score is 7.10 (95% CI 7.08 to 7.11) in the most deprived tenth and 6.97 (95% CI 6.94 to 7.00) in the least deprived, so the ladder is 0.13 (95% CI 0.10 to 0.16) points higher in the most deprived tenth. The ZIP-area measure follows the same steps, from −0.21 (95% CI −0.25 to −0.17) with education to 0.07 (95% CI 0.04 to 0.10) higher with income. With a six-level education item the final gap is 0.13 (95% CI 0.10 to 0.16).

The last step assumes that one nominal bracket means one standard of living in every county, and the data do not support that. The interviews in the least deprived tenth come from counties where the HUD two-bedroom Fair Market Rent averages $2,697 (95% CI $2,594 to $2,800) a month, against $943 (95% CI $932 to $953) in the most deprived tenth. The rents are for fiscal year 2025 and stand in for relative price levels in 2008 to 2017, and counties in a metro share one value.

With a quintile of county rent added to the cells the gap is −0.02 (95% CI −0.05 to 0.00) points, and with the logarithm of rent as a linear control it is −0.04 (95% CI −0.07 to −0.01). At the ZIP-area scale the quintile version gives −0.06 (95% CI −0.09 to −0.03). Counting each county once, instead of each interview, gives +0.02 (95% CI −0.03 to 0.07). The sign of the reversal therefore depends on prices and on weighting, and the data cannot separate a place difference from a price difference.

Leaving the 2008 ladder in changes none of this. With it, the county drop is 0.26 (95% CI 0.19 to 0.33), the nominal-income gap +0.12 (95% CI 0.09 to 0.15), the rent-quintile gap −0.03 (95% CI −0.06 to 0.00) and the pooled income slope 0.256 (95% CI 0.252 to 0.259).

This comparison does not say that places have no bearing on how people rate their lives. Household income is one of the things a deprived place has less of, and the deprivation index is itself built from income, education, employment and housing, so comparing respondents within an income bracket removes part of what the index measures. In these data the ladder gradient across places is gone once respondents are compared with others of the same education and income, and whether a gap of the opposite sign remains depends on the price of living.

### Daily feelings follow place less evenly than the ladder

The seven feelings do not share one gradient (Table 3). Physical pain was reported by 20.8% (95% CI 20.0% to 21.7%) of respondents in the least deprived tenth of counties and 29.6% (95% CI 29.2% to 30.0%) in the most deprived, a gap of 8.8 (95% CI 7.9 to 9.7) points. The gap is 11.0 (95% CI 10.3 to 11.7) points when places are tenths of ZIP areas. Apart from pain, no feeling differs by more than 2.1 points between the extreme tenths.

Sadness differs by 1.9 (95% CI 0.7 to 3.2) points and stress by −2.1 (95% CI −3.1 to −1.2) points, so sadness is more common in the most deprived tenth and stress is less common there. Enjoyment differs by −1.2 (95% CI −2.1 to −0.4), happiness by −0.6 (95% CI −1.4 to 0.2), worry by −1.1 (95% CI −2.3 to 0.2) and anger by −0.7 (95% CI −1.7 to 0.3).

Table 3. Physical pain is 8.8 percentage points more common in the most deprived tenth of counties, the largest gap of the seven feelings, and 1.8 at the same characteristics and 1.2 with county rent also allowed for

Comparing respondents within cells at nominal income shrinks the pain gap to 1.8 (95% CI 1.4 to 2.3) points, so that 80% (95% CI 77% to 84%) of the raw gap lines up with who lives in each tenth, and the remainder keeps its direction.

The other gaps change as well. Sadness becomes −1.7 (95% CI −2.4 to −1.0) points, worry moves from −1.1 (95% CI −2.3 to 0.2) to −3.7 (95% CI −4.6 to −2.8), anger to −1.3 (95% CI −1.9 to −0.6), and stress stays below zero at −2.3 (95% CI −3.0 to −1.6).

Enjoyment and happiness move above zero, to 1.6 (95% CI 1.0 to 2.2) and 1.8 (95% CI 1.2 to 2.4). Those reversals are the nominal-income reversals of the ladder, and they go with it when county rent is allowed for. With rent allowed for, the pain gap becomes 1.2 (95% CI 0.8 to 1.7) points, 1.1 (95% CI 0.6 to 1.5) with the logarithm of rent as the control and 1.7 (95% CI 1.1 to 2.3) with each county counted once.

No other gap is above zero once rent is allowed for. Enjoyment is −0.7, happiness −0.5, worry 0.5, stress 0.4, sadness 0.5 and anger 0.4, each within a point of zero. Stress is less common in the most deprived tenth as interviewed and at nominal income, and these data do not say why.

### The ladder rises more steeply with income than the day does

The ladder score rises from 6.28 (95% CI 6.26 to 6.29) in the poorest household income bracket to 7.60 (95% CI 7.59 to 7.61) in the richest (Table 4). To compare slopes across items, the figure scales each item by its own standard deviation.

In those units the richest bracket sits 0.68 (95% CI 0.67 to 0.69) above the poorest on the ladder, against −0.51 (95% CI −0.52 to −0.50) for pain, −0.48 (95% CI −0.49 to −0.48) for sadness, −0.34 (95% CI −0.35 to −0.33) for worry and 0.37 (95% CI 0.37 to 0.38) for enjoyment. Stress differs by only −0.12 (95% CI −0.13 to −0.11).

Stress is common in the poorest bracket, 43.6% (95% CI 43.4% to 43.9%) of respondents against 37.7% (95% CI 37.3% to 38.0%) in the richest, and varies by at most 2.3 points across the other five brackets. Pain falls from 37.8% (95% CI 37.4% to 38.3%) of respondents in the poorest bracket to 16.1% (95% CI 15.9% to 16.3%) in the richest.

Table 4. The richest income bracket scores 0.68 standard deviations above the poorest on the ladder and −0.51 on physical pain, but only −0.12 on stress

Kahneman and Deaton found a similar split in the Gallup-Healthways Well-Being Index. Life evaluation kept rising with the logarithm of income, while a measure of emotional well-being rose with income and showed no further progress above about $75,000 a year.[3] Killingsworth, Kahneman and Mellers later reconciled that result with experience-sampling data in which well-being kept rising with income for most people and flattened above about $100,000 a year only for an unhappy minority.[4] The top bracket in the Gallup extract starts at $90,000, so these data cannot show whether any item flattens at higher incomes.

The expected ladder rises with income as the ladder now does, from 6.98 (95% CI 6.95 to 7.02) in the poorest bracket to 8.11 (95% CI 8.10 to 8.12) in the richest. Graham and Pinto, using Gallup surveys, report that optimism differs most by race and place among low-income Americans.[5]

### The income gradient is nearly the same in every kind of county

The ladder's income gradient is nearly the same in all three bands of county deprivation (Table 5). Across all respondents the ladder rises 0.250 (95% CI 0.246 to 0.253) points for each step up the six income brackets. With county fixed effects, which compare each respondent only with others in the same county, it rises 0.251 (95% CI 0.247 to 0.255) points per step. Within any one income bracket, the three bands of counties differ by at most 0.12 points in the mean ladder score. In the poorest bracket the most deprived three tenths of counties score −0.09 (95% CI −0.13 to −0.05) points relative to the least deprived three tenths, and in the richest bracket 0.03 (95% CI 0.01 to 0.05).

Table 5. The income gradient of the ladder is nearly the same in all three bands of counties, 0.250 points per bracket across respondents and 0.251 within counties

Counties hold very little of the variation. The share of the variance in the ladder score that lies between counties, with states included, is 0.51% (95% CI 0.46% to 0.56%), and the rest lies between people in the same county. Chapter 10.1, Gradients, slopes and variance (https://vishalsingh.org/zipcode/gradients-slopes-and-variance) splits the variance of tract life expectancy, where states hold 13.7% and counties within states 15.9%, together 29.7% (the parts are rounded). The two shares are not strictly comparable, because a tract mean averages many people and a ladder score is one person's answer, so the county share of life expectancy would fall if it were measured for people.

Pain separates the bands, mostly among the poorest. In the poorest bracket pain is 6.6 (95% CI 5.7 to 7.5) points more common in the most deprived three tenths of counties than in the least deprived three tenths, and in the richest bracket 1.7 (95% CI 1.2 to 2.1) points.

> Causal evidence: Whether an offer to move to a less poor neighborhood changes how happy adults say they are (Randomized housing-voucher experiment)
>
> The comparisons above are between different people, so they cannot say what a place does to the people in it. The Moving to Opportunity experiment assigned low-income families in public housing in five cities, by lottery, to be offered a voucher usable only in a lower-poverty neighborhood, to be offered an unrestricted voucher, or to stay where they were. Ludwig and coauthors surveyed the adults a decade or more later. Pooled over the two voucher groups, the offer of a move to a less disadvantaged area raised a standardized answer to one question about how happy things are these days by 0.098 (95% CI 0.022 to 0.174) standard deviations, with no detectable change in adult economic self-sufficiency.[6]
>
> That is the effect of an offer, not of a move, on one happiness question asked of adults who began in public housing in very poor neighborhoods, and the estimate is small. It shows that an offer to move to a lower-poverty neighborhood can change a self-reported well-being measure without changing income. It does not cover the ladder or daily feelings, and it does not say how much of the gradient across Gallup counties is a place effect.

### The ladder and the day move together, more tightly across counties

Respondents who reported happiness during a lot of the day yesterday rated their lives 1.57 (95% CI 1.55 to 1.59) ladder points higher than those who did not, and respondents who reported enjoyment rated them 1.46 (95% CI 1.44 to 1.47) higher (Table 6). Those who reported sadness rated their lives 1.33 (95% CI 1.32 to 1.35) points lower, and those who reported physical pain 0.91 (95% CI 0.90 to 0.93) lower.

Correlations are smaller than these gaps suggest, since enjoyment was reported by 86% of respondents and sadness by only 17%. Across respondents the ladder correlates +0.26 (95% CI 0.26 to 0.26) with enjoyment, −0.28 (95% CI −0.28 to −0.28) with worry, the strongest of the seven, −0.20 (95% CI −0.20 to −0.20) with pain and −0.18 (95% CI −0.18 to −0.18) with anger, the weakest.

Table 6. Across respondents the ladder correlates −0.28 with worry and +0.26 with enjoyment, and across county means it moves together with most feelings more tightly

Across counties the pairs move together more tightly. The county mean ladder correlates +0.59 (95% CI 0.56 to 0.63) with the county share reporting enjoyment, −0.62 (95% CI −0.66 to −0.58) with pain, −0.57 (95% CI −0.61 to −0.53) with sadness and −0.20 (95% CI −0.26 to −0.14) with stress, the one feeling whose county correlation is not larger than its correlation across respondents. County means need at least 100 respondents with each item. Averaging to counties removes variation between people, and a stronger correlation between averages is the usual result of doing so. It says nothing about how closely any one person's life rating and day are linked.

The measures also differ in what they follow. The ladder rises with income more steeply than any feeling does, and pain tracks both income and county deprivation, while stress follows neither much. The ladder answers a question about a life and the feelings answer a question about a day, and they share a direction without sharing a shape. Chapter 6.2, Civic life, loneliness and partnership (https://vishalsingh.org/zipcode/civic-life-loneliness-and-partnership) uses the same ladder to compare partnered and other adults, and Chapter 6.4, Well-being across the life span (https://vishalsingh.org/zipcode/well-being-across-the-life-span) takes up how the ladder and the feelings change with age, which this chapter holds fixed.

### Pain is the feeling that tracks county deaths of despair

Felt despair is not an item in the Gallup survey. This chapter reads it from the items Gallup does ask: the pain, sadness, worry and stress that respondents reported for yesterday, and the ladder now and the expected ladder. The comparison is with the county rate of deaths of despair for 2015, from the Joint Economic Committee's compilation of CDC data for larger counties. Deaths of despair, in Case and Deaton's term, are deaths from suicide, drugs and alcohol.[7] Case and Deaton documented rising reports of chronic pain alongside rising mortality from suicide, drugs and alcohol among middle-aged white non-Hispanic Americans from 1999 to 2013.[8] Chapter 8.2, Overdose, alcohol and suicide (https://vishalsingh.org/zipcode/overdose-alcohol-and-suicide) maps the deaths themselves.

The Joint Economic Committee's file keeps the 2015 rate as a cross-check series, and it does not document the rate's population base or whether it is age adjusted. It is used here because it is the only county rate from the Gallup years, and it can be checked against CDC WONDER. The check uses the complete counties of the chapter on overdose, alcohol and suicide, those whose drug, alcohol and suicide counts are published in every year from 2018 to 2023. On those counties the two rates correlate +0.81 (95% CI 0.77 to 0.84) across 534 counties, weighted by respondents. The WONDER rate sums three crude rates, includes suicides by drug poisoning in both the drug and the suicide rate, and comes a decade after the interviews.

The county share reporting pain correlates +0.69 (95% CI 0.64 to 0.74) with the Joint Economic Committee rate on those counties and +0.58 (95% CI 0.51 to 0.64) with the WONDER rate, and the share reporting stress −0.14 (95% CI −0.23 to −0.05) with the WONDER rate.

Respondents in the lowest fifth of the county rate reported physical pain at 20.2% (95% CI 19.8% to 20.6%) and respondents in the highest fifth at 26.8% (95% CI 26.3% to 27.3%), a difference of 6.6 (95% CI 6.0 to 7.3) points, and the share rises at each step (Table 7).

The ladder falls from 7.15 (95% CI 7.12 to 7.18) to 7.00 (95% CI 6.98 to 7.02). The share reporting stress hardly moves, 37.4% (95% CI 36.9% to 37.9%) in the lowest fifth and 36.4% (95% CI 36.1% to 36.8%) in the highest.

Table 7. Physical pain is the feeling most closely tied to the county deaths-of-despair rate under every weighting and comparison, with a correlation of +0.65 across counties

The second tab compares county averages with the rate in 1,337 larger counties. With counties weighted by their respondents, the county share reporting pain correlates +0.65 (95% CI 0.61 to 0.69) with the rate, the ladder now −0.40 (95% CI −0.47 to −0.33), the expected ladder −0.54 (95% CI −0.60 to −0.49) and sadness +0.31 (95% CI 0.20 to 0.41).

Worry (+0.06 (95% CI −0.05 to 0.17)) and stress (−0.09 (95% CI −0.17 to −0.01)) barely correlate under that weighting. Counting each county once changes that. Worry rises to +0.36 (95% CI 0.29 to 0.45), stress to +0.19 (95% CI 0.11 to 0.28) and sadness to +0.45 (95% CI 0.39 to 0.52), while pain remains the strongest at +0.58 (95% CI 0.54 to 0.63). Correlations that give every county equal weight are dominated by the many small counties, and those weighted by respondents by the large ones. The statement that stress and worry do not follow the despair map therefore holds for the counties where most respondents live and not for all counties.

County averages also mix places with people. Respondents in the highest fifth of the rate average 55.5 (95% CI 55.1 to 56.0) years of age against 51.9 (95% CI 51.3 to 52.5) in the lowest. Comparing respondents within cells of age and sex before averaging to counties leaves the pain correlation at +0.62 (95% CI 0.58 to 0.66), so age and sex do not account for it.

Comparing within cells of age, sex, race, education and household income brings it to +0.43 (95% CI 0.38 to 0.48), still the highest of the feelings and the only negative feeling whose interval stays above zero. The correlation of the ladder changes sign to +0.14 (95% CI 0.07 to 0.22) in that comparison, which includes household income and education, while the comparison on age and sex alone leaves it at −0.47 (95% CI −0.53 to −0.40). Pain remains tied to the county rate of deaths of despair after respondents are compared on age, sex, race, education and household income, and no other negative feeling does.

### The ladder gradient follows all four rankings of counties

Over the 3,121 counties that have every ranking, the ADI baseline gives a ladder score of 7.17 (95% CI 7.11 to 7.23) in the least deprived tenth of interviews and 6.93 (95% CI 6.91 to 6.94) in the most deprived, with a rank correlation of −0.98 between tenth and mean (Table 8). Under the standardized index, which applies the ADI's coefficients after standardizing each of its measures, the score runs from 7.22 (95% CI 7.19 to 7.25) to 6.94 (95% CI 6.92 to 6.96) with a correlation of −1.00. Ranking counties by median household income gives 7.22 (95% CI 7.19 to 7.24) and 6.94 (95% CI 6.92 to 6.95) (−1.00), and ranking them by the bachelor's degree share gives 7.24 (95% CI 7.21 to 7.26) and 6.90 (95% CI 6.89 to 6.92) (−0.96). Under the rule of Chapter 9.1, How far one ranking goes (https://vishalsingh.org/zipcode/how-far-one-ranking-goes) none of the three counts as a change, since the ends differ under each ranking and every correlation stays beyond 0.9 in absolute value.

Table 8. The ladder score is steadily worse toward the more deprived end under the ADI and under each of the three other rankings

### What these data cannot show

The unit of the chapter is the person, but the place measures describe counties and ZIP areas, so a gap between tenths of places mixes the places with the people in them. Comparing respondents within cells handles only the characteristics the survey records. It does not handle health, disability, work or family, which go with pain. Household income is in nominal dollars and the rent file is a single fiscal year 2025 standard shared within metros, so the price comparison is a rough proxy, and neither it nor the nominal comparison identifies what places do to residents. The randomized Moving to Opportunity experiment is the evidence on that question, and it covers neither the ladder nor these feelings.

The survey has no weights, and the interviews are a repeated cross-section, not a panel, so the chapter shows differences between people and not changes within them. Pooled figures weight 2009 to 2012 most for the ladder and 2008 to 2012 for the feelings, and the chapter does not remove differences between years. Each feeling is a yes or no answer to whether the respondent had it during a lot of the day yesterday, which records neither intensity nor duration, and the ladder is a whole-number rating whose rungs different people may anchor differently. Answers given in one interview can color each other, as the 2008 ladder shows, and the survey cannot separate that for the other items.

The county is derived from the ZIP code and is approximate near county lines. ZIP-area deprivation treats a ZIP code as the ZCTA of the same number and drops Connecticut. County deprivation is built from Area Deprivation Index ranks that include income and education, so the index overlaps with the respondent characteristics used in the comparisons. The index is the 2023 release, built on ACS data for 2019 to 2023, and the interviews are from 2008 to 2017, which assumes that the ranking of places persisted. The deaths-of-despair rate is a single year for larger counties, a cross-check series with an undocumented population base and age adjustment, and its correlations are across counties, so they do not show that the people who report pain are the people who die.

Chapter 4.5, Mental distress in young adults (https://vishalsingh.org/zipcode/mental-distress-in-young-adults) measures frequent mental distress in a different survey with survey weights, and its estimates are not directly comparable with the unweighted Gallup levels here.

## Tables

### Table 1. The ladder runs from 2009 to 2017 and the feelings from 2008, but happiness and sadness stop after 2016 and anger was not asked in 2014, 2015 or 2017

Gallup U.S. Daily items used in this chapter: the years each was asked, interviews with a county and an answer, and the share of the interviews in those years that carry the item

| Item | Kind | Years asked | Interviews with an answer | Share of interviews in those years (%) |
| --- | --- | --- | ---: | ---: |
| Ladder now | Life evaluation | 2009 to 2017 | 2,217,485 | 70 |
| Ladder in five years | Life evaluation | 2009 to 2017 | 2,124,998 | 67 |
| Enjoyment | Positive feeling | 2008 to 2017 | 2,625,590 | 75 |
| Happiness | Positive feeling | 2008 to 2016 | 2,464,754 | 78 |
| Worry | Negative feeling | 2008 to 2017 | 2,629,572 | 75 |
| Stress | Negative feeling | 2008 to 2017 | 2,629,276 | 75 |
| Sadness | Negative feeling | 2008 to 2016 | 2,469,420 | 78 |
| Anger | Negative feeling | 2008 to 2013 and 2016 | 2,099,828 | 85 |
| Physical pain | Negative feeling | 2008 to 2017 | 2,629,184 | 75 |

Note: The 2008 ladder answers are left out: political questions preceded the ladder in 2008 and the 2008 ladder runs low (Deaton 2012). Gallup removed the political questions for part of the sample in early January 2009 and added a transition question in April 2009. The feelings are not affected.

Note: Happiness and anger each have a few stray answers in years when the item was not asked; they are dropped.

Note: Feelings are answers to whether the respondent experienced the feeling during a lot of the day yesterday. The two ladder items are scored 0 to 10; the second asks where the respondent expects to stand in about five years.

Source: Gallup U.S. Daily, ladder 2009-2017 and feelings 2008-2017; unweighted counts of interviews with a county

Data: https://vishalsingh.org/zipcode/report/life-evaluation-and-daily-feelings/figures/item_coverage.json

### Table 2. The ladder score is 0.24 points lower in the most deprived tenth of counties, 0.13 points higher at the same nominal household income and other characteristics, and −0.02 once county rent levels also enter

Mean Cantril ladder score by decile of the deprivation of the respondent's county and ZIP area, as interviewed and after comparing respondents within cells, with the gap at each step

#### Table 2.1. As interviewed

The ladder score falls from 7.17 to 6.93 across deciles of county deprivation and from 7.29 to 6.87 across deciles of ZIP-area deprivation

Unweighted mean Cantril ladder score by decile of the Area Deprivation Index of the respondent's county and of the respondent's ZIP area, Gallup U.S. Daily interviews 2009 to 2017

| conf_high | conf_low | decile | estimate | n | n_units | place |
| ---: | ---: | ---: | ---: | ---: | ---: | --- |
| 7.23 | 7.11 | 1 | 7.17 | 218,776 | 45 | County deprivation |
| 7.17 | 7.09 | 2 | 7.13 | 220,210 | 85 | County deprivation |
| 7.15 | 7.08 | 3 | 7.11 | 225,715 | 110 | County deprivation |
| 7.14 | 7.07 | 4 | 7.10 | 215,982 | 134 | County deprivation |
| 7.12 | 7.07 | 5 | 7.10 | 220,790 | 168 | County deprivation |
| 7.12 | 7.06 | 6 | 7.09 | 223,139 | 198 | County deprivation |
| 7.08 | 7.03 | 7 | 7.06 | 222,438 | 266 | County deprivation |
| 7.06 | 7.02 | 8 | 7.04 | 223,727 | 350 | County deprivation |
| 7.00 | 6.96 | 9 | 6.98 | 222,937 | 563 | County deprivation |
| 6.94 | 6.91 | 10 | 6.93 | 223,260 | 1,205 | County deprivation |
| 7.34 | 7.24 | 1 | 7.29 | 213,516 | 203 | ZIP-area deprivation |
| 7.24 | 7.15 | 2 | 7.20 | 215,235 | 343 | ZIP-area deprivation |
| 7.19 | 7.13 | 3 | 7.16 | 217,052 | 465 | ZIP-area deprivation |
| 7.15 | 7.10 | 4 | 7.13 | 217,305 | 614 | ZIP-area deprivation |
| 7.12 | 7.08 | 5 | 7.10 | 216,938 | 767 | ZIP-area deprivation |
| 7.07 | 7.03 | 6 | 7.05 | 217,457 | 964 | ZIP-area deprivation |
| 7.04 | 7.01 | 7 | 7.03 | 217,491 | 1,197 | ZIP-area deprivation |
| 7.00 | 6.97 | 8 | 6.98 | 217,224 | 1,451 | ZIP-area deprivation |
| 6.93 | 6.91 | 9 | 6.92 | 217,621 | 1,656 | ZIP-area deprivation |
| 6.88 | 6.85 | 10 | 6.87 | 217,263 | 1,806 | ZIP-area deprivation |

#### Table 2.2. Same characteristics

Among respondents of the same age, sex, race, education and nominal household income, the ladder score is 7.10 in the most deprived county decile and 6.97 in the least

Mean ladder score after comparing respondents within cells of five-year age band, sex, race, three-level education and household income bracket, by decile of county and ZIP-area deprivation, 2009 to 2017

| conf_high | conf_low | decile | estimate | n | n_units | place |
| ---: | ---: | ---: | ---: | ---: | ---: | --- |
| 7.00 | 6.94 | 1 | 6.97 | 173,060 | 44 | County deprivation |
| 6.99 | 6.93 | 2 | 6.96 | 175,650 | 85 | County deprivation |
| 7.01 | 6.97 | 3 | 6.99 | 179,718 | 109 | County deprivation |
| 7.03 | 6.98 | 4 | 7.00 | 171,208 | 134 | County deprivation |
| 7.04 | 6.99 | 5 | 7.01 | 174,311 | 168 | County deprivation |
| 7.06 | 7.03 | 6 | 7.05 | 177,633 | 198 | County deprivation |
| 7.07 | 7.03 | 7 | 7.05 | 176,592 | 266 | County deprivation |
| 7.08 | 7.05 | 8 | 7.07 | 177,789 | 350 | County deprivation |
| 7.09 | 7.05 | 9 | 7.07 | 176,681 | 563 | County deprivation |
| 7.11 | 7.08 | 10 | 7.10 | 177,139 | 1,205 | County deprivation |
| 7.02 | 6.97 | 1 | 6.99 | 169,075 | 199 | ZIP-area deprivation |
| 7.01 | 6.97 | 2 | 6.99 | 172,355 | 337 | ZIP-area deprivation |
| 7.02 | 6.99 | 3 | 7.00 | 174,458 | 454 | ZIP-area deprivation |
| 7.02 | 6.99 | 4 | 7.01 | 174,430 | 607 | ZIP-area deprivation |
| 7.04 | 7.01 | 5 | 7.03 | 174,417 | 755 | ZIP-area deprivation |
| 7.04 | 7.01 | 6 | 7.03 | 174,768 | 954 | ZIP-area deprivation |
| 7.06 | 7.04 | 7 | 7.05 | 174,854 | 1,179 | ZIP-area deprivation |
| 7.06 | 7.04 | 8 | 7.05 | 174,142 | 1,432 | ZIP-area deprivation |
| 7.07 | 7.04 | 9 | 7.05 | 174,544 | 1,632 | ZIP-area deprivation |
| 7.08 | 7.05 | 10 | 7.06 | 174,290 | 1,781 | ZIP-area deprivation |

#### Table 2.3. Gap after each adjustment

The ladder gap between the most and least deprived county tenth narrows to −0.11 with education, turns to +0.13 with nominal household income and is −0.02 once county rent levels enter

Difference in mean ladder score, most deprived minus least deprived tenth of interviews, after successive comparisons within cells of respondent characteristics, 2009 to 2017

| conf_high | conf_low | estimate | n | n_units | place | step |
| ---: | ---: | ---: | ---: | ---: | --- | --- |
| −0.205 | −0.335 | −0.270 | 350,199 | 1,249 | County deprivation | No adjustment |
| −0.234 | −0.360 | −0.297 | 350,199 | 1,249 | County deprivation | Same age and sex |
| −0.069 | −0.151 | −0.110 | 350,199 | 1,249 | County deprivation | Plus education |
| −0.039 | −0.134 | −0.087 | 350,199 | 1,249 | County deprivation | Plus race and ethnicity |
| 0.157 | 0.099 | 0.128 | 350,199 | 1,249 | County deprivation | Plus household income |
| 0.004 | −0.054 | −0.025 | 350,199 | 1,249 | County deprivation | Plus county rent level |
| −0.394 | −0.499 | −0.446 | 343,365 | 1,917 | ZIP-area deprivation | No adjustment |
| −0.413 | −0.515 | −0.464 | 343,365 | 1,917 | ZIP-area deprivation | Same age and sex |
| −0.166 | −0.250 | −0.208 | 343,365 | 1,917 | ZIP-area deprivation | Plus education |
| −0.156 | −0.243 | −0.200 | 343,365 | 1,917 | ZIP-area deprivation | Plus race and ethnicity |
| 0.102 | 0.040 | 0.071 | 343,365 | 1,917 | ZIP-area deprivation | Plus household income |
| −0.034 | −0.092 | −0.063 | 343,365 | 1,917 | ZIP-area deprivation | Plus county rent level |

Note: Gallup's extract has no survey weights, so every estimate is unweighted and levels can differ from the weighted figures Gallup publishes. County comes from the respondent's ZIP code and is approximate near county lines. The ladder is the evaluative measure and the feelings are experiential measures; they are never combined into one index.

Note: Deciles are cut once on every interview that has the deprivation measure, so each holds about a tenth of interviews and every item uses the same cut points. County deprivation is the mean of the tract Area Deprivation Index national ranks (1 = least deprived), weighted by tract population. The ZIP code is treated as the ZCTA of the same number, whose index is an area-weighted approximation; Connecticut has no ZCTAs in the project files and drops out of the ZIP-area line.

Note: The second and third tabs use one sample at every step, the interviews that report age, sex, race, education, household income and a county rent. Education has three levels and ages above the oldest five-year band share it. Household income is in nominal dollars and does not reflect differences in the cost of living across counties; the top bracket has no upper limit.

Note: The rent step adds a quintile of the county's HUD two-bedroom Fair Market Rent for fiscal year 2025, which stands in for relative price levels in the interview years. Counties in a metro share the metro's rent. Where the rent is allowed for, the data cannot separate a place difference from a price difference.

Source: Gallup U.S. Daily, ladder 2009-2017; unweighted, intervals clustered by county; Area Deprivation Index 2023 and ACS 2019-2023 tract population joined on the respondent's county or ZIP code; unweighted

Data: https://vishalsingh.org/zipcode/report/life-evaluation-and-daily-feelings/figures/ladder_by_deprivation.json

### Table 3. Physical pain is 8.8 percentage points more common in the most deprived tenth of counties, the largest gap of the seven feelings, and 1.8 at the same characteristics and 1.2 with county rent also allowed for

Difference in the share reporting each feeling a lot of the day yesterday, most deprived minus least deprived tenth of interviews by county deprivation, Gallup U.S. Daily 2008 to 2017

| basis | conf_high | conf_low | estimate | feeling | n | n_units |
| --- | ---: | ---: | ---: | --- | ---: | ---: |
| As interviewed | −0.350 | −2.10 | −1.23 | Enjoyment | 525,461 | 1,250 |
| Same age, sex, race, education and nominal income | 2.16 | 0.990 | 1.58 | Enjoyment | 412,511 | 1,249 |
| Plus county rent level | −0.090 | −1.23 | −0.660 | Enjoyment | 412,511 | 1,249 |
| As interviewed | 0.200 | −1.41 | −0.610 | Happiness | 493,755 | 1,250 |
| Same age, sex, race, education and nominal income | 2.43 | 1.17 | 1.76 | Happiness | 385,768 | 1,249 |
| Plus county rent level | 0.150 | −1.09 | −0.490 | Happiness | 385,768 | 1,249 |
| As interviewed | 0.190 | −2.33 | −1.07 | Worry | 526,425 | 1,250 |
| Same age, sex, race, education and nominal income | −2.78 | −4.62 | −3.72 | Worry | 413,010 | 1,249 |
| Plus county rent level | 1.48 | −0.400 | 0.520 | Worry | 413,010 | 1,249 |
| As interviewed | −1.19 | −3.08 | −2.13 | Stress | 526,330 | 1,250 |
| Same age, sex, race, education and nominal income | −1.58 | −3 | −2.33 | Stress | 412,916 | 1,249 |
| Plus county rent level | 1.18 | −0.310 | 0.370 | Stress | 412,916 | 1,249 |
| As interviewed | 3.19 | 0.660 | 1.93 | Sadness | 494,753 | 1,250 |
| Same age, sex, race, education and nominal income | −0.960 | −2.40 | −1.73 | Sadness | 386,323 | 1,249 |
| Plus county rent level | 1.25 | −0.190 | 0.480 | Sadness | 386,323 | 1,249 |
| As interviewed | 0.250 | −1.67 | −0.710 | Anger | 421,120 | 1,250 |
| Same age, sex, race, education and nominal income | −0.630 | −1.87 | −1.29 | Anger | 326,006 | 1,248 |
| Plus county rent level | 1.10 | −0.230 | 0.400 | Anger | 326,006 | 1,248 |
| As interviewed | 9.68 | 7.87 | 8.77 | Physical pain | 526,330 | 1,250 |
| Same age, sex, race, education and nominal income | 2.26 | 1.38 | 1.81 | Physical pain | 412,916 | 1,249 |
| Plus county rent level | 1.70 | 0.820 | 1.23 | Physical pain | 412,916 | 1,249 |

Note: Gallup's extract has no survey weights, so every estimate is unweighted and levels can differ from the weighted figures Gallup publishes. County comes from the respondent's ZIP code and is approximate near county lines. The ladder is the evaluative measure and the feelings are experiential measures; they are never combined into one index.

Note: Each feeling is counted in the years its item was asked (see the table of items). The two adjusted dots use one sample, the interviews that report every characteristic and a county rent, so it is smaller. Deprivation is the county Area Deprivation Index, in deciles cut on all interviews. The rent is the FY2025 HUD two-bedroom Fair Market Rent of the county, a stand-in for relative price levels in 2008 to 2017.

Source: Gallup U.S. Daily, 2008-2017; unweighted, intervals clustered by county; Area Deprivation Index 2023 and ACS 2019-2023 tract population joined on the respondent's county or ZIP code

Data: https://vishalsingh.org/zipcode/report/life-evaluation-and-daily-feelings/figures/feelings_by_deprivation.json

### Table 4. The richest income bracket scores 0.68 standard deviations above the poorest on the ladder and −0.51 on physical pain, but only −0.12 on stress

Unweighted mean of each item in standard deviations of the item, by household income bracket, Gallup U.S. Daily interviews (ladder 2009 to 2017, feelings 2008 to 2017)

| Item | Income bracket (1 = lowest) | Mean in standard deviations | Raw mean (ladder points or share reporting) | Respondents |
| --- | ---: | ---: | ---: | ---: |
| Ladder now | 1 | −0.41 | 627.82 | 373,989 |
| Ladder now | 2 | −0.19 | 670.23 | 241,269 |
| Ladder now | 3 | −0.07 | 693.71 | 209,896 |
| Ladder now | 4 | 0.02 | 711.83 | 194,839 |
| Ladder now | 5 | 0.12 | 729.75 | 333,711 |
| Ladder now | 6 | 0.27 | 760.19 | 453,101 |
| Enjoyment | 1 | −0.26 | 76.78 | 440,439 |
| Enjoyment | 2 | −0.05 | 84.03 | 282,706 |
| Enjoyment | 3 | 0.02 | 86.46 | 247,271 |
| Enjoyment | 4 | 0.05 | 87.63 | 227,521 |
| Enjoyment | 5 | 0.09 | 88.86 | 390,056 |
| Enjoyment | 6 | 0.11 | 89.82 | 527,783 |
| Worry | 1 | 0.26 | 40.73 | 441,514 |
| Worry | 2 | 0.06 | 31.70 | 283,002 |
| Worry | 3 | −0.01 | 28.55 | 247,491 |
| Worry | 4 | −0.05 | 26.87 | 227,727 |
| Worry | 5 | −0.07 | 25.75 | 390,212 |
| Worry | 6 | −0.08 | 25.45 | 528,050 |
| Stress | 1 | 0.15 | 43.64 | 441,398 |
| Stress | 2 | 0.02 | 37.17 | 283,002 |
| Stress | 3 | −0.01 | 35.66 | 247,449 |
| Stress | 4 | −0.02 | 35.36 | 227,700 |
| Stress | 5 | −0.01 | 35.77 | 390,164 |
| Stress | 6 | 0.03 | 37.67 | 528,004 |
| Sadness | 1 | 0.33 | 28.83 | 419,214 |
| Sadness | 2 | 0.07 | 19.08 | 266,659 |
| Sadness | 3 | −0.02 | 15.85 | 235,369 |
| Sadness | 4 | −0.07 | 13.85 | 214,543 |
| Sadness | 5 | −0.12 | 12.12 | 363,770 |
| Sadness | 6 | −0.15 | 10.83 | 479,863 |
| Physical pain | 1 | 0.32 | 37.82 | 441,363 |
| Physical pain | 2 | 0.10 | 28.12 | 282,945 |
| Physical pain | 3 | 0.01 | 24.35 | 247,422 |
| Physical pain | 4 | −0.06 | 21.60 | 227,696 |
| Physical pain | 5 | −0.12 | 19.07 | 390,175 |
| Physical pain | 6 | −0.18 | 16.11 | 528,048 |

Note: Gallup's extract has no survey weights, so every estimate is unweighted and levels can differ from the weighted figures Gallup publishes. County comes from the respondent's ZIP code and is approximate near county lines. The ladder is the evaluative measure and the feelings are experiential measures; they are never combined into one index.

Note: Each item is scaled by its own mean and standard deviation over all respondents with the item, so the lines compare slopes, not levels, and the ladder is not combined with the feelings. Happiness and anger, asked in fewer years, are left out of this chart. The top bracket has no upper limit, so the chart cannot show whether an item flattens at higher incomes. Household income is reported for most but not all interviews, and brackets are in nominal dollars.

Source: Gallup U.S. Daily, ladder 2009-2017 and feelings 2008-2017; unweighted, intervals clustered by county; household income is the respondent's reported bracket

Data: https://vishalsingh.org/zipcode/report/life-evaluation-and-daily-feelings/figures/income_gradient.json

### Table 5. The income gradient of the ladder is nearly the same in all three bands of counties, 0.250 points per bracket across respondents and 0.251 within counties

Mean ladder score and share reporting physical pain by household income bracket, in three bands of county deprivation, Gallup U.S. Daily interviews (ladder 2009 to 2017, pain 2008 to 2017)

#### Table 5.1. Ladder score

Within each household income bracket, the ladder score differs by at most 0.12 points across three bands of county deprivation

Unweighted mean ladder score by household income bracket, in three bands of county deprivation (tenths 1 to 3, 4 to 7 and 8 to 10 of interviews), Gallup U.S. Daily 2009 to 2017

| bracket | conf_high | conf_low | estimate | n | n_units | place |
| ---: | ---: | ---: | ---: | ---: | ---: | --- |
| 1 | 6.37 | 6.29 | 6.33 | 86,715 | 237 | Least deprived three tenths of counties |
| 2 | 6.72 | 6.67 | 6.69 | 58,576 | 236 | Least deprived three tenths of counties |
| 3 | 6.92 | 6.88 | 6.90 | 54,352 | 238 | Least deprived three tenths of counties |
| 4 | 7.08 | 7.03 | 7.06 | 54,434 | 236 | Least deprived three tenths of counties |
| 5 | 7.26 | 7.21 | 7.24 | 105,652 | 238 | Least deprived three tenths of counties |
| 6 | 7.60 | 7.57 | 7.58 | 183,790 | 237 | Least deprived three tenths of counties |
| 1 | 6.31 | 6.26 | 6.29 | 139,826 | 763 | Middle four tenths |
| 2 | 6.72 | 6.68 | 6.70 | 95,714 | 763 | Middle four tenths |
| 3 | 6.95 | 6.91 | 6.93 | 85,958 | 761 | Middle four tenths |
| 4 | 7.14 | 7.10 | 7.12 | 80,130 | 760 | Middle four tenths |
| 5 | 7.32 | 7.29 | 7.30 | 137,439 | 763 | Middle four tenths |
| 6 | 7.63 | 7.60 | 7.61 | 178,686 | 763 | Middle four tenths |
| 1 | 6.26 | 6.22 | 6.24 | 147,372 | 2,109 | Most deprived three tenths |
| 2 | 6.72 | 6.69 | 6.71 | 86,910 | 2,104 | Most deprived three tenths |
| 3 | 6.99 | 6.95 | 6.97 | 69,541 | 2,092 | Most deprived three tenths |
| 4 | 7.19 | 7.15 | 7.17 | 60,226 | 2,078 | Most deprived three tenths |
| 5 | 7.37 | 7.34 | 7.36 | 90,515 | 2,097 | Most deprived three tenths |
| 6 | 7.63 | 7.60 | 7.62 | 90,545 | 2,090 | Most deprived three tenths |

#### Table 5.2. Physical pain

In the poorest income bracket, physical pain is 6.6 percentage points more common in the most deprived three tenths of counties than in the least deprived three tenths, against 1.7 in the richest bracket

Unweighted share reporting physical pain a lot of the day yesterday by household income bracket, in three bands of county deprivation, Gallup U.S. Daily 2008 to 2017

| bracket | conf_high | conf_low | estimate | n | n_units | place |
| ---: | ---: | ---: | ---: | ---: | ---: | --- |
| 1 | 35.02 | 33.38 | 34.20 | 102,306 | 237 | Least deprived three tenths of counties |
| 2 | 27.73 | 26.35 | 27.04 | 68,623 | 237 | Least deprived three tenths of counties |
| 3 | 24.53 | 23.38 | 23.95 | 64,124 | 238 | Least deprived three tenths of counties |
| 4 | 21.60 | 20.44 | 21.02 | 63,809 | 236 | Least deprived three tenths of counties |
| 5 | 19.54 | 18.78 | 19.16 | 124,159 | 238 | Least deprived three tenths of counties |
| 6 | 16.05 | 15.35 | 15.70 | 215,333 | 237 | Least deprived three tenths of counties |
| 1 | 37.55 | 36.26 | 36.91 | 164,464 | 763 | Middle four tenths |
| 2 | 28.16 | 27.20 | 27.68 | 112,141 | 763 | Middle four tenths |
| 3 | 24.53 | 23.68 | 24.11 | 101,201 | 762 | Middle four tenths |
| 4 | 21.77 | 20.97 | 21.37 | 93,670 | 761 | Middle four tenths |
| 5 | 19.13 | 18.50 | 18.82 | 160,630 | 763 | Middle four tenths |
| 6 | 16.22 | 15.58 | 15.90 | 207,669 | 764 | Middle four tenths |
| 1 | 41.26 | 40.34 | 40.80 | 174,503 | 2,111 | Most deprived three tenths |
| 2 | 29.71 | 28.98 | 29.34 | 102,096 | 2,106 | Most deprived three tenths |
| 3 | 25.30 | 24.64 | 24.97 | 82,038 | 2,096 | Most deprived three tenths |
| 4 | 22.77 | 22.09 | 22.43 | 70,157 | 2,088 | Most deprived three tenths |
| 5 | 19.64 | 19.06 | 19.35 | 105,265 | 2,098 | Most deprived three tenths |
| 6 | 17.65 | 17.08 | 17.37 | 104,953 | 2,096 | Most deprived three tenths |

Note: Gallup's extract has no survey weights, so every estimate is unweighted and levels can differ from the weighted figures Gallup publishes. County comes from the respondent's ZIP code and is approximate near county lines. The ladder is the evaluative measure and the feelings are experiential measures; they are never combined into one index.

Note: Bands cut the deprivation deciles into the least deprived three, the middle four and the most deprived three tenths of interviews by county Area Deprivation Index. The slopes are coefficients of the ladder score on the bracket number (1 to 6) with errors clustered on county; the second has county fixed effects. Brackets are not equal in dollar width.

Source: Gallup U.S. Daily, ladder 2009-2017 and feelings 2008-2017; unweighted, intervals clustered by county; Area Deprivation Index 2023 and ACS 2019-2023 tract population joined on the respondent's county or ZIP code

Data: https://vishalsingh.org/zipcode/report/life-evaluation-and-daily-feelings/figures/income_within_places.json

### Table 6. Across respondents the ladder correlates −0.28 with worry and +0.26 with enjoyment, and across county means it moves together with most feelings more tightly

Cantril ladder scores against feelings yesterday, by whether the respondent reported each feeling, and the correlation of the two across respondents and across county means

#### Table 6.1. Ladder by feeling reported

Respondents who reported happiness rated their lives 1.57 ladder points higher, and those who reported sadness rated them 1.33 lower

Unweighted mean ladder score of respondents who did and did not report each feeling a lot of the day yesterday, Gallup U.S. Daily (ladder 2009 to 2017, feelings 2008 to 2017)

| feeling | group | n_no | n_yes | no | no_hi | no_lo | yes | yes_hi | yes_lo |
| --- | --- | ---: | ---: | ---: | ---: | ---: | ---: | ---: | ---: |
| Enjoyment | Positive feelings | 310,895 | 1,900,405 | 5.82 | 5.84 | 5.80 | 7.28 | 7.29 | 7.27 |
| Happiness | Positive feelings | 230,060 | 1,821,129 | 5.67 | 5.69 | 5.65 | 7.24 | 7.25 | 7.23 |
| Worry | Negative feelings | 1,573,559 | 640,621 | 7.42 | 7.43 | 7.41 | 6.22 | 6.24 | 6.20 |
| Stress | Negative feelings | 1,407,513 | 806,482 | 7.43 | 7.44 | 7.43 | 6.44 | 6.46 | 6.42 |
| Sadness | Negative feelings | 1,713,579 | 341,114 | 7.29 | 7.29 | 7.28 | 5.95 | 5.97 | 5.93 |
| Anger | Negative feelings | 1,483,704 | 202,942 | 7.18 | 7.19 | 7.17 | 6.11 | 6.13 | 6.09 |
| Physical pain | Negative feelings | 1,677,403 | 536,374 | 7.29 | 7.30 | 7.28 | 6.38 | 6.39 | 6.37 |

#### Table 6.2. Correlations

The ladder correlates +0.26 with enjoyment across respondents and +0.59 across county means, and −0.20 with pain across respondents and −0.62 across county means

Correlation of the Cantril ladder score with each feeling, across respondents and across county means, Gallup U.S. Daily (ladder 2009 to 2017, feelings 2008 to 2017)

| basis | conf_high | conf_low | estimate | feeling | n | n_units |
| --- | ---: | ---: | ---: | --- | ---: | ---: |
| Across respondents | 0.262 | 0.258 | 0.260 | Enjoyment | 2,211,300 | 3,141 |
| Across county means | 0.630 | 0.558 | 0.595 | Enjoyment | 2,175,289 | 2,356 |
| Across respondents | 0.257 | 0.253 | 0.255 | Happiness | 2,051,189 | 3,140 |
| Across county means | 0.588 | 0.499 | 0.544 | Happiness | 2,175,189 | 2,355 |
| Across respondents | −0.277 | −0.282 | −0.279 | Worry | 2,214,180 | 3,141 |
| Across county means | −0.323 | −0.436 | −0.381 | Worry | 2,175,289 | 2,356 |
| Across respondents | −0.244 | −0.249 | −0.247 | Stress | 2,213,995 | 3,141 |
| Across county means | −0.138 | −0.263 | −0.201 | Stress | 2,175,289 | 2,356 |
| Across respondents | −0.253 | −0.257 | −0.255 | Sadness | 2,054,693 | 3,140 |
| Across county means | −0.528 | −0.612 | −0.572 | Sadness | 2,175,189 | 2,355 |
| Across respondents | −0.178 | −0.182 | −0.180 | Anger | 1,686,646 | 3,139 |
| Across county means | −0.279 | −0.394 | −0.337 | Anger | 2,169,173 | 2,298 |
| Across respondents | −0.199 | −0.204 | −0.201 | Physical pain | 2,213,777 | 3,141 |
| Across county means | −0.583 | −0.662 | −0.623 | Physical pain | 2,175,289 | 2,356 |

Note: Gallup's extract has no survey weights, so every estimate is unweighted and levels can differ from the weighted figures Gallup publishes. County comes from the respondent's ZIP code and is approximate near county lines. The ladder is the evaluative measure and the feelings are experiential measures; they are never combined into one index.

Note: The ladder and the feelings were asked of the same respondents, so the comparison uses interviews with both answers. County means need at least the county floor of respondents with each item. Anger is asked in fewer years than the other feelings (see the table of items).

Source: Gallup U.S. Daily, ladder 2009-2017 and feelings 2008-2017; unweighted, intervals clustered by county

Data: https://vishalsingh.org/zipcode/report/life-evaluation-and-daily-feelings/figures/ladder_and_feelings.json

### Table 7. Physical pain is the feeling most closely tied to the county deaths-of-despair rate under every weighting and comparison, with a correlation of +0.65 across counties

Feelings and life ratings of Gallup U.S. Daily respondents against the deaths-of-despair rate of their county (ladder 2009 to 2017, feelings 2008 to 2017)

#### Table 7.1. By fifth of the county rate

The share reporting physical pain rises from 20.2% to 26.8% across fifths of the county deaths-of-despair rate, while stress stays near 37.4%

Unweighted share reporting each feeling a lot of the day yesterday, by fifth of interviews ranked by the deaths-of-despair rate of the respondent's county, Gallup U.S. Daily 2008 to 2017

| conf_high | conf_low | estimate | measure | n | n_units | quintile |
| ---: | ---: | ---: | --- | ---: | ---: | ---: |
| 20.55 | 19.78 | 20.16 | Physical pain | 465,440 | 150 | 1 |
| 22.69 | 21.60 | 22.15 | Physical pain | 466,441 | 190 | 2 |
| 24.28 | 23.45 | 23.86 | Physical pain | 474,417 | 278 | 3 |
| 25.77 | 24.80 | 25.28 | Physical pain | 470,320 | 277 | 4 |
| 27.32 | 26.27 | 26.80 | Physical pain | 474,232 | 446 | 5 |
| 16.44 | 15.29 | 15.86 | Sadness | 434,567 | 150 | 1 |
| 17.05 | 15.38 | 16.22 | Sadness | 436,885 | 190 | 2 |
| 16.51 | 15.98 | 16.25 | Sadness | 445,971 | 278 | 3 |
| 17.36 | 16.76 | 17.06 | Sadness | 442,076 | 277 | 4 |
| 17.72 | 17.18 | 17.45 | Sadness | 447,105 | 446 | 5 |
| 37.92 | 36.88 | 37.40 | Stress | 465,362 | 150 | 1 |
| 37.34 | 36.25 | 36.79 | Stress | 466,398 | 190 | 2 |
| 36.39 | 35.62 | 36 | Stress | 474,426 | 278 | 3 |
| 36.50 | 35.85 | 36.17 | Stress | 470,344 | 277 | 4 |
| 36.83 | 36.06 | 36.45 | Stress | 474,370 | 446 | 5 |

#### Table 7.2. Correlations across counties

Across counties, the share reporting pain correlates +0.65 with the deaths-of-despair rate, the highest of the feelings, and +0.43 once respondents are compared within cells

Correlation across counties between the county mean of each item and the county rate of deaths of despair (CDC, 2015, as compiled by the Joint Economic Committee), counties with enough respondents, Gallup U.S. Daily (ladder 2009 to 2017, feelings 2008 to 2017)

| conf_high | conf_low | estimate | group | measure | n_units | respondents |
| ---: | ---: | ---: | --- | --- | ---: | ---: |
| 0.691 | 0.615 | 0.654 | Weighted by respondents | Physical pain | 1,337 | 2350453 |
| 0.627 | 0.536 | 0.582 | Each county once | Physical pain | 1,337 | 2350453 |
| 0.480 | 0.380 | 0.433 | Weighted, same age, sex, race, education and income | Physical pain | 1,331 | 1844740 |
| 0.410 | 0.197 | 0.308 | Weighted by respondents | Sadness | 1,337 | 2206239 |
| 0.522 | 0.385 | 0.453 | Each county once | Sadness | 1,337 | 2206239 |
| 0.061 | −0.118 | −0.030 | Weighted, same age, sex, race, education and income | Sadness | 1,331 | 1722909 |
| 0.167 | −0.046 | 0.059 | Weighted by respondents | Worry | 1,337 | 2350793 |
| 0.447 | 0.286 | 0.364 | Each county once | Worry | 1,337 | 2350793 |
| −0.048 | −0.213 | −0.134 | Weighted, same age, sex, race, education and income | Worry | 1,331 | 1845031 |
| 0.113 | −0.099 | 0.007 | Weighted by respondents | Anger | 1,337 | 1875987 |
| 0.331 | 0.158 | 0.244 | Each county once | Anger | 1,337 | 1875987 |
| 0.029 | −0.145 | −0.062 | Weighted, same age, sex, race, education and income | Anger | 1,331 | 1452370 |
| −0.009 | −0.169 | −0.088 | Weighted by respondents | Stress | 1,337 | 2350501 |
| 0.282 | 0.110 | 0.194 | Each county once | Stress | 1,337 | 2350501 |
| −0.001 | −0.151 | −0.077 | Weighted, same age, sex, race, education and income | Stress | 1,331 | 1844805 |
| 0.010 | −0.221 | −0.102 | Weighted by respondents | Happiness | 1,337 | 2202015 |
| −0.249 | −0.420 | −0.332 | Each county once | Happiness | 1,337 | 2202015 |
| 0.272 | 0.104 | 0.192 | Weighted, same age, sex, race, education and income | Happiness | 1,331 | 1720588 |
| −0.027 | −0.245 | −0.133 | Weighted by respondents | Enjoyment | 1,337 | 2347130 |
| −0.272 | −0.447 | −0.358 | Each county once | Enjoyment | 1,337 | 2347130 |
| 0.230 | 0.062 | 0.149 | Weighted, same age, sex, race, education and income | Enjoyment | 1,331 | 1843168 |
| −0.326 | −0.475 | −0.405 | Weighted by respondents | Ladder now | 1,337 | 1981094 |
| −0.355 | −0.481 | −0.420 | Each county once | Ladder now | 1,337 | 1981094 |
| 0.217 | 0.067 | 0.143 | Weighted, same age, sex, race, education and income | Ladder now | 1,331 | 1570386 |
| −0.488 | −0.596 | −0.542 | Weighted by respondents | Ladder in five years | 1,337 | 1900184 |
| −0.391 | −0.509 | −0.452 | Each county once | Ladder in five years | 1,337 | 1900184 |
| −0.045 | −0.172 | −0.108 | Weighted, same age, sex, race, education and income | Ladder in five years | 1,331 | 1522554 |

Note: Gallup's extract has no survey weights, so every estimate is unweighted and levels can differ from the weighted figures Gallup publishes. County comes from the respondent's ZIP code and is approximate near county lines. The ladder is the evaluative measure and the feelings are experiential measures; they are never combined into one index.

Note: The deaths-of-despair rate is a single 2015 cross-section covering the larger counties (suicide, alcohol and drug deaths, CDC multiple cause of death, compiled by the Joint Economic Committee). The file keeps it as a cross-check series and does not document its population base or age adjustment, so it is used here only to rank counties; the text checks it against CDC WONDER rates for 2018 to 2023. Fifths are cut on interviews in counties with a rate.

Note: County means need a floor of respondents with each item, and the same counties enter every dot. The third dot of each row compares respondents within cells of five-year age band, sex, race, three-level education and household income bracket before averaging to counties, so it uses interviews that report every characteristic.

Source: Gallup U.S. Daily, ladder 2009-2017 and feelings 2008-2017; unweighted, intervals clustered by county; county deaths of despair from the Joint Economic Committee compilation of CDC data for 2015

Data: https://vishalsingh.org/zipcode/report/life-evaluation-and-daily-feelings/figures/felt_despair.json

### Table 8. The ladder score is steadily worse toward the more deprived end under the ADI and under each of the three other rankings

Mean ladder score (0 to 10) by tenth of interviews, with the respondent's county ranked four ways, over the 3,121 counties that have all four rankings

| bin | conf_high | conf_low | estimate | n_units | ranking |
| ---: | ---: | ---: | ---: | ---: | --- |
| 1 | 7.23 | 7.11 | 7.17 | 41 | Area Deprivation Index |
| 2 | 7.17 | 7.10 | 7.13 | 88 | Area Deprivation Index |
| 3 | 7.13 | 7.07 | 7.10 | 110 | Area Deprivation Index |
| 4 | 7.15 | 7.08 | 7.11 | 142 | Area Deprivation Index |
| 5 | 7.12 | 7.07 | 7.10 | 158 | Area Deprivation Index |
| 6 | 7.13 | 7.07 | 7.10 | 197 | Area Deprivation Index |
| 7 | 7.08 | 7.03 | 7.06 | 284 | Area Deprivation Index |
| 8 | 7.06 | 7.02 | 7.04 | 329 | Area Deprivation Index |
| 9 | 7.00 | 6.96 | 6.98 | 572 | Area Deprivation Index |
| 10 | 6.94 | 6.91 | 6.93 | 1,200 | Area Deprivation Index |
| 1 | 7.25 | 7.19 | 7.22 | 101 | Standardized index |
| 2 | 7.19 | 7.14 | 7.16 | 100 | Standardized index |
| 3 | 7.15 | 7.10 | 7.12 | 168 | Standardized index |
| 4 | 7.13 | 7.09 | 7.11 | 188 | Standardized index |
| 5 | 7.10 | 7.05 | 7.07 | 185 | Standardized index |
| 6 | 7.07 | 7.03 | 7.05 | 248 | Standardized index |
| 7 | 7.08 | 7.00 | 7.04 | 240 | Standardized index |
| 8 | 7.06 | 6.99 | 7.03 | 321 | Standardized index |
| 9 | 6.98 | 6.95 | 6.96 | 579 | Standardized index |
| 10 | 6.96 | 6.92 | 6.94 | 991 | Standardized index |
| 1 | 7.24 | 7.19 | 7.22 | 83 | Median household income |
| 2 | 7.18 | 7.12 | 7.15 | 95 | Median household income |
| 3 | 7.14 | 7.08 | 7.11 | 131 | Median household income |
| 4 | 7.12 | 7.05 | 7.09 | 130 | Median household income |
| 5 | 7.11 | 7.04 | 7.07 | 190 | Median household income |
| 6 | 7.11 | 7.04 | 7.07 | 201 | Median household income |
| 7 | 7.08 | 7.03 | 7.06 | 310 | Median household income |
| 8 | 7.06 | 7.01 | 7.03 | 348 | Median household income |
| 9 | 7.00 | 6.95 | 6.97 | 524 | Median household income |
| 10 | 6.95 | 6.92 | 6.94 | 1,109 | Median household income |
| 1 | 7.26 | 7.21 | 7.24 | 91 | Bachelor's degree share |
| 2 | 7.19 | 7.14 | 7.16 | 119 | Bachelor's degree share |
| 3 | 7.16 | 7.10 | 7.13 | 97 | Bachelor's degree share |
| 4 | 7.13 | 7.08 | 7.10 | 125 | Bachelor's degree share |
| 5 | 7.11 | 7.05 | 7.08 | 157 | Bachelor's degree share |
| 6 | 7.07 | 7.02 | 7.04 | 148 | Bachelor's degree share |
| 7 | 7.12 | 7.05 | 7.08 | 238 | Bachelor's degree share |
| 8 | 7.02 | 6.98 | 7.00 | 414 | Bachelor's degree share |
| 9 | 6.99 | 6.95 | 6.97 | 615 | Bachelor's degree share |
| 10 | 6.92 | 6.89 | 6.90 | 1,117 | Bachelor's degree share |

Note: Least and most deprived tenth: ADI 7.17 and 6.93, rank correlation −0.98; standardized index 7.22 and 6.94, −1.00; median household income 7.22 and 6.94, −1.00; bachelor's degree share 7.24 and 6.90, −0.96.

Note: The standardized index applies the ADI's coefficients to its measures after standardizing each one, so home value and income carry less weight than in the ADI. Income and degree share are single census measures, each ranked from the highest value to the lowest.

Source: Gallup U.S. Daily, ladder 2009-2017; unweighted, intervals clustered by county; Area Deprivation Index 2023 and ACS 2019-2023 tract population joined on the respondent's county or ZIP code; unweighted; ADI 2023 and ACS 2019-2023 tract tables (B19013, B15003 and the 15 rebuilt ADI measures)

Data: https://vishalsingh.org/zipcode/report/life-evaluation-and-daily-feelings/figures/alt_rankings.json

## All estimates

Every fact in facts.json (157), sorted by id.

| Fact id | Value | Unit | Estimand | Population | Period | Interval | n | Weights | Source | Note |
| --- | ---: | --- | --- | --- | --- | --- | ---: | --- | --- | --- |
| age_q1 | 51.9 | years of age | unweighted mean age, interval clustered by the respondent's county | respondents in the lowest fifth of interviews, ranked by the deaths-of-despair rate of the respondent's county | interviews 2008 to 2017; deaths of despair 2015 | 95% CI 51.3 to 52.5 | 617,488 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| age_q5 | 55.5 | years of age | unweighted mean age, interval clustered by the respondent's county | respondents in the highest fifth of interviews, ranked by the deaths-of-despair rate of the respondent's county | interviews 2008 to 2017; deaths of despair 2015 | 95% CI 55.1 to 56.0 | 619,036 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| alt_adi_least | 7.17 | Mean ladder score (0 to 10) | unweighted mean ladder score; 95% intervals clustered by the respondent's county | respondents in the least deprived tenth of interviews, with the respondent's county ranked by Area Deprivation Index, among interviews whose county has all four rankings | interviews in 2009 to 2017 | 95% CI 7.11 to 7.23 | 41 counties | none (unweighted) | Gallup U.S. Daily Poll |   |
| alt_adi_most | 6.93 | Mean ladder score (0 to 10) | unweighted mean ladder score; 95% intervals clustered by the respondent's county | respondents in the most deprived tenth of interviews, with the respondent's county ranked by Area Deprivation Index, among interviews whose county has all four rankings | interviews in 2009 to 2017 | 95% CI 6.91 to 6.94 | 1,200 counties | none (unweighted) | Gallup U.S. Daily Poll |   |
| alt_adi_rho | −0.98 | rank correlation | Spearman rank correlation between bin order and bin mean, each bin counted once | the tenths of interviews, with the respondent's county ranked by Area Deprivation Index | interviews in 2009 to 2017 |   |   | each bin counted once | Gallup U.S. Daily Poll |   |
| alt_degree_least | 7.24 | Mean ladder score (0 to 10) | unweighted mean ladder score; 95% intervals clustered by the respondent's county | respondents in the least deprived tenth of interviews, with the respondent's county ranked by Bachelor's degree share, among interviews whose county has all four rankings | interviews in 2009 to 2017 | 95% CI 7.21 to 7.26 | 91 counties | none (unweighted) | Gallup U.S. Daily Poll |   |
| alt_degree_most | 6.90 | Mean ladder score (0 to 10) | unweighted mean ladder score; 95% intervals clustered by the respondent's county | respondents in the most deprived tenth of interviews, with the respondent's county ranked by Bachelor's degree share, among interviews whose county has all four rankings | interviews in 2009 to 2017 | 95% CI 6.89 to 6.92 | 1,117 counties | none (unweighted) | Gallup U.S. Daily Poll |   |
| alt_degree_rho | −0.96 | rank correlation | Spearman rank correlation between bin order and bin mean, each bin counted once | the tenths of interviews, with the respondent's county ranked by Bachelor's degree share | interviews in 2009 to 2017 |   |   | each bin counted once | Gallup U.S. Daily Poll |   |
| alt_income_least | 7.22 | Mean ladder score (0 to 10) | unweighted mean ladder score; 95% intervals clustered by the respondent's county | respondents in the least deprived tenth of interviews, with the respondent's county ranked by Median household income, among interviews whose county has all four rankings | interviews in 2009 to 2017 | 95% CI 7.19 to 7.24 | 83 counties | none (unweighted) | Gallup U.S. Daily Poll |   |
| alt_income_most | 6.94 | Mean ladder score (0 to 10) | unweighted mean ladder score; 95% intervals clustered by the respondent's county | respondents in the most deprived tenth of interviews, with the respondent's county ranked by Median household income, among interviews whose county has all four rankings | interviews in 2009 to 2017 | 95% CI 6.92 to 6.95 | 1,109 counties | none (unweighted) | Gallup U.S. Daily Poll |   |
| alt_income_rho | −1.00 | rank correlation | Spearman rank correlation between bin order and bin mean, each bin counted once | the tenths of interviews, with the respondent's county ranked by Median household income | interviews in 2009 to 2017 |   |   | each bin counted once | Gallup U.S. Daily Poll |   |
| alt_n_places | 3,121 | counties | count of counties in the common sample of the ranking comparison | counties with the outcome and all four rankings | interviews in 2009 to 2017 |   |   | none (full count) | Gallup U.S. Daily Poll |   |
| alt_rho_cut | 0.9 | absolute rank correlation | threshold between a steady and an uneven gradient | the rule for changed gradients (docs/RANKING_SWITCH_RULE.md) | interviews in 2009 to 2017 |   |   | none (a fixed rule) | Gallup U.S. Daily Poll |   |
| alt_rho_margin | 0.1 | absolute rank correlation | smallest move in the rank correlation that counts as a change of shape | the rule for changed gradients (docs/RANKING_SWITCH_RULE.md) | interviews in 2009 to 2017 |   |   | none (a fixed rule) | Gallup U.S. Daily Poll |   |
| alt_std_least | 7.22 | Mean ladder score (0 to 10) | unweighted mean ladder score; 95% intervals clustered by the respondent's county | respondents in the least deprived tenth of interviews, with the respondent's county ranked by Standardized index, among interviews whose county has all four rankings | interviews in 2009 to 2017 | 95% CI 7.19 to 7.25 | 101 counties | none (unweighted) | Gallup U.S. Daily Poll |   |
| alt_std_most | 6.94 | Mean ladder score (0 to 10) | unweighted mean ladder score; 95% intervals clustered by the respondent's county | respondents in the most deprived tenth of interviews, with the respondent's county ranked by Standardized index, among interviews whose county has all four rankings | interviews in 2009 to 2017 | 95% CI 6.92 to 6.96 | 991 counties | none (unweighted) | Gallup U.S. Daily Poll |   |
| alt_std_rho | −1.00 | rank correlation | Spearman rank correlation between bin order and bin mean, each bin counted once | the tenths of interviews, with the respondent's county ranked by Standardized index | interviews in 2009 to 2017 |   |   | each bin counted once | Gallup U.S. Daily Poll |   |
| county_floor | 100 | respondents | minimum number of respondents with an item for a county mean to enter (a definition of the sample) | counties in the county-level correlations | interviews: ladder 2009 to 2017, feelings 2008 to 2017 |   |   | none (definition) | Gallup U.S. Daily Poll |   |
| expected_inc1 | 6.98 | expected ladder score (0 worst possible life, 10 best possible life) | unweighted mean of the ladder score the respondent expects to have in the future, interval clustered by the respondent's county | respondents in the poorest household income bracket with an expected ladder score | interviews in 2009 to 2017 | 95% CI 6.95 to 7.02 | 348,488 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| expected_inc6 | 8.11 | expected ladder score (0 worst possible life, 10 best possible life) | unweighted mean of the ladder score the respondent expects to have in the future, interval clustered by the respondent's county | respondents in the richest household income bracket with an expected ladder score | interviews in 2009 to 2017 | 95% CI 8.10 to 8.12 | 446,942 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| gap_adj_anger | −1.3 | percentage points, most deprived tenth minus least deprived tenth | the same difference after comparing respondents within cells of five-year age band (age capped at 85), sex, race, three-level education and household income bracket, interval clustered by the respondent's county | respondents with age, sex, race, education, household income and a county rent and an answer to the anger item in the two extreme county tenths | interviews in the years the item was asked | 95% CI −1.9 to −0.6 | 326,006 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| gap_adj_enjoyment | 1.6 | percentage points, most deprived tenth minus least deprived tenth | the same difference after comparing respondents within cells of five-year age band (age capped at 85), sex, race, three-level education and household income bracket, interval clustered by the respondent's county | respondents with age, sex, race, education, household income and a county rent and an answer to the enjoyment item in the two extreme county tenths | interviews in the years the item was asked | 95% CI 1.0 to 2.2 | 412,511 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| gap_adj_happiness_yest | 1.8 | percentage points, most deprived tenth minus least deprived tenth | the same difference after comparing respondents within cells of five-year age band (age capped at 85), sex, race, three-level education and household income bracket, interval clustered by the respondent's county | respondents with age, sex, race, education, household income and a county rent and an answer to the happiness item in the two extreme county tenths | interviews in the years the item was asked | 95% CI 1.2 to 2.4 | 385,768 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| gap_adj_physical_pain | 1.8 | percentage points, most deprived tenth minus least deprived tenth | the same difference after comparing respondents within cells of five-year age band (age capped at 85), sex, race, three-level education and household income bracket, interval clustered by the respondent's county | respondents with age, sex, race, education, household income and a county rent and an answer to the physical pain item in the two extreme county tenths | interviews in the years the item was asked | 95% CI 1.4 to 2.3 | 412,916 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| gap_adj_sadness | −1.7 | percentage points, most deprived tenth minus least deprived tenth | the same difference after comparing respondents within cells of five-year age band (age capped at 85), sex, race, three-level education and household income bracket, interval clustered by the respondent's county | respondents with age, sex, race, education, household income and a county rent and an answer to the sadness item in the two extreme county tenths | interviews in the years the item was asked | 95% CI −2.4 to −1.0 | 386,323 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| gap_adj_stress | −2.3 | percentage points, most deprived tenth minus least deprived tenth | the same difference after comparing respondents within cells of five-year age band (age capped at 85), sex, race, three-level education and household income bracket, interval clustered by the respondent's county | respondents with age, sex, race, education, household income and a county rent and an answer to the stress item in the two extreme county tenths | interviews in the years the item was asked | 95% CI −3.0 to −1.6 | 412,916 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| gap_adj_worry | −3.7 | percentage points, most deprived tenth minus least deprived tenth | the same difference after comparing respondents within cells of five-year age band (age capped at 85), sex, race, three-level education and household income bracket, interval clustered by the respondent's county | respondents with age, sex, race, education, household income and a county rent and an answer to the worry item in the two extreme county tenths | interviews in the years the item was asked | 95% CI −4.6 to −2.8 | 413,010 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| gap_anger | −0.7 | percentage points, most deprived tenth minus least deprived tenth | difference between the unweighted shares reporting the feeling a lot of the day yesterday in the most and the least deprived tenth of interviews, ranked by county ADI, interval clustered by the respondent's county | respondents with an answer to the anger item in the two extreme county tenths | interviews in the years the item was asked | 95% CI −1.7 to 0.3 | 421,120 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| gap_enjoyment | −1.2 | percentage points, most deprived tenth minus least deprived tenth | difference between the unweighted shares reporting the feeling a lot of the day yesterday in the most and the least deprived tenth of interviews, ranked by county ADI, interval clustered by the respondent's county | respondents with an answer to the enjoyment item in the two extreme county tenths | interviews in the years the item was asked | 95% CI −2.1 to −0.4 | 525,461 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| gap_happiness_yest | −0.6 | percentage points, most deprived tenth minus least deprived tenth | difference between the unweighted shares reporting the feeling a lot of the day yesterday in the most and the least deprived tenth of interviews, ranked by county ADI, interval clustered by the respondent's county | respondents with an answer to the happiness item in the two extreme county tenths | interviews in the years the item was asked | 95% CI −1.4 to 0.2 | 493,755 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| gap_physical_pain | 8.8 | percentage points, most deprived tenth minus least deprived tenth | difference between the unweighted shares reporting the feeling a lot of the day yesterday in the most and the least deprived tenth of interviews, ranked by county ADI, interval clustered by the respondent's county | respondents with an answer to the physical pain item in the two extreme county tenths | interviews in the years the item was asked | 95% CI 7.9 to 9.7 | 526,330 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| gap_rent_anger | 0.4 | percentage points, most deprived tenth minus least deprived tenth | the same difference after comparing respondents within cells of five-year age band (age capped at 85), sex, race, three-level education and household income bracket and a quintile of county rent, interval clustered by the respondent's county | respondents with age, sex, race, education, household income, a county rent and an answer to the anger item in the two extreme county tenths | interviews in the years the item was asked | 95% CI −0.2 to 1.1 | 326,006 respondents | none (unweighted) | HUD Fair Market Rents |   |
| gap_rent_enjoyment | −0.7 | percentage points, most deprived tenth minus least deprived tenth | the same difference after comparing respondents within cells of five-year age band (age capped at 85), sex, race, three-level education and household income bracket and a quintile of county rent, interval clustered by the respondent's county | respondents with age, sex, race, education, household income, a county rent and an answer to the enjoyment item in the two extreme county tenths | interviews in the years the item was asked | 95% CI −1.2 to −0.1 | 412,511 respondents | none (unweighted) | HUD Fair Market Rents |   |
| gap_rent_happiness_yest | −0.5 | percentage points, most deprived tenth minus least deprived tenth | the same difference after comparing respondents within cells of five-year age band (age capped at 85), sex, race, three-level education and household income bracket and a quintile of county rent, interval clustered by the respondent's county | respondents with age, sex, race, education, household income, a county rent and an answer to the happiness item in the two extreme county tenths | interviews in the years the item was asked | 95% CI −1.1 to 0.2 | 385,768 respondents | none (unweighted) | HUD Fair Market Rents |   |
| gap_rent_physical_pain | 1.2 | percentage points, most deprived tenth minus least deprived tenth | the same difference after comparing respondents within cells of five-year age band (age capped at 85), sex, race, three-level education and household income bracket and a quintile of county rent, interval clustered by the respondent's county | respondents with age, sex, race, education, household income, a county rent and an answer to the physical pain item in the two extreme county tenths | interviews in the years the item was asked | 95% CI 0.8 to 1.7 | 412,916 respondents | none (unweighted) | HUD Fair Market Rents |   |
| gap_rent_sadness | 0.5 | percentage points, most deprived tenth minus least deprived tenth | the same difference after comparing respondents within cells of five-year age band (age capped at 85), sex, race, three-level education and household income bracket and a quintile of county rent, interval clustered by the respondent's county | respondents with age, sex, race, education, household income, a county rent and an answer to the sadness item in the two extreme county tenths | interviews in the years the item was asked | 95% CI −0.2 to 1.2 | 386,323 respondents | none (unweighted) | HUD Fair Market Rents |   |
| gap_rent_stress | 0.4 | percentage points, most deprived tenth minus least deprived tenth | the same difference after comparing respondents within cells of five-year age band (age capped at 85), sex, race, three-level education and household income bracket and a quintile of county rent, interval clustered by the respondent's county | respondents with age, sex, race, education, household income, a county rent and an answer to the stress item in the two extreme county tenths | interviews in the years the item was asked | 95% CI −0.3 to 1.2 | 412,916 respondents | none (unweighted) | HUD Fair Market Rents |   |
| gap_rent_worry | 0.5 | percentage points, most deprived tenth minus least deprived tenth | the same difference after comparing respondents within cells of five-year age band (age capped at 85), sex, race, three-level education and household income bracket and a quintile of county rent, interval clustered by the respondent's county | respondents with age, sex, race, education, household income, a county rent and an answer to the worry item in the two extreme county tenths | interviews in the years the item was asked | 95% CI −0.4 to 1.5 | 413,010 respondents | none (unweighted) | HUD Fair Market Rents |   |
| gap_sadness | 1.9 | percentage points, most deprived tenth minus least deprived tenth | difference between the unweighted shares reporting the feeling a lot of the day yesterday in the most and the least deprived tenth of interviews, ranked by county ADI, interval clustered by the respondent's county | respondents with an answer to the sadness item in the two extreme county tenths | interviews in the years the item was asked | 95% CI 0.7 to 3.2 | 494,753 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| gap_stress | −2.1 | percentage points, most deprived tenth minus least deprived tenth | difference between the unweighted shares reporting the feeling a lot of the day yesterday in the most and the least deprived tenth of interviews, ranked by county ADI, interval clustered by the respondent's county | respondents with an answer to the stress item in the two extreme county tenths | interviews in the years the item was asked | 95% CI −3.1 to −1.2 | 526,330 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| gap_worry | −1.1 | percentage points, most deprived tenth minus least deprived tenth | difference between the unweighted shares reporting the feeling a lot of the day yesterday in the most and the least deprived tenth of interviews, ranked by county ADI, interval clustered by the respondent's county | respondents with an answer to the worry item in the two extreme county tenths | interviews in the years the item was asked | 95% CI −2.3 to 0.2 | 526,425 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| gap_zip_physical_pain | 11.0 | percentage points, most deprived tenth minus least deprived tenth | difference between the unweighted shares reporting physical pain in the most and the least deprived tenth of interviews, ranked by ZIP-area ADI, interval clustered by the respondent's county | respondents with an answer to the pain item in the two extreme ZIP-area tenths | interviews 2008 to 2017 | 95% CI 10.3 to 11.7 | 513,469 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| icc_ladder | 0.51% | percent of the variance of the ladder score | intraclass correlation of the ladder score by county (one-way ANOVA estimator), bootstrap over counties | interviews with a ladder score and a county | interviews in 2009 to 2017 | 95% CI 0.46% to 0.56% | 3,141 counties | none (unweighted) | Gallup U.S. Daily Poll |   |
| inc_gap_enjoyment | 0.37 | standard deviations, richest bracket minus poorest bracket | difference between the unweighted means of the item in the richest and the poorest household income bracket, in standard deviations of the item over all respondents, interval clustered by the respondent's county | respondents with a household income bracket and an answer to the enjoyment item, in the poorest and the richest bracket | interviews in 2008 to 2017 | 95% CI 0.37 to 0.38 | 968,222 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| inc_gap_ladder | 0.68 | standard deviations, richest bracket minus poorest bracket | difference between the unweighted means of the item in the richest and the poorest household income bracket, in standard deviations of the item over all respondents, interval clustered by the respondent's county | respondents with a household income bracket and an answer to the ladder score item, in the poorest and the richest bracket | interviews in 2009 to 2017 | 95% CI 0.67 to 0.69 | 827,090 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| inc_gap_physical_pain | −0.51 | standard deviations, richest bracket minus poorest bracket | difference between the unweighted means of the item in the richest and the poorest household income bracket, in standard deviations of the item over all respondents, interval clustered by the respondent's county | respondents with a household income bracket and an answer to the physical pain item, in the poorest and the richest bracket | interviews in 2008 to 2017 | 95% CI −0.52 to −0.50 | 969,411 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| inc_gap_sadness | −0.48 | standard deviations, richest bracket minus poorest bracket | difference between the unweighted means of the item in the richest and the poorest household income bracket, in standard deviations of the item over all respondents, interval clustered by the respondent's county | respondents with a household income bracket and an answer to the sadness item, in the poorest and the richest bracket | interviews in 2008 to 2017 | 95% CI −0.49 to −0.48 | 899,077 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| inc_gap_stress | −0.12 | standard deviations, richest bracket minus poorest bracket | difference between the unweighted means of the item in the richest and the poorest household income bracket, in standard deviations of the item over all respondents, interval clustered by the respondent's county | respondents with a household income bracket and an answer to the stress item, in the poorest and the richest bracket | interviews in 2008 to 2017 | 95% CI −0.13 to −0.11 | 969,402 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| inc_gap_worry | −0.34 | standard deviations, richest bracket minus poorest bracket | difference between the unweighted means of the item in the richest and the poorest household income bracket, in standard deviations of the item over all respondents, interval clustered by the respondent's county | respondents with a household income bracket and an answer to the worry item, in the poorest and the richest bracket | interviews in 2008 to 2017 | 95% CI −0.35 to −0.33 | 969,564 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| income_top_cut | $90,000 | dollars of annual household income | the lower limit of the top household income bracket (a definition of the bracket) | the richest household income bracket in the Gallup extract (this amount and over) | interviews in 2008 to 2017 |   |   | none (definition) | Gallup U.S. Daily Poll |   |
| kd_threshold | $75,000 | dollars of annual household income | annual income above which a measure of emotional well-being stopped rising with log income in Kahneman and Deaton (2010); copied from the paper | respondents in the Gallup-Healthways Well-Being Index analyzed by Kahneman and Deaton | 2008 to 2009 |   |   | as in the paper | Published estimates from cited studies | A published figure from the paper, not a result of this build. |
| kw_threshold | $100,000 | dollars of annual household income | annual income above which well-being flattened for an unhappy minority in Killingsworth, Kahneman and Mellers (2023); copied from the paper | the unhappiest fifth of respondents in the experience-sampling data analyzed by Killingsworth, Kahneman and Mellers | 2009 to 2022 |   |   | as in the paper | Published estimates from cited studies | A published figure from the paper, not a result of this build. |
| ladder_adj_d10_county | 7.10 | Cantril ladder score (0 worst possible life, 10 best possible life) | mean ladder score after subtracting the mean of the respondent's cell (five-year age band (age capped at 85), sex, race, three-level education and household income bracket) and adding the grand mean, interval clustered by the respondent's county | respondents with age, sex, race, education, household income and a county rent in the most deprived tenth of interviews by county deprivation | interviews in 2009 to 2017 | 95% CI 7.08 to 7.11 | 177,139 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| ladder_adj_d1_county | 6.97 | Cantril ladder score (0 worst possible life, 10 best possible life) | mean ladder score after subtracting the mean of the respondent's cell (five-year age band (age capped at 85), sex, race, three-level education and household income bracket) and adding the grand mean, interval clustered by the respondent's county | respondents with age, sex, race, education, household income and a county rent in the least deprived tenth of interviews by county deprivation | interviews in 2009 to 2017 | 95% CI 6.94 to 7.00 | 173,060 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| ladder_d10_county | 6.93 | Cantril ladder score (0 worst possible life, 10 best possible life) | unweighted mean ladder score in a tenth of interviews ranked by the deprivation of the respondent's county (county ADI is the mean of its tracts' ADI 2023 national ranks weighted by 2019-2023 tract population), interval clustered by the respondent's county | respondents in the most deprived tenth of interviews, ranked by the Area Deprivation Index of the respondent's county | interviews in 2009 to 2017 | 95% CI 6.91 to 6.94 | 223,260 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| ladder_d10_zip | 6.87 | Cantril ladder score (0 worst possible life, 10 best possible life) | unweighted mean ladder score in a tenth of interviews ranked by the deprivation of the respondent's ZIP area (ZIP-area ADI is the 2023 national rank of the ZCTA with the ZIP code's number, an area-weighted approximation), interval clustered by the respondent's county | respondents in the most deprived tenth of interviews, ranked by the Area Deprivation Index of the respondent's ZIP area | interviews in 2009 to 2017 | 95% CI 6.85 to 6.88 | 217,263 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| ladder_d1_county | 7.17 | Cantril ladder score (0 worst possible life, 10 best possible life) | unweighted mean ladder score in a tenth of interviews ranked by the deprivation of the respondent's county (county ADI is the mean of its tracts' ADI 2023 national ranks weighted by 2019-2023 tract population), interval clustered by the respondent's county | respondents in the least deprived tenth of interviews, ranked by the Area Deprivation Index of the respondent's county | interviews in 2009 to 2017 | 95% CI 7.11 to 7.23 | 218,776 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| ladder_d1_zip | 7.29 | Cantril ladder score (0 worst possible life, 10 best possible life) | unweighted mean ladder score in a tenth of interviews ranked by the deprivation of the respondent's ZIP area (ZIP-area ADI is the 2023 national rank of the ZCTA with the ZIP code's number, an area-weighted approximation), interval clustered by the respondent's county | respondents in the least deprived tenth of interviews, ranked by the Area Deprivation Index of the respondent's ZIP area | interviews in 2009 to 2017 | 95% CI 7.24 to 7.34 | 213,516 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| ladder_drop_county | 0.24 | ladder points, least deprived tenth minus most deprived tenth | difference between the unweighted mean ladder scores of the least and the most deprived tenth (county ADI is the mean of its tracts' ADI 2023 national ranks weighted by 2019-2023 tract population), interval clustered by the respondent's county | respondents in the two extreme tenths of interviews, ranked by the Area Deprivation Index of the respondent's county | interviews in 2009 to 2017 | 95% CI 0.18 to 0.31 | 442,036 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| ladder_drop_county_with08 | 0.26 | ladder points, least deprived tenth minus most deprived tenth | the county drop of the chapter's main comparison recomputed with the 2008 ladder answers left in, interval clustered by the respondent's county | respondents in the two extreme tenths of interviews, ranked by the Area Deprivation Index of the respondent's county, including 2008 | interviews in 2008 to 2017 including the 2008 ladder, which carries a question-order artifact | 95% CI 0.19 to 0.33 | 513,776 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| ladder_drop_zip | 0.42 | ladder points, least deprived tenth minus most deprived tenth | difference between the unweighted mean ladder scores of the least and the most deprived tenth (ZIP-area ADI is the 2023 national rank of the ZCTA with the ZIP code's number, an area-weighted approximation), interval clustered by the respondent's county | respondents in the two extreme tenths of interviews, ranked by the Area Deprivation Index of the respondent's ZIP area | interviews in 2009 to 2017 | 95% CI 0.37 to 0.48 | 430,779 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| ladder_gap_adj_county | 0.13 | ladder points, most deprived tenth minus least deprived tenth | gap between the extreme tenths after comparing respondents within cells of five-year age band (age capped at 85), sex, race, three-level education and household income bracket, interval clustered by the respondent's county | respondents with age, sex, race, education, household income and a county rent in the two extreme county tenths | interviews in 2009 to 2017 | 95% CI 0.10 to 0.16 | 350,199 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| ladder_gap_adj_county_with08 | +0.12 | ladder points, most deprived tenth minus least deprived tenth | the nominal-income gap of the chapter's main comparison recomputed with the 2008 ladder answers left in, interval clustered by the respondent's county | respondents with age, sex, race, education, household income and a county rent in the two extreme county tenths, including 2008 | interviews in 2008 to 2017 including the 2008 ladder, which carries a question-order artifact | 95% CI 0.09 to 0.15 | 403,911 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| ladder_gap_adj_zip | 0.07 | ladder points, most deprived tenth minus least deprived tenth | gap between the extreme tenths after comparing respondents within cells of five-year age band (age capped at 85), sex, race, three-level education and household income bracket, interval clustered by the respondent's county | respondents with age, sex, race, education, household income and a county rent in the two extreme ZIP area tenths | interviews in 2009 to 2017 | 95% CI 0.04 to 0.10 | 343,365 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| ladder_gap_edu6 | 0.13 | ladder points, most deprived tenth minus least deprived tenth | the nominal-income gap with the six-level education item in the cells in place of the three-level one, interval clustered by the respondent's county | respondents with age, sex, race, six-level education, household income and a county rent in the two extreme county tenths | interviews in 2009 to 2017 | 95% CI 0.10 to 0.16 | 350,199 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| ladder_gap_eqc | +0.02 | ladder points, most deprived tenth minus least deprived tenth | gap between the extreme tenths after comparing respondents within cells of five-year age band (age capped at 85), sex, race, three-level education and household income bracket, with each county counted once (counties with at least the survey floor of interviews with the item), bootstrap over counties | respondents with age, sex, race, education, household income and a county rent in the two extreme county tenths | interviews in 2009 to 2017 | 95% CI −0.03 to 0.07 | 347,376 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| ladder_gap_lnrent | −0.04 | ladder points, most deprived tenth minus least deprived tenth | gap between the extreme tenths after comparing respondents within cells of five-year age band (age capped at 85), sex, race, three-level education and household income bracket, with the logarithm of county rent also a linear control, interval clustered by the respondent's county | respondents with age, sex, race, education, household income and a county rent in the two extreme county tenths | interviews in 2009 to 2017 | 95% CI −0.07 to −0.01 | 350,199 respondents | none (unweighted) | HUD Fair Market Rents |   |
| ladder_gap_rentq | −0.02 | ladder points, most deprived tenth minus least deprived tenth | gap between the extreme tenths after comparing respondents within cells of five-year age band (age capped at 85), sex, race, three-level education and household income bracket, with a quintile of county rent also in the cells, interval clustered by the respondent's county | respondents with age, sex, race, education, household income and a county rent in the two extreme county tenths | interviews in 2009 to 2017 | 95% CI −0.05 to 0.00 | 350,199 respondents | none (unweighted) | HUD Fair Market Rents |   |
| ladder_gap_rentq_with08 | −0.03 | ladder points, most deprived tenth minus least deprived tenth | the rent-quintile gap of the chapter's main comparison recomputed with the 2008 ladder answers left in, interval clustered by the respondent's county | respondents with age, sex, race, education, household income and a county rent in the two extreme county tenths, including 2008 | interviews in 2008 to 2017 including the 2008 ladder, which carries a question-order artifact | 95% CI −0.06 to 0.00 | 403,911 respondents | none (unweighted) | HUD Fair Market Rents |   |
| ladder_gap_rentq_zip | −0.06 | ladder points, most deprived tenth minus least deprived tenth | gap between the extreme ZIP-area tenths after comparing respondents within cells of five-year age band (age capped at 85), sex, race, three-level education and household income bracket and a quintile of county rent, interval clustered by the respondent's county | respondents with age, sex, race, education, household income and a county rent in the two extreme ZIP-area tenths | interviews in 2009 to 2017 | 95% CI −0.09 to −0.03 | 343,365 respondents | none (unweighted) | HUD Fair Market Rents |   |
| ladder_inc1 | 6.28 | Cantril ladder score (0 worst possible life, 10 best possible life) | unweighted mean ladder score, interval clustered by the respondent's county | respondents in the poorest household income bracket | interviews in 2009 to 2017 | 95% CI 6.26 to 6.29 | 373,989 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| ladder_inc6 | 7.60 | Cantril ladder score (0 worst possible life, 10 best possible life) | unweighted mean ladder score, interval clustered by the respondent's county | respondents in the richest household income bracket | interviews in 2009 to 2017 | 95% CI 7.59 to 7.61 | 453,101 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| ladder_q1 | 7.15 | Cantril ladder score (0 worst possible life, 10 best possible life) | unweighted mean ladder score, interval clustered by the respondent's county | respondents in the lowest fifth of interviews, ranked by the deaths-of-despair rate of the respondent's county | interviews 2009 to 2017; deaths of despair 2015 | 95% CI 7.12 to 7.18 | 392,393 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| ladder_q5 | 7.00 | Cantril ladder score (0 worst possible life, 10 best possible life) | unweighted mean ladder score, interval clustered by the respondent's county | respondents in the highest fifth of interviews, ranked by the deaths-of-despair rate of the respondent's county | interviews 2009 to 2017; deaths of despair 2015 | 95% CI 6.98 to 7.02 | 399,284 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| ladder_share_2009 | 85% | percent of the year's interviews that carry a ladder score | share of the year's interviews with a ladder score (unweighted count) | interviews with a county in 2009 | 2009 |   | 353,849 respondents | none (unweighted count) | Gallup U.S. Daily Poll |   |
| ladder_share_late | 49% | percent of interviews that carry a ladder score | share of interviews with a ladder score (unweighted count) | interviews with a county from 2013 to 2017 | 2013 to 2017 |   | 1,750,460 respondents | none (unweighted count) | Gallup U.S. Daily Poll |   |
| lf_enjoy | 1.46 | ladder points higher among respondents who reported the feeling | difference between the unweighted mean ladder scores of respondents who did and did not report the feeling a lot of the day yesterday, interval clustered by the respondent's county | respondents with a ladder score and an answer to the enjoyment item | interviews in the years the ladder (2009 to 2017) and the feeling were both asked | 95% CI 1.44 to 1.47 | 2,211,300 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| lf_happy | 1.57 | ladder points higher among respondents who reported the feeling | difference between the unweighted mean ladder scores of respondents who did and did not report the feeling a lot of the day yesterday, interval clustered by the respondent's county | respondents with a ladder score and an answer to the happiness item | interviews in the years the ladder (2009 to 2017) and the feeling were both asked | 95% CI 1.55 to 1.59 | 2,051,189 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| lf_pain | 0.91 | ladder points lower among respondents who reported the feeling | difference between the unweighted mean ladder scores of respondents who did and did not report the feeling a lot of the day yesterday, interval clustered by the respondent's county | respondents with a ladder score and an answer to the physical pain item | interviews in the years the ladder (2009 to 2017) and the feeling were both asked | 95% CI 0.90 to 0.93 | 2,213,777 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| lf_sad | 1.33 | ladder points lower among respondents who reported the feeling | difference between the unweighted mean ladder scores of respondents who did and did not report the feeling a lot of the day yesterday, interval clustered by the respondent's county | respondents with a ladder score and an answer to the sadness item | interviews in the years the ladder (2009 to 2017) and the feeling were both asked | 95% CI 1.32 to 1.35 | 2,054,693 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| mto_swb | 0.098 | standard deviations of a three-point happiness question | intention-to-treat effect of the offer to move to a less disadvantaged area on a standardized answer to 'how would you say things are these days: very happy, pretty happy, not too happy' in the Moving to Opportunity experiment (Ludwig and coauthors 2012); the estimate and standard error (0.039) are copied from the paper and the interval is the estimate plus or minus 1.96 standard errors | low-income adults in public housing in five cities (Baltimore, Boston, Chicago, Los Angeles, New York) who were offered a housing voucher to move to a lower-poverty neighborhood or an unrestricted voucher, or stayed, by random assignment | long-term follow-up survey, 10 to 15 years after random assignment | 95% CI 0.022 to 0.174 |   | as in the paper | Published estimates from cited studies | A published estimate for the study's adult sample, not a result of this build. |
| n_cell_sample | 1,759,964 | interviews | count of interviews in the one sample used for every comparison within cells (the cell means come from these interviews at every step) | interviews with a county, a ladder score, an age, a sex, a race, an education level, a household income bracket and a county rent | interviews in 2009 to 2017 |   |   | none (unweighted count) | Gallup U.S. Daily Poll |   |
| n_counties | 3,142 | counties | count of distinct county codes assigned to respondents (derived from the ZIP code) | counties of the interviews in the analysis | 2008 to 2017 |   |   | none (unweighted count) | Gallup U.S. Daily Poll |   |
| n_counties_d1 | 45 | counties | count of distinct counties among the interviews in the tenth (a tenth holds about a tenth of interviews, not of counties) | counties of the interviews in the least deprived tenth of interviews by county deprivation | interviews in 2009 to 2017 |   |   | none (unweighted count) | Gallup U.S. Daily Poll |   |
| n_counties_d10 | 1,205 | counties | count of distinct counties among the interviews in the tenth (a tenth holds about a tenth of interviews, not of counties) | counties of the interviews in the most deprived tenth of interviews by county deprivation | interviews in 2009 to 2017 |   |   | none (unweighted count) | Gallup U.S. Daily Poll |   |
| n_despair_counties | 1,337 | counties | count of counties in the correlations with the deaths-of-despair rate | counties with a deaths-of-despair rate and at least 100 respondents with each item | interviews: ladder 2009 to 2017, feelings 2008 to 2017; deaths of despair 2015 |   |   | none (full count) | Gallup U.S. Daily Poll |   |
| n_interviews | 3,519,539 | interviews | count of interviews with a county code (derived from the ZIP code); the others are dropped | interviews with a county | 2008 to 2017 |   |   | none (unweighted count) | Gallup U.S. Daily Poll |   |
| n_interviews_all | 3,530,270 | interviews | count of interviews in the extract (unweighted) | all interviews in the Gallup U.S. Daily extract | 2008 to 2017 |   |   | none (unweighted count) | Gallup U.S. Daily Poll |   |
| n_ladder | 2,217,485 | interviews | count of interviews with a ladder score | interviews with a county and a ladder score | interviews in 2009 to 2017 |   |   | none (unweighted count) | Gallup U.S. Daily Poll |   |
| n_no_county | 10,731 | interviews | count of interviews dropped because the extract has no county for them | interviews with no county code | 2008 to 2017 |   |   | none (unweighted count) | Gallup U.S. Daily Poll |   |
| n_no_county_adi | 784 | interviews | count of interviews left out of the county deprivation comparisons because the county has no index or is a split county | interviews with a county whose county has no Area Deprivation Index | ADI 2023 |   |   | none (unweighted count) | Gallup U.S. Daily Poll |   |
| n_wonder_counties | 534 | counties | count of counties whose drug, alcohol and suicide counts are published in every year 2018 to 2023 (complete counties, as in the chapter on overdose, alcohol and suicide) and that have a JEC rate and enough respondents | counties with a JEC rate, a complete WONDER rate and at least 100 respondents with each of the ladder, pain and stress items | interviews: ladder 2009 to 2017, feelings 2008 to 2017; JEC deaths of despair 2015; WONDER deaths 2018 to 2023 |   |   | none (full count) | CDC WONDER underlying cause of death |   |
| other_gap_max | 2.1 | percentage points (absolute value) | the largest absolute difference between the most and the least deprived county tenth among the six feelings other than physical pain | respondents in the two extreme county tenths | interviews in the years each item was asked |   |   | none (unweighted) | Gallup U.S. Daily Poll |   |
| pain_d10_county | 29.6% | percent reporting physical pain a lot of the day yesterday | unweighted share in a tenth of interviews ranked by county ADI, interval clustered by the respondent's county | respondents in the most deprived tenth of interviews, ranked by the Area Deprivation Index of the respondent's county | interviews 2008 to 2017 | 95% CI 29.2% to 30.0% | 264,787 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| pain_d1_county | 20.8% | percent reporting physical pain a lot of the day yesterday | unweighted share in a tenth of interviews ranked by county ADI, interval clustered by the respondent's county | respondents in the least deprived tenth of interviews, ranked by the Area Deprivation Index of the respondent's county | interviews 2008 to 2017 | 95% CI 20.0% to 21.7% | 261,543 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| pain_gap_eqc | 1.7 | percentage points, most deprived tenth minus least deprived tenth | the pain gap after comparing respondents within cells of five-year age band (age capped at 85), sex, race, three-level education and household income bracket, with each county counted once (counties with at least the survey floor of interviews with the item), bootstrap over counties | respondents with age, sex, race, education, household income, a county rent and an answer to the pain item in the two extreme county tenths | interviews 2008 to 2017 | 95% CI 1.1 to 2.3 | 410,598 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| pain_gap_explained | 80% | percent of the raw gap | share of the gap in the pain share between the extreme county tenths that disappears when respondents are compared within cells, bootstrap over counties with shared draws | respondents with age, sex, race, education, household income and a county rent and an answer to the pain item in the two extreme county tenths | interviews 2008 to 2017 | 95% CI 77% to 84% | 412,916 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| pain_gap_lnrent | 1.1 | percentage points, most deprived tenth minus least deprived tenth | the pain gap after comparing respondents within cells of five-year age band (age capped at 85), sex, race, three-level education and household income bracket, with the logarithm of county rent also a linear control, interval clustered by the respondent's county | respondents with age, sex, race, education, household income, a county rent and an answer to the pain item in the two extreme county tenths | interviews 2008 to 2017 | 95% CI 0.6 to 1.5 | 412,916 respondents | none (unweighted) | HUD Fair Market Rents |   |
| pain_gap_rentq | 1.2 | percentage points, most deprived tenth minus least deprived tenth | the pain gap after comparing respondents within cells of five-year age band (age capped at 85), sex, race, three-level education and household income bracket, with a quintile of county rent also in the cells, interval clustered by the respondent's county | respondents with age, sex, race, education, household income, a county rent and an answer to the pain item in the two extreme county tenths | interviews 2008 to 2017 | 95% CI 0.8 to 1.7 | 412,916 respondents | none (unweighted) | HUD Fair Market Rents |   |
| pain_inc1 | 37.8% | percent reporting physical pain a lot of the day yesterday | unweighted share, interval clustered by the respondent's county | respondents in the poorest household income bracket | interviews in 2008 to 2017 | 95% CI 37.4% to 38.3% | 441,363 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| pain_inc6 | 16.1% | percent reporting physical pain a lot of the day yesterday | unweighted share, interval clustered by the respondent's county | respondents in the richest household income bracket | interviews in 2008 to 2017 | 95% CI 15.9% to 16.3% | 528,048 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| pain_q1 | 20.2% | percent reporting physical pain a lot of the day yesterday | unweighted share, interval clustered by the respondent's county | respondents in the lowest fifth of interviews, ranked by the deaths-of-despair rate of the respondent's county | interviews 2008 to 2017; deaths of despair 2015 | 95% CI 19.8% to 20.6% | 465,440 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| pain_q5 | 26.8% | percent reporting physical pain a lot of the day yesterday | unweighted share, interval clustered by the respondent's county | respondents in the highest fifth of interviews, ranked by the deaths-of-despair rate of the respondent's county | interviews 2008 to 2017; deaths of despair 2015 | 95% CI 26.3% to 27.3% | 474,232 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| pain_q_gap | 6.6 | percentage points, highest fifth minus lowest fifth | difference between the unweighted shares reporting physical pain, interval clustered by the respondent's county | respondents in the two extreme fifths of interviews by county deaths-of-despair rate | interviews 2008 to 2017; deaths of despair 2015 | 95% CI 6.0 to 7.3 | 939,672 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| place_spread_ladder | 0.12 | ladder points (largest spread over the six brackets) | largest, over the six household income brackets, of the difference between the highest and the lowest of the three band means | respondents in each household income bracket, in the least, middle and most deprived county bands | interviews in 2009 to 2017 |   |   | none (unweighted) | Gallup U.S. Daily Poll |   |
| placegap_ladder_inc1 | −0.09 | ladder points, most deprived three tenths minus least deprived three tenths | difference between the unweighted means (shares for pain) in the most and the least deprived band of counties (tenths 8 to 10 and 1 to 3 of interviews by county ADI), interval clustered by the respondent's county | respondents in the poorest household income bracket, in the two extreme county bands | interviews in 2009 to 2017 | 95% CI −0.13 to −0.05 | 234,087 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| placegap_ladder_inc6 | 0.03 | ladder points, most deprived three tenths minus least deprived three tenths | difference between the unweighted means (shares for pain) in the most and the least deprived band of counties (tenths 8 to 10 and 1 to 3 of interviews by county ADI), interval clustered by the respondent's county | respondents in the richest household income bracket, in the two extreme county bands | interviews in 2009 to 2017 | 95% CI 0.01 to 0.05 | 274,335 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| placegap_pain_inc1 | 6.6 | percentage points, most deprived three tenths minus least deprived three tenths | difference between the unweighted means (shares for pain) in the most and the least deprived band of counties (tenths 8 to 10 and 1 to 3 of interviews by county ADI), interval clustered by the respondent's county | respondents in the poorest household income bracket, in the two extreme county bands | interviews in 2008 to 2017 | 95% CI 5.7 to 7.5 | 276,809 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| placegap_pain_inc6 | 1.7 | percentage points, most deprived three tenths minus least deprived three tenths | difference between the unweighted means (shares for pain) in the most and the least deprived band of counties (tenths 8 to 10 and 1 to 3 of interviews by county ADI), interval clustered by the respondent's county | respondents in the richest household income bracket, in the two extreme county bands | interviews in 2008 to 2017 | 95% CI 1.2 to 2.1 | 320,286 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| r_county_enjoy | +0.59 | correlation coefficient | correlation across counties between the county mean ladder score and the county share reporting the feeling, counties weighted by their respondents, bootstrap over counties | counties with at least 100 respondents with a ladder score and an answer to the enjoyment item | interviews in the years the ladder (2009 to 2017) and the feeling were both asked | 95% CI 0.56 to 0.63 | 2,356 counties | respondents with the item in each county | Gallup U.S. Daily Poll |   |
| r_county_pain | −0.62 | correlation coefficient | correlation across counties between the county mean ladder score and the county share reporting the feeling, counties weighted by their respondents, bootstrap over counties | counties with at least 100 respondents with a ladder score and an answer to the physical pain item | interviews in the years the ladder (2009 to 2017) and the feeling were both asked | 95% CI −0.66 to −0.58 | 2,356 counties | respondents with the item in each county | Gallup U.S. Daily Poll |   |
| r_county_sad | −0.57 | correlation coefficient | correlation across counties between the county mean ladder score and the county share reporting the feeling, counties weighted by their respondents, bootstrap over counties | counties with at least 100 respondents with a ladder score and an answer to the sadness item | interviews in the years the ladder (2009 to 2017) and the feeling were both asked | 95% CI −0.61 to −0.53 | 2,355 counties | respondents with the item in each county | Gallup U.S. Daily Poll |   |
| r_county_stress | −0.20 | correlation coefficient | correlation across counties between the county mean ladder score and the county share reporting the feeling, counties weighted by their respondents, bootstrap over counties | counties with at least 100 respondents with a ladder score and an answer to the stress item | interviews in the years the ladder (2009 to 2017) and the feeling were both asked | 95% CI −0.26 to −0.14 | 2,356 counties | respondents with the item in each county | Gallup U.S. Daily Poll |   |
| r_des_ladder | −0.40 | correlation coefficient | correlation across counties between the county mean of the item and the JEC county rate of deaths of despair (CDC multiple cause of death, 2015), counties weighted by their respondents with the item, bootstrap over counties | counties with a deaths-of-despair rate and at least 100 respondents with each item | interviews 2009 to 2017; deaths of despair 2015 | 95% CI −0.47 to −0.33 | 1,337 counties | respondents with the item in each county | Gallup U.S. Daily Poll |   |
| r_des_ladder5 | −0.54 | correlation coefficient | correlation across counties between the county mean of the item and the JEC county rate of deaths of despair (CDC multiple cause of death, 2015), counties weighted by their respondents with the item, bootstrap over counties | counties with a deaths-of-despair rate and at least 100 respondents with each item | interviews 2009 to 2017; deaths of despair 2015 | 95% CI −0.60 to −0.49 | 1,337 counties | respondents with the item in each county | Gallup U.S. Daily Poll |   |
| r_des_ladder_adj | +0.14 | correlation coefficient | correlation across counties between the county mean of the cell-adjusted item (five-year age band (age capped at 85), sex, race, three-level education and household income bracket) and the JEC county rate of deaths of despair, counties weighted by their respondents, bootstrap over counties | counties with a deaths-of-despair rate and at least 100 respondents with each item (respondents with age, sex, race, education, household income and a county rent) | interviews 2009 to 2017; deaths of despair 2015 | 95% CI 0.07 to 0.22 | 1,331 counties | respondents with the item in each county | Gallup U.S. Daily Poll |   |
| r_des_ladder_agesex | −0.47 | correlation coefficient | correlation across counties between the county mean ladder score after comparing respondents within cells of five-year age band and sex and the JEC county rate of deaths of despair, counties weighted by their respondents, bootstrap over counties | counties with a deaths-of-despair rate and at least 100 respondents with each item (respondents with age and sex) | interviews 2009 to 2017; deaths of despair 2015 | 95% CI −0.53 to −0.40 | 1,331 counties | respondents with the item in each county | Gallup U.S. Daily Poll |   |
| r_des_pain | +0.65 | correlation coefficient | correlation across counties between the county mean of the item and the JEC county rate of deaths of despair (CDC multiple cause of death, 2015), counties weighted by their respondents with the item, bootstrap over counties | counties with a deaths-of-despair rate and at least 100 respondents with each item | interviews 2008 to 2017; deaths of despair 2015 | 95% CI 0.61 to 0.69 | 1,337 counties | respondents with the item in each county | Gallup U.S. Daily Poll |   |
| r_des_pain_adj | +0.43 | correlation coefficient | correlation across counties between the county mean of the cell-adjusted item (five-year age band (age capped at 85), sex, race, three-level education and household income bracket) and the JEC county rate of deaths of despair, counties weighted by their respondents, bootstrap over counties | counties with a deaths-of-despair rate and at least 100 respondents with each item (respondents with age, sex, race, education, household income and a county rent) | interviews 2008 to 2017; deaths of despair 2015 | 95% CI 0.38 to 0.48 | 1,331 counties | respondents with the item in each county | Gallup U.S. Daily Poll |   |
| r_des_pain_agesex | +0.62 | correlation coefficient | correlation across counties between the county mean of the item after comparing respondents within cells of five-year age band and sex and the JEC county rate of deaths of despair, counties weighted by their respondents, bootstrap over counties | counties with a deaths-of-despair rate and at least 100 respondents with each item (respondents with age and sex) | interviews 2008 to 2017; deaths of despair 2015 | 95% CI 0.58 to 0.66 | 1,331 counties | respondents with the item in each county | Gallup U.S. Daily Poll |   |
| r_des_pain_unw | +0.58 | correlation coefficient | correlation across counties between the county mean of the item and the JEC county rate of deaths of despair (CDC multiple cause of death, 2015), each county counted once, bootstrap over counties | counties with a deaths-of-despair rate and at least 100 respondents with each item | interviews 2008 to 2017; deaths of despair 2015 | 95% CI 0.54 to 0.63 | 1,337 counties | none (each county counts once) | Gallup U.S. Daily Poll |   |
| r_des_sad | +0.31 | correlation coefficient | correlation across counties between the county mean of the item and the JEC county rate of deaths of despair (CDC multiple cause of death, 2015), counties weighted by their respondents with the item, bootstrap over counties | counties with a deaths-of-despair rate and at least 100 respondents with each item | interviews 2008 to 2017; deaths of despair 2015 | 95% CI 0.20 to 0.41 | 1,337 counties | respondents with the item in each county | Gallup U.S. Daily Poll |   |
| r_des_sad_unw | +0.45 | correlation coefficient | correlation across counties between the county mean of the item and the JEC county rate of deaths of despair (CDC multiple cause of death, 2015), each county counted once, bootstrap over counties | counties with a deaths-of-despair rate and at least 100 respondents with each item | interviews 2008 to 2017; deaths of despair 2015 | 95% CI 0.39 to 0.52 | 1,337 counties | none (each county counts once) | Gallup U.S. Daily Poll |   |
| r_des_stress | −0.09 | correlation coefficient | correlation across counties between the county mean of the item and the JEC county rate of deaths of despair (CDC multiple cause of death, 2015), counties weighted by their respondents with the item, bootstrap over counties | counties with a deaths-of-despair rate and at least 100 respondents with each item | interviews 2008 to 2017; deaths of despair 2015 | 95% CI −0.17 to −0.01 | 1,337 counties | respondents with the item in each county | Gallup U.S. Daily Poll |   |
| r_des_stress_unw | +0.19 | correlation coefficient | correlation across counties between the county mean of the item and the JEC county rate of deaths of despair (CDC multiple cause of death, 2015), each county counted once, bootstrap over counties | counties with a deaths-of-despair rate and at least 100 respondents with each item | interviews 2008 to 2017; deaths of despair 2015 | 95% CI 0.11 to 0.28 | 1,337 counties | none (each county counts once) | Gallup U.S. Daily Poll |   |
| r_des_worry | +0.06 | correlation coefficient | correlation across counties between the county mean of the item and the JEC county rate of deaths of despair (CDC multiple cause of death, 2015), counties weighted by their respondents with the item, bootstrap over counties | counties with a deaths-of-despair rate and at least 100 respondents with each item | interviews 2008 to 2017; deaths of despair 2015 | 95% CI −0.05 to 0.17 | 1,337 counties | respondents with the item in each county | Gallup U.S. Daily Poll |   |
| r_des_worry_unw | +0.36 | correlation coefficient | correlation across counties between the county mean of the item and the JEC county rate of deaths of despair (CDC multiple cause of death, 2015), each county counted once, bootstrap over counties | counties with a deaths-of-despair rate and at least 100 respondents with each item | interviews 2008 to 2017; deaths of despair 2015 | 95% CI 0.29 to 0.45 | 1,337 counties | none (each county counts once) | Gallup U.S. Daily Poll |   |
| r_jec_wonder | +0.81 | correlation coefficient | correlation across these counties between the JEC 2015 rate of deaths of despair and the WONDER rate, counties weighted by their respondents with the pain item, bootstrap over counties; the WONDER rate is the sum of the county crude rates of drug, alcohol and suicide deaths per 100,000, 2018 to 2023, which includes suicides by drug poisoning in both the drug and the suicide rate | counties with a JEC rate, a complete WONDER rate and at least 100 respondents with each of the ladder, pain and stress items | interviews: ladder 2009 to 2017, feelings 2008 to 2017; JEC deaths of despair 2015; WONDER deaths 2018 to 2023 | 95% CI 0.77 to 0.84 | 534 counties | respondents with the pain item in each county | CDC WONDER underlying cause of death |   |
| r_pain_jec_same | +0.69 | correlation coefficient | correlation across these counties between the county share reporting physical pain and the JEC 2015 rate, counties weighted by their respondents with the pain item, bootstrap over counties; the WONDER rate is the sum of the county crude rates of drug, alcohol and suicide deaths per 100,000, 2018 to 2023, which includes suicides by drug poisoning in both the drug and the suicide rate | counties with a JEC rate, a complete WONDER rate and at least 100 respondents with each of the ladder, pain and stress items | interviews: ladder 2009 to 2017, feelings 2008 to 2017; JEC deaths of despair 2015; WONDER deaths 2018 to 2023 | 95% CI 0.64 to 0.74 | 534 counties | respondents with the pain item in each county | Gallup U.S. Daily Poll |   |
| r_pain_wonder | +0.58 | correlation coefficient | correlation across these counties between the county share reporting physical pain and the WONDER rate, counties weighted by their respondents with the pain item, bootstrap over counties; the WONDER rate is the sum of the county crude rates of drug, alcohol and suicide deaths per 100,000, 2018 to 2023, which includes suicides by drug poisoning in both the drug and the suicide rate | counties with a JEC rate, a complete WONDER rate and at least 100 respondents with each of the ladder, pain and stress items | interviews: ladder 2009 to 2017, feelings 2008 to 2017; JEC deaths of despair 2015; WONDER deaths 2018 to 2023 | 95% CI 0.51 to 0.64 | 534 counties | respondents with the pain item in each county | CDC WONDER underlying cause of death |   |
| r_person_anger | −0.18 | correlation coefficient | correlation between the ladder score and the 0/1 answer across respondents, unweighted, bootstrap over counties | respondents with a ladder score and an answer to the anger item | interviews in the years the ladder (2009 to 2017) and the feeling were both asked | 95% CI −0.18 to −0.18 | 1,686,646 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| r_person_enjoy | +0.26 | correlation coefficient | correlation between the ladder score and the 0/1 answer across respondents, unweighted, bootstrap over counties | respondents with a ladder score and an answer to the enjoyment item | interviews in the years the ladder (2009 to 2017) and the feeling were both asked | 95% CI 0.26 to 0.26 | 2,211,300 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| r_person_pain | −0.20 | correlation coefficient | correlation between the ladder score and the 0/1 answer across respondents, unweighted, bootstrap over counties | respondents with a ladder score and an answer to the physical pain item | interviews in the years the ladder (2009 to 2017) and the feeling were both asked | 95% CI −0.20 to −0.20 | 2,213,777 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| r_person_worry | −0.28 | correlation coefficient | correlation between the ladder score and the 0/1 answer across respondents, unweighted, bootstrap over counties | respondents with a ladder score and an answer to the worry item | interviews in the years the ladder (2009 to 2017) and the feeling were both asked | 95% CI −0.28 to −0.28 | 2,214,180 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| r_stress_wonder | −0.14 | correlation coefficient | correlation across these counties between the county share reporting stress and the WONDER rate, counties weighted by their respondents with the pain item, bootstrap over counties; the WONDER rate is the sum of the county crude rates of drug, alcohol and suicide deaths per 100,000, 2018 to 2023, which includes suicides by drug poisoning in both the drug and the suicide rate | counties with a JEC rate, a complete WONDER rate and at least 100 respondents with each of the ladder, pain and stress items | interviews: ladder 2009 to 2017, feelings 2008 to 2017; JEC deaths of despair 2015; WONDER deaths 2018 to 2023 | 95% CI −0.23 to −0.05 | 534 counties | respondents with the pain item in each county | CDC WONDER underlying cause of death |   |
| rent_d1 | $2,697 | dollars per month, HUD two-bedroom Fair Market Rent | unweighted mean over interviews of the FY2025 two-bedroom Fair Market Rent of the respondent's county (a stand-in for relative price levels in the interview years), interval clustered by the respondent's county | respondents in the least deprived tenth of interviews, ranked by the Area Deprivation Index of the respondent's county | HUD fiscal year 2025 | 95% CI $2,594 to $2,800 | 353,555 respondents | none (unweighted) | HUD Fair Market Rents | The rent is an FY2025 area standard, shared by the counties of a metro, used here only as a proxy for relative price levels. |
| rent_d10 | $943 | dollars per month, HUD two-bedroom Fair Market Rent | unweighted mean over interviews of the FY2025 two-bedroom Fair Market Rent of the respondent's county (a stand-in for relative price levels in the interview years), interval clustered by the respondent's county | respondents in the most deprived tenth of interviews, ranked by the Area Deprivation Index of the respondent's county | HUD fiscal year 2025 | 95% CI $932 to $953 | 351,962 respondents | none (unweighted) | HUD Fair Market Rents | The rent is an FY2025 area standard, shared by the counties of a metro, used here only as a proxy for relative price levels. |
| share_enjoy | 86% | percent reporting enjoyment a lot of the day yesterday | unweighted share of respondents who reported the feeling | respondents with a ladder score and an answer to the enjoyment item | interviews in the years the item was asked |   | 2,211,300 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| share_income_reported | 81% | percent of interviews that report a household income bracket | share of interviews with a household income bracket (unweighted count) | interviews with a county | 2008 to 2017 |   | 3,519,539 respondents | none (unweighted count) | Gallup U.S. Daily Poll |   |
| share_sad | 17% | percent reporting sadness a lot of the day yesterday | unweighted share of respondents who reported the feeling | respondents with a ladder score and an answer to the sadness item | interviews in the years the item was asked |   | 2,054,693 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| share_zip_adi | 97.6% | percent of interviews with a county whose ZIP code has a ZCTA Area Deprivation Index | share of interviews that take a ZIP-area deprivation rank (the ZIP code is treated as the ZCTA of the same number) | interviews with a county | ADI 2023 |   | 3,519,539 respondents | none (unweighted count) | Gallup U.S. Daily Poll |   |
| slope_pooled | 0.250 | ladder points per household income bracket step | coefficient on the household income bracket (1 to 6) in a regression of the ladder score with no other terms, unweighted, errors clustered on county | respondents with a household income bracket and a ladder score | interviews in 2009 to 2017 | 95% CI 0.246 to 0.253 | 1,806,805 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| slope_pooled_with08 | 0.256 | ladder points per household income bracket step | the pooled income slope of the ladder recomputed with the 2008 ladder answers left in, unweighted, errors clustered on county | respondents with a household income bracket and a ladder score, including 2008 | interviews in 2008 to 2017 including the 2008 ladder, which carries a question-order artifact | 95% CI 0.252 to 0.259 | 2,074,160 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| slope_within | 0.251 | ladder points per household income bracket step | coefficient on the household income bracket (1 to 6) in a regression of the ladder score with county fixed effects, unweighted, errors clustered on county | respondents with a household income bracket and a ladder score | interviews in 2009 to 2017 | 95% CI 0.247 to 0.255 | 1,806,803 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| step_age_sex_county | −0.30 | ladder points, most deprived tenth minus least deprived tenth | gap between the extreme tenths after comparing respondents within cells of five-year age band and sex, interval clustered by the respondent's county | respondents with age, sex, race, education, household income and a county rent in the two extreme county tenths | interviews in 2009 to 2017 | 95% CI −0.36 to −0.23 | 350,199 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| step_edu_county | −0.11 | ladder points, most deprived tenth minus least deprived tenth | gap between the extreme tenths after comparing respondents within cells of five-year age band (age capped at 85), sex and three-level education, interval clustered by the respondent's county | respondents with age, sex, race, education, household income and a county rent in the two extreme county tenths | interviews in 2009 to 2017 | 95% CI −0.15 to −0.07 | 350,199 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| step_edu_zip | −0.21 | ladder points, most deprived tenth minus least deprived tenth | gap between the extreme tenths after comparing respondents within cells of five-year age band (age capped at 85), sex and three-level education, interval clustered by the respondent's county | respondents with age, sex, race, education, household income and a county rent in the two extreme ZIP area tenths | interviews in 2009 to 2017 | 95% CI −0.25 to −0.17 | 343,365 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| step_income_county | +0.13 | ladder points, most deprived tenth minus least deprived tenth | gap between the extreme tenths after comparing respondents within cells of five-year age band (age capped at 85), sex, race, three-level education and household income bracket, interval clustered by the respondent's county | respondents with age, sex, race, education, household income and a county rent in the two extreme county tenths | interviews in 2009 to 2017 | 95% CI 0.10 to 0.16 | 350,199 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| step_none_county | −0.27 | ladder points, most deprived tenth minus least deprived tenth | gap between the extreme tenths on the interviews that report every characteristic, before any adjustment, interval clustered by the respondent's county | respondents with age, sex, race, education, household income and a county rent in the two extreme county tenths | interviews in 2009 to 2017 | 95% CI −0.34 to −0.21 | 350,199 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| stress_inc1 | 43.6% | percent reporting stress a lot of the day yesterday | unweighted share, interval clustered by the respondent's county | respondents in the poorest household income bracket | interviews in 2008 to 2017 | 95% CI 43.4% to 43.9% | 441,398 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| stress_inc6 | 37.7% | percent reporting stress a lot of the day yesterday | unweighted share, interval clustered by the respondent's county | respondents in the richest household income bracket | interviews in 2008 to 2017 | 95% CI 37.3% to 38.0% | 528,004 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| stress_q1 | 37.4% | percent reporting stress a lot of the day yesterday | unweighted share, interval clustered by the respondent's county | respondents in the lowest fifth of interviews, ranked by the deaths-of-despair rate of the respondent's county | interviews 2008 to 2017; deaths of despair 2015 | 95% CI 36.9% to 37.9% | 465,362 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| stress_q5 | 36.4% | percent reporting stress a lot of the day yesterday | unweighted share, interval clustered by the respondent's county | respondents in the highest fifth of interviews, ranked by the deaths-of-despair rate of the respondent's county | interviews 2008 to 2017; deaths of despair 2015 | 95% CI 36.1% to 36.8% | 474,370 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| stress_range_2to6 | 2.3 | percentage points (highest minus lowest bracket) | the highest minus the lowest unweighted share reporting stress among the second to sixth household income brackets | respondents in the five household income brackets above the poorest | interviews in 2008 to 2017 |   |   | none (unweighted) | Gallup U.S. Daily Poll |   |
| vsplit_between | 29.7 | percent of the variance of tract life expectancy | share of the population-weighted variance of tract life expectancy at birth between counties, counting states and counties within states together (the sum of the two canonical shares from b2.canon.variance_split) | residents of non-institutional tracts with a life expectancy estimate (50 states and DC) | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts) |   | 78,008 tracts | ACS 2019-2023 tract population | USALEEP life expectancy |   |
| vsplit_county | 15.9 | percent of the variance of tract life expectancy | share of the population-weighted variance of tract life expectancy at birth between counties within states (sum-of-squares decomposition) | residents of non-institutional tracts with a life expectancy estimate (50 states and DC) | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts) |   | 78,008 tracts | ACS 2019-2023 tract population | USALEEP life expectancy | Connecticut's nine planning regions stand in for its counties. The tract estimates are modeled; their mean squared standard error equals 26.7 percent of the total variance and sits inside the within-county share. |
| vsplit_state | 13.7 | percent of the variance of tract life expectancy | share of the population-weighted variance of tract life expectancy at birth between states (sum-of-squares decomposition) | residents of non-institutional tracts with a life expectancy estimate (50 states and DC) | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts) |   | 78,008 tracts | ACS 2019-2023 tract population | USALEEP life expectancy | Connecticut's nine planning regions stand in for its counties. The tract estimates are modeled; their mean squared standard error equals 26.7 percent of the total variance and sits inside the within-county share. |

## Definitions

- Survey weights: Numbers attached to survey respondents that say how many people in the population each one represents, so that results describe the whole population and not only the people who answered.
- Cantril ladder: A survey question that asks people to place their life on a ladder whose bottom step is the worst possible life and whose top step is the best possible. It measures how people evaluate their life as a whole, not how they felt yesterday.
- Area Deprivation Index (ADI): A score that ranks neighborhoods by income, education, employment and housing conditions drawn from the American Community Survey. In this book a higher national percentile means a more deprived neighborhood.
- ZCTA: ZIP Code Tabulation Area, the Census Bureau's approximation of a ZIP code as a set of census blocks. It lets census data be reported by ZIP code, though some ZIP codes have no matching ZCTA.
- ZIP code: A postal delivery code assigned by the U.S. Postal Service. It describes mail routes, not a fixed area, so ZIP-keyed data must be matched to ZCTAs before they are mapped with census data.
- SD units (SEDA): Test scores stated in units of the typical spread of scores around the national average, so that a score reads as the number of such spreads above or below the national average. SEDA also reports scores in grade levels, which are a separate scale.
- Fixed effects: A regression adjustment that gives each group, such as each county, its own baseline. The estimate then rests on differences within groups.

## Methods

This chapter compares Gallup interviewees across tenths of county deprivation, first as they are and then within cells of age, sex, race, education and household income. The extract has no survey weights, so every estimate is unweighted, and an interval around an unweighted average does not make it nationally representative. The ladder gap between the most and least deprived tenths changes sign with the income measure: it turns positive at the same nominal household income and is near zero or below once county rent levels are allowed for. The ladder and the feelings are asked in different years and are never combined.

### Interview sample

Gallup respondent rows are read inside Box and aggregated; nothing at respondent level is published. Every Gallup estimate is unweighted, because the extract has no survey weights, and levels can differ from the weighted figures Gallup publishes.

Every comparison within cells uses one sample, the interviews that report every cell variable and a county rent, so the cell means come from the same interviews at every step.

### Questions and years

The ladder is scored 0 to 10; the second ladder item asks where the respondent expects to stand in about five years. Both ladder items are cut to 2009 to 2017, because political questions preceded the ladder in 2008 and the 2008 ladder runs low (Deaton 2012; Gallup removed the questions for part of the sample in early January 2009 and added a transition question in April 2009). The ladder is asked of nearly all respondents in 2010 to 2012, of 85% in 2009 and of 49% from 2013 to 2017, so the pooled ladder figures weight 2009 to 2012 most.

Feelings are answers (yes or no) to whether the respondent experienced the feeling during a lot of the day yesterday. The feelings are not affected by the 2008 question order and keep 2008, so the pooled feelings figures weight 2008 to 2012 most. Each item is counted in the years it was asked, which are listed in the item-coverage figure, and a few stray answers in years when an item was not asked are dropped.

The ladder and the feelings are never combined into one index.

Explains Table 1.

### County assignment

The county of an interview comes from the respondent's ZIP code and is approximate near county lines; interviews with no county are dropped.

County deprivation is the mean of the Area Deprivation Index 2023 national ranks of a county's tracts, weighted by 2019-2023 tract population (Connecticut tracts move to the eight legacy counties, the retired Alaska and South Dakota codes move to their successors, Puerto Rico is excluded, and the split county of Valdez-Cordova has no value). ZIP-area deprivation is the 2023 ZCTA national rank of the ZIP code's number, an area-weighted approximation; a ZIP code is not a ZCTA, the ZIP-to-ZCTA crosswalk is not staged, and Connecticut has no ZCTAs. Deciles of deprivation are cut once on every interview that has the measure, so each holds about a tenth of interviews.

County means for the correlations need at least 100 respondents with the item, and the same counties enter every row. The deaths-of-despair rate is a single 2015 cross-section for the larger counties, joined on the county code. The Joint Economic Committee's file keeps it as a cross-check series and does not document its population base or age adjustment, so the chapter checks it against CDC WONDER: on counties whose drug, alcohol and suicide counts are published in every year 2018 to 2023 (complete counties, as in the chapter on overdose, alcohol and suicide), the two rates correlate +0.81 across 534 counties; the WONDER rate sums three crude rates and includes suicides by drug poisoning in both the drug and the suicide rate.

### Adjusted comparisons

Comparing respondents within cells subtracts from each respondent's value the mean of everyone in the same cell (five-year age band (age capped at 85), sex, race, three-level education and household income bracket) and adds the overall mean; the adjusted group means then compare like with like. Education has three levels in the cells (a six-level item gives 0.13 for the main comparison, against +0.13 with three levels) and ages above the oldest five-year band share it.

The sign of the ladder gap depends on how income is measured. Household income is the harmonized six-bracket variable, in nominal dollars; the top bracket has no upper limit. The price sensitivities add a quintile of the county's HUD two-bedroom Fair Market Rent for fiscal year 2025, or its logarithm as a linear control, to the cells; counties in a metro share one rent, New England towns are averaged to legacy counties with HUD's 2022 population, and the FY2025 rent stands in for relative price levels in the interview years.

The equal-county version averages the cell-adjusted values within each county and then compares the means of counties with at least the survey floor of interviews, each county counted once, with a bootstrap over counties.

Explains Table 2.

### Uncertainty

Intervals for means, shares and differences are 95% intervals clustered by county (t distribution with the number of counties minus one). A bootstrap over counties resamples whole counties to see how much an estimate varies; the correlations across counties, the intraclass correlation and the share of a gap that adjustment removes use a bootstrap over whole counties with 2,000 draws (seed 12345, shared draws within a statistic). The intraclass correlation is the one-way ANOVA estimator.

### Four rankings of the same places

The comparison follows the book's ranking switch rule, fixed on October 28, 2026 before any of these gradients was computed. The main gradient of the chapter is recomputed under four rankings of the same places, with the outcome, period, weights, bins and intervals unchanged: the published ADI; a standardized index that applies the ADI's coefficients to its measures after standardizing each one across tracts; median household income; and the share of adults 25 and older with a bachelor's degree or more. The three alternatives come from ACS tract estimates for 2019 to 2023, the period of the ADI.

All four rankings use the same places, the 3,121 counties that have the outcome and all four rankings, so the ADI panel can differ slightly from the chapter's main figure. Tracts enter only if they have every input of the standardized index, which leaves out tracts without a median rent, home value or mortgage cost. A respondent's county takes the population-weighted mean of the ranks of its common-sample tracts, with Connecticut tracts moved to the eight legacy counties and retired Alaska and South Dakota codes moved to their successors, and the tenths are tenths of interviews, cut once on every interview whose county has all four rankings, as in the chapter's main figure.

A pattern counts as changed when the worse end switches sides, when the two ends differ under one ranking and not the other, or when the rank correlation of bin order and bin mean falls on the other side of 0.9 and moves by at least 0.1.

Explains Table 8.

## What these data cannot show

The unit of the chapter is the person, but the place measures describe counties and ZIP areas, so a gap between tenths of places mixes the places with the people in them. Comparing respondents within cells handles only the characteristics the survey records. It does not handle health, disability, work or family, which go with pain. Household income is in nominal dollars and the rent file is a single fiscal year 2025 standard shared within metros, so the price comparison is a rough proxy, and neither it nor the nominal comparison identifies what places do to residents. The randomized Moving to Opportunity experiment is the evidence on that question, and it covers neither the ladder nor these feelings.

The survey has no weights, and the interviews are a repeated cross-section, not a panel, so the chapter shows differences between people and not changes within them. Pooled figures weight 2009 to 2012 most for the ladder and 2008 to 2012 for the feelings, and the chapter does not remove differences between years. Each feeling is a yes or no answer to whether the respondent had it during a lot of the day yesterday, which records neither intensity nor duration, and the ladder is a whole-number rating whose rungs different people may anchor differently. Answers given in one interview can color each other, as the 2008 ladder shows, and the survey cannot separate that for the other items.

The county is derived from the ZIP code and is approximate near county lines. ZIP-area deprivation treats a ZIP code as the ZCTA of the same number and drops Connecticut. County deprivation is built from Area Deprivation Index ranks that include income and education, so the index overlaps with the respondent characteristics used in the comparisons. The index is the 2023 release, built on ACS data for 2019 to 2023, and the interviews are from 2008 to 2017, which assumes that the ranking of places persisted. The deaths-of-despair rate is a single year for larger counties, a cross-check series with an undocumented population base and age adjustment, and its correlations are across counties, so they do not show that the people who report pain are the people who die.

Chapter 4.5, Mental distress in young adults (https://vishalsingh.org/zipcode/mental-distress-in-young-adults) measures frequent mental distress in a different survey with survey weights, and its estimates are not directly comparable with the unweighted Gallup levels here.

## References

1. Cantril, Hadley (1965). The Pattern of Human Concerns. Rutgers University Press, New Brunswick, NJ.
2. Deaton, Angus (2012). The financial crisis and the well-being of Americans: 2011 OEP Hicks Lecture. Oxford Economic Papers 64(1): 1-26. https://doi.org/10.1093/oep/gpr051
3. Kahneman, Daniel; Deaton, Angus (2010). High income improves evaluation of life but not emotional well-being. Proceedings of the National Academy of Sciences 107(38): 16489-16493. https://doi.org/10.1073/pnas.1011492107
4. Killingsworth, Matthew A.; Kahneman, Daniel; Mellers, Barbara (2023). Income and emotional well-being: A conflict resolved. Proceedings of the National Academy of Sciences 120(10): e2208661120. https://doi.org/10.1073/pnas.2208661120
5. Graham, Carol; Pinto, Sergio (2019). Unequal hopes and lives in the USA: optimism, race, place, and premature mortality. Journal of Population Economics 32(2): 665-733. https://doi.org/10.1007/s00148-018-0687-y
6. Ludwig, Jens; Duncan, Greg J.; Gennetian, Lisa A.; Katz, Lawrence F.; Kessler, Ronald C.; Kling, Jeffrey R.; Sanbonmatsu, Lisa (2012). Neighborhood effects on the long-term well-being of low-income adults. Science 337(6101): 1505-1510. https://doi.org/10.1126/science.1224648
7. Case, Anne; Deaton, Angus (2020). Deaths of Despair and the Future of Capitalism. Princeton University Press, Princeton, NJ.
8. Case, Anne; Deaton, Angus (2015). Rising morbidity and mortality in midlife among white non-Hispanic Americans in the 21st century. Proceedings of the National Academy of Sciences 112(49): 15078-15083. https://doi.org/10.1073/pnas.1518393112

## Provenance

- Pipeline version: 0.1.0
- Build script: book2/chapters/life-evaluation-and-daily-feelings/build.py (sha256 c10703bb40e85185d9a6f2ec0f51d6520d5ce5834aa06016094a33af1474a9a4)
- Built on: 2026-10-07
- Site commit: c2b1ff3309cb4a10833f884da91ac0057a66ad0a

Inputs:

| File | Size (bytes) | SHA-256, first 1 MB | SHA-256, last 1 MB |
| --- | ---: | --- | --- |
| book2/chapters/life-evaluation-and-daily-feelings/wellbeing_data.py | 43,672 | dc8fb82fa180829f225d1c6f6435480a8f83c924147a32d66d7f5510eec2b225 | dc8fb82fa180829f225d1c6f6435480a8f83c924147a32d66d7f5510eec2b225 |
| data/analytic/tract_atlas.parquet | 39,659,925 | 29d5e4d943552d54aeb6ba92aadc939d9bd15bdea553c78357dbd516757be627 | 1e10d1ecf722340f23ee2212d4999a90250f85b65fa89b39e32db06d9472e22e |
| data/geo/dim_geo.parquet | 1,180,921 | 29d5e4d943552d54aeb6ba92aadc939d9bd15bdea553c78357dbd516757be627 | e58c8c488df54ab3187f5461f179322afdc49c9bd2315404003fe1272eea32c3 |
| data/raw/acs/deprivation_components_2019_2023/tract.parquet | 13,197,793 | 5545cde2cbe96a8b56e468d65388248a241aa6bd300d3c9e325d08d3f86b11da | 5c8ec454d066988b0f1a01aaa6ed34ebfc05d162e01cf9c6bb01f9198cbaff12 |
| data/raw/gallup/gallup_us.parquet | 209,530,597 | b57c719f2c1d4749e43b1a70ca6dbd97f0480a0b538983091eea7eadfec40551 | b8167235969e50da6143a0e8d3e74ffb8f7c231743acc9fb2a78130a602af6f5 |
| data/staged/adi/adi_atlas_tract2020_2023.parquet | 798,078 | 5e409bc212ff871def780fa27b4cf8b616d9da420b2b9f6455517eddcd8c1a95 | 5e409bc212ff871def780fa27b4cf8b616d9da420b2b9f6455517eddcd8c1a95 |
| data/staged/adi/adi_atlas_zcta_2023.parquet | 766,397 | 5f53b97a981f9e88dd58af9e756554119580de309d750b7af6c2206bd79b6415 | 5f53b97a981f9e88dd58af9e756554119580de309d750b7af6c2206bd79b6415 |
| data/staged/crosswalks/ct_tract2020_to_old_county.parquet | 15,842 | 961ef39f31596ef65fca2c67d5f2f63e533c5dc9b49745d3bb6437844e4b831d | 961ef39f31596ef65fca2c67d5f2f63e533c5dc9b49745d3bb6437844e4b831d |
| data/staged/housing/hud_fmr_area_fy2025.parquet | 220,184 | e8acd06036b1bf0c46d350385c9cf08de423cd7145ead42ac2b6bf9367361b7d | e8acd06036b1bf0c46d350385c9cf08de423cd7145ead42ac2b6bf9367361b7d |
| data/staged/life_expectancy/usaleep_tract2020.parquet | 1,316,169 | 9c9c7f103791b75eddba74e68a89804efdf2486b6b66bb4bc02a46ba5bf8bc23 | 27317107be2465d9fd5e83271dea0cd6e915ee419e2802bedc28747d0a41d38f |
| data/staged/mortality/wonder_ucd_alcohol_county_2018_2024.parquet | 289,631 | 478b6b5f8ffa061bcca1df84162d1e169babeb7832bc9818379af9be382a778f | 478b6b5f8ffa061bcca1df84162d1e169babeb7832bc9818379af9be382a778f |
| data/staged/mortality/wonder_ucd_drug_county_2018_2024.parquet | 308,988 | b0e5cb6dfb44fc1ce53a192faf74cb84277bb5d601a3fc32b65078a7ec03bbf1 | b0e5cb6dfb44fc1ce53a192faf74cb84277bb5d601a3fc32b65078a7ec03bbf1 |
| data/staged/mortality/wonder_ucd_suicide_county_2018_2024.parquet | 290,192 | 448558540a3f42296acf7f66a945f7c6794fd6f4dc9c7032a18a3d105d0541f1 | 448558540a3f42296acf7f66a945f7c6794fd6f4dc9c7032a18a3d105d0541f1 |
| data/staged/mortality/wonder_ucd_suicide_state_2018_2024.parquet | 15,948 | 7588fe51ddda48e57e88852172442b4d2bf64689882e600ebf342668ff79ea1a | 7588fe51ddda48e57e88852172442b4d2bf64689882e600ebf342668ff79ea1a |
| data/staged/ses/acs5_tract2020_2019_2023.parquet | 7,232,424 | 5fe6f730a35225f14d5d8abfc6b9bce58b52e545f309a7eac74f1dae74d7a31c | 7b4a2475b3852d3f91804716a04566ef8ceadacec2be9fb85f5eb642078efac6 |
| data/staged/social_capital/jec_social_capital_county.parquet | 460,297 | 1f042e817f25c79bc96cf0a256133c4068e655bd10caff55c2ae0dc3c8f7df22 | 1f042e817f25c79bc96cf0a256133c4068e655bd10caff55c2ae0dc3c8f7df22 |
| data/staged/svi/svi_tract2020_2022.parquet | 17,740,415 | cb3860d8edc2aaf1f2a02211a58f7a0b80457eb6ad9a37b5ec8abb2d5da9da10 | e371a4a696d5bed05d54ad7c5b6ca2504508805b847ec415c2322be7f66ec2b2 |

Outputs:

| File | SHA-256 |
| --- | --- |
| facts.json | ed1377d46cea9f92d315518cf23107c167643246e6f909612c77af42aa4f2748 |
| figures/alt_rankings.json | 3eac056a094f131714b25d7ccf86a56f742087fb29f3672721187b0e41f7d2a0 |
| figures/feelings_by_deprivation.json | 99a6721eb790cdd7c6b7d160ce88464b3cfb671b6df07e039b85abf61a2de409 |
| figures/felt_despair.json | 8e31b0d405922f512bfc224fcf4184fb48724c411bd9dc41b1d0fe4b5dbc4c2f |
| figures/income_gradient.json | 6bffedc2e663b1b57dd9961bbe3d220ac8b4c2bcb4ed58c2b17d7928f249cb1e |
| figures/income_within_places.json | eaedcd97d39a29a27cc3c95c9288ac734e1d2cdb3949e539e95d2170ed115c49 |
| figures/item_coverage.json | c235091f38314038f947f7941c5b788c9fcad1b44670c4debc284c5e533629bb |
| figures/ladder_and_feelings.json | 9fc753bc28fa69b4cd62807b730eaef4dd10cc64b46f0968579cec32d41bb428 |
| figures/ladder_by_deprivation.json | b563618cd2af45b968f7d7d33c2b7b1c36d5f02d2f51832cd765483997aff533 |
| meta.json | 46f9840e1a9f0ed18e57c20ba6706658e4b1834dd324c6a9821293151c5bf339 |
| refs.json | af3d3761f67ed01c066f1a1224342ebede46f2a79cc9e31ae92c4bc16acb298b |

## Machine formats

- Report (HTML): https://vishalsingh.org/zipcode/life-evaluation-and-daily-feelings/report
- Report (Markdown): https://vishalsingh.org/zipcode/life-evaluation-and-daily-feelings/report.md
- Every estimate (JSON): https://vishalsingh.org/zipcode/report/life-evaluation-and-daily-feelings/facts.json
- Table 1, figure "item_coverage" (JSON): https://vishalsingh.org/zipcode/report/life-evaluation-and-daily-feelings/figures/item_coverage.json
- Table 2, figure "ladder_by_deprivation" (JSON): https://vishalsingh.org/zipcode/report/life-evaluation-and-daily-feelings/figures/ladder_by_deprivation.json
- Table 3, figure "feelings_by_deprivation" (JSON): https://vishalsingh.org/zipcode/report/life-evaluation-and-daily-feelings/figures/feelings_by_deprivation.json
- Table 4, figure "income_gradient" (JSON): https://vishalsingh.org/zipcode/report/life-evaluation-and-daily-feelings/figures/income_gradient.json
- Table 5, figure "income_within_places" (JSON): https://vishalsingh.org/zipcode/report/life-evaluation-and-daily-feelings/figures/income_within_places.json
- Table 6, figure "ladder_and_feelings" (JSON): https://vishalsingh.org/zipcode/report/life-evaluation-and-daily-feelings/figures/ladder_and_feelings.json
- Table 7, figure "felt_despair" (JSON): https://vishalsingh.org/zipcode/report/life-evaluation-and-daily-feelings/figures/felt_despair.json
- Table 8, figure "alt_rankings" (JSON): https://vishalsingh.org/zipcode/report/life-evaluation-and-daily-feelings/figures/alt_rankings.json
- Index of every report: https://vishalsingh.org/zipcode/llms.txt; all reports in one file: https://vishalsingh.org/zipcode/llms-full.txt

---

# Well-being across the life span: report

Chapter 6.4 of ZIP Code Destiny, by Vishal Singh (NYU Stern). Generated from the chapter's data files; the reading version is at https://vishalsingh.org/zipcode/well-being-across-the-life-span. Table n of this report is figure n of the chapter.

## Identification

- Title: Well-being across the life span
- Label: Two clocks of a life
- Chapter: 6.4 (Part VI, Environment, connection and well-being; An American life)
- Question: How do life evaluation and daily emotions change with age, and does the pattern differ between richer and poorer places?
- Dek: In Gallup interviews from 2009 to 2017, life evaluation averaged 6.84 at ages 45 to 54 and 7.38 at ages 75 to 94, and the lowest point of the unadjusted profile by year of age falls at age 54.
- Grain: individuals
- Status: published
- Published: 2026-10-09
- Updated: not revised
- Data sources:
  - Gallup U.S. Daily Poll. Gallup U.S. Daily tracking poll, 2008-2017: Cantril ladder score (2009-2017), feelings yesterday, age, household income, sex, marital status, education and self-rated health; aggregates only; years: 2008-2017; weights: none (unweighted)
  - Area Deprivation Index. Area Deprivation Index 2023, national rank, tract ranks averaged to counties and joined to the county of each interview (2008-2017); years: 2023 release (version 4.0.1; built on ACS 2019-2023); weights: 2019-2023 tract population
  - American Community Survey. ACS 2019-2023 tract population, the weight for averaging tract deprivation ranks to counties; years: 2019-2023 (5-year); weights: 2019-2023 tract population
- Canonical URL: https://vishalsingh.org/zipcode/well-being-across-the-life-span
- Report: https://vishalsingh.org/zipcode/well-being-across-the-life-span/report
- Markdown: https://vishalsingh.org/zipcode/well-being-across-the-life-span/report.md
- BibTeX key: singh2026wellbeingacrossthelifespan

```bibtex
@misc{singh2026wellbeingacrossthelifespan,
  author = {Singh, Vishal},
  title = {Well-being across the life span},
  year = {2026},
  url = {https://vishalsingh.org/zipcode/well-being-across-the-life-span},
  note = {In ZIP Code Destiny. Published 9 October 2026.}
}
```

## Principal estimates

The facts the dek and the opening prose cite, in order of appearance.

| Estimand | Population | Period | Value | Interval | n | Weights | Source | Fact id |
| --- | --- | --- | ---: | --- | ---: | --- | --- | --- |
| unweighted mean of the 0 to 10 ladder score, interval clustered by the respondent's county | adults aged 18 to 94 interviewed by Gallup who answered the item and whose county is known (age 45 to 54) | interviews in 2009 to 2017 | 6.84 | 95% CI 6.82 to 6.86 | 380,027 respondents | none (unweighted) | Gallup U.S. Daily Poll | ladder_45_54 |
| definition used in this chapter, not an estimate | an age group of adults defined in this chapter | ladder interviews in 2009 to 2017 and feelings interviews in 2008 to 2017 | 45 to 54 |   |   | none (definition) | Gallup U.S. Daily Poll | band_4554 |
| unweighted mean of the 0 to 10 ladder score, interval clustered by the respondent's county | adults aged 18 to 94 interviewed by Gallup who answered the item and whose county is known (age 75 to 94) | interviews in 2009 to 2017 | 7.38 | 95% CI 7.37 to 7.39 | 289,487 respondents | none (unweighted) | Gallup U.S. Daily Poll | ladder_75_94 |
| definition used in this chapter, not an estimate | an age group of adults defined in this chapter | ladder interviews in 2009 to 2017 and feelings interviews in 2008 to 2017 | 75 to 94 |   |   | none (definition) | Gallup U.S. Daily Poll | band_7594 |
| age of the lowest point of the five-year moving average of the mean ladder score by year of age (ages 20 to 92 searched, refined between whole years by a parabola through the lowest point and its neighbours), interval from a bootstrap over counties | adults aged 18 to 94 interviewed by Gallup who answered the item and whose county is known | interviews in 2009 to 2017 | 54 | 95% CI 51 to 54 | 2,180,433 respondents | none (unweighted) | Gallup U.S. Daily Poll | ladder_turn |
| unweighted share of interviews that answered yes to the item, interval clustered by the respondent's county | adults aged 18 to 94 interviewed by Gallup who answered the item and whose county is known (age 25 to 34), stress | interviews in 2008 to 2017 | 48.6% | 95% CI 48.3% to 48.9% | 269,066 respondents | none (unweighted) | Gallup U.S. Daily Poll | stress_2534 |
| definition used in this chapter, not an estimate | an age group of adults defined in this chapter | ladder interviews in 2009 to 2017 and feelings interviews in 2008 to 2017 | 25 to 34 |   |   | none (definition) | Gallup U.S. Daily Poll | band_2534 |
| unweighted share of interviews that answered yes to the item, interval clustered by the respondent's county | adults aged 18 to 94 interviewed by Gallup who answered the item and whose county is known (age 75 to 94), stress | interviews in 2008 to 2017 | 18.0% | 95% CI 17.8% to 18.2% | 342,420 respondents | none (unweighted) | Gallup U.S. Daily Poll | stress_7594 |
| definition used in this chapter, not an estimate (Gallup's harmonized brackets, not adjusted for inflation) | the upper limit of the lowest household income group | interviews in 2009 to 2014 | $36,000 |   |   | none (definition) | Gallup U.S. Daily Poll | inc_cut_low |
| definition used in this chapter, not an estimate (Gallup's harmonized brackets, not adjusted for inflation) | the lower limit of the highest household income group | interviews in 2009 to 2014 | $90,000 |   |   | none (definition) | Gallup U.S. Daily Poll | inc_cut_high |
| definition used in this chapter, not an estimate | an age group of adults defined in this chapter | ladder interviews in 2009 to 2017 and feelings interviews in 2008 to 2017 | 35 to 44 |   |   | none (definition) | Gallup U.S. Daily Poll | band_3544 |
| age of the lowest point of the five-year moving average of the mean ladder score by year of age (ages 20 to 92 searched, refined between whole years by a parabola through the lowest point and its neighbours), interval from a bootstrap over counties | adults with a ladder score, sex, marital status, education and household income | interviews in 2009 to 2014 | 54 | 95% CI 51 to 55 | 1,355,746 respondents | none (unweighted) | Gallup U.S. Daily Poll | turn_cc_raw |
| age of the lowest point of the five-year moving average of the mean ladder score by year of age (ages 20 to 92 searched, refined between whole years by a parabola through the lowest point and its neighbours), interval from a bootstrap over counties | adults with a ladder score, sex, marital status, education and household income | interviews in 2009 to 2014 | 48 | 95% CI 48 to 54 | 1,355,746 respondents | none (unweighted) | Gallup U.S. Daily Poll | turn_all5 |
| age at which a quadratic in age fitted to the ladder score turns, interval from a bootstrap over counties | adults with a ladder score, sex, marital status, education and household income | interviews in 2009 to 2014 | 44 | 95% CI 44 to 45 | 1,355,746 respondents | none (unweighted) | Gallup U.S. Daily Poll | quad_raw |

## Findings

In Gallup interviews from 2009 to 2017, adults rated their lives 6.84 (95% CI 6.82 to 6.86) on average at ages 45 to 54 and 7.38 (95% CI 7.37 to 7.39) at ages 75 to 94, on the Cantril ladder, which runs from the worst possible life to the best possible one. The lowest point of the profile by year of age falls at age 54 (95% CI 51 to 54). Daily feelings follow other schedules. On the day before the interview, 48.6% of adults aged 25 to 34 reported stress for a lot of the day, against 18.0% of those aged 75 to 94.

These profiles compare people of different ages interviewed in the same years. A person aged 75 to 94 in these data was born earlier than a person aged 45 to 54, so a difference between them reflects age and birth cohort together, the cohort being the group of people born in the same years, and the oldest interviewees are also the survivors of their cohort. None of the profiles is the path of one person's life.

Two results qualify the shape of the ladder profile. The midlife fall is much larger in households with annual income under $36,000 than in households of $90,000 or more, where the ladder is level from ages 25 to 34 to ages 35 to 44 and falls only modestly afterward (Table 4). The age of the lowest point also depends on what is held fixed. It is 54 (95% CI 51 to 55) in the raw profile of interviews with household income, 48 (95% CI 48 to 54) after adjusting for sex, marital status, education, household income and survey year, and 44 (95% CI 44 to 45) in a quadratic fit (Table 5).

### What the poll can compare

The file holds 3,457,670 interviews of adults aged 18 to 94 with a known county, in 3,142 counties, 2,180,433 of them with a usable ladder score. The extract does not top-code age, and the cells at the oldest ages are smaller, so their intervals are wider. Gallup's extract has no survey weights, so every estimate here is unweighted and its levels can differ from the weighted figures Gallup publishes. Intervals are clustered by the respondent's county, which is derived from the ZIP code. The poll drew a new sample every day, so it is a repeated cross-section and not a panel, and no respondent is followed from one year to the next.

The ladder asks respondents to rate their present life as a whole. The feelings are yes or no answers about whether the respondent had a feeling for a lot of the day before the interview: enjoyment, happiness, worry, stress, sadness, anger and physical pain. The two kinds of measure are never combined into an index. The ladder, enjoyment, worry, stress and physical pain were asked in 2008 to 2017, happiness and sadness in 2008 to 2016, and anger in 2008 to 2013 and 2016. From 2013 each item was asked of 49% of interviews, and the ladder reached 85% of interviews in 2009.

The ladder of 2008 is left out of every estimate. In 2008 Gallup asked political questions before the ladder, which lowered the ladder scores, and it removed them for part of the sample in January 2009 and added a transition question in April 2009.[1] The 2008 ladder averaged 6.69 (95% CI 6.68 to 6.71), against 7.03 (95% CI 7.02 to 7.04) in 2009, and the difference is between 0.29 (95% CI 0.26 to 0.32) and 0.38 (95% CI 0.35 to 0.41) ladder points in every age band. The extract has no interview month, so early 2009 interviews may carry part of the same artifact, and the 2009 ladder is kept with that caveat. The feelings keep 2008, because the question order affected them less. The largest difference between their 2008 and 2009 levels is −1.6 (95% CI −1.8 to −1.3) points, for worry, which was lower in 2008 and not higher.

### The ladder is lowest in midlife and highest in old age

Table 1 plots the mean ladder score for each year of age. The score falls from 7.25 (95% CI 7.23 to 7.28) at age 18 to 6.93 (95% CI 6.91 to 6.96) at age 25 and changes little from ages 25 to 34 to ages 35 to 44. It falls again to its lowest five-year average at age 54 (95% CI 51 to 54) and then rises to its highest values among the oldest adults.

By age band the mean is 6.84 (95% CI 6.82 to 6.86) at ages 45 to 54, a change of −0.15 (95% CI −0.16 to −0.13) ladder points from the mean of 6.99 (95% CI 6.97 to 7.00) at ages 25 to 34. At ages 75 to 94 it is 7.38 (95% CI 7.37 to 7.39), a change of +0.54 (95% CI 0.53 to 0.55) from midlife. The rise from midlife to old age is larger than the fall into midlife.

Table 1. Mean life evaluation falls from 7.03 among adults 18 to 24 to 6.84 among adults 45 to 54 and is 7.38 among adults 75 to 94

Stone and colleagues found the same shape in the 2008 Gallup-Healthways survey, and Blanchflower and Oswald and later Blanchflower report a low in middle age across many countries.[2][3][4] The profile here is a wave more than a parabola. It begins high at age 18, with a mean of 7.03 (95% CI 7.02 to 7.05) at ages 18 to 24, and drops to a first low among young adults. It then flattens, falls to the lowest point around age 54 and climbs to a higher level than any it had before. Frijters and Beatton describe a similar wave in the raw profiles of three household panels.[5]

### Daily feelings follow other schedules

Table 2 shows the seven feelings. Enjoyment and happiness reach their lowest points in midlife, at ages 52 (95% CI 52 to 53) and 54 (95% CI 54 to 56), close to the ladder's lowest point. They differ from the ladder at the other end, where the ladder is higher at ages 75 to 94 than at ages 18 to 24.

Enjoyment is 88.3% (95% CI 88.1% to 88.6%) at ages 18 to 24, 82.6% (95% CI 82.4% to 82.8%) at ages 45 to 54 and 88.5% (95% CI 88.3% to 88.7%) at ages 75 to 94, back at its youngest level. Happiness goes from 91.9% (95% CI 91.7% to 92.1%) to 86.6% (95% CI 86.4% to 86.8%) and 90.0% (95% CI 89.8% to 90.2%), staying below its youngest level.

The negative experiences do not share that shape. Stress is highest in early adulthood, at age 26 (95% CI 25 to 33). It is 48.6% (95% CI 48.3% to 48.9%) at ages 25 to 34, 44.6% (95% CI 44.3% to 44.8%) at ages 45 to 54 and 18.0% (95% CI 17.8% to 18.2%) at ages 75 to 94.

Worry peaks in midlife, at age 48 (95% CI 47 to 49), with 35.4% (95% CI 35.2% to 35.7%) of adults at ages 45 to 54 and 18.6% (95% CI 18.4% to 18.8%) at ages 75 to 94. Anger falls from the youngest ages, from 16.5% (95% CI 16.3% to 16.8%) to 6.2% (95% CI 6.0% to 6.3%).

Sadness moves in a narrow range, 14.9% (95% CI 14.6% to 15.2%) at ages 25 to 34, 18.6% (95% CI 18.3% to 18.8%) at ages 45 to 54 and 16.1% (95% CI 15.9% to 16.3%) at ages 75 to 94, and peaks near age 55 (95% CI 55 to 55).

Physical pain rises from 15.5% at ages 18 to 24 to 28.5% at ages 55 to 64 and stays near that level afterward. It peaks near age 60 (95% CI 58 to 61) and is 26.4% (95% CI 26.2% to 26.6%) at ages 75 to 94.

Table 2. Stress, worry and anger are lower among older adults, enjoyment and happiness are lowest in midlife, and physical pain peaks near age 60

The contrast is therefore between two groups of measures. The ladder, enjoyment and happiness reach their lowest points in midlife, and stress, worry and anger are lower among older adults than among younger ones, while pain is more common. Stone and colleagues reported the same patterns for stress, anger and worry in the 2008 survey.[2] The daily feelings of older adults are not brighter in every respect, but they include fewer days of stress and worry, and the pain profile reaches a plateau after its peak and does not keep climbing.

### Richer and poorer places

To ask whether the profile differs between richer and poorer places, each interview is placed in one of three groups of about equal numbers of interviews by the Area Deprivation Index of the respondent's county, the book's measure of neighborhood deprivation. The groups are thirds of interviews and not of counties. Interviews come mostly from large counties, so the least deprived third comes from 274 counties, the middle third from 655 and the most deprived third from 2,193. The 2023 version of the index is joined to interviews from 2008 to 2017, so it describes each county as it was later. 931 interviews in counties with no rank are left out. Chapter 6.3, Life evaluation and daily feelings (https://vishalsingh.org/zipcode/life-evaluation-and-daily-feelings) asks how the ladder and the feelings differ across places at all ages. The question here is whether the age profile changes with place.

It has a midlife low and a later rise in all three groups, and the midlife fall is deeper in the most deprived third. The fall from ages 25 to 34 to ages 45 to 54 is −0.11 (95% CI −0.14 to −0.09) ladder points in the least deprived third and −0.17 (95% CI −0.19 to −0.15) in the most deprived, with intervals that do not overlap, and the difference between the two falls is −0.06 (95% CI −0.09 to −0.03) ladder points.

The lowest ladder point is at age 54 (95% CI 51 to 56) in the least deprived third, 54 (95% CI 52 to 55) in the middle third and 51 (95% CI 48 to 54) in the most deprived, so it is earlier in the most deprived third, although the intervals overlap (Table 3).

The ladder gap between the most and the least deprived third is −0.07 (95% CI −0.10 to −0.04) ladder points at ages 18 to 24, −0.21 (95% CI −0.24 to −0.17) at ages 45 to 54 and −0.12 (95% CI −0.15 to −0.09) at ages 75 to 94. It is widest in midlife, where scores average 6.94 (95% CI 6.90 to 6.97) in the least deprived third and 6.73 (95% CI 6.71 to 6.75) in the most deprived.

Table 3. The ladder falls into midlife and rises after it in every third of interviews by county deprivation, and the fall from ages 25 to 34 to ages 45 to 54 is −0.17 points in the most deprived third and −0.11 in the least

Most feelings show little place gap (the last tab of Table 3). The largest difference between the extreme thirds in any age band is 1.3 points for stress, 1.5 for enjoyment, 2.1 for sadness and 2.9 for worry.

Physical pain is the exception. It is more common in the most deprived third, with gaps of +0.9 (95% CI 0.3 to 1.4) points at ages 18 to 24 and +6.8 (95% CI 6.0 to 7.6) points at ages 45 to 54. The groups describe where a person lives, and counties differ in income, age mix and much else, so the comparison does not say what place does to well-being.

### The midlife fall is much larger in lower-income households

Table 4 repeats the profiles within three groups of households: those with annual income under $36,000, those between $36,000 and $90,000, and those at $90,000 or more. Income is the household's and includes a partner's earnings. It is reported in dollars of the interview year, and 81% of interviews report it. Each line holds the household's income bracket fixed, so it compares people of similar income at different ages. The figure uses interviews in 2009 to 2014 only.

The share of income reporters at $90,000 or more was 23% (95% CI 22% to 24%) in 2014 and 35% (95% CI 34% to 36%) in 2017, a jump larger than inflation can explain. The income groups therefore mean something different in 2015 to 2017, and those years are reported separately below. The smallest cell of the income profiles, the highest income group at the oldest ages, holds 51 interviews, so the intervals there are wide.

Within households under $36,000, the ladder averages 6.85 (95% CI 6.83 to 6.88) at ages 18 to 24, 5.89 (95% CI 5.86 to 5.91) at ages 45 to 54 and 7.04 (95% CI 7.03 to 7.06) at ages 75 to 94. The fall from ages 25 to 34 to ages 45 to 54 is −0.59 (95% CI −0.62 to −0.56) ladder points there and −0.27 (95% CI −0.29 to −0.26) between the cutoffs, where the mean at ages 45 to 54 is 6.86 (95% CI 6.85 to 6.88).

The profile of households of $90,000 or more is different but not flat. Its mean is 7.50 (95% CI 7.48 to 7.52) at ages 25 to 34, 7.51 (95% CI 7.50 to 7.53) at ages 35 to 44 and 7.45 (95% CI 7.43 to 7.46) at ages 45 to 54, so the fall from ages 25 to 34 to ages 45 to 54 is −0.05 (95% CI −0.07 to −0.03) ladder points.

The fall from ages 35 to 44 to ages 45 to 54 is −0.07 (95% CI −0.08 to −0.05) in the highest group, against −0.27 (95% CI −0.30 to −0.24) and −0.12 (95% CI −0.13 to −0.10) in the other two groups. The midlife fall in the highest group is far smaller and not absent. The rise after midlife is +1.15 (95% CI 1.13 to 1.18) in the lowest group, +0.69 (95% CI 0.67 to 0.71) in the middle group and +0.37 (95% CI 0.33 to 0.40) in the highest.

Table 4. The fall in life evaluation from ages 25 to 34 to ages 45 to 54 is −0.59 ladder points in households under $36,000 and −0.05 in households of $90,000 or more, where the fall from ages 35 to 44 is −0.07

In 2015 to 2017 the same contrasts for the highest group are +0.09 (95% CI 0.06 to 0.12) from ages 25 to 34 to ages 45 to 54 and −0.02 (95% CI −0.04 to 0.01) from ages 35 to 44 to ages 45 to 54. The first changes sign between the periods and the second loses its size, so a pooled value would hide a difference between periods. The change in the bracket mix makes the later years unsuitable for this comparison.

The feelings show the same split, except for stress. Physical pain at ages 45 to 54 is 41.2% (95% CI 40.6% to 41.8%) in households under $36,000 and 15.7% (95% CI 15.4% to 16.0%) in households of $90,000 or more, and enjoyment is 72.7% (95% CI 72.3% to 73.0%) and 87.6% (95% CI 87.3% to 87.9%). The pain gap narrows at older ages, to 30.5% (95% CI 30.2% to 30.9%) and 20.9% (95% CI 20.3% to 21.5%) at ages 75 to 94.

Stress is higher in the lowest income group in every age band, but it falls after midlife in every group. In the lowest group it goes from 52.6% (95% CI 52.1% to 53.2%) at ages 45 to 54 to 20.2% (95% CI 19.9% to 20.5%) at ages 75 to 94, and in the highest from 42.4% (95% CI 42.0% to 42.8%) to 16.6% (95% CI 15.9% to 17.2%). The decline of stress after midlife does not depend on income.

Holding income fixed removes one source of composition, since the mix of incomes differs by age. It does not remove all of it. Who is in the lowest income group in midlife differs from who is in it in early adulthood, and income is partly a consequence of health and well-being as well as a source of them. Chapter 7.4, Disability, retirement and Medicare (https://vishalsingh.org/zipcode/disability-retirement-and-medicare) covers income, disability and retirement at older ages.

### What changes the age of the lowest point

Table 5 reports the age of the lowest point under each adjustment. In the 1,355,746 interviews from 2009 to 2014 that report sex, marital status, education and household income, the lowest point of the raw profile is at age 54 (95% CI 51 to 55). Adjusting for sex leaves it at 54 (95% CI 51 to 55), for marital status at 54 (95% CI 48 to 54) and for household income at 48 (95% CI 48 to 54), and adjusting for all five gives 48 (95% CI 48 to 54).

The point estimates for income and for all five are earlier than the raw age, and their intervals reach almost to it, because the bottom of the profile is flat with a low stretch near each age and the bootstrap lands on one or the other. The raw five-year average varies by only 0.06 ladder points between ages 46 and 56, so small differences in the sample move the lowest point within that range.

Adding self-rated health to the five, in the 1,354,627 interviews that report it, gives 48 (95% CI 47 to 48), a narrower interval. A quadratic in age puts the turning age at 44 (95% CI 44 to 45) raw and 47 (95% CI 47 to 47) adjusted. The profile is not symmetric, so a parabola turns earlier than the profile does, and the narrow intervals of the quadratic describe the fit and not whether the shape is right.

Within income groups, the lowest point is at age 54 (95% CI 54 to 55) in households under $36,000 and 48 (95% CI 48 to 51) between the cutoffs. The highest group is not given a lowest point. The search starts at age 20, and its lowest five-year average over the whole search is at age 24, in early adulthood, so its midlife fall is described by the change from ages 35 to 44 to ages 45 to 54 instead.

Table 5. Adjusting for household income moves the lowest point of the ladder profile from age 54 to age 48 with overlapping intervals, and a quadratic fit puts it at age 44, so no single age is the turning point

The adjustments deepen the dip as well as moving its low point earlier. The fall from ages 25 to 34 to ages 45 to 54 is −0.20 (95% CI −0.22 to −0.19) ladder points in the raw profile of these interviews and −0.32 (95% CI −0.33 to −0.30) after the five adjustments. The fall from ages 18 to 24 to ages 45 to 54 is −0.28 (95% CI −0.30 to −0.26) raw and −0.71 (95% CI −0.73 to −0.69) adjusted.

The rise from ages 45 to 54 to ages 75 to 94 is +0.56 (95% CI 0.54 to 0.57) raw and +0.91 (95% CI 0.90 to 0.93) adjusted. Adjusting for marital status alone changes the fall to −0.23 (95% CI −0.25 to −0.22), the largest change among the single adjustments after household income. The raw profile already has a midlife low without any adjustment, and the adjustments deepen it, mostly through household income.

This is where the literature disagrees. Blanchflower and Oswald report a low in middle age after holding other characteristics fixed.[3] Glenn argues that part of the U shape comes from the controls, marital status above all, because happier people are more likely to marry and stay married, and a variable influenced by the outcome should not be held fixed.[6] The same objection applies to household income and to self-rated health, which are partly consequences of well-being. The adjusted profile in Table 5 answers what the profile would be among adults with the same mix of characteristics. It does not isolate age.

> Study evidence: Whether the midlife low happens within a person's own life (Individual fixed-effects panels)
>
> A cross-section cannot say whether the dip happens to individuals. Two studies follow the same people. Frijters and Beatton used three household panels, the German Socio-Economic Panel, the British Household Panel Survey and the Household, Income and Labour Dynamics in Australia survey. Once each person's own average level was removed, the middle-age U shape disappeared in all three. The largest age effect was a rise in happiness around age 60, followed by a decline after age 75, with almost no change between early adulthood and middle age.[5] Cheng, Powdthavee and Oswald used four data sets and only within-person changes, and reported support for a U shape.[7]
>
> The two studies compare people with themselves over time, which removes differences between people that do not change. Neither randomizes anything. A within-person design needs an assumption that separates age from calendar year, because the two move together for everyone, and the studies reach opposite results on the shape. Neither uses the Gallup poll, whose respondents are interviewed once.

### Birth cohorts

A cross-section mixes age with cohort, but repeated cross-sections can follow a cohort on average. Table 6 follows six birth cohorts, each covering a decade of birth years, across the interview years. People born in 1950 to 1959 were aged 50 to 59 in 2009 and 58 to 67 in 2017. Their stress changed by −7.9 (95% CI −8.5 to −7.2) points between those years, against −10.6 (95% CI −11.1 to −10.1) points between the two age groups in the 2009 cross-section.

The comparison with all adults must allow for the sample's age. The interview sample was younger in 2017 (mean age 52.3) than in 2009 (55.1), and a change in the average age of the sample moves averages that differ by age for reasons unrelated to aging. The change among all adults is therefore taken at each single year of age from 50 to 67, the ages the cohort passes through, and averaged. Stress changed by +1.7 (95% CI 1.3 to 2.2) points at those ages over the same years, and relative to that the cohort's change was −9.6 (95% CI −10.1 to −9.1).

The ladder score of the cohort changed by +0.41 (95% CI 0.39 to 0.44) ladder points, against +0.28 (95% CI 0.26 to 0.30) in the cross-section and +0.19 (95% CI 0.17 to 0.21) for adults at the same ages, and relative to them the cohort's change was +0.22 (95% CI 0.20 to 0.24).

Table 6. Relative to adults at the same ages, stress among people born in 1950 to 1959 changed by −9.6 points and their ladder score by +0.22 points between 2009 and 2017, against −10.6 and +0.28 across the same ages in 2009

For stress and for the ladder, the changes within the cohort, relative to adults at the same ages, have the sign and most of the size of the differences across ages in one year. They are somewhat smaller, −9.6 against −10.6 for stress and +0.22 against +0.28 for the ladder, with intervals that do not overlap for the ladder, so a cohort difference of that size is possible. The cross-sectional differences in these two measures are not only cohort differences.

Physical pain is different. It changed by +5.9 (95% CI 5.3 to 6.5) points within the cohort and by +5.1 (95% CI 4.6 to 5.6) points among adults at the same ages, which leaves +0.8 (95% CI 0.3 to 1.2) points, against 0.0 (95% CI −0.4 to 0.4) across the same ages in 2009. Pain rose within every cohort in Table 6. It also rose among adults at ages 50 to 67 by almost as much as in this cohort, so nearly all of this cohort's rise is a rise in the period and not aging. The cross-sectional plateau after the peak is not contradicted by the cohort lines. The check covers one cohort, and the correction using all adults is rough because all adults include the cohort itself.

### What these data cannot show

The profiles are differences between people of different ages in a repeated cross-section. They mix age, birth cohort and the years in which each group was interviewed, and the three cannot be separated with data from one decade. The oldest adults are survivors, so the oldest cells describe the people who lived to be interviewed there, and these data cannot say in which direction survival moves a profile. The panel studies in the box reach opposite conclusions about the middle-age dip, and these data cannot arbitrate between them.

Every estimate is unweighted, and the poll was a telephone survey, so the levels describe the people interviewed and not the national population. County comes from the respondent's ZIP code and is approximate near county lines. The feelings are yes or no answers about the previous day, so they measure how often a feeling was present and not how strong it was. The ladder of 2008 is left out and the 2009 ladder may carry part of the same question-order artifact, which the extract cannot separate without an interview month.

The adjusted profiles are statistical constructs. Household income, marital status and self-rated health are partly consequences of well-being, so adjusting for them does not isolate the part of the profile that belongs to age, and the profile is flat near its low point, so the age of that low is less certain than its bootstrap intervals suggest. Household income is nominal, includes a partner's earnings and was missing for a share of interviews. The mix of income brackets changes in 2015 to 2017 for reasons the extract does not document, so results by income use 2009 to 2014, and the cutoffs between groups are fixed dollar amounts across that period. The deprivation groups describe the county where a respondent lives and not the respondent, the groups hold very different numbers of counties, the index describes counties as of 2023, and the Area Deprivation Index understates deprivation in high-cost cities.

The poll ends in 2017, and later years may differ. Chapter 4.5, Mental distress in young adults (https://vishalsingh.org/zipcode/mental-distress-in-young-adults) follows frequent mental distress among young adults from 2014 in a different survey, and Chapter 6.2, Civic life, loneliness and partnership (https://vishalsingh.org/zipcode/civic-life-loneliness-and-partnership) shows how the ladder gap between partnered and other adults varies by age band in the same Gallup data. No figure in this chapter follows a person through life.

## Tables

### Table 1. Mean life evaluation falls from 7.03 among adults 18 to 24 to 6.84 among adults 45 to 54 and is 7.38 among adults 75 to 94

Unweighted mean Cantril ladder score by year of age, adults aged 18 to 94 interviewed by Gallup, 2009 to 2017

| age | conf_high | conf_low | estimate | n |
| ---: | ---: | ---: | ---: | ---: |
| 18 | 7.28 | 7.23 | 7.25 | 25,104 |
| 19 | 7.16 | 7.11 | 7.14 | 20,764 |
| 20 | 7.06 | 7.01 | 7.03 | 20,566 |
| 21 | 7.04 | 6.98 | 7.01 | 21,404 |
| 22 | 6.99 | 6.93 | 6.96 | 20,267 |
| 23 | 6.93 | 6.88 | 6.91 | 20,868 |
| 24 | 6.91 | 6.85 | 6.88 | 20,651 |
| 25 | 6.96 | 6.91 | 6.93 | 22,756 |
| 26 | 6.95 | 6.89 | 6.92 | 20,582 |
| 27 | 6.97 | 6.92 | 6.94 | 21,275 |
| 28 | 7.01 | 6.96 | 6.98 | 22,607 |
| 29 | 7.02 | 6.97 | 6.99 | 21,955 |
| 30 | 7.04 | 6.99 | 7.01 | 26,817 |
| 31 | 7.05 | 7.00 | 7.02 | 21,509 |
| 32 | 7.03 | 6.98 | 7.00 | 25,123 |
| 33 | 7.05 | 7 | 7.03 | 22,981 |
| 34 | 7.04 | 6.98 | 7.01 | 23,449 |
| 35 | 7.06 | 7.01 | 7.04 | 27,289 |
| 36 | 7.06 | 7.01 | 7.04 | 23,837 |
| 37 | 7.04 | 6.99 | 7.01 | 24,074 |
| 38 | 7.02 | 6.97 | 7.00 | 25,474 |
| 39 | 7.05 | 7.00 | 7.02 | 25,043 |
| 40 | 7.01 | 6.96 | 6.99 | 33,427 |
| 41 | 7.02 | 6.96 | 6.99 | 24,465 |
| 42 | 6.97 | 6.92 | 6.94 | 30,521 |
| 43 | 6.93 | 6.88 | 6.91 | 28,188 |
| 44 | 6.94 | 6.89 | 6.91 | 27,722 |
| 45 | 6.94 | 6.88 | 6.91 | 34,003 |
| 46 | 6.88 | 6.83 | 6.86 | 31,175 |
| 47 | 6.86 | 6.81 | 6.84 | 33,763 |
| 48 | 6.87 | 6.82 | 6.84 | 34,633 |
| 49 | 6.86 | 6.81 | 6.83 | 34,790 |
| 50 | 6.87 | 6.83 | 6.85 | 49,187 |
| 51 | 6.87 | 6.82 | 6.84 | 35,575 |
| 52 | 6.84 | 6.79 | 6.82 | 43,598 |
| 53 | 6.82 | 6.77 | 6.79 | 41,271 |
| 54 | 6.84 | 6.80 | 6.82 | 42,032 |
| 55 | 6.86 | 6.82 | 6.84 | 47,470 |
| 56 | 6.85 | 6.80 | 6.83 | 43,108 |
| 57 | 6.87 | 6.83 | 6.85 | 44,152 |
| 58 | 6.89 | 6.84 | 6.87 | 45,107 |
| 59 | 6.93 | 6.89 | 6.91 | 44,432 |
| 60 | 7.00 | 6.96 | 6.98 | 54,582 |
| 61 | 7.03 | 6.99 | 7.01 | 41,935 |
| 62 | 7.09 | 7.05 | 7.07 | 51,583 |
| 63 | 7.12 | 7.08 | 7.10 | 46,017 |
| 64 | 7.18 | 7.13 | 7.16 | 46,280 |
| 65 | 7.22 | 7.19 | 7.21 | 53,072 |
| 66 | 7.28 | 7.24 | 7.26 | 43,111 |
| 67 | 7.30 | 7.26 | 7.28 | 44,121 |
| 68 | 7.31 | 7.27 | 7.29 | 42,066 |
| 69 | 7.34 | 7.30 | 7.32 | 38,080 |
| 70 | 7.37 | 7.33 | 7.35 | 46,117 |
| 71 | 7.36 | 7.31 | 7.33 | 33,149 |
| 72 | 7.37 | 7.32 | 7.34 | 35,737 |
| 73 | 7.37 | 7.32 | 7.34 | 32,279 |
| 74 | 7.38 | 7.33 | 7.36 | 29,803 |
| 75 | 7.38 | 7.33 | 7.36 | 32,359 |
| 76 | 7.38 | 7.33 | 7.35 | 26,175 |
| 77 | 7.38 | 7.33 | 7.36 | 25,447 |
| 78 | 7.39 | 7.33 | 7.36 | 24,365 |
| 79 | 7.40 | 7.35 | 7.37 | 21,647 |
| 80 | 7.44 | 7.39 | 7.41 | 26,170 |
| 81 | 7.44 | 7.38 | 7.41 | 18,844 |
| 82 | 7.42 | 7.35 | 7.38 | 19,071 |
| 83 | 7.38 | 7.31 | 7.35 | 17,079 |
| 84 | 7.42 | 7.35 | 7.39 | 14,958 |
| 85 | 7.39 | 7.32 | 7.36 | 14,462 |
| 86 | 7.41 | 7.33 | 7.37 | 10,947 |
| 87 | 7.41 | 7.32 | 7.37 | 9,423 |
| 88 | 7.42 | 7.33 | 7.38 | 8,210 |
| 89 | 7.47 | 7.36 | 7.42 | 5,902 |
| 90 | 7.58 | 7.47 | 7.52 | 5,455 |
| 91 | 7.57 | 7.43 | 7.50 | 3,335 |
| 92 | 7.53 | 7.36 | 7.45 | 2,641 |
| 93 | 7.56 | 7.36 | 7.46 | 1,754 |
| 94 | 7.61 | 7.35 | 7.48 | 1,243 |

Note: Each age is a different group of people interviewed in the same years, so a profile mixes age with birth cohort and does not follow anyone through life.

Note: Cells at the oldest ages hold fewer interviews, so their intervals are wider; the extract has no interviews above age 94.

Note: The ladder of 2008 is not used. Political questions that came before the ladder lowered it in 2008 and in the first months of 2009, and the extract has no interview month, so early 2009 interviews may carry part of the effect. Scores averaged 6.69 in 2008 and 7.03 in 2009, and the 2009 minus 2008 difference is between 0.29 and 0.38 ladder points in every age band.

Note: The ladder was asked of 85% of interviews in 2009, of nearly all in 2010 to 2012 and of 49% from 2013 to 2017.

Note: Gallup's extract has no survey weights, so every estimate is unweighted and levels can differ from the weighted figures Gallup publishes. County comes from the respondent's ZIP code and is approximate near county lines.

Source: Gallup U.S. Daily, ladder 2009-2017; unweighted, intervals clustered by county

Data: https://vishalsingh.org/zipcode/report/well-being-across-the-life-span/figures/ladder_by_age.json

### Table 2. Stress, worry and anger are lower among older adults, enjoyment and happiness are lowest in midlife, and physical pain peaks near age 60

Share of adults who reported each feeling during a lot of the day yesterday, by year of age, and the age at which each profile turns, Gallup 2008 to 2017

#### Table 2.1. Positive feelings

Enjoyment was 88.3% among adults 18 to 24, 82.6% among adults 45 to 54 and 88.5% among adults 75 to 94

Share of adults who reported each feeling during a lot of the day yesterday, by year of age, Gallup interviews in the years each item was asked

| age | conf_high | conf_low | estimate | group | n |
| ---: | ---: | ---: | ---: | --- | ---: |
| 18 | 90.98 | 90.23 | 90.60 | Enjoyment | 28,557 |
| 19 | 89.84 | 89.06 | 89.45 | Enjoyment | 23,263 |
| 20 | 88.74 | 87.85 | 88.30 | Enjoyment | 23,009 |
| 21 | 88.61 | 87.65 | 88.13 | Enjoyment | 23,783 |
| 22 | 87.78 | 86.85 | 87.31 | Enjoyment | 22,818 |
| 23 | 87.72 | 86.80 | 87.26 | Enjoyment | 23,592 |
| 24 | 87.27 | 86.39 | 86.83 | Enjoyment | 23,437 |
| 25 | 87.10 | 86.20 | 86.65 | Enjoyment | 26,005 |
| 26 | 86.76 | 85.73 | 86.25 | Enjoyment | 23,861 |
| 27 | 86.46 | 85.54 | 86 | Enjoyment | 24,759 |
| 28 | 86.47 | 85.53 | 86 | Enjoyment | 26,621 |
| 29 | 86.20 | 85.25 | 85.72 | Enjoyment | 25,876 |
| 30 | 86.16 | 85.35 | 85.76 | Enjoyment | 31,422 |
| 31 | 86.64 | 85.74 | 86.19 | Enjoyment | 25,431 |
| 32 | 86 | 85.18 | 85.59 | Enjoyment | 29,597 |
| 33 | 86.08 | 85.15 | 85.62 | Enjoyment | 27,389 |
| 34 | 85.80 | 84.97 | 85.38 | Enjoyment | 27,933 |
| 35 | 85.49 | 84.64 | 85.06 | Enjoyment | 32,578 |
| 36 | 85.41 | 84.61 | 85.01 | Enjoyment | 28,964 |
| 37 | 85.25 | 84.41 | 84.83 | Enjoyment | 30,009 |
| 38 | 84.82 | 83.99 | 84.40 | Enjoyment | 31,592 |
| 39 | 84.86 | 84.02 | 84.44 | Enjoyment | 30,597 |
| 40 | 84.20 | 83.48 | 83.84 | Enjoyment | 40,260 |
| 41 | 84.69 | 83.81 | 84.25 | Enjoyment | 29,985 |
| 42 | 84.05 | 83.25 | 83.65 | Enjoyment | 37,089 |
| 43 | 84.06 | 83.25 | 83.66 | Enjoyment | 34,666 |
| 44 | 83.72 | 82.91 | 83.31 | Enjoyment | 34,208 |
| 45 | 83.99 | 83.17 | 83.58 | Enjoyment | 41,548 |
| 46 | 83.64 | 82.82 | 83.23 | Enjoyment | 38,143 |
| 47 | 83.27 | 82.50 | 82.88 | Enjoyment | 41,423 |
| 48 | 82.74 | 81.97 | 82.36 | Enjoyment | 42,340 |
| 49 | 83.18 | 82.34 | 82.76 | Enjoyment | 42,520 |
| 50 | 82.88 | 82.15 | 82.52 | Enjoyment | 59,198 |
| 51 | 82.88 | 82.13 | 82.50 | Enjoyment | 43,366 |
| 52 | 83.03 | 82.29 | 82.66 | Enjoyment | 52,428 |
| 53 | 82.23 | 81.44 | 81.83 | Enjoyment | 49,670 |
| 54 | 82.41 | 81.63 | 82.02 | Enjoyment | 50,885 |
| 55 | 83.22 | 82.55 | 82.88 | Enjoyment | 56,720 |
| 56 | 83.17 | 82.49 | 82.83 | Enjoyment | 51,455 |
| 57 | 83.74 | 83.08 | 83.41 | Enjoyment | 52,780 |
| 58 | 83.39 | 82.70 | 83.04 | Enjoyment | 54,014 |
| 59 | 84.32 | 83.65 | 83.98 | Enjoyment | 52,975 |
| 60 | 84.90 | 84.24 | 84.57 | Enjoyment | 64,982 |
| 61 | 85.46 | 84.73 | 85.10 | Enjoyment | 50,789 |
| 62 | 86.08 | 85.48 | 85.78 | Enjoyment | 60,227 |
| 63 | 86.48 | 85.83 | 86.16 | Enjoyment | 52,955 |
| 64 | 87.68 | 87.03 | 87.35 | Enjoyment | 53,663 |
| 65 | 88.18 | 87.58 | 87.88 | Enjoyment | 62,084 |
| 66 | 88.83 | 88.15 | 88.49 | Enjoyment | 50,067 |
| 67 | 88.96 | 88.36 | 88.66 | Enjoyment | 50,877 |
| 68 | 89.36 | 88.72 | 89.04 | Enjoyment | 48,587 |
| 69 | 89.83 | 89.24 | 89.54 | Enjoyment | 43,823 |
| 70 | 89.93 | 89.33 | 89.63 | Enjoyment | 53,175 |
| 71 | 89.91 | 89.26 | 89.58 | Enjoyment | 38,510 |
| 72 | 89.54 | 88.85 | 89.20 | Enjoyment | 41,607 |
| 73 | 90 | 89.33 | 89.67 | Enjoyment | 37,700 |
| 74 | 89.92 | 89.20 | 89.56 | Enjoyment | 34,644 |
| 75 | 89.95 | 89.29 | 89.62 | Enjoyment | 37,805 |
| 76 | 89.79 | 89.09 | 89.44 | Enjoyment | 30,749 |
| 77 | 89.65 | 88.93 | 89.29 | Enjoyment | 30,079 |
| 78 | 89.09 | 88.34 | 88.72 | Enjoyment | 28,996 |
| 79 | 89.22 | 88.41 | 88.81 | Enjoyment | 25,522 |
| 80 | 89.21 | 88.37 | 88.79 | Enjoyment | 31,021 |
| 81 | 89.12 | 88.21 | 88.67 | Enjoyment | 22,351 |
| 82 | 89.09 | 88.21 | 88.65 | Enjoyment | 22,581 |
| 83 | 88.28 | 87.32 | 87.80 | Enjoyment | 20,144 |
| 84 | 88.18 | 87.17 | 87.68 | Enjoyment | 17,710 |
| 85 | 88.27 | 87.30 | 87.78 | Enjoyment | 17,007 |
| 86 | 87.63 | 86.47 | 87.05 | Enjoyment | 12,976 |
| 87 | 87.93 | 86.65 | 87.29 | Enjoyment | 10,983 |
| 88 | 87.16 | 85.73 | 86.44 | Enjoyment | 9,723 |
| 89 | 87.87 | 86.15 | 87.01 | Enjoyment | 6,788 |
| 90 | 87.71 | 86.09 | 86.90 | Enjoyment | 6,359 |
| 91 | 88.42 | 86.24 | 87.33 | Enjoyment | 3,876 |
| 92 | 88.17 | 85.94 | 87.06 | Enjoyment | 3,083 |
| 93 | 88.01 | 85 | 86.51 | Enjoyment | 2,031 |
| 94 | 88.14 | 84.60 | 86.37 | Enjoyment | 1,438 |
| 18 | 93.51 | 92.83 | 93.17 | Happiness | 26,328 |
| 19 | 92.84 | 92.07 | 92.46 | Happiness | 21,421 |
| 20 | 92.27 | 91.45 | 91.86 | Happiness | 21,096 |
| 21 | 92 | 91.20 | 91.60 | Happiness | 21,766 |
| 22 | 91.75 | 90.97 | 91.36 | Happiness | 20,896 |
| 23 | 91.83 | 90.92 | 91.37 | Happiness | 21,564 |
| 24 | 91.48 | 90.70 | 91.09 | Happiness | 21,419 |
| 25 | 91.39 | 90.45 | 90.92 | Happiness | 23,779 |
| 26 | 91.05 | 90.19 | 90.62 | Happiness | 21,871 |
| 27 | 90.70 | 89.87 | 90.28 | Happiness | 22,747 |
| 28 | 90.75 | 89.98 | 90.36 | Happiness | 24,555 |
| 29 | 90.87 | 90.08 | 90.48 | Happiness | 23,865 |
| 30 | 90.27 | 89.55 | 89.91 | Happiness | 29,131 |
| 31 | 91.28 | 90.49 | 90.88 | Happiness | 23,608 |
| 32 | 90.63 | 89.88 | 90.26 | Happiness | 27,463 |
| 33 | 90.96 | 90.18 | 90.57 | Happiness | 25,404 |
| 34 | 90.82 | 90.01 | 90.41 | Happiness | 25,881 |
| 35 | 90.07 | 89.31 | 89.69 | Happiness | 30,143 |
| 36 | 90.37 | 89.61 | 89.99 | Happiness | 26,869 |
| 37 | 89.98 | 89.19 | 89.58 | Happiness | 28,044 |
| 38 | 90.08 | 89.26 | 89.67 | Happiness | 29,670 |
| 39 | 89.62 | 88.79 | 89.20 | Happiness | 28,734 |
| 40 | 88.85 | 88.15 | 88.50 | Happiness | 37,679 |
| 41 | 89.24 | 88.43 | 88.83 | Happiness | 28,383 |
| 42 | 88.70 | 87.95 | 88.33 | Happiness | 34,991 |
| 43 | 88.70 | 87.93 | 88.32 | Happiness | 32,863 |
| 44 | 88.33 | 87.58 | 87.96 | Happiness | 32,405 |
| 45 | 88.31 | 87.62 | 87.97 | Happiness | 39,173 |
| 46 | 87.93 | 87.19 | 87.56 | Happiness | 36,081 |
| 47 | 87.67 | 86.93 | 87.30 | Happiness | 39,096 |
| 48 | 87.12 | 86.37 | 86.75 | Happiness | 40,072 |
| 49 | 87.11 | 86.35 | 86.73 | Happiness | 40,405 |
| 50 | 86.54 | 85.85 | 86.19 | Happiness | 56,025 |
| 51 | 86.69 | 86.03 | 86.36 | Happiness | 41,286 |
| 52 | 86.59 | 85.92 | 86.25 | Happiness | 49,682 |
| 53 | 86.02 | 85.30 | 85.66 | Happiness | 47,090 |
| 54 | 85.96 | 85.25 | 85.60 | Happiness | 48,184 |
| 55 | 86.12 | 85.47 | 85.80 | Happiness | 53,676 |
| 56 | 86.16 | 85.52 | 85.84 | Happiness | 48,612 |
| 57 | 86.63 | 85.98 | 86.31 | Happiness | 49,896 |
| 58 | 86.29 | 85.63 | 85.96 | Happiness | 51,055 |
| 59 | 86.69 | 86.07 | 86.38 | Happiness | 50,127 |
| 60 | 87.18 | 86.55 | 86.87 | Happiness | 61,244 |
| 61 | 87.56 | 86.91 | 87.23 | Happiness | 48,038 |
| 62 | 87.87 | 87.27 | 87.57 | Happiness | 56,916 |
| 63 | 88.18 | 87.55 | 87.87 | Happiness | 49,996 |
| 64 | 89.10 | 88.50 | 88.80 | Happiness | 50,602 |
| 65 | 89.24 | 88.63 | 88.94 | Happiness | 58,470 |
| 66 | 90 | 89.36 | 89.68 | Happiness | 47,133 |
| 67 | 90.17 | 89.52 | 89.84 | Happiness | 47,727 |
| 68 | 90.42 | 89.79 | 90.11 | Happiness | 45,528 |
| 69 | 90.84 | 90.23 | 90.54 | Happiness | 40,978 |
| 70 | 90.79 | 90.22 | 90.51 | Happiness | 49,145 |
| 71 | 90.95 | 90.33 | 90.64 | Happiness | 36,212 |
| 72 | 90.68 | 90.04 | 90.36 | Happiness | 39,095 |
| 73 | 91.06 | 90.38 | 90.72 | Happiness | 35,328 |
| 74 | 90.95 | 90.29 | 90.62 | Happiness | 32,368 |
| 75 | 90.94 | 90.28 | 90.61 | Happiness | 35,569 |
| 76 | 90.93 | 90.18 | 90.56 | Happiness | 28,994 |
| 77 | 90.89 | 90.14 | 90.51 | Happiness | 28,398 |
| 78 | 90.46 | 89.69 | 90.08 | Happiness | 27,483 |
| 79 | 90.77 | 90 | 90.39 | Happiness | 24,133 |
| 80 | 90.47 | 89.65 | 90.06 | Happiness | 29,435 |
| 81 | 90.84 | 89.96 | 90.40 | Happiness | 21,245 |
| 82 | 90.75 | 89.87 | 90.31 | Happiness | 21,502 |
| 83 | 89.83 | 88.88 | 89.36 | Happiness | 19,117 |
| 84 | 90.12 | 89.08 | 89.60 | Happiness | 16,855 |
| 85 | 90.05 | 89.04 | 89.55 | Happiness | 16,147 |
| 86 | 89.85 | 88.76 | 89.31 | Happiness | 12,324 |
| 87 | 89.90 | 88.69 | 89.30 | Happiness | 10,388 |
| 88 | 89.58 | 88.27 | 88.92 | Happiness | 9,227 |
| 89 | 90.09 | 88.58 | 89.34 | Happiness | 6,433 |
| 90 | 90.27 | 88.64 | 89.45 | Happiness | 5,954 |
| 91 | 90.93 | 88.86 | 89.90 | Happiness | 3,643 |
| 92 | 90.06 | 87.92 | 88.99 | Happiness | 2,869 |
| 93 | 88.37 | 85.43 | 86.90 | Happiness | 1,878 |
| 94 | 91.68 | 88.41 | 90.05 | Happiness | 1,326 |

#### Table 2.2. Negative feelings

Stress fell from 48.6% among adults 25 to 34 to 18.0% among adults 75 to 94, while physical pain rose from 15.5% among adults 18 to 24 to 28.5% among adults 55 to 64

Share of adults who reported each feeling during a lot of the day yesterday, by year of age, Gallup interviews in the years each item was asked

| age | conf_high | conf_low | estimate | group | n |
| ---: | ---: | ---: | ---: | --- | ---: |
| 18 | 27.60 | 26.52 | 27.06 | Worry | 28,569 |
| 19 | 29.74 | 28.58 | 29.16 | Worry | 23,263 |
| 20 | 30.78 | 29.53 | 30.16 | Worry | 23,026 |
| 21 | 31.96 | 30.71 | 31.33 | Worry | 23,793 |
| 22 | 33.12 | 31.73 | 32.42 | Worry | 22,822 |
| 23 | 33 | 31.79 | 32.40 | Worry | 23,611 |
| 24 | 33.28 | 32.06 | 32.67 | Worry | 23,454 |
| 25 | 33.11 | 31.91 | 32.51 | Worry | 26,027 |
| 26 | 33.62 | 32.28 | 32.95 | Worry | 23,871 |
| 27 | 33.45 | 32.18 | 32.82 | Worry | 24,772 |
| 28 | 33.50 | 32.17 | 32.84 | Worry | 26,629 |
| 29 | 33.33 | 32.18 | 32.75 | Worry | 25,885 |
| 30 | 33.30 | 32.11 | 32.71 | Worry | 31,469 |
| 31 | 33.94 | 32.64 | 33.29 | Worry | 25,449 |
| 32 | 33.92 | 32.75 | 33.33 | Worry | 29,620 |
| 33 | 34.01 | 32.75 | 33.38 | Worry | 27,399 |
| 34 | 33.96 | 32.68 | 33.32 | Worry | 27,969 |
| 35 | 33.74 | 32.53 | 33.14 | Worry | 32,611 |
| 36 | 33.98 | 32.74 | 33.36 | Worry | 28,991 |
| 37 | 34.97 | 33.80 | 34.39 | Worry | 30,040 |
| 38 | 34.78 | 33.64 | 34.21 | Worry | 31,629 |
| 39 | 34.98 | 33.69 | 34.33 | Worry | 30,627 |
| 40 | 34.67 | 33.67 | 34.17 | Worry | 40,304 |
| 41 | 34.99 | 33.78 | 34.38 | Worry | 29,998 |
| 42 | 35.32 | 34.18 | 34.75 | Worry | 37,118 |
| 43 | 35.46 | 34.39 | 34.93 | Worry | 34,694 |
| 44 | 35.63 | 34.50 | 35.07 | Worry | 34,227 |
| 45 | 35.10 | 34.03 | 34.57 | Worry | 41,579 |
| 46 | 36.38 | 35.17 | 35.77 | Worry | 38,186 |
| 47 | 36.45 | 35.47 | 35.96 | Worry | 41,477 |
| 48 | 36.69 | 35.63 | 36.16 | Worry | 42,392 |
| 49 | 36.12 | 35.05 | 35.59 | Worry | 42,567 |
| 50 | 35.37 | 34.48 | 34.92 | Worry | 59,303 |
| 51 | 36.14 | 35.07 | 35.61 | Worry | 43,404 |
| 52 | 35.77 | 34.87 | 35.32 | Worry | 52,497 |
| 53 | 36.32 | 35.38 | 35.85 | Worry | 49,716 |
| 54 | 35.42 | 34.49 | 34.95 | Worry | 50,951 |
| 55 | 34.27 | 33.41 | 33.84 | Worry | 56,795 |
| 56 | 34.39 | 33.45 | 33.92 | Worry | 51,529 |
| 57 | 34.12 | 33.23 | 33.67 | Worry | 52,847 |
| 58 | 33.57 | 32.70 | 33.13 | Worry | 54,077 |
| 59 | 32.66 | 31.70 | 32.18 | Worry | 53,045 |
| 60 | 30.61 | 29.78 | 30.20 | Worry | 65,048 |
| 61 | 30.69 | 29.70 | 30.19 | Worry | 50,838 |
| 62 | 28.83 | 27.93 | 28.38 | Worry | 60,289 |
| 63 | 28.34 | 27.43 | 27.89 | Worry | 53,014 |
| 64 | 26.56 | 25.67 | 26.11 | Worry | 53,740 |
| 65 | 24.70 | 23.83 | 24.27 | Worry | 62,206 |
| 66 | 23.70 | 22.87 | 23.28 | Worry | 50,147 |
| 67 | 23.09 | 22.16 | 22.63 | Worry | 50,938 |
| 68 | 22.23 | 21.34 | 21.78 | Worry | 48,671 |
| 69 | 21.73 | 20.86 | 21.29 | Worry | 43,869 |
| 70 | 20.90 | 20.14 | 20.52 | Worry | 53,277 |
| 71 | 20.84 | 19.97 | 20.41 | Worry | 38,563 |
| 72 | 20.59 | 19.76 | 20.18 | Worry | 41,700 |
| 73 | 20.09 | 19.21 | 19.65 | Worry | 37,771 |
| 74 | 20.33 | 19.46 | 19.90 | Worry | 34,722 |
| 75 | 19.52 | 18.73 | 19.13 | Worry | 37,892 |
| 76 | 20.19 | 19.23 | 19.71 | Worry | 30,818 |
| 77 | 19.65 | 18.73 | 19.19 | Worry | 30,150 |
| 78 | 19.58 | 18.61 | 19.10 | Worry | 29,088 |
| 79 | 19.75 | 18.79 | 19.27 | Worry | 25,569 |
| 80 | 18.68 | 17.74 | 18.21 | Worry | 31,131 |
| 81 | 19.24 | 18.15 | 18.69 | Worry | 22,423 |
| 82 | 19.27 | 18.21 | 18.74 | Worry | 22,659 |
| 83 | 19.19 | 18.07 | 18.63 | Worry | 20,215 |
| 84 | 19.25 | 18.03 | 18.64 | Worry | 17,779 |
| 85 | 18.18 | 16.98 | 17.58 | Worry | 17,067 |
| 86 | 18.40 | 17.04 | 17.72 | Worry | 13,031 |
| 87 | 18.38 | 16.92 | 17.65 | Worry | 11,048 |
| 88 | 18.19 | 16.62 | 17.40 | Worry | 9,774 |
| 89 | 18.66 | 16.74 | 17.70 | Worry | 6,819 |
| 90 | 17.21 | 15.30 | 16.26 | Worry | 6,398 |
| 91 | 17.56 | 15.27 | 16.41 | Worry | 3,906 |
| 92 | 17.32 | 14.63 | 15.97 | Worry | 3,099 |
| 93 | 16.11 | 13.06 | 14.59 | Worry | 2,050 |
| 94 | 16.80 | 13.15 | 14.98 | Worry | 1,449 |
| 18 | 42.52 | 41.08 | 41.80 | Stress | 28,568 |
| 19 | 45.50 | 44.04 | 44.77 | Stress | 23,262 |
| 20 | 47.90 | 46.48 | 47.19 | Stress | 23,022 |
| 21 | 48.51 | 47.10 | 47.81 | Stress | 23,787 |
| 22 | 49.34 | 47.93 | 48.64 | Stress | 22,821 |
| 23 | 48.63 | 47.30 | 47.96 | Stress | 23,605 |
| 24 | 49.67 | 48.35 | 49.01 | Stress | 23,454 |
| 25 | 49.11 | 47.73 | 48.42 | Stress | 26,036 |
| 26 | 49.86 | 48.41 | 49.13 | Stress | 23,861 |
| 27 | 50.14 | 48.81 | 49.48 | Stress | 24,771 |
| 28 | 49.04 | 47.75 | 48.39 | Stress | 26,630 |
| 29 | 49.05 | 47.82 | 48.43 | Stress | 25,891 |
| 30 | 47.37 | 46.10 | 46.73 | Stress | 31,457 |
| 31 | 50.22 | 48.90 | 49.56 | Stress | 25,455 |
| 32 | 49.47 | 48.16 | 48.82 | Stress | 29,613 |
| 33 | 49.37 | 48.13 | 48.75 | Stress | 27,392 |
| 34 | 49.04 | 47.78 | 48.41 | Stress | 27,960 |
| 35 | 48.01 | 46.84 | 47.43 | Stress | 32,600 |
| 36 | 48.25 | 47.02 | 47.64 | Stress | 28,987 |
| 37 | 48.95 | 47.68 | 48.31 | Stress | 30,020 |
| 38 | 48.55 | 47.32 | 47.94 | Stress | 31,620 |
| 39 | 48.36 | 47.13 | 47.75 | Stress | 30,615 |
| 40 | 46.75 | 45.61 | 46.18 | Stress | 40,304 |
| 41 | 48.57 | 47.28 | 47.93 | Stress | 30,003 |
| 42 | 47.35 | 46.27 | 46.81 | Stress | 37,111 |
| 43 | 47.65 | 46.56 | 47.11 | Stress | 34,684 |
| 44 | 47.31 | 46.18 | 46.75 | Stress | 34,226 |
| 45 | 45.60 | 44.55 | 45.08 | Stress | 41,584 |
| 46 | 46.64 | 45.64 | 46.14 | Stress | 38,180 |
| 47 | 46.34 | 45.34 | 45.84 | Stress | 41,478 |
| 48 | 46.07 | 45.04 | 45.56 | Stress | 42,378 |
| 49 | 45.86 | 44.88 | 45.37 | Stress | 42,564 |
| 50 | 43.78 | 42.90 | 43.34 | Stress | 59,277 |
| 51 | 45.16 | 44.17 | 44.66 | Stress | 43,409 |
| 52 | 44.27 | 43.36 | 43.82 | Stress | 52,483 |
| 53 | 44.61 | 43.68 | 44.14 | Stress | 49,715 |
| 54 | 43.44 | 42.55 | 42.99 | Stress | 50,958 |
| 55 | 42 | 41.18 | 41.59 | Stress | 56,778 |
| 56 | 41.92 | 41.02 | 41.47 | Stress | 51,513 |
| 57 | 41.04 | 40.17 | 40.61 | Stress | 52,848 |
| 58 | 40.21 | 39.31 | 39.76 | Stress | 54,058 |
| 59 | 38.88 | 38.02 | 38.45 | Stress | 53,028 |
| 60 | 36.30 | 35.47 | 35.88 | Stress | 65,039 |
| 61 | 35.63 | 34.67 | 35.15 | Stress | 50,826 |
| 62 | 33.38 | 32.48 | 32.93 | Stress | 60,276 |
| 63 | 32.23 | 31.36 | 31.80 | Stress | 52,997 |
| 64 | 30.20 | 29.39 | 29.79 | Stress | 53,732 |
| 65 | 27.41 | 26.56 | 26.98 | Stress | 62,199 |
| 66 | 26.22 | 25.37 | 25.79 | Stress | 50,147 |
| 67 | 25.36 | 24.44 | 24.90 | Stress | 50,933 |
| 68 | 23.94 | 23.06 | 23.50 | Stress | 48,661 |
| 69 | 23.29 | 22.39 | 22.84 | Stress | 43,876 |
| 70 | 22 | 21.30 | 21.65 | Stress | 53,287 |
| 71 | 21.72 | 20.84 | 21.28 | Stress | 38,553 |
| 72 | 21.21 | 20.41 | 20.81 | Stress | 41,693 |
| 73 | 20.79 | 19.93 | 20.36 | Stress | 37,746 |
| 74 | 20.65 | 19.79 | 20.22 | Stress | 34,729 |
| 75 | 19.90 | 19.06 | 19.48 | Stress | 37,917 |
| 76 | 19.96 | 19.02 | 19.49 | Stress | 30,825 |
| 77 | 20.22 | 19.24 | 19.73 | Stress | 30,161 |
| 78 | 19.32 | 18.39 | 18.86 | Stress | 29,100 |
| 79 | 19.14 | 18.16 | 18.65 | Stress | 25,570 |
| 80 | 18.45 | 17.53 | 17.99 | Stress | 31,115 |
| 81 | 18.65 | 17.62 | 18.14 | Stress | 22,419 |
| 82 | 18 | 17.01 | 17.50 | Stress | 22,663 |
| 83 | 18.22 | 17.13 | 17.68 | Stress | 20,214 |
| 84 | 17.62 | 16.55 | 17.08 | Stress | 17,771 |
| 85 | 17.44 | 16.34 | 16.89 | Stress | 17,078 |
| 86 | 17.24 | 15.88 | 16.56 | Stress | 13,037 |
| 87 | 16.66 | 15.29 | 15.97 | Stress | 11,068 |
| 88 | 16.03 | 14.62 | 15.32 | Stress | 9,776 |
| 89 | 16.60 | 14.80 | 15.70 | Stress | 6,814 |
| 90 | 15.78 | 14.09 | 14.93 | Stress | 6,396 |
| 91 | 14.95 | 12.67 | 13.81 | Stress | 3,896 |
| 92 | 15.09 | 12.60 | 13.84 | Stress | 3,106 |
| 93 | 15.41 | 12.20 | 13.80 | Stress | 2,050 |
| 94 | 13.64 | 10.18 | 11.91 | Stress | 1,444 |
| 18 | 15.36 | 14.45 | 14.91 | Sadness | 26,311 |
| 19 | 16.06 | 15.07 | 15.57 | Sadness | 21,413 |
| 20 | 16.31 | 15.27 | 15.79 | Sadness | 21,082 |
| 21 | 15.79 | 14.84 | 15.32 | Sadness | 21,759 |
| 22 | 16.05 | 15 | 15.52 | Sadness | 20,899 |
| 23 | 15.40 | 14.35 | 14.87 | Sadness | 21,557 |
| 24 | 15.85 | 14.86 | 15.35 | Sadness | 21,409 |
| 25 | 16.01 | 14.88 | 15.45 | Sadness | 23,760 |
| 26 | 15.70 | 14.58 | 15.14 | Sadness | 21,863 |
| 27 | 15.52 | 14.48 | 15 | Sadness | 22,750 |
| 28 | 15.18 | 14.05 | 14.61 | Sadness | 24,537 |
| 29 | 15.32 | 14.17 | 14.74 | Sadness | 23,867 |
| 30 | 15.24 | 14.33 | 14.79 | Sadness | 29,131 |
| 31 | 15.36 | 14.32 | 14.84 | Sadness | 23,599 |
| 32 | 15.45 | 14.32 | 14.88 | Sadness | 27,453 |
| 33 | 15.17 | 14.21 | 14.69 | Sadness | 25,403 |
| 34 | 15.44 | 14.38 | 14.91 | Sadness | 25,880 |
| 35 | 15.56 | 14.65 | 15.11 | Sadness | 30,147 |
| 36 | 15.14 | 14.13 | 14.63 | Sadness | 26,874 |
| 37 | 16.03 | 15.02 | 15.53 | Sadness | 28,038 |
| 38 | 15.80 | 14.72 | 15.26 | Sadness | 29,685 |
| 39 | 15.98 | 14.97 | 15.48 | Sadness | 28,746 |
| 40 | 16.48 | 15.40 | 15.94 | Sadness | 37,697 |
| 41 | 16.03 | 15.07 | 15.55 | Sadness | 28,404 |
| 42 | 16.73 | 15.72 | 16.23 | Sadness | 35,013 |
| 43 | 17.32 | 16.38 | 16.85 | Sadness | 32,880 |
| 44 | 17.28 | 16.33 | 16.81 | Sadness | 32,434 |
| 45 | 17.27 | 16.35 | 16.81 | Sadness | 39,202 |
| 46 | 17.82 | 16.89 | 17.35 | Sadness | 36,089 |
| 47 | 18.34 | 17.40 | 17.87 | Sadness | 39,128 |
| 48 | 18.63 | 17.76 | 18.20 | Sadness | 40,112 |
| 49 | 18.81 | 17.85 | 18.33 | Sadness | 40,449 |
| 50 | 19.03 | 18.24 | 18.63 | Sadness | 56,078 |
| 51 | 19.42 | 18.51 | 18.96 | Sadness | 41,315 |
| 52 | 19.48 | 18.65 | 19.07 | Sadness | 49,742 |
| 53 | 20.40 | 19.56 | 19.98 | Sadness | 47,128 |
| 54 | 20.18 | 19.36 | 19.77 | Sadness | 48,240 |
| 55 | 19.65 | 18.89 | 19.27 | Sadness | 53,727 |
| 56 | 19.60 | 18.80 | 19.20 | Sadness | 48,665 |
| 57 | 20.24 | 19.48 | 19.86 | Sadness | 49,991 |
| 58 | 19.76 | 18.94 | 19.35 | Sadness | 51,098 |
| 59 | 19.31 | 18.51 | 18.91 | Sadness | 50,180 |
| 60 | 18.49 | 17.75 | 18.12 | Sadness | 61,303 |
| 61 | 18.79 | 18.01 | 18.40 | Sadness | 48,109 |
| 62 | 18.01 | 17.29 | 17.65 | Sadness | 56,973 |
| 63 | 17.88 | 17.14 | 17.51 | Sadness | 50,073 |
| 64 | 16.82 | 16.09 | 16.45 | Sadness | 50,682 |
| 65 | 16.16 | 15.41 | 15.79 | Sadness | 58,576 |
| 66 | 15.61 | 14.90 | 15.26 | Sadness | 47,212 |
| 67 | 15.78 | 15.05 | 15.41 | Sadness | 47,791 |
| 68 | 15.11 | 14.37 | 14.74 | Sadness | 45,596 |
| 69 | 14.81 | 14.09 | 14.45 | Sadness | 41,075 |
| 70 | 14.53 | 13.82 | 14.17 | Sadness | 49,252 |
| 71 | 15.08 | 14.35 | 14.72 | Sadness | 36,279 |
| 72 | 15.45 | 14.66 | 15.05 | Sadness | 39,182 |
| 73 | 14.96 | 14.17 | 14.56 | Sadness | 35,411 |
| 74 | 15.63 | 14.78 | 15.21 | Sadness | 32,439 |
| 75 | 15.24 | 14.47 | 14.85 | Sadness | 35,644 |
| 76 | 15.44 | 14.57 | 15.01 | Sadness | 29,070 |
| 77 | 15.81 | 14.94 | 15.38 | Sadness | 28,463 |
| 78 | 16.33 | 15.42 | 15.87 | Sadness | 27,566 |
| 79 | 16.86 | 15.89 | 16.38 | Sadness | 24,200 |
| 80 | 16.22 | 15.36 | 15.79 | Sadness | 29,500 |
| 81 | 16.86 | 15.85 | 16.36 | Sadness | 21,308 |
| 82 | 16.80 | 15.77 | 16.29 | Sadness | 21,557 |
| 83 | 17.31 | 16.14 | 16.73 | Sadness | 19,173 |
| 84 | 17.45 | 16.22 | 16.83 | Sadness | 16,941 |
| 85 | 17.39 | 16.26 | 16.82 | Sadness | 16,206 |
| 86 | 18.02 | 16.63 | 17.33 | Sadness | 12,375 |
| 87 | 18.75 | 17.25 | 18 | Sadness | 10,439 |
| 88 | 17.43 | 15.95 | 16.69 | Sadness | 9,274 |
| 89 | 17.85 | 15.99 | 16.92 | Sadness | 6,459 |
| 90 | 17.86 | 15.97 | 16.91 | Sadness | 5,984 |
| 91 | 16.87 | 14.46 | 15.67 | Sadness | 3,657 |
| 92 | 17.48 | 14.65 | 16.06 | Sadness | 2,889 |
| 93 | 19.09 | 15.64 | 17.36 | Sadness | 1,889 |
| 94 | 16.40 | 12.58 | 14.49 | Sadness | 1,325 |
| 18 | 17.90 | 16.84 | 17.37 | Anger | 21,725 |
| 19 | 17.82 | 16.64 | 17.23 | Anger | 17,505 |
| 20 | 17.39 | 16.22 | 16.80 | Anger | 17,176 |
| 21 | 16.62 | 15.50 | 16.06 | Anger | 17,653 |
| 22 | 16.70 | 15.53 | 16.11 | Anger | 16,893 |
| 23 | 16.50 | 15.34 | 15.92 | Anger | 17,519 |
| 24 | 16.70 | 15.54 | 16.12 | Anger | 17,415 |
| 25 | 17 | 15.89 | 16.45 | Anger | 19,452 |
| 26 | 16.54 | 15.42 | 15.98 | Anger | 18,145 |
| 27 | 16.64 | 15.48 | 16.06 | Anger | 18,881 |
| 28 | 16.39 | 15.38 | 15.89 | Anger | 20,638 |
| 29 | 16.45 | 15.39 | 15.92 | Anger | 20,017 |
| 30 | 16.47 | 15.54 | 16.01 | Anger | 24,372 |
| 31 | 16.28 | 15.20 | 15.74 | Anger | 19,885 |
| 32 | 16.53 | 15.52 | 16.02 | Anger | 23,096 |
| 33 | 16.61 | 15.62 | 16.11 | Anger | 21,430 |
| 34 | 16.34 | 15.31 | 15.82 | Anger | 21,732 |
| 35 | 16.63 | 15.68 | 16.15 | Anger | 25,373 |
| 36 | 15.95 | 14.96 | 15.45 | Anger | 22,886 |
| 37 | 16.67 | 15.70 | 16.18 | Anger | 24,139 |
| 38 | 16.81 | 15.90 | 16.35 | Anger | 25,720 |
| 39 | 16.07 | 15.10 | 15.59 | Anger | 24,912 |
| 40 | 16.14 | 15.31 | 15.73 | Anger | 32,752 |
| 41 | 16.29 | 15.36 | 15.82 | Anger | 24,943 |
| 42 | 16.10 | 15.22 | 15.66 | Anger | 30,455 |
| 43 | 16.07 | 15.11 | 15.59 | Anger | 28,440 |
| 44 | 15.89 | 15.02 | 15.45 | Anger | 27,888 |
| 45 | 15.47 | 14.68 | 15.08 | Anger | 33,803 |
| 46 | 15.60 | 14.75 | 15.17 | Anger | 31,452 |
| 47 | 15.23 | 14.43 | 14.83 | Anger | 34,243 |
| 48 | 15.50 | 14.67 | 15.09 | Anger | 35,141 |
| 49 | 14.97 | 14.13 | 14.55 | Anger | 35,309 |
| 50 | 14.18 | 13.50 | 13.84 | Anger | 48,561 |
| 51 | 14.44 | 13.65 | 14.05 | Anger | 35,897 |
| 52 | 13.91 | 13.16 | 13.54 | Anger | 42,908 |
| 53 | 14.25 | 13.55 | 13.90 | Anger | 40,709 |
| 54 | 13.60 | 12.90 | 13.25 | Anger | 41,424 |
| 55 | 13.37 | 12.70 | 13.03 | Anger | 46,109 |
| 56 | 13.12 | 12.41 | 12.77 | Anger | 41,967 |
| 57 | 12.84 | 12.12 | 12.48 | Anger | 42,926 |
| 58 | 12.45 | 11.81 | 12.13 | Anger | 43,889 |
| 59 | 12.04 | 11.35 | 11.70 | Anger | 43,045 |
| 60 | 11.80 | 11.18 | 11.49 | Anger | 52,398 |
| 61 | 11.37 | 10.74 | 11.06 | Anger | 41,472 |
| 62 | 11.04 | 10.45 | 10.74 | Anger | 48,829 |
| 63 | 10.79 | 10.16 | 10.47 | Anger | 42,777 |
| 64 | 10.34 | 9.72 | 10.03 | Anger | 43,060 |
| 65 | 9.81 | 9.20 | 9.51 | Anger | 49,538 |
| 66 | 9.44 | 8.84 | 9.14 | Anger | 39,393 |
| 67 | 9.03 | 8.37 | 8.70 | Anger | 39,366 |
| 68 | 9.04 | 8.43 | 8.73 | Anger | 37,753 |
| 69 | 8.56 | 7.96 | 8.26 | Anger | 34,814 |
| 70 | 8.37 | 7.79 | 8.08 | Anger | 41,384 |
| 71 | 8.33 | 7.73 | 8.03 | Anger | 30,318 |
| 72 | 8.02 | 7.42 | 7.72 | Anger | 32,575 |
| 73 | 7.92 | 7.29 | 7.61 | Anger | 29,791 |
| 74 | 7.54 | 6.89 | 7.22 | Anger | 27,289 |
| 75 | 7.41 | 6.80 | 7.11 | Anger | 30,195 |
| 76 | 7.39 | 6.77 | 7.08 | Anger | 24,672 |
| 77 | 7.20 | 6.56 | 6.88 | Anger | 24,157 |
| 78 | 7.06 | 6.41 | 6.73 | Anger | 23,509 |
| 79 | 6.71 | 5.97 | 6.34 | Anger | 20,765 |
| 80 | 6.52 | 5.90 | 6.21 | Anger | 25,434 |
| 81 | 6.63 | 5.88 | 6.25 | Anger | 18,439 |
| 82 | 6 | 5.29 | 5.64 | Anger | 18,622 |
| 83 | 6.21 | 5.46 | 5.83 | Anger | 16,473 |
| 84 | 5.81 | 4.98 | 5.39 | Anger | 14,523 |
| 85 | 6.28 | 5.44 | 5.86 | Anger | 13,922 |
| 86 | 5.66 | 4.82 | 5.24 | Anger | 10,594 |
| 87 | 5.77 | 4.82 | 5.29 | Anger | 8,862 |
| 88 | 5.07 | 4.20 | 4.64 | Anger | 7,914 |
| 89 | 5.57 | 4.36 | 4.96 | Anger | 5,481 |
| 90 | 5.50 | 4.26 | 4.88 | Anger | 5,041 |
| 91 | 4.84 | 3.49 | 4.16 | Anger | 3,074 |
| 92 | 5.37 | 3.72 | 4.55 | Anger | 2,420 |
| 93 | 5.46 | 3.44 | 4.45 | Anger | 1,574 |
| 94 | 5.51 | 3.23 | 4.37 | Anger | 1,099 |
| 18 | 14.98 | 14.21 | 14.59 | Physical pain | 28,574 |
| 19 | 15.53 | 14.63 | 15.08 | Physical pain | 23,261 |
| 20 | 16.19 | 15.21 | 15.70 | Physical pain | 23,027 |
| 21 | 16.13 | 15.12 | 15.63 | Physical pain | 23,787 |
| 22 | 16.41 | 15.39 | 15.90 | Physical pain | 22,818 |
| 23 | 16.14 | 15.11 | 15.62 | Physical pain | 23,611 |
| 24 | 16.64 | 15.58 | 16.11 | Physical pain | 23,454 |
| 25 | 16.70 | 15.67 | 16.19 | Physical pain | 26,046 |
| 26 | 16.59 | 15.54 | 16.06 | Physical pain | 23,861 |
| 27 | 17.39 | 16.33 | 16.86 | Physical pain | 24,770 |
| 28 | 17.06 | 16.09 | 16.57 | Physical pain | 26,621 |
| 29 | 17.21 | 16.22 | 16.71 | Physical pain | 25,890 |
| 30 | 17.31 | 16.36 | 16.83 | Physical pain | 31,454 |
| 31 | 17.80 | 16.70 | 17.25 | Physical pain | 25,454 |
| 32 | 18.09 | 17.13 | 17.61 | Physical pain | 29,622 |
| 33 | 18.26 | 17.22 | 17.74 | Physical pain | 27,400 |
| 34 | 18.09 | 17.06 | 17.58 | Physical pain | 27,960 |
| 35 | 18.25 | 17.28 | 17.76 | Physical pain | 32,611 |
| 36 | 18.91 | 17.90 | 18.41 | Physical pain | 28,995 |
| 37 | 19.27 | 18.28 | 18.77 | Physical pain | 30,039 |
| 38 | 19.02 | 18.07 | 18.55 | Physical pain | 31,640 |
| 39 | 19.61 | 18.53 | 19.07 | Physical pain | 30,629 |
| 40 | 19.67 | 18.74 | 19.20 | Physical pain | 40,310 |
| 41 | 20.37 | 19.32 | 19.85 | Physical pain | 30,012 |
| 42 | 20.76 | 19.73 | 20.25 | Physical pain | 37,134 |
| 43 | 21.82 | 20.81 | 21.31 | Physical pain | 34,706 |
| 44 | 22.33 | 21.17 | 21.75 | Physical pain | 34,244 |
| 45 | 22.23 | 21.19 | 21.71 | Physical pain | 41,586 |
| 46 | 23.47 | 22.41 | 22.94 | Physical pain | 38,198 |
| 47 | 24.38 | 23.27 | 23.83 | Physical pain | 41,474 |
| 48 | 25.31 | 24.25 | 24.78 | Physical pain | 42,376 |
| 49 | 25.66 | 24.59 | 25.12 | Physical pain | 42,562 |
| 50 | 25.61 | 24.57 | 25.09 | Physical pain | 59,291 |
| 51 | 27.47 | 26.29 | 26.88 | Physical pain | 43,410 |
| 52 | 26.88 | 25.80 | 26.34 | Physical pain | 52,500 |
| 53 | 28.15 | 27.05 | 27.60 | Physical pain | 49,738 |
| 54 | 28.39 | 27.34 | 27.86 | Physical pain | 50,954 |
| 55 | 28.16 | 27.12 | 27.64 | Physical pain | 56,786 |
| 56 | 29.02 | 27.98 | 28.50 | Physical pain | 51,524 |
| 57 | 29.27 | 28.22 | 28.74 | Physical pain | 52,831 |
| 58 | 29.50 | 28.51 | 29 | Physical pain | 54,069 |
| 59 | 29.48 | 28.46 | 28.97 | Physical pain | 53,045 |
| 60 | 28.67 | 27.69 | 28.18 | Physical pain | 65,070 |
| 61 | 29.39 | 28.40 | 28.90 | Physical pain | 50,842 |
| 62 | 29.29 | 28.36 | 28.83 | Physical pain | 60,276 |
| 63 | 29.33 | 28.37 | 28.85 | Physical pain | 53,010 |
| 64 | 28.36 | 27.42 | 27.89 | Physical pain | 53,742 |
| 65 | 26.96 | 26.07 | 26.51 | Physical pain | 62,191 |
| 66 | 27.15 | 26.26 | 26.71 | Physical pain | 50,128 |
| 67 | 27.40 | 26.52 | 26.96 | Physical pain | 50,937 |
| 68 | 26.92 | 25.96 | 26.44 | Physical pain | 48,643 |
| 69 | 27.01 | 26.08 | 26.54 | Physical pain | 43,860 |
| 70 | 25.76 | 24.88 | 25.32 | Physical pain | 53,255 |
| 71 | 26.55 | 25.59 | 26.07 | Physical pain | 38,542 |
| 72 | 26.42 | 25.46 | 25.94 | Physical pain | 41,689 |
| 73 | 26.54 | 25.59 | 26.07 | Physical pain | 37,738 |
| 74 | 26.55 | 25.54 | 26.04 | Physical pain | 34,693 |
| 75 | 26.47 | 25.54 | 26.01 | Physical pain | 37,873 |
| 76 | 27.11 | 26.11 | 26.61 | Physical pain | 30,802 |
| 77 | 27.23 | 26.15 | 26.69 | Physical pain | 30,148 |
| 78 | 26.90 | 25.84 | 26.37 | Physical pain | 29,063 |
| 79 | 27.05 | 25.91 | 26.48 | Physical pain | 25,544 |
| 80 | 25.98 | 24.98 | 25.48 | Physical pain | 31,099 |
| 81 | 27.08 | 25.88 | 26.48 | Physical pain | 22,405 |
| 82 | 27.61 | 26.41 | 27.01 | Physical pain | 22,630 |
| 83 | 27.75 | 26.48 | 27.12 | Physical pain | 20,190 |
| 84 | 27.44 | 26.10 | 26.77 | Physical pain | 17,760 |
| 85 | 27.04 | 25.74 | 26.39 | Physical pain | 17,070 |
| 86 | 27.71 | 26.23 | 26.97 | Physical pain | 13,006 |
| 87 | 28.73 | 27.05 | 27.89 | Physical pain | 11,042 |
| 88 | 26.85 | 25.16 | 26.01 | Physical pain | 9,766 |
| 89 | 27.81 | 25.54 | 26.67 | Physical pain | 6,827 |
| 90 | 25.95 | 23.82 | 24.89 | Physical pain | 6,381 |
| 91 | 27.16 | 24.38 | 25.77 | Physical pain | 3,896 |
| 92 | 27.92 | 24.78 | 26.35 | Physical pain | 3,101 |
| 93 | 26.80 | 23.17 | 24.99 | Physical pain | 2,045 |
| 94 | 26.09 | 21.57 | 23.83 | Physical pain | 1,448 |

#### Table 2.3. Age at the turn

Worry peaks near age 48, sadness near age 55 and physical pain near age 60, enjoyment and happiness are lowest near ages 52 and 54, and stress peaks earlier, near age 26

Age of the highest or lowest point of the five-year moving average of each profile, with 95% intervals from a bootstrap over counties

| age | conf_high | conf_low | group | label | n |
| ---: | ---: | ---: | --- | --- | ---: |
| 52.20 | 52.80 | 51.80 | Lowest point | Enjoyment | 2,581,678 |
| 54.40 | 55.60 | 53.90 | Lowest point | Happiness | 2,424,735 |
| 48.20 | 49.10 | 46.90 | Highest point | Worry | 2,585,390 |
| 25.90 | 32.80 | 25.10 | Highest point | Stress | 2,585,120 |
| 55.10 | 55.40 | 54.80 | Highest point | Sadness | 2,427,607 |
| 59.90 | 60.80 | 58.40 | Highest point | Physical pain | 2,584,986 |

Note: Each age is a different group of people interviewed in the same years, so a profile mixes age with birth cohort and does not follow anyone through life.

Note: Cells at the oldest ages hold fewer interviews, so their intervals are wider; the extract has no interviews above age 94.

Note: Happiness and sadness were asked in 2008 to 2016, anger in 2008 to 2013 and 2016 and the other feelings in 2008 to 2017. From 2013 each item was asked of 49% of interviews. The feelings keep 2008: the largest difference from 2009 among the seven is −1.6 points.

Note: Gallup's extract has no survey weights, so every estimate is unweighted and levels can differ from the weighted figures Gallup publishes. County comes from the respondent's ZIP code and is approximate near county lines.

Source: Gallup U.S. Daily, ladder 2009-2017 and feelings 2008-2017; unweighted, intervals clustered by county

Data: https://vishalsingh.org/zipcode/report/well-being-across-the-life-span/figures/feelings_by_age.json

### Table 3. The ladder falls into midlife and rises after it in every third of interviews by county deprivation, and the fall from ages 25 to 34 to ages 45 to 54 is −0.17 points in the most deprived third and −0.11 in the least

Cantril ladder score and feelings yesterday by year of age, for the least deprived, middle and most deprived third of interviews ranked by the Area Deprivation Index of the respondent's county, Gallup ladder 2009 to 2017 and feelings 2008 to 2017

#### Table 3.1. Ladder

The ladder is lower in the most deprived third of interviews in every age band, with gaps of −0.07 points at ages 18 to 24, −0.21 at ages 45 to 54 and −0.12 at ages 75 to 94

Unweighted mean Cantril ladder score by year of age and group of interviews by county deprivation, Gallup 2009 to 2017

| age | conf_high | conf_low | estimate | group | n |
| ---: | ---: | ---: | ---: | --- | ---: |
| 18 | 7.32 | 7.23 | 7.27 | Least deprived third | 9,352 |
| 19 | 7.19 | 7.11 | 7.15 | Least deprived third | 7,498 |
| 20 | 7.12 | 7.04 | 7.08 | Least deprived third | 7,314 |
| 21 | 7.10 | 7.01 | 7.05 | Least deprived third | 7,576 |
| 22 | 7.06 | 6.96 | 7.01 | Least deprived third | 7,398 |
| 23 | 7.00 | 6.91 | 6.96 | Least deprived third | 7,627 |
| 24 | 7.00 | 6.92 | 6.96 | Least deprived third | 7,574 |
| 25 | 7.03 | 6.95 | 6.99 | Least deprived third | 8,549 |
| 26 | 7.02 | 6.93 | 6.97 | Least deprived third | 7,666 |
| 27 | 7.00 | 6.92 | 6.96 | Least deprived third | 7,878 |
| 28 | 7.08 | 7.00 | 7.04 | Least deprived third | 8,371 |
| 29 | 7.10 | 7.01 | 7.06 | Least deprived third | 8,012 |
| 30 | 7.11 | 7.04 | 7.08 | Least deprived third | 10,053 |
| 31 | 7.16 | 7.06 | 7.11 | Least deprived third | 7,791 |
| 32 | 7.11 | 7.03 | 7.07 | Least deprived third | 9,162 |
| 33 | 7.13 | 7.06 | 7.10 | Least deprived third | 8,365 |
| 34 | 7.15 | 7.06 | 7.11 | Least deprived third | 8,587 |
| 35 | 7.16 | 7.08 | 7.12 | Least deprived third | 10,273 |
| 36 | 7.15 | 7.06 | 7.11 | Least deprived third | 8,688 |
| 37 | 7.13 | 7.05 | 7.09 | Least deprived third | 8,703 |
| 38 | 7.12 | 7.04 | 7.08 | Least deprived third | 9,269 |
| 39 | 7.14 | 7.06 | 7.10 | Least deprived third | 8,995 |
| 40 | 7.09 | 7.02 | 7.06 | Least deprived third | 12,478 |
| 41 | 7.15 | 7.06 | 7.10 | Least deprived third | 8,782 |
| 42 | 7.08 | 6.99 | 7.03 | Least deprived third | 10,953 |
| 43 | 7.06 | 6.97 | 7.01 | Least deprived third | 9,898 |
| 44 | 7.06 | 6.97 | 7.01 | Least deprived third | 9,810 |
| 45 | 7.07 | 6.98 | 7.02 | Least deprived third | 12,018 |
| 46 | 7.02 | 6.93 | 6.97 | Least deprived third | 10,751 |
| 47 | 6.98 | 6.89 | 6.93 | Least deprived third | 11,610 |
| 48 | 6.98 | 6.90 | 6.94 | Least deprived third | 11,867 |
| 49 | 6.96 | 6.86 | 6.91 | Least deprived third | 11,710 |
| 50 | 6.97 | 6.89 | 6.93 | Least deprived third | 17,124 |
| 51 | 7.00 | 6.92 | 6.96 | Least deprived third | 11,747 |
| 52 | 6.95 | 6.86 | 6.91 | Least deprived third | 14,482 |
| 53 | 6.94 | 6.84 | 6.89 | Least deprived third | 13,416 |
| 54 | 6.96 | 6.87 | 6.91 | Least deprived third | 13,484 |
| 55 | 7.00 | 6.91 | 6.96 | Least deprived third | 15,505 |
| 56 | 6.95 | 6.87 | 6.91 | Least deprived third | 13,706 |
| 57 | 6.95 | 6.87 | 6.91 | Least deprived third | 13,995 |
| 58 | 6.97 | 6.89 | 6.93 | Least deprived third | 14,135 |
| 59 | 7.04 | 6.95 | 6.99 | Least deprived third | 13,973 |
| 60 | 7.10 | 7.03 | 7.07 | Least deprived third | 17,630 |
| 61 | 7.10 | 7.00 | 7.05 | Least deprived third | 13,070 |
| 62 | 7.18 | 7.12 | 7.15 | Least deprived third | 16,250 |
| 63 | 7.22 | 7.13 | 7.17 | Least deprived third | 14,101 |
| 64 | 7.28 | 7.20 | 7.24 | Least deprived third | 14,308 |
| 65 | 7.30 | 7.22 | 7.26 | Least deprived third | 16,522 |
| 66 | 7.36 | 7.28 | 7.32 | Least deprived third | 13,195 |
| 67 | 7.43 | 7.35 | 7.39 | Least deprived third | 13,502 |
| 68 | 7.43 | 7.34 | 7.38 | Least deprived third | 12,771 |
| 69 | 7.46 | 7.38 | 7.42 | Least deprived third | 11,352 |
| 70 | 7.49 | 7.40 | 7.45 | Least deprived third | 14,159 |
| 71 | 7.49 | 7.41 | 7.45 | Least deprived third | 9,805 |
| 72 | 7.48 | 7.39 | 7.43 | Least deprived third | 10,531 |
| 73 | 7.46 | 7.38 | 7.42 | Least deprived third | 9,433 |
| 74 | 7.49 | 7.39 | 7.44 | Least deprived third | 8,662 |
| 75 | 7.49 | 7.41 | 7.45 | Least deprived third | 9,452 |
| 76 | 7.48 | 7.39 | 7.43 | Least deprived third | 7,525 |
| 77 | 7.48 | 7.39 | 7.44 | Least deprived third | 7,373 |
| 78 | 7.48 | 7.38 | 7.43 | Least deprived third | 6,971 |
| 79 | 7.46 | 7.36 | 7.41 | Least deprived third | 6,340 |
| 80 | 7.50 | 7.41 | 7.45 | Least deprived third | 7,924 |
| 81 | 7.51 | 7.41 | 7.46 | Least deprived third | 5,532 |
| 82 | 7.50 | 7.40 | 7.45 | Least deprived third | 5,740 |
| 83 | 7.43 | 7.31 | 7.37 | Least deprived third | 5,160 |
| 84 | 7.49 | 7.37 | 7.43 | Least deprived third | 4,468 |
| 85 | 7.46 | 7.34 | 7.40 | Least deprived third | 4,535 |
| 86 | 7.50 | 7.36 | 7.43 | Least deprived third | 3,367 |
| 87 | 7.46 | 7.30 | 7.38 | Least deprived third | 2,921 |
| 88 | 7.49 | 7.33 | 7.41 | Least deprived third | 2,641 |
| 89 | 7.53 | 7.33 | 7.43 | Least deprived third | 1,849 |
| 90 | 7.62 | 7.42 | 7.52 | Least deprived third | 1,809 |
| 91 | 7.73 | 7.51 | 7.62 | Least deprived third | 1,082 |
| 92 | 7.49 | 7.19 | 7.34 | Least deprived third | 858 |
| 93 | 7.57 | 7.22 | 7.40 | Least deprived third | 603 |
| 94 | 7.50 | 7.06 | 7.28 | Least deprived third | 419 |
| 18 | 7.28 | 7.20 | 7.24 | Middle third | 8,347 |
| 19 | 7.18 | 7.10 | 7.14 | Middle third | 6,868 |
| 20 | 7.07 | 6.98 | 7.02 | Middle third | 6,846 |
| 21 | 7.05 | 6.96 | 7.01 | Middle third | 7,168 |
| 22 | 6.97 | 6.89 | 6.93 | Middle third | 6,617 |
| 23 | 6.89 | 6.81 | 6.85 | Middle third | 6,982 |
| 24 | 6.89 | 6.80 | 6.84 | Middle third | 6,881 |
| 25 | 6.96 | 6.87 | 6.91 | Middle third | 7,385 |
| 26 | 6.95 | 6.87 | 6.91 | Middle third | 6,774 |
| 27 | 7.01 | 6.92 | 6.96 | Middle third | 7,012 |
| 28 | 7.05 | 6.97 | 7.01 | Middle third | 7,479 |
| 29 | 7.02 | 6.93 | 6.97 | Middle third | 7,269 |
| 30 | 7.07 | 6.98 | 7.03 | Middle third | 8,783 |
| 31 | 7.07 | 6.98 | 7.03 | Middle third | 7,099 |
| 32 | 7.07 | 6.99 | 7.03 | Middle third | 8,355 |
| 33 | 7.08 | 7.00 | 7.04 | Middle third | 7,621 |
| 34 | 7.04 | 6.95 | 7.00 | Middle third | 7,850 |
| 35 | 7.08 | 7.00 | 7.04 | Middle third | 8,925 |
| 36 | 7.11 | 7.02 | 7.07 | Middle third | 8,082 |
| 37 | 7.07 | 6.98 | 7.03 | Middle third | 8,174 |
| 38 | 7.04 | 6.95 | 7.00 | Middle third | 8,426 |
| 39 | 7.08 | 6.99 | 7.04 | Middle third | 8,268 |
| 40 | 7.07 | 6.98 | 7.03 | Middle third | 11,000 |
| 41 | 7.04 | 6.93 | 6.98 | Middle third | 8,098 |
| 42 | 7.01 | 6.92 | 6.96 | Middle third | 10,173 |
| 43 | 6.96 | 6.88 | 6.92 | Middle third | 9,331 |
| 44 | 6.96 | 6.88 | 6.92 | Middle third | 9,065 |
| 45 | 6.94 | 6.85 | 6.90 | Middle third | 11,236 |
| 46 | 6.92 | 6.83 | 6.88 | Middle third | 10,369 |
| 47 | 6.90 | 6.80 | 6.85 | Middle third | 11,259 |
| 48 | 6.90 | 6.82 | 6.86 | Middle third | 11,628 |
| 49 | 6.92 | 6.83 | 6.88 | Middle third | 11,452 |
| 50 | 6.91 | 6.83 | 6.87 | Middle third | 16,013 |
| 51 | 6.89 | 6.81 | 6.85 | Middle third | 11,787 |
| 52 | 6.88 | 6.80 | 6.84 | Middle third | 14,572 |
| 53 | 6.84 | 6.76 | 6.80 | Middle third | 13,756 |
| 54 | 6.86 | 6.78 | 6.82 | Middle third | 14,096 |
| 55 | 6.88 | 6.81 | 6.84 | Middle third | 15,749 |
| 56 | 6.86 | 6.78 | 6.82 | Middle third | 14,495 |
| 57 | 6.92 | 6.85 | 6.88 | Middle third | 14,703 |
| 58 | 6.91 | 6.84 | 6.87 | Middle third | 14,995 |
| 59 | 6.97 | 6.90 | 6.94 | Middle third | 14,668 |
| 60 | 7.03 | 6.96 | 7.00 | Middle third | 18,154 |
| 61 | 7.07 | 7 | 7.03 | Middle third | 14,035 |
| 62 | 7.10 | 7.04 | 7.07 | Middle third | 17,202 |
| 63 | 7.15 | 7.08 | 7.12 | Middle third | 15,501 |
| 64 | 7.19 | 7.12 | 7.16 | Middle third | 15,486 |
| 65 | 7.28 | 7.21 | 7.24 | Middle third | 18,127 |
| 66 | 7.34 | 7.27 | 7.31 | Middle third | 14,659 |
| 67 | 7.30 | 7.23 | 7.26 | Middle third | 15,038 |
| 68 | 7.34 | 7.27 | 7.31 | Middle third | 14,396 |
| 69 | 7.39 | 7.31 | 7.35 | Middle third | 12,952 |
| 70 | 7.38 | 7.31 | 7.34 | Middle third | 15,675 |
| 71 | 7.40 | 7.32 | 7.36 | Middle third | 11,443 |
| 72 | 7.42 | 7.34 | 7.38 | Middle third | 12,228 |
| 73 | 7.42 | 7.34 | 7.38 | Middle third | 10,982 |
| 74 | 7.42 | 7.33 | 7.38 | Middle third | 10,095 |
| 75 | 7.42 | 7.34 | 7.38 | Middle third | 11,066 |
| 76 | 7.41 | 7.32 | 7.37 | Middle third | 8,960 |
| 77 | 7.41 | 7.32 | 7.37 | Middle third | 8,668 |
| 78 | 7.46 | 7.37 | 7.41 | Middle third | 8,380 |
| 79 | 7.45 | 7.36 | 7.40 | Middle third | 7,402 |
| 80 | 7.49 | 7.41 | 7.45 | Middle third | 8,957 |
| 81 | 7.47 | 7.36 | 7.42 | Middle third | 6,371 |
| 82 | 7.47 | 7.36 | 7.42 | Middle third | 6,481 |
| 83 | 7.43 | 7.33 | 7.38 | Middle third | 5,825 |
| 84 | 7.48 | 7.36 | 7.42 | Middle third | 5,095 |
| 85 | 7.44 | 7.32 | 7.38 | Middle third | 4,898 |
| 86 | 7.42 | 7.28 | 7.34 | Middle third | 3,731 |
| 87 | 7.52 | 7.34 | 7.43 | Middle third | 3,244 |
| 88 | 7.48 | 7.33 | 7.41 | Middle third | 2,741 |
| 89 | 7.55 | 7.37 | 7.46 | Middle third | 2,063 |
| 90 | 7.62 | 7.42 | 7.51 | Middle third | 1,827 |
| 91 | 7.57 | 7.36 | 7.46 | Middle third | 1,146 |
| 92 | 7.69 | 7.41 | 7.55 | Middle third | 915 |
| 93 | 7.69 | 7.35 | 7.52 | Middle third | 589 |
| 94 | 7.85 | 7.42 | 7.64 | Middle third | 431 |
| 18 | 7.29 | 7.21 | 7.25 | Most deprived third | 7,397 |
| 19 | 7.17 | 7.07 | 7.12 | Most deprived third | 6,397 |
| 20 | 7.03 | 6.95 | 6.99 | Most deprived third | 6,403 |
| 21 | 7.00 | 6.92 | 6.96 | Most deprived third | 6,655 |
| 22 | 6.99 | 6.89 | 6.94 | Most deprived third | 6,249 |
| 23 | 6.94 | 6.85 | 6.90 | Most deprived third | 6,257 |
| 24 | 6.87 | 6.77 | 6.82 | Most deprived third | 6,188 |
| 25 | 6.93 | 6.83 | 6.88 | Most deprived third | 6,815 |
| 26 | 6.91 | 6.80 | 6.86 | Most deprived third | 6,129 |
| 27 | 6.93 | 6.84 | 6.89 | Most deprived third | 6,377 |
| 28 | 6.92 | 6.83 | 6.88 | Most deprived third | 6,754 |
| 29 | 6.98 | 6.89 | 6.94 | Most deprived third | 6,661 |
| 30 | 6.97 | 6.88 | 6.92 | Most deprived third | 7,974 |
| 31 | 6.97 | 6.87 | 6.92 | Most deprived third | 6,612 |
| 32 | 6.94 | 6.84 | 6.89 | Most deprived third | 7,600 |
| 33 | 6.97 | 6.88 | 6.92 | Most deprived third | 6,992 |
| 34 | 6.96 | 6.87 | 6.91 | Most deprived third | 7,006 |
| 35 | 6.97 | 6.88 | 6.93 | Most deprived third | 8,085 |
| 36 | 6.97 | 6.88 | 6.92 | Most deprived third | 7,063 |
| 37 | 6.94 | 6.85 | 6.90 | Most deprived third | 7,190 |
| 38 | 6.95 | 6.85 | 6.90 | Most deprived third | 7,775 |
| 39 | 6.96 | 6.88 | 6.92 | Most deprived third | 7,775 |
| 40 | 6.90 | 6.82 | 6.86 | Most deprived third | 9,938 |
| 41 | 6.91 | 6.82 | 6.87 | Most deprived third | 7,575 |
| 42 | 6.86 | 6.77 | 6.81 | Most deprived third | 9,387 |
| 43 | 6.82 | 6.73 | 6.77 | Most deprived third | 8,954 |
| 44 | 6.84 | 6.75 | 6.80 | Most deprived third | 8,845 |
| 45 | 6.84 | 6.76 | 6.80 | Most deprived third | 10,741 |
| 46 | 6.77 | 6.68 | 6.72 | Most deprived third | 10,046 |
| 47 | 6.76 | 6.68 | 6.72 | Most deprived third | 10,883 |
| 48 | 6.76 | 6.68 | 6.72 | Most deprived third | 11,130 |
| 49 | 6.76 | 6.68 | 6.72 | Most deprived third | 11,621 |
| 50 | 6.78 | 6.71 | 6.74 | Most deprived third | 16,033 |
| 51 | 6.76 | 6.68 | 6.72 | Most deprived third | 12,032 |
| 52 | 6.75 | 6.67 | 6.71 | Most deprived third | 14,532 |
| 53 | 6.74 | 6.67 | 6.70 | Most deprived third | 14,083 |
| 54 | 6.77 | 6.70 | 6.73 | Most deprived third | 14,436 |
| 55 | 6.77 | 6.70 | 6.73 | Most deprived third | 16,199 |
| 56 | 6.79 | 6.72 | 6.75 | Most deprived third | 14,892 |
| 57 | 6.80 | 6.73 | 6.77 | Most deprived third | 15,441 |
| 58 | 6.84 | 6.77 | 6.80 | Most deprived third | 15,961 |
| 59 | 6.85 | 6.79 | 6.82 | Most deprived third | 15,776 |
| 60 | 6.92 | 6.86 | 6.89 | Most deprived third | 18,786 |
| 61 | 6.98 | 6.91 | 6.94 | Most deprived third | 14,814 |
| 62 | 7.02 | 6.96 | 6.99 | Most deprived third | 18,116 |
| 63 | 7.04 | 6.98 | 7.01 | Most deprived third | 16,397 |
| 64 | 7.11 | 7.05 | 7.08 | Most deprived third | 16,468 |
| 65 | 7.15 | 7.09 | 7.12 | Most deprived third | 18,407 |
| 66 | 7.21 | 7.14 | 7.17 | Most deprived third | 15,247 |
| 67 | 7.22 | 7.16 | 7.19 | Most deprived third | 15,574 |
| 68 | 7.23 | 7.16 | 7.19 | Most deprived third | 14,892 |
| 69 | 7.25 | 7.18 | 7.22 | Most deprived third | 13,762 |
| 70 | 7.30 | 7.23 | 7.27 | Most deprived third | 16,259 |
| 71 | 7.25 | 7.17 | 7.21 | Most deprived third | 11,890 |
| 72 | 7.27 | 7.20 | 7.23 | Most deprived third | 12,971 |
| 73 | 7.28 | 7.20 | 7.24 | Most deprived third | 11,858 |
| 74 | 7.31 | 7.24 | 7.27 | Most deprived third | 11,042 |
| 75 | 7.30 | 7.22 | 7.26 | Most deprived third | 11,833 |
| 76 | 7.31 | 7.23 | 7.27 | Most deprived third | 9,683 |
| 77 | 7.33 | 7.24 | 7.28 | Most deprived third | 9,399 |
| 78 | 7.30 | 7.21 | 7.26 | Most deprived third | 9,009 |
| 79 | 7.36 | 7.27 | 7.32 | Most deprived third | 7,892 |
| 80 | 7.39 | 7.30 | 7.34 | Most deprived third | 9,286 |
| 81 | 7.40 | 7.30 | 7.35 | Most deprived third | 6,937 |
| 82 | 7.35 | 7.24 | 7.30 | Most deprived third | 6,844 |
| 83 | 7.35 | 7.24 | 7.30 | Most deprived third | 6,092 |
| 84 | 7.38 | 7.26 | 7.32 | Most deprived third | 5,395 |
| 85 | 7.36 | 7.24 | 7.30 | Most deprived third | 5,029 |
| 86 | 7.41 | 7.28 | 7.35 | Most deprived third | 3,847 |
| 87 | 7.37 | 7.22 | 7.29 | Most deprived third | 3,256 |
| 88 | 7.40 | 7.25 | 7.32 | Most deprived third | 2,828 |
| 89 | 7.46 | 7.26 | 7.36 | Most deprived third | 1,989 |
| 90 | 7.63 | 7.44 | 7.53 | Most deprived third | 1,816 |
| 91 | 7.57 | 7.29 | 7.43 | Most deprived third | 1,106 |
| 92 | 7.61 | 7.28 | 7.44 | Most deprived third | 868 |
| 93 | 7.65 | 7.28 | 7.47 | Most deprived third | 562 |
| 94 | 7.73 | 7.29 | 7.51 | Most deprived third | 393 |

#### Table 3.2. Physical pain

Physical pain is more common in the most deprived third of interviews, with gaps of +0.9 points at ages 18 to 24 and +6.8 at ages 45 to 54

Share of adults who reported physical pain during a lot of the day yesterday by year of age and group of interviews by county deprivation, Gallup 2008 to 2017

| age | conf_high | conf_low | estimate | group | n |
| ---: | ---: | ---: | ---: | --- | ---: |
| 18 | 14.99 | 13.82 | 14.41 | Least deprived third | 10,495 |
| 19 | 15.77 | 14.27 | 15.02 | Least deprived third | 8,307 |
| 20 | 16.20 | 14.60 | 15.40 | Least deprived third | 8,142 |
| 21 | 16.45 | 14.73 | 15.59 | Least deprived third | 8,395 |
| 22 | 16.14 | 14.43 | 15.29 | Least deprived third | 8,275 |
| 23 | 15.95 | 14.24 | 15.10 | Least deprived third | 8,545 |
| 24 | 16.46 | 14.63 | 15.54 | Least deprived third | 8,571 |
| 25 | 16.30 | 14.53 | 15.42 | Least deprived third | 9,743 |
| 26 | 16.24 | 14.30 | 15.27 | Least deprived third | 8,780 |
| 27 | 16.82 | 15.33 | 16.07 | Least deprived third | 9,052 |
| 28 | 16.14 | 14.77 | 15.45 | Least deprived third | 9,726 |
| 29 | 16.64 | 14.90 | 15.77 | Least deprived third | 9,341 |
| 30 | 16.39 | 14.98 | 15.68 | Least deprived third | 11,680 |
| 31 | 16.87 | 15.21 | 16.04 | Least deprived third | 9,072 |
| 32 | 17.33 | 15.85 | 16.59 | Least deprived third | 10,773 |
| 33 | 17.07 | 15.32 | 16.19 | Least deprived third | 9,943 |
| 34 | 17.74 | 15.97 | 16.86 | Least deprived third | 10,210 |
| 35 | 16.56 | 15.14 | 15.85 | Least deprived third | 12,205 |
| 36 | 17.81 | 16.17 | 16.99 | Least deprived third | 10,524 |
| 37 | 17.93 | 16.30 | 17.12 | Least deprived third | 10,873 |
| 38 | 17.21 | 15.78 | 16.49 | Least deprived third | 11,495 |
| 39 | 18.69 | 16.93 | 17.81 | Least deprived third | 11,034 |
| 40 | 18.23 | 16.63 | 17.43 | Least deprived third | 15,083 |
| 41 | 18.02 | 16.37 | 17.20 | Least deprived third | 10,788 |
| 42 | 19.01 | 17.20 | 18.11 | Least deprived third | 13,322 |
| 43 | 19.90 | 18.17 | 19.04 | Least deprived third | 12,233 |
| 44 | 20.39 | 18.39 | 19.39 | Least deprived third | 12,153 |
| 45 | 19.91 | 18.12 | 19.01 | Least deprived third | 14,674 |
| 46 | 20.85 | 19.10 | 19.97 | Least deprived third | 13,268 |
| 47 | 21.76 | 20.11 | 20.94 | Least deprived third | 14,381 |
| 48 | 23.43 | 21.51 | 22.47 | Least deprived third | 14,573 |
| 49 | 23.35 | 21.33 | 22.34 | Least deprived third | 14,418 |
| 50 | 22.91 | 21.21 | 22.06 | Least deprived third | 20,738 |
| 51 | 24.42 | 22.51 | 23.46 | Least deprived third | 14,384 |
| 52 | 23.92 | 22.30 | 23.11 | Least deprived third | 17,473 |
| 53 | 25.51 | 23.52 | 24.51 | Least deprived third | 16,297 |
| 54 | 25.74 | 23.84 | 24.79 | Least deprived third | 16,579 |
| 55 | 25.20 | 23.45 | 24.33 | Least deprived third | 18,609 |
| 56 | 26.60 | 24.88 | 25.74 | Least deprived third | 16,500 |
| 57 | 26.82 | 25.06 | 25.94 | Least deprived third | 16,888 |
| 58 | 26.94 | 25.09 | 26.01 | Least deprived third | 17,118 |
| 59 | 26.86 | 24.90 | 25.88 | Least deprived third | 16,738 |
| 60 | 26.03 | 24.27 | 25.15 | Least deprived third | 21,162 |
| 61 | 27.13 | 25.27 | 26.20 | Least deprived third | 15,964 |
| 62 | 27.20 | 25.44 | 26.32 | Least deprived third | 19,057 |
| 63 | 26.87 | 24.94 | 25.91 | Least deprived third | 16,382 |
| 64 | 26.45 | 24.62 | 25.53 | Least deprived third | 16,687 |
| 65 | 25.13 | 23.38 | 24.25 | Least deprived third | 19,473 |
| 66 | 25.55 | 24.06 | 24.80 | Least deprived third | 15,341 |
| 67 | 25.89 | 24.22 | 25.05 | Least deprived third | 15,535 |
| 68 | 25.30 | 23.56 | 24.43 | Least deprived third | 14,761 |
| 69 | 25.72 | 23.93 | 24.82 | Least deprived third | 13,052 |
| 70 | 24.54 | 22.91 | 23.73 | Least deprived third | 16,437 |
| 71 | 25.33 | 23.52 | 24.42 | Least deprived third | 11,386 |
| 72 | 25.01 | 23.21 | 24.11 | Least deprived third | 12,377 |
| 73 | 25.17 | 23.36 | 24.27 | Least deprived third | 11,036 |
| 74 | 25.55 | 23.65 | 24.60 | Least deprived third | 10,078 |
| 75 | 25.68 | 23.85 | 24.77 | Least deprived third | 11,177 |
| 76 | 26.11 | 24.36 | 25.23 | Least deprived third | 8,882 |
| 77 | 25.92 | 23.92 | 24.92 | Least deprived third | 8,777 |
| 78 | 26.59 | 24.67 | 25.63 | Least deprived third | 8,341 |
| 79 | 25.81 | 23.84 | 24.83 | Least deprived third | 7,472 |
| 80 | 25.48 | 23.80 | 24.64 | Least deprived third | 9,502 |
| 81 | 27.31 | 25.03 | 26.17 | Least deprived third | 6,569 |
| 82 | 26.39 | 24.25 | 25.32 | Least deprived third | 6,881 |
| 83 | 27.30 | 25 | 26.15 | Least deprived third | 6,126 |
| 84 | 26.91 | 24.52 | 25.72 | Least deprived third | 5,374 |
| 85 | 26.40 | 24.16 | 25.28 | Least deprived third | 5,372 |
| 86 | 27.37 | 24.79 | 26.08 | Least deprived third | 4,007 |
| 87 | 28.05 | 25.27 | 26.66 | Least deprived third | 3,421 |
| 88 | 26.26 | 23.70 | 24.98 | Least deprived third | 3,143 |
| 89 | 28.11 | 24.25 | 26.18 | Least deprived third | 2,135 |
| 90 | 26.61 | 22.80 | 24.70 | Least deprived third | 2,105 |
| 91 | 28.62 | 23.45 | 26.03 | Least deprived third | 1,283 |
| 92 | 30.18 | 24.26 | 27.22 | Least deprived third | 1,014 |
| 93 | 28.17 | 22.18 | 25.18 | Least deprived third | 703 |
| 94 | 27.40 | 19.42 | 23.41 | Least deprived third | 487 |
| 18 | 15.53 | 14.16 | 14.85 | Middle third | 9,511 |
| 19 | 15.81 | 14.30 | 15.06 | Middle third | 7,650 |
| 20 | 15.88 | 14.33 | 15.10 | Middle third | 7,660 |
| 21 | 16.29 | 14.60 | 15.44 | Middle third | 7,932 |
| 22 | 16.61 | 14.90 | 15.75 | Middle third | 7,459 |
| 23 | 16.41 | 14.59 | 15.50 | Middle third | 7,891 |
| 24 | 16.61 | 14.91 | 15.76 | Middle third | 7,803 |
| 25 | 16.58 | 14.99 | 15.79 | Middle third | 8,451 |
| 26 | 16.07 | 14.53 | 15.30 | Middle third | 7,837 |
| 27 | 17.60 | 15.54 | 16.57 | Middle third | 8,189 |
| 28 | 17.35 | 15.73 | 16.54 | Middle third | 8,844 |
| 29 | 17.13 | 15.57 | 16.35 | Middle third | 8,598 |
| 30 | 17.23 | 15.57 | 16.40 | Middle third | 10,288 |
| 31 | 17.76 | 15.64 | 16.70 | Middle third | 8,466 |
| 32 | 18.09 | 16.36 | 17.23 | Middle third | 9,834 |
| 33 | 18.57 | 16.78 | 17.67 | Middle third | 9,088 |
| 34 | 18.16 | 16.39 | 17.27 | Middle third | 9,320 |
| 35 | 18.29 | 16.63 | 17.46 | Middle third | 10,665 |
| 36 | 18.95 | 17.25 | 18.10 | Middle third | 9,807 |
| 37 | 19.04 | 17.46 | 18.25 | Middle third | 10,098 |
| 38 | 19.11 | 17.45 | 18.28 | Middle third | 10,522 |
| 39 | 19.05 | 17.17 | 18.11 | Middle third | 10,139 |
| 40 | 19.30 | 17.80 | 18.55 | Middle third | 13,219 |
| 41 | 20.65 | 18.94 | 19.79 | Middle third | 9,897 |
| 42 | 20.26 | 18.70 | 19.48 | Middle third | 12,408 |
| 43 | 21.50 | 19.82 | 20.66 | Middle third | 11,471 |
| 44 | 21.98 | 19.99 | 20.98 | Middle third | 11,248 |
| 45 | 22.10 | 20.38 | 21.24 | Middle third | 13,712 |
| 46 | 23.38 | 21.68 | 22.53 | Middle third | 12,610 |
| 47 | 24.08 | 22.33 | 23.20 | Middle third | 13,787 |
| 48 | 24.76 | 22.97 | 23.87 | Middle third | 14,129 |
| 49 | 25 | 23.21 | 24.10 | Middle third | 14,027 |
| 50 | 24.92 | 23.31 | 24.11 | Middle third | 19,306 |
| 51 | 27.44 | 25.40 | 26.42 | Middle third | 14,332 |
| 52 | 26.44 | 24.68 | 25.56 | Middle third | 17,418 |
| 53 | 27.90 | 26.15 | 27.02 | Middle third | 16,515 |
| 54 | 27.78 | 26.24 | 27.01 | Middle third | 16,938 |
| 55 | 27.71 | 26.10 | 26.90 | Middle third | 18,884 |
| 56 | 28.45 | 26.74 | 27.60 | Middle third | 17,316 |
| 57 | 28.93 | 27.12 | 28.03 | Middle third | 17,520 |
| 58 | 28.74 | 27.14 | 27.94 | Middle third | 17,953 |
| 59 | 28.87 | 27.28 | 28.07 | Middle third | 17,486 |
| 60 | 28.03 | 26.52 | 27.27 | Middle third | 21,614 |
| 61 | 28.72 | 27.21 | 27.97 | Middle third | 16,996 |
| 62 | 28.20 | 26.71 | 27.45 | Middle third | 20,147 |
| 63 | 28.64 | 27.18 | 27.91 | Middle third | 17,843 |
| 64 | 27.43 | 25.87 | 26.65 | Middle third | 17,923 |
| 65 | 26.15 | 24.83 | 25.49 | Middle third | 21,135 |
| 66 | 26.40 | 24.96 | 25.68 | Middle third | 16,921 |
| 67 | 26.84 | 25.47 | 26.16 | Middle third | 17,304 |
| 68 | 26.01 | 24.52 | 25.27 | Middle third | 16,566 |
| 69 | 26.75 | 25.13 | 25.94 | Middle third | 14,855 |
| 70 | 25.98 | 24.52 | 25.25 | Middle third | 17,959 |
| 71 | 26.10 | 24.57 | 25.34 | Middle third | 13,281 |
| 72 | 26.62 | 25.10 | 25.86 | Middle third | 14,214 |
| 73 | 26.27 | 24.75 | 25.51 | Middle third | 12,795 |
| 74 | 26.69 | 24.92 | 25.81 | Middle third | 11,722 |
| 75 | 25.87 | 24.40 | 25.14 | Middle third | 12,906 |
| 76 | 26.41 | 24.69 | 25.55 | Middle third | 10,505 |
| 77 | 27.33 | 25.51 | 26.42 | Middle third | 10,231 |
| 78 | 26.41 | 24.64 | 25.52 | Middle third | 9,959 |
| 79 | 27.89 | 25.98 | 26.94 | Middle third | 8,720 |
| 80 | 26.11 | 24.36 | 25.23 | Middle third | 10,605 |
| 81 | 26.08 | 24.10 | 25.09 | Middle third | 7,550 |
| 82 | 27.77 | 25.94 | 26.86 | Middle third | 7,633 |
| 83 | 27.57 | 25.45 | 26.51 | Middle third | 6,832 |
| 84 | 27.15 | 24.89 | 26.02 | Middle third | 5,992 |
| 85 | 27.20 | 24.96 | 26.08 | Middle third | 5,739 |
| 86 | 28.94 | 26.24 | 27.59 | Middle third | 4,455 |
| 87 | 29.64 | 26.66 | 28.15 | Middle third | 3,801 |
| 88 | 27.65 | 24.56 | 26.11 | Middle third | 3,271 |
| 89 | 27.71 | 23.92 | 25.81 | Middle third | 2,363 |
| 90 | 26.37 | 22.65 | 24.51 | Middle third | 2,146 |
| 91 | 27.88 | 23.49 | 25.68 | Middle third | 1,316 |
| 92 | 26.82 | 21.95 | 24.39 | Middle third | 1,058 |
| 93 | 27.25 | 20.86 | 24.06 | Middle third | 690 |
| 94 | 26.68 | 18.87 | 22.77 | Middle third | 483 |
| 18 | 15.27 | 13.77 | 14.52 | Most deprived third | 8,560 |
| 19 | 16.01 | 14.31 | 15.16 | Most deprived third | 7,302 |
| 20 | 17.63 | 15.71 | 16.67 | Most deprived third | 7,222 |
| 21 | 16.77 | 14.95 | 15.86 | Most deprived third | 7,455 |
| 22 | 17.66 | 15.90 | 16.78 | Most deprived third | 7,081 |
| 23 | 17.24 | 15.55 | 16.40 | Most deprived third | 7,172 |
| 24 | 18.13 | 16.24 | 17.19 | Most deprived third | 7,070 |
| 25 | 18.45 | 16.65 | 17.55 | Most deprived third | 7,845 |
| 26 | 18.79 | 16.93 | 17.86 | Most deprived third | 7,230 |
| 27 | 19.04 | 17.23 | 18.14 | Most deprived third | 7,521 |
| 28 | 18.82 | 17.09 | 17.95 | Most deprived third | 8,048 |
| 29 | 19.11 | 17.36 | 18.24 | Most deprived third | 7,935 |
| 30 | 19.58 | 17.88 | 18.73 | Most deprived third | 9,477 |
| 31 | 20.10 | 18.33 | 19.21 | Most deprived third | 7,907 |
| 32 | 20.14 | 18.39 | 19.26 | Most deprived third | 9,007 |
| 33 | 20.53 | 18.78 | 19.66 | Most deprived third | 8,364 |
| 34 | 19.67 | 17.89 | 18.78 | Most deprived third | 8,423 |
| 35 | 21.32 | 19.69 | 20.50 | Most deprived third | 9,731 |
| 36 | 21.39 | 19.54 | 20.46 | Most deprived third | 8,659 |
| 37 | 22.18 | 20.49 | 21.34 | Most deprived third | 9,060 |
| 38 | 22.15 | 20.46 | 21.30 | Most deprived third | 9,618 |
| 39 | 22.40 | 20.73 | 21.57 | Most deprived third | 9,450 |
| 40 | 22.91 | 21.37 | 22.14 | Most deprived third | 11,995 |
| 41 | 23.83 | 22.11 | 22.97 | Most deprived third | 9,316 |
| 42 | 24.42 | 22.78 | 23.60 | Most deprived third | 11,394 |
| 43 | 25.39 | 23.65 | 24.52 | Most deprived third | 10,995 |
| 44 | 26.09 | 24.31 | 25.20 | Most deprived third | 10,840 |
| 45 | 25.98 | 24.42 | 25.20 | Most deprived third | 13,192 |
| 46 | 27.43 | 25.67 | 26.55 | Most deprived third | 12,309 |
| 47 | 28.43 | 26.72 | 27.58 | Most deprived third | 13,290 |
| 48 | 28.98 | 27.38 | 28.18 | Most deprived third | 13,664 |
| 49 | 29.79 | 28.14 | 28.97 | Most deprived third | 14,109 |
| 50 | 30.06 | 28.62 | 29.34 | Most deprived third | 19,225 |
| 51 | 31.54 | 29.82 | 30.68 | Most deprived third | 14,683 |
| 52 | 31.11 | 29.57 | 30.34 | Most deprived third | 17,593 |
| 53 | 31.97 | 30.38 | 31.17 | Most deprived third | 16,905 |
| 54 | 32.45 | 30.83 | 31.64 | Most deprived third | 17,418 |
| 55 | 32.28 | 30.83 | 31.56 | Most deprived third | 19,271 |
| 56 | 32.74 | 31.14 | 31.94 | Most deprived third | 17,691 |
| 57 | 32.76 | 31.23 | 32 | Most deprived third | 18,409 |
| 58 | 33.44 | 31.98 | 32.71 | Most deprived third | 18,977 |
| 59 | 33.27 | 31.80 | 32.54 | Most deprived third | 18,805 |
| 60 | 32.64 | 31.24 | 31.94 | Most deprived third | 22,276 |
| 61 | 32.95 | 31.44 | 32.20 | Most deprived third | 17,862 |
| 62 | 33.13 | 31.71 | 32.42 | Most deprived third | 21,055 |
| 63 | 33.03 | 31.62 | 32.32 | Most deprived third | 18,765 |
| 64 | 31.78 | 30.40 | 31.09 | Most deprived third | 19,112 |
| 65 | 30.23 | 28.86 | 29.55 | Most deprived third | 21,564 |
| 66 | 30.02 | 28.61 | 29.32 | Most deprived third | 17,854 |
| 67 | 30.07 | 28.68 | 29.37 | Most deprived third | 18,088 |
| 68 | 30.03 | 28.54 | 29.28 | Most deprived third | 17,307 |
| 69 | 29.22 | 27.83 | 28.52 | Most deprived third | 15,939 |
| 70 | 27.44 | 26.10 | 26.77 | Most deprived third | 18,834 |
| 71 | 28.85 | 27.36 | 28.10 | Most deprived third | 13,863 |
| 72 | 28.24 | 26.77 | 27.51 | Most deprived third | 15,090 |
| 73 | 28.77 | 27.26 | 28.02 | Most deprived third | 13,898 |
| 74 | 28.11 | 26.62 | 27.37 | Most deprived third | 12,888 |
| 75 | 28.57 | 27.08 | 27.83 | Most deprived third | 13,781 |
| 76 | 29.46 | 27.82 | 28.64 | Most deprived third | 11,408 |
| 77 | 29.15 | 27.46 | 28.31 | Most deprived third | 11,132 |
| 78 | 28.58 | 26.85 | 27.71 | Most deprived third | 10,757 |
| 79 | 28.28 | 26.41 | 27.35 | Most deprived third | 9,339 |
| 80 | 27.31 | 25.60 | 26.46 | Most deprived third | 10,988 |
| 81 | 28.93 | 27.05 | 27.99 | Most deprived third | 8,281 |
| 82 | 29.60 | 27.60 | 28.60 | Most deprived third | 8,109 |
| 83 | 29.57 | 27.47 | 28.52 | Most deprived third | 7,230 |
| 84 | 29.47 | 27.25 | 28.36 | Most deprived third | 6,392 |
| 85 | 28.81 | 26.56 | 27.69 | Most deprived third | 5,959 |
| 86 | 28.37 | 25.93 | 27.15 | Most deprived third | 4,542 |
| 87 | 30.18 | 27.34 | 28.76 | Most deprived third | 3,818 |
| 88 | 28.40 | 25.36 | 26.88 | Most deprived third | 3,352 |
| 89 | 30.02 | 25.91 | 27.96 | Most deprived third | 2,328 |
| 90 | 27.19 | 23.68 | 25.43 | Most deprived third | 2,127 |
| 91 | 27.95 | 23.13 | 25.54 | Most deprived third | 1,296 |
| 92 | 30.22 | 24.79 | 27.50 | Most deprived third | 1,029 |
| 93 | 29.04 | 22.49 | 25.77 | Most deprived third | 652 |
| 94 | 29.13 | 21.50 | 25.31 | Most deprived third | 478 |

#### Table 3.3. Gap in the feelings

Except for physical pain, feelings differ by at most 2.9 points between the most and least deprived thirds of interviews in any age band, and physical pain has a gap of +6.8 points at ages 45 to 54

Difference in the share reporting each feeling yesterday, most deprived third minus least deprived third of interviews, by age band, with 95% intervals clustered by county, Gallup 2008 to 2017

| conf_high | conf_low | estimate | group | label | n |
| ---: | ---: | ---: | --- | --- | ---: |
| 0.570 | −0.580 | 0 | Enjoyment | Ages 18 to 24 | 112,542 |
| 0.510 | −0.630 | −0.060 | Enjoyment | Ages 25 to 34 | 179,962 |
| −0.530 | −1.49 | −1.01 | Enjoyment | Ages 35 to 44 | 220,497 |
| −0.970 | −2.05 | −1.51 | Enjoyment | Ages 45 to 54 | 308,777 |
| −0.490 | −1.50 | −0.990 | Enjoyment | Ages 55 to 64 | 366,897 |
| −0.650 | −1.72 | −1.19 | Enjoyment | Ages 65 to 74 | 304,381 |
| 0.560 | −0.500 | 0.030 | Enjoyment | Ages 75 to 94 | 225,148 |
| −2.16 | −3.57 | −2.86 | Worry | Ages 18 to 24 | 112,597 |
| −0.680 | −2.13 | −1.41 | Worry | Ages 25 to 34 | 180,083 |
| 0.320 | −1.10 | −0.390 | Worry | Ages 35 to 44 | 220,697 |
| 0.490 | −0.930 | −0.220 | Worry | Ages 45 to 54 | 309,157 |
| −0.110 | −1.44 | −0.770 | Worry | Ages 55 to 64 | 367,349 |
| 0.280 | −1.03 | −0.380 | Worry | Ages 65 to 74 | 304,938 |
| 0.270 | −0.980 | −0.350 | Worry | Ages 75 to 94 | 225,950 |
| −0.210 | −2.45 | −1.33 | Stress | Ages 18 to 24 | 112,582 |
| 0.430 | −1.06 | −0.320 | Stress | Ages 25 to 34 | 180,065 |
| 0.520 | −0.700 | −0.090 | Stress | Ages 35 to 44 | 220,640 |
| 0.790 | −0.290 | 0.250 | Stress | Ages 45 to 54 | 309,120 |
| −0.330 | −1.36 | −0.840 | Stress | Ages 55 to 64 | 367,272 |
| −0.360 | −1.49 | −0.930 | Stress | Ages 65 to 74 | 304,911 |
| −0.230 | −1.24 | −0.730 | Stress | Ages 75 to 94 | 225,958 |
| −0.400 | −1.49 | −0.950 | Sadness | Ages 18 to 24 | 103,310 |
| 0.940 | −0.590 | 0.180 | Sadness | Ages 25 to 34 | 166,160 |
| 2.26 | 0.600 | 1.43 | Sadness | Ages 35 to 44 | 207,288 |
| 2.93 | 1.36 | 2.15 | Sadness | Ages 45 to 54 | 292,866 |
| 2.11 | 0.820 | 1.46 | Sadness | Ages 55 to 64 | 347,212 |
| 1.66 | 0.550 | 1.11 | Sadness | Ages 65 to 74 | 285,849 |
| 1.11 | 0.020 | 0.560 | Sadness | Ages 75 to 94 | 213,833 |
| 1.39 | 0.340 | 0.870 | Physical pain | Ages 18 to 24 | 112,592 |
| 3.21 | 2.03 | 2.62 | Physical pain | Ages 25 to 34 | 180,077 |
| 5.52 | 4.28 | 4.90 | Physical pain | Ages 35 to 44 | 220,768 |
| 7.62 | 6.03 | 6.83 | Physical pain | Ages 45 to 54 | 309,173 |
| 7.17 | 5.61 | 6.39 | Physical pain | Ages 55 to 64 | 367,328 |
| 4.62 | 3.41 | 4.01 | Physical pain | Ages 65 to 74 | 304,801 |
| 2.90 | 1.98 | 2.44 | Physical pain | Ages 75 to 94 | 225,769 |

Note: Each age is a different group of people interviewed in the same years, so a profile mixes age with birth cohort and does not follow anyone through life.

Note: Cells at the oldest ages hold fewer interviews, so their intervals are wider; the extract has no interviews above age 94.

Note: County deprivation is the mean Area Deprivation Index 2023 national rank of the county's tracts, weighted by tract population (1 is least deprived), joined to interviews from 2008 to 2017, so it describes counties as they were later than the interviews. Interviews are ranked on it and cut into three groups of about equal size; a county sits in one group. 931 interviews are in counties with no rank and are left out of this figure.

Note: The ladder of 2008 is not used. Political questions that came before the ladder lowered it in 2008 and in the first months of 2009, and the extract has no interview month, so early 2009 interviews may carry part of the effect. Scores averaged 6.69 in 2008 and 7.03 in 2009, and the 2009 minus 2008 difference is between 0.29 and 0.38 ladder points in every age band.

Note: Gallup's extract has no survey weights, so every estimate is unweighted and levels can differ from the weighted figures Gallup publishes. County comes from the respondent's ZIP code and is approximate near county lines.

Source: Gallup U.S. Daily, ladder 2009-2017 and feelings 2008-2017; unweighted, intervals clustered by county; county Area Deprivation Index 2023 joined on the county of the interview

Data: https://vishalsingh.org/zipcode/report/well-being-across-the-life-span/figures/profiles_by_place.json

### Table 4. The fall in life evaluation from ages 25 to 34 to ages 45 to 54 is −0.59 ladder points in households under $36,000 and −0.05 in households of $90,000 or more, where the fall from ages 35 to 44 is −0.07

Cantril ladder score and feelings yesterday by year of age, for interviews in 2009 to 2014 in households with annual income under the lower cutoff, between the cutoffs and above the upper cutoff, Gallup

#### Table 4.1. Ladder

Ladder scores fall to 5.89 at ages 45 to 54 in households under $36,000, while in households of $90,000 or more they are 7.51 at ages 35 to 44 and 7.45 at ages 45 to 54

Unweighted mean Cantril ladder score by year of age and household income group, Gallup 2009 to 2014

| age | conf_high | conf_low | estimate | group | n |
| ---: | ---: | ---: | ---: | --- | ---: |
| 18 | 7.14 | 7.04 | 7.09 | Under $36,000 | 5,686 |
| 19 | 7.00 | 6.91 | 6.96 | Under $36,000 | 5,858 |
| 20 | 6.97 | 6.87 | 6.92 | Under $36,000 | 6,592 |
| 21 | 6.94 | 6.85 | 6.89 | Under $36,000 | 7,219 |
| 22 | 6.85 | 6.76 | 6.81 | Under $36,000 | 6,854 |
| 23 | 6.75 | 6.66 | 6.71 | Under $36,000 | 6,753 |
| 24 | 6.71 | 6.60 | 6.65 | Under $36,000 | 6,544 |
| 25 | 6.71 | 6.62 | 6.66 | Under $36,000 | 6,878 |
| 26 | 6.60 | 6.50 | 6.55 | Under $36,000 | 5,908 |
| 27 | 6.56 | 6.45 | 6.50 | Under $36,000 | 5,783 |
| 28 | 6.57 | 6.46 | 6.52 | Under $36,000 | 6,052 |
| 29 | 6.49 | 6.37 | 6.43 | Under $36,000 | 5,352 |
| 30 | 6.56 | 6.43 | 6.49 | Under $36,000 | 6,511 |
| 31 | 6.47 | 6.33 | 6.40 | Under $36,000 | 4,588 |
| 32 | 6.46 | 6.33 | 6.40 | Under $36,000 | 5,552 |
| 33 | 6.43 | 6.30 | 6.37 | Under $36,000 | 4,611 |
| 34 | 6.41 | 6.27 | 6.34 | Under $36,000 | 4,532 |
| 35 | 6.41 | 6.28 | 6.34 | Under $36,000 | 5,368 |
| 36 | 6.38 | 6.22 | 6.30 | Under $36,000 | 4,409 |
| 37 | 6.29 | 6.14 | 6.21 | Under $36,000 | 4,486 |
| 38 | 6.29 | 6.16 | 6.23 | Under $36,000 | 4,752 |
| 39 | 6.25 | 6.12 | 6.19 | Under $36,000 | 4,443 |
| 40 | 6.28 | 6.16 | 6.22 | Under $36,000 | 6,404 |
| 41 | 6.15 | 6.01 | 6.08 | Under $36,000 | 4,289 |
| 42 | 6.08 | 5.95 | 6.02 | Under $36,000 | 5,550 |
| 43 | 6.06 | 5.93 | 5.99 | Under $36,000 | 5,208 |
| 44 | 6.10 | 5.97 | 6.04 | Under $36,000 | 5,001 |
| 45 | 6.03 | 5.91 | 5.97 | Under $36,000 | 6,302 |
| 46 | 5.95 | 5.83 | 5.89 | Under $36,000 | 5,860 |
| 47 | 5.97 | 5.85 | 5.91 | Under $36,000 | 6,436 |
| 48 | 5.94 | 5.82 | 5.88 | Under $36,000 | 6,715 |
| 49 | 5.95 | 5.83 | 5.89 | Under $36,000 | 6,802 |
| 50 | 5.96 | 5.86 | 5.91 | Under $36,000 | 9,698 |
| 51 | 5.97 | 5.86 | 5.92 | Under $36,000 | 7,178 |
| 52 | 5.93 | 5.83 | 5.88 | Under $36,000 | 8,699 |
| 53 | 5.87 | 5.77 | 5.82 | Under $36,000 | 8,438 |
| 54 | 5.90 | 5.80 | 5.85 | Under $36,000 | 8,402 |
| 55 | 5.93 | 5.83 | 5.88 | Under $36,000 | 9,412 |
| 56 | 5.87 | 5.77 | 5.82 | Under $36,000 | 8,579 |
| 57 | 5.95 | 5.84 | 5.89 | Under $36,000 | 9,010 |
| 58 | 5.96 | 5.86 | 5.91 | Under $36,000 | 8,977 |
| 59 | 6.05 | 5.94 | 5.99 | Under $36,000 | 8,763 |
| 60 | 6.10 | 6 | 6.05 | Under $36,000 | 10,465 |
| 61 | 6.20 | 6.09 | 6.14 | Under $36,000 | 8,360 |
| 62 | 6.28 | 6.19 | 6.23 | Under $36,000 | 10,210 |
| 63 | 6.31 | 6.22 | 6.26 | Under $36,000 | 9,070 |
| 64 | 6.43 | 6.34 | 6.39 | Under $36,000 | 8,947 |
| 65 | 6.53 | 6.44 | 6.48 | Under $36,000 | 10,348 |
| 66 | 6.59 | 6.50 | 6.54 | Under $36,000 | 8,602 |
| 67 | 6.62 | 6.53 | 6.57 | Under $36,000 | 8,867 |
| 68 | 6.65 | 6.56 | 6.60 | Under $36,000 | 8,512 |
| 69 | 6.73 | 6.64 | 6.68 | Under $36,000 | 8,076 |
| 70 | 6.78 | 6.70 | 6.74 | Under $36,000 | 9,664 |
| 71 | 6.81 | 6.71 | 6.76 | Under $36,000 | 7,672 |
| 72 | 6.84 | 6.74 | 6.79 | Under $36,000 | 8,342 |
| 73 | 6.85 | 6.75 | 6.80 | Under $36,000 | 7,729 |
| 74 | 6.91 | 6.81 | 6.86 | Under $36,000 | 7,424 |
| 75 | 6.94 | 6.85 | 6.89 | Under $36,000 | 8,139 |
| 76 | 6.93 | 6.83 | 6.88 | Under $36,000 | 6,945 |
| 77 | 7.00 | 6.90 | 6.95 | Under $36,000 | 6,834 |
| 78 | 7.03 | 6.93 | 6.98 | Under $36,000 | 6,649 |
| 79 | 7.01 | 6.90 | 6.96 | Under $36,000 | 6,142 |
| 80 | 7.11 | 7.01 | 7.06 | Under $36,000 | 7,443 |
| 81 | 7.11 | 7 | 7.05 | Under $36,000 | 5,617 |
| 82 | 7.14 | 7.02 | 7.08 | Under $36,000 | 5,792 |
| 83 | 7.10 | 6.98 | 7.04 | Under $36,000 | 5,125 |
| 84 | 7.18 | 7.06 | 7.12 | Under $36,000 | 4,628 |
| 85 | 7.18 | 7.05 | 7.12 | Under $36,000 | 4,351 |
| 86 | 7.28 | 7.13 | 7.20 | Under $36,000 | 3,386 |
| 87 | 7.29 | 7.12 | 7.21 | Under $36,000 | 2,842 |
| 88 | 7.26 | 7.09 | 7.18 | Under $36,000 | 2,474 |
| 89 | 7.31 | 7.10 | 7.21 | Under $36,000 | 1,867 |
| 90 | 7.37 | 7.16 | 7.26 | Under $36,000 | 1,616 |
| 91 | 7.61 | 7.34 | 7.47 | Under $36,000 | 981 |
| 92 | 7.46 | 7.13 | 7.29 | Under $36,000 | 755 |
| 93 | 7.53 | 7.15 | 7.34 | Under $36,000 | 521 |
| 94 | 7.70 | 7.21 | 7.45 | Under $36,000 | 310 |
| 18 | 7.38 | 7.29 | 7.34 | $36,000 to $89,999 | 4,315 |
| 19 | 7.25 | 7.14 | 7.19 | $36,000 to $89,999 | 3,545 |
| 20 | 7.22 | 7.12 | 7.17 | $36,000 to $89,999 | 3,432 |
| 21 | 7.22 | 7.11 | 7.16 | $36,000 to $89,999 | 3,725 |
| 22 | 7.21 | 7.11 | 7.16 | $36,000 to $89,999 | 3,852 |
| 23 | 7.16 | 7.07 | 7.12 | $36,000 to $89,999 | 4,534 |
| 24 | 7.12 | 7.03 | 7.07 | $36,000 to $89,999 | 4,926 |
| 25 | 7.20 | 7.11 | 7.16 | $36,000 to $89,999 | 5,718 |
| 26 | 7.18 | 7.10 | 7.14 | $36,000 to $89,999 | 5,657 |
| 27 | 7.22 | 7.14 | 7.18 | $36,000 to $89,999 | 6,142 |
| 28 | 7.21 | 7.13 | 7.17 | $36,000 to $89,999 | 6,743 |
| 29 | 7.19 | 7.12 | 7.15 | $36,000 to $89,999 | 6,811 |
| 30 | 7.17 | 7.10 | 7.14 | $36,000 to $89,999 | 7,970 |
| 31 | 7.18 | 7.11 | 7.14 | $36,000 to $89,999 | 6,785 |
| 32 | 7.14 | 7.07 | 7.10 | $36,000 to $89,999 | 7,577 |
| 33 | 7.15 | 7.08 | 7.11 | $36,000 to $89,999 | 7,028 |
| 34 | 7.13 | 7.05 | 7.09 | $36,000 to $89,999 | 7,113 |
| 35 | 7.13 | 7.06 | 7.09 | $36,000 to $89,999 | 7,780 |
| 36 | 7.09 | 7.00 | 7.05 | $36,000 to $89,999 | 7,073 |
| 37 | 7.05 | 6.97 | 7.01 | $36,000 to $89,999 | 7,115 |
| 38 | 7.05 | 6.98 | 7.02 | $36,000 to $89,999 | 7,564 |
| 39 | 7.05 | 6.98 | 7.02 | $36,000 to $89,999 | 7,347 |
| 40 | 7.01 | 6.94 | 6.98 | $36,000 to $89,999 | 9,294 |
| 41 | 7.02 | 6.95 | 6.99 | $36,000 to $89,999 | 7,481 |
| 42 | 6.95 | 6.88 | 6.91 | $36,000 to $89,999 | 8,922 |
| 43 | 6.92 | 6.84 | 6.88 | $36,000 to $89,999 | 8,202 |
| 44 | 6.93 | 6.85 | 6.89 | $36,000 to $89,999 | 7,958 |
| 45 | 6.96 | 6.88 | 6.92 | $36,000 to $89,999 | 9,379 |
| 46 | 6.90 | 6.83 | 6.86 | $36,000 to $89,999 | 8,579 |
| 47 | 6.87 | 6.80 | 6.84 | $36,000 to $89,999 | 9,507 |
| 48 | 6.88 | 6.81 | 6.85 | $36,000 to $89,999 | 9,611 |
| 49 | 6.86 | 6.79 | 6.82 | $36,000 to $89,999 | 9,826 |
| 50 | 6.90 | 6.83 | 6.87 | $36,000 to $89,999 | 12,927 |
| 51 | 6.92 | 6.84 | 6.88 | $36,000 to $89,999 | 9,986 |
| 52 | 6.87 | 6.80 | 6.83 | $36,000 to $89,999 | 12,058 |
| 53 | 6.90 | 6.83 | 6.87 | $36,000 to $89,999 | 11,152 |
| 54 | 6.93 | 6.86 | 6.89 | $36,000 to $89,999 | 11,437 |
| 55 | 6.91 | 6.84 | 6.88 | $36,000 to $89,999 | 12,738 |
| 56 | 6.94 | 6.87 | 6.90 | $36,000 to $89,999 | 12,024 |
| 57 | 6.97 | 6.90 | 6.93 | $36,000 to $89,999 | 12,157 |
| 58 | 6.98 | 6.91 | 6.95 | $36,000 to $89,999 | 12,474 |
| 59 | 7.04 | 6.98 | 7.01 | $36,000 to $89,999 | 12,419 |
| 60 | 7.12 | 7.06 | 7.09 | $36,000 to $89,999 | 14,550 |
| 61 | 7.16 | 7.09 | 7.13 | $36,000 to $89,999 | 11,886 |
| 62 | 7.19 | 7.13 | 7.16 | $36,000 to $89,999 | 14,832 |
| 63 | 7.24 | 7.17 | 7.21 | $36,000 to $89,999 | 13,321 |
| 64 | 7.29 | 7.23 | 7.26 | $36,000 to $89,999 | 13,534 |
| 65 | 7.33 | 7.27 | 7.30 | $36,000 to $89,999 | 14,666 |
| 66 | 7.41 | 7.34 | 7.38 | $36,000 to $89,999 | 11,987 |
| 67 | 7.43 | 7.37 | 7.40 | $36,000 to $89,999 | 11,910 |
| 68 | 7.44 | 7.37 | 7.40 | $36,000 to $89,999 | 10,820 |
| 69 | 7.48 | 7.41 | 7.44 | $36,000 to $89,999 | 9,902 |
| 70 | 7.50 | 7.43 | 7.46 | $36,000 to $89,999 | 11,374 |
| 71 | 7.54 | 7.46 | 7.50 | $36,000 to $89,999 | 8,453 |
| 72 | 7.54 | 7.46 | 7.50 | $36,000 to $89,999 | 8,691 |
| 73 | 7.55 | 7.47 | 7.51 | $36,000 to $89,999 | 7,438 |
| 74 | 7.58 | 7.49 | 7.53 | $36,000 to $89,999 | 6,774 |
| 75 | 7.55 | 7.47 | 7.51 | $36,000 to $89,999 | 7,325 |
| 76 | 7.56 | 7.47 | 7.52 | $36,000 to $89,999 | 5,726 |
| 77 | 7.57 | 7.48 | 7.53 | $36,000 to $89,999 | 5,546 |
| 78 | 7.61 | 7.51 | 7.56 | $36,000 to $89,999 | 5,173 |
| 79 | 7.67 | 7.57 | 7.62 | $36,000 to $89,999 | 4,559 |
| 80 | 7.65 | 7.55 | 7.60 | $36,000 to $89,999 | 5,116 |
| 81 | 7.66 | 7.54 | 7.60 | $36,000 to $89,999 | 3,742 |
| 82 | 7.64 | 7.51 | 7.58 | $36,000 to $89,999 | 3,591 |
| 83 | 7.62 | 7.48 | 7.55 | $36,000 to $89,999 | 3,174 |
| 84 | 7.66 | 7.52 | 7.59 | $36,000 to $89,999 | 2,650 |
| 85 | 7.57 | 7.42 | 7.50 | $36,000 to $89,999 | 2,556 |
| 86 | 7.61 | 7.44 | 7.52 | $36,000 to $89,999 | 1,838 |
| 87 | 7.64 | 7.44 | 7.54 | $36,000 to $89,999 | 1,561 |
| 88 | 7.57 | 7.37 | 7.47 | $36,000 to $89,999 | 1,303 |
| 89 | 7.78 | 7.52 | 7.65 | $36,000 to $89,999 | 890 |
| 90 | 7.79 | 7.49 | 7.64 | $36,000 to $89,999 | 773 |
| 91 | 7.92 | 7.62 | 7.77 | $36,000 to $89,999 | 505 |
| 92 | 7.81 | 7.43 | 7.62 | $36,000 to $89,999 | 397 |
| 93 | 7.81 | 7.30 | 7.56 | $36,000 to $89,999 | 226 |
| 94 | 7.78 | 7.20 | 7.49 | $36,000 to $89,999 | 198 |
| 18 | 7.58 | 7.47 | 7.52 | $90,000 or more | 2,731 |
| 19 | 7.61 | 7.47 | 7.54 | $90,000 or more | 2,035 |
| 20 | 7.45 | 7.31 | 7.38 | $90,000 or more | 1,825 |
| 21 | 7.46 | 7.31 | 7.38 | $90,000 or more | 1,817 |
| 22 | 7.31 | 7.16 | 7.24 | $90,000 or more | 1,638 |
| 23 | 7.24 | 7.08 | 7.16 | $90,000 or more | 1,635 |
| 24 | 7.27 | 7.11 | 7.19 | $90,000 or more | 1,547 |
| 25 | 7.30 | 7.15 | 7.22 | $90,000 or more | 1,718 |
| 26 | 7.41 | 7.27 | 7.34 | $90,000 or more | 1,673 |
| 27 | 7.47 | 7.34 | 7.41 | $90,000 or more | 2,028 |
| 28 | 7.55 | 7.44 | 7.50 | $90,000 or more | 2,337 |
| 29 | 7.57 | 7.45 | 7.51 | $90,000 or more | 2,627 |
| 30 | 7.60 | 7.49 | 7.54 | $90,000 or more | 3,223 |
| 31 | 7.59 | 7.50 | 7.54 | $90,000 or more | 3,225 |
| 32 | 7.58 | 7.49 | 7.54 | $90,000 or more | 3,704 |
| 33 | 7.59 | 7.50 | 7.55 | $90,000 or more | 3,774 |
| 34 | 7.58 | 7.50 | 7.54 | $90,000 or more | 4,031 |
| 35 | 7.58 | 7.50 | 7.54 | $90,000 or more | 4,750 |
| 36 | 7.61 | 7.53 | 7.57 | $90,000 or more | 4,481 |
| 37 | 7.60 | 7.53 | 7.57 | $90,000 or more | 4,744 |
| 38 | 7.56 | 7.49 | 7.53 | $90,000 or more | 5,116 |
| 39 | 7.57 | 7.50 | 7.53 | $90,000 or more | 5,415 |
| 40 | 7.56 | 7.48 | 7.52 | $90,000 or more | 6,509 |
| 41 | 7.55 | 7.47 | 7.51 | $90,000 or more | 5,541 |
| 42 | 7.52 | 7.46 | 7.49 | $90,000 or more | 6,799 |
| 43 | 7.48 | 7.41 | 7.45 | $90,000 or more | 6,276 |
| 44 | 7.50 | 7.43 | 7.46 | $90,000 or more | 6,156 |
| 45 | 7.50 | 7.43 | 7.46 | $90,000 or more | 7,061 |
| 46 | 7.46 | 7.39 | 7.43 | $90,000 or more | 6,785 |
| 47 | 7.47 | 7.39 | 7.43 | $90,000 or more | 7,090 |
| 48 | 7.45 | 7.39 | 7.42 | $90,000 or more | 7,421 |
| 49 | 7.47 | 7.40 | 7.44 | $90,000 or more | 7,424 |
| 50 | 7.46 | 7.40 | 7.43 | $90,000 or more | 9,246 |
| 51 | 7.49 | 7.42 | 7.45 | $90,000 or more | 7,276 |
| 52 | 7.51 | 7.45 | 7.48 | $90,000 or more | 8,649 |
| 53 | 7.48 | 7.41 | 7.45 | $90,000 or more | 8,237 |
| 54 | 7.50 | 7.43 | 7.47 | $90,000 or more | 8,045 |
| 55 | 7.50 | 7.43 | 7.46 | $90,000 or more | 8,899 |
| 56 | 7.49 | 7.43 | 7.46 | $90,000 or more | 7,975 |
| 57 | 7.52 | 7.46 | 7.49 | $90,000 or more | 8,058 |
| 58 | 7.55 | 7.48 | 7.51 | $90,000 or more | 8,057 |
| 59 | 7.58 | 7.52 | 7.55 | $90,000 or more | 7,770 |
| 60 | 7.62 | 7.55 | 7.58 | $90,000 or more | 8,768 |
| 61 | 7.68 | 7.61 | 7.65 | $90,000 or more | 6,864 |
| 62 | 7.71 | 7.65 | 7.68 | $90,000 or more | 7,960 |
| 63 | 7.72 | 7.65 | 7.68 | $90,000 or more | 6,937 |
| 64 | 7.76 | 7.69 | 7.73 | $90,000 or more | 6,546 |
| 65 | 7.83 | 7.75 | 7.79 | $90,000 or more | 6,757 |
| 66 | 7.85 | 7.77 | 7.81 | $90,000 or more | 5,407 |
| 67 | 7.88 | 7.79 | 7.83 | $90,000 or more | 5,156 |
| 68 | 7.95 | 7.86 | 7.90 | $90,000 or more | 4,582 |
| 69 | 7.93 | 7.83 | 7.88 | $90,000 or more | 3,858 |
| 70 | 7.95 | 7.85 | 7.90 | $90,000 or more | 4,316 |
| 71 | 7.98 | 7.86 | 7.92 | $90,000 or more | 3,016 |
| 72 | 7.95 | 7.83 | 7.89 | $90,000 or more | 3,036 |
| 73 | 7.91 | 7.78 | 7.84 | $90,000 or more | 2,624 |
| 74 | 8.02 | 7.88 | 7.95 | $90,000 or more | 2,227 |
| 75 | 7.94 | 7.79 | 7.87 | $90,000 or more | 2,316 |
| 76 | 7.94 | 7.78 | 7.86 | $90,000 or more | 1,798 |
| 77 | 7.98 | 7.81 | 7.90 | $90,000 or more | 1,747 |
| 78 | 7.87 | 7.69 | 7.78 | $90,000 or more | 1,550 |
| 79 | 7.87 | 7.67 | 7.77 | $90,000 or more | 1,297 |
| 80 | 7.93 | 7.75 | 7.84 | $90,000 or more | 1,568 |
| 81 | 8 | 7.80 | 7.90 | $90,000 or more | 1,108 |
| 82 | 7.93 | 7.73 | 7.83 | $90,000 or more | 1,101 |
| 83 | 7.79 | 7.54 | 7.67 | $90,000 or more | 892 |
| 84 | 7.88 | 7.64 | 7.76 | $90,000 or more | 756 |
| 85 | 7.95 | 7.68 | 7.82 | $90,000 or more | 712 |
| 86 | 7.91 | 7.58 | 7.75 | $90,000 or more | 518 |
| 87 | 7.82 | 7.46 | 7.64 | $90,000 or more | 449 |
| 88 | 7.98 | 7.59 | 7.79 | $90,000 or more | 402 |
| 89 | 7.89 | 7.37 | 7.63 | $90,000 or more | 275 |
| 90 | 8.02 | 7.48 | 7.75 | $90,000 or more | 253 |
| 91 | 7.75 | 7.14 | 7.45 | $90,000 or more | 166 |
| 92 | 8.13 | 7.43 | 7.78 | $90,000 or more | 129 |
| 93 | 8.16 | 7.23 | 7.70 | $90,000 or more | 69 |
| 94 | 8.27 | 7.22 | 7.75 | $90,000 or more | 51 |

#### Table 4.2. Enjoyment

Enjoyment is lowest in midlife in every income group, at 72.7% in households under $36,000 against 87.6% in households of $90,000 or more

Share of adults who reported enjoyment during a lot of the day yesterday by year of age and household income group, Gallup 2009 to 2014

| age | conf_high | conf_low | estimate | group | n |
| ---: | ---: | ---: | ---: | --- | ---: |
| 18 | 89.94 | 88.31 | 89.12 | Under $36,000 | 5,802 |
| 19 | 88.85 | 87.22 | 88.04 | Under $36,000 | 5,976 |
| 20 | 87.51 | 85.90 | 86.70 | Under $36,000 | 6,724 |
| 21 | 87.02 | 85.36 | 86.19 | Under $36,000 | 7,362 |
| 22 | 86.66 | 84.97 | 85.82 | Under $36,000 | 7,001 |
| 23 | 86.04 | 84.36 | 85.20 | Under $36,000 | 6,919 |
| 24 | 86.08 | 84.33 | 85.21 | Under $36,000 | 6,693 |
| 25 | 84.95 | 83.27 | 84.11 | Under $36,000 | 7,056 |
| 26 | 84.31 | 82.09 | 83.20 | Under $36,000 | 6,054 |
| 27 | 82.79 | 80.81 | 81.80 | Under $36,000 | 5,918 |
| 28 | 83.11 | 81.25 | 82.18 | Under $36,000 | 6,218 |
| 29 | 80.60 | 78.51 | 79.56 | Under $36,000 | 5,523 |
| 30 | 81.67 | 79.72 | 80.70 | Under $36,000 | 6,672 |
| 31 | 80.74 | 78.41 | 79.58 | Under $36,000 | 4,725 |
| 32 | 81.18 | 79.13 | 80.15 | Under $36,000 | 5,694 |
| 33 | 80.86 | 78.63 | 79.74 | Under $36,000 | 4,749 |
| 34 | 79.80 | 77.39 | 78.59 | Under $36,000 | 4,681 |
| 35 | 79.75 | 77.39 | 78.57 | Under $36,000 | 5,521 |
| 36 | 78.99 | 76.47 | 77.73 | Under $36,000 | 4,549 |
| 37 | 77.78 | 75.40 | 76.59 | Under $36,000 | 4,639 |
| 38 | 77.88 | 75.51 | 76.69 | Under $36,000 | 4,938 |
| 39 | 76.76 | 74.32 | 75.54 | Under $36,000 | 4,608 |
| 40 | 77.38 | 75.26 | 76.32 | Under $36,000 | 6,609 |
| 41 | 77.34 | 74.88 | 76.11 | Under $36,000 | 4,446 |
| 42 | 76.02 | 73.78 | 74.90 | Under $36,000 | 5,725 |
| 43 | 76.57 | 74.27 | 75.42 | Under $36,000 | 5,366 |
| 44 | 75.62 | 73.25 | 74.44 | Under $36,000 | 5,187 |
| 45 | 75.99 | 73.82 | 74.90 | Under $36,000 | 6,519 |
| 46 | 74.30 | 72.18 | 73.24 | Under $36,000 | 6,043 |
| 47 | 74.53 | 72.39 | 73.46 | Under $36,000 | 6,696 |
| 48 | 74.20 | 72.18 | 73.19 | Under $36,000 | 6,948 |
| 49 | 73.55 | 71.50 | 72.52 | Under $36,000 | 7,031 |
| 50 | 73.36 | 71.54 | 72.45 | Under $36,000 | 9,984 |
| 51 | 73.68 | 71.76 | 72.72 | Under $36,000 | 7,415 |
| 52 | 73.59 | 71.75 | 72.67 | Under $36,000 | 8,965 |
| 53 | 72.65 | 70.65 | 71.65 | Under $36,000 | 8,695 |
| 54 | 71.97 | 70.03 | 71 | Under $36,000 | 8,678 |
| 55 | 73.80 | 72.07 | 72.93 | Under $36,000 | 9,675 |
| 56 | 73.71 | 71.84 | 72.78 | Under $36,000 | 8,853 |
| 57 | 74.53 | 72.68 | 73.61 | Under $36,000 | 9,256 |
| 58 | 73.47 | 71.56 | 72.52 | Under $36,000 | 9,264 |
| 59 | 75.54 | 73.72 | 74.63 | Under $36,000 | 9,042 |
| 60 | 76.55 | 74.82 | 75.69 | Under $36,000 | 10,805 |
| 61 | 77.53 | 75.72 | 76.63 | Under $36,000 | 8,621 |
| 62 | 77.79 | 76.17 | 76.98 | Under $36,000 | 10,561 |
| 63 | 79.08 | 77.32 | 78.20 | Under $36,000 | 9,322 |
| 64 | 80.21 | 78.50 | 79.35 | Under $36,000 | 9,232 |
| 65 | 81.03 | 79.45 | 80.24 | Under $36,000 | 10,637 |
| 66 | 82.72 | 81.08 | 81.90 | Under $36,000 | 8,852 |
| 67 | 82.14 | 80.56 | 81.35 | Under $36,000 | 9,125 |
| 68 | 83.78 | 82.09 | 82.94 | Under $36,000 | 8,767 |
| 69 | 84.82 | 83.14 | 83.98 | Under $36,000 | 8,307 |
| 70 | 85.37 | 83.86 | 84.62 | Under $36,000 | 9,958 |
| 71 | 85.75 | 84.07 | 84.91 | Under $36,000 | 7,925 |
| 72 | 85.74 | 84.15 | 84.95 | Under $36,000 | 8,616 |
| 73 | 85.83 | 84.25 | 85.04 | Under $36,000 | 8,046 |
| 74 | 86.11 | 84.52 | 85.32 | Under $36,000 | 7,661 |
| 75 | 86.39 | 84.86 | 85.62 | Under $36,000 | 8,396 |
| 76 | 86.39 | 84.76 | 85.57 | Under $36,000 | 7,188 |
| 77 | 86.65 | 85.04 | 85.85 | Under $36,000 | 7,072 |
| 78 | 86.45 | 84.74 | 85.59 | Under $36,000 | 6,907 |
| 79 | 86.43 | 84.75 | 85.59 | Under $36,000 | 6,342 |
| 80 | 87.02 | 85.41 | 86.22 | Under $36,000 | 7,734 |
| 81 | 87.56 | 85.84 | 86.70 | Under $36,000 | 5,799 |
| 82 | 87.68 | 86.01 | 86.85 | Under $36,000 | 5,991 |
| 83 | 86.51 | 84.58 | 85.55 | Under $36,000 | 5,341 |
| 84 | 87.03 | 84.97 | 86 | Under $36,000 | 4,772 |
| 85 | 86.53 | 84.54 | 85.54 | Under $36,000 | 4,501 |
| 86 | 86.89 | 84.56 | 85.72 | Under $36,000 | 3,523 |
| 87 | 87.91 | 85.20 | 86.56 | Under $36,000 | 2,938 |
| 88 | 85.92 | 83.10 | 84.51 | Under $36,000 | 2,602 |
| 89 | 86.23 | 83.07 | 84.65 | Under $36,000 | 1,948 |
| 90 | 87.14 | 83.93 | 85.54 | Under $36,000 | 1,687 |
| 91 | 89.57 | 85.39 | 87.48 | Under $36,000 | 1,030 |
| 92 | 90.19 | 85.94 | 88.07 | Under $36,000 | 796 |
| 93 | 87.72 | 81.60 | 84.66 | Under $36,000 | 541 |
| 94 | 89.18 | 81.73 | 85.45 | Under $36,000 | 330 |
| 18 | 92.70 | 90.90 | 91.80 | $36,000 to $89,999 | 4,365 |
| 19 | 92.59 | 90.88 | 91.73 | $36,000 to $89,999 | 3,581 |
| 20 | 91.74 | 89.77 | 90.76 | $36,000 to $89,999 | 3,494 |
| 21 | 91.20 | 89.36 | 90.28 | $36,000 to $89,999 | 3,785 |
| 22 | 91.30 | 89.47 | 90.39 | $36,000 to $89,999 | 3,921 |
| 23 | 90.30 | 88.44 | 89.37 | $36,000 to $89,999 | 4,619 |
| 24 | 89.57 | 87.83 | 88.70 | $36,000 to $89,999 | 5,044 |
| 25 | 90.40 | 88.84 | 89.62 | $36,000 to $89,999 | 5,859 |
| 26 | 89.57 | 87.91 | 88.74 | $36,000 to $89,999 | 5,798 |
| 27 | 89.28 | 87.71 | 88.50 | $36,000 to $89,999 | 6,293 |
| 28 | 89.52 | 87.97 | 88.74 | $36,000 to $89,999 | 6,930 |
| 29 | 89.09 | 87.53 | 88.31 | $36,000 to $89,999 | 7,015 |
| 30 | 89.08 | 87.68 | 88.38 | $36,000 to $89,999 | 8,193 |
| 31 | 89.66 | 88.14 | 88.90 | $36,000 to $89,999 | 7,009 |
| 32 | 88.24 | 86.52 | 87.38 | $36,000 to $89,999 | 7,806 |
| 33 | 88.10 | 86.58 | 87.34 | $36,000 to $89,999 | 7,251 |
| 34 | 88.10 | 86.61 | 87.35 | $36,000 to $89,999 | 7,322 |
| 35 | 87.19 | 85.63 | 86.41 | $36,000 to $89,999 | 8,049 |
| 36 | 87.43 | 85.82 | 86.63 | $36,000 to $89,999 | 7,314 |
| 37 | 87.01 | 85.32 | 86.16 | $36,000 to $89,999 | 7,364 |
| 38 | 86.28 | 84.70 | 85.49 | $36,000 to $89,999 | 7,836 |
| 39 | 86.89 | 85.31 | 86.10 | $36,000 to $89,999 | 7,597 |
| 40 | 85.66 | 84.26 | 84.96 | $36,000 to $89,999 | 9,581 |
| 41 | 85.67 | 84.08 | 84.87 | $36,000 to $89,999 | 7,715 |
| 42 | 85.25 | 83.76 | 84.51 | $36,000 to $89,999 | 9,230 |
| 43 | 86.06 | 84.51 | 85.29 | $36,000 to $89,999 | 8,510 |
| 44 | 85.04 | 83.49 | 84.26 | $36,000 to $89,999 | 8,217 |
| 45 | 85.59 | 84.12 | 84.86 | $36,000 to $89,999 | 9,667 |
| 46 | 85.28 | 83.83 | 84.55 | $36,000 to $89,999 | 8,875 |
| 47 | 84.93 | 83.48 | 84.20 | $36,000 to $89,999 | 9,838 |
| 48 | 84.32 | 82.88 | 83.60 | $36,000 to $89,999 | 9,884 |
| 49 | 85.22 | 83.66 | 84.44 | $36,000 to $89,999 | 10,168 |
| 50 | 84.68 | 83.24 | 83.96 | $36,000 to $89,999 | 13,286 |
| 51 | 85.02 | 83.67 | 84.34 | $36,000 to $89,999 | 10,333 |
| 52 | 85.26 | 83.97 | 84.62 | $36,000 to $89,999 | 12,402 |
| 53 | 84.49 | 83.15 | 83.82 | $36,000 to $89,999 | 11,519 |
| 54 | 85.37 | 83.98 | 84.68 | $36,000 to $89,999 | 11,825 |
| 55 | 85.55 | 84.37 | 84.96 | $36,000 to $89,999 | 13,123 |
| 56 | 85.42 | 84.13 | 84.77 | $36,000 to $89,999 | 12,394 |
| 57 | 86.55 | 85.29 | 85.92 | $36,000 to $89,999 | 12,486 |
| 58 | 86.17 | 84.96 | 85.57 | $36,000 to $89,999 | 12,867 |
| 59 | 86.91 | 85.72 | 86.31 | $36,000 to $89,999 | 12,815 |
| 60 | 87.15 | 86.06 | 86.60 | $36,000 to $89,999 | 15,018 |
| 61 | 88.06 | 86.91 | 87.49 | $36,000 to $89,999 | 12,242 |
| 62 | 88.99 | 87.92 | 88.45 | $36,000 to $89,999 | 15,287 |
| 63 | 88.95 | 87.89 | 88.42 | $36,000 to $89,999 | 13,609 |
| 64 | 90.59 | 89.58 | 90.08 | $36,000 to $89,999 | 13,874 |
| 65 | 91.04 | 90.13 | 90.59 | $36,000 to $89,999 | 15,032 |
| 66 | 91.37 | 90.33 | 90.85 | $36,000 to $89,999 | 12,305 |
| 67 | 91.87 | 90.88 | 91.37 | $36,000 to $89,999 | 12,218 |
| 68 | 92.13 | 91.09 | 91.61 | $36,000 to $89,999 | 11,112 |
| 69 | 93 | 91.94 | 92.47 | $36,000 to $89,999 | 10,146 |
| 70 | 92.82 | 91.84 | 92.33 | $36,000 to $89,999 | 11,664 |
| 71 | 92.71 | 91.55 | 92.13 | $36,000 to $89,999 | 8,654 |
| 72 | 92.75 | 91.60 | 92.17 | $36,000 to $89,999 | 8,905 |
| 73 | 92.79 | 91.54 | 92.16 | $36,000 to $89,999 | 7,615 |
| 74 | 93.32 | 92.03 | 92.68 | $36,000 to $89,999 | 6,936 |
| 75 | 92.66 | 91.46 | 92.06 | $36,000 to $89,999 | 7,504 |
| 76 | 92.66 | 91.19 | 91.92 | $36,000 to $89,999 | 5,869 |
| 77 | 92.65 | 91.29 | 91.97 | $36,000 to $89,999 | 5,705 |
| 78 | 92.50 | 91.01 | 91.76 | $36,000 to $89,999 | 5,313 |
| 79 | 92.78 | 91.23 | 92.01 | $36,000 to $89,999 | 4,666 |
| 80 | 92.57 | 91.09 | 91.83 | $36,000 to $89,999 | 5,263 |
| 81 | 91.80 | 89.88 | 90.84 | $36,000 to $89,999 | 3,820 |
| 82 | 91.15 | 89.17 | 90.16 | $36,000 to $89,999 | 3,678 |
| 83 | 90.80 | 88.73 | 89.76 | $36,000 to $89,999 | 3,263 |
| 84 | 90.84 | 88.66 | 89.75 | $36,000 to $89,999 | 2,722 |
| 85 | 91.37 | 88.98 | 90.17 | $36,000 to $89,999 | 2,635 |
| 86 | 90.77 | 88.08 | 89.43 | $36,000 to $89,999 | 1,882 |
| 87 | 90.70 | 87.75 | 89.22 | $36,000 to $89,999 | 1,596 |
| 88 | 90.86 | 87.61 | 89.24 | $36,000 to $89,999 | 1,338 |
| 89 | 92.62 | 88.62 | 90.62 | $36,000 to $89,999 | 906 |
| 90 | 91.85 | 87.85 | 89.85 | $36,000 to $89,999 | 798 |
| 91 | 91.12 | 85.50 | 88.31 | $36,000 to $89,999 | 522 |
| 92 | 91.73 | 85.72 | 88.73 | $36,000 to $89,999 | 408 |
| 93 | 88.20 | 78.75 | 83.47 | $36,000 to $89,999 | 236 |
| 94 | 88.68 | 78.48 | 83.58 | $36,000 to $89,999 | 201 |
| 18 | 94.90 | 93.21 | 94.05 | $90,000 or more | 2,775 |
| 19 | 93.84 | 91.55 | 92.69 | $90,000 or more | 2,066 |
| 20 | 93.04 | 90.62 | 91.83 | $90,000 or more | 1,848 |
| 21 | 92.67 | 90.04 | 91.36 | $90,000 or more | 1,851 |
| 22 | 92.60 | 89.90 | 91.25 | $90,000 or more | 1,669 |
| 23 | 91.32 | 88.45 | 89.89 | $90,000 or more | 1,671 |
| 24 | 91.28 | 88.50 | 89.89 | $90,000 or more | 1,573 |
| 25 | 91.74 | 89.05 | 90.40 | $90,000 or more | 1,760 |
| 26 | 91.88 | 89.08 | 90.48 | $90,000 or more | 1,712 |
| 27 | 92.66 | 89.91 | 91.29 | $90,000 or more | 2,077 |
| 28 | 91.16 | 88.63 | 89.89 | $90,000 or more | 2,394 |
| 29 | 91.02 | 88.78 | 89.90 | $90,000 or more | 2,703 |
| 30 | 90.78 | 88.76 | 89.77 | $90,000 or more | 3,303 |
| 31 | 89.87 | 87.76 | 88.82 | $90,000 or more | 3,317 |
| 32 | 89.95 | 88.05 | 89 | $90,000 or more | 3,828 |
| 33 | 90.34 | 88.40 | 89.37 | $90,000 or more | 3,895 |
| 34 | 90.27 | 88.42 | 89.35 | $90,000 or more | 4,140 |
| 35 | 89.56 | 87.81 | 88.68 | $90,000 or more | 4,913 |
| 36 | 90 | 88.27 | 89.14 | $90,000 or more | 4,630 |
| 37 | 90.25 | 88.41 | 89.33 | $90,000 or more | 4,930 |
| 38 | 89.40 | 87.70 | 88.55 | $90,000 or more | 5,310 |
| 39 | 89.22 | 87.51 | 88.37 | $90,000 or more | 5,570 |
| 40 | 89.56 | 88.09 | 88.83 | $90,000 or more | 6,722 |
| 41 | 89.04 | 87.42 | 88.23 | $90,000 or more | 5,701 |
| 42 | 88.25 | 86.64 | 87.44 | $90,000 or more | 7,017 |
| 43 | 88.37 | 86.68 | 87.52 | $90,000 or more | 6,484 |
| 44 | 88.18 | 86.52 | 87.35 | $90,000 or more | 6,362 |
| 45 | 88.77 | 87.22 | 88 | $90,000 or more | 7,297 |
| 46 | 88.65 | 87.12 | 87.89 | $90,000 or more | 7,017 |
| 47 | 88.42 | 86.96 | 87.69 | $90,000 or more | 7,296 |
| 48 | 87.35 | 85.88 | 86.62 | $90,000 or more | 7,674 |
| 49 | 88.52 | 86.90 | 87.71 | $90,000 or more | 7,666 |
| 50 | 88.53 | 87.19 | 87.86 | $90,000 or more | 9,556 |
| 51 | 88.28 | 86.71 | 87.49 | $90,000 or more | 7,500 |
| 52 | 88.43 | 87.02 | 87.72 | $90,000 or more | 8,918 |
| 53 | 88.08 | 86.64 | 87.36 | $90,000 or more | 8,498 |
| 54 | 88.62 | 86.99 | 87.81 | $90,000 or more | 8,285 |
| 55 | 88.83 | 87.55 | 88.19 | $90,000 or more | 9,154 |
| 56 | 88.05 | 86.48 | 87.26 | $90,000 or more | 8,204 |
| 57 | 89.43 | 88.10 | 88.76 | $90,000 or more | 8,304 |
| 58 | 88.95 | 87.58 | 88.26 | $90,000 or more | 8,317 |
| 59 | 89.69 | 88.31 | 89 | $90,000 or more | 7,990 |
| 60 | 90.43 | 89.19 | 89.81 | $90,000 or more | 9,008 |
| 61 | 90.68 | 89.26 | 89.97 | $90,000 or more | 7,079 |
| 62 | 91.35 | 89.96 | 90.65 | $90,000 or more | 8,185 |
| 63 | 91.67 | 90.23 | 90.95 | $90,000 or more | 7,086 |
| 64 | 92.47 | 91.08 | 91.78 | $90,000 or more | 6,712 |
| 65 | 92.86 | 91.58 | 92.22 | $90,000 or more | 6,913 |
| 66 | 93.73 | 92.38 | 93.05 | $90,000 or more | 5,528 |
| 67 | 94.03 | 92.63 | 93.33 | $90,000 or more | 5,276 |
| 68 | 94.14 | 92.71 | 93.42 | $90,000 or more | 4,697 |
| 69 | 94.79 | 93.36 | 94.08 | $90,000 or more | 3,951 |
| 70 | 94.16 | 92.64 | 93.40 | $90,000 or more | 4,423 |
| 71 | 94.84 | 93.18 | 94.01 | $90,000 or more | 3,104 |
| 72 | 93.98 | 92.13 | 93.05 | $90,000 or more | 3,095 |
| 73 | 95.24 | 93.58 | 94.41 | $90,000 or more | 2,683 |
| 74 | 94.91 | 92.91 | 93.91 | $90,000 or more | 2,299 |
| 75 | 94.50 | 92.57 | 93.53 | $90,000 or more | 2,366 |
| 76 | 94.83 | 92.69 | 93.76 | $90,000 or more | 1,842 |
| 77 | 94.38 | 91.96 | 93.17 | $90,000 or more | 1,800 |
| 78 | 92.80 | 90 | 91.40 | $90,000 or more | 1,593 |
| 79 | 92.96 | 89.91 | 91.44 | $90,000 or more | 1,331 |
| 80 | 92.64 | 89.81 | 91.23 | $90,000 or more | 1,607 |
| 81 | 93.07 | 89.81 | 91.44 | $90,000 or more | 1,133 |
| 82 | 92.85 | 89.65 | 91.25 | $90,000 or more | 1,132 |
| 83 | 91.81 | 87.70 | 89.76 | $90,000 or more | 908 |
| 84 | 92.67 | 88.57 | 90.62 | $90,000 or more | 789 |
| 85 | 93.44 | 89.17 | 91.30 | $90,000 or more | 736 |
| 86 | 91.85 | 86.43 | 89.14 | $90,000 or more | 534 |
| 87 | 91.88 | 86.38 | 89.13 | $90,000 or more | 460 |
| 88 | 90.11 | 83.48 | 86.80 | $90,000 or more | 409 |
| 89 | 92.93 | 86.17 | 89.55 | $90,000 or more | 287 |
| 90 | 92.42 | 85.28 | 88.85 | $90,000 or more | 260 |
| 91 | 93.68 | 83.20 | 88.44 | $90,000 or more | 173 |
| 92 | 94.09 | 83.52 | 88.81 | $90,000 or more | 134 |
| 93 | 97.06 | 83.23 | 90.14 | $90,000 or more | 71 |
| 94 | 97.07 | 80.29 | 88.68 | $90,000 or more | 53 |

#### Table 4.3. Stress

Stress is highest in the lowest income group in every age band and falls after midlife in all three groups, from 52.6% at ages 45 to 54 to 20.2% at ages 75 to 94 in households under $36,000

Share of adults who reported stress during a lot of the day yesterday by year of age and household income group, Gallup 2009 to 2014

| age | conf_high | conf_low | estimate | group | n |
| ---: | ---: | ---: | ---: | --- | ---: |
| 18 | 45.12 | 42.36 | 43.74 | Under $36,000 | 5,805 |
| 19 | 48.65 | 45.54 | 47.10 | Under $36,000 | 5,973 |
| 20 | 51.15 | 48.68 | 49.91 | Under $36,000 | 6,728 |
| 21 | 51.07 | 48.66 | 49.86 | Under $36,000 | 7,364 |
| 22 | 52.26 | 49.64 | 50.95 | Under $36,000 | 7,001 |
| 23 | 51.48 | 49.03 | 50.25 | Under $36,000 | 6,919 |
| 24 | 52.67 | 50.25 | 51.46 | Under $36,000 | 6,706 |
| 25 | 51.36 | 48.68 | 50.02 | Under $36,000 | 7,063 |
| 26 | 53.49 | 50.69 | 52.09 | Under $36,000 | 6,053 |
| 27 | 54.73 | 51.95 | 53.34 | Under $36,000 | 5,924 |
| 28 | 52.91 | 50.39 | 51.65 | Under $36,000 | 6,223 |
| 29 | 52.38 | 49.57 | 50.98 | Under $36,000 | 5,528 |
| 30 | 50.17 | 47.45 | 48.81 | Under $36,000 | 6,681 |
| 31 | 55.29 | 52.30 | 53.79 | Under $36,000 | 4,731 |
| 32 | 53.23 | 50.35 | 51.79 | Under $36,000 | 5,700 |
| 33 | 53.24 | 50 | 51.62 | Under $36,000 | 4,752 |
| 34 | 52.36 | 49.20 | 50.78 | Under $36,000 | 4,691 |
| 35 | 52.14 | 49.08 | 50.61 | Under $36,000 | 5,529 |
| 36 | 54.12 | 50.69 | 52.40 | Under $36,000 | 4,555 |
| 37 | 54.65 | 51.47 | 53.06 | Under $36,000 | 4,642 |
| 38 | 53.78 | 50.64 | 52.21 | Under $36,000 | 4,938 |
| 39 | 54.50 | 51.51 | 53 | Under $36,000 | 4,611 |
| 40 | 52.99 | 49.73 | 51.36 | Under $36,000 | 6,618 |
| 41 | 55.71 | 52.75 | 54.23 | Under $36,000 | 4,455 |
| 42 | 54.61 | 51.90 | 53.25 | Under $36,000 | 5,733 |
| 43 | 54.33 | 51.56 | 52.94 | Under $36,000 | 5,372 |
| 44 | 54.47 | 51.49 | 52.98 | Under $36,000 | 5,189 |
| 45 | 53.27 | 50.33 | 51.80 | Under $36,000 | 6,527 |
| 46 | 55.18 | 52.57 | 53.88 | Under $36,000 | 6,045 |
| 47 | 54.12 | 51.59 | 52.86 | Under $36,000 | 6,707 |
| 48 | 54.03 | 51.57 | 52.80 | Under $36,000 | 6,953 |
| 49 | 54.42 | 52.06 | 53.24 | Under $36,000 | 7,042 |
| 50 | 53.07 | 50.81 | 51.94 | Under $36,000 | 10,011 |
| 51 | 53.93 | 51.52 | 52.73 | Under $36,000 | 7,427 |
| 52 | 52.93 | 50.67 | 51.80 | Under $36,000 | 8,981 |
| 53 | 55.09 | 52.74 | 53.91 | Under $36,000 | 8,714 |
| 54 | 52.99 | 50.89 | 51.94 | Under $36,000 | 8,689 |
| 55 | 51.01 | 48.87 | 49.94 | Under $36,000 | 9,693 |
| 56 | 52.04 | 49.90 | 50.97 | Under $36,000 | 8,864 |
| 57 | 50.45 | 48.40 | 49.42 | Under $36,000 | 9,277 |
| 58 | 49.84 | 47.68 | 48.76 | Under $36,000 | 9,288 |
| 59 | 48.38 | 46.25 | 47.31 | Under $36,000 | 9,063 |
| 60 | 45.55 | 43.62 | 44.58 | Under $36,000 | 10,818 |
| 61 | 44.79 | 42.67 | 43.73 | Under $36,000 | 8,640 |
| 62 | 41.42 | 39.51 | 40.46 | Under $36,000 | 10,575 |
| 63 | 40.55 | 38.58 | 39.57 | Under $36,000 | 9,326 |
| 64 | 39.20 | 37.33 | 38.26 | Under $36,000 | 9,254 |
| 65 | 34.45 | 32.50 | 33.48 | Under $36,000 | 10,658 |
| 66 | 33.92 | 31.90 | 32.91 | Under $36,000 | 8,864 |
| 67 | 32.36 | 30.44 | 31.40 | Under $36,000 | 9,134 |
| 68 | 31.19 | 29.30 | 30.24 | Under $36,000 | 8,785 |
| 69 | 29.65 | 27.76 | 28.70 | Under $36,000 | 8,313 |
| 70 | 28.37 | 26.62 | 27.50 | Under $36,000 | 9,983 |
| 71 | 27.34 | 25.33 | 26.33 | Under $36,000 | 7,929 |
| 72 | 26.76 | 24.90 | 25.83 | Under $36,000 | 8,641 |
| 73 | 25.98 | 24.09 | 25.04 | Under $36,000 | 8,056 |
| 74 | 25.45 | 23.48 | 24.47 | Under $36,000 | 7,672 |
| 75 | 24.57 | 22.78 | 23.67 | Under $36,000 | 8,431 |
| 76 | 24.04 | 22.05 | 23.04 | Under $36,000 | 7,200 |
| 77 | 24.40 | 22.34 | 23.37 | Under $36,000 | 7,086 |
| 78 | 22.54 | 20.66 | 21.60 | Under $36,000 | 6,934 |
| 79 | 22.78 | 20.77 | 21.77 | Under $36,000 | 6,361 |
| 80 | 20.92 | 19.25 | 20.08 | Under $36,000 | 7,738 |
| 81 | 21.15 | 19.12 | 20.14 | Under $36,000 | 5,810 |
| 82 | 19.62 | 17.67 | 18.64 | Under $36,000 | 6,007 |
| 83 | 21.44 | 19.26 | 20.35 | Under $36,000 | 5,351 |
| 84 | 19.70 | 17.51 | 18.60 | Under $36,000 | 4,790 |
| 85 | 19.11 | 16.83 | 17.97 | Under $36,000 | 4,519 |
| 86 | 19.12 | 16.61 | 17.87 | Under $36,000 | 3,537 |
| 87 | 18.55 | 15.88 | 17.21 | Under $36,000 | 2,963 |
| 88 | 16.64 | 13.84 | 15.24 | Under $36,000 | 2,612 |
| 89 | 17.29 | 14.05 | 15.67 | Under $36,000 | 1,946 |
| 90 | 17.02 | 13.55 | 15.29 | Under $36,000 | 1,694 |
| 91 | 17.85 | 13.42 | 15.64 | Under $36,000 | 1,036 |
| 92 | 15.91 | 11.20 | 13.56 | Under $36,000 | 804 |
| 93 | 18.20 | 12.34 | 15.27 | Under $36,000 | 550 |
| 94 | 14.39 | 7.43 | 10.91 | Under $36,000 | 330 |
| 18 | 43.79 | 40.81 | 42.30 | $36,000 to $89,999 | 4,364 |
| 19 | 45.71 | 42.71 | 44.21 | $36,000 to $89,999 | 3,583 |
| 20 | 47.20 | 43.96 | 45.58 | $36,000 to $89,999 | 3,493 |
| 21 | 48.18 | 45.16 | 46.67 | $36,000 to $89,999 | 3,784 |
| 22 | 47.84 | 44.79 | 46.32 | $36,000 to $89,999 | 3,923 |
| 23 | 48.15 | 45.36 | 46.75 | $36,000 to $89,999 | 4,620 |
| 24 | 48.57 | 45.74 | 47.15 | $36,000 to $89,999 | 5,043 |
| 25 | 49.26 | 46.78 | 48.02 | $36,000 to $89,999 | 5,862 |
| 26 | 48.67 | 46.04 | 47.35 | $36,000 to $89,999 | 5,797 |
| 27 | 48.79 | 46.25 | 47.52 | $36,000 to $89,999 | 6,292 |
| 28 | 48.09 | 45.80 | 46.95 | $36,000 to $89,999 | 6,927 |
| 29 | 49.16 | 46.78 | 47.97 | $36,000 to $89,999 | 7,015 |
| 30 | 47.84 | 45.62 | 46.73 | $36,000 to $89,999 | 8,200 |
| 31 | 49.72 | 47.33 | 48.52 | $36,000 to $89,999 | 7,013 |
| 32 | 50.37 | 48.15 | 49.26 | $36,000 to $89,999 | 7,804 |
| 33 | 49.66 | 47.37 | 48.52 | $36,000 to $89,999 | 7,247 |
| 34 | 48.86 | 46.51 | 47.69 | $36,000 to $89,999 | 7,327 |
| 35 | 48.16 | 45.99 | 47.07 | $36,000 to $89,999 | 8,051 |
| 36 | 47.41 | 45.17 | 46.29 | $36,000 to $89,999 | 7,315 |
| 37 | 49.32 | 47.14 | 48.23 | $36,000 to $89,999 | 7,369 |
| 38 | 49.04 | 46.78 | 47.91 | $36,000 to $89,999 | 7,841 |
| 39 | 48.69 | 46.36 | 47.53 | $36,000 to $89,999 | 7,598 |
| 40 | 48.19 | 46.17 | 47.18 | $36,000 to $89,999 | 9,581 |
| 41 | 48.09 | 45.91 | 47 | $36,000 to $89,999 | 7,717 |
| 42 | 47.54 | 45.51 | 46.53 | $36,000 to $89,999 | 9,233 |
| 43 | 47.62 | 45.40 | 46.51 | $36,000 to $89,999 | 8,505 |
| 44 | 47.54 | 45.38 | 46.46 | $36,000 to $89,999 | 8,224 |
| 45 | 44.81 | 42.87 | 43.84 | $36,000 to $89,999 | 9,674 |
| 46 | 45.94 | 43.86 | 44.90 | $36,000 to $89,999 | 8,882 |
| 47 | 46.02 | 44.02 | 45.02 | $36,000 to $89,999 | 9,844 |
| 48 | 45.56 | 43.62 | 44.59 | $36,000 to $89,999 | 9,890 |
| 49 | 44.96 | 42.94 | 43.95 | $36,000 to $89,999 | 10,166 |
| 50 | 42.80 | 41.09 | 41.94 | $36,000 to $89,999 | 13,299 |
| 51 | 44.13 | 42.17 | 43.15 | $36,000 to $89,999 | 10,334 |
| 52 | 43.70 | 41.98 | 42.84 | $36,000 to $89,999 | 12,410 |
| 53 | 43.11 | 41.19 | 42.15 | $36,000 to $89,999 | 11,526 |
| 54 | 42.50 | 40.69 | 41.60 | $36,000 to $89,999 | 11,828 |
| 55 | 40.85 | 39.19 | 40.02 | $36,000 to $89,999 | 13,125 |
| 56 | 40.67 | 38.89 | 39.78 | $36,000 to $89,999 | 12,398 |
| 57 | 39 | 37.19 | 38.09 | $36,000 to $89,999 | 12,498 |
| 58 | 38.78 | 37 | 37.89 | $36,000 to $89,999 | 12,866 |
| 59 | 37.44 | 35.78 | 36.61 | $36,000 to $89,999 | 12,816 |
| 60 | 35.27 | 33.74 | 34.51 | $36,000 to $89,999 | 15,018 |
| 61 | 33.56 | 31.75 | 32.66 | $36,000 to $89,999 | 12,246 |
| 62 | 31.63 | 30.15 | 30.89 | $36,000 to $89,999 | 15,295 |
| 63 | 30.09 | 28.54 | 29.32 | $36,000 to $89,999 | 13,621 |
| 64 | 27.25 | 25.73 | 26.49 | $36,000 to $89,999 | 13,874 |
| 65 | 25.62 | 24.20 | 24.91 | $36,000 to $89,999 | 15,045 |
| 66 | 23.88 | 22.26 | 23.07 | $36,000 to $89,999 | 12,317 |
| 67 | 23.34 | 21.79 | 22.57 | $36,000 to $89,999 | 12,221 |
| 68 | 21.24 | 19.67 | 20.46 | $36,000 to $89,999 | 11,116 |
| 69 | 21.03 | 19.44 | 20.23 | $36,000 to $89,999 | 10,156 |
| 70 | 20.28 | 18.83 | 19.56 | $36,000 to $89,999 | 11,673 |
| 71 | 18.82 | 17.10 | 17.96 | $36,000 to $89,999 | 8,654 |
| 72 | 19.06 | 17.47 | 18.27 | $36,000 to $89,999 | 8,913 |
| 73 | 18.29 | 16.59 | 17.44 | $36,000 to $89,999 | 7,615 |
| 74 | 18.30 | 16.54 | 17.42 | $36,000 to $89,999 | 6,939 |
| 75 | 18.40 | 16.67 | 17.53 | $36,000 to $89,999 | 7,506 |
| 76 | 18.43 | 16.44 | 17.43 | $36,000 to $89,999 | 5,879 |
| 77 | 18.42 | 16.44 | 17.43 | $36,000 to $89,999 | 5,714 |
| 78 | 17.04 | 15.12 | 16.08 | $36,000 to $89,999 | 5,317 |
| 79 | 17.60 | 15.50 | 16.55 | $36,000 to $89,999 | 4,672 |
| 80 | 17.10 | 15.06 | 16.08 | $36,000 to $89,999 | 5,273 |
| 81 | 17.40 | 15.02 | 16.21 | $36,000 to $89,999 | 3,831 |
| 82 | 18.03 | 15.53 | 16.78 | $36,000 to $89,999 | 3,684 |
| 83 | 16.55 | 13.99 | 15.27 | $36,000 to $89,999 | 3,261 |
| 84 | 17.14 | 14.53 | 15.84 | $36,000 to $89,999 | 2,728 |
| 85 | 17.96 | 15.18 | 16.57 | $36,000 to $89,999 | 2,638 |
| 86 | 16.64 | 13.42 | 15.03 | $36,000 to $89,999 | 1,883 |
| 87 | 16.31 | 13.03 | 14.67 | $36,000 to $89,999 | 1,602 |
| 88 | 17.03 | 13.17 | 15.10 | $36,000 to $89,999 | 1,344 |
| 89 | 14.91 | 10.50 | 12.71 | $36,000 to $89,999 | 905 |
| 90 | 17.42 | 11.94 | 14.68 | $36,000 to $89,999 | 797 |
| 91 | 15.24 | 9.52 | 12.38 | $36,000 to $89,999 | 525 |
| 92 | 15.21 | 9.12 | 12.17 | $36,000 to $89,999 | 411 |
| 93 | 15.79 | 7.19 | 11.49 | $36,000 to $89,999 | 235 |
| 94 | 14.13 | 5.68 | 9.90 | $36,000 to $89,999 | 202 |
| 18 | 44.68 | 40.75 | 42.72 | $90,000 or more | 2,774 |
| 19 | 47.48 | 43.21 | 45.34 | $90,000 or more | 2,062 |
| 20 | 48.50 | 44.03 | 46.27 | $90,000 or more | 1,848 |
| 21 | 47.79 | 43.03 | 45.41 | $90,000 or more | 1,852 |
| 22 | 49.17 | 44.11 | 46.64 | $90,000 or more | 1,668 |
| 23 | 48.76 | 44.01 | 46.38 | $90,000 or more | 1,673 |
| 24 | 50.84 | 46.18 | 48.51 | $90,000 or more | 1,573 |
| 25 | 50.06 | 45.28 | 47.67 | $90,000 or more | 1,760 |
| 26 | 51.09 | 45.93 | 48.51 | $90,000 or more | 1,713 |
| 27 | 51.51 | 47.05 | 49.28 | $90,000 or more | 2,078 |
| 28 | 49.13 | 45.35 | 47.24 | $90,000 or more | 2,392 |
| 29 | 50.25 | 46.68 | 48.46 | $90,000 or more | 2,703 |
| 30 | 47.74 | 44.49 | 46.11 | $90,000 or more | 3,307 |
| 31 | 50.96 | 47.65 | 49.31 | $90,000 or more | 3,320 |
| 32 | 49.63 | 46.38 | 48 | $90,000 or more | 3,831 |
| 33 | 50.03 | 46.89 | 48.46 | $90,000 or more | 3,896 |
| 34 | 48.69 | 45.69 | 47.19 | $90,000 or more | 4,141 |
| 35 | 48.41 | 45.67 | 47.04 | $90,000 or more | 4,913 |
| 36 | 48.57 | 45.62 | 47.09 | $90,000 or more | 4,629 |
| 37 | 49.10 | 46.40 | 47.75 | $90,000 or more | 4,930 |
| 38 | 49.50 | 46.64 | 48.07 | $90,000 or more | 5,307 |
| 39 | 48.35 | 45.74 | 47.05 | $90,000 or more | 5,569 |
| 40 | 47.18 | 44.58 | 45.88 | $90,000 or more | 6,724 |
| 41 | 48.54 | 45.81 | 47.18 | $90,000 or more | 5,700 |
| 42 | 47.50 | 45.10 | 46.30 | $90,000 or more | 7,019 |
| 43 | 48.48 | 46.12 | 47.30 | $90,000 or more | 6,488 |
| 44 | 46.08 | 43.66 | 44.87 | $90,000 or more | 6,361 |
| 45 | 44.60 | 42.37 | 43.49 | $90,000 or more | 7,299 |
| 46 | 45.72 | 43.34 | 44.53 | $90,000 or more | 7,018 |
| 47 | 44.92 | 42.56 | 43.74 | $90,000 or more | 7,298 |
| 48 | 44.42 | 42.29 | 43.36 | $90,000 or more | 7,676 |
| 49 | 44.67 | 42.52 | 43.59 | $90,000 or more | 7,671 |
| 50 | 43.44 | 41.38 | 42.41 | $90,000 or more | 9,555 |
| 51 | 44.27 | 41.93 | 43.10 | $90,000 or more | 7,506 |
| 52 | 42 | 39.90 | 40.95 | $90,000 or more | 8,918 |
| 53 | 41.81 | 39.59 | 40.70 | $90,000 or more | 8,491 |
| 54 | 40.47 | 38.43 | 39.45 | $90,000 or more | 8,289 |
| 55 | 40.58 | 38.57 | 39.57 | $90,000 or more | 9,155 |
| 56 | 39.55 | 37.46 | 38.51 | $90,000 or more | 8,209 |
| 57 | 39.78 | 37.51 | 38.64 | $90,000 or more | 8,304 |
| 58 | 38.87 | 36.78 | 37.82 | $90,000 or more | 8,323 |
| 59 | 36.45 | 34.27 | 35.36 | $90,000 or more | 7,995 |
| 60 | 34.92 | 32.96 | 33.94 | $90,000 or more | 9,016 |
| 61 | 33.04 | 30.73 | 31.88 | $90,000 or more | 7,076 |
| 62 | 31.11 | 28.98 | 30.04 | $90,000 or more | 8,188 |
| 63 | 30.61 | 28.40 | 29.51 | $90,000 or more | 7,090 |
| 64 | 28.13 | 25.94 | 27.03 | $90,000 or more | 6,714 |
| 65 | 25.65 | 23.57 | 24.61 | $90,000 or more | 6,920 |
| 66 | 24.01 | 21.79 | 22.90 | $90,000 or more | 5,529 |
| 67 | 22.03 | 19.83 | 20.93 | $90,000 or more | 5,274 |
| 68 | 22.24 | 19.80 | 21.02 | $90,000 or more | 4,700 |
| 69 | 20.58 | 17.99 | 19.28 | $90,000 or more | 3,952 |
| 70 | 20.03 | 17.68 | 18.85 | $90,000 or more | 4,429 |
| 71 | 19.54 | 16.83 | 18.18 | $90,000 or more | 3,102 |
| 72 | 19.07 | 16.49 | 17.78 | $90,000 or more | 3,099 |
| 73 | 18.99 | 16.16 | 17.57 | $90,000 or more | 2,686 |
| 74 | 19.07 | 15.94 | 17.51 | $90,000 or more | 2,302 |
| 75 | 19.18 | 16.11 | 17.64 | $90,000 or more | 2,369 |
| 76 | 18.34 | 15.09 | 16.71 | $90,000 or more | 1,843 |
| 77 | 19.96 | 16.33 | 18.15 | $90,000 or more | 1,802 |
| 78 | 19.43 | 15.76 | 17.60 | $90,000 or more | 1,597 |
| 79 | 18.95 | 14.70 | 16.83 | $90,000 or more | 1,337 |
| 80 | 18.93 | 15.29 | 17.11 | $90,000 or more | 1,613 |
| 81 | 18.88 | 14.37 | 16.62 | $90,000 or more | 1,131 |
| 82 | 17.11 | 13.21 | 15.16 | $90,000 or more | 1,141 |
| 83 | 19.31 | 14.43 | 16.87 | $90,000 or more | 913 |
| 84 | 17.38 | 12.53 | 14.96 | $90,000 or more | 789 |
| 85 | 16.51 | 11.41 | 13.96 | $90,000 or more | 738 |
| 86 | 16.47 | 10.40 | 13.43 | $90,000 or more | 536 |
| 87 | 21.38 | 14.63 | 18 | $90,000 or more | 461 |
| 88 | 19.06 | 12.08 | 15.57 | $90,000 or more | 411 |
| 89 | 16.62 | 8.98 | 12.80 | $90,000 or more | 289 |
| 90 | 19.99 | 11.31 | 15.65 | $90,000 or more | 262 |
| 91 | 14.52 | 6.29 | 10.40 | $90,000 or more | 173 |
| 92 | 21.53 | 8.09 | 14.81 | $90,000 or more | 135 |
| 93 | 27.23 | 9.39 | 18.31 | $90,000 or more | 71 |
| 94 | 17.02 | 1.49 | 9.26 | $90,000 or more | 54 |

#### Table 4.4. Physical pain

Physical pain was 41.2% at ages 45 to 54 in households under $36,000, against 15.7% in households of $90,000 or more

Share of adults who reported physical pain during a lot of the day yesterday by year of age and household income group, Gallup 2009 to 2014

| age | conf_high | conf_low | estimate | group | n |
| ---: | ---: | ---: | ---: | --- | ---: |
| 18 | 17.91 | 15.95 | 16.93 | Under $36,000 | 5,807 |
| 19 | 17.74 | 15.88 | 16.81 | Under $36,000 | 5,973 |
| 20 | 18.67 | 16.61 | 17.64 | Under $36,000 | 6,729 |
| 21 | 17.60 | 15.85 | 16.73 | Under $36,000 | 7,360 |
| 22 | 18.54 | 16.64 | 17.59 | Under $36,000 | 6,998 |
| 23 | 18.65 | 16.75 | 17.70 | Under $36,000 | 6,921 |
| 24 | 19.76 | 17.84 | 18.80 | Under $36,000 | 6,703 |
| 25 | 20.67 | 18.80 | 19.74 | Under $36,000 | 7,067 |
| 26 | 22.10 | 20.03 | 21.07 | Under $36,000 | 6,057 |
| 27 | 23.33 | 21.09 | 22.21 | Under $36,000 | 5,920 |
| 28 | 23.29 | 21.23 | 22.26 | Under $36,000 | 6,222 |
| 29 | 25.07 | 22.96 | 24.02 | Under $36,000 | 5,525 |
| 30 | 25.03 | 22.95 | 23.99 | Under $36,000 | 6,678 |
| 31 | 27.36 | 24.83 | 26.09 | Under $36,000 | 4,729 |
| 32 | 27.74 | 25.28 | 26.51 | Under $36,000 | 5,696 |
| 33 | 27.43 | 24.75 | 26.09 | Under $36,000 | 4,757 |
| 34 | 29.10 | 26.46 | 27.78 | Under $36,000 | 4,690 |
| 35 | 29.66 | 27.12 | 28.39 | Under $36,000 | 5,533 |
| 36 | 30.98 | 28.07 | 29.52 | Under $36,000 | 4,559 |
| 37 | 32.81 | 30.18 | 31.50 | Under $36,000 | 4,645 |
| 38 | 32.73 | 29.92 | 31.32 | Under $36,000 | 4,945 |
| 39 | 33.02 | 30.10 | 31.56 | Under $36,000 | 4,613 |
| 40 | 32.95 | 30.17 | 31.56 | Under $36,000 | 6,619 |
| 41 | 34.87 | 32.13 | 33.50 | Under $36,000 | 4,454 |
| 42 | 35.14 | 32.54 | 33.84 | Under $36,000 | 5,739 |
| 43 | 36.54 | 33.93 | 35.24 | Under $36,000 | 5,375 |
| 44 | 38.22 | 35.36 | 36.79 | Under $36,000 | 5,197 |
| 45 | 37.54 | 35.07 | 36.30 | Under $36,000 | 6,523 |
| 46 | 39.68 | 37.09 | 38.38 | Under $36,000 | 6,052 |
| 47 | 40.48 | 37.88 | 39.18 | Under $36,000 | 6,702 |
| 48 | 41.54 | 39.11 | 40.32 | Under $36,000 | 6,954 |
| 49 | 42.11 | 39.48 | 40.79 | Under $36,000 | 7,038 |
| 50 | 41.56 | 39.42 | 40.49 | Under $36,000 | 10,005 |
| 51 | 44.02 | 41.42 | 42.72 | Under $36,000 | 7,432 |
| 52 | 43.62 | 41.17 | 42.39 | Under $36,000 | 8,978 |
| 53 | 45.41 | 43.02 | 44.21 | Under $36,000 | 8,712 |
| 54 | 45.59 | 43.44 | 44.52 | Under $36,000 | 8,691 |
| 55 | 44.46 | 42.42 | 43.44 | Under $36,000 | 9,689 |
| 56 | 46.17 | 43.93 | 45.05 | Under $36,000 | 8,865 |
| 57 | 45.34 | 43.14 | 44.24 | Under $36,000 | 9,272 |
| 58 | 46.29 | 44.22 | 45.26 | Under $36,000 | 9,285 |
| 59 | 45.31 | 43.08 | 44.20 | Under $36,000 | 9,055 |
| 60 | 44.69 | 42.67 | 43.68 | Under $36,000 | 10,823 |
| 61 | 43.89 | 41.68 | 42.79 | Under $36,000 | 8,636 |
| 62 | 43.03 | 41.01 | 42.02 | Under $36,000 | 10,576 |
| 63 | 42.94 | 40.85 | 41.89 | Under $36,000 | 9,331 |
| 64 | 41.57 | 39.62 | 40.59 | Under $36,000 | 9,250 |
| 65 | 38.84 | 36.99 | 37.91 | Under $36,000 | 10,653 |
| 66 | 37.90 | 35.85 | 36.88 | Under $36,000 | 8,862 |
| 67 | 38.15 | 36.10 | 37.13 | Under $36,000 | 9,139 |
| 68 | 36.90 | 34.79 | 35.84 | Under $36,000 | 8,774 |
| 69 | 35.74 | 33.67 | 34.71 | Under $36,000 | 8,307 |
| 70 | 34.02 | 32.09 | 33.06 | Under $36,000 | 9,968 |
| 71 | 34.57 | 32.50 | 33.54 | Under $36,000 | 7,932 |
| 72 | 33.49 | 31.48 | 32.48 | Under $36,000 | 8,644 |
| 73 | 33.95 | 31.93 | 32.94 | Under $36,000 | 8,045 |
| 74 | 33.06 | 30.92 | 31.99 | Under $36,000 | 7,671 |
| 75 | 32.96 | 30.85 | 31.91 | Under $36,000 | 8,412 |
| 76 | 33.11 | 30.96 | 32.04 | Under $36,000 | 7,198 |
| 77 | 32.68 | 30.45 | 31.56 | Under $36,000 | 7,084 |
| 78 | 31.62 | 29.43 | 30.53 | Under $36,000 | 6,922 |
| 79 | 32.48 | 30.16 | 31.32 | Under $36,000 | 6,347 |
| 80 | 30.38 | 28.30 | 29.34 | Under $36,000 | 7,739 |
| 81 | 31.60 | 29.19 | 30.40 | Under $36,000 | 5,810 |
| 82 | 31.77 | 29.45 | 30.61 | Under $36,000 | 6,005 |
| 83 | 32.32 | 29.88 | 31.10 | Under $36,000 | 5,341 |
| 84 | 32.11 | 29.42 | 30.76 | Under $36,000 | 4,785 |
| 85 | 31.10 | 28.40 | 29.75 | Under $36,000 | 4,511 |
| 86 | 31.21 | 28.04 | 29.63 | Under $36,000 | 3,534 |
| 87 | 32.20 | 28.94 | 30.57 | Under $36,000 | 2,957 |
| 88 | 29.28 | 25.88 | 27.58 | Under $36,000 | 2,607 |
| 89 | 31.21 | 27.08 | 29.14 | Under $36,000 | 1,956 |
| 90 | 28.86 | 24.78 | 26.82 | Under $36,000 | 1,689 |
| 91 | 31.95 | 26.14 | 29.04 | Under $36,000 | 1,033 |
| 92 | 32.68 | 26.27 | 29.48 | Under $36,000 | 804 |
| 93 | 33.81 | 25.93 | 29.87 | Under $36,000 | 549 |
| 94 | 30.24 | 20.52 | 25.38 | Under $36,000 | 331 |
| 18 | 15.12 | 13.01 | 14.07 | $36,000 to $89,999 | 4,365 |
| 19 | 15.22 | 13.04 | 14.13 | $36,000 to $89,999 | 3,581 |
| 20 | 16.26 | 13.95 | 15.11 | $36,000 to $89,999 | 3,495 |
| 21 | 15.29 | 12.97 | 14.13 | $36,000 to $89,999 | 3,786 |
| 22 | 15.28 | 13.12 | 14.20 | $36,000 to $89,999 | 3,922 |
| 23 | 13.89 | 11.99 | 12.94 | $36,000 to $89,999 | 4,620 |
| 24 | 14.15 | 12.14 | 13.14 | $36,000 to $89,999 | 5,044 |
| 25 | 13.35 | 11.73 | 12.54 | $36,000 to $89,999 | 5,863 |
| 26 | 13.14 | 11.39 | 12.26 | $36,000 to $89,999 | 5,797 |
| 27 | 14.23 | 12.53 | 13.38 | $36,000 to $89,999 | 6,294 |
| 28 | 13.47 | 12.03 | 12.75 | $36,000 to $89,999 | 6,927 |
| 29 | 13.95 | 12.22 | 13.08 | $36,000 to $89,999 | 7,016 |
| 30 | 14.37 | 12.80 | 13.59 | $36,000 to $89,999 | 8,200 |
| 31 | 15.60 | 13.95 | 14.77 | $36,000 to $89,999 | 7,013 |
| 32 | 15.37 | 13.76 | 14.56 | $36,000 to $89,999 | 7,807 |
| 33 | 15.89 | 14.22 | 15.06 | $36,000 to $89,999 | 7,253 |
| 34 | 15.59 | 13.94 | 14.77 | $36,000 to $89,999 | 7,328 |
| 35 | 15.40 | 13.86 | 14.63 | $36,000 to $89,999 | 8,051 |
| 36 | 17 | 15.17 | 16.09 | $36,000 to $89,999 | 7,316 |
| 37 | 18.10 | 16.38 | 17.24 | $36,000 to $89,999 | 7,372 |
| 38 | 16 | 14.44 | 15.22 | $36,000 to $89,999 | 7,846 |
| 39 | 17.70 | 15.92 | 16.81 | $36,000 to $89,999 | 7,602 |
| 40 | 17.53 | 16.06 | 16.80 | $36,000 to $89,999 | 9,585 |
| 41 | 17.89 | 16.26 | 17.07 | $36,000 to $89,999 | 7,719 |
| 42 | 18.82 | 17.14 | 17.98 | $36,000 to $89,999 | 9,238 |
| 43 | 19.61 | 17.92 | 18.76 | $36,000 to $89,999 | 8,511 |
| 44 | 19.71 | 17.94 | 18.82 | $36,000 to $89,999 | 8,229 |
| 45 | 19.51 | 17.88 | 18.69 | $36,000 to $89,999 | 9,677 |
| 46 | 21.56 | 19.88 | 20.72 | $36,000 to $89,999 | 8,881 |
| 47 | 22 | 20.28 | 21.14 | $36,000 to $89,999 | 9,844 |
| 48 | 22.99 | 21.27 | 22.13 | $36,000 to $89,999 | 9,891 |
| 49 | 23.45 | 21.77 | 22.61 | $36,000 to $89,999 | 10,168 |
| 50 | 23.09 | 21.64 | 22.36 | $36,000 to $89,999 | 13,299 |
| 51 | 24.38 | 22.57 | 23.47 | $36,000 to $89,999 | 10,336 |
| 52 | 23.41 | 21.92 | 22.67 | $36,000 to $89,999 | 12,415 |
| 53 | 24.67 | 23.15 | 23.91 | $36,000 to $89,999 | 11,525 |
| 54 | 25.20 | 23.59 | 24.39 | $36,000 to $89,999 | 11,824 |
| 55 | 24.61 | 23.13 | 23.87 | $36,000 to $89,999 | 13,130 |
| 56 | 25.59 | 24.12 | 24.85 | $36,000 to $89,999 | 12,400 |
| 57 | 26.06 | 24.48 | 25.27 | $36,000 to $89,999 | 12,493 |
| 58 | 25.83 | 24.20 | 25.02 | $36,000 to $89,999 | 12,871 |
| 59 | 25.39 | 23.83 | 24.61 | $36,000 to $89,999 | 12,819 |
| 60 | 25.20 | 23.85 | 24.53 | $36,000 to $89,999 | 15,021 |
| 61 | 27.04 | 25.45 | 26.25 | $36,000 to $89,999 | 12,246 |
| 62 | 25.61 | 24.16 | 24.88 | $36,000 to $89,999 | 15,295 |
| 63 | 26.18 | 24.63 | 25.41 | $36,000 to $89,999 | 13,610 |
| 64 | 24.12 | 22.70 | 23.41 | $36,000 to $89,999 | 13,870 |
| 65 | 23.52 | 22.13 | 22.83 | $36,000 to $89,999 | 15,044 |
| 66 | 24.08 | 22.60 | 23.34 | $36,000 to $89,999 | 12,317 |
| 67 | 24.42 | 22.97 | 23.70 | $36,000 to $89,999 | 12,220 |
| 68 | 23.68 | 21.98 | 22.83 | $36,000 to $89,999 | 11,109 |
| 69 | 23.44 | 21.87 | 22.65 | $36,000 to $89,999 | 10,149 |
| 70 | 22.62 | 21.09 | 21.86 | $36,000 to $89,999 | 11,667 |
| 71 | 23.28 | 21.45 | 22.36 | $36,000 to $89,999 | 8,652 |
| 72 | 22.80 | 21.17 | 21.99 | $36,000 to $89,999 | 8,919 |
| 73 | 23.05 | 21.22 | 22.14 | $36,000 to $89,999 | 7,621 |
| 74 | 22.12 | 20.17 | 21.14 | $36,000 to $89,999 | 6,934 |
| 75 | 22.92 | 20.94 | 21.93 | $36,000 to $89,999 | 7,501 |
| 76 | 23.94 | 21.83 | 22.88 | $36,000 to $89,999 | 5,873 |
| 77 | 24.46 | 22.09 | 23.28 | $36,000 to $89,999 | 5,705 |
| 78 | 23.18 | 20.95 | 22.07 | $36,000 to $89,999 | 5,316 |
| 79 | 24.39 | 22.07 | 23.23 | $36,000 to $89,999 | 4,667 |
| 80 | 23.13 | 20.94 | 22.04 | $36,000 to $89,999 | 5,268 |
| 81 | 24.62 | 22.06 | 23.34 | $36,000 to $89,999 | 3,826 |
| 82 | 25.80 | 22.99 | 24.40 | $36,000 to $89,999 | 3,681 |
| 83 | 24.40 | 21.40 | 22.90 | $36,000 to $89,999 | 3,262 |
| 84 | 24.12 | 20.77 | 22.45 | $36,000 to $89,999 | 2,722 |
| 85 | 24.94 | 21.61 | 23.28 | $36,000 to $89,999 | 2,638 |
| 86 | 24.74 | 21.06 | 22.90 | $36,000 to $89,999 | 1,878 |
| 87 | 28.47 | 24.18 | 26.33 | $36,000 to $89,999 | 1,603 |
| 88 | 25.41 | 21.32 | 23.36 | $36,000 to $89,999 | 1,344 |
| 89 | 25.88 | 20.54 | 23.21 | $36,000 to $89,999 | 909 |
| 90 | 23.56 | 17.65 | 20.60 | $36,000 to $89,999 | 796 |
| 91 | 26.74 | 19.94 | 23.34 | $36,000 to $89,999 | 527 |
| 92 | 30.40 | 22.28 | 26.34 | $36,000 to $89,999 | 410 |
| 93 | 26.71 | 16.51 | 21.61 | $36,000 to $89,999 | 236 |
| 94 | 28.32 | 16.23 | 22.28 | $36,000 to $89,999 | 202 |
| 18 | 15.50 | 13.06 | 14.28 | $90,000 or more | 2,773 |
| 19 | 14.82 | 12.01 | 13.41 | $90,000 or more | 2,065 |
| 20 | 14.25 | 10.96 | 12.60 | $90,000 or more | 1,849 |
| 21 | 13.02 | 10.11 | 11.56 | $90,000 or more | 1,851 |
| 22 | 15.35 | 12 | 13.68 | $90,000 or more | 1,667 |
| 23 | 14.13 | 10.98 | 12.55 | $90,000 or more | 1,673 |
| 24 | 14.32 | 10.86 | 12.59 | $90,000 or more | 1,573 |
| 25 | 14.73 | 11.52 | 13.12 | $90,000 or more | 1,760 |
| 26 | 12.78 | 9.87 | 11.33 | $90,000 or more | 1,713 |
| 27 | 11.86 | 9.14 | 10.50 | $90,000 or more | 2,077 |
| 28 | 12.44 | 9.62 | 11.03 | $90,000 or more | 2,393 |
| 29 | 11.81 | 9.43 | 10.62 | $90,000 or more | 2,702 |
| 30 | 11.59 | 9.59 | 10.59 | $90,000 or more | 3,305 |
| 31 | 11.45 | 9.33 | 10.39 | $90,000 or more | 3,320 |
| 32 | 11.21 | 9.14 | 10.17 | $90,000 or more | 3,833 |
| 33 | 12.45 | 10.49 | 11.47 | $90,000 or more | 3,896 |
| 34 | 11.62 | 9.69 | 10.65 | $90,000 or more | 4,139 |
| 35 | 12.21 | 10.54 | 11.38 | $90,000 or more | 4,914 |
| 36 | 12.63 | 10.77 | 11.70 | $90,000 or more | 4,633 |
| 37 | 11.49 | 9.88 | 10.69 | $90,000 or more | 4,932 |
| 38 | 12.72 | 10.98 | 11.85 | $90,000 or more | 5,308 |
| 39 | 12.14 | 10.58 | 11.36 | $90,000 or more | 5,571 |
| 40 | 11.95 | 10.29 | 11.12 | $90,000 or more | 6,728 |
| 41 | 13.42 | 11.67 | 12.54 | $90,000 or more | 5,700 |
| 42 | 13.18 | 11.60 | 12.39 | $90,000 or more | 7,022 |
| 43 | 14.05 | 12.39 | 13.22 | $90,000 or more | 6,490 |
| 44 | 14.74 | 13.07 | 13.90 | $90,000 or more | 6,365 |
| 45 | 14.48 | 12.88 | 13.68 | $90,000 or more | 7,302 |
| 46 | 14.57 | 12.85 | 13.71 | $90,000 or more | 7,016 |
| 47 | 15.64 | 14.04 | 14.84 | $90,000 or more | 7,297 |
| 48 | 16.11 | 14.53 | 15.32 | $90,000 or more | 7,676 |
| 49 | 16.34 | 14.66 | 15.50 | $90,000 or more | 7,665 |
| 50 | 16.93 | 15.30 | 16.12 | $90,000 or more | 9,555 |
| 51 | 18.26 | 16.45 | 17.36 | $90,000 or more | 7,507 |
| 52 | 17.51 | 15.88 | 16.69 | $90,000 or more | 8,919 |
| 53 | 17.11 | 15.62 | 16.36 | $90,000 or more | 8,500 |
| 54 | 17.52 | 15.92 | 16.72 | $90,000 or more | 8,288 |
| 55 | 18.80 | 17.06 | 17.93 | $90,000 or more | 9,154 |
| 56 | 19.29 | 17.56 | 18.42 | $90,000 or more | 8,212 |
| 57 | 19.03 | 17.26 | 18.15 | $90,000 or more | 8,305 |
| 58 | 19.65 | 18.02 | 18.83 | $90,000 or more | 8,320 |
| 59 | 20.06 | 18.36 | 19.21 | $90,000 or more | 7,997 |
| 60 | 19.57 | 17.98 | 18.78 | $90,000 or more | 9,022 |
| 61 | 19.82 | 18.06 | 18.94 | $90,000 or more | 7,081 |
| 62 | 19.92 | 18.11 | 19.02 | $90,000 or more | 8,193 |
| 63 | 19.83 | 17.95 | 18.89 | $90,000 or more | 7,090 |
| 64 | 20.25 | 18.32 | 19.29 | $90,000 or more | 6,720 |
| 65 | 18.39 | 16.57 | 17.48 | $90,000 or more | 6,917 |
| 66 | 20.33 | 18.22 | 19.28 | $90,000 or more | 5,525 |
| 67 | 18.73 | 16.64 | 17.69 | $90,000 or more | 5,275 |
| 68 | 18.53 | 16.35 | 17.44 | $90,000 or more | 4,702 |
| 69 | 20.83 | 18.30 | 19.56 | $90,000 or more | 3,951 |
| 70 | 18.96 | 16.62 | 17.79 | $90,000 or more | 4,429 |
| 71 | 20.28 | 17.55 | 18.92 | $90,000 or more | 3,103 |
| 72 | 20.08 | 17.31 | 18.70 | $90,000 or more | 3,097 |
| 73 | 21.54 | 18.51 | 20.02 | $90,000 or more | 2,687 |
| 74 | 21.43 | 18.26 | 19.84 | $90,000 or more | 2,298 |
| 75 | 20.03 | 16.68 | 18.35 | $90,000 or more | 2,370 |
| 76 | 21.25 | 17.64 | 19.45 | $90,000 or more | 1,841 |
| 77 | 23.29 | 19.35 | 21.32 | $90,000 or more | 1,801 |
| 78 | 23.61 | 19.50 | 21.55 | $90,000 or more | 1,596 |
| 79 | 21.56 | 16.97 | 19.27 | $90,000 or more | 1,339 |
| 80 | 21.83 | 17.89 | 19.86 | $90,000 or more | 1,611 |
| 81 | 23.94 | 19.35 | 21.64 | $90,000 or more | 1,132 |
| 82 | 22.56 | 17.93 | 20.25 | $90,000 or more | 1,136 |
| 83 | 27.41 | 21.60 | 24.51 | $90,000 or more | 910 |
| 84 | 24.55 | 19.05 | 21.80 | $90,000 or more | 789 |
| 85 | 24.94 | 18.69 | 21.82 | $90,000 or more | 738 |
| 86 | 23.72 | 16.51 | 20.11 | $90,000 or more | 532 |
| 87 | 29.98 | 21.76 | 25.87 | $90,000 or more | 460 |
| 88 | 31.77 | 23.70 | 27.74 | $90,000 or more | 411 |
| 89 | 25.96 | 16.40 | 21.18 | $90,000 or more | 288 |
| 90 | 27.09 | 17.36 | 22.22 | $90,000 or more | 261 |
| 91 | 27.75 | 16.43 | 22.09 | $90,000 or more | 172 |
| 92 | 34.34 | 18.99 | 26.67 | $90,000 or more | 135 |
| 93 | 31.19 | 11.06 | 21.13 | $90,000 or more | 71 |
| 94 | 33.43 | 11.01 | 22.22 | $90,000 or more | 54 |

Note: Each age is a different group of people interviewed in the same years, so a profile mixes age with birth cohort and does not follow anyone through life.

Note: Cells at the oldest ages hold fewer interviews, so their intervals are wider; the extract has no interviews above age 94.

Note: This figure uses interviews in 2009 to 2014. The share of income reporters at $90,000 or more was 23% in 2014 and 35% in 2017, more than inflation explains, so household income is not comparable across 2015 to 2017. 81% of interviews in 2009 to 2014 report a household income bracket. Gallup's brackets are in dollars of the interview year and are not adjusted for inflation; the groups are the harmonized six brackets collapsed to three.

Note: Gallup's extract has no survey weights, so every estimate is unweighted and levels can differ from the weighted figures Gallup publishes. County comes from the respondent's ZIP code and is approximate near county lines.

Source: Gallup U.S. Daily, 2009-2014 (household income is not comparable across 2015-2017); unweighted, intervals clustered by county

Data: https://vishalsingh.org/zipcode/report/well-being-across-the-life-span/figures/profiles_by_income.json

### Table 5. Adjusting for household income moves the lowest point of the ladder profile from age 54 to age 48 with overlapping intervals, and a quadratic fit puts it at age 44, so no single age is the turning point

Cantril ladder score by year of age, raw and adjusted, the age of its lowest point under different adjustments and for groups, and the fall and rise between age bands, Gallup 2009 to 2017

#### Table 5.1. Raw and adjusted profiles

Adjusting for sex, marital status, education, household income and survey year deepens the fall from ages 25 to 34 to ages 45 to 54 to −0.32 ladder points from −0.20

Cantril ladder score by year of age, raw and adjusted for sex, marital status, education, household income and survey year, adults with all of these reported, Gallup 2009 to 2014

| age | conf_high | conf_low | estimate | group | n |
| ---: | ---: | ---: | ---: | --- | ---: |
| 18 | 7.30 | 7.23 | 7.27 | Raw | 12,653 |
| 19 | 7.17 | 7.10 | 7.13 | Raw | 11,389 |
| 20 | 7.09 | 7.03 | 7.06 | Raw | 11,795 |
| 21 | 7.07 | 7.01 | 7.04 | Raw | 12,694 |
| 22 | 7.01 | 6.94 | 6.98 | Raw | 12,286 |
| 23 | 6.94 | 6.88 | 6.91 | Raw | 12,871 |
| 24 | 6.91 | 6.83 | 6.87 | Raw | 12,949 |
| 25 | 6.96 | 6.90 | 6.93 | Raw | 14,254 |
| 26 | 6.94 | 6.87 | 6.90 | Raw | 13,185 |
| 27 | 6.96 | 6.90 | 6.93 | Raw | 13,893 |
| 28 | 6.99 | 6.93 | 6.96 | Raw | 15,054 |
| 29 | 6.99 | 6.92 | 6.96 | Raw | 14,739 |
| 30 | 7.00 | 6.95 | 6.98 | Raw | 17,613 |
| 31 | 7.03 | 6.97 | 7.00 | Raw | 14,543 |
| 32 | 7.00 | 6.93 | 6.97 | Raw | 16,745 |
| 33 | 7.03 | 6.97 | 7.00 | Raw | 15,325 |
| 34 | 7.02 | 6.96 | 6.99 | Raw | 15,611 |
| 35 | 7.02 | 6.96 | 6.99 | Raw | 17,824 |
| 36 | 7.02 | 6.96 | 6.99 | Raw | 15,895 |
| 37 | 6.98 | 6.92 | 6.95 | Raw | 16,277 |
| 38 | 6.98 | 6.92 | 6.95 | Raw | 17,359 |
| 39 | 7.00 | 6.94 | 6.97 | Raw | 17,119 |
| 40 | 6.95 | 6.89 | 6.92 | Raw | 22,087 |
| 41 | 6.96 | 6.90 | 6.93 | Raw | 17,268 |
| 42 | 6.89 | 6.84 | 6.87 | Raw | 21,197 |
| 43 | 6.86 | 6.80 | 6.83 | Raw | 19,594 |
| 44 | 6.88 | 6.82 | 6.85 | Raw | 19,040 |
| 45 | 6.86 | 6.80 | 6.83 | Raw | 22,646 |
| 46 | 6.81 | 6.75 | 6.78 | Raw | 21,122 |
| 47 | 6.79 | 6.73 | 6.76 | Raw | 22,931 |
| 48 | 6.78 | 6.73 | 6.75 | Raw | 23,652 |
| 49 | 6.78 | 6.72 | 6.75 | Raw | 23,942 |
| 50 | 6.77 | 6.71 | 6.74 | Raw | 31,687 |
| 51 | 6.80 | 6.74 | 6.77 | Raw | 24,325 |
| 52 | 6.77 | 6.71 | 6.74 | Raw | 29,269 |
| 53 | 6.75 | 6.70 | 6.72 | Raw | 27,701 |
| 54 | 6.77 | 6.72 | 6.75 | Raw | 27,744 |
| 55 | 6.77 | 6.72 | 6.75 | Raw | 30,907 |
| 56 | 6.76 | 6.71 | 6.73 | Raw | 28,442 |
| 57 | 6.79 | 6.74 | 6.77 | Raw | 29,066 |
| 58 | 6.82 | 6.76 | 6.79 | Raw | 29,341 |
| 59 | 6.88 | 6.82 | 6.85 | Raw | 28,808 |
| 60 | 6.92 | 6.88 | 6.90 | Raw | 33,569 |
| 61 | 6.99 | 6.93 | 6.96 | Raw | 26,962 |
| 62 | 7.03 | 6.98 | 7.00 | Raw | 32,851 |
| 63 | 7.05 | 7.00 | 7.03 | Raw | 29,174 |
| 64 | 7.12 | 7.08 | 7.10 | Raw | 28,851 |
| 65 | 7.16 | 7.11 | 7.14 | Raw | 31,582 |
| 66 | 7.22 | 7.17 | 7.19 | Raw | 25,857 |
| 67 | 7.23 | 7.18 | 7.21 | Raw | 25,770 |
| 68 | 7.24 | 7.19 | 7.22 | Raw | 23,753 |
| 69 | 7.27 | 7.21 | 7.24 | Raw | 21,674 |
| 70 | 7.29 | 7.24 | 7.27 | Raw | 25,178 |
| 71 | 7.30 | 7.24 | 7.27 | Raw | 19,021 |
| 72 | 7.30 | 7.23 | 7.27 | Raw | 19,910 |
| 73 | 7.29 | 7.23 | 7.26 | Raw | 17,652 |
| 74 | 7.33 | 7.26 | 7.29 | Raw | 16,282 |
| 75 | 7.31 | 7.25 | 7.28 | Raw | 17,652 |
| 76 | 7.29 | 7.22 | 7.25 | Raw | 14,355 |
| 77 | 7.33 | 7.26 | 7.29 | Raw | 13,990 |
| 78 | 7.33 | 7.26 | 7.30 | Raw | 13,249 |
| 79 | 7.33 | 7.26 | 7.30 | Raw | 11,898 |
| 80 | 7.38 | 7.31 | 7.35 | Raw | 13,990 |
| 81 | 7.38 | 7.30 | 7.34 | Raw | 10,362 |
| 82 | 7.37 | 7.29 | 7.33 | Raw | 10,372 |
| 83 | 7.32 | 7.23 | 7.27 | Raw | 9,106 |
| 84 | 7.39 | 7.30 | 7.34 | Raw | 7,941 |
| 85 | 7.36 | 7.27 | 7.31 | Raw | 7,534 |
| 86 | 7.41 | 7.30 | 7.36 | Raw | 5,655 |
| 87 | 7.41 | 7.29 | 7.35 | Raw | 4,787 |
| 88 | 7.39 | 7.26 | 7.33 | Raw | 4,116 |
| 89 | 7.45 | 7.30 | 7.37 | Raw | 2,983 |
| 90 | 7.51 | 7.33 | 7.42 | Raw | 2,596 |
| 91 | 7.66 | 7.46 | 7.57 | Raw | 1,626 |
| 92 | 7.56 | 7.32 | 7.44 | Raw | 1,256 |
| 93 | 7.59 | 7.28 | 7.43 | Raw | 802 |
| 94 | 7.67 | 7.31 | 7.50 | Raw | 548 |
| 18 | 7.60 | 7.53 | 7.56 | Adjusted | 12,653 |
| 19 | 7.54 | 7.47 | 7.50 | Adjusted | 11,389 |
| 20 | 7.50 | 7.43 | 7.47 | Adjusted | 11,795 |
| 21 | 7.47 | 7.40 | 7.44 | Adjusted | 12,694 |
| 22 | 7.34 | 7.26 | 7.30 | Adjusted | 12,286 |
| 23 | 7.20 | 7.14 | 7.17 | Adjusted | 12,871 |
| 24 | 7.13 | 7.05 | 7.09 | Adjusted | 12,949 |
| 25 | 7.14 | 7.08 | 7.11 | Adjusted | 14,254 |
| 26 | 7.07 | 7.01 | 7.04 | Adjusted | 13,185 |
| 27 | 7.05 | 7.00 | 7.02 | Adjusted | 13,893 |
| 28 | 7.05 | 6.99 | 7.02 | Adjusted | 15,054 |
| 29 | 6.99 | 6.93 | 6.96 | Adjusted | 14,739 |
| 30 | 7.00 | 6.94 | 6.97 | Adjusted | 17,613 |
| 31 | 6.96 | 6.89 | 6.92 | Adjusted | 14,543 |
| 32 | 6.93 | 6.87 | 6.90 | Adjusted | 16,745 |
| 33 | 6.92 | 6.86 | 6.89 | Adjusted | 15,325 |
| 34 | 6.90 | 6.84 | 6.87 | Adjusted | 15,611 |
| 35 | 6.90 | 6.84 | 6.87 | Adjusted | 17,824 |
| 36 | 6.88 | 6.82 | 6.85 | Adjusted | 15,895 |
| 37 | 6.83 | 6.77 | 6.80 | Adjusted | 16,277 |
| 38 | 6.83 | 6.78 | 6.80 | Adjusted | 17,359 |
| 39 | 6.82 | 6.76 | 6.79 | Adjusted | 17,119 |
| 40 | 6.82 | 6.76 | 6.79 | Adjusted | 22,087 |
| 41 | 6.78 | 6.72 | 6.75 | Adjusted | 17,268 |
| 42 | 6.73 | 6.67 | 6.70 | Adjusted | 21,197 |
| 43 | 6.70 | 6.64 | 6.67 | Adjusted | 19,594 |
| 44 | 6.72 | 6.67 | 6.70 | Adjusted | 19,040 |
| 45 | 6.72 | 6.67 | 6.69 | Adjusted | 22,646 |
| 46 | 6.67 | 6.61 | 6.64 | Adjusted | 21,122 |
| 47 | 6.67 | 6.61 | 6.64 | Adjusted | 22,931 |
| 48 | 6.66 | 6.60 | 6.63 | Adjusted | 23,652 |
| 49 | 6.67 | 6.61 | 6.64 | Adjusted | 23,942 |
| 50 | 6.67 | 6.63 | 6.65 | Adjusted | 31,687 |
| 51 | 6.69 | 6.64 | 6.67 | Adjusted | 24,325 |
| 52 | 6.67 | 6.63 | 6.65 | Adjusted | 29,269 |
| 53 | 6.66 | 6.61 | 6.64 | Adjusted | 27,701 |
| 54 | 6.68 | 6.63 | 6.66 | Adjusted | 27,744 |
| 55 | 6.68 | 6.64 | 6.66 | Adjusted | 30,907 |
| 56 | 6.68 | 6.63 | 6.65 | Adjusted | 28,442 |
| 57 | 6.71 | 6.66 | 6.69 | Adjusted | 29,066 |
| 58 | 6.74 | 6.69 | 6.71 | Adjusted | 29,341 |
| 59 | 6.80 | 6.75 | 6.77 | Adjusted | 28,808 |
| 60 | 6.85 | 6.80 | 6.83 | Adjusted | 33,569 |
| 61 | 6.92 | 6.87 | 6.89 | Adjusted | 26,962 |
| 62 | 6.96 | 6.92 | 6.94 | Adjusted | 32,851 |
| 63 | 7.00 | 6.95 | 6.97 | Adjusted | 29,174 |
| 64 | 7.07 | 7.03 | 7.05 | Adjusted | 28,851 |
| 65 | 7.13 | 7.09 | 7.11 | Adjusted | 31,582 |
| 66 | 7.20 | 7.15 | 7.17 | Adjusted | 25,857 |
| 67 | 7.23 | 7.18 | 7.20 | Adjusted | 25,770 |
| 68 | 7.26 | 7.21 | 7.23 | Adjusted | 23,753 |
| 69 | 7.30 | 7.25 | 7.28 | Adjusted | 21,674 |
| 70 | 7.34 | 7.29 | 7.32 | Adjusted | 25,178 |
| 71 | 7.38 | 7.32 | 7.35 | Adjusted | 19,021 |
| 72 | 7.39 | 7.33 | 7.36 | Adjusted | 19,910 |
| 73 | 7.41 | 7.35 | 7.38 | Adjusted | 17,652 |
| 74 | 7.47 | 7.40 | 7.43 | Adjusted | 16,282 |
| 75 | 7.46 | 7.40 | 7.43 | Adjusted | 17,652 |
| 76 | 7.46 | 7.40 | 7.43 | Adjusted | 14,355 |
| 77 | 7.51 | 7.45 | 7.48 | Adjusted | 13,990 |
| 78 | 7.54 | 7.47 | 7.50 | Adjusted | 13,249 |
| 79 | 7.55 | 7.49 | 7.52 | Adjusted | 11,898 |
| 80 | 7.62 | 7.55 | 7.59 | Adjusted | 13,990 |
| 81 | 7.64 | 7.56 | 7.60 | Adjusted | 10,362 |
| 82 | 7.64 | 7.56 | 7.60 | Adjusted | 10,372 |
| 83 | 7.61 | 7.52 | 7.57 | Adjusted | 9,106 |
| 84 | 7.70 | 7.61 | 7.66 | Adjusted | 7,941 |
| 85 | 7.68 | 7.59 | 7.63 | Adjusted | 7,534 |
| 86 | 7.75 | 7.64 | 7.70 | Adjusted | 5,655 |
| 87 | 7.76 | 7.64 | 7.70 | Adjusted | 4,787 |
| 88 | 7.75 | 7.62 | 7.69 | Adjusted | 4,116 |
| 89 | 7.84 | 7.68 | 7.76 | Adjusted | 2,983 |
| 90 | 7.89 | 7.72 | 7.80 | Adjusted | 2,596 |
| 91 | 8.05 | 7.84 | 7.94 | Adjusted | 1,626 |
| 92 | 7.94 | 7.70 | 7.82 | Adjusted | 1,256 |
| 93 | 7.99 | 7.69 | 7.84 | Adjusted | 802 |
| 94 | 8.04 | 7.67 | 7.86 | Adjusted | 548 |

#### Table 5.2. Age of the lowest point

The lowest point of the ladder profile is at age 54 raw and 48 adjusted for five controls, and a quadratic fit puts it at 44 raw and 47 adjusted

Age of the lowest point of the five-year moving average of the mean ladder score by year of age, or the turning age of a fitted quadratic, with 95% intervals from a bootstrap over counties

| age | conf_high | conf_low | group | label | n |
| ---: | ---: | ---: | --- | --- | ---: |
| 53.90 | 54.30 | 51.50 | Lowest point of the profile | All interviews with a ladder score, raw, 2009 to 2017 | 2,180,433 |
| 54.10 | 54.60 | 51 | Lowest point of the profile | Interviews with complete controls, raw | 1,355,746 |
| 54.10 | 54.60 | 51.10 | Lowest point of the profile | Adjusted for sex | 1,355,746 |
| 53.90 | 54.20 | 48.10 | Lowest point of the profile | Adjusted for marital status | 1,355,746 |
| 54.20 | 54.70 | 48.10 | Lowest point of the profile | Adjusted for education | 1,355,746 |
| 48.20 | 54 | 47.90 | Lowest point of the profile | Adjusted for household income | 1,355,746 |
| 54.10 | 54.60 | 51 | Lowest point of the profile | Adjusted for survey year | 1,355,746 |
| 48.10 | 53.90 | 47.80 | Lowest point of the profile | Adjusted for all five | 1,355,746 |
| 54.10 | 54.60 | 50.90 | Lowest point of the profile | Interviews with a health rating, raw | 1,354,627 |
| 47.70 | 48 | 47 | Lowest point of the profile | Adjusted for the five and self-rated health | 1,354,627 |
| 44.40 | 44.80 | 44 | Quadratic fit | Raw, quadratic fit | 1,355,746 |
| 47.10 | 47.40 | 46.90 | Quadratic fit | Adjusted for all five, quadratic fit | 1,355,746 |
| 54.40 | 55 | 53.90 | Within household income groups, 2009 to 2014 | Household income under the lower cutoff | 485,159 |
| 48.20 | 50.80 | 47.80 | Within household income groups, 2009 to 2014 | Household income between the cutoffs | 567,870 |
| 54.40 | 55.50 | 50.80 | Within county deprivation thirds | Least deprived third of interviews | 718,005 |
| 53.90 | 54.50 | 52.40 | Within county deprivation thirds | Middle third of interviews | 730,419 |
| 51.10 | 53.70 | 48 | Within county deprivation thirds | Most deprived third of interviews | 731,406 |

#### Table 5.3. Fall and rise

The fall in the ladder from ages 25 to 34 to ages 45 to 54 is −0.59 points in households under $36,000 and −0.05 in households of $90,000 or more, where the fall from ages 35 to 44 is −0.07

Change in the mean ladder score between age bands, as the mean of the profile in the later band minus the earlier band, with 95% intervals from a bootstrap over counties

| conf_high | conf_low | estimate | group | label | n |
| ---: | ---: | ---: | --- | --- | ---: |
| −0.189 | −0.219 | −0.205 | Fall, ages 25 to 34 to ages 45 to 54 | All adults, raw | 1,355,746 |
| −0.150 | −0.175 | −0.162 | Fall, ages 35 to 44 to ages 45 to 54 | All adults, raw | 1,355,746 |
| 0.574 | 0.541 | 0.557 | Rise, ages 45 to 54 to ages 75 to 94 | All adults, raw | 1,355,746 |
| −0.303 | −0.333 | −0.318 | Fall, ages 25 to 34 to ages 45 to 54 | All adults, adjusted for five controls | 1,355,746 |
| −0.106 | −0.133 | −0.119 | Fall, ages 35 to 44 to ages 45 to 54 | All adults, adjusted for five controls | 1,355,746 |
| 0.932 | 0.898 | 0.914 | Rise, ages 45 to 54 to ages 75 to 94 | All adults, adjusted for five controls | 1,355,746 |
| −0.559 | −0.618 | −0.589 | Fall, ages 25 to 34 to ages 45 to 54 | Household income under the lower cutoff | 485,159 |
| −0.242 | −0.302 | −0.272 | Fall, ages 35 to 44 to ages 45 to 54 | Household income under the lower cutoff | 485,159 |
| 1.18 | 1.13 | 1.15 | Rise, ages 45 to 54 to ages 75 to 94 | Household income under the lower cutoff | 485,159 |
| −0.255 | −0.290 | −0.273 | Fall, ages 25 to 34 to ages 45 to 54 | Household income between the cutoffs | 567,870 |
| −0.100 | −0.131 | −0.116 | Fall, ages 35 to 44 to ages 45 to 54 | Household income between the cutoffs | 567,870 |
| 0.713 | 0.670 | 0.693 | Rise, ages 45 to 54 to ages 75 to 94 | Household income between the cutoffs | 567,870 |
| −0.029 | −0.069 | −0.050 | Fall, ages 25 to 34 to ages 45 to 54 | Household income above the upper cutoff | 310,559 |
| −0.051 | −0.080 | −0.066 | Fall, ages 35 to 44 to ages 45 to 54 | Household income above the upper cutoff | 310,559 |
| 0.398 | 0.332 | 0.365 | Rise, ages 45 to 54 to ages 75 to 94 | Household income above the upper cutoff | 310,559 |
| −0.366 | −0.473 | −0.419 | Fall, ages 25 to 34 to ages 45 to 54 | Under the lower cutoff, 2015 to 2017 | 126,328 |
| −0.152 | −0.268 | −0.211 | Fall, ages 35 to 44 to ages 45 to 54 | Under the lower cutoff, 2015 to 2017 | 126,328 |
| 1.07 | 0.957 | 1.01 | Rise, ages 45 to 54 to ages 75 to 94 | Under the lower cutoff, 2015 to 2017 | 126,328 |
| −0.141 | −0.207 | −0.174 | Fall, ages 25 to 34 to ages 45 to 54 | Between the cutoffs, 2015 to 2017 | 166,352 |
| −0.059 | −0.126 | −0.093 | Fall, ages 35 to 44 to ages 45 to 54 | Between the cutoffs, 2015 to 2017 | 166,352 |
| 0.793 | 0.723 | 0.759 | Rise, ages 45 to 54 to ages 75 to 94 | Between the cutoffs, 2015 to 2017 | 166,352 |
| 0.122 | 0.065 | 0.093 | Fall, ages 25 to 34 to ages 45 to 54 | Above the upper cutoff, 2015 to 2017 | 139,811 |
| 0.008 | −0.041 | −0.017 | Fall, ages 35 to 44 to ages 45 to 54 | Above the upper cutoff, 2015 to 2017 | 139,811 |
| 0.490 | 0.396 | 0.444 | Rise, ages 45 to 54 to ages 75 to 94 | Above the upper cutoff, 2015 to 2017 | 139,811 |
| −0.085 | −0.139 | −0.113 | Fall, ages 25 to 34 to ages 45 to 54 | Least deprived third of interviews | 718,005 |
| −0.116 | −0.151 | −0.134 | Fall, ages 35 to 44 to ages 45 to 54 | Least deprived third of interviews | 718,005 |
| 0.524 | 0.470 | 0.497 | Rise, ages 45 to 54 to ages 75 to 94 | Least deprived third of interviews | 718,005 |
| −0.116 | −0.163 | −0.139 | Fall, ages 25 to 34 to ages 45 to 54 | Middle third of interviews | 730,419 |
| −0.128 | −0.161 | −0.144 | Fall, ages 35 to 44 to ages 45 to 54 | Middle third of interviews | 730,419 |
| 0.579 | 0.527 | 0.553 | Rise, ages 45 to 54 to ages 75 to 94 | Middle third of interviews | 730,419 |
| −0.153 | −0.193 | −0.172 | Fall, ages 25 to 34 to ages 45 to 54 | Most deprived third of interviews | 731,406 |
| −0.116 | −0.152 | −0.134 | Fall, ages 35 to 44 to ages 45 to 54 | Most deprived third of interviews | 731,406 |
| 0.600 | 0.560 | 0.580 | Rise, ages 45 to 54 to ages 75 to 94 | Most deprived third of interviews | 731,406 |

Note: Each age is a different group of people interviewed in the same years, so a profile mixes age with birth cohort and does not follow anyone through life.

Note: Adjusted rows and income groups use interviews in 2009 to 2014, because the income distribution of reporters shifts in 2015 to 2017; the rows marked 2015 to 2017 are shown separately. Adjusted rows use the 1,355,746 interviews that report sex, marital status, education and household income; the row with self-rated health uses the 1,354,627 interviews that also report health.

Note: The age of the lowest point is searched over ages 20 to 92 on the five-year moving average and refined between whole years; the intervals come from a bootstrap over counties (replicates and seed in the methods note).

Note: The ladder of 2008 is not used. Political questions that came before the ladder lowered it in 2008 and in the first months of 2009, and the extract has no interview month, so early 2009 interviews may carry part of the effect. Scores averaged 6.69 in 2008 and 7.03 in 2009, and the 2009 minus 2008 difference is between 0.29 and 0.38 ladder points in every age band.

Note: Gallup's extract has no survey weights, so every estimate is unweighted and levels can differ from the weighted figures Gallup publishes. County comes from the respondent's ZIP code and is approximate near county lines.

Source: Gallup U.S. Daily, ladder 2009-2017 (2009-2014 where household income is used, 2015-2017 where stated); unweighted, intervals clustered by county or from a bootstrap over counties; county Area Deprivation Index 2023 joined on the county of the interview

Data: https://vishalsingh.org/zipcode/report/well-being-across-the-life-span/figures/ladder_turning_points.json

### Table 6. Relative to adults at the same ages, stress among people born in 1950 to 1959 changed by −9.6 points and their ladder score by +0.22 points between 2009 and 2017, against −10.6 and +0.28 across the same ages in 2009

Birth cohorts of a decade followed across Gallup interview years, for the ladder (2009 to 2017), stress and physical pain (2008 to 2017), with the profile by year of age from all interviews

#### Table 6.1. Ladder

The ladder score of people born in 1950 to 1959 changed by +0.41 points between 2009 and 2017, as they went from ages 50 to 59 to ages 58 to 67, against +0.28 points across the same ages in 2009 and +0.19 points among all adults at ages 50 to 67

Mean Cantril ladder score for six birth cohorts in each interview year from 2009 to 2017, and for all interviews by year of age

| age | conf_high | conf_low | estimate | group | n |
| ---: | ---: | ---: | ---: | --- | ---: |
| 74.10 | 7.33 | 7.29 | 7.31 | Born 1930 to 1939 | 41,505 |
| 75.20 | 7.36 | 7.32 | 7.34 | Born 1930 to 1939 | 46,954 |
| 76.10 | 7.33 | 7.29 | 7.31 | Born 1930 to 1939 | 42,886 |
| 77.10 | 7.36 | 7.31 | 7.33 | Born 1930 to 1939 | 41,176 |
| 78 | 7.41 | 7.34 | 7.38 | Born 1930 to 1939 | 17,029 |
| 78.80 | 7.41 | 7.35 | 7.38 | Born 1930 to 1939 | 17,391 |
| 79.90 | 7.45 | 7.38 | 7.42 | Born 1930 to 1939 | 15,911 |
| 80.80 | 7.48 | 7.41 | 7.45 | Born 1930 to 1939 | 14,280 |
| 81.70 | 7.49 | 7.41 | 7.45 | Born 1930 to 1939 | 10,517 |
| 64.10 | 7.18 | 7.14 | 7.16 | Born 1940 to 1949 | 62,066 |
| 65.30 | 7.20 | 7.17 | 7.18 | Born 1940 to 1949 | 72,110 |
| 66.20 | 7.17 | 7.14 | 7.16 | Born 1940 to 1949 | 68,765 |
| 67.20 | 7.22 | 7.19 | 7.20 | Born 1940 to 1949 | 69,475 |
| 68 | 7.33 | 7.28 | 7.31 | Born 1940 to 1949 | 32,334 |
| 69 | 7.33 | 7.28 | 7.31 | Born 1940 to 1949 | 32,845 |
| 70.10 | 7.41 | 7.36 | 7.38 | Born 1940 to 1949 | 32,230 |
| 71 | 7.46 | 7.41 | 7.44 | Born 1940 to 1949 | 31,175 |
| 71.90 | 7.51 | 7.45 | 7.48 | Born 1940 to 1949 | 24,982 |
| 54.50 | 6.82 | 6.78 | 6.80 | Born 1950 to 1959 | 63,588 |
| 55.70 | 6.83 | 6.80 | 6.82 | Born 1950 to 1959 | 73,724 |
| 56.60 | 6.83 | 6.80 | 6.82 | Born 1950 to 1959 | 71,179 |
| 57.60 | 6.86 | 6.83 | 6.84 | Born 1950 to 1959 | 72,420 |
| 58.60 | 6.91 | 6.86 | 6.89 | Born 1950 to 1959 | 36,053 |
| 59.50 | 6.99 | 6.94 | 6.97 | Born 1950 to 1959 | 35,610 |
| 60.70 | 7.11 | 7.07 | 7.09 | Born 1950 to 1959 | 35,426 |
| 61.60 | 7.17 | 7.12 | 7.14 | Born 1950 to 1959 | 35,866 |
| 62.50 | 7.24 | 7.19 | 7.22 | Born 1950 to 1959 | 31,360 |
| 44.70 | 6.86 | 6.81 | 6.83 | Born 1960 to 1969 | 48,157 |
| 46 | 6.87 | 6.83 | 6.85 | Born 1960 to 1969 | 55,779 |
| 46.90 | 6.83 | 6.78 | 6.80 | Born 1960 to 1969 | 54,906 |
| 48 | 6.81 | 6.76 | 6.78 | Born 1960 to 1969 | 54,915 |
| 48.90 | 6.87 | 6.82 | 6.84 | Born 1960 to 1969 | 28,541 |
| 49.80 | 6.92 | 6.86 | 6.88 | Born 1960 to 1969 | 28,214 |
| 50.90 | 6.98 | 6.93 | 6.95 | Born 1960 to 1969 | 27,677 |
| 51.80 | 7.02 | 6.97 | 7.00 | Born 1960 to 1969 | 29,045 |
| 52.70 | 7.10 | 7.04 | 7.07 | Born 1960 to 1969 | 26,572 |
| 34.80 | 7.00 | 6.95 | 6.98 | Born 1970 to 1979 | 33,172 |
| 36 | 7.06 | 7.01 | 7.04 | Born 1970 to 1979 | 38,714 |
| 36.70 | 6.99 | 6.95 | 6.97 | Born 1970 to 1979 | 39,699 |
| 37.80 | 6.97 | 6.92 | 6.95 | Born 1970 to 1979 | 39,318 |
| 38.70 | 7.00 | 6.94 | 6.97 | Born 1970 to 1979 | 21,440 |
| 39.60 | 7.05 | 6.99 | 7.02 | Born 1970 to 1979 | 20,381 |
| 40.70 | 7.10 | 7.04 | 7.07 | Born 1970 to 1979 | 19,588 |
| 41.70 | 7.12 | 7.05 | 7.08 | Born 1970 to 1979 | 21,500 |
| 42.60 | 7.16 | 7.10 | 7.13 | Born 1970 to 1979 | 20,619 |
| 24.90 | 6.94 | 6.89 | 6.91 | Born 1980 to 1989 | 21,259 |
| 25.90 | 7.04 | 6.98 | 7.01 | Born 1980 to 1989 | 26,119 |
| 26.60 | 6.99 | 6.94 | 6.96 | Born 1980 to 1989 | 34,816 |
| 27.60 | 6.99 | 6.94 | 6.97 | Born 1980 to 1989 | 35,952 |
| 28.60 | 6.99 | 6.93 | 6.96 | Born 1980 to 1989 | 20,750 |
| 29.60 | 7.00 | 6.94 | 6.97 | Born 1980 to 1989 | 19,525 |
| 30.60 | 7.04 | 6.98 | 7.01 | Born 1980 to 1989 | 18,831 |
| 31.60 | 7.06 | 7.01 | 7.04 | Born 1980 to 1989 | 20,494 |
| 32.50 | 7.08 | 7.02 | 7.05 | Born 1980 to 1989 | 21,095 |
| 18 | 7.28 | 7.23 | 7.25 | All interviews, by age | 25,104 |
| 19 | 7.16 | 7.11 | 7.14 | All interviews, by age | 20,764 |
| 20 | 7.06 | 7.01 | 7.03 | All interviews, by age | 20,566 |
| 21 | 7.04 | 6.98 | 7.01 | All interviews, by age | 21,404 |
| 22 | 6.99 | 6.93 | 6.96 | All interviews, by age | 20,267 |
| 23 | 6.93 | 6.88 | 6.91 | All interviews, by age | 20,868 |
| 24 | 6.91 | 6.85 | 6.88 | All interviews, by age | 20,651 |
| 25 | 6.96 | 6.91 | 6.93 | All interviews, by age | 22,756 |
| 26 | 6.95 | 6.89 | 6.92 | All interviews, by age | 20,582 |
| 27 | 6.97 | 6.92 | 6.94 | All interviews, by age | 21,275 |
| 28 | 7.01 | 6.96 | 6.98 | All interviews, by age | 22,607 |
| 29 | 7.02 | 6.97 | 6.99 | All interviews, by age | 21,955 |
| 30 | 7.04 | 6.99 | 7.01 | All interviews, by age | 26,817 |
| 31 | 7.05 | 7.00 | 7.02 | All interviews, by age | 21,509 |
| 32 | 7.03 | 6.98 | 7.00 | All interviews, by age | 25,123 |
| 33 | 7.05 | 7 | 7.03 | All interviews, by age | 22,981 |
| 34 | 7.04 | 6.98 | 7.01 | All interviews, by age | 23,449 |
| 35 | 7.06 | 7.01 | 7.04 | All interviews, by age | 27,289 |
| 36 | 7.06 | 7.01 | 7.04 | All interviews, by age | 23,837 |
| 37 | 7.04 | 6.99 | 7.01 | All interviews, by age | 24,074 |
| 38 | 7.02 | 6.97 | 7.00 | All interviews, by age | 25,474 |
| 39 | 7.05 | 7.00 | 7.02 | All interviews, by age | 25,043 |
| 40 | 7.01 | 6.96 | 6.99 | All interviews, by age | 33,427 |
| 41 | 7.02 | 6.96 | 6.99 | All interviews, by age | 24,465 |
| 42 | 6.97 | 6.92 | 6.94 | All interviews, by age | 30,521 |
| 43 | 6.93 | 6.88 | 6.91 | All interviews, by age | 28,188 |
| 44 | 6.94 | 6.89 | 6.91 | All interviews, by age | 27,722 |
| 45 | 6.94 | 6.88 | 6.91 | All interviews, by age | 34,003 |
| 46 | 6.88 | 6.83 | 6.86 | All interviews, by age | 31,175 |
| 47 | 6.86 | 6.81 | 6.84 | All interviews, by age | 33,763 |
| 48 | 6.87 | 6.82 | 6.84 | All interviews, by age | 34,633 |
| 49 | 6.86 | 6.81 | 6.83 | All interviews, by age | 34,790 |
| 50 | 6.87 | 6.83 | 6.85 | All interviews, by age | 49,187 |
| 51 | 6.87 | 6.82 | 6.84 | All interviews, by age | 35,575 |
| 52 | 6.84 | 6.79 | 6.82 | All interviews, by age | 43,598 |
| 53 | 6.82 | 6.77 | 6.79 | All interviews, by age | 41,271 |
| 54 | 6.84 | 6.80 | 6.82 | All interviews, by age | 42,032 |
| 55 | 6.86 | 6.82 | 6.84 | All interviews, by age | 47,470 |
| 56 | 6.85 | 6.80 | 6.83 | All interviews, by age | 43,108 |
| 57 | 6.87 | 6.83 | 6.85 | All interviews, by age | 44,152 |
| 58 | 6.89 | 6.84 | 6.87 | All interviews, by age | 45,107 |
| 59 | 6.93 | 6.89 | 6.91 | All interviews, by age | 44,432 |
| 60 | 7.00 | 6.96 | 6.98 | All interviews, by age | 54,582 |
| 61 | 7.03 | 6.99 | 7.01 | All interviews, by age | 41,935 |
| 62 | 7.09 | 7.05 | 7.07 | All interviews, by age | 51,583 |
| 63 | 7.12 | 7.08 | 7.10 | All interviews, by age | 46,017 |
| 64 | 7.18 | 7.13 | 7.16 | All interviews, by age | 46,280 |
| 65 | 7.22 | 7.19 | 7.21 | All interviews, by age | 53,072 |
| 66 | 7.28 | 7.24 | 7.26 | All interviews, by age | 43,111 |
| 67 | 7.30 | 7.26 | 7.28 | All interviews, by age | 44,121 |
| 68 | 7.31 | 7.27 | 7.29 | All interviews, by age | 42,066 |
| 69 | 7.34 | 7.30 | 7.32 | All interviews, by age | 38,080 |
| 70 | 7.37 | 7.33 | 7.35 | All interviews, by age | 46,117 |
| 71 | 7.36 | 7.31 | 7.33 | All interviews, by age | 33,149 |
| 72 | 7.37 | 7.32 | 7.34 | All interviews, by age | 35,737 |
| 73 | 7.37 | 7.32 | 7.34 | All interviews, by age | 32,279 |
| 74 | 7.38 | 7.33 | 7.36 | All interviews, by age | 29,803 |
| 75 | 7.38 | 7.33 | 7.36 | All interviews, by age | 32,359 |
| 76 | 7.38 | 7.33 | 7.35 | All interviews, by age | 26,175 |
| 77 | 7.38 | 7.33 | 7.36 | All interviews, by age | 25,447 |
| 78 | 7.39 | 7.33 | 7.36 | All interviews, by age | 24,365 |
| 79 | 7.40 | 7.35 | 7.37 | All interviews, by age | 21,647 |
| 80 | 7.44 | 7.39 | 7.41 | All interviews, by age | 26,170 |
| 81 | 7.44 | 7.38 | 7.41 | All interviews, by age | 18,844 |
| 82 | 7.42 | 7.35 | 7.38 | All interviews, by age | 19,071 |
| 83 | 7.38 | 7.31 | 7.35 | All interviews, by age | 17,079 |
| 84 | 7.42 | 7.35 | 7.39 | All interviews, by age | 14,958 |
| 85 | 7.39 | 7.32 | 7.36 | All interviews, by age | 14,462 |
| 86 | 7.41 | 7.33 | 7.37 | All interviews, by age | 10,947 |
| 87 | 7.41 | 7.32 | 7.37 | All interviews, by age | 9,423 |
| 88 | 7.42 | 7.33 | 7.38 | All interviews, by age | 8,210 |
| 89 | 7.47 | 7.36 | 7.42 | All interviews, by age | 5,902 |
| 90 | 7.58 | 7.47 | 7.52 | All interviews, by age | 5,455 |
| 91 | 7.57 | 7.43 | 7.50 | All interviews, by age | 3,335 |
| 92 | 7.53 | 7.36 | 7.45 | All interviews, by age | 2,641 |
| 93 | 7.56 | 7.36 | 7.46 | All interviews, by age | 1,754 |
| 94 | 7.61 | 7.35 | 7.48 | All interviews, by age | 1,243 |

#### Table 6.2. Stress

Stress among people born in 1950 to 1959 changed by −7.9 points between 2009 and 2017, against −10.6 points across the same ages in 2009 and +1.7 points among all adults at ages 50 to 67

Share of adults who reported stress during a lot of the day yesterday for six birth cohorts in each interview year from 2008 to 2017, and for all interviews by year of age

| age | conf_high | conf_low | estimate | group | n |
| ---: | ---: | ---: | ---: | --- | ---: |
| 73.20 | 19.52 | 18.71 | 19.11 | Born 1930 to 1939 | 45,875 |
| 74.10 | 19.96 | 19.20 | 19.58 | Born 1930 to 1939 | 48,535 |
| 75.20 | 19.63 | 18.79 | 19.21 | Born 1930 to 1939 | 47,215 |
| 76.10 | 20.17 | 19.42 | 19.79 | Born 1930 to 1939 | 43,159 |
| 77.10 | 19.84 | 19.04 | 19.44 | Born 1930 to 1939 | 41,441 |
| 78 | 19.46 | 18.26 | 18.86 | Born 1930 to 1939 | 17,150 |
| 78.80 | 19.66 | 18.47 | 19.06 | Born 1930 to 1939 | 17,546 |
| 79.90 | 19.47 | 18.19 | 18.83 | Born 1930 to 1939 | 16,054 |
| 80.80 | 19.19 | 17.85 | 18.52 | Born 1930 to 1939 | 14,401 |
| 81.70 | 20.71 | 19.12 | 19.92 | Born 1930 to 1939 | 10,599 |
| 63.20 | 29.52 | 28.76 | 29.14 | Born 1940 to 1949 | 69,436 |
| 64.10 | 29.04 | 28.17 | 28.61 | Born 1940 to 1949 | 72,158 |
| 65.30 | 27.68 | 26.96 | 27.32 | Born 1940 to 1949 | 72,279 |
| 66.20 | 27.21 | 26.48 | 26.85 | Born 1940 to 1949 | 68,892 |
| 67.20 | 26.14 | 25.41 | 25.77 | Born 1940 to 1949 | 69,677 |
| 68 | 24.71 | 23.70 | 24.21 | Born 1940 to 1949 | 32,462 |
| 69 | 24.10 | 23.15 | 23.63 | Born 1940 to 1949 | 32,955 |
| 70.10 | 23.43 | 22.39 | 22.91 | Born 1940 to 1949 | 32,326 |
| 71 | 23.05 | 22 | 22.53 | Born 1940 to 1949 | 31,275 |
| 71.90 | 24.80 | 23.72 | 24.26 | Born 1940 to 1949 | 25,053 |
| 53.50 | 41.07 | 40.31 | 40.69 | Born 1950 to 1959 | 75,994 |
| 54.50 | 42.55 | 41.79 | 42.17 | Born 1950 to 1959 | 74,461 |
| 55.70 | 42.08 | 41.30 | 41.69 | Born 1950 to 1959 | 73,817 |
| 56.60 | 41.27 | 40.51 | 40.89 | Born 1950 to 1959 | 71,287 |
| 57.60 | 39.97 | 39.21 | 39.59 | Born 1950 to 1959 | 72,548 |
| 58.60 | 38.97 | 37.92 | 38.45 | Born 1950 to 1959 | 36,117 |
| 59.50 | 37.66 | 36.60 | 37.13 | Born 1950 to 1959 | 35,666 |
| 60.70 | 36.04 | 34.94 | 35.49 | Born 1950 to 1959 | 35,466 |
| 61.60 | 34.29 | 33.24 | 33.76 | Born 1950 to 1959 | 35,937 |
| 62.50 | 34.89 | 33.70 | 34.30 | Born 1950 to 1959 | 31,409 |
| 43.80 | 45.09 | 44.26 | 44.67 | Born 1960 to 1969 | 59,220 |
| 44.70 | 47.27 | 46.33 | 46.80 | Born 1960 to 1969 | 56,539 |
| 46 | 46.58 | 45.72 | 46.15 | Born 1960 to 1969 | 55,838 |
| 46.90 | 46.84 | 45.89 | 46.36 | Born 1960 to 1969 | 54,976 |
| 48 | 46.29 | 45.33 | 45.81 | Born 1960 to 1969 | 55,023 |
| 48.90 | 45.87 | 44.61 | 45.24 | Born 1960 to 1969 | 28,602 |
| 49.80 | 44.80 | 43.58 | 44.19 | Born 1960 to 1969 | 28,237 |
| 50.90 | 44.41 | 43.27 | 43.84 | Born 1960 to 1969 | 27,717 |
| 51.80 | 43.86 | 42.63 | 43.24 | Born 1960 to 1969 | 29,100 |
| 52.70 | 44.03 | 42.80 | 43.42 | Born 1960 to 1969 | 26,619 |
| 33.90 | 47.89 | 46.81 | 47.35 | Born 1970 to 1979 | 42,406 |
| 34.80 | 49.14 | 48.03 | 48.59 | Born 1970 to 1979 | 39,045 |
| 36 | 47.89 | 46.83 | 47.36 | Born 1970 to 1979 | 38,767 |
| 36.70 | 49.24 | 48.15 | 48.69 | Born 1970 to 1979 | 39,747 |
| 37.80 | 49.07 | 47.87 | 48.47 | Born 1970 to 1979 | 39,394 |
| 38.70 | 47.94 | 46.55 | 47.24 | Born 1970 to 1979 | 21,480 |
| 39.60 | 48.14 | 46.69 | 47.41 | Born 1970 to 1979 | 20,405 |
| 40.70 | 47.20 | 45.75 | 46.47 | Born 1970 to 1979 | 19,612 |
| 41.70 | 46.88 | 45.38 | 46.13 | Born 1970 to 1979 | 21,542 |
| 42.60 | 46.57 | 45.25 | 45.91 | Born 1970 to 1979 | 20,662 |
| 24 | 48.40 | 47.16 | 47.78 | Born 1980 to 1989 | 25,752 |
| 24.90 | 49.23 | 47.91 | 48.57 | Born 1980 to 1989 | 24,886 |
| 25.90 | 47.64 | 46.33 | 46.98 | Born 1980 to 1989 | 26,177 |
| 26.60 | 48.81 | 47.77 | 48.29 | Born 1980 to 1989 | 34,882 |
| 27.60 | 48.81 | 47.52 | 48.16 | Born 1980 to 1989 | 36,021 |
| 28.60 | 49.51 | 48.06 | 48.79 | Born 1980 to 1989 | 20,795 |
| 29.60 | 48.85 | 47.39 | 48.12 | Born 1980 to 1989 | 19,548 |
| 30.60 | 49.68 | 48.27 | 48.98 | Born 1980 to 1989 | 18,860 |
| 31.60 | 49.65 | 48.22 | 48.93 | Born 1980 to 1989 | 20,528 |
| 32.50 | 50.19 | 48.72 | 49.46 | Born 1980 to 1989 | 21,126 |
| 18 | 42.52 | 41.08 | 41.80 | All interviews, by age | 28,568 |
| 19 | 45.50 | 44.04 | 44.77 | All interviews, by age | 23,262 |
| 20 | 47.90 | 46.48 | 47.19 | All interviews, by age | 23,022 |
| 21 | 48.51 | 47.10 | 47.81 | All interviews, by age | 23,787 |
| 22 | 49.34 | 47.93 | 48.64 | All interviews, by age | 22,821 |
| 23 | 48.63 | 47.30 | 47.96 | All interviews, by age | 23,605 |
| 24 | 49.67 | 48.35 | 49.01 | All interviews, by age | 23,454 |
| 25 | 49.11 | 47.73 | 48.42 | All interviews, by age | 26,036 |
| 26 | 49.86 | 48.41 | 49.13 | All interviews, by age | 23,861 |
| 27 | 50.14 | 48.81 | 49.48 | All interviews, by age | 24,771 |
| 28 | 49.04 | 47.75 | 48.39 | All interviews, by age | 26,630 |
| 29 | 49.05 | 47.82 | 48.43 | All interviews, by age | 25,891 |
| 30 | 47.37 | 46.10 | 46.73 | All interviews, by age | 31,457 |
| 31 | 50.22 | 48.90 | 49.56 | All interviews, by age | 25,455 |
| 32 | 49.47 | 48.16 | 48.82 | All interviews, by age | 29,613 |
| 33 | 49.37 | 48.13 | 48.75 | All interviews, by age | 27,392 |
| 34 | 49.04 | 47.78 | 48.41 | All interviews, by age | 27,960 |
| 35 | 48.01 | 46.84 | 47.43 | All interviews, by age | 32,600 |
| 36 | 48.25 | 47.02 | 47.64 | All interviews, by age | 28,987 |
| 37 | 48.95 | 47.68 | 48.31 | All interviews, by age | 30,020 |
| 38 | 48.55 | 47.32 | 47.94 | All interviews, by age | 31,620 |
| 39 | 48.36 | 47.13 | 47.75 | All interviews, by age | 30,615 |
| 40 | 46.75 | 45.61 | 46.18 | All interviews, by age | 40,304 |
| 41 | 48.57 | 47.28 | 47.93 | All interviews, by age | 30,003 |
| 42 | 47.35 | 46.27 | 46.81 | All interviews, by age | 37,111 |
| 43 | 47.65 | 46.56 | 47.11 | All interviews, by age | 34,684 |
| 44 | 47.31 | 46.18 | 46.75 | All interviews, by age | 34,226 |
| 45 | 45.60 | 44.55 | 45.08 | All interviews, by age | 41,584 |
| 46 | 46.64 | 45.64 | 46.14 | All interviews, by age | 38,180 |
| 47 | 46.34 | 45.34 | 45.84 | All interviews, by age | 41,478 |
| 48 | 46.07 | 45.04 | 45.56 | All interviews, by age | 42,378 |
| 49 | 45.86 | 44.88 | 45.37 | All interviews, by age | 42,564 |
| 50 | 43.78 | 42.90 | 43.34 | All interviews, by age | 59,277 |
| 51 | 45.16 | 44.17 | 44.66 | All interviews, by age | 43,409 |
| 52 | 44.27 | 43.36 | 43.82 | All interviews, by age | 52,483 |
| 53 | 44.61 | 43.68 | 44.14 | All interviews, by age | 49,715 |
| 54 | 43.44 | 42.55 | 42.99 | All interviews, by age | 50,958 |
| 55 | 42 | 41.18 | 41.59 | All interviews, by age | 56,778 |
| 56 | 41.92 | 41.02 | 41.47 | All interviews, by age | 51,513 |
| 57 | 41.04 | 40.17 | 40.61 | All interviews, by age | 52,848 |
| 58 | 40.21 | 39.31 | 39.76 | All interviews, by age | 54,058 |
| 59 | 38.88 | 38.02 | 38.45 | All interviews, by age | 53,028 |
| 60 | 36.30 | 35.47 | 35.88 | All interviews, by age | 65,039 |
| 61 | 35.63 | 34.67 | 35.15 | All interviews, by age | 50,826 |
| 62 | 33.38 | 32.48 | 32.93 | All interviews, by age | 60,276 |
| 63 | 32.23 | 31.36 | 31.80 | All interviews, by age | 52,997 |
| 64 | 30.20 | 29.39 | 29.79 | All interviews, by age | 53,732 |
| 65 | 27.41 | 26.56 | 26.98 | All interviews, by age | 62,199 |
| 66 | 26.22 | 25.37 | 25.79 | All interviews, by age | 50,147 |
| 67 | 25.36 | 24.44 | 24.90 | All interviews, by age | 50,933 |
| 68 | 23.94 | 23.06 | 23.50 | All interviews, by age | 48,661 |
| 69 | 23.29 | 22.39 | 22.84 | All interviews, by age | 43,876 |
| 70 | 22 | 21.30 | 21.65 | All interviews, by age | 53,287 |
| 71 | 21.72 | 20.84 | 21.28 | All interviews, by age | 38,553 |
| 72 | 21.21 | 20.41 | 20.81 | All interviews, by age | 41,693 |
| 73 | 20.79 | 19.93 | 20.36 | All interviews, by age | 37,746 |
| 74 | 20.65 | 19.79 | 20.22 | All interviews, by age | 34,729 |
| 75 | 19.90 | 19.06 | 19.48 | All interviews, by age | 37,917 |
| 76 | 19.96 | 19.02 | 19.49 | All interviews, by age | 30,825 |
| 77 | 20.22 | 19.24 | 19.73 | All interviews, by age | 30,161 |
| 78 | 19.32 | 18.39 | 18.86 | All interviews, by age | 29,100 |
| 79 | 19.14 | 18.16 | 18.65 | All interviews, by age | 25,570 |
| 80 | 18.45 | 17.53 | 17.99 | All interviews, by age | 31,115 |
| 81 | 18.65 | 17.62 | 18.14 | All interviews, by age | 22,419 |
| 82 | 18 | 17.01 | 17.50 | All interviews, by age | 22,663 |
| 83 | 18.22 | 17.13 | 17.68 | All interviews, by age | 20,214 |
| 84 | 17.62 | 16.55 | 17.08 | All interviews, by age | 17,771 |
| 85 | 17.44 | 16.34 | 16.89 | All interviews, by age | 17,078 |
| 86 | 17.24 | 15.88 | 16.56 | All interviews, by age | 13,037 |
| 87 | 16.66 | 15.29 | 15.97 | All interviews, by age | 11,068 |
| 88 | 16.03 | 14.62 | 15.32 | All interviews, by age | 9,776 |
| 89 | 16.60 | 14.80 | 15.70 | All interviews, by age | 6,814 |
| 90 | 15.78 | 14.09 | 14.93 | All interviews, by age | 6,396 |
| 91 | 14.95 | 12.67 | 13.81 | All interviews, by age | 3,896 |
| 92 | 15.09 | 12.60 | 13.84 | All interviews, by age | 3,106 |
| 93 | 15.41 | 12.20 | 13.80 | All interviews, by age | 2,050 |
| 94 | 13.64 | 10.18 | 11.91 | All interviews, by age | 1,444 |

#### Table 6.3. Physical pain

Physical pain among people born in 1950 to 1959 changed by +5.9 points between 2009 and 2017, against +5.1 points among all adults at ages 50 to 67 and 0.0 points across the same ages in 2009

Share of adults who reported physical pain during a lot of the day yesterday for six birth cohorts in each interview year from 2008 to 2017, and for all interviews by year of age

| age | conf_high | conf_low | estimate | group | n |
| ---: | ---: | ---: | ---: | --- | ---: |
| 73.20 | 24.88 | 24.04 | 24.46 | Born 1930 to 1939 | 45,858 |
| 74.10 | 25.69 | 24.84 | 25.26 | Born 1930 to 1939 | 48,518 |
| 75.20 | 25.60 | 24.73 | 25.16 | Born 1930 to 1939 | 47,189 |
| 76.10 | 26.12 | 25.22 | 25.67 | Born 1930 to 1939 | 43,121 |
| 77.10 | 26.41 | 25.53 | 25.97 | Born 1930 to 1939 | 41,386 |
| 78 | 26.46 | 25.11 | 25.78 | Born 1930 to 1939 | 17,139 |
| 78.80 | 28.06 | 26.69 | 27.37 | Born 1930 to 1939 | 17,516 |
| 79.90 | 27.52 | 26.13 | 26.82 | Born 1930 to 1939 | 16,024 |
| 80.80 | 28.56 | 26.99 | 27.77 | Born 1930 to 1939 | 14,382 |
| 81.70 | 30.53 | 28.76 | 29.64 | Born 1930 to 1939 | 10,589 |
| 63.20 | 26.74 | 25.88 | 26.31 | Born 1940 to 1949 | 69,474 |
| 64.10 | 26.66 | 25.86 | 26.26 | Born 1940 to 1949 | 72,179 |
| 65.30 | 27.04 | 26.18 | 26.61 | Born 1940 to 1949 | 72,272 |
| 66.20 | 27.31 | 26.51 | 26.91 | Born 1940 to 1949 | 68,891 |
| 67.20 | 26.77 | 25.90 | 26.34 | Born 1940 to 1949 | 69,645 |
| 68 | 27.15 | 26.09 | 26.62 | Born 1940 to 1949 | 32,433 |
| 69 | 27.68 | 26.58 | 27.13 | Born 1940 to 1949 | 32,925 |
| 70.10 | 28.32 | 27.27 | 27.80 | Born 1940 to 1949 | 32,304 |
| 71 | 28.19 | 27.14 | 27.67 | Born 1940 to 1949 | 31,270 |
| 71.90 | 31.95 | 30.69 | 31.32 | Born 1940 to 1949 | 25,043 |
| 53.50 | 25.88 | 24.99 | 25.44 | Born 1950 to 1959 | 76,023 |
| 54.50 | 27.20 | 26.24 | 26.72 | Born 1950 to 1959 | 74,452 |
| 55.70 | 28.13 | 27.14 | 27.64 | Born 1950 to 1959 | 73,850 |
| 56.60 | 28.55 | 27.62 | 28.08 | Born 1950 to 1959 | 71,280 |
| 57.60 | 28.30 | 27.37 | 27.83 | Born 1950 to 1959 | 72,546 |
| 58.60 | 30.65 | 29.46 | 30.05 | Born 1950 to 1959 | 36,103 |
| 59.50 | 31.01 | 29.81 | 30.41 | Born 1950 to 1959 | 35,677 |
| 60.70 | 30.48 | 29.28 | 29.88 | Born 1950 to 1959 | 35,475 |
| 61.60 | 29.90 | 28.77 | 29.34 | Born 1950 to 1959 | 35,949 |
| 62.50 | 33.22 | 31.94 | 32.58 | Born 1950 to 1959 | 31,405 |
| 43.80 | 20.99 | 20.09 | 20.54 | Born 1960 to 1969 | 59,244 |
| 44.70 | 22.08 | 21.20 | 21.64 | Born 1960 to 1969 | 56,561 |
| 46 | 23.09 | 22.12 | 22.60 | Born 1960 to 1969 | 55,851 |
| 46.90 | 24.20 | 23.23 | 23.72 | Born 1960 to 1969 | 54,985 |
| 48 | 24.66 | 23.59 | 24.13 | Born 1960 to 1969 | 55,019 |
| 48.90 | 26.41 | 25.17 | 25.79 | Born 1960 to 1969 | 28,607 |
| 49.80 | 26.75 | 25.38 | 26.07 | Born 1960 to 1969 | 28,269 |
| 50.90 | 27.87 | 26.50 | 27.19 | Born 1960 to 1969 | 27,720 |
| 51.80 | 27.56 | 26.21 | 26.89 | Born 1960 to 1969 | 29,098 |
| 52.70 | 30.77 | 29.32 | 30.04 | Born 1960 to 1969 | 26,619 |
| 33.90 | 16.58 | 15.82 | 16.20 | Born 1970 to 1979 | 42,416 |
| 34.80 | 18 | 17.13 | 17.56 | Born 1970 to 1979 | 39,061 |
| 36 | 17.97 | 17.08 | 17.53 | Born 1970 to 1979 | 38,777 |
| 36.70 | 18.87 | 17.95 | 18.41 | Born 1970 to 1979 | 39,774 |
| 37.80 | 19.06 | 18.01 | 18.54 | Born 1970 to 1979 | 39,411 |
| 38.70 | 20.73 | 19.54 | 20.13 | Born 1970 to 1979 | 21,485 |
| 39.60 | 21.37 | 20.11 | 20.74 | Born 1970 to 1979 | 20,414 |
| 40.70 | 22.04 | 20.71 | 21.37 | Born 1970 to 1979 | 19,617 |
| 41.70 | 21.72 | 20.34 | 21.03 | Born 1970 to 1979 | 21,548 |
| 42.60 | 24.43 | 23.01 | 23.72 | Born 1970 to 1979 | 20,663 |
| 24 | 16.49 | 15.52 | 16 | Born 1980 to 1989 | 25,759 |
| 24.90 | 16.45 | 15.52 | 15.99 | Born 1980 to 1989 | 24,888 |
| 25.90 | 16.22 | 15.26 | 15.74 | Born 1980 to 1989 | 26,168 |
| 26.60 | 17.15 | 16.29 | 16.72 | Born 1980 to 1989 | 34,879 |
| 27.60 | 16.91 | 15.87 | 16.39 | Born 1980 to 1989 | 36,016 |
| 28.60 | 17.57 | 16.36 | 16.96 | Born 1980 to 1989 | 20,796 |
| 29.60 | 17.97 | 16.82 | 17.40 | Born 1980 to 1989 | 19,556 |
| 30.60 | 17.88 | 16.64 | 17.26 | Born 1980 to 1989 | 18,854 |
| 31.60 | 18.60 | 17.45 | 18.03 | Born 1980 to 1989 | 20,536 |
| 32.50 | 20.84 | 19.46 | 20.15 | Born 1980 to 1989 | 21,139 |
| 18 | 14.98 | 14.21 | 14.59 | All interviews, by age | 28,574 |
| 19 | 15.53 | 14.63 | 15.08 | All interviews, by age | 23,261 |
| 20 | 16.19 | 15.21 | 15.70 | All interviews, by age | 23,027 |
| 21 | 16.13 | 15.12 | 15.63 | All interviews, by age | 23,787 |
| 22 | 16.41 | 15.39 | 15.90 | All interviews, by age | 22,818 |
| 23 | 16.14 | 15.11 | 15.62 | All interviews, by age | 23,611 |
| 24 | 16.64 | 15.58 | 16.11 | All interviews, by age | 23,454 |
| 25 | 16.70 | 15.67 | 16.19 | All interviews, by age | 26,046 |
| 26 | 16.59 | 15.54 | 16.06 | All interviews, by age | 23,861 |
| 27 | 17.39 | 16.33 | 16.86 | All interviews, by age | 24,770 |
| 28 | 17.06 | 16.09 | 16.57 | All interviews, by age | 26,621 |
| 29 | 17.21 | 16.22 | 16.71 | All interviews, by age | 25,890 |
| 30 | 17.31 | 16.36 | 16.83 | All interviews, by age | 31,454 |
| 31 | 17.80 | 16.70 | 17.25 | All interviews, by age | 25,454 |
| 32 | 18.09 | 17.13 | 17.61 | All interviews, by age | 29,622 |
| 33 | 18.26 | 17.22 | 17.74 | All interviews, by age | 27,400 |
| 34 | 18.09 | 17.06 | 17.58 | All interviews, by age | 27,960 |
| 35 | 18.25 | 17.28 | 17.76 | All interviews, by age | 32,611 |
| 36 | 18.91 | 17.90 | 18.41 | All interviews, by age | 28,995 |
| 37 | 19.27 | 18.28 | 18.77 | All interviews, by age | 30,039 |
| 38 | 19.02 | 18.07 | 18.55 | All interviews, by age | 31,640 |
| 39 | 19.61 | 18.53 | 19.07 | All interviews, by age | 30,629 |
| 40 | 19.67 | 18.74 | 19.20 | All interviews, by age | 40,310 |
| 41 | 20.37 | 19.32 | 19.85 | All interviews, by age | 30,012 |
| 42 | 20.76 | 19.73 | 20.25 | All interviews, by age | 37,134 |
| 43 | 21.82 | 20.81 | 21.31 | All interviews, by age | 34,706 |
| 44 | 22.33 | 21.17 | 21.75 | All interviews, by age | 34,244 |
| 45 | 22.23 | 21.19 | 21.71 | All interviews, by age | 41,586 |
| 46 | 23.47 | 22.41 | 22.94 | All interviews, by age | 38,198 |
| 47 | 24.38 | 23.27 | 23.83 | All interviews, by age | 41,474 |
| 48 | 25.31 | 24.25 | 24.78 | All interviews, by age | 42,376 |
| 49 | 25.66 | 24.59 | 25.12 | All interviews, by age | 42,562 |
| 50 | 25.61 | 24.57 | 25.09 | All interviews, by age | 59,291 |
| 51 | 27.47 | 26.29 | 26.88 | All interviews, by age | 43,410 |
| 52 | 26.88 | 25.80 | 26.34 | All interviews, by age | 52,500 |
| 53 | 28.15 | 27.05 | 27.60 | All interviews, by age | 49,738 |
| 54 | 28.39 | 27.34 | 27.86 | All interviews, by age | 50,954 |
| 55 | 28.16 | 27.12 | 27.64 | All interviews, by age | 56,786 |
| 56 | 29.02 | 27.98 | 28.50 | All interviews, by age | 51,524 |
| 57 | 29.27 | 28.22 | 28.74 | All interviews, by age | 52,831 |
| 58 | 29.50 | 28.51 | 29 | All interviews, by age | 54,069 |
| 59 | 29.48 | 28.46 | 28.97 | All interviews, by age | 53,045 |
| 60 | 28.67 | 27.69 | 28.18 | All interviews, by age | 65,070 |
| 61 | 29.39 | 28.40 | 28.90 | All interviews, by age | 50,842 |
| 62 | 29.29 | 28.36 | 28.83 | All interviews, by age | 60,276 |
| 63 | 29.33 | 28.37 | 28.85 | All interviews, by age | 53,010 |
| 64 | 28.36 | 27.42 | 27.89 | All interviews, by age | 53,742 |
| 65 | 26.96 | 26.07 | 26.51 | All interviews, by age | 62,191 |
| 66 | 27.15 | 26.26 | 26.71 | All interviews, by age | 50,128 |
| 67 | 27.40 | 26.52 | 26.96 | All interviews, by age | 50,937 |
| 68 | 26.92 | 25.96 | 26.44 | All interviews, by age | 48,643 |
| 69 | 27.01 | 26.08 | 26.54 | All interviews, by age | 43,860 |
| 70 | 25.76 | 24.88 | 25.32 | All interviews, by age | 53,255 |
| 71 | 26.55 | 25.59 | 26.07 | All interviews, by age | 38,542 |
| 72 | 26.42 | 25.46 | 25.94 | All interviews, by age | 41,689 |
| 73 | 26.54 | 25.59 | 26.07 | All interviews, by age | 37,738 |
| 74 | 26.55 | 25.54 | 26.04 | All interviews, by age | 34,693 |
| 75 | 26.47 | 25.54 | 26.01 | All interviews, by age | 37,873 |
| 76 | 27.11 | 26.11 | 26.61 | All interviews, by age | 30,802 |
| 77 | 27.23 | 26.15 | 26.69 | All interviews, by age | 30,148 |
| 78 | 26.90 | 25.84 | 26.37 | All interviews, by age | 29,063 |
| 79 | 27.05 | 25.91 | 26.48 | All interviews, by age | 25,544 |
| 80 | 25.98 | 24.98 | 25.48 | All interviews, by age | 31,099 |
| 81 | 27.08 | 25.88 | 26.48 | All interviews, by age | 22,405 |
| 82 | 27.61 | 26.41 | 27.01 | All interviews, by age | 22,630 |
| 83 | 27.75 | 26.48 | 27.12 | All interviews, by age | 20,190 |
| 84 | 27.44 | 26.10 | 26.77 | All interviews, by age | 17,760 |
| 85 | 27.04 | 25.74 | 26.39 | All interviews, by age | 17,070 |
| 86 | 27.71 | 26.23 | 26.97 | All interviews, by age | 13,006 |
| 87 | 28.73 | 27.05 | 27.89 | All interviews, by age | 11,042 |
| 88 | 26.85 | 25.16 | 26.01 | All interviews, by age | 9,766 |
| 89 | 27.81 | 25.54 | 26.67 | All interviews, by age | 6,827 |
| 90 | 25.95 | 23.82 | 24.89 | All interviews, by age | 6,381 |
| 91 | 27.16 | 24.38 | 25.77 | All interviews, by age | 3,896 |
| 92 | 27.92 | 24.78 | 26.35 | All interviews, by age | 3,101 |
| 93 | 26.80 | 23.17 | 24.99 | All interviews, by age | 2,045 |
| 94 | 26.09 | 21.57 | 23.83 | All interviews, by age | 1,448 |

Note: A cohort is the interview year minus age, in groups of ten birth years. The cohorts shown were aged 18 to 94 in every interview year from 2008 to 2017; cohorts born before 1930 leave the age range and those born after 1989 enter it, so they are left out. The comparison of the 1950 to 1959 cohort starts in 2009 for every measure, because the ladder of 2008 is not used.

Note: The comparison group is all adults at ages 50 to 67, averaging the change at each single year of age, because the interview sample was younger in 2017 (mean age 52.3) than in 2009 (55.1), and a younger sample would bias a change in all adults. All adults include the cohort itself, and the correction cannot separate aging from the changes that affected every age.

Note: The ladder of 2008 is not used. Political questions that came before the ladder lowered it in 2008 and in the first months of 2009, and the extract has no interview month, so early 2009 interviews may carry part of the effect. Scores averaged 6.69 in 2008 and 7.03 in 2009, and the 2009 minus 2008 difference is between 0.29 and 0.38 ladder points in every age band.

Note: Gallup's extract has no survey weights, so every estimate is unweighted and levels can differ from the weighted figures Gallup publishes. County comes from the respondent's ZIP code and is approximate near county lines.

Source: Gallup U.S. Daily, ladder 2009-2017 and feelings 2008-2017; unweighted, intervals clustered by county

Data: https://vishalsingh.org/zipcode/report/well-being-across-the-life-span/figures/cohort_lines.json

## All estimates

Every fact in facts.json (151), sorted by id.

| Fact id | Value | Unit | Estimand | Population | Period | Interval | n | Weights | Source | Note |
| --- | ---: | --- | --- | --- | --- | --- | ---: | --- | --- | --- |
| age_18 | 18 | years of age | definition used in this chapter, not an estimate | an age used in this chapter | ladder interviews in 2009 to 2017 and feelings interviews in 2008 to 2017 |   |   | none (definition) | Gallup U.S. Daily Poll |   |
| age_20 | 20 | years of age | definition used in this chapter, not an estimate | an age used in this chapter | ladder interviews in 2009 to 2017 and feelings interviews in 2008 to 2017 |   |   | none (definition) | Gallup U.S. Daily Poll |   |
| age_25 | 25 | years of age | definition used in this chapter, not an estimate | an age used in this chapter | ladder interviews in 2009 to 2017 and feelings interviews in 2008 to 2017 |   |   | none (definition) | Gallup U.S. Daily Poll |   |
| age_46 | 46 | years of age | definition used in this chapter, not an estimate | an age used in this chapter | ladder interviews in 2009 to 2017 and feelings interviews in 2008 to 2017 |   |   | none (definition) | Gallup U.S. Daily Poll |   |
| age_56 | 56 | years of age | definition used in this chapter, not an estimate | an age used in this chapter | ladder interviews in 2009 to 2017 and feelings interviews in 2008 to 2017 |   |   | none (definition) | Gallup U.S. Daily Poll |   |
| age_94 | 94 | years of age | definition used in this chapter, not an estimate | an age used in this chapter | ladder interviews in 2009 to 2017 and feelings interviews in 2008 to 2017 |   |   | none (definition) | Gallup U.S. Daily Poll |   |
| anger_1824 | 16.5% | percent of adults who reported the feeling during a lot of the day yesterday | unweighted share of interviews that answered yes to the item, interval clustered by the respondent's county | adults aged 18 to 94 interviewed by Gallup who answered the item and whose county is known (age 18 to 24), anger | interviews in 2008 to 2013 and 2016 | 95% CI 16.3% to 16.8% | 125,886 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| anger_7594 | 6.2% | percent of adults who reported the feeling during a lot of the day yesterday | unweighted share of interviews that answered yes to the item, interval clustered by the respondent's county | adults aged 18 to 94 interviewed by Gallup who answered the item and whose county is known (age 75 to 94), anger | interviews in 2008 to 2013 and 2016 | 95% CI 6.0% to 6.3% | 276,770 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| band_1824 | 18 to 24 | years of age | definition used in this chapter, not an estimate | an age group of adults defined in this chapter | ladder interviews in 2009 to 2017 and feelings interviews in 2008 to 2017 |   |   | none (definition) | Gallup U.S. Daily Poll |   |
| band_2534 | 25 to 34 | years of age | definition used in this chapter, not an estimate | an age group of adults defined in this chapter | ladder interviews in 2009 to 2017 and feelings interviews in 2008 to 2017 |   |   | none (definition) | Gallup U.S. Daily Poll |   |
| band_3544 | 35 to 44 | years of age | definition used in this chapter, not an estimate | an age group of adults defined in this chapter | ladder interviews in 2009 to 2017 and feelings interviews in 2008 to 2017 |   |   | none (definition) | Gallup U.S. Daily Poll |   |
| band_4554 | 45 to 54 | years of age | definition used in this chapter, not an estimate | an age group of adults defined in this chapter | ladder interviews in 2009 to 2017 and feelings interviews in 2008 to 2017 |   |   | none (definition) | Gallup U.S. Daily Poll |   |
| band_5564 | 55 to 64 | years of age | definition used in this chapter, not an estimate | an age group of adults defined in this chapter | ladder interviews in 2009 to 2017 and feelings interviews in 2008 to 2017 |   |   | none (definition) | Gallup U.S. Daily Poll |   |
| band_7594 | 75 to 94 | years of age | definition used in this chapter, not an estimate | an age group of adults defined in this chapter | ladder interviews in 2009 to 2017 and feelings interviews in 2008 to 2017 |   |   | none (definition) | Gallup U.S. Daily Poll |   |
| coh_ages_end | 58 to 67 | years of age | definition used in this chapter, not an estimate | an age group defined in this chapter: the ages of people born in 1950 to 1959 in 2017 | 2017 |   |   | none (definition) | Gallup U.S. Daily Poll |   |
| coh_ages_start | 50 to 59 | years of age | definition used in this chapter, not an estimate | an age group defined in this chapter: the ages of people born in 1950 to 1959 in 2009 | 2009 |   |   | none (definition) | Gallup U.S. Daily Poll |   |
| coh_cross_ladder | +0.28 | ladder points | unweighted mean at ages 58 to 67 minus unweighted mean at ages 50 to 59 among people interviewed in 2009, interval from a bootstrap over counties | adults aged 50 to 67 interviewed in 2009 | 2009 | 95% CI 0.26 to 0.30 | 128,524 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| coh_cross_pain | 0.0 | percentage points | unweighted mean at ages 58 to 67 minus unweighted mean at ages 50 to 59 among people interviewed in 2009, interval from a bootstrap over counties | adults aged 50 to 67 interviewed in 2009 | 2009 | 95% CI −0.4 to 0.4 | 150,105 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| coh_cross_stress | −10.6 | percentage points | unweighted mean at ages 58 to 67 minus unweighted mean at ages 50 to 59 among people interviewed in 2009, interval from a bootstrap over counties | adults aged 50 to 67 interviewed in 2009 | 2009 | 95% CI −11.1 to −10.1 | 150,096 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| coh_net_ladder | +0.22 | ladder points | change in the cohort's unweighted mean between 2009 and 2017 minus the age-standardized change among all adults at ages 50 to 67, interval from a bootstrap over counties | adults born in 1950 to 1959, interviewed in 2009 and 2017 | 2009 and 2017 | 95% CI 0.20 to 0.24 | 94,948 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| coh_net_pain | +0.8 | percentage points | change in the cohort's unweighted mean between 2009 and 2017 minus the age-standardized change among all adults at ages 50 to 67, interval from a bootstrap over counties | adults born in 1950 to 1959, interviewed in 2009 and 2017 | 2009 and 2017 | 95% CI 0.3 to 1.2 | 105,857 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| coh_net_stress | −9.6 | percentage points | change in the cohort's unweighted mean between 2009 and 2017 minus the age-standardized change among all adults at ages 50 to 67, interval from a bootstrap over counties | adults born in 1950 to 1959, interviewed in 2009 and 2017 | 2009 and 2017 | 95% CI −10.1 to −9.1 | 105,870 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| coh_period_ladder | +0.19 | ladder points | age-standardized change among all adults: the mean over the single years of age 50 to 67 of the unweighted mean in 2017 minus the unweighted mean in 2009 at that age, interval from a bootstrap over counties | adults aged 50 to 67 interviewed in 2009 and 2017 | 2009 and 2017 | 95% CI 0.17 to 0.21 | 169,188 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| coh_period_pain | +5.1 | percentage points | age-standardized change among all adults: the mean over the single years of age 50 to 67 of the unweighted mean in 2017 minus the unweighted mean in 2009 at that age, interval from a bootstrap over counties | adults aged 50 to 67 interviewed in 2009 and 2017 | 2009 and 2017 | 95% CI 4.6 to 5.6 | 188,628 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| coh_period_stress | +1.7 | percentage points | age-standardized change among all adults: the mean over the single years of age 50 to 67 of the unweighted mean in 2017 minus the unweighted mean in 2009 at that age, interval from a bootstrap over counties | adults aged 50 to 67 interviewed in 2009 and 2017 | 2009 and 2017 | 95% CI 1.3 to 2.2 | 188,623 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| coh_std_ages | 50 to 67 | years of age | definition used in this chapter, not an estimate | the ages the 1950 to 1959 cohort passes through between 2009 and 2017 | interviews in 2009 and 2017 |   |   | none (definition) | Gallup U.S. Daily Poll |   |
| coh_within_ladder | +0.41 | ladder points | unweighted mean in 2017 minus unweighted mean in 2009 among people born in 1950 to 1959, interval from a bootstrap over counties | adults born in 1950 to 1959, interviewed in 2009 and 2017 | 2009 and 2017 | 95% CI 0.39 to 0.44 | 94,948 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| coh_within_pain | +5.9 | percentage points | unweighted mean in 2017 minus unweighted mean in 2009 among people born in 1950 to 1959, interval from a bootstrap over counties | adults born in 1950 to 1959, interviewed in 2009 and 2017 | 2009 and 2017 | 95% CI 5.3 to 6.5 | 105,857 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| coh_within_stress | −7.9 | percentage points | unweighted mean in 2017 minus unweighted mean in 2009 among people born in 1950 to 1959, interval from a bootstrap over counties | adults born in 1950 to 1959, interviewed in 2009 and 2017 | 2009 and 2017 | 95% CI −8.5 to −7.2 | 105,870 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| enjoy_1824 | 88.3% | percent of adults who reported the feeling during a lot of the day yesterday | unweighted share of interviews that answered yes to the item, interval clustered by the respondent's county | adults aged 18 to 94 interviewed by Gallup who answered the item and whose county is known (age 18 to 24), enjoyment | interviews in 2008 to 2017 | 95% CI 88.1% to 88.6% | 168,459 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| enjoy_4554 | 82.6% | percent of adults who reported the feeling during a lot of the day yesterday | unweighted share of interviews that answered yes to the item, interval clustered by the respondent's county | adults aged 18 to 94 interviewed by Gallup who answered the item and whose county is known (age 45 to 54), enjoyment | interviews in 2008 to 2017 | 95% CI 82.4% to 82.8% | 461,521 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| enjoy_7594 | 88.5% | percent of adults who reported the feeling during a lot of the day yesterday | unweighted share of interviews that answered yes to the item, interval clustered by the respondent's county | adults aged 18 to 94 interviewed by Gallup who answered the item and whose county is known (age 75 to 94), enjoyment | interviews in 2008 to 2017 | 95% CI 88.3% to 88.7% | 341,222 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| enjoy_inc_high_4554 | 87.6% | percent of adults who reported the feeling during a lot of the day yesterday | unweighted share of interviews that answered yes to the item, interval clustered by the respondent's county | adults aged 18 to 94 interviewed by Gallup who answered the item and whose county is known with household income $90,000 or more (age 45 to 54), enjoyment | interviews in 2009 to 2014 | 95% CI 87.3% to 87.9% | 79,707 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| enjoy_inc_low_4554 | 72.7% | percent of adults who reported the feeling during a lot of the day yesterday | unweighted share of interviews that answered yes to the item, interval clustered by the respondent's county | adults aged 18 to 94 interviewed by Gallup who answered the item and whose county is known with household income Under $36,000 (age 45 to 54), enjoyment | interviews in 2009 to 2014 | 95% CI 72.3% to 73.0% | 76,974 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| enjoy_low | 52 | years of age | age of the lowest point of the five-year moving average of the share reporting the feeling by year of age (ages 20 to 92 searched, refined between whole years by a parabola), interval from a bootstrap over counties | adults aged 18 to 94 interviewed by Gallup who answered the item and whose county is known, enjoyment | interviews in 2008 to 2017 | 95% CI 52 to 53 | 2,581,678 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| fall_adj | −0.32 | ladder points, later age band minus earlier age band | mean of the ladder profile at ages 45 to 54 minus the mean at ages 25 to 34, each year of age weighted by its number of interviews (a raw profile gives the difference of the two band means), interval from a bootstrap over counties; adjusted for sex, marital status, education, household income and survey year | adults with a ladder score, sex, marital status, education and household income | interviews in 2009 to 2014 | 95% CI −0.33 to −0.30 | 1,355,746 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| fall_all | −0.15 | ladder points, later age band minus earlier age band | mean of the ladder profile at ages 45 to 54 minus the mean at ages 25 to 34, each year of age weighted by its number of interviews (a raw profile gives the difference of the two band means), interval from a bootstrap over counties | adults aged 18 to 94 interviewed by Gallup who answered the item and whose county is known | interviews in 2009 to 2017 | 95% CI −0.16 to −0.13 | 2,180,433 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| fall_cc_raw | −0.20 | ladder points, later age band minus earlier age band | mean of the ladder profile at ages 45 to 54 minus the mean at ages 25 to 34, each year of age weighted by its number of interviews (a raw profile gives the difference of the two band means), interval from a bootstrap over counties | adults with a ladder score, sex, marital status, education and household income | interviews in 2009 to 2014 | 95% CI −0.22 to −0.19 | 1,355,746 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| fall_inc_high | −0.05 | ladder points, later age band minus earlier age band | mean of the ladder profile at ages 45 to 54 minus the mean at ages 25 to 34, each year of age weighted by its number of interviews (a raw profile gives the difference of the two band means), interval from a bootstrap over counties | adults with a ladder score and household income $90,000 or more | interviews in 2009 to 2014 | 95% CI −0.07 to −0.03 | 310,559 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| fall_inc_high_late | +0.09 | ladder points, later age band minus earlier age band | mean of the ladder profile at ages 45 to 54 minus the mean at ages 25 to 34, each year of age weighted by its number of interviews (a raw profile gives the difference of the two band means), interval from a bootstrap over counties | adults with a ladder score and household income $90,000 or more | interviews in 2015 to 2017 | 95% CI 0.06 to 0.12 | 139,811 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| fall_inc_low | −0.59 | ladder points, later age band minus earlier age band | mean of the ladder profile at ages 45 to 54 minus the mean at ages 25 to 34, each year of age weighted by its number of interviews (a raw profile gives the difference of the two band means), interval from a bootstrap over counties | adults with a ladder score and household income Under $36,000 | interviews in 2009 to 2014 | 95% CI −0.62 to −0.56 | 485,159 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| fall_inc_mid | −0.27 | ladder points, later age band minus earlier age band | mean of the ladder profile at ages 45 to 54 minus the mean at ages 25 to 34, each year of age weighted by its number of interviews (a raw profile gives the difference of the two band means), interval from a bootstrap over counties | adults with a ladder score and household income $36,000 to $89,999 | interviews in 2009 to 2014 | 95% CI −0.29 to −0.26 | 567,870 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| fall_marital | −0.23 | ladder points, later age band minus earlier age band | mean of the ladder profile at ages 45 to 54 minus the mean at ages 25 to 34, each year of age weighted by its number of interviews (a raw profile gives the difference of the two band means), interval from a bootstrap over counties; adjusted for marital status only | adults with a ladder score, sex, marital status, education and household income | interviews in 2009 to 2014 | 95% CI −0.25 to −0.22 | 1,355,746 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| fall_place_least | −0.11 | ladder points, later age band minus earlier age band | mean of the ladder profile at ages 45 to 54 minus the mean at ages 25 to 34, each year of age weighted by its number of interviews (a raw profile gives the difference of the two band means), interval from a bootstrap over counties | adults with a ladder score in the least deprived third of interviews by county Area Deprivation Index | interviews in 2009 to 2017 | 95% CI −0.14 to −0.09 | 718,005 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| fall_place_most | −0.17 | ladder points, later age band minus earlier age band | mean of the ladder profile at ages 45 to 54 minus the mean at ages 25 to 34, each year of age weighted by its number of interviews (a raw profile gives the difference of the two band means), interval from a bootstrap over counties | adults with a ladder score in the most deprived third of interviews by county Area Deprivation Index | interviews in 2009 to 2017 | 95% CI −0.19 to −0.15 | 731,406 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| fb_decline_age | 75 | years of age | approximate age around which happiness declines in the fixed-effects results of Frijters and Beatton (2012), as stated in the abstract of the paper | adults in the German Socio-Economic Panel, the British Household Panel Survey and the Household, Income and Labour Dynamics in Australia survey | panel waves analyzed by Frijters and Beatton (2012) |   |   | none (published estimate) | Published estimates from cited studies | Quoted from the abstract of the paper, which gives approximate ages. |
| fb_rise_age | 60 | years of age | approximate age around which happiness rises in the fixed-effects results of Frijters and Beatton (2012), as stated in the abstract of the paper | adults in the German Socio-Economic Panel, the British Household Panel Survey and the Household, Income and Labour Dynamics in Australia survey | panel waves analyzed by Frijters and Beatton (2012) |   |   | none (published estimate) | Published estimates from cited studies | Quoted from the abstract of the paper, which gives approximate ages. |
| feel_gap_2008_max | −1.6 | percentage points, 2008 minus 2009 | unweighted difference in the share reporting the feeling between 2008 and 2009, for the feeling with the largest absolute difference of the seven, interval clustered by county | adults aged 18 to 94 interviewed by Gallup in 2008 or 2009 who answered the item (worry) | 2008 and 2009 | 95% CI −1.8 to −1.3 | 696,084 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| flat_range | 0.06 | ladder points | difference between the highest and the lowest value of the five-year moving average of the raw mean ladder score at ages 46 to 56 | adults with a ladder score, sex, marital status, education and household income | interviews in 2009 to 2014 |   | 1,355,746 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| gap2008_hi | 0.38 | ladder points, 2009 minus 2008 | unweighted difference between the mean ladder score in 2009 and in 2008 within the age band with the largest difference of the seven bands, interval clustered by county | adults aged 18 to 94 interviewed by Gallup in 2008 or 2009 who answered the ladder (the age band with the largest difference) | 2008 and 2009 | 95% CI 0.35 to 0.41 | 106,145 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| gap2008_lo | 0.29 | ladder points, 2009 minus 2008 | unweighted difference between the mean ladder score in 2009 and in 2008 within the age band with the smallest difference of the seven bands, interval clustered by county | adults aged 18 to 94 interviewed by Gallup in 2008 or 2009 who answered the ladder (the age band with the smallest difference) | 2008 and 2009 | 95% CI 0.26 to 0.32 | 83,551 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| happy_1824 | 91.9% | percent of adults who reported the feeling during a lot of the day yesterday | unweighted share of interviews that answered yes to the item, interval clustered by the respondent's county | adults aged 18 to 94 interviewed by Gallup who answered the item and whose county is known (age 18 to 24), happiness | interviews in 2008 to 2016 | 95% CI 91.7% to 92.1% | 154,490 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| happy_4554 | 86.6% | percent of adults who reported the feeling during a lot of the day yesterday | unweighted share of interviews that answered yes to the item, interval clustered by the respondent's county | adults aged 18 to 94 interviewed by Gallup who answered the item and whose county is known (age 45 to 54), happiness | interviews in 2008 to 2016 | 95% CI 86.4% to 86.8% | 437,094 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| happy_7594 | 90.0% | percent of adults who reported the feeling during a lot of the day yesterday | unweighted share of interviews that answered yes to the item, interval clustered by the respondent's county | adults aged 18 to 94 interviewed by Gallup who answered the item and whose county is known (age 75 to 94), happiness | interviews in 2008 to 2016 | 95% CI 89.8% to 90.2% | 322,920 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| happy_low | 54 | years of age | age of the lowest point of the five-year moving average of the share reporting the feeling by year of age (ages 20 to 92 searched, refined between whole years by a parabola), interval from a bootstrap over counties | adults aged 18 to 94 interviewed by Gallup who answered the item and whose county is known, happiness | interviews in 2008 to 2016 | 95% CI 54 to 56 | 2,424,735 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| high_share_2014 | 23% | percent of interviews that report a household income bracket | unweighted share reporting household income of $90,000 or more, interval clustered by county | adults aged 18 to 94 interviewed by Gallup in 2014 who report a household income bracket | 2014 | 95% CI 22% to 24% | 284,323 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| high_share_2017 | 35% | percent of interviews that report a household income bracket | unweighted share reporting household income of $90,000 or more, interval clustered by county | adults aged 18 to 94 interviewed by Gallup in 2017 who report a household income bracket | 2017 | 95% CI 34% to 36% | 287,211 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| inc_cut_high | $90,000 | dollars of annual household income, as reported in the year of the interview | definition used in this chapter, not an estimate (Gallup's harmonized brackets, not adjusted for inflation) | the lower limit of the highest household income group | interviews in 2009 to 2014 |   |   | none (definition) | Gallup U.S. Daily Poll |   |
| inc_cut_low | $36,000 | dollars of annual household income, as reported in the year of the interview | definition used in this chapter, not an estimate (Gallup's harmonized brackets, not adjusted for inflation) | the upper limit of the lowest household income group | interviews in 2009 to 2014 |   |   | none (definition) | Gallup U.S. Daily Poll |   |
| income_known_share | 81% | percent of interviews with a household income bracket | share of interviews that report a household income bracket (unweighted count) | Gallup interviews of adults aged 18 to 94 with a known county | interviews in 2009 to 2014 |   | 1,691,589 respondents | none (unweighted count) | Gallup U.S. Daily Poll |   |
| ladder_18_24 | 7.03 | Cantril ladder score (0 worst possible life, 10 best possible life) | unweighted mean of the 0 to 10 ladder score, interval clustered by the respondent's county | adults aged 18 to 94 interviewed by Gallup who answered the item and whose county is known (age 18 to 24) | interviews in 2009 to 2017 | 95% CI 7.02 to 7.05 | 149,624 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| ladder_25_34 | 6.99 | Cantril ladder score (0 worst possible life, 10 best possible life) | unweighted mean of the 0 to 10 ladder score, interval clustered by the respondent's county | adults aged 18 to 94 interviewed by Gallup who answered the item and whose county is known (age 25 to 34) | interviews in 2009 to 2017 | 95% CI 6.97 to 7.00 | 229,054 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| ladder_4554_least | 6.94 | Cantril ladder score (0 worst possible life, 10 best possible life) | unweighted mean of the 0 to 10 ladder score, interval clustered by the respondent's county | adults aged 18 to 94 interviewed by Gallup who answered the item and whose county is known in the least deprived third of interviews by county Area Deprivation Index (age 45 to 54) | interviews in 2009 to 2017 | 95% CI 6.90 to 6.97 | 128,209 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| ladder_4554_most | 6.73 | Cantril ladder score (0 worst possible life, 10 best possible life) | unweighted mean of the 0 to 10 ladder score, interval clustered by the respondent's county | adults aged 18 to 94 interviewed by Gallup who answered the item and whose county is known in the most deprived third of interviews by county Area Deprivation Index (age 45 to 54) | interviews in 2009 to 2017 | 95% CI 6.71 to 6.75 | 125,537 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| ladder_45_54 | 6.84 | Cantril ladder score (0 worst possible life, 10 best possible life) | unweighted mean of the 0 to 10 ladder score, interval clustered by the respondent's county | adults aged 18 to 94 interviewed by Gallup who answered the item and whose county is known (age 45 to 54) | interviews in 2009 to 2017 | 95% CI 6.82 to 6.86 | 380,027 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| ladder_75_94 | 7.38 | Cantril ladder score (0 worst possible life, 10 best possible life) | unweighted mean of the 0 to 10 ladder score, interval clustered by the respondent's county | adults aged 18 to 94 interviewed by Gallup who answered the item and whose county is known (age 75 to 94) | interviews in 2009 to 2017 | 95% CI 7.37 to 7.39 | 289,487 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| ladder_at_18 | 7.25 | Cantril ladder score (0 worst possible life, 10 best possible life) | unweighted mean of the 0 to 10 ladder score, interval clustered by the respondent's county | adults aged 18 to 94 interviewed by Gallup who answered the item and whose county is known (age 18) | interviews in 2009 to 2017 | 95% CI 7.23 to 7.28 | 25,104 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| ladder_at_25 | 6.93 | Cantril ladder score (0 worst possible life, 10 best possible life) | unweighted mean of the 0 to 10 ladder score, interval clustered by the respondent's county | adults aged 18 to 94 interviewed by Gallup who answered the item and whose county is known (age 25) | interviews in 2009 to 2017 | 95% CI 6.91 to 6.96 | 22,756 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| ladder_inc_high_2534 | 7.50 | Cantril ladder score (0 worst possible life, 10 best possible life) | unweighted mean of the 0 to 10 ladder score, interval clustered by the respondent's county | adults aged 18 to 94 interviewed by Gallup who answered the item and whose county is known with household income $90,000 or more (age 25 to 34) | interviews in 2009 to 2014 | 95% CI 7.48 to 7.52 | 28,340 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| ladder_inc_high_3544 | 7.51 | Cantril ladder score (0 worst possible life, 10 best possible life) | unweighted mean of the 0 to 10 ladder score, interval clustered by the respondent's county | adults aged 18 to 94 interviewed by Gallup who answered the item and whose county is known with household income $90,000 or more (age 35 to 44) | interviews in 2009 to 2014 | 95% CI 7.50 to 7.53 | 55,787 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| ladder_inc_high_4554 | 7.45 | Cantril ladder score (0 worst possible life, 10 best possible life) | unweighted mean of the 0 to 10 ladder score, interval clustered by the respondent's county | adults aged 18 to 94 interviewed by Gallup who answered the item and whose county is known with household income $90,000 or more (age 45 to 54) | interviews in 2009 to 2014 | 95% CI 7.43 to 7.46 | 77,234 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| ladder_inc_low_1824 | 6.85 | Cantril ladder score (0 worst possible life, 10 best possible life) | unweighted mean of the 0 to 10 ladder score, interval clustered by the respondent's county | adults aged 18 to 94 interviewed by Gallup who answered the item and whose county is known with household income Under $36,000 (age 18 to 24) | interviews in 2009 to 2014 | 95% CI 6.83 to 6.88 | 45,506 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| ladder_inc_low_4554 | 5.89 | Cantril ladder score (0 worst possible life, 10 best possible life) | unweighted mean of the 0 to 10 ladder score, interval clustered by the respondent's county | adults aged 18 to 94 interviewed by Gallup who answered the item and whose county is known with household income Under $36,000 (age 45 to 54) | interviews in 2009 to 2014 | 95% CI 5.86 to 5.91 | 74,530 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| ladder_inc_low_7594 | 7.04 | Cantril ladder score (0 worst possible life, 10 best possible life) | unweighted mean of the 0 to 10 ladder score, interval clustered by the respondent's county | adults aged 18 to 94 interviewed by Gallup who answered the item and whose county is known with household income Under $36,000 (age 75 to 94) | interviews in 2009 to 2014 | 95% CI 7.03 to 7.06 | 82,417 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| ladder_inc_mid_4554 | 6.86 | Cantril ladder score (0 worst possible life, 10 best possible life) | unweighted mean of the 0 to 10 ladder score, interval clustered by the respondent's county | adults aged 18 to 94 interviewed by Gallup who answered the item and whose county is known with household income $36,000 to $89,999 (age 45 to 54) | interviews in 2009 to 2014 | 95% CI 6.85 to 6.88 | 104,462 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| ladder_share_2009 | 85% | percent of the year's interviews that carry a ladder score | share of the year's interviews with a ladder score (before any filter) | all Gallup interviews in 2009 | 2009 |   |   | none (unweighted count) | Gallup U.S. Daily Poll |   |
| ladder_share_late | 49% | percent of interviews that carry a ladder score | mean over the five years of the share of the year's interviews with a ladder score (before any filter) | all Gallup interviews from 2013 to 2017 | 2013 to 2017 |   |   | none (unweighted count) | Gallup U.S. Daily Poll |   |
| ladder_turn | 54 | years of age | age of the lowest point of the five-year moving average of the mean ladder score by year of age (ages 20 to 92 searched, refined between whole years by a parabola through the lowest point and its neighbours), interval from a bootstrap over counties | adults aged 18 to 94 interviewed by Gallup who answered the item and whose county is known | interviews in 2009 to 2017 | 95% CI 51 to 54 | 2,180,433 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| ladder_year_2008 | 6.69 | Cantril ladder score (0 worst possible life, 10 best possible life) | unweighted mean of the 0 to 10 ladder score, interval clustered by the respondent's county (all ages; the 2008 value is shown to document the artifact and is not used elsewhere) | adults aged 18 to 94 interviewed by Gallup who answered the ladder and whose county is known | 2008 | 95% CI 6.68 to 6.71 | 346,485 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| ladder_year_2009 | 7.03 | Cantril ladder score (0 worst possible life, 10 best possible life) | unweighted mean of the 0 to 10 ladder score, interval clustered by the respondent's county (all ages; the 2008 value is shown to document the artifact and is not used elsewhere) | adults aged 18 to 94 interviewed by Gallup who answered the ladder and whose county is known | 2009 | 95% CI 7.02 to 7.04 | 296,836 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| mean_age_2009 | 55.1 | years of age | mean age of the interviews | adults aged 18 to 94 interviewed by Gallup in 2009 with a known county | 2009 |   |   | none (unweighted count) | Gallup U.S. Daily Poll |   |
| mean_age_2017 | 52.3 | years of age | mean age of the interviews | adults aged 18 to 94 interviewed by Gallup in 2017 with a known county | 2017 |   |   | none (unweighted count) | Gallup U.S. Daily Poll |   |
| midfall_inc_high | −0.07 | ladder points, later age band minus earlier age band | mean of the ladder profile at ages 45 to 54 minus the mean at ages 35 to 44, each year of age weighted by its number of interviews (a raw profile gives the difference of the two band means), interval from a bootstrap over counties | adults with a ladder score and household income $90,000 or more | interviews in 2009 to 2014 | 95% CI −0.08 to −0.05 | 310,559 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| midfall_inc_high_late | −0.02 | ladder points, later age band minus earlier age band | mean of the ladder profile at ages 45 to 54 minus the mean at ages 35 to 44, each year of age weighted by its number of interviews (a raw profile gives the difference of the two band means), interval from a bootstrap over counties | adults with a ladder score and household income $90,000 or more | interviews in 2015 to 2017 | 95% CI −0.04 to 0.01 | 139,811 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| midfall_inc_low | −0.27 | ladder points, later age band minus earlier age band | mean of the ladder profile at ages 45 to 54 minus the mean at ages 35 to 44, each year of age weighted by its number of interviews (a raw profile gives the difference of the two band means), interval from a bootstrap over counties | adults with a ladder score and household income Under $36,000 | interviews in 2009 to 2014 | 95% CI −0.30 to −0.24 | 485,159 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| midfall_inc_mid | −0.12 | ladder points, later age band minus earlier age band | mean of the ladder profile at ages 45 to 54 minus the mean at ages 35 to 44, each year of age weighted by its number of interviews (a raw profile gives the difference of the two band means), interval from a bootstrap over counties | adults with a ladder score and household income $36,000 to $89,999 | interviews in 2009 to 2014 | 95% CI −0.13 to −0.10 | 567,870 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| min_cell_income | 51 | interviews | size of the smallest cell of the ladder profiles by household income group (the highest income group at the oldest ages) | adults with a ladder score in the smallest single-year-of-age cell by household income group, 2009 to 2014 | interviews in 2009 to 2014 |   | 51 respondents | none (unweighted count) | Gallup U.S. Daily Poll |   |
| n_cc | 1,355,746 | interviews | count of interviews used for the adjusted profiles | adults with a ladder score, sex, marital status, education and household income | interviews in 2009 to 2014 |   |   | none (unweighted count) | Gallup U.S. Daily Poll |   |
| n_counties | 3,142 | counties | count of distinct counties assigned to respondents (county is derived from the ZIP code) | counties of the respondents' ZIP codes in the analysis | interviews in 2008 to 2017 |   |   | none (unweighted count) | Gallup U.S. Daily Poll |   |
| n_counties_least | 274 | counties | count of distinct counties whose interviews fall in the group (a county is in one group) | counties in the least deprived third of interviews by county Area Deprivation Index | interviews in 2008 to 2017 |   |   | none (unweighted count) | Gallup U.S. Daily Poll |   |
| n_counties_mid | 655 | counties | count of distinct counties whose interviews fall in the group (a county is in one group) | counties in the middle third of interviews by county Area Deprivation Index | interviews in 2008 to 2017 |   |   | none (unweighted count) | Gallup U.S. Daily Poll |   |
| n_counties_most | 2,193 | counties | count of distinct counties whose interviews fall in the group (a county is in one group) | counties in the most deprived third of interviews by county Area Deprivation Index | interviews in 2008 to 2017 |   |   | none (unweighted count) | Gallup U.S. Daily Poll |   |
| n_frame | 3,457,670 | interviews | count of interviews (not all answered every item) | Gallup interviews of adults aged 18 to 94 with a known county | interviews in 2008 to 2017 |   |   | none (unweighted count) | Gallup U.S. Daily Poll |   |
| n_health | 1,354,627 | interviews | count of interviews used for the adjusted profiles with self-rated health | adults with a ladder score, sex, marital status, education, household income and a self-rated health answer | interviews in 2009 to 2014 |   |   | none (unweighted count) | Gallup U.S. Daily Poll |   |
| n_ladder | 2,180,433 | interviews | count of interviews with a ladder score (the ladder of 2008 is not used) | adults aged 18 to 94 interviewed by Gallup who answered the item and whose county is known | interviews in 2009 to 2017 |   |   | none (unweighted count) | Gallup U.S. Daily Poll |   |
| n_no_place | 931 | interviews | count of interviews left out of the deprivation groups | Gallup interviews of adults aged 18 to 94 in counties with no Area Deprivation Index rank | interviews in 2008 to 2017 |   |   | none (unweighted count) | Gallup U.S. Daily Poll |   |
| pain_1824 | 15.5% | percent of adults who reported the feeling during a lot of the day yesterday | unweighted share of interviews that answered yes to the item, interval clustered by the respondent's county | adults aged 18 to 94 interviewed by Gallup who answered the item and whose county is known (age 18 to 24), physical pain | interviews in 2008 to 2017 | 95% CI 15.3% to 15.7% | 168,532 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| pain_5564 | 28.5% | percent of adults who reported the feeling during a lot of the day yesterday | unweighted share of interviews that answered yes to the item, interval clustered by the respondent's county | adults aged 18 to 94 interviewed by Gallup who answered the item and whose county is known (age 55 to 64), physical pain | interviews in 2008 to 2017 | 95% CI 28.2% to 28.9% | 551,195 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| pain_7594 | 26.4% | percent of adults who reported the feeling during a lot of the day yesterday | unweighted share of interviews that answered yes to the item, interval clustered by the respondent's county | adults aged 18 to 94 interviewed by Gallup who answered the item and whose county is known (age 75 to 94), physical pain | interviews in 2008 to 2017 | 95% CI 26.2% to 26.6% | 342,096 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| pain_inc_high_4554 | 15.7% | percent of adults who reported the feeling during a lot of the day yesterday | unweighted share of interviews that answered yes to the item, interval clustered by the respondent's county | adults aged 18 to 94 interviewed by Gallup who answered the item and whose county is known with household income $90,000 or more (age 45 to 54), physical pain | interviews in 2009 to 2014 | 95% CI 15.4% to 16.0% | 79,725 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| pain_inc_high_7594 | 20.9% | percent of adults who reported the feeling during a lot of the day yesterday | unweighted share of interviews that answered yes to the item, interval clustered by the respondent's county | adults aged 18 to 94 interviewed by Gallup who answered the item and whose county is known with household income $90,000 or more (age 75 to 94), physical pain | interviews in 2009 to 2014 | 95% CI 20.3% to 21.5% | 17,647 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| pain_inc_low_4554 | 41.2% | percent of adults who reported the feeling during a lot of the day yesterday | unweighted share of interviews that answered yes to the item, interval clustered by the respondent's county | adults aged 18 to 94 interviewed by Gallup who answered the item and whose county is known with household income Under $36,000 (age 45 to 54), physical pain | interviews in 2009 to 2014 | 95% CI 40.6% to 41.8% | 77,087 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| pain_inc_low_7594 | 30.5% | percent of adults who reported the feeling during a lot of the day yesterday | unweighted share of interviews that answered yes to the item, interval clustered by the respondent's county | adults aged 18 to 94 interviewed by Gallup who answered the item and whose county is known with household income Under $36,000 (age 75 to 94), physical pain | interviews in 2009 to 2014 | 95% CI 30.2% to 30.9% | 85,614 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| pain_peak | 60 | years of age | age of the highest point of the five-year moving average of the share reporting the feeling by year of age (ages 20 to 92 searched, refined between whole years by a parabola), interval from a bootstrap over counties | adults aged 18 to 94 interviewed by Gallup who answered the item and whose county is known, physical pain | interviews in 2008 to 2017 | 95% CI 58 to 61 | 2,584,986 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| place_fall_diff | −0.06 | ladder points, most deprived third minus least deprived third | difference between the two groups in the change of the unweighted mean ladder score from ages 25 to 34 to ages 45 to 54, interval from a bootstrap over counties | adults with a ladder score aged 25 to 54 in the extreme thirds of interviews by county Area Deprivation Index | interviews in 2009 to 2017 | 95% CI −0.09 to −0.03 | 407,100 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| place_gap_ladder_1824 | −0.07 | ladder points, most deprived third minus least deprived third | unweighted difference in the mean ladder score between the two groups of interviews, interval clustered by county | adults aged 18 to 94 interviewed by Gallup who answered the item and whose county is known (age 18 to 24), in the extreme thirds of interviews by county Area Deprivation Index | interviews in 2009 to 2017 | 95% CI −0.10 to −0.04 | 99,885 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| place_gap_ladder_4554 | −0.21 | ladder points, most deprived third minus least deprived third | unweighted difference in the mean ladder score between the two groups of interviews, interval clustered by county | adults aged 18 to 94 interviewed by Gallup who answered the item and whose county is known (age 45 to 54), in the extreme thirds of interviews by county Area Deprivation Index | interviews in 2009 to 2017 | 95% CI −0.24 to −0.17 | 253,746 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| place_gap_ladder_7594 | −0.12 | ladder points, most deprived third minus least deprived third | unweighted difference in the mean ladder score between the two groups of interviews, interval clustered by county | adults aged 18 to 94 interviewed by Gallup who answered the item and whose county is known (age 75 to 94), in the extreme thirds of interviews by county Area Deprivation Index | interviews in 2009 to 2017 | 95% CI −0.15 to −0.09 | 190,633 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| place_gap_pain_1824 | +0.9 | percentage points, most deprived third minus least deprived third | unweighted difference in the mean indicator of physical pain yesterday between the two groups of interviews, interval clustered by county | adults aged 18 to 94 interviewed by Gallup who answered the item and whose county is known (age 18 to 24), in the extreme thirds of interviews by county Area Deprivation Index | interviews in 2008 to 2017 | 95% CI 0.3 to 1.4 | 112,592 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| place_gap_pain_4554 | +6.8 | percentage points, most deprived third minus least deprived third | unweighted difference in the mean indicator of physical pain yesterday between the two groups of interviews, interval clustered by county | adults aged 18 to 94 interviewed by Gallup who answered the item and whose county is known (age 45 to 54), in the extreme thirds of interviews by county Area Deprivation Index | interviews in 2008 to 2017 | 95% CI 6.0 to 7.6 | 309,173 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| place_max_enjoy | 1.5 | percentage points, absolute difference | largest absolute difference across the seven age bands between the most and least deprived thirds in the share reporting the feeling yesterday (n is the smallest band) | adults aged 18 to 94 interviewed by Gallup who answered the item and whose county is known (enjoyment), in the extreme thirds of interviews by county Area Deprivation Index, the largest of the seven age bands | interviews in 2008 to 2017 |   | 112,542 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| place_max_sad | 2.1 | percentage points, absolute difference | largest absolute difference across the seven age bands between the most and least deprived thirds in the share reporting the feeling yesterday (n is the smallest band) | adults aged 18 to 94 interviewed by Gallup who answered the item and whose county is known (sadness), in the extreme thirds of interviews by county Area Deprivation Index, the largest of the seven age bands | interviews in 2008 to 2016 |   | 103,310 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| place_max_stress | 1.3 | percentage points, absolute difference | largest absolute difference across the seven age bands between the most and least deprived thirds in the share reporting the feeling yesterday (n is the smallest band) | adults aged 18 to 94 interviewed by Gallup who answered the item and whose county is known (stress), in the extreme thirds of interviews by county Area Deprivation Index, the largest of the seven age bands | interviews in 2008 to 2017 |   | 112,582 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| place_max_worry | 2.9 | percentage points, absolute difference | largest absolute difference across the seven age bands between the most and least deprived thirds in the share reporting the feeling yesterday (n is the smallest band) | adults aged 18 to 94 interviewed by Gallup who answered the item and whose county is known (worry), in the extreme thirds of interviews by county Area Deprivation Index, the largest of the seven age bands | interviews in 2008 to 2017 |   | 112,597 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| quad_all5 | 47 | years of age | age at which a quadratic in age fitted to the ladder score turns (with fixed effects for sex, marital status, education, household income and survey year), interval from a bootstrap over counties | adults with a ladder score, sex, marital status, education and household income | interviews in 2009 to 2014 | 95% CI 47 to 47 | 1,355,746 respondents | none (unweighted) | Gallup U.S. Daily Poll | A parabola forces a symmetric shape; the interval describes sampling variation of this fit and not whether the shape is right. |
| quad_raw | 44 | years of age | age at which a quadratic in age fitted to the ladder score turns, interval from a bootstrap over counties | adults with a ladder score, sex, marital status, education and household income | interviews in 2009 to 2014 | 95% CI 44 to 45 | 1,355,746 respondents | none (unweighted) | Gallup U.S. Daily Poll | A parabola forces a symmetric shape; the interval describes sampling variation of this fit and not whether the shape is right. |
| rise_adj | +0.91 | ladder points, later age band minus earlier age band | mean of the ladder profile at ages 75 to 94 minus the mean at ages 45 to 54, each year of age weighted by its number of interviews (a raw profile gives the difference of the two band means), interval from a bootstrap over counties; adjusted for sex, marital status, education, household income and survey year | adults with a ladder score, sex, marital status, education and household income | interviews in 2009 to 2014 | 95% CI 0.90 to 0.93 | 1,355,746 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| rise_all | +0.54 | ladder points, later age band minus earlier age band | mean of the ladder profile at ages 75 to 94 minus the mean at ages 45 to 54, each year of age weighted by its number of interviews (a raw profile gives the difference of the two band means), interval from a bootstrap over counties | adults aged 18 to 94 interviewed by Gallup who answered the item and whose county is known | interviews in 2009 to 2017 | 95% CI 0.53 to 0.55 | 2,180,433 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| rise_cc_raw | +0.56 | ladder points, later age band minus earlier age band | mean of the ladder profile at ages 75 to 94 minus the mean at ages 45 to 54, each year of age weighted by its number of interviews (a raw profile gives the difference of the two band means), interval from a bootstrap over counties | adults with a ladder score, sex, marital status, education and household income | interviews in 2009 to 2014 | 95% CI 0.54 to 0.57 | 1,355,746 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| rise_inc_high | +0.37 | ladder points, later age band minus earlier age band | mean of the ladder profile at ages 75 to 94 minus the mean at ages 45 to 54, each year of age weighted by its number of interviews (a raw profile gives the difference of the two band means), interval from a bootstrap over counties | adults with a ladder score and household income $90,000 or more | interviews in 2009 to 2014 | 95% CI 0.33 to 0.40 | 310,559 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| rise_inc_low | +1.15 | ladder points, later age band minus earlier age band | mean of the ladder profile at ages 75 to 94 minus the mean at ages 45 to 54, each year of age weighted by its number of interviews (a raw profile gives the difference of the two band means), interval from a bootstrap over counties | adults with a ladder score and household income Under $36,000 | interviews in 2009 to 2014 | 95% CI 1.13 to 1.18 | 485,159 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| rise_inc_mid | +0.69 | ladder points, later age band minus earlier age band | mean of the ladder profile at ages 75 to 94 minus the mean at ages 45 to 54, each year of age weighted by its number of interviews (a raw profile gives the difference of the two band means), interval from a bootstrap over counties | adults with a ladder score and household income $36,000 to $89,999 | interviews in 2009 to 2014 | 95% CI 0.67 to 0.71 | 567,870 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| sad_2534 | 14.9% | percent of adults who reported the feeling during a lot of the day yesterday | unweighted share of interviews that answered yes to the item, interval clustered by the respondent's county | adults aged 18 to 94 interviewed by Gallup who answered the item and whose county is known (age 25 to 34), sadness | interviews in 2008 to 2016 | 95% CI 14.6% to 15.2% | 248,243 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| sad_4554 | 18.6% | percent of adults who reported the feeling during a lot of the day yesterday | unweighted share of interviews that answered yes to the item, interval clustered by the respondent's county | adults aged 18 to 94 interviewed by Gallup who answered the item and whose county is known (age 45 to 54), sadness | interviews in 2008 to 2016 | 95% CI 18.3% to 18.8% | 437,483 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| sad_7594 | 16.1% | percent of adults who reported the feeling during a lot of the day yesterday | unweighted share of interviews that answered yes to the item, interval clustered by the respondent's county | adults aged 18 to 94 interviewed by Gallup who answered the item and whose county is known (age 75 to 94), sadness | interviews in 2008 to 2016 | 95% CI 15.9% to 16.3% | 323,919 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| sadness_peak | 55 | years of age | age of the highest point of the five-year moving average of the share reporting the feeling by year of age (ages 20 to 92 searched, refined between whole years by a parabola), interval from a bootstrap over counties | adults aged 18 to 94 interviewed by Gallup who answered the item and whose county is known, sadness | interviews in 2008 to 2016 | 95% CI 55 to 55 | 2,427,607 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| stress_2534 | 48.6% | percent of adults who reported the feeling during a lot of the day yesterday | unweighted share of interviews that answered yes to the item, interval clustered by the respondent's county | adults aged 18 to 94 interviewed by Gallup who answered the item and whose county is known (age 25 to 34), stress | interviews in 2008 to 2017 | 95% CI 48.3% to 48.9% | 269,066 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| stress_4554 | 44.6% | percent of adults who reported the feeling during a lot of the day yesterday | unweighted share of interviews that answered yes to the item, interval clustered by the respondent's county | adults aged 18 to 94 interviewed by Gallup who answered the item and whose county is known (age 45 to 54), stress | interviews in 2008 to 2017 | 95% CI 44.3% to 44.8% | 462,026 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| stress_7594 | 18.0% | percent of adults who reported the feeling during a lot of the day yesterday | unweighted share of interviews that answered yes to the item, interval clustered by the respondent's county | adults aged 18 to 94 interviewed by Gallup who answered the item and whose county is known (age 75 to 94), stress | interviews in 2008 to 2017 | 95% CI 17.8% to 18.2% | 342,420 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| stress_inc_high_4554 | 42.4% | percent of adults who reported the feeling during a lot of the day yesterday | unweighted share of interviews that answered yes to the item, interval clustered by the respondent's county | adults aged 18 to 94 interviewed by Gallup who answered the item and whose county is known with household income $90,000 or more (age 45 to 54), stress | interviews in 2009 to 2014 | 95% CI 42.0% to 42.8% | 79,721 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| stress_inc_high_7594 | 16.6% | percent of adults who reported the feeling during a lot of the day yesterday | unweighted share of interviews that answered yes to the item, interval clustered by the respondent's county | adults aged 18 to 94 interviewed by Gallup who answered the item and whose county is known with household income $90,000 or more (age 75 to 94), stress | interviews in 2009 to 2014 | 95% CI 15.9% to 17.2% | 17,665 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| stress_inc_low_4554 | 52.6% | percent of adults who reported the feeling during a lot of the day yesterday | unweighted share of interviews that answered yes to the item, interval clustered by the respondent's county | adults aged 18 to 94 interviewed by Gallup who answered the item and whose county is known with household income Under $36,000 (age 45 to 54), stress | interviews in 2009 to 2014 | 95% CI 52.1% to 53.2% | 77,096 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| stress_inc_low_7594 | 20.2% | percent of adults who reported the feeling during a lot of the day yesterday | unweighted share of interviews that answered yes to the item, interval clustered by the respondent's county | adults aged 18 to 94 interviewed by Gallup who answered the item and whose county is known with household income Under $36,000 (age 75 to 94), stress | interviews in 2009 to 2014 | 95% CI 19.9% to 20.5% | 85,699 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| stress_peak | 26 | years of age | age of the highest point of the five-year moving average of the share reporting the feeling by year of age (ages 20 to 92 searched, refined between whole years by a parabola), interval from a bootstrap over counties | adults aged 18 to 94 interviewed by Gallup who answered the item and whose county is known, stress | interviews in 2008 to 2017 | 95% CI 25 to 33 | 2,585,120 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| turn_all5 | 48 | years of age | age of the lowest point of the five-year moving average of the mean ladder score by year of age (ages 20 to 92 searched, refined between whole years by a parabola through the lowest point and its neighbours), interval from a bootstrap over counties | adults with a ladder score, sex, marital status, education and household income | interviews in 2009 to 2014 | 95% CI 48 to 54 | 1,355,746 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| turn_cc_raw | 54 | years of age | age of the lowest point of the five-year moving average of the mean ladder score by year of age (ages 20 to 92 searched, refined between whole years by a parabola through the lowest point and its neighbours), interval from a bootstrap over counties | adults with a ladder score, sex, marital status, education and household income | interviews in 2009 to 2014 | 95% CI 51 to 55 | 1,355,746 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| turn_h_plus | 48 | years of age | age of the lowest point of the five-year moving average of the mean ladder score by year of age (ages 20 to 92 searched, refined between whole years by a parabola through the lowest point and its neighbours), interval from a bootstrap over counties | adults with a ladder score, sex, marital status, education and household income and a self-rated health answer | interviews in 2009 to 2014 | 95% CI 47 to 48 | 1,354,627 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| turn_inc_high | 24 | years of age | age of the lowest point of the five-year moving average of the mean ladder score by year of age (ages 20 to 92 searched, refined between whole years by a parabola through the lowest point and its neighbours), interval from a bootstrap over counties | adults with a ladder score and household income $90,000 or more | interviews in 2009 to 2014 | 95% CI 23 to 24 | 310,559 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| turn_inc_low | 54 | years of age | age of the lowest point of the five-year moving average of the mean ladder score by year of age (ages 20 to 92 searched, refined between whole years by a parabola through the lowest point and its neighbours), interval from a bootstrap over counties | adults with a ladder score and household income Under $36,000 | interviews in 2009 to 2014 | 95% CI 54 to 55 | 485,159 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| turn_inc_mid | 48 | years of age | age of the lowest point of the five-year moving average of the mean ladder score by year of age (ages 20 to 92 searched, refined between whole years by a parabola through the lowest point and its neighbours), interval from a bootstrap over counties | adults with a ladder score and household income $36,000 to $89,999 | interviews in 2009 to 2014 | 95% CI 48 to 51 | 567,870 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| turn_income | 48 | years of age | age of the lowest point of the five-year moving average of the mean ladder score by year of age (ages 20 to 92 searched, refined between whole years by a parabola through the lowest point and its neighbours), interval from a bootstrap over counties | adults with a ladder score, sex, marital status, education and household income | interviews in 2009 to 2014 | 95% CI 48 to 54 | 1,355,746 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| turn_marital | 54 | years of age | age of the lowest point of the five-year moving average of the mean ladder score by year of age (ages 20 to 92 searched, refined between whole years by a parabola through the lowest point and its neighbours), interval from a bootstrap over counties | adults with a ladder score, sex, marital status, education and household income | interviews in 2009 to 2014 | 95% CI 48 to 54 | 1,355,746 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| turn_place_least | 54 | years of age | age of the lowest point of the five-year moving average of the mean ladder score by year of age (ages 20 to 92 searched, refined between whole years by a parabola through the lowest point and its neighbours), interval from a bootstrap over counties | adults with a ladder score in the least deprived third of interviews by county Area Deprivation Index | interviews in 2009 to 2017 | 95% CI 51 to 56 | 718,005 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| turn_place_mid | 54 | years of age | age of the lowest point of the five-year moving average of the mean ladder score by year of age (ages 20 to 92 searched, refined between whole years by a parabola through the lowest point and its neighbours), interval from a bootstrap over counties | adults with a ladder score in the middle third of interviews by county Area Deprivation Index | interviews in 2009 to 2017 | 95% CI 52 to 55 | 730,419 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| turn_place_most | 51 | years of age | age of the lowest point of the five-year moving average of the mean ladder score by year of age (ages 20 to 92 searched, refined between whole years by a parabola through the lowest point and its neighbours), interval from a bootstrap over counties | adults with a ladder score in the most deprived third of interviews by county Area Deprivation Index | interviews in 2009 to 2017 | 95% CI 48 to 54 | 731,406 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| turn_sex | 54 | years of age | age of the lowest point of the five-year moving average of the mean ladder score by year of age (ages 20 to 92 searched, refined between whole years by a parabola through the lowest point and its neighbours), interval from a bootstrap over counties | adults with a ladder score, sex, marital status, education and household income | interviews in 2009 to 2014 | 95% CI 51 to 55 | 1,355,746 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| worry_4554 | 35.4% | percent of adults who reported the feeling during a lot of the day yesterday | unweighted share of interviews that answered yes to the item, interval clustered by the respondent's county | adults aged 18 to 94 interviewed by Gallup who answered the item and whose county is known (age 45 to 54), worry | interviews in 2008 to 2017 | 95% CI 35.2% to 35.7% | 462,072 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| worry_7594 | 18.6% | percent of adults who reported the feeling during a lot of the day yesterday | unweighted share of interviews that answered yes to the item, interval clustered by the respondent's county | adults aged 18 to 94 interviewed by Gallup who answered the item and whose county is known (age 75 to 94), worry | interviews in 2008 to 2017 | 95% CI 18.4% to 18.8% | 342,365 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| worry_peak | 48 | years of age | age of the highest point of the five-year moving average of the share reporting the feeling by year of age (ages 20 to 92 searched, refined between whole years by a parabola), interval from a bootstrap over counties | adults aged 18 to 94 interviewed by Gallup who answered the item and whose county is known, worry | interviews in 2008 to 2017 | 95% CI 47 to 49 | 2,585,390 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| young_fall_adj | −0.71 | ladder points, later age band minus earlier age band | mean of the ladder profile at ages 45 to 54 minus the mean at ages 18 to 24, each year of age weighted by its number of interviews (a raw profile gives the difference of the two band means), interval from a bootstrap over counties; adjusted for sex, marital status, education, household income and survey year | adults with a ladder score, sex, marital status, education and household income | interviews in 2009 to 2014 | 95% CI −0.73 to −0.69 | 1,355,746 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |
| young_fall_cc_raw | −0.28 | ladder points, later age band minus earlier age band | mean of the ladder profile at ages 45 to 54 minus the mean at ages 18 to 24, each year of age weighted by its number of interviews (a raw profile gives the difference of the two band means), interval from a bootstrap over counties | adults with a ladder score, sex, marital status, education and household income | interviews in 2009 to 2014 | 95% CI −0.30 to −0.26 | 1,355,746 respondents | none (unweighted) | Gallup U.S. Daily Poll |   |

## Definitions

- Cantril ladder: A survey question that asks people to place their life on a ladder whose bottom step is the worst possible life and whose top step is the best possible. It measures how people evaluate their life as a whole, not how they felt yesterday.
- Survey weights: Numbers attached to survey respondents that say how many people in the population each one represents, so that results describe the whole population and not only the people who answered.
- Area Deprivation Index (ADI): A score that ranks neighborhoods by income, education, employment and housing conditions drawn from the American Community Survey. In this book a higher national percentile means a more deprived neighborhood.

## Methods

This chapter compares Gallup interviewees by year of age, age band, household income group, deprivation third and birth cohort. The extract has no survey weights, so all estimates are unweighted, and the county deprivation measure is dated after the interviews. The ladder of 2008 is set to missing, the income groups and adjusted profiles use interviews in 2009 to 2014 only, and the ladder and the feelings are different measures asked in different years and are never combined. Intervals are clustered by the respondent's county.

### Who was interviewed

Gallup respondent rows are read inside Box through the project loader and aggregated; nothing at respondent level is published. All estimates are unweighted.

Age is whole years at interview, 18 to 94 (the extract has no age above 94 and does not top-code it); interviews with no county are left out.

Explains Table 1 and Table 2.

### Which questions and years each measure covers

The ladder of 2008 is set to missing for every estimate, because political questions that preceded the ladder in 2008 and the first months of 2009 lowered it (Deaton 2012); the extract has no interview month, so early 2009 interviews may carry part of the effect. The feelings keep 2008.

The ladder was asked of 85% of interviews in 2009, of nearly all in 2010 to 2012 and of 49% from 2013 to 2017; the feelings follow a similar pattern in the years they were asked, so the pooled profiles weight 2008 to 2012 most.

The ladder (an evaluation of life as a whole) and the feelings (yes or no answers about the previous day) are different measures and are never combined.

### Why household income covers 2009 to 2014

Everything that uses household income, the income groups and the adjusted profiles, uses interviews in 2009 to 2014, because the share of income reporters at $90,000 or more rises from 23% in 2014 to 35% in 2017; the same contrasts for 2015 to 2017 are shown separately.

Household income is Gallup's harmonized six brackets in dollars of the interview year, collapsed to under $36,000, $36,000 to $89,999 and $90,000 or more.

Explains Table 4.

### How deprivation thirds are built

Deprivation thirds rank interviews by the Area Deprivation Index 2023 of the respondent's county (the mean of the county's tract national ranks weighted by 2019-2023 tract population; Connecticut tracts move to the eight legacy counties; Puerto Rico is excluded), a measure dated after the interviews, and cut them into three groups of about equal numbers of interviews, with a county in one group; the groups hold different numbers of counties.

Explains Table 3.

### How profiles, turning ages and cohorts are built

The age of the lowest or highest point of a profile is the minimum or maximum of its five-year moving average (each age averaged with the two ages on either side), searched over ages 20 to 92 and refined between whole years by the vertex of a parabola through the extreme and its two neighbours (at most half a year).

Adjusted profiles come from an unweighted least-squares regression of the outcome on a fixed effect for each year of age and fixed effects for sex, marital status (three groups), education (three groups), household income (six brackets) and survey year, and are evaluated at the sample's mix of the controls. The quadratic fit replaces the age effects with a constant, (age - 50) / 10 and its square, and its turning age is 50 - 5b/c for linear coefficient b and quadratic coefficient c, defined when c is positive.

The difference between age bands in a profile is the mean of the profile over the later band minus the mean over the earlier band, each year of age weighted by its number of interviews, so for a raw profile it is the difference of the two band means.

A birth cohort is the interview year minus age, in groups of ten birth years. The change among all adults that a cohort is compared with is age-standardized: the mean over single years of age 50 to 67 of the change at that age between 2009 and 2017, because the interview sample was younger in 2017 than in 2009.

Explains Table 5 and Table 6.

### How uncertainty is measured

Means and shares by year of age, age band, household income group, deprivation third or birth cohort have 95% intervals clustered by the respondent's county (derived from the ZIP code), from the linearized variance with counties as clusters and t distributions with the number of counties minus one degrees of freedom.

Intervals for turning ages, adjusted profiles and differences between age bands come from a bootstrap that resamples whole counties with replacement (2,000 replicates, seed 12345) and refits the regression or recomputes the group means in every replicate from the redrawn county sums.

## What these data cannot show

The profiles are differences between people of different ages in a repeated cross-section. They mix age, birth cohort and the years in which each group was interviewed, and the three cannot be separated with data from one decade. The oldest adults are survivors, so the oldest cells describe the people who lived to be interviewed there, and these data cannot say in which direction survival moves a profile. The panel studies in the box reach opposite conclusions about the middle-age dip, and these data cannot arbitrate between them.

Every estimate is unweighted, and the poll was a telephone survey, so the levels describe the people interviewed and not the national population. County comes from the respondent's ZIP code and is approximate near county lines. The feelings are yes or no answers about the previous day, so they measure how often a feeling was present and not how strong it was. The ladder of 2008 is left out and the 2009 ladder may carry part of the same question-order artifact, which the extract cannot separate without an interview month.

The adjusted profiles are statistical constructs. Household income, marital status and self-rated health are partly consequences of well-being, so adjusting for them does not isolate the part of the profile that belongs to age, and the profile is flat near its low point, so the age of that low is less certain than its bootstrap intervals suggest. Household income is nominal, includes a partner's earnings and was missing for a share of interviews. The mix of income brackets changes in 2015 to 2017 for reasons the extract does not document, so results by income use 2009 to 2014, and the cutoffs between groups are fixed dollar amounts across that period. The deprivation groups describe the county where a respondent lives and not the respondent, the groups hold very different numbers of counties, the index describes counties as of 2023, and the Area Deprivation Index understates deprivation in high-cost cities.

The poll ends in 2017, and later years may differ. Chapter 4.5, Mental distress in young adults (https://vishalsingh.org/zipcode/mental-distress-in-young-adults) follows frequent mental distress among young adults from 2014 in a different survey, and Chapter 6.2, Civic life, loneliness and partnership (https://vishalsingh.org/zipcode/civic-life-loneliness-and-partnership) shows how the ladder gap between partnered and other adults varies by age band in the same Gallup data. No figure in this chapter follows a person through life.

## References

1. Deaton, Angus (2012). The financial crisis and the well-being of Americans: 2011 OEP Hicks Lecture. Oxford Economic Papers 64(1): 1-26. https://doi.org/10.1093/oep/gpr051
2. Stone, Arthur A.; Schwartz, Joseph E.; Broderick, Joan E.; Deaton, Angus (2010). A snapshot of the age distribution of psychological well-being in the United States. Proceedings of the National Academy of Sciences 107(22): 9985-9990. https://doi.org/10.1073/pnas.1003744107
3. Blanchflower, David G.; Oswald, Andrew J. (2008). Is well-being U-shaped over the life cycle?. Social Science & Medicine 66(8): 1733-1749. https://doi.org/10.1016/j.socscimed.2008.01.030
4. Blanchflower, David G. (2021). Is happiness U-shaped everywhere? Age and subjective well-being in 145 countries. Journal of Population Economics 34(2): 575-624. https://doi.org/10.1007/s00148-020-00797-z
5. Frijters, Paul; Beatton, Tony (2012). The mystery of the U-shaped relationship between happiness and age. Journal of Economic Behavior & Organization 82(2-3): 525-542. https://doi.org/10.1016/j.jebo.2012.03.008
6. Glenn, Norval (2009). Is the apparent U-shape of well-being over the life course a result of inappropriate use of control variables? A commentary on Blanchflower and Oswald (66: 8, 2008, 1733-1749). Social Science & Medicine 69(4): 481-485. https://doi.org/10.1016/j.socscimed.2009.05.038
7. Cheng, Terence C.; Powdthavee, Nattavudh; Oswald, Andrew J. (2017). Longitudinal evidence for a midlife nadir in human well-being: results from four data sets. The Economic Journal 127(599): 126-142. https://doi.org/10.1111/ecoj.12256

## Provenance

- Pipeline version: 0.1.0
- Build script: book2/chapters/well-being-across-the-life-span/build.py (sha256 61acab29fa739f6467d5a08604479301bae5fe915e5b5939211633ffed034a57)
- Built on: 2026-10-07
- Site commit: c2b1ff3309cb4a10833f884da91ac0057a66ad0a

Inputs:

| File | Size (bytes) | SHA-256, first 1 MB | SHA-256, last 1 MB |
| --- | ---: | --- | --- |
| book2/chapters/well-being-across-the-life-span/age_data.py | 26,462 | ee6277fc08150c892abca17eed77a7a8987d6cc84ee016572d4e216fe50cf22e | ee6277fc08150c892abca17eed77a7a8987d6cc84ee016572d4e216fe50cf22e |
| data/analytic/tract_atlas.parquet | 39,659,925 | 29d5e4d943552d54aeb6ba92aadc939d9bd15bdea553c78357dbd516757be627 | 1e10d1ecf722340f23ee2212d4999a90250f85b65fa89b39e32db06d9472e22e |
| data/raw/gallup/gallup_us.parquet | 209,530,597 | b57c719f2c1d4749e43b1a70ca6dbd97f0480a0b538983091eea7eadfec40551 | b8167235969e50da6143a0e8d3e74ffb8f7c231743acc9fb2a78130a602af6f5 |
| data/staged/crosswalks/ct_tract2020_to_old_county.parquet | 15,842 | 961ef39f31596ef65fca2c67d5f2f63e533c5dc9b49745d3bb6437844e4b831d | 961ef39f31596ef65fca2c67d5f2f63e533c5dc9b49745d3bb6437844e4b831d |

Outputs:

| File | SHA-256 |
| --- | --- |
| facts.json | e17859489292c7610c9434e1c6eedcf4e5553f6255e457941076f413e3e3def5 |
| figures/cohort_lines.json | 9613d36a3bc83060273c5921a889be73c5c5ca968f14abc19bd8ef75efccc98f |
| figures/feelings_by_age.json | c36e1400b5722a7bee533f0d8bc7837d21e63b88922cdfe45cfc13c21efa213c |
| figures/ladder_by_age.json | 075f2582500b359a06a0e78157b386b161b7d5a46e6f57d3bb7e609d7fd0021f |
| figures/ladder_turning_points.json | 72017ea5c8022cdcf4e0e281bdd0704a62d213058881efb9aab8b0b3713fb98e |
| figures/profiles_by_income.json | 36a8d4b4e605720b3639e990eedb6482a78766e0171a60ad61aa12d02c8b3b06 |
| figures/profiles_by_place.json | e526a1644fe967dcc12dacac235a4be853de5f78613fa02f08ddd694dbb66b2a |
| meta.json | 602937578088c2357483d704f8328b0bb370e427afd1ed8ba89b45bb5834a2b1 |
| refs.json | b061ac27ab739f809f8ed3769b7a1c470af341fe7efbe8d983c66d7c6ca314d2 |

## Machine formats

- Report (HTML): https://vishalsingh.org/zipcode/well-being-across-the-life-span/report
- Report (Markdown): https://vishalsingh.org/zipcode/well-being-across-the-life-span/report.md
- Every estimate (JSON): https://vishalsingh.org/zipcode/report/well-being-across-the-life-span/facts.json
- Table 1, figure "ladder_by_age" (JSON): https://vishalsingh.org/zipcode/report/well-being-across-the-life-span/figures/ladder_by_age.json
- Table 2, figure "feelings_by_age" (JSON): https://vishalsingh.org/zipcode/report/well-being-across-the-life-span/figures/feelings_by_age.json
- Table 3, figure "profiles_by_place" (JSON): https://vishalsingh.org/zipcode/report/well-being-across-the-life-span/figures/profiles_by_place.json
- Table 4, figure "profiles_by_income" (JSON): https://vishalsingh.org/zipcode/report/well-being-across-the-life-span/figures/profiles_by_income.json
- Table 5, figure "ladder_turning_points" (JSON): https://vishalsingh.org/zipcode/report/well-being-across-the-life-span/figures/ladder_turning_points.json
- Table 6, figure "cohort_lines" (JSON): https://vishalsingh.org/zipcode/report/well-being-across-the-life-span/figures/cohort_lines.json
- Index of every report: https://vishalsingh.org/zipcode/llms.txt; all reports in one file: https://vishalsingh.org/zipcode/llms-full.txt

---

# Health behaviors and the supply of care: report

Chapter 7.1 of ZIP Code Destiny, by Vishal Singh (NYU Stern). Generated from the chapter's data files; the reading version is at https://vishalsingh.org/zipcode/health-behaviors-and-the-supply-of-care. Table n of this report is figure n of the chapter.

## Identification

- Title: Health behaviors and the supply of care
- Label: The body keeps the address
- Chapter: 7.1 (Part VII, Health and care; An American life)
- Question: How steeply do health behaviors and chronic conditions rise with neighborhood deprivation, and does the supply of doctors and dentists follow need?
- Dek: In the most deprived tenth of census tracts, the modeled 2022 share of adults with a past-year dental visit was 24.4 points below the least deprived tenth, and the routine-checkup share 3.6 points above.
- Grain: neighborhoods
- Status: published
- Published: 2026-10-09
- Updated: not revised
- Data sources:
  - CDC PLACES. CDC PLACES 2024 release, census tracts: modeled prevalence among adults for 40 measures (BRFSS 2022; 2021 for four measures), crude prevalence; years: 2024 release (tracts); 2025 release (ZCTAs, 2023 data year); weights: adult population of each tract (PLACES)
  - Area Deprivation Index. Neighborhood Atlas Area Deprivation Index 2023, national percentile, tract and county (county mean of tract ranks); years: 2023 release (version 4.0.1; built on ACS 2019-2023); weights: adult population of each tract; ACS 2019-2023 population for county means
  - County Health Rankings. County Health Rankings 2024: provider counts (physicians 2021, dentists 2022, mental health providers 2023), preventable hospital stays 2021, food environment index 2019 and 2021, percent rural; years: 2024 release; weights: pooled ratios of counts; county population for county means and correlations
  - American Community Survey. ACS 2019-2023 tracts: median household income and the shares of residents who are ages 65 and over, Black alone and Hispanic; years: 2019-2023 (5-year); weights: none (controls and the income axis)
- Canonical URL: https://vishalsingh.org/zipcode/health-behaviors-and-the-supply-of-care
- Report: https://vishalsingh.org/zipcode/health-behaviors-and-the-supply-of-care/report
- Markdown: https://vishalsingh.org/zipcode/health-behaviors-and-the-supply-of-care/report.md
- BibTeX key: singh2026healthbehaviorsandthesupplyofcare

```bibtex
@misc{singh2026healthbehaviorsandthesupplyofcare,
  author = {Singh, Vishal},
  title = {Health behaviors and the supply of care},
  year = {2026},
  url = {https://vishalsingh.org/zipcode/health-behaviors-and-the-supply-of-care},
  note = {In ZIP Code Destiny. Published 9 October 2026.}
}
```

## Principal estimates

The facts the dek and the opening prose cite, in order of appearance.

| Estimand | Population | Period | Value | Interval | n | Weights | Source | Fact id |
| --- | --- | --- | ---: | --- | ---: | --- | --- | --- |
| difference between the weighted mean prevalence of the least deprived and the most deprived tenth of tracts by ADI national percentile, bootstrap interval over counties (draws shared by the two means) | adults: dental visit in the past year | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 24.4 | 95% CI 22.5 to 26.5 | 8,213 tracts | adult population of the tract (PLACES) | CDC PLACES | flat_dental_delta |
| difference between the weighted mean prevalence of the least deprived and the most deprived tenth of tracts by ADI national percentile, bootstrap interval over counties (draws shared by the two means) | adults: routine checkup in the past year | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 3.6 | 95% CI 2.4 to 4.6 | 8,213 tracts | adult population of the tract (PLACES) | CDC PLACES | flat_checkup_delta |
| weighted mean of the tract prevalence in a tenth of tracts (equal numbers of tracts), bootstrap interval over counties | adults in the most deprived tenth of tracts by ADI national percentile: dental visit in the past year | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 45.8% | 95% CI 45.0% to 46.6% | 8,213 tracts | adult population of the tract (PLACES) | CDC PLACES | flat_dental_d10 |
| weighted mean of the tract prevalence in a tenth of tracts (equal numbers of tracts), bootstrap interval over counties | adults in the least deprived tenth of tracts by ADI national percentile: dental visit in the past year | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 70.3% | 95% CI 68.5% to 72.1% | 8,274 tracts | adult population of the tract (PLACES) | CDC PLACES | flat_dental_d1 |
| weighted mean of the tract prevalence in a tenth of tracts (equal numbers of tracts), bootstrap interval over counties | adults in the most deprived tenth of tracts by ADI national percentile: routine checkup in the past year | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 77.8% | 95% CI 77.3% to 78.2% | 8,213 tracts | adult population of the tract (PLACES) | CDC PLACES | checkup_hi |
| weighted mean of the tract prevalence in a tenth of tracts (equal numbers of tracts), bootstrap interval over counties | adults in the least deprived tenth of tracts by ADI national percentile: routine checkup in the past year | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 74.2% | 95% CI 73.2% to 75.3% | 8,274 tracts | adult population of the tract (PLACES) | CDC PLACES | checkup_lo |

## Findings

The share of adults who visited a dentist in the past year was 45.8% in the most deprived tenth of census tracts and 70.3% in the least deprived tenth. The share who had a routine checkup in the past year went the other way, 77.8% in the most deprived tenth against 74.2% in the least deprived. Both are modeled estimates from the 2024 release of CDC PLACES, which uses survey responses from 2022, and the tenths rank tracts on the Area Deprivation Index (ADI).[1][2] The chapter places those two measures among the other measures of health that rise with deprivation, the behaviors that sort by income, and the supply of physicians, dentists and mental health providers in deprived counties.

### Health and access measures rise with deprivation, and the checkup rises least

The share of adults who smoke was 9.1% (95% CI 8.5% to 9.6%) in the least deprived tenth of tracts and 22.8% (95% CI 22.5% to 23.1%) in the most deprived, 2.5 (95% CI 2.4 to 2.7) times as high. The share with diagnosed diabetes rose from 9.5% to 17.5%.

PLACES estimates 83,522 tracts, and 82,781 of them have an ADI rank, so the 741 without one are left out. Tenths hold from 8,118 to 8,511 tracts, because tracts with tied ranks fall in the same tenth.

Table 1 follows 12 measures of health, behavior and access across the whole index. They are this book's selection from the 40 that PLACES publishes for tracts: conditions and behaviors that the first edition followed, insurance, food insecurity and the checkup. For 11 of the 12 measures, the largest step between neighboring twentieths of tracts is the last one, at the deprived end of the index.

For the 11 measures other than the annual checkup, the ratio of prevalence in the most deprived tenth to the least deprived runs from 1.25 (95% CI 1.21 to 1.30) (depression) to 3.6 (95% CI 3.1 to 4.1) (all teeth lost among adults ages 65 and over).

Table 1. 11 measures are at least 4.8 points more common in the most deprived tenth of tracts than in the least deprived, and the annual checkup is 3.6 points more common

Among the twelve, the annual checkup rises least in ratio terms. Its rate was 74.2% (95% CI 73.2% to 75.3%) in the least deprived tenth and 77.8% (95% CI 77.3% to 78.2%) in the most deprived, a ratio of 1.05 (95% CI 1.03 to 1.06). The next smallest ratio of the twelve is 1.25 for depression. Of all 40 PLACES measures, 34 are higher in the most deprived tenth, and the checkup has the smallest ratio of them, with 1.08 for blood pressure medicine next.

In points the ranking differs. The checkup rises 3.6 (95% CI 2.4 to 4.6) points, and 4 measures outside the twelve (asthma, stroke, high cholesterol and social isolation) rise by less. Depression rises by 4.8 (95% CI 4.1 to 5.6) points, close to the checkup in points but from a lower base. The second view of the figure, in multiples of the least deprived tenth, separates the two.

Part of every PLACES gradient in this chapter is built in. PLACES does not count anyone. It fits a model to responses from the Behavioral Risk Factor Surveillance System (BRFSS) and applies the model to each tract's population. The model uses county poverty, the age, sex, race and ethnicity of the tract's residents and their education, with state and county random effects. The inputs come from the census and the American Community Survey.[1] The Area Deprivation Index is built from American Community Survey data as well,[2] so part of the association between the model's estimates and the index is mechanical. The shared inputs do not guarantee a contrast between two measures, such as the checkup and the dental visit. The model can still produce one through the demographic mix of tracts, because the survey responses it is fitted to differ by age, race and education. Within a county, the model gives tracts different values only through the demographic composition of their residents. Slopes within counties, and in part slopes within states and the coefficient zoo, therefore show how the model maps composition and county inputs into prevalence, and not what the residents of those tracts do. The sections below ask how much of each pattern survives when the state, the county and part of the mix are held fixed.

 (Note: The PLACES 2024 release uses BRFSS responses from 2022, and from 2021 for blood pressure, blood pressure medicine, high cholesterol and cholesterol screening. Values are crude prevalence among adults, so the age structure of a tract moves them. The tract file carries no intervals, and the bands in this chapter come from resampling counties.)

### Behaviors and conditions fall with income in every state except binge drinking, which rises

One reading of the gradient is that deprived places are sicker because their residents behave worse. The analysis uses 6 PLACES measures that bear on that reading, and they do not sort the same way. For each state, the prevalence of each across the state's tracts is regressed on the log of tract median household income, with the shares of residents who are older, Black and Hispanic held fixed, and each coefficient is scaled to a doubling of income. A set of such coefficients, 306 here, one per state and measure, is a coefficient zoo (Table 2), and Chapter 10.1, Gradients, slopes and variance (https://vishalsingh.org/zipcode/gradients-slopes-and-variance) takes up the method.

Smoking, obesity, diabetes, frequent mental distress and physical inactivity all have a negative slope in all 51 states. A doubling of tract income goes with a median of −4.7 points of smoking, −4.3 points of obesity and −6.7 points of physical inactivity.

For each of these five measures the interval lies below zero in at least 47 of the 51 states. Binge drinking has the opposite sign. Its slope is positive in 50 states, with a median of +1.2 points per doubling and an interval above zero in 41. Maryland is the one state with a negative estimate, and its interval includes zero.

Table 2. Binge drinking is higher in higher-income tracts in 50 of 51 states, while the other measures are lower in higher-income tracts in all 51 states

These are differences between tracts, and they do not say who drinks. Tracts with higher incomes have a higher modeled share of adults who binge drink, and tract values cannot show whether the drinkers are the high-income residents of those tracts. Survey estimates for individuals, with design-based intervals, are the subject of Chapter 7.2, Health behaviors, person by person (https://vishalsingh.org/zipcode/health-behaviors-person-by-person). The PLACES model and the census income measure share inputs, so the coefficients show what those inputs and the survey responses give together. Whatever the model contributes, the six measures do not share one direction with income, and an index of deprivation does not rank places on all of them alike.

### The annual checkup does not follow the gradient that the dental visit does

The checkup measures contact with a physician, and the dental visit measures routine dental care, which is mostly preventive. Table 3 plots both against the index. The dental visit rate falls from 70.3% (95% CI 68.5% to 72.1%) in the least deprived tenth of tracts to 45.8% (95% CI 45.0% to 46.6%) in the most deprived, 24.4 (95% CI 22.5 to 26.5) points lower. The checkup rate rises from 74.2% to 77.8%, 3.6 (95% CI 2.4 to 4.6) points higher. Across the most deprived tenth of the index the dental line falls fastest and the checkup line rises fastest.

Table 3. From the least to the most deprived tenth of tracts, the dental visit rate falls from 70.3% to 45.8% while the annual checkup rate rises from 74.2% to 77.8%

The same split appears across a wider ledger of care (Table 4). Among women ages 50 to 74, mammography is 7.0 (95% CI 6.3 to 7.8) points lower in the most deprived tenth. Cholesterol screening is 6.9 (95% CI 6.2 to 7.5) points lower, and colorectal cancer screening among adults ages 45 to 75 is 6.2 (95% CI 4.4 to 8.2) points lower.

Among adults ages 18 to 64, the share without health insurance is 17.4% (95% CI 16.1% to 18.9%) in the most deprived tenth against 6.3% (95% CI 5.7% to 6.8%) in the least deprived, 11.1 (95% CI 9.7 to 12.7) points higher.

The share of adults without reliable transportation is 9.7 (95% CI 9.0 to 10.5) points higher. Among adults with high blood pressure, the share taking medicine for it is 5.9 (95% CI 4.8 to 6.7) points higher, and the checkup, as above, is 3.6 points higher. Tract means weight each tract by all its adults, including for the measures that cover only some ages, because the file carries no population for them. The figure groups the measures as checkup and treatment, dental visit and screening, and barriers. The grouping is this book's, since PLACES files the checkup, blood pressure medicine, the dental visit and insurance together under prevention. Within it, the dental visit and the screening tests fall with deprivation, and the checkup and blood pressure medicine rise.

Table 4. In the most deprived tenth of tracts, the dental visit and screening tests are less common by 6.2 to 24.4 points and barriers to care are more common by 9.7 and 11.1 points

### Within states and counties the checkup slope stays near zero

The national checkup difference could reflect where deprived tracts are, because the checkup rate differs across states. Table 5 compares four ways of drawing the slope across 82,779 tracts that have an ADI rank, both measures and the three census shares. Across all tracts, the checkup rate rises +0.41 (95% CI 0.29 to 0.54) points for each ten-point step in the ADI percentile. Within states, with a separate baseline for each state (state fixed effects), the slope is −0.01 (95% CI −0.08 to 0.05), and within counties it is −0.27 (95% CI −0.35 to −0.19), a decline. Holding the tract's shares of older, Black and Hispanic residents fixed within states gives −0.05 (95% CI −0.10 to 0.01).

The dental slope is −2.37 (95% CI −2.66 to −2.08) across all tracts, −2.97 (95% CI −3.14 to −2.80) within states, −3.86 (95% CI −3.99 to −3.73) within counties and −2.38 (95% CI −2.51 to −2.24) within states with the controls. Within a state or a county the checkup slope is at most 0.27 in size, and the smallest dental slope is 8.8 times as large.

Table 5. The checkup slope is within 0.5 points of zero in all four specifications, and the dental slope is between −2.37 and −3.86 points

Fitting the slope separately in each state gives the same contrast (Table 6). The dental slope is negative in 51 of 51 states and has an interval below zero in 50, with a median of −3.2 points per ten-point step. It is within 0.5 points of zero in 0 states.

The checkup slope is within 0.5 points of zero in 42 states, with a median of +0.03. It is not zero everywhere. The interval lies below zero in 18 states, above zero in 10 and across zero in 23.

The largest checkup estimate is +1.7 (95% CI 1.3 to 2.1) in the District of Columbia, which is a single county, so its interval is not clustered. Every state has at least 158 tracts, above the floor of 60 tracts for entering the comparison, so none is dropped.

Table 6. The dental visit rate falls with deprivation in all 51 states, and the checkup slope is within 0.5 points of zero in 42

None of this removes the model's inputs, and a slope within counties still shows how the model maps composition into prevalence. What the comparison does show is that two measures that share every input, the checkup and the dental visit, have slopes that differ in size by a factor of 8.8 or more in every specification that holds the state or county fixed.

### The supply of providers is thinner in deprived counties

In 1971 the physician Julian Tudor Hart observed that the availability of good medical care tends to vary inversely with the need for it in the population served.[3] Read descriptively, the inverse care law predicts that counties with more deprivation, where measures of illness are higher, have fewer providers for their residents. County Health Rankings (CHR) counts providers by county: primary care physicians in 2021, dentists in 2022 and mental health providers in 2023.[4] Primary care physicians are practicing non-federal physicians in general practice, family medicine, internal medicine and pediatrics. Dentists are those with a National Provider Identifier. Mental health providers include psychiatrists, psychologists, social workers, counselors and therapists, and nurse practitioners who specialize in mental health. Across the 3,143 counties there is one primary care physician for every 1,334 residents, one dentist for every 1,361 and one mental health provider for every 319.

Ranking 3,128 counties on a county ADI, the population-weighted mean of tract ranks, and pooling each tenth, residents per primary care physician rise from 1,159 (95% CI 1,093 to 1,205) in the least deprived tenth of counties to 2,730 (95% CI 2,452 to 3,018) in the most deprived, a ratio of 2.4 (95% CI 2.1 to 2.6). For dentists the figures are 1,131 and 3,055, a ratio of 2.7 (95% CI 2.3 to 3.1).

For mental health providers they are 250 and 550, a ratio of 2.2 (95% CI 1.5 to 3.5). The most deprived tenth, though, has fewer residents per mental health provider than the ninth, and its interval is wide (Table 7).

Weighted by county residents, the county ADI correlates −0.41 (95% CI −0.46 to −0.35) with physicians per hundred thousand residents, −0.52 (95% CI −0.61 to −0.41) with dentists and −0.42 (95% CI −0.53 to −0.26) with mental health providers. Counting each county once, the correlations are −0.37 (95% CI −0.43 to −0.32), −0.35 (95% CI −0.45 to −0.27) and −0.33 (95% CI −0.44 to −0.23), smaller in size but still negative.

Table 7. Residents per primary care physician rise from 1,159 to 2,730 and per dentist from 1,131 to 3,055 between the least and most deprived tenths of counties

 (Note: Residents per provider here is a county group's residents divided by its providers, so a county with no provider adds residents and no providers. CHR leaves the ratio blank for a county with no provider when its population is above 2,000 for physicians, 4,000 for dentists and 1,000 for mental health providers, and reports a rate of zero for smaller counties. The first edition of this book dropped the blank ones and averaged the county ratios. That average is larger in every tenth for all three provider types, because small counties with one or two providers have very large ratios.)

Counties with no provider stay in these figures. Among all 3,143 counties there are 232 with no recorded primary care physician, 239 with no dentist and 216 with no mental health provider; the 15 counties without a deprivation rank are outside the tenths.

Table 8 maps residents per mental health provider. In 275 counties the ratio is above 3,200, and 91.6% of those counties are majority rural. Most lie in the South and the Midwest (96.0%). The 216 gray counties have no recorded provider, so the ratio is undefined for them and their supply is the thinnest of all.

Table 8. Residents per mental health provider exceed 3,200 in 275 counties, and 91.6% of those counties have a mostly rural population

Two other county measures move with the same index. Preventable hospital stays, admissions for conditions that outpatient care can often manage, were 2,370 (95% CI 2,188 to 2,599) per hundred thousand Medicare enrollees in the least deprived tenth of counties and 4,012 (95% CI 3,608 to 4,435) in the most deprived, and the county ADI correlates +0.46 (95% CI 0.31 to 0.56) with the rate, +0.38 (95% CI 0.30 to 0.46) counting each county once.

The CHR food environment index, a county score that combines access to healthy food with food insecurity and on which a higher value is better, falls from 8.6 (95% CI 8.5 to 8.8) to 6.1 (95% CI 5.7 to 6.6) across the same tenths, and correlates −0.66 (95% CI −0.74 to −0.59) with the county ADI, −0.51 (95% CI −0.58 to −0.45) counting each county once.

### Evidence on coverage and capacity

> Causal evidence: Coverage expansions and community health centers (Randomized lottery; difference-in-differences)
>
> The gradients above describe places. Four studies ask whether adding coverage or capacity changes the use of care and health for the people who gain it.
>
> In 2008 Oregon drew names by lottery for the chance to apply for Medicaid, so the people selected could be compared with the people who were not. In the first year, Medicaid coverage raised the use of health care, including primary and preventive care, lowered out-of-pocket medical spending and medical debt, and improved self-reported health.[5] At the later follow-up, coverage raised the use of preventive services, raised the detection and management of diabetes and lowered the rate of positive screens for depression, but produced no significant change in measured blood pressure, cholesterol or glycated hemoglobin.[6]
>
> Miller, Johnson and Wherry linked survey records to death records and compared near-elderly adults in states that expanded Medicaid under the Affordable Care Act with those in states that did not, before and after the expansions. This difference-in-differences design, checked with an event study of the years before the expansions, found lower annual mortality in the expansion states.[7] Bailey and Goodman-Bacon applied a difference-in-differences event study to the 1965 to 1974 rollout of community health centers, comparing counties that opened a center earlier with counties that opened one later, and found lower mortality among older adults in the counties that opened a center earlier.[8]
>
> The Oregon and Medicaid studies change insurance coverage, which is not the same as changing the number of dentists or mental health providers in a county, and the health center study changes clinic capacity. The estimates apply to the people and years each study covers, and none of the four measures what adding providers to the counties in the figures would do.

### Residents per primary care physician rise under all four rankings

Over the 3,125 counties that have all four rankings, the ADI baseline runs from 1,164 (95% CI 1,100 to 1,210) residents per primary care physician in the least deprived tenth of counties to 2,749 (95% CI 2,473 to 3,038) in the most deprived, with a rank correlation of +1.00 between tenth and ratio (Table 9). The standardized index, which applies the ADI's coefficients after standardizing each measure, gives 1,094 (95% CI 1,037 to 1,170) and 2,687 (95% CI 2,514 to 2,973), with a correlation of +0.99. Ranked by median household income the tenths run from 1,104 (95% CI 1,041 to 1,175) to 2,615 (95% CI 2,386 to 2,921), and ranked by the bachelor's degree share from 1,010 (95% CI 965 to 1,060) to 3,520 (95% CI 3,228 to 3,805). Under each ranking the two ends differ and the ratio rises steadily toward the deprived end, so none of the three alternatives changes the gradient under the book's rule (Chapter 9.1, How far one ranking goes (https://vishalsingh.org/zipcode/how-far-one-ranking-goes)).

Table 9. Residents per primary care physician is steadily worse toward the more deprived end under the ADI and under each of the three other rankings

### What these data cannot show

PLACES values are modeled estimates, and the tract file has no intervals, so the bands in this chapter describe the sampling of counties and not the model's own uncertainty. The model and the Area Deprivation Index draw on overlapping census inputs, which builds part of each gradient into the numbers, and the data cannot say how much. The values are crude prevalence among adults and are not adjusted for age. CDC advises against using them to rank the overall health of an area or to track change at the local level, which this chapter does not do.[9] The slopes and coefficients compare tracts and say nothing about the behavior of a person. Within a county the model gives tracts different values only through their demographic composition, so the comparison of the checkup and the dental visit within counties cannot separate behavior from composition at all.

The checkup measure is the share of adults who report a routine checkup in the past year, and the dental measure is the share who report a dental visit. Neither says what happened at the visit, whether the care was timely or whether it was affordable. Dental coverage is a separate matter from medical coverage, and the chapter has no measure of it.

Provider counts locate providers where they practice and divide by residents. They do not measure capacity, wait times or whether a provider accepts a patient's insurance, and county averages hide deserts inside counties and care that crosses county lines. The three provider counts come from different years, 2021 to 2023, and the PLACES estimates describe 2022 on an index built from 2019 to 2023 census data, so the sources do not describe one moment. Connecticut appears in PLACES under planning-region codes, which serve as its counties for the bootstrap, and in CHR under its eight legacy counties. 15 small counties have no deprivation rank and are left out of the tenths in the provider figures, and Puerto Rico is covered by neither PLACES nor the maps.

The association between supply and deprivation could run in either direction or through a third factor such as rurality, and the figures do not separate them. The studies in the box concern coverage and clinic capacity, and none of them estimates how a change in the supply of providers would alter the county gradients.

## Tables

### Table 1. 11 measures are at least 4.8 points more common in the most deprived tenth of tracts than in the least deprived, and the annual checkup is 3.6 points more common

Adults with each condition or behavior, percentage points above the least deprived tenth of tracts, by twentieth of tracts ranked on the Area Deprivation Index national percentile, 2022

| Measure | Mean ADI percentile | Prevalence (%) | Points above least deprived tenth | Lower limit of the 95% interval | Upper limit of the 95% interval | Multiple of least deprived tenth | Tracts |
| --- | ---: | ---: | ---: | ---: | ---: | ---: | ---: |
| Diagnosed diabetes | 3.5 | 9.1 | −0.3 | −0.5 | −0.1 | 0.96 | 4,307 |
| Diagnosed diabetes | 8.7 | 9.9 | 0.4 | 0.2 | 0.6 | 1.04 | 3,967 |
| Diagnosed diabetes | 13.7 | 10.2 | 0.7 | 0.3 | 1.1 | 1.07 | 4,246 |
| Diagnosed diabetes | 18.5 | 10.5 | 1.1 | 0.5 | 1.6 | 1.11 | 3,872 |
| Diagnosed diabetes | 23.3 | 10.3 | 0.8 | 0.5 | 1.2 | 1.09 | 4,314 |
| Diagnosed diabetes | 28.2 | 10.6 | 1.1 | 0.7 | 1.7 | 1.12 | 4,197 |
| Diagnosed diabetes | 33.0 | 10.7 | 1.2 | 0.8 | 1.7 | 1.13 | 4,082 |
| Diagnosed diabetes | 37.7 | 11.0 | 1.5 | 1.1 | 2.0 | 1.16 | 4,179 |
| Diagnosed diabetes | 42.5 | 11.3 | 1.8 | 1.4 | 2.3 | 1.19 | 3,944 |
| Diagnosed diabetes | 47.3 | 11.6 | 2.1 | 1.7 | 2.6 | 1.22 | 4,245 |
| Diagnosed diabetes | 52.1 | 11.9 | 2.4 | 2.0 | 3.0 | 1.26 | 4,090 |
| Diagnosed diabetes | 56.8 | 12.3 | 2.8 | 2.3 | 3.3 | 1.29 | 4,228 |
| Diagnosed diabetes | 61.5 | 12.6 | 3.1 | 2.6 | 3.7 | 1.33 | 4,008 |
| Diagnosed diabetes | 66.2 | 13.0 | 3.5 | 3.0 | 4.1 | 1.37 | 4,297 |
| Diagnosed diabetes | 71.0 | 13.4 | 3.9 | 3.4 | 4.5 | 1.41 | 4,044 |
| Diagnosed diabetes | 75.8 | 13.8 | 4.4 | 3.9 | 4.9 | 1.46 | 4,194 |
| Diagnosed diabetes | 80.7 | 14.4 | 4.9 | 4.4 | 5.5 | 1.52 | 4,171 |
| Diagnosed diabetes | 85.9 | 15.2 | 5.7 | 5.2 | 6.3 | 1.60 | 4,183 |
| Diagnosed diabetes | 91.3 | 16.4 | 6.9 | 6.4 | 7.5 | 1.73 | 4,206 |
| Diagnosed diabetes | 97.1 | 18.9 | 9.5 | 8.9 | 10.1 | 2.00 | 4,007 |
| Obesity | 3.5 | 22.5 | −1.4 | −1.7 | −1.1 | 0.94 | 4,307 |
| Obesity | 8.7 | 25.4 | 1.5 | 1.0 | 2.0 | 1.06 | 3,967 |
| Obesity | 13.7 | 27.4 | 3.5 | 2.9 | 4.2 | 1.15 | 4,246 |
| Obesity | 18.5 | 28.8 | 4.9 | 4.1 | 5.6 | 1.21 | 3,872 |
| Obesity | 23.3 | 29.8 | 6.0 | 5.1 | 6.7 | 1.25 | 4,314 |
| Obesity | 28.2 | 31.1 | 7.3 | 6.3 | 8.1 | 1.30 | 4,197 |
| Obesity | 33.0 | 31.7 | 7.9 | 7.0 | 8.7 | 1.33 | 4,082 |
| Obesity | 37.7 | 32.6 | 8.7 | 7.8 | 9.5 | 1.37 | 4,179 |
| Obesity | 42.5 | 33.5 | 9.6 | 8.7 | 10.4 | 1.40 | 3,944 |
| Obesity | 47.3 | 34.2 | 10.4 | 9.4 | 11.1 | 1.43 | 4,245 |
| Obesity | 52.1 | 35.0 | 11.1 | 10.2 | 11.9 | 1.47 | 4,090 |
| Obesity | 56.8 | 35.8 | 11.9 | 11.0 | 12.7 | 1.50 | 4,228 |
| Obesity | 61.5 | 36.4 | 12.5 | 11.6 | 13.3 | 1.53 | 4,008 |
| Obesity | 66.2 | 37.3 | 13.4 | 12.5 | 14.2 | 1.56 | 4,297 |
| Obesity | 71.0 | 38.0 | 14.2 | 13.3 | 15.0 | 1.59 | 4,044 |
| Obesity | 75.8 | 38.7 | 14.8 | 13.9 | 15.6 | 1.62 | 4,194 |
| Obesity | 80.7 | 39.6 | 15.8 | 14.9 | 16.6 | 1.66 | 4,171 |
| Obesity | 85.9 | 40.6 | 16.8 | 15.9 | 17.5 | 1.70 | 4,183 |
| Obesity | 91.3 | 42.3 | 18.4 | 17.5 | 19.2 | 1.77 | 4,206 |
| Obesity | 97.1 | 45.8 | 21.9 | 20.9 | 22.8 | 1.92 | 4,007 |
| Current smoking | 3.5 | 8.4 | −0.6 | −0.8 | −0.4 | 0.93 | 4,307 |
| Current smoking | 8.7 | 9.7 | 0.7 | 0.5 | 0.9 | 1.07 | 3,967 |
| Current smoking | 13.7 | 10.5 | 1.4 | 0.8 | 1.8 | 1.16 | 4,246 |
| Current smoking | 18.5 | 11.0 | 2.0 | 1.4 | 2.5 | 1.22 | 3,872 |
| Current smoking | 23.3 | 11.3 | 2.2 | 1.6 | 2.7 | 1.25 | 4,314 |
| Current smoking | 28.2 | 11.9 | 2.9 | 2.3 | 3.5 | 1.32 | 4,197 |
| Current smoking | 33.0 | 12.3 | 3.3 | 2.7 | 3.8 | 1.36 | 4,082 |
| Current smoking | 37.7 | 13.0 | 3.9 | 3.3 | 4.5 | 1.43 | 4,179 |
| Current smoking | 42.5 | 13.5 | 4.5 | 3.9 | 5.0 | 1.49 | 3,944 |
| Current smoking | 47.3 | 14.2 | 5.2 | 4.6 | 5.7 | 1.57 | 4,245 |
| Current smoking | 52.1 | 14.8 | 5.7 | 5.2 | 6.3 | 1.63 | 4,090 |
| Current smoking | 56.8 | 15.4 | 6.3 | 5.8 | 6.9 | 1.70 | 4,228 |
| Current smoking | 61.5 | 16.0 | 6.9 | 6.4 | 7.5 | 1.76 | 4,008 |
| Current smoking | 66.2 | 16.8 | 7.7 | 7.2 | 8.3 | 1.85 | 4,297 |
| Current smoking | 71.0 | 17.4 | 8.3 | 7.8 | 8.9 | 1.92 | 4,044 |
| Current smoking | 75.8 | 18.1 | 9.1 | 8.5 | 9.6 | 2.00 | 4,194 |
| Current smoking | 80.7 | 19.0 | 9.9 | 9.3 | 10.4 | 2.09 | 4,171 |
| Current smoking | 85.9 | 20.0 | 10.9 | 10.3 | 11.5 | 2.20 | 4,183 |
| Current smoking | 91.3 | 21.4 | 12.3 | 11.7 | 12.9 | 2.36 | 4,206 |
| Current smoking | 97.1 | 24.6 | 15.5 | 14.8 | 16.2 | 2.71 | 4,007 |
| Depression | 3.5 | 19.3 | −0.2 | −0.5 | 0.0 | 0.99 | 4,307 |
| Depression | 8.7 | 19.8 | 0.3 | 0.0 | 0.6 | 1.01 | 3,967 |
| Depression | 13.7 | 20.4 | 0.8 | 0.4 | 1.3 | 1.04 | 4,246 |
| Depression | 18.5 | 20.7 | 1.1 | 0.7 | 1.6 | 1.06 | 3,872 |
| Depression | 23.3 | 21.3 | 1.8 | 1.2 | 2.4 | 1.09 | 4,314 |
| Depression | 28.2 | 21.4 | 1.9 | 1.2 | 2.5 | 1.09 | 4,197 |
| Depression | 33.0 | 21.7 | 2.2 | 1.5 | 2.9 | 1.11 | 4,082 |
| Depression | 37.7 | 22.0 | 2.4 | 1.8 | 3.1 | 1.12 | 4,179 |
| Depression | 42.5 | 22.2 | 2.6 | 1.9 | 3.3 | 1.13 | 3,944 |
| Depression | 47.3 | 22.5 | 3.0 | 2.3 | 3.7 | 1.15 | 4,245 |
| Depression | 52.1 | 22.8 | 3.2 | 2.5 | 4.0 | 1.17 | 4,090 |
| Depression | 56.8 | 23.1 | 3.5 | 2.8 | 4.2 | 1.18 | 4,228 |
| Depression | 61.5 | 23.2 | 3.6 | 2.9 | 4.4 | 1.19 | 4,008 |
| Depression | 66.2 | 23.5 | 4.0 | 3.3 | 4.7 | 1.20 | 4,297 |
| Depression | 71.0 | 23.7 | 4.2 | 3.5 | 4.9 | 1.22 | 4,044 |
| Depression | 75.8 | 24.0 | 4.4 | 3.7 | 5.2 | 1.23 | 4,194 |
| Depression | 80.7 | 24.2 | 4.7 | 4.0 | 5.4 | 1.24 | 4,171 |
| Depression | 85.9 | 24.4 | 4.9 | 4.2 | 5.6 | 1.25 | 4,183 |
| Depression | 91.3 | 24.4 | 4.9 | 4.2 | 5.6 | 1.25 | 4,206 |
| Depression | 97.1 | 24.3 | 4.8 | 4.1 | 5.6 | 1.25 | 4,007 |
| Frequent mental distress | 3.5 | 13.4 | −0.4 | −0.6 | −0.2 | 0.97 | 4,307 |
| Frequent mental distress | 8.7 | 14.3 | 0.5 | 0.3 | 0.7 | 1.03 | 3,967 |
| Frequent mental distress | 13.7 | 15.0 | 1.1 | 0.7 | 1.5 | 1.08 | 4,246 |
| Frequent mental distress | 18.5 | 15.3 | 1.5 | 1.0 | 2.0 | 1.11 | 3,872 |
| Frequent mental distress | 23.3 | 15.5 | 1.7 | 1.3 | 2.0 | 1.12 | 4,314 |
| Frequent mental distress | 28.2 | 15.7 | 1.9 | 1.4 | 2.3 | 1.13 | 4,197 |
| Frequent mental distress | 33.0 | 15.9 | 2.1 | 1.6 | 2.5 | 1.15 | 4,082 |
| Frequent mental distress | 37.7 | 16.2 | 2.4 | 1.9 | 2.9 | 1.17 | 4,179 |
| Frequent mental distress | 42.5 | 16.5 | 2.6 | 2.2 | 3.1 | 1.19 | 3,944 |
| Frequent mental distress | 47.3 | 16.7 | 2.9 | 2.5 | 3.4 | 1.21 | 4,245 |
| Frequent mental distress | 52.1 | 17.1 | 3.3 | 2.9 | 3.8 | 1.24 | 4,090 |
| Frequent mental distress | 56.8 | 17.4 | 3.5 | 3.1 | 4.0 | 1.25 | 4,228 |
| Frequent mental distress | 61.5 | 17.7 | 3.9 | 3.4 | 4.3 | 1.28 | 4,008 |
| Frequent mental distress | 66.2 | 18.0 | 4.2 | 3.8 | 4.7 | 1.30 | 4,297 |
| Frequent mental distress | 71.0 | 18.4 | 4.5 | 4.1 | 5.0 | 1.33 | 4,044 |
| Frequent mental distress | 75.8 | 18.7 | 4.9 | 4.5 | 5.4 | 1.35 | 4,194 |
| Frequent mental distress | 80.7 | 19.0 | 5.2 | 4.8 | 5.7 | 1.37 | 4,171 |
| Frequent mental distress | 85.9 | 19.5 | 5.7 | 5.3 | 6.2 | 1.41 | 4,183 |
| Frequent mental distress | 91.3 | 20.3 | 6.4 | 6.0 | 6.9 | 1.46 | 4,206 |
| Frequent mental distress | 97.1 | 21.9 | 8.1 | 7.6 | 8.6 | 1.58 | 4,007 |
| No leisure-time physical activity | 3.5 | 16.8 | −1.0 | −1.5 | −0.5 | 0.94 | 4,307 |
| No leisure-time physical activity | 8.7 | 18.9 | 1.1 | 0.5 | 1.5 | 1.06 | 3,967 |
| No leisure-time physical activity | 13.7 | 19.8 | 2.0 | 0.6 | 3.3 | 1.11 | 4,246 |
| No leisure-time physical activity | 18.5 | 20.8 | 3.0 | 1.3 | 4.6 | 1.17 | 3,872 |
| No leisure-time physical activity | 23.3 | 20.5 | 2.7 | 1.2 | 3.9 | 1.15 | 4,314 |
| No leisure-time physical activity | 28.2 | 21.4 | 3.5 | 2.0 | 5.3 | 1.20 | 4,197 |
| No leisure-time physical activity | 33.0 | 21.5 | 3.7 | 2.3 | 5.0 | 1.21 | 4,082 |
| No leisure-time physical activity | 37.7 | 22.3 | 4.5 | 3.1 | 5.9 | 1.25 | 4,179 |
| No leisure-time physical activity | 42.5 | 23.0 | 5.2 | 3.8 | 6.6 | 1.29 | 3,944 |
| No leisure-time physical activity | 47.3 | 23.8 | 6.0 | 4.5 | 7.3 | 1.33 | 4,245 |
| No leisure-time physical activity | 52.1 | 24.7 | 6.9 | 5.5 | 8.3 | 1.39 | 4,090 |
| No leisure-time physical activity | 56.8 | 25.4 | 7.6 | 6.2 | 9.0 | 1.43 | 4,228 |
| No leisure-time physical activity | 61.5 | 26.3 | 8.5 | 7.1 | 9.9 | 1.48 | 4,008 |
| No leisure-time physical activity | 66.2 | 27.3 | 9.5 | 8.1 | 10.9 | 1.53 | 4,297 |
| No leisure-time physical activity | 71.0 | 28.2 | 10.4 | 9.0 | 11.8 | 1.59 | 4,044 |
| No leisure-time physical activity | 75.8 | 29.2 | 11.4 | 10.1 | 12.8 | 1.64 | 4,194 |
| No leisure-time physical activity | 80.7 | 30.4 | 12.6 | 11.2 | 14.0 | 1.71 | 4,171 |
| No leisure-time physical activity | 85.9 | 32.0 | 14.2 | 12.8 | 15.6 | 1.80 | 4,183 |
| No leisure-time physical activity | 91.3 | 34.3 | 16.5 | 15.1 | 17.9 | 1.93 | 4,206 |
| No leisure-time physical activity | 97.1 | 38.7 | 20.9 | 19.5 | 22.3 | 2.17 | 4,007 |
| Short sleep duration | 3.5 | 31.9 | −0.8 | −1.1 | −0.5 | 0.97 | 4,307 |
| Short sleep duration | 8.7 | 33.6 | 0.9 | 0.5 | 1.3 | 1.03 | 3,967 |
| Short sleep duration | 13.7 | 34.2 | 1.5 | 0.5 | 2.2 | 1.04 | 4,246 |
| Short sleep duration | 18.5 | 34.6 | 1.8 | 0.7 | 2.8 | 1.06 | 3,872 |
| Short sleep duration | 23.3 | 34.4 | 1.6 | 0.5 | 2.7 | 1.05 | 4,314 |
| Short sleep duration | 28.2 | 34.7 | 2.0 | 0.8 | 3.2 | 1.06 | 4,197 |
| Short sleep duration | 33.0 | 34.8 | 2.1 | 0.9 | 3.2 | 1.06 | 4,082 |
| Short sleep duration | 37.7 | 35.3 | 2.6 | 1.4 | 3.8 | 1.08 | 4,179 |
| Short sleep duration | 42.5 | 35.5 | 2.8 | 1.6 | 4.0 | 1.09 | 3,944 |
| Short sleep duration | 47.3 | 35.9 | 3.2 | 2.0 | 4.3 | 1.10 | 4,245 |
| Short sleep duration | 52.1 | 36.4 | 3.7 | 2.5 | 4.9 | 1.11 | 4,090 |
| Short sleep duration | 56.8 | 36.7 | 4.0 | 2.8 | 5.2 | 1.12 | 4,228 |
| Short sleep duration | 61.5 | 37.2 | 4.5 | 3.2 | 5.6 | 1.14 | 4,008 |
| Short sleep duration | 66.2 | 37.5 | 4.8 | 3.6 | 6.0 | 1.15 | 4,297 |
| Short sleep duration | 71.0 | 38.0 | 5.3 | 4.1 | 6.5 | 1.16 | 4,044 |
| Short sleep duration | 75.8 | 38.4 | 5.7 | 4.5 | 6.9 | 1.18 | 4,194 |
| Short sleep duration | 80.7 | 38.9 | 6.2 | 5.0 | 7.4 | 1.19 | 4,171 |
| Short sleep duration | 85.9 | 39.7 | 7.0 | 5.8 | 8.2 | 1.21 | 4,183 |
| Short sleep duration | 91.3 | 41.1 | 8.4 | 7.2 | 9.6 | 1.26 | 4,206 |
| Short sleep duration | 97.1 | 44.4 | 11.7 | 10.5 | 13.1 | 1.36 | 4,007 |
| Food insecurity | 3.5 | 8.3 | −1.0 | −1.7 | −0.3 | 0.89 | 3,228 |
| Food insecurity | 8.7 | 10.5 | 1.2 | 0.3 | 1.9 | 1.13 | 2,713 |
| Food insecurity | 13.7 | 12.3 | 2.9 | 1.1 | 4.9 | 1.31 | 2,931 |
| Food insecurity | 18.5 | 13.4 | 4.1 | 1.8 | 6.6 | 1.44 | 2,860 |
| Food insecurity | 23.3 | 12.4 | 3.0 | 1.8 | 4.7 | 1.32 | 3,253 |
| Food insecurity | 28.2 | 12.4 | 3.0 | 2.1 | 4.2 | 1.32 | 3,173 |
| Food insecurity | 33.0 | 12.8 | 3.5 | 2.5 | 4.9 | 1.38 | 3,183 |
| Food insecurity | 37.7 | 13.6 | 4.3 | 3.0 | 6.0 | 1.46 | 3,348 |
| Food insecurity | 42.5 | 14.1 | 4.8 | 3.6 | 6.4 | 1.51 | 3,125 |
| Food insecurity | 47.3 | 14.6 | 5.3 | 3.9 | 7.0 | 1.57 | 3,362 |
| Food insecurity | 52.1 | 15.4 | 6.1 | 4.7 | 7.8 | 1.65 | 3,211 |
| Food insecurity | 56.8 | 15.9 | 6.6 | 5.1 | 8.4 | 1.71 | 3,394 |
| Food insecurity | 61.5 | 16.9 | 7.5 | 6.0 | 9.3 | 1.80 | 3,104 |
| Food insecurity | 66.2 | 17.6 | 8.2 | 6.6 | 10.2 | 1.88 | 3,329 |
| Food insecurity | 71.0 | 18.3 | 9.0 | 7.5 | 10.7 | 1.96 | 3,194 |
| Food insecurity | 75.8 | 19.4 | 10.0 | 8.5 | 11.8 | 2.08 | 3,238 |
| Food insecurity | 80.7 | 20.2 | 10.8 | 9.3 | 12.6 | 2.16 | 3,183 |
| Food insecurity | 85.9 | 22.1 | 12.8 | 11.3 | 14.5 | 2.37 | 3,193 |
| Food insecurity | 91.3 | 25.7 | 16.4 | 14.8 | 18.3 | 2.75 | 3,247 |
| Food insecurity | 97.1 | 33.8 | 24.4 | 22.7 | 26.3 | 3.62 | 3,101 |
| No health insurance, ages 18 to 64 | 3.5 | 5.6 | −0.8 | −1.1 | −0.4 | 0.88 | 4,307 |
| No health insurance, ages 18 to 64 | 8.7 | 7.1 | 0.8 | 0.4 | 1.2 | 1.13 | 3,967 |
| No health insurance, ages 18 to 64 | 13.7 | 8.2 | 1.9 | 0.8 | 3.0 | 1.29 | 4,246 |
| No health insurance, ages 18 to 64 | 18.5 | 9.3 | 3.0 | 1.7 | 4.6 | 1.48 | 3,872 |
| No health insurance, ages 18 to 64 | 23.3 | 9.1 | 2.8 | 1.8 | 3.8 | 1.44 | 4,314 |
| No health insurance, ages 18 to 64 | 28.2 | 9.7 | 3.4 | 2.5 | 4.3 | 1.54 | 4,197 |
| No health insurance, ages 18 to 64 | 33.0 | 10.0 | 3.7 | 2.9 | 4.5 | 1.58 | 4,082 |
| No health insurance, ages 18 to 64 | 37.7 | 10.6 | 4.3 | 3.4 | 5.3 | 1.68 | 4,179 |
| No health insurance, ages 18 to 64 | 42.5 | 11.0 | 4.7 | 3.8 | 5.6 | 1.74 | 3,944 |
| No health insurance, ages 18 to 64 | 47.3 | 11.2 | 4.9 | 4.0 | 5.8 | 1.77 | 4,245 |
| No health insurance, ages 18 to 64 | 52.1 | 11.8 | 5.5 | 4.5 | 6.5 | 1.87 | 4,090 |
| No health insurance, ages 18 to 64 | 56.8 | 12.1 | 5.8 | 4.7 | 6.9 | 1.92 | 4,228 |
| No health insurance, ages 18 to 64 | 61.5 | 12.8 | 6.5 | 5.3 | 7.7 | 2.03 | 4,008 |
| No health insurance, ages 18 to 64 | 66.2 | 12.8 | 6.5 | 5.3 | 7.8 | 2.02 | 4,297 |
| No health insurance, ages 18 to 64 | 71.0 | 13.1 | 6.8 | 5.7 | 8.1 | 2.08 | 4,044 |
| No health insurance, ages 18 to 64 | 75.8 | 13.6 | 7.3 | 6.1 | 8.7 | 2.16 | 4,194 |
| No health insurance, ages 18 to 64 | 80.7 | 13.9 | 7.6 | 6.5 | 8.9 | 2.21 | 4,171 |
| No health insurance, ages 18 to 64 | 85.9 | 14.8 | 8.4 | 7.3 | 9.8 | 2.34 | 4,183 |
| No health insurance, ages 18 to 64 | 91.3 | 16.5 | 10.2 | 8.7 | 11.9 | 2.62 | 4,206 |
| No health insurance, ages 18 to 64 | 97.1 | 18.5 | 12.2 | 10.8 | 14.0 | 2.94 | 4,007 |
| All teeth lost, ages 65 and over | 3.5 | 6.6 | −0.6 | −0.9 | −0.3 | 0.92 | 4,307 |
| All teeth lost, ages 65 and over | 8.7 | 7.8 | 0.7 | 0.3 | 0.9 | 1.09 | 3,967 |
| All teeth lost, ages 65 and over | 13.7 | 8.5 | 1.4 | 0.2 | 2.3 | 1.19 | 4,246 |
| All teeth lost, ages 65 and over | 18.5 | 9.2 | 2.0 | 0.7 | 3.0 | 1.28 | 3,872 |
| All teeth lost, ages 65 and over | 23.3 | 9.3 | 2.1 | 0.9 | 3.1 | 1.29 | 4,314 |
| All teeth lost, ages 65 and over | 28.2 | 9.9 | 2.7 | 1.4 | 3.9 | 1.37 | 4,197 |
| All teeth lost, ages 65 and over | 33.0 | 10.1 | 2.9 | 1.8 | 3.7 | 1.40 | 4,082 |
| All teeth lost, ages 65 and over | 37.7 | 10.6 | 3.4 | 2.2 | 4.3 | 1.48 | 4,179 |
| All teeth lost, ages 65 and over | 42.5 | 11.3 | 4.1 | 2.9 | 5.0 | 1.57 | 3,944 |
| All teeth lost, ages 65 and over | 47.3 | 11.9 | 4.7 | 3.6 | 5.7 | 1.66 | 4,245 |
| All teeth lost, ages 65 and over | 52.1 | 12.7 | 5.5 | 4.4 | 6.5 | 1.77 | 4,090 |
| All teeth lost, ages 65 and over | 56.8 | 13.4 | 6.3 | 5.1 | 7.2 | 1.87 | 4,228 |
| All teeth lost, ages 65 and over | 61.5 | 14.3 | 7.1 | 5.9 | 8.0 | 1.98 | 4,008 |
| All teeth lost, ages 65 and over | 66.2 | 15.3 | 8.1 | 6.9 | 9.1 | 2.13 | 4,297 |
| All teeth lost, ages 65 and over | 71.0 | 16.2 | 9.0 | 7.8 | 9.9 | 2.25 | 4,044 |
| All teeth lost, ages 65 and over | 75.8 | 17.4 | 10.2 | 9.1 | 11.1 | 2.42 | 4,194 |
| All teeth lost, ages 65 and over | 80.7 | 18.5 | 11.3 | 10.1 | 12.2 | 2.57 | 4,171 |
| All teeth lost, ages 65 and over | 85.9 | 20.3 | 13.1 | 12.0 | 14.0 | 2.83 | 4,183 |
| All teeth lost, ages 65 and over | 91.3 | 23.0 | 15.8 | 14.6 | 16.7 | 3.20 | 4,206 |
| All teeth lost, ages 65 and over | 97.1 | 29.5 | 22.3 | 21.1 | 23.4 | 4.11 | 4,007 |
| Fair or poor health | 3.5 | 12.7 | −0.8 | −1.2 | −0.3 | 0.94 | 4,307 |
| Fair or poor health | 8.7 | 14.3 | 0.8 | 0.3 | 1.2 | 1.06 | 3,967 |
| Fair or poor health | 13.7 | 15.2 | 1.7 | 0.6 | 2.9 | 1.13 | 4,246 |
| Fair or poor health | 18.5 | 16.1 | 2.7 | 1.2 | 4.3 | 1.20 | 3,872 |
| Fair or poor health | 23.3 | 15.7 | 2.2 | 1.2 | 3.2 | 1.16 | 4,314 |
| Fair or poor health | 28.2 | 16.2 | 2.8 | 1.7 | 4.1 | 1.21 | 4,197 |
| Fair or poor health | 33.0 | 16.2 | 2.8 | 1.9 | 3.8 | 1.21 | 4,082 |
| Fair or poor health | 37.7 | 16.9 | 3.4 | 2.5 | 4.5 | 1.25 | 4,179 |
| Fair or poor health | 42.5 | 17.4 | 3.9 | 3.0 | 5.0 | 1.29 | 3,944 |
| Fair or poor health | 47.3 | 18.0 | 4.5 | 3.6 | 5.6 | 1.33 | 4,245 |
| Fair or poor health | 52.1 | 18.7 | 5.2 | 4.3 | 6.3 | 1.39 | 4,090 |
| Fair or poor health | 56.8 | 19.3 | 5.8 | 4.9 | 7.0 | 1.43 | 4,228 |
| Fair or poor health | 61.5 | 20.2 | 6.7 | 5.7 | 7.8 | 1.50 | 4,008 |
| Fair or poor health | 66.2 | 20.9 | 7.5 | 6.5 | 8.6 | 1.55 | 4,297 |
| Fair or poor health | 71.0 | 21.7 | 8.2 | 7.3 | 9.3 | 1.61 | 4,044 |
| Fair or poor health | 75.8 | 22.7 | 9.2 | 8.3 | 10.4 | 1.69 | 4,194 |
| Fair or poor health | 80.7 | 23.7 | 10.3 | 9.3 | 11.4 | 1.76 | 4,171 |
| Fair or poor health | 85.9 | 25.3 | 11.9 | 10.9 | 13.0 | 1.88 | 4,183 |
| Fair or poor health | 91.3 | 27.8 | 14.4 | 13.3 | 15.6 | 2.07 | 4,206 |
| Fair or poor health | 97.1 | 32.9 | 19.5 | 18.4 | 20.6 | 2.45 | 4,007 |
| Annual checkup | 3.5 | 74.0 | −0.1 | −0.6 | 0.4 | 1.00 | 4,307 |
| Annual checkup | 8.7 | 74.3 | 0.1 | −0.4 | 0.8 | 1.00 | 3,967 |
| Annual checkup | 13.7 | 74.3 | 0.2 | −0.7 | 1.2 | 1.00 | 4,246 |
| Annual checkup | 18.5 | 74.2 | 0.1 | −0.9 | 1.2 | 1.00 | 3,872 |
| Annual checkup | 23.3 | 74.5 | 0.3 | −0.6 | 1.3 | 1.00 | 4,314 |
| Annual checkup | 28.2 | 74.8 | 0.7 | −0.4 | 1.7 | 1.01 | 4,197 |
| Annual checkup | 33.0 | 74.8 | 0.7 | −0.4 | 1.6 | 1.01 | 4,082 |
| Annual checkup | 37.7 | 75.0 | 0.8 | −0.3 | 1.8 | 1.01 | 4,179 |
| Annual checkup | 42.5 | 75.3 | 1.1 | 0.0 | 2.1 | 1.01 | 3,944 |
| Annual checkup | 47.3 | 75.6 | 1.5 | 0.3 | 2.5 | 1.02 | 4,245 |
| Annual checkup | 52.1 | 75.7 | 1.6 | 0.4 | 2.5 | 1.02 | 4,090 |
| Annual checkup | 56.8 | 76.0 | 1.8 | 0.7 | 2.8 | 1.02 | 4,228 |
| Annual checkup | 61.5 | 76.0 | 1.9 | 0.7 | 2.9 | 1.02 | 4,008 |
| Annual checkup | 66.2 | 76.4 | 2.2 | 1.1 | 3.2 | 1.03 | 4,297 |
| Annual checkup | 71.0 | 76.6 | 2.4 | 1.2 | 3.4 | 1.03 | 4,044 |
| Annual checkup | 75.8 | 76.7 | 2.5 | 1.4 | 3.5 | 1.03 | 4,194 |
| Annual checkup | 80.7 | 77.0 | 2.9 | 1.7 | 3.9 | 1.04 | 4,171 |
| Annual checkup | 85.9 | 77.2 | 3.1 | 1.9 | 4.1 | 1.04 | 4,183 |
| Annual checkup | 91.3 | 77.4 | 3.3 | 2.1 | 4.3 | 1.04 | 4,206 |
| Annual checkup | 97.1 | 78.2 | 4.0 | 2.8 | 5.0 | 1.05 | 4,007 |

Views: Points above the least deprived tenth (every row); Multiple of the least deprived tenth (every row).

Note: PLACES values are modeled estimates. They are built from BRFSS survey responses with census and American Community Survey inputs (county poverty, the age, sex, race and ethnicity of each tract's residents, and education), with state and county random effects, and the Area Deprivation Index is built from American Community Survey data too, so part of each gradient here is mechanical. Within a county, tract values differ through the demographic composition of tracts. The tract file has no intervals; the bands are 95% intervals for the weighted mean from a bootstrap over counties (replicates and seed in the methods note).

Note: PLACES 2024 release: BRFSS 2022 for most measures and BRFSS 2021 for blood pressure, blood pressure medicine, high cholesterol and cholesterol screening. Values are crude prevalence among adults, not age-adjusted, so differences in age structure move them.

Note: The zero line is the weighted mean of the least deprived tenth of tracts. Tenths hold 8,118 to 8,511 tracts, because tied ranks fall in the same tenth, and twentieths hold fewer, so the first points sit slightly below the line. The food insecurity panel covers fewer tracts, because that question was asked in only some states, so its first points reflect which states fall in each twentieth.

Note: The twelve measures are this book's selection from the 40 that PLACES publishes for tracts: conditions and behaviors that the first edition followed, insurance and food insecurity, and the checkup. Tract means weight each tract by all its adults, including for the measures that cover only some ages, such as insurance and tooth loss, because the file carries no population for them.

Note: Panel scales are shared, so the prevalence in each twentieth is in the table under Show data.

Source: CDC PLACES 2024 release (BRFSS 2022) and Area Deprivation Index 2023; modeled estimates, population-weighted (adults in each tract)

Data: https://vishalsingh.org/zipcode/report/health-behaviors-and-the-supply-of-care/figures/twelve_outcome_gradients.json

### Table 2. Binge drinking is higher in higher-income tracts in 50 of 51 states, while the other measures are lower in higher-income tracts in all 51 states

Points of prevalence per doubling of tract median household income, one dot per state and measure, adults, with controls for the age, Black and Hispanic shares of residents, 2022

| Behavior | State | Points per doubling | Lower limit of the 95% interval | Upper limit of the 95% interval | Tracts |
| --- | --- | ---: | ---: | ---: | ---: |
| Current smoking | Alabama | −5.29 | −6.54 | −4.05 | 1,423 |
| Current smoking | Alaska | −8.13 | −10.80 | −5.45 | 173 |
| Current smoking | Arizona | −5.12 | −7.05 | −3.18 | 1,725 |
| Current smoking | Arkansas | −6.01 | −7.17 | −4.85 | 818 |
| Current smoking | California | −3.66 | −4.10 | −3.23 | 9,003 |
| Current smoking | Colorado | −2.88 | −3.61 | −2.15 | 1,423 |
| Current smoking | Connecticut | −3.76 | −4.87 | −2.65 | 870 |
| Current smoking | Delaware | −3.63 | −5.14 | −2.13 | 256 |
| Current smoking | District of Columbia | −2.38 | −3.45 | −1.32 | 201 |
| Current smoking | Florida | −4.95 | −5.54 | −4.36 | 5,053 |
| Current smoking | Georgia | −5.70 | −6.27 | −5.14 | 2,766 |
| Current smoking | Hawaii | −3.96 | −4.52 | −3.40 | 419 |
| Current smoking | Idaho | −3.30 | −4.00 | −2.59 | 453 |
| Current smoking | Illinois | −4.67 | −5.93 | −3.42 | 3,239 |
| Current smoking | Indiana | −5.07 | −6.55 | −3.60 | 1,678 |
| Current smoking | Iowa | −4.47 | −5.64 | −3.31 | 890 |
| Current smoking | Kansas | −5.16 | −6.46 | −3.87 | 811 |
| Current smoking | Kentucky | −7.14 | −7.89 | −6.38 | 1,287 |
| Current smoking | Louisiana | −6.01 | −6.60 | −5.43 | 1,352 |
| Current smoking | Maine | −5.80 | −6.68 | −4.91 | 399 |
| Current smoking | Maryland | −4.84 | −5.75 | −3.92 | 1,449 |
| Current smoking | Massachusetts | −3.95 | −5.70 | −2.20 | 1,585 |
| Current smoking | Michigan | −5.85 | −6.69 | −5.01 | 2,869 |
| Current smoking | Minnesota | −4.56 | −5.34 | −3.78 | 1,497 |
| Current smoking | Mississippi | −5.77 | −7.25 | −4.29 | 865 |
| Current smoking | Missouri | −6.34 | −7.36 | −5.31 | 1,644 |
| Current smoking | Montana | −6.65 | −8.29 | −5.00 | 315 |
| Current smoking | Nebraska | −3.80 | −4.23 | −3.37 | 550 |
| Current smoking | Nevada | −4.54 | −5.64 | −3.44 | 767 |
| Current smoking | New Hampshire | −4.38 | −5.14 | −3.62 | 347 |
| Current smoking | New Jersey | −4.31 | −4.87 | −3.75 | 2,154 |
| Current smoking | New Mexico | −6.93 | −8.55 | −5.31 | 594 |
| Current smoking | New York | −4.70 | −5.23 | −4.17 | 5,182 |
| Current smoking | North Carolina | −5.40 | −6.02 | −4.78 | 2,628 |
| Current smoking | North Dakota | −4.04 | −6.23 | −1.86 | 227 |
| Current smoking | Ohio | −6.77 | −7.24 | −6.30 | 3,129 |
| Current smoking | Oklahoma | −6.30 | −7.23 | −5.36 | 1,189 |
| Current smoking | Oregon | −4.57 | −5.48 | −3.66 | 991 |
| Current smoking | Pennsylvania | −5.51 | −5.98 | −5.05 | 3,377 |
| Current smoking | Rhode Island | −4.53 | −5.11 | −3.94 | 246 |
| Current smoking | South Carolina | −5.61 | −6.44 | −4.78 | 1,301 |
| Current smoking | South Dakota | −7.73 | −10.72 | −4.74 | 242 |
| Current smoking | Tennessee | −6.64 | −7.92 | −5.36 | 1,675 |
| Current smoking | Texas | −4.35 | −4.98 | −3.73 | 6,802 |
| Current smoking | Utah | −1.76 | −2.66 | −0.86 | 707 |
| Current smoking | Vermont | −3.95 | −5.08 | −2.83 | 192 |
| Current smoking | Virginia | −4.66 | −5.25 | −4.07 | 2,152 |
| Current smoking | Washington | −3.93 | −4.54 | −3.32 | 1,767 |
| Current smoking | West Virginia | −5.86 | −7.73 | −3.99 | 543 |
| Current smoking | Wisconsin | −4.04 | −4.66 | −3.42 | 1,523 |
| Current smoking | Wyoming | −2.44 | −5.45 | 0.56 | 159 |
| Obesity | Alabama | −4.19 | −5.38 | −3.00 | 1,423 |
| Obesity | Alaska | −5.44 | −6.79 | −4.08 | 173 |
| Obesity | Arizona | −2.94 | −4.29 | −1.60 | 1,725 |
| Obesity | Arkansas | −4.27 | −5.48 | −3.06 | 818 |
| Obesity | California | −4.02 | −5.36 | −2.69 | 9,003 |
| Obesity | Colorado | −1.90 | −3.21 | −0.58 | 1,423 |
| Obesity | Connecticut | −5.18 | −7.75 | −2.61 | 870 |
| Obesity | Delaware | −3.57 | −4.88 | −2.26 | 256 |
| Obesity | District of Columbia | −1.46 | −2.58 | −0.33 | 201 |
| Obesity | Florida | −5.09 | −5.94 | −4.24 | 5,053 |
| Obesity | Georgia | −6.05 | −7.26 | −4.83 | 2,766 |
| Obesity | Hawaii | −3.12 | −3.40 | −2.84 | 419 |
| Obesity | Idaho | −3.29 | −4.95 | −1.62 | 453 |
| Obesity | Illinois | −4.31 | −5.85 | −2.77 | 3,239 |
| Obesity | Indiana | −3.79 | −5.61 | −1.96 | 1,678 |
| Obesity | Iowa | −3.37 | −4.61 | −2.12 | 890 |
| Obesity | Kansas | −4.54 | −5.89 | −3.19 | 811 |
| Obesity | Kentucky | −5.33 | −6.31 | −4.34 | 1,287 |
| Obesity | Louisiana | −5.19 | −6.09 | −4.30 | 1,352 |
| Obesity | Maine | −6.92 | −8.25 | −5.60 | 399 |
| Obesity | Maryland | −5.11 | −7.78 | −2.45 | 1,449 |
| Obesity | Massachusetts | −3.57 | −5.94 | −1.20 | 1,585 |
| Obesity | Michigan | −5.40 | −6.53 | −4.27 | 2,869 |
| Obesity | Minnesota | −5.36 | −6.53 | −4.18 | 1,497 |
| Obesity | Mississippi | −4.27 | −5.67 | −2.86 | 865 |
| Obesity | Missouri | −4.95 | −6.98 | −2.93 | 1,644 |
| Obesity | Montana | −5.67 | −8.49 | −2.85 | 315 |
| Obesity | Nebraska | −3.15 | −4.74 | −1.56 | 550 |
| Obesity | Nevada | −3.67 | −4.87 | −2.48 | 767 |
| Obesity | New Hampshire | −2.80 | −4.04 | −1.56 | 347 |
| Obesity | New Jersey | −5.19 | −6.17 | −4.22 | 2,154 |
| Obesity | New Mexico | −5.71 | −6.82 | −4.61 | 594 |
| Obesity | New York | −5.19 | −6.80 | −3.59 | 5,182 |
| Obesity | North Carolina | −4.68 | −5.96 | −3.40 | 2,628 |
| Obesity | North Dakota | −2.65 | −4.32 | −0.98 | 227 |
| Obesity | Ohio | −5.73 | −6.28 | −5.18 | 3,129 |
| Obesity | Oklahoma | −4.94 | −6.09 | −3.79 | 1,189 |
| Obesity | Oregon | −3.21 | −4.36 | −2.07 | 991 |
| Obesity | Pennsylvania | −4.76 | −5.47 | −4.05 | 3,377 |
| Obesity | Rhode Island | −3.64 | −5.39 | −1.90 | 246 |
| Obesity | South Carolina | −5.64 | −6.83 | −4.46 | 1,301 |
| Obesity | South Dakota | −4.30 | −5.52 | −3.08 | 242 |
| Obesity | Tennessee | −5.61 | −7.06 | −4.16 | 1,675 |
| Obesity | Texas | −3.87 | −4.48 | −3.25 | 6,802 |
| Obesity | Utah | −1.56 | −2.61 | −0.51 | 707 |
| Obesity | Vermont | −3.52 | −4.60 | −2.43 | 192 |
| Obesity | Virginia | −4.49 | −5.82 | −3.17 | 2,152 |
| Obesity | Washington | −5.22 | −7.57 | −2.88 | 1,767 |
| Obesity | West Virginia | −3.40 | −4.67 | −2.14 | 543 |
| Obesity | Wisconsin | −3.44 | −4.32 | −2.57 | 1,523 |
| Obesity | Wyoming | −3.13 | −7.58 | 1.33 | 159 |
| Diagnosed diabetes | Alabama | −2.33 | −3.32 | −1.34 | 1,423 |
| Diagnosed diabetes | Alaska | −2.87 | −3.89 | −1.84 | 173 |
| Diagnosed diabetes | Arizona | −2.22 | −3.76 | −0.69 | 1,725 |
| Diagnosed diabetes | Arkansas | −2.85 | −3.87 | −1.84 | 818 |
| Diagnosed diabetes | California | −1.35 | −1.71 | −0.98 | 9,003 |
| Diagnosed diabetes | Colorado | −0.76 | −1.28 | −0.25 | 1,423 |
| Diagnosed diabetes | Connecticut | −1.76 | −2.78 | −0.74 | 870 |
| Diagnosed diabetes | Delaware | −0.86 | −2.34 | 0.62 | 256 |
| Diagnosed diabetes | District of Columbia | −0.39 | −0.92 | 0.13 | 201 |
| Diagnosed diabetes | Florida | −2.46 | −3.01 | −1.90 | 5,053 |
| Diagnosed diabetes | Georgia | −2.57 | −3.04 | −2.09 | 2,766 |
| Diagnosed diabetes | Hawaii | −1.82 | −2.94 | −0.70 | 419 |
| Diagnosed diabetes | Idaho | −1.23 | −2.01 | −0.46 | 453 |
| Diagnosed diabetes | Illinois | −1.71 | −2.59 | −0.83 | 3,239 |
| Diagnosed diabetes | Indiana | −1.57 | −2.43 | −0.72 | 1,678 |
| Diagnosed diabetes | Iowa | −1.28 | −2.25 | −0.31 | 890 |
| Diagnosed diabetes | Kansas | −1.75 | −2.43 | −1.07 | 811 |
| Diagnosed diabetes | Kentucky | −3.07 | −3.50 | −2.63 | 1,287 |
| Diagnosed diabetes | Louisiana | −3.08 | −3.65 | −2.51 | 1,352 |
| Diagnosed diabetes | Maine | −2.38 | −2.91 | −1.86 | 399 |
| Diagnosed diabetes | Maryland | −1.72 | −2.41 | −1.03 | 1,449 |
| Diagnosed diabetes | Massachusetts | −1.43 | −2.19 | −0.66 | 1,585 |
| Diagnosed diabetes | Michigan | −2.11 | −2.63 | −1.60 | 2,869 |
| Diagnosed diabetes | Minnesota | −1.59 | −1.96 | −1.22 | 1,497 |
| Diagnosed diabetes | Mississippi | −3.08 | −4.13 | −2.04 | 865 |
| Diagnosed diabetes | Missouri | −2.30 | −2.85 | −1.76 | 1,644 |
| Diagnosed diabetes | Montana | −2.87 | −3.85 | −1.90 | 315 |
| Diagnosed diabetes | Nebraska | −0.99 | −1.47 | −0.52 | 550 |
| Diagnosed diabetes | Nevada | −2.04 | −2.93 | −1.14 | 767 |
| Diagnosed diabetes | New Hampshire | −1.35 | −1.94 | −0.75 | 347 |
| Diagnosed diabetes | New Jersey | −1.46 | −1.94 | −0.99 | 2,154 |
| Diagnosed diabetes | New Mexico | −3.61 | −4.53 | −2.69 | 594 |
| Diagnosed diabetes | New York | −2.24 | −2.87 | −1.61 | 5,182 |
| Diagnosed diabetes | North Carolina | −2.41 | −2.95 | −1.87 | 2,628 |
| Diagnosed diabetes | North Dakota | −1.70 | −3.17 | −0.23 | 227 |
| Diagnosed diabetes | Ohio | −2.58 | −2.99 | −2.18 | 3,129 |
| Diagnosed diabetes | Oklahoma | −2.95 | −3.68 | −2.23 | 1,189 |
| Diagnosed diabetes | Oregon | −1.38 | −2.13 | −0.62 | 991 |
| Diagnosed diabetes | Pennsylvania | −1.91 | −2.22 | −1.59 | 3,377 |
| Diagnosed diabetes | Rhode Island | −2.01 | −2.55 | −1.47 | 246 |
| Diagnosed diabetes | South Carolina | −2.82 | −3.47 | −2.18 | 1,301 |
| Diagnosed diabetes | South Dakota | −3.14 | −4.61 | −1.67 | 242 |
| Diagnosed diabetes | Tennessee | −2.84 | −3.76 | −1.91 | 1,675 |
| Diagnosed diabetes | Texas | −1.86 | −2.42 | −1.29 | 6,802 |
| Diagnosed diabetes | Utah | −0.38 | −1.20 | 0.45 | 707 |
| Diagnosed diabetes | Vermont | −1.09 | −1.61 | −0.57 | 192 |
| Diagnosed diabetes | Virginia | −1.49 | −1.94 | −1.05 | 2,152 |
| Diagnosed diabetes | Washington | −1.41 | −1.93 | −0.89 | 1,767 |
| Diagnosed diabetes | West Virginia | −2.40 | −3.84 | −0.95 | 543 |
| Diagnosed diabetes | Wisconsin | −1.03 | −1.50 | −0.55 | 1,523 |
| Diagnosed diabetes | Wyoming | −1.02 | −2.99 | 0.94 | 159 |
| Frequent mental distress | Alabama | −4.47 | −4.82 | −4.12 | 1,423 |
| Frequent mental distress | Alaska | −4.04 | −5.03 | −3.04 | 173 |
| Frequent mental distress | Arizona | −4.21 | −5.30 | −3.11 | 1,725 |
| Frequent mental distress | Arkansas | −4.24 | −4.78 | −3.70 | 818 |
| Frequent mental distress | California | −3.34 | −3.76 | −2.93 | 9,003 |
| Frequent mental distress | Colorado | −3.44 | −4.15 | −2.73 | 1,423 |
| Frequent mental distress | Connecticut | −3.41 | −3.88 | −2.93 | 870 |
| Frequent mental distress | Delaware | −3.95 | −5.59 | −2.31 | 256 |
| Frequent mental distress | District of Columbia | −3.02 | −4.58 | −1.45 | 201 |
| Frequent mental distress | Florida | −3.51 | −3.82 | −3.20 | 5,053 |
| Frequent mental distress | Georgia | −4.03 | −4.25 | −3.80 | 2,766 |
| Frequent mental distress | Hawaii | −2.41 | −2.82 | −2.00 | 419 |
| Frequent mental distress | Idaho | −4.19 | −5.20 | −3.17 | 453 |
| Frequent mental distress | Illinois | −3.42 | −3.79 | −3.05 | 3,239 |
| Frequent mental distress | Indiana | −4.33 | −4.68 | −3.98 | 1,678 |
| Frequent mental distress | Iowa | −4.31 | −5.00 | −3.62 | 890 |
| Frequent mental distress | Kansas | −3.96 | −4.35 | −3.57 | 811 |
| Frequent mental distress | Kentucky | −4.11 | −4.41 | −3.80 | 1,287 |
| Frequent mental distress | Louisiana | −3.71 | −4.17 | −3.25 | 1,352 |
| Frequent mental distress | Maine | −3.60 | −4.13 | −3.07 | 399 |
| Frequent mental distress | Maryland | −3.93 | −4.17 | −3.69 | 1,449 |
| Frequent mental distress | Massachusetts | −3.54 | −3.83 | −3.26 | 1,585 |
| Frequent mental distress | Michigan | −4.30 | −4.59 | −4.00 | 2,869 |
| Frequent mental distress | Minnesota | −3.95 | −4.32 | −3.58 | 1,497 |
| Frequent mental distress | Mississippi | −3.82 | −4.30 | −3.34 | 865 |
| Frequent mental distress | Missouri | −4.22 | −4.65 | −3.79 | 1,644 |
| Frequent mental distress | Montana | −4.06 | −4.83 | −3.29 | 315 |
| Frequent mental distress | Nebraska | −3.52 | −3.99 | −3.06 | 550 |
| Frequent mental distress | Nevada | −2.82 | −3.21 | −2.42 | 767 |
| Frequent mental distress | New Hampshire | −3.46 | −4.09 | −2.83 | 347 |
| Frequent mental distress | New Jersey | −3.18 | −3.63 | −2.73 | 2,154 |
| Frequent mental distress | New Mexico | −4.02 | −4.70 | −3.34 | 594 |
| Frequent mental distress | New York | −2.93 | −3.35 | −2.50 | 5,182 |
| Frequent mental distress | North Carolina | −3.89 | −4.14 | −3.64 | 2,628 |
| Frequent mental distress | North Dakota | −3.91 | −4.69 | −3.13 | 227 |
| Frequent mental distress | Ohio | −4.71 | −4.99 | −4.43 | 3,129 |
| Frequent mental distress | Oklahoma | −4.33 | −4.60 | −4.06 | 1,189 |
| Frequent mental distress | Oregon | −4.19 | −4.87 | −3.50 | 991 |
| Frequent mental distress | Pennsylvania | −3.60 | −3.82 | −3.38 | 3,377 |
| Frequent mental distress | Rhode Island | −2.93 | −3.65 | −2.22 | 246 |
| Frequent mental distress | South Carolina | −3.80 | −4.32 | −3.28 | 1,301 |
| Frequent mental distress | South Dakota | −4.69 | −5.57 | −3.81 | 242 |
| Frequent mental distress | Tennessee | −4.32 | −4.50 | −4.14 | 1,675 |
| Frequent mental distress | Texas | −3.83 | −4.11 | −3.55 | 6,802 |
| Frequent mental distress | Utah | −3.80 | −5.22 | −2.38 | 707 |
| Frequent mental distress | Vermont | −4.37 | −5.55 | −3.19 | 192 |
| Frequent mental distress | Virginia | −3.86 | −4.08 | −3.63 | 2,152 |
| Frequent mental distress | Washington | −4.30 | −4.64 | −3.96 | 1,767 |
| Frequent mental distress | West Virginia | −4.42 | −5.08 | −3.77 | 543 |
| Frequent mental distress | Wisconsin | −4.26 | −4.76 | −3.76 | 1,523 |
| Frequent mental distress | Wyoming | −3.49 | −4.43 | −2.55 | 159 |
| No leisure-time physical activity | Alabama | −8.26 | −9.49 | −7.02 | 1,423 |
| No leisure-time physical activity | Alaska | −7.43 | −9.28 | −5.57 | 173 |
| No leisure-time physical activity | Arizona | −6.01 | −7.81 | −4.22 | 1,725 |
| No leisure-time physical activity | Arkansas | −9.36 | −10.76 | −7.97 | 818 |
| No leisure-time physical activity | California | −5.16 | −5.83 | −4.48 | 9,003 |
| No leisure-time physical activity | Colorado | −3.81 | −4.54 | −3.07 | 1,423 |
| No leisure-time physical activity | Connecticut | −5.60 | −6.98 | −4.22 | 870 |
| No leisure-time physical activity | Delaware | −6.14 | −7.34 | −4.95 | 256 |
| No leisure-time physical activity | District of Columbia | −3.05 | −3.82 | −2.27 | 201 |
| No leisure-time physical activity | Florida | −7.40 | −8.10 | −6.70 | 5,053 |
| No leisure-time physical activity | Georgia | −8.61 | −9.48 | −7.73 | 2,766 |
| No leisure-time physical activity | Hawaii | −4.83 | −7.09 | −2.57 | 419 |
| No leisure-time physical activity | Idaho | −6.49 | −7.48 | −5.50 | 453 |
| No leisure-time physical activity | Illinois | −6.54 | −7.55 | −5.53 | 3,239 |
| No leisure-time physical activity | Indiana | −7.30 | −8.70 | −5.91 | 1,678 |
| No leisure-time physical activity | Iowa | −7.56 | −8.65 | −6.46 | 890 |
| No leisure-time physical activity | Kansas | −6.53 | −7.42 | −5.63 | 811 |
| No leisure-time physical activity | Kentucky | −9.02 | −9.63 | −8.41 | 1,287 |
| No leisure-time physical activity | Louisiana | −8.32 | −9.14 | −7.50 | 1,352 |
| No leisure-time physical activity | Maine | −8.03 | −9.35 | −6.70 | 399 |
| No leisure-time physical activity | Maryland | −6.69 | −7.77 | −5.62 | 1,449 |
| No leisure-time physical activity | Massachusetts | −5.93 | −7.56 | −4.29 | 1,585 |
| No leisure-time physical activity | Michigan | −7.22 | −8.06 | −6.37 | 2,869 |
| No leisure-time physical activity | Minnesota | −6.25 | −6.85 | −5.65 | 1,497 |
| No leisure-time physical activity | Mississippi | −8.38 | −10.04 | −6.72 | 865 |
| No leisure-time physical activity | Missouri | −8.13 | −8.97 | −7.29 | 1,644 |
| No leisure-time physical activity | Montana | −8.17 | −9.75 | −6.58 | 315 |
| No leisure-time physical activity | Nebraska | −5.56 | −6.21 | −4.91 | 550 |
| No leisure-time physical activity | Nevada | −5.65 | −7.08 | −4.22 | 767 |
| No leisure-time physical activity | New Hampshire | −5.29 | −6.12 | −4.46 | 347 |
| No leisure-time physical activity | New Jersey | −5.76 | −6.59 | −4.92 | 2,154 |
| No leisure-time physical activity | New Mexico | −7.85 | −9.10 | −6.59 | 594 |
| No leisure-time physical activity | New York | −6.95 | −8.08 | −5.82 | 5,182 |
| No leisure-time physical activity | North Carolina | −7.53 | −8.25 | −6.82 | 2,628 |
| No leisure-time physical activity | North Dakota | −6.37 | −8.10 | −4.63 | 227 |
| No leisure-time physical activity | Ohio | −8.44 | −8.88 | −8.00 | 3,129 |
| No leisure-time physical activity | Oklahoma | −9.23 | −10.46 | −7.99 | 1,189 |
| No leisure-time physical activity | Oregon | −5.06 | −5.98 | −4.14 | 991 |
| No leisure-time physical activity | Pennsylvania | −6.93 | −7.46 | −6.40 | 3,377 |
| No leisure-time physical activity | Rhode Island | −5.88 | −6.26 | −5.50 | 246 |
| No leisure-time physical activity | South Carolina | −8.46 | −9.27 | −7.64 | 1,301 |
| No leisure-time physical activity | South Dakota | −7.28 | −9.25 | −5.30 | 242 |
| No leisure-time physical activity | Tennessee | −8.52 | −9.83 | −7.21 | 1,675 |
| No leisure-time physical activity | Texas | −6.93 | −7.78 | −6.07 | 6,802 |
| No leisure-time physical activity | Utah | −3.44 | −4.31 | −2.56 | 707 |
| No leisure-time physical activity | Vermont | −5.43 | −6.39 | −4.47 | 192 |
| No leisure-time physical activity | Virginia | −5.79 | −6.42 | −5.16 | 2,152 |
| No leisure-time physical activity | Washington | −5.31 | −6.22 | −4.40 | 1,767 |
| No leisure-time physical activity | West Virginia | −8.03 | −9.68 | −6.37 | 543 |
| No leisure-time physical activity | Wisconsin | −6.16 | −6.64 | −5.68 | 1,523 |
| No leisure-time physical activity | Wyoming | −4.84 | −8.10 | −1.57 | 159 |
| Binge drinking | Alabama | 1.26 | 0.97 | 1.54 | 1,423 |
| Binge drinking | Alaska | 1.02 | 0.43 | 1.61 | 173 |
| Binge drinking | Arizona | 0.91 | 0.37 | 1.44 | 1,725 |
| Binge drinking | Arkansas | 1.97 | 1.23 | 2.71 | 818 |
| Binge drinking | California | 0.66 | −0.07 | 1.40 | 9,003 |
| Binge drinking | Colorado | 0.65 | −0.18 | 1.47 | 1,423 |
| Binge drinking | Connecticut | 0.15 | −0.40 | 0.70 | 870 |
| Binge drinking | Delaware | 0.95 | 0.63 | 1.28 | 256 |
| Binge drinking | District of Columbia | 0.92 | 0.13 | 1.70 | 201 |
| Binge drinking | Florida | 1.18 | 0.86 | 1.49 | 5,053 |
| Binge drinking | Georgia | 1.35 | 0.96 | 1.75 | 2,766 |
| Binge drinking | Hawaii | 0.97 | 0.41 | 1.53 | 419 |
| Binge drinking | Idaho | 1.16 | 0.71 | 1.61 | 453 |
| Binge drinking | Illinois | 1.54 | 0.70 | 2.38 | 3,239 |
| Binge drinking | Indiana | 1.20 | 0.61 | 1.79 | 1,678 |
| Binge drinking | Iowa | 1.49 | 1.23 | 1.76 | 890 |
| Binge drinking | Kansas | 0.65 | 0.22 | 1.07 | 811 |
| Binge drinking | Kentucky | 1.99 | 1.73 | 2.24 | 1,287 |
| Binge drinking | Louisiana | 2.33 | 1.81 | 2.84 | 1,352 |
| Binge drinking | Maine | 1.74 | 0.89 | 2.59 | 399 |
| Binge drinking | Maryland | −0.77 | −1.71 | 0.17 | 1,449 |
| Binge drinking | Massachusetts | 0.51 | −0.26 | 1.27 | 1,585 |
| Binge drinking | Michigan | 1.12 | 0.41 | 1.84 | 2,869 |
| Binge drinking | Minnesota | 0.63 | −0.05 | 1.32 | 1,497 |
| Binge drinking | Mississippi | 1.41 | 0.90 | 1.92 | 865 |
| Binge drinking | Missouri | 1.88 | 1.42 | 2.34 | 1,644 |
| Binge drinking | Montana | 2.18 | 1.44 | 2.93 | 315 |
| Binge drinking | Nebraska | 0.91 | 0.58 | 1.23 | 550 |
| Binge drinking | Nevada | 0.89 | 0.19 | 1.59 | 767 |
| Binge drinking | New Hampshire | 1.12 | 0.05 | 2.20 | 347 |
| Binge drinking | New Jersey | 0.36 | −0.19 | 0.90 | 2,154 |
| Binge drinking | New Mexico | 1.59 | 1.17 | 2.02 | 594 |
| Binge drinking | New York | 1.37 | 0.26 | 2.48 | 5,182 |
| Binge drinking | North Carolina | 1.57 | 0.96 | 2.18 | 2,628 |
| Binge drinking | North Dakota | 1.94 | 1.42 | 2.46 | 227 |
| Binge drinking | Ohio | 1.67 | 1.25 | 2.09 | 3,129 |
| Binge drinking | Oklahoma | 1.68 | 0.95 | 2.41 | 1,189 |
| Binge drinking | Oregon | 1.01 | 0.59 | 1.43 | 991 |
| Binge drinking | Pennsylvania | 1.16 | 0.68 | 1.63 | 3,377 |
| Binge drinking | Rhode Island | 1.13 | 0.64 | 1.62 | 246 |
| Binge drinking | South Carolina | 2.25 | 1.24 | 3.26 | 1,301 |
| Binge drinking | South Dakota | 1.32 | 0.46 | 2.18 | 242 |
| Binge drinking | Tennessee | 1.99 | 1.39 | 2.60 | 1,675 |
| Binge drinking | Texas | 0.89 | 0.49 | 1.28 | 6,802 |
| Binge drinking | Utah | 1.44 | 0.99 | 1.89 | 707 |
| Binge drinking | Vermont | 1.21 | 0.28 | 2.15 | 192 |
| Binge drinking | Virginia | 0.04 | −0.72 | 0.79 | 2,152 |
| Binge drinking | Washington | 0.37 | −0.17 | 0.92 | 1,767 |
| Binge drinking | West Virginia | 1.62 | 1.11 | 2.14 | 543 |
| Binge drinking | Wisconsin | 1.43 | 0.91 | 1.95 | 1,523 |
| Binge drinking | Wyoming | 0.82 | −0.66 | 2.30 | 159 |

Note: Each dot is a regression across the tracts of one state, and every state has at least 60 tracts with all the variables. The intervals use counties as clusters; Delaware and the District of Columbia use HC3 errors because they have too few counties for cluster-robust errors. The strip medians are printed on the chart.

Note: PLACES values are modeled estimates. They are built from BRFSS survey responses with census and American Community Survey inputs (county poverty, the age, sex, race and ethnicity of each tract's residents, and education), with state and county random effects, and the Area Deprivation Index is built from American Community Survey data too, so part of each gradient here is mechanical. Within a county, tract values differ through the demographic composition of tracts. Income comes from the same survey, and the controls remove only the age, Black and Hispanic shares. Race-alone shares include Hispanic residents, and median household income is top-coded in the survey.

Note: These are differences between tracts within a state. They do not describe what any person does.

Source: CDC PLACES 2024 release (BRFSS 2022) and ACS 2019-2023; modeled estimates, population-weighted (adults in each tract)

Data: https://vishalsingh.org/zipcode/report/health-behaviors-and-the-supply-of-care/figures/coefficient_zoo.json

### Table 3. From the least to the most deprived tenth of tracts, the dental visit rate falls from 70.3% to 45.8% while the annual checkup rate rises from 74.2% to 77.8%

Adults who visited a dentist or had a routine checkup in the past year, percent, by Area Deprivation Index national percentile of the tract, 2022

| conf_high | conf_low | estimate | measure | n_units | percentile |
| ---: | ---: | ---: | --- | ---: | ---: |
| 76.10 | 73 | 74.40 | Annual checkup | 693 | 1 |
| 75.42 | 71.84 | 73.47 | Annual checkup | 779 | 2 |
| 75.32 | 72.22 | 73.62 | Annual checkup | 587 | 3 |
| 75.30 | 72.99 | 74.01 | Annual checkup | 925 | 4 |
| 75.49 | 73.33 | 74.33 | Annual checkup | 636 | 5 |
| 75.67 | 73.40 | 74.46 | Annual checkup | 979 | 6 |
| 75.45 | 73.35 | 74.30 | Annual checkup | 607 | 7 |
| 75.31 | 73.27 | 74.18 | Annual checkup | 915 | 8 |
| 75.30 | 73.30 | 74.22 | Annual checkup | 677 | 9 |
| 75.64 | 73.39 | 74.42 | Annual checkup | 1,000 | 10 |
| 75.45 | 73.49 | 74.38 | Annual checkup | 598 | 11 |
| 75.94 | 73.39 | 74.58 | Annual checkup | 1,019 | 12 |
| 75.39 | 72.90 | 74.03 | Annual checkup | 693 | 13 |
| 75.74 | 73.59 | 74.63 | Annual checkup | 1,007 | 14 |
| 75.36 | 72.98 | 74.08 | Annual checkup | 656 | 15 |
| 75.32 | 73.07 | 74.13 | Annual checkup | 1,054 | 16 |
| 75.65 | 72.97 | 74.29 | Annual checkup | 684 | 17 |
| 75.20 | 73.14 | 74.12 | Annual checkup | 1,033 | 18 |
| 74.96 | 72.70 | 73.80 | Annual checkup | 702 | 19 |
| 75.70 | 73.47 | 74.58 | Annual checkup | 1,019 | 20 |
| 75.18 | 73.25 | 74.18 | Annual checkup | 655 | 21 |
| 75.39 | 73.71 | 74.55 | Annual checkup | 1,098 | 22 |
| 75.52 | 73.77 | 74.64 | Annual checkup | 652 | 23 |
| 75.17 | 73.45 | 74.31 | Annual checkup | 1,050 | 24 |
| 75.28 | 73.86 | 74.56 | Annual checkup | 679 | 25 |
| 75.39 | 74.06 | 74.72 | Annual checkup | 1,028 | 26 |
| 75.60 | 73.91 | 74.77 | Annual checkup | 678 | 27 |
| 75.54 | 74.17 | 74.86 | Annual checkup | 1,029 | 28 |
| 75.67 | 74.13 | 74.90 | Annual checkup | 659 | 29 |
| 75.64 | 74.31 | 74.98 | Annual checkup | 1,112 | 30 |
| 75.35 | 73.93 | 74.62 | Annual checkup | 668 | 31 |
| 75.39 | 74.19 | 74.79 | Annual checkup | 1,010 | 32 |
| 75.39 | 74.01 | 74.70 | Annual checkup | 689 | 33 |
| 75.41 | 74.28 | 74.84 | Annual checkup | 1,053 | 34 |
| 75.74 | 74.51 | 75.13 | Annual checkup | 672 | 35 |
| 75.51 | 74.30 | 74.88 | Annual checkup | 1,098 | 36 |
| 75.33 | 74.10 | 74.71 | Annual checkup | 730 | 37 |
| 75.58 | 74.42 | 75.01 | Annual checkup | 1,031 | 38 |
| 76.08 | 74.90 | 75.48 | Annual checkup | 620 | 39 |
| 75.63 | 74.54 | 75.08 | Annual checkup | 995 | 40 |
| 75.69 | 74.46 | 75.08 | Annual checkup | 709 | 41 |
| 75.70 | 74.73 | 75.22 | Annual checkup | 1,045 | 42 |
| 75.85 | 74.69 | 75.25 | Annual checkup | 690 | 43 |
| 75.92 | 74.82 | 75.38 | Annual checkup | 1,044 | 44 |
| 76.03 | 75.03 | 75.53 | Annual checkup | 689 | 45 |
| 76.22 | 75.30 | 75.74 | Annual checkup | 1,008 | 46 |
| 76.33 | 75.20 | 75.74 | Annual checkup | 655 | 47 |
| 75.99 | 75 | 75.50 | Annual checkup | 1,070 | 48 |
| 76.34 | 75.26 | 75.80 | Annual checkup | 652 | 49 |
| 75.98 | 75.02 | 75.48 | Annual checkup | 1,066 | 50 |
| 76.16 | 75.04 | 75.58 | Annual checkup | 667 | 51 |
| 76.44 | 75.51 | 75.98 | Annual checkup | 1,106 | 52 |
| 76.19 | 75.32 | 75.75 | Annual checkup | 721 | 53 |
| 76.28 | 75.26 | 75.75 | Annual checkup | 1,047 | 54 |
| 76.55 | 75.55 | 76.04 | Annual checkup | 670 | 55 |
| 76.34 | 75.42 | 75.87 | Annual checkup | 1,069 | 56 |
| 76.32 | 75.32 | 75.82 | Annual checkup | 664 | 57 |
| 76.63 | 75.65 | 76.15 | Annual checkup | 1,046 | 58 |
| 76.44 | 75.45 | 75.94 | Annual checkup | 756 | 59 |
| 76.47 | 75.54 | 75.98 | Annual checkup | 1,056 | 60 |
| 76.50 | 75.52 | 76 | Annual checkup | 672 | 61 |
| 76.44 | 75.45 | 75.96 | Annual checkup | 1,107 | 62 |
| 76.55 | 75.51 | 76.01 | Annual checkup | 679 | 63 |
| 76.69 | 75.70 | 76.21 | Annual checkup | 1,061 | 64 |
| 76.72 | 75.49 | 76.14 | Annual checkup | 699 | 65 |
| 76.91 | 76.02 | 76.48 | Annual checkup | 1,041 | 66 |
| 76.96 | 76 | 76.49 | Annual checkup | 696 | 67 |
| 77 | 76.17 | 76.60 | Annual checkup | 1,028 | 68 |
| 76.97 | 76.14 | 76.55 | Annual checkup | 679 | 69 |
| 76.83 | 76.08 | 76.46 | Annual checkup | 1,030 | 70 |
| 77.05 | 76.15 | 76.62 | Annual checkup | 711 | 71 |
| 77.10 | 76.28 | 76.70 | Annual checkup | 1,040 | 72 |
| 76.94 | 76.02 | 76.48 | Annual checkup | 678 | 73 |
| 76.97 | 76.17 | 76.59 | Annual checkup | 1,061 | 74 |
| 77.10 | 76.06 | 76.60 | Annual checkup | 721 | 75 |
| 77.14 | 76.28 | 76.72 | Annual checkup | 1,001 | 76 |
| 77.34 | 76.36 | 76.87 | Annual checkup | 666 | 77 |
| 77.25 | 76.42 | 76.83 | Annual checkup | 1,025 | 78 |
| 77.42 | 76.64 | 77.05 | Annual checkup | 698 | 79 |
| 77.40 | 76.62 | 77.01 | Annual checkup | 1,024 | 80 |
| 77.44 | 76.56 | 77.01 | Annual checkup | 663 | 81 |
| 77.47 | 76.73 | 77.10 | Annual checkup | 983 | 82 |
| 77.57 | 76.70 | 77.14 | Annual checkup | 657 | 83 |
| 77.59 | 76.84 | 77.23 | Annual checkup | 906 | 84 |
| 77.60 | 76.68 | 77.12 | Annual checkup | 629 | 85 |
| 77.55 | 76.65 | 77.12 | Annual checkup | 984 | 86 |
| 77.54 | 76.69 | 77.11 | Annual checkup | 619 | 87 |
| 77.85 | 77.17 | 77.51 | Annual checkup | 911 | 88 |
| 77.71 | 76.73 | 77.22 | Annual checkup | 648 | 89 |
| 77.84 | 76.96 | 77.40 | Annual checkup | 941 | 90 |
| 78.03 | 77.07 | 77.55 | Annual checkup | 578 | 91 |
| 78.03 | 77.10 | 77.57 | Annual checkup | 878 | 92 |
| 77.95 | 77.04 | 77.49 | Annual checkup | 589 | 93 |
| 77.98 | 76.88 | 77.43 | Annual checkup | 794 | 94 |
| 78.49 | 77.38 | 77.94 | Annual checkup | 532 | 95 |
| 78.56 | 77.55 | 78.05 | Annual checkup | 761 | 96 |
| 78.69 | 77.23 | 77.94 | Annual checkup | 536 | 97 |
| 78.79 | 77.53 | 78.16 | Annual checkup | 725 | 98 |
| 79.20 | 77.82 | 78.56 | Annual checkup | 549 | 99 |
| 79.66 | 77.53 | 78.70 | Annual checkup | 682 | 100 |
| 75.95 | 73.21 | 74.77 | Dental visit | 693 | 1 |
| 74.95 | 71.66 | 73.03 | Dental visit | 779 | 2 |
| 74.03 | 69.83 | 71.73 | Dental visit | 587 | 3 |
| 73.10 | 69.50 | 71.27 | Dental visit | 925 | 4 |
| 72.41 | 67.16 | 69.78 | Dental visit | 636 | 5 |
| 71.85 | 66.40 | 69.10 | Dental visit | 979 | 6 |
| 72.28 | 68.03 | 69.95 | Dental visit | 607 | 7 |
| 71.49 | 67.30 | 69.14 | Dental visit | 915 | 8 |
| 70.93 | 66.28 | 68.51 | Dental visit | 677 | 9 |
| 71.20 | 65.54 | 68.13 | Dental visit | 1,000 | 10 |
| 70.92 | 66.74 | 68.66 | Dental visit | 598 | 11 |
| 71.10 | 65.61 | 68.31 | Dental visit | 1,019 | 12 |
| 70.67 | 64.96 | 67.82 | Dental visit | 693 | 13 |
| 69.96 | 65.82 | 67.76 | Dental visit | 1,007 | 14 |
| 69.56 | 65.06 | 67.17 | Dental visit | 656 | 15 |
| 68.75 | 64.39 | 66.47 | Dental visit | 1,054 | 16 |
| 69.25 | 63.74 | 66.50 | Dental visit | 684 | 17 |
| 68.72 | 64.09 | 66.41 | Dental visit | 1,033 | 18 |
| 68.01 | 62.54 | 65.35 | Dental visit | 702 | 19 |
| 68.93 | 63.42 | 66.35 | Dental visit | 1,019 | 20 |
| 68.80 | 64.71 | 66.89 | Dental visit | 655 | 21 |
| 68.55 | 64.87 | 66.88 | Dental visit | 1,098 | 22 |
| 68.13 | 64.36 | 66.42 | Dental visit | 652 | 23 |
| 68.45 | 64.67 | 66.80 | Dental visit | 1,050 | 24 |
| 68.55 | 65.97 | 67.31 | Dental visit | 679 | 25 |
| 67.72 | 65.33 | 66.54 | Dental visit | 1,028 | 26 |
| 68.37 | 64.46 | 66.58 | Dental visit | 678 | 27 |
| 67.19 | 64.49 | 65.94 | Dental visit | 1,029 | 28 |
| 67.13 | 63.11 | 65.31 | Dental visit | 659 | 29 |
| 66.91 | 63.97 | 65.49 | Dental visit | 1,112 | 30 |
| 67.01 | 64.12 | 65.58 | Dental visit | 668 | 31 |
| 66.85 | 64.91 | 65.87 | Dental visit | 1,010 | 32 |
| 66.15 | 64.07 | 65.12 | Dental visit | 689 | 33 |
| 66.43 | 64.52 | 65.46 | Dental visit | 1,053 | 34 |
| 66.47 | 64.56 | 65.50 | Dental visit | 672 | 35 |
| 65.61 | 63.65 | 64.60 | Dental visit | 1,098 | 36 |
| 65.35 | 63.25 | 64.33 | Dental visit | 730 | 37 |
| 65.20 | 63.21 | 64.22 | Dental visit | 1,031 | 38 |
| 65.94 | 63.57 | 64.81 | Dental visit | 620 | 39 |
| 65.01 | 62.95 | 63.98 | Dental visit | 995 | 40 |
| 64.49 | 62.47 | 63.45 | Dental visit | 709 | 41 |
| 64.73 | 62.72 | 63.73 | Dental visit | 1,045 | 42 |
| 64.07 | 62.06 | 63.06 | Dental visit | 690 | 43 |
| 64.47 | 62.63 | 63.54 | Dental visit | 1,044 | 44 |
| 63.79 | 61.68 | 62.72 | Dental visit | 689 | 45 |
| 64.37 | 61.92 | 63.18 | Dental visit | 1,008 | 46 |
| 64.27 | 62.43 | 63.32 | Dental visit | 655 | 47 |
| 63.22 | 61.36 | 62.31 | Dental visit | 1,070 | 48 |
| 63.60 | 61.77 | 62.69 | Dental visit | 652 | 49 |
| 62.95 | 61.06 | 61.98 | Dental visit | 1,066 | 50 |
| 63.09 | 61.02 | 62.02 | Dental visit | 667 | 51 |
| 62.68 | 60.73 | 61.70 | Dental visit | 1,106 | 52 |
| 62.23 | 60.18 | 61.18 | Dental visit | 721 | 53 |
| 61.92 | 59.70 | 60.81 | Dental visit | 1,047 | 54 |
| 61.91 | 59.65 | 60.80 | Dental visit | 670 | 55 |
| 61.76 | 59.66 | 60.68 | Dental visit | 1,069 | 56 |
| 61.68 | 59.81 | 60.75 | Dental visit | 664 | 57 |
| 61.60 | 59.55 | 60.60 | Dental visit | 1,046 | 58 |
| 60.95 | 58.48 | 59.74 | Dental visit | 756 | 59 |
| 60.71 | 58.91 | 59.80 | Dental visit | 1,056 | 60 |
| 60.96 | 58.76 | 59.82 | Dental visit | 672 | 61 |
| 59.68 | 57.43 | 58.55 | Dental visit | 1,107 | 62 |
| 60.09 | 57.55 | 58.83 | Dental visit | 679 | 63 |
| 59.80 | 57.28 | 58.60 | Dental visit | 1,061 | 64 |
| 59.44 | 56.75 | 58.17 | Dental visit | 699 | 65 |
| 60.07 | 57.81 | 59.01 | Dental visit | 1,041 | 66 |
| 59.18 | 56.90 | 58.09 | Dental visit | 696 | 67 |
| 58.98 | 57.04 | 57.99 | Dental visit | 1,028 | 68 |
| 59.27 | 57.22 | 58.28 | Dental visit | 679 | 69 |
| 58.04 | 56.06 | 57.05 | Dental visit | 1,030 | 70 |
| 58.22 | 55.94 | 57.06 | Dental visit | 711 | 71 |
| 58.05 | 55.98 | 56.99 | Dental visit | 1,040 | 72 |
| 57.70 | 55.43 | 56.57 | Dental visit | 678 | 73 |
| 56.97 | 55.23 | 56.10 | Dental visit | 1,061 | 74 |
| 57.37 | 55.13 | 56.25 | Dental visit | 721 | 75 |
| 56.19 | 54.20 | 55.22 | Dental visit | 1,001 | 76 |
| 56.62 | 54.18 | 55.43 | Dental visit | 666 | 77 |
| 55.97 | 54.23 | 55.16 | Dental visit | 1,025 | 78 |
| 56.19 | 54.31 | 55.21 | Dental visit | 698 | 79 |
| 56.09 | 54.19 | 55.15 | Dental visit | 1,024 | 80 |
| 55.11 | 52.96 | 54.09 | Dental visit | 663 | 81 |
| 54.97 | 53.06 | 54.04 | Dental visit | 983 | 82 |
| 53.68 | 51.27 | 52.51 | Dental visit | 657 | 83 |
| 54 | 52.03 | 53.06 | Dental visit | 906 | 84 |
| 53.16 | 50.52 | 51.88 | Dental visit | 629 | 85 |
| 52.58 | 50.51 | 51.60 | Dental visit | 984 | 86 |
| 52.83 | 50.87 | 51.82 | Dental visit | 619 | 87 |
| 51.78 | 49.96 | 50.88 | Dental visit | 911 | 88 |
| 50.53 | 48.08 | 49.35 | Dental visit | 648 | 89 |
| 51.08 | 49.04 | 50.09 | Dental visit | 941 | 90 |
| 50.20 | 47.56 | 48.95 | Dental visit | 578 | 91 |
| 49.48 | 47.27 | 48.38 | Dental visit | 878 | 92 |
| 48.71 | 46.46 | 47.55 | Dental visit | 589 | 93 |
| 46.61 | 44.20 | 45.45 | Dental visit | 794 | 94 |
| 46.74 | 44.23 | 45.52 | Dental visit | 532 | 95 |
| 45.02 | 42.81 | 43.95 | Dental visit | 761 | 96 |
| 44.18 | 41.51 | 42.78 | Dental visit | 536 | 97 |
| 42.57 | 38.99 | 40.94 | Dental visit | 725 | 98 |
| 42.64 | 39.48 | 41.10 | Dental visit | 549 | 99 |
| 40.76 | 37.25 | 39.14 | Dental visit | 682 | 100 |

Note: PLACES values are modeled estimates. They are built from BRFSS survey responses with census and American Community Survey inputs (county poverty, the age, sex, race and ethnicity of each tract's residents, and education), with state and county random effects, and the Area Deprivation Index is built from American Community Survey data too, so part of each gradient here is mechanical. Within a county, tract values differ through the demographic composition of tracts. The tract file has no intervals; the bands are 95% intervals for the weighted mean from a bootstrap over counties (replicates and seed in the methods note).

Note: PLACES 2024 release: BRFSS 2022 for most measures and BRFSS 2021 for blood pressure, blood pressure medicine, high cholesterol and cholesterol screening. Values are crude prevalence among adults, not age-adjusted, so differences in age structure move them.

Source: CDC PLACES 2024 release (BRFSS 2022) and Area Deprivation Index 2023; modeled estimates, population-weighted (adults in each tract)

Data: https://vishalsingh.org/zipcode/report/health-behaviors-and-the-supply-of-care/figures/checkup_vs_dental.json

### Table 4. In the most deprived tenth of tracts, the dental visit and screening tests are less common by 6.2 to 24.4 points and barriers to care are more common by 9.7 and 11.1 points

Difference in the share of adults, most deprived tenth of tracts minus least deprived tenth, percentage points, eight measures of care, 2021 and 2022

| Measure | Group | Least deprived tenth (%) | Most deprived tenth (%) | Difference (points) | Lower limit of the 95% interval | Upper limit of the 95% interval | Tracts in the most deprived tenth |
| --- | --- | ---: | ---: | ---: | ---: | ---: | ---: |
| Annual checkup | Checkup and treatment | 74.2 | 77.8 | 3.6 | 2.4 | 4.6 | 8,213 |
| Blood pressure medicine, adults with high blood pressure | Checkup and treatment | 73.6 | 79.5 | 5.9 | 4.8 | 6.7 | 7,946 |
| Dental visit | Dental visit and screening | 70.3 | 45.8 | −24.4 | −26.5 | −22.5 | 8,213 |
| Mammography, women ages 50 to 74 | Dental visit and screening | 78.8 | 71.7 | −7.0 | −7.8 | −6.3 | 8,213 |
| Cholesterol screening | Dental visit and screening | 87.9 | 81.0 | −6.9 | −7.5 | −6.3 | 7,946 |
| Colorectal cancer screening, ages 45 to 75 | Dental visit and screening | 65.3 | 59.1 | −6.2 | −8.3 | −4.4 | 8,213 |
| No reliable transportation | Barriers | 6.1 | 15.9 | 9.7 | 9.0 | 10.6 | 6,348 |
| No health insurance, ages 18 to 64 | Barriers | 6.3 | 17.4 | 11.1 | 9.7 | 12.8 | 8,213 |

Note: Blood pressure medicine is the share of adults with high blood pressure who take medicine for it, and blood pressure medicine and cholesterol screening are BRFSS 2021 estimates; the other measures are BRFSS 2022. The transportation question was asked in only some states, so that row covers fewer tracts. The grouping into checkup and treatment, dental visit and screening, and barriers is this book's: PLACES files the checkup, blood pressure medicine, the dental visit and insurance together under prevention. Some measures cover only some ages (insurance, mammography, colorectal screening), and tract means weight each tract by all its adults.

Note: PLACES values are modeled estimates. They are built from BRFSS survey responses with census and American Community Survey inputs (county poverty, the age, sex, race and ethnicity of each tract's residents, and education), with state and county random effects, and the Area Deprivation Index is built from American Community Survey data too, so part of each gradient here is mechanical. Within a county, tract values differ through the demographic composition of tracts. The tract file has no intervals; the bands are 95% intervals for the weighted mean from a bootstrap over counties (replicates and seed in the methods note).

Source: CDC PLACES 2024 release (BRFSS 2021 and 2022) and Area Deprivation Index 2023; modeled estimates, population-weighted (adults in each tract)

Data: https://vishalsingh.org/zipcode/report/health-behaviors-and-the-supply-of-care/figures/care_ledger.json

### Table 5. The checkup slope is within 0.5 points of zero in all four specifications, and the dental slope is between −2.37 and −3.86 points

Slope of tract prevalence on the Area Deprivation Index national percentile, points of prevalence per ten percentiles, adults, four specifications, 2022

| conf_high | conf_low | estimate | measure | n | specification |
| ---: | ---: | ---: | --- | ---: | --- |
| 0.541 | 0.287 | 0.414 | Annual checkup | 82,779 | All tracts |
| −2.08 | −2.67 | −2.37 | Dental visit | 82,779 | All tracts |
| 0.054 | −0.080 | −0.013 | Annual checkup | 82,779 | Within states |
| −2.81 | −3.14 | −2.97 | Dental visit | 82,779 | Within states |
| −0.190 | −0.348 | −0.269 | Annual checkup | 82,779 | Within counties |
| −3.73 | −3.99 | −3.86 | Dental visit | 82,779 | Within counties |
| 0.006 | −0.100 | −0.047 | Annual checkup | 82,779 | Within states, with age and race controls |
| −2.24 | −2.51 | −2.38 | Dental visit | 82,779 | Within states, with age and race controls |

Note: Within states and within counties, the comparison is between tracts of one state or county, with a separate baseline for each. The last row also holds constant the tract shares of residents who are ages 65 and over, Black alone and Hispanic. The standard error treats the tracts of a county as one cluster.

Note: PLACES values are modeled estimates. They are built from BRFSS survey responses with census and American Community Survey inputs (county poverty, the age, sex, race and ethnicity of each tract's residents, and education), with state and county random effects, and the Area Deprivation Index is built from American Community Survey data too, so part of each gradient here is mechanical. Within a county, tract values differ through the demographic composition of tracts. The tract file has no intervals; the bands are 95% intervals for the weighted mean from a bootstrap over counties (replicates and seed in the methods note).

Source: CDC PLACES 2024 release (BRFSS 2022) and Area Deprivation Index 2023; modeled estimates, population-weighted (adults in each tract)

Data: https://vishalsingh.org/zipcode/report/health-behaviors-and-the-supply-of-care/figures/checkup_specifications.json

### Table 6. The dental visit rate falls with deprivation in all 51 states, and the checkup slope is within 0.5 points of zero in 42

Slope of tract prevalence on the Area Deprivation Index national percentile within each state, points of prevalence per ten percentiles, adults, 2022

| conf_high | conf_low | estimate | measure | n | state |
| ---: | ---: | ---: | --- | ---: | --- |
| 0.287 | −0.038 | 0.124 | Annual checkup | 1,421 | AL |
| −3.48 | −4.00 | −3.74 | Dental visit | 1,421 | AL |
| −0.261 | −0.862 | −0.561 | Annual checkup | 172 | AK |
| −3.07 | −4.31 | −3.69 | Dental visit | 172 | AK |
| −0.227 | −0.620 | −0.423 | Annual checkup | 1,723 | AZ |
| −3.07 | −3.86 | −3.46 | Dental visit | 1,723 | AZ |
| 0.808 | 0.004 | 0.406 | Annual checkup | 818 | AR |
| −3.02 | −4.22 | −3.62 | Dental visit | 818 | AR |
| −0.276 | −0.857 | −0.566 | Annual checkup | 8,982 | CA |
| −3.36 | −4.11 | −3.74 | Dental visit | 8,982 | CA |
| 0.182 | −0.677 | −0.248 | Annual checkup | 1,420 | CO |
| −2.96 | −3.81 | −3.38 | Dental visit | 1,420 | CO |
| −0.135 | −0.604 | −0.369 | Annual checkup | 872 | CT |
| −2.67 | −4.46 | −3.57 | Dental visit | 872 | CT |
| −0.197 | −0.605 | −0.401 | Annual checkup | 253 | DE |
| −3.36 | −4.10 | −3.73 | Dental visit | 253 | DE |
| 2.11 | 1.32 | 1.71 | Annual checkup | 201 | DC |
| −5.57 | −7.38 | −6.47 | Dental visit | 201 | DC |
| 0.188 | −0.244 | −0.028 | Annual checkup | 5,034 | FL |
| −2.59 | −3.24 | −2.92 | Dental visit | 5,034 | FL |
| 0.604 | 0.140 | 0.372 | Annual checkup | 2,767 | GA |
| −3.09 | −3.42 | −3.26 | Dental visit | 2,767 | GA |
| −0.061 | −0.477 | −0.269 | Annual checkup | 414 | HI |
| −2.46 | −3.31 | −2.88 | Dental visit | 414 | HI |
| −0.009 | −1.13 | −0.567 | Annual checkup | 453 | ID |
| −1.34 | −3.51 | −2.43 | Dental visit | 453 | ID |
| 0.631 | −0.232 | 0.199 | Annual checkup | 3,239 | IL |
| −2.34 | −3.09 | −2.71 | Dental visit | 3,239 | IL |
| −0.035 | −0.262 | −0.149 | Annual checkup | 1,671 | IN |
| −3.12 | −3.83 | −3.47 | Dental visit | 1,671 | IN |
| 0.241 | −0.148 | 0.047 | Annual checkup | 887 | IA |
| −2.41 | −3.12 | −2.77 | Dental visit | 887 | IA |
| −0.075 | −0.353 | −0.214 | Annual checkup | 811 | KS |
| −2.67 | −3.47 | −3.07 | Dental visit | 811 | KS |
| −0.028 | −0.136 | −0.082 | Annual checkup | 1,285 | KY |
| −3.35 | −3.87 | −3.61 | Dental visit | 1,285 | KY |
| 0.488 | 0.144 | 0.316 | Annual checkup | 1,352 | LA |
| −3.52 | −4.38 | −3.95 | Dental visit | 1,352 | LA |
| 0.492 | −0.013 | 0.239 | Annual checkup | 399 | ME |
| −2.43 | −2.95 | −2.69 | Dental visit | 399 | ME |
| 0.577 | 0.043 | 0.310 | Annual checkup | 1,451 | MD |
| −3.08 | −3.73 | −3.41 | Dental visit | 1,451 | MD |
| 0.696 | −0.341 | 0.177 | Annual checkup | 1,576 | MA |
| −2.09 | −3.89 | −2.99 | Dental visit | 1,576 | MA |
| 0.273 | −0.206 | 0.034 | Annual checkup | 2,872 | MI |
| −2.62 | −3.82 | −3.22 | Dental visit | 2,872 | MI |
| 0.025 | −0.332 | −0.154 | Annual checkup | 1,492 | MN |
| −2.27 | −3.09 | −2.68 | Dental visit | 1,492 | MN |
| 0.600 | 0.173 | 0.387 | Annual checkup | 867 | MS |
| −1.43 | −1.88 | −1.66 | Dental visit | 867 | MS |
| 0.092 | −0.300 | −0.104 | Annual checkup | 1,641 | MO |
| −3.25 | −3.68 | −3.47 | Dental visit | 1,641 | MO |
| 0.947 | −0.482 | 0.232 | Annual checkup | 316 | MT |
| −1.91 | −2.95 | −2.43 | Dental visit | 316 | MT |
| −0.204 | −0.497 | −0.350 | Annual checkup | 547 | NE |
| −2.60 | −3.78 | −3.19 | Dental visit | 547 | NE |
| −0.251 | −0.963 | −0.607 | Annual checkup | 767 | NV |
| −3.17 | −4.51 | −3.84 | Dental visit | 767 | NV |
| −0.002 | −1.06 | −0.530 | Annual checkup | 346 | NH |
| −2.23 | −3.10 | −2.67 | Dental visit | 346 | NH |
| 0.441 | −0.353 | 0.044 | Annual checkup | 2,155 | NJ |
| −2.61 | −4.18 | −3.40 | Dental visit | 2,155 | NJ |
| −0.335 | −0.969 | −0.652 | Annual checkup | 595 | NM |
| −3.23 | −4.19 | −3.71 | Dental visit | 595 | NM |
| 0.446 | −0.068 | 0.189 | Annual checkup | 5,196 | NY |
| 0.164 | −1.22 | −0.526 | Dental visit | 5,196 | NY |
| 0.287 | 0.009 | 0.148 | Annual checkup | 2,625 | NC |
| −2.88 | −3.25 | −3.06 | Dental visit | 2,625 | NC |
| 0.909 | 0.246 | 0.577 | Annual checkup | 225 | ND |
| −1.50 | −2.73 | −2.11 | Dental visit | 225 | ND |
| 0.211 | −0.126 | 0.043 | Annual checkup | 3,131 | OH |
| −3.34 | −3.77 | −3.56 | Dental visit | 3,131 | OH |
| −0.009 | −0.331 | −0.170 | Annual checkup | 1,186 | OK |
| −3.65 | −4.21 | −3.93 | Dental visit | 1,186 | OK |
| 0.642 | −0.513 | 0.065 | Annual checkup | 990 | OR |
| −2.56 | −3.59 | −3.07 | Dental visit | 990 | OR |
| 0.236 | −0.079 | 0.079 | Annual checkup | 3,374 | PA |
| −2.40 | −3.33 | −2.86 | Dental visit | 3,374 | PA |
| −0.463 | −0.992 | −0.727 | Annual checkup | 244 | RI |
| −3.26 | −5.87 | −4.56 | Dental visit | 244 | RI |
| 0.431 | 0.043 | 0.237 | Annual checkup | 1,295 | SC |
| −2.89 | −3.56 | −3.22 | Dental visit | 1,295 | SC |
| 0.101 | −0.535 | −0.217 | Annual checkup | 242 | SD |
| −2.33 | −3.80 | −3.07 | Dental visit | 242 | SD |
| 0.378 | −0.036 | 0.171 | Annual checkup | 1,669 | TN |
| −2.51 | −3.76 | −3.14 | Dental visit | 1,669 | TN |
| 0.009 | −0.387 | −0.189 | Annual checkup | 6,789 | TX |
| −3.33 | −3.95 | −3.64 | Dental visit | 6,789 | TX |
| 0.010 | −0.821 | −0.405 | Annual checkup | 705 | UT |
| −2.99 | −3.90 | −3.45 | Dental visit | 705 | UT |
| 0.796 | −0.182 | 0.307 | Annual checkup | 192 | VT |
| −1.99 | −2.73 | −2.36 | Dental visit | 192 | VT |
| 0.495 | 0.226 | 0.361 | Annual checkup | 2,141 | VA |
| −2.27 | −2.67 | −2.47 | Dental visit | 2,141 | VA |
| −0.002 | −0.490 | −0.246 | Annual checkup | 1,757 | WA |
| −2.96 | −3.68 | −3.32 | Dental visit | 1,757 | WA |
| 0.392 | −0.083 | 0.154 | Annual checkup | 543 | WV |
| −2.31 | −3.79 | −3.05 | Dental visit | 543 | WV |
| 0.676 | −0.430 | 0.123 | Annual checkup | 1,520 | WI |
| −2.36 | −4.15 | −3.25 | Dental visit | 1,520 | WI |
| −0.048 | −0.920 | −0.484 | Annual checkup | 158 | WY |
| −1.72 | −2.84 | −2.28 | Dental visit | 158 | WY |

Note: States are shown by their postal codes. Delaware, Hawaii and the District of Columbia have too few counties for cluster-robust errors, so their intervals use HC3 errors, and the District of Columbia is a single county. Every state has at least 158 tracts, above the floor of 60, so none is dropped.

Note: PLACES values are modeled estimates. They are built from BRFSS survey responses with census and American Community Survey inputs (county poverty, the age, sex, race and ethnicity of each tract's residents, and education), with state and county random effects, and the Area Deprivation Index is built from American Community Survey data too, so part of each gradient here is mechanical. Within a county, tract values differ through the demographic composition of tracts. The tract file has no intervals; the bands are 95% intervals for the weighted mean from a bootstrap over counties (replicates and seed in the methods note).

Source: CDC PLACES 2024 release (BRFSS 2022) and Area Deprivation Index 2023; modeled estimates, population-weighted (adults in each tract)

Data: https://vishalsingh.org/zipcode/report/health-behaviors-and-the-supply-of-care/figures/checkup_slopes_by_state.json

### Table 7. Residents per primary care physician rise from 1,159 to 2,730 and per dentist from 1,131 to 3,055 between the least and most deprived tenths of counties

Residents per provider, pooled over the counties in each tenth of counties ranked on the county Area Deprivation Index, providers counted in 2021 (physicians), 2022 (dentists) and 2023 (mental health providers)

| Provider | County deprivation decile | Residents per provider | Lower limit of the 95% interval | Upper limit of the 95% interval | Providers | Residents | Counties | Counties with no provider in the decile |
| --- | ---: | ---: | ---: | ---: | ---: | ---: | ---: | ---: |
| Primary care physicians | 1 | 1,159 | 1,093 | 1,205 | 113,749 | 131,811,420 | 312 | 7 |
| Primary care physicians | 2 | 1,335 | 1,221 | 1,449 | 46,027 | 61,434,044 | 313 | 5 |
| Primary care physicians | 3 | 1,389 | 1,286 | 1,511 | 30,365 | 42,189,945 | 313 | 7 |
| Primary care physicians | 4 | 1,411 | 1,309 | 1,512 | 22,741 | 32,087,932 | 312 | 10 |
| Primary care physicians | 5 | 1,529 | 1,365 | 1,737 | 13,522 | 20,679,797 | 314 | 17 |
| Primary care physicians | 6 | 1,656 | 1,458 | 1,877 | 8,739 | 14,471,394 | 312 | 29 |
| Primary care physicians | 7 | 1,894 | 1,660 | 2,148 | 5,756 | 10,903,268 | 313 | 19 |
| Primary care physicians | 8 | 2,083 | 1,905 | 2,249 | 4,293 | 8,942,931 | 315 | 38 |
| Primary care physicians | 9 | 2,593 | 2,394 | 2,828 | 2,130 | 5,522,704 | 312 | 35 |
| Primary care physicians | 10 | 2,730 | 2,452 | 3,018 | 1,403 | 3,830,042 | 312 | 53 |
| Dentists | 1 | 1,131 | 1,074 | 1,232 | 116,844 | 132,162,193 | 312 | 10 |
| Dentists | 2 | 1,439 | 1,304 | 1,583 | 43,139 | 62,071,203 | 313 | 4 |
| Dentists | 3 | 1,425 | 1,351 | 1,552 | 29,869 | 42,550,714 | 313 | 13 |
| Dentists | 4 | 1,494 | 1,388 | 1,624 | 21,574 | 32,224,541 | 312 | 11 |
| Dentists | 5 | 1,622 | 1,473 | 1,848 | 12,745 | 20,677,282 | 314 | 21 |
| Dentists | 6 | 1,776 | 1,586 | 1,984 | 8,145 | 14,463,884 | 312 | 27 |
| Dentists | 7 | 1,949 | 1,775 | 2,169 | 5,585 | 10,883,473 | 313 | 22 |
| Dentists | 8 | 2,454 | 2,196 | 2,718 | 3,638 | 8,928,681 | 315 | 30 |
| Dentists | 9 | 2,713 | 2,533 | 2,920 | 2,026 | 5,496,600 | 312 | 31 |
| Dentists | 10 | 3,055 | 2,638 | 3,533 | 1,240 | 3,788,578 | 312 | 58 |
| Mental health providers | 1 | 250 | 221 | 307 | 528,315 | 132,162,193 | 312 | 2 |
| Mental health providers | 2 | 351 | 290 | 415 | 176,791 | 62,071,203 | 313 | 2 |
| Mental health providers | 3 | 372 | 312 | 436 | 114,486 | 42,550,714 | 313 | 11 |
| Mental health providers | 4 | 364 | 310 | 425 | 88,447 | 32,224,541 | 312 | 9 |
| Mental health providers | 5 | 402 | 339 | 517 | 51,406 | 20,677,282 | 314 | 16 |
| Mental health providers | 6 | 463 | 363 | 604 | 31,271 | 14,463,884 | 312 | 26 |
| Mental health providers | 7 | 483 | 407 | 612 | 22,524 | 10,883,473 | 313 | 18 |
| Mental health providers | 8 | 551 | 426 | 746 | 16,210 | 8,928,681 | 315 | 32 |
| Mental health providers | 9 | 641 | 505 | 837 | 8,572 | 5,496,600 | 312 | 38 |
| Mental health providers | 10 | 550 | 382 | 882 | 6,883 | 3,788,578 | 312 | 52 |

Note: Counties with no recorded provider stay in: they add residents and no providers. County Health Rankings leaves their published ratio blank above a population cutoff that differs by provider type and reports a rate of 0 below it, and the first edition of this chapter dropped them. Primary care physicians are practicing non-federal physicians under age 75 in general practice, family medicine, internal medicine and pediatrics; dentists are those with a National Provider Identifier; mental health providers include psychiatrists, psychologists, social workers, counselors, therapists and mental health nurse practitioners. Providers are counted where they practice, not by the care patients receive.

Note: Deciles hold about the same number of counties, 15 small counties with no deprivation rank are outside them (so the counts of counties with no provider in the table are smaller than the counts in the text, which cover all counties), and Puerto Rico is not covered.

Note: Across all the counties in the comparison, there are 1,334 residents per primary care physician.

Source: County Health Rankings 2024 (provider counts from AMA, CMS and NPPES files, 2021-2023) and Area Deprivation Index 2023; full counts, pooled ratios

Data: https://vishalsingh.org/zipcode/report/health-behaviors-and-the-supply-of-care/figures/people_per_provider.json

### Table 8. Residents per mental health provider exceed 3,200 in 275 counties, and 91.6% of those counties have a mostly rural population

Residents per mental health provider by county, providers counted in 2023, on a scale that doubles at each step

No inline rows: the values of Table 8 are in https://vishalsingh.org/zipcode/data/health-behaviors-and-the-supply-of-care/mental_health_provider_map.json.

Note: Gray counties have no mental health provider recorded in 2023; County Health Rankings leaves the ratio blank for them. They are not missing data, and their tooltip says so. Puerto Rico is not drawn. Alaska is drawn, and Connecticut appears under its eight legacy counties.

Note: Providers are counted where they practice and each county's residents are the denominator, so a county next to a large city can look thin while its residents are served across the county line.

Source: County Health Rankings 2024 (CMS National Provider Identifier file, 2023); full counts, county ratios

Data: https://vishalsingh.org/zipcode/report/health-behaviors-and-the-supply-of-care/figures/mental_health_provider_map.json

### Table 9. Residents per primary care physician is steadily worse toward the more deprived end under the ADI and under each of the three other rankings

Residents per primary care physician by tenth of counties ranked four ways, over the 3,125 counties that have all four rankings

| bin | conf_high | conf_low | estimate | n_units | ranking |
| ---: | ---: | ---: | ---: | ---: | --- |
| 1 | 1210.2 | 1099.9 | 1163.7 | 312 | Area Deprivation Index |
| 2 | 1446.7 | 1202.1 | 1322.5 | 313 | Area Deprivation Index |
| 3 | 1486.1 | 1275.4 | 1368.9 | 312 | Area Deprivation Index |
| 4 | 1525.9 | 1332.6 | 1430.3 | 313 | Area Deprivation Index |
| 5 | 1763.5 | 1353.0 | 1532.7 | 312 | Area Deprivation Index |
| 6 | 1799.8 | 1472.4 | 1618.0 | 313 | Area Deprivation Index |
| 7 | 2167.6 | 1702.9 | 1921.7 | 312 | Area Deprivation Index |
| 8 | 2183.3 | 1878 | 2036.3 | 313 | Area Deprivation Index |
| 9 | 2899.6 | 2479.4 | 2673.4 | 312 | Area Deprivation Index |
| 10 | 3037.9 | 2473.2 | 2748.7 | 313 | Area Deprivation Index |
| 1 | 1170.1 | 1036.5 | 1093.8 | 312 | Standardized index |
| 2 | 1442.1 | 1209.2 | 1316.6 | 313 | Standardized index |
| 3 | 1373.7 | 1215.9 | 1279.7 | 312 | Standardized index |
| 4 | 1515.3 | 1281.3 | 1409.3 | 313 | Standardized index |
| 5 | 1638.0 | 1450.9 | 1532.5 | 312 | Standardized index |
| 6 | 1806.3 | 1462.2 | 1620.4 | 313 | Standardized index |
| 7 | 1851.3 | 1470.4 | 1648.1 | 312 | Standardized index |
| 8 | 2220.0 | 1813.4 | 2033.2 | 313 | Standardized index |
| 9 | 2549.1 | 2062.8 | 2275.0 | 312 | Standardized index |
| 10 | 2973.2 | 2513.8 | 2687.5 | 313 | Standardized index |
| 1 | 1174.8 | 1041.1 | 1104.0 | 312 | Median household income |
| 2 | 1480.2 | 1288.7 | 1377.4 | 313 | Median household income |
| 3 | 1541.3 | 1308.0 | 1422.2 | 312 | Median household income |
| 4 | 1442.7 | 1254.4 | 1346.0 | 313 | Median household income |
| 5 | 1621.0 | 1397.2 | 1501.7 | 312 | Median household income |
| 6 | 1766.7 | 1431.4 | 1588.2 | 313 | Median household income |
| 7 | 1870.8 | 1494.3 | 1664.4 | 312 | Median household income |
| 8 | 2100.5 | 1780.2 | 1953.9 | 313 | Median household income |
| 9 | 2679.2 | 2321.4 | 2495.9 | 312 | Median household income |
| 10 | 2920.6 | 2385.9 | 2615.3 | 313 | Median household income |
| 1 | 1060.4 | 965.3 | 1009.7 | 312 | Bachelor's degree share |
| 2 | 1354.3 | 1220.3 | 1279.7 | 313 | Bachelor's degree share |
| 3 | 1510.0 | 1342.3 | 1422.8 | 312 | Bachelor's degree share |
| 4 | 1801.7 | 1592.9 | 1690.5 | 313 | Bachelor's degree share |
| 5 | 1937.6 | 1628.6 | 1766.4 | 312 | Bachelor's degree share |
| 6 | 2287.0 | 1875.0 | 2058.4 | 313 | Bachelor's degree share |
| 7 | 2417.3 | 2074.7 | 2235.9 | 312 | Bachelor's degree share |
| 8 | 2668.3 | 2334.0 | 2495.9 | 313 | Bachelor's degree share |
| 9 | 2881.0 | 2467.9 | 2651.0 | 312 | Bachelor's degree share |
| 10 | 3804.9 | 3227.7 | 3519.8 | 313 | Bachelor's degree share |

Note: Least and most deprived tenth: ADI 1,164 and 2,749, rank correlation +1.00; standardized index 1,094 and 2,687, +0.99; median household income 1,104 and 2,615, +0.96; bachelor's degree share 1,010 and 3,520, +1.00.

Note: The standardized index applies the ADI's coefficients to its measures after standardizing each one, so home value and income carry less weight than in the ADI. Income and degree share are single census measures, each ranked from the highest value to the lowest.

Source: County Health Rankings 2024 (provider counts from AMA, CMS and NPPES files, 2021-2023) and Area Deprivation Index 2023; full counts, pooled ratios; ADI 2023 and ACS 2019-2023 tract tables (B19013, B15003 and the 15 rebuilt ADI measures)

Data: https://vishalsingh.org/zipcode/report/health-behaviors-and-the-supply-of-care/figures/alt_rankings.json

## All estimates

Every fact in facts.json (134), sorted by id.

| Fact id | Value | Unit | Estimand | Population | Period | Interval | n | Weights | Source | Note |
| --- | ---: | --- | --- | --- | --- | --- | ---: | --- | --- | --- |
| age_col_hi | 75 | years of age | age limit of the measure's population (a definition) | adults covered by the PLACES measure of colorectal cancer screening | BRFSS 2022 (PLACES 2024 release); ADI 2023 |   |   | none (definition) | CDC PLACES |   |
| age_col_lo | 45 | years of age | age limit of the measure's population (a definition) | adults covered by the PLACES measure of colorectal cancer screening | BRFSS 2022 (PLACES 2024 release); ADI 2023 |   |   | none (definition) | CDC PLACES |   |
| age_ins_hi | 64 | years of age | age limit of the measure's population (a definition) | adults covered by the PLACES measure of no health insurance | BRFSS 2022 (PLACES 2024 release); ADI 2023 |   |   | none (definition) | CDC PLACES |   |
| age_ins_lo | 18 | years of age | age limit of the measure's population (a definition) | adults covered by the PLACES measure of no health insurance | BRFSS 2022 (PLACES 2024 release); ADI 2023 |   |   | none (definition) | CDC PLACES |   |
| age_mam_hi | 74 | years of age | age limit of the measure's population (a definition) | adults covered by the PLACES measure of mammography | BRFSS 2022 (PLACES 2024 release); ADI 2023 |   |   | none (definition) | CDC PLACES |   |
| age_mam_lo | 50 | years of age | age limit of the measure's population (a definition) | adults covered by the PLACES measure of mammography | BRFSS 2022 (PLACES 2024 release); ADI 2023 |   |   | none (definition) | CDC PLACES |   |
| age_teeth | 65 | years of age | age limit of the measure's population (a definition) | adults covered by the PLACES measure of all teeth lost | BRFSS 2022 (PLACES 2024 release); ADI 2023 |   |   | none (definition) | CDC PLACES |   |
| alt_adi_least | 1,164 | Residents per primary care physician | residents of the counties divided by their primary care physicians (a pooled ratio; counties with no physician add residents and no physicians); 95% intervals from a bootstrap over states | the least deprived tenth of counties ranked by Area Deprivation Index, among the counties with all four rankings | primary care physicians in 2021; residents in the same year; rankings from ACS 2019-2023 | 95% CI 1,100 to 1,210 | 312 counties | none (pooled ratio of counts) | County Health Rankings |   |
| alt_adi_most | 2,749 | Residents per primary care physician | residents of the counties divided by their primary care physicians (a pooled ratio; counties with no physician add residents and no physicians); 95% intervals from a bootstrap over states | the most deprived tenth of counties ranked by Area Deprivation Index, among the counties with all four rankings | primary care physicians in 2021; residents in the same year; rankings from ACS 2019-2023 | 95% CI 2,473 to 3,038 | 313 counties | none (pooled ratio of counts) | County Health Rankings |   |
| alt_adi_rho | +1.00 | rank correlation | Spearman rank correlation between bin order and bin mean, each bin counted once | the bins of counties ranked by Area Deprivation Index | primary care physicians in 2021; residents in the same year; rankings from ACS 2019-2023 |   |   | each bin counted once | County Health Rankings |   |
| alt_degree_least | 1,010 | Residents per primary care physician | residents of the counties divided by their primary care physicians (a pooled ratio; counties with no physician add residents and no physicians); 95% intervals from a bootstrap over states | the least deprived tenth of counties ranked by Bachelor's degree share, among the counties with all four rankings | primary care physicians in 2021; residents in the same year; rankings from ACS 2019-2023 | 95% CI 965 to 1,060 | 312 counties | none (pooled ratio of counts) | County Health Rankings |   |
| alt_degree_most | 3,520 | Residents per primary care physician | residents of the counties divided by their primary care physicians (a pooled ratio; counties with no physician add residents and no physicians); 95% intervals from a bootstrap over states | the most deprived tenth of counties ranked by Bachelor's degree share, among the counties with all four rankings | primary care physicians in 2021; residents in the same year; rankings from ACS 2019-2023 | 95% CI 3,228 to 3,805 | 313 counties | none (pooled ratio of counts) | County Health Rankings |   |
| alt_degree_rho | +1.00 | rank correlation | Spearman rank correlation between bin order and bin mean, each bin counted once | the bins of counties ranked by Bachelor's degree share | primary care physicians in 2021; residents in the same year; rankings from ACS 2019-2023 |   |   | each bin counted once | County Health Rankings |   |
| alt_income_least | 1,104 | Residents per primary care physician | residents of the counties divided by their primary care physicians (a pooled ratio; counties with no physician add residents and no physicians); 95% intervals from a bootstrap over states | the least deprived tenth of counties ranked by Median household income, among the counties with all four rankings | primary care physicians in 2021; residents in the same year; rankings from ACS 2019-2023 | 95% CI 1,041 to 1,175 | 312 counties | none (pooled ratio of counts) | County Health Rankings |   |
| alt_income_most | 2,615 | Residents per primary care physician | residents of the counties divided by their primary care physicians (a pooled ratio; counties with no physician add residents and no physicians); 95% intervals from a bootstrap over states | the most deprived tenth of counties ranked by Median household income, among the counties with all four rankings | primary care physicians in 2021; residents in the same year; rankings from ACS 2019-2023 | 95% CI 2,386 to 2,921 | 313 counties | none (pooled ratio of counts) | County Health Rankings |   |
| alt_income_rho | +0.96 | rank correlation | Spearman rank correlation between bin order and bin mean, each bin counted once | the bins of counties ranked by Median household income | primary care physicians in 2021; residents in the same year; rankings from ACS 2019-2023 |   |   | each bin counted once | County Health Rankings |   |
| alt_n_places | 3,125 | counties | count of counties in the common sample of the ranking comparison | counties with the outcome and all four rankings | primary care physicians in 2021; residents in the same year; rankings from ACS 2019-2023 |   |   | none (full count) | County Health Rankings |   |
| alt_rho_cut | 0.9 | absolute rank correlation | threshold between a steady and an uneven gradient | the rule for changed gradients (docs/RANKING_SWITCH_RULE.md) | primary care physicians in 2021; residents in the same year; rankings from ACS 2019-2023 |   |   | none (a fixed rule) | County Health Rankings |   |
| alt_rho_margin | 0.1 | absolute rank correlation | smallest move in the rank correlation that counts as a change of shape | the rule for changed gradients (docs/RANKING_SWITCH_RULE.md) | primary care physicians in 2021; residents in the same year; rankings from ACS 2019-2023 |   |   | none (a fixed rule) | County Health Rankings |   |
| alt_std_least | 1,094 | Residents per primary care physician | residents of the counties divided by their primary care physicians (a pooled ratio; counties with no physician add residents and no physicians); 95% intervals from a bootstrap over states | the least deprived tenth of counties ranked by Standardized index, among the counties with all four rankings | primary care physicians in 2021; residents in the same year; rankings from ACS 2019-2023 | 95% CI 1,037 to 1,170 | 312 counties | none (pooled ratio of counts) | County Health Rankings |   |
| alt_std_most | 2,687 | Residents per primary care physician | residents of the counties divided by their primary care physicians (a pooled ratio; counties with no physician add residents and no physicians); 95% intervals from a bootstrap over states | the most deprived tenth of counties ranked by Standardized index, among the counties with all four rankings | primary care physicians in 2021; residents in the same year; rankings from ACS 2019-2023 | 95% CI 2,514 to 2,973 | 313 counties | none (pooled ratio of counts) | County Health Rankings |   |
| alt_std_rho | +0.99 | rank correlation | Spearman rank correlation between bin order and bin mean, each bin counted once | the bins of counties ranked by Standardized index | primary care physicians in 2021; residents in the same year; rankings from ACS 2019-2023 |   |   | each bin counted once | County Health Rankings |   |
| bpmed_ratio | 1.08 | ratio of prevalence, most deprived tenth to least deprived tenth | weighted mean prevalence in the most deprived tenth of tracts over that in the least deprived tenth, bootstrap interval over counties | adults: taking medicine for high blood pressure, among adults with high blood pressure | BRFSS 2021 (PLACES 2024 release); ADI 2023 | 95% CI 1.06 to 1.09 | 7,946 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| care_dent_all | 1,361 | residents per dentist | residents of all counties divided by their providers, bootstrap interval over states | residents of all counties in County Health Rankings 2024 | dentists in 2022 | 95% CI 1,261 to 1,476 | 3,143 counties | none (pooled ratio of counts) | County Health Rankings |   |
| care_dent_d1 | 1,131 | residents per dentist | residents of the counties divided by their providers (a pooled ratio; counties with no provider add residents and no providers), bootstrap interval over states | residents of counties in the least deprived tenth of counties by county ADI | dentists in 2022; residents in the same year | 95% CI 1,074 to 1,232 | 312 counties | none (pooled ratio of counts) | County Health Rankings |   |
| care_dent_d10 | 3,055 | residents per dentist | residents of the counties divided by their providers (a pooled ratio; counties with no provider add residents and no providers), bootstrap interval over states | residents of counties in the most deprived tenth of counties by county ADI | dentists in 2022; residents in the same year | 95% CI 2,638 to 3,533 | 312 counties | none (pooled ratio of counts) | County Health Rankings |   |
| care_dent_mult | 2.7 | ratio of residents per dentist, most deprived tenth of counties to least deprived tenth | pooled residents per provider in the most deprived tenth of counties over that in the least deprived tenth, bootstrap interval over states | residents of counties | dentists in 2022 | 95% CI 2.3 to 3.1 | 312 counties | none (pooled ratio of counts) | County Health Rankings |   |
| care_dent_r | −0.52 | correlation coefficient | county-population-weighted correlation between the county ADI and providers per 100,000 residents (counties with no provider count as zero), bootstrap interval over states | residents of counties with an ADI rank: county ADI against dentists per 100,000 residents | dentists in 2022 | 95% CI −0.61 to −0.41 | 3,128 counties | county residents (CHR denominator for the provider measure) | County Health Rankings |   |
| care_dent_r_unw | −0.35 | correlation coefficient | unweighted correlation between the county ADI and providers per 100,000 residents (counties with no provider count as zero), bootstrap interval over states | counties with an ADI rank: county ADI against dentists per 100,000 residents, each county counted once | dentists in 2022 | 95% CI −0.45 to −0.27 | 3,128 counties | none (each county counts once) | County Health Rankings |   |
| care_fei_bot | 6.1 | food environment index (0 worst, 10 best) | county-population-weighted mean of the county index, bootstrap interval over states | residents of the most deprived tenth of counties | 2019 and 2021 | 95% CI 5.7 to 6.6 | 306 counties | county population | County Health Rankings |   |
| care_fei_r | −0.66 | correlation coefficient | county-population-weighted correlation, bootstrap interval over states | residents of counties with an ADI rank: county ADI against the food environment index | 2019 and 2021 | 95% CI −0.74 to −0.59 | 3,094 counties | county population | County Health Rankings |   |
| care_fei_r_unw | −0.51 | correlation coefficient | unweighted correlation, bootstrap interval over states | counties with an ADI rank: county ADI against the food environment index, each county counted once | 2019 and 2021 | 95% CI −0.58 to −0.45 | 3,094 counties | none (each county counts once) | County Health Rankings |   |
| care_fei_top | 8.6 | food environment index (0 worst, 10 best) | county-population-weighted mean of the county index, bootstrap interval over states | residents of the least deprived tenth of counties | 2019 and 2021 | 95% CI 8.5 to 8.8 | 308 counties | county population | County Health Rankings |   |
| care_mh_all | 319 | residents per mental health provider | residents of all counties divided by their providers, bootstrap interval over states | residents of all counties in County Health Rankings 2024 | mental health providers in 2023 | 95% CI 274 to 382 | 3,143 counties | none (pooled ratio of counts) | County Health Rankings |   |
| care_mh_d1 | 250 | residents per mental health provider | residents of the counties divided by their providers (a pooled ratio; counties with no provider add residents and no providers), bootstrap interval over states | residents of counties in the least deprived tenth of counties by county ADI | mental health providers in 2023; residents in the same year | 95% CI 221 to 307 | 312 counties | none (pooled ratio of counts) | County Health Rankings |   |
| care_mh_d10 | 550 | residents per mental health provider | residents of the counties divided by their providers (a pooled ratio; counties with no provider add residents and no providers), bootstrap interval over states | residents of counties in the most deprived tenth of counties by county ADI | mental health providers in 2023; residents in the same year | 95% CI 382 to 882 | 312 counties | none (pooled ratio of counts) | County Health Rankings |   |
| care_mh_mult | 2.2 | ratio of residents per mental health provider, most deprived tenth of counties to least deprived tenth | pooled residents per provider in the most deprived tenth of counties over that in the least deprived tenth, bootstrap interval over states | residents of counties | mental health providers in 2023 | 95% CI 1.5 to 3.5 | 312 counties | none (pooled ratio of counts) | County Health Rankings |   |
| care_mh_r | −0.42 | correlation coefficient | county-population-weighted correlation between the county ADI and providers per 100,000 residents (counties with no provider count as zero), bootstrap interval over states | residents of counties with an ADI rank: county ADI against mental health providers per 100,000 residents | mental health providers in 2023 | 95% CI −0.53 to −0.26 | 3,128 counties | county residents (CHR denominator for the provider measure) | County Health Rankings |   |
| care_mh_r_unw | −0.33 | correlation coefficient | unweighted correlation between the county ADI and providers per 100,000 residents (counties with no provider count as zero), bootstrap interval over states | counties with an ADI rank: county ADI against mental health providers per 100,000 residents, each county counted once | mental health providers in 2023 | 95% CI −0.44 to −0.23 | 3,128 counties | none (each county counts once) | County Health Rankings |   |
| care_pcp_all | 1,334 | residents per primary care physician | residents of all counties divided by their providers, bootstrap interval over states | residents of all counties in County Health Rankings 2024 | primary care physicians in 2021 | 95% CI 1,276 to 1,406 | 3,143 counties | none (pooled ratio of counts) | County Health Rankings |   |
| care_pcp_d1 | 1,159 | residents per primary care physician | residents of the counties divided by their providers (a pooled ratio; counties with no provider add residents and no providers), bootstrap interval over states | residents of counties in the least deprived tenth of counties by county ADI | primary care physicians in 2021; residents in the same year | 95% CI 1,093 to 1,205 | 312 counties | none (pooled ratio of counts) | County Health Rankings |   |
| care_pcp_d10 | 2,730 | residents per primary care physician | residents of the counties divided by their providers (a pooled ratio; counties with no provider add residents and no providers), bootstrap interval over states | residents of counties in the most deprived tenth of counties by county ADI | primary care physicians in 2021; residents in the same year | 95% CI 2,452 to 3,018 | 312 counties | none (pooled ratio of counts) | County Health Rankings |   |
| care_pcp_mult | 2.4 | ratio of residents per primary care physician, most deprived tenth of counties to least deprived tenth | pooled residents per provider in the most deprived tenth of counties over that in the least deprived tenth, bootstrap interval over states | residents of counties | primary care physicians in 2021 | 95% CI 2.1 to 2.6 | 312 counties | none (pooled ratio of counts) | County Health Rankings |   |
| care_pcp_r | −0.41 | correlation coefficient | county-population-weighted correlation between the county ADI and providers per 100,000 residents (counties with no provider count as zero), bootstrap interval over states | residents of counties with an ADI rank: county ADI against primary care physicians per 100,000 residents | primary care physicians in 2021 | 95% CI −0.46 to −0.35 | 3,128 counties | county residents (CHR denominator for the provider measure) | County Health Rankings |   |
| care_pcp_r_unw | −0.37 | correlation coefficient | unweighted correlation between the county ADI and providers per 100,000 residents (counties with no provider count as zero), bootstrap interval over states | counties with an ADI rank: county ADI against primary care physicians per 100,000 residents, each county counted once | primary care physicians in 2021 | 95% CI −0.43 to −0.32 | 3,128 counties | none (each county counts once) | County Health Rankings |   |
| care_ph_hi | 4,012 | preventable hospital stays per 100,000 Medicare enrollees | county-population-weighted mean of the county rate, bootstrap interval over states | Medicare enrollees in the most deprived tenth of counties | 2021 | 95% CI 3,608 to 4,435 | 305 counties | county population (enrollee counts are not in the staged file) | County Health Rankings |   |
| care_ph_lo | 2,370 | preventable hospital stays per 100,000 Medicare enrollees | county-population-weighted mean of the county rate, bootstrap interval over states | Medicare enrollees in the least deprived tenth of counties | 2021 | 95% CI 2,188 to 2,599 | 304 counties | county population (enrollee counts are not in the staged file) | County Health Rankings |   |
| care_ph_r | +0.46 | correlation coefficient | county-population-weighted correlation, bootstrap interval over states | Medicare enrollees in counties with an ADI rank: county ADI against preventable hospital stays | 2021 | 95% CI 0.31 to 0.56 | 3,067 counties | county population | County Health Rankings |   |
| care_ph_r_unw | +0.38 | correlation coefficient | unweighted correlation, bootstrap interval over states | counties with an ADI rank: county ADI against preventable hospital stays, each county counted once | 2021 | 95% CI 0.30 to 0.46 | 3,067 counties | none (each county counts once) | County Health Rankings |   |
| cat_band | 0.5 | points of prevalence per ten-point step in the ADI national percentile | half-width of the band around zero used to call a slope small (a definition) | the checkup slopes of the states | BRFSS 2022 (PLACES 2024 release); ADI 2023 |   |   | none (rule) | CDC PLACES |   |
| cat_checkup_maxabs | +1.7 | points of prevalence per ten-point step in the ADI national percentile | population-weighted slope of tract prevalence on the ADI national percentile, times ten; the interval is HC3 because the District of Columbia is a single county | the state with the largest routine checkup slope in absolute value (District of Columbia), adults | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 95% CI 1.3 to 2.1 | 201 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| cat_checkup_med | +0.03 | points of prevalence per ten-point step in the ADI national percentile | median of the state estimates, each a population-weighted slope of tract prevalence on the ADI national percentile, times ten | states and the District of Columbia with at least 60 tracts that have an ADI rank and both measures, routine checkup in the past year | BRFSS 2022 (PLACES 2024 release); ADI 2023 |   | 51 states | adult population of the tract (PLACES) | CDC PLACES | A median of 51 estimates; each state's interval is in the figure. |
| cat_checkup_signeg | 18 | states | count of states whose 95% interval for the routine checkup slope lies below zero | states and the District of Columbia with at least 60 tracts that have an ADI rank and both measures | BRFSS 2022 (PLACES 2024 release); ADI 2023 |   | 51 states | none (full count) | CDC PLACES |   |
| cat_checkup_sigpos | 10 | states | count of states whose 95% interval for the routine checkup slope lies above zero | states and the District of Columbia with at least 60 tracts that have an ADI rank and both measures | BRFSS 2022 (PLACES 2024 release); ADI 2023 |   | 51 states | none (full count) | CDC PLACES |   |
| cat_checkup_small | 42 | states | count of states whose routine checkup slope is within the band around zero | states and the District of Columbia with at least 60 tracts that have an ADI rank and both measures | BRFSS 2022 (PLACES 2024 release); ADI 2023 |   | 51 states | none (full count) | CDC PLACES |   |
| cat_checkup_span | 23 | states | count of states whose 95% interval for the routine checkup slope includes zero | states and the District of Columbia with at least 60 tracts that have an ADI rank and both measures | BRFSS 2022 (PLACES 2024 release); ADI 2023 |   | 51 states | none (full count) | CDC PLACES |   |
| cat_dental_med | −3.2 | points of prevalence per ten-point step in the ADI national percentile | median of the state estimates, each a population-weighted slope of tract prevalence on the ADI national percentile, times ten | states and the District of Columbia with at least 60 tracts that have an ADI rank and both measures, dental visit in the past year | BRFSS 2022 (PLACES 2024 release); ADI 2023 |   | 51 states | adult population of the tract (PLACES) | CDC PLACES | A median of 51 estimates; each state's interval is in the figure. |
| cat_dental_negn | 51 | states | count of states whose dental visit slope on the ADI is negative | states and the District of Columbia with at least 60 tracts that have an ADI rank and both measures | BRFSS 2022 (PLACES 2024 release); ADI 2023 |   | 51 states | none (full count) | CDC PLACES |   |
| cat_dental_signeg | 50 | states | count of states whose 95% interval for the dental visit slope lies below zero | states and the District of Columbia with at least 60 tracts that have an ADI rank and both measures | BRFSS 2022 (PLACES 2024 release); ADI 2023 |   | 51 states | none (full count) | CDC PLACES |   |
| cat_dental_small | 0 | states | count of states whose dental visit slope is within the band around zero | states and the District of Columbia with at least 60 tracts that have an ADI rank and both measures | BRFSS 2022 (PLACES 2024 release); ADI 2023 |   | 51 states | none (full count) | CDC PLACES |   |
| cat_floor | 60 | tracts | minimum number of tracts with every variable for a state to enter (an inclusion rule that drops no state) | states in the state slopes and the coefficient zoo | BRFSS 2022 (PLACES 2024 release); ADI 2023 |   |   | none (rule) | CDC PLACES |   |
| cat_min_tracts | 158 | tracts | smallest state tract count | the state with the fewest tracts that have an ADI rank and both measures | BRFSS 2022 (PLACES 2024 release); ADI 2023 |   |   | none (full count) | CDC PLACES |   |
| cat_n_states | 51 | states | count of states and the District of Columbia | states and the District of Columbia with at least 60 tracts that have an ADI rank and both measures | BRFSS 2022 (PLACES 2024 release); ADI 2023 |   |   | none (full count) | CDC PLACES |   |
| checkup_hi | 77.8% | percent of adults | weighted mean of the tract prevalence in a tenth of tracts (equal numbers of tracts), bootstrap interval over counties | adults in the most deprived tenth of tracts by ADI national percentile: routine checkup in the past year | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 95% CI 77.3% to 78.2% | 8,213 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| checkup_lo | 74.2% | percent of adults | weighted mean of the tract prevalence in a tenth of tracts (equal numbers of tracts), bootstrap interval over counties | adults in the least deprived tenth of tracts by ADI national percentile: routine checkup in the past year | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 95% CI 73.2% to 75.3% | 8,274 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| checkup_ratio | 1.05 | ratio of prevalence, most deprived tenth to least deprived tenth | weighted mean prevalence in the most deprived tenth of tracts over that in the least deprived tenth, bootstrap interval over counties | adults: routine checkup in the past year | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 95% CI 1.03 to 1.06 | 8,213 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| chr_cut_dent | 4,000 | residents | population above which the staged ratio is blank and below which it holds a negative population, read from the file and rounded to the nearest thousand | counties with no recorded dentist | dentists in 2022 |   |   | none (rule) | County Health Rankings |   |
| chr_cut_mh | 1,000 | residents | population above which the staged ratio is blank and below which it holds a negative population, read from the file and rounded to the nearest thousand | counties with no recorded mental health provider | mental health providers in 2023 |   |   | none (rule) | County Health Rankings |   |
| chr_cut_pcp | 2,000 | residents | population above which the staged ratio is blank and below which it holds a negative population, read from the file and rounded to the nearest thousand | counties with no recorded primary care physician | primary care physicians in 2021 |   |   | none (rule) | County Health Rankings |   |
| dep_delta | 4.8 | percentage points of adults, higher in the most deprived tenth of tracts | difference between the weighted mean prevalence of the least deprived and the most deprived tenth of tracts by ADI national percentile, bootstrap interval over counties (draws shared by the two means) | adults: depression | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 95% CI 4.1 to 5.6 | 8,213 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| dep_ratio | 1.25 | ratio of prevalence, most deprived tenth to least deprived tenth | weighted mean prevalence in the most deprived tenth of tracts over that in the least deprived tenth, bootstrap interval over counties | adults: depression | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 95% CI 1.21 to 1.30 | 8,213 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| diab_hi | 17.5% | percent of adults | weighted mean of the tract prevalence in a tenth of tracts (equal numbers of tracts), bootstrap interval over counties | adults in the most deprived tenth of tracts by ADI national percentile: diagnosed diabetes | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 95% CI 17.3% to 17.9% | 8,213 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| diab_lo | 9.5% | percent of adults | weighted mean of the tract prevalence in a tenth of tracts (equal numbers of tracts), bootstrap interval over counties | adults in the least deprived tenth of tracts by ADI national percentile: diagnosed diabetes | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 95% CI 9.0% to 9.9% | 8,274 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| flat_bpmed_delta | 5.9 | percentage points of adults, higher in the most deprived tenth of tracts | difference between the weighted mean prevalence of the least deprived and the most deprived tenth of tracts by ADI national percentile, bootstrap interval over counties (draws shared by the two means) | adults: taking medicine for high blood pressure, among adults with high blood pressure | BRFSS 2021 (PLACES 2024 release); ADI 2023 | 95% CI 4.8 to 6.7 | 7,946 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| flat_checkup_delta | 3.6 | percentage points of adults, higher in the most deprived tenth of tracts | difference between the weighted mean prevalence of the least deprived and the most deprived tenth of tracts by ADI national percentile, bootstrap interval over counties (draws shared by the two means) | adults: routine checkup in the past year | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 95% CI 2.4 to 4.6 | 8,213 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| flat_chol_delta | 6.9 | percentage points of adults, lower in the most deprived tenth of tracts | difference between the weighted mean prevalence of the least deprived and the most deprived tenth of tracts by ADI national percentile, bootstrap interval over counties (draws shared by the two means) | adults: cholesterol screening in the past five years | BRFSS 2021 (PLACES 2024 release); ADI 2023 | 95% CI 6.2 to 7.5 | 7,946 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| flat_colon_delta | 6.2 | percentage points of adults, lower in the most deprived tenth of tracts | difference between the weighted mean prevalence of the least deprived and the most deprived tenth of tracts by ADI national percentile, bootstrap interval over counties (draws shared by the two means) | adults: colorectal cancer screening up to date, ages 45 to 75 | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 95% CI 4.4 to 8.2 | 8,213 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| flat_dental_d1 | 70.3% | percent of adults | weighted mean of the tract prevalence in a tenth of tracts (equal numbers of tracts), bootstrap interval over counties | adults in the least deprived tenth of tracts by ADI national percentile: dental visit in the past year | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 95% CI 68.5% to 72.1% | 8,274 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| flat_dental_d10 | 45.8% | percent of adults | weighted mean of the tract prevalence in a tenth of tracts (equal numbers of tracts), bootstrap interval over counties | adults in the most deprived tenth of tracts by ADI national percentile: dental visit in the past year | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 95% CI 45.0% to 46.6% | 8,213 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| flat_dental_delta | 24.4 | percentage points of adults, lower in the most deprived tenth of tracts | difference between the weighted mean prevalence of the least deprived and the most deprived tenth of tracts by ADI national percentile, bootstrap interval over counties (draws shared by the two means) | adults: dental visit in the past year | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 95% CI 22.5 to 26.5 | 8,213 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| flat_mammo_delta | 7.0 | percentage points of adults, lower in the most deprived tenth of tracts | difference between the weighted mean prevalence of the least deprived and the most deprived tenth of tracts by ADI national percentile, bootstrap interval over counties (draws shared by the two means) | adults: mammography in the past two years, women ages 50 to 74 | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 95% CI 6.3 to 7.8 | 8,213 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| flat_notrans_delta | 9.7 | percentage points of adults, higher in the most deprived tenth of tracts | difference between the weighted mean prevalence of the least deprived and the most deprived tenth of tracts by ADI national percentile, bootstrap interval over counties (draws shared by the two means) | adults: lack of reliable transportation in the past 12 months | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 95% CI 9.0 to 10.5 | 6,348 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| flat_uninsured_d1 | 6.3% | percent of adults | weighted mean of the tract prevalence in a tenth of tracts (equal numbers of tracts), bootstrap interval over counties | adults in the least deprived tenth of tracts by ADI national percentile: no health insurance, ages 18 to 64 | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 95% CI 5.7% to 6.8% | 8,274 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| flat_uninsured_d10 | 17.4% | percent of adults | weighted mean of the tract prevalence in a tenth of tracts (equal numbers of tracts), bootstrap interval over counties | adults in the most deprived tenth of tracts by ADI national percentile: no health insurance, ages 18 to 64 | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 95% CI 16.1% to 18.9% | 8,213 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| flat_uninsured_delta | 11.1 | percentage points of adults, higher in the most deprived tenth of tracts | difference between the weighted mean prevalence of the least deprived and the most deprived tenth of tracts by ADI national percentile, bootstrap interval over counties (draws shared by the two means) | adults: no health insurance, ages 18 to 64 | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 95% CI 9.7 to 12.7 | 8,213 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| map_hi_n | 275 | counties | count of counties with more residents per provider than the break | counties with at least one recorded mental health provider | mental health providers in 2023 |   |   | none (full count) | County Health Rankings |   |
| map_hi_rural | 91.6% | percent of those counties | share of the counties that are majority rural (CHR percent rural above 50, 2020 census) | counties with more than 3,200 residents per mental health provider | mental health providers in 2023; rural share 2020 census |   | 275 counties | none (each county counts once) | County Health Rankings |   |
| map_hi_sm | 96.0% | percent of those counties | share of the counties in the Midwest or South Census regions | counties with more than 3,200 residents per mental health provider | mental health providers in 2023 |   | 275 counties | none (each county counts once) | County Health Rankings |   |
| map_hi_thr | 3,200 | residents per mental health provider | a break of the map's key (a definition) | counties on the map | mental health providers in 2023 |   |   | none (rule) | County Health Rankings |   |
| map_n_gray | 216 | counties | count of counties with no recorded mental health provider | counties drawn on the map (Puerto Rico not drawn) | mental health providers in 2023 |   |   | none (full count) | County Health Rankings |   |
| n_adi_tracts | 82,781 | tracts | count of tracts | census tracts with a PLACES estimate and an Area Deprivation Index rank | BRFSS 2022 (PLACES 2024 release); ADI 2023 |   |   | none (full count) | CDC PLACES |   |
| n_care_counties | 3,128 | counties | count of counties | counties and county equivalents in County Health Rankings 2024 with an ADI rank | 2021 to 2023 |   |   | none (full count) | County Health Rankings |   |
| n_chr_counties | 3,143 | counties | count of counties | counties and county equivalents in County Health Rankings 2024 (Puerto Rico not included) | 2021 to 2023 |   |   | none (full count) | County Health Rankings |   |
| n_higher_measures | 34 | measures | count of measures whose 95% interval for the ratio of the most deprived tenth to the least deprived tenth lies above 1 | PLACES measures in the tract file | BRFSS 2022 (PLACES 2024 release); ADI 2023 |   | 40 measures | adult population of the tract (PLACES) | CDC PLACES |   |
| n_last_step | 11 | measures | count of measures whose largest step between neighboring twentieths of tracts, ranked by ADI, is the step into the most deprived twentieth | the twelve PLACES measures of the twelve-outcome figure | BRFSS 2022 (PLACES 2024 release); ADI 2023 |   | 12 measures | adult population of the tract (PLACES) | CDC PLACES | Point estimates of the twentieth means; the intervals of the means are in the figure. |
| n_less_points | 4 | measures | count of measures, none of them plotted in the twelve-outcome figure, whose difference in points between the most deprived and the least deprived tenth is smaller than the checkup's | PLACES measures that are higher in the most deprived tenth | BRFSS 2022 (PLACES 2024 release); ADI 2023 |   | 34 measures | adult population of the tract (PLACES) | CDC PLACES |   |
| n_no_adi | 741 | tracts | count of tracts | census tracts with a PLACES estimate and no Area Deprivation Index rank | BRFSS 2022 (PLACES 2024 release); ADI 2023 |   |   | none (full count) | Area Deprivation Index | The ADI is missing for tracts whose block groups are all suppressed. |
| n_no_adi_counties | 15 | counties | count of counties with no tract that has a deprivation rank and a population | counties in County Health Rankings 2024 | tract ranks from ADI 2023, ACS 2019-2023 population |   |   | none (full count) | Area Deprivation Index |   |
| n_panels | 12 | measures | count of measures | PLACES measures in the twelve-outcome figure | BRFSS 2022 (PLACES 2024 release); ADI 2023 |   |   | none (rule) | CDC PLACES |   |
| n_places_measures | 40 | measures | count of measures (rule) | PLACES measures in the tract file | BRFSS 2022 (PLACES 2024 release); ADI 2023 |   |   | none (rule) | CDC PLACES |   |
| n_places_tracts | 83,522 | tracts | count of tracts | 2020 census tracts with a CDC PLACES 2024 release estimate | PLACES 2024 release |   |   | none (full count) | CDC PLACES |   |
| n_rising | 11 | measures | count of measures (rule) | the twelve PLACES measures of the twelve-outcome figure other than the annual checkup | BRFSS 2022 (PLACES 2024 release); ADI 2023 |   |   | none (rule) | CDC PLACES |   |
| n_spec_tracts | 82,779 | tracts | count of tracts | tracts with an ADI rank, both measures and the three ACS controls | BRFSS 2022 (PLACES 2024 release); ADI 2023 |   |   | none (full count) | CDC PLACES |   |
| n_zero_dent | 239 | counties | count of counties with no recorded dentist | counties in County Health Rankings 2024 | dentists in 2022 |   |   | none (full count) | County Health Rankings |   |
| n_zero_mh | 216 | counties | count of counties with no recorded mental health provider | counties in County Health Rankings 2024 | mental health providers in 2023 |   |   | none (full count) | County Health Rankings |   |
| n_zero_pcp | 232 | counties | count of counties with no recorded primary care physician | counties in County Health Rankings 2024 | primary care physicians in 2021 |   |   | none (full count) | County Health Rankings |   |
| smk_hi | 22.8% | percent of adults | weighted mean of the tract prevalence in a tenth of tracts (equal numbers of tracts), bootstrap interval over counties | adults in the most deprived tenth of tracts by ADI national percentile: current cigarette smoking | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 95% CI 22.5% to 23.1% | 8,213 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| smk_lo | 9.1% | percent of adults | weighted mean of the tract prevalence in a tenth of tracts (equal numbers of tracts), bootstrap interval over counties | adults in the least deprived tenth of tracts by ADI national percentile: current cigarette smoking | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 95% CI 8.5% to 9.6% | 8,274 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| smk_mult | 2.5 | ratio of prevalence, most deprived tenth to least deprived tenth | weighted mean prevalence in the most deprived tenth of tracts over that in the least deprived tenth, bootstrap interval over counties | adults: current cigarette smoking | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 95% CI 2.4 to 2.7 | 8,213 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| spec_checkup_all | +0.41 | points of prevalence per ten-point step in the ADI national percentile | population-weighted least-squares slope of tract prevalence on the ADI national percentile, times ten; all tracts; cluster-robust interval with counties as clusters | adults: routine checkup in the past year, all tracts | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 95% CI 0.29 to 0.54 | 82,779 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| spec_checkup_county | −0.27 | points of prevalence per ten-point step in the ADI national percentile | population-weighted least-squares slope of tract prevalence on the ADI national percentile, times ten; county fixed effects; cluster-robust interval with counties as clusters | adults: routine checkup in the past year, within counties | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 95% CI −0.35 to −0.19 | 82,779 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| spec_checkup_ctrl | −0.05 | points of prevalence per ten-point step in the ADI national percentile | population-weighted least-squares slope of tract prevalence on the ADI national percentile, times ten; state fixed effects and tract shares of residents ages 65 and over, Black alone and Hispanic; cluster-robust interval with counties as clusters | adults: routine checkup in the past year, within states, with age and race controls | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 95% CI −0.10 to 0.01 | 82,779 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| spec_checkup_maxwithin | 0.27 | points of prevalence per ten-point step in the ADI national percentile | largest absolute checkup slope among the specifications that hold the state or county fixed | adults: routine checkup, the three specifications within states or counties | BRFSS 2022 (PLACES 2024 release); ADI 2023 |   | 82,779 tracts | adult population of the tract (PLACES) | CDC PLACES | A point estimate; the intervals are in the figure. |
| spec_checkup_state | −0.01 | points of prevalence per ten-point step in the ADI national percentile | population-weighted least-squares slope of tract prevalence on the ADI national percentile, times ten; state fixed effects; cluster-robust interval with counties as clusters | adults: routine checkup in the past year, within states | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 95% CI −0.08 to 0.05 | 82,779 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| spec_dental_all | −2.37 | points of prevalence per ten-point step in the ADI national percentile | population-weighted least-squares slope of tract prevalence on the ADI national percentile, times ten; all tracts; cluster-robust interval with counties as clusters | adults: dental visit in the past year, all tracts | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 95% CI −2.66 to −2.08 | 82,779 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| spec_dental_county | −3.86 | points of prevalence per ten-point step in the ADI national percentile | population-weighted least-squares slope of tract prevalence on the ADI national percentile, times ten; county fixed effects; cluster-robust interval with counties as clusters | adults: dental visit in the past year, within counties | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 95% CI −3.99 to −3.73 | 82,779 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| spec_dental_ctrl | −2.38 | points of prevalence per ten-point step in the ADI national percentile | population-weighted least-squares slope of tract prevalence on the ADI national percentile, times ten; state fixed effects and tract shares of residents ages 65 and over, Black alone and Hispanic; cluster-robust interval with counties as clusters | adults: dental visit in the past year, within states, with age and race controls | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 95% CI −2.51 to −2.24 | 82,779 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| spec_dental_state | −2.97 | points of prevalence per ten-point step in the ADI national percentile | population-weighted least-squares slope of tract prevalence on the ADI national percentile, times ten; state fixed effects; cluster-robust interval with counties as clusters | adults: dental visit in the past year, within states | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 95% CI −3.14 to −2.80 | 82,779 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| spec_ratio_min | 8.8 | ratio of absolute slopes | smallest absolute dental slope over the largest absolute checkup slope among the specifications that hold the state or county fixed | adults: dental visit slope over routine checkup slope, the three specifications within states or counties | BRFSS 2022 (PLACES 2024 release); ADI 2023 |   | 82,779 tracts | adult population of the tract (PLACES) | CDC PLACES | A ratio of two point estimates. |
| teeth_ratio | 3.6 | ratio of prevalence, most deprived tenth to least deprived tenth | weighted mean prevalence in the most deprived tenth of tracts over that in the least deprived tenth, bootstrap interval over counties | adults: all teeth lost, among adults ages 65 and over | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 95% CI 3.1 to 4.1 | 8,213 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| tenth_max | 8,511 | tracts | largest count of tracts in a tenth (tied ranks fall in the same tenth) | tracts in the largest tenth by ADI national percentile | BRFSS 2022 (PLACES 2024 release); ADI 2023 |   |   | none (full count) | CDC PLACES |   |
| tenth_min | 8,118 | tracts | smallest count of tracts in a tenth (tied ranks fall in the same tenth) | tracts in the smallest tenth by ADI national percentile | BRFSS 2022 (PLACES 2024 release); ADI 2023 |   |   | none (full count) | CDC PLACES |   |
| us_pcp | 1,334 | residents per primary care physician | residents of all the counties of the tenths divided by their providers (a pooled ratio; counties with no provider add residents and no providers), bootstrap interval over states (the draws of the tenths) | residents of all the counties in the comparison: counties in County Health Rankings 2024 with an ADI rank | primary care physicians in 2021; residents in the same year | 95% CI 1,276 to 1,406 | 3,128 counties | none (pooled ratio of counts) | County Health Rankings |   |
| zoo_binge_med | +1.2 | points of prevalence per doubling of tract median household income | median of the state estimates, each a population-weighted regression across the state's tracts of tract prevalence on the log of median household income, with tract shares of residents ages 65 and over, Black alone and Hispanic as controls, times the natural log of 2 | adults: binge drinking, states and the District of Columbia with at least 60 tracts that have income, the three controls and the measure | BRFSS 2022 (PLACES 2024 release); ACS 2019-2023 |   | 51 states | adult population of the tract (PLACES) | CDC PLACES | A median of the state estimates; each state's interval is in the table. |
| zoo_binge_posn | 50 | states | count of states where binge drinking is higher in higher-income tracts | states and the District of Columbia with at least 60 tracts that have income, the three controls and the measure | BRFSS 2022 (PLACES 2024 release); ACS 2019-2023 |   | 51 states | none (full count) | CDC PLACES |   |
| zoo_binge_sig | 41 | states | count of states whose 95% interval for the binge drinking slope lies above zero | states and the District of Columbia with at least 60 tracts that have income, the three controls and the measure | BRFSS 2022 (PLACES 2024 release); ACS 2019-2023 |   | 51 states | none (full count) | CDC PLACES |   |
| zoo_five_negn | 51 | states | count of states where all five other behaviors have a negative slope on income | states and the District of Columbia with at least 60 tracts that have income, the three controls and the measure | BRFSS 2022 (PLACES 2024 release); ACS 2019-2023 |   | 51 states | none (full count) | CDC PLACES |   |
| zoo_five_sigmin | 47 | states | smallest, over the five other behaviors, of the count of states whose 95% interval for the slope lies below zero | states and the District of Columbia with at least 60 tracts that have income, the three controls and the measure | BRFSS 2022 (PLACES 2024 release); ACS 2019-2023 |   | 51 states | none (full count) | CDC PLACES |   |
| zoo_inact_med | −6.7 | points of prevalence per doubling of tract median household income | median of the state estimates, each a population-weighted regression across the state's tracts of tract prevalence on the log of median household income, with tract shares of residents ages 65 and over, Black alone and Hispanic as controls, times the natural log of 2 | adults: no leisure-time physical activity, states and the District of Columbia with at least 60 tracts that have income, the three controls and the measure | BRFSS 2022 (PLACES 2024 release); ACS 2019-2023 |   | 51 states | adult population of the tract (PLACES) | CDC PLACES | A median of the state estimates; each state's interval is in the table. |
| zoo_n_measures | 6 | measures | count of measures (rule) | PLACES measures in the coefficient zoo | BRFSS 2022 (PLACES 2024 release); ACS 2019-2023 |   |   | none (rule) | CDC PLACES |   |
| zoo_n_models | 306 | regressions | count of state-by-behavior regressions | states and the District of Columbia with at least 60 tracts that have income, the three controls and the measure | BRFSS 2022 (PLACES 2024 release); ACS 2019-2023 |   |   | none (full count) | CDC PLACES |   |
| zoo_n_states | 51 | states | count of states and the District of Columbia | states and the District of Columbia with at least 60 tracts that have income, the three controls and the measure | BRFSS 2022 (PLACES 2024 release); ACS 2019-2023 |   |   | none (full count) | CDC PLACES |   |
| zoo_obes_med | −4.3 | points of prevalence per doubling of tract median household income | median of the state estimates, each a population-weighted regression across the state's tracts of tract prevalence on the log of median household income, with tract shares of residents ages 65 and over, Black alone and Hispanic as controls, times the natural log of 2 | adults: obesity, states and the District of Columbia with at least 60 tracts that have income, the three controls and the measure | BRFSS 2022 (PLACES 2024 release); ACS 2019-2023 |   | 51 states | adult population of the tract (PLACES) | CDC PLACES | A median of the state estimates; each state's interval is in the table. |
| zoo_smoke_med | −4.7 | points of prevalence per doubling of tract median household income | median of the state estimates, each a population-weighted regression across the state's tracts of tract prevalence on the log of median household income, with tract shares of residents ages 65 and over, Black alone and Hispanic as controls, times the natural log of 2 | adults: current cigarette smoking, states and the District of Columbia with at least 60 tracts that have income, the three controls and the measure | BRFSS 2022 (PLACES 2024 release); ACS 2019-2023 |   | 51 states | adult population of the tract (PLACES) | CDC PLACES | A median of the state estimates; each state's interval is in the table. |

## Definitions

- Modeled estimate: A figure produced by a statistical model instead of a direct count or measurement, usually for small areas where direct data are thin, borrowing strength from surveys, neighboring areas or other data. Neighboring areas can share errors, so single extremes call for caution.
- Census tract: A small area drawn by the Census Bureau to stand in for a neighborhood. Boundaries are redrawn after each decennial census.
- Area Deprivation Index (ADI): A score that ranks neighborhoods by income, education, employment and housing conditions drawn from the American Community Survey. In this book a higher national percentile means a more deprived neighborhood.
- Fixed effects: A regression adjustment that gives each group, such as each county, its own baseline. The estimate then rests on differences within groups.
- Inverse care law: The pattern that people who need medical care most tend to receive the least of it. The physician Julian Tudor Hart named it in 1971.

## Methods

This chapter compares adults in tenths of census tracts ranked by the Area Deprivation Index 2023, and counties ranked the same way, on health behaviors, access to care and provider supply. The tract values are model-based estimates from CDC PLACES, built from BRFSS 2022 (2021 for four measures) with poverty, demographic and education inputs, not direct counts, and the staged tract file has no intervals. Outcome, deprivation and provider counts come from different years (provider counts are for 2021 to 2023). Tracts and counties without a deprivation rank are left out (741 tracts and 15 small counties), and a county with no recorded provider adds residents and no providers to a provider ratio. Means are weighted by the tract's adult population.

### How tract prevalence is estimated

Tract prevalence is CDC PLACES, 2024 release, crude prevalence among adults: model-based estimates built from BRFSS 2022 (2021 for blood pressure, blood pressure medicine, high cholesterol and cholesterol screening) with county poverty, the age, sex, race and ethnicity of each tract's residents and education from the American Community Survey and the census, and state and county random effects, so that within a county tract values differ through demographic composition.

The staged tract file has no intervals.

Explains Table 1, Table 3 and Table 4.

### How tracts are ranked, grouped and weighted

Deprivation is the Area Deprivation Index 2023 national percentile (1 least, 100 most deprived), and 741 tracts have no rank and are left out. Tenths hold 8,118 to 8,511 tracts (tied ranks fall in the same tenth) and twentieths fewer.

Means are weighted by the tract's adult population, including for measures that cover only some ages because the file has no population for them; counting each tract once gives similar differences.

Explains Table 1, Table 3 and Table 4.

### How provider supply is measured

Provider supply is from County Health Rankings 2024.

Residents per provider in a group of counties is the group's residents divided by its providers, so a county with no recorded provider adds residents and no providers; the county ratios that County Health Rankings publishes are not used.

Counties are ranked on the county Area Deprivation Index, the mean of the tract percentiles weighted by 2019-2023 population, with Connecticut tracts assigned to its eight legacy counties; 15 small counties have no rank. County percent rural is from the 2020 census.

Explains Table 7 and Table 8.

### How the map is shaded and what is left out

The map shades residents per mental health provider on a scale that doubles at each step and draws counties with no recorded provider in gray.

Counties are not suppressed anywhere in this chapter, and no tract or county estimate is set to zero.

Explains Table 8.

### How uncertainty is measured

Intervals for means, differences and ratios are percentile intervals from 2,000 draws of whole counties with replacement (seed 12345); differences and ratios share the draws of their two means.

Intervals for pooled ratios, county means and correlations come from the same draw scheme applied to whole states (seed 12345); correlations are reported weighted by county residents and with each county counted once.

### How the slopes and the coefficient zoo are fitted

Slopes are population-weighted least-squares fits with cluster-robust errors and t intervals, counties as clusters; fixed effects are absorbed by weighted demeaning and checked against a regression with dummy variables.

The coefficient zoo fits, for each state with at least 60 tracts that have every variable, the tract prevalence of a behavior on the natural log of median household income (ACS 2019-2023) with the shares of residents ages 65 and over, Black alone and Hispanic as controls, and reports the coefficient times the natural log of 2. States with fewer than five counties use HC3 errors.

Explains Table 2, Table 5 and Table 6.

### Four rankings of the same places

The comparison follows the book's ranking switch rule, fixed on October 28, 2026 before any of these gradients was computed. The main gradient of the chapter is recomputed under four rankings of the same places, with the outcome, period, weights, bins and intervals unchanged: the published ADI; a standardized index that applies the ADI's coefficients to its measures after standardizing each one across tracts; median household income; and the share of adults 25 and older with a bachelor's degree or more. The three alternatives come from ACS tract estimates for 2019 to 2023, the period of the ADI.

All four rankings use the same places, the 3,125 counties that have the outcome and all four rankings, so the ADI panel can differ slightly from the chapter's main figure. Tracts enter only if they have every input of the standardized index, which leaves out tracts without a median rent, home value or mortgage cost. A county takes the population-weighted mean of the ranks of its common-sample tracts, with Connecticut tracts assigned to its eight legacy counties, as for the county ADI.

A pattern counts as changed when the worse end switches sides, when the two ends differ under one ranking and not the other, or when the rank correlation of bin order and bin mean falls on the other side of 0.9 and moves by at least 0.1.

Explains Table 9.

## What these data cannot show

PLACES values are modeled estimates, and the tract file has no intervals, so the bands in this chapter describe the sampling of counties and not the model's own uncertainty. The model and the Area Deprivation Index draw on overlapping census inputs, which builds part of each gradient into the numbers, and the data cannot say how much. The values are crude prevalence among adults and are not adjusted for age. CDC advises against using them to rank the overall health of an area or to track change at the local level, which this chapter does not do.[9] The slopes and coefficients compare tracts and say nothing about the behavior of a person. Within a county the model gives tracts different values only through their demographic composition, so the comparison of the checkup and the dental visit within counties cannot separate behavior from composition at all.

The checkup measure is the share of adults who report a routine checkup in the past year, and the dental measure is the share who report a dental visit. Neither says what happened at the visit, whether the care was timely or whether it was affordable. Dental coverage is a separate matter from medical coverage, and the chapter has no measure of it.

Provider counts locate providers where they practice and divide by residents. They do not measure capacity, wait times or whether a provider accepts a patient's insurance, and county averages hide deserts inside counties and care that crosses county lines. The three provider counts come from different years, 2021 to 2023, and the PLACES estimates describe 2022 on an index built from 2019 to 2023 census data, so the sources do not describe one moment. Connecticut appears in PLACES under planning-region codes, which serve as its counties for the bootstrap, and in CHR under its eight legacy counties. 15 small counties have no deprivation rank and are left out of the tenths in the provider figures, and Puerto Rico is covered by neither PLACES nor the maps.

The association between supply and deprivation could run in either direction or through a third factor such as rurality, and the figures do not separate them. The studies in the box concern coverage and clinic capacity, and none of them estimates how a change in the supply of providers would alter the county gradients.

## References

1. Greenlund, Kurt J.; Lu, Hua; Wang, Yan; Matthews, Kevin A.; LeClercq, Jennifer M.; Lee, Benjamin; Carlson, Susan A. (2022). PLACES: Local Data for Better Health. Preventing Chronic Disease 19: 210459. https://doi.org/10.5888/pcd19.210459
2. Kind, Amy J. H.; Buckingham, William R. (2018). Making Neighborhood-Disadvantage Metrics Accessible: The Neighborhood Atlas. New England Journal of Medicine 378(26): 2456-2458. https://doi.org/10.1056/NEJMp1802313
3. Tudor Hart, Julian (1971). The Inverse Care Law. The Lancet 297(7696): 405-412. https://doi.org/10.1016/S0140-6736(71)92410-X
4. University of Wisconsin Population Health Institute (2024). County Health Rankings & Roadmaps Technical Documentation, Version 2. University of Wisconsin Population Health Institute (updated March 18, 2024). https://www.countyhealthrankings.org/sites/default/files/media/document/2024%20CHRR%20Technical%20Document_1.pdf
5. Finkelstein, Amy; Taubman, Sarah; Wright, Bill; Bernstein, Mira; Gruber, Jonathan; Newhouse, Joseph P.; Allen, Heidi; Baicker, Katherine; Oregon Health Study Group (2012). The Oregon Health Insurance Experiment: Evidence from the First Year. The Quarterly Journal of Economics 127(3): 1057-1106. https://doi.org/10.1093/qje/qjs020
6. Baicker, Katherine; Taubman, Sarah L.; Allen, Heidi L.; Bernstein, Mira; Gruber, Jonathan H.; Newhouse, Joseph P.; Schneider, Eric C.; Wright, Bill J.; Zaslavsky, Alan M.; Finkelstein, Amy N. (2013). The Oregon Experiment: Effects of Medicaid on Clinical Outcomes. New England Journal of Medicine 368(18): 1713-1722. https://doi.org/10.1056/NEJMsa1212321
7. Miller, Sarah; Johnson, Norman; Wherry, Laura R. (2021). Medicaid and Mortality: New Evidence from Linked Survey and Administrative Data. The Quarterly Journal of Economics 136(3): 1783-1829. https://doi.org/10.1093/qje/qjab004
8. Bailey, Martha J.; Goodman-Bacon, Andrew (2015). The War on Poverty's Experiment in Public Medicine: Community Health Centers and the Mortality of Older Americans. American Economic Review 105(3): 1067-1104. https://doi.org/10.1257/aer.20120070
9. Centers for Disease Control and Prevention (2024). PLACES Frequently Asked Questions: Using the Data. CDC PLACES website, page last reviewed August 29, 2024. https://www.cdc.gov/places/faqs/using-data/index.html

## Provenance

- Pipeline version: 0.1.0
- Build script: book2/chapters/health-behaviors-and-the-supply-of-care/build.py (sha256 4dc2e47fd80746fd69982d9c8af4e068cf34d6389e02b624fecc581717ec6e6b)
- Built on: 2026-10-07
- Site commit: c2b1ff3309cb4a10833f884da91ac0057a66ad0a

Inputs:

| File | Size (bytes) | SHA-256, first 1 MB | SHA-256, last 1 MB |
| --- | ---: | --- | --- |
| book2/chapters/health-behaviors-and-the-supply-of-care/gradients.py | 11,771 | 1fe448a81ab7fbb49ae49ade9c68d93a5617e95e103be3e322c47d29884ef32f | 1fe448a81ab7fbb49ae49ade9c68d93a5617e95e103be3e322c47d29884ef32f |
| book2/chapters/health-behaviors-and-the-supply-of-care/supply.py | 7,234 | 315c7b776f033f7288149ae1f98c8d58b68a2483b02ad53a7d885db567199566 | 315c7b776f033f7288149ae1f98c8d58b68a2483b02ad53a7d885db567199566 |
| book2/chapters/health-behaviors-and-the-supply-of-care/zoo.py | 430 | 447ef1b703ca8e0d04d14a482e62396d4a42a497ca22ce22e92280bcb5f0bb3a | 447ef1b703ca8e0d04d14a482e62396d4a42a497ca22ce22e92280bcb5f0bb3a |
| data/analytic/tract_atlas.parquet | 39,659,925 | 29d5e4d943552d54aeb6ba92aadc939d9bd15bdea553c78357dbd516757be627 | 1e10d1ecf722340f23ee2212d4999a90250f85b65fa89b39e32db06d9472e22e |
| data/geo/dim_geo.parquet | 1,180,921 | 29d5e4d943552d54aeb6ba92aadc939d9bd15bdea553c78357dbd516757be627 | e58c8c488df54ab3187f5461f179322afdc49c9bd2315404003fe1272eea32c3 |
| data/raw/acs/deprivation_components_2019_2023/tract.parquet | 13,197,793 | 5545cde2cbe96a8b56e468d65388248a241aa6bd300d3c9e325d08d3f86b11da | 5c8ec454d066988b0f1a01aaa6ed34ebfc05d162e01cf9c6bb01f9198cbaff12 |
| data/staged/adi/adi_atlas_tract2020_2023.parquet | 798,078 | 5e409bc212ff871def780fa27b4cf8b616d9da420b2b9f6455517eddcd8c1a95 | 5e409bc212ff871def780fa27b4cf8b616d9da420b2b9f6455517eddcd8c1a95 |
| data/staged/crosswalks/ct_tract2020_to_old_county.parquet | 15,842 | 961ef39f31596ef65fca2c67d5f2f63e533c5dc9b49745d3bb6437844e4b831d | 961ef39f31596ef65fca2c67d5f2f63e533c5dc9b49745d3bb6437844e4b831d |
| data/staged/health/chr_county_2024.parquet | 8,776,933 | 619795f7ecfec8aff10694b649eaee5e421dbeffbe4be47af4161b51ceefa5cd | c69357f1bcf630407bb9716cb445618838999558c019aaaf77d4681b6bf45c56 |
| data/staged/health/places_tract2020_latest.parquet | 5,919,159 | af4c3b0d1205b7d84b81676a9529ef73323b2dbeb9d54e235f5efeebfad83177 | a2f19029ee2ef032be87e0d5288d88d489afd52379279833f178baa021df3dbf |
| data/staged/ses/acs5_tract2020_2019_2023.parquet | 7,232,424 | 5fe6f730a35225f14d5d8abfc6b9bce58b52e545f309a7eac74f1dae74d7a31c | 7b4a2475b3852d3f91804716a04566ef8ceadacec2be9fb85f5eb642078efac6 |
| data/staged/svi/svi_tract2020_2022.parquet | 17,740,415 | cb3860d8edc2aaf1f2a02211a58f7a0b80457eb6ad9a37b5ec8abb2d5da9da10 | e371a4a696d5bed05d54ad7c5b6ca2504508805b847ec415c2322be7f66ec2b2 |

Outputs:

| File | SHA-256 |
| --- | --- |
| facts.json | cb3c3e891db62b647d899ac6fb4e094e6cfb9eba07c7713713a4648aa23d0e0f |
| figures/alt_rankings.json | 0e6a524e2d02f3d85fc9200381f43c0c1d9992221f13382a6b8759f7451f7cba |
| figures/care_ledger.json | 9264b088611727b7684db2e297b63c2a18b15f90d97472765d0266ebf7ca579d |
| figures/checkup_slopes_by_state.json | 0f2764a46daa079fb2098294ce69267ea7fc5f0365fc27d1459b59a75e04d2a8 |
| figures/checkup_specifications.json | 68902599b95a1f8f5dc1c994eac81a71a5ad566f8221b4b402c6fe25e0863cd6 |
| figures/checkup_vs_dental.json | b16bc37d663753dd00b0c0baf5b2292305c6844a6efaf1b3a8311a90b056345f |
| figures/coefficient_zoo.json | 597a9227664276b82f7078ef41ed4c63c1e039c4a9c1d3355a2ce8d515269bfd |
| figures/mental_health_provider_map.json | 163018ff6047d0171e658516aebfa3e41b52300b0148fc2995948a1fb6c945cb |
| figures/people_per_provider.json | afc13ef7ae3be5189fe198b6a6f18723b761935c3e00a6a816c76d7fbc30dbb5 |
| figures/twelve_outcome_gradients.json | 4b7354f894718ec4f0e4ba1957343e755e6d8d2cbddbf609292392f8c050798d |
| meta.json | 9ffc14279db5ea06a06de8e2a5fd759c65f7bb837ad43072646de38ec54c01f8 |
| public/data/health-behaviors-and-the-supply-of-care/mental_health_provider_map.json | 7f9714b1991ea8122cdb3e5663400ce3287d322f3fe3097f6cba257d65340c16 |
| refs.json | 24a7bbefb95f1ddf599fef65b29c35bfa0bdcd416815464da5e3130cd0ff49e8 |

## Machine formats

- Report (HTML): https://vishalsingh.org/zipcode/health-behaviors-and-the-supply-of-care/report
- Report (Markdown): https://vishalsingh.org/zipcode/health-behaviors-and-the-supply-of-care/report.md
- Every estimate (JSON): https://vishalsingh.org/zipcode/report/health-behaviors-and-the-supply-of-care/facts.json
- Table 1, figure "twelve_outcome_gradients" (JSON): https://vishalsingh.org/zipcode/report/health-behaviors-and-the-supply-of-care/figures/twelve_outcome_gradients.json
- Table 2, figure "coefficient_zoo" (JSON): https://vishalsingh.org/zipcode/report/health-behaviors-and-the-supply-of-care/figures/coefficient_zoo.json
- Table 3, figure "checkup_vs_dental" (JSON): https://vishalsingh.org/zipcode/report/health-behaviors-and-the-supply-of-care/figures/checkup_vs_dental.json
- Table 4, figure "care_ledger" (JSON): https://vishalsingh.org/zipcode/report/health-behaviors-and-the-supply-of-care/figures/care_ledger.json
- Table 5, figure "checkup_specifications" (JSON): https://vishalsingh.org/zipcode/report/health-behaviors-and-the-supply-of-care/figures/checkup_specifications.json
- Table 6, figure "checkup_slopes_by_state" (JSON): https://vishalsingh.org/zipcode/report/health-behaviors-and-the-supply-of-care/figures/checkup_slopes_by_state.json
- Table 7, figure "people_per_provider" (JSON): https://vishalsingh.org/zipcode/report/health-behaviors-and-the-supply-of-care/figures/people_per_provider.json
- Table 8, figure "mental_health_provider_map" (JSON): https://vishalsingh.org/zipcode/report/health-behaviors-and-the-supply-of-care/figures/mental_health_provider_map.json
- Table 9, figure "alt_rankings" (JSON): https://vishalsingh.org/zipcode/report/health-behaviors-and-the-supply-of-care/figures/alt_rankings.json
- Payload of "mental_health_provider_map" (JSON): https://vishalsingh.org/zipcode/data/health-behaviors-and-the-supply-of-care/mental_health_provider_map.json
- Index of every report: https://vishalsingh.org/zipcode/llms.txt; all reports in one file: https://vishalsingh.org/zipcode/llms-full.txt

---

# Health behaviors, person by person: report

Chapter 7.2 of ZIP Code Destiny, by Vishal Singh (NYU Stern). Generated from the chapter's data files; the reading version is at https://vishalsingh.org/zipcode/health-behaviors-person-by-person. Table n of this report is figure n of the chapter.

## Identification

- Title: Health behaviors, person by person
- Label: The rich person's vice
- Chapter: 7.2 (Part VII, Health and care; An American life)
- Question: Do the neighborhood gradients in smoking and drinking hold for individuals once income, race and age are measured person by person?
- Dek: Among adults, the share who binged on alcohol in the past month was 13.3% in the lowest of five household income bands and 20.4% in the highest, while smoking, obesity and physical inactivity were each less common in the highest band.
- Grain: individuals
- Status: published
- Published: 2026-10-09
- Updated: not revised
- Data sources:
  - BRFSS. CDC Behavioral Risk Factor Surveillance System public-use files, 2014-2024, adults in the 50 states and the District of Columbia; binge drinking asked from 2015; years: 2014-2024; weights: BRFSS final weight with stratum and primary sampling unit (design-based estimates)
- Canonical URL: https://vishalsingh.org/zipcode/health-behaviors-person-by-person
- Report: https://vishalsingh.org/zipcode/health-behaviors-person-by-person/report
- Markdown: https://vishalsingh.org/zipcode/health-behaviors-person-by-person/report.md
- BibTeX key: singh2026healthbehaviorspersonbyperson

```bibtex
@misc{singh2026healthbehaviorspersonbyperson,
  author = {Singh, Vishal},
  title = {Health behaviors, person by person},
  year = {2026},
  url = {https://vishalsingh.org/zipcode/health-behaviors-person-by-person},
  note = {In ZIP Code Destiny. Published 9 October 2026.}
}
```

## Principal estimates

The facts the dek and the opening prose cite, in order of appearance.

| Estimand | Population | Period | Value | Interval | n | Weights | Source | Fact id |
| --- | --- | --- | ---: | --- | ---: | --- | --- | --- |
| design-based weighted share of respondents, Taylor-linearized 95% interval | adults in the 50 states and the District of Columbia in the lowest of five household income bands: binge drinking in the past 30 days | BRFSS 2015-2024 | 13.3% | 95% CI 13.0% to 13.5% | 450,791 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS | binge_poor |
| design-based weighted share of respondents, Taylor-linearized 95% interval | adults in the 50 states and the District of Columbia in the highest of five household income bands: binge drinking in the past 30 days | BRFSS 2015-2024 | 20.4% | 95% CI 20.3% to 20.6% | 1,254,573 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS | binge_rich |
| design-based weighted share of respondents, Taylor-linearized 95% interval | adults in the 50 states and the District of Columbia in the highest of five household income bands: current cigarette smoking | BRFSS 2014-2024 | 8.5% | 95% CI 8.4% to 8.6% | 1,392,590 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS | smoke_rich |
| design-based weighted share of respondents, Taylor-linearized 95% interval | adults in the 50 states and the District of Columbia in the highest of five household income bands: obesity (BMI 30 or more, self-reported height and weight) | BRFSS 2014-2024 | 28.4% | 95% CI 28.3% to 28.6% | 1,365,694 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS | obese_rich |
| design-based weighted share of respondents, Taylor-linearized 95% interval | adults in the 50 states and the District of Columbia in the highest of five household income bands: no leisure-time physical activity in the past 30 days | BRFSS 2014-2024 | 13.6% | 95% CI 13.5% to 13.7% | 1,411,844 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS | inactive_rich |
| design-based difference between two weighted shares of respondents, Taylor-linearized 95% interval | White non-Hispanic adults in the 50 states and the District of Columbia: binge drinking in the past 30 days | BRFSS 2015-2024 | 8.6 | 95% CI 8.3 to 8.9 | 1,319,140 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS | race_gap_white |
| design-based difference between two weighted shares of respondents, Taylor-linearized 95% interval | Hispanic adults in the 50 states and the District of Columbia: binge drinking in the past 30 days | BRFSS 2015-2024 | 10.3 | 95% CI 9.3 to 11.2 | 119,344 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS | race_gap_hisp |
| design-based difference between two weighted shares of respondents, Taylor-linearized 95% interval | Black non-Hispanic adults in the 50 states and the District of Columbia: binge drinking in the past 30 days | BRFSS 2015-2024 | 0.5 | 95% CI −0.3 to 1.2 | 120,974 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS | race_gap_black |
| definition: the smallest number of respondents behind any published cell | the smallest number of respondents behind any published cell | 2014-2024 | 30 |   |   | none (definition) | BRFSS | cell_floor |

## Findings

The share of adults who binged on alcohol in the past month was 13.3% (95% CI 13.0% to 13.5%) in the lowest of five household income bands and 20.4% (95% CI 20.3% to 20.6%) in the highest. Current smoking, obesity and physical inactivity were each lower in the highest band, at 8.5%, 28.4% and 13.6%. The estimates come from the Behavioral Risk Factor Surveillance System (BRFSS), a telephone survey that CDC runs with state health departments, pooled over survey years from 2014 to 2024.[1] The rise in binge drinking is an average over groups that differ. The difference between the highest band and the lowest is 8.6 points among non-Hispanic White adults, 10.3 among Hispanic adults and 0.5 (95% CI −0.3 to 1.2) among non-Hispanic Black adults.

 (Note: The BRFSS files also hold interviews in Guam, Puerto Rico and the Virgin Islands. They are outside every estimate in this chapter, which describes adults in the states and the District of Columbia.)

Chapter 7.1, Health behaviors and the supply of care (https://vishalsingh.org/zipcode/health-behaviors-and-the-supply-of-care) follows the same behaviors across census tracts with modeled estimates, and a tract value cannot show whether the adults who drink in a tract are the ones with high incomes (ecological inference). A survey that asks each adult about drinking and household income can. The cost is place, because BRFSS locates a respondent by state and, for some respondents, by metro status, and nowhere finer. Every estimate below is a share of respondents weighted with the survey's survey weights, and every 95% interval uses the survey's strata and sampling units. No cell rests on fewer than 30 respondents, and no respondent-level row is published.

### Binge drinking rises with household income while smoking, obesity and inactivity fall

A binge is 5 or more drinks on one occasion for a man, or 4 or more for a woman, at least once in the past 30 days.[2] The question is in the survey from 2015, so binge estimates cover 10 years and the other measures cover 11.

The pooled file holds 4,790,742 interviews with adults in the states and the District of Columbia. The binge drinking estimates by income band rest on 3,336,677 of them, because the question was not asked in 2014 and some adults gave no income.

Across the 5 income bands the share of adults who binged rises at every step, from 13.3% (95% CI 13.0% to 13.5%) in the lowest band to 20.4% (95% CI 20.3% to 20.6%) in the highest, a difference of 7.2 (95% CI 6.9 to 7.5) points (Table 1).

The other three behaviors fall at every step. Current smokers, adults who have smoked at least 100 cigarettes in their lives and now smoke every day or some days,[1] are 25.7% (95% CI 25.4% to 25.9%) of adults in the lowest band and 8.5% (95% CI 8.4% to 8.6%) in the highest, a difference of 17.2 (95% CI 16.9 to 17.4) points.

Obesity, a body mass index of 30 or more from self-reported height and weight, goes from 36.0% to 28.4%, a difference of 7.6 (95% CI 7.2 to 7.9) points. The share who did no leisure-time physical activity in the past month goes from 38.2% to 13.6%, a difference of 24.6 (95% CI 24.3 to 24.9) points.

The lowest band is household income under $20,000 a year and the highest is $75,000 or more. BRFSS recorded income in eight categories through 2020 and in eleven from 2021, which split the top category, so the five bands merge the categories that did not change and keep the highest band whole across the break. The bands are in nominal dollars, and the highest band held 30.2% of the respondents who gave an income in 2014 and 46.1% in 2024.

Part of that shift is a step at the 2021 change in the question, when the lowest band went from 15.8% of respondents who gave an income in 2020 to 11.8% in 2021, a larger one-year drop than in any earlier year. A further 19.4% (95% CI 19.3% to 19.5%) of adults did not know their household income or declined to give it, and they are not in any band.

Table 1. Binge drinking rose from 13.3% to 20.4% of adults across the five income bands while smoking, obesity and inactivity fell

For 2021 to 2024 the highest band can be divided into four, and the second view of Table 1 shows them. Across those four bands binge drinking rises from 18.2% (95% CI 17.8% to 18.6%) to 22.1% (95% CI 21.5% to 22.6%) and smoking falls from 9.7% to 4.3%. Obesity falls most between the two highest bands, from 30.1% to 23.2%.

A rise in drinking with income appears in surveys from other settings. A review of studies of socioeconomic status and alcohol concluded that people of higher status drink similar or greater amounts, while people of lower status bear more of the consequences.[3] In a national survey of 2001 and 2002, the prevalence of alcohol abuse by the DSM-IV definition rose with income while alcohol dependence fell with it. The abuse result came from the hazardous use criterion, mostly driving after drinking.[4] BRFSS measures the quantity drunk on one occasion and does not measure consequences, so it cannot say who bears the harm.

### The rise holds within age, education and state, not among non-Hispanic Black adults

Adults in different income bands differ in age, schooling and race, so a rise across pooled bands could come from who is in the bands. Table 2 repeats the comparison inside groups. In each of four age groups the rate is higher in the highest band than in the lowest, by 10.2 (95% CI 9.4 to 11.0) points at ages 18 to 29 and by 2.6 (95% CI 2.3 to 2.9) points at 65 and older. In the oldest group the rate goes from 4.4% to 6.9%. Within each education level the rate also rises with income, by 12.7 (95% CI 11.1 to 14.2) points among adults who did not finish high school and 3.5 (95% CI 2.9 to 4.1) among college graduates.

Table 2. The rise in binge drinking with income holds within age groups and education levels and differs by race and ethnicity

The education lines cross. In the lowest band, college graduates binge more often than adults who did not finish high school (15.2% against 11.7%), and in the highest band less often (18.7% against 24.4%). At a given income, then, schooling does not move binge drinking in one direction, and a single measure of socioeconomic status that combined income and schooling would hide the crossing.

The race and ethnicity tab shows a rise that is not shared by every group. Among White non-Hispanic adults the rate goes from 13.0% to 21.7%, a difference of 8.6 (95% CI 8.3 to 8.9) points, and among Hispanic adults the difference is 10.3 (95% CI 9.3 to 11.2) points. Among multiracial non-Hispanic adults it is 5.5 (95% CI 3.7 to 7.3) points.

Among Black non-Hispanic adults the rate is 13.3% in the lowest band and 13.8% in the highest, a difference of 0.5 (95% CI −0.3 to 1.2) points whose interval includes zero. Among adults of another race, non-Hispanic, the difference is 1.0 (95% CI 0.0 to 2.1) points.

The pooled difference of 7.2 points therefore averages groups in which the rise is large with groups in which it is absent or small, so the pattern behind this chapter's label belongs to White, Hispanic and multiracial adults. The survey cannot say why the groups differ. The review cited above lists race and ethnicity among the factors that moderate the link between socioeconomic status and alcohol use.[3]

The rise is also not the product of a few states. Comparing the highest band with the lowest within each state, the difference in binge drinking is above zero in 51 of the 51 jurisdictions, the states and the District of Columbia, and its interval is above zero in 51 (Table 3). It runs from 2.7 (95% CI 1.6 to 3.7) points in Utah to 13.1 (95% CI 10.7 to 15.5) points in the District of Columbia.

The smoking difference is below zero, with an interval below zero, in 51 as well, from 9.1 (95% CI 8.2 to 9.9) points lower in California to 29.9 (95% CI 27.2 to 32.6) points lower in Alaska. In the tract regressions of Chapter 7.1, Health behaviors and the supply of care (https://vishalsingh.org/zipcode/health-behaviors-and-the-supply-of-care), Maryland is the one state with a negative estimate for binge drinking, and its interval includes zero. Among Maryland's respondents the difference is 5.7 (95% CI 4.4 to 6.9) points.

Table 3. Binge drinking is higher in the highest income band than in the lowest in all 51 jurisdictions, and smoking is lower in all 51

The difference is positive in each of the 10 survey years from 2015 to 2024, with an interval above zero in 10 of them. The yearly differences run from 6.7 (95% CI 5.8 to 7.7) points in 2024 to 8.3 (95% CI 7.3 to 9.2) points in 2022, and those two intervals overlap. The difference is also similar on either side of the 2021 change in the income question, at 7.3 (95% CI 6.9 to 7.6) points for 2015 to 2020 and 7.5 (95% CI 7.0 to 8.0) points for 2021 to 2024.

### Men binge drink more than women in every income band, by about the same number of points

Men's binge drinking rate rises from 18.3% in the lowest band to 24.3% in the highest, and women's rises from 9.7% to 15.8% (Table 4).

The difference between men and women is between 8.5 (95% CI 8.2 to 8.8) and 9.2 (95% CI 8.7 to 9.6) points in every band, so the lines stay about the same distance apart. As both rates rise by a similar number of points, the ratio of men's rate to women's falls from 1.9 (95% CI 1.8 to 2.0) in the lowest band to 1.5 (95% CI 1.5 to 1.6) in the highest, and stays above one.

Table 4. Men binge drink more than women in every income band, by 8.5 to 9.2 points

Obesity shows a different pattern by sex. Women are more often obese than men in the lowest band, by 8.6 (95% CI 8.0 to 9.2) points (39.7% against 31.1%). They are less often obese in the highest band, by 5.6 (95% CI 5.2 to 5.9) points (25.3% against 30.9%), as women's rate falls across the bands and men's does not. Smoking falls with income for both sexes and the difference narrows, from 6.7 (95% CI 6.2 to 7.2) points in the lowest band to 2.9 (95% CI 2.7 to 3.1) in the highest.

### The slope with income differs by behavior and, for obesity, by sex

Table 5 sets the rate in the highest band against the rate in the lowest for six measures, separately for women and men. Binge drinking is the only measure that is higher in the highest band, and it is higher by 6.0 (95% CI 5.6 to 6.5) points for men and 6.2 (95% CI 5.8 to 6.5) for women, a difference between the sexes of 0.1 (95% CI −0.4 to 0.7) points whose interval includes zero.

For obesity the difference is −14.4 (95% CI −14.8 to −13.9) points for women and −0.2 (95% CI −0.7 to 0.3) for men, and the men's interval includes zero. The women's difference is 14.2 (95% CI 13.5 to 14.8) points steeper than the men's.

Smoking is steeper for men, with a difference of −19.7 (95% CI −20.1 to −19.3) points against −15.9 (95% CI −16.2 to −15.6) for women, which is 3.8 (95% CI 3.2 to 4.3) points steeper. For fair or poor health the two differences are close, at −29.7 for men and −30.6 for women.

Table 5. The obesity difference between the highest and the lowest income band is −14.4 points for women and −0.2 for men

Reviews of the literature report an inverse association between socioeconomic status and obesity in high-income countries,[5][6] and the earlier of the two found it most clearly among women.[5] The survey shows the same sex difference for income in the United States over 2014 to 2024. The obesity rate of men is nearly the same in the lowest and highest bands, and the rate of women is not.

### Smoking separates by education and, between the sexes, by race and ethnicity

The smoking rate is highest among adults with the least schooling and the lowest income, and the table of rates by schooling and income shows a row in which the rate falls least with income (Table 6). Among adults who did not finish high school the rate is 28.0% in the lowest band and 23.2% in the highest, a difference of 4.8 (95% CI 3.4 to 6.2) points, and it does not fall at every step.

Among college graduates it goes from 13.6% to 4.4%, a difference of 9.1 (95% CI 8.7 to 9.6) points. Across all incomes the rate is 24.7% among adults who did not finish high school and 5.8% among college graduates.

Table 6. Smoking is 28.0% among adults who did not finish high school in the lowest income band and 4.4% among college graduates in the highest

Men smoke more than women, 16.5% of men against 12.8% of women, but the difference depends on race and ethnicity (Table 7). It is 1.7 (95% CI 1.6 to 1.9) points among White non-Hispanic adults and 8.1 (95% CI 7.7 to 8.5) among Hispanic adults, with Black non-Hispanic adults at 7.0 (95% CI 6.5 to 7.4), adults of another race, non-Hispanic, at 5.9 (95% CI 5.4 to 6.4) and multiracial non-Hispanic adults at 3.3 (95% CI 2.2 to 4.4).

Across the five groups, men's rates run from 13.5% (95% CI 13.1% to 14.0%) to 21.9% (95% CI 21.2% to 22.7%) and women's from 7.6% (95% CI 7.4% to 7.9%) to 18.6% (95% CI 17.9% to 19.4%), with multiracial non-Hispanic adults highest for both sexes. Women's rates differ more across groups than men's, so the size of the difference follows the women's rates.

White women smoke at 14.5% and Hispanic women at 7.6%, against 16.3% and 15.8% for men. Men's rate is above women's in all 25 cells of race and ethnicity by income band, with an interval above zero in 25.

Table 7. Men smoke more than women by 1.7 points among non-Hispanic White adults and 8.1 among Hispanic adults

Cross-national work describes the difference between the sexes in smoking as a stage in the spread of cigarettes, which reach men before women.[7][8] The pattern across groups here fits that account, but one survey period cannot show a diffusion over time.

### Evidence on whether income changes drinking and smoking

> Causal evidence: Lottery winnings, smoking and drinking (Natural experiment: size of lottery winnings among winners)
>
> The comparisons above cannot say whether more income changes what a person drinks or smokes. Apouey and Clark studied this with a British household panel in which respondents reported winnings from the national lottery, football pools and other gambling. Winners differ from non-winners in who plays, so they compared larger winners with smaller ones, on the argument that the amount won is largely outside the player's control.[9] Larger winnings were associated with more smoking and with more frequent social drinking, meaning going out for a drink at a pub or club. The paper reports a positive effect of winnings on mental health and no significant effect on self-assessed general health.[9]
>
> The study covers gambling windfalls among British adults in the 1990s and 2000s. Its drinking outcome is how often a person goes out for a drink, not binge drinking, and it does not estimate what a lasting difference in household income of the size between the BRFSS bands would do in the United States. The BRFSS comparisons in this chapter are descriptive.

### What these data cannot show

Every measure in this chapter is reported by the respondent in a telephone interview, so the estimates describe what adults say. Respondents may under-report their drinking and their weight, and the data cannot show whether that differs by income, sex or race. The 2024 survey year is provisional. The intervals reflect the survey's sampling design and weights. They do not cover the adults who could not be reached or declined to take part, or the 19.4% of adults who gave no household income and are left out of every band.

The data are a repeated cross-section, not a panel. A person in the highest band and a person in the lowest are different people, so the gradients say how behavior differs across incomes and not how it changes when one person's income changes. The bands are household income in nominal dollars, not adjusted for household size or for prices. The highest band holds 46.1% of respondents who gave an income in 2024 against 30.2% in 2014, and part of that shift is a step at the 2021 change in the income question, so pooling the years mixes survey years with different bands. The binge drinking difference between the highest band and the lowest is similar before and after that change. Income, age, schooling and race are related. The age, education and race and ethnicity panels compare within one of them at a time and only for binge drinking, and the other figures are not age-adjusted.

The groups are those BRFSS defines. The Hispanic group may be of any race, and the other-race group combines populations whose drinking and smoking differ. Binge drinking is a measure of quantity on one occasion, and a person who binges once in a month is counted with a person who binges every week. BRFSS locates respondents by state and, for some respondents, by metro status, which is missing for a large share of respondents and is not used here, so the results say nothing about neighborhoods. The person-level results here and the tract-level results of Chapter 7.1, Health behaviors and the supply of care (https://vishalsingh.org/zipcode/health-behaviors-and-the-supply-of-care) come from different measures, survey responses in one and modeled estimates in the other, so neither confirms the other. Whether the drinking measured here relates to the alcohol-related deaths mapped in Chapter 8.2, Overdose, alcohol and suicide (https://vishalsingh.org/zipcode/overdose-alcohol-and-suicide) is outside what BRFSS can show.

## Tables

### Table 1. Binge drinking rose from 13.3% to 20.4% of adults across the five income bands while smoking, obesity and inactivity fell

Percent of adults in the states and DC with each behavior, by household income band, BRFSS 2015 to 2024 for binge drinking and 2014 to 2024 for the other three, with 95% confidence intervals

| Behavior | Income band | Percent of adults | Lower limit of the 95% interval | Upper limit of the 95% interval | Respondents |
| --- | --- | ---: | ---: | ---: | ---: |
| Binge drinking | Under $20K | 13.3 | 13.0 | 13.5 | 450,791 |
| Binge drinking | $20K to $35K | 14.6 | 14.4 | 14.8 | 614,141 |
| Binge drinking | $35K to $50K | 16.2 | 15.9 | 16.4 | 464,109 |
| Binge drinking | $50K to $75K | 17.6 | 17.4 | 17.9 | 553,063 |
| Binge drinking | $75K or more | 20.4 | 20.3 | 20.6 | 1,254,573 |
| Binge drinking | Under $20K | 12.5 | 12.0 | 12.9 | 121,580 |
| Binge drinking | $20K to $35K | 13.8 | 13.4 | 14.1 | 222,223 |
| Binge drinking | $35K to $50K | 14.8 | 14.4 | 15.2 | 176,553 |
| Binge drinking | $50K to $75K | 16.3 | 16.0 | 16.6 | 221,026 |
| Binge drinking | $75K to $100K | 18.2 | 17.8 | 18.6 | 184,694 |
| Binge drinking | $100K to $150K | 19.7 | 19.4 | 20.1 | 193,860 |
| Binge drinking | $150K to $200K | 21.3 | 20.7 | 21.8 | 87,864 |
| Binge drinking | $200K or more | 22.1 | 21.5 | 22.6 | 93,005 |
| Current smoking | Under $20K | 25.7 | 25.4 | 25.9 | 531,816 |
| Current smoking | $20K to $35K | 20.1 | 19.9 | 20.3 | 707,394 |
| Current smoking | $35K to $50K | 16.5 | 16.2 | 16.7 | 529,537 |
| Current smoking | $50K to $75K | 13.6 | 13.4 | 13.8 | 624,088 |
| Current smoking | $75K or more | 8.5 | 8.4 | 8.6 | 1,392,590 |
| Current smoking | Under $20K | 23.4 | 22.9 | 23.9 | 126,031 |
| Current smoking | $20K to $35K | 18.1 | 17.8 | 18.5 | 229,441 |
| Current smoking | $35K to $50K | 14.6 | 14.3 | 15.0 | 181,626 |
| Current smoking | $50K to $75K | 12.1 | 11.8 | 12.4 | 226,612 |
| Current smoking | $75K to $100K | 9.7 | 9.4 | 10.0 | 189,175 |
| Current smoking | $100K to $150K | 7.8 | 7.6 | 8.1 | 198,439 |
| Current smoking | $150K to $200K | 6.1 | 5.8 | 6.4 | 89,948 |
| Current smoking | $200K or more | 4.3 | 4.0 | 4.6 | 95,390 |
| Obesity | Under $20K | 36.0 | 35.7 | 36.3 | 512,939 |
| Obesity | $20K to $35K | 34.5 | 34.2 | 34.7 | 687,969 |
| Obesity | $35K to $50K | 33.4 | 33.1 | 33.7 | 516,714 |
| Obesity | $50K to $75K | 33.0 | 32.7 | 33.2 | 609,779 |
| Obesity | $75K or more | 28.4 | 28.3 | 28.6 | 1,365,694 |
| Obesity | Under $20K | 37.9 | 37.3 | 38.5 | 121,701 |
| Obesity | $20K to $35K | 36.8 | 36.4 | 37.3 | 223,523 |
| Obesity | $35K to $50K | 35.7 | 35.2 | 36.2 | 177,690 |
| Obesity | $50K to $75K | 35.4 | 35.0 | 35.8 | 222,115 |
| Obesity | $75K to $100K | 34.0 | 33.5 | 34.5 | 185,249 |
| Obesity | $100K to $150K | 33.4 | 32.9 | 33.8 | 194,989 |
| Obesity | $150K to $200K | 30.1 | 29.5 | 30.8 | 88,636 |
| Obesity | $200K or more | 23.2 | 22.7 | 23.8 | 94,433 |
| No leisure-time physical activity | Under $20K | 38.2 | 37.9 | 38.5 | 537,736 |
| No leisure-time physical activity | $20K to $35K | 32.2 | 32.0 | 32.5 | 718,497 |
| No leisure-time physical activity | $35K to $50K | 26.1 | 25.8 | 26.3 | 536,832 |
| No leisure-time physical activity | $50K to $75K | 21.1 | 20.8 | 21.3 | 632,292 |
| No leisure-time physical activity | $75K or more | 13.6 | 13.5 | 13.7 | 1,411,844 |
| No leisure-time physical activity | Under $20K | 39.4 | 38.8 | 40.0 | 131,164 |
| No leisure-time physical activity | $20K to $35K | 33.4 | 33.0 | 33.8 | 238,962 |
| No leisure-time physical activity | $35K to $50K | 27.2 | 26.7 | 27.6 | 188,133 |
| No leisure-time physical activity | $50K to $75K | 21.3 | 21.0 | 21.7 | 233,764 |
| No leisure-time physical activity | $75K to $100K | 17.5 | 17.1 | 17.9 | 194,948 |
| No leisure-time physical activity | $100K to $150K | 13.5 | 13.2 | 13.9 | 204,250 |
| No leisure-time physical activity | $150K to $200K | 10.9 | 10.5 | 11.4 | 92,704 |
| No leisure-time physical activity | $200K or more | 8.1 | 7.7 | 8.5 | 98,299 |

Views: Five income bands, 2014 to 2024 (view = five); Eight income bands, 2021 to 2024 (view = eight).

Note: Respondents who did not know their household income or declined to give it are left out of the bands, 19.4% of adults. The five bands group the income categories that BRFSS used in every year. The categories are in nominal dollars, so the highest band holds 30.2% of respondents who gave an income in 2014 and 46.1% in 2024, and part of that shift is a step at the 2021 change in the income question.

Note: Intervals are 95% confidence intervals from the survey's stratum and primary sampling unit (design-based, Taylor linearization). No cell rests on fewer than 30 respondents. The survey is a repeated cross-section of adults in the states and the District of Columbia; Guam, Puerto Rico and the Virgin Islands are not included.

Note: Every measure is self-reported by telephone. Estimates describe the people who answered, and BRFSS geography stops at the state, so they say nothing about tracts.

Source: BRFSS 2014-2024 public-use files (binge drinking from 2015), weighted (BRFSS final weight) with design-based intervals

Data: https://vishalsingh.org/zipcode/report/health-behaviors-person-by-person/figures/behaviors_by_income.json

### Table 2. The rise in binge drinking with income holds within age groups and education levels and differs by race and ethnicity

Percent of adults who binged in the past month by household income band within age group, education level, and race and ethnicity, BRFSS 2015 to 2024

#### Table 2.1. By age group

In each age group binge drinking is higher in the highest income band than in the lowest, by 10.2 points at ages 18 to 29 and 65 and older

Percent of adults who binged in the past month by household income band, four age groups, BRFSS 2015 to 2024

| Age group | Income band | Percent of adults | Lower limit of the 95% interval | Upper limit of the 95% interval | Respondents |
| --- | --- | ---: | ---: | ---: | ---: |
| 18 to 29 | Under $20K | 22.3 | 21.7 | 23.0 | 53,527 |
| 18 to 29 | $20K to $35K | 24.7 | 24.1 | 25.2 | 75,758 |
| 18 to 29 | $35K to $50K | 26.6 | 26.0 | 27.3 | 55,347 |
| 18 to 29 | $50K to $75K | 28.5 | 27.9 | 29.2 | 58,025 |
| 18 to 29 | $75K or more | 32.5 | 32.0 | 33.0 | 98,374 |
| 30 to 44 | Under $20K | 15.7 | 15.2 | 16.2 | 68,510 |
| 30 to 44 | $20K to $35K | 18.4 | 17.9 | 18.9 | 93,637 |
| 30 to 44 | $35K to $50K | 21.0 | 20.5 | 21.6 | 74,973 |
| 30 to 44 | $50K to $75K | 23.2 | 22.7 | 23.7 | 102,182 |
| 30 to 44 | $75K or more | 25.9 | 25.6 | 26.2 | 294,181 |
| 45 to 64 | Under $20K | 11.3 | 11.0 | 11.7 | 162,187 |
| 45 to 64 | $20K to $35K | 12.6 | 12.2 | 13.0 | 170,206 |
| 45 to 64 | $35K to $50K | 14.1 | 13.6 | 14.5 | 134,364 |
| 45 to 64 | $50K to $75K | 14.5 | 14.1 | 14.9 | 186,248 |
| 45 to 64 | $75K or more | 16.9 | 16.7 | 17.1 | 540,009 |
| 65 and older | Under $20K | 4.4 | 4.2 | 4.6 | 163,483 |
| 65 and older | $20K to $35K | 4.9 | 4.7 | 5.1 | 270,018 |
| 65 and older | $35K to $50K | 5.2 | 5.0 | 5.5 | 195,888 |
| 65 and older | $50K to $75K | 6.0 | 5.8 | 6.3 | 202,563 |
| 65 and older | $75K or more | 6.9 | 6.7 | 7.1 | 311,719 |

#### Table 2.2. By education

Within each education level binge drinking is higher in the highest income band, by 12.7 points for adults who did not finish high school and 3.5 for college graduates

Percent of adults who binged in the past month by household income band, four education levels, BRFSS 2015 to 2024

| Education level | Income band | Percent of adults | Lower limit of the 95% interval | Upper limit of the 95% interval | Respondents |
| --- | --- | ---: | ---: | ---: | ---: |
| Did not finish high school | Under $20K | 11.7 | 11.3 | 12.2 | 89,482 |
| Did not finish high school | $20K to $35K | 14.7 | 14.1 | 15.3 | 59,975 |
| Did not finish high school | $35K to $50K | 17.3 | 16.3 | 18.4 | 20,704 |
| Did not finish high school | $50K to $75K | 20.0 | 18.7 | 21.5 | 12,176 |
| Did not finish high school | $75K or more | 24.4 | 23.0 | 26.0 | 13,268 |
| High school graduate | Under $20K | 12.7 | 12.3 | 13.1 | 176,340 |
| High school graduate | $20K to $35K | 14.3 | 14.0 | 14.6 | 231,398 |
| High school graduate | $35K to $50K | 16.6 | 16.2 | 17.0 | 141,970 |
| High school graduate | $50K to $75K | 18.6 | 18.1 | 19.0 | 126,631 |
| High school graduate | $75K or more | 23.3 | 22.9 | 23.8 | 162,563 |
| Some college | Under $20K | 15.4 | 15.0 | 15.9 | 123,117 |
| Some college | $20K to $35K | 14.8 | 14.4 | 15.2 | 195,972 |
| Some college | $35K to $50K | 15.9 | 15.5 | 16.3 | 153,035 |
| Some college | $50K to $75K | 17.5 | 17.1 | 17.8 | 172,120 |
| Some college | $75K or more | 21.7 | 21.3 | 22.0 | 278,348 |
| College graduate | Under $20K | 15.2 | 14.6 | 15.8 | 60,504 |
| College graduate | $20K to $35K | 15.3 | 14.8 | 15.7 | 125,543 |
| College graduate | $35K to $50K | 15.4 | 15.0 | 15.8 | 147,709 |
| College graduate | $50K to $75K | 16.6 | 16.3 | 16.9 | 241,590 |
| College graduate | $75K or more | 18.7 | 18.5 | 18.8 | 799,249 |

#### Table 2.3. By race and ethnicity

Binge drinking is higher in the highest income band by 8.6 points among non-Hispanic White adults and 10.3 among Hispanic adults, and by 0.5 among non-Hispanic Black adults

Percent of adults who binged in the past month by household income band, five race and ethnicity groups as BRFSS defines them, BRFSS 2015 to 2024

| Race and ethnicity | Income band | Percent of adults | Lower limit of the 95% interval | Upper limit of the 95% interval | Respondents |
| --- | --- | ---: | ---: | ---: | ---: |
| White, non-Hispanic | Under $20K | 13.0 | 12.8 | 13.3 | 275,676 |
| White, non-Hispanic | $20K to $35K | 13.7 | 13.5 | 13.9 | 433,459 |
| White, non-Hispanic | $35K to $50K | 15.8 | 15.5 | 16.0 | 356,207 |
| White, non-Hispanic | $50K to $75K | 17.8 | 17.5 | 18.0 | 446,702 |
| White, non-Hispanic | $75K or more | 21.7 | 21.5 | 21.8 | 1,043,464 |
| Black, non-Hispanic | Under $20K | 13.3 | 12.8 | 13.8 | 61,641 |
| Black, non-Hispanic | $20K to $35K | 13.5 | 13.0 | 14.1 | 58,764 |
| Black, non-Hispanic | $35K to $50K | 14.9 | 14.1 | 15.6 | 36,009 |
| Black, non-Hispanic | $50K to $75K | 13.9 | 13.2 | 14.7 | 35,243 |
| Black, non-Hispanic | $75K or more | 13.8 | 13.2 | 14.3 | 59,333 |
| Other race, non-Hispanic | Under $20K | 11.7 | 10.9 | 12.6 | 28,856 |
| Other race, non-Hispanic | $20K to $35K | 12.1 | 11.3 | 13.0 | 30,152 |
| Other race, non-Hispanic | $35K to $50K | 12.1 | 11.2 | 13.1 | 20,159 |
| Other race, non-Hispanic | $50K to $75K | 13.0 | 12.0 | 14.0 | 22,237 |
| Other race, non-Hispanic | $75K or more | 12.7 | 12.2 | 13.3 | 57,224 |
| Multiracial, non-Hispanic | Under $20K | 15.6 | 14.3 | 17.1 | 12,949 |
| Multiracial, non-Hispanic | $20K to $35K | 17.2 | 16.0 | 18.5 | 14,072 |
| Multiracial, non-Hispanic | $35K to $50K | 20.1 | 18.5 | 21.7 | 9,793 |
| Multiracial, non-Hispanic | $50K to $75K | 19.0 | 17.6 | 20.4 | 10,918 |
| Multiracial, non-Hispanic | $75K or more | 21.2 | 20.1 | 22.3 | 22,847 |
| Hispanic | Under $20K | 13.9 | 13.3 | 14.4 | 63,202 |
| Hispanic | $20K to $35K | 18.0 | 17.4 | 18.6 | 67,969 |
| Hispanic | $35K to $50K | 20.0 | 19.2 | 20.9 | 35,225 |
| Hispanic | $50K to $75K | 22.3 | 21.3 | 23.2 | 30,905 |
| Hispanic | $75K or more | 24.1 | 23.4 | 24.9 | 56,142 |

Note: Respondents who did not know their household income or declined to give it are left out of the bands, 19.4% of adults. The five bands group the income categories that BRFSS used in every year. The categories are in nominal dollars, so the highest band holds 30.2% of respondents who gave an income in 2014 and 46.1% in 2024, and part of that shift is a step at the 2021 change in the income question.

Note: Intervals are 95% confidence intervals from the survey's stratum and primary sampling unit (design-based, Taylor linearization). No cell rests on fewer than 30 respondents. The survey is a repeated cross-section of adults in the states and the District of Columbia; Guam, Puerto Rico and the Virgin Islands are not included.

Note: Age groups, education levels and race and ethnicity groups are the BRFSS calculated categories. The four education levels are no high school diploma, a high school diploma, some college or technical school, and a college or technical school degree. Race and ethnicity has five groups; the Hispanic group may be of any race.

Source: BRFSS 2015-2024 public-use files, weighted (BRFSS final weight) with design-based intervals

Data: https://vishalsingh.org/zipcode/report/health-behaviors-person-by-person/figures/binge_within_groups.json

### Table 3. Binge drinking is higher in the highest income band than in the lowest in all 51 jurisdictions, and smoking is lower in all 51

Rate in the highest household income band minus the rate in the lowest, percentage points of adults, by state and the District of Columbia, BRFSS 2015 to 2024 for binge drinking and 2014 to 2024 for smoking

| Measure | State | Difference (points) | Lower limit of the 95% interval | Upper limit of the 95% interval | Respondents in the two bands |
| --- | --- | ---: | ---: | ---: | ---: |
| Binge drinking | Alabama | 5.1 | 3.8 | 6.5 | 22,132 |
| Binge drinking | Alaska | 4.7 | 2.3 | 7.1 | 18,795 |
| Binge drinking | Arizona | 5.2 | 3.9 | 6.6 | 37,600 |
| Binge drinking | Arkansas | 4.3 | 2.5 | 6.1 | 17,920 |
| Binge drinking | California | 6.0 | 4.9 | 7.0 | 45,940 |
| Binge drinking | Colorado | 6.7 | 5.5 | 7.9 | 44,591 |
| Binge drinking | Connecticut | 6.2 | 4.9 | 7.5 | 43,306 |
| Binge drinking | Delaware | 7.0 | 5.2 | 8.8 | 15,780 |
| Binge drinking | District of Columbia | 13.1 | 10.7 | 15.5 | 18,026 |
| Binge drinking | Florida | 7.5 | 6.1 | 8.9 | 52,789 |
| Binge drinking | Georgia | 5.2 | 3.8 | 6.6 | 27,715 |
| Binge drinking | Hawaii | 3.6 | 2.1 | 5.0 | 34,030 |
| Binge drinking | Idaho | 6.4 | 4.7 | 8.0 | 19,584 |
| Binge drinking | Illinois | 6.7 | 5.2 | 8.1 | 25,131 |
| Binge drinking | Indiana | 8.5 | 7.3 | 9.7 | 36,284 |
| Binge drinking | Iowa | 11.2 | 9.9 | 12.6 | 33,575 |
| Binge drinking | Kansas | 6.5 | 5.4 | 7.6 | 51,800 |
| Binge drinking | Kentucky | 10.2 | 8.7 | 11.8 | 22,343 |
| Binge drinking | Louisiana | 10.6 | 9.1 | 12.0 | 19,904 |
| Binge drinking | Maine | 6.3 | 4.9 | 7.8 | 39,103 |
| Binge drinking | Maryland | 5.7 | 4.4 | 6.9 | 68,318 |
| Binge drinking | Massachusetts | 8.4 | 7.0 | 9.7 | 36,575 |
| Binge drinking | Michigan | 8.1 | 7.0 | 9.3 | 38,454 |
| Binge drinking | Minnesota | 8.4 | 7.4 | 9.4 | 67,343 |
| Binge drinking | Mississippi | 7.7 | 6.2 | 9.2 | 18,082 |
| Binge drinking | Missouri | 9.6 | 8.1 | 11.0 | 29,469 |
| Binge drinking | Montana | 6.1 | 4.6 | 7.7 | 23,613 |
| Binge drinking | Nebraska | 10.6 | 9.4 | 11.7 | 53,614 |
| Binge drinking | Nevada | 7.9 | 5.6 | 10.1 | 11,971 |
| Binge drinking | New Hampshire | 9.6 | 7.8 | 11.4 | 25,428 |
| Binge drinking | New Jersey | 6.4 | 4.8 | 8.0 | 34,305 |
| Binge drinking | New Mexico | 4.6 | 3.1 | 6.1 | 22,177 |
| Binge drinking | New York | 7.5 | 6.7 | 8.4 | 95,165 |
| Binge drinking | North Carolina | 7.0 | 5.6 | 8.3 | 18,903 |
| Binge drinking | North Dakota | 7.6 | 5.6 | 9.6 | 22,318 |
| Binge drinking | Ohio | 7.5 | 6.3 | 8.8 | 47,777 |
| Binge drinking | Oklahoma | 6.7 | 5.4 | 8.0 | 21,765 |
| Binge drinking | Oregon | 5.8 | 4.4 | 7.1 | 22,172 |
| Binge drinking | Pennsylvania | 8.2 | 6.7 | 9.7 | 21,333 |
| Binge drinking | Rhode Island | 7.1 | 5.4 | 8.9 | 23,732 |
| Binge drinking | South Carolina | 9.4 | 8.2 | 10.7 | 34,360 |
| Binge drinking | South Dakota | 5.2 | 2.6 | 7.8 | 23,776 |
| Binge drinking | Tennessee | 8.1 | 6.6 | 9.7 | 18,485 |
| Binge drinking | Texas | 7.4 | 5.9 | 8.8 | 47,898 |
| Binge drinking | Utah | 2.7 | 1.6 | 3.7 | 46,474 |
| Binge drinking | Vermont | 6.1 | 4.3 | 7.9 | 27,137 |
| Binge drinking | Virginia | 5.4 | 4.1 | 6.6 | 38,022 |
| Binge drinking | Washington | 5.9 | 4.9 | 6.8 | 74,985 |
| Binge drinking | West Virginia | 7.9 | 6.7 | 9.2 | 21,040 |
| Binge drinking | Wisconsin | 7.7 | 5.9 | 9.5 | 27,748 |
| Binge drinking | Wyoming | 4.7 | 2.5 | 6.8 | 16,577 |
| Current smoking | Alabama | −23.1 | −24.6 | −21.7 | 25,862 |
| Current smoking | Alaska | −29.9 | −32.6 | −27.2 | 21,314 |
| Current smoking | Arizona | −15.4 | −16.7 | −14.1 | 43,911 |
| Current smoking | Arkansas | −26.5 | −28.3 | −24.6 | 20,251 |
| Current smoking | California | −9.1 | −9.9 | −8.2 | 51,083 |
| Current smoking | Colorado | −18.7 | −19.8 | −17.5 | 50,721 |
| Current smoking | Connecticut | −17.0 | −18.3 | −15.6 | 47,952 |
| Current smoking | Delaware | −17.8 | −19.8 | −15.9 | 17,868 |
| Current smoking | District of Columbia | −22.7 | −24.9 | −20.5 | 20,541 |
| Current smoking | Florida | −14.9 | −16.2 | −13.6 | 58,131 |
| Current smoking | Georgia | −18.5 | −19.9 | −17.1 | 31,002 |
| Current smoking | Hawaii | −14.3 | −15.6 | −13.0 | 37,541 |
| Current smoking | Idaho | −18.5 | −20.1 | −16.9 | 21,900 |
| Current smoking | Illinois | −15.3 | −16.8 | −13.8 | 27,952 |
| Current smoking | Indiana | −23.1 | −24.3 | −21.8 | 41,592 |
| Current smoking | Iowa | −20.5 | −21.9 | −19.2 | 37,013 |
| Current smoking | Kansas | −24.1 | −25.3 | −23.0 | 57,589 |
| Current smoking | Kentucky | −27.5 | −29.2 | −25.8 | 26,966 |
| Current smoking | Louisiana | −19.1 | −20.7 | −17.6 | 23,376 |
| Current smoking | Maine | −26.8 | −28.2 | −25.4 | 43,531 |
| Current smoking | Maryland | −17.2 | −18.6 | −15.8 | 76,105 |
| Current smoking | Massachusetts | −16.3 | −17.7 | −14.9 | 43,965 |
| Current smoking | Michigan | −25.2 | −26.4 | −23.9 | 42,477 |
| Current smoking | Minnesota | −18.9 | −19.9 | −17.9 | 75,694 |
| Current smoking | Mississippi | −19.1 | −20.7 | −17.5 | 20,273 |
| Current smoking | Missouri | −26.1 | −27.6 | −24.6 | 32,584 |
| Current smoking | Montana | −26.3 | −27.9 | −24.8 | 26,674 |
| Current smoking | Nebraska | −18.7 | −19.8 | −17.6 | 62,552 |
| Current smoking | Nevada | −15.7 | −18.1 | −13.4 | 13,704 |
| Current smoking | New Hampshire | −26.5 | −28.6 | −24.4 | 28,302 |
| Current smoking | New Jersey | −12.0 | −13.6 | −10.3 | 41,041 |
| Current smoking | New Mexico | −15.7 | −17.0 | −14.3 | 26,116 |
| Current smoking | New York | −12.1 | −12.9 | −11.2 | 101,119 |
| Current smoking | North Carolina | −19.5 | −21.1 | −17.9 | 22,204 |
| Current smoking | North Dakota | −17.1 | −19.0 | −15.2 | 25,970 |
| Current smoking | Ohio | −26.6 | −27.9 | −25.3 | 52,993 |
| Current smoking | Oklahoma | −23.9 | −25.3 | −22.4 | 25,772 |
| Current smoking | Oregon | −21.3 | −22.8 | −19.9 | 24,599 |
| Current smoking | Pennsylvania | −20.2 | −21.8 | −18.6 | 26,060 |
| Current smoking | Rhode Island | −16.0 | −17.7 | −14.4 | 27,026 |
| Current smoking | South Carolina | −20.6 | −21.9 | −19.3 | 39,654 |
| Current smoking | South Dakota | −20.9 | −23.4 | −18.4 | 26,826 |
| Current smoking | Tennessee | −25.2 | −26.9 | −23.5 | 20,741 |
| Current smoking | Texas | −11.2 | −12.5 | −10.0 | 55,625 |
| Current smoking | Utah | −13.3 | −14.4 | −12.2 | 53,042 |
| Current smoking | Vermont | −22.9 | −24.6 | −21.3 | 30,207 |
| Current smoking | Virginia | −19.2 | −20.5 | −17.9 | 42,791 |
| Current smoking | Washington | −19.2 | −20.3 | −18.1 | 80,378 |
| Current smoking | West Virginia | −28.5 | −29.9 | −27.1 | 23,595 |
| Current smoking | Wisconsin | −20.2 | −21.9 | −18.5 | 31,091 |
| Current smoking | Wyoming | −19.9 | −22.1 | −17.8 | 19,130 |

Views: Binge drinking (measure = Binge drinking); Current smoking (measure = Current smoking).

Note: Respondents who did not know their household income or declined to give it are left out of the bands, 19.4% of adults. The five bands group the income categories that BRFSS used in every year. The categories are in nominal dollars, so the highest band holds 30.2% of respondents who gave an income in 2014 and 46.1% in 2024, and part of that shift is a step at the 2021 change in the income question.

Note: Intervals are 95% confidence intervals from the survey's stratum and primary sampling unit (design-based, Taylor linearization). No cell rests on fewer than 30 respondents. The survey is a repeated cross-section of adults in the states and the District of Columbia; Guam, Puerto Rico and the Virgin Islands are not included.

Source: BRFSS 2014-2024 public-use files (binge drinking from 2015), weighted (BRFSS final weight) with design-based intervals

Data: https://vishalsingh.org/zipcode/report/health-behaviors-person-by-person/figures/gradient_by_state.json

### Table 4. Men binge drink more than women in every income band, by 8.5 to 9.2 points

Percent of adults who binged in the past month by sex and household income band, BRFSS 2015 to 2024, with 95% confidence intervals

| Measure | Sex | Income band | Percent of adults | Lower limit of the 95% interval | Upper limit of the 95% interval | Respondents |
| --- | --- | --- | ---: | ---: | ---: | ---: |
| Binge drinking | Men | Under $20K | 18.3 | 17.9 | 18.7 | 170,518 |
| Binge drinking | Men | $20K to $35K | 19.6 | 19.3 | 20.0 | 253,196 |
| Binge drinking | Men | $35K to $50K | 20.5 | 20.2 | 20.9 | 211,271 |
| Binge drinking | Men | $50K to $75K | 22.1 | 21.7 | 22.4 | 265,095 |
| Binge drinking | Men | $75K or more | 24.3 | 24.1 | 24.5 | 657,924 |
| Binge drinking | Women | Under $20K | 9.7 | 9.4 | 9.9 | 280,085 |
| Binge drinking | Women | $20K to $35K | 10.5 | 10.2 | 10.7 | 360,747 |
| Binge drinking | Women | $35K to $50K | 11.9 | 11.6 | 12.2 | 252,750 |
| Binge drinking | Women | $50K to $75K | 13.0 | 12.7 | 13.3 | 287,884 |
| Binge drinking | Women | $75K or more | 15.8 | 15.6 | 16.0 | 596,461 |
| Obesity | Men | Under $20K | 31.1 | 30.7 | 31.6 | 198,642 |
| Obesity | Men | $20K to $35K | 31.3 | 31.0 | 31.7 | 291,809 |
| Obesity | Men | $35K to $50K | 32.0 | 31.6 | 32.4 | 242,454 |
| Obesity | Men | $50K to $75K | 32.8 | 32.5 | 33.2 | 300,974 |
| Obesity | Men | $75K or more | 30.9 | 30.7 | 31.1 | 733,927 |
| Obesity | Women | Under $20K | 39.7 | 39.4 | 40.1 | 314,114 |
| Obesity | Women | $20K to $35K | 37.3 | 37.0 | 37.7 | 395,965 |
| Obesity | Women | $35K to $50K | 34.9 | 34.5 | 35.3 | 274,183 |
| Obesity | Women | $50K to $75K | 33.2 | 32.8 | 33.5 | 308,724 |
| Obesity | Women | $75K or more | 25.3 | 25.1 | 25.6 | 631,594 |
| Current smoking | Men | Under $20K | 29.5 | 29.1 | 29.9 | 200,372 |
| Current smoking | Men | $20K to $35K | 23.1 | 22.8 | 23.4 | 291,391 |
| Current smoking | Men | $35K to $50K | 18.9 | 18.6 | 19.3 | 240,761 |
| Current smoking | Men | $50K to $75K | 15.7 | 15.5 | 16.0 | 298,532 |
| Current smoking | Men | $75K or more | 9.8 | 9.7 | 10.0 | 728,129 |
| Current smoking | Women | Under $20K | 22.8 | 22.5 | 23.1 | 331,241 |
| Current smoking | Women | $20K to $35K | 17.5 | 17.2 | 17.7 | 415,794 |
| Current smoking | Women | $35K to $50K | 14.0 | 13.7 | 14.3 | 288,686 |
| Current smoking | Women | $50K to $75K | 11.4 | 11.2 | 11.7 | 325,465 |
| Current smoking | Women | $75K or more | 6.9 | 6.8 | 7.0 | 664,260 |

Views: Binge drinking (measure = Binge drinking); Obesity (measure = Obesity); Current smoking (measure = Current smoking).

Note: Respondents who did not know their household income or declined to give it are left out of the bands, 19.4% of adults. The five bands group the income categories that BRFSS used in every year. The categories are in nominal dollars, so the highest band holds 30.2% of respondents who gave an income in 2014 and 46.1% in 2024, and part of that shift is a step at the 2021 change in the income question.

Note: Intervals are 95% confidence intervals from the survey's stratum and primary sampling unit (design-based, Taylor linearization). No cell rests on fewer than 30 respondents. The survey is a repeated cross-section of adults in the states and the District of Columbia; Guam, Puerto Rico and the Virgin Islands are not included.

Source: BRFSS 2014-2024 public-use files (binge drinking from 2015), weighted (BRFSS final weight) with design-based intervals

Data: https://vishalsingh.org/zipcode/report/health-behaviors-person-by-person/figures/sex_by_income.json

### Table 5. The obesity difference between the highest and the lowest income band is −14.4 points for women and −0.2 for men

Rate in the highest household income band minus the rate in the lowest, percentage points of adults, by measure and sex, BRFSS 2014 to 2024 (binge drinking 2015 to 2024)

| Measure | Sex | Difference (points) | Lower limit of the 95% interval | Upper limit of the 95% interval | Respondents in the two bands |
| --- | --- | ---: | ---: | ---: | ---: |
| Binge drinking | Women | 6.2 | 5.8 | 6.5 | 876,546 |
| Binge drinking | Men | 6.0 | 5.6 | 6.5 | 828,442 |
| Current smoking | Women | −15.9 | −16.2 | −15.6 | 995,501 |
| Current smoking | Men | −19.7 | −20.1 | −19.3 | 928,501 |
| Obesity | Women | −14.4 | −14.8 | −13.9 | 945,708 |
| Obesity | Men | −0.2 | −0.7 | 0.3 | 932,569 |
| Diabetes | Women | −11.8 | −12.1 | −11.5 | 1,006,571 |
| Diabetes | Men | −8.5 | −8.9 | −8.2 | 952,929 |
| No leisure-time physical activity | Women | −25.7 | −26.1 | −25.3 | 1,009,313 |
| No leisure-time physical activity | Men | −22.6 | −23.1 | −22.1 | 939,848 |
| Fair or poor health | Women | −30.6 | −31.0 | −30.3 | 1,023,026 |
| Fair or poor health | Men | −29.7 | −30.2 | −29.3 | 952,294 |

Note: Respondents who did not know their household income or declined to give it are left out of the bands, 19.4% of adults. The five bands group the income categories that BRFSS used in every year. The categories are in nominal dollars, so the highest band holds 30.2% of respondents who gave an income in 2014 and 46.1% in 2024, and part of that shift is a step at the 2021 change in the income question.

Note: Intervals are 95% confidence intervals from the survey's stratum and primary sampling unit (design-based, Taylor linearization). No cell rests on fewer than 30 respondents. The survey is a repeated cross-section of adults in the states and the District of Columbia; Guam, Puerto Rico and the Virgin Islands are not included.

Note: Obesity is a body mass index of 30 or more from self-reported height and weight. Diabetes is a diagnosis reported by the respondent: prediabetes counts as no diabetes, and respondents with diabetes only during pregnancy are left out of the denominator. Fair or poor health is the respondent's rating of general health.

Source: BRFSS 2014-2024 public-use files (binge drinking from 2015), weighted (BRFSS final weight) with design-based intervals

Data: https://vishalsingh.org/zipcode/report/health-behaviors-person-by-person/figures/gradient_by_behavior_and_sex.json

### Table 6. Smoking is 28.0% among adults who did not finish high school in the lowest income band and 4.4% among college graduates in the highest

Percent of adults who currently smoke by education level and household income band, BRFSS 2014 to 2024

| band | conf_high | conf_low | education | estimate | n |
| --- | ---: | ---: | --- | ---: | ---: |
| Under $20K | 28.48 | 27.47 | Did not finish high school | 27.97 | 107,166 |
| $20K to $35K | 25.20 | 23.98 | Did not finish high school | 24.59 | 69,781 |
| $35K to $50K | 24.96 | 22.91 | Did not finish high school | 23.92 | 23,739 |
| $50K to $75K | 27.78 | 25.07 | Did not finish high school | 26.40 | 13,858 |
| $75K or more | 24.51 | 21.90 | Did not finish high school | 23.18 | 15,060 |
| Under $20K | 27.62 | 26.84 | High school graduate | 27.23 | 208,946 |
| $20K to $35K | 21.99 | 21.34 | High school graduate | 21.66 | 268,241 |
| $35K to $50K | 19.63 | 18.85 | High school graduate | 19.24 | 162,535 |
| $50K to $75K | 18.15 | 17.37 | High school graduate | 17.76 | 143,885 |
| $75K or more | 14.85 | 14.19 | High school graduate | 14.52 | 181,529 |
| Under $20K | 24.80 | 23.91 | Some college | 24.35 | 144,056 |
| $20K to $35K | 19.46 | 18.81 | Some college | 19.13 | 224,209 |
| $35K to $50K | 16.75 | 16.04 | Some college | 16.39 | 174,273 |
| $50K to $75K | 14.54 | 13.92 | Some college | 14.23 | 193,584 |
| $75K or more | 11.09 | 10.66 | Some college | 10.87 | 309,247 |
| Under $20K | 14.06 | 13.15 | College graduate | 13.60 | 70,037 |
| $20K to $35K | 11.21 | 10.56 | College graduate | 10.88 | 143,684 |
| $35K to $50K | 8.70 | 8.18 | College graduate | 8.44 | 168,202 |
| $50K to $75K | 7.19 | 6.79 | College graduate | 6.99 | 272,154 |
| $75K or more | 4.54 | 4.37 | College graduate | 4.45 | 885,429 |

Note: Respondents who did not know their household income or declined to give it are left out of the bands, 19.4% of adults. The five bands group the income categories that BRFSS used in every year. The categories are in nominal dollars, so the highest band holds 30.2% of respondents who gave an income in 2014 and 46.1% in 2024, and part of that shift is a step at the 2021 change in the income question.

Note: Intervals are 95% confidence intervals from the survey's stratum and primary sampling unit (design-based, Taylor linearization). No cell rests on fewer than 30 respondents. The survey is a repeated cross-section of adults in the states and the District of Columbia; Guam, Puerto Rico and the Virgin Islands are not included.

Note: Current smoking is at least 100 cigarettes in a lifetime and smoking every day or some days now. Education is the level the respondent completed; college includes technical school.

Source: BRFSS 2014-2024 public-use files, weighted (BRFSS final weight) with design-based intervals

Data: https://vishalsingh.org/zipcode/report/health-behaviors-person-by-person/figures/smoking_by_education_and_income.json

### Table 7. Men smoke more than women by 1.7 points among non-Hispanic White adults and 8.1 among Hispanic adults

Percent of adults who currently smoke by sex within five race and ethnicity groups, BRFSS 2014 to 2024, with 95% confidence intervals

| group | hi_m | hi_w | lo_m | lo_w | men | n_m | n_w | women |
| --- | ---: | ---: | ---: | ---: | ---: | ---: | ---: | ---: |
| White, non-Hispanic | 16.37 | 14.61 | 16.13 | 14.41 | 16.25 | 1,553,398 | 1,898,894 | 14.51 |
| Black, non-Hispanic | 20.61 | 13.54 | 19.84 | 12.99 | 20.22 | 134,857 | 211,073 | 13.26 |
| Other race, non-Hispanic | 13.98 | 7.97 | 13.12 | 7.34 | 13.55 | 107,849 | 107,257 | 7.65 |
| Multiracial, non-Hispanic | 22.67 | 19.37 | 21.15 | 17.88 | 21.90 | 44,344 | 48,941 | 18.62 |
| Hispanic | 16.11 | 7.88 | 15.42 | 7.41 | 15.76 | 161,479 | 185,066 | 7.64 |

Note: Intervals are 95% confidence intervals from the survey's stratum and primary sampling unit (design-based, Taylor linearization). No cell rests on fewer than 30 respondents. The survey is a repeated cross-section of adults in the states and the District of Columbia; Guam, Puerto Rico and the Virgin Islands are not included.

Note: Groups are those BRFSS defines: White, Black, another race and multiracial adults are non-Hispanic, and Hispanic adults may be of any race. The Hispanic and other-race groups each combine populations whose smoking differs.

Source: BRFSS 2014-2024 public-use files, weighted (BRFSS final weight) with design-based intervals

Data: https://vishalsingh.org/zipcode/report/health-behaviors-person-by-person/figures/smoking_gap_by_race.json

## All estimates

Every fact in facts.json (124), sorted by id.

| Fact id | Value | Unit | Estimand | Population | Period | Interval | n | Weights | Source | Note |
| --- | ---: | --- | --- | --- | --- | --- | ---: | --- | --- | --- |
| age_gap_old | 2.6 | percentage points of adults, higher in the highest income band | design-based difference between two weighted shares of respondents, Taylor-linearized 95% interval | adults in the 50 states and the District of Columbia ages 65 and older: binge drinking in the past 30 days | BRFSS 2015-2024 | 95% CI 2.3 to 2.9 | 475,202 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| age_gap_young | 10.2 | percentage points of adults, higher in the highest income band | design-based difference between two weighted shares of respondents, Taylor-linearized 95% interval | adults in the 50 states and the District of Columbia ages 18 to 29: binge drinking in the past 30 days | BRFSS 2015-2024 | 95% CI 9.4 to 11.0 | 151,901 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| age_old | 65 | count in a definition | definition: the youngest age in the oldest group | the youngest age in the oldest group | 2014-2024 |   |   | none (definition) | BRFSS |   |
| age_young_hi | 29 | count in a definition | definition: the oldest age in the youngest group | the oldest age in the youngest group | 2014-2024 |   |   | none (definition) | BRFSS |   |
| age_young_lo | 18 | count in a definition | definition: the youngest age in the survey and in the youngest group | the youngest age in the survey and in the youngest group | 2014-2024 |   |   | none (definition) | BRFSS |   |
| binge8_75 | 18.2% | percent of adults | design-based weighted share of respondents, Taylor-linearized 95% interval | adults in the 50 states and the District of Columbia in the $75,000 to $99,999 band: binge drinking in the past 30 days | BRFSS 2021-2024 | 95% CI 17.8% to 18.6% | 184,694 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| binge8_top | 22.1% | percent of adults | design-based weighted share of respondents, Taylor-linearized 95% interval | adults in the 50 states and the District of Columbia in the $200,000 or more band: binge drinking in the past 30 days | BRFSS 2021-2024 | 95% CI 21.5% to 22.6% | 93,005 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| binge_black_poor | 13.3% | percent of adults | design-based weighted share of respondents, Taylor-linearized 95% interval | Black non-Hispanic adults in the lowest of five household income bands: binge drinking in the past 30 days | BRFSS 2015-2024 | 95% CI 12.8% to 13.8% | 61,641 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| binge_black_rich | 13.8% | percent of adults | design-based weighted share of respondents, Taylor-linearized 95% interval | Black non-Hispanic adults in the highest of five household income bands: binge drinking in the past 30 days | BRFSS 2015-2024 | 95% CI 13.2% to 14.3% | 59,333 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| binge_coll_poor | 15.2% | percent of adults | design-based weighted share of respondents, Taylor-linearized 95% interval | adults in the 50 states and the District of Columbia who graduated from college or technical school in the lowest of five household income bands: binge drinking in the past 30 days | BRFSS 2015-2024 | 95% CI 14.6% to 15.8% | 60,504 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| binge_coll_rich | 18.7% | percent of adults | design-based weighted share of respondents, Taylor-linearized 95% interval | adults in the 50 states and the District of Columbia who graduated from college or technical school in the highest of five household income bands: binge drinking in the past 30 days | BRFSS 2015-2024 | 95% CI 18.5% to 18.8% | 799,249 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| binge_gap | 7.2 | percentage points of adults, higher in the highest income band | design-based difference between two weighted shares of respondents, Taylor-linearized 95% interval | adults in the 50 states and the District of Columbia in the highest and lowest of five household income bands: binge drinking in the past 30 days | BRFSS 2015-2024 | 95% CI 6.9 to 7.5 | 1,705,364 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| binge_men_poor | 18.3% | percent of adults | design-based weighted share of respondents, Taylor-linearized 95% interval | adults in the 50 states and the District of Columbia who are men in the lowest of five household income bands: binge drinking in the past 30 days | BRFSS 2015-2024 | 95% CI 17.9% to 18.7% | 170,518 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| binge_men_rich | 24.3% | percent of adults | design-based weighted share of respondents, Taylor-linearized 95% interval | adults in the 50 states and the District of Columbia who are men in the highest of five household income bands: binge drinking in the past 30 days | BRFSS 2015-2024 | 95% CI 24.1% to 24.5% | 657,924 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| binge_nohs_poor | 11.7% | percent of adults | design-based weighted share of respondents, Taylor-linearized 95% interval | adults in the 50 states and the District of Columbia who did not finish high school in the lowest of five household income bands: binge drinking in the past 30 days | BRFSS 2015-2024 | 95% CI 11.3% to 12.2% | 89,482 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| binge_nohs_rich | 24.4% | percent of adults | design-based weighted share of respondents, Taylor-linearized 95% interval | adults in the 50 states and the District of Columbia who did not finish high school in the highest of five household income bands: binge drinking in the past 30 days | BRFSS 2015-2024 | 95% CI 23.0% to 26.0% | 13,268 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| binge_old_poor | 4.4% | percent of adults | design-based weighted share of respondents, Taylor-linearized 95% interval | adults in the 50 states and the District of Columbia ages 65 and older in the lowest of five household income bands: binge drinking in the past 30 days | BRFSS 2015-2024 | 95% CI 4.2% to 4.6% | 163,483 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| binge_old_rich | 6.9% | percent of adults | design-based weighted share of respondents, Taylor-linearized 95% interval | adults in the 50 states and the District of Columbia ages 65 and older in the highest of five household income bands: binge drinking in the past 30 days | BRFSS 2015-2024 | 95% CI 6.7% to 7.1% | 311,719 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| binge_poor | 13.3% | percent of adults | design-based weighted share of respondents, Taylor-linearized 95% interval | adults in the 50 states and the District of Columbia in the lowest of five household income bands: binge drinking in the past 30 days | BRFSS 2015-2024 | 95% CI 13.0% to 13.5% | 450,791 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| binge_rich | 20.4% | percent of adults | design-based weighted share of respondents, Taylor-linearized 95% interval | adults in the 50 states and the District of Columbia in the highest of five household income bands: binge drinking in the past 30 days | BRFSS 2015-2024 | 95% CI 20.3% to 20.6% | 1,254,573 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| binge_sex_diff | 0.1 | percentage points, women's difference between the highest and the lowest income band minus men's | difference between the women's and the men's income-band differences; standard errors added in quadrature (disjoint groups of respondents) | adults in the 50 states and the District of Columbia: binge drinking in the past 30 days | BRFSS 2015-2024 | 95% CI −0.4 to 0.7 | 1,704,988 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| binge_white_poor | 13.0% | percent of adults | design-based weighted share of respondents, Taylor-linearized 95% interval | White non-Hispanic adults in the lowest of five household income bands: binge drinking in the past 30 days | BRFSS 2015-2024 | 95% CI 12.8% to 13.3% | 275,676 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| binge_white_rich | 21.7% | percent of adults | design-based weighted share of respondents, Taylor-linearized 95% interval | White non-Hispanic adults in the highest of five household income bands: binge drinking in the past 30 days | BRFSS 2015-2024 | 95% CI 21.5% to 21.8% | 1,043,464 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| binge_women_poor | 9.7% | percent of adults | design-based weighted share of respondents, Taylor-linearized 95% interval | adults in the 50 states and the District of Columbia who are women in the lowest of five household income bands: binge drinking in the past 30 days | BRFSS 2015-2024 | 95% CI 9.4% to 9.9% | 280,085 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| binge_women_rich | 15.8% | percent of adults | design-based weighted share of respondents, Taylor-linearized 95% interval | adults in the 50 states and the District of Columbia who are women in the highest of five household income bands: binge drinking in the past 30 days | BRFSS 2015-2024 | 95% CI 15.6% to 16.0% | 596,461 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| bmi_cut | 30 | count in a definition | definition: the body mass index at which an adult counts as obese | the body mass index at which an adult counts as obese | 2014-2024 |   |   | none (definition) | BRFSS |   |
| cell_floor | 30 | count in a definition | definition: the smallest number of respondents behind any published cell | the smallest number of respondents behind any published cell | 2014-2024 |   |   | none (definition) | BRFSS |   |
| cells_gap_n | 25 | cells | count of cells: five groups by five income bands | race and ethnicity by income band cells | BRFSS 2014-2024 |   |   | none (count) | BRFSS |   |
| cells_gap_pos | 25 | cells | count of cells where the 95% interval of the smoking difference between men and women is above zero | race and ethnicity by income band cells | BRFSS 2014-2024 |   |   | none (count) | BRFSS |   |
| cigs_life | 100 | count in a definition | definition: the lifetime cigarettes that make a smoker | the lifetime cigarettes that make a smoker | 2014-2024 |   |   | none (definition) | BRFSS |   |
| cut_low | $20,000 | dollars of annual household income | definition: income band edge | upper limit of the lowest income band | 2014-2024 |   |   | none (definition) | BRFSS |   |
| cut_top | $75,000 | dollars of annual household income | definition: income band edge | lower limit of the highest income band | 2014-2024 |   |   | none (definition) | BRFSS |   |
| days_past | 30 | count in a definition | definition: days in the period the binge drinking question covers | days in the period the binge drinking question covers | 2014-2024 |   |   | none (definition) | BRFSS |   |
| drinks_men | 5 | count in a definition | definition: drinks on one occasion that make a binge for men | drinks on one occasion that make a binge for men | 2014-2024 |   |   | none (definition) | BRFSS |   |
| drinks_women | 4 | count in a definition | definition: drinks on one occasion that make a binge for women | drinks on one occasion that make a binge for women | 2014-2024 |   |   | none (definition) | BRFSS |   |
| edu_gap_coll | 3.5 | percentage points of adults, higher in the highest income band | design-based difference between two weighted shares of respondents, Taylor-linearized 95% interval | adults in the 50 states and the District of Columbia who graduated from college or technical school: binge drinking in the past 30 days | BRFSS 2015-2024 | 95% CI 2.9 to 4.1 | 859,753 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| edu_gap_nohs | 12.7 | percentage points of adults, higher in the highest income band | design-based difference between two weighted shares of respondents, Taylor-linearized 95% interval | adults in the 50 states and the District of Columbia who did not finish high school: binge drinking in the past 30 days | BRFSS 2015-2024 | 95% CI 11.1 to 14.2 | 102,750 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| gap_after | 7.5 | percentage points of adults, higher in the highest income band | design-based difference between two weighted shares of respondents, Taylor-linearized 95% interval; before and after BRFSS changed its income question in 2021 | adults in the 50 states and the District of Columbia in the highest and lowest of five household income bands: binge drinking in the past 30 days | 2021-2024 | 95% CI 7.0 to 8.0 | 681,003 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| gap_before | 7.3 | percentage points of adults, higher in the highest income band | design-based difference between two weighted shares of respondents, Taylor-linearized 95% interval; before and after BRFSS changed its income question in 2021 | adults in the 50 states and the District of Columbia in the highest and lowest of five household income bands: binge drinking in the past 30 days | 2015-2020 | 95% CI 6.9 to 7.6 | 1,024,361 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| grad_binge_men | 6.0 | percentage points of adults, richest minus poorest of five income bands | design-based difference between two weighted shares of respondents, Taylor-linearized 95% interval | adults in the 50 states and the District of Columbia who are men: binge drinking in the past 30 days | BRFSS 2015-2024 | 95% CI 5.6 to 6.5 | 828,442 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| grad_binge_women | 6.2 | percentage points of adults, richest minus poorest of five income bands | design-based difference between two weighted shares of respondents, Taylor-linearized 95% interval | adults in the 50 states and the District of Columbia who are women: binge drinking in the past 30 days | BRFSS 2015-2024 | 95% CI 5.8 to 6.5 | 876,546 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| grad_fairpoor_men | −29.7 | percentage points of adults, richest minus poorest of five income bands | design-based difference between two weighted shares of respondents, Taylor-linearized 95% interval | adults in the 50 states and the District of Columbia who are men: fair or poor self-rated health | BRFSS 2014-2024 | 95% CI −30.2 to −29.3 | 952,294 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| grad_fairpoor_women | −30.6 | percentage points of adults, richest minus poorest of five income bands | design-based difference between two weighted shares of respondents, Taylor-linearized 95% interval | adults in the 50 states and the District of Columbia who are women: fair or poor self-rated health | BRFSS 2014-2024 | 95% CI −31.0 to −30.3 | 1,023,026 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| grad_obese_men | −0.2 | percentage points of adults, richest minus poorest of five income bands | design-based difference between two weighted shares of respondents, Taylor-linearized 95% interval | adults in the 50 states and the District of Columbia who are men: obesity (BMI 30 or more, self-reported height and weight) | BRFSS 2014-2024 | 95% CI −0.7 to 0.3 | 932,569 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| grad_obese_women | −14.4 | percentage points of adults, richest minus poorest of five income bands | design-based difference between two weighted shares of respondents, Taylor-linearized 95% interval | adults in the 50 states and the District of Columbia who are women: obesity (BMI 30 or more, self-reported height and weight) | BRFSS 2014-2024 | 95% CI −14.8 to −13.9 | 945,708 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| grad_smoke_men | −19.7 | percentage points of adults, richest minus poorest of five income bands | design-based difference between two weighted shares of respondents, Taylor-linearized 95% interval | adults in the 50 states and the District of Columbia who are men: current cigarette smoking | BRFSS 2014-2024 | 95% CI −20.1 to −19.3 | 928,501 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| grad_smoke_women | −15.9 | percentage points of adults, richest minus poorest of five income bands | design-based difference between two weighted shares of respondents, Taylor-linearized 95% interval | adults in the 50 states and the District of Columbia who are women: current cigarette smoking | BRFSS 2014-2024 | 95% CI −16.2 to −15.6 | 995,501 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| inactive_gap | 24.6 | percentage points of adults, lower in the highest income band | design-based difference between two weighted shares of respondents, Taylor-linearized 95% interval | adults in the 50 states and the District of Columbia in the highest and lowest of five household income bands: no leisure-time physical activity in the past 30 days | BRFSS 2014-2024 | 95% CI 24.3 to 24.9 | 1,949,580 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| inactive_poor | 38.2% | percent of adults | design-based weighted share of respondents, Taylor-linearized 95% interval | adults in the 50 states and the District of Columbia in the lowest of five household income bands: no leisure-time physical activity in the past 30 days | BRFSS 2014-2024 | 95% CI 37.9% to 38.5% | 537,736 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| inactive_rich | 13.6% | percent of adults | design-based weighted share of respondents, Taylor-linearized 95% interval | adults in the 50 states and the District of Columbia in the highest of five household income bands: no leisure-time physical activity in the past 30 days | BRFSS 2014-2024 | 95% CI 13.5% to 13.7% | 1,411,844 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| inc_missing | 19.4% | percent of adults | design-based weighted share of respondents, Taylor-linearized 95% interval; the share who did not know or refused to give a household income, or left it blank | adults in the 50 states and the District of Columbia | BRFSS 2014-2024 | 95% CI 19.3% to 19.5% | 4,790,742 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| low_share_20 | 15.8% | percent of adults | design-based weighted share of respondents, Taylor-linearized 95% interval; share in the lowest of five bands (bands are in nominal dollars) | adults in the 50 states and the District of Columbia who reported a household income band | 2020 | 95% CI 15.4% to 16.1% | 316,015 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| low_share_21 | 11.8% | percent of adults | design-based weighted share of respondents, Taylor-linearized 95% interval; share in the lowest of five bands (bands are in nominal dollars) | adults in the 50 states and the District of Columbia who reported a household income band | 2021 | 95% CI 11.6% to 12.1% | 338,675 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| n_adults | 4,790,742 | respondents | count of interviews in the public-use files | adults in the 50 states and the District of Columbia | 2014-2024 |   |   | none (count) | BRFSS |   |
| n_binge_bands | 3,336,677 | respondents | count of respondents behind the five binge drinking estimates by income band | adults in the 50 states and the District of Columbia with an income band and an answer to the binge drinking question | BRFSS 2015-2024 |   |   | none (count) | BRFSS |   |
| n_binge_years | 10 | survey years | definition: the binge drinking question is in the files from 2015 | BRFSS public-use files | 2015-2024 |   |   | none (count) | BRFSS |   |
| n_income_bands | 5 | income bands | definition: five bands common to all years | household income bands used in the chapter | 2014-2024 |   |   | none (count) | BRFSS |   |
| n_jur | 51 | states and the District of Columbia | count of jurisdictions in the analysis | the 50 states and the District of Columbia | 2014-2024 |   |   | none (count) | BRFSS | Guam, Puerto Rico and the Virgin Islands are in the BRFSS files and outside the analysis. |
| n_years | 11 | survey years | definition: survey years 2014 to 2024 | BRFSS public-use files | 2014-2024 |   |   | none (count) | BRFSS |   |
| obese8_150 | 30.1% | percent of adults | design-based weighted share of respondents, Taylor-linearized 95% interval | adults in the 50 states and the District of Columbia in the $150,000 to $199,999 band: obesity (BMI 30 or more, self-reported height and weight) | BRFSS 2021-2024 | 95% CI 29.5% to 30.8% | 88,636 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| obese8_top | 23.2% | percent of adults | design-based weighted share of respondents, Taylor-linearized 95% interval | adults in the 50 states and the District of Columbia in the $200,000 or more band: obesity (BMI 30 or more, self-reported height and weight) | BRFSS 2021-2024 | 95% CI 22.7% to 23.8% | 94,433 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| obese_gap | 7.6 | percentage points of adults, lower in the highest income band | design-based difference between two weighted shares of respondents, Taylor-linearized 95% interval | adults in the 50 states and the District of Columbia in the highest and lowest of five household income bands: obesity (BMI 30 or more, self-reported height and weight) | BRFSS 2014-2024 | 95% CI 7.2 to 7.9 | 1,878,633 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| obese_men_poor | 31.1% | percent of adults | design-based weighted share of respondents, Taylor-linearized 95% interval | adults in the 50 states and the District of Columbia who are men in the lowest of five household income bands: obesity (BMI 30 or more, self-reported height and weight) | BRFSS 2014-2024 | 95% CI 30.7% to 31.6% | 198,642 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| obese_men_rich | 30.9% | percent of adults | design-based weighted share of respondents, Taylor-linearized 95% interval | adults in the 50 states and the District of Columbia who are men in the highest of five household income bands: obesity (BMI 30 or more, self-reported height and weight) | BRFSS 2014-2024 | 95% CI 30.7% to 31.1% | 733,927 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| obese_mw_rich | 5.6 | percentage points of adults, higher among men than among women | design-based difference between two weighted shares of respondents, Taylor-linearized 95% interval | adults in the 50 states and the District of Columbia in the highest of five household income bands: obesity (BMI 30 or more, self-reported height and weight) | BRFSS 2014-2024 | 95% CI 5.2 to 5.9 | 1,365,521 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| obese_poor | 36.0% | percent of adults | design-based weighted share of respondents, Taylor-linearized 95% interval | adults in the 50 states and the District of Columbia in the lowest of five household income bands: obesity (BMI 30 or more, self-reported height and weight) | BRFSS 2014-2024 | 95% CI 35.7% to 36.3% | 512,939 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| obese_rich | 28.4% | percent of adults | design-based weighted share of respondents, Taylor-linearized 95% interval | adults in the 50 states and the District of Columbia in the highest of five household income bands: obesity (BMI 30 or more, self-reported height and weight) | BRFSS 2014-2024 | 95% CI 28.3% to 28.6% | 1,365,694 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| obese_steeper | 14.2 | percentage points steeper for women than for men (richest minus poorest band, women's difference minus men's, sign reversed) | difference between the women's and the men's income-band differences; standard errors added in quadrature (disjoint groups of respondents) | adults in the 50 states and the District of Columbia: obesity (BMI 30 or more, self-reported height and weight) | BRFSS 2014-2024 | 95% CI 13.5 to 14.8 | 1,878,277 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| obese_wm_poor | 8.6 | percentage points of adults, higher among women than among men | design-based difference between two weighted shares of respondents, Taylor-linearized 95% interval | adults in the 50 states and the District of Columbia in the lowest of five household income bands: obesity (BMI 30 or more, self-reported height and weight) | BRFSS 2014-2024 | 95% CI 8.0 to 9.2 | 512,756 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| obese_women_poor | 39.7% | percent of adults | design-based weighted share of respondents, Taylor-linearized 95% interval | adults in the 50 states and the District of Columbia who are women in the lowest of five household income bands: obesity (BMI 30 or more, self-reported height and weight) | BRFSS 2014-2024 | 95% CI 39.4% to 40.1% | 314,114 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| obese_women_rich | 25.3% | percent of adults | design-based weighted share of respondents, Taylor-linearized 95% interval | adults in the 50 states and the District of Columbia who are women in the highest of five household income bands: obesity (BMI 30 or more, self-reported height and weight) | BRFSS 2014-2024 | 95% CI 25.1% to 25.6% | 631,594 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| race_gap_black | 0.5 | percentage points of adults, higher in the highest income band | design-based difference between two weighted shares of respondents, Taylor-linearized 95% interval | Black non-Hispanic adults in the 50 states and the District of Columbia: binge drinking in the past 30 days | BRFSS 2015-2024 | 95% CI −0.3 to 1.2 | 120,974 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| race_gap_hisp | 10.3 | percentage points of adults, higher in the highest income band | design-based difference between two weighted shares of respondents, Taylor-linearized 95% interval | Hispanic adults in the 50 states and the District of Columbia: binge drinking in the past 30 days | BRFSS 2015-2024 | 95% CI 9.3 to 11.2 | 119,344 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| race_gap_multi | 5.5 | percentage points of adults, higher in the highest income band | design-based difference between two weighted shares of respondents, Taylor-linearized 95% interval | multiracial non-Hispanic adults in the 50 states and the District of Columbia: binge drinking in the past 30 days | BRFSS 2015-2024 | 95% CI 3.7 to 7.3 | 35,796 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| race_gap_other | 1.0 | percentage points of adults, higher in the highest income band | design-based difference between two weighted shares of respondents, Taylor-linearized 95% interval | adults of another race, non-Hispanic in the 50 states and the District of Columbia: binge drinking in the past 30 days | BRFSS 2015-2024 | 95% CI 0.0 to 2.1 | 86,080 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| race_gap_white | 8.6 | percentage points of adults, higher in the highest income band | design-based difference between two weighted shares of respondents, Taylor-linearized 95% interval | White non-Hispanic adults in the 50 states and the District of Columbia: binge drinking in the past 30 days | BRFSS 2015-2024 | 95% CI 8.3 to 8.9 | 1,319,140 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| ratio_poor | 1.9 | ratio of rates, men to women | ratio of two weighted shares from different groups of respondents, log-scale interval from the two standard errors in quadrature | adults in the 50 states and the District of Columbia in the lowest of five household income bands: binge drinking in the past 30 days | BRFSS 2015-2024 | 95% CI 1.8 to 2.0 | 450,603 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| ratio_rich | 1.5 | ratio of rates, men to women | ratio of two weighted shares from different groups of respondents, log-scale interval from the two standard errors in quadrature | adults in the 50 states and the District of Columbia in the highest of five household income bands: binge drinking in the past 30 days | BRFSS 2015-2024 | 95% CI 1.5 to 1.6 | 1,254,385 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| sexgap_max | 9.2 | percentage points of adults, higher among men than among women | design-based difference between two weighted shares of respondents, Taylor-linearized 95% interval | adults in the 50 states and the District of Columbia in the income band with the largest difference: binge drinking in the past 30 days | BRFSS 2015-2024 | 95% CI 8.7 to 9.6 | 613,943 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| sexgap_min | 8.5 | percentage points of adults, higher among men than among women | design-based difference between two weighted shares of respondents, Taylor-linearized 95% interval | adults in the 50 states and the District of Columbia in the income band with the smallest difference: binge drinking in the past 30 days | BRFSS 2015-2024 | 95% CI 8.2 to 8.8 | 1,254,385 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| smoke8_75 | 9.7% | percent of adults | design-based weighted share of respondents, Taylor-linearized 95% interval | adults in the 50 states and the District of Columbia in the $75,000 to $99,999 band: current cigarette smoking | BRFSS 2021-2024 | 95% CI 9.4% to 10.0% | 189,175 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| smoke8_top | 4.3% | percent of adults | design-based weighted share of respondents, Taylor-linearized 95% interval | adults in the 50 states and the District of Columbia in the $200,000 or more band: current cigarette smoking | BRFSS 2021-2024 | 95% CI 4.0% to 4.6% | 95,390 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| smoke_coll_poor | 13.6% | percent of adults | design-based weighted share of respondents, Taylor-linearized 95% interval | adults in the 50 states and the District of Columbia who graduated from college or technical school in the lowest of five household income bands: current cigarette smoking | BRFSS 2014-2024 | 95% CI 13.1% to 14.1% | 70,037 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| smoke_coll_rich | 4.4% | percent of adults | design-based weighted share of respondents, Taylor-linearized 95% interval | adults in the 50 states and the District of Columbia who graduated from college or technical school in the highest of five household income bands: current cigarette smoking | BRFSS 2014-2024 | 95% CI 4.4% to 4.5% | 885,429 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| smoke_edu_coll | 5.8% | percent of adults | design-based weighted share of respondents, Taylor-linearized 95% interval | adults in the 50 states and the District of Columbia who graduated from college or technical school: current cigarette smoking | BRFSS 2014-2024 | 95% CI 5.8% to 5.9% | 1,783,216 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| smoke_edu_nohs | 24.7% | percent of adults | design-based weighted share of respondents, Taylor-linearized 95% interval | adults in the 50 states and the District of Columbia who did not finish high school: current cigarette smoking | BRFSS 2014-2024 | 95% CI 24.4% to 25.0% | 301,438 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| smoke_gap | 17.2 | percentage points of adults, lower in the highest income band | design-based difference between two weighted shares of respondents, Taylor-linearized 95% interval | adults in the 50 states and the District of Columbia in the highest and lowest of five household income bands: current cigarette smoking | BRFSS 2014-2024 | 95% CI 16.9 to 17.4 | 1,924,406 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| smoke_gap_coll | 9.1 | percentage points of adults, lower in the highest income band | design-based difference between two weighted shares of respondents, Taylor-linearized 95% interval | adults in the 50 states and the District of Columbia who graduated from college or technical school: current cigarette smoking | BRFSS 2014-2024 | 95% CI 8.7 to 9.6 | 955,466 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| smoke_gap_nohs | 4.8 | percentage points of adults, lower in the highest income band | design-based difference between two weighted shares of respondents, Taylor-linearized 95% interval | adults in the 50 states and the District of Columbia who did not finish high school: current cigarette smoking | BRFSS 2014-2024 | 95% CI 3.4 to 6.2 | 122,226 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| smoke_men_all | 16.5% | percent of adults | design-based weighted share of respondents, Taylor-linearized 95% interval | adults in the 50 states and the District of Columbia who are men: current cigarette smoking | BRFSS 2014-2024 | 95% CI 16.4% to 16.6% | 2,049,062 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| smoke_men_hi | 21.9% | percent of adults | design-based weighted share of respondents, Taylor-linearized 95% interval; the highest rate of the five race and ethnicity groups | adults in the 50 states and the District of Columbia who are men and multiracial non-Hispanic: current cigarette smoking | BRFSS 2014-2024 | 95% CI 21.2% to 22.7% | 44,344 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| smoke_men_hisp | 15.8% | percent of adults | design-based weighted share of respondents, Taylor-linearized 95% interval | adults in the 50 states and the District of Columbia who are men and Hispanic: current cigarette smoking | BRFSS 2014-2024 | 95% CI 15.4% to 16.1% | 161,479 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| smoke_men_lo | 13.5% | percent of adults | design-based weighted share of respondents, Taylor-linearized 95% interval; the lowest rate of the five race and ethnicity groups | adults in the 50 states and the District of Columbia who are men and of another race, non-Hispanic: current cigarette smoking | BRFSS 2014-2024 | 95% CI 13.1% to 14.0% | 107,849 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| smoke_men_white | 16.3% | percent of adults | design-based weighted share of respondents, Taylor-linearized 95% interval | adults in the 50 states and the District of Columbia who are men and White non-Hispanic: current cigarette smoking | BRFSS 2014-2024 | 95% CI 16.1% to 16.4% | 1,553,398 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| smoke_nohs_poor | 28.0% | percent of adults | design-based weighted share of respondents, Taylor-linearized 95% interval | adults in the 50 states and the District of Columbia who did not finish high school in the lowest of five household income bands: current cigarette smoking | BRFSS 2014-2024 | 95% CI 27.5% to 28.5% | 107,166 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| smoke_nohs_rich | 23.2% | percent of adults | design-based weighted share of respondents, Taylor-linearized 95% interval | adults in the 50 states and the District of Columbia who did not finish high school in the highest of five household income bands: current cigarette smoking | BRFSS 2014-2024 | 95% CI 21.9% to 24.5% | 15,060 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| smoke_poor | 25.7% | percent of adults | design-based weighted share of respondents, Taylor-linearized 95% interval | adults in the 50 states and the District of Columbia in the lowest of five household income bands: current cigarette smoking | BRFSS 2014-2024 | 95% CI 25.4% to 25.9% | 531,816 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| smoke_rich | 8.5% | percent of adults | design-based weighted share of respondents, Taylor-linearized 95% interval | adults in the 50 states and the District of Columbia in the highest of five household income bands: current cigarette smoking | BRFSS 2014-2024 | 95% CI 8.4% to 8.6% | 1,392,590 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| smoke_sexgap_black | 7.0 | percentage points of adults, higher among men than among women | design-based difference between two weighted shares of respondents, Taylor-linearized 95% interval | Black non-Hispanic adults: current cigarette smoking | BRFSS 2014-2024 | 95% CI 6.5 to 7.4 | 345,930 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| smoke_sexgap_hisp | 8.1 | percentage points of adults, higher among men than among women | design-based difference between two weighted shares of respondents, Taylor-linearized 95% interval | Hispanic adults: current cigarette smoking | BRFSS 2014-2024 | 95% CI 7.7 to 8.5 | 346,545 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| smoke_sexgap_multi | 3.3 | percentage points of adults, higher among men than among women | design-based difference between two weighted shares of respondents, Taylor-linearized 95% interval | multiracial non-Hispanic adults: current cigarette smoking | BRFSS 2014-2024 | 95% CI 2.2 to 4.4 | 93,285 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| smoke_sexgap_other | 5.9 | percentage points of adults, higher among men than among women | design-based difference between two weighted shares of respondents, Taylor-linearized 95% interval | adults of another race, non-Hispanic: current cigarette smoking | BRFSS 2014-2024 | 95% CI 5.4 to 6.4 | 215,106 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| smoke_sexgap_poor | 6.7 | percentage points of adults, higher among men than among women | design-based difference between two weighted shares of respondents, Taylor-linearized 95% interval | adults in the 50 states and the District of Columbia in the lowest of five household income bands: current cigarette smoking | BRFSS 2014-2024 | 95% CI 6.2 to 7.2 | 531,613 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| smoke_sexgap_rich | 2.9 | percentage points of adults, higher among men than among women | design-based difference between two weighted shares of respondents, Taylor-linearized 95% interval | adults in the 50 states and the District of Columbia in the highest of five household income bands: current cigarette smoking | BRFSS 2014-2024 | 95% CI 2.7 to 3.1 | 1,392,389 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| smoke_sexgap_white | 1.7 | percentage points of adults, higher among men than among women | design-based difference between two weighted shares of respondents, Taylor-linearized 95% interval | White non-Hispanic adults: current cigarette smoking | BRFSS 2014-2024 | 95% CI 1.6 to 1.9 | 3,452,292 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| smoke_steeper_men | 3.8 | percentage points steeper for men than for women (richest minus poorest band, women's difference minus men's) | difference between the women's and the men's income-band differences; standard errors added in quadrature (disjoint groups of respondents) | adults in the 50 states and the District of Columbia: current cigarette smoking | BRFSS 2014-2024 | 95% CI 3.2 to 4.3 | 1,924,002 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| smoke_women_all | 12.8% | percent of adults | design-based weighted share of respondents, Taylor-linearized 95% interval | adults in the 50 states and the District of Columbia who are women: current cigarette smoking | BRFSS 2014-2024 | 95% CI 12.7% to 12.9% | 2,492,615 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| smoke_women_hi | 18.6% | percent of adults | design-based weighted share of respondents, Taylor-linearized 95% interval; the highest rate of the five race and ethnicity groups | adults in the 50 states and the District of Columbia who are women and multiracial non-Hispanic: current cigarette smoking | BRFSS 2014-2024 | 95% CI 17.9% to 19.4% | 48,941 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| smoke_women_hisp | 7.6% | percent of adults | design-based weighted share of respondents, Taylor-linearized 95% interval | adults in the 50 states and the District of Columbia who are women and Hispanic: current cigarette smoking | BRFSS 2014-2024 | 95% CI 7.4% to 7.9% | 185,066 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| smoke_women_lo | 7.6% | percent of adults | design-based weighted share of respondents, Taylor-linearized 95% interval; the lowest rate of the five race and ethnicity groups | adults in the 50 states and the District of Columbia who are women and Hispanic: current cigarette smoking | BRFSS 2014-2024 | 95% CI 7.4% to 7.9% | 185,066 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| smoke_women_white | 14.5% | percent of adults | design-based weighted share of respondents, Taylor-linearized 95% interval | adults in the 50 states and the District of Columbia who are women and White non-Hispanic: current cigarette smoking | BRFSS 2014-2024 | 95% CI 14.4% to 14.6% | 1,898,894 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| state_max | 13.1 | percentage points of adults, higher in the highest income band | design-based difference between two weighted shares of respondents, Taylor-linearized 95% interval; the jurisdiction with the largest difference | adults in the District of Columbia: binge drinking in the past 30 days | BRFSS 2015-2024 | 95% CI 10.7 to 15.5 | 18,026 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| state_md | 5.7 | percentage points of adults, higher in the highest income band | design-based difference between two weighted shares of respondents, Taylor-linearized 95% interval | adults in Maryland: binge drinking in the past 30 days | BRFSS 2015-2024 | 95% CI 4.4 to 6.9 | 68,318 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| state_min | 2.7 | percentage points of adults, higher in the highest income band | design-based difference between two weighted shares of respondents, Taylor-linearized 95% interval; the jurisdiction with the smallest difference | adults in Utah: binge drinking in the past 30 days | BRFSS 2015-2024 | 95% CI 1.6 to 3.7 | 46,474 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| state_n_pos | 51 | states and the District of Columbia | count of jurisdictions where the binge drinking rate is higher in the highest income band than in the lowest | the 50 states and the District of Columbia | BRFSS 2015-2024 |   |   | none (count) | BRFSS |   |
| state_n_sig | 51 | states and the District of Columbia | count of jurisdictions where the 95% interval of that difference is above zero | the 50 states and the District of Columbia | BRFSS 2015-2024 |   |   | none (count) | BRFSS |   |
| state_smoke_hi | 29.9 | percentage points of adults, lower in the highest income band | design-based difference between two weighted shares of respondents, Taylor-linearized 95% interval; the jurisdiction with the largest difference | adults in Alaska: current cigarette smoking | BRFSS 2014-2024 | 95% CI 27.2 to 32.6 | 21,314 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| state_smoke_lo | 9.1 | percentage points of adults, lower in the highest income band | design-based difference between two weighted shares of respondents, Taylor-linearized 95% interval; the jurisdiction with the smallest difference | adults in California: current cigarette smoking | BRFSS 2014-2024 | 95% CI 8.2 to 9.9 | 51,083 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| state_smoke_n_neg | 51 | states and the District of Columbia | count of jurisdictions where the 95% interval of the smoking difference (highest minus lowest income band) is below zero | the 50 states and the District of Columbia | BRFSS 2014-2024 |   |   | none (count) | BRFSS |   |
| top_share_14 | 30.2% | percent of adults | design-based weighted share of respondents, Taylor-linearized 95% interval; share in the highest of five bands (bands are in nominal dollars) | adults in the 50 states and the District of Columbia who reported a household income band | 2014 | 95% CI 29.9% to 30.5% | 386,053 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| top_share_24 | 46.1% | percent of adults | design-based weighted share of respondents, Taylor-linearized 95% interval; share in the highest of five bands (bands are in nominal dollars) | adults in the 50 states and the District of Columbia who reported a household income band | 2024 | 95% CI 45.7% to 46.4% | 364,440 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| year_max | 8.3 | percentage points of adults, higher in the highest income band | design-based difference between two weighted shares of respondents, Taylor-linearized 95% interval; the survey year with the largest difference | adults in the 50 states and the District of Columbia, survey year 2022: binge drinking in the past 30 days | 2022 | 95% CI 7.3 to 9.2 | 164,332 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| year_min | 6.7 | percentage points of adults, higher in the highest income band | design-based difference between two weighted shares of respondents, Taylor-linearized 95% interval; the survey year with the smallest difference | adults in the 50 states and the District of Columbia, survey year 2024: binge drinking in the past 30 days | 2024 | 95% CI 5.8 to 7.7 | 182,498 respondents | BRFSS final weight (xllcpwt); stratum and primary sampling unit for the interval | BRFSS |   |
| year_n_pos | 10 | survey years | count of survey years in which the 95% interval of the binge drinking difference (highest minus lowest income band) is above zero | adults in the 50 states and the District of Columbia | BRFSS 2015-2024 |   |   | none (count) | BRFSS |   |

## Definitions

- Modeled estimate: A figure produced by a statistical model instead of a direct count or measurement, usually for small areas where direct data are thin, borrowing strength from surveys, neighboring areas or other data. Neighboring areas can share errors, so single extremes call for caution.
- Ecological inference: Drawing conclusions about individuals from data about places or groups. A pattern across counties need not hold for the people inside them.
- Survey weights: Numbers attached to survey respondents that say how many people in the population each one represents, so that results describe the whole population and not only the people who answered.

## Methods

These estimates compare adults interviewed in successive annual surveys. Survey weights account for how the sample represents the covered population, and the uncertainty calculation uses the survey design. Different questions and income comparisons use different eligible subsets; the interviews do not follow the same adults over time. The public files locate a respondent no more finely than the state, so none of these comparisons is a neighborhood comparison.

### Who was interviewed

Estimates use the CDC Behavioral Risk Factor Surveillance System (BRFSS) public-use files for 2014 to 2024, 4,874,919 respondents in all, of whom 4,790,742 are adults in the 50 states and the District of Columbia; the 84,177 respondents in Guam, Puerto Rico and the Virgin Islands are left out.

The survey is a repeated cross-section, not a panel, and its geography stops at the state (metro status is missing for most respondents).

Explains Table 3.

### What the behavior measures mean

Binge drinking is the BRFSS calculated variable for five or more drinks for men or four or more for women on one occasion in the past 30 days.

Current smoking, obesity (BMI of 30 or more from self-reported height and weight), no leisure-time physical activity, fair or poor health and race and ethnicity are BRFSS calculated variables; diabetes is a reported diagnosis, prediabetes counts as no diabetes, and diabetes only during pregnancy leaves the denominator.

Explains Table 5 and Table 7.

### Which years each question covers

Binge drinking is in the files from 2015, so its estimates cover 2015 to 2024 and the other measures cover 2014 to 2024.

### Making income bands comparable

Household income was recorded in eight categories through 2020 and in eleven from 2021, which split the top category. The five bands group categories that exist in every year (under $20,000, $20,000 to $34,999, $35,000 to $49,999, $50,000 to $74,999, $75,000 or more) and the eight-band view uses the 2021 to 2024 categories. Respondents who did not know or refused to give income are left out of the bands.

Gradients are the rate in the highest income band minus the rate in the lowest. The 2021 change in the income question is checked by estimating the binge difference separately for 2015 to 2020 and 2021 to 2024.

Explains Table 1 and Table 4.

### Weights and uncertainty

Every estimate is a weighted share with the BRFSS final weight, computed on the full frame with the group or domain as a subpopulation, so that strata and sampling units outside a group keep their place in the variance.

A cell is published only with 30 or more respondents. Strata with a single sampling unit are merged with a neighbor in the same state and year.

Intervals are 95% intervals from Taylor linearization with the survey's strata (made year-specific) and primary sampling units; shares use a logit interval, differences a normal interval.

In the public files each respondent is a primary sampling unit, so two groups of respondents are independent and the standard error of a difference of two group estimates is the two standard errors added in quadrature; the build checks this against the design-based gap estimator to within 0.03%.

## What these data cannot show

Every measure in this chapter is reported by the respondent in a telephone interview, so the estimates describe what adults say. Respondents may under-report their drinking and their weight, and the data cannot show whether that differs by income, sex or race. The 2024 survey year is provisional. The intervals reflect the survey's sampling design and weights. They do not cover the adults who could not be reached or declined to take part, or the 19.4% of adults who gave no household income and are left out of every band.

The data are a repeated cross-section, not a panel. A person in the highest band and a person in the lowest are different people, so the gradients say how behavior differs across incomes and not how it changes when one person's income changes. The bands are household income in nominal dollars, not adjusted for household size or for prices. The highest band holds 46.1% of respondents who gave an income in 2024 against 30.2% in 2014, and part of that shift is a step at the 2021 change in the income question, so pooling the years mixes survey years with different bands. The binge drinking difference between the highest band and the lowest is similar before and after that change. Income, age, schooling and race are related. The age, education and race and ethnicity panels compare within one of them at a time and only for binge drinking, and the other figures are not age-adjusted.

The groups are those BRFSS defines. The Hispanic group may be of any race, and the other-race group combines populations whose drinking and smoking differ. Binge drinking is a measure of quantity on one occasion, and a person who binges once in a month is counted with a person who binges every week. BRFSS locates respondents by state and, for some respondents, by metro status, which is missing for a large share of respondents and is not used here, so the results say nothing about neighborhoods. The person-level results here and the tract-level results of Chapter 7.1, Health behaviors and the supply of care (https://vishalsingh.org/zipcode/health-behaviors-and-the-supply-of-care) come from different measures, survey responses in one and modeled estimates in the other, so neither confirms the other. Whether the drinking measured here relates to the alcohol-related deaths mapped in Chapter 8.2, Overdose, alcohol and suicide (https://vishalsingh.org/zipcode/overdose-alcohol-and-suicide) is outside what BRFSS can show.

## References

1. Centers for Disease Control and Prevention (2023). Calculated Variables in the 2022 BRFSS Data File. Behavioral Risk Factor Surveillance System, technical documentation, 29 September 2023. https://www.cdc.gov/brfss/annual_data/2022/pdf/2022-calculated-variables-version4-508.pdf
2. Bohm, Michele K.; Liu, Yong; Esser, Marissa B.; Mesnick, Jessica B.; Lu, Hua; Pan, Yi; Greenlund, Kurt J. (2021). Binge Drinking Among Adults, by Select Characteristics and State - United States, 2018. MMWR Morbidity and Mortality Weekly Report 70(41): 1441-1446. https://doi.org/10.15585/mmwr.mm7041a2
3. Collins, Susan E. (2016). Associations Between Socioeconomic Factors and Alcohol Outcomes. Alcohol Research: Current Reviews 38(1): 83-94. https://doi.org/10.35946/arcr.v38.1.11
4. Keyes, Katherine M.; Hasin, Deborah S. (2008). Socio-economic status and problem alcohol use: the positive relationship between income and the DSM-IV alcohol abuse diagnosis. Addiction 103(7): 1120-1130. https://doi.org/10.1111/j.1360-0443.2008.02218.x
5. Sobal, Jeffery; Stunkard, Albert J. (1989). Socioeconomic status and obesity: A review of the literature. Psychological Bulletin 105(2): 260-275. https://doi.org/10.1037/0033-2909.105.2.260
6. McLaren, L. (2007). Socioeconomic Status and Obesity. Epidemiologic Reviews 29: 29-48. https://doi.org/10.1093/epirev/mxm001
7. Lopez, A. D.; Collishaw, N. E.; Piha, T. (1994). A descriptive model of the cigarette epidemic in developed countries. Tobacco Control 3(3): 242-247. https://doi.org/10.1136/tc.3.3.242
8. Pampel, Fred C. (2006). Global Patterns and Determinants of Sex Differences in Smoking. International Journal of Comparative Sociology 47(6): 466-487. https://doi.org/10.1177/0020715206070267
9. Apouey, Benedicte; Clark, Andrew E. (2015). Winning big but feeling no better? The effect of lottery prizes on physical and mental health. Health Economics 24(5): 516-538. https://doi.org/10.1002/hec.3035

## Provenance

- Pipeline version: 0.1.0
- Build script: book2/chapters/health-behaviors-person-by-person/build.py (sha256 841ecebcc35363bb5a1dd456307d73de44c0b27f8c25cd4fc773e3cabe4ddab8)
- Built on: 2026-10-07
- Site commit: c2b1ff3309cb4a10833f884da91ac0057a66ad0a

Inputs:

| File | Size (bytes) | SHA-256, first 1 MB | SHA-256, last 1 MB |
| --- | ---: | --- | --- |
| book2/chapters/health-behaviors-person-by-person/compute.py | 6,233 | 255c772d43009a45081e2bd14aab8555040b209143e24ef49533dd650cb5b31d | 255c772d43009a45081e2bd14aab8555040b209143e24ef49533dd650cb5b31d |
| book2/chapters/health-behaviors-person-by-person/survey.py | 7,022 | de7a95901b95620dd0b52eeb4dd2330496e5508893f76c99bc092b9f0cc76eec | de7a95901b95620dd0b52eeb4dd2330496e5508893f76c99bc092b9f0cc76eec |

Outputs:

| File | SHA-256 |
| --- | --- |
| facts.json | 57ddf5dd12fa5810873f9a306edf28e7911d21de5b1eb9503cdbe624c4742d4b |
| figures/behaviors_by_income.json | 5bcba1357eda7c97bcd11b599dbf66f1dd494a61d92e2a03d18df287c8c8eae1 |
| figures/binge_within_groups.json | ada479126f6e66cd1b0cc4d3fa2063f300de915b299e01c63f14236b22dcc7c5 |
| figures/gradient_by_behavior_and_sex.json | c56eb60c776db16ee7cf2d91ac3162288059152383fc41dcfc96d5f8367ab227 |
| figures/gradient_by_state.json | 5bf8b8de803954b95d7065370f3824faa39add297075757affe6b8108e4b7962 |
| figures/sex_by_income.json | 97f9b81291dbd91dd17517f7c95b4bc46e90970d03b43bcb33cca658e005cec0 |
| figures/smoking_by_education_and_income.json | e8923099e4a2e2015405a66015b8a147d0700a41937bad4ed9c58eec2d31bda8 |
| figures/smoking_gap_by_race.json | 44681fc98c22a3ce902383c1d70aa4d6b46d0bc2826b7587a0b0a08ed421cb27 |
| meta.json | 5680f2b6c400e4ee9a4a1436151dce75b158f3711e030184e05e44d9a0f60f25 |
| refs.json | bd44e4f82a27501a2097869bf4cdda64ef6c1ae34a6d541abe2338b210794bb4 |

## Machine formats

- Report (HTML): https://vishalsingh.org/zipcode/health-behaviors-person-by-person/report
- Report (Markdown): https://vishalsingh.org/zipcode/health-behaviors-person-by-person/report.md
- Every estimate (JSON): https://vishalsingh.org/zipcode/report/health-behaviors-person-by-person/facts.json
- Table 1, figure "behaviors_by_income" (JSON): https://vishalsingh.org/zipcode/report/health-behaviors-person-by-person/figures/behaviors_by_income.json
- Table 2, figure "binge_within_groups" (JSON): https://vishalsingh.org/zipcode/report/health-behaviors-person-by-person/figures/binge_within_groups.json
- Table 3, figure "gradient_by_state" (JSON): https://vishalsingh.org/zipcode/report/health-behaviors-person-by-person/figures/gradient_by_state.json
- Table 4, figure "sex_by_income" (JSON): https://vishalsingh.org/zipcode/report/health-behaviors-person-by-person/figures/sex_by_income.json
- Table 5, figure "gradient_by_behavior_and_sex" (JSON): https://vishalsingh.org/zipcode/report/health-behaviors-person-by-person/figures/gradient_by_behavior_and_sex.json
- Table 6, figure "smoking_by_education_and_income" (JSON): https://vishalsingh.org/zipcode/report/health-behaviors-person-by-person/figures/smoking_by_education_and_income.json
- Table 7, figure "smoking_gap_by_race" (JSON): https://vishalsingh.org/zipcode/report/health-behaviors-person-by-person/figures/smoking_gap_by_race.json
- Index of every report: https://vishalsingh.org/zipcode/llms.txt; all reports in one file: https://vishalsingh.org/zipcode/llms-full.txt

---

# Multiple chronic conditions: report

Chapter 7.3 of ZIP Code Destiny, by Vishal Singh (NYU Stern). Generated from the chapter's data files; the reading version is at https://vishalsingh.org/zipcode/multiple-chronic-conditions. Table n of this report is figure n of the chapter.

## Identification

- Title: Multiple chronic conditions
- Label: Where illness keeps company
- Chapter: 7.3 (Part VII, Health and care; An American life)
- Question: Which chronic conditions are high in the same neighborhoods, and how does that overlap change with deprivation?
- Dek: Weighted by adults, census tracts in the most deprived tenth were high on 5.7 of nine modeled chronic-condition prevalences on average in 2022, against 0.5 in the least deprived tenth, a count for tracts and not people.
- Grain: neighborhoods
- Status: published
- Published: 2026-10-09
- Updated: not revised
- Data sources:
  - CDC PLACES. CDC PLACES 2024 release, census tracts, crude prevalence among adults of nine conditions (BRFSS 2022); years: 2024 release (tracts); 2025 release (ZCTAs, 2023 data year); weights: adult population of the tract
  - Area Deprivation Index. Neighborhood Atlas Area Deprivation Index, 2023, national percentile; years: 2023 release (version 4.0.1; built on ACS 2019-2023); weights: adult population of the tract
  - American Community Survey. ACS 2019-2023 tract shares used to approximate the model's inputs; years: 2019-2023 (5-year); weights: adult population of the tract
  - USALEEP life expectancy. NCHS small-area life expectancy at birth, 2010-2015, 2020 tracts; years: 2010-2015; weights: ACS 2019-2023 tract population
  - Published estimates from cited studies. Estimates quoted in the causal box (Ludwig and coauthors 2011); years: as reported in each study; weights: as reported in the study
- Canonical URL: https://vishalsingh.org/zipcode/multiple-chronic-conditions
- Report: https://vishalsingh.org/zipcode/multiple-chronic-conditions/report
- Markdown: https://vishalsingh.org/zipcode/multiple-chronic-conditions/report.md
- BibTeX key: singh2026multiplechronicconditions

```bibtex
@misc{singh2026multiplechronicconditions,
  author = {Singh, Vishal},
  title = {Multiple chronic conditions},
  year = {2026},
  url = {https://vishalsingh.org/zipcode/multiple-chronic-conditions},
  note = {In ZIP Code Destiny. Published 9 October 2026.}
}
```

## Principal estimates

The facts the dek and the opening prose cite, in order of appearance.

| Estimand | Population | Period | Value | Interval | n | Weights | Source | Fact id |
| --- | --- | --- | ---: | --- | ---: | --- | --- | --- |
| mean number of the nine conditions that are high, population-weighted mean over a tenth of tracts, bootstrap interval over counties | adults in the most deprived tenth of tracts by ADI national percentile | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 5.7 | 95% CI 5.5 to 5.8 | 8,213 tracts | adult population of the tract (PLACES) | CDC PLACES | cnt_d10 |
| mean number of the nine conditions that are high, population-weighted mean over a tenth of tracts, bootstrap interval over counties | adults in the least deprived tenth of tracts by ADI national percentile | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 0.5 | 95% CI 0.4 to 0.6 | 8,274 tracts | adult population of the tract (PLACES) | CDC PLACES | cnt_d1 |
| definition: a tract is high on a condition when its prevalence is in the highest 25% of tracts, ties at their mean rank | census tracts with an estimate for all nine conditions and an ADI rank | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 25% |   |   | none (a definition) | CDC PLACES | cut_pct |
| difference between the population-weighted mean count of high conditions in the most and the least deprived tenth of tracts, bootstrap interval over counties (draws shared) | adults in the most deprived tenth of tracts compared with the least deprived tenth | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 5.2 | 95% CI 5.0 to 5.4 | 82,781 tracts | adult population of the tract (PLACES) | CDC PLACES | cnt_diff |
| count of tracts | census tracts with an estimate for all nine conditions and an ADI rank | BRFSS 2022 (PLACES 2024 release); ADI 2023; ACS 2019-2023 | 82,781 |   |   | none (full count) | CDC PLACES | n_tracts |
| mean of the population-weighted Pearson correlations over the 36 pairs of conditions, bootstrap interval over counties | adults living in census tracts with an estimate for all nine conditions and an ADI rank | BRFSS 2022 (PLACES 2024 release); ADI 2023 | +0.49 | 95% CI 0.48 to 0.51 | 82,781 tracts | adult population of the tract (PLACES) | CDC PLACES | r_mean |
| count of measure ids in the staged tract file | health measures that PLACES publishes for tracts | PLACES 2024 release | 40 |   |   | none (full count) | CDC PLACES | n_places_measures |
| definition: the number of conditions in the count | PLACES health outcome measures that cover all adults and have an estimate in every state | PLACES 2024 release | 9 |   |   | none (a definition) | CDC PLACES | n_conditions |

## Findings

Weighted by adults, a tract in the most deprived tenth of census tracts is in the highest 25% of all tracts for the prevalence of 5.7 (95% CI 5.5 to 5.8) of nine chronic conditions on average, against 0.5 (95% CI 0.4 to 0.6) in the least deprived tenth, a difference of 5.2 (95% CI 5.0 to 5.4) conditions. Across 82,781 tracts, the average correlation between the prevalences of two of the conditions is +0.49 (95% CI 0.48 to 0.51). This chapter asks which conditions are high in the same neighborhoods and how that overlap changes with deprivation. It is a chapter about places and says nothing about how many conditions any one person has.

The data are modeled estimates from CDC PLACES, which gives each tract one prevalence for each condition among adults and holds no record of any person.[1] In surveys that ask each adult about each condition, multiple chronic conditions means having two or more of them,[2] and a tract estimate cannot be turned into that count. A tract can be high on several conditions while no resident has more than one, and a tract can be high on none while some residents have several. A count of conditions in this chapter is a count of tract prevalences above a stated cutoff.

PLACES publishes 40 measures for tracts. The 9 conditions used here are the health outcome measures that cover all adults and have a tract estimate in every state: arthritis, cancer (non-skin) or melanoma, current asthma, coronary heart disease, COPD, depression, diagnosed diabetes, obesity and stroke. PLACES lists obesity, a measure of body weight, with the health outcomes. High blood pressure and high cholesterol (among adults ever screened) are left out because the 2024 release has no tract estimate of either for Florida. The last section gives the number of tracts and a check on high blood pressure outside Florida.

A tract is high on a condition when its prevalence is in the highest 25% of tracts, with each tract counted once, and the count of a tract is the number of the nine conditions on which it is high. Means, correlations and shares of adults are population-weighted by the adults in each tract, and counts and shares of tracts count each tract once. Deprivation is ranked on the Area Deprivation Index (ADI), a national percentile built from American Community Survey (ACS) data.[3]

 (Note: The PLACES 2024 release uses survey responses from 2022, and from 2021 for blood pressure. Values are crude prevalence among adults, so the age structure of a tract moves them. The tract file carries no intervals, and the bands in this chapter come from resampling counties.)

### Stroke, diabetes, COPD and heart disease are high in the same tracts

The strongest correlations join stroke, diabetes, COPD, coronary heart disease and arthritis (Table 1). Across tracts the prevalence of stroke correlates +0.90 (95% CI 0.89 to 0.92) with diabetes, +0.90 (95% CI 0.89 to 0.90) with COPD and +0.86 (95% CI 0.85 to 0.87) with heart disease, and the pairs among stroke, diabetes, COPD and heart disease run from +0.74 to +0.90.

Arthritis correlates +0.90 (95% CI 0.89 to 0.91) with heart disease. At the other end, cancer and depression have the weakest correlations with the rest, and depression correlates only +0.06 (95% CI 0.01 to 0.11) with diabetes. The mean correlation of cancer with the other eight conditions is +0.22 (95% CI 0.19 to 0.25) and that of depression is +0.26 (95% CI 0.22 to 0.30), against +0.68 (95% CI 0.66 to 0.69) for COPD.

Of the 36 pairs, 32 have an interval entirely above zero and 2 have one entirely below it, cancer with asthma (−0.13 (95% CI −0.20 to −0.07)) and cancer with obesity (−0.11 (95% CI −0.18 to −0.05)).

One principal component accounts for 59.0% (95% CI 57.9% to 60.2%) of the variance of the nine standardized prevalences, so most of the difference between tracts in these conditions follows a single shared pattern. Every correlation here is between modeled estimates and is partly built in by the inputs the model shares across conditions, a point the section on the model's inputs takes up.

Table 1. Stroke, diabetes, COPD and heart disease are high in the same tracts, with correlations from +0.74 to +0.90, while cancer correlates −0.13 with asthma

The correlations of cancer follow the age of the residents. PLACES gives crude prevalence, and across tracts the prevalence of cancer correlates +0.81 (95% CI 0.79 to 0.83) with the ACS share of residents aged 65 and over, against +0.64 (95% CI 0.61 to 0.66) for arthritis, +0.64 (95% CI 0.61 to 0.66) for heart disease, +0.23 (95% CI 0.19 to 0.26) for diabetes and −0.17 (95% CI −0.21 to −0.14) for depression.

Cancer correlates 0.00 (95% CI −0.07 to 0.07) with the ADI. It correlates +0.74 (95% CI 0.72 to 0.75) with arthritis and +0.63 (95% CI 0.60 to 0.66) with heart disease, the two other conditions that line up with age, and close to zero with diabetes (+0.05 (95% CI 0.00 to 0.11)) and depression (0.00 (95% CI −0.07 to 0.06)). Two features of a tract therefore line up with the conditions. One is the age of its residents, which lines up with cancer, arthritis and heart disease. The other is deprivation, taken up in the sections below.

Asthma correlates +0.64 (95% CI 0.59 to 0.69) with obesity. The modeled asthma estimates are set against air pollution in Chapter 6.1, Air pollution (https://vishalsingh.org/zipcode/air-pollution), and here asthma is one of nine conditions.

The same pattern can be counted. For the tracts that are high on stroke, the mean number of the other eight conditions that are also high is 4.9 (95% CI 4.8 to 5.0), and for tracts high on COPD it is 4.9 (95% CI 4.9 to 5.0) (Table 2). For tracts high on depression it is 3.1 (95% CI 3.0 to 3.3) and for tracts high on cancer 2.5 (95% CI 2.4 to 2.6). If the conditions were unrelated across tracts, each mean would be 2.0. Weighting tracts by adults instead of counting each once moves each mean by 0.34 conditions or less.

Table 2. Tracts high on stroke are also high on 4.9 of the other eight conditions on average, against 2.5 for tracts high on cancer and 2.0 if the conditions were independent

### More tracts are high on all nine or on none than unrelated conditions give

Table 3 counts, for each tract, the conditions on which it is high. 34.0% (95% CI 30.7% to 37.1%) of tracts are high on none of the nine, where unrelated conditions with the same shares of high tracts would leave 7.4%. At the other end 892 tracts, 1.1% (95% CI 0.9% to 1.3%) of the total, are high on all nine, where unrelated conditions would give 0.3 tracts.

Tracts high on 5 or more of the nine make up 21.3% (95% CI 19.3% to 23.3%) of tracts against 4.9% for unrelated conditions. The benchmark is a reference point and not a hypothesis anyone held. It is the distribution that independent draws would give, with each condition's observed share of high tracts. Real tracts are not independent draws, in part because the model fits every condition to the same inputs.

Weighted by adults, 38.1% of adults live in tracts that are high on none and 17.1% in tracts that are high on 5 or more. Large tracts are therefore somewhat less likely to be high on many conditions.

Table 3. 34.0% of tracts are high on none of the nine conditions and 892 tracts are high on all nine, where independent conditions would give 0.3 tracts

The cutoff is a choice, and the numbers depend on it. By construction a tract is high on 2.3 conditions on average at the chapter's cutoff, and ties at the cutoff move the share of tracts that is high on a single condition between 24.0% and 25.6%.

Table 4 repeats the count with other cutoffs. With the stricter cutoff of the highest 10% of tracts, 64.9% of tracts are high on none and 14 are high on all nine, against 0.0 expected for unrelated conditions.

With the looser cutoff of the highest 33.3% the figures are 23.4%, 2,679 and 4.5. The number of tracts that are high on all nine exceeds the unrelated benchmark under every cutoff, and the share that is high on none depends strongly on the cutoff.

Table 4. Tracts are high on all nine conditions more often than independent conditions would give under every cutoff, from 14 tracts against 0.0 at the strictest to 2,679 against 4.5 at the loosest

### The count rises across the deprivation index, and the overlap does not strengthen

Table 5 orders tracts on the ADI and shows the mean count in each tenth. It is 0.5 (95% CI 0.4 to 0.6) in the least deprived tenth and 5.7 (95% CI 5.5 to 5.8) in the most deprived, a difference of 5.2 (95% CI 5.0 to 5.4), and it rises at every step, with the two largest steps at the deprived end. The share of adults who live in tracts high on 5 or more conditions is 0.8% (95% CI 0.3% to 1.6%) in the least deprived tenth and 72.5% (95% CI 69.9% to 75.1%) in the most deprived.

Part of this gradient is mechanical. The ADI is built from ACS data, and the PLACES model applies survey responses to each tract's census population using inputs from the same census sources, so conditions and index move together partly by design. The gradient of each condition taken one at a time belongs to Chapter 7.1, Health behaviors and the supply of care (https://vishalsingh.org/zipcode/health-behaviors-and-the-supply-of-care), and the count here adds only that the conditions which rise are high in the same tracts.

Table 5. Weighted by adults, a tract in the most deprived tenth is high on 5.7 of nine conditions on average, against 0.5 in the least deprived tenth

The conditions do not go together more closely where deprivation is higher. Inside the least deprived tenth the mean correlation among the nine is +0.40 (95% CI 0.37 to 0.44), and inside the most deprived tenth it is +0.34 (95% CI 0.32 to 0.36), a difference of +0.06 (95% CI 0.03 to 0.11) with the least deprived tenth higher. Inside the second to the ninth tenth it lies between +0.35 and +0.36.

After the mean of each ADI tenth is subtracted, the average correlation across all tracts is +0.36 (95% CI 0.35 to 0.38), against +0.49 before. Holding the ADI tenth fixed therefore removes part of the overlap, and a correlation remains in every tenth. The count of high conditions rises with deprivation, while the overlap among the conditions inside a tenth is smaller in the most deprived tenth than in the least deprived.

### An approximation of the model's census inputs leaves most of the overlap in place

Part of every correlation among the PLACES estimates in this chapter is built into the estimates. PLACES fits a model to survey responses, using the age, sex, race, ethnicity and education of each respondent, a county measure of poverty, and state and county effects, and applies it to the census population of each tract.[1] Conditions that are modeled from the same inputs move together wherever the inputs move, whatever the residents do. Within a county the model gives tracts different values only through the composition of their residents.

The data cannot take the model apart, but the ACS can approximate its inputs. The approximation needs ACS shares, and the ACS file lacks three of them for the District of Columbia, so this section leaves out the 203 tracts that lack a share, among them the 201 tracts of the District of Columbia. Every other statistic in the chapter includes them.

For each condition, a weighted regression on 8 ACS tract shares, their squares and pairwise products, a county poverty rate and state effects reproduces 79% to 90% of the variance of tract prevalence, counting the part the state effects carry. The average correlation among the residuals is +0.47 (95% CI 0.46 to 0.48), against +0.49 (95% CI 0.48 to 0.51) for the conditions themselves on the same tracts and +0.49 on all tracts (Table 6).

With county effects in place of state effects and county poverty, the fit reproduces 62% to 85% of the variance within counties. The average correlation among its residuals is +0.52 (95% CI 0.51 to 0.52), higher than the raw value on the same tracts.

Gradient-boosted trees fitted to the same inputs, scored on counties they did not see, reproduce 75% to 90% of the variance and leave a mean correlation of +0.45 (95% CI 0.44 to 0.47). Holding the state fixed gives +0.48 (95% CI 0.47 to 0.50) and holding the county fixed gives +0.43 (95% CI 0.42 to 0.43).

Table 6. An approximation of the model's census inputs leaves the average correlation among the nine conditions at +0.47, against +0.49 in the raw data

These numbers show that the inputs which can be measured here do not remove the overlap. They do not show that the overlap is real. The fits approximate the model and are not the model. The ACS file has no sex shares and no age groups beyond the shares under 18 and at 65 and over, the model works from individual respondents and not tract shares, and its county and state effects are fitted to each condition separately from the same survey. Whatever the conditions share and the fit misses, such as the age mix inside the older group, stays in the residuals and keeps them correlated. What the comparison supports is narrower. The shared part of the PLACES estimates is not a simple function of the census shares that were available, and the data cannot say how much of it comes from the model's other inputs and how much from the neighborhoods.

### Where tracts are high on many conditions

Tracts that are high on many conditions are not spread evenly. Of the variance of the count, 22.4% lies between states and 29.8% between counties within states, a total of 52.2% between counties, and 47.8% lies between tracts within counties. As the section on the model's inputs explains, the within-county part reflects the composition of residents.

Table 7 maps, for each county, the share of adults who live in tracts that are high on 5 or more conditions. In 1,229 of the 3,128 counties with a value, 50% or more of adults do.

By Census region the share is 25.8% of adults in the South, 20.2% in the Midwest, 8.5% in the Northeast and 7.0% in the West, against 17.1% nationally, and in West Virginia it is 81.7%. The counties with large shares cluster in the South and in parts of the Midwest.

Table 7. In 1,229 counties, 50% or more of adults live in tracts that are high on 5 or more conditions, and 25.8% of adults in the South do, against 7.0% in the West

### Life expectancy is shorter in tracts that are high on more conditions

Death records give a check that does not depend on the PLACES model. USALEEP life expectancy at birth, built from deaths in 2010 to 2015, is 80.2 (95% CI 80.0 to 80.4) years in tracts that are high on none of the nine and 74.1 (95% CI 73.9 to 74.3) years in tracts that are high on eight, a gap of 6.1 (95% CI 5.8 to 6.3) years (Table 8). In tracts that are high on all nine it is 75.2 (95% CI 74.9 to 75.4) years, 1.0 (95% CI 0.8 to 1.3) years above tracts that are high on eight, and the data here do not say why.

Tracts high on all nine differ in composition. The share of residents aged 65 and over is 22.6% (95% CI 22.3% to 22.9%) in tracts high on all nine against 18.9% (95% CI 18.6% to 19.2%) at eight, and the Black alone share is 2.7% (95% CI 2.2% to 3.1%) against 14.9% (95% CI 12.9% to 16.9%).

Across 77,934 tracts the count correlates −0.48 (95% CI −0.49 to −0.46) with life expectancy, and −0.25 (95% CI −0.27 to −0.22) once each tract is compared with the mean of its county. Within a county, tracts that are high on seven conditions differ from the county mean by −1.9 (95% CI −2.2 to −1.7) years, against +0.6 (95% CI 0.5 to 0.7) years for tracts that are high on none. Within a county the count varies only through the composition of tract residents, which also moves life expectancy, so this comparison does not isolate the conditions.

Life expectancy at birth is built from death rates at each age, so the age mix of a tract does not move it the way it moves crude prevalence, which makes this a check on whether the count reflects more than age. The two sources cover different years, 2010 to 2015 and 2022, and a correlation between them does not show that the conditions shorten lives.

Table 8. Life expectancy at birth is 6.1 years shorter in tracts that are high on eight conditions than in tracts that are high on none

### What a randomized offer to move shows about two of the conditions

> Causal evidence: Moving to Opportunity and measured obesity and diabetes (Randomized housing-voucher experiment)
>
> The tract patterns above do not say whether living in a high-poverty neighborhood changes health. One randomized study measured two of the nine conditions, with a definition that matches PLACES for one and not for the other. From 1994 to 1998 the Department of Housing and Urban Development randomly offered women with children in public housing in high-poverty tracts one of three things: a voucher usable only in low-poverty tracts, with moving counseling; an ordinary voucher; or no new offer. Of the women offered the low-poverty voucher, 48% used it to move, and every estimate below compares women by the offer they were randomized to (intention-to-treat) and not by whether they moved. A survey from 2008 to 2010 measured height, weight and glycated hemoglobin.[4]
>
> Compared with the group offered no voucher, the group offered the low-poverty voucher differed in the prevalence of extreme obesity (a body mass index of 40 or more) by −3.4 (95% CI −6.4 to −0.4) points and in the prevalence of glycated hemoglobin of 6.5% or more by −4.3 (95% CI −7.8 to −0.8) points, intervals that exclude zero. Obesity as PLACES defines it, a body mass index of 30 or more, differed by −1.2 (95% CI −5.4 to 3.0) points, an interval that includes zero. The glycated hemoglobin cutoff counts diabetes that has not been diagnosed, while PLACES counts diagnosed diabetes. The group offered ordinary vouchers did not differ significantly from the group offered nothing.
>
> The study measured two conditions and not a count of conditions, enrolled volunteers for a program, and does not show what a move would do to the number of conditions that are high in a tract.

### The count of high conditions rises under all four rankings

Table 9 repeats the count by tenth of tracts under four rankings, over the 75,596 tracts that have all four. Ranked by the ADI, the mean count is 0.4 (95% CI 0.3 to 0.5) in the least deprived tenth and 5.7 (95% CI 5.5 to 5.8) in the most deprived, and under the standardized index it runs from 0.5 (95% CI 0.4 to 0.5) to 4.5 (95% CI 4.1 to 4.8), lower in the most deprived tenth than under the ADI. Ranked by median household income it runs from 0.4 (95% CI 0.3 to 0.4) to 5.2 (95% CI 5.0 to 5.3), and by bachelor's degree share from 0.5 (95% CI 0.4 to 0.5) to 4.3 (95% CI 3.8 to 4.6), and under each ranking the tenth means rise in order (rank correlations of +1.00, +1.00 and +1.00). None of the three changes the gradient under the book's rule (Chapter 9.1, How far one ranking goes (https://vishalsingh.org/zipcode/how-far-one-ranking-goes)), and all four rankings, like the PLACES model, draw on ACS measures of the same tracts.

Table 9. The count of chronic conditions that are high in a tract is steadily worse toward the more deprived end under the ADI and under each of the three other rankings

### What these data cannot show

A count of high conditions describes a tract and not a person. A correlation across places can differ widely in size from the correlation among the people who live in them,[5] so the overlap here cannot be read as people carrying several conditions at once, and no figure in this chapter says how many conditions an adult has. Drawing that conclusion from place data is ecological inference. The count of conditions per person comes from surveys that ask each adult, and Chapter 7.2, Health behaviors, person by person (https://vishalsingh.org/zipcode/health-behaviors-person-by-person) uses one for behaviors.

The PLACES values are modeled estimates, the tract file has no intervals, and the bands in this chapter describe the sampling of counties and not the model's own uncertainty. CDC advises against using the estimates to rank the overall health of an area or to track change at the local level, and this chapter does neither.[6] It does compare conditions with each other and with the ADI, and every such comparison carries the shared inputs described above. The approximations show that measured census shares do not remove the overlap, but they cannot separate what the model builds in from what the neighborhoods contribute. The staged tract file holds crude prevalence only, so age is part of every pattern, most visibly for cancer, arthritis and heart disease.

High is relative. A tract in the highest 25% of tracts for a condition is high compared with other tracts and not high by any clinical standard, and the cutoff moves every count, as the table shows. High blood pressure and high cholesterol are left out because the 2024 release has no estimate of either for the 5,077 tracts of Florida. Outside Florida the mean correlation of high blood pressure with the nine conditions is +0.65 (95% CI 0.64 to 0.67), against +0.51 among the nine on the same tracts, so it follows the same shared pattern.

Of the 83,522 tracts with PLACES estimates, 82,781 also have an ADI rank, and the other 741 are left out. Counties are the clusters of the bootstrap, 3,129 of them, with Connecticut's planning regions standing in for its counties, while the map draws Connecticut as its eight legacy counties. The map has 15 gray counties with no tract in the sample, and 990 counties have fewer than 5 tracts, so their shares move in large steps. Puerto Rico is not covered by PLACES and is not drawn.

The conditions are measured once, in 2022, so the data say nothing about which condition comes first, whether tracts move into or out of the high group, or whether living in a tract changes any of the prevalences. The ADI describes 2019 to 2023 and the life expectancy check describes 2010 to 2015, so the sources do not describe one moment.

## Tables

### Table 1. Stroke, diabetes, COPD and heart disease are high in the same tracts, with correlations from +0.74 to +0.90, while cancer correlates −0.13 with asthma

Population-weighted Pearson correlations between the tract prevalence of nine chronic conditions among adults, 82,781 census tracts, 2022

| conf_high | conf_low | measure | n | other | r |
| ---: | ---: | --- | ---: | --- | ---: |
| n/a | n/a | Cancer | n/a | Cancer | 1 |
| 0.755 | 0.718 | Cancer | 82,781 | Arthritis | 0.738 |
| 0.655 | 0.601 | Cancer | 82,781 | Coronary heart disease | 0.629 |
| 0.362 | 0.297 | Cancer | 82,781 | COPD | 0.332 |
| 0.287 | 0.203 | Cancer | 82,781 | Stroke | 0.246 |
| 0.106 | −0.002 | Cancer | 82,781 | Diabetes | 0.052 |
| −0.050 | −0.180 | Cancer | 82,781 | Obesity | −0.112 |
| −0.069 | −0.196 | Cancer | 82,781 | Asthma | −0.127 |
| 0.057 | −0.067 | Cancer | 82,781 | Depression | −0.002 |
| 0.755 | 0.718 | Arthritis | 82,781 | Cancer | 0.738 |
| n/a | n/a | Arthritis | n/a | Arthritis | 1 |
| 0.914 | 0.888 | Arthritis | 82,781 | Coronary heart disease | 0.902 |
| 0.823 | 0.800 | Arthritis | 82,781 | COPD | 0.812 |
| 0.759 | 0.713 | Arthritis | 82,781 | Stroke | 0.736 |
| 0.635 | 0.520 | Arthritis | 82,781 | Diabetes | 0.577 |
| 0.531 | 0.421 | Arthritis | 82,781 | Obesity | 0.479 |
| 0.422 | 0.315 | Arthritis | 82,781 | Asthma | 0.374 |
| 0.305 | 0.197 | Arthritis | 82,781 | Depression | 0.253 |
| 0.655 | 0.601 | Coronary heart disease | 82,781 | Cancer | 0.629 |
| 0.914 | 0.888 | Coronary heart disease | 82,781 | Arthritis | 0.902 |
| n/a | n/a | Coronary heart disease | n/a | Coronary heart disease | 1 |
| 0.891 | 0.874 | Coronary heart disease | 82,781 | COPD | 0.883 |
| 0.875 | 0.851 | Coronary heart disease | 82,781 | Stroke | 0.863 |
| 0.763 | 0.710 | Coronary heart disease | 82,781 | Diabetes | 0.737 |
| 0.518 | 0.433 | Coronary heart disease | 82,781 | Obesity | 0.477 |
| 0.382 | 0.282 | Coronary heart disease | 82,781 | Asthma | 0.335 |
| 0.223 | 0.117 | Coronary heart disease | 82,781 | Depression | 0.172 |
| 0.362 | 0.297 | COPD | 82,781 | Cancer | 0.332 |
| 0.823 | 0.800 | COPD | 82,781 | Arthritis | 0.812 |
| 0.891 | 0.874 | COPD | 82,781 | Coronary heart disease | 0.883 |
| n/a | n/a | COPD | n/a | COPD | 1 |
| 0.904 | 0.887 | COPD | 82,781 | Stroke | 0.896 |
| 0.803 | 0.738 | COPD | 82,781 | Diabetes | 0.771 |
| 0.719 | 0.666 | COPD | 82,781 | Obesity | 0.694 |
| 0.641 | 0.569 | COPD | 82,781 | Asthma | 0.607 |
| 0.462 | 0.374 | COPD | 82,781 | Depression | 0.420 |
| 0.287 | 0.203 | Stroke | 82,781 | Cancer | 0.246 |
| 0.759 | 0.713 | Stroke | 82,781 | Arthritis | 0.736 |
| 0.875 | 0.851 | Stroke | 82,781 | Coronary heart disease | 0.863 |
| 0.904 | 0.887 | Stroke | 82,781 | COPD | 0.896 |
| n/a | n/a | Stroke | n/a | Stroke | 1 |
| 0.917 | 0.887 | Stroke | 82,781 | Diabetes | 0.902 |
| 0.675 | 0.623 | Stroke | 82,781 | Obesity | 0.649 |
| 0.580 | 0.507 | Stroke | 82,781 | Asthma | 0.545 |
| 0.193 | 0.109 | Stroke | 82,781 | Depression | 0.151 |
| 0.106 | −0.002 | Diabetes | 82,781 | Cancer | 0.052 |
| 0.635 | 0.520 | Diabetes | 82,781 | Arthritis | 0.577 |
| 0.763 | 0.710 | Diabetes | 82,781 | Coronary heart disease | 0.737 |
| 0.803 | 0.738 | Diabetes | 82,781 | COPD | 0.771 |
| 0.917 | 0.887 | Diabetes | 82,781 | Stroke | 0.902 |
| n/a | n/a | Diabetes | n/a | Diabetes | 1 |
| 0.734 | 0.647 | Diabetes | 82,781 | Obesity | 0.691 |
| 0.456 | 0.356 | Diabetes | 82,781 | Asthma | 0.403 |
| 0.105 | 0.010 | Diabetes | 82,781 | Depression | 0.058 |
| −0.050 | −0.180 | Obesity | 82,781 | Cancer | −0.112 |
| 0.531 | 0.421 | Obesity | 82,781 | Arthritis | 0.479 |
| 0.518 | 0.433 | Obesity | 82,781 | Coronary heart disease | 0.477 |
| 0.719 | 0.666 | Obesity | 82,781 | COPD | 0.694 |
| 0.675 | 0.623 | Obesity | 82,781 | Stroke | 0.649 |
| 0.734 | 0.647 | Obesity | 82,781 | Diabetes | 0.691 |
| n/a | n/a | Obesity | n/a | Obesity | 1 |
| 0.686 | 0.592 | Obesity | 82,781 | Asthma | 0.641 |
| 0.459 | 0.350 | Obesity | 82,781 | Depression | 0.407 |
| −0.069 | −0.196 | Asthma | 82,781 | Cancer | −0.127 |
| 0.422 | 0.315 | Asthma | 82,781 | Arthritis | 0.374 |
| 0.382 | 0.282 | Asthma | 82,781 | Coronary heart disease | 0.335 |
| 0.641 | 0.569 | Asthma | 82,781 | COPD | 0.607 |
| 0.580 | 0.507 | Asthma | 82,781 | Stroke | 0.545 |
| 0.456 | 0.356 | Asthma | 82,781 | Diabetes | 0.403 |
| 0.686 | 0.592 | Asthma | 82,781 | Obesity | 0.641 |
| n/a | n/a | Asthma | n/a | Asthma | 1 |
| 0.660 | 0.572 | Asthma | 82,781 | Depression | 0.618 |
| 0.057 | −0.067 | Depression | 82,781 | Cancer | −0.002 |
| 0.305 | 0.197 | Depression | 82,781 | Arthritis | 0.253 |
| 0.223 | 0.117 | Depression | 82,781 | Coronary heart disease | 0.172 |
| 0.462 | 0.374 | Depression | 82,781 | COPD | 0.420 |
| 0.193 | 0.109 | Depression | 82,781 | Stroke | 0.151 |
| 0.105 | 0.010 | Depression | 82,781 | Diabetes | 0.058 |
| 0.459 | 0.350 | Depression | 82,781 | Obesity | 0.407 |
| 0.660 | 0.572 | Depression | 82,781 | Asthma | 0.618 |
| n/a | n/a | Depression | n/a | Depression | 1 |

Note: PLACES values are modeled estimates. A model fitted to survey responses is applied to each tract's census population, so every condition's estimate depends on the same age, sex, race, ethnicity and education inputs and on county and state effects, and correlations across tracts are partly built in. The tract file carries no intervals, and the intervals shown here come from a bootstrap over counties.

Note: The diagonal is one by construction and has no interval. Values are crude prevalence among adults, not adjusted for age, so the age mix of a tract moves them.

Note: The nine conditions are the PLACES health outcome measures that cover all adults and have an estimate in every state. High blood pressure and high cholesterol are left out because the 2024 tract release has no estimate of either for 5,077 Florida tracts; outside Florida the mean correlation of high blood pressure with the nine is +0.65.

Source: CDC PLACES 2024 release (BRFSS 2022) and ADI 2023; modeled estimates, population-weighted (adults in each tract)

Data: https://vishalsingh.org/zipcode/report/multiple-chronic-conditions/figures/condition_correlations.json

### Table 2. Tracts high on stroke are also high on 4.9 of the other eight conditions on average, against 2.5 for tracts high on cancer and 2.0 if the conditions were independent

Mean number of the other eight conditions that are also high in a tract, for tracts in the highest 25% of tracts on each condition, 2022

| condition | conf_high | conf_low | estimate | n |
| --- | ---: | ---: | ---: | ---: |
| Arthritis | 4.65 | 4.48 | 4.57 | 20,620 |
| Cancer | 2.64 | 2.39 | 2.51 | 21,087 |
| Asthma | 3.91 | 3.55 | 3.73 | 20,589 |
| Coronary heart disease | 4.82 | 4.63 | 4.73 | 21,177 |
| COPD | 4.98 | 4.86 | 4.92 | 21,048 |
| Depression | 3.31 | 2.96 | 3.12 | 20,971 |
| Diabetes | 4.69 | 4.03 | 4.35 | 20,549 |
| Obesity | 4.15 | 3.86 | 3.99 | 20,885 |
| Stroke | 5.04 | 4.77 | 4.90 | 19,866 |

Note: PLACES values are modeled estimates. A model fitted to survey responses is applied to each tract's census population, so every condition's estimate depends on the same age, sex, race, ethnicity and education inputs and on county and state effects, and correlations across tracts are partly built in. The tract file carries no intervals, and the intervals shown here come from a bootstrap over counties.

Note: Weighting tracts by adults moves each mean by 0.34 conditions or less.

Source: CDC PLACES 2024 release (BRFSS 2022) and ADI 2023; modeled estimates, unweighted (each tract counted once)

Data: https://vishalsingh.org/zipcode/report/multiple-chronic-conditions/figures/others_high.json

### Table 3. 34.0% of tracts are high on none of the nine conditions and 892 tracts are high on all nine, where independent conditions would give 0.3 tracts

Tracts by the number of the nine conditions whose prevalence is in the highest 25% of tracts, observed and expected if the conditions were independent, 82,781 tracts, 2022

| conf_high | conf_low | count | n | series | share |
| ---: | ---: | --- | ---: | --- | ---: |
| 37.10 | 30.75 | 0 | 82,781 | Observed | 33.98 |
| 21.61 | 19.41 | 1 | 82,781 | Observed | 20.52 |
| 12.31 | 10.49 | 2 | 82,781 | Observed | 11.37 |
| 7.64 | 6.64 | 3 | 82,781 | Observed | 7.13 |
| 6.23 | 5.26 | 4 | 82,781 | Observed | 5.74 |
| 5.77 | 4.83 | 5 | 82,781 | Observed | 5.30 |
| 6.61 | 5.37 | 6 | 82,781 | Observed | 5.97 |
| 6.03 | 4.69 | 7 | 82,781 | Observed | 5.31 |
| 4.12 | 3.16 | 8 | 82,781 | Observed | 3.61 |
| 1.29 | 0.900 | 9 | 82,781 | Observed | 1.08 |
| n/a | n/a | 0 | n/a | If conditions were independent | 7.44 |
| n/a | n/a | 1 | n/a | If conditions were independent | 22.42 |
| n/a | n/a | 2 | n/a | If conditions were independent | 30.01 |
| n/a | n/a | 3 | n/a | If conditions were independent | 23.43 |
| n/a | n/a | 4 | n/a | If conditions were independent | 11.76 |
| n/a | n/a | 5 | n/a | If conditions were independent | 3.93 |
| n/a | n/a | 6 | n/a | If conditions were independent | 0.880 |
| n/a | n/a | 7 | n/a | If conditions were independent | 0.130 |
| n/a | n/a | 8 | n/a | If conditions were independent | 0.010 |
| n/a | n/a | 9 | n/a | If conditions were independent | 0 |

Note: The count is a count of tract prevalences above a cutoff, not a count of conditions in any person. PLACES values are modeled estimates. A model fitted to survey responses is applied to each tract's census population, so every condition's estimate depends on the same age, sex, race, ethnicity and education inputs and on county and state effects, and correlations across tracts are partly built in. The tract file carries no intervals, and the intervals shown here come from a bootstrap over counties.

Note: Tied values fall on one side of the cutoff, so the share of tracts that is high on one condition runs from 24.0% to 25.6%. The benchmark is a Poisson-binomial distribution with those shares. Shares of adults living in these tracts differ from shares of tracts (for example 38.1% of adults live in tracts that are high on none).

Source: CDC PLACES 2024 release (BRFSS 2022) and ADI 2023; modeled estimates, unweighted (each tract counted once)

Data: https://vishalsingh.org/zipcode/report/multiple-chronic-conditions/figures/high_count_distribution.json

### Table 4. Tracts are high on all nine conditions more often than independent conditions would give under every cutoff, from 14 tracts against 0.0 at the strictest to 2,679 against 4.5 at the loosest

Count of high conditions per tract when a tract is high on a condition if it is in the highest 10%, 20%, 25% or 33.3% of tracts, 82,781 tracts, 2022

| A tract is high on a condition if it is in the | Mean high conditions | Tracts high on none (%) | Tracts high on five or more (%) | Expected if independent (%) | Tracts high on all nine | Expected if independent |
| --- | ---: | ---: | ---: | ---: | ---: | ---: |
| Highest 10% of tracts | 0.90 | 64.9 | 6.2 | 0.1 | 14 | 0.00 |
| Highest 20% of tracts | 1.80 | 42.7 | 15.8 | 2.0 | 360 | 0.04 |
| Highest 25% of tracts (used in this chapter) | 2.26 | 34.0 | 21.3 | 4.9 | 892 | 0.32 |
| Highest third of tracts | 3.02 | 23.4 | 30.9 | 14.8 | 2,679 | 4.47 |

Note: The count is a count of tract prevalences above a cutoff, not a count of conditions in any person. PLACES values are modeled estimates. A model fitted to survey responses is applied to each tract's census population, so every condition's estimate depends on the same age, sex, race, ethnicity and education inputs and on county and state effects, and correlations across tracts are partly built in. The tract file carries no intervals, and the intervals shown here come from a bootstrap over counties.

Source: CDC PLACES 2024 release (BRFSS 2022) and ADI 2023; modeled estimates, unweighted (each tract counted once)

Data: https://vishalsingh.org/zipcode/report/multiple-chronic-conditions/figures/cutoff_sensitivity.json

### Table 5. Weighted by adults, a tract in the most deprived tenth is high on 5.7 of nine conditions on average, against 0.5 in the least deprived tenth

Population-weighted mean number of the nine conditions that are high, by tenth of tracts on the ADI national percentile, 82,781 tracts, 2022

| conf_high | conf_low | decile | estimate | measure | n_units |
| ---: | ---: | ---: | ---: | --- | ---: |
| 0.550 | 0.380 | 1 | 0.450 | count | 8,274 |
| 0.700 | 0.550 | 2 | 0.620 | count | 8,118 |
| 1 | 0.780 | 3 | 0.880 | count | 8,511 |
| 1.10 | 0.920 | 4 | 1.01 | count | 8,261 |
| 1.39 | 1.19 | 5 | 1.29 | count | 8,189 |
| 1.83 | 1.60 | 6 | 1.71 | count | 8,318 |
| 2.46 | 2.22 | 7 | 2.34 | count | 8,305 |
| 3.34 | 3.07 | 8 | 3.21 | count | 8,238 |
| 4.49 | 4.21 | 9 | 4.35 | count | 8,354 |
| 5.81 | 5.52 | 10 | 5.66 | count | 8,213 |
| 1.57 | 0.310 | 1 | 0.800 | share | 8,274 |
| 1.62 | 0.850 | 2 | 1.19 | share | 8,118 |
| 3.33 | 2.08 | 3 | 2.67 | share | 8,511 |
| 4.93 | 3.49 | 4 | 4.19 | share | 8,261 |
| 7.79 | 5.88 | 5 | 6.79 | share | 8,189 |
| 12.89 | 10.27 | 6 | 11.53 | share | 8,318 |
| 20.66 | 17.07 | 7 | 18.80 | share | 8,305 |
| 33.81 | 29.22 | 8 | 31.48 | share | 8,238 |
| 51.73 | 46.65 | 9 | 49.11 | share | 8,354 |
| 75.09 | 69.88 | 10 | 72.53 | share | 8,213 |

Views: Mean number of high conditions (measure = count); Adults in tracts high on five or more (measure = share).

Note: PLACES values are modeled estimates. A model fitted to survey responses is applied to each tract's census population, so every condition's estimate depends on the same age, sex, race, ethnicity and education inputs and on county and state effects, and correlations across tracts are partly built in. The tract file carries no intervals, and the intervals shown here come from a bootstrap over counties. The ADI is built from American Community Survey data that overlap the model's inputs, so part of each association with it is mechanical.

Note: Tenths hold equal numbers of tracts except where tied ranks fall in one tenth. Values are crude prevalence among adults, not adjusted for age, so the age mix of a tract moves them.

Note: Across all the tracts in the comparison, weighted by adults, a tract is high on 2.0 of the nine conditions on average.

Source: CDC PLACES 2024 release (BRFSS 2022) and ADI 2023; modeled estimates, population-weighted (adults in each tract)

Data: https://vishalsingh.org/zipcode/report/multiple-chronic-conditions/figures/high_count_by_adi.json

### Table 6. An approximation of the model's census inputs leaves the average correlation among the nine conditions at +0.47, against +0.49 in the raw data

Mean of the 36 pairwise population-weighted correlations among the nine conditions across 82,781 tracts, 2022, under seven comparisons

| conf_high | conf_low | estimate | group | n | stage |
| ---: | ---: | ---: | --- | ---: | --- |
| 0.511 | 0.477 | 0.495 | Compared within groups of tracts | 82,781 | All tracts |
| 0.498 | 0.471 | 0.484 | Compared within groups of tracts | 82,781 | Within states |
| 0.435 | 0.418 | 0.427 | Compared within groups of tracts | 82,781 | Within counties |
| 0.377 | 0.346 | 0.361 | Compared within groups of tracts | 82,781 | Within tenths of the ADI |
| 0.476 | 0.460 | 0.468 | After approximating the model's inputs | 82,578 | Census shares, county poverty and state effects removed |
| 0.521 | 0.511 | 0.516 | After approximating the model's inputs | 82,578 | Census shares removed within counties |
| 0.469 | 0.441 | 0.455 | After approximating the model's inputs | 82,578 | Flexible fit (trees) on the same inputs and state |

Note: The approximation regresses each condition on 8 ACS tract shares (ages 65 and over and under 18, white, Black, Asian and Hispanic residents, and adults with a college degree and with no more than high school), with squares and pairwise products, and removes state or county effects. It uses a county poverty rate in place of the model's measure of county poverty. The ACS file has no sex shares, and the model also uses fitted county and state effects for each condition. The fit explains 79% to 90% of the variance of each condition, counting the state effects, and the tree fit 75% to 90% out of sample. With county effects the fit explains 62% to 85% of the variance within counties. The three adjusted rows leave out the 203 tracts that lack an ACS input, among them the 201 tracts of the District of Columbia, whose ACS file lacks three of the shares; on the tracts they use, the raw average correlation is +0.49.

Note: PLACES values are modeled estimates. A model fitted to survey responses is applied to each tract's census population, so every condition's estimate depends on the same age, sex, race, ethnicity and education inputs and on county and state effects, and correlations across tracts are partly built in. The tract file carries no intervals, and the intervals shown here come from a bootstrap over counties.

Source: CDC PLACES 2024 release (BRFSS 2022), ADI 2023 and ACS 2019-2023; modeled estimates, population-weighted (adults in each tract)

Data: https://vishalsingh.org/zipcode/report/multiple-chronic-conditions/figures/correlation_stages.json

### Table 7. In 1,229 counties, 50% or more of adults live in tracts that are high on 5 or more conditions, and 25.8% of adults in the South do, against 7.0% in the West

Share of adults living in tracts that are high on 5 or more of the nine conditions, by county, population-weighted, 2022

No inline rows: the values of Table 7 are in https://vishalsingh.org/zipcode/data/multiple-chronic-conditions/adults_high_count_map.json.

Note: PLACES values are modeled estimates. A model fitted to survey responses is applied to each tract's census population, so every condition's estimate depends on the same age, sex, race, ethnicity and education inputs and on county and state effects, and correlations across tracts are partly built in. The tract file carries no intervals, and the intervals shown here come from a bootstrap over counties.

Note: A county's share is a full count of its tracts and moves in large steps when the county has few tracts: 990 counties have fewer than 5 tracts. Gray counties have no tract in the sample. Puerto Rico is not covered by PLACES and is not drawn. Alaska is drawn, and Connecticut appears under its eight legacy counties, to which its tracts are assigned.

Source: CDC PLACES 2024 release (BRFSS 2022), ADI 2023 and ACS 2019-2023; modeled estimates, population-weighted (adults in each tract)

Data: https://vishalsingh.org/zipcode/report/multiple-chronic-conditions/figures/adults_high_count_map.json

### Table 8. Life expectancy at birth is 6.1 years shorter in tracts that are high on eight conditions than in tracts that are high on none

Population-weighted mean tract life expectancy at birth by number of high conditions, 77,934 tracts, deaths in 2010 to 2015, institutional tracts dropped

| conf_high | conf_low | count | estimate | measure | n_units |
| ---: | ---: | ---: | ---: | --- | ---: |
| 80.37 | 79.98 | 0 | 80.18 | level | 26,431 |
| 80 | 79.64 | 1 | 79.82 | level | 15,948 |
| 78.50 | 78.18 | 2 | 78.33 | level | 8,632 |
| 78.19 | 77.78 | 3 | 77.97 | level | 5,599 |
| 77.32 | 76.79 | 4 | 77.06 | level | 4,541 |
| 76.72 | 76.18 | 5 | 76.45 | level | 4,162 |
| 76.38 | 75.96 | 6 | 76.16 | level | 4,602 |
| 74.84 | 74.35 | 7 | 74.59 | level | 4,248 |
| 74.30 | 73.94 | 8 | 74.12 | level | 2,898 |
| 75.41 | 74.93 | 9 | 75.17 | level | 873 |
| 0.660 | 0.460 | 0 | 0.560 | dev | 26,417 |
| 0.800 | 0.610 | 1 | 0.700 | dev | 15,927 |
| −0.230 | −0.480 | 2 | −0.350 | dev | 8,612 |
| −0.180 | −0.560 | 3 | −0.360 | dev | 5,567 |
| −0.770 | −1.25 | 4 | −1 | dev | 4,511 |
| −1.02 | −1.50 | 5 | −1.26 | dev | 4,135 |
| −0.980 | −1.29 | 6 | −1.14 | dev | 4,572 |
| −1.70 | −2.16 | 7 | −1.94 | dev | 4,219 |
| −1.60 | −2 | 8 | −1.80 | dev | 2,889 |
| −0.290 | −0.640 | 9 | −0.470 | dev | 871 |

Views: Life expectancy (years) (measure = level); Difference from the county mean (measure = dev).

Note: Tract life expectancy covers 2010 to 2015 and the conditions cover 2022, so the two describe different years. USALEEP estimates were area-weighted from 2010 tracts onto 2020 tracts and suit comparisons of groups of tracts, not single tracts. The within-county view leaves out counties with a single USALEEP tract in the sample. Connecticut tracts carry planning-region codes in PLACES, which serve as its counties for the bootstrap.

Note: Within a county the count varies only through the composition of tract residents, which also moves life expectancy. Life expectancy is not affected by the age mix of a tract the way crude prevalence is, which makes it a check on whether the count reflects more than age. The count itself comes from PLACES modeled estimates and carries the caveats in the notes to the earlier figures.

Source: CDC PLACES 2024 release (BRFSS 2022), USALEEP life expectancy 2010-2015 (area-weighted onto 2020 tracts) and ACS 2019-2023; modeled estimates, population-weighted (ACS tract population)

Data: https://vishalsingh.org/zipcode/report/multiple-chronic-conditions/figures/life_expectancy_by_count.json

### Table 9. The count of chronic conditions that are high in a tract is steadily worse toward the more deprived end under the ADI and under each of the three other rankings

Mean number of high conditions (of nine) by tenth of tracts ranked four ways, over the 75,596 tracts that have all four rankings

| bin | conf_high | conf_low | estimate | n_units | ranking |
| ---: | ---: | ---: | ---: | ---: | --- |
| 1 | 0.490 | 0.339 | 0.402 | 7,551 | Area Deprivation Index |
| 2 | 0.701 | 0.537 | 0.611 | 7,563 | Area Deprivation Index |
| 3 | 0.936 | 0.765 | 0.845 | 7,489 | Area Deprivation Index |
| 4 | 1.09 | 0.907 | 0.998 | 7,757 | Area Deprivation Index |
| 5 | 1.37 | 1.17 | 1.27 | 7,485 | Area Deprivation Index |
| 6 | 1.85 | 1.61 | 1.73 | 7,513 | Area Deprivation Index |
| 7 | 2.46 | 2.23 | 2.35 | 7,658 | Area Deprivation Index |
| 8 | 3.38 | 3.10 | 3.24 | 7,419 | Area Deprivation Index |
| 9 | 4.49 | 4.20 | 4.34 | 7,618 | Area Deprivation Index |
| 10 | 5.81 | 5.53 | 5.66 | 7,543 | Area Deprivation Index |
| 1 | 0.534 | 0.420 | 0.474 | 7,559 | Standardized index |
| 2 | 0.732 | 0.576 | 0.654 | 7,560 | Standardized index |
| 3 | 0.978 | 0.813 | 0.898 | 7,559 | Standardized index |
| 4 | 1.29 | 1.09 | 1.19 | 7,560 | Standardized index |
| 5 | 1.69 | 1.45 | 1.57 | 7,560 | Standardized index |
| 6 | 2.22 | 1.91 | 2.06 | 7,559 | Standardized index |
| 7 | 2.80 | 2.41 | 2.60 | 7,560 | Standardized index |
| 8 | 3.40 | 2.89 | 3.15 | 7,559 | Standardized index |
| 9 | 3.90 | 3.30 | 3.60 | 7,560 | Standardized index |
| 10 | 4.81 | 4.12 | 4.47 | 7,560 | Standardized index |
| 1 | 0.435 | 0.344 | 0.386 | 7,560 | Median household income |
| 2 | 0.646 | 0.502 | 0.569 | 7,558 | Median household income |
| 3 | 0.866 | 0.691 | 0.775 | 7,560 | Median household income |
| 4 | 1.18 | 0.972 | 1.07 | 7,560 | Median household income |
| 5 | 1.53 | 1.30 | 1.41 | 7,560 | Median household income |
| 6 | 2.05 | 1.78 | 1.91 | 7,560 | Median household income |
| 7 | 2.77 | 2.45 | 2.60 | 7,555 | Median household income |
| 8 | 3.34 | 2.98 | 3.16 | 7,564 | Median household income |
| 9 | 4.23 | 3.87 | 4.05 | 7,560 | Median household income |
| 10 | 5.32 | 5.00 | 5.16 | 7,559 | Median household income |
| 1 | 0.522 | 0.405 | 0.458 | 7,558 | Bachelor's degree share |
| 2 | 0.714 | 0.569 | 0.640 | 7,557 | Bachelor's degree share |
| 3 | 0.928 | 0.758 | 0.843 | 7,566 | Bachelor's degree share |
| 4 | 1.19 | 1.00 | 1.10 | 7,560 | Bachelor's degree share |
| 5 | 1.60 | 1.37 | 1.49 | 7,553 | Bachelor's degree share |
| 6 | 2.10 | 1.83 | 1.96 | 7,559 | Bachelor's degree share |
| 7 | 2.78 | 2.45 | 2.62 | 7,562 | Bachelor's degree share |
| 8 | 3.46 | 3.03 | 3.25 | 7,569 | Bachelor's degree share |
| 9 | 4.22 | 3.65 | 3.94 | 7,548 | Bachelor's degree share |
| 10 | 4.64 | 3.85 | 4.25 | 7,564 | Bachelor's degree share |

Note: Least and most deprived tenth: ADI 0.4 and 5.7, rank correlation +1.00; standardized index 0.5 and 4.5, +1.00; median household income 0.4 and 5.2, +1.00; bachelor's degree share 0.5 and 4.3, +1.00.

Note: The standardized index applies the ADI's coefficients to its measures after standardizing each one, so home value and income carry less weight than in the ADI. Income and degree share are single census measures, each ranked from the highest value to the lowest.

Source: CDC PLACES 2024 release (BRFSS 2022) and ADI 2023; modeled estimates, population-weighted (adults in each tract); ADI 2023 and ACS 2019-2023 tract tables (B19013, B15003 and the 15 rebuilt ADI measures)

Data: https://vishalsingh.org/zipcode/report/multiple-chronic-conditions/figures/alt_rankings.json

## All estimates

Every fact in facts.json (146), sorted by id.

| Fact id | Value | Unit | Estimand | Population | Period | Interval | n | Weights | Source | Note |
| --- | ---: | --- | --- | --- | --- | --- | ---: | --- | --- | --- |
| age_old | 65 | years of age | definition: the age at which the ACS share of older residents starts | residents in the ACS share used as an input | ACS 2019-2023 |   |   | none (a definition) | American Community Survey |   |
| age_young | 18 | years of age | definition: the age below which the ACS share of children is counted | residents in the ACS share used as an input | ACS 2019-2023 |   |   | none (a definition) | American Community Survey |   |
| all_exp | 0.3 | tracts | number of tracts expected to be high on all nine conditions if they were independent across tracts (Poisson-binomial with the observed share high on each) | census tracts with an estimate for all nine conditions and an ADI rank | BRFSS 2022 (PLACES 2024 release); ADI 2023 |   |   | none (a benchmark) | CDC PLACES |   |
| all_n | 892 | tracts | count of tracts that are high on all nine conditions | census tracts with an estimate for all nine conditions and an ADI rank | BRFSS 2022 (PLACES 2024 release); ADI 2023 |   |   | none (full count) | CDC PLACES |   |
| all_share | 1.1% | percent of tracts | share of tracts that are high on all nine conditions, bootstrap interval over counties | census tracts with an estimate for all nine conditions and an ADI rank | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 95% CI 0.9% to 1.3% | 82,781 tracts | none (each tract counts once) | CDC PLACES |   |
| alt_adi_least | 0.4 | Mean number of high conditions (of nine) | adult-weighted mean number of the nine conditions that are high; 95% intervals from a bootstrap over counties | the least deprived tenth of tracts ranked by Area Deprivation Index, among the tracts with all four rankings | BRFSS 2022 (PLACES 2024 release); ADI 2023; ACS 2019-2023 | 95% CI 0.3 to 0.5 | 7,551 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| alt_adi_most | 5.7 | Mean number of high conditions (of nine) | adult-weighted mean number of the nine conditions that are high; 95% intervals from a bootstrap over counties | the most deprived tenth of tracts ranked by Area Deprivation Index, among the tracts with all four rankings | BRFSS 2022 (PLACES 2024 release); ADI 2023; ACS 2019-2023 | 95% CI 5.5 to 5.8 | 7,543 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| alt_adi_rho | +1.00 | rank correlation | Spearman rank correlation between bin order and bin mean, each bin counted once | the bins of tracts ranked by Area Deprivation Index | BRFSS 2022 (PLACES 2024 release); ADI 2023; ACS 2019-2023 |   |   | each bin counted once | CDC PLACES |   |
| alt_degree_least | 0.5 | Mean number of high conditions (of nine) | adult-weighted mean number of the nine conditions that are high; 95% intervals from a bootstrap over counties | the least deprived tenth of tracts ranked by Bachelor's degree share, among the tracts with all four rankings | BRFSS 2022 (PLACES 2024 release); ADI 2023; ACS 2019-2023 | 95% CI 0.4 to 0.5 | 7,558 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| alt_degree_most | 4.3 | Mean number of high conditions (of nine) | adult-weighted mean number of the nine conditions that are high; 95% intervals from a bootstrap over counties | the most deprived tenth of tracts ranked by Bachelor's degree share, among the tracts with all four rankings | BRFSS 2022 (PLACES 2024 release); ADI 2023; ACS 2019-2023 | 95% CI 3.8 to 4.6 | 7,564 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| alt_degree_rho | +1.00 | rank correlation | Spearman rank correlation between bin order and bin mean, each bin counted once | the bins of tracts ranked by Bachelor's degree share | BRFSS 2022 (PLACES 2024 release); ADI 2023; ACS 2019-2023 |   |   | each bin counted once | CDC PLACES |   |
| alt_income_least | 0.4 | Mean number of high conditions (of nine) | adult-weighted mean number of the nine conditions that are high; 95% intervals from a bootstrap over counties | the least deprived tenth of tracts ranked by Median household income, among the tracts with all four rankings | BRFSS 2022 (PLACES 2024 release); ADI 2023; ACS 2019-2023 | 95% CI 0.3 to 0.4 | 7,560 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| alt_income_most | 5.2 | Mean number of high conditions (of nine) | adult-weighted mean number of the nine conditions that are high; 95% intervals from a bootstrap over counties | the most deprived tenth of tracts ranked by Median household income, among the tracts with all four rankings | BRFSS 2022 (PLACES 2024 release); ADI 2023; ACS 2019-2023 | 95% CI 5.0 to 5.3 | 7,559 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| alt_income_rho | +1.00 | rank correlation | Spearman rank correlation between bin order and bin mean, each bin counted once | the bins of tracts ranked by Median household income | BRFSS 2022 (PLACES 2024 release); ADI 2023; ACS 2019-2023 |   |   | each bin counted once | CDC PLACES |   |
| alt_n_places | 75,596 | tracts | count of tracts in the common sample of the ranking comparison | tracts with the outcome and all four rankings | BRFSS 2022 (PLACES 2024 release); ADI 2023; ACS 2019-2023 |   |   | none (full count) | CDC PLACES |   |
| alt_rho_cut | 0.9 | absolute rank correlation | threshold between a steady and an uneven gradient | the rule for changed gradients (docs/RANKING_SWITCH_RULE.md) | BRFSS 2022 (PLACES 2024 release); ADI 2023; ACS 2019-2023 |   |   | none (a fixed rule) | CDC PLACES |   |
| alt_rho_margin | 0.1 | absolute rank correlation | smallest move in the rank correlation that counts as a change of shape | the rule for changed gradients (docs/RANKING_SWITCH_RULE.md) | BRFSS 2022 (PLACES 2024 release); ADI 2023; ACS 2019-2023 |   |   | none (a fixed rule) | CDC PLACES |   |
| alt_std_least | 0.5 | Mean number of high conditions (of nine) | adult-weighted mean number of the nine conditions that are high; 95% intervals from a bootstrap over counties | the least deprived tenth of tracts ranked by Standardized index, among the tracts with all four rankings | BRFSS 2022 (PLACES 2024 release); ADI 2023; ACS 2019-2023 | 95% CI 0.4 to 0.5 | 7,559 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| alt_std_most | 4.5 | Mean number of high conditions (of nine) | adult-weighted mean number of the nine conditions that are high; 95% intervals from a bootstrap over counties | the most deprived tenth of tracts ranked by Standardized index, among the tracts with all four rankings | BRFSS 2022 (PLACES 2024 release); ADI 2023; ACS 2019-2023 | 95% CI 4.1 to 4.8 | 7,560 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| alt_std_rho | +1.00 | rank correlation | Spearman rank correlation between bin order and bin mean, each bin counted once | the bins of tracts ranked by Standardized index | BRFSS 2022 (PLACES 2024 release); ADI 2023; ACS 2019-2023 |   |   | each bin counted once | CDC PLACES |   |
| blk_c8 | 14.9% | percent of residents Black alone | population-weighted mean of the ACS share of residents Black alone, bootstrap interval over counties | residents of tracts that are high on 8 of the nine conditions, USALEEP tracts in the sample | ACS 2019-2023 | 95% CI 12.9% to 16.9% | 2,898 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| blk_c9 | 2.7% | percent of residents Black alone | population-weighted mean of the ACS share of residents Black alone, bootstrap interval over counties | residents of tracts that are high on 9 of the nine conditions, USALEEP tracts in the sample | ACS 2019-2023 | 95% CI 2.2% to 3.1% | 873 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| bp_florida | 5,077 | tracts | count of tracts with no high blood pressure estimate | Florida census tracts | PLACES 2024 release (BRFSS 2021 for blood pressure) |   |   | none (full count) | CDC PLACES |   |
| bp_nine_r | +0.51 | correlation coefficient (mean of the 36 pairs) | mean pairwise correlation among the nine conditions on the same tracts | adults living in tracts outside Florida with all ten estimates | BRFSS 2022 (PLACES 2024 release); ADI 2023 |   | 77,544 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| bp_r | +0.65 | correlation coefficient (mean over the nine conditions) | mean of the correlations of high blood pressure with the nine conditions, bootstrap interval over counties | adults living in tracts outside Florida with all ten estimates | BRFSS 2021 and 2022 (PLACES 2024 release) | 95% CI 0.64 to 0.67 | 77,544 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| cnt_d1 | 0.5 | conditions (of nine) | mean number of the nine conditions that are high, population-weighted mean over a tenth of tracts, bootstrap interval over counties | adults in the least deprived tenth of tracts by ADI national percentile | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 95% CI 0.4 to 0.6 | 8,274 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| cnt_d10 | 5.7 | conditions (of nine) | mean number of the nine conditions that are high, population-weighted mean over a tenth of tracts, bootstrap interval over counties | adults in the most deprived tenth of tracts by ADI national percentile | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 95% CI 5.5 to 5.8 | 8,213 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| cnt_diff | 5.2 | conditions (of nine) | difference between the population-weighted mean count of high conditions in the most and the least deprived tenth of tracts, bootstrap interval over counties (draws shared) | adults in the most deprived tenth of tracts compared with the least deprived tenth | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 95% CI 5.0 to 5.4 | 82,781 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| cnt_mean | 2.3 | conditions (of nine) | mean number of high conditions per tract; it equals nine times the share high, so a result of the cutoff | census tracts with an estimate for all nine conditions and an ADI rank | BRFSS 2022 (PLACES 2024 release); ADI 2023 |   | 82,781 tracts | none (each tract counts once) | CDC PLACES |   |
| cut_fifth | 20% | percent of tracts | definition: a cutoff of the sensitivity check, the highest 20% of tracts | census tracts with an estimate for all nine conditions and an ADI rank | BRFSS 2022 (PLACES 2024 release); ADI 2023 |   |   | none (a definition) | CDC PLACES |   |
| cut_pct | 25% | percent of tracts | definition: a tract is high on a condition when its prevalence is in the highest 25% of tracts, ties at their mean rank | census tracts with an estimate for all nine conditions and an ADI rank | BRFSS 2022 (PLACES 2024 release); ADI 2023 |   |   | none (a definition) | CDC PLACES |   |
| cut_tenth | 10% | percent of tracts | definition: a cutoff of the sensitivity check, the highest 10% of tracts | census tracts with an estimate for all nine conditions and an ADI rank | BRFSS 2022 (PLACES 2024 release); ADI 2023 |   |   | none (a definition) | CDC PLACES |   |
| cut_third | 33.3% | percent of tracts | definition: a cutoff of the sensitivity check, the highest 33.3% of tracts | census tracts with an estimate for all nine conditions and an ADI rank | BRFSS 2022 (PLACES 2024 release); ADI 2023 |   |   | none (a definition) | CDC PLACES |   |
| dc_tracts | 201 | tracts | count of tracts; the ACS file lacks the under-18, white-alone and Asian-alone shares for the District of Columbia | tracts of the District of Columbia in the sample | ACS 2019-2023 |   |   | none (full count) | American Community Survey |   |
| few_k | 5 | tracts | definition: a county with fewer tracts than this has a share that moves in large steps | counties on the map | BRFSS 2022 (PLACES 2024 release); ADI 2023 |   |   | none (a definition) | CDC PLACES |   |
| ge5_d1 | 0.8% | percent of adults | share of adults living in tracts that are high on five or more conditions, population-weighted mean over a tenth of tracts, bootstrap interval over counties | adults in the least deprived tenth of tracts by ADI national percentile | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 95% CI 0.3% to 1.6% | 8,274 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| ge5_d10 | 72.5% | percent of adults | share of adults living in tracts that are high on five or more conditions, population-weighted mean over a tenth of tracts, bootstrap interval over counties | adults in the most deprived tenth of tracts by ADI national percentile | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 95% CI 69.9% to 75.1% | 8,213 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| ge5_exp | 4.9% | percent of tracts | share of tracts expected to be high on five or more conditions if they were independent across tracts | census tracts with an estimate for all nine conditions and an ADI rank | BRFSS 2022 (PLACES 2024 release); ADI 2023 |   |   | none (a benchmark) | CDC PLACES |   |
| ge5_share | 21.3% | percent of tracts | share of tracts that are high on five or more conditions, bootstrap interval over counties | census tracts with an estimate for all nine conditions and an ADI rank | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 95% CI 19.3% to 23.3% | 82,781 tracts | none (each tract counts once) | CDC PLACES |   |
| ge5_share_w | 17.1% | percent of adults | share of adults living in tracts that are high on five or more conditions | adults living in census tracts with an estimate for all nine conditions and an ADI rank | BRFSS 2022 (PLACES 2024 release); ADI 2023 |   | 82,781 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| half_pct | 50% | percent of adults | definition: a county where half of adults or more live in such tracts | counties where the share of adults in tracts high on five or more conditions reaches this level | BRFSS 2022 (PLACES 2024 release); ADI 2023 |   |   | none (a definition) | CDC PLACES |   |
| high_share_max | 25.6% | percent of tracts | largest share of tracts that is high on one condition | census tracts with an estimate for all nine conditions and an ADI rank | BRFSS 2022 (PLACES 2024 release); ADI 2023 |   | 82,781 tracts | none (each tract counts once) | CDC PLACES |   |
| high_share_min | 24.0% | percent of tracts | smallest share of tracts that is high on one condition; tied values fall on one side of the cutoff | census tracts with an estimate for all nine conditions and an ADI rank | BRFSS 2022 (PLACES 2024 release); ADI 2023 |   | 82,781 tracts | none (each tract counts once) | CDC PLACES |   |
| le_c0 | 80.2 | years of life expectancy at birth | population-weighted mean tract life expectancy, bootstrap interval over counties | residents of tracts that are high on 0 of the nine conditions, USALEEP tracts in the sample, institutional tracts dropped | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); PLACES 2024 release | 95% CI 80.0 to 80.4 | 26,431 tracts | ACS 2019-2023 tract population | USALEEP life expectancy |   |
| le_c8 | 74.1 | years of life expectancy at birth | population-weighted mean tract life expectancy, bootstrap interval over counties | residents of tracts that are high on 8 of the nine conditions, USALEEP tracts in the sample, institutional tracts dropped | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); PLACES 2024 release | 95% CI 73.9 to 74.3 | 2,898 tracts | ACS 2019-2023 tract population | USALEEP life expectancy |   |
| le_c9 | 75.2 | years of life expectancy at birth | population-weighted mean tract life expectancy, bootstrap interval over counties | residents of tracts that are high on 9 of the nine conditions, USALEEP tracts in the sample, institutional tracts dropped | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); PLACES 2024 release | 95% CI 74.9 to 75.4 | 873 tracts | ACS 2019-2023 tract population | USALEEP life expectancy |   |
| le_gap | 6.1 | years of life expectancy at birth | difference in population-weighted mean tract life expectancy, bootstrap interval over counties (draws shared) | residents of tracts that are high on none of the nine conditions compared with tracts that are high on eight | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); PLACES 2024 release | 95% CI 5.8 to 6.3 | 29,329 tracts | ACS 2019-2023 tract population | USALEEP life expectancy |   |
| le_r | −0.48 | correlation coefficient | population-weighted correlation of the count of high conditions with tract life expectancy, bootstrap interval over counties | residents of USALEEP tracts in the sample | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); PLACES 2024 release | 95% CI −0.49 to −0.46 | 77,934 tracts | ACS 2019-2023 tract population | USALEEP life expectancy |   |
| le_r_dev | −0.25 | correlation coefficient | population-weighted correlation of the count of high conditions with tract life expectancy minus the county mean, bootstrap interval over counties | residents of USALEEP tracts in counties with at least two such tracts | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); PLACES 2024 release | 95% CI −0.27 to −0.22 | 77,720 tracts | ACS 2019-2023 tract population | USALEEP life expectancy |   |
| le_rebound | 1.0 | years of life expectancy at birth | difference in population-weighted mean tract life expectancy, bootstrap interval over counties (draws shared) | residents of tracts high on all nine conditions compared with tracts high on eight | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); PLACES 2024 release | 95% CI 0.8 to 1.3 | 3,771 tracts | ACS 2019-2023 tract population | USALEEP life expectancy |   |
| led_c0 | +0.6 | years of life expectancy at birth, relative to the mean of the tract's county | population-weighted mean difference between tract life expectancy and the population-weighted mean of its county, bootstrap interval over counties | residents of tracts that are high on 0 of the nine conditions, in counties with at least two USALEEP tracts | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); PLACES 2024 release | 95% CI 0.5 to 0.7 | 26,417 tracts | ACS 2019-2023 tract population | USALEEP life expectancy |   |
| led_c7 | −1.9 | years of life expectancy at birth, relative to the mean of the tract's county | population-weighted mean difference between tract life expectancy and the population-weighted mean of its county, bootstrap interval over counties | residents of tracts that are high on 7 of the nine conditions, in counties with at least two USALEEP tracts | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); PLACES 2024 release | 95% CI −2.2 to −1.7 | 4,219 tracts | ACS 2019-2023 tract population | USALEEP life expectancy |   |
| many_k | 5 | conditions | definition: five or more of the nine conditions, a majority | census tracts with an estimate for all nine conditions and an ADI rank | BRFSS 2022 (PLACES 2024 release); ADI 2023 |   |   | none (a definition) | CDC PLACES |   |
| mto_bmi_extreme | 40 | body mass index | definition: the lowest body mass index counted as extreme obesity | women with children in public housing in high-poverty tracts, randomized to a voucher offer from 1994 to 1998, measured 2008 to 2010 | measured 2008 to 2010 |   |   | none (a definition) | Published estimates from cited studies |   |
| mto_bmi_obese | 30 | body mass index | definition: the lowest body mass index counted as obesity, the PLACES definition | women with children in public housing in high-poverty tracts, randomized to a voucher offer from 1994 to 1998, measured 2008 to 2010 | measured 2008 to 2010 |   |   | none (a definition) | Published estimates from cited studies |   |
| mto_hba1c_cut | 6.5% | percent glycated hemoglobin | definition: the lowest glycated hemoglobin counted as diabetes in the study | women with children in public housing in high-poverty tracts, randomized to a voucher offer from 1994 to 1998, measured 2008 to 2010 | measured 2008 to 2010 |   |   | none (a definition) | Published estimates from cited studies |   |
| mto_itt_extreme | −3.4 | percentage points of women with body mass index of 40 or more | intention-to-treat difference in prevalence between women randomly offered a voucher usable only in low-poverty tracts (with counseling) and women offered no voucher, reported by Ludwig and coauthors (2011, New England Journal of Medicine 365(16)): 4,498 women with children in public housing in high-poverty tracts in Baltimore, Boston, Chicago, Los Angeles and New York, randomized 1994 to 1998 and measured in a survey from 2008 to 2010; copied from the paper and not computed in this build | women with children in public housing in high-poverty tracts, randomized to a voucher offer from 1994 to 1998, measured 2008 to 2010 | randomized 1994 to 1998, measured 2008 to 2010 | 95% CI −6.4 to −0.4 |   | none (reported estimate) | Published estimates from cited studies | A published estimate, not a result of this build. |
| mto_itt_hba1c | −4.3 | percentage points of women with glycated hemoglobin of 6.5% or more | intention-to-treat difference in prevalence between women randomly offered a voucher usable only in low-poverty tracts (with counseling) and women offered no voucher, reported by Ludwig and coauthors (2011, New England Journal of Medicine 365(16)): 4,498 women with children in public housing in high-poverty tracts in Baltimore, Boston, Chicago, Los Angeles and New York, randomized 1994 to 1998 and measured in a survey from 2008 to 2010; copied from the paper and not computed in this build | women with children in public housing in high-poverty tracts, randomized to a voucher offer from 1994 to 1998, measured 2008 to 2010 | randomized 1994 to 1998, measured 2008 to 2010 | 95% CI −7.8 to −0.8 |   | none (reported estimate) | Published estimates from cited studies | A published estimate, not a result of this build. |
| mto_itt_obese | −1.2 | percentage points of women with body mass index of 30 or more | intention-to-treat difference in prevalence between women randomly offered a voucher usable only in low-poverty tracts (with counseling) and women offered no voucher, reported by Ludwig and coauthors (2011, New England Journal of Medicine 365(16)): 4,498 women with children in public housing in high-poverty tracts in Baltimore, Boston, Chicago, Los Angeles and New York, randomized 1994 to 1998 and measured in a survey from 2008 to 2010; copied from the paper and not computed in this build | women with children in public housing in high-poverty tracts, randomized to a voucher offer from 1994 to 1998, measured 2008 to 2010 | randomized 1994 to 1998, measured 2008 to 2010 | 95% CI −5.4 to 3.0 |   | none (reported estimate) | Published estimates from cited studies | A published estimate, not a result of this build. |
| mto_takeup | 48% | percent of women offered the low-poverty voucher | share of the low-poverty voucher group that used the voucher to move, reported by Ludwig and coauthors (2011, New England Journal of Medicine 365(16)); copied from the paper and not computed in this build | women with children in public housing in high-poverty tracts, randomized to a voucher offer from 1994 to 1998, measured 2008 to 2010 | randomized 1994 to 1998 |   |   | none (reported estimate) | Published estimates from cited studies | A published estimate, not a result of this build. |
| n_acs_dropped | 203 | tracts | count of sample tracts left out of the approximation only | sample tracts that lack at least one of the ACS inputs of the approximation | ACS 2019-2023 |   |   | none (full count) | American Community Survey |   |
| n_acs_shares | 8 | ACS tract shares | definition: the number of ACS shares in the approximation | census tract shares used to approximate the model's inputs | ACS 2019-2023 |   |   | none (a definition) | American Community Survey |   |
| n_cl_counties | 3,129 | counties | number of bootstrap clusters | counties (Connecticut planning regions) with a tract in the sample | BRFSS 2022 (PLACES 2024 release); ADI 2023 |   |   | none (full count) | CDC PLACES |   |
| n_conditions | 9 | conditions | definition: the number of conditions in the count | PLACES health outcome measures that cover all adults and have an estimate in every state | PLACES 2024 release |   |   | none (a definition) | CDC PLACES |   |
| n_dropped | 741 | tracts | count of tracts left out | census tracts with a PLACES estimate that have no ADI rank | BRFSS 2022 (PLACES 2024 release); ADI 2023 |   |   | none (full count) | Area Deprivation Index |   |
| n_few_tracts | 990 | counties | count of counties with fewer tracts in the sample than the definition above, where the share moves in large steps | counties on the map | BRFSS 2022 (PLACES 2024 release); ADI 2023 |   |   | none (full count) | CDC PLACES |   |
| n_half | 1,229 | counties | count of counties with a population-weighted tract share of 50% or more | counties where at least half of adults live in tracts that are high on five or more conditions | BRFSS 2022 (PLACES 2024 release); ADI 2023 |   |   | adult population of the tract (PLACES) | CDC PLACES |   |
| n_le_tracts | 77,934 | tracts | count of tracts | tracts in the sample with a USALEEP life expectancy estimate that is not institutional | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); PLACES 2024 release |   |   | none (full count) | USALEEP life expectancy |   |
| n_map_counties | 3,128 | counties | count of counties drawn with a value (Connecticut as its eight legacy counties) | counties and county equivalents on the map with at least one tract in the sample | BRFSS 2022 (PLACES 2024 release); ADI 2023 |   |   | none (full count) | CDC PLACES |   |
| n_map_gray | 15 | counties | count of counties drawn gray | counties on the map sprite with no tract in the sample | BRFSS 2022 (PLACES 2024 release); ADI 2023 |   |   | none (full count) | CDC PLACES |   |
| n_neg_pairs | 2 | pairs of conditions | count of pairs whose 95% bootstrap interval lies below zero | pairs among the nine conditions with an interval entirely below zero | BRFSS 2022 (PLACES 2024 release); ADI 2023 |   | 36 pairs of conditions | adult population of the tract (PLACES) | CDC PLACES |   |
| n_pairs | 36 | pairs of conditions | definition: nine choose two | pairs among the nine conditions | BRFSS 2022 (PLACES 2024 release); ADI 2023 |   |   | none (a definition) | CDC PLACES |   |
| n_places_measures | 40 | measures | count of measure ids in the staged tract file | health measures that PLACES publishes for tracts | PLACES 2024 release |   |   | none (full count) | CDC PLACES |   |
| n_places_tracts | 83,522 | tracts | count of tracts | 2020 census tracts with a CDC PLACES 2024 release estimate for all nine conditions | PLACES 2024 release |   |   | none (full count) | CDC PLACES |   |
| n_pos_pairs | 32 | pairs of conditions | count of pairs whose 95% bootstrap interval lies above zero | pairs among the nine conditions with an interval entirely above zero | BRFSS 2022 (PLACES 2024 release); ADI 2023 |   | 36 pairs of conditions | adult population of the tract (PLACES) | CDC PLACES |   |
| n_tracts | 82,781 | tracts | count of tracts | census tracts with an estimate for all nine conditions and an ADI rank | BRFSS 2022 (PLACES 2024 release); ADI 2023; ACS 2019-2023 |   |   | none (full count) | CDC PLACES |   |
| none_exp | 7.4% | percent of tracts | share of tracts expected to be high on none if the nine conditions were independent across tracts (Poisson-binomial with the observed share high on each) | census tracts with an estimate for all nine conditions and an ADI rank | BRFSS 2022 (PLACES 2024 release); ADI 2023 |   |   | none (a benchmark) | CDC PLACES |   |
| none_share | 34.0% | percent of tracts | share of tracts that are high on none of the nine conditions, bootstrap interval over counties | census tracts with an estimate for all nine conditions and an ADI rank | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 95% CI 30.7% to 37.1% | 82,781 tracts | none (each tract counts once) | CDC PLACES |   |
| none_share_w | 38.1% | percent of adults | share of adults living in tracts that are high on none of the nine conditions | adults living in census tracts with an estimate for all nine conditions and an ADI rank | BRFSS 2022 (PLACES 2024 release); ADI 2023 |   | 82,781 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| oh_cancer | 2.5 | other conditions (of eight) that are also high | mean number of the other eight conditions that are also high, bootstrap interval over counties | tracts high on cancer | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 95% CI 2.4 to 2.6 | 21,087 tracts | none (each tract counts once) | CDC PLACES |   |
| oh_copd | 4.9 | other conditions (of eight) that are also high | mean number of the other eight conditions that are also high, bootstrap interval over counties | tracts high on COPD | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 95% CI 4.9 to 5.0 | 21,048 tracts | none (each tract counts once) | CDC PLACES |   |
| oh_depression | 3.1 | other conditions (of eight) that are also high | mean number of the other eight conditions that are also high, bootstrap interval over counties | tracts high on depression | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 95% CI 3.0 to 3.3 | 20,971 tracts | none (each tract counts once) | CDC PLACES |   |
| oh_indep | 2.0 | other conditions (of eight) that are also high | expected number of the other eight conditions that are also high if the conditions were independent across tracts | a tract that is high on one condition | BRFSS 2022 (PLACES 2024 release); ADI 2023 |   |   | none (a benchmark) | CDC PLACES |   |
| oh_stroke | 4.9 | other conditions (of eight) that are also high | mean number of the other eight conditions that are also high, bootstrap interval over counties | tracts high on stroke | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 95% CI 4.8 to 5.0 | 19,866 tracts | none (each tract counts once) | CDC PLACES |   |
| oh_wdiff | 0.34 | other conditions (of eight) that are also high | largest absolute difference, over the nine conditions, between the mean weighted by adults and the mean counting each tract once | tracts high on one condition | BRFSS 2022 (PLACES 2024 release); ADI 2023 |   |   | adult population of the tract (PLACES) | CDC PLACES |   |
| p65_c8 | 18.9% | percent of residents aged 65 and over | population-weighted mean of the ACS share of residents aged 65 and over, bootstrap interval over counties | residents of tracts that are high on 8 of the nine conditions, USALEEP tracts in the sample | ACS 2019-2023 | 95% CI 18.6% to 19.2% | 2,898 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| p65_c9 | 22.6% | percent of residents aged 65 and over | population-weighted mean of the ACS share of residents aged 65 and over, bootstrap interval over counties | residents of tracts that are high on 9 of the nine conditions, USALEEP tracts in the sample | ACS 2019-2023 | 95% CI 22.3% to 22.9% | 873 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| pc1_share | 59.0% | percent of the variance of the standardized prevalences | share of the variance on the first principal component of the nine standardized prevalences, bootstrap interval over counties | adults living in census tracts with an estimate for all nine conditions and an ADI rank | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 95% CI 57.9% to 60.2% | 82,781 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| r2_flex_max | 90% | percent of the variance of tract prevalence | largest weighted out-of-fold R squared of the nine tree fits, folds by county | adults living in census tracts with an estimate for all nine conditions and an ADI rank and every ACS input of the approximation | BRFSS 2022 (PLACES 2024 release); ADI 2023; ACS 2019-2023 |   | 82,578 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| r2_flex_min | 75% | percent of the variance of tract prevalence | smallest weighted out-of-fold R squared of the nine tree fits, folds by county | adults living in census tracts with an estimate for all nine conditions and an ADI rank and every ACS input of the approximation | BRFSS 2022 (PLACES 2024 release); ADI 2023; ACS 2019-2023 |   | 82,578 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| r2_max | 90% | percent of the variance of tract prevalence | largest weighted R squared of the nine fits on the ACS shares, county poverty and state effects | adults living in census tracts with an estimate for all nine conditions and an ADI rank and every ACS input of the approximation | BRFSS 2022 (PLACES 2024 release); ADI 2023; ACS 2019-2023 |   | 82,578 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| r2_min | 79% | percent of the variance of tract prevalence | smallest weighted R squared of the nine fits on the ACS shares, county poverty and state effects | adults living in census tracts with an estimate for all nine conditions and an ADI rank and every ACS input of the approximation | BRFSS 2022 (PLACES 2024 release); ADI 2023; ACS 2019-2023 |   | 82,578 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| r2w_max | 85% | percent of the variance of tract prevalence within counties | largest weighted within-county R squared of the nine fits on the ACS shares with county effects | adults living in census tracts with an estimate for all nine conditions and an ADI rank and every ACS input of the approximation | BRFSS 2022 (PLACES 2024 release); ADI 2023; ACS 2019-2023 |   | 82,578 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| r2w_min | 62% | percent of the variance of tract prevalence within counties | smallest weighted within-county R squared of the nine fits on the ACS shares with county effects | adults living in census tracts with an estimate for all nine conditions and an ADI rank and every ACS input of the approximation | BRFSS 2022 (PLACES 2024 release); ADI 2023; ACS 2019-2023 |   | 82,578 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| r65_arth | +0.64 | correlation coefficient | population-weighted correlation of the tract prevalence of arthritis with the ACS share of residents aged 65 and over, bootstrap interval over counties | adults living in census tracts with an estimate for all nine conditions and an ADI rank | BRFSS 2022 (PLACES 2024 release); ADI 2023; ACS 2019-2023 | 95% CI 0.61 to 0.66 | 82,779 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| r65_cancer | +0.81 | correlation coefficient | population-weighted correlation of the tract prevalence of cancer with the ACS share of residents aged 65 and over, bootstrap interval over counties | adults living in census tracts with an estimate for all nine conditions and an ADI rank | BRFSS 2022 (PLACES 2024 release); ADI 2023; ACS 2019-2023 | 95% CI 0.79 to 0.83 | 82,779 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| r65_chd | +0.64 | correlation coefficient | population-weighted correlation of the tract prevalence of coronary heart disease with the ACS share of residents aged 65 and over, bootstrap interval over counties | adults living in census tracts with an estimate for all nine conditions and an ADI rank | BRFSS 2022 (PLACES 2024 release); ADI 2023; ACS 2019-2023 | 95% CI 0.61 to 0.66 | 82,779 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| r65_depr | −0.17 | correlation coefficient | population-weighted correlation of the tract prevalence of depression with the ACS share of residents aged 65 and over, bootstrap interval over counties | adults living in census tracts with an estimate for all nine conditions and an ADI rank | BRFSS 2022 (PLACES 2024 release); ADI 2023; ACS 2019-2023 | 95% CI −0.21 to −0.14 | 82,779 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| r65_diab | +0.23 | correlation coefficient | population-weighted correlation of the tract prevalence of diagnosed diabetes with the ACS share of residents aged 65 and over, bootstrap interval over counties | adults living in census tracts with an estimate for all nine conditions and an ADI rank | BRFSS 2022 (PLACES 2024 release); ADI 2023; ACS 2019-2023 | 95% CI 0.19 to 0.26 | 82,779 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| r_arth_chd | +0.90 | correlation coefficient | population-weighted Pearson correlation of tract prevalence of arthritis and coronary heart disease, bootstrap interval over counties | adults living in census tracts with an estimate for all nine conditions and an ADI rank | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 95% CI 0.89 to 0.91 | 82,781 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| r_asth_obes | +0.64 | correlation coefficient | population-weighted Pearson correlation of tract prevalence of current asthma and obesity, bootstrap interval over counties | adults living in census tracts with an estimate for all nine conditions and an ADI rank | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 95% CI 0.59 to 0.69 | 82,781 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| r_cancer_arth | +0.74 | correlation coefficient | population-weighted Pearson correlation of tract prevalence of cancer and arthritis, bootstrap interval over counties | adults living in census tracts with an estimate for all nine conditions and an ADI rank | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 95% CI 0.72 to 0.75 | 82,781 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| r_cancer_asth | −0.13 | correlation coefficient | population-weighted Pearson correlation of tract prevalence of cancer and current asthma, bootstrap interval over counties | adults living in census tracts with an estimate for all nine conditions and an ADI rank | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 95% CI −0.20 to −0.07 | 82,781 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| r_cancer_chd | +0.63 | correlation coefficient | population-weighted Pearson correlation of tract prevalence of cancer and coronary heart disease, bootstrap interval over counties | adults living in census tracts with an estimate for all nine conditions and an ADI rank | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 95% CI 0.60 to 0.66 | 82,781 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| r_cancer_depr | 0.00 | correlation coefficient | population-weighted Pearson correlation of tract prevalence of cancer and depression, bootstrap interval over counties | adults living in census tracts with an estimate for all nine conditions and an ADI rank | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 95% CI −0.07 to 0.06 | 82,781 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| r_cancer_diab | +0.05 | correlation coefficient | population-weighted Pearson correlation of tract prevalence of cancer and diagnosed diabetes, bootstrap interval over counties | adults living in census tracts with an estimate for all nine conditions and an ADI rank | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 95% CI 0.00 to 0.11 | 82,781 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| r_cancer_obes | −0.11 | correlation coefficient | population-weighted Pearson correlation of tract prevalence of cancer and obesity, bootstrap interval over counties | adults living in census tracts with an estimate for all nine conditions and an ADI rank | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 95% CI −0.18 to −0.05 | 82,781 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| r_core_max | +0.90 | correlation coefficient | largest population-weighted correlation among the six pairs of coronary heart disease, COPD, diagnosed diabetes and stroke (Diabetes and Stroke), bootstrap interval over counties | adults living in census tracts with an estimate for all nine conditions and an ADI rank | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 95% CI 0.89 to 0.92 | 82,781 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| r_core_min | +0.74 | correlation coefficient | smallest population-weighted correlation among the six pairs of coronary heart disease, COPD, diagnosed diabetes and stroke (Coronary heart disease and Diabetes), bootstrap interval over counties | adults living in census tracts with an estimate for all nine conditions and an ADI rank | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 95% CI 0.71 to 0.76 | 82,781 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| r_depr_diab | +0.06 | correlation coefficient | population-weighted Pearson correlation of tract prevalence of depression and diagnosed diabetes, bootstrap interval over counties | adults living in census tracts with an estimate for all nine conditions and an ADI rank | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 95% CI 0.01 to 0.11 | 82,781 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| r_mean | +0.49 | correlation coefficient (mean of the 36 pairs) | mean of the population-weighted Pearson correlations over the 36 pairs of conditions, bootstrap interval over counties | adults living in census tracts with an estimate for all nine conditions and an ADI rank | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 95% CI 0.48 to 0.51 | 82,781 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| r_mean_acs | +0.49 | correlation coefficient (mean of the 36 pairs) | mean pairwise correlation among the nine conditions on the tracts the approximation uses, bootstrap interval over counties | adults living in census tracts with an estimate for all nine conditions and an ADI rank and every ACS input of the approximation | BRFSS 2022 (PLACES 2024 release); ADI 2023; ACS 2019-2023 | 95% CI 0.48 to 0.51 | 82,578 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| r_stroke_chd | +0.86 | correlation coefficient | population-weighted Pearson correlation of tract prevalence of stroke and coronary heart disease, bootstrap interval over counties | adults living in census tracts with an estimate for all nine conditions and an ADI rank | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 95% CI 0.85 to 0.87 | 82,781 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| r_stroke_copd | +0.90 | correlation coefficient | population-weighted Pearson correlation of tract prevalence of stroke and COPD, bootstrap interval over counties | adults living in census tracts with an estimate for all nine conditions and an ADI rank | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 95% CI 0.89 to 0.90 | 82,781 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| r_stroke_diab | +0.90 | correlation coefficient | population-weighted Pearson correlation of tract prevalence of stroke and diagnosed diabetes, bootstrap interval over counties | adults living in census tracts with an estimate for all nine conditions and an ADI rank | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 95% CI 0.89 to 0.92 | 82,781 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| r_within_adi | +0.36 | correlation coefficient (mean of the 36 pairs) | mean pairwise correlation among the nine conditions after subtracting the mean of each ADI tenth, bootstrap interval over counties | adults living in census tracts with an estimate for all nine conditions and an ADI rank | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 95% CI 0.35 to 0.38 | 82,781 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| radi_cancer | 0.00 | correlation coefficient | population-weighted correlation of the tract prevalence of cancer with the ADI national percentile, bootstrap interval over counties | adults living in census tracts with an estimate for all nine conditions and an ADI rank | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 95% CI −0.07 to 0.07 | 82,781 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| rdec_d1 | +0.40 | correlation coefficient (mean of the 36 pairs) | mean pairwise correlation among the nine conditions inside one tenth of tracts, bootstrap interval over counties | adults in the least deprived tenth of tracts | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 95% CI 0.37 to 0.44 | 8,274 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| rdec_d10 | +0.34 | correlation coefficient (mean of the 36 pairs) | mean pairwise correlation among the nine conditions inside one tenth of tracts, bootstrap interval over counties | adults in the most deprived tenth of tracts | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 95% CI 0.32 to 0.36 | 8,213 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| rdec_diff | +0.06 | correlation coefficient (mean of the 36 pairs), least deprived tenth minus most deprived tenth | mean pairwise correlation inside the least deprived tenth minus that inside the most deprived tenth, bootstrap interval over counties (draws shared) | adults in the least and the most deprived tenth of tracts | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 95% CI 0.03 to 0.11 | 82,781 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| rdec_mid_max | +0.36 | correlation coefficient (mean of the 36 pairs) | largest of the eight mean pairwise correlations inside the second to ninth tenth | adults in the second to ninth tenth of tracts | BRFSS 2022 (PLACES 2024 release); ADI 2023 |   | 82,781 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| rdec_mid_min | +0.35 | correlation coefficient (mean of the 36 pairs) | smallest of the eight mean pairwise correlations inside the second to ninth tenth | adults in the second to ninth tenth of tracts | BRFSS 2022 (PLACES 2024 release); ADI 2023 |   | 82,781 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| reg_midwest | 20.2% | percent of adults | share of adults living in tracts that are high on five or more conditions | adults in the Midwest Census region living in the analysis tracts | BRFSS 2022 (PLACES 2024 release); ADI 2023 |   |   | adult population of the tract (PLACES) | CDC PLACES |   |
| reg_northeast | 8.5% | percent of adults | share of adults living in tracts that are high on five or more conditions | adults in the Northeast Census region living in the analysis tracts | BRFSS 2022 (PLACES 2024 release); ADI 2023 |   |   | adult population of the tract (PLACES) | CDC PLACES |   |
| reg_south | 25.8% | percent of adults | share of adults living in tracts that are high on five or more conditions | adults in the South Census region living in the analysis tracts | BRFSS 2022 (PLACES 2024 release); ADI 2023 |   |   | adult population of the tract (PLACES) | CDC PLACES |   |
| reg_west | 7.0% | percent of adults | share of adults living in tracts that are high on five or more conditions | adults in the West Census region living in the analysis tracts | BRFSS 2022 (PLACES 2024 release); ADI 2023 |   |   | adult population of the tract (PLACES) | CDC PLACES |   |
| rowmean_cancer | +0.22 | correlation coefficient (mean over the other eight conditions) | mean of the correlations of cancer with the other eight conditions, bootstrap interval over counties | adults living in census tracts with an estimate for all nine conditions and an ADI rank | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 95% CI 0.19 to 0.25 | 82,781 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| rowmean_copd | +0.68 | correlation coefficient (mean over the other eight conditions) | mean of the correlations of COPD with the other eight conditions, bootstrap interval over counties | adults living in census tracts with an estimate for all nine conditions and an ADI rank | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 95% CI 0.66 to 0.69 | 82,781 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| rowmean_depression | +0.26 | correlation coefficient (mean over the other eight conditions) | mean of the correlations of depression with the other eight conditions, bootstrap interval over counties | adults living in census tracts with an estimate for all nine conditions and an ADI rank | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 95% CI 0.22 to 0.30 | 82,781 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| s_adj_county | +0.52 | correlation coefficient (mean of the 36 pairs) | mean pairwise correlation among the nine conditions after removing a fit on eight ACS tract shares and county effects; the fit is not refitted inside the bootstrap, so the interval conditions on it | adults living in census tracts with an estimate for all nine conditions and an ADI rank and every ACS input of the approximation | BRFSS 2022 (PLACES 2024 release); ADI 2023; ACS 2019-2023 | 95% CI 0.51 to 0.52 | 82,578 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| s_adj_state | +0.47 | correlation coefficient (mean of the 36 pairs) | mean pairwise correlation among the nine conditions after removing a fit on eight ACS tract shares, the county poverty rate and state effects; the fit is not refitted inside the bootstrap, so the interval conditions on it | adults living in census tracts with an estimate for all nine conditions and an ADI rank and every ACS input of the approximation | BRFSS 2022 (PLACES 2024 release); ADI 2023; ACS 2019-2023 | 95% CI 0.46 to 0.48 | 82,578 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| s_county | +0.43 | correlation coefficient (mean of the 36 pairs) | mean pairwise correlation among the nine conditions after subtracting the mean of each county; the fit is not refitted inside the bootstrap, so the interval conditions on it | adults living in census tracts with an estimate for all nine conditions and an ADI rank | BRFSS 2022 (PLACES 2024 release); ADI 2023; ACS 2019-2023 | 95% CI 0.42 to 0.43 | 82,781 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| s_flex | +0.45 | correlation coefficient (mean of the 36 pairs) | mean pairwise correlation among the nine conditions after removing a gradient-boosted-trees fit on the ACS tract shares, the county poverty rate and the state; the fit is not refitted inside the bootstrap, so the interval conditions on it | adults living in census tracts with an estimate for all nine conditions and an ADI rank and every ACS input of the approximation | BRFSS 2022 (PLACES 2024 release); ADI 2023; ACS 2019-2023 | 95% CI 0.44 to 0.47 | 82,578 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| s_state | +0.48 | correlation coefficient (mean of the 36 pairs) | mean pairwise correlation among the nine conditions after subtracting the mean of each state; the fit is not refitted inside the bootstrap, so the interval conditions on it | adults living in census tracts with an estimate for all nine conditions and an ADI rank | BRFSS 2022 (PLACES 2024 release); ADI 2023; ACS 2019-2023 | 95% CI 0.47 to 0.50 | 82,781 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| sens_all_tenth | 14 | tracts | count of tracts high on all nine conditions when a tract is high if it is in the highest 10% of tracts | census tracts with an estimate for all nine conditions and an ADI rank | BRFSS 2022 (PLACES 2024 release); ADI 2023 |   |   | none (full count) | CDC PLACES |   |
| sens_all_third | 2,679 | tracts | count of tracts high on all nine conditions when a tract is high if it is in the highest 33.3% of tracts | census tracts with an estimate for all nine conditions and an ADI rank | BRFSS 2022 (PLACES 2024 release); ADI 2023 |   |   | none (full count) | CDC PLACES |   |
| sens_exp_tenth | 0.0 | tracts | number of tracts expected to be high on all nine conditions if they were independent, with the highest 10% of tracts counted as high | census tracts with an estimate for all nine conditions and an ADI rank | BRFSS 2022 (PLACES 2024 release); ADI 2023 |   |   | none (a benchmark) | CDC PLACES |   |
| sens_exp_third | 4.5 | tracts | number of tracts expected to be high on all nine conditions if they were independent, with the highest 33.3% of tracts counted as high | census tracts with an estimate for all nine conditions and an ADI rank | BRFSS 2022 (PLACES 2024 release); ADI 2023 |   |   | none (a benchmark) | CDC PLACES |   |
| sens_none_tenth | 64.9% | percent of tracts | share of tracts high on none of the nine conditions when a tract is high if it is in the highest 10% of tracts | census tracts with an estimate for all nine conditions and an ADI rank | BRFSS 2022 (PLACES 2024 release); ADI 2023 |   | 82,781 tracts | none (each tract counts once) | CDC PLACES |   |
| sens_none_third | 23.4% | percent of tracts | share of tracts high on none of the nine conditions when a tract is high if it is in the highest 33.3% of tracts | census tracts with an estimate for all nine conditions and an ADI rank | BRFSS 2022 (PLACES 2024 release); ADI 2023 |   | 82,781 tracts | none (each tract counts once) | CDC PLACES |   |
| us_chronic_high | 2.0 | conditions (of nine) | mean number of the nine conditions that are high, population-weighted mean over all the tracts of the tenths pooled, bootstrap interval over counties (the draws of the tenths) | adults in all the tracts of the comparison: census tracts with an estimate for all nine conditions and an ADI rank | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 95% CI 1.8 to 2.1 | 82,781 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| vc_between | 52.2% | percent of the variance of the tract count of high conditions | population-weighted share of the variance of the count of high conditions that lies between counties, including between states; a decomposition of every tract | adults living in census tracts with an estimate for all nine conditions and an ADI rank | BRFSS 2022 (PLACES 2024 release); ADI 2023 |   | 82,781 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| vc_county | 29.8% | percent of the variance of the tract count of high conditions | population-weighted share of the variance of the count of high conditions that lies between counties within states; a decomposition of every tract, so it has no sampling interval | adults living in census tracts with an estimate for all nine conditions and an ADI rank | BRFSS 2022 (PLACES 2024 release); ADI 2023 |   | 82,781 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| vc_state | 22.4% | percent of the variance of the tract count of high conditions | population-weighted share of the variance of the count of high conditions that lies between states; a decomposition of every tract, so it has no sampling interval | adults living in census tracts with an estimate for all nine conditions and an ADI rank | BRFSS 2022 (PLACES 2024 release); ADI 2023 |   | 82,781 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| vc_within | 47.8% | percent of the variance of the tract count of high conditions | population-weighted share of the variance of the count of high conditions that lies between tracts within counties; a decomposition of every tract, so it has no sampling interval | adults living in census tracts with an estimate for all nine conditions and an ADI rank | BRFSS 2022 (PLACES 2024 release); ADI 2023 |   | 82,781 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| wv_share | 81.7% | percent of adults | share of adults living in tracts that are high on five or more conditions | adults in West Virginia living in the analysis tracts | BRFSS 2022 (PLACES 2024 release); ADI 2023 |   |   | adult population of the tract (PLACES) | CDC PLACES |   |

## Definitions

- Census tract: A small area drawn by the Census Bureau to stand in for a neighborhood. Boundaries are redrawn after each decennial census.
- Area Deprivation Index (ADI): A score that ranks neighborhoods by income, education, employment and housing conditions drawn from the American Community Survey. In this book a higher national percentile means a more deprived neighborhood.
- Modeled estimate: A figure produced by a statistical model instead of a direct count or measurement, usually for small areas where direct data are thin, borrowing strength from surveys, neighboring areas or other data. Neighboring areas can share errors, so single extremes call for caution.
- Population-weighted: Averaged so that each place counts in proportion to its number of residents, which lets a large county move the result more than a small one. An unweighted average counts every place equally.
- Principal components: A method that combines many correlated measures into a few summary scores, ordered by how much of the variation among places each captures. The first component is the single weighted combination that captures the most.
- Fixed effects: A regression adjustment that gives each group, such as each county, its own baseline. The estimate then rests on differences within groups.
- USALEEP: The U.S. Small-area Life Expectancy Estimates Project, a National Center for Health Statistics release of life expectancy at birth for most census tracts. It covers deaths from 2010 to 2015 only and has not been updated.
- Life expectancy at birth: The average number of years a newborn would live if death rates at every age stayed as they were in the period measured. It summarizes the death rates of a period and is not a forecast for any child.
- Ecological inference: Drawing conclusions about individuals from data about places or groups. A pattern across counties need not hold for the people inside them.
- Bootstrap interval: A range for an estimate found by recomputing it on many random resamples of the data and reading the range from the results. It shows how much the estimate would vary from sample to sample.

## Methods

The chapter compares modeled prevalences of nine chronic conditions across census tracts, from the CDC PLACES 2024 release, with correlations, means and shares of adults weighted by the adult population of each tract. A tract is high on a condition when its prevalence is in the highest 25% of tracts, and the count of high conditions describes tracts and not people. The estimates come from a model that fits every condition to shared inputs, so part of the correlation among conditions is built in; the adjustments here approximate those inputs from census data and are not the model. Life expectancy covers deaths in 2010 to 2015, a different period from the PLACES estimates.

### Which tracts and conditions are compared

The tract file of CDC PLACES (2024 release) gives one modeled prevalence per condition per tract. The chapter uses nine health outcome measures that cover all adults and have an estimate in every state: arthritis, cancer (non-skin) or melanoma, current asthma, coronary heart disease, COPD, depression, diagnosed diabetes, obesity and stroke.

The sample is the tracts with all nine conditions and an ADI rank. The approximations of the model's inputs use the tracts with every ACS input, which leaves out the tracts of the District of Columbia and a few others.

Explains Table 1.

### What counts as high on a condition

A tract is high on a condition when its prevalence is in the highest 25% of tracts, with tied values at their mean rank, and the count is the number of conditions on which a tract is high; it describes tracts and not people.

The benchmark for independent conditions is a Poisson-binomial distribution with each condition's observed share of high tracts.

Explains Table 2, Table 3 and Table 4.

### How tracts are weighted and where life expectancy comes from

Correlations, means and shares of adults are weighted by the tract's adult population; counts and shares of tracts count each tract once.

Life expectancy is the USALEEP estimate for 2010 to 2015 on 2020 tracts, weighted by ACS 2019-2023 tract population.

Explains Table 5 and Table 8.

### How the model's inputs are approximated

The adjustments for the model's inputs are weighted least-squares fits of each condition on eight ACS tract shares with squares and pairwise products, with state effects and a county poverty rate (or county effects), and a gradient-boosted-trees fit with folds by county; they are fitted once and not refitted inside the bootstrap, so their intervals condition on the fit.

They approximate the PLACES model, which uses individual age, sex, race, ethnicity and education, county poverty, and county and state effects.

Explains Table 6.

### How uncertainty is measured

The variance split of the count is a weighted decomposition of every tract and has no interval.

Intervals are 95% percentile intervals from a bootstrap that draws whole counties with replacement (3,129 clusters; 2,000 replicates, seed 12345), with the same draws for every statistic, so a difference has an interval from shared draws.

Explains Table 7.

### Four rankings of the same places

The comparison follows the book's ranking switch rule, fixed on October 28, 2026 before any of these gradients was computed. The main gradient of the chapter is recomputed under four rankings of the same places, with the outcome, period, weights, bins and intervals unchanged: the published ADI; a standardized index that applies the ADI's coefficients to its measures after standardizing each one across tracts; median household income; and the share of adults 25 and older with a bachelor's degree or more. The three alternatives come from ACS tract estimates for 2019 to 2023, the period of the ADI.

All four rankings use the same places, the 75,596 tracts that have the outcome and all four rankings, so the ADI panel can differ slightly from the chapter's main figure. Tracts enter only if they have every input of the standardized index, which leaves out tracts without a median rent, home value or mortgage cost.

A pattern counts as changed when the worse end switches sides, when the two ends differ under one ranking and not the other, or when the rank correlation of bin order and bin mean falls on the other side of 0.9 and moves by at least 0.1.

Explains Table 9.

## What these data cannot show

A count of high conditions describes a tract and not a person. A correlation across places can differ widely in size from the correlation among the people who live in them,[5] so the overlap here cannot be read as people carrying several conditions at once, and no figure in this chapter says how many conditions an adult has. Drawing that conclusion from place data is ecological inference. The count of conditions per person comes from surveys that ask each adult, and Chapter 7.2, Health behaviors, person by person (https://vishalsingh.org/zipcode/health-behaviors-person-by-person) uses one for behaviors.

The PLACES values are modeled estimates, the tract file has no intervals, and the bands in this chapter describe the sampling of counties and not the model's own uncertainty. CDC advises against using the estimates to rank the overall health of an area or to track change at the local level, and this chapter does neither.[6] It does compare conditions with each other and with the ADI, and every such comparison carries the shared inputs described above. The approximations show that measured census shares do not remove the overlap, but they cannot separate what the model builds in from what the neighborhoods contribute. The staged tract file holds crude prevalence only, so age is part of every pattern, most visibly for cancer, arthritis and heart disease.

High is relative. A tract in the highest 25% of tracts for a condition is high compared with other tracts and not high by any clinical standard, and the cutoff moves every count, as the table shows. High blood pressure and high cholesterol are left out because the 2024 release has no estimate of either for the 5,077 tracts of Florida. Outside Florida the mean correlation of high blood pressure with the nine conditions is +0.65 (95% CI 0.64 to 0.67), against +0.51 among the nine on the same tracts, so it follows the same shared pattern.

Of the 83,522 tracts with PLACES estimates, 82,781 also have an ADI rank, and the other 741 are left out. Counties are the clusters of the bootstrap, 3,129 of them, with Connecticut's planning regions standing in for its counties, while the map draws Connecticut as its eight legacy counties. The map has 15 gray counties with no tract in the sample, and 990 counties have fewer than 5 tracts, so their shares move in large steps. Puerto Rico is not covered by PLACES and is not drawn.

The conditions are measured once, in 2022, so the data say nothing about which condition comes first, whether tracts move into or out of the high group, or whether living in a tract changes any of the prevalences. The ADI describes 2019 to 2023 and the life expectancy check describes 2010 to 2015, so the sources do not describe one moment.

## References

1. Greenlund, Kurt J.; Lu, Hua; Wang, Yan; Matthews, Kevin A.; LeClercq, Jennifer M.; Lee, Benjamin; Carlson, Susan A. (2022). PLACES: Local Data for Better Health. Preventing Chronic Disease 19: 210459. https://doi.org/10.5888/pcd19.210459
2. Ward, Brian W.; Schiller, Jeannine S. (2013). Prevalence of Multiple Chronic Conditions Among US Adults: Estimates From the National Health Interview Survey, 2010. Preventing Chronic Disease 10: 120203. https://doi.org/10.5888/pcd10.120203
3. Kind, Amy J. H.; Buckingham, William R. (2018). Making Neighborhood-Disadvantage Metrics Accessible: The Neighborhood Atlas. New England Journal of Medicine 378(26): 2456-2458. https://doi.org/10.1056/NEJMp1802313
4. Ludwig, Jens; Sanbonmatsu, Lisa; Gennetian, Lisa; Adam, Emma; Duncan, Greg J.; Katz, Lawrence F.; Kessler, Ronald C.; Kling, Jeffrey R.; Lindau, Stacy Tessler; Whitaker, Robert C.; McDade, Thomas W. (2011). Neighborhoods, Obesity, and Diabetes: A Randomized Social Experiment. New England Journal of Medicine 365(16): 1509-1519. https://doi.org/10.1056/NEJMsa1103216
5. Robinson, W. S. (1950). Ecological Correlations and the Behavior of Individuals. American Sociological Review 15(3): 351-357. https://doi.org/10.2307/2087176
6. Centers for Disease Control and Prevention (2024). PLACES Frequently Asked Questions: Using the Data. CDC PLACES website, page last reviewed August 29, 2024. https://www.cdc.gov/places/faqs/using-data/index.html

## Provenance

- Pipeline version: 0.1.0
- Build script: book2/chapters/multiple-chronic-conditions/build.py (sha256 68473ef38c471eddc146bfd87b4d1844dd1a3d188dd8f118c39f671c4aabdda6)
- Built on: 2026-10-09
- Site commit: c2b1ff3309cb4a10833f884da91ac0057a66ad0a

Inputs:

| File | Size (bytes) | SHA-256, first 1 MB | SHA-256, last 1 MB |
| --- | ---: | --- | --- |
| book2/chapters/multiple-chronic-conditions/conditions.py | 16,370 | c914a81074e57f22cee1d2ec8d1fd3e3092778970e57ba2ba1dd6513d228a6cd | c914a81074e57f22cee1d2ec8d1fd3e3092778970e57ba2ba1dd6513d228a6cd |
| data/analytic/tract_atlas.parquet | 39,659,925 | 29d5e4d943552d54aeb6ba92aadc939d9bd15bdea553c78357dbd516757be627 | 1e10d1ecf722340f23ee2212d4999a90250f85b65fa89b39e32db06d9472e22e |
| data/geo/dim_geo.parquet | 1,180,921 | 29d5e4d943552d54aeb6ba92aadc939d9bd15bdea553c78357dbd516757be627 | e58c8c488df54ab3187f5461f179322afdc49c9bd2315404003fe1272eea32c3 |
| data/raw/acs/deprivation_components_2019_2023/tract.parquet | 13,197,793 | 5545cde2cbe96a8b56e468d65388248a241aa6bd300d3c9e325d08d3f86b11da | 5c8ec454d066988b0f1a01aaa6ed34ebfc05d162e01cf9c6bb01f9198cbaff12 |
| data/staged/adi/adi_atlas_tract2020_2023.parquet | 798,078 | 5e409bc212ff871def780fa27b4cf8b616d9da420b2b9f6455517eddcd8c1a95 | 5e409bc212ff871def780fa27b4cf8b616d9da420b2b9f6455517eddcd8c1a95 |
| data/staged/crosswalks/ct_tract2020_to_old_county.parquet | 15,842 | 961ef39f31596ef65fca2c67d5f2f63e533c5dc9b49745d3bb6437844e4b831d | 961ef39f31596ef65fca2c67d5f2f63e533c5dc9b49745d3bb6437844e4b831d |
| data/staged/health/places_tract2020_latest.parquet | 5,919,159 | af4c3b0d1205b7d84b81676a9529ef73323b2dbeb9d54e235f5efeebfad83177 | a2f19029ee2ef032be87e0d5288d88d489afd52379279833f178baa021df3dbf |
| data/staged/life_expectancy/usaleep_tract2020.parquet | 1,316,169 | 9c9c7f103791b75eddba74e68a89804efdf2486b6b66bb4bc02a46ba5bf8bc23 | 27317107be2465d9fd5e83271dea0cd6e915ee419e2802bedc28747d0a41d38f |
| data/staged/ses/acs5_tract2020_2019_2023.parquet | 7,232,424 | 5fe6f730a35225f14d5d8abfc6b9bce58b52e545f309a7eac74f1dae74d7a31c | 7b4a2475b3852d3f91804716a04566ef8ceadacec2be9fb85f5eb642078efac6 |
| data/staged/svi/svi_tract2020_2022.parquet | 17,740,415 | cb3860d8edc2aaf1f2a02211a58f7a0b80457eb6ad9a37b5ec8abb2d5da9da10 | e371a4a696d5bed05d54ad7c5b6ca2504508805b847ec415c2322be7f66ec2b2 |

Outputs:

| File | SHA-256 |
| --- | --- |
| facts.json | 4e51b97b62ba016b5bafc582e374b83382ed094ee0257de05d7d6736e14c222f |
| figures/adults_high_count_map.json | 1486fc45cb621174c3c7e83ce34526bbd7eecbc017bdc2ae588544c7939d6902 |
| figures/alt_rankings.json | 5c95d266eb3e2f097c8cf6025e189e39af38e1c1609ca88792c64fe2e055896f |
| figures/condition_correlations.json | 9794c4a769bbad7289cafc87765582217f5e5756bb0699f9dea84f9fa1a7fc8c |
| figures/correlation_stages.json | 537492838d3812cdc91f6d2297e08490e84477f1fe0daec54175561dceea173a |
| figures/cutoff_sensitivity.json | 844cb702276951831dd9fc3a6f2f3e951d3dbad95237e3f21a99cfc76dc508c8 |
| figures/high_count_by_adi.json | fa06d9dd565c8bd59515de2b4ecca0fdc50503679b1b5e1f6f7fc8aeb2686ac8 |
| figures/high_count_distribution.json | 347bffb3a04563835b9651a4efc1fdfe8a4ae41f63a71cd4c007f9873c01218e |
| figures/life_expectancy_by_count.json | b01186ea160253386dad8661c62220db64e1eb83d1eebb7af007fbcc61349f08 |
| figures/others_high.json | 80fdfbcb15740d87cc147f79bd7f1674d718dd43de42094ef245cc913fc83d64 |
| meta.json | 6178ca31c08233752668d7c2cef86587894ed6b7c195c100704db6bf0070cced |
| public/data/multiple-chronic-conditions/adults_high_count_map.json | 2f4e829c2701ecc1904b87f42b2742f16e4ed11af286db1b98875d439d65f7d2 |
| refs.json | 28ec62e2f073067cf5c0e5a00f90474110c06bf19c8ce7c6dcf10c770fffffb3 |

## Machine formats

- Report (HTML): https://vishalsingh.org/zipcode/multiple-chronic-conditions/report
- Report (Markdown): https://vishalsingh.org/zipcode/multiple-chronic-conditions/report.md
- Every estimate (JSON): https://vishalsingh.org/zipcode/report/multiple-chronic-conditions/facts.json
- Table 1, figure "condition_correlations" (JSON): https://vishalsingh.org/zipcode/report/multiple-chronic-conditions/figures/condition_correlations.json
- Table 2, figure "others_high" (JSON): https://vishalsingh.org/zipcode/report/multiple-chronic-conditions/figures/others_high.json
- Table 3, figure "high_count_distribution" (JSON): https://vishalsingh.org/zipcode/report/multiple-chronic-conditions/figures/high_count_distribution.json
- Table 4, figure "cutoff_sensitivity" (JSON): https://vishalsingh.org/zipcode/report/multiple-chronic-conditions/figures/cutoff_sensitivity.json
- Table 5, figure "high_count_by_adi" (JSON): https://vishalsingh.org/zipcode/report/multiple-chronic-conditions/figures/high_count_by_adi.json
- Table 6, figure "correlation_stages" (JSON): https://vishalsingh.org/zipcode/report/multiple-chronic-conditions/figures/correlation_stages.json
- Table 7, figure "adults_high_count_map" (JSON): https://vishalsingh.org/zipcode/report/multiple-chronic-conditions/figures/adults_high_count_map.json
- Table 8, figure "life_expectancy_by_count" (JSON): https://vishalsingh.org/zipcode/report/multiple-chronic-conditions/figures/life_expectancy_by_count.json
- Table 9, figure "alt_rankings" (JSON): https://vishalsingh.org/zipcode/report/multiple-chronic-conditions/figures/alt_rankings.json
- Payload of "adults_high_count_map" (JSON): https://vishalsingh.org/zipcode/data/multiple-chronic-conditions/adults_high_count_map.json
- Index of every report: https://vishalsingh.org/zipcode/llms.txt; all reports in one file: https://vishalsingh.org/zipcode/llms-full.txt

---

# Disability, retirement and Medicare: report

Chapter 7.4 of ZIP Code Destiny, by Vishal Singh (NYU Stern). Generated from the chapter's data files; the reading version is at https://vishalsingh.org/zipcode/disability-retirement-and-medicare. Table n of this report is figure n of the chapter.

## Identification

- Title: Disability, retirement and Medicare
- Label: The twilight
- Chapter: 7.4 (Part VII, Health and care; An American life)
- Question: How do disability, the mix of Social Security benefits, Medicare spending and life expectancy in old age differ by neighborhood and county income?
- Dek: In CDC PLACES model estimates from 2022 survey responses, the share of adults with any disability was 23.3% in the least deprived tenth of census tracts and 42.9% in the most deprived.
- Grain: neighborhoods
- Status: published
- Published: 2026-10-09
- Updated: not revised
- Data sources:
  - CDC PLACES. Any disability and all teeth lost among adults, census tracts, 2024 release (BRFSS 2022 responses); years: 2024 release (tracts); 2025 release (ZCTAs, 2023 data year); weights: adult population of the tract; estimated residents aged 65 and over for tooth loss
  - Area Deprivation Index. Area Deprivation Index national rank, 2023, tracts and ZCTAs; years: 2023 release (version 4.0.1; built on ACS 2019-2023); weights: none (index); the county rank is the population-weighted mean of tract ranks
  - American Community Survey. ACS 2019-2023 tract and ZCTA population, share aged 65 and over and median household income; years: 2019-2023 (5-year); weights: ACS population
  - SSA beneficiaries by ZIP code. OASDI beneficiaries and benefits by ZIP code, December 2024; years: December 2024; weights: none (full counts); beneficiaries or retired workers where pooled
  - IRS Statistics of Income. Taxable pensions and annuities by ZIP code, tax year 2022; years: tax year 2022; weights: none (full counts); dollars where pooled
  - Medicare Geographic Variation. Price-standardized Medicare payments, emergency department visits and service lines per Original Medicare beneficiary, county, 2024; years: 2024 (from the 2014-2024 file); weights: fee-for-service beneficiaries
  - County Health Rankings. County Health Rankings 2024: life expectancy (deaths 2019-2021), population and county measures; years: 2024 release; weights: county residents; each county once where stated
  - Health Inequality Project. Expected age at death at age 40 by national household-income quartile and sex, county, 2001-2014 (Online Data Table 11); years: 2001-2014 (pooled); weights: county residents (County Health Rankings 2024)
- Canonical URL: https://vishalsingh.org/zipcode/disability-retirement-and-medicare
- Report: https://vishalsingh.org/zipcode/disability-retirement-and-medicare/report
- Markdown: https://vishalsingh.org/zipcode/disability-retirement-and-medicare/report.md
- BibTeX key: singh2026disabilityretirementandmedicare

```bibtex
@misc{singh2026disabilityretirementandmedicare,
  author = {Singh, Vishal},
  title = {Disability, retirement and Medicare},
  year = {2026},
  url = {https://vishalsingh.org/zipcode/disability-retirement-and-medicare},
  note = {In ZIP Code Destiny. Published 9 October 2026.}
}
```

## Principal estimates

The facts the dek and the opening prose cite, in order of appearance.

| Estimand | Population | Period | Value | Interval | n | Weights | Source | Fact id |
| --- | --- | --- | ---: | --- | ---: | --- | --- | --- |
| mean of tract prevalence (crude, modeled), weighted by adult population, bootstrap interval over counties | adults in the least deprived tenth of tracts by ADI national rank | 2024 release of CDC PLACES (BRFSS responses from 2022); ADI 2023 | 23.3% | 95% CI 22.3% to 24.1% | 8,274 tracts | adult population of the tract (PLACES) | CDC PLACES | dis_d1 |
| mean of tract prevalence (crude, modeled), weighted by adult population, bootstrap interval over counties | adults in the most deprived tenth of tracts by ADI national rank | 2024 release of CDC PLACES (BRFSS responses from 2022); ADI 2023 | 42.9% | 95% CI 42.5% to 43.2% | 8,213 tracts | adult population of the tract (PLACES) | CDC PLACES | dis_d10 |
| share of the fifth's beneficiaries, pooled over its ZIP codes, bootstrap interval over counties | OASDI beneficiaries in the most deprived fifth of ZIP codes by ADI national rank | December 2024 | 16.8% | 95% CI 16.4% to 17.3% | 6,363 ZIP codes | beneficiaries (each ZIP code counts by its beneficiary count) | SSA beneficiaries by ZIP code | dis_q5 |
| share of the fifth's beneficiaries, pooled over its ZIP codes, bootstrap interval over counties | OASDI beneficiaries in the least deprived fifth of ZIP codes by ADI national rank | December 2024 | 7.2% | 95% CI 6.9% to 7.5% | 6,363 ZIP codes | beneficiaries (each ZIP code counts by its beneficiary count) | SSA beneficiaries by ZIP code | dis_q1 |

## Findings

The share of adults with any disability, as CDC PLACES models it from 2022 survey responses, was 42.9% (95% CI 42.5% to 43.2%) in the most deprived tenth of census tracts and 23.3% (95% CI 22.3% to 24.1%) in the least deprived. The PLACES model and the deprivation index both draw on American Community Survey data, so part of that gradient is built in. The share of Social Security beneficiaries who were disabled workers in December 2024 was 16.8% (95% CI 16.4% to 17.3%) in the most deprived fifth of ZIP codes and 7.2% (95% CI 6.9% to 7.5%) in the least deprived. This chapter follows those two gradients into the programs of old age. It asks whether Medicare payments per beneficiary and the expected age at death of the poorest income quartile also move with deprivation, at the grain of tracts, ZIP codes and counties.

Deprivation is measured by the Area Deprivation Index (ADI), a national percentile rank on which a higher value means a more deprived neighborhood.[1] The sources differ in kind. The Social Security and Medicare files are counts and payments from administrative records, while PLACES and the Health Inequality Project report modeled estimates. Every figure names its unit and its weights.

### Disability rises with neighborhood deprivation, and age accounts for little of it

CDC PLACES estimates, for every census tract, the share of adults aged 18 and over who report difficulty with hearing, vision, thinking, walking, self-care or living independently, using survey responses from 2022 in its 2024 release.[2] Across 82,779 tracts, weighted by adult population, the share with any disability was 23.3% (95% CI 22.3% to 24.1%) in the least deprived tenth of tracts and 42.9% (95% CI 42.5% to 43.2%) in the most deprived, 1.8 (95% CI 1.8 to 1.9) times as high (Table 1).

Among adults aged 65 and over, the share who have lost all their teeth was 6.9% (95% CI 6.2% to 7.9%) and 25.1% (95% CI 24.7% to 25.6%), 3.6 (95% CI 3.2 to 4.0) times as high. The means of both measures rise at every step across the tenths, though the intervals of some neighboring tenths overlap, and tooth loss rises fastest at the deprived end. Across tracts, disability correlates +0.69 (95% CI 0.65 to 0.73) with the ADI rank.

PLACES does not count anyone. It fits a model to survey responses and applies it to each tract using census and American Community Survey inputs, and the ADI is built from American Community Survey data as well, so part of these gradients is built in. The tract file carries no intervals of its own, and the intervals here come from resampling counties.

 (Note: A person has any disability if they report difficulty in at least one of the six areas. Values are crude prevalence, not adjusted for age, so a tract with more older residents would have a higher value if age mattered much.)

The age of residents does not account for the gradient. Each ten-point step in the ADI rank goes with 2.0 (95% CI 1.8 to 2.1) points more disability, and holding the share of older residents fixed leaves the slope at 2.0 (95% CI 1.8 to 2.1).

Residents aged 65 and over were 17.3% (95% CI 16.6% to 18.0%) of the population in the least deprived tenth of tracts and 16.6% (95% CI 16.1% to 17.0%) in the most deprived. Across tracts, disability correlates only +0.07 (95% CI 0.04 to 0.10) with the share of older residents.

Disability was 30.5% (95% CI 29.6% to 31.3%) in the tenth of tracts with the fewest older residents (5.9% of residents) and 32.3% (95% CI 31.8% to 32.7%) in the tenth with the most (35.8%), a difference of 1.8 (95% CI 0.9 to 2.7) points.

Table 1. Adult disability is 23.3% in the least deprived tenth of census tracts and 42.9% in the most deprived, and it differs by 1.8 points between tenths of tracts with few and many older residents

### Disabled workers are a larger share of beneficiaries in deprived ZIP codes

Old-Age, Survivors and Disability Insurance pays benefits to retired workers, to disabled workers, to survivors (widows, widowers and parents of deceased workers), and to spouses and children. The Social Security Administration counts its beneficiaries by ZIP code. In December 2024 the 31,815 ZIP codes in the states and the District of Columbia that match a ZIP code area with an ADI rank held 65.9 million beneficiaries.

In the least deprived fifth of these ZIP codes, 80.5% (95% CI 79.9% to 81.1%) of beneficiaries were retired workers and 7.2% (95% CI 6.9% to 7.5%) were disabled workers. In the most deprived fifth, 65.9% (95% CI 65.3% to 66.4%) were retired workers and 16.8% (95% CI 16.4% to 17.3%) were disabled workers (Table 2).

The share of widows, widowers and parents is also larger in the most deprived fifth, 6.6% against 4.8%. So is the share of spouses and children, 10.7% against 7.5%.

A disabled worker's benefit becomes a retired worker's benefit at full retirement age. The retired-worker share therefore includes people who first qualified through disability, and the disabled-worker share counts only those still below full retirement age.

Table 2. Disabled workers are 16.8% of Social Security beneficiaries in the most deprived fifth of ZIP codes and 7.2% in the least deprived, and retired workers are 65.9% against 80.5%

Across ZIP codes, the disabled-worker share correlates +0.68 (95% CI 0.64 to 0.71) with the ADI rank when ZIP codes are weighted by population, and +0.52 (95% CI 0.50 to 0.54) when each counts once. Small ZIP codes, whose shares rest on few beneficiaries and counts that SSA rounds, count for more in the second correlation.

Among the 1,838 counties with at least 5,000 beneficiaries in their ZIP codes, the disabled-worker share is highest in Pike County, Kentucky (27.2%). The 12 counties with the highest shares, down to 23.0%, are all in Kentucky, Virginia and West Virginia.

The ranking depends on the cutoff. With a minimum of 10,000 beneficiaries, the 12 highest counties include counties in Alabama, Arkansas and Oklahoma as well as Kentucky and Virginia.

Deprivation and income carry separate information about the share. In a regression across ZIP codes with the ADI rank, median household income and the share of residents aged 65 and over, each in standard-deviation units, weighted by population, the disabled-worker share has a coefficient of 0.46 (95% CI 0.40 to 0.52) on the ADI rank and −0.36 (95% CI −0.42 to −0.30) on income.

### The size of a retired worker's benefit follows ZIP code income

The Social Security benefit formula starts from a worker's earnings over a career, so the average benefit in a ZIP code should rise with the incomes of its residents. The average monthly retired-worker benefit was $1,554 (95% CI $1,523 to $1,586) in the tenth of ZIP codes with the lowest median household income and $2,370 (95% CI $2,354 to $2,386) in the tenth with the highest (Table 3).

Across 28,547 ZIP codes with at least 50 retired workers, the correlation between the average benefit and median household income is +0.84 (95% CI 0.83 to 0.85) weighted by population and +0.80 (95% CI 0.79 to 0.82) counting each ZIP code once.

In the regression above, with the average retired-worker benefit as the outcome, the coefficient on income is 0.89 (95% CI 0.80 to 0.98), the coefficient on the share of residents aged 65 and over is 0.16 (95% CI 0.14 to 0.18) and the coefficient on the ADI rank is 0.08 (95% CI −0.02 to 0.18), an interval that includes zero. The benefit amount moves with income and not with the ADI rank, while the disabled-worker share moves with both.

Table 3. The average monthly retired-worker benefit rises with ZIP code income, from $1,554 in the lowest tenth of ZIP codes to $2,370 in the highest, a correlation of +0.84

Social Security is one layer of retirement income. Taxable pensions and annuities reported on tax returns for 2022 include pensions from private and public employers. They were 41.7% (95% CI 41.0% to 42.4%) of the sum of those pensions and annualized Social Security benefits in the least deprived fifth of ZIP codes and 28.0% (95% CI 27.5% to 28.4%) in the most deprived. The two sources cover different years and different people, tax filers and beneficiaries, so the share is a rough index of how much retirement income comes from outside Social Security.

### Medicare payments show no detectable difference across county deprivation, but their mix differs

The Medicare Geographic Variation file reports, for each county, Part A and Part B payments per beneficiary with local prices removed, called price-standardized payments.[3] The payments are per beneficiary enrolled in Original Medicare, the fee-for-service program, with both parts. Medicare Advantage enrollees, who were 54.5% of beneficiaries with both parts in these counties, are not in them. The file counts beneficiaries of all ages, including people under 65 who qualify through disability, and standardization does not adjust for how sick beneficiaries are. Across 3,128 counties with an ADI rank, covering 27.6 million fee-for-service beneficiaries, payments averaged $12,585 (95% CI $12,125 to $12,987) per beneficiary in 2024.

Ranking counties on a county ADI, the mean of its tracts' ranks weighted by population, payments were $12,680 (95% CI $11,997 to $13,194) in the least deprived fifth of counties and $13,176 (95% CI $12,517 to $13,785) in the most deprived. The difference of $496 (95% CI −$290 to $1406) has an interval that includes zero, so the data show no detectable difference, though the interval is wide (Table 4). The correlation between a county's payments and its ADI is −0.04 (95% CI −0.19 to 0.21) weighted by fee-for-service beneficiaries and +0.30 (95% CI 0.20 to 0.39) counting each county once.

The fifths are different kinds of counties. 87.3% of the beneficiaries in the most deprived fifth live in counties with fewer than 5,000 fee-for-service beneficiaries, against 3.1% in the least deprived fifth, so the comparison of fifth means sets mostly small counties against mostly large ones.

Among counties below that size, payments rise from $10,893 (95% CI $10,526 to $11,270) in the least deprived fifth to $13,295 (95% CI $12,601 to $13,903) in the most deprived, and the intervals do not overlap. Among larger counties they go from $12,736 (95% CI $12,046 to $13,250) to $12,359 (95% CI $11,830 to $12,916), and the intervals overlap. Small counties make up 97% of the counties in the most deprived fifth and carry little weight in the weighted correlation, which is part of why the two correlations disagree.

The total hides changes in its parts. Between the least and the most deprived fifth, payments for procedures, tests, imaging and treatments differ by −$738 (95% CI −$971 to −$475), from $2,508 to $1,770. Payments for evaluation and management visits, the category CMS uses for clinician visits in offices, hospitals and other settings, differ by −$430 (95% CI −$543 to −$296), from $1,333 to $903, which is 32% (95% CI 24% to 39%) lower.

Payments for skilled nursing facilities are $437 (95% CI $214 to $653) higher, from $865 to $1,302. Payments for hospital outpatient departments and surgery centers are $410 (95% CI $206 to $601) higher and for all other services $480 (95% CI $243 to $736) higher. The differences for inpatient hospital stays ($134 (95% CI −$12 to $315)) and for home health and hospice ($224 (95% CI −$56 to $511)) have intervals that include zero.

Emergency department visits are higher in the more deprived fifths, 571 (95% CI 556 to 584) per thousand beneficiaries in the least deprived and 677 (95% CI 656 to 694) in the most deprived, and a county's visit rate correlates +0.37 (95% CI 0.26 to 0.47) with its ADI. The means rise at every step, though the intervals of neighboring fifths overlap. Chapter 7.1, Health behaviors and the supply of care (https://vishalsingh.org/zipcode/health-behaviors-and-the-supply-of-care) reports the supply of primary care physicians and the rate of preventable hospital stays by county deprivation.

Table 4. Price-standardized Medicare payments are $12,680 per fee-for-service beneficiary in the least deprived fifth of counties and $13,176 in the most deprived, a difference whose interval includes zero

Payments vary across counties far more than they vary across the fifths. Weighted by beneficiaries, the county at the 90th percentile of payments, $14,770, is at 1.42 (95% CI 1.34 to 1.49) times the county at the 10th percentile, $10,433. Fisher and colleagues compared Medicare beneficiaries across regions grouped by spending in a cohort study and found that those in higher-spending regions received more care without better health outcomes.[4]

Across the 3,057 counties that also have a life expectancy at birth in County Health Rankings, for deaths in 2019 to 2021, the correlation between county payments and life expectancy is −0.07 (95% CI −0.27 to 0.05) weighted by beneficiaries, with a slope of −0.12 (95% CI −0.51 to 0.08) years per thousand dollars. Counting each county once, the correlation is −0.35 (95% CI −0.43 to −0.25).

Holding the county ADI fixed, the slope is −0.17 (95% CI −0.33 to −0.01) years per thousand dollars weighted by beneficiaries and −0.28 (95% CI −0.40 to −0.16) counting each county once, so at a given level of deprivation life expectancy is slightly lower where payments are higher. The county ADI correlates −0.82 (95% CI −0.85 to −0.77) with life expectancy.

### Weighted by residents, the poorest income quartile's expected age at death varies more across counties

The Health Inequality Project linked federal tax records to Social Security death records to estimate expected age at death for people who reach age 40, by household income, sex and county.[5] The file used here gives the estimate for each quartile of the national distribution of household income, from the poorest to the richest, in 1,559 larger counties, for 2001 to 2014, adjusted to the national racial composition. The file does not state its size threshold. Its smallest county has 15,304 residents and its median county 69,380, in the County Health Rankings 2024 release (a Census Bureau estimate for 2022).

The chapter averages the estimates for women and men. Weighted by county residents, the poorest quartile's expected age at death averaged 79.5 (95% CI 79.1 to 79.9) across counties and the richest quartile's 86.3 (95% CI 86.1 to 86.4), a difference of 6.7 (95% CI 6.3 to 7.2) years.

The poorest quartile's values run from 77.5 in the county at the 5th percentile to 81.8 in the county at the 95th (Table 5). Population-weighted state averages of the county values are lowest in Oklahoma, Nevada, Indiana, Wyoming, Kentucky, Michigan and Arkansas and highest in New York, Montana, California, Vermont, Massachusetts, Maine and Idaho, and Nevada and Wyoming have few counties in the file. The Project estimated only larger counties, 1,559 of them on this map, so the map is gray for the rest.

Table 5. Expected age at death at 40 for the poorest income quartile runs from 77.5 in the county at the fifth percentile to 81.8 in the county at the ninety-fifth

The richest quartile's expected age at death is higher than the poorest quartile's in 1,558 of the 1,559 counties (Table 6), but the poorest quartile's values vary more from county to county. Weighted by residents, the range from the 5th to the 95th percentile of counties is 4.3 (95% CI 3.7 to 4.9) years for the poorest quartile and 3.0 (95% CI 2.7 to 3.6) for the richest, a difference of 1.3 (95% CI 0.3 to 1.9) years whose interval excludes zero.

Among the 283 counties with at least 250,000 residents, the ranges are 4.2 (95% CI 3.5 to 5.0) and 2.6 (95% CI 2.3 to 3.4), a difference of 1.6 (95% CI 0.7 to 2.5) years.

Table 6. Weighted by residents, county values of expected age at death at 40 span 4.3 years between the 5th and 95th percentiles for the poorest income quartile and 3.0 for the richest

The comparison depends on the weights. Counting each county once, the ranges are 4.0 (95% CI 3.7 to 4.4) years for the poorest quartile and 4.3 (95% CI 3.9 to 4.6) for the richest, and the difference of −0.3 (95% CI −0.7 to 0.2) includes zero.

Estimates for the richest quartile rest on fewer deaths in small counties and are less precise, with a mean standard error of 0.9 years against 0.5 for the poorest quartile. Removing the mean squared standard error from the variance of the county values, and counting each county once, leaves a standard deviation of 1.13 (95% CI 1.00 to 1.24) years for the poorest quartile and 0.91 (95% CI 0.77 to 1.02) for the richest, a difference of 0.22 (95% CI 0.08 to 0.38) years whose interval excludes zero.

Without the race adjustment, the ranges are 6.4 (95% CI 4.7 to 6.7) years for the poorest quartile and 3.0 (95% CI 2.7 to 3.4) for the richest, so the adjustment shortens the poorest quartile's range and leaves the richest quartile's nearly unchanged.

The figure below sets expected age at death beside 15 county measures, the county ADI and measures from County Health Rankings (Table 7). Weighted by residents, the poorest quartile's expected age at death correlates −0.67 (95% CI −0.74 to −0.51) with the county ADI, −0.64 (95% CI −0.71 to −0.51) with the share of adults with obesity, −0.58 (95% CI −0.67 to −0.45) with the smoking rate, +0.48 (95% CI 0.38 to 0.55) with median household income and +0.32 (95% CI 0.25 to 0.40) with the share of adults with some college.

Counting each county once, the correlations are −0.44 (95% CI −0.53 to −0.34) with the ADI, −0.45 (95% CI −0.54 to −0.34) with obesity and −0.34 (95% CI −0.44 to −0.24) with smoking. The share of people under 65 without health insurance correlates −0.15 (95% CI −0.28 to 0.06) weighted and −0.16 (95% CI −0.29 to 0.01) counting each county once, intervals that include zero. Primary care physicians per hundred thousand residents correlate +0.28 (95% CI 0.20 to 0.35).

For the richest quartile, the county ADI correlates −0.22 (95% CI −0.43 to −0.07), the uninsured share −0.27 (95% CI −0.37 to −0.18) and physical inactivity −0.50 (95% CI −0.60 to −0.41). The Project's own analysis of local areas reported a correlation of −0.69 between life expectancy in the lowest income quartile and smoking, and no significant correlation with access to medical care.[5]

Table 7. Across counties weighted by residents, the poorest quartile's expected age at death correlates −0.67 with the county ADI, −0.64 with obesity and −0.58 with smoking, and −0.15 with the uninsured share

> Causal evidence: What eligibility rules and cash benefits change for the people they reach (Regression discontinuity; instrumental variables (examiner assignment); difference-in-differences)
>
> The comparisons above are between places. Four studies ask what Medicare and Social Security change for the people who receive them, each with its own design.
>
> Card, Dobkin and Maestas compared patients just below and just above age 65, when most people become eligible for Medicare. The patients were admitted to California hospitals through the emergency department with conditions whose admission rates did not differ between weekdays and weekends. In this regression discontinuity design, patients just past the threshold received more treatment and had lower mortality in the days after admission.[6]
>
> Maestas, Mullen and Strand used the assignment of disability insurance applications to examiners who differ in how often they allow claims. An examiner's allowance rate served as an instrument for whether an applicant received benefits, an instrumental variables design, and applicants on the margin of allowance would have worked more had they not received benefits.[7] Deshpande, Gross and Su used two age thresholds in the disability determination rules at which the chance of allowance rises, in a design similar to a regression discontinuity, and found that allowance lowered the chances of bankruptcy and foreclosure for the applicants the thresholds affected.[8]
>
> Snyder and Evans used the Social Security benefit notch, a change in the benefit formula that gave people born just before January 1917 larger benefits than people born just after it. They compared mortality across those birth cohorts, with other cohorts as controls, in a difference-in-differences design that the published paper pairs with a regression discontinuity. The cohort with the larger benefits had higher mortality, and the authors link the lower mortality of the smaller-benefit cohort to more work after retirement.[9]
>
> These designs estimate what a program changes for the people it reaches at a threshold or on the margin of a decision. None of them estimates whether the programs would narrow the differences between places in this chapter.

### Adult disability rises across tracts under all four rankings

Table 8 repeats the disability gradient under four rankings, over the 75,596 tracts that have all four. Ranked by the ADI, the share of adults with any disability is 23.1% (95% CI 22.1% to 23.9%) in the least deprived tenth and 42.6% (95% CI 42.2% to 43.0%) in the most deprived, and under the standardized index it is 21.4% (95% CI 21.0% to 21.8%) and 43.4% (95% CI 42.8% to 43.9%). Ranked by median household income it runs from 21.4% (95% CI 21.0% to 21.8%) to 42.9% (95% CI 42.6% to 43.3%), and by bachelor's degree share from 20.7% (95% CI 20.3% to 21.1%) to 42.0% (95% CI 41.6% to 42.5%), and under each ranking the tenth means rise in order (rank correlations of +1.00, +1.00 and +1.00). None of the three changes the gradient under the book's rule (Chapter 9.1, How far one ranking goes (https://vishalsingh.org/zipcode/how-far-one-ranking-goes)), and the PLACES model draws on ACS inputs that all four rankings share.

Table 8. Adult disability is steadily worse toward the more deprived end under the ADI and under each of the three other rankings

### What these data cannot show

Every pattern here describes tracts, ZIP codes or counties, and a pattern across places need not hold for the people inside them (ecological inference). A ZIP code with a high disabled-worker share says nothing about whether any resident should receive a disability benefit, and a county's expected age at death for the poorest quartile is an estimate for a group of people who share an income rank, not a forecast for one of them.

The measures differ in vintage and in how they are made. PLACES values are modeled from 2022 survey responses, and their link with the ADI is partly built in. The Social Security counts are for December 2024 and are rounded by SSA, and the ZIP code is matched to the ZIP code area of the same number, which leaves out ZIP codes such as post office boxes. The Medicare payments are for 2024, cover only fee-for-service beneficiaries, and so leave out the Medicare Advantage enrollees, who are 54.5% of beneficiaries with both parts in these counties. They are not adjusted for the health of the beneficiaries, so equal payments do not mean equal care or equal need. The life-expectancy estimates describe deaths from 2001 to 2014 and end before the later years of this book, and household income is measured on a national scale that does not adjust for the cost of living where a household lives.

The correlates are measured years after the deaths, and smoking and obesity are County Health Rankings' modeled estimates for all adults, not for the poorest quartile. The Project's authors found that the share of immigrants was among the strongest correlates of low-income longevity across local areas, a pattern that tract and county measures of this kind cannot separate from place itself (see Chapter 9.5, Immigrants and the gradient (https://vishalsingh.org/zipcode/immigrants-and-the-gradient; planned)). Weights matter for several results here. The Medicare correlations with deprivation and with life expectancy and the spread of the richest quartile's expected age at death all change when each county counts once, and both versions are reported.

None of the gradients tells what would move a number. The causal evidence above concerns eligibility rules, examiner decisions and a benefit notch, and the places in this chapter differ in many ways that those studies do not address.

## Tables

### Table 1. Adult disability is 23.3% in the least deprived tenth of census tracts and 42.9% in the most deprived, and it differs by 1.8 points between tenths of tracts with few and many older residents

Percent of adults with any disability and percent of older adults who have lost all their teeth, by tenth of the Area Deprivation Index, and disability against the share of older residents, census tracts, 2024 PLACES release

#### Table 1.1. By deprivation

Disability and tooth loss both rise across tenths of tracts by deprivation, tooth loss from 6.9% to 25.1% and disability from 23.3% to 42.9%

Percent of adults with any disability and percent of older adults who have lost all their teeth, by tenth of the Area Deprivation Index national rank, census tracts

| conf_high | conf_low | decile | estimate | n | series |
| ---: | ---: | ---: | ---: | ---: | --- |
| 24.07 | 22.30 | 1 | 23.28 | 8,274 | Any disability among adults |
| 27.45 | 24.50 | 2 | 25.97 | 8,118 | Any disability among adults |
| 27.99 | 26.08 | 3 | 27 | 8,509 | Any disability among adults |
| 28.73 | 27.60 | 4 | 28.17 | 8,261 | Any disability among adults |
| 30.32 | 29.39 | 5 | 29.84 | 8,189 | Any disability among adults |
| 32.06 | 31.21 | 6 | 31.62 | 8,318 | Any disability among adults |
| 33.88 | 33.16 | 7 | 33.53 | 8,305 | Any disability among adults |
| 35.81 | 35.15 | 8 | 35.48 | 8,238 | Any disability among adults |
| 38.39 | 37.74 | 9 | 38.06 | 8,354 | Any disability among adults |
| 43.21 | 42.50 | 10 | 42.86 | 8,213 | Any disability among adults |
| 7.91 | 6.24 | 1 | 6.94 | 8,258 | All teeth lost among older adults |
| 9.17 | 7.72 | 2 | 8.48 | 8,111 | All teeth lost among older adults |
| 9.65 | 8.65 | 3 | 9.12 | 8,498 | All teeth lost among older adults |
| 10.29 | 9.66 | 4 | 9.97 | 8,251 | All teeth lost among older adults |
| 11.53 | 10.90 | 5 | 11.20 | 8,179 | All teeth lost among older adults |
| 12.90 | 12.32 | 6 | 12.61 | 8,299 | All teeth lost among older adults |
| 14.49 | 13.95 | 7 | 14.22 | 8,298 | All teeth lost among older adults |
| 16.51 | 15.96 | 8 | 16.23 | 8,214 | All teeth lost among older adults |
| 19.15 | 18.58 | 9 | 18.87 | 8,341 | All teeth lost among older adults |
| 25.56 | 24.65 | 10 | 25.10 | 8,213 | All teeth lost among older adults |

#### Table 1.2. By age structure

Tracts with many older residents have little more disability than tracts with few, 32.3% in the tenth with the most against 30.5% in the tenth with the fewest

Percent of adults with any disability against the share of residents who are older adults, one dot per sampled census tract, with means for tenths of tracts by age share

No inline rows: the values of Table 1.2 are in https://vishalsingh.org/zipcode/data/disability-retirement-and-medicare/disability_by_age.json.

Note: PLACES values are modeled estimates built from survey responses with census and ACS inputs that overlap the Area Deprivation Index, so part of each gradient is built in. Values are crude prevalence among adults, not adjusted for age. The tract file has no intervals, and the bands and whiskers come from resampling counties.

Note: Any disability is difficulty with hearing, vision, thinking, walking, self-care or independent living reported by adults aged 18 and over, and older adults are those aged 65 and over. Tenths hold equal numbers of tracts, and a tract with no tooth-loss value is left out of that line.

Note: The dots in the second view are a random sample of 2,000 tracts inside the plotted range, drawn with a fixed seed (stated in the methods note); 398 tracts fall outside that range, and every estimate uses all tracts.

Note: Across all the tracts in the comparison, 30.9% of adults report any disability.

Source: CDC PLACES 2024 release (BRFSS 2022), ACS 2019-2023 and Area Deprivation Index 2023; modeled estimates, weighted by adult population

Data: https://vishalsingh.org/zipcode/report/disability-retirement-and-medicare/figures/disability_gradient.json

### Table 2. Disabled workers are 16.8% of Social Security beneficiaries in the most deprived fifth of ZIP codes and 7.2% in the least deprived, and retired workers are 65.9% against 80.5%

Share of OASDI beneficiaries by type, percent, ZIP codes in fifths of the Area Deprivation Index national rank, December 2024

| Fifth of ZIP codes by ADI | Type of beneficiary | Share of beneficiaries (%) | Lower limit of the 95% interval | Upper limit of the 95% interval | ZIP codes | Beneficiaries |
| --- | --- | ---: | ---: | ---: | ---: | ---: |
| 1 (least deprived) | Retired workers | 80.5 | 79.9 | 81.0 | 6,363 | 23,011,380 |
| 1 (least deprived) | Disabled workers | 7.2 | 6.9 | 7.5 | 6,363 | 23,011,380 |
| 1 (least deprived) | Widowed | 4.8 | 4.7 | 4.9 | 6,363 | 23,011,380 |
| 1 (least deprived) | Spouses and children | 7.5 | 7.2 | 7.8 | 6,363 | 23,011,380 |
| 2 | Retired workers | 77.2 | 76.9 | 77.6 | 6,363 | 17,037,990 |
| 2 | Disabled workers | 10.1 | 9.8 | 10.3 | 6,363 | 17,037,990 |
| 2 | Widowed | 5.1 | 5.0 | 5.1 | 6,363 | 17,037,990 |
| 2 | Spouses and children | 7.6 | 7.5 | 7.8 | 6,363 | 17,037,990 |
| 3 | Retired workers | 73.9 | 73.6 | 74.3 | 6,363 | 11,839,725 |
| 3 | Disabled workers | 12.3 | 12.1 | 12.5 | 6,363 | 11,839,725 |
| 3 | Widowed | 5.5 | 5.5 | 5.6 | 6,363 | 11,839,725 |
| 3 | Spouses and children | 8.3 | 8.1 | 8.4 | 6,363 | 11,839,725 |
| 4 | Retired workers | 70.4 | 70.0 | 70.7 | 6,363 | 8,506,535 |
| 4 | Disabled workers | 14.4 | 14.2 | 14.7 | 6,363 | 8,506,535 |
| 4 | Widowed | 6.1 | 6.0 | 6.2 | 6,363 | 8,506,535 |
| 4 | Spouses and children | 9.1 | 8.9 | 9.2 | 6,363 | 8,506,535 |
| 5 (most deprived) | Retired workers | 65.8 | 65.3 | 66.4 | 6,363 | 5,518,365 |
| 5 (most deprived) | Disabled workers | 16.8 | 16.4 | 17.3 | 6,363 | 5,518,365 |
| 5 (most deprived) | Widowed | 6.6 | 6.5 | 6.8 | 6,363 | 5,518,365 |
| 5 (most deprived) | Spouses and children | 10.7 | 10.4 | 10.9 | 6,363 | 5,518,365 |

Note: Types are the groups SSA reports: retired workers, disabled workers, widows, widowers and parents (survivors), and spouses and children. Disabled-worker benefits become retired-worker benefits at full retirement age, so former disabled workers are counted as retired workers after it.

Note: The figure covers 31,815 ZIP codes and 65.9 million beneficiaries that match a ZCTA with an ADI rank. A ZIP code is matched to the ZCTA of the same number (a ZIP code is not a ZCTA), and ZIP codes with no ZCTA, such as post office boxes, are left out. Connecticut ZIP codes are included. SSA rounds its counts and does not list ZIP codes below a disclosure threshold.

Source: SSA OASDI beneficiaries by ZIP code, December 2024, and Area Deprivation Index 2023; full counts, pooled over ZIP codes, intervals from a bootstrap over counties

Data: https://vishalsingh.org/zipcode/report/disability-retirement-and-medicare/figures/oasdi_mix.json

### Table 3. The average monthly retired-worker benefit rises with ZIP code income, from $1,554 in the lowest tenth of ZIP codes to $2,370 in the highest, a correlation of +0.84

Average monthly OASDI retired-worker benefit in dollars against median household income in dollars (log scale), one dot per sampled ZIP code area, with means for tenths of ZIP codes by income

No inline rows: the values of Table 3 are in https://vishalsingh.org/zipcode/data/disability-retirement-and-medicare/benefit_vs_income.json.

Note: The average is the monthly benefit dollars paid to retired workers in a ZIP code divided by its retired workers, for December 2024. SSA reports benefit totals in thousands of dollars, converted here, and rounds them. ZIP code income is the ACS 2019-2023 median household income of the matching ZCTA, which the Census Bureau tops out at a ceiling, and ZIP codes without an income are left out.

Note: The dots are drawn at random from the ZIP codes inside the plotted range with a fixed seed (stated in the methods note); every estimate uses all ZIP codes. The correlation weights ZIP codes by population; counting each ZIP code once gives +0.80.

Source: SSA OASDI beneficiaries by ZIP code, December 2024, and ACS 2019-2023 ZCTA median household income; full counts of benefits, weighted by retired workers, intervals from a bootstrap over counties

Data: https://vishalsingh.org/zipcode/report/disability-retirement-and-medicare/figures/benefit_vs_income.json

### Table 4. Price-standardized Medicare payments are $12,680 per fee-for-service beneficiary in the least deprived fifth of counties and $13,176 in the most deprived, a difference whose interval includes zero

Medicare Part A and Part B payments per Original Medicare beneficiary with local prices removed, by fifth of the county Area Deprivation Index, service lines, emergency department visits and county life expectancy, 2024

#### Table 4.1. Service lines

Payments for procedures, tests, imaging and treatments fall from $2,508 to $1,770 across fifths of counties while payments for skilled nursing facilities rise from $865 to $1,302

Price-standardized payments per fee-for-service beneficiary by service line, dollars per year, fifths of counties by the county Area Deprivation Index

| Fifth of counties by ADI | Service line | Dollars per beneficiary | Lower limit of the 95% interval | Upper limit of the 95% interval | Counties |
| --- | --- | ---: | ---: | ---: | ---: |
| 1 (least deprived) | Inpatient hospital stays | 2,871 | 2,712 | 3,001 | 605 |
| 2 | Inpatient hospital stays | 2,869 | 2,765 | 2,963 | 585 |
| 3 | Inpatient hospital stays | 2,913 | 2,812 | 2,993 | 571 |
| 4 | Inpatient hospital stays | 2,955 | 2,858 | 3,036 | 565 |
| 5 (most deprived) | Inpatient hospital stays | 3,005 | 2,919 | 3,073 | 541 |
| 1 (least deprived) | Hospital outpatient and surgery centers | 2,516 | 2,401 | 2,684 | 605 |
| 2 | Hospital outpatient and surgery centers | 2,780 | 2,618 | 2,953 | 585 |
| 3 | Hospital outpatient and surgery centers | 2,864 | 2,708 | 3,035 | 571 |
| 4 | Hospital outpatient and surgery centers | 3,002 | 2,793 | 3,201 | 565 |
| 5 (most deprived) | Hospital outpatient and surgery centers | 2,926 | 2,768 | 3,098 | 541 |
| 1 (least deprived) | Skilled nursing facilities | 865 | 712 | 986 | 605 |
| 2 | Skilled nursing facilities | 771 | 721 | 829 | 585 |
| 3 | Skilled nursing facilities | 879 | 831 | 928 | 571 |
| 4 | Skilled nursing facilities | 1,004 | 925 | 1,081 | 565 |
| 5 (most deprived) | Skilled nursing facilities | 1,302 | 1,122 | 1,462 | 541 |
| 1 (least deprived) | Home health and hospice | 1,010 | 839 | 1,148 | 605 |
| 2 | Home health and hospice | 962 | 853 | 1,066 | 585 |
| 3 | Home health and hospice | 1,018 | 864 | 1,187 | 571 |
| 4 | Home health and hospice | 1,060 | 896 | 1,228 | 565 |
| 5 (most deprived) | Home health and hospice | 1,234 | 975 | 1,460 | 541 |
| 1 (least deprived) | Evaluation and management visits | 1,333 | 1,218 | 1,426 | 605 |
| 2 | Evaluation and management visits | 1,146 | 1,062 | 1,219 | 585 |
| 3 | Evaluation and management visits | 1,054 | 992 | 1,102 | 571 |
| 4 | Evaluation and management visits | 967 | 902 | 1,028 | 565 |
| 5 (most deprived) | Evaluation and management visits | 903 | 831 | 967 | 541 |
| 1 (least deprived) | Procedures, tests, imaging and treatments | 2,508 | 2,275 | 2,723 | 605 |
| 2 | Procedures, tests, imaging and treatments | 2,204 | 2,011 | 2,399 | 585 |
| 3 | Procedures, tests, imaging and treatments | 1,938 | 1,808 | 2,073 | 571 |
| 4 | Procedures, tests, imaging and treatments | 1,786 | 1,655 | 1,936 | 565 |
| 5 (most deprived) | Procedures, tests, imaging and treatments | 1,770 | 1,635 | 1,884 | 541 |
| 1 (least deprived) | All other services | 1,578 | 1,375 | 1,745 | 605 |
| 2 | All other services | 1,639 | 1,456 | 1,830 | 585 |
| 3 | All other services | 1,724 | 1,544 | 1,922 | 571 |
| 4 | All other services | 1,776 | 1,622 | 1,931 | 565 |
| 5 (most deprived) | All other services | 2,058 | 1,863 | 2,226 | 541 |

#### Table 4.2. Life expectancy

Weighted by beneficiaries, county life expectancy has no clear relation to payments per beneficiary, a correlation of −0.07, but counting each county once the correlation is −0.35

County life expectancy at birth in years against price-standardized Medicare payments per fee-for-service beneficiary in dollars, one dot per county, dot area is beneficiaries, with means for tenths of counties by payments

No inline rows: the values of Table 4.2 are in https://vishalsingh.org/zipcode/data/disability-retirement-and-medicare/spend_vs_life_expectancy.json.

#### Table 4.3. Emergency visits

Emergency department visits are higher in the more deprived fifths of counties, from 571 in the least deprived to 677 in the most deprived per thousand fee-for-service beneficiaries

Emergency department visits per thousand fee-for-service beneficiaries per year, fifths of counties by the county Area Deprivation Index

| conf_high | conf_low | fifth | n | visits |
| ---: | ---: | --- | ---: | ---: |
| 584 | 555.5 | 1 (least deprived) | 625 | 570.5 |
| 617.4 | 582.1 | 2 | 625 | 598.9 |
| 635.5 | 603 | 3 | 626 | 620.1 |
| 655.4 | 627.3 | 4 | 628 | 642.1 |
| 693.6 | 656.4 | 5 (most deprived) | 624 | 676.7 |

Note: Payments are CMS price-standardized Part A and Part B payments per capita for beneficiaries enrolled in Original Medicare with both parts, so Medicare Advantage enrollees, who are 54.5% of beneficiaries with both parts in these counties, are not in them. Standardization removes local price differences and does not adjust for how sick beneficiaries are. The file counts beneficiaries of all ages, including those under 65 who qualify through disability.

Note: A fifth holds equal numbers of counties, ranked on the county Area Deprivation Index (the mean of its tracts' ranks, weighted by population). Intervals resample whole states. Connecticut is by planning region. The service lines view leaves out counties for which CMS blanks a line, and its last segment is the total minus the lines shown. Evaluation and management visits are CMS's category for visits to physicians and other clinicians in many settings, among them offices, hospitals, nursing facilities, homes and emergency departments.

Note: Life expectancy is the County Health Rankings 2024 estimate computed from NCHS death counts. Connecticut is left out of the second view because that file uses its legacy counties.

Note: Across all the counties in the comparison, there are 592 emergency department visits per thousand fee-for-service beneficiaries per year.

Source: CMS Medicare Geographic Variation, county, 2024, and Area Deprivation Index 2023; price-standardized payments per fee-for-service beneficiary, weighted by fee-for-service beneficiaries, full counts (life expectancy from County Health Rankings 2024, deaths 2019-2021, full counts)

Data: https://vishalsingh.org/zipcode/report/disability-retirement-and-medicare/figures/medicare_spending.json

### Table 5. Expected age at death at 40 for the poorest income quartile runs from 77.5 in the county at the fifth percentile to 81.8 in the county at the ninety-fifth

Expected age at death in years for people who reach age 40 in the lowest quartile of national household income, 2001 to 2014, counties in the Health Inequality Project file

No inline rows: the values of Table 5 are in https://vishalsingh.org/zipcode/data/disability-retirement-and-medicare/poor_le_map.json.

Note: The Project estimated only the larger counties, so 1,559 counties are drawn with a value and the rest are gray with the reason in the tooltip. A gray county was not estimated; it is not a county with a low value.

Note: Each value comes from a bootstrap estimate with a standard error given in the tooltip, and small counties are noisy. Connecticut is shown by its eight legacy counties, as in the Project's file. Puerto Rico is not drawn.

Source: Health Inequality Project county expected age at death, 2001-2014 (Online Data Table 11); modeled estimates, race-adjusted, average of women and men; county residents from the County Health Rankings 2024 release

Data: https://vishalsingh.org/zipcode/report/disability-retirement-and-medicare/figures/poor_le_map.json

### Table 6. Weighted by residents, county values of expected age at death at 40 span 4.3 years between the 5th and 95th percentiles for the poorest income quartile and 3.0 for the richest

Expected age at death in years for people who reach age 40 in the poorest quartile against the richest quartile of national household income, one dot per county, dot area is county residents, 2001 to 2014

No inline rows: the values of Table 6 are in https://vishalsingh.org/zipcode/data/disability-retirement-and-medicare/poor_vs_rich_le_at_40.json.

Note: The richest quartile's estimates rest on fewer deaths in small counties and carry larger standard errors: the mean standard error of a county's value is 0.9 years for the richest quartile and 0.5 years for the poorest. Counting each county once, the range is 4.0 years for the poorest quartile and 4.3 for the richest, and among counties of at least 250,000 residents it is 4.2 against 2.6.

Note: Values are adjusted to the national racial composition and averaged over women and men. Without the race adjustment, the range across counties is 6.4 years for the poorest quartile and 3.0 for the richest. The ranges run from the 5th to the 95th percentile of counties, weighted by county residents, and their intervals resample whole states.

Note: Weights and dot areas are county residents in the County Health Rankings 2024 release, a Census Bureau estimate for 2022 and the nearest population available, while the estimates are for 2001 to 2014. The file covers 1,559 larger counties.

Source: Health Inequality Project county expected age at death, 2001-2014 (Online Data Table 11); modeled estimates, race-adjusted, average of women and men; weighted by county residents (County Health Rankings 2024 release)

Data: https://vishalsingh.org/zipcode/report/disability-retirement-and-medicare/figures/poor_vs_rich_le_at_40.json

### Table 7. Across counties weighted by residents, the poorest quartile's expected age at death correlates −0.67 with the county ADI, −0.64 with obesity and −0.58 with smoking, and −0.15 with the uninsured share

Correlation across counties of expected age at death at age 40 in the poorest and the richest income quartile with 15 county measures, 2001 to 2014 deaths and 2018 to 2023 measures, with 95% intervals

| conf_high | conf_low | estimate | measure | n | quartile | weighting |
| ---: | ---: | ---: | --- | ---: | --- | --- |
| 0.547 | 0.376 | 0.478 | Median household income | 1,559 | Poorest quartile | Population-weighted |
| 0.366 | 0.197 | 0.290 | Median household income | 1,559 | Poorest quartile | Each county once |
| 0.382 | 0.171 | 0.266 | Median household income | 1,559 | Richest quartile | Population-weighted |
| 0.355 | 0.255 | 0.308 | Median household income | 1,559 | Richest quartile | Each county once |
| 0.403 | 0.247 | 0.324 | Adults with some college | 1,559 | Poorest quartile | Population-weighted |
| 0.305 | 0.156 | 0.235 | Adults with some college | 1,559 | Poorest quartile | Each county once |
| 0.527 | 0.380 | 0.456 | Adults with some college | 1,559 | Richest quartile | Population-weighted |
| 0.453 | 0.356 | 0.407 | Adults with some college | 1,559 | Richest quartile | Each county once |
| 0.346 | 0.205 | 0.277 | Primary care physicians per 100,000 residents | 1,559 | Poorest quartile | Population-weighted |
| 0.330 | 0.199 | 0.264 | Primary care physicians per 100,000 residents | 1,559 | Poorest quartile | Each county once |
| 0.491 | 0.339 | 0.418 | Primary care physicians per 100,000 residents | 1,559 | Richest quartile | Population-weighted |
| 0.407 | 0.317 | 0.363 | Primary care physicians per 100,000 residents | 1,559 | Richest quartile | Each county once |
| 0.468 | 0.025 | 0.273 | Residents who are Hispanic | 1,559 | Poorest quartile | Population-weighted |
| 0.250 | −0.138 | 0.032 | Residents who are Hispanic | 1,559 | Poorest quartile | Each county once |
| −0.117 | −0.404 | −0.300 | Residents who are Hispanic | 1,559 | Richest quartile | Population-weighted |
| 0.040 | −0.270 | −0.125 | Residents who are Hispanic | 1,559 | Richest quartile | Each county once |
| 0.321 | −0.010 | 0.195 | Ratio of 80th to 20th percentile household income | 1,545 | Poorest quartile | Population-weighted |
| 0.098 | −0.094 | 0 | Ratio of 80th to 20th percentile household income | 1,545 | Poorest quartile | Each county once |
| −0.008 | −0.216 | −0.125 | Ratio of 80th to 20th percentile household income | 1,545 | Richest quartile | Population-weighted |
| −0.051 | −0.222 | −0.145 | Ratio of 80th to 20th percentile household income | 1,545 | Richest quartile | Each county once |
| 0.169 | −0.198 | 0.025 | Unemployment rate | 1,559 | Poorest quartile | Population-weighted |
| 0.086 | −0.147 | −0.023 | Unemployment rate | 1,559 | Poorest quartile | Each county once |
| −0.232 | −0.443 | −0.343 | Unemployment rate | 1,559 | Richest quartile | Population-weighted |
| −0.135 | −0.366 | −0.249 | Unemployment rate | 1,559 | Richest quartile | Each county once |
| 0.026 | −0.271 | −0.126 | Residents who are non-Hispanic Black | 1,559 | Poorest quartile | Population-weighted |
| −0.005 | −0.179 | −0.084 | Residents who are non-Hispanic Black | 1,559 | Poorest quartile | Each county once |
| −0.038 | −0.248 | −0.130 | Residents who are non-Hispanic Black | 1,559 | Richest quartile | Population-weighted |
| −0.075 | −0.255 | −0.162 | Residents who are non-Hispanic Black | 1,559 | Richest quartile | Each county once |
| 0.057 | −0.280 | −0.151 | People under 65 without health insurance | 1,559 | Poorest quartile | Population-weighted |
| 0.011 | −0.290 | −0.157 | People under 65 without health insurance | 1,559 | Poorest quartile | Each county once |
| −0.179 | −0.369 | −0.272 | People under 65 without health insurance | 1,559 | Richest quartile | Population-weighted |
| −0.165 | −0.363 | −0.271 | People under 65 without health insurance | 1,559 | Richest quartile | Each county once |
| −0.050 | −0.316 | −0.218 | Residents living in rural areas | 1,559 | Poorest quartile | Population-weighted |
| 0.133 | −0.076 | 0.023 | Residents living in rural areas | 1,559 | Poorest quartile | Each county once |
| 0.060 | −0.158 | −0.046 | Residents living in rural areas | 1,559 | Richest quartile | Population-weighted |
| −0.068 | −0.204 | −0.136 | Residents living in rural areas | 1,559 | Richest quartile | Each county once |
| −0.122 | −0.358 | −0.241 | Children in poverty | 1,559 | Poorest quartile | Population-weighted |
| −0.138 | −0.287 | −0.211 | Children in poverty | 1,559 | Poorest quartile | Each county once |
| −0.309 | −0.454 | −0.383 | Children in poverty | 1,559 | Richest quartile | Population-weighted |
| −0.335 | −0.450 | −0.395 | Children in poverty | 1,559 | Richest quartile | Each county once |
| −0.155 | −0.532 | −0.355 | Adults with no leisure-time physical activity | 1,559 | Poorest quartile | Population-weighted |
| −0.311 | −0.478 | −0.401 | Adults with no leisure-time physical activity | 1,559 | Poorest quartile | Each county once |
| −0.412 | −0.603 | −0.504 | Adults with no leisure-time physical activity | 1,559 | Richest quartile | Population-weighted |
| −0.428 | −0.533 | −0.485 | Adults with no leisure-time physical activity | 1,559 | Richest quartile | Each county once |
| −0.223 | −0.511 | −0.379 | Preventable hospital stays per 100,000 Medicare enrollees | 1,557 | Poorest quartile | Population-weighted |
| −0.281 | −0.439 | −0.360 | Preventable hospital stays per 100,000 Medicare enrollees | 1,557 | Poorest quartile | Each county once |
| −0.279 | −0.485 | −0.375 | Preventable hospital stays per 100,000 Medicare enrollees | 1,557 | Richest quartile | Population-weighted |
| −0.274 | −0.387 | −0.333 | Preventable hospital stays per 100,000 Medicare enrollees | 1,557 | Richest quartile | Each county once |
| −0.451 | −0.670 | −0.576 | Adults who smoke | 1,559 | Poorest quartile | Population-weighted |
| −0.243 | −0.436 | −0.343 | Adults who smoke | 1,559 | Poorest quartile | Each county once |
| −0.137 | −0.453 | −0.273 | Adults who smoke | 1,559 | Richest quartile | Population-weighted |
| −0.298 | −0.442 | −0.372 | Adults who smoke | 1,559 | Richest quartile | Each county once |
| −0.510 | −0.707 | −0.642 | Adults with obesity | 1,559 | Poorest quartile | Population-weighted |
| −0.340 | −0.536 | −0.446 | Adults with obesity | 1,559 | Poorest quartile | Each county once |
| −0.204 | −0.476 | −0.320 | Adults with obesity | 1,559 | Richest quartile | Population-weighted |
| −0.320 | −0.449 | −0.388 | Adults with obesity | 1,559 | Richest quartile | Each county once |
| −0.510 | −0.744 | −0.670 | Area Deprivation Index of the county (national rank) | 1,559 | Poorest quartile | Population-weighted |
| −0.338 | −0.532 | −0.442 | Area Deprivation Index of the county (national rank) | 1,559 | Poorest quartile | Each county once |
| −0.075 | −0.427 | −0.218 | Area Deprivation Index of the county (national rank) | 1,559 | Richest quartile | Population-weighted |
| −0.277 | −0.414 | −0.350 | Area Deprivation Index of the county (national rank) | 1,559 | Richest quartile | Each county once |

Views: Weighted by county residents (weighting = Population-weighted); Each county once (weighting = Each county once).

Note: The measures are for 2018 to 2023, after the 2001 to 2014 window of the life-expectancy estimates, and smoking, obesity and physical inactivity are County Health Rankings' modeled estimates for all adults, not for the poorest quartile. The county ADI is the mean of its tracts' ranks (2023), and part of the other measures overlaps its inputs. A correlation across counties does not say that smoking or obesity shortens the lives of the poor.

Note: Expected age at death is adjusted to the national racial composition and averaged over women and men. The two views use the same 1,559 counties wherever a measure exists.

Source: Health Inequality Project county expected age at death, 2001-2014 (Online Data Table 11); modeled estimates, race-adjusted, average of women and men; county measures from County Health Rankings 2024; population-weighted or unweighted as labeled

Data: https://vishalsingh.org/zipcode/report/disability-retirement-and-medicare/figures/poor_le_correlates.json

### Table 8. Adult disability is steadily worse toward the more deprived end under the ADI and under each of the three other rankings

Adults with any disability (percent of adults) by tenth of tracts ranked four ways, over the 75,596 tracts that have all four rankings

| bin | conf_high | conf_low | estimate | n_units | ranking |
| ---: | ---: | ---: | ---: | ---: | --- |
| 1 | 23.86 | 22.11 | 23.11 | 7,551 | Area Deprivation Index |
| 2 | 27.45 | 24.67 | 26.07 | 7,563 | Area Deprivation Index |
| 3 | 27.80 | 26.08 | 26.89 | 7,489 | Area Deprivation Index |
| 4 | 28.81 | 27.74 | 28.27 | 7,757 | Area Deprivation Index |
| 5 | 30.38 | 29.43 | 29.89 | 7,485 | Area Deprivation Index |
| 6 | 32.07 | 31.22 | 31.62 | 7,513 | Area Deprivation Index |
| 7 | 33.81 | 33.13 | 33.47 | 7,658 | Area Deprivation Index |
| 8 | 35.76 | 35.11 | 35.45 | 7,419 | Area Deprivation Index |
| 9 | 38.27 | 37.59 | 37.93 | 7,618 | Area Deprivation Index |
| 10 | 42.96 | 42.22 | 42.60 | 7,543 | Area Deprivation Index |
| 1 | 21.79 | 21.00 | 21.39 | 7,559 | Standardized index |
| 2 | 24.62 | 23.97 | 24.29 | 7,560 | Standardized index |
| 3 | 26.93 | 26.29 | 26.61 | 7,559 | Standardized index |
| 4 | 28.80 | 28.25 | 28.53 | 7,560 | Standardized index |
| 5 | 30.60 | 30.08 | 30.33 | 7,560 | Standardized index |
| 6 | 32.31 | 31.76 | 32.03 | 7,559 | Standardized index |
| 7 | 34.19 | 33.53 | 33.85 | 7,560 | Standardized index |
| 8 | 36.26 | 35.58 | 35.92 | 7,559 | Standardized index |
| 9 | 38.87 | 38.05 | 38.48 | 7,560 | Standardized index |
| 10 | 43.92 | 42.84 | 43.38 | 7,560 | Standardized index |
| 1 | 21.83 | 21.04 | 21.41 | 7,560 | Median household income |
| 2 | 25.04 | 24.40 | 24.72 | 7,558 | Median household income |
| 3 | 27.25 | 26.66 | 26.95 | 7,560 | Median household income |
| 4 | 29.02 | 28.45 | 28.74 | 7,560 | Median household income |
| 5 | 30.72 | 30.15 | 30.44 | 7,560 | Median household income |
| 6 | 32.41 | 31.87 | 32.14 | 7,560 | Median household income |
| 7 | 34.19 | 33.68 | 33.94 | 7,555 | Median household income |
| 8 | 36.11 | 35.56 | 35.84 | 7,564 | Median household income |
| 9 | 38.72 | 38.13 | 38.44 | 7,560 | Median household income |
| 10 | 43.26 | 42.57 | 42.92 | 7,559 | Median household income |
| 1 | 21.12 | 20.26 | 20.67 | 7,558 | Bachelor's degree share |
| 2 | 24.64 | 24.06 | 24.35 | 7,557 | Bachelor's degree share |
| 3 | 27.00 | 26.49 | 26.75 | 7,566 | Bachelor's degree share |
| 4 | 28.94 | 28.48 | 28.70 | 7,560 | Bachelor's degree share |
| 5 | 30.85 | 30.36 | 30.60 | 7,553 | Bachelor's degree share |
| 6 | 32.51 | 32.04 | 32.27 | 7,559 | Bachelor's degree share |
| 7 | 34.31 | 33.85 | 34.09 | 7,562 | Bachelor's degree share |
| 8 | 36.23 | 35.73 | 35.98 | 7,569 | Bachelor's degree share |
| 9 | 38.57 | 37.96 | 38.28 | 7,548 | Bachelor's degree share |
| 10 | 42.46 | 41.62 | 42.05 | 7,564 | Bachelor's degree share |

Note: Least and most deprived tenth: ADI 23.1% and 42.6%, rank correlation +1.00; standardized index 21.4% and 43.4%, +1.00; median household income 21.4% and 42.9%, +1.00; bachelor's degree share 20.7% and 42.0%, +1.00.

Note: The standardized index applies the ADI's coefficients to its measures after standardizing each one, so home value and income carry less weight than in the ADI. Income and degree share are single census measures, each ranked from the highest value to the lowest.

Source: CDC PLACES 2024 release (BRFSS 2022), ACS 2019-2023 and Area Deprivation Index 2023; modeled estimates, weighted by adult population; ADI 2023 and ACS 2019-2023 tract tables (B19013, B15003 and the 15 rebuilt ADI measures)

Data: https://vishalsingh.org/zipcode/report/disability-retirement-and-medicare/figures/alt_rankings.json

## All estimates

Every fact in facts.json (165), sorted by id.

| Fact id | Value | Unit | Estimand | Population | Period | Interval | n | Weights | Source | Note |
| --- | ---: | --- | --- | --- | --- | --- | ---: | --- | --- | --- |
| age65_d1 | 17.3% | percent of residents aged 65 and over | mean of the tract share aged 65 and over, weighted by tract population, bootstrap interval over counties | residents of the least deprived tenth of tracts by ADI national rank | ACS 2019-2023; ADI 2023 | 95% CI 16.6% to 18.0% | 8,274 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| age65_d10 | 16.6% | percent of residents aged 65 and over | mean of the tract share aged 65 and over, weighted by tract population, bootstrap interval over counties | residents of the most deprived tenth of tracts by ADI national rank | ACS 2019-2023; ADI 2023 | 95% CI 16.1% to 17.0% | 8,213 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| age_adult | 18 | years of age | age at which a person is counted as an adult in PLACES (a definition) | adults counted in the PLACES disability measure | 2024 release of CDC PLACES (BRFSS responses from 2022); ADI 2023 |   |   | none (definition) | CDC PLACES |   |
| age_dec_hi | 35.8% | percent of residents aged 65 and over | mean of the tract share aged 65 and over, weighted by adult population | residents of the tenth of tracts with the most residents aged 65 and over | ACS 2019-2023 |   | 8,286 tracts | adult population of the tract (PLACES) | American Community Survey |   |
| age_dec_lo | 5.9% | percent of residents aged 65 and over | mean of the tract share aged 65 and over, weighted by adult population | residents of the tenth of tracts with the fewest residents aged 65 and over | ACS 2019-2023 |   | 8,275 tracts | adult population of the tract (PLACES) | American Community Survey |   |
| age_senior | 65 | years of age | age at which an adult is counted as older in these measures (a definition) | adults counted as older adults in PLACES tooth loss, in the ACS age share and in the Social Security aged-65-and-over counts | 2024 release of CDC PLACES (BRFSS responses from 2022); ADI 2023 |   |   | none (definition) | CDC PLACES |   |
| alt_adi_least | 23.1% | Adults with any disability (percent of adults) | adult-weighted mean of tract disability prevalence (crude, modeled); 95% intervals from a bootstrap over counties | the least deprived tenth of tracts ranked by Area Deprivation Index, among the tracts with all four rankings | 2024 release of CDC PLACES (BRFSS responses from 2022); ADI 2023; ACS 2019-2023 | 95% CI 22.1% to 23.9% | 7,551 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| alt_adi_most | 42.6% | Adults with any disability (percent of adults) | adult-weighted mean of tract disability prevalence (crude, modeled); 95% intervals from a bootstrap over counties | the most deprived tenth of tracts ranked by Area Deprivation Index, among the tracts with all four rankings | 2024 release of CDC PLACES (BRFSS responses from 2022); ADI 2023; ACS 2019-2023 | 95% CI 42.2% to 43.0% | 7,543 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| alt_adi_rho | +1.00 | rank correlation | Spearman rank correlation between bin order and bin mean, each bin counted once | the bins of tracts ranked by Area Deprivation Index | 2024 release of CDC PLACES (BRFSS responses from 2022); ADI 2023; ACS 2019-2023 |   |   | each bin counted once | CDC PLACES |   |
| alt_degree_least | 20.7% | Adults with any disability (percent of adults) | adult-weighted mean of tract disability prevalence (crude, modeled); 95% intervals from a bootstrap over counties | the least deprived tenth of tracts ranked by Bachelor's degree share, among the tracts with all four rankings | 2024 release of CDC PLACES (BRFSS responses from 2022); ADI 2023; ACS 2019-2023 | 95% CI 20.3% to 21.1% | 7,558 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| alt_degree_most | 42.0% | Adults with any disability (percent of adults) | adult-weighted mean of tract disability prevalence (crude, modeled); 95% intervals from a bootstrap over counties | the most deprived tenth of tracts ranked by Bachelor's degree share, among the tracts with all four rankings | 2024 release of CDC PLACES (BRFSS responses from 2022); ADI 2023; ACS 2019-2023 | 95% CI 41.6% to 42.5% | 7,564 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| alt_degree_rho | +1.00 | rank correlation | Spearman rank correlation between bin order and bin mean, each bin counted once | the bins of tracts ranked by Bachelor's degree share | 2024 release of CDC PLACES (BRFSS responses from 2022); ADI 2023; ACS 2019-2023 |   |   | each bin counted once | CDC PLACES |   |
| alt_income_least | 21.4% | Adults with any disability (percent of adults) | adult-weighted mean of tract disability prevalence (crude, modeled); 95% intervals from a bootstrap over counties | the least deprived tenth of tracts ranked by Median household income, among the tracts with all four rankings | 2024 release of CDC PLACES (BRFSS responses from 2022); ADI 2023; ACS 2019-2023 | 95% CI 21.0% to 21.8% | 7,560 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| alt_income_most | 42.9% | Adults with any disability (percent of adults) | adult-weighted mean of tract disability prevalence (crude, modeled); 95% intervals from a bootstrap over counties | the most deprived tenth of tracts ranked by Median household income, among the tracts with all four rankings | 2024 release of CDC PLACES (BRFSS responses from 2022); ADI 2023; ACS 2019-2023 | 95% CI 42.6% to 43.3% | 7,559 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| alt_income_rho | +1.00 | rank correlation | Spearman rank correlation between bin order and bin mean, each bin counted once | the bins of tracts ranked by Median household income | 2024 release of CDC PLACES (BRFSS responses from 2022); ADI 2023; ACS 2019-2023 |   |   | each bin counted once | CDC PLACES |   |
| alt_n_places | 75,596 | tracts | count of tracts in the common sample of the ranking comparison | tracts with the outcome and all four rankings | 2024 release of CDC PLACES (BRFSS responses from 2022); ADI 2023; ACS 2019-2023 |   |   | none (full count) | CDC PLACES |   |
| alt_rho_cut | 0.9 | absolute rank correlation | threshold between a steady and an uneven gradient | the rule for changed gradients (docs/RANKING_SWITCH_RULE.md) | 2024 release of CDC PLACES (BRFSS responses from 2022); ADI 2023; ACS 2019-2023 |   |   | none (a fixed rule) | CDC PLACES |   |
| alt_rho_margin | 0.1 | absolute rank correlation | smallest move in the rank correlation that counts as a change of shape | the rule for changed gradients (docs/RANKING_SWITCH_RULE.md) | 2024 release of CDC PLACES (BRFSS responses from 2022); ADI 2023; ACS 2019-2023 |   |   | none (a fixed rule) | CDC PLACES |   |
| alt_std_least | 21.4% | Adults with any disability (percent of adults) | adult-weighted mean of tract disability prevalence (crude, modeled); 95% intervals from a bootstrap over counties | the least deprived tenth of tracts ranked by Standardized index, among the tracts with all four rankings | 2024 release of CDC PLACES (BRFSS responses from 2022); ADI 2023; ACS 2019-2023 | 95% CI 21.0% to 21.8% | 7,559 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| alt_std_most | 43.4% | Adults with any disability (percent of adults) | adult-weighted mean of tract disability prevalence (crude, modeled); 95% intervals from a bootstrap over counties | the most deprived tenth of tracts ranked by Standardized index, among the tracts with all four rankings | 2024 release of CDC PLACES (BRFSS responses from 2022); ADI 2023; ACS 2019-2023 | 95% CI 42.8% to 43.9% | 7,560 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| alt_std_rho | +1.00 | rank correlation | Spearman rank correlation between bin order and bin mean, each bin counted once | the bins of tracts ranked by Standardized index | 2024 release of CDC PLACES (BRFSS responses from 2022); ADI 2023; ACS 2019-2023 |   |   | each bin counted once | CDC PLACES |   |
| belt_k | 12 | counties | number of counties whose names the text gives | counties with the highest disabled-worker share | December 2024 |   |   | none (rule) | SSA beneficiaries by ZIP code |   |
| belt_kth | 23.0% | percent of OASDI beneficiaries who are disabled workers | disabled workers divided by all beneficiaries in the county's ZIP codes | OASDI beneficiaries in the county with the twelfth-highest disabled-worker share among counties with at least 5,000 beneficiaries | December 2024 |   | 6,980 OASDI beneficiaries | none (full count) | SSA beneficiaries by ZIP code |   |
| belt_min | 5,000 | OASDI beneficiaries | minimum number of beneficiaries for a county to enter the ranking | counties in the county ranking of disabled-worker shares | December 2024 |   |   | none (inclusion rule) | SSA beneficiaries by ZIP code |   |
| belt_min_alt | 10,000 | OASDI beneficiaries | alternative minimum number of beneficiaries for a county to enter the ranking | counties in the check of the county ranking of disabled-worker shares | December 2024 |   |   | none (inclusion rule) | SSA beneficiaries by ZIP code |   |
| belt_top | 27.2% | percent of OASDI beneficiaries who are disabled workers | disabled workers divided by all beneficiaries in the county's ZIP codes | OASDI beneficiaries in the county with the highest disabled-worker share among counties with at least 5,000 beneficiaries | December 2024 |   | 15,925 OASDI beneficiaries | none (full count) | SSA beneficiaries by ZIP code |   |
| ben_inc_d1 | $1,554 | dollars per month, average retired-worker benefit | benefit dollars divided by retired workers, pooled over the tenth's ZIP codes (benefit totals in thousands of dollars converted to dollars; SSA rounds the totals), bootstrap interval over counties | retired workers in the lowest tenth of ZIP codes by median household income | December 2024 | 95% CI $1,523 to $1,586 | 2,863 ZIP codes | retired workers | SSA beneficiaries by ZIP code |   |
| ben_inc_d10 | $2,370 | dollars per month, average retired-worker benefit | benefit dollars divided by retired workers, pooled over the tenth's ZIP codes (benefit totals in thousands of dollars converted to dollars; SSA rounds the totals), bootstrap interval over counties | retired workers in the highest tenth of ZIP codes by median household income | December 2024 | 95% CI $2,354 to $2,386 | 2,855 ZIP codes | retired workers | SSA beneficiaries by ZIP code |   |
| beta_ben_adi | 0.08 | standard deviations of the outcome per standard deviation of the predictor | standardized coefficient on the ADI national rank for the average monthly retired-worker benefit in a regression on the ADI rank, median income and share aged 65 and over, weighted by ZIP-code population, errors clustered on county | ZIP codes in the 50 states and DC with the outcome, ADI, median household income and share aged 65 and over | OASDI December 2024; ACS 2019-2023; ADI 2023 | 95% CI −0.02 to 0.18 | 28,547 ZIP codes | ACS 2019-2023 ZCTA population | SSA beneficiaries by ZIP code |   |
| beta_ben_age | 0.16 | standard deviations of the outcome per standard deviation of the predictor | standardized coefficient on the share of residents aged 65 and over for the average monthly retired-worker benefit in a regression on the ADI rank, median income and share aged 65 and over, weighted by ZIP-code population, errors clustered on county | ZIP codes in the 50 states and DC with the outcome, ADI, median household income and share aged 65 and over | OASDI December 2024; ACS 2019-2023; ADI 2023 | 95% CI 0.14 to 0.18 | 28,547 ZIP codes | ACS 2019-2023 ZCTA population | SSA beneficiaries by ZIP code |   |
| beta_ben_inc | 0.89 | standard deviations of the outcome per standard deviation of the predictor | standardized coefficient on ZIP median household income for the average monthly retired-worker benefit in a regression on the ADI rank, median income and share aged 65 and over, weighted by ZIP-code population, errors clustered on county | ZIP codes in the 50 states and DC with the outcome, ADI, median household income and share aged 65 and over | OASDI December 2024; ACS 2019-2023; ADI 2023 | 95% CI 0.80 to 0.98 | 28,547 ZIP codes | ACS 2019-2023 ZCTA population | SSA beneficiaries by ZIP code |   |
| beta_dis_adi | 0.46 | standard deviations of the outcome per standard deviation of the predictor | standardized coefficient on the ADI national rank for a ZIP code's disabled-worker share of OASDI beneficiaries in a regression on the ADI rank, median income and share aged 65 and over, weighted by ZIP-code population, errors clustered on county | ZIP codes in the 50 states and DC with the outcome, ADI, median household income and share aged 65 and over | OASDI December 2024; ACS 2019-2023; ADI 2023 | 95% CI 0.40 to 0.52 | 30,103 ZIP codes | ACS 2019-2023 ZCTA population | SSA beneficiaries by ZIP code |   |
| beta_dis_inc | −0.36 | standard deviations of the outcome per standard deviation of the predictor | standardized coefficient on ZIP median household income for a ZIP code's disabled-worker share of OASDI beneficiaries in a regression on the ADI rank, median income and share aged 65 and over, weighted by ZIP-code population, errors clustered on county | ZIP codes in the 50 states and DC with the outcome, ADI, median household income and share aged 65 and over | OASDI December 2024; ACS 2019-2023; ADI 2023 | 95% CI −0.42 to −0.30 | 30,103 ZIP codes | ACS 2019-2023 ZCTA population | SSA beneficiaries by ZIP code |   |
| big_pop | 250,000 | residents | county population at or above which a county is called large in the check | counties in the larger-county check | 2022 estimate in the County Health Rankings 2024 release |   |   | none (cutoff) | County Health Rankings |   |
| chetty_r_smoke | −0.69 | correlation coefficient | correlation between life expectancy in the lowest income quartile and smoking across local areas, reported in the abstract of Chetty and coauthors (2016); copied from it and not computed in this build | local areas in the Health Inequality Project's analysis of life expectancy in the lowest income quartile | life expectancy 2001 to 2014; area measures as in the paper |   |   | as in the paper | Published estimates from cited studies | A published estimate for the local areas the paper analyzes, not a result of this build. |
| corr_adi | −0.67 | correlation coefficient | correlation across counties of expected age at death at 40 (average of the women's and men's estimates, adjusted to the national racial composition) with the county Area Deprivation Index, population-weighted, bootstrap interval over states | counties in the Health Inequality Project file with the measure; poorest quartile of national household income | expected age at death 2001 to 2014; ADI 2023, mean of tract ranks weighted by ACS 2019-2023 population | 95% CI −0.74 to −0.51 | 1,559 counties | county residents (County Health Rankings 2024) | Health Inequality Project |   |
| corr_adi_rich | −0.22 | correlation coefficient | correlation across counties of expected age at death at 40 (average of the women's and men's estimates, adjusted to the national racial composition) with the county Area Deprivation Index, population-weighted, bootstrap interval over states | counties in the Health Inequality Project file with the measure; richest quartile of national household income | expected age at death 2001 to 2014; ADI 2023, mean of tract ranks weighted by ACS 2019-2023 population | 95% CI −0.43 to −0.07 | 1,559 counties | county residents (County Health Rankings 2024) | Health Inequality Project |   |
| corr_adi_unw | −0.44 | correlation coefficient | correlation across counties of expected age at death at 40 (average of the women's and men's estimates, adjusted to the national racial composition) with the county Area Deprivation Index, each county once, bootstrap interval over states | counties in the Health Inequality Project file with the measure; poorest quartile of national household income | expected age at death 2001 to 2014; ADI 2023, mean of tract ranks weighted by ACS 2019-2023 population | 95% CI −0.53 to −0.34 | 1,559 counties | none (each county counts once) | Health Inequality Project |   |
| corr_college | +0.32 | correlation coefficient | correlation across counties of expected age at death at 40 (average of the women's and men's estimates, adjusted to the national racial composition) with the share of adults with some college, population-weighted, bootstrap interval over states | counties in the Health Inequality Project file with the measure; poorest quartile of national household income | expected age at death 2001 to 2014; ACS 2018-2022 | 95% CI 0.25 to 0.40 | 1,559 counties | county residents (County Health Rankings 2024) | Health Inequality Project |   |
| corr_inactive_rich | −0.50 | correlation coefficient | correlation across counties of expected age at death at 40 (average of the women's and men's estimates, adjusted to the national racial composition) with the share of adults with no leisure-time physical activity, population-weighted, bootstrap interval over states | counties in the Health Inequality Project file with the measure; richest quartile of national household income | expected age at death 2001 to 2014; BRFSS 2021 as modeled by County Health Rankings 2024 | 95% CI −0.60 to −0.41 | 1,559 counties | county residents (County Health Rankings 2024) | Health Inequality Project |   |
| corr_income | +0.48 | correlation coefficient | correlation across counties of expected age at death at 40 (average of the women's and men's estimates, adjusted to the national racial composition) with median household income, population-weighted, bootstrap interval over states | counties in the Health Inequality Project file with the measure; poorest quartile of national household income | expected age at death 2001 to 2014; income from SAIPE 2022 | 95% CI 0.38 to 0.55 | 1,559 counties | county residents (County Health Rankings 2024) | Health Inequality Project |   |
| corr_obese | −0.64 | correlation coefficient | correlation across counties of expected age at death at 40 (average of the women's and men's estimates, adjusted to the national racial composition) with the share of adults with obesity, population-weighted, bootstrap interval over states | counties in the Health Inequality Project file with the measure; poorest quartile of national household income | expected age at death 2001 to 2014; obesity from BRFSS 2021 as modeled by County Health Rankings 2024 | 95% CI −0.71 to −0.51 | 1,559 counties | county residents (County Health Rankings 2024) | Health Inequality Project |   |
| corr_obese_unw | −0.45 | correlation coefficient | correlation across counties of expected age at death at 40 (average of the women's and men's estimates, adjusted to the national racial composition) with the share of adults with obesity, each county once, bootstrap interval over states | counties in the Health Inequality Project file with the measure; poorest quartile of national household income | expected age at death 2001 to 2014; obesity from BRFSS 2021 as modeled by County Health Rankings 2024 | 95% CI −0.54 to −0.34 | 1,559 counties | none (each county counts once) | Health Inequality Project |   |
| corr_pcp | +0.28 | correlation coefficient | correlation across counties of expected age at death at 40 (average of the women's and men's estimates, adjusted to the national racial composition) with primary care physicians per 100,000 residents, population-weighted, bootstrap interval over states | counties in the Health Inequality Project file with the measure; poorest quartile of national household income | expected age at death 2001 to 2014; AMA and Area Health Resources File 2021 | 95% CI 0.20 to 0.35 | 1,559 counties | county residents (County Health Rankings 2024) | Health Inequality Project |   |
| corr_smoke | −0.58 | correlation coefficient | correlation across counties of expected age at death at 40 (average of the women's and men's estimates, adjusted to the national racial composition) with the share of adults who smoke, population-weighted, bootstrap interval over states | counties in the Health Inequality Project file with the measure; poorest quartile of national household income | expected age at death 2001 to 2014; smoking from BRFSS 2021 as modeled by County Health Rankings 2024 | 95% CI −0.67 to −0.45 | 1,559 counties | county residents (County Health Rankings 2024) | Health Inequality Project |   |
| corr_smoke_unw | −0.34 | correlation coefficient | correlation across counties of expected age at death at 40 (average of the women's and men's estimates, adjusted to the national racial composition) with the share of adults who smoke, each county once, bootstrap interval over states | counties in the Health Inequality Project file with the measure; poorest quartile of national household income | expected age at death 2001 to 2014; smoking from BRFSS 2021 as modeled by County Health Rankings 2024 | 95% CI −0.44 to −0.24 | 1,559 counties | none (each county counts once) | Health Inequality Project |   |
| corr_uninsured | −0.15 | correlation coefficient | correlation across counties of expected age at death at 40 (average of the women's and men's estimates, adjusted to the national racial composition) with the share of people under 65 without health insurance, population-weighted, bootstrap interval over states | counties in the Health Inequality Project file with the measure; poorest quartile of national household income | expected age at death 2001 to 2014; SAHIE 2021 | 95% CI −0.28 to 0.06 | 1,559 counties | county residents (County Health Rankings 2024) | Health Inequality Project |   |
| corr_uninsured_rich | −0.27 | correlation coefficient | correlation across counties of expected age at death at 40 (average of the women's and men's estimates, adjusted to the national racial composition) with the share of people under 65 without health insurance, population-weighted, bootstrap interval over states | counties in the Health Inequality Project file with the measure; richest quartile of national household income | expected age at death 2001 to 2014; SAHIE 2021 | 95% CI −0.37 to −0.18 | 1,559 counties | county residents (County Health Rankings 2024) | Health Inequality Project |   |
| corr_uninsured_unw | −0.16 | correlation coefficient | correlation across counties of expected age at death at 40 (average of the women's and men's estimates, adjusted to the national racial composition) with the share of people under 65 without health insurance, each county once, bootstrap interval over states | counties in the Health Inequality Project file with the measure; poorest quartile of national household income | expected age at death 2001 to 2014; SAHIE 2021 | 95% CI −0.29 to 0.01 | 1,559 counties | none (each county counts once) | Health Inequality Project |   |
| dis_age_diff | 1.8 | percentage points of adult disability prevalence, tenth of tracts with the most residents aged 65 and over minus the tenth with the fewest | difference of two adult-population-weighted means (tenths of tracts by age share), bootstrap interval over counties (shared draws) | adults aged 18 and over in tracts with a PLACES disability estimate, an ADI rank and an ACS age share | 2024 release of CDC PLACES (BRFSS responses from 2022); ADI 2023 | 95% CI 0.9 to 2.7 | 82,779 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| dis_age_hi | 32.3% | percent of adults with any disability | mean of tract prevalence, weighted by adult population, bootstrap interval over counties (tenths hold equal numbers of tracts) | adults in the tenth of tracts with the most residents aged 65 and over | 2024 release of CDC PLACES (BRFSS responses from 2022); ADI 2023 | 95% CI 31.8% to 32.7% | 8,286 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| dis_age_lo | 30.5% | percent of adults with any disability | mean of tract prevalence, weighted by adult population, bootstrap interval over counties (tenths hold equal numbers of tracts) | adults in the tenth of tracts with the fewest residents aged 65 and over | 2024 release of CDC PLACES (BRFSS responses from 2022); ADI 2023 | 95% CI 29.6% to 31.3% | 8,275 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| dis_d1 | 23.3% | percent of adults with any disability | mean of tract prevalence (crude, modeled), weighted by adult population, bootstrap interval over counties | adults in the least deprived tenth of tracts by ADI national rank | 2024 release of CDC PLACES (BRFSS responses from 2022); ADI 2023 | 95% CI 22.3% to 24.1% | 8,274 tracts | adult population of the tract (PLACES) | CDC PLACES | PLACES models these values from survey responses and census inputs that overlap the ADI, so part of the gradient is built in. |
| dis_d10 | 42.9% | percent of adults with any disability | mean of tract prevalence (crude, modeled), weighted by adult population, bootstrap interval over counties | adults in the most deprived tenth of tracts by ADI national rank | 2024 release of CDC PLACES (BRFSS responses from 2022); ADI 2023 | 95% CI 42.5% to 43.2% | 8,213 tracts | adult population of the tract (PLACES) | CDC PLACES | PLACES models these values from survey responses and census inputs that overlap the ADI, so part of the gradient is built in. |
| dis_q1 | 7.2% | percent of OASDI beneficiaries who are disabled workers | share of the fifth's beneficiaries, pooled over its ZIP codes, bootstrap interval over counties | OASDI beneficiaries in the least deprived fifth of ZIP codes by ADI national rank | December 2024 | 95% CI 6.9% to 7.5% | 6,363 ZIP codes | beneficiaries (each ZIP code counts by its beneficiary count) | SSA beneficiaries by ZIP code | Disabled-worker benefits become retired-worker benefits at full retirement age, so former disabled workers are counted as retired workers after it. |
| dis_q5 | 16.8% | percent of OASDI beneficiaries who are disabled workers | share of the fifth's beneficiaries, pooled over its ZIP codes, bootstrap interval over counties | OASDI beneficiaries in the most deprived fifth of ZIP codes by ADI national rank | December 2024 | 95% CI 16.4% to 17.3% | 6,363 ZIP codes | beneficiaries (each ZIP code counts by its beneficiary count) | SSA beneficiaries by ZIP code | Disabled-worker benefits become retired-worker benefits at full retirement age, so former disabled workers are counted as retired workers after it. |
| dis_ratio | 1.8 | ratio of disability prevalence in the most deprived tenth of tracts to that in the least deprived tenth | ratio of two adult-population-weighted means, bootstrap interval over counties (shared draws) | adults aged 18 and over in tracts with a PLACES disability estimate, an ADI rank and an ACS age share | 2024 release of CDC PLACES (BRFSS responses from 2022); ADI 2023 | 95% CI 1.8 to 1.9 | 82,779 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| em_diff | −$430 | dollars per fee-for-service beneficiary per year, most deprived fifth minus least deprived fifth, payments for evaluation and management visits | difference of two fee-for-service-beneficiary-weighted means of county payments per capita, bootstrap interval over states (shared draws) | fee-for-service beneficiaries in counties with every shown service line reported | 2024 | 95% CI −$543 to −$296 | 2,867 counties | fee-for-service beneficiaries | Medicare Geographic Variation |   |
| em_drop | 32% | percent lower payments for evaluation and management visits per beneficiary in the most deprived fifth of counties than in the least deprived fifth | one minus the ratio of the most deprived fifth's weighted mean to the least deprived fifth's, bootstrap interval over states (shared draws) | fee-for-service beneficiaries in counties with every shown service line reported | 2024 | 95% CI 24% to 39% | 2,867 counties | fee-for-service beneficiaries | Medicare Geographic Variation |   |
| em_q1 | $1,333 | dollars per fee-for-service beneficiary per year, price-standardized payments for evaluation and management visits | mean of county payments per capita, weighted by fee-for-service beneficiaries, bootstrap interval over states | fee-for-service beneficiaries in the least deprived fifth of counties by county ADI, counties with every shown service line reported | 2024 | 95% CI $1,218 to $1,426 | 605 counties | fee-for-service beneficiaries | Medicare Geographic Variation |   |
| em_q5 | $903 | dollars per fee-for-service beneficiary per year, price-standardized payments for evaluation and management visits | mean of county payments per capita, weighted by fee-for-service beneficiaries, bootstrap interval over states | fee-for-service beneficiaries in the most deprived fifth of counties by county ADI, counties with every shown service line reported | 2024 | 95% CI $831 to $967 | 541 counties | fee-for-service beneficiaries | Medicare Geographic Variation |   |
| er_q1 | 571 | emergency department visits per 1,000 fee-for-service beneficiaries per year | mean of county rates, weighted by fee-for-service beneficiaries, bootstrap interval over states | fee-for-service beneficiaries in the least deprived fifth of counties by county ADI | 2024 | 95% CI 556 to 584 | 625 counties | fee-for-service beneficiaries | Medicare Geographic Variation |   |
| er_q5 | 677 | emergency department visits per 1,000 fee-for-service beneficiaries per year | mean of county rates, weighted by fee-for-service beneficiaries, bootstrap interval over states | fee-for-service beneficiaries in the most deprived fifth of counties by county ADI | 2024 | 95% CI 656 to 694 | 624 counties | fee-for-service beneficiaries | Medicare Geographic Variation |   |
| er_r | +0.37 | correlation coefficient | correlation of the county emergency department visit rate with the county ADI, weighted by fee-for-service beneficiaries, bootstrap interval over states | fee-for-service (Original Medicare) beneficiaries in counties with a standardized payment and an ADI rank | 2024 | 95% CI 0.26 to 0.47 | 3,128 counties | fee-for-service beneficiaries | Medicare Geographic Variation |   |
| hh_diff | $224 | dollars per fee-for-service beneficiary per year, most deprived fifth minus least deprived fifth, payments for home health and hospice | difference of two fee-for-service-beneficiary-weighted means of county payments per capita, bootstrap interval over states (shared draws) | fee-for-service beneficiaries in counties with every shown service line reported | 2024 | 95% CI −$56 to $511 | 2,867 counties | fee-for-service beneficiaries | Medicare Geographic Variation |   |
| hip_age | 40 | years of age | age at which expected age at death is measured (a definition) | people whose expected age at death the Project estimates | 2001 to 2014 |   |   | none (definition) | Health Inequality Project |   |
| hip_median_pop | 69,380 | residents | median county population in the County Health Rankings 2024 release (a Census Bureau estimate for 2022) among the counties in the file | counties in the Health Inequality Project file | 2022 estimate in the County Health Rankings 2024 release |   | 1,559 counties | none (median over counties) | County Health Rankings |   |
| hip_min_pop | 15,304 | residents | smallest county population in the County Health Rankings 2024 release (a Census Bureau estimate for 2022) among the counties in the file | the smallest county in the Health Inequality Project file | 2022 estimate in the County Health Rankings 2024 release |   |   | none (one county) | County Health Rankings |   |
| income_gap | 6.7 | years of expected age at death for people who reach age 40, richest quartile minus poorest | difference of two county-population-weighted means (average of the women's and men's estimates, adjusted to the national racial composition), bootstrap interval over states (shared draws) | residents who reach age 40 in the counties of the Health Inequality Project file | 2001 to 2014 | 95% CI 6.3 to 7.2 | 1,559 counties | county residents (County Health Rankings 2024) | Health Inequality Project |   |
| ip_diff | $134 | dollars per fee-for-service beneficiary per year, most deprived fifth minus least deprived fifth, payments for inpatient hospital stays | difference of two fee-for-service-beneficiary-weighted means of county payments per capita, bootstrap interval over states (shared draws) | fee-for-service beneficiaries in counties with every shown service line reported | 2024 | 95% CI −$12 to $315 | 2,867 counties | fee-for-service beneficiaries | Medicare Geographic Variation |   |
| le_r_adi | −0.82 | correlation coefficient | correlation of county life expectancy at birth with the county ADI, weighted by fee-for-service beneficiaries, bootstrap interval over states | fee-for-service beneficiaries in counties with a standardized payment, an ADI rank and a County Health Rankings life expectancy (Connecticut left out) | payments 2024; life expectancy for deaths in 2019 to 2021 | 95% CI −0.85 to −0.77 | 3,057 counties | fee-for-service beneficiaries | County Health Rankings |   |
| le_r_spend | −0.07 | correlation coefficient | correlation of county price-standardized payments per capita with county life expectancy at birth, weighted by fee-for-service beneficiaries, bootstrap interval over states | fee-for-service beneficiaries in counties with a standardized payment, an ADI rank and a County Health Rankings life expectancy (Connecticut left out) | payments 2024; life expectancy for deaths in 2019 to 2021 | 95% CI −0.27 to 0.05 | 3,057 counties | fee-for-service beneficiaries | County Health Rankings |   |
| le_r_spend_unw | −0.35 | correlation coefficient | correlation of county price-standardized payments per capita with county life expectancy at birth, each county counted once, bootstrap interval over states | fee-for-service beneficiaries in counties with a standardized payment, an ADI rank and a County Health Rankings life expectancy (Connecticut left out) | payments 2024; life expectancy for deaths in 2019 to 2021 | 95% CI −0.43 to −0.25 | 3,057 counties | none (each county counts once) | County Health Rankings |   |
| le_slope_1k | −0.12 | years of life expectancy at birth per 1,000 dollars of price-standardized payments per beneficiary | beneficiary-weighted least-squares slope of county life expectancy on county payments per capita, bootstrap interval over states | fee-for-service beneficiaries in counties with a standardized payment, an ADI rank and a County Health Rankings life expectancy (Connecticut left out) | payments 2024; life expectancy for deaths in 2019 to 2021 | 95% CI −0.51 to 0.08 | 3,057 counties | fee-for-service beneficiaries | County Health Rankings |   |
| le_slope_ctrl | −0.17 | years of life expectancy at birth per 1,000 dollars of price-standardized payments per beneficiary, holding the county ADI fixed | coefficient on payments in a least-squares regression of county life expectancy on payments and the county ADI, weighted by fee-for-service beneficiaries, errors clustered on state | fee-for-service beneficiaries in counties with a standardized payment, an ADI rank and a County Health Rankings life expectancy (Connecticut left out) | payments 2024; life expectancy for deaths in 2019 to 2021 | 95% CI −0.33 to −0.01 | 3,057 counties | fee-for-service beneficiaries | County Health Rankings |   |
| le_slope_ctrl_unw | −0.28 | years of life expectancy at birth per 1,000 dollars of price-standardized payments per beneficiary, holding the county ADI fixed | coefficient on payments in a least-squares regression of county life expectancy on payments and the county ADI, each county counted once, errors clustered on state | fee-for-service beneficiaries in counties with a standardized payment, an ADI rank and a County Health Rankings life expectancy (Connecticut left out) | payments 2024; life expectancy for deaths in 2019 to 2021 | 95% CI −0.40 to −0.16 | 3,057 counties | none (each county counts once) | County Health Rankings |   |
| ma_share | 54.5% | percent of beneficiaries with Parts A and B who are in Medicare Advantage | Medicare Advantage beneficiaries divided by Medicare Advantage plus Original Medicare beneficiaries | beneficiaries with Parts A and B in the counties counted | 2024 |   |   | none (full count) | Medicare Geographic Variation |   |
| medicare_age | 65 | years of age | age at which eligibility for Medicare begins (a definition) | people who become eligible for Medicare through age | 2024 |   |   | none (definition) | Medicare Geographic Variation |   |
| min_retired | 50 | retired workers | minimum number of retired workers for a ZIP code to enter the benefit analysis | ZIP codes in the benefit analysis | December 2024 |   |   | none (inclusion rule) | SSA beneficiaries by ZIP code |   |
| n_belt | 1,838 | counties | count of counties | counties with at least 5,000 OASDI beneficiaries in the ZIP codes the atlas assigns to them | December 2024 |   |   | none (full count) | SSA beneficiaries by ZIP code | A ZIP code is counted in one county, the one the atlas assigns to it, so a county total is approximate where ZIP codes cross county lines. |
| n_benes_ssa | 65.9 million | OASDI beneficiaries | sum of beneficiary counts (SSA controlled-rounded counts), ZIP codes of the same state line summed | OASDI beneficiaries living in the ZIP codes counted | December 2024 |   |   | none (full count) | SSA beneficiaries by ZIP code |   |
| n_big_cty | 283 | counties | count of counties | counties in the Health Inequality Project file with at least 250,000 residents in the County Health Rankings 2024 release | 2001 to 2014 |   |   | none (full count) | Health Inequality Project |   |
| n_covariates | 15 | county measures | number of County Health Rankings measures in the correlation figure | county measures correlated with expected age at death | 2018 to 2023 |   |   | none (count) | County Health Rankings |   |
| n_cty_med | 3,128 | counties | count of counties | counties and county equivalents (Connecticut by planning region) with a standardized payment, fee-for-service beneficiaries and an ADI rank | 2024 |   |   | none (full count) | Medicare Geographic Variation |   |
| n_ffs | 27.6 million | fee-for-service beneficiaries | sum of Original Medicare beneficiaries (the denominator of the per-capita payments) | beneficiaries enrolled in Original Medicare with Parts A and B in the counties counted | 2024 |   |   | none (full count) | Medicare Geographic Variation |   |
| n_hip | 1,559 | counties | count of counties | counties in the Health Inequality Project file (the larger counties; Connecticut by legacy county) | 2001 to 2014 |   |   | none (full count) | Health Inequality Project |   |
| n_le_cty | 3,057 | counties | count of counties | counties with a standardized payment, an ADI rank and a County Health Rankings life expectancy (Connecticut left out) | payments 2024; life expectancy for deaths in 2019 to 2021 |   |   | none (full count) | County Health Rankings |   |
| n_map_counties | 1,559 | counties | count of sprite counties with an estimate | counties drawn with a value on the county map | 2001 to 2014 |   |   | none (full count) | Health Inequality Project |   |
| n_outside_age | 398 | tracts | count of tracts outside the plotted range | tracts outside the plotted range of the age-structure scatter plot | 2024 release of CDC PLACES (BRFSS responses from 2022); ADI 2023 |   |   | none (full count) | CDC PLACES |   |
| n_outside_ben | 102 | ZIP codes | count of ZIP codes outside the plotted range | ZIP codes outside the plotted range of the benefit scatter plot | December 2024 |   |   | none (full count) | SSA beneficiaries by ZIP code |   |
| n_outside_le | 34 | counties | count of counties outside the plotted range | counties outside the plotted range of the payments-and-life-expectancy scatter plot | payments 2024; life expectancy for deaths in 2019 to 2021 |   |   | none (full count) | County Health Rankings |   |
| n_rich_higher | 1,558 | counties | count of counties (average of the women's and men's estimates, adjusted to the national racial composition) | counties in the Health Inequality Project file in which the richest quartile's expected age at death exceeds the poorest quartile's | 2001 to 2014 |   | 1,559 counties | none (full count) | Health Inequality Project |   |
| n_sample_age | 2,000 | tracts | size of a random sample drawn with a fixed seed from the tracts inside the plotted range | tracts drawn at random for the scatter plot | 2024 release of CDC PLACES (BRFSS responses from 2022); ADI 2023 |   | 82,779 tracts | none (sample size) | CDC PLACES |   |
| n_sample_ben | 2,000 | ZIP codes | size of a random sample drawn with a fixed seed from the ZIP codes inside the plotted range | ZIP codes drawn at random for the scatter plot | December 2024 |   | 28,547 ZIP codes | none (sample size) | SSA beneficiaries by ZIP code |   |
| n_tracts_dis | 82,779 | tracts | count of tracts | 2020 census tracts with a PLACES any-disability estimate, an ADI rank and an ACS 2019-2023 share aged 65 and over | 2024 release of CDC PLACES (BRFSS responses from 2022); ADI 2023 |   |   | none (full count) | CDC PLACES |   |
| n_zips_ben | 28,547 | ZIP codes | count of ZIP codes | ZIP codes with at least 50 retired workers, a median household income and an ADI rank | December 2024 |   |   | none (full count) | SSA beneficiaries by ZIP code |   |
| n_zips_ssa | 31,815 | ZIP codes | count of ZIP codes | ZIP codes in the 50 states and DC with OASDI beneficiaries that match a ZCTA with an ADI rank, a population and a county | December 2024 |   |   | none (full count) | SSA beneficiaries by ZIP code |   |
| op_diff | $410 | dollars per fee-for-service beneficiary per year, most deprived fifth minus least deprived fifth, payments for hospital outpatient departments and surgery centers | difference of two fee-for-service-beneficiary-weighted means of county payments per capita, bootstrap interval over states (shared draws) | fee-for-service beneficiaries in counties with every shown service line reported | 2024 | 95% CI $206 to $601 | 2,867 counties | fee-for-service beneficiaries | Medicare Geographic Variation |   |
| oth_diff | $480 | dollars per fee-for-service beneficiary per year, most deprived fifth minus least deprived fifth, payments for all other services (the total minus the lines shown) | difference of two fee-for-service-beneficiary-weighted means of county payments per capita, bootstrap interval over states (shared draws) | fee-for-service beneficiaries in counties with every shown service line reported | 2024 | 95% CI $243 to $736 | 2,867 counties | fee-for-service beneficiaries | Medicare Geographic Variation |   |
| pen_q1 | 41.7% | percent of the sum of taxable pensions and annuities and annualized OASDI benefits | pooled dollars of taxable pensions and annuities (IRS line for pensions and annuities in adjusted gross income) divided by those dollars plus 12 times the December 2024 monthly OASDI benefits, bootstrap interval over counties | ZIP codes in the least deprived fifth by ADI national rank with IRS and SSA values | IRS tax year 2022 (amounts in thousands of dollars, converted); OASDI December 2024 benefits at an annual rate | 95% CI 41.0% to 42.4% | 5,847 ZIP codes | dollars (pooled over ZIP codes) | IRS Statistics of Income | Pensions here are private, government and military pensions and annuities reported on tax returns. The two sources cover different years and different people, so the share is a rough index. |
| pen_q5 | 28.0% | percent of the sum of taxable pensions and annuities and annualized OASDI benefits | pooled dollars of taxable pensions and annuities (IRS line for pensions and annuities in adjusted gross income) divided by those dollars plus 12 times the December 2024 monthly OASDI benefits, bootstrap interval over counties | ZIP codes in the most deprived fifth by ADI national rank with IRS and SSA values | IRS tax year 2022 (amounts in thousands of dollars, converted); OASDI December 2024 benefits at an annual rate | 95% CI 27.5% to 28.4% | 4,962 ZIP codes | dollars (pooled over ZIP codes) | IRS Statistics of Income | Pensions here are private, government and military pensions and annuities reported on tax returns. The two sources cover different years and different people, so the share is a rough index. |
| poor_mean | 79.5 | years, expected age at death for people who reach age 40 | mean of county values (average of the women's and men's estimates, adjusted to the national racial composition), weighted by county population (County Health Rankings 2024 release, a 2022 estimate), bootstrap interval over states | residents who reach age 40 in the counties of the Health Inequality Project file in the poorest quartile of national household income | 2001 to 2014 | 95% CI 79.1 to 79.9 | 1,559 counties | county residents (County Health Rankings 2024) | Health Inequality Project |   |
| poor_p5 | 77.5 | years, expected age at death for people who reach age 40 | 5th percentile of county values (average of the women's and men's estimates, adjusted to the national racial composition), weighted by county population | poorest quartile of national household income, all counties in the file | 2001 to 2014 |   | 1,559 counties | county residents (County Health Rankings 2024) | Health Inequality Project |   |
| poor_p95 | 81.8 | years, expected age at death for people who reach age 40 | 95th percentile of county values (average of the women's and men's estimates, adjusted to the national racial composition), weighted by county population | poorest quartile of national household income, all counties in the file | 2001 to 2014 |   | 1,559 counties | county residents (County Health Rankings 2024) | Health Inequality Project |   |
| poor_sd | 1.13 | years, standard deviation of county values of expected age at death at 40 with sampling variance removed | square root of the variance of county values (average of the women's and men's estimates, adjusted to the national racial composition) minus the mean squared bootstrap standard error, each county counted once, bootstrap interval over states | poorest quartile of national household income, all counties in the file | 2001 to 2014 | 95% CI 1.00 to 1.24 | 1,559 counties | none (each county counts once) | Health Inequality Project |   |
| poor_spread | 4.3 | years from the 5th to the 95th percentile of county expected age at death at 40 | width of the 5th-to-95th percentile range of county values (average of the women's and men's estimates, adjusted to the national racial composition), weighted by county population, bootstrap interval over states | poorest quartile of national household income, all counties in the file | 2001 to 2014 | 95% CI 3.7 to 4.9 | 1,559 counties | county residents (County Health Rankings 2024) | Health Inequality Project |   |
| poor_spread_agg | 6.4 | years from the 5th to the 95th percentile of county expected age at death at 40, not adjusted for race | width of the 5th-to-95th percentile range of county values (average of the women's and men's estimates, not race-adjusted), weighted by county population, bootstrap interval over states | poorest quartile of national household income, all counties in the file | 2001 to 2014 | 95% CI 4.7 to 6.7 | 1,559 counties | county residents (County Health Rankings 2024) | Health Inequality Project |   |
| poor_spread_big | 4.2 | years from the 5th to the 95th percentile of county expected age at death at 40 | width of the 5th-to-95th percentile range of county values (average of the women's and men's estimates, adjusted to the national racial composition), weighted by county population, bootstrap interval over states | poorest quartile of national household income, counties of at least 250,000 residents | 2001 to 2014 | 95% CI 3.5 to 5.0 | 283 counties | county residents (County Health Rankings 2024) | Health Inequality Project |   |
| poor_spread_unw | 4.0 | years from the 5th to the 95th percentile of county expected age at death at 40 | width of the 5th-to-95th percentile range of county values (average of the women's and men's estimates, adjusted to the national racial composition), each county counted once, bootstrap interval over states | poorest quartile of national household income, all counties in the file | 2001 to 2014 | 95% CI 3.7 to 4.4 | 1,559 counties | none (each county counts once) | Health Inequality Project |   |
| proc_diff | −$738 | dollars per fee-for-service beneficiary per year, most deprived fifth minus least deprived fifth, payments for procedures, tests, imaging and treatments | difference of two fee-for-service-beneficiary-weighted means of county payments per capita, bootstrap interval over states (shared draws) | fee-for-service beneficiaries in counties with every shown service line reported | 2024 | 95% CI −$971 to −$475 | 2,867 counties | fee-for-service beneficiaries | Medicare Geographic Variation |   |
| proc_q1 | $2,508 | dollars per fee-for-service beneficiary per year, price-standardized payments for procedures, tests, imaging and treatments | mean of county payments per capita, weighted by fee-for-service beneficiaries, bootstrap interval over states | fee-for-service beneficiaries in the least deprived fifth of counties by county ADI, counties with every shown service line reported | 2024 | 95% CI $2,275 to $2,723 | 605 counties | fee-for-service beneficiaries | Medicare Geographic Variation |   |
| proc_q5 | $1,770 | dollars per fee-for-service beneficiary per year, price-standardized payments for procedures, tests, imaging and treatments | mean of county payments per capita, weighted by fee-for-service beneficiaries, bootstrap interval over states | fee-for-service beneficiaries in the most deprived fifth of counties by county ADI, counties with every shown service line reported | 2024 | 95% CI $1,635 to $1,884 | 541 counties | fee-for-service beneficiaries | Medicare Geographic Variation |   |
| r_ben_inc | +0.84 | correlation coefficient | correlation of a ZIP code's average monthly retired-worker benefit with its median household income, weighted by ZIP-code population, bootstrap interval over counties | ZIP codes with at least 50 retired workers and a median household income | OASDI December 2024; ACS 2019-2023 | 95% CI 0.83 to 0.85 | 28,547 ZIP codes | ACS 2019-2023 ZCTA population | SSA beneficiaries by ZIP code |   |
| r_ben_inc_unw | +0.80 | correlation coefficient | correlation of a ZIP code's average monthly retired-worker benefit with its median household income, each ZIP code counted once, bootstrap interval over counties | ZIP codes with at least 50 retired workers and a median household income | OASDI December 2024; ACS 2019-2023 | 95% CI 0.79 to 0.82 | 28,547 ZIP codes | none (each ZIP code counts once) | SSA beneficiaries by ZIP code |   |
| r_dis_adi | +0.69 | correlation coefficient | adult-population-weighted correlation of tract disability prevalence with the ADI national rank, bootstrap interval over counties | adults aged 18 and over in tracts with a PLACES disability estimate, an ADI rank and an ACS age share | 2024 release of CDC PLACES (BRFSS responses from 2022); ADI 2023 | 95% CI 0.65 to 0.73 | 82,779 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| r_dis_age | +0.07 | correlation coefficient | adult-population-weighted correlation of tract disability prevalence with the share of residents aged 65 and over, bootstrap interval over counties | adults aged 18 and over in tracts with a PLACES disability estimate, an ADI rank and an ACS age share | 2024 release of CDC PLACES (BRFSS responses from 2022); ADI 2023 | 95% CI 0.04 to 0.10 | 82,779 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| r_dis_zip | +0.68 | correlation coefficient | correlation of a ZIP code's disabled-worker share of OASDI beneficiaries with its ADI national rank, weighted by the ZIP code area's population, bootstrap interval over counties | OASDI beneficiaries in ZIP codes in the 50 states and DC that match a ZCTA with an ADI rank, a population and a county | December 2024 | 95% CI 0.64 to 0.71 | 31,815 ZIP codes | ACS 2019-2023 ZCTA population | SSA beneficiaries by ZIP code |   |
| r_dis_zip_unw | +0.52 | correlation coefficient | correlation of a ZIP code's disabled-worker share of OASDI beneficiaries with its ADI national rank, with each ZIP code counted once, bootstrap interval over counties | OASDI beneficiaries in ZIP codes in the 50 states and DC that match a ZCTA with an ADI rank, a population and a county | December 2024 | 95% CI 0.50 to 0.54 | 31,815 ZIP codes | none (each ZIP code counts once) | SSA beneficiaries by ZIP code |   |
| ratio_90_10 | 1.42 | ratio of the 90th to the 10th percentile of county price-standardized payments per capita | beneficiary-weighted percentile ratio across counties, bootstrap interval over states | fee-for-service (Original Medicare) beneficiaries in counties with a standardized payment and an ADI rank | 2024 | 95% CI 1.34 to 1.49 | 3,128 counties | fee-for-service beneficiaries | Medicare Geographic Variation |   |
| ret_q1 | 80.5% | percent of OASDI beneficiaries who are retired workers | share of the fifth's beneficiaries, pooled over its ZIP codes, bootstrap interval over counties | OASDI beneficiaries in the least deprived fifth of ZIP codes by ADI national rank | December 2024 | 95% CI 79.9% to 81.1% | 6,363 ZIP codes | beneficiaries (each ZIP code counts by its beneficiary count) | SSA beneficiaries by ZIP code |   |
| ret_q5 | 65.9% | percent of OASDI beneficiaries who are retired workers | share of the fifth's beneficiaries, pooled over its ZIP codes, bootstrap interval over counties | OASDI beneficiaries in the most deprived fifth of ZIP codes by ADI national rank | December 2024 | 95% CI 65.3% to 66.4% | 6,363 ZIP codes | beneficiaries (each ZIP code counts by its beneficiary count) | SSA beneficiaries by ZIP code |   |
| rich_mean | 86.3 | years, expected age at death for people who reach age 40 | mean of county values (average of the women's and men's estimates, adjusted to the national racial composition), weighted by county population (County Health Rankings 2024 release, a 2022 estimate), bootstrap interval over states | residents who reach age 40 in the counties of the Health Inequality Project file in the richest quartile of national household income | 2001 to 2014 | 95% CI 86.1 to 86.4 | 1,559 counties | county residents (County Health Rankings 2024) | Health Inequality Project |   |
| rich_p5 | 84.7 | years, expected age at death for people who reach age 40 | 5th percentile of county values (average of the women's and men's estimates, adjusted to the national racial composition), weighted by county population | richest quartile of national household income, all counties in the file | 2001 to 2014 |   | 1,559 counties | county residents (County Health Rankings 2024) | Health Inequality Project |   |
| rich_p95 | 87.7 | years, expected age at death for people who reach age 40 | 95th percentile of county values (average of the women's and men's estimates, adjusted to the national racial composition), weighted by county population | richest quartile of national household income, all counties in the file | 2001 to 2014 |   | 1,559 counties | county residents (County Health Rankings 2024) | Health Inequality Project |   |
| rich_sd | 0.91 | years, standard deviation of county values of expected age at death at 40 with sampling variance removed | square root of the variance of county values (average of the women's and men's estimates, adjusted to the national racial composition) minus the mean squared bootstrap standard error, each county counted once, bootstrap interval over states | richest quartile of national household income, all counties in the file | 2001 to 2014 | 95% CI 0.77 to 1.02 | 1,559 counties | none (each county counts once) | Health Inequality Project |   |
| rich_spread | 3.0 | years from the 5th to the 95th percentile of county expected age at death at 40 | width of the 5th-to-95th percentile range of county values (average of the women's and men's estimates, adjusted to the national racial composition), weighted by county population, bootstrap interval over states | richest quartile of national household income, all counties in the file | 2001 to 2014 | 95% CI 2.7 to 3.6 | 1,559 counties | county residents (County Health Rankings 2024) | Health Inequality Project |   |
| rich_spread_agg | 3.0 | years from the 5th to the 95th percentile of county expected age at death at 40, not adjusted for race | width of the 5th-to-95th percentile range of county values (average of the women's and men's estimates, not race-adjusted), weighted by county population, bootstrap interval over states | richest quartile of national household income, all counties in the file | 2001 to 2014 | 95% CI 2.7 to 3.4 | 1,559 counties | county residents (County Health Rankings 2024) | Health Inequality Project |   |
| rich_spread_big | 2.6 | years from the 5th to the 95th percentile of county expected age at death at 40 | width of the 5th-to-95th percentile range of county values (average of the women's and men's estimates, adjusted to the national racial composition), weighted by county population, bootstrap interval over states | richest quartile of national household income, counties of at least 250,000 residents | 2001 to 2014 | 95% CI 2.3 to 3.4 | 283 counties | county residents (County Health Rankings 2024) | Health Inequality Project |   |
| rich_spread_unw | 4.3 | years from the 5th to the 95th percentile of county expected age at death at 40 | width of the 5th-to-95th percentile range of county values (average of the women's and men's estimates, adjusted to the national racial composition), each county counted once, bootstrap interval over states | richest quartile of national household income, all counties in the file | 2001 to 2014 | 95% CI 3.9 to 4.6 | 1,559 counties | none (each county counts once) | Health Inequality Project |   |
| sc_q1 | 7.5% | percent of OASDI beneficiaries who are spouses and children | share of the fifth's beneficiaries, pooled over its ZIP codes, bootstrap interval over counties | OASDI beneficiaries in the least deprived fifth of ZIP codes by ADI national rank | December 2024 | 95% CI 7.2% to 7.8% | 6,363 ZIP codes | beneficiaries (each ZIP code counts by its beneficiary count) | SSA beneficiaries by ZIP code |   |
| sc_q5 | 10.7% | percent of OASDI beneficiaries who are spouses and children | share of the fifth's beneficiaries, pooled over its ZIP codes, bootstrap interval over counties | OASDI beneficiaries in the most deprived fifth of ZIP codes by ADI national rank | December 2024 | 95% CI 10.4% to 10.9% | 6,363 ZIP codes | beneficiaries (each ZIP code counts by its beneficiary count) | SSA beneficiaries by ZIP code |   |
| sd_diff | 0.22 | years, standard deviation for the poorest quartile minus that for the richest, sampling variance removed | difference of the two standard deviations of county values (average of the women's and men's estimates, adjusted to the national racial composition) after removing the mean squared bootstrap standard error, each county counted once, bootstrap interval over states (shared draws) | both quartiles, all counties in the file | 2001 to 2014 | 95% CI 0.08 to 0.38 | 1,559 counties | none (each county counts once) | Health Inequality Project |   |
| se_poor | 0.5 | years, bootstrap standard error of a county's expected age at death | mean over counties of the Project's bootstrap standard errors (the women's and men's combined as independent), race-adjusted | poorest quartile of national household income, all counties in the file | 2001 to 2014 |   | 1,559 counties | none (each county counts once) | Health Inequality Project |   |
| se_rich | 0.9 | years, bootstrap standard error of a county's expected age at death | mean over counties of the Project's bootstrap standard errors (the women's and men's combined as independent), race-adjusted | richest quartile of national household income, all counties in the file | 2001 to 2014 |   | 1,559 counties | none (each county counts once) | Health Inequality Project |   |
| slope_adi | 2.0 | percentage points of adult disability prevalence per ten points of ADI national rank | adult-population-weighted least-squares slope of tract disability prevalence on the ADI rank, errors clustered on county | adults aged 18 and over in tracts with a PLACES disability estimate, an ADI rank and an ACS age share | 2024 release of CDC PLACES (BRFSS responses from 2022); ADI 2023 | 95% CI 1.8 to 2.1 | 82,779 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| slope_adi_ctrl | 2.0 | percentage points of adult disability prevalence per ten points of ADI national rank, holding the share aged 65 and over fixed | adult-population-weighted least-squares slope on the ADI rank in a regression that also includes the tract's share aged 65 and over, errors clustered on county | adults aged 18 and over in tracts with a PLACES disability estimate, an ADI rank and an ACS age share | 2024 release of CDC PLACES (BRFSS responses from 2022); ADI 2023 | 95% CI 1.8 to 2.1 | 82,779 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| small_ben_q1 | 3.1% | percent of the fifth's fee-for-service beneficiaries who live in counties below the size cutoff | beneficiaries in counties with fewer than the cutoff divided by all beneficiaries of the fifth | fee-for-service beneficiaries in the least deprived fifth of counties by county ADI | 2024 |   | 625 counties | fee-for-service beneficiaries | Medicare Geographic Variation |   |
| small_ben_q5 | 87.3% | percent of the fifth's fee-for-service beneficiaries who live in counties below the size cutoff | beneficiaries in counties with fewer than the cutoff divided by all beneficiaries of the fifth | fee-for-service beneficiaries in the most deprived fifth of counties by county ADI | 2024 |   | 624 counties | fee-for-service beneficiaries | Medicare Geographic Variation |   |
| small_cut | 5,000 | fee-for-service beneficiaries | county size below which a county is called small | counties in the size check of the spending comparison | 2024 |   |   | none (cutoff) | Medicare Geographic Variation |   |
| small_share_q5 | 97% | percent of counties in the most deprived fifth that are small | share of counties with fewer than the cutoff of fee-for-service beneficiaries | counties in the most deprived fifth by county ADI | 2024 |   | 624 counties | none (each county counts once) | Medicare Geographic Variation |   |
| snf_diff | $437 | dollars per fee-for-service beneficiary per year, most deprived fifth minus least deprived fifth, payments for skilled nursing facilities | difference of two fee-for-service-beneficiary-weighted means of county payments per capita, bootstrap interval over states (shared draws) | fee-for-service beneficiaries in counties with every shown service line reported | 2024 | 95% CI $214 to $653 | 2,867 counties | fee-for-service beneficiaries | Medicare Geographic Variation |   |
| snf_q1 | $865 | dollars per fee-for-service beneficiary per year, price-standardized payments for skilled nursing facilities | mean of county payments per capita, weighted by fee-for-service beneficiaries, bootstrap interval over states | fee-for-service beneficiaries in the least deprived fifth of counties by county ADI, counties with every shown service line reported | 2024 | 95% CI $712 to $986 | 605 counties | fee-for-service beneficiaries | Medicare Geographic Variation |   |
| snf_q5 | $1,302 | dollars per fee-for-service beneficiary per year, price-standardized payments for skilled nursing facilities | mean of county payments per capita, weighted by fee-for-service beneficiaries, bootstrap interval over states | fee-for-service beneficiaries in the most deprived fifth of counties by county ADI, counties with every shown service line reported | 2024 | 95% CI $1,122 to $1,462 | 541 counties | fee-for-service beneficiaries | Medicare Geographic Variation |   |
| spend_diff | $496 | dollars per fee-for-service beneficiary per year | mean in the most deprived fifth of counties minus mean in the least deprived fifth, weighted by fee-for-service beneficiaries, bootstrap interval over states (shared draws) | fee-for-service (Original Medicare) beneficiaries in counties with a standardized payment and an ADI rank | 2024 | 95% CI −$290 to $1406 | 3,128 counties | fee-for-service beneficiaries | Medicare Geographic Variation |   |
| spend_large_q1 | $12,736 | dollars per fee-for-service beneficiary per year, price-standardized | mean of county price-standardized payments per capita, weighted by fee-for-service beneficiaries, bootstrap interval over states | counties with at least 5,000 fee-for-service beneficiaries, least deprived fifth by county ADI | 2024 | 95% CI $12,046 to $13,250 | 444 counties | fee-for-service beneficiaries | Medicare Geographic Variation |   |
| spend_large_q5 | $12,359 | dollars per fee-for-service beneficiary per year, price-standardized | mean of county price-standardized payments per capita, weighted by fee-for-service beneficiaries, bootstrap interval over states | counties with at least 5,000 fee-for-service beneficiaries, most deprived fifth by county ADI | 2024 | 95% CI $11,830 to $12,916 | 21 counties | fee-for-service beneficiaries | Medicare Geographic Variation |   |
| spend_mean | $12,585 | dollars per fee-for-service beneficiary per year, price-standardized | mean of county price-standardized Part A and Part B payments per capita, weighted by fee-for-service beneficiaries, bootstrap interval over states | fee-for-service (Original Medicare) beneficiaries in counties with a standardized payment and an ADI rank | 2024 | 95% CI $12,125 to $12,987 | 3,128 counties | fee-for-service beneficiaries | Medicare Geographic Variation |   |
| spend_p10 | $10,433 | dollars per fee-for-service beneficiary per year, price-standardized | 10th percentile of county price-standardized payments per capita, weighted by fee-for-service beneficiaries | fee-for-service (Original Medicare) beneficiaries in counties with a standardized payment and an ADI rank | 2024 |   | 3,128 counties | fee-for-service beneficiaries | Medicare Geographic Variation |   |
| spend_p90 | $14,770 | dollars per fee-for-service beneficiary per year, price-standardized | 90th percentile of county price-standardized payments per capita, weighted by fee-for-service beneficiaries | fee-for-service (Original Medicare) beneficiaries in counties with a standardized payment and an ADI rank | 2024 |   | 3,128 counties | fee-for-service beneficiaries | Medicare Geographic Variation |   |
| spend_q1 | $12,680 | dollars per fee-for-service beneficiary per year, price-standardized | mean of county price-standardized payments per capita, weighted by fee-for-service beneficiaries, bootstrap interval over states (fifths hold equal numbers of counties) | fee-for-service beneficiaries in the least deprived fifth of counties by county ADI | 2024 | 95% CI $11,997 to $13,194 | 625 counties | fee-for-service beneficiaries | Medicare Geographic Variation |   |
| spend_q5 | $13,176 | dollars per fee-for-service beneficiary per year, price-standardized | mean of county price-standardized payments per capita, weighted by fee-for-service beneficiaries, bootstrap interval over states (fifths hold equal numbers of counties) | fee-for-service beneficiaries in the most deprived fifth of counties by county ADI | 2024 | 95% CI $12,517 to $13,785 | 624 counties | fee-for-service beneficiaries | Medicare Geographic Variation |   |
| spend_r | −0.04 | correlation coefficient | correlation of county price-standardized payments per capita with the county ADI, weighted by fee-for-service beneficiaries, bootstrap interval over states | fee-for-service (Original Medicare) beneficiaries in counties with a standardized payment and an ADI rank | 2024 | 95% CI −0.19 to 0.21 | 3,128 counties | fee-for-service beneficiaries | Medicare Geographic Variation |   |
| spend_r_unw | +0.30 | correlation coefficient | correlation of county price-standardized payments per capita with the county ADI, each county counted once, bootstrap interval over states | fee-for-service (Original Medicare) beneficiaries in counties with a standardized payment and an ADI rank | 2024 | 95% CI 0.20 to 0.39 | 3,128 counties | none (each county counts once) | Medicare Geographic Variation |   |
| spend_small_q1 | $10,893 | dollars per fee-for-service beneficiary per year, price-standardized | mean of county price-standardized payments per capita, weighted by fee-for-service beneficiaries, bootstrap interval over states | counties with fewer than 5,000 fee-for-service beneficiaries, least deprived fifth by county ADI | 2024 | 95% CI $10,526 to $11,270 | 181 counties | fee-for-service beneficiaries | Medicare Geographic Variation |   |
| spend_small_q5 | $13,295 | dollars per fee-for-service beneficiary per year, price-standardized | mean of county price-standardized payments per capita, weighted by fee-for-service beneficiaries, bootstrap interval over states | counties with fewer than 5,000 fee-for-service beneficiaries, most deprived fifth by county ADI | 2024 | 95% CI $12,601 to $13,903 | 603 counties | fee-for-service beneficiaries | Medicare Geographic Variation |   |
| spread_diff | 1.3 | years, width for the poorest quartile minus width for the richest | difference of the two 5th-to-95th percentile widths (average of the women's and men's estimates, adjusted to the national racial composition), weighted by county population, bootstrap interval over states (shared draws) | both quartiles, all counties in the file | 2001 to 2014 | 95% CI 0.3 to 1.9 | 1,559 counties | county residents (County Health Rankings 2024) | Health Inequality Project |   |
| spread_diff_big | 1.6 | years, width for the poorest quartile minus width for the richest | difference of the two 5th-to-95th percentile widths (average of the women's and men's estimates, adjusted to the national racial composition), weighted by county population, bootstrap interval over states (shared draws) | both quartiles, counties of at least 250,000 residents | 2001 to 2014 | 95% CI 0.7 to 2.5 | 283 counties | county residents (County Health Rankings 2024) | Health Inequality Project |   |
| spread_diff_unw | −0.3 | years, width for the poorest quartile minus width for the richest | difference of the two 5th-to-95th percentile widths (average of the women's and men's estimates, adjusted to the national racial composition), each county counted once, bootstrap interval over states (shared draws) | both quartiles, all counties in the file | 2001 to 2014 | 95% CI −0.7 to 0.2 | 1,559 counties | none (each county counts once) | Health Inequality Project |   |
| teeth_d1 | 6.9% | percent of adults aged 65 and over who have lost all their teeth | mean of tract prevalence (crude, modeled), weighted by the tract's estimated residents aged 65 and over, bootstrap interval over counties | adults aged 65 and over in the least deprived tenth of tracts by ADI national rank | 2024 release of CDC PLACES (BRFSS responses from 2022); ADI 2023 | 95% CI 6.2% to 7.9% | 8,258 tracts | residents aged 65 and over (ACS 2019-2023 share times ACS population) | CDC PLACES | PLACES models these values from survey responses and census inputs that overlap the ADI, so part of the gradient is built in. |
| teeth_d10 | 25.1% | percent of adults aged 65 and over who have lost all their teeth | mean of tract prevalence (crude, modeled), weighted by the tract's estimated residents aged 65 and over, bootstrap interval over counties | adults aged 65 and over in the most deprived tenth of tracts by ADI national rank | 2024 release of CDC PLACES (BRFSS responses from 2022); ADI 2023 | 95% CI 24.7% to 25.6% | 8,213 tracts | residents aged 65 and over (ACS 2019-2023 share times ACS population) | CDC PLACES | PLACES models these values from survey responses and census inputs that overlap the ADI, so part of the gradient is built in. |
| teeth_ratio | 3.6 | ratio of tooth loss in the most deprived tenth of tracts to that in the least deprived tenth | ratio of two weighted means, bootstrap interval over counties (shared draws) | adults aged 65 and over in tracts with a PLACES tooth-loss estimate, an ADI rank and an ACS age share | 2024 release of CDC PLACES (BRFSS responses from 2022); ADI 2023 | 95% CI 3.2 to 4.0 | 82,662 tracts | residents aged 65 and over (ACS 2019-2023 share times ACS population) | CDC PLACES |   |
| us_disability | 30.9% | percent of adults with any disability | mean of tract prevalence (crude, modeled) over all the tracts of the tenths pooled, weighted by adult population, bootstrap interval over counties (the draws of the tenths) | adults aged 18 and over in all the tracts of the comparison: tracts with a PLACES disability estimate, an ADI rank and an ACS age share | 2024 release of CDC PLACES (BRFSS responses from 2022); ADI 2023 | 95% CI 30.5% to 31.3% | 82,779 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| us_ed_visits | 592 | emergency department visits per 1,000 fee-for-service beneficiaries per year | mean of county rates over all the counties of the fifths pooled, weighted by fee-for-service beneficiaries, bootstrap interval over states (the draws of the fifths) | fee-for-service beneficiaries in all the counties of the comparison: counties with a standardized payment and an ADI rank | 2024 | 95% CI 581 to 602 | 3,128 counties | fee-for-service beneficiaries | Medicare Geographic Variation |   |
| wid_q1 | 4.8% | percent of OASDI beneficiaries who are widows, widowers and parents | share of the fifth's beneficiaries, pooled over its ZIP codes, bootstrap interval over counties | OASDI beneficiaries in the least deprived fifth of ZIP codes by ADI national rank | December 2024 | 95% CI 4.7% to 4.9% | 6,363 ZIP codes | beneficiaries (each ZIP code counts by its beneficiary count) | SSA beneficiaries by ZIP code |   |
| wid_q5 | 6.6% | percent of OASDI beneficiaries who are widows, widowers and parents | share of the fifth's beneficiaries, pooled over its ZIP codes, bootstrap interval over counties | OASDI beneficiaries in the most deprived fifth of ZIP codes by ADI national rank | December 2024 | 95% CI 6.5% to 6.8% | 6,363 ZIP codes | beneficiaries (each ZIP code counts by its beneficiary count) | SSA beneficiaries by ZIP code |   |

## Definitions

- Area Deprivation Index (ADI): A score that ranks neighborhoods by income, education, employment and housing conditions drawn from the American Community Survey. In this book a higher national percentile means a more deprived neighborhood.
- Modeled estimate: A figure produced by a statistical model instead of a direct count or measurement, usually for small areas where direct data are thin, borrowing strength from surveys, neighboring areas or other data. Neighboring areas can share errors, so single extremes call for caution.
- OASDI: Old-Age, Survivors and Disability Insurance, the Social Security programs that pay retirement, survivor and disability benefits.
- ZCTA: ZIP Code Tabulation Area, the Census Bureau's approximation of a ZIP code as a set of census blocks. It lets census data be reported by ZIP code, though some ZIP codes have no matching ZCTA.
- Price-standardized spending: Medicare spending adjusted to remove differences in local prices and payment rules, so that counties are compared on the care paid for rather than on local prices. It does not adjust for how sick the beneficiaries are.
- Life expectancy at birth: The average number of years a newborn would live if death rates at every age stayed as they were in the period measured. It summarizes the death rates of a period and is not a forecast for any child.
- Ecological inference: Drawing conclusions about individuals from data about places or groups. A pattern across counties need not hold for the people inside them.

## Methods

This chapter compares tracts, ZIP codes and counties across tenths or fifths of the Area Deprivation Index (2023 national rank) on disability, tooth loss, Social Security beneficiaries and benefits, pensions, Medicare payments and expected age at death at 40. The sources differ in kind and period: PLACES prevalence is modeled with census and ACS inputs from BRFSS 2022 responses, the Social Security counts are for December 2024, pensions are for tax year 2022, and Medicare payments for 2024 are for Original Medicare beneficiaries with both parts. ZIP codes with no ZCTA are left out, and intervals come from a bootstrap over whole counties or states.

### How disability and tooth loss are measured

Tract disability and tooth loss are CDC PLACES crude prevalence among adults (2024 release, BRFSS 2022 responses, modeled with census and ACS inputs), weighted by adult population (disability) and by estimated residents aged 65 and over (tooth loss).

Explains Table 1.

### How places are ranked and grouped by deprivation

Tenths and fifths hold equal numbers of tracts, ZIP codes or counties, ranked on the Area Deprivation Index national rank (2023). A county's rank is the mean of its tracts' ranks weighted by ACS population.

### How Social Security and pension data are matched to ZIP codes

Social Security counts and benefits are SSA's December 2024 ZIP code table for the 50 states and DC: ZIP codes that cross a state line are summed, benefit totals in thousands of dollars are converted to dollars, and a ZIP code is matched to the ZCTA of the same number in the project's ZCTA atlas (ZIP codes with no ZCTA are left out; Connecticut's are matched like the rest).

Pensions are taxable pensions and annuities from IRS Statistics of Income for tax year 2022 (thousands of dollars, converted), set against 12 times the December 2024 monthly benefits.

Explains Table 2 and Table 3.

### What the Medicare payments cover

Medicare payments are CMS price-standardized Part A and Part B payments per capita for Original Medicare beneficiaries with both parts, 2024; the file has one row per county, and Connecticut appears by planning region.

Counties with a blank service line are left out of the stacked bars and the last segment is the total minus the lines shown.

Explains Table 4.

### How expected age at death is measured and weighted

Expected age at death at 40 is the Health Inequality Project's county estimate by national household-income quartile; the chapter uses the race-adjusted column, averaged over women and men, and weights counties by their residents in the County Health Rankings 2024 release (a Census Bureau estimate for 2022).

Explains Table 5, Table 6 and Table 7.

### How intervals, ranges and plotted samples are computed

Intervals are 95% percentile intervals from 2,000 draws of whole clusters with replacement (seed 12345): counties for tract and ZIP code estimates and states for county estimates, computed by the same routine for means, shares, ratios, differences, correlations and slopes; the regressions use cluster-robust errors on county.

The width of the range of county values is the distance from the 5th to the 95th weighted percentile.

Scatter plots show random samples (tracts, ZIP codes) inside the plotted range with the same seed, or every county.

### Four rankings of the same places

The comparison follows the book's ranking switch rule, fixed on October 28, 2026 before any of these gradients was computed. The main gradient of the chapter is recomputed under four rankings of the same places, with the outcome, period, weights, bins and intervals unchanged: the published ADI; a standardized index that applies the ADI's coefficients to its measures after standardizing each one across tracts; median household income; and the share of adults 25 and older with a bachelor's degree or more. The three alternatives come from ACS tract estimates for 2019 to 2023, the period of the ADI.

All four rankings use the same places, the 75,596 tracts that have the outcome and all four rankings, so the ADI panel can differ slightly from the chapter's main figure. Tracts enter only if they have every input of the standardized index, which leaves out tracts without a median rent, home value or mortgage cost.

A pattern counts as changed when the worse end switches sides, when the two ends differ under one ranking and not the other, or when the rank correlation of bin order and bin mean falls on the other side of 0.9 and moves by at least 0.1.

Explains Table 8.

## What these data cannot show

Every pattern here describes tracts, ZIP codes or counties, and a pattern across places need not hold for the people inside them (ecological inference). A ZIP code with a high disabled-worker share says nothing about whether any resident should receive a disability benefit, and a county's expected age at death for the poorest quartile is an estimate for a group of people who share an income rank, not a forecast for one of them.

The measures differ in vintage and in how they are made. PLACES values are modeled from 2022 survey responses, and their link with the ADI is partly built in. The Social Security counts are for December 2024 and are rounded by SSA, and the ZIP code is matched to the ZIP code area of the same number, which leaves out ZIP codes such as post office boxes. The Medicare payments are for 2024, cover only fee-for-service beneficiaries, and so leave out the Medicare Advantage enrollees, who are 54.5% of beneficiaries with both parts in these counties. They are not adjusted for the health of the beneficiaries, so equal payments do not mean equal care or equal need. The life-expectancy estimates describe deaths from 2001 to 2014 and end before the later years of this book, and household income is measured on a national scale that does not adjust for the cost of living where a household lives.

The correlates are measured years after the deaths, and smoking and obesity are County Health Rankings' modeled estimates for all adults, not for the poorest quartile. The Project's authors found that the share of immigrants was among the strongest correlates of low-income longevity across local areas, a pattern that tract and county measures of this kind cannot separate from place itself (see Chapter 9.5, Immigrants and the gradient (https://vishalsingh.org/zipcode/immigrants-and-the-gradient; planned)). Weights matter for several results here. The Medicare correlations with deprivation and with life expectancy and the spread of the richest quartile's expected age at death all change when each county counts once, and both versions are reported.

None of the gradients tells what would move a number. The causal evidence above concerns eligibility rules, examiner decisions and a benefit notch, and the places in this chapter differ in many ways that those studies do not address.

## References

1. Kind, Amy J. H.; Buckingham, William R. (2018). Making Neighborhood-Disadvantage Metrics Accessible: The Neighborhood Atlas. New England Journal of Medicine 378(26): 2456-2458. https://doi.org/10.1056/NEJMp1802313
2. Greenlund, Kurt J.; Lu, Hua; Wang, Yan; Matthews, Kevin A.; LeClercq, Jennifer M.; Lee, Benjamin; Carlson, Susan A. (2022). PLACES: Local Data for Better Health. Preventing Chronic Disease 19: 210459. https://doi.org/10.5888/pcd19.210459
3. Centers for Medicare & Medicaid Services (2026). Medicare Data for the Original Medicare Geographic Variation Public Use File: A Methodological Overview (May 2026 update). CMS Office of Enterprise Data and Analytics. https://data.cms.gov/sites/default/files/2026-05/Original%20Medicare%20Geographic%20Variation%20Public%20Use%20File%20Methods%20Paper.pdf
4. Fisher, Elliott S.; Wennberg, David E.; Stukel, Therese A.; Gottlieb, Daniel J.; Lucas, F. L.; Pinder, Etoile L. (2003). The Implications of Regional Variations in Medicare Spending. Part 2: Health Outcomes and Satisfaction with Care. Annals of Internal Medicine 138(4): 288-298. https://doi.org/10.7326/0003-4819-138-4-200302180-00007
5. Chetty, Raj; Stepner, Michael; Abraham, Sarah; Lin, Shelby; Scuderi, Benjamin; Turner, Nicholas; Bergeron, Augustin; Cutler, David (2016). The Association Between Income and Life Expectancy in the United States, 2001-2014. JAMA 315(16): 1750-1766. https://doi.org/10.1001/jama.2016.4226
6. Card, David; Dobkin, Carlos; Maestas, Nicole (2009). Does Medicare Save Lives?. The Quarterly Journal of Economics 124(2): 597-636. https://doi.org/10.1162/qjec.2009.124.2.597
7. Maestas, Nicole; Mullen, Kathleen J.; Strand, Alexander (2013). Does Disability Insurance Receipt Discourage Work? Using Examiner Assignment to Estimate Causal Effects of SSDI Receipt. American Economic Review 103(5): 1797-1829. https://doi.org/10.1257/aer.103.5.1797
8. Deshpande, Manasi; Gross, Tal; Su, Yalun (2021). Disability and Distress: The Effect of Disability Programs on Financial Outcomes. American Economic Journal: Applied Economics 13(2): 151-178. https://doi.org/10.1257/app.20190709
9. Snyder, Stephen E.; Evans, William N. (2006). The Effect of Income on Mortality: Evidence from the Social Security Notch. The Review of Economics and Statistics 88(3): 482-495. https://doi.org/10.1162/rest.88.3.482

## Provenance

- Pipeline version: 0.1.0
- Build script: book2/chapters/disability-retirement-and-medicare/build.py (sha256 448318b8effa37f391716e2671a8ac83f461cfe2f31c3b3cdb74a7f1b658a33f)
- Built on: 2026-10-07
- Site commit: c2b1ff3309cb4a10833f884da91ac0057a66ad0a

Inputs:

| File | Size (bytes) | SHA-256, first 1 MB | SHA-256, last 1 MB |
| --- | ---: | --- | --- |
| book2/chapters/disability-retirement-and-medicare/disab.py | 6,033 | 00439b80f701784359cee1fe2044ca72f1eb4006baf7f11aa9f605aa89bece63 | 00439b80f701784359cee1fe2044ca72f1eb4006baf7f11aa9f605aa89bece63 |
| book2/chapters/disability-retirement-and-medicare/hip.py | 11,372 | 9d83106b301ea2bc54e11155a9f2d28a2d71cb78e1ae18831786da97f846be70 | 9d83106b301ea2bc54e11155a9f2d28a2d71cb78e1ae18831786da97f846be70 |
| book2/chapters/disability-retirement-and-medicare/medi.py | 7,713 | c8c8e45d1d90f6738eab633a2d8073dbcdb0345313f059ae8d19d30a7c332189 | c8c8e45d1d90f6738eab633a2d8073dbcdb0345313f059ae8d19d30a7c332189 |
| book2/chapters/disability-retirement-and-medicare/oasdi.py | 11,254 | 820b615b7769a4de0ba9b5e043228e87d0ea02172e3b87e52eb58a29cc0f8807 | 820b615b7769a4de0ba9b5e043228e87d0ea02172e3b87e52eb58a29cc0f8807 |
| data/analytic/tract_atlas.parquet | 39,659,925 | 29d5e4d943552d54aeb6ba92aadc939d9bd15bdea553c78357dbd516757be627 | 1e10d1ecf722340f23ee2212d4999a90250f85b65fa89b39e32db06d9472e22e |
| data/analytic/zcta_atlas.parquet | 674,348,737 | 0abfad485b74b74eb956fdc57c88e504abe10d601a2e99a3f16467469826578c | 0e37a4ce00d8569d49dccd0b2dfdc2c8435934f9c8a606bbca2ca1dad2c9effe |
| data/raw/acs/deprivation_components_2019_2023/tract.parquet | 13,197,793 | 5545cde2cbe96a8b56e468d65388248a241aa6bd300d3c9e325d08d3f86b11da | 5c8ec454d066988b0f1a01aaa6ed34ebfc05d162e01cf9c6bb01f9198cbaff12 |
| data/staged/adi/adi_atlas_tract2020_2023.parquet | 798,078 | 5e409bc212ff871def780fa27b4cf8b616d9da420b2b9f6455517eddcd8c1a95 | 5e409bc212ff871def780fa27b4cf8b616d9da420b2b9f6455517eddcd8c1a95 |
| data/staged/crosswalks/ct_tract2020_to_old_county.parquet | 15,842 | 961ef39f31596ef65fca2c67d5f2f63e533c5dc9b49745d3bb6437844e4b831d | 961ef39f31596ef65fca2c67d5f2f63e533c5dc9b49745d3bb6437844e4b831d |
| data/staged/health/chr_county_2024.parquet | 8,776,933 | 619795f7ecfec8aff10694b649eaee5e421dbeffbe4be47af4161b51ceefa5cd | c69357f1bcf630407bb9716cb445618838999558c019aaaf77d4681b6bf45c56 |
| data/staged/health/places_tract2020_latest.parquet | 5,919,159 | af4c3b0d1205b7d84b81676a9529ef73323b2dbeb9d54e235f5efeebfad83177 | a2f19029ee2ef032be87e0d5288d88d489afd52379279833f178baa021df3dbf |
| data/staged/income/irs_soi_zip_2022.parquet | 29,261,256 | ae6a84e1c187f7b0e1085cc5a2e7048aa3e3ccc4e50030b4ce8e5db11ce09375 | 2cf093d54788fa8198767866bd9291043066ff9338449c0a9bdfdf93b8189482 |
| data/staged/life_expectancy/health_ineq_county_le_income.parquet | 601,631 | 43cd39b714e143eefe5a3eba108864435ec5d2606f896be1323fc62764bff55a | 43cd39b714e143eefe5a3eba108864435ec5d2606f896be1323fc62764bff55a |
| data/staged/medicare/cms_geo_variation_county_2024.parquet | 3,771,529 | 6fbd23b41d69de745227bdc34cd831c523294c6142eb67ff4466e04cd63d398a | 04a6064b1322641cf997c02ccc80f58fcca8f359d5147d6f11026ccdc22ce8c4 |
| data/staged/ses/acs5_tract2020_2019_2023.parquet | 7,232,424 | 5fe6f730a35225f14d5d8abfc6b9bce58b52e545f309a7eac74f1dae74d7a31c | 7b4a2475b3852d3f91804716a04566ef8ceadacec2be9fb85f5eb642078efac6 |
| data/staged/social_security/ssa_oasdi_zip_2024.parquet | 844,914 | 9609b3bd5d1838d1c918db5c3519d6fb0d7df3793c1b62b6b70d0b0a4feb9375 | 9609b3bd5d1838d1c918db5c3519d6fb0d7df3793c1b62b6b70d0b0a4feb9375 |
| data/staged/svi/svi_tract2020_2022.parquet | 17,740,415 | cb3860d8edc2aaf1f2a02211a58f7a0b80457eb6ad9a37b5ec8abb2d5da9da10 | e371a4a696d5bed05d54ad7c5b6ca2504508805b847ec415c2322be7f66ec2b2 |

Outputs:

| File | SHA-256 |
| --- | --- |
| facts.json | 9ff34bbcae9bdf00191de17cf396fec86f8485d0c6e8ca7921d5ae0c4b04fbf6 |
| figures/alt_rankings.json | 2f174623d60889d3db6dc07afecd828379c703363700dd058f5e9f54c0852012 |
| figures/benefit_vs_income.json | c3ae37a7cd33c2790a14646974d4ec95b0e203b203490fc81825067b34df94bc |
| figures/disability_gradient.json | 3275b35c2b470ec413dc94c8610de3a0433d0c045c74557c1d39e7cfbc820406 |
| figures/medicare_spending.json | 2b213e89a79fdd6668c7cd45010381cad8f5536df3e7a1175ffbff6437cf9597 |
| figures/oasdi_mix.json | 284dedf467972c99fca829dc2c230f82f6e4572a98781a3cd9c769b612ef4ea8 |
| figures/poor_le_correlates.json | 5a229bd2369a84f7a63461f87d0b1aa4c509d00c87d2e85892e68995e6807448 |
| figures/poor_le_map.json | 6cc362ada14fbbe49908982708cd2b6d332ce73954d8eab1c4aef4b895a03e08 |
| figures/poor_vs_rich_le_at_40.json | c8831dcba7eba9a66412914bc287f223bc45c67f0bc7071917116f7d79b11792 |
| meta.json | 82938135b4a0616d19b01d7d80b96405aa990154aa8238c14c73c7dd1c517ee4 |
| public/data/disability-retirement-and-medicare/benefit_vs_income.json | 08a2ca13dc889b122c14e3eb94cc17ee02becd0f445af52796d411c48bf87127 |
| public/data/disability-retirement-and-medicare/disability_by_age.json | f1eb842f96495215dad71700b36ca35a61a81b9b488af430a9544f3cb7448948 |
| public/data/disability-retirement-and-medicare/poor_le_map.json | 21ba92bdb1a41dc88d2bcad68f8eb5bd4a9934000c5e9f2dca61605796882faf |
| public/data/disability-retirement-and-medicare/poor_vs_rich_le_at_40.json | a0bab32a076197cf2e54e3de3a1be10213001700e23f1f12af1c3bf49e36b96b |
| public/data/disability-retirement-and-medicare/spend_vs_life_expectancy.json | d10d2bd3f517299b673834b2bf16dbbe712209e03d542449a70f7d211c9532a3 |
| refs.json | 79173fe5e0203880ecf85d35f7defa871b3823208a115150c822b6de2be49030 |

## Machine formats

- Report (HTML): https://vishalsingh.org/zipcode/disability-retirement-and-medicare/report
- Report (Markdown): https://vishalsingh.org/zipcode/disability-retirement-and-medicare/report.md
- Every estimate (JSON): https://vishalsingh.org/zipcode/report/disability-retirement-and-medicare/facts.json
- Table 1, figure "disability_gradient" (JSON): https://vishalsingh.org/zipcode/report/disability-retirement-and-medicare/figures/disability_gradient.json
- Table 2, figure "oasdi_mix" (JSON): https://vishalsingh.org/zipcode/report/disability-retirement-and-medicare/figures/oasdi_mix.json
- Table 3, figure "benefit_vs_income" (JSON): https://vishalsingh.org/zipcode/report/disability-retirement-and-medicare/figures/benefit_vs_income.json
- Table 4, figure "medicare_spending" (JSON): https://vishalsingh.org/zipcode/report/disability-retirement-and-medicare/figures/medicare_spending.json
- Table 5, figure "poor_le_map" (JSON): https://vishalsingh.org/zipcode/report/disability-retirement-and-medicare/figures/poor_le_map.json
- Table 6, figure "poor_vs_rich_le_at_40" (JSON): https://vishalsingh.org/zipcode/report/disability-retirement-and-medicare/figures/poor_vs_rich_le_at_40.json
- Table 7, figure "poor_le_correlates" (JSON): https://vishalsingh.org/zipcode/report/disability-retirement-and-medicare/figures/poor_le_correlates.json
- Table 8, figure "alt_rankings" (JSON): https://vishalsingh.org/zipcode/report/disability-retirement-and-medicare/figures/alt_rankings.json
- Payload of "disability_gradient" (JSON): https://vishalsingh.org/zipcode/data/disability-retirement-and-medicare/disability_by_age.json
- Payload of "benefit_vs_income" (JSON): https://vishalsingh.org/zipcode/data/disability-retirement-and-medicare/benefit_vs_income.json
- Payload of "medicare_spending" (JSON): https://vishalsingh.org/zipcode/data/disability-retirement-and-medicare/spend_vs_life_expectancy.json
- Payload of "poor_le_map" (JSON): https://vishalsingh.org/zipcode/data/disability-retirement-and-medicare/poor_le_map.json
- Payload of "poor_vs_rich_le_at_40" (JSON): https://vishalsingh.org/zipcode/data/disability-retirement-and-medicare/poor_vs_rich_le_at_40.json
- Index of every report: https://vishalsingh.org/zipcode/llms.txt; all reports in one file: https://vishalsingh.org/zipcode/llms-full.txt

---

# Gun homicide and the causes of early death: report

Chapter 8.1 of ZIP Code Destiny, by Vishal Singh (NYU Stern). Generated from the chapter's data files; the reading version is at https://vishalsingh.org/zipcode/gun-homicide-and-causes-of-early-death. Table n of this report is figure n of the chapter.

## Identification

- Title: Gun homicide and the causes of early death
- Label: The sliver
- Chapter: 8.1 (Part VIII, Early death; An American life)
- Question: How concentrated are gun homicides in a few neighborhoods, and which causes account for early death at each level of deprivation?
- Dek: Outside New York City, the tract clusters with the highest firearm homicide rates hold 50% of the deaths that CDC's rates imply for 2022 to 2024 and 9.5% of residents.
- Grain: neighborhoods
- Status: published
- Published: 2026-10-09
- Updated: not revised
- Data sources:
  - CDC injury and violence mapping. CDC firearm homicide rates for tract clusters, 2022 to 2024 (modeled estimates for 1 to 9 deaths, zero for none, direct rates for 10 or more), and the county file, 2021 to 2024; years: 2019-2024 for counties, 2022-2024 for tracts (data as of 2026-05-14); weights: ACS 2019-2023 residents of each cluster
  - American Community Survey. Total population of 2020 tracts, 2019-2023; years: 2019-2023 (5-year); weights: none (full count)
  - USALEEP life expectancy. Life expectancy at birth by tract, deaths in 2010 to 2015, 2020 tracts; years: 2010-2015; weights: 2020 tract population
  - Area Deprivation Index. Area Deprivation Index national rank, 2023, tracts rolled up to counties; years: 2023 release (version 4.0.1; built on ACS 2019-2023); weights: 2019-2023 tract population
  - County Health Rankings. County death rates for premature mortality (age-adjusted), suicide (crude and age-adjusted), overdose, homicide, motor-vehicle crashes and firearms (crude), 2024 release, with race-specific homicide and suicide rates and rural share; years: 2024 release; weights: county population
- Canonical URL: https://vishalsingh.org/zipcode/gun-homicide-and-causes-of-early-death
- Report: https://vishalsingh.org/zipcode/gun-homicide-and-causes-of-early-death/report
- Markdown: https://vishalsingh.org/zipcode/gun-homicide-and-causes-of-early-death/report.md
- BibTeX key: singh2026gunhomicideandcausesofearlydeath

```bibtex
@misc{singh2026gunhomicideandcausesofearlydeath,
  author = {Singh, Vishal},
  title = {Gun homicide and the causes of early death},
  year = {2026},
  url = {https://vishalsingh.org/zipcode/gun-homicide-and-causes-of-early-death},
  note = {In ZIP Code Destiny. Published 9 October 2026.}
}
```

## Principal estimates

The facts the dek and the opening prose cite, in order of appearance.

| Estimand | Population | Period | Value | Interval | n | Weights | Source | Fact id |
| --- | --- | --- | ---: | --- | ---: | --- | --- | --- |
| threshold that defines the concentration statistic (definitional, not an estimate) | all deaths implied by the cluster rates | annual rates for 2022, 2023 and 2024, averaged | 50% |   |   | none (definition) | CDC injury and violence mapping | thr_half |
| share of residents in the highest-rate clusters that together hold half of implied firearm homicide deaths; clusters ranked by mean annual rate; county bootstrap interval | residents of tract clusters outside New York City with a published rate in each of 2022, 2023 and 2024 | annual rates for 2022, 2023 and 2024, averaged | 9.5% | 95% CI 8.3% to 10.8% | 17,814 tract clusters | ACS 2019-2023 residents of each tract cluster (rate times residents gives implied deaths) | CDC injury and violence mapping | nat_half |
| share of residents in clusters with zero recorded deaths in each of the three years (a rate of 0 from a count of 0); county bootstrap interval | residents of tract clusters outside New York City with a published rate in each of 2022, 2023 and 2024 | annual rates for 2022, 2023 and 2024, averaged | 30.7% | 95% CI 29.0% to 32.6% | 17,814 tract clusters | ACS 2019-2023 residents of each tract cluster (rate times residents gives implied deaths) | CDC injury and violence mapping | nat_zero |

## Findings

The tract clusters with the highest firearm homicide rates hold 50% of the deaths that CDC's rates imply for 2022 to 2024 outside New York City, and 9.5% (95% CI 8.3% to 10.8%) of residents. At the other end of the ranking, 30.7% (95% CI 29.0% to 32.6%) of residents live in clusters with no recorded firearm homicide in any year from 2022 to 2024. A rate for a state, a county or a city averages over both ends, and the neighborhood is where they separate.

This chapter follows that concentration down to Cook County, the county that holds Chicago and its suburbs, and then widens to the other kinds of early death. It changes grain once. The first four figures describe tract clusters, the unit in which CDC publishes neighborhood death rates. The last four describe counties, the finest grain at which County Health Rankings reports deaths from overdose, suicide, homicide and motor-vehicle crashes. Neither grain describes individuals.

### Firearm homicide deaths concentrate in a small share of tract clusters

CDC's injury and violence dashboard publishes death rates for census tracts, but not for single tracts. It combines small neighboring tracts into clusters large enough to report a stable rate and repeats the cluster's rate on every member tract. The median tract belongs to a cluster of 5 tracts, and the largest cluster has 20. A cluster-year with a few deaths gets a modeled estimate with an interval, one with none gets a rate of zero, one with many gets a direct rate, and CDC withholds the rate when the model is unstable. The unit in this section is the cluster, and each of the 17,814 clusters in the national curve is counted once.

To draw the curve, rank every cluster from its highest to its lowest mean annual rate for 2022 to 2024, multiply each rate by the cluster's residents to get implied deaths, and accumulate residents and deaths down the ranking. Equal shares would follow the diagonal. The curve in Table 1 bends above it. The clusters that hold the highest-rate tenth of residents hold 51.3% (95% CI 48.1% to 54.4%) of implied deaths. Reaching 50% of deaths takes 9.5% (95% CI 8.3% to 10.8%) of residents, and reaching 75% takes 24.1% (95% CI 22.2% to 26.0%).

Table 1. The tract clusters with the highest firearm homicide rates hold 50% of implied deaths and 9.5% of residents outside New York City

The interval around each share comes from resampling whole counties, so it describes which places are in the file. It leaves out the modeling error in CDC's rates. Redrawing every modeled rate from CDC's own interval, a separate check, shifts the estimate upward. The share of residents holding 50% of deaths comes out between 9.6% and 9.8%, above the estimate of 9.5% and inside the resampling interval.

Studies of crime incidents on street segments in several U.S. cities find that a small share of segments accounts for a large share of incidents, a regularity Weisburd calls the law of crime concentration.[1] Averaging over a cluster can only lower measured concentration, so the curve in Table 1 is a lower bound for what block-level rates would show if CDC published them.

Two groups of residents are not on the curve. New York City is left out because CDC's file records almost no deaths there, and the clusters where CDC withholds the rate in at least one year from 2022 to 2024 hold 4.7% of the remaining residents. The data for 2024 are provisional, and CDC withholds more rates for that year (455 clusters, against 312 in 2022).

The curve covers 308.7 million residents. Filling the withheld years from the count range CDC publishes for them puts the share of residents who hold 50% of deaths between 8.9% and 10.2%, so the result does not rest on the exclusion.

The file also undercounts deaths outside New York City. With the same residents on both sides, the deaths implied by the cluster rates come to 78.6% of what the county file implies for the same counties, so the curve describes how deaths are distributed and not how many there are.

The shortfall is uneven. In Essex County, New Jersey, and Suffolk County, Massachusetts, the clusters imply 7.3% and 8.7% of the county file's deaths, and in San Diego and Santa Clara counties, California, 27.6% and 28.1%. These are the counties where the clusters imply under 30% of a county-file toll of at least 15 deaths a year.

Dropping their clusters moves the share holding 50% of deaths from 9.5% to 9.6%.

### Counties differ in how steeply the curve bends

The same curve can be drawn inside one county. Table 2 shows it for the 8 counties with the most implied firearm homicide deaths a year. In Cook County, 50% of deaths sit with 10.5% (95% CI 9.1% to 12.2%) of residents. The share is larger in every other county shown, and in Shelby County, Tennessee, around Memphis, it reaches 24.4% (95% CI 20.4% to 29.3%). Harris County, Texas, around Houston, comes close, and its interval overlaps Shelby's.

Table 2. In Cook County the clusters holding 50% of firearm homicide deaths hold 10.5% of residents, against 24.4% in the county with the largest share

Each curve rests on the clusters with a rate in every year from 2022 to 2024, which is at least 80.2% of residents in every county shown and 87.7% in Cook County. A county line is not a neighborhood boundary. A county that includes a low-rate suburban ring holds more residents at the low end of the curve than a county whose line follows the city limits, so a comparison of the curves compares county definitions as well as places.

### Chicago, cluster by cluster

Table 3 maps the clusters of Cook County, with each tract carrying its cluster's rate. Of the 1,163 tracts with a rate in every year from 2022 to 2024, 24.2% sit in clusters with no recorded firearm homicide, and the highest cluster rate is 89 deaths per hundred thousand residents a year. Of the residents in the highest-rate tenth of clusters, 90.8% live in tracts south of the Loop, judged by tract centroid, and the tracts well north of the Loop sit at the low end of the rate range. The remaining 168 tracts are gray because CDC withholds the rate in at least one year or because no one lives there.

Table 3. In Cook County, 24.2% of tracts with an estimate sit in clusters with no recorded firearm homicide, while the highest cluster rate reaches 89 per hundred thousand residents

A cluster's rate averages over several tracts and over the years 2022 to 2024, so the highest figure on the map is a floor for the rate on the worst blocks and not a ceiling. The map describes clusters, and nothing on it describes a person who lives in one.

### The highest-rate tenths of Cook County are also its most deprived and shortest-lived

Rank Cook County residents by the firearm homicide rate of their cluster and cut the ranking into tenths (Table 4). Clusters with no recorded deaths hold 27.0% (95% CI 22.1% to 32.2%) of the 87.7% of residents who live in clusters with a rate in every year from 2022 to 2024, so the lowest two tenths and part of the third all have a rate of zero and are ordered at random.

The highest tenth lives in clusters with a mean rate of 60 (95% CI 56 to 65) deaths per hundred thousand residents a year and a mean Area Deprivation Index national rank of 74 (95% CI 71 to 78), on a scale where the highest rank is the most deprived. Life expectancy at birth for deaths in 2010 to 2015 is 71.7 (95% CI 71.0 to 72.4) years in that tenth and 81.2 (95% CI 80.7 to 81.7) years in the clusters with no recorded deaths, a gap of 9.5 (95% CI 8.6 to 10.4) years within one county. The ninth tenth sits with the tenth: its life expectancy and its deprivation rank both lie beyond the intervals of the other eight tenths.

Table 4. In Cook County, residents of the highest-rate tenth of clusters have a life expectancy of 71.7 years (deaths in 2010 to 2015), 9.5 fewer than residents of clusters with no recorded deaths

The life expectancy estimates come from USALEEP and cover deaths in 2010 to 2015, a decade before the firearm homicide rates. The comparison shows that the same places have high homicide rates and short lives. It does not say how much of the gap firearm homicide accounts for, and it describes places, not people. The tenths are formed on the rate itself, so the first tab shows how far apart the groups were sorted, and the other two tabs carry the information. The gradient of life expectancy across deprivation deciles for the whole country is in Chapter 8.4, Place and the length of a life (https://vishalsingh.org/zipcode/place-and-the-length-of-a-life), and Chapter 8.5, Violence and the life-expectancy gap (https://vishalsingh.org/zipcode/violence-and-the-life-expectancy-gap) estimates how much of the national gap homicide accounts for.

### Premature mortality rises with county deprivation

Firearm homicide is one of several ways to die young, and the others follow deprivation as well. County Health Rankings reports age-adjusted deaths of residents before age 75, pooled over 2019 to 2021 from National Center for Health Statistics records. Ranking counties by their Area Deprivation Index, the least deprived tenth of counties has 307 (95% CI 296 to 318) premature deaths per hundred thousand residents and the most deprived tenth has 701 (95% CI 660 to 740). Across 3,083 counties, the population-weighted correlation between county deprivation and premature mortality is +0.78 (95% CI 0.74 to 0.83).

Table 5. Premature mortality rises from 307 to 701 per hundred thousand residents across deciles of county deprivation

County Health Rankings is a different source from the CDC panel, at a coarser grain. Its rates are counts of deaths pooled over several years, and a county with fewer than 10 deaths from a cause in the window gets no rate (20 for premature mortality). A missing rate means too few deaths to report, not a rate of zero. Chapter 10.3, Suppressed counties in death data (https://vishalsingh.org/zipcode/suppressed-counties-in-death-data) covers what that does to death data.

### The mix of deaths differs between the least and most deprived fifths

Take four kinds of death: drug overdose, suicide, homicide and motor-vehicle crashes. Adding their crude rates, which are deaths per hundred thousand residents with no age adjustment, gives a toll for a group of counties, and each kind's share of that toll is its place in the mix. The four overlap a little, because a suicide by drug poisoning is counted as both an overdose and a suicide, so the toll includes such a death in both of those shares. The 1,248 counties that publish all four rates hold 89.6% of county residents, and Table 6 sorts them into fifths of residents by county deprivation.

The toll is larger in the most deprived fifth than in the least deprived, 76 (95% CI 69 to 84) deaths per hundred thousand residents against 42 (95% CI 37 to 47), and the mix differs between those end fifths. Overdose is 51.0% (95% CI 46.9% to 57.6%) of the toll in the least deprived fifth and 43.6% (95% CI 38.4% to 48.2%) in the most deprived, a change of −7.4 percentage points (95% CI −15.5 percentage points to −1.3 percentage points).

Homicide rises from 8.8% (95% CI 5.9% to 10.2%) to 13.1% (95% CI 11.7% to 14.7%) of the toll (+4.3 percentage points (95% CI 2.2 percentage points to 7.5 percentage points)). Motor-vehicle crashes rise from 16.0% (95% CI 12.6% to 17.9%) to 21.4% (95% CI 18.4% to 24.6%) (+5.4 percentage points (95% CI 1.8 percentage points to 10.5 percentage points)).

Suicide holds a similar share in the end fifths, 24.2% (95% CI 21.6% to 29.5%) and 21.9% (95% CI 20.1% to 23.9%). The figure also shows the fifths in between, and the overdose and motor-vehicle shares do not follow a steady line across all five, so it supports a comparison of the two ends and no more.

Table 6. Overdose is 51.0% of the four-cause toll in the least deprived fifth of residents and 43.6% in the most, while motor-vehicle crashes go from 16.0% to 21.4%

The rates cover different windows: 2019 to 2021 for overdose, 2017 to 2021 for suicide, and 2015 to 2021 for homicide and motor-vehicle crashes. The mix therefore compares rates measured over different spans, and counties with few deaths have noisy rates. Overdose enters this chapter only in this mix, in the map of the highest-rate kind of death and in a cross-check against the CDC panel. The overdose maps and gradients are in Chapter 8.2, Overdose, alcohol and suicide (https://vishalsingh.org/zipcode/overdose-alcohol-and-suicide), and the overdose trend since 2018 is in Chapter 8.3, Mortality trends, 2018 to 2024 (https://vishalsingh.org/zipcode/mortality-trends-2018-2024).

### Rural share moves with suicide and motor-vehicle deaths, and homicide depends on the weighting

Table 7 correlates seven County Health Rankings measures with county deprivation and with the rural share of residents. The first two views weight counties by population and the last two count every county once. Weighted by population, all seven correlations with deprivation are positive. The largest are for premature mortality, child mortality and motor-vehicle crashes, and the interval excludes zero for every measure except the share of crash deaths that involve alcohol. The homicide rate has a correlation of +0.41 (95% CI 0.34 to 0.48) with county deprivation, and +0.49 (95% CI 0.42 to 0.55) when every county counts once.

Table 7. Homicide and other causes of early death are higher in more deprived counties, with a population-weighted correlation of +0.41 for homicide

On the rural-share views, suicide and motor-vehicle death rates are higher where more residents live outside urban areas. The population-weighted correlations are +0.53 (95% CI 0.45 to 0.60) and +0.66 (95% CI 0.58 to 0.72), and the unweighted ones are +0.37 (95% CI 0.32 to 0.44) and +0.60 (95% CI 0.57 to 0.64). County deprivation and rural share are themselves correlated (+0.56 (95% CI 0.48 to 0.63)), so the deprivation and rural-share views overlap.

The homicide estimate depends on the weighting, −0.05 (95% CI −0.18 to 0.05) weighted by population and +0.25 (95% CI 0.12 to 0.35) counting every county once. Which counties enter matters for homicide. Only the 1,376 counties with enough homicides to publish a rate are in, and a county with few homicides gets none, so these counties are less rural than counties as a whole, with a mean rural share of 43.3% against 64.1%. Among the counties that report, the homicide rate is near zero against rural share when large counties dominate and modestly positive when every county counts equally. Neither estimate describes all counties, because the counties missing from the sample are the ones with the fewest homicides.

### Homicide and suicide rates differ by race in opposite directions

County Health Rankings also publishes rates for Black and white residents, neither group including anyone who identifies as Hispanic, in counties where a group has enough deaths to report. These are county rates averaged over the counties that report, and they are not national rates by race. County Health Rankings publishes the homicide rate as a crude rate and the suicide rate as an age-adjusted rate, so the homicide ratio below also carries any difference in age structure between the two groups. In the 486 counties that report both groups, the homicide rate is 27.3 (95% CI 23.8 to 31.8) deaths per hundred thousand Black residents and 2.8 (95% CI 2.5 to 3.2) per hundred thousand white residents, a ratio of 9.6 (95% CI 8.1 to 11.6).

The suicide rate runs the other way, 16.4 (95% CI 14.8 to 18.0) for white residents and 7.8 (95% CI 7.1 to 8.5) for Black residents in the 297 counties that report both, a ratio of 2.1 (95% CI 1.9 to 2.3). A county reports a group's rate only when the group has enough deaths, so the reporting counties are larger and less rural than the rest, and each comparison is between groups in the same counties. The ratio describes those counties. It leaves out Black residents of counties with too few deaths to report, and it says nothing about the risk to any one person.

### The kind of death with the highest rate differs by county

Table 8 labels each county by whichever of the four kinds of death has the highest crude rate there. Among the 1,248 counties that publish all four rates, overdose is highest in 829, suicide in 207, motor-vehicle crashes in 201 and homicide in 11. The gray counties lack at least one rate and are mostly small, so the map speaks for the labeled counties, which hold 89.6% of county residents, and not for the land they sit on.

Table 8. Overdose has the highest rate of the four causes in 829 of 1,248 counties, and homicide in only 11

The counties where homicide is highest are Dallas and Montgomery in Alabama; Crittenden, Jefferson and Mississippi in Arkansas; Bibb and Muscogee in Georgia; Caddo Parish in Louisiana; and Adams, Hinds and Yazoo in Mississippi. Non-Hispanic Black residents are a median 56.1% of residents in these counties and 7.1% across the labeled counties, which fits the racial gap in homicide rates above. The comparison describes county populations and says nothing about any individual.

County Health Rankings and the CDC injury panel both draw on the nation's death certificates, so their agreement checks the processing and not the underlying counts. Population-weighted correlations between the two county rates, with the panel's modeled estimate for 2021, run from +0.87 (95% CI 0.83 to 0.90) for suicide to +0.95 (95% CI 0.94 to 0.96) for drug overdose, which measures agreement between the two sources and not a gradient. Counting every county once, the suicide correlation falls to +0.64 (95% CI 0.58 to 0.71), so the agreement is lower when small counties count equally.

### What place-based evidence shows

Concentration makes it possible to aim programs at the places where violence is highest, and evaluations of such programs differ in design. A meta-analysis of quasi-experimental evaluations of focused deterrence, which pairs a warning to the people most involved in violence with offers of services, finds a moderate average reduction in crime in treated areas, with smaller estimates in evaluations that use stronger designs.[2] The evaluation of CeaseFire in Chicago compared most program sites with matched comparison areas and found shooting hot spots shrinking at most sites, with the decrease linked to the program at some of them.[3] Both designs compare treated areas with areas that were not assigned at random. A systematic review of Cure Violence evaluations, written by staff of Cure Violence Global who note that this presents a bias, reports reductions in most findings but statistical significance in a minority, with results that differ by site.[4] The box below reports a trial in which assignment was random.

> Causal evidence: A randomized test of remediating vacant lots (Cluster randomized controlled trial)
>
> In Philadelphia, Branas and colleagues randomly assigned vacant lots, grouped into clusters of neighboring lots, to remediation (cleaning, grading, planting grass and trees, and low fencing), to trash cleanup only, or to no treatment. Because assignment was random, differences in outcomes between treated and untreated clusters can be attributed to the treatment and not to which neighborhoods were treated. Police-reported gun assaults fell around treated lots compared with untreated lots, with the clearest reductions in neighborhoods below the poverty line, and residents near treated lots reported feeling safer.[5] The reduction across all neighborhoods was smaller.
>
> The trial tested a low-cost change to the built environment and not a violence prevention program, in one city. It does not show that the same change would lower gun violence elsewhere, or that any program aimed at concentrated violence works.

### Premature mortality rises with county deprivation under all four rankings

Over the 3,080 counties that have a premature mortality rate and all four rankings, the ADI baseline runs from 307 (95% CI 296 to 318) premature deaths per hundred thousand residents in the least deprived tenth of counties to 698 (95% CI 655 to 741) in the most deprived, with a rank correlation of +1.00 between tenth and rate (Table 9). The standardized index, which applies the ADI's coefficients after standardizing each measure, gives 280 (95% CI 270 to 291) and 650 (95% CI 574 to 723), with a correlation of +1.00. Ranked by median household income the tenths run from 291 (95% CI 284 to 300) to 680 (95% CI 647 to 709), and ranked by the bachelor's degree share from 297 (95% CI 282 to 315) to 627 (95% CI 597 to 656). Under each ranking the two ends differ and the rate rises steadily toward the deprived end, so none of the three alternatives changes the gradient under the book's rule (Chapter 9.1, How far one ranking goes (https://vishalsingh.org/zipcode/how-far-one-ranking-goes)).

Table 9. The premature death rate is steadily worse toward the more deprived end under the ADI and under each of the three other rankings

### What these data cannot show

The firearm homicide figures describe tract clusters, and CDC's rates for them are modeled estimates, so they say nothing about single tracts, blocks or people. The curves and the Cook County map average over clusters of neighboring tracts, which lowers measured concentration, and the cluster file's implied deaths fall short of the county file's, so the figures describe shares and not totals. New York City is absent, four more counties have clusters that imply a small share of the county file's deaths, and the clusters where CDC withholds a rate are left out, with more withheld for the provisional year 2024. The fill check bounds the last exclusion and cannot remove it.

Every comparison across places is ecological. The life expectancy of Cook County's highest-rate tenth comes from deaths in 2010 to 2015 and the rates from 2022 to 2024, so the data cannot say whether firearm homicide shortened these lives or whether the same conditions underlie both. County gradients and correlations describe counties, and the homicide and rural-share correlation differs with the weighting and depends on which counties report, so a correlation across counties says little about the people in them. The Area Deprivation Index understates deprivation in high-cost cities, and it is built from survey measures that overlap with several outcome measures.

County Health Rankings leaves out counties with few deaths, so the mix of deaths and the map of the highest-rate kind speak for the larger counties. The rates in those two figures are crude, so a county with an older population has a higher rate for the same risk, and the correlation figure mixes crude and age-adjusted measures, which its row labels state. The race-specific rates are a composite of reporting counties and not national rates, and the cross-check against the injury panel compares two compilations of the same death records. Outside the box above, nothing in this chapter supports a claim about what changes violence.

## Tables

### Table 1. The tract clusters with the highest firearm homicide rates hold 50% of implied deaths and 9.5% of residents outside New York City

Cumulative share of implied firearm homicide deaths against cumulative share of residents, tract clusters ranked from the highest to the lowest mean annual rate, 2022 to 2024, outside New York City

| conf_high | conf_low | pct_residents | series | share_deaths |
| ---: | ---: | ---: | --- | ---: |
| n/a | n/a | 0 | Firearm homicide deaths | 0 |
| 15.02 | 12.36 | 1 | Firearm homicide deaths | 13.93 |
| 23.60 | 19.23 | 2 | Firearm homicide deaths | 21.70 |
| 29.75 | 24.60 | 3 | Firearm homicide deaths | 27.45 |
| 34.76 | 29.14 | 4 | Firearm homicide deaths | 32.20 |
| 39.03 | 33.12 | 5 | Firearm homicide deaths | 36.28 |
| 42.74 | 36.67 | 6 | Firearm homicide deaths | 39.89 |
| 46.09 | 39.89 | 7 | Firearm homicide deaths | 43.16 |
| 49.12 | 42.80 | 8 | Firearm homicide deaths | 46.10 |
| 51.87 | 45.55 | 9 | Firearm homicide deaths | 48.79 |
| 54.39 | 48.10 | 10 | Firearm homicide deaths | 51.31 |
| 56.71 | 50.48 | 11 | Firearm homicide deaths | 53.66 |
| 58.84 | 52.73 | 12 | Firearm homicide deaths | 55.87 |
| 60.89 | 54.85 | 13 | Firearm homicide deaths | 57.94 |
| 62.78 | 56.87 | 14 | Firearm homicide deaths | 59.89 |
| 64.60 | 58.77 | 15 | Firearm homicide deaths | 61.74 |
| 66.28 | 60.63 | 16 | Firearm homicide deaths | 63.49 |
| 67.85 | 62.36 | 17 | Firearm homicide deaths | 65.15 |
| 69.35 | 64.01 | 18 | Firearm homicide deaths | 66.73 |
| 70.79 | 65.59 | 19 | Firearm homicide deaths | 68.24 |
| 72.19 | 67.12 | 20 | Firearm homicide deaths | 69.68 |
| 73.50 | 68.57 | 21 | Firearm homicide deaths | 71.06 |
| 74.74 | 69.96 | 22 | Firearm homicide deaths | 72.38 |
| 75.92 | 71.31 | 23 | Firearm homicide deaths | 73.65 |
| 77.04 | 72.59 | 24 | Firearm homicide deaths | 74.85 |
| 78.11 | 73.83 | 25 | Firearm homicide deaths | 76.01 |
| 79.15 | 75.01 | 26 | Firearm homicide deaths | 77.12 |
| 80.16 | 76.16 | 27 | Firearm homicide deaths | 78.18 |
| 81.11 | 77.24 | 28 | Firearm homicide deaths | 79.21 |
| 82.04 | 78.29 | 29 | Firearm homicide deaths | 80.20 |
| 82.91 | 79.30 | 30 | Firearm homicide deaths | 81.16 |
| 83.78 | 80.28 | 31 | Firearm homicide deaths | 82.07 |
| 84.59 | 81.22 | 32 | Firearm homicide deaths | 82.96 |
| 85.42 | 82.13 | 33 | Firearm homicide deaths | 83.82 |
| 86.17 | 83.01 | 34 | Firearm homicide deaths | 84.65 |
| 86.92 | 83.87 | 35 | Firearm homicide deaths | 85.45 |
| 87.64 | 84.70 | 36 | Firearm homicide deaths | 86.22 |
| 88.31 | 85.50 | 37 | Firearm homicide deaths | 86.96 |
| 88.97 | 86.27 | 38 | Firearm homicide deaths | 87.68 |
| 89.60 | 87 | 39 | Firearm homicide deaths | 88.36 |
| 90.22 | 87.72 | 40 | Firearm homicide deaths | 89.02 |
| 90.81 | 88.41 | 41 | Firearm homicide deaths | 89.66 |
| 91.36 | 89.07 | 42 | Firearm homicide deaths | 90.27 |
| 91.90 | 89.71 | 43 | Firearm homicide deaths | 90.85 |
| 92.41 | 90.32 | 44 | Firearm homicide deaths | 91.41 |
| 92.91 | 90.90 | 45 | Firearm homicide deaths | 91.94 |
| 93.38 | 91.45 | 46 | Firearm homicide deaths | 92.45 |
| 93.84 | 91.99 | 47 | Firearm homicide deaths | 92.95 |
| 94.28 | 92.50 | 48 | Firearm homicide deaths | 93.42 |
| 94.71 | 93 | 49 | Firearm homicide deaths | 93.88 |
| 95.12 | 93.47 | 50 | Firearm homicide deaths | 94.32 |
| 95.52 | 93.93 | 51 | Firearm homicide deaths | 94.74 |
| 95.89 | 94.37 | 52 | Firearm homicide deaths | 95.15 |
| 96.26 | 94.79 | 53 | Firearm homicide deaths | 95.55 |
| 96.61 | 95.21 | 54 | Firearm homicide deaths | 95.93 |
| 96.95 | 95.60 | 55 | Firearm homicide deaths | 96.30 |
| 97.28 | 95.98 | 56 | Firearm homicide deaths | 96.66 |
| 97.59 | 96.36 | 57 | Firearm homicide deaths | 97 |
| 97.90 | 96.72 | 58 | Firearm homicide deaths | 97.33 |
| 98.19 | 97.06 | 59 | Firearm homicide deaths | 97.65 |
| 98.47 | 97.40 | 60 | Firearm homicide deaths | 97.96 |
| 98.75 | 97.71 | 61 | Firearm homicide deaths | 98.25 |
| 99 | 98.02 | 62 | Firearm homicide deaths | 98.52 |
| 99.23 | 98.32 | 63 | Firearm homicide deaths | 98.79 |
| 99.44 | 98.59 | 64 | Firearm homicide deaths | 99.04 |
| 99.64 | 98.85 | 65 | Firearm homicide deaths | 99.27 |
| 99.80 | 99.10 | 66 | Firearm homicide deaths | 99.48 |
| 99.95 | 99.32 | 67 | Firearm homicide deaths | 99.67 |
| 100 | 99.53 | 68 | Firearm homicide deaths | 99.84 |
| 100 | 99.72 | 69 | Firearm homicide deaths | 99.97 |
| 100 | 99.88 | 70 | Firearm homicide deaths | 100 |
| n/a | n/a | 71 | Firearm homicide deaths | 100 |
| n/a | n/a | 72 | Firearm homicide deaths | 100 |
| n/a | n/a | 73 | Firearm homicide deaths | 100 |
| n/a | n/a | 74 | Firearm homicide deaths | 100 |
| n/a | n/a | 75 | Firearm homicide deaths | 100 |
| n/a | n/a | 76 | Firearm homicide deaths | 100 |
| n/a | n/a | 77 | Firearm homicide deaths | 100 |
| n/a | n/a | 78 | Firearm homicide deaths | 100 |
| n/a | n/a | 79 | Firearm homicide deaths | 100 |
| n/a | n/a | 80 | Firearm homicide deaths | 100 |
| n/a | n/a | 81 | Firearm homicide deaths | 100 |
| n/a | n/a | 82 | Firearm homicide deaths | 100 |
| n/a | n/a | 83 | Firearm homicide deaths | 100 |
| n/a | n/a | 84 | Firearm homicide deaths | 100 |
| n/a | n/a | 85 | Firearm homicide deaths | 100 |
| n/a | n/a | 86 | Firearm homicide deaths | 100 |
| n/a | n/a | 87 | Firearm homicide deaths | 100 |
| n/a | n/a | 88 | Firearm homicide deaths | 100 |
| n/a | n/a | 89 | Firearm homicide deaths | 100 |
| n/a | n/a | 90 | Firearm homicide deaths | 100 |
| n/a | n/a | 91 | Firearm homicide deaths | 100 |
| n/a | n/a | 92 | Firearm homicide deaths | 100 |
| n/a | n/a | 93 | Firearm homicide deaths | 100 |
| n/a | n/a | 94 | Firearm homicide deaths | 100 |
| n/a | n/a | 95 | Firearm homicide deaths | 100 |
| n/a | n/a | 96 | Firearm homicide deaths | 100 |
| n/a | n/a | 97 | Firearm homicide deaths | 100 |
| n/a | n/a | 98 | Firearm homicide deaths | 100 |
| n/a | n/a | 99 | Firearm homicide deaths | 100 |
| n/a | n/a | 100 | Firearm homicide deaths | 100 |
| n/a | n/a | 0 | Equal shares | 0 |
| n/a | n/a | 1 | Equal shares | 1 |
| n/a | n/a | 2 | Equal shares | 2 |
| n/a | n/a | 3 | Equal shares | 3 |
| n/a | n/a | 4 | Equal shares | 4 |
| n/a | n/a | 5 | Equal shares | 5 |
| n/a | n/a | 6 | Equal shares | 6 |
| n/a | n/a | 7 | Equal shares | 7 |
| n/a | n/a | 8 | Equal shares | 8 |
| n/a | n/a | 9 | Equal shares | 9 |
| n/a | n/a | 10 | Equal shares | 10 |
| n/a | n/a | 11 | Equal shares | 11 |
| n/a | n/a | 12 | Equal shares | 12 |
| n/a | n/a | 13 | Equal shares | 13 |
| n/a | n/a | 14 | Equal shares | 14 |
| n/a | n/a | 15 | Equal shares | 15 |
| n/a | n/a | 16 | Equal shares | 16 |
| n/a | n/a | 17 | Equal shares | 17 |
| n/a | n/a | 18 | Equal shares | 18 |
| n/a | n/a | 19 | Equal shares | 19 |
| n/a | n/a | 20 | Equal shares | 20 |
| n/a | n/a | 21 | Equal shares | 21 |
| n/a | n/a | 22 | Equal shares | 22 |
| n/a | n/a | 23 | Equal shares | 23 |
| n/a | n/a | 24 | Equal shares | 24 |
| n/a | n/a | 25 | Equal shares | 25 |
| n/a | n/a | 26 | Equal shares | 26 |
| n/a | n/a | 27 | Equal shares | 27 |
| n/a | n/a | 28 | Equal shares | 28 |
| n/a | n/a | 29 | Equal shares | 29 |
| n/a | n/a | 30 | Equal shares | 30 |
| n/a | n/a | 31 | Equal shares | 31 |
| n/a | n/a | 32 | Equal shares | 32 |
| n/a | n/a | 33 | Equal shares | 33 |
| n/a | n/a | 34 | Equal shares | 34 |
| n/a | n/a | 35 | Equal shares | 35 |
| n/a | n/a | 36 | Equal shares | 36 |
| n/a | n/a | 37 | Equal shares | 37 |
| n/a | n/a | 38 | Equal shares | 38 |
| n/a | n/a | 39 | Equal shares | 39 |
| n/a | n/a | 40 | Equal shares | 40 |
| n/a | n/a | 41 | Equal shares | 41 |
| n/a | n/a | 42 | Equal shares | 42 |
| n/a | n/a | 43 | Equal shares | 43 |
| n/a | n/a | 44 | Equal shares | 44 |
| n/a | n/a | 45 | Equal shares | 45 |
| n/a | n/a | 46 | Equal shares | 46 |
| n/a | n/a | 47 | Equal shares | 47 |
| n/a | n/a | 48 | Equal shares | 48 |
| n/a | n/a | 49 | Equal shares | 49 |
| n/a | n/a | 50 | Equal shares | 50 |
| n/a | n/a | 51 | Equal shares | 51 |
| n/a | n/a | 52 | Equal shares | 52 |
| n/a | n/a | 53 | Equal shares | 53 |
| n/a | n/a | 54 | Equal shares | 54 |
| n/a | n/a | 55 | Equal shares | 55 |
| n/a | n/a | 56 | Equal shares | 56 |
| n/a | n/a | 57 | Equal shares | 57 |
| n/a | n/a | 58 | Equal shares | 58 |
| n/a | n/a | 59 | Equal shares | 59 |
| n/a | n/a | 60 | Equal shares | 60 |
| n/a | n/a | 61 | Equal shares | 61 |
| n/a | n/a | 62 | Equal shares | 62 |
| n/a | n/a | 63 | Equal shares | 63 |
| n/a | n/a | 64 | Equal shares | 64 |
| n/a | n/a | 65 | Equal shares | 65 |
| n/a | n/a | 66 | Equal shares | 66 |
| n/a | n/a | 67 | Equal shares | 67 |
| n/a | n/a | 68 | Equal shares | 68 |
| n/a | n/a | 69 | Equal shares | 69 |
| n/a | n/a | 70 | Equal shares | 70 |
| n/a | n/a | 71 | Equal shares | 71 |
| n/a | n/a | 72 | Equal shares | 72 |
| n/a | n/a | 73 | Equal shares | 73 |
| n/a | n/a | 74 | Equal shares | 74 |
| n/a | n/a | 75 | Equal shares | 75 |
| n/a | n/a | 76 | Equal shares | 76 |
| n/a | n/a | 77 | Equal shares | 77 |
| n/a | n/a | 78 | Equal shares | 78 |
| n/a | n/a | 79 | Equal shares | 79 |
| n/a | n/a | 80 | Equal shares | 80 |
| n/a | n/a | 81 | Equal shares | 81 |
| n/a | n/a | 82 | Equal shares | 82 |
| n/a | n/a | 83 | Equal shares | 83 |
| n/a | n/a | 84 | Equal shares | 84 |
| n/a | n/a | 85 | Equal shares | 85 |
| n/a | n/a | 86 | Equal shares | 86 |
| n/a | n/a | 87 | Equal shares | 87 |
| n/a | n/a | 88 | Equal shares | 88 |
| n/a | n/a | 89 | Equal shares | 89 |
| n/a | n/a | 90 | Equal shares | 90 |
| n/a | n/a | 91 | Equal shares | 91 |
| n/a | n/a | 92 | Equal shares | 92 |
| n/a | n/a | 93 | Equal shares | 93 |
| n/a | n/a | 94 | Equal shares | 94 |
| n/a | n/a | 95 | Equal shares | 95 |
| n/a | n/a | 96 | Equal shares | 96 |
| n/a | n/a | 97 | Equal shares | 97 |
| n/a | n/a | 98 | Equal shares | 98 |
| n/a | n/a | 99 | Equal shares | 99 |
| n/a | n/a | 100 | Equal shares | 100 |

Note: A cluster is a group of neighboring tracts for which CDC publishes one rate, repeated on each member tract; the median tract belongs to a cluster of 5 tracts and the largest cluster has 20. The curve counts each of the 17,814 clusters once, so its unit is the cluster, not the tract.

Note: Left out: New York City, where CDC records almost no deaths in this file, and the clusters where CDC withholds the rate in at least one year from 2022 to 2024 (4.7% of residents outside New York City). The data for 2024 are provisional, and CDC withholds more rates for that year (455 clusters, against 312 in 2022). Filling the withheld years from the published count range gives between 8.9% and 10.2% of residents holding 50% of deaths.

Note: Deaths are each cluster's annual rate times its residents, averaged over the three years. With the same residents on both sides they come to 78.6% of the deaths the county file implies for the same counties, so the curve describes shares and not totals. The shortfall is uneven: the clusters imply 7.3% of the county file's deaths in Essex County, New Jersey, 8.7% in Suffolk County, Massachusetts, 27.6% in San Diego County, California, and 28.1% in Santa Clara County, California, the counties where they imply under 30% of a county-file toll of at least 15 deaths a year. Dropping the clusters of those four counties moves the share holding 50% of deaths to 9.6%.

Note: Intervals are 95% intervals from a bootstrap that draws whole counties with replacement (replicates and seed are in the methods note). They describe which places are in the file and do not include the modeling error in CDC's rates. Puerto Rico is absent from the tract file.

Source: CDC Mapping Injury, Overdose, and Violence, tract file, firearm homicide, 2022-2024; modeled estimates for tract clusters; ACS 2019-2023 population

Data: https://vishalsingh.org/zipcode/report/gun-homicide-and-causes-of-early-death/figures/national_concentration.json

### Table 2. In Cook County the clusters holding 50% of firearm homicide deaths hold 10.5% of residents, against 24.4% in the county with the largest share

Cumulative share of implied firearm homicide deaths against cumulative share of residents, tract clusters ranked from the highest rate, the counties with the most implied deaths, 2022 to 2024

| conf_high | conf_low | county | pct_residents | share_deaths |
| ---: | ---: | --- | ---: | ---: |
| n/a | n/a | Cook County, IL | 0 | 0 |
| 8.22 | 5.76 | Cook County, IL | 1 | 6.98 |
| 15.40 | 10.93 | Cook County, IL | 2 | 13.07 |
| 21.47 | 15.61 | Cook County, IL | 3 | 18.45 |
| 26.99 | 19.96 | Cook County, IL | 4 | 23.18 |
| 32.16 | 24.05 | Cook County, IL | 5 | 27.79 |
| 36.99 | 28.03 | Cook County, IL | 6 | 32.23 |
| 41.37 | 31.92 | Cook County, IL | 7 | 36.51 |
| 45.56 | 35.67 | Cook County, IL | 8 | 40.57 |
| 49.53 | 39.28 | Cook County, IL | 9 | 44.45 |
| 52.93 | 42.85 | Cook County, IL | 10 | 48.15 |
| 56.01 | 46.28 | Cook County, IL | 11 | 51.48 |
| 58.85 | 49.41 | Cook County, IL | 12 | 54.47 |
| 61.40 | 52.16 | Cook County, IL | 13 | 57.12 |
| 63.77 | 54.80 | Cook County, IL | 14 | 59.59 |
| 65.94 | 57.09 | Cook County, IL | 15 | 61.75 |
| 67.92 | 59.10 | Cook County, IL | 16 | 63.71 |
| 69.69 | 61.04 | Cook County, IL | 17 | 65.47 |
| 71.37 | 62.79 | Cook County, IL | 18 | 67.16 |
| 72.95 | 64.43 | Cook County, IL | 19 | 68.76 |
| 74.46 | 66.01 | Cook County, IL | 20 | 70.28 |
| 75.92 | 67.53 | Cook County, IL | 21 | 71.76 |
| 77.36 | 69.03 | Cook County, IL | 22 | 73.23 |
| 78.77 | 70.47 | Cook County, IL | 23 | 74.65 |
| 80.10 | 71.84 | Cook County, IL | 24 | 76.04 |
| 81.34 | 73.21 | Cook County, IL | 25 | 77.37 |
| 82.50 | 74.57 | Cook County, IL | 26 | 78.68 |
| 83.61 | 75.87 | Cook County, IL | 27 | 79.91 |
| 84.69 | 77.14 | Cook County, IL | 28 | 81.10 |
| 85.74 | 78.35 | Cook County, IL | 29 | 82.23 |
| 86.75 | 79.46 | Cook County, IL | 30 | 83.32 |
| 87.59 | 80.64 | Cook County, IL | 31 | 84.38 |
| 88.36 | 81.69 | Cook County, IL | 32 | 85.37 |
| 89.13 | 82.73 | Cook County, IL | 33 | 86.28 |
| 89.83 | 83.75 | Cook County, IL | 34 | 87.13 |
| 90.49 | 84.75 | Cook County, IL | 35 | 87.94 |
| 91.06 | 85.61 | Cook County, IL | 36 | 88.70 |
| 91.65 | 86.49 | Cook County, IL | 37 | 89.39 |
| 92.21 | 87.28 | Cook County, IL | 38 | 90.03 |
| 92.76 | 88.01 | Cook County, IL | 39 | 90.64 |
| 93.27 | 88.68 | Cook County, IL | 40 | 91.23 |
| 93.76 | 89.36 | Cook County, IL | 41 | 91.77 |
| 94.21 | 89.98 | Cook County, IL | 42 | 92.27 |
| 94.64 | 90.57 | Cook County, IL | 43 | 92.75 |
| 95.06 | 91.16 | Cook County, IL | 44 | 93.22 |
| 95.41 | 91.67 | Cook County, IL | 45 | 93.68 |
| 95.76 | 92.21 | Cook County, IL | 46 | 94.12 |
| 96.10 | 92.69 | Cook County, IL | 47 | 94.56 |
| 96.40 | 93.16 | Cook County, IL | 48 | 94.97 |
| 96.70 | 93.60 | Cook County, IL | 49 | 95.36 |
| 96.97 | 94.03 | Cook County, IL | 50 | 95.72 |
| n/a | n/a | Wayne County, MI | 0 | 0 |
| 5.45 | 3.32 | Wayne County, MI | 1 | 4.42 |
| 10.36 | 6.53 | Wayne County, MI | 2 | 8.33 |
| 15.01 | 9.61 | Wayne County, MI | 3 | 12.07 |
| 19.25 | 12.63 | Wayne County, MI | 4 | 15.75 |
| 23.38 | 15.57 | Wayne County, MI | 5 | 19.30 |
| 27.12 | 18.32 | Wayne County, MI | 6 | 22.48 |
| 30.77 | 21.03 | Wayne County, MI | 7 | 25.55 |
| 34.17 | 23.57 | Wayne County, MI | 8 | 28.45 |
| 37.52 | 26.02 | Wayne County, MI | 9 | 31.31 |
| 40.80 | 28.41 | Wayne County, MI | 10 | 34.05 |
| 43.96 | 30.78 | Wayne County, MI | 11 | 36.77 |
| 46.93 | 33.10 | Wayne County, MI | 12 | 39.48 |
| 49.86 | 35.42 | Wayne County, MI | 13 | 42.17 |
| 52.68 | 37.70 | Wayne County, MI | 14 | 44.76 |
| 55.44 | 39.93 | Wayne County, MI | 15 | 47.23 |
| 57.91 | 42.11 | Wayne County, MI | 16 | 49.67 |
| 60.30 | 44.21 | Wayne County, MI | 17 | 52.05 |
| 62.45 | 46.37 | Wayne County, MI | 18 | 54.42 |
| 64.54 | 48.51 | Wayne County, MI | 19 | 56.70 |
| 66.62 | 50.64 | Wayne County, MI | 20 | 58.95 |
| 68.77 | 52.64 | Wayne County, MI | 21 | 61.06 |
| 70.64 | 54.49 | Wayne County, MI | 22 | 63.03 |
| 72.33 | 56.47 | Wayne County, MI | 23 | 64.82 |
| 74.12 | 58.30 | Wayne County, MI | 24 | 66.54 |
| 75.73 | 60.19 | Wayne County, MI | 25 | 68.23 |
| 77.31 | 62.04 | Wayne County, MI | 26 | 69.91 |
| 78.91 | 63.69 | Wayne County, MI | 27 | 71.59 |
| 80.38 | 65.26 | Wayne County, MI | 28 | 73.26 |
| 81.69 | 66.93 | Wayne County, MI | 29 | 74.80 |
| 82.88 | 68.46 | Wayne County, MI | 30 | 76.26 |
| 83.98 | 69.95 | Wayne County, MI | 31 | 77.60 |
| 84.98 | 71.43 | Wayne County, MI | 32 | 78.85 |
| 86 | 72.78 | Wayne County, MI | 33 | 80.05 |
| 86.84 | 74.14 | Wayne County, MI | 34 | 81.21 |
| 87.63 | 75.47 | Wayne County, MI | 35 | 82.36 |
| 88.33 | 76.66 | Wayne County, MI | 36 | 83.39 |
| 88.95 | 77.89 | Wayne County, MI | 37 | 84.38 |
| 89.65 | 79.07 | Wayne County, MI | 38 | 85.31 |
| 90.30 | 80.17 | Wayne County, MI | 39 | 86.18 |
| 90.87 | 81.41 | Wayne County, MI | 40 | 87.03 |
| 91.43 | 82.47 | Wayne County, MI | 41 | 87.88 |
| 91.95 | 83.43 | Wayne County, MI | 42 | 88.60 |
| 92.46 | 84.37 | Wayne County, MI | 43 | 89.27 |
| 92.93 | 85.28 | Wayne County, MI | 44 | 89.88 |
| 93.36 | 86.07 | Wayne County, MI | 45 | 90.45 |
| 93.77 | 86.91 | Wayne County, MI | 46 | 90.98 |
| 94.19 | 87.69 | Wayne County, MI | 47 | 91.47 |
| 94.58 | 88.41 | Wayne County, MI | 48 | 91.94 |
| 94.89 | 89.01 | Wayne County, MI | 49 | 92.40 |
| 95.19 | 89.62 | Wayne County, MI | 50 | 92.85 |
| n/a | n/a | Los Angeles County, CA | 0 | 0 |
| 6.11 | 4.77 | Los Angeles County, CA | 1 | 5.48 |
| 11.12 | 8.79 | Los Angeles County, CA | 2 | 10.10 |
| 15.42 | 12.28 | Los Angeles County, CA | 3 | 13.96 |
| 19.22 | 15.45 | Los Angeles County, CA | 4 | 17.37 |
| 22.58 | 18.34 | Los Angeles County, CA | 5 | 20.47 |
| 25.66 | 21.06 | Los Angeles County, CA | 6 | 23.36 |
| 28.49 | 23.61 | Los Angeles County, CA | 7 | 26.08 |
| 31.14 | 26 | Los Angeles County, CA | 8 | 28.59 |
| 33.62 | 28.29 | Los Angeles County, CA | 9 | 30.96 |
| 35.94 | 30.54 | Los Angeles County, CA | 10 | 33.24 |
| 38.15 | 32.68 | Los Angeles County, CA | 11 | 35.44 |
| 40.28 | 34.78 | Los Angeles County, CA | 12 | 37.56 |
| 42.38 | 36.79 | Los Angeles County, CA | 13 | 39.63 |
| 44.42 | 38.78 | Los Angeles County, CA | 14 | 41.65 |
| 46.40 | 40.73 | Los Angeles County, CA | 15 | 43.59 |
| 48.34 | 42.57 | Los Angeles County, CA | 16 | 45.48 |
| 50.18 | 44.39 | Los Angeles County, CA | 17 | 47.34 |
| 52.01 | 46.20 | Los Angeles County, CA | 18 | 49.16 |
| 53.79 | 47.98 | Los Angeles County, CA | 19 | 50.90 |
| 55.54 | 49.70 | Los Angeles County, CA | 20 | 52.62 |
| 57.24 | 51.36 | Los Angeles County, CA | 21 | 54.31 |
| 58.85 | 53.02 | Los Angeles County, CA | 22 | 55.96 |
| 60.48 | 54.65 | Los Angeles County, CA | 23 | 57.59 |
| 62.06 | 56.25 | Los Angeles County, CA | 24 | 59.17 |
| 63.56 | 57.77 | Los Angeles County, CA | 25 | 60.71 |
| 65.03 | 59.36 | Los Angeles County, CA | 26 | 62.21 |
| 66.46 | 60.84 | Los Angeles County, CA | 27 | 63.67 |
| 67.87 | 62.26 | Los Angeles County, CA | 28 | 65.08 |
| 69.21 | 63.67 | Los Angeles County, CA | 29 | 66.47 |
| 70.54 | 65.04 | Los Angeles County, CA | 30 | 67.81 |
| 71.83 | 66.37 | Los Angeles County, CA | 31 | 69.11 |
| 73.12 | 67.68 | Los Angeles County, CA | 32 | 70.38 |
| 74.33 | 68.95 | Los Angeles County, CA | 33 | 71.63 |
| 75.52 | 70.17 | Los Angeles County, CA | 34 | 72.85 |
| 76.69 | 71.38 | Los Angeles County, CA | 35 | 74.04 |
| 77.82 | 72.57 | Los Angeles County, CA | 36 | 75.21 |
| 78.93 | 73.73 | Los Angeles County, CA | 37 | 76.34 |
| 79.96 | 74.85 | Los Angeles County, CA | 38 | 77.44 |
| 80.98 | 75.98 | Los Angeles County, CA | 39 | 78.51 |
| 81.99 | 77.09 | Los Angeles County, CA | 40 | 79.55 |
| 82.96 | 78.15 | Los Angeles County, CA | 41 | 80.56 |
| 83.89 | 79.17 | Los Angeles County, CA | 42 | 81.54 |
| 84.78 | 80.14 | Los Angeles County, CA | 43 | 82.50 |
| 85.65 | 81.10 | Los Angeles County, CA | 44 | 83.43 |
| 86.47 | 82.06 | Los Angeles County, CA | 45 | 84.33 |
| 87.27 | 82.96 | Los Angeles County, CA | 46 | 85.19 |
| 88.01 | 83.83 | Los Angeles County, CA | 47 | 86.02 |
| 88.70 | 84.69 | Los Angeles County, CA | 48 | 86.81 |
| 89.36 | 85.52 | Los Angeles County, CA | 49 | 87.55 |
| 89.99 | 86.29 | Los Angeles County, CA | 50 | 88.24 |
| n/a | n/a | Maricopa County, AZ | 0 | 0 |
| 5.14 | 3.39 | Maricopa County, AZ | 1 | 4.36 |
| 9.34 | 6.49 | Maricopa County, AZ | 2 | 7.95 |
| 12.92 | 9.36 | Maricopa County, AZ | 3 | 11.20 |
| 16.15 | 12.07 | Maricopa County, AZ | 4 | 14.14 |
| 19.20 | 14.71 | Maricopa County, AZ | 5 | 16.94 |
| 22.18 | 17.25 | Maricopa County, AZ | 6 | 19.62 |
| 24.99 | 19.73 | Maricopa County, AZ | 7 | 22.27 |
| 27.70 | 22.20 | Maricopa County, AZ | 8 | 24.85 |
| 30.27 | 24.55 | Maricopa County, AZ | 9 | 27.40 |
| 32.76 | 26.90 | Maricopa County, AZ | 10 | 29.87 |
| 35.28 | 29.17 | Maricopa County, AZ | 11 | 32.24 |
| 37.64 | 31.32 | Maricopa County, AZ | 12 | 34.51 |
| 39.95 | 33.43 | Maricopa County, AZ | 13 | 36.70 |
| 42.14 | 35.45 | Maricopa County, AZ | 14 | 38.81 |
| 44.20 | 37.44 | Maricopa County, AZ | 15 | 40.82 |
| 46.26 | 39.35 | Maricopa County, AZ | 16 | 42.80 |
| 48.23 | 41.27 | Maricopa County, AZ | 17 | 44.73 |
| 50.15 | 43.05 | Maricopa County, AZ | 18 | 46.59 |
| 51.93 | 44.73 | Maricopa County, AZ | 19 | 48.43 |
| 53.70 | 46.44 | Maricopa County, AZ | 20 | 50.17 |
| 55.44 | 48.13 | Maricopa County, AZ | 21 | 51.84 |
| 57.05 | 49.75 | Maricopa County, AZ | 22 | 53.42 |
| 58.70 | 51.29 | Maricopa County, AZ | 23 | 54.99 |
| 60.25 | 52.79 | Maricopa County, AZ | 24 | 56.54 |
| 61.81 | 54.30 | Maricopa County, AZ | 25 | 58.06 |
| 63.31 | 55.84 | Maricopa County, AZ | 26 | 59.54 |
| 64.80 | 57.25 | Maricopa County, AZ | 27 | 61.01 |
| 66.25 | 58.69 | Maricopa County, AZ | 28 | 62.46 |
| 67.64 | 60.06 | Maricopa County, AZ | 29 | 63.89 |
| 69 | 61.48 | Maricopa County, AZ | 30 | 65.28 |
| 70.35 | 62.84 | Maricopa County, AZ | 31 | 66.63 |
| 71.58 | 64.22 | Maricopa County, AZ | 32 | 67.96 |
| 72.79 | 65.48 | Maricopa County, AZ | 33 | 69.24 |
| 74.02 | 66.74 | Maricopa County, AZ | 34 | 70.49 |
| 75.17 | 68.01 | Maricopa County, AZ | 35 | 71.70 |
| 76.36 | 69.22 | Maricopa County, AZ | 36 | 72.87 |
| 77.52 | 70.39 | Maricopa County, AZ | 37 | 74.01 |
| 78.60 | 71.55 | Maricopa County, AZ | 38 | 75.11 |
| 79.66 | 72.68 | Maricopa County, AZ | 39 | 76.21 |
| 80.71 | 73.76 | Maricopa County, AZ | 40 | 77.27 |
| 81.73 | 74.88 | Maricopa County, AZ | 41 | 78.29 |
| 82.68 | 75.97 | Maricopa County, AZ | 42 | 79.30 |
| 83.63 | 76.95 | Maricopa County, AZ | 43 | 80.29 |
| 84.51 | 77.99 | Maricopa County, AZ | 44 | 81.24 |
| 85.36 | 78.99 | Maricopa County, AZ | 45 | 82.17 |
| 86.25 | 79.97 | Maricopa County, AZ | 46 | 83.10 |
| 87.10 | 80.89 | Maricopa County, AZ | 47 | 84.01 |
| 87.94 | 81.79 | Maricopa County, AZ | 48 | 84.89 |
| 88.67 | 82.69 | Maricopa County, AZ | 49 | 85.74 |
| 89.43 | 83.59 | Maricopa County, AZ | 50 | 86.57 |
| n/a | n/a | Dallas County, TX | 0 | 0 |
| 7.67 | 3.10 | Dallas County, TX | 1 | 5.88 |
| 13.39 | 6.14 | Dallas County, TX | 2 | 9.17 |
| 17.35 | 9.14 | Dallas County, TX | 3 | 12.39 |
| 20.46 | 12.12 | Dallas County, TX | 4 | 15.56 |
| 23.48 | 15.05 | Dallas County, TX | 5 | 18.72 |
| 26.37 | 17.84 | Dallas County, TX | 6 | 21.85 |
| 29.17 | 20.44 | Dallas County, TX | 7 | 24.80 |
| 31.64 | 22.77 | Dallas County, TX | 8 | 27.44 |
| 34.10 | 24.92 | Dallas County, TX | 9 | 29.92 |
| 36.48 | 27.02 | Dallas County, TX | 10 | 32.14 |
| 38.70 | 28.82 | Dallas County, TX | 11 | 34.22 |
| 40.78 | 30.57 | Dallas County, TX | 12 | 36.10 |
| 42.76 | 32.29 | Dallas County, TX | 13 | 37.86 |
| 44.53 | 33.98 | Dallas County, TX | 14 | 39.50 |
| 46.27 | 35.62 | Dallas County, TX | 15 | 41.06 |
| 47.84 | 37.20 | Dallas County, TX | 16 | 42.61 |
| 49.48 | 38.79 | Dallas County, TX | 17 | 44.15 |
| 50.96 | 40.39 | Dallas County, TX | 18 | 45.66 |
| 52.44 | 41.90 | Dallas County, TX | 19 | 47.17 |
| 53.87 | 43.41 | Dallas County, TX | 20 | 48.67 |
| 55.28 | 44.87 | Dallas County, TX | 21 | 50.16 |
| 56.65 | 46.34 | Dallas County, TX | 22 | 51.61 |
| 58.03 | 47.80 | Dallas County, TX | 23 | 52.99 |
| 59.41 | 49.19 | Dallas County, TX | 24 | 54.36 |
| 60.83 | 50.61 | Dallas County, TX | 25 | 55.73 |
| 62.10 | 52 | Dallas County, TX | 26 | 57.07 |
| 63.37 | 53.36 | Dallas County, TX | 27 | 58.37 |
| 64.59 | 54.72 | Dallas County, TX | 28 | 59.67 |
| 65.81 | 56.02 | Dallas County, TX | 29 | 60.96 |
| 66.97 | 57.32 | Dallas County, TX | 30 | 62.23 |
| 68.14 | 58.61 | Dallas County, TX | 31 | 63.49 |
| 69.27 | 59.91 | Dallas County, TX | 32 | 64.74 |
| 70.34 | 61.10 | Dallas County, TX | 33 | 65.98 |
| 71.45 | 62.36 | Dallas County, TX | 34 | 67.13 |
| 72.47 | 63.55 | Dallas County, TX | 35 | 68.24 |
| 73.44 | 64.70 | Dallas County, TX | 36 | 69.30 |
| 74.37 | 65.88 | Dallas County, TX | 37 | 70.36 |
| 75.32 | 67.02 | Dallas County, TX | 38 | 71.37 |
| 76.26 | 68.09 | Dallas County, TX | 39 | 72.36 |
| 77.14 | 69.16 | Dallas County, TX | 40 | 73.34 |
| 78.09 | 70.21 | Dallas County, TX | 41 | 74.28 |
| 78.96 | 71.17 | Dallas County, TX | 42 | 75.22 |
| 79.82 | 72.14 | Dallas County, TX | 43 | 76.12 |
| 80.66 | 73.12 | Dallas County, TX | 44 | 77.02 |
| 81.48 | 74.03 | Dallas County, TX | 45 | 77.90 |
| 82.28 | 74.95 | Dallas County, TX | 46 | 78.75 |
| 83.07 | 75.84 | Dallas County, TX | 47 | 79.59 |
| 83.83 | 76.71 | Dallas County, TX | 48 | 80.40 |
| 84.60 | 77.58 | Dallas County, TX | 49 | 81.20 |
| 85.35 | 78.42 | Dallas County, TX | 50 | 81.99 |
| n/a | n/a | Philadelphia, PA | 0 | 0 |
| 4.86 | 2.49 | Philadelphia, PA | 1 | 4.05 |
| 9.03 | 4.95 | Philadelphia, PA | 2 | 6.90 |
| 12.33 | 7.35 | Philadelphia, PA | 3 | 9.67 |
| 15.30 | 9.72 | Philadelphia, PA | 4 | 12.40 |
| 18.15 | 12.02 | Philadelphia, PA | 5 | 14.95 |
| 20.94 | 14.24 | Philadelphia, PA | 6 | 17.37 |
| 23.56 | 16.44 | Philadelphia, PA | 7 | 19.76 |
| 26.22 | 18.59 | Philadelphia, PA | 8 | 22.13 |
| 28.74 | 20.72 | Philadelphia, PA | 9 | 24.49 |
| 31.05 | 22.74 | Philadelphia, PA | 10 | 26.77 |
| 33.41 | 24.70 | Philadelphia, PA | 11 | 28.93 |
| 35.79 | 26.58 | Philadelphia, PA | 12 | 31.09 |
| 37.81 | 28.50 | Philadelphia, PA | 13 | 33.25 |
| 40.03 | 30.35 | Philadelphia, PA | 14 | 35.30 |
| 41.97 | 32.22 | Philadelphia, PA | 15 | 37.27 |
| 43.97 | 34 | Philadelphia, PA | 16 | 39.20 |
| 45.92 | 35.78 | Philadelphia, PA | 17 | 41.12 |
| 47.71 | 37.43 | Philadelphia, PA | 18 | 42.87 |
| 49.45 | 39.16 | Philadelphia, PA | 19 | 44.55 |
| 51.14 | 40.80 | Philadelphia, PA | 20 | 46.16 |
| 52.82 | 42.40 | Philadelphia, PA | 21 | 47.74 |
| 54.44 | 43.98 | Philadelphia, PA | 22 | 49.31 |
| 56.15 | 45.48 | Philadelphia, PA | 23 | 50.87 |
| 57.78 | 46.99 | Philadelphia, PA | 24 | 52.43 |
| 59.31 | 48.49 | Philadelphia, PA | 25 | 53.97 |
| 60.72 | 49.90 | Philadelphia, PA | 26 | 55.45 |
| 62.11 | 51.33 | Philadelphia, PA | 27 | 56.88 |
| 63.51 | 52.74 | Philadelphia, PA | 28 | 58.20 |
| 64.86 | 54.05 | Philadelphia, PA | 29 | 59.49 |
| 66.19 | 55.39 | Philadelphia, PA | 30 | 60.76 |
| 67.52 | 56.63 | Philadelphia, PA | 31 | 62.03 |
| 68.80 | 57.85 | Philadelphia, PA | 32 | 63.28 |
| 70.05 | 59.01 | Philadelphia, PA | 33 | 64.51 |
| 71.16 | 60.22 | Philadelphia, PA | 34 | 65.74 |
| 72.29 | 61.42 | Philadelphia, PA | 35 | 66.96 |
| 73.45 | 62.60 | Philadelphia, PA | 36 | 68.15 |
| 74.57 | 63.79 | Philadelphia, PA | 37 | 69.34 |
| 75.65 | 64.95 | Philadelphia, PA | 38 | 70.51 |
| 76.70 | 66.06 | Philadelphia, PA | 39 | 71.62 |
| 77.68 | 67.11 | Philadelphia, PA | 40 | 72.68 |
| 78.66 | 68.17 | Philadelphia, PA | 41 | 73.74 |
| 79.64 | 69.24 | Philadelphia, PA | 42 | 74.74 |
| 80.57 | 70.20 | Philadelphia, PA | 43 | 75.71 |
| 81.49 | 71.20 | Philadelphia, PA | 44 | 76.66 |
| 82.39 | 72.22 | Philadelphia, PA | 45 | 77.57 |
| 83.23 | 73.17 | Philadelphia, PA | 46 | 78.49 |
| 84.05 | 74.11 | Philadelphia, PA | 47 | 79.39 |
| 84.83 | 75.07 | Philadelphia, PA | 48 | 80.28 |
| 85.60 | 76.03 | Philadelphia, PA | 49 | 81.17 |
| 86.33 | 76.93 | Philadelphia, PA | 50 | 82.01 |
| n/a | n/a | Harris County, TX | 0 | 0 |
| 4.53 | 3.20 | Harris County, TX | 1 | 4.19 |
| 8.62 | 6.23 | Harris County, TX | 2 | 7.65 |
| 12.14 | 9.08 | Harris County, TX | 3 | 10.70 |
| 15.32 | 11.64 | Harris County, TX | 4 | 13.57 |
| 18.21 | 13.95 | Harris County, TX | 5 | 16.30 |
| 20.87 | 16.11 | Harris County, TX | 6 | 18.66 |
| 23.36 | 18.17 | Harris County, TX | 7 | 20.89 |
| 25.58 | 20.15 | Harris County, TX | 8 | 22.96 |
| 27.75 | 22.10 | Harris County, TX | 9 | 24.96 |
| 29.84 | 23.99 | Harris County, TX | 10 | 26.88 |
| 31.88 | 25.80 | Harris County, TX | 11 | 28.77 |
| 33.81 | 27.59 | Harris County, TX | 12 | 30.64 |
| 35.60 | 29.33 | Harris County, TX | 13 | 32.40 |
| 37.39 | 31.05 | Harris County, TX | 14 | 34.14 |
| 39.16 | 32.72 | Harris County, TX | 15 | 35.85 |
| 40.84 | 34.31 | Harris County, TX | 16 | 37.54 |
| 42.47 | 35.90 | Harris County, TX | 17 | 39.18 |
| 44.04 | 37.48 | Harris County, TX | 18 | 40.72 |
| 45.57 | 39.01 | Harris County, TX | 19 | 42.25 |
| 47.09 | 40.52 | Harris County, TX | 20 | 43.76 |
| 48.64 | 42.03 | Harris County, TX | 21 | 45.27 |
| 50.12 | 43.51 | Harris County, TX | 22 | 46.76 |
| 51.59 | 44.96 | Harris County, TX | 23 | 48.23 |
| 53.04 | 46.39 | Harris County, TX | 24 | 49.67 |
| 54.44 | 47.79 | Harris County, TX | 25 | 51.08 |
| 55.84 | 49.16 | Harris County, TX | 26 | 52.47 |
| 57.25 | 50.51 | Harris County, TX | 27 | 53.81 |
| 58.54 | 51.80 | Harris County, TX | 28 | 55.14 |
| 59.88 | 53.10 | Harris County, TX | 29 | 56.46 |
| 61.19 | 54.39 | Harris County, TX | 30 | 57.76 |
| 62.48 | 55.68 | Harris County, TX | 31 | 59.01 |
| 63.75 | 56.94 | Harris County, TX | 32 | 60.24 |
| 64.97 | 58.15 | Harris County, TX | 33 | 61.47 |
| 66.17 | 59.36 | Harris County, TX | 34 | 62.67 |
| 67.41 | 60.58 | Harris County, TX | 35 | 63.87 |
| 68.60 | 61.78 | Harris County, TX | 36 | 65.07 |
| 69.79 | 62.97 | Harris County, TX | 37 | 66.24 |
| 70.95 | 64.14 | Harris County, TX | 38 | 67.40 |
| 72.11 | 65.27 | Harris County, TX | 39 | 68.54 |
| 73.17 | 66.41 | Harris County, TX | 40 | 69.67 |
| 74.22 | 67.55 | Harris County, TX | 41 | 70.78 |
| 75.31 | 68.68 | Harris County, TX | 42 | 71.87 |
| 76.39 | 69.77 | Harris County, TX | 43 | 72.95 |
| 77.43 | 70.86 | Harris County, TX | 44 | 74 |
| 78.41 | 71.91 | Harris County, TX | 45 | 75.03 |
| 79.39 | 72.98 | Harris County, TX | 46 | 76.04 |
| 80.34 | 74 | Harris County, TX | 47 | 77.02 |
| 81.27 | 74.97 | Harris County, TX | 48 | 78 |
| 82.18 | 75.95 | Harris County, TX | 49 | 78.96 |
| 83.08 | 76.92 | Harris County, TX | 50 | 79.90 |
| n/a | n/a | Shelby County, TN | 0 | 0 |
| 3.35 | 2.13 | Shelby County, TN | 1 | 2.76 |
| 6.68 | 4.25 | Shelby County, TN | 2 | 5.52 |
| 9.65 | 6.32 | Shelby County, TN | 3 | 8.03 |
| 12.70 | 8.34 | Shelby County, TN | 4 | 10.54 |
| 15.44 | 10.35 | Shelby County, TN | 5 | 13.05 |
| 18.15 | 12.33 | Shelby County, TN | 6 | 15.41 |
| 20.72 | 14.28 | Shelby County, TN | 7 | 17.65 |
| 23.23 | 16.16 | Shelby County, TN | 8 | 19.74 |
| 25.83 | 18.07 | Shelby County, TN | 9 | 21.82 |
| 28.34 | 19.94 | Shelby County, TN | 10 | 23.88 |
| 30.73 | 21.73 | Shelby County, TN | 11 | 25.94 |
| 32.99 | 23.47 | Shelby County, TN | 12 | 27.95 |
| 35.28 | 25.18 | Shelby County, TN | 13 | 29.93 |
| 37.60 | 26.90 | Shelby County, TN | 14 | 31.91 |
| 39.58 | 28.63 | Shelby County, TN | 15 | 33.87 |
| 41.60 | 30.26 | Shelby County, TN | 16 | 35.82 |
| 43.59 | 31.92 | Shelby County, TN | 17 | 37.73 |
| 45.55 | 33.59 | Shelby County, TN | 18 | 39.54 |
| 47.46 | 35.22 | Shelby County, TN | 19 | 41.31 |
| 49.31 | 36.81 | Shelby County, TN | 20 | 43.05 |
| 51.04 | 38.36 | Shelby County, TN | 21 | 44.78 |
| 52.73 | 39.81 | Shelby County, TN | 22 | 46.34 |
| 54.49 | 41.28 | Shelby County, TN | 23 | 47.91 |
| 56.08 | 42.67 | Shelby County, TN | 24 | 49.44 |
| 57.73 | 44.14 | Shelby County, TN | 25 | 50.96 |
| 59.35 | 45.53 | Shelby County, TN | 26 | 52.36 |
| 60.91 | 46.94 | Shelby County, TN | 27 | 53.75 |
| 62.40 | 48.26 | Shelby County, TN | 28 | 55.14 |
| 63.91 | 49.58 | Shelby County, TN | 29 | 56.48 |
| 65.29 | 50.86 | Shelby County, TN | 30 | 57.81 |
| 66.75 | 52.10 | Shelby County, TN | 31 | 59.13 |
| 68.12 | 53.35 | Shelby County, TN | 32 | 60.45 |
| 69.48 | 54.60 | Shelby County, TN | 33 | 61.75 |
| 70.76 | 55.83 | Shelby County, TN | 34 | 63.03 |
| 72.05 | 57.05 | Shelby County, TN | 35 | 64.32 |
| 73.29 | 58.24 | Shelby County, TN | 36 | 65.58 |
| 74.51 | 59.41 | Shelby County, TN | 37 | 66.83 |
| 75.68 | 60.54 | Shelby County, TN | 38 | 67.99 |
| 76.80 | 61.74 | Shelby County, TN | 39 | 69.15 |
| 78.05 | 62.78 | Shelby County, TN | 40 | 70.29 |
| 79.14 | 63.88 | Shelby County, TN | 41 | 71.42 |
| 80.16 | 65.08 | Shelby County, TN | 42 | 72.47 |
| 81.19 | 66.11 | Shelby County, TN | 43 | 73.53 |
| 82.23 | 67.14 | Shelby County, TN | 44 | 74.58 |
| 83.24 | 68.27 | Shelby County, TN | 45 | 75.63 |
| 84.18 | 69.31 | Shelby County, TN | 46 | 76.67 |
| 85.11 | 70.40 | Shelby County, TN | 47 | 77.70 |
| 85.94 | 71.36 | Shelby County, TN | 48 | 78.69 |
| 86.79 | 72.37 | Shelby County, TN | 49 | 79.66 |
| 87.66 | 73.37 | Shelby County, TN | 50 | 80.63 |

Note: The 8 counties are those with the most implied firearm homicide deaths a year in the cluster file; New York City is left out. A cluster that crosses a county line counts in each county with only the residents who live there.

Note: The unit is the tract cluster, a group of neighboring tracts with one CDC rate. In every county shown at least 80.2% of residents live in a cluster with a rate in every year from 2022 to 2024, and 87.7% do in Cook County; the others are left out.

Note: Each curve carries a 95% interval from a bootstrap that draws clusters within the county with replacement (replicates and seed are in the methods note); the intervals are in the tooltip and the data table. A county line cuts through metropolitan areas, so a county that includes suburbs holds more low-rate residents than one that does not.

Source: CDC Mapping Injury, Overdose, and Violence, tract file, firearm homicide, 2022-2024; modeled estimates for tract clusters; ACS 2019-2023 population

Data: https://vishalsingh.org/zipcode/report/gun-homicide-and-causes-of-early-death/figures/county_concentration.json

### Table 3. In Cook County, 24.2% of tracts with an estimate sit in clusters with no recorded firearm homicide, while the highest cluster rate reaches 89 per hundred thousand residents

Mean firearm homicide rate per 100,000 residents, 2022 to 2024, of the tract cluster each Cook County tract belongs to

No inline rows: the values of Table 3 are in https://vishalsingh.org/zipcode/data/gun-homicide-and-causes-of-early-death/chicago_homicide_map.json.

Note: The unit of this map is the tract cluster. CDC publishes one rate for each group of neighboring tracts (the median tract sits in a cluster of 5) and the file repeats it on every member tract, so a shade covers a cluster and the map does not describe single tracts. 1,163 tracts are shaded and 168 are gray.

Note: A cluster-year with 1 to 9 deaths gets a CDC modeled estimate, one with no deaths a rate of zero from a count of zero, and one with 10 or more deaths a direct rate. A mean rate is shown only when CDC publishes a rate in each of 2022, 2023 and 2024, and the data for 2024 are provisional. The highest cluster rate is an average over several tracts, so a single block can run higher.

Note: Life expectancy is the USALEEP estimate for deaths in 2010 to 2015, area-weighted onto 2020 tracts, with institutional tracts left blank; the deprivation rank is the 2023 national percentile, 1 least and 100 most deprived.

Source: CDC Mapping Injury, Overdose, and Violence, tract file, firearm homicide, 2022-2024; modeled estimates for tract clusters; USALEEP 2010-2015; ADI 2023

Data: https://vishalsingh.org/zipcode/report/gun-homicide-and-causes-of-early-death/figures/chicago_homicide_map.json

### Table 4. In Cook County, residents of the highest-rate tenth of clusters have a life expectancy of 71.7 years (deaths in 2010 to 2015), 9.5 fewer than residents of clusters with no recorded deaths

Cook County residents in tenths by the firearm homicide rate of their tract cluster, 2022 to 2024: the rate, life expectancy at birth for deaths in 2010 to 2015, and the Area Deprivation Index

#### Table 4.1. Firearm homicide rate

The highest tenth of Cook County residents lives in clusters with a rate of 60 per hundred thousand residents

Mean firearm homicide deaths per 100,000 residents a year, 2022 to 2024, by tenth of residents

| conf_high | conf_low | estimate | n_units | tenth |
| ---: | ---: | ---: | ---: | ---: |
| n/a | n/a | 0 | 30 | 1 |
| n/a | n/a | 0 | 30 | 2 |
| 0.490 | 0.110 | 0.300 | 27 | 3 |
| 2.13 | 1.85 | 1.99 | 27 | 4 |
| 3.25 | 2.87 | 3.05 | 27 | 5 |
| 5.83 | 5.30 | 5.57 | 26 | 6 |
| 10.65 | 9.18 | 9.91 | 30 | 7 |
| 17.03 | 15.55 | 16.29 | 25 | 8 |
| 30.45 | 25.20 | 27.75 | 31 | 9 |
| 65.05 | 55.83 | 60.03 | 26 | 10 |

#### Table 4.2. Life expectancy

Life expectancy for deaths in 2010 to 2015 falls to 71.7 years in the highest tenth of Cook County clusters

Population-weighted mean life expectancy at birth, deaths in 2010 to 2015, by tenth of residents ranked by firearm homicide rate of the cluster, 2022 to 2024

| conf_high | conf_low | estimate | n_units | tenth |
| ---: | ---: | ---: | ---: | ---: |
| 81.38 | 79.72 | 80.59 | 30 | 1 |
| 82.42 | 80.94 | 81.65 | 30 | 2 |
| 81.92 | 80.26 | 81.05 | 27 | 3 |
| 81.02 | 79.85 | 80.43 | 27 | 4 |
| 80.24 | 78.69 | 79.48 | 25 | 5 |
| 80.21 | 78.93 | 79.56 | 26 | 6 |
| 78.58 | 77.17 | 77.92 | 30 | 7 |
| 78.08 | 76.49 | 77.27 | 25 | 8 |
| 75.93 | 74.39 | 75.19 | 31 | 9 |
| 72.39 | 70.99 | 71.69 | 26 | 10 |

#### Table 4.3. Deprivation

The deprivation rank is 74 in the highest tenth of Cook County clusters

Population-weighted mean Area Deprivation Index national rank, 2023, by tenth of residents ranked by firearm homicide rate of the cluster, 2022 to 2024

| conf_high | conf_low | estimate | n_units | tenth |
| ---: | ---: | ---: | ---: | ---: |
| 39.52 | 30.90 | 35.31 | 30 | 1 |
| 34.07 | 23.53 | 29.05 | 30 | 2 |
| 37.86 | 27.25 | 32.88 | 27 | 3 |
| 44.38 | 31.99 | 37.79 | 27 | 4 |
| 44.79 | 33.73 | 39.49 | 27 | 5 |
| 46.10 | 34.81 | 40.18 | 26 | 6 |
| 56.36 | 48.78 | 52.70 | 30 | 7 |
| 63.59 | 55.39 | 59.58 | 25 | 8 |
| 74.67 | 64.85 | 69.67 | 31 | 9 |
| 78.28 | 71.04 | 74.37 | 26 | 10 |

Note: The unit is the tract cluster. CDC publishes one rate for each group of neighboring tracts, and tracts inherit it. Tenths hold equal numbers of residents among the 87.7% of Cook County residents who live in clusters with a rate in every year from 2022 to 2024. Clusters with zero recorded deaths hold 27.0% of those residents and fill the lowest two tenths and part of the third, ordered among themselves at random with the book's fixed seed, so the first three tenths differ only in who lives there.

Note: The tenths are formed on the rate itself, so the first tab shows how far apart the groups were sorted. Life expectancy is the USALEEP estimate for deaths in 2010 to 2015, area-weighted onto 2020 tracts, with institutional tracts left out; the rates are for 2022 to 2024.

Note: Intervals are 95% intervals from a bootstrap that draws clusters with replacement (replicates and seed are in the methods note).

Source: CDC Mapping Injury, Overdose, and Violence, tract file, firearm homicide, 2022-2024; modeled estimates for tract clusters; USALEEP 2010-2015; ADI 2023; population-weighted

Data: https://vishalsingh.org/zipcode/report/gun-homicide-and-causes-of-early-death/figures/cook_deciles.json

### Table 5. Premature mortality rises from 307 to 701 per hundred thousand residents across deciles of county deprivation

Population-weighted mean age-adjusted deaths before age 75 per 100,000 residents, by decile of the county Area Deprivation Index, counties, County Health Rankings 2024 release

| conf_high | conf_low | decile | estimate | n_units |
| ---: | ---: | ---: | ---: | ---: |
| 317.6 | 296.1 | 1 | 306.9 | 308 |
| 387.5 | 356.6 | 2 | 372.7 | 308 |
| 438.9 | 399.7 | 3 | 417.1 | 308 |
| 493.3 | 439.2 | 4 | 465.4 | 309 |
| 510.6 | 465.7 | 5 | 490.1 | 308 |
| 536.2 | 483.2 | 6 | 505.8 | 308 |
| 563.4 | 502 | 7 | 531.3 | 309 |
| 588.4 | 508.8 | 8 | 543.4 | 308 |
| 631.6 | 568.9 | 9 | 604 | 307 |
| 739.9 | 659.7 | 10 | 701.1 | 310 |

Note: Counties are ranked by the population-weighted mean of their tract Area Deprivation Index ranks and split into ten groups with equal numbers of counties; the means are weighted by county population. 3,083 counties have both a rate and a rank; County Health Rankings leaves out a county with fewer than 20 premature deaths in the window. Puerto Rico is not in County Health Rankings and is not drawn.

Note: The band is a 95% interval for the mean, from a bootstrap that draws whole states with replacement (replicates and seed are in the methods note); it is not the spread of counties within a decile. The rate is County Health Rankings premature age-adjusted mortality, deaths of residents under 75 per hundred thousand, pooled over 2019 to 2021.

Note: Across all the counties in the comparison the rate is 392 per hundred thousand residents.

Source: County Health Rankings 2024 release, NCHS death counts 2019-2021, age-adjusted rates from full counts, population-weighted; Area Deprivation Index 2023

Data: https://vishalsingh.org/zipcode/report/gun-homicide-and-causes-of-early-death/figures/premature_death_gradient.json

### Table 6. Overdose is 51.0% of the four-cause toll in the least deprived fifth of residents and 43.6% in the most, while motor-vehicle crashes go from 16.0% to 21.4%

Each kind of death's share of the sum of four crude death rates (overdose, suicide, homicide, motor-vehicle crash), by fifth of residents ranked by county deprivation, counties with all four rates published

| cause | conf_high | conf_low | n_units | quintile | rate | share |
| --- | ---: | ---: | ---: | --- | ---: | ---: |
| Drug overdose | 57.56 | 46.95 | 71 | 1 (least deprived) | 21.58 | 50.97 |
| Suicide | 29.49 | 21.62 | 71 | 1 (least deprived) | 10.23 | 24.16 |
| Homicide | 10.21 | 5.90 | 71 | 1 (least deprived) | 3.74 | 8.83 |
| Motor-vehicle crash | 17.93 | 12.58 | 71 | 1 (least deprived) | 6.79 | 16.03 |
| Drug overdose | 51.82 | 43.30 | 117 | 2 | 24.84 | 47.08 |
| Suicide | 28.86 | 22.59 | 117 | 2 | 13.46 | 25.51 |
| Homicide | 10.24 | 7.57 | 117 | 2 | 4.78 | 9.06 |
| Motor-vehicle crash | 20.68 | 15.55 | 117 | 2 | 9.68 | 18.35 |
| Drug overdose | 50.35 | 42.32 | 189 | 3 | 27.79 | 46.56 |
| Suicide | 27.08 | 21.39 | 189 | 3 | 14.26 | 23.88 |
| Homicide | 13.28 | 8.39 | 189 | 3 | 6.31 | 10.58 |
| Motor-vehicle crash | 21.38 | 16.21 | 189 | 3 | 11.32 | 18.97 |
| Drug overdose | 52.09 | 42.41 | 253 | 4 | 32.31 | 47.43 |
| Suicide | 24.50 | 20.81 | 253 | 4 | 15.36 | 22.54 |
| Homicide | 13.37 | 9.49 | 253 | 4 | 7.85 | 11.53 |
| Motor-vehicle crash | 20.92 | 15.90 | 253 | 4 | 12.60 | 18.50 |
| Drug overdose | 48.22 | 38.43 | 618 | 5 (most deprived) | 33.30 | 43.58 |
| Suicide | 23.85 | 20.06 | 618 | 5 (most deprived) | 16.71 | 21.87 |
| Homicide | 14.67 | 11.67 | 618 | 5 (most deprived) | 10.05 | 13.15 |
| Motor-vehicle crash | 24.65 | 18.44 | 618 | 5 (most deprived) | 16.36 | 21.41 |

Note: Only counties with all four rates published enter (1,248 counties, 89.6% of county residents): County Health Rankings leaves a county out of a cause when it has fewer than 10 deaths in the window, so small counties are missing and a missing rate means too few deaths, not zero. Fifths hold equal numbers of residents, not counties.

Note: The four rates are crude, so they are comparable with one another, and they overlap a little: a suicide by drug poisoning counts as both an overdose and a suicide. Firearm deaths are left out because they sit inside suicide and homicide, and child mortality has a different denominator. The windows differ by cause (overdose 2019 to 2021, suicide 2017 to 2021, homicide and motor-vehicle crashes 2015 to 2021). Overdose appears in this chapter only in this mix, the leading-kind map and a cross-check against the CDC panel; the overdose maps, gradients and trend are in the chapters on overdose, alcohol and suicide and on mortality trends.

Note: Shares are the cause's population-weighted mean rate divided by the sum of the four means, with 95% intervals from a bootstrap that draws whole states with replacement (replicates and seed are in the methods note). The total toll is 42 per hundred thousand in the least deprived fifth and 76 in the most.

Source: County Health Rankings 2024 release, NCHS death counts 2015-2021 in windows that differ by cause, crude rates from full counts, population-weighted; Area Deprivation Index 2023

Data: https://vishalsingh.org/zipcode/report/gun-homicide-and-causes-of-early-death/figures/cause_mix_by_quintile.json

### Table 7. Homicide and other causes of early death are higher in more deprived counties, with a population-weighted correlation of +0.41 for homicide

Correlation across counties of each death measure with the county Area Deprivation Index and with the rural share of residents, weighted by county population or counting every county once, County Health Rankings 2024 release

| conf_high | conf_low | estimate | group | measure | n_units | view | weighting |
| ---: | ---: | ---: | --- | --- | ---: | --- | --- |
| 0.584 | 0.281 | 0.464 | Suicide (age-adjusted rate) | Suicide (age-adjusted rate) | 2,427 | deprivation | population-weighted |
| 0.317 | 0.098 | 0.207 | Suicide (age-adjusted rate) | Suicide (age-adjusted rate) | 2,427 | deprivation | unweighted |
| 0.604 | 0.446 | 0.527 | Suicide (age-adjusted rate) | Suicide (age-adjusted rate) | 2,427 | rural | population-weighted |
| 0.438 | 0.323 | 0.374 | Suicide (age-adjusted rate) | Suicide (age-adjusted rate) | 2,427 | rural | unweighted |
| 0.225 | −0.105 | 0.083 | Alcohol-impaired driving (share of crash deaths) | Alcohol-impaired driving (share of crash deaths) | 3,102 | deprivation | population-weighted |
| 0.005 | −0.131 | −0.062 | Alcohol-impaired driving (share of crash deaths) | Alcohol-impaired driving (share of crash deaths) | 3,102 | deprivation | unweighted |
| 0.221 | −0.029 | 0.104 | Alcohol-impaired driving (share of crash deaths) | Alcohol-impaired driving (share of crash deaths) | 3,116 | rural | population-weighted |
| 0.099 | −0.029 | 0.032 | Alcohol-impaired driving (share of crash deaths) | Alcohol-impaired driving (share of crash deaths) | 3,116 | rural | unweighted |
| 0.677 | 0.478 | 0.600 | Firearm death, any intent (crude rate) | Firearm death, any intent (crude rate) | 2,279 | deprivation | population-weighted |
| 0.516 | 0.325 | 0.423 | Firearm death, any intent (crude rate) | Firearm death, any intent (crude rate) | 2,279 | deprivation | unweighted |
| 0.367 | 0.129 | 0.265 | Firearm death, any intent (crude rate) | Firearm death, any intent (crude rate) | 2,279 | rural | population-weighted |
| 0.425 | 0.294 | 0.359 | Firearm death, any intent (crude rate) | Firearm death, any intent (crude rate) | 2,279 | rural | unweighted |
| 0.482 | 0.342 | 0.413 | Homicide (crude rate) | Homicide (crude rate) | 1,376 | deprivation | population-weighted |
| 0.554 | 0.415 | 0.490 | Homicide (crude rate) | Homicide (crude rate) | 1,376 | deprivation | unweighted |
| 0.053 | −0.175 | −0.055 | Homicide (crude rate) | Homicide (crude rate) | 1,376 | rural | population-weighted |
| 0.345 | 0.121 | 0.246 | Homicide (crude rate) | Homicide (crude rate) | 1,376 | rural | unweighted |
| 0.705 | 0.562 | 0.624 | Motor-vehicle crash (crude rate) | Motor-vehicle crash (crude rate) | 2,688 | deprivation | population-weighted |
| 0.621 | 0.494 | 0.556 | Motor-vehicle crash (crude rate) | Motor-vehicle crash (crude rate) | 2,688 | deprivation | unweighted |
| 0.718 | 0.585 | 0.655 | Motor-vehicle crash (crude rate) | Motor-vehicle crash (crude rate) | 2,688 | rural | population-weighted |
| 0.643 | 0.566 | 0.602 | Motor-vehicle crash (crude rate) | Motor-vehicle crash (crude rate) | 2,688 | rural | unweighted |
| 0.756 | 0.602 | 0.693 | Child mortality (crude rate) | Child mortality (crude rate) | 2,038 | deprivation | population-weighted |
| 0.610 | 0.499 | 0.552 | Child mortality (crude rate) | Child mortality (crude rate) | 2,038 | deprivation | unweighted |
| 0.419 | 0.201 | 0.324 | Child mortality (crude rate) | Child mortality (crude rate) | 2,038 | rural | population-weighted |
| 0.489 | 0.352 | 0.423 | Child mortality (crude rate) | Child mortality (crude rate) | 2,038 | rural | unweighted |
| 0.827 | 0.737 | 0.781 | Premature mortality (age-adjusted rate) | Premature mortality (age-adjusted rate) | 3,083 | deprivation | population-weighted |
| 0.722 | 0.574 | 0.644 | Premature mortality (age-adjusted rate) | Premature mortality (age-adjusted rate) | 3,083 | deprivation | unweighted |
| 0.495 | 0.317 | 0.417 | Premature mortality (age-adjusted rate) | Premature mortality (age-adjusted rate) | 3,088 | rural | population-weighted |
| 0.327 | 0.183 | 0.251 | Premature mortality (age-adjusted rate) | Premature mortality (age-adjusted rate) | 3,088 | rural | unweighted |

Views: Deprivation, population-weighted (view = deprivation, weighting = population-weighted); Rural share, population-weighted (view = rural, weighting = population-weighted); Deprivation, unweighted (view = deprivation, weighting = unweighted); Rural share, unweighted (view = rural, weighting = unweighted).

Note: Each correlation uses the counties with a published rate for that measure (County Health Rankings leaves out a county with fewer than 10 deaths in the window, 20 for premature mortality), so the counties differ across rows and the number is in the tooltip. The 1,376 counties with a homicide rate are less rural than counties as a whole (43.3% against 64.1% mean rural share), because a county with few homicides gets no rate, so the homicide rows describe the counties that report and not all counties. Alcohol-impaired driving is a share of motor-vehicle crash deaths, not a rate per resident. Correlations across counties describe places, not people.

Note: Intervals are 95% intervals from a bootstrap that draws whole states with replacement (replicates and seed are in the methods note). Rural share is the County Health Rankings measure of the share of residents living outside urban areas. County deprivation and rural share correlate at +0.56 across counties, weighted by population.

Source: County Health Rankings 2024 release, NCHS death counts 2015-2021 in windows that differ by cause, crude rates for four measures, age-adjusted rates for two and a share of crash deaths for one, from full counts, population-weighted or unweighted as each view states; Area Deprivation Index 2023

Data: https://vishalsingh.org/zipcode/report/gun-homicide-and-causes-of-early-death/figures/cause_correlations.json

### Table 8. Overdose has the highest rate of the four causes in 829 of 1,248 counties, and homicide in only 11

Kind of death with the highest crude death rate among overdose, suicide, homicide and motor-vehicle crash, counties with all four rates published, County Health Rankings 2024 release

No inline rows: the values of Table 8 are in https://vishalsingh.org/zipcode/data/gun-homicide-and-causes-of-early-death/dominant_cause_map.json.

Note: Gray means at least one of the four rates is missing. County Health Rankings leaves out a county's rate when it has fewer than 10 deaths from that cause in the window, so a missing rate means too few deaths, not zero, and the gray counties are mostly small ones. The map answers only for the 1,248 counties with all four rates, which hold 89.6% of county residents.

Note: Counties are compared on rates computed over windows of different lengths (overdose 2019 to 2021, suicide 2017 to 2021, homicide and motor-vehicle crashes 2015 to 2021), and small counties have noisy rates, so a label in a county with few deaths can change with one year of data. Puerto Rico is not on the map. Connecticut is drawn as its eight legacy counties.

Source: County Health Rankings 2024 release, NCHS death counts 2015-2021 in windows that differ by cause, crude rates from full counts

Data: https://vishalsingh.org/zipcode/report/gun-homicide-and-causes-of-early-death/figures/dominant_cause_map.json

### Table 9. The premature death rate is steadily worse toward the more deprived end under the ADI and under each of the three other rankings

Premature deaths per 100,000 residents, age-adjusted by tenth of counties ranked four ways, over the 3,080 counties that have all four rankings

| bin | conf_high | conf_low | estimate | n_units | ranking |
| ---: | ---: | ---: | ---: | ---: | --- |
| 1 | 318.3 | 295.9 | 307.3 | 308 | Area Deprivation Index |
| 2 | 386.6 | 354.3 | 371.4 | 308 | Area Deprivation Index |
| 3 | 438.1 | 402.9 | 419.7 | 308 | Area Deprivation Index |
| 4 | 487.6 | 431.0 | 457.7 | 308 | Area Deprivation Index |
| 5 | 508.7 | 464.0 | 487.1 | 308 | Area Deprivation Index |
| 6 | 540.2 | 491.0 | 515.3 | 308 | Area Deprivation Index |
| 7 | 554.7 | 495.4 | 523.1 | 308 | Area Deprivation Index |
| 8 | 590.6 | 509.2 | 545.3 | 308 | Area Deprivation Index |
| 9 | 632.4 | 571.6 | 605.7 | 308 | Area Deprivation Index |
| 10 | 740.7 | 654.9 | 698.0 | 308 | Area Deprivation Index |
| 1 | 291.5 | 270.5 | 280 | 308 | Standardized index |
| 2 | 381.3 | 359.5 | 370.3 | 308 | Standardized index |
| 3 | 406.4 | 358.1 | 379.4 | 308 | Standardized index |
| 4 | 434.3 | 395.8 | 412.9 | 308 | Standardized index |
| 5 | 496.9 | 437.5 | 469.7 | 308 | Standardized index |
| 6 | 545.1 | 497.1 | 521.8 | 308 | Standardized index |
| 7 | 549.3 | 500.3 | 525.1 | 308 | Standardized index |
| 8 | 575.2 | 514.0 | 542.4 | 308 | Standardized index |
| 9 | 619.0 | 532.2 | 573.6 | 308 | Standardized index |
| 10 | 723.5 | 574.3 | 650.1 | 308 | Standardized index |
| 1 | 300.4 | 284.3 | 290.6 | 308 | Median household income |
| 2 | 378.0 | 351.9 | 363.8 | 308 | Median household income |
| 3 | 416.3 | 386.1 | 400.2 | 308 | Median household income |
| 4 | 451.2 | 415.8 | 433.8 | 308 | Median household income |
| 5 | 500.4 | 463.9 | 483.1 | 308 | Median household income |
| 6 | 533.1 | 492.5 | 511.2 | 308 | Median household income |
| 7 | 559.8 | 513.9 | 538.9 | 308 | Median household income |
| 8 | 586.0 | 522.5 | 552.7 | 308 | Median household income |
| 9 | 636.7 | 596.5 | 618.2 | 308 | Median household income |
| 10 | 708.7 | 647.3 | 679.8 | 308 | Median household income |
| 1 | 315.5 | 282.1 | 296.7 | 308 | Bachelor's degree share |
| 2 | 389.6 | 357.5 | 371.7 | 308 | Bachelor's degree share |
| 3 | 453.0 | 381.2 | 414.2 | 308 | Bachelor's degree share |
| 4 | 463.8 | 422.8 | 442.2 | 308 | Bachelor's degree share |
| 5 | 502.6 | 445.6 | 477.7 | 308 | Bachelor's degree share |
| 6 | 508.9 | 460.7 | 480.1 | 308 | Bachelor's degree share |
| 7 | 547.0 | 489.3 | 516.1 | 308 | Bachelor's degree share |
| 8 | 571.7 | 517.4 | 545.2 | 308 | Bachelor's degree share |
| 9 | 619.5 | 534.8 | 579.0 | 308 | Bachelor's degree share |
| 10 | 655.8 | 597.4 | 627.1 | 308 | Bachelor's degree share |

Note: Least and most deprived tenth: ADI 307 and 698, rank correlation +1.00; standardized index 280 and 650, +1.00; median household income 291 and 680, +1.00; bachelor's degree share 297 and 627, +1.00.

Note: The standardized index applies the ADI's coefficients to its measures after standardizing each one, so home value and income carry less weight than in the ADI. Income and degree share are single census measures, each ranked from the highest value to the lowest.

Source: County Health Rankings 2024 release, NCHS death counts 2019-2021, age-adjusted rates from full counts, population-weighted; Area Deprivation Index 2023; ADI 2023 and ACS 2019-2023 tract tables (B19013, B15003 and the 15 rebuilt ADI measures)

Data: https://vishalsingh.org/zipcode/report/gun-homicide-and-causes-of-early-death/figures/alt_rankings.json

## All estimates

Every fact in facts.json (112), sorted by id.

| Fact id | Value | Unit | Estimand | Population | Period | Interval | n | Weights | Source | Note |
| --- | ---: | --- | --- | --- | --- | --- | ---: | --- | --- | --- |
| adi_rural_r | +0.56 | correlation coefficient | population-weighted correlation of county ADI national rank with rural share; state-cluster bootstrap interval | residents of counties with an ADI rank and a rural share | ADI 2023 release; County Health Rankings 2024 release (rural share of residents) | 95% CI 0.48 to 0.63 | 3,128 counties | county population | County Health Rankings |   |
| all_rural | 64.1% | percent of residents living in rural areas | simple mean over counties of the rural share of residents | all counties in the file | County Health Rankings 2024 release (rural share of residents) |   | 3,143 counties | none (each county counts once) | County Health Rankings |   |
| alt_adi_least | 307 | Premature deaths per 100,000 residents, age-adjusted | population-weighted mean of county age-adjusted premature mortality rates; 95% intervals from a bootstrap over states | the least deprived tenth of counties ranked by Area Deprivation Index, among the counties with all four rankings | deaths in 2019 to 2021 (County Health Rankings 2024 release); rankings from ACS 2019-2023 | 95% CI 296 to 318 | 308 counties | county population | County Health Rankings |   |
| alt_adi_most | 698 | Premature deaths per 100,000 residents, age-adjusted | population-weighted mean of county age-adjusted premature mortality rates; 95% intervals from a bootstrap over states | the most deprived tenth of counties ranked by Area Deprivation Index, among the counties with all four rankings | deaths in 2019 to 2021 (County Health Rankings 2024 release); rankings from ACS 2019-2023 | 95% CI 655 to 741 | 308 counties | county population | County Health Rankings |   |
| alt_adi_rho | +1.00 | rank correlation | Spearman rank correlation between bin order and bin mean, each bin counted once | the bins of counties ranked by Area Deprivation Index | deaths in 2019 to 2021 (County Health Rankings 2024 release); rankings from ACS 2019-2023 |   |   | each bin counted once | County Health Rankings |   |
| alt_degree_least | 297 | Premature deaths per 100,000 residents, age-adjusted | population-weighted mean of county age-adjusted premature mortality rates; 95% intervals from a bootstrap over states | the least deprived tenth of counties ranked by Bachelor's degree share, among the counties with all four rankings | deaths in 2019 to 2021 (County Health Rankings 2024 release); rankings from ACS 2019-2023 | 95% CI 282 to 315 | 308 counties | county population | County Health Rankings |   |
| alt_degree_most | 627 | Premature deaths per 100,000 residents, age-adjusted | population-weighted mean of county age-adjusted premature mortality rates; 95% intervals from a bootstrap over states | the most deprived tenth of counties ranked by Bachelor's degree share, among the counties with all four rankings | deaths in 2019 to 2021 (County Health Rankings 2024 release); rankings from ACS 2019-2023 | 95% CI 597 to 656 | 308 counties | county population | County Health Rankings |   |
| alt_degree_rho | +1.00 | rank correlation | Spearman rank correlation between bin order and bin mean, each bin counted once | the bins of counties ranked by Bachelor's degree share | deaths in 2019 to 2021 (County Health Rankings 2024 release); rankings from ACS 2019-2023 |   |   | each bin counted once | County Health Rankings |   |
| alt_income_least | 291 | Premature deaths per 100,000 residents, age-adjusted | population-weighted mean of county age-adjusted premature mortality rates; 95% intervals from a bootstrap over states | the least deprived tenth of counties ranked by Median household income, among the counties with all four rankings | deaths in 2019 to 2021 (County Health Rankings 2024 release); rankings from ACS 2019-2023 | 95% CI 284 to 300 | 308 counties | county population | County Health Rankings |   |
| alt_income_most | 680 | Premature deaths per 100,000 residents, age-adjusted | population-weighted mean of county age-adjusted premature mortality rates; 95% intervals from a bootstrap over states | the most deprived tenth of counties ranked by Median household income, among the counties with all four rankings | deaths in 2019 to 2021 (County Health Rankings 2024 release); rankings from ACS 2019-2023 | 95% CI 647 to 709 | 308 counties | county population | County Health Rankings |   |
| alt_income_rho | +1.00 | rank correlation | Spearman rank correlation between bin order and bin mean, each bin counted once | the bins of counties ranked by Median household income | deaths in 2019 to 2021 (County Health Rankings 2024 release); rankings from ACS 2019-2023 |   |   | each bin counted once | County Health Rankings |   |
| alt_n_places | 3,080 | counties | count of counties in the common sample of the ranking comparison | counties with the outcome and all four rankings | deaths in 2019 to 2021 (County Health Rankings 2024 release); rankings from ACS 2019-2023 |   |   | none (full count) | County Health Rankings |   |
| alt_rho_cut | 0.9 | absolute rank correlation | threshold between a steady and an uneven gradient | the rule for changed gradients (docs/RANKING_SWITCH_RULE.md) | deaths in 2019 to 2021 (County Health Rankings 2024 release); rankings from ACS 2019-2023 |   |   | none (a fixed rule) | County Health Rankings |   |
| alt_rho_margin | 0.1 | absolute rank correlation | smallest move in the rank correlation that counts as a change of shape | the rule for changed gradients (docs/RANKING_SWITCH_RULE.md) | deaths in 2019 to 2021 (County Health Rankings 2024 release); rankings from ACS 2019-2023 |   |   | none (a fixed rule) | County Health Rankings |   |
| alt_std_least | 280 | Premature deaths per 100,000 residents, age-adjusted | population-weighted mean of county age-adjusted premature mortality rates; 95% intervals from a bootstrap over states | the least deprived tenth of counties ranked by Standardized index, among the counties with all four rankings | deaths in 2019 to 2021 (County Health Rankings 2024 release); rankings from ACS 2019-2023 | 95% CI 270 to 291 | 308 counties | county population | County Health Rankings |   |
| alt_std_most | 650 | Premature deaths per 100,000 residents, age-adjusted | population-weighted mean of county age-adjusted premature mortality rates; 95% intervals from a bootstrap over states | the most deprived tenth of counties ranked by Standardized index, among the counties with all four rankings | deaths in 2019 to 2021 (County Health Rankings 2024 release); rankings from ACS 2019-2023 | 95% CI 574 to 723 | 308 counties | county population | County Health Rankings |   |
| alt_std_rho | +1.00 | rank correlation | Spearman rank correlation between bin order and bin mean, each bin counted once | the bins of counties ranked by Standardized index | deaths in 2019 to 2021 (County Health Rankings 2024 release); rankings from ACS 2019-2023 |   |   | each bin counted once | County Health Rankings |   |
| axis_stop | 50% | percent of residents | where the horizontal axis of the county curves stops (definitional) | residents of each of the eight counties | annual rates for 2022, 2023 and 2024, averaged |   |   | none (definition) | CDC injury and violence mapping |   |
| chi_max_fa | 89 | firearm homicide deaths per 100,000 residents per year | highest mean annual rate of any cluster | Cook County tract clusters shown on the map | annual rates for 2022, 2023 and 2024, averaged |   |   | none (single highest cluster) | CDC injury and violence mapping |   |
| chi_n_gray | 168 | tracts | tracts with no cluster rate shown: CDC withholds the rate in at least one year, or the tract has no residents | Cook County tracts on the map | annual rates for 2022, 2023 and 2024, averaged |   |   | none (full count) | CDC injury and violence mapping |   |
| chi_n_tracts | 1,163 | tracts | tracts shown with a cluster rate | Cook County tracts on the map | annual rates for 2022, 2023 and 2024, averaged |   |   | none (full count) | CDC injury and violence mapping |   |
| chi_top_south | 90.8% | percent of residents | share living in tracts whose centroid lies south of the Chicago Loop (a tract centroid south of the Loop's position on the map) | Cook County residents in the tenth living in the highest-rate clusters | annual rates for 2022, 2023 and 2024, averaged |   | 174 tracts | ACS 2019-2023 tract population | CDC injury and violence mapping |   |
| chi_zero_pct | 24.2% | percent of tracts | share of tracts whose cluster has zero recorded deaths in each of the three years | Cook County tracts shown with a cluster rate | annual rates for 2022, 2023 and 2024, averaged |   |   | none (each tract counts once) | CDC injury and violence mapping |   |
| chr_floor | 10 | deaths in the window | County Health Rankings leaves a county's rate out when it has fewer deaths than this (definitional; checked: the smallest published count is 10) | counties in the file | County Health Rankings 2024 release; deaths pooled over windows that end in 2021 and differ by cause (overdose and premature mortality 2019 to 2021, suicide and firearm deaths 2017 to 2021, homicide and motor-vehicle crashes 2015 to 2021) |   |   | none (definition) | County Health Rankings |   |
| chr_floor_pmm | 20 | deaths in the window | the same floor for premature mortality (definitional; checked: the smallest published count is 20) | counties in the file | County Health Rankings 2024 release; deaths pooled over windows that end in 2021 and differ by cause (overdose and premature mortality 2019 to 2021, suicide and firearm deaths 2017 to 2021, homicide and motor-vehicle crashes 2015 to 2021) |   |   | none (definition) | County Health Rankings |   |
| clu_tract_max | 20 | tracts per cluster | largest number of tracts in one cluster | tracts outside New York City | annual rates for 2022, 2023 and 2024, averaged |   |   | none (single largest cluster) | CDC injury and violence mapping |   |
| clu_tract_median | 5 | tracts per cluster | median, over tracts, of the number of tracts in the tract's cluster | tracts outside New York City | annual rates for 2022, 2023 and 2024, averaged |   |   | none (each tract counts once) | CDC injury and violence mapping |   |
| cook_cov | 87.7% | percent of residents | share living in a cluster with a published rate in all three years | residents of Cook County | annual rates for 2022, 2023 and 2024, averaged |   |   | none (full count) | CDC injury and violence mapping |   |
| cook_half | 10.5% | percent of residents | share of residents in the highest-rate clusters that hold half of implied firearm homicide deaths; clusters resampled with replacement | residents of Cook County tract clusters with a published rate in each of 2022, 2023 and 2024 | annual rates for 2022, 2023 and 2024, averaged | 95% CI 9.1% to 12.2% | 279 tract clusters | ACS 2019-2023 residents of each tract cluster (rate times residents gives implied deaths) | CDC injury and violence mapping |   |
| cook_le_gap | 9.5 | years of life expectancy at birth | difference of two population-weighted means of tract life expectancy; clusters resampled with replacement | residents of Cook County tract clusters with zero recorded deaths, minus residents in the highest-rate tenth | deaths in 2010 to 2015 | 95% CI 8.6 to 10.4 | 106 tract clusters | 2020 tract population (ACS 2019-2023) | USALEEP life expectancy |   |
| cook_top_adi | 74 | Area Deprivation Index national rank, 1 least to 100 most deprived | population-weighted mean of tract ranks; clusters resampled with replacement | residents of Cook County in the tenth living in the highest-rate clusters | ADI 2023 release, built on ACS 2019-2023 | 95% CI 71 to 78 | 26 tract clusters | ACS 2019-2023 tract population | Area Deprivation Index |   |
| cook_top_fa | 60 | firearm homicide deaths per 100,000 residents per year | population-weighted mean of cluster rates; clusters resampled with replacement | residents of Cook County in the tenth of residents living in the highest-rate clusters | annual rates for 2022, 2023 and 2024, averaged | 95% CI 56 to 65 | 26 tract clusters | ACS 2019-2023 residents of each tract cluster (rate times residents gives implied deaths) | CDC injury and violence mapping |   |
| cook_top_le | 71.7 | years of life expectancy at birth | population-weighted mean of tract life expectancy, institutional tracts dropped; clusters resampled with replacement | residents of Cook County in the tenth living in the highest-rate clusters | deaths in 2010 to 2015 | 95% CI 71.0 to 72.4 | 26 tract clusters | 2020 tract population (ACS 2019-2023) | USALEEP life expectancy |   |
| cook_zero | 27.0% | percent of residents | share of residents in clusters with zero recorded deaths in each of the three years; clusters resampled with replacement | residents of Cook County tract clusters with a published rate in each of 2022, 2023 and 2024 | annual rates for 2022, 2023 and 2024, averaged | 95% CI 22.1% to 32.2% | 279 tract clusters | ACS 2019-2023 residents of each tract cluster (rate times residents gives implied deaths) | CDC injury and violence mapping |   |
| cook_zero_le | 81.2 | years of life expectancy at birth | population-weighted mean of tract life expectancy, institutional tracts dropped; clusters resampled with replacement | residents of Cook County tract clusters with zero recorded deaths in each of the three years | deaths in 2010 to 2015 | 95% CI 80.7 to 81.7 | 80 tract clusters | 2020 tract population (ACS 2019-2023) | USALEEP life expectancy |   |
| county_max_half | 24.4% | percent of residents | share of residents in the highest-rate clusters that hold half of implied firearm homicide deaths; clusters resampled with replacement | residents of tract clusters in the county with the largest share among the 8, with a modeled rate in each year | annual rates for 2022, 2023 and 2024, averaged | 95% CI 20.4% to 29.3% | 46 tract clusters | ACS 2019-2023 residents of each tract cluster (rate times residents gives implied deaths) | CDC injury and violence mapping |   |
| cty_at50_max | 95.7% | percent of implied firearm homicide deaths | cumulative share of deaths at the stop of the horizontal axis; largest of the 8 counties; clusters resampled with replacement | residents of Cook County, IL in clusters with a published rate in every year | annual rates for 2022, 2023 and 2024, averaged | 95% CI 94.0% to 97.0% | 279 tract clusters | ACS 2019-2023 residents of each tract cluster (rate times residents gives implied deaths) | CDC injury and violence mapping |   |
| cty_at50_min | 79.9% | percent of implied firearm homicide deaths | cumulative share of deaths at the stop of the horizontal axis; smallest of the 8 counties; clusters resampled with replacement | residents of Harris County, TX in clusters with a published rate in every year | annual rates for 2022, 2023 and 2024, averaged | 95% CI 76.9% to 83.1% | 267 tract clusters | ACS 2019-2023 residents of each tract cluster (rate times residents gives implied deaths) | CDC injury and violence mapping |   |
| cty_cov_min | 80.2% | percent of residents | smallest share of a county's residents who live in a cluster with a published rate in all three years | residents of the 8 counties | annual rates for 2022, 2023 and 2024, averaged |   |   | none (full count) | CDC injury and violence mapping |   |
| dom_hom | 11 | counties | counties where homicide has the highest of the four crude rates | counties with all four cause rates published | County Health Rankings 2024 release; deaths pooled over windows that end in 2021 and differ by cause (overdose and premature mortality 2019 to 2021, suicide and firearm deaths 2017 to 2021, homicide and motor-vehicle crashes 2015 to 2021) |   |   | none (full count) | County Health Rankings |   |
| dom_hom_black | 56.1% | percent of residents who are non-Hispanic Black | median county share | the counties where homicide has the highest of the four rates | ACS 2017-2021 as reported by County Health Rankings 2024 |   | 11 counties | none (each county counts once) | County Health Rankings |   |
| dom_mv | 201 | counties | counties where motor-vehicle crash has the highest of the four crude rates | counties with all four cause rates published | County Health Rankings 2024 release; deaths pooled over windows that end in 2021 and differ by cause (overdose and premature mortality 2019 to 2021, suicide and firearm deaths 2017 to 2021, homicide and motor-vehicle crashes 2015 to 2021) |   |   | none (full count) | County Health Rankings |   |
| dom_od | 829 | counties | counties where overdose has the highest of the four crude rates | counties with all four cause rates published | County Health Rankings 2024 release; deaths pooled over windows that end in 2021 and differ by cause (overdose and premature mortality 2019 to 2021, suicide and firearm deaths 2017 to 2021, homicide and motor-vehicle crashes 2015 to 2021) |   |   | none (full count) | County Health Rankings |   |
| dom_suic | 207 | counties | counties where suicide has the highest of the four crude rates | counties with all four cause rates published | County Health Rankings 2024 release; deaths pooled over windows that end in 2021 and differ by cause (overdose and premature mortality 2019 to 2021, suicide and firearm deaths 2017 to 2021, homicide and motor-vehicle crashes 2015 to 2021) |   |   | none (full count) | County Health Rankings |   |
| hom_black | 27.3 | homicide deaths per 100,000 non-Hispanic Black residents (crude rate) | mean of county race-specific rates weighted by group residents; state-cluster bootstrap interval; a composite of reporting counties, not a national rate | residents of the 486 counties that publish both a Black and a white rate for homicide, each group weighted by its residents | County Health Rankings 2024 release; deaths pooled over windows that end in 2021 and differ by cause (overdose and premature mortality 2019 to 2021, suicide and firearm deaths 2017 to 2021, homicide and motor-vehicle crashes 2015 to 2021) | 95% CI 23.8 to 31.8 | 486 counties | county population times the group's share of residents | County Health Rankings |   |
| hom_pub_rural | 43.3% | percent of residents living in rural areas | simple mean over counties of the rural share of residents | counties with a published homicide rate | County Health Rankings 2024 release (rural share of residents) |   | 1,376 counties | none (each county counts once) | County Health Rankings |   |
| hom_radi | +0.41 | correlation coefficient | population-weighted correlation of the county rate with county ADI national rank; state-cluster bootstrap interval | residents of counties with a published homicide rate | County Health Rankings 2024 release; deaths pooled over windows that end in 2021 and differ by cause (overdose and premature mortality 2019 to 2021, suicide and firearm deaths 2017 to 2021, homicide and motor-vehicle crashes 2015 to 2021) | 95% CI 0.34 to 0.48 | 1,376 counties | county population | County Health Rankings |   |
| hom_radi_unw | +0.49 | correlation coefficient | unweighted correlation of the county rate with county ADI national rank; state-cluster bootstrap interval | the counties with a published homicide rate | County Health Rankings 2024 release; deaths pooled over windows that end in 2021 and differ by cause (overdose and premature mortality 2019 to 2021, suicide and firearm deaths 2017 to 2021, homicide and motor-vehicle crashes 2015 to 2021) | 95% CI 0.42 to 0.55 | 1,376 counties | none (each county counts once) | County Health Rankings |   |
| hom_ratio | 9.6 | ratio of Black to white homicide rates | ratio of the two group means; state-cluster bootstrap interval | residents of the 486 counties that publish both a Black and a white rate for homicide, each group weighted by its residents | County Health Rankings 2024 release; deaths pooled over windows that end in 2021 and differ by cause (overdose and premature mortality 2019 to 2021, suicide and firearm deaths 2017 to 2021, homicide and motor-vehicle crashes 2015 to 2021) | 95% CI 8.1 to 11.6 | 486 counties | county population times the group's share of residents | County Health Rankings |   |
| hom_rrural | −0.05 | correlation coefficient | population-weighted correlation of the county rate with county rural share of residents; state-cluster bootstrap interval | residents of counties with a published homicide rate | County Health Rankings 2024 release; deaths pooled over windows that end in 2021 and differ by cause (overdose and premature mortality 2019 to 2021, suicide and firearm deaths 2017 to 2021, homicide and motor-vehicle crashes 2015 to 2021) | 95% CI −0.18 to 0.05 | 1,376 counties | county population | County Health Rankings |   |
| hom_rrural_unw | +0.25 | correlation coefficient | unweighted correlation of the county rate with county rural share of residents; state-cluster bootstrap interval | the counties with a published homicide rate | County Health Rankings 2024 release; deaths pooled over windows that end in 2021 and differ by cause (overdose and premature mortality 2019 to 2021, suicide and firearm deaths 2017 to 2021, homicide and motor-vehicle crashes 2015 to 2021) | 95% CI 0.12 to 0.35 | 1,376 counties | none (each county counts once) | County Health Rankings |   |
| hom_sh_gap | +4.3 percentage points | percentage points of the four-kind crude death rate (homicide), most deprived fifth minus least deprived fifth | difference between the cause's share in the two fifths; state-cluster bootstrap interval from the same replicates | residents of counties in the least and most deprived fifths of residents among counties with all four rates published | County Health Rankings 2024 release; deaths pooled over windows that end in 2021 and differ by cause (overdose and premature mortality 2019 to 2021, suicide and firearm deaths 2017 to 2021, homicide and motor-vehicle crashes 2015 to 2021) | 95% CI 2.2 percentage points to 7.5 percentage points | 689 counties | county population | County Health Rankings |   |
| hom_sh_q1 | 8.8% | percent of the four-kind crude death rate (homicide) | crude rate of one kind of death as a share of the sum of the four means (overdose, suicide, homicide, motor-vehicle crash); state-cluster bootstrap interval | residents of counties in the least deprived fifth of residents among counties with all four rates published | County Health Rankings 2024 release; deaths pooled over windows that end in 2021 and differ by cause (overdose and premature mortality 2019 to 2021, suicide and firearm deaths 2017 to 2021, homicide and motor-vehicle crashes 2015 to 2021) | 95% CI 5.9% to 10.2% | 71 counties | county population | County Health Rankings |   |
| hom_sh_q5 | 13.1% | percent of the four-kind crude death rate (homicide) | crude rate of one kind of death as a share of the sum of the four means (overdose, suicide, homicide, motor-vehicle crash); state-cluster bootstrap interval | residents of counties in the most deprived fifth of residents among counties with all four rates published | County Health Rankings 2024 release; deaths pooled over windows that end in 2021 and differ by cause (overdose and premature mortality 2019 to 2021, suicide and firearm deaths 2017 to 2021, homicide and motor-vehicle crashes 2015 to 2021) | 95% CI 11.7% to 14.7% | 618 counties | county population | County Health Rankings |   |
| hom_white | 2.8 | homicide deaths per 100,000 non-Hispanic white residents (crude rate) | mean of county race-specific rates weighted by group residents; state-cluster bootstrap interval; a composite of reporting counties, not a national rate | residents of the 486 counties that publish both a Black and a white rate for homicide, each group weighted by its residents | County Health Rankings 2024 release; deaths pooled over windows that end in 2021 and differ by cause (overdose and premature mortality 2019 to 2021, suicide and firearm deaths 2017 to 2021, homicide and motor-vehicle crashes 2015 to 2021) | 95% CI 2.5 to 3.2 | 486 counties | county population times the group's share of residents | County Health Rankings |   |
| mix_black | 7.1% | percent of residents who are non-Hispanic Black | median county share | counties with all four cause rates published | ACS 2017-2021 as reported by County Health Rankings 2024 |   | 1,248 counties | none (each county counts once) | County Health Rankings |   |
| mix_pop | 89.6% | percent of county residents | share of residents living in counties with all four rates published | residents of the 3143 counties in the file | County Health Rankings 2024 release; deaths pooled over windows that end in 2021 and differ by cause (overdose and premature mortality 2019 to 2021, suicide and firearm deaths 2017 to 2021, homicide and motor-vehicle crashes 2015 to 2021) |   |   | county population | County Health Rankings |   |
| mv_rrural | +0.66 | correlation coefficient | population-weighted correlation of the county rate with county rural share of residents; state-cluster bootstrap interval | residents of counties with a published motor-vehicle crash rate | County Health Rankings 2024 release; deaths pooled over windows that end in 2021 and differ by cause (overdose and premature mortality 2019 to 2021, suicide and firearm deaths 2017 to 2021, homicide and motor-vehicle crashes 2015 to 2021) | 95% CI 0.58 to 0.72 | 2,688 counties | county population | County Health Rankings |   |
| mv_rrural_unw | +0.60 | correlation coefficient | unweighted correlation of the county rate with county rural share of residents; state-cluster bootstrap interval | the counties with a published motor-vehicle crash rate | County Health Rankings 2024 release; deaths pooled over windows that end in 2021 and differ by cause (overdose and premature mortality 2019 to 2021, suicide and firearm deaths 2017 to 2021, homicide and motor-vehicle crashes 2015 to 2021) | 95% CI 0.57 to 0.64 | 2,688 counties | none (each county counts once) | County Health Rankings |   |
| mv_sh_gap | +5.4 percentage points | percentage points of the four-kind crude death rate (motor-vehicle), most deprived fifth minus least deprived fifth | difference between the cause's share in the two fifths; state-cluster bootstrap interval from the same replicates | residents of counties in the least and most deprived fifths of residents among counties with all four rates published | County Health Rankings 2024 release; deaths pooled over windows that end in 2021 and differ by cause (overdose and premature mortality 2019 to 2021, suicide and firearm deaths 2017 to 2021, homicide and motor-vehicle crashes 2015 to 2021) | 95% CI 1.8 percentage points to 10.5 percentage points | 689 counties | county population | County Health Rankings |   |
| mv_sh_q1 | 16.0% | percent of the four-kind crude death rate (motor-vehicle) | crude rate of one kind of death as a share of the sum of the four means (overdose, suicide, homicide, motor-vehicle crash); state-cluster bootstrap interval | residents of counties in the least deprived fifth of residents among counties with all four rates published | County Health Rankings 2024 release; deaths pooled over windows that end in 2021 and differ by cause (overdose and premature mortality 2019 to 2021, suicide and firearm deaths 2017 to 2021, homicide and motor-vehicle crashes 2015 to 2021) | 95% CI 12.6% to 17.9% | 71 counties | county population | County Health Rankings |   |
| mv_sh_q5 | 21.4% | percent of the four-kind crude death rate (motor-vehicle) | crude rate of one kind of death as a share of the sum of the four means (overdose, suicide, homicide, motor-vehicle crash); state-cluster bootstrap interval | residents of counties in the most deprived fifth of residents among counties with all four rates published | County Health Rankings 2024 release; deaths pooled over windows that end in 2021 and differ by cause (overdose and premature mortality 2019 to 2021, suicide and firearm deaths 2017 to 2021, homicide and motor-vehicle crashes 2015 to 2021) | 95% CI 18.4% to 24.6% | 618 counties | county population | County Health Rankings |   |
| n_curve_counties | 8 | counties | the counties with the most implied firearm homicide deaths a year, which get a curve | counties outside New York City | annual rates for 2022, 2023 and 2024, averaged |   |   | none (full count) | CDC injury and violence mapping |   |
| n_dom | 1,248 | counties | count of counties with all four rates published | counties with all four cause rates published | County Health Rankings 2024 release; deaths pooled over windows that end in 2021 and differ by cause (overdose and premature mortality 2019 to 2021, suicide and firearm deaths 2017 to 2021, homicide and motor-vehicle crashes 2015 to 2021) |   |   | none (full count) | County Health Rankings |   |
| n_hom_pub | 1,376 | counties | counties with a published homicide rate | counties in the file | County Health Rankings 2024 release; deaths pooled over windows that end in 2021 and differ by cause (overdose and premature mortality 2019 to 2021, suicide and firearm deaths 2017 to 2021, homicide and motor-vehicle crashes 2015 to 2021) |   |   | none (full count) | County Health Rankings |   |
| n_mix | 1,248 | counties | count of counties in the cause mix | counties with all four cause rates published and an ADI rank | County Health Rankings 2024 release; deaths pooled over windows that end in 2021 and differ by cause (overdose and premature mortality 2019 to 2021, suicide and firearm deaths 2017 to 2021, homicide and motor-vehicle crashes 2015 to 2021) |   |   | none (full count) | County Health Rankings |   |
| n_race_hom | 486 | counties | count of counties in the comparison | counties that publish both a Black and a white homicide rate | County Health Rankings 2024 release; deaths pooled over windows that end in 2021 and differ by cause (overdose and premature mortality 2019 to 2021, suicide and firearm deaths 2017 to 2021, homicide and motor-vehicle crashes 2015 to 2021) |   |   | none (full count) | County Health Rankings |   |
| n_race_suic | 297 | counties | count of counties in the comparison | counties that publish both a Black and a white suicide rate | County Health Rankings 2024 release; deaths pooled over windows that end in 2021 and differ by cause (overdose and premature mortality 2019 to 2021, suicide and firearm deaths 2017 to 2021, homicide and motor-vehicle crashes 2015 to 2021) |   |   | none (full count) | County Health Rankings |   |
| nat_covered | 308.7 million | residents | sum of cluster residents | residents of tract clusters outside New York City with a published rate in each of 2022, 2023 and 2024 | annual rates for 2022, 2023 and 2024, averaged |   |   | none (full count) | CDC injury and violence mapping |   |
| nat_fill_max | 10.2% | percent of residents | sensitivity: share holding half of deaths when a withheld year takes the lowest, middle or highest deaths in its published range of 1 to 9; largest of the three | residents of tract clusters outside New York City, withheld years filled | annual rates for 2022, 2023 and 2024, averaged |   |   | ACS 2019-2023 residents of each tract cluster (rate times residents gives implied deaths) | CDC injury and violence mapping |   |
| nat_fill_min | 8.9% | percent of residents | sensitivity: share holding half of deaths when a withheld year takes the lowest, middle or highest deaths in its published range of 1 to 9; smallest of the three | residents of tract clusters outside New York City, withheld years filled | annual rates for 2022, 2023 and 2024, averaged |   |   | ACS 2019-2023 residents of each tract cluster (rate times residents gives implied deaths) | CDC injury and violence mapping |   |
| nat_half | 9.5% | percent of residents | share of residents in the highest-rate clusters that together hold half of implied firearm homicide deaths; clusters ranked by mean annual rate; county bootstrap interval | residents of tract clusters outside New York City with a published rate in each of 2022, 2023 and 2024 | annual rates for 2022, 2023 and 2024, averaged | 95% CI 8.3% to 10.8% | 17,814 tract clusters | ACS 2019-2023 residents of each tract cluster (rate times residents gives implied deaths) | CDC injury and violence mapping |   |
| nat_half_excl | 9.6% | percent of residents | sensitivity: share holding half of implied deaths after dropping the clusters whose main county is undercounted; no interval | residents of tract clusters outside New York City with a published rate in every year, without the clusters of the 4 undercounted counties | annual rates for 2022, 2023 and 2024, averaged |   |   | ACS 2019-2023 residents of each tract cluster (rate times residents gives implied deaths) | CDC injury and violence mapping |   |
| nat_n_clusters | 17,814 | tract clusters | count of clusters in the curve | residents of tract clusters outside New York City with a published rate in each of 2022, 2023 and 2024 | annual rates for 2022, 2023 and 2024, averaged |   |   | none (full count) | CDC injury and violence mapping |   |
| nat_redraw_hi | 9.8% | percent of residents | check: share of residents holding half of deaths when every modeled cluster-year rate is redrawn from CDC's published 95% interval (log-normal, 500 replicates); 97.5th percentile | residents of tract clusters outside New York City with a published rate in each of 2022, 2023 and 2024 | annual rates for 2022, 2023 and 2024, averaged |   |   | ACS 2019-2023 residents of each tract cluster (rate times residents gives implied deaths) | CDC injury and violence mapping |   |
| nat_redraw_lo | 9.6% | percent of residents | check: share of residents holding half of deaths when every modeled cluster-year rate is redrawn from CDC's published 95% interval (log-normal, 500 replicates); 2.5th percentile | residents of tract clusters outside New York City with a published rate in each of 2022, 2023 and 2024 | annual rates for 2022, 2023 and 2024, averaged |   |   | ACS 2019-2023 residents of each tract cluster (rate times residents gives implied deaths) | CDC injury and violence mapping |   |
| nat_three_q | 24.1% | percent of residents | share of residents in the highest-rate clusters that together hold three quarters of implied firearm homicide deaths; clusters ranked by mean annual rate; county bootstrap interval | residents of tract clusters outside New York City with a published rate in each of 2022, 2023 and 2024 | annual rates for 2022, 2023 and 2024, averaged | 95% CI 22.2% to 26.0% | 17,814 tract clusters | ACS 2019-2023 residents of each tract cluster (rate times residents gives implied deaths) | CDC injury and violence mapping |   |
| nat_top10 | 51.3% | percent of implied firearm homicide deaths | share of deaths in the clusters that hold the highest-rate tenth of residents; county bootstrap interval | residents of tract clusters outside New York City with a published rate in each of 2022, 2023 and 2024 | annual rates for 2022, 2023 and 2024, averaged | 95% CI 48.1% to 54.4% | 17,814 tract clusters | ACS 2019-2023 residents of each tract cluster (rate times residents gives implied deaths) | CDC injury and violence mapping |   |
| nat_tract_ratio | 78.6% | percent of implied deaths | implied annual deaths (rate times ACS residents) from the cluster rates as a share of the county file's implied deaths for the same residents (county rate times the residents of the same clusters) | tract clusters with a published rate and the residents of those clusters, in counties outside New York City with a county-file rate | annual rates for 2022, 2023 and 2024, averaged |   |   | ACS 2019-2023 residents | CDC injury and violence mapping |   |
| nat_withheld | 4.7% | percent of residents | share living in clusters where CDC withholds the rate in at least one of the three years | residents of tract clusters outside New York City | annual rates for 2022, 2023 and 2024, averaged |   |   | none (full count) | CDC injury and violence mapping |   |
| nat_zero | 30.7% | percent of residents | share of residents in clusters with zero recorded deaths in each of the three years (a rate of 0 from a count of 0); county bootstrap interval | residents of tract clusters outside New York City with a published rate in each of 2022, 2023 and 2024 | annual rates for 2022, 2023 and 2024, averaged | 95% CI 29.0% to 32.6% | 17,814 tract clusters | ACS 2019-2023 residents of each tract cluster (rate times residents gives implied deaths) | CDC injury and violence mapping |   |
| od_sh_gap | −7.4 percentage points | percentage points of the four-kind crude death rate (overdose), most deprived fifth minus least deprived fifth | difference between the cause's share in the two fifths; state-cluster bootstrap interval from the same replicates | residents of counties in the least and most deprived fifths of residents among counties with all four rates published | County Health Rankings 2024 release; deaths pooled over windows that end in 2021 and differ by cause (overdose and premature mortality 2019 to 2021, suicide and firearm deaths 2017 to 2021, homicide and motor-vehicle crashes 2015 to 2021) | 95% CI −15.5 percentage points to −1.3 percentage points | 689 counties | county population | County Health Rankings |   |
| od_sh_q1 | 51.0% | percent of the four-kind crude death rate (overdose) | crude rate of one kind of death as a share of the sum of the four means (overdose, suicide, homicide, motor-vehicle crash); state-cluster bootstrap interval | residents of counties in the least deprived fifth of residents among counties with all four rates published | County Health Rankings 2024 release; deaths pooled over windows that end in 2021 and differ by cause (overdose and premature mortality 2019 to 2021, suicide and firearm deaths 2017 to 2021, homicide and motor-vehicle crashes 2015 to 2021) | 95% CI 46.9% to 57.6% | 71 counties | county population | County Health Rankings |   |
| od_sh_q5 | 43.6% | percent of the four-kind crude death rate (overdose) | crude rate of one kind of death as a share of the sum of the four means (overdose, suicide, homicide, motor-vehicle crash); state-cluster bootstrap interval | residents of counties in the most deprived fifth of residents among counties with all four rates published | County Health Rankings 2024 release; deaths pooled over windows that end in 2021 and differ by cause (overdose and premature mortality 2019 to 2021, suicide and firearm deaths 2017 to 2021, homicide and motor-vehicle crashes 2015 to 2021) | 95% CI 38.4% to 48.2% | 618 counties | county population | County Health Rankings |   |
| pmm_age | 75 | years of age | upper age of the premature-death definition (definitional, not an estimate) | residents who die before this age in County Health Rankings premature mortality | County Health Rankings 2024 release; deaths pooled over windows that end in 2021 and differ by cause (overdose and premature mortality 2019 to 2021, suicide and firearm deaths 2017 to 2021, homicide and motor-vehicle crashes 2015 to 2021) |   |   | none (definition) | County Health Rankings |   |
| pmm_d1 | 307 | premature deaths (under age 75) per 100,000 residents, age-adjusted | population-weighted mean of county rates; state-cluster bootstrap interval | residents of counties in the least deprived tenth of counties by Area Deprivation Index | County Health Rankings 2024 release; deaths pooled over windows that end in 2021 and differ by cause (overdose and premature mortality 2019 to 2021, suicide and firearm deaths 2017 to 2021, homicide and motor-vehicle crashes 2015 to 2021) | 95% CI 296 to 318 | 308 counties | county population | County Health Rankings |   |
| pmm_d10 | 701 | premature deaths (under age 75) per 100,000 residents, age-adjusted | population-weighted mean of county rates; state-cluster bootstrap interval | residents of counties in the most deprived tenth of counties by Area Deprivation Index | County Health Rankings 2024 release; deaths pooled over windows that end in 2021 and differ by cause (overdose and premature mortality 2019 to 2021, suicide and firearm deaths 2017 to 2021, homicide and motor-vehicle crashes 2015 to 2021) | 95% CI 660 to 740 | 310 counties | county population | County Health Rankings |   |
| pmm_n | 3,083 | counties | count of counties in the gradient | counties with a premature mortality rate and an ADI rank | County Health Rankings 2024 release; deaths pooled over windows that end in 2021 and differ by cause (overdose and premature mortality 2019 to 2021, suicide and firearm deaths 2017 to 2021, homicide and motor-vehicle crashes 2015 to 2021) |   |   | none (full count) | County Health Rankings |   |
| pmm_r | +0.78 | correlation coefficient | population-weighted correlation of county premature mortality with county ADI; state-cluster bootstrap interval | residents of counties with a premature mortality rate and an ADI rank | County Health Rankings 2024 release; deaths pooled over windows that end in 2021 and differ by cause (overdose and premature mortality 2019 to 2021, suicide and firearm deaths 2017 to 2021, homicide and motor-vehicle crashes 2015 to 2021) | 95% CI 0.74 to 0.83 | 3,083 counties | county population | County Health Rankings |   |
| suic_black | 7.8 | suicide deaths per 100,000 non-Hispanic Black residents (age-adjusted rate) | mean of county race-specific rates weighted by group residents; state-cluster bootstrap interval; a composite of reporting counties, not a national rate | residents of the 297 counties that publish both a Black and a white rate for suicide, each group weighted by its residents | County Health Rankings 2024 release; deaths pooled over windows that end in 2021 and differ by cause (overdose and premature mortality 2019 to 2021, suicide and firearm deaths 2017 to 2021, homicide and motor-vehicle crashes 2015 to 2021) | 95% CI 7.1 to 8.5 | 297 counties | county population times the group's share of residents | County Health Rankings |   |
| suic_ratio | 2.1 | ratio of white to Black suicide rates | ratio of the two group means; state-cluster bootstrap interval | residents of the 297 counties that publish both a Black and a white rate for suicide, each group weighted by its residents | County Health Rankings 2024 release; deaths pooled over windows that end in 2021 and differ by cause (overdose and premature mortality 2019 to 2021, suicide and firearm deaths 2017 to 2021, homicide and motor-vehicle crashes 2015 to 2021) | 95% CI 1.9 to 2.3 | 297 counties | county population times the group's share of residents | County Health Rankings |   |
| suic_rrural | +0.53 | correlation coefficient | population-weighted correlation of the county rate with county rural share of residents; state-cluster bootstrap interval | residents of counties with a published suicide rate | County Health Rankings 2024 release; deaths pooled over windows that end in 2021 and differ by cause (overdose and premature mortality 2019 to 2021, suicide and firearm deaths 2017 to 2021, homicide and motor-vehicle crashes 2015 to 2021) | 95% CI 0.45 to 0.60 | 2,427 counties | county population | County Health Rankings |   |
| suic_rrural_unw | +0.37 | correlation coefficient | unweighted correlation of the county rate with county rural share of residents; state-cluster bootstrap interval | the counties with a published suicide rate | County Health Rankings 2024 release; deaths pooled over windows that end in 2021 and differ by cause (overdose and premature mortality 2019 to 2021, suicide and firearm deaths 2017 to 2021, homicide and motor-vehicle crashes 2015 to 2021) | 95% CI 0.32 to 0.44 | 2,427 counties | none (each county counts once) | County Health Rankings |   |
| suic_sh_q1 | 24.2% | percent of the four-kind crude death rate (suicide) | crude rate of one kind of death as a share of the sum of the four means (overdose, suicide, homicide, motor-vehicle crash); state-cluster bootstrap interval | residents of counties in the least deprived fifth of residents among counties with all four rates published | County Health Rankings 2024 release; deaths pooled over windows that end in 2021 and differ by cause (overdose and premature mortality 2019 to 2021, suicide and firearm deaths 2017 to 2021, homicide and motor-vehicle crashes 2015 to 2021) | 95% CI 21.6% to 29.5% | 71 counties | county population | County Health Rankings |   |
| suic_sh_q5 | 21.9% | percent of the four-kind crude death rate (suicide) | crude rate of one kind of death as a share of the sum of the four means (overdose, suicide, homicide, motor-vehicle crash); state-cluster bootstrap interval | residents of counties in the most deprived fifth of residents among counties with all four rates published | County Health Rankings 2024 release; deaths pooled over windows that end in 2021 and differ by cause (overdose and premature mortality 2019 to 2021, suicide and firearm deaths 2017 to 2021, homicide and motor-vehicle crashes 2015 to 2021) | 95% CI 20.1% to 23.9% | 618 counties | county population | County Health Rankings |   |
| suic_white | 16.4 | suicide deaths per 100,000 non-Hispanic white residents (age-adjusted rate) | mean of county race-specific rates weighted by group residents; state-cluster bootstrap interval; a composite of reporting counties, not a national rate | residents of the 297 counties that publish both a Black and a white rate for suicide, each group weighted by its residents | County Health Rankings 2024 release; deaths pooled over windows that end in 2021 and differ by cause (overdose and premature mortality 2019 to 2021, suicide and firearm deaths 2017 to 2021, homicide and motor-vehicle crashes 2015 to 2021) | 95% CI 14.8 to 18.0 | 297 counties | county population times the group's share of residents | County Health Rankings |   |
| thr_half | 50% | percent of implied firearm homicide deaths | threshold that defines the concentration statistic (definitional, not an estimate) | all deaths implied by the cluster rates | annual rates for 2022, 2023 and 2024, averaged |   |   | none (definition) | CDC injury and violence mapping |   |
| thr_q3 | 75% | percent of implied firearm homicide deaths | threshold that defines the concentration statistic (definitional, not an estimate) | all deaths implied by the cluster rates | annual rates for 2022, 2023 and 2024, averaged |   |   | none (definition) | CDC injury and violence mapping |   |
| toll_q1 | 42 | deaths per 100,000 residents per year, crude rates, four kinds of death summed | sum of the population-weighted mean crude rates of overdose, suicide, homicide and motor-vehicle crash deaths; state-cluster bootstrap interval | residents of counties in the least deprived fifth of residents among counties with all four rates published | County Health Rankings 2024 release; deaths pooled over windows that end in 2021 and differ by cause (overdose and premature mortality 2019 to 2021, suicide and firearm deaths 2017 to 2021, homicide and motor-vehicle crashes 2015 to 2021) | 95% CI 37 to 47 | 71 counties | county population | County Health Rankings |   |
| toll_q5 | 76 | deaths per 100,000 residents per year, crude rates, four kinds of death summed | sum of the population-weighted mean crude rates of overdose, suicide, homicide and motor-vehicle crash deaths; state-cluster bootstrap interval | residents of counties in the most deprived fifth of residents among counties with all four rates published | County Health Rankings 2024 release; deaths pooled over windows that end in 2021 and differ by cause (overdose and premature mortality 2019 to 2021, suicide and firearm deaths 2017 to 2021, homicide and motor-vehicle crashes 2015 to 2021) | 95% CI 69 to 84 | 618 counties | county population | County Health Rankings |   |
| under_cut | 30% | percent of implied deaths | cutoff that defines an undercounted county: clusters imply less than this share of the county file's deaths (definitional) | counties outside New York City | annual rates for 2022, 2023 and 2024, averaged |   |   | none (definition) | CDC injury and violence mapping |   |
| under_essex | 7.3% | percent of implied deaths | implied annual deaths from the cluster rates as a share of the county file's implied deaths for the same residents | residents of Essex County, New Jersey in clusters with a published rate in every year | annual rates for 2022, 2023 and 2024, averaged |   |   | ACS 2019-2023 residents | CDC injury and violence mapping |   |
| under_floor | 15 | implied deaths a year in the county file | smallest county-file toll for which the undercount rule applies (definitional) | counties outside New York City | annual rates for 2022, 2023 and 2024, averaged |   |   | none (definition) | CDC injury and violence mapping |   |
| under_sc | 28.1% | percent of implied deaths | implied annual deaths from the cluster rates as a share of the county file's implied deaths for the same residents | residents of Santa Clara County, California in clusters with a published rate in every year | annual rates for 2022, 2023 and 2024, averaged |   |   | ACS 2019-2023 residents | CDC injury and violence mapping |   |
| under_sd | 27.6% | percent of implied deaths | implied annual deaths from the cluster rates as a share of the county file's implied deaths for the same residents | residents of San Diego County, California in clusters with a published rate in every year | annual rates for 2022, 2023 and 2024, averaged |   |   | ACS 2019-2023 residents | CDC injury and violence mapping |   |
| under_suffolk | 8.7% | percent of implied deaths | implied annual deaths from the cluster rates as a share of the county file's implied deaths for the same residents | residents of Suffolk County, Massachusetts in clusters with a published rate in every year | annual rates for 2022, 2023 and 2024, averaged |   |   | ACS 2019-2023 residents | CDC injury and violence mapping |   |
| us_premature | 392 | premature deaths (under age 75) per 100,000 residents, age-adjusted | population-weighted mean of county rates; state-cluster bootstrap interval | residents of all counties with a premature mortality rate and an ADI rank | County Health Rankings 2024 release; deaths pooled over windows that end in 2021 and differ by cause (overdose and premature mortality 2019 to 2021, suicide and firearm deaths 2017 to 2021, homicide and motor-vehicle crashes 2015 to 2021) | 95% CI 369 to 416 | 3,083 counties | county population | County Health Rankings |   |
| wh_clu_2022 | 312 | tract clusters | clusters where CDC withholds the rate in 2022 (marked unstable) | tract clusters outside New York City | 2022 |   |   | none (full count) | CDC injury and violence mapping |   |
| wh_clu_2024 | 455 | tract clusters | clusters where CDC withholds the rate in 2024 (marked unstable) | tract clusters outside New York City | 2024 |   |   | none (full count) | CDC injury and violence mapping |   |
| xc_hi | +0.95 | correlation coefficient | population-weighted correlation of the two county rates; state-cluster bootstrap interval | residents of counties with a rate in both County Health Rankings and the CDC injury panel | County Health Rankings 2024 windows against the injury panel's 2021 modeled rate | 95% CI 0.94 to 0.96 | 1,882 counties | county population | County Health Rankings |   |
| xc_lo | +0.87 | correlation coefficient | population-weighted correlation of the two county rates; state-cluster bootstrap interval | residents of counties with a rate in both County Health Rankings and the CDC injury panel | County Health Rankings 2024 windows against the injury panel's 2021 modeled rate | 95% CI 0.83 to 0.90 | 2,264 counties | county population | County Health Rankings |   |
| xc_lo_unw | +0.64 | correlation coefficient | unweighted correlation of the two county suicide rates; state-cluster bootstrap interval | the counties with a suicide rate in both County Health Rankings and the CDC injury panel | County Health Rankings 2024 windows against the injury panel's 2021 modeled rate | 95% CI 0.58 to 0.71 | 2,264 counties | none (each county counts once) | County Health Rankings |   |

## Definitions

- Census tract: A small area drawn by the Census Bureau to stand in for a neighborhood. Boundaries are redrawn after each decennial census.
- Modeled estimate: A figure produced by a statistical model instead of a direct count or measurement, usually for small areas where direct data are thin, borrowing strength from surveys, neighboring areas or other data. Neighboring areas can share errors, so single extremes call for caution.
- Area Deprivation Index (ADI): A score that ranks neighborhoods by income, education, employment and housing conditions drawn from the American Community Survey. In this book a higher national percentile means a more deprived neighborhood.
- USALEEP: The U.S. Small-area Life Expectancy Estimates Project, a National Center for Health Statistics release of life expectancy at birth for most census tracts. It covers deaths from 2010 to 2015 only and has not been updated.
- Age-adjusted rate: A rate recalculated as if every place had the same mix of ages, so that places can be compared without the effect of having older or younger residents.
- Population-weighted: Averaged so that each place counts in proportion to its number of residents, which lets a large county move the result more than a small one. An unweighted average counts every place equally.

## Methods

The first four figures describe tract clusters, the unit in which CDC publishes neighborhood firearm homicide rates, using modeled rates averaged over 2022, 2023 and 2024; each cluster's rate is repeated on its member tracts, and New York City is left out. The last four describe counties, using County Health Rankings death rates pooled over windows that end in 2021 and differ by cause, so the mix of deaths compares rates measured over different spans, and a county with too few deaths has no rate. Life expectancy covers deaths in 2010 to 2015, earlier than the rates it is compared with. Every comparison describes places and not people.

### Which clusters and years enter the concentration curves

Firearm homicide rates come from the CDC injury and violence tract file, which publishes one rate for each cluster of neighboring tracts (the median tract sits in a cluster of 5) and repeats it on every member tract.

The chapter works on clusters: it keeps one row per cluster, sets a cluster's population to the sum of its tracts' 2019-2023 American Community Survey population, and averages the annual rates for 2022, 2023 and 2024.

A cluster enters only if CDC publishes a rate in all three years; New York City, where the file records almost no deaths, is left out, and a sensitivity check drops the clusters of the counties where the cluster rates imply under 30% of the county file's deaths.

Explains Table 1, Table 2 and Table 3.

### How implied deaths and the curves are computed

Implied deaths are the rate times the cluster's residents. Clusters are ranked from the highest rate and the curve accumulates residents and implied deaths.

Explains Table 1, Table 2 and Table 4.

### Where the county death rates come from

County Health Rankings rates are deaths per 100,000 from National Center for Health Statistics records, pooled over windows that end in 2021 and differ by cause (2019 to 2021 for overdose and premature mortality, 2017 to 2021 for suicide, 2015 to 2021 for homicide and motor-vehicle crashes); premature mortality and suicide are age-adjusted, and the overdose, homicide, motor-vehicle, firearm and child mortality rates are crude (the cause mix and the leading-kind map use CHR's crude suicide rate so that four rates share one basis).

A county is missing a rate when it has fewer than 10 deaths in the window (20 for premature mortality).

Race-specific rates are county rates for non-Hispanic Black and white residents, averaged over counties that publish both, each group weighted by its residents.

Explains Table 5, Table 6, Table 7 and Table 8.

### How deprivation and life expectancy are measured

County deprivation is the mean of tract Area Deprivation Index national ranks weighted by tract population; Connecticut tracts map to the eight legacy counties.

Life expectancy is the National Center for Health Statistics estimate for deaths in 2010 to 2015, area-weighted onto 2020 tracts, with institutional tracts left out.

Explains Table 4 and Table 5.

### How uncertainty is measured

Intervals for national concentration statistics come from a bootstrap that draws whole counties with replacement, and those for one county from a bootstrap that draws clusters, 2,000 replicates with seed 12345, as percentile intervals.

Decile means, shares and correlations across counties come from a bootstrap that draws whole states with replacement, 2,000 replicates, seed 12345, weighted by county population.

Explains Table 1, Table 2, Table 6 and Table 7.

### Four rankings of the same places

The comparison follows the book's ranking switch rule, fixed on October 28, 2026 before any of these gradients was computed. The main gradient of the chapter is recomputed under four rankings of the same places, with the outcome, period, weights, bins and intervals unchanged: the published ADI; a standardized index that applies the ADI's coefficients to its measures after standardizing each one across tracts; median household income; and the share of adults 25 and older with a bachelor's degree or more. The three alternatives come from ACS tract estimates for 2019 to 2023, the period of the ADI.

All four rankings use the same places, the 3,080 counties that have the outcome and all four rankings, so the ADI panel can differ slightly from the chapter's main figure. Tracts enter only if they have every input of the standardized index, which leaves out tracts without a median rent, home value or mortgage cost. A county takes the population-weighted mean of the ranks of its common-sample tracts, with Connecticut tracts assigned to its eight legacy counties, as for the county ADI.

A pattern counts as changed when the worse end switches sides, when the two ends differ under one ranking and not the other, or when the rank correlation of bin order and bin mean falls on the other side of 0.9 and moves by at least 0.1.

Explains Table 9.

## What these data cannot show

The firearm homicide figures describe tract clusters, and CDC's rates for them are modeled estimates, so they say nothing about single tracts, blocks or people. The curves and the Cook County map average over clusters of neighboring tracts, which lowers measured concentration, and the cluster file's implied deaths fall short of the county file's, so the figures describe shares and not totals. New York City is absent, four more counties have clusters that imply a small share of the county file's deaths, and the clusters where CDC withholds a rate are left out, with more withheld for the provisional year 2024. The fill check bounds the last exclusion and cannot remove it.

Every comparison across places is ecological. The life expectancy of Cook County's highest-rate tenth comes from deaths in 2010 to 2015 and the rates from 2022 to 2024, so the data cannot say whether firearm homicide shortened these lives or whether the same conditions underlie both. County gradients and correlations describe counties, and the homicide and rural-share correlation differs with the weighting and depends on which counties report, so a correlation across counties says little about the people in them. The Area Deprivation Index understates deprivation in high-cost cities, and it is built from survey measures that overlap with several outcome measures.

County Health Rankings leaves out counties with few deaths, so the mix of deaths and the map of the highest-rate kind speak for the larger counties. The rates in those two figures are crude, so a county with an older population has a higher rate for the same risk, and the correlation figure mixes crude and age-adjusted measures, which its row labels state. The race-specific rates are a composite of reporting counties and not national rates, and the cross-check against the injury panel compares two compilations of the same death records. Outside the box above, nothing in this chapter supports a claim about what changes violence.

## References

1. Weisburd, David (2015). The law of crime concentration and the criminology of place. Criminology 53(2): 133-157. https://doi.org/10.1111/1745-9125.12070
2. Braga, Anthony A.; Weisburd, David; Turchan, Brandon (2018). Focused deterrence strategies and crime control: an updated systematic review and meta-analysis of the empirical evidence. Criminology & Public Policy 17(1): 205-250. https://doi.org/10.1111/1745-9133.12353
3. Skogan, Wesley G.; Hartnett, Susan M.; Bump, Natalie; Dubois, Jill (2008). Evaluation of CeaseFire-Chicago. National Institute of Justice (NCJ 227181); Northwestern University Institute for Policy Research. https://www.ojp.gov/ncjrs/virtual-library/abstracts/evaluation-ceasefire-chicago
4. Ransford, Charles; Williams, Monique; Slutkin, Gary (2025). A systematic review on the effectiveness of the Cure Violence approach. INQUIRY: The Journal of Health Care Organization, Provision, and Financing 62: 00469580251366142. https://doi.org/10.1177/00469580251366142
5. Branas, Charles C.; South, Eugenia; Kondo, Michelle C.; Hohl, Bernadette C.; Bourgois, Philippe; Wiebe, Douglas J.; MacDonald, John M. (2018). Citywide cluster randomized trial to restore blighted vacant land and its effects on violence, crime, and fear. Proceedings of the National Academy of Sciences 115(12): 2946-2951. https://doi.org/10.1073/pnas.1718503115

## Provenance

- Pipeline version: 0.1.0
- Build script: book2/chapters/gun-homicide-and-causes-of-early-death/build.py (sha256 50358ebb69b0c63622881101bbd7717cac3204b209675430919d65ee9192cdce)
- Built on: 2026-10-07
- Site commit: c2b1ff3309cb4a10833f884da91ac0057a66ad0a

Inputs:

| File | Size (bytes) | SHA-256, first 1 MB | SHA-256, last 1 MB |
| --- | ---: | --- | --- |
| data/analytic/tract_atlas.parquet | 39,659,925 | 29d5e4d943552d54aeb6ba92aadc939d9bd15bdea553c78357dbd516757be627 | 1e10d1ecf722340f23ee2212d4999a90250f85b65fa89b39e32db06d9472e22e |
| data/raw/acs/deprivation_components_2019_2023/tract.parquet | 13,197,793 | 5545cde2cbe96a8b56e468d65388248a241aa6bd300d3c9e325d08d3f86b11da | 5c8ec454d066988b0f1a01aaa6ed34ebfc05d162e01cf9c6bb01f9198cbaff12 |
| data/staged/adi/adi_atlas_tract2020_2023.parquet | 798,078 | 5e409bc212ff871def780fa27b4cf8b616d9da420b2b9f6455517eddcd8c1a95 | 5e409bc212ff871def780fa27b4cf8b616d9da420b2b9f6455517eddcd8c1a95 |
| data/staged/crosswalks/ct_tract2020_to_old_county.parquet | 15,842 | 961ef39f31596ef65fca2c67d5f2f63e533c5dc9b49745d3bb6437844e4b831d | 961ef39f31596ef65fca2c67d5f2f63e533c5dc9b49745d3bb6437844e4b831d |
| data/staged/health/chr_county_2024.parquet | 8,776,933 | 619795f7ecfec8aff10694b649eaee5e421dbeffbe4be47af4161b51ceefa5cd | c69357f1bcf630407bb9716cb445618838999558c019aaaf77d4681b6bf45c56 |
| data/staged/life_expectancy/usaleep_tract2020.parquet | 1,316,169 | 9c9c7f103791b75eddba74e68a89804efdf2486b6b66bb4bc02a46ba5bf8bc23 | 27317107be2465d9fd5e83271dea0cd6e915ee419e2802bedc28747d0a41d38f |
| data/staged/mortality/injury_violence_county.parquet | 722,427 | d0dfd060a3ab9eaa3a87b35d1b24b5a1e9f21af3513c10367edb651c46f40403 | d0dfd060a3ab9eaa3a87b35d1b24b5a1e9f21af3513c10367edb651c46f40403 |
| data/staged/mortality/injury_violence_tract2020.parquet | 8,059,686 | 48f2a99135501cb7c8c7af7c85361808bd998eb4ab06825759340a1befc5f4f1 | e2e5d28442025676279f349563e0eef9235b013d40007e081ba9671697d04b49 |
| data/staged/ses/acs5_tract2020_2019_2023.parquet | 7,232,424 | 5fe6f730a35225f14d5d8abfc6b9bce58b52e545f309a7eac74f1dae74d7a31c | 7b4a2475b3852d3f91804716a04566ef8ceadacec2be9fb85f5eb642078efac6 |
| data/staged/svi/svi_tract2020_2022.parquet | 17,740,415 | cb3860d8edc2aaf1f2a02211a58f7a0b80457eb6ad9a37b5ec8abb2d5da9da10 | e371a4a696d5bed05d54ad7c5b6ca2504508805b847ec415c2322be7f66ec2b2 |

Outputs:

| File | SHA-256 |
| --- | --- |
| facts.json | 3e69869e30dd34d1912643bf04bf67c57c394e53032fca8b0d878efe9f9a2242 |
| figures/alt_rankings.json | caae3d0ba756606aa3c3be21d3cb4a42d1b4f1c5a131b8213e6ceb0d65616748 |
| figures/cause_correlations.json | 31df8dd62dca3ea410ee4e0133badba7219b35aaf5d760242ac7540f41423e1c |
| figures/cause_mix_by_quintile.json | 56d6d080aa8bca7f95329e1fb06eb381518b3f3b7de7da88a660e632b31c0f1b |
| figures/chicago_homicide_map.json | 191f47e20373da8319145c987449663f86facd5eefa9e18418538c970d2d9610 |
| figures/cook_deciles.json | 7e88b81b16c56e130a2ba528c99f92c56f8fb4d9fc286b80e479c4f8ff1d87cb |
| figures/county_concentration.json | 931ee02bf8258befcf001fc0f0b4f6312e284bfa0820abdd1445e2f099afa0f5 |
| figures/dominant_cause_map.json | 3fd940d7605940e2ab1b58e0ed55ebf5af0756d6b8ad5f47e3c471154b0099b9 |
| figures/national_concentration.json | c3d61facf072989e1e93e9546f0acc3929a0f51eb568464e58d5d175977f4ff4 |
| figures/premature_death_gradient.json | 9f6caebfdf9bf42e3502ddab55fa820d9a160aa265130d037db8ac7f0cdd33f3 |
| meta.json | 5e1e38da016a6630107880e0c507280877bc3c6400125f64f03753a4f964dcb0 |
| public/data/gun-homicide-and-causes-of-early-death/chicago_homicide_map.json | f137e61f1d729372d707ce651b766566744577940436937023b5f801e6724b46 |
| public/data/gun-homicide-and-causes-of-early-death/dominant_cause_map.json | 02083449d1f78f996a0b0a68fba9fbda4367ec85910a12bf9b69347daa9cc21b |
| refs.json | a5bd6c46293c4572475be2397d03e1a3cb3d77e234f4d9b70191af695160bb22 |

## Machine formats

- Report (HTML): https://vishalsingh.org/zipcode/gun-homicide-and-causes-of-early-death/report
- Report (Markdown): https://vishalsingh.org/zipcode/gun-homicide-and-causes-of-early-death/report.md
- Every estimate (JSON): https://vishalsingh.org/zipcode/report/gun-homicide-and-causes-of-early-death/facts.json
- Table 1, figure "national_concentration" (JSON): https://vishalsingh.org/zipcode/report/gun-homicide-and-causes-of-early-death/figures/national_concentration.json
- Table 2, figure "county_concentration" (JSON): https://vishalsingh.org/zipcode/report/gun-homicide-and-causes-of-early-death/figures/county_concentration.json
- Table 3, figure "chicago_homicide_map" (JSON): https://vishalsingh.org/zipcode/report/gun-homicide-and-causes-of-early-death/figures/chicago_homicide_map.json
- Table 4, figure "cook_deciles" (JSON): https://vishalsingh.org/zipcode/report/gun-homicide-and-causes-of-early-death/figures/cook_deciles.json
- Table 5, figure "premature_death_gradient" (JSON): https://vishalsingh.org/zipcode/report/gun-homicide-and-causes-of-early-death/figures/premature_death_gradient.json
- Table 6, figure "cause_mix_by_quintile" (JSON): https://vishalsingh.org/zipcode/report/gun-homicide-and-causes-of-early-death/figures/cause_mix_by_quintile.json
- Table 7, figure "cause_correlations" (JSON): https://vishalsingh.org/zipcode/report/gun-homicide-and-causes-of-early-death/figures/cause_correlations.json
- Table 8, figure "dominant_cause_map" (JSON): https://vishalsingh.org/zipcode/report/gun-homicide-and-causes-of-early-death/figures/dominant_cause_map.json
- Table 9, figure "alt_rankings" (JSON): https://vishalsingh.org/zipcode/report/gun-homicide-and-causes-of-early-death/figures/alt_rankings.json
- Payload of "chicago_homicide_map" (JSON): https://vishalsingh.org/zipcode/data/gun-homicide-and-causes-of-early-death/chicago_homicide_map.json
- Payload of "dominant_cause_map" (JSON): https://vishalsingh.org/zipcode/data/gun-homicide-and-causes-of-early-death/dominant_cause_map.json
- Index of every report: https://vishalsingh.org/zipcode/llms.txt; all reports in one file: https://vishalsingh.org/zipcode/llms-full.txt

---

# Overdose, alcohol and suicide: report

Chapter 8.2 of ZIP Code Destiny, by Vishal Singh (NYU Stern). Generated from the chapter's data files; the reading version is at https://vishalsingh.org/zipcode/overdose-alcohol-and-suicide. Table n of this report is figure n of the chapter.

## Identification

- Title: Overdose, alcohol and suicide
- Label: Two despairs
- Chapter: 8.2 (Part VIII, Early death; An American life)
- Question: Where are overdose, alcohol and suicide deaths highest, and how does each change with county deprivation?
- Dek: Across counties with a stable modeled rate for suicide, overdose and homicide in 2019 to 2023, suicide death rates are only weakly related to overdose and homicide rates (population-weighted correlations of +0.13 and −0.02).
- Grain: counties
- Status: published
- Published: 2026-10-09
- Updated: not revised
- Data sources:
  - CDC injury and violence mapping. CDC Mapping Injury, Overdose, and Violence county file: modeled overdose, suicide and homicide death rates, annual 2019 to 2023; years: 2019-2024 for counties, 2022-2024 for tracts (data as of 2026-05-14); weights: 2019-2023 ACS residents of each county, and every county once
  - CDC WONDER underlying cause of death. CDC WONDER underlying cause of death: drug poisoning (overdose), alcohol-induced causes and suicide, counties 2018 to 2024; years: 2018-2024; weights: none (crude rates from counts); WONDER county population
  - American Community Survey. Total population of 2020 tracts, 2019-2023, summed to counties as weights for the panel's county rates; years: 2019-2023 (5-year); weights: none (full count)
  - Area Deprivation Index. Neighborhood Atlas Area Deprivation Index 2023, tract national rank averaged to counties; years: 2023 release (version 4.0.1; built on ACS 2019-2023); weights: 2019-2023 tract population
  - Rural-Urban Continuum Codes. USDA Rural-Urban Continuum Codes 2023, county; years: 2023; weights: none (classification)
  - County Health Rankings. County Health Rankings 2024, share of residents who are American Indian or Alaska Native (Census Population Estimates 2022); years: 2024 release; weights: none (published county measure)
- Canonical URL: https://vishalsingh.org/zipcode/overdose-alcohol-and-suicide
- Report: https://vishalsingh.org/zipcode/overdose-alcohol-and-suicide/report
- Markdown: https://vishalsingh.org/zipcode/overdose-alcohol-and-suicide/report.md
- BibTeX key: singh2026overdosealcoholandsuicide

```bibtex
@misc{singh2026overdosealcoholandsuicide,
  author = {Singh, Vishal},
  title = {Overdose, alcohol and suicide},
  year = {2026},
  url = {https://vishalsingh.org/zipcode/overdose-alcohol-and-suicide},
  note = {In ZIP Code Destiny. Published 9 October 2026.}
}
```

## Principal estimates

The facts the dek and the opening prose cite, in order of appearance.

| Estimand | Population | Period | Value | Interval | n | Weights | Source | Fact id |
| --- | --- | --- | ---: | --- | ---: | --- | --- | --- |
| correlation of the county overdose and suicide rates, population-weighted; bootstrap over counties | counties with a stable rate for all three kinds of death | annual modeled rates for 2019 to 2023, averaged | +0.13 | 95% CI 0.05 to 0.21 | 1,899 counties | 2019-2023 ACS residents | CDC injury and violence mapping | p_corr_od_su |
| correlation of the county suicide and homicide rates, population-weighted; bootstrap over counties | counties with a stable rate for all three kinds of death | annual modeled rates for 2019 to 2023, averaged | −0.02 | 95% CI −0.10 to 0.06 | 1,899 counties | 2019-2023 ACS residents | CDC injury and violence mapping | p_corr_su_hm |
| count of counties | counties in CDC's county file with a stable rate for overdose, suicide and homicide in every year | 2019 to 2023 | 1,899 |   |   | none (full count) | CDC injury and violence mapping | p_common_n |
| share of residents living in counties with a stable rate for all three kinds of death | residents of the 50 states and DC | 2019-2023 ACS residents | 93.5% |   |   | 2019-2023 ACS residents | CDC injury and violence mapping | p_common_pop |
| correlation of the county overdose and homicide rates, population-weighted; bootstrap over counties | counties with a stable rate for all three kinds of death | annual modeled rates for 2019 to 2023, averaged | +0.48 | 95% CI 0.35 to 0.58 | 1,899 counties | 2019-2023 ACS residents | CDC injury and violence mapping | p_corr_od_hm |
| definition: WONDER withholds a county-year count above zero and below this number | county-years in CDC WONDER | deaths in 2018 to 2024 | 10 |   |   | none (definition) | CDC WONDER underlying cause of death | w_floor |
| count of mapped counties | counties complete for drug and alcohol with at least 20 pooled deaths of each | deaths in 2018 to 2024 | 545 |   |   | none (full count) | CDC WONDER underlying cause of death | dom_n |
| definition: the least number of pooled deaths for a county to be mapped | a county drawn on a map of WONDER counts | deaths in 2018 to 2024 | 20 |   |   | none (definition) | CDC WONDER underlying cause of death | map_floor |
| count of mapped counties with a higher crude overdose than alcohol-induced death rate | mapped counties | deaths in 2018 to 2024 | 479 |   | 545 counties | none (full count) | CDC WONDER underlying cause of death | dom_od_higher |
| count of mapped counties with a higher crude alcohol-induced than overdose death rate | mapped counties | deaths in 2018 to 2024 | 66 |   | 545 counties | none (full count) | CDC WONDER underlying cause of death | dom_alc_higher |

## Findings

Suicide, overdose and homicide death rates are highest in different counties, and suicide is the most separate. Across the 1,899 counties with a stable modeled rate for all three in 2019 to 2023, which hold 93.5% of residents, the county overdose rate has a population-weighted correlation of +0.13 (95% CI 0.05 to 0.21) with the suicide rate. The suicide rate has a correlation of −0.02 (95% CI −0.10 to 0.06) with the homicide rate. Overdose and homicide rates are more closely related, with a correlation of +0.48 (95% CI 0.35 to 0.58).

Drug and alcohol deaths separate as well, but the counts behind that result come from a selected set of counties. CDC WONDER publishes a county's count in every year only when the count is zero or at least 10 each year, which favors larger counties with high rates for their size. In the 545 such counties where each cause has at least 20 deaths, the overdose death rate is higher than the alcohol-induced death rate in 479 counties and lower in 66.

Case and Deaton found that mortality among middle-aged white non-Hispanic Americans rose between 1999 and 2013, mostly through drug and alcohol poisonings, suicide and chronic liver disease.[1] They later grouped suicide, drug overdose and alcoholic liver disease as deaths of despair.[2] The grouping treats three kinds of death as one problem. This chapter asks whether they share an address, using death rates for counties. A county rate is deaths per hundred thousand residents, crude and not adjusted for age, and it says nothing about who died or why. Whether people in these places feel despair is a person-level question, taken up in Chapter 6.3, Life evaluation and daily feelings (https://vishalsingh.org/zipcode/life-evaluation-and-daily-feelings).

### Suicide, overdose and homicide rates peak in different counties

The first three figures use the county file of CDC's injury and violence dashboard, which carries modeled estimates of death rates and is the only source in this chapter that also covers homicide. A model borrows strength across counties where deaths are few, so the maps can show counties that CDC WONDER's public counts leave blank. A county is drawn when CDC publishes a stable rate in each year from 2019 to 2023, and its rate is the mean of the five annual rates. That leaves 2,714 counties with an overdose rate, 2,273 with a suicide rate and 2,807 with a homicide rate, holding 97.7%, 94.9% and 97.6% of residents. CDC's file has 3,135 counties and leaves out Connecticut's. The suicide map has the fewest counties because CDC marks more small-county suicide rates unstable.

Table 1. Across counties, the population-weighted correlation of suicide rates is +0.13 with overdose rates and −0.02 with homicide rates, against +0.48 between overdose and homicide rates

The counties with the highest rates lie in different places. In the highest tenth of overdose rates, 68.0% of the counties are in the South, which holds 45.7% of counties with an overdose rate. The South also holds 85.8% of the highest tenth for homicide, against 47.7% of counties with a homicide rate. The highest tenth for suicide lies elsewhere, with 50.0% of its counties in the West, which holds 13.7% of counties with a suicide rate.

The correlations in the opening paragraph give the same picture county by county. Counting every county once, the three correlations are smaller and in the same order: +0.19, +0.02 and +0.27. The suicide rate shares little with the overdose rate or the homicide rate, and the interval for its correlation with the homicide rate includes zero under both weightings. Overdose and homicide share more, but much less when every county counts once than when large counties carry their weight. The panel counts only unintentional and undetermined-intent drug poisonings as overdose, so an overdose recorded as a suicide appears in the suicide rate and not in the overdose rate. Homicide counts every method. Gun homicide is the subject of Chapter 8.1, Gun homicide and the causes of early death (https://vishalsingh.org/zipcode/gun-homicide-and-causes-of-early-death).

### Homicide, suicide and overdose are higher in the most deprived tenth of counties

Weighted by population, all three rates are higher in the most deprived tenth than in the least, and the homicide and overdose lines do not rise steadily in between. Rank counties by the Area Deprivation Index of their residents and cut the ranking into tenths with equal numbers of counties (Table 2). A county's rank is the population-weighted mean of the national ranks of its tracts. Weighted by population, the homicide rate is 4.8 (95% CI 4.2 to 5.4) deaths per hundred thousand residents in the least deprived tenth and 13.6 (95% CI 11.1 to 16.2) in the most deprived, a ratio of 2.83 (95% CI 2.24 to 3.58). Suicide goes from 12.2 (95% CI 11.3 to 13.2) to 17.4 (95% CI 15.7 to 19.4), a ratio of 1.43 (95% CI 1.25 to 1.63), and the line is level over the more deprived tenths.

Overdose goes from 23.8 (95% CI 22.2 to 25.6) to 31.1 (95% CI 25.7 to 36.5), a ratio of 1.30 (95% CI 1.07 to 1.55). Across all counties with a rate, the population-weighted correlation with deprivation is +0.46 (95% CI 0.37 to 0.53) for suicide, +0.30 (95% CI 0.25 to 0.36) for homicide and +0.22 (95% CI 0.15 to 0.28) for overdose.

Homicide and overdose do not climb steadily. The point estimates of the homicide line fall in 4 of the nine steps between tenths, and those of the overdose line in 4. The overdose line reaches 36.1 (95% CI 30.6 to 42.0) in tenth 4, which is 5.0 (95% CI −2.6 to 13.2) above the most deprived tenth. The interval for that difference includes zero, and a peak chosen from ten noisy means overstates its lead, so the data do not show that overdose rates are higher in the middle of the ranking than at the deprived end.

Table 2. Weighted by population, homicide rates in the most deprived tenth of counties are 2.83 times those in the least deprived, suicide rates 1.43 times and overdose rates 1.30 times, but the homicide and overdose lines do not rise steadily

Counting every county once changes the picture for two of the three. Homicide is 3.4 (95% CI 3.1 to 3.7) in the least deprived tenth and 9.5 (95% CI 8.2 to 10.9) in the most deprived, a ratio of 2.78 (95% CI 2.34 to 3.30), close to the weighted ratio. Suicide is 15.3 (95% CI 14.6 to 16.1) in the least deprived tenth and 14.2 (95% CI 13.2 to 15.3) in the most deprived, a ratio of 0.93 (95% CI 0.85 to 1.02).

Overdose is 22.2 (95% CI 21.1 to 23.2) and 23.0 (95% CI 19.5 to 26.4), a ratio of 1.03 (95% CI 0.88 to 1.20). Neither the suicide ratio's interval nor the overdose ratio's excludes equal rates in the two tenths. The correlation with deprivation is −0.01 (95% CI −0.04 to 0.03) for overdose and +0.01 (95% CI −0.04 to 0.05) for suicide, against +0.18 (95% CI 0.16 to 0.21) for homicide. The weighted rise in suicide and overdose comes from the larger counties, and the small counties that make up most of the most deprived tenth do not show it.

Three features of the tenths limit what they show. The tenths cover the 3,120 counties with a deprivation rank and hold equal numbers of counties, so the most deprived tenth is mostly small rural counties. A county without a stable rate for a kind of death is left out of that line. Those counties are also small ones, and the suicide line has the fewest counties in its most deprived tenth for that reason. The Area Deprivation Index understates deprivation in high-cost cities (Chapter 1.1, What the deprivation index measures (https://vishalsingh.org/zipcode/anatomy-of-deprivation)), so a high-cost county can sit in a less deprived tenth than its tracts would suggest. The homicide correlation with deprivation here differs from the one in Chapter 8.1, Gun homicide and the causes of early death (https://vishalsingh.org/zipcode/gun-homicide-and-causes-of-early-death), which uses County Health Rankings' homicide rates, available only for counties with enough homicides, and a different set of counties.

### Suicide is higher in more urban nonmetro counties, homicide in large metro counties

Table 3 groups counties by the 2023 Rural-Urban Continuum Code. Weighted by population, suicide is 19.7 (95% CI 19.2 to 20.2) deaths per hundred thousand residents in nonmetro counties with an urban population of 5,000 or more (more urban nonmetro counties) and 12.2 (95% CI 11.6 to 12.9) in large metro counties, those in metro areas of 1 million or more. The ratio is 1.6 (95% CI 1.5 to 1.7).

Homicide runs the other way, 7.6 (95% CI 6.7 to 8.7) in large metro counties against 4.6 (95% CI 4.2 to 5.1) in nonmetro counties with a smaller urban population (least urban nonmetro counties), a ratio of 1.6 (95% CI 1.4 to 1.9). The overdose rate is 27.8 (95% CI 25.7 to 30.0) in large metro counties and 21.8 (95% CI 20.2 to 23.5) in least urban nonmetro counties, a ratio of 1.3 (95% CI 1.1 to 1.4).

Table 3. Weighted by population, suicide rates in more urban nonmetro counties are 1.6 times those in large metro counties, and homicide rates in large metro counties are 1.6 times those in the least urban nonmetro counties

Counting every county once, suicide is 19.6 (95% CI 19.2 to 20.0) in more urban nonmetro counties against 14.6 (95% CI 14.1 to 15.0) in large metro counties, a ratio of 1.3 (95% CI 1.3 to 1.4). Homicide is 6.2 (95% CI 5.6 to 6.9) in large metro counties against 3.5 (95% CI 3.2 to 3.9) in least urban nonmetro counties, a ratio of 1.8 (95% CI 1.5 to 2.0). Both contrasts hold.

Two other comparisons do not hold. Suicide is no higher in least urban nonmetro counties than in large metro counties when every county counts once. The overdose ratio is 1.8 (95% CI 1.6 to 2.0), with 28.3 (95% CI 26.8 to 29.9) in large metro counties and 16.0 (95% CI 14.8 to 17.2) in least urban nonmetro counties, a larger gap than the weighted one. Overdose rates are lower in least urban nonmetro counties under both weightings, and they differ more by county type than the weighted figures suggest.

These are crude rates. Age structure differs between rural and urban counties and suicide rates differ by age, so part of a gap between county types can reflect who lives in the county and not a different risk at the same age. The panel's rates are also modeled for small counties, which can borrow strength from neighboring areas, so a nonmetro county's rate may partly reflect its neighbors.

### Overdose deaths exceed alcohol-induced deaths in most counties where both can be mapped

Alcohol-induced deaths appear in this chapter's data only through CDC WONDER, which counts deaths by the underlying cause on the death certificate. Here drug means drug poisoning of any intent (unintentional, suicide, homicide or undetermined). Alcohol-induced means WONDER's category of alcohol-induced deaths, which includes alcoholic liver disease and alcohol poisoning, and suicide means intentional self-harm by any method. A suicide by drug poisoning is in both the drug count and the suicide count. All three are crude rates per hundred thousand residents.

WONDER withholds any county-year count above zero and below 10, and it publishes a count of zero, so a rate pooled over the published years of a county is biased. This chapter does not impute, bound or back out the withheld counts. It uses the counties whose count is published in every year, which it calls complete counties, for which the rate pooled over 2018 to 2024 is exact. Complete counties are selected on the outcome as well as on size. Among counties of 30,000 to 60,000 residents, the mean overdose rate in CDC's modeled panel is 50.9 (95% CI 46.1 to 56.4) in counties with a published count in every year and 22.5 (95% CI 21.5 to 23.4) in the rest. The maps and gradient of WONDER counts therefore describe these counties and not counties in general. Chapter 10.3, Suppressed counties in death data (https://vishalsingh.org/zipcode/suppressed-counties-in-death-data) covers how many county-years are withheld, which counties lose their counts and what the selection does to comparisons.

Table 4 maps the ratio of the two rates, on a log scale, for the 545 counties that are complete for both and have at least 20 pooled deaths of each. They hold 76.6% of residents outside Connecticut. The overdose rate is the higher one in 479 of them and the alcohol-induced rate in 66.

Each class of the key is a doubling of the ratio. The overdose rate is at least 2 times the alcohol-induced rate in 234 mapped counties, and the alcohol-induced rate is at least 2 times the overdose rate in 6. Across the 569 counties that publish both counts every year, the two rates have a correlation of +0.25 (95% CI 0.16 to 0.35), and +0.30 (95% CI 0.20 to 0.43) counting every county once.

The counties where alcohol is higher are mostly in the West and Midwest, which hold 78.8% of them. Alcohol is the higher rate in 28.0% of mapped counties in the West and 16.4% in the Midwest, against 6.5% in the South and 0.0% in the Northeast.

Table 4. Overdose deaths exceed alcohol-induced deaths in 479 of 545 mapped counties, and the 66 counties where alcohol is higher lie mostly in the West and Midwest

Several of the counties with the highest alcohol-induced rates have large American Indian and Alaska Native populations. Of the ten mapped counties with the highest alcohol-induced rate, 6 are counties where American Indian or Alaska Native residents are at least 10% of residents, by County Health Rankings' figures from the American Community Survey. Only 10 of the 133 counties in WONDER of that kind are mapped, and alcohol is the higher rate in 6 of those.

Those counties' pooled alcohol-induced rate is 57.3 deaths per hundred thousand residents, against 13.6 in the other mapped counties. The cutoff matters. At 5% instead, there are 27 such counties and alcohol is the higher rate in 8 of them. These county totals are not divided by race, so the comparison describes counties and not the people who died.

### Weighted by population, suicide and overdose rates correlate more with deprivation than alcohol-induced rates

Table 5 repeats the gradient for the WONDER rates, in the 525 counties whose drug, alcohol and suicide counts are published in every year, which hold 76.0% of residents outside Connecticut. The tenths are tenths of these counties, which are selected as described above, so they do not line up with the tenths of Table 2. The correlation of the county rate with the deprivation rank, weighted by person-years, is +0.42 (95% CI 0.30 to 0.51) for suicide, +0.35 (95% CI 0.24 to 0.44) for overdose and +0.13 (95% CI 0.04 to 0.22) for alcohol-induced deaths.

Counting every county once, the correlations are +0.28 (95% CI 0.19 to 0.36), +0.26 (95% CI 0.19 to 0.34) and +0.15 (95% CI 0.05 to 0.22). From the same bootstrap draws, the weighted correlation for overdose is higher than the one for alcohol by +0.21 (95% CI 0.09 to 0.33), and the one for suicide by +0.28 (95% CI 0.18 to 0.39).

In the least deprived tenth the rates are 22.0 (95% CI 20.0 to 24.8), 12.5 (95% CI 10.8 to 14.0) and 9.9 (95% CI 9.1 to 11.1) for overdose, alcohol-induced deaths and suicide. In the most deprived tenth they are 32.3 (95% CI 25.2 to 40.0), 17.6 (95% CI 13.9 to 22.7) and 16.5 (95% CI 14.2 to 18.9).

The overdose rate is higher in the most deprived tenth than in the least deprived by 10.3 (95% CI 2.5 to 18.3) deaths per hundred thousand residents and the suicide rate by 6.6 (95% CI 4.0 to 9.2). These are differences between the two tenths' pooled rates, with intervals from bootstrap draws that the two tenths share.

Table 5. Among counties with published counts in every year, the population-weighted correlation of deprivation is +0.42 with suicide and +0.35 with overdose death rates, against +0.13 with alcohol-induced death rates

The overdose line here keeps rising into more deprived tenths than the panel's line in Table 2, which is a different set of counties ranked in a different way. In all counties with a modeled rate the population-weighted correlation with deprivation for overdose is +0.22 (95% CI 0.15 to 0.28), below the +0.35 (95% CI 0.24 to 0.44) here, so the selection matters for the overdose result. The alcohol-induced line is level until the most deprived tenth, and the jump there depends on a few counties. 5 of the tenth's counties are ones where American Indian or Alaska Native residents are at least 10% of residents, and without them the tenth's rate is 14.4 (95% CI 12.1 to 16.6), in line with the other tenths.

The panel and WONDER give the same county rates where both can be compared. Across counties whose WONDER counts are published in every year, the county rate in the two sources has a correlation of +0.995 (95% CI 0.993 to 0.996) for overdose and +0.995 (95% CI 0.994 to 0.996) for suicide. The overdose agreement holds although WONDER's overdose count includes overdoses recorded as suicides and the panel's does not.

In those counties the panel's rates are modeled in 0.6% of overdose county-years and 0.0% of suicide county-years, against 44.9% and 54.0% in all counties. The agreement therefore checks the processing of the same death records where the panel's rates are direct and does not validate the modeled rates for small counties behind the first three figures.

### Change over time is in the chapter on mortality trends

Chapter 8.3, Mortality trends, 2018 to 2024 (https://vishalsingh.org/zipcode/mortality-trends-2018-2024) follows the national overdose, alcohol-induced and suicide rates from 2018 to 2024, including the size of the overdose peak and the fall since. It also checks WONDER's crude series against CDC's modeled panel, which gives a different size for the rise. A study of death records describes the fall in overdose deaths from 2023 to 2024 as concentrated in deaths involving illicit fentanyl. The study also reports that deaths involving stimulants without fentanyl and deaths involving xylazine continued to make up a growing share of overdose deaths.[3] That study is descriptive, and a fall in a national rate says nothing about which counties it reached. This chapter does not map change over time.

Explaining the fall is a causal question, and the maps cannot answer it. One policy lever has been studied with designs that compare states by when they adopted it.

> Causal evidence: Naloxone access laws and opioid deaths (Difference-in-differences, staggered state adoption)
>
> Two studies use the staggered timing of state laws that widened access to naloxone, the drug that reverses an opioid overdose. Abouk, Pacula and Powell compared states before and after they adopted each type of law with states that had not adopted it, using state and time fixed effects and the years since adoption, for 2005 to 2016.[4] Laws that gave pharmacists direct authority to dispense naloxone were followed by fewer fatal opioid overdoses, and laws of other types showed no association with deaths. Doleac and Mukherjee used the same variation in timing and found more opioid-related emergency room visits and more opioid-related theft, with no net reduction in opioid-related deaths.[5] The two studies disagree about the same kind of policy, so the evidence on naloxone laws is mixed, and both studies end well before 2024, so neither can say how much of the fall after 2023 naloxone, treatment or changes in the drug supply account for.

### Only the income ranking separates the two ends of the overdose gradient

Over the 2,699 counties that have a stable overdose rate and all four rankings, the ADI baseline gives 24.0 (95% CI 22.3 to 25.7) overdose deaths per hundred thousand residents in the least deprived tenth of counties and 30.0 (95% CI 24.8 to 35.3) in the most deprived, with a rank correlation of +0.14 between tenth and rate, and on this common sample the two end intervals overlap, unlike those in Table 2 (Table 6). The standardized index, which applies the ADI's coefficients after standardizing each measure, gives 21.7 (95% CI 20.3 to 23.2) and 27.0 (95% CI 20.3 to 34.6), with a correlation of +0.28, and its end intervals overlap as well. Ranked by median household income the tenths run from 22.7 (95% CI 21.2 to 24.4) to 33.1 (95% CI 28.7 to 37.6), intervals that do not overlap, with a correlation of +0.43, and ranked by the bachelor's degree share they run from 24.3 (95% CI 22.3 to 26.3) to 29.4 (95% CI 26.0 to 33.0), with overlapping ends. Under the book's rule the income ranking counts as a change, since its ends differ and the baseline's do not, while the standardized index and the degree share leave the gradient unchanged (Chapter 9.1, How far one ranking goes (https://vishalsingh.org/zipcode/how-far-one-ranking-goes)).

Table 6. The overdose death rate does not differ detectably between the least and the most deprived places under the ADI; it is worse at the more deprived end but uneven along the way under the income ranking

### What these data cannot show

Every figure describes counties. A county rate is an average over all of its residents, so it cannot say who died, and a correlation across counties does not describe the people in them (ecological inference). The data do not measure despair. Deaths from suicide, overdose and alcohol are counted, and whether the people who live in these counties feel despair is a person-level question for Chapter 6.3, Life evaluation and daily feelings (https://vishalsingh.org/zipcode/life-evaluation-and-daily-feelings). Survey measures of drinking are in Chapter 7.2, Health behaviors, person by person (https://vishalsingh.org/zipcode/health-behaviors-person-by-person).

The WONDER figures cover a selected set of counties. Counties whose counts are published in every year hold most residents but are a minority of counties, they are larger, and they have higher rates than other counties of similar size. The correlations with deprivation among the 525 complete counties, +0.42 (95% CI 0.30 to 0.51) for suicide, +0.35 (95% CI 0.24 to 0.44) for overdose and +0.13 (95% CI 0.04 to 0.22) for alcohol-induced deaths, describe those counties only. Most counties where American Indian or Alaska Native residents are at least 10% of residents are outside the maps, as are other small counties with high rates. Withheld counts could hold up to the shares given in the methods note, and no method in this chapter recovers them.

North Carolina's drug-poisoning count for 2023 in WONDER predates the state's amended death certificates and is low. It is 15.8% of the published deaths of the 48 complete North Carolina counties in the window. Raising each of their published 2023 counts by 10% to 19% raises their pooled overdose rate by 1.6% to 3.0%.

At the larger increase the correlation of overdose with deprivation is +0.35 (95% CI 0.24 to 0.44) and the overdose gap between the most and least deprived tenths is 10.4 (95% CI 2.6 to 18.4), and no comparison in this chapter changes direction. The modeled panel covers nearly every county but borrows strength across them, so its rate for a small county is partly its neighbors' rate, and its agreement with WONDER is checked only where its rates are direct. Alcohol-induced deaths are deaths whose underlying cause is alcohol, such as alcoholic liver disease, and they leave out deaths from other conditions in which alcohol was involved.

Several results depend on the weighting. In all counties with a modeled rate, the population-weighted correlations with deprivation are +0.46 (95% CI 0.37 to 0.53) for suicide and +0.22 (95% CI 0.15 to 0.28) for overdose, and counting every county once they are +0.01 (95% CI −0.04 to 0.05) and −0.01 (95% CI −0.04 to 0.03). Weighted by population, suicide and overdose rates are higher in the most deprived tenth of counties and suicide rates are higher outside large metro counties. Counting every county once, the first holds for homicide only and the second for smaller metro and more urban nonmetro counties only. Rates are crude, so a county with an older population has a higher rate for the same risk at each age, and comparisons of rural and urban counties carry that difference. The Area Deprivation Index understates deprivation in high-cost cities. The tenths of counties are relative to the counties in each figure. The three kinds of death also overlap a little, since a suicide by drug poisoning is counted as both a suicide and a drug death. Outside the box above, nothing in this chapter supports a claim about what lowers any of these death rates.

## Tables

### Table 1. Across counties, the population-weighted correlation of suicide rates is +0.13 with overdose rates and −0.02 with homicide rates, against +0.48 between overdose and homicide rates

CDC modeled death rates per 100,000 residents, mean of annual rates for 2019 to 2023, by county: overdose, suicide and homicide in three views

No inline rows: the values of Table 1 are in https://vishalsingh.org/zipcode/data/overdose-alcohol-and-suicide/three_causes_overdose.json.

Views: Overdose (every row); Suicide (every row); Homicide (every row).

Note: Each map plots a county's mean annual modeled rate for 2019 to 2023. A county is drawn when CDC publishes a stable rate in each of the five years; gray hatched counties have a year in which CDC marks the rate unstable (fewer than 10 deaths and no stable model) or, for two Alaska areas that did not exist in 2019, no row, and the tooltip says which. The overdose map has 2,714 counties with a rate, the suicide map 2,273 and the homicide map 2,807, and they hold 97.7%, 94.9% and 97.6% of residents.

Note: These are CDC modeled estimates, which borrow strength across counties where deaths are few, so the WONDER floor of 20 pooled deaths does not apply here; it applies to the maps of WONDER counts in Figure 4. CDC publishes counts under 10 as ranges, and the tooltip gives the lowest and highest five-year totals those ranges allow. Overdose is unintentional or undetermined-intent drug overdose, so it leaves out overdoses recorded as suicides; suicide and homicide include every method. Rates are crude, not age-adjusted. The panel's 2024 values are provisional and left out.

Note: Connecticut's eight counties are gray because CDC's county file does not carry them. Puerto Rico is not drawn. Alaska is drawn.

Source: CDC Mapping Injury, Overdose, and Violence county file, 2019-2023; modeled estimates, crude rates

Data: https://vishalsingh.org/zipcode/report/overdose-alcohol-and-suicide/figures/three_causes_maps.json

### Table 2. Weighted by population, homicide rates in the most deprived tenth of counties are 2.83 times those in the least deprived, suicide rates 1.43 times and overdose rates 1.30 times, but the homicide and overdose lines do not rise steadily

CDC modeled death rates per 100,000 residents by tenth of counties ranked by Area Deprivation Index, mean of annual rates for 2019 to 2023, county rates weighted by population

| cause | conf_high | conf_low | decile | estimate | n_units | weighting |
| --- | ---: | ---: | ---: | ---: | ---: | --- |
| Overdose | 25.58 | 22.17 | 1 | 23.84 | 302 | Population-weighted |
| Overdose | 29.93 | 24.57 | 2 | 27.28 | 296 | Population-weighted |
| Overdose | 34.96 | 26.90 | 3 | 30.74 | 289 | Population-weighted |
| Overdose | 42.01 | 30.60 | 4 | 36.13 | 283 | Population-weighted |
| Overdose | 35.49 | 27.61 | 5 | 31.86 | 270 | Population-weighted |
| Overdose | 31.25 | 24.58 | 6 | 27.90 | 268 | Population-weighted |
| Overdose | 35.13 | 27.89 | 7 | 31.39 | 270 | Population-weighted |
| Overdose | 33.13 | 21.71 | 8 | 27.17 | 260 | Population-weighted |
| Overdose | 30.12 | 23.86 | 9 | 26.83 | 248 | Population-weighted |
| Overdose | 36.49 | 25.66 | 10 | 31.09 | 215 | Population-weighted |
| Suicide | 13.16 | 11.31 | 1 | 12.18 | 267 | Population-weighted |
| Suicide | 15.20 | 13.23 | 2 | 14.15 | 255 | Population-weighted |
| Suicide | 17.11 | 14.83 | 3 | 15.83 | 247 | Population-weighted |
| Suicide | 17.04 | 15.09 | 4 | 16.03 | 239 | Population-weighted |
| Suicide | 17.94 | 16.04 | 5 | 16.91 | 243 | Population-weighted |
| Suicide | 18.76 | 16.57 | 6 | 17.63 | 225 | Population-weighted |
| Suicide | 19.09 | 17.60 | 7 | 18.32 | 214 | Population-weighted |
| Suicide | 18.94 | 14.20 | 8 | 16.62 | 211 | Population-weighted |
| Suicide | 18.83 | 17.33 | 9 | 18.10 | 183 | Population-weighted |
| Suicide | 19.41 | 15.71 | 10 | 17.37 | 179 | Population-weighted |
| Homicide | 5.44 | 4.19 | 1 | 4.82 | 301 | Population-weighted |
| Homicide | 9.17 | 5.65 | 2 | 7.31 | 292 | Population-weighted |
| Homicide | 9.02 | 6.40 | 3 | 7.86 | 281 | Population-weighted |
| Homicide | 14.24 | 8.53 | 4 | 11.44 | 282 | Population-weighted |
| Homicide | 12.33 | 7.48 | 5 | 9.89 | 277 | Population-weighted |
| Homicide | 9.99 | 6.87 | 6 | 8.45 | 275 | Population-weighted |
| Homicide | 10.00 | 6.19 | 7 | 7.86 | 275 | Population-weighted |
| Homicide | 7.73 | 5.66 | 8 | 6.63 | 280 | Population-weighted |
| Homicide | 10.31 | 7.40 | 9 | 8.79 | 273 | Population-weighted |
| Homicide | 16.17 | 11.14 | 10 | 13.63 | 257 | Population-weighted |
| Overdose | 23.25 | 21.05 | 1 | 22.19 | 302 | Each county counted once |
| Overdose | 25.61 | 22.63 | 2 | 24.07 | 296 | Each county counted once |
| Overdose | 25.72 | 22.45 | 3 | 24.04 | 289 | Each county counted once |
| Overdose | 27.06 | 22.97 | 4 | 25.01 | 283 | Each county counted once |
| Overdose | 24.59 | 20.70 | 5 | 22.56 | 270 | Each county counted once |
| Overdose | 23.20 | 19.55 | 6 | 21.40 | 268 | Each county counted once |
| Overdose | 26.06 | 21.64 | 7 | 23.81 | 270 | Each county counted once |
| Overdose | 25.14 | 20.33 | 8 | 22.68 | 260 | Each county counted once |
| Overdose | 23.62 | 19.26 | 9 | 21.33 | 248 | Each county counted once |
| Overdose | 26.39 | 19.55 | 10 | 22.96 | 215 | Each county counted once |
| Suicide | 16.12 | 14.57 | 1 | 15.34 | 267 | Each county counted once |
| Suicide | 18.19 | 16.75 | 2 | 17.46 | 255 | Each county counted once |
| Suicide | 18.63 | 17.09 | 3 | 17.85 | 247 | Each county counted once |
| Suicide | 18.27 | 16.91 | 4 | 17.58 | 239 | Each county counted once |
| Suicide | 18.45 | 16.62 | 5 | 17.46 | 243 | Each county counted once |
| Suicide | 18.36 | 16.64 | 6 | 17.49 | 225 | Each county counted once |
| Suicide | 18.40 | 16.95 | 7 | 17.66 | 214 | Each county counted once |
| Suicide | 17.53 | 15.88 | 8 | 16.72 | 211 | Each county counted once |
| Suicide | 16.89 | 15.09 | 9 | 16.04 | 183 | Each county counted once |
| Suicide | 15.33 | 13.24 | 10 | 14.22 | 179 | Each county counted once |
| Homicide | 3.74 | 3.11 | 1 | 3.41 | 301 | Each county counted once |
| Homicide | 4.88 | 3.94 | 2 | 4.34 | 292 | Each county counted once |
| Homicide | 4.76 | 3.86 | 3 | 4.29 | 281 | Each county counted once |
| Homicide | 5.48 | 4.14 | 4 | 4.79 | 282 | Each county counted once |
| Homicide | 5.48 | 4.17 | 5 | 4.79 | 277 | Each county counted once |
| Homicide | 5.10 | 3.90 | 6 | 4.48 | 275 | Each county counted once |
| Homicide | 6.06 | 4.63 | 7 | 5.31 | 275 | Each county counted once |
| Homicide | 5.55 | 4.23 | 8 | 4.91 | 280 | Each county counted once |
| Homicide | 7.24 | 5.54 | 9 | 6.36 | 273 | Each county counted once |
| Homicide | 10.86 | 8.16 | 10 | 9.47 | 257 | Each county counted once |

Views: Population-weighted (weighting = Population-weighted); Each county counted once (weighting = Each county counted once).

Note: The deprivation rank of a county is the population-weighted mean of the Area Deprivation Index national ranks of its tracts, so a county ranked as least deprived can contain deprived tracts. The index understates deprivation in high-cost areas.

Note: Counties without a stable rate in all five years are left out of that cause's line: 2,714 counties for overdose, 2,273 for suicide and 2,807 for homicide, out of 3,120 counties with a deprivation rank. The suicide line is thinnest in the most deprived tenth, where CDC marks the most small-county rates unstable. Connecticut is not in CDC's county file.

Note: The weighted and unweighted views differ for suicide and overdose: counting every county once, the ratio of the most deprived tenth to the least deprived is 0.93 for suicide and 1.03 for overdose, against 2.78 for homicide, so the weighted rise comes from the larger counties.

Note: Across all the counties in the comparison, weighted by population, the rate is 27.7 per hundred thousand residents for overdose, 14.3 for suicide and 7.1 for homicide.

Source: CDC Mapping Injury, Overdose, and Violence county file, 2019-2023, and Area Deprivation Index 2023; modeled estimates, crude rates, population-weighted and unweighted

Data: https://vishalsingh.org/zipcode/report/overdose-alcohol-and-suicide/figures/three_causes_by_decile.json

### Table 3. Weighted by population, suicide rates in more urban nonmetro counties are 1.6 times those in large metro counties, and homicide rates in large metro counties are 1.6 times those in the least urban nonmetro counties

CDC modeled death rates per 100,000 residents by 2023 Rural-Urban Continuum Code group of county, mean of annual rates for 2019 to 2023, county rates weighted by population

| cause | conf_high | conf_low | county_type | estimate | n_units | weighting |
| --- | ---: | ---: | --- | ---: | ---: | --- |
| Overdose | 30.00 | 25.67 | Large metro | 27.79 | 428 | Population-weighted |
| Overdose | 29.93 | 26.91 | Smaller metro | 28.41 | 702 | Population-weighted |
| Overdose | 27.72 | 25.21 | Nonmetro, more urban | 26.45 | 851 | Population-weighted |
| Overdose | 23.48 | 20.20 | Nonmetro, least urban | 21.77 | 733 | Population-weighted |
| Suicide | 12.88 | 11.57 | Large metro | 12.21 | 387 | Population-weighted |
| Suicide | 17.15 | 16.00 | Smaller metro | 16.58 | 629 | Population-weighted |
| Suicide | 20.19 | 19.23 | Nonmetro, more urban | 19.69 | 668 | Population-weighted |
| Suicide | 18.54 | 17.01 | Nonmetro, least urban | 17.74 | 589 | Population-weighted |
| Homicide | 8.69 | 6.65 | Large metro | 7.62 | 431 | Population-weighted |
| Homicide | 7.29 | 6.15 | Smaller metro | 6.69 | 704 | Population-weighted |
| Homicide | 6.30 | 5.24 | Nonmetro, more urban | 5.76 | 846 | Population-weighted |
| Homicide | 5.13 | 4.19 | Nonmetro, least urban | 4.65 | 826 | Population-weighted |
| Overdose | 29.92 | 26.84 | Large metro | 28.34 | 428 | Each county counted once |
| Overdose | 27.02 | 24.61 | Smaller metro | 25.81 | 702 | Each county counted once |
| Overdose | 24.96 | 22.88 | Nonmetro, more urban | 23.88 | 851 | Each county counted once |
| Overdose | 17.23 | 14.75 | Nonmetro, least urban | 15.96 | 733 | Each county counted once |
| Suicide | 15.03 | 14.14 | Large metro | 14.59 | 387 | Each county counted once |
| Suicide | 17.86 | 17.08 | Smaller metro | 17.47 | 629 | Each county counted once |
| Suicide | 20.02 | 19.15 | Nonmetro, more urban | 19.58 | 668 | Each county counted once |
| Suicide | 14.99 | 13.69 | Nonmetro, least urban | 14.34 | 589 | Each county counted once |
| Homicide | 6.87 | 5.62 | Large metro | 6.20 | 431 | Each county counted once |
| Homicide | 6.34 | 5.50 | Smaller metro | 5.90 | 704 | Each county counted once |
| Homicide | 6.00 | 5.08 | Nonmetro, more urban | 5.53 | 846 | Each county counted once |
| Homicide | 3.89 | 3.16 | Nonmetro, least urban | 3.53 | 826 | Each county counted once |

Views: Population-weighted (weighting = Population-weighted); Each county counted once (weighting = Each county counted once).

Note: The groups follow the Rural-Urban Continuum Code: large metro is counties in metro areas of 1 million or more, smaller metro is the other metro counties, more urban nonmetro counties have an urban population of 5,000 or more, and least urban nonmetro counties have fewer. Connecticut is not in CDC's county file.

Note: Rates are crude, not age-adjusted. Age structure differs between rural and urban counties, and suicide rates differ by age, so part of a difference between types of county can reflect age structure and not the same risk at every age.

Source: CDC Mapping Injury, Overdose, and Violence county file, 2019-2023, and USDA Rural-Urban Continuum Codes 2023; modeled estimates, crude rates, population-weighted and unweighted

Data: https://vishalsingh.org/zipcode/report/overdose-alcohol-and-suicide/figures/rate_by_county_type.json

### Table 4. Overdose deaths exceed alcohol-induced deaths in 479 of 545 mapped counties, and the 66 counties where alcohol is higher lie mostly in the West and Midwest

Crude alcohol-induced death rate relative to crude overdose death rate, log base 2 of the ratio, counties with a published count in every year and at least 20 pooled deaths of each cause, 2018 to 2024

No inline rows: the values of Table 4 are in https://vishalsingh.org/zipcode/data/overdose-alcohol-and-suicide/drug_vs_alcohol_dominance.json.

Note: A county is drawn when WONDER publishes its drug and alcohol counts in every year from 2018 to 2024 and it has at least 20 pooled deaths of each. WONDER withholds county-year counts above zero and below 10, so a rate built from the published years of a county with withheld years would leave out its low-count years; the map uses only counties with no withheld year. That leaves 545 of the counties, which hold 76.6% of residents outside Connecticut, and the tooltip gives the reason for every gray county.

Note: Those counties are selected on the outcome as well as on size. Among counties of 30,000 to 60,000 residents, the mean overdose rate in CDC's modeled panel is 50.9 in counties whose WONDER overdose count is published in every year and 22.5 in the rest: a county whose deaths reach 10 or more in every year has a high rate for its size. The map describes those counties and not counties in general.

Note: Rates are crude, not age-adjusted. Drug deaths are drug poisonings of any intent (unintentional, suicide, homicide and undetermined), alcohol-induced deaths are WONDER's alcohol-induced category, and the county ratio compares counts and populations from the same WONDER query. Connecticut's counties have no population in WONDER for 2022 to 2024, so they are gray. Valdez-Cordova AK is coded by its old code and draws on both successor areas; Kusilvak AK and Oglala Lakota SD are drawn under their current codes. Puerto Rico is not drawn.

Source: CDC WONDER underlying cause of death, 2018-2024; crude rates, counties with a published count in every year

Data: https://vishalsingh.org/zipcode/report/overdose-alcohol-and-suicide/figures/drug_vs_alcohol_dominance.json

### Table 5. Among counties with published counts in every year, the population-weighted correlation of deprivation is +0.42 with suicide and +0.35 with overdose death rates, against +0.13 with alcohol-induced death rates

Crude death rates per 100,000 residents by tenth of counties ranked by Area Deprivation Index, counties with a published count in every year for all three kinds of death, 2018 to 2024, weighted by person-years

| cause | conf_high | conf_low | decile | estimate | n_units |
| --- | ---: | ---: | ---: | ---: | ---: |
| Overdose | 24.80 | 20.04 | 1 | 21.99 | 52 |
| Overdose | 27.34 | 20.94 | 2 | 24.01 | 53 |
| Overdose | 31.36 | 23.49 | 3 | 27.64 | 52 |
| Overdose | 29.42 | 23.00 | 4 | 26.06 | 53 |
| Overdose | 30.54 | 22.71 | 5 | 27.06 | 52 |
| Overdose | 39.78 | 26.38 | 6 | 32.06 | 53 |
| Overdose | 41.98 | 27.94 | 7 | 34.55 | 52 |
| Overdose | 45.37 | 30.72 | 8 | 37.26 | 53 |
| Overdose | 39.62 | 29.73 | 9 | 34.91 | 52 |
| Overdose | 40.00 | 25.20 | 10 | 32.29 | 53 |
| Alcohol-induced causes | 14.04 | 10.78 | 1 | 12.54 | 52 |
| Alcohol-induced causes | 15.52 | 11.99 | 2 | 13.76 | 53 |
| Alcohol-induced causes | 16.56 | 12.24 | 3 | 14.44 | 52 |
| Alcohol-induced causes | 14.22 | 10.93 | 4 | 12.30 | 53 |
| Alcohol-induced causes | 15.58 | 12.07 | 5 | 13.42 | 52 |
| Alcohol-induced causes | 17.39 | 11.97 | 6 | 14.30 | 53 |
| Alcohol-induced causes | 15.22 | 12.22 | 7 | 13.48 | 52 |
| Alcohol-induced causes | 16.86 | 14.36 | 8 | 15.57 | 53 |
| Alcohol-induced causes | 16.05 | 13.60 | 9 | 14.71 | 52 |
| Alcohol-induced causes | 22.74 | 13.88 | 10 | 17.57 | 53 |
| Suicide | 11.05 | 9.07 | 1 | 9.90 | 52 |
| Suicide | 14.56 | 11.76 | 2 | 13.07 | 53 |
| Suicide | 16.09 | 12.55 | 3 | 14.35 | 52 |
| Suicide | 14.53 | 12.38 | 4 | 13.27 | 53 |
| Suicide | 15.88 | 11.98 | 5 | 13.72 | 52 |
| Suicide | 16.51 | 13.24 | 6 | 14.61 | 53 |
| Suicide | 16.50 | 14.06 | 7 | 15.08 | 52 |
| Suicide | 17.56 | 15.05 | 8 | 16.27 | 53 |
| Suicide | 16.99 | 14.71 | 9 | 15.73 | 52 |
| Suicide | 18.93 | 14.24 | 10 | 16.50 | 53 |

Note: The figure uses the 525 counties whose drug, alcohol and suicide counts WONDER publishes in every year from 2018 to 2024, which hold 76.0% of residents outside Connecticut. WONDER withholds county-year counts above zero and below 10, which would drop the low-count years of a small county and bias its rate. The tenths are tenths of these counties ranked by deprivation, with equal numbers of counties, so the most deprived tenth here is not the most deprived tenth of all counties.

Note: These counties are selected on the outcome as well as on size (see Figure 4), so the correlations describe them and not counties in general. In all counties with a modeled rate, the population-weighted correlation with deprivation is +0.46 for suicide and +0.22 for overdose (Figure 2), and +0.01 and −0.01 counting every county once.

Note: Rates are crude, not age-adjusted. A suicide by drug poisoning is counted in both the overdose and the suicide rate. The most deprived tenth includes counties with large American Indian and Alaska Native populations, where alcohol-induced death rates are high.

Source: CDC WONDER underlying cause of death, 2018-2024, and Area Deprivation Index 2023; crude rates, population-weighted (person-years)

Data: https://vishalsingh.org/zipcode/report/overdose-alcohol-and-suicide/figures/despair_gradient.json

### Table 6. The overdose death rate does not differ detectably between the least and the most deprived places under the ADI; it is worse at the more deprived end but uneven along the way under the income ranking

Overdose deaths per 100,000 residents by tenth of counties ranked four ways, over the 2,699 counties that have all four rankings

| bin | conf_high | conf_low | estimate | n_units | ranking |
| ---: | ---: | ---: | ---: | ---: | --- |
| 1 | 25.73 | 22.32 | 23.98 | 301 | Area Deprivation Index |
| 2 | 29.71 | 24.37 | 27.05 | 294 | Area Deprivation Index |
| 3 | 37.57 | 27.61 | 32.22 | 290 | Area Deprivation Index |
| 4 | 39.38 | 29.56 | 34.01 | 281 | Area Deprivation Index |
| 5 | 33.46 | 26.73 | 30.04 | 272 | Area Deprivation Index |
| 6 | 34.52 | 25.38 | 30.27 | 269 | Area Deprivation Index |
| 7 | 34.66 | 27.24 | 30.94 | 268 | Area Deprivation Index |
| 8 | 34.44 | 22.27 | 28.15 | 261 | Area Deprivation Index |
| 9 | 30.71 | 24.25 | 27.22 | 242 | Area Deprivation Index |
| 10 | 35.31 | 24.80 | 30.05 | 221 | Area Deprivation Index |
| 1 | 23.20 | 20.29 | 21.74 | 300 | Standardized index |
| 2 | 30.97 | 25.61 | 28.42 | 296 | Standardized index |
| 3 | 34.32 | 26.26 | 30.03 | 273 | Standardized index |
| 4 | 29.67 | 23.00 | 25.92 | 260 | Standardized index |
| 5 | 37.89 | 29.35 | 33.55 | 283 | Standardized index |
| 6 | 40.27 | 31.70 | 36.37 | 265 | Standardized index |
| 7 | 48.75 | 26.81 | 37.39 | 261 | Standardized index |
| 8 | 30.39 | 24.17 | 27.13 | 271 | Standardized index |
| 9 | 36.80 | 24.06 | 30.61 | 247 | Standardized index |
| 10 | 34.57 | 20.33 | 26.97 | 243 | Standardized index |
| 1 | 24.36 | 21.19 | 22.68 | 300 | Median household income |
| 2 | 29.37 | 24.46 | 27.04 | 295 | Median household income |
| 3 | 33.05 | 25.53 | 28.89 | 284 | Median household income |
| 4 | 33.50 | 25.01 | 29.39 | 275 | Median household income |
| 5 | 40.44 | 32.14 | 36.53 | 267 | Median household income |
| 6 | 47.88 | 29.17 | 38.58 | 264 | Median household income |
| 7 | 35.61 | 28.82 | 32.13 | 262 | Median household income |
| 8 | 33.75 | 23.00 | 28.01 | 259 | Median household income |
| 9 | 31.52 | 26.39 | 28.87 | 246 | Median household income |
| 10 | 37.60 | 28.75 | 33.15 | 247 | Median household income |
| 1 | 26.33 | 22.27 | 24.29 | 298 | Bachelor's degree share |
| 2 | 31.45 | 26.22 | 28.93 | 295 | Bachelor's degree share |
| 3 | 35.98 | 25.74 | 30.09 | 285 | Bachelor's degree share |
| 4 | 33.74 | 27.06 | 30.26 | 266 | Bachelor's degree share |
| 5 | 32.82 | 25.50 | 29.05 | 249 | Bachelor's degree share |
| 6 | 29.74 | 22.80 | 25.95 | 277 | Bachelor's degree share |
| 7 | 32.85 | 26.02 | 29.49 | 258 | Bachelor's degree share |
| 8 | 32.12 | 26.81 | 29.38 | 263 | Bachelor's degree share |
| 9 | 30.65 | 24.86 | 27.68 | 257 | Bachelor's degree share |
| 10 | 32.97 | 25.96 | 29.41 | 251 | Bachelor's degree share |

Note: Least and most deprived tenth: ADI 24.0 and 30.0, rank correlation +0.14; standardized index 21.7 and 27.0, +0.28; median household income 22.7 and 33.1, +0.43; bachelor's degree share 24.3 and 29.4, +0.14.

Note: The standardized index applies the ADI's coefficients to its measures after standardizing each one, so home value and income carry less weight than in the ADI. Income and degree share are single census measures, each ranked from the highest value to the lowest.

Source: CDC Mapping Injury, Overdose, and Violence county file, 2019-2023, and Area Deprivation Index 2023; modeled estimates, crude rates, population-weighted; ADI 2023 and ACS 2019-2023 tract tables (B19013, B15003 and the 15 rebuilt ADI measures)

Data: https://vishalsingh.org/zipcode/report/overdose-alcohol-and-suicide/figures/alt_rankings.json

### Table 7. Weighted by residents, the county overdose rate rises weakly with deprivation (+0.22), and counted once each, counties show no relation (−0.01)

Overdose deaths per 100,000 residents, 2019 to 2023 average, modeled county rates, against county Area Deprivation Index; dot area is residents

No inline rows: the values of Table 7 are in https://vishalsingh.org/zipcode/data/overdose-alcohol-and-suicide/overdose_map_scatter.json.

Note: Counties without a stable modeled rate in every year from 2019 to 2023 are gray on the map and have no dot; Connecticut's counties are not in CDC's county file.

Source: CDC Mapping Injury, Overdose, and Violence county file, 2019-2023, and Area Deprivation Index 2023; modeled estimates, crude rates

Data: https://vishalsingh.org/zipcode/report/overdose-alcohol-and-suicide/figures/overdose_map_scatter.json

## All estimates

Every fact in facts.json (162), sorted by id.

| Fact id | Value | Unit | Estimand | Population | Period | Interval | n | Weights | Source | Note |
| --- | ---: | --- | --- | --- | --- | --- | ---: | --- | --- | --- |
| a_n | 3,120 | counties | count of counties | counties in CDC's county file with an Area Deprivation Index rank | ADI 2023 |   |   | none (full count) | Area Deprivation Index |   |
| aian5_alc_higher | 8 | counties | count with a higher crude alcohol-induced than overdose death rate | mapped counties where American Indian or Alaska Native residents are at least the lower cutoff share | deaths in 2018 to 2024 |   | 27 counties | none (full count) | CDC WONDER underlying cause of death |   |
| aian5_cut | 5% | percent of residents who are American Indian or Alaska Native | definition: the lower cutoff used to check how the counts change | counties | Census Population Estimates 2022 as reported by County Health Rankings 2024 |   |   | none (definition) | County Health Rankings |   |
| aian5_n | 27 | counties | count of counties | mapped counties where American Indian or Alaska Native residents are at least the lower cutoff share | Census Population Estimates 2022 as reported by County Health Rankings 2024 |   |   | none (full count) | County Health Rankings |   |
| aian_alc_higher | 6 | counties | count with a higher crude alcohol-induced than overdose death rate | mapped counties where American Indian or Alaska Native residents are at least the cutoff share | deaths in 2018 to 2024 |   | 10 counties | none (full count) | CDC WONDER underlying cause of death |   |
| aian_all_n | 133 | counties | count of counties | counties in CDC WONDER where American Indian or Alaska Native residents are at least the cutoff share (Alaska and South Dakota codes mapped to their successors) | Census Population Estimates 2022 as reported by County Health Rankings 2024 |   |   | none (full count) | County Health Rankings |   |
| aian_cut | 10% | percent of residents who are American Indian or Alaska Native | definition: a county counts as American Indian or Alaska Native heavy at this share of residents | counties | Census Population Estimates 2022 as reported by County Health Rankings 2024 |   |   | none (definition) | County Health Rankings |   |
| aian_n | 10 | counties | count of counties | mapped counties where American Indian or Alaska Native residents are at least the cutoff share | Census Population Estimates 2022 as reported by County Health Rankings 2024 |   |   | none (full count) | County Health Rankings |   |
| aian_rate | 57.3 | alcohol-induced deaths per 100,000 residents | pooled crude rate | residents of mapped counties where American Indian or Alaska Native residents are at least the cutoff share | deaths in 2018 to 2024 |   | 10 counties | none (crude rate from counts) | CDC WONDER underlying cause of death |   |
| aian_rate_other | 13.6 | alcohol-induced deaths per 100,000 residents | pooled crude rate | residents of the other mapped counties | deaths in 2018 to 2024 |   | 535 counties | none (crude rate from counts) | CDC WONDER underlying cause of death |   |
| aian_top10 | 6 | counties | count of those counties where American Indian or Alaska Native residents are at least the cutoff share | the ten mapped counties with the highest crude alcohol-induced death rate | deaths in 2018 to 2024 |   | 10 counties | none (full count) | CDC WONDER underlying cause of death |   |
| all_hm | 47.7% | percent of counties | share of counties with a stable homicide rate in the South region (Census regions) | counties in CDC's county file with a stable modeled rate in each year 2019 to 2023 | annual modeled rates for 2019 to 2023, averaged |   | 2,807 counties | none (each county counts once) | CDC injury and violence mapping |   |
| all_od | 45.7% | percent of counties | share of counties with a stable overdose rate in the South region (Census regions) | counties in CDC's county file with a stable modeled rate in each year 2019 to 2023 | annual modeled rates for 2019 to 2023, averaged |   | 2,714 counties | none (each county counts once) | CDC injury and violence mapping |   |
| all_su | 13.7% | percent of counties | share of counties with a stable suicide rate in the West region (Census regions) | counties in CDC's county file with a stable modeled rate in each year 2019 to 2023 | annual modeled rates for 2019 to 2023, averaged |   | 2,273 counties | none (each county counts once) | CDC injury and violence mapping |   |
| alt_adi_least | 24.0 | Overdose deaths per 100,000 residents | population-weighted mean of county modeled overdose rates; 95% intervals from a bootstrap over counties | the least deprived tenth of counties ranked by Area Deprivation Index, among the counties with all four rankings | annual modeled rates for 2019 to 2023, averaged; rankings from ACS 2019-2023 | 95% CI 22.3 to 25.7 | 301 counties | 2019-2023 ACS residents | CDC injury and violence mapping |   |
| alt_adi_most | 30.0 | Overdose deaths per 100,000 residents | population-weighted mean of county modeled overdose rates; 95% intervals from a bootstrap over counties | the most deprived tenth of counties ranked by Area Deprivation Index, among the counties with all four rankings | annual modeled rates for 2019 to 2023, averaged; rankings from ACS 2019-2023 | 95% CI 24.8 to 35.3 | 221 counties | 2019-2023 ACS residents | CDC injury and violence mapping |   |
| alt_adi_rho | +0.14 | rank correlation | Spearman rank correlation between bin order and bin mean, each bin counted once | the bins of counties ranked by Area Deprivation Index | annual modeled rates for 2019 to 2023, averaged; rankings from ACS 2019-2023 |   |   | each bin counted once | CDC injury and violence mapping |   |
| alt_degree_least | 24.3 | Overdose deaths per 100,000 residents | population-weighted mean of county modeled overdose rates; 95% intervals from a bootstrap over counties | the least deprived tenth of counties ranked by Bachelor's degree share, among the counties with all four rankings | annual modeled rates for 2019 to 2023, averaged; rankings from ACS 2019-2023 | 95% CI 22.3 to 26.3 | 298 counties | 2019-2023 ACS residents | CDC injury and violence mapping |   |
| alt_degree_most | 29.4 | Overdose deaths per 100,000 residents | population-weighted mean of county modeled overdose rates; 95% intervals from a bootstrap over counties | the most deprived tenth of counties ranked by Bachelor's degree share, among the counties with all four rankings | annual modeled rates for 2019 to 2023, averaged; rankings from ACS 2019-2023 | 95% CI 26.0 to 33.0 | 251 counties | 2019-2023 ACS residents | CDC injury and violence mapping |   |
| alt_degree_rho | +0.14 | rank correlation | Spearman rank correlation between bin order and bin mean, each bin counted once | the bins of counties ranked by Bachelor's degree share | annual modeled rates for 2019 to 2023, averaged; rankings from ACS 2019-2023 |   |   | each bin counted once | CDC injury and violence mapping |   |
| alt_income_least | 22.7 | Overdose deaths per 100,000 residents | population-weighted mean of county modeled overdose rates; 95% intervals from a bootstrap over counties | the least deprived tenth of counties ranked by Median household income, among the counties with all four rankings | annual modeled rates for 2019 to 2023, averaged; rankings from ACS 2019-2023 | 95% CI 21.2 to 24.4 | 300 counties | 2019-2023 ACS residents | CDC injury and violence mapping |   |
| alt_income_most | 33.1 | Overdose deaths per 100,000 residents | population-weighted mean of county modeled overdose rates; 95% intervals from a bootstrap over counties | the most deprived tenth of counties ranked by Median household income, among the counties with all four rankings | annual modeled rates for 2019 to 2023, averaged; rankings from ACS 2019-2023 | 95% CI 28.7 to 37.6 | 247 counties | 2019-2023 ACS residents | CDC injury and violence mapping |   |
| alt_income_rho | +0.43 | rank correlation | Spearman rank correlation between bin order and bin mean, each bin counted once | the bins of counties ranked by Median household income | annual modeled rates for 2019 to 2023, averaged; rankings from ACS 2019-2023 |   |   | each bin counted once | CDC injury and violence mapping |   |
| alt_n_places | 2,699 | counties | count of counties in the common sample of the ranking comparison | counties with the outcome and all four rankings | annual modeled rates for 2019 to 2023, averaged; rankings from ACS 2019-2023 |   |   | none (full count) | CDC injury and violence mapping |   |
| alt_rho_cut | 0.9 | absolute rank correlation | threshold between a steady and an uneven gradient | the rule for changed gradients (docs/RANKING_SWITCH_RULE.md) | annual modeled rates for 2019 to 2023, averaged; rankings from ACS 2019-2023 |   |   | none (a fixed rule) | CDC injury and violence mapping |   |
| alt_rho_margin | 0.1 | absolute rank correlation | smallest move in the rank correlation that counts as a change of shape | the rule for changed gradients (docs/RANKING_SWITCH_RULE.md) | annual modeled rates for 2019 to 2023, averaged; rankings from ACS 2019-2023 |   |   | none (a fixed rule) | CDC injury and violence mapping |   |
| alt_std_least | 21.7 | Overdose deaths per 100,000 residents | population-weighted mean of county modeled overdose rates; 95% intervals from a bootstrap over counties | the least deprived tenth of counties ranked by Standardized index, among the counties with all four rankings | annual modeled rates for 2019 to 2023, averaged; rankings from ACS 2019-2023 | 95% CI 20.3 to 23.2 | 300 counties | 2019-2023 ACS residents | CDC injury and violence mapping |   |
| alt_std_most | 27.0 | Overdose deaths per 100,000 residents | population-weighted mean of county modeled overdose rates; 95% intervals from a bootstrap over counties | the most deprived tenth of counties ranked by Standardized index, among the counties with all four rankings | annual modeled rates for 2019 to 2023, averaged; rankings from ACS 2019-2023 | 95% CI 20.3 to 34.6 | 243 counties | 2019-2023 ACS residents | CDC injury and violence mapping |   |
| alt_std_rho | +0.28 | rank correlation | Spearman rank correlation between bin order and bin mean, each bin counted once | the bins of counties ranked by Standardized index | annual modeled rates for 2019 to 2023, averaged; rankings from ACS 2019-2023 |   |   | each bin counted once | CDC injury and violence mapping |   |
| dom_alc_2x | 6 | counties | count of mapped counties with an alcohol-induced rate at least twice the overdose rate | mapped counties | deaths in 2018 to 2024 |   | 545 counties | none (full count) | CDC WONDER underlying cause of death |   |
| dom_alc_higher | 66 | counties | count of mapped counties with a higher crude alcohol-induced than overdose death rate | mapped counties | deaths in 2018 to 2024 |   | 545 counties | none (full count) | CDC WONDER underlying cause of death |   |
| dom_alc_midwest | 16.4% | percent of mapped counties | share of mapped counties with a higher crude alcohol-induced than overdose death rate | mapped counties in the Midwest Census region | deaths in 2018 to 2024 |   | 116 counties | none (each county counts once) | CDC WONDER underlying cause of death |   |
| dom_alc_ne | 0.0% | percent of mapped counties | share of mapped counties with a higher crude alcohol-induced than overdose death rate | mapped counties in the Northeast Census region | deaths in 2018 to 2024 |   | 94 counties | none (each county counts once) | CDC WONDER underlying cause of death |   |
| dom_alc_south | 6.5% | percent of mapped counties | share of mapped counties with a higher crude alcohol-induced than overdose death rate | mapped counties in the South Census region | deaths in 2018 to 2024 |   | 217 counties | none (each county counts once) | CDC WONDER underlying cause of death |   |
| dom_alc_west | 28.0% | percent of mapped counties | share of mapped counties with a higher crude alcohol-induced than overdose death rate | mapped counties in the West Census region | deaths in 2018 to 2024 |   | 118 counties | none (each county counts once) | CDC WONDER underlying cause of death |   |
| dom_both | 569 | counties | count of counties | counties whose drug and alcohol counts are both published in every year | deaths in 2018 to 2024 |   |   | none (full count) | CDC WONDER underlying cause of death |   |
| dom_corr | +0.25 | correlation coefficient | correlation of the county crude overdose and alcohol-induced death rates, weighted by county population; bootstrap over counties | counties whose drug and alcohol counts are both published in every year | deaths in 2018 to 2024 | 95% CI 0.16 to 0.35 | 569 counties | WONDER county population | CDC WONDER underlying cause of death |   |
| dom_corr_u | +0.30 | correlation coefficient | correlation of the county crude overdose and alcohol-induced death rates, counting every county once; bootstrap over counties | counties whose drug and alcohol counts are both published in every year | deaths in 2018 to 2024 | 95% CI 0.20 to 0.43 | 569 counties | none (each county counts once) | CDC WONDER underlying cause of death |   |
| dom_f2 | 2 | times | definition: the ratio at which the first class boundaries of the dominance map fall | county overdose and alcohol-induced death rates | deaths in 2018 to 2024 |   |   | none (definition) | CDC WONDER underlying cause of death |   |
| dom_f4 | 4 | times | definition: the ratio at which the second class boundaries of the dominance map fall | county overdose and alcohol-induced death rates | deaths in 2018 to 2024 |   |   | none (definition) | CDC WONDER underlying cause of death |   |
| dom_n | 545 | counties | count of mapped counties | counties complete for drug and alcohol with at least 20 pooled deaths of each | deaths in 2018 to 2024 |   |   | none (full count) | CDC WONDER underlying cause of death |   |
| dom_od_2x | 234 | counties | count of mapped counties with an overdose rate at least twice the alcohol-induced rate | mapped counties | deaths in 2018 to 2024 |   | 545 counties | none (full count) | CDC WONDER underlying cause of death |   |
| dom_od_higher | 479 | counties | count of mapped counties with a higher crude overdose than alcohol-induced death rate | mapped counties | deaths in 2018 to 2024 |   | 545 counties | none (full count) | CDC WONDER underlying cause of death |   |
| dom_pop | 76.6% | percent of residents | share of those residents (person-years) living in the mapped counties | residents of the 49 states and DC outside Connecticut, which has no WONDER county population for 2022 to 2024 | WONDER population, 2018 to 2024 |   |   | WONDER county population | CDC WONDER underlying cause of death |   |
| dom_wm_share | 78.8% | percent of mapped counties with a higher alcohol-induced rate | share in the West and Midwest Census regions | mapped counties with a higher alcohol-induced than overdose rate | deaths in 2018 to 2024 |   | 66 counties | none (each county counts once) | CDC WONDER underlying cause of death |   |
| g_adi_alc | +0.13 | correlation coefficient | correlation of the county crude alcohol-induced rate with the Area Deprivation Index rank, weighted by person-years; bootstrap over counties | counties in the gradient | deaths in 2018 to 2024 | 95% CI 0.04 to 0.22 | 525 counties | person-years in the county | CDC WONDER underlying cause of death |   |
| g_adi_alc_u | +0.15 | correlation coefficient | correlation of the county crude alcohol-induced rate with the Area Deprivation Index rank, counting every county once; bootstrap over counties | counties in the gradient | deaths in 2018 to 2024 | 95% CI 0.05 to 0.22 | 525 counties | none (each county counts once) | CDC WONDER underlying cause of death |   |
| g_adi_od | +0.35 | correlation coefficient | correlation of the county crude overdose rate with the Area Deprivation Index rank, weighted by person-years; bootstrap over counties | counties in the gradient | deaths in 2018 to 2024 | 95% CI 0.24 to 0.44 | 525 counties | person-years in the county | CDC WONDER underlying cause of death |   |
| g_adi_od_u | +0.26 | correlation coefficient | correlation of the county crude overdose rate with the Area Deprivation Index rank, counting every county once; bootstrap over counties | counties in the gradient | deaths in 2018 to 2024 | 95% CI 0.19 to 0.34 | 525 counties | none (each county counts once) | CDC WONDER underlying cause of death |   |
| g_adi_su | +0.42 | correlation coefficient | correlation of the county crude suicide rate with the Area Deprivation Index rank, weighted by person-years; bootstrap over counties | counties in the gradient | deaths in 2018 to 2024 | 95% CI 0.30 to 0.51 | 525 counties | person-years in the county | CDC WONDER underlying cause of death |   |
| g_adi_su_u | +0.28 | correlation coefficient | correlation of the county crude suicide rate with the Area Deprivation Index rank, counting every county once; bootstrap over counties | counties in the gradient | deaths in 2018 to 2024 | 95% CI 0.19 to 0.36 | 525 counties | none (each county counts once) | CDC WONDER underlying cause of death |   |
| g_alc_d1 | 12.5 | deaths per 100,000 residents | pooled crude alcohol-induced rate; bootstrap over counties | residents of the least deprived tenth of the counties in the gradient | deaths in 2018 to 2024 | 95% CI 10.8 to 14.0 | 52 counties | none (crude rate from counts) | CDC WONDER underlying cause of death |   |
| g_alc_d10 | 17.6 | deaths per 100,000 residents | pooled crude alcohol-induced rate; bootstrap over counties | residents of the most deprived tenth of the counties in the gradient | deaths in 2018 to 2024 | 95% CI 13.9 to 22.7 | 53 counties | none (crude rate from counts) | CDC WONDER underlying cause of death |   |
| g_alc_d10_ex | 14.4 | deaths per 100,000 residents | pooled crude alcohol-induced rate; bootstrap over counties | residents of the most deprived tenth of the gradient, without its counties where American Indian or Alaska Native residents are at least the cutoff share | deaths in 2018 to 2024 | 95% CI 12.1 to 16.6 | 48 counties | none (crude rate from counts) | CDC WONDER underlying cause of death |   |
| g_d10_aian_n | 5 | counties | count of counties | counties in the most deprived tenth of the gradient where American Indian or Alaska Native residents are at least the cutoff share | deaths in 2018 to 2024 |   | 53 counties | none (full count) | CDC WONDER underlying cause of death |   |
| g_diff_od_alc | +0.21 | difference between two correlation coefficients | correlation with deprivation for overdose minus that for alcohol-induced causes, weighted by person-years, from one set of bootstrap draws over counties | counties in the gradient | deaths in 2018 to 2024 | 95% CI 0.09 to 0.33 | 525 counties | person-years in the county | CDC WONDER underlying cause of death |   |
| g_diff_su_alc | +0.28 | difference between two correlation coefficients | correlation with deprivation for suicide minus that for alcohol-induced causes, weighted by person-years, from one set of bootstrap draws over counties | counties in the gradient | deaths in 2018 to 2024 | 95% CI 0.18 to 0.39 | 525 counties | person-years in the county | CDC WONDER underlying cause of death |   |
| g_n | 525 | counties | count of counties in the gradient | counties whose drug, alcohol and suicide counts are published in every year and that have a deprivation rank | deaths in 2018 to 2024 |   |   | none (full count) | CDC WONDER underlying cause of death |   |
| g_od_d1 | 22.0 | deaths per 100,000 residents | pooled crude overdose rate; bootstrap over counties | residents of the least deprived tenth of the counties in the gradient | deaths in 2018 to 2024 | 95% CI 20.0 to 24.8 | 52 counties | none (crude rate from counts) | CDC WONDER underlying cause of death |   |
| g_od_d10 | 32.3 | deaths per 100,000 residents | pooled crude overdose rate; bootstrap over counties | residents of the most deprived tenth of the counties in the gradient | deaths in 2018 to 2024 | 95% CI 25.2 to 40.0 | 53 counties | none (crude rate from counts) | CDC WONDER underlying cause of death |   |
| g_od_diff | 10.3 | deaths per 100,000 residents | pooled crude overdose rate of the most deprived tenth minus that of the least deprived tenth, weighted by person-years; bootstrap over counties with draws shared by the two tenths | residents of the most deprived tenth of the gradient relative to the least deprived tenth | deaths in 2018 to 2024 | 95% CI 2.5 to 18.3 | 525 counties | person-years in the county | CDC WONDER underlying cause of death |   |
| g_pop | 76.0% | percent of residents | share of those residents (person-years) living in the counties of the gradient | residents of the 49 states and DC outside Connecticut, which has no WONDER county population for 2022 to 2024 | WONDER population, 2018 to 2024 |   |   | WONDER county population | CDC WONDER underlying cause of death |   |
| g_su_d1 | 9.9 | deaths per 100,000 residents | pooled crude suicide rate; bootstrap over counties | residents of the least deprived tenth of the counties in the gradient | deaths in 2018 to 2024 | 95% CI 9.1 to 11.1 | 52 counties | none (crude rate from counts) | CDC WONDER underlying cause of death |   |
| g_su_d10 | 16.5 | deaths per 100,000 residents | pooled crude suicide rate; bootstrap over counties | residents of the most deprived tenth of the counties in the gradient | deaths in 2018 to 2024 | 95% CI 14.2 to 18.9 | 53 counties | none (crude rate from counts) | CDC WONDER underlying cause of death |   |
| g_su_diff | 6.6 | deaths per 100,000 residents | pooled crude suicide rate of the most deprived tenth minus that of the least deprived tenth, weighted by person-years; bootstrap over counties with draws shared by the two tenths | residents of the most deprived tenth of the gradient relative to the least deprived tenth | deaths in 2018 to 2024 | 95% CI 4.0 to 9.2 | 525 counties | person-years in the county | CDC WONDER underlying cause of death |   |
| map_floor | 20 | pooled deaths | definition: the least number of pooled deaths for a county to be mapped | a county drawn on a map of WONDER counts | deaths in 2018 to 2024 |   |   | none (definition) | CDC WONDER underlying cause of death |   |
| nc_adi_od_hi | +0.35 | correlation coefficient | correlation of the county crude overdose rate with the Area Deprivation Index rank, weighted by person-years, with the larger rise applied to North Carolina's published 2023 counts | counties in the gradient | deaths in 2018 to 2024 | 95% CI 0.24 to 0.44 | 525 counties | person-years in the county | CDC WONDER underlying cause of death |   |
| nc_diff_od_hi | 10.4 | deaths per 100,000 residents | pooled crude overdose rate of the most deprived tenth minus that of the least deprived tenth, with the larger rise applied to North Carolina's published 2023 counts; bootstrap over counties with shared draws | residents of the most deprived tenth of the gradient relative to the least deprived tenth | deaths in 2018 to 2024 | 95% CI 2.6 to 18.4 | 525 counties | person-years in the county | CDC WONDER underlying cause of death |   |
| nc_hi | 19% | percent of the published count | definition: the larger rise, which brings North Carolina's 2022-to-2023 change in the state file in line with the nation's | published 2023 drug counts of those counties | deaths in 2023 |   |   | none (definition) | CDC WONDER underlying cause of death |   |
| nc_lo | 10% | percent of the published count | definition: the smaller rise applied to each published county count in the sensitivity check (a withheld cell is never changed) | published 2023 drug counts of those counties | deaths in 2023 |   |   | none (definition) | CDC WONDER underlying cause of death |   |
| nc_n | 48 | counties | count of counties | North Carolina counties whose WONDER drug count is published in every year | deaths in 2018 to 2024 |   |   | none (full count) | CDC WONDER underlying cause of death |   |
| nc_share23 | 15.8% | percent of deaths | share of their published deaths in 2018 to 2024 that fall in 2023 | drug-poisoning deaths in those counties | deaths in 2018 to 2024 |   |   | none (full count) | CDC WONDER underlying cause of death |   |
| nc_shift_hi | 3.0% | percent change in the pooled rate | change in the pooled overdose rate of those counties if each published 2023 count were raised by the larger rise | North Carolina counties whose WONDER drug count is published in every year | deaths in 2018 to 2024 |   |   | none (crude rate from counts) | CDC WONDER underlying cause of death |   |
| nc_shift_lo | 1.6% | percent change in the pooled rate | change in the pooled overdose rate of those counties if each published 2023 count were raised by the smaller rise | North Carolina counties whose WONDER drug count is published in every year | deaths in 2018 to 2024 |   |   | none (crude rate from counts) | CDC WONDER underlying cause of death |   |
| p_adi_hm | +0.30 | correlation coefficient | correlation of the county homicide rate with the county Area Deprivation Index rank, population-weighted; bootstrap over counties | counties with a stable modeled homicide rate and a deprivation rank | annual modeled rates for 2019 to 2023, averaged | 95% CI 0.25 to 0.36 | 2,793 counties | 2019-2023 ACS residents | CDC injury and violence mapping |   |
| p_adi_hm_u | +0.18 | correlation coefficient | correlation of the county homicide rate with the county Area Deprivation Index rank, counting every county once; bootstrap over counties | counties with a stable modeled homicide rate and a deprivation rank | annual modeled rates for 2019 to 2023, averaged | 95% CI 0.16 to 0.21 | 2,793 counties | none (each county counts once) | CDC injury and violence mapping |   |
| p_adi_od | +0.22 | correlation coefficient | correlation of the county overdose rate with the county Area Deprivation Index rank, population-weighted; bootstrap over counties | counties with a stable modeled overdose rate and a deprivation rank | annual modeled rates for 2019 to 2023, averaged | 95% CI 0.15 to 0.28 | 2,701 counties | 2019-2023 ACS residents | CDC injury and violence mapping |   |
| p_adi_od_u | −0.01 | correlation coefficient | correlation of the county overdose rate with the county Area Deprivation Index rank, counting every county once; bootstrap over counties | counties with a stable modeled overdose rate and a deprivation rank | annual modeled rates for 2019 to 2023, averaged | 95% CI −0.04 to 0.03 | 2,701 counties | none (each county counts once) | CDC injury and violence mapping |   |
| p_adi_su | +0.46 | correlation coefficient | correlation of the county suicide rate with the county Area Deprivation Index rank, population-weighted; bootstrap over counties | counties with a stable modeled suicide rate and a deprivation rank | annual modeled rates for 2019 to 2023, averaged | 95% CI 0.37 to 0.53 | 2,263 counties | 2019-2023 ACS residents | CDC injury and violence mapping |   |
| p_adi_su_u | +0.01 | correlation coefficient | correlation of the county suicide rate with the county Area Deprivation Index rank, counting every county once; bootstrap over counties | counties with a stable modeled suicide rate and a deprivation rank | annual modeled rates for 2019 to 2023, averaged | 95% CI −0.04 to 0.05 | 2,263 counties | none (each county counts once) | CDC injury and violence mapping |   |
| p_common_n | 1,899 | counties | count of counties | counties in CDC's county file with a stable rate for overdose, suicide and homicide in every year | 2019 to 2023 |   |   | none (full count) | CDC injury and violence mapping |   |
| p_common_pop | 93.5% | percent of residents | share of residents living in counties with a stable rate for all three kinds of death | residents of the 50 states and DC | 2019-2023 ACS residents |   |   | 2019-2023 ACS residents | CDC injury and violence mapping |   |
| p_corr_od_hm | +0.48 | correlation coefficient | correlation of the county overdose and homicide rates, population-weighted; bootstrap over counties | counties with a stable rate for all three kinds of death | annual modeled rates for 2019 to 2023, averaged | 95% CI 0.35 to 0.58 | 1,899 counties | 2019-2023 ACS residents | CDC injury and violence mapping |   |
| p_corr_od_hm_u | +0.27 | correlation coefficient | correlation of the county overdose and homicide rates, counting every county once; bootstrap over counties | counties with a stable rate for all three kinds of death | annual modeled rates for 2019 to 2023, averaged | 95% CI 0.20 to 0.33 | 1,899 counties | none (each county counts once) | CDC injury and violence mapping |   |
| p_corr_od_su | +0.13 | correlation coefficient | correlation of the county overdose and suicide rates, population-weighted; bootstrap over counties | counties with a stable rate for all three kinds of death | annual modeled rates for 2019 to 2023, averaged | 95% CI 0.05 to 0.21 | 1,899 counties | 2019-2023 ACS residents | CDC injury and violence mapping |   |
| p_corr_od_su_u | +0.19 | correlation coefficient | correlation of the county overdose and suicide rates, counting every county once; bootstrap over counties | counties with a stable rate for all three kinds of death | annual modeled rates for 2019 to 2023, averaged | 95% CI 0.14 to 0.23 | 1,899 counties | none (each county counts once) | CDC injury and violence mapping |   |
| p_corr_su_hm | −0.02 | correlation coefficient | correlation of the county suicide and homicide rates, population-weighted; bootstrap over counties | counties with a stable rate for all three kinds of death | annual modeled rates for 2019 to 2023, averaged | 95% CI −0.10 to 0.06 | 1,899 counties | 2019-2023 ACS residents | CDC injury and violence mapping |   |
| p_corr_su_hm_u | +0.02 | correlation coefficient | correlation of the county suicide and homicide rates, counting every county once; bootstrap over counties | counties with a stable rate for all three kinds of death | annual modeled rates for 2019 to 2023, averaged | 95% CI −0.02 to 0.06 | 1,899 counties | none (each county counts once) | CDC injury and violence mapping |   |
| p_hm_n | 2,807 | counties | count of counties with a stable modeled homicide rate in each of the five years | counties in CDC's county file | 2019 to 2023 |   |   | none (full count) | CDC injury and violence mapping |   |
| p_hm_pop | 97.6% | percent of residents | share of residents living in counties with a stable modeled homicide rate in each year | residents of the 50 states and DC | 2019-2023 ACS residents |   |   | 2019-2023 ACS residents | CDC injury and violence mapping |   |
| p_n | 3,135 | counties | count of counties in the file | counties and county equivalents in CDC's county file (Connecticut's eight counties are absent) | 2019 to 2023 |   |   | none (full count) | CDC injury and violence mapping |   |
| p_od_n | 2,714 | counties | count of counties with a stable modeled overdose rate in each of the five years | counties in CDC's county file | 2019 to 2023 |   |   | none (full count) | CDC injury and violence mapping |   |
| p_od_pop | 97.7% | percent of residents | share of residents living in counties with a stable modeled overdose rate in each year | residents of the 50 states and DC | 2019-2023 ACS residents |   |   | 2019-2023 ACS residents | CDC injury and violence mapping |   |
| p_su_n | 2,273 | counties | count of counties with a stable modeled suicide rate in each of the five years | counties in CDC's county file | 2019 to 2023 |   |   | none (full count) | CDC injury and violence mapping |   |
| p_su_pop | 94.9% | percent of residents | share of residents living in counties with a stable modeled suicide rate in each year | residents of the 50 states and DC | 2019-2023 ACS residents |   |   | 2019-2023 ACS residents | CDC injury and violence mapping |   |
| pd_hm_d1 | 4.8 | deaths per 100,000 residents | mean of county homicide rates, population-weighted; bootstrap over counties | residents of the least deprived tenth of counties by deprivation with a stable homicide rate | annual modeled rates for 2019 to 2023, averaged | 95% CI 4.2 to 5.4 | 301 counties | 2019-2023 ACS residents | CDC injury and violence mapping |   |
| pd_hm_d10 | 13.6 | deaths per 100,000 residents | mean of county homicide rates, population-weighted; bootstrap over counties | residents of the most deprived tenth of counties by deprivation with a stable homicide rate | annual modeled rates for 2019 to 2023, averaged | 95% CI 11.1 to 16.2 | 257 counties | 2019-2023 ACS residents | CDC injury and violence mapping |   |
| pd_hm_falls | 4 | steps between adjacent tenths | count of steps where the population-weighted mean homicide rate is lower in the more deprived tenth (point estimates) | the nine steps between ten tenths of counties | annual modeled rates for 2019 to 2023, averaged |   |   | 2019-2023 ACS residents | CDC injury and violence mapping |   |
| pd_od_d1 | 23.8 | deaths per 100,000 residents | mean of county overdose rates, population-weighted; bootstrap over counties | residents of the least deprived tenth of counties by deprivation with a stable overdose rate | annual modeled rates for 2019 to 2023, averaged | 95% CI 22.2 to 25.6 | 302 counties | 2019-2023 ACS residents | CDC injury and violence mapping |   |
| pd_od_d10 | 31.1 | deaths per 100,000 residents | mean of county overdose rates, population-weighted; bootstrap over counties | residents of the most deprived tenth of counties by deprivation with a stable overdose rate | annual modeled rates for 2019 to 2023, averaged | 95% CI 25.7 to 36.5 | 215 counties | 2019-2023 ACS residents | CDC injury and violence mapping |   |
| pd_od_falls | 4 | steps between adjacent tenths | count of steps where the population-weighted mean overdose rate is lower in the more deprived tenth (point estimates) | the nine steps between ten tenths of counties | annual modeled rates for 2019 to 2023, averaged |   |   | 2019-2023 ACS residents | CDC injury and violence mapping |   |
| pd_od_peak | 36.1 | deaths per 100,000 residents | population-weighted mean of county overdose rates; bootstrap over counties | residents of the tenth of counties with the highest overdose rate among the ten tenths | annual modeled rates for 2019 to 2023, averaged | 95% CI 30.6 to 42.0 | 283 counties | 2019-2023 ACS residents | CDC injury and violence mapping |   |
| pd_od_peak_dec | 4 | tenth of counties by deprivation (1 = least deprived) | number of the tenth with the highest population-weighted mean overdose rate | tenths of counties in CDC's county file ranked by deprivation | annual modeled rates for 2019 to 2023, averaged |   |   | none (position in the ranking) | CDC injury and violence mapping |   |
| pd_od_peak_gap | 5.0 | deaths per 100,000 residents | difference of the two population-weighted means; bootstrap over counties with shared draws (the peak is chosen after seeing the data, so the contrast can only overstate) | residents of the tenth of counties with the highest overdose rate, relative to the most deprived tenth | annual modeled rates for 2019 to 2023, averaged | 95% CI −2.6 to 13.2 | 2,701 counties | 2019-2023 ACS residents | CDC injury and violence mapping |   |
| pd_ratio_hm | 2.83 | ratio of rates | ratio of the mean county homicide rates in the two tenths, population-weighted; bootstrap over counties | the most deprived tenth of counties relative to the least deprived tenth | annual modeled rates for 2019 to 2023, averaged | 95% CI 2.24 to 3.58 | 2,793 counties | 2019-2023 ACS residents | CDC injury and violence mapping |   |
| pd_ratio_od | 1.30 | ratio of rates | ratio of the mean county overdose rates in the two tenths, population-weighted; bootstrap over counties | the most deprived tenth of counties relative to the least deprived tenth | annual modeled rates for 2019 to 2023, averaged | 95% CI 1.07 to 1.55 | 2,701 counties | 2019-2023 ACS residents | CDC injury and violence mapping |   |
| pd_ratio_su | 1.43 | ratio of rates | ratio of the mean county suicide rates in the two tenths, population-weighted; bootstrap over counties | the most deprived tenth of counties relative to the least deprived tenth | annual modeled rates for 2019 to 2023, averaged | 95% CI 1.25 to 1.63 | 2,263 counties | 2019-2023 ACS residents | CDC injury and violence mapping |   |
| pd_su_d1 | 12.2 | deaths per 100,000 residents | mean of county suicide rates, population-weighted; bootstrap over counties | residents of the least deprived tenth of counties by deprivation with a stable suicide rate | annual modeled rates for 2019 to 2023, averaged | 95% CI 11.3 to 13.2 | 267 counties | 2019-2023 ACS residents | CDC injury and violence mapping |   |
| pd_su_d10 | 17.4 | deaths per 100,000 residents | mean of county suicide rates, population-weighted; bootstrap over counties | residents of the most deprived tenth of counties by deprivation with a stable suicide rate | annual modeled rates for 2019 to 2023, averaged | 95% CI 15.7 to 19.4 | 179 counties | 2019-2023 ACS residents | CDC injury and violence mapping |   |
| pu_hm_d1 | 3.4 | deaths per 100,000 residents | mean of county homicide rates, counting every county once; bootstrap over counties | counties of the least deprived tenth of counties by deprivation with a stable homicide rate | annual modeled rates for 2019 to 2023, averaged | 95% CI 3.1 to 3.7 | 301 counties | none (each county counts once) | CDC injury and violence mapping |   |
| pu_hm_d10 | 9.5 | deaths per 100,000 residents | mean of county homicide rates, counting every county once; bootstrap over counties | counties of the most deprived tenth of counties by deprivation with a stable homicide rate | annual modeled rates for 2019 to 2023, averaged | 95% CI 8.2 to 10.9 | 257 counties | none (each county counts once) | CDC injury and violence mapping |   |
| pu_od_d1 | 22.2 | deaths per 100,000 residents | mean of county overdose rates, counting every county once; bootstrap over counties | counties of the least deprived tenth of counties by deprivation with a stable overdose rate | annual modeled rates for 2019 to 2023, averaged | 95% CI 21.1 to 23.2 | 302 counties | none (each county counts once) | CDC injury and violence mapping |   |
| pu_od_d10 | 23.0 | deaths per 100,000 residents | mean of county overdose rates, counting every county once; bootstrap over counties | counties of the most deprived tenth of counties by deprivation with a stable overdose rate | annual modeled rates for 2019 to 2023, averaged | 95% CI 19.5 to 26.4 | 215 counties | none (each county counts once) | CDC injury and violence mapping |   |
| pu_ratio_hm | 2.78 | ratio of rates | ratio of the mean county homicide rates in the two tenths, counting every county once; bootstrap over counties | the most deprived tenth of counties relative to the least deprived tenth | annual modeled rates for 2019 to 2023, averaged | 95% CI 2.34 to 3.30 | 2,793 counties | none (each county counts once) | CDC injury and violence mapping |   |
| pu_ratio_od | 1.03 | ratio of rates | ratio of the mean county overdose rates in the two tenths, counting every county once; bootstrap over counties | the most deprived tenth of counties relative to the least deprived tenth | annual modeled rates for 2019 to 2023, averaged | 95% CI 0.88 to 1.20 | 2,701 counties | none (each county counts once) | CDC injury and violence mapping |   |
| pu_ratio_su | 0.93 | ratio of rates | ratio of the mean county suicide rates in the two tenths, counting every county once; bootstrap over counties | the most deprived tenth of counties relative to the least deprived tenth | annual modeled rates for 2019 to 2023, averaged | 95% CI 0.85 to 1.02 | 2,263 counties | none (each county counts once) | CDC injury and violence mapping |   |
| pu_su_d1 | 15.3 | deaths per 100,000 residents | mean of county suicide rates, counting every county once; bootstrap over counties | counties of the least deprived tenth of counties by deprivation with a stable suicide rate | annual modeled rates for 2019 to 2023, averaged | 95% CI 14.6 to 16.1 | 267 counties | none (each county counts once) | CDC injury and violence mapping |   |
| pu_su_d10 | 14.2 | deaths per 100,000 residents | mean of county suicide rates, counting every county once; bootstrap over counties | counties of the most deprived tenth of counties by deprivation with a stable suicide rate | annual modeled rates for 2019 to 2023, averaged | 95% CI 13.2 to 15.3 | 179 counties | none (each county counts once) | CDC injury and violence mapping |   |
| rucc_big | 1 million | residents | definition: the metro area size of code 1 | metro areas in the 2023 Rural-Urban Continuum Code | 2023 |   |   | none (definition) | Rural-Urban Continuum Codes |   |
| rucc_town | 5,000 | residents | definition: the urban population that separates codes 4 to 7 from codes 8 and 9 | nonmetro counties in the 2023 Rural-Urban Continuum Code | 2023 |   |   | none (definition) | Rural-Urban Continuum Codes |   |
| sel_hi | 60,000 | residents | definition: upper end of the county size band | counties in the comparison of complete counties with the rest | 2019-2023 ACS residents |   |   | none (definition) | American Community Survey |   |
| sel_lo | 30,000 | residents | definition: lower end of the county size band | counties in the comparison of complete counties with the rest | 2019-2023 ACS residents |   |   | none (definition) | American Community Survey |   |
| sel_od_in | 50.9 | deaths per 100,000 residents | mean of the CDC panel's modeled county overdose rates, counting every county once; bootstrap over counties | counties of 30,000 to 60,000 residents whose WONDER overdose count is published in every year | annual modeled rates for 2019 to 2023, averaged | 95% CI 46.1 to 56.4 | 73 counties | none (each county counts once) | CDC injury and violence mapping |   |
| sel_od_out | 22.5 | deaths per 100,000 residents | mean of the CDC panel's modeled county overdose rates, counting every county once; bootstrap over counties | counties of 30,000 to 60,000 residents whose WONDER overdose count is not published in every year | annual modeled rates for 2019 to 2023, averaged | 95% CI 21.5 to 23.4 | 480 counties | none (each county counts once) | CDC injury and violence mapping |   |
| top_hm | 85.8% | percent of counties | share of those counties in the South region (Census regions) | counties in the highest tenth of homicide rates among counties with a stable rate | annual modeled rates for 2019 to 2023, averaged |   | 281 counties | none (each county counts once) | CDC injury and violence mapping |   |
| top_od | 68.0% | percent of counties | share of those counties in the South region (Census regions) | counties in the highest tenth of overdose rates among counties with a stable rate | annual modeled rates for 2019 to 2023, averaged |   | 272 counties | none (each county counts once) | CDC injury and violence mapping |   |
| top_su | 50.0% | percent of counties | share of those counties in the West region (Census regions) | counties in the highest tenth of suicide rates among counties with a stable rate | annual modeled rates for 2019 to 2023, averaged |   | 228 counties | none (each county counts once) | CDC injury and violence mapping |   |
| ty_hm_large | 7.6 | deaths per 100,000 residents | mean of county homicide rates, population-weighted; bootstrap over counties | residents of large metro counties with a stable modeled homicide rate | annual modeled rates for 2019 to 2023, averaged | 95% CI 6.7 to 8.7 | 431 counties | 2019-2023 ACS residents | CDC injury and violence mapping |   |
| ty_hm_least | 4.6 | deaths per 100,000 residents | mean of county homicide rates, population-weighted; bootstrap over counties | residents of least urban nonmetro counties with a stable modeled homicide rate | annual modeled rates for 2019 to 2023, averaged | 95% CI 4.2 to 5.1 | 826 counties | 2019-2023 ACS residents | CDC injury and violence mapping |   |
| ty_hm_ratio | 1.6 | ratio of rates | ratio of mean county homicide rates between the two types, population-weighted; bootstrap over counties | the two county types compared | annual modeled rates for 2019 to 2023, averaged | 95% CI 1.4 to 1.9 | 2,807 counties | 2019-2023 ACS residents | CDC injury and violence mapping |   |
| ty_od_large | 27.8 | deaths per 100,000 residents | mean of county overdose rates, population-weighted; bootstrap over counties | residents of large metro counties with a stable modeled overdose rate | annual modeled rates for 2019 to 2023, averaged | 95% CI 25.7 to 30.0 | 428 counties | 2019-2023 ACS residents | CDC injury and violence mapping |   |
| ty_od_least | 21.8 | deaths per 100,000 residents | mean of county overdose rates, population-weighted; bootstrap over counties | residents of least urban nonmetro counties with a stable modeled overdose rate | annual modeled rates for 2019 to 2023, averaged | 95% CI 20.2 to 23.5 | 733 counties | 2019-2023 ACS residents | CDC injury and violence mapping |   |
| ty_od_ratio | 1.3 | ratio of rates | ratio of mean county overdose rates between the two types, population-weighted; bootstrap over counties | the two county types compared | annual modeled rates for 2019 to 2023, averaged | 95% CI 1.1 to 1.4 | 2,714 counties | 2019-2023 ACS residents | CDC injury and violence mapping |   |
| ty_su_large | 12.2 | deaths per 100,000 residents | mean of county suicide rates, population-weighted; bootstrap over counties | residents of large metro counties with a stable modeled suicide rate | annual modeled rates for 2019 to 2023, averaged | 95% CI 11.6 to 12.9 | 387 counties | 2019-2023 ACS residents | CDC injury and violence mapping |   |
| ty_su_more | 19.7 | deaths per 100,000 residents | mean of county suicide rates, population-weighted; bootstrap over counties | residents of more urban nonmetro counties with a stable modeled suicide rate | annual modeled rates for 2019 to 2023, averaged | 95% CI 19.2 to 20.2 | 668 counties | 2019-2023 ACS residents | CDC injury and violence mapping |   |
| ty_su_ratio | 1.6 | ratio of rates | ratio of mean county suicide rates between the two types, population-weighted; bootstrap over counties | the two county types compared | annual modeled rates for 2019 to 2023, averaged | 95% CI 1.5 to 1.7 | 2,273 counties | 2019-2023 ACS residents | CDC injury and violence mapping |   |
| tyu_hm_large | 6.2 | deaths per 100,000 residents | mean of county homicide rates, counting every county once; bootstrap over counties | counties of large metro counties with a stable modeled homicide rate | annual modeled rates for 2019 to 2023, averaged | 95% CI 5.6 to 6.9 | 431 counties | none (each county counts once) | CDC injury and violence mapping |   |
| tyu_hm_least | 3.5 | deaths per 100,000 residents | mean of county homicide rates, counting every county once; bootstrap over counties | counties of least urban nonmetro counties with a stable modeled homicide rate | annual modeled rates for 2019 to 2023, averaged | 95% CI 3.2 to 3.9 | 826 counties | none (each county counts once) | CDC injury and violence mapping |   |
| tyu_hm_ratio | 1.8 | ratio of rates | ratio of mean county homicide rates between the two types, counting every county once; bootstrap over counties | the two county types compared | annual modeled rates for 2019 to 2023, averaged | 95% CI 1.5 to 2.0 | 2,807 counties | none (each county counts once) | CDC injury and violence mapping |   |
| tyu_od_large | 28.3 | deaths per 100,000 residents | mean of county overdose rates, counting every county once; bootstrap over counties | counties of large metro counties with a stable modeled overdose rate | annual modeled rates for 2019 to 2023, averaged | 95% CI 26.8 to 29.9 | 428 counties | none (each county counts once) | CDC injury and violence mapping |   |
| tyu_od_least | 16.0 | deaths per 100,000 residents | mean of county overdose rates, counting every county once; bootstrap over counties | counties of least urban nonmetro counties with a stable modeled overdose rate | annual modeled rates for 2019 to 2023, averaged | 95% CI 14.8 to 17.2 | 733 counties | none (each county counts once) | CDC injury and violence mapping |   |
| tyu_od_ratio | 1.8 | ratio of rates | ratio of mean county overdose rates between the two types, counting every county once; bootstrap over counties | the two county types compared | annual modeled rates for 2019 to 2023, averaged | 95% CI 1.6 to 2.0 | 2,714 counties | none (each county counts once) | CDC injury and violence mapping |   |
| tyu_su_large | 14.6 | deaths per 100,000 residents | mean of county suicide rates, counting every county once; bootstrap over counties | counties of large metro counties with a stable modeled suicide rate | annual modeled rates for 2019 to 2023, averaged | 95% CI 14.1 to 15.0 | 387 counties | none (each county counts once) | CDC injury and violence mapping |   |
| tyu_su_more | 19.6 | deaths per 100,000 residents | mean of county suicide rates, counting every county once; bootstrap over counties | counties of more urban nonmetro counties with a stable modeled suicide rate | annual modeled rates for 2019 to 2023, averaged | 95% CI 19.2 to 20.0 | 668 counties | none (each county counts once) | CDC injury and violence mapping |   |
| tyu_su_ratio | 1.3 | ratio of rates | ratio of mean county suicide rates between the two types, counting every county once; bootstrap over counties | the two county types compared | annual modeled rates for 2019 to 2023, averaged | 95% CI 1.3 to 1.4 | 2,273 counties | none (each county counts once) | CDC injury and violence mapping |   |
| us_homicide_cty | 7.1 | deaths per 100,000 residents | mean of county homicide rates, population-weighted; bootstrap over counties | residents of all counties with a stable modeled homicide rate and a deprivation rank | annual modeled rates for 2019 to 2023, averaged | 95% CI 6.5 to 7.7 | 2,793 counties | 2019-2023 ACS residents | CDC injury and violence mapping |   |
| us_overdose | 27.7 | deaths per 100,000 residents | mean of county overdose rates, population-weighted; bootstrap over counties | residents of all counties with a stable modeled overdose rate and a deprivation rank | annual modeled rates for 2019 to 2023, averaged | 95% CI 26.4 to 29.0 | 2,701 counties | 2019-2023 ACS residents | CDC injury and violence mapping |   |
| us_suicide | 14.3 | deaths per 100,000 residents | mean of county suicide rates, population-weighted; bootstrap over counties | residents of all counties with a stable modeled suicide rate and a deprivation rank | annual modeled rates for 2019 to 2023, averaged | 95% CI 13.8 to 14.9 | 2,263 counties | 2019-2023 ACS residents | CDC injury and violence mapping |   |
| w_cell_max | 9 | deaths in a county-year | definition: the largest county-year count that WONDER withholds | county-years in CDC WONDER | deaths in 2018 to 2024 |   |   | none (definition) | CDC WONDER underlying cause of death |   |
| w_floor | 10 | deaths in a county-year | definition: WONDER withholds a county-year count above zero and below this number | county-years in CDC WONDER | deaths in 2018 to 2024 |   |   | none (definition) | CDC WONDER underlying cause of death |   |
| w_max_alc | 29.0% | percent of deaths in the county files | largest share of the county file's alcohol-induced deaths that the withheld cells could hold, at the largest withheld count each | deaths in county-years in CDC WONDER | 2018 to 2024 |   |   | none (bound from the public threshold) | CDC WONDER underlying cause of death |   |
| w_max_od | 13.7% | percent of deaths in the county files | largest share of the county file's drug overdose deaths that the withheld cells could hold, at the largest withheld count each | deaths in county-years in CDC WONDER | 2018 to 2024 |   |   | none (bound from the public threshold) | CDC WONDER underlying cause of death |   |
| w_max_su | 28.3% | percent of deaths in the county files | largest share of the county file's suicide deaths that the withheld cells could hold, at the largest withheld count each | deaths in county-years in CDC WONDER | 2018 to 2024 |   |   | none (bound from the public threshold) | CDC WONDER underlying cause of death |   |
| w_wh_alc | 57.9% | percent of county-years | share of county-years whose alcohol-induced count WONDER withholds | county-years in CDC WONDER | 2018 to 2024 |   | 12,729 county-years | none (full count) | CDC WONDER underlying cause of death |   |
| w_wh_od | 47.3% | percent of county-years | share of county-years whose drug overdose count WONDER withholds | county-years in CDC WONDER | 2018 to 2024 |   | 10,405 county-years | none (full count) | CDC WONDER underlying cause of death |   |
| w_wh_su | 57.4% | percent of county-years | share of county-years whose suicide count WONDER withholds | county-years in CDC WONDER | 2018 to 2024 |   | 12,616 county-years | none (full count) | CDC WONDER underlying cause of death |   |
| xc_mod_all_od | 44.9% | percent of county-years that are modeled | share of county-years with a rate that CDC marks as modeled and not direct | county-years of the panel's overdose rates in all counties | 2019 to 2023 |   | 15,109 county-years | none (full count) | CDC injury and violence mapping |   |
| xc_mod_all_su | 54.0% | percent of county-years that are modeled | share of county-years with a rate that CDC marks as modeled and not direct | county-years of the panel's suicide rates in all counties | 2019 to 2023 |   | 14,432 county-years | none (full count) | CDC injury and violence mapping |   |
| xc_mod_od | 0.6% | percent of county-years that are modeled | share of county-years with a rate that CDC marks as modeled and not direct | county-years of the panel's overdose rates in counties whose WONDER count is published in every year | 2019 to 2023 |   | 4,445 county-years | none (full count) | CDC injury and violence mapping |   |
| xc_mod_su | 0.0% | percent of county-years that are modeled | share of county-years with a rate that CDC marks as modeled and not direct | county-years of the panel's suicide rates in counties whose WONDER count is published in every year | 2019 to 2023 |   | 3,590 county-years | none (full count) | CDC injury and violence mapping |   |
| xc_od | +0.995 | correlation coefficient | correlation of the county WONDER crude rate with the CDC panel's county rate, weighted by county population; bootstrap over counties | counties whose WONDER overdose count is published in every year | WONDER 2018 to 2024 against the panel's 2019 to 2023 | 95% CI 0.993 to 0.996 | 889 counties | WONDER county population | CDC WONDER underlying cause of death |   |
| xc_su | +0.995 | correlation coefficient | correlation of the county WONDER crude rate with the CDC panel's county rate, weighted by county population; bootstrap over counties | counties whose WONDER suicide count is published in every year | WONDER 2018 to 2024 against the panel's 2019 to 2023 | 95% CI 0.994 to 0.996 | 718 counties | WONDER county population | CDC WONDER underlying cause of death |   |

## Definitions

- Crude rate: Deaths or births divided by the population, with no adjustment for age. A place with older residents has a higher crude death rate even if health at each age is the same.
- Modeled estimate: A figure produced by a statistical model instead of a direct count or measurement, usually for small areas where direct data are thin, borrowing strength from surveys, neighboring areas or other data. Neighboring areas can share errors, so single extremes call for caution.
- Area Deprivation Index (ADI): A score that ranks neighborhoods by income, education, employment and housing conditions drawn from the American Community Survey. In this book a higher national percentile means a more deprived neighborhood.
- Population-weighted: Averaged so that each place counts in proportion to its number of residents, which lets a large county move the result more than a small one. An unweighted average counts every place equally.
- Suppression: The withholding of a number because the count behind it is too small to publish without risking the identification of a person. A suppressed value is missing, not zero.
- Ecological inference: Drawing conclusions about individuals from data about places or groups. A pattern across counties need not hold for the people inside them.

## Methods

The chapter compares death rates across counties from two sources: the CDC injury and violence panel, which carries modeled overdose, suicide and homicide rates for 2019 to 2023, and CDC WONDER counts for 2018 to 2024. WONDER withholds small nonzero counts, so the WONDER results use only counties whose counts are published in every year; these are larger counties, selected on the outcome as well as on size, with higher rates than other counties of similar size. Statistics across counties are computed twice, weighted by residents and counting every county once, and several results differ between the two. The panel and WONDER are compared only in the counties that are complete in WONDER.

### How the panel's county rates are built

Two sources serve different jobs. The CDC injury and violence panel (county file, data as of May 2026) carries modeled overdose, suicide and homicide death rates for every county except Connecticut's.

A county's rate is the mean of its annual rates for 2019 to 2023 and is kept only if CDC publishes a stable rate in each of the five years; 2024 is provisional and left out.

Overdose in the panel is unintentional or undetermined-intent drug overdose; suicide and homicide include every method.

Explains Table 1, Table 2 and Table 3.

### Why the WONDER results use only complete counties

CDC WONDER underlying cause of death supplies drug poisoning (overdose, any intent), alcohol-induced causes and suicide, all crude rates per 100,000 residents.

WONDER withholds a county-year count above zero and below 10 and publishes a count of 0, so the published years of a county leave out its low-count years and keep its zero years; a rate pooled over those years is biased. Nothing is imputed, bounded or derived from totals.

The chapter instead uses counties whose count is published in every year 2018 to 2024 (complete counties), for which the pooled rate is exact, and it reports how many counties and how many residents that leaves.

Complete counties are larger counties and are selected on the outcome as well as on size: among counties of 30,000 to 60,000 residents the panel's mean overdose rate is 50.9 in complete counties and 22.5 in the rest.

Withheld county-years are 47.3% of county-years for overdose, 57.9% for alcohol-induced causes and 57.4% for suicide; at 9 deaths each, they could hold at most 13.7%, 29.0% and 28.3% of the deaths in the county files.

Explains Table 4 and Table 5.

### How counties are ranked, grouped and mapped

County deprivation is the mean of the Area Deprivation Index national ranks of a county's tracts, weighted by 2019-2023 ACS tract population, with Connecticut tracts assigned to the eight legacy counties; tenths of counties hold equal numbers of counties.

County types follow the 2023 Rural-Urban Continuum Code: large metro (code 1), smaller metro (2 and 3), more urban nonmetro with an urban population of 5,000 or more (4 to 7) and least urban nonmetro with fewer (8 and 9).

Connecticut's counties have no WONDER population for 2022 to 2024 and are left out, and the shares of residents that the maps and the gradient hold divide by the residents of the other 49 states and DC.

WONDER codes Valdez-Cordova AK, Kusilvak AK and Oglala Lakota SD by their old codes, so tracts are merged to those codes for deprivation and county type, and County Health Rankings rows are reached through the successor codes. Maps need at least 20 pooled deaths.

The share of residents who are American Indian or Alaska Native comes from County Health Rankings (Census Population Estimates 2022).

Explains Table 1, Table 2, Table 3, Table 4 and Table 5.

### How counties are weighted and the two sources compared

Statistics across counties are computed twice, weighted by 2019-2023 ACS residents and counting every county once, and both are reported where they differ.

The panel and WONDER are compared only in complete counties, where the panel's rates are modeled in 0.6% of overdose county-years and 0.0% of suicide county-years, against 44.9% and 54.0% in all counties.

Explains Table 2, Table 3 and Table 5.

### How North Carolina's low 2023 drug count is handled

North Carolina's 2023 drug count in WONDER predates amended death certificates and is low; it is 15.8% of the published deaths of its 48 complete counties in the window.

Raising each of their published 2023 counts by 10% to 19% (a withheld cell is never changed) raises their pooled overdose rate by 1.6% to 3.0%, and every comparison asserted in this build holds at both increases.

Explains Table 4 and Table 5.

### How uncertainty is measured

Intervals for means, ratios, correlations and differences are 95% percentile intervals from a bootstrap that resamples whole counties (2,000 replicates, seed 12345), with shared draws for ratios and differences.

### Four rankings of the same places

The comparison follows the book's ranking switch rule, fixed on October 28, 2026 before any of these gradients was computed. The main gradient of the chapter is recomputed under four rankings of the same places, with the outcome, period, weights, bins and intervals unchanged: the published ADI; a standardized index that applies the ADI's coefficients to its measures after standardizing each one across tracts; median household income; and the share of adults 25 and older with a bachelor's degree or more. The three alternatives come from ACS tract estimates for 2019 to 2023, the period of the ADI.

All four rankings use the same places, the 2,699 counties that have the outcome and all four rankings, so the ADI panel can differ slightly from the chapter's main figure. Tracts enter only if they have every input of the standardized index, which leaves out tracts without a median rent, home value or mortgage cost. A county takes the population-weighted mean of the ranks of its common-sample tracts under the county codes of CDC's county file, which does not carry Connecticut, and the tenths are cut among the panel counties with all four rankings before those without a stable overdose rate are set aside, as in the chapter's figure.

A pattern counts as changed when the worse end switches sides, when the two ends differ under one ranking and not the other, or when the rank correlation of bin order and bin mean falls on the other side of 0.9 and moves by at least 0.1.

Explains Table 6.

## What these data cannot show

Every figure describes counties. A county rate is an average over all of its residents, so it cannot say who died, and a correlation across counties does not describe the people in them (ecological inference). The data do not measure despair. Deaths from suicide, overdose and alcohol are counted, and whether the people who live in these counties feel despair is a person-level question for Chapter 6.3, Life evaluation and daily feelings (https://vishalsingh.org/zipcode/life-evaluation-and-daily-feelings). Survey measures of drinking are in Chapter 7.2, Health behaviors, person by person (https://vishalsingh.org/zipcode/health-behaviors-person-by-person).

The WONDER figures cover a selected set of counties. Counties whose counts are published in every year hold most residents but are a minority of counties, they are larger, and they have higher rates than other counties of similar size. The correlations with deprivation among the 525 complete counties, +0.42 (95% CI 0.30 to 0.51) for suicide, +0.35 (95% CI 0.24 to 0.44) for overdose and +0.13 (95% CI 0.04 to 0.22) for alcohol-induced deaths, describe those counties only. Most counties where American Indian or Alaska Native residents are at least 10% of residents are outside the maps, as are other small counties with high rates. Withheld counts could hold up to the shares given in the methods note, and no method in this chapter recovers them.

North Carolina's drug-poisoning count for 2023 in WONDER predates the state's amended death certificates and is low. It is 15.8% of the published deaths of the 48 complete North Carolina counties in the window. Raising each of their published 2023 counts by 10% to 19% raises their pooled overdose rate by 1.6% to 3.0%.

At the larger increase the correlation of overdose with deprivation is +0.35 (95% CI 0.24 to 0.44) and the overdose gap between the most and least deprived tenths is 10.4 (95% CI 2.6 to 18.4), and no comparison in this chapter changes direction. The modeled panel covers nearly every county but borrows strength across them, so its rate for a small county is partly its neighbors' rate, and its agreement with WONDER is checked only where its rates are direct. Alcohol-induced deaths are deaths whose underlying cause is alcohol, such as alcoholic liver disease, and they leave out deaths from other conditions in which alcohol was involved.

Several results depend on the weighting. In all counties with a modeled rate, the population-weighted correlations with deprivation are +0.46 (95% CI 0.37 to 0.53) for suicide and +0.22 (95% CI 0.15 to 0.28) for overdose, and counting every county once they are +0.01 (95% CI −0.04 to 0.05) and −0.01 (95% CI −0.04 to 0.03). Weighted by population, suicide and overdose rates are higher in the most deprived tenth of counties and suicide rates are higher outside large metro counties. Counting every county once, the first holds for homicide only and the second for smaller metro and more urban nonmetro counties only. Rates are crude, so a county with an older population has a higher rate for the same risk at each age, and comparisons of rural and urban counties carry that difference. The Area Deprivation Index understates deprivation in high-cost cities. The tenths of counties are relative to the counties in each figure. The three kinds of death also overlap a little, since a suicide by drug poisoning is counted as both a suicide and a drug death. Outside the box above, nothing in this chapter supports a claim about what lowers any of these death rates.

## References

1. Case, Anne; Deaton, Angus (2015). Rising morbidity and mortality in midlife among white non-Hispanic Americans in the 21st century. Proceedings of the National Academy of Sciences 112(49): 15078-15083. https://doi.org/10.1073/pnas.1518393112
2. Case, Anne; Deaton, Angus (2020). Deaths of Despair and the Future of Capitalism. Princeton University Press.
3. Friedman, Joseph R.; Palamar, Joseph J.; Ciccarone, Daniel; Gaines, Tommi L.; Borquez, Annick; Shover, Chelsea L.; Strathdee, Steffanie A. (2026). Charting the decline of the fourth wave: US overdose deaths by race, ethnicity and substance involvement. Addiction 121(9): 2552-2558. https://doi.org/10.1111/add.70472
4. Abouk, Rahi; Pacula, Rosalie Liccardo; Powell, David (2019). Association between state laws facilitating pharmacy distribution of naloxone and risk of fatal overdose. JAMA Internal Medicine 179(6): 805-811. https://doi.org/10.1001/jamainternmed.2019.0272
5. Doleac, Jennifer L.; Mukherjee, Anita (2022). The effects of naloxone access laws on opioid abuse, mortality, and crime. Journal of Law and Economics 65(2): 211-238. https://doi.org/10.1086/719588

## Provenance

- Pipeline version: 0.1.0
- Build script: book2/chapters/overdose-alcohol-and-suicide/build.py (sha256 5d8c87039fd8fd14bafa06333bbff92057aec1ab1a0f84d76553e726b2db2dbe)
- Built on: 2026-10-07
- Site commit: c2b1ff3309cb4a10833f884da91ac0057a66ad0a

Inputs:

| File | Size (bytes) | SHA-256, first 1 MB | SHA-256, last 1 MB |
| --- | ---: | --- | --- |
| data/analytic/tract_atlas.parquet | 39,659,925 | 29d5e4d943552d54aeb6ba92aadc939d9bd15bdea553c78357dbd516757be627 | 1e10d1ecf722340f23ee2212d4999a90250f85b65fa89b39e32db06d9472e22e |
| data/raw/acs/deprivation_components_2019_2023/tract.parquet | 13,197,793 | 5545cde2cbe96a8b56e468d65388248a241aa6bd300d3c9e325d08d3f86b11da | 5c8ec454d066988b0f1a01aaa6ed34ebfc05d162e01cf9c6bb01f9198cbaff12 |
| data/staged/adi/adi_atlas_tract2020_2023.parquet | 798,078 | 5e409bc212ff871def780fa27b4cf8b616d9da420b2b9f6455517eddcd8c1a95 | 5e409bc212ff871def780fa27b4cf8b616d9da420b2b9f6455517eddcd8c1a95 |
| data/staged/crosswalks/ct_tract2020_to_old_county.parquet | 15,842 | 961ef39f31596ef65fca2c67d5f2f63e533c5dc9b49745d3bb6437844e4b831d | 961ef39f31596ef65fca2c67d5f2f63e533c5dc9b49745d3bb6437844e4b831d |
| data/staged/crosswalks/ers_rucc_county_2023.parquet | 67,278 | ad01d36a53664a2bfafbcbebb0f5a921f5b81122ffa80cf96333db23b41c191e | ad01d36a53664a2bfafbcbebb0f5a921f5b81122ffa80cf96333db23b41c191e |
| data/staged/health/chr_county_2024.parquet | 8,776,933 | 619795f7ecfec8aff10694b649eaee5e421dbeffbe4be47af4161b51ceefa5cd | c69357f1bcf630407bb9716cb445618838999558c019aaaf77d4681b6bf45c56 |
| data/staged/mortality/injury_violence_county.parquet | 722,427 | d0dfd060a3ab9eaa3a87b35d1b24b5a1e9f21af3513c10367edb651c46f40403 | d0dfd060a3ab9eaa3a87b35d1b24b5a1e9f21af3513c10367edb651c46f40403 |
| data/staged/mortality/wonder_ucd_alcohol_county_2018_2024.parquet | 289,631 | 478b6b5f8ffa061bcca1df84162d1e169babeb7832bc9818379af9be382a778f | 478b6b5f8ffa061bcca1df84162d1e169babeb7832bc9818379af9be382a778f |
| data/staged/mortality/wonder_ucd_alcohol_state_2018_2024.parquet | 16,162 | 94c821fe0bce60f99f2a300acf8e929b857a8b0f780f5634cadce0698fd665a4 | 94c821fe0bce60f99f2a300acf8e929b857a8b0f780f5634cadce0698fd665a4 |
| data/staged/mortality/wonder_ucd_drug_county_2018_2024.parquet | 308,988 | b0e5cb6dfb44fc1ce53a192faf74cb84277bb5d601a3fc32b65078a7ec03bbf1 | b0e5cb6dfb44fc1ce53a192faf74cb84277bb5d601a3fc32b65078a7ec03bbf1 |
| data/staged/mortality/wonder_ucd_drug_state_2018_2024.parquet | 17,336 | 72a7ecbb0c49a8ba9b90b22f6fad1a6d19c025c256c72799a7acf2369ebabe44 | 72a7ecbb0c49a8ba9b90b22f6fad1a6d19c025c256c72799a7acf2369ebabe44 |
| data/staged/mortality/wonder_ucd_suicide_county_2018_2024.parquet | 290,192 | 448558540a3f42296acf7f66a945f7c6794fd6f4dc9c7032a18a3d105d0541f1 | 448558540a3f42296acf7f66a945f7c6794fd6f4dc9c7032a18a3d105d0541f1 |
| data/staged/mortality/wonder_ucd_suicide_state_2018_2024.parquet | 15,948 | 7588fe51ddda48e57e88852172442b4d2bf64689882e600ebf342668ff79ea1a | 7588fe51ddda48e57e88852172442b4d2bf64689882e600ebf342668ff79ea1a |
| data/staged/ses/acs5_tract2020_2019_2023.parquet | 7,232,424 | 5fe6f730a35225f14d5d8abfc6b9bce58b52e545f309a7eac74f1dae74d7a31c | 7b4a2475b3852d3f91804716a04566ef8ceadacec2be9fb85f5eb642078efac6 |
| data/staged/svi/svi_tract2020_2022.parquet | 17,740,415 | cb3860d8edc2aaf1f2a02211a58f7a0b80457eb6ad9a37b5ec8abb2d5da9da10 | e371a4a696d5bed05d54ad7c5b6ca2504508805b847ec415c2322be7f66ec2b2 |

Outputs:

| File | SHA-256 |
| --- | --- |
| facts.json | 63d63863e78931b4e81f3b1a9139d7c8f996c5de86982d97219967944bc2f0e4 |
| figures/alt_rankings.json | d65045934802ba7de6c7fac3a6b91557460db6f900774a99b519294cb9fe8c5f |
| figures/despair_gradient.json | c5c1c87291b998d6bcb231aa0ee1f74b912268eecc99fda0d2f3f3145fc796ca |
| figures/drug_vs_alcohol_dominance.json | c81916df228a9b72d8c295da0e569c7e589a99a5204439e4a892798ac0886fca |
| figures/overdose_map_scatter.json | c42a48640410c56392168f48bbbcfb8e4e26f353e035d82429c486e86f5d9d64 |
| figures/rate_by_county_type.json | 075430d88ebd37427085b6604a1505e781377290ad9c32a3f730c27149c02187 |
| figures/three_causes_by_decile.json | 92ef9f8f26ff64531896da0f23322aae3d67e8acc808bdc546fa5587efa9bbd9 |
| figures/three_causes_maps.json | 4542cdc2b9468265db8f773c2f3c0d845887100f9876eb58b0a751aab7afb971 |
| meta.json | 8bd9610c5b742da958e023caf9204247185872271d0e2c342567f9f86816e85b |
| public/data/overdose-alcohol-and-suicide/drug_vs_alcohol_dominance.json | ff08aed3fb3b7f71f78ad152bec4c5c58ded87ff20aff2116156db2abe11caf7 |
| public/data/overdose-alcohol-and-suicide/overdose_map_scatter.json | 677dad3020d8c19fc6cac8822f5da06c0002e9828c99822d6a8fbd3a223b0f7f |
| public/data/overdose-alcohol-and-suicide/three_causes_homicide.json | c00726c24312e475ffe280f22184cbbb038fdb74ef43f5b9c9b1c679af0c8f4f |
| public/data/overdose-alcohol-and-suicide/three_causes_overdose.json | 9278d4edcf9a442668529049c116b5a08df06c6eb83859f48415fb2faccb13da |
| public/data/overdose-alcohol-and-suicide/three_causes_suicide.json | 8b54afc2684c642847b373c3268f60a35db95773d5b01d563361b4ded230a524 |
| refs.json | 4b62041a65570f9e671d959231c1e3654d5860b40f4924928183eebb7a7b8a70 |

## Machine formats

- Report (HTML): https://vishalsingh.org/zipcode/overdose-alcohol-and-suicide/report
- Report (Markdown): https://vishalsingh.org/zipcode/overdose-alcohol-and-suicide/report.md
- Every estimate (JSON): https://vishalsingh.org/zipcode/report/overdose-alcohol-and-suicide/facts.json
- Table 1, figure "three_causes_maps" (JSON): https://vishalsingh.org/zipcode/report/overdose-alcohol-and-suicide/figures/three_causes_maps.json
- Table 2, figure "three_causes_by_decile" (JSON): https://vishalsingh.org/zipcode/report/overdose-alcohol-and-suicide/figures/three_causes_by_decile.json
- Table 3, figure "rate_by_county_type" (JSON): https://vishalsingh.org/zipcode/report/overdose-alcohol-and-suicide/figures/rate_by_county_type.json
- Table 4, figure "drug_vs_alcohol_dominance" (JSON): https://vishalsingh.org/zipcode/report/overdose-alcohol-and-suicide/figures/drug_vs_alcohol_dominance.json
- Table 5, figure "despair_gradient" (JSON): https://vishalsingh.org/zipcode/report/overdose-alcohol-and-suicide/figures/despair_gradient.json
- Table 6, figure "alt_rankings" (JSON): https://vishalsingh.org/zipcode/report/overdose-alcohol-and-suicide/figures/alt_rankings.json
- Table 7, figure "overdose_map_scatter" (JSON): https://vishalsingh.org/zipcode/report/overdose-alcohol-and-suicide/figures/overdose_map_scatter.json
- Payload of "three_causes_maps" (JSON): https://vishalsingh.org/zipcode/data/overdose-alcohol-and-suicide/three_causes_overdose.json
- Payload of "drug_vs_alcohol_dominance" (JSON): https://vishalsingh.org/zipcode/data/overdose-alcohol-and-suicide/drug_vs_alcohol_dominance.json
- Payload of "overdose_map_scatter" (JSON): https://vishalsingh.org/zipcode/data/overdose-alcohol-and-suicide/overdose_map_scatter.json
- Index of every report: https://vishalsingh.org/zipcode/llms.txt; all reports in one file: https://vishalsingh.org/zipcode/llms-full.txt

---

# Mortality trends, 2018 to 2024: report

Chapter 8.3 of ZIP Code Destiny, by Vishal Singh (NYU Stern). Generated from the chapter's data files; the reading version is at https://vishalsingh.org/zipcode/mortality-trends-2018-2024. Table n of this report is figure n of the chapter.

## Identification

- Title: Mortality trends, 2018 to 2024
- Label: Last call
- Chapter: 8.3 (Part VIII, Early death; An American life)
- Question: How did deaths from overdose, alcohol, suicide and homicide rise and fall between 2018 and 2024, and did the changes differ by county deprivation?
- Dek: In CDC WONDER, the national overdose death rate peaked in 2022 at 50.5% above its 2019 level and was 8.5% above it in 2024.
- Grain: counties
- Status: published
- Published: 2026-10-09
- Updated: not revised
- Data sources:
  - CDC WONDER underlying cause of death. CDC WONDER underlying cause of death: drug poisoning (overdose, any intent), alcohol-induced deaths and suicide; states summed to the nation and counties published in every year, 2018 to 2024; years: 2018-2024; weights: none (crude rates from counts); WONDER population of the year
  - CDC injury and violence mapping. CDC Mapping Injury, Overdose, and Violence county file: modeled overdose, suicide and homicide death rates, annual 2019 to 2024 (2024 provisional); years: 2019-2024 for counties, 2022-2024 for tracts (data as of 2026-05-14); weights: WONDER county population of the year
  - Area Deprivation Index. Neighborhood Atlas Area Deprivation Index 2023, tract national rank averaged to counties; years: 2023 release (version 4.0.1; built on ACS 2019-2023); weights: 2019-2023 ACS tract population
  - Published estimates from cited studies. Shares quoted from NCHS Data Brief 522 and a research letter on alcohol-related deaths; years: as reported in each study; weights: as reported in each source
- Canonical URL: https://vishalsingh.org/zipcode/mortality-trends-2018-2024
- Report: https://vishalsingh.org/zipcode/mortality-trends-2018-2024/report
- Markdown: https://vishalsingh.org/zipcode/mortality-trends-2018-2024/report.md
- BibTeX key: singh2026mortalitytrends20182024

```bibtex
@misc{singh2026mortalitytrends20182024,
  author = {Singh, Vishal},
  title = {Mortality trends, 2018 to 2024},
  year = {2026},
  url = {https://vishalsingh.org/zipcode/mortality-trends-2018-2024},
  note = {In ZIP Code Destiny. Published 9 October 2026.}
}
```

## Principal estimates

The facts the dek and the opening prose cite, in order of appearance.

| Estimand | Population | Period | Value | Interval | n | Weights | Source | Fact id |
| --- | --- | --- | ---: | --- | ---: | --- | --- | --- |
| national crude drug overdose death rate divided by the 2019 rate, minus 1; normal interval on the log ratio of the two death counts | residents of the United States | 2022 against 2019 | 50.5% | 95% CI 49.1% to 51.9% | 107,941 deaths | none (crude rate from counts) | CDC WONDER underlying cause of death | w_od_p22 |
| national crude drug overdose death rate divided by the 2019 rate, minus 1; normal interval on the log ratio of the two death counts | residents of the United States | 2024 against 2019 | 8.5% | 95% CI 7.4% to 9.6% | 79,384 deaths | none (crude rate from counts) | CDC WONDER underlying cause of death | w_od_p24 |
| national crude drug overdose death rate: the sum of the 51 state counts over the sum of state populations; interval from the count | residents of the United States | 2019 | 21.5 | 95% CI 21.4 to 21.7 | 70,630 deaths | none (crude rate from counts) | CDC WONDER underlying cause of death | w_od_r19 |
| national crude drug overdose death rate: the sum of the 51 state counts over the sum of state populations; interval from the count | residents of the United States | 2022 | 32.4 | 95% CI 32.2 to 32.6 | 107,941 deaths | none (crude rate from counts) | CDC WONDER underlying cause of death | w_od_r22 |
| national crude drug overdose death rate: the sum of the 51 state counts over the sum of state populations; interval from the count | residents of the United States | 2024 | 23.3 | 95% CI 23.2 to 23.5 | 79,384 deaths | none (crude rate from counts) | CDC WONDER underlying cause of death | w_od_r24 |
| count of counties in the file | counties in CDC's county file (Connecticut's eight counties are absent) | 2019 to 2024 | 3,135 |   |   | none (full count) | CDC injury and violence mapping | p_n_file |
| weighted mean of county drug overdose rates in the year divided by the 2019 value, minus 1; bootstrap over counties with shared draws | residents of counties with a stable modeled drug overdose rate that year | 2022 against 2019 | 54.7% | 95% CI 50.0% to 59.7% | 3,040 counties | WONDER county population of the year | CDC injury and violence mapping | p_od_p22 |
| the panel's percent rise in the overdose rate minus WONDER's, 2019 to 2022 | residents of the United States | 2022 against 2019 | 4.2 |   |   | see the two series | CDC injury and violence mapping | rec_pts |
| weighted mean of county homicide rates in the year divided by the 2019 value, minus 1; bootstrap over counties with shared draws | residents of counties with a stable modeled homicide rate that year | 2021 against 2019 | 35.2% | 95% CI 31.2% to 39.1% | 3,051 counties | WONDER county population of the year | CDC injury and violence mapping | p_hm_p21 |

## Findings

The national overdose death rate was 21.5 (95% CI 21.4 to 21.7) per hundred thousand residents in 2019, 32.4 (95% CI 32.2 to 32.6) in 2022 and 23.3 (95% CI 23.2 to 23.5) in 2024. The 2022 rate is 50.5% (95% CI 49.1% to 51.9%) above the 2019 rate, and the 2024 rate is 8.5% (95% CI 7.4% to 9.6%) above it. These are crude rates from CDC WONDER, which counts every drug poisoning whatever its intent. CDC's modeled injury panel, which counts unintentional and undetermined-intent drug overdoses in 3,135 counties, puts the rise to 2022 at 54.7% (95% CI 50.0% to 59.7%), which is 4.2 points more. Neither number is wrong. The two sources count different sets of deaths. The section on them below finds that the difference tracks a roughly constant number of deaths added to two different bases, which fits the way the panel defines overdose but cannot be proved from the files used here.

Alcohol-induced deaths and suicide moved on other schedules. Homicide, which only the panel measures, peaked in 2021 at 35.2% (95% CI 31.2% to 39.1%) above 2019 and was back within the interval of its 2019 level in 2024. Case and Deaton group suicide, drug overdose and alcoholic liver disease as deaths of despair.[1] Chapter 8.2, Overdose, alcohol and suicide (https://vishalsingh.org/zipcode/overdose-alcohol-and-suicide) asks whether they share an address, and the paths in this chapter show that they do not share a clock.

The chapter follows the paths from 2018 to 2024 and asks which counties they reached. National figures sum counties, and the county figures split them by deprivation. The panel's 2024 values are provisional and WONDER's are final. The data used here do not include COVID deaths, so the marker for 2020 on the charts is a date and not a measured cause.

### Overdose deaths peaked in 2022 and alcohol-induced deaths in 2021, while suicide stayed near its 2019 rate

The alcohol-induced rate peaked a year before overdose, at 16.3 in 2021, 37.4% (95% CI 35.7% to 39.2%) above 2019, and was still 15.5% (95% CI 13.9% to 17.0%) above its 2019 level in 2024. The suicide rate stayed within 3.6% of its 2019 level in every year.

Table 1 uses three cuts of CDC WONDER's underlying-cause-of-death files: drug poisoning of any intent, which this chapter calls overdose, alcohol-induced deaths and suicide. National counts are the sum of the state rows, none of which has a withheld cell, and the 2022 suicide count of 49,476 equals the count CDC reports. Rates are crude. A drug poisoning recorded as a suicide is in both the overdose and the suicide counts.[2]

Table 1. The national overdose death rate rose from 21.5 per hundred thousand residents in 2019 to 32.4 in 2022 and was 23.3 in 2024, while the suicide rate stayed within 3.6% of its 2019 level

The overdose rate rose in 2020 and 2021 and stayed near its peak from 2021 to 2023, at 32.1 in 2021, 32.4 in 2022 and 31.4 in 2023, before it fell to 23.3 in 2024. The 2018 rate was 4.3% (95% CI 3.3% to 5.3%) below the 2019 rate, so the rise had begun before 2020.

The alcohol-induced rate went from 11.9 in 2019 to 14.9 in 2020 and 16.3 in 2021, and then fell to 13.7 in 2024.

The number of alcohol-induced deaths rose 25.7% (95% CI 24.0% to 27.3%) between 2019 and 2020, from 39,043 to 49,061. A study of alcohol-related deaths, a broader count that includes deaths where an alcohol-induced cause was a contributing cause and not the underlying one, reported a rise of 25.5% over the same year,[3] so the two measures rose by similar amounts.

The suicide rate was 14.5 in 2019, fell 3.6% in 2020 to 14.0, and was 14.8 in 2022.

The 2024 values carry two cautions. WONDER's 2024 counts are final, but its 2024 population estimate comes from a new vintage, and it grows 1.6% over 2023, against yearly growth of 0.3% to 0.7% in each year from 2019 to 2023. Had it grown at the mean yearly rate of 2019 to 2023, the 2024 rates would be 1.1% higher, about 1.2 points of the overdose index. The fall in 2024 is therefore a little overstated by the denominator. The 2023 overdose count is low, because North Carolina amended many 2023 overdose death certificates after WONDER's final 2023 file, so the national 2023 overdose point is lower than a count from current records would give.

### Overdose and alcohol-induced deaths ran above the 2019 rate in every year from 2020 to 2024, and suicide did not

One way to put the three paths on a common scale counts the deaths above the number a flat 2019 rate would give in each year's population (Table 2). From 2020 to 2024 that is 131,549 (95% CI 128,564 to 134,533) overdose deaths and 50,558 (95% CI 48,358 to 52,757) alcohol-induced deaths. For suicide it is 99 (95% CI −2279 to 2476), an interval that includes zero.

The overdose excess peaked at 36,225 deaths in 2022 and was still 6,200 (95% CI 5,427 to 6,972) in 2024. The alcohol-induced excess peaked at 14,780 in 2021 and was 6,259 (95% CI 5,675 to 6,842) in 2024.

Table 2. From 2020 to 2024 the United States had 131,549 more overdose deaths than a flat 2019 rate gives

These totals are arithmetic. They do not estimate how many deaths any one event accounts for. The benchmark also ignores the rise that was already under way. The 2018 rates were 4.3% and 4.1% below the 2019 rates for overdose and alcohol-induced deaths, so a benchmark that continued the earlier rise would give smaller totals. Because the rates are crude, a change in the age structure of the population would also move them.

### The CDC panel shows the same overdose path, a homicide peak a year earlier and a flat suicide line

Homicide is not in the WONDER files used here, so Table 3 turns to the county file of CDC's injury and violence dashboard, called the panel in this chapter. It carries modeled annual death rates for 3,135 counties; Connecticut's are absent. Where a county has few deaths CDC publishes a modeled rate, and where the model is unstable it publishes none. In 2019 the file has a stable rate for 2,977 counties for overdose, 3,027 for homicide and 2,880 for suicide. Overdose in the panel is unintentional or undetermined-intent drug poisoning, and suicide and homicide count every method. The national line is the mean of county rates weighted by county population, over the counties with a stable rate that year, with intervals from resampling counties.

Table 3. In CDC's modeled county panel the overdose rate peaked in 2022 at 54.7% above its 2019 level and the homicide rate in 2021 at 35.2%, while the suicide rate stayed within 4.0% of its 2019 level

On this measure the overdose rate peaked in 2022 at 54.7% (95% CI 50.0% to 59.7%) above its 2019 level and was 12.0% (95% CI 7.4% to 17.2%) above it in 2024. The homicide rate peaked a year earlier, in 2021, at 35.2% (95% CI 31.2% to 39.1%) above 2019, and its 2024 value, 3.0% (95% CI −0.2% to 6.1%) above, has an interval that includes the 2019 level. The suicide rate stayed within 4.0% of 2019.

In rates, the panel's overdose rate was 20.0 (95% CI 18.8 to 21.2) per hundred thousand residents in 2019 and 30.9 (95% CI 29.5 to 32.4) in 2022. The homicide rate was 5.8 (95% CI 5.3 to 6.3) in 2019 and 7.8 (95% CI 7.1 to 8.6) in 2021, and the suicide rate 14.4 (95% CI 13.8 to 15.0) in 2019 and 14.5 (95% CI 13.9 to 15.1) in 2024.

Restricting overdose to the counties with a stable rate in every year gives a 2022 rise of 54.5% (95% CI 49.7% to 59.1%), close to the full-county figure. A CDC report on firearm homicide, the subject of Chapter 8.1, Gun homicide and the causes of early death (https://vishalsingh.org/zipcode/gun-homicide-and-causes-of-early-death), found that the age-adjusted rate rose 34.6% from 2019 to 2020.[4] The panel's all-method homicide rate rose 27.9% (95% CI 24.9% to 30.8%) over the same year, a smaller rise in a different measure.

### The two sources give different rises for overdose, and the difference tracks deaths that only WONDER counts

WONDER's overdose rate rose 50.5% (95% CI 49.1% to 51.9%) between 2019 and 2022 and the panel's rose 54.7% (95% CI 50.0% to 59.7%). The panel's interval, which resamples counties, includes WONDER's rise, and WONDER's interval, which reflects only the randomness of the counts, does not include the panel's. The two intervals answer different questions. The difference of 4.2 points has an accounting explanation more than a statistical one.

The series differ in three ways. First, WONDER counts drug poisonings of every intent, and the panel counts unintentional and undetermined-intent poisonings. Poisonings recorded as suicide or homicide are therefore in WONDER and not in the panel. NCHS reports that in 2023 4.4% of drug overdose deaths were suicides and under 1.0% were homicides.[2] Second, the panel leaves out Connecticut, which accounts for 1,214 of WONDER's deaths in 2019, and leaving Connecticut out of WONDER moves its rise from 50.5% to 51.0% (95% CI 49.6% to 52.5%). Third, the panel has no rate for the county-years CDC marks unstable and models the rest. Counties with a stable rate in 2019 hold 99.5% of residents outside Connecticut, so the omitted counties are few.

The files used here do not record intent, so the chapter cannot split the difference by these three. It can show how the difference behaves (Table 4). The panel's deaths, implied by multiplying each county's rate by WONDER's county population, are 64,511 in 2019 and 101,642 in 2022. WONDER's count outside Connecticut is 69,416 and 106,459. The gap is 4,905 deaths in 2019 and 4,817 in 2022, and it stays within a narrow band in the years between.

Both series add about the same number of deaths between 2019 and 2022, 37,043 in WONDER and 37,132 in the panel, so the panel's smaller base turns the same increase into a larger percentage. Adding the 2019 gap to the panel's deaths in every year lowers its rise from 54.7% to 51.2%, close to WONDER's 51.0% outside Connecticut.

By 2023 the gap is 4.4% of WONDER's count outside Connecticut, which is within what the NCHS split for that year implies, 4.4% for suicides plus under 1.0% for homicides. The match covers one year only. The gap was 7.1% of WONDER's count in 2019 and 4.5% in 2022, and the intent split for 2019 is not in the files used here, so the match in 2023 is consistent with the intent explanation and does not prove it.

Table 4. WONDER's overdose rate rose 51.0% from 2019 to 2022 outside Connecticut and the panel's rose 54.7%, a difference that matches a roughly constant number of deaths the panel does not count

The chapter uses WONDER for national counts and rates and the panel for homicide and for the rates of small counties, and the text names the source each time.

### More deprived counties have higher rates; a larger overdose rise appears only in withheld counties

Table 5 splits the series by county deprivation. Counties are ranked by the Area Deprivation Index of their residents and cut into thirds with equal numbers of counties. WONDER publishes a county's count only when it is zero or at least 10, and the chapter does not impute withheld counts. It uses complete counties, those whose count is published in every year from 2018 to 2024, where a count of zero counts as published. There are 889 complete counties for overdose, 629 for alcohol-induced deaths and 718 for suicide.

Complete counties are selected on the outcome as well as on size. Chapter 10.3, Suppressed counties in death data (https://vishalsingh.org/zipcode/suppressed-counties-in-death-data) covers how WONDER's rule removes counties and why pooling a county's published years is biased, and it shows that among counties of similar size, complete counties have much higher modeled rates than the rest, except among the smallest. The set includes counties with no deaths. Of the complete counties, 99 for overdose, 39 for alcohol-induced deaths and 24 for suicide have zero deaths in every year from 2018 to 2024, and 76 of the overdose ones are in the most deprived third.

For national totals the selection matters little. The overdose rate pooled over complete counties was 48.9% (95% CI 44.0% to 53.9%) above 2019 in 2022, against 50.5% (95% CI 49.1% to 51.9%) for the nation. For comparisons between counties it matters more, so the figure also shows the panel, whose modeled rates cover nearly every county. The thirds are thirds of the counties in each version, so the WONDER thirds and the panel thirds are different sets of counties. Counties with no index value are left out of the thirds, 9 of the 889 complete overdose counties and 13 of the 2,648 panel counties for overdose.

Table 5. Among counties with published counts in every year, death rates in 2022 were higher in the most deprived third of counties than in the least deprived third, by 1.49 times for overdose, 1.22 for alcohol-induced deaths and 1.37 for suicide

In complete counties the most deprived third, which holds 293 counties for overdose, 207 for alcohol-induced deaths and 238 for suicide, had higher rates than the least deprived third in every year for all three kinds of death. In 2022 the ratio was 1.49 (95% CI 1.33 to 1.66) for overdose, 1.22 (95% CI 1.10 to 1.37) for alcohol-induced deaths and 1.37 (95% CI 1.25 to 1.49) for suicide.

The overdose rate rose by about the same proportion in the least and most deprived thirds, 48.4% (95% CI 40.7% to 56.0%) above 2019 in the least deprived third and 50.3% (95% CI 41.8% to 62.1%) in the most deprived. The difference of +1.9 (95% CI −9.4 to 16.4) index points includes zero. The same holds for alcohol-induced deaths in 2021, at 33.9% (95% CI 30.0% to 37.7%) and 38.5% (95% CI 33.4% to 44.1%).

By 2024 the overdose rate in the most deprived third was 28.9 per hundred thousand residents, level with its 2019 value of 28.9, while the least deprived third was 13.4% (95% CI 5.1% to 21.5%) above its 2019 level. The most deprived third's index minus the least deprived third's was then −13.4 (95% CI −23.4 to −1.0) index points, an interval that excludes zero.

Dropping the complete counties with no deaths and cutting the thirds again leaves the 2022 result in place. The index is then 49.0% (95% CI 41.1% to 56.7%) above 2019 in the least deprived third and 54.1% (95% CI 45.8% to 63.9%) in the most deprived, and the intervals overlap.

The panel's overdose rates for all of its counties tell a different story. Its least and most deprived thirds had the same overdose rate in 2019, 19.0 (95% CI 17.7 to 20.3) and 19.2 (95% CI 17.1 to 21.3), a ratio of 1.01 (95% CI 0.88 to 1.15). The most deprived third then rose further. In 2022 its rate was 71.0% (95% CI 62.3% to 80.7%) above its 2019 level, against 51.4% (95% CI 45.2% to 57.6%) in the least deprived third, a difference of +19.6 (95% CI 8.6 to 31.3) index points that excludes zero, and the ratio was 1.14 (95% CI 1.01 to 1.27).

In 2024 the most and least deprived thirds were 18.7% (95% CI 11.8% to 26.3%) and 13.2% (95% CI 6.6% to 20.0%) above 2019, and the difference was +5.5 (95% CI −3.9 to 15.5) index points, an interval that includes zero.

That difference between the two sources is a difference between sets of counties. Restricted to the 880 counties that are complete in WONDER and cut into WONDER's thirds, the panel's overdose rate in 2022 is 52.3% (95% CI 43.9% to 60.5%) above 2019 in the least deprived third and 52.3% (95% CI 43.0% to 64.8%) in the most deprived, a difference of +0.1 (95% CI −12.1 to 16.1) index points. The panel's 2019 ratio in those counties is 1.51 (95% CI 1.28 to 1.75), against 1.48 (95% CI 1.26 to 1.69) in WONDER itself.

In 2024 the panel inside those counties has a difference of −16.2 (95% CI −27.2 to −2.6) index points, which also excludes zero and has the sign of WONDER's. Where WONDER publishes every year, the two sources give the same answer, and Chapter 8.2, Overdose, alcohol and suicide (https://vishalsingh.org/zipcode/overdose-alcohol-and-suicide) shows the same for county rates.

In the other 1,759 balanced panel counties, those WONDER withholds in at least one year, the most deprived third rose 94.9% (95% CI 85.1% to 105.0%) above 2019 in 2022 against 60.2% (95% CI 52.4% to 68.3%) in the least deprived third, a difference of +34.7 (95% CI 21.1 to 47.3) index points. The difference was still +25.2 (95% CI 14.3 to 36.0) in 2024. Only the modeled panel covers those counties, and each has a county-year with between one death and 9 deaths. Whether the larger rise in their most deprived third is real or the model smoothing small deprived counties toward their neighbors cannot be tested with these files.

Homicide, which only the panel measures, was higher in the most deprived third throughout, with a ratio of 1.34 (95% CI 1.13 to 1.60) in 2019 and 1.48 (95% CI 1.23 to 1.80) in 2024. Both thirds rose by similar proportions to 2021, 36.2% (95% CI 30.9% to 41.3%) and 40.8% (95% CI 32.3% to 49.8%) above 2019. By 2024 the least deprived third was 2.7% (95% CI −1.2% to 6.6%) above its 2019 level, an interval that includes the 2019 level, and the most deprived third was still 13.2% (95% CI 5.9% to 20.8%) above.

Suicide rates are higher in the most deprived third, with a ratio of 1.26 (95% CI 1.15 to 1.38) in 2019 and 1.37 (95% CI 1.25 to 1.49) in 2022 in WONDER. The next section shows that the difference between the thirds in suicide rates was as large in 2018 as in 2020.

### Within the same counties, the overdose rise starts in 2020 and the fall comes in 2024

An event study asks how an outcome moved around a date, relative to a reference year. Here the date is the arrival of the pandemic in 2020 and the reference year is 2019, the last year before it. Every county met the date together, so no county is unexposed, and the design is descriptive. It shows how death rates moved within the same counties, with each county's own level removed by a fixed effect, and it does not isolate what the pandemic did. Each coefficient is the population-weighted average change in a county's death rate since 2019, in deaths per hundred thousand residents. The model has a fixed effect for every county and every year, weights each county-year by its population and clusters errors by state (DC counts as one), so that counties in one state may share shocks. It is fitted with pyfixest. Clustering by county gives intervals about 2.0 times narrower for the 2022 overdose coefficient, and the chapter reports the wider ones.

Table 6. Within the same counties, the overdose rate in 2022 was 11.2 deaths per hundred thousand residents above its 2019 level and the alcohol-induced rate in 2021 4.3 above, while suicide rates never moved by more than about 0.6 deaths per hundred thousand residents

In complete counties the overdose rate in 2022 was 11.2 (95% CI 9.5 to 13.0) deaths per hundred thousand residents above its 2019 level, with 10.9 in 2021 and 10.3 in 2023. In 2024 the coefficient was 1.8 (95% CI −1.1 to 4.8), an interval that includes zero. The alcohol-induced rate was 4.3 (95% CI 3.5 to 5.0) above in 2021 and still 1.7 (95% CI 1.2 to 2.1) above in 2024. Suicide never moved by more than about 0.6 deaths per hundred thousand residents.

The panel version uses the counties with a stable rate in every year, 2,648 for overdose, 2,755 for homicide and 2,184 for suicide. It gives the same overdose path, with 11.0 (95% CI 9.3 to 12.7) in 2022 and 2.5 (95% CI −0.1 to 5.2) in 2024, and for homicide 2.1 (95% CI 1.7 to 2.5) in 2021 and 0.2 (95% CI 0.0 to 0.5) in 2024.

WONDER's files reach back to 2018, so the complete-county version has a pre-period, and the 2018 coefficient is the check. It is −1.0 (95% CI −1.6 to −0.5) for overdose and −0.4 (95% CI −0.7 to −0.1) for alcohol-induced deaths. Both rates were already rising before 2020, though by less than in 2022. A pre-period coefficient is a check on the design and not proof of it.[5] The panel starts in 2019, so it has no pre-period.

A second version asks whether the most deprived third of counties moved differently from the least deprived third. It interacts the year indicators with membership of the most deprived third and adds year fixed effects, so each coefficient is the change since 2019 in the most deprived third minus the change in the least deprived third. This compares two groups' changes, and it supports a causal reading only if the groups would have moved alike without the shock. The chapter uses it descriptively.

Table 7. From 2019 to 2022 the overdose rate rose 5.1 deaths per hundred thousand residents more in the most deprived third of counties with published counts than in the least deprived third, with no detectable gap in 2018

In complete counties the overdose rate rose 5.1 (95% CI 1.0 to 9.2) deaths per hundred thousand residents more in the most deprived third than in the least deprived third between 2019 and 2022, and the 2018 coefficient, −0.1 (95% CI −1.7 to 1.6), shows no detectable gap before 2020. The proportional rises in the least and most deprived thirds are alike (see the previous section), so the absolute gap reflects the different 2019 bases, 28.9 deaths per hundred thousand residents in the most deprived third against 19.6 in the least deprived.

The alcohol-induced gap in 2022 was 1.3 (95% CI 0.1 to 2.5), with a 2018 coefficient of −0.2 (95% CI −0.9 to 0.5). For suicide the 2018 coefficient is already 0.8 (95% CI 0.3 to 1.3), close in size to the 2020 coefficient of 1.0 (95% CI 0.2 to 1.7). The difference in suicide rates between the thirds was as large in 2018 as in 2020, with 2019 the low point, so the later coefficients cannot be read as a response to the pandemic. In the panel the overdose gap in 2022 is 3.8 (95% CI 0.0 to 7.6) and the homicide gap is 1.0 (95% CI −0.2 to 2.3), both with intervals that reach zero.

### The fall since 2022 reached most counties with deaths, with little difference by deprivation

Between 2022 and 2024 the panel's overdose rate fell in 2,104 of the 2,624 counties that have a stable rate in both years and at least one overdose death in one of them, or 80.2% (95% CI 78.7% to 81.7%) of them (Table 8). Weighted by population, the county rate changed by −8.6 (95% CI −9.4 to −7.8) deaths per hundred thousand residents, and 86.0% (95% CI 82.5% to 89.1%) of residents live in counties where it fell.

The file has 3,135 counties, and 2,927 have a stable rate in both years. The other 303 have a rate of zero in both years, with no deaths, so their change has no direction; they are gray on the map and left out of every share. They are concentrated in the most deprived third (164 counties against 30 in the least deprived), and counting them as not lower would build a deprivation gradient out of counties with no deaths.

The share of counties with a fall was 80.0% (95% CI 77.4% to 82.5%) in the least deprived third, 82.6% (95% CI 80.0% to 85.0%) in the middle third and 77.8% (95% CI 75.0% to 80.7%) in the most deprived third. The differences by deprivation are small, and the intervals of the least and most deprived thirds overlap. The difference by region is large. The share was lowest in the West, at 62.7% (95% CI 57.8% to 67.7%), and highest in the Northeast, at 91.8% (95% CI 87.9% to 95.2%).

Table 8. The modeled overdose death rate was lower in 2024 than in 2022 in 80.2% of counties with any overdose deaths in 2022 or 2024, from 62.7% in the West to 91.8% in the Northeast

The count depends on the model. CDC models the rate of a county with few deaths, and the change for such a county is a model's change. Among the 986 counties with direct rates in both years, a rate lower in 2024 holds for 84.1% (95% CI 81.7% to 86.3%), against 77.8% (95% CI 75.7% to 79.8%) of the 1,638 with a modeled rate in at least one year, so the modeled counties do not inflate the share.

A county with the same positive rate in both years counts as no fall, which applies to 12 counties, and 14 counties have no deprivation value and are left out of the thirds. A study of death records describes the fall in overdose deaths from 2023 to 2024 as concentrated in deaths involving fentanyl. The study also reports that deaths involving stimulants without fentanyl continued to make up a growing share of overdose deaths.[6] That description is national and says nothing about which counties the fall reached, and no analysis here explains it. Chapter 8.2, Overdose, alcohol and suicide (https://vishalsingh.org/zipcode/overdose-alcohol-and-suicide) summarizes the studies of state naloxone access laws, whose results are mixed.

### What these data cannot show

Every figure describes counties and the nation. A county rate is an average over all of its residents, so it cannot say who died, and a pattern across counties need not hold for the people in them (ecological inference). The data do not measure mechanisms. The event studies describe how rates moved in the same counties around 2020 and cannot separate deaths from the virus, changes in the supply of drugs and alcohol, disrupted treatment and services, or anything else that changed in the same years. The data do not include COVID deaths.

The panel's 2024 values are provisional and may change, and WONDER's 2024 population estimate comes from a new vintage. WONDER's 2024 counts are final, but its 2023 overdose count is low for North Carolina. The size of the fall in overdose deaths in 2024 may therefore change. WONDER's county series cover complete counties, which are selected on their counts (large enough to reach the threshold every year, or very small with no deaths), and WONDER's withheld counts are not recoverable. The panel covers nearly every county but models the rates of counties with few deaths, so its rate for a small county is partly its neighbors' rate. Where WONDER publishes every year the two sources agree on the overdose rise by deprivation. In the counties WONDER withholds, only the panel has a rate, and the data cannot test whether its larger rise in the most deprived third is real or the model smoothing small counties toward their neighbors. The difference between WONDER's and the panel's national rises is consistent with the intent definitions, but the files used here do not record intent, so it is not proved.

Rates are crude, and a county with an older population has a higher rate for the same risk at each age. Thirds of counties are relative to the counties in each figure, and county deprivation is measured once, from the Area Deprivation Index, which understates deprivation in high-cost cities (Chapter 1.1, What the deprivation index measures (https://vishalsingh.org/zipcode/anatomy-of-deprivation)). The event study has one pre-period year at most, and the differential version for suicide fails that check. Homicide is available only from the panel and counts every method, so it differs from the firearm homicide in Chapter 8.1, Gun homicide and the causes of early death (https://vishalsingh.org/zipcode/gun-homicide-and-causes-of-early-death). Alcohol-induced deaths are deaths whose underlying cause is alcohol, and they leave out deaths from other conditions in which alcohol played a part. Outside the studies cited, nothing in this chapter supports a claim about what raised or lowered any of these death rates.

## Tables

### Table 1. The national overdose death rate rose from 21.5 per hundred thousand residents in 2019 to 32.4 in 2022 and was 23.3 in 2024, while the suicide rate stayed within 3.6% of its 2019 level

Crude death rates per 100,000 residents for drug poisoning of any intent (overdose), alcohol-induced deaths and suicide, United States, 2018 to 2024

| cause | conf_high | conf_low | index | index_hi | index_lo | n | rate | year |
| --- | ---: | ---: | ---: | ---: | ---: | ---: | ---: | ---: |
| Overdose | 20.75 | 20.44 | 95.69 | 96.71 | 94.69 | 67,367 | 20.59 | 2018 |
| Overdose | 21.68 | 21.36 | 100 | n/a | n/a | 70,630 | 21.52 | 2019 |
| Overdose | 28.04 | 27.68 | 129.5 | 130.8 | 128.2 | 91,799 | 27.86 | 2020 |
| Overdose | 32.34 | 31.96 | 149.4 | 150.8 | 148.0 | 106,699 | 32.15 | 2021 |
| Overdose | 32.58 | 32.19 | 150.5 | 151.9 | 149.1 | 107,941 | 32.39 | 2022 |
| Overdose | 31.54 | 31.16 | 145.7 | 147.1 | 144.3 | 105,007 | 31.35 | 2023 |
| Overdose | 23.50 | 23.18 | 108.5 | 109.6 | 107.4 | 79,384 | 23.34 | 2024 |
| Alcohol-induced deaths | 11.53 | 11.29 | 95.92 | 97.29 | 94.57 | 37,329 | 11.41 | 2018 |
| Alcohol-induced deaths | 12.01 | 11.78 | 100 | n/a | n/a | 39,043 | 11.89 | 2019 |
| Alcohol-induced deaths | 15.02 | 14.76 | 125.2 | 126.9 | 123.5 | 49,061 | 14.89 | 2020 |
| Alcohol-induced deaths | 16.49 | 16.21 | 137.4 | 139.2 | 135.7 | 54,258 | 16.35 | 2021 |
| Alcohol-induced deaths | 15.49 | 15.23 | 129.1 | 130.8 | 127.4 | 51,191 | 15.36 | 2022 |
| Alcohol-induced deaths | 14.44 | 14.19 | 120.3 | 122.0 | 118.7 | 47,938 | 14.31 | 2023 |
| Alcohol-induced deaths | 13.86 | 13.61 | 115.5 | 117.0 | 113.9 | 46,714 | 13.73 | 2024 |
| Suicide | 14.91 | 14.64 | 102.1 | 103.4 | 100.8 | 48,344 | 14.78 | 2018 |
| Suicide | 14.60 | 14.34 | 100 | n/a | n/a | 47,511 | 14.47 | 2019 |
| Suicide | 14.08 | 13.83 | 96.41 | 97.65 | 95.18 | 45,979 | 13.95 | 2020 |
| Suicide | 14.65 | 14.39 | 100.3 | 101.6 | 99.03 | 48,183 | 14.52 | 2021 |
| Suicide | 14.98 | 14.71 | 102.6 | 103.9 | 101.3 | 49,476 | 14.84 | 2022 |
| Suicide | 14.85 | 14.60 | 101.7 | 103.0 | 100.5 | 49,316 | 14.72 | 2023 |
| Suicide | 14.48 | 14.23 | 99.18 | 100.4 | 97.93 | 48,824 | 14.36 | 2024 |

Views: Deaths per 100,000 (every row); Index, 2019 = 100 (every row).

Note: The rates are WONDER crude rates summed from the state rows, which have no withheld cell, so the national series is complete. The 2022 suicide count is 49,476, the count CDC reports. WONDER's 2024 counts are final.

Note: WONDER's 2024 population estimate grows 1.6% over 2023, against yearly growth of 0.3% to 0.7% in each year from 2019 to 2023, because it comes from a new population vintage. Had it grown at the mean yearly rate of 2019 to 2023, the 2024 rates would be 1.1% higher, about 1.2 index points for overdose.

Note: Drug poisoning means poisoning of any intent (unintentional, suicide, homicide and undetermined), alcohol-induced deaths are WONDER's alcohol-induced category, and suicide means intentional self-harm by any method. A suicide by drug poisoning is in both the drug and the suicide counts. Rates are crude, not age-adjusted.

Note: North Carolina amended many 2023 overdose death certificates after WONDER's final 2023 file, so WONDER's 2023 overdose counts for that state, and the 2023 overdose points built from them, are low.

Source: CDC WONDER underlying cause of death, 2018-2024; crude rates, full counts

Data: https://vishalsingh.org/zipcode/report/mortality-trends-2018-2024/figures/national_rates.json

### Table 2. From 2020 to 2024 the United States had 131,549 more overdose deaths than a flat 2019 rate gives

Observed national crude death rate per 100,000 residents and the 2019 rate held constant, United States, 2018 to 2024

| cause | rate | series | year |
| --- | ---: | --- | ---: |
| Overdose | 20.59 | Observed | 2018 |
| Overdose | 21.52 | At the 2019 rate | 2018 |
| Overdose | 21.52 | Observed | 2019 |
| Overdose | 21.52 | At the 2019 rate | 2019 |
| Overdose | 27.86 | Observed | 2020 |
| Overdose | 21.52 | At the 2019 rate | 2020 |
| Overdose | 32.15 | Observed | 2021 |
| Overdose | 21.52 | At the 2019 rate | 2021 |
| Overdose | 32.39 | Observed | 2022 |
| Overdose | 21.52 | At the 2019 rate | 2022 |
| Overdose | 31.35 | Observed | 2023 |
| Overdose | 21.52 | At the 2019 rate | 2023 |
| Overdose | 23.34 | Observed | 2024 |
| Overdose | 21.52 | At the 2019 rate | 2024 |
| Alcohol-induced deaths | 11.41 | Observed | 2018 |
| Alcohol-induced deaths | 11.89 | At the 2019 rate | 2018 |
| Alcohol-induced deaths | 11.89 | Observed | 2019 |
| Alcohol-induced deaths | 11.89 | At the 2019 rate | 2019 |
| Alcohol-induced deaths | 14.89 | Observed | 2020 |
| Alcohol-induced deaths | 11.89 | At the 2019 rate | 2020 |
| Alcohol-induced deaths | 16.35 | Observed | 2021 |
| Alcohol-induced deaths | 11.89 | At the 2019 rate | 2021 |
| Alcohol-induced deaths | 15.36 | Observed | 2022 |
| Alcohol-induced deaths | 11.89 | At the 2019 rate | 2022 |
| Alcohol-induced deaths | 14.31 | Observed | 2023 |
| Alcohol-induced deaths | 11.89 | At the 2019 rate | 2023 |
| Alcohol-induced deaths | 13.73 | Observed | 2024 |
| Alcohol-induced deaths | 11.89 | At the 2019 rate | 2024 |
| Suicide | 14.78 | Observed | 2018 |
| Suicide | 14.47 | At the 2019 rate | 2018 |
| Suicide | 14.47 | Observed | 2019 |
| Suicide | 14.47 | At the 2019 rate | 2019 |
| Suicide | 13.95 | Observed | 2020 |
| Suicide | 14.47 | At the 2019 rate | 2020 |
| Suicide | 14.52 | Observed | 2021 |
| Suicide | 14.47 | At the 2019 rate | 2021 |
| Suicide | 14.84 | Observed | 2022 |
| Suicide | 14.47 | At the 2019 rate | 2022 |
| Suicide | 14.72 | Observed | 2023 |
| Suicide | 14.47 | At the 2019 rate | 2023 |
| Suicide | 14.36 | Observed | 2024 |
| Suicide | 14.47 | At the 2019 rate | 2024 |

Views: Overdose (cause = Overdose); Alcohol-induced deaths (cause = Alcohol-induced deaths); Suicide (cause = Suicide).

Note: The benchmark ignores the rise in the overdose and alcohol-induced rates before 2019 (the 2018 rates were 4.3% and 4.1% below the 2019 rates) and any change in the age structure of the population, since the rates are crude. A different benchmark gives a different total.

Note: WONDER's national population estimate for 2024 comes from a new vintage and grows 1.6% over 2023, which lowers the 2024 rates and the 2024 excess.

Note: North Carolina amended many 2023 overdose death certificates after WONDER's final 2023 file, so WONDER's 2023 overdose counts for that state, and the 2023 overdose points built from them, are low. The 2023 excess for overdose is therefore low.

Source: CDC WONDER underlying cause of death, 2018-2024; crude rates, full counts

Data: https://vishalsingh.org/zipcode/report/mortality-trends-2018-2024/figures/excess_deaths.json

### Table 3. In CDC's modeled county panel the overdose rate peaked in 2022 at 54.7% above its 2019 level and the homicide rate in 2021 at 35.2%, while the suicide rate stayed within 4.0% of its 2019 level

Mean of county death rates weighted by county population, relative to 2019 (2019 = 100), for overdose, homicide and suicide, counties with a stable modeled rate that year, United States, 2019 to 2024 (2024 provisional)

| cause | index | n_units | provisional | rate | rate_hi | rate_lo | year |
| --- | ---: | ---: | --- | ---: | ---: | ---: | ---: |
| Overdose | 100 | 2,977 | no | 19.97 | 21.18 | 18.77 | 2019 |
| Overdose | 131.9 | 3,012 | no | 26.35 | 27.77 | 25.02 | 2020 |
| Overdose | 154.2 | 3,045 | no | 30.80 | 32.31 | 29.36 | 2021 |
| Overdose | 154.7 | 3,040 | no | 30.90 | 32.41 | 29.51 | 2022 |
| Overdose | 150.1 | 3,035 | no | 29.97 | 31.39 | 28.62 | 2023 |
| Overdose | 112.0 | 3,006 | yes | 22.38 | 23.46 | 21.36 | 2024 |
| Homicide | 100 | 3,027 | no | 5.80 | 6.29 | 5.32 | 2019 |
| Homicide | 127.9 | 3,067 | no | 7.41 | 8.06 | 6.78 | 2020 |
| Homicide | 135.2 | 3,051 | no | 7.83 | 8.55 | 7.14 | 2021 |
| Homicide | 128.4 | 3,069 | no | 7.44 | 8.13 | 6.80 | 2022 |
| Homicide | 117.1 | 3,063 | no | 6.79 | 7.37 | 6.21 | 2023 |
| Homicide | 103.0 | 3,043 | yes | 5.97 | 6.46 | 5.51 | 2024 |
| Suicide | 100 | 2,880 | no | 14.42 | 14.99 | 13.84 | 2019 |
| Suicide | 95.97 | 2,903 | no | 13.84 | 14.38 | 13.29 | 2020 |
| Suicide | 100.3 | 2,888 | no | 14.47 | 15.04 | 13.87 | 2021 |
| Suicide | 102.7 | 2,891 | no | 14.81 | 15.36 | 14.23 | 2022 |
| Suicide | 101.8 | 2,870 | no | 14.69 | 15.24 | 14.12 | 2023 |
| Suicide | 100.7 | 2,870 | yes | 14.53 | 15.11 | 13.91 | 2024 |

Views: Index, 2019 = 100 (every row); Deaths per 100,000 (every row).

Note: The panel's rates are CDC modeled estimates: a county with few deaths gets a modeled rate, and a county where the model is unstable gets none. A county enters a year when it has a stable rate that year. In 2019 that is 2,977 counties for overdose, 3,027 for homicide and 2,880 for suicide, out of 3,135 counties in the file. Connecticut's counties are not in the file, and Puerto Rico is not included.

Note: Weights are WONDER's county populations of the year. Using only the 2,648 counties with a stable overdose rate in every year gives a 2022 overdose rate 54.5% above 2019. 2024 is provisional.

Note: Rates are crude, not age-adjusted.

Source: CDC Mapping Injury, Overdose, and Violence county file, 2019-2024, and CDC WONDER county populations; modeled estimates, population-weighted

Data: https://vishalsingh.org/zipcode/report/mortality-trends-2018-2024/figures/panel_three_clocks.json

### Table 4. WONDER's overdose rate rose 51.0% from 2019 to 2022 outside Connecticut and the panel's rose 54.7%, a difference that matches a roughly constant number of deaths the panel does not count

Drug overdose deaths and crude rates, CDC WONDER (poisoning of any intent) and the CDC county panel (unintentional or undetermined intent), 2019 to 2024

#### Table 4.1. Deaths per year

Both series add about the same number of overdose deaths from 2019 to 2022, 37,043 in WONDER outside Connecticut and 37,132 in the panel

Overdose deaths per year, CDC WONDER drug poisoning of any intent outside Connecticut and the deaths implied by the CDC panel's modeled county rates, 2019 to 2024 (the panel's 2024 values are provisional)

| deaths | provisional | series | year |
| ---: | --- | --- | ---: |
| 69416 | no | CDC WONDER, any intent, outside Connecticut | 2019 |
| 64510.6 | no | CDC panel, unintentional or undetermined intent | 2019 |
| 90428 | no | CDC WONDER, any intent, outside Connecticut | 2020 |
| 85588.1 | no | CDC panel, unintentional or undetermined intent | 2020 |
| 105147 | no | CDC WONDER, any intent, outside Connecticut | 2021 |
| 100904 | no | CDC panel, unintentional or undetermined intent | 2021 |
| 106459 | no | CDC WONDER, any intent, outside Connecticut | 2022 |
| 101642 | no | CDC panel, unintentional or undetermined intent | 2022 |
| 103689 | no | CDC WONDER, any intent, outside Connecticut | 2023 |
| 99080.1 | no | CDC panel, unintentional or undetermined intent | 2023 |
| 78389 | no | CDC WONDER, any intent, outside Connecticut | 2024 |
| 75033.4 | yes | CDC panel, unintentional or undetermined intent | 2024 |

#### Table 4.2. Rise in the rate

WONDER's overdose rate rose 51.0% outside Connecticut and the panel's rose 54.7%; adding the 2019 gap back to the panel's deaths brings its rise to 51.2%

Overdose deaths and crude rates per 100,000 residents in 2019 and 2022, CDC WONDER and the CDC panel, and the rise in the rate between the two years

| Series | Deaths, 2019 | Deaths, 2022 | Change in deaths | Rate, 2019 | Rate, 2022 | Rise in the rate (%) |
| --- | ---: | ---: | ---: | ---: | ---: | ---: |
| WONDER, all states and DC | 70,630 | 107,941 | +37,311 | 21.5 | 32.4 | 50.5 |
| WONDER, outside Connecticut | 69,416 | 106,459 | +37,043 | 21.4 | 32.3 | 51.0 |
| Panel, counties with a stable rate | 64,511 | 101,642 | +37,132 | 20.0 | 30.9 | 54.7 |
| WONDER outside Connecticut minus the panel | 4,905 | 4,817 | −89 | n/a | n/a | n/a |
| Panel plus the 2019 gap in both years | 69,416 | 106,548 | +37,132 | 21.4 | 32.3 | 51.2 |

Note: The files used here do not record intent, so the gap cannot be split here. The panel leaves out drug poisonings recorded as suicide or homicide, Connecticut, and the county-years CDC marks unstable, and it models the rates of counties with few deaths. Counties with a stable rate in 2019 hold 99.5% of residents outside Connecticut.

Note: The panel's implied deaths use WONDER's county populations, so the two series share a denominator. The panel's 2024 values are provisional (open circles) and WONDER's 2024 counts are final.

Note: North Carolina amended many 2023 overdose death certificates after WONDER's final 2023 file, so WONDER's 2023 overdose counts for that state, and the 2023 overdose points built from them, are low.

Source: CDC WONDER underlying cause of death, 2019-2024, and CDC Mapping Injury, Overdose, and Violence county file, 2019-2024; crude rates, modeled estimates, population-weighted

Data: https://vishalsingh.org/zipcode/report/mortality-trends-2018-2024/figures/overdose_two_sources.json

### Table 5. Among counties with published counts in every year, death rates in 2022 were higher in the most deprived third of counties than in the least deprived third, by 1.49 times for overdose, 1.22 for alcohol-induced deaths and 1.37 for suicide

Pooled crude death rates per 100,000 residents by third of counties ranked by Area Deprivation Index, counties whose count CDC WONDER publishes in every year 2018 to 2024, United States

| cause | conf_high | conf_low | index | index_hi | index_lo | n_units | provisional | rate | source | third | year |
| --- | ---: | ---: | ---: | ---: | ---: | ---: | --- | ---: | --- | --- | ---: |
| Overdose | 20.41 | 17.14 | 95.64 | 98.36 | 93.06 | 294 | no | 18.76 | CDC WONDER, counties with published counts in every year | Least deprived third | 2018 |
| Overdose | 21.25 | 18.06 | 100 | n/a | n/a | 294 | no | 19.61 | CDC WONDER, counties with published counts in every year | Least deprived third | 2019 |
| Overdose | 27.08 | 23.60 | 128.7 | 133.0 | 124.4 | 294 | no | 25.25 | CDC WONDER, counties with published counts in every year | Least deprived third | 2020 |
| Overdose | 30.62 | 26.97 | 146.6 | 152.9 | 140.4 | 294 | no | 28.74 | CDC WONDER, counties with published counts in every year | Least deprived third | 2021 |
| Overdose | 31.02 | 27.35 | 148.4 | 156.0 | 140.7 | 294 | no | 29.11 | CDC WONDER, counties with published counts in every year | Least deprived third | 2022 |
| Overdose | 31.14 | 27.22 | 148.4 | 157.7 | 139.2 | 294 | no | 29.10 | CDC WONDER, counties with published counts in every year | Least deprived third | 2023 |
| Overdose | 23.98 | 20.63 | 113.4 | 121.5 | 105.1 | 294 | no | 22.23 | CDC WONDER, counties with published counts in every year | Least deprived third | 2024 |
| Overdose | 28.90 | 22.89 | 94.74 | 97.22 | 92.23 | 293 | no | 25.68 | CDC WONDER, counties with published counts in every year | Middle third | 2018 |
| Overdose | 30.46 | 24.10 | 100 | n/a | n/a | 293 | no | 27.11 | CDC WONDER, counties with published counts in every year | Middle third | 2019 |
| Overdose | 39.24 | 31.57 | 130.2 | 134.5 | 125.8 | 293 | no | 35.29 | CDC WONDER, counties with published counts in every year | Middle third | 2020 |
| Overdose | 44.63 | 36.57 | 149.1 | 156.0 | 142.0 | 293 | no | 40.40 | CDC WONDER, counties with published counts in every year | Middle third | 2021 |
| Overdose | 44.57 | 36.69 | 149.2 | 156.2 | 142.4 | 293 | no | 40.45 | CDC WONDER, counties with published counts in every year | Middle third | 2022 |
| Overdose | 41.87 | 34.36 | 139.8 | 147.2 | 132.6 | 293 | no | 37.90 | CDC WONDER, counties with published counts in every year | Middle third | 2023 |
| Overdose | 30.13 | 25.11 | 101.3 | 107.7 | 95.31 | 293 | no | 27.47 | CDC WONDER, counties with published counts in every year | Middle third | 2024 |
| Overdose | 31.44 | 24.16 | 96.83 | 100.5 | 92.89 | 293 | no | 28.02 | CDC WONDER, counties with published counts in every year | Most deprived third | 2018 |
| Overdose | 32.29 | 24.93 | 100 | n/a | n/a | 293 | no | 28.94 | CDC WONDER, counties with published counts in every year | Most deprived third | 2019 |
| Overdose | 40.80 | 33.10 | 128.4 | 135.5 | 122.8 | 293 | no | 37.15 | CDC WONDER, counties with published counts in every year | Most deprived third | 2020 |
| Overdose | 47.40 | 38.89 | 149.8 | 160.9 | 141.9 | 293 | no | 43.36 | CDC WONDER, counties with published counts in every year | Most deprived third | 2021 |
| Overdose | 47.35 | 39.19 | 150.3 | 162.1 | 141.8 | 293 | no | 43.50 | CDC WONDER, counties with published counts in every year | Most deprived third | 2022 |
| Overdose | 45.63 | 37.48 | 144.4 | 154.2 | 137.2 | 293 | no | 41.80 | CDC WONDER, counties with published counts in every year | Most deprived third | 2023 |
| Overdose | 31.22 | 26.33 | 99.94 | 109.6 | 93.10 | 293 | no | 28.92 | CDC WONDER, counties with published counts in every year | Most deprived third | 2024 |
| Alcohol-induced deaths | 11.95 | 10.41 | 96.41 | 98.63 | 94.15 | 208 | no | 11.17 | CDC WONDER, counties with published counts in every year | Least deprived third | 2018 |
| Alcohol-induced deaths | 12.41 | 10.77 | 100 | n/a | n/a | 208 | no | 11.59 | CDC WONDER, counties with published counts in every year | Least deprived third | 2019 |
| Alcohol-induced deaths | 15.14 | 13.21 | 122.1 | 125.1 | 119.2 | 208 | no | 14.15 | CDC WONDER, counties with published counts in every year | Least deprived third | 2020 |
| Alcohol-induced deaths | 16.75 | 14.28 | 133.9 | 137.7 | 130.0 | 208 | no | 15.52 | CDC WONDER, counties with published counts in every year | Least deprived third | 2021 |
| Alcohol-induced deaths | 15.71 | 13.51 | 125.9 | 129.2 | 122.6 | 208 | no | 14.60 | CDC WONDER, counties with published counts in every year | Least deprived third | 2022 |
| Alcohol-induced deaths | 14.83 | 12.70 | 118.7 | 121.8 | 115.3 | 208 | no | 13.76 | CDC WONDER, counties with published counts in every year | Least deprived third | 2023 |
| Alcohol-induced deaths | 14.08 | 11.98 | 112.4 | 115.3 | 109.0 | 208 | no | 13.03 | CDC WONDER, counties with published counts in every year | Least deprived third | 2024 |
| Alcohol-induced deaths | 12.68 | 10.44 | 97.27 | 100.9 | 93.91 | 207 | no | 11.45 | CDC WONDER, counties with published counts in every year | Middle third | 2018 |
| Alcohol-induced deaths | 12.95 | 10.80 | 100 | n/a | n/a | 207 | no | 11.77 | CDC WONDER, counties with published counts in every year | Middle third | 2019 |
| Alcohol-induced deaths | 16.37 | 13.78 | 127.1 | 130.6 | 123.7 | 207 | no | 14.96 | CDC WONDER, counties with published counts in every year | Middle third | 2020 |
| Alcohol-induced deaths | 17.40 | 14.84 | 135.8 | 140.3 | 131.0 | 207 | no | 15.98 | CDC WONDER, counties with published counts in every year | Middle third | 2021 |
| Alcohol-induced deaths | 16.33 | 14.00 | 127.9 | 132.7 | 123.0 | 207 | no | 15.05 | CDC WONDER, counties with published counts in every year | Middle third | 2022 |
| Alcohol-induced deaths | 15.18 | 13.01 | 119.2 | 123.5 | 114.8 | 207 | no | 14.03 | CDC WONDER, counties with published counts in every year | Middle third | 2023 |
| Alcohol-induced deaths | 14.48 | 12.43 | 113.5 | 118.7 | 108.5 | 207 | no | 13.36 | CDC WONDER, counties with published counts in every year | Middle third | 2024 |
| Alcohol-induced deaths | 13.99 | 12.03 | 95.42 | 99.55 | 91.73 | 207 | no | 12.95 | CDC WONDER, counties with published counts in every year | Most deprived third | 2018 |
| Alcohol-induced deaths | 14.55 | 12.62 | 100 | n/a | n/a | 207 | no | 13.57 | CDC WONDER, counties with published counts in every year | Most deprived third | 2019 |
| Alcohol-induced deaths | 18.68 | 15.89 | 126.7 | 131.5 | 121.9 | 207 | no | 17.20 | CDC WONDER, counties with published counts in every year | Most deprived third | 2020 |
| Alcohol-induced deaths | 20.41 | 17.37 | 138.5 | 144.1 | 133.4 | 207 | no | 18.79 | CDC WONDER, counties with published counts in every year | Most deprived third | 2021 |
| Alcohol-induced deaths | 19.44 | 16.44 | 131.7 | 137.2 | 126.3 | 207 | no | 17.87 | CDC WONDER, counties with published counts in every year | Most deprived third | 2022 |
| Alcohol-induced deaths | 17.45 | 15.13 | 119.8 | 124.6 | 114.9 | 207 | no | 16.26 | CDC WONDER, counties with published counts in every year | Most deprived third | 2023 |
| Alcohol-induced deaths | 16.85 | 14.79 | 116.3 | 121.3 | 111.5 | 207 | no | 15.78 | CDC WONDER, counties with published counts in every year | Most deprived third | 2024 |
| Suicide | 13.58 | 11.82 | 101.3 | 103.4 | 99.14 | 239 | no | 12.66 | CDC WONDER, counties with published counts in every year | Least deprived third | 2018 |
| Suicide | 13.43 | 11.66 | 100 | n/a | n/a | 239 | no | 12.50 | CDC WONDER, counties with published counts in every year | Least deprived third | 2019 |
| Suicide | 12.70 | 11.07 | 94.53 | 96.58 | 92.48 | 239 | no | 11.82 | CDC WONDER, counties with published counts in every year | Least deprived third | 2020 |
| Suicide | 13.07 | 11.30 | 97.12 | 99.14 | 94.95 | 239 | no | 12.14 | CDC WONDER, counties with published counts in every year | Least deprived third | 2021 |
| Suicide | 13.48 | 11.76 | 100.5 | 102.5 | 98.25 | 239 | no | 12.56 | CDC WONDER, counties with published counts in every year | Least deprived third | 2022 |
| Suicide | 13.25 | 11.54 | 98.56 | 100.8 | 96.21 | 239 | no | 12.32 | CDC WONDER, counties with published counts in every year | Least deprived third | 2023 |
| Suicide | 12.92 | 11.09 | 95.73 | 98.00 | 93.52 | 239 | no | 11.96 | CDC WONDER, counties with published counts in every year | Least deprived third | 2024 |
| Suicide | 16.26 | 13.89 | 100.9 | 103.6 | 98.43 | 238 | no | 15.01 | CDC WONDER, counties with published counts in every year | Middle third | 2018 |
| Suicide | 16.08 | 13.85 | 100 | n/a | n/a | 238 | no | 14.87 | CDC WONDER, counties with published counts in every year | Middle third | 2019 |
| Suicide | 15.01 | 13.10 | 94.06 | 96.40 | 91.55 | 238 | no | 13.99 | CDC WONDER, counties with published counts in every year | Middle third | 2020 |
| Suicide | 15.86 | 13.87 | 99.47 | 102.4 | 96.58 | 238 | no | 14.79 | CDC WONDER, counties with published counts in every year | Middle third | 2021 |
| Suicide | 15.97 | 14.10 | 100.9 | 104.2 | 97.32 | 238 | no | 15.01 | CDC WONDER, counties with published counts in every year | Middle third | 2022 |
| Suicide | 15.94 | 14.07 | 100.6 | 103.6 | 97.43 | 238 | no | 14.96 | CDC WONDER, counties with published counts in every year | Middle third | 2023 |
| Suicide | 15.66 | 13.65 | 98.27 | 101.0 | 95.41 | 238 | no | 14.62 | CDC WONDER, counties with published counts in every year | Middle third | 2024 |
| Suicide | 17.74 | 15.83 | 106.2 | 109.6 | 103.0 | 238 | no | 16.74 | CDC WONDER, counties with published counts in every year | Most deprived third | 2018 |
| Suicide | 16.60 | 14.99 | 100 | n/a | n/a | 238 | no | 15.77 | CDC WONDER, counties with published counts in every year | Most deprived third | 2019 |
| Suicide | 16.95 | 15.11 | 101.6 | 105.0 | 98.31 | 238 | no | 16.02 | CDC WONDER, counties with published counts in every year | Most deprived third | 2020 |
| Suicide | 17.43 | 15.59 | 104.5 | 108.6 | 100.9 | 238 | no | 16.48 | CDC WONDER, counties with published counts in every year | Most deprived third | 2021 |
| Suicide | 18.16 | 16.29 | 109.1 | 113.1 | 105.4 | 238 | no | 17.20 | CDC WONDER, counties with published counts in every year | Most deprived third | 2022 |
| Suicide | 18.25 | 16.46 | 109.7 | 113.8 | 106.3 | 238 | no | 17.30 | CDC WONDER, counties with published counts in every year | Most deprived third | 2023 |
| Suicide | 17.58 | 15.87 | 105.8 | 109.8 | 102.2 | 238 | no | 16.69 | CDC WONDER, counties with published counts in every year | Most deprived third | 2024 |
| Overdose | 20.26 | 17.68 | 100 | n/a | n/a | 879 | no | 18.99 | CDC panel, counties with a stable rate in every year | Least deprived third | 2019 |
| Overdose | 26.50 | 23.46 | 131.4 | 134.8 | 127.7 | 879 | no | 24.95 | CDC panel, counties with a stable rate in every year | Least deprived third | 2020 |
| Overdose | 30.43 | 27.21 | 151.7 | 157.0 | 146.7 | 879 | no | 28.79 | CDC panel, counties with a stable rate in every year | Least deprived third | 2021 |
| Overdose | 30.30 | 27.19 | 151.4 | 157.6 | 145.2 | 879 | no | 28.75 | CDC panel, counties with a stable rate in every year | Least deprived third | 2022 |
| Overdose | 29.83 | 26.69 | 148.8 | 156.1 | 141.5 | 879 | no | 28.26 | CDC panel, counties with a stable rate in every year | Least deprived third | 2023 |
| Overdose | 22.77 | 20.26 | 113.2 | 120.0 | 106.6 | 879 | yes | 21.49 | CDC panel, counties with a stable rate in every year | Least deprived third | 2024 |
| Overdose | 26.97 | 21.01 | 100 | n/a | n/a | 878 | no | 23.86 | CDC panel, counties with a stable rate in every year | Middle third | 2019 |
| Overdose | 34.45 | 28.17 | 130.8 | 136.4 | 125.9 | 878 | no | 31.19 | CDC panel, counties with a stable rate in every year | Middle third | 2020 |
| Overdose | 40.28 | 33.69 | 154.9 | 163.8 | 147.2 | 878 | no | 36.95 | CDC panel, counties with a stable rate in every year | Middle third | 2021 |
| Overdose | 41.21 | 34.54 | 158.5 | 168.2 | 150.0 | 878 | no | 37.81 | CDC panel, counties with a stable rate in every year | Middle third | 2022 |
| Overdose | 38.62 | 32.05 | 147.9 | 156.3 | 140.7 | 878 | no | 35.29 | CDC panel, counties with a stable rate in every year | Middle third | 2023 |
| Overdose | 28.12 | 23.28 | 107.2 | 114.1 | 101.3 | 878 | yes | 25.57 | CDC panel, counties with a stable rate in every year | Middle third | 2024 |
| Overdose | 21.32 | 17.12 | 100 | n/a | n/a | 878 | no | 19.19 | CDC panel, counties with a stable rate in every year | Most deprived third | 2019 |
| Overdose | 29.54 | 23.91 | 139.4 | 145.0 | 134.3 | 878 | no | 26.75 | CDC panel, counties with a stable rate in every year | Most deprived third | 2020 |
| Overdose | 37.00 | 29.90 | 175.0 | 184.8 | 166.0 | 878 | no | 33.59 | CDC panel, counties with a stable rate in every year | Most deprived third | 2021 |
| Overdose | 35.94 | 29.39 | 171.0 | 180.7 | 162.3 | 878 | no | 32.81 | CDC panel, counties with a stable rate in every year | Most deprived third | 2022 |
| Overdose | 35.48 | 29.10 | 168.9 | 178.1 | 160.6 | 878 | no | 32.41 | CDC panel, counties with a stable rate in every year | Most deprived third | 2023 |
| Overdose | 24.81 | 20.66 | 118.7 | 126.3 | 111.8 | 878 | yes | 22.77 | CDC panel, counties with a stable rate in every year | Most deprived third | 2024 |
| Homicide | 5.74 | 4.64 | 100 | n/a | n/a | 914 | no | 5.15 | CDC panel, counties with a stable rate in every year | Least deprived third | 2019 |
| Homicide | 7.40 | 5.83 | 127.1 | 130.6 | 123.2 | 914 | no | 6.55 | CDC panel, counties with a stable rate in every year | Least deprived third | 2020 |
| Homicide | 7.97 | 6.20 | 136.2 | 141.3 | 130.9 | 914 | no | 7.02 | CDC panel, counties with a stable rate in every year | Least deprived third | 2021 |
| Homicide | 7.50 | 5.91 | 128.8 | 133.1 | 124.3 | 914 | no | 6.64 | CDC panel, counties with a stable rate in every year | Least deprived third | 2022 |
| Homicide | 6.74 | 5.36 | 116.5 | 120.5 | 112.3 | 914 | no | 6.00 | CDC panel, counties with a stable rate in every year | Least deprived third | 2023 |
| Homicide | 5.92 | 4.75 | 102.7 | 106.6 | 98.82 | 914 | yes | 5.29 | CDC panel, counties with a stable rate in every year | Least deprived third | 2024 |
| Homicide | 9.24 | 6.80 | 100 | n/a | n/a | 914 | no | 7.95 | CDC panel, counties with a stable rate in every year | Middle third | 2019 |
| Homicide | 11.90 | 8.72 | 128.9 | 135.7 | 122.2 | 914 | no | 10.25 | CDC panel, counties with a stable rate in every year | Middle third | 2020 |
| Homicide | 12.05 | 8.94 | 131.4 | 138.6 | 124.7 | 914 | no | 10.45 | CDC panel, counties with a stable rate in every year | Middle third | 2021 |
| Homicide | 11.59 | 8.56 | 125.9 | 133.1 | 118.7 | 914 | no | 10.01 | CDC panel, counties with a stable rate in every year | Middle third | 2022 |
| Homicide | 10.73 | 7.86 | 116.0 | 124.9 | 107.8 | 914 | no | 9.23 | CDC panel, counties with a stable rate in every year | Middle third | 2023 |
| Homicide | 9.29 | 6.94 | 101.5 | 108.6 | 94.92 | 914 | yes | 8.07 | CDC panel, counties with a stable rate in every year | Middle third | 2024 |
| Homicide | 7.81 | 6.05 | 100 | n/a | n/a | 914 | no | 6.90 | CDC panel, counties with a stable rate in every year | Most deprived third | 2019 |
| Homicide | 10.38 | 7.87 | 130.7 | 138.5 | 122.6 | 914 | no | 9.02 | CDC panel, counties with a stable rate in every year | Most deprived third | 2020 |
| Homicide | 11.28 | 8.43 | 140.8 | 149.8 | 132.3 | 914 | no | 9.72 | CDC panel, counties with a stable rate in every year | Most deprived third | 2021 |
| Homicide | 10.80 | 8.23 | 135.8 | 144.8 | 127.5 | 914 | no | 9.37 | CDC panel, counties with a stable rate in every year | Most deprived third | 2022 |
| Homicide | 10.11 | 7.69 | 127.7 | 136.2 | 119.3 | 914 | no | 8.81 | CDC panel, counties with a stable rate in every year | Most deprived third | 2023 |
| Homicide | 9.08 | 6.79 | 113.2 | 120.8 | 105.9 | 914 | yes | 7.81 | CDC panel, counties with a stable rate in every year | Most deprived third | 2024 |
| Suicide | 14.20 | 12.80 | 100 | n/a | n/a | 725 | no | 13.47 | CDC panel, counties with a stable rate in every year | Least deprived third | 2019 |
| Suicide | 13.35 | 12.04 | 94.06 | 95.56 | 92.45 | 725 | no | 12.67 | CDC panel, counties with a stable rate in every year | Least deprived third | 2020 |
| Suicide | 13.95 | 12.59 | 98.19 | 99.84 | 96.56 | 725 | no | 13.22 | CDC panel, counties with a stable rate in every year | Least deprived third | 2021 |
| Suicide | 14.26 | 12.90 | 100.5 | 102.3 | 98.63 | 725 | no | 13.54 | CDC panel, counties with a stable rate in every year | Least deprived third | 2022 |
| Suicide | 14.09 | 12.77 | 99.46 | 101.1 | 97.62 | 725 | no | 13.39 | CDC panel, counties with a stable rate in every year | Least deprived third | 2023 |
| Suicide | 14.05 | 12.64 | 98.85 | 100.6 | 97.08 | 725 | yes | 13.31 | CDC panel, counties with a stable rate in every year | Least deprived third | 2024 |
| Suicide | 16.86 | 15.64 | 100 | n/a | n/a | 724 | no | 16.20 | CDC panel, counties with a stable rate in every year | Middle third | 2019 |
| Suicide | 16.81 | 15.45 | 99.39 | 102.1 | 96.62 | 724 | no | 16.11 | CDC panel, counties with a stable rate in every year | Middle third | 2020 |
| Suicide | 17.55 | 16.12 | 103.6 | 106.6 | 100.7 | 724 | no | 16.79 | CDC panel, counties with a stable rate in every year | Middle third | 2021 |
| Suicide | 18.05 | 16.65 | 106.9 | 109.7 | 104.1 | 724 | no | 17.32 | CDC panel, counties with a stable rate in every year | Middle third | 2022 |
| Suicide | 18.01 | 16.58 | 106.6 | 109.7 | 103.6 | 724 | no | 17.27 | CDC panel, counties with a stable rate in every year | Middle third | 2023 |
| Suicide | 17.59 | 16.28 | 104.3 | 107.2 | 101.5 | 724 | yes | 16.90 | CDC panel, counties with a stable rate in every year | Middle third | 2024 |
| Suicide | 17.95 | 16.00 | 100 | n/a | n/a | 725 | no | 17.03 | CDC panel, counties with a stable rate in every year | Most deprived third | 2019 |
| Suicide | 18.00 | 15.84 | 99.81 | 103.4 | 96.23 | 725 | no | 17.00 | CDC panel, counties with a stable rate in every year | Most deprived third | 2020 |
| Suicide | 19.03 | 16.74 | 105.5 | 109.5 | 101.6 | 725 | no | 17.96 | CDC panel, counties with a stable rate in every year | Most deprived third | 2021 |
| Suicide | 18.98 | 16.79 | 105.5 | 109.3 | 101.6 | 725 | no | 17.96 | CDC panel, counties with a stable rate in every year | Most deprived third | 2022 |
| Suicide | 19.07 | 17.05 | 106.4 | 109.8 | 103.0 | 725 | no | 18.12 | CDC panel, counties with a stable rate in every year | Most deprived third | 2023 |
| Suicide | 18.88 | 16.76 | 104.9 | 108.6 | 101.4 | 725 | yes | 17.88 | CDC panel, counties with a stable rate in every year | Most deprived third | 2024 |
| Overdose, counties complete in WONDER | 19.68 | 16.61 | 100 | n/a | n/a | 294 | no | 18.11 | CDC panel, overdose, counties complete in WONDER and other counties | Least deprived third | 2019 |
| Overdose, counties complete in WONDER | 25.48 | 22.06 | 130.9 | 135.6 | 126.3 | 294 | no | 23.70 | CDC panel, overdose, counties complete in WONDER and other counties | Least deprived third | 2020 |
| Overdose, counties complete in WONDER | 29.21 | 25.65 | 151.1 | 158.0 | 144.5 | 294 | no | 27.36 | CDC panel, overdose, counties complete in WONDER and other counties | Least deprived third | 2021 |
| Overdose, counties complete in WONDER | 29.45 | 25.77 | 152.3 | 160.5 | 143.9 | 294 | no | 27.57 | CDC panel, overdose, counties complete in WONDER and other counties | Least deprived third | 2022 |
| Overdose, counties complete in WONDER | 29.68 | 25.82 | 152.8 | 163.1 | 143.0 | 294 | no | 27.67 | CDC panel, overdose, counties complete in WONDER and other counties | Least deprived third | 2023 |
| Overdose, counties complete in WONDER | 22.97 | 19.68 | 117.4 | 126.2 | 108.6 | 294 | yes | 21.25 | CDC panel, overdose, counties complete in WONDER and other counties | Least deprived third | 2024 |
| Overdose, counties complete in WONDER | 28.83 | 22.63 | 100 | n/a | n/a | 293 | no | 25.57 | CDC panel, overdose, counties complete in WONDER and other counties | Middle third | 2019 |
| Overdose, counties complete in WONDER | 37.56 | 30.06 | 131.9 | 136.7 | 127.0 | 293 | no | 33.72 | CDC panel, overdose, counties complete in WONDER and other counties | Middle third | 2020 |
| Overdose, counties complete in WONDER | 43.15 | 35.17 | 152.4 | 160.1 | 144.8 | 293 | no | 38.96 | CDC panel, overdose, counties complete in WONDER and other counties | Middle third | 2021 |
| Overdose, counties complete in WONDER | 43.07 | 35.24 | 152.4 | 160.0 | 144.9 | 293 | no | 38.97 | CDC panel, overdose, counties complete in WONDER and other counties | Middle third | 2022 |
| Overdose, counties complete in WONDER | 40.38 | 32.97 | 142.5 | 150.6 | 134.7 | 293 | no | 36.45 | CDC panel, overdose, counties complete in WONDER and other counties | Middle third | 2023 |
| Overdose, counties complete in WONDER | 29.32 | 24.27 | 104.0 | 110.9 | 97.49 | 293 | yes | 26.59 | CDC panel, overdose, counties complete in WONDER and other counties | Middle third | 2024 |
| Overdose, counties complete in WONDER | 30.78 | 23.47 | 100 | n/a | n/a | 293 | no | 27.41 | CDC panel, overdose, counties complete in WONDER and other counties | Most deprived third | 2019 |
| Overdose, counties complete in WONDER | 39.04 | 31.47 | 129.3 | 137.6 | 122.8 | 293 | no | 35.44 | CDC panel, overdose, counties complete in WONDER and other counties | Most deprived third | 2020 |
| Overdose, counties complete in WONDER | 45.82 | 37.56 | 152.6 | 164.5 | 143.8 | 293 | no | 41.84 | CDC panel, overdose, counties complete in WONDER and other counties | Most deprived third | 2021 |
| Overdose, counties complete in WONDER | 45.52 | 37.53 | 152.3 | 164.8 | 143.0 | 293 | no | 41.75 | CDC panel, overdose, counties complete in WONDER and other counties | Most deprived third | 2022 |
| Overdose, counties complete in WONDER | 43.99 | 35.96 | 146.5 | 157.0 | 138.6 | 293 | no | 40.14 | CDC panel, overdose, counties complete in WONDER and other counties | Most deprived third | 2023 |
| Overdose, counties complete in WONDER | 29.98 | 25.27 | 101.2 | 111.5 | 93.85 | 293 | yes | 27.73 | CDC panel, overdose, counties complete in WONDER and other counties | Most deprived third | 2024 |
| Overdose, other counties | 13.94 | 12.55 | 100 | n/a | n/a | 585 | no | 13.22 | CDC panel, overdose, counties complete in WONDER and other counties | Least deprived third | 2019 |
| Overdose, other counties | 18.12 | 16.29 | 130.0 | 135.4 | 124.7 | 585 | no | 17.18 | CDC panel, overdose, counties complete in WONDER and other counties | Least deprived third | 2020 |
| Overdose, other counties | 22.24 | 20.05 | 159.8 | 167.2 | 152.4 | 585 | no | 21.13 | CDC panel, overdose, counties complete in WONDER and other counties | Least deprived third | 2021 |
| Overdose, other counties | 22.40 | 20.11 | 160.2 | 168.3 | 152.4 | 585 | no | 21.19 | CDC panel, overdose, counties complete in WONDER and other counties | Least deprived third | 2022 |
| Overdose, other counties | 20.77 | 18.71 | 148.7 | 156.8 | 141.3 | 585 | no | 19.66 | CDC panel, overdose, counties complete in WONDER and other counties | Least deprived third | 2023 |
| Overdose, other counties | 16.17 | 14.50 | 115.7 | 123.0 | 108.9 | 585 | yes | 15.30 | CDC panel, overdose, counties complete in WONDER and other counties | Least deprived third | 2024 |
| Overdose, other counties | 14.61 | 13.28 | 100 | n/a | n/a | 585 | no | 13.93 | CDC panel, overdose, counties complete in WONDER and other counties | Middle third | 2019 |
| Overdose, other counties | 20.58 | 18.35 | 139.8 | 145.8 | 133.7 | 585 | no | 19.47 | CDC panel, overdose, counties complete in WONDER and other counties | Middle third | 2020 |
| Overdose, other counties | 25.35 | 22.62 | 172.5 | 180.4 | 164.2 | 585 | no | 24.02 | CDC panel, overdose, counties complete in WONDER and other counties | Middle third | 2021 |
| Overdose, other counties | 25.49 | 22.64 | 173.0 | 181.0 | 164.5 | 585 | no | 24.09 | CDC panel, overdose, counties complete in WONDER and other counties | Middle third | 2022 |
| Overdose, other counties | 23.65 | 21.27 | 161.2 | 168.6 | 153.9 | 585 | no | 22.45 | CDC panel, overdose, counties complete in WONDER and other counties | Middle third | 2023 |
| Overdose, other counties | 17.95 | 16.27 | 122.7 | 128.9 | 117.0 | 585 | yes | 17.09 | CDC panel, overdose, counties complete in WONDER and other counties | Middle third | 2024 |
| Overdose, other counties | 15.05 | 13.43 | 100 | n/a | n/a | 585 | no | 14.26 | CDC panel, overdose, counties complete in WONDER and other counties | Most deprived third | 2019 |
| Overdose, other counties | 22.72 | 19.91 | 149.9 | 157.6 | 142.7 | 585 | no | 21.38 | CDC panel, overdose, counties complete in WONDER and other counties | Most deprived third | 2020 |
| Overdose, other counties | 30.08 | 26.02 | 197.6 | 209.0 | 187.2 | 585 | no | 28.18 | CDC panel, overdose, counties complete in WONDER and other counties | Most deprived third | 2021 |
| Overdose, other counties | 29.50 | 26.06 | 194.9 | 205.0 | 185.1 | 585 | no | 27.80 | CDC panel, overdose, counties complete in WONDER and other counties | Most deprived third | 2022 |
| Overdose, other counties | 29 | 24.98 | 189.5 | 200.6 | 178.0 | 585 | no | 27.03 | CDC panel, overdose, counties complete in WONDER and other counties | Most deprived third | 2023 |
| Overdose, other counties | 21.34 | 18.84 | 140.9 | 149.4 | 132.7 | 585 | yes | 20.09 | CDC panel, overdose, counties complete in WONDER and other counties | Most deprived third | 2024 |

Views: WONDER, rate (source = CDC WONDER, counties with published counts in every year); WONDER, index (source = CDC WONDER, counties with published counts in every year); Panel, rate (source = CDC panel, counties with a stable rate in every year); Panel, index (source = CDC panel, counties with a stable rate in every year); Panel, overdose in two sets of counties (source = CDC panel, overdose, counties complete in WONDER and other counties).

Note: Counties are ranked by the mean of the Area Deprivation Index national ranks of their tracts, weighted by tract population, and cut into thirds with equal numbers of counties. The index understates deprivation in high-cost cities, so a high-cost county can sit in a less deprived third than its tracts would suggest.

Note: The WONDER tabs use counties whose count is published in every year from 2018 to 2024: 889 counties for overdose, 629 for alcohol-induced deaths and 718 for suicide. They are larger counties and are selected on the outcome as well as on size (as the text explains). The panel tabs use counties with a stable modeled rate in every year: 2,648 for overdose, 2,755 for homicide and 2,184 for suicide.

Note: Counties with no Area Deprivation Index value are left out of the thirds: 9 of the 889 complete overdose counties and 13 of the 2,648 panel counties for overdose.

Note: Rates are crude, not age-adjusted. The panel's 2024 values are provisional (open circles in the panel tabs) and WONDER's 2024 counts are final.

Note: North Carolina amended many 2023 overdose death certificates after WONDER's final 2023 file, so WONDER's 2023 overdose counts for that state, and the 2023 overdose points built from them, are low. The WONDER overdose tabs include the counties of that state.

Source: CDC WONDER underlying cause of death, 2018-2024, CDC Mapping Injury, Overdose, and Violence county file, 2019-2024, and Area Deprivation Index 2023; crude rates, modeled estimates, population-weighted

Data: https://vishalsingh.org/zipcode/report/mortality-trends-2018-2024/figures/trend_by_deprivation.json

### Table 6. Within the same counties, the overdose rate in 2022 was 11.2 deaths per hundred thousand residents above its 2019 level and the alcohol-induced rate in 2021 4.3 above, while suicide rates never moved by more than about 0.6 deaths per hundred thousand residents

Two-way fixed-effects event study: population-weighted average change in a county's crude death rate since 2019, deaths per 100,000 residents, counties whose count CDC WONDER publishes in every year, 2018 to 2024, errors clustered by state

| cause | conf_high | conf_low | estimate | n_units | provisional | source | year |
| --- | ---: | ---: | ---: | ---: | --- | --- | ---: |
| Overdose | −0.456 | −1.63 | −1.04 | 889 | no | CDC WONDER, counties with published counts in every year | 2018 |
| Overdose | n/a | n/a | 0 | 889 | no | CDC WONDER, counties with published counts in every year | 2019 |
| Overdose | 7.68 | 5.65 | 6.66 | 889 | no | CDC WONDER, counties with published counts in every year | 2020 |
| Overdose | 12.52 | 9.35 | 10.94 | 889 | no | CDC WONDER, counties with published counts in every year | 2021 |
| Overdose | 12.95 | 9.50 | 11.23 | 889 | no | CDC WONDER, counties with published counts in every year | 2022 |
| Overdose | 12.64 | 8.02 | 10.33 | 889 | no | CDC WONDER, counties with published counts in every year | 2023 |
| Overdose | 4.82 | −1.14 | 1.84 | 889 | no | CDC WONDER, counties with published counts in every year | 2024 |
| Alcohol-induced deaths | −0.126 | −0.707 | −0.416 | 629 | no | CDC WONDER, counties with published counts in every year | 2018 |
| Alcohol-induced deaths | n/a | n/a | 0 | 629 | no | CDC WONDER, counties with published counts in every year | 2019 |
| Alcohol-induced deaths | 3.35 | 2.51 | 2.93 | 629 | no | CDC WONDER, counties with published counts in every year | 2020 |
| Alcohol-induced deaths | 5.00 | 3.51 | 4.26 | 629 | no | CDC WONDER, counties with published counts in every year | 2021 |
| Alcohol-induced deaths | 3.91 | 2.76 | 3.34 | 629 | no | CDC WONDER, counties with published counts in every year | 2022 |
| Alcohol-induced deaths | 2.79 | 1.84 | 2.31 | 629 | no | CDC WONDER, counties with published counts in every year | 2023 |
| Alcohol-induced deaths | 2.13 | 1.21 | 1.67 | 629 | no | CDC WONDER, counties with published counts in every year | 2024 |
| Suicide | 0.498 | 0.114 | 0.306 | 718 | no | CDC WONDER, counties with published counts in every year | 2018 |
| Suicide | n/a | n/a | 0 | 718 | no | CDC WONDER, counties with published counts in every year | 2019 |
| Suicide | −0.310 | −0.902 | −0.606 | 718 | no | CDC WONDER, counties with published counts in every year | 2020 |
| Suicide | 0.234 | −0.461 | −0.114 | 718 | no | CDC WONDER, counties with published counts in every year | 2021 |
| Suicide | 0.628 | −0.074 | 0.277 | 718 | no | CDC WONDER, counties with published counts in every year | 2022 |
| Suicide | 0.479 | −0.206 | 0.136 | 718 | no | CDC WONDER, counties with published counts in every year | 2023 |
| Suicide | 0.156 | −0.662 | −0.253 | 718 | no | CDC WONDER, counties with published counts in every year | 2024 |
| Overdose | n/a | n/a | 0 | 2,648 | no | CDC panel, counties with a stable rate in every year | 2019 |
| Overdose | 7.45 | 5.32 | 6.39 | 2,648 | no | CDC panel, counties with a stable rate in every year | 2020 |
| Overdose | 12.50 | 9.28 | 10.89 | 2,648 | no | CDC panel, counties with a stable rate in every year | 2021 |
| Overdose | 12.69 | 9.35 | 11.02 | 2,648 | no | CDC panel, counties with a stable rate in every year | 2022 |
| Overdose | 12.23 | 8.02 | 10.12 | 2,648 | no | CDC panel, counties with a stable rate in every year | 2023 |
| Overdose | 5.18 | −0.078 | 2.55 | 2,648 | yes | CDC panel, counties with a stable rate in every year | 2024 |
| Homicide | n/a | n/a | 0 | 2,755 | no | CDC panel, counties with a stable rate in every year | 2019 |
| Homicide | 1.88 | 1.38 | 1.63 | 2,755 | no | CDC panel, counties with a stable rate in every year | 2020 |
| Homicide | 2.47 | 1.71 | 2.09 | 2,755 | no | CDC panel, counties with a stable rate in every year | 2021 |
| Homicide | 2.04 | 1.38 | 1.71 | 2,755 | no | CDC panel, counties with a stable rate in every year | 2022 |
| Homicide | 1.37 | 0.751 | 1.06 | 2,755 | no | CDC panel, counties with a stable rate in every year | 2023 |
| Homicide | 0.508 | −0.012 | 0.248 | 2,755 | yes | CDC panel, counties with a stable rate in every year | 2024 |
| Suicide | n/a | n/a | 0 | 2,184 | no | CDC panel, counties with a stable rate in every year | 2019 |
| Suicide | −0.322 | −0.911 | −0.617 | 2,184 | no | CDC panel, counties with a stable rate in every year | 2020 |
| Suicide | 0.330 | −0.331 | −0.001 | 2,184 | no | CDC panel, counties with a stable rate in every year | 2021 |
| Suicide | 0.652 | −0.028 | 0.312 | 2,184 | no | CDC panel, counties with a stable rate in every year | 2022 |
| Suicide | 0.523 | −0.129 | 0.197 | 2,184 | no | CDC panel, counties with a stable rate in every year | 2023 |
| Suicide | 0.474 | −0.374 | 0.050 | 2,184 | yes | CDC panel, counties with a stable rate in every year | 2024 |

Views: CDC WONDER, 2018 to 2024 (source = CDC WONDER, counties with published counts in every year); CDC panel, 2019 to 2024 (source = CDC panel, counties with a stable rate in every year).

Note: The event is the arrival of the pandemic in 2020, treated as one common shock: every county is exposed in the same year, so there is no unexposed comparison group, and the coefficients describe how county death rates moved and not what any one thing did. County fixed effects remove each county's own level, and the weights are county populations of the year.

Note: The WONDER tab covers 889 counties for overdose, 629 for alcohol-induced deaths and 718 for suicide, in 50 state clusters (DC counts as one). The panel tab covers the counties with a stable modeled rate in every year (2,648, 2,755 and 2,184) and starts in 2019, so it has no pre-period. Clustering by county gives intervals about 2.0 times narrower for the 2022 overdose coefficient than clustering by state, and the chapter reports the wider ones.

Note: The panel's 2024 values are provisional (open circles); WONDER's 2024 counts are final.

Note: North Carolina amended many 2023 overdose death certificates after WONDER's final 2023 file, so WONDER's 2023 overdose counts for that state, and the 2023 overdose points built from them, are low. The WONDER overdose tab includes the counties of that state.

Source: CDC WONDER underlying cause of death, 2018-2024, and CDC Mapping Injury, Overdose, and Violence county file, 2019-2024; crude rates, modeled estimates, population-weighted regressions

Data: https://vishalsingh.org/zipcode/report/mortality-trends-2018-2024/figures/event_study.json

### Table 7. From 2019 to 2022 the overdose rate rose 5.1 deaths per hundred thousand residents more in the most deprived third of counties with published counts than in the least deprived third, with no detectable gap in 2018

Difference in the change in the crude death rate since 2019 between the most deprived and the least deprived third of counties, deaths per 100,000 residents, counties whose count CDC WONDER publishes in every year, 2018 to 2024, errors clustered by state

| cause | conf_high | conf_low | estimate | n_units | provisional | source | year |
| --- | ---: | ---: | ---: | ---: | --- | --- | ---: |
| Overdose | 1.56 | −1.73 | −0.084 | 587 | no | CDC WONDER, counties with published counts in every year | 2018 |
| Overdose | n/a | n/a | 0 | 587 | no | CDC WONDER, counties with published counts in every year | 2019 |
| Overdose | 4.82 | 0.370 | 2.60 | 587 | no | CDC WONDER, counties with published counts in every year | 2020 |
| Overdose | 9.27 | 1.34 | 5.30 | 587 | no | CDC WONDER, counties with published counts in every year | 2021 |
| Overdose | 9.19 | 1.03 | 5.11 | 587 | no | CDC WONDER, counties with published counts in every year | 2022 |
| Overdose | 7.75 | −0.871 | 3.44 | 587 | no | CDC WONDER, counties with published counts in every year | 2023 |
| Overdose | 2.64 | −7.75 | −2.56 | 587 | no | CDC WONDER, counties with published counts in every year | 2024 |
| Alcohol-induced deaths | 0.459 | −0.895 | −0.218 | 415 | no | CDC WONDER, counties with published counts in every year | 2018 |
| Alcohol-induced deaths | n/a | n/a | 0 | 415 | no | CDC WONDER, counties with published counts in every year | 2019 |
| Alcohol-induced deaths | 1.99 | 0.163 | 1.08 | 415 | no | CDC WONDER, counties with published counts in every year | 2020 |
| Alcohol-induced deaths | 2.78 | −0.172 | 1.30 | 415 | no | CDC WONDER, counties with published counts in every year | 2021 |
| Alcohol-induced deaths | 2.54 | 0.072 | 1.31 | 415 | no | CDC WONDER, counties with published counts in every year | 2022 |
| Alcohol-induced deaths | 1.44 | −0.370 | 0.535 | 415 | no | CDC WONDER, counties with published counts in every year | 2023 |
| Alcohol-induced deaths | 1.56 | −0.033 | 0.765 | 415 | no | CDC WONDER, counties with published counts in every year | 2024 |
| Suicide | 1.30 | 0.286 | 0.793 | 477 | no | CDC WONDER, counties with published counts in every year | 2018 |
| Suicide | n/a | n/a | 0 | 477 | no | CDC WONDER, counties with published counts in every year | 2019 |
| Suicide | 1.67 | 0.241 | 0.956 | 477 | no | CDC WONDER, counties with published counts in every year | 2020 |
| Suicide | 1.79 | 0.368 | 1.08 | 477 | no | CDC WONDER, counties with published counts in every year | 2021 |
| Suicide | 2.16 | 0.615 | 1.39 | 477 | no | CDC WONDER, counties with published counts in every year | 2022 |
| Suicide | 2.42 | 1.06 | 1.74 | 477 | no | CDC WONDER, counties with published counts in every year | 2023 |
| Suicide | 2.17 | 0.760 | 1.46 | 477 | no | CDC WONDER, counties with published counts in every year | 2024 |
| Overdose | n/a | n/a | 0 | 1,757 | no | CDC panel, counties with a stable rate in every year | 2019 |
| Overdose | 4.49 | −1.25 | 1.62 | 1,757 | no | CDC panel, counties with a stable rate in every year | 2020 |
| Overdose | 9.36 | −0.291 | 4.53 | 1,757 | no | CDC panel, counties with a stable rate in every year | 2021 |
| Overdose | 7.60 | −0.007 | 3.80 | 1,757 | no | CDC panel, counties with a stable rate in every year | 2022 |
| Overdose | 8.49 | −0.678 | 3.91 | 1,757 | no | CDC panel, counties with a stable rate in every year | 2023 |
| Overdose | 4.55 | −2.45 | 1.05 | 1,757 | yes | CDC panel, counties with a stable rate in every year | 2024 |
| Homicide | n/a | n/a | 0 | 1,828 | no | CDC panel, counties with a stable rate in every year | 2019 |
| Homicide | 1.86 | −0.372 | 0.744 | 1,828 | no | CDC panel, counties with a stable rate in every year | 2020 |
| Homicide | 2.66 | −0.687 | 0.987 | 1,828 | no | CDC panel, counties with a stable rate in every year | 2021 |
| Homicide | 2.27 | −0.172 | 1.05 | 1,828 | no | CDC panel, counties with a stable rate in every year | 2022 |
| Homicide | 2.33 | −0.073 | 1.13 | 1,828 | no | CDC panel, counties with a stable rate in every year | 2023 |
| Homicide | 2.13 | −0.388 | 0.869 | 1,828 | yes | CDC panel, counties with a stable rate in every year | 2024 |
| Suicide | n/a | n/a | 0 | 1,450 | no | CDC panel, counties with a stable rate in every year | 2019 |
| Suicide | 1.44 | 0.146 | 0.792 | 1,450 | no | CDC panel, counties with a stable rate in every year | 2020 |
| Suicide | 2.03 | 0.357 | 1.20 | 1,450 | no | CDC panel, counties with a stable rate in every year | 2021 |
| Suicide | 1.65 | 0.161 | 0.906 | 1,450 | no | CDC panel, counties with a stable rate in every year | 2022 |
| Suicide | 1.92 | 0.530 | 1.23 | 1,450 | no | CDC panel, counties with a stable rate in every year | 2023 |
| Suicide | 1.79 | 0.320 | 1.05 | 1,450 | yes | CDC panel, counties with a stable rate in every year | 2024 |

Views: CDC WONDER, 2018 to 2024 (source = CDC WONDER, counties with published counts in every year); CDC panel, 2019 to 2024 (source = CDC panel, counties with a stable rate in every year).

Note: The comparison keeps only the least and most deprived thirds, so the middle third is left out. A difference between two groups' changes is a difference-in-differences with a fixed characteristic and needs the groups to have moved alike without the pandemic; the 2018 coefficient is the only available check, and the suicide line fails it.

Note: The WONDER tab covers 293 counties in the most deprived third for overdose, 207 for alcohol-induced deaths and 238 for suicide, with equal numbers in the least deprived third. The panel tab has no pre-period. The panel's 2024 values are provisional (open circles); WONDER's 2024 counts are final.

Source: CDC WONDER underlying cause of death, 2018-2024, and CDC Mapping Injury, Overdose, and Violence county file, 2019-2024; crude rates, modeled estimates, population-weighted regressions

Data: https://vishalsingh.org/zipcode/report/mortality-trends-2018-2024/figures/event_study_deprivation.json

### Table 8. The modeled overdose death rate was lower in 2024 than in 2022 in 80.2% of counties with any overdose deaths in 2022 or 2024, from 62.7% in the West to 91.8% in the Northeast

Change in the CDC panel's modeled overdose death rate per 100,000 residents from 2022 to 2024 (2024 provisional), counties with a stable rate in both years and at least one death

No inline rows: the values of Table 8 are in https://vishalsingh.org/zipcode/data/mortality-trends-2018-2024/overdose_turn.json.

Note: The map covers the 2,624 of 2,927 counties with a stable rate in both years that have at least one death in 2022 or 2024. The 303 counties with no deaths in either year have no direction of change; they are gray, with the reason in the tooltip, and they are left out of every share. They are concentrated in the most deprived third (164 counties against 30 in the least deprived), so counting them as not lower would build a deprivation gradient out of counties with no deaths. 12 counties have the same positive rate in both years and count as not lower.

Note: CDC models the rate of a county with few deaths, so the change for such a county is a model's change. Among the 986 counties with direct rates in both years, a rate lower in 2024 holds for 84.1% of counties, against 77.8% of the 1,638 with a modeled rate in at least one year. 14 counties have no Area Deprivation Index value and are left out of the thirds.

Note: The panel's 2024 values are provisional and may change as investigations are completed. The 20-death floor for maps of WONDER counts does not apply, since these are modeled rates. Thirds of deprivation rank the counties with a stable rate in both years by the mean Area Deprivation Index of their tracts.

Note: Connecticut's eight counties are gray because CDC's county file does not carry them. Puerto Rico is not drawn. Alaska is drawn.

Source: CDC Mapping Injury, Overdose, and Violence county file, 2022 and 2024; modeled estimates, crude rates

Data: https://vishalsingh.org/zipcode/report/mortality-trends-2018-2024/figures/overdose_turn_map.json

## All estimates

Every fact in facts.json (181), sorted by id.

| Fact id | Value | Unit | Estimand | Population | Period | Interval | n | Weights | Source | Note |
| --- | ---: | --- | --- | --- | --- | --- | ---: | --- | --- | --- |
| adi_miss_p | 13 | counties | count of those counties with no Area Deprivation Index value, left out of the thirds | counties with a stable modeled overdose rate in every year 2019 to 2024 | 2019 to 2024 |   | 2,648 counties | none (full count) | CDC injury and violence mapping |   |
| adi_miss_t | 14 | counties | count of those counties with no Area Deprivation Index value, left out of the thirds | counties with a stable modeled overdose rate in both 2022 and 2024 | 2022 and 2024 |   | 2,927 counties | none (full count) | CDC injury and violence mapping |   |
| adi_miss_w_od | 9 | counties | count of those counties with no Area Deprivation Index value, left out of the thirds | counties whose drug overdose count is published in every year 2018 to 2024 | 2018 to 2024 |   | 889 counties | none (full count) | CDC WONDER underlying cause of death |   |
| bench_su22 | 49,476 | suicide deaths | sum of the 51 state rows of CDC WONDER; equals the count CDC reports for 2022 | residents of the United States | 2022 |   |   | none (full count) | CDC WONDER underlying cause of death |   |
| dp_od_i22 | +19.6 | index points (2019 = 100) | index of the most deprived third's pooled overdose rate (2019 = 100) minus the same index of the least deprived third; bootstrap over counties with shared draws | counties with a stable modeled overdose rate in every year, outer thirds | 2022 against 2019 | 95% CI 8.6 to 31.3 | 1,757 counties | WONDER county population of the year | CDC injury and violence mapping |   |
| dp_od_i24 | +5.5 | index points (2019 = 100) | index of the most deprived third's pooled overdose rate (2019 = 100) minus the same index of the least deprived third; bootstrap over counties with shared draws | counties with a stable modeled overdose rate in every year, outer thirds | 2024 against 2019 | 95% CI −3.9 to 15.5 | 1,757 counties | WONDER county population of the year | CDC injury and violence mapping |   |
| dw_od_i22 | +1.9 | index points (2019 = 100) | index of the most deprived third's pooled overdose rate (2019 = 100) minus the same index of the least deprived third; bootstrap over counties with shared draws | counties whose overdose count is published in every year, outer thirds | 2022 against 2019 | 95% CI −9.4 to 16.4 | 587 counties | none (crude rates from counts) | CDC WONDER underlying cause of death |   |
| dw_od_i24 | −13.4 | index points (2019 = 100) | index of the most deprived third's pooled overdose rate (2019 = 100) minus the same index of the least deprived third; bootstrap over counties with shared draws | counties whose overdose count is published in every year, outer thirds | 2024 against 2019 | 95% CI −23.4 to −1.0 | 587 counties | none (crude rates from counts) | CDC WONDER underlying cause of death |   |
| eg_p_hm_22 | 1.0 | deaths per 100,000 residents | two-way fixed-effects coefficient, homicide, change since 2019 in the most deprived third minus the same change in the least deprived third; county-population weights, errors clustered by state | balanced counties with homicide counts | 2022 against 2019 | 95% CI −0.2 to 2.3 | 1,828 counties | county population of the year | CDC injury and violence mapping |   |
| eg_p_od_22 | 3.8 | deaths per 100,000 residents | two-way fixed-effects coefficient, drug overdose, change since 2019 in the most deprived third minus the same change in the least deprived third; county-population weights, errors clustered by state | balanced counties with drug overdose counts | 2022 against 2019 | 95% CI 0.0 to 7.6 | 1,757 counties | county population of the year | CDC injury and violence mapping |   |
| eg_w_alc_18 | −0.2 | deaths per 100,000 residents | two-way fixed-effects coefficient, alcohol-induced, change since 2019 in the most deprived third minus the same change in the least deprived third; county-population weights, errors clustered by state | complete counties with alcohol-induced counts | 2018 against 2019 | 95% CI −0.9 to 0.5 | 415 counties | county population of the year | CDC WONDER underlying cause of death |   |
| eg_w_alc_22 | 1.3 | deaths per 100,000 residents | two-way fixed-effects coefficient, alcohol-induced, change since 2019 in the most deprived third minus the same change in the least deprived third; county-population weights, errors clustered by state | complete counties with alcohol-induced counts | 2022 against 2019 | 95% CI 0.1 to 2.5 | 415 counties | county population of the year | CDC WONDER underlying cause of death |   |
| eg_w_od_18 | −0.1 | deaths per 100,000 residents | two-way fixed-effects coefficient, drug overdose, change since 2019 in the most deprived third minus the same change in the least deprived third; county-population weights, errors clustered by state | complete counties with drug overdose counts | 2018 against 2019 | 95% CI −1.7 to 1.6 | 587 counties | county population of the year | CDC WONDER underlying cause of death |   |
| eg_w_od_22 | 5.1 | deaths per 100,000 residents | two-way fixed-effects coefficient, drug overdose, change since 2019 in the most deprived third minus the same change in the least deprived third; county-population weights, errors clustered by state | complete counties with drug overdose counts | 2022 against 2019 | 95% CI 1.0 to 9.2 | 587 counties | county population of the year | CDC WONDER underlying cause of death |   |
| eg_w_su_18 | 0.8 | deaths per 100,000 residents | two-way fixed-effects coefficient, suicide, change since 2019 in the most deprived third minus the same change in the least deprived third; county-population weights, errors clustered by state | complete counties with suicide counts | 2018 against 2019 | 95% CI 0.3 to 1.3 | 477 counties | county population of the year | CDC WONDER underlying cause of death |   |
| eg_w_su_20 | 1.0 | deaths per 100,000 residents | two-way fixed-effects coefficient, suicide, change since 2019 in the most deprived third minus the same change in the least deprived third; county-population weights, errors clustered by state | complete counties with suicide counts | 2020 against 2019 | 95% CI 0.2 to 1.7 | 477 counties | county population of the year | CDC WONDER underlying cause of death |   |
| es_p_hm_21 | 2.1 | deaths per 100,000 residents | two-way fixed-effects coefficient, homicide, change in the county death rate since 2019; county-population weights, errors clustered by state | balanced counties with homicide counts | 2021 against 2019 | 95% CI 1.7 to 2.5 | 2,755 counties | county population of the year | CDC injury and violence mapping |   |
| es_p_hm_24 | 0.2 | deaths per 100,000 residents | two-way fixed-effects coefficient, homicide, change in the county death rate since 2019; county-population weights, errors clustered by state | balanced counties with homicide counts | 2024 against 2019 (2024 provisional) | 95% CI 0.0 to 0.5 | 2,755 counties | county population of the year | CDC injury and violence mapping |   |
| es_p_od_22 | 11.0 | deaths per 100,000 residents | two-way fixed-effects coefficient, drug overdose, change in the county death rate since 2019; county-population weights, errors clustered by state | balanced counties with drug overdose counts | 2022 against 2019 | 95% CI 9.3 to 12.7 | 2,648 counties | county population of the year | CDC injury and violence mapping |   |
| es_p_od_24 | 2.5 | deaths per 100,000 residents | two-way fixed-effects coefficient, drug overdose, change in the county death rate since 2019; county-population weights, errors clustered by state | balanced counties with drug overdose counts | 2024 against 2019 (2024 provisional) | 95% CI −0.1 to 5.2 | 2,648 counties | county population of the year | CDC injury and violence mapping |   |
| es_se_ratio | 2.0 | ratio of standard errors | standard error of the 2022 coefficient clustered by state over the standard error clustered by county | the WONDER overdose event study | 2022 against 2019 |   |   | see the model | CDC WONDER underlying cause of death |   |
| es_states | 50 | state clusters | count of the clusters in the event study | states and DC with a complete county in the WONDER overdose event study | 2018 to 2024 |   |   | none (full count) | CDC WONDER underlying cause of death |   |
| es_w_alc_18 | −0.4 | deaths per 100,000 residents | two-way fixed-effects coefficient, alcohol-induced, change in the county death rate since 2019; county-population weights, errors clustered by state | complete counties with alcohol-induced counts | 2018 against 2019 | 95% CI −0.7 to −0.1 | 629 counties | county population of the year | CDC WONDER underlying cause of death |   |
| es_w_alc_21 | 4.3 | deaths per 100,000 residents | two-way fixed-effects coefficient, alcohol-induced, change in the county death rate since 2019; county-population weights, errors clustered by state | complete counties with alcohol-induced counts | 2021 against 2019 | 95% CI 3.5 to 5.0 | 629 counties | county population of the year | CDC WONDER underlying cause of death |   |
| es_w_alc_24 | 1.7 | deaths per 100,000 residents | two-way fixed-effects coefficient, alcohol-induced, change in the county death rate since 2019; county-population weights, errors clustered by state | complete counties with alcohol-induced counts | 2024 against 2019 | 95% CI 1.2 to 2.1 | 629 counties | county population of the year | CDC WONDER underlying cause of death |   |
| es_w_od_18 | −1.0 | deaths per 100,000 residents | two-way fixed-effects coefficient, drug overdose, change in the county death rate since 2019; county-population weights, errors clustered by state | complete counties with drug overdose counts | 2018 against 2019 | 95% CI −1.6 to −0.5 | 889 counties | county population of the year | CDC WONDER underlying cause of death |   |
| es_w_od_21 | 10.9 | deaths per 100,000 residents | two-way fixed-effects coefficient, drug overdose, change in the county death rate since 2019; county-population weights, errors clustered by state | complete counties with drug overdose counts | 2021 against 2019 | 95% CI 9.3 to 12.5 | 889 counties | county population of the year | CDC WONDER underlying cause of death |   |
| es_w_od_22 | 11.2 | deaths per 100,000 residents | two-way fixed-effects coefficient, drug overdose, change in the county death rate since 2019; county-population weights, errors clustered by state | complete counties with drug overdose counts | 2022 against 2019 | 95% CI 9.5 to 13.0 | 889 counties | county population of the year | CDC WONDER underlying cause of death |   |
| es_w_od_23 | 10.3 | deaths per 100,000 residents | two-way fixed-effects coefficient, drug overdose, change in the county death rate since 2019; county-population weights, errors clustered by state | complete counties with drug overdose counts | 2023 against 2019 | 95% CI 8.0 to 12.6 | 889 counties | county population of the year | CDC WONDER underlying cause of death |   |
| es_w_od_24 | 1.8 | deaths per 100,000 residents | two-way fixed-effects coefficient, drug overdose, change in the county death rate since 2019; county-population weights, errors clustered by state | complete counties with drug overdose counts | 2024 against 2019 | 95% CI −1.1 to 4.8 | 889 counties | county population of the year | CDC WONDER underlying cause of death |   |
| es_w_su_max | 0.6 | deaths per 100,000 residents | largest absolute event-study coefficient for suicide, change in the county death rate since 2019 | counties with suicide counts in the event study | 2018 to 2024 |   |   | county population of the year | CDC WONDER underlying cause of death |   |
| kegler_rise | 34.6% | percent rise in the firearm homicide rate | rise in the age-adjusted firearm homicide rate reported by Kegler and colleagues (2022) from NCHS mortality data | residents of the United States | 2020 against 2019 |   |   | as reported in the study | Published estimates from cited studies |   |
| nchs_hom | 1.0% | percent of drug overdose deaths | upper bound on the share of drug overdose deaths that were homicides (reported as less than 1.0 percent), NCHS Data Brief 522 (Garnett and Minino 2024) | residents of the United States | 2023 |   |   | as reported in the study | Published estimates from cited studies |   |
| nchs_suicide | 4.4% | percent of drug overdose deaths | share of drug overdose deaths that were suicides, reported in NCHS Data Brief 522 (Garnett and Minino 2024) from final 2023 death certificates | residents of the United States | 2023 |   |   | as reported in the study | Published estimates from cited studies |   |
| p_hm_n | 3,027 | counties | count of counties with a stable modeled homicide rate in 2019 | counties in CDC's county file | 2019 |   |   | none (full count) | CDC injury and violence mapping |   |
| p_hm_nb | 2,755 | counties | count of counties with a stable modeled homicide rate and a population in every year | counties in CDC's county file | 2019 to 2024 |   |   | none (full count) | CDC injury and violence mapping |   |
| p_hm_p20 | 27.9% | percent above the 2019 rate | weighted mean of county homicide rates in the year divided by the 2019 value, minus 1; bootstrap over counties with shared draws | residents of counties with a stable modeled homicide rate that year | 2020 against 2019 | 95% CI 24.9% to 30.8% | 3,067 counties | WONDER county population of the year | CDC injury and violence mapping |   |
| p_hm_p21 | 35.2% | percent above the 2019 rate | weighted mean of county homicide rates in the year divided by the 2019 value, minus 1; bootstrap over counties with shared draws | residents of counties with a stable modeled homicide rate that year | 2021 against 2019 | 95% CI 31.2% to 39.1% | 3,051 counties | WONDER county population of the year | CDC injury and violence mapping |   |
| p_hm_p24 | 3.0% | percent above the 2019 rate | weighted mean of county homicide rates in the year divided by the 2019 value, minus 1; bootstrap over counties with shared draws | residents of counties with a stable modeled homicide rate that year | 2024 against 2019 (2024 provisional) | 95% CI −0.2% to 6.1% | 3,043 counties | WONDER county population of the year | CDC injury and violence mapping |   |
| p_hm_r19 | 5.8 | deaths per 100,000 residents | mean of county homicide rates weighted by county population; bootstrap over counties | residents of counties with a stable modeled homicide rate that year | 2019 | 95% CI 5.3 to 6.3 | 3,027 counties | WONDER county population of the year | CDC injury and violence mapping |   |
| p_hm_r21 | 7.8 | deaths per 100,000 residents | mean of county homicide rates weighted by county population; bootstrap over counties | residents of counties with a stable modeled homicide rate that year | 2021 | 95% CI 7.1 to 8.6 | 3,051 counties | WONDER county population of the year | CDC injury and violence mapping |   |
| p_n_file | 3,135 | counties | count of counties in the file | counties in CDC's county file (Connecticut's eight counties are absent) | 2019 to 2024 |   |   | none (full count) | CDC injury and violence mapping |   |
| p_od_n | 2,977 | counties | count of counties with a stable modeled drug overdose rate in 2019 | counties in CDC's county file | 2019 |   |   | none (full count) | CDC injury and violence mapping |   |
| p_od_nb | 2,648 | counties | count of counties with a stable modeled drug overdose rate and a population in every year | counties in CDC's county file | 2019 to 2024 |   |   | none (full count) | CDC injury and violence mapping |   |
| p_od_p22 | 54.7% | percent above the 2019 rate | weighted mean of county drug overdose rates in the year divided by the 2019 value, minus 1; bootstrap over counties with shared draws | residents of counties with a stable modeled drug overdose rate that year | 2022 against 2019 | 95% CI 50.0% to 59.7% | 3,040 counties | WONDER county population of the year | CDC injury and violence mapping |   |
| p_od_p22_bal | 54.5% | percent above the 2019 rate | weighted mean of county overdose rates in 2022 divided by the 2019 value, minus 1, over counties with a stable rate in all six years; bootstrap over counties | residents of counties with a stable modeled overdose rate in every year | 2022 against 2019 | 95% CI 49.7% to 59.1% | 2,648 counties | WONDER county population of the year | CDC injury and violence mapping |   |
| p_od_p24 | 12.0% | percent above the 2019 rate | weighted mean of county drug overdose rates in the year divided by the 2019 value, minus 1; bootstrap over counties with shared draws | residents of counties with a stable modeled drug overdose rate that year | 2024 against 2019 (2024 provisional) | 95% CI 7.4% to 17.2% | 3,006 counties | WONDER county population of the year | CDC injury and violence mapping |   |
| p_od_r19 | 20.0 | deaths per 100,000 residents | mean of county drug overdose rates weighted by county population; bootstrap over counties | residents of counties with a stable modeled drug overdose rate that year | 2019 | 95% CI 18.8 to 21.2 | 2,977 counties | WONDER county population of the year | CDC injury and violence mapping |   |
| p_od_r22 | 30.9 | deaths per 100,000 residents | mean of county drug overdose rates weighted by county population; bootstrap over counties | residents of counties with a stable modeled drug overdose rate that year | 2022 | 95% CI 29.5 to 32.4 | 3,040 counties | WONDER county population of the year | CDC injury and violence mapping |   |
| p_su_dev | 4.0% | percent of the 2019 rate | largest absolute difference of the weighted mean county suicide rate from its 2019 value | residents of counties with a stable modeled suicide rate that year | 2019 to 2024 against 2019 (2024 provisional) |   |   | WONDER county population of the year | CDC injury and violence mapping |   |
| p_su_n | 2,880 | counties | count of counties with a stable modeled suicide rate in 2019 | counties in CDC's county file | 2019 |   |   | none (full count) | CDC injury and violence mapping |   |
| p_su_nb | 2,184 | counties | count of counties with a stable modeled suicide rate and a population in every year | counties in CDC's county file | 2019 to 2024 |   |   | none (full count) | CDC injury and violence mapping |   |
| p_su_r19 | 14.4 | deaths per 100,000 residents | mean of county suicide rates weighted by county population; bootstrap over counties | residents of counties with a stable modeled suicide rate that year | 2019 | 95% CI 13.8 to 15.0 | 2,880 counties | WONDER county population of the year | CDC injury and violence mapping |   |
| p_su_r24 | 14.5 | deaths per 100,000 residents | mean of county suicide rates weighted by county population; bootstrap over counties | residents of counties with a stable modeled suicide rate that year | 2024 (provisional) | 95% CI 13.9 to 15.1 | 2,870 counties | WONDER county population of the year | CDC injury and violence mapping |   |
| pop_effect | 1.1% | percent of the 2024 rate | how much higher the 2024 national rates would be had the population grown at the mean yearly rate of 2019 to 2023 | residents of the United States | 2024 |   |   | none (population estimate) | CDC WONDER underlying cause of death |   |
| pop_effect_pts | 1.2 | index points (2019 = 100) | the same difference for the overdose index of 2024, in index points | residents of the United States | 2024 |   |   | none (population estimate) | CDC WONDER underlying cause of death |   |
| pop_g24 | 1.6% | percent change in the national population estimate | growth of WONDER's national population estimate from 2023 to 2024 | residents of the United States | 2023 to 2024 |   |   | none (population estimate) | CDC WONDER underlying cause of death |   |
| pop_g_hi | 0.7% | percent change in the national population estimate | largest yearly growth of WONDER's national population estimate in 2019 to 2023 | residents of the United States | largest yearly change 2019 to 2023 |   |   | none (population estimate) | CDC WONDER underlying cause of death |   |
| pop_g_lo | 0.3% | percent change in the national population estimate | smallest yearly growth of WONDER's national population estimate in 2019 to 2023 | residents of the United States | smallest yearly change 2019 to 2023 |   |   | none (population estimate) | CDC WONDER underlying cause of death |   |
| px_c_d22 | +0.1 | index points (2019 = 100) | index of the most deprived third's pooled overdose rate (2019 = 100) minus the same index of the least deprived third; bootstrap over counties with shared draws | balanced panel counties that are complete in WONDER's overdose cut, with WONDER's thirds | 2022 against 2019 | 95% CI −12.1 to 16.1 | 587 counties | WONDER county population of the year | CDC injury and violence mapping |   |
| px_c_d24 | −16.2 | index points (2019 = 100) | index of the most deprived third's pooled overdose rate (2019 = 100) minus the same index of the least deprived third; bootstrap over counties with shared draws | balanced panel counties that are complete in WONDER's overdose cut, with WONDER's thirds | 2024 against 2019 | 95% CI −27.2 to −2.6 | 587 counties | WONDER county population of the year | CDC injury and violence mapping |   |
| px_c_i1_22 | 52.3% | percent above the 2019 rate | pooled modeled overdose rate of the third in the year over its 2019 rate, minus 1; bootstrap over counties with shared draws | residents of the least deprived third of the balanced panel counties that are complete in WONDER's overdose cut, with WONDER's thirds | 2022 against 2019 | 95% CI 43.9% to 60.5% | 294 counties | WONDER county population of the year | CDC injury and violence mapping |   |
| px_c_i3_22 | 52.3% | percent above the 2019 rate | pooled modeled overdose rate of the third in the year over its 2019 rate, minus 1; bootstrap over counties with shared draws | residents of the most deprived third of the balanced panel counties that are complete in WONDER's overdose cut, with WONDER's thirds | 2022 against 2019 | 95% CI 43.0% to 64.8% | 293 counties | WONDER county population of the year | CDC injury and violence mapping |   |
| px_c_n | 880 | counties | count of counties | balanced panel counties that are complete in WONDER's overdose cut, with WONDER's thirds | 2019 to 2024 |   |   | none (full count) | CDC injury and violence mapping |   |
| px_c_ratio19 | 1.51 | ratio of rates | ratio of the pooled modeled overdose rates of the two thirds; bootstrap over counties | the most deprived third relative to the least deprived third of the balanced panel counties that are complete in WONDER's overdose cut, with WONDER's thirds | 2019 | 95% CI 1.28 to 1.75 | 587 counties | WONDER county population of the year | CDC injury and violence mapping |   |
| px_o_d22 | +34.7 | index points (2019 = 100) | index of the most deprived third's pooled overdose rate (2019 = 100) minus the same index of the least deprived third; bootstrap over counties with shared draws | balanced panel counties that are not complete in WONDER's overdose cut, with thirds cut among them | 2022 against 2019 | 95% CI 21.1 to 47.3 | 1,170 counties | WONDER county population of the year | CDC injury and violence mapping |   |
| px_o_d24 | +25.2 | index points (2019 = 100) | index of the most deprived third's pooled overdose rate (2019 = 100) minus the same index of the least deprived third; bootstrap over counties with shared draws | balanced panel counties that are not complete in WONDER's overdose cut, with thirds cut among them | 2024 against 2019 | 95% CI 14.3 to 36.0 | 1,170 counties | WONDER county population of the year | CDC injury and violence mapping |   |
| px_o_i1_22 | 60.2% | percent above the 2019 rate | pooled modeled overdose rate of the third in the year over its 2019 rate, minus 1; bootstrap over counties with shared draws | residents of the least deprived third of the balanced panel counties that are not complete in WONDER's overdose cut, with thirds cut among them | 2022 against 2019 | 95% CI 52.4% to 68.3% | 585 counties | WONDER county population of the year | CDC injury and violence mapping |   |
| px_o_i3_22 | 94.9% | percent above the 2019 rate | pooled modeled overdose rate of the third in the year over its 2019 rate, minus 1; bootstrap over counties with shared draws | residents of the most deprived third of the balanced panel counties that are not complete in WONDER's overdose cut, with thirds cut among them | 2022 against 2019 | 95% CI 85.1% to 105.0% | 585 counties | WONDER county population of the year | CDC injury and violence mapping |   |
| px_o_n | 1,759 | counties | count of counties | balanced panel counties that are not complete in WONDER's overdose cut, with thirds cut among them | 2019 to 2024 |   |   | none (full count) | CDC injury and violence mapping |   |
| rate_gamma_max | 100 | deaths | definition: counts below this number get gamma limits, counts at or above it a normal interval | national and state death counts in CDC WONDER | 2018 to 2024 |   |   | none (definition) | CDC WONDER underlying cause of death |   |
| rec_add_p22 | 51.2% | percent above the 2019 rate | the panel's implied deaths plus the 2019 gap in both years, over WONDER's population outside Connecticut: 2022 rate over the 2019 rate, minus 1 | residents of the 50 states and DC outside Connecticut | 2022 against 2019 |   |   | see the two series | CDC injury and violence mapping |   |
| rec_cover | 99.5% | percent of residents | share of residents living in counties with a stable modeled overdose rate in 2019 | residents of the 50 states and DC outside Connecticut | 2019 |   |   | WONDER county population | CDC injury and violence mapping |   |
| rec_ct19 | 1,214 | drug overdose deaths | WONDER drug poisoning deaths of any intent in Connecticut | Connecticut residents | 2019 |   |   | none (full count) | CDC WONDER underlying cause of death |   |
| rec_gap19 | 4,905 | drug overdose deaths | WONDER count outside Connecticut minus the panel's implied deaths | residents of the 50 states and DC outside Connecticut | 2019 |   |   | see the two series | CDC WONDER underlying cause of death |   |
| rec_gap22 | 4,817 | drug overdose deaths | WONDER count outside Connecticut minus the panel's implied deaths | residents of the 50 states and DC outside Connecticut | 2022 |   |   | see the two series | CDC WONDER underlying cause of death |   |
| rec_gap23_share | 4.4% | percent of WONDER's count | WONDER count outside Connecticut minus the panel's implied deaths, over WONDER's count | residents of the 50 states and DC outside Connecticut | 2023 |   |   | see the two series | CDC WONDER underlying cause of death |   |
| rec_gapsh19 | 7.1% | percent of WONDER's count | WONDER count outside Connecticut minus the panel's implied deaths, over WONDER's count | residents of the 50 states and DC outside Connecticut | 2019 |   |   | see the two series | CDC WONDER underlying cause of death |   |
| rec_gapsh22 | 4.5% | percent of WONDER's count | WONDER count outside Connecticut minus the panel's implied deaths, over WONDER's count | residents of the 50 states and DC outside Connecticut | 2022 |   |   | see the two series | CDC WONDER underlying cause of death |   |
| rec_inc_p | 37,132 | drug overdose deaths added | the panel's implied deaths in 2022 minus those in 2019 | residents of counties with a stable modeled overdose rate that year | 2022 against 2019 |   |   | WONDER county population of the year | CDC injury and violence mapping |   |
| rec_inc_w | 37,043 | drug overdose deaths added | WONDER count outside Connecticut in 2022 minus the count in 2019 | residents of the 50 states and DC outside Connecticut | 2022 against 2019 |   |   | none (full counts) | CDC WONDER underlying cause of death |   |
| rec_p19 | 64,511 | drug overdose deaths implied | sum over counties of the modeled overdose rate times WONDER's county population, over 100,000 | residents of counties with a stable modeled overdose rate that year | 2019 |   |   | WONDER county population of the year | CDC injury and violence mapping |   |
| rec_p22 | 101,642 | drug overdose deaths implied | sum over counties of the modeled overdose rate times WONDER's county population, over 100,000 | residents of counties with a stable modeled overdose rate that year | 2022 |   |   | WONDER county population of the year | CDC injury and violence mapping |   |
| rec_pts | 4.2 | percentage points | the panel's percent rise in the overdose rate minus WONDER's, 2019 to 2022 | residents of the United States | 2022 against 2019 |   |   | see the two series | CDC injury and violence mapping |   |
| rec_pts_xct | 3.7 | percentage points | the panel's percent rise in the overdose rate minus WONDER's outside Connecticut, 2019 to 2022 | residents of the 50 states and DC outside Connecticut | 2022 against 2019 |   |   | see the two series | CDC injury and violence mapping |   |
| rec_w19 | 69,416 | drug overdose deaths | WONDER drug poisoning deaths of any intent, the 50 state rows other than Connecticut summed | residents of the 50 states and DC outside Connecticut | 2019 |   |   | none (full count) | CDC WONDER underlying cause of death |   |
| rec_w22 | 106,459 | drug overdose deaths | WONDER drug poisoning deaths of any intent, the 50 state rows other than Connecticut summed | residents of the 50 states and DC outside Connecticut | 2022 |   |   | none (full count) | CDC WONDER underlying cause of death |   |
| tn_all | 3,135 | counties | count of counties in the file | counties in CDC's county file | 2022 and 2024 |   |   | none (full count) | CDC injury and violence mapping |   |
| tn_direct | 84.1% | percent of counties | share of counties whose rate was lower in 2024 than in 2022; bootstrap over counties | counties whose overdose rate is direct (from 10 or more deaths) in both 2022 and 2024, with at least one overdose death | 2022 to 2024 | 95% CI 81.7% to 86.3% | 986 counties | none (each county counts once) | CDC injury and violence mapping |   |
| tn_direct_n | 986 | counties | count of counties | counties whose overdose rate is direct (from 10 or more deaths) in both 2022 and 2024, with at least one overdose death | 2022 and 2024 |   | 2,624 counties | none (full count) | CDC injury and violence mapping |   |
| tn_fell | 80.2% | percent of counties | share of counties whose modeled overdose rate was lower in 2024 than in 2022; bootstrap over counties | counties with a stable modeled overdose rate in both 2022 and 2024 and at least one overdose death in one of the two years | 2022 to 2024 | 95% CI 78.7% to 81.7% | 2,624 counties | none (each county counts once) | CDC injury and violence mapping |   |
| tn_fell_n | 2,104 | counties | count of counties whose rate was lower in 2024 than in 2022 | counties with a stable modeled overdose rate in both 2022 and 2024 and at least one overdose death in one of the two years | 2022 to 2024 |   | 2,624 counties | none (full count) | CDC injury and violence mapping |   |
| tn_modeled | 77.8% | percent of counties | share of counties whose rate was lower in 2024 than in 2022; bootstrap over counties | counties whose overdose rate is modeled in at least one of 2022 and 2024, with at least one overdose death | 2022 to 2024 | 95% CI 75.7% to 79.8% | 1,638 counties | none (each county counts once) | CDC injury and violence mapping |   |
| tn_modeled_n | 1,638 | counties | count of counties | counties whose overdose rate is modeled in at least one of 2022 and 2024, with at least one overdose death | 2022 and 2024 |   | 2,624 counties | none (full count) | CDC injury and violence mapping |   |
| tn_n | 2,624 | counties | count of counties | counties with a stable modeled overdose rate in both 2022 and 2024 and at least one overdose death in one of the two years | 2022 and 2024 |   |   | none (full count) | CDC injury and violence mapping |   |
| tn_ne | 91.8% | percent of counties | share of counties whose modeled overdose rate was lower in 2024 than in 2022; bootstrap over counties | counties in the Northeast Census region with a stable modeled overdose rate in both 2022 and 2024 and at least one overdose death in one of the two years | 2022 to 2024 | 95% CI 87.9% to 95.2% | 207 counties | none (each county counts once) | CDC injury and violence mapping |   |
| tn_pooled | −8.6 | deaths per 100,000 residents | weighted mean of county overdose rates in 2024 minus the same in 2022; bootstrap over counties | residents of counties with a stable modeled overdose rate in both 2022 and 2024 and at least one overdose death in one of the two years | 2024 minus 2022 (2024 provisional) | 95% CI −9.4 to −7.8 | 2,624 counties | WONDER county population of the year | CDC injury and violence mapping |   |
| tn_popfell | 86.0% | percent of residents | share of residents living in counties whose modeled overdose rate was lower in 2024 than in 2022; bootstrap over counties | residents of counties with a stable modeled overdose rate in both 2022 and 2024 and at least one overdose death in one of the two years | 2022 population | 95% CI 82.5% to 89.1% | 2,624 counties | WONDER county population | CDC injury and violence mapping |   |
| tn_same | 12 | counties | count of counties whose rate was the same positive number in both years; counted as not lower | counties with a stable modeled overdose rate in both 2022 and 2024 and at least one overdose death in one of the two years | 2022 to 2024 |   | 2,624 counties | none (full count) | CDC injury and violence mapping |   |
| tn_stable | 2,927 | counties | count of counties with a stable modeled overdose rate in both years | counties in CDC's county file | 2022 and 2024 |   |   | none (full count) | CDC injury and violence mapping |   |
| tn_t1 | 80.0% | percent of counties | share of counties whose modeled overdose rate was lower in 2024 than in 2022; bootstrap over counties | counties in the least deprived third of those with a stable modeled overdose rate in both 2022 and 2024 and at least one overdose death in one of the two years | 2022 to 2024 | 95% CI 77.4% to 82.5% | 941 counties | none (each county counts once) | CDC injury and violence mapping |   |
| tn_t2 | 82.6% | percent of counties | share of counties whose modeled overdose rate was lower in 2024 than in 2022; bootstrap over counties | counties in the middle third of those with a stable modeled overdose rate in both 2022 and 2024 and at least one overdose death in one of the two years | 2022 to 2024 | 95% CI 80.0% to 85.0% | 874 counties | none (each county counts once) | CDC injury and violence mapping |   |
| tn_t3 | 77.8% | percent of counties | share of counties whose modeled overdose rate was lower in 2024 than in 2022; bootstrap over counties | counties in the most deprived third of those with a stable modeled overdose rate in both 2022 and 2024 and at least one overdose death in one of the two years | 2022 to 2024 | 95% CI 75.0% to 80.7% | 807 counties | none (each county counts once) | CDC injury and violence mapping |   |
| tn_we | 62.7% | percent of counties | share of counties whose modeled overdose rate was lower in 2024 than in 2022; bootstrap over counties | counties in the West Census region with a stable modeled overdose rate in both 2022 and 2024 and at least one overdose death in one of the two years | 2022 to 2024 | 95% CI 57.8% to 67.7% | 365 counties | none (each county counts once) | CDC injury and violence mapping |   |
| tn_zero_rate | 303 | counties | count of counties with a rate of zero in both years (no overdose deaths in either), left out of every share because their change has no direction | counties with a stable modeled overdose rate in both 2022 and 2024 | 2022 and 2024 |   | 2,927 counties | none (full count) | CDC injury and violence mapping |   |
| tn_zero_t1 | 30 | counties | count of counties with a rate of zero in both years | counties in the least deprived third of those with a stable modeled overdose rate in both years | 2022 and 2024 |   |   | none (full count) | CDC injury and violence mapping |   |
| tn_zero_t3 | 164 | counties | count of counties with a rate of zero in both years | counties in the most deprived third of those with a stable modeled overdose rate in both years | 2022 and 2024 |   |   | none (full count) | CDC injury and violence mapping |   |
| tp_hm_i1_21 | 36.2% | percent above the 2019 rate | pooled homicide rate of the least deprived third of counties in the year over its 2019 rate, minus 1; bootstrap over counties with shared draws | residents of the least deprived third of counties with a stable modeled rate in every year (homicide) | 2021 against 2019 | 95% CI 30.9% to 41.3% | 914 counties | WONDER county population of the year | CDC injury and violence mapping |   |
| tp_hm_i1_24 | 2.7% | percent above the 2019 rate | pooled homicide rate of the least deprived third of counties in the year over its 2019 rate, minus 1; bootstrap over counties with shared draws | residents of the least deprived third of counties with a stable modeled rate in every year (homicide) | 2024 against 2019 (2024 provisional) | 95% CI −1.2% to 6.6% | 914 counties | WONDER county population of the year | CDC injury and violence mapping |   |
| tp_hm_i3_21 | 40.8% | percent above the 2019 rate | pooled homicide rate of the most deprived third of counties in the year over its 2019 rate, minus 1; bootstrap over counties with shared draws | residents of the most deprived third of counties with a stable modeled rate in every year (homicide) | 2021 against 2019 | 95% CI 32.3% to 49.8% | 914 counties | WONDER county population of the year | CDC injury and violence mapping |   |
| tp_hm_i3_24 | 13.2% | percent above the 2019 rate | pooled homicide rate of the most deprived third of counties in the year over its 2019 rate, minus 1; bootstrap over counties with shared draws | residents of the most deprived third of counties with a stable modeled rate in every year (homicide) | 2024 against 2019 (2024 provisional) | 95% CI 5.9% to 20.8% | 914 counties | WONDER county population of the year | CDC injury and violence mapping |   |
| tp_hm_ratio19 | 1.34 | ratio of rates | ratio of the pooled homicide rates of the two thirds; bootstrap over counties | the most deprived third of counties with a stable modeled rate in every year relative to the least deprived third (homicide) | 2019 | 95% CI 1.13 to 1.60 | 1,828 counties | WONDER county population of the year | CDC injury and violence mapping |   |
| tp_hm_ratio24 | 1.48 | ratio of rates | ratio of the pooled homicide rates of the two thirds; bootstrap over counties | the most deprived third of counties with a stable modeled rate in every year relative to the least deprived third (homicide) | 2024 (provisional) | 95% CI 1.23 to 1.80 | 1,828 counties | WONDER county population of the year | CDC injury and violence mapping |   |
| tp_od_i1_22 | 51.4% | percent above the 2019 rate | pooled drug overdose rate of the least deprived third of counties in the year over its 2019 rate, minus 1; bootstrap over counties with shared draws | residents of the least deprived third of counties with a stable modeled rate in every year (drug overdose) | 2022 against 2019 | 95% CI 45.2% to 57.6% | 879 counties | WONDER county population of the year | CDC injury and violence mapping |   |
| tp_od_i1_24 | 13.2% | percent above the 2019 rate | pooled drug overdose rate of the least deprived third of counties in the year over its 2019 rate, minus 1; bootstrap over counties with shared draws | residents of the least deprived third of counties with a stable modeled rate in every year (drug overdose) | 2024 against 2019 (2024 provisional) | 95% CI 6.6% to 20.0% | 879 counties | WONDER county population of the year | CDC injury and violence mapping |   |
| tp_od_i3_22 | 71.0% | percent above the 2019 rate | pooled drug overdose rate of the most deprived third of counties in the year over its 2019 rate, minus 1; bootstrap over counties with shared draws | residents of the most deprived third of counties with a stable modeled rate in every year (drug overdose) | 2022 against 2019 | 95% CI 62.3% to 80.7% | 878 counties | WONDER county population of the year | CDC injury and violence mapping |   |
| tp_od_i3_24 | 18.7% | percent above the 2019 rate | pooled drug overdose rate of the most deprived third of counties in the year over its 2019 rate, minus 1; bootstrap over counties with shared draws | residents of the most deprived third of counties with a stable modeled rate in every year (drug overdose) | 2024 against 2019 (2024 provisional) | 95% CI 11.8% to 26.3% | 878 counties | WONDER county population of the year | CDC injury and violence mapping |   |
| tp_od_r1_19 | 19.0 | deaths per 100,000 residents | pooled drug overdose rate of the least deprived third of counties; bootstrap over counties | residents of the least deprived third of counties with a stable modeled rate in every year (drug overdose) | 2019 | 95% CI 17.7 to 20.3 | 879 counties | WONDER county population of the year | CDC injury and violence mapping |   |
| tp_od_r3_19 | 19.2 | deaths per 100,000 residents | pooled drug overdose rate of the most deprived third of counties; bootstrap over counties | residents of the most deprived third of counties with a stable modeled rate in every year (drug overdose) | 2019 | 95% CI 17.1 to 21.3 | 878 counties | WONDER county population of the year | CDC injury and violence mapping |   |
| tp_od_ratio19 | 1.01 | ratio of rates | ratio of the pooled drug overdose rates of the two thirds; bootstrap over counties | the most deprived third of counties with a stable modeled rate in every year relative to the least deprived third (drug overdose) | 2019 | 95% CI 0.88 to 1.15 | 1,757 counties | WONDER county population of the year | CDC injury and violence mapping |   |
| tp_od_ratio22 | 1.14 | ratio of rates | ratio of the pooled drug overdose rates of the two thirds; bootstrap over counties | the most deprived third of counties with a stable modeled rate in every year relative to the least deprived third (drug overdose) | 2022 | 95% CI 1.01 to 1.27 | 1,757 counties | WONDER county population of the year | CDC injury and violence mapping |   |
| tw_alc_i1_21 | 33.9% | percent above the 2019 rate | pooled alcohol-induced rate of the least deprived third of counties in the year over its 2019 rate, minus 1; bootstrap over counties with shared draws | residents of the least deprived third of counties whose count is published in every year (alcohol-induced) | 2021 against 2019 | 95% CI 30.0% to 37.7% | 208 counties | none (crude rates from counts) | CDC WONDER underlying cause of death |   |
| tw_alc_i3_21 | 38.5% | percent above the 2019 rate | pooled alcohol-induced rate of the most deprived third of counties in the year over its 2019 rate, minus 1; bootstrap over counties with shared draws | residents of the most deprived third of counties whose count is published in every year (alcohol-induced) | 2021 against 2019 | 95% CI 33.4% to 44.1% | 207 counties | none (crude rates from counts) | CDC WONDER underlying cause of death |   |
| tw_alc_n | 207 | counties | count of counties in the third | counties in the most deprived third (alcohol-induced) | 2018 to 2024 |   |   | none (full count) | CDC WONDER underlying cause of death |   |
| tw_alc_ratio22 | 1.22 | ratio of rates | ratio of the pooled alcohol-induced rates of the two thirds; bootstrap over counties | the most deprived third of counties whose count is published in every year relative to the least deprived third (alcohol-induced) | 2022 | 95% CI 1.10 to 1.37 | 415 counties | none (crude rates from counts) | CDC WONDER underlying cause of death |   |
| tw_od_i1_22 | 48.4% | percent above the 2019 rate | pooled drug overdose rate of the least deprived third of counties in the year over its 2019 rate, minus 1; bootstrap over counties with shared draws | residents of the least deprived third of counties whose count is published in every year (drug overdose) | 2022 against 2019 | 95% CI 40.7% to 56.0% | 294 counties | none (crude rates from counts) | CDC WONDER underlying cause of death |   |
| tw_od_i1_24 | 13.4% | percent above the 2019 rate | pooled drug overdose rate of the least deprived third of counties in the year over its 2019 rate, minus 1; bootstrap over counties with shared draws | residents of the least deprived third of counties whose count is published in every year (drug overdose) | 2024 against 2019 | 95% CI 5.1% to 21.5% | 294 counties | none (crude rates from counts) | CDC WONDER underlying cause of death |   |
| tw_od_i3_22 | 50.3% | percent above the 2019 rate | pooled drug overdose rate of the most deprived third of counties in the year over its 2019 rate, minus 1; bootstrap over counties with shared draws | residents of the most deprived third of counties whose count is published in every year (drug overdose) | 2022 against 2019 | 95% CI 41.8% to 62.1% | 293 counties | none (crude rates from counts) | CDC WONDER underlying cause of death |   |
| tw_od_n | 293 | counties | count of counties in the third | counties in the most deprived third (drug overdose) | 2018 to 2024 |   |   | none (full count) | CDC WONDER underlying cause of death |   |
| tw_od_r1_19 | 19.6 | deaths per 100,000 residents | pooled drug overdose rate of the least deprived third of counties; bootstrap over counties | residents of the least deprived third of counties whose count is published in every year (drug overdose) | 2019 | 95% CI 18.1 to 21.2 | 294 counties | none (crude rates from counts) | CDC WONDER underlying cause of death |   |
| tw_od_r3_19 | 28.9 | deaths per 100,000 residents | pooled drug overdose rate of the most deprived third of counties; bootstrap over counties | residents of the most deprived third of counties whose count is published in every year (drug overdose) | 2019 | 95% CI 24.9 to 32.3 | 293 counties | none (crude rates from counts) | CDC WONDER underlying cause of death |   |
| tw_od_r3_24 | 28.9 | deaths per 100,000 residents | pooled drug overdose rate of the most deprived third of counties; bootstrap over counties | residents of the most deprived third of counties whose count is published in every year (drug overdose) | 2024 | 95% CI 26.3 to 31.2 | 293 counties | none (crude rates from counts) | CDC WONDER underlying cause of death |   |
| tw_od_ratio19 | 1.48 | ratio of rates | ratio of the pooled drug overdose rates of the two thirds; bootstrap over counties | the most deprived third of counties whose count is published in every year relative to the least deprived third (drug overdose) | 2019 | 95% CI 1.26 to 1.69 | 587 counties | none (crude rates from counts) | CDC WONDER underlying cause of death |   |
| tw_od_ratio22 | 1.49 | ratio of rates | ratio of the pooled drug overdose rates of the two thirds; bootstrap over counties | the most deprived third of counties whose count is published in every year relative to the least deprived third (drug overdose) | 2022 | 95% CI 1.33 to 1.66 | 587 counties | none (crude rates from counts) | CDC WONDER underlying cause of death |   |
| tw_su_n | 238 | counties | count of counties in the third | counties in the most deprived third (suicide) | 2018 to 2024 |   |   | none (full count) | CDC WONDER underlying cause of death |   |
| tw_su_ratio19 | 1.26 | ratio of rates | ratio of the pooled suicide rates of the two thirds; bootstrap over counties | the most deprived third of counties whose count is published in every year relative to the least deprived third (suicide) | 2019 | 95% CI 1.15 to 1.38 | 477 counties | none (crude rates from counts) | CDC WONDER underlying cause of death |   |
| tw_su_ratio22 | 1.37 | ratio of rates | ratio of the pooled suicide rates of the two thirds; bootstrap over counties | the most deprived third of counties whose count is published in every year relative to the least deprived third (suicide) | 2022 | 95% CI 1.25 to 1.49 | 477 counties | none (crude rates from counts) | CDC WONDER underlying cause of death |   |
| w_alc_b18 | 4.1% | percent below the 2019 rate | 2019 national crude alcohol-induced death rate minus the 2018 rate, over the 2019 rate | residents of the United States | 2018 against 2019 | 95% CI 2.7% to 5.4% | 37,329 deaths | none (crude rate from counts) | CDC WONDER underlying cause of death |   |
| w_alc_d19 | 39,043 | alcohol-induced deaths | national count of alcohol-induced deaths: the sum of the 51 state rows | residents of the United States | 2019 |   |   | none (full count) | CDC WONDER underlying cause of death |   |
| w_alc_d20 | 49,061 | alcohol-induced deaths | national count of alcohol-induced deaths: the sum of the 51 state rows | residents of the United States | 2020 |   |   | none (full count) | CDC WONDER underlying cause of death |   |
| w_alc_dp20 | 25.7% | percent rise in the number of deaths | national count of alcohol-induced deaths in 2020 over the 2019 count, minus 1; normal interval on the log ratio | residents of the United States | 2020 against 2019 | 95% CI 24.0% to 27.3% | 49,061 deaths | none (full counts) | CDC WONDER underlying cause of death |   |
| w_alc_p21 | 37.4% | percent above the 2019 rate | national crude alcohol-induced death rate divided by the 2019 rate, minus 1; normal interval on the log ratio of the two death counts | residents of the United States | 2021 against 2019 | 95% CI 35.7% to 39.2% | 54,258 deaths | none (crude rate from counts) | CDC WONDER underlying cause of death |   |
| w_alc_p24 | 15.5% | percent above the 2019 rate | national crude alcohol-induced death rate divided by the 2019 rate, minus 1; normal interval on the log ratio of the two death counts | residents of the United States | 2024 against 2019 | 95% CI 13.9% to 17.0% | 46,714 deaths | none (crude rate from counts) | CDC WONDER underlying cause of death |   |
| w_alc_r19 | 11.9 | deaths per 100,000 residents | national crude alcohol-induced death rate: the sum of the 51 state counts over the sum of state populations; interval from the count | residents of the United States | 2019 | 95% CI 11.8 to 12.0 | 39,043 deaths | none (crude rate from counts) | CDC WONDER underlying cause of death |   |
| w_alc_r20 | 14.9 | deaths per 100,000 residents | national crude alcohol-induced death rate: the sum of the 51 state counts over the sum of state populations; interval from the count | residents of the United States | 2020 | 95% CI 14.8 to 15.0 | 49,061 deaths | none (crude rate from counts) | CDC WONDER underlying cause of death |   |
| w_alc_r21 | 16.3 | deaths per 100,000 residents | national crude alcohol-induced death rate: the sum of the 51 state counts over the sum of state populations; interval from the count | residents of the United States | 2021 | 95% CI 16.2 to 16.5 | 54,258 deaths | none (crude rate from counts) | CDC WONDER underlying cause of death |   |
| w_alc_r24 | 13.7 | deaths per 100,000 residents | national crude alcohol-induced death rate: the sum of the 51 state counts over the sum of state populations; interval from the count | residents of the United States | 2024 | 95% CI 13.6 to 13.9 | 46,714 deaths | none (crude rate from counts) | CDC WONDER underlying cause of death |   |
| w_cell_max | 9 | deaths in a county-year | definition: the largest county-year count that WONDER withholds | county-years in CDC WONDER | deaths in 2018 to 2024 |   |   | none (definition) | CDC WONDER underlying cause of death |   |
| w_floor | 10 | deaths in a county-year | definition: WONDER withholds a county-year count above zero and below this number | county-years in CDC WONDER | deaths in 2018 to 2024 |   |   | none (definition) | CDC WONDER underlying cause of death |   |
| w_od_b18 | 4.3% | percent below the 2019 rate | 2019 national crude drug overdose death rate minus the 2018 rate, over the 2019 rate | residents of the United States | 2018 against 2019 | 95% CI 3.3% to 5.3% | 67,367 deaths | none (crude rate from counts) | CDC WONDER underlying cause of death |   |
| w_od_p22 | 50.5% | percent above the 2019 rate | national crude drug overdose death rate divided by the 2019 rate, minus 1; normal interval on the log ratio of the two death counts | residents of the United States | 2022 against 2019 | 95% CI 49.1% to 51.9% | 107,941 deaths | none (crude rate from counts) | CDC WONDER underlying cause of death |   |
| w_od_p24 | 8.5% | percent above the 2019 rate | national crude drug overdose death rate divided by the 2019 rate, minus 1; normal interval on the log ratio of the two death counts | residents of the United States | 2024 against 2019 | 95% CI 7.4% to 9.6% | 79,384 deaths | none (crude rate from counts) | CDC WONDER underlying cause of death |   |
| w_od_r19 | 21.5 | deaths per 100,000 residents | national crude drug overdose death rate: the sum of the 51 state counts over the sum of state populations; interval from the count | residents of the United States | 2019 | 95% CI 21.4 to 21.7 | 70,630 deaths | none (crude rate from counts) | CDC WONDER underlying cause of death |   |
| w_od_r21 | 32.1 | deaths per 100,000 residents | national crude drug overdose death rate: the sum of the 51 state counts over the sum of state populations; interval from the count | residents of the United States | 2021 | 95% CI 32.0 to 32.3 | 106,699 deaths | none (crude rate from counts) | CDC WONDER underlying cause of death |   |
| w_od_r22 | 32.4 | deaths per 100,000 residents | national crude drug overdose death rate: the sum of the 51 state counts over the sum of state populations; interval from the count | residents of the United States | 2022 | 95% CI 32.2 to 32.6 | 107,941 deaths | none (crude rate from counts) | CDC WONDER underlying cause of death |   |
| w_od_r23 | 31.4 | deaths per 100,000 residents | national crude drug overdose death rate: the sum of the 51 state counts over the sum of state populations; interval from the count | residents of the United States | 2023 | 95% CI 31.2 to 31.5 | 105,007 deaths | none (crude rate from counts) | CDC WONDER underlying cause of death |   |
| w_od_r24 | 23.3 | deaths per 100,000 residents | national crude drug overdose death rate: the sum of the 51 state counts over the sum of state populations; interval from the count | residents of the United States | 2024 | 95% CI 23.2 to 23.5 | 79,384 deaths | none (crude rate from counts) | CDC WONDER underlying cause of death |   |
| w_su_b20 | 3.6% | percent below the 2019 rate | 2019 national crude suicide rate minus the 2020 rate, over the 2019 rate | residents of the United States | 2020 against 2019 | 95% CI 2.3% to 4.8% | 45,979 deaths | none (crude rate from counts) | CDC WONDER underlying cause of death |   |
| w_su_dev | 3.6% | percent of the 2019 rate | largest absolute difference of the national crude suicide rate from its 2019 rate | residents of the United States | 2018 to 2024 against 2019 |   |   | none (crude rate from counts) | CDC WONDER underlying cause of death |   |
| w_su_r19 | 14.5 | deaths per 100,000 residents | national crude suicide death rate: the sum of the 51 state counts over the sum of state populations; interval from the count | residents of the United States | 2019 | 95% CI 14.3 to 14.6 | 47,511 deaths | none (crude rate from counts) | CDC WONDER underlying cause of death |   |
| w_su_r20 | 14.0 | deaths per 100,000 residents | national crude suicide death rate: the sum of the 51 state counts over the sum of state populations; interval from the count | residents of the United States | 2020 | 95% CI 13.8 to 14.1 | 45,979 deaths | none (crude rate from counts) | CDC WONDER underlying cause of death |   |
| w_su_r22 | 14.8 | deaths per 100,000 residents | national crude suicide death rate: the sum of the 51 state counts over the sum of state populations; interval from the count | residents of the United States | 2022 | 95% CI 14.7 to 15.0 | 49,476 deaths | none (crude rate from counts) | CDC WONDER underlying cause of death |   |
| wc_alc_n | 629 | counties | count of counties whose alcohol-induced count is published in every year and that have a population in every year | counties and county equivalents in CDC WONDER | 2018 to 2024 |   |   | none (full count) | CDC WONDER underlying cause of death |   |
| wc_od_n | 889 | counties | count of counties whose drug overdose count is published in every year and that have a population in every year | counties and county equivalents in CDC WONDER | 2018 to 2024 |   |   | none (full count) | CDC WONDER underlying cause of death |   |
| wc_od_p22 | 48.9% | percent above the 2019 rate | pooled crude drug overdose rate of the complete counties in the year divided by the 2019 rate, minus 1; bootstrap over counties | residents of counties whose drug overdose count is published in every year | 2022 against 2019 | 95% CI 44.0% to 53.9% | 889 counties | none (crude rates from counts) | CDC WONDER underlying cause of death |   |
| wc_su_n | 718 | counties | count of counties whose suicide count is published in every year and that have a population in every year | counties and county equivalents in CDC WONDER | 2018 to 2024 |   |   | none (full count) | CDC WONDER underlying cause of death |   |
| white_rise | 25.5% | percent rise in the number of deaths | rise in alcohol-related deaths (an alcohol-induced cause listed as the underlying or a contributing cause, ages 16 and over) reported by White, Castle, Powell, Hingson and Koob (2022) from NCHS mortality data | residents of the United States aged 16 and over | 2020 against 2019 |   |   | as reported in the study | Published estimates from cited studies |   |
| wx_od_p22 | 51.0% | percent above the 2019 rate | crude drug overdose death rate outside Connecticut in 2022 divided by the 2019 rate, minus 1; normal interval on the log ratio of the two death counts | residents of the 50 states and DC outside Connecticut | 2022 against 2019 | 95% CI 49.6% to 52.5% | 106,459 deaths | none (crude rate from counts) | CDC WONDER underlying cause of death |   |
| wz_alc_n | 39 | counties | count of those counties with zero deaths in all seven years | counties whose alcohol-induced count is published in every year 2018 to 2024 | 2018 to 2024 |   | 629 counties | none (full count) | CDC WONDER underlying cause of death |   |
| wz_od_n | 99 | counties | count of those counties with zero deaths in all seven years | counties whose drug overdose count is published in every year 2018 to 2024 | 2018 to 2024 |   | 889 counties | none (full count) | CDC WONDER underlying cause of death |   |
| wz_od_t3 | 76 | counties | count of those counties in the most deprived third of the complete overdose counties | complete overdose counties with zero deaths in all seven years | 2018 to 2024 |   |   | none (full count) | CDC WONDER underlying cause of death |   |
| wz_su_n | 24 | counties | count of those counties with zero deaths in all seven years | counties whose suicide count is published in every year 2018 to 2024 | 2018 to 2024 |   | 718 counties | none (full count) | CDC WONDER underlying cause of death |   |
| wzr_od_i1_22 | 49.0% | percent above the 2019 rate | pooled crude overdose rate of the third in 2022 over its 2019 rate, minus 1; bootstrap over counties with shared draws | residents of the least deprived third of complete overdose counties with at least one death, thirds cut again among them | 2022 against 2019 | 95% CI 41.1% to 56.7% | 264 counties | none (crude rates from counts) | CDC WONDER underlying cause of death |   |
| wzr_od_i3_22 | 54.1% | percent above the 2019 rate | pooled crude overdose rate of the third in 2022 over its 2019 rate, minus 1; bootstrap over counties with shared draws | residents of the most deprived third of complete overdose counties with at least one death, thirds cut again among them | 2022 against 2019 | 95% CI 45.8% to 63.9% | 263 counties | none (crude rates from counts) | CDC WONDER underlying cause of death |   |
| x_alc_24 | 6,259 | deaths above the 2019 rate | observed alcohol-induced deaths minus the 2019 crude rate times the 2024 population; Poisson interval | residents of the United States | 2024 | 95% CI 5,675 to 6,842 | 46,714 deaths | none (full counts) | CDC WONDER underlying cause of death |   |
| x_alc_peak | 14,780 | deaths above the 2019 rate | largest yearly difference between observed alcohol-induced deaths and the 2019 rate times the year's population | residents of the United States | the year of the largest excess in 2020 to 2024 |   |   | none (full counts) | CDC WONDER underlying cause of death |   |
| x_alc_total | 50,558 | deaths above the 2019 rate | observed alcohol-induced deaths minus the 2019 crude rate times each year's population, summed over 2020 to 2024; Poisson interval | residents of the United States | 2020 to 2024 | 95% CI 48,358 to 52,757 | 249,162 deaths | none (full counts) | CDC WONDER underlying cause of death |   |
| x_od_24 | 6,200 | deaths above the 2019 rate | observed drug overdose deaths minus the 2019 crude rate times the 2024 population; Poisson interval | residents of the United States | 2024 | 95% CI 5,427 to 6,972 | 79,384 deaths | none (full counts) | CDC WONDER underlying cause of death |   |
| x_od_peak | 36,225 | deaths above the 2019 rate | largest yearly difference between observed drug overdose deaths and the 2019 rate times the year's population | residents of the United States | the year of the largest excess in 2020 to 2024 |   |   | none (full counts) | CDC WONDER underlying cause of death |   |
| x_od_total | 131,549 | deaths above the 2019 rate | observed drug overdose deaths minus the 2019 crude rate times each year's population, summed over 2020 to 2024; Poisson interval | residents of the United States | 2020 to 2024 | 95% CI 128,564 to 134,533 | 490,830 deaths | none (full counts) | CDC WONDER underlying cause of death |   |
| x_su_total | 99 | deaths above the 2019 rate | observed suicide deaths minus the 2019 crude rate times each year's population, summed over 2020 to 2024; Poisson interval | residents of the United States | 2020 to 2024 | 95% CI −2279 to 2476 | 241,778 deaths | none (full counts) | CDC WONDER underlying cause of death |   |

## Definitions

- Suppression: The withholding of a number because the count behind it is too small to publish without risking the identification of a person. A suppressed value is missing, not zero.
- Crude rate: Deaths or births divided by the population, with no adjustment for age. A place with older residents has a higher crude death rate even if health at each age is the same.
- Modeled estimate: A figure produced by a statistical model instead of a direct count or measurement, usually for small areas where direct data are thin, borrowing strength from surveys, neighboring areas or other data. Neighboring areas can share errors, so single extremes call for caution.
- Area Deprivation Index (ADI): A score that ranks neighborhoods by income, education, employment and housing conditions drawn from the American Community Survey. In this book a higher national percentile means a more deprived neighborhood.
- Complete county: A county whose death count is published in every year of a window, possibly as zero, so its rate over the window is exact. Counties that are complete are selected by their counts, so as a group they are not typical of counties their size.
- Fixed effects: A regression adjustment that gives each group, such as each county, its own baseline. The estimate then rests on differences within groups.
- Ecological inference: Drawing conclusions about individuals from data about places or groups. A pattern across counties need not hold for the people inside them.

## Methods

This chapter follows death rates from overdose, alcohol-induced causes and suicide for the nation and for counties, 2018 to 2024, from two sources: CDC WONDER death counts and CDC's modeled county injury and violence panel. WONDER withholds small county-year counts, so its county series use only complete counties, those with a published count in every year, and these counties are selected on the outcome as well as on size. The panel's rates for small counties are modeled and its 2024 values are provisional, while WONDER's 2024 counts are final; Connecticut's counties are in neither county series. County deprivation is grouped in thirds, and rates pooled over counties are weighted by the county population of the year.

### National series and the age of the data

National series for CDC WONDER sum the 51 state rows, which have no withheld cell, and divide by the sum of WONDER's state populations. Rates are crude, per 100,000 residents.

The 2022 suicide count is 49,476, as CDC reports. WONDER's 2024 counts are final, and the panel's 2024 values are provisional. North Carolina amended many 2023 overdose death certificates after WONDER's final 2023 file, so WONDER's 2023 overdose counts for that state, and the 2023 overdose points built from them, are low.

Explains Table 1 and Table 2.

### Which counties the WONDER county series cover

WONDER withholds a county-year count above zero and below 10 and publishes a count of 0, so a rate pooled over the published years of a county is biased. Nothing is imputed, bounded or derived from totals.

County series use complete counties, those whose count is published in every year 2018 to 2024, as a count of zero or at least 10, and that have a population in every year: 889 for overdose, 629 for alcohol-induced deaths and 718 for suicide. The chapters on overdose, alcohol and suicide and on suppressed counties in death data use the window 2018 to 2023, so their sets of complete counties are larger than these, which also require 2024.

Complete counties are selected on the outcome as well as on size, and they include counties with zero deaths in all seven years: 99 for overdose, 39 for alcohol-induced deaths and 24 for suicide. Connecticut's eight counties have no WONDER population for 2022 to 2024 and are never complete.

Explains Table 5 and Table 4.

### CDC's modeled county panel

The CDC injury and violence panel (county file, data as of May 2026) carries a modeled rate for nearly every county except Connecticut's; overdose is unintentional or undetermined-intent drug overdose, and suicide and homicide include every method. A county counts toward a panel series in a year when CDC publishes a stable rate. The panel's national series is the mean of county rates weighted by WONDER's county population of the year, over the counties with a stable rate that year. Event studies and thirds use the balanced counties, those with a stable rate in all six years.

The change in the overdose rate from 2022 to 2024 uses counties with a stable rate in both years; a county with no deaths in either year (303 counties) has no direction of change and is left out of every share and mean, and thirds are cut before that exclusion.

Explains Table 3, Table 8 and Table 4.

### Population weights and county deprivation

Weights for rates pooled over counties are WONDER's county populations of the year, with Chugach and Copper River (WONDER's Valdez-Cordova) sharing its population by 2019-2023 ACS residents. The panel's implied deaths are the county rate times that population over 100,000. The population benchmark for 2024 uses the yearly changes of 2019 to 2023.

County deprivation is the mean of the Area Deprivation Index national ranks of a county's tracts, weighted by 2019-2023 ACS tract population, with Connecticut tracts assigned to the eight legacy counties; thirds hold equal numbers of counties among those in each series that have a value, and counties without one are left out of the thirds (9 of the complete overdose counties, 13 of the panel's balanced overdose counties and 14 counties in the map).

Explains Table 5 and Table 7.

### How uncertainty is measured

Intervals use gamma limits under 100 deaths and a normal approximation on the count above that; an index divides two rates and takes a normal interval on the log of the ratio of the two death counts.

Rates pooled over counties, ratios and differences of pooled rates, and shares of counties have 95% percentile intervals from a bootstrap that resamples whole counties (2,000 replicates, seed 12345), with shared draws across years.

### Event studies and excess deaths

Event studies are two-way fixed-effects regressions of the county death rate on year indicators, with a fixed effect for every county, weights equal to the county population of the year and errors clustered by state (50 clusters: the states and DC), fitted with pyfixest; the reference year is 2019, and the differential versions interact the year indicators with an indicator for the most deprived third, against the least deprived third, with year fixed effects.

Excess deaths subtract the 2019 crude rate times each year's population from the year's deaths; their interval treats the 2019 count and each year's count as Poisson.

Explains Table 6, Table 7 and Table 2.

## What these data cannot show

Every figure describes counties and the nation. A county rate is an average over all of its residents, so it cannot say who died, and a pattern across counties need not hold for the people in them (ecological inference). The data do not measure mechanisms. The event studies describe how rates moved in the same counties around 2020 and cannot separate deaths from the virus, changes in the supply of drugs and alcohol, disrupted treatment and services, or anything else that changed in the same years. The data do not include COVID deaths.

The panel's 2024 values are provisional and may change, and WONDER's 2024 population estimate comes from a new vintage. WONDER's 2024 counts are final, but its 2023 overdose count is low for North Carolina. The size of the fall in overdose deaths in 2024 may therefore change. WONDER's county series cover complete counties, which are selected on their counts (large enough to reach the threshold every year, or very small with no deaths), and WONDER's withheld counts are not recoverable. The panel covers nearly every county but models the rates of counties with few deaths, so its rate for a small county is partly its neighbors' rate. Where WONDER publishes every year the two sources agree on the overdose rise by deprivation. In the counties WONDER withholds, only the panel has a rate, and the data cannot test whether its larger rise in the most deprived third is real or the model smoothing small counties toward their neighbors. The difference between WONDER's and the panel's national rises is consistent with the intent definitions, but the files used here do not record intent, so it is not proved.

Rates are crude, and a county with an older population has a higher rate for the same risk at each age. Thirds of counties are relative to the counties in each figure, and county deprivation is measured once, from the Area Deprivation Index, which understates deprivation in high-cost cities (Chapter 1.1, What the deprivation index measures (https://vishalsingh.org/zipcode/anatomy-of-deprivation)). The event study has one pre-period year at most, and the differential version for suicide fails that check. Homicide is available only from the panel and counts every method, so it differs from the firearm homicide in Chapter 8.1, Gun homicide and the causes of early death (https://vishalsingh.org/zipcode/gun-homicide-and-causes-of-early-death). Alcohol-induced deaths are deaths whose underlying cause is alcohol, and they leave out deaths from other conditions in which alcohol played a part. Outside the studies cited, nothing in this chapter supports a claim about what raised or lowered any of these death rates.

## References

1. Case, Anne; Deaton, Angus (2020). Deaths of Despair and the Future of Capitalism. Princeton University Press.
2. Garnett, Matthew F.; Minino, Arialdi M. (2024). Drug overdose deaths in the United States, 2003-2023. NCHS Data Brief No. 522, National Center for Health Statistics. https://doi.org/10.15620/cdc/170565
3. White, Aaron M.; Castle, I-Jen P.; Powell, Patricia A.; Hingson, Ralph W.; Koob, George F. (2022). Alcohol-related deaths during the COVID-19 pandemic. JAMA 327(17): 1704-1706. https://doi.org/10.1001/jama.2022.4308
4. Kegler, Scott R.; Simon, Thomas R.; Zwald, Marissa L.; Chen, May S.; Mercy, James A.; Jones, Christopher M.; Mercado-Crespo, Melissa C.; Blair, Janet M.; Stone, Deborah M.; Ottley, Phyllis G.; Dills, Jennifer (2022). Vital Signs: Changes in firearm homicide and suicide rates, United States, 2019-2020. MMWR Morbidity and Mortality Weekly Report 71(19): 656-663. https://doi.org/10.15585/mmwr.mm7119e1
5. Roth, Jonathan; Sant'Anna, Pedro H. C.; Bilinski, Alyssa; Poe, John (2023). What's trending in difference-in-differences? A synthesis of the recent econometrics literature. Journal of Econometrics 235(2): 2218-2244. https://doi.org/10.1016/j.jeconom.2023.03.008
6. Friedman, Joseph R.; Palamar, Joseph J.; Ciccarone, Daniel; Gaines, Tommi L.; Borquez, Annick; Shover, Chelsea L.; Strathdee, Steffanie A. (2026). Charting the decline of the fourth wave: US overdose deaths by race, ethnicity and substance involvement. Addiction 121(9): 2552-2558. https://doi.org/10.1111/add.70472

## Provenance

- Pipeline version: 0.1.0
- Build script: book2/chapters/mortality-trends-2018-2024/build.py (sha256 bcfad2efb8097eccadced4c52c3fa55377d2df4bde70dda892aa93b79b679af7)
- Built on: 2026-10-07
- Site commit: c2b1ff3309cb4a10833f884da91ac0057a66ad0a

Inputs:

| File | Size (bytes) | SHA-256, first 1 MB | SHA-256, last 1 MB |
| --- | ---: | --- | --- |
| data/analytic/tract_atlas.parquet | 39,659,925 | 29d5e4d943552d54aeb6ba92aadc939d9bd15bdea553c78357dbd516757be627 | 1e10d1ecf722340f23ee2212d4999a90250f85b65fa89b39e32db06d9472e22e |
| data/staged/crosswalks/ct_tract2020_to_old_county.parquet | 15,842 | 961ef39f31596ef65fca2c67d5f2f63e533c5dc9b49745d3bb6437844e4b831d | 961ef39f31596ef65fca2c67d5f2f63e533c5dc9b49745d3bb6437844e4b831d |
| data/staged/mortality/injury_violence_county.parquet | 722,427 | d0dfd060a3ab9eaa3a87b35d1b24b5a1e9f21af3513c10367edb651c46f40403 | d0dfd060a3ab9eaa3a87b35d1b24b5a1e9f21af3513c10367edb651c46f40403 |
| data/staged/mortality/wonder_ucd_alcohol_county_2018_2024.parquet | 289,631 | 478b6b5f8ffa061bcca1df84162d1e169babeb7832bc9818379af9be382a778f | 478b6b5f8ffa061bcca1df84162d1e169babeb7832bc9818379af9be382a778f |
| data/staged/mortality/wonder_ucd_alcohol_state_2018_2024.parquet | 16,162 | 94c821fe0bce60f99f2a300acf8e929b857a8b0f780f5634cadce0698fd665a4 | 94c821fe0bce60f99f2a300acf8e929b857a8b0f780f5634cadce0698fd665a4 |
| data/staged/mortality/wonder_ucd_drug_county_2018_2024.parquet | 308,988 | b0e5cb6dfb44fc1ce53a192faf74cb84277bb5d601a3fc32b65078a7ec03bbf1 | b0e5cb6dfb44fc1ce53a192faf74cb84277bb5d601a3fc32b65078a7ec03bbf1 |
| data/staged/mortality/wonder_ucd_drug_state_2018_2024.parquet | 17,336 | 72a7ecbb0c49a8ba9b90b22f6fad1a6d19c025c256c72799a7acf2369ebabe44 | 72a7ecbb0c49a8ba9b90b22f6fad1a6d19c025c256c72799a7acf2369ebabe44 |
| data/staged/mortality/wonder_ucd_suicide_county_2018_2024.parquet | 290,192 | 448558540a3f42296acf7f66a945f7c6794fd6f4dc9c7032a18a3d105d0541f1 | 448558540a3f42296acf7f66a945f7c6794fd6f4dc9c7032a18a3d105d0541f1 |
| data/staged/mortality/wonder_ucd_suicide_state_2018_2024.parquet | 15,948 | 7588fe51ddda48e57e88852172442b4d2bf64689882e600ebf342668ff79ea1a | 7588fe51ddda48e57e88852172442b4d2bf64689882e600ebf342668ff79ea1a |

Outputs:

| File | SHA-256 |
| --- | --- |
| facts.json | 773e2fb72bb8e26c13ee6c6cb84a4cfd2b19184b1fbc92ffcca4eb5d1b20cccd |
| figures/event_study.json | 80c9773780d1203cdb212b3888ef522e28f612f4f821839cd0cf39d8d17f6ad7 |
| figures/event_study_deprivation.json | 547c348e8dbe92acea6715fd47af0ad62c6affa2e9338ab7ba199b2ee1489bf1 |
| figures/excess_deaths.json | fcd1e94858c682fe45ce5fd152127716185ede50cf83a9ffad5d92bf014acf4c |
| figures/national_rates.json | b070d31bbce8e2333efe43fdc1791f2466b605ccc1faefa06654ff9c75f31dcb |
| figures/overdose_turn_map.json | 5376e1e1291fcc70ecb9aa8e7ffb5ea8e67aa43692a464a8e567c70c81b7b171 |
| figures/overdose_two_sources.json | a2c9312c2d7435c09d6d1b8a06356927dbb191e36abf471316822fd7cc7aada3 |
| figures/panel_three_clocks.json | 0beb2690fa4d282b6852b6a275e4885aac7cca87894402b1f6268e414fa8244a |
| figures/trend_by_deprivation.json | 724999678f46e40e01b30a9875bd23a36cdf31162b051f988eed713fe579e20c |
| meta.json | 8a12998b3adebf671bdbc7ccaada4e43fef3fe78ab62cc34a50a8bc269580372 |
| public/data/mortality-trends-2018-2024/overdose_turn.json | 08c248d24ca2870397734acf85fc8275c298a6df9fe1cdd4e1decd6a7dfd8d51 |
| refs.json | 72d7c6081d1b9b077d366a670dc84939f5eb0ad89d75cc9450db1acc1e6934da |

## Machine formats

- Report (HTML): https://vishalsingh.org/zipcode/mortality-trends-2018-2024/report
- Report (Markdown): https://vishalsingh.org/zipcode/mortality-trends-2018-2024/report.md
- Every estimate (JSON): https://vishalsingh.org/zipcode/report/mortality-trends-2018-2024/facts.json
- Table 1, figure "national_rates" (JSON): https://vishalsingh.org/zipcode/report/mortality-trends-2018-2024/figures/national_rates.json
- Table 2, figure "excess_deaths" (JSON): https://vishalsingh.org/zipcode/report/mortality-trends-2018-2024/figures/excess_deaths.json
- Table 3, figure "panel_three_clocks" (JSON): https://vishalsingh.org/zipcode/report/mortality-trends-2018-2024/figures/panel_three_clocks.json
- Table 4, figure "overdose_two_sources" (JSON): https://vishalsingh.org/zipcode/report/mortality-trends-2018-2024/figures/overdose_two_sources.json
- Table 5, figure "trend_by_deprivation" (JSON): https://vishalsingh.org/zipcode/report/mortality-trends-2018-2024/figures/trend_by_deprivation.json
- Table 6, figure "event_study" (JSON): https://vishalsingh.org/zipcode/report/mortality-trends-2018-2024/figures/event_study.json
- Table 7, figure "event_study_deprivation" (JSON): https://vishalsingh.org/zipcode/report/mortality-trends-2018-2024/figures/event_study_deprivation.json
- Table 8, figure "overdose_turn_map" (JSON): https://vishalsingh.org/zipcode/report/mortality-trends-2018-2024/figures/overdose_turn_map.json
- Payload of "overdose_turn_map" (JSON): https://vishalsingh.org/zipcode/data/mortality-trends-2018-2024/overdose_turn.json
- Index of every report: https://vishalsingh.org/zipcode/llms.txt; all reports in one file: https://vishalsingh.org/zipcode/llms-full.txt

---

# Place and the length of a life: report

Chapter 8.4 of ZIP Code Destiny, by Vishal Singh (NYU Stern). Generated from the chapter's data files; the reading version is at https://vishalsingh.org/zipcode/place-and-the-length-of-a-life. Table n of this report is figure n of the chapter.

## Identification

- Title: Place and the length of a life
- Label: Two addresses
- Chapter: 8.4 (Part VIII, Early death; An American life)
- Question: How far apart are life expectancies in nearby neighborhoods, and how steadily does life expectancy fall with neighborhood deprivation?
- Dek: Life expectancy at birth was 82.3 years in the least deprived tenth of census tracts and 74.0 years in the most deprived, for deaths recorded in 2010 to 2015.
- Grain: neighborhoods
- Status: published
- Published: 2026-10-09
- Updated: 2026-10-11
- Data sources:
  - USALEEP life expectancy. USALEEP life expectancy at birth with standard errors, 2010-2015: native 2010 tracts for the pairs, 2020 tracts (area-weighted from 2010 tracts, institutional tracts dropped) for everything else; years: 2010-2015; weights: ACS 2019-2023 tract population
  - Area Deprivation Index. Neighborhood Atlas Area Deprivation Index 2023, national percentile, 2020 tracts; years: 2023 release (version 4.0.1; built on ACS 2019-2023); weights: ACS 2019-2023 tract population, also for the ADI's split by county (over the frame of tracts of the chapter on gradients, slopes and variance)
  - American Community Survey. ACS 2019-2023 tracts: population (weights and the population floor of the pairs); years: 2019-2023 (5-year); weights: tract population
  - Census TIGER/Line tracts. Census 2020 tracts: internal points (distances in the pairs) and tract names; Census 2020 county file: the core-based statistical area of each county; years: 2020; weights: none (locations)
  - CDC PLACES. CDC PLACES 2024 release, tracts: current smoking and routine checkup among adults; years: 2024 release (tracts); 2025 release (ZCTAs, 2023 data year); weights: adult population of each tract (PLACES)
  - Published estimates from cited studies. Estimates from Ludwig and coauthors (2011) and Chetty, Hendren and Katz (2016), quoted in the box on the Moving to Opportunity experiment; years: as reported in each study; weights: as reported in each study
- Canonical URL: https://vishalsingh.org/zipcode/place-and-the-length-of-a-life
- Report: https://vishalsingh.org/zipcode/place-and-the-length-of-a-life/report
- Markdown: https://vishalsingh.org/zipcode/place-and-the-length-of-a-life/report.md
- BibTeX key: singh2026placeandthelengthofalife

```bibtex
@misc{singh2026placeandthelengthofalife,
  author = {Singh, Vishal},
  title = {Place and the length of a life},
  year = {2026},
  url = {https://vishalsingh.org/zipcode/place-and-the-length-of-a-life},
  note = {In ZIP Code Destiny. Published 9 October 2026; updated 11 October 2026.}
}
```

## Principal estimates

The facts the dek and the opening prose cite, in order of appearance.

| Estimand | Population | Period | Value | Interval | n | Weights | Source | Fact id |
| --- | --- | --- | ---: | --- | ---: | --- | --- | --- |
| population-weighted mean of tract life expectancy in a tenth of tracts (rank bins of the ADI; tracts with tied ranks fall in the same tenth), bootstrap interval over counties | residents of tracts in the least deprived tenth of tracts by ADI national percentile | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population | 82.3 | 95% CI 82.0 to 82.6 | 7,820 tracts | ACS 2019-2023 tract population | USALEEP life expectancy | le_d1 |
| population-weighted mean of tract life expectancy in a tenth of tracts (rank bins of the ADI; tracts with tied ranks fall in the same tenth), bootstrap interval over counties | residents of tracts in the most deprived tenth of tracts by ADI national percentile | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population | 74.0 | 95% CI 73.7 to 74.3 | 7,844 tracts | ACS 2019-2023 tract population | USALEEP life expectancy | le_d10 |
| count of tracts in the gradient | residents of non-institutional tracts in the 50 states and DC with a life expectancy estimate and an ADI rank | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population | 77,936 |   |   | none (full count) | USALEEP life expectancy | n_gradient_tracts |
| difference between the population-weighted mean life expectancy of the least and the most deprived tenth of tracts, bootstrap interval over counties (draws shared by the two means) | residents of non-institutional tracts in the 50 states and DC with a life expectancy estimate and an ADI rank | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population | 8.3 | 95% CI 7.9 to 8.7 | 7,844 tracts | ACS 2019-2023 tract population | USALEEP life expectancy | le_decile_drop |

## Findings

Life expectancy at birth was 82.3 years in the least deprived tenth of census tracts and 74.0 years in the most deprived, for deaths recorded in 2010 to 2015.[1] Both are population-weighted averages of tenths cut from the 77,936 non-institutional tracts in the states and DC that have a life expectancy estimate and a deprivation rank. The difference between them is 8.3 (95% CI 7.9 to 8.7) years, with an interval from resampling whole counties. This chapter asks how steadily life expectancy falls with deprivation, what a county average leaves out, how far apart nearby neighborhoods are, whether the ordering repeats inside metros and counties, and what an experiment adds about moving. The introduction (https://vishalsingh.org/zipcode/place-across-an-american-life) places the comparison among the book's other outcomes and sources.

A life expectancy estimate summarizes the death rates of a place during a period. It is the average age at death of newborns who would meet, at every age, the death rates that the tract recorded in 2010 to 2015, and it does not predict the lifespan of a particular child. Deprivation is measured separately, by the Area Deprivation Index (ADI), a national percentile built from census measures of income, education, employment and housing.[2] The ADI describes 2019 to 2023, after the deaths behind the life expectancy estimates, so the opening comparison sets an earlier pattern of mortality beside a later description of neighborhood conditions.

The distance between the two averages calls for a closer look. Comparing only the ends of a scale leaves the middle unseen. Comparing distant places mixes differences in their residents with differences in their surroundings, and a detailed map can exaggerate a contrast when each neighborhood's estimate is uncertain. The sections below take up those problems in turn.

### A national pattern needs a closer look

Arranging the tracts from least to most deprived and cutting the ranking into tenths gives a mean life expectancy that is lower in every tenth than in the one before (Table 1). The means are weighted by tract population. The ordering holds along the whole scale, and the size of the step changes along the way. The step between the two least deprived tenths is 1.3 (95% CI 1.0 to 1.6) years and the step between the two most deprived is 1.9 (95% CI 1.6 to 2.1) years, and the steps are largest at the two ends. Between neighboring ADI percentiles the line sometimes rises by a fraction of a year. In the units of the index itself, life expectancy is 0.84 (95% CI 0.75 to 0.92) years lower for each ten-point step in the ADI percentile across all tracts.

Tracts with similar ranks still differ. A tract's estimate has a median standard error of 1.7 years, so part of the spread among tracts at one rank is estimation error, and part is differences that the deprivation rank does not summarize. The line describes an ordering of averages and leaves much about individual places unexplained.

Table 1. Life expectancy is lower in each higher tenth of tracts by deprivation, and smoking rises with it while the annual checkup rises little

The figure keeps these ideas apart. The line is the mean at each ADI percentile. The shaded band is the interquartile range of tract estimates, a spread of places that includes estimation error and spans several years at every percentile. The markers on the tenths carry intervals from a bootstrap over counties, which describe uncertainty about those averages. Averaging over thousands of tracts shrinks the error of a mean to a small interval, so a wide band of neighborhoods can surround a precisely estimated line, and the band is not the uncertainty of the line. A single dot in the first tab says little on its own.

### County averages leave differences out

A county is a useful administrative unit, and it can contain neighborhoods with different incomes, housing and health. New York City makes the problem visible, because each of its five boroughs is also a county. Table 2 shades the 2,074 non-institutional tracts of the city that have an estimate and an ADI rank. In every borough the 5th to 95th percentile of tract life expectancy spans at least 9.3 years, while the borough means lie within 4.2 years of one another.

The shades need care. The estimates describe deaths in 2010 to 2015 and were moved by area from the 2010 tract boundaries onto the 2020 outlines drawn here. The published standard errors, which understate the noise, put estimation error at 37% of the variance of these estimates, so the map tends to exaggerate the contrast between neighbors. It gives a view inside the borough averages, and the edge between two shades is not a precise boundary in residents' prospects. A gray tract has no estimate or was excluded, and it is not a zero.

Table 2. Within every New York City borough, the 5th to 95th percentile of tract life expectancy spans at least 9.3 years, while the borough means lie within 4.2 years of one another

A comparison of nearby neighborhoods needs a selection rule as well. Tract life expectancy is a modeled estimate with a standard error, and picking the pairs with the largest apparent gaps would also pick the largest estimation errors. The pairs in Table 3 are chosen instead by their difference in deprivation, in the 20 most populous counties only, subject to requirements on distance, population, boundary stability and precision. The rule never compares the life expectancy of the tracts in a pair; it uses the estimate only to require one with a small enough standard error. Each pair uses the native estimates for deaths in 2010 to 2015 on the 2010 tract boundaries and the ADI for 2019 to 2023, so the deprivation measures date from about a decade after the deaths.

Table 3. In 17 large counties, the pair of tracts within 5 kilometers with the widest ADI difference had higher life expectancy in the less deprived tract in 15, and a median gap over the pairs of 5.4 years

The pairs are selected by deprivation differences rather than by their life expectancy gaps, so the selection rule does not directly choose the largest outcome estimates.

Method note, How the neighborhoods were paired: A tract qualifies if its boundary did not change between the 2010 and 2020 censuses, it is not an institutional tract, it has at least 2,000 residents and the standard error of its estimate is at most 2.0 years. Among qualifying tracts within 5 kilometers of each other, a county's pair is the one whose ADI differs most, and it counts if the difference reaches 50 percentile points.

In 15 of the 17 pairs the less deprived tract has the higher estimate, and in 12 the 95% interval for the gap lies above zero. The median gap over all 17 pairs, including those whose gap is not positive, is 5.4 (95% CI 3.0 to 6.9) years, with an interval from a bootstrap over pairs. The gap treats the two standard errors as independent.

This is a result about the pairs that meet the rule in the large counties searched. It does not estimate the gap between a typical pair of neighbors in the country, and it depends on the settings of the rule, above all the radius. Pairs chosen within 3 kilometers have a median gap of 2.7 years (see [alternative choices](https://vishalsingh.org/zipcode/place-and-the-length-of-a-life/report#methods-alternative-choices) in Methods for the other settings).

Showing the complete set keeps the widest contrast from standing in for every comparison. The widest pair is in Cook County, Illinois, and its tracts lie 5.0 kilometers apart. The more deprived one (ADI 94) had a life expectancy of 65.8 (95% CI 62.5 to 69.1) years and the less deprived one (ADI 24) 78.2 (95% CI 74.7 to 81.7) years, a gap of 12.4 (95% CI 7.6 to 17.2) years. The widest pair of a set is also the one most likely to have been widened by estimation error.

Wider gaps may exist among tracts the rule does not pair, because the ADI ranks some tracts alike that differ widely in life expectancy, as it does in high-cost cities (Chapter 1.1, What the deprivation index measures (https://vishalsingh.org/zipcode/anatomy-of-deprivation)). The largest gaps among noisy estimates are also the least reliable, so the rule does not seek them.

### Most variation lies between tracts in the same county

Splitting the variance of tract life expectancy into a part between states, a part between counties in a state and a part between tracts in a county gives 13.7%, 15.9% and 70.3% (Table 4). The split comes from every one of 78,008 tracts in 3,112 counties and Connecticut planning regions and has no sampling interval.

The within-county share includes estimation error. The tract estimates are modeled, and their mean squared standard error equals 26.7% of the total variance, all of it inside the within-county share. Taking the published errors at face value, 60% of the variance that is not estimation error lies within counties. Because the 2020-tract errors are mapping grade and understate the noise, the true share is lower.

The ADI is the contrast. Only 34% of its variance lies within counties, so the index that ranks tracts for the gradient places most of its variation between counties, while life expectancy places most of its variation inside them. Chapter 1.1, What the deprivation index measures (https://vishalsingh.org/zipcode/anatomy-of-deprivation) examines what the ADI is made of.

Table 4. 70.3 percent of the variance of tract life expectancy lies between tracts in the same county, 26.7 points of it estimation error, against 34% percent for the ADI

Table 5 shows the range inside the 30 most populous counties. The 5th to 95th percentile range of tract life expectancy has a median of 10.5 years across them. The widest, 13.9 years in Cook County, is wider than the 8.3-year difference between the means of the nation's least and most deprived tenths, though a range of tract estimates and a difference between means are not the same kind of contrast.

Part of every range is estimation error. Shrinking each tract estimate toward its county mean approximates the range net of that error and gives a median range of 7.7 years. The approximation takes the published errors at face value and does not recover the true differences between tracts exactly, and because those errors understate the noise, the shrunken range is an upper bound. The median share of the variance within a county that is estimation error is 43% across these counties, against 38% for all counties together, and each estimate is shrunk by the square root of one minus its county's share.

Table 5. In the 30 most populous counties, tract life expectancy spans a median of 10.5 years between the 5th and 95th percentile, 7.7 net of estimation error

A map that paints each county one color shows the county mean and hides the range around it. Chapter 1.2, What a county average leaves out (https://vishalsingh.org/zipcode/deprivation-at-two-zooms) builds the deprivation index at two zooms to see how much a county-level measure keeps.

### Repeating the comparison changes the question

The national line compares neighborhoods across the country, and it could arise from where deprived tracts are, with the least deprived concentrated in some regions and the most deprived in others. A different comparison asks whether the ordering also appears among neighborhoods of the same area.

Fitting the line separately in each metro, meaning a core-based statistical area of the Census Bureau, tests that. The 267 metros that have at least 40 tracts with an estimate and an ADI rank hold 82% of the residents in the sample. All of them show lower life expectancy in more deprived tracts (Table 6), and the 95% interval lies above zero in 260 of them. The median metro has life expectancy 1.20 (95% CI 1.14 to 1.23) years lower per ten-point step.

At the 5th percentile of metros the difference is 0.70 years and at the 95th 1.67. Those percentiles overstate the real spread of slopes, and a ranking of local slopes would put the noisiest estimates at the extremes. Estimation error makes up 48% of the variance of the estimated slopes, and the order of the dots in the figure puts the least precise slopes at both ends, so no metro is named steepest or flattest. Chapter 10.1, Gradients, slopes and variance (https://vishalsingh.org/zipcode/gradients-slopes-and-variance) shrinks the estimates and reports the same metro slopes.

Table 6. Life expectancy is lower in more deprived tracts in all 267 metros with enough tracts, by a median of 1.20 years per ten-point step

Comparing only tracts in the same county, with a separate baseline for each county (fixed effects), gives a slope of 1.12 (95% CI 1.07 to 1.16) years per ten-point step, against 0.84 (95% CI 0.75 to 0.92) across all tracts. The within-county slope is the larger of the two, and its interval lies entirely beyond the interval of the national slope, so the same difference in ADI goes with a larger difference in life expectancy inside a county than across the country. These comparisons describe the geography more closely and leave the causal question open. Chapter 10.1, Gradients, slopes and variance (https://vishalsingh.org/zipcode/gradients-slopes-and-variance) sets out how the slopes and the split of variation are built (the procedures used here are in [Methods](https://vishalsingh.org/zipcode/place-and-the-length-of-a-life/report#methods-local-slopes-and-variance)).

### Some outcomes follow a different pattern

Other chapters ask whether the ordering seen in longevity holds for other measures of life, and where it breaks down. The second tab of Table 1 gives a first example. In the PLACES 2024 release, which models survey responses from 2022, the share of adults who smoke was 9.1% (95% CI 8.5% to 9.6%) in the least deprived tenth of tracts and 22.8% (95% CI 22.5% to 23.1%) in the most deprived, 13.7 (95% CI 13.1 to 14.4) points higher.

The share who had a routine checkup in the past year was 74.2% and 77.8%, only 3.6 (95% CI 2.4 to 4.6) points higher in the most deprived tenth. The checkup records contact with a physician and not the state of anyone's health. Its small difference across the scale is the exception that Chapter 7.1, Health behaviors and the supply of care (https://vishalsingh.org/zipcode/health-behaviors-and-the-supply-of-care) examines beside the dental visit, which falls with deprivation. The smoking and checkup lines are both modeled estimates from CDC PLACES, which draws on survey responses and on census and American Community Survey inputs.[3] The ADI is built from the American Community Survey as well, so part of each of those two gradients is built in.

The choice of comparison matters as well. Traffic pollution is lower in more deprived tracts across the country and higher in more deprived tracts within a county (Chapter 6.1, Air pollution (https://vishalsingh.org/zipcode/air-pollution)). The choice of deprivation measure matters too, because a national ranking combines different kinds of places, and its measures of housing cost can rank expensive neighborhoods in ways that obscure hardship (Chapter 1.1, What the deprivation index measures (https://vishalsingh.org/zipcode/anatomy-of-deprivation)).

Chapter 7.2, Health behaviors, person by person (https://vishalsingh.org/zipcode/health-behaviors-person-by-person) changes the unit. A survey can ask the same respondent about household income and drinking instead of inferring that relationship from neighborhood averages, and among adults measured one at a time, binge drinking rises with household income except among non-Hispanic Black adults. The survey gains information about individuals and loses geographic detail, because its public files do not support the tract comparisons of the maps. PLACES draws on the same survey, so agreement between the two is not an independent replication.

### Moving is a different question

The comparisons so far describe places. A family deciding whether to move faces another question, what would change for that family after the move, and the residents already living in a destination differ from potential movers in ways that a map cannot remove. This is the one section of the chapter that asks what moving does, and it reports a randomized experiment.

> Causal evidence: Whether moving to a lower-poverty neighborhood changes health and later income (Randomized housing-voucher experiment)
>
> Every comparison above is between places, and none says whether a family that moved to a less deprived tract would live longer. The Moving to Opportunity experiment tested part of that question. From 1994 to 1998 the U.S. Department of Housing and Urban Development randomly assigned families in public housing in high-poverty census tracts to a voucher usable only in a low-poverty tract, to an unrestricted voucher, or to no voucher.[4]
>
> In a survey in 2008 to 2010 of the women who had enrolled, the prevalence of a glycated hemoglobin level of 6.5% or more, a marker of diabetes, was 4.3 (95% CI 0.8 to 7.8) points lower in the group offered the low-poverty voucher than in the control group, and the prevalence of a body-mass index of 40 or more was 3.4 (95% CI 0.4 to 6.4) points lower. The unrestricted voucher showed no significant difference.[4]
>
> Chetty, Hendren and Katz followed the children through tax records. Children assigned the experimental voucher before age 13 had an annual income 14% higher in their mid-twenties than the control group, an intention-to-treat comparison that includes families who did not move. Only 48% of the families took up the voucher, and scaling the comparison by take-up gives 31% for the children of families that did.[5]
>
> Neither study measured life expectancy. The health results are for women who volunteered for the program from public housing in the poorest tracts, and the health results and the first income comparison are between groups assigned an offer of a voucher, which is not the same as every family moving. They do not give the size of the change along the full ADI scale.

The book sets this evidence beside its descriptive comparisons and does not treat the experiment as an explanation of the whole gradient.

### Life expectancy falls across tenths of tracts under all four rankings

Over the 71,771 tracts that have every ranking, the ADI baseline runs from 82.3 (95% CI 82.1 to 82.6) years in the least deprived tenth to 74.0 (95% CI 73.8 to 74.3) in the most deprived, with a rank correlation of −1.00 between tenth and mean (Table 7). Under the standardized index, which applies the ADI's coefficients after standardizing each of its measures, the tenths run from 82.4 (95% CI 82.2 to 82.5) to 74.7 (95% CI 74.2 to 75.1) years with a correlation of −1.00. Ranking tracts by median household income gives 82.2 (95% CI 82.0 to 82.4) and 74.2 (95% CI 73.9 to 74.4) years (−1.00), and ranking them by the bachelor's degree share gives 82.3 (95% CI 82.0 to 82.5) and 75.3 (95% CI 74.9 to 75.7) years (−1.00). Under the rule of Chapter 9.1, How far one ranking goes (https://vishalsingh.org/zipcode/how-far-one-ranking-goes) none of the three counts as a change, since the ends differ under each ranking and every correlation stays beyond 0.9 in absolute value.

Table 7. Life expectancy at birth is steadily worse toward the more deprived end under the ADI and under each of the three other rankings

### Where to read next

The rest of this part follows deaths before old age by cause. Chapter 8.1, Gun homicide and the causes of early death (https://vishalsingh.org/zipcode/gun-homicide-and-causes-of-early-death) follows firearm homicide into the few neighborhoods where it concentrates, Chapter 8.2, Overdose, alcohol and suicide (https://vishalsingh.org/zipcode/overdose-alcohol-and-suicide) maps overdose, alcohol-induced deaths and suicide by county, Chapter 8.3, Mortality trends, 2018 to 2024 (https://vishalsingh.org/zipcode/mortality-trends-2018-2024) follows those deaths from 2018 to 2024, and Chapter 8.5, Violence and the life-expectancy gap (https://vishalsingh.org/zipcode/violence-and-the-life-expectancy-gap) asks how much of this chapter's gap between the least and the most deprived tenths homicide accounts for.

Reading paths connect to this chapter's comparisons. [The argument in eight chapters](https://vishalsingh.org/zipcode/path/core) follows the main line through the volumes, [differences inside one county](https://vishalsingh.org/zipcode/path/the-hyper-local-map) follows the within-county comparison, and [outcomes that do not follow the gradient](https://vishalsingh.org/zipcode/path/the-exception) collects the exceptions. The [contents page](https://vishalsingh.org/zipcode/contents) lists every chapter with its finding.

### What these data cannot show

Outside the box, every number here describes places. A tract with a lower life expectancy and a higher ADI is a tract whose residents, taken together, died younger in 2010 to 2015, and the data do not say which of them did, or whether a person who moved would have died later (ecological inference). The pairs show that nearby tracts differ, and the data cannot separate what the places do from who lives in them. The experiment in the box covers women and children who enrolled in one program, and it does not say how much of the gradient in the figures would change if a place changed.

The periods differ. The life expectancy estimates are modeled for deaths in 2010 to 2015 and were never updated. The ADI is a national percentile built on survey data for 2019 to 2023, and it may rank a tract differently from the way the same tract looked in 2010.

The geography is matched imperfectly. The 2020-tract file used for the gradient is area-weighted from 2010 tracts, and its errors are mapping grade, approximations that understate the uncertainty. Because the pairs are chosen on the ADI among large counties, their median describes that rule, as the sensitivity to the radius shows, and says nothing about counties that are small or rural. The counties left out are Los Angeles, Santa Clara and Alameda, all in California, and the ADI understates deprivation in high-cost cities, so a qualifying pair there may exist and not be found. The distances are straight lines and not rides. Puerto Rico has no life expectancy estimates and is not included, and Connecticut's planning regions stand in for its counties. Institutional tracts are dropped, and a tract without an estimate is absent and not zero.

The estimates are modeled. A tract's life expectancy has a median standard error of 1.7 years, which the intervals in Table 3 show. The intervals in the figures come from resampling counties of published estimates, and they leave out error in the life expectancy model beyond its standard errors, error from moving estimates onto 2020 tracts and uncertainty in the ADI. The PLACES measures are modeled estimates built partly from the same census and survey inputs as the ADI, so the lines for smoking and the checkup show how the model maps those inputs into prevalence as well as how adults differ.

## Tables

### Table 1. Life expectancy is lower in each higher tenth of tracts by deprivation, and smoking rises with it while the annual checkup rises little

Tract life expectancy at birth by Area Deprivation Index percentile, and two modeled health measures by tenth of tracts

#### Table 1.1. Life expectancy

Life expectancy at birth falls from 82.3 years in the least deprived tenth of tracts to 74.0 years in the most deprived

Life expectancy at birth against the Area Deprivation Index national percentile, one dot per sampled tract, with the population-weighted mean and the interquartile range of tracts at each percentile, deaths in 2010 to 2015

No inline rows: the values of Table 1.1 are in https://vishalsingh.org/zipcode/data/place-and-the-length-of-a-life/le_by_adi_tracts.json.

#### Table 1.2. Smoking and checkups

Smoking rises from 9.1% to 22.8% across tenths of tracts by deprivation, and the annual checkup moves from 74.2% to 77.8%

Percent of adults who currently smoke and percent who had a routine checkup in the past year, by tenth of tracts by ADI national percentile, modeled tract estimates, BRFSS 2022

| conf_high | conf_low | decile | estimate | n | series |
| ---: | ---: | ---: | ---: | ---: | --- |
| 9.59 | 8.53 | 1 | 9.06 | 8,274 | Current smoking |
| 11.25 | 10.20 | 2 | 10.74 | 8,118 | Current smoking |
| 11.99 | 11.28 | 3 | 11.62 | 8,511 | Current smoking |
| 12.88 | 12.45 | 4 | 12.67 | 8,261 | Current smoking |
| 14.09 | 13.70 | 5 | 13.89 | 8,189 | Current smoking |
| 15.28 | 14.93 | 6 | 15.10 | 8,318 | Current smoking |
| 16.54 | 16.25 | 7 | 16.39 | 8,305 | Current smoking |
| 17.94 | 17.61 | 8 | 17.78 | 8,238 | Current smoking |
| 19.62 | 19.27 | 9 | 19.45 | 8,354 | Current smoking |
| 23.14 | 22.46 | 10 | 22.81 | 8,213 | Current smoking |
| 75.29 | 73.23 | 1 | 74.16 | 8,274 | Annual checkup |
| 75.40 | 73.25 | 2 | 74.27 | 8,118 | Annual checkup |
| 75.35 | 74 | 3 | 74.67 | 8,511 | Annual checkup |
| 75.41 | 74.40 | 4 | 74.90 | 8,261 | Annual checkup |
| 75.89 | 75.06 | 5 | 75.46 | 8,189 | Annual checkup |
| 76.26 | 75.45 | 6 | 75.85 | 8,318 | Annual checkup |
| 76.59 | 75.81 | 7 | 76.21 | 8,305 | Annual checkup |
| 76.96 | 76.26 | 8 | 76.62 | 8,238 | Annual checkup |
| 77.44 | 76.83 | 9 | 77.14 | 8,354 | Annual checkup |
| 78.18 | 77.34 | 10 | 77.76 | 8,213 | Annual checkup |

Source: USALEEP, 2010-2015; Area Deprivation Index, 2023; CDC PLACES 2024 release (BRFSS 2022); population-weighted, modeled estimates

Data: https://vishalsingh.org/zipcode/report/place-and-the-length-of-a-life/figures/le_gradient.json

### Table 2. Within every New York City borough, the 5th to 95th percentile of tract life expectancy spans at least 9.3 years, while the borough means lie within 4.2 years of one another

Life expectancy at birth of 2,074 census tracts in the five boroughs of New York City, deaths in 2010 to 2015, on 2020 tracts; 250 tracts are gray because they have no value

No inline rows: the values of Table 2 are in https://vishalsingh.org/zipcode/data/place-and-the-length-of-a-life/nyc_le_map.json.

Note: The unit is the 2020 census tract. Life expectancy is the USALEEP estimate for deaths in 2010 to 2015, which the National Center for Health Statistics built on 2010 tracts and which is area-weighted onto 2020 tracts here, so the estimate for a 2020 tract blends the 2010 tracts that overlap it and its standard error is mapping grade, which understates the error. Every shade is a modeled estimate with a standard error (the median is 2.0 years), and the interval in the tooltip is a normal-approximation 95% interval built from the standard error.

Note: Taking the published standard errors at face value, estimation error makes up 37% of the variance of these estimates. If the errors of neighboring tracts are independent, the difference between two neighbors carries the sum of their error variances, so a map of noisy estimates exaggerates the contrast between neighbors. That is why the pairs figure chooses its pairs on the ADI, which shares no sampling error with life expectancy. This map marks no tract.

Note: 2,074 tracts are shaded: the non-institutional tracts of the five counties of New York City (Bronx, Kings, New York, Queens and Richmond) that have a life expectancy estimate, an ACS 2019-2023 population above zero and an ADI rank, which is the sample of the gradient. 250 are gray because the tract has no residents in the survey, no published estimate, an institutional population or no ADI rank; the tooltip names which.

Note: Ranges are population-weighted 5th and 95th percentiles of tract estimates, and borough means are population-weighted. A range of tract estimates and a spread of borough means are not the same kind of contrast, and the range includes estimation error. The range net of error shrinks each tract estimate toward its borough mean by the square root of one minus the borough's share of variance that is estimation error, as in the county spread figure, and is an upper bound because the 2020-tract errors are mapping grade.

Source: USALEEP life expectancy at birth, 2010-2015, area-weighted onto 2020 tracts; Area Deprivation Index, 2023; ACS 2019-2023 population; population-weighted ranges; modeled estimates

Data: https://vishalsingh.org/zipcode/report/place-and-the-length-of-a-life/figures/nyc_le_map.json

### Table 3. In 17 large counties, the pair of tracts within 5 kilometers with the widest ADI difference had higher life expectancy in the less deprived tract in 15, and a median gap over the pairs of 5.4 years

Life expectancy at birth in the two tracts of one pair per county, the pair with the widest ADI difference, in 17 of the 20 most populous counties, native 2010 tracts, deaths in 2010 to 2015, 95% intervals

| County and tracts | Life expectancy, more deprived tract (years) | Life expectancy, less deprived tract (years) | Gap (years) | ADI, more deprived tract | ADI, less deprived tract | Distance (kilometers) |
| --- | ---: | ---: | ---: | ---: | ---: | ---: |
| Cook County, IL (tracts 4910 and 7201) | 65.8 | 78.2 | 12.4 | 94 | 24 | 5.0 |
| Harris County, TX (tracts 4222 and 4207) | 74.7 | 80.9 | 6.2 | 100 | 14 | 4.8 |
| Maricopa County, AZ (tracts 4204.01 and 4204.02) | 73.2 | 80.4 | 7.2 | 94 | 14 | 4.7 |
| San Diego County, CA (tracts 202.09 and 207.06) | 79.2 | 78.5 | −0.7 | 69 | 5 | 4.4 |
| Orange County, CA (tracts 748.02 and 751) | 76.2 | 82.3 | 6.1 | 61 | 10 | 2.2 |
| Miami-Dade County, FL (tracts 1.24 and 11.04) | 75.9 | 81.3 | 5.4 | 88 | 9 | 4.5 |
| Kings County, NY (tracts 596 and 422) | 78.5 | 81.0 | 2.5 | 67 | 1 | 3.1 |
| Riverside County, CA (tracts 422.09 and 306.03) | 75.8 | 80.1 | 4.3 | 90 | 11 | 4.7 |
| Dallas County, TX (tracts 136.21 and 136.11) | 77.6 | 80.6 | 3.0 | 72 | 6 | 2.0 |
| Queens County, NY (tracts 240 and 1333) | 76.8 | 83.7 | 6.9 | 66 | 5 | 3.9 |
| King County, WA (tracts 308.01 and 299.02) | 72.8 | 79.4 | 6.6 | 89 | 14 | 4.1 |
| San Bernardino County, CA (tracts 87.10 and 87.08) | 74.2 | 79.1 | 4.9 | 98 | 24 | 2.1 |
| Clark County, NV (tracts 29.65 and 29.47) | 79.1 | 79.6 | 0.5 | 86 | 18 | 3.8 |
| Tarrant County, TX (tracts 1025 and 1054.05) | 69.1 | 79.1 | 10.0 | 82 | 6 | 2.4 |
| Broward County, FL (tracts 104.02 and 105.02) | 79.9 | 79.9 | 0.0 | 92 | 20 | 3.1 |
| Bexar County, TX (tracts 1702 and 1902) | 76.7 | 81.0 | 4.3 | 96 | 30 | 4.6 |
| Wayne County, MI (tracts 5015 and 5506) | 70.9 | 78.3 | 7.4 | 94 | 10 | 4.0 |

Note: The two tracts of a pair are chosen on the ADI, not on life expectancy, so the differences are not the largest that could be found among noisy estimates. A county with no pair that reaches an ADI difference of 50 percentile points is left out (3 of the 20 counties).

Note: Life expectancy is the native estimate for the 2010 tract, used only where the boundary did not change by 2020, meaning that the 2010 and 2020 tracts with one GEOID share at least 99% of the land area of each, so the ADI and the ACS population of the 2020 tract describe the same ground. Distances are straight lines between the Census internal points of the tracts, and the data hold no transit routes. Intervals are normal-approximation 95% intervals from the standard error; the gap treats the two errors as independent. Institutional tracts are excluded.

Source: USALEEP life expectancy at birth on 2010 tracts, 2010-2015, with standard errors; Area Deprivation Index, 2023; ACS 2019-2023; modeled estimates

Data: https://vishalsingh.org/zipcode/report/place-and-the-length-of-a-life/figures/neighborhood_pairs.json

### Table 4. 70.3 percent of the variance of tract life expectancy lies between tracts in the same county, 26.7 points of it estimation error, against 34% percent for the ADI

Share of the population-weighted variance of tract life expectancy and of the ADI national percentile that lies between states, between counties in a state and between tracts in a county, percent

| level | measure | share |
| --- | --- | ---: |
| Between states | Tract life expectancy | 13.75 |
| Between counties in a state | Tract life expectancy | 15.92 |
| Between tracts in a county | Tract life expectancy | 43.62 |
| Estimation error within counties | Tract life expectancy | 26.71 |
| Between states | ADI national percentile | 36.64 |
| Between counties in a state | ADI national percentile | 29.38 |
| Between tracts in a county | ADI national percentile | 33.98 |

Note: Tract life expectancy is modeled. The mean squared standard error of the tract estimates equals 26.7 percent of the total variance and sits inside the within-county share, as the two right-hand segments show; the 2020-tract standard errors are mapping grade and understate the noise, so the share of the true differences that lies between tracts in a county is lower than the net share.

Note: The life expectancy bar uses 78,008 tracts in 3,112 counties and Connecticut planning regions; the ADI bar uses the tracts of the index in the chapter on gradients, slopes and variance. Connecticut's planning regions stand in for its counties.

Source: USALEEP life expectancy, 2010-2015; ACS 2019-2023; Area Deprivation Index, 2023; population-weighted decomposition of the variance, full counts of tracts

Data: https://vishalsingh.org/zipcode/report/place-and-the-length-of-a-life/figures/variance_split.json

### Table 5. In the 30 most populous counties, tract life expectancy spans a median of 10.5 years between the 5th and 95th percentile, 7.7 net of estimation error

Population-weighted mean life expectancy at birth of each county (dot) and the 5th to 95th percentile of its tracts (whisker), sorted by the width of the range, deaths in 2010 to 2015

| County | Mean life expectancy (years) | 5th percentile of tracts | 95th percentile of tracts | Range (years) | Range net of estimation error (years) | Estimation error, share of variance (%) | Tracts | Residents |
| --- | ---: | ---: | ---: | ---: | ---: | ---: | ---: | ---: |
| Cook County, IL | 78.5 | 70.7 | 84.6 | 13.9 | 12.1 | 24 | 1,221 | 4,864,100 |
| Wayne County, MI | 75.4 | 68.9 | 82.2 | 13.3 | 11.7 | 23 | 555 | 1,707,514 |
| New York County, NY | 82.4 | 75.6 | 88.6 | 13.0 | 10.8 | 31 | 250 | 1,405,675 |
| Palm Beach County, FL | 80.6 | 74.0 | 86.9 | 12.9 | 10.8 | 30 | 330 | 1,399,959 |
| Philadelphia County, PA | 75.4 | 69.0 | 81.5 | 12.5 | 10.9 | 25 | 364 | 1,502,783 |
| Hillsborough County, FL | 78.2 | 71.4 | 83.4 | 12.0 | 9.7 | 35 | 287 | 1,317,655 |
| Miami-Dade County, FL | 80.5 | 74.3 | 86.1 | 11.8 | 9.9 | 30 | 655 | 2,544,531 |
| Kings County, NY | 79.9 | 73.8 | 85.3 | 11.6 | 8.8 | 42 | 734 | 2,521,198 |
| Alameda County, CA | 81.0 | 75.3 | 86.5 | 11.2 | 8.6 | 41 | 362 | 1,598,446 |
| Broward County, FL | 79.5 | 73.6 | 84.7 | 11.1 | 8.7 | 39 | 392 | 1,847,256 |
| Dallas County, TX | 77.8 | 72.0 | 82.8 | 10.9 | 8.4 | 40 | 561 | 2,324,374 |
| King County, WA | 81.3 | 76.1 | 86.9 | 10.8 | 8.1 | 44 | 475 | 2,184,985 |
| Harris County, TX | 78.3 | 72.7 | 83.5 | 10.8 | 8.1 | 44 | 1,000 | 4,292,763 |
| Queens County, NY | 81.7 | 76.6 | 87.1 | 10.5 | 7.3 | 52 | 656 | 2,250,510 |
| Orange County, FL | 79.3 | 73.6 | 84.1 | 10.5 | 8.0 | 42 | 248 | 1,339,363 |
| Los Angeles County, CA | 80.6 | 75.4 | 85.9 | 10.5 | 7.8 | 45 | 2,350 | 9,461,723 |
| Clark County, NV | 77.7 | 72.2 | 82.5 | 10.3 | 7.2 | 52 | 475 | 2,040,360 |
| Bronx County, NY | 78.3 | 73.5 | 83.9 | 10.3 | 7.4 | 49 | 326 | 1,369,766 |
| Riverside County, CA | 79.4 | 74.1 | 84.3 | 10.2 | 7.6 | 45 | 492 | 2,352,240 |
| San Diego County, CA | 80.9 | 75.9 | 85.8 | 9.9 | 7.2 | 47 | 681 | 3,038,849 |
| Middlesex County, MA | 81.8 | 77.4 | 87.1 | 9.7 | 7.6 | 40 | 338 | 1,544,211 |
| San Bernardino County, CA | 78.4 | 73.5 | 83.2 | 9.7 | 7.1 | 47 | 447 | 2,093,159 |
| Sacramento County, CA | 78.8 | 74.2 | 83.9 | 9.7 | 7.6 | 39 | 343 | 1,515,289 |
| Bexar County, TX | 78.2 | 73.1 | 82.8 | 9.7 | 7.3 | 43 | 334 | 1,812,879 |
| Maricopa County, AZ | 78.8 | 73.9 | 83.5 | 9.6 | 6.8 | 50 | 865 | 3,980,946 |
| Tarrant County, TX | 78.0 | 73.0 | 82.4 | 9.4 | 6.5 | 52 | 402 | 1,931,602 |
| Orange County, CA | 81.7 | 77.6 | 86.1 | 8.5 | 4.9 | 67 | 575 | 2,964,928 |
| Nassau County, NY | 82.4 | 78.6 | 87.0 | 8.4 | 6.3 | 43 | 272 | 1,358,476 |
| Santa Clara County, CA | 82.8 | 78.8 | 87.1 | 8.3 | 4.2 | 75 | 393 | 1,835,064 |
| Suffolk County, NY | 80.7 | 77.0 | 84.5 | 7.5 | 5.0 | 55 | 359 | 1,400,449 |

Views: Tract estimates (every row); Net of estimation error (every row).

Note: Tract estimates are modeled, so the range of the estimates is wider than the range of the true values. The second view takes the published standard errors at face value: the median share of the variance within a county that is estimation error is 43%, against 38% for all counties together. The shrinkage multiplies the distance of the 5th and 95th percentiles from the county mean by the square root of one minus that share. The 2020-tract errors are mapping grade, so the range net of error is an upper bound. The range of tract estimates is not like-for-like with a difference between means of tenths.

Note: Counties are chosen by population and not by range. With 15 counties the median range is 10.5 years and with 60 it is 10.5. Institutional tracts are dropped. Connecticut's planning regions stand in for its counties.

Source: USALEEP life expectancy, 2010-2015; ACS 2019-2023 population; population-weighted percentiles of tract values, modeled estimates

Data: https://vishalsingh.org/zipcode/report/place-and-the-length-of-a-life/figures/county_spread.json

### Table 6. Life expectancy is lower in more deprived tracts in all 267 metros with enough tracts, by a median of 1.20 years per ten-point step

One regression of tract life expectancy on the ADI national percentile in each metro with enough tracts, shown by metro and as a distribution, deaths in 2010 to 2015

#### Table 6.1. Slope by metro

Life expectancy is lower in more deprived tracts in every metro, from 0.70 years per ten-point step at the 5th percentile of metros to 1.67 at the 95th

Slope of tract life expectancy at birth on the ADI national percentile within each metro, years lower per ten-point step, one dot per metro labeled by its most populous county, deaths in 2010 to 2015

| conf_high | conf_low | errors | estimate | metro | n |
| ---: | ---: | --- | ---: | --- | ---: |
| 1.46 | 1.09 | HC3 errors | 1.27 | Summit, OH | 167 |
| 1.73 | 0.711 | HC3 errors | 1.22 | Dougherty, GA | 43 |
| 1.55 | 0.922 | Clustered on counties | 1.24 | Albany, NY | 232 |
| 1.17 | 0.770 | HC3 errors | 0.968 | Bernalillo, NM | 223 |
| 1.90 | 0.894 | HC3 errors | 1.39 | Rapides, LA | 42 |
| 1.13 | 0.802 | HC3 errors | 0.965 | Lehigh, PA | 188 |
| 1.30 | 0.637 | Clustered on counties | 0.970 | Randall, TX | 64 |
| 1.84 | 1.07 | HC3 errors | 1.45 | Anchorage, AK | 79 |
| 1.71 | 1.15 | HC3 errors | 1.43 | Washtenaw, MI | 86 |
| 1.63 | 0.507 | HC3 errors | 1.07 | Outagamie, WI | 52 |
| 0.934 | 0.363 | HC3 errors | 0.649 | Buncombe, NC | 112 |
| 1.59 | 0.881 | HC3 errors | 1.23 | Clarke, GA | 53 |
| 1.19 | 0.949 | Clustered on counties | 1.07 | Fulton, GA | 1,354 |
| 1.18 | 0.542 | HC3 errors | 0.859 | Atlantic, NJ | 69 |
| 1.06 | −0.236 | HC3 errors | 0.411 | Lee, AL | 43 |
| 1.26 | 0.917 | Clustered on counties | 1.09 | Richmond, GA | 134 |
| 0.846 | 0.511 | Clustered on counties | 0.679 | Travis, TX | 442 |
| 1.16 | 0.751 | HC3 errors | 0.957 | Kern, CA | 212 |
| 1.67 | 1.14 | Clustered on counties | 1.40 | Baltimore, MD | 670 |
| 2.99 | 0.846 | HC3 errors | 1.92 | Barnstable, MA | 52 |
| 1.87 | 0.796 | Clustered on counties | 1.33 | East Baton Rouge, LA | 192 |
| 1.49 | 0.976 | HC3 errors | 1.23 | Jefferson, TX | 105 |
| 1.60 | 0.018 | HC3 errors | 0.807 | Whatcom, WA | 50 |
| 2.18 | 0.665 | HC3 errors | 1.42 | Deschutes, OR | 42 |
| 2.00 | 0.799 | HC3 errors | 1.40 | Yellowstone, MT | 42 |
| 1.88 | 0.700 | HC3 errors | 1.29 | Broome, NY | 64 |
| 1.45 | 1.17 | Clustered on counties | 1.31 | Jefferson, AL | 289 |
| 1.63 | 0.624 | HC3 errors | 1.13 | McLean, IL | 40 |
| 1.17 | 0.843 | Clustered on counties | 1.00 | Ada, ID | 190 |
| 1.80 | 0.798 | Clustered on counties | 1.30 | Middlesex, MA | 1,045 |
| 1.98 | 0.652 | HC3 errors | 1.31 | Boulder, CO | 64 |
| 2.98 | 1.76 | HC3 errors | 2.37 | Kitsap, WA | 53 |
| 1.28 | 0.983 | HC3 errors | 1.13 | Western Connecticut, CT | 212 |
| 1.73 | 1.00 | HC3 errors | 1.36 | Cameron, TX | 118 |
| 1.49 | 1.17 | HC3 errors | 1.33 | Erie, NY | 307 |
| 1.76 | 0.633 | HC3 errors | 1.20 | Chittenden, VT | 50 |
| 1.84 | 1.31 | HC3 errors | 1.58 | Stark, OH | 95 |
| 1.02 | 0.583 | HC3 errors | 0.799 | Lee, FL | 190 |
| 1.63 | 0.653 | HC3 errors | 1.14 | Linn, IA | 65 |
| 2.34 | 1.40 | HC3 errors | 1.87 | Champaign, IL | 43 |
| 1.29 | 0.654 | Clustered on counties | 0.972 | Kanawha, WV | 76 |
| 1.06 | 0.708 | HC3 errors | 0.884 | Charleston, SC | 159 |
| 1.28 | 0.986 | Clustered on counties | 1.13 | Mecklenburg, NC | 599 |
| 1.76 | 0.435 | Clustered on counties | 1.10 | Albemarle, VA | 51 |
| 1.60 | 1.01 | Clustered on counties | 1.31 | Hamilton, TN | 130 |
| 1.23 | 1.06 | Clustered on counties | 1.15 | Cook, IL | 2,183 |
| 2.11 | 1.14 | HC3 errors | 1.63 | Butte, CA | 51 |
| 1.69 | 1.45 | Clustered on counties | 1.57 | Hamilton, OH | 522 |
| 1.22 | 0.711 | HC3 errors | 0.964 | Montgomery, TN | 52 |
| 1.46 | 1.19 | Clustered on counties | 1.33 | Cuyahoga, OH | 589 |
| 1.16 | 0.549 | HC3 errors | 0.854 | Brazos, TX | 60 |
| 1.81 | 1.09 | HC3 errors | 1.45 | El Paso, CO | 155 |
| 1.46 | 0.677 | HC3 errors | 1.07 | Boone, MO | 40 |
| 1.33 | 0.942 | Clustered on counties | 1.14 | Richland, SC | 205 |
| 1.53 | 1.34 | Clustered on counties | 1.43 | Muscogee, GA | 87 |
| 1.74 | 0.972 | Clustered on counties | 1.36 | Franklin, OH | 474 |
| 1.18 | 0.518 | HC3 errors | 0.848 | Nueces, TX | 103 |
| 1.33 | 0.752 | HC3 errors | 1.04 | Okaloosa, FL | 63 |
| 0.986 | 0.834 | Clustered on counties | 0.910 | Dallas, TX | 1,501 |
| 1.36 | 0.726 | HC3 errors | 1.04 | Baldwin, AL | 41 |
| 1.51 | 0.487 | HC3 errors | 0.996 | Scott, IA | 107 |
| 1.87 | 1.45 | HC3 errors | 1.66 | Montgomery, OH | 209 |
| 1.57 | 0.953 | HC3 errors | 1.26 | Volusia, FL | 143 |
| 1.43 | 0.939 | Clustered on counties | 1.19 | Denver, CO | 635 |
| 1.38 | 0.681 | Clustered on counties | 1.03 | Polk, IA | 154 |
| 1.40 | 1.20 | Clustered on counties | 1.30 | Wayne, MI | 1,238 |
| 1.51 | 0.670 | HC3 errors | 1.09 | St. Louis, MN | 93 |
| 1.75 | 0.308 | Clustered on counties | 1.03 | Durham, NC | 131 |
| 1.79 | −0.045 | HC3 errors | 0.871 | Monroe, PA | 46 |
| 2.17 | 1.36 | HC3 errors | 1.77 | Elkhart, IN | 44 |
| 0.851 | 0.416 | HC3 errors | 0.633 | El Paso, TX | 175 |
| 2.08 | 0.935 | HC3 errors | 1.50 | Erie, PA | 65 |
| 1.91 | 0.820 | HC3 errors | 1.36 | Lane, OR | 86 |
| 2.13 | 1.46 | HC3 errors | 1.80 | Vanderburgh, IN | 88 |
| 2.06 | 1.28 | HC3 errors | 1.67 | Cass, ND | 58 |
| 1.65 | 0.765 | HC3 errors | 1.21 | Cumberland, NC | 107 |
| 0.985 | 0.401 | HC3 errors | 0.693 | Benton, AR | 109 |
| 1.57 | 1.07 | HC3 errors | 1.32 | Genesee, MI | 124 |
| 1.62 | 0.646 | HC3 errors | 1.13 | Florence, SC | 52 |
| 2.02 | 0.669 | HC3 errors | 1.34 | Lauderdale, AL | 40 |
| 2.56 | −0.146 | HC3 errors | 1.21 | Larimer, CO | 76 |
| 1.79 | 0.738 | HC3 errors | 1.27 | Sebastian, AR | 63 |
| 1.96 | 1.20 | HC3 errors | 1.58 | Allen, IN | 100 |
| 1.26 | 0.827 | HC3 errors | 1.04 | Fresno, CA | 218 |
| 1.77 | 1.21 | HC3 errors | 1.49 | Alachua, FL | 60 |
| 1.19 | 0.487 | HC3 errors | 0.840 | Hall, GA | 48 |
| 1.26 | 0.142 | HC3 errors | 0.701 | Warren, NY | 42 |
| 1.24 | 0.877 | HC3 errors | 1.06 | Kent, MI | 228 |
| 1.53 | 0.618 | HC3 errors | 1.07 | Weld, CO | 69 |
| 1.57 | 0.572 | HC3 errors | 1.07 | Brown, WI | 69 |
| 1.64 | 1.20 | HC3 errors | 1.42 | Guilford, NC | 169 |
| 1.36 | 0.988 | HC3 errors | 1.18 | Greenville, SC | 221 |
| 1.26 | 0.390 | HC3 errors | 0.827 | Harrison, MS | 137 |
| 1.98 | 1.13 | HC3 errors | 1.56 | Washington, MD | 65 |
| 1.55 | 0.981 | HC3 errors | 1.27 | Dauphin, PA | 126 |
| 1.29 | 0.997 | HC3 errors | 1.15 | Capitol, CT | 272 |
| 1.66 | 0.607 | HC3 errors | 1.13 | Forrest, MS | 45 |
| 1.36 | 0.395 | HC3 errors | 0.879 | Catawba, NC | 78 |
| 1.52 | 0.392 | HC3 errors | 0.955 | Hawaii, HI | 43 |
| 1.36 | 0.503 | HC3 errors | 0.930 | Beaufort, SC | 42 |
| 1.51 | 0.360 | HC3 errors | 0.936 | Citrus, FL | 41 |
| 1.13 | 0.623 | HC3 errors | 0.874 | Terrebonne, LA | 59 |
| 0.892 | 0.758 | Clustered on counties | 0.825 | Harris, TX | 1,457 |
| 2.12 | 0.348 | Clustered on counties | 1.24 | Cabell, WV | 102 |
| 1.53 | 1.08 | HC3 errors | 1.30 | Madison, AL | 108 |
| 1.48 | 1.18 | Clustered on counties | 1.33 | Marion, IN | 453 |
| 1.59 | 0.656 | HC3 errors | 1.12 | Jackson, MI | 45 |
| 1.50 | 1.00 | Clustered on counties | 1.25 | Hinds, MS | 149 |
| 1.83 | 0.819 | HC3 errors | 1.32 | Madison, TN | 50 |
| 1.53 | 1.23 | Clustered on counties | 1.38 | Duval, FL | 319 |
| 1.47 | 0.562 | HC3 errors | 1.02 | Washington, TN | 53 |
| 2.26 | 0.816 | HC3 errors | 1.54 | Jasper, MO | 42 |
| 2.53 | 0.560 | HC3 errors | 1.54 | Maui, HI | 42 |
| 1.55 | 0.897 | HC3 errors | 1.22 | Kalamazoo, MI | 63 |
| 1.50 | 1.05 | Clustered on counties | 1.27 | Jackson, MO | 556 |
| 1.21 | 0.656 | HC3 errors | 0.934 | Benton, WA | 63 |
| 1.00 | 0.383 | HC3 errors | 0.694 | Bell, TX | 91 |
| 2.04 | 0.991 | Clustered on counties | 1.51 | Sullivan, TN | 77 |
| 1.88 | 0.315 | HC3 errors | 1.10 | Ulster, NY | 41 |
| 1.59 | 1.29 | Clustered on counties | 1.44 | Knox, TN | 198 |
| 1.44 | 0.826 | Clustered on counties | 1.13 | Lafayette, LA | 125 |
| 1.98 | 0.909 | HC3 errors | 1.44 | Tippecanoe, IN | 49 |
| 1.48 | 0.741 | HC3 errors | 1.11 | Calcasieu, LA | 52 |
| 1.29 | 0.490 | HC3 errors | 0.892 | Mohave, AZ | 55 |
| 1.51 | 0.881 | HC3 errors | 1.20 | Polk, FL | 162 |
| 1.57 | 0.750 | HC3 errors | 1.16 | Lancaster, PA | 112 |
| 1.33 | 0.912 | HC3 errors | 1.12 | Ingham, MI | 135 |
| 1.34 | 0.641 | HC3 errors | 0.991 | Webb, TX | 61 |
| 0.970 | −0.013 | HC3 errors | 0.478 | Doña Ana, NM | 57 |
| 1.03 | 0.735 | HC3 errors | 0.885 | Clark, NV | 475 |
| 1.53 | 0.543 | HC3 errors | 1.03 | Grafton, NH | 66 |
| 1.84 | 0.989 | Clustered on counties | 1.41 | Fayette, KY | 124 |
| 1.65 | 0.994 | HC3 errors | 1.32 | Lancaster, NE | 74 |
| 1.82 | 0.949 | Clustered on counties | 1.38 | Pulaski, AR | 177 |
| 1.98 | 0.429 | HC3 errors | 1.20 | Laurel, KY | 43 |
| 1.02 | 0.470 | HC3 errors | 0.744 | Gregg, TX | 64 |
| 1.45 | 1.15 | HC3 errors | 1.30 | Los Angeles, CA | 2,923 |
| 1.73 | 1.29 | Clustered on counties | 1.51 | Jefferson, KY | 320 |
| 1.25 | 0.097 | HC3 errors | 0.675 | Lubbock, TX | 97 |
| 1.90 | 0.911 | Clustered on counties | 1.40 | Bedford, VA | 64 |
| 1.68 | 1.12 | Clustered on counties | 1.40 | Bibb, GA | 64 |
| 1.15 | 0.419 | HC3 errors | 0.782 | Dane, WI | 123 |
| 1.55 | 0.681 | HC3 errors | 1.12 | Hillsborough, NH | 99 |
| 0.888 | 0.374 | HC3 errors | 0.631 | Hidalgo, TX | 208 |
| 1.83 | 0.570 | HC3 errors | 1.20 | Jackson, OR | 51 |
| 1.45 | 1.12 | Clustered on counties | 1.29 | Shelby, TN | 316 |
| 1.32 | 0.358 | HC3 errors | 0.839 | Merced, CA | 60 |
| 0.922 | 0.725 | HC3 errors | 0.823 | Miami-Dade, FL | 1,377 |
| 1.32 | 1.08 | HC3 errors | 1.20 | Milwaukee, WI | 374 |
| 1.52 | 0.938 | Clustered on counties | 1.23 | Hennepin, MN | 847 |
| 1.61 | 1.02 | HC3 errors | 1.31 | Mobile, AL | 128 |
| 1.35 | 0.694 | HC3 errors | 1.02 | Stanislaus, CA | 111 |
| 1.93 | 1.04 | HC3 errors | 1.49 | Ouachita, LA | 57 |
| 2.14 | 1.11 | HC3 errors | 1.62 | Monroe, MI | 40 |
| 1.49 | 0.964 | HC3 errors | 1.23 | Montgomery, AL | 94 |
| 1.98 | 0.493 | HC3 errors | 1.24 | Skagit, WA | 41 |
| 1.92 | 1.19 | HC3 errors | 1.56 | Muskegon, MI | 44 |
| 1.15 | 0.593 | HC3 errors | 0.872 | Horry, SC | 110 |
| 1.38 | 0.423 | HC3 errors | 0.902 | Collier, FL | 94 |
| 1.55 | 0.985 | Clustered on counties | 1.27 | Davidson, TN | 384 |
| 1.31 | 0.818 | HC3 errors | 1.06 | South Central Connecticut, CT | 128 |
| 1.38 | 0.948 | Clustered on counties | 1.16 | Jefferson, LA | 379 |
| 1.12 | 0.421 | Clustered on counties | 0.771 | Kings, NY | 4,563 |
| 1.92 | 1.12 | HC3 errors | 1.52 | Berrien, MI | 51 |
| 1.19 | 0.714 | HC3 errors | 0.953 | Sarasota, FL | 204 |
| 1.66 | 0.818 | HC3 errors | 1.24 | Southeastern Connecticut, CT | 62 |
| 1.27 | 0.541 | HC3 errors | 0.906 | Marion, FL | 77 |
| 1.96 | 1.32 | HC3 errors | 1.64 | Davis, UT | 134 |
| 1.49 | 1.11 | Clustered on counties | 1.30 | Oklahoma, OK | 380 |
| 2.18 | 1.21 | HC3 errors | 1.69 | Thurston, WA | 60 |
| 1.31 | 0.980 | Clustered on counties | 1.14 | Douglas, NE | 247 |
| 1.03 | 0.689 | HC3 errors | 0.859 | Orange, FL | 460 |
| 1.49 | 0.593 | HC3 errors | 1.04 | Ventura, CA | 182 |
| 1.47 | 1.01 | HC3 errors | 1.24 | Brevard, FL | 140 |
| 1.23 | 0.379 | HC3 errors | 0.803 | Bay, FL | 46 |
| 1.73 | 1.17 | HC3 errors | 1.45 | Escambia, FL | 106 |
| 1.41 | 1.07 | Clustered on counties | 1.24 | Peoria, IL | 104 |
| 1.27 | 1.13 | Clustered on counties | 1.20 | Philadelphia, PA | 1,455 |
| 0.989 | 0.800 | HC3 errors | 0.894 | Maricopa, AZ | 948 |
| 1.21 | 0.893 | Clustered on counties | 1.05 | Allegheny, PA | 674 |
| 1.45 | 0.677 | HC3 errors | 1.06 | Cumberland, ME | 127 |
| 1.51 | 1.21 | Clustered on counties | 1.36 | Multnomah, OR | 537 |
| 1.67 | 1.00 | HC3 errors | 1.34 | St. Lucie, FL | 102 |
| 2.22 | 1.04 | HC3 errors | 1.63 | Schuylkill, PA | 40 |
| 1.02 | 0.313 | HC3 errors | 0.666 | Orange, NY | 162 |
| 1.26 | 0.223 | HC3 errors | 0.740 | Yavapai, AZ | 70 |
| 1.46 | 1.07 | Clustered on counties | 1.26 | Providence, RI | 357 |
| 1.53 | 0.768 | HC3 errors | 1.15 | Utah, UT | 140 |
| 1.79 | 1.14 | HC3 errors | 1.47 | Pueblo, CO | 55 |
| 1.58 | 0.643 | HC3 errors | 1.11 | Charlotte, FL | 40 |
| 1.42 | 0.528 | HC3 errors | 0.976 | Racine, WI | 41 |
| 1.04 | 0.745 | HC3 errors | 0.890 | Wake, NC | 268 |
| 1.39 | 0.967 | HC3 errors | 1.18 | Berks, PA | 103 |
| 1.73 | 0.732 | HC3 errors | 1.23 | Shasta, CA | 49 |
| 1.67 | 0.812 | HC3 errors | 1.24 | Washoe, NV | 117 |
| 1.36 | 1.12 | Clustered on counties | 1.24 | Chesterfield, VA | 298 |
| 1.15 | 0.890 | HC3 errors | 1.02 | Riverside, CA | 939 |
| 1.82 | 0.799 | Clustered on counties | 1.31 | Roanoke, VA | 75 |
| 1.41 | 0.373 | HC3 errors | 0.892 | Olmsted, MN | 52 |
| 1.39 | 1.05 | Clustered on counties | 1.22 | Monroe, NY | 288 |
| 1.91 | 1.01 | HC3 errors | 1.46 | Winnebago, IL | 88 |
| 1.61 | 1.27 | HC3 errors | 1.44 | Sacramento, CA | 519 |
| 2.09 | 1.35 | HC3 errors | 1.72 | Saginaw, MI | 59 |
| 1.86 | 0.503 | HC3 errors | 1.18 | Stearns, MN | 43 |
| 1.43 | 1.07 | Clustered on counties | 1.25 | St. Louis, MO | 669 |
| 2.15 | 0.667 | HC3 errors | 1.41 | Marion, OR | 78 |
| 1.58 | 0.320 | HC3 errors | 0.950 | Monterey, CA | 96 |
| 1.21 | 0.759 | HC3 errors | 0.986 | Sussex, DE | 105 |
| 1.51 | 0.894 | HC3 errors | 1.20 | Salt Lake, UT | 247 |
| 1.05 | 0.647 | Clustered on counties | 0.849 | Bexar, TX | 453 |
| 1.34 | 0.826 | HC3 errors | 1.08 | San Diego, CA | 681 |
| 3.04 | 1.67 | Clustered on counties | 2.36 | Alameda, CA | 1,044 |
| 1.09 | 0.161 | HC3 errors | 0.624 | Santa Clara, CA | 405 |
| 1.48 | −0.040 | HC3 errors | 0.721 | San Luis Obispo, CA | 61 |
| 1.73 | −0.548 | HC3 errors | 0.592 | Santa Cruz, CA | 63 |
| 1.03 | 0.374 | HC3 errors | 0.703 | Santa Fe, NM | 52 |
| 1.64 | 0.787 | HC3 errors | 1.22 | Santa Barbara, CA | 98 |
| 2.12 | 0.547 | HC3 errors | 1.34 | Sonoma, CA | 116 |
| 1.28 | 0.806 | HC3 errors | 1.04 | Chatham, GA | 101 |
| 1.64 | 1.08 | HC3 errors | 1.36 | Luzerne, PA | 162 |
| 1.79 | 1.34 | HC3 errors | 1.56 | King, WA | 829 |
| 1.27 | 0.555 | HC3 errors | 0.914 | Indian River, FL | 41 |
| 1.51 | 1.01 | HC3 errors | 1.26 | Caddo, LA | 114 |
| 1.87 | 0.709 | HC3 errors | 1.29 | Minnehaha, SD | 60 |
| 1.51 | 1.05 | HC3 errors | 1.28 | St. Joseph, IN | 92 |
| 1.63 | 0.905 | HC3 errors | 1.27 | Spartanburg, SC | 85 |
| 1.84 | 1.26 | HC3 errors | 1.55 | Spokane, WA | 140 |
| 2.00 | 1.14 | HC3 errors | 1.57 | Sangamon, IL | 56 |
| 1.67 | 1.27 | HC3 errors | 1.47 | Hampden, MA | 151 |
| 2.08 | 0.848 | Clustered on counties | 1.46 | Greene, MO | 111 |
| 2.70 | 1.57 | HC3 errors | 2.13 | Clark, OH | 41 |
| 1.76 | 1.15 | HC3 errors | 1.46 | San Joaquin, CA | 169 |
| 1.43 | 0.879 | HC3 errors | 1.15 | Onondaga, NY | 183 |
| 1.22 | 0.803 | HC3 errors | 1.01 | Leon, FL | 82 |
| 1.10 | 0.888 | HC3 errors | 0.995 | Hillsborough, FL | 692 |
| 2.40 | 1.28 | Clustered on counties | 1.84 | Vigo, IN | 53 |
| 1.64 | 0.254 | HC3 errors | 0.949 | Bowie, TX | 42 |
| 1.42 | 1.05 | HC3 errors | 1.23 | Lucas, OH | 213 |
| 1.32 | 1.12 | Clustered on counties | 1.22 | Shawnee, KS | 59 |
| 1.42 | 0.785 | HC3 errors | 1.10 | Grand Traverse, MI | 43 |
| 1.58 | 1.11 | HC3 errors | 1.34 | Mercer, NJ | 79 |
| 1.30 | 0.873 | HC3 errors | 1.08 | Pima, AZ | 246 |
| 1.48 | 0.994 | Clustered on counties | 1.24 | Tulsa, OK | 304 |
| 1.84 | 0.561 | HC3 errors | 1.20 | Lee, MS | 40 |
| 1.28 | 0.561 | HC3 errors | 0.920 | Tuscaloosa, AL | 69 |
| 1.45 | 0.295 | HC3 errors | 0.875 | Smith, TX | 53 |
| 1.57 | 0.764 | HC3 errors | 1.16 | Honolulu, HI | 286 |
| 2.16 | 1.08 | HC3 errors | 1.62 | Oneida, NY | 83 |
| 2.86 | 1.47 | HC3 errors | 2.16 | Solano, CA | 96 |
| 1.68 | 1.04 | Clustered on counties | 1.36 | Virginia Beach, VA | 436 |
| 1.44 | 0.731 | HC3 errors | 1.08 | Tulare, CA | 101 |
| 1.71 | 0.963 | HC3 errors | 1.34 | McLennan, TX | 55 |
| 1.37 | 0.235 | HC3 errors | 0.804 | Houston, GA | 44 |
| 1.34 | 0.861 | Clustered on counties | 1.10 | Fairfax, VA | 1,350 |
| 1.34 | 0.879 | HC3 errors | 1.11 | Naugatuck Valley, CT | 95 |
| 1.67 | 0.512 | HC3 errors | 1.09 | Black Hawk, IA | 46 |
| 2.51 | 0.881 | HC3 errors | 1.69 | Marathon, WI | 41 |
| 2.19 | 1.04 | HC3 errors | 1.61 | Belmont, OH | 48 |
| 1.51 | 1.07 | HC3 errors | 1.29 | Sedgwick, KS | 164 |
| 2.15 | 0.833 | HC3 errors | 1.49 | Wichita, TX | 44 |
| 1.27 | 0.688 | HC3 errors | 0.978 | New Hanover, NC | 61 |
| 1.59 | 0.891 | Clustered on counties | 1.24 | Forsyth, NC | 161 |
| 1.24 | 0.783 | HC3 errors | 1.01 | Worcester, MA | 184 |
| 1.41 | 0.812 | HC3 errors | 1.11 | Yakima, WA | 56 |
| 1.75 | 1.10 | HC3 errors | 1.43 | York, PA | 105 |
| 2.14 | 1.27 | HC3 errors | 1.70 | Mahoning, OH | 155 |
| 0.485 | −0.212 | HC3 errors | 0.137 | Yuma, AZ | 60 |

#### Table 6.2. Distribution

Metro slopes cluster around a median of 1.20 years per ten-point step, with few metros at either end

Number of metros by slope of life expectancy on the ADI, years lower per ten-point step, in bins of equal width

| metros | series | x0 | x1 |
| ---: | --- | ---: | ---: |
| 1 | Estimated slope | 0 | 0.200 |
| 0 | Estimated slope | 0.200 | 0.400 |
| 3 | Estimated slope | 0.400 | 0.600 |
| 17 | Estimated slope | 0.600 | 0.800 |
| 48 | Estimated slope | 0.800 | 1 |
| 63 | Estimated slope | 1 | 1.20 |
| 76 | Estimated slope | 1.20 | 1.40 |
| 38 | Estimated slope | 1.40 | 1.60 |
| 14 | Estimated slope | 1.60 | 1.80 |
| 3 | Estimated slope | 1.80 | 2 |
| 2 | Estimated slope | 2 | 2.20 |
| 2 | Estimated slope | 2.20 | 2.40 |
| 0 | Estimated slope | 2.40 | 2.60 |
| 0 | Estimated slope | 2.60 | 2.80 |

Source: USALEEP life expectancy, 2010-2015; Area Deprivation Index, 2023; ACS 2019-2023; population-weighted regression slopes, modeled estimates

Data: https://vishalsingh.org/zipcode/report/place-and-the-length-of-a-life/figures/metro_slopes.json

### Table 7. Life expectancy at birth is steadily worse toward the more deprived end under the ADI and under each of the three other rankings

Life expectancy at birth (years) by tenth of tracts ranked four ways, over the 71,771 tracts that have all four rankings

| bin | conf_high | conf_low | estimate | n_units | ranking |
| ---: | ---: | ---: | ---: | ---: | --- |
| 1 | 82.56 | 82.11 | 82.33 | 7,179 | Area Deprivation Index |
| 2 | 81.18 | 80.73 | 80.93 | 7,168 | Area Deprivation Index |
| 3 | 80.50 | 80.11 | 80.30 | 7,220 | Area Deprivation Index |
| 4 | 79.87 | 79.54 | 79.71 | 7,108 | Area Deprivation Index |
| 5 | 79.25 | 78.93 | 79.09 | 7,229 | Area Deprivation Index |
| 6 | 78.54 | 78.25 | 78.40 | 7,042 | Area Deprivation Index |
| 7 | 77.83 | 77.56 | 77.70 | 7,298 | Area Deprivation Index |
| 8 | 77.05 | 76.77 | 76.90 | 7,109 | Area Deprivation Index |
| 9 | 76.04 | 75.76 | 75.90 | 7,212 | Area Deprivation Index |
| 10 | 74.34 | 73.79 | 74.04 | 7,206 | Area Deprivation Index |
| 1 | 82.54 | 82.16 | 82.37 | 7,177 | Standardized index |
| 2 | 81.23 | 80.97 | 81.10 | 7,177 | Standardized index |
| 3 | 80.45 | 80.20 | 80.32 | 7,177 | Standardized index |
| 4 | 79.73 | 79.48 | 79.61 | 7,177 | Standardized index |
| 5 | 79.12 | 78.84 | 78.97 | 7,177 | Standardized index |
| 6 | 78.43 | 78.09 | 78.26 | 7,177 | Standardized index |
| 7 | 77.84 | 77.46 | 77.65 | 7,177 | Standardized index |
| 8 | 77.15 | 76.65 | 76.89 | 7,177 | Standardized index |
| 9 | 76.32 | 75.72 | 76.00 | 7,177 | Standardized index |
| 10 | 75.14 | 74.24 | 74.69 | 7,178 | Standardized index |
| 1 | 82.43 | 82.00 | 82.23 | 7,177 | Median household income |
| 2 | 81.08 | 80.82 | 80.95 | 7,176 | Median household income |
| 3 | 80.39 | 80.13 | 80.26 | 7,177 | Median household income |
| 4 | 79.77 | 79.49 | 79.63 | 7,176 | Median household income |
| 5 | 79.12 | 78.85 | 78.99 | 7,179 | Median household income |
| 6 | 78.53 | 78.23 | 78.38 | 7,178 | Median household income |
| 7 | 77.84 | 77.56 | 77.70 | 7,176 | Median household income |
| 8 | 77.16 | 76.83 | 76.99 | 7,177 | Median household income |
| 9 | 76.01 | 75.62 | 75.81 | 7,177 | Median household income |
| 10 | 74.44 | 73.92 | 74.18 | 7,178 | Median household income |
| 1 | 82.51 | 82.02 | 82.27 | 7,178 | Bachelor's degree share |
| 2 | 81.27 | 80.92 | 81.09 | 7,181 | Bachelor's degree share |
| 3 | 80.47 | 80.15 | 80.31 | 7,166 | Bachelor's degree share |
| 4 | 79.73 | 79.42 | 79.57 | 7,182 | Bachelor's degree share |
| 5 | 78.99 | 78.68 | 78.84 | 7,185 | Bachelor's degree share |
| 6 | 78.40 | 78.05 | 78.22 | 7,171 | Bachelor's degree share |
| 7 | 77.82 | 77.45 | 77.63 | 7,175 | Bachelor's degree share |
| 8 | 77.19 | 76.81 | 77.00 | 7,178 | Bachelor's degree share |
| 9 | 76.49 | 76.01 | 76.24 | 7,180 | Bachelor's degree share |
| 10 | 75.71 | 74.93 | 75.29 | 7,175 | Bachelor's degree share |

Note: Least and most deprived tenth: ADI 82.3 and 74.0, rank correlation −1.00; standardized index 82.4 and 74.7, −1.00; median household income 82.2 and 74.2, −1.00; bachelor's degree share 82.3 and 75.3, −1.00.

Note: The standardized index applies the ADI's coefficients to its measures after standardizing each one, so home value and income carry less weight than in the ADI. Income and degree share are single census measures, each ranked from the highest value to the lowest.

Source: USALEEP life expectancy at birth, 2010-2015, and Area Deprivation Index, 2023; population-weighted (ACS 2019-2023 tract population), modeled estimates; ADI 2023 and ACS 2019-2023 tract tables (B19013, B15003 and the 15 rebuilt ADI measures)

Data: https://vishalsingh.org/zipcode/report/place-and-the-length-of-a-life/figures/alt_rankings.json

### Table 8. In all 49 states whose tracts reach across at least 4 national tenths, life expectancy is lower in the state's most deprived tenth than in its least deprived tenth

Life expectancy at birth by national tenth of tracts by ADI, within each state, population-weighted, deaths in 2010 to 2015; the faint line is all states together

| bin | conf_high | conf_low | estimate | n_units | state | suppressed |
| ---: | ---: | ---: | ---: | ---: | --- | --- |
| 1 | 82.56 | 82.11 | 82.33 | 7,179 | US | no |
| 2 | 81.18 | 80.73 | 80.93 | 7,168 | US | no |
| 3 | 80.50 | 80.11 | 80.30 | 7,220 | US | no |
| 4 | 79.87 | 79.54 | 79.71 | 7,108 | US | no |
| 5 | 79.25 | 78.93 | 79.09 | 7,229 | US | no |
| 6 | 78.54 | 78.25 | 78.40 | 7,042 | US | no |
| 7 | 77.83 | 77.56 | 77.70 | 7,298 | US | no |
| 8 | 77.05 | 76.77 | 76.90 | 7,109 | US | no |
| 9 | 76.04 | 75.76 | 75.90 | 7,212 | US | no |
| 10 | 74.34 | 73.79 | 74.04 | 7,206 | US | no |
| 1 | n/a | n/a | n/a | 1 | AK | yes |
| 2 | n/a | n/a | n/a | 10 | AK | yes |
| 3 | n/a | n/a | n/a | 18 | AK | yes |
| 4 | 81.01 | 79.04 | 79.57 | 36 | AK | no |
| 5 | 80.76 | 77.98 | 79.21 | 44 | AK | no |
| 6 | n/a | n/a | n/a | 18 | AK | yes |
| 7 | n/a | n/a | n/a | 9 | AK | yes |
| 8 | n/a | n/a | n/a | 7 | AK | yes |
| 9 | n/a | n/a | n/a | 2 | AK | yes |
| 10 | n/a | n/a | n/a | 8 | AK | yes |
| 1 | n/a | n/a | n/a | 7 | AL | yes |
| 2 | n/a | n/a | n/a | 13 | AL | yes |
| 3 | 81.45 | 80.28 | 80.93 | 34 | AL | no |
| 4 | 80.01 | 78.94 | 79.60 | 50 | AL | no |
| 5 | 78.81 | 77.59 | 78.27 | 83 | AL | no |
| 6 | 77.71 | 76.26 | 76.97 | 106 | AL | no |
| 7 | 76.92 | 75.32 | 76.12 | 149 | AL | no |
| 8 | 75.76 | 74.96 | 75.35 | 181 | AL | no |
| 9 | 74.74 | 73.87 | 74.29 | 254 | AL | no |
| 10 | 73.36 | 72.04 | 72.61 | 320 | AL | no |
| 1 | n/a | n/a | n/a | 1 | AR | yes |
| 2 | n/a | n/a | n/a | 4 | AR | yes |
| 3 | n/a | n/a | n/a | 15 | AR | yes |
| 4 | n/a | n/a | n/a | 17 | AR | yes |
| 5 | 79.27 | 77.41 | 78.50 | 43 | AR | no |
| 6 | 78.94 | 77 | 77.99 | 46 | AR | no |
| 7 | 77.32 | 76.41 | 76.87 | 84 | AR | no |
| 8 | 76.52 | 75.65 | 76.07 | 125 | AR | no |
| 9 | 75.86 | 74.85 | 75.32 | 165 | AR | no |
| 10 | 74.62 | 73.38 | 73.91 | 241 | AR | no |
| 1 | 87.50 | 82.79 | 82.90 | 60 | AZ | no |
| 2 | 82.55 | 80.51 | 80.77 | 171 | AZ | no |
| 3 | 81.91 | 79.60 | 80.08 | 205 | AZ | no |
| 4 | 81.01 | 78.80 | 79.30 | 220 | AZ | no |
| 5 | 79.90 | 77.36 | 77.92 | 197 | AZ | no |
| 6 | 78.46 | 76.88 | 77.21 | 157 | AZ | no |
| 7 | 79.09 | 76.89 | 77.43 | 134 | AZ | no |
| 8 | 77.68 | 75.94 | 76.59 | 105 | AZ | no |
| 9 | 76.48 | 74.69 | 75.42 | 82 | AZ | no |
| 10 | 77.48 | 74.77 | 76.04 | 71 | AZ | no |
| 1 | 82.49 | 82.04 | 82.24 | 3,373 | CA | no |
| 2 | 80.65 | 80.15 | 80.33 | 2,033 | CA | no |
| 3 | 79.57 | 78.81 | 79.09 | 1,064 | CA | no |
| 4 | 78.81 | 77.83 | 78.22 | 619 | CA | no |
| 5 | 78.10 | 76.73 | 77.31 | 387 | CA | no |
| 6 | 77.28 | 75.98 | 76.67 | 256 | CA | no |
| 7 | 76.97 | 75.78 | 76.37 | 187 | CA | no |
| 8 | 76.95 | 75.08 | 75.83 | 100 | CA | no |
| 9 | 77.19 | 74.04 | 75.19 | 40 | CA | no |
| 10 | 77.75 | 74.74 | 76.42 | 33 | CA | no |
| 1 | 82.90 | 81.64 | 82.18 | 182 | CO | no |
| 2 | 81.84 | 80.19 | 81 | 304 | CO | no |
| 3 | 80.51 | 78.92 | 79.57 | 258 | CO | no |
| 4 | 79.43 | 78 | 78.64 | 183 | CO | no |
| 5 | 79.13 | 77.34 | 77.96 | 97 | CO | no |
| 6 | 79.15 | 75.72 | 76.96 | 52 | CO | no |
| 7 | 78.86 | 76.27 | 77.31 | 50 | CO | no |
| 8 | 78.56 | 75.81 | 76.96 | 27 | CO | no |
| 9 | 79.04 | 74.10 | 75.82 | 29 | CO | no |
| 10 | n/a | n/a | n/a | 13 | CO | yes |
| 1 | n/a | n/a | 84.03 | 62 | CT | no |
| 2 | n/a | n/a | 82.80 | 64 | CT | no |
| 3 | n/a | n/a | 82 | 117 | CT | no |
| 4 | n/a | n/a | 80.81 | 122 | CT | no |
| 5 | n/a | n/a | 80.19 | 122 | CT | no |
| 6 | n/a | n/a | 79.04 | 93 | CT | no |
| 7 | n/a | n/a | 77.57 | 78 | CT | no |
| 8 | n/a | n/a | 76.60 | 70 | CT | no |
| 9 | n/a | n/a | 76.22 | 24 | CT | no |
| 10 | n/a | n/a | n/a | 5 | CT | yes |
| 1 | n/a | n/a | 79.57 | 81 | DC | no |
| 2 | n/a | n/a | 77.47 | 41 | DC | no |
| 3 | n/a | n/a | n/a | 18 | DC | yes |
| 4 | n/a | n/a | 71.99 | 20 | DC | no |
| 5 | n/a | n/a | n/a | 5 | DC | yes |
| 6 | n/a | n/a | n/a | 1 | DC | yes |
| 7 | n/a | n/a | n/a | 0 | DC | yes |
| 8 | n/a | n/a | n/a | 0 | DC | yes |
| 9 | n/a | n/a | n/a | 0 | DC | yes |
| 10 | n/a | n/a | n/a | 0 | DC | yes |
| 1 | n/a | n/a | n/a | 5 | DE | yes |
| 2 | n/a | n/a | n/a | 10 | DE | yes |
| 3 | n/a | n/a | 80.41 | 28 | DE | no |
| 4 | n/a | n/a | 79.06 | 33 | DE | no |
| 5 | n/a | n/a | 77.84 | 46 | DE | no |
| 6 | n/a | n/a | 77.45 | 32 | DE | no |
| 7 | n/a | n/a | 76.45 | 35 | DE | no |
| 8 | n/a | n/a | n/a | 11 | DE | yes |
| 9 | n/a | n/a | n/a | 4 | DE | yes |
| 10 | n/a | n/a | n/a | 5 | DE | yes |
| 1 | 83.56 | 81.85 | 82.87 | 162 | FL | no |
| 2 | 82.19 | 81.28 | 81.80 | 352 | FL | no |
| 3 | 81.52 | 80.65 | 81.08 | 519 | FL | no |
| 4 | 80.38 | 79.63 | 80.04 | 655 | FL | no |
| 5 | 79.28 | 78.62 | 78.95 | 644 | FL | no |
| 6 | 78.65 | 77.72 | 78.18 | 592 | FL | no |
| 7 | 77.59 | 76.59 | 77.07 | 498 | FL | no |
| 8 | 76.62 | 75.26 | 75.93 | 352 | FL | no |
| 9 | 76.03 | 74.43 | 75.22 | 278 | FL | no |
| 10 | 75.36 | 72.97 | 74.24 | 205 | FL | no |
| 1 | 82.33 | 80.57 | 81.45 | 52 | GA | no |
| 2 | 81.23 | 80.48 | 80.91 | 133 | GA | no |
| 3 | 80.53 | 79.44 | 79.92 | 189 | GA | no |
| 4 | 79.64 | 78.59 | 79.21 | 227 | GA | no |
| 5 | 78.27 | 77.21 | 77.79 | 288 | GA | no |
| 6 | 77.48 | 76.62 | 77.03 | 306 | GA | no |
| 7 | 76.66 | 76.01 | 76.32 | 293 | GA | no |
| 8 | 76.13 | 75.33 | 75.72 | 261 | GA | no |
| 9 | 75.18 | 74.45 | 74.81 | 275 | GA | no |
| 10 | 74.13 | 73.13 | 73.58 | 260 | GA | no |
| 1 | n/a | n/a | 82.37 | 213 | HI | no |
| 2 | n/a | n/a | 80.52 | 75 | HI | no |
| 3 | n/a | n/a | 81.22 | 43 | HI | no |
| 4 | n/a | n/a | 79.69 | 23 | HI | no |
| 5 | n/a | n/a | n/a | 6 | HI | yes |
| 6 | n/a | n/a | n/a | 2 | HI | yes |
| 7 | n/a | n/a | n/a | 5 | HI | yes |
| 8 | n/a | n/a | n/a | 0 | HI | yes |
| 9 | n/a | n/a | n/a | 1 | HI | yes |
| 10 | n/a | n/a | n/a | 0 | HI | yes |
| 1 | n/a | n/a | n/a | 0 | IA | yes |
| 2 | n/a | n/a | n/a | 3 | IA | yes |
| 3 | n/a | n/a | n/a | 17 | IA | yes |
| 4 | 82.42 | 80.91 | 81.53 | 46 | IA | no |
| 5 | 82.85 | 80.67 | 81.56 | 69 | IA | no |
| 6 | 82.33 | 80.78 | 81.46 | 98 | IA | no |
| 7 | 80.94 | 79.85 | 80.38 | 123 | IA | no |
| 8 | 79.40 | 78.19 | 78.78 | 175 | IA | no |
| 9 | 78.20 | 77.10 | 77.61 | 199 | IA | no |
| 10 | 76.99 | 75.46 | 76.14 | 104 | IA | no |
| 1 | n/a | n/a | n/a | 18 | ID | yes |
| 2 | 82.38 | 80.27 | 81.38 | 42 | ID | no |
| 3 | 80.86 | 79.43 | 80.52 | 83 | ID | no |
| 4 | 80.17 | 78.64 | 79.56 | 83 | ID | no |
| 5 | 80.23 | 78.83 | 79.42 | 69 | ID | no |
| 6 | 79.38 | 77.87 | 78.47 | 53 | ID | no |
| 7 | 79.24 | 76.90 | 77.88 | 42 | ID | no |
| 8 | 77.86 | 75.07 | 76.12 | 24 | ID | no |
| 9 | n/a | n/a | n/a | 3 | ID | yes |
| 10 | n/a | n/a | n/a | 0 | ID | yes |
| 1 | 84.36 | 81.93 | 82.83 | 83 | IL | no |
| 2 | 82.95 | 81.38 | 81.84 | 163 | IL | no |
| 3 | 82.29 | 80.91 | 81.39 | 209 | IL | no |
| 4 | 81.17 | 80.11 | 80.37 | 296 | IL | no |
| 5 | 80.82 | 79.37 | 79.86 | 354 | IL | no |
| 6 | 80.04 | 78.38 | 78.86 | 308 | IL | no |
| 7 | 79.63 | 77.49 | 78.18 | 334 | IL | no |
| 8 | 78.90 | 76.64 | 77.66 | 339 | IL | no |
| 9 | 77.95 | 75.54 | 76.67 | 377 | IL | no |
| 10 | 75.03 | 74.03 | 74.46 | 375 | IL | no |
| 1 | n/a | n/a | n/a | 3 | IN | yes |
| 2 | n/a | n/a | n/a | 14 | IN | yes |
| 3 | 81.62 | 80.16 | 80.94 | 44 | IN | no |
| 4 | 80.88 | 78.48 | 79.77 | 51 | IN | no |
| 5 | 80.87 | 79.31 | 80.07 | 138 | IN | no |
| 6 | 79.91 | 78.91 | 79.38 | 184 | IN | no |
| 7 | 78.78 | 77.64 | 78.11 | 233 | IN | no |
| 8 | 78.09 | 76.88 | 77.45 | 288 | IN | no |
| 9 | 76.49 | 75.19 | 75.74 | 273 | IN | no |
| 10 | 73.51 | 72.04 | 72.76 | 290 | IN | no |
| 1 | n/a | n/a | n/a | 2 | KS | yes |
| 2 | n/a | n/a | n/a | 15 | KS | yes |
| 3 | 82.74 | 80.32 | 81.92 | 35 | KS | no |
| 4 | 81.66 | 80.63 | 81.06 | 50 | KS | no |
| 5 | 80.83 | 79.64 | 80.33 | 60 | KS | no |
| 6 | 80.57 | 78.87 | 79.56 | 61 | KS | no |
| 7 | 80.25 | 79.02 | 79.55 | 85 | KS | no |
| 8 | 78.68 | 77.45 | 78.02 | 97 | KS | no |
| 9 | 77.80 | 76.72 | 77.17 | 139 | KS | no |
| 10 | 76.77 | 74.63 | 75.45 | 193 | KS | no |
| 1 | n/a | n/a | n/a | 1 | KY | yes |
| 2 | n/a | n/a | n/a | 9 | KY | yes |
| 3 | 82.78 | 78.73 | 81.68 | 34 | KY | no |
| 4 | 80.54 | 78.39 | 79.79 | 51 | KY | no |
| 5 | 79.74 | 78.68 | 79.29 | 75 | KY | no |
| 6 | 78.42 | 77.56 | 77.95 | 114 | KY | no |
| 7 | 77.17 | 76.35 | 76.78 | 134 | KY | no |
| 8 | 75.95 | 75.30 | 75.60 | 185 | KY | no |
| 9 | 74.94 | 74.12 | 74.52 | 258 | KY | no |
| 10 | 73.59 | 72.49 | 73.04 | 286 | KY | no |
| 1 | n/a | n/a | n/a | 12 | LA | yes |
| 2 | n/a | n/a | n/a | 18 | LA | yes |
| 3 | 81.39 | 78.41 | 79.97 | 43 | LA | no |
| 4 | 80.15 | 78.59 | 79.46 | 55 | LA | no |
| 5 | 78.94 | 77.63 | 78.38 | 77 | LA | no |
| 6 | 78.29 | 76.99 | 77.61 | 104 | LA | no |
| 7 | 76.95 | 75.95 | 76.42 | 176 | LA | no |
| 8 | 75.97 | 74.99 | 75.44 | 170 | LA | no |
| 9 | 75.14 | 74.11 | 74.59 | 190 | LA | no |
| 10 | 73.82 | 72.59 | 73.17 | 223 | LA | no |
| 1 | 83.32 | 81.35 | 82.72 | 321 | MA | no |
| 2 | 81.58 | 80.34 | 81.01 | 343 | MA | no |
| 3 | 80.82 | 79.88 | 80.32 | 292 | MA | no |
| 4 | 80.63 | 78.74 | 79.61 | 182 | MA | no |
| 5 | 80.16 | 77.34 | 78.50 | 104 | MA | no |
| 6 | 79.66 | 77.14 | 78.34 | 71 | MA | no |
| 7 | 78.21 | 77.12 | 77.45 | 48 | MA | no |
| 8 | n/a | n/a | n/a | 19 | MA | yes |
| 9 | n/a | n/a | n/a | 8 | MA | yes |
| 10 | n/a | n/a | n/a | 1 | MA | yes |
| 1 | 84.55 | 81.55 | 83.91 | 110 | MD | no |
| 2 | 82.73 | 80.68 | 81.78 | 195 | MD | no |
| 3 | 80.90 | 79.81 | 80.41 | 263 | MD | no |
| 4 | 79.95 | 78.71 | 79.43 | 225 | MD | no |
| 5 | 78.87 | 77.45 | 78.15 | 172 | MD | no |
| 6 | 78.30 | 76.30 | 77.06 | 104 | MD | no |
| 7 | 76.86 | 73.95 | 75.48 | 59 | MD | no |
| 8 | 75.45 | 73.64 | 74.17 | 66 | MD | no |
| 9 | 76.06 | 71.74 | 73.34 | 35 | MD | no |
| 10 | 74.38 | 69.18 | 70.28 | 42 | MD | no |
| 1 | n/a | n/a | n/a | 4 | ME | yes |
| 2 | n/a | n/a | n/a | 19 | ME | yes |
| 3 | 81.52 | 80.06 | 80.52 | 31 | ME | no |
| 4 | 82.03 | 78.74 | 79.37 | 34 | ME | no |
| 5 | 79.75 | 78.25 | 78.89 | 41 | ME | no |
| 6 | 79.92 | 77.77 | 78.76 | 49 | ME | no |
| 7 | 78.66 | 77.04 | 77.78 | 39 | ME | no |
| 8 | 77.72 | 76.49 | 77 | 43 | ME | no |
| 9 | 77.98 | 75.74 | 77.03 | 44 | ME | no |
| 10 | n/a | n/a | n/a | 15 | ME | yes |
| 1 | n/a | n/a | n/a | 14 | MI | yes |
| 2 | 84.40 | 82.09 | 82.67 | 40 | MI | no |
| 3 | 81.92 | 81.10 | 81.46 | 109 | MI | no |
| 4 | 81.46 | 80.45 | 80.93 | 195 | MI | no |
| 5 | 80.80 | 79.44 | 80.02 | 225 | MI | no |
| 6 | 79.70 | 78.86 | 79.23 | 304 | MI | no |
| 7 | 78.89 | 77.83 | 78.26 | 359 | MI | no |
| 8 | 78.23 | 77.21 | 77.65 | 377 | MI | no |
| 9 | 77.16 | 75.56 | 76.28 | 391 | MI | no |
| 10 | 74.83 | 72.72 | 73.36 | 451 | MI | no |
| 1 | n/a | n/a | n/a | 14 | MN | yes |
| 2 | 83.49 | 81.98 | 83.10 | 71 | MN | no |
| 3 | 82.39 | 81.66 | 82.04 | 152 | MN | no |
| 4 | 82.21 | 81.36 | 81.71 | 230 | MN | no |
| 5 | 81.88 | 80.31 | 80.92 | 266 | MN | no |
| 6 | 80.72 | 78.97 | 79.71 | 201 | MN | no |
| 7 | 80.42 | 77.89 | 78.99 | 175 | MN | no |
| 8 | 80.30 | 78.89 | 79.55 | 142 | MN | no |
| 9 | 79.47 | 77.47 | 78.40 | 78 | MN | no |
| 10 | n/a | n/a | n/a | 17 | MN | yes |
| 1 | n/a | n/a | n/a | 5 | MO | yes |
| 2 | 83.04 | 81.12 | 81.69 | 26 | MO | no |
| 3 | 82.38 | 80.07 | 81.58 | 52 | MO | no |
| 4 | 81.60 | 80.05 | 80.98 | 93 | MO | no |
| 5 | 80.31 | 79.08 | 79.82 | 136 | MO | no |
| 6 | 79.09 | 77.97 | 78.62 | 165 | MO | no |
| 7 | 78.46 | 77.31 | 77.84 | 233 | MO | no |
| 8 | 77.31 | 76.47 | 76.87 | 249 | MO | no |
| 9 | 76.19 | 75.14 | 75.68 | 257 | MO | no |
| 10 | 74.27 | 72.35 | 73.22 | 258 | MO | no |
| 1 | n/a | n/a | n/a | 1 | MS | yes |
| 2 | n/a | n/a | n/a | 2 | MS | yes |
| 3 | n/a | n/a | n/a | 5 | MS | yes |
| 4 | n/a | n/a | n/a | 16 | MS | yes |
| 5 | 80.41 | 78.28 | 79.51 | 30 | MS | no |
| 6 | 79.08 | 76.57 | 77.71 | 58 | MS | no |
| 7 | 77.88 | 76.53 | 77.29 | 66 | MS | no |
| 8 | 76.58 | 75.30 | 75.98 | 105 | MS | no |
| 9 | 75.21 | 74.31 | 74.76 | 161 | MS | no |
| 10 | 74.41 | 73.46 | 73.92 | 277 | MS | no |
| 1 | n/a | n/a | n/a | 7 | MT | yes |
| 2 | 81.50 | 80.07 | 80.98 | 23 | MT | no |
| 3 | 81.11 | 79.91 | 80.39 | 34 | MT | no |
| 4 | 80.45 | 79.19 | 79.65 | 46 | MT | no |
| 5 | 80.20 | 78.37 | 79.18 | 50 | MT | no |
| 6 | 80.21 | 77.86 | 78.84 | 35 | MT | no |
| 7 | 79.32 | 76.27 | 77.68 | 32 | MT | no |
| 8 | 78.02 | 75.90 | 76.62 | 35 | MT | no |
| 9 | 76.97 | 73.49 | 75.02 | 23 | MT | no |
| 10 | n/a | n/a | n/a | 7 | MT | yes |
| 1 | 82.71 | 81.42 | 81.96 | 44 | NC | no |
| 2 | 81.93 | 80.85 | 81.47 | 107 | NC | no |
| 3 | 80.52 | 79.53 | 80.14 | 194 | NC | no |
| 4 | 80.39 | 79.63 | 80.05 | 198 | NC | no |
| 5 | 79.24 | 78.38 | 78.84 | 245 | NC | no |
| 6 | 78.55 | 77.71 | 78.14 | 257 | NC | no |
| 7 | 77.42 | 76.71 | 77.02 | 321 | NC | no |
| 8 | 76.65 | 75.93 | 76.30 | 362 | NC | no |
| 9 | 75.99 | 75.32 | 75.66 | 317 | NC | no |
| 10 | 74.62 | 73.58 | 74.08 | 243 | NC | no |
| 1 | n/a | n/a | n/a | 0 | ND | yes |
| 2 | n/a | n/a | n/a | 0 | ND | yes |
| 3 | n/a | n/a | n/a | 9 | ND | yes |
| 4 | n/a | n/a | n/a | 19 | ND | yes |
| 5 | 81.93 | 80.34 | 81.29 | 31 | ND | no |
| 6 | 81.57 | 79.28 | 80.49 | 33 | ND | no |
| 7 | 79.32 | 77.13 | 78.44 | 36 | ND | no |
| 8 | 79.35 | 77.08 | 78.16 | 28 | ND | no |
| 9 | 80.02 | 77.82 | 78.92 | 40 | ND | no |
| 10 | n/a | n/a | n/a | 13 | ND | yes |
| 1 | n/a | n/a | n/a | 1 | NE | yes |
| 2 | n/a | n/a | n/a | 4 | NE | yes |
| 3 | 82.98 | 80.72 | 81.26 | 25 | NE | no |
| 4 | 82.22 | 80.41 | 81.11 | 44 | NE | no |
| 5 | 82.88 | 80.34 | 81.24 | 42 | NE | no |
| 6 | 81.22 | 79.95 | 80.50 | 61 | NE | no |
| 7 | 79.89 | 78.45 | 79.03 | 81 | NE | no |
| 8 | 79.65 | 77.83 | 78.63 | 102 | NE | no |
| 9 | 79.04 | 76.52 | 77.54 | 96 | NE | no |
| 10 | 78.38 | 73.43 | 75.21 | 42 | NE | no |
| 1 | n/a | n/a | n/a | 10 | NH | yes |
| 2 | 81.89 | 80.34 | 80.95 | 28 | NH | no |
| 3 | 81.11 | 80.11 | 80.57 | 71 | NH | no |
| 4 | 82.85 | 79.84 | 80.80 | 66 | NH | no |
| 5 | 81.77 | 79.12 | 80.11 | 60 | NH | no |
| 6 | 80.26 | 77.22 | 78.48 | 43 | NH | no |
| 7 | 78.38 | 76.38 | 77.34 | 24 | NH | no |
| 8 | n/a | n/a | n/a | 10 | NH | yes |
| 9 | n/a | n/a | n/a | 3 | NH | yes |
| 10 | n/a | n/a | n/a | 1 | NH | yes |
| 1 | 83.66 | 82.15 | 83.05 | 206 | NJ | no |
| 2 | 82.40 | 80.87 | 81.67 | 325 | NJ | no |
| 3 | 81.15 | 80.17 | 80.68 | 406 | NJ | no |
| 4 | 79.92 | 79.19 | 79.52 | 334 | NJ | no |
| 5 | 78.82 | 77.68 | 78.24 | 248 | NJ | no |
| 6 | 78.02 | 76.09 | 77.06 | 135 | NJ | no |
| 7 | 78.01 | 76.40 | 77.21 | 96 | NJ | no |
| 8 | 76.75 | 75.07 | 75.99 | 43 | NJ | no |
| 9 | 79.07 | 74.79 | 76.16 | 25 | NJ | no |
| 10 | 74.11 | 71.71 | 72.80 | 24 | NJ | no |
| 1 | n/a | n/a | n/a | 9 | NM | yes |
| 2 | n/a | n/a | n/a | 16 | NM | yes |
| 3 | n/a | n/a | n/a | 19 | NM | yes |
| 4 | 82.54 | 79.45 | 81.03 | 36 | NM | no |
| 5 | 81.14 | 79.26 | 80.07 | 61 | NM | no |
| 6 | 80.03 | 78.58 | 79.18 | 57 | NM | no |
| 7 | 80.18 | 77.52 | 78.44 | 62 | NM | no |
| 8 | 79.77 | 76.79 | 77.80 | 75 | NM | no |
| 9 | 77.80 | 76.17 | 76.99 | 73 | NM | no |
| 10 | 77.61 | 75.48 | 76.43 | 85 | NM | no |
| 1 | 82.12 | 79.81 | 81 | 37 | NV | no |
| 2 | 81.09 | 79.53 | 80.43 | 71 | NV | no |
| 3 | 79.84 | 77.94 | 78.93 | 127 | NV | no |
| 4 | 79.50 | 77.45 | 78.42 | 134 | NV | no |
| 5 | 78.17 | 75.99 | 76.75 | 96 | NV | no |
| 6 | 78.09 | 75.06 | 75.63 | 78 | NV | no |
| 7 | 76.35 | 74 | 75.04 | 46 | NV | no |
| 8 | 77.92 | 74.48 | 75.97 | 26 | NV | no |
| 9 | n/a | n/a | n/a | 13 | NV | yes |
| 10 | n/a | n/a | n/a | 11 | NV | yes |
| 1 | 83.52 | 81.21 | 82.23 | 1,082 | NY | no |
| 2 | 81.54 | 80.12 | 80.86 | 904 | NY | no |
| 3 | 81.44 | 79.88 | 80.58 | 405 | NY | no |
| 4 | 82.20 | 80.53 | 81.34 | 260 | NY | no |
| 5 | 81.95 | 80.87 | 81.38 | 223 | NY | no |
| 6 | 81.54 | 80.35 | 80.93 | 227 | NY | no |
| 7 | 81.03 | 79.83 | 80.48 | 286 | NY | no |
| 8 | 80.84 | 79.31 | 80.04 | 289 | NY | no |
| 9 | 79.49 | 77.79 | 78.67 | 318 | NY | no |
| 10 | 78.03 | 75.34 | 76.53 | 253 | NY | no |
| 1 | n/a | n/a | n/a | 16 | OH | yes |
| 2 | 83.79 | 80.85 | 82.37 | 40 | OH | no |
| 3 | 82.03 | 80.89 | 81.41 | 105 | OH | no |
| 4 | 80.77 | 79.70 | 80.12 | 133 | OH | no |
| 5 | 80.62 | 80.03 | 80.31 | 225 | OH | no |
| 6 | 79.46 | 78.67 | 79.04 | 309 | OH | no |
| 7 | 78.62 | 77.68 | 78.13 | 422 | OH | no |
| 8 | 77.71 | 76.59 | 77.14 | 458 | OH | no |
| 9 | 76.13 | 74.94 | 75.57 | 464 | OH | no |
| 10 | 73.25 | 72.24 | 72.75 | 599 | OH | no |
| 1 | n/a | n/a | n/a | 3 | OK | yes |
| 2 | n/a | n/a | n/a | 9 | OK | yes |
| 3 | n/a | n/a | n/a | 17 | OK | yes |
| 4 | 81.15 | 78.56 | 80.29 | 37 | OK | no |
| 5 | 80.54 | 78.87 | 79.67 | 78 | OK | no |
| 6 | 78.79 | 77.42 | 78.25 | 99 | OK | no |
| 7 | 78.02 | 77.11 | 77.57 | 134 | OK | no |
| 8 | 76.40 | 75.43 | 75.86 | 152 | OK | no |
| 9 | 76.07 | 74.92 | 75.44 | 286 | OK | no |
| 10 | 73.81 | 72.47 | 72.93 | 265 | OK | no |
| 1 | 83.62 | 81.64 | 82.64 | 69 | OR | no |
| 2 | 82.06 | 79.93 | 80.87 | 194 | OR | no |
| 3 | 80.33 | 78.92 | 79.57 | 207 | OR | no |
| 4 | 79.74 | 78.01 | 78.89 | 172 | OR | no |
| 5 | 78.30 | 77.16 | 77.68 | 132 | OR | no |
| 6 | 77.37 | 75.92 | 76.57 | 74 | OR | no |
| 7 | 78.33 | 76.28 | 77.27 | 43 | OR | no |
| 8 | n/a | n/a | n/a | 16 | OR | yes |
| 9 | n/a | n/a | n/a | 8 | OR | yes |
| 10 | n/a | n/a | n/a | 2 | OR | yes |
| 1 | 82.85 | 80.71 | 82.15 | 50 | PA | no |
| 2 | 82.06 | 80.72 | 81.66 | 137 | PA | no |
| 3 | 81.49 | 80.49 | 81.06 | 250 | PA | no |
| 4 | 81.14 | 79.39 | 80.26 | 247 | PA | no |
| 5 | 80.52 | 79.02 | 79.81 | 330 | PA | no |
| 6 | 80.02 | 78.72 | 79.38 | 354 | PA | no |
| 7 | 79.09 | 77.85 | 78.49 | 420 | PA | no |
| 8 | 78.40 | 77.20 | 77.81 | 406 | PA | no |
| 9 | 77.60 | 75.92 | 76.69 | 401 | PA | no |
| 10 | 74.67 | 73.04 | 73.75 | 421 | PA | no |
| 1 | n/a | n/a | n/a | 8 | RI | yes |
| 2 | n/a | n/a | 81.66 | 23 | RI | no |
| 3 | n/a | n/a | 80.52 | 38 | RI | no |
| 4 | n/a | n/a | 80.07 | 42 | RI | no |
| 5 | n/a | n/a | 78.34 | 62 | RI | no |
| 6 | n/a | n/a | 76.56 | 28 | RI | no |
| 7 | n/a | n/a | n/a | 10 | RI | yes |
| 8 | n/a | n/a | n/a | 4 | RI | yes |
| 9 | n/a | n/a | n/a | 1 | RI | yes |
| 10 | n/a | n/a | n/a | 0 | RI | yes |
| 1 | 82.76 | 80.47 | 81.45 | 21 | SC | no |
| 2 | 83.66 | 79.79 | 81.09 | 37 | SC | no |
| 3 | 80.93 | 79.62 | 80.16 | 51 | SC | no |
| 4 | 80.72 | 79.01 | 79.89 | 83 | SC | no |
| 5 | 79.26 | 78.24 | 78.83 | 95 | SC | no |
| 6 | 78.31 | 77.34 | 77.86 | 137 | SC | no |
| 7 | 77.61 | 76.37 | 76.98 | 165 | SC | no |
| 8 | 76.50 | 75.50 | 76 | 140 | SC | no |
| 9 | 75.50 | 74.58 | 74.99 | 183 | SC | no |
| 10 | 74.38 | 73.49 | 73.90 | 236 | SC | no |
| 1 | n/a | n/a | n/a | 0 | SD | yes |
| 2 | n/a | n/a | n/a | 2 | SD | yes |
| 3 | n/a | n/a | n/a | 8 | SD | yes |
| 4 | 83.30 | 81.05 | 82.39 | 21 | SD | no |
| 5 | 84.12 | 80.92 | 81.95 | 22 | SD | no |
| 6 | 81.24 | 79.61 | 80.39 | 34 | SD | no |
| 7 | 80.72 | 79.13 | 79.82 | 48 | SD | no |
| 8 | 80.46 | 77.29 | 78.75 | 32 | SD | no |
| 9 | 78.97 | 76.54 | 77.55 | 38 | SD | no |
| 10 | n/a | n/a | n/a | 18 | SD | yes |
| 1 | 81.97 | 80 | 81.73 | 26 | TN | no |
| 2 | 81.43 | 77.87 | 80.09 | 40 | TN | no |
| 3 | 80.41 | 78.22 | 79.19 | 86 | TN | no |
| 4 | 79.32 | 77.42 | 78.46 | 137 | TN | no |
| 5 | 78.25 | 76.78 | 77.47 | 148 | TN | no |
| 6 | 77.09 | 75.97 | 76.55 | 208 | TN | no |
| 7 | 75.97 | 75.16 | 75.58 | 190 | TN | no |
| 8 | 75.23 | 74.41 | 74.80 | 213 | TN | no |
| 9 | 73.88 | 73.31 | 73.61 | 237 | TN | no |
| 10 | 73.01 | 71.59 | 72.04 | 177 | TN | no |
| 1 | 82.22 | 80.86 | 81.75 | 189 | TX | no |
| 2 | 81.04 | 80.40 | 80.69 | 303 | TX | no |
| 3 | 80.93 | 80.08 | 80.49 | 445 | TX | no |
| 4 | 79.93 | 79.35 | 79.64 | 529 | TX | no |
| 5 | 79.58 | 78.76 | 79.15 | 596 | TX | no |
| 6 | 78.67 | 78 | 78.31 | 658 | TX | no |
| 7 | 78.42 | 77.52 | 77.97 | 650 | TX | no |
| 8 | 77.79 | 76.63 | 77.11 | 712 | TX | no |
| 9 | 77.03 | 75.94 | 76.46 | 706 | TX | no |
| 10 | 77.15 | 74.69 | 75.92 | 756 | TX | no |
| 1 | 81.98 | 81.48 | 81.73 | 45 | UT | no |
| 2 | 82.33 | 80.65 | 81.14 | 130 | UT | no |
| 3 | 81.26 | 79.43 | 80.18 | 150 | UT | no |
| 4 | 79.98 | 77.80 | 78.50 | 133 | UT | no |
| 5 | 78.81 | 76.80 | 77.56 | 83 | UT | no |
| 6 | 80.73 | 77.14 | 78.38 | 48 | UT | no |
| 7 | n/a | n/a | n/a | 18 | UT | yes |
| 8 | n/a | n/a | n/a | 3 | UT | yes |
| 9 | n/a | n/a | n/a | 2 | UT | yes |
| 10 | n/a | n/a | n/a | 3 | UT | yes |
| 1 | 83.30 | 81.77 | 82.74 | 213 | VA | no |
| 2 | 81.82 | 80.41 | 81.33 | 219 | VA | no |
| 3 | 80.68 | 79.59 | 80.09 | 232 | VA | no |
| 4 | 80 | 79.04 | 79.53 | 196 | VA | no |
| 5 | 79.08 | 78.08 | 78.62 | 213 | VA | no |
| 6 | 78.04 | 76.63 | 77.36 | 206 | VA | no |
| 7 | 76.85 | 75.83 | 76.34 | 173 | VA | no |
| 8 | 75.72 | 74.58 | 75.15 | 146 | VA | no |
| 9 | 75.44 | 74.37 | 74.94 | 137 | VA | no |
| 10 | 74.41 | 72.82 | 73.55 | 101 | VA | no |
| 1 | n/a | n/a | n/a | 0 | VT | yes |
| 2 | n/a | n/a | n/a | 9 | VT | yes |
| 3 | n/a | n/a | n/a | 16 | VT | yes |
| 4 | 84.08 | 80.76 | 81.68 | 31 | VT | no |
| 5 | 83.33 | 80.77 | 81.91 | 33 | VT | no |
| 6 | 82.72 | 80.59 | 81.71 | 40 | VT | no |
| 7 | 81.17 | 79.21 | 80.20 | 27 | VT | no |
| 8 | 78.96 | 77.03 | 78.08 | 21 | VT | no |
| 9 | n/a | n/a | n/a | 4 | VT | yes |
| 10 | n/a | n/a | n/a | 0 | VT | yes |
| 1 | 82.94 | 81.68 | 82.34 | 343 | WA | no |
| 2 | 81.79 | 80.29 | 80.63 | 358 | WA | no |
| 3 | 80.57 | 79.20 | 79.64 | 351 | WA | no |
| 4 | 79.33 | 77.60 | 78.35 | 240 | WA | no |
| 5 | 78.87 | 77.60 | 78.15 | 129 | WA | no |
| 6 | 78.84 | 77.07 | 77.94 | 97 | WA | no |
| 7 | 77.91 | 75.46 | 76.60 | 70 | WA | no |
| 8 | 79.58 | 75.16 | 76.99 | 25 | WA | no |
| 9 | n/a | n/a | n/a | 8 | WA | yes |
| 10 | n/a | n/a | n/a | 5 | WA | yes |
| 1 | n/a | n/a | n/a | 6 | WI | yes |
| 2 | n/a | n/a | n/a | 16 | WI | yes |
| 3 | 81.74 | 80.02 | 81.34 | 81 | WI | no |
| 4 | 81.65 | 80.67 | 81.03 | 133 | WI | no |
| 5 | 80.79 | 79.87 | 80.31 | 200 | WI | no |
| 6 | 81 | 79.99 | 80.48 | 218 | WI | no |
| 7 | 79.85 | 78.56 | 79.15 | 256 | WI | no |
| 8 | 78.34 | 76.98 | 77.56 | 194 | WI | no |
| 9 | 77.93 | 76.53 | 77.17 | 144 | WI | no |
| 10 | 76.68 | 73.67 | 74.17 | 75 | WI | no |

The first 500 of 520 rows; every row is in https://vishalsingh.org/zipcode/report/place-and-the-length-of-a-life/figures/le_state_tiles.json.

Note: A state's tenth with fewer than 20 tracts is left empty, so lines break. Intervals are from a bootstrap over counties; a state with too few counties for the bootstrap (Connecticut, Delaware, Hawaii, Rhode Island and the District of Columbia) has no band.

Source: USALEEP life expectancy at birth, 2010-2015, and Area Deprivation Index, 2023; population-weighted (ACS 2019-2023 tract population), modeled estimates

Data: https://vishalsingh.org/zipcode/report/place-and-the-length-of-a-life/figures/le_state_tiles.json

## All estimates

Every fact in facts.json (113), sorted by id.

| Fact id | Value | Unit | Estimand | Population | Period | Interval | n | Weights | Source | Note |
| --- | ---: | --- | --- | --- | --- | --- | ---: | --- | --- | --- |
| adi_within | 34% | percent of the variance of the ADI national percentile | share of the population-weighted variance of the ADI national percentile that lies between tracts in the same county (sum-of-squares decomposition) | tracts in the 50 states and DC with all nine measures, an ACS population above zero and an ADI rank | ACS 2019-2023; ADI 2023 |   | 82,545 tracts | ACS 2019-2023 tract population | Area Deprivation Index |   |
| alt_adi_least | 82.3 | Life expectancy at birth (years) | population-weighted mean of tract life expectancy at birth; 95% intervals from a bootstrap over counties | the least deprived tenth of tracts ranked by Area Deprivation Index, among the tracts with all four rankings | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population | 95% CI 82.1 to 82.6 | 7,179 tracts | ACS 2019-2023 tract population | USALEEP life expectancy |   |
| alt_adi_most | 74.0 | Life expectancy at birth (years) | population-weighted mean of tract life expectancy at birth; 95% intervals from a bootstrap over counties | the most deprived tenth of tracts ranked by Area Deprivation Index, among the tracts with all four rankings | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population | 95% CI 73.8 to 74.3 | 7,206 tracts | ACS 2019-2023 tract population | USALEEP life expectancy |   |
| alt_adi_rho | −1.00 | rank correlation | Spearman rank correlation between bin order and bin mean, each bin counted once | the bins of tracts ranked by Area Deprivation Index | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population |   |   | each bin counted once | USALEEP life expectancy |   |
| alt_degree_least | 82.3 | Life expectancy at birth (years) | population-weighted mean of tract life expectancy at birth; 95% intervals from a bootstrap over counties | the least deprived tenth of tracts ranked by Bachelor's degree share, among the tracts with all four rankings | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population | 95% CI 82.0 to 82.5 | 7,178 tracts | ACS 2019-2023 tract population | USALEEP life expectancy |   |
| alt_degree_most | 75.3 | Life expectancy at birth (years) | population-weighted mean of tract life expectancy at birth; 95% intervals from a bootstrap over counties | the most deprived tenth of tracts ranked by Bachelor's degree share, among the tracts with all four rankings | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population | 95% CI 74.9 to 75.7 | 7,175 tracts | ACS 2019-2023 tract population | USALEEP life expectancy |   |
| alt_degree_rho | −1.00 | rank correlation | Spearman rank correlation between bin order and bin mean, each bin counted once | the bins of tracts ranked by Bachelor's degree share | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population |   |   | each bin counted once | USALEEP life expectancy |   |
| alt_income_least | 82.2 | Life expectancy at birth (years) | population-weighted mean of tract life expectancy at birth; 95% intervals from a bootstrap over counties | the least deprived tenth of tracts ranked by Median household income, among the tracts with all four rankings | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population | 95% CI 82.0 to 82.4 | 7,177 tracts | ACS 2019-2023 tract population | USALEEP life expectancy |   |
| alt_income_most | 74.2 | Life expectancy at birth (years) | population-weighted mean of tract life expectancy at birth; 95% intervals from a bootstrap over counties | the most deprived tenth of tracts ranked by Median household income, among the tracts with all four rankings | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population | 95% CI 73.9 to 74.4 | 7,178 tracts | ACS 2019-2023 tract population | USALEEP life expectancy |   |
| alt_income_rho | −1.00 | rank correlation | Spearman rank correlation between bin order and bin mean, each bin counted once | the bins of tracts ranked by Median household income | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population |   |   | each bin counted once | USALEEP life expectancy |   |
| alt_n_places | 71,771 | tracts | count of tracts in the common sample of the ranking comparison | tracts with the outcome and all four rankings | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population |   |   | none (full count) | USALEEP life expectancy |   |
| alt_rho_cut | 0.9 | absolute rank correlation | threshold between a steady and an uneven gradient | the rule for changed gradients (docs/RANKING_SWITCH_RULE.md) | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population |   |   | none (a fixed rule) | USALEEP life expectancy |   |
| alt_rho_margin | 0.1 | absolute rank correlation | smallest move in the rank correlation that counts as a change of shape | the rule for changed gradients (docs/RANKING_SWITCH_RULE.md) | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population |   |   | none (a fixed rule) | USALEEP life expectancy |   |
| alt_std_least | 82.4 | Life expectancy at birth (years) | population-weighted mean of tract life expectancy at birth; 95% intervals from a bootstrap over counties | the least deprived tenth of tracts ranked by Standardized index, among the tracts with all four rankings | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population | 95% CI 82.2 to 82.5 | 7,177 tracts | ACS 2019-2023 tract population | USALEEP life expectancy |   |
| alt_std_most | 74.7 | Life expectancy at birth (years) | population-weighted mean of tract life expectancy at birth; 95% intervals from a bootstrap over counties | the most deprived tenth of tracts ranked by Standardized index, among the tracts with all four rankings | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population | 95% CI 74.2 to 75.1 | 7,178 tracts | ACS 2019-2023 tract population | USALEEP life expectancy |   |
| alt_std_rho | −1.00 | rank correlation | Spearman rank correlation between bin order and bin mean, each bin counted once | the bins of tracts ranked by Standardized index | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population |   |   | each bin counted once | USALEEP life expectancy |   |
| chk_delta | 3.6 | percentage points of adults, higher in the most deprived tenth of tracts | difference between the weighted mean prevalence of the least deprived and the most deprived tenth of tracts by ADI national percentile, bootstrap interval over counties (draws shared by the two means) | adults: routine checkup in the past year | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 95% CI 2.4 to 4.6 | 8,213 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| chk_hi | 77.8% | percent of adults | weighted mean of the tract prevalence in a tenth of tracts (rank bins of the ADI; tracts with tied ranks fall in the same tenth), bootstrap interval over counties | adults in the most deprived tenth of tracts by ADI national percentile: routine checkup in the past year | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 95% CI 77.3% to 78.2% | 8,213 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| chk_lo | 74.2% | percent of adults | weighted mean of the tract prevalence in a tenth of tracts (rank bins of the ADI; tracts with tied ranks fall in the same tenth), bootstrap interval over counties | adults in the least deprived tenth of tracts by ADI national percentile: routine checkup in the past year | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 95% CI 73.2% to 75.3% | 8,274 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| ladder_county | 1.12 | years of life expectancy at birth lower per ten-point step in the ADI national percentile | population-weighted least-squares slope of tract life expectancy on the ADI national percentile, times ten, sign reversed, with a constant for each county; cluster-robust interval (t on the number of counties minus one) | residents of non-institutional tracts in the 50 states and DC with a life expectancy estimate and an ADI rank | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population | 95% CI 1.07 to 1.16 | 77,722 tracts | ACS 2019-2023 tract population | USALEEP life expectancy |   |
| ladder_raw | 0.84 | years of life expectancy at birth lower per ten-point step in the ADI national percentile | population-weighted least-squares slope of tract life expectancy on the ADI national percentile, times ten, sign reversed; cluster-robust interval (t on the number of states minus one) | residents of non-institutional tracts in the 50 states and DC with a life expectancy estimate and an ADI rank | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population | 95% CI 0.75 to 0.92 | 77,936 tracts | ACS 2019-2023 tract population | USALEEP life expectancy |   |
| le_d1 | 82.3 | years of life expectancy at birth | population-weighted mean of tract life expectancy in a tenth of tracts (rank bins of the ADI; tracts with tied ranks fall in the same tenth), bootstrap interval over counties | residents of tracts in the least deprived tenth of tracts by ADI national percentile | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population | 95% CI 82.0 to 82.6 | 7,820 tracts | ACS 2019-2023 tract population | USALEEP life expectancy |   |
| le_d10 | 74.0 | years of life expectancy at birth | population-weighted mean of tract life expectancy in a tenth of tracts (rank bins of the ADI; tracts with tied ranks fall in the same tenth), bootstrap interval over counties | residents of tracts in the most deprived tenth of tracts by ADI national percentile | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population | 95% CI 73.7 to 74.3 | 7,844 tracts | ACS 2019-2023 tract population | USALEEP life expectancy |   |
| le_decile_drop | 8.3 | years of life expectancy at birth lower in the most deprived tenth of tracts | difference between the population-weighted mean life expectancy of the least and the most deprived tenth of tracts, bootstrap interval over counties (draws shared by the two means) | residents of non-institutional tracts in the 50 states and DC with a life expectancy estimate and an ADI rank | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population | 95% CI 7.9 to 8.7 | 7,844 tracts | ACS 2019-2023 tract population | USALEEP life expectancy |   |
| le_step_first | 1.3 | years of life expectancy at birth, first tenth minus second tenth of tracts by ADI | difference between the population-weighted mean life expectancy of the least deprived tenth and the next tenth, bootstrap interval over counties (draws shared) | residents of non-institutional tracts in the 50 states and DC with a life expectancy estimate and an ADI rank | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population | 95% CI 1.0 to 1.6 | 7,820 tracts | ACS 2019-2023 tract population | USALEEP life expectancy |   |
| le_step_last | 1.9 | years of life expectancy at birth, ninth tenth minus tenth tenth of tracts by ADI | difference between the population-weighted mean life expectancy of the ninth tenth and the most deprived tenth, bootstrap interval over counties (draws shared) | residents of non-institutional tracts in the 50 states and DC with a life expectancy estimate and an ADI rank | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population | 95% CI 1.6 to 2.1 | 7,844 tracts | ACS 2019-2023 tract population | USALEEP life expectancy |   |
| metro_flat_edge | 0.70 | years of life expectancy at birth lower per ten-point step in the ADI national percentile | 5th percentile over metros of the estimated metro slope, sign reversed | residents of metros (core-based statistical areas) with at least 40 tracts that have a life expectancy estimate and an ADI rank | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population |   | 267 metros | none (each metro counts once) | USALEEP life expectancy | A percentile of estimates, which are stretched by estimation error. |
| metro_floor | 40 | tracts | minimum number of tracts with a life expectancy estimate and an ADI rank for a metro to enter (rule) | metros | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population |   |   | none (definition) | USALEEP life expectancy |   |
| metro_median | 1.20 | years of life expectancy at birth lower per ten-point step in the ADI national percentile | median over metros of the metro slope, sign reversed; bootstrap interval over metros | residents of metros (core-based statistical areas) with at least 40 tracts that have a life expectancy estimate and an ADI rank | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population | 95% CI 1.14 to 1.23 | 267 metros | none (each metro counts once) | USALEEP life expectancy |   |
| metro_n_lower | 267 | metros | count of metros with life expectancy lower in more deprived tracts: the reported slope (years lower per ten-point step) is above zero | residents of metros (core-based statistical areas) with at least 40 tracts that have a life expectancy estimate and an ADI rank | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population |   | 267 metros | none (full count) | USALEEP life expectancy |   |
| metro_n_sig | 260 | metros | count of metros whose 95% interval for the reported slope lies above zero | residents of metros (core-based statistical areas) with at least 40 tracts that have a life expectancy estimate and an ADI rank | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population |   | 267 metros | none (full count) | USALEEP life expectancy |   |
| metro_noise | 48% | percent of the variance of the estimated metro slopes | mean squared standard error of the metro slopes over the variance of the estimated slopes across metros | residents of metros (core-based statistical areas) with at least 40 tracts that have a life expectancy estimate and an ADI rank | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population |   | 267 metros | none (each metro counts once) | USALEEP life expectancy | The standard errors treat tracts as independent in metros with fewer than five counties, so this share may be too small. |
| metro_pop_share | 82% | percent of residents | share of residents of the tract sample who live in the metros of the analysis | residents of non-institutional tracts in the 50 states and DC with a life expectancy estimate and an ADI rank | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population |   | 267 metros | ACS 2019-2023 tract population | USALEEP life expectancy |   |
| metro_steep_edge | 1.67 | years of life expectancy at birth lower per ten-point step in the ADI national percentile | 95th percentile over metros of the estimated metro slope, sign reversed | residents of metros (core-based statistical areas) with at least 40 tracts that have a life expectancy estimate and an ADI rank | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population |   | 267 metros | none (each metro counts once) | USALEEP life expectancy | A percentile of estimates, which are stretched by estimation error. |
| mto_age_cut | 13 | years of age | the age below which Chetty, Hendren and Katz (2016) find the gain in income (definition from the study) | children in the Moving to Opportunity experiment | age when the family moved |   |   | none (definition) | Published estimates from cited studies |   |
| mto_bmi_cut | 40 | kilograms per square meter (body-mass index) | the body-mass index at or above which Ludwig and coauthors (2011) count a woman as extremely obese (definition from the study) | women in the Moving to Opportunity experiment | health measured in 2008 to 2010 |   |   | none (definition) | Published estimates from cited studies |   |
| mto_diab_diff | 4.3 | percentage points lower prevalence of glycated hemoglobin of 6.5% or more in the low-poverty voucher group than in the control group | Ludwig and coauthors (2011), New England Journal of Medicine: 4,498 women with children living in public housing in high-poverty census tracts, randomly assigned in 1994 to 1998 to a voucher usable only in a low-poverty tract, an unrestricted voucher or no voucher; health measured in 2008 to 2010 (intention to treat); reported absolute difference in the prevalence of a glycated hemoglobin level of 6.5% or more, copied from the abstract and not computed in this build | women in the Moving to Opportunity experiment | health measured in 2008 to 2010 | 95% CI 0.8 to 7.8 |   | none (reported estimate) | Published estimates from cited studies | A published estimate, not a result of this build. |
| mto_hba1c_cut | 6.5% | percent glycated hemoglobin | the glycated hemoglobin level at or above which Ludwig and coauthors (2011) count a woman as having diabetes (definition from the study) | women in the Moving to Opportunity experiment | health measured in 2008 to 2010 |   |   | none (definition) | Published estimates from cited studies |   |
| mto_income_itt | 14% | percent higher annual income in the mid-twenties than the control group (intention to treat, $1,624) | Chetty, Hendren and Katz (2016), American Economic Review (NBER working paper 21156): children below age 13 at random assignment, individual earnings in the mid-twenties, control mean $11,270; intention-to-treat effect of assignment to the experimental voucher group, $1,624, a 14 percent increase; read in the text of the working paper and not computed in this build | children below age 13 at random assignment to the experimental voucher group of the Moving to Opportunity experiment | income in the mid-twenties |   |   | none (reported estimate) | Published estimates from cited studies | A published estimate, not a result of this build. |
| mto_income_tot | 31% | percent higher annual income in the mid-twenties among families that took up the voucher (treatment on the treated, $3,477) | Chetty, Hendren and Katz (2016), American Economic Review (NBER working paper 21156): children below age 13 at random assignment, individual earnings in the mid-twenties, control mean $11,270; treatment-on-the-treated effect for families that took up the experimental voucher, the intention-to-treat effect scaled by the take-up rate, $3,477, a 31 percent increase; read in the text of the working paper and not computed in this build | children below age 13 whose families took up the experimental voucher in the Moving to Opportunity experiment | income in the mid-twenties |   |   | none (reported estimate) | Published estimates from cited studies | A published estimate, not a result of this build. |
| mto_obese_diff | 3.4 | percentage points lower prevalence of a body-mass index of 40 or more in the low-poverty voucher group than in the control group | Ludwig and coauthors (2011), New England Journal of Medicine: 4,498 women with children living in public housing in high-poverty census tracts, randomly assigned in 1994 to 1998 to a voucher usable only in a low-poverty tract, an unrestricted voucher or no voucher; health measured in 2008 to 2010 (intention to treat); reported absolute difference in the prevalence of a body-mass index of 40 or more, copied from the abstract and not computed in this build | women in the Moving to Opportunity experiment | health measured in 2008 to 2010 | 95% CI 0.4 to 6.4 |   | none (reported estimate) | Published estimates from cited studies | A published estimate, not a result of this build. |
| mto_takeup | 48% | percent of families assigned the experimental voucher who took it up | Chetty, Hendren and Katz (2016): take-up rate of the experimental voucher; read in the text of the working paper (NBER w21156) and not computed in this build | families in the experimental voucher group of the Moving to Opportunity experiment | take-up of the voucher |   |   | none (reported estimate) | Published estimates from cited studies | A published estimate, not a result of this build. |
| n_cloud | 4,000 | tracts | size of the random sample of tracts drawn with a fixed seed for the scatter (rule) | residents of non-institutional tracts in the 50 states and DC with a life expectancy estimate and an ADI rank | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population |   | 77,936 tracts | none (definition) | USALEEP life expectancy |   |
| n_gradient_tracts | 77,936 | tracts | count of tracts in the gradient | residents of non-institutional tracts in the 50 states and DC with a life expectancy estimate and an ADI rank | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population |   |   | none (full count) | USALEEP life expectancy |   |
| n_metros | 267 | metros | count of metros | residents of metros (core-based statistical areas) with at least 40 tracts that have a life expectancy estimate and an ADI rank | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population |   |   | none (full count) | USALEEP life expectancy | Metropolitan and micropolitan statistical areas; each tract takes the area of its county (Census county file). |
| n_no_pair | 3 | counties | count of counties with no pair whose ADI difference reaches the minimum | the 20 most populous counties | deaths in 2010 to 2015 (USALEEP, native 2010 tracts); ADI 2023 and ACS 2019-2023 on the same tracts |   |   | none (full count) | USALEEP life expectancy |   |
| n_pair_pool | 4,917 | tracts | count of tracts from which the pairs are drawn | tracts of the 20 most populous counties that pass the floors of the rule: boundary unchanged since 2010, not institutional, at least 2,000 residents, a standard error of at most 2.0 years, an ADI rank | deaths in 2010 to 2015 (USALEEP, native 2010 tracts); ADI 2023 and ACS 2019-2023 on the same tracts |   |   | none (full count) | USALEEP life expectancy |   |
| n_pair_tracts | 13,625 | tracts | count of tracts searched | 2020 tracts of the 20 most populous counties with an ACS population above zero and a life expectancy estimate | deaths in 2010 to 2015 (USALEEP, native 2010 tracts); ADI 2023 and ACS 2019-2023 on the same tracts |   |   | none (full count) | USALEEP life expectancy |   |
| n_pairs | 17 | pairs of tracts | count of counties with such a pair (one pair per county) | counties among the 20 most populous with a pair that meets the rule | deaths in 2010 to 2015 (USALEEP, native 2010 tracts); ADI 2023 and ACS 2019-2023 on the same tracts |   |   | none (full count) | USALEEP life expectancy |   |
| n_spread_counties | 30 | counties | number of most populous counties in the county spread (rule) | counties and county equivalents of the 50 states and DC | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population |   |   | none (definition) | USALEEP life expectancy |   |
| nyc_borough_mean_range | 4.2 | years of life expectancy at birth between the highest and the lowest borough mean of tract life expectancy | largest minus smallest of the five boroughs' population-weighted mean of tract life expectancy | residents of non-institutional tracts in the five boroughs of New York City (Bronx, Kings, New York, Queens and Richmond counties) that have a life expectancy estimate and an ADI rank | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population |   | 5 boroughs | ACS 2019-2023 tract population | USALEEP life expectancy | A spread of five means and a range of tract estimates are not the same kind of contrast. |
| nyc_borough_range_min | 9.3 | years of life expectancy at birth between the 5th and 95th percentile of tracts in a borough | smallest of the five boroughs' population-weighted 95th minus 5th percentile of tract life expectancy (it is Richmond County's), each borough counted once | residents of non-institutional tracts in the five boroughs of New York City (Bronx, Kings, New York, Queens and Richmond counties) that have a life expectancy estimate and an ADI rank | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population |   | 5 boroughs | none (each borough counts once) | USALEEP life expectancy | The percentiles are of modeled tract estimates, so part of the range is estimation error. |
| nyc_borough_range_min_net | 6.9 | years of life expectancy at birth between the 5th and 95th percentile of tracts in a borough, net of estimation error | smallest of the five boroughs' 5th to 95th percentile range after shrinking the tract estimates toward the borough mean by the square root of one minus the share of the variance that is estimation error (published standard errors, normal approximation), each borough counted once | residents of non-institutional tracts in the five boroughs of New York City (Bronx, Kings, New York, Queens and Richmond counties) that have a life expectancy estimate and an ADI rank | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population |   | 5 boroughs | none (each borough counts once) | USALEEP life expectancy | The 2020-tract standard errors are mapping grade and understate the noise, so the range net of error is an upper bound. |
| nyc_n_gray | 250 | tracts | count of gray tracts on the map: no residents in the ACS, no life expectancy estimate, an institutional population, or no ADI rank | 2020 tracts of the five boroughs of New York City (Bronx, Kings, New York, Queens and Richmond counties) that are not in the gradient sample | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population |   |   | none (full count) | USALEEP life expectancy |   |
| nyc_n_shaded | 2,074 | tracts | count of tracts shaded on the New York City map: the tracts of the gradient sample that lie in the five boroughs | non-institutional tracts in the five boroughs of New York City (Bronx, Kings, New York, Queens and Richmond counties) with a life expectancy estimate, an ACS 2019-2023 population above zero and an ADI rank | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population |   |   | none (full count) | USALEEP life expectancy |   |
| nyc_noise | 37% | percent of the variance of tract life expectancy across the five boroughs | population-weighted mean squared standard error of the tract estimates over their population-weighted variance (the published errors taken at face value) | residents of non-institutional tracts in the five boroughs of New York City (Bronx, Kings, New York, Queens and Richmond counties) that have a life expectancy estimate and an ADI rank | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population |   | 2,074 tracts | ACS 2019-2023 tract population | USALEEP life expectancy | The 2020-tract errors are mapping grade and understate the noise, so the share is a lower bound. |
| nyc_se_median | 2.0 | years of life expectancy | median over the shaded tracts of the standard error of tract life expectancy, each tract counted once | residents of non-institutional tracts in the five boroughs of New York City (Bronx, Kings, New York, Queens and Richmond counties) that have a life expectancy estimate and an ADI rank | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population |   | 2,074 tracts | none (each tract counts once) | USALEEP life expectancy | The 2020-tract standard errors are mapping grade and understate the error of the area-weighted estimates. |
| pair_adi_min | 50 | percentile points of the ADI national percentile | smallest difference in ADI between the two tracts for a pair to count (rule) | pairs of tracts | deaths in 2010 to 2015 (USALEEP, native 2010 tracts); ADI 2023 and ACS 2019-2023 on the same tracts |   |   | none (definition) | USALEEP life expectancy |   |
| pair_n_counties | 20 | counties | number of most populous counties searched for a pair (rule) | counties and county equivalents of the 50 states and DC | deaths in 2010 to 2015 (USALEEP, native 2010 tracts); ADI 2023 and ACS 2019-2023 on the same tracts |   |   | none (definition) | USALEEP life expectancy |   |
| pair_pop_floor | 2,000 | residents | smallest ACS 2019-2023 population of a tract that enters a pair (rule) | tracts in a pair | deaths in 2010 to 2015 (USALEEP, native 2010 tracts); ADI 2023 and ACS 2019-2023 on the same tracts |   |   | none (definition) | USALEEP life expectancy |   |
| pair_radius | 5 | kilometers | largest straight-line distance between the Census internal points of the two tracts (rule) | pairs of tracts | deaths in 2010 to 2015 (USALEEP, native 2010 tracts); ADI 2023 and ACS 2019-2023 on the same tracts |   |   | none (definition) | USALEEP life expectancy |   |
| pair_se_cap | 2.0 | years of life expectancy | largest standard error of the USALEEP life expectancy of a tract that enters a pair (rule) | tracts in a pair | deaths in 2010 to 2015 (USALEEP, native 2010 tracts); ADI 2023 and ACS 2019-2023 on the same tracts |   |   | none (definition) | USALEEP life expectancy |   |
| pair_stable_share | 99% | percent of land area | smallest share of the land area of the 2010 tract and of the 2020 tract that two tracts with one GEOID must share for the boundary to count as unchanged (rule) | tracts present in both the 2010 and the 2020 census | deaths in 2010 to 2015 (USALEEP, native 2010 tracts); ADI 2023 and ACS 2019-2023 on the same tracts |   |   | none (definition) | USALEEP life expectancy |   |
| pairs_median_gap | 5.4 | years of life expectancy at birth, less deprived tract minus more deprived tract | median over the pairs of the difference in life expectancy, each pair counted once; bootstrap interval over pairs | residents of the tracts of one pair per county: in each of the 20 most populous counties, the two tracts within 5 kilometers of each other with the widest difference in ADI, where that difference is at least 50 percentile points | deaths in 2010 to 2015 (USALEEP, native 2010 tracts); ADI 2023 and ACS 2019-2023 on the same tracts | 95% CI 3.0 to 6.9 | 17 pairs of tracts | none (each pair counts once) | USALEEP life expectancy | The two tracts of a pair are chosen on the ADI and not on life expectancy, so the difference is not selected for size. |
| pairs_median_km3 | 2.7 | years of life expectancy at birth, less deprived tract minus more deprived tract | median over pairs of the difference in life expectancy, each pair counted once | pairs chosen by the same rule with 3 kilometers in place of the main setting, in counties where a pair meets the ADI minimum | deaths in 2010 to 2015 (USALEEP, native 2010 tracts); ADI 2023 and ACS 2019-2023 on the same tracts |   | 17 pairs of tracts | none (each pair counts once) | USALEEP life expectancy |   |
| pairs_median_km8 | 5.1 | years of life expectancy at birth, less deprived tract minus more deprived tract | median over pairs of the difference in life expectancy, each pair counted once | pairs chosen by the same rule with 8 kilometers in place of the main setting, in counties where a pair meets the ADI minimum | deaths in 2010 to 2015 (USALEEP, native 2010 tracts); ADI 2023 and ACS 2019-2023 on the same tracts |   | 18 pairs of tracts | none (each pair counts once) | USALEEP life expectancy |   |
| pairs_median_pop4000 | 5.5 | years of life expectancy at birth, less deprived tract minus more deprived tract | median over pairs of the difference in life expectancy, each pair counted once | pairs chosen by the same rule with a population floor of 4,000 in place of the main setting, in counties where a pair meets the ADI minimum | deaths in 2010 to 2015 (USALEEP, native 2010 tracts); ADI 2023 and ACS 2019-2023 on the same tracts |   | 16 pairs of tracts | none (each pair counts once) | USALEEP life expectancy |   |
| pairs_median_se15 | 4.3 | years of life expectancy at birth, less deprived tract minus more deprived tract | median over pairs of the difference in life expectancy, each pair counted once | pairs chosen by the same rule with a standard-error cap of 1.5 years in place of the main setting, in counties where a pair meets the ADI minimum | deaths in 2010 to 2015 (USALEEP, native 2010 tracts); ADI 2023 and ACS 2019-2023 on the same tracts |   | 15 pairs of tracts | none (each pair counts once) | USALEEP life expectancy |   |
| pairs_median_se30 | 6.2 | years of life expectancy at birth, less deprived tract minus more deprived tract | median over pairs of the difference in life expectancy, each pair counted once | pairs chosen by the same rule with a standard-error cap of 3.0 years in place of the main setting, in counties where a pair meets the ADI minimum | deaths in 2010 to 2015 (USALEEP, native 2010 tracts); ADI 2023 and ACS 2019-2023 on the same tracts |   | 17 pairs of tracts | none (each pair counts once) | USALEEP life expectancy |   |
| pairs_median_senone | 4.6 | years of life expectancy at birth, less deprived tract minus more deprived tract | median over pairs of the difference in life expectancy, each pair counted once | pairs chosen by the same rule with no standard-error cap in place of the main setting, in counties where a pair meets the ADI minimum | deaths in 2010 to 2015 (USALEEP, native 2010 tracts); ADI 2023 and ACS 2019-2023 on the same tracts |   | 18 pairs of tracts | none (each pair counts once) | USALEEP life expectancy |   |
| pairs_n_clear | 12 | pairs of tracts | count of pairs whose 95% interval for the difference, which treats the two standard errors as independent, lies above zero | residents of the tracts of one pair per county: in each of the 20 most populous counties, the two tracts within 5 kilometers of each other with the widest difference in ADI, where that difference is at least 50 percentile points | deaths in 2010 to 2015 (USALEEP, native 2010 tracts); ADI 2023 and ACS 2019-2023 on the same tracts |   | 17 pairs of tracts | none (full count) | USALEEP life expectancy |   |
| pairs_n_higher | 15 | pairs of tracts | count of pairs in which the less deprived tract has the higher life expectancy estimate | residents of the tracts of one pair per county: in each of the 20 most populous counties, the two tracts within 5 kilometers of each other with the widest difference in ADI, where that difference is at least 50 percentile points | deaths in 2010 to 2015 (USALEEP, native 2010 tracts); ADI 2023 and ACS 2019-2023 on the same tracts |   | 17 pairs of tracts | none (full count) | USALEEP life expectancy |   |
| pairs_n_se15 | 15 | pairs of tracts | count of counties with such a pair (one pair per county) | pairs chosen by the same rule with a standard-error cap of 1.5 years, in counties where a pair meets the ADI minimum | deaths in 2010 to 2015 (USALEEP, native 2010 tracts); ADI 2023 and ACS 2019-2023 on the same tracts |   |   | none (full count) | USALEEP life expectancy |   |
| pairs_n_senone | 18 | pairs of tracts | count of counties with such a pair (one pair per county) | pairs chosen by the same rule with no standard-error cap, in counties where a pair meets the ADI minimum | deaths in 2010 to 2015 (USALEEP, native 2010 tracts); ADI 2023 and ACS 2019-2023 on the same tracts |   |   | none (full count) | USALEEP life expectancy |   |
| pairs_pop_alt | 4,000 | residents | larger population floor tried (sensitivity) | tracts in a pair | deaths in 2010 to 2015 (USALEEP, native 2010 tracts); ADI 2023 and ACS 2019-2023 on the same tracts |   |   | none (definition) | USALEEP life expectancy |   |
| pairs_radius_km3 | 3 | kilometers | smaller radius tried (sensitivity) | pairs of tracts | deaths in 2010 to 2015 (USALEEP, native 2010 tracts); ADI 2023 and ACS 2019-2023 on the same tracts |   |   | none (definition) | USALEEP life expectancy |   |
| pairs_radius_km8 | 8 | kilometers | larger radius tried (sensitivity) | pairs of tracts | deaths in 2010 to 2015 (USALEEP, native 2010 tracts); ADI 2023 and ACS 2019-2023 on the same tracts |   |   | none (definition) | USALEEP life expectancy |   |
| pairs_se_alt_hi | 3.0 | years of life expectancy | larger standard-error cap tried (sensitivity) | tracts in a pair | deaths in 2010 to 2015 (USALEEP, native 2010 tracts); ADI 2023 and ACS 2019-2023 on the same tracts |   |   | none (definition) | USALEEP life expectancy |   |
| pairs_se_alt_lo | 1.5 | years of life expectancy | smaller standard-error cap tried (sensitivity) | tracts in a pair | deaths in 2010 to 2015 (USALEEP, native 2010 tracts); ADI 2023 and ACS 2019-2023 on the same tracts |   |   | none (definition) | USALEEP life expectancy |   |
| pw_adi_hi | 94 | ADI national percentile | Area Deprivation Index national percentile of one tract (mean of its block groups' ranks), rounded | the more deprived tract of the widest pair | deaths in 2010 to 2015 (USALEEP, native 2010 tracts); ADI 2023 and ACS 2019-2023 on the same tracts |   | 1 tracts | none (one tract) | USALEEP life expectancy |   |
| pw_adi_lo | 24 | ADI national percentile | Area Deprivation Index national percentile of one tract (mean of its block groups' ranks), rounded | the less deprived tract of the widest pair | deaths in 2010 to 2015 (USALEEP, native 2010 tracts); ADI 2023 and ACS 2019-2023 on the same tracts |   | 1 tracts | none (one tract) | USALEEP life expectancy |   |
| pw_gap | 12.4 | years of life expectancy at birth, less deprived tract minus more deprived tract | difference between the USALEEP life expectancy of the less deprived and the more deprived tract; 95% interval treats the two standard errors as independent | residents of two tracts of Cook County, IL, within 5 kilometers of each other (the pair of that county with the widest ADI difference) | deaths in 2010 to 2015 (USALEEP, native 2010 tracts); ADI 2023 and ACS 2019-2023 on the same tracts | 95% CI 7.6 to 17.2 | 2 tracts | none (two tracts) | USALEEP life expectancy | The widest of the 17 pairs by life expectancy, so its estimate is the likeliest to include noise that raised the gap. |
| pw_km | 5.0 | kilometers | straight-line distance between the Census internal points of the two tracts | residents of two tracts of Cook County, IL, within 5 kilometers of each other (the pair of that county with the widest ADI difference) | deaths in 2010 to 2015 (USALEEP, native 2010 tracts); ADI 2023 and ACS 2019-2023 on the same tracts |   |   | none (one pair) | USALEEP life expectancy |   |
| pw_le_hi | 65.8 | years of life expectancy at birth | USALEEP life expectancy at birth of one tract (native 2010 tract), 95% interval of 1.96 standard errors | residents of the more deprived tract of the pair, Census Tract 4910 | deaths in 2010 to 2015 (USALEEP, native 2010 tracts); ADI 2023 and ACS 2019-2023 on the same tracts | 95% CI 62.5 to 69.1 | 1 tracts | none (one tract) | USALEEP life expectancy |   |
| pw_le_lo | 78.2 | years of life expectancy at birth | USALEEP life expectancy at birth of one tract (native 2010 tract), 95% interval of 1.96 standard errors | residents of the less deprived tract of the pair, Census Tract 7201 | deaths in 2010 to 2015 (USALEEP, native 2010 tracts); ADI 2023 and ACS 2019-2023 on the same tracts | 95% CI 74.7 to 81.7 | 1 tracts | none (one tract) | USALEEP life expectancy |   |
| smk_delta | 13.7 | percentage points of adults, higher in the most deprived tenth of tracts | difference between the weighted mean prevalence of the least deprived and the most deprived tenth of tracts by ADI national percentile, bootstrap interval over counties (draws shared by the two means) | adults: current cigarette smoking | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 95% CI 13.1 to 14.4 | 8,213 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| smk_hi | 22.8% | percent of adults | weighted mean of the tract prevalence in a tenth of tracts (rank bins of the ADI; tracts with tied ranks fall in the same tenth), bootstrap interval over counties | adults in the most deprived tenth of tracts by ADI national percentile: current cigarette smoking | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 95% CI 22.5% to 23.1% | 8,213 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| smk_lo | 9.1% | percent of adults | weighted mean of the tract prevalence in a tenth of tracts (rank bins of the ADI; tracts with tied ranks fall in the same tenth), bootstrap interval over counties | adults in the least deprived tenth of tracts by ADI national percentile: current cigarette smoking | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 95% CI 8.5% to 9.6% | 8,274 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| spread_median | 10.5 | years of life expectancy at birth between the 5th and 95th percentile of tracts | median over counties of the population-weighted 95th minus 5th percentile of tract life expectancy, each county counted once | residents of the 30 most populous counties and Connecticut planning regions | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population |   | 30 counties | none (each county counts once) | USALEEP life expectancy | The percentiles are of modeled tract estimates, so part of the range is estimation error. |
| spread_median_15 | 10.5 | years of life expectancy at birth between the 5th and 95th percentile of tracts | median over counties of the population-weighted 95th minus 5th percentile of tract life expectancy, each county counted once | residents of the 15 most populous counties and Connecticut planning regions | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population |   | 15 counties | none (each county counts once) | USALEEP life expectancy |   |
| spread_median_60 | 10.5 | years of life expectancy at birth between the 5th and 95th percentile of tracts | median over counties of the population-weighted 95th minus 5th percentile of tract life expectancy, each county counted once | residents of the 60 most populous counties and Connecticut planning regions | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population |   | 60 counties | none (each county counts once) | USALEEP life expectancy |   |
| spread_median_net | 7.7 | years of life expectancy at birth between the 5th and 95th percentile of tracts, net of estimation error | median over counties of the 5th to 95th percentile range after shrinking the tract estimates toward the county mean by the square root of one minus the share of the variance that is estimation error (published standard errors, normal approximation) | residents of the 30 most populous counties and Connecticut planning regions | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population |   | 30 counties | none (each county counts once) | USALEEP life expectancy | The 2020-tract standard errors are mapping grade and understate the noise, so the range net of error is an upper bound. |
| spread_n_15 | 15 | counties | smaller or larger number of counties tried (sensitivity) | counties and county equivalents of the 50 states and DC | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population |   |   | none (definition) | USALEEP life expectancy |   |
| spread_n_60 | 60 | counties | smaller or larger number of counties tried (sensitivity) | counties and county equivalents of the 50 states and DC | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population |   |   | none (definition) | USALEEP life expectancy |   |
| spread_noise_median | 43% | percent of the variance of tract life expectancy within the county | median over counties of the mean squared standard error of tract life expectancy over the population-weighted variance of tract life expectancy in the county | residents of the 30 most populous counties and Connecticut planning regions | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population |   | 30 counties | none (each county counts once) | USALEEP life expectancy |   |
| spread_widest | 13.9 | years of life expectancy at birth between the 5th and 95th percentile of tracts | population-weighted 95th minus 5th percentile of tract life expectancy in the county with the widest range among the 30 most populous | residents of Cook County, IL | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population |   | 1,221 tracts | ACS 2019-2023 tract population | USALEEP life expectancy | Includes estimation error. |
| st_min_tracts | 20 | tracts | definition: the fewest tracts a state's tenth needs for an estimate | a state's tenth of tracts | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population |   |   | none (definition) | USALEEP life expectancy |   |
| st_span | 4 | national tenths | definition: the fewest national tenths, first to last, a state's estimated tenths must span to be counted | a state's tracts | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population |   |   | none (definition) | USALEEP life expectancy |   |
| st_with_span | 49 | states | count of states (50 and DC) meeting the stated rule, from population-weighted tract means by national ADI tenth | states whose estimated national ADI tenths of tracts run at least 3 tenths apart | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population |   |   | none (full count) | USALEEP life expectancy |   |
| tenth_max | 8,031 | tracts | largest count of tracts in a tenth of the gradient by ADI national percentile (tracts with tied ranks fall in the same tenth) | residents of non-institutional tracts in the 50 states and DC with a life expectancy estimate and an ADI rank | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population |   |   | none (full count) | USALEEP life expectancy |   |
| tenth_min | 7,613 | tracts | smallest count of tracts in a tenth of the gradient by ADI national percentile (tracts with tied ranks fall in the same tenth) | residents of non-institutional tracts in the 50 states and DC with a life expectancy estimate and an ADI rank | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population |   |   | none (full count) | USALEEP life expectancy |   |
| tract_se_median | 1.7 | years of life expectancy | median over tracts of the standard error of tract life expectancy, each tract counted once | residents of non-institutional tracts in the 50 states and DC with a life expectancy estimate and an ADI rank | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population |   | 77,936 tracts | none (each tract counts once) | USALEEP life expectancy | The 2020-tract standard errors are mapping grade and understate the error of the area-weighted estimates. |
| us_checkup | 75.6% | percent of adults | weighted mean of the tract prevalence over all the tracts of the tenths pooled, bootstrap interval over counties (the draws of the tenths) | adults in all tracts with an ADI rank: routine checkup in the past year | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 95% CI 75.1% to 76.0% | 82,781 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| us_le | 78.8 | years of life expectancy at birth | population-weighted mean of tract life expectancy over all the tracts of the tenths pooled, bootstrap interval over counties (the draws of the tenths) | residents of all the tracts in the comparison: non-institutional tracts in the 50 states and DC with a life expectancy estimate and an ADI rank | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population | 95% CI 78.6 to 79.0 | 77,936 tracts | ACS 2019-2023 tract population | USALEEP life expectancy |   |
| us_smoking | 14.5% | percent of adults | weighted mean of the tract prevalence over all the tracts of the tenths pooled, bootstrap interval over counties (the draws of the tenths) | adults in all tracts with an ADI rank: current cigarette smoking | BRFSS 2022 (PLACES 2024 release); ADI 2023 | 95% CI 14.2% to 14.8% | 82,781 tracts | adult population of the tract (PLACES) | CDC PLACES |   |
| vsplit_county | 15.9 | percent of the variance of tract life expectancy | share of the population-weighted variance of tract life expectancy at birth between counties within states (sum-of-squares decomposition) | residents of non-institutional tracts with a life expectancy estimate (50 states and DC) | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts) |   | 78,008 tracts | ACS 2019-2023 tract population | USALEEP life expectancy | Connecticut's nine planning regions stand in for its counties. The tract estimates are modeled; their mean squared standard error equals 26.7 percent of the total variance and sits inside the within-county share. |
| vsplit_n_counties | 3,112 | counties and Connecticut planning regions | count of counties with at least one tract in the variance split | residents of non-institutional tracts with a life expectancy estimate (50 states and DC) | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts) |   |   | none (full count) | USALEEP life expectancy |   |
| vsplit_n_tracts | 78,008 | tracts | count of tracts in the variance split | residents of non-institutional tracts with a life expectancy estimate (50 states and DC) | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts) |   |   | none (full count) | USALEEP life expectancy |   |
| vsplit_noise | 26.7 | percent of the variance of tract life expectancy | population-weighted mean squared standard error of tract life expectancy as a share of the total variance | residents of non-institutional tracts with a life expectancy estimate (50 states and DC) | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts) |   | 78,008 tracts | ACS 2019-2023 tract population | USALEEP life expectancy | The 2020-tract standard errors are mapping grade and understate the error. |
| vsplit_noise_within | 38% | percent of the within-county variance of tract life expectancy | mean squared standard error of tract life expectancy over the within-county part of the variance (the canonical noise share over the canonical within-county share), taking the published errors at face value | residents of non-institutional tracts with a life expectancy estimate (50 states and DC) | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts) |   | 78,008 tracts | ACS 2019-2023 tract population | USALEEP life expectancy | The same denominator as the county shares of estimation error in the county spread figure. |
| vsplit_state | 13.7 | percent of the variance of tract life expectancy | share of the population-weighted variance of tract life expectancy at birth between states (sum-of-squares decomposition) | residents of non-institutional tracts with a life expectancy estimate (50 states and DC) | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts) |   | 78,008 tracts | ACS 2019-2023 tract population | USALEEP life expectancy | Connecticut's nine planning regions stand in for its counties. The tract estimates are modeled; their mean squared standard error equals 26.7 percent of the total variance and sits inside the within-county share. |
| vsplit_within | 70.3 | percent of the variance of tract life expectancy | share of the population-weighted variance of tract life expectancy at birth between tracts within counties (sum-of-squares decomposition) | residents of non-institutional tracts with a life expectancy estimate (50 states and DC) | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts) |   | 78,008 tracts | ACS 2019-2023 tract population | USALEEP life expectancy | Connecticut's nine planning regions stand in for its counties. The tract estimates are modeled; their mean squared standard error equals 26.7 percent of the total variance and sits inside the within-county share. |
| vsplit_within_signal | 60% | percent of the variance net of estimation error | within-county share of the variance of tract life expectancy after subtracting the mean squared standard error from the total and from the within-county part | residents of non-institutional tracts with a life expectancy estimate (50 states and DC) | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts) |   | 78,008 tracts | ACS 2019-2023 tract population | USALEEP life expectancy | Takes the published standard errors at face value; the 2020-tract errors are mapping grade and understate the noise, so the share of the true differences that lies within counties is lower. |

## Definitions

- Census tract: A small area drawn by the Census Bureau to stand in for a neighborhood. Boundaries are redrawn after each decennial census.
- Area Deprivation Index (ADI): A score that ranks neighborhoods by income, education, employment and housing conditions drawn from the American Community Survey. In this book a higher national percentile means a more deprived neighborhood.
- Modeled estimate: A figure produced by a statistical model instead of a direct count or measurement, usually for small areas where direct data are thin, borrowing strength from surveys, neighboring areas or other data. Neighboring areas can share errors, so single extremes call for caution.
- Fixed effects: A regression adjustment that gives each group, such as each county, its own baseline. The estimate then rests on differences within groups.
- Ecological inference: Drawing conclusions about individuals from data about places or groups. A pattern across counties need not hold for the people inside them.

## Methods

This chapter compares neighborhood estimates of life expectancy with a deprivation index and repeats the comparison at smaller geographic scales. The deaths behind the estimates (2010 to 2015) precede the survey data behind the index (2019 to 2023) by about a decade, the tract estimates are modeled and the displayed intervals do not carry all of their error, and the nearby pairs come from a rule applied to the most populous counties only. No result estimates what moving would do to a person's lifespan; the box on the Moving to Opportunity experiment quotes cited studies.

### Sample and period

Tract life expectancy at birth is the USALEEP estimate for deaths in 2010 to 2015. The gradient, the metro slopes, the variance split, the county spread and the New York City map use the file on 2020 tracts, with institutional tracts (at least 25 percent of residents in group quarters, or a known case) dropped and an ACS 2019-2023 population above zero, in the 50 states and DC. The gradient covers the 77,936 such tracts that have an estimate and an ADI rank.

The Area Deprivation Index is the 2023 release of the national percentile (1 least, 100 most deprived), built on the ACS for 2019 to 2023, joined by tract. The deaths therefore precede the ACS data the ADI uses by about a decade.

The nearby pairs use a separate sample, the native 2010 tracts of the most populous counties whose boundaries did not change (How the neighborhoods were paired, below). Smoking and the checkup are CDC PLACES 2024 estimates, which model survey responses from 2022. The box on the Moving to Opportunity experiment quotes estimates from the cited studies; they are not results of this build and keep the populations, outcomes and periods the studies report.

The New York City map draws the tracts of the gradient sample that lie in the five counties of New York City (Bronx, Kings, New York, Queens and Richmond) on the 2020 tract outlines of the city, and every other tract of the outlines is gray with a reason: no residents in the ACS, no life expectancy estimate, an institutional tract or no ADI rank. No tract is chosen or marked.

Explains Table 1, Table 2, Table 6, Table 4 and Table 5.

### Matching places across years

USALEEP built its estimates on 2010 tracts. The file on 2020 tracts used for the gradient and the comparisons of scale is area-weighted from the 2010 tracts, and its standard errors are mapping grade, so they understate the uncertainty of an estimate on a 2020 tract.

A pair uses the native estimate and standard error for the 2010 tract, not the interpolated file, and only for tracts whose boundary did not change. The 2010 and 2020 tracts with one GEOID must share at least 99% of the land area of each (Census 2010-to-2020 relationship file), so the ADI and the ACS population of the 2020 tract describe the same ground.

### Weights and comparisons

Weights are the ACS 2019-2023 tract population for the means of tenths and percentiles, the metro slopes, the county spread and the borough figures, so those results describe residents, and so are the variance splits of life expectancy and of the ADI. Medians across metros, counties and boroughs count each place once, and each pair counts once in the median gap over pairs.

Tenths of tracts are rank bins of the ADI with tracts of tied rank in the same tenth, so they hold from 7,613 to 8,031 tracts. The dots in the first tab of the gradient figure are a random sample of 4,000 tracts drawn with a fixed seed; the line, the band and the markers use every tract.

Smoking and the checkup are the estimates of the chapter on health behaviors and the supply of care, from its function and its tenths of tracts, weighted by adults in each tract, with county-bootstrap intervals. PLACES values are modeled estimates built partly from the ACS, as the ADI is, so part of each gradient is built in.

### Uncertainty

Each tract estimate has a published standard error, with a median of 1.7 years. The interval of a tract is 1.96 standard errors, and the interval of a gap treats the two errors as independent.

Intervals for the means of tenths and percentiles come from repeatedly drawing whole counties with replacement and recomputing the mean, a bootstrap over counties; they are percentile intervals from 2,000 draws with seed 12345. The median gap over pairs has a bootstrap interval over pairs, with the same draws and seed. Metro slopes have errors clustered on counties where a metro has at least five and HC3 errors otherwise.

The band in the gradient figure is the population-weighted interquartile range of tracts at each ADI percentile. It describes the spread of places, includes estimation error and is not an interval for the line.

The bootstrap resamples counties of published estimates. It does not add the error of the USALEEP model beyond the published standard errors, the error of moving estimates onto 2020 tracts or uncertainty in the ADI.

Explains Table 1 and Table 3.

### How the neighborhoods were paired

Each of the 20 most populous counties, by ACS population of the tracts in the sample, is searched. A tract enters if its boundary did not change between the 2010 and 2020 censuses (Matching places across years, above), if it is not an institutional tract such as a prison or a dormitory, if it has at least 2,000 residents and if the standard error of its life expectancy is at most 2.0 years. 4,917 of the 13,625 non-institutional tracts of these counties that have a life expectancy estimate qualify.

Among qualifying tracts whose Census internal points are at most 5 kilometers apart (a straight line, projected to EPSG:5070), the county's pair is the one whose ADI differs most, with ties going to the shorter distance, and it counts if the difference reaches 50 percentile points. No pair reaches that difference in 3 counties, and they are left out. The rule never looks at life expectancy.

The data hold no transit routes, so the pairs are a short distance apart and not a measured ride. The pairs are the result of a rule applied to large counties and are not a sample of neighborhoods.

Explains Table 3.

### How the local slopes and variance split were calculated

The metro slopes and the fixed-effects ladder are the functions of the chapter on gradients, slopes and variance. A core-based statistical area is coded by the Census county file, a metro needs at least 40 tracts, and each slope is a population-weighted least-squares fit. The within-county slope adds a separate baseline for each county.

The variance split is b2.canon.variance_split(), the function the chapter on gradients, slopes and variance calls, and the split of the ADI is the one in the chapter on gradients, slopes and variance. Estimation error is the population-weighted mean squared standard error of the tract estimates as a share of the total variance; the 2020-tract standard errors are mapping grade and understate it.

The county spread gives, for the 30 most populous counties, the population-weighted 5th and 95th percentiles of tract life expectancy and the population-weighted mean. The view net of estimation error shrinks each tract estimate toward the county mean by the square root of one minus the county's share of variance that is estimation error. The shrinkage is an approximation that takes the published errors at face value; it does not recover the true differences between tracts.

For the New York City map, the 5th to 95th percentile of tract life expectancy in a borough and the mean of a borough are population-weighted, and the shrinkage toward the borough mean and the share of the variance that is estimation error are the rules of the county spread, applied to the five boroughs and taking the published standard errors at face value.

Explains Table 6, Table 4, Table 5 and Table 2.

### How the results change under alternative choices

The median gap over pairs depends on the settings of the rule, above all the radius. Pairs chosen within 3 kilometers have a median gap of 2.7 years, and within 8 kilometers it is 5.1, against 5.4 within 5 kilometers. Raising the population floor to 4,000 residents gives 5.5.

A cap of 1.5 years on the standard error gives a median of 4.3 over 15 pairs, a cap of 3.0 years gives 6.2, and no cap gives 4.6 over 18 pairs.

The number of counties in the county spread matters little. The median range is 10.5 years among the 15 most populous counties and 10.5 among the 60 most populous, against 10.5 among 30.

Explains Table 3 and Table 5.

### Four rankings of the same places

The comparison follows the book's ranking switch rule, fixed on October 28, 2026 before any of these gradients was computed. The main gradient of the chapter is recomputed under four rankings of the same places, with the outcome, period, weights, bins and intervals unchanged: the published ADI; a standardized index that applies the ADI's coefficients to its measures after standardizing each one across tracts; median household income; and the share of adults 25 and older with a bachelor's degree or more. The three alternatives come from ACS tract estimates for 2019 to 2023, the period of the ADI.

All four rankings use the same places, the 71,771 tracts that have the outcome and all four rankings, so the ADI panel can differ slightly from the chapter's main figure. Tracts enter only if they have every input of the standardized index, which leaves out tracts without a median rent, home value or mortgage cost.

A pattern counts as changed when the worse end switches sides, when the two ends differ under one ranking and not the other, or when the rank correlation of bin order and bin mean falls on the other side of 0.9 and moves by at least 0.1.

Explains Table 7.

### Rankings and subsets of places

The file splits.json in this chapter's data folder repeats the tenths of tracts under five rankings (the ADI, the standardized index, median household income, the bachelor's degree share and the median home value) and within subsets of places: the four Census regions, counties where at least 25% of residents are Black alone and the other counties (ACS 2019-2023), and metro and nonmetro counties (Rural-Urban Continuum Codes 2023). Each place keeps its national tenth, so an end bin means the same level of deprivation in every subset. The outcome, weights and bootstrap are those of the chapter's gradient, over the places with all five rankings. A tenth with fewer than 20 tracts or 5 counties is left empty.

## What these data cannot show

Outside the box, every number here describes places. A tract with a lower life expectancy and a higher ADI is a tract whose residents, taken together, died younger in 2010 to 2015, and the data do not say which of them did, or whether a person who moved would have died later (ecological inference). The pairs show that nearby tracts differ, and the data cannot separate what the places do from who lives in them. The experiment in the box covers women and children who enrolled in one program, and it does not say how much of the gradient in the figures would change if a place changed.

The periods differ. The life expectancy estimates are modeled for deaths in 2010 to 2015 and were never updated. The ADI is a national percentile built on survey data for 2019 to 2023, and it may rank a tract differently from the way the same tract looked in 2010.

The geography is matched imperfectly. The 2020-tract file used for the gradient is area-weighted from 2010 tracts, and its errors are mapping grade, approximations that understate the uncertainty. Because the pairs are chosen on the ADI among large counties, their median describes that rule, as the sensitivity to the radius shows, and says nothing about counties that are small or rural. The counties left out are Los Angeles, Santa Clara and Alameda, all in California, and the ADI understates deprivation in high-cost cities, so a qualifying pair there may exist and not be found. The distances are straight lines and not rides. Puerto Rico has no life expectancy estimates and is not included, and Connecticut's planning regions stand in for its counties. Institutional tracts are dropped, and a tract without an estimate is absent and not zero.

The estimates are modeled. A tract's life expectancy has a median standard error of 1.7 years, which the intervals in Table 3 show. The intervals in the figures come from resampling counties of published estimates, and they leave out error in the life expectancy model beyond its standard errors, error from moving estimates onto 2020 tracts and uncertainty in the ADI. The PLACES measures are modeled estimates built partly from the same census and survey inputs as the ADI, so the lines for smoking and the checkup show how the model maps those inputs into prevalence as well as how adults differ.

## References

1. Arias, Elizabeth; Escobedo, Luis A.; Kennedy, Jiaquan; Fu, Chenxi; Cisewski, Jodi (2018). U.S. Small-area Life Expectancy Estimates Project: Methodology and Results Summary. Vital and Health Statistics, Series 2, Number 181, National Center for Health Statistics. https://www.cdc.gov/nchs/data/series/sr_02/sr02_181.pdf
2. Kind, Amy J. H.; Buckingham, William R. (2018). Making Neighborhood-Disadvantage Metrics Accessible: The Neighborhood Atlas. New England Journal of Medicine 378(26): 2456-2458. https://doi.org/10.1056/NEJMp1802313
3. Greenlund, Kurt J.; Lu, Hua; Wang, Yan; Matthews, Kevin A.; LeClercq, Jennifer M.; Lee, Benjamin; Carlson, Susan A. (2022). PLACES: Local Data for Better Health. Preventing Chronic Disease 19: 210459. https://doi.org/10.5888/pcd19.210459
4. Ludwig, Jens; Sanbonmatsu, Lisa; Gennetian, Lisa; Adam, Emma; Duncan, Greg J.; Katz, Lawrence F.; Kessler, Ronald C.; Kling, Jeffrey R.; Lindau, Stacy Tessler; Whitaker, Robert C.; McDade, Thomas W. (2011). Neighborhoods, Obesity, and Diabetes: A Randomized Social Experiment. New England Journal of Medicine 365(16): 1509-1519. https://doi.org/10.1056/NEJMsa1103216
5. Chetty, Raj; Hendren, Nathaniel; Katz, Lawrence F. (2016). The Effects of Exposure to Better Neighborhoods on Children: New Evidence from the Moving to Opportunity Experiment. American Economic Review 106(4): 855-902. https://doi.org/10.1257/aer.20150572

## Provenance

- Pipeline version: 0.1.0
- Build script: book2/chapters/place-and-the-length-of-a-life/build.py (sha256 42fcdf636a9813540d6eaae0848093bac35cb3a434f0fded640ca760a3a405e9)
- Built on: 2026-10-07
- Site commit: c2b1ff3309cb4a10833f884da91ac0057a66ad0a

Inputs:

| File | Size (bytes) | SHA-256, first 1 MB | SHA-256, last 1 MB |
| --- | ---: | --- | --- |
| book2/chapters/gradients-slopes-and-variance/f_pca.py | 9,345 | 5353ebb9cba70e3b87433e822f6caa660a2b724c02b4d6e0cfe56e455575bf30 | 5353ebb9cba70e3b87433e822f6caa660a2b724c02b4d6e0cfe56e455575bf30 |
| book2/chapters/gradients-slopes-and-variance/f_sample.py | 3,332 | 68db6c906b0076863c96cc3ae3b2ca569e20a783b4043a6a278cbcf78a298192 | 68db6c906b0076863c96cc3ae3b2ca569e20a783b4043a6a278cbcf78a298192 |
| book2/chapters/gradients-slopes-and-variance/f_scale.py | 7,541 | 21813ba176531a15307e31ddac981de8b5383e726f3cba3b569ff001639e1545 | 21813ba176531a15307e31ddac981de8b5383e726f3cba3b569ff001639e1545 |
| book2/chapters/gradients-slopes-and-variance/f_slopes.py | 8,806 | 4a4adf3f306b504412faa7538c4ce2983dce6f1006e7e07e8e0d4c4450c5ba6c | 4a4adf3f306b504412faa7538c4ce2983dce6f1006e7e07e8e0d4c4450c5ba6c |
| book2/chapters/health-behaviors-and-the-supply-of-care/gradients.py | 11,771 | 1fe448a81ab7fbb49ae49ade9c68d93a5617e95e103be3e322c47d29884ef32f | 1fe448a81ab7fbb49ae49ade9c68d93a5617e95e103be3e322c47d29884ef32f |
| book2/chapters/place-and-the-length-of-a-life/i_nyc.py | 6,614 | 26eb60243fc45a586b9e89492728b6d34500fabd44f9275a7c21c5bc83f4fcdd | 26eb60243fc45a586b9e89492728b6d34500fabd44f9275a7c21c5bc83f4fcdd |
| book2/chapters/place-and-the-length-of-a-life/i_pairs.py | 8,863 | 7c1151befeb600c7870f45cc0174d00901cbca5c1c4c77eab21f236451cc2ea9 | 7c1151befeb600c7870f45cc0174d00901cbca5c1c4c77eab21f236451cc2ea9 |
| data/analytic/tract_atlas.parquet | 39,659,925 | 29d5e4d943552d54aeb6ba92aadc939d9bd15bdea553c78357dbd516757be627 | 1e10d1ecf722340f23ee2212d4999a90250f85b65fa89b39e32db06d9472e22e |
| data/geo/counties_2020.parquet | 123,034,155 | 2fce837cdc2db1f2c229f2ea453c6793338a462770307e2c64e8d1fd713ec663 | dae928bf1042f3d174d2d64d6e3ec923fa3daa40aee1bb4feb104aac2fe3c04a |
| data/geo/dim_geo.parquet | 1,180,921 | 29d5e4d943552d54aeb6ba92aadc939d9bd15bdea553c78357dbd516757be627 | e58c8c488df54ab3187f5461f179322afdc49c9bd2315404003fe1272eea32c3 |
| data/geo/tracts_2020.parquet | 619,858,391 | f4f05c8c7a63d1785ec2aad283de3f6fa839b6c437f13a1b96cea31d3344bf0f | 7b3811247c6b3e322954f24a4a2384af52aef6ed8688f8dbdb0407d21ef80a59 |
| data/raw/acs/deprivation_components_2019_2023/county_core.parquet | 498,878 | 8b177857027e73c1f404f7b62191ab78811e95ee54c4f07383e9e9e17da1a345 | 8b177857027e73c1f404f7b62191ab78811e95ee54c4f07383e9e9e17da1a345 |
| data/raw/acs/deprivation_components_2019_2023/tract.parquet | 13,197,793 | 5545cde2cbe96a8b56e468d65388248a241aa6bd300d3c9e325d08d3f86b11da | 5c8ec454d066988b0f1a01aaa6ed34ebfc05d162e01cf9c6bb01f9198cbaff12 |
| data/staged/adi/adi_atlas_tract2020_2023.parquet | 798,078 | 5e409bc212ff871def780fa27b4cf8b616d9da420b2b9f6455517eddcd8c1a95 | 5e409bc212ff871def780fa27b4cf8b616d9da420b2b9f6455517eddcd8c1a95 |
| data/staged/crosswalks/ers_rucc_county_2023.parquet | 67,278 | ad01d36a53664a2bfafbcbebb0f5a921f5b81122ffa80cf96333db23b41c191e | ad01d36a53664a2bfafbcbebb0f5a921f5b81122ffa80cf96333db23b41c191e |
| data/staged/crosswalks/tract_2010_to_2020.parquet | 2,602,636 | f7526f0d1bac31dbc540f6d733a850c9e0ee4c7622011214c04afedfbfa1e0a5 | 9e573492f9804b571b9944a53cc59fadab4051a175952b058f77cc5a644a98ff |
| data/staged/health/places_tract2020_latest.parquet | 5,919,159 | af4c3b0d1205b7d84b81676a9529ef73323b2dbeb9d54e235f5efeebfad83177 | a2f19029ee2ef032be87e0d5288d88d489afd52379279833f178baa021df3dbf |
| data/staged/life_expectancy/usaleep_tract2010.parquet | 771,963 | fccbc4183992062a22c87cfb9644475b69e56b93e3b58b8035379144d8d5fc4e | fccbc4183992062a22c87cfb9644475b69e56b93e3b58b8035379144d8d5fc4e |
| data/staged/life_expectancy/usaleep_tract2020.parquet | 1,316,169 | 9c9c7f103791b75eddba74e68a89804efdf2486b6b66bb4bc02a46ba5bf8bc23 | 27317107be2465d9fd5e83271dea0cd6e915ee419e2802bedc28747d0a41d38f |
| data/staged/ses/acs5_tract2020_2019_2023.parquet | 7,232,424 | 5fe6f730a35225f14d5d8abfc6b9bce58b52e545f309a7eac74f1dae74d7a31c | 7b4a2475b3852d3f91804716a04566ef8ceadacec2be9fb85f5eb642078efac6 |
| data/staged/svi/svi_tract2020_2022.parquet | 17,740,415 | cb3860d8edc2aaf1f2a02211a58f7a0b80457eb6ad9a37b5ec8abb2d5da9da10 | e371a4a696d5bed05d54ad7c5b6ca2504508805b847ec415c2322be7f66ec2b2 |

Outputs:

| File | SHA-256 |
| --- | --- |
| facts.json | 9eed1ad6694ae817ab5800fd787f79ec8a01a706b2d211bccb978f17d11e6e6d |
| figures/alt_rankings.json | 09e34b229fe46e41cd294da684e89b20b3cfa35d04a5f3688c6995e96678516e |
| figures/county_spread.json | e7264a20e88e45945dda904deb0295fd9cf9f8f9629530d58f590cd201d7ce14 |
| figures/le_gradient.json | fec8e001d57c3dc7cdce2699846d13a9b38ae12d162c014ea0adf565f67f48ee |
| figures/le_state_tiles.json | 353890582755910d41349e48f84466bc96d3226a986b1ba491e79bdc6b977f8c |
| figures/metro_slopes.json | 6828971481a384c1cd77181ce323ad134c86df75e927a9454a415e253221d917 |
| figures/neighborhood_pairs.json | a351ec74a8c8fe67ccd8d866f8b591ecc80d0559539cc1196cd8943a1916b3b5 |
| figures/nyc_le_map.json | 7c2ea29ada631687412387561ab61d15ba9abe1c499bba627b7b4efd26108901 |
| figures/variance_split.json | b593bd67f1860a4c7eb7b601fd57c654de59efeac52138f06720bb029d77ba04 |
| meta.json | bee1a755cc118018bb92b9645379a902c8464586320564bf21defbc311a76ee7 |
| public/data/place-and-the-length-of-a-life/le_by_adi_tracts.json | c7d1c7d6edc208acaf1de63281fa09caeabc7c4256ea9b2dd32153f5cfac50cd |
| public/data/place-and-the-length-of-a-life/nyc_le_map.json | 9403c0f4c4f07e8cc999821cccaa7cb8a632036ff44c84847015dc3cea44c649 |
| public/data/place-and-the-length-of-a-life/splits.json | 63e1adbb1d7ae3354bb1e224c982a8c535e1e1137340d5ae18d3d5e613b3c353 |
| refs.json | 2d976463c36bc44fd12832d343cc0a844526c07065c1c3444516db9b007dd7f8 |

## Machine formats

- Report (HTML): https://vishalsingh.org/zipcode/place-and-the-length-of-a-life/report
- Report (Markdown): https://vishalsingh.org/zipcode/place-and-the-length-of-a-life/report.md
- Every estimate (JSON): https://vishalsingh.org/zipcode/report/place-and-the-length-of-a-life/facts.json
- Table 1, figure "le_gradient" (JSON): https://vishalsingh.org/zipcode/report/place-and-the-length-of-a-life/figures/le_gradient.json
- Table 2, figure "nyc_le_map" (JSON): https://vishalsingh.org/zipcode/report/place-and-the-length-of-a-life/figures/nyc_le_map.json
- Table 3, figure "neighborhood_pairs" (JSON): https://vishalsingh.org/zipcode/report/place-and-the-length-of-a-life/figures/neighborhood_pairs.json
- Table 4, figure "variance_split" (JSON): https://vishalsingh.org/zipcode/report/place-and-the-length-of-a-life/figures/variance_split.json
- Table 5, figure "county_spread" (JSON): https://vishalsingh.org/zipcode/report/place-and-the-length-of-a-life/figures/county_spread.json
- Table 6, figure "metro_slopes" (JSON): https://vishalsingh.org/zipcode/report/place-and-the-length-of-a-life/figures/metro_slopes.json
- Table 7, figure "alt_rankings" (JSON): https://vishalsingh.org/zipcode/report/place-and-the-length-of-a-life/figures/alt_rankings.json
- Table 8, figure "le_state_tiles" (JSON): https://vishalsingh.org/zipcode/report/place-and-the-length-of-a-life/figures/le_state_tiles.json
- Payload of "le_gradient" (JSON): https://vishalsingh.org/zipcode/data/place-and-the-length-of-a-life/le_by_adi_tracts.json
- Payload of "nyc_le_map" (JSON): https://vishalsingh.org/zipcode/data/place-and-the-length-of-a-life/nyc_le_map.json
- Index of every report: https://vishalsingh.org/zipcode/llms.txt; all reports in one file: https://vishalsingh.org/zipcode/llms-full.txt

---

# Violence and the life-expectancy gap: report

Chapter 8.5 of ZIP Code Destiny, by Vishal Singh (NYU Stern). Generated from the chapter's data files; the reading version is at https://vishalsingh.org/zipcode/violence-and-the-life-expectancy-gap. Table n of this report is figure n of the chapter.

## Identification

- Title: Violence and the life-expectancy gap
- Label: The lost years
- Chapter: 8.5 (Part VIII, Early death; An American life)
- Question: How much of the life-expectancy gap between the most and least deprived neighborhoods comes from violent deaths?
- Dek: A rough accounting that assumes 41.5 years of life lost per homicide attributes 4.8% of the gap of 8.3 years in life expectancy between the most and the least deprived tenths of tracts to homicide.
- Grain: neighborhoods
- Status: published
- Published: 2026-10-09
- Updated: not revised
- Data sources:
  - CDC injury and violence mapping. CDC homicide rates for tract clusters, annual 2022, 2023 and 2024 (modeled estimates for 1 to 9 deaths, zero for none, direct rates for 10 or more); county file 2019 and 2022 to 2024 for one check; years: 2019-2024 for counties, 2022-2024 for tracts (data as of 2026-05-14); weights: ACS 2019-2023 residents of each tract
  - USALEEP life expectancy. Life expectancy at birth by tract, deaths in 2010 to 2015, area-weighted onto 2020 tracts; years: 2010-2015; weights: 2019-2023 tract population
  - Area Deprivation Index. Area Deprivation Index national rank, 2023, for the tenths of tracts; years: 2023 release (version 4.0.1; built on ACS 2019-2023); weights: none (rank)
  - American Community Survey. Tract population (weights) and age shares, 2019-2023; years: 2019-2023 (5-year); weights: none (full count)
  - Published estimates from cited studies. Mean years of potential life lost per homicide victim and per suicide decedent (Zimmerman, Fridel and Trovato 2024) and a decomposition of the Black-white gap in male life expectancy in Washington, D.C. (Roberts, Reither and Lim 2020); years: as reported in each study; weights: none (published estimates)
- Canonical URL: https://vishalsingh.org/zipcode/violence-and-the-life-expectancy-gap
- Report: https://vishalsingh.org/zipcode/violence-and-the-life-expectancy-gap/report
- Markdown: https://vishalsingh.org/zipcode/violence-and-the-life-expectancy-gap/report.md
- BibTeX key: singh2026violenceandthelifeexpectancygap

```bibtex
@misc{singh2026violenceandthelifeexpectancygap,
  author = {Singh, Vishal},
  title = {Violence and the life-expectancy gap},
  year = {2026},
  url = {https://vishalsingh.org/zipcode/violence-and-the-life-expectancy-gap},
  note = {In ZIP Code Destiny. Published 9 October 2026.}
}
```

## Principal estimates

The facts the dek and the opening prose cite, in order of appearance.

| Estimand | Population | Period | Value | Interval | n | Weights | Source | Fact id |
| --- | --- | --- | ---: | --- | ---: | --- | --- | --- |
| mean of life expectancy at birth (specific to the victim's race, ethnicity and sex, 2019) minus age at death, from Zimmerman, Fridel and Trovato (2024), table 1; used here as an assumed setting for every tenth of tracts | homicide victims in the National Violent Death Reporting System, 44 states and DC | deaths in 2003 to 2019 | 41.5 |   | 98,617 homicide victims | none (mean over victims) | Published estimates from cited studies | yll_setting |
| rough accounting: the part of the gap attributed to homicide over the whole gap, bootstrap interval over counties (draws shared) | residents of non-institutional tracts in the 50 states and DC with a life expectancy estimate and an ADI rank | homicide rates for 2022 to 2024 with life expectancy for deaths in 2010 to 2015 | 4.8% | 95% CI 4.1% to 5.4% | 14,089 tracts | ACS 2019-2023 tract population | CDC injury and violence mapping | viol_gap_share |
| difference between the population-weighted mean life expectancy of the least and the most deprived tenth of tracts, bootstrap interval over counties (draws shared by the two means) | residents of non-institutional tracts in the 50 states and DC with a life expectancy estimate and an ADI rank | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population | 8.3 | 95% CI 7.9 to 8.7 | 7,844 tracts | ACS 2019-2023 tract population | USALEEP life expectancy | le_decile_drop |
| the gap in life expectancy less the part attributed to homicide by rough accounting (a remainder, all other causes) | residents of non-institutional tracts in the 50 states and DC with a life expectancy estimate and an ADI rank | homicide rates for 2022 to 2024 with life expectancy for deaths in 2010 to 2015 | 7.9 | 95% CI 7.5 to 8.3 | 14,089 tracts | ACS 2019-2023 tract population | CDC injury and violence mapping | other_gap_years |
| the remainder over the whole gap: the share not attributed to homicide | residents of non-institutional tracts in the 50 states and DC with a life expectancy estimate and an ADI rank | homicide rates for 2022 to 2024 with life expectancy for deaths in 2010 to 2015 | 95.2% | 95% CI 94.6% to 95.9% | 14,089 tracts | ACS 2019-2023 tract population | CDC injury and violence mapping | other_gap_share |
| rough accounting with years of life lost per homicide set at the largest possible setting: the life expectancy of the most deprived tenth, bootstrap interval over counties | residents of non-institutional tracts in the 50 states and DC with a life expectancy estimate and an ADI rank | homicide rates for 2022 to 2024 with life expectancy for deaths in 2010 to 2015 | 8.5% | 95% CI 7.3% to 9.7% | 14,089 tracts | ACS 2019-2023 tract population | CDC injury and violence mapping | share_bound |
| share of residents in the highest group, mean over the three ways of leaving out a year | residents of those tracts that sit in a CDC cluster with a published homicide rate in each year from 2022 to 2024, outside New York City | annual homicide rates for 2022, 2023 and 2024, averaged (2024 provisional); ADI 2023; ACS 2019-2023 population | 1.3% |   |   | none (full count) | CDC injury and violence mapping | band_top_pop_share |
| part of the difference attributed to homicide by rough accounting over the whole difference, bootstrap interval over counties (draws shared) | residents of tracts in the highest group of clusters by homicide rate | homicide rates for 2022 to 2024 with life expectancy for deaths in 2010 to 2015 | 18.5% | 95% CI 16.5% to 20.3% | 1,337 tracts | ACS 2019-2023 tract population | CDC injury and violence mapping | band_top_share |
| difference between the population-weighted mean life expectancy of the lowest and the highest group, bootstrap interval over counties (draws shared) | residents of tracts in the highest group of clusters by homicide rate | homicide rates for 2022 to 2024 with life expectancy for deaths in 2010 to 2015 | 8.1 | 95% CI 7.7 to 8.4 | 1,337 tracts | ACS 2019-2023 tract population | CDC injury and violence mapping | band_top_gap |
| counties with a tract in the highest group; the smallest over the three ways of leaving out a year | residents of tracts in the highest group of clusters by homicide rate | annual homicide rates for 2022, 2023 and 2024, averaged (2024 provisional); ADI 2023; ACS 2019-2023 population | 54 |   |   | none (full count) | CDC injury and violence mapping | band_top_counties_min |
| counties with a tract in the highest group; the largest over the three ways of leaving out a year | residents of tracts in the highest group of clusters by homicide rate | annual homicide rates for 2022, 2023 and 2024, averaged (2024 provisional); ADI 2023; ACS 2019-2023 population | 67 |   |   | none (full count) | CDC injury and violence mapping | band_top_counties_max |

## Findings

Homicide accounts for 4.8% (95% CI 4.1% to 5.4%) of the 8.3 (95% CI 7.9 to 8.7) years by which life expectancy at birth, for deaths in 2010 to 2015, is lower in the most deprived tenth of tracts than in the least deprived. The figure comes from a rough accounting that takes each tenth's homicide rate for 2022 to 2024 and assumes that each death takes 41.5 years of life. The remaining 7.9 (95% CI 7.5 to 8.3) years, 95.2% (95% CI 94.6% to 95.9%) of the gap, are left to every other cause of death.

Neither source gives death rates by age for a tract. This is therefore a rough accounting and not an age-specific decomposition of the kind Arriaga described.[1] It multiplies a crude homicide rate by life expectancy and by an assumed number of years of life per homicide. The two sources also cover different periods, life expectancy for deaths in 2010 to 2015 and homicide rates for 2022 to 2024, and come from different models. The share is proportional to the assumed years and reaches 8.5% (95% CI 7.3% to 9.7%) at the largest value that one setting shared by every tenth could take.

CDC publishes one homicide rate for each cluster of about five neighboring tracts. In the clusters with the highest rates, which hold 1.3% of residents, homicide accounts for 18.5% (95% CI 16.5% to 20.3%) of a difference of 8.1 (95% CI 7.7 to 8.4) years from the clusters with the lowest rates. That group spans only 54 to 67 counties.

The concentration of gun homicide in a few clusters, and the mix of causes of early death, belong to Chapter 8.1, Gun homicide and the causes of early death (https://vishalsingh.org/zipcode/gun-homicide-and-causes-of-early-death). The gradient of life expectancy across deprivation tenths belongs to Chapter 8.4, Place and the length of a life (https://vishalsingh.org/zipcode/place-and-the-length-of-a-life), whose tract sample and gap this chapter reuses. Overdose, alcohol and suicide are in Chapter 8.2, Overdose, alcohol and suicide (https://vishalsingh.org/zipcode/overdose-alcohol-and-suicide), and their change since 2018 is in Chapter 8.3, Mortality trends, 2018 to 2024 (https://vishalsingh.org/zipcode/mortality-trends-2018-2024). The unit here is the tract, with each tract carrying the rate of its cluster, and nothing below describes an individual.

### Homicide rates rise across deprivation tenths, with the steepest step at the top

This chapter counts a violent death as a homicide by any means, as the CDC injury and violence dashboard defines it, and says "homicide" for what it counts. The National Violent Death Reporting System, the source of the years lost per death used below, counts suicide among violent deaths; this chapter does not. Suicides of every means are not counted. The dashboard's data dictionary names deaths in legal intervention only inside its all-firearm measure, which this chapter does not use. One check below adds firearm suicide. Firearms account for 75.1% of the homicide deaths that the rates imply.

The dashboard's tract file gives one rate for each cluster of about five neighboring tracts and repeats it on every tract of the cluster. The rate is a modeled estimate for clusters with a few deaths, zero for none and a direct rate for many, and Chapter 8.1, Gun homicide and the causes of early death (https://vishalsingh.org/zipcode/gun-homicide-and-causes-of-early-death) describes the file. The rates are crude rates per resident. For the clusters where CDC publishes an exact count of deaths for 2022, the rate times the cluster's residents divided by the count has a median of 0.99, as in the check in that chapter. A tract here carries the mean of its cluster's annual rates for 2022, 2023 and 2024, and the 2024 values are provisional.

The tenths are those of Chapter 8.4, Place and the length of a life (https://vishalsingh.org/zipcode/place-and-the-length-of-a-life), 77,936 tracts ranked on the Area Deprivation Index and cut into ten groups with equal numbers of tracts, not of residents. The least deprived tenth holds 32.2 million residents and the most deprived 23.0 million.

The mean rate rises from 1.8 (95% CI 1.3 to 2.2) homicide deaths per hundred thousand residents a year in the least deprived tenth to 14.8 (95% CI 12.8 to 16.9) in the most deprived, a ratio of 8.4 (95% CI 6.5 to 11.3) (Table 1). The rate is higher in each tenth than in the one before, and the smallest step among the first nine is 0.4 (95% CI 0.1 to 0.7). The step from the ninth tenth to the most deprived, 6.4 (95% CI 4.8 to 8.0), is larger than the next-largest step by 5.3 (95% CI 3.6 to 7.0).

Table 1. Homicide deaths per hundred thousand residents a year rise from 1.8 in the least deprived tenth of tracts to 14.8 in the most deprived

Rates are available for 92.8% of the residents in the gradient. New York City holds 2.6% of them and is left out, because its tract rates are near zero. The mean is 0.17 per hundred thousand in the tract file against 4.9 in the county file for the same boroughs, and 75.0% of its tracts sit at exactly zero.

The city holds 12.1% of the least deprived tenth, where the index, which understates deprivation in high-cost cities, ranks many of its tracts. Giving its tracts the county file's rate for their borough raises the least deprived tenth's rate from 1.8 to 2.1 (95% CI 1.5 to 2.6) and moves the share of the gap from 4.8% to 4.6% (95% CI 4.0% to 5.3%).

A further 4.7% of residents live in clusters where CDC withholds the rate in at least one of 2022, 2023 and 2024. The gap uses all of the gradient's tracts, the rates cover 72,225 of them, and each tenth's rate is applied to the whole tenth. Measuring the gap over only the tracts with a rate gives 4.7% (95% CI 4.0% to 5.3%).

### A homicide rate becomes years of life through two assumptions

For a group of tracts $$g$$, the years of life expectancy lost to homicide are approximated by

$$
\text{years lost}_g \;=\; r_g \times e_{0,g} \times L,
$$

where $$r_g$$ is the group's homicide rate per resident per year, $$e_{0,g}$$ is its life expectancy at birth and $$L$$ is the years of life a homicide takes. In a population with constant births and constant age-specific death rates, the crude death rate times life expectancy at birth equals the lifetime probability of dying of that cause. Under this approximation a newborn's lifetime risk of dying of homicide is 0.14% (95% CI 0.11% to 0.18%) in the least deprived tenth and 1.10% (95% CI 0.95% to 1.24%) in the most deprived. Multiplying the risk by the years a death takes gives, to a first approximation, the life expectancy that removing the cause would add. The approximation ignores competing risks, the chance that a person dies of another cause first, and the age mix of each tract. Residents of working age are 61.4% (95% CI 60.1% to 62.3%) of the least deprived tenth and 58.5% (95% CI 58.2% to 58.7%) of the most deprived, and scaling the rates to a common working-age share moves the answer to 5.0% (95% CI 4.3% to 5.7%).

The value of $$L$$ comes from outside both files. Zimmerman, Fridel and Trovato subtract the age at death from a life expectancy specific to the victim's race, ethnicity and sex for homicide victims in the National Violent Death Reporting System between 2003 and 2019. They find a mean of 41.5 years, with the loss floored at zero for a victim who died older than the life expectancy.[2] Most of the records come from 17 states. The accounting applies that national mean to every tenth. In that study Hispanic, Asian and Black victims lose more years on average than white victims, so the years lost would be above the mean where most victims come from those groups and below it elsewhere. Scaling the mean by each tenth's life expectancy over the national one, which lowers it for the most deprived tenth, gives 4.4% (95% CI 3.8% to 5.0%). The sweep below varies the setting.

### Most of the gap belongs to causes of death other than homicide

Across all tracts with a rate, the accounting puts the average loss to homicide at 0.18 (95% CI 0.16 to 0.20) years of life expectancy. Tenth by tenth, the loss is 0.06 (95% CI 0.05 to 0.07) years in the least deprived tenth and 0.45 (95% CI 0.39 to 0.52) in the most deprived.

The difference of 0.39 (95% CI 0.33 to 0.46) years is 4.8% (95% CI 4.1% to 5.4%) of the gap of 8.3 (95% CI 7.9 to 8.7) years between life expectancies of 82.3 (95% CI 82.0 to 82.6) and 74.0 (95% CI 73.7 to 74.3) years that Chapter 8.4, Place and the length of a life (https://vishalsingh.org/zipcode/place-and-the-length-of-a-life) reports. The other 95.2% (95% CI 94.6% to 95.9%) is the remainder, which includes every other cause of death, among them the overdose and suicide deaths in Chapter 8.2, Overdose, alcohol and suicide (https://vishalsingh.org/zipcode/overdose-alcohol-and-suicide) (Table 2). The two lines rest on different tracts: the observed line uses every tract of each tenth, and the homicide line uses those with a rate.

Table 2. Homicide accounts for 0.39 of the 8.3 years by which life expectancy is lower in the most deprived tenth of tracts than in the least deprived

The homicide part is concentrated at the top of the ranking. The homicide rate in the most deprived tenth is 6.4 (95% CI 4.8 to 8.0) higher than in the ninth, so the step in life expectancy between them contains 0.19 (95% CI 0.14 to 0.24) years of homicide, 10.2% (95% CI 8.0% to 12.3%) of that step's 1.9 (95% CI 1.6 to 2.1) years. That is a larger share than across the gradient as a whole. The accounting says how many years of life homicide removes, and says nothing about which of the other causes of death account for the rest.

### The share is proportional to the years of life each homicide takes

Because the accounting multiplies by $$L$$, the share rises in a straight line with that setting (Table 3). At 30 years per death homicide accounts for 3.4% (95% CI 2.9% to 3.9%) of the gap, and at 55 years for 6.3% (95% CI 5.4% to 7.2%).

A death cannot take more years than a person has. The sweep applies one value to every tenth, so the largest value it can take is the life expectancy of the most deprived tenth, 74.0 years, which would mean a death at birth in that tenth. That value gives 8.5% (95% CI 7.3% to 9.7%). No single value shared by every tenth puts the share above that. If every homicide in the most deprived tenth took that tenth's whole life expectancy and none in the least deprived tenth took any, the share would reach 9.8% (95% CI 8.6% to 11.0%).

Table 3. Homicide accounts for 3.4% of the gap if each takes 30 years of life and 6.3% if each takes 55, and for at most 8.5% when one value applies to every tenth

The other settings move the answer less than $$L$$ does (Table 4), and across the other changes tested the share ranges from 3.9% to 6.2%. Rates for 2022 and 2023 alone, the years that are final, give 5.1% (95% CI 4.4% to 5.9%). Neither file covers 2010 to 2015, the years behind the life expectancy estimates, and the county file reaches back only to 2019. Scaling each tenth's accounting to the 2019 county rates gives 3.9% (95% CI 3.3% to 4.5%), and counting only firearm homicide, the measure in Chapter 8.1, Gun homicide and the causes of early death (https://vishalsingh.org/zipcode/gun-homicide-and-causes-of-early-death), gives 4.1% (95% CI 3.4% to 4.7%). These are the two lowest shares in the table, and they cannot be told apart.

Adding firearm suicide, at the shorter 29.5 years that Zimmerman and colleagues find for suicide decedents, gives 6.2% (95% CI 5.6% to 6.9%), the highest. The rows that fill withheld rates with the low or the high end of the range CDC publishes for them bound what those rates could add, and both stay inside the range. So do the rows that use national life expectancy in place of each tenth's and count each tract once.

Table 4. Changing the rate window, the set of deaths counted or the conversion moves the share of the gap between 3.9% and 6.2%

### In the clusters with the highest homicide rates, homicide accounts for a larger share

A national gap averages over a wide range of places. Grouping clusters by their own homicide rate shows the other end. The clusters fall into six groups, from those with rates up to 2 to those over 40 deaths per hundred thousand residents a year, and Table 5 compares the life expectancy of each group with that of the lowest. If the annual errors in a cluster's rate are independent, chance would raise the measured rate of a group chosen for its high rate. Each cluster is therefore placed by two of the three annual rates, and the accounting uses the third year's, scaled to the level of the 2022 to 2024 mean. Forming the highest group and measuring it on the same years would put its mean rate at 59.5 instead of the 52.0 (95% CI 46.9 to 56.1) used here.

Life expectancy is 80.2 (95% CI 80.0 to 80.4) years in the lowest-rate group, whose mean rate is 1.8 (95% CI 1.7 to 1.9), and 72.1 (95% CI 71.8 to 72.4) years in the group over 40. That group holds 3.6 million residents, 1.3% of the residents of tracts with a rate.

The difference of 8.1 (95% CI 7.7 to 8.4) years is close to the gap of 8.3 years between the least and most deprived tenths. By the accounting, homicide costs this group 1.6 (95% CI 1.4 to 1.7) years of life expectancy, and the part beyond the lowest-rate group's loss, 1.5 (95% CI 1.3 to 1.6) years, is 18.5% (95% CI 16.5% to 20.3%) of the difference.

The share is higher in each higher group, from 4.5% (95% CI 3.9% to 5.1%) in the second-lowest to 12.4% (95% CI 11.4% to 13.4%) in the second-highest and 18.5% (95% CI 16.5% to 20.3%) in the highest.

Table 5. Life expectancy is 8.1 years lower in clusters with homicide rates over 40 per hundred thousand than in clusters with rates up to 2, and homicide accounts for 18.5% of the difference

The highest group is a narrow slice and its membership shifts with the year left out. It holds between 1,165 and 1,550 tracts, in between 194 and 264 clusters and between 54 and 67 counties, depending on the split.

The 10 counties with most of its residents hold between 52.7% and 57.1% of them, and Cook County alone between 12.2% and 15.1%. The interval for its share is a bootstrap over those counties.

About 62% of its residents have a 2024 rate that is a modeled estimate and provisional. Using 2022 and 2023 alone, with one year placing a cluster and the other measuring its rate, gives 20.0% (95% CI 17.9% to 21.9%) for the highest group.

An age-specific decomposition from one place gives a figure for scale. In Washington, D.C., Roberts, Reither and Lim attribute 2.43 of the 17.23 years by which Black men's life expectancy trailed white men's for deaths in 2015 to 2017 to homicide, or 14.1%.[3] That is a gap between groups of men in one city, not a gap between kinds of tracts. Light and Vachuska find that homicide contributed more than any other cause of death to the national gap between Black and white men's life expectancy in 2020 and 2021.[4] Chapter 8.1, Gun homicide and the causes of early death (https://vishalsingh.org/zipcode/gun-homicide-and-causes-of-early-death) shows the same pairing of high rates and short lives inside Cook County.

### Within a deprivation tenth, life expectancy is lower where homicide rates are higher

Deprivation and homicide move together, so the comparison of tenths mixes them. Splitting each tenth at its median homicide rate, among its tracts with a rate, holds the deprivation rank nearly fixed. In every tenth, life expectancy is lower in the group with the higher rate (Table 6), by 0.8 (95% CI 0.5 to 1.1) years in the least deprived tenth and 2.1 (95% CI 1.7 to 2.6) in the most deprived. In the most deprived tenth the two groups' mean rates are 5.0 (95% CI 4.7 to 5.4) and 24.6 (95% CI 21.7 to 27.3) per hundred thousand.

Table 6. Within each tenth of tracts on the deprivation index, life expectancy is lower in the group with the higher homicide rate, by 0.8 years in the least deprived tenth and 2.1 in the most deprived

The accounting attributes 0.6 (95% CI 0.5 to 0.7) years of the most deprived tenth's difference to homicide, 27.5% (95% CI 22.4% to 33.8%), and 6.3% (95% CI 4.9% to 8.4%) of the least deprived tenth's. Most of each difference is left to other causes of death. The index is built from census measures of income, education, employment and housing, so tracts of the same rank can differ in conditions it does not measure, and homicide may mark them. The pattern also fits homicide being tied to other deaths, for example among residents who live with it. The data cannot separate the two.

### The homicide rate rises with deprivation under all four rankings

The tenths are again those of the life expectancy chapter, cut over its tracts that have every ranking, and the rate comes from the 66,773 of them with a usable rate outside New York City. Under the ADI baseline the rate runs from 1.8 (95% CI 1.3 to 2.2) homicide deaths per hundred thousand residents a year in the least deprived tenth to 14.3 (95% CI 12.4 to 16.3) in the most deprived, with a rank correlation of +1.00 between tenth and mean (Table 7). The standardized index, which applies the ADI's coefficients after standardizing each of its measures, gives 1.6 (95% CI 1.5 to 1.8) and 15.3 (95% CI 13.0 to 17.9) (+1.00), median household income gives 1.7 (95% CI 1.5 to 1.9) and 14.8 (95% CI 12.9 to 16.9) (+1.00), and the bachelor's degree share gives 2.3 (95% CI 2.0 to 2.7) and 10.8 (95% CI 9.6 to 12.1) (+1.00). Under the rule of Chapter 9.1, How far one ranking goes (https://vishalsingh.org/zipcode/how-far-one-ranking-goes) none of the three counts as a change, since the ends differ under each ranking and every correlation stays at or above 0.9.

Table 7. The homicide death rate is steadily worse toward the more deprived end under the ADI and under each of the three other rankings

### What these data cannot show

The accounting is not an age-specific decomposition. It applies one value of $$L$$ to every tenth, although victims' ages differ by place, and that value is a national mean for victims, most of them from 17 states, specific to each victim's race, ethnicity and sex and floored at zero. It converts a crude rate to a lifetime risk by assuming a stationary population. The share scales with $$L$$, so a reader who believes victims lose more or fewer years should read the answer off Table 3. The intervals around each share treat $$L$$ as known, and the sweep is the answer to that uncertainty. Arriaga's method needs rates by age and cause for each group, and neither file provides them below the county.

The two sources do not describe the same years. Life expectancy is the USALEEP estimate for deaths in 2010 to 2015, area-weighted onto 2020 tracts with standard errors that are good for mapping only, and the homicide rates are for 2022 to 2024. The county file's rates for 2019 are below its rates for 2022 to 2024 in every tenth, and no file here describes 2010 to 2015, so the share for those years could be lower or higher than shown. The rates are CDC's modeled estimates for clusters of neighboring tracts, repeated on each tract, which averages over neighborhoods and lowers measured concentration. That they are crude per-resident rates is inferred from the check against exact counts and not from CDC documentation. The 2024 values are provisional, and the rows that fill withheld rates from CDC's count ranges give bounds and not estimates. The 95% intervals come from a bootstrap that resamples counties, so they describe which counties are in the file and not the error of either model.

The comparisons are of places. A tract's life expectancy and its homicide rate come from different people and different decades, and a difference between a higher-rate and a lower-rate group does not show what homicide does to the people who live there. The accounting counts deaths from homicide and no other consequence of it, so it leaves out any association of violence with other deaths and with health short of death. The Area Deprivation Index understates deprivation in high-cost cities, and New York City, which holds a large share of the least deprived tenth, has published rates near zero that the accounting leaves out. Nothing in this chapter shows what reducing violence would do to life expectancy; Chapter 8.1, Gun homicide and the causes of early death (https://vishalsingh.org/zipcode/gun-homicide-and-causes-of-early-death) reviews evidence on programs aimed at violence.

## Tables

### Table 1. Homicide deaths per hundred thousand residents a year rise from 1.8 in the least deprived tenth of tracts to 14.8 in the most deprived

Mean annual homicide deaths per 100,000 residents, 2022 to 2024, by tenth of tracts on the Area Deprivation Index, tracts outside New York City with a published rate

| conf_high | conf_low | estimate | n_units | tenth |
| ---: | ---: | ---: | ---: | ---: |
| 2.16 | 1.35 | 1.76 | 6,479 | 1 |
| 3.18 | 2.07 | 2.62 | 6,827 | 2 |
| 3.48 | 2.63 | 3.03 | 7,072 | 3 |
| 4.18 | 3.29 | 3.70 | 7,220 | 4 |
| 4.69 | 3.97 | 4.33 | 7,178 | 5 |
| 5.63 | 4.68 | 5.16 | 7,602 | 6 |
| 7.09 | 5.55 | 6.27 | 7,317 | 7 |
| 8.32 | 6.48 | 7.34 | 7,495 | 8 |
| 9.59 | 7.41 | 8.43 | 7,425 | 9 |
| 16.85 | 12.76 | 14.81 | 7,610 | 10 |

Note: The homicide rate is CDC's modeled estimate for a cluster of about five neighboring tracts, repeated on each tract; a tract's rate is the mean of its cluster's annual rates for 2022 to 2024. 2024 is provisional.

Note: Homicide is deaths from homicide by any means, as CDC's dashboard defines it. Suicides of every means and deaths in legal intervention are not counted; the panel publishes legal intervention only inside all firearm deaths.

Note: New York City is left out of every rate: its tracts hold 2.6% of residents in the gradient and carry rates near zero in CDC's tract file, a mean of 0.17 per hundred thousand against 4.9 in the county file, with 75.0% of its tracts at exactly zero. A further 4.7% of residents live in clusters where CDC withholds the rate in at least one year, and 92.8% are in tracts with a rate.

Note: Intervals are 95% intervals from a bootstrap that draws whole counties with replacement (replicates and seed are in the methods note).

Note: Across all the tracts in the comparison the rate is 5.4 per hundred thousand residents a year.

Source: CDC Mapping Injury, Overdose, and Violence, tract file, homicide, 2022-2024; modeled estimates for tract clusters; population-weighted

Data: https://vishalsingh.org/zipcode/report/violence-and-the-life-expectancy-gap/figures/homicide_by_tenth.json

### Table 2. Homicide accounts for 0.39 of the 8.3 years by which life expectancy is lower in the most deprived tenth of tracts than in the least deprived

Years of life expectancy at birth below the least deprived tenth, by tenth of tracts on the Area Deprivation Index: observed for deaths in 2010 to 2015, and the part attributed to homicide by a rough accounting with homicide rates for 2022 to 2024

| conf_high | conf_low | decile | estimate | n_units | series |
| ---: | ---: | ---: | ---: | ---: | --- |
| n/a | n/a | 1 | 0 | 7,820 | All causes (observed) |
| n/a | n/a | 1 | 0 | 6,479 | Homicide (rough accounting) |
| 1.55 | 1.03 | 2 | 1.33 | 7,860 | All causes (observed) |
| 0.040 | 0.015 | 2 | 0.028 | 6,827 | Homicide (rough accounting) |
| 2.25 | 1.65 | 3 | 1.98 | 7,737 | All causes (observed) |
| 0.052 | 0.029 | 3 | 0.041 | 7,072 | Homicide (rough accounting) |
| 2.82 | 2.23 | 4 | 2.54 | 7,757 | All causes (observed) |
| 0.077 | 0.048 | 4 | 0.062 | 7,220 | Homicide (rough accounting) |
| 3.44 | 2.85 | 5 | 3.17 | 7,613 | All causes (observed) |
| 0.097 | 0.066 | 5 | 0.082 | 7,178 | Homicide (rough accounting) |
| 4.16 | 3.59 | 6 | 3.89 | 8,031 | All causes (observed) |
| 0.128 | 0.088 | 6 | 0.108 | 7,602 | Homicide (rough accounting) |
| 4.87 | 4.29 | 7 | 4.60 | 7,722 | All causes (observed) |
| 0.172 | 0.115 | 7 | 0.142 | 7,317 | Homicide (rough accounting) |
| 5.69 | 5.09 | 8 | 5.41 | 7,844 | All causes (observed) |
| 0.209 | 0.143 | 8 | 0.174 | 7,495 | Homicide (rough accounting) |
| 6.70 | 6.09 | 9 | 6.42 | 7,708 | All causes (observed) |
| 0.245 | 0.170 | 9 | 0.205 | 7,425 | Homicide (rough accounting) |
| 8.66 | 7.87 | 10 | 8.28 | 7,844 | All causes (observed) |
| 0.459 | 0.331 | 10 | 0.395 | 7,610 | Homicide (rough accounting) |

Note: Life expectancy is the USALEEP estimate for deaths in 2010 to 2015, area-weighted from 2010 onto 2020 tracts; tracts where group quarters hold a large share of residents are left out. Tenths hold equal numbers of tracts and means are weighted by residents.

Note: The accounting is a rough one, not an age-specific decomposition: it multiplies a crude homicide rate by life expectancy to get the lifetime risk of dying of homicide, and by the years of life a death takes, assumed to be 41.5 for every tenth. The rates are for 2022 to 2024 and life expectancy is for deaths in 2010 to 2015.

Note: The two lines rest on different tracts. The observed line uses every tract of each tenth, and the homicide line uses the tracts with a published rate, applying each tenth's rate to the whole tenth.

Note: New York City is left out of every rate: its tracts hold 2.6% of residents in the gradient and carry rates near zero in CDC's tract file, a mean of 0.17 per hundred thousand against 4.9 in the county file, with 75.0% of its tracts at exactly zero. A further 4.7% of residents live in clusters where CDC withholds the rate in at least one year, and 92.8% are in tracts with a rate.

Note: The homicide rate is CDC's modeled estimate for a cluster of about five neighboring tracts, repeated on each tract; a tract's rate is the mean of its cluster's annual rates for 2022 to 2024. 2024 is provisional.

Note: Intervals are 95% intervals from a bootstrap that draws whole counties with replacement (replicates and seed are in the methods note).

Source: CDC Mapping Injury, Overdose, and Violence, tract file, homicide, 2022-2024 (modeled estimates for tract clusters); USALEEP life expectancy, 2010-2015; Area Deprivation Index 2023; years of life lost per death assumed from published estimates; population-weighted

Data: https://vishalsingh.org/zipcode/report/violence-and-the-life-expectancy-gap/figures/gap_by_tenth.json

### Table 3. Homicide accounts for 3.4% of the gap if each takes 30 years of life and 6.3% if each takes 55, and for at most 8.5% when one value applies to every tenth

Share of the gap in life expectancy between the least and the most deprived tenths of tracts attributed to homicide, by the years of life assumed lost per homicide

| conf_high | conf_low | estimate | years_lost |
| ---: | ---: | ---: | ---: |
| n/a | n/a | 0 | 0 |
| 0.131 | 0.098 | 0.115 | 1 |
| 0.262 | 0.197 | 0.230 | 2 |
| 0.394 | 0.295 | 0.345 | 3 |
| 0.525 | 0.393 | 0.460 | 4 |
| 0.656 | 0.491 | 0.574 | 5 |
| 0.787 | 0.590 | 0.689 | 6 |
| 0.919 | 0.688 | 0.804 | 7 |
| 1.05 | 0.786 | 0.919 | 8 |
| 1.18 | 0.885 | 1.03 | 9 |
| 1.31 | 0.983 | 1.15 | 10 |
| 1.44 | 1.08 | 1.26 | 11 |
| 1.57 | 1.18 | 1.38 | 12 |
| 1.71 | 1.28 | 1.49 | 13 |
| 1.84 | 1.38 | 1.61 | 14 |
| 1.97 | 1.47 | 1.72 | 15 |
| 2.10 | 1.57 | 1.84 | 16 |
| 2.23 | 1.67 | 1.95 | 17 |
| 2.36 | 1.77 | 2.07 | 18 |
| 2.49 | 1.87 | 2.18 | 19 |
| 2.62 | 1.97 | 2.30 | 20 |
| 2.76 | 2.06 | 2.41 | 21 |
| 2.89 | 2.16 | 2.53 | 22 |
| 3.02 | 2.26 | 2.64 | 23 |
| 3.15 | 2.36 | 2.76 | 24 |
| 3.28 | 2.46 | 2.87 | 25 |
| 3.41 | 2.56 | 2.99 | 26 |
| 3.54 | 2.65 | 3.10 | 27 |
| 3.67 | 2.75 | 3.22 | 28 |
| 3.81 | 2.85 | 3.33 | 29 |
| 3.94 | 2.95 | 3.45 | 30 |
| 4.07 | 3.05 | 3.56 | 31 |
| 4.20 | 3.15 | 3.68 | 32 |
| 4.33 | 3.24 | 3.79 | 33 |
| 4.46 | 3.34 | 3.91 | 34 |
| 4.59 | 3.44 | 4.02 | 35 |
| 4.72 | 3.54 | 4.14 | 36 |
| 4.86 | 3.64 | 4.25 | 37 |
| 4.99 | 3.73 | 4.37 | 38 |
| 5.12 | 3.83 | 4.48 | 39 |
| 5.25 | 3.93 | 4.60 | 40 |
| 5.38 | 4.03 | 4.71 | 41 |
| 5.51 | 4.13 | 4.83 | 42 |
| 5.64 | 4.23 | 4.94 | 43 |
| 5.77 | 4.33 | 5.05 | 44 |
| 5.91 | 4.42 | 5.17 | 45 |
| 6.04 | 4.52 | 5.29 | 46 |
| 6.17 | 4.62 | 5.40 | 47 |
| 6.30 | 4.72 | 5.51 | 48 |
| 6.43 | 4.82 | 5.63 | 49 |
| 6.56 | 4.91 | 5.75 | 50 |
| 6.69 | 5.01 | 5.86 | 51 |
| 6.82 | 5.11 | 5.97 | 52 |
| 6.96 | 5.21 | 6.09 | 53 |
| 7.09 | 5.31 | 6.20 | 54 |
| 7.22 | 5.41 | 6.32 | 55 |
| 7.35 | 5.50 | 6.43 | 56 |
| 7.48 | 5.60 | 6.55 | 57 |
| 7.61 | 5.70 | 6.66 | 58 |
| 7.74 | 5.80 | 6.78 | 59 |
| 7.87 | 5.90 | 6.89 | 60 |
| 8.00 | 6.00 | 7.01 | 61 |
| 8.14 | 6.09 | 7.12 | 62 |
| 8.27 | 6.19 | 7.24 | 63 |
| 8.40 | 6.29 | 7.35 | 64 |
| 8.53 | 6.39 | 7.47 | 65 |
| 8.66 | 6.49 | 7.58 | 66 |
| 8.79 | 6.58 | 7.70 | 67 |
| 8.92 | 6.68 | 7.81 | 68 |
| 9.05 | 6.78 | 7.93 | 69 |
| 9.19 | 6.88 | 8.04 | 70 |
| 9.32 | 6.98 | 8.16 | 71 |
| 9.45 | 7.08 | 8.27 | 72 |
| 9.58 | 7.17 | 8.39 | 73 |
| 9.71 | 7.27 | 8.50 | 74 |

Note: The accounting multiplies each tenth's homicide rate for 2022 to 2024 by its life expectancy and by the years lost per death, so the share is proportional to that setting. The largest value that can apply to every tenth is the life expectancy of the most deprived tenth, which would mean that every homicide took a whole life. If every homicide in the most deprived tenth took that tenth's whole life expectancy and none in the least deprived tenth took any, which no single value gives, the share would reach 9.8%.

Note: Intervals are 95% intervals from a bootstrap that draws whole counties with replacement (replicates and seed are in the methods note).

Source: CDC Mapping Injury, Overdose, and Violence, tract file, homicide, 2022-2024 (modeled estimates for tract clusters); USALEEP life expectancy, 2010-2015; Area Deprivation Index 2023; years of life lost per death assumed from published estimates; population-weighted

Data: https://vishalsingh.org/zipcode/report/violence-and-the-life-expectancy-gap/figures/years_lost_sweep.json

### Table 4. Changing the rate window, the set of deaths counted or the conversion moves the share of the gap between 3.9% and 6.2%

Part of the gap in life expectancy between the least and the most deprived tenths of tracts attributed to homicide, in the main setting and in each variation, at the published years of life lost per death

| Setting | Years attributed to homicide | Gap (years) | Share of the gap (%) | Interval, low (%) | Interval, high (%) | Tracts with a rate | Of these, in the end tenths |
| --- | ---: | ---: | ---: | ---: | ---: | ---: | ---: |
| Main setting: homicide rates for 2022 to 2024, all tracts with a rate | 0.40 | 8.28 | 4.8 | 4.1 | 5.5 | 72,225 | 14,089 |
| Rates for 2022 and 2023 only (final data) | 0.42 | 8.28 | 5.1 | 4.4 | 5.9 | 73,581 | 14,275 |
| Withheld rates filled with the low end of CDC's count range | 0.39 | 8.28 | 4.7 | 4.0 | 5.3 | 75,862 | 14,636 |
| Withheld rates filled with the high end of CDC's count range | 0.37 | 8.28 | 4.5 | 3.8 | 5.1 | 75,862 | 14,636 |
| Rates scaled to the 2019 level, from the county file | 0.32 | 8.28 | 3.9 | 3.3 | 4.5 | 70,374 | 13,779 |
| Firearm homicide only | 0.34 | 8.28 | 4.1 | 3.5 | 4.7 | 72,285 | 14,115 |
| Homicide plus firearm suicide | 0.52 | 8.28 | 6.2 | 5.6 | 6.9 | 50,213 | 10,102 |
| National life expectancy in place of each tenth's | 0.43 | 8.28 | 5.2 | 4.4 | 5.9 | 72,225 | 14,089 |
| Rates scaled to a common share of residents aged 18 to 64 | 0.41 | 8.28 | 5.0 | 4.3 | 5.7 | 72,225 | 14,089 |
| Gap taken over the tracts that have a rate | 0.40 | 8.44 | 4.7 | 4.0 | 5.3 | 72,225 | 14,089 |
| Each tract counted once, not weighted by residents | 0.41 | 8.35 | 5.0 | 4.2 | 5.7 | 72,225 | 14,089 |
| New York City tracts at the county file's rate | 0.38 | 8.28 | 4.6 | 4.0 | 5.3 | 74,299 | 15,117 |
| Years lost scaled by each tenth's life expectancy over the national | 0.36 | 8.28 | 4.4 | 3.8 | 5.0 | 72,225 | 14,089 |

Note: Firearm suicide enters at the mean years of life lost per suicide, 29.5, and only for tracts with a published rate for both. Rates scaled to the 2019 level use the county file's ratio of the 2019 rate to the 2022 to 2024 rate in each tenth, so they describe counties and not clusters. The two rows that fill withheld rates use the low and the high end of the count range CDC publishes for them, so they bound what the withheld rates could add and are not estimates. The row for New York City gives the city's tracts the county file's rate for their borough. The row that scales the years lost multiplies them by each tenth's life expectancy over the national one. The rows that measure the gap over the tracts that have a rate and that count each tract once have gaps of their own, which the table gives.

Source: CDC Mapping Injury, Overdose, and Violence, tract file, homicide, 2022-2024 (modeled estimates for tract clusters); USALEEP life expectancy, 2010-2015; Area Deprivation Index 2023; years of life lost per death assumed from published estimates; population-weighted

Data: https://vishalsingh.org/zipcode/report/violence-and-the-life-expectancy-gap/figures/setting_table.json

### Table 5. Life expectancy is 8.1 years lower in clusters with homicide rates over 40 per hundred thousand than in clusters with rates up to 2, and homicide accounts for 18.5% of the difference

Years of life expectancy at birth below the lowest-rate group of tract clusters, by group of clusters ranked on the homicide rate for 2022 to 2024: the observed difference split into the part attributed to homicide and a remainder

| band | component | conf_high | conf_low | years |
| --- | --- | ---: | ---: | ---: |
| Over 2 to 5 | Homicide (rough accounting) | 0.055 | 0.046 | 0.051 |
| Over 2 to 5 | All other causes (remainder) | 1.25 | 0.940 | 1.09 |
| Over 5 to 10 | Homicide (rough accounting) | 0.140 | 0.127 | 0.134 |
| Over 5 to 10 | All other causes (remainder) | 2.42 | 2.03 | 2.22 |
| Over 10 to 20 | Homicide (rough accounting) | 0.334 | 0.305 | 0.319 |
| Over 10 to 20 | All other causes (remainder) | 3.90 | 3.43 | 3.67 |
| Over 20 to 40 | Homicide (rough accounting) | 0.755 | 0.675 | 0.718 |
| Over 20 to 40 | All other causes (remainder) | 5.42 | 4.74 | 5.09 |
| Over 40 | Homicide (rough accounting) | 1.62 | 1.34 | 1.50 |
| Over 40 | All other causes (remainder) | 6.98 | 6.17 | 6.58 |

Note: Clusters are ranked on the mean of two of the three annual rates, and the rate used in the accounting is the third year's, put on the level of the three-year mean by one national factor. The figure averages the three ways of leaving out a year. If the annual errors in a cluster's rate are independent, this keeps chance from inflating the rate of a group chosen for its high rate: the highest group's rate is 52.0 this way and would be 59.5 if the same years chose the group and measured it. About 62% of the highest group's residents have a 2024 rate that is a modeled, provisional estimate.

Note: The groups differ between the three ways of leaving out a year. The highest group holds 1,165 to 1,550 tracts, 194 to 264 clusters and 54 to 67 counties, and the 10 counties with most of its residents hold 52.7% to 57.1% of them, so its interval rests on few counties. Using only 2022 and 2023, one year placing a cluster and the other measuring it, the highest group's share is 20.0%.

Note: Groups hold the residents of tracts outside New York City whose cluster has a rate in every year. The lowest group, rates up to 2, has a mean rate of 1.8 and a life expectancy of 80.2 years.

Note: Life expectancy is the USALEEP estimate for deaths in 2010 to 2015, area-weighted from 2010 onto 2020 tracts; tracts where group quarters hold a large share of residents are left out. Tenths hold equal numbers of tracts and means are weighted by residents.

Note: Intervals are 95% intervals from a bootstrap that draws whole counties with replacement (replicates and seed are in the methods note).

Source: CDC Mapping Injury, Overdose, and Violence, tract file, homicide, 2022-2024 (modeled estimates for tract clusters); USALEEP life expectancy, 2010-2015; Area Deprivation Index 2023; years of life lost per death assumed from published estimates; population-weighted

Data: https://vishalsingh.org/zipcode/report/violence-and-the-life-expectancy-gap/figures/gap_by_violence_band.json

### Table 6. Within each tenth of tracts on the deprivation index, life expectancy is lower in the group with the higher homicide rate, by 0.8 years in the least deprived tenth and 2.1 in the most deprived

Population-weighted mean life expectancy at birth for deaths in 2010 to 2015 in the lower-rate and higher-rate groups of each tenth of tracts on the Area Deprivation Index, split on the homicide rate for 2022 to 2024

| hi1 | hi2 | higher | lo1 | lo2 | lower | n1 | n2 | tenth |
| ---: | ---: | ---: | ---: | ---: | ---: | ---: | ---: | --- |
| 82.98 | 82.28 | 81.97 | 82.63 | 81.73 | 82.81 | 14320000 | 12679000 | 1 (least deprived) |
| 81.64 | 80.94 | 80.57 | 81.31 | 80.29 | 81.47 | 15296000 | 15016000 | 2 |
| 80.99 | 80.19 | 79.90 | 80.73 | 79.64 | 80.86 | 16173000 | 15971000 | 3 |
| 80.42 | 79.42 | 79.22 | 80.09 | 79.01 | 80.26 | 16346000 | 16076000 | 4 |
| 79.79 | 78.81 | 78.61 | 79.50 | 78.40 | 79.64 | 15852000 | 15625000 | 5 |
| 79.18 | 77.95 | 77.74 | 78.90 | 77.55 | 79.04 | 15939000 | 15798000 | 6 |
| 78.45 | 77.25 | 77.07 | 78.16 | 76.91 | 78.30 | 14581000 | 14508000 | 7 |
| 77.77 | 76.32 | 76.15 | 77.47 | 75.98 | 77.62 | 13908000 | 13808000 | 8 |
| 76.75 | 75.34 | 75.14 | 76.45 | 74.94 | 76.59 | 12765000 | 12689000 | 9 |
| 75.45 | 73.16 | 72.90 | 74.71 | 72.68 | 75.04 | 11177000 | 11140000 | 10 (most deprived) |

Note: The groups are formed by cross-fitting, as in the figure of groups: each tenth is split on the mean of two of the three annual rates and the figure averages the three ways of leaving out a year. The n columns give residents. Each tenth is split among the tracts with a usable homicide rate, so a tenth has fewer tracts here than in the other figures.

Note: The split is on the homicide rate and the outcome is life expectancy from another source and another period, so the difference is not built by the sorting. It compares places and does not show what violence does to the people in them.

Note: Life expectancy is the USALEEP estimate for deaths in 2010 to 2015, area-weighted from 2010 onto 2020 tracts; tracts where group quarters hold a large share of residents are left out. Tenths hold equal numbers of tracts and means are weighted by residents.

Note: Intervals are 95% intervals from a bootstrap that draws whole counties with replacement (replicates and seed are in the methods note).

Source: CDC Mapping Injury, Overdose, and Violence, tract file, homicide, 2022-2024 (modeled estimates for tract clusters); USALEEP life expectancy, 2010-2015; Area Deprivation Index 2023; population-weighted

Data: https://vishalsingh.org/zipcode/report/violence-and-the-life-expectancy-gap/figures/same_deprivation.json

### Table 7. The homicide death rate is steadily worse toward the more deprived end under the ADI and under each of the three other rankings

Homicide deaths per 100,000 residents a year by tenth of tracts ranked four ways, over the 66,773 tracts that have all four rankings

| bin | conf_high | conf_low | estimate | n_units | ranking |
| ---: | ---: | ---: | ---: | ---: | --- |
| 1 | 2.19 | 1.33 | 1.77 | 6,021 | Area Deprivation Index |
| 2 | 3.16 | 2.08 | 2.61 | 6,243 | Area Deprivation Index |
| 3 | 3.50 | 2.66 | 3.08 | 6,712 | Area Deprivation Index |
| 4 | 4.16 | 3.33 | 3.73 | 6,635 | Area Deprivation Index |
| 5 | 4.68 | 3.99 | 4.33 | 6,822 | Area Deprivation Index |
| 6 | 5.55 | 4.65 | 5.08 | 6,656 | Area Deprivation Index |
| 7 | 6.96 | 5.50 | 6.15 | 6,942 | Area Deprivation Index |
| 8 | 8.01 | 6.31 | 7.09 | 6,791 | Area Deprivation Index |
| 9 | 9.35 | 7.29 | 8.24 | 6,956 | Area Deprivation Index |
| 10 | 16.28 | 12.40 | 14.34 | 6,995 | Area Deprivation Index |
| 1 | 1.83 | 1.45 | 1.63 | 6,743 | Standardized index |
| 2 | 2.68 | 2.29 | 2.49 | 6,684 | Standardized index |
| 3 | 3.36 | 2.93 | 3.15 | 6,654 | Standardized index |
| 4 | 3.82 | 3.39 | 3.60 | 6,709 | Standardized index |
| 5 | 4.45 | 3.95 | 4.20 | 6,720 | Standardized index |
| 6 | 5.05 | 4.43 | 4.74 | 6,659 | Standardized index |
| 7 | 5.97 | 5.15 | 5.55 | 6,665 | Standardized index |
| 8 | 7.40 | 6.36 | 6.85 | 6,677 | Standardized index |
| 9 | 9.99 | 8.28 | 9.08 | 6,605 | Standardized index |
| 10 | 17.88 | 13.03 | 15.35 | 6,657 | Standardized index |
| 1 | 1.94 | 1.49 | 1.70 | 6,631 | Median household income |
| 2 | 2.84 | 2.41 | 2.63 | 6,584 | Median household income |
| 3 | 3.47 | 2.99 | 3.23 | 6,603 | Median household income |
| 4 | 4.08 | 3.58 | 3.83 | 6,644 | Median household income |
| 5 | 4.72 | 4.11 | 4.41 | 6,627 | Median household income |
| 6 | 5.39 | 4.67 | 5.03 | 6,643 | Median household income |
| 7 | 6.20 | 5.24 | 5.70 | 6,695 | Median household income |
| 8 | 7.46 | 6.23 | 6.83 | 6,750 | Median household income |
| 9 | 9.89 | 8.09 | 8.96 | 6,771 | Median household income |
| 10 | 16.87 | 12.85 | 14.79 | 6,825 | Median household income |
| 1 | 2.71 | 1.99 | 2.33 | 6,499 | Bachelor's degree share |
| 2 | 3.12 | 2.55 | 2.82 | 6,623 | Bachelor's degree share |
| 3 | 3.65 | 3.07 | 3.35 | 6,587 | Bachelor's degree share |
| 4 | 4.53 | 3.79 | 4.14 | 6,595 | Bachelor's degree share |
| 5 | 5.23 | 4.42 | 4.81 | 6,631 | Bachelor's degree share |
| 6 | 5.96 | 4.95 | 5.43 | 6,699 | Bachelor's degree share |
| 7 | 6.71 | 5.49 | 6.06 | 6,727 | Bachelor's degree share |
| 8 | 7.38 | 6.16 | 6.75 | 6,775 | Bachelor's degree share |
| 9 | 9.25 | 7.37 | 8.26 | 6,791 | Bachelor's degree share |
| 10 | 12.14 | 9.55 | 10.79 | 6,846 | Bachelor's degree share |

Note: Least and most deprived tenth: ADI 1.8 and 14.3, rank correlation +1.00; standardized index 1.6 and 15.3, +1.00; median household income 1.7 and 14.8, +1.00; bachelor's degree share 2.3 and 10.8, +1.00.

Note: The standardized index applies the ADI's coefficients to its measures after standardizing each one, so home value and income carry less weight than in the ADI. Income and degree share are single census measures, each ranked from the highest value to the lowest.

Source: CDC Mapping Injury, Overdose, and Violence, tract file, homicide, 2022-2024; modeled estimates for tract clusters; population-weighted; ADI 2023 and ACS 2019-2023 tract tables (B19013, B15003 and the 15 rebuilt ADI measures)

Data: https://vishalsingh.org/zipcode/report/violence-and-the-life-expectancy-gap/figures/alt_rankings.json

## All estimates

Every fact in facts.json (111), sorted by id.

| Fact id | Value | Unit | Estimand | Population | Period | Interval | n | Weights | Source | Note |
| --- | ---: | --- | --- | --- | --- | --- | ---: | --- | --- | --- |
| alt_adi_least | 1.8 | Homicide deaths per 100,000 residents a year | population-weighted mean of the tract homicide rate (each tract carries its cluster's rate); 95% intervals from a bootstrap over counties | the least deprived tenth of tracts ranked by Area Deprivation Index, among the tracts with all four rankings | annual homicide rates for 2022, 2023 and 2024, averaged (2024 provisional); ADI 2023; ACS 2019-2023 population | 95% CI 1.3 to 2.2 | 6,021 tracts | ACS 2019-2023 tract population | CDC injury and violence mapping |   |
| alt_adi_most | 14.3 | Homicide deaths per 100,000 residents a year | population-weighted mean of the tract homicide rate (each tract carries its cluster's rate); 95% intervals from a bootstrap over counties | the most deprived tenth of tracts ranked by Area Deprivation Index, among the tracts with all four rankings | annual homicide rates for 2022, 2023 and 2024, averaged (2024 provisional); ADI 2023; ACS 2019-2023 population | 95% CI 12.4 to 16.3 | 6,995 tracts | ACS 2019-2023 tract population | CDC injury and violence mapping |   |
| alt_adi_rho | +1.00 | rank correlation | Spearman rank correlation between bin order and bin mean, each bin counted once | the bins of tracts ranked by Area Deprivation Index | annual homicide rates for 2022, 2023 and 2024, averaged (2024 provisional); ADI 2023; ACS 2019-2023 population |   |   | each bin counted once | CDC injury and violence mapping |   |
| alt_degree_least | 2.3 | Homicide deaths per 100,000 residents a year | population-weighted mean of the tract homicide rate (each tract carries its cluster's rate); 95% intervals from a bootstrap over counties | the least deprived tenth of tracts ranked by Bachelor's degree share, among the tracts with all four rankings | annual homicide rates for 2022, 2023 and 2024, averaged (2024 provisional); ADI 2023; ACS 2019-2023 population | 95% CI 2.0 to 2.7 | 6,499 tracts | ACS 2019-2023 tract population | CDC injury and violence mapping |   |
| alt_degree_most | 10.8 | Homicide deaths per 100,000 residents a year | population-weighted mean of the tract homicide rate (each tract carries its cluster's rate); 95% intervals from a bootstrap over counties | the most deprived tenth of tracts ranked by Bachelor's degree share, among the tracts with all four rankings | annual homicide rates for 2022, 2023 and 2024, averaged (2024 provisional); ADI 2023; ACS 2019-2023 population | 95% CI 9.6 to 12.1 | 6,846 tracts | ACS 2019-2023 tract population | CDC injury and violence mapping |   |
| alt_degree_rho | +1.00 | rank correlation | Spearman rank correlation between bin order and bin mean, each bin counted once | the bins of tracts ranked by Bachelor's degree share | annual homicide rates for 2022, 2023 and 2024, averaged (2024 provisional); ADI 2023; ACS 2019-2023 population |   |   | each bin counted once | CDC injury and violence mapping |   |
| alt_income_least | 1.7 | Homicide deaths per 100,000 residents a year | population-weighted mean of the tract homicide rate (each tract carries its cluster's rate); 95% intervals from a bootstrap over counties | the least deprived tenth of tracts ranked by Median household income, among the tracts with all four rankings | annual homicide rates for 2022, 2023 and 2024, averaged (2024 provisional); ADI 2023; ACS 2019-2023 population | 95% CI 1.5 to 1.9 | 6,631 tracts | ACS 2019-2023 tract population | CDC injury and violence mapping |   |
| alt_income_most | 14.8 | Homicide deaths per 100,000 residents a year | population-weighted mean of the tract homicide rate (each tract carries its cluster's rate); 95% intervals from a bootstrap over counties | the most deprived tenth of tracts ranked by Median household income, among the tracts with all four rankings | annual homicide rates for 2022, 2023 and 2024, averaged (2024 provisional); ADI 2023; ACS 2019-2023 population | 95% CI 12.9 to 16.9 | 6,825 tracts | ACS 2019-2023 tract population | CDC injury and violence mapping |   |
| alt_income_rho | +1.00 | rank correlation | Spearman rank correlation between bin order and bin mean, each bin counted once | the bins of tracts ranked by Median household income | annual homicide rates for 2022, 2023 and 2024, averaged (2024 provisional); ADI 2023; ACS 2019-2023 population |   |   | each bin counted once | CDC injury and violence mapping |   |
| alt_n_places | 66,773 | tracts | count of tracts in the common sample of the ranking comparison | tracts with the outcome and all four rankings | annual homicide rates for 2022, 2023 and 2024, averaged (2024 provisional); ADI 2023; ACS 2019-2023 population |   |   | none (full count) | CDC injury and violence mapping |   |
| alt_rho_cut | 0.9 | absolute rank correlation | threshold between a steady and an uneven gradient | the rule for changed gradients (docs/RANKING_SWITCH_RULE.md) | annual homicide rates for 2022, 2023 and 2024, averaged (2024 provisional); ADI 2023; ACS 2019-2023 population |   |   | none (a fixed rule) | CDC injury and violence mapping |   |
| alt_rho_margin | 0.1 | absolute rank correlation | smallest move in the rank correlation that counts as a change of shape | the rule for changed gradients (docs/RANKING_SWITCH_RULE.md) | annual homicide rates for 2022, 2023 and 2024, averaged (2024 provisional); ADI 2023; ACS 2019-2023 population |   |   | none (a fixed rule) | CDC injury and violence mapping |   |
| alt_std_least | 1.6 | Homicide deaths per 100,000 residents a year | population-weighted mean of the tract homicide rate (each tract carries its cluster's rate); 95% intervals from a bootstrap over counties | the least deprived tenth of tracts ranked by Standardized index, among the tracts with all four rankings | annual homicide rates for 2022, 2023 and 2024, averaged (2024 provisional); ADI 2023; ACS 2019-2023 population | 95% CI 1.5 to 1.8 | 6,743 tracts | ACS 2019-2023 tract population | CDC injury and violence mapping |   |
| alt_std_most | 15.3 | Homicide deaths per 100,000 residents a year | population-weighted mean of the tract homicide rate (each tract carries its cluster's rate); 95% intervals from a bootstrap over counties | the most deprived tenth of tracts ranked by Standardized index, among the tracts with all four rankings | annual homicide rates for 2022, 2023 and 2024, averaged (2024 provisional); ADI 2023; ACS 2019-2023 population | 95% CI 13.0 to 17.9 | 6,657 tracts | ACS 2019-2023 tract population | CDC injury and violence mapping |   |
| alt_std_rho | +1.00 | rank correlation | Spearman rank correlation between bin order and bin mean, each bin counted once | the bins of tracts ranked by Standardized index | annual homicide rates for 2022, 2023 and 2024, averaged (2024 provisional); ADI 2023; ACS 2019-2023 population |   |   | each bin counted once | CDC injury and violence mapping |   |
| band_cut_ref | 2 | homicide deaths per 100,000 residents a year | upper limit of the lowest group of clusters, the comparison group (definition) | clusters of tracts | annual homicide rates for 2022, 2023 and 2024, averaged (2024 provisional); ADI 2023; ACS 2019-2023 population |   |   | none (definition) | CDC injury and violence mapping |   |
| band_cut_top | 40 | homicide deaths per 100,000 residents a year | cut point between the two highest groups of clusters (definition) | clusters of tracts | annual homicide rates for 2022, 2023 and 2024, averaged (2024 provisional); ADI 2023; ACS 2019-2023 population |   |   | none (definition) | CDC injury and violence mapping |   |
| band_low_share | 4.5% | percent of the difference in life expectancy at birth between the lowest and the second lowest group | part of the difference attributed to homicide by rough accounting over the whole difference, bootstrap interval over counties (draws shared) | residents of tracts in the second lowest group of clusters by homicide rate | homicide rates for 2022 to 2024 with life expectancy for deaths in 2010 to 2015 | 95% CI 3.9% to 5.1% | 19,808 tracts | ACS 2019-2023 tract population | CDC injury and violence mapping |   |
| band_mid_share | 12.4% | percent of the difference in life expectancy at birth between the lowest and the second highest group | part of the difference attributed to homicide by rough accounting over the whole difference, bootstrap interval over counties (draws shared) | residents of tracts in the second highest group of clusters by homicide rate | homicide rates for 2022 to 2024 with life expectancy for deaths in 2010 to 2015 | 95% CI 11.4% to 13.4% | 2,913 tracts | ACS 2019-2023 tract population | CDC injury and violence mapping |   |
| band_ref_le | 80.2 | years of life expectancy at birth | population-weighted mean of tract life expectancy, mean over the three ways of leaving out a year; bootstrap interval over counties | residents of tracts in the lowest group of clusters by homicide rate | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population | 95% CI 80.0 to 80.4 | 26,042 tracts | ACS 2019-2023 tract population | USALEEP life expectancy | The groups differ a little between the three ways of leaving out a year; n is the mean number of tracts in the group over the three. The highest group holds 1,165 to 1,550 tracts, 194 to 264 clusters and 54 to 67 counties, so its interval rests on few counties. |
| band_ref_rate | 1.8 | homicide deaths per 100,000 residents a year | population-weighted mean rate for the year left out of forming the groups, put on the level of the three-year mean, averaged over the three ways of leaving out a year; bootstrap interval over counties | residents of tracts in the lowest group of clusters by homicide rate | annual homicide rates for 2022, 2023 and 2024, averaged (2024 provisional); ADI 2023; ACS 2019-2023 population | 95% CI 1.7 to 1.9 | 26,042 tracts | ACS 2019-2023 tract population | CDC injury and violence mapping | The groups differ a little between the three ways of leaving out a year; n is the mean number of tracts in the group over the three. The highest group holds 1,165 to 1,550 tracts, 194 to 264 clusters and 54 to 67 counties, so its interval rests on few counties. |
| band_top_clusters_max | 264 | tract clusters | clusters in the highest group; the largest over the three ways of leaving out a year | residents of tracts in the highest group of clusters by homicide rate | annual homicide rates for 2022, 2023 and 2024, averaged (2024 provisional); ADI 2023; ACS 2019-2023 population |   |   | none (full count) | CDC injury and violence mapping |   |
| band_top_clusters_min | 194 | tract clusters | clusters in the highest group; the smallest over the three ways of leaving out a year | residents of tracts in the highest group of clusters by homicide rate | annual homicide rates for 2022, 2023 and 2024, averaged (2024 provisional); ADI 2023; ACS 2019-2023 population |   |   | none (full count) | CDC injury and violence mapping |   |
| band_top_cook_max | 15.1% | percent of the residents of the highest group | share of the group's residents who live in Cook County, Illinois; the largest over the three ways of leaving out a year | residents of tracts in the highest group of clusters by homicide rate | annual homicide rates for 2022, 2023 and 2024, averaged (2024 provisional); ADI 2023; ACS 2019-2023 population |   |   | none (full count) | CDC injury and violence mapping |   |
| band_top_cook_min | 12.2% | percent of the residents of the highest group | share of the group's residents who live in Cook County, Illinois; the smallest over the three ways of leaving out a year | residents of tracts in the highest group of clusters by homicide rate | annual homicide rates for 2022, 2023 and 2024, averaged (2024 provisional); ADI 2023; ACS 2019-2023 population |   |   | none (full count) | CDC injury and violence mapping |   |
| band_top_counties_max | 67 | counties and Connecticut planning regions | counties with a tract in the highest group; the largest over the three ways of leaving out a year | residents of tracts in the highest group of clusters by homicide rate | annual homicide rates for 2022, 2023 and 2024, averaged (2024 provisional); ADI 2023; ACS 2019-2023 population |   |   | none (full count) | CDC injury and violence mapping |   |
| band_top_counties_min | 54 | counties and Connecticut planning regions | counties with a tract in the highest group; the smallest over the three ways of leaving out a year | residents of tracts in the highest group of clusters by homicide rate | annual homicide rates for 2022, 2023 and 2024, averaged (2024 provisional); ADI 2023; ACS 2019-2023 population |   |   | none (full count) | CDC injury and violence mapping |   |
| band_top_gap | 8.1 | years of life expectancy at birth below the lowest group of clusters by homicide rate | difference between the population-weighted mean life expectancy of the lowest and the highest group, bootstrap interval over counties (draws shared) | residents of tracts in the highest group of clusters by homicide rate | homicide rates for 2022 to 2024 with life expectancy for deaths in 2010 to 2015 | 95% CI 7.7 to 8.4 | 1,337 tracts | ACS 2019-2023 tract population | CDC injury and violence mapping | The groups differ a little between the three ways of leaving out a year; n is the mean number of tracts in the group over the three. The highest group holds 1,165 to 1,550 tracts, 194 to 264 clusters and 54 to 67 counties, so its interval rests on few counties. |
| band_top_le | 72.1 | years of life expectancy at birth | population-weighted mean of tract life expectancy, mean over the three ways of leaving out a year; bootstrap interval over counties | residents of tracts in the highest group of clusters by homicide rate | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population | 95% CI 71.8 to 72.4 | 1,337 tracts | ACS 2019-2023 tract population | USALEEP life expectancy | The groups differ a little between the three ways of leaving out a year; n is the mean number of tracts in the group over the three. The highest group holds 1,165 to 1,550 tracts, 194 to 264 clusters and 54 to 67 counties, so its interval rests on few counties. |
| band_top_loss | 1.6 | years of life expectancy at birth lost to homicide, by rough accounting | rough accounting: homicide rate x life expectancy at birth x assumed years of life lost per death, tenth of tracts by ADI, bootstrap interval over counties (draws shared with the gap); groups formed by cross-fitting | residents of tracts in the highest group of clusters by homicide rate | homicide rates for 2022 to 2024 with life expectancy for deaths in 2010 to 2015 | 95% CI 1.4 to 1.7 | 1,337 tracts | ACS 2019-2023 tract population | CDC injury and violence mapping | The groups differ a little between the three ways of leaving out a year; n is the mean number of tracts in the group over the three. The highest group holds 1,165 to 1,550 tracts, 194 to 264 clusters and 54 to 67 counties, so its interval rests on few counties. |
| band_top_modeled | 62% | percent of the residents of the highest group | share of the group's residents whose 2024 cluster rate is a CDC modeled estimate (1 to 9 deaths) and not a direct rate | residents of tracts in the highest group of clusters by homicide rate when 2022 and 2023 place the cluster | 2024 (provisional) |   |   | none (full count) | CDC injury and violence mapping |   |
| band_top_pop | 3.6 million | residents | residents in the highest group, mean over the three ways of leaving out a year (cross-fitting) | residents of tracts in the highest group of clusters by homicide rate | annual homicide rates for 2022, 2023 and 2024, averaged (2024 provisional); ADI 2023; ACS 2019-2023 population |   |   | none (full count) | CDC injury and violence mapping |   |
| band_top_pop_share | 1.3% | percent of residents of tracts with a usable homicide rate | share of residents in the highest group, mean over the three ways of leaving out a year | residents of those tracts that sit in a CDC cluster with a published homicide rate in each year from 2022 to 2024, outside New York City | annual homicide rates for 2022, 2023 and 2024, averaged (2024 provisional); ADI 2023; ACS 2019-2023 population |   |   | none (full count) | CDC injury and violence mapping |   |
| band_top_rate | 52.0 | homicide deaths per 100,000 residents a year | population-weighted mean rate for the year left out of forming the groups, put on the level of the three-year mean, averaged over the three ways of leaving out a year; bootstrap interval over counties | residents of tracts in the highest group of clusters by homicide rate | annual homicide rates for 2022, 2023 and 2024, averaged (2024 provisional); ADI 2023; ACS 2019-2023 population | 95% CI 46.9 to 56.1 | 1,337 tracts | ACS 2019-2023 tract population | CDC injury and violence mapping | The groups differ a little between the three ways of leaving out a year; n is the mean number of tracts in the group over the three. The highest group holds 1,165 to 1,550 tracts, 194 to 264 clusters and 54 to 67 counties, so its interval rests on few counties. |
| band_top_rate_naive | 59.5 | homicide deaths per 100,000 residents a year | population-weighted mean of the three-year rate in the group formed on the same three-year rate (not used in the accounting) | residents of tracts in the highest group of clusters by homicide rate | annual homicide rates for 2022, 2023 and 2024, averaged (2024 provisional); ADI 2023; ACS 2019-2023 population |   | 1,358 tracts | ACS 2019-2023 tract population | CDC injury and violence mapping |   |
| band_top_share | 18.5% | percent of the difference in life expectancy at birth between the lowest and the highest group | part of the difference attributed to homicide by rough accounting over the whole difference, bootstrap interval over counties (draws shared) | residents of tracts in the highest group of clusters by homicide rate | homicide rates for 2022 to 2024 with life expectancy for deaths in 2010 to 2015 | 95% CI 16.5% to 20.3% | 1,337 tracts | ACS 2019-2023 tract population | CDC injury and violence mapping | The groups differ a little between the three ways of leaving out a year; n is the mean number of tracts in the group over the three. The highest group holds 1,165 to 1,550 tracts, 194 to 264 clusters and 54 to 67 counties, so its interval rests on few counties. |
| band_top_share_final | 20.0% | percent of the difference in life expectancy at birth between the lowest and the highest group | part of the difference attributed to homicide by rough accounting over the whole difference, one of 2022 and 2023 placing a cluster and the other measuring its rate, the two ways averaged; bootstrap interval over counties (draws shared) | residents of tracts in the highest group of clusters by homicide rate, groups from the 2022 and 2023 rates only | homicide rates for 2022 to 2024 with life expectancy for deaths in 2010 to 2015 | 95% CI 17.9% to 21.9% | 1,576 tracts | ACS 2019-2023 tract population | CDC injury and violence mapping |   |
| band_top_top10_max | 57.1% | percent of the residents of the highest group | share of the group's residents who live in the ten counties with most of them; the largest over the three ways of leaving out a year | residents of tracts in the highest group of clusters by homicide rate | annual homicide rates for 2022, 2023 and 2024, averaged (2024 provisional); ADI 2023; ACS 2019-2023 population |   |   | none (full count) | CDC injury and violence mapping |   |
| band_top_top10_min | 52.7% | percent of the residents of the highest group | share of the group's residents who live in the ten counties with most of them; the smallest over the three ways of leaving out a year | residents of tracts in the highest group of clusters by homicide rate | annual homicide rates for 2022, 2023 and 2024, averaged (2024 provisional); ADI 2023; ACS 2019-2023 population |   |   | none (full count) | CDC injury and violence mapping |   |
| band_top_topn | 10 | counties | number of counties with most residents of the highest group that the concentration shares count (definition) | residents of tracts in the highest group of clusters by homicide rate | annual homicide rates for 2022, 2023 and 2024, averaged (2024 provisional); ADI 2023; ACS 2019-2023 population |   |   | none (full count) | CDC injury and violence mapping |   |
| band_top_tracts_max | 1,550 | tracts | tracts in the highest group; the largest over the three ways of leaving out a year | residents of tracts in the highest group of clusters by homicide rate | annual homicide rates for 2022, 2023 and 2024, averaged (2024 provisional); ADI 2023; ACS 2019-2023 population |   |   | none (full count) | CDC injury and violence mapping |   |
| band_top_tracts_min | 1,165 | tracts | tracts in the highest group; the smallest over the three ways of leaving out a year | residents of tracts in the highest group of clusters by homicide rate | annual homicide rates for 2022, 2023 and 2024, averaged (2024 provisional); ADI 2023; ACS 2019-2023 population |   |   | none (full count) | CDC injury and violence mapping |   |
| band_top_viol | 1.5 | years of life expectancy at birth attributed to homicide by rough accounting, highest group below the lowest | rough accounting: homicide rate x life expectancy at birth x assumed years of life lost per death, tenth of tracts by ADI, bootstrap interval over counties (draws shared with the gap); groups formed by cross-fitting | residents of tracts in the highest group of clusters by homicide rate | homicide rates for 2022 to 2024 with life expectancy for deaths in 2010 to 2015 | 95% CI 1.3 to 1.6 | 1,337 tracts | ACS 2019-2023 tract population | CDC injury and violence mapping | The groups differ a little between the three ways of leaving out a year; n is the mean number of tracts in the group over the three. The highest group holds 1,165 to 1,550 tracts, 194 to 264 clusters and 54 to 67 counties, so its interval rests on few counties. |
| cover_pop | 92.8% | percent of residents of the tracts in the gradient | share of residents who live in a tract of a cluster with a published homicide rate in every year from 2022 to 2024, outside New York City | residents of non-institutional tracts in the 50 states and DC with a life expectancy estimate and an ADI rank | annual homicide rates for 2022, 2023 and 2024, averaged (2024 provisional); ADI 2023; ACS 2019-2023 population |   | 77,936 tracts | ACS 2019-2023 tract population | CDC injury and violence mapping |   |
| dc_gap_years | 17.23 | years of life expectancy at birth, Black men below white men | age-and-cause decomposition of the Black-white gap in male life expectancy at birth, Roberts, Reither and Lim (2020), table 1 | men in Washington, D.C. | deaths in 2015 to 2017 |   |   | none (published estimate) | Published estimates from cited studies |   |
| dc_hom_share | 14.1% | percent of the Black-white gap in male life expectancy at birth | age-and-cause decomposition of the Black-white gap in male life expectancy at birth, Roberts, Reither and Lim (2020), table 1 | men in Washington, D.C. | deaths in 2015 to 2017 |   |   | none (published estimate) | Published estimates from cited studies |   |
| dc_hom_years | 2.43 | years of life expectancy at birth attributed to homicide | age-and-cause decomposition of the Black-white gap in male life expectancy at birth, Roberts, Reither and Lim (2020), table 1 | men in Washington, D.C. | deaths in 2015 to 2017 |   |   | none (published estimate) | Published estimates from cited studies |   |
| fa_share | 75.1% | percent of homicide deaths implied by the cluster rates | firearm homicide deaths implied by the rates (rate x residents) over all homicide deaths implied, tracts with both rates | residents of those tracts that sit in a CDC cluster with a published homicide rate in each year from 2022 to 2024, outside New York City | annual homicide rates for 2022, 2023 and 2024, averaged (2024 provisional); ADI 2023; ACS 2019-2023 population |   | 72,225 tracts | tract residents (implied deaths) | CDC injury and violence mapping |   |
| half_diff_d1 | 0.8 | years of life expectancy at birth lower in the higher-rate group of the tenth | population-weighted mean life expectancy of the lower-rate group minus that of the higher-rate group within the tenth (split at the weighted median rate), mean over the three ways of leaving out a year; bootstrap interval over counties | residents of tracts in the least deprived tenth of tracts with a usable homicide rate | homicide rates for 2022 to 2024 with life expectancy for deaths in 2010 to 2015 | 95% CI 0.5 to 1.1 | 6,479 tracts | ACS 2019-2023 tract population | CDC injury and violence mapping | The tenth has 7,820 tracts; the two groups are formed among the 6,479 with a usable rate, and the life expectancy difference is measured over those tracts. |
| half_diff_d10 | 2.1 | years of life expectancy at birth lower in the higher-rate group of the tenth | population-weighted mean life expectancy of the lower-rate group minus that of the higher-rate group within the tenth (split at the weighted median rate), mean over the three ways of leaving out a year; bootstrap interval over counties | residents of tracts in the most deprived tenth of tracts with a usable homicide rate | homicide rates for 2022 to 2024 with life expectancy for deaths in 2010 to 2015 | 95% CI 1.7 to 2.6 | 7,610 tracts | ACS 2019-2023 tract population | CDC injury and violence mapping | The tenth has 7,844 tracts; the two groups are formed among the 7,610 with a usable rate, and the life expectancy difference is measured over those tracts. |
| half_rate_hi_d10 | 24.6 | homicide deaths per 100,000 residents a year | population-weighted mean rate for the year left out of the split, put on the level of the three-year mean, averaged over the three ways of leaving out a year; bootstrap interval over counties | residents of tracts in the higher-rate group of the most deprived tenth of tracts | annual homicide rates for 2022, 2023 and 2024, averaged (2024 provisional); ADI 2023; ACS 2019-2023 population | 95% CI 21.7 to 27.3 | 7,610 tracts | ACS 2019-2023 tract population | CDC injury and violence mapping | The tenth has 7,844 tracts; the two groups are formed among the 7,610 with a usable rate, and the life expectancy difference is measured over those tracts. |
| half_rate_lo_d10 | 5.0 | homicide deaths per 100,000 residents a year | population-weighted mean rate for the year left out of the split, put on the level of the three-year mean, averaged over the three ways of leaving out a year; bootstrap interval over counties | residents of tracts in the lower-rate group of the most deprived tenth of tracts | annual homicide rates for 2022, 2023 and 2024, averaged (2024 provisional); ADI 2023; ACS 2019-2023 population | 95% CI 4.7 to 5.4 | 7,610 tracts | ACS 2019-2023 tract population | CDC injury and violence mapping | The tenth has 7,844 tracts; the two groups are formed among the 7,610 with a usable rate, and the life expectancy difference is measured over those tracts. |
| half_share_d1 | 6.3% | percent of the difference in life expectancy at birth between the two groups of the tenth | part of the difference attributed to homicide by rough accounting over the whole difference, bootstrap interval over counties (draws shared) | residents of tracts in the least deprived tenth of tracts with a usable homicide rate | homicide rates for 2022 to 2024 with life expectancy for deaths in 2010 to 2015 | 95% CI 4.9% to 8.4% | 6,479 tracts | ACS 2019-2023 tract population | CDC injury and violence mapping | The tenth has 7,820 tracts; the two groups are formed among the 6,479 with a usable rate, and the life expectancy difference is measured over those tracts. |
| half_share_d10 | 27.5% | percent of the difference in life expectancy at birth between the two groups of the tenth | part of the difference attributed to homicide by rough accounting over the whole difference, bootstrap interval over counties (draws shared) | residents of tracts in the most deprived tenth of tracts with a usable homicide rate | homicide rates for 2022 to 2024 with life expectancy for deaths in 2010 to 2015 | 95% CI 22.4% to 33.8% | 7,610 tracts | ACS 2019-2023 tract population | CDC injury and violence mapping | The tenth has 7,844 tracts; the two groups are formed among the 7,610 with a usable rate, and the life expectancy difference is measured over those tracts. |
| half_viol_d10 | 0.6 | years of life expectancy at birth attributed to homicide by rough accounting, higher-rate group below the lower-rate group | rough accounting: homicide rate x life expectancy at birth x assumed years of life lost per death, tenth of tracts by ADI, bootstrap interval over counties (draws shared with the gap); groups formed by cross-fitting | residents of tracts in the most deprived tenth of tracts with a usable homicide rate | homicide rates for 2022 to 2024 with life expectancy for deaths in 2010 to 2015 | 95% CI 0.5 to 0.7 | 7,610 tracts | ACS 2019-2023 tract population | CDC injury and violence mapping | The tenth has 7,844 tracts; the two groups are formed among the 7,610 with a usable rate, and the life expectancy difference is measured over those tracts. |
| hom_d1 | 1.8 | homicide deaths per 100,000 residents a year | population-weighted mean of the tract homicide rate (each tract carries its cluster's rate) in a tenth of tracts by ADI, bootstrap interval over counties | residents of tracts in the least deprived tenth of tracts with a usable homicide rate | annual homicide rates for 2022, 2023 and 2024, averaged (2024 provisional); ADI 2023; ACS 2019-2023 population | 95% CI 1.3 to 2.2 | 6,479 tracts | ACS 2019-2023 tract population | CDC injury and violence mapping |   |
| hom_d10 | 14.8 | homicide deaths per 100,000 residents a year | population-weighted mean of the tract homicide rate (each tract carries its cluster's rate) in a tenth of tracts by ADI, bootstrap interval over counties | residents of tracts in the most deprived tenth of tracts with a usable homicide rate | annual homicide rates for 2022, 2023 and 2024, averaged (2024 provisional); ADI 2023; ACS 2019-2023 population | 95% CI 12.8 to 16.9 | 7,610 tracts | ACS 2019-2023 tract population | CDC injury and violence mapping |   |
| hom_d1_nyc | 2.1 | homicide deaths per 100,000 residents a year | population-weighted mean homicide rate of the least deprived tenth when New York City tracts carry the county file's rate for their borough, bootstrap interval over counties | residents of tracts in the least deprived tenth of tracts, New York City tracts at the county file's rate | annual homicide rates for 2022, 2023 and 2024, averaged (2024 provisional); ADI 2023; ACS 2019-2023 population | 95% CI 1.5 to 2.6 | 7,493 tracts | ACS 2019-2023 tract population | CDC injury and violence mapping |   |
| hom_ratio | 8.4 | ratio of the homicide rate in the most deprived tenth to that in the least deprived | ratio of population-weighted mean homicide rates, most deprived tenth over least deprived, bootstrap interval over counties (draws shared) | residents of the most and the least deprived tenths of tracts with a usable homicide rate | annual homicide rates for 2022, 2023 and 2024, averaged (2024 provisional); ADI 2023; ACS 2019-2023 population | 95% CI 6.5 to 11.3 | 14,089 tracts | ACS 2019-2023 tract population | CDC injury and violence mapping |   |
| hom_step_gap | 5.3 | homicide deaths per 100,000 residents a year, last step minus the next-largest step | the step from the ninth to the most deprived tenth minus the largest other step by point estimate, bootstrap interval over counties (draws shared) | residents of the tenths of tracts with a usable homicide rate | annual homicide rates for 2022, 2023 and 2024, averaged (2024 provisional); ADI 2023; ACS 2019-2023 population | 95% CI 3.6 to 7.0 | 15,035 tracts | ACS 2019-2023 tract population | CDC injury and violence mapping |   |
| hom_step_last | 6.4 | homicide deaths per 100,000 residents a year, most deprived tenth minus ninth tenth | difference between population-weighted mean homicide rates of the most deprived and the ninth tenth of tracts, bootstrap interval over counties (draws shared) | residents of the ninth and the most deprived tenths of tracts with a usable homicide rate | annual homicide rates for 2022, 2023 and 2024, averaged (2024 provisional); ADI 2023; ACS 2019-2023 population | 95% CI 4.8 to 8.0 | 15,035 tracts | ACS 2019-2023 tract population | CDC injury and violence mapping |   |
| hom_step_min | 0.4 | homicide deaths per 100,000 residents a year, smallest step between neighboring tenths among the first nine | difference between population-weighted mean homicide rates of neighboring tenths, bootstrap interval over counties (draws shared); the smallest step by point estimate | residents of two neighboring tenths of tracts with a usable homicide rate | annual homicide rates for 2022, 2023 and 2024, averaged (2024 provisional); ADI 2023; ACS 2019-2023 population | 95% CI 0.1 to 0.7 | 13,899 tracts | ACS 2019-2023 tract population | CDC injury and violence mapping |   |
| le_d1 | 82.3 | years of life expectancy at birth | population-weighted mean of tract life expectancy in a tenth of tracts (rank bins of the ADI; tracts with tied ranks fall in the same tenth), bootstrap interval over counties | residents of tracts in the least deprived tenth of tracts by ADI national percentile | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population | 95% CI 82.0 to 82.6 | 7,820 tracts | ACS 2019-2023 tract population | USALEEP life expectancy |   |
| le_d10 | 74.0 | years of life expectancy at birth | population-weighted mean of tract life expectancy in a tenth of tracts (rank bins of the ADI; tracts with tied ranks fall in the same tenth), bootstrap interval over counties | residents of tracts in the most deprived tenth of tracts by ADI national percentile | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population | 95% CI 73.7 to 74.3 | 7,844 tracts | ACS 2019-2023 tract population | USALEEP life expectancy |   |
| le_decile_drop | 8.3 | years of life expectancy at birth lower in the most deprived tenth of tracts | difference between the population-weighted mean life expectancy of the least and the most deprived tenth of tracts, bootstrap interval over counties (draws shared by the two means) | residents of non-institutional tracts in the 50 states and DC with a life expectancy estimate and an ADI rank | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population | 95% CI 7.9 to 8.7 | 7,844 tracts | ACS 2019-2023 tract population | USALEEP life expectancy |   |
| le_step_last | 1.9 | years of life expectancy at birth, ninth tenth minus tenth tenth of tracts by ADI | difference between the population-weighted mean life expectancy of the ninth tenth and the most deprived tenth, bootstrap interval over counties (draws shared) | residents of non-institutional tracts in the 50 states and DC with a life expectancy estimate and an ADI rank | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population | 95% CI 1.6 to 2.1 | 7,844 tracts | ACS 2019-2023 tract population | USALEEP life expectancy |   |
| n_counties | 3,104 | counties and Connecticut planning regions | count of counties with a tract in the gradient (the units resampled in every bootstrap) | residents of non-institutional tracts in the 50 states and DC with a life expectancy estimate and an ADI rank | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population |   |   | none (full count) | USALEEP life expectancy |   |
| n_gradient_tracts | 77,936 | tracts | count of tracts in the gradient | residents of non-institutional tracts in the 50 states and DC with a life expectancy estimate and an ADI rank | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population |   |   | none (full count) | USALEEP life expectancy |   |
| n_rate_tracts | 72,225 | tracts | count of tracts with a usable homicide rate | residents of those tracts that sit in a CDC cluster with a published homicide rate in each year from 2022 to 2024, outside New York City | annual homicide rates for 2022, 2023 and 2024, averaged (2024 provisional); ADI 2023; ACS 2019-2023 population |   |   | none (full count) | CDC injury and violence mapping |   |
| nyc_county_rate | 4.9 | homicide deaths per 100,000 residents a year | population-weighted mean of the county file's rate (mean of 2022, 2023 and 2024) for the borough of each New York City tract | residents of the New York City tracts in the gradient | annual homicide rates for 2022, 2023 and 2024, averaged (2024 provisional); ADI 2023; ACS 2019-2023 population |   | 2,074 tracts | ACS 2019-2023 tract population | CDC injury and violence mapping |   |
| nyc_pop | 2.6% | percent of residents of the tracts in the gradient | share of residents who live in New York City, whose tract rates are left out | residents of non-institutional tracts in the 50 states and DC with a life expectancy estimate and an ADI rank | annual homicide rates for 2022, 2023 and 2024, averaged (2024 provisional); ADI 2023; ACS 2019-2023 population |   | 77,936 tracts | ACS 2019-2023 tract population | CDC injury and violence mapping |   |
| nyc_tenth1 | 12.1% | percent of residents of the least deprived tenth of tracts | share of the residents of the least deprived tenth who live in New York City, whose tract rates are left out | residents of the least deprived tenth of tracts by ADI national percentile | annual homicide rates for 2022, 2023 and 2024, averaged (2024 provisional); ADI 2023; ACS 2019-2023 population |   | 7,820 tracts | ACS 2019-2023 tract population | CDC injury and violence mapping |   |
| nyc_tract_rate | 0.17 | homicide deaths per 100,000 residents a year | population-weighted mean of the three-year tract-file rate over New York City tracts with a published rate | residents of the New York City tracts in the gradient | annual homicide rates for 2022, 2023 and 2024, averaged (2024 provisional); ADI 2023; ACS 2019-2023 population |   | 2,034 tracts | ACS 2019-2023 tract population | CDC injury and violence mapping |   |
| nyc_zero_share | 75.0% | percent of New York City tracts | share of New York City tracts whose three-year mean rate in the tract file is exactly zero (each tract counted once) | residents of the New York City tracts in the gradient | annual homicide rates for 2022, 2023 and 2024, averaged (2024 provisional); ADI 2023; ACS 2019-2023 population |   | 2,034 tracts | ACS 2019-2023 tract population | CDC injury and violence mapping |   |
| other_gap_share | 95.2% | percent of the gap in life expectancy at birth between the least and the most deprived tenth | the remainder over the whole gap: the share not attributed to homicide | residents of non-institutional tracts in the 50 states and DC with a life expectancy estimate and an ADI rank | homicide rates for 2022 to 2024 with life expectancy for deaths in 2010 to 2015 | 95% CI 94.6% to 95.9% | 14,089 tracts | ACS 2019-2023 tract population | CDC injury and violence mapping | The gap uses all 15,664 tracts of the two end tenths, and the part attributed to homicide uses the 14,089 of them with a published rate; each tenth's rate is applied to the whole tenth. |
| other_gap_years | 7.9 | years of life expectancy at birth, most deprived tenth below the least deprived, from causes other than homicide | the gap in life expectancy less the part attributed to homicide by rough accounting (a remainder, all other causes) | residents of non-institutional tracts in the 50 states and DC with a life expectancy estimate and an ADI rank | homicide rates for 2022 to 2024 with life expectancy for deaths in 2010 to 2015 | 95% CI 7.5 to 8.3 | 14,089 tracts | ACS 2019-2023 tract population | CDC injury and violence mapping | The gap uses all 15,664 tracts of the two end tenths, and the part attributed to homicide uses the 14,089 of them with a published rate; each tenth's rate is applied to the whole tenth. |
| pop_d1 | 32.2 million | residents | sum of tract population in a tenth of tracts (tenths hold equal numbers of tracts, not of residents) | residents of tracts in the least deprived tenth of tracts by ADI national percentile | ACS 2019-2023; ADI 2023 |   | 7,820 tracts | none (full count) | American Community Survey |   |
| pop_d10 | 23.0 million | residents | sum of tract population in a tenth of tracts (tenths hold equal numbers of tracts, not of residents) | residents of tracts in the most deprived tenth of tracts by ADI national percentile | ACS 2019-2023; ADI 2023 |   | 7,844 tracts | none (full count) | American Community Survey |   |
| rate_basis_ratio | 0.99 | ratio of the deaths implied by the rate to CDC's exact count | median over clusters of the annual rate x ACS residents of the cluster / 100,000, divided by CDC's exact count of deaths (a value near one means the rates are crude per-resident rates) | tract clusters outside New York City with an exact 2022 count of deaths at the publication threshold or above | 2022 |   | 229 clusters | none (median over clusters) | CDC injury and violence mapping |   |
| risk_d1 | 0.14% | percent of newborns who would die of homicide, stationary-population approximation | homicide rate per resident per year x life expectancy at birth of the tenth (crude rate times life expectancy equals the lifetime risk in a stationary population), bootstrap interval over counties | residents of tracts in the least deprived tenth of tracts with a usable homicide rate | homicide rates for 2022 to 2024 with life expectancy for deaths in 2010 to 2015 | 95% CI 0.11% to 0.18% | 6,479 tracts | ACS 2019-2023 tract population | CDC injury and violence mapping |   |
| risk_d10 | 1.10% | percent of newborns who would die of homicide, stationary-population approximation | homicide rate per resident per year x life expectancy at birth of the tenth (crude rate times life expectancy equals the lifetime risk in a stationary population), bootstrap interval over counties | residents of tracts in the most deprived tenth of tracts with a usable homicide rate | homicide rates for 2022 to 2024 with life expectancy for deaths in 2010 to 2015 | 95% CI 0.95% to 1.24% | 7,610 tracts | ACS 2019-2023 tract population | CDC injury and violence mapping |   |
| share_2019 | 3.9% | percent of the gap in life expectancy at birth between the least and the most deprived tenth | rough accounting at the published years of life lost, rates scaled to the 2019 level from the county file, bootstrap interval over counties | residents of non-institutional tracts in the 50 states and DC with a life expectancy estimate and an ADI rank | homicide rates for 2022 to 2024 with life expectancy for deaths in 2010 to 2015 | 95% CI 3.3% to 4.5% | 13,779 tracts | ACS 2019-2023 tract population | CDC injury and violence mapping | n counts the tracts of the two end tenths that carry a rate in this setting; the gap in the main setting uses all 15,664 tracts of those tenths. |
| share_agemix | 5.0% | percent of the gap in life expectancy at birth between the least and the most deprived tenth | rough accounting at the published years of life lost, rates scaled to a common share of residents aged 18 to 64, bootstrap interval over counties | residents of non-institutional tracts in the 50 states and DC with a life expectancy estimate and an ADI rank | homicide rates for 2022 to 2024 with life expectancy for deaths in 2010 to 2015 | 95% CI 4.3% to 5.7% | 14,089 tracts | ACS 2019-2023 tract population | CDC injury and violence mapping | n counts the tracts of the two end tenths that carry a rate in this setting; the gap in the main setting uses all 15,664 tracts of those tenths. |
| share_bound | 8.5% | percent of the gap in life expectancy at birth between the least and the most deprived tenth | rough accounting with years of life lost per homicide set at the largest possible setting: the life expectancy of the most deprived tenth, bootstrap interval over counties | residents of non-institutional tracts in the 50 states and DC with a life expectancy estimate and an ADI rank | homicide rates for 2022 to 2024 with life expectancy for deaths in 2010 to 2015 | 95% CI 7.3% to 9.7% | 14,089 tracts | ACS 2019-2023 tract population | CDC injury and violence mapping | The gap uses all 15,664 tracts of the two end tenths, and the part attributed to homicide uses the 14,089 of them with a published rate; each tenth's rate is applied to the whole tenth. |
| share_bound_tenth | 9.8% | percent of the gap in life expectancy at birth between the least and the most deprived tenth | rough accounting with the years lost set at the most deprived tenth's whole life expectancy there and at zero in the least deprived tenth (the largest share any tenth-specific setting could give), bootstrap interval over counties | residents of non-institutional tracts in the 50 states and DC with a life expectancy estimate and an ADI rank | homicide rates for 2022 to 2024 with life expectancy for deaths in 2010 to 2015 | 95% CI 8.6% to 11.0% | 14,089 tracts | ACS 2019-2023 tract population | CDC injury and violence mapping | The gap uses all 15,664 tracts of the two end tenths, and the part attributed to homicide uses the 14,089 of them with a published rate; each tenth's rate is applied to the whole tenth. |
| share_final | 5.1% | percent of the gap in life expectancy at birth between the least and the most deprived tenth | rough accounting at the published years of life lost, rates for 2022 and 2023 only, bootstrap interval over counties | residents of non-institutional tracts in the 50 states and DC with a life expectancy estimate and an ADI rank | homicide rates for 2022 to 2024 with life expectancy for deaths in 2010 to 2015 | 95% CI 4.4% to 5.9% | 14,275 tracts | ACS 2019-2023 tract population | CDC injury and violence mapping | n counts the tracts of the two end tenths that carry a rate in this setting; the gap in the main setting uses all 15,664 tracts of those tenths. |
| share_firearm | 4.1% | percent of the gap in life expectancy at birth between the least and the most deprived tenth | rough accounting at the published years of life lost, firearm homicide only, bootstrap interval over counties | residents of non-institutional tracts in the 50 states and DC with a life expectancy estimate and an ADI rank | homicide rates for 2022 to 2024 with life expectancy for deaths in 2010 to 2015 | 95% CI 3.4% to 4.7% | 14,115 tracts | ACS 2019-2023 tract population | CDC injury and violence mapping | n counts the tracts of the two end tenths that carry a rate in this setting; the gap in the main setting uses all 15,664 tracts of those tenths. |
| share_fsuicide | 6.2% | percent of the gap in life expectancy at birth between the least and the most deprived tenth | rough accounting at the published years of life lost, homicide plus firearm suicide, bootstrap interval over counties | residents of non-institutional tracts in the 50 states and DC with a life expectancy estimate and an ADI rank | homicide rates for 2022 to 2024 with life expectancy for deaths in 2010 to 2015 | 95% CI 5.6% to 6.9% | 10,102 tracts | ACS 2019-2023 tract population | CDC injury and violence mapping | n counts the tracts of the two end tenths that carry a rate in this setting; the gap in the main setting uses all 15,664 tracts of those tenths. |
| share_high | 6.3% | percent of the gap in life expectancy at birth between the least and the most deprived tenth | rough accounting with years of life lost per homicide set at the higher round setting, bootstrap interval over counties | residents of non-institutional tracts in the 50 states and DC with a life expectancy estimate and an ADI rank | homicide rates for 2022 to 2024 with life expectancy for deaths in 2010 to 2015 | 95% CI 5.4% to 7.2% | 14,089 tracts | ACS 2019-2023 tract population | CDC injury and violence mapping | The gap uses all 15,664 tracts of the two end tenths, and the part attributed to homicide uses the 14,089 of them with a published rate; each tenth's rate is applied to the whole tenth. |
| share_low | 3.4% | percent of the gap in life expectancy at birth between the least and the most deprived tenth | rough accounting with years of life lost per homicide set at the lower round setting, bootstrap interval over counties | residents of non-institutional tracts in the 50 states and DC with a life expectancy estimate and an ADI rank | homicide rates for 2022 to 2024 with life expectancy for deaths in 2010 to 2015 | 95% CI 2.9% to 3.9% | 14,089 tracts | ACS 2019-2023 tract population | CDC injury and violence mapping | The gap uses all 15,664 tracts of the two end tenths, and the part attributed to homicide uses the 14,089 of them with a published rate; each tenth's rate is applied to the whole tenth. |
| share_lscale | 4.4% | percent of the gap in life expectancy at birth between the least and the most deprived tenth | rough accounting at the published years of life lost, years lost scaled by each tenth's life expectancy over the national life expectancy, bootstrap interval over counties | residents of non-institutional tracts in the 50 states and DC with a life expectancy estimate and an ADI rank | homicide rates for 2022 to 2024 with life expectancy for deaths in 2010 to 2015 | 95% CI 3.8% to 5.0% | 14,089 tracts | ACS 2019-2023 tract population | CDC injury and violence mapping | n counts the tracts of the two end tenths that carry a rate in this setting; the gap in the main setting uses all 15,664 tracts of those tenths. |
| share_nyc | 4.6% | percent of the gap in life expectancy at birth between the least and the most deprived tenth | rough accounting at the published years of life lost, New York City tracts at the county file's rate for their borough, bootstrap interval over counties | residents of non-institutional tracts in the 50 states and DC with a life expectancy estimate and an ADI rank | homicide rates for 2022 to 2024 with life expectancy for deaths in 2010 to 2015 | 95% CI 4.0% to 5.3% | 15,117 tracts | ACS 2019-2023 tract population | CDC injury and violence mapping | n counts the tracts of the two end tenths that carry a rate in this setting; the gap in the main setting uses all 15,664 tracts of those tenths. |
| share_same_tracts | 4.7% | percent of the gap in life expectancy at birth between the least and the most deprived tenth | rough accounting at the published years of life lost, the gap measured over the tracts that have a rate, bootstrap interval over counties | residents of non-institutional tracts in the 50 states and DC with a life expectancy estimate and an ADI rank | homicide rates for 2022 to 2024 with life expectancy for deaths in 2010 to 2015 | 95% CI 4.0% to 5.3% | 14,089 tracts | ACS 2019-2023 tract population | CDC injury and violence mapping | n counts the tracts of the two end tenths that carry a rate in this setting; the gap in the main setting uses all 15,664 tracts of those tenths. |
| us_homicide_tract | 5.4 | homicide deaths per 100,000 residents a year | population-weighted mean of the tract homicide rate (each tract carries its cluster's rate) over all tracts of the gradient with a rate, bootstrap interval over counties | residents of all tracts with a usable homicide rate, outside New York City | annual homicide rates for 2022, 2023 and 2024, averaged (2024 provisional); ADI 2023; ACS 2019-2023 population | 95% CI 4.9 to 6.0 | 72,225 tracts | ACS 2019-2023 tract population | CDC injury and violence mapping |   |
| variant_max | 6.2% | percent of the gap in life expectancy at birth between the least and the most deprived tenth | largest share over the settings of the table (point estimates) | residents of non-institutional tracts in the 50 states and DC with a life expectancy estimate and an ADI rank | homicide rates for 2022 to 2024 with life expectancy for deaths in 2010 to 2015 |   |   | ACS 2019-2023 tract population | CDC injury and violence mapping |   |
| variant_min | 3.9% | percent of the gap in life expectancy at birth between the least and the most deprived tenth | smallest share over the settings of the table (point estimates) | residents of non-institutional tracts in the 50 states and DC with a life expectancy estimate and an ADI rank | homicide rates for 2022 to 2024 with life expectancy for deaths in 2010 to 2015 |   |   | ACS 2019-2023 tract population | CDC injury and violence mapping |   |
| viol_gap_share | 4.8% | percent of the gap in life expectancy at birth between the least and the most deprived tenth | rough accounting: the part of the gap attributed to homicide over the whole gap, bootstrap interval over counties (draws shared) | residents of non-institutional tracts in the 50 states and DC with a life expectancy estimate and an ADI rank | homicide rates for 2022 to 2024 with life expectancy for deaths in 2010 to 2015 | 95% CI 4.1% to 5.4% | 14,089 tracts | ACS 2019-2023 tract population | CDC injury and violence mapping | The gap uses all 15,664 tracts of the two end tenths, and the part attributed to homicide uses the 14,089 of them with a published rate; each tenth's rate is applied to the whole tenth. |
| viol_gap_years | 0.39 | years of life expectancy at birth, most deprived tenth below the least deprived, attributed to homicide by rough accounting | rough accounting: homicide rate x life expectancy at birth x assumed years of life lost per death, tenth of tracts by ADI, bootstrap interval over counties (draws shared with the gap) | residents of the most and the least deprived tenths of tracts | homicide rates for 2022 to 2024 with life expectancy for deaths in 2010 to 2015 | 95% CI 0.33 to 0.46 | 14,089 tracts | ACS 2019-2023 tract population | CDC injury and violence mapping | The gap uses all 15,664 tracts of the two end tenths, and the part attributed to homicide uses the 14,089 of them with a published rate; each tenth's rate is applied to the whole tenth. |
| viol_loss_all | 0.18 | years of life expectancy at birth lost to homicide, by rough accounting | rough accounting: homicide rate x life expectancy at birth x assumed years of life lost per death, all tracts with a rate, bootstrap interval over counties (draws shared with the gap) | residents of those tracts that sit in a CDC cluster with a published homicide rate in each year from 2022 to 2024, outside New York City | homicide rates for 2022 to 2024 with life expectancy for deaths in 2010 to 2015 | 95% CI 0.16 to 0.20 | 72,225 tracts | ACS 2019-2023 tract population | CDC injury and violence mapping |   |
| viol_loss_d1 | 0.06 | years of life expectancy at birth lost to homicide, by rough accounting | rough accounting: homicide rate x life expectancy at birth x assumed years of life lost per death, tenth of tracts by ADI, bootstrap interval over counties (draws shared with the gap) | residents of tracts in the least deprived tenth of tracts with a usable homicide rate | homicide rates for 2022 to 2024 with life expectancy for deaths in 2010 to 2015 | 95% CI 0.05 to 0.07 | 6,479 tracts | ACS 2019-2023 tract population | CDC injury and violence mapping |   |
| viol_loss_d10 | 0.45 | years of life expectancy at birth lost to homicide, by rough accounting | rough accounting: homicide rate x life expectancy at birth x assumed years of life lost per death, tenth of tracts by ADI, bootstrap interval over counties (draws shared with the gap) | residents of tracts in the most deprived tenth of tracts with a usable homicide rate | homicide rates for 2022 to 2024 with life expectancy for deaths in 2010 to 2015 | 95% CI 0.39 to 0.52 | 7,610 tracts | ACS 2019-2023 tract population | CDC injury and violence mapping |   |
| viol_step_last | 0.19 | years of life expectancy at birth, ninth tenth to most deprived tenth, attributed to homicide by rough accounting | rough accounting: homicide rate x life expectancy at birth x assumed years of life lost per death, tenth of tracts by ADI, bootstrap interval over counties (draws shared with the gap) | residents of the ninth and the most deprived tenths of tracts | homicide rates for 2022 to 2024 with life expectancy for deaths in 2010 to 2015 | 95% CI 0.14 to 0.24 | 15,035 tracts | ACS 2019-2023 tract population | CDC injury and violence mapping | The step in life expectancy uses all tracts of the two tenths; the part attributed to homicide uses the tracts with a published rate. |
| viol_step_share | 10.2% | percent of the step in life expectancy between the ninth and the most deprived tenth | part of the ninth-to-tenth step attributed to homicide by rough accounting over the whole step | residents of the ninth and the most deprived tenths of tracts | homicide rates for 2022 to 2024 with life expectancy for deaths in 2010 to 2015 | 95% CI 8.0% to 12.3% | 15,035 tracts | ACS 2019-2023 tract population | CDC injury and violence mapping |   |
| wa_d1 | 61.4% | percent of residents aged 18 to 64 | population-weighted mean of the tract share of residents aged 18 to 64 (one minus the shares under 18 and 65 and over), bootstrap interval over counties | residents of tracts in the least deprived tenth of tracts | ACS 2019-2023; ADI 2023 | 95% CI 60.1% to 62.3% | 7,820 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| wa_d10 | 58.5% | percent of residents aged 18 to 64 | population-weighted mean of the tract share of residents aged 18 to 64 (one minus the shares under 18 and 65 and over), bootstrap interval over counties | residents of tracts in the most deprived tenth of tracts | ACS 2019-2023; ADI 2023 | 95% CI 58.2% to 58.7% | 7,844 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| withheld_pop | 4.7% | percent of residents of the tracts in the gradient | share of residents outside New York City whose cluster has no published homicide rate in at least one year from 2022 to 2024 | residents of non-institutional tracts in the 50 states and DC with a life expectancy estimate and an ADI rank | annual homicide rates for 2022, 2023 and 2024, averaged (2024 provisional); ADI 2023; ACS 2019-2023 population |   | 77,936 tracts | ACS 2019-2023 tract population | CDC injury and violence mapping |   |
| yll_high | 55 | years of life lost per homicide | higher round setting tried around the published mean (sensitivity) | homicide victims | homicide rates for 2022 to 2024 with life expectancy for deaths in 2010 to 2015 |   |   | none (assumed setting) | CDC injury and violence mapping |   |
| yll_low | 30 | years of life lost per homicide | lower round setting tried around the published mean (sensitivity) | homicide victims | homicide rates for 2022 to 2024 with life expectancy for deaths in 2010 to 2015 |   |   | none (assumed setting) | CDC injury and violence mapping |   |
| yll_setting | 41.5 | years of potential life lost per homicide victim | mean of life expectancy at birth (specific to the victim's race, ethnicity and sex, 2019) minus age at death, from Zimmerman, Fridel and Trovato (2024), table 1; used here as an assumed setting for every tenth of tracts | homicide victims in the National Violent Death Reporting System, 44 states and DC | deaths in 2003 to 2019 |   | 98,617 homicide victims | none (mean over victims) | Published estimates from cited studies |   |
| yll_states | 17 | states | number of states that supply most of the records of the sample (the paper's statement; the other reporting states supply the rest) | states in the National Violent Death Reporting System sample of Zimmerman, Fridel and Trovato (2024) | deaths in 2003 to 2019 |   |   | none (published count) | Published estimates from cited studies |   |
| yll_suicide | 29.5 | years of potential life lost per suicide decedent | mean of life expectancy at birth (specific to the decedent's race, ethnicity and sex, 2019) minus age at death, from Zimmerman, Fridel and Trovato (2024), table 1; used only to add firearm suicide in one check | suicide decedents in the National Violent Death Reporting System, 44 states and DC | deaths in 2003 to 2019 |   | 230,527 suicide decedents | none (mean over decedents) | Published estimates from cited studies |   |

## Definitions

- Census tract: A small area drawn by the Census Bureau to stand in for a neighborhood. Boundaries are redrawn after each decennial census.
- Crude rate: Deaths or births divided by the population, with no adjustment for age. A place with older residents has a higher crude death rate even if health at each age is the same.
- Modeled estimate: A figure produced by a statistical model instead of a direct count or measurement, usually for small areas where direct data are thin, borrowing strength from surveys, neighboring areas or other data. Neighboring areas can share errors, so single extremes call for caution.
- Area Deprivation Index (ADI): A score that ranks neighborhoods by income, education, employment and housing conditions drawn from the American Community Survey. In this book a higher national percentile means a more deprived neighborhood.
- USALEEP: The U.S. Small-area Life Expectancy Estimates Project, a National Center for Health Statistics release of life expectancy at birth for most census tracts. It covers deaths from 2010 to 2015 only and has not been updated.
- Bootstrap interval: A range for an estimate found by recomputing it on many random resamples of the data and reading the range from the results. It shows how much the estimate would vary from sample to sample.

## Methods

This chapter approximates how much of the life-expectancy gap between the most and the least deprived tenths of tracts homicide accounts for. It is a rough accounting and not an age-specific decomposition, because neither source gives age-specific death rates for a tract, and the two cover different periods (life expectancy for deaths in 2010 to 2015, homicide rates for 2022 to 2024) and come from different models. A tract enters only if CDC publishes the homicide rate of its cluster in all three years; New York City is left out, and suicide is not counted. The years of life lost per homicide are an assumption taken from a published study, and the chapter varies it.

### Tracts, periods and homicide rates

The sample, tenths, weights and gap are those of the chapter Place and the length of a life: tracts with a USALEEP life expectancy for deaths in 2010 to 2015 (area-weighted onto 2020 tracts), institutional tracts (group quarters at least 25 percent of residents) dropped, an ADI 2023 national rank, tenths of tracts with equal numbers of tracts, means weighted by ACS 2019-2023 tract population.

The homicide rate is CDC's rate for the cluster of about five tracts that holds the tract, repeated on each member tract; a tract carries the mean of its cluster's annual rates for 2022, 2023 and 2024 and is used only if CDC publishes all three, which 72,225 tracts outside New York City satisfy. New York City is left out because the file records almost no deaths there. Withheld rates are not filled in the main setting; two checks fill them from the published count range.

Explains Table 1 and Table 6.

### How the accounting approximates the share of the gap

The chapter is a rough accounting and not an age-specific decomposition. Neither source gives age-specific death rates for a tract, and the two cover different periods and come from different models, so the part of a life-expectancy gap that homicide accounts for is approximated by the product of three terms: the homicide rate, life expectancy at birth, and the mean years of life a homicide takes.

The first two give the lifetime risk of dying of homicide in a stationary population, where the crude death rate times life expectancy equals the lifetime probability of dying of that cause; the third converts the risk to years. The product is a first-order approximation to the life expectancy that would be gained if the cause were removed, and it leaves out competing risks and the age mix of each tract; it also treats a victim's life expectancy at birth minus age at death, the published measure, as the remaining life expectancy at the age of death. Arriaga's method (1984) needs death rates by age and cause for each group, which neither file provides below the county.

Explains Table 2.

### Years of life lost per homicide are an assumption

The years of life lost per homicide are an assumption: 41.5 years, the mean of life expectancy (specific to race, ethnicity and sex) minus age at death over homicide victims in the National Violent Death Reporting System, 2003 to 2019, from Zimmerman, Fridel and Trovato (2024); the chapter varies it from zero to the life expectancy of the most deprived tenth.

Explains Table 3.

### What counts as a violent death

In this chapter a violent death means a homicide by any means as the CDC injury dashboard defines it (its All_Homicide measure). The National Violent Death Reporting System, the source of the years lost, counts suicide among violent deaths and this chapter does not. Suicides and deaths in legal intervention are not included; one check adds firearm suicide.

### How uncertainty is measured

Intervals are percentile intervals from a bootstrap that draws whole counties (3,104 counties and Connecticut planning regions) with replacement, 2,000 replicates, seed 12345, shared by every statistic so that a share or a difference has one interval; the gap and its interval equal that chapter's.

### Groups formed on the homicide rate and the 2019 check

For groups formed on the homicide rate, each cluster is placed by the mean of two of the three annual rates and the rate in the accounting is the third year's, scaled by one national factor to the level of the three-year mean; the three ways of leaving out a year are averaged. The 2019 check scales each tenth's accounting by the ratio of the county file's 2019 rate to its 2022 to 2024 rate.

Explains Table 5 and Table 4.

### Four rankings of the same places

The comparison follows the book's ranking switch rule, fixed on October 28, 2026 before any of these gradients was computed. The main gradient of the chapter is recomputed under four rankings of the same places, with the outcome, period, weights, bins and intervals unchanged: the published ADI; a standardized index that applies the ADI's coefficients to its measures after standardizing each one across tracts; median household income; and the share of adults 25 and older with a bachelor's degree or more. The three alternatives come from ACS tract estimates for 2019 to 2023, the period of the ADI.

All four rankings use the same places, the 66,773 tracts that have the outcome and all four rankings, so the ADI panel can differ slightly from the chapter's main figure. Tracts enter only if they have every input of the standardized index, which leaves out tracts without a median rent, home value or mortgage cost.

A pattern counts as changed when the worse end switches sides, when the two ends differ under one ranking and not the other, or when the rank correlation of bin order and bin mean falls on the other side of 0.9 and moves by at least 0.1.

Explains Table 7.

## What these data cannot show

The accounting is not an age-specific decomposition. It applies one value of $$L$$ to every tenth, although victims' ages differ by place, and that value is a national mean for victims, most of them from 17 states, specific to each victim's race, ethnicity and sex and floored at zero. It converts a crude rate to a lifetime risk by assuming a stationary population. The share scales with $$L$$, so a reader who believes victims lose more or fewer years should read the answer off Table 3. The intervals around each share treat $$L$$ as known, and the sweep is the answer to that uncertainty. Arriaga's method needs rates by age and cause for each group, and neither file provides them below the county.

The two sources do not describe the same years. Life expectancy is the USALEEP estimate for deaths in 2010 to 2015, area-weighted onto 2020 tracts with standard errors that are good for mapping only, and the homicide rates are for 2022 to 2024. The county file's rates for 2019 are below its rates for 2022 to 2024 in every tenth, and no file here describes 2010 to 2015, so the share for those years could be lower or higher than shown. The rates are CDC's modeled estimates for clusters of neighboring tracts, repeated on each tract, which averages over neighborhoods and lowers measured concentration. That they are crude per-resident rates is inferred from the check against exact counts and not from CDC documentation. The 2024 values are provisional, and the rows that fill withheld rates from CDC's count ranges give bounds and not estimates. The 95% intervals come from a bootstrap that resamples counties, so they describe which counties are in the file and not the error of either model.

The comparisons are of places. A tract's life expectancy and its homicide rate come from different people and different decades, and a difference between a higher-rate and a lower-rate group does not show what homicide does to the people who live there. The accounting counts deaths from homicide and no other consequence of it, so it leaves out any association of violence with other deaths and with health short of death. The Area Deprivation Index understates deprivation in high-cost cities, and New York City, which holds a large share of the least deprived tenth, has published rates near zero that the accounting leaves out. Nothing in this chapter shows what reducing violence would do to life expectancy; Chapter 8.1, Gun homicide and the causes of early death (https://vishalsingh.org/zipcode/gun-homicide-and-causes-of-early-death) reviews evidence on programs aimed at violence.

## References

1. Arriaga, Eduardo E. (1984). Measuring and explaining the change in life expectancies. Demography 21(1): 83-96. https://doi.org/10.2307/2061029
2. Zimmerman, Gregory M.; Fridel, Emma E.; Trovato, Daniel (2024). Disproportionate burden of violence: Explaining racial and ethnic disparities in potential years of life lost among homicide victims, suicide decedents, and homicide-suicide perpetrators. PLOS ONE 19(2): e0297346. https://doi.org/10.1371/journal.pone.0297346
3. Roberts, Max; Reither, Eric N.; Lim, Sojung (2020). Contributors to the black-white life expectancy gap in Washington D.C.. Scientific Reports 10: 13416. https://doi.org/10.1038/s41598-020-70046-6
4. Light, Michael T.; Vachuska, Karl (2024). Increased homicide played a key role in driving Black-White disparities in life expectancy among men during the COVID-19 pandemic. PLOS ONE 19(8): e0308105. https://doi.org/10.1371/journal.pone.0308105

## Provenance

- Pipeline version: 0.1.0
- Build script: book2/chapters/violence-and-the-life-expectancy-gap/build.py (sha256 1a06e4f1501b9a6282fe6d46efe39e8c5fd7a49d408c7e21642113171e360838)
- Built on: 2026-10-09
- Site commit: c2b1ff3309cb4a10833f884da91ac0057a66ad0a

Inputs:

| File | Size (bytes) | SHA-256, first 1 MB | SHA-256, last 1 MB |
| --- | ---: | --- | --- |
| data/analytic/tract_atlas.parquet | 39,659,925 | 29d5e4d943552d54aeb6ba92aadc939d9bd15bdea553c78357dbd516757be627 | 1e10d1ecf722340f23ee2212d4999a90250f85b65fa89b39e32db06d9472e22e |
| data/geo/counties_2020.parquet | 123,034,155 | 2fce837cdc2db1f2c229f2ea453c6793338a462770307e2c64e8d1fd713ec663 | dae928bf1042f3d174d2d64d6e3ec923fa3daa40aee1bb4feb104aac2fe3c04a |
| data/geo/dim_geo.parquet | 1,180,921 | 29d5e4d943552d54aeb6ba92aadc939d9bd15bdea553c78357dbd516757be627 | e58c8c488df54ab3187f5461f179322afdc49c9bd2315404003fe1272eea32c3 |
| data/raw/acs/deprivation_components_2019_2023/tract.parquet | 13,197,793 | 5545cde2cbe96a8b56e468d65388248a241aa6bd300d3c9e325d08d3f86b11da | 5c8ec454d066988b0f1a01aaa6ed34ebfc05d162e01cf9c6bb01f9198cbaff12 |
| data/staged/adi/adi_atlas_tract2020_2023.parquet | 798,078 | 5e409bc212ff871def780fa27b4cf8b616d9da420b2b9f6455517eddcd8c1a95 | 5e409bc212ff871def780fa27b4cf8b616d9da420b2b9f6455517eddcd8c1a95 |
| data/staged/life_expectancy/usaleep_tract2020.parquet | 1,316,169 | 9c9c7f103791b75eddba74e68a89804efdf2486b6b66bb4bc02a46ba5bf8bc23 | 27317107be2465d9fd5e83271dea0cd6e915ee419e2802bedc28747d0a41d38f |
| data/staged/mortality/injury_violence_county.parquet | 722,427 | d0dfd060a3ab9eaa3a87b35d1b24b5a1e9f21af3513c10367edb651c46f40403 | d0dfd060a3ab9eaa3a87b35d1b24b5a1e9f21af3513c10367edb651c46f40403 |
| data/staged/mortality/injury_violence_tract2020.parquet | 8,059,686 | 48f2a99135501cb7c8c7af7c85361808bd998eb4ab06825759340a1befc5f4f1 | e2e5d28442025676279f349563e0eef9235b013d40007e081ba9671697d04b49 |
| data/staged/ses/acs5_tract2020_2019_2023.parquet | 7,232,424 | 5fe6f730a35225f14d5d8abfc6b9bce58b52e545f309a7eac74f1dae74d7a31c | 7b4a2475b3852d3f91804716a04566ef8ceadacec2be9fb85f5eb642078efac6 |
| data/staged/svi/svi_tract2020_2022.parquet | 17,740,415 | cb3860d8edc2aaf1f2a02211a58f7a0b80457eb6ad9a37b5ec8abb2d5da9da10 | e371a4a696d5bed05d54ad7c5b6ca2504508805b847ec415c2322be7f66ec2b2 |

Outputs:

| File | SHA-256 |
| --- | --- |
| facts.json | 4e4801a65b00f2df89da49274a1b39684d11d8a0ddd7a43db764c7e090b9ebde |
| figures/alt_rankings.json | 6bd94f9150a3728846137c9bfb01005366e66051f66f74f12d70780c2b6e1986 |
| figures/gap_by_tenth.json | e6b0c1b66fbc252b1da010c58cde39a8ea20e0741ded55f634938b4fc1f8d4e7 |
| figures/gap_by_violence_band.json | 00d5c34f399a4fcf8c87fb9ceb8a5df94a13d11e557303e11894481c2200f97a |
| figures/homicide_by_tenth.json | 70f57e2d7ce7c5f1c7ec3c733a50b8e094d8a482f41e5bd3958c36d192f3ca47 |
| figures/same_deprivation.json | dc8831e5bf215727f5e2a4155c727e975d606b0b9b90d6a4702eec85b8817c06 |
| figures/setting_table.json | 1f7839a7cb80f6f645fceb7944186d3f4f3c283bdb17c4a4aa471b2c6bde4470 |
| figures/years_lost_sweep.json | 872d92d4e8c5106802bf384cfafed29377fa236e38b6abcc3e3ed93c48951e74 |
| meta.json | 5823f0444142f8328fe76539c2ed863a80902ddc1daa57c5ffa7191d4f1109ba |
| refs.json | e86a987f426691128511d6d87fc4e32e43c80b1dd444ce04c152fc9b1c15119f |

## Machine formats

- Report (HTML): https://vishalsingh.org/zipcode/violence-and-the-life-expectancy-gap/report
- Report (Markdown): https://vishalsingh.org/zipcode/violence-and-the-life-expectancy-gap/report.md
- Every estimate (JSON): https://vishalsingh.org/zipcode/report/violence-and-the-life-expectancy-gap/facts.json
- Table 1, figure "homicide_by_tenth" (JSON): https://vishalsingh.org/zipcode/report/violence-and-the-life-expectancy-gap/figures/homicide_by_tenth.json
- Table 2, figure "gap_by_tenth" (JSON): https://vishalsingh.org/zipcode/report/violence-and-the-life-expectancy-gap/figures/gap_by_tenth.json
- Table 3, figure "years_lost_sweep" (JSON): https://vishalsingh.org/zipcode/report/violence-and-the-life-expectancy-gap/figures/years_lost_sweep.json
- Table 4, figure "setting_table" (JSON): https://vishalsingh.org/zipcode/report/violence-and-the-life-expectancy-gap/figures/setting_table.json
- Table 5, figure "gap_by_violence_band" (JSON): https://vishalsingh.org/zipcode/report/violence-and-the-life-expectancy-gap/figures/gap_by_violence_band.json
- Table 6, figure "same_deprivation" (JSON): https://vishalsingh.org/zipcode/report/violence-and-the-life-expectancy-gap/figures/same_deprivation.json
- Table 7, figure "alt_rankings" (JSON): https://vishalsingh.org/zipcode/report/violence-and-the-life-expectancy-gap/figures/alt_rankings.json
- Index of every report: https://vishalsingh.org/zipcode/llms.txt; all reports in one file: https://vishalsingh.org/zipcode/llms-full.txt

---

# How far one ranking goes: report

Chapter 9.1 of ZIP Code Destiny, by Vishal Singh (NYU Stern). Generated from the chapter's data files; the reading version is at https://vishalsingh.org/zipcode/how-far-one-ranking-goes. Table n of this report is figure n of the chapter.

## Identification

- Title: How far one ranking goes
- Label: One ranking, many outcomes
- Chapter: 9.1 (Part IX, Reading the patterns; Reading the patterns)
- Question: Across the outcomes in this book, how often does an outcome change steadily from the least to the most deprived places, and where does it change little or run the other way?
- Dek: Of 30 outcome gradients drawn from 18 chapters, 17 are steadily worse toward the more deprived end of the ranking each chapter uses, 5 are better there, and 2 show no detectable difference between the least and the most deprived places.
- Grain: neighborhoods
- Status: draft
- Published: 2026-10-28
- Updated: not revised
- Data sources:
  - Area Deprivation Index. Area Deprivation Index tenths, fifths or quarters of tracts or counties, as each chapter defines them; district income tenths for the gender gaps; years: 2023 release (version 4.0.1; built on ACS 2019-2023); weights: as in each chapter
  - Published estimates from cited studies. None in the table; the chapter reads only this book's own figures; years: as reported in each study; weights: not applicable
- Canonical URL: https://vishalsingh.org/zipcode/how-far-one-ranking-goes
- Report: https://vishalsingh.org/zipcode/how-far-one-ranking-goes/report
- Markdown: https://vishalsingh.org/zipcode/how-far-one-ranking-goes/report.md
- BibTeX key: singh2026howfaronerankinggoes

```bibtex
@misc{singh2026howfaronerankinggoes,
  author = {Singh, Vishal},
  title = {How far one ranking goes},
  year = {2026},
  url = {https://vishalsingh.org/zipcode/how-far-one-ranking-goes},
  note = {In ZIP Code Destiny. Published 28 October 2026.}
}
```

## Principal estimates

The facts the dek and the opening prose cite, in order of appearance.

| Estimand | Population | Period | Value | Interval | n | Weights | Source | Fact id |
| --- | --- | --- | ---: | --- | ---: | --- | --- | --- |
| count of rows of the pattern table that meet the stated rule | outcome gradients that the chapters of this book publish by bins of a ranking of places | as in each chapter | 30 |   |   | none (count of rows) | Area Deprivation Index | n_rows |
| count of rows of the pattern table that meet the stated rule | outcome gradients that the chapters of this book publish by bins of a ranking of places | as in each chapter | 18 |   |   | none (count of rows) | Area Deprivation Index | n_chapters |
| count of rows of the pattern table that meet the stated rule | outcome gradients that the chapters of this book publish by bins of a ranking of places | as in each chapter | 17 |   |   | none (count of rows) | Area Deprivation Index | n_steady_worse |
| count of rows of the pattern table that meet the stated rule | outcome gradients that the chapters of this book publish by bins of a ranking of places | as in each chapter | 5 |   |   | none (count of rows) | Area Deprivation Index | n_better |
| count of rows of the pattern table that meet the stated rule | outcome gradients that the chapters of this book publish by bins of a ranking of places | as in each chapter | 2 |   |   | none (count of rows) | Area Deprivation Index | n_flat |
| count of rows of the pattern table that meet the stated rule | outcome gradients that the chapters of this book publish by bins of a ranking of places | as in each chapter | 3 |   |   | none (count of rows) | Area Deprivation Index | n_uneven_worse |

## Findings

Of 30 outcome gradients that 18 chapters of this book publish by bins of a ranking of places, 17 are steadily worse toward the more deprived end, 3 are worse there but uneven along the way, 5 are better at the deprived end, and 2 show no detectable difference between the least and the most deprived places (Table 1). The common ordering of places describes much of the book, and the exceptions are concentrated in housing costs, air pollution, the use of care and the county geography of overdose, suicide and homicide.

This chapter reads those gradients side by side. Each keeps its own units, places, period and ranking. Most rows rank tracts or counties by the Area Deprivation Index, and the gender gaps in test scores rank school districts by median household income. The rows describe different places in different years, so the chapter counts patterns under a stated rule and does not compare the size of one outcome's difference with another's.

Table 1. Of 30 outcome gradients, 17 are steadily worse toward the more deprived end, 5 are better there and 2 show no detectable difference between the ends

### The rule that sorts the gradients

A gradient is a list of means, one for each tenth, fifth or quarter of places from the least to the most deprived. The chapter asks two things of each list. First, whether the two ends differ, which it answers by whether the 95% intervals of the first and last bins overlap. Second, whether the means change steadily along the way, which it answers with the rank correlation between the order of the bins and their means. A row whose ends differ is steady when that correlation is at least 0.9 in absolute value, and uneven otherwise.

The classification depends on two choices, an interval test at the ends and a threshold on the rank correlation, and a different choice would move rows near the boundary.

Method note, How a gradient is classified: Non-overlapping intervals at the two ends are a conservative test of a difference. The rank correlation is computed over the published bin means, each bin counted once. Where higher or lower values are worse for residents, the row records which end is worse; outcomes without such a reading are classified by shape alone.

Where an outcome has a direction that is worse for residents, such as a higher death rate or a lower turnout, the table records whether the deprived end is worse or better. 4 rows carry no such reading: the gender gaps in reading and math, the predicted share of women with a teen birth and emergency visits per Medicare beneficiary.

### Birth, health and civic participation follow the ranking step by step

The steady rows run across the life course. Low birthweight, preterm birth and infant mortality rise across tenths of counties. Distance to a four-year college, incarceration among children from low-income families, mortgage denials, smoking, the number of chronic conditions that are high in a tract, adult disability and the number of residents per primary care physician rise across tenths of tracts or counties. Voter turnout, census self-response, the ladder score of life evaluation and life expectancy fall (Table 2).

Table 2. Within each domain, most outcomes move in one direction along the ranking, and the exceptions differ by domain

Several of these rows are not independent of the ranking. The PLACES health estimates and the CDC injury panel are modeled partly from census measures that the deprivation index also uses, so part of their agreement with it is built in. The steady rows are a count of chapters' findings and not a test that one process links them.

### Some outcomes run the other way

5 rows are better at the deprived end. Traffic proximity falls from a mean national percentile of 74.7 (95% CI 68.0 to 79.2) in the least deprived tenth of tracts to 39.0 (95% CI 35.4 to 42.1) in the most deprived, and fine particulate matter falls with it. Both comparisons run across the whole country, and Chapter 6.1, Air pollution (https://vishalsingh.org/zipcode/air-pollution) finds that the traffic comparison reverses within counties. The share of adults with a routine checkup is 74.2 (95% CI 73.2 to 75.3) in the least deprived tenth and 77.8 (95% CI 77.3 to 78.2) in the most deprived, while the dental visit and screening tests that Chapter 7.1, Health behaviors and the supply of care (https://vishalsingh.org/zipcode/health-behaviors-and-the-supply-of-care) follows fall.

Housing costs make up the rest. Owner households are less often over the cost-burden line in more deprived tracts, and so are low-income renters, while the share of all renter households over the line shows no detectable difference between the ends. Gross rent, home value, the mortgage cost and family income are ingredients of the index (Chapter 1.1, What the deprivation index measures (https://vishalsingh.org/zipcode/anatomy-of-deprivation)), so a housing-cost ratio compared along it is sorted partly on its own parts.

### Some death rates are worse at the deprived end without rising step by step

The overdose death rate is 23.8 (95% CI 22.2 to 25.6) in the least deprived tenth of counties and 31.1 (95% CI 25.7 to 36.5) in the most deprived, yet the rank correlation across the tenths is +0.09: the middle tenths do not lie between the ends. Suicide and homicide by county are also worse at the deprived end and uneven. Homicide shows how the scale of the comparison matters. Across tenths of counties its rank correlation is +0.47, and across tenths of tracts it is +1.00 (Chapter 8.5, Violence and the life-expectancy gap (https://vishalsingh.org/zipcode/violence-and-the-life-expectancy-gap)). The two rows come from different periods and estimates, and a county average mixes tracts with very different rates.

### The most deprived places usually stand farther from the national value than the least deprived

Each end of a ranking can also be set against the country as a whole. For the 25 rows whose values have a natural zero, Table 3 divides the mean in the least and in the most deprived bin by the national value that the row's chapter publishes, the same estimate over all the places it ranks. The most deprived end lies farther from that value in 17 of them and the least deprived end in 8. The distance to a four-year college, residents per primary care physician and the county homicide rate depart most at the most deprived end, above the national value. For chronic conditions, homicide by tract, smoking and the predicted teen birth share, the least deprived end departs more, below it. Renter burden, the annual checkup and life expectancy stay close to the national value at both ends. The national value weights places by population, births or households, so it lies nearer the places that hold more of that weight, and the ratios carry no intervals.

Table 3. In 17 of 25 outcomes the most deprived places stand farther from the national value than the least deprived places do

### Under three other rankings, one main gradient changes

17 chapters that rank places by the deprivation index recomputed their main gradient under three other rankings of the same places (Table 4). The standardized index applies the deprivation index's coefficients to its measures after standardizing each one, so home value and income weigh less in it. Median household income and the share of adults with a bachelor's degree are single census measures. Each chapter kept its outcome, weights, bins and intervals and used only the places that have all four rankings, and its report shows the four panels.

Under the standardized index, 1 of the 16 counted gradients changes class and no gradient reverses. The book set its threshold before the comparison: the standardized index would replace the deprivation index if it changed at least 4 counted gradients or reversed at least 2. The deprivation index therefore stays the book's primary ranking. The gradient that changes is traffic proximity. Across the country it falls toward the more deprived end under the deprivation index, and under the standardized index the least and the most deprived tenths of tracts no longer differ, with the middle tenths below both ends (Chapter 6.1, Air pollution (https://vishalsingh.org/zipcode/air-pollution)).

The two checks move a few rows. Median household income changes 2 counted gradients, traffic proximity and the county overdose rate, whose ends differ only under income. The degree share changes 1, low birthweight, whose ends no longer differ. On the places with all four rankings, the county overdose rate shows no detectable difference between the ends under the deprivation index itself, where the table above, read from the chapter's full sample, finds a small one. Renting is recomputed and left out of the count, because rent and income are inputs to both indices; under the standardized index and under income the share of renters over the cost-burden line rises steadily toward the more deprived end.

Table 4. Under the standardized index the main gradient changes class in 1 of 16 counted chapters, below the threshold of 4 that would replace the deprivation index

### What the count leaves out

The table holds only outcomes that a chapter publishes in bins with intervals. Upward mobility, test score levels and the jobs held by residents are reported against other rankings or as scatters and are not in it. 4 rows are uneven and 20 are worse at the deprived end in total, counts that would change with a different threshold or a different interval test, and with a different ranking. The ranking comparison above covers one main gradient per chapter and not every row of the table.

### What these data cannot show

The rows come from different sources, places, periods and units. Births are ranked by county, life expectancy by tract, and the gender gaps by district, and the outcome periods differ from one row to the next and from the period of the index, 2019 to 2023. The count is not an estimate for one population and does not measure how strongly places shape any outcome. A steady row describes an ordering of places and not the experience of every resident, and the classification says nothing about why an outcome follows the ranking or departs from it.

## Tables

### Table 1. Of 30 outcome gradients, 17 are steadily worse toward the more deprived end, 5 are better there and 2 show no detectable difference between the ends

One row per outcome: the mean in the least and the most deprived bin of the ranking its chapter uses, in that chapter's units, the rank correlation between bin order and bin mean, and the pattern under the rule in the methods

| Domain | Outcome | Chapter | Least deprived bin | Most deprived bin | Rank correlation with bin order | Pattern | More deprived end | Bins | Unit | Ranking |
| --- | --- | --- | --- | --- | ---: | --- | --- | ---: | --- | --- |
| Birth and school | Low birthweight | Birthweight and preterm birth | 7.4 | 9.7 | +1.00 | Steady | Worse at the deprived end | 10 | Share of live births (%) | Decile of county deprivation rank (1 = least deprived tenth of counties, 10 = most deprived) |
| Birth and school | Preterm birth | Birthweight and preterm birth | 8.8 | 12.1 | +1.00 | Steady | Worse at the deprived end | 10 | Share of live births (%) | Decile of county deprivation rank (1 = least deprived tenth of counties, 10 = most deprived) |
| Birth and school | Infant mortality | Infant mortality | 4.0 | 8.8 | +0.95 | Steady | Worse at the deprived end | 10 | Infant deaths per 1,000 live births | Decile of county deprivation rank (1 = least deprived tenth of counties, 10 = most deprived) |
| Birth and school | Reading gap, boys minus girls | Gender gaps in reading and math | −0.87 | −0.79 | +0.39 | No detectable difference between the ends | No reading | 10 | Mean male minus female gap (grade levels) | Tenth of districts by median household income (1 = lowest, 10 = highest), read from the highest bin to the lowest |
| Birth and school | Math gap, boys minus girls | Gender gaps in reading and math | 0.08 | −0.20 | −0.93 | Steady | No reading | 10 | Mean male minus female gap (grade levels) | Tenth of districts by median household income (1 = lowest, 10 = highest), read from the highest bin to the lowest |
| Becoming an adult | Distance to a four-year college | College access, price and earnings | 6.1 | 29.3 | +1.00 | Steady | Worse at the deprived end | 10 | Miles to the nearest four-year campus, average resident | Tenth of counties by ADI national rank (1 = least deprived, 10 = most deprived) |
| Becoming an adult | Predicted teen birth share | Teen births | 13.0 | 38.5 | +1.00 | Steady | No reading | 10 | Predicted share of women with a teen birth (percent) | Tenth of places, means of the points by Area Deprivation Index national rank (1 = least deprived, 100 = most deprived) |
| Becoming an adult | Incarceration, Black children from low-income families | School discipline and incarceration | 3.8 | 6.4 | +0.99 | Steady | Worse at the deprived end | 10 | Percent incarcerated on April 1, 2010 | Tenth of tracts by deprivation rank (1 = least deprived, 10 = most deprived) |
| Becoming an adult | Incarceration, white children from low-income families | School discipline and incarceration | 1.23 | 2.59 | +1.00 | Steady | Worse at the deprived end | 10 | Percent incarcerated on April 1, 2010 | Tenth of tracts by deprivation rank (1 = least deprived, 10 = most deprived) |
| Home, work and money | Mortgage denial rate | Mortgage denials | 7.8 | 13.8 | +0.95 | Steady | Worse at the deprived end | 10 | Denied share of decided applications (%) | Tenth of places, means of the points by Area Deprivation Index national rank (1 = least deprived, 100 = most deprived) |
| Home, work and money | Renter households over the cost-burden line | Renting, affordability and eviction | 44.5 | 47.6 | −0.30 | No detectable difference between the ends | Neither | 10 | Share paying more than 30% of income (%) | Area Deprivation Index tenth of tracts (1 = least deprived, 10 = most deprived) |
| Home, work and money | Owner households over the cost-burden line | Renting, affordability and eviction | 25.6 | 19.8 | −0.88 | Uneven | Better at the deprived end | 10 | Share paying more than 30% of income (%) | Area Deprivation Index tenth of tracts (1 = least deprived, 10 = most deprived) |
| Home, work and money | Low-income renters over the cost-burden line | Renting, affordability and eviction | 75.3 | 62.5 | −1.00 | Steady | Better at the deprived end | 10 | Low-income households paying more than 30% of income (%) | Area Deprivation Index tenth of tracts (1 = least deprived, 10 = most deprived) |
| Environment and connection | Traffic proximity | Air pollution | 74.7 | 39.0 | −0.96 | Steady | Better at the deprived end | 10 | Mean national percentile of the tract (0 to 100) | Decile of the Area Deprivation Index national rank (1 = least deprived, 10 = most deprived) |
| Environment and connection | Fine particulate matter | Air pollution | 64.3 | 47.8 | −0.93 | Steady | Better at the deprived end | 10 | Mean national percentile of the tract (0 to 100) | Decile of the Area Deprivation Index national rank (1 = least deprived, 10 = most deprived) |
| Environment and connection | Voter turnout, 2020 | Civic life, loneliness and partnership | 72.1 | 56.4 | −1.00 | Steady | Worse at the deprived end | 10 | Turnout or self-response (%) | Decile of county deprivation (1 = least deprived tenth of counties, 10 = most deprived) |
| Environment and connection | Census self-response, 2020 | Civic life, loneliness and partnership | 68.6 | 47.6 | −1.00 | Steady | Worse at the deprived end | 10 | Turnout or self-response (%) | Decile of county deprivation (1 = least deprived tenth of counties, 10 = most deprived) |
| Environment and connection | Life evaluation (ladder score) | Life evaluation and daily feelings | 7.17 | 6.93 | −1.00 | Steady | Worse at the deprived end | 10 | Mean ladder score (0 to 10) | Decile of deprivation (1 = least deprived tenth of interviews, 10 = most deprived) |
| Health and care | Current smoking | Place and the length of a life | 9.1 | 22.8 | +1.00 | Steady | Worse at the deprived end | 10 | Percent of adults | Tenth of tracts by Area Deprivation Index (1 = least deprived, 10 = most deprived) |
| Health and care | Annual checkup | Place and the length of a life | 74.2 | 77.8 | +1.00 | Steady | Better at the deprived end | 10 | Percent of adults | Tenth of tracts by Area Deprivation Index (1 = least deprived, 10 = most deprived) |
| Health and care | Chronic conditions that are high | Multiple chronic conditions | 0.5 | 5.7 | +1.00 | Steady | Worse at the deprived end | 10 | Mean number of high conditions (of nine) | Tenth of tracts by ADI national percentile (1 = least deprived, 10 = most deprived) |
| Health and care | Adult disability | Disability, retirement and Medicare | 23.3 | 42.9 | +1.00 | Steady | Worse at the deprived end | 10 | Percent of adults | Tenth of tracts by Area Deprivation Index (1 = least deprived, 10 = most deprived) |
| Health and care | Residents per primary care physician | Health behaviors and the supply of care | 1,159 | 2,730 | +1.00 | Steady | Worse at the deprived end | 10 | Residents per provider | County deprivation decile (1 = least deprived, 10 = most deprived) |
| Health and care | Emergency visits per Medicare beneficiary | Disability, retirement and Medicare | 571 | 677 | +1.00 | Steady | No reading | 5 | Visits per thousand beneficiaries per year | Fifth of counties by ADI |
| Early death | Premature death rate | Gun homicide and the causes of early death | 307 | 701 | +1.00 | Steady | Worse at the deprived end | 10 | Premature deaths per 100,000 residents, age-adjusted | Decile of the county area deprivation index (1 least deprived, 10 most deprived) |
| Early death | Overdose death rate | Overdose, alcohol and suicide | 23.8 | 31.1 | +0.09 | Uneven | Worse at the deprived end | 10 | Deaths per 100,000 residents | Tenth of counties by Area Deprivation Index (1 = least deprived, 10 = most deprived) |
| Early death | Suicide death rate | Overdose, alcohol and suicide | 12.2 | 17.4 | +0.81 | Uneven | Worse at the deprived end | 10 | Deaths per 100,000 residents | Tenth of counties by Area Deprivation Index (1 = least deprived, 10 = most deprived) |
| Early death | Homicide death rate, counties | Overdose, alcohol and suicide | 4.8 | 13.6 | +0.47 | Uneven | Worse at the deprived end | 10 | Deaths per 100,000 residents | Tenth of counties by Area Deprivation Index (1 = least deprived, 10 = most deprived) |
| Early death | Homicide death rate, tracts | Violence and the life-expectancy gap | 1.8 | 14.8 | +1.00 | Steady | Worse at the deprived end | 10 | Homicide deaths per 100,000 residents a year | Tenth of tracts by Area Deprivation Index (1 = least deprived, 10 = most deprived) |
| Early death | Life expectancy at birth | Place and the length of a life | 82.3 | 74.0 | −1.00 | Steady | Worse at the deprived end | 10 | Life expectancy at birth (years) | Tenth of places, means of the points by Area Deprivation Index national percentile (1 = least deprived, 100 = most deprived) |

Note: Gender gaps in scores, the predicted teen birth share and emergency visits have no reading of better or worse and carry none.

Note: The district income tenths of the gender-gap rows run from the highest to the lowest income here, so that the first bin is the most advantaged, as in the other rows.

Note: Bins are tenths, fifths or quarters of places, as each chapter defines them; the units of the rows are named in the Unit column.

Source: Figures of the chapters named in each row, data years 1978 to 2024 as each chapter states; estimates as published there: full counts, crude rates, population-weighted means, unweighted survey means or modeled estimates.

Data: https://vishalsingh.org/zipcode/report/how-far-one-ranking-goes/figures/pattern_table.json

### Table 2. Within each domain, most outcomes move in one direction along the ranking, and the exceptions differ by domain

Mean and 95% interval for each bin of the ranking each chapter uses, one panel per outcome, each in its own units

#### Table 2.1. Birth and school

Birth and school outcomes by bin of the ranking each chapter uses

Mean and 95% interval for each bin, least deprived at the top, each panel in its own units

| bin | conf_high | conf_low | estimate | outcome |
| --- | ---: | ---: | ---: | --- |
| 1, least deprived | 7.68 | 7.18 | 7.43 | Low birthweight |
| 2 | 8.00 | 7.32 | 7.63 | Low birthweight |
| 3 | 8.36 | 7.55 | 7.96 | Low birthweight |
| 4 | 8.67 | 7.84 | 8.29 | Low birthweight |
| 5 | 8.65 | 7.97 | 8.37 | Low birthweight |
| 6 | 8.98 | 8.29 | 8.69 | Low birthweight |
| 7 | 9.50 | 8.57 | 9.05 | Low birthweight |
| 8 | 9.85 | 8.89 | 9.40 | Low birthweight |
| 9 | 10.18 | 9.00 | 9.63 | Low birthweight |
| 10, most deprived | 10.26 | 9.20 | 9.68 | Low birthweight |
| 1, least deprived | 9.01 | 8.59 | 8.82 | Preterm birth |
| 2 | 9.48 | 9.08 | 9.28 | Preterm birth |
| 3 | 10.12 | 9.47 | 9.82 | Preterm birth |
| 4 | 10.29 | 9.80 | 10.05 | Preterm birth |
| 5 | 10.53 | 9.98 | 10.27 | Preterm birth |
| 6 | 11.28 | 10.26 | 10.80 | Preterm birth |
| 7 | 11.22 | 10.56 | 10.92 | Preterm birth |
| 8 | 11.59 | 10.92 | 11.27 | Preterm birth |
| 9 | 11.82 | 11.01 | 11.42 | Preterm birth |
| 10, most deprived | 12.66 | 11.32 | 12.05 | Preterm birth |
| 1, least deprived | 4.19 | 3.83 | 4.00 | Infant mortality |
| 2 | 5.32 | 4.88 | 5.11 | Infant mortality |
| 3 | 5.97 | 5.51 | 5.78 | Infant mortality |
| 4 | 6.34 | 5.86 | 6.09 | Infant mortality |
| 5 | 7.06 | 6.37 | 6.72 | Infant mortality |
| 6 | 7.34 | 6.70 | 7.05 | Infant mortality |
| 7 | 8.12 | 7.12 | 7.65 | Infant mortality |
| 8 | 7.96 | 6.73 | 7.38 | Infant mortality |
| 9 | 7.92 | 6.49 | 7.24 | Infant mortality |
| 10, most deprived | 9.46 | 8.05 | 8.77 | Infant mortality |
| 1, least deprived | −0.781 | −0.971 | −0.867 | Reading gap, boys minus girls |
| 2 | −0.735 | −0.881 | −0.806 | Reading gap, boys minus girls |
| 3 | −0.725 | −0.854 | −0.799 | Reading gap, boys minus girls |
| 4 | −0.687 | −0.858 | −0.767 | Reading gap, boys minus girls |
| 5 | −0.692 | −0.841 | −0.760 | Reading gap, boys minus girls |
| 6 | −0.678 | −0.805 | −0.737 | Reading gap, boys minus girls |
| 7 | −0.701 | −0.855 | −0.782 | Reading gap, boys minus girls |
| 8 | −0.704 | −0.861 | −0.780 | Reading gap, boys minus girls |
| 9 | −0.718 | −0.880 | −0.795 | Reading gap, boys minus girls |
| 10, most deprived | −0.714 | −0.885 | −0.791 | Reading gap, boys minus girls |
| 1, least deprived | 0.119 | 0.038 | 0.080 | Math gap, boys minus girls |
| 2 | 0.063 | −0.032 | 0.008 | Math gap, boys minus girls |
| 3 | 0.001 | −0.101 | −0.061 | Math gap, boys minus girls |
| 4 | −0.017 | −0.114 | −0.063 | Math gap, boys minus girls |
| 5 | −0.032 | −0.111 | −0.071 | Math gap, boys minus girls |
| 6 | −0.013 | −0.086 | −0.051 | Math gap, boys minus girls |
| 7 | −0.049 | −0.154 | −0.102 | Math gap, boys minus girls |
| 8 | −0.093 | −0.196 | −0.143 | Math gap, boys minus girls |
| 9 | −0.102 | −0.190 | −0.145 | Math gap, boys minus girls |
| 10, most deprived | −0.153 | −0.259 | −0.203 | Math gap, boys minus girls |

#### Table 2.2. Becoming an adult

Becoming an adult outcomes by bin of the ranking each chapter uses

Mean and 95% interval for each bin, least deprived at the top, each panel in its own units

| bin | conf_high | conf_low | estimate | outcome |
| --- | ---: | ---: | ---: | --- |
| 1, least deprived | 6.91 | 5.27 | 6.10 | Distance to a four-year college |
| 2 | 10.37 | 7.33 | 8.53 | Distance to a four-year college |
| 3 | 12.34 | 7.70 | 9.36 | Distance to a four-year college |
| 4 | 11.67 | 7.94 | 9.54 | Distance to a four-year college |
| 5 | 20.68 | 10.77 | 14.49 | Distance to a four-year college |
| 6 | 18.94 | 12.67 | 15.25 | Distance to a four-year college |
| 7 | 21.96 | 14.73 | 17.97 | Distance to a four-year college |
| 8 | 24.30 | 18.35 | 21.28 | Distance to a four-year college |
| 9 | 29.50 | 22.45 | 25.60 | Distance to a four-year college |
| 10, most deprived | 35.02 | 24.64 | 29.34 | Distance to a four-year college |
| 1, least deprived | 14.01 | 11.85 | 12.95 | Predicted teen birth share |
| 2 | 19.34 | 16.06 | 17.73 | Predicted teen birth share |
| 3 | 21.14 | 18.97 | 20.03 | Predicted teen birth share |
| 4 | 23.01 | 21.62 | 22.31 | Predicted teen birth share |
| 5 | 25.17 | 23.94 | 24.55 | Predicted teen birth share |
| 6 | 27.15 | 25.90 | 26.54 | Predicted teen birth share |
| 7 | 29.25 | 28.03 | 28.65 | Predicted teen birth share |
| 8 | 31.26 | 29.92 | 30.57 | Predicted teen birth share |
| 9 | 33.37 | 32.12 | 32.75 | Predicted teen birth share |
| 10, most deprived | 39.27 | 37.79 | 38.54 | Predicted teen birth share |
| 1, least deprived | 4.17 | 3.40 | 3.80 | Incarceration, Black children from low-income families |
| 2 | 4.80 | 4.13 | 4.46 | Incarceration, Black children from low-income families |
| 3 | 5.08 | 4.51 | 4.79 | Incarceration, Black children from low-income families |
| 4 | 5.30 | 4.75 | 5.02 | Incarceration, Black children from low-income families |
| 5 | 5.61 | 5.08 | 5.33 | Incarceration, Black children from low-income families |
| 6 | 5.76 | 5.31 | 5.52 | Incarceration, Black children from low-income families |
| 7 | 5.74 | 5.32 | 5.52 | Incarceration, Black children from low-income families |
| 8 | 6.03 | 5.55 | 5.79 | Incarceration, Black children from low-income families |
| 9 | 6.10 | 5.59 | 5.84 | Incarceration, Black children from low-income families |
| 10, most deprived | 6.68 | 6.06 | 6.37 | Incarceration, Black children from low-income families |
| 1, least deprived | 1.41 | 1.09 | 1.23 | Incarceration, white children from low-income families |
| 2 | 1.61 | 1.33 | 1.47 | Incarceration, white children from low-income families |
| 3 | 1.75 | 1.52 | 1.64 | Incarceration, white children from low-income families |
| 4 | 2.03 | 1.80 | 1.92 | Incarceration, white children from low-income families |
| 5 | 2.23 | 1.87 | 2.05 | Incarceration, white children from low-income families |
| 6 | 2.29 | 1.97 | 2.13 | Incarceration, white children from low-income families |
| 7 | 2.39 | 2.07 | 2.23 | Incarceration, white children from low-income families |
| 8 | 2.42 | 2.10 | 2.25 | Incarceration, white children from low-income families |
| 9 | 2.45 | 2.17 | 2.32 | Incarceration, white children from low-income families |
| 10, most deprived | 2.81 | 2.40 | 2.59 | Incarceration, white children from low-income families |

#### Table 2.3. Home, work and money

Home, work and money outcomes by bin of the ranking each chapter uses

Mean and 95% interval for each bin, least deprived at the top, each panel in its own units

| bin | conf_high | conf_low | estimate | outcome |
| --- | ---: | ---: | ---: | --- |
| 1, least deprived | 8.44 | 7.17 | 7.81 | Mortgage denial rate |
| 2 | 8.23 | 7.20 | 7.69 | Mortgage denial rate |
| 3 | 8.02 | 7.25 | 7.63 | Mortgage denial rate |
| 4 | 8.62 | 7.79 | 8.19 | Mortgage denial rate |
| 5 | 9.04 | 8.14 | 8.59 | Mortgage denial rate |
| 6 | 9.63 | 8.68 | 9.18 | Mortgage denial rate |
| 7 | 9.99 | 9.10 | 9.55 | Mortgage denial rate |
| 8 | 10.51 | 9.59 | 10.04 | Mortgage denial rate |
| 9 | 11.35 | 10.54 | 10.91 | Mortgage denial rate |
| 10, most deprived | 14.33 | 13.33 | 13.81 | Mortgage denial rate |
| 1, least deprived | 46.68 | 41.91 | 44.45 | Renter households over the cost-burden line |
| 2 | 50.08 | 45 | 47.68 | Renter households over the cost-burden line |
| 3 | 48.50 | 44.76 | 46.65 | Renter households over the cost-burden line |
| 4 | 46.71 | 44.48 | 45.61 | Renter households over the cost-burden line |
| 5 | 46.16 | 44.14 | 45.17 | Renter households over the cost-burden line |
| 6 | 46.02 | 44.16 | 45.12 | Renter households over the cost-burden line |
| 7 | 45.57 | 43.59 | 44.64 | Renter households over the cost-burden line |
| 8 | 45.27 | 43.51 | 44.42 | Renter households over the cost-burden line |
| 9 | 44.95 | 43.26 | 44.13 | Renter households over the cost-burden line |
| 10, most deprived | 48.42 | 46.63 | 47.58 | Renter households over the cost-burden line |
| 1, least deprived | 27.22 | 23.74 | 25.55 | Owner households over the cost-burden line |
| 2 | 25.42 | 22.31 | 23.87 | Owner households over the cost-burden line |
| 3 | 22.64 | 20.98 | 21.79 | Owner households over the cost-burden line |
| 4 | 21.70 | 20.36 | 21.03 | Owner households over the cost-burden line |
| 5 | 20.85 | 19.59 | 20.20 | Owner households over the cost-burden line |
| 6 | 20 | 18.90 | 19.45 | Owner households over the cost-burden line |
| 7 | 19.56 | 18.45 | 18.97 | Owner households over the cost-burden line |
| 8 | 18.89 | 17.95 | 18.42 | Owner households over the cost-burden line |
| 9 | 18.74 | 17.90 | 18.31 | Owner households over the cost-burden line |
| 10, most deprived | 20.28 | 19.25 | 19.78 | Owner households over the cost-burden line |
| 1, least deprived | 78.06 | 72.60 | 75.33 | Low-income renters over the cost-burden line |
| 2 | 76.22 | 74.15 | 75.05 | Low-income renters over the cost-burden line |
| 3 | 75.27 | 72.86 | 74.01 | Low-income renters over the cost-burden line |
| 4 | 74.23 | 72.10 | 73.20 | Low-income renters over the cost-burden line |
| 5 | 72.19 | 69.90 | 71.11 | Low-income renters over the cost-burden line |
| 6 | 70.06 | 67.89 | 68.99 | Low-income renters over the cost-burden line |
| 7 | 67.66 | 65.52 | 66.62 | Low-income renters over the cost-burden line |
| 8 | 65.44 | 63.63 | 64.56 | Low-income renters over the cost-burden line |
| 9 | 63.27 | 61.66 | 62.51 | Low-income renters over the cost-burden line |
| 10, most deprived | 63.27 | 61.58 | 62.46 | Low-income renters over the cost-burden line |

#### Table 2.4. Environment and connection

Environment and connection outcomes by bin of the ranking each chapter uses

Mean and 95% interval for each bin, least deprived at the top, each panel in its own units

| bin | conf_high | conf_low | estimate | outcome |
| --- | ---: | ---: | ---: | --- |
| 1, least deprived | 79.18 | 68.05 | 74.70 | Traffic proximity |
| 2 | 71.29 | 58.96 | 65.23 | Traffic proximity |
| 3 | 60.23 | 53.49 | 56.92 | Traffic proximity |
| 4 | 53.77 | 48.87 | 51.45 | Traffic proximity |
| 5 | 50.34 | 44.93 | 47.72 | Traffic proximity |
| 6 | 47.13 | 41.66 | 44.43 | Traffic proximity |
| 7 | 44.32 | 38.17 | 41.19 | Traffic proximity |
| 8 | 40.55 | 35.39 | 38.01 | Traffic proximity |
| 9 | 37.53 | 32.20 | 34.92 | Traffic proximity |
| 10, most deprived | 42.13 | 35.42 | 38.99 | Traffic proximity |
| 1, least deprived | 73.24 | 52.34 | 64.27 | Fine particulate matter |
| 2 | 65.84 | 47.61 | 56.82 | Fine particulate matter |
| 3 | 56.99 | 47.69 | 52.29 | Fine particulate matter |
| 4 | 54.11 | 47.20 | 50.75 | Fine particulate matter |
| 5 | 51.99 | 46.09 | 49.21 | Fine particulate matter |
| 6 | 51.28 | 44.87 | 48.16 | Fine particulate matter |
| 7 | 50.70 | 44.20 | 47.40 | Fine particulate matter |
| 8 | 47.38 | 41.64 | 44.47 | Fine particulate matter |
| 9 | 46.62 | 41.50 | 44.06 | Fine particulate matter |
| 10, most deprived | 50.97 | 44.29 | 47.76 | Fine particulate matter |
| 1, least deprived | 73.80 | 70.58 | 72.06 | Voter turnout, 2020 |
| 2 | 71 | 67.93 | 69.33 | Voter turnout, 2020 |
| 3 | 68.58 | 64.62 | 66.59 | Voter turnout, 2020 |
| 4 | 65.57 | 62.07 | 63.95 | Voter turnout, 2020 |
| 5 | 64.15 | 61.43 | 62.84 | Voter turnout, 2020 |
| 6 | 62.08 | 59.10 | 60.63 | Voter turnout, 2020 |
| 7 | 60.94 | 58.84 | 59.85 | Voter turnout, 2020 |
| 8 | 60.52 | 56.90 | 58.66 | Voter turnout, 2020 |
| 9 | 59.01 | 57.18 | 58.06 | Voter turnout, 2020 |
| 10, most deprived | 57.59 | 55.37 | 56.43 | Voter turnout, 2020 |
| 1, least deprived | 69.94 | 67.39 | 68.60 | Census self-response, 2020 |
| 2 | 68.15 | 65.91 | 66.97 | Census self-response, 2020 |
| 3 | 68.04 | 65 | 66.45 | Census self-response, 2020 |
| 4 | 66.25 | 63.61 | 64.94 | Census self-response, 2020 |
| 5 | 64.80 | 62.56 | 63.80 | Census self-response, 2020 |
| 6 | 62.88 | 60.24 | 61.63 | Census self-response, 2020 |
| 7 | 61.85 | 59.72 | 60.76 | Census self-response, 2020 |
| 8 | 58.76 | 55.75 | 57.15 | Census self-response, 2020 |
| 9 | 55.21 | 53.18 | 54.20 | Census self-response, 2020 |
| 10, most deprived | 49.12 | 46.11 | 47.63 | Census self-response, 2020 |
| 1, least deprived | 7.23 | 7.11 | 7.17 | Life evaluation (ladder score) |
| 2 | 7.17 | 7.09 | 7.13 | Life evaluation (ladder score) |
| 3 | 7.15 | 7.08 | 7.11 | Life evaluation (ladder score) |
| 4 | 7.14 | 7.07 | 7.10 | Life evaluation (ladder score) |
| 5 | 7.12 | 7.07 | 7.10 | Life evaluation (ladder score) |
| 6 | 7.12 | 7.06 | 7.09 | Life evaluation (ladder score) |
| 7 | 7.08 | 7.03 | 7.06 | Life evaluation (ladder score) |
| 8 | 7.06 | 7.02 | 7.04 | Life evaluation (ladder score) |
| 9 | 7.00 | 6.96 | 6.98 | Life evaluation (ladder score) |
| 10, most deprived | 6.94 | 6.91 | 6.93 | Life evaluation (ladder score) |

#### Table 2.5. Health and care

Health and care outcomes by bin of the ranking each chapter uses

Mean and 95% interval for each bin, least deprived at the top, each panel in its own units

| bin | conf_high | conf_low | estimate | outcome |
| --- | ---: | ---: | ---: | --- |
| 1, least deprived | 9.59 | 8.53 | 9.06 | Current smoking |
| 2 | 11.25 | 10.20 | 10.74 | Current smoking |
| 3 | 11.99 | 11.28 | 11.62 | Current smoking |
| 4 | 12.88 | 12.45 | 12.67 | Current smoking |
| 5 | 14.09 | 13.70 | 13.89 | Current smoking |
| 6 | 15.28 | 14.93 | 15.10 | Current smoking |
| 7 | 16.54 | 16.25 | 16.39 | Current smoking |
| 8 | 17.94 | 17.61 | 17.78 | Current smoking |
| 9 | 19.62 | 19.27 | 19.45 | Current smoking |
| 10, most deprived | 23.14 | 22.46 | 22.81 | Current smoking |
| 1, least deprived | 75.29 | 73.23 | 74.16 | Annual checkup |
| 2 | 75.40 | 73.25 | 74.27 | Annual checkup |
| 3 | 75.35 | 74 | 74.67 | Annual checkup |
| 4 | 75.41 | 74.40 | 74.90 | Annual checkup |
| 5 | 75.89 | 75.06 | 75.46 | Annual checkup |
| 6 | 76.26 | 75.45 | 75.85 | Annual checkup |
| 7 | 76.59 | 75.81 | 76.21 | Annual checkup |
| 8 | 76.96 | 76.26 | 76.62 | Annual checkup |
| 9 | 77.44 | 76.83 | 77.14 | Annual checkup |
| 10, most deprived | 78.18 | 77.34 | 77.76 | Annual checkup |
| 1, least deprived | 0.550 | 0.380 | 0.450 | Chronic conditions that are high |
| 2 | 0.700 | 0.550 | 0.620 | Chronic conditions that are high |
| 3 | 1 | 0.780 | 0.880 | Chronic conditions that are high |
| 4 | 1.10 | 0.920 | 1.01 | Chronic conditions that are high |
| 5 | 1.39 | 1.19 | 1.29 | Chronic conditions that are high |
| 6 | 1.83 | 1.60 | 1.71 | Chronic conditions that are high |
| 7 | 2.46 | 2.22 | 2.34 | Chronic conditions that are high |
| 8 | 3.34 | 3.07 | 3.21 | Chronic conditions that are high |
| 9 | 4.49 | 4.21 | 4.35 | Chronic conditions that are high |
| 10, most deprived | 5.81 | 5.52 | 5.66 | Chronic conditions that are high |
| 1, least deprived | 24.07 | 22.30 | 23.28 | Adult disability |
| 2 | 27.45 | 24.50 | 25.97 | Adult disability |
| 3 | 27.99 | 26.08 | 27 | Adult disability |
| 4 | 28.73 | 27.60 | 28.17 | Adult disability |
| 5 | 30.32 | 29.39 | 29.84 | Adult disability |
| 6 | 32.06 | 31.21 | 31.62 | Adult disability |
| 7 | 33.88 | 33.16 | 33.53 | Adult disability |
| 8 | 35.81 | 35.15 | 35.48 | Adult disability |
| 9 | 38.39 | 37.74 | 38.06 | Adult disability |
| 10, most deprived | 43.21 | 42.50 | 42.86 | Adult disability |
| 1, least deprived | 1205 | 1093 | 1159 | Residents per primary care physician |
| 2 | 1449 | 1221 | 1335 | Residents per primary care physician |
| 3 | 1511 | 1286 | 1389 | Residents per primary care physician |
| 4 | 1512 | 1309 | 1411 | Residents per primary care physician |
| 5 | 1737 | 1365 | 1529 | Residents per primary care physician |
| 6 | 1877 | 1458 | 1656 | Residents per primary care physician |
| 7 | 2148 | 1660 | 1894 | Residents per primary care physician |
| 8 | 2249 | 1905 | 2083 | Residents per primary care physician |
| 9 | 2828 | 2394 | 2593 | Residents per primary care physician |
| 10, most deprived | 3018 | 2452 | 2730 | Residents per primary care physician |
| 1, least deprived | 584 | 555.5 | 570.5 | Emergency visits per Medicare beneficiary |
| 2 | 617.4 | 582.1 | 598.9 | Emergency visits per Medicare beneficiary |
| 3 | 635.5 | 603 | 620.1 | Emergency visits per Medicare beneficiary |
| 4 | 655.4 | 627.3 | 642.1 | Emergency visits per Medicare beneficiary |
| 5, most deprived | 693.6 | 656.4 | 676.7 | Emergency visits per Medicare beneficiary |

#### Table 2.6. Early death

Early death outcomes by bin of the ranking each chapter uses

Mean and 95% interval for each bin, least deprived at the top, each panel in its own units

| bin | conf_high | conf_low | estimate | outcome |
| --- | ---: | ---: | ---: | --- |
| 1, least deprived | 317.6 | 296.1 | 306.9 | Premature death rate |
| 2 | 387.5 | 356.6 | 372.7 | Premature death rate |
| 3 | 438.9 | 399.7 | 417.1 | Premature death rate |
| 4 | 493.3 | 439.2 | 465.4 | Premature death rate |
| 5 | 510.6 | 465.7 | 490.1 | Premature death rate |
| 6 | 536.2 | 483.2 | 505.8 | Premature death rate |
| 7 | 563.4 | 502 | 531.3 | Premature death rate |
| 8 | 588.4 | 508.8 | 543.4 | Premature death rate |
| 9 | 631.6 | 568.9 | 604 | Premature death rate |
| 10, most deprived | 739.9 | 659.7 | 701.1 | Premature death rate |
| 1, least deprived | 25.58 | 22.17 | 23.84 | Overdose death rate |
| 2 | 29.93 | 24.57 | 27.28 | Overdose death rate |
| 3 | 34.96 | 26.90 | 30.74 | Overdose death rate |
| 4 | 42.01 | 30.60 | 36.13 | Overdose death rate |
| 5 | 35.49 | 27.61 | 31.86 | Overdose death rate |
| 6 | 31.25 | 24.58 | 27.90 | Overdose death rate |
| 7 | 35.13 | 27.89 | 31.39 | Overdose death rate |
| 8 | 33.13 | 21.71 | 27.17 | Overdose death rate |
| 9 | 30.12 | 23.86 | 26.83 | Overdose death rate |
| 10, most deprived | 36.49 | 25.66 | 31.09 | Overdose death rate |
| 1, least deprived | 13.16 | 11.31 | 12.18 | Suicide death rate |
| 2 | 15.20 | 13.23 | 14.15 | Suicide death rate |
| 3 | 17.11 | 14.83 | 15.83 | Suicide death rate |
| 4 | 17.04 | 15.09 | 16.03 | Suicide death rate |
| 5 | 17.94 | 16.04 | 16.91 | Suicide death rate |
| 6 | 18.76 | 16.57 | 17.63 | Suicide death rate |
| 7 | 19.09 | 17.60 | 18.32 | Suicide death rate |
| 8 | 18.94 | 14.20 | 16.62 | Suicide death rate |
| 9 | 18.83 | 17.33 | 18.10 | Suicide death rate |
| 10, most deprived | 19.41 | 15.71 | 17.37 | Suicide death rate |
| 1, least deprived | 5.44 | 4.19 | 4.82 | Homicide death rate, counties |
| 2 | 9.17 | 5.65 | 7.31 | Homicide death rate, counties |
| 3 | 9.02 | 6.40 | 7.86 | Homicide death rate, counties |
| 4 | 14.24 | 8.53 | 11.44 | Homicide death rate, counties |
| 5 | 12.33 | 7.48 | 9.89 | Homicide death rate, counties |
| 6 | 9.99 | 6.87 | 8.45 | Homicide death rate, counties |
| 7 | 10.00 | 6.19 | 7.86 | Homicide death rate, counties |
| 8 | 7.73 | 5.66 | 6.63 | Homicide death rate, counties |
| 9 | 10.31 | 7.40 | 8.79 | Homicide death rate, counties |
| 10, most deprived | 16.17 | 11.14 | 13.63 | Homicide death rate, counties |
| 1, least deprived | 2.16 | 1.35 | 1.76 | Homicide death rate, tracts |
| 2 | 3.18 | 2.07 | 2.62 | Homicide death rate, tracts |
| 3 | 3.48 | 2.63 | 3.03 | Homicide death rate, tracts |
| 4 | 4.18 | 3.29 | 3.70 | Homicide death rate, tracts |
| 5 | 4.69 | 3.97 | 4.33 | Homicide death rate, tracts |
| 6 | 5.63 | 4.68 | 5.16 | Homicide death rate, tracts |
| 7 | 7.09 | 5.55 | 6.27 | Homicide death rate, tracts |
| 8 | 8.32 | 6.48 | 7.34 | Homicide death rate, tracts |
| 9 | 9.59 | 7.41 | 8.43 | Homicide death rate, tracts |
| 10, most deprived | 16.85 | 12.76 | 14.81 | Homicide death rate, tracts |
| 1, least deprived | 82.55 | 81.99 | 82.29 | Life expectancy at birth |
| 2 | 81.21 | 80.73 | 80.95 | Life expectancy at birth |
| 3 | 80.52 | 80.07 | 80.30 | Life expectancy at birth |
| 4 | 79.91 | 79.58 | 79.74 | Life expectancy at birth |
| 5 | 79.27 | 78.97 | 79.12 | Life expectancy at birth |
| 6 | 78.54 | 78.25 | 78.39 | Life expectancy at birth |
| 7 | 77.82 | 77.56 | 77.69 | Life expectancy at birth |
| 8 | 77.03 | 76.75 | 76.88 | Life expectancy at birth |
| 9 | 76.03 | 75.72 | 75.87 | Life expectancy at birth |
| 10, most deprived | 74.32 | 73.72 | 74.01 | Life expectancy at birth |

Source: Figures of the chapters named in each row, data years 1978 to 2024 as each chapter states; estimates as published there: full counts, crude rates, population-weighted means, unweighted survey means or modeled estimates.

Data: https://vishalsingh.org/zipcode/report/how-far-one-ranking-goes/figures/gradients_by_domain.json

### Table 3. In 17 of 25 outcomes the most deprived places stand farther from the national value than the least deprived places do

Mean at each end of the ranking each chapter uses, divided by the value for all the places of the comparison, on a log scale so that twice and half the national value sit the same distance from it

| Domain | Outcome | Least deprived end | United States | Most deprived end | Least deprived end divided by the US value | Most deprived end divided by the US value | Unit |
| --- | --- | --- | --- | --- | ---: | ---: | --- |
| Birth and school | Infant mortality | 4.0 | 5.7 | 8.8 | 0.71 | 1.55 | Infant deaths per 1,000 live births |
| Birth and school | Preterm birth | 8.8 | 10.1% | 12.1 | 0.87 | 1.19 | Share of live births (%) |
| Birth and school | Low birthweight | 7.4 | 8.3% | 9.7 | 0.89 | 1.16 | Share of live births (%) |
| Becoming an adult | Distance to a four-year college | 6.1 | 9.6 | 29.3 | 0.64 | 3.05 | Miles to the nearest four-year campus, average resident |
| Becoming an adult | Predicted teen birth share | 13.0 | 25.0% | 38.5 | 0.52 | 1.54 | Predicted share of women with a teen birth (percent) |
| Becoming an adult | Incarceration, white children from low-income families | 1.23 | 1.9% | 2.59 | 0.64 | 1.33 | Percent incarcerated on April 1, 2010 |
| Becoming an adult | Incarceration, Black children from low-income families | 3.8 | 5.2% | 6.4 | 0.73 | 1.23 | Percent incarcerated on April 1, 2010 |
| Home, work and money | Mortgage denial rate | 7.8 | 9.0% | 13.8 | 0.87 | 1.54 | Denied share of decided applications (%) |
| Home, work and money | Renter households over the cost-burden line | 44.5 | 45.5% | 47.6 | 0.98 | 1.04 | Share paying more than 30% of income (%) |
| Home, work and money | Owner households over the cost-burden line | 25.6 | 20.8% | 19.8 | 1.23 | 0.95 | Share paying more than 30% of income (%) |
| Home, work and money | Low-income renters over the cost-burden line | 75.3 | 69.0% | 62.5 | 1.09 | 0.91 | Low-income households paying more than 30% of income (%) |
| Environment and connection | Voter turnout, 2020 | 72.1 | 67.8% | 56.4 | 1.06 | 0.83 | Turnout or self-response (%) |
| Environment and connection | Census self-response, 2020 | 68.6 | 66.0% | 47.6 | 1.04 | 0.72 | Turnout or self-response (%) |
| Health and care | Chronic conditions that are high | 0.5 | 2.0 | 5.7 | 0.23 | 2.90 | Mean number of high conditions (of nine) |
| Health and care | Residents per primary care physician | 1,159 | 1,334 | 2,730 | 0.87 | 2.05 | Residents per provider |
| Health and care | Current smoking | 9.1 | 14.5% | 22.8 | 0.63 | 1.57 | Percent of adults |
| Health and care | Adult disability | 23.3 | 30.9% | 42.9 | 0.75 | 1.39 | Percent of adults |
| Health and care | Emergency visits per Medicare beneficiary | 571 | 592 | 677 | 0.96 | 1.14 | Visits per thousand beneficiaries per year |
| Health and care | Annual checkup | 74.2 | 75.6% | 77.8 | 0.98 | 1.03 | Percent of adults |
| Early death | Homicide death rate, tracts | 1.8 | 5.4 | 14.8 | 0.32 | 2.72 | Homicide deaths per 100,000 residents a year |
| Early death | Homicide death rate, counties | 4.8 | 7.1 | 13.6 | 0.68 | 1.93 | Deaths per 100,000 residents |
| Early death | Premature death rate | 307 | 392 | 701 | 0.78 | 1.79 | Premature deaths per 100,000 residents, age-adjusted |
| Early death | Suicide death rate | 12.2 | 14.3 | 17.4 | 0.85 | 1.21 | Deaths per 100,000 residents |
| Early death | Overdose death rate | 23.8 | 27.7 | 31.1 | 0.86 | 1.12 | Deaths per 100,000 residents |
| Early death | Life expectancy at birth | 82.3 | 78.8 | 74.0 | 1.04 | 0.94 | Life expectancy at birth (years) |

Note: The national value pools all the places of each comparison into one bin, with the estimator and weights of its chapter, and is published there. It covers the places that the chapter ranks, so places without a rank or an estimate are left out.

Note: Rows with no natural zero are left out: the gender gaps in scores, the traffic and fine particulate percentiles and the ladder score. They are in the pattern table.

Note: The ratios have no intervals; the intervals of each bin and of each national value are in its chapter.

Source: Figures of the chapters named in each row, data years 1978 to 2024 as each chapter states; estimates as published there: full counts, crude rates, population-weighted means, unweighted survey means or modeled estimates.

Data: https://vishalsingh.org/zipcode/report/how-far-one-ranking-goes/figures/ends_vs_us.json

### Table 4. Under the standardized index the main gradient changes class in 1 of 16 counted chapters, below the threshold of 4 that would replace the deprivation index

One row per chapter: the class of its main gradient under the ADI over the places with all four rankings, and whether each alternative ranking changes it, with the rank correlation of bin order and bin mean

| Chapter | Main gradient | Under the ADI | Standardized index | Median household income | Bachelor's degree share | Counted |
| --- | --- | --- | --- | --- | --- | --- |
| Birthweight and preterm birth | Low birthweight | Steady, worse at the deprived end (+0.99) | Unchanged (+0.99) | Unchanged (+0.96) | Ends differ under one ranking only (+0.68) | Yes |
| Infant mortality | Infant mortality | Steady, worse at the deprived end (+0.95) | Unchanged (+0.92) | Unchanged (+1.00) | Unchanged (+0.99) | Yes |
| College access, price and earnings | Distance to a four-year college | Steady, worse at the deprived end (+0.99) | Unchanged (+0.93) | Unchanged (+0.99) | Unchanged (+0.99) | Yes |
| Teen births | Predicted teen birth share | Steady (+1.00) | Unchanged (+1.00) | Unchanged (+1.00) | Unchanged (+1.00) | Yes |
| School discipline and incarceration | Incarceration, Black children from low-income families | Steady, worse at the deprived end (+1.00) | Unchanged (+0.98) | Unchanged (+1.00) | Unchanged (+1.00) | Yes |
| Mortgage denials | Mortgage denial rate | Steady, worse at the deprived end (+0.95) | Unchanged (+1.00) | Unchanged (+1.00) | Unchanged (+1.00) | Yes |
| Renting, affordability and eviction | Renter households over the cost-burden line | Uneven, worse at the deprived end (−0.30) | Shape changed (+0.99) | Shape changed (+1.00) | Unchanged (+0.79) | No, shares inputs with the rankings |
| Air pollution | Traffic proximity | Steady, better at the deprived end (−0.96) | Ends differ under one ranking only (−0.31) | Shape changed (−0.70) | Unchanged (−0.87) | Yes |
| Civic life, loneliness and partnership | Voter turnout, 2020 | Steady, worse at the deprived end (−1.00) | Unchanged (−0.98) | Unchanged (−0.99) | Unchanged (−0.98) | Yes |
| Life evaluation and daily feelings | Life evaluation (ladder score) | Steady, worse at the deprived end (−0.98) | Unchanged (−1.00) | Unchanged (−1.00) | Unchanged (−0.96) | Yes |
| Place and the length of a life | Life expectancy at birth | Steady, worse at the deprived end (−1.00) | Unchanged (−1.00) | Unchanged (−1.00) | Unchanged (−1.00) | Yes |
| Multiple chronic conditions | Chronic conditions that are high | Steady, worse at the deprived end (+1.00) | Unchanged (+1.00) | Unchanged (+1.00) | Unchanged (+1.00) | Yes |
| Disability, retirement and Medicare | Adult disability | Steady, worse at the deprived end (+1.00) | Unchanged (+1.00) | Unchanged (+1.00) | Unchanged (+1.00) | Yes |
| Health behaviors and the supply of care | Residents per primary care physician | Steady, worse at the deprived end (+1.00) | Unchanged (+0.99) | Unchanged (+0.96) | Unchanged (+1.00) | Yes |
| Gun homicide and the causes of early death | Premature death rate | Steady, worse at the deprived end (+1.00) | Unchanged (+1.00) | Unchanged (+1.00) | Unchanged (+1.00) | Yes |
| Overdose, alcohol and suicide | Overdose death rate | No detectable difference between the ends (+0.14) | Unchanged (+0.28) | Ends differ under one ranking only (+0.43) | Unchanged (+0.14) | Yes |
| Violence and the life-expectancy gap | Homicide death rate, tracts | Steady, worse at the deprived end (+1.00) | Unchanged (+1.00) | Unchanged (+1.00) | Unchanged (+1.00) | Yes |

Note: Each chapter recomputes its own gradient with its outcome, period, weights, bins and intervals; this table only classifies the published rows.

Note: Renting is reported and left out of the count, because rent and income are inputs to the ADI and to the standardized index.

Source: Figures named "alt_rankings" in the chapters listed, data years 1978 to 2024 as each chapter states; ADI 2023 and ACS 2019-2023 tract tables; estimates weighted as in each chapter (population-weighted, births-weighted or unweighted means, crude rates or modeled estimates)

Data: https://vishalsingh.org/zipcode/report/how-far-one-ranking-goes/figures/ranking_check.json

### Table 5. Under either ranking and in each set of places, the least deprived tenth of tracts lives longer and smokes less than all places together, and the most deprived tenth lives shorter and smokes more

Mean at each end of the ranking, among the selected places, divided by the value for all tracts of the comparison, on a log scale; the tick marks the selected places pooled

| Ranking | Places | Outcome | Least deprived end | Selected places pooled | All places pooled | Most deprived end | Least deprived end divided by all places | Most deprived end divided by all places | Selected places divided by all places | Tracts at the least deprived end | Tracts at the most deprived end | Unit |
| --- | --- | --- | --- | --- | --- | --- | ---: | ---: | ---: | ---: | ---: | --- |
| Area Deprivation Index | All places | Life expectancy at birth | 82.3 | 78.8 | 78.8 | 74.0 | 1.04 | 0.94 | n/a | 7,179 | 7,206 | Life expectancy at birth (years) |
| Area Deprivation Index | South | Life expectancy at birth | 82.3 | 77.7 | 78.8 | 74.0 | 1.04 | 0.94 | 0.99 | 928 | 3,810 | Life expectancy at birth (years) |
| Area Deprivation Index | Counties at least 25% Black | Life expectancy at birth | 81.0 | 77.0 | 78.8 | 73.0 | 1.03 | 0.93 | 0.98 | 690 | 2,040 | Life expectancy at birth (years) |
| Median household income | All places | Life expectancy at birth | 82.2 | 78.8 | 78.8 | 74.2 | 1.04 | 0.94 | n/a | 7,177 | 7,178 | Life expectancy at birth (years) |
| Median household income | South | Life expectancy at birth | 81.5 | 77.7 | 78.8 | 74.0 | 1.03 | 0.94 | 0.99 | 1,889 | 3,939 | Life expectancy at birth (years) |
| Median household income | Counties at least 25% Black | Life expectancy at birth | 81.4 | 77.0 | 78.8 | 73.3 | 1.03 | 0.93 | 0.98 | 655 | 2,146 | Life expectancy at birth (years) |
| Area Deprivation Index | All places | Current smoking | 9.0 | 14.5 | 14.5 | 22.7 | 0.62 | 1.56 | n/a | 7,551 | 7,543 | Percent of adults |
| Area Deprivation Index | South | Current smoking | 8.0 | 15.5 | 14.5 | 22.4 | 0.55 | 1.54 | 1.07 | 1,012 | 4,001 | Percent of adults |
| Area Deprivation Index | Counties at least 25% Black | Current smoking | 10.6 | 16.0 | 14.5 | 23.4 | 0.73 | 1.61 | 1.10 | 715 | 2,180 | Percent of adults |
| Median household income | All places | Current smoking | 8.7 | 14.5 | 14.5 | 21.8 | 0.60 | 1.50 | n/a | 7,560 | 7,559 | Percent of adults |
| Median household income | South | Current smoking | 9.0 | 15.5 | 14.5 | 21.8 | 0.62 | 1.50 | 1.07 | 2,039 | 4,155 | Percent of adults |
| Median household income | Counties at least 25% Black | Current smoking | 8.6 | 16.0 | 14.5 | 22.8 | 0.60 | 1.57 | 1.10 | 687 | 2,288 | Percent of adults |
| Area Deprivation Index | All places | Annual checkup | 74.0 | 75.6 | 75.6 | 77.8 | 0.98 | 1.03 | n/a | 7,551 | 7,543 | Percent of adults |
| Area Deprivation Index | South | Annual checkup | 76.5 | 76.8 | 75.6 | 78.1 | 1.01 | 1.03 | 1.02 | 1,012 | 4,001 | Percent of adults |
| Area Deprivation Index | Counties at least 25% Black | Annual checkup | 75.2 | 78.1 | 75.6 | 80.6 | 1.00 | 1.07 | 1.03 | 715 | 2,180 | Percent of adults |
| Median household income | All places | Annual checkup | 75.4 | 75.6 | 75.6 | 76.9 | 1.00 | 1.02 | n/a | 7,560 | 7,559 | Percent of adults |
| Median household income | South | Annual checkup | 76.4 | 76.8 | 75.6 | 77.8 | 1.01 | 1.03 | 1.02 | 2,039 | 4,155 | Percent of adults |
| Median household income | Counties at least 25% Black | Annual checkup | 76.4 | 78.1 | 75.6 | 80.2 | 1.01 | 1.06 | 1.03 | 687 | 2,288 | Percent of adults |

Views: ADI, All places (rk = adi, sub = all); ADI, South (rk = adi, sub = south); ADI, Counties at least a quarter Black (rk = adi, sub = black25); Household income, All places (rk = income, sub = all); Household income, South (rk = income, sub = south); Household income, Counties at least a quarter Black (rk = income, sub = black25).

Note: A prototype on the three outcomes of one chapter, Place and the length of a life, before the other rows of the chapter get the same treatment.

Note: Each tract keeps the tenth it holds among all tracts under the chosen ranking, so the ends mean the same level of deprivation or income in every set of places. Household income ranks tracts by ACS median household income, lowest income at the most deprived end.

Note: The places are the Census South region and counties where at least 25% of residents are Black alone (ACS county estimates). A tract takes the class of its county.

Note: All tracts with every ranking enter, so the values for all places can differ slightly from the chapter's figures, which use every tract with an ADI rank. Weights and intervals are the chapter's: adults for smoking and the checkup, residents for life expectancy, a bootstrap over counties.

Source: USALEEP life expectancy at birth, 2010-2015, and CDC PLACES 2024 (BRFSS 2022) via the chapter Place and the length of a life; ADI 2023 and ACS 2019-2023 tract and county tables; population- and adult-weighted means of modeled estimates

Data: https://vishalsingh.org/zipcode/report/how-far-one-ranking-goes/figures/ends_splits.json

### Table 6. 17 of the 30 gradients are steadily worse toward the more deprived end of the ranking each chapter uses

Each outcome by bin of its chapter's ranking of places, least to most deprived, in its own units, with 95% intervals

| bin | conf_high | conf_low | digits | domain | estimate | least | most | outcome | pattern | rho | unit |
| ---: | ---: | ---: | ---: | --- | ---: | --- | --- | --- | --- | ---: | --- |
| 1 | 7.68 | 7.18 | 1 | Birth and school | 7.43 | 7.4 | 9.7 | Low birthweight | Steady, worse at the deprived end | 1 | Share of live births (%) |
| 2 | 8.00 | 7.32 | 1 | Birth and school | 7.63 | 7.4 | 9.7 | Low birthweight | Steady, worse at the deprived end | 1 | Share of live births (%) |
| 3 | 8.36 | 7.55 | 1 | Birth and school | 7.96 | 7.4 | 9.7 | Low birthweight | Steady, worse at the deprived end | 1 | Share of live births (%) |
| 4 | 8.67 | 7.84 | 1 | Birth and school | 8.29 | 7.4 | 9.7 | Low birthweight | Steady, worse at the deprived end | 1 | Share of live births (%) |
| 5 | 8.65 | 7.97 | 1 | Birth and school | 8.37 | 7.4 | 9.7 | Low birthweight | Steady, worse at the deprived end | 1 | Share of live births (%) |
| 6 | 8.98 | 8.29 | 1 | Birth and school | 8.69 | 7.4 | 9.7 | Low birthweight | Steady, worse at the deprived end | 1 | Share of live births (%) |
| 7 | 9.50 | 8.57 | 1 | Birth and school | 9.05 | 7.4 | 9.7 | Low birthweight | Steady, worse at the deprived end | 1 | Share of live births (%) |
| 8 | 9.85 | 8.89 | 1 | Birth and school | 9.40 | 7.4 | 9.7 | Low birthweight | Steady, worse at the deprived end | 1 | Share of live births (%) |
| 9 | 10.18 | 9.00 | 1 | Birth and school | 9.63 | 7.4 | 9.7 | Low birthweight | Steady, worse at the deprived end | 1 | Share of live births (%) |
| 10 | 10.26 | 9.20 | 1 | Birth and school | 9.68 | 7.4 | 9.7 | Low birthweight | Steady, worse at the deprived end | 1 | Share of live births (%) |
| 1 | 9.01 | 8.59 | 1 | Birth and school | 8.82 | 8.8 | 12.1 | Preterm birth | Steady, worse at the deprived end | 1 | Share of live births (%) |
| 2 | 9.48 | 9.08 | 1 | Birth and school | 9.28 | 8.8 | 12.1 | Preterm birth | Steady, worse at the deprived end | 1 | Share of live births (%) |
| 3 | 10.12 | 9.47 | 1 | Birth and school | 9.82 | 8.8 | 12.1 | Preterm birth | Steady, worse at the deprived end | 1 | Share of live births (%) |
| 4 | 10.29 | 9.80 | 1 | Birth and school | 10.05 | 8.8 | 12.1 | Preterm birth | Steady, worse at the deprived end | 1 | Share of live births (%) |
| 5 | 10.53 | 9.98 | 1 | Birth and school | 10.27 | 8.8 | 12.1 | Preterm birth | Steady, worse at the deprived end | 1 | Share of live births (%) |
| 6 | 11.28 | 10.26 | 1 | Birth and school | 10.80 | 8.8 | 12.1 | Preterm birth | Steady, worse at the deprived end | 1 | Share of live births (%) |
| 7 | 11.22 | 10.56 | 1 | Birth and school | 10.92 | 8.8 | 12.1 | Preterm birth | Steady, worse at the deprived end | 1 | Share of live births (%) |
| 8 | 11.59 | 10.92 | 1 | Birth and school | 11.27 | 8.8 | 12.1 | Preterm birth | Steady, worse at the deprived end | 1 | Share of live births (%) |
| 9 | 11.82 | 11.01 | 1 | Birth and school | 11.42 | 8.8 | 12.1 | Preterm birth | Steady, worse at the deprived end | 1 | Share of live births (%) |
| 10 | 12.66 | 11.32 | 1 | Birth and school | 12.05 | 8.8 | 12.1 | Preterm birth | Steady, worse at the deprived end | 1 | Share of live births (%) |
| 1 | 4.19 | 3.83 | 1 | Birth and school | 4.00 | 4.0 | 8.8 | Infant mortality | Steady, worse at the deprived end | 0.950 | Infant deaths per 1,000 live births |
| 2 | 5.32 | 4.88 | 1 | Birth and school | 5.11 | 4.0 | 8.8 | Infant mortality | Steady, worse at the deprived end | 0.950 | Infant deaths per 1,000 live births |
| 3 | 5.97 | 5.51 | 1 | Birth and school | 5.78 | 4.0 | 8.8 | Infant mortality | Steady, worse at the deprived end | 0.950 | Infant deaths per 1,000 live births |
| 4 | 6.34 | 5.86 | 1 | Birth and school | 6.09 | 4.0 | 8.8 | Infant mortality | Steady, worse at the deprived end | 0.950 | Infant deaths per 1,000 live births |
| 5 | 7.06 | 6.37 | 1 | Birth and school | 6.72 | 4.0 | 8.8 | Infant mortality | Steady, worse at the deprived end | 0.950 | Infant deaths per 1,000 live births |
| 6 | 7.34 | 6.70 | 1 | Birth and school | 7.05 | 4.0 | 8.8 | Infant mortality | Steady, worse at the deprived end | 0.950 | Infant deaths per 1,000 live births |
| 7 | 8.12 | 7.12 | 1 | Birth and school | 7.65 | 4.0 | 8.8 | Infant mortality | Steady, worse at the deprived end | 0.950 | Infant deaths per 1,000 live births |
| 8 | 7.96 | 6.73 | 1 | Birth and school | 7.38 | 4.0 | 8.8 | Infant mortality | Steady, worse at the deprived end | 0.950 | Infant deaths per 1,000 live births |
| 9 | 7.92 | 6.49 | 1 | Birth and school | 7.24 | 4.0 | 8.8 | Infant mortality | Steady, worse at the deprived end | 0.950 | Infant deaths per 1,000 live births |
| 10 | 9.46 | 8.05 | 1 | Birth and school | 8.77 | 4.0 | 8.8 | Infant mortality | Steady, worse at the deprived end | 0.950 | Infant deaths per 1,000 live births |
| 1 | −0.781 | −0.971 | 2 | Birth and school | −0.867 | −0.87 | −0.79 | Reading gap, boys minus girls | No detectable difference between the ends | 0.390 | Mean male minus female gap (grade levels) |
| 2 | −0.735 | −0.881 | 2 | Birth and school | −0.806 | −0.87 | −0.79 | Reading gap, boys minus girls | No detectable difference between the ends | 0.390 | Mean male minus female gap (grade levels) |
| 3 | −0.725 | −0.854 | 2 | Birth and school | −0.799 | −0.87 | −0.79 | Reading gap, boys minus girls | No detectable difference between the ends | 0.390 | Mean male minus female gap (grade levels) |
| 4 | −0.687 | −0.858 | 2 | Birth and school | −0.767 | −0.87 | −0.79 | Reading gap, boys minus girls | No detectable difference between the ends | 0.390 | Mean male minus female gap (grade levels) |
| 5 | −0.692 | −0.841 | 2 | Birth and school | −0.760 | −0.87 | −0.79 | Reading gap, boys minus girls | No detectable difference between the ends | 0.390 | Mean male minus female gap (grade levels) |
| 6 | −0.678 | −0.805 | 2 | Birth and school | −0.737 | −0.87 | −0.79 | Reading gap, boys minus girls | No detectable difference between the ends | 0.390 | Mean male minus female gap (grade levels) |
| 7 | −0.701 | −0.855 | 2 | Birth and school | −0.782 | −0.87 | −0.79 | Reading gap, boys minus girls | No detectable difference between the ends | 0.390 | Mean male minus female gap (grade levels) |
| 8 | −0.704 | −0.861 | 2 | Birth and school | −0.780 | −0.87 | −0.79 | Reading gap, boys minus girls | No detectable difference between the ends | 0.390 | Mean male minus female gap (grade levels) |
| 9 | −0.718 | −0.880 | 2 | Birth and school | −0.795 | −0.87 | −0.79 | Reading gap, boys minus girls | No detectable difference between the ends | 0.390 | Mean male minus female gap (grade levels) |
| 10 | −0.714 | −0.885 | 2 | Birth and school | −0.791 | −0.87 | −0.79 | Reading gap, boys minus girls | No detectable difference between the ends | 0.390 | Mean male minus female gap (grade levels) |
| 1 | 0.119 | 0.038 | 2 | Birth and school | 0.080 | 0.08 | −0.20 | Math gap, boys minus girls | Steady | −0.930 | Mean male minus female gap (grade levels) |
| 2 | 0.063 | −0.032 | 2 | Birth and school | 0.008 | 0.08 | −0.20 | Math gap, boys minus girls | Steady | −0.930 | Mean male minus female gap (grade levels) |
| 3 | 0.001 | −0.101 | 2 | Birth and school | −0.061 | 0.08 | −0.20 | Math gap, boys minus girls | Steady | −0.930 | Mean male minus female gap (grade levels) |
| 4 | −0.017 | −0.114 | 2 | Birth and school | −0.063 | 0.08 | −0.20 | Math gap, boys minus girls | Steady | −0.930 | Mean male minus female gap (grade levels) |
| 5 | −0.032 | −0.111 | 2 | Birth and school | −0.071 | 0.08 | −0.20 | Math gap, boys minus girls | Steady | −0.930 | Mean male minus female gap (grade levels) |
| 6 | −0.013 | −0.086 | 2 | Birth and school | −0.051 | 0.08 | −0.20 | Math gap, boys minus girls | Steady | −0.930 | Mean male minus female gap (grade levels) |
| 7 | −0.049 | −0.154 | 2 | Birth and school | −0.102 | 0.08 | −0.20 | Math gap, boys minus girls | Steady | −0.930 | Mean male minus female gap (grade levels) |
| 8 | −0.093 | −0.196 | 2 | Birth and school | −0.143 | 0.08 | −0.20 | Math gap, boys minus girls | Steady | −0.930 | Mean male minus female gap (grade levels) |
| 9 | −0.102 | −0.190 | 2 | Birth and school | −0.145 | 0.08 | −0.20 | Math gap, boys minus girls | Steady | −0.930 | Mean male minus female gap (grade levels) |
| 10 | −0.153 | −0.259 | 2 | Birth and school | −0.203 | 0.08 | −0.20 | Math gap, boys minus girls | Steady | −0.930 | Mean male minus female gap (grade levels) |
| 1 | 6.91 | 5.27 | 1 | Becoming an adult | 6.10 | 6.1 | 29.3 | Distance to a four-year college | Steady, worse at the deprived end | 1 | Miles to the nearest four-year campus, average resident |
| 2 | 10.37 | 7.33 | 1 | Becoming an adult | 8.53 | 6.1 | 29.3 | Distance to a four-year college | Steady, worse at the deprived end | 1 | Miles to the nearest four-year campus, average resident |
| 3 | 12.34 | 7.70 | 1 | Becoming an adult | 9.36 | 6.1 | 29.3 | Distance to a four-year college | Steady, worse at the deprived end | 1 | Miles to the nearest four-year campus, average resident |
| 4 | 11.67 | 7.94 | 1 | Becoming an adult | 9.54 | 6.1 | 29.3 | Distance to a four-year college | Steady, worse at the deprived end | 1 | Miles to the nearest four-year campus, average resident |
| 5 | 20.68 | 10.77 | 1 | Becoming an adult | 14.49 | 6.1 | 29.3 | Distance to a four-year college | Steady, worse at the deprived end | 1 | Miles to the nearest four-year campus, average resident |
| 6 | 18.94 | 12.67 | 1 | Becoming an adult | 15.25 | 6.1 | 29.3 | Distance to a four-year college | Steady, worse at the deprived end | 1 | Miles to the nearest four-year campus, average resident |
| 7 | 21.96 | 14.73 | 1 | Becoming an adult | 17.97 | 6.1 | 29.3 | Distance to a four-year college | Steady, worse at the deprived end | 1 | Miles to the nearest four-year campus, average resident |
| 8 | 24.30 | 18.35 | 1 | Becoming an adult | 21.28 | 6.1 | 29.3 | Distance to a four-year college | Steady, worse at the deprived end | 1 | Miles to the nearest four-year campus, average resident |
| 9 | 29.50 | 22.45 | 1 | Becoming an adult | 25.60 | 6.1 | 29.3 | Distance to a four-year college | Steady, worse at the deprived end | 1 | Miles to the nearest four-year campus, average resident |
| 10 | 35.02 | 24.64 | 1 | Becoming an adult | 29.34 | 6.1 | 29.3 | Distance to a four-year college | Steady, worse at the deprived end | 1 | Miles to the nearest four-year campus, average resident |
| 1 | 14.01 | 11.85 | 1 | Becoming an adult | 12.95 | 13.0 | 38.5 | Predicted teen birth share | Steady | 1 | Predicted share of women with a teen birth (percent) |
| 2 | 19.34 | 16.06 | 1 | Becoming an adult | 17.73 | 13.0 | 38.5 | Predicted teen birth share | Steady | 1 | Predicted share of women with a teen birth (percent) |
| 3 | 21.14 | 18.97 | 1 | Becoming an adult | 20.03 | 13.0 | 38.5 | Predicted teen birth share | Steady | 1 | Predicted share of women with a teen birth (percent) |
| 4 | 23.01 | 21.62 | 1 | Becoming an adult | 22.31 | 13.0 | 38.5 | Predicted teen birth share | Steady | 1 | Predicted share of women with a teen birth (percent) |
| 5 | 25.17 | 23.94 | 1 | Becoming an adult | 24.55 | 13.0 | 38.5 | Predicted teen birth share | Steady | 1 | Predicted share of women with a teen birth (percent) |
| 6 | 27.15 | 25.90 | 1 | Becoming an adult | 26.54 | 13.0 | 38.5 | Predicted teen birth share | Steady | 1 | Predicted share of women with a teen birth (percent) |
| 7 | 29.25 | 28.03 | 1 | Becoming an adult | 28.65 | 13.0 | 38.5 | Predicted teen birth share | Steady | 1 | Predicted share of women with a teen birth (percent) |
| 8 | 31.26 | 29.92 | 1 | Becoming an adult | 30.57 | 13.0 | 38.5 | Predicted teen birth share | Steady | 1 | Predicted share of women with a teen birth (percent) |
| 9 | 33.37 | 32.12 | 1 | Becoming an adult | 32.75 | 13.0 | 38.5 | Predicted teen birth share | Steady | 1 | Predicted share of women with a teen birth (percent) |
| 10 | 39.27 | 37.79 | 1 | Becoming an adult | 38.54 | 13.0 | 38.5 | Predicted teen birth share | Steady | 1 | Predicted share of women with a teen birth (percent) |
| 1 | 4.17 | 3.40 | 1 | Becoming an adult | 3.80 | 3.8 | 6.4 | Incarceration, Black children from low-income families | Steady, worse at the deprived end | 0.990 | Percent incarcerated on April 1, 2010 |
| 2 | 4.80 | 4.13 | 1 | Becoming an adult | 4.46 | 3.8 | 6.4 | Incarceration, Black children from low-income families | Steady, worse at the deprived end | 0.990 | Percent incarcerated on April 1, 2010 |
| 3 | 5.08 | 4.51 | 1 | Becoming an adult | 4.79 | 3.8 | 6.4 | Incarceration, Black children from low-income families | Steady, worse at the deprived end | 0.990 | Percent incarcerated on April 1, 2010 |
| 4 | 5.30 | 4.75 | 1 | Becoming an adult | 5.02 | 3.8 | 6.4 | Incarceration, Black children from low-income families | Steady, worse at the deprived end | 0.990 | Percent incarcerated on April 1, 2010 |
| 5 | 5.61 | 5.08 | 1 | Becoming an adult | 5.33 | 3.8 | 6.4 | Incarceration, Black children from low-income families | Steady, worse at the deprived end | 0.990 | Percent incarcerated on April 1, 2010 |
| 6 | 5.76 | 5.31 | 1 | Becoming an adult | 5.52 | 3.8 | 6.4 | Incarceration, Black children from low-income families | Steady, worse at the deprived end | 0.990 | Percent incarcerated on April 1, 2010 |
| 7 | 5.74 | 5.32 | 1 | Becoming an adult | 5.52 | 3.8 | 6.4 | Incarceration, Black children from low-income families | Steady, worse at the deprived end | 0.990 | Percent incarcerated on April 1, 2010 |
| 8 | 6.03 | 5.55 | 1 | Becoming an adult | 5.79 | 3.8 | 6.4 | Incarceration, Black children from low-income families | Steady, worse at the deprived end | 0.990 | Percent incarcerated on April 1, 2010 |
| 9 | 6.10 | 5.59 | 1 | Becoming an adult | 5.84 | 3.8 | 6.4 | Incarceration, Black children from low-income families | Steady, worse at the deprived end | 0.990 | Percent incarcerated on April 1, 2010 |
| 10 | 6.68 | 6.06 | 1 | Becoming an adult | 6.37 | 3.8 | 6.4 | Incarceration, Black children from low-income families | Steady, worse at the deprived end | 0.990 | Percent incarcerated on April 1, 2010 |
| 1 | 1.41 | 1.09 | 2 | Becoming an adult | 1.23 | 1.23 | 2.59 | Incarceration, white children from low-income families | Steady, worse at the deprived end | 1 | Percent incarcerated on April 1, 2010 |
| 2 | 1.61 | 1.33 | 2 | Becoming an adult | 1.47 | 1.23 | 2.59 | Incarceration, white children from low-income families | Steady, worse at the deprived end | 1 | Percent incarcerated on April 1, 2010 |
| 3 | 1.75 | 1.52 | 2 | Becoming an adult | 1.64 | 1.23 | 2.59 | Incarceration, white children from low-income families | Steady, worse at the deprived end | 1 | Percent incarcerated on April 1, 2010 |
| 4 | 2.03 | 1.80 | 2 | Becoming an adult | 1.92 | 1.23 | 2.59 | Incarceration, white children from low-income families | Steady, worse at the deprived end | 1 | Percent incarcerated on April 1, 2010 |
| 5 | 2.23 | 1.87 | 2 | Becoming an adult | 2.05 | 1.23 | 2.59 | Incarceration, white children from low-income families | Steady, worse at the deprived end | 1 | Percent incarcerated on April 1, 2010 |
| 6 | 2.29 | 1.97 | 2 | Becoming an adult | 2.13 | 1.23 | 2.59 | Incarceration, white children from low-income families | Steady, worse at the deprived end | 1 | Percent incarcerated on April 1, 2010 |
| 7 | 2.39 | 2.07 | 2 | Becoming an adult | 2.23 | 1.23 | 2.59 | Incarceration, white children from low-income families | Steady, worse at the deprived end | 1 | Percent incarcerated on April 1, 2010 |
| 8 | 2.42 | 2.10 | 2 | Becoming an adult | 2.25 | 1.23 | 2.59 | Incarceration, white children from low-income families | Steady, worse at the deprived end | 1 | Percent incarcerated on April 1, 2010 |
| 9 | 2.45 | 2.17 | 2 | Becoming an adult | 2.32 | 1.23 | 2.59 | Incarceration, white children from low-income families | Steady, worse at the deprived end | 1 | Percent incarcerated on April 1, 2010 |
| 10 | 2.81 | 2.40 | 2 | Becoming an adult | 2.59 | 1.23 | 2.59 | Incarceration, white children from low-income families | Steady, worse at the deprived end | 1 | Percent incarcerated on April 1, 2010 |
| 1 | 8.44 | 7.17 | 1 | Home, work and money | 7.81 | 7.8 | 13.8 | Mortgage denial rate | Steady, worse at the deprived end | 0.950 | Denied share of decided applications (%) |
| 2 | 8.23 | 7.20 | 1 | Home, work and money | 7.69 | 7.8 | 13.8 | Mortgage denial rate | Steady, worse at the deprived end | 0.950 | Denied share of decided applications (%) |
| 3 | 8.02 | 7.25 | 1 | Home, work and money | 7.63 | 7.8 | 13.8 | Mortgage denial rate | Steady, worse at the deprived end | 0.950 | Denied share of decided applications (%) |
| 4 | 8.62 | 7.79 | 1 | Home, work and money | 8.19 | 7.8 | 13.8 | Mortgage denial rate | Steady, worse at the deprived end | 0.950 | Denied share of decided applications (%) |
| 5 | 9.04 | 8.14 | 1 | Home, work and money | 8.59 | 7.8 | 13.8 | Mortgage denial rate | Steady, worse at the deprived end | 0.950 | Denied share of decided applications (%) |
| 6 | 9.63 | 8.68 | 1 | Home, work and money | 9.18 | 7.8 | 13.8 | Mortgage denial rate | Steady, worse at the deprived end | 0.950 | Denied share of decided applications (%) |
| 7 | 9.99 | 9.10 | 1 | Home, work and money | 9.55 | 7.8 | 13.8 | Mortgage denial rate | Steady, worse at the deprived end | 0.950 | Denied share of decided applications (%) |
| 8 | 10.51 | 9.59 | 1 | Home, work and money | 10.04 | 7.8 | 13.8 | Mortgage denial rate | Steady, worse at the deprived end | 0.950 | Denied share of decided applications (%) |
| 9 | 11.35 | 10.54 | 1 | Home, work and money | 10.91 | 7.8 | 13.8 | Mortgage denial rate | Steady, worse at the deprived end | 0.950 | Denied share of decided applications (%) |
| 10 | 14.33 | 13.33 | 1 | Home, work and money | 13.81 | 7.8 | 13.8 | Mortgage denial rate | Steady, worse at the deprived end | 0.950 | Denied share of decided applications (%) |
| 1 | 46.68 | 41.91 | 1 | Home, work and money | 44.45 | 44.5 | 47.6 | Renter households over the cost-burden line | No detectable difference between the ends | −0.300 | Share paying more than 30% of income (%) |
| 2 | 50.08 | 45 | 1 | Home, work and money | 47.68 | 44.5 | 47.6 | Renter households over the cost-burden line | No detectable difference between the ends | −0.300 | Share paying more than 30% of income (%) |
| 3 | 48.50 | 44.76 | 1 | Home, work and money | 46.65 | 44.5 | 47.6 | Renter households over the cost-burden line | No detectable difference between the ends | −0.300 | Share paying more than 30% of income (%) |
| 4 | 46.71 | 44.48 | 1 | Home, work and money | 45.61 | 44.5 | 47.6 | Renter households over the cost-burden line | No detectable difference between the ends | −0.300 | Share paying more than 30% of income (%) |
| 5 | 46.16 | 44.14 | 1 | Home, work and money | 45.17 | 44.5 | 47.6 | Renter households over the cost-burden line | No detectable difference between the ends | −0.300 | Share paying more than 30% of income (%) |
| 6 | 46.02 | 44.16 | 1 | Home, work and money | 45.12 | 44.5 | 47.6 | Renter households over the cost-burden line | No detectable difference between the ends | −0.300 | Share paying more than 30% of income (%) |
| 7 | 45.57 | 43.59 | 1 | Home, work and money | 44.64 | 44.5 | 47.6 | Renter households over the cost-burden line | No detectable difference between the ends | −0.300 | Share paying more than 30% of income (%) |
| 8 | 45.27 | 43.51 | 1 | Home, work and money | 44.42 | 44.5 | 47.6 | Renter households over the cost-burden line | No detectable difference between the ends | −0.300 | Share paying more than 30% of income (%) |
| 9 | 44.95 | 43.26 | 1 | Home, work and money | 44.13 | 44.5 | 47.6 | Renter households over the cost-burden line | No detectable difference between the ends | −0.300 | Share paying more than 30% of income (%) |
| 10 | 48.42 | 46.63 | 1 | Home, work and money | 47.58 | 44.5 | 47.6 | Renter households over the cost-burden line | No detectable difference between the ends | −0.300 | Share paying more than 30% of income (%) |
| 1 | 27.22 | 23.74 | 1 | Home, work and money | 25.55 | 25.6 | 19.8 | Owner households over the cost-burden line | Uneven, better at the deprived end | −0.880 | Share paying more than 30% of income (%) |
| 2 | 25.42 | 22.31 | 1 | Home, work and money | 23.87 | 25.6 | 19.8 | Owner households over the cost-burden line | Uneven, better at the deprived end | −0.880 | Share paying more than 30% of income (%) |
| 3 | 22.64 | 20.98 | 1 | Home, work and money | 21.79 | 25.6 | 19.8 | Owner households over the cost-burden line | Uneven, better at the deprived end | −0.880 | Share paying more than 30% of income (%) |
| 4 | 21.70 | 20.36 | 1 | Home, work and money | 21.03 | 25.6 | 19.8 | Owner households over the cost-burden line | Uneven, better at the deprived end | −0.880 | Share paying more than 30% of income (%) |
| 5 | 20.85 | 19.59 | 1 | Home, work and money | 20.20 | 25.6 | 19.8 | Owner households over the cost-burden line | Uneven, better at the deprived end | −0.880 | Share paying more than 30% of income (%) |
| 6 | 20 | 18.90 | 1 | Home, work and money | 19.45 | 25.6 | 19.8 | Owner households over the cost-burden line | Uneven, better at the deprived end | −0.880 | Share paying more than 30% of income (%) |
| 7 | 19.56 | 18.45 | 1 | Home, work and money | 18.97 | 25.6 | 19.8 | Owner households over the cost-burden line | Uneven, better at the deprived end | −0.880 | Share paying more than 30% of income (%) |
| 8 | 18.89 | 17.95 | 1 | Home, work and money | 18.42 | 25.6 | 19.8 | Owner households over the cost-burden line | Uneven, better at the deprived end | −0.880 | Share paying more than 30% of income (%) |
| 9 | 18.74 | 17.90 | 1 | Home, work and money | 18.31 | 25.6 | 19.8 | Owner households over the cost-burden line | Uneven, better at the deprived end | −0.880 | Share paying more than 30% of income (%) |
| 10 | 20.28 | 19.25 | 1 | Home, work and money | 19.78 | 25.6 | 19.8 | Owner households over the cost-burden line | Uneven, better at the deprived end | −0.880 | Share paying more than 30% of income (%) |
| 1 | 78.06 | 72.60 | 1 | Home, work and money | 75.33 | 75.3 | 62.5 | Low-income renters over the cost-burden line | Steady, better at the deprived end | −1 | Low-income households paying more than 30% of income (%) |
| 2 | 76.22 | 74.15 | 1 | Home, work and money | 75.05 | 75.3 | 62.5 | Low-income renters over the cost-burden line | Steady, better at the deprived end | −1 | Low-income households paying more than 30% of income (%) |
| 3 | 75.27 | 72.86 | 1 | Home, work and money | 74.01 | 75.3 | 62.5 | Low-income renters over the cost-burden line | Steady, better at the deprived end | −1 | Low-income households paying more than 30% of income (%) |
| 4 | 74.23 | 72.10 | 1 | Home, work and money | 73.20 | 75.3 | 62.5 | Low-income renters over the cost-burden line | Steady, better at the deprived end | −1 | Low-income households paying more than 30% of income (%) |
| 5 | 72.19 | 69.90 | 1 | Home, work and money | 71.11 | 75.3 | 62.5 | Low-income renters over the cost-burden line | Steady, better at the deprived end | −1 | Low-income households paying more than 30% of income (%) |
| 6 | 70.06 | 67.89 | 1 | Home, work and money | 68.99 | 75.3 | 62.5 | Low-income renters over the cost-burden line | Steady, better at the deprived end | −1 | Low-income households paying more than 30% of income (%) |
| 7 | 67.66 | 65.52 | 1 | Home, work and money | 66.62 | 75.3 | 62.5 | Low-income renters over the cost-burden line | Steady, better at the deprived end | −1 | Low-income households paying more than 30% of income (%) |
| 8 | 65.44 | 63.63 | 1 | Home, work and money | 64.56 | 75.3 | 62.5 | Low-income renters over the cost-burden line | Steady, better at the deprived end | −1 | Low-income households paying more than 30% of income (%) |
| 9 | 63.27 | 61.66 | 1 | Home, work and money | 62.51 | 75.3 | 62.5 | Low-income renters over the cost-burden line | Steady, better at the deprived end | −1 | Low-income households paying more than 30% of income (%) |
| 10 | 63.27 | 61.58 | 1 | Home, work and money | 62.46 | 75.3 | 62.5 | Low-income renters over the cost-burden line | Steady, better at the deprived end | −1 | Low-income households paying more than 30% of income (%) |
| 1 | 79.18 | 68.05 | 1 | Environment and connection | 74.70 | 74.7 | 39.0 | Traffic proximity | Steady, better at the deprived end | −0.960 | Mean national percentile of the tract (0 to 100) |
| 2 | 71.29 | 58.96 | 1 | Environment and connection | 65.23 | 74.7 | 39.0 | Traffic proximity | Steady, better at the deprived end | −0.960 | Mean national percentile of the tract (0 to 100) |
| 3 | 60.23 | 53.49 | 1 | Environment and connection | 56.92 | 74.7 | 39.0 | Traffic proximity | Steady, better at the deprived end | −0.960 | Mean national percentile of the tract (0 to 100) |
| 4 | 53.77 | 48.87 | 1 | Environment and connection | 51.45 | 74.7 | 39.0 | Traffic proximity | Steady, better at the deprived end | −0.960 | Mean national percentile of the tract (0 to 100) |
| 5 | 50.34 | 44.93 | 1 | Environment and connection | 47.72 | 74.7 | 39.0 | Traffic proximity | Steady, better at the deprived end | −0.960 | Mean national percentile of the tract (0 to 100) |
| 6 | 47.13 | 41.66 | 1 | Environment and connection | 44.43 | 74.7 | 39.0 | Traffic proximity | Steady, better at the deprived end | −0.960 | Mean national percentile of the tract (0 to 100) |
| 7 | 44.32 | 38.17 | 1 | Environment and connection | 41.19 | 74.7 | 39.0 | Traffic proximity | Steady, better at the deprived end | −0.960 | Mean national percentile of the tract (0 to 100) |
| 8 | 40.55 | 35.39 | 1 | Environment and connection | 38.01 | 74.7 | 39.0 | Traffic proximity | Steady, better at the deprived end | −0.960 | Mean national percentile of the tract (0 to 100) |
| 9 | 37.53 | 32.20 | 1 | Environment and connection | 34.92 | 74.7 | 39.0 | Traffic proximity | Steady, better at the deprived end | −0.960 | Mean national percentile of the tract (0 to 100) |
| 10 | 42.13 | 35.42 | 1 | Environment and connection | 38.99 | 74.7 | 39.0 | Traffic proximity | Steady, better at the deprived end | −0.960 | Mean national percentile of the tract (0 to 100) |
| 1 | 73.24 | 52.34 | 1 | Environment and connection | 64.27 | 64.3 | 47.8 | Fine particulate matter | Steady, better at the deprived end | −0.930 | Mean national percentile of the tract (0 to 100) |
| 2 | 65.84 | 47.61 | 1 | Environment and connection | 56.82 | 64.3 | 47.8 | Fine particulate matter | Steady, better at the deprived end | −0.930 | Mean national percentile of the tract (0 to 100) |
| 3 | 56.99 | 47.69 | 1 | Environment and connection | 52.29 | 64.3 | 47.8 | Fine particulate matter | Steady, better at the deprived end | −0.930 | Mean national percentile of the tract (0 to 100) |
| 4 | 54.11 | 47.20 | 1 | Environment and connection | 50.75 | 64.3 | 47.8 | Fine particulate matter | Steady, better at the deprived end | −0.930 | Mean national percentile of the tract (0 to 100) |
| 5 | 51.99 | 46.09 | 1 | Environment and connection | 49.21 | 64.3 | 47.8 | Fine particulate matter | Steady, better at the deprived end | −0.930 | Mean national percentile of the tract (0 to 100) |
| 6 | 51.28 | 44.87 | 1 | Environment and connection | 48.16 | 64.3 | 47.8 | Fine particulate matter | Steady, better at the deprived end | −0.930 | Mean national percentile of the tract (0 to 100) |
| 7 | 50.70 | 44.20 | 1 | Environment and connection | 47.40 | 64.3 | 47.8 | Fine particulate matter | Steady, better at the deprived end | −0.930 | Mean national percentile of the tract (0 to 100) |
| 8 | 47.38 | 41.64 | 1 | Environment and connection | 44.47 | 64.3 | 47.8 | Fine particulate matter | Steady, better at the deprived end | −0.930 | Mean national percentile of the tract (0 to 100) |
| 9 | 46.62 | 41.50 | 1 | Environment and connection | 44.06 | 64.3 | 47.8 | Fine particulate matter | Steady, better at the deprived end | −0.930 | Mean national percentile of the tract (0 to 100) |
| 10 | 50.97 | 44.29 | 1 | Environment and connection | 47.76 | 64.3 | 47.8 | Fine particulate matter | Steady, better at the deprived end | −0.930 | Mean national percentile of the tract (0 to 100) |
| 1 | 73.80 | 70.58 | 1 | Environment and connection | 72.06 | 72.1 | 56.4 | Voter turnout, 2020 | Steady, worse at the deprived end | −1 | Turnout or self-response (%) |
| 2 | 71 | 67.93 | 1 | Environment and connection | 69.33 | 72.1 | 56.4 | Voter turnout, 2020 | Steady, worse at the deprived end | −1 | Turnout or self-response (%) |
| 3 | 68.58 | 64.62 | 1 | Environment and connection | 66.59 | 72.1 | 56.4 | Voter turnout, 2020 | Steady, worse at the deprived end | −1 | Turnout or self-response (%) |
| 4 | 65.57 | 62.07 | 1 | Environment and connection | 63.95 | 72.1 | 56.4 | Voter turnout, 2020 | Steady, worse at the deprived end | −1 | Turnout or self-response (%) |
| 5 | 64.15 | 61.43 | 1 | Environment and connection | 62.84 | 72.1 | 56.4 | Voter turnout, 2020 | Steady, worse at the deprived end | −1 | Turnout or self-response (%) |
| 6 | 62.08 | 59.10 | 1 | Environment and connection | 60.63 | 72.1 | 56.4 | Voter turnout, 2020 | Steady, worse at the deprived end | −1 | Turnout or self-response (%) |
| 7 | 60.94 | 58.84 | 1 | Environment and connection | 59.85 | 72.1 | 56.4 | Voter turnout, 2020 | Steady, worse at the deprived end | −1 | Turnout or self-response (%) |
| 8 | 60.52 | 56.90 | 1 | Environment and connection | 58.66 | 72.1 | 56.4 | Voter turnout, 2020 | Steady, worse at the deprived end | −1 | Turnout or self-response (%) |
| 9 | 59.01 | 57.18 | 1 | Environment and connection | 58.06 | 72.1 | 56.4 | Voter turnout, 2020 | Steady, worse at the deprived end | −1 | Turnout or self-response (%) |
| 10 | 57.59 | 55.37 | 1 | Environment and connection | 56.43 | 72.1 | 56.4 | Voter turnout, 2020 | Steady, worse at the deprived end | −1 | Turnout or self-response (%) |
| 1 | 69.94 | 67.39 | 1 | Environment and connection | 68.60 | 68.6 | 47.6 | Census self-response, 2020 | Steady, worse at the deprived end | −1 | Turnout or self-response (%) |
| 2 | 68.15 | 65.91 | 1 | Environment and connection | 66.97 | 68.6 | 47.6 | Census self-response, 2020 | Steady, worse at the deprived end | −1 | Turnout or self-response (%) |
| 3 | 68.04 | 65 | 1 | Environment and connection | 66.45 | 68.6 | 47.6 | Census self-response, 2020 | Steady, worse at the deprived end | −1 | Turnout or self-response (%) |
| 4 | 66.25 | 63.61 | 1 | Environment and connection | 64.94 | 68.6 | 47.6 | Census self-response, 2020 | Steady, worse at the deprived end | −1 | Turnout or self-response (%) |
| 5 | 64.80 | 62.56 | 1 | Environment and connection | 63.80 | 68.6 | 47.6 | Census self-response, 2020 | Steady, worse at the deprived end | −1 | Turnout or self-response (%) |
| 6 | 62.88 | 60.24 | 1 | Environment and connection | 61.63 | 68.6 | 47.6 | Census self-response, 2020 | Steady, worse at the deprived end | −1 | Turnout or self-response (%) |
| 7 | 61.85 | 59.72 | 1 | Environment and connection | 60.76 | 68.6 | 47.6 | Census self-response, 2020 | Steady, worse at the deprived end | −1 | Turnout or self-response (%) |
| 8 | 58.76 | 55.75 | 1 | Environment and connection | 57.15 | 68.6 | 47.6 | Census self-response, 2020 | Steady, worse at the deprived end | −1 | Turnout or self-response (%) |
| 9 | 55.21 | 53.18 | 1 | Environment and connection | 54.20 | 68.6 | 47.6 | Census self-response, 2020 | Steady, worse at the deprived end | −1 | Turnout or self-response (%) |
| 10 | 49.12 | 46.11 | 1 | Environment and connection | 47.63 | 68.6 | 47.6 | Census self-response, 2020 | Steady, worse at the deprived end | −1 | Turnout or self-response (%) |
| 1 | 7.23 | 7.11 | 2 | Environment and connection | 7.17 | 7.17 | 6.93 | Life evaluation (ladder score) | Steady, worse at the deprived end | −1 | Mean ladder score (0 to 10) |
| 2 | 7.17 | 7.09 | 2 | Environment and connection | 7.13 | 7.17 | 6.93 | Life evaluation (ladder score) | Steady, worse at the deprived end | −1 | Mean ladder score (0 to 10) |
| 3 | 7.15 | 7.08 | 2 | Environment and connection | 7.11 | 7.17 | 6.93 | Life evaluation (ladder score) | Steady, worse at the deprived end | −1 | Mean ladder score (0 to 10) |
| 4 | 7.14 | 7.07 | 2 | Environment and connection | 7.10 | 7.17 | 6.93 | Life evaluation (ladder score) | Steady, worse at the deprived end | −1 | Mean ladder score (0 to 10) |
| 5 | 7.12 | 7.07 | 2 | Environment and connection | 7.10 | 7.17 | 6.93 | Life evaluation (ladder score) | Steady, worse at the deprived end | −1 | Mean ladder score (0 to 10) |
| 6 | 7.12 | 7.06 | 2 | Environment and connection | 7.09 | 7.17 | 6.93 | Life evaluation (ladder score) | Steady, worse at the deprived end | −1 | Mean ladder score (0 to 10) |
| 7 | 7.08 | 7.03 | 2 | Environment and connection | 7.06 | 7.17 | 6.93 | Life evaluation (ladder score) | Steady, worse at the deprived end | −1 | Mean ladder score (0 to 10) |
| 8 | 7.06 | 7.02 | 2 | Environment and connection | 7.04 | 7.17 | 6.93 | Life evaluation (ladder score) | Steady, worse at the deprived end | −1 | Mean ladder score (0 to 10) |
| 9 | 7.00 | 6.96 | 2 | Environment and connection | 6.98 | 7.17 | 6.93 | Life evaluation (ladder score) | Steady, worse at the deprived end | −1 | Mean ladder score (0 to 10) |
| 10 | 6.94 | 6.91 | 2 | Environment and connection | 6.93 | 7.17 | 6.93 | Life evaluation (ladder score) | Steady, worse at the deprived end | −1 | Mean ladder score (0 to 10) |
| 1 | 9.59 | 8.53 | 1 | Health and care | 9.06 | 9.1 | 22.8 | Current smoking | Steady, worse at the deprived end | 1 | Percent of adults |
| 2 | 11.25 | 10.20 | 1 | Health and care | 10.74 | 9.1 | 22.8 | Current smoking | Steady, worse at the deprived end | 1 | Percent of adults |
| 3 | 11.99 | 11.28 | 1 | Health and care | 11.62 | 9.1 | 22.8 | Current smoking | Steady, worse at the deprived end | 1 | Percent of adults |
| 4 | 12.88 | 12.45 | 1 | Health and care | 12.67 | 9.1 | 22.8 | Current smoking | Steady, worse at the deprived end | 1 | Percent of adults |
| 5 | 14.09 | 13.70 | 1 | Health and care | 13.89 | 9.1 | 22.8 | Current smoking | Steady, worse at the deprived end | 1 | Percent of adults |
| 6 | 15.28 | 14.93 | 1 | Health and care | 15.10 | 9.1 | 22.8 | Current smoking | Steady, worse at the deprived end | 1 | Percent of adults |
| 7 | 16.54 | 16.25 | 1 | Health and care | 16.39 | 9.1 | 22.8 | Current smoking | Steady, worse at the deprived end | 1 | Percent of adults |
| 8 | 17.94 | 17.61 | 1 | Health and care | 17.78 | 9.1 | 22.8 | Current smoking | Steady, worse at the deprived end | 1 | Percent of adults |
| 9 | 19.62 | 19.27 | 1 | Health and care | 19.45 | 9.1 | 22.8 | Current smoking | Steady, worse at the deprived end | 1 | Percent of adults |
| 10 | 23.14 | 22.46 | 1 | Health and care | 22.81 | 9.1 | 22.8 | Current smoking | Steady, worse at the deprived end | 1 | Percent of adults |
| 1 | 75.29 | 73.23 | 1 | Health and care | 74.16 | 74.2 | 77.8 | Annual checkup | Steady, better at the deprived end | 1 | Percent of adults |
| 2 | 75.40 | 73.25 | 1 | Health and care | 74.27 | 74.2 | 77.8 | Annual checkup | Steady, better at the deprived end | 1 | Percent of adults |
| 3 | 75.35 | 74 | 1 | Health and care | 74.67 | 74.2 | 77.8 | Annual checkup | Steady, better at the deprived end | 1 | Percent of adults |
| 4 | 75.41 | 74.40 | 1 | Health and care | 74.90 | 74.2 | 77.8 | Annual checkup | Steady, better at the deprived end | 1 | Percent of adults |
| 5 | 75.89 | 75.06 | 1 | Health and care | 75.46 | 74.2 | 77.8 | Annual checkup | Steady, better at the deprived end | 1 | Percent of adults |
| 6 | 76.26 | 75.45 | 1 | Health and care | 75.85 | 74.2 | 77.8 | Annual checkup | Steady, better at the deprived end | 1 | Percent of adults |
| 7 | 76.59 | 75.81 | 1 | Health and care | 76.21 | 74.2 | 77.8 | Annual checkup | Steady, better at the deprived end | 1 | Percent of adults |
| 8 | 76.96 | 76.26 | 1 | Health and care | 76.62 | 74.2 | 77.8 | Annual checkup | Steady, better at the deprived end | 1 | Percent of adults |
| 9 | 77.44 | 76.83 | 1 | Health and care | 77.14 | 74.2 | 77.8 | Annual checkup | Steady, better at the deprived end | 1 | Percent of adults |
| 10 | 78.18 | 77.34 | 1 | Health and care | 77.76 | 74.2 | 77.8 | Annual checkup | Steady, better at the deprived end | 1 | Percent of adults |
| 1 | 0.550 | 0.380 | 1 | Health and care | 0.450 | 0.5 | 5.7 | Chronic conditions that are high | Steady, worse at the deprived end | 1 | Mean number of high conditions (of nine) |
| 2 | 0.700 | 0.550 | 1 | Health and care | 0.620 | 0.5 | 5.7 | Chronic conditions that are high | Steady, worse at the deprived end | 1 | Mean number of high conditions (of nine) |
| 3 | 1 | 0.780 | 1 | Health and care | 0.880 | 0.5 | 5.7 | Chronic conditions that are high | Steady, worse at the deprived end | 1 | Mean number of high conditions (of nine) |
| 4 | 1.10 | 0.920 | 1 | Health and care | 1.01 | 0.5 | 5.7 | Chronic conditions that are high | Steady, worse at the deprived end | 1 | Mean number of high conditions (of nine) |
| 5 | 1.39 | 1.19 | 1 | Health and care | 1.29 | 0.5 | 5.7 | Chronic conditions that are high | Steady, worse at the deprived end | 1 | Mean number of high conditions (of nine) |
| 6 | 1.83 | 1.60 | 1 | Health and care | 1.71 | 0.5 | 5.7 | Chronic conditions that are high | Steady, worse at the deprived end | 1 | Mean number of high conditions (of nine) |
| 7 | 2.46 | 2.22 | 1 | Health and care | 2.34 | 0.5 | 5.7 | Chronic conditions that are high | Steady, worse at the deprived end | 1 | Mean number of high conditions (of nine) |
| 8 | 3.34 | 3.07 | 1 | Health and care | 3.21 | 0.5 | 5.7 | Chronic conditions that are high | Steady, worse at the deprived end | 1 | Mean number of high conditions (of nine) |
| 9 | 4.49 | 4.21 | 1 | Health and care | 4.35 | 0.5 | 5.7 | Chronic conditions that are high | Steady, worse at the deprived end | 1 | Mean number of high conditions (of nine) |
| 10 | 5.81 | 5.52 | 1 | Health and care | 5.66 | 0.5 | 5.7 | Chronic conditions that are high | Steady, worse at the deprived end | 1 | Mean number of high conditions (of nine) |
| 1 | 24.07 | 22.30 | 1 | Health and care | 23.28 | 23.3 | 42.9 | Adult disability | Steady, worse at the deprived end | 1 | Percent of adults |
| 2 | 27.45 | 24.50 | 1 | Health and care | 25.97 | 23.3 | 42.9 | Adult disability | Steady, worse at the deprived end | 1 | Percent of adults |
| 3 | 27.99 | 26.08 | 1 | Health and care | 27 | 23.3 | 42.9 | Adult disability | Steady, worse at the deprived end | 1 | Percent of adults |
| 4 | 28.73 | 27.60 | 1 | Health and care | 28.17 | 23.3 | 42.9 | Adult disability | Steady, worse at the deprived end | 1 | Percent of adults |
| 5 | 30.32 | 29.39 | 1 | Health and care | 29.84 | 23.3 | 42.9 | Adult disability | Steady, worse at the deprived end | 1 | Percent of adults |
| 6 | 32.06 | 31.21 | 1 | Health and care | 31.62 | 23.3 | 42.9 | Adult disability | Steady, worse at the deprived end | 1 | Percent of adults |
| 7 | 33.88 | 33.16 | 1 | Health and care | 33.53 | 23.3 | 42.9 | Adult disability | Steady, worse at the deprived end | 1 | Percent of adults |
| 8 | 35.81 | 35.15 | 1 | Health and care | 35.48 | 23.3 | 42.9 | Adult disability | Steady, worse at the deprived end | 1 | Percent of adults |
| 9 | 38.39 | 37.74 | 1 | Health and care | 38.06 | 23.3 | 42.9 | Adult disability | Steady, worse at the deprived end | 1 | Percent of adults |
| 10 | 43.21 | 42.50 | 1 | Health and care | 42.86 | 23.3 | 42.9 | Adult disability | Steady, worse at the deprived end | 1 | Percent of adults |
| 1 | 1205 | 1093 | 0 | Health and care | 1159 | 1,159 | 2,730 | Residents per primary care physician | Steady, worse at the deprived end | 1 | Residents per provider |
| 2 | 1449 | 1221 | 0 | Health and care | 1335 | 1,159 | 2,730 | Residents per primary care physician | Steady, worse at the deprived end | 1 | Residents per provider |
| 3 | 1511 | 1286 | 0 | Health and care | 1389 | 1,159 | 2,730 | Residents per primary care physician | Steady, worse at the deprived end | 1 | Residents per provider |
| 4 | 1512 | 1309 | 0 | Health and care | 1411 | 1,159 | 2,730 | Residents per primary care physician | Steady, worse at the deprived end | 1 | Residents per provider |
| 5 | 1737 | 1365 | 0 | Health and care | 1529 | 1,159 | 2,730 | Residents per primary care physician | Steady, worse at the deprived end | 1 | Residents per provider |
| 6 | 1877 | 1458 | 0 | Health and care | 1656 | 1,159 | 2,730 | Residents per primary care physician | Steady, worse at the deprived end | 1 | Residents per provider |
| 7 | 2148 | 1660 | 0 | Health and care | 1894 | 1,159 | 2,730 | Residents per primary care physician | Steady, worse at the deprived end | 1 | Residents per provider |
| 8 | 2249 | 1905 | 0 | Health and care | 2083 | 1,159 | 2,730 | Residents per primary care physician | Steady, worse at the deprived end | 1 | Residents per provider |
| 9 | 2828 | 2394 | 0 | Health and care | 2593 | 1,159 | 2,730 | Residents per primary care physician | Steady, worse at the deprived end | 1 | Residents per provider |
| 10 | 3018 | 2452 | 0 | Health and care | 2730 | 1,159 | 2,730 | Residents per primary care physician | Steady, worse at the deprived end | 1 | Residents per provider |
| 1 | 584 | 555.5 | 0 | Health and care | 570.5 | 571 | 677 | Emergency visits per Medicare beneficiary | Steady | 1 | Visits per thousand beneficiaries per year |
| 2 | 617.4 | 582.1 | 0 | Health and care | 598.9 | 571 | 677 | Emergency visits per Medicare beneficiary | Steady | 1 | Visits per thousand beneficiaries per year |
| 3 | 635.5 | 603 | 0 | Health and care | 620.1 | 571 | 677 | Emergency visits per Medicare beneficiary | Steady | 1 | Visits per thousand beneficiaries per year |
| 4 | 655.4 | 627.3 | 0 | Health and care | 642.1 | 571 | 677 | Emergency visits per Medicare beneficiary | Steady | 1 | Visits per thousand beneficiaries per year |
| 5 | 693.6 | 656.4 | 0 | Health and care | 676.7 | 571 | 677 | Emergency visits per Medicare beneficiary | Steady | 1 | Visits per thousand beneficiaries per year |
| 1 | 317.6 | 296.1 | 0 | Early death | 306.9 | 307 | 701 | Premature death rate | Steady, worse at the deprived end | 1 | Premature deaths per 100,000 residents, age-adjusted |
| 2 | 387.5 | 356.6 | 0 | Early death | 372.7 | 307 | 701 | Premature death rate | Steady, worse at the deprived end | 1 | Premature deaths per 100,000 residents, age-adjusted |
| 3 | 438.9 | 399.7 | 0 | Early death | 417.1 | 307 | 701 | Premature death rate | Steady, worse at the deprived end | 1 | Premature deaths per 100,000 residents, age-adjusted |
| 4 | 493.3 | 439.2 | 0 | Early death | 465.4 | 307 | 701 | Premature death rate | Steady, worse at the deprived end | 1 | Premature deaths per 100,000 residents, age-adjusted |
| 5 | 510.6 | 465.7 | 0 | Early death | 490.1 | 307 | 701 | Premature death rate | Steady, worse at the deprived end | 1 | Premature deaths per 100,000 residents, age-adjusted |
| 6 | 536.2 | 483.2 | 0 | Early death | 505.8 | 307 | 701 | Premature death rate | Steady, worse at the deprived end | 1 | Premature deaths per 100,000 residents, age-adjusted |
| 7 | 563.4 | 502 | 0 | Early death | 531.3 | 307 | 701 | Premature death rate | Steady, worse at the deprived end | 1 | Premature deaths per 100,000 residents, age-adjusted |
| 8 | 588.4 | 508.8 | 0 | Early death | 543.4 | 307 | 701 | Premature death rate | Steady, worse at the deprived end | 1 | Premature deaths per 100,000 residents, age-adjusted |
| 9 | 631.6 | 568.9 | 0 | Early death | 604 | 307 | 701 | Premature death rate | Steady, worse at the deprived end | 1 | Premature deaths per 100,000 residents, age-adjusted |
| 10 | 739.9 | 659.7 | 0 | Early death | 701.1 | 307 | 701 | Premature death rate | Steady, worse at the deprived end | 1 | Premature deaths per 100,000 residents, age-adjusted |
| 1 | 25.58 | 22.17 | 1 | Early death | 23.84 | 23.8 | 31.1 | Overdose death rate | Uneven, worse at the deprived end | 0.090 | Deaths per 100,000 residents |
| 2 | 29.93 | 24.57 | 1 | Early death | 27.28 | 23.8 | 31.1 | Overdose death rate | Uneven, worse at the deprived end | 0.090 | Deaths per 100,000 residents |
| 3 | 34.96 | 26.90 | 1 | Early death | 30.74 | 23.8 | 31.1 | Overdose death rate | Uneven, worse at the deprived end | 0.090 | Deaths per 100,000 residents |
| 4 | 42.01 | 30.60 | 1 | Early death | 36.13 | 23.8 | 31.1 | Overdose death rate | Uneven, worse at the deprived end | 0.090 | Deaths per 100,000 residents |
| 5 | 35.49 | 27.61 | 1 | Early death | 31.86 | 23.8 | 31.1 | Overdose death rate | Uneven, worse at the deprived end | 0.090 | Deaths per 100,000 residents |
| 6 | 31.25 | 24.58 | 1 | Early death | 27.90 | 23.8 | 31.1 | Overdose death rate | Uneven, worse at the deprived end | 0.090 | Deaths per 100,000 residents |
| 7 | 35.13 | 27.89 | 1 | Early death | 31.39 | 23.8 | 31.1 | Overdose death rate | Uneven, worse at the deprived end | 0.090 | Deaths per 100,000 residents |
| 8 | 33.13 | 21.71 | 1 | Early death | 27.17 | 23.8 | 31.1 | Overdose death rate | Uneven, worse at the deprived end | 0.090 | Deaths per 100,000 residents |
| 9 | 30.12 | 23.86 | 1 | Early death | 26.83 | 23.8 | 31.1 | Overdose death rate | Uneven, worse at the deprived end | 0.090 | Deaths per 100,000 residents |
| 10 | 36.49 | 25.66 | 1 | Early death | 31.09 | 23.8 | 31.1 | Overdose death rate | Uneven, worse at the deprived end | 0.090 | Deaths per 100,000 residents |
| 1 | 13.16 | 11.31 | 1 | Early death | 12.18 | 12.2 | 17.4 | Suicide death rate | Uneven, worse at the deprived end | 0.810 | Deaths per 100,000 residents |
| 2 | 15.20 | 13.23 | 1 | Early death | 14.15 | 12.2 | 17.4 | Suicide death rate | Uneven, worse at the deprived end | 0.810 | Deaths per 100,000 residents |
| 3 | 17.11 | 14.83 | 1 | Early death | 15.83 | 12.2 | 17.4 | Suicide death rate | Uneven, worse at the deprived end | 0.810 | Deaths per 100,000 residents |
| 4 | 17.04 | 15.09 | 1 | Early death | 16.03 | 12.2 | 17.4 | Suicide death rate | Uneven, worse at the deprived end | 0.810 | Deaths per 100,000 residents |
| 5 | 17.94 | 16.04 | 1 | Early death | 16.91 | 12.2 | 17.4 | Suicide death rate | Uneven, worse at the deprived end | 0.810 | Deaths per 100,000 residents |
| 6 | 18.76 | 16.57 | 1 | Early death | 17.63 | 12.2 | 17.4 | Suicide death rate | Uneven, worse at the deprived end | 0.810 | Deaths per 100,000 residents |
| 7 | 19.09 | 17.60 | 1 | Early death | 18.32 | 12.2 | 17.4 | Suicide death rate | Uneven, worse at the deprived end | 0.810 | Deaths per 100,000 residents |
| 8 | 18.94 | 14.20 | 1 | Early death | 16.62 | 12.2 | 17.4 | Suicide death rate | Uneven, worse at the deprived end | 0.810 | Deaths per 100,000 residents |
| 9 | 18.83 | 17.33 | 1 | Early death | 18.10 | 12.2 | 17.4 | Suicide death rate | Uneven, worse at the deprived end | 0.810 | Deaths per 100,000 residents |
| 10 | 19.41 | 15.71 | 1 | Early death | 17.37 | 12.2 | 17.4 | Suicide death rate | Uneven, worse at the deprived end | 0.810 | Deaths per 100,000 residents |
| 1 | 5.44 | 4.19 | 1 | Early death | 4.82 | 4.8 | 13.6 | Homicide death rate, counties | Uneven, worse at the deprived end | 0.470 | Deaths per 100,000 residents |
| 2 | 9.17 | 5.65 | 1 | Early death | 7.31 | 4.8 | 13.6 | Homicide death rate, counties | Uneven, worse at the deprived end | 0.470 | Deaths per 100,000 residents |
| 3 | 9.02 | 6.40 | 1 | Early death | 7.86 | 4.8 | 13.6 | Homicide death rate, counties | Uneven, worse at the deprived end | 0.470 | Deaths per 100,000 residents |
| 4 | 14.24 | 8.53 | 1 | Early death | 11.44 | 4.8 | 13.6 | Homicide death rate, counties | Uneven, worse at the deprived end | 0.470 | Deaths per 100,000 residents |
| 5 | 12.33 | 7.48 | 1 | Early death | 9.89 | 4.8 | 13.6 | Homicide death rate, counties | Uneven, worse at the deprived end | 0.470 | Deaths per 100,000 residents |
| 6 | 9.99 | 6.87 | 1 | Early death | 8.45 | 4.8 | 13.6 | Homicide death rate, counties | Uneven, worse at the deprived end | 0.470 | Deaths per 100,000 residents |
| 7 | 10.00 | 6.19 | 1 | Early death | 7.86 | 4.8 | 13.6 | Homicide death rate, counties | Uneven, worse at the deprived end | 0.470 | Deaths per 100,000 residents |
| 8 | 7.73 | 5.66 | 1 | Early death | 6.63 | 4.8 | 13.6 | Homicide death rate, counties | Uneven, worse at the deprived end | 0.470 | Deaths per 100,000 residents |
| 9 | 10.31 | 7.40 | 1 | Early death | 8.79 | 4.8 | 13.6 | Homicide death rate, counties | Uneven, worse at the deprived end | 0.470 | Deaths per 100,000 residents |
| 10 | 16.17 | 11.14 | 1 | Early death | 13.63 | 4.8 | 13.6 | Homicide death rate, counties | Uneven, worse at the deprived end | 0.470 | Deaths per 100,000 residents |
| 1 | 2.16 | 1.35 | 1 | Early death | 1.76 | 1.8 | 14.8 | Homicide death rate, tracts | Steady, worse at the deprived end | 1 | Homicide deaths per 100,000 residents a year |
| 2 | 3.18 | 2.07 | 1 | Early death | 2.62 | 1.8 | 14.8 | Homicide death rate, tracts | Steady, worse at the deprived end | 1 | Homicide deaths per 100,000 residents a year |
| 3 | 3.48 | 2.63 | 1 | Early death | 3.03 | 1.8 | 14.8 | Homicide death rate, tracts | Steady, worse at the deprived end | 1 | Homicide deaths per 100,000 residents a year |
| 4 | 4.18 | 3.29 | 1 | Early death | 3.70 | 1.8 | 14.8 | Homicide death rate, tracts | Steady, worse at the deprived end | 1 | Homicide deaths per 100,000 residents a year |
| 5 | 4.69 | 3.97 | 1 | Early death | 4.33 | 1.8 | 14.8 | Homicide death rate, tracts | Steady, worse at the deprived end | 1 | Homicide deaths per 100,000 residents a year |
| 6 | 5.63 | 4.68 | 1 | Early death | 5.16 | 1.8 | 14.8 | Homicide death rate, tracts | Steady, worse at the deprived end | 1 | Homicide deaths per 100,000 residents a year |
| 7 | 7.09 | 5.55 | 1 | Early death | 6.27 | 1.8 | 14.8 | Homicide death rate, tracts | Steady, worse at the deprived end | 1 | Homicide deaths per 100,000 residents a year |
| 8 | 8.32 | 6.48 | 1 | Early death | 7.34 | 1.8 | 14.8 | Homicide death rate, tracts | Steady, worse at the deprived end | 1 | Homicide deaths per 100,000 residents a year |
| 9 | 9.59 | 7.41 | 1 | Early death | 8.43 | 1.8 | 14.8 | Homicide death rate, tracts | Steady, worse at the deprived end | 1 | Homicide deaths per 100,000 residents a year |
| 10 | 16.85 | 12.76 | 1 | Early death | 14.81 | 1.8 | 14.8 | Homicide death rate, tracts | Steady, worse at the deprived end | 1 | Homicide deaths per 100,000 residents a year |
| 1 | 82.55 | 81.99 | 1 | Early death | 82.29 | 82.3 | 74.0 | Life expectancy at birth | Steady, worse at the deprived end | −1 | Life expectancy at birth (years) |
| 2 | 81.21 | 80.73 | 1 | Early death | 80.95 | 82.3 | 74.0 | Life expectancy at birth | Steady, worse at the deprived end | −1 | Life expectancy at birth (years) |
| 3 | 80.52 | 80.07 | 1 | Early death | 80.30 | 82.3 | 74.0 | Life expectancy at birth | Steady, worse at the deprived end | −1 | Life expectancy at birth (years) |
| 4 | 79.91 | 79.58 | 1 | Early death | 79.74 | 82.3 | 74.0 | Life expectancy at birth | Steady, worse at the deprived end | −1 | Life expectancy at birth (years) |
| 5 | 79.27 | 78.97 | 1 | Early death | 79.12 | 82.3 | 74.0 | Life expectancy at birth | Steady, worse at the deprived end | −1 | Life expectancy at birth (years) |
| 6 | 78.54 | 78.25 | 1 | Early death | 78.39 | 82.3 | 74.0 | Life expectancy at birth | Steady, worse at the deprived end | −1 | Life expectancy at birth (years) |
| 7 | 77.82 | 77.56 | 1 | Early death | 77.69 | 82.3 | 74.0 | Life expectancy at birth | Steady, worse at the deprived end | −1 | Life expectancy at birth (years) |
| 8 | 77.03 | 76.75 | 1 | Early death | 76.88 | 82.3 | 74.0 | Life expectancy at birth | Steady, worse at the deprived end | −1 | Life expectancy at birth (years) |
| 9 | 76.03 | 75.72 | 1 | Early death | 75.87 | 82.3 | 74.0 | Life expectancy at birth | Steady, worse at the deprived end | −1 | Life expectancy at birth (years) |
| 10 | 74.32 | 73.72 | 1 | Early death | 74.01 | 82.3 | 74.0 | Life expectancy at birth | Steady, worse at the deprived end | −1 | Life expectancy at birth (years) |

Source: Figures of the chapters named in each row, data years 1978 to 2024 as each chapter states; estimates as published there: full counts, crude rates, population-weighted means, unweighted survey means or modeled estimates.

Data: https://vishalsingh.org/zipcode/report/how-far-one-ranking-goes/figures/gradient_sparklines.json

## All estimates

Every fact in facts.json (30), sorted by id.

| Fact id | Value | Unit | Estimand | Population | Period | Interval | n | Weights | Source | Note |
| --- | ---: | --- | --- | --- | --- | --- | ---: | --- | --- | --- |
| alt_changed_degree | 1 | chapters | counted main gradients that change under the bachelor's degree share | the main gradients of the chapters counted by the ranking switch rule | as in each chapter |   |   | none (count of chapters) | Area Deprivation Index |   |
| alt_changed_income | 2 | chapters | counted main gradients that change under the median household income | the main gradients of the chapters counted by the ranking switch rule | as in each chapter |   |   | none (count of chapters) | Area Deprivation Index |   |
| alt_changed_std | 1 | chapters | counted main gradients that change under the standardized index | the main gradients of the chapters counted by the ranking switch rule | as in each chapter |   |   | none (count of chapters) | Area Deprivation Index |   |
| alt_n_counted | 16 | chapters | chapters whose main gradient enters the count of the rule | the main gradients of the chapters counted by the ranking switch rule | as in each chapter |   |   | none (count of chapters) | Area Deprivation Index |   |
| alt_n_recomputed | 17 | chapters | chapters whose main gradient was recomputed under four rankings | the main gradients of the chapters counted by the ranking switch rule | as in each chapter |   |   | none (count of chapters) | Area Deprivation Index |   |
| alt_switch_changes | 4 | chapters | rule: changed main gradients under the standardized index that switch the ranking | the main gradients of the chapters counted by the ranking switch rule | as in each chapter |   |   | none (count of chapters) | Area Deprivation Index |   |
| alt_switch_reversals | 2 | chapters | rule: reversed main gradients under the standardized index that switch the ranking | the main gradients of the chapters counted by the ranking switch rule | as in each chapter |   |   | none (count of chapters) | Area Deprivation Index |   |
| checkup_least | 74.2 | percent of adults with a routine checkup in the past year | bin mean as published in Place and the length of a life | least deprived bin of tenth of tracts by Area Deprivation Index (1 = least deprived, 10 = most deprived) | as in Place and the length of a life | 95% CI 73.2 to 75.3 |   | as in Place and the length of a life | CDC PLACES | Copied from the figure rows of Place and the length of a life. |
| checkup_most | 77.8 | percent of adults with a routine checkup in the past year | bin mean as published in Place and the length of a life | most deprived bin of tenth of tracts by Area Deprivation Index (1 = least deprived, 10 = most deprived) | as in Place and the length of a life | 95% CI 77.3 to 78.2 |   | as in Place and the length of a life | CDC PLACES | Copied from the figure rows of Place and the length of a life. |
| hom_cty_rho | +0.47 | rank correlation between bin order and bin mean | Spearman correlation over the published bins | homicide death rate, counties, bins of tenth of counties by Area Deprivation Index (1 = least deprived, 10 = most deprived) | as in Overdose, alcohol and suicide |   |   | none (bins counted once) | CDC injury and violence mapping |   |
| hom_tract_rho | +1.00 | rank correlation between bin order and bin mean | Spearman correlation over the published bins | homicide death rate, tracts, bins of tenth of tracts by Area Deprivation Index (1 = least deprived, 10 = most deprived) | as in Violence and the life-expectancy gap |   |   | none (bins counted once) | CDC injury and violence mapping |   |
| n_better | 5 | outcome gradients | count of rows of the pattern table that meet the stated rule | outcome gradients that the chapters of this book publish by bins of a ranking of places | as in each chapter |   |   | none (count of rows) | Area Deprivation Index |   |
| n_chapters | 18 | chapters | count of rows of the pattern table that meet the stated rule | outcome gradients that the chapters of this book publish by bins of a ranking of places | as in each chapter |   |   | none (count of rows) | Area Deprivation Index |   |
| n_far_least | 8 | outcome gradients | count of ratio-scale rows of the pattern table meeting the stated rule (log distance of each end from the middle of the ranking) | outcome gradients that the chapters of this book publish by bins of a ranking of places | as in each chapter |   |   | none (count of rows) | Area Deprivation Index |   |
| n_far_most | 17 | outcome gradients | count of ratio-scale rows of the pattern table meeting the stated rule (log distance of each end from the middle of the ranking) | outcome gradients that the chapters of this book publish by bins of a ranking of places | as in each chapter |   |   | none (count of rows) | Area Deprivation Index |   |
| n_flat | 2 | outcome gradients | count of rows of the pattern table that meet the stated rule | outcome gradients that the chapters of this book publish by bins of a ranking of places | as in each chapter |   |   | none (count of rows) | Area Deprivation Index |   |
| n_no_reading | 4 | outcome gradients | count of rows of the pattern table that meet the stated rule | outcome gradients that the chapters of this book publish by bins of a ranking of places | as in each chapter |   |   | none (count of rows) | Area Deprivation Index |   |
| n_ratio_rows | 25 | outcome gradients | count of ratio-scale rows of the pattern table meeting the stated rule (log distance of each end from the middle of the ranking) | outcome gradients that the chapters of this book publish by bins of a ranking of places | as in each chapter |   |   | none (count of rows) | Area Deprivation Index |   |
| n_rows | 30 | outcome gradients | count of rows of the pattern table that meet the stated rule | outcome gradients that the chapters of this book publish by bins of a ranking of places | as in each chapter |   |   | none (count of rows) | Area Deprivation Index |   |
| n_steady_worse | 17 | outcome gradients | count of rows of the pattern table that meet the stated rule | outcome gradients that the chapters of this book publish by bins of a ranking of places | as in each chapter |   |   | none (count of rows) | Area Deprivation Index |   |
| n_uneven | 4 | outcome gradients | count of rows of the pattern table that meet the stated rule | outcome gradients that the chapters of this book publish by bins of a ranking of places | as in each chapter |   |   | none (count of rows) | Area Deprivation Index |   |
| n_uneven_worse | 3 | outcome gradients | count of rows of the pattern table that meet the stated rule | outcome gradients that the chapters of this book publish by bins of a ranking of places | as in each chapter |   |   | none (count of rows) | Area Deprivation Index |   |
| n_worse | 20 | outcome gradients | count of rows of the pattern table that meet the stated rule | outcome gradients that the chapters of this book publish by bins of a ranking of places | as in each chapter |   |   | none (count of rows) | Area Deprivation Index |   |
| od_least | 23.8 | overdose deaths per 100,000 residents | bin mean as published in Overdose, alcohol and suicide | least deprived bin of tenth of counties by Area Deprivation Index (1 = least deprived, 10 = most deprived) | as in Overdose, alcohol and suicide | 95% CI 22.2 to 25.6 |   | as in Overdose, alcohol and suicide | CDC injury and violence mapping | Copied from the figure rows of Overdose, alcohol and suicide. |
| od_most | 31.1 | overdose deaths per 100,000 residents | bin mean as published in Overdose, alcohol and suicide | most deprived bin of tenth of counties by Area Deprivation Index (1 = least deprived, 10 = most deprived) | as in Overdose, alcohol and suicide | 95% CI 25.7 to 36.5 |   | as in Overdose, alcohol and suicide | CDC injury and violence mapping | Copied from the figure rows of Overdose, alcohol and suicide. |
| od_rho | +0.09 | rank correlation between bin order and bin mean | Spearman correlation over the published bins | overdose death rate, bins of tenth of counties by Area Deprivation Index (1 = least deprived, 10 = most deprived) | as in Overdose, alcohol and suicide |   |   | none (bins counted once) | CDC injury and violence mapping |   |
| rho_cut | 0.9 | absolute rank correlation between bin order and bin mean | definition: threshold for a steady gradient | the rule of this chapter | not applicable (definition) |   |   | none (definition) | Area Deprivation Index |   |
| split_black_cut | 25% | percent of county residents who are Black alone | definition: the share of residents at or above which a county counts as having a large Black population (docs/RANKINGS_AND_SPLITS_DESIGN.md) | counties of the 50 states and DC | ACS 2019-2023 |   |   | none (definition) | American Community Survey |   |
| traffic_least | 74.7 | mean national percentile of tract traffic proximity | bin mean as published in Air pollution | least deprived bin of decile of the Area Deprivation Index national rank (1 = least deprived, 10 = most deprived) | as in Air pollution | 95% CI 68.0 to 79.2 |   | as in Air pollution | EPA EJScreen | Copied from the figure rows of Air pollution. |
| traffic_most | 39.0 | mean national percentile of tract traffic proximity | bin mean as published in Air pollution | most deprived bin of decile of the Area Deprivation Index national rank (1 = least deprived, 10 = most deprived) | as in Air pollution | 95% CI 35.4 to 42.1 |   | as in Air pollution | EPA EJScreen | Copied from the figure rows of Air pollution. |

## Definitions

- Area Deprivation Index (ADI): A score that ranks neighborhoods by income, education, employment and housing conditions drawn from the American Community Survey. In this book a higher national percentile means a more deprived neighborhood.

## Methods

This chapter reads the binned gradients that other chapters publish and computes no new estimate from data. The rows come from different sources, places, periods and units, so the chapter counts patterns under a stated rule and does not compare their sizes.

### Which gradients are included

One row per outcome whose chapter publishes a mean and a 95% interval for each tenth, fifth or quarter of places ranked by the Area Deprivation Index, or, for the gender gaps in test scores, by district median household income. Outcomes that a chapter reports only as two ends, as a scatter without binned means or against another ranking were left out, among them upward mobility, test score levels and the jobs held by residents.

Each row keeps the bins, weights and interval method of its chapter. The table names the ranking and the unit of every row.

Explains Table 1.

### The classification rule

The ends differ when the 95% intervals of the least and the most deprived bins do not overlap. This is conservative: two estimates whose intervals touch can still differ.

A row whose ends differ is steady when the rank correlation between the order of the bins and the bin means is at least 0.9 in absolute value, and uneven otherwise. A row is worse or better at the deprived end when its chapter's outcome has a direction that is worse for residents (a higher death rate, a lower turnout); gender gaps in scores, the predicted teen birth share and emergency visits per Medicare beneficiary carry no such reading.

Explains Table 1 and Table 2.

### The two ends against the national value

For each row whose values have a natural zero, the chapter divides the mean of the least and of the most deprived bin by the national value that the row's own chapter publishes: the same estimator, weights, outcome and years applied to all the places of the comparison as one bin, a population-weighted mean, a pooled rate or a pooled ratio as the bins are. The places are those the chapter ranks, so a place without a rank or an estimate is in neither the bins nor the national value.

The end that lies farther from the national value is the one whose ratio is farther from one on a log scale, so that twice and half the national value count as the same distance. Because the national value weights places by their population, births or households, it sits nearer the larger, less deprived places when they hold more of the weight. The ratios carry no intervals. The gender gaps in scores, the two pollution percentiles and the ladder score have no natural zero and are left out.

Explains Table 3.

### The ranking switch rule

The rule was fixed on October 28, 2026, before any alternative gradient was computed. Each chapter that ranks places by the deprivation index recomputes its main gradient under the published index, a standardized index, median household income and the bachelor's degree share, over the places that have all four, with its own outcome, weights, bins and intervals. This chapter classifies those rows with the rule above.

An alternative changes a gradient when the worse end switches sides, when the ends differ under one ranking and not the other, or when the rank correlation crosses 0.9 and moves by a stated margin. The standardized index would replace the deprivation index as the book's primary ranking if it changed at least 4 of the 16 counted gradients or reversed at least 2. Income and degree share are checks and cannot switch the ranking. Renting is recomputed and left out of the count, because rent and income are inputs to both indices.

Explains Table 4.

### Shared inputs and comparability

Several rows are modeled estimates built partly from census measures that the deprivation index also uses (the PLACES health measures, the CDC injury panel), and the housing-cost rows divide rent or owner costs by income, two ingredients of the index. Agreement between such a row and the ranking is partly built in.

The rows describe tracts, counties, districts or interviews in different years, so no row is a sample from the same places as another, and a count of steady rows is a count of chapters' findings, not an estimate for one population.

## What these data cannot show

The rows come from different sources, places, periods and units. Births are ranked by county, life expectancy by tract, and the gender gaps by district, and the outcome periods differ from one row to the next and from the period of the index, 2019 to 2023. The count is not an estimate for one population and does not measure how strongly places shape any outcome. A steady row describes an ordering of places and not the experience of every resident, and the classification says nothing about why an outcome follows the ranking or departs from it.

## References

The chapter cites no reference.

## Provenance

- Pipeline version: 0.1.0
- Build script: book2/chapters/how-far-one-ranking-goes/build.py (sha256 fe832d8cc9cd0650417e733e53391cd8081f26368ea3152b4962a900f3511719)
- Built on: 2026-10-28
- Site commit: c2b1ff3309cb4a10833f884da91ac0057a66ad0a

Inputs:

| File | Size (bytes) | SHA-256, first 1 MB | SHA-256, last 1 MB |
| --- | ---: | --- | --- |
| C:/github/zipcode-book/content/chapters/air-pollution/data/figures/air_vs_deprivation.json | 9,202 | f39920e02e942de8030bc8e4d33c42062e9a7775403ac2ca2f7adc7647c2dcb9 | f39920e02e942de8030bc8e4d33c42062e9a7775403ac2ca2f7adc7647c2dcb9 |
| C:/github/zipcode-book/content/chapters/air-pollution/data/figures/alt_rankings.json | 9,867 | fb9b7c1a5a852c52c286a80d9fd5a275f943fc01e710b5e3f23814aacf966268 | fb9b7c1a5a852c52c286a80d9fd5a275f943fc01e710b5e3f23814aacf966268 |
| C:/github/zipcode-book/content/chapters/birthweight-and-preterm-birth/data/facts.json | 39,621 | 9332a0015b74d0cb3babb0d72c1b75129ee521b3719bb6e2d1abf87321e9308a | 9332a0015b74d0cb3babb0d72c1b75129ee521b3719bb6e2d1abf87321e9308a |
| C:/github/zipcode-book/content/chapters/birthweight-and-preterm-birth/data/figures/alt_rankings.json | 9,623 | 2b4facd8450b8f46d33c77517e55ab89b6e87e3383c8cbbc365b108150ce1206 | 2b4facd8450b8f46d33c77517e55ab89b6e87e3383c8cbbc365b108150ce1206 |
| C:/github/zipcode-book/content/chapters/birthweight-and-preterm-birth/data/figures/lbw_by_deprivation.json | 109,035 | d98d50597d29c7dc3a12ce69ff8d5abf68a205051a62824ffb30791b0c827796 | d98d50597d29c7dc3a12ce69ff8d5abf68a205051a62824ffb30791b0c827796 |
| C:/github/zipcode-book/content/chapters/civic-life-loneliness-and-partnership/data/facts.json | 72,807 | 885b22dae15bf5507523280f10d2ec8ad50a4009d4607d9ea4e70c0c6fa8d1c6 | 885b22dae15bf5507523280f10d2ec8ad50a4009d4607d9ea4e70c0c6fa8d1c6 |
| C:/github/zipcode-book/content/chapters/civic-life-loneliness-and-partnership/data/figures/alt_rankings.json | 9,589 | ee02e7e5c4deefeb5d2b37f55efc742af66468028d2cd7818d70a22f684b901b | ee02e7e5c4deefeb5d2b37f55efc742af66468028d2cd7818d70a22f684b901b |
| C:/github/zipcode-book/content/chapters/civic-life-loneliness-and-partnership/data/figures/civic_gradient.json | 14,822 | 16152938d8fc8a1900bb13ce35aeea159621130649fbd0567a20ee5cb1ee1795 | 16152938d8fc8a1900bb13ce35aeea159621130649fbd0567a20ee5cb1ee1795 |
| C:/github/zipcode-book/content/chapters/college-access-price-and-earnings/data/facts.json | 78,352 | a880df2b6f42b0dd852789f716c41a3f6a0d06c99f5404b1c9ff66f4f10896c8 | a880df2b6f42b0dd852789f716c41a3f6a0d06c99f5404b1c9ff66f4f10896c8 |
| C:/github/zipcode-book/content/chapters/college-access-price-and-earnings/data/figures/alt_rankings.json | 9,595 | 565c3e97445628662450b630a7df8f411aca8527313130186ce0589f5066fbc7 | 565c3e97445628662450b630a7df8f411aca8527313130186ce0589f5066fbc7 |
| C:/github/zipcode-book/content/chapters/college-access-price-and-earnings/data/figures/college_desert_gradient.json | 12,004 | 836f2c34fa05985f2ddb99b36c36eae630814edf58392f81ec21fafc12ac0327 | 836f2c34fa05985f2ddb99b36c36eae630814edf58392f81ec21fafc12ac0327 |
| C:/github/zipcode-book/content/chapters/disability-retirement-and-medicare/data/facts.json | 104,878 | 9ff34bbcae9bdf00191de17cf396fec86f8485d0c6e8ca7921d5ae0c4b04fbf6 | 9ff34bbcae9bdf00191de17cf396fec86f8485d0c6e8ca7921d5ae0c4b04fbf6 |
| C:/github/zipcode-book/content/chapters/disability-retirement-and-medicare/data/figures/alt_rankings.json | 9,603 | 2f174623d60889d3db6dc07afecd828379c703363700dd058f5e9f54c0852012 | 2f174623d60889d3db6dc07afecd828379c703363700dd058f5e9f54c0852012 |
| C:/github/zipcode-book/content/chapters/disability-retirement-and-medicare/data/figures/disability_gradient.json | 13,775 | 3275b35c2b470ec413dc94c8610de3a0433d0c045c74557c1d39e7cfbc820406 | 3275b35c2b470ec413dc94c8610de3a0433d0c045c74557c1d39e7cfbc820406 |
| C:/github/zipcode-book/content/chapters/disability-retirement-and-medicare/data/figures/medicare_spending.json | 25,405 | 2b213e89a79fdd6668c7cd45010381cad8f5536df3e7a1175ffbff6437cf9597 | 2b213e89a79fdd6668c7cd45010381cad8f5536df3e7a1175ffbff6437cf9597 |
| C:/github/zipcode-book/content/chapters/gender-gaps-in-reading-and-math/data/figures/gap_by_income_decile.json | 7,769 | a03a2c5902a4a1d61ad9cede1dec83114f09eeed7325a98fa43fdd93d3af6087 | a03a2c5902a4a1d61ad9cede1dec83114f09eeed7325a98fa43fdd93d3af6087 |
| C:/github/zipcode-book/content/chapters/gun-homicide-and-causes-of-early-death/data/facts.json | 70,205 | 3e69869e30dd34d1912643bf04bf67c57c394e53032fca8b0d878efe9f9a2242 | 3e69869e30dd34d1912643bf04bf67c57c394e53032fca8b0d878efe9f9a2242 |
| C:/github/zipcode-book/content/chapters/gun-homicide-and-causes-of-early-death/data/figures/alt_rankings.json | 9,622 | caae3d0ba756606aa3c3be21d3cb4a42d1b4f1c5a131b8213e6ceb0d65616748 | caae3d0ba756606aa3c3be21d3cb4a42d1b4f1c5a131b8213e6ceb0d65616748 |
| C:/github/zipcode-book/content/chapters/gun-homicide-and-causes-of-early-death/data/figures/premature_death_gradient.json | 4,093 | 9f6caebfdf9bf42e3502ddab55fa820d9a160aa265130d037db8ac7f0cdd33f3 | 9f6caebfdf9bf42e3502ddab55fa820d9a160aa265130d037db8ac7f0cdd33f3 |
| C:/github/zipcode-book/content/chapters/health-behaviors-and-the-supply-of-care/data/facts.json | 74,804 | cb3c3e891db62b647d899ac6fb4e094e6cfb9eba07c7713713a4648aa23d0e0f | cb3c3e891db62b647d899ac6fb4e094e6cfb9eba07c7713713a4648aa23d0e0f |
| C:/github/zipcode-book/content/chapters/health-behaviors-and-the-supply-of-care/data/figures/alt_rankings.json | 9,667 | 0e6a524e2d02f3d85fc9200381f43c0c1d9992221f13382a6b8759f7451f7cba | 0e6a524e2d02f3d85fc9200381f43c0c1d9992221f13382a6b8759f7451f7cba |
| C:/github/zipcode-book/content/chapters/health-behaviors-and-the-supply-of-care/data/figures/people_per_provider.json | 11,766 | afc13ef7ae3be5189fe198b6a6f18723b761935c3e00a6a816c76d7fbc30dbb5 | afc13ef7ae3be5189fe198b6a6f18723b761935c3e00a6a816c76d7fbc30dbb5 |
| C:/github/zipcode-book/content/chapters/infant-mortality/data/facts.json | 53,465 | 692b8c0a99be5f6003eee48dd12459e7f65edc244298bf9b5c1f6044d810cfc6 | 692b8c0a99be5f6003eee48dd12459e7f65edc244298bf9b5c1f6044d810cfc6 |
| C:/github/zipcode-book/content/chapters/infant-mortality/data/figures/alt_rankings.json | 9,554 | f65e1e27fd5688264db1374dadbdc45f9ec310c1c979dcb715318361b537b178 | f65e1e27fd5688264db1374dadbdc45f9ec310c1c979dcb715318361b537b178 |
| C:/github/zipcode-book/content/chapters/infant-mortality/data/figures/imr_by_deprivation.json | 63,114 | ebe5d39824c8881f3b66cf5662e8da7229643c1894815e73029362570c2ff062 | ebe5d39824c8881f3b66cf5662e8da7229643c1894815e73029362570c2ff062 |
| C:/github/zipcode-book/content/chapters/life-evaluation-and-daily-feelings/data/figures/alt_rankings.json | 9,668 | 3eac056a094f131714b25d7ccf86a56f742087fb29f3672721187b0e41f7d2a0 | 3eac056a094f131714b25d7ccf86a56f742087fb29f3672721187b0e41f7d2a0 |
| C:/github/zipcode-book/content/chapters/life-evaluation-and-daily-feelings/data/figures/ladder_by_deprivation.json | 22,947 | b563618cd2af45b968f7d7d33c2b7b1c36d5f02d2f51832cd765483997aff533 | b563618cd2af45b968f7d7d33c2b7b1c36d5f02d2f51832cd765483997aff533 |
| C:/github/zipcode-book/content/chapters/mortgage-denials/data/facts.json | 83,097 | 753a34074702cd60ad13949eee9ff31cd4f2a6b4b4433b30561ff0352bff0f01 | 753a34074702cd60ad13949eee9ff31cd4f2a6b4b4433b30561ff0352bff0f01 |
| C:/github/zipcode-book/content/chapters/mortgage-denials/data/figures/alt_rankings.json | 9,641 | 773da62209fd79716ba73b2fab4d60c0e69d89cc89b3e144c91194c4f7af73ec | 773da62209fd79716ba73b2fab4d60c0e69d89cc89b3e144c91194c4f7af73ec |
| C:/github/zipcode-book/content/chapters/mortgage-denials/data/figures/denial_vs_deprivation.json | 4,203 | 1bea4e523acf3849a3eb019195ce80b6b66a41af203a003863112da81f25fec0 | 1bea4e523acf3849a3eb019195ce80b6b66a41af203a003863112da81f25fec0 |
| C:/github/zipcode-book/content/chapters/multiple-chronic-conditions/data/facts.json | 81,354 | 4e51b97b62ba016b5bafc582e374b83382ed094ee0257de05d7d6736e14c222f | 4e51b97b62ba016b5bafc582e374b83382ed094ee0257de05d7d6736e14c222f |
| C:/github/zipcode-book/content/chapters/multiple-chronic-conditions/data/figures/alt_rankings.json | 9,650 | 5c95d266eb3e2f097c8cf6025e189e39af38e1c1609ca88792c64fe2e055896f | 5c95d266eb3e2f097c8cf6025e189e39af38e1c1609ca88792c64fe2e055896f |
| C:/github/zipcode-book/content/chapters/multiple-chronic-conditions/data/figures/high_count_by_adi.json | 6,432 | fa06d9dd565c8bd59515de2b4ecca0fdc50503679b1b5e1f6f7fc8aeb2686ac8 | fa06d9dd565c8bd59515de2b4ecca0fdc50503679b1b5e1f6f7fc8aeb2686ac8 |
| C:/github/zipcode-book/content/chapters/overdose-alcohol-and-suicide/data/facts.json | 81,979 | 63d63863e78931b4e81f3b1a9139d7c8f996c5de86982d97219967944bc2f0e4 | 63d63863e78931b4e81f3b1a9139d7c8f996c5de86982d97219967944bc2f0e4 |
| C:/github/zipcode-book/content/chapters/overdose-alcohol-and-suicide/data/figures/alt_rankings.json | 9,707 | d65045934802ba7de6c7fac3a6b91557460db6f900774a99b519294cb9fe8c5f | d65045934802ba7de6c7fac3a6b91557460db6f900774a99b519294cb9fe8c5f |
| C:/github/zipcode-book/content/chapters/overdose-alcohol-and-suicide/data/figures/three_causes_by_decile.json | 17,224 | 92ef9f8f26ff64531896da0f23322aae3d67e8acc808bdc546fa5587efa9bbd9 | 92ef9f8f26ff64531896da0f23322aae3d67e8acc808bdc546fa5587efa9bbd9 |
| C:/github/zipcode-book/content/chapters/place-and-the-length-of-a-life/data/facts.json | 71,140 | 9eed1ad6694ae817ab5800fd787f79ec8a01a706b2d211bccb978f17d11e6e6d | 9eed1ad6694ae817ab5800fd787f79ec8a01a706b2d211bccb978f17d11e6e6d |
| C:/github/zipcode-book/content/chapters/place-and-the-length-of-a-life/data/figures/alt_rankings.json | 9,596 | 09e34b229fe46e41cd294da684e89b20b3cfa35d04a5f3688c6995e96678516e | 09e34b229fe46e41cd294da684e89b20b3cfa35d04a5f3688c6995e96678516e |
| C:/github/zipcode-book/content/chapters/place-and-the-length-of-a-life/data/figures/le_gradient.json | 32,837 | fec8e001d57c3dc7cdce2699846d13a9b38ae12d162c014ea0adf565f67f48ee | fec8e001d57c3dc7cdce2699846d13a9b38ae12d162c014ea0adf565f67f48ee |
| C:/github/zipcode-book/content/chapters/renting-affordability-and-eviction/data/facts.json | 101,510 | e406f8a10f637011935e5a498813d757cff8c7fa2eb2a6eec5f4d034d769f947 | e406f8a10f637011935e5a498813d757cff8c7fa2eb2a6eec5f4d034d769f947 |
| C:/github/zipcode-book/content/chapters/renting-affordability-and-eviction/data/figures/alt_rankings.json | 10,143 | ad5422a3e2801029f9a69d147554ac6c36d9a6fa9064f8ac23a0418bf8a2dcd4 | ad5422a3e2801029f9a69d147554ac6c36d9a6fa9064f8ac23a0418bf8a2dcd4 |
| C:/github/zipcode-book/content/chapters/renting-affordability-and-eviction/data/figures/burden_by_deprivation.json | 22,558 | a0f766e578479473605b0a240a240514814de8fe7aab2168bc7207dd044e2eee | a0f766e578479473605b0a240a240514814de8fe7aab2168bc7207dd044e2eee |
| C:/github/zipcode-book/content/chapters/school-discipline-and-incarceration/data/facts.json | 60,956 | 88a92432b6bedeb487d217f56ccb7d2d909106e9fd65f23139263eec30bcf089 | 88a92432b6bedeb487d217f56ccb7d2d909106e9fd65f23139263eec30bcf089 |
| C:/github/zipcode-book/content/chapters/school-discipline-and-incarceration/data/figures/alt_rankings.json | 9,779 | aadc8b9fe818480dbeadcf72b8d936746dcbf813936dd21af727ebb3dde2fb1f | aadc8b9fe818480dbeadcf72b8d936746dcbf813936dd21af727ebb3dde2fb1f |
| C:/github/zipcode-book/content/chapters/school-discipline-and-incarceration/data/figures/incarceration_gradient.json | 29,100 | 0d8a4e6e4eb5ac7f097793036c26d3d7c89093f9a6905319b871d1af4212228d | 0d8a4e6e4eb5ac7f097793036c26d3d7c89093f9a6905319b871d1af4212228d |
| C:/github/zipcode-book/content/chapters/teen-births/data/facts.json | 92,192 | 58432d8c5da5f66f3d277a165542e5ff751f6f1d34af62ba767f6d81a2d88c12 | 58432d8c5da5f66f3d277a165542e5ff751f6f1d34af62ba767f6d81a2d88c12 |
| C:/github/zipcode-book/content/chapters/teen-births/data/figures/alt_rankings.json | 9,737 | 0a0cf476dc955f9ffece13e9d3fd238082a9c44fd2ec2d441ba0800c394cd651 | 0a0cf476dc955f9ffece13e9d3fd238082a9c44fd2ec2d441ba0800c394cd651 |
| C:/github/zipcode-book/content/chapters/teen-births/data/figures/teen_birth_by_deprivation_tracts.json | 4,830 | bc9c3f68f15ce4f0a78770d0e8cf7d9f94ca648adbe42d20fe9bb7fbae6cdd78 | bc9c3f68f15ce4f0a78770d0e8cf7d9f94ca648adbe42d20fe9bb7fbae6cdd78 |
| C:/github/zipcode-book/content/chapters/violence-and-the-life-expectancy-gap/data/facts.json | 79,485 | 4e4801a65b00f2df89da49274a1b39684d11d8a0ddd7a43db764c7e090b9ebde | 4e4801a65b00f2df89da49274a1b39684d11d8a0ddd7a43db764c7e090b9ebde |
| C:/github/zipcode-book/content/chapters/violence-and-the-life-expectancy-gap/data/figures/alt_rankings.json | 9,630 | 6bd94f9150a3728846137c9bfb01005366e66051f66f74f12d70780c2b6e1986 | 6bd94f9150a3728846137c9bfb01005366e66051f66f74f12d70780c2b6e1986 |
| C:/github/zipcode-book/content/chapters/violence-and-the-life-expectancy-gap/data/figures/homicide_by_tenth.json | 4,141 | 70f57e2d7ce7c5f1c7ec3c733a50b8e094d8a482f41e5bd3958c36d192f3ca47 | 70f57e2d7ce7c5f1c7ec3c733a50b8e094d8a482f41e5bd3958c36d192f3ca47 |
| C:/github/zipcode-book/public/data/place-and-the-length-of-a-life/splits.json | 274,392 | 63e1adbb1d7ae3354bb1e224c982a8c535e1e1137340d5ae18d3d5e613b3c353 | 63e1adbb1d7ae3354bb1e224c982a8c535e1e1137340d5ae18d3d5e613b3c353 |

Outputs:

| File | SHA-256 |
| --- | --- |
| facts.json | d9eee2e8e34b82b69d1588ea09d316fcfd6359a427dca3c2b8d6df2536c0a23f |
| figures/ends_splits.json | b8233611105529501f84abdf718d5e533b8b5e75b2c6d09bc334d4be65b740db |
| figures/ends_vs_us.json | bd0996f5b4db2ac8bf6454f650dda1df440fe65dc2455d87a8d83dca8b71ff7b |
| figures/gradient_sparklines.json | a6696520b765c3d6660e804121f80b8b61f89b8705aef819a18bad14985af308 |
| figures/gradients_by_domain.json | ca072bde78f63590464ff557f947ed7dc500899c5064249764dd1672f82a4e9e |
| figures/pattern_table.json | e46e415e8cf8aee1a8f11726c708aa4982b3ca52a737c556610c644c8d27f788 |
| figures/ranking_check.json | c8b77eeb16009d8a6dee80ee5df52ea301124c46b1652b500032cc618e402ac6 |
| meta.json | bf04cdaaccb5b3d7a557d12e9bbcf782e17b9b13699423e3533f839b18243c88 |
| refs.json | ca3d163bab055381827226140568f3bef7eaac187cebd76878e0b63e9e442356 |

## Machine formats

- Report (HTML): https://vishalsingh.org/zipcode/how-far-one-ranking-goes/report
- Report (Markdown): https://vishalsingh.org/zipcode/how-far-one-ranking-goes/report.md
- Every estimate (JSON): https://vishalsingh.org/zipcode/report/how-far-one-ranking-goes/facts.json
- Table 1, figure "pattern_table" (JSON): https://vishalsingh.org/zipcode/report/how-far-one-ranking-goes/figures/pattern_table.json
- Table 2, figure "gradients_by_domain" (JSON): https://vishalsingh.org/zipcode/report/how-far-one-ranking-goes/figures/gradients_by_domain.json
- Table 3, figure "ends_vs_us" (JSON): https://vishalsingh.org/zipcode/report/how-far-one-ranking-goes/figures/ends_vs_us.json
- Table 4, figure "ranking_check" (JSON): https://vishalsingh.org/zipcode/report/how-far-one-ranking-goes/figures/ranking_check.json
- Table 5, figure "ends_splits" (JSON): https://vishalsingh.org/zipcode/report/how-far-one-ranking-goes/figures/ends_splits.json
- Table 6, figure "gradient_sparklines" (JSON): https://vishalsingh.org/zipcode/report/how-far-one-ranking-goes/figures/gradient_sparklines.json
- Index of every report: https://vishalsingh.org/zipcode/llms.txt; all reports in one file: https://vishalsingh.org/zipcode/llms-full.txt

---

# Eight types of neighborhood: report

Chapter 9.2 of ZIP Code Destiny, by Vishal Singh (NYU Stern). Generated from the chapter's data files; the reading version is at https://vishalsingh.org/zipcode/eight-types-of-neighborhood. Table n of this report is figure n of the chapter.

## Identification

- Title: Eight types of neighborhood
- Label: The eight Americas
- Chapter: 9.2 (Part IX, Reading the patterns; Reading the patterns)
- Question: Can American tracts be sorted into a small number of types, and how far apart are their outcomes?
- Dek: Mean life expectancy for deaths in 2010 to 2015 runs from 74.7 to 82.1 years across eight overlapping types of census tract, built by k-means from 81,343 tracts on 7 measures of income, poverty, education, family structure, tenure, age and density, with no outcome among them.
- Grain: neighborhoods
- Status: published
- Published: 2026-10-09
- Updated: 2026-10-28
- Data sources:
  - American Community Survey. ACS 5-year estimates 2019-2023 for 2020 tracts: income, poverty, education, households, tenure, age and population (the inputs), and race and ethnicity (compared by type, not an input); years: 2019-2023 (5-year); weights: ACS 2019-2023 tract population
  - Census TIGER/Line tracts. 2020 tract land area for population density; years: 2020; weights: none (land area)
  - CDC/ATSDR Social Vulnerability Index. CDC/ATSDR Social Vulnerability Index 2022: share of residents in group quarters, for the institutional-tract flag; years: 2022 release (ACS 2018-2022); weights: none (flag)
  - USALEEP life expectancy. NCHS small-area life expectancy at birth, deaths in 2010 to 2015, 2020 tracts (area-weighted); an outcome compared by type, not an input; years: 2010-2015; weights: ACS 2019-2023 tract population
  - Opportunity Atlas. Adult income rank, teen births and incarceration of children born 1978 to 1983 raised in families at the 25th income percentile, 2020 tracts (interpolated by land area); outcomes compared by type; years: 1978-1983 birth cohorts (2018 release); weights: ACS 2019-2023 tract population
  - CDC PLACES. CDC PLACES 2024 release (BRFSS 2022): diabetes, smoking, frequent mental distress and obesity among adults, modeled crude prevalence; outcomes compared by type; years: 2024 release (tracts); 2025 release (ZCTAs, 2023 data year); weights: ACS 2019-2023 tract population
  - Area Deprivation Index. Area Deprivation Index 2023 national rank, a profile column and not an input; years: 2023 release (version 4.0.1; built on ACS 2019-2023); weights: ACS 2019-2023 tract population
  - Published estimates from cited studies. The exposure effect quoted in the causal box (Chetty and Hendren 2018); years: as reported in each study; weights: as reported in the study
- Canonical URL: https://vishalsingh.org/zipcode/eight-types-of-neighborhood
- Report: https://vishalsingh.org/zipcode/eight-types-of-neighborhood/report
- Markdown: https://vishalsingh.org/zipcode/eight-types-of-neighborhood/report.md
- BibTeX key: singh2026eighttypesofneighborhood

```bibtex
@misc{singh2026eighttypesofneighborhood,
  author = {Singh, Vishal},
  title = {Eight types of neighborhood},
  year = {2026},
  url = {https://vishalsingh.org/zipcode/eight-types-of-neighborhood},
  note = {In ZIP Code Destiny. Published 9 October 2026; updated 28 October 2026.}
}
```

## Principal estimates

The facts the dek and the opening prose cite, in order of appearance.

| Estimand | Population | Period | Value | Interval | n | Weights | Source | Fact id |
| --- | --- | --- | ---: | --- | ---: | --- | --- | --- |
| population-weighted mean of the tract life expectancy over the tracts of the type, bootstrap interval over counties | residents of the tracts of type 8 (High-poverty, single-parent) that have the measure | deaths in 2010 to 2015 | 74.7 | 95% CI 74.2 to 75.1 | 4,595 tracts | ACS 2019-2023 tract population | USALEEP life expectancy | t8_le |
| population-weighted mean of the tract life expectancy over the tracts of the type, bootstrap interval over counties | residents of the tracts of type 1 (High-income owners) that have the measure | deaths in 2010 to 2015 | 82.1 | 95% CI 81.9 to 82.3 | 10,184 tracts | ACS 2019-2023 tract population | USALEEP life expectancy | t1_le |
| count of tracts in the typology | census tracts with every input, not institutional, with more residents than the floor | tract boundaries of 2020; inputs from ACS 2019-2023 | 81,343 |   |   | none (full count) | American Community Survey | n_tracts |
| count of inputs of the typology (definition) | measures standardized and given to k-means | ACS 2019-2023, 2020 land area | 7 |   |   | none (definition) | American Community Survey | n_features |
| difference between the population-weighted mean life expectancy of type 1 minus type 8 (High-income owners minus High-poverty, single-parent), bootstrap interval over counties with shared draws | residents of the tracts of types 1 and 8 that have a life expectancy | deaths in 2010 to 2015 | 7.5 | 95% CI 7.1 to 7.9 | 14,779 tracts | ACS 2019-2023 tract population | USALEEP life expectancy | le_gap |
| count of inputs of the earlier typology (definition) | measures standardized and given to k-means in the earlier typology | 2010 to 2023 | 13 |   |   | none (definition) | American Community Survey | n_features_old |
| count of typed tracts that have all thirteen inputs of the earlier typology (the tracts of the comparison) | census tracts in the typology | tract inputs from 2010 to 2023 | 77,745 |   |   | none (full count) | American Community Survey | n_compare |
| highest minus lowest population-weighted mean life expectancy across the groups of the chapter's earlier k-means fit on thirteen inputs, published until October 28, 2026, (the seven inputs of the typology plus the Black and Hispanic shares, life expectancy, the adult income rank, diabetes and smoking; same seed, groups and starts), on the tracts of the comparison, bootstrap interval over counties | residents of the typed tracts that have all thirteen inputs of the earlier typology | deaths in 2010 to 2015 | 10.0 | 95% CI 9.8 to 10.2 | 77,745 tracts | ACS 2019-2023 tract population | USALEEP life expectancy | le_spread_old |
| highest minus lowest population-weighted mean life expectancy across the groups of the eight types of this chapter, on the tracts of the comparison, bootstrap interval over counties | residents of the typed tracts that have all thirteen inputs of the earlier typology | deaths in 2010 to 2015 | 7.5 | 95% CI 7.1 to 7.9 | 77,745 tracts | ACS 2019-2023 tract population | USALEEP life expectancy | le_spread_c |
| mean silhouette of the eight types on a random sample of tracts drawn with a fixed seed | a random sample of census tracts in the typology | ACS 2019-2023, 2020 land area | 0.19 |   | 20,000 tracts | none (each tract counts once) | American Community Survey | sil_k8 |
| mean silhouette of an 8-group k-means fit to simulated data with no clusters: a multivariate normal with the correlations of the inputs and one draw per tract | a random sample of census tracts in the typology | ACS 2019-2023, 2020 land area | 0.14 |   | 20,000 tracts | none (each tract counts once) | American Community Survey | sil_gauss |

## Findings

Across eight types of census tract formed from income, poverty, education, family structure, tenure, age and density alone, mean life expectancy for deaths in 2010 to 2015 runs from 74.7 (95% CI 74.2 to 75.1) years in the high-poverty, single-parent type to 82.1 (95% CI 81.9 to 82.3) years among high-income owners, a gap of 7.5 (95% CI 7.1 to 7.9) years. K-means builds the types from 7 measures of the American Community Survey (ACS) for 2019 to 2023 on 81,343 tracts, and none of the inputs is an outcome or a measure of race or ethnicity. The chapter's earlier typology had life expectancy, mobility, diabetes, smoking and the Black and Hispanic shares among its 13 inputs, and on the same 77,745 tracts its types spanned 10.0 (95% CI 9.8 to 10.2) years of life expectancy, against 7.5 (95% CI 7.1 to 7.9) years for the types here. The types are not sharply separated. The silhouette of the typology, a score of how well tracts fit their own type against the nearest other, is 0.19, against 0.14 for simulated data that contain no clusters.

The names are this book's labels for groups of tracts that look alike on the inputs. The overlap suggests that the types are regions of a continuum. This chapter builds the types, maps them by county, compares them on eight outcomes and on the Black and Hispanic shares of their residents, sets them beside the earlier fit and draws the overlap.

Three other chapters hold related material. The split of life-expectancy variance among states, counties and tracts, and a principal-components analysis of nine measures, are in Chapter 10.1, Gradients, slopes and variance (https://vishalsingh.org/zipcode/gradients-slopes-and-variance). The parts of a deprivation index are in Chapter 1.1, What the deprivation index measures (https://vishalsingh.org/zipcode/anatomy-of-deprivation). A second clustering, built by k-means from the deprivation measures of the Area Deprivation Index, is in Chapter 10.2, Families of places (https://vishalsingh.org/zipcode/families-of-places). Neither clustering uses an outcome. The types here differ from those families because the share of older residents and population density are inputs. These two describe how a place is settled, and they set older owners, low-density owners and dense, college-educated renters apart from the other types.

### How the types were built

The typology starts from the 85,396 tracts of the 2020 census. Puerto Rico's 981 tracts are left out, because the county map and the outcome measures cover only the states and the District of Columbia. A further 1,498 tracts lack at least one input or have no residents, and 1,541 are institutional tracts with a large share of residents in group quarters such as prisons, nursing homes or dormitories. The institutional flag comes from the group-quarters share in the Social Vulnerability Index, so it does not depend on whether a tract has a life expectancy estimate.

Another 33 tracts have 200 or fewer residents. What remains is 81,343 tracts, which hold 324.7 million residents, 97.7% of the residents of the states and the District of Columbia (Table 1). A tract enters the typology whether or not it has an outcome. Life expectancy exists for 77,763 of the typed tracts, and each outcome below is compared over the typed tracts that have it.

Table 1. The typology covers 81,343 of the 85,396 census tracts

The 7 inputs are the log of median household income, the poverty rate, the share of adults with a college degree, the single-parent share of households, the homeownership rate, the share of residents aged 65 and over and the log of population density, from the ACS for 2019 to 2023 and 2020 tract land area. The first five describe resources and family structure. Age and density describe how a place is settled, and they are not outcomes. Life expectancy, the adult income rank of children raised in low-income families, teen births, incarceration, diabetes, smoking, frequent mental distress and obesity are all outside the inputs, and so are the Black and Hispanic shares of residents.

Each input is standardized to mean zero and standard deviation one across tracts, and each tract counts once. K-means then places eight centers, assigns every tract to the nearest and repeats until the assignments settle, from many random starts. Eight is the book's choice, kept from the first edition, and the data do not select it. Across fits with 2 to 10 types the mean silhouette is highest at 2 types, where it is 0.27 against 0.24 for simulated data with no clusters. At no number of types do the tracts score more than 0.08 above that benchmark.

Refits on bootstrap resamples of tracts agree with the typology at a median of 0.95 on a scale where one means identical partitions and zero means chance, and the least similar refit agrees at 0.55. A refit that weights each tract by its population agrees at 0.84. The partition is therefore one of several that the data support. Chapter 10.2, Families of places (https://vishalsingh.org/zipcode/families-of-places) sets out the method and its stability checks for a related clustering.

Types are numbered by the population-weighted mean of the first principal component of the inputs, a combined resource score that rises with income, college share and homeownership and falls with poverty and the single-parent share. The numbering runs from the highest score to the lowest and does not depend on the starting seed. The names are the book's labels for the profiles in Table 2, written into the build with a check that each profile still fits its name. A name describes what the inputs measure and names no race, ethnicity or outcome. The names label tracts and not the people who live in them. The chapter's label echoes the Eight Americas of Murray and colleagues, who divided the population by race and county to compare mortality.[1] These types group tracts by a profile of resources and settlement and share only the count of groups.

### The types differ on age and density as well as income

Table 2 profiles each type. High-income owners hold 48.5 million residents, with tract median household incomes averaging 157 (95% CI 153 to 160) thousand dollars, a college share of 63.4%, a homeownership rate of 83.8% and a poverty rate of 4.0%.

Middle-income owners are the most populous type, with 79.5 million residents. Their incomes average 98 thousand dollars, the second highest of the eight, with a college share of 35.3% and a homeownership rate of 78.8%.

Table 2. High-income owners have the highest incomes and the high-poverty, single-parent type the highest poverty and single-parent shares, and their mean life expectancy differs by 7.5 years

Two owner types are set apart by settlement more than by income. Older owners hold 11.3 million residents, of whom 42.3% are aged 65 and over, in tracts where the homeownership rate is 78.9% and incomes average 84 thousand dollars. Low-density owners hold 52.3 million residents at a typical density of 55 (95% CI 53 to 57) residents per square mile, the lowest of the eight, with a homeownership rate of 80.1% and incomes averaging 68 thousand dollars.

Dense, college-educated renters hold 29.8 million residents at a typical density of 8,616 (95% CI 6,403 to 11,073) residents per square mile, the highest of the eight. Their college share of 56.6% is second only to that of high-income owners, and 36.1% of their homes are owner-occupied.

The moderate-income, mixed-tenure type holds 63.3 million residents, with incomes averaging 68 thousand dollars and a poverty rate of 14.5%. Its homeownership rate of 57.7% lies between those of the owner types and the renter types.

The last two types have similar incomes and poverty rates and differ in family structure and education. Lower-income renters hold 22.7 million residents, with a poverty rate of 29.8%, a homeownership rate of 35.8%, a single-parent share of 10.3% and a college share of 21.7%. The high-poverty, single-parent type holds 17.3 million, with a poverty rate of 32.4%, a homeownership rate of 35.2%, a single-parent share of 24.9% and a college share of 13.7%.

The Black and Hispanic shares are not inputs, and they differ across the types. Both are higher in each of the last three types than in any of the first five. The Black share averages 35.2% (95% CI 30.9% to 39.6%) and the Hispanic share 38.3% (95% CI 32.6% to 44.0%) in the high-poverty, single-parent type, against 5.0% and 9.6% among high-income owners and 5.7% and 8.1% among low-density owners. The moderate-income, mixed-tenure type has a Hispanic share of 32.0% (95% CI 27.8% to 36.1%), above its Black share. These shares describe who lives in each type and played no part in forming it.

### Where the types are most common

Coloring each county by its most common type draws a map dominated by one type. Low-density owners are the most common type in 2,207 of the 3,110 counties that have a type, and middle-income owners in 438 (Table 3). Low-density owners lead in most counties of the South, the Midwest, the Plains and the Mountain West, and middle-income owners in most counties of Connecticut, Delaware, Hawaii and Rhode Island. The moderate-income, mixed-tenure type leads in 222 counties, most often in Texas and California. High-income owners lead in 103 counties and the high-poverty, single-parent type in 28, most of them in Texas, Georgia, Mississippi, South Dakota and Louisiana. The types other than the two most common lead mainly in counties with large populations.

Table 3. Low-density owners are the most common type in 2,207 of the 3,110 counties with a type, and middle-income owners in 438

A county's color is the type of its largest group of residents, and a county's residents are mixed. In the median county the modal type holds 68.1% of the typed residents. Across all counties with a type, 42.9% of typed residents live in a tract of their own county's modal type, which leaves more residents in a tract of another type.

Among the 173 counties with at least 100 typed tracts, a median of 3 types each hold at least 10% of the typed residents.

Counties are typed from their tracts. A county gets a type when its typed tracts hold at least 50% of its residents. 13 counties have no tract that is not institutional, has every input and has more than 200 residents. In another 20, typed tracts hold a smaller share. Both groups are gray on the map, and the tooltip gives the reason.

Residents are counted from the ACS. A tie in residents between two types goes to the lower-numbered type, and 1 county has one. Of the counties with a type, 485 rest on fewer than 3 typed tracts. Connecticut is drawn as its eight legacy counties, and Puerto Rico has no typed tract and is not drawn.

### The high-poverty, single-parent type is least favorable on all eight outcomes

None of the eight outcomes is an input, so their order across the types is not built in (Table 4). The first tab holds life expectancy, the adult income rank of children raised in low-income families, and teen births and incarceration among the same Opportunity Atlas cohorts. The second holds diabetes, smoking, frequent mental distress and obesity among adults. The high-poverty, single-parent type has the least favorable mean on all eight. High-income owners have the most favorable mean on every outcome except diabetes, which is lowest among dense, college-educated renters (8.7% (95% CI 8.4% to 9.0%) of adults, against 9.4% (95% CI 9.2% to 9.5%)).

Table 4. The high-poverty, single-parent type has the least favorable mean on all eight outcomes, and high-income owners the most favorable on every outcome except diabetes

The share of women raised in low-income families who had a teen birth is 13.9% (95% CI 13.1% to 14.7%) among high-income owners and 38.9% (95% CI 37.8% to 40.0%) in the high-poverty, single-parent type. The share of children from those families who were incarcerated on Census Day 2010 is 1.3% (95% CI 1.2% to 1.4%) in the first type and 3.5% (95% CI 3.3% to 3.8%) in the last. The Opportunity Atlas measures share no file with any input. PLACES estimates are modeled partly from ACS covariates such as poverty and education, so the agreement between the types and the adult health measures is partly mechanical.

Between the two ends the types do not line up in one order. Dense, college-educated renters, numbered fifth, have a life expectancy of 80.3 (95% CI 79.9 to 80.6) years, which is 2.5 (95% CI 2.1 to 2.8) years above that of low-density owners, numbered fourth, although the two types sit at about the same place on the resource score. They also live longer than middle-income owners, at 79.6 (95% CI 79.5 to 79.7) years, and about as long as older owners, at 80.4 (95% CI 80.1 to 80.6) years, whose interval overlaps theirs. Life expectancy averages 77.8 (95% CI 77.7 to 77.9) years among low-density owners, 77.3 (95% CI 77.1 to 77.6) in the moderate-income, mixed-tenure type and 75.3 (95% CI 74.9 to 75.8) among lower-income renters. The intervals are narrow next to most gaps between types, since each type holds thousands of tracts. They describe how the means would vary with the sample of counties given the types, not with the choice of types.

The label foretells part of a tract's outcome and leaves the rest. The types account for 32% (95% CI 30% to 34%) of the population-weighted variance of tract life expectancy, 32% (95% CI 30% to 34%) of that of the teen-birth share and 8% (95% CI 7% to 9%) of that of the incarceration share. They account for 58% (95% CI 56% to 60%) of the variance of smoking and 59% (95% CI 57% to 61%) of that of frequent mental distress, two modeled PLACES measures whose agreement with the types is partly mechanical. Tracts of the same type still differ, and they differ most on incarceration.

### The earlier types spread life expectancy wider, with outcomes among their inputs

The chapter's earlier typology used 13 inputs. It added the Black and Hispanic shares, life expectancy, the adult income rank, diabetes and smoking to the inputs here. The comparison refits it with the same k-means, seed and starts on the 77,745 typed tracts that have all of its inputs, which are the tracts it typed. It agrees with the types of this chapter at 0.41 on the agreement scale of the bootstrap refits, so the two partitions overlap in part.

On those tracts the earlier types span 10.0 (95% CI 9.8 to 10.2) years of life expectancy from the highest to the lowest type mean, and the types here span 7.5 (95% CI 7.1 to 7.9) years. The earlier types account for 49% (95% CI 47% to 51%) of the variance of tract life expectancy, against 32% (95% CI 30% to 34%) for the types here, and for 46% (95% CI 44% to 48%) of the variance of the teen-birth share, against 33% (95% CI 30% to 35%). Part of the earlier spread came from sorting tracts on life expectancy itself and on measures that move with it. The types here sort life expectancy and teen births with neither among their inputs, less sharply than the earlier fit did.

### The types are not sharply separated

A silhouette of 0.19 is low, and two comparisons give it a scale. The first is simulated data with the correlations of the inputs and no clusters, in which k-means still finds eight groups. Their silhouette is 0.14, and 56.3% of the variance lies between the groups.

The tract data score 0.19 and 62.8%, with 37.2% of the variance within the types, a step above what a cloud without clusters gives. The benchmark is a guide and not a proof, because the simulated margins are symmetric and tract incomes, densities and rates are not.

The second comparison is the tracts themselves. In a sample of 20,000 tracts, 6.6% have a negative silhouette, which means that on average they sit closer to the tracts of another type than to those of their own.

Table 5 shows the overlap on the first two principal components, which capture 68.4% of the variance in the inputs. The first, 43.9%, weights income, college share, homeownership and the share of older adults positively and poverty and the single-parent share negatively, and it is the resource score that numbers the types. The second, 24.5%, weights density and college share positively and homeownership and the share of older adults negatively, and it separates dense renters from older and low-density owners. The plotted sample holds 6,000 tracts, and the ranges from the fifth to the ninety-fifth percentile of tracts overlap for every pair of neighboring types along the resource score. The overlap suggests that the types are bands of a continuum, and a tract near a border could have been assigned to the neighbor by a small change in the inputs. The resource score of this figure is a different object from the nine-measure analysis of Chapter 10.1, Gradients, slopes and variance (https://vishalsingh.org/zipcode/gradients-slopes-and-variance), which uses other measures.

Table 5. Neighboring types overlap along the first component, and 6.6% of 20,000 sampled tracts sit nearer another type than their own

> Causal evidence: Whether growing up in a different place changes a child's adult income (Movers design)
>
> The comparisons above describe tracts and cannot show that a tract's type changes anyone's life. Chetty and Hendren used federal tax records for children born in 1980 to 1988 whose families moved once between commuting zones, and compared children who moved at different ages, including siblings in the same family. The identifying assumption is that the selection that sends a family to a better or a worse place does not vary with the child's age at the move. They test it with sibling comparisons, with controls for changes in parental income and marital status, and with moves caused by displacement shocks. Children's adult income ranks move toward the ranks of children of permanent residents of the destination who had parents at the same income, by 0.04 percentile for each year of childhood spent there and each percentile by which the destination's ranks are higher.[2]
>
> The design applies to commuting zones and counties, and its benchmark is permanent residents of the destination at the same parent income. It does not test the eight types, and it does not say that moving a child from one type to another changes an outcome by the gap between the types' means.

### What these data cannot show

The types describe tracts, and every comparison is ecological: the life expectancy of the high-poverty, single-parent type is an average over tracts and says little about any one resident of one (ecological inference). The names are the book's labels for the profiles. A name such as high-poverty, single-parent summarizes two inputs of a profile that also has a Black share of 35.2% and a Hispanic share of 38.3%, and it does not describe the people who live there.

The types are built without outcomes, and the comparisons still do not show that a tract's resources or settlement produced any outcome. They show how closely conditions and outcomes move together across tracts. The measures come from different periods, with the ACS inputs for 2019 to 2023, life expectancy for deaths in 2010 to 2015, mobility for children born in 1978 to 1983 and PLACES built on 2022 survey data, so an outcome can predate the conditions it is compared with. The USALEEP estimates and the Opportunity Atlas ranks for 2020 tracts are interpolated from 2010 tracts by land area, not by population. The USALEEP estimates suit maps and gradients, the ACS file carries no margins of error, and PLACES values are modeled from covariates that include ACS measures and carry no intervals at the tract level.

The intervals in this chapter condition on the types. Tracts that lack an input are not typed. The 3,072 untyped tracts hold 7.7 million residents, and 39% of them lack a life expectancy estimate. Typed tracts without a life expectancy stay in the types and drop out of the life-expectancy comparisons.

The untyped tracts differ from the typed tracts. Their poverty rate averages 25.8% against 12.3%, and their Black share 19.4% against 11.9%. The residents of typed tracts, 97.7% of all residents, are therefore not a random sample of the country. The county map shows one type per county and hides the mix, and Puerto Rico is outside the typology and the map.

## Tables

### Table 1. The typology covers 81,343 of the 85,396 census tracts

Census tracts at each step from the 2020 census to the tracts with every input, counted at the first step that removes them

| Step | Tracts |
| --- | ---: |
| Tracts of the 2020 census | 85,396 |
| Puerto Rico (outside the county map and the outcome measures) | 981 |
| Other tracts missing at least one input or with no residents | 1,498 |
| Institutional tracts (25 percent or more of residents in group quarters) | 1,541 |
| Tracts with 200 or fewer residents | 33 |
| Tracts in the typology | 81,343 |

Note: Puerto Rico is left out because the county map and the outcome measures cover the 50 states and DC. A tract is institutional when at least 25% of its residents live in group quarters such as prisons, nursing homes or dormitories (SVI 2022), or when it is a known chronic-care tract. Tracts with 200 or fewer residents are left out.

Note: A typed tract need not have an outcome. Life expectancy exists for 77,763 of the typed tracts, and each outcome is compared over the typed tracts that have it.

Source: 2020 census tracts with ACS 2019-2023 inputs, 2020 TIGER land area and SVI 2022 group-quarters shares; counts of tracts, full count

Data: https://vishalsingh.org/zipcode/report/eight-types-of-neighborhood/figures/sample_flow.json

### Table 2. High-income owners have the highest incomes and the high-poverty, single-parent type the highest poverty and single-parent shares, and their mean life expectancy differs by 7.5 years

Population-weighted mean of tract values by neighborhood type, types numbered by the first principal component of the inputs, largest first

| Type | Residents (millions) | Tracts | Household income (thousands of dollars) | Poverty (%) | College degree (%) | Single-parent households (%) | Homeowners (%) | Aged 65 and over (%) | Density (residents per square mile) | Black (%) | Hispanic (%) | ADI national rank | Life expectancy (years) |
| --- | ---: | ---: | ---: | ---: | ---: | ---: | ---: | ---: | ---: | ---: | ---: | ---: | ---: |
| High-income owners | 48.5 | 10,661 | 157 | 4.0 | 63.4 | 4.6 | 83.8 | 17.3 | 2,017.3 | 5.0 | 9.6 | 17.3 | 82.1 |
| Older owners | 11.3 | 3,560 | 84 | 8.4 | 41.8 | 3.2 | 78.9 | 42.3 | 955.7 | 4.3 | 9.1 | 40.3 | 80.4 |
| Middle-income owners | 79.5 | 17,400 | 98 | 6.9 | 35.3 | 6.7 | 78.8 | 17.1 | 1,582.6 | 8.7 | 16.5 | 40.5 | 79.6 |
| Low-density owners | 52.3 | 15,491 | 68 | 13.0 | 20.4 | 7.0 | 80.1 | 20.8 | 55.33 | 5.7 | 8.1 | 68.2 | 77.8 |
| Dense, college-educated renters | 29.8 | 7,720 | 96 | 11.1 | 56.6 | 5.7 | 36.1 | 12.8 | 8,616.3 | 11.2 | 18.7 | 26.3 | 80.3 |
| Moderate-income, mixed-tenure | 63.3 | 14,865 | 68 | 14.5 | 22.3 | 13.4 | 57.7 | 13.7 | 3,362.8 | 16.9 | 32.0 | 56.9 | 77.3 |
| Lower-income renters | 22.7 | 6,759 | 43 | 29.8 | 21.7 | 10.3 | 35.8 | 14.7 | 4,376.0 | 24.8 | 28.5 | 67.4 | 75.3 |
| High-poverty, single-parent | 17.3 | 4,887 | 43 | 32.4 | 13.7 | 24.9 | 35.2 | 10.5 | 5,103.7 | 35.2 | 38.3 | 70.5 | 74.7 |

Note: The type names are the book's labels for groups of tracts that resemble one another on the inputs, not categories in the data. They describe income, poverty, education, family structure, tenure, age and density. No outcome and no measure of race or ethnicity is an input; the Black and Hispanic shares are shown to compare the composition of the types.

Note: Intervals for these cells and for the outcomes are bootstrap intervals over counties; the cells describe the tracts in each type.

Note: The ACS columns cover 2019 to 2023, and density is the weighted geometric mean of tract residents per square mile of 2020 land area, so a few very dense tracts do not set the value. Life expectancy is for deaths in 2010 to 2015 on tracts of 2020 (area-weighted), over the 77,763 typed tracts that have it. The ADI is the 2023 national rank (1 least and 100 most deprived).

Note: Black residents are Black alone and may be Hispanic; Hispanic residents are of any race.

Source: ACS 2019-2023, USALEEP life expectancy 2010-2015, Opportunity Atlas (children born 1978 to 1983), PLACES 2024 release and ADI 2023; tract values weighted by ACS population; population-weighted

Data: https://vishalsingh.org/zipcode/report/eight-types-of-neighborhood/figures/type_profiles.json

### Table 3. Low-density owners are the most common type in 2,207 of the 3,110 counties with a type, and middle-income owners in 438

Neighborhood type that holds the most residents in each county, tracts typed by k-means on 7 measures, ACS 2019-2023 population

No inline rows: the values of Table 3 are in https://vishalsingh.org/zipcode/data/eight-types-of-neighborhood/modal_type_map.json.

Note: A county gets a type when its typed tracts hold at least 50% of its residents. 13 counties have no typed tract (a tract that is not institutional, has every input and has more than the floor of residents) and 20 have typed tracts that hold a smaller share, so they are gray with the reason in the tooltip. A county with a tie in residents takes the lower-numbered type, and 1 county has a tie. 485 counties with a type rest on fewer than 3 typed tracts.

Note: Connecticut is drawn as its eight legacy counties, with tracts assigned through the exact tract crosswalk. Puerto Rico is outside the typology and is not drawn.

Note: Colors mark the eight types and have no order.

Source: k-means types of tracts on seven ACS 2019-2023 measures and 2020 land area, summed to counties by ACS population; population-weighted

Data: https://vishalsingh.org/zipcode/report/eight-types-of-neighborhood/figures/modal_type_map.json

### Table 4. The high-poverty, single-parent type has the least favorable mean on all eight outcomes, and high-income owners the most favorable on every outcome except diabetes

Population-weighted mean of tract values by neighborhood type with 95% intervals from a bootstrap over counties, each panel in its own units

#### Table 4.1. Life expectancy, mobility and justice

Life expectancy and adult income rank are highest and teen births and incarceration lowest among high-income owners, and the reverse holds in the high-poverty, single-parent type

Life expectancy, the adult income rank, teen births and incarceration, population-weighted type means with 95% intervals

| conf_high | conf_low | estimate | label | n | panel | type |
| ---: | ---: | ---: | --- | ---: | --- | --- |
| 82.30 | 81.93 | 82.13 | High-income owners | 10,184 | Life expectancy at birth (years) | High-income owners |
| 80.64 | 80.10 | 80.36 | Older owners | 3,221 | Life expectancy at birth (years) | Older owners |
| 79.73 | 79.47 | 79.60 | Middle-income owners | 16,921 | Life expectancy at birth (years) | Middle-income owners |
| 77.92 | 77.71 | 77.81 | Low-density owners | 15,201 | Life expectancy at birth (years) | Low-density owners |
| 80.62 | 79.92 | 80.31 | Dense, college-educated renters | 6,980 | Life expectancy at birth (years) | Dense, college-educated renters |
| 77.58 | 77.08 | 77.33 | Moderate-income, mixed-tenure | 14,393 | Life expectancy at birth (years) | Moderate-income, mixed-tenure |
| 75.82 | 74.88 | 75.34 | Lower-income renters | 6,268 | Life expectancy at birth (years) | Lower-income renters |
| 75.12 | 74.22 | 74.67 | High-poverty, single-parent | 4,595 | Life expectancy at birth (years) | High-poverty, single-parent |
| 50.12 | 49.23 | 49.68 | High-income owners | 10,651 | Adult income rank of children from low-income families (percentile) | High-income owners |
| 45.44 | 44.44 | 44.94 | Older owners | 3,543 | Adult income rank of children from low-income families (percentile) | Older owners |
| 45.00 | 44.42 | 44.71 | Middle-income owners | 17,392 | Adult income rank of children from low-income families (percentile) | Middle-income owners |
| 42.77 | 42.37 | 42.57 | Low-density owners | 15,490 | Adult income rank of children from low-income families (percentile) | Low-density owners |
| 45.08 | 43.73 | 44.44 | Dense, college-educated renters | 7,706 | Adult income rank of children from low-income families (percentile) | Dense, college-educated renters |
| 40.91 | 40.34 | 40.64 | Moderate-income, mixed-tenure | 14,860 | Adult income rank of children from low-income families (percentile) | Moderate-income, mixed-tenure |
| 38.43 | 37.36 | 37.89 | Lower-income renters | 6,757 | Adult income rank of children from low-income families (percentile) | Lower-income renters |
| 36.42 | 35.18 | 35.81 | High-poverty, single-parent | 4,884 | Adult income rank of children from low-income families (percentile) | High-poverty, single-parent |
| 14.67 | 13.10 | 13.86 | High-income owners | 10,629 | Teen births among daughters of low-income parents (%) | High-income owners |
| 19.97 | 18.31 | 19.12 | Older owners | 3,504 | Teen births among daughters of low-income parents (%) | Older owners |
| 22.93 | 21.84 | 22.38 | Middle-income owners | 17,382 | Teen births among daughters of low-income parents (%) | Middle-income owners |
| 27.88 | 27.21 | 27.54 | Low-density owners | 15,474 | Teen births among daughters of low-income parents (%) | Low-density owners |
| 22.10 | 19.14 | 20.51 | Dense, college-educated renters | 7,685 | Teen births among daughters of low-income parents (%) | Dense, college-educated renters |
| 30.43 | 29.15 | 29.77 | Moderate-income, mixed-tenure | 14,852 | Teen births among daughters of low-income parents (%) | Moderate-income, mixed-tenure |
| 35.04 | 32.46 | 33.76 | Lower-income renters | 6,745 | Teen births among daughters of low-income parents (%) | Lower-income renters |
| 39.98 | 37.77 | 38.86 | High-poverty, single-parent | 4,883 | Teen births among daughters of low-income parents (%) | High-poverty, single-parent |
| 1.35 | 1.18 | 1.26 | High-income owners | 10,648 | Incarcerated on Census Day 2010, children of low-income parents (%) | High-income owners |
| 1.86 | 1.63 | 1.75 | Older owners | 3,533 | Incarcerated on Census Day 2010, children of low-income parents (%) | Older owners |
| 1.90 | 1.76 | 1.83 | Middle-income owners | 17,390 | Incarcerated on Census Day 2010, children of low-income parents (%) | Middle-income owners |
| 1.98 | 1.89 | 1.94 | Low-density owners | 15,488 | Incarcerated on Census Day 2010, children of low-income parents (%) | Low-density owners |
| 2.14 | 1.69 | 1.90 | Dense, college-educated renters | 7,695 | Incarcerated on Census Day 2010, children of low-income parents (%) | Dense, college-educated renters |
| 2.59 | 2.36 | 2.48 | Moderate-income, mixed-tenure | 14,858 | Incarcerated on Census Day 2010, children of low-income parents (%) | Moderate-income, mixed-tenure |
| 3.25 | 2.83 | 3.04 | Lower-income renters | 6,754 | Incarcerated on Census Day 2010, children of low-income parents (%) | Lower-income renters |
| 3.81 | 3.28 | 3.54 | High-poverty, single-parent | 4,884 | Incarcerated on Census Day 2010, children of low-income parents (%) | High-poverty, single-parent |

#### Table 4.2. Adult health

Diabetes, smoking, mental distress and obesity are highest in the high-poverty, single-parent type and lowest among high-income owners, except diabetes, which is lowest among dense, college-educated renters

Modeled PLACES prevalence among adults, population-weighted type means with 95% intervals

| conf_high | conf_low | estimate | label | n | panel | type |
| ---: | ---: | ---: | --- | ---: | --- | --- |
| 9.51 | 9.18 | 9.35 | High-income owners | 10,660 | Diabetes among adults (%) | High-income owners |
| 13.92 | 13.35 | 13.64 | Older owners | 3,560 | Diabetes among adults (%) | Older owners |
| 11.24 | 10.95 | 11.09 | Middle-income owners | 17,400 | Diabetes among adults (%) | Middle-income owners |
| 13.90 | 13.72 | 13.81 | Low-density owners | 15,491 | Diabetes among adults (%) | Low-density owners |
| 9.02 | 8.44 | 8.74 | Dense, college-educated renters | 7,719 | Diabetes among adults (%) | Dense, college-educated renters |
| 13.32 | 12.90 | 13.11 | Moderate-income, mixed-tenure | 14,865 | Diabetes among adults (%) | Moderate-income, mixed-tenure |
| 15.69 | 14.89 | 15.30 | Lower-income renters | 6,759 | Diabetes among adults (%) | Lower-income renters |
| 17.26 | 16.63 | 16.94 | High-poverty, single-parent | 4,887 | Diabetes among adults (%) | High-poverty, single-parent |
| 9.24 | 8.86 | 9.03 | High-income owners | 10,660 | Smoking among adults (%) | High-income owners |
| 11.57 | 10.98 | 11.28 | Older owners | 3,560 | Smoking among adults (%) | Older owners |
| 13.33 | 12.94 | 13.13 | Middle-income owners | 17,400 | Smoking among adults (%) | Middle-income owners |
| 18.24 | 18.02 | 18.13 | Low-density owners | 15,491 | Smoking among adults (%) | Low-density owners |
| 10.82 | 10.27 | 10.53 | Dense, college-educated renters | 7,719 | Smoking among adults (%) | Dense, college-educated renters |
| 16.68 | 16.04 | 16.36 | Moderate-income, mixed-tenure | 14,865 | Smoking among adults (%) | Moderate-income, mixed-tenure |
| 20.13 | 19.11 | 19.60 | Lower-income renters | 6,759 | Smoking among adults (%) | Lower-income renters |
| 22.65 | 21.39 | 22.00 | High-poverty, single-parent | 4,887 | Smoking among adults (%) | High-poverty, single-parent |
| 13.27 | 12.99 | 13.12 | High-income owners | 10,660 | Frequent mental distress among adults (%) | High-income owners |
| 13.39 | 12.95 | 13.18 | Older owners | 3,560 | Frequent mental distress among adults (%) | Older owners |
| 16.05 | 15.86 | 15.95 | Middle-income owners | 17,400 | Frequent mental distress among adults (%) | Middle-income owners |
| 17.70 | 17.55 | 17.62 | Low-density owners | 15,491 | Frequent mental distress among adults (%) | Low-density owners |
| 16.03 | 15.53 | 15.75 | Dense, college-educated renters | 7,719 | Frequent mental distress among adults (%) | Dense, college-educated renters |
| 18.73 | 18.46 | 18.60 | Moderate-income, mixed-tenure | 14,865 | Frequent mental distress among adults (%) | Moderate-income, mixed-tenure |
| 21.29 | 20.76 | 21.02 | Lower-income renters | 6,759 | Frequent mental distress among adults (%) | Lower-income renters |
| 22.36 | 21.91 | 22.14 | High-poverty, single-parent | 4,887 | Frequent mental distress among adults (%) | High-poverty, single-parent |
| 27.89 | 26.48 | 27.12 | High-income owners | 10,660 | Obesity among adults (%) | High-income owners |
| 30.42 | 29.27 | 29.83 | Older owners | 3,560 | Obesity among adults (%) | Older owners |
| 33.31 | 32.34 | 32.84 | Middle-income owners | 17,400 | Obesity among adults (%) | Middle-income owners |
| 38.11 | 37.77 | 37.95 | Low-density owners | 15,491 | Obesity among adults (%) | Low-density owners |
| 28.45 | 26.39 | 27.30 | Dense, college-educated renters | 7,719 | Obesity among adults (%) | Dense, college-educated renters |
| 37.30 | 35.91 | 36.63 | Moderate-income, mixed-tenure | 14,865 | Obesity among adults (%) | Moderate-income, mixed-tenure |
| 40.30 | 38.19 | 39.26 | Lower-income renters | 6,759 | Obesity among adults (%) | Lower-income renters |
| 43.84 | 41.87 | 42.89 | High-poverty, single-parent | 4,887 | Obesity among adults (%) | High-poverty, single-parent |

Note: No outcome is an input to the typology. PLACES estimates are modeled partly from ACS covariates such as poverty and education, so the agreement between the types and the adult health measures is partly mechanical. The Opportunity Atlas measures share no file with any input.

Note: Opportunity Atlas measures are for children born 1978 to 1983 raised in families at the 25th income percentile, interpolated from 2010 to 2020 tracts by land area, and tract population stands in for the number of children. Life expectancy is for deaths in 2010 to 2015 and covers 77,763 of the typed tracts. PLACES values are modeled crude prevalence for adults, with BRFSS 2022 data. Each panel uses the typed tracts that have the measure.

Note: Intervals show how the means would vary with the sample of counties, given the types; they do not include uncertainty about the typology itself (a bootstrap over counties; replicates and seed in the methods note).

Source: ACS 2019-2023, USALEEP life expectancy 2010-2015, Opportunity Atlas (children born 1978 to 1983), PLACES 2024 release and ADI 2023; tract values weighted by ACS population; population-weighted, modeled estimates for the PLACES measures

Data: https://vishalsingh.org/zipcode/report/eight-types-of-neighborhood/figures/type_outcomes.json

### Table 5. Neighboring types overlap along the first component, and 6.6% of 20,000 sampled tracts sit nearer another type than their own

A random sample of tracts on the first two principal components of the 7 standardized inputs, colored by type, with each type's centroid labeled

No inline rows: the values of Table 5 are in https://vishalsingh.org/zipcode/data/eight-types-of-neighborhood/pca_continuum.json.

Note: The dots are a random sample of 6,000 of the typed tracts drawn with a fixed seed (replicates and seed in the methods note); the components, the types and the centroids use every typed tract. Components are in standardized units, and their signs are set so that log income and log density rise with the first and second component.

Note: The first component weights income, college share, homeownership and the share of older adults positively and poverty and the single-parent share negatively. The second component weights density and college share positively and homeownership and the share of older adults negatively.

Source: k-means types of tracts on seven ACS 2019-2023 measures and 2020 land area; principal components of the standardized inputs; unweighted sample of tracts, centroids population-weighted

Data: https://vishalsingh.org/zipcode/report/eight-types-of-neighborhood/figures/pca_continuum.json

## All estimates

Every fact in facts.json (131), sorted by id.

| Fact id | Value | Unit | Estimand | Population | Period | Interval | n | Weights | Source | Note |
| --- | ---: | --- | --- | --- | --- | --- | ---: | --- | --- | --- |
| age_old | 65 | years of age | definition: the age at and above which a resident counts as an older adult in the share-aged-65-and-over input | residents | 2019-2023 |   |   | none (definition) | American Community Survey |   |
| ari_boot_median | 0.95 | adjusted Rand index (1 = identical partitions, 0 = chance agreement) | agreement of the typology with the median of the refits on bootstrap resamples of tracts | census tracts in the typology | ACS 2019-2023, 2020 land area |   | 81,343 tracts | none (each tract counts once) | American Community Survey |   |
| ari_boot_min | 0.55 | adjusted Rand index (1 = identical partitions, 0 = chance agreement) | agreement of the typology with the lowest of the refits on bootstrap resamples of tracts | census tracts in the typology | ACS 2019-2023, 2020 land area |   | 81,343 tracts | none (each tract counts once) | American Community Survey |   |
| ari_old | 0.41 | adjusted Rand index (1 = identical partitions, 0 = chance agreement) | agreement of the typology with the chapter's earlier k-means fit on thirteen inputs, published until October 28, 2026, (the seven inputs of the typology plus the Black and Hispanic shares, life expectancy, the adult income rank, diabetes and smoking; same seed, groups and starts) | census tracts in the typology that have all thirteen inputs of the earlier typology | tract inputs from 2010 to 2023 |   | 77,745 tracts | none (each tract counts once) | American Community Survey |   |
| ari_weighted | 0.84 | adjusted Rand index (1 = identical partitions, 0 = chance agreement) | agreement of the typology with a refit that weights each tract by its population | census tracts in the typology | ACS 2019-2023, 2020 land area |   | 81,343 tracts | none (each tract counts once) | American Community Survey |   |
| big_min_tracts | 100 | tracts | definition: the least number of typed tracts for the types-per-county count | counties | 2019-2023 |   |   | none (definition) | American Community Survey |   |
| big_types_median | 3 | types | median across counties of the number of types that hold at least a tenth of the county's typed residents | counties with at least 100 typed tracts | 2019-2023 |   | 173 counties | none (each county counts once) | American Community Survey |   |
| ch_exposure_yr | 0.04 | percentiles of adult income rank per year of childhood exposure, per percentile of difference in permanent residents' outcomes | exposure effect reported by Chetty and Hendren (2018, Quarterly Journal of Economics 133(3)): federal tax records, children born 1980 to 1988, baseline sample of one-time movers between commuting zones; movers design that compares children who moved at different ages, with sibling comparisons (family fixed effects) and moves caused by displacement shocks; the benchmark is the adult income rank of children of permanent residents of the destination at the same parent income; copied from the paper and not computed in this build | children in the study's baseline sample whose families moved once between commuting zones | children born 1980 to 1988, moves from 1997 to 2010 |   |   | none (reported estimate) | Published estimates from cited studies | A published estimate, not a result of this build. |
| cover_min | 50% | percent of a county's residents | definition: a county gets a modal type when its typed tracts hold at least this share of its residents | counties | 2019-2023 |   |   | none (definition) | American Community Survey |   |
| few_tracts | 3 | tracts | definition: a county with fewer typed tracts than this is flagged as resting on very few tracts | counties | 2019-2023 |   |   | none (definition) | American Community Survey |   |
| inst_gq_pct | 25% | percent of a tract's residents | definition: a tract is institutional when at least this share of its residents live in group quarters (SVI 2022 ep_groupq) | census tracts | SVI 2022 (ACS 2018-2022) |   |   | none (definition) | CDC/ATSDR Social Vulnerability Index |   |
| k_best | 2 | types | number of types with the highest mean silhouette in the comparison | k-means fits compared for the silhouette | ACS 2019-2023, 2020 land area |   |   | none (definition) | American Community Survey |   |
| k_hi | 10 | types | definition: the most types in the silhouette comparison | k-means fits compared for the silhouette | ACS 2019-2023, 2020 land area |   |   | none (definition) | American Community Survey |   |
| k_lo | 2 | types | definition: the fewest types in the silhouette comparison | k-means fits compared for the silhouette | ACS 2019-2023, 2020 land area |   |   | none (definition) | American Community Survey |   |
| le_gap | 7.5 | years of life expectancy at birth | difference between the population-weighted mean life expectancy of type 1 minus type 8 (High-income owners minus High-poverty, single-parent), bootstrap interval over counties with shared draws | residents of the tracts of types 1 and 8 that have a life expectancy | deaths in 2010 to 2015 | 95% CI 7.1 to 7.9 | 14,779 tracts | ACS 2019-2023 tract population | USALEEP life expectancy |   |
| le_spread_c | 7.5 | years of life expectancy at birth | highest minus lowest population-weighted mean life expectancy across the groups of the eight types of this chapter, on the tracts of the comparison, bootstrap interval over counties | residents of the typed tracts that have all thirteen inputs of the earlier typology | deaths in 2010 to 2015 | 95% CI 7.1 to 7.9 | 77,745 tracts | ACS 2019-2023 tract population | USALEEP life expectancy |   |
| le_spread_old | 10.0 | years of life expectancy at birth | highest minus lowest population-weighted mean life expectancy across the groups of the chapter's earlier k-means fit on thirteen inputs, published until October 28, 2026, (the seven inputs of the typology plus the Black and Hispanic shares, life expectancy, the adult income rank, diabetes and smoking; same seed, groups and starts), on the tracts of the comparison, bootstrap interval over counties | residents of the typed tracts that have all thirteen inputs of the earlier typology | deaths in 2010 to 2015 | 95% CI 9.8 to 10.2 | 77,745 tracts | ACS 2019-2023 tract population | USALEEP life expectancy |   |
| med_modal_share | 68.1% | percent of a county's typed residents | median across counties of the share of typed residents who live in tracts of the county's modal type | counties with a modal type | 2019-2023 |   | 3,110 counties | none (each county counts once) | American Community Survey |   |
| n_big | 173 | counties | count of counties with at least 100 typed tracts (the cutoff is part of the definition) | counties with at least 100 typed tracts | 2019-2023 |   |   | none (full count) | American Community Survey |   |
| n_compare | 77,745 | tracts | count of typed tracts that have all thirteen inputs of the earlier typology (the tracts of the comparison) | census tracts in the typology | tract inputs from 2010 to 2023 |   |   | none (full count) | American Community Survey |   |
| n_counties_few | 485 | counties | count of counties with a modal type that rests on fewer than three typed tracts | counties and county equivalents on the county map (3,143, Connecticut as eight legacy counties) | 2019-2023 |   |   | none (full count) | American Community Survey |   |
| n_counties_low_cover | 20 | counties | count of counties whose typed tracts hold under half of the county's residents | counties and county equivalents on the county map (3,143, Connecticut as eight legacy counties) | 2019-2023 |   |   | none (full count) | American Community Survey |   |
| n_counties_no_tract | 13 | counties | count of counties with no tract that is not institutional, has every input and has more than the floor of residents | counties and county equivalents on the county map (3,143, Connecticut as eight legacy counties) | 2019-2023 |   |   | none (full count) | American Community Survey |   |
| n_counties_typed | 3,110 | counties | count of counties with a modal type | counties and county equivalents on the county map (3,143, Connecticut as eight legacy counties) | 2019-2023 |   |   | none (full count) | American Community Survey |   |
| n_county_ties | 1 | counties | count of counties where two types hold exactly the same number of typed residents (the county takes the lower-numbered type) | counties and county equivalents on the county map (3,143, Connecticut as eight legacy counties) | 2019-2023 |   |   | none (full count) | American Community Survey |   |
| n_features | 7 | inputs | count of inputs of the typology (definition) | measures standardized and given to k-means | ACS 2019-2023, 2020 land area |   |   | none (definition) | American Community Survey |   |
| n_features_old | 13 | inputs | count of inputs of the earlier typology (definition) | measures standardized and given to k-means in the earlier typology | 2010 to 2023 |   |   | none (definition) | American Community Survey |   |
| n_le | 77,763 | tracts | count of typed tracts with a USALEEP life expectancy estimate (the tracts over which life expectancy is compared) | census tracts in the typology | deaths in 2010 to 2015 |   |   | none (full count) | USALEEP life expectancy |   |
| n_modal_t1 | 103 | counties | count of counties where type 1 (High-income owners) holds the most residents | counties and county equivalents on the county map (3,143, Connecticut as eight legacy counties) | 2019-2023 |   |   | none (full count) | American Community Survey |   |
| n_modal_t3 | 438 | counties | count of counties where type 3 (Middle-income owners) holds the most residents | counties and county equivalents on the county map (3,143, Connecticut as eight legacy counties) | 2019-2023 |   |   | none (full count) | American Community Survey |   |
| n_modal_t4 | 2,207 | counties | count of counties where type 4 (Low-density owners) holds the most residents | counties and county equivalents on the county map (3,143, Connecticut as eight legacy counties) | 2019-2023 |   |   | none (full count) | American Community Survey |   |
| n_modal_t6 | 222 | counties | count of counties where type 6 (Moderate-income, mixed-tenure) holds the most residents | counties and county equivalents on the county map (3,143, Connecticut as eight legacy counties) | 2019-2023 |   |   | none (full count) | American Community Survey |   |
| n_modal_t8 | 28 | counties | count of counties where type 8 (High-poverty, single-parent) holds the most residents | counties and county equivalents on the county map (3,143, Connecticut as eight legacy counties) | 2019-2023 |   |   | none (full count) | American Community Survey |   |
| n_sample | 6,000 | tracts | size of a random sample drawn with a fixed seed for the scatter plot | census tracts in the typology | ACS 2019-2023, 2020 land area |   | 81,343 tracts | none (sample size) | American Community Survey |   |
| n_tracts | 81,343 | tracts | count of tracts in the typology | census tracts with every input, not institutional, with more residents than the floor | tract boundaries of 2020; inputs from ACS 2019-2023 |   |   | none (full count) | American Community Survey |   |
| n_tracts_all | 85,396 | tracts | count of 2020 census tracts in the atlas | census tracts of the 2020 census (50 states, DC and Puerto Rico) | tract boundaries of 2020; inputs from ACS 2019-2023 |   |   | none (full count) | American Community Survey |   |
| n_tracts_inst | 1,541 | tracts | count of tracts flagged institutional (SVI 2022 group-quarters share of 25 percent or more, plus one known tract) | census tracts of the 50 states and DC with every input | tract boundaries of 2020; inputs from ACS 2019-2023 |   |   | none (full count) | American Community Survey |   |
| n_tracts_missing | 1,498 | tracts | count of tracts missing at least one of the seven inputs or with no residents | census tracts of the 50 states and DC | tract boundaries of 2020; inputs from ACS 2019-2023 |   |   | none (full count) | American Community Survey |   |
| n_tracts_pr | 981 | tracts | count of Puerto Rico tracts, left out because the county map and the outcome measures cover the 50 states and DC | census tracts of Puerto Rico | tract boundaries of 2020; inputs from ACS 2019-2023 |   |   | none (full count) | American Community Survey |   |
| n_tracts_small | 33 | tracts | count of tracts with the population floor or fewer residents | census tracts of the 50 states and DC with every input, not institutional | tract boundaries of 2020; inputs from ACS 2019-2023 |   |   | none (full count) | American Community Survey |   |
| n_untyped | 3,072 | tracts | count of tracts missing an input, institutional or with the floor of residents or fewer | census tracts of the 50 states and DC that are not in the typology | 2019-2023 |   |   | none (full count) | American Community Survey |   |
| pc12_evr | 68.4% | percent of the total variance of the standardized inputs | share of variance captured by the first two principal components | census tracts in the typology | ACS 2019-2023, 2020 land area |   | 81,343 tracts | none (each tract counts once) | American Community Survey |   |
| pc1_evr | 43.9% | percent of the total variance of the standardized inputs | share of variance captured by the first principal component | census tracts in the typology | ACS 2019-2023, 2020 land area |   | 81,343 tracts | none (each tract counts once) | American Community Survey |   |
| pc2_evr | 24.5% | percent of the total variance of the standardized inputs | share of variance captured by the second principal component | census tracts in the typology | ACS 2019-2023, 2020 land area |   | 81,343 tracts | none (each tract counts once) | American Community Survey |   |
| pop_floor | 200 | residents | definition: a tract with this many residents or fewer is left out of the typology | census tracts | 2019-2023 |   |   | none (definition) | American Community Survey |   |
| pop_in_modal | 42.9% | percent of typed residents | share of residents who live in a tract of their own county's modal type | residents of typed tracts in counties with a modal type | 2019-2023 |   | 81,314 tracts | none (full count of estimates) | American Community Survey |   |
| pop_typed_share | 97.7% | percent of residents | residents of tracts in the typology as a share of all residents of the 50 states and DC | residents of the 50 states and DC | 2019-2023 |   | 81,343 tracts | none (full count of estimates) | American Community Survey |   |
| r2_jail | 8% | percent of the population-weighted variance of tract incarceration share | R squared of the eight type means: variance between types over total variance of the tract measure, bootstrap interval over counties | residents of the tracts in the typology that have the measure | children born 1978 to 1983, incarceration on April 1, 2010 | 95% CI 7% to 9% | 81,250 tracts | ACS 2019-2023 tract population | Opportunity Atlas |   |
| r2_le | 32% | percent of the population-weighted variance of tract life expectancy | R squared of the eight type means: variance between types over total variance of the tract measure, bootstrap interval over counties | residents of the tracts in the typology that have the measure | deaths in 2010 to 2015 | 95% CI 30% to 34% | 77,763 tracts | ACS 2019-2023 tract population | USALEEP life expectancy |   |
| r2_le_c | 32% | percent of the population-weighted variance of tract life expectancy | R squared of the group means of the eight types of this chapter on the tracts of the comparison that have the measure, bootstrap interval over counties | residents of the typed tracts that have all thirteen inputs of the earlier typology | deaths in 2010 to 2015 | 95% CI 30% to 34% | 77,745 tracts | ACS 2019-2023 tract population | USALEEP life expectancy |   |
| r2_le_old | 49% | percent of the population-weighted variance of tract life expectancy | R squared of the group means of the chapter's earlier k-means fit on thirteen inputs, published until October 28, 2026, (the seven inputs of the typology plus the Black and Hispanic shares, life expectancy, the adult income rank, diabetes and smoking; same seed, groups and starts) on the tracts of the comparison that have the measure, bootstrap interval over counties | residents of the typed tracts that have all thirteen inputs of the earlier typology | deaths in 2010 to 2015 | 95% CI 47% to 51% | 77,745 tracts | ACS 2019-2023 tract population | USALEEP life expectancy |   |
| r2_mhlth | 59% | percent of the population-weighted variance of tract frequent-mental-distress rate | R squared of the eight type means: variance between types over total variance of the tract measure, bootstrap interval over counties | residents of the tracts in the typology that have the measure | PLACES 2024 release, BRFSS 2022 | 95% CI 57% to 61% | 81,341 tracts | ACS 2019-2023 tract population | CDC PLACES |   |
| r2_smk | 58% | percent of the population-weighted variance of tract smoking rate | R squared of the eight type means: variance between types over total variance of the tract measure, bootstrap interval over counties | residents of the tracts in the typology that have the measure | PLACES 2024 release, BRFSS 2022 | 95% CI 56% to 60% | 81,341 tracts | ACS 2019-2023 tract population | CDC PLACES |   |
| r2_teen | 32% | percent of the population-weighted variance of tract teen-birth share | R squared of the eight type means: variance between types over total variance of the tract measure, bootstrap interval over counties | residents of the tracts in the typology that have the measure | children born 1978 to 1983 | 95% CI 30% to 34% | 81,154 tracts | ACS 2019-2023 tract population | Opportunity Atlas |   |
| r2_teen_c | 33% | percent of the population-weighted variance of tract teen-birth share | R squared of the group means of the eight types of this chapter on the tracts of the comparison that have the measure, bootstrap interval over counties | residents of the typed tracts that have all thirteen inputs of the earlier typology | children born 1978 to 1983 | 95% CI 30% to 35% | 77,687 tracts | ACS 2019-2023 tract population | Opportunity Atlas |   |
| r2_teen_old | 46% | percent of the population-weighted variance of tract teen-birth share | R squared of the group means of the chapter's earlier k-means fit on thirteen inputs, published until October 28, 2026, (the seven inputs of the typology plus the Black and Hispanic shares, life expectancy, the adult income rank, diabetes and smoking; same seed, groups and starts) on the tracts of the comparison that have the measure, bootstrap interval over counties | residents of the typed tracts that have all thirteen inputs of the earlier typology | children born 1978 to 1983 | 95% CI 44% to 48% | 77,687 tracts | ACS 2019-2023 tract population | Opportunity Atlas |   |
| sil_best | 0.27 | silhouette (-1 to 1) | mean silhouette of the k-means fit (10 starts) with the number of types that scores highest | a random sample of census tracts in the typology | ACS 2019-2023, 2020 land area |   | 20,000 tracts | none (each tract counts once) | American Community Survey |   |
| sil_best_bench | 0.24 | silhouette (-1 to 1) | mean silhouette of the same number of k-means groups (10 starts) in simulated data with no clusters: a multivariate normal with the correlations of the inputs and one draw per tract | a random sample of census tracts in the typology | ACS 2019-2023, 2020 land area |   | 20,000 tracts | none (each tract counts once) | American Community Survey |   |
| sil_gauss | 0.14 | silhouette (-1 to 1) | mean silhouette of an 8-group k-means fit to simulated data with no clusters: a multivariate normal with the correlations of the inputs and one draw per tract | a random sample of census tracts in the typology | ACS 2019-2023, 2020 land area |   | 20,000 tracts | none (each tract counts once) | American Community Survey |   |
| sil_k8 | 0.19 | silhouette (-1 to 1) | mean silhouette of the eight types on a random sample of tracts drawn with a fixed seed | a random sample of census tracts in the typology | ACS 2019-2023, 2020 land area |   | 20,000 tracts | none (each tract counts once) | American Community Survey |   |
| sil_margin_max | 0.08 | silhouette (-1 to 1) | largest difference, over fits with 2 to 10 types, between the mean silhouette of the tracts and that of the simulated data with no clusters | a random sample of census tracts in the typology | ACS 2019-2023, 2020 land area |   | 20,000 tracts | none (each tract counts once) | American Community Survey |   |
| sil_neg_share | 6.6% | percent of sampled tracts | share of sampled tracts with a negative silhouette (nearer to another type than to their own) | a random sample of census tracts in the typology | ACS 2019-2023, 2020 land area |   | 20,000 tracts | none (each tract counts once) | American Community Survey |   |
| sil_sample | 20,000 | tracts | size of the random sample of tracts, drawn with a fixed seed, on which the silhouettes are computed | census tracts in the typology | ACS 2019-2023, 2020 land area |   | 81,343 tracts | none (sample size) | American Community Survey |   |
| ss_between | 62.8% | percent of the total variance of the standardized inputs | share of total variance of the standardized inputs that lies between the eight types | census tracts in the typology | ACS 2019-2023, 2020 land area |   | 81,343 tracts | none (each tract counts once) | American Community Survey |   |
| ss_between_gauss | 56.3% | percent of the total variance of the standardized inputs | share of variance between eight k-means groups in simulated data with no clusters (multivariate normal with the correlations of the inputs) | census tracts in the typology | ACS 2019-2023, 2020 land area |   | 81,343 tracts | none (each tract counts once) | American Community Survey |   |
| ss_within | 37.2% | percent of the total variance of the standardized inputs | share of total variance of the standardized inputs that lies within the eight types | census tracts in the typology | ACS 2019-2023, 2020 land area |   | 81,343 tracts | none (each tract counts once) | American Community Survey |   |
| t1_blk | 5.0% | percent of residents who are Black alone, Hispanic residents included | population-weighted mean of the tract Black share over the tracts of the type, bootstrap interval over counties | residents of the tracts of type 1 (High-income owners) that have the measure | 2019-2023 | 95% CI 4.3% to 5.7% | 10,661 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| t1_col | 63.4% | percent of adults 25 and over with a bachelor's degree or more | population-weighted mean of the tract college share over the tracts of the type, bootstrap interval over counties | residents of the tracts of type 1 (High-income owners) that have the measure | 2019-2023 | 95% CI 62.5% to 64.1% | 10,661 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| t1_diab | 9.4% | percent of adults with diabetes (modeled) | population-weighted mean of the tract diabetes rate over the tracts of the type, bootstrap interval over counties | residents of the tracts of type 1 (High-income owners) that have the measure | PLACES 2024 release, BRFSS 2022 | 95% CI 9.2% to 9.5% | 10,660 tracts | ACS 2019-2023 tract population | CDC PLACES |   |
| t1_his | 9.6% | percent of residents who are Hispanic of any race | population-weighted mean of the tract Hispanic share over the tracts of the type, bootstrap interval over counties | residents of the tracts of type 1 (High-income owners) that have the measure | 2019-2023 | 95% CI 8.7% to 10.5% | 10,661 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| t1_income | 157 | thousands of dollars of tract median household income | population-weighted mean of the tract tract median household income over the tracts of the type, bootstrap interval over counties | residents of the tracts of type 1 (High-income owners) that have the measure | 2019-2023 | 95% CI 153 to 160 | 10,661 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| t1_jail | 1.3% | percent of children raised in low-income families incarcerated on Census Day 2010 | population-weighted mean of the tract incarceration share over the tracts of the type, bootstrap interval over counties | residents of the tracts of type 1 (High-income owners) that have the measure | children born 1978 to 1983, incarceration on April 1, 2010 | 95% CI 1.2% to 1.4% | 10,648 tracts | ACS 2019-2023 tract population | Opportunity Atlas |   |
| t1_le | 82.1 | years of life expectancy at birth | population-weighted mean of the tract life expectancy over the tracts of the type, bootstrap interval over counties | residents of the tracts of type 1 (High-income owners) that have the measure | deaths in 2010 to 2015 | 95% CI 81.9 to 82.3 | 10,184 tracts | ACS 2019-2023 tract population | USALEEP life expectancy |   |
| t1_own | 83.8% | percent of occupied homes that are owner-occupied | population-weighted mean of the tract homeownership rate over the tracts of the type, bootstrap interval over counties | residents of the tracts of type 1 (High-income owners) that have the measure | 2019-2023 | 95% CI 83.0% to 84.7% | 10,661 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| t1_pop | 48.5 million | residents | sum of ACS 2019-2023 tract population over the tracts of the type | residents of the tracts of type 1 (High-income owners) | 2019-2023 |   | 10,661 tracts | none (full count of estimates) | American Community Survey |   |
| t1_pov | 4.0% | percent of residents below the poverty line | population-weighted mean of the tract poverty rate over the tracts of the type, bootstrap interval over counties | residents of the tracts of type 1 (High-income owners) that have the measure | 2019-2023 | 95% CI 3.9% to 4.1% | 10,661 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| t1_teen | 13.9% | percent of women raised in low-income families who had a teen birth | population-weighted mean of the tract teen-birth share over the tracts of the type, bootstrap interval over counties | residents of the tracts of type 1 (High-income owners) that have the measure | children born 1978 to 1983 | 95% CI 13.1% to 14.7% | 10,629 tracts | ACS 2019-2023 tract population | Opportunity Atlas |   |
| t2_income | 84 | thousands of dollars of tract median household income | population-weighted mean of the tract tract median household income over the tracts of the type, bootstrap interval over counties | residents of the tracts of type 2 (Older owners) that have the measure | 2019-2023 | 95% CI 82 to 86 | 3,560 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| t2_le | 80.4 | years of life expectancy at birth | population-weighted mean of the tract life expectancy over the tracts of the type, bootstrap interval over counties | residents of the tracts of type 2 (Older owners) that have the measure | deaths in 2010 to 2015 | 95% CI 80.1 to 80.6 | 3,221 tracts | ACS 2019-2023 tract population | USALEEP life expectancy |   |
| t2_own | 78.9% | percent of occupied homes that are owner-occupied | population-weighted mean of the tract homeownership rate over the tracts of the type, bootstrap interval over counties | residents of the tracts of type 2 (Older owners) that have the measure | 2019-2023 | 95% CI 78.0% to 79.7% | 3,560 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| t2_p65 | 42.3% | percent of residents aged 65 and over | population-weighted mean of the tract share aged 65 and over over the tracts of the type, bootstrap interval over counties | residents of the tracts of type 2 (Older owners) that have the measure | 2019-2023 | 95% CI 40.8% to 43.8% | 3,560 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| t2_pop | 11.3 million | residents | sum of ACS 2019-2023 tract population over the tracts of the type | residents of the tracts of type 2 (Older owners) | 2019-2023 |   | 3,560 tracts | none (full count of estimates) | American Community Survey |   |
| t3_col | 35.3% | percent of adults 25 and over with a bachelor's degree or more | population-weighted mean of the tract college share over the tracts of the type, bootstrap interval over counties | residents of the tracts of type 3 (Middle-income owners) that have the measure | 2019-2023 | 95% CI 34.9% to 35.7% | 17,400 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| t3_income | 98 | thousands of dollars of tract median household income | population-weighted mean of the tract tract median household income over the tracts of the type, bootstrap interval over counties | residents of the tracts of type 3 (Middle-income owners) that have the measure | 2019-2023 | 95% CI 97 to 99 | 17,400 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| t3_le | 79.6 | years of life expectancy at birth | population-weighted mean of the tract life expectancy over the tracts of the type, bootstrap interval over counties | residents of the tracts of type 3 (Middle-income owners) that have the measure | deaths in 2010 to 2015 | 95% CI 79.5 to 79.7 | 16,921 tracts | ACS 2019-2023 tract population | USALEEP life expectancy |   |
| t3_own | 78.8% | percent of occupied homes that are owner-occupied | population-weighted mean of the tract homeownership rate over the tracts of the type, bootstrap interval over counties | residents of the tracts of type 3 (Middle-income owners) that have the measure | 2019-2023 | 95% CI 78.4% to 79.2% | 17,400 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| t3_pop | 79.5 million | residents | sum of ACS 2019-2023 tract population over the tracts of the type | residents of the tracts of type 3 (Middle-income owners) | 2019-2023 |   | 17,400 tracts | none (full count of estimates) | American Community Survey |   |
| t4_blk | 5.7% | percent of residents who are Black alone, Hispanic residents included | population-weighted mean of the tract Black share over the tracts of the type, bootstrap interval over counties | residents of the tracts of type 4 (Low-density owners) that have the measure | 2019-2023 | 95% CI 5.3% to 6.1% | 15,491 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| t4_dens | 55 | residents per square mile of land | population-weighted geometric mean of tract density (exponential of the weighted mean of log density), bootstrap interval over counties | residents of the tracts of type 4 (Low-density owners) | 2019-2023 population, 2020 land area | 95% CI 53 to 57 | 15,491 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| t4_his | 8.1% | percent of residents who are Hispanic of any race | population-weighted mean of the tract Hispanic share over the tracts of the type, bootstrap interval over counties | residents of the tracts of type 4 (Low-density owners) that have the measure | 2019-2023 | 95% CI 7.6% to 8.7% | 15,491 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| t4_income | 68 | thousands of dollars of tract median household income | population-weighted mean of the tract tract median household income over the tracts of the type, bootstrap interval over counties | residents of the tracts of type 4 (Low-density owners) that have the measure | 2019-2023 | 95% CI 67 to 68 | 15,491 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| t4_le | 77.8 | years of life expectancy at birth | population-weighted mean of the tract life expectancy over the tracts of the type, bootstrap interval over counties | residents of the tracts of type 4 (Low-density owners) that have the measure | deaths in 2010 to 2015 | 95% CI 77.7 to 77.9 | 15,201 tracts | ACS 2019-2023 tract population | USALEEP life expectancy |   |
| t4_own | 80.1% | percent of occupied homes that are owner-occupied | population-weighted mean of the tract homeownership rate over the tracts of the type, bootstrap interval over counties | residents of the tracts of type 4 (Low-density owners) that have the measure | 2019-2023 | 95% CI 79.9% to 80.3% | 15,491 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| t4_pop | 52.3 million | residents | sum of ACS 2019-2023 tract population over the tracts of the type | residents of the tracts of type 4 (Low-density owners) | 2019-2023 |   | 15,491 tracts | none (full count of estimates) | American Community Survey |   |
| t5_col | 56.6% | percent of adults 25 and over with a bachelor's degree or more | population-weighted mean of the tract college share over the tracts of the type, bootstrap interval over counties | residents of the tracts of type 5 (Dense, college-educated renters) that have the measure | 2019-2023 | 95% CI 54.9% to 58.3% | 7,720 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| t5_dens | 8,616 | residents per square mile of land | population-weighted geometric mean of tract density (exponential of the weighted mean of log density), bootstrap interval over counties | residents of the tracts of type 5 (Dense, college-educated renters) | 2019-2023 population, 2020 land area | 95% CI 6,403 to 11,073 | 7,720 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| t5_diab | 8.7% | percent of adults with diabetes (modeled) | population-weighted mean of the tract diabetes rate over the tracts of the type, bootstrap interval over counties | residents of the tracts of type 5 (Dense, college-educated renters) that have the measure | PLACES 2024 release, BRFSS 2022 | 95% CI 8.4% to 9.0% | 7,719 tracts | ACS 2019-2023 tract population | CDC PLACES |   |
| t5_le | 80.3 | years of life expectancy at birth | population-weighted mean of the tract life expectancy over the tracts of the type, bootstrap interval over counties | residents of the tracts of type 5 (Dense, college-educated renters) that have the measure | deaths in 2010 to 2015 | 95% CI 79.9 to 80.6 | 6,980 tracts | ACS 2019-2023 tract population | USALEEP life expectancy |   |
| t5_own | 36.1% | percent of occupied homes that are owner-occupied | population-weighted mean of the tract homeownership rate over the tracts of the type, bootstrap interval over counties | residents of the tracts of type 5 (Dense, college-educated renters) that have the measure | 2019-2023 | 95% CI 34.6% to 37.7% | 7,720 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| t5_pop | 29.8 million | residents | sum of ACS 2019-2023 tract population over the tracts of the type | residents of the tracts of type 5 (Dense, college-educated renters) | 2019-2023 |   | 7,720 tracts | none (full count of estimates) | American Community Survey |   |
| t5_t4_le_gap | 2.5 | years of life expectancy at birth | difference between the population-weighted mean life expectancy of type 5 minus type 4 (Dense, college-educated renters minus Low-density owners), bootstrap interval over counties with shared draws | residents of the tracts of types 5 and 4 that have a life expectancy | deaths in 2010 to 2015 | 95% CI 2.1 to 2.8 | 22,181 tracts | ACS 2019-2023 tract population | USALEEP life expectancy |   |
| t6_his | 32.0% | percent of residents who are Hispanic of any race | population-weighted mean of the tract Hispanic share over the tracts of the type, bootstrap interval over counties | residents of the tracts of type 6 (Moderate-income, mixed-tenure) that have the measure | 2019-2023 | 95% CI 27.8% to 36.1% | 14,865 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| t6_income | 68 | thousands of dollars of tract median household income | population-weighted mean of the tract tract median household income over the tracts of the type, bootstrap interval over counties | residents of the tracts of type 6 (Moderate-income, mixed-tenure) that have the measure | 2019-2023 | 95% CI 67 to 69 | 14,865 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| t6_le | 77.3 | years of life expectancy at birth | population-weighted mean of the tract life expectancy over the tracts of the type, bootstrap interval over counties | residents of the tracts of type 6 (Moderate-income, mixed-tenure) that have the measure | deaths in 2010 to 2015 | 95% CI 77.1 to 77.6 | 14,393 tracts | ACS 2019-2023 tract population | USALEEP life expectancy |   |
| t6_own | 57.7% | percent of occupied homes that are owner-occupied | population-weighted mean of the tract homeownership rate over the tracts of the type, bootstrap interval over counties | residents of the tracts of type 6 (Moderate-income, mixed-tenure) that have the measure | 2019-2023 | 95% CI 56.6% to 58.7% | 14,865 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| t6_pop | 63.3 million | residents | sum of ACS 2019-2023 tract population over the tracts of the type | residents of the tracts of type 6 (Moderate-income, mixed-tenure) | 2019-2023 |   | 14,865 tracts | none (full count of estimates) | American Community Survey |   |
| t6_pov | 14.5% | percent of residents below the poverty line | population-weighted mean of the tract poverty rate over the tracts of the type, bootstrap interval over counties | residents of the tracts of type 6 (Moderate-income, mixed-tenure) that have the measure | 2019-2023 | 95% CI 14.3% to 14.6% | 14,865 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| t7_col | 21.7% | percent of adults 25 and over with a bachelor's degree or more | population-weighted mean of the tract college share over the tracts of the type, bootstrap interval over counties | residents of the tracts of type 7 (Lower-income renters) that have the measure | 2019-2023 | 95% CI 20.9% to 22.5% | 6,759 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| t7_le | 75.3 | years of life expectancy at birth | population-weighted mean of the tract life expectancy over the tracts of the type, bootstrap interval over counties | residents of the tracts of type 7 (Lower-income renters) that have the measure | deaths in 2010 to 2015 | 95% CI 74.9 to 75.8 | 6,268 tracts | ACS 2019-2023 tract population | USALEEP life expectancy |   |
| t7_own | 35.8% | percent of occupied homes that are owner-occupied | population-weighted mean of the tract homeownership rate over the tracts of the type, bootstrap interval over counties | residents of the tracts of type 7 (Lower-income renters) that have the measure | 2019-2023 | 95% CI 33.6% to 38.0% | 6,759 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| t7_pop | 22.7 million | residents | sum of ACS 2019-2023 tract population over the tracts of the type | residents of the tracts of type 7 (Lower-income renters) | 2019-2023 |   | 6,759 tracts | none (full count of estimates) | American Community Survey |   |
| t7_pov | 29.8% | percent of residents below the poverty line | population-weighted mean of the tract poverty rate over the tracts of the type, bootstrap interval over counties | residents of the tracts of type 7 (Lower-income renters) that have the measure | 2019-2023 | 95% CI 29.3% to 30.4% | 6,759 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| t7_sp | 10.3% | percent of households that are single-parent households | population-weighted mean of the tract single-parent share over the tracts of the type, bootstrap interval over counties | residents of the tracts of type 7 (Lower-income renters) that have the measure | 2019-2023 | 95% CI 10.1% to 10.6% | 6,759 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| t8_blk | 35.2% | percent of residents who are Black alone, Hispanic residents included | population-weighted mean of the tract Black share over the tracts of the type, bootstrap interval over counties | residents of the tracts of type 8 (High-poverty, single-parent) that have the measure | 2019-2023 | 95% CI 30.9% to 39.6% | 4,887 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| t8_col | 13.7% | percent of adults 25 and over with a bachelor's degree or more | population-weighted mean of the tract college share over the tracts of the type, bootstrap interval over counties | residents of the tracts of type 8 (High-poverty, single-parent) that have the measure | 2019-2023 | 95% CI 13.1% to 14.3% | 4,887 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| t8_his | 38.3% | percent of residents who are Hispanic of any race | population-weighted mean of the tract Hispanic share over the tracts of the type, bootstrap interval over counties | residents of the tracts of type 8 (High-poverty, single-parent) that have the measure | 2019-2023 | 95% CI 32.6% to 44.0% | 4,887 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| t8_jail | 3.5% | percent of children raised in low-income families incarcerated on Census Day 2010 | population-weighted mean of the tract incarceration share over the tracts of the type, bootstrap interval over counties | residents of the tracts of type 8 (High-poverty, single-parent) that have the measure | children born 1978 to 1983, incarceration on April 1, 2010 | 95% CI 3.3% to 3.8% | 4,884 tracts | ACS 2019-2023 tract population | Opportunity Atlas |   |
| t8_le | 74.7 | years of life expectancy at birth | population-weighted mean of the tract life expectancy over the tracts of the type, bootstrap interval over counties | residents of the tracts of type 8 (High-poverty, single-parent) that have the measure | deaths in 2010 to 2015 | 95% CI 74.2 to 75.1 | 4,595 tracts | ACS 2019-2023 tract population | USALEEP life expectancy |   |
| t8_own | 35.2% | percent of occupied homes that are owner-occupied | population-weighted mean of the tract homeownership rate over the tracts of the type, bootstrap interval over counties | residents of the tracts of type 8 (High-poverty, single-parent) that have the measure | 2019-2023 | 95% CI 32.5% to 37.9% | 4,887 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| t8_pop | 17.3 million | residents | sum of ACS 2019-2023 tract population over the tracts of the type | residents of the tracts of type 8 (High-poverty, single-parent) | 2019-2023 |   | 4,887 tracts | none (full count of estimates) | American Community Survey |   |
| t8_pov | 32.4% | percent of residents below the poverty line | population-weighted mean of the tract poverty rate over the tracts of the type, bootstrap interval over counties | residents of the tracts of type 8 (High-poverty, single-parent) that have the measure | 2019-2023 | 95% CI 31.7% to 33.2% | 4,887 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| t8_sp | 24.9% | percent of households that are single-parent households | population-weighted mean of the tract single-parent share over the tracts of the type, bootstrap interval over counties | residents of the tracts of type 8 (High-poverty, single-parent) that have the measure | 2019-2023 | 95% CI 24.6% to 25.3% | 4,887 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| t8_teen | 38.9% | percent of women raised in low-income families who had a teen birth | population-weighted mean of the tract teen-birth share over the tracts of the type, bootstrap interval over counties | residents of the tracts of type 8 (High-poverty, single-parent) that have the measure | children born 1978 to 1983 | 95% CI 37.8% to 40.0% | 4,883 tracts | ACS 2019-2023 tract population | Opportunity Atlas |   |
| total_pop | 324.7 million | residents | sum of ACS 2019-2023 tract population over the tracts in the typology | residents of the tracts in the typology | 2019-2023 |   | 81,343 tracts | none (full count of estimates) | American Community Survey |   |
| type_share_min | 10% | percent of a county's typed residents | definition: a type is present in a county when it holds at least this share of the typed residents | counties | 2019-2023 |   |   | none (definition) | American Community Survey |   |
| typed_blk | 11.9% | percent of residents who are Black alone, Hispanic residents included | population-weighted mean of the tract Black share over the tracts in the typology | residents of the tracts in the typology that have the measure | 2019-2023 |   | 81,343 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| typed_pov | 12.3% | percent of residents below the poverty line | population-weighted mean of the tract poverty rate over the tracts in the typology | residents of the tracts in the typology that have the measure | 2019-2023 |   | 81,343 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| untyped_blk | 19.4% | percent of residents who are Black alone, Hispanic residents included | population-weighted mean of the tract Black share over the tracts not in the typology | residents of the tracts not in the typology that have the measure | 2019-2023 |   | 2,268 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| untyped_no_le | 39% | percent of tracts | share of the tracts not in the typology that have no USALEEP life expectancy estimate | census tracts of the 50 states and DC that are not in the typology | deaths in 2010 to 2015 |   | 3,072 tracts | none (each tract counts once) | USALEEP life expectancy |   |
| untyped_pop | 7.7 million | residents | sum of ACS 2019-2023 tract population over the tracts not in the typology | residents of the tracts of the 50 states and DC that are not in the typology | 2019-2023 |   | 3,072 tracts | none (full count of estimates) | American Community Survey |   |
| untyped_pov | 25.8% | percent of residents below the poverty line | population-weighted mean of the tract poverty rate over the tracts not in the typology | residents of the tracts not in the typology that have the measure | 2019-2023 |   | 2,078 tracts | ACS 2019-2023 tract population | American Community Survey |   |

## Definitions

- K-means: A method that sorts places into a chosen number of groups so that places in a group resemble one another on the measures used. The analyst sets the number of groups.
- Silhouette: A score for how well a place fits the group it was assigned in a clustering. Higher values mean the place sits close to its own group and far from the others.
- Principal components: A method that combines many correlated measures into a few summary scores, ordered by how much of the variation among places each captures. The first component is the single weighted combination that captures the most.
- Ecological inference: Drawing conclusions about individuals from data about places or groups. A pattern across counties need not hold for the people inside them.

## Methods

The typology groups tracts of the 50 states and DC by 7 ACS 2019-2023 measures of resources and settlement: log median household income, the poverty rate, the college share of adults, the single-parent share of households, the homeownership rate, the share aged 65 and over and log population density. No outcome and no race or ethnicity measure is an input. The eight outcomes and the Black and Hispanic shares are compared across the types, and the earlier fit on 13 inputs, which included four outcomes and the two race and ethnicity shares, is refit as a comparison. The clustering counts every tract once while type means are weighted by tract population, and the intervals come from a bootstrap over counties that conditions on the types and does not carry uncertainty about the typology.

### Which tracts are in the typology

The typology covers the tracts of the 50 states and DC that have residents and all 7 inputs, are not institutional and have more than 200 residents. A tract is institutional when its SVI 2022 group-quarters share is 25 percent or more or it is one known chronic-care tract, the rule the book applies to USALEEP tracts; the flag is taken from SVI because USALEEP has rows only where a life expectancy exists. A typed tract need not have an outcome: each outcome is compared over the typed tracts that have it, and life expectancy exists for 77,763 of them. Puerto Rico is left out because the county map and the outcome measures cover the 50 states and DC.

Explains Table 1.

### What the inputs measure and when

The inputs are log median household income, the poverty rate, the college share of adults 25 and over, the single-parent share of households, the homeownership rate and the share aged 65 and over (ACS 2019-2023), and the log of residents per square mile of land (ACS population, 2020 TIGER land area). Density and age describe settlement and are not outcomes; they are what separates these types from the four families of the deprivation-based clustering in Families of places, which uses the fifteen ADI measures.

The outcomes are life expectancy at birth (USALEEP, deaths 2010 to 2015), the adult income rank, teen births and incarceration of children from low-income families (Opportunity Atlas, children born 1978 to 1983, interpolated from 2010 to 2020 tracts by land area), and diabetes, smoking, frequent mental distress and obesity among adults (PLACES, modeled). The Black share (Black alone, Hispanic residents included), the Hispanic share and the ADI national rank are profile columns.

Explains Table 2.

### How the types are numbered and named

Types are numbered by the population-weighted mean of the first principal component of the inputs, oriented to rise with log income, largest first, so the numbers do not depend on the initialization.

The names are the book's labels and are written in types_data.py, with an assertion that each profile fits its name. They describe income, poverty, education, family structure, tenure, age and density and name no race, ethnicity or outcome.

Explains Table 2 and Table 5.

### How a county gets a type

A county's type is the type that holds the most typed residents; a county has one when its typed tracts hold at least 50% of its residents, and a tie goes to the lower-numbered type. The count of types per county uses counties with at least 100 typed tracts and counts a type when it holds at least 10% of the typed residents.

Explains Table 3.

### Weights and intervals

Type means are weighted by ACS 2019-2023 tract population, over the typed tracts that have the measure.

Intervals are 95% percentile intervals from 2,000 draws of whole counties with replacement (seed 12345), with Connecticut tracts assigned to the eight legacy counties; differences between two types and the shares of variance between types use the same draws. The intervals condition on the types and do not carry uncertainty about the typology.

The spread of life expectancy is the highest minus the lowest population-weighted group mean, and the R squared is the variance between groups over the total, both with the county bootstrap above.

Explains Table 4.

### How the clustering was fitted and checked

The inputs are standardized with each tract counted once and clustered by k-means (scikit-learn, 8 groups, 20 starts, seed 12345, one thread). Eight groups is the book's choice, kept from the first edition, and the data do not select it. The fit weights every tract equally; a fit weighted by tract population agrees at 0.84.

Refits on 8 bootstrap resamples of tracts, assigned back to every tract by nearest centroid, agree with the typology at a median of 0.95 and at least 0.55.

The mean silhouette is computed on a random sample of 20,000 tracts (seed 12345) for the fit with 8 groups (20 starts) and for fits with 2 to 10 groups (10 starts each); it is highest at 2 groups. At each number of groups the same k-means is fit to a benchmark with no clusters, a multivariate normal with the correlations of the inputs and one draw per tract, scored on the same sample positions; the tracts score above the benchmark by at most 0.08. The 8-group benchmark with 20 starts gives 0.14. The simulated margins are symmetric and tract margins are not, so the benchmark is a guide.

The comparison refits the earlier k-means on its 13 inputs (the seven inputs, the Black and Hispanic shares, life expectancy, the adult income rank, diabetes and smoking; 8 groups, 20 starts, seed 12345, standardized over its own tracts) on the 77,745 typed tracts that have all of them, which are the tracts of the earlier typology. Its agreement with the typology, its spread of life expectancy and its R squared values are computed on those tracts, and so are the values for this chapter's types that are set beside them.

The principal-components figure shows a random sample of 6,000 typed tracts (seed 12345).

Explains Table 5.

## What these data cannot show

The types describe tracts, and every comparison is ecological: the life expectancy of the high-poverty, single-parent type is an average over tracts and says little about any one resident of one (ecological inference). The names are the book's labels for the profiles. A name such as high-poverty, single-parent summarizes two inputs of a profile that also has a Black share of 35.2% and a Hispanic share of 38.3%, and it does not describe the people who live there.

The types are built without outcomes, and the comparisons still do not show that a tract's resources or settlement produced any outcome. They show how closely conditions and outcomes move together across tracts. The measures come from different periods, with the ACS inputs for 2019 to 2023, life expectancy for deaths in 2010 to 2015, mobility for children born in 1978 to 1983 and PLACES built on 2022 survey data, so an outcome can predate the conditions it is compared with. The USALEEP estimates and the Opportunity Atlas ranks for 2020 tracts are interpolated from 2010 tracts by land area, not by population. The USALEEP estimates suit maps and gradients, the ACS file carries no margins of error, and PLACES values are modeled from covariates that include ACS measures and carry no intervals at the tract level.

The intervals in this chapter condition on the types. Tracts that lack an input are not typed. The 3,072 untyped tracts hold 7.7 million residents, and 39% of them lack a life expectancy estimate. Typed tracts without a life expectancy stay in the types and drop out of the life-expectancy comparisons.

The untyped tracts differ from the typed tracts. Their poverty rate averages 25.8% against 12.3%, and their Black share 19.4% against 11.9%. The residents of typed tracts, 97.7% of all residents, are therefore not a random sample of the country. The county map shows one type per county and hides the mix, and Puerto Rico is outside the typology and the map.

## References

1. Murray, Christopher J. L.; Kulkarni, Sandeep C.; Michaud, Catherine; Tomijima, Niels; Bulzacchelli, Maria T.; Iandiorio, Terrell J.; Ezzati, Majid (2006). Eight Americas: Investigating Mortality Disparities across Races, Counties, and Race-Counties in the United States. PLoS Medicine 3(9): e260. https://doi.org/10.1371/journal.pmed.0030260
2. Chetty, Raj; Hendren, Nathaniel (2018). The Impacts of Neighborhoods on Intergenerational Mobility I: Childhood Exposure Effects. The Quarterly Journal of Economics 133(3): 1107-1162. https://doi.org/10.1093/qje/qjy007

## Provenance

- Pipeline version: 0.1.0
- Build script: book2/chapters/eight-types-of-neighborhood/build.py (sha256 0401bf488cb85b31eb791cd96bfa7c70d74e205cd1aea5763da3d96c8ab3a064)
- Built on: 2026-10-28
- Site commit: c2b1ff3309cb4a10833f884da91ac0057a66ad0a

Inputs:

| File | Size (bytes) | SHA-256, first 1 MB | SHA-256, last 1 MB |
| --- | ---: | --- | --- |
| book2/chapters/eight-types-of-neighborhood/types_data.py | 25,686 | 6d7e5d157e6c1d2a6ce404220d256e0bbd7dc6b8136bafdff53ceb0a07468122 | 6d7e5d157e6c1d2a6ce404220d256e0bbd7dc6b8136bafdff53ceb0a07468122 |
| data/analytic/tract_atlas.parquet | 39,659,925 | 29d5e4d943552d54aeb6ba92aadc939d9bd15bdea553c78357dbd516757be627 | 1e10d1ecf722340f23ee2212d4999a90250f85b65fa89b39e32db06d9472e22e |
| data/geo/tracts_2020.parquet | 619,858,391 | f4f05c8c7a63d1785ec2aad283de3f6fa839b6c437f13a1b96cea31d3344bf0f | 7b3811247c6b3e322954f24a4a2384af52aef6ed8688f8dbdb0407d21ef80a59 |
| data/staged/crosswalks/ct_tract2020_to_old_county.parquet | 15,842 | 961ef39f31596ef65fca2c67d5f2f63e533c5dc9b49745d3bb6437844e4b831d | 961ef39f31596ef65fca2c67d5f2f63e533c5dc9b49745d3bb6437844e4b831d |
| data/staged/svi/svi_tract2020_2022.parquet | 17,740,415 | cb3860d8edc2aaf1f2a02211a58f7a0b80457eb6ad9a37b5ec8abb2d5da9da10 | e371a4a696d5bed05d54ad7c5b6ca2504508805b847ec415c2322be7f66ec2b2 |

Outputs:

| File | SHA-256 |
| --- | --- |
| facts.json | 9b5fa324c28f5faf9951ff17d7d2a7b52cc0f8db1da35566a157c0f14b5565f9 |
| figures/modal_type_map.json | e7bf6b97530b641ddf30ea60305a69369d2b4ce25de0d2449b9ef08d03f1aa43 |
| figures/pca_continuum.json | 061b33a95bc751d1dbfdc8181fb4d4eb3edea046e4cd6e8104287a3fef7cf1c4 |
| figures/sample_flow.json | 0149d90223074f3db14f5b38f1f46a74ba476a066dd1617ff307173984c0f2c3 |
| figures/type_outcomes.json | b01096122d02bb7d52d28939eda8ec2b8154a0a4b162f284033cb5c337bae210 |
| figures/type_profiles.json | dbb555769ecef5f8a9d2d73d9a9615fa52fffdf60909ac3480f4a0803d2f92de |
| meta.json | 8287fdd8357c1fca35bbcbe6612351bb4cda7451a13ef437ae1d950c34a91412 |
| public/data/eight-types-of-neighborhood/modal_type_map.json | b433c7cd3f9a4c204d18126d8c943e69701be6e917e51ee96e1f9ec9813390b0 |
| public/data/eight-types-of-neighborhood/pca_continuum.json | 06c5add45f70ed23946415d3f1a277bf6cd241a750837c5360192e7dd0b36553 |
| refs.json | 0d466e104a404555a8699785fede9eb5fddf2508077d824b4daf5465aa7fde04 |

## Machine formats

- Report (HTML): https://vishalsingh.org/zipcode/eight-types-of-neighborhood/report
- Report (Markdown): https://vishalsingh.org/zipcode/eight-types-of-neighborhood/report.md
- Every estimate (JSON): https://vishalsingh.org/zipcode/report/eight-types-of-neighborhood/facts.json
- Table 1, figure "sample_flow" (JSON): https://vishalsingh.org/zipcode/report/eight-types-of-neighborhood/figures/sample_flow.json
- Table 2, figure "type_profiles" (JSON): https://vishalsingh.org/zipcode/report/eight-types-of-neighborhood/figures/type_profiles.json
- Table 3, figure "modal_type_map" (JSON): https://vishalsingh.org/zipcode/report/eight-types-of-neighborhood/figures/modal_type_map.json
- Table 4, figure "type_outcomes" (JSON): https://vishalsingh.org/zipcode/report/eight-types-of-neighborhood/figures/type_outcomes.json
- Table 5, figure "pca_continuum" (JSON): https://vishalsingh.org/zipcode/report/eight-types-of-neighborhood/figures/pca_continuum.json
- Payload of "modal_type_map" (JSON): https://vishalsingh.org/zipcode/data/eight-types-of-neighborhood/modal_type_map.json
- Payload of "pca_continuum" (JSON): https://vishalsingh.org/zipcode/data/eight-types-of-neighborhood/pca_continuum.json
- Index of every report: https://vishalsingh.org/zipcode/llms.txt; all reports in one file: https://vishalsingh.org/zipcode/llms-full.txt

---

# Places that beat the model: report

Chapter 9.3 of ZIP Code Destiny, by Vishal Singh (NYU Stern). Generated from the chapter's data files; the reading version is at https://vishalsingh.org/zipcode/places-that-beat-the-model. Table n of this report is figure n of the chapter.

## Identification

- Title: Places that beat the model
- Label: Coda
- Chapter: 9.3 (Part IX, Reading the patterns; Reading the patterns)
- Question: Where do children from low-income families rise further than five census measures predict, and do people move toward those places?
- Dek: Five census measures from 2019 to 2023 account for 44.0% of the variation in tract upward mobility for children born in 1978 to 1983, and counties below that prediction cluster in the South.
- Grain: counties
- Status: published
- Published: 2026-10-09
- Updated: 2026-10-28
- Data sources:
  - Opportunity Atlas. Adult household income rank of children born 1978 to 1983 to parents at the 25th income percentile, 2020-tract file (the model's outcome) and county file (the migration comparison); years: 1978-1983 birth cohorts (2018 release); weights: ACS 2019-2023 tract population; county population
  - American Community Survey. ACS 2019 to 2023 tract measures (poverty, college, single-parent share, median household income, race and ethnicity, homeownership, population) and tract median home values, averaged to counties; years: 2019-2023 (5-year); weights: 2019-2023 tract population
  - Census TIGER/Line tracts. 2020 tract land area, for residents per square mile; years: 2020; weights: none (land area)
  - CDC PLACES. Current smoking among adults, 2024 release, tracts; years: 2024 release (tracts); 2025 release (ZCTAs, 2023 data year); weights: 2019-2023 tract population
  - EPA EJScreen. Share of housing units built before 1960, 2024 release, tracts; years: 2024 (indicators); ACS 2018-2022 (demographics); weights: 2019-2023 tract population
  - USALEEP life expectancy. Life expectancy at birth by tract, deaths in 2010 to 2015, area-weighted onto 2020 tracts, for the life-expectancy version of the model and county means; years: 2010-2015; weights: 2019-2023 tract population
  - Area Deprivation Index. Area Deprivation Index 2023, national rank, tracts averaged to counties; years: 2023 release (version 4.0.1; built on ACS 2019-2023); weights: 2019-2023 tract population
  - IRS migration data. County-to-county migration, filing years 2022 and 2023: arriving and leaving returns and adjusted gross income, all origins and destinations; years: 2022-2023 (filing years); weights: none (full counts of returns)
  - IRS Statistics of Income. IRS Statistics of Income county file, tax year 2022: returns filed, summed over the income brackets, the denominator of the net migration rate; years: tax year 2022; weights: none (full counts of returns)
- Canonical URL: https://vishalsingh.org/zipcode/places-that-beat-the-model
- Report: https://vishalsingh.org/zipcode/places-that-beat-the-model/report
- Markdown: https://vishalsingh.org/zipcode/places-that-beat-the-model/report.md
- BibTeX key: singh2026placesthatbeatthemodel

```bibtex
@misc{singh2026placesthatbeatthemodel,
  author = {Singh, Vishal},
  title = {Places that beat the model},
  year = {2026},
  url = {https://vishalsingh.org/zipcode/places-that-beat-the-model},
  note = {In ZIP Code Destiny. Published 9 October 2026; updated 28 October 2026.}
}
```

## Principal estimates

The facts the dek and the opening prose cite, in order of appearance.

| Estimand | Population | Period | Value | Interval | n | Weights | Source | Fact id |
| --- | --- | --- | ---: | --- | ---: | --- | --- | --- |
| R-squared of the population-weighted least-squares fit of tract mobility on poverty, college share, single-parent share, log median income and Black share, county-bootstrap interval | 2020 census tracts with more than 200 residents, a mobility estimate and all five predictors | children born 1978 to 1983, household income measured in 2014 to 2015; predictors ACS 2019 to 2023 | 44.0% | 95% CI 41.8% to 46.1% | 82,791 tracts | ACS 2019-2023 tract population | Opportunity Atlas | r2 |
| count of tracts in the model | 2020 census tracts with more than 200 residents, a mobility estimate and all five predictors | children born 1978 to 1983, household income measured in 2014 to 2015; predictors ACS 2019 to 2023 | 82,791 |   |   | none (full count) | Opportunity Atlas | model_n |
| one minus the R-squared of the five-measure model, county-bootstrap interval | 2020 census tracts with more than 200 residents, a mobility estimate and all five predictors | children born 1978 to 1983, household income measured in 2014 to 2015; predictors ACS 2019 to 2023 | 56.0% | 95% CI 53.9% to 58.2% | 82,791 tracts | ACS 2019-2023 tract population | Opportunity Atlas | model_unexplained |
| correlation across counties of the county mean residual of mobility with the net migration rate, weighted by county population, bootstrap interval over states | counties with a residual and a net migration rate | returns filed in 2022 matched with returns filed in 2023 (tax years 2021 and 2022); mobility for children born 1978 to 1983, household income measured in 2014 to 2015; predictors ACS 2019 to 2023 | −0.44 | 95% CI −0.57 to −0.28 | 1,576 counties | ACS 2019-2023 county population | IRS migration data | r_resid_rate |

## Findings

Five census measures account for 44.0% (95% CI 41.8% to 46.1%) of the variation in upward mobility across 82,791 tracts, where upward mobility is the adult household income rank of children raised in low-income families.[1] The children were born in 1978 to 1983, and the five measures describe the tracts in 2019 to 2023. The remaining 56.0% is not scattered at random. Averaged to counties, it lies below the prediction in the South and above it in the largest counties around New York. The counties above the prediction lost tax filers on net in filing years 2022 and 2023, a correlation of −0.44 (95% CI −0.57 to −0.28) between the county residual and the net migration rate. Part of that correlation is built in by timing, since the five measures describe counties after many of the movers had arrived.

The chapter takes two steps. The first fits the model, averages its residuals to counties and describes the counties at both ends. The second follows tax filers between counties and asks whether net migration runs toward the places where children from low-income families rise furthest. The life-expectancy gradient that the book starts from is in Chapter 8.4, Place and the length of a life (https://vishalsingh.org/zipcode/place-and-the-length-of-a-life), and the Opportunity Atlas map of mobility is in Chapter 4.3, Race, income and mobility (https://vishalsingh.org/zipcode/race-income-and-mobility). A residual is the part of mobility that these five measures leave unexplained. It is not a measure of what a place does to a child.

### A five-measure model of upward mobility

The outcome is the position in the national distribution of household income that children born in 1978 to 1983 to parents at the 25th income percentile had reached by 2014 to 2015, as the Opportunity Atlas estimates it. The predictors are the poverty rate, the share of adults with a bachelor's degree or more, the single-parent household share, the natural log of median household income and the Black share of residents (Black alone, of any ethnicity). All five come from the American Community Survey (ACS) for 2019 to 2023. A weighted least-squares line, with each tract counted in proportion to its population, fits 82,791 tracts in 3,136 counties after leaving out tracts with 200 residents or fewer. It accounts for 44.0% (95% CI 41.8% to 46.1%) of the variance in tract mobility, and 56.0% (95% CI 53.9% to 58.2%) remains. The interval comes from a bootstrap that resamples whole counties.

The outcome and the predictors describe different decades. The children grew up in the 1980s and 1990s, and the predictors describe 2019 to 2023. Each prediction therefore asks what a tract's census profile today implies for children raised there a generation ago. Table 1 plots the actual rank against the prediction. Tenths of tracts ranked by prediction lie close to the diagonal, so the line is not tilted at either end. The dots spread around it with a population-weighted standard deviation of 5.0 percentile points. The gap between a tract's actual and predicted rank is its residual.

Table 1. Five census measures account for 44.0% of the variation in tract upward mobility, and the population-weighted standard deviation of the gap between actual and predicted rank is 5.0 percentile points

A tract's residual mixes things the data cannot separate. The Atlas estimate for a tract rests on a modest number of children and carries added privacy noise. The five measures omit much that differs between places, such as schools and local labor markets. The outcome and the predictors also describe different decades. The next section asks whether residuals averaged to counties behave like noise.

### Counties above and below the prediction

The residual of a county is the mean of its tract residuals, weighted by tract population. A county needs at least 8 tracts in the model, which leaves 1,576 counties that hold 94% of the model's population (Table 2). County residuals run from −4.8 percentile points at the fifth percentile of counties to +5.0 at the ninety-fifth.

Ranking counties on a noisy estimate can produce a list of extremes that is mostly noise, so two checks come first. The typical standard error of a county residual is 0.7 percentile points, small beside that spread. It treats the tracts of a county as independent and leaves out errors that all of them share. The second check splits the tracts of each county at random into two subsets, ranks counties on one subset and reads the residual off the other. Across counties the residuals of the two subsets correlate +0.85 (95% CI 0.80 to 0.88). The tenth of counties highest in the first subset has a mean residual of +5.0 (95% CI 3.7 to 6.1) in the second, against −4.4 (95% CI −5.0 to −3.4) for the lowest tenth. The ranking is therefore not an artifact of the noise in a subset.

The floor matters for the extremes. A floor of 5 tracts maps 2,140 counties and puts the ninety-fifth percentile of county residuals at +6.6. A floor of 15 maps 944 and puts it at +3.9. The correlation of the residual with net migration, below, keeps its sign at both floors.

Among the 281 counties with at least 250,000 residents, the largest residuals, in percentile points of adult income rank, are in Richmond County, New York, which is Staten Island (+6.2 (95% CI 5.3 to 7.2)), Nassau County, New York (+6.1 (95% CI 5.4 to 6.8)), Bergen County, New Jersey (+6.1 (95% CI 5.5 to 6.7)) and Queens (+5.6 (95% CI 5.2 to 6.1)).

The smallest residuals are in St. Johns County, Florida (−5.8 (95% CI −7.3 to −4.4)), Denver (−5.4 (95% CI −6.1 to −4.6)) and Buncombe County, North Carolina, which holds Asheville (−5.4 (95% CI −6.1 to −4.6)).

Of the 10 counties of that size with the largest residuals, 9 lie in New York, New Jersey, Pennsylvania, Massachusetts or Virginia. Of the 10 smallest, 8 lie in the South, a Census region that includes the District of Columbia.

Across the mapped counties, the South holds the bottom tenth. Of its counties, 81% lie in the South, which holds 46% of all mapped counties. The share is 83% with a floor of 15 tracts.

The top tenth depends on the floor. With the floor of 8 tracts, 54% of the top tenth lie in the Midwest, which holds 28% of the counties. Another 14% lie in the Northeast, which holds 13%.

With a floor of 15 tracts the Midwest share of the top tenth is 37% against 23% of the counties. The Northeast share is 28% against 17%. The Midwestern part of the top tenth is therefore made of counties with few tracts, and the largest counties above the prediction are the ones near New York.

Table 2. Counties below the prediction cluster in the South, and the largest counties above it lie around New York

 (Note: Counties that gained many college-educated adults after the 1990s have high predictors for children who grew up there earlier, and they would sit below the line for that reason alone. The staged data hold no census measures from before 2019, so the chapter cannot test that explanation.)

### Older housing and fewer smokers mark the counties above the prediction

Table 3 compares the tenth of counties with the largest residuals, home to 18.1 million people, with the tenth with the smallest, home to 17.2 million people. The table under the figure gives each top minus bottom difference with an interval from draws shared by the two groups. Five differences have intervals that exclude zero. Housing built before 1960 makes up 38% (95% CI 28% to 47%) of housing units in the top tenth and 18% (95% CI 15% to 22%) in the bottom tenth, a difference of +20.1 (95% CI 9.8 to 29.2) points.

The smoking share of adults is 13.5% (95% CI 12.6% to 14.8%) in the top tenth and 15.9% (95% CI 15.2% to 16.6%) in the bottom tenth, a difference of −2.4 (95% CI −3.5 to −0.9) points. The Hispanic share differs by +5.2 (95% CI 0.7 to 9.3) points and the poverty rate by −1.9 (95% CI −3.4 to −0.2) points.

The average resident of the top tenth lives at 9,090 (95% CI 2,838 to 18,233) people per square mile, against 2,286 (95% CI 1,149 to 3,952) in the bottom tenth, a difference of +6,803 (95% CI 97 to 15,794), whose lower limit is close to zero. The differences in the Black share, in ownership and in college include zero.

Table 3. Counties in the top tenth by residual have more pre-1960 housing, a larger Hispanic share, less poverty and fewer smokers than the bottom tenth, and the differences in the Black share, ownership and college include zero

The same five measures account for 44.2% (95% CI 42.0% to 46.5%) of the variance of tract life expectancy at birth for deaths in 2010 to 2015. County residuals for the two outcomes move together. Across counties, the residual of mobility correlates +0.39 (95% CI 0.27 to 0.51) with the residual of life expectancy. Counties above the prediction for children raised in the 1980s therefore tend to be above it for the life expectancy of residents in 2010 to 2015. What the five measures leave out is partly shared between the two outcomes, and these data do not say what it is.

### Net income moved toward the South in filing years 2022 and 2023

The Internal Revenue Service matches the individual returns filed in 2022 with those filed in 2023 on the filer's taxpayer identification number. It assigns each return a county from the full ZIP code on it and counts a return as a migrant when the county differs between the two filings.[2] The unit is the filer, usually a household, and not the person. People who do not file are missing, and the address on a return is a mailing address. Arrivals include moves from abroad and departures include moves abroad. Arriving returns equal 5.6% of the returns filed for tax year 2022. Adjusted gross income is the amount on the 2023 return, and net income is arriving minus leaving.

Of the 3,086 counties with both directions published, 1,719 gained net income and 1,367 lost it, and the gaining counties together received 71.6 billion dollars net (Table 4).

Among counties of 50,000 residents or more, San Francisco lost the most per resident, 2,488 dollars, and Collier County, Florida, gained the most, 5,793 dollars.

Pooled by Census region, net income per resident was 270 (95% CI 22 to 528) dollars in the South, −332 (95% CI −475 to −54) in the Northeast and −172 (95% CI −296 to −60) in the Midwest. The interval for the West includes zero.

Table 4. Net adjusted gross income was positive in 1,719 counties and negative in 1,367, with the largest losses per resident in San Francisco and Boston's Suffolk County and the largest gains in Florida

### Net migration is lower where childhood mobility is higher

Table 5 sets each county's net migration rate, arriving minus leaving returns per thousand returns filed, against the Atlas mobility rank of children raised there. Across 3,086 counties the correlation is −0.19 (95% CI −0.31 to −0.02) when each county counts in proportion to its population and −0.23 (95% CI −0.33 to −0.10) when each counts once. The slope of the fitted line is −0.39 (95% CI −0.68 to −0.05) returns per thousand for each percentile point of the mobility rank.

The sign is the same against life expectancy for deaths in 2010 to 2015, −0.25 (95% CI −0.37 to −0.09) weighted and −0.13 (95% CI −0.23 to −0.01) unweighted. Counties where children from low-income families rose further and adults lived longer gained fewer filers.

The tenths show where the association sits. The two tenths highest on mobility had net rates of −5.2 (95% CI −7.9 to −1.6) and −3.6 (95% CI −4.7 to −1.9) returns per thousand, with intervals below zero. The lowest tenths gained in the point estimates, +0.9 (95% CI −2.7 to 3.6) and +3.6 (95% CI −0.7 to 7.3), but their intervals include zero.

Deprivation gives a less consistent picture, since the correlation is +0.19 (95% CI −0.05 to 0.34) weighted and −0.19 (95% CI −0.28 to −0.09) unweighted. The most deprived tenth of counties had a net rate of −4.5 (95% CI −6.7 to −2.3) returns per thousand, the least deprived tenth had −2.4 (95% CI −4.6 to 1.5), and the second least deprived had +2.6 (95% CI −0.1 to 5.1). Net out-migration of filers therefore shows at the most deprived end with an interval below zero and at the least deprived end only as a point estimate.

Table 5. Net migration of tax filers is lower in counties with higher childhood mobility, a population-weighted correlation of −0.19, and the two highest tenths by mobility lost filers on net

The county residual from the model moves with net migration more strongly than mobility itself does. Across the counties with a residual, the correlation of the residual with the net migration rate is −0.44 (95% CI −0.57 to −0.28) weighted and −0.39 (95% CI −0.49 to −0.27) counting each county once. Part of that link is built in by timing. The predictors describe 2019 to 2023, after many movers had arrived, so a county that gained college-educated residents has high predictors and, if its children's outcomes predate those residents, a negative residual. The data cannot separate that mechanical link from any other.

### Home values and net income per resident

Table 6 plots net income per resident against county home value. Across 3,084 counties the correlation is −0.30 (95% CI −0.47 to 0.02) weighted by population and +0.18 (95% CI 0.04 to 0.33) counting each county once. The first interval includes zero and the second lies above zero. The two differ in sign because the weighted version counts each county in proportion to its population.

Net income per resident did not decline steadily with home value. In the lowest tenth of counties by home value, net income was −109 (95% CI −142 to −70) dollars per resident, the only tenth with an interval entirely below zero. In the top tenth it was −103 (95% CI −298 to 238), and the ninth tenth received 103 (95% CI −129 to 382). Both of those intervals include zero. The point estimates are lowest in the cheapest tenth and the most expensive tenth and highest in the ninth. The relation of life expectancy to home value, and the years of life that go with each doubling of it, are in Chapter 5.2, Home value, income and redlining (https://vishalsingh.org/zipcode/home-value-income-and-redlining).

Table 6. Net income per resident does not fall steadily with county home value, with a population-weighted correlation of −0.30 and +0.18 when each county counts once

### Pooled by deprivation, arriving and leaving returns report similar incomes

One account of these flows is income sorting, in which richer filers replace poorer ones. Table 7 compares the average adjusted gross income of arriving and leaving returns, pooled over tenths of counties by deprivation. Both averages fall by nearly the same amount from the least to the most deprived tenth. The arriving minus leaving gap runs from −2,110 to 2,486 dollars across the tenths. Its interval includes zero in 9 of the ten, among them the most deprived tenth, where the gap is −337 (95% CI −1126 to 559) dollars. Only the ninth tenth has arriving returns that report more than leaving returns by an amount the interval separates from zero.

Counties taken one by one give a different answer from pooling. The correlation of the county gap with deprivation is +0.02 (95% CI −0.13 to 0.13) weighted by population, a value near zero, and −0.21 (95% CI −0.29 to −0.14) counting each county once. With each county counted once, arriving returns report less than leaving returns where deprivation is higher. In the median county, arriving returns report 54,902 (95% CI 52,722 to 57,194) dollars of income and leaving returns 52,444 (95% CI 50,430 to 54,356), a median gap of 2,514 (95% CI 1,466 to 3,433) dollars. Similar incomes therefore hold for the pooled tenths and the population-weighted correlation, and not for the median county or the unweighted correlation. The averages describe returns, and they hide who within a county moves.

Table 7. In pooled tenths of counties by deprivation, returns arriving report about the same income as returns leaving, though the median county's arriving returns report 2,514 dollars more

> Causal evidence: Whether places change children's incomes and whether housing costs steer migration (Movers design; Bartik-type comparisons of state income convergence, split by predetermined land-use regulation and by land availability)
>
> Two studies ask the causal questions that this chapter's comparisons point to.
>
> Chetty and Hendren followed children whose families moved between counties and commuting zones, using variation in the age of the child at the move, including siblings in the same family. The outcomes of children who moved to a county with higher incomes among its permanent residents moved toward those of permanent residents in proportion to the years of childhood they spent there.[3] The county-level companion paper estimates the causal contribution of each county and identifies affordable counties with strong outcomes.[4] The design identifies how much a move changes a child's outcome. It does not identify why a county sits above or below a census-based prediction.
>
> Ganong and Shoag count the state supreme and appellate court cases per capita that mention land use, as a measure of land-use regulation, and relate the convergence of state incomes to it. To check that regulation is not a result of slower convergence, they split states by the same measure for 1956 to 1965, before the slowdown, and by land availability, a Bartik-type use of historical differences across places with a national change. Convergence slowed where regulation was tighter, and rising housing prices in high-income states deterred the migration of low-skill workers toward them.[5] The study covers states from 1940 to 2010. It does not test whether housing cost steers the county flows of 2022 and 2023.

### What these data cannot show

None of the residuals measures what a place does for a child. A county residual combines noise in the Atlas estimates, measures the model omits and the gap between the decades of the outcome and of the predictors. The standard error of a county residual leaves out errors that all of a county's tracts share. The floor of 8 tracts excludes 1,567 map units, so the map says nothing about them. The traits in Table 3 describe counties selected on the residual, and their intervals treat the two groups as fixed.

The migration figures cover one pair of filing years. Internal migration in the United States has declined since the 1980s.[6] Arriving returns equal only 5.6% of the returns filed here, so most filers in every county did not change county. The two filings follow the pandemic years, in which demand for housing shifted from the centers of large cities toward their outskirts,[7] and one pair of years may not describe other periods. The unit is the county, so the correlations compare places and say nothing about why any household moved. Several of them change sign with the weighting. Connecticut's flows are published by planning region and are spread over its legacy counties by population, which is approximate. A county's home value is the mean of its tracts' median home values, weighted by population, and a tract median at the Census top code enters at the code. Puerto Rico is left out of the model and the maps.

The two halves of the chapter do not measure the same people. The residual scores the adult outcomes of children raised in the 1980s and 1990s, and the migration data score the movements of filers in 2022 and 2023. Reading one against the other compares where two different populations sit on the same map. The adjusted gross income of a return is the amount on the 2023 return, for calendar 2022, and a move between the two filings can come after part or all of that income was earned.

## Tables

### Table 1. Five census measures account for 44.0% of the variation in tract upward mobility, and the population-weighted standard deviation of the gap between actual and predicted rank is 5.0 percentile points

Adult household income rank (percentile) of children raised in low-income families against the rank predicted from poverty, college, single-parent share, income and Black share, one dot per tract (random sample)

No inline rows: the values of Table 1 are in https://vishalsingh.org/zipcode/data/places-that-beat-the-model/predicted_vs_actual.json.

Note: The dots are a random sample of the tracts inside the plotted range, drawn with a fixed seed (stated in the methods note); the R-squared, the residuals and the markers use every tract in the model.

Note: Predictors are the poverty rate, the share of adults with a bachelor's degree or more, the single-parent household share, the natural log of median household income and the Black share of residents (Black alone, of any ethnicity), all from the ACS for 2019 to 2023. Tracts with 200 residents or fewer are left out, and each tract is weighted by its ACS population.

Note: The outcome describes children raised in the 1980s and 1990s and the predictors describe 2019 to 2023, so the residual includes whatever changed in between. Ranks are percentiles of the national distribution of household income for children born in 1978 to 1983, measured in 2014 to 2015. The Opportunity Atlas adds noise to its estimates to protect privacy and the staged values are never clipped. The file on 2020 tracts is interpolated from 2010 tracts by land area.

Source: Opportunity Atlas tract outcomes, 1978-1983 birth cohorts (2020 tracts), and ACS 2019-2023 predictors; population-weighted (ACS tract population), modeled estimates

Data: https://vishalsingh.org/zipcode/report/places-that-beat-the-model/figures/predicted_vs_actual.json

### Table 2. Counties below the prediction cluster in the South, and the largest counties above it lie around New York

Population-weighted mean residual of tract upward mobility (actual minus predicted rank, percentile points) by county, counties with at least 8 tracts in the model

No inline rows: the values of Table 2 are in https://vishalsingh.org/zipcode/data/places-that-beat-the-model/residual_map.json.

Note: A county is mapped when at least 8 of its tracts are in the model, which covers 94% of the model's population. Each gray county gives its reason in the tooltip.

Note: The interval in the tooltip treats a county's tracts as independent draws, so it leaves out errors that all tracts of a county share. The typical standard error is 0.7 percentile points.

Note: Connecticut is drawn as its eight legacy counties; its tracts are assigned to them exactly. Alaska is drawn. Puerto Rico is left out of the model and is not drawn. Ranks are percentiles of the national distribution of household income for children born in 1978 to 1983, measured in 2014 to 2015. The Opportunity Atlas adds noise to its estimates to protect privacy and the staged values are never clipped. The file on 2020 tracts is interpolated from 2010 tracts by land area.

Source: Opportunity Atlas tract outcomes, 1978-1983 birth cohorts, and ACS 2019-2023; population-weighted county means of model residuals, modeled estimates

Data: https://vishalsingh.org/zipcode/report/places-that-beat-the-model/figures/residual_map.json

### Table 3. Counties in the top tenth by residual have more pre-1960 housing, a larger Hispanic share, less poverty and fewer smokers than the bottom tenth, and the differences in the Black share, ownership and college include zero

Population-weighted tract means in the bottom and top tenths of counties by mean residual (equal numbers of counties), with 95% intervals from a bootstrap over counties

| Trait | Bottom tenth | Top tenth | Top minus bottom (points) | Lower limit of the 95% interval | Upper limit of the 95% interval |
| --- | ---: | ---: | ---: | ---: | ---: |
| Housing built before 1960 (share of housing units) | 18.1 | 38.3 | +20.1 | +9.8 | +29.2 |
| Hispanic residents (share of residents) | 10.5 | 15.7 | +5.2 | +0.7 | +9.3 |
| Black-alone residents (share of residents) | 11.8 | 8.6 | −3.2 | −7.2 | +0.3 |
| Adults who currently smoke (crude prevalence, share of adults) | 15.9 | 13.5 | −2.4 | −3.5 | −0.9 |
| Owner-occupied homes (share of households) | 70.2 | 68.2 | −1.9 | −8.4 | +4.0 |
| Adults with a bachelor's degree or more (share of adults) | 33.9 | 37.6 | +3.6 | −2.5 | +9.2 |
| Residents below the poverty line (share of residents) | 12.6 | 10.7 | −1.9 | −3.4 | −0.2 |

Note: Tenths hold equal numbers of counties ranked on the mean residual, and the tracts of those counties form the groups. The groups are fixed before the draws, so the intervals leave out the uncertainty of the ranking.

Note: Race shares are race alone and include Hispanic residents of any race. Smoking is the PLACES crude prevalence among adults, a model-based estimate built partly from ACS inputs. The pre-1960 share is from EJScreen, which draws it from the ACS for 2018 to 2022. Tracts without a value drop out of that row.

Note: The poverty, college and Black-share rows are predictors in the model, so their gaps are differences between whole counties; inside a tract the residual is unrelated to them by construction.

Source: ACS 2019-2023, CDC PLACES 2024 release (smoking, modeled estimates) and EPA EJScreen 2024 (housing age); population-weighted (ACS tract population) tract means by group of counties

Data: https://vishalsingh.org/zipcode/report/places-that-beat-the-model/figures/overperformer_traits.json

### Table 4. Net adjusted gross income was positive in 1,719 counties and negative in 1,367, with the largest losses per resident in San Francisco and Boston's Suffolk County and the largest gains in Florida

Net adjusted gross income (arriving minus leaving returns) in dollars per resident by county, returns filed in 2022 matched with returns filed in 2023

No inline rows: the values of Table 4 are in https://vishalsingh.org/zipcode/data/places-that-beat-the-model/net_agi_map.json.

Note: Arrivals and departures are tax returns matched across the 2022 and 2023 filings on the filer's taxpayer identification number, with a county assigned from the ZIP+4 on each return, so the unit is the filer and not the person. People who do not file are missing, and the IRS withholds small flows; a withheld direction leaves the county gray, never at zero. Income is adjusted gross income on the 2023 return, converted from thousands of dollars.

Note: Arrivals include moves from abroad, departures include moves abroad, and both include moves to and from the territories. Net income is divided by the ACS 2019-2023 population, which counts people who do not file.

Note: Connecticut's IRS flows are published by planning region and are allocated to its legacy counties in proportion to 2020 population, which is approximate. Alaska is drawn. Puerto Rico is not drawn.

Source: IRS Statistics of Income county-to-county migration data, filing years 2022 and 2023, and ACS 2019-2023 population; full counts of returns, not weighted

Data: https://vishalsingh.org/zipcode/report/places-that-beat-the-model/figures/net_agi_map.json

### Table 5. Net migration of tax filers is lower in counties with higher childhood mobility, a population-weighted correlation of −0.19, and the two highest tenths by mobility lost filers on net

Net migrating returns per 1,000 returns filed for tax year 2022, by county, against the Opportunity Atlas upward-mobility rank, returns filed in 2022 matched with returns filed in 2023

#### Table 5.1. Counties

Counties with a higher upward-mobility rank had lower net migration, a slope of −0.39 returns per thousand for each percentile point

Net migrating returns per 1,000 returns filed against the mean adult income rank of children raised in low-income families, one dot per county sized by population

No inline rows: the values of Table 5.1 are in https://vishalsingh.org/zipcode/data/places-that-beat-the-model/migration_vs_mobility.json.

#### Table 5.2. Tenths by mobility

The two highest tenths of counties by mobility had net rates of −5.2 and −3.6 returns per thousand, with intervals below zero

Pooled net migrating returns per 1,000 returns filed, by tenth of counties ranked on upward mobility

| conf_high | conf_low | decile | estimate | n_units |
| ---: | ---: | --- | ---: | ---: |
| 3.64 | −2.67 | 1 (lowest) | 0.880 | 308 |
| 7.30 | −0.750 | 2 | 3.63 | 309 |
| 5.34 | −2.12 | 3 | 1.62 | 308 |
| 3.01 | −0.430 | 4 | 1.50 | 309 |
| 0.710 | −2.70 | 5 | −1.01 | 309 |
| 1.46 | −4.64 | 6 | −2.61 | 308 |
| 5.12 | −3.15 | 7 | 0.630 | 309 |
| 2.10 | −3.03 | 8 | −0.790 | 308 |
| −1.56 | −7.86 | 9 | −5.22 | 309 |
| −1.87 | −4.70 | 10 (highest) | −3.55 | 309 |

#### Table 5.3. Tenths by deprivation

The most deprived tenth of counties had a net rate of −4.5 returns per thousand and the least deprived tenth −2.4, with positive point estimates in the second to fourth tenths

Pooled net migrating returns per 1,000 returns filed, by tenth of counties ranked on the Area Deprivation Index (1 = least deprived)

| conf_high | conf_low | decile | estimate | n_units |
| ---: | ---: | --- | ---: | ---: |
| 1.52 | −4.56 | 1 (lowest) | −2.38 | 308 |
| 5.13 | −0.070 | 2 | 2.56 | 308 |
| 4.73 | 0.260 | 3 | 2.08 | 308 |
| 5.01 | −1.43 | 4 | 1.53 | 308 |
| 1.69 | −1.71 | 5 | −0.240 | 308 |
| 2.19 | −1.57 | 6 | 0.010 | 308 |
| 2.30 | −1.13 | 7 | 0.420 | 308 |
| 1.64 | −1.95 | 8 | −0.490 | 308 |
| 0.530 | −2.54 | 9 | −1.13 | 308 |
| −2.29 | −6.65 | 10 (highest) | −4.52 | 308 |

Note: The net migration rate is arriving minus leaving returns per thousand returns filed in the county for tax year 2022. Returns are filers, not people; arrivals include moves from abroad and departures include moves abroad.

Note: The scatter shows 3,086 counties. The correlation counts each county in proportion to its population. Counting each county once gives −0.23.

Note: Tenths hold equal numbers of the counties in the analysis. A bar is the sum of net returns over the counties of a tenth divided by the sum of their returns filed, so large counties count more. The intervals resample states, whose counties resemble each other.

Source: IRS Statistics of Income county migration data, filing years 2022 and 2023, IRS SOI county file, tax year 2022, Opportunity Atlas county outcomes (1978-1983 birth cohorts) and ADI 2023; population-weighted fit, full counts of returns

Data: https://vishalsingh.org/zipcode/report/places-that-beat-the-model/figures/migration_vs_mobility.json

### Table 6. Net income per resident does not fall steadily with county home value, with a population-weighted correlation of −0.30 and +0.18 when each county counts once

Net adjusted gross income per resident in dollars against county home value in dollars (log scale), one dot per county sized by population, with pooled means for tenths of counties

No inline rows: the values of Table 6 are in https://vishalsingh.org/zipcode/data/places-that-beat-the-model/migration_vs_home_value.json.

Note: The county home value is the mean of the ACS 2019-2023 median home values of the county's tracts, weighted by tract population, so it is a mean of medians and not the median of the county's homes. Connecticut tracts are assigned to the legacy counties. The Census Bureau top-codes tract medians, and a tract at the top code enters at that value.

Note: The correlation counts each county in proportion to its population, and its interval includes zero. Counting each county once, it is +0.18, with an interval above zero. Both use 3,084 counties.

Note: Tenths hold equal numbers of counties, and each marker pools the counties of a tenth, so large counties count more. Intervals resample states.

Source: IRS Statistics of Income county migration data, filing years 2022 and 2023, ACS 2019-2023 population and tract median home values; population-weighted, full counts of returns

Data: https://vishalsingh.org/zipcode/report/places-that-beat-the-model/figures/migration_vs_home_value.json

### Table 7. In pooled tenths of counties by deprivation, returns arriving report about the same income as returns leaving, though the median county's arriving returns report 2,514 dollars more

Average adjusted gross income per return of arriving and of leaving returns in dollars, by tenth of counties ranked on the Area Deprivation Index (1 = least deprived), returns filed in 2022 matched with 2023

| conf_high | conf_low | decile | estimate | n_units | series |
| ---: | ---: | ---: | ---: | ---: | --- |
| 101858 | 89963 | 1 | 96045 | 308 | Returns arriving |
| 100399 | 88774 | 1 | 95914 | 308 | Returns leaving |
| 81801 | 71936 | 2 | 77560 | 308 | Returns arriving |
| 81840 | 69588 | 2 | 75074 | 308 | Returns leaving |
| 67604 | 62926 | 3 | 65531 | 308 | Returns arriving |
| 70059 | 62338 | 3 | 66572 | 308 | Returns leaving |
| 61978 | 58963 | 4 | 60616 | 308 | Returns arriving |
| 65788 | 59809 | 4 | 62568 | 308 | Returns leaving |
| 57473 | 53277 | 5 | 55646 | 308 | Returns arriving |
| 60851 | 54038 | 5 | 57756 | 308 | Returns leaving |
| 54966 | 51571 | 6 | 53303 | 308 | Returns arriving |
| 55552 | 51644 | 6 | 53594 | 308 | Returns leaving |
| 52838 | 49110 | 7 | 50810 | 308 | Returns arriving |
| 51851 | 48448 | 7 | 50127 | 308 | Returns leaving |
| 51626 | 47803 | 8 | 49699 | 308 | Returns arriving |
| 49903 | 46304 | 8 | 48257 | 308 | Returns leaving |
| 47507 | 44521 | 9 | 45946 | 308 | Returns arriving |
| 45728 | 43273 | 9 | 44383 | 308 | Returns leaving |
| 42970 | 40300 | 10 | 41459 | 308 | Returns arriving |
| 43325 | 40678 | 10 | 41796 | 308 | Returns leaving |

Note: Adjusted gross income is the amount on the 2023 return of the matched pair of returns. A return is a filer, usually a household, and the averages weight each return equally.

Note: Counties need both directions published, so the 57 map units without a net figure are left out. The median county is the middle county when each counts once.

Note: Pooled averages hide who moves within a county, and pooling (which weights counties by their returns) gives a different answer from counting each county once.

Source: IRS Statistics of Income county migration data, filing years 2022 and 2023, and Area Deprivation Index 2023; pooled over counties with both directions published, full counts of returns

Data: https://vishalsingh.org/zipcode/report/places-that-beat-the-model/figures/arriver_vs_leaver_income.json

## All estimates

Every fact in facts.json (112), sorted by id.

| Fact id | Value | Unit | Estimand | Population | Period | Interval | n | Weights | Source | Note |
| --- | ---: | --- | --- | --- | --- | --- | ---: | --- | --- | --- |
| adi_d1 | −2.4 | net migrating returns per 1,000 returns filed | net migrating returns summed over the tenth's counties, divided by returns filed for tax year 2022 and multiplied by 1,000, bootstrap interval over states | counties in the first tenth by Area Deprivation Index rank | returns filed in 2022 matched with returns filed in 2023 (tax years 2021 and 2022) | 95% CI −4.6 to 1.5 | 308 counties | none (counts of returns) | IRS migration data | Tenths hold equal numbers of counties. |
| adi_d10 | −4.5 | net migrating returns per 1,000 returns filed | net migrating returns summed over the tenth's counties, divided by returns filed for tax year 2022 and multiplied by 1,000, bootstrap interval over states | counties in the tenth tenth by Area Deprivation Index rank | returns filed in 2022 matched with returns filed in 2023 (tax years 2021 and 2022) | 95% CI −6.7 to −2.3 | 308 counties | none (counts of returns) | IRS migration data | Tenths hold equal numbers of counties. |
| adi_d2 | +2.6 | net migrating returns per 1,000 returns filed | net migrating returns summed over the tenth's counties, divided by returns filed for tax year 2022 and multiplied by 1,000, bootstrap interval over states | counties in the second tenth by Area Deprivation Index rank | returns filed in 2022 matched with returns filed in 2023 (tax years 2021 and 2022) | 95% CI −0.1 to 5.1 | 308 counties | none (counts of returns) | IRS migration data | Tenths hold equal numbers of counties. |
| agi_moved | 71.6 | billions of dollars of adjusted gross income | sum of net adjusted gross income (arriving minus leaving) over the counties that gained | counties and county equivalents with both directions of the IRS flow published, gaining counties only | returns filed in 2022 matched with returns filed in 2023 (tax years 2021 and 2022) |   |   | none (full counts of returns) | IRS migration data | Dollars are adjusted gross income on the 2023 return, converted from thousands. |
| agi_mw | −172 | dollars of net adjusted gross income per resident, positive for a gain | net adjusted gross income summed over the region's counties divided by ACS population, bootstrap interval over states | residents of the Midwest Census region | returns filed in 2022 matched with returns filed in 2023 (tax years 2021 and 2022) | 95% CI −296 to −60 | 1,030 counties | ACS 2019-2023 county population | IRS migration data |   |
| agi_ne | −332 | dollars of net adjusted gross income per resident, positive for a gain | net adjusted gross income summed over the region's counties divided by ACS population, bootstrap interval over states | residents of the Northeast Census region | returns filed in 2022 matched with returns filed in 2023 (tax years 2021 and 2022) | 95% CI −475 to −54 | 217 counties | ACS 2019-2023 county population | IRS migration data |   |
| agi_s | 270 | dollars of net adjusted gross income per resident, positive for a gain | net adjusted gross income summed over the region's counties divided by ACS population, bootstrap interval over states | residents of the South Census region | returns filed in 2022 matched with returns filed in 2023 (tax years 2021 and 2022) | 95% CI 22 to 528 | 1,411 counties | ACS 2019-2023 county population | IRS migration data |   |
| big_bot_south | 8 | counties | how many lie in the South Census region (the Census Bureau's definition, which includes the District of Columbia) | the counties of that size with the smallest residuals | children born 1978 to 1983, household income measured in 2014 to 2015; predictors ACS 2019 to 2023 |   | 10 counties | none (full count) | Opportunity Atlas |   |
| big_k | 10 | counties | number of counties taken from each end of the ranking by residual | counties with at least 8 tracts in the model and at least 250,000 residents | children born 1978 to 1983, household income measured in 2014 to 2015; predictors ACS 2019 to 2023 |   |   | none (inclusion rule) | Opportunity Atlas |   |
| big_place | 250,000 | residents | county size above which counties are named (inclusion rule) | counties | ACS 2019 to 2023 |   |   | none (inclusion rule) | American Community Survey |   |
| big_place_mig | 50,000 | residents | county size above which counties are named in the migration text (inclusion rule) | counties | ACS 2019 to 2023 |   |   | none (inclusion rule) | American Community Survey |   |
| big_top_corridor | 9 | counties | how many lie in New York, New Jersey, Pennsylvania, Massachusetts or Virginia | the counties of that size with the largest residuals | children born 1978 to 1983, household income measured in 2014 to 2015; predictors ACS 2019 to 2023 |   | 10 counties | none (full count) | Opportunity Atlas |   |
| bot_v | 2,488 | dollars of net adjusted gross income lost per resident | largest net loss per resident (leaving minus arriving adjusted gross income over ACS population) among those counties: San Francisco, CA | counties with at least 50,000 residents and a net figure | returns filed in 2022 matched with returns filed in 2023 (tax years 2021 and 2022) |   | 998 counties | none (a single county) | IRS migration data |   |
| dens_hi | 9,090 | residents per square mile of land | population-weighted mean over the tracts of the group, county-bootstrap interval; the groups are fixed before the draws | residents of the tracts in the top tenth of counties by mean residual | ACS 2019 to 2023 population over 2020 census land area | 95% CI 2,838 to 18,233 | 4,512 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| dens_lo | 2,286 | residents per square mile of land | population-weighted mean over the tracts of the group, county-bootstrap interval; the groups are fixed before the draws | residents of the tracts in the bottom tenth of counties by mean residual | ACS 2019 to 2023 population over 2020 census land area | 95% CI 1,149 to 3,952 | 4,207 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| diff_dens | +6,803 | residents per square mile of land | top-tenth mean minus bottom-tenth mean, population-weighted over tracts, interval from a county bootstrap with the same draws for both groups; the groups are fixed before the draws | residents of the tracts in the top and bottom tenths of counties by mean residual | ACS 2019 to 2023 population over 2020 census land area | 95% CI 97 to 15,794 | 8,719 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| diff_hisp | +5.2 | percentage points of residents who are Hispanic | top-tenth mean minus bottom-tenth mean, population-weighted over tracts, interval from a county bootstrap with the same draws for both groups; the groups are fixed before the draws | residents of the tracts in the top and bottom tenths of counties by mean residual | ACS 2019 to 2023 | 95% CI 0.7 to 9.3 | 8,719 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| diff_poverty | −1.9 | percentage points of residents below the poverty line | top-tenth mean minus bottom-tenth mean, population-weighted over tracts, interval from a county bootstrap with the same draws for both groups; the groups are fixed before the draws | residents of the tracts in the top and bottom tenths of counties by mean residual | ACS 2019 to 2023 | 95% CI −3.4 to −0.2 | 8,719 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| diff_pre1960 | +20.1 | percentage points of housing units built before 1960 | top-tenth mean minus bottom-tenth mean, population-weighted over tracts, interval from a county bootstrap with the same draws for both groups; the groups are fixed before the draws | residents of the tracts in the top and bottom tenths of counties by mean residual | EJScreen 2024 release (ACS 2018 to 2022) | 95% CI 9.8 to 29.2 | 8,719 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| diff_smoke | −2.4 | percentage points of adults who currently smoke (crude prevalence) | top-tenth mean minus bottom-tenth mean, population-weighted over tracts, interval from a county bootstrap with the same draws for both groups; the groups are fixed before the draws | residents of the tracts in the top and bottom tenths of counties by mean residual | PLACES 2024 release (BRFSS 2022) | 95% CI −3.5 to −0.9 | 8,719 tracts | ACS 2019-2023 tract population | CDC PLACES |   |
| floor_high | 15 | tracts | another minimum number of tracts in the model tried for a county residual (sensitivity) | counties | children born 1978 to 1983, household income measured in 2014 to 2015; predictors ACS 2019 to 2023 |   |   | none (inclusion rule) | Opportunity Atlas |   |
| floor_low | 5 | tracts | another minimum number of tracts in the model tried for a county residual (sensitivity) | counties | children born 1978 to 1983, household income measured in 2014 to 2015; predictors ACS 2019 to 2023 |   |   | none (inclusion rule) | Opportunity Atlas |   |
| gap_d10 | −337 | dollars of adjusted gross income per return, arriving minus leaving | summed adjusted gross income of arriving returns over summed arriving returns, minus the same for leaving returns, bootstrap interval over states (shared draws) | the most deprived tenth of counties with both directions published | returns filed in 2022 matched with returns filed in 2023 (tax years 2021 and 2022); ADI 2023 release | 95% CI −1126 to 559 | 308 counties | none (counts of returns) | IRS migration data |   |
| gap_max | 2,486 | dollars of adjusted gross income per return, arriving minus leaving | highest tenth value of the arriving minus leaving average adjusted gross income per return | the tenth of counties with the highest gap | returns filed in 2022 matched with returns filed in 2023 (tax years 2021 and 2022) |   |   | none (counts of returns) | IRS migration data |   |
| gap_min | −2,110 | dollars of adjusted gross income per return, arriving minus leaving | lowest tenth value of the arriving minus leaving average adjusted gross income per return | the tenth of counties with the lowest gap | returns filed in 2022 matched with returns filed in 2023 (tax years 2021 and 2022) |   |   | none (counts of returns) | IRS migration data |   |
| gap_n_zero | 9 | tenths of counties | tenths whose 95% interval for the arriving minus leaving gap includes zero | tenths of counties by Area Deprivation Index rank | returns filed in 2022 matched with returns filed in 2023 (tax years 2021 and 2022) |   |   | none (full count) | IRS migration data |   |
| hv_bot | −109 | dollars of net adjusted gross income per resident, negative for a loss | net adjusted gross income summed over the group's counties divided by ACS population, bootstrap interval over states | residents of the bottom tenth of counties by home value | returns filed in 2022 matched with returns filed in 2023 (tax years 2021 and 2022); county home values from ACS 2019 to 2023 tract median home values | 95% CI −142 to −70 | 308 counties | ACS 2019-2023 county population | IRS migration data | Tenths hold equal numbers of counties. |
| hv_n | 3,084 | counties | count of counties in the scatter | counties with a home value and a net figure | returns filed in 2022 matched with returns filed in 2023 (tax years 2021 and 2022); county home values from ACS 2019 to 2023 tract median home values |   |   | none (full count) | IRS migration data |   |
| hv_ninth | 103 | dollars of net adjusted gross income per resident, negative for a loss | net adjusted gross income summed over the group's counties divided by ACS population, bootstrap interval over states | residents of the ninth tenth of counties by home value | returns filed in 2022 matched with returns filed in 2023 (tax years 2021 and 2022); county home values from ACS 2019 to 2023 tract median home values | 95% CI −129 to 382 | 308 counties | ACS 2019-2023 county population | IRS migration data | Tenths hold equal numbers of counties. |
| hv_top | −103 | dollars of net adjusted gross income per resident, negative for a loss | net adjusted gross income summed over the group's counties divided by ACS population, bootstrap interval over states | residents of the top tenth of counties by home value | returns filed in 2022 matched with returns filed in 2023 (tax years 2021 and 2022); county home values from ACS 2019 to 2023 tract median home values | 95% CI −298 to 238 | 309 counties | ACS 2019-2023 county population | IRS migration data | Tenths hold equal numbers of counties. |
| kfr_d1 | +0.9 | net migrating returns per 1,000 returns filed | net migrating returns summed over the tenth's counties, divided by returns filed for tax year 2022 and multiplied by 1,000, bootstrap interval over states | counties in the first tenth by upward-mobility rank | returns filed in 2022 matched with returns filed in 2023 (tax years 2021 and 2022) | 95% CI −2.7 to 3.6 | 308 counties | none (counts of returns) | IRS migration data | Tenths hold equal numbers of counties. |
| kfr_d10 | −3.6 | net migrating returns per 1,000 returns filed | net migrating returns summed over the tenth's counties, divided by returns filed for tax year 2022 and multiplied by 1,000, bootstrap interval over states | counties in the tenth tenth by upward-mobility rank | returns filed in 2022 matched with returns filed in 2023 (tax years 2021 and 2022) | 95% CI −4.7 to −1.9 | 309 counties | none (counts of returns) | IRS migration data | Tenths hold equal numbers of counties. |
| kfr_d2 | +3.6 | net migrating returns per 1,000 returns filed | net migrating returns summed over the tenth's counties, divided by returns filed for tax year 2022 and multiplied by 1,000, bootstrap interval over states | counties in the second tenth by upward-mobility rank | returns filed in 2022 matched with returns filed in 2023 (tax years 2021 and 2022) | 95% CI −0.7 to 7.3 | 309 counties | none (counts of returns) | IRS migration data | Tenths hold equal numbers of counties. |
| kfr_d9 | −5.2 | net migrating returns per 1,000 returns filed | net migrating returns summed over the tenth's counties, divided by returns filed for tax year 2022 and multiplied by 1,000, bootstrap interval over states | counties in the ninth tenth by upward-mobility rank | returns filed in 2022 matched with returns filed in 2023 (tax years 2021 and 2022) | 95% CI −7.9 to −1.6 | 309 counties | none (counts of returns) | IRS migration data | Tenths hold equal numbers of counties. |
| le_r2 | 44.2% | percent of the variance of tract life expectancy | R-squared of the same five-measure, population-weighted fit with tract life expectancy at birth as the outcome, county-bootstrap interval | non-institutional 2020 tracts in the mobility model (a mobility estimate and all five predictors) that also have a life expectancy estimate | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); predictors ACS 2019 to 2023 | 95% CI 42.0% to 46.5% | 77,746 tracts | ACS 2019-2023 tract population | USALEEP life expectancy |   |
| le_resid_r | +0.39 | correlation coefficient | correlation across counties of the county mean residual of mobility with the county mean residual of life expectancy, weighted by county population, bootstrap interval over states | counties with at least 8 tracts in each of the two models | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); predictors ACS 2019 to 2023; mobility for children born 1978 to 1983, household income measured in 2014 to 2015; predictors ACS 2019 to 2023 | 95% CI 0.27 to 0.51 | 1,529 counties | ACS 2019-2023 county population | USALEEP life expectancy |   |
| map_clamp | 2,000 | dollars of net adjusted gross income per resident | net income per resident at the ends of the map key (display range) | counties on the map | returns filed in 2022 matched with returns filed in 2023 (tax years 2021 and 2022) |   |   | none (display range) | IRS migration data |   |
| med_gap | 2,514 | dollars of adjusted gross income per return, arriving minus leaving | median across counties of the county's average adjusted gross income per return (arriving minus leaving), bootstrap interval over states | counties with both directions of the IRS flow published | returns filed in 2022 matched with returns filed in 2023 (tax years 2021 and 2022) | 95% CI 1,466 to 3,433 | 3,086 counties | none (each county counts once) | IRS migration data | Adjusted gross income is on the 2023 return. |
| med_in | 54,902 | dollars of adjusted gross income per return, arriving returns | median across counties of the county's average adjusted gross income per return (arriving returns), bootstrap interval over states | counties with both directions of the IRS flow published | returns filed in 2022 matched with returns filed in 2023 (tax years 2021 and 2022) | 95% CI 52,722 to 57,194 | 3,086 counties | none (each county counts once) | IRS migration data | Adjusted gross income is on the 2023 return. |
| med_out | 52,444 | dollars of adjusted gross income per return, leaving returns | median across counties of the county's average adjusted gross income per return (leaving returns), bootstrap interval over states | counties with both directions of the IRS flow published | returns filed in 2022 matched with returns filed in 2023 (tax years 2021 and 2022) | 95% CI 50,430 to 54,356 | 3,086 counties | none (each county counts once) | IRS migration data | Adjusted gross income is on the 2023 return. |
| midwest_all_share | 28% | percent of counties | share of all counties with a residual that are in the Midwest Census region, under a floor of 8 tracts | counties with at least 8 tracts in the model | children born 1978 to 1983, household income measured in 2014 to 2015; predictors ACS 2019 to 2023 |   | 1,576 counties | none (each county counts once) | Opportunity Atlas |   |
| midwest_all_share_high | 23% | percent of counties | share of all counties with a residual that are in the Midwest Census region, under a floor of 15 tracts | counties with at least 15 tracts in the model | children born 1978 to 1983, household income measured in 2014 to 2015; predictors ACS 2019 to 2023 |   | 944 counties | none (each county counts once) | Opportunity Atlas |   |
| midwest_top_share | 54% | percent of counties | share of those counties in the Midwest Census region, under a floor of 8 tracts | the top tenth of counties with a residual, ranked by mean residual | children born 1978 to 1983, household income measured in 2014 to 2015; predictors ACS 2019 to 2023 |   | 158 counties | none (each county counts once) | Opportunity Atlas |   |
| midwest_top_share_high | 37% | percent of counties | share of those counties in the Midwest Census region, under a floor of 15 tracts | the top tenth of counties with a residual, ranked by mean residual | children born 1978 to 1983, household income measured in 2014 to 2015; predictors ACS 2019 to 2023 |   | 95 counties | none (each county counts once) | Opportunity Atlas |   |
| min_pop_tract | 200 | residents | a tract needs more than this many residents to enter the model (inclusion rule) | 2020 census tracts | children born 1978 to 1983, household income measured in 2014 to 2015; predictors ACS 2019 to 2023 |   |   | none (inclusion rule) | Opportunity Atlas |   |
| min_tracts_resid | 8 | tracts | minimum number of tracts in the model for a county to have a residual (inclusion rule) | counties | children born 1978 to 1983, household income measured in 2014 to 2015; predictors ACS 2019 to 2023 |   |   | none (inclusion rule) | Opportunity Atlas |   |
| mob_slope | −0.39 | net migrating returns per 1,000 returns filed, per percentile point of upward mobility | slope of the net migration rate on the county mobility rank, weighted by county population, bootstrap interval over states | counties with a net migration rate and a mobility estimate | returns filed in 2022 matched with returns filed in 2023 (tax years 2021 and 2022); mobility for children born 1978 to 1983, household income measured in 2014 to 2015; predictors ACS 2019 to 2023 | 95% CI −0.68 to −0.05 | 3,086 counties | ACS 2019-2023 county population | Opportunity Atlas |   |
| model_counties | 3,136 | counties | count of counties | counties and county equivalents with a tract in the model | children born 1978 to 1983, household income measured in 2014 to 2015; predictors ACS 2019 to 2023 |   |   | none (full count) | Opportunity Atlas | Connecticut counts as its eight legacy counties. |
| model_n | 82,791 | tracts | count of tracts in the model | 2020 census tracts with more than 200 residents, a mobility estimate and all five predictors | children born 1978 to 1983, household income measured in 2014 to 2015; predictors ACS 2019 to 2023 |   |   | none (full count) | Opportunity Atlas |   |
| model_unexplained | 56.0% | percent of the variance of tract upward mobility | one minus the R-squared of the five-measure model, county-bootstrap interval | 2020 census tracts with more than 200 residents, a mobility estimate and all five predictors | children born 1978 to 1983, household income measured in 2014 to 2015; predictors ACS 2019 to 2023 | 95% CI 53.9% to 58.2% | 82,791 tracts | ACS 2019-2023 tract population | Opportunity Atlas |   |
| mov_gain | 1,719 | counties | counties where arriving adjusted gross income exceeds leaving | counties and county equivalents with both directions of the IRS flow published | returns filed in 2022 matched with returns filed in 2023 (tax years 2021 and 2022) |   |   | none (full count) | IRS migration data |   |
| mov_lose | 1,367 | counties | counties where leaving adjusted gross income exceeds arriving | counties and county equivalents with both directions of the IRS flow published | returns filed in 2022 matched with returns filed in 2023 (tax years 2021 and 2022) |   |   | none (full count) | IRS migration data |   |
| mov_n | 3,086 | counties | count of counties with a net figure | counties and county equivalents with both directions of the IRS flow published | returns filed in 2022 matched with returns filed in 2023 (tax years 2021 and 2022) |   |   | none (full count) | IRS migration data |   |
| mover_share | 5.6% | percent of returns filed for tax year 2022 | returns that arrived in a county from another county or from abroad, as a share of all returns filed | individual income tax returns filed for tax year 2022 | returns filed in 2022 matched with returns filed in 2023 (tax years 2021 and 2022) |   |   | none (full counts of returns) | IRS migration data | Arriving returns are summed over the counties with an arrivals figure; filed returns are the IRS SOI county totals for tax year 2022. |
| n_at_floor_high | 944 | counties | count of counties with a residual under this floor | counties with at least 15 tracts in the model | children born 1978 to 1983, household income measured in 2014 to 2015; predictors ACS 2019 to 2023 |   |   | none (full count) | Opportunity Atlas |   |
| n_at_floor_low | 2,140 | counties | count of counties with a residual under this floor | counties with at least 5 tracts in the model | children born 1978 to 1983, household income measured in 2014 to 2015; predictors ACS 2019 to 2023 |   |   | none (full count) | Opportunity Atlas |   |
| n_beyond | 55 | counties | counties whose net income per resident lies beyond the ends of the map key | counties and county equivalents with both directions of the IRS flow published | returns filed in 2022 matched with returns filed in 2023 (tax years 2021 and 2022) |   |   | none (full count) | IRS migration data |   |
| n_big | 281 | counties | count of those counties | counties with at least 8 tracts in the model and at least 250,000 residents | children born 1978 to 1983, household income measured in 2014 to 2015; predictors ACS 2019 to 2023 |   |   | none (full count) | Opportunity Atlas |   |
| n_nodata_mig | 57 | counties | map units without a net figure: no IRS row, one direction withheld, or no population | counties and county equivalents on the map | returns filed in 2022 matched with returns filed in 2023 (tax years 2021 and 2022) |   |   | none (full count) | IRS migration data |   |
| n_out_hv | 8 | counties | counties outside the plotted range of the home-value scatter | counties with a home value and a net figure | returns filed in 2022 matched with returns filed in 2023 (tax years 2021 and 2022) |   |   | none (full count) | IRS migration data |   |
| n_out_mob | 21 | counties | counties outside the plotted range of the mobility scatter | counties with a net migration rate and a mobility estimate | returns filed in 2022 matched with returns filed in 2023 (tax years 2021 and 2022) |   |   | none (full count) | IRS migration data |   |
| n_outside_scatter | 132 | tracts | tracts outside the plotted range of the prediction scatter | 2020 census tracts with more than 200 residents, a mobility estimate and all five predictors | children born 1978 to 1983, household income measured in 2014 to 2015; predictors ACS 2019 to 2023 |   |   | none (full count) | Opportunity Atlas |   |
| n_resid_beyond | 148 | counties | counties whose mean residual lies beyond the ends of the map key (display range) | counties with at least 8 tracts in the model | children born 1978 to 1983, household income measured in 2014 to 2015; predictors ACS 2019 to 2023 |   |   | none (full count) | Opportunity Atlas |   |
| n_resid_counties | 1,576 | counties | count of counties with a residual | counties with at least 8 tracts in the model | children born 1978 to 1983, household income measured in 2014 to 2015; predictors ACS 2019 to 2023 |   |   | none (full count) | Opportunity Atlas |   |
| n_resid_nodata | 1,567 | counties | map units without a county residual, counting those with fewer tracts than the floor and those with no tract in the model | counties and county equivalents on the map | children born 1978 to 1983, household income measured in 2014 to 2015; predictors ACS 2019 to 2023 |   |   | none (full count) | Opportunity Atlas |   |
| n_sample | 4,000 | tracts | size of a random sample of tracts drawn with a fixed seed, inside the plotted range | 2020 census tracts with more than 200 residents, a mobility estimate and all five predictors | children born 1978 to 1983, household income measured in 2014 to 2015; predictors ACS 2019 to 2023 |   |   | none (sample size) | Opportunity Atlas |   |
| northeast_all_share | 13% | percent of counties | share of all counties with a residual that are in the Northeast Census region, under a floor of 8 tracts | counties with at least 8 tracts in the model | children born 1978 to 1983, household income measured in 2014 to 2015; predictors ACS 2019 to 2023 |   | 1,576 counties | none (each county counts once) | Opportunity Atlas |   |
| northeast_all_share_high | 17% | percent of counties | share of all counties with a residual that are in the Northeast Census region, under a floor of 15 tracts | counties with at least 15 tracts in the model | children born 1978 to 1983, household income measured in 2014 to 2015; predictors ACS 2019 to 2023 |   | 944 counties | none (each county counts once) | Opportunity Atlas |   |
| northeast_top_share | 14% | percent of counties | share of those counties in the Northeast Census region, under a floor of 8 tracts | the top tenth of counties with a residual, ranked by mean residual | children born 1978 to 1983, household income measured in 2014 to 2015; predictors ACS 2019 to 2023 |   | 158 counties | none (each county counts once) | Opportunity Atlas |   |
| northeast_top_share_high | 28% | percent of counties | share of those counties in the Northeast Census region, under a floor of 15 tracts | the top tenth of counties with a residual, ranked by mean residual | children born 1978 to 1983, household income measured in 2014 to 2015; predictors ACS 2019 to 2023 |   | 95 counties | none (each county counts once) | Opportunity Atlas |   |
| p95_at_floor_high | +3.9 | percentile points of national household income rank in adulthood, actual minus predicted | ninety-fifth percentile of the county mean residual under this floor, each county counted once | counties with at least 15 tracts in the model | children born 1978 to 1983, household income measured in 2014 to 2015; predictors ACS 2019 to 2023 |   | 944 counties | none (each county counts once) | Opportunity Atlas |   |
| p95_at_floor_low | +6.6 | percentile points of national household income rank in adulthood, actual minus predicted | ninety-fifth percentile of the county mean residual under this floor, each county counted once | counties with at least 5 tracts in the model | children born 1978 to 1983, household income measured in 2014 to 2015; predictors ACS 2019 to 2023 |   | 2,140 counties | none (each county counts once) | Opportunity Atlas |   |
| pop_high | 18.1 million | residents | sum of ACS tract population over the tracts of the group | residents of the tracts in the top tenth of counties by mean residual | ACS 2019 to 2023 |   |   | none (full count) | American Community Survey |   |
| pop_low | 17.2 million | residents | sum of ACS tract population over the tracts of the group | residents of the tracts in the bottom tenth of counties by mean residual | ACS 2019 to 2023 |   |   | none (full count) | American Community Survey |   |
| pre1960_hi | 38% | percent of housing units built before 1960 | population-weighted mean over the tracts of the group, county-bootstrap interval; the groups are fixed before the draws | residents of the tracts in the top tenth of counties by mean residual | EJScreen 2024 release (ACS 2018 to 2022) | 95% CI 28% to 47% | 4,512 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| pre1960_lo | 18% | percent of housing units built before 1960 | population-weighted mean over the tracts of the group, county-bootstrap interval; the groups are fixed before the draws | residents of the tracts in the bottom tenth of counties by mean residual | EJScreen 2024 release (ACS 2018 to 2022) | 95% CI 15% to 22% | 4,207 tracts | ACS 2019-2023 tract population | EPA EJScreen |   |
| r2 | 44.0% | percent of the variance of tract upward mobility | R-squared of the population-weighted least-squares fit of tract mobility on poverty, college share, single-parent share, log median income and Black share, county-bootstrap interval | 2020 census tracts with more than 200 residents, a mobility estimate and all five predictors | children born 1978 to 1983, household income measured in 2014 to 2015; predictors ACS 2019 to 2023 | 95% CI 41.8% to 46.1% | 82,791 tracts | ACS 2019-2023 tract population | Opportunity Atlas |   |
| r_adi | +0.19 | correlation coefficient | correlation across counties of the net migration rate with the county Area Deprivation Index rank, weighted by county population, bootstrap interval over states | counties with a net migration rate and a Area Deprivation Index rank | returns filed in 2022 matched with returns filed in 2023 (tax years 2021 and 2022); ADI 2023 release | 95% CI −0.05 to 0.34 | 3,080 counties | ACS 2019-2023 county population | Area Deprivation Index |   |
| r_adi_unw | −0.19 | correlation coefficient | correlation across counties of the net migration rate with the county Area Deprivation Index rank, each county counted once, bootstrap interval over states | counties with a net migration rate and a Area Deprivation Index rank | returns filed in 2022 matched with returns filed in 2023 (tax years 2021 and 2022); ADI 2023 release | 95% CI −0.28 to −0.09 | 3,080 counties | none (each county counts once) | Area Deprivation Index |   |
| r_gap_adi | +0.02 | correlation coefficient | correlation across counties of the arriving minus leaving average adjusted gross income per return with the Area Deprivation Index rank, weighted by county population, bootstrap interval over states | counties with both directions of the IRS flow published and an Area Deprivation Index rank | returns filed in 2022 matched with returns filed in 2023 (tax years 2021 and 2022); ADI 2023 release | 95% CI −0.13 to 0.13 | 3,080 counties | ACS 2019-2023 county population | IRS migration data |   |
| r_gap_adi_unw | −0.21 | correlation coefficient | correlation across counties of the arriving minus leaving average adjusted gross income per return with the Area Deprivation Index rank, each county counted once, bootstrap interval over states | counties with both directions of the IRS flow published and an Area Deprivation Index rank | returns filed in 2022 matched with returns filed in 2023 (tax years 2021 and 2022); ADI 2023 release | 95% CI −0.29 to −0.14 | 3,080 counties | none (each county counts once) | IRS migration data |   |
| r_hv | −0.30 | correlation coefficient | correlation across counties of net adjusted gross income per resident with the county home value, weighted by county population, bootstrap interval over states | counties with a home value and a net figure | returns filed in 2022 matched with returns filed in 2023 (tax years 2021 and 2022); county home values from ACS 2019 to 2023 tract median home values | 95% CI −0.47 to 0.02 | 3,084 counties | ACS 2019-2023 county population | IRS migration data | The county home value is the mean of the ACS 2019-2023 median home values of the county's tracts, weighted by ACS 2019-2023 tract population; Connecticut tracts are assigned to the legacy counties through the tract crosswalk. A tract at the Census top code of 2,000,001 dollars enters at that value. |
| r_hv_unw | +0.18 | correlation coefficient | correlation across counties of net adjusted gross income per resident with the county home value, each county counted once, bootstrap interval over states | counties with a home value and a net figure | returns filed in 2022 matched with returns filed in 2023 (tax years 2021 and 2022); county home values from ACS 2019 to 2023 tract median home values | 95% CI 0.04 to 0.33 | 3,084 counties | none (each county counts once) | IRS migration data |   |
| r_kfr | −0.19 | correlation coefficient | correlation across counties of the net migration rate with the county upward-mobility rank, weighted by county population, bootstrap interval over states | counties with a net migration rate and a upward-mobility rank | returns filed in 2022 matched with returns filed in 2023 (tax years 2021 and 2022); mobility for children born 1978 to 1983, household income measured in 2014 to 2015; predictors ACS 2019 to 2023 | 95% CI −0.31 to −0.02 | 3,086 counties | ACS 2019-2023 county population | Opportunity Atlas |   |
| r_kfr_unw | −0.23 | correlation coefficient | correlation across counties of the net migration rate with the county upward-mobility rank, each county counted once, bootstrap interval over states | counties with a net migration rate and a upward-mobility rank | returns filed in 2022 matched with returns filed in 2023 (tax years 2021 and 2022); mobility for children born 1978 to 1983, household income measured in 2014 to 2015; predictors ACS 2019 to 2023 | 95% CI −0.33 to −0.10 | 3,086 counties | none (each county counts once) | Opportunity Atlas |   |
| r_le | −0.25 | correlation coefficient | correlation across counties of the net migration rate with the county life expectancy, weighted by county population, bootstrap interval over states | counties with a net migration rate and a life expectancy | returns filed in 2022 matched with returns filed in 2023 (tax years 2021 and 2022); life expectancy for deaths in 2010 to 2015 | 95% CI −0.37 to −0.09 | 3,075 counties | ACS 2019-2023 county population | USALEEP life expectancy |   |
| r_le_unw | −0.13 | correlation coefficient | correlation across counties of the net migration rate with the county life expectancy, each county counted once, bootstrap interval over states | counties with a net migration rate and a life expectancy | returns filed in 2022 matched with returns filed in 2023 (tax years 2021 and 2022); life expectancy for deaths in 2010 to 2015 | 95% CI −0.23 to −0.01 | 3,075 counties | none (each county counts once) | USALEEP life expectancy |   |
| r_resid_rate | −0.44 | correlation coefficient | correlation across counties of the county mean residual of mobility with the net migration rate, weighted by county population, bootstrap interval over states | counties with a residual and a net migration rate | returns filed in 2022 matched with returns filed in 2023 (tax years 2021 and 2022); mobility for children born 1978 to 1983, household income measured in 2014 to 2015; predictors ACS 2019 to 2023 | 95% CI −0.57 to −0.28 | 1,576 counties | ACS 2019-2023 county population | IRS migration data |   |
| r_resid_rate_unw | −0.39 | correlation coefficient | correlation across counties of the county mean residual of mobility with the net migration rate, each county counted once, bootstrap interval over states | counties with a residual and a net migration rate | returns filed in 2022 matched with returns filed in 2023 (tax years 2021 and 2022); mobility for children born 1978 to 1983, household income measured in 2014 to 2015; predictors ACS 2019 to 2023 | 95% CI −0.49 to −0.27 | 1,576 counties | none (each county counts once) | IRS migration data |   |
| resid_bergen | +6.1 | percentile points of national household income rank in adulthood, actual minus predicted | population-weighted mean of the tract residuals in the county; the interval is 1.96 standard errors from the spread of tract residuals within the county | children raised in low-income families (parents at the 25th percentile of the national income distribution) in Bergen, NJ | children born 1978 to 1983, household income measured in 2014 to 2015; predictors ACS 2019 to 2023 | 95% CI 5.5 to 6.7 | 203 tracts | ACS 2019-2023 tract population | Opportunity Atlas |   |
| resid_buncombe | −5.4 | percentile points of national household income rank in adulthood, actual minus predicted | population-weighted mean of the tract residuals in the county; the interval is 1.96 standard errors from the spread of tract residuals within the county | children raised in low-income families (parents at the 25th percentile of the national income distribution) in Buncombe, NC | children born 1978 to 1983, household income measured in 2014 to 2015; predictors ACS 2019 to 2023 | 95% CI −6.1 to −4.6 | 64 tracts | ACS 2019-2023 tract population | Opportunity Atlas |   |
| resid_denver | −5.4 | percentile points of national household income rank in adulthood, actual minus predicted | population-weighted mean of the tract residuals in the county; the interval is 1.96 standard errors from the spread of tract residuals within the county | children raised in low-income families (parents at the 25th percentile of the national income distribution) in Denver, CO | children born 1978 to 1983, household income measured in 2014 to 2015; predictors ACS 2019 to 2023 | 95% CI −6.1 to −4.6 | 175 tracts | ACS 2019-2023 tract population | Opportunity Atlas |   |
| resid_key | 5 | percentile points of national household income rank in adulthood | residual at the ends of the map key (display range) | counties on the map | children born 1978 to 1983, household income measured in 2014 to 2015; predictors ACS 2019 to 2023 |   |   | none (display range) | Opportunity Atlas |   |
| resid_nassau | +6.1 | percentile points of national household income rank in adulthood, actual minus predicted | population-weighted mean of the tract residuals in the county; the interval is 1.96 standard errors from the spread of tract residuals within the county | children raised in low-income families (parents at the 25th percentile of the national income distribution) in Nassau, NY | children born 1978 to 1983, household income measured in 2014 to 2015; predictors ACS 2019 to 2023 | 95% CI 5.4 to 6.8 | 278 tracts | ACS 2019-2023 tract population | Opportunity Atlas |   |
| resid_p5 | −4.8 | percentile points of national household income rank in adulthood, actual minus predicted | fifth percentile of the county mean residual, each county counted once | counties with at least 8 tracts in the model | children born 1978 to 1983, household income measured in 2014 to 2015; predictors ACS 2019 to 2023 |   | 1,576 counties | none (each county counts once) | Opportunity Atlas |   |
| resid_p95 | +5.0 | percentile points of national household income rank in adulthood, actual minus predicted | ninety-fifth percentile of the county mean residual, each county counted once | counties with at least 8 tracts in the model | children born 1978 to 1983, household income measured in 2014 to 2015; predictors ACS 2019 to 2023 |   | 1,576 counties | none (each county counts once) | Opportunity Atlas |   |
| resid_pop_share | 94% | percent of the population of the tracts in the model | share of the model's population living in counties with a residual | 2020 census tracts with more than 200 residents, a mobility estimate and all five predictors | children born 1978 to 1983, household income measured in 2014 to 2015; predictors ACS 2019 to 2023 |   | 1,576 counties | ACS 2019-2023 tract population | Opportunity Atlas |   |
| resid_queens | +5.6 | percentile points of national household income rank in adulthood, actual minus predicted | population-weighted mean of the tract residuals in the county; the interval is 1.96 standard errors from the spread of tract residuals within the county | children raised in low-income families (parents at the 25th percentile of the national income distribution) in Queens, NY | children born 1978 to 1983, household income measured in 2014 to 2015; predictors ACS 2019 to 2023 | 95% CI 5.2 to 6.1 | 673 tracts | ACS 2019-2023 tract population | Opportunity Atlas |   |
| resid_sd | 5.0 | percentile points of national household income rank in adulthood | population-weighted standard deviation of the tract residual (actual minus predicted rank) | 2020 census tracts with more than 200 residents, a mobility estimate and all five predictors | children born 1978 to 1983, household income measured in 2014 to 2015; predictors ACS 2019 to 2023 |   | 82,791 tracts | ACS 2019-2023 tract population | Opportunity Atlas | Full counts of the tracts in the model; no sampling interval is drawn. |
| resid_se_median | 0.7 | percentile points of national household income rank in adulthood | median across counties of the standard error of the county mean residual (spread of tract residuals within the county) | counties with at least 8 tracts in the model | children born 1978 to 1983, household income measured in 2014 to 2015; predictors ACS 2019 to 2023 |   | 1,576 counties | none (each county counts once) | Opportunity Atlas | The standard error treats a county's tracts as independent draws. It leaves out errors shared by all tracts of a county. |
| resid_stjohns | −5.8 | percentile points of national household income rank in adulthood, actual minus predicted | population-weighted mean of the tract residuals in the county; the interval is 1.96 standard errors from the spread of tract residuals within the county | children raised in low-income families (parents at the 25th percentile of the national income distribution) in St. Johns, FL | children born 1978 to 1983, household income measured in 2014 to 2015; predictors ACS 2019 to 2023 | 95% CI −7.3 to −4.4 | 49 tracts | ACS 2019-2023 tract population | Opportunity Atlas |   |
| resid_top | +6.2 | percentile points of national household income rank in adulthood, actual minus predicted | population-weighted mean of the tract residuals in the county; the interval is 1.96 standard errors from the spread of tract residuals within the county | children raised in low-income families (parents at the 25th percentile of the national income distribution) in Richmond, NY | children born 1978 to 1983, household income measured in 2014 to 2015; predictors ACS 2019 to 2023 | 95% CI 5.3 to 7.2 | 117 tracts | ACS 2019-2023 tract population | Opportunity Atlas |   |
| scatter_n | 3,086 | counties | count of counties in the scatter | counties with a net migration rate and a mobility estimate | returns filed in 2022 matched with returns filed in 2023 (tax years 2021 and 2022) |   |   | none (full count) | IRS migration data |   |
| smoke_hi | 13.5% | percent of adults who currently smoke | population-weighted mean over the tracts of the group, county-bootstrap interval; the groups are fixed before the draws | residents of the tracts in the top tenth of counties by mean residual | PLACES 2024 release (BRFSS 2022) | 95% CI 12.6% to 14.8% | 4,512 tracts | ACS 2019-2023 tract population | CDC PLACES |   |
| smoke_lo | 15.9% | percent of adults who currently smoke | population-weighted mean over the tracts of the group, county-bootstrap interval; the groups are fixed before the draws | residents of the tracts in the bottom tenth of counties by mean residual | PLACES 2024 release (BRFSS 2022) | 95% CI 15.2% to 16.6% | 4,207 tracts | ACS 2019-2023 tract population | CDC PLACES |   |
| south_all_share | 46% | percent of counties | share of all counties with a residual that are in the South Census region, under a floor of 8 tracts | counties with at least 8 tracts in the model | children born 1978 to 1983, household income measured in 2014 to 2015; predictors ACS 2019 to 2023 |   | 1,576 counties | none (each county counts once) | Opportunity Atlas |   |
| south_bottom_share | 81% | percent of counties | share of those counties in the South Census region, under a floor of 8 tracts | the bottom tenth of counties with a residual, ranked by mean residual | children born 1978 to 1983, household income measured in 2014 to 2015; predictors ACS 2019 to 2023 |   | 157 counties | none (each county counts once) | Opportunity Atlas |   |
| south_bottom_share_high | 83% | percent of counties | share of those counties in the South Census region, under a floor of 15 tracts | the bottom tenth of counties with a residual, ranked by mean residual | children born 1978 to 1983, household income measured in 2014 to 2015; predictors ACS 2019 to 2023 |   | 94 counties | none (each county counts once) | Opportunity Atlas |   |
| split_bottom | −4.4 | percentile points of national household income rank in adulthood, actual minus predicted | mean residual in the second random half of tracts of the bottom tenth of counties ranked on the first half, each county counted once, bootstrap interval over states | counties with at least 8 tracts in the model and at least 3 tracts in each random half | children born 1978 to 1983, household income measured in 2014 to 2015; predictors ACS 2019 to 2023 | 95% CI −5.0 to −3.4 | 1,486 counties | none (each county counts once) | Opportunity Atlas |   |
| split_r | +0.85 | correlation coefficient | correlation across counties between the mean residual of one random half of a county's tracts and the other half, each county counted once, bootstrap interval over states | counties with at least 8 tracts in the model and at least 3 tracts in each random half | children born 1978 to 1983, household income measured in 2014 to 2015; predictors ACS 2019 to 2023 | 95% CI 0.80 to 0.88 | 1,486 counties | none (each county counts once) | Opportunity Atlas |   |
| split_top | +5.0 | percentile points of national household income rank in adulthood, actual minus predicted | mean residual in the second random half of tracts of the top tenth of counties ranked on the first half, each county counted once, bootstrap interval over states | counties with at least 8 tracts in the model and at least 3 tracts in each random half | children born 1978 to 1983, household income measured in 2014 to 2015; predictors ACS 2019 to 2023 | 95% CI 3.7 to 6.1 | 1,486 counties | none (each county counts once) | Opportunity Atlas |   |
| top_v | 5,793 | dollars of net adjusted gross income gained per resident | largest net gain per resident (arriving minus leaving adjusted gross income over ACS population) among those counties: Collier, FL | counties with at least 50,000 residents and a net figure | returns filed in 2022 matched with returns filed in 2023 (tax years 2021 and 2022) |   | 998 counties | none (a single county) | IRS migration data |   |

## Definitions

- Income rank (kfr): The position that adult children reach in the national distribution of household income, from the lowest rank to the highest, averaged over children who grew up in a place with parents at a stated income level. The Opportunity Atlas labels this measure kfr.
- Census tract: A small area drawn by the Census Bureau to stand in for a neighborhood. Boundaries are redrawn after each decennial census.
- Bootstrap interval: A range for an estimate found by recomputing it on many random resamples of the data and reading the range from the results. It shows how much the estimate would vary from sample to sample.
- Life expectancy at birth: The average number of years a newborn would live if death rates at every age stayed as they were in the period measured. It summarizes the death rates of a period and is not a forecast for any child.
- Area Deprivation Index (ADI): A score that ranks neighborhoods by income, education, employment and housing conditions drawn from the American Community Survey. In this book a higher national percentile means a more deprived neighborhood.

## Methods

A population-weighted line across 2020 census tracts predicts the Opportunity Atlas adult income rank of children born 1978 to 1983 from five ACS 2019-2023 measures, and a county's residual is the actual rank minus the prediction, averaged over its tracts and reported only for counties with at least 8 tracts in the model. The migration comparison uses IRS returns matched between filing years 2022 and 2023, so it describes returns that changed county, with moves from abroad and moves abroad in the county totals. County correlations are population-weighted and, where weighting changes the sign, also reported with each county counted once; intervals resample states, and the county home value is a population-weighted mean of ACS 2019-2023 tract median home values.

### The tract model and its sample

A population-weighted least-squares line across 2020 census tracts regresses the Opportunity Atlas mean adult household income rank of children born 1978 to 1983 to parents at the 25th income percentile on five ACS 2019-2023 measures: the poverty rate, the share of adults with a bachelor's degree or more, the single-parent household share, the natural log of median household income and the Black share of residents. Tracts with 200 residents or fewer, tracts missing a variable and Puerto Rico are left out. The weight is the ACS tract population.

The life-expectancy version fits the same five measures to USALEEP life expectancy at birth (deaths in 2010 to 2015, institutional tracts left out) and correlates county mean residuals across counties, weighted by county population, with an interval that resamples states.

Explains Table 1.

### County residuals, floors and groups

The tract residual is the actual rank minus the prediction. A county's residual is the population-weighted mean of its tracts' residuals, reported for counties with at least 8 tracts in the model; a floor of 5 tracts would map 2,140 counties and a floor of 15 would map 944. Its standard error treats the tracts as independent draws within the county.

Connecticut tracts are assigned to its eight legacy counties through the staged tract crosswalk.

Groups are tenths of counties with equal numbers of counties; their traits are population-weighted tract means, with intervals that resample counties and treat the groups as fixed.

Explains Table 2 and Table 3.

### Migration measures and county variables

The IRS matches individual returns filed in 2022 (tax year 2021) with returns filed in 2023 (tax year 2022) on the filer's taxpayer identification number, assigns each return a county from its ZIP+4, and counts a return as a migrant when the county differs; the county totals used here are the IRS rows for all origins and destinations, so moves from abroad are in the arrivals and moves abroad in the departures, and adjusted gross income is the amount on the 2023 return, converted from thousands of dollars. Net is arriving minus leaving, computed from both sides, and a county with one direction withheld has no net figure.

The net migration rate divides net returns by the returns filed for tax year 2022 (IRS SOI county file, summed over the income brackets) and multiplies by 1,000; net income per resident divides by the ACS 2019-2023 population. Connecticut's planning regions are allocated to the legacy counties in proportion to 2020 population.

County mobility is the Atlas county estimate, deprivation the population-weighted mean of tract ranks, life expectancy the population-weighted mean of non-institutional tracts, and home value the population-weighted mean of the ACS 2019-2023 median home values of the county's tracts (Connecticut tracts assigned to the legacy counties through the tract crosswalk; a tract at the Census top code of 2,000,001 dollars enters at that value). Tenths of counties are cut on the counties in the analysis.

Explains Table 4, Table 7, Table 5 and Table 6.

### Weights and uncertainty

The R-squared interval and the interval for every statistic of tracts come from a bootstrap that resamples whole counties.

Correlations and fits across counties are weighted by ACS population and, where weighting changes the sign, also reported with each county counted once; intervals resample states. Bars and markers for tenths pool the counties of a group (summed numerator over summed denominator). Differences between groups use the same resampled counties for both groups.

### Split-half check, replicates and seeds

The split-half check assigns each tract of a county to one of two halves at random (seed 12345), ranks counties on one half and reads the residual off the other, for counties with at least 3 tracts in each half; intervals resample states.

The dots of the prediction scatter are a random sample of tracts drawn with seed 12345. Bootstrap intervals use 2,000 replicates and seed 12345.

## What these data cannot show

None of the residuals measures what a place does for a child. A county residual combines noise in the Atlas estimates, measures the model omits and the gap between the decades of the outcome and of the predictors. The standard error of a county residual leaves out errors that all of a county's tracts share. The floor of 8 tracts excludes 1,567 map units, so the map says nothing about them. The traits in Table 3 describe counties selected on the residual, and their intervals treat the two groups as fixed.

The migration figures cover one pair of filing years. Internal migration in the United States has declined since the 1980s.[6] Arriving returns equal only 5.6% of the returns filed here, so most filers in every county did not change county. The two filings follow the pandemic years, in which demand for housing shifted from the centers of large cities toward their outskirts,[7] and one pair of years may not describe other periods. The unit is the county, so the correlations compare places and say nothing about why any household moved. Several of them change sign with the weighting. Connecticut's flows are published by planning region and are spread over its legacy counties by population, which is approximate. A county's home value is the mean of its tracts' median home values, weighted by population, and a tract median at the Census top code enters at the code. Puerto Rico is left out of the model and the maps.

The two halves of the chapter do not measure the same people. The residual scores the adult outcomes of children raised in the 1980s and 1990s, and the migration data score the movements of filers in 2022 and 2023. Reading one against the other compares where two different populations sit on the same map. The adjusted gross income of a return is the amount on the 2023 return, for calendar 2022, and a move between the two filings can come after part or all of that income was earned.

## References

1. Chetty, Raj; Friedman, John N.; Hendren, Nathaniel; Jones, Maggie R.; Porter, Sonya R. (2018). The Opportunity Atlas: Mapping the Childhood Roots of Social Mobility. NBER Working Paper 25147. https://doi.org/10.3386/w25147
2. Internal Revenue Service, Statistics of Income Division (2025). 2022-2023 Migration Data Users Guide. IRS Statistics of Income, County-to-County Migration Data. https://www.irs.gov/pub/irs-soi/2223inpublicmigdoc.pdf
3. Chetty, Raj; Hendren, Nathaniel (2018). The Impacts of Neighborhoods on Intergenerational Mobility I: Childhood Exposure Effects. The Quarterly Journal of Economics 133(3): 1107-1162. https://doi.org/10.1093/qje/qjy007
4. Chetty, Raj; Hendren, Nathaniel (2018). The Impacts of Neighborhoods on Intergenerational Mobility II: County-Level Estimates. The Quarterly Journal of Economics 133(3): 1163-1228. https://doi.org/10.1093/qje/qjy006
5. Ganong, Peter; Shoag, Daniel (2017). Why Has Regional Income Convergence in the U.S. Declined?. Journal of Urban Economics 102: 76-90. https://doi.org/10.1016/j.jue.2017.07.002
6. Molloy, Raven; Smith, Christopher L.; Wozniak, Abigail (2011). Internal Migration in the United States. Journal of Economic Perspectives 25(3): 173-196. https://doi.org/10.1257/jep.25.3.173
7. Ramani, Arjun; Bloom, Nicholas (2021). The Donut Effect of COVID-19 on Cities. NBER Working Paper 28876. https://doi.org/10.3386/w28876

## Provenance

- Pipeline version: 0.1.0
- Build script: book2/chapters/places-that-beat-the-model/build.py (sha256 35c06d8e38ab77487ea45be6c64943e4dd636a9d30ff0c56546a8f23e61a7408)
- Built on: 2026-10-28
- Site commit: c2b1ff3309cb4a10833f884da91ac0057a66ad0a

Inputs:

| File | Size (bytes) | SHA-256, first 1 MB | SHA-256, last 1 MB |
| --- | ---: | --- | --- |
| book2/chapters/places-that-beat-the-model/migr.py | 10,542 | b4b0bcd6ede6140c0f5d665592529c9edc7f3d96f099dc6be2e00cefece276a3 | b4b0bcd6ede6140c0f5d665592529c9edc7f3d96f099dc6be2e00cefece276a3 |
| book2/chapters/places-that-beat-the-model/resid.py | 18,449 | 41ad684e1a933aabc9410401ed0198bf53711fa3265170d466b43ee9c6626c67 | 41ad684e1a933aabc9410401ed0198bf53711fa3265170d466b43ee9c6626c67 |
| data/analytic/tract_atlas.parquet | 39,659,925 | 29d5e4d943552d54aeb6ba92aadc939d9bd15bdea553c78357dbd516757be627 | 1e10d1ecf722340f23ee2212d4999a90250f85b65fa89b39e32db06d9472e22e |
| data/geo/dim_geo.parquet | 1,180,921 | 29d5e4d943552d54aeb6ba92aadc939d9bd15bdea553c78357dbd516757be627 | e58c8c488df54ab3187f5461f179322afdc49c9bd2315404003fe1272eea32c3 |
| data/geo/tracts_2020.parquet | 619,858,391 | f4f05c8c7a63d1785ec2aad283de3f6fa839b6c437f13a1b96cea31d3344bf0f | 7b3811247c6b3e322954f24a4a2384af52aef6ed8688f8dbdb0407d21ef80a59 |
| data/raw/acs/deprivation_components_2019_2023/tract.parquet | 13,197,793 | 5545cde2cbe96a8b56e468d65388248a241aa6bd300d3c9e325d08d3f86b11da | 5c8ec454d066988b0f1a01aaa6ed34ebfc05d162e01cf9c6bb01f9198cbaff12 |
| data/staged/adi/adi_atlas_tract2020_2023.parquet | 798,078 | 5e409bc212ff871def780fa27b4cf8b616d9da420b2b9f6455517eddcd8c1a95 | 5e409bc212ff871def780fa27b4cf8b616d9da420b2b9f6455517eddcd8c1a95 |
| data/staged/crosswalks/ct_old_county_to_planning_region.parquet | 5,797 | 5fb0d280ff6c9a9f160af7f03934db4682efc3c5fda11ae27b7f614d6ad664bb | 5fb0d280ff6c9a9f160af7f03934db4682efc3c5fda11ae27b7f614d6ad664bb |
| data/staged/crosswalks/ct_tract2020_to_old_county.parquet | 15,842 | 961ef39f31596ef65fca2c67d5f2f63e533c5dc9b49745d3bb6437844e4b831d | 961ef39f31596ef65fca2c67d5f2f63e533c5dc9b49745d3bb6437844e4b831d |
| data/staged/environment/ejscreen_tract2020_2024.parquet | 10,891,402 | 8b728fb8c87df1d639a9652a31e22cc07d51aad31f3390a8ff6e82e5cebb6a86 | ea8df449172c133db5eb83c0c0df93aaaf8b781b373b75a0056fbde1b1d32824 |
| data/staged/health/places_tract2020_latest.parquet | 5,919,159 | af4c3b0d1205b7d84b81676a9529ef73323b2dbeb9d54e235f5efeebfad83177 | a2f19029ee2ef032be87e0d5288d88d489afd52379279833f178baa021df3dbf |
| data/staged/income/irs_soi_county_2022.parquet | 7,590,532 | 827be995a4744d9198cf6da5afa6c099fe5ae7da5e7e6fc6b2eaa475b17e65f1 | 2e341a6e8456554ed604a4d357031fc08815a16f5bb7b740db85b583aa56b5f1 |
| data/staged/life_expectancy/usaleep_tract2020.parquet | 1,316,169 | 9c9c7f103791b75eddba74e68a89804efdf2486b6b66bb4bc02a46ba5bf8bc23 | 27317107be2465d9fd5e83271dea0cd6e915ee419e2802bedc28747d0a41d38f |
| data/staged/migration/irs_migration_county_2022_2023.parquet | 154,824 | 280ed853a1d696b5366ce1d77e98118807e974192631b1fadb3427f485bb606f | 280ed853a1d696b5366ce1d77e98118807e974192631b1fadb3427f485bb606f |
| data/staged/mobility/oa_module1_tract2020.parquet | 16,362,438 | 26cb468a319ad2e727c5eac733b4a47b77a83781ae189e1828e8c40bc0e3ace8 | 5702bad38b81b5ee0b3d01ba9d1ece7d23328985eafba269720a0af5ef6820d4 |
| data/staged/mobility/oa_outcomes_county.parquet | 721,343 | fd7bc55294fbd3f7a646f27083500c1f12415775db7a3607878b522d277f8140 | fd7bc55294fbd3f7a646f27083500c1f12415775db7a3607878b522d277f8140 |
| data/staged/ses/acs5_tract2020_2019_2023.parquet | 7,232,424 | 5fe6f730a35225f14d5d8abfc6b9bce58b52e545f309a7eac74f1dae74d7a31c | 7b4a2475b3852d3f91804716a04566ef8ceadacec2be9fb85f5eb642078efac6 |
| data/staged/svi/svi_tract2020_2022.parquet | 17,740,415 | cb3860d8edc2aaf1f2a02211a58f7a0b80457eb6ad9a37b5ec8abb2d5da9da10 | e371a4a696d5bed05d54ad7c5b6ca2504508805b847ec415c2322be7f66ec2b2 |

Outputs:

| File | SHA-256 |
| --- | --- |
| facts.json | 6cefcc3077a83109695b9e419b4da1f701e776e38b9464bac12e370858ef05c3 |
| figures/arriver_vs_leaver_income.json | 316d5bc9a18af2ec063047f2a42fe2091998aec217d9c20a34b7f4bbc258bd2b |
| figures/migration_vs_home_value.json | e9abb63d7910f2fe12eff67ad4d34e9a37e62032be9882254fee7699a2fb7be3 |
| figures/migration_vs_mobility.json | c946a048fd88b5db74b8484d00b047eed28ea16fd8b73b39f31799faa70afe45 |
| figures/net_agi_map.json | a3ffe8711debbcf5b1be47d24600e3242496bf509037139f88646a15e528254f |
| figures/overperformer_traits.json | 1ba33eb34937056848e2910cb0a981f906304970efce4099abd732cb227e4919 |
| figures/predicted_vs_actual.json | d114178f9728367227a75f7ec100d6f519e261713ad686437b9fc6df6739f124 |
| figures/residual_map.json | 84cad600e2056df37e38053325b1535d07cd9805f5ad04b75c90c34694de9c5c |
| meta.json | b3a3dbc0a0bd1f4f89e96d5ce443d218e0dccbc59bbe5ec40eefa312cd500d20 |
| public/data/places-that-beat-the-model/migration_vs_home_value.json | 046824aaa73f11fa0e55f3f7e9e0cea6a14e4bddd0c5bfb8af24b173a7ca9035 |
| public/data/places-that-beat-the-model/migration_vs_mobility.json | 4b16655633f5e5529d209d9378d707db19972e43a1c68455efc456f16a5bb171 |
| public/data/places-that-beat-the-model/net_agi_map.json | 4dc752d023e5fe72193a1a5122ac2363db3a773477014c87b2a1cc62ac9f044a |
| public/data/places-that-beat-the-model/predicted_vs_actual.json | 99da6c3cb25739ee541d56eda865117fa57be5591d6866b632cc293cff531265 |
| public/data/places-that-beat-the-model/residual_map.json | b53c2022a2e7d130b5822e83ef0b41ed08daf6c6cbc5c5bd630eefb46ee50a2d |
| refs.json | 9bbb72943e4ec3ea678e798893e45b68d9ed2d62c24287cd51fa8b241467bb73 |

## Machine formats

- Report (HTML): https://vishalsingh.org/zipcode/places-that-beat-the-model/report
- Report (Markdown): https://vishalsingh.org/zipcode/places-that-beat-the-model/report.md
- Every estimate (JSON): https://vishalsingh.org/zipcode/report/places-that-beat-the-model/facts.json
- Table 1, figure "predicted_vs_actual" (JSON): https://vishalsingh.org/zipcode/report/places-that-beat-the-model/figures/predicted_vs_actual.json
- Table 2, figure "residual_map" (JSON): https://vishalsingh.org/zipcode/report/places-that-beat-the-model/figures/residual_map.json
- Table 3, figure "overperformer_traits" (JSON): https://vishalsingh.org/zipcode/report/places-that-beat-the-model/figures/overperformer_traits.json
- Table 4, figure "net_agi_map" (JSON): https://vishalsingh.org/zipcode/report/places-that-beat-the-model/figures/net_agi_map.json
- Table 5, figure "migration_vs_mobility" (JSON): https://vishalsingh.org/zipcode/report/places-that-beat-the-model/figures/migration_vs_mobility.json
- Table 6, figure "migration_vs_home_value" (JSON): https://vishalsingh.org/zipcode/report/places-that-beat-the-model/figures/migration_vs_home_value.json
- Table 7, figure "arriver_vs_leaver_income" (JSON): https://vishalsingh.org/zipcode/report/places-that-beat-the-model/figures/arriver_vs_leaver_income.json
- Payload of "predicted_vs_actual" (JSON): https://vishalsingh.org/zipcode/data/places-that-beat-the-model/predicted_vs_actual.json
- Payload of "residual_map" (JSON): https://vishalsingh.org/zipcode/data/places-that-beat-the-model/residual_map.json
- Payload of "net_agi_map" (JSON): https://vishalsingh.org/zipcode/data/places-that-beat-the-model/net_agi_map.json
- Payload of "migration_vs_mobility" (JSON): https://vishalsingh.org/zipcode/data/places-that-beat-the-model/migration_vs_mobility.json
- Payload of "migration_vs_home_value" (JSON): https://vishalsingh.org/zipcode/data/places-that-beat-the-model/migration_vs_home_value.json
- Index of every report: https://vishalsingh.org/zipcode/llms.txt; all reports in one file: https://vishalsingh.org/zipcode/llms-full.txt

---

# What moving can change: report

Chapter 9.6 of ZIP Code Destiny, by Vishal Singh (NYU Stern). Generated from the chapter's data files; the reading version is at https://vishalsingh.org/zipcode/what-moving-can-change. Table n of this report is figure n of the chapter.

## Identification

- Title: What moving can change
- Label: Offers, moves and policies
- Chapter: 9.6 (Part IX, Reading the patterns; Reading the patterns)
- Question: What do randomized experiments and other designs with a stated identification strategy find about moving families or changing places, and how does that evidence differ from the comparisons in this book?
- Dek: Randomized offers of housing vouchers, studies of children whose families moved at different ages, and a trial of help with the housing search find changes for particular families under particular conditions, from 15% to 53% of families moving to high-mobility areas in one trial, and none of them measures what the gradients in this book would become if places changed.
- Grain: individuals
- Status: draft
- Published: 2026-10-28
- Updated: not revised
- Data sources:
  - Published estimates from cited studies. Estimates reported in the cited studies, most of them copied from the chapters that first report them; years: as reported in each study; weights: as reported in each study
  - Opportunity Atlas. The share of variation across neighborhoods attributed to causal effects, as reported by the Opportunity Atlas authors and in Race, income and mobility; years: 1978-1983 birth cohorts (2018 release); weights: as reported
- Canonical URL: https://vishalsingh.org/zipcode/what-moving-can-change
- Report: https://vishalsingh.org/zipcode/what-moving-can-change/report
- Markdown: https://vishalsingh.org/zipcode/what-moving-can-change/report.md
- BibTeX key: singh2026whatmovingcanchange

```bibtex
@misc{singh2026whatmovingcanchange,
  author = {Singh, Vishal},
  title = {What moving can change},
  year = {2026},
  url = {https://vishalsingh.org/zipcode/what-moving-can-change},
  note = {In ZIP Code Destiny. Published 28 October 2026.}
}
```

## Principal estimates

The facts the dek and the opening prose cite, in order of appearance.

| Estimand | Population | Period | Value | Interval | n | Weights | Source | Fact id |
| --- | --- | --- | ---: | --- | ---: | --- | --- | --- |
| Bergman, Chetty, DeLuca, Hendren, Katz and Palmer (2024), American Economic Review 114(5): randomized trial among families with children issued housing vouchers in Seattle and King County; the treatment combined information, short-term financial assistance and customized search assistance; share of families who moved to a high-upward-mobility area, copied from the abstract and not computed in this build (control group) | families with children issued housing vouchers in Seattle and King County, control group of the Creating Moves to Opportunity trial | moves after random assignment, as reported in the paper | 15% |   |   | none (reported estimate) | Published estimates from cited studies | cmto_control |
| Bergman, Chetty, DeLuca, Hendren, Katz and Palmer (2024), American Economic Review 114(5): randomized trial among families with children issued housing vouchers in Seattle and King County; the treatment combined information, short-term financial assistance and customized search assistance; share of families who moved to a high-upward-mobility area, copied from the abstract and not computed in this build (treatment group) | families with children issued housing vouchers in Seattle and King County, treatment group of the Creating Moves to Opportunity trial | moves after random assignment, as reported in the paper | 53% |   |   | none (reported estimate) | Published estimates from cited studies | cmto_treat |

## Findings

In a randomized trial among families with children issued housing vouchers in Seattle and King County, 53% of the families offered information, short-term financial help and customized help with the housing search moved to a neighborhood with high upward mobility, against 15% of the families in the control group.[1] That result answers a question the rest of this book cannot. The earlier chapters compare places and the people who already live in them. They show where outcomes are higher or lower and how they move with deprivation, and they cannot say what would happen to a family that moved, or to a place that changed.

This chapter reports the studies the book cites whose designs separate what a place, or a change to it, does for a family from the selection of families into places. It groups them by what was changed: an offer of a voucher, the years a child spent in a place, help with the search for housing, a forced move, and a change to a place or an institution. Each study describes its own population and period, and none of the estimates is pooled with another or attached to a gradient in an earlier chapter.

### Questions the designs answer

| What was changed | Design | Question it can answer | What it cannot answer |
| --- | --- | --- | --- |
| An offer of a housing voucher | Randomized experiment (Moving to Opportunity) | What changed for the families offered the voucher, compared with families not offered it | What any move to any less poor neighborhood would do |
| The years a child spent in a place | Movers design, comparing children who moved at different ages | How much of the difference between places a child's outcome takes on per year of exposure | Which feature of a place produces the change |
| Help with the housing search | Randomized experiment (Creating Moves to Opportunity) | Whether assistance changes where families with vouchers move | What the moves do to children's adult outcomes, which the trial has not yet observed |
| A forced move from public housing | Comparison of children in demolished and nearby buildings that were not demolished | What a displacement out of public housing changed for children | What a voluntary move would change |

A fifth group of studies changes a place or an institution rather than who lives there. Those studies are reported in the chapters on the outcomes they measure and are listed in the last section.

### An offer of a voucher changed some outcomes for some members of the family

> Causal evidence: Moving to Opportunity: adults' health and well-being and children's later income (Randomized housing-voucher experiment)
>
> From 1994 to 1998 the Department of Housing and Urban Development randomly assigned families in public housing in high-poverty tracts to a voucher usable only in a low-poverty tract, with counseling, to an unrestricted voucher, or to no voucher. Only 48% of the families offered the low-poverty voucher used it to move.
>
> In a survey of the women in 2008 to 2010, the prevalence of a glycated hemoglobin level of 6.5% or more, a marker of diabetes, was 4.3 (95% CI 0.8 to 7.8) points lower in the group offered the low-poverty voucher than in the control group, and the prevalence of a body-mass index of 40 or more was 3.4 (95% CI 0.4 to 6.4) points lower.[2] A decade or more after assignment, the offer of a move raised a standardized answer to one question about how happy things are by 0.098 (95% CI 0.022 to 0.174) standard deviations, with no detectable change in the adults' economic self-sufficiency.[3]
>
> The children's adult incomes depended on their age at the move. Children assigned the experimental voucher before age 13 had an annual income 14% higher in their mid-twenties than the control group, an intention-to-treat comparison that includes families who did not move. Scaled by take-up, the estimate for children of families who used the voucher is 31%. The estimated effects for children who were teenagers at the move were negative.[4]

The estimates compare groups by the offer they received, and the offer went to volunteers living in public housing in the poorest tracts of large cities. They describe what an offer did for those families. They do not give the change that a move along the full range of the deprivation index would produce, and none of the studies measured life expectancy.

### A child's adult income moves toward the destination's with each year spent there

> Causal evidence: Childhood exposure to a place and adult income (Movers design)
>
> Chetty and Hendren compared children whose families moved between areas when the children were at different ages, including siblings in the same family. The identifying assumption is that the selection that sends a family to a better or worse place does not vary with the child's age at the move, which they test with sibling comparisons, controls for changes in parental income and marital status, and moves that followed displacement shocks. Children's adult income ranks move toward those of children of permanent residents of the destination by 0.04 percentile for each year of childhood spent there and each percentile by which the destination's ranks are higher.[5] A companion paper estimates the causal contribution of each county.[6]
>
> The Opportunity Atlas authors combine the experimental and movers estimates and report that 60% of the variation in outcomes across neighborhoods is driven by causal effects of the neighborhoods.[7]

The design estimates how much of a place's difference a child takes on, and its benchmark is the children of permanent residents at the same parent income. It does not say which feature of a place, its schools, its neighbors, its jobs or its safety, carries the effect, and the share attributed to causal effects belongs to the Atlas outcome and sample, not to the outcomes of other chapters.

### Help with the housing search changed where families moved

> Causal evidence: Creating Moves to Opportunity: assistance with the housing search (Randomized controlled trial)
>
> Bergman and coauthors randomly assigned families with children issued housing vouchers in Seattle and King County to the usual voucher process or to a program that combined information about high-opportunity neighborhoods, short-term financial assistance, customized help with the search and connections to landlords. The program raised the share of families who moved to a high-upward-mobility area from 15% to 53%, and a second trial attributes most of the effect to the customized search assistance. The authors conclude that many low-income families do not strongly prefer low-mobility areas, and that barriers in the search for housing add to the segregation of families by income.[1]

The trial measures where families moved, not how their children fare as adults. Whether the moves change the children's outcomes is a question for the movers design and for later follow-up.

### A forced move out of public housing

> Causal evidence: Public housing demolitions in Chicago and children's later outcomes (Comparison of children in demolished buildings with children in nearby buildings that were not demolished)
>
> The demolition of public housing in Chicago required families to move, with housing vouchers, to less disadvantaged neighborhoods. Chyn compared children whose buildings were demolished with children in nearby buildings that were not. As young adults the displaced children were more likely to be employed, earned more and had fewer arrests for violent crime, and those displaced at young ages dropped out of high school less often.[8]

The families did not choose to move, so the comparison avoids the selection of volunteers in a voucher experiment. It describes one city's public housing in one period, and the change bundles a new neighborhood with the loss of the old building and its community.

### Changes to places and institutions

Moving a family and changing a place are different interventions. Several chapters report designs that study a change to a place or to an institution, each beside the outcome it measures.

| Chapter | What changed | Design |
| --- | --- | --- |
| Chapter 8.1, Gun homicide and the causes of early death (https://vishalsingh.org/zipcode/gun-homicide-and-causes-of-early-death) | Remediation of vacant lots | Cluster randomized trial |
| Chapter 4.2, Teen births (https://vishalsingh.org/zipcode/teen-births) | Access to long-acting contraception | Difference-in-differences |
| Chapter 2.3, Smoking in pregnancy (https://vishalsingh.org/zipcode/smoking-in-pregnancy) | Cigarette taxes and state income support | Matched comparison states; difference-in-differences |
| Chapter 2.1, Birthweight and preterm birth (https://vishalsingh.org/zipcode/birthweight-and-preterm-birth) | Income support and Medicaid eligibility | Difference-in-differences; simulated eligibility |
| Chapter 3.1, Test scores, class and race (https://vishalsingh.org/zipcode/test-scores-class-and-race) | Head Start funding and school finance reforms | Regression discontinuity; event study |
| Chapter 7.1, Health behaviors and the supply of care (https://vishalsingh.org/zipcode/health-behaviors-and-the-supply-of-care) | Insurance coverage and community health centers | Randomized lottery; difference-in-differences |
| Chapter 8.2, Overdose, alcohol and suicide (https://vishalsingh.org/zipcode/overdose-alcohol-and-suicide) | Naloxone access laws | Staggered difference-in-differences |
| Chapter 5.2, Home value, income and redlining (https://vishalsingh.org/zipcode/home-value-income-and-redlining) | The grading of neighborhoods on the HOLC maps | Boundary design |
| Chapter 6.1, Air pollution (https://vishalsingh.org/zipcode/air-pollution) | Redlining and traffic near the home | Boundary design; difference-in-differences |
| Chapter 5.4, Renting, affordability and eviction (https://vishalsingh.org/zipcode/renting-affordability-and-eviction) | An eviction order | Random assignment of cases to judges |
| Chapter 3.4, School discipline and incarceration (https://vishalsingh.org/zipcode/school-discipline-and-incarceration) | Suspension and juvenile incarceration | Boundary change; random assignment of cases to judges |
| Chapter 7.4, Disability, retirement and Medicare (https://vishalsingh.org/zipcode/disability-retirement-and-medicare) | Medicare eligibility, disability allowance and Social Security benefits | Regression discontinuity; examiner assignment; difference-in-differences |

Each study estimates what one policy or decision changed for the population it reached. None estimates how much a gradient across places would narrow if the policy were applied everywhere, and the estimates are not comparable with one another, since their outcomes, units and populations differ. The [index of studies](https://vishalsingh.org/zipcode/reference/studies) lists every study box in the book with its design, population and outcome.

### What these data cannot show

The studies estimate effects for volunteers in public housing, for children whose families moved, for families issued vouchers in one metro area and for children displaced in one city, in the periods they cover. None follows the populations the earlier chapters compare. An effect of an offer is not an effect of a move, and an effect of a move is not an effect of changing the place a family left. The movers design identifies how much a child's outcome follows a place, not which feature of the place is responsible. No study here gives the share of any gradient in this book that a policy could remove, and the book reports none.

## Tables

The chapter has no figures.

## All estimates

Every fact in facts.json (13), sorted by id.

| Fact id | Value | Unit | Estimand | Population | Period | Interval | n | Weights | Source | Note |
| --- | ---: | --- | --- | --- | --- | --- | ---: | --- | --- | --- |
| ch_exposure_yr | 0.04 | percentiles of adult income rank per year of childhood exposure, per percentile of difference in permanent residents' outcomes | exposure effect reported by Chetty and Hendren (2018, Quarterly Journal of Economics 133(3)): federal tax records, children born 1980 to 1988, baseline sample of one-time movers between commuting zones; movers design that compares children who moved at different ages, with sibling comparisons (family fixed effects) and moves caused by displacement shocks; the benchmark is the adult income rank of children of permanent residents of the destination at the same parent income; copied from the paper and not computed in this build | children in the study's baseline sample whose families moved once between commuting zones | children born 1980 to 1988, moves from 1997 to 2010 |   |   | none (reported estimate) | Published estimates from cited studies | A published estimate, not a result of this build. |
| cmto_control | 15% | percent of families who moved to a high-upward-mobility area | Bergman, Chetty, DeLuca, Hendren, Katz and Palmer (2024), American Economic Review 114(5): randomized trial among families with children issued housing vouchers in Seattle and King County; the treatment combined information, short-term financial assistance and customized search assistance; share of families who moved to a high-upward-mobility area, copied from the abstract and not computed in this build (control group) | families with children issued housing vouchers in Seattle and King County, control group of the Creating Moves to Opportunity trial | moves after random assignment, as reported in the paper |   |   | none (reported estimate) | Published estimates from cited studies | A published estimate, not a result of this build. |
| cmto_treat | 53% | percent of families who moved to a high-upward-mobility area | Bergman, Chetty, DeLuca, Hendren, Katz and Palmer (2024), American Economic Review 114(5): randomized trial among families with children issued housing vouchers in Seattle and King County; the treatment combined information, short-term financial assistance and customized search assistance; share of families who moved to a high-upward-mobility area, copied from the abstract and not computed in this build (treatment group) | families with children issued housing vouchers in Seattle and King County, treatment group of the Creating Moves to Opportunity trial | moves after random assignment, as reported in the paper |   |   | none (reported estimate) | Published estimates from cited studies | A published estimate, not a result of this build. |
| mto_age_cut | 13 | years of age | the age below which Chetty, Hendren and Katz (2016) find the gain in income (definition from the study) | children in the Moving to Opportunity experiment | age when the family moved |   |   | none (definition) | Published estimates from cited studies |   |
| mto_bmi_cut | 40 | kilograms per square meter (body-mass index) | the body-mass index at or above which Ludwig and coauthors (2011) count a woman as extremely obese (definition from the study) | women in the Moving to Opportunity experiment | health measured in 2008 to 2010 |   |   | none (definition) | Published estimates from cited studies |   |
| mto_diab_diff | 4.3 | percentage points lower prevalence of glycated hemoglobin of 6.5% or more in the low-poverty voucher group than in the control group | Ludwig and coauthors (2011), New England Journal of Medicine: 4,498 women with children living in public housing in high-poverty census tracts, randomly assigned in 1994 to 1998 to a voucher usable only in a low-poverty tract, an unrestricted voucher or no voucher; health measured in 2008 to 2010 (intention to treat); reported absolute difference in the prevalence of a glycated hemoglobin level of 6.5% or more, copied from the abstract and not computed in this build | women in the Moving to Opportunity experiment | health measured in 2008 to 2010 | 95% CI 0.8 to 7.8 |   | none (reported estimate) | Published estimates from cited studies | A published estimate, not a result of this build. |
| mto_hba1c_cut | 6.5% | percent glycated hemoglobin | the glycated hemoglobin level at or above which Ludwig and coauthors (2011) count a woman as having diabetes (definition from the study) | women in the Moving to Opportunity experiment | health measured in 2008 to 2010 |   |   | none (definition) | Published estimates from cited studies |   |
| mto_income_itt | 14% | percent higher annual income in the mid-twenties than the control group (intention to treat, $1,624) | Chetty, Hendren and Katz (2016), American Economic Review (NBER working paper 21156): children below age 13 at random assignment, individual earnings in the mid-twenties, control mean $11,270; intention-to-treat effect of assignment to the experimental voucher group, $1,624, a 14 percent increase; read in the text of the working paper and not computed in this build | children below age 13 at random assignment to the experimental voucher group of the Moving to Opportunity experiment | income in the mid-twenties |   |   | none (reported estimate) | Published estimates from cited studies | A published estimate, not a result of this build. |
| mto_income_tot | 31% | percent higher annual income in the mid-twenties among families that took up the voucher (treatment on the treated, $3,477) | Chetty, Hendren and Katz (2016), American Economic Review (NBER working paper 21156): children below age 13 at random assignment, individual earnings in the mid-twenties, control mean $11,270; treatment-on-the-treated effect for families that took up the experimental voucher, the intention-to-treat effect scaled by the take-up rate, $3,477, a 31 percent increase; read in the text of the working paper and not computed in this build | children below age 13 whose families took up the experimental voucher in the Moving to Opportunity experiment | income in the mid-twenties |   |   | none (reported estimate) | Published estimates from cited studies | A published estimate, not a result of this build. |
| mto_obese_diff | 3.4 | percentage points lower prevalence of a body-mass index of 40 or more in the low-poverty voucher group than in the control group | Ludwig and coauthors (2011), New England Journal of Medicine: 4,498 women with children living in public housing in high-poverty census tracts, randomly assigned in 1994 to 1998 to a voucher usable only in a low-poverty tract, an unrestricted voucher or no voucher; health measured in 2008 to 2010 (intention to treat); reported absolute difference in the prevalence of a body-mass index of 40 or more, copied from the abstract and not computed in this build | women in the Moving to Opportunity experiment | health measured in 2008 to 2010 | 95% CI 0.4 to 6.4 |   | none (reported estimate) | Published estimates from cited studies | A published estimate, not a result of this build. |
| mto_swb | 0.098 | standard deviations of a three-point happiness question | intention-to-treat effect of the offer to move to a less disadvantaged area on a standardized answer to 'how would you say things are these days: very happy, pretty happy, not too happy' in the Moving to Opportunity experiment (Ludwig and coauthors 2012); the estimate and standard error (0.039) are copied from the paper and the interval is the estimate plus or minus 1.96 standard errors | low-income adults in public housing in five cities (Baltimore, Boston, Chicago, Los Angeles, New York) who were offered a housing voucher to move to a lower-poverty neighborhood or an unrestricted voucher, or stayed, by random assignment | long-term follow-up survey, 10 to 15 years after random assignment | 95% CI 0.022 to 0.174 |   | as in the paper | Published estimates from cited studies | A published estimate for the study's adult sample, not a result of this build. |
| mto_takeup | 48% | percent of families assigned the experimental voucher who took it up | Chetty, Hendren and Katz (2016): take-up rate of the experimental voucher; read in the text of the working paper (NBER w21156) and not computed in this build | families in the experimental voucher group of the Moving to Opportunity experiment | take-up of the voucher |   |   | none (reported estimate) | Published estimates from cited studies | A published estimate, not a result of this build. |
| oa_causal_share | 60% | percent of the variation in adult outcomes across neighborhoods | share of the variation across neighborhoods that experimental and quasi-experimental estimates attribute to causal effects of neighborhoods; reported in the abstract of Chetty, Friedman, Hendren, Jones and Porter (2018), copied from it and not computed in this build | census tracts of the United States, children born 1978 to 1983 | children born 1978 to 1983, household income measured in 2014 to 2015 |   |   | none (reported estimate) | Opportunity Atlas | A published estimate, not a result of this build. |

## Definitions

The chapter marks no glossary term.

## Methods

This chapter computes nothing. It reports studies whose designs assign an offer by lottery or compare children who moved at different ages, and every estimate describes the study's population and period, not the places compared elsewhere in the book.

### Where the numbers come from

The Moving to Opportunity health and income estimates are copied from Place and the length of a life, the well-being estimate from Life evaluation and daily feelings, the exposure rate of the movers design from Eight types of neighborhood, and the Opportunity Atlas share from Race, income and mobility, with their populations, periods and intervals.

The move rates of the Creating Moves to Opportunity trial are copied from the abstract of Bergman and coauthors (2024). The study of Chicago public housing demolitions is reported in words, from the abstract of Chyn (2018).

### Offers, moves and scaling

An intention-to-treat estimate compares groups by the offer they were randomized to, including families who did not move. A treatment-on-the-treated estimate divides it by the share that took up the offer, which assumes the offer changed nothing for families who did not use it. The chapter names which estimate each number is.

## What these data cannot show

The studies estimate effects for volunteers in public housing, for children whose families moved, for families issued vouchers in one metro area and for children displaced in one city, in the periods they cover. None follows the populations the earlier chapters compare. An effect of an offer is not an effect of a move, and an effect of a move is not an effect of changing the place a family left. The movers design identifies how much a child's outcome follows a place, not which feature of the place is responsible. No study here gives the share of any gradient in this book that a policy could remove, and the book reports none.

## References

1. Bergman, Peter; Chetty, Raj; DeLuca, Stefanie; Hendren, Nathaniel; Katz, Lawrence F.; Palmer, Christopher (2024). Creating Moves to Opportunity: Experimental Evidence on Barriers to Neighborhood Choice. American Economic Review 114(5): 1281-1337. https://doi.org/10.1257/aer.20200407
2. Ludwig, Jens; Sanbonmatsu, Lisa; Gennetian, Lisa; Adam, Emma; Duncan, Greg J.; Katz, Lawrence F.; Kessler, Ronald C.; Kling, Jeffrey R.; Lindau, Stacy Tessler; Whitaker, Robert C.; McDade, Thomas W. (2011). Neighborhoods, Obesity, and Diabetes: A Randomized Social Experiment. New England Journal of Medicine 365(16): 1509-1519. https://doi.org/10.1056/NEJMsa1103216
3. Ludwig, Jens; Duncan, Greg J.; Gennetian, Lisa A.; Katz, Lawrence F.; Kessler, Ronald C.; Kling, Jeffrey R.; Sanbonmatsu, Lisa (2012). Neighborhood effects on the long-term well-being of low-income adults. Science 337(6101): 1505-1510. https://doi.org/10.1126/science.1224648
4. Chetty, Raj; Hendren, Nathaniel; Katz, Lawrence F. (2016). The Effects of Exposure to Better Neighborhoods on Children: New Evidence from the Moving to Opportunity Experiment. American Economic Review 106(4): 855-902. https://doi.org/10.1257/aer.20150572
5. Chetty, Raj; Hendren, Nathaniel (2018). The Impacts of Neighborhoods on Intergenerational Mobility I: Childhood Exposure Effects. The Quarterly Journal of Economics 133(3): 1107-1162. https://doi.org/10.1093/qje/qjy007
6. Chetty, Raj; Hendren, Nathaniel (2018). The Impacts of Neighborhoods on Intergenerational Mobility II: County-Level Estimates. The Quarterly Journal of Economics 133(3): 1163-1228. https://doi.org/10.1093/qje/qjy006
7. Chetty, Raj; Friedman, John N.; Hendren, Nathaniel; Jones, Maggie R.; Porter, Sonya R. (2018). The Opportunity Atlas: Mapping the Childhood Roots of Social Mobility. NBER Working Paper 25147. https://doi.org/10.3386/w25147
8. Chyn, Eric (2018). Moved to Opportunity: The Long-Run Effects of Public Housing Demolition on Children. American Economic Review 108(10): 3028-3056. https://doi.org/10.1257/aer.20161352

## Provenance

- Pipeline version: 0.1.0
- Build script: book2/chapters/what-moving-can-change/build.py (sha256 34775c7caa18b1b1aabda540dae962103933b123ad488162b8123f05b7370254)
- Built on: 2026-10-28
- Site commit: c2b1ff3309cb4a10833f884da91ac0057a66ad0a

Inputs:

| File | Size (bytes) | SHA-256, first 1 MB | SHA-256, last 1 MB |
| --- | ---: | --- | --- |
| C:/github/zipcode-book/content/chapters/eight-types-of-neighborhood/data/facts.json | 67,536 | 9b5fa324c28f5faf9951ff17d7d2a7b52cc0f8db1da35566a157c0f14b5565f9 | 9b5fa324c28f5faf9951ff17d7d2a7b52cc0f8db1da35566a157c0f14b5565f9 |
| C:/github/zipcode-book/content/chapters/life-evaluation-and-daily-feelings/data/facts.json | 105,948 | ed1377d46cea9f92d315518cf23107c167643246e6f909612c77af42aa4f2748 | ed1377d46cea9f92d315518cf23107c167643246e6f909612c77af42aa4f2748 |
| C:/github/zipcode-book/content/chapters/life-evaluation-and-daily-feelings/data/refs.json | 2,366 | af3d3761f67ed01c066f1a1224342ebede46f2a79cc9e31ae92c4bc16acb298b | af3d3761f67ed01c066f1a1224342ebede46f2a79cc9e31ae92c4bc16acb298b |
| C:/github/zipcode-book/content/chapters/place-and-the-length-of-a-life/data/facts.json | 71,140 | 9eed1ad6694ae817ab5800fd787f79ec8a01a706b2d211bccb978f17d11e6e6d | 9eed1ad6694ae817ab5800fd787f79ec8a01a706b2d211bccb978f17d11e6e6d |
| C:/github/zipcode-book/content/chapters/place-and-the-length-of-a-life/data/refs.json | 1,878 | 2d976463c36bc44fd12832d343cc0a844526c07065c1c3444516db9b007dd7f8 | 2d976463c36bc44fd12832d343cc0a844526c07065c1c3444516db9b007dd7f8 |
| C:/github/zipcode-book/content/chapters/race-income-and-mobility/data/facts.json | 58,319 | 78f6e8841f063c454c4a54f624a06a916d204af05d3a711bcd457b0043f68ca0 | 78f6e8841f063c454c4a54f624a06a916d204af05d3a711bcd457b0043f68ca0 |
| C:/github/zipcode-book/content/chapters/race-income-and-mobility/data/refs.json | 2,424 | 52c81dac44ace635a783210962a409d0d152a5168429e5b8da06018c4ce74a7d | 52c81dac44ace635a783210962a409d0d152a5168429e5b8da06018c4ce74a7d |

Outputs:

| File | SHA-256 |
| --- | --- |
| facts.json | 3c6b4662f85282e76f5c5661b1d80b5da85d4d8ee2fefbf758aa2564c16cfd94 |
| meta.json | 3178bc7e7b451e021b5dad69f900d55b6df329f8b101a6ad4f6efd66b2ba2cef |
| refs.json | d5f4017f3b5fa104e57ce95078364a1594a4ce7dc1c46131a8694c015fa3c5be |

## Machine formats

- Report (HTML): https://vishalsingh.org/zipcode/what-moving-can-change/report
- Report (Markdown): https://vishalsingh.org/zipcode/what-moving-can-change/report.md
- Every estimate (JSON): https://vishalsingh.org/zipcode/report/what-moving-can-change/facts.json
- Index of every report: https://vishalsingh.org/zipcode/llms.txt; all reports in one file: https://vishalsingh.org/zipcode/llms-full.txt

---

# Gradients, slopes and variance: report

Chapter 10.1 of ZIP Code Destiny, by Vishal Singh (NYU Stern). Generated from the chapter's data files; the reading version is at https://vishalsingh.org/zipcode/gradients-slopes-and-variance. Table n of this report is figure n of the chapter.

## Identification

- Title: Gradients, slopes and variance
- Label: The factory floor
- Chapter: 10.1 (Part X, How the estimates are made; Methods and sources)
- Question: How are the book's gradients estimated, and how much of the variation lies between states, between counties and within counties?
- Dek: For deaths in 2010 to 2015, tract life expectancy is 1.12 years lower per ten-point step of the ADI within counties, and lower in more deprived tracts in all 267 metros.
- Grain: neighborhoods
- Status: published
- Published: 2026-10-09
- Updated: not revised
- Data sources:
  - USALEEP life expectancy. USALEEP life expectancy at birth, 2010-2015, on 2020 tracts (area-weighted from 2010 tracts), institutional tracts dropped; years: 2010-2015; weights: ACS 2019-2023 tract population
  - Area Deprivation Index. Neighborhood Atlas Area Deprivation Index 2023, national percentile, tracts; years: 2023 release (version 4.0.1; built on ACS 2019-2023); weights: ACS 2019-2023 tract population (each tract once for the index split by county)
  - American Community Survey. ACS 2019-2023 tracts: population, the nine measures of the index and the measures of the correlations; years: 2019-2023 (5-year); weights: tract population; each tract once in the index
  - Census TIGER/Line tracts. Census 2020 county file: the core-based statistical area code of each county (CBSAFP); years: 2020; weights: none (assignment of counties to metros)
  - CDC PLACES. CDC PLACES 2024 release, tracts: six measures of the coefficient zoo; years: 2024 release (tracts); 2025 release (ZCTAs, 2023 data year); weights: adult population of each tract (PLACES)
  - CDC WONDER natality. CDC WONDER natality 2018-2024, county by birthweight class: counts of withheld cells; years: 2018-2024; weights: none (full counts of cells)
- Canonical URL: https://vishalsingh.org/zipcode/gradients-slopes-and-variance
- Report: https://vishalsingh.org/zipcode/gradients-slopes-and-variance/report
- Markdown: https://vishalsingh.org/zipcode/gradients-slopes-and-variance/report.md
- BibTeX key: singh2026gradientsslopesandvariance

```bibtex
@misc{singh2026gradientsslopesandvariance,
  author = {Singh, Vishal},
  title = {Gradients, slopes and variance},
  year = {2026},
  url = {https://vishalsingh.org/zipcode/gradients-slopes-and-variance},
  note = {In ZIP Code Destiny. Published 9 October 2026.}
}
```

## Principal estimates

The facts the dek and the opening prose cite, in order of appearance.

| Estimand | Population | Period | Value | Interval | n | Weights | Source | Fact id |
| --- | --- | --- | ---: | --- | ---: | --- | --- | --- |
| population-weighted least-squares slope of tract life expectancy on the ADI national percentile, times ten, sign reversed, with a constant for each county; cluster-robust interval (t on the number of counties minus one) | residents of non-institutional tracts in the 50 states and DC with a life expectancy estimate and an ADI rank | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population | 1.12 | 95% CI 1.07 to 1.16 | 77,722 tracts | ACS 2019-2023 tract population | USALEEP life expectancy | ladder_county |
| count of metros | residents of metros (core-based statistical areas) with at least 40 tracts that have a life expectancy estimate and an ADI rank | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population | 267 |   |   | none (full count) | USALEEP life expectancy | n_metros |
| population-weighted least-squares slope of tract life expectancy on the ADI national percentile, times ten, sign reversed; cluster-robust interval (t on the number of states minus one) | residents of non-institutional tracts in the 50 states and DC with a life expectancy estimate and an ADI rank | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population | 0.84 | 95% CI 0.75 to 0.92 | 77,936 tracts | ACS 2019-2023 tract population | USALEEP life expectancy | ladder_raw |
| count of tracts in the regressions | residents of non-institutional tracts in the 50 states and DC with a life expectancy estimate and an ADI rank | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population | 77,936 |   |   | none (full count) | USALEEP life expectancy | n_ladder_tracts |

## Findings

Life expectancy at birth is 0.84 (95% CI 0.75 to 0.92) years lower for each ten-point step up the Area Deprivation Index (ADI) national percentile across the 77,936 census tracts that have an ADI rank and a USALEEP estimate for deaths in 2010 to 2015.[1][2] Within counties the difference is 1.12 (95% CI 1.07 to 1.16) years, and a separate regression in each of 267 metros finds lower life expectancy in more deprived tracts in every one. This chapter shows how those estimates are made and where they stop being informative. It covers the regression and its fixed effects, one regression per metro and per state, and a nine-measure index. It then takes up the change in a correlation with the size of the area, the split of variance between states, counties and tracts, and the suppressed cells of CDC WONDER. The gradient by decile of deprivation is the subject of Chapter 8.4, Place and the length of a life (https://vishalsingh.org/zipcode/place-and-the-length-of-a-life).

### The slope is steeper within counties than across all tracts

The regression behind most figures in this book is a population-weighted least-squares line through tracts,

$$
Y_i = \alpha + \beta\,\mathrm{ADI}_i + \varepsilon_i,
$$

where $$Y_i$$ is the life expectancy of tract $$i$$ in years, $$\mathrm{ADI}_i$$ is its national percentile (higher is more deprived), $$\alpha$$ is the intercept, $$\beta$$ is the slope and $$\varepsilon_i$$ is the part of life expectancy the line leaves unexplained. The line minimizes the sum of squared residuals with each tract weighted by its 2019-2023 population, so a tract with few residents counts for less. The book reports the estimated $$\beta$$ with its sign reversed and scaled to a ten-point step. A reported slope is therefore the years of life expectancy lower for a ten-point step, and a slope above zero means life expectancy is lower where the ADI is higher. Counting each tract once moves none of the three slopes in Table 1 by more than 0.015 years, so the weights do not produce the result.

A national slope can come from composition, with the least deprived tracts concentrated in some states and the most deprived in others. The fixed-effects ladder tests for that by giving each state, and then each county, its own intercept,

$$
Y_{ig} = \beta\,\mathrm{ADI}_{ig} + \mu_g + \varepsilon_{ig},
$$

where $$g$$ indexes states in the second row of the figure and counties in the third, and $$\mu_g$$ is the constant of group $$g$$. The slope $$\beta$$ then compares only tracts in the same state or county. The fits subtract each group's weighted mean from life expectancy and from the ADI before fitting, which gives the same slope as a dummy variable for every group. Errors cluster on the group that was absorbed, states for the first two rows and counties for the third. The intervals use the t distribution with one fewer degree of freedom than there are clusters.[3]

Across all tracts, life expectancy is 0.84 years lower per step, within states 0.92 and within counties 1.12, so the within-county slope is 1.3 times the national one. A composition story predicts the opposite order. The county fit uses 77,722 tracts, because a tract that is alone in its county adds no contrast.

Table 1. Life expectancy is 1.12 years lower per ten-point step of the ADI within counties, against 0.84 years across all tracts

A slope that survives these comparisons shows that rich and poor states at opposite ends of one line do not produce the pattern. It still compares tracts, and families sort themselves among tracts for reasons the regression does not observe, so it does not say what would happen to a family that moved.

### Life expectancy is lower in more deprived tracts in every metro

A single slope assumes that the relationship is the same everywhere, and fitting the regression separately in each metro measures how much it differs. A metro here is a core-based statistical area, metropolitan or micropolitan, as the Census Bureau's county file codes it. Every tract takes the area of its county, which keeps Connecticut, whose planning regions stand in for counties. The file has no names, so Table 2 labels each metro by its most populous county, and Connecticut metros by a planning region. A metro enters if it has at least 40 tracts with a life expectancy and an ADI rank. 267 do, and they hold 82% of the residents in the sample.

A floor of 25 tracts gives 403 metros and a floor of 100 gives 134. Life expectancy is lower in more deprived tracts in every metro at each of the three floors.

The model is the first one, fitted within each metro $$m$$ with the same tract weights,

$$
Y_{im} = \alpha_m + \beta_m\,\mathrm{ADI}_{im} + \varepsilon_{im},
$$

where $$\alpha_m$$ and $$\beta_m$$ are the intercept and slope of metro $$m$$. Errors cluster on counties in the 64 metros that have at least 5 counties. The rest use HC3 errors, which let the variance differ across tracts but treat tracts as independent, so their intervals may be too narrow where neighboring tracts resemble one another.

All 267 reported slopes are above zero, and the 95% interval of 260 of them lies above zero. The median metro has a life expectancy 1.20 (95% CI 1.14 to 1.23) years lower per ten-point step, with an interval from a bootstrap over metros.

At the 5th percentile of metros the difference is 0.70 years and at the 95th it is 1.67, a ratio of 2.4. Part of that range is estimation error. Let $$\hat\beta_m$$ be the estimated slope of metro $$m$$ and $$v_m$$ its squared standard error, let $$s^2$$ be the variance of the $$\hat\beta_m$$ across metros and let $$\bar v$$ be the mean of the $$v_m$$. The variance of the true slopes is estimated by $$\tau^2 = s^2 - \bar v$$, and each estimate can be shrunk toward the mean $$\bar\beta$$ in proportion to its reliability,

$$
\tilde\beta_m = \bar\beta + \frac{\tau^2}{\tau^2 + v_m}\left(\hat\beta_m - \bar\beta\right).
$$

Estimation error accounts for 48% of the variance of the estimated slopes. The standard deviation of the true slopes is 0.22 years against 0.31 for the estimates, and after shrinkage the 5th and 95th percentiles move to 0.88 and 1.46. The second tab of the figure sorts the estimates, which puts the metros with the least precise slopes at both ends. The order of neighbors therefore carries no information, and the chapter names no metro as the steepest or the flattest.

Table 2. Life expectancy is lower in more deprived tracts in all 267 metros, by a median of 1.20 years per ten-point step

### Metro traits are not detectably related to the spread

Table 3 plots the metro slope against a metro's median household income, taken as the population-weighted median of its tracts' median incomes. Each correlation is computed weighted by metro residents and counting each metro once, with the slope measured as years lower per step. The correlation with income is +0.13 (95% CI −0.24 to 0.46) weighted and +0.05 (95% CI −0.11 to 0.19) unweighted.

Two further traits are the dispersion of tract ADI in the metro (the population-weighted standard deviation) and the metro's population. The correlation with ADI dispersion is −0.21 (95% CI −0.46 to 0.12) weighted and −0.14 (95% CI −0.27 to 0.01) unweighted, and with population −0.20 (95% CI −0.47 to 0.10) and −0.06 (95% CI −0.18 to 0.08). Each correlation has a bootstrap interval over metros, and all six intervals include zero, the one for ADI dispersion counting each metro once only just. No line is drawn through the cloud.

This is a failure to detect a relationship and not evidence that none exists. The slopes carry estimation error, 48% of their variance, which pulls every correlation toward zero. The weighted intervals are wide enough to leave a moderate correlation of either sign for income, dispersion or population. The traits also do not exhaust the candidates, which include health systems and housing, and the chapter tests no others.

Table 3. Metro slopes are not detectably related to metro median income, a correlation of +0.05 counting each metro once

### The coefficient zoo returns coefficients of both signs

The method does not depend on life expectancy. For six behaviors and conditions in CDC PLACES, model-based estimates of the share of adults in each tract (modeled estimates),[4] the regression

$$
P_i = a_s + \gamma_s\,\ln(\mathrm{income}_i) + \boldsymbol{\delta}_s'\,\mathbf{c}_i + \varepsilon_i
$$

is fitted in each state $$s$$ across the tracts $$i$$ in it. Here $$P_i$$ is the share of adults with the condition or behavior, $$\mathrm{income}_i$$ is the tract's median household income, $$\mathbf{c}_i$$ holds the shares of residents who are ages 65 and over, Black alone and Hispanic, and the weight is the tract's adult population. The intercept $$a_s$$, the income coefficient $$\gamma_s$$ and the control coefficients $$\boldsymbol{\delta}_s$$ are specific to state $$s$$. The coefficient reported is $$\gamma_s \ln 2$$, the change in points of prevalence for a doubling of income. A state enters if it has at least 60 tracts with every variable. All 51 states, counting the District of Columbia as one, do, which gives 306 coefficients. Chapter 7.1, Health behaviors and the supply of care (https://vishalsingh.org/zipcode/health-behaviors-and-the-supply-of-care) reads what they say about behavior, and Table 4 shows the same dots.

The coefficients for smoking, obesity, diabetes, frequent mental distress and physical inactivity are negative in all 51 states, and the median for smoking is −4.7 points per doubling. The coefficient for binge drinking is positive in 50 states, with a median of +1.2 points.

The state coefficients can themselves be analyzed. States where smoking is more common have more negative smoking coefficients, a correlation of −0.73 (95% CI −0.84 to −0.60) across states. Most of that is scale, since the same proportional gradient is more points where the level is higher. Dividing each coefficient by the state's smoking level gives a correlation of −0.18 (95% CI −0.42 to 0.08).

Table 4. The same regression in every state gives coefficients below zero for five measures in all 51 states and above zero for binge drinking in 50

PLACES draws on the same census inputs as the income measure, and within a county it separates tracts only through their demographic mix. The coefficients therefore describe how the model maps composition into prevalence as well as how prevalence varies with income. A positive tract-level coefficient for binge drinking does not say who drinks.

### A nine-measure index keeps one component and discards a second

The ADI is the deprivation measure behind this book's gradients, and Chapter 1.1, What the deprivation index measures (https://vishalsingh.org/zipcode/anatomy-of-deprivation) examines how it is built. To study how an index of this kind behaves, this chapter builds a second one from nine tract measures in the American Community Survey (ACS) for 2019-2023. They are the poverty rate, the share of households on cash assistance or SNAP, the unemployment rate, the share of adults with a high school diploma or less, the share of households headed by a single parent, the log of median household income, the share of adults with a bachelor's degree or more, labor-force participation and homeownership. A tract needs all nine and an ADI rank, which leaves 82,545 tracts. Another 1,066 tracts with residents are dropped for lacking a measure or a rank. Each measure is standardized to mean zero and standard deviation one across tracts, each tract counted once as the ADI counts tracts, and the principal components are the eigenvectors of the correlation matrix,[5]

$$
z_{ik} = \frac{x_{ik}-\bar x_k}{s_k}, \qquad R\,\ell_j = \lambda_j\,\ell_j, \qquad \mathrm{index}_i = \text{percentile rank of } \sum_{k}\ell_{1k}\,z_{ik},
$$

where $$x_{ik}$$ is measure $$k$$ in tract $$i$$, $$\bar x_k$$ and $$s_k$$ are its mean and standard deviation across tracts, $$R$$ is the correlation matrix, $$\ell_j$$ is the loading vector of component $$j$$ and $$\lambda_j$$ is its eigenvalue, the variance of the component in units of one standardized measure.

The first component holds 51.3% (95% CI 50.4% to 52.1%) of the variance and the second 16.0% (95% CI 15.5% to 16.4%), together 67.2% (95% CI 66.5% to 68.0%) because the shares are rounded separately. Weighting tracts by population gives 51.4% for the first. The Kaiser rule keeps components with an eigenvalue above one, and 2 clear it.[6] The rule is a convention with no statistical test behind it.

The measures overlap. The share of adults with a bachelor's degree and the share with a high school diploma or less correlate at −0.90 (95% CI −0.91 to −0.90). Cronbach's alpha for the nine, signed so that higher means more deprived, is 0.87 (95% CI 0.86 to 0.87).[7]

Table 5. The first of nine components holds 51.3% of the variance in nine census measures and the second holds 16.0%

The weights matter little. The index correlates with the score of a one-factor factor analysis at +0.994 (95% CI 0.993 to 0.994) and with an equal-weight average of the nine signed measures at +0.997 (95% CI 0.996 to 0.997), both rank correlations. Dropping any one measure and refitting moves the median tract by at most 2.3 percentile points, and the largest absolute loading on the first component is 0.41, so no measure dominates the score. The choice of measures matters more, since the ADI correlates with the index at +0.68 (95% CI 0.61 to 0.74).

The second component is orthogonal to the first by construction, so an index built on the first discards it. It contrasts homeownership (loading −0.59) with labor-force participation (+0.56) and the college share (+0.35), and tracts with the same index value can differ on it. That the nine measures cohere statistically does not show that the index measures what a family in the tract would call deprivation. Chapter 1.2, What a county average leaves out (https://vishalsingh.org/zipcode/deprivation-at-two-zooms) rebuilds a deprivation index for counties and Chapter 10.2, Families of places (https://vishalsingh.org/zipcode/families-of-places) groups tracts instead of ranking them.

### A correlation depends on the size of the area

The same data can give different correlations at different grains, a case of the modifiable areal unit problem.[8] Robinson showed in 1950 that a correlation across states can differ in size and sign from the correlation across the people inside them.[9][10] Table 6 computes six pairs of tract measures across tracts, across counties and across states, all on the same 77,775 tracts, which lie in 3,112 counties and Connecticut planning regions. A county or state value is the population-weighted mean of its tracts' values. Each correlation is computed both weighted by the population of its unit and counting each unit once, with a bootstrap interval that draws counties for the tract correlations and states for the others. The state correlations rest on 51 states, so their intervals are wide.

Weighted by population, 4 of the 6 pairs are larger in absolute value across states than across tracts, and counting each unit once 4 are, but not the same pairs. College share and income go from +0.73 (95% CI 0.71 to 0.74) across tracts to +0.83 (95% CI 0.73 to 0.94) across states when weighted.

Homeownership and poverty fall toward zero under both weightings. Weighted, they go from −0.53 (95% CI −0.55 to −0.51) across tracts through −0.40 (95% CI −0.53 to −0.26) across counties to −0.05 (95% CI −0.30 to 0.17) across states. Counting each unit once they go from −0.52 (95% CI −0.54 to −0.50) to −0.11 (95% CI −0.32 to 0.11), so both state intervals include zero.

The weights decide the result for 2 of the 6 pairs. Unemployment and poverty rise from +0.43 (95% CI 0.41 to 0.46) across tracts to +0.59 (95% CI 0.52 to 0.66) across counties and fall to +0.33 (95% CI 0.08 to 0.59) across states when weighted. Counting each unit once they are +0.45 (95% CI 0.42 to 0.47), +0.53 (95% CI 0.44 to 0.61) and +0.50 (95% CI 0.32 to 0.65).

College share and life expectancy rise from +0.56 (95% CI 0.54 to 0.58) across tracts to +0.71 (95% CI 0.55 to 0.83) across states when weighted and fall from +0.55 (95% CI 0.53 to 0.57) to +0.48 (95% CI 0.13 to 0.81) counting each unit once. A weighted correlation across states gives populous states most of the weight, so the two weightings answer different questions. Averaging within an area removes the variation inside it, so a correlation of averages describes how the areas differ and says nothing about how the tracts or people inside them do.

Table 6. Weighted by population, the same pair of measures has a different correlation at each grain, and homeownership and poverty go from −0.53 across tracts to −0.05 across states

### Most variation in life expectancy lies between tracts in the same county

Split the total variance of a tract measure $$y$$, with population weights $$w_i$$, into a part between states, a part between counties within states and a part between tracts within counties,

$$
\underbrace{\sum_i w_i (y_i-\bar y)^2}_{\text{total}} = \underbrace{\sum_i w_i (\bar y_{s(i)}-\bar y)^2}_{\text{between states}} + \underbrace{\sum_i w_i (\bar y_{c(i)}-\bar y_{s(i)})^2}_{\text{between counties}} + \underbrace{\sum_i w_i (y_i-\bar y_{c(i)})^2}_{\text{between tracts}},
$$

where $$\bar y$$ is the weighted mean of all tracts and $$\bar y_{s(i)}$$ and $$\bar y_{c(i)}$$ are the weighted means of the state and the county that contain tract $$i$$. Each part over the total is a share of the variance, and the three shares sum to the whole. The split of life expectancy at birth uses the 78,008 tracts of 3,112 counties and Connecticut planning regions.

For life expectancy, the shares are 13.7% between states, 15.9% between counties in the same state and 70.3% between tracts in the same county. The split covers every tract in the sample and has no sampling interval, but it needs a caveat. The tract estimates are modeled, and their mean squared standard error equals 26.7% of the total variance, which sits inside the within-county share. Taking the published errors at face value, 60% of the variance that is not estimation error lies within counties. The errors of the 2020-tract file are mapping grade and understate the noise, so the share of the true differences that lies within counties is lower than that.

Table 7. 70.3 percent of the variance of tract life expectancy lies between tracts in the same county, 26.7 points of it estimation error, against 34% percent for the ADI

The ADI is the exception among the measures in the figure. Only 34% of its variance lies within counties and 37% lies between states, against 72% within counties for the nine-measure index. The two deprivation measures put their variation at different levels, and Chapter 1.1, What the deprivation index measures (https://vishalsingh.org/zipcode/anatomy-of-deprivation) examines what the ADI is made of. A county map of life expectancy paints the county average and hides the largest share of its variance.

### Large counties differ in how much of the ADI scale their tracts cover

A county's ADI range is the distance between the tract at the 10th percentile of ADI and the tract at the 90th, weighted by population. Across the 757 counties and Connecticut planning regions with at least 20 tracts that have an ADI rank, the median range is 39 percentile points.

The floor is a choice. With a floor of 10 tracts there are 1,319 counties and the median is 34 points. With a floor of 40 there are 421 and the median is 43.

Among the 30 most populous counties (Table 8), the range runs from 9 points in Santa Clara County, California, to 69 in Philadelphia County, Pennsylvania. Of the 30, 11 have a range of 25 points or less.

The median range in life expectancy between tracts at the 10th and 90th percentiles in those counties is 8.0 years. The counties are chosen by population and not by range, and tract life expectancy is modeled, so part of that spread in years is estimation error.

A county's mean ADI and its range are related, and how closely depends on the floor. Counting each county once, the correlation is −0.08 (95% CI −0.25 to 0.08) at a floor of 10 tracts, +0.27 (95% CI 0.06 to 0.44) at 20 and +0.62 (95% CI 0.45 to 0.74) at 40. Counting each county once, a county's mean says little about its range when small counties are included and more among large counties.

Weighted by county population the correlation is +0.47 (95% CI 0.16 to 0.63), +0.63 (95% CI 0.37 to 0.75) and +0.77 (95% CI 0.58 to 0.85) at the three floors. The figure shows the county averages as dots, and a county map shows only those and none of the whiskers.

Table 8. In each of the 30 most populous counties, tracts at the 10th and 90th percentile of the ADI are between 9 and 69 percentile points apart

### Suppressed cells are not zeros

CDC WONDER withholds any cell with fewer than 10 events, so a suppressed cell holds at most 9 births, and a published zero is a real zero.

The natality file by county and birthweight class for 2018 to 2024 has 11,250 withheld cells of 48,552, or 23.2%, in the 578 counties that WONDER names in every year. Those are counties of at least 100,000 residents.

The share depends on the class (Table 9). It is 71.3% for births of 499 grams or less, 77.6% for births of 5,000 grams or more and 65.1% for births whose weight is not stated. The five middle classes, where counts are large, have 0 withheld cells.

Table 9. WONDER withholds 71.3% of county-year cells in the lightest birthweight class and 77.6% in the heaviest, against 0.0% in the five middle classes

In the lightest class 2,885 cells are withheld and 350 are published as zero. A withheld cell is never a zero, so reading the blanks as zeros would misstate each of those withheld cells, and in this class they outnumber the real zeros.

If every withheld cell held 9 births, the withheld cells of the lightest class would hold 60% of its births and those of the heaviest class 76%. That limit comes from the public threshold and the births in published cells, and it says nothing about any one cell. A count of births in a rare class summed over published cells covers only the county-years where the class is large enough to publish.

The book counts withheld cells, leaves them out of both numerator and denominator and derives no value for them. Gradients use no suppression floor, and a map of WONDER deaths needs at least 20 pooled deaths and says so. Chapter 2.1, Birthweight and preterm birth (https://vishalsingh.org/zipcode/birthweight-and-preterm-birth) uses this file, and Chapter 10.3, Suppressed counties in death data (https://vishalsingh.org/zipcode/suppressed-counties-in-death-data) takes up the same problem in death counts.

### What these data cannot show

Every estimate in this chapter describes how an outcome is arranged across places. The slopes, the metro regressions, the index and the variance split compare tracts, counties and states, and none says what would happen to a person who moved or to a place that changed. A gradient of tract life expectancy on the ADI says nothing about individuals, which is the ecological inference problem. The steeper slope within counties is compatible with more than one explanation that the data cannot separate. Life expectancy is a modeled estimate for deaths in 2010 to 2015 that has not been updated. The 2020-tract file is area-weighted from 2010 tracts, so the estimate for a single tract is not reliable.

The metro slopes cover the residents of metros with enough tracts and say nothing about the rest. Their intervals use HC3 errors in most metros and may be too narrow, which would make the share of the spread that is estimation error larger. The ADI is a national percentile and understates deprivation in high-cost cities (Chapter 1.1, What the deprivation index measures (https://vishalsingh.org/zipcode/anatomy-of-deprivation)), which bears on the metro slopes. Puerto Rico is not covered, and Connecticut's planning regions stand in for its counties in the metros, the variance split and the county ranges.

The index is internally coherent, and the data hold no independent measure of what residents would call deprivation, so they cannot show that it is valid. The ACS values behind the index, the ADI and the area-size correlations are survey estimates with sampling error that is largest in small tracts. The staged file carries no margins of error, so every analysis here treats them as exact. The ADI and the PLACES estimates draw on the same census inputs as the measures they are compared with, so part of each association is built in, and the data cannot say how much.

Weights can decide a result. The correlations across states for unemployment and poverty, and for the college share and life expectancy, change direction between the two weightings in the area-size section, and neither weighting is the correct one. The suppression figure counts withheld cells and cannot show what they hold. Counties below 100,000 residents are pooled by WONDER and do not appear as cells, and the bound on the share held in withheld cells is the largest value the threshold allows.

## Tables

### Table 1. Life expectancy is 1.12 years lower per ten-point step of the ADI within counties, against 0.84 years across all tracts

Slope of tract life expectancy at birth on the Area Deprivation Index national percentile, years lower per ten-point step, three specifications, deaths in 2010 to 2015

| Specification | Weights | Years lower per ten points | Lower limit of the 95% interval | Upper limit of the 95% interval | Clusters | Tracts |
| --- | --- | ---: | ---: | ---: | ---: | ---: |
| All tracts | Population-weighted | 0.84 | 0.75 | 0.92 | 51 | 77,936 |
| All tracts | Each tract once | 0.84 | 0.75 | 0.93 | 51 | 77,936 |
| Within states | Population-weighted | 0.92 | 0.78 | 1.05 | 51 | 77,936 |
| Within states | Each tract once | 0.91 | 0.77 | 1.05 | 51 | 77,936 |
| Within counties | Population-weighted | 1.12 | 1.07 | 1.16 | 2,890 | 77,722 |
| Within counties | Each tract once | 1.13 | 1.09 | 1.17 | 2,890 | 77,722 |

Note: Intervals use cluster-robust errors, clustered on states for the first two rows and on counties for the third, with t intervals on one fewer degree of freedom than there are clusters. The county fit leaves out tracts that are alone in their county, so it uses 77,722 of 77,936 tracts.

Note: Institutional tracts, such as prisons and dormitories, are dropped. USALEEP covers deaths in 2010 to 2015 and the 2020-tract file is area-weighted from 2010 tracts.

Source: USALEEP life expectancy, 2010-2015; Area Deprivation Index, 2023; ACS 2019-2023; population-weighted and unweighted regression slopes, modeled estimates

Data: https://vishalsingh.org/zipcode/report/gradients-slopes-and-variance/figures/fe_ladder.json

### Table 2. Life expectancy is lower in more deprived tracts in all 267 metros, by a median of 1.20 years per ten-point step

One regression of tract life expectancy on the ADI national percentile in each metro with enough tracts, shown by metro and as a distribution, deaths in 2010 to 2015

#### Table 2.1. Distribution

After shrinkage the 5th and 95th percentile metro slopes move from 0.70 and 1.67 years to 0.88 and 1.46

Number of metros by slope of life expectancy on the ADI, years lower per ten-point step, estimated and shrunken, in bins of equal width

| metros | series | x0 | x1 |
| ---: | --- | ---: | ---: |
| 1 | Estimated slope | 0 | 0.200 |
| 0 | Estimated slope | 0.200 | 0.400 |
| 3 | Estimated slope | 0.400 | 0.600 |
| 17 | Estimated slope | 0.600 | 0.800 |
| 48 | Estimated slope | 0.800 | 1 |
| 63 | Estimated slope | 1 | 1.20 |
| 76 | Estimated slope | 1.20 | 1.40 |
| 38 | Estimated slope | 1.40 | 1.60 |
| 14 | Estimated slope | 1.60 | 1.80 |
| 3 | Estimated slope | 1.80 | 2 |
| 2 | Estimated slope | 2 | 2.20 |
| 2 | Estimated slope | 2.20 | 2.40 |
| 0 | Estimated slope | 2.40 | 2.60 |
| 0 | Estimated slope | 2.60 | 2.80 |
| 0 | Shrunken slope | 0 | 0.200 |
| 0 | Shrunken slope | 0.200 | 0.400 |
| 1 | Shrunken slope | 0.400 | 0.600 |
| 3 | Shrunken slope | 0.600 | 0.800 |
| 30 | Shrunken slope | 0.800 | 1 |
| 104 | Shrunken slope | 1 | 1.20 |
| 103 | Shrunken slope | 1.20 | 1.40 |
| 25 | Shrunken slope | 1.40 | 1.60 |
| 1 | Shrunken slope | 1.60 | 1.80 |
| 0 | Shrunken slope | 1.80 | 2 |
| 0 | Shrunken slope | 2 | 2.20 |
| 0 | Shrunken slope | 2.20 | 2.40 |
| 0 | Shrunken slope | 2.40 | 2.60 |
| 0 | Shrunken slope | 2.60 | 2.80 |

#### Table 2.2. Slope by metro

Life expectancy is lower in more deprived tracts in every metro, from 0.70 years per ten-point step at the 5th percentile of metros to 1.67 at the 95th

Slope of tract life expectancy at birth on the ADI national percentile within each metro, years lower per ten-point step, one dot per metro labeled by its most populous county, deaths in 2010 to 2015

| conf_high | conf_low | errors | estimate | metro | n |
| ---: | ---: | --- | ---: | --- | ---: |
| 1.46 | 1.09 | HC3 errors | 1.27 | Summit, OH | 167 |
| 1.73 | 0.711 | HC3 errors | 1.22 | Dougherty, GA | 43 |
| 1.55 | 0.922 | Clustered on counties | 1.24 | Albany, NY | 232 |
| 1.17 | 0.770 | HC3 errors | 0.968 | Bernalillo, NM | 223 |
| 1.90 | 0.894 | HC3 errors | 1.39 | Rapides, LA | 42 |
| 1.13 | 0.802 | HC3 errors | 0.965 | Lehigh, PA | 188 |
| 1.30 | 0.637 | Clustered on counties | 0.970 | Randall, TX | 64 |
| 1.84 | 1.07 | HC3 errors | 1.45 | Anchorage, AK | 79 |
| 1.71 | 1.15 | HC3 errors | 1.43 | Washtenaw, MI | 86 |
| 1.63 | 0.507 | HC3 errors | 1.07 | Outagamie, WI | 52 |
| 0.934 | 0.363 | HC3 errors | 0.649 | Buncombe, NC | 112 |
| 1.59 | 0.881 | HC3 errors | 1.23 | Clarke, GA | 53 |
| 1.19 | 0.949 | Clustered on counties | 1.07 | Fulton, GA | 1,354 |
| 1.18 | 0.542 | HC3 errors | 0.859 | Atlantic, NJ | 69 |
| 1.06 | −0.236 | HC3 errors | 0.411 | Lee, AL | 43 |
| 1.26 | 0.917 | Clustered on counties | 1.09 | Richmond, GA | 134 |
| 0.846 | 0.511 | Clustered on counties | 0.679 | Travis, TX | 442 |
| 1.16 | 0.751 | HC3 errors | 0.957 | Kern, CA | 212 |
| 1.67 | 1.14 | Clustered on counties | 1.40 | Baltimore, MD | 670 |
| 2.99 | 0.846 | HC3 errors | 1.92 | Barnstable, MA | 52 |
| 1.87 | 0.796 | Clustered on counties | 1.33 | East Baton Rouge, LA | 192 |
| 1.49 | 0.976 | HC3 errors | 1.23 | Jefferson, TX | 105 |
| 1.60 | 0.018 | HC3 errors | 0.807 | Whatcom, WA | 50 |
| 2.18 | 0.665 | HC3 errors | 1.42 | Deschutes, OR | 42 |
| 2.00 | 0.799 | HC3 errors | 1.40 | Yellowstone, MT | 42 |
| 1.88 | 0.700 | HC3 errors | 1.29 | Broome, NY | 64 |
| 1.45 | 1.17 | Clustered on counties | 1.31 | Jefferson, AL | 289 |
| 1.63 | 0.624 | HC3 errors | 1.13 | McLean, IL | 40 |
| 1.17 | 0.843 | Clustered on counties | 1.00 | Ada, ID | 190 |
| 1.80 | 0.798 | Clustered on counties | 1.30 | Middlesex, MA | 1,045 |
| 1.98 | 0.652 | HC3 errors | 1.31 | Boulder, CO | 64 |
| 2.98 | 1.76 | HC3 errors | 2.37 | Kitsap, WA | 53 |
| 1.28 | 0.983 | HC3 errors | 1.13 | Western Connecticut, CT | 212 |
| 1.73 | 1.00 | HC3 errors | 1.36 | Cameron, TX | 118 |
| 1.49 | 1.17 | HC3 errors | 1.33 | Erie, NY | 307 |
| 1.76 | 0.633 | HC3 errors | 1.20 | Chittenden, VT | 50 |
| 1.84 | 1.31 | HC3 errors | 1.58 | Stark, OH | 95 |
| 1.02 | 0.583 | HC3 errors | 0.799 | Lee, FL | 190 |
| 1.63 | 0.653 | HC3 errors | 1.14 | Linn, IA | 65 |
| 2.34 | 1.40 | HC3 errors | 1.87 | Champaign, IL | 43 |
| 1.29 | 0.654 | Clustered on counties | 0.972 | Kanawha, WV | 76 |
| 1.06 | 0.708 | HC3 errors | 0.884 | Charleston, SC | 159 |
| 1.28 | 0.986 | Clustered on counties | 1.13 | Mecklenburg, NC | 599 |
| 1.76 | 0.435 | Clustered on counties | 1.10 | Albemarle, VA | 51 |
| 1.60 | 1.01 | Clustered on counties | 1.31 | Hamilton, TN | 130 |
| 1.23 | 1.06 | Clustered on counties | 1.15 | Cook, IL | 2,183 |
| 2.11 | 1.14 | HC3 errors | 1.63 | Butte, CA | 51 |
| 1.69 | 1.45 | Clustered on counties | 1.57 | Hamilton, OH | 522 |
| 1.22 | 0.711 | HC3 errors | 0.964 | Montgomery, TN | 52 |
| 1.46 | 1.19 | Clustered on counties | 1.33 | Cuyahoga, OH | 589 |
| 1.16 | 0.549 | HC3 errors | 0.854 | Brazos, TX | 60 |
| 1.81 | 1.09 | HC3 errors | 1.45 | El Paso, CO | 155 |
| 1.46 | 0.677 | HC3 errors | 1.07 | Boone, MO | 40 |
| 1.33 | 0.942 | Clustered on counties | 1.14 | Richland, SC | 205 |
| 1.53 | 1.34 | Clustered on counties | 1.43 | Muscogee, GA | 87 |
| 1.74 | 0.972 | Clustered on counties | 1.36 | Franklin, OH | 474 |
| 1.18 | 0.518 | HC3 errors | 0.848 | Nueces, TX | 103 |
| 1.33 | 0.752 | HC3 errors | 1.04 | Okaloosa, FL | 63 |
| 0.986 | 0.834 | Clustered on counties | 0.910 | Dallas, TX | 1,501 |
| 1.36 | 0.726 | HC3 errors | 1.04 | Baldwin, AL | 41 |
| 1.51 | 0.487 | HC3 errors | 0.996 | Scott, IA | 107 |
| 1.87 | 1.45 | HC3 errors | 1.66 | Montgomery, OH | 209 |
| 1.57 | 0.953 | HC3 errors | 1.26 | Volusia, FL | 143 |
| 1.43 | 0.939 | Clustered on counties | 1.19 | Denver, CO | 635 |
| 1.38 | 0.681 | Clustered on counties | 1.03 | Polk, IA | 154 |
| 1.40 | 1.20 | Clustered on counties | 1.30 | Wayne, MI | 1,238 |
| 1.51 | 0.670 | HC3 errors | 1.09 | St. Louis, MN | 93 |
| 1.75 | 0.308 | Clustered on counties | 1.03 | Durham, NC | 131 |
| 1.79 | −0.045 | HC3 errors | 0.871 | Monroe, PA | 46 |
| 2.17 | 1.36 | HC3 errors | 1.77 | Elkhart, IN | 44 |
| 0.851 | 0.416 | HC3 errors | 0.633 | El Paso, TX | 175 |
| 2.08 | 0.935 | HC3 errors | 1.50 | Erie, PA | 65 |
| 1.91 | 0.820 | HC3 errors | 1.36 | Lane, OR | 86 |
| 2.13 | 1.46 | HC3 errors | 1.80 | Vanderburgh, IN | 88 |
| 2.06 | 1.28 | HC3 errors | 1.67 | Cass, ND | 58 |
| 1.65 | 0.765 | HC3 errors | 1.21 | Cumberland, NC | 107 |
| 0.985 | 0.401 | HC3 errors | 0.693 | Benton, AR | 109 |
| 1.57 | 1.07 | HC3 errors | 1.32 | Genesee, MI | 124 |
| 1.62 | 0.646 | HC3 errors | 1.13 | Florence, SC | 52 |
| 2.02 | 0.669 | HC3 errors | 1.34 | Lauderdale, AL | 40 |
| 2.56 | −0.146 | HC3 errors | 1.21 | Larimer, CO | 76 |
| 1.79 | 0.738 | HC3 errors | 1.27 | Sebastian, AR | 63 |
| 1.96 | 1.20 | HC3 errors | 1.58 | Allen, IN | 100 |
| 1.26 | 0.827 | HC3 errors | 1.04 | Fresno, CA | 218 |
| 1.77 | 1.21 | HC3 errors | 1.49 | Alachua, FL | 60 |
| 1.19 | 0.487 | HC3 errors | 0.840 | Hall, GA | 48 |
| 1.26 | 0.142 | HC3 errors | 0.701 | Warren, NY | 42 |
| 1.24 | 0.877 | HC3 errors | 1.06 | Kent, MI | 228 |
| 1.53 | 0.618 | HC3 errors | 1.07 | Weld, CO | 69 |
| 1.57 | 0.572 | HC3 errors | 1.07 | Brown, WI | 69 |
| 1.64 | 1.20 | HC3 errors | 1.42 | Guilford, NC | 169 |
| 1.36 | 0.988 | HC3 errors | 1.18 | Greenville, SC | 221 |
| 1.26 | 0.390 | HC3 errors | 0.827 | Harrison, MS | 137 |
| 1.98 | 1.13 | HC3 errors | 1.56 | Washington, MD | 65 |
| 1.55 | 0.981 | HC3 errors | 1.27 | Dauphin, PA | 126 |
| 1.29 | 0.997 | HC3 errors | 1.15 | Capitol, CT | 272 |
| 1.66 | 0.607 | HC3 errors | 1.13 | Forrest, MS | 45 |
| 1.36 | 0.395 | HC3 errors | 0.879 | Catawba, NC | 78 |
| 1.52 | 0.392 | HC3 errors | 0.955 | Hawaii, HI | 43 |
| 1.36 | 0.503 | HC3 errors | 0.930 | Beaufort, SC | 42 |
| 1.51 | 0.360 | HC3 errors | 0.936 | Citrus, FL | 41 |
| 1.13 | 0.623 | HC3 errors | 0.874 | Terrebonne, LA | 59 |
| 0.892 | 0.758 | Clustered on counties | 0.825 | Harris, TX | 1,457 |
| 2.12 | 0.348 | Clustered on counties | 1.24 | Cabell, WV | 102 |
| 1.53 | 1.08 | HC3 errors | 1.30 | Madison, AL | 108 |
| 1.48 | 1.18 | Clustered on counties | 1.33 | Marion, IN | 453 |
| 1.59 | 0.656 | HC3 errors | 1.12 | Jackson, MI | 45 |
| 1.50 | 1.00 | Clustered on counties | 1.25 | Hinds, MS | 149 |
| 1.83 | 0.819 | HC3 errors | 1.32 | Madison, TN | 50 |
| 1.53 | 1.23 | Clustered on counties | 1.38 | Duval, FL | 319 |
| 1.47 | 0.562 | HC3 errors | 1.02 | Washington, TN | 53 |
| 2.26 | 0.816 | HC3 errors | 1.54 | Jasper, MO | 42 |
| 2.53 | 0.560 | HC3 errors | 1.54 | Maui, HI | 42 |
| 1.55 | 0.897 | HC3 errors | 1.22 | Kalamazoo, MI | 63 |
| 1.50 | 1.05 | Clustered on counties | 1.27 | Jackson, MO | 556 |
| 1.21 | 0.656 | HC3 errors | 0.934 | Benton, WA | 63 |
| 1.00 | 0.383 | HC3 errors | 0.694 | Bell, TX | 91 |
| 2.04 | 0.991 | Clustered on counties | 1.51 | Sullivan, TN | 77 |
| 1.88 | 0.315 | HC3 errors | 1.10 | Ulster, NY | 41 |
| 1.59 | 1.29 | Clustered on counties | 1.44 | Knox, TN | 198 |
| 1.44 | 0.826 | Clustered on counties | 1.13 | Lafayette, LA | 125 |
| 1.98 | 0.909 | HC3 errors | 1.44 | Tippecanoe, IN | 49 |
| 1.48 | 0.741 | HC3 errors | 1.11 | Calcasieu, LA | 52 |
| 1.29 | 0.490 | HC3 errors | 0.892 | Mohave, AZ | 55 |
| 1.51 | 0.881 | HC3 errors | 1.20 | Polk, FL | 162 |
| 1.57 | 0.750 | HC3 errors | 1.16 | Lancaster, PA | 112 |
| 1.33 | 0.912 | HC3 errors | 1.12 | Ingham, MI | 135 |
| 1.34 | 0.641 | HC3 errors | 0.991 | Webb, TX | 61 |
| 0.970 | −0.013 | HC3 errors | 0.478 | Doña Ana, NM | 57 |
| 1.03 | 0.735 | HC3 errors | 0.885 | Clark, NV | 475 |
| 1.53 | 0.543 | HC3 errors | 1.03 | Grafton, NH | 66 |
| 1.84 | 0.989 | Clustered on counties | 1.41 | Fayette, KY | 124 |
| 1.65 | 0.994 | HC3 errors | 1.32 | Lancaster, NE | 74 |
| 1.82 | 0.949 | Clustered on counties | 1.38 | Pulaski, AR | 177 |
| 1.98 | 0.429 | HC3 errors | 1.20 | Laurel, KY | 43 |
| 1.02 | 0.470 | HC3 errors | 0.744 | Gregg, TX | 64 |
| 1.45 | 1.15 | HC3 errors | 1.30 | Los Angeles, CA | 2,923 |
| 1.73 | 1.29 | Clustered on counties | 1.51 | Jefferson, KY | 320 |
| 1.25 | 0.097 | HC3 errors | 0.675 | Lubbock, TX | 97 |
| 1.90 | 0.911 | Clustered on counties | 1.40 | Bedford, VA | 64 |
| 1.68 | 1.12 | Clustered on counties | 1.40 | Bibb, GA | 64 |
| 1.15 | 0.419 | HC3 errors | 0.782 | Dane, WI | 123 |
| 1.55 | 0.681 | HC3 errors | 1.12 | Hillsborough, NH | 99 |
| 0.888 | 0.374 | HC3 errors | 0.631 | Hidalgo, TX | 208 |
| 1.83 | 0.570 | HC3 errors | 1.20 | Jackson, OR | 51 |
| 1.45 | 1.12 | Clustered on counties | 1.29 | Shelby, TN | 316 |
| 1.32 | 0.358 | HC3 errors | 0.839 | Merced, CA | 60 |
| 0.922 | 0.725 | HC3 errors | 0.823 | Miami-Dade, FL | 1,377 |
| 1.32 | 1.08 | HC3 errors | 1.20 | Milwaukee, WI | 374 |
| 1.52 | 0.938 | Clustered on counties | 1.23 | Hennepin, MN | 847 |
| 1.61 | 1.02 | HC3 errors | 1.31 | Mobile, AL | 128 |
| 1.35 | 0.694 | HC3 errors | 1.02 | Stanislaus, CA | 111 |
| 1.93 | 1.04 | HC3 errors | 1.49 | Ouachita, LA | 57 |
| 2.14 | 1.11 | HC3 errors | 1.62 | Monroe, MI | 40 |
| 1.49 | 0.964 | HC3 errors | 1.23 | Montgomery, AL | 94 |
| 1.98 | 0.493 | HC3 errors | 1.24 | Skagit, WA | 41 |
| 1.92 | 1.19 | HC3 errors | 1.56 | Muskegon, MI | 44 |
| 1.15 | 0.593 | HC3 errors | 0.872 | Horry, SC | 110 |
| 1.38 | 0.423 | HC3 errors | 0.902 | Collier, FL | 94 |
| 1.55 | 0.985 | Clustered on counties | 1.27 | Davidson, TN | 384 |
| 1.31 | 0.818 | HC3 errors | 1.06 | South Central Connecticut, CT | 128 |
| 1.38 | 0.948 | Clustered on counties | 1.16 | Jefferson, LA | 379 |
| 1.12 | 0.421 | Clustered on counties | 0.771 | Kings, NY | 4,563 |
| 1.92 | 1.12 | HC3 errors | 1.52 | Berrien, MI | 51 |
| 1.19 | 0.714 | HC3 errors | 0.953 | Sarasota, FL | 204 |
| 1.66 | 0.818 | HC3 errors | 1.24 | Southeastern Connecticut, CT | 62 |
| 1.27 | 0.541 | HC3 errors | 0.906 | Marion, FL | 77 |
| 1.96 | 1.32 | HC3 errors | 1.64 | Davis, UT | 134 |
| 1.49 | 1.11 | Clustered on counties | 1.30 | Oklahoma, OK | 380 |
| 2.18 | 1.21 | HC3 errors | 1.69 | Thurston, WA | 60 |
| 1.31 | 0.980 | Clustered on counties | 1.14 | Douglas, NE | 247 |
| 1.03 | 0.689 | HC3 errors | 0.859 | Orange, FL | 460 |
| 1.49 | 0.593 | HC3 errors | 1.04 | Ventura, CA | 182 |
| 1.47 | 1.01 | HC3 errors | 1.24 | Brevard, FL | 140 |
| 1.23 | 0.379 | HC3 errors | 0.803 | Bay, FL | 46 |
| 1.73 | 1.17 | HC3 errors | 1.45 | Escambia, FL | 106 |
| 1.41 | 1.07 | Clustered on counties | 1.24 | Peoria, IL | 104 |
| 1.27 | 1.13 | Clustered on counties | 1.20 | Philadelphia, PA | 1,455 |
| 0.989 | 0.800 | HC3 errors | 0.894 | Maricopa, AZ | 948 |
| 1.21 | 0.893 | Clustered on counties | 1.05 | Allegheny, PA | 674 |
| 1.45 | 0.677 | HC3 errors | 1.06 | Cumberland, ME | 127 |
| 1.51 | 1.21 | Clustered on counties | 1.36 | Multnomah, OR | 537 |
| 1.67 | 1.00 | HC3 errors | 1.34 | St. Lucie, FL | 102 |
| 2.22 | 1.04 | HC3 errors | 1.63 | Schuylkill, PA | 40 |
| 1.02 | 0.313 | HC3 errors | 0.666 | Orange, NY | 162 |
| 1.26 | 0.223 | HC3 errors | 0.740 | Yavapai, AZ | 70 |
| 1.46 | 1.07 | Clustered on counties | 1.26 | Providence, RI | 357 |
| 1.53 | 0.768 | HC3 errors | 1.15 | Utah, UT | 140 |
| 1.79 | 1.14 | HC3 errors | 1.47 | Pueblo, CO | 55 |
| 1.58 | 0.643 | HC3 errors | 1.11 | Charlotte, FL | 40 |
| 1.42 | 0.528 | HC3 errors | 0.976 | Racine, WI | 41 |
| 1.04 | 0.745 | HC3 errors | 0.890 | Wake, NC | 268 |
| 1.39 | 0.967 | HC3 errors | 1.18 | Berks, PA | 103 |
| 1.73 | 0.732 | HC3 errors | 1.23 | Shasta, CA | 49 |
| 1.67 | 0.812 | HC3 errors | 1.24 | Washoe, NV | 117 |
| 1.36 | 1.12 | Clustered on counties | 1.24 | Chesterfield, VA | 298 |
| 1.15 | 0.890 | HC3 errors | 1.02 | Riverside, CA | 939 |
| 1.82 | 0.799 | Clustered on counties | 1.31 | Roanoke, VA | 75 |
| 1.41 | 0.373 | HC3 errors | 0.892 | Olmsted, MN | 52 |
| 1.39 | 1.05 | Clustered on counties | 1.22 | Monroe, NY | 288 |
| 1.91 | 1.01 | HC3 errors | 1.46 | Winnebago, IL | 88 |
| 1.61 | 1.27 | HC3 errors | 1.44 | Sacramento, CA | 519 |
| 2.09 | 1.35 | HC3 errors | 1.72 | Saginaw, MI | 59 |
| 1.86 | 0.503 | HC3 errors | 1.18 | Stearns, MN | 43 |
| 1.43 | 1.07 | Clustered on counties | 1.25 | St. Louis, MO | 669 |
| 2.15 | 0.667 | HC3 errors | 1.41 | Marion, OR | 78 |
| 1.58 | 0.320 | HC3 errors | 0.950 | Monterey, CA | 96 |
| 1.21 | 0.759 | HC3 errors | 0.986 | Sussex, DE | 105 |
| 1.51 | 0.894 | HC3 errors | 1.20 | Salt Lake, UT | 247 |
| 1.05 | 0.647 | Clustered on counties | 0.849 | Bexar, TX | 453 |
| 1.34 | 0.826 | HC3 errors | 1.08 | San Diego, CA | 681 |
| 3.04 | 1.67 | Clustered on counties | 2.36 | Alameda, CA | 1,044 |
| 1.09 | 0.161 | HC3 errors | 0.624 | Santa Clara, CA | 405 |
| 1.48 | −0.040 | HC3 errors | 0.721 | San Luis Obispo, CA | 61 |
| 1.73 | −0.548 | HC3 errors | 0.592 | Santa Cruz, CA | 63 |
| 1.03 | 0.374 | HC3 errors | 0.703 | Santa Fe, NM | 52 |
| 1.64 | 0.787 | HC3 errors | 1.22 | Santa Barbara, CA | 98 |
| 2.12 | 0.547 | HC3 errors | 1.34 | Sonoma, CA | 116 |
| 1.28 | 0.806 | HC3 errors | 1.04 | Chatham, GA | 101 |
| 1.64 | 1.08 | HC3 errors | 1.36 | Luzerne, PA | 162 |
| 1.79 | 1.34 | HC3 errors | 1.56 | King, WA | 829 |
| 1.27 | 0.555 | HC3 errors | 0.914 | Indian River, FL | 41 |
| 1.51 | 1.01 | HC3 errors | 1.26 | Caddo, LA | 114 |
| 1.87 | 0.709 | HC3 errors | 1.29 | Minnehaha, SD | 60 |
| 1.51 | 1.05 | HC3 errors | 1.28 | St. Joseph, IN | 92 |
| 1.63 | 0.905 | HC3 errors | 1.27 | Spartanburg, SC | 85 |
| 1.84 | 1.26 | HC3 errors | 1.55 | Spokane, WA | 140 |
| 2.00 | 1.14 | HC3 errors | 1.57 | Sangamon, IL | 56 |
| 1.67 | 1.27 | HC3 errors | 1.47 | Hampden, MA | 151 |
| 2.08 | 0.848 | Clustered on counties | 1.46 | Greene, MO | 111 |
| 2.70 | 1.57 | HC3 errors | 2.13 | Clark, OH | 41 |
| 1.76 | 1.15 | HC3 errors | 1.46 | San Joaquin, CA | 169 |
| 1.43 | 0.879 | HC3 errors | 1.15 | Onondaga, NY | 183 |
| 1.22 | 0.803 | HC3 errors | 1.01 | Leon, FL | 82 |
| 1.10 | 0.888 | HC3 errors | 0.995 | Hillsborough, FL | 692 |
| 2.40 | 1.28 | Clustered on counties | 1.84 | Vigo, IN | 53 |
| 1.64 | 0.254 | HC3 errors | 0.949 | Bowie, TX | 42 |
| 1.42 | 1.05 | HC3 errors | 1.23 | Lucas, OH | 213 |
| 1.32 | 1.12 | Clustered on counties | 1.22 | Shawnee, KS | 59 |
| 1.42 | 0.785 | HC3 errors | 1.10 | Grand Traverse, MI | 43 |
| 1.58 | 1.11 | HC3 errors | 1.34 | Mercer, NJ | 79 |
| 1.30 | 0.873 | HC3 errors | 1.08 | Pima, AZ | 246 |
| 1.48 | 0.994 | Clustered on counties | 1.24 | Tulsa, OK | 304 |
| 1.84 | 0.561 | HC3 errors | 1.20 | Lee, MS | 40 |
| 1.28 | 0.561 | HC3 errors | 0.920 | Tuscaloosa, AL | 69 |
| 1.45 | 0.295 | HC3 errors | 0.875 | Smith, TX | 53 |
| 1.57 | 0.764 | HC3 errors | 1.16 | Honolulu, HI | 286 |
| 2.16 | 1.08 | HC3 errors | 1.62 | Oneida, NY | 83 |
| 2.86 | 1.47 | HC3 errors | 2.16 | Solano, CA | 96 |
| 1.68 | 1.04 | Clustered on counties | 1.36 | Virginia Beach, VA | 436 |
| 1.44 | 0.731 | HC3 errors | 1.08 | Tulare, CA | 101 |
| 1.71 | 0.963 | HC3 errors | 1.34 | McLennan, TX | 55 |
| 1.37 | 0.235 | HC3 errors | 0.804 | Houston, GA | 44 |
| 1.34 | 0.861 | Clustered on counties | 1.10 | Fairfax, VA | 1,350 |
| 1.34 | 0.879 | HC3 errors | 1.11 | Naugatuck Valley, CT | 95 |
| 1.67 | 0.512 | HC3 errors | 1.09 | Black Hawk, IA | 46 |
| 2.51 | 0.881 | HC3 errors | 1.69 | Marathon, WI | 41 |
| 2.19 | 1.04 | HC3 errors | 1.61 | Belmont, OH | 48 |
| 1.51 | 1.07 | HC3 errors | 1.29 | Sedgwick, KS | 164 |
| 2.15 | 0.833 | HC3 errors | 1.49 | Wichita, TX | 44 |
| 1.27 | 0.688 | HC3 errors | 0.978 | New Hanover, NC | 61 |
| 1.59 | 0.891 | Clustered on counties | 1.24 | Forsyth, NC | 161 |
| 1.24 | 0.783 | HC3 errors | 1.01 | Worcester, MA | 184 |
| 1.41 | 0.812 | HC3 errors | 1.11 | Yakima, WA | 56 |
| 1.75 | 1.10 | HC3 errors | 1.43 | York, PA | 105 |
| 2.14 | 1.27 | HC3 errors | 1.70 | Mahoning, OH | 155 |
| 0.485 | −0.212 | HC3 errors | 0.137 | Yuma, AZ | 60 |

Source: USALEEP life expectancy, 2010-2015; Area Deprivation Index, 2023; ACS 2019-2023; population-weighted regression slopes, modeled estimates

Data: https://vishalsingh.org/zipcode/report/gradients-slopes-and-variance/figures/metro_slopes_forest.json

### Table 3. Metro slopes are not detectably related to metro median income, a correlation of +0.05 counting each metro once

Slope of life expectancy on the ADI, years lower per ten-point step, against the population-weighted median of tract median household income, one dot per metro, dot area is residents, ACS 2019-2023

| Metro (most populous county) | Median household income (dollars) | Years lower per ten points | Residents |
| --- | ---: | ---: | ---: |
| Summit, OH | 78,539 | 1.27 | 669,646 |
| Dougherty, GA | 58,616 | 1.22 | 137,062 |
| Albany, NY | 90,562 | 1.24 | 836,414 |
| Bernalillo, NM | 67,143 | 0.97 | 893,158 |
| Rapides, LA | 55,364 | 1.40 | 140,716 |
| Lehigh, PA | 84,487 | 0.96 | 833,485 |
| Randall, TX | 68,494 | 0.97 | 237,500 |
| Anchorage, AK | 98,834 | 1.45 | 380,453 |
| Washtenaw, MI | 96,960 | 1.43 | 304,946 |
| Outagamie, WI | 79,222 | 1.07 | 219,811 |
| Buncombe, NC | 68,432 | 0.65 | 441,184 |
| Clarke, GA | 69,709 | 1.23 | 199,046 |
| Fulton, GA | 87,864 | 1.07 | 5,630,116 |
| Atlantic, NJ | 82,927 | 0.86 | 268,098 |
| Lee, AL | 70,429 | 0.41 | 164,528 |
| Richmond, GA | 69,885 | 1.09 | 574,113 |
| Travis, TX | 100,981 | 0.68 | 2,125,441 |
| Kern, CA | 65,661 | 0.96 | 825,064 |
| Baltimore, MD | 105,865 | 1.40 | 2,720,602 |
| Barnstable, MA | 95,323 | 1.92 | 216,294 |
| East Baton Rouge, LA | 72,773 | 1.33 | 773,655 |
| Jefferson, TX | 69,211 | 1.23 | 370,574 |
| Whatcom, WA | 86,204 | 0.81 | 209,575 |
| Deschutes, OR | 87,928 | 1.42 | 186,931 |
| Yellowstone, MT | 74,303 | 1.40 | 180,017 |
| Broome, NY | 69,856 | 1.29 | 229,131 |
| Jefferson, AL | 73,171 | 1.31 | 1,073,001 |
| McLean, IL | 93,924 | 1.13 | 146,591 |
| Ada, ID | 82,830 | 1.00 | 770,927 |
| Middlesex, MA | 115,783 | 1.30 | 4,655,005 |
| Boulder, CO | 111,280 | 1.31 | 269,739 |
| Kitsap, WA | 96,978 | 2.37 | 254,129 |
| Western Connecticut, CT | 119,204 | 1.13 | 906,110 |
| Cameron, TX | 49,397 | 1.36 | 420,423 |
| Erie, NY | 74,430 | 1.33 | 1,114,712 |
| Chittenden, VT | 97,573 | 1.20 | 212,127 |
| Stark, OH | 67,916 | 1.58 | 393,729 |
| Lee, FL | 73,765 | 0.80 | 706,110 |
| Linn, IA | 79,604 | 1.14 | 267,683 |
| Champaign, IL | 78,227 | 1.87 | 189,134 |
| Kanawha, WV | 56,842 | 0.97 | 240,764 |
| Charleston, SC | 84,750 | 0.88 | 755,397 |
| Mecklenburg, NC | 79,411 | 1.13 | 2,441,459 |
| Albemarle, VA | 99,461 | 1.10 | 205,622 |
| Hamilton, TN | 66,386 | 1.31 | 550,749 |
| Cook, IL | 89,535 | 1.15 | 9,078,307 |
| Butte, CA | 66,437 | 1.63 | 199,170 |
| Hamilton, OH | 83,974 | 1.56 | 2,187,695 |
| Montgomery, TN | 69,185 | 0.96 | 216,994 |
| Cuyahoga, OH | 75,421 | 1.32 | 1,974,832 |
| Brazos, TX | 64,066 | 0.85 | 215,331 |
| El Paso, CO | 90,786 | 1.45 | 683,216 |
| Boone, MO | 75,882 | 1.07 | 177,863 |
| Richland, SC | 68,108 | 1.14 | 774,726 |
| Muscogee, GA | 56,087 | 1.44 | 307,364 |
| Franklin, OH | 84,228 | 1.36 | 1,996,447 |
| Nueces, TX | 62,659 | 0.85 | 400,651 |
| Okaloosa, FL | 81,407 | 1.04 | 282,957 |
| Dallas, TX | 89,919 | 0.91 | 6,987,097 |
| Baldwin, AL | 72,143 | 1.04 | 233,001 |
| Scott, IA | 73,210 | 1.00 | 371,800 |
| Montgomery, OH | 71,585 | 1.66 | 767,209 |
| Volusia, FL | 68,729 | 1.26 | 654,841 |
| Denver, CO | 103,222 | 1.19 | 2,715,379 |
| Polk, IA | 88,621 | 1.03 | 699,409 |
| Wayne, MI | 78,215 | 1.30 | 4,220,000 |
| St. Louis, MN | 73,474 | 1.09 | 269,181 |
| Durham, NC | 80,667 | 1.03 | 577,271 |
| Monroe, PA | 82,889 | 0.87 | 150,264 |
| Elkhart, IN | 74,014 | 1.77 | 202,494 |
| El Paso, TX | 59,731 | 0.63 | 815,804 |
| Erie, PA | 68,900 | 1.50 | 241,679 |
| Lane, OR | 75,900 | 1.36 | 356,564 |
| Vanderburgh, IN | 70,479 | 1.80 | 307,015 |
| Cass, ND | 81,328 | 1.67 | 245,120 |
| Cumberland, NC | 63,983 | 1.21 | 481,759 |
| Benton, AR | 79,398 | 0.69 | 532,009 |
| Genesee, MI | 67,173 | 1.32 | 386,955 |
| Florence, SC | 51,228 | 1.13 | 194,799 |
| Lauderdale, AL | 62,256 | 1.34 | 150,584 |
| Larimer, CO | 96,086 | 1.21 | 335,193 |
| Sebastian, AR | 57,749 | 1.27 | 246,136 |
| Allen, IN | 74,450 | 1.58 | 419,042 |
| Fresno, CA | 68,789 | 1.04 | 972,223 |
| Alachua, FL | 65,453 | 1.49 | 274,469 |
| Hall, GA | 78,409 | 0.84 | 197,646 |
| Warren, NY | 73,926 | 0.70 | 120,895 |
| Kent, MI | 79,418 | 1.06 | 1,042,581 |
| Weld, CO | 91,034 | 1.07 | 303,989 |
| Brown, WI | 78,555 | 1.07 | 314,505 |
| Guilford, NC | 64,720 | 1.42 | 734,908 |
| Greenville, SC | 67,109 | 1.18 | 907,427 |
| Harrison, MS | 63,333 | 0.83 | 403,294 |
| Washington, MD | 81,677 | 1.56 | 285,382 |
| Dauphin, PA | 80,314 | 1.27 | 577,920 |
| Capitol, CT | 101,384 | 1.15 | 1,084,341 |
| Forrest, MS | 57,582 | 1.13 | 168,319 |
| Catawba, NC | 60,922 | 0.88 | 349,670 |
| Hawaii, HI | 82,193 | 0.95 | 203,684 |
| Beaufort, SC | 80,222 | 0.93 | 208,704 |
| Citrus, FL | 52,582 | 0.94 | 158,693 |
| Terrebonne, LA | 66,584 | 0.87 | 201,316 |
| Harris, TX | 82,729 | 0.82 | 6,642,463 |
| Cabell, WV | 59,798 | 1.24 | 342,756 |
| Madison, AL | 90,881 | 1.30 | 458,766 |
| Marion, IN | 82,707 | 1.33 | 2,032,367 |
| Jackson, MI | 73,094 | 1.12 | 153,400 |
| Hinds, MS | 63,404 | 1.25 | 552,763 |
| Madison, TN | 60,270 | 1.32 | 173,169 |
| Duval, FL | 79,227 | 1.38 | 1,564,210 |
| Washington, TN | 55,932 | 1.02 | 207,033 |
| Jasper, MO | 64,575 | 1.54 | 183,022 |
| Maui, HI | 99,412 | 1.54 | 159,226 |
| Kalamazoo, MI | 76,432 | 1.22 | 245,459 |
| Jackson, MO | 84,494 | 1.27 | 2,096,382 |
| Benton, WA | 84,220 | 0.93 | 299,764 |
| Bell, TX | 69,611 | 0.69 | 405,407 |
| Sullivan, TN | 55,583 | 1.51 | 300,116 |
| Ulster, NY | 84,187 | 1.10 | 142,011 |
| Knox, TN | 72,138 | 1.44 | 809,385 |
| Lafayette, LA | 66,671 | 1.13 | 448,515 |
| Tippecanoe, IN | 66,937 | 1.44 | 195,160 |
| Calcasieu, LA | 66,533 | 1.11 | 202,151 |
| Mohave, AZ | 53,143 | 0.89 | 184,833 |
| Polk, FL | 65,270 | 1.20 | 714,860 |
| Lancaster, PA | 85,303 | 1.16 | 537,447 |
| Ingham, MI | 73,858 | 1.12 | 491,538 |
| Webb, TX | 56,078 | 0.99 | 245,692 |
| Doña Ana, NM | 52,415 | 0.48 | 210,924 |
| Clark, NV | 76,382 | 0.89 | 2,040,360 |
| Grafton, NH | 80,713 | 1.03 | 214,011 |
| Fayette, KY | 74,213 | 1.41 | 469,966 |
| Lancaster, NE | 78,349 | 1.32 | 309,128 |
| Pulaski, AR | 66,246 | 1.39 | 705,921 |
| Laurel, KY | 48,033 | 1.20 | 135,185 |
| Gregg, TX | 66,104 | 0.74 | 264,952 |
| Los Angeles, CA | 93,203 | 1.30 | 12,424,893 |
| Jefferson, KY | 75,097 | 1.51 | 1,237,206 |
| Lubbock, TX | 67,469 | 0.68 | 281,286 |
| Bedford, VA | 68,254 | 1.40 | 246,474 |
| Bibb, GA | 63,730 | 1.40 | 218,037 |
| Dane, WI | 95,609 | 0.78 | 572,839 |
| Hillsborough, NH | 111,157 | 1.12 | 406,867 |
| Hidalgo, TX | 51,427 | 0.63 | 862,862 |
| Jackson, OR | 69,562 | 1.20 | 219,490 |
| Shelby, TN | 66,331 | 1.29 | 1,235,221 |
| Merced, CA | 62,533 | 0.84 | 270,880 |
| Miami-Dade, FL | 74,414 | 0.82 | 5,791,746 |
| Milwaukee, WI | 82,816 | 1.20 | 1,378,567 |
| Hennepin, MN | 100,713 | 1.23 | 3,551,467 |
| Mobile, AL | 60,264 | 1.31 | 406,513 |
| Stanislaus, CA | 79,423 | 1.02 | 548,836 |
| Ouachita, LA | 53,701 | 1.49 | 194,368 |
| Monroe, MI | 73,561 | 1.62 | 155,001 |
| Montgomery, AL | 63,880 | 1.23 | 333,773 |
| Skagit, WA | 84,454 | 1.24 | 130,407 |
| Muskegon, MI | 65,723 | 1.56 | 172,022 |
| Horry, SC | 67,828 | 0.87 | 489,212 |
| Collier, FL | 86,133 | 0.90 | 346,920 |
| Davidson, TN | 83,658 | 1.27 | 1,838,601 |
| South Central Connecticut, CT | 93,307 | 1.06 | 534,415 |
| Jefferson, LA | 67,691 | 1.16 | 1,171,586 |
| Kings, NY | 100,747 | 0.77 | 18,857,008 |
| Berrien, MI | 68,611 | 1.52 | 146,726 |
| Sarasota, FL | 80,341 | 0.95 | 831,108 |
| Southeastern Connecticut, CT | 94,811 | 1.24 | 260,834 |
| Marion, FL | 61,597 | 0.91 | 333,673 |
| Davis, UT | 100,390 | 1.64 | 684,276 |
| Oklahoma, OK | 73,307 | 1.30 | 1,362,954 |
| Thurston, WA | 97,796 | 1.69 | 289,042 |
| Douglas, NE | 84,841 | 1.14 | 906,982 |
| Orange, FL | 78,256 | 0.86 | 2,563,054 |
| Ventura, CA | 104,381 | 1.04 | 814,047 |
| Brevard, FL | 77,672 | 1.24 | 605,284 |
| Bay, FL | 74,261 | 0.80 | 173,076 |
| Escambia, FL | 75,748 | 1.45 | 475,092 |
| Peoria, IL | 71,097 | 1.24 | 389,448 |
| Philadelphia, PA | 95,391 | 1.20 | 5,987,126 |
| Maricopa, AZ | 86,916 | 0.89 | 4,383,220 |
| Allegheny, PA | 76,103 | 1.05 | 2,255,125 |
| Cumberland, ME | 90,084 | 1.06 | 484,107 |
| Multnomah, OR | 96,267 | 1.36 | 2,415,437 |
| St. Lucie, FL | 73,760 | 1.33 | 492,192 |
| Schuylkill, PA | 66,594 | 1.63 | 134,044 |
| Orange, NY | 100,646 | 0.67 | 654,415 |
| Yavapai, AZ | 65,085 | 0.74 | 239,019 |
| Providence, RI | 89,419 | 1.26 | 1,583,138 |
| Utah, UT | 101,864 | 1.15 | 622,944 |
| Pueblo, CO | 67,283 | 1.47 | 165,819 |
| Charlotte, FL | 63,346 | 1.11 | 166,856 |
| Racine, WI | 79,997 | 0.98 | 176,535 |
| Wake, NC | 94,676 | 0.89 | 1,312,735 |
| Berks, PA | 84,157 | 1.18 | 415,678 |
| Shasta, CA | 73,870 | 1.23 | 177,622 |
| Washoe, NV | 91,819 | 1.24 | 436,487 |
| Chesterfield, VA | 87,203 | 1.24 | 1,220,266 |
| Riverside, CA | 86,902 | 1.02 | 4,445,399 |
| Roanoke, VA | 68,774 | 1.31 | 304,511 |
| Olmsted, MN | 90,266 | 0.89 | 227,252 |
| Monroe, NY | 75,298 | 1.22 | 1,017,611 |
| Winnebago, IL | 72,373 | 1.46 | 333,345 |
| Sacramento, CA | 95,171 | 1.44 | 2,275,221 |
| Saginaw, MI | 62,664 | 1.72 | 178,008 |
| Stearns, MN | 78,277 | 1.18 | 191,054 |
| St. Louis, MO | 80,401 | 1.25 | 2,698,897 |
| Marion, OR | 80,970 | 1.41 | 420,236 |
| Monterey, CA | 88,897 | 0.95 | 412,418 |
| Sussex, DE | 76,838 | 0.99 | 409,522 |
| Salt Lake, UT | 98,169 | 1.20 | 1,180,115 |
| Bexar, TX | 73,970 | 0.85 | 2,362,483 |
| San Diego, CA | 105,089 | 1.08 | 3,038,849 |
| Alameda, CA | 139,249 | 2.35 | 4,509,425 |
| Santa Clara, CA | 161,876 | 0.62 | 1,901,120 |
| San Luis Obispo, CA | 96,635 | 0.72 | 245,057 |
| Santa Cruz, CA | 119,621 | 0.59 | 237,745 |
| Santa Fe, NM | 73,619 | 0.70 | 148,272 |
| Santa Barbara, CA | 99,372 | 1.22 | 412,638 |
| Sonoma, CA | 102,512 | 1.34 | 464,875 |
| Chatham, GA | 80,415 | 1.04 | 386,024 |
| Luzerne, PA | 64,824 | 1.36 | 547,160 |
| King, WA | 116,777 | 1.56 | 3,901,567 |
| Indian River, FL | 69,670 | 0.91 | 159,904 |
| Caddo, LA | 58,198 | 1.26 | 387,561 |
| Minnehaha, SD | 81,909 | 1.29 | 259,327 |
| St. Joseph, IN | 66,265 | 1.28 | 306,731 |
| Spartanburg, SC | 68,253 | 1.26 | 332,585 |
| Spokane, WA | 73,703 | 1.55 | 571,186 |
| Sangamon, IL | 81,653 | 1.57 | 206,373 |
| Hampden, MA | 78,217 | 1.47 | 661,944 |
| Greene, MO | 67,133 | 1.46 | 460,198 |
| Clark, OH | 60,631 | 2.13 | 124,716 |
| San Joaquin, CA | 93,350 | 1.46 | 765,093 |
| Onondaga, NY | 78,968 | 1.15 | 571,929 |
| Leon, FL | 73,029 | 1.01 | 313,008 |
| Hillsborough, FL | 72,580 | 0.99 | 2,915,720 |
| Vigo, IN | 62,079 | 1.84 | 173,868 |
| Bowie, TX | 56,942 | 0.95 | 132,963 |
| Lucas, OH | 69,543 | 1.23 | 618,783 |
| Shawnee, KS | 72,654 | 1.22 | 230,593 |
| Grand Traverse, MI | 80,819 | 1.10 | 152,857 |
| Mercer, NJ | 99,710 | 1.34 | 365,389 |
| Pima, AZ | 68,805 | 1.08 | 970,791 |
| Tulsa, OK | 67,647 | 1.24 | 992,751 |
| Lee, MS | 60,211 | 1.20 | 154,901 |
| Tuscaloosa, AL | 66,848 | 0.92 | 245,890 |
| Smith, TX | 72,525 | 0.88 | 229,377 |
| Honolulu, HI | 114,427 | 1.17 | 920,726 |
| Oneida, NY | 73,753 | 1.62 | 269,625 |
| Solano, CA | 100,671 | 2.16 | 438,687 |
| Virginia Beach, VA | 82,992 | 1.36 | 1,641,154 |
| Tulare, CA | 67,494 | 1.08 | 471,294 |
| McLennan, TX | 64,453 | 1.34 | 256,286 |
| Houston, GA | 83,283 | 0.80 | 189,105 |
| Fairfax, VA | 129,801 | 1.10 | 5,859,377 |
| Naugatuck Valley, CT | 93,202 | 1.11 | 434,486 |
| Black Hawk, IA | 75,013 | 1.09 | 155,456 |
| Marathon, WI | 77,127 | 1.69 | 163,997 |
| Belmont, OH | 60,282 | 1.61 | 129,839 |
| Sedgwick, KS | 71,185 | 1.29 | 639,062 |
| Wichita, TX | 65,991 | 1.49 | 129,389 |
| New Hanover, NC | 74,780 | 0.98 | 269,411 |
| Forsyth, NC | 66,974 | 1.24 | 653,947 |
| Worcester, MA | 99,647 | 1.01 | 840,052 |
| Yakima, WA | 68,633 | 1.11 | 253,792 |
| York, PA | 84,907 | 1.43 | 444,752 |
| Mahoning, OH | 58,912 | 1.70 | 512,839 |
| Yuma, AZ | 59,195 | 0.14 | 190,651 |

Note: The income of a metro is the population-weighted median of its tracts' median household incomes, a proxy built from tract values. Slopes are estimates with the intervals shown in the metro figure, which this plot does not draw.

Note: The correlations in the text are weighted by metro residents and counting each metro once, with bootstrap intervals over metros; all six intervals include zero.

Source: USALEEP life expectancy, 2010-2015; ACS 2019-2023; Area Deprivation Index, 2023; metro slopes, population-weighted within metros, modeled estimates

Data: https://vishalsingh.org/zipcode/report/gradients-slopes-and-variance/figures/slope_heterogeneity.json

### Table 4. The same regression in every state gives coefficients below zero for five measures in all 51 states and above zero for binge drinking in 50

Points of prevalence per doubling of tract median household income, one dot per state and measure, adults, with controls for the age, Black and Hispanic shares of residents, 2022

| Measure | State | Points per doubling | Lower limit of the 95% interval | Upper limit of the 95% interval | Tracts |
| --- | --- | ---: | ---: | ---: | ---: |
| Current smoking | Alabama | −5.29 | −6.54 | −4.05 | 1,423 |
| Current smoking | Alaska | −8.13 | −10.80 | −5.45 | 173 |
| Current smoking | Arizona | −5.12 | −7.05 | −3.18 | 1,725 |
| Current smoking | Arkansas | −6.01 | −7.17 | −4.85 | 818 |
| Current smoking | California | −3.66 | −4.10 | −3.23 | 9,003 |
| Current smoking | Colorado | −2.88 | −3.61 | −2.15 | 1,423 |
| Current smoking | Connecticut | −3.76 | −4.87 | −2.65 | 870 |
| Current smoking | Delaware | −3.63 | −5.14 | −2.13 | 256 |
| Current smoking | District of Columbia | −2.38 | −3.45 | −1.32 | 201 |
| Current smoking | Florida | −4.95 | −5.54 | −4.36 | 5,053 |
| Current smoking | Georgia | −5.70 | −6.27 | −5.14 | 2,766 |
| Current smoking | Hawaii | −3.96 | −4.52 | −3.40 | 419 |
| Current smoking | Idaho | −3.30 | −4.00 | −2.59 | 453 |
| Current smoking | Illinois | −4.67 | −5.93 | −3.42 | 3,239 |
| Current smoking | Indiana | −5.07 | −6.55 | −3.60 | 1,678 |
| Current smoking | Iowa | −4.47 | −5.64 | −3.31 | 890 |
| Current smoking | Kansas | −5.16 | −6.46 | −3.87 | 811 |
| Current smoking | Kentucky | −7.14 | −7.89 | −6.38 | 1,287 |
| Current smoking | Louisiana | −6.01 | −6.60 | −5.43 | 1,352 |
| Current smoking | Maine | −5.80 | −6.68 | −4.91 | 399 |
| Current smoking | Maryland | −4.84 | −5.75 | −3.92 | 1,449 |
| Current smoking | Massachusetts | −3.95 | −5.70 | −2.20 | 1,585 |
| Current smoking | Michigan | −5.85 | −6.69 | −5.01 | 2,869 |
| Current smoking | Minnesota | −4.56 | −5.34 | −3.78 | 1,497 |
| Current smoking | Mississippi | −5.77 | −7.25 | −4.29 | 865 |
| Current smoking | Missouri | −6.34 | −7.36 | −5.31 | 1,644 |
| Current smoking | Montana | −6.65 | −8.29 | −5.00 | 315 |
| Current smoking | Nebraska | −3.80 | −4.23 | −3.37 | 550 |
| Current smoking | Nevada | −4.54 | −5.64 | −3.44 | 767 |
| Current smoking | New Hampshire | −4.38 | −5.14 | −3.62 | 347 |
| Current smoking | New Jersey | −4.31 | −4.87 | −3.75 | 2,154 |
| Current smoking | New Mexico | −6.93 | −8.55 | −5.31 | 594 |
| Current smoking | New York | −4.70 | −5.23 | −4.17 | 5,182 |
| Current smoking | North Carolina | −5.40 | −6.02 | −4.78 | 2,628 |
| Current smoking | North Dakota | −4.04 | −6.23 | −1.86 | 227 |
| Current smoking | Ohio | −6.77 | −7.24 | −6.30 | 3,129 |
| Current smoking | Oklahoma | −6.30 | −7.23 | −5.36 | 1,189 |
| Current smoking | Oregon | −4.57 | −5.48 | −3.66 | 991 |
| Current smoking | Pennsylvania | −5.51 | −5.98 | −5.05 | 3,377 |
| Current smoking | Rhode Island | −4.53 | −5.11 | −3.94 | 246 |
| Current smoking | South Carolina | −5.61 | −6.44 | −4.78 | 1,301 |
| Current smoking | South Dakota | −7.73 | −10.72 | −4.74 | 242 |
| Current smoking | Tennessee | −6.64 | −7.92 | −5.36 | 1,675 |
| Current smoking | Texas | −4.35 | −4.98 | −3.73 | 6,802 |
| Current smoking | Utah | −1.76 | −2.66 | −0.86 | 707 |
| Current smoking | Vermont | −3.95 | −5.08 | −2.83 | 192 |
| Current smoking | Virginia | −4.66 | −5.25 | −4.07 | 2,152 |
| Current smoking | Washington | −3.93 | −4.54 | −3.32 | 1,767 |
| Current smoking | West Virginia | −5.86 | −7.73 | −3.99 | 543 |
| Current smoking | Wisconsin | −4.04 | −4.66 | −3.42 | 1,523 |
| Current smoking | Wyoming | −2.44 | −5.45 | 0.56 | 159 |
| Obesity | Alabama | −4.19 | −5.38 | −3.00 | 1,423 |
| Obesity | Alaska | −5.44 | −6.79 | −4.08 | 173 |
| Obesity | Arizona | −2.94 | −4.29 | −1.60 | 1,725 |
| Obesity | Arkansas | −4.27 | −5.48 | −3.06 | 818 |
| Obesity | California | −4.02 | −5.36 | −2.69 | 9,003 |
| Obesity | Colorado | −1.90 | −3.21 | −0.58 | 1,423 |
| Obesity | Connecticut | −5.18 | −7.75 | −2.61 | 870 |
| Obesity | Delaware | −3.57 | −4.88 | −2.26 | 256 |
| Obesity | District of Columbia | −1.46 | −2.58 | −0.33 | 201 |
| Obesity | Florida | −5.09 | −5.94 | −4.24 | 5,053 |
| Obesity | Georgia | −6.05 | −7.26 | −4.83 | 2,766 |
| Obesity | Hawaii | −3.12 | −3.40 | −2.84 | 419 |
| Obesity | Idaho | −3.29 | −4.95 | −1.62 | 453 |
| Obesity | Illinois | −4.31 | −5.85 | −2.77 | 3,239 |
| Obesity | Indiana | −3.79 | −5.61 | −1.96 | 1,678 |
| Obesity | Iowa | −3.37 | −4.61 | −2.12 | 890 |
| Obesity | Kansas | −4.54 | −5.89 | −3.19 | 811 |
| Obesity | Kentucky | −5.33 | −6.31 | −4.34 | 1,287 |
| Obesity | Louisiana | −5.19 | −6.09 | −4.30 | 1,352 |
| Obesity | Maine | −6.92 | −8.25 | −5.60 | 399 |
| Obesity | Maryland | −5.11 | −7.78 | −2.45 | 1,449 |
| Obesity | Massachusetts | −3.57 | −5.94 | −1.20 | 1,585 |
| Obesity | Michigan | −5.40 | −6.53 | −4.27 | 2,869 |
| Obesity | Minnesota | −5.36 | −6.53 | −4.18 | 1,497 |
| Obesity | Mississippi | −4.27 | −5.67 | −2.86 | 865 |
| Obesity | Missouri | −4.95 | −6.98 | −2.93 | 1,644 |
| Obesity | Montana | −5.67 | −8.49 | −2.85 | 315 |
| Obesity | Nebraska | −3.15 | −4.74 | −1.56 | 550 |
| Obesity | Nevada | −3.67 | −4.87 | −2.48 | 767 |
| Obesity | New Hampshire | −2.80 | −4.04 | −1.56 | 347 |
| Obesity | New Jersey | −5.19 | −6.17 | −4.22 | 2,154 |
| Obesity | New Mexico | −5.71 | −6.82 | −4.61 | 594 |
| Obesity | New York | −5.19 | −6.80 | −3.59 | 5,182 |
| Obesity | North Carolina | −4.68 | −5.96 | −3.40 | 2,628 |
| Obesity | North Dakota | −2.65 | −4.32 | −0.98 | 227 |
| Obesity | Ohio | −5.73 | −6.28 | −5.18 | 3,129 |
| Obesity | Oklahoma | −4.94 | −6.09 | −3.79 | 1,189 |
| Obesity | Oregon | −3.21 | −4.36 | −2.07 | 991 |
| Obesity | Pennsylvania | −4.76 | −5.47 | −4.05 | 3,377 |
| Obesity | Rhode Island | −3.64 | −5.39 | −1.90 | 246 |
| Obesity | South Carolina | −5.64 | −6.83 | −4.46 | 1,301 |
| Obesity | South Dakota | −4.30 | −5.52 | −3.08 | 242 |
| Obesity | Tennessee | −5.61 | −7.06 | −4.16 | 1,675 |
| Obesity | Texas | −3.87 | −4.48 | −3.25 | 6,802 |
| Obesity | Utah | −1.56 | −2.61 | −0.51 | 707 |
| Obesity | Vermont | −3.52 | −4.60 | −2.43 | 192 |
| Obesity | Virginia | −4.49 | −5.82 | −3.17 | 2,152 |
| Obesity | Washington | −5.22 | −7.57 | −2.88 | 1,767 |
| Obesity | West Virginia | −3.40 | −4.67 | −2.14 | 543 |
| Obesity | Wisconsin | −3.44 | −4.32 | −2.57 | 1,523 |
| Obesity | Wyoming | −3.13 | −7.58 | 1.33 | 159 |
| Diagnosed diabetes | Alabama | −2.33 | −3.32 | −1.34 | 1,423 |
| Diagnosed diabetes | Alaska | −2.87 | −3.89 | −1.84 | 173 |
| Diagnosed diabetes | Arizona | −2.22 | −3.76 | −0.69 | 1,725 |
| Diagnosed diabetes | Arkansas | −2.85 | −3.87 | −1.84 | 818 |
| Diagnosed diabetes | California | −1.35 | −1.71 | −0.98 | 9,003 |
| Diagnosed diabetes | Colorado | −0.76 | −1.28 | −0.25 | 1,423 |
| Diagnosed diabetes | Connecticut | −1.76 | −2.78 | −0.74 | 870 |
| Diagnosed diabetes | Delaware | −0.86 | −2.34 | 0.62 | 256 |
| Diagnosed diabetes | District of Columbia | −0.39 | −0.92 | 0.13 | 201 |
| Diagnosed diabetes | Florida | −2.46 | −3.01 | −1.90 | 5,053 |
| Diagnosed diabetes | Georgia | −2.57 | −3.04 | −2.09 | 2,766 |
| Diagnosed diabetes | Hawaii | −1.82 | −2.94 | −0.70 | 419 |
| Diagnosed diabetes | Idaho | −1.23 | −2.01 | −0.46 | 453 |
| Diagnosed diabetes | Illinois | −1.71 | −2.59 | −0.83 | 3,239 |
| Diagnosed diabetes | Indiana | −1.57 | −2.43 | −0.72 | 1,678 |
| Diagnosed diabetes | Iowa | −1.28 | −2.25 | −0.31 | 890 |
| Diagnosed diabetes | Kansas | −1.75 | −2.43 | −1.07 | 811 |
| Diagnosed diabetes | Kentucky | −3.07 | −3.50 | −2.63 | 1,287 |
| Diagnosed diabetes | Louisiana | −3.08 | −3.65 | −2.51 | 1,352 |
| Diagnosed diabetes | Maine | −2.38 | −2.91 | −1.86 | 399 |
| Diagnosed diabetes | Maryland | −1.72 | −2.41 | −1.03 | 1,449 |
| Diagnosed diabetes | Massachusetts | −1.43 | −2.19 | −0.66 | 1,585 |
| Diagnosed diabetes | Michigan | −2.11 | −2.63 | −1.60 | 2,869 |
| Diagnosed diabetes | Minnesota | −1.59 | −1.96 | −1.22 | 1,497 |
| Diagnosed diabetes | Mississippi | −3.08 | −4.13 | −2.04 | 865 |
| Diagnosed diabetes | Missouri | −2.30 | −2.85 | −1.76 | 1,644 |
| Diagnosed diabetes | Montana | −2.87 | −3.85 | −1.90 | 315 |
| Diagnosed diabetes | Nebraska | −0.99 | −1.47 | −0.52 | 550 |
| Diagnosed diabetes | Nevada | −2.04 | −2.93 | −1.14 | 767 |
| Diagnosed diabetes | New Hampshire | −1.35 | −1.94 | −0.75 | 347 |
| Diagnosed diabetes | New Jersey | −1.46 | −1.94 | −0.99 | 2,154 |
| Diagnosed diabetes | New Mexico | −3.61 | −4.53 | −2.69 | 594 |
| Diagnosed diabetes | New York | −2.24 | −2.87 | −1.61 | 5,182 |
| Diagnosed diabetes | North Carolina | −2.41 | −2.95 | −1.87 | 2,628 |
| Diagnosed diabetes | North Dakota | −1.70 | −3.17 | −0.23 | 227 |
| Diagnosed diabetes | Ohio | −2.58 | −2.99 | −2.18 | 3,129 |
| Diagnosed diabetes | Oklahoma | −2.95 | −3.68 | −2.23 | 1,189 |
| Diagnosed diabetes | Oregon | −1.38 | −2.13 | −0.62 | 991 |
| Diagnosed diabetes | Pennsylvania | −1.91 | −2.22 | −1.59 | 3,377 |
| Diagnosed diabetes | Rhode Island | −2.01 | −2.55 | −1.47 | 246 |
| Diagnosed diabetes | South Carolina | −2.82 | −3.47 | −2.18 | 1,301 |
| Diagnosed diabetes | South Dakota | −3.14 | −4.61 | −1.67 | 242 |
| Diagnosed diabetes | Tennessee | −2.84 | −3.76 | −1.91 | 1,675 |
| Diagnosed diabetes | Texas | −1.86 | −2.42 | −1.29 | 6,802 |
| Diagnosed diabetes | Utah | −0.38 | −1.20 | 0.45 | 707 |
| Diagnosed diabetes | Vermont | −1.09 | −1.61 | −0.57 | 192 |
| Diagnosed diabetes | Virginia | −1.49 | −1.94 | −1.05 | 2,152 |
| Diagnosed diabetes | Washington | −1.41 | −1.93 | −0.89 | 1,767 |
| Diagnosed diabetes | West Virginia | −2.40 | −3.84 | −0.95 | 543 |
| Diagnosed diabetes | Wisconsin | −1.03 | −1.50 | −0.55 | 1,523 |
| Diagnosed diabetes | Wyoming | −1.02 | −2.99 | 0.94 | 159 |
| Frequent mental distress | Alabama | −4.47 | −4.82 | −4.12 | 1,423 |
| Frequent mental distress | Alaska | −4.04 | −5.03 | −3.04 | 173 |
| Frequent mental distress | Arizona | −4.21 | −5.30 | −3.11 | 1,725 |
| Frequent mental distress | Arkansas | −4.24 | −4.78 | −3.70 | 818 |
| Frequent mental distress | California | −3.34 | −3.76 | −2.93 | 9,003 |
| Frequent mental distress | Colorado | −3.44 | −4.15 | −2.73 | 1,423 |
| Frequent mental distress | Connecticut | −3.41 | −3.88 | −2.93 | 870 |
| Frequent mental distress | Delaware | −3.95 | −5.59 | −2.31 | 256 |
| Frequent mental distress | District of Columbia | −3.02 | −4.58 | −1.45 | 201 |
| Frequent mental distress | Florida | −3.51 | −3.82 | −3.20 | 5,053 |
| Frequent mental distress | Georgia | −4.03 | −4.25 | −3.80 | 2,766 |
| Frequent mental distress | Hawaii | −2.41 | −2.82 | −2.00 | 419 |
| Frequent mental distress | Idaho | −4.19 | −5.20 | −3.17 | 453 |
| Frequent mental distress | Illinois | −3.42 | −3.79 | −3.05 | 3,239 |
| Frequent mental distress | Indiana | −4.33 | −4.68 | −3.98 | 1,678 |
| Frequent mental distress | Iowa | −4.31 | −5.00 | −3.62 | 890 |
| Frequent mental distress | Kansas | −3.96 | −4.35 | −3.57 | 811 |
| Frequent mental distress | Kentucky | −4.11 | −4.41 | −3.80 | 1,287 |
| Frequent mental distress | Louisiana | −3.71 | −4.17 | −3.25 | 1,352 |
| Frequent mental distress | Maine | −3.60 | −4.13 | −3.07 | 399 |
| Frequent mental distress | Maryland | −3.93 | −4.17 | −3.69 | 1,449 |
| Frequent mental distress | Massachusetts | −3.54 | −3.83 | −3.26 | 1,585 |
| Frequent mental distress | Michigan | −4.30 | −4.59 | −4.00 | 2,869 |
| Frequent mental distress | Minnesota | −3.95 | −4.32 | −3.58 | 1,497 |
| Frequent mental distress | Mississippi | −3.82 | −4.30 | −3.34 | 865 |
| Frequent mental distress | Missouri | −4.22 | −4.65 | −3.79 | 1,644 |
| Frequent mental distress | Montana | −4.06 | −4.83 | −3.29 | 315 |
| Frequent mental distress | Nebraska | −3.52 | −3.99 | −3.06 | 550 |
| Frequent mental distress | Nevada | −2.82 | −3.21 | −2.42 | 767 |
| Frequent mental distress | New Hampshire | −3.46 | −4.09 | −2.83 | 347 |
| Frequent mental distress | New Jersey | −3.18 | −3.63 | −2.73 | 2,154 |
| Frequent mental distress | New Mexico | −4.02 | −4.70 | −3.34 | 594 |
| Frequent mental distress | New York | −2.93 | −3.35 | −2.50 | 5,182 |
| Frequent mental distress | North Carolina | −3.89 | −4.14 | −3.64 | 2,628 |
| Frequent mental distress | North Dakota | −3.91 | −4.69 | −3.13 | 227 |
| Frequent mental distress | Ohio | −4.71 | −4.99 | −4.43 | 3,129 |
| Frequent mental distress | Oklahoma | −4.33 | −4.60 | −4.06 | 1,189 |
| Frequent mental distress | Oregon | −4.19 | −4.87 | −3.50 | 991 |
| Frequent mental distress | Pennsylvania | −3.60 | −3.82 | −3.38 | 3,377 |
| Frequent mental distress | Rhode Island | −2.93 | −3.65 | −2.22 | 246 |
| Frequent mental distress | South Carolina | −3.80 | −4.32 | −3.28 | 1,301 |
| Frequent mental distress | South Dakota | −4.69 | −5.57 | −3.81 | 242 |
| Frequent mental distress | Tennessee | −4.32 | −4.50 | −4.14 | 1,675 |
| Frequent mental distress | Texas | −3.83 | −4.11 | −3.55 | 6,802 |
| Frequent mental distress | Utah | −3.80 | −5.22 | −2.38 | 707 |
| Frequent mental distress | Vermont | −4.37 | −5.55 | −3.19 | 192 |
| Frequent mental distress | Virginia | −3.86 | −4.08 | −3.63 | 2,152 |
| Frequent mental distress | Washington | −4.30 | −4.64 | −3.96 | 1,767 |
| Frequent mental distress | West Virginia | −4.42 | −5.08 | −3.77 | 543 |
| Frequent mental distress | Wisconsin | −4.26 | −4.76 | −3.76 | 1,523 |
| Frequent mental distress | Wyoming | −3.49 | −4.43 | −2.55 | 159 |
| No leisure-time physical activity | Alabama | −8.26 | −9.49 | −7.02 | 1,423 |
| No leisure-time physical activity | Alaska | −7.43 | −9.28 | −5.57 | 173 |
| No leisure-time physical activity | Arizona | −6.01 | −7.81 | −4.22 | 1,725 |
| No leisure-time physical activity | Arkansas | −9.36 | −10.76 | −7.97 | 818 |
| No leisure-time physical activity | California | −5.16 | −5.83 | −4.48 | 9,003 |
| No leisure-time physical activity | Colorado | −3.81 | −4.54 | −3.07 | 1,423 |
| No leisure-time physical activity | Connecticut | −5.60 | −6.98 | −4.22 | 870 |
| No leisure-time physical activity | Delaware | −6.14 | −7.34 | −4.95 | 256 |
| No leisure-time physical activity | District of Columbia | −3.05 | −3.82 | −2.27 | 201 |
| No leisure-time physical activity | Florida | −7.40 | −8.10 | −6.70 | 5,053 |
| No leisure-time physical activity | Georgia | −8.61 | −9.48 | −7.73 | 2,766 |
| No leisure-time physical activity | Hawaii | −4.83 | −7.09 | −2.57 | 419 |
| No leisure-time physical activity | Idaho | −6.49 | −7.48 | −5.50 | 453 |
| No leisure-time physical activity | Illinois | −6.54 | −7.55 | −5.53 | 3,239 |
| No leisure-time physical activity | Indiana | −7.30 | −8.70 | −5.91 | 1,678 |
| No leisure-time physical activity | Iowa | −7.56 | −8.65 | −6.46 | 890 |
| No leisure-time physical activity | Kansas | −6.53 | −7.42 | −5.63 | 811 |
| No leisure-time physical activity | Kentucky | −9.02 | −9.63 | −8.41 | 1,287 |
| No leisure-time physical activity | Louisiana | −8.32 | −9.14 | −7.50 | 1,352 |
| No leisure-time physical activity | Maine | −8.03 | −9.35 | −6.70 | 399 |
| No leisure-time physical activity | Maryland | −6.69 | −7.77 | −5.62 | 1,449 |
| No leisure-time physical activity | Massachusetts | −5.93 | −7.56 | −4.29 | 1,585 |
| No leisure-time physical activity | Michigan | −7.22 | −8.06 | −6.37 | 2,869 |
| No leisure-time physical activity | Minnesota | −6.25 | −6.85 | −5.65 | 1,497 |
| No leisure-time physical activity | Mississippi | −8.38 | −10.04 | −6.72 | 865 |
| No leisure-time physical activity | Missouri | −8.13 | −8.97 | −7.29 | 1,644 |
| No leisure-time physical activity | Montana | −8.17 | −9.75 | −6.58 | 315 |
| No leisure-time physical activity | Nebraska | −5.56 | −6.21 | −4.91 | 550 |
| No leisure-time physical activity | Nevada | −5.65 | −7.08 | −4.22 | 767 |
| No leisure-time physical activity | New Hampshire | −5.29 | −6.12 | −4.46 | 347 |
| No leisure-time physical activity | New Jersey | −5.76 | −6.59 | −4.92 | 2,154 |
| No leisure-time physical activity | New Mexico | −7.85 | −9.10 | −6.59 | 594 |
| No leisure-time physical activity | New York | −6.95 | −8.08 | −5.82 | 5,182 |
| No leisure-time physical activity | North Carolina | −7.53 | −8.25 | −6.82 | 2,628 |
| No leisure-time physical activity | North Dakota | −6.37 | −8.10 | −4.63 | 227 |
| No leisure-time physical activity | Ohio | −8.44 | −8.88 | −8.00 | 3,129 |
| No leisure-time physical activity | Oklahoma | −9.23 | −10.46 | −7.99 | 1,189 |
| No leisure-time physical activity | Oregon | −5.06 | −5.98 | −4.14 | 991 |
| No leisure-time physical activity | Pennsylvania | −6.93 | −7.46 | −6.40 | 3,377 |
| No leisure-time physical activity | Rhode Island | −5.88 | −6.26 | −5.50 | 246 |
| No leisure-time physical activity | South Carolina | −8.46 | −9.27 | −7.64 | 1,301 |
| No leisure-time physical activity | South Dakota | −7.28 | −9.25 | −5.30 | 242 |
| No leisure-time physical activity | Tennessee | −8.52 | −9.83 | −7.21 | 1,675 |
| No leisure-time physical activity | Texas | −6.93 | −7.78 | −6.07 | 6,802 |
| No leisure-time physical activity | Utah | −3.44 | −4.31 | −2.56 | 707 |
| No leisure-time physical activity | Vermont | −5.43 | −6.39 | −4.47 | 192 |
| No leisure-time physical activity | Virginia | −5.79 | −6.42 | −5.16 | 2,152 |
| No leisure-time physical activity | Washington | −5.31 | −6.22 | −4.40 | 1,767 |
| No leisure-time physical activity | West Virginia | −8.03 | −9.68 | −6.37 | 543 |
| No leisure-time physical activity | Wisconsin | −6.16 | −6.64 | −5.68 | 1,523 |
| No leisure-time physical activity | Wyoming | −4.84 | −8.10 | −1.57 | 159 |
| Binge drinking | Alabama | 1.26 | 0.97 | 1.54 | 1,423 |
| Binge drinking | Alaska | 1.02 | 0.43 | 1.61 | 173 |
| Binge drinking | Arizona | 0.91 | 0.37 | 1.44 | 1,725 |
| Binge drinking | Arkansas | 1.97 | 1.23 | 2.71 | 818 |
| Binge drinking | California | 0.66 | −0.07 | 1.40 | 9,003 |
| Binge drinking | Colorado | 0.65 | −0.18 | 1.47 | 1,423 |
| Binge drinking | Connecticut | 0.15 | −0.40 | 0.70 | 870 |
| Binge drinking | Delaware | 0.95 | 0.63 | 1.28 | 256 |
| Binge drinking | District of Columbia | 0.92 | 0.13 | 1.70 | 201 |
| Binge drinking | Florida | 1.18 | 0.86 | 1.49 | 5,053 |
| Binge drinking | Georgia | 1.35 | 0.96 | 1.75 | 2,766 |
| Binge drinking | Hawaii | 0.97 | 0.41 | 1.53 | 419 |
| Binge drinking | Idaho | 1.16 | 0.71 | 1.61 | 453 |
| Binge drinking | Illinois | 1.54 | 0.70 | 2.38 | 3,239 |
| Binge drinking | Indiana | 1.20 | 0.61 | 1.79 | 1,678 |
| Binge drinking | Iowa | 1.49 | 1.23 | 1.76 | 890 |
| Binge drinking | Kansas | 0.65 | 0.22 | 1.07 | 811 |
| Binge drinking | Kentucky | 1.99 | 1.73 | 2.24 | 1,287 |
| Binge drinking | Louisiana | 2.33 | 1.81 | 2.84 | 1,352 |
| Binge drinking | Maine | 1.74 | 0.89 | 2.59 | 399 |
| Binge drinking | Maryland | −0.77 | −1.71 | 0.17 | 1,449 |
| Binge drinking | Massachusetts | 0.51 | −0.26 | 1.27 | 1,585 |
| Binge drinking | Michigan | 1.12 | 0.41 | 1.84 | 2,869 |
| Binge drinking | Minnesota | 0.63 | −0.05 | 1.32 | 1,497 |
| Binge drinking | Mississippi | 1.41 | 0.90 | 1.92 | 865 |
| Binge drinking | Missouri | 1.88 | 1.42 | 2.34 | 1,644 |
| Binge drinking | Montana | 2.18 | 1.44 | 2.93 | 315 |
| Binge drinking | Nebraska | 0.91 | 0.58 | 1.23 | 550 |
| Binge drinking | Nevada | 0.89 | 0.19 | 1.59 | 767 |
| Binge drinking | New Hampshire | 1.12 | 0.05 | 2.20 | 347 |
| Binge drinking | New Jersey | 0.36 | −0.19 | 0.90 | 2,154 |
| Binge drinking | New Mexico | 1.59 | 1.17 | 2.02 | 594 |
| Binge drinking | New York | 1.37 | 0.26 | 2.48 | 5,182 |
| Binge drinking | North Carolina | 1.57 | 0.96 | 2.18 | 2,628 |
| Binge drinking | North Dakota | 1.94 | 1.42 | 2.46 | 227 |
| Binge drinking | Ohio | 1.67 | 1.25 | 2.09 | 3,129 |
| Binge drinking | Oklahoma | 1.68 | 0.95 | 2.41 | 1,189 |
| Binge drinking | Oregon | 1.01 | 0.59 | 1.43 | 991 |
| Binge drinking | Pennsylvania | 1.16 | 0.68 | 1.63 | 3,377 |
| Binge drinking | Rhode Island | 1.13 | 0.64 | 1.62 | 246 |
| Binge drinking | South Carolina | 2.25 | 1.24 | 3.26 | 1,301 |
| Binge drinking | South Dakota | 1.32 | 0.46 | 2.18 | 242 |
| Binge drinking | Tennessee | 1.99 | 1.39 | 2.60 | 1,675 |
| Binge drinking | Texas | 0.89 | 0.49 | 1.28 | 6,802 |
| Binge drinking | Utah | 1.44 | 0.99 | 1.89 | 707 |
| Binge drinking | Vermont | 1.21 | 0.28 | 2.15 | 192 |
| Binge drinking | Virginia | 0.04 | −0.72 | 0.79 | 2,152 |
| Binge drinking | Washington | 0.37 | −0.17 | 0.92 | 1,767 |
| Binge drinking | West Virginia | 1.62 | 1.11 | 2.14 | 543 |
| Binge drinking | Wisconsin | 1.43 | 0.91 | 1.95 | 1,523 |
| Binge drinking | Wyoming | 0.82 | −0.66 | 2.30 | 159 |

Note: Each dot is a regression across the tracts of one state, and every state has at least 60 tracts with all the variables. The intervals use counties as clusters; Delaware and the District of Columbia use HC3 errors because they have too few counties for cluster-robust errors.

Note: PLACES values are modeled from BRFSS responses with census and ACS inputs, and median household income comes from the ACS, so part of each coefficient reflects shared inputs. These are differences between tracts within a state and do not describe what any person does.

Note: The highlighted dot is Maryland, the one state with a negative binge drinking coefficient; its interval includes zero.

Source: CDC PLACES 2024 release (BRFSS 2022) and ACS 2019-2023; modeled estimates, population-weighted (adults in each tract)

Data: https://vishalsingh.org/zipcode/report/gradients-slopes-and-variance/figures/coefficient_zoo.json

### Table 5. The first of nine components holds 51.3% of the variance in nine census measures and the second holds 16.0%

Principal components of poverty, cash assistance, unemployment, education, single-parent households, income, labor-force participation and homeownership across tracts, 2019-2023

#### Table 5.1. Scree

Two of nine components have an eigenvalue above one, and the first holds 51.3% of the variance

Eigenvalue of each principal component of the correlation matrix of nine standardized tract measures, with the Kaiser line at one

| component | conf_high | conf_low | eigenvalue |
| --- | ---: | ---: | ---: |
| PC1 | 4.69 | 4.54 | 4.61 |
| PC2 | 1.48 | 1.40 | 1.44 |
| PC3 | 0.907 | 0.854 | 0.880 |
| PC4 | 0.743 | 0.699 | 0.720 |
| PC5 | 0.543 | 0.520 | 0.532 |
| PC6 | 0.303 | 0.274 | 0.286 |
| PC7 | 0.283 | 0.263 | 0.276 |
| PC8 | 0.186 | 0.167 | 0.177 |
| PC9 | 0.082 | 0.073 | 0.077 |

#### Table 5.2. First component

The first component gives weight to all nine measures, with a largest absolute loading of 0.41

Loading of each measure on the first principal component, signed so that a positive loading means the measure rises with deprivation, with 95% intervals

| conf_high | conf_low | loading | measure |
| ---: | ---: | ---: | --- |
| 0.391 | 0.386 | 0.388 | Poverty rate |
| 0.404 | 0.400 | 0.402 | Households on cash assistance or SNAP |
| 0.262 | 0.242 | 0.252 | Unemployment rate |
| 0.384 | 0.372 | 0.378 | Adults with a high school diploma or less |
| 0.310 | 0.299 | 0.304 | Single-parent households |
| −0.410 | −0.416 | −0.412 | Median household income (log) |
| −0.359 | −0.370 | −0.364 | Adults with a bachelor's degree or more |
| −0.161 | −0.192 | −0.177 | Labor-force participation |
| −0.218 | −0.247 | −0.233 | Homeownership |

#### Table 5.3. Second component

The second component separates homeownership (−0.59) from labor-force participation (+0.56) and the college share (+0.35)

Loading of each measure on the second principal component, signed so that the college share loads positively, with 95% intervals

| conf_high | conf_low | loading | measure |
| ---: | ---: | ---: | --- |
| 0.169 | 0.132 | 0.152 | Poverty rate |
| 0.134 | 0.098 | 0.116 | Households on cash assistance or SNAP |
| 0.211 | 0.148 | 0.181 | Unemployment rate |
| −0.285 | −0.323 | −0.304 | Adults with a high school diploma or less |
| 0.264 | 0.195 | 0.228 | Single-parent households |
| 0.059 | 0.021 | 0.040 | Median household income (log) |
| 0.372 | 0.334 | 0.354 | Adults with a bachelor's degree or more |
| 0.584 | 0.524 | 0.555 | Labor-force participation |
| −0.573 | −0.610 | −0.593 | Homeownership |

Source: ACS 2019-2023, census tracts; principal components of nine standardized measures, unweighted (each tract counted once)

Data: https://vishalsingh.org/zipcode/report/gradients-slopes-and-variance/figures/nine_measure_pca.json

### Table 6. Weighted by population, the same pair of measures has a different correlation at each grain, and homeownership and poverty go from −0.53 across tracts to −0.05 across states

Correlation of six pairs of tract measures across tracts, across counties and across states, weighted by population or counting each unit once, with 95% intervals from a bootstrap over counties or states, 2019-2023

| Pair of measures | Tracts, weighted | Counties, weighted | States, weighted | Tracts, each once | Counties, each once | States, each once | States, weighted, lower limit | States, weighted, upper limit | States, each once, lower limit | States, each once, upper limit |
| --- | ---: | ---: | ---: | ---: | ---: | ---: | ---: | ---: | ---: | ---: |
| College share and income | +0.73 | +0.79 | +0.83 | +0.72 | +0.71 | +0.81 | +0.73 | +0.94 | +0.73 | +0.91 |
| Poverty rate and college share | −0.50 | −0.56 | −0.69 | −0.49 | −0.50 | −0.52 | −0.84 | −0.50 | −0.81 | −0.20 |
| Homeownership and poverty rate | −0.53 | −0.40 | −0.05 | −0.52 | −0.36 | −0.11 | −0.30 | +0.17 | −0.32 | +0.12 |
| Unemployment and poverty rate | +0.43 | +0.58 | +0.33 | +0.45 | +0.53 | +0.50 | +0.08 | +0.58 | +0.32 | +0.65 |
| College share and life expectancy | +0.56 | +0.64 | +0.70 | +0.55 | +0.51 | +0.48 | +0.55 | +0.83 | +0.13 | +0.81 |
| Poverty rate and life expectancy | −0.50 | −0.60 | −0.66 | −0.52 | −0.57 | −0.78 | −0.88 | −0.45 | −0.89 | −0.65 |

Views: Weighted, tracts to counties (every row); Weighted, counties to states (every row); Each unit once, tracts to counties (every row); Each unit once, counties to states (every row).

Note: Every correlation uses the same 77,775 tracts. County and state values are population-weighted means of the tract values; the weighted views weight each correlation by the population of its unit and the others count each unit once. The county correlations rest on 3,112 counties and the state correlations on 51 states, so the state intervals are wide.

Note: The correlation with life expectancy is between tract life expectancy, a modeled estimate, and census measures.

Note: ACS tract values are survey estimates with sampling error, and the staged file has no margins of error, so they are treated as exact.

Source: ACS 2019-2023 tracts and USALEEP life expectancy, 2010-2015; correlations of tract values and of population-weighted county and state means, population-weighted and unweighted

Data: https://vishalsingh.org/zipcode/report/gradients-slopes-and-variance/figures/maup_correlations.json

### Table 7. 70.3 percent of the variance of tract life expectancy lies between tracts in the same county, 26.7 points of it estimation error, against 34% percent for the ADI

Share of the population-weighted variance of three tract measures between states, between counties in a state and between tracts in a county, percent

| level | measure | share |
| --- | --- | ---: |
| Between states | Tract life expectancy | 13.75 |
| Between counties in a state | Tract life expectancy | 15.92 |
| Between tracts in a county | Tract life expectancy | 70.33 |
| Between states | ADI national percentile | 36.64 |
| Between counties in a state | ADI national percentile | 29.38 |
| Between tracts in a county | ADI national percentile | 33.98 |
| Between states | Nine-measure index percentile | 5.13 |
| Between counties in a state | Nine-measure index percentile | 23.29 |
| Between tracts in a county | Nine-measure index percentile | 71.59 |

Note: Tract life expectancy is modeled. The mean squared standard error of the tract estimates equals 26.7 percent of the total variance and sits inside the between-tract segment; the 2020-tract standard errors are mapping grade and understate the noise, so the share of the true differences that lies between tracts is lower than the net share in the text.

Note: The life expectancy bar uses 78,008 tracts in 3,112 counties and Connecticut planning regions; the ADI and index bars use the 82,545 tracts of the index. Connecticut's planning regions stand in for its counties.

Source: USALEEP life expectancy, 2010-2015; ACS 2019-2023; Area Deprivation Index, 2023; population-weighted decomposition of the variance, full counts of tracts

Data: https://vishalsingh.org/zipcode/report/gradients-slopes-and-variance/figures/variance_split.json

### Table 8. In each of the 30 most populous counties, tracts at the 10th and 90th percentile of the ADI are between 9 and 69 percentile points apart

Population-weighted mean ADI national percentile of each county (dot) and the 10th to 90th percentile range of its tracts' ADI (whisker), the most populous counties sorted by range, ACS 2019-2023 population

| County | Mean ADI | 10th percentile of tracts | 90th percentile of tracts | Range | Tracts | Residents |
| --- | ---: | ---: | ---: | ---: | ---: | ---: |
| Philadelphia, PA | 62.4 | 25.0 | 93.8 | 68.8 | 386 | 1,578,494 |
| Wayne, MI | 72.8 | 35.3 | 99.0 | 63.7 | 579 | 1,772,852 |
| Bexar, TX | 59.2 | 26.9 | 89.0 | 62.1 | 369 | 2,023,627 |
| Dallas, TX | 51.8 | 17.4 | 78.6 | 61.1 | 641 | 2,597,456 |
| Cook, IL | 47.1 | 17.0 | 77.0 | 60.0 | 1,322 | 5,158,123 |
| Harris, TX | 55.6 | 24.5 | 84.3 | 59.8 | 1,106 | 4,739,169 |
| Palm Beach, FL | 38.8 | 12.0 | 69.7 | 57.7 | 363 | 1,503,596 |
| Tarrant, TX | 48.9 | 22.1 | 77.6 | 55.4 | 445 | 2,123,556 |
| Broward, FL | 41.0 | 14.2 | 69.0 | 54.8 | 415 | 1,946,125 |
| Hillsborough, FL | 46.0 | 20.3 | 71.8 | 51.6 | 322 | 1,476,792 |
| Orange, FL | 44.1 | 20.0 | 70.4 | 50.4 | 262 | 1,413,087 |
| Maricopa, AZ | 36.1 | 13.0 | 60.7 | 47.7 | 985 | 4,468,217 |
| Bronx, NY | 31.1 | 15.2 | 61.0 | 45.7 | 340 | 1,411,947 |
| Miami-Dade, FL | 37.2 | 15.7 | 61.2 | 45.5 | 691 | 2,663,978 |
| San Bernardino, CA | 31.9 | 11.9 | 54.2 | 42.3 | 461 | 2,166,515 |
| Clark, NV | 39.2 | 20.0 | 62.2 | 42.2 | 534 | 2,293,278 |
| Riverside, CA | 28.4 | 11.0 | 51.3 | 40.3 | 514 | 2,435,341 |
| Sacramento, CA | 27.9 | 11.3 | 47.4 | 36.1 | 361 | 1,577,120 |
| Queens, NY | 16.4 | 6.0 | 31.1 | 25.1 | 677 | 2,325,406 |
| Middlesex, MA | 14.5 | 3.7 | 27.8 | 24.0 | 353 | 1,619,270 |
| Suffolk, NY | 19.7 | 7.8 | 31.5 | 23.7 | 368 | 1,437,246 |
| King, WA | 12.9 | 3.0 | 25.5 | 22.5 | 491 | 2,252,223 |
| San Diego, CA | 14.0 | 3.3 | 25.5 | 22.2 | 723 | 3,203,557 |
| Los Angeles, CA | 14.2 | 3.0 | 25.0 | 22.0 | 2,447 | 9,781,315 |
| New York, NY | 8.9 | 1.0 | 20.5 | 19.5 | 294 | 1,614,981 |
| Nassau, NY | 13.8 | 4.7 | 22.3 | 17.6 | 277 | 1,384,110 |
| Orange, CA | 11.2 | 3.0 | 20.0 | 17.0 | 611 | 3,164,006 |
| Kings, NY | 10.0 | 2.0 | 19.0 | 17.0 | 773 | 2,642,467 |
| Alameda, CA | 7.8 | 2.0 | 18.0 | 16.0 | 373 | 1,649,334 |
| Santa Clara, CA | 5.0 | 1.0 | 10.0 | 9.0 | 407 | 1,900,063 |

Note: Tract ADI is a national percentile, so a county whose tracts are all among the least deprived has a narrow range by construction. Counties are chosen by population, not by range.

Note: Connecticut's planning regions stand in for its counties. Puerto Rico is not included.

Source: Area Deprivation Index, 2023; ACS 2019-2023 population; population-weighted percentiles of tract values, full counts of tracts

Data: https://vishalsingh.org/zipcode/report/gradients-slopes-and-variance/figures/county_adi_spread.json

### Table 9. WONDER withholds 71.3% of county-year cells in the lightest birthweight class and 77.6% in the heaviest, against 0.0% in the five middle classes

Share of county-year cells that are withheld, published as zero or published with at least 10 births, by birthweight class, 578 counties named in every year, 2018 to 2024, percent

| Birthweight class | Cell status | Share of cells (%) | Cells | Cells in the class | Births in published cells of the class | Largest share of the class's births in withheld cells (%) |
| --- | --- | ---: | ---: | ---: | ---: | ---: |
| 499 g or less | Withheld (1 to 9 births) | 71.3 | 2,885 | 4,046 | 17,300 | 60.0 |
| 499 g or less | Published zero | 8.7 | 350 | 4,046 | 17,300 | 60.0 |
| 499 g or less | Published, 10 or more births | 20.0 | 811 | 4,046 | 17,300 | 60.0 |
| 500 to 999 g | Withheld (1 to 9 births) | 36.4 | 1,474 | 4,046 | 93,575 | 12.4 |
| 500 to 999 g | Published zero | 0.4 | 16 | 4,046 | 93,575 | 12.4 |
| 500 to 999 g | Published, 10 or more births | 63.2 | 2,556 | 4,046 | 93,575 | 12.4 |
| 1,000 to 1,499 g | Withheld (1 to 9 births) | 18.3 | 741 | 4,046 | 144,348 | 4.4 |
| 1,000 to 1,499 g | Published zero | 0.0 | 0 | 4,046 | 144,348 | 4.4 |
| 1,000 to 1,499 g | Published, 10 or more births | 81.7 | 3,305 | 4,046 | 144,348 | 4.4 |
| 1,500 to 1,999 g | Withheld (1 to 9 births) | 1.1 | 43 | 4,046 | 334,713 | 0.1 |
| 1,500 to 1,999 g | Published zero | 0.0 | 0 | 4,046 | 334,713 | 0.1 |
| 1,500 to 1,999 g | Published, 10 or more births | 98.9 | 4,003 | 4,046 | 334,713 | 0.1 |
| 2,000 to 2,499 g | Withheld (1 to 9 births) | 0.0 | 0 | 4,046 | 1,107,825 | 0.0 |
| 2,000 to 2,499 g | Published zero | 0.0 | 0 | 4,046 | 1,107,825 | 0.0 |
| 2,000 to 2,499 g | Published, 10 or more births | 100.0 | 4,046 | 4,046 | 1,107,825 | 0.0 |
| 2,500 to 2,999 g | Withheld (1 to 9 births) | 0.0 | 0 | 4,046 | 3,979,246 | 0.0 |
| 2,500 to 2,999 g | Published zero | 0.0 | 0 | 4,046 | 3,979,246 | 0.0 |
| 2,500 to 2,999 g | Published, 10 or more births | 100.0 | 4,046 | 4,046 | 3,979,246 | 0.0 |
| 3,000 to 3,499 g | Withheld (1 to 9 births) | 0.0 | 0 | 4,046 | 7,991,060 | 0.0 |
| 3,000 to 3,499 g | Published zero | 0.0 | 0 | 4,046 | 7,991,060 | 0.0 |
| 3,000 to 3,499 g | Published, 10 or more births | 100.0 | 4,046 | 4,046 | 7,991,060 | 0.0 |
| 3,500 to 3,999 g | Withheld (1 to 9 births) | 0.0 | 0 | 4,046 | 5,263,258 | 0.0 |
| 3,500 to 3,999 g | Published zero | 0.0 | 0 | 4,046 | 5,263,258 | 0.0 |
| 3,500 to 3,999 g | Published, 10 or more births | 100.0 | 4,046 | 4,046 | 5,263,258 | 0.0 |
| 4,000 to 4,499 g | Withheld (1 to 9 births) | 0.0 | 0 | 4,046 | 1,289,498 | 0.0 |
| 4,000 to 4,499 g | Published zero | 0.0 | 0 | 4,046 | 1,289,498 | 0.0 |
| 4,000 to 4,499 g | Published, 10 or more births | 100.0 | 4,046 | 4,046 | 1,289,498 | 0.0 |
| 4,500 to 4,999 g | Withheld (1 to 9 births) | 8.3 | 334 | 4,046 | 167,374 | 1.8 |
| 4,500 to 4,999 g | Published zero | 0.0 | 0 | 4,046 | 167,374 | 1.8 |
| 4,500 to 4,999 g | Published, 10 or more births | 91.7 | 3,712 | 4,046 | 167,374 | 1.8 |
| 5,000 g or more | Withheld (1 to 9 births) | 77.6 | 3,141 | 4,046 | 8,863 | 76.1 |
| 5,000 g or more | Published zero | 10.9 | 441 | 4,046 | 8,863 | 76.1 |
| 5,000 g or more | Published, 10 or more births | 11.5 | 464 | 4,046 | 8,863 | 76.1 |
| Weight not stated | Withheld (1 to 9 births) | 65.0 | 2,632 | 4,046 | 9,365 | 71.7 |
| Weight not stated | Published zero | 27.2 | 1,100 | 4,046 | 9,365 | 71.7 |
| Weight not stated | Published, 10 or more births | 7.8 | 314 | 4,046 | 9,365 | 71.7 |

Note: WONDER withholds a cell with fewer than 10 births, so a withheld cell holds at most 9. The counts here are of cells only; nothing is derived for a withheld cell, and the county totals are not used.

Note: The bound in the table is the births in the published cells of a class plus 9 for each withheld cell, which is the largest share of the class's births that the withheld cells could hold.

Note: Cells are for the 578 counties that WONDER names in every year, which have at least 100,000 residents; pooled rows for other counties are not cells of a county.

Source: CDC WONDER natality, county by birthweight class, 2018-2024; full counts of cells

Data: https://vishalsingh.org/zipcode/report/gradients-slopes-and-variance/figures/wonder_suppression_by_birthweight.json

## All estimates

Every fact in facts.json (134), sorted by id.

| Fact id | Value | Unit | Estimand | Population | Period | Interval | n | Weights | Source | Note |
| --- | ---: | --- | --- | --- | --- | --- | ---: | --- | --- | --- |
| adi_state | 37% | percent of the variance of the ADI national percentile | share of the population-weighted variance of the ADI national percentile that lies between states (sum-of-squares decomposition) | tracts in the 50 states and DC with all nine measures, an ACS population above zero and an ADI rank | ACS 2019-2023; ADI 2023 |   | 82,545 tracts | ACS 2019-2023 tract population | Area Deprivation Index |   |
| adi_within | 34% | percent of the variance of the ADI national percentile | share of the population-weighted variance of the ADI national percentile that lies between tracts in the same county (sum-of-squares decomposition) | tracts in the 50 states and DC with all nine measures, an ACS population above zero and an ADI rank | ACS 2019-2023; ADI 2023 |   | 82,545 tracts | ACS 2019-2023 tract population | Area Deprivation Index |   |
| big_le_spread | 8.0 | years of life expectancy between the 10th and 90th percentile of tract life expectancy | median over counties of the population-weighted 90th minus 10th percentile of tract life expectancy | the 30 most populous counties among those with an ADI rank for at least 20 tracts | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population |   | 30 counties | none (each county counts once) | USALEEP life expectancy | Tract life expectancies are modeled, so part of this spread is estimation error. |
| big_max | 69 | ADI percentile points between the 10th and 90th percentile of tract ADI | largest population-weighted 90th minus 10th percentile of tract ADI among the counties | the 30 most populous counties among those with an ADI rank for at least 20 tracts | ACS 2019-2023; ADI 2023 |   | 30 counties | none (each county counts once) | Area Deprivation Index |   |
| big_min | 9 | ADI percentile points between the 10th and 90th percentile of tract ADI | smallest population-weighted 90th minus 10th percentile of tract ADI among the counties | the 30 most populous counties among those with an ADI rank for at least 20 tracts | ACS 2019-2023; ADI 2023 |   | 30 counties | none (each county counts once) | Area Deprivation Index |   |
| big_n | 30 | counties | number of counties shown (rule) | most populous counties and Connecticut planning regions (ACS 2019-2023) | ACS 2019-2023; ADI 2023 |   |   | none (definition) | Area Deprivation Index |   |
| big_narrow_cut | 25 | ADI percentile points | a range of this many percentile points or fewer counts as narrow (definition) | the 30 most populous counties among those with an ADI rank for at least 20 tracts | ACS 2019-2023; ADI 2023 |   |   | none (definition) | Area Deprivation Index |   |
| big_narrow_n | 11 | counties | count of the counties whose 90th minus 10th percentile of tract ADI is at most the narrow cut | the 30 most populous counties among those with an ADI rank for at least 20 tracts | ACS 2019-2023; ADI 2023 |   | 30 counties | none (full count) | Area Deprivation Index |   |
| binge_median | +1.2 | points of prevalence per doubling of tract median household income | median over states of the state coefficient: a population-weighted regression across the state's tracts of tract prevalence on the log of median household income, with tract shares of residents ages 65 and over, Black alone and Hispanic as controls, times the natural log of 2 | adults: binge drinking, states and the District of Columbia with at least 60 tracts that have income, the three controls and the measure | BRFSS 2022 (PLACES 2024 release); ACS 2019-2023 |   | 51 states | adult population of the tract (PLACES) | CDC PLACES | A median of the state estimates. |
| binge_pos_n | 50 | states | count of states where the binge drinking coefficient is above zero | states and the District of Columbia with at least 60 tracts that have income, the three controls and the measure | BRFSS 2022 (PLACES 2024 release); ACS 2019-2023 |   | 51 states | none (full count) | CDC PLACES |   |
| bw_heavy_min | 5,000 | grams | lower limit of the heaviest class (definition) | births in the heaviest birthweight class of the WONDER file | 2018 to 2024 |   |   | none (definition) | CDC WONDER natality |   |
| bw_light_max | 499 | grams | upper limit of the lightest class (definition) | births in the lightest birthweight class of the WONDER file | 2018 to 2024 |   |   | none (definition) | CDC WONDER natality |   |
| idx_within | 72% | percent of the variance of the nine-measure index percentile | share of the population-weighted variance of the first-component index percentile that lies between tracts in the same county (sum-of-squares decomposition) | tracts in the 50 states and DC with all nine measures, an ACS population above zero and an ADI rank | ACS 2019-2023; ADI 2023 |   | 82,545 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| ladder_county | 1.12 | years of life expectancy at birth lower per ten-point step in the ADI national percentile | population-weighted least-squares slope of tract life expectancy on the ADI national percentile, times ten, sign reversed, with a constant for each county; cluster-robust interval (t on the number of counties minus one) | residents of non-institutional tracts in the 50 states and DC with a life expectancy estimate and an ADI rank | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population | 95% CI 1.07 to 1.16 | 77,722 tracts | ACS 2019-2023 tract population | USALEEP life expectancy |   |
| ladder_ratio | 1.3 | ratio of slopes | within-county slope over the slope across all tracts | residents of non-institutional tracts in the 50 states and DC with a life expectancy estimate and an ADI rank | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population |   | 77,936 tracts | ACS 2019-2023 tract population | USALEEP life expectancy | A ratio of two point estimates. |
| ladder_raw | 0.84 | years of life expectancy at birth lower per ten-point step in the ADI national percentile | population-weighted least-squares slope of tract life expectancy on the ADI national percentile, times ten, sign reversed; cluster-robust interval (t on the number of states minus one) | residents of non-institutional tracts in the 50 states and DC with a life expectancy estimate and an ADI rank | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population | 95% CI 0.75 to 0.92 | 77,936 tracts | ACS 2019-2023 tract population | USALEEP life expectancy |   |
| ladder_state | 0.92 | years of life expectancy at birth lower per ten-point step in the ADI national percentile | population-weighted least-squares slope of tract life expectancy on the ADI national percentile, times ten, sign reversed, with a constant for each state; cluster-robust interval (t on the number of states minus one) | residents of non-institutional tracts in the 50 states and DC with a life expectancy estimate and an ADI rank | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population | 95% CI 0.78 to 1.05 | 77,936 tracts | ACS 2019-2023 tract population | USALEEP life expectancy |   |
| ladder_unw_maxdiff | 0.015 | years of life expectancy per ten-point step | largest absolute difference, over the three specifications, between the population-weighted slope and the slope that counts each tract once | residents of non-institutional tracts in the 50 states and DC with a life expectancy estimate and an ADI rank | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population |   | 77,936 tracts | ACS 2019-2023 tract population | USALEEP life expectancy | A difference of point estimates. |
| map_floor | 20 | pooled deaths | minimum number of pooled deaths for a county to be mapped (rule) | counties on a map of WONDER deaths | 2018 to 2024 |   |   | none (definition) | CDC WONDER underlying cause of death |   |
| maup_cle_state | +0.71 | correlation coefficient | population-weighted correlation of the college share and life expectancy across states (values are population-weighted means of tract values); bootstrap interval over states | tracts in the 50 states and DC with a life expectancy estimate and all five ACS measures of the correlations | ACS 2019-2023; deaths in 2010 to 2015 (USALEEP) | 95% CI 0.55 to 0.83 | 51 states | population of the unit (ACS 2019-2023) | American Community Survey |   |
| maup_cle_state_u | +0.48 | correlation coefficient | unweighted (each unit counted once) correlation of the college share and life expectancy across states (values are population-weighted means of tract values); bootstrap interval over states | tracts in the 50 states and DC with a life expectancy estimate and all five ACS measures of the correlations | ACS 2019-2023; deaths in 2010 to 2015 (USALEEP) | 95% CI 0.13 to 0.81 | 51 states | none (each unit counts once) | American Community Survey |   |
| maup_cle_tract | +0.56 | correlation coefficient | population-weighted correlation of the college share and life expectancy across tracts (values are population-weighted means of tract values); bootstrap interval over counties | tracts in the 50 states and DC with a life expectancy estimate and all five ACS measures of the correlations | ACS 2019-2023; deaths in 2010 to 2015 (USALEEP) | 95% CI 0.54 to 0.58 | 77,775 tracts | population of the unit (ACS 2019-2023) | American Community Survey |   |
| maup_cle_tract_u | +0.55 | correlation coefficient | unweighted (each unit counted once) correlation of the college share and life expectancy across tracts (values are population-weighted means of tract values); bootstrap interval over counties | tracts in the 50 states and DC with a life expectancy estimate and all five ACS measures of the correlations | ACS 2019-2023; deaths in 2010 to 2015 (USALEEP) | 95% CI 0.53 to 0.57 | 77,775 tracts | none (each unit counts once) | American Community Survey |   |
| maup_coll_state | +0.83 | correlation coefficient | population-weighted correlation of the college share and the log of median household income across states (values are population-weighted means of tract values); bootstrap interval over states | tracts in the 50 states and DC with a life expectancy estimate and all five ACS measures of the correlations | ACS 2019-2023; deaths in 2010 to 2015 (USALEEP) | 95% CI 0.73 to 0.94 | 51 states | population of the unit (ACS 2019-2023) | American Community Survey |   |
| maup_coll_tract | +0.73 | correlation coefficient | population-weighted correlation of the college share and the log of median household income across tracts (values are population-weighted means of tract values); bootstrap interval over counties | tracts in the 50 states and DC with a life expectancy estimate and all five ACS measures of the correlations | ACS 2019-2023; deaths in 2010 to 2015 (USALEEP) | 95% CI 0.71 to 0.74 | 77,775 tracts | population of the unit (ACS 2019-2023) | American Community Survey |   |
| maup_n_counties | 3,112 | counties and Connecticut planning regions | count of counties in the correlations | tracts in the 50 states and DC with a life expectancy estimate and all five ACS measures of the correlations | ACS 2019-2023; deaths in 2010 to 2015 (USALEEP) |   |   | none (full count) | American Community Survey |   |
| maup_n_flip | 2 | pairs of measures | count of pairs for which the correlation is larger across states than across tracts under one weighting and not under the other | tracts in the 50 states and DC with a life expectancy estimate and all five ACS measures of the correlations | ACS 2019-2023; deaths in 2010 to 2015 (USALEEP) |   | 6 pairs | none (full count) | American Community Survey |   |
| maup_n_larger | 4 | pairs of measures | count of pairs whose correlation, weighted by population, is larger in absolute value across states than across tracts | tracts in the 50 states and DC with a life expectancy estimate and all five ACS measures of the correlations | ACS 2019-2023; deaths in 2010 to 2015 (USALEEP) |   | 6 pairs | none (full count) | American Community Survey |   |
| maup_n_larger_u | 4 | pairs of measures | count of pairs whose correlation, counting each unit once, is larger in absolute value across states than across tracts | tracts in the 50 states and DC with a life expectancy estimate and all five ACS measures of the correlations | ACS 2019-2023; deaths in 2010 to 2015 (USALEEP) |   | 6 pairs | none (full count) | American Community Survey |   |
| maup_n_pairs | 6 | pairs of measures | count of pairs (rule) | tracts in the 50 states and DC with a life expectancy estimate and all five ACS measures of the correlations | ACS 2019-2023; deaths in 2010 to 2015 (USALEEP) |   |   | none (definition) | American Community Survey |   |
| maup_n_states | 51 | states | count of states in the correlations | tracts in the 50 states and DC with a life expectancy estimate and all five ACS measures of the correlations | ACS 2019-2023; deaths in 2010 to 2015 (USALEEP) |   |   | none (full count) | American Community Survey |   |
| maup_n_tracts | 77,775 | tracts | count of tracts in the correlations | tracts in the 50 states and DC with a life expectancy estimate and all five ACS measures of the correlations | ACS 2019-2023; deaths in 2010 to 2015 (USALEEP) |   |   | none (full count) | American Community Survey |   |
| maup_own_county | −0.40 | correlation coefficient | population-weighted correlation of homeownership and the poverty rate across counties (values are population-weighted means of tract values); bootstrap interval over states | tracts in the 50 states and DC with a life expectancy estimate and all five ACS measures of the correlations | ACS 2019-2023; deaths in 2010 to 2015 (USALEEP) | 95% CI −0.53 to −0.26 | 3,112 counties | population of the unit (ACS 2019-2023) | American Community Survey |   |
| maup_own_state | −0.05 | correlation coefficient | population-weighted correlation of homeownership and the poverty rate across states (values are population-weighted means of tract values); bootstrap interval over states | tracts in the 50 states and DC with a life expectancy estimate and all five ACS measures of the correlations | ACS 2019-2023; deaths in 2010 to 2015 (USALEEP) | 95% CI −0.30 to 0.17 | 51 states | population of the unit (ACS 2019-2023) | American Community Survey |   |
| maup_own_state_u | −0.11 | correlation coefficient | unweighted (each unit counted once) correlation of homeownership and the poverty rate across states (values are population-weighted means of tract values); bootstrap interval over states | tracts in the 50 states and DC with a life expectancy estimate and all five ACS measures of the correlations | ACS 2019-2023; deaths in 2010 to 2015 (USALEEP) | 95% CI −0.32 to 0.11 | 51 states | none (each unit counts once) | American Community Survey |   |
| maup_own_tract | −0.53 | correlation coefficient | population-weighted correlation of homeownership and the poverty rate across tracts (values are population-weighted means of tract values); bootstrap interval over counties | tracts in the 50 states and DC with a life expectancy estimate and all five ACS measures of the correlations | ACS 2019-2023; deaths in 2010 to 2015 (USALEEP) | 95% CI −0.55 to −0.51 | 77,775 tracts | population of the unit (ACS 2019-2023) | American Community Survey |   |
| maup_own_tract_u | −0.52 | correlation coefficient | unweighted (each unit counted once) correlation of homeownership and the poverty rate across tracts (values are population-weighted means of tract values); bootstrap interval over counties | tracts in the 50 states and DC with a life expectancy estimate and all five ACS measures of the correlations | ACS 2019-2023; deaths in 2010 to 2015 (USALEEP) | 95% CI −0.54 to −0.50 | 77,775 tracts | none (each unit counts once) | American Community Survey |   |
| maup_unemp_county | +0.59 | correlation coefficient | population-weighted correlation of the unemployment rate and the poverty rate across counties (values are population-weighted means of tract values); bootstrap interval over states | tracts in the 50 states and DC with a life expectancy estimate and all five ACS measures of the correlations | ACS 2019-2023; deaths in 2010 to 2015 (USALEEP) | 95% CI 0.52 to 0.66 | 3,112 counties | population of the unit (ACS 2019-2023) | American Community Survey |   |
| maup_unemp_county_u | +0.53 | correlation coefficient | unweighted (each unit counted once) correlation of the unemployment rate and the poverty rate across counties (values are population-weighted means of tract values); bootstrap interval over states | tracts in the 50 states and DC with a life expectancy estimate and all five ACS measures of the correlations | ACS 2019-2023; deaths in 2010 to 2015 (USALEEP) | 95% CI 0.44 to 0.61 | 3,112 counties | none (each unit counts once) | American Community Survey |   |
| maup_unemp_state | +0.33 | correlation coefficient | population-weighted correlation of the unemployment rate and the poverty rate across states (values are population-weighted means of tract values); bootstrap interval over states | tracts in the 50 states and DC with a life expectancy estimate and all five ACS measures of the correlations | ACS 2019-2023; deaths in 2010 to 2015 (USALEEP) | 95% CI 0.08 to 0.59 | 51 states | population of the unit (ACS 2019-2023) | American Community Survey |   |
| maup_unemp_state_u | +0.50 | correlation coefficient | unweighted (each unit counted once) correlation of the unemployment rate and the poverty rate across states (values are population-weighted means of tract values); bootstrap interval over states | tracts in the 50 states and DC with a life expectancy estimate and all five ACS measures of the correlations | ACS 2019-2023; deaths in 2010 to 2015 (USALEEP) | 95% CI 0.32 to 0.65 | 51 states | none (each unit counts once) | American Community Survey |   |
| maup_unemp_tract | +0.43 | correlation coefficient | population-weighted correlation of the unemployment rate and the poverty rate across tracts (values are population-weighted means of tract values); bootstrap interval over counties | tracts in the 50 states and DC with a life expectancy estimate and all five ACS measures of the correlations | ACS 2019-2023; deaths in 2010 to 2015 (USALEEP) | 95% CI 0.41 to 0.46 | 77,775 tracts | population of the unit (ACS 2019-2023) | American Community Survey |   |
| maup_unemp_tract_u | +0.45 | correlation coefficient | unweighted (each unit counted once) correlation of the unemployment rate and the poverty rate across tracts (values are population-weighted means of tract values); bootstrap interval over counties | tracts in the 50 states and DC with a life expectancy estimate and all five ACS measures of the correlations | ACS 2019-2023; deaths in 2010 to 2015 (USALEEP) | 95% CI 0.42 to 0.47 | 77,775 tracts | none (each unit counts once) | American Community Survey |   |
| metro_cluster_min | 5 | counties | minimum number of counties in a metro for cluster-robust errors on counties (rule); a metro with fewer uses HC3 errors | residents of metros (core-based statistical areas) with at least 40 tracts that have a life expectancy estimate and an ADI rank | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population |   |   | none (definition) | USALEEP life expectancy |   |
| metro_flat_edge | 0.70 | years of life expectancy at birth lower per ten-point step in the ADI national percentile | 5th percentile over metros of the estimated metro slope, sign reversed | residents of metros (core-based statistical areas) with at least 40 tracts that have a life expectancy estimate and an ADI rank | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population |   | 267 metros | none (each metro counts once) | USALEEP life expectancy | A percentile of estimates, which are stretched by estimation error. |
| metro_flat_edge_sh | 0.88 | years of life expectancy at birth lower per ten-point step in the ADI national percentile | 5th percentile over metros of the metro slope shrunk toward the mean by tau2 over tau2 plus its squared standard error, sign reversed | residents of metros (core-based statistical areas) with at least 40 tracts that have a life expectancy estimate and an ADI rank | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population |   | 267 metros | none (each metro counts once) | USALEEP life expectancy |   |
| metro_floor | 40 | tracts | minimum number of tracts with a life expectancy estimate and an ADI rank for a metro to enter (rule) | metros | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population |   |   | none (definition) | USALEEP life expectancy |   |
| metro_floor_high | 100 | tracts | highest minimum number of tracts tried (rule) | metros | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population |   |   | none (definition) | USALEEP life expectancy |   |
| metro_floor_low | 25 | tracts | lowest minimum number of tracts tried (rule) | metros | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population |   |   | none (definition) | USALEEP life expectancy |   |
| metro_median | 1.20 | years of life expectancy at birth lower per ten-point step in the ADI national percentile | median over metros of the metro slope, sign reversed; bootstrap interval over metros | residents of metros (core-based statistical areas) with at least 40 tracts that have a life expectancy estimate and an ADI rank | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population | 95% CI 1.14 to 1.23 | 267 metros | none (each metro counts once) | USALEEP life expectancy |   |
| metro_n_clustered | 64 | metros | count of metros with at least five counties, whose errors are clustered on counties (the others use HC3 errors) | residents of metros (core-based statistical areas) with at least 40 tracts that have a life expectancy estimate and an ADI rank | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population |   |   | none (full count) | USALEEP life expectancy |   |
| metro_n_lower | 267 | metros | count of metros with life expectancy lower in more deprived tracts: the reported slope (years lower per ten-point step) is above zero | residents of metros (core-based statistical areas) with at least 40 tracts that have a life expectancy estimate and an ADI rank | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population |   | 267 metros | none (full count) | USALEEP life expectancy |   |
| metro_n_sig | 260 | metros | count of metros whose 95% interval for the reported slope lies above zero | residents of metros (core-based statistical areas) with at least 40 tracts that have a life expectancy estimate and an ADI rank | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population |   | 267 metros | none (full count) | USALEEP life expectancy |   |
| metro_noise | 48% | percent of the variance of the estimated metro slopes | mean squared standard error of the metro slopes over the variance of the estimated slopes across metros | residents of metros (core-based statistical areas) with at least 40 tracts that have a life expectancy estimate and an ADI rank | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population |   | 267 metros | none (each metro counts once) | USALEEP life expectancy | The standard errors treat tracts as independent in metros with fewer than five counties, so this share may be too small. |
| metro_pop_share | 82% | percent of residents | share of residents of the tract sample who live in the metros of the analysis | residents of non-institutional tracts in the 50 states and DC with a life expectancy estimate and an ADI rank | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population |   | 267 metros | ACS 2019-2023 tract population | USALEEP life expectancy |   |
| metro_ratio | 2.4 | ratio of slopes | 95th over 5th percentile of the estimated metro slopes, sign reversed | residents of metros (core-based statistical areas) with at least 40 tracts that have a life expectancy estimate and an ADI rank | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population |   | 267 metros | none (each metro counts once) | USALEEP life expectancy | A ratio of two percentiles. |
| metro_sd | 0.31 | years of life expectancy at birth lower per ten-point step in the ADI national percentile | standard deviation across metros of the estimated metro slope | residents of metros (core-based statistical areas) with at least 40 tracts that have a life expectancy estimate and an ADI rank | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population |   | 267 metros | none (each metro counts once) | USALEEP life expectancy |   |
| metro_steep_edge | 1.67 | years of life expectancy at birth lower per ten-point step in the ADI national percentile | 95th percentile over metros of the estimated metro slope, sign reversed | residents of metros (core-based statistical areas) with at least 40 tracts that have a life expectancy estimate and an ADI rank | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population |   | 267 metros | none (each metro counts once) | USALEEP life expectancy | A percentile of estimates, which are stretched by estimation error. |
| metro_steep_edge_sh | 1.46 | years of life expectancy at birth lower per ten-point step in the ADI national percentile | 95th percentile over metros of the metro slope shrunk toward the mean by tau2 over tau2 plus its squared standard error, sign reversed | residents of metros (core-based statistical areas) with at least 40 tracts that have a life expectancy estimate and an ADI rank | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population |   | 267 metros | none (each metro counts once) | USALEEP life expectancy |   |
| metro_tau | 0.22 | years of life expectancy at birth lower per ten-point step in the ADI national percentile | estimated standard deviation across metros of the true metro slope: the square root of the variance of the estimates minus the mean squared standard error | residents of metros (core-based statistical areas) with at least 40 tracts that have a life expectancy estimate and an ADI rank | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population |   | 267 metros | none (each metro counts once) | USALEEP life expectancy |   |
| n_ladder_county_tracts | 77,722 | tracts | count of tracts in the county fixed-effects regression | residents of non-institutional tracts in the 50 states and DC with a life expectancy estimate and an ADI rank, in counties with at least two such tracts | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population |   |   | none (full count) | USALEEP life expectancy | A tract that is alone in its county has no within-county contrast and leaves the fit. |
| n_ladder_tracts | 77,936 | tracts | count of tracts in the regressions | residents of non-institutional tracts in the 50 states and DC with a life expectancy estimate and an ADI rank | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population |   |   | none (full count) | USALEEP life expectancy |   |
| n_metros | 267 | metros | count of metros | residents of metros (core-based statistical areas) with at least 40 tracts that have a life expectancy estimate and an ADI rank | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population |   |   | none (full count) | USALEEP life expectancy | Metropolitan and micropolitan statistical areas; each tract takes the area of its county (Census county file). |
| n_metros_100 | 134 | metros | count of metros at this minimum number of tracts | core-based statistical areas with at least 100 tracts that have a life expectancy estimate and an ADI rank | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population |   |   | none (full count) | USALEEP life expectancy |   |
| n_metros_25 | 403 | metros | count of metros at this minimum number of tracts | core-based statistical areas with at least 25 tracts that have a life expectancy estimate and an ADI rank | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population |   |   | none (full count) | USALEEP life expectancy |   |
| pca1_max | 0.41 | absolute loading on the first component | largest absolute loading of a measure on the first component | tracts in the 50 states and DC with an ACS population above zero, all nine measures and an ADI rank | ACS 2019-2023; ADI 2023 |   | 82,545 tracts | none (each tract counts once) | American Community Survey |   |
| pca2_col | +0.35 | loading on the second component | loading of adults with a bachelor's degree or more on the second component, oriented so that the college share loads positively | tracts in the 50 states and DC with an ACS population above zero, all nine measures and an ADI rank | ACS 2019-2023; ADI 2023 |   | 82,545 tracts | none (each tract counts once) | American Community Survey |   |
| pca2_lfp | +0.56 | loading on the second component | loading of labor-force participation on the second component, oriented so that the college share loads positively | tracts in the 50 states and DC with an ACS population above zero, all nine measures and an ADI rank | ACS 2019-2023; ADI 2023 |   | 82,545 tracts | none (each tract counts once) | American Community Survey |   |
| pca2_own | −0.59 | loading on the second component | loading of homeownership on the second component, oriented so that the college share loads positively | tracts in the 50 states and DC with an ACS population above zero, all nine measures and an ADI rank | ACS 2019-2023; ADI 2023 |   | 82,545 tracts | none (each tract counts once) | American Community Survey |   |
| pca_alpha | 0.87 | standardized Cronbach's alpha | alpha from the mean correlation between the nine measures after each is signed so that higher means more deprived; bootstrap interval over counties | tracts in the 50 states and DC with an ACS population above zero, all nine measures and an ADI rank | ACS 2019-2023; ADI 2023 | 95% CI 0.86 to 0.87 | 82,545 tracts | none (each tract counts once) | American Community Survey |   |
| pca_cum12 | 67.2% | percent of the variance of the nine standardized measures | sum of the first two eigenvalues of the correlation matrix of the nine standardized measures over nine (each tract counted once); bootstrap interval over counties | tracts in the 50 states and DC with an ACS population above zero, all nine measures and an ADI rank | ACS 2019-2023; ADI 2023 | 95% CI 66.5% to 68.0% | 82,545 tracts | none (each tract counts once) | American Community Survey |   |
| pca_dims | 2 | components | count of components with an eigenvalue above one (the Kaiser rule) | tracts in the 50 states and DC with an ACS population above zero, all nine measures and an ADI rank | ACS 2019-2023; ADI 2023 |   | 82,545 tracts | none (full count) | American Community Survey |   |
| pca_evr1 | 51.3% | percent of the variance of the nine standardized measures | first eigenvalue of the correlation matrix of the nine standardized measures over nine (each tract counted once); bootstrap interval over counties | tracts in the 50 states and DC with an ACS population above zero, all nine measures and an ADI rank | ACS 2019-2023; ADI 2023 | 95% CI 50.4% to 52.1% | 82,545 tracts | none (each tract counts once) | American Community Survey |   |
| pca_evr1_w | 51.4% | percent of the variance of the nine standardized measures | first eigenvalue over nine, from the correlation matrix weighted by ACS tract population | tracts in the 50 states and DC with an ACS population above zero, all nine measures and an ADI rank | ACS 2019-2023; ADI 2023 |   | 82,545 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| pca_evr2 | 16.0% | percent of the variance of the nine standardized measures | second eigenvalue of the correlation matrix of the nine standardized measures over nine (each tract counted once); bootstrap interval over counties | tracts in the 50 states and DC with an ACS population above zero, all nine measures and an ADI rank | ACS 2019-2023; ADI 2023 | 95% CI 15.5% to 16.4% | 82,545 tracts | none (each tract counts once) | American Community Survey |   |
| pca_jack_max | 2.3 | percentile points | largest, over the nine measures, of the median absolute change in a tract's index percentile when that measure is dropped and the first component is refitted | tracts in the 50 states and DC with an ACS population above zero, all nine measures and an ADI rank | ACS 2019-2023; ADI 2023 |   | 82,545 tracts | none (each tract counts once) | American Community Survey |   |
| pca_n | 82,545 | tracts | count of tracts in the index | tracts in the 50 states and DC with an ACS population above zero, all nine measures and an ADI rank | ACS 2019-2023; ADI 2023 |   |   | none (full count) | American Community Survey |   |
| pca_n_dropped | 1,066 | tracts | count of such tracts that lack a measure or an ADI rank | tracts in the 50 states and DC with an ACS population above zero | ACS 2019-2023; ADI 2023 |   |   | none (full count) | American Community Survey |   |
| pca_r_collhs | −0.90 | correlation coefficient | correlation across tracts of the share of adults with a bachelor's degree or more with the share with a high school diploma or less, each tract counted once; bootstrap interval over counties | tracts in the 50 states and DC with an ACS population above zero, all nine measures and an ADI rank | ACS 2019-2023; ADI 2023 | 95% CI −0.91 to −0.90 | 82,545 tracts | none (each tract counts once) | American Community Survey |   |
| pca_rho_adi | +0.68 | Spearman rank correlation | rank correlation across tracts of the first-component index with the ADI national percentile, each tract counted once; bootstrap interval over counties holding the ranks fixed | tracts in the 50 states and DC with an ACS population above zero, all nine measures and an ADI rank | ACS 2019-2023; ADI 2023 | 95% CI 0.61 to 0.74 | 82,545 tracts | none (each tract counts once) | American Community Survey |   |
| pca_rho_avg | +0.997 | Spearman rank correlation | rank correlation across tracts of the first-component index with the equal-weight average of the nine signed standardized measures, each tract counted once; bootstrap interval over counties holding the ranks fixed | tracts in the 50 states and DC with an ACS population above zero, all nine measures and an ADI rank | ACS 2019-2023; ADI 2023 | 95% CI 0.996 to 0.997 | 82,545 tracts | none (each tract counts once) | American Community Survey |   |
| pca_rho_fa | +0.994 | Spearman rank correlation | rank correlation across tracts of the first-component index with the score of a one-factor factor analysis of the nine measures, each tract counted once; bootstrap interval over counties holding the ranks fixed | tracts in the 50 states and DC with an ACS population above zero, all nine measures and an ADI rank | ACS 2019-2023; ADI 2023 | 95% CI 0.993 to 0.994 | 82,545 tracts | none (each tract counts once) | American Community Survey |   |
| r_disp_u | −0.14 | correlation coefficient | correlation across metros of the metro slope (years lower per ten-point step) with the dispersion of tract ADI in the metro (population-weighted standard deviation), each metro counted once; bootstrap interval over metros | residents of metros (core-based statistical areas) with at least 40 tracts that have a life expectancy estimate and an ADI rank | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population | 95% CI −0.27 to 0.01 | 267 metros | none (each metro counts once) | USALEEP life expectancy |   |
| r_disp_w | −0.21 | correlation coefficient | correlation across metros of the metro slope (years lower per ten-point step) with the dispersion of tract ADI in the metro (population-weighted standard deviation), weighted by metro population; bootstrap interval over metros | residents of metros (core-based statistical areas) with at least 40 tracts that have a life expectancy estimate and an ADI rank | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population | 95% CI −0.46 to 0.12 | 267 metros | ACS 2019-2023 tract population summed to the metro | USALEEP life expectancy |   |
| r_income_u | +0.05 | correlation coefficient | correlation across metros of the metro slope (years lower per ten-point step) with the log of the metro median household income (the population-weighted median of its tracts' median incomes), each metro counted once; bootstrap interval over metros | residents of metros (core-based statistical areas) with at least 40 tracts that have a life expectancy estimate and an ADI rank | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population | 95% CI −0.11 to 0.19 | 267 metros | none (each metro counts once) | USALEEP life expectancy |   |
| r_income_w | +0.13 | correlation coefficient | correlation across metros of the metro slope (years lower per ten-point step) with the log of the metro median household income (the population-weighted median of its tracts' median incomes), weighted by metro population; bootstrap interval over metros | residents of metros (core-based statistical areas) with at least 40 tracts that have a life expectancy estimate and an ADI rank | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population | 95% CI −0.24 to 0.46 | 267 metros | ACS 2019-2023 tract population summed to the metro | USALEEP life expectancy |   |
| r_size_u | −0.06 | correlation coefficient | correlation across metros of the metro slope (years lower per ten-point step) with the log of the metro population, each metro counted once; bootstrap interval over metros | residents of metros (core-based statistical areas) with at least 40 tracts that have a life expectancy estimate and an ADI rank | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population | 95% CI −0.18 to 0.08 | 267 metros | none (each metro counts once) | USALEEP life expectancy |   |
| r_size_w | −0.20 | correlation coefficient | correlation across metros of the metro slope (years lower per ten-point step) with the log of the metro population, weighted by metro population; bootstrap interval over metros | residents of metros (core-based statistical areas) with at least 40 tracts that have a life expectancy estimate and an ADI rank | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts); ADI 2023; ACS 2019-2023 population | 95% CI −0.47 to 0.10 | 267 metros | ACS 2019-2023 tract population summed to the metro | USALEEP life expectancy |   |
| range_floor | 20 | tracts | minimum number of tracts with an ADI rank for a county to enter (rule) | counties and Connecticut planning regions with at least 20 tracts that have an ADI rank | ACS 2019-2023; ADI 2023 |   |   | none (definition) | Area Deprivation Index |   |
| range_floor_high | 40 | tracts | a minimum number of tracts with an ADI rank tried in the sweep (rule) | counties and Connecticut planning regions with at least 20 tracts that have an ADI rank | ACS 2019-2023; ADI 2023 |   |   | none (definition) | Area Deprivation Index |   |
| range_floor_low | 10 | tracts | a minimum number of tracts with an ADI rank tried in the sweep (rule) | counties and Connecticut planning regions with at least 20 tracts that have an ADI rank | ACS 2019-2023; ADI 2023 |   |   | none (definition) | Area Deprivation Index |   |
| range_median | 39 | ADI percentile points between the 10th and 90th percentile of tract ADI | median over counties of the population-weighted 90th minus 10th percentile of tract ADI | counties and Connecticut planning regions with at least 20 tracts that have an ADI rank | ACS 2019-2023; ADI 2023 |   | 757 counties | none (each county counts once) | Area Deprivation Index |   |
| range_median_10 | 34 | ADI percentile points between the 10th and 90th percentile of tract ADI | median over counties of the population-weighted 90th minus 10th percentile of tract ADI | counties and Connecticut planning regions with at least 10 tracts that have an ADI rank | ACS 2019-2023; ADI 2023 |   | 1,319 counties | none (each county counts once) | Area Deprivation Index |   |
| range_median_40 | 43 | ADI percentile points between the 10th and 90th percentile of tract ADI | median over counties of the population-weighted 90th minus 10th percentile of tract ADI | counties and Connecticut planning regions with at least 40 tracts that have an ADI rank | ACS 2019-2023; ADI 2023 |   | 421 counties | none (each county counts once) | Area Deprivation Index |   |
| range_n_counties | 757 | counties and Connecticut planning regions | count of counties | counties and Connecticut planning regions with at least 20 tracts that have an ADI rank | ACS 2019-2023; ADI 2023 |   |   | none (full count) | Area Deprivation Index |   |
| range_n_counties_10 | 1,319 | counties and Connecticut planning regions | count of counties | counties and Connecticut planning regions with at least 10 tracts that have an ADI rank | ACS 2019-2023; ADI 2023 |   |   | none (full count) | Area Deprivation Index |   |
| range_n_counties_40 | 421 | counties and Connecticut planning regions | count of counties | counties and Connecticut planning regions with at least 40 tracts that have an ADI rank | ACS 2019-2023; ADI 2023 |   |   | none (full count) | Area Deprivation Index |   |
| range_r_unw | +0.27 | correlation coefficient | correlation across counties of the population-weighted mean tract ADI with the 90th minus 10th percentile of tract ADI, each county counted once; bootstrap interval over states | counties and Connecticut planning regions with at least 20 tracts that have an ADI rank | ACS 2019-2023; ADI 2023 | 95% CI 0.06 to 0.44 | 757 counties | none (each county counts once) | Area Deprivation Index |   |
| range_r_unw_10 | −0.08 | correlation coefficient | correlation across counties of the population-weighted mean tract ADI with the 90th minus 10th percentile of tract ADI, each county counted once; bootstrap interval over states | counties and Connecticut planning regions with at least 10 tracts that have an ADI rank | ACS 2019-2023; ADI 2023 | 95% CI −0.25 to 0.08 | 1,319 counties | none (each county counts once) | Area Deprivation Index |   |
| range_r_unw_40 | +0.62 | correlation coefficient | correlation across counties of the population-weighted mean tract ADI with the 90th minus 10th percentile of tract ADI, each county counted once; bootstrap interval over states | counties and Connecticut planning regions with at least 40 tracts that have an ADI rank | ACS 2019-2023; ADI 2023 | 95% CI 0.45 to 0.74 | 421 counties | none (each county counts once) | Area Deprivation Index |   |
| range_r_w | +0.63 | correlation coefficient | correlation across counties of the population-weighted mean tract ADI with the 90th minus 10th percentile of tract ADI, weighted by county population; bootstrap interval over states | counties and Connecticut planning regions with at least 20 tracts that have an ADI rank | ACS 2019-2023; ADI 2023 | 95% CI 0.37 to 0.75 | 757 counties | ACS 2019-2023 county population | Area Deprivation Index |   |
| range_r_w_10 | +0.47 | correlation coefficient | correlation across counties of the population-weighted mean tract ADI with the 90th minus 10th percentile of tract ADI, weighted by county population; bootstrap interval over states | counties and Connecticut planning regions with at least 10 tracts that have an ADI rank | ACS 2019-2023; ADI 2023 | 95% CI 0.16 to 0.63 | 1,319 counties | ACS 2019-2023 county population | Area Deprivation Index |   |
| range_r_w_40 | +0.77 | correlation coefficient | correlation across counties of the population-weighted mean tract ADI with the 90th minus 10th percentile of tract ADI, weighted by county population; bootstrap interval over states | counties and Connecticut planning regions with at least 40 tracts that have an ADI rank | ACS 2019-2023; ADI 2023 | 95% CI 0.58 to 0.85 | 421 counties | ACS 2019-2023 county population | Area Deprivation Index |   |
| smoke_median | −4.7 | points of prevalence per doubling of tract median household income | median over states of the state coefficient: a population-weighted regression across the state's tracts of tract prevalence on the log of median household income, with tract shares of residents ages 65 and over, Black alone and Hispanic as controls, times the natural log of 2 | adults: current cigarette smoking, states and the District of Columbia with at least 60 tracts that have income, the three controls and the measure | BRFSS 2022 (PLACES 2024 release); ACS 2019-2023 |   | 51 states | adult population of the tract (PLACES) | CDC PLACES | A median of the state estimates. |
| supp_cell_max | 9 | births | most births a withheld cell can hold (the published threshold minus one) | withheld county-year cells | 2018 to 2024 |   |   | none (definition) | CDC WONDER natality |   |
| supp_floor | 10 | births | CDC WONDER withholds a cell with fewer births than this (rule) | county-year cells | 2018 to 2024 |   |   | none (definition) | CDC WONDER natality |   |
| supp_heavy | 77.6% | percent of cells | share of the cells of this class that WONDER withholds | county-year cells of the 578 counties that WONDER names in every year in the heaviest class | 2018 to 2024 |   | 4,046 cells | none (full count) | CDC WONDER natality |   |
| supp_light | 71.3% | percent of cells | share of the cells of this class that WONDER withholds | county-year cells of the 578 counties that WONDER names in every year in the lightest class | 2018 to 2024 |   | 4,046 cells | none (full count) | CDC WONDER natality |   |
| supp_max_heavy | 76% | percent of the births in the class | largest share of the class's births the withheld cells could hold: 9 births in each withheld cell, over the births in published cells plus those | county-year cells of the 578 counties that WONDER names in every year in the heaviest class | 2018 to 2024 |   | 3,141 cells | none (an upper limit from the cell count) | CDC WONDER natality |   |
| supp_max_light | 60% | percent of the births in the class | largest share of the class's births the withheld cells could hold: 9 births in each withheld cell, over the births in published cells plus those | county-year cells of the 578 counties that WONDER names in every year in the lightest class | 2018 to 2024 |   | 2,885 cells | none (an upper limit from the cell count) | CDC WONDER natality |   |
| supp_middle | 0 | cells | count of withheld cells in the five middle birthweight classes | county-year cells of the 578 counties that WONDER names in every year in the five birthweight classes from 2,000 to 4,499 grams | 2018 to 2024 |   | 20,230 cells | none (full count) | CDC WONDER natality |   |
| supp_middle_share | 0.0% | percent of cells | share of the cells of the five middle classes that WONDER withholds | county-year cells of the 578 counties that WONDER names in every year in the five birthweight classes from 2,000 to 4,499 grams | 2018 to 2024 |   | 20,230 cells | none (full count) | CDC WONDER natality |   |
| supp_n | 11,250 | cells | count of cells withheld by WONDER (1 to 9 births) | county-year cells of the 578 counties that WONDER names in every year by birthweight class | 2018 to 2024 |   | 48,552 cells | none (full count) | CDC WONDER natality |   |
| supp_n_cells | 48,552 | cells | count of county-year-class cells | county-year cells of the 578 counties that WONDER names in every year by birthweight class | 2018 to 2024 |   |   | none (full count) | CDC WONDER natality |   |
| supp_n_counties | 578 | counties | count of counties | counties that WONDER names in every year 2018 to 2024 | 2018 to 2024 |   |   | none (full count) | CDC WONDER natality |   |
| supp_nostated | 65.1% | percent of cells | share of the cells of this class that WONDER withholds | county-year cells of the 578 counties that WONDER names in every year in the class for weight not stated | 2018 to 2024 |   | 4,046 cells | none (full count) | CDC WONDER natality |   |
| supp_overall | 23.2% | percent of cells | share of county-year-class cells withheld | county-year cells of the 578 counties that WONDER names in every year by birthweight class | 2018 to 2024 |   | 48,552 cells | none (full count) | CDC WONDER natality |   |
| supp_withheld_light | 2,885 | cells | count of cells withheld (1 to 9 births) | county-year cells of the 578 counties that WONDER names in every year in the lightest class | 2018 to 2024 |   | 4,046 cells | none (full count) | CDC WONDER natality |   |
| supp_zero_light | 350 | cells | count of cells published as zero births | county-year cells of the 578 counties that WONDER names in every year in the lightest class | 2018 to 2024 |   | 4,046 cells | none (full count) | CDC WONDER natality |   |
| vsplit_county | 15.9 | percent of the variance of tract life expectancy | share of the population-weighted variance of tract life expectancy at birth between counties within states (sum-of-squares decomposition) | residents of non-institutional tracts with a life expectancy estimate (50 states and DC) | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts) |   | 78,008 tracts | ACS 2019-2023 tract population | USALEEP life expectancy | Connecticut's nine planning regions stand in for its counties. The tract estimates are modeled; their mean squared standard error equals 26.7 percent of the total variance and sits inside the within-county share. |
| vsplit_n_counties | 3,112 | counties and Connecticut planning regions | count of counties with at least one tract in the variance split | residents of non-institutional tracts with a life expectancy estimate (50 states and DC) | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts) |   |   | none (full count) | USALEEP life expectancy |   |
| vsplit_n_tracts | 78,008 | tracts | count of tracts in the variance split | residents of non-institutional tracts with a life expectancy estimate (50 states and DC) | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts) |   |   | none (full count) | USALEEP life expectancy |   |
| vsplit_noise | 26.7 | percent of the variance of tract life expectancy | population-weighted mean squared standard error of tract life expectancy as a share of the total variance | residents of non-institutional tracts with a life expectancy estimate (50 states and DC) | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts) |   | 78,008 tracts | ACS 2019-2023 tract population | USALEEP life expectancy | The 2020-tract standard errors are mapping grade and understate the error. |
| vsplit_state | 13.7 | percent of the variance of tract life expectancy | share of the population-weighted variance of tract life expectancy at birth between states (sum-of-squares decomposition) | residents of non-institutional tracts with a life expectancy estimate (50 states and DC) | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts) |   | 78,008 tracts | ACS 2019-2023 tract population | USALEEP life expectancy | Connecticut's nine planning regions stand in for its counties. The tract estimates are modeled; their mean squared standard error equals 26.7 percent of the total variance and sits inside the within-county share. |
| vsplit_within | 70.3 | percent of the variance of tract life expectancy | share of the population-weighted variance of tract life expectancy at birth between tracts within counties (sum-of-squares decomposition) | residents of non-institutional tracts with a life expectancy estimate (50 states and DC) | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts) |   | 78,008 tracts | ACS 2019-2023 tract population | USALEEP life expectancy | Connecticut's nine planning regions stand in for its counties. The tract estimates are modeled; their mean squared standard error equals 26.7 percent of the total variance and sits inside the within-county share. |
| vsplit_within_signal | 60% | percent of the variance net of estimation error | within-county share of the variance of tract life expectancy after subtracting the mean squared standard error from the total and from the within-county part | residents of non-institutional tracts with a life expectancy estimate (50 states and DC) | deaths in 2010 to 2015 (USALEEP, area-weighted onto 2020 tracts) |   | 78,008 tracts | ACS 2019-2023 tract population | USALEEP life expectancy | Takes the published standard errors at face value; the 2020-tract errors are mapping grade and understate the noise, so the share of the true differences that lies within counties is lower. |
| wonder_county_min | 100,000 | residents | WONDER names a county only if it has at least this many residents (rule) | counties named individually in CDC WONDER natality | 2018 to 2024 |   |   | none (definition) | CDC WONDER natality |   |
| zoo_age_cut | 65 | years of age | residents of this age and over are counted in the first control (definition) | states and the District of Columbia with at least 60 tracts that have income, the three controls and the measure | BRFSS 2022 (PLACES 2024 release); ACS 2019-2023 |   |   | none (definition) | American Community Survey |   |
| zoo_floor | 60 | tracts | minimum number of tracts with every variable for a state to enter (rule) | states and the District of Columbia with at least 60 tracts that have income, the three controls and the measure | BRFSS 2022 (PLACES 2024 release); ACS 2019-2023 |   |   | none (definition) | CDC PLACES |   |
| zoo_n_fits | 306 | regressions | count of state-by-measure regressions | states and the District of Columbia with at least 60 tracts that have income, the three controls and the measure | BRFSS 2022 (PLACES 2024 release); ACS 2019-2023 |   |   | none (full count) | CDC PLACES |   |
| zoo_n_states | 51 | states | count of states, counting the District of Columbia as one | states and the District of Columbia with at least 60 tracts that have income, the three controls and the measure | BRFSS 2022 (PLACES 2024 release); ACS 2019-2023 |   |   | none (full count) | CDC PLACES |   |
| zoo_neg_five | 51 | states | count of states where the coefficients for smoking, obesity, diabetes, frequent mental distress and physical inactivity are all below zero | states and the District of Columbia with at least 60 tracts that have income, the three controls and the measure | BRFSS 2022 (PLACES 2024 release); ACS 2019-2023 |   | 51 states | none (full count) | CDC PLACES |   |
| zoo_r_het | −0.73 | correlation coefficient | correlation across states of the smoking coefficient with the state's population-weighted mean smoking prevalence, each state counted once; bootstrap interval over states | adults: current cigarette smoking, states and the District of Columbia with at least 60 tracts that have income, the three controls and the measure | BRFSS 2022 (PLACES 2024 release); ACS 2019-2023 | 95% CI −0.84 to −0.60 | 51 states | none (each state counts once) | CDC PLACES |   |
| zoo_r_het_rel | −0.18 | correlation coefficient | correlation across states of the smoking coefficient divided by the state's population-weighted mean smoking prevalence with that mean, each state counted once; bootstrap interval over states | adults: current cigarette smoking, states and the District of Columbia with at least 60 tracts that have income, the three controls and the measure | BRFSS 2022 (PLACES 2024 release); ACS 2019-2023 | 95% CI −0.42 to 0.08 | 51 states | none (each state counts once) | CDC PLACES |   |

## Definitions

- Area Deprivation Index (ADI): A score that ranks neighborhoods by income, education, employment and housing conditions drawn from the American Community Survey. In this book a higher national percentile means a more deprived neighborhood.
- USALEEP: The U.S. Small-area Life Expectancy Estimates Project, a National Center for Health Statistics release of life expectancy at birth for most census tracts. It covers deaths from 2010 to 2015 only and has not been updated.
- Census tract: A small area drawn by the Census Bureau to stand in for a neighborhood. Boundaries are redrawn after each decennial census.
- Fixed effects: A regression adjustment that gives each group, such as each county, its own baseline. The estimate then rests on differences within groups.
- Modeled estimate: A figure produced by a statistical model instead of a direct count or measurement, usually for small areas where direct data are thin, borrowing strength from surveys, neighboring areas or other data. Neighboring areas can share errors, so single extremes call for caution.
- Principal components: A method that combines many correlated measures into a few summary scores, ordered by how much of the variation among places each captures. The first component is the single weighted combination that captures the most.
- MAUP (modifiable areal unit problem): The same data can show different patterns depending on how the map is divided, for example into tracts, counties or states.
- Suppression: The withholding of a number because the count behind it is too small to publish without risking the identification of a person. A suppressed value is missing, not zero.
- Ecological inference: Drawing conclusions about individuals from data about places or groups. A pattern across counties need not hold for the people inside them.

## Methods

Life expectancy at birth for deaths in 2010 to 2015 is regressed on the 2023 Area Deprivation Index (ADI) across tracts, within counties and within metros, so the outcome and the deprivation measure come from different periods. Tract life expectancy is a modeled estimate area-weighted from 2010 onto 2020 tracts, and the sample is the 77,936 tracts with an ADI rank in the 50 states and DC after institutional tracts are dropped. The variance split has no sampling interval, and the coefficient zoo and the count of withheld birth cells use other sources, described under their own headings.

### Which tracts and which years

Tract life expectancy at birth is the USALEEP estimate for deaths in 2010 to 2015, area-weighted from 2010 tracts onto 2020 tracts. Institutional tracts (at least 25 percent of residents in group quarters, or a known case) are dropped, and a tract needs an ACS 2019-2023 population above zero and a county, which leaves 77,936 tracts with an ADI rank (78,008 counting those without one) in the 50 states and DC. Puerto Rico has no USALEEP estimates.

Explains Table 1.

### Measures and weights

Weights are the ACS 2019-2023 tract population. The ADI is the 2023 national percentile (1 least, 100 most deprived).

The nine measures of the index are the poverty rate, the share of households on cash assistance or SNAP, the unemployment rate, the share of adults with a high school diploma or less, the share of households headed by a single parent, the log of median household income, the share of adults with a bachelor's degree or more, labor-force participation and homeownership, from ACS 2019-2023 tracts. Each is standardized across the 82,545 tracts, each counted once; components are the eigenvectors of the correlation matrix.

Explains Table 5.

### Metros, counties and grains of geography

A metro is a core-based statistical area (metropolitan or micropolitan) coded on the Census 2020 county file; each tract takes the area of its county, so Connecticut's planning regions work as counties. A metro needs at least 40 tracts with a life expectancy and an ADI rank.

The six pairs of measures use one sample; county and state values are population-weighted means of tract values, and each correlation is computed weighted by its unit's population and counting each unit once, with bootstrap intervals over counties (tracts) or states (counties and states).

The variance split is the population-weighted sum-of-squares split of b2.canon.variance_split(), computed in the same way for the ADI and the index; it has no sampling interval.

County ranges are population-weighted 10th and 90th percentiles of tract ADI for counties with at least 20 tracts that have an ADI rank; the count of counties, the median range and the correlation of mean ADI with the range are also given at floors of 10 and 40 tracts.

Explains Table 2, Table 8, Table 6 and Table 7.

### The coefficient zoo and withheld birth cells

The coefficient zoo is b2.canon.coefficient_zoo(), the same function the chapter on health behaviors and the supply of care calls: for each state with at least 60 tracts that have every variable, a weighted regression of PLACES prevalence on the log of ACS median household income with the shares of residents ages 65 and over, Black alone and Hispanic as controls, weighted by adults, with county-clustered errors (HC3 where a state has fewer than five counties), times the natural log of 2.

The suppression count uses WONDER natality, county by birthweight class, for the 578 counties named in every year 2018 to 2024; the file lists a withheld cell as missing, and a published 0 is a real zero. Withheld cells are counted and left out; the bound uses only the published threshold of 10 births and the births in published cells.

Explains Table 4 and Table 9.

### How the slopes and their errors are estimated

Every slope is a weighted least-squares slope of life expectancy on the ADI, times ten. The ladder uses pyfixest with absorbed fixed effects and CRV1 errors, clustered on states for the first two rows and counties for the third, with t intervals on one fewer degree of freedom than there are clusters; the first two are checked against b2.stats.wls with state dummies and the county fit against weighted demeaning.

Each metro slope comes from b2.stats.wls, clustered on counties when the metro has at least five counties and with HC3 errors otherwise.

The variance of the true metro slopes is the variance of the estimated slopes minus their mean squared standard error, and each slope is shrunk toward the mean by that variance over itself plus its squared standard error.

Explains Table 1, Table 2 and Table 3.

### Bootstrap intervals, draws and seeds

Metro correlations carry bootstrap intervals over metros (2,000 draws, seed 12345).

Intervals for eigenvalues, loadings, alpha and the correlation come from a bootstrap over counties (2,000 draws, seed 12345) that recomputes the correlation matrix. Rank correlations hold ranks fixed in the bootstrap.

## What these data cannot show

Every estimate in this chapter describes how an outcome is arranged across places. The slopes, the metro regressions, the index and the variance split compare tracts, counties and states, and none says what would happen to a person who moved or to a place that changed. A gradient of tract life expectancy on the ADI says nothing about individuals, which is the ecological inference problem. The steeper slope within counties is compatible with more than one explanation that the data cannot separate. Life expectancy is a modeled estimate for deaths in 2010 to 2015 that has not been updated. The 2020-tract file is area-weighted from 2010 tracts, so the estimate for a single tract is not reliable.

The metro slopes cover the residents of metros with enough tracts and say nothing about the rest. Their intervals use HC3 errors in most metros and may be too narrow, which would make the share of the spread that is estimation error larger. The ADI is a national percentile and understates deprivation in high-cost cities (Chapter 1.1, What the deprivation index measures (https://vishalsingh.org/zipcode/anatomy-of-deprivation)), which bears on the metro slopes. Puerto Rico is not covered, and Connecticut's planning regions stand in for its counties in the metros, the variance split and the county ranges.

The index is internally coherent, and the data hold no independent measure of what residents would call deprivation, so they cannot show that it is valid. The ACS values behind the index, the ADI and the area-size correlations are survey estimates with sampling error that is largest in small tracts. The staged file carries no margins of error, so every analysis here treats them as exact. The ADI and the PLACES estimates draw on the same census inputs as the measures they are compared with, so part of each association is built in, and the data cannot say how much.

Weights can decide a result. The correlations across states for unemployment and poverty, and for the college share and life expectancy, change direction between the two weightings in the area-size section, and neither weighting is the correct one. The suppression figure counts withheld cells and cannot show what they hold. Counties below 100,000 residents are pooled by WONDER and do not appear as cells, and the bound on the share held in withheld cells is the largest value the threshold allows.

## References

1. Kind, Amy J. H.; Buckingham, William R. (2018). Making Neighborhood-Disadvantage Metrics Accessible: The Neighborhood Atlas. New England Journal of Medicine 378(26): 2456-2458. https://doi.org/10.1056/NEJMp1802313
2. Arias, Elizabeth; Escobedo, Luis A.; Kennedy, Jiaquan; Fu, Chenxi; Cisewski, Jodi (2018). U.S. Small-area Life Expectancy Estimates Project: Methodology and Results Summary. Vital and Health Statistics, Series 2, Number 181, National Center for Health Statistics. https://www.cdc.gov/nchs/data/series/sr_02/sr02_181.pdf
3. Cameron, A. Colin; Miller, Douglas L. (2015). A Practitioner's Guide to Cluster-Robust Inference. Journal of Human Resources 50(2): 317-372. https://doi.org/10.3368/jhr.50.2.317
4. Greenlund, Kurt J.; Lu, Hua; Wang, Yan; Matthews, Kevin A.; LeClercq, Jennifer M.; Lee, Benjamin; Carlson, Susan A. (2022). PLACES: Local Data for Better Health. Preventing Chronic Disease 19: 210459. https://doi.org/10.5888/pcd19.210459
5. Messer, Lynne C.; Laraia, Barbara A.; Kaufman, Jay S.; Eyster, Janet; Holzman, Claudia; Culhane, Jennifer; Elo, Irma; Burke, Jessica G.; O'Campo, Patricia (2006). The Development of a Standardized Neighborhood Deprivation Index. Journal of Urban Health 83(6): 1041-1062. https://doi.org/10.1007/s11524-006-9094-x
6. Kaiser, Henry F. (1960). The Application of Electronic Computers to Factor Analysis. Educational and Psychological Measurement 20(1): 141-151. https://doi.org/10.1177/001316446002000116
7. Cronbach, Lee J. (1951). Coefficient Alpha and the Internal Structure of Tests. Psychometrika 16(3): 297-334. https://doi.org/10.1007/BF02310555
8. Openshaw, Stan (1984). The Modifiable Areal Unit Problem. CATMOG 38, Concepts and Techniques in Modern Geography, Geo Books, Norwich.
9. Robinson, W. S. (1950). Ecological Correlations and the Behavior of Individuals. American Sociological Review 15(3): 351-357. https://doi.org/10.2307/2087176
10. Subramanian, S. V.; Jones, Kelvyn; Kaddour, Afamia; Krieger, Nancy (2009). Revisiting Robinson: The Perils of Individualistic and Ecologic Fallacy. International Journal of Epidemiology 38(2): 342-360. https://doi.org/10.1093/ije/dyn359

## Provenance

- Pipeline version: 0.1.0
- Build script: book2/chapters/gradients-slopes-and-variance/build.py (sha256 e3c0eb69d17fd0c8f4f36ceab0c447935b0800f5cb97899e4c225561ef54b28a)
- Built on: 2026-10-07
- Site commit: c2b1ff3309cb4a10833f884da91ac0057a66ad0a

Inputs:

| File | Size (bytes) | SHA-256, first 1 MB | SHA-256, last 1 MB |
| --- | ---: | --- | --- |
| book2/chapters/gradients-slopes-and-variance/f_pca.py | 9,345 | 5353ebb9cba70e3b87433e822f6caa660a2b724c02b4d6e0cfe56e455575bf30 | 5353ebb9cba70e3b87433e822f6caa660a2b724c02b4d6e0cfe56e455575bf30 |
| book2/chapters/gradients-slopes-and-variance/f_sample.py | 3,332 | 68db6c906b0076863c96cc3ae3b2ca569e20a783b4043a6a278cbcf78a298192 | 68db6c906b0076863c96cc3ae3b2ca569e20a783b4043a6a278cbcf78a298192 |
| book2/chapters/gradients-slopes-and-variance/f_scale.py | 7,541 | 21813ba176531a15307e31ddac981de8b5383e726f3cba3b569ff001639e1545 | 21813ba176531a15307e31ddac981de8b5383e726f3cba3b569ff001639e1545 |
| book2/chapters/gradients-slopes-and-variance/f_slopes.py | 8,806 | 4a4adf3f306b504412faa7538c4ce2983dce6f1006e7e07e8e0d4c4450c5ba6c | 4a4adf3f306b504412faa7538c4ce2983dce6f1006e7e07e8e0d4c4450c5ba6c |
| book2/chapters/gradients-slopes-and-variance/f_supp.py | 3,227 | d06aab1dbc3521d9f5243f092021f968582f4168792e14960dc2aa62a0447017 | d06aab1dbc3521d9f5243f092021f968582f4168792e14960dc2aa62a0447017 |
| data/analytic/tract_atlas.parquet | 39,659,925 | 29d5e4d943552d54aeb6ba92aadc939d9bd15bdea553c78357dbd516757be627 | 1e10d1ecf722340f23ee2212d4999a90250f85b65fa89b39e32db06d9472e22e |
| data/geo/counties_2020.parquet | 123,034,155 | 2fce837cdc2db1f2c229f2ea453c6793338a462770307e2c64e8d1fd713ec663 | dae928bf1042f3d174d2d64d6e3ec923fa3daa40aee1bb4feb104aac2fe3c04a |
| data/geo/dim_geo.parquet | 1,180,921 | 29d5e4d943552d54aeb6ba92aadc939d9bd15bdea553c78357dbd516757be627 | e58c8c488df54ab3187f5461f179322afdc49c9bd2315404003fe1272eea32c3 |
| data/staged/adi/adi_atlas_tract2020_2023.parquet | 798,078 | 5e409bc212ff871def780fa27b4cf8b616d9da420b2b9f6455517eddcd8c1a95 | 5e409bc212ff871def780fa27b4cf8b616d9da420b2b9f6455517eddcd8c1a95 |
| data/staged/births/wonder_natality_county_birthweight_2018_2024.parquet | 115,259 | 17a46844906a1ab03d02c98fd4e2aca0535a3ad4da186e0feaafb17a03dfa9ef | 17a46844906a1ab03d02c98fd4e2aca0535a3ad4da186e0feaafb17a03dfa9ef |
| data/staged/births/wonder_natality_county_total_2018_2024.parquet | 41,774 | e85d351fa72c022a05415d332185113b6badcaff1119da1d6f45c76ffe16d86d | e85d351fa72c022a05415d332185113b6badcaff1119da1d6f45c76ffe16d86d |
| data/staged/health/places_tract2020_latest.parquet | 5,919,159 | af4c3b0d1205b7d84b81676a9529ef73323b2dbeb9d54e235f5efeebfad83177 | a2f19029ee2ef032be87e0d5288d88d489afd52379279833f178baa021df3dbf |
| data/staged/life_expectancy/usaleep_tract2020.parquet | 1,316,169 | 9c9c7f103791b75eddba74e68a89804efdf2486b6b66bb4bc02a46ba5bf8bc23 | 27317107be2465d9fd5e83271dea0cd6e915ee419e2802bedc28747d0a41d38f |
| data/staged/ses/acs5_tract2020_2019_2023.parquet | 7,232,424 | 5fe6f730a35225f14d5d8abfc6b9bce58b52e545f309a7eac74f1dae74d7a31c | 7b4a2475b3852d3f91804716a04566ef8ceadacec2be9fb85f5eb642078efac6 |
| data/staged/svi/svi_tract2020_2022.parquet | 17,740,415 | cb3860d8edc2aaf1f2a02211a58f7a0b80457eb6ad9a37b5ec8abb2d5da9da10 | e371a4a696d5bed05d54ad7c5b6ca2504508805b847ec415c2322be7f66ec2b2 |

Outputs:

| File | SHA-256 |
| --- | --- |
| facts.json | 318e84614c33eaa882eec84cd8bce9f60c8f355f3c67c98d5a29aa26cdeb93cf |
| figures/coefficient_zoo.json | 644555fe41b4157a0841b8a37c8d543273ce469b0dc3067d4346bb97acf31bf0 |
| figures/county_adi_spread.json | f6454ca1e281f5e8a649b5cabc08c85c2d61abf44445b66e7a693af5efa409dd |
| figures/fe_ladder.json | e5477b2b5672b53bb02a5401ff3d4fce98dc85b843e74b98caa800bf0adf851b |
| figures/maup_correlations.json | 28cce465fe4429b41eaaa7d47597135a041799e7e041e5c7eafea7f50d909d87 |
| figures/metro_slopes_forest.json | 261d8e3e45f05da6ad4da94af932eadcde69f8d9be7511468f31fe36d14f38b8 |
| figures/nine_measure_pca.json | 5e29e1ac6c3b2dff95909ae56e795abe1ff8f6326fe6652cde3b6c2b8a32f39a |
| figures/slope_heterogeneity.json | 1a3f5ebd417cf63a702c258822e622bfee7a821a7c64866cc9b4c428a7710da6 |
| figures/variance_split.json | 5f6d353f9319dc62e2a886546d14398a61bf5d366ce1e59f01ce123f591d94c5 |
| figures/wonder_suppression_by_birthweight.json | 2c3e6b55c3719f5c39bcb5fb395dab64e63ddd20d0a3bfde02e079b93c545a5a |
| meta.json | dee469e5a873abf617d4405e8998402c4464d7cfe72cb4002f96a7743ccebfdc |
| refs.json | ac8e900619c72fda0850597fc4e7b43a50aa28644f2322fa3fdeabad2f039f8f |

## Machine formats

- Report (HTML): https://vishalsingh.org/zipcode/gradients-slopes-and-variance/report
- Report (Markdown): https://vishalsingh.org/zipcode/gradients-slopes-and-variance/report.md
- Every estimate (JSON): https://vishalsingh.org/zipcode/report/gradients-slopes-and-variance/facts.json
- Table 1, figure "fe_ladder" (JSON): https://vishalsingh.org/zipcode/report/gradients-slopes-and-variance/figures/fe_ladder.json
- Table 2, figure "metro_slopes_forest" (JSON): https://vishalsingh.org/zipcode/report/gradients-slopes-and-variance/figures/metro_slopes_forest.json
- Table 3, figure "slope_heterogeneity" (JSON): https://vishalsingh.org/zipcode/report/gradients-slopes-and-variance/figures/slope_heterogeneity.json
- Table 4, figure "coefficient_zoo" (JSON): https://vishalsingh.org/zipcode/report/gradients-slopes-and-variance/figures/coefficient_zoo.json
- Table 5, figure "nine_measure_pca" (JSON): https://vishalsingh.org/zipcode/report/gradients-slopes-and-variance/figures/nine_measure_pca.json
- Table 6, figure "maup_correlations" (JSON): https://vishalsingh.org/zipcode/report/gradients-slopes-and-variance/figures/maup_correlations.json
- Table 7, figure "variance_split" (JSON): https://vishalsingh.org/zipcode/report/gradients-slopes-and-variance/figures/variance_split.json
- Table 8, figure "county_adi_spread" (JSON): https://vishalsingh.org/zipcode/report/gradients-slopes-and-variance/figures/county_adi_spread.json
- Table 9, figure "wonder_suppression_by_birthweight" (JSON): https://vishalsingh.org/zipcode/report/gradients-slopes-and-variance/figures/wonder_suppression_by_birthweight.json
- Index of every report: https://vishalsingh.org/zipcode/llms.txt; all reports in one file: https://vishalsingh.org/zipcode/llms-full.txt

---

# Families of places: report

Chapter 10.2 of ZIP Code Destiny, by Vishal Singh (NYU Stern). Generated from the chapter's data files; the reading version is at https://vishalsingh.org/zipcode/families-of-places. Table n of this report is figure n of the chapter.

## Identification

- Title: Families of places
- Label: Sorting instead of ranking
- Chapter: 10.2 (Part X, How the estimates are made; Methods and sources)
- Question: Which groups of neighborhoods share a deprivation profile, and how do the groups differ?
- Dek: Among 4 families of census tracts sorted on deprivation measures alone, life expectancy for deaths in 2010 to 2015 runs from 74.7 years in the high-poverty family to 81.7 years in the high-income family.
- Grain: neighborhoods
- Status: published
- Published: 2026-10-09
- Updated: not revised
- Data sources:
  - American Community Survey. ACS 5-year estimates 2019-2023 for 2020 tracts and counties: the 15 deprivation measures, from the staged tract file and the Census API pull of housing, occupation, income-distribution and schooling tables; years: 2019-2023 (5-year); weights: ACS 2019-2023 tract population
  - Area Deprivation Index. Area Deprivation Index 2023 (v4.0.1) national rank, tract values averaged from block groups, used to number the families and to form deciles; years: 2023 release (version 4.0.1; built on ACS 2019-2023); weights: ACS 2019-2023 tract population
  - USALEEP life expectancy. NCHS small-area life expectancy at birth, deaths in 2010 to 2015, 2020 tracts (area-weighted), institutional tracts left out; years: 2010-2015; weights: ACS 2019-2023 tract population
  - Opportunity Atlas. Adult household income rank of children born 1978 to 1983 raised in families at the 25th income percentile, 2020 tracts; years: 1978-1983 birth cohorts (2018 release); weights: ACS 2019-2023 tract population
  - CDC PLACES. CDC PLACES 2024 release (BRFSS 2022): diabetes, current smoking and fair or poor health among adults, modeled prevalence; years: 2024 release (tracts); 2025 release (ZCTAs, 2023 data year); weights: ACS 2019-2023 tract population
- Canonical URL: https://vishalsingh.org/zipcode/families-of-places
- Report: https://vishalsingh.org/zipcode/families-of-places/report
- Markdown: https://vishalsingh.org/zipcode/families-of-places/report.md
- BibTeX key: singh2026familiesofplaces

```bibtex
@misc{singh2026familiesofplaces,
  author = {Singh, Vishal},
  title = {Families of places},
  year = {2026},
  url = {https://vishalsingh.org/zipcode/families-of-places},
  note = {In ZIP Code Destiny. Published 9 October 2026.}
}
```

## Principal estimates

The facts the dek and the opening prose cite, in order of appearance.

| Estimand | Population | Period | Value | Interval | n | Weights | Source | Fact id |
| --- | --- | --- | ---: | --- | ---: | --- | --- | --- |
| definition: the number of families the chapter uses | k-means fits compared | ACS 2019-2023 | 4 |   |   | none (definition) | American Community Survey | k |
| population-weighted mean of the tract life expectancy over the tracts of the family, bootstrap interval over counties | residents of tracts in the high poverty family | deaths in 2010 to 2015 | 74.7 | 95% CI 74.5 to 74.9 | 12,903 tracts | ACS 2019-2023 tract population | USALEEP life expectancy | f4_le |
| population-weighted mean of the tract life expectancy over the tracts of the family, bootstrap interval over counties | residents of tracts in the high income family | deaths in 2010 to 2015 | 81.7 | 95% CI 81.5 to 81.8 | 16,221 tracts | ACS 2019-2023 tract population | USALEEP life expectancy | f1_le |
| count of tracts clustered | census tracts in the frame (2020 tracts with a published ADI rank, residents and all 15 measures) | inputs from ACS 2019-2023 | 75,717 |   |   | none (full count) | American Community Survey | n_tracts |
| count of measures (definition) | deprivation measures given to k-means | ACS 2019-2023 | 15 |   |   | none (definition) | American Community Survey | n_measures |
| difference between the population-weighted mean life expectancy of the high-income and the high-poverty family, bootstrap interval over counties with shared draws | residents of tracts in the high-income and high-poverty families | deaths in 2010 to 2015 | 6.9 | 95% CI 6.7 to 7.2 | 75,717 tracts | ACS 2019-2023 tract population | USALEEP life expectancy | le_gap_14 |
| population-weighted mean of the tract life expectancy over the tracts of the family, bootstrap interval over counties | residents of tracts in the crowded renters family | deaths in 2010 to 2015 | 78.6 | 95% CI 78.1 to 79.0 | 6,587 tracts | ACS 2019-2023 tract population | USALEEP life expectancy | f2_le |
| population-weighted mean of the tract life expectancy over the tracts of the family, bootstrap interval over counties | residents of tracts in the mid-range family | deaths in 2010 to 2015 | 78.6 | 95% CI 78.5 to 78.7 | 36,163 tracts | ACS 2019-2023 tract population | USALEEP life expectancy | f3_le |
| difference between the population-weighted mean ADI national rank of the mid-range and the crowded-renter family, bootstrap interval over counties with shared draws | residents of tracts in the mid-range and crowded-renter families | 2023 release on ACS 2019-2023 | 16.3 | 95% CI 8.4 to 22.6 | 75,717 tracts | ACS 2019-2023 tract population | Area Deprivation Index | adi_gap_32 |
| mean silhouette of the four families, interval from the sampling error of the sample | a seeded random sample of tracts in the frame | ACS 2019-2023 | 0.21 | 95% CI 0.21 to 0.21 | 10,000 tracts | none (each tract counts once) | American Community Survey | sil_k |
| mean silhouette of the 4-group k-means fit to simulated data with no clusters (mean of 3 draws) | a seeded random sample of tracts in the frame | ACS 2019-2023 | 0.12 |   | 10,000 tracts | none (each tract counts once) | American Community Survey | sil_null_k |
| within-group sum of squares of the best different arrangement relative to the fit that defines the families | census tracts in the frame | ACS 2019-2023 | 1.1% |   | 75,717 tracts | none (each tract counts once) | American Community Survey | alt_excess |

## Findings

K-means sorts the 75,717 census tracts of the frame into 4 families on 15 of the measures that the Area Deprivation Index (ADI) combines. The families differ by 6.9 (95% CI 6.7 to 7.2) years of life expectancy at birth between the longest-lived and the shortest-lived, for deaths in 2010 to 2015, and the sort never used life expectancy.

Two of the families show no detectable difference in life expectancy, 78.6 years for the crowded renters and 78.6 years for the mid-range family. The crowded renters also do no better on adult income rank or on modeled estimates of diabetes, smoking and self-rated health, although the ADI scores them as less deprived than the mid-range family by 16.3 (95% CI 8.4 to 22.6) points on its national rank.

The families are regions of a continuum and not separate clusters. The silhouette, a score of how well tracts fit their own group against the nearest other, is 0.21 for the four families, against 0.12 for simulated data with no clusters. A second arrangement of four groups has only 1.1% more within-group variance.

This chapter chooses the number of families, describes them, compares them with outcomes held out of the sort and with deciles of the ADI, and labels counties by the family they resemble.

Related material is in other chapters. Chapter 1.1, What the deprivation index measures (https://vishalsingh.org/zipcode/anatomy-of-deprivation) rebuilds the ADI from these measures and shows how much of its weight falls on home value. Chapter 9.2, Eight types of neighborhood (https://vishalsingh.org/zipcode/eight-types-of-neighborhood) sorts tracts on income, poverty, education, family structure, tenure, age and density, with no outcome among its inputs, so its types separate places by how they are settled as well as by their resources, while the families here use deprivation measures alone. Chapter 10.1, Gradients, slopes and variance (https://vishalsingh.org/zipcode/gradients-slopes-and-variance) covers the slopes of outcomes on the ADI and the split of variance among states, counties and tracts, and Chapter 1.2, What a county average leaves out (https://vishalsingh.org/zipcode/deprivation-at-two-zooms) repeats the index for counties.

### The measures and how they are sorted

The frame is the one that Chapter 1.1, What the deprivation index measures (https://vishalsingh.org/zipcode/anatomy-of-deprivation) uses, the 75,717 tracts of the 2020 census that have a published ADI rank, residents and all 15 measures. They hold 307.1 million residents, 93.0% of the residents of tracts with a rank in the states and the District of Columbia. Puerto Rico's 981 tracts are outside the frame, as they are in that chapter. The measures come from the American Community Survey (ACS) for 2019 to 2023. The measures of advantage are median family income, median home value, median gross rent, the monthly mortgage cost, the white-collar share of workers and the homeownership rate. The measures of hardship are the poverty rate, the near-poverty share, income disparity (the log ratio of low-income to high-income households), unemployment, the single-parent share, the share of households without a vehicle, crowding, homes without complete plumbing and the share of adults with under 9 years of schooling.

Each measure is standardized, which means subtracting its mean across tracts and dividing by its standard deviation, so that a dollar amount and a proportion count equally. K-means then places 4 centers in the space of the standardized measures so that the total squared distance from each tract to its nearest center is as small as it can be made,[1]

$$
\min_{c,\ \mu}\ \sum_{i} \lVert z_i - \mu_{c(i)} \rVert^{2},
$$

where $$z_i$$ is the vector of standardized measures for tract $$i$$, $$\mu_j$$ is the center of family $$j$$ and $$c(i)$$ is the family that tract $$i$$ is assigned to. The fit starts from many random sets of centers and keeps the best. Each tract counts once, so a family is a family of places and not of people. Residents return as weights in every profile and outcome mean below.

The ADI is not an input, but it is built from the same ACS measures, so the sort and the index share their raw material. Life expectancy, adult income rank and the health measures from PLACES are also outside the sort. They come back below as held-out outcomes.

### Four families depart most from a cloud with no clusters

The data do not say how many groups there are, and k-means returns a partition for any number. Table 1 compares fits with 2 to 10 groups on the silhouette, computed on a seeded sample of tracts.[2] The silhouette of the tract data is 0.22 for two groups, 0.21 for four and 0.15 for five, so it is no higher for four groups than for two and drops after four. Taken alone it favors two groups.

A positive silhouette does not show that groups exist, because k-means divides any cloud into parts and some of them look separate. The build therefore fits the same k-means to simulated data as well. The simulated clouds keep the distribution of each measure, including its skew and its top codes, and the correlations among measures, but their dependence is a single multivariate normal cloud with one dense core. At two groups they score 0.21, nearly the same as the tract data, which means the high silhouette at two groups comes from the shape of the cloud and not from groups in it. The sample silhouette has a margin of error of at most 0.003 either way, and the 3 simulated clouds differ from one another by at most 0.007.

At four groups the simulated score is 0.12, and the tract data exceed it by 0.09, against at most 0.03 at any other number of groups. This chapter uses four families because that is where the data depart most from a cloud without clusters. The choice is a judgment about where to draw lines on a continuum, and it does not count the kinds of neighborhood that exist.

Table 1. The tract data beat a cloud with no clusters by 0.09 in silhouette at four families, more than at any other number of families

The departure is modest. The four families hold 41.8% of the variance of the standardized measures between them, against 38.3% for four groups cut from the simulated cloud, and 13.4% of the sampled tracts have a negative silhouette, which means they sit closer to another family than to their own.

### The partition changes with weights, scale and the sample of counties

A fit can be reproducible and still have soft borders. The adjusted Rand index compares two partitions of the same tracts, with the maximum score for identical partitions and a score of zero for the agreement of random ones.[3] Refits from 6 other starting seeds agree with the families at 0.99 or higher. The result depends more on the choices in the sort. A refit that weights each tract by its population agrees at 0.74, so whether tracts or residents count changes the partition. A refit with the four dollar amounts replaced by their logs agrees at only 0.62, so the families depend on how the dollar measures are scaled, as the ADI does (Chapter 1.1, What the deprivation index measures (https://vishalsingh.org/zipcode/anatomy-of-deprivation)).

Resampling whole counties tests stability with the county as the unit of sampling, because tracts in a county resemble one another.[4] The second tab of Table 1 refits k-means on 200 resamples that draw counties with replacement, at each number of groups and with as many starts as the main fit, and assigns every tract to the nearest refit center.

The median agreement with the families at four groups is 0.88, but 17.5% (95% CI 12.9% to 23.4%) of the refits find a different arrangement, which this chapter defines as an agreement below 0.5. That is more often than at three groups, where the share is 0.0% (95% CI 0.0% to 1.9%), or at five, where it is 2.0% (95% CI 0.8% to 5.0%).

The different arrangement is in the full data too. Of 30 fits that start from a single random set of centers, 13 end there, and its within-group variance is only 1.1% higher than that of the families. It puts 76.9% of the crowded-renter tracts in one group with 47.6% of the high-poverty tracts, so a separate crowded-renter family is the least secure part of the partition.

Median agreement is between 0.88 and 0.92 up to five groups, falls to 0.80 at six and 0.54 at seven, and is 0.78 at eight. Reproducibility therefore does not favor four groups, and the benchmark above is the reason for choosing them.

### What defines each family

Table 2 shows the profile of each family on the 15 measures, and Table 3 gives the same families in the units of the measures. The four families are the high-income family, the crowded renters, the mid-range family and the high-poverty family, numbered by the mean ADI rank of their tracts from the least to the most deprived. The names describe the measures that set each family apart and are the book's labels. They say nothing about who lives in the tracts, and no measure in the sort is a race, an ethnicity or a place of birth.

Table 2. Crowded renters average 2.2 standard deviations above the mean tract on crowding, and the high-poverty family 1.4 above on poverty

The high-income family holds 76.6 million residents (24.9% of the frame) in 17,284 tracts. The mean tract median family income is 162 (95% CI 159 to 166) thousand dollars and the mean median home value 714 (95% CI 654 to 774) thousand dollars, with a poverty rate of 5.7% (95% CI 5.4% to 5.9%). Its tracts sit above the mean tract on income, housing costs and white-collar work and below it on every hardship measure, although the interval for households without a vehicle includes zero.

The crowded-renter family is the smallest, with 29.9 million residents in 6,870 tracts. Its tracts average 14.9% (95% CI 13.4% to 16.2%) of homes with more than one person per room, 43.3% (95% CI 41.0% to 46.1%) homeownership and 17.1% (95% CI 15.8% to 18.2%) of adults with under 9 years of schooling. Crowding sits 2.2 (95% CI 1.9 to 2.5) standard deviations above the mean tract.

Housing costs in the crowded-renter family are above the mean tract, with a home value of 472 (95% CI 389 to 540) thousand dollars, while family incomes are below it, at 74 (95% CI 71 to 76) thousand dollars. The family is concentrated by state, with 42.6% of its residents in California, 13.8% in New York and 11.8% in Texas.

The mid-range family is the largest, with 153.1 million residents in 37,746 tracts. Its column in Table 2 stays within 0.34 standard deviations of zero on every measure, with a family income of 95 (95% CI 94 to 95) thousand dollars and a poverty rate of 9.9% (95% CI 9.8% to 10.0%).

The high-poverty family holds 47.4 million residents in 13,817 tracts, with a poverty rate of 26.5% (95% CI 26.2% to 26.8%), a family income of 58 (95% CI 57 to 58) thousand dollars and a home value of 168 (95% CI 162 to 175) thousand dollars, so low income and low housing cost go together.

Table 3. Mean tract family income runs from 58 thousand dollars in the high-poverty family to 162 thousand dollars in the high-income family

Counting residents instead of tracts changes the profile little. Every cell of Table 2 differs from the center that k-means found, which counts each tract once, by at most 0.2 standard deviations.

### Outcomes the sort did not use

Table 4 compares the families on measures that were not inputs. Life expectancy runs from 81.7 (95% CI 81.5 to 81.8) years in the high-income family to 74.7 (95% CI 74.5 to 74.9) years in the high-poverty family, a difference of 6.9 (95% CI 6.7 to 7.2) years. The adult income rank of children raised in low-income families, which the Opportunity Atlas reports as a percentile, runs from 48.1 (95% CI 47.7 to 48.6) to 37.2 (95% CI 36.9 to 37.6). Both outcomes are older than the measures. Life expectancy is for deaths in 2010 to 2015, mobility is for children born in 1978 to 1983, and the families are drawn on conditions in 2019 to 2023. The comparison shows that present conditions line up with past outcomes, and it does not show that the families produce them.

Table 4. Life expectancy runs from 81.7 years in the high-income family to 74.7 years in the high-poverty family, and the crowded-renter and mid-range families do not differ detectably

The two middle families do not follow the ADI. The ADI scores the crowded renters at a mean rank of 38.5 (95% CI 32.0 to 46.2) and the mid-range family at 54.7 (95% CI 53.8 to 55.7), a difference of 16.3 (95% CI 8.4 to 22.6) points. Their life expectancy differs by 0.0 (95% CI −0.5 to 0.4) years, an interval that includes zero.

The adult income rank of the crowded renters is 1.9 (95% CI 1.1 to 2.7) percentiles below that of the mid-range family, and their modeled rate of fair or poor health is 8.2 (95% CI 7.2 to 9.2) points above it. The ADI follows home value closely, as Chapter 1.1, What the deprivation index measures (https://vishalsingh.org/zipcode/anatomy-of-deprivation) shows, and the crowded renters live in tracts with higher home values than the mid-range family does, 472 thousand dollars against 271 thousand dollars.

The health measures in the second tab are modeled. The PLACES model combines each survey respondent's age, sex, race and education with the poverty rate of the respondent's county. It applies the result to the age, race and sex counts of each tract. Differences among the families in that tab therefore reflect county effects and the demographic make-up of tracts, and no tract-level input of the families enters the model. Life expectancy and adult income rank are not modeled in that way. They come from death records and from tax records, interpolated from 2010 tracts to 2020 tracts by area. The intervals in the figure are bootstrap intervals over counties given the families, and they do not carry the uncertainty about the families themselves, which the refits above address.

### Deciles of the ADI mix the families

A ranking puts tracts in order, and a decile of the ranking holds whatever kinds of tract land there. Table 5 splits each decile of the ADI national rank by family. The deciles are close to equal in size, from 7,430 to 7,764 tracts, because ties in the published rank put tracts of equal rank in one decile.

In the most deprived decile, 79.2% of residents live in tracts of the high-poverty family and 16.1% in tracts of the mid-range family. The mid-range residents live 2.4 (95% CI 2.1 to 2.7) years longer than the high-poverty residents of the same decile.

No decile is a single family. The most common family holds between 50.7% and 88.3% of a decile's residents, and a second family holds at least 8.2% in every decile. Across the ten deciles, 72.0% of residents live in a tract of their decile's most common family. The crowded renters appear in every decile, from 4.7% of the most deprived decile's residents to 19.3% of the second.

Table 5. In the most deprived decile of tracts, 79.2% of residents live in high-poverty tracts and 16.1% in mid-range tracts

The mix matters for how much of the variation in outcomes a grouping carries. Ten deciles of the ADI account for 36.6% (95% CI 34.3% to 39.0%) of the variance of tract life expectancy and the four families for 33.3% (95% CI 31.4% to 35.2%). The deciles carry 3.3 (95% CI 2.0 to 4.5) points more, so four groups formed from the measures come close to ten groups formed from the rank. For adult income rank the families carry more, 27.7% (95% CI 25.6% to 30.0%) against 23.3% (95% CI 21.0% to 26.2%) for the deciles, a difference of 4.4 (95% CI 2.5 to 6.2) points.

The cells formed by decile and family account for 41.3% (95% CI 39.2% to 43.3%) of the variance of life expectancy and 30.9% (95% CI 28.7% to 33.4%) of adult income rank. The families add 4.7 (95% CI 4.1 to 5.4) points for life expectancy and 7.6 (95% CI 6.3 to 9.1) points for adult income rank to what the deciles carry, and the deciles add 8.0 (95% CI 7.3 to 8.8) and 3.2 (95% CI 2.6 to 4.0) points to what the families carry.

A grouping with more cells accounts for more variance even when it is unrelated to the outcome. If 40 cells were unrelated to life expectancy and each county counted as a single observation, which gives the most that chance could carry, they would account for 1.3%, below each of the gains above.

### At county scale the families blur

A county has one set of measures and many kinds of tract. Each of the 3,127 counties is labeled with the family whose center lies nearest the county's own measures, standardized as the tracts are. The labels give 96 high-income counties, 29 crowded-renter counties, 2,261 mid-range counties and 741 high-poverty counties (Table 6).

The distance uses 14 of the measures, without the single-parent share, because the county tables define it differently from the tract file. Each tract then takes its county's label, and the agreement between the county labels and the tract families is 0.22 on the adjusted Rand index. Across the country, 62.5% of residents live in a tract of the most common family of their county's label, and the share runs from 50.7% to 70.6% across the four labels.

Table 6. Crowded-renter tracts hold 50.7% of the residents of counties that resemble the crowded-renter family, and high-income tracts hold 34.6%

The crowded-renter and high-poverty labels are the weakest. The 29 counties with the crowded-renter label hold 21.8 million residents, and 50.7% of them live in crowded-renter tracts while 34.6% live in high-income tracts. In the 741 counties labeled high-poverty, 52.9% of residents live in high-poverty tracts.

The crowded-renter family is also not contained in its counties. 37.0% of its residents live in counties with that label, 37.2% in counties labeled mid-range and 22.3% in counties labeled high-income. A county average combines crowding, renting and costly housing with the tracts around them, so a county label does not recover the family. Chapter 1.2, What a county average leaves out (https://vishalsingh.org/zipcode/deprivation-at-two-zooms) makes the same comparison for the index itself.

### What these data cannot show

The families describe tracts, and every comparison in this chapter is ecological. The life expectancy of a family is an average over tracts, and it says little about any one resident of one tract (ecological inference). Nothing here shows that living in a family's tracts changes a person's outcome. The families are arrangements of census measures, and the outcomes that line up with them reflect selection, history, prices and policy that these data do not separate.

The number and the names of the families are choices. The silhouette favors no number of groups on its own, and the benchmark picks four by a stated rule. Refits find another arrangement more often at four groups than at three or five, and a second arrangement of four groups has only 1.1% more within-group variance than the first. Weighting by residents or rescaling the dollar measures changes the partition. The crowded-renter family depends most on those choices. The frame leaves out tracts that lack a median rent, home value or mortgage cost, which the ACS withholds for tracts with almost no renters or almost no owners, and it leaves out Puerto Rico, so none of these results describes the island.

The measures are tract estimates for 2019 to 2023 without margins of error. The tract file defines the share of single-parent households differently from the county tables, while the 819 tracts of Utah and the District of Columbia carry the county definition. Life expectancy is for deaths in 2010 to 2015 and was interpolated from 2010 to 2020 tracts by area, and the USALEEP estimates suit maps and gradients and not single tracts. Adult income rank describes children born in 1978 to 1983 and was interpolated to 2020 tracts by area. PLACES values are modeled, and the ADI averages block-group ranks over tracts without population weights. The county labels use county estimates of the same measures and treat Connecticut's planning regions as counties.

## Tables

### Table 1. The tract data beat a cloud with no clusters by 0.09 in silhouette at four families, more than at any other number of families

Mean silhouette of k-means fits to the 15 standardized deprivation measures, tract data and simulated data with no clusters, and agreement of refits on resamples of counties

#### Table 1.1. Silhouette

The tract data beat simulated data with no clusters by 0.09 at four families and by at most 0.03 at any other number

Mean silhouette by number of families, tract data and simulated data with no clusters, with 95% intervals for the sample

| conf_high | conf_low | k | n | series | silhouette |
| ---: | ---: | ---: | ---: | --- | ---: |
| 0.221 | 0.216 | 2 | 10,000 | Tract data | 0.218 |
| 0.208 | 0.205 | 2 | 10,000 | Simulated, no clusters | 0.207 |
| 0.214 | 0.207 | 3 | 10,000 | Tract data | 0.210 |
| 0.182 | 0.179 | 3 | 10,000 | Simulated, no clusters | 0.180 |
| 0.212 | 0.206 | 4 | 10,000 | Tract data | 0.209 |
| 0.122 | 0.120 | 4 | 10,000 | Simulated, no clusters | 0.121 |
| 0.153 | 0.148 | 5 | 10,000 | Tract data | 0.151 |
| 0.126 | 0.123 | 5 | 10,000 | Simulated, no clusters | 0.124 |
| 0.156 | 0.151 | 6 | 10,000 | Tract data | 0.153 |
| 0.127 | 0.124 | 6 | 10,000 | Simulated, no clusters | 0.125 |
| 0.131 | 0.126 | 7 | 10,000 | Tract data | 0.128 |
| 0.125 | 0.123 | 7 | 10,000 | Simulated, no clusters | 0.124 |
| 0.134 | 0.129 | 8 | 10,000 | Tract data | 0.132 |
| 0.122 | 0.120 | 8 | 10,000 | Simulated, no clusters | 0.121 |
| 0.140 | 0.135 | 9 | 10,000 | Tract data | 0.137 |
| 0.113 | 0.110 | 9 | 10,000 | Simulated, no clusters | 0.111 |
| 0.142 | 0.137 | 10 | 10,000 | Tract data | 0.140 |
| 0.112 | 0.109 | 10 | 10,000 | Simulated, no clusters | 0.110 |

#### Table 1.2. Agreement of refits

At four groups 17.5% of refits on resampled counties find a different arrangement, more than at three or five groups

Adjusted Rand index between the fit on all tracts and refits on resamples of counties, by number of families, median and tenth to ninetieth percentile

| k | median | p10 | p90 | refits |
| ---: | ---: | ---: | ---: | ---: |
| 2 | 0.884 | 0.738 | 0.978 | 200 |
| 3 | 0.919 | 0.855 | 0.961 | 200 |
| 4 | 0.882 | 0.352 | 0.953 | 200 |
| 5 | 0.881 | 0.753 | 0.947 | 200 |
| 6 | 0.798 | 0.645 | 0.922 | 200 |
| 7 | 0.543 | 0.467 | 0.777 | 200 |
| 8 | 0.780 | 0.677 | 0.917 | 200 |

Note: The silhouette uses a seeded random sample of tracts; the intervals show its sampling error, at most 0.003 in either direction. The simulated line is the mean of 3 draws that differ by at most 0.007 at any number of families.

Note: Simulated data are drawn from a Gaussian copula: each measure keeps its distribution in the tract data, and the dependence among measures is a single multivariate normal cloud, so k-means has no clusters to find.

Note: The band in the second tab runs from the tenth to the ninetieth percentile of the refits, each a k-means fit on a resample that draws whole counties with replacement and assigns every tract to its nearest center (replicates and seed in the methods note).

Source: ACS 2019-2023 deprivation measures, k-means fits to all tracts, silhouette on a sample of tracts; unweighted (each tract counts once)

Data: https://vishalsingh.org/zipcode/report/families-of-places/figures/number_of_families.json

### Table 2. Crowded renters average 2.2 standard deviations above the mean tract on crowding, and the high-poverty family 1.4 above on poverty

Population-weighted mean z-score of each of the 15 measures by family, tracts, ACS 2019 to 2023; zero is the mean tract

| conf_high | conf_low | family | measure | n | z |
| ---: | ---: | --- | --- | ---: | ---: |
| 1.44 | 1.29 | High income | Family income | 17,284 | 1.36 |
| −0.551 | −0.654 | Crowded renters | Family income | 6,870 | −0.599 |
| −0.121 | −0.148 | Mid-range | Family income | 37,746 | −0.134 |
| −0.942 | −0.966 | High poverty | Family income | 13,817 | −0.954 |
| 1.32 | 0.927 | High income | Home value | 17,284 | 1.12 |
| 0.553 | 0.058 | Crowded renters | Home value | 6,870 | 0.329 |
| −0.307 | −0.349 | Mid-range | Home value | 37,746 | −0.328 |
| −0.644 | −0.686 | High poverty | Home value | 13,817 | −0.666 |
| 1.36 | 1.14 | High income | Gross rent | 17,284 | 1.25 |
| 0.333 | 0.096 | Crowded renters | Gross rent | 6,870 | 0.225 |
| −0.231 | −0.315 | Mid-range | Gross rent | 37,746 | −0.273 |
| −0.644 | −0.707 | High poverty | Gross rent | 13,817 | −0.674 |
| 1.38 | 1.17 | High income | Mortgage cost | 17,284 | 1.28 |
| 0.587 | 0.166 | Crowded renters | Mortgage cost | 6,870 | 0.399 |
| −0.316 | −0.367 | Mid-range | Mortgage cost | 37,746 | −0.342 |
| −0.748 | −0.814 | High poverty | Mortgage cost | 13,817 | −0.782 |
| 1.23 | 1.10 | High income | White-collar jobs | 17,284 | 1.17 |
| −0.779 | −0.970 | Crowded renters | White-collar jobs | 6,870 | −0.877 |
| −0.061 | −0.102 | Mid-range | White-collar jobs | 37,746 | −0.082 |
| −0.776 | −0.827 | High poverty | White-collar jobs | 13,817 | −0.802 |
| 0.400 | 0.222 | High income | Homeownership | 17,284 | 0.309 |
| −0.917 | −1.15 | Crowded renters | Homeownership | 6,870 | −1.05 |
| 0.353 | 0.304 | Mid-range | Homeownership | 37,746 | 0.328 |
| −0.651 | −0.733 | High poverty | Homeownership | 13,817 | −0.692 |
| −0.722 | −0.779 | High income | Poverty | 17,284 | −0.750 |
| 0.669 | 0.492 | Crowded renters | Poverty | 6,870 | 0.572 |
| −0.307 | −0.327 | Mid-range | Poverty | 37,746 | −0.316 |
| 1.42 | 1.35 | High poverty | Poverty | 13,817 | 1.38 |
| −0.864 | −0.925 | High income | Below 150% of poverty | 17,284 | −0.895 |
| 0.821 | 0.645 | Crowded renters | Below 150% of poverty | 6,870 | 0.726 |
| −0.271 | −0.296 | Mid-range | Below 150% of poverty | 37,746 | −0.283 |
| 1.41 | 1.36 | High poverty | Below 150% of poverty | 13,817 | 1.39 |
| −0.683 | −0.806 | High income | Income disparity | 17,284 | −0.743 |
| 0.271 | 0.162 | Crowded renters | Income disparity | 6,870 | 0.217 |
| −0.086 | −0.119 | Mid-range | Income disparity | 37,746 | −0.102 |
| 0.925 | 0.890 | High poverty | Income disparity | 13,817 | 0.907 |
| −0.226 | −0.331 | High income | Unemployment | 17,284 | −0.282 |
| 0.522 | 0.366 | Crowded renters | Unemployment | 6,870 | 0.449 |
| −0.231 | −0.273 | Mid-range | Unemployment | 37,746 | −0.252 |
| 0.844 | 0.692 | High poverty | Unemployment | 13,817 | 0.764 |
| −0.538 | −0.581 | High income | Single parents | 17,284 | −0.559 |
| 0.853 | 0.631 | Crowded renters | Single parents | 6,870 | 0.735 |
| −0.120 | −0.156 | Mid-range | Single parents | 37,746 | −0.137 |
| 0.987 | 0.894 | High poverty | Single parents | 13,817 | 0.940 |
| 0.020 | −0.322 | High income | No vehicle | 17,284 | −0.174 |
| 1.29 | 0.331 | Crowded renters | No vehicle | 6,870 | 0.757 |
| −0.301 | −0.331 | Mid-range | No vehicle | 37,746 | −0.317 |
| 0.575 | 0.400 | High poverty | No vehicle | 13,817 | 0.480 |
| −0.180 | −0.304 | High income | Crowding | 17,284 | −0.241 |
| 2.49 | 1.94 | Crowded renters | Crowding | 6,870 | 2.23 |
| −0.244 | −0.284 | Mid-range | Crowding | 37,746 | −0.264 |
| 0.094 | 0.018 | High poverty | Crowding | 13,817 | 0.054 |
| −0.409 | −0.425 | High income | No plumbing | 17,284 | −0.417 |
| −0.227 | −0.290 | Crowded renters | No plumbing | 6,870 | −0.261 |
| −0.103 | −0.145 | Mid-range | No plumbing | 37,746 | −0.125 |
| 0.655 | 0.512 | High poverty | No plumbing | 13,817 | 0.580 |
| −0.382 | −0.462 | High income | Under 9 years of school | 17,284 | −0.421 |
| 2.24 | 1.83 | Crowded renters | Under 9 years of school | 6,870 | 2.05 |
| −0.235 | −0.279 | Mid-range | Under 9 years of school | 37,746 | −0.256 |
| 0.310 | 0.215 | High poverty | Under 9 years of school | 13,817 | 0.262 |

Note: Values are in standard deviations of the tract distribution of each measure, with each tract counted once, so zero is the mean tract and not the mean resident. Weighting by residents moves a cell by at most 0.2 from the k-means center.

Note: Income, home value, rent, mortgage cost, white-collar work and homeownership are measures of advantage, and the rest are measures of hardship. Intervals are 95% intervals from a bootstrap over counties, given the families.

Note: The share of single-parent households is defined differently in the tract file and in the county tables, and the tracts of Utah and the District of Columbia, which came from the Census API, follow the county definition. They are 819 of the tracts in the frame.

Source: ACS 2019-2023 deprivation measures (15) and Area Deprivation Index 2023; k-means families; population-weighted (z-scores use the unweighted mean and standard deviation across tracts)

Data: https://vishalsingh.org/zipcode/report/families-of-places/figures/family_profiles.json

### Table 3. Mean tract family income runs from 58 thousand dollars in the high-poverty family to 162 thousand dollars in the high-income family

Population-weighted means of tract values by family, in the units of each column, tracts and residents of the frame, ACS 2019 to 2023

| Family | Tracts | Residents (millions) | ADI national rank (mean) | Family income (thousands of dollars) | Home value (thousands of dollars) | Poverty (%) | Crowded homes (%) | Homeownership (%) | Adults with under 9 years of school (%) |
| --- | ---: | ---: | ---: | ---: | ---: | ---: | ---: | ---: | ---: |
| High income | 17,284 | 76.6 | 16.9 | 162 | 714 | 5.7 | 2.3 | 72.4 | 2.3 |
| Crowded renters | 6,870 | 29.9 | 38.5 | 74 | 472 | 18.6 | 14.9 | 43.3 | 17.1 |
| Mid-range | 37,746 | 153.1 | 54.7 | 95 | 271 | 9.9 | 2.2 | 72.8 | 3.3 |
| High poverty | 13,817 | 47.4 | 79.3 | 58 | 168 | 26.5 | 3.8 | 50.9 | 6.4 |

Note: The ADI is not an input to the families. Families are numbered from least to most deprived by the population-weighted mean ADI of their tracts. The names describe the measures that define each family and are the book's labels, not categories in the data.

Note: Dollar amounts are 2019-2023 ACS medians of tracts, averaged over tracts with residents as weights; they are not medians of the family.

Source: ACS 2019-2023 deprivation measures (15) and Area Deprivation Index 2023; k-means families; population-weighted

Data: https://vishalsingh.org/zipcode/report/families-of-places/figures/family_table.json

### Table 4. Life expectancy runs from 81.7 years in the high-income family to 74.7 years in the high-poverty family, and the crowded-renter and mid-range families do not differ detectably

Population-weighted mean of tract values by family with 95% intervals from a bootstrap over counties; none of these measures was used to form the families

#### Table 4.1. Measured outcomes

Life expectancy and adult income rank are highest in the high-income family and lowest in the high-poverty family

Life expectancy at birth for deaths in 2010 to 2015 and adult income rank of children born 1978 to 1983, population-weighted family means with 95% intervals

| conf_high | conf_low | estimate | family | label | n | panel |
| ---: | ---: | ---: | --- | --- | ---: | --- |
| 81.81 | 81.50 | 81.66 | High income | High income | 16,221 | Life expectancy at birth (years) |
| 78.99 | 78.09 | 78.58 | Crowded renters | Crowded renters | 6,587 | Life expectancy at birth (years) |
| 78.67 | 78.50 | 78.59 | Mid-range | Mid-range | 36,163 | Life expectancy at birth (years) |
| 74.94 | 74.55 | 74.74 | High poverty | High poverty | 12,903 | Life expectancy at birth (years) |
| 48.57 | 47.71 | 48.13 | High income | High income | 17,256 | Adult income rank, children from low-income families (percentile) |
| 42.07 | 40.51 | 41.30 | Crowded renters | Crowded renters | 6,869 | Adult income rank, children from low-income families (percentile) |
| 43.41 | 42.98 | 43.20 | Mid-range | Mid-range | 37,728 | Adult income rank, children from low-income families (percentile) |
| 37.58 | 36.90 | 37.23 | High poverty | High poverty | 13,814 | Adult income rank, children from low-income families (percentile) |

#### Table 4.2. Modeled health measures

Diabetes, smoking and fair or poor health are lowest in the high-income family and highest in the high-poverty family

PLACES modeled prevalence among adults, population-weighted family means with 95% intervals

| conf_high | conf_low | estimate | family | label | n | panel |
| ---: | ---: | ---: | --- | --- | ---: | --- |
| 9.66 | 9.19 | 9.43 | High income | High income | 17,284 | Diabetes among adults (%) |
| 14.23 | 13.35 | 13.78 | Crowded renters | Crowded renters | 6,870 | Diabetes among adults (%) |
| 12.08 | 11.85 | 11.97 | Mid-range | Mid-range | 37,746 | Diabetes among adults (%) |
| 16.09 | 15.60 | 15.86 | High poverty | High poverty | 13,817 | Diabetes among adults (%) |
| 9.70 | 9.33 | 9.51 | High income | High income | 17,284 | Smoking among adults (%) |
| 16.53 | 15.33 | 15.83 | Crowded renters | Crowded renters | 6,870 | Smoking among adults (%) |
| 15.14 | 14.83 | 14.99 | Mid-range | Mid-range | 37,746 | Smoking among adults (%) |
| 21.09 | 20.55 | 20.82 | High poverty | High poverty | 13,817 | Smoking among adults (%) |
| 13.09 | 12.46 | 12.78 | High income | High income | 17,284 | Fair or poor health among adults (%) |
| 27.35 | 25.27 | 26.39 | Crowded renters | Crowded renters | 6,870 | Fair or poor health among adults (%) |
| 18.31 | 18.00 | 18.16 | Mid-range | Mid-range | 37,746 | Fair or poor health among adults (%) |
| 27.92 | 27.24 | 27.59 | High poverty | High poverty | 13,817 | Fair or poor health among adults (%) |

Note: Life expectancy is for deaths in 2010 to 2015 on tracts of 2020 (area-weighted from 2010 tracts), with institutional tracts left out. The adult income rank is for children born in 1978 to 1983, interpolated by land area from 2010 tracts to 2020 tracts, and tract population stands in for the number of children. Both describe the past, while the families are built on measures for 2019 to 2023.

Note: PLACES values are modeled estimates. The model combines each survey respondent's age, sex, race and education with the poverty rate of the respondent's county and applies the result to the age, race and sex counts of each tract, so differences among the families in the second tab reflect county effects and the demographic make-up of tracts, not tract-level inputs of the families. Intervals describe how the means would vary with the sample of counties, given the families, and not the choice of families.

Source: USALEEP 2010-2015, Opportunity Atlas (children born 1978 to 1983) and PLACES 2024 release; families from ACS 2019-2023; population-weighted, modeled estimates for PLACES

Data: https://vishalsingh.org/zipcode/report/families-of-places/figures/held_out_outcomes.json

### Table 5. In the most deprived decile of tracts, 79.2% of residents live in high-poverty tracts and 16.1% in mid-range tracts

Share of residents in each decile of the ADI national rank by family of their tract, percent of residents, tracts of the frame, ACS 2019 to 2023

| decile | family | share | tracts |
| --- | --- | ---: | ---: |
| 1 | High income | 88.30 | 6606 |
| 1 | Crowded renters | 11.09 | 901 |
| 1 | Mid-range | 0.378 | 39 |
| 1 | High poverty | 0.234 | 15 |
| 2 | High income | 70.94 | 5256 |
| 2 | Crowded renters | 19.28 | 1508 |
| 2 | Mid-range | 9.17 | 770 |
| 2 | High poverty | 0.617 | 54 |
| 3 | High income | 50.70 | 3698 |
| 3 | Crowded renters | 12.78 | 960 |
| 3 | Mid-range | 35.16 | 2708 |
| 3 | High poverty | 1.36 | 123 |
| 4 | High income | 18.56 | 1418 |
| 4 | Crowded renters | 10.33 | 795 |
| 4 | Mid-range | 68.03 | 5269 |
| 4 | High poverty | 3.07 | 282 |
| 5 | High income | 3.32 | 256 |
| 5 | Crowded renters | 8.23 | 589 |
| 5 | Mid-range | 82.74 | 6171 |
| 5 | High poverty | 5.72 | 477 |
| 6 | High income | 0.440 | 42 |
| 6 | Crowded renters | 8.14 | 559 |
| 6 | Mid-range | 82.69 | 6223 |
| 6 | High poverty | 8.73 | 699 |
| 7 | High income | 0.048 | 6 |
| 7 | Crowded renters | 7.61 | 516 |
| 7 | Mid-range | 77.17 | 5966 |
| 7 | High poverty | 15.18 | 1177 |
| 8 | High income | 0.015 | 1 |
| 8 | Crowded renters | 6.78 | 430 |
| 8 | Mid-range | 68.51 | 5193 |
| 8 | High poverty | 24.70 | 1806 |
| 9 | High income | 0.006 | 1 |
| 9 | Crowded renters | 5.13 | 331 |
| 9 | Mid-range | 52.11 | 4127 |
| 9 | High poverty | 42.76 | 3176 |
| 10 | High income | 0 | 0 |
| 10 | Crowded renters | 4.71 | 281 |
| 10 | Mid-range | 16.14 | 1280 |
| 10 | High poverty | 79.16 | 6008 |

Note: Shares are of residents (ACS 2019-2023 tract population) of the tracts in the frame, not of tracts. Deciles of the frame's tracts are close to equal in size, from 7,430 to 7,764 tracts, because ties in the published tract ADI national rank put tracts of equal rank in one decile. The shares are counts of ACS estimates and carry no sampling interval here.

Source: ACS 2019-2023 tract population by family; Area Deprivation Index 2023; full counts of ACS estimates

Data: https://vishalsingh.org/zipcode/report/families-of-places/figures/adi_decile_composition.json

### Table 6. Crowded-renter tracts hold 50.7% of the residents of counties that resemble the crowded-renter family, and high-income tracts hold 34.6%

Share of residents by family of their tract, in counties labeled by the nearest family center, percent of residents, ACS 2019 to 2023

| county_label | family | share |
| --- | --- | ---: |
| High-income counties | High income | 70.56 |
| High-income counties | Crowded renters | 14.05 |
| High-income counties | Mid-range | 14.06 |
| High-income counties | High poverty | 1.33 |
| Crowded-renter counties | High income | 34.63 |
| Crowded-renter counties | Crowded renters | 50.73 |
| Crowded-renter counties | Mid-range | 9.97 |
| Crowded-renter counties | High poverty | 4.67 |
| Mid-range counties | High income | 16.29 |
| Mid-range counties | Crowded renters | 5.20 |
| Mid-range counties | Mid-range | 63.01 |
| Mid-range counties | High poverty | 15.49 |
| High-poverty counties | High income | 2.91 |
| High-poverty counties | Crowded renters | 4.43 |
| High-poverty counties | Mid-range | 39.72 |
| High-poverty counties | High poverty | 52.93 |

Note: A county takes the label of the family whose center lies nearest the county's own 14 measures, standardized with the mean and standard deviation of the tract distribution, without the single-parent share, which is defined differently for counties. No clustering is run on counties. Counties include county equivalents and Connecticut's planning regions, as the ACS tables report them.

Note: Shares are of residents of the tracts in the frame and are counts of ACS estimates without a sampling interval. The county measures are medians and rates for the whole county, not averages of its tracts.

Source: ACS 2019-2023 county measures and tract population by family; labels from the nearest tract-family center; full counts of ACS estimates

Data: https://vishalsingh.org/zipcode/report/families-of-places/figures/county_group_composition.json

## All estimates

Every fact in facts.json (130), sorted by id.

| Fact id | Value | Unit | Estimand | Population | Period | Interval | n | Weights | Source | Note |
| --- | ---: | --- | --- | --- | --- | --- | ---: | --- | --- | --- |
| adi_gap_32 | 16.3 | points of the ADI national rank | difference between the population-weighted mean ADI national rank of the mid-range and the crowded-renter family, bootstrap interval over counties with shared draws | residents of tracts in the mid-range and crowded-renter families | 2023 release on ACS 2019-2023 | 95% CI 8.4 to 22.6 | 75,717 tracts | ACS 2019-2023 tract population | Area Deprivation Index |   |
| alt_cr_share | 76.9% | percent of the family's tracts | share of the family's tracts that fall in one group of the best different arrangement | tracts in the crowded-renter family | ACS 2019-2023 |   | 6,870 tracts | none (each tract counts once) | American Community Survey |   |
| alt_excess | 1.1% | percent more within-group variance | within-group sum of squares of the best different arrangement relative to the fit that defines the families | census tracts in the frame | ACS 2019-2023 |   | 75,717 tracts | none (each tract counts once) | American Community Survey |   |
| alt_f4_share | 47.6% | percent of the family's tracts | share of the family's tracts that fall in the same group of the best different arrangement as most crowded-renter tracts | tracts in the high-poverty family | ACS 2019-2023 |   | 13,817 tracts | none (each tract counts once) | American Community Survey |   |
| ari_boot_median | 0.88 | adjusted Rand index (1 = identical partitions, 0 = chance agreement) | median agreement with the families of refits on resamples that draw whole counties | census tracts in the frame | ACS 2019-2023 |   | 75,717 tracts | none (each tract counts once) | American Community Survey |   |
| ari_cut | 0.5 | adjusted Rand index | definition: a refit whose agreement with the families is below this cutoff counts as a different arrangement of the groups | refits on resamples of counties | ACS 2019-2023 |   |   | none (definition) | American Community Survey |   |
| ari_k6 | 0.80 | adjusted Rand index (1 = identical partitions, 0 = chance agreement) | median agreement of refits on resamples of counties at 6 groups | census tracts in the frame | ACS 2019-2023 |   | 75,717 tracts | none (each tract counts once) | American Community Survey |   |
| ari_k7 | 0.54 | adjusted Rand index (1 = identical partitions, 0 = chance agreement) | median agreement of refits on resamples of counties at 7 groups | census tracts in the frame | ACS 2019-2023 |   | 75,717 tracts | none (each tract counts once) | American Community Survey |   |
| ari_k8 | 0.78 | adjusted Rand index (1 = identical partitions, 0 = chance agreement) | median agreement of refits on resamples of counties at 8 groups | census tracts in the frame | ACS 2019-2023 |   | 75,717 tracts | none (each tract counts once) | American Community Survey |   |
| ari_log | 0.62 | adjusted Rand index (1 = identical partitions, 0 = chance agreement) | agreement with the families of a refit that replaces the four dollar measures by their logs | census tracts in the frame | ACS 2019-2023 |   | 75,717 tracts | none (each tract counts once) | American Community Survey |   |
| ari_seed_min | 0.99 | adjusted Rand index (1 = identical partitions, 0 = chance agreement) | the lowest agreement with the families among refits with other starting seeds (10 starts each) | census tracts in the frame | ACS 2019-2023 |   | 75,717 tracts | none (each tract counts once) | American Community Survey |   |
| ari_stab_hi | 0.92 | adjusted Rand index (1 = identical partitions, 0 = chance agreement) | the highest median agreement of refits on resamples of counties among two to five groups | census tracts in the frame | ACS 2019-2023 |   | 75,717 tracts | none (each tract counts once) | American Community Survey |   |
| ari_stab_lo | 0.88 | adjusted Rand index (1 = identical partitions, 0 = chance agreement) | the lowest median agreement of refits on resamples of counties among two to five groups | census tracts in the frame | ACS 2019-2023 |   | 75,717 tracts | none (each tract counts once) | American Community Survey |   |
| ari_weighted | 0.74 | adjusted Rand index (1 = identical partitions, 0 = chance agreement) | agreement with the families of a refit that weights each tract by its population | census tracts in the frame | ACS 2019-2023 |   | 75,717 tracts | none (each tract counts once) | American Community Survey |   |
| boot_other_share | 17.5% | percent of refits | share of refits whose agreement with the families is below the cutoff, Wilson interval that treats the refits as independent draws | refits on resamples of counties at four groups | ACS 2019-2023 | 95% CI 12.9% to 23.4% | 200 refits | none (each refit counts once) | American Community Survey |   |
| boot_other_share_k3 | 0.0% | percent of refits | share of refits whose agreement with the families is below the cutoff, Wilson interval that treats the refits as independent | refits on resamples of counties at 3 groups | ACS 2019-2023 | 95% CI 0.0% to 1.9% | 200 refits | none (each refit counts once) | American Community Survey |   |
| boot_other_share_k5 | 2.0% | percent of refits | share of refits whose agreement with the families is below the cutoff, Wilson interval that treats the refits as independent | refits on resamples of counties at 5 groups | ACS 2019-2023 | 95% CI 0.8% to 5.0% | 200 refits | none (each refit counts once) | American Community Survey |   |
| cg1_n | 96 | counties | count of counties whose measures lie nearest the center of the high income family | counties and county equivalents of the 50 states and DC with all 15 measures | ACS 2019-2023 |   |   | none (full count) | American Community Survey |   |
| cg2_f1 | 34.6% | percent of the label's residents | share of the label's residents who live in tracts of the high income family | residents of tracts in the crowded-renter counties | ACS 2019-2023 |   | 29 counties | none (full count of estimates) | American Community Survey |   |
| cg2_f2 | 50.7% | percent of the label's residents | share of the label's residents who live in tracts of the crowded renters family | residents of tracts in the crowded-renter counties | ACS 2019-2023 |   | 29 counties | none (full count of estimates) | American Community Survey |   |
| cg2_n | 29 | counties | count of counties whose measures lie nearest the center of the crowded renters family | counties and county equivalents of the 50 states and DC with all 15 measures | ACS 2019-2023 |   |   | none (full count) | American Community Survey |   |
| cg2_pop | 21.8 million | residents | residents of the frame who live in counties nearest the crowded-renter center | residents of tracts in counties labeled crowded-renter | ACS 2019-2023 |   | 29 counties | none (full count of estimates) | American Community Survey |   |
| cg3_n | 2,261 | counties | count of counties whose measures lie nearest the center of the mid-range family | counties and county equivalents of the 50 states and DC with all 15 measures | ACS 2019-2023 |   |   | none (full count) | American Community Survey |   |
| cg4_f4 | 52.9% | percent of the label's residents | share of the label's residents who live in tracts of the high poverty family | residents of tracts in the high-poverty counties | ACS 2019-2023 |   | 741 counties | none (full count of estimates) | American Community Survey |   |
| cg4_n | 741 | counties | count of counties whose measures lie nearest the center of the high poverty family | counties and county equivalents of the 50 states and DC with all 15 measures | ACS 2019-2023 |   |   | none (full count) | American Community Survey |   |
| cg_ari | 0.22 | adjusted Rand index (1 = identical partitions, 0 = chance agreement) | agreement between the tract families and the county labels, each tract taking its county's label | census tracts in the frame | ACS 2019-2023 |   | 75,717 tracts | none (each tract counts once) | American Community Survey |   |
| cg_diag_max | 70.6% | percent of a label's residents | the largest share, over the four county labels, of a label's residents who live in tracts of the family that the label names | residents of tracts in the frame | ACS 2019-2023 |   | 75,717 tracts | none (full count of estimates) | American Community Survey |   |
| cg_diag_min | 50.7% | percent of a label's residents | the smallest share, over the four county labels, of a label's residents who live in tracts of the family that the label names | residents of tracts in the frame | ACS 2019-2023 |   | 75,717 tracts | none (full count of estimates) | American Community Survey |   |
| cg_purity | 62.5% | percent of residents | share of residents who live in a tract of the most common tract family of their county's label | residents of tracts in the frame | ACS 2019-2023 |   | 75,717 tracts | none (full count of estimates) | American Community Survey |   |
| cr_dec_max | 19.3% | percent of residents of the decile | the largest share, over the ten deciles, of a decile's residents who live in tracts of the crowded-renter family | residents of tracts in the frame | ACS 2019-2023 |   | 75,717 tracts | none (full count of estimates) | American Community Survey |   |
| cr_dec_min | 4.7% | percent of residents of the decile | the smallest share, over the ten deciles, of a decile's residents who live in tracts of the crowded-renter family | residents of tracts in the frame | ACS 2019-2023 |   | 75,717 tracts | none (full count of estimates) | American Community Survey |   |
| cr_in_lab1 | 22.3% | percent of the family's residents | share of the crowded-renter family's residents who live in counties labeled high-income counties | residents of tracts in the crowded renters family | ACS 2019-2023 |   | 6,870 tracts | none (full count of estimates) | American Community Survey |   |
| cr_in_lab2 | 37.0% | percent of the family's residents | share of the crowded-renter family's residents who live in counties labeled crowded-renter counties | residents of tracts in the crowded renters family | ACS 2019-2023 |   | 6,870 tracts | none (full count of estimates) | American Community Survey |   |
| cr_in_lab3 | 37.2% | percent of the family's residents | share of the crowded-renter family's residents who live in counties labeled mid-range counties | residents of tracts in the crowded renters family | ACS 2019-2023 |   | 6,870 tracts | none (full count of estimates) | American Community Survey |   |
| dec10_sh_3 | 16.1% | percent of residents of the decile | share of the decile's residents who live in tracts of the mid-range family | residents of tracts in the most deprived decile of ADI ranks | ACS 2019-2023 |   | 7,569 tracts | none (full count of estimates) | American Community Survey |   |
| dec10_sh_4 | 79.2% | percent of residents of the decile | share of the decile's residents who live in tracts of the high-poverty family | residents of tracts in the most deprived decile of ADI ranks | ACS 2019-2023 |   | 7,569 tracts | none (full count of estimates) | American Community Survey |   |
| dec_modal_max | 88.3% | percent of residents of the decile | the largest share, over the ten deciles, of a decile's residents who live in tracts of its most common family | residents of tracts in the frame | ACS 2019-2023 |   | 75,717 tracts | none (full count of estimates) | American Community Survey |   |
| dec_modal_min | 50.7% | percent of residents of the decile | the smallest share, over the ten deciles, of a decile's residents who live in tracts of its most common family | residents of tracts in the frame | ACS 2019-2023 |   | 75,717 tracts | none (full count of estimates) | American Community Survey |   |
| dec_purity | 72.0% | percent of residents | share of residents who live in a tract of the most common family of their decile | residents of tracts in the frame | ACS 2019-2023 |   | 75,717 tracts | none (full count of estimates) | American Community Survey |   |
| dec_second_min | 8.2% | percent of residents of the decile | the smallest share, over the ten deciles, of a decile's residents who live in tracts of its second most common family | residents of tracts in the frame | ACS 2019-2023 |   | 75,717 tracts | none (full count of estimates) | American Community Survey |   |
| dec_tracts_max | 7,764 | tracts | the most tracts in a decile of the frame | census tracts in the frame (2020 tracts with a published ADI rank, residents and all 15 measures) | inputs from ACS 2019-2023 |   |   | none (full count) | American Community Survey |   |
| dec_tracts_min | 7,430 | tracts | the fewest tracts in a decile of the frame | census tracts in the frame (2020 tracts with a published ADI rank, residents and all 15 measures) | inputs from ACS 2019-2023 |   |   | none (full count) | American Community Survey |   |
| edu_years | 9 | years of schooling | definition: the schooling threshold of the measure, the share of adults who stopped before this many years | adults 25 and over | ACS 2019-2023 |   |   | none (definition) | American Community Survey |   |
| f1_hv | 714 | thousands of dollars of tract median home value | population-weighted mean of the tract median home value over the tracts of the family, bootstrap interval over counties | residents of tracts in the high income family | ACS 2019-2023 | 95% CI 654 to 774 | 17,284 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| f1_inc | 162 | thousands of dollars of tract median family income | population-weighted mean of the tract median family income over the tracts of the family, bootstrap interval over counties | residents of tracts in the high income family | ACS 2019-2023 | 95% CI 159 to 166 | 17,284 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| f1_kfr | 48.1 | percentile of the national household income rank in adulthood, children raised in families at the 25th income percentile | population-weighted mean of the tract adult income rank over the tracts of the family, bootstrap interval over counties | residents of tracts in the high income family | children born 1978 to 1983, income measured around 2014 to 2015 | 95% CI 47.7 to 48.6 | 17,256 tracts | ACS 2019-2023 tract population | Opportunity Atlas |   |
| f1_le | 81.7 | years of life expectancy at birth | population-weighted mean of the tract life expectancy over the tracts of the family, bootstrap interval over counties | residents of tracts in the high income family | deaths in 2010 to 2015 | 95% CI 81.5 to 81.8 | 16,221 tracts | ACS 2019-2023 tract population | USALEEP life expectancy |   |
| f1_pop | 76.6 million | residents | sum of ACS 2019-2023 tract population over the tracts of the family | residents of tracts in the high income family | ACS 2019-2023 |   | 17,284 tracts | none (full count of estimates) | American Community Survey |   |
| f1_pov | 5.7% | percent of persons below the poverty line | population-weighted mean of the tract poverty rate over the tracts of the family, bootstrap interval over counties | residents of tracts in the high income family | ACS 2019-2023 | 95% CI 5.4% to 5.9% | 17,284 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| f1_share | 24.9% | percent of residents of the frame | residents of the high income family as a share of residents of the frame | residents of the tracts in the frame | ACS 2019-2023 |   | 17,284 tracts | none (full count of estimates) | American Community Survey |   |
| f1_tracts | 17,284 | tracts | count of tracts in the high income family | census tracts in the frame (2020 tracts with a published ADI rank, residents and all 15 measures) | inputs from ACS 2019-2023 |   |   | none (full count) | American Community Survey |   |
| f2_adi | 38.5 | points of the ADI national rank (1 least, 100 most deprived) | population-weighted mean of the tract ADI national rank, bootstrap interval over counties | residents of tracts in the crowded renters family | 2023 release on ACS 2019-2023 | 95% CI 32.0 to 46.2 | 6,870 tracts | ACS 2019-2023 tract population | Area Deprivation Index |   |
| f2_ca | 42.6% | percent of the family's residents | share of the family's residents who live in CA | residents of tracts in the crowded renters family | ACS 2019-2023 |   | 6,870 tracts | none (full count of estimates) | American Community Survey |   |
| f2_crowd | 14.9% | percent of households with more than one person per room | population-weighted mean of the tract crowding over the tracts of the family, bootstrap interval over counties | residents of tracts in the crowded renters family | ACS 2019-2023 | 95% CI 13.4% to 16.2% | 6,870 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| f2_edu | 17.1% | percent of adults 25 and over with under 9 years of schooling | population-weighted mean of the tract share of adults with under 9 years of schooling over the tracts of the family, bootstrap interval over counties | residents of tracts in the crowded renters family | ACS 2019-2023 | 95% CI 15.8% to 18.2% | 6,870 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| f2_hv | 472 | thousands of dollars of tract median home value | population-weighted mean of the tract median home value over the tracts of the family, bootstrap interval over counties | residents of tracts in the crowded renters family | ACS 2019-2023 | 95% CI 389 to 540 | 6,870 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| f2_inc | 74 | thousands of dollars of tract median family income | population-weighted mean of the tract median family income over the tracts of the family, bootstrap interval over counties | residents of tracts in the crowded renters family | ACS 2019-2023 | 95% CI 71 to 76 | 6,870 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| f2_le | 78.6 | years of life expectancy at birth | population-weighted mean of the tract life expectancy over the tracts of the family, bootstrap interval over counties | residents of tracts in the crowded renters family | deaths in 2010 to 2015 | 95% CI 78.1 to 79.0 | 6,587 tracts | ACS 2019-2023 tract population | USALEEP life expectancy |   |
| f2_ny | 13.8% | percent of the family's residents | share of the family's residents who live in NY | residents of tracts in the crowded renters family | ACS 2019-2023 |   | 6,870 tracts | none (full count of estimates) | American Community Survey |   |
| f2_own | 43.3% | percent of occupied homes that are owner-occupied | population-weighted mean of the tract homeownership rate over the tracts of the family, bootstrap interval over counties | residents of tracts in the crowded renters family | ACS 2019-2023 | 95% CI 41.0% to 46.1% | 6,870 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| f2_pop | 29.9 million | residents | sum of ACS 2019-2023 tract population over the tracts of the family | residents of tracts in the crowded renters family | ACS 2019-2023 |   | 6,870 tracts | none (full count of estimates) | American Community Survey |   |
| f2_tracts | 6,870 | tracts | count of tracts in the crowded renters family | census tracts in the frame (2020 tracts with a published ADI rank, residents and all 15 measures) | inputs from ACS 2019-2023 |   |   | none (full count) | American Community Survey |   |
| f2_tx | 11.8% | percent of the family's residents | share of the family's residents who live in TX | residents of tracts in the crowded renters family | ACS 2019-2023 |   | 6,870 tracts | none (full count of estimates) | American Community Survey |   |
| f3_adi | 54.7 | points of the ADI national rank (1 least, 100 most deprived) | population-weighted mean of the tract ADI national rank, bootstrap interval over counties | residents of tracts in the mid-range family | 2023 release on ACS 2019-2023 | 95% CI 53.8 to 55.7 | 37,746 tracts | ACS 2019-2023 tract population | Area Deprivation Index |   |
| f3_hv | 271 | thousands of dollars of tract median home value | population-weighted mean of the tract median home value over the tracts of the family, bootstrap interval over counties | residents of tracts in the mid-range family | ACS 2019-2023 | 95% CI 265 to 278 | 37,746 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| f3_inc | 95 | thousands of dollars of tract median family income | population-weighted mean of the tract median family income over the tracts of the family, bootstrap interval over counties | residents of tracts in the mid-range family | ACS 2019-2023 | 95% CI 94 to 95 | 37,746 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| f3_le | 78.6 | years of life expectancy at birth | population-weighted mean of the tract life expectancy over the tracts of the family, bootstrap interval over counties | residents of tracts in the mid-range family | deaths in 2010 to 2015 | 95% CI 78.5 to 78.7 | 36,163 tracts | ACS 2019-2023 tract population | USALEEP life expectancy |   |
| f3_pop | 153.1 million | residents | sum of ACS 2019-2023 tract population over the tracts of the family | residents of tracts in the mid-range family | ACS 2019-2023 |   | 37,746 tracts | none (full count of estimates) | American Community Survey |   |
| f3_pov | 9.9% | percent of persons below the poverty line | population-weighted mean of the tract poverty rate over the tracts of the family, bootstrap interval over counties | residents of tracts in the mid-range family | ACS 2019-2023 | 95% CI 9.8% to 10.0% | 37,746 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| f3_tracts | 37,746 | tracts | count of tracts in the mid-range family | census tracts in the frame (2020 tracts with a published ADI rank, residents and all 15 measures) | inputs from ACS 2019-2023 |   |   | none (full count) | American Community Survey |   |
| f4_hv | 168 | thousands of dollars of tract median home value | population-weighted mean of the tract median home value over the tracts of the family, bootstrap interval over counties | residents of tracts in the high poverty family | ACS 2019-2023 | 95% CI 162 to 175 | 13,817 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| f4_inc | 58 | thousands of dollars of tract median family income | population-weighted mean of the tract median family income over the tracts of the family, bootstrap interval over counties | residents of tracts in the high poverty family | ACS 2019-2023 | 95% CI 57 to 58 | 13,817 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| f4_kfr | 37.2 | percentile of the national household income rank in adulthood, children raised in families at the 25th income percentile | population-weighted mean of the tract adult income rank over the tracts of the family, bootstrap interval over counties | residents of tracts in the high poverty family | children born 1978 to 1983, income measured around 2014 to 2015 | 95% CI 36.9 to 37.6 | 13,814 tracts | ACS 2019-2023 tract population | Opportunity Atlas |   |
| f4_le | 74.7 | years of life expectancy at birth | population-weighted mean of the tract life expectancy over the tracts of the family, bootstrap interval over counties | residents of tracts in the high poverty family | deaths in 2010 to 2015 | 95% CI 74.5 to 74.9 | 12,903 tracts | ACS 2019-2023 tract population | USALEEP life expectancy |   |
| f4_pop | 47.4 million | residents | sum of ACS 2019-2023 tract population over the tracts of the family | residents of tracts in the high poverty family | ACS 2019-2023 |   | 13,817 tracts | none (full count of estimates) | American Community Survey |   |
| f4_pov | 26.5% | percent of persons below the poverty line | population-weighted mean of the tract poverty rate over the tracts of the family, bootstrap interval over counties | residents of tracts in the high poverty family | ACS 2019-2023 | 95% CI 26.2% to 26.8% | 13,817 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| f4_tracts | 13,817 | tracts | count of tracts in the high poverty family | census tracts in the frame (2020 tracts with a published ADI rank, residents and all 15 measures) | inputs from ACS 2019-2023 |   |   | none (full count) | American Community Survey |   |
| frame_pop_share | 93.0% | percent of residents of tracts with a published ADI rank | residents of the tracts in the frame as a share of residents of tracts with a published rank | residents of tracts with a published ADI rank and residents | ACS 2019-2023 |   | 75,717 tracts | none (full count of estimates) | American Community Survey |   |
| ghlth_gap_23 | 8.2 | percent of adults in fair or poor health (modeled) | difference between the population-weighted mean rate of fair or poor health of the crowded-renter and the mid-range family, bootstrap interval over counties with shared draws | residents of tracts in the crowded-renter and mid-range families | PLACES 2024 release, BRFSS 2022 | 95% CI 7.2 to 9.2 | 75,717 tracts | ACS 2019-2023 tract population | CDC PLACES |   |
| k | 4 | families | definition: the number of families the chapter uses | k-means fits compared | ACS 2019-2023 |   |   | none (definition) | American Community Survey |   |
| k_hi | 10 | families | definition: the most families in the comparison | k-means fits compared | ACS 2019-2023 |   |   | none (definition) | American Community Survey |   |
| k_lo | 2 | families | definition: the fewest families in the comparison | k-means fits compared | ACS 2019-2023 |   |   | none (definition) | American Community Survey |   |
| k_stab_hi | 8 | families | definition: the most families in the comparison of refits | k-means fits compared | ACS 2019-2023 |   |   | none (definition) | American Community Survey |   |
| kfr_gap_32 | 1.9 | percentile of the national household income rank in adulthood, children raised in families at the 25th income percentile | difference between the population-weighted mean adult income rank of the mid-range and the crowded-renter family, bootstrap interval over counties with shared draws | residents of tracts in the mid-range and crowded-renter families | children born 1978 to 1983, income measured around 2014 to 2015 | 95% CI 1.1 to 2.7 | 75,717 tracts | ACS 2019-2023 tract population | Opportunity Atlas |   |
| le_gap_14 | 6.9 | years of life expectancy at birth | difference between the population-weighted mean life expectancy of the high-income and the high-poverty family, bootstrap interval over counties with shared draws | residents of tracts in the high-income and high-poverty families | deaths in 2010 to 2015 | 95% CI 6.7 to 7.2 | 75,717 tracts | ACS 2019-2023 tract population | USALEEP life expectancy |   |
| le_gap_23 | 0.0 | years of life expectancy at birth | difference between the population-weighted mean life expectancy of the crowded-renter and the mid-range family, bootstrap interval over counties with shared draws | residents of tracts in the crowded-renter and mid-range families | deaths in 2010 to 2015 | 95% CI −0.5 to 0.4 | 75,717 tracts | ACS 2019-2023 tract population | USALEEP life expectancy |   |
| le_gap_dec10 | 2.4 | years of life expectancy at birth | difference between the population-weighted mean life expectancy of the mid-range and the high-poverty family within the most deprived decile of ADI ranks, bootstrap interval over counties with shared draws | residents of tracts in the most deprived decile of tracts that belong to the mid-range and high-poverty families | deaths in 2010 to 2015 | 95% CI 2.1 to 2.7 | 75,717 tracts | ACS 2019-2023 tract population | USALEEP life expectancy |   |
| n_adi_measures | 17 | measures | definition: count of measures in the index formula (Singh 2003, as printed in Petterson 2023), of which the chapter on what the deprivation index measures rebuilds the 15 used here | measures combined by the Area Deprivation Index | 2023 release on ACS 2019-2023 |   |   | none (definition) | Area Deprivation Index |   |
| n_boot_refits | 200 | refits | count of refits at each number of groups (definition) | refits on resamples of counties at each number of groups | ACS 2019-2023 |   |   | none (definition) | American Community Survey |   |
| n_cells | 40 | cells | count of cells (definition): ten deciles times four families | cells formed by decile and family | ACS 2019-2023 |   |   | none (definition) | American Community Survey |   |
| n_counties | 3,127 | counties | count of counties labeled | counties and county equivalents of the 50 states and DC with all 15 measures | ACS 2019-2023 |   |   | none (full count) | American Community Survey |   |
| n_county_measures | 14 | measures | count of measures (definition): the 15 measures without the single-parent share | measures in the distance that labels a county | ACS 2019-2023 |   |   | none (definition) | American Community Survey |   |
| n_measures | 15 | measures | count of measures (definition) | deprivation measures given to k-means | ACS 2019-2023 |   |   | none (definition) | American Community Survey |   |
| n_null | 3 | simulated clouds | count of simulated clouds (definition) | simulated data with no clusters | ACS 2019-2023 |   |   | none (definition) | American Community Survey |   |
| n_seeds | 6 | refits | count of refits (definition) | refits with other starting seeds | ACS 2019-2023 |   |   | none (definition) | American Community Survey |   |
| n_single | 30 | fits | count of fits (definition) | single-start fits of four groups | ACS 2019-2023 |   |   | none (definition) | American Community Survey |   |
| n_single_other | 13 | fits | count of single-start fits whose agreement with the families is below one half | single-start fits of four groups | ACS 2019-2023 |   |   | none (full count) | American Community Survey |   |
| n_tracts | 75,717 | tracts | count of tracts clustered | census tracts in the frame (2020 tracts with a published ADI rank, residents and all 15 measures) | inputs from ACS 2019-2023 |   |   | none (full count) | American Community Survey |   |
| n_tracts_pr | 981 | tracts | count of Puerto Rico tracts, left out of the frame (the frame of the chapter on what the deprivation index measures) | census tracts of Puerto Rico | 2020 census tracts |   |   | none (full count) | American Community Survey |   |
| n_tracts_ut_dc | 819 | tracts | count of tracts of Utah and the District of Columbia in the frame, whose single-parent share follows the county definition | census tracts in the frame (2020 tracts with a published ADI rank, residents and all 15 measures) | inputs from ACS 2019-2023 |   |   | none (full count) | American Community Survey |   |
| null_draw_gap | 0.007 | silhouette (-1 to 1) | the largest difference between the highest and lowest of the simulated draws at any number of groups | a seeded random sample of tracts in the frame | ACS 2019-2023 |   | 10,000 tracts | none (each tract counts once) | American Community Survey |   |
| r2_chance | 1.3% | percent of the variance of tract life expectancy | the most that 40 groups unrelated to the measure could account for by chance: groups minus one over counties minus one, which treats each county as one observation | residents of tracts with a life expectancy estimate | deaths in 2010 to 2015 |   | 3,101 counties | ACS 2019-2023 tract population | USALEEP life expectancy |   |
| r2_kfr_add_dec | 3.2 | percentage points of the population-weighted variance of tract adult income rank | difference between the R squared of the cells formed by decile and family over the four families, bootstrap interval over counties with shared draws | residents of tracts with the measure | children born 1978 to 1983, income measured around 2014 to 2015 | 95% CI 2.6 to 4.0 | 75,667 tracts | ACS 2019-2023 tract population | Opportunity Atlas |   |
| r2_kfr_add_fam | 7.6 | percentage points of the population-weighted variance of tract adult income rank | difference between the R squared of the cells formed by decile and family over the ten deciles, bootstrap interval over counties with shared draws | residents of tracts with the measure | children born 1978 to 1983, income measured around 2014 to 2015 | 95% CI 6.3 to 9.1 | 75,667 tracts | ACS 2019-2023 tract population | Opportunity Atlas |   |
| r2_kfr_both | 30.9% | percent of the population-weighted variance of tract adult income rank | R squared of the group means: variance between the cells formed by decile and family over the total variance of the tract measure, bootstrap interval over counties | residents of tracts with the measure | children born 1978 to 1983, income measured around 2014 to 2015 | 95% CI 28.7% to 33.4% | 75,667 tracts | ACS 2019-2023 tract population | Opportunity Atlas |   |
| r2_kfr_dec | 23.3% | percent of the population-weighted variance of tract adult income rank | R squared of the group means: variance between the ten deciles of the ADI rank over the total variance of the tract measure, bootstrap interval over counties | residents of tracts with the measure | children born 1978 to 1983, income measured around 2014 to 2015 | 95% CI 21.0% to 26.2% | 75,667 tracts | ACS 2019-2023 tract population | Opportunity Atlas |   |
| r2_kfr_fam | 27.7% | percent of the population-weighted variance of tract adult income rank | R squared of the group means: variance between the four families over the total variance of the tract measure, bootstrap interval over counties | residents of tracts with the measure | children born 1978 to 1983, income measured around 2014 to 2015 | 95% CI 25.6% to 30.0% | 75,667 tracts | ACS 2019-2023 tract population | Opportunity Atlas |   |
| r2_kfr_fam_minus_dec | 4.4 | percentage points of the population-weighted variance of tract adult income rank | difference between the R squared of the four families over the ten deciles, bootstrap interval over counties with shared draws | residents of tracts with the measure | children born 1978 to 1983, income measured around 2014 to 2015 | 95% CI 2.5 to 6.2 | 75,667 tracts | ACS 2019-2023 tract population | Opportunity Atlas |   |
| r2_le_add_dec | 8.0 | percentage points of the population-weighted variance of tract life expectancy | difference between the R squared of the cells formed by decile and family over the four families, bootstrap interval over counties with shared draws | residents of tracts with the measure | deaths in 2010 to 2015 | 95% CI 7.3 to 8.8 | 71,874 tracts | ACS 2019-2023 tract population | USALEEP life expectancy |   |
| r2_le_add_fam | 4.7 | percentage points of the population-weighted variance of tract life expectancy | difference between the R squared of the cells formed by decile and family over the ten deciles, bootstrap interval over counties with shared draws | residents of tracts with the measure | deaths in 2010 to 2015 | 95% CI 4.1 to 5.4 | 71,874 tracts | ACS 2019-2023 tract population | USALEEP life expectancy |   |
| r2_le_both | 41.3% | percent of the population-weighted variance of tract life expectancy | R squared of the group means: variance between the cells formed by decile and family over the total variance of the tract measure, bootstrap interval over counties | residents of tracts with the measure | deaths in 2010 to 2015 | 95% CI 39.2% to 43.3% | 71,874 tracts | ACS 2019-2023 tract population | USALEEP life expectancy |   |
| r2_le_dec | 36.6% | percent of the population-weighted variance of tract life expectancy | R squared of the group means: variance between the ten deciles of the ADI rank over the total variance of the tract measure, bootstrap interval over counties | residents of tracts with the measure | deaths in 2010 to 2015 | 95% CI 34.3% to 39.0% | 71,874 tracts | ACS 2019-2023 tract population | USALEEP life expectancy |   |
| r2_le_dec_minus_fam | 3.3 | percentage points of the population-weighted variance of tract life expectancy | difference between the R squared of the ten deciles over the four families, bootstrap interval over counties with shared draws | residents of tracts with the measure | deaths in 2010 to 2015 | 95% CI 2.0 to 4.5 | 71,874 tracts | ACS 2019-2023 tract population | USALEEP life expectancy |   |
| r2_le_fam | 33.3% | percent of the population-weighted variance of tract life expectancy | R squared of the group means: variance between the four families over the total variance of the tract measure, bootstrap interval over counties | residents of tracts with the measure | deaths in 2010 to 2015 | 95% CI 31.4% to 35.2% | 71,874 tracts | ACS 2019-2023 tract population | USALEEP life expectancy |   |
| sil_half_max | 0.003 | silhouette (-1 to 1) | largest half-width of the 95% interval for the sample silhouette over the fits compared | a seeded random sample of tracts in the frame | ACS 2019-2023 |   | 10,000 tracts | none (each tract counts once) | American Community Survey |   |
| sil_k | 0.21 | silhouette (-1 to 1) | mean silhouette of the four families, interval from the sampling error of the sample | a seeded random sample of tracts in the frame | ACS 2019-2023 | 95% CI 0.21 to 0.21 | 10,000 tracts | none (each tract counts once) | American Community Survey |   |
| sil_k2 | 0.22 | silhouette (-1 to 1) | mean silhouette of the 2-family k-means fit to the tract data, interval from the sampling error of the sample | a seeded random sample of tracts in the frame | ACS 2019-2023 | 95% CI 0.22 to 0.22 | 10,000 tracts | none (each tract counts once) | American Community Survey |   |
| sil_k5 | 0.15 | silhouette (-1 to 1) | mean silhouette of the 5-family k-means fit to the tract data, interval from the sampling error of the sample | a seeded random sample of tracts in the frame | ACS 2019-2023 | 95% CI 0.15 to 0.15 | 10,000 tracts | none (each tract counts once) | American Community Survey |   |
| sil_margin_k | 0.09 | silhouette (-1 to 1) | silhouette of the tract data minus the silhouette of simulated data with no clusters, four groups | a seeded random sample of tracts in the frame | ACS 2019-2023 |   | 10,000 tracts | none (each tract counts once) | American Community Survey |   |
| sil_margin_other | 0.03 | silhouette (-1 to 1) | the largest silhouette margin over simulated data at any number of groups other than four | a seeded random sample of tracts in the frame | ACS 2019-2023 |   | 10,000 tracts | none (each tract counts once) | American Community Survey |   |
| sil_neg_share | 13.4% | percent of sampled tracts | share of sampled tracts with a negative silhouette in the four families (nearer to another family than to their own) | a seeded random sample of tracts in the frame | ACS 2019-2023 |   | 10,000 tracts | none (each tract counts once) | American Community Survey |   |
| sil_null_k | 0.12 | silhouette (-1 to 1) | mean silhouette of the 4-group k-means fit to simulated data with no clusters (mean of 3 draws) | a seeded random sample of tracts in the frame | ACS 2019-2023 |   | 10,000 tracts | none (each tract counts once) | American Community Survey |   |
| sil_null_k2 | 0.21 | silhouette (-1 to 1) | mean silhouette of the 2-group k-means fit to simulated data with no clusters (mean of 3 draws) | a seeded random sample of tracts in the frame | ACS 2019-2023 |   | 10,000 tracts | none (each tract counts once) | American Community Survey |   |
| ss_between | 41.8% | percent of the total variance of the standardized measures | share of the total variance of the 15 standardized measures that lies between the four families | census tracts in the frame | ACS 2019-2023 |   | 75,717 tracts | none (each tract counts once) | American Community Survey |   |
| ss_between_null | 38.3% | percent of the total variance of the standardized measures | share of the total variance that lies between four k-means groups in simulated data with no clusters (mean of the draws) | census tracts in the frame | ACS 2019-2023 |   | 75,717 tracts | none (each tract counts once) | American Community Survey |   |
| total_pop | 307.1 million | residents | sum of ACS 2019-2023 tract population over the tracts in the frame | residents of the tracts in the frame | ACS 2019-2023 |   | 75,717 tracts | none (full count of estimates) | American Community Survey |   |
| z_f2_crowded | 2.2 | standard deviations above the mean tract | population-weighted mean of the tract crowding z-score over the tracts of the family, bootstrap interval over counties | residents of tracts in the crowded renters family | ACS 2019-2023 | 95% CI 1.9 to 2.5 | 6,870 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| z_f4_poverty | 1.4 | standard deviations above the mean tract | population-weighted mean of the tract poverty z-score over the tracts of the family, bootstrap interval over counties | residents of tracts in the high poverty family | ACS 2019-2023 | 95% CI 1.4 to 1.4 | 13,817 tracts | ACS 2019-2023 tract population | American Community Survey |   |
| z_gap_max | 0.2 | standard deviations | the largest difference, over families and measures, between the unweighted mean z-score (the k-means center) and the population-weighted mean z-score | census tracts in the frame | ACS 2019-2023 |   | 75,717 tracts | none (each tract counts once) | American Community Survey |   |
| z_mid_max | 0.34 | standard deviations from the mean tract | the largest absolute population-weighted mean z-score of the mid-range family over the 15 measures | residents of tracts in the mid-range family | ACS 2019-2023 |   | 37,746 tracts | ACS 2019-2023 tract population | American Community Survey |   |

## Definitions

- K-means: A method that sorts places into a chosen number of groups so that places in a group resemble one another on the measures used. The analyst sets the number of groups.
- Area Deprivation Index (ADI): A score that ranks neighborhoods by income, education, employment and housing conditions drawn from the American Community Survey. In this book a higher national percentile means a more deprived neighborhood.
- Life expectancy at birth: The average number of years a newborn would live if death rates at every age stayed as they were in the period measured. It summarizes the death rates of a period and is not a forecast for any child.
- Income rank (kfr): The position that adult children reach in the national distribution of household income, from the lowest rank to the highest, averaged over children who grew up in a place with parents at a stated income level. The Opportunity Atlas labels this measure kfr.
- Silhouette: A score for how well a place fits the group it was assigned in a clustering. Higher values mean the place sits close to its own group and far from the others.
- Modeled estimate: A figure produced by a statistical model instead of a direct count or measurement, usually for small areas where direct data are thin, borrowing strength from surveys, neighboring areas or other data. Neighboring areas can share errors, so single extremes call for caution.
- Ecological inference: Drawing conclusions about individuals from data about places or groups. A pattern across counties need not hold for the people inside them.
- USALEEP: The U.S. Small-area Life Expectancy Estimates Project, a National Center for Health Statistics release of life expectancy at birth for most census tracts. It covers deaths from 2010 to 2015 only and has not been updated.

## Methods

This chapter sorts the 75,717 census tracts of the frame into 4 families with k-means on 15 ACS 2019-2023 measures, and describes each family with outcomes that were not inputs. The fit counts each tract once and is unweighted by design, family means use tract population as weights, and the intervals resample whole counties and do not carry uncertainty about the fit. The held-out outcomes come from other periods and models: life expectancy for deaths in 2010 to 2015, adult income rank for children born 1978 to 1983, and CDC PLACES health measures that are modeled estimates. Puerto Rico's tracts are outside the frame, and county labels come from each county's nearest center, since no clustering is run on counties.

### Which tracts and measures are used

The frame is the 2020 tracts that have a published ADI rank, residents and all 15 measures (the frame of the chapter on the anatomy of a deprivation index, loaded by its loader). The 15 measures are a subset of the 17 that the Area Deprivation Index combines. Puerto Rico's tracts are not in the frame, as in that chapter.

The tracts of Utah and the District of Columbia carry the county definition of the single-parent share, which differs from the tract file's definition.  The measures are in their own units: dollars for the four dollar amounts, proportions for the other ten, and the log ratio of low-income to high-income households.

### How the families are fitted

Each measure is standardized across tracts with each tract counted once, and k-means (scikit-learn, Lloyd's algorithm, k-means++ starts, 4 groups, 20 starts, seed 12345, one thread) places the centers. The fit is unweighted by design. Families are numbered by the population-weighted mean of the published ADI, least deprived first.

### How the number of families is chosen

For 2 to 10 groups the build fits k-means with 10 starts and computes the mean silhouette on a seeded random sample of 10,000 tracts (seed 12345). The same fits are run on 3 simulated clouds of the same size that keep each measure's distribution and the correlations of normal scores among measures (a Gaussian copula, seeds 12345 to 12347) and have no clusters.

Explains Table 1.

### Held-out outcomes, deciles and county labels

Life expectancy leaves out institutional tracts (25 percent or more of residents in group quarters). Adult income rank, life expectancy and PLACES measures are not inputs. PLACES values are modeled from individual age, sex, race and education and county poverty, post-stratified to tract counts.

Deciles of the published ADI national rank are formed on the tracts of the frame and are close to equal in size, because tied ranks fall in one decile.

Each county takes the label of the family whose center lies nearest its own 14 measures (all but the single-parent share, which is defined differently for tracts and counties), standardized with the tract mean and standard deviation; no clustering is run on counties, and each tract takes its county's label.

Explains Table 4, Table 5 and Table 6.

### Means and intervals

Family means use ACS 2019-2023 tract population as weights. Intervals are 95% percentile intervals from 2,000 draws of whole counties with replacement (seed 12345); differences between two families and the variance explained by groups use the same draws. Intervals condition on the families and do not carry uncertainty about the fit.

Explains Table 2 and Table 3.

### Stability refits

Stability is tested with refits using 6 other seeds; a refit weighted by tract population; a refit with the four dollar amounts logged; 30 single-start fits to find other arrangements; and 200 refits at each number of groups from 2 to 8 (20 starts each, as many as the main fit) on resamples that draw whole counties with replacement (Connecticut tracts are assigned to its eight legacy counties; the generator of resample b is seeded with 12345 and b, and the same resamples serve every number of groups), each assigning every tract to its nearest center.

Agreement is the adjusted Rand index. The share of refits with an agreement below 0.5 has a Wilson interval that treats the refits as independent.

The refits run in worker processes; each depends only on its seed, so the result does not depend on the number of workers.

## What these data cannot show

The families describe tracts, and every comparison in this chapter is ecological. The life expectancy of a family is an average over tracts, and it says little about any one resident of one tract (ecological inference). Nothing here shows that living in a family's tracts changes a person's outcome. The families are arrangements of census measures, and the outcomes that line up with them reflect selection, history, prices and policy that these data do not separate.

The number and the names of the families are choices. The silhouette favors no number of groups on its own, and the benchmark picks four by a stated rule. Refits find another arrangement more often at four groups than at three or five, and a second arrangement of four groups has only 1.1% more within-group variance than the first. Weighting by residents or rescaling the dollar measures changes the partition. The crowded-renter family depends most on those choices. The frame leaves out tracts that lack a median rent, home value or mortgage cost, which the ACS withholds for tracts with almost no renters or almost no owners, and it leaves out Puerto Rico, so none of these results describes the island.

The measures are tract estimates for 2019 to 2023 without margins of error. The tract file defines the share of single-parent households differently from the county tables, while the 819 tracts of Utah and the District of Columbia carry the county definition. Life expectancy is for deaths in 2010 to 2015 and was interpolated from 2010 to 2020 tracts by area, and the USALEEP estimates suit maps and gradients and not single tracts. Adult income rank describes children born in 1978 to 1983 and was interpolated to 2020 tracts by area. PLACES values are modeled, and the ADI averages block-group ranks over tracts without population weights. The county labels use county estimates of the same measures and treat Connecticut's planning regions as counties.

## References

1. MacQueen, J. (1967). Some methods for classification and analysis of multivariate observations. Proceedings of the Fifth Berkeley Symposium on Mathematical Statistics and Probability, Volume 1: Statistics, 281-297.
2. Rousseeuw, Peter J. (1987). Silhouettes: a graphical aid to the interpretation and validation of cluster analysis. Journal of Computational and Applied Mathematics 20: 53-65. https://doi.org/10.1016/0377-0427(87)90125-7
3. Hubert, Lawrence; Arabie, Phipps (1985). Comparing partitions. Journal of Classification 2(1): 193-218. https://doi.org/10.1007/BF01908075
4. Hennig, Christian (2007). Cluster-wise assessment of cluster stability. Computational Statistics & Data Analysis 52(1): 258-271. https://doi.org/10.1016/j.csda.2006.11.025

## Provenance

- Pipeline version: 0.1.0
- Build script: book2/chapters/families-of-places/build.py (sha256 fcb706d743590f60d4d04f5e5603684f782888d93d17e4f0133be0d0d400603a)
- Built on: 2026-10-07
- Site commit: c2b1ff3309cb4a10833f884da91ac0057a66ad0a

Inputs:

| File | Size (bytes) | SHA-256, first 1 MB | SHA-256, last 1 MB |
| --- | ---: | --- | --- |
| book2/chapters/anatomy-of-deprivation/deprivation.py | 18,028 | 092694cfd30446c54292ca7182d4d784e95f28766c0db63fedf53f84c7bb1b7d | 092694cfd30446c54292ca7182d4d784e95f28766c0db63fedf53f84c7bb1b7d |
| book2/chapters/families-of-places/families_data.py | 23,256 | 41dc810414ea1eccb3899088e8c6ab91a208263239defaa5b7a65305cedb7eb9 | 41dc810414ea1eccb3899088e8c6ab91a208263239defaa5b7a65305cedb7eb9 |
| data/analytic/tract_atlas.parquet | 39,659,925 | 29d5e4d943552d54aeb6ba92aadc939d9bd15bdea553c78357dbd516757be627 | 1e10d1ecf722340f23ee2212d4999a90250f85b65fa89b39e32db06d9472e22e |
| data/raw/acs/deprivation_components_2019_2023/county.parquet | 1,004,968 | c894ba6cf96378244f47cc4271a24bff087d15ba27f09e49a4ebc0a4786c6f9d | c894ba6cf96378244f47cc4271a24bff087d15ba27f09e49a4ebc0a4786c6f9d |
| data/raw/acs/deprivation_components_2019_2023/county_core.parquet | 498,878 | 8b177857027e73c1f404f7b62191ab78811e95ee54c4f07383e9e9e17da1a345 | 8b177857027e73c1f404f7b62191ab78811e95ee54c4f07383e9e9e17da1a345 |
| data/raw/acs/deprivation_components_2019_2023/tract.parquet | 13,197,793 | 5545cde2cbe96a8b56e468d65388248a241aa6bd300d3c9e325d08d3f86b11da | 5c8ec454d066988b0f1a01aaa6ed34ebfc05d162e01cf9c6bb01f9198cbaff12 |
| data/staged/adi/adi_atlas_tract2020_2023.parquet | 798,078 | 5e409bc212ff871def780fa27b4cf8b616d9da420b2b9f6455517eddcd8c1a95 | 5e409bc212ff871def780fa27b4cf8b616d9da420b2b9f6455517eddcd8c1a95 |
| data/staged/crosswalks/ct_tract2020_to_old_county.parquet | 15,842 | 961ef39f31596ef65fca2c67d5f2f63e533c5dc9b49745d3bb6437844e4b831d | 961ef39f31596ef65fca2c67d5f2f63e533c5dc9b49745d3bb6437844e4b831d |
| data/staged/life_expectancy/usaleep_tract2020.parquet | 1,316,169 | 9c9c7f103791b75eddba74e68a89804efdf2486b6b66bb4bc02a46ba5bf8bc23 | 27317107be2465d9fd5e83271dea0cd6e915ee419e2802bedc28747d0a41d38f |
| data/staged/ses/acs5_tract2020_2019_2023.parquet | 7,232,424 | 5fe6f730a35225f14d5d8abfc6b9bce58b52e545f309a7eac74f1dae74d7a31c | 7b4a2475b3852d3f91804716a04566ef8ceadacec2be9fb85f5eb642078efac6 |
| data/staged/svi/svi_tract2020_2022.parquet | 17,740,415 | cb3860d8edc2aaf1f2a02211a58f7a0b80457eb6ad9a37b5ec8abb2d5da9da10 | e371a4a696d5bed05d54ad7c5b6ca2504508805b847ec415c2322be7f66ec2b2 |

Outputs:

| File | SHA-256 |
| --- | --- |
| facts.json | 08466a4097eaac08b5d2e1dfae8c1ca703031b8ce6b3fe47fd0e305e521a3f3d |
| figures/adi_decile_composition.json | 34f80f19cbc2a4685e3de9a8a0a6c10b859ad49e65a42955f28e6629bf290e83 |
| figures/county_group_composition.json | 5d7e485b59f3d518877365296f46bd417c57461077af415982c2546771926f94 |
| figures/family_profiles.json | dba608e1df40e9909de89b387b1512a95d929d8e035cbed1e1b4ae3e1149af0c |
| figures/family_table.json | 092d8c9e5f55f7b33bc05832500e0c8e2ff1ab9ad7794e3c288b86f1edd9e4b3 |
| figures/held_out_outcomes.json | 9715ccb8c0bd1dc8215395e79723bf22944973ed875d1a7d31234d34c25696e5 |
| figures/number_of_families.json | 61ac7ef104efb337fd1cde276c8e0babe1f0855aa962fc699d1810e1433504cd |
| meta.json | bd419ebe0b487c5bec3758ecde3907369233241d8120b55436f20624942dd381 |
| refs.json | 979ce72ce7e56ad2f2fa0b196a3e6c986a3807d5b36c8a3d1d0532d742e6c62c |

## Machine formats

- Report (HTML): https://vishalsingh.org/zipcode/families-of-places/report
- Report (Markdown): https://vishalsingh.org/zipcode/families-of-places/report.md
- Every estimate (JSON): https://vishalsingh.org/zipcode/report/families-of-places/facts.json
- Table 1, figure "number_of_families" (JSON): https://vishalsingh.org/zipcode/report/families-of-places/figures/number_of_families.json
- Table 2, figure "family_profiles" (JSON): https://vishalsingh.org/zipcode/report/families-of-places/figures/family_profiles.json
- Table 3, figure "family_table" (JSON): https://vishalsingh.org/zipcode/report/families-of-places/figures/family_table.json
- Table 4, figure "held_out_outcomes" (JSON): https://vishalsingh.org/zipcode/report/families-of-places/figures/held_out_outcomes.json
- Table 5, figure "adi_decile_composition" (JSON): https://vishalsingh.org/zipcode/report/families-of-places/figures/adi_decile_composition.json
- Table 6, figure "county_group_composition" (JSON): https://vishalsingh.org/zipcode/report/families-of-places/figures/county_group_composition.json
- Index of every report: https://vishalsingh.org/zipcode/llms.txt; all reports in one file: https://vishalsingh.org/zipcode/llms-full.txt

---

# Suppressed counties in death data: report

Chapter 10.3 of ZIP Code Destiny, by Vishal Singh (NYU Stern). Generated from the chapter's data files; the reading version is at https://vishalsingh.org/zipcode/suppressed-counties-in-death-data. Table n of this report is figure n of the chapter.

## Identification

- Title: Suppressed counties in death data
- Label: The counties that don't count
- Chapter: 10.3 (Part X, How the estimates are made; Methods and sources)
- Question: Which counties have death counts withheld for privacy, and how does the withholding bias comparisons between counties?
- Dek: CDC WONDER withheld the death count in 47.1% to 57.9% of county-years for overdose, alcohol-induced deaths and suicide from 2018 to 2023, and the counties it published in every year are selected on their death rates.
- Grain: counties
- Status: published
- Published: 2026-10-09
- Updated: not revised
- Data sources:
  - CDC WONDER underlying cause of death. CDC WONDER underlying cause of death, county files for drug poisoning (overdose), alcohol-induced causes and suicide, 2018 to 2024, and state files for the same causes; years: 2018-2024; weights: none (full counts of deaths; crude rates); WONDER county population
  - CDC WONDER natality. CDC WONDER natality, county totals 2018 to 2024: births in the counties that WONDER names and in the pooled Unidentified Counties rows; years: 2018-2024; weights: none (full counts of births)
  - CDC WONDER linked birth and infant death. CDC WONDER linked birth and infant death, county totals 2018 to 2022: births in the counties that WONDER names and in the pooled rows; years: 2018-2022; weights: none (full counts of births)
  - CDC injury and violence mapping. CDC Mapping Injury, Overdose, and Violence county file: modeled overdose and suicide death rates, annual 2019 to 2023; years: 2019-2024 for counties, 2022-2024 for tracts (data as of 2026-05-14); weights: mean WONDER population of each county
  - County Health Rankings. County Health Rankings 2024: counties with no rate for overdose, suicide, homicide, firearm deaths and premature death, and race and ethnicity shares of residents (Census Population Estimates); years: 2024 release; weights: county population in County Health Rankings
  - Area Deprivation Index. Neighborhood Atlas Area Deprivation Index 2023, tract national rank averaged to counties; years: 2023 release (version 4.0.1; built on ACS 2019-2023); weights: 2019-2023 tract population
  - Rural-Urban Continuum Codes. USDA Rural-Urban Continuum Codes 2023, county; years: 2023; weights: none (classification)
  - American Community Survey. Total population of 2020 tracts, 2019-2023, summed to counties as population for deprivation and for the shares of residents in the table; years: 2019-2023 (5-year); weights: none (full count)
- Canonical URL: https://vishalsingh.org/zipcode/suppressed-counties-in-death-data
- Report: https://vishalsingh.org/zipcode/suppressed-counties-in-death-data/report
- Markdown: https://vishalsingh.org/zipcode/suppressed-counties-in-death-data/report.md
- BibTeX key: singh2026suppressedcountiesindeathdata

```bibtex
@misc{singh2026suppressedcountiesindeathdata,
  author = {Singh, Vishal},
  title = {Suppressed counties in death data},
  year = {2026},
  url = {https://vishalsingh.org/zipcode/suppressed-counties-in-death-data},
  note = {In ZIP Code Destiny. Published 9 October 2026.}
}
```

## Principal estimates

The facts the dek and the opening prose cite, in order of appearance.

| Estimand | Population | Period | Value | Interval | n | Weights | Source | Fact id |
| --- | --- | --- | ---: | --- | ---: | --- | --- | --- |
| share of county-years whose drug overdose count WONDER withholds | county-years in the CDC WONDER county files, each kind of death counted separately | county-years in 2018 to 2023 | 47.1% |   | 18,852 county-years | none (full count) | CDC WONDER underlying cause of death | w_wh_od |
| share of county-years whose alcohol-induced count WONDER withholds | county-years in the CDC WONDER county files, each kind of death counted separately | county-years in 2018 to 2023 | 57.9% |   | 18,852 county-years | none (full count) | CDC WONDER underlying cause of death | w_wh_alc |
| share of county-years whose suicide count WONDER withholds | county-years in the CDC WONDER county files, each kind of death counted separately | county-years in 2018 to 2023 | 57.5% |   | 18,852 county-years | none (full count) | CDC WONDER underlying cause of death | w_wh_su |
| share of the summed county populations that sit in county-years whose drug overdose count is withheld (WONDER publishes the population) | residents counted in the county-years of the WONDER county files | county-years in 2018 to 2023 | 10.1% |   |   | none (full count) | CDC WONDER underlying cause of death | w_pop_od |
| share of the summed county populations that sit in county-years whose alcohol-induced count is withheld (WONDER publishes the population) | residents counted in the county-years of the WONDER county files | county-years in 2018 to 2023 | 14.8% |   |   | none (full count) | CDC WONDER underlying cause of death | w_pop_alc |
| share of the summed county populations that sit in county-years whose suicide count is withheld (WONDER publishes the population) | residents counted in the county-years of the WONDER county files | county-years in 2018 to 2023 | 12.2% |   |   | none (full count) | CDC WONDER underlying cause of death | w_pop_su |
| definition: WONDER withholds a county-year count above zero and below this number | county-years in the CDC WONDER county files, each kind of death counted separately | not applicable (definition) | 10 |   |   | none (definition) | CDC WONDER underlying cause of death | w_floor |

## Findings

CDC WONDER, the federal database of death certificates, withheld the death count for 47.1% of county-years for drug overdose from 2018 to 2023, 57.9% for alcohol-induced deaths and 57.5% for suicide. The withheld county-years hold only 10.1%, 14.8% and 12.2% of residents, so the blanks cover many places and few people. They are also a selected set of places. WONDER publishes a county-year count when it is zero or at least 10 deaths, so a county enters the published data in the years when its count is high. A rate pooled over those years is biased. Here that means it differs from the county's true rate in a systematic direction that depends on the size of the county. It falls below the rate in CDC's modeled injury and violence panel in the smallest counties and rises above it in counties of middling small size.

This chapter explains the mechanism behind a rule that every chapter using WONDER follows. A withheld cell is counted and left out of both the numerator and the denominator of a rate, and the book never fills it in. It discloses only the largest share of deaths that the withheld cells could hold at the public limit. County rates in other chapters use complete counties, those whose counts are published in every year, or a modeled source. The sections below take up the rule, the counties it removes, the bias it produces and a second rule, which decides which counties WONDER names at all in its birth products. The cause maps belong to Chapter 8.2, Overdose, alcohol and suicide (https://vishalsingh.org/zipcode/overdose-alcohol-and-suicide), the birth outcomes to Chapter 2.1, Birthweight and preterm birth (https://vishalsingh.org/zipcode/birthweight-and-preterm-birth), and a figure of suppressed cells by birthweight class to Chapter 10.1, Gradients, slopes and variance (https://vishalsingh.org/zipcode/gradients-slopes-and-variance).

### WONDER withholds small counts and publishes zeros on request

The county files of WONDER's Underlying Cause of Death database show a count above zero and below 10 as Suppressed, and the database's terms bar users from presenting a death count of 9 or fewer, or a rate built on one. A suppressed cell says only that the count is small. It is missing, and a count of zero is published as zero. Zeros appear only when a query asks for them. A default WONDER query omits county-years with no deaths, and the pulls behind this book used the option that shows zero values, so a zero here is a published count and not an absent row.

Agencies withhold small counts for two reasons that are easy to confuse.[1] One is confidentiality, which sets the limit of 10 for deaths and births in subnational tables.[2] The other is reliability. Until recently NCHS suppressed or flagged any rate based on fewer than 20 events, the count at which the relative standard error of a Poisson count reaches 22.4%. Its current standard asks for at least 10 events and a 95% confidence interval no wider than 160% of the estimate.[2] The first rule protects the people counted, and the second protects the reader of the rate.

Maps in this book keep the earlier floor of 20 pooled deaths for every year, so that every map follows one rule that is stricter than the current NCHS minimum. Gradients in other chapters use no floor and weight each place by its population.

Publication clears WONDER's limit and not the reliability bar. Among county-years that WONDER publishes with at least 10 deaths, 35.4% for overdose, 43.8% for alcohol-induced deaths and 44.8% for suicide have fewer than 20 deaths.

This chapter uses the county files of the Underlying Cause of Death database for 2018 to 2024 (single race), which carry a row for each of 3,142 counties and each year, for three kinds of death. They are drug poisoning of any intent, which the chapter calls overdose, alcohol-induced deaths and suicide. Over 2018 to 2023 that is 18,852 county-years for each kind of death. Rates are crude. Every year from 2018 to 2024 is final in the June 2026 pull, although North Carolina's 2023 overdose counts predate the state's amended death certificates and are low. WONDER publishes a county's population whether or not it withholds the deaths, which makes it possible to measure how many residents live in a county-year with a blank.

Table 1. In every year from 2018 to 2024 WONDER withheld between 44.6% and 59.9% of county-year death counts, in county-years that hold between 8.6% and 17.4% of residents

In every year from 2018 to 2024 the withheld share ran from 44.6% to 59.9% of county-years across the three kinds of death, and the withheld county-years held between 8.6% and 17.4% of residents.

The share of residents fell for overdose, from 12.3% in 2018 to 8.6% in its lowest year, and for alcohol-induced deaths, from 17.4% to 12.5%, while for suicide it stayed between 11.8% and 12.7%.

A blank is not a zero. Once a query asks for zeros, they are a large part of what WONDER publishes. Counts of zero are 30.0% of the published overdose county-years, 34.2% of the alcohol-induced ones and 24.7% of the suicide ones.

A rate pooled over a county's published years keeps its years with no deaths and drops its years with a few, which is the bias that the later sections measure. For suicide, the pooled rate is 0.18 (95% CI 0.00 to 0.42) times the panel's rate in counties under 10,000 residents and 1.51 (95% CI 1.47 to 1.55) times in counties of 25,000 to 50,000 residents (Table 7).

### Withholding is heaviest in counties of middling small size

How often a count is withheld depends on how many deaths a county expects. Table 2 sorts counties by their mean population over 2018 to 2023 into six classes, with boundaries at 10,000, 25,000, 50,000, 100,000 and 250,000 residents. The last two are the populations at which WONDER names a county in its birth products, taken up below.

Table 2. WONDER withholds 82.3% of overdose county-years in counties of 10,000 to 25,000 residents and 45.2% in smaller counties, where 54.7% are published as zero

Counties of 10,000 to 25,000 residents are withheld most often, in 82.3% of overdose county-years, 84.3% of alcohol-induced county-years and 93.1% of suicide county-years.

Counties under 10,000 residents are withheld less, in 45.2%, 54.9% and 61.3%, because many of their county-years have no deaths and WONDER publishes the zero. Published zeros are 54.7%, 44.8% and 38.5% of their county-years.

At the large end, counties of 100,000 to 250,000 residents are withheld in 2.5% of overdose county-years, 9.7% of alcohol-induced county-years and 1.5% of suicide county-years. Counties of 250,000 or more are withheld in 0.0% of overdose, 0.1% of alcohol-induced and 0.0% of suicide county-years. A county that expects more than a handful of deaths in a year but fewer than the limit is withheld in most years, and the section on bias below works out how much that selects.

### Withheld years are fewest in the Northeast and the Pacific states

Table 3 counts, for each county, the years from 2018 to 2023 in which its count is withheld. A county published in every year has a count, possibly zero, and a population in each of those years. That is a complete county. Other than Connecticut's counties there are 942 for overdose, 644 for alcohol-induced deaths and 745 for suicide. At the other end, 591 counties for overdose, 782 for alcohol-induced deaths and 919 for suicide are withheld in every year.

The overdose map shades 950 counties at no withheld year. They are the 942 complete counties and Connecticut's 8 counties, which have deaths but no population in 2022 and 2023 and so are not complete. The alcohol-induced and suicide maps shade 652 and 752 counties at no withheld year, again the complete counties and the Connecticut counties with no withheld year.

Table 3. WONDER withheld the overdose count in every year from 2018 to 2023 in 591 counties and in no year in 950

Counties in New England, the Middle Atlantic and the Pacific division average fewer withheld years than counties in any other Census division, for each kind of death. For suicide counts, 32.7% of counties in the South and 31.1% in the Midwest are withheld in every year, against 22.5% in the West and 12.0% in the Northeast. The map shows where blanks occur, and it says nothing about whether the deaths behind a blank are many or few.

### Residents of withheld counties live in smaller, more rural and more deprived places

The counties fall into three groups by how often WONDER withholds their count: in every year, in some years or in no year. Take overdose. The 942 counties published in every year hold 83.8% of the residents of the 3,134 counties other than Connecticut's.

The 1,601 counties withheld in some years are 51.1% of counties and hold 12.1% of residents. The 591 counties withheld in every year hold 4.1% (Table 4).

Table 4. Counties whose overdose count WONDER publishes in every year hold 83.8% of residents, and counties withheld in every year hold 4.1% and are smaller, more rural and more deprived

The counties withheld in every year are small and rural. Their median county has 19,761 residents, against 134,390 in counties published every year. 75.5% (95% CI 71.2% to 79.9%) of their residents live in nonmetro counties, those with a Rural-Urban Continuum Code of 4 or higher, against 4.2% (95% CI 3.4% to 5.1%) in counties published every year. They are more deprived. The population-weighted mean Area Deprivation Index rank is 70.2 (95% CI 68.5 to 71.8) against 43.0 (95% CI 39.9 to 46.0), where a higher rank is more deprived.

By County Health Rankings' figures, residents of the counties withheld in every year are less often Hispanic or non-Hispanic Black and more often non-Hispanic white or American Indian or Alaska Native. 10.0% (95% CI 8.4% to 11.7%) are Hispanic against 21.0% (95% CI 18.2% to 24.0%), and 8.1% (95% CI 7.1% to 9.2%) are non-Hispanic Black against 13.4% (95% CI 12.1% to 14.7%). 77.2% (95% CI 75.3% to 79.1%) are non-Hispanic white against 55.3% (95% CI 52.1% to 58.5%), and 2.1% (95% CI 1.7% to 2.6%) are American Indian or Alaska Native against 1.2% (95% CI 1.0% to 1.3%).

The alcohol-induced and suicide views show differences between counties published every year and counties withheld in every year in the same direction on every measure. For alcohol-induced deaths the counties withheld in every year number 782 and hold 6.4% of residents, and for suicide they number 919 and hold 5.8%. The counties withheld in some years number 1,708 for alcohol-induced deaths and 1,470 for suicide.

These are properties of counties and of their residents, and not of the people who died (ecological inference). For a comparison of counties, the consequence is that published counts describe the larger, more urban and less deprived counties, with their higher shares of Hispanic and Black residents, and leave out the rural counties where deaths are few.

### Counties published every year have higher rates, except the smallest

Among the smallest counties the order reverses. A county's count is published when it is high, so the counties published in every year are the ones with high counts. Testing that needs a rate that does not depend on WONDER's rule. CDC's Mapping Injury, Overdose, and Violence panel supplies one for overdose and suicide. It gives modeled estimates of death rates for every county except Connecticut's, and where a count is under 10 a model that borrows from neighboring counties replaces it. A county's rate here is the mean of its annual rates for 2019 to 2023, kept when CDC publishes a stable rate in each year. This comparison uses the same years for WONDER counts, so a county counts as published every year when its count is published in each year from 2019 to 2023.

Table 5 compares the panel's rate in counties that WONDER publishes in every year with the rate in counties where it withholds the count at least once, within a size of county. The second group includes counties that WONDER never publishes.

Among counties of 25,000 to 50,000 residents the overdose rate is 52.3 (95% CI 47.3 to 57.8) deaths per hundred thousand residents where WONDER publishes every year and 22.3 (95% CI 21.5 to 23.2) where it withholds the count at least once. Among counties of 50,000 to 100,000 residents the rates are 35.8 (95% CI 33.5 to 38.5) and 17.5 (95% CI 16.5 to 18.5), and among counties of 100,000 to 250,000 residents they are 27.6 (95% CI 26.2 to 29.2) and 11.3 (95% CI 10.2 to 12.6).

For suicide the rates are 30.9 (95% CI 27.5 to 34.9) and 19.4 (95% CI 19.0 to 19.8) among counties of 25,000 to 50,000 residents, 21.6 (95% CI 20.7 to 22.5) and 17.0 (95% CI 16.6 to 17.4) among counties of 50,000 to 100,000 residents, and 17.0 (95% CI 16.4 to 17.5) and 12.4 (95% CI 11.3 to 13.7) among counties of 100,000 to 250,000 residents.

A group needs at least 10 counties to be plotted, and few counties of the middle sizes are published in every year, which is itself a result of the rule.

Table 5. Among counties of 25,000 to 50,000 residents, the panel's overdose rate is 52.3 where WONDER publishes every year and 22.3 where it withholds the count in at least one year, and the order reverses under 10,000 residents

The order reverses in the smallest counties. A county under 10,000 residents almost never reaches 10 deaths, so it is published in every year only if its count is zero in every year. The panel's rate for those counties is 0.0 for overdose and 0.0 for suicide, against 13.0 (95% CI 11.7 to 14.3) and 14.2 (95% CI 13.2 to 15.4) where the count is withheld at least once. The rule selects on the outcome in both directions. A county of middling size has to have a high count in every year to be complete, and a very small county has to have none.

Part of each gap is a direct rate set against a model. In counties published every year the panel's overdose rates are modeled in 1.1% of county-years and its suicide rates in 0.0%, against 70.2% and 72.5% in counties where WONDER withholds the count at least once. A modeled rate borrows from neighbors, and the rates of small counties are noisy before any suppression, so a map of them can mislead on its own.[3] The comparison shows the direction and rough size of the selection. It does not give the true rate of any withheld county.

### The bias of a pooled rate follows from the rule

The rule alone fixes the sign of the bias, and the sign depends on how many deaths a county expects. Suppose a county has a constant true rate and its deaths in a year are a Poisson count, the standard model for deaths in a small population.[4][2] WONDER publishes the year when the count is zero or at least 10. Table 6 plots, for counties that share an expected number of deaths a year, the rate pooled over their published years divided by their true rate.

The ratio is far below one when a county expects only a few deaths, because nearly every published county-year is then a zero. It reaches one near 4.2 expected deaths. It peaks at 1.8 at 5.4 expected deaths, where only 5.3% of county-years are published and nearly all of them have a high count. Above 14.2 expected deaths it is within 5% of one.

Table 6. A rate pooled over published county-years is 1.8 times the true rate at 5.4 expected deaths a year, below it under 4.2 and within 5% of it above 14.2

The rate of a complete county is exact for that county, so complete counties escape the bias in their own rates. They do not escape it as a group. Under the same model the counties published in every year are a selected set whose pooled rate is the same multiple of the true rate as the rate pooled over published county-years, so the set overstates the typical county of its size.

Table 7 compares the model with data. For each size of county it divides two pooled WONDER rates for 2019 to 2023 by the panel's population-weighted rate for the same years in all counties of that size with a stable panel rate. One is the rate pooled over published county-years, and the other is the rate pooled over counties published every year. A county that WONDER never publishes adds nothing to the pooled WONDER rate and stays in the panel rate.

The data follow the sign pattern of the model. In counties under 10,000 residents the rate pooled over published county-years is 0.18 (95% CI 0.00 to 0.42) times the panel's rate for suicide and 0.05 (95% CI 0.00 to 0.13) times for overdose. Among counties of 10,000 to 25,000 residents the ratio is 0.78 (95% CI 0.50 to 1.07) for suicide and 1.63 (95% CI 1.45 to 1.77) for overdose. Overdose is above one and suicide below it, although the suicide interval reaches one, which fits the model because a county of that size expects more overdose deaths than suicides.

Among counties of 25,000 to 50,000 residents the ratios are 1.51 (95% CI 1.47 to 1.55) and 1.63 (95% CI 1.58 to 1.68), and the overdose rate pooled over counties published every year is 2.09 (95% CI 1.92 to 2.28) times the panel's. Among counties of 50,000 to 100,000 residents the ratios are 1.09 (95% CI 1.07 to 1.10) and 1.18 (95% CI 1.16 to 1.20), with 1.30 (95% CI 1.26 to 1.36) for overdose pooled over counties published every year.

Among counties of 100,000 to 250,000 residents the ratios are 1.00 (95% CI 1.00 to 1.01) and 1.07 (95% CI 1.06 to 1.08), with 1.09 (95% CI 1.07 to 1.10) for the complete counties. In counties of 250,000 or more residents, where every count is published, the overdose ratio is 1.05 (95% CI 1.05 to 1.06). That is the difference between the two definitions of overdose, since WONDER counts drug poisoning of any intent and the panel counts unintentional and undetermined-intent poisonings.

Table 7. A suicide rate pooled over published county-years is 0.18 times the panel's rate in counties under 10,000 residents and 1.51 times in counties of 25,000 to 50,000 residents

The data match the sign pattern and not the sizes of the model, which assumes one rate for every county of a given expected count. The panel is itself a model, so these ratios compare WONDER's rule with a modeled rate and not with the true rate. The panel's rates for counties that WONDER never publishes are modeled in every year. Restricting both sides to counties with at least one published county-year, the ratios among counties of 25,000 to 50,000 residents are 1.34 (95% CI 1.31 to 1.37) for suicide and 1.38 (95% CI 1.35 to 1.41) for overdose, lower than the ratios above but still over one.

### WONDER names only large counties in its birth products

The rule above acts on cells. A second rule acts on whole counties in WONDER's birth products. Natality names a county only if it has 100,000 or more residents, and linked birth and infant death only if it has 250,000 or more, as the documentation of the two products states. WONDER pools every other county in a state into one row labeled Unidentified Counties, so smaller counties have no cell at all, and there is nothing to suppress or to recover.

Natality lists 580 county identifiers, 578 of them with births, and those counties hold 79.6% of births in 2018 to 2023. It pools 2,564 counties, counting as pooled the identifiers with no births, which are described below. Linked birth and infant death lists 263 identifiers, 261 with births, which hold 64.0% of births in 2018 to 2022, and pools 2,881 counties.

A county's size class in this chapter is not the same as being named. 606 counties have a mean population of 100,000 or more, against 578 named in natality, and 277 have 250,000 or more, against 261 named in linked birth and infant death.

Cell suppression then applies inside the named counties. A count of births or infant deaths above zero and below 10 is withheld, as Chapter 10.1, Gradients, slopes and variance (https://vishalsingh.org/zipcode/gradients-slopes-and-variance) shows for birthweight classes in natality. Chapter 2.1, Birthweight and preterm birth (https://vishalsingh.org/zipcode/birthweight-and-preterm-birth) and Chapter 2.2, Infant mortality (https://vishalsingh.org/zipcode/infant-mortality) work with those files. The mortality county files differ. They name every county and rely on the cell rule alone, so a missing county there is a blank inside a named county and not a pooled row.

The line between named and pooled moves slightly. Some identifiers are listed with no births of their own in any year. In natality 2 identifiers are such counties (Cape May County NJ and Portsmouth city VA), and in linked birth and infant death 2 are (Hinds County MS and Mahoning County OH). WONDER shows their births as missing or zero in every year because it has moved them into the state's pooled row. A zero for these counties is not a count. The book's loaders set them to missing, which is why the balanced natality panel has 578 counties and not 580.

### What the book does instead

A withheld cell is counted and left out of the numerator and the denominator. It is never replaced by a value, a midpoint or the remainder of a total, because a total minus the published cells pins a withheld cell exactly wherever only one cell is withheld. Every WONDER chapter reports how many cells were withheld and the largest share of events they could hold, as Chapter 2.1, Birthweight and preterm birth (https://vishalsingh.org/zipcode/birthweight-and-preterm-birth), Chapter 2.2, Infant mortality (https://vishalsingh.org/zipcode/infant-mortality) and Chapter 8.2, Overdose, alcohol and suicide (https://vishalsingh.org/zipcode/overdose-alcohol-and-suicide) do. At the public limit of 9 deaths each, the withheld cells could hold at most 13.4% of the deaths in the county files for overdose, 29.1% for alcohol-induced deaths and 28.4% for suicide.

County death rates in those chapters use complete counties. For overdose, the 942 complete counties hold 83.8% of residents. The alcohol-induced and suicide counts give 644 counties holding 77.9% and 745 counties holding 82.0%. A rate for a complete county is exact for all its years, and the set of complete counties is selected on the outcome, so a map or gradient of complete counties says so. A map of WONDER counts needs at least 20 pooled deaths in a county to draw it.

Comparisons across all counties use a modeled source such as the CDC panel, which has no stable overdose rate in some year for 428 counties holding 2.3% of residents. National series are summed from the state files, which have no withheld cell. All 357 state-years of each kind of death are published, and the number of withheld state-years is 0.

County Health Rankings pools several years into one rate, which brings in more counties than a single county-year does. It still leaves a county without a rate when the window holds fewer than 10 deaths. That removes 716 counties for suicide and 1,244 for overdose, with 4.0% of residents. It removes 1,767 counties for homicide, with 9.7%. For premature death the floor is 20 deaths, and 55 counties have no rate.

The book does not model the withheld counts. One study treats them as censored data and estimates county rates with spatial Bayesian models, and it finds that ignoring the suppression biases inference.[5] That approach produces a value for a cell that WONDER withholds, and the modeled sources this book uses are published by CDC itself.

Table 8. In every source the counties left without a value are a larger share of counties than of residents, from 55 counties without a premature death rate to 2,881 that WONDER linked birth and infant death does not name

### What these data cannot show

The comparisons of complete counties with the rest use panel rates that are modeled for the counties WONDER withholds. They show the direction of the selection and cannot give the true rate of a withheld county. The two sources also define overdose differently. The panel counts unintentional and undetermined-intent drug poisonings, and WONDER counts drug poisoning of any intent, so levels differ a little by definition.

The calculation of the bias treats a county's deaths as Poisson counts with a constant rate and independent years. Real counts vary more than that, and rates move over time, so the curve gives direction and rough size and does not correct any estimate in this book. The sizes of county and their boundaries are choices, and the figures move with them.

Withheld counts cannot be recovered. A withheld cell is known only to lie above zero and below the limit, and the bound printed here is a ceiling set by the public limit and the published cells, and it says nothing about any one cell. The group comparisons describe counties and their residents, and County Health Rankings' shares of residents come from Census population estimates with that source's categories. Connecticut's counties are left out of every county comparison because WONDER has no population for them in 2022 and 2023. North Carolina's 2023 overdose counts predate the state's amended death certificates and are low, which enters the 2018 to 2023 window of the overdose counts. A pull that left out WONDER's option to show zero values would turn county-years with no deaths into absent rows, and the below-one ratios in the smallest counties depend on keeping them. The chapter covers three kinds of death. Kinds of death with more or fewer deaths would move the sizes of county at which the rule bites, which the Poisson calculation implies and these data do not test.

## Tables

### Table 1. In every year from 2018 to 2024 WONDER withheld between 44.6% and 59.9% of county-year death counts, in county-years that hold between 8.6% and 17.4% of residents

Share of county-years whose death count CDC WONDER withholds, and share of residents living in a withheld county-year, by year, for drug overdose, alcohol-induced causes and suicide, all counties in the county files

| cause | measure | n | value | year |
| --- | --- | ---: | ---: | ---: |
| Drug overdose | counties | 1,541 | 49.05 | 2018 |
| Drug overdose | residents | 1,541 | 12.31 | 2018 |
| Drug overdose | counties | 1,586 | 50.48 | 2019 |
| Drug overdose | residents | 1,586 | 12.32 | 2019 |
| Drug overdose | counties | 1,504 | 47.87 | 2020 |
| Drug overdose | residents | 1,504 | 10.09 | 2020 |
| Drug overdose | counties | 1,400 | 44.56 | 2021 |
| Drug overdose | residents | 1,400 | 8.79 | 2021 |
| Drug overdose | counties | 1,413 | 44.97 | 2022 |
| Drug overdose | residents | 1,413 | 8.56 | 2022 |
| Drug overdose | counties | 1,426 | 45.39 | 2023 |
| Drug overdose | residents | 1,426 | 8.64 | 2023 |
| Drug overdose | counties | 1,535 | 48.85 | 2024 |
| Drug overdose | residents | 1,535 | 10.78 | 2024 |
| Alcohol-induced | counties | 1,863 | 59.29 | 2018 |
| Alcohol-induced | residents | 1,863 | 17.42 | 2018 |
| Alcohol-induced | counties | 1,881 | 59.87 | 2019 |
| Alcohol-induced | residents | 1,881 | 16.91 | 2019 |
| Alcohol-induced | counties | 1,805 | 57.45 | 2020 |
| Alcohol-induced | residents | 1,805 | 14.26 | 2020 |
| Alcohol-induced | counties | 1,762 | 56.08 | 2021 |
| Alcohol-induced | residents | 1,762 | 12.52 | 2021 |
| Alcohol-induced | counties | 1,801 | 57.32 | 2022 |
| Alcohol-induced | residents | 1,801 | 13.68 | 2022 |
| Alcohol-induced | counties | 1,797 | 57.19 | 2023 |
| Alcohol-induced | residents | 1,797 | 14.10 | 2023 |
| Alcohol-induced | counties | 1,820 | 57.92 | 2024 |
| Alcohol-induced | residents | 1,820 | 13.96 | 2024 |
| Suicide | counties | 1,798 | 57.22 | 2018 |
| Suicide | residents | 1,798 | 12.26 | 2018 |
| Suicide | counties | 1,820 | 57.92 | 2019 |
| Suicide | residents | 1,820 | 12.42 | 2019 |
| Suicide | counties | 1,832 | 58.31 | 2020 |
| Suicide | residents | 1,832 | 12.67 | 2020 |
| Suicide | counties | 1,801 | 57.32 | 2021 |
| Suicide | residents | 1,801 | 11.90 | 2021 |
| Suicide | counties | 1,792 | 57.03 | 2022 |
| Suicide | residents | 1,792 | 12.14 | 2022 |
| Suicide | counties | 1,792 | 57.03 | 2023 |
| Suicide | residents | 1,792 | 11.97 | 2023 |
| Suicide | counties | 1,781 | 56.68 | 2024 |
| Suicide | residents | 1,781 | 11.81 | 2024 |

Views: Share of county-years (measure = counties); Share of residents (measure = residents).

Note: These are full counts of the 3,142 counties in WONDER's county files, so the lines carry no sampling interval. A county-year is withheld when its count is above zero and below 10; a count of zero is published.

Note: Residents come from the county population that WONDER publishes with every cell, withheld or not. Connecticut's 8 counties have no WONDER population in 2022 to 2024 and are left out of the resident shares in those years.

Note: Rates are not plotted. Drug overdose is drug poisoning of any intent, so an overdose recorded as a suicide is in both the drug and the suicide counts. North Carolina's 2023 overdose counts in these files predate the state's amended death certificates and are low.

Source: CDC WONDER underlying cause of death (single race), county files, 2018-2024; full counts

Data: https://vishalsingh.org/zipcode/report/suppressed-counties-in-death-data/figures/withheld_by_year.json

### Table 2. WONDER withholds 82.3% of overdose county-years in counties of 10,000 to 25,000 residents and 45.2% in smaller counties, where 54.7% are published as zero

Share of county-years withheld, published as zero and published with at least 10 deaths, by size of county (mean residents 2018 to 2023), drug overdose, alcohol-induced causes and suicide

| cause | counties | county_years | segment | share | size |
| --- | ---: | ---: | --- | ---: | --- |
| Drug overdose | 729 | 4,374 | Withheld | 45.18 | Under 10,000 |
| Drug overdose | 729 | 4,374 | Published, zero deaths | 54.66 | Under 10,000 |
| Drug overdose | 729 | 4,374 | Published, 10 or more deaths | 0.160 | Under 10,000 |
| Drug overdose | 799 | 4,794 | Withheld | 82.29 | 10,000 to 24,999 |
| Drug overdose | 799 | 4,794 | Published, zero deaths | 11.22 | 10,000 to 24,999 |
| Drug overdose | 799 | 4,794 | Published, 10 or more deaths | 6.49 | 10,000 to 24,999 |
| Drug overdose | 619 | 3,714 | Withheld | 64.38 | 25,000 to 49,999 |
| Drug overdose | 619 | 3,714 | Published, zero deaths | 1.72 | 25,000 to 49,999 |
| Drug overdose | 619 | 3,714 | Published, 10 or more deaths | 33.90 | 25,000 to 49,999 |
| Drug overdose | 389 | 2,334 | Withheld | 21.77 | 50,000 to 99,999 |
| Drug overdose | 389 | 2,334 | Published, zero deaths | 0.090 | 50,000 to 99,999 |
| Drug overdose | 389 | 2,334 | Published, 10 or more deaths | 78.15 | 50,000 to 99,999 |
| Drug overdose | 329 | 1,974 | Withheld | 2.53 | 100,000 to 249,999 |
| Drug overdose | 329 | 1,974 | Published, zero deaths | 0 | 100,000 to 249,999 |
| Drug overdose | 329 | 1,974 | Published, 10 or more deaths | 97.47 | 100,000 to 249,999 |
| Drug overdose | 277 | 1,662 | Withheld | 0 | 250,000 or more |
| Drug overdose | 277 | 1,662 | Published, zero deaths | 0 | 250,000 or more |
| Drug overdose | 277 | 1,662 | Published, 10 or more deaths | 100 | 250,000 or more |
| Alcohol-induced | 729 | 4,374 | Withheld | 54.94 | Under 10,000 |
| Alcohol-induced | 729 | 4,374 | Published, zero deaths | 44.79 | Under 10,000 |
| Alcohol-induced | 729 | 4,374 | Published, 10 or more deaths | 0.270 | Under 10,000 |
| Alcohol-induced | 799 | 4,794 | Withheld | 84.27 | 10,000 to 24,999 |
| Alcohol-induced | 799 | 4,794 | Published, zero deaths | 14.12 | 10,000 to 24,999 |
| Alcohol-induced | 799 | 4,794 | Published, 10 or more deaths | 1.61 | 10,000 to 24,999 |
| Alcohol-induced | 619 | 3,714 | Withheld | 84.17 | 25,000 to 49,999 |
| Alcohol-induced | 619 | 3,714 | Published, zero deaths | 1.99 | 25,000 to 49,999 |
| Alcohol-induced | 619 | 3,714 | Published, 10 or more deaths | 13.84 | 25,000 to 49,999 |
| Alcohol-induced | 389 | 2,334 | Withheld | 49.14 | 50,000 to 99,999 |
| Alcohol-induced | 389 | 2,334 | Published, zero deaths | 0.390 | 50,000 to 99,999 |
| Alcohol-induced | 389 | 2,334 | Published, 10 or more deaths | 50.47 | 50,000 to 99,999 |
| Alcohol-induced | 329 | 1,974 | Withheld | 9.73 | 100,000 to 249,999 |
| Alcohol-induced | 329 | 1,974 | Published, zero deaths | 0 | 100,000 to 249,999 |
| Alcohol-induced | 329 | 1,974 | Published, 10 or more deaths | 90.27 | 100,000 to 249,999 |
| Alcohol-induced | 277 | 1,662 | Withheld | 0.060 | 250,000 or more |
| Alcohol-induced | 277 | 1,662 | Published, zero deaths | 0 | 250,000 or more |
| Alcohol-induced | 277 | 1,662 | Published, 10 or more deaths | 99.94 | 250,000 or more |
| Suicide | 729 | 4,374 | Withheld | 61.34 | Under 10,000 |
| Suicide | 729 | 4,374 | Published, zero deaths | 38.45 | Under 10,000 |
| Suicide | 729 | 4,374 | Published, 10 or more deaths | 0.210 | Under 10,000 |
| Suicide | 799 | 4,794 | Withheld | 93.12 | 10,000 to 24,999 |
| Suicide | 799 | 4,794 | Published, zero deaths | 5.74 | 10,000 to 24,999 |
| Suicide | 799 | 4,794 | Published, 10 or more deaths | 1.15 | 10,000 to 24,999 |
| Suicide | 619 | 3,714 | Withheld | 80.67 | 25,000 to 49,999 |
| Suicide | 619 | 3,714 | Published, zero deaths | 0.590 | 25,000 to 49,999 |
| Suicide | 619 | 3,714 | Published, 10 or more deaths | 18.74 | 25,000 to 49,999 |
| Suicide | 389 | 2,334 | Withheld | 28.36 | 50,000 to 99,999 |
| Suicide | 389 | 2,334 | Published, zero deaths | 0.040 | 50,000 to 99,999 |
| Suicide | 389 | 2,334 | Published, 10 or more deaths | 71.59 | 50,000 to 99,999 |
| Suicide | 329 | 1,974 | Withheld | 1.52 | 100,000 to 249,999 |
| Suicide | 329 | 1,974 | Published, zero deaths | 0 | 100,000 to 249,999 |
| Suicide | 329 | 1,974 | Published, 10 or more deaths | 98.48 | 100,000 to 249,999 |
| Suicide | 277 | 1,662 | Withheld | 0 | 250,000 or more |
| Suicide | 277 | 1,662 | Published, zero deaths | 0 | 250,000 or more |
| Suicide | 277 | 1,662 | Published, 10 or more deaths | 100 | 250,000 or more |

Views: Drug overdose (cause = Drug overdose); Alcohol-induced (cause = Alcohol-induced); Suicide (cause = Suicide).

Note: Size is a county's mean WONDER population over 2018 to 2023. The boundaries are 10,000, 25,000, 50,000, 100,000 and 250,000 residents; the last two are the populations at which WONDER names a county in its birth products, although a county's class here and its being named do not coincide exactly.

Note: These are shares of all county-years in each class, including Connecticut's 8 counties, so the bars are full counts with no interval. A published zero is a real zero; a withheld count is missing.

Source: CDC WONDER underlying cause of death (single race), county files, 2018-2023; full counts

Data: https://vishalsingh.org/zipcode/report/suppressed-counties-in-death-data/figures/withheld_by_size.json

### Table 3. WONDER withheld the overdose count in every year from 2018 to 2023 in 591 counties and in no year in 950

Number of years from 2018 to 2023 in which CDC WONDER withholds a county's death count, by county, for drug overdose, alcohol-induced causes and suicide

No inline rows: the values of Table 3 are in https://vishalsingh.org/zipcode/data/suppressed-counties-in-death-data/years_withheld_od.json.

Views: Drug overdose (every row); Alcohol-induced (every row); Suicide (every row).

Note: A count is withheld when it is above zero and below 10, and the key counts the years it is withheld out of the years from 2018 to 2023. A county with no withheld year can still have a count of zero in some years.

Note: The overdose view shades 950 counties at no withheld year: the 942 complete counties and Connecticut's 8, which have deaths but no population in 2022 and 2023.

Note: WONDER codes Valdez-Cordova AK as one county, so its counts are drawn on both successor areas, Chugach and Copper River; Kusilvak AK and Oglala Lakota SD are drawn under their current codes. Connecticut's 8 counties are drawn from their legacy codes, and WONDER has no population for them in 2022 and 2023. Puerto Rico is not drawn.

Source: CDC WONDER underlying cause of death (single race), county files, 2018-2023; full counts

Data: https://vishalsingh.org/zipcode/report/suppressed-counties-in-death-data/figures/years_withheld_map.json

### Table 4. Counties whose overdose count WONDER publishes in every year hold 83.8% of residents, and counties withheld in every year hold 4.1% and are smaller, more rural and more deprived

Counties by how many years from 2018 to 2023 WONDER withholds the death count: share of counties and residents, median size, nonmetro residents, Area Deprivation Index rank and race and ethnicity of residents, other than Connecticut's counties

| cause | estimate | group | measure | n_units |
| --- | ---: | --- | --- | ---: |
| Drug overdose | 30.06 | Published every year | Share of counties (%) | 942 |
| Drug overdose | 83.81 | Published every year | Share of residents (%) | 942 |
| Drug overdose | 134.4 | Published every year | Median county residents (thousands) | 942 |
| Drug overdose | 4.16 | Published every year | Residents in nonmetro counties (%) | 942 |
| Drug overdose | 43.04 | Published every year | Mean Area Deprivation Index rank (1 least, 100 most deprived) | 942 |
| Drug overdose | 20.97 | Published every year | Residents who are Hispanic (%) | 942 |
| Drug overdose | 13.37 | Published every year | Residents who are non-Hispanic Black (%) | 942 |
| Drug overdose | 55.34 | Published every year | Residents who are non-Hispanic white (%) | 942 |
| Drug overdose | 1.18 | Published every year | Residents who are American Indian or Alaska Native (%) | 942 |
| Drug overdose | 51.08 | Withheld in some years | Share of counties (%) | 1,601 |
| Drug overdose | 12.09 | Withheld in some years | Share of residents (%) | 1,601 |
| Drug overdose | 16.08 | Withheld in some years | Median county residents (thousands) | 1,601 |
| Drug overdose | 61.06 | Withheld in some years | Residents in nonmetro counties (%) | 1,601 |
| Drug overdose | 67.71 | Withheld in some years | Mean Area Deprivation Index rank (1 least, 100 most deprived) | 1,601 |
| Drug overdose | 9.66 | Withheld in some years | Residents who are Hispanic (%) | 1,601 |
| Drug overdose | 9.38 | Withheld in some years | Residents who are non-Hispanic Black (%) | 1,601 |
| Drug overdose | 75.91 | Withheld in some years | Residents who are non-Hispanic white (%) | 1,601 |
| Drug overdose | 2.12 | Withheld in some years | Residents who are American Indian or Alaska Native (%) | 1,601 |
| Drug overdose | 18.86 | Withheld in every year | Share of counties (%) | 591 |
| Drug overdose | 4.10 | Withheld in every year | Share of residents (%) | 591 |
| Drug overdose | 19.76 | Withheld in every year | Median county residents (thousands) | 591 |
| Drug overdose | 75.52 | Withheld in every year | Residents in nonmetro counties (%) | 591 |
| Drug overdose | 70.17 | Withheld in every year | Mean Area Deprivation Index rank (1 least, 100 most deprived) | 591 |
| Drug overdose | 10.00 | Withheld in every year | Residents who are Hispanic (%) | 591 |
| Drug overdose | 8.11 | Withheld in every year | Residents who are non-Hispanic Black (%) | 591 |
| Drug overdose | 77.24 | Withheld in every year | Residents who are non-Hispanic white (%) | 591 |
| Drug overdose | 2.13 | Withheld in every year | Residents who are American Indian or Alaska Native (%) | 591 |
| Alcohol-induced | 20.55 | Published every year | Share of counties (%) | 644 |
| Alcohol-induced | 77.87 | Published every year | Share of residents (%) | 644 |
| Alcohol-induced | 207.1 | Published every year | Median county residents (thousands) | 644 |
| Alcohol-induced | 2.07 | Published every year | Residents in nonmetro counties (%) | 644 |
| Alcohol-induced | 41.45 | Published every year | Mean Area Deprivation Index rank (1 least, 100 most deprived) | 644 |
| Alcohol-induced | 22.08 | Published every year | Residents who are Hispanic (%) | 644 |
| Alcohol-induced | 13.34 | Published every year | Residents who are non-Hispanic Black (%) | 644 |
| Alcohol-induced | 53.77 | Published every year | Residents who are non-Hispanic white (%) | 644 |
| Alcohol-induced | 1.26 | Published every year | Residents who are American Indian or Alaska Native (%) | 644 |
| Alcohol-induced | 54.50 | Withheld in some years | Share of counties (%) | 1,708 |
| Alcohol-induced | 15.76 | Withheld in some years | Share of residents (%) | 1,708 |
| Alcohol-induced | 15.46 | Withheld in some years | Median county residents (thousands) | 1,708 |
| Alcohol-induced | 50.04 | Withheld in some years | Residents in nonmetro counties (%) | 1,708 |
| Alcohol-induced | 65.48 | Withheld in some years | Mean Area Deprivation Index rank (1 least, 100 most deprived) | 1,708 |
| Alcohol-induced | 8.85 | Withheld in some years | Residents who are Hispanic (%) | 1,708 |
| Alcohol-induced | 10.62 | Withheld in some years | Residents who are non-Hispanic Black (%) | 1,708 |
| Alcohol-induced | 75.66 | Withheld in some years | Residents who are non-Hispanic white (%) | 1,708 |
| Alcohol-induced | 1.54 | Withheld in some years | Residents who are American Indian or Alaska Native (%) | 1,708 |
| Alcohol-induced | 24.95 | Withheld in every year | Share of counties (%) | 782 |
| Alcohol-induced | 6.37 | Withheld in every year | Share of residents (%) | 782 |
| Alcohol-induced | 23.70 | Withheld in every year | Median county residents (thousands) | 782 |
| Alcohol-induced | 70.16 | Withheld in every year | Residents in nonmetro counties (%) | 782 |
| Alcohol-induced | 71.23 | Withheld in every year | Mean Area Deprivation Index rank (1 least, 100 most deprived) | 782 |
| Alcohol-induced | 8.84 | Withheld in every year | Residents who are Hispanic (%) | 782 |
| Alcohol-induced | 9.60 | Withheld in every year | Residents who are non-Hispanic Black (%) | 782 |
| Alcohol-induced | 77.47 | Withheld in every year | Residents who are non-Hispanic white (%) | 782 |
| Alcohol-induced | 1.62 | Withheld in every year | Residents who are American Indian or Alaska Native (%) | 782 |
| Suicide | 23.77 | Published every year | Share of counties (%) | 745 |
| Suicide | 81.97 | Published every year | Share of residents (%) | 745 |
| Suicide | 176.5 | Published every year | Median county residents (thousands) | 745 |
| Suicide | 2.89 | Published every year | Residents in nonmetro counties (%) | 745 |
| Suicide | 42.37 | Published every year | Mean Area Deprivation Index rank (1 least, 100 most deprived) | 745 |
| Suicide | 21.38 | Published every year | Residents who are Hispanic (%) | 745 |
| Suicide | 13.36 | Published every year | Residents who are non-Hispanic Black (%) | 745 |
| Suicide | 54.77 | Published every year | Residents who are non-Hispanic white (%) | 745 |
| Suicide | 1.22 | Published every year | Residents who are American Indian or Alaska Native (%) | 745 |
| Suicide | 46.90 | Withheld in some years | Share of counties (%) | 1,470 |
| Suicide | 12.23 | Withheld in some years | Share of residents (%) | 1,470 |
| Suicide | 16.53 | Withheld in some years | Median county residents (thousands) | 1,470 |
| Suicide | 60.19 | Withheld in some years | Residents in nonmetro counties (%) | 1,470 |
| Suicide | 67.54 | Withheld in some years | Mean Area Deprivation Index rank (1 least, 100 most deprived) | 1,470 |
| Suicide | 9.02 | Withheld in some years | Residents who are Hispanic (%) | 1,470 |
| Suicide | 9.58 | Withheld in some years | Residents who are non-Hispanic Black (%) | 1,470 |
| Suicide | 76.38 | Withheld in some years | Residents who are non-Hispanic white (%) | 1,470 |
| Suicide | 1.89 | Withheld in some years | Residents who are American Indian or Alaska Native (%) | 1,470 |
| Suicide | 29.32 | Withheld in every year | Share of counties (%) | 919 |
| Suicide | 5.80 | Withheld in every year | Share of residents (%) | 919 |
| Suicide | 18.89 | Withheld in every year | Median county residents (thousands) | 919 |
| Suicide | 73.05 | Withheld in every year | Residents in nonmetro counties (%) | 919 |
| Suicide | 71.44 | Withheld in every year | Mean Area Deprivation Index rank (1 least, 100 most deprived) | 919 |
| Suicide | 9.12 | Withheld in every year | Residents who are Hispanic (%) | 919 |
| Suicide | 9.38 | Withheld in every year | Residents who are non-Hispanic Black (%) | 919 |
| Suicide | 77.41 | Withheld in every year | Residents who are non-Hispanic white (%) | 919 |
| Suicide | 1.72 | Withheld in every year | Residents who are American Indian or Alaska Native (%) | 919 |

Views: Drug overdose (cause = Drug overdose); Alcohol-induced (cause = Alcohol-induced); Suicide (cause = Suicide).

Note: Shares of counties and residents and the median are full counts of the 3,134 counties other than Connecticut's 8 and carry no interval. Residents are the mean WONDER population over 2018 to 2023. The other panels are population-weighted means across counties, so an interval describes the spread of county values.

Note: Deprivation is the mean of the Area Deprivation Index national ranks of a county's tracts, weighted by 2019-2023 tract population, and the index understates deprivation in high-cost counties. Nonmetro counties have a Rural-Urban Continuum Code of 4 or higher.

Note: Race and ethnicity shares come from County Health Rankings 2024 (Census Population Estimates) and use that source's categories: Hispanic of any race, and non-Hispanic Black and white. They describe a county's residents, not the people who died.

Source: CDC WONDER underlying cause of death, county files, 2018-2023, Area Deprivation Index 2023, Rural-Urban Continuum Codes 2023 and County Health Rankings 2024; full counts and population-weighted means

Data: https://vishalsingh.org/zipcode/report/suppressed-counties-in-death-data/figures/who_lives_in_withheld.json

### Table 5. Among counties of 25,000 to 50,000 residents, the panel's overdose rate is 52.3 where WONDER publishes every year and 22.3 where it withholds the count in at least one year, and the order reverses under 10,000 residents

CDC modeled death rate per 100,000 residents, mean of annual rates for 2019 to 2023, in counties whose WONDER count is published in each year 2019 to 2023 and in counties where it is withheld in at least one of those years, including counties never published, by size of county, population-weighted

| cause | estimate | n_units | size | status |
| --- | ---: | ---: | --- | --- |
| Drug overdose | 0 | 128 | Under 10,000 | Published in every year |
| Drug overdose | 12.97 | 326 | Under 10,000 | Withheld in at least one year |
| Drug overdose | 78.76 | 11 | 10,000 to 24,999 | Published in every year |
| Drug overdose | 20.98 | 662 | 10,000 to 24,999 | Withheld in at least one year |
| Drug overdose | 52.28 | 66 | 25,000 to 49,999 | Published in every year |
| Drug overdose | 22.33 | 536 | 25,000 to 49,999 | Withheld in at least one year |
| Drug overdose | 35.83 | 227 | 50,000 to 99,999 | Published in every year |
| Drug overdose | 17.54 | 160 | 50,000 to 99,999 | Withheld in at least one year |
| Drug overdose | 27.63 | 306 | 100,000 to 249,999 | Published in every year |
| Drug overdose | 11.26 | 19 | 100,000 to 249,999 | Withheld in at least one year |
| Drug overdose | 28.41 | 273 | 250,000 or more | Published in every year |
| Suicide | 0 | 41 | Under 10,000 | Published in every year |
| Suicide | 14.23 | 365 | Under 10,000 | Withheld in at least one year |
| Suicide | 18.27 | 406 | 10,000 to 24,999 | Withheld in at least one year |
| Suicide | 30.91 | 12 | 25,000 to 49,999 | Published in every year |
| Suicide | 19.41 | 472 | 25,000 to 49,999 | Withheld in at least one year |
| Suicide | 21.57 | 145 | 50,000 to 99,999 | Published in every year |
| Suicide | 17.03 | 234 | 50,000 to 99,999 | Withheld in at least one year |
| Suicide | 17.01 | 304 | 100,000 to 249,999 | Published in every year |
| Suicide | 12.40 | 21 | 100,000 to 249,999 | Withheld in at least one year |
| Suicide | 12.54 | 273 | 250,000 or more | Published in every year |

Views: Drug overdose (cause = Drug overdose); Suicide (cause = Suicide).

Note: The comparison uses CDC's injury and violence panel because its rates do not depend on WONDER's publication rule. They are modeled estimates, built partly from neighboring counties where deaths are few. Where WONDER withholds a county-year, 70.2% of the panel's overdose county-years are modeled, against 1.1% in counties published every year, so part of each gap is a direct rate against a model.

Note: A group of fewer than 10 counties is not plotted, and an interval is absent where every county has a rate of zero. Overdose in the panel is unintentional or undetermined-intent drug overdose. Connecticut is not in the panel. WONDER counts and panel rates both cover 2019 to 2023, and the group withheld in at least one year includes counties that WONDER never publishes in those years.

Note: In the smallest counties, counties published every year are the counties with no deaths in any year, because a count of 10 is almost out of reach there.

Source: CDC WONDER underlying cause of death, county files, 2019-2023, and CDC Mapping Injury, Overdose, and Violence county file, 2019-2023; crude rates, modeled estimates, population-weighted

Data: https://vishalsingh.org/zipcode/report/suppressed-counties-in-death-data/figures/panel_rate_by_status.json

### Table 6. A rate pooled over published county-years is 1.8 times the true rate at 5.4 expected deaths a year, below it under 4.2 and within 5% of it above 14.2

Expected pooled rate over the county-years WONDER publishes (counts of zero and of at least 10) divided by the true rate, against the expected number of deaths in a county-year, Poisson counts and a constant rate

| lam | published | ratio |
| ---: | ---: | ---: |
| 0.100 | 90.48 | 0 |
| 0.200 | 81.87 | 0 |
| 0.300 | 74.08 | 0 |
| 0.400 | 67.03 | 0 |
| 0.500 | 60.65 | 0 |
| 0.600 | 54.88 | 0 |
| 0.700 | 49.66 | 0 |
| 0.800 | 44.93 | 0 |
| 0.900 | 40.66 | 0 |
| 1 | 36.79 | 0 |
| 1.10 | 33.29 | 0 |
| 1.20 | 30.12 | 0 |
| 1.30 | 27.25 | 0 |
| 1.40 | 24.66 | 0.000 |
| 1.50 | 22.31 | 0.000 |
| 1.60 | 20.19 | 0.000 |
| 1.70 | 18.27 | 0.000 |
| 1.80 | 16.53 | 0.001 |
| 1.90 | 14.96 | 0.001 |
| 2 | 13.54 | 0.002 |
| 2.10 | 12.25 | 0.003 |
| 2.20 | 11.09 | 0.004 |
| 2.30 | 10.04 | 0.006 |
| 2.40 | 9.09 | 0.009 |
| 2.50 | 8.24 | 0.014 |
| 2.60 | 7.46 | 0.020 |
| 2.70 | 6.77 | 0.028 |
| 2.80 | 6.15 | 0.040 |
| 2.90 | 5.59 | 0.055 |
| 3 | 5.09 | 0.075 |
| 3.10 | 4.64 | 0.101 |
| 3.20 | 4.25 | 0.134 |
| 3.30 | 3.91 | 0.177 |
| 3.40 | 3.61 | 0.230 |
| 3.50 | 3.35 | 0.295 |
| 3.60 | 3.13 | 0.372 |
| 3.70 | 2.96 | 0.463 |
| 3.80 | 2.82 | 0.567 |
| 3.90 | 2.71 | 0.683 |
| 4 | 2.65 | 0.808 |
| 4.10 | 2.61 | 0.938 |
| 4.20 | 2.61 | 1.07 |
| 4.30 | 2.65 | 1.20 |
| 4.40 | 2.72 | 1.32 |
| 4.50 | 2.82 | 1.43 |
| 4.60 | 2.96 | 1.52 |
| 4.70 | 3.13 | 1.61 |
| 4.80 | 3.34 | 1.67 |
| 4.90 | 3.58 | 1.73 |
| 5 | 3.86 | 1.77 |
| 5.10 | 4.17 | 1.79 |
| 5.20 | 4.52 | 1.81 |
| 5.30 | 4.91 | 1.82 |
| 5.40 | 5.33 | 1.83 |
| 5.50 | 5.79 | 1.83 |
| 5.60 | 6.28 | 1.82 |
| 5.70 | 6.82 | 1.81 |
| 5.80 | 7.39 | 1.80 |
| 5.90 | 8.00 | 1.78 |
| 6 | 8.64 | 1.77 |
| 6.10 | 9.32 | 1.75 |
| 6.20 | 10.04 | 1.73 |
| 6.30 | 10.80 | 1.72 |
| 6.40 | 11.59 | 1.70 |
| 6.50 | 12.41 | 1.68 |
| 6.60 | 13.27 | 1.66 |
| 6.70 | 14.17 | 1.64 |
| 6.80 | 15.09 | 1.62 |
| 6.90 | 16.05 | 1.61 |
| 7 | 17.04 | 1.59 |
| 7.10 | 18.06 | 1.57 |
| 7.20 | 19.11 | 1.56 |
| 7.30 | 20.19 | 1.54 |
| 7.40 | 21.29 | 1.52 |
| 7.50 | 22.41 | 1.51 |
| 7.60 | 23.57 | 1.49 |
| 7.70 | 24.74 | 1.48 |
| 7.80 | 25.93 | 1.46 |
| 7.90 | 27.14 | 1.45 |
| 8 | 28.37 | 1.44 |
| 8.10 | 29.62 | 1.42 |
| 8.20 | 30.88 | 1.41 |
| 8.30 | 32.15 | 1.40 |
| 8.40 | 33.43 | 1.39 |
| 8.50 | 34.72 | 1.37 |
| 8.60 | 36.02 | 1.36 |
| 8.70 | 37.33 | 1.35 |
| 8.80 | 38.64 | 1.34 |
| 8.90 | 39.96 | 1.33 |
| 9 | 41.27 | 1.32 |
| 9.10 | 42.59 | 1.31 |
| 9.20 | 43.90 | 1.30 |
| 9.30 | 45.22 | 1.29 |
| 9.40 | 46.52 | 1.28 |
| 9.50 | 47.83 | 1.27 |
| 9.60 | 49.12 | 1.26 |
| 9.70 | 50.41 | 1.25 |
| 9.80 | 51.69 | 1.25 |
| 9.90 | 52.95 | 1.24 |
| 10 | 54.21 | 1.23 |
| 10.10 | 55.46 | 1.22 |
| 10.20 | 56.69 | 1.22 |
| 10.30 | 57.90 | 1.21 |
| 10.40 | 59.10 | 1.20 |
| 10.50 | 60.29 | 1.20 |
| 10.60 | 61.46 | 1.19 |
| 10.70 | 62.61 | 1.18 |
| 10.80 | 63.74 | 1.18 |
| 10.90 | 64.86 | 1.17 |
| 11 | 65.95 | 1.16 |
| 11.10 | 67.03 | 1.16 |
| 11.20 | 68.08 | 1.15 |
| 11.30 | 69.11 | 1.15 |
| 11.40 | 70.13 | 1.14 |
| 11.50 | 71.12 | 1.14 |
| 11.60 | 72.09 | 1.13 |
| 11.70 | 73.04 | 1.13 |
| 11.80 | 73.97 | 1.12 |
| 11.90 | 74.88 | 1.12 |
| 12 | 75.76 | 1.12 |
| 12.10 | 76.62 | 1.11 |
| 12.20 | 77.47 | 1.11 |
| 12.30 | 78.28 | 1.10 |
| 12.40 | 79.08 | 1.10 |
| 12.50 | 79.86 | 1.10 |
| 12.60 | 80.61 | 1.09 |
| 12.70 | 81.34 | 1.09 |
| 12.80 | 82.06 | 1.09 |
| 12.90 | 82.75 | 1.08 |
| 13 | 83.42 | 1.08 |
| 13.10 | 84.07 | 1.08 |
| 13.20 | 84.70 | 1.07 |
| 13.30 | 85.31 | 1.07 |
| 13.40 | 85.90 | 1.07 |
| 13.50 | 86.47 | 1.07 |
| 13.60 | 87.03 | 1.06 |
| 13.70 | 87.56 | 1.06 |
| 13.80 | 88.08 | 1.06 |
| 13.90 | 88.58 | 1.06 |
| 14 | 89.06 | 1.05 |
| 14.10 | 89.53 | 1.05 |
| 14.20 | 89.97 | 1.05 |
| 14.30 | 90.41 | 1.05 |
| 14.40 | 90.82 | 1.04 |
| 14.50 | 91.22 | 1.04 |
| 14.60 | 91.61 | 1.04 |
| 14.70 | 91.98 | 1.04 |
| 14.80 | 92.34 | 1.04 |
| 14.90 | 92.68 | 1.04 |
| 15 | 93.02 | 1.03 |
| 15.10 | 93.33 | 1.03 |
| 15.20 | 93.64 | 1.03 |
| 15.30 | 93.93 | 1.03 |
| 15.40 | 94.21 | 1.03 |
| 15.50 | 94.48 | 1.03 |
| 15.60 | 94.74 | 1.03 |
| 15.70 | 94.99 | 1.03 |
| 15.80 | 95.22 | 1.02 |
| 15.90 | 95.45 | 1.02 |
| 16 | 95.67 | 1.02 |
| 16.10 | 95.88 | 1.02 |
| 16.20 | 96.08 | 1.02 |
| 16.30 | 96.27 | 1.02 |
| 16.40 | 96.45 | 1.02 |
| 16.50 | 96.63 | 1.02 |
| 16.60 | 96.79 | 1.02 |
| 16.70 | 96.95 | 1.02 |
| 16.80 | 97.10 | 1.02 |
| 16.90 | 97.25 | 1.01 |
| 17 | 97.39 | 1.01 |
| 17.10 | 97.52 | 1.01 |
| 17.20 | 97.65 | 1.01 |
| 17.30 | 97.77 | 1.01 |
| 17.40 | 97.88 | 1.01 |
| 17.50 | 97.99 | 1.01 |
| 17.60 | 98.09 | 1.01 |
| 17.70 | 98.19 | 1.01 |
| 17.80 | 98.29 | 1.01 |
| 17.90 | 98.38 | 1.01 |
| 18 | 98.46 | 1.01 |
| 18.10 | 98.54 | 1.01 |
| 18.20 | 98.62 | 1.01 |
| 18.30 | 98.69 | 1.01 |
| 18.40 | 98.76 | 1.01 |
| 18.50 | 98.83 | 1.01 |
| 18.60 | 98.89 | 1.01 |
| 18.70 | 98.95 | 1.01 |
| 18.80 | 99.01 | 1.01 |
| 18.90 | 99.06 | 1.01 |
| 19 | 99.11 | 1.00 |
| 19.10 | 99.16 | 1.00 |
| 19.20 | 99.21 | 1.00 |
| 19.30 | 99.25 | 1.00 |
| 19.40 | 99.29 | 1.00 |
| 19.50 | 99.33 | 1.00 |
| 19.60 | 99.37 | 1.00 |
| 19.70 | 99.41 | 1.00 |
| 19.80 | 99.44 | 1.00 |
| 19.90 | 99.47 | 1.00 |
| 20 | 99.50 | 1.00 |
| 20.10 | 99.53 | 1.00 |
| 20.20 | 99.56 | 1.00 |
| 20.30 | 99.58 | 1.00 |
| 20.40 | 99.61 | 1.00 |
| 20.50 | 99.63 | 1.00 |
| 20.60 | 99.65 | 1.00 |
| 20.70 | 99.67 | 1.00 |
| 20.80 | 99.69 | 1.00 |
| 20.90 | 99.71 | 1.00 |
| 21 | 99.72 | 1.00 |
| 21.10 | 99.74 | 1.00 |
| 21.20 | 99.75 | 1.00 |
| 21.30 | 99.77 | 1.00 |
| 21.40 | 99.78 | 1.00 |
| 21.50 | 99.80 | 1.00 |
| 21.60 | 99.81 | 1.00 |
| 21.70 | 99.82 | 1.00 |
| 21.80 | 99.83 | 1.00 |
| 21.90 | 99.84 | 1.00 |
| 22 | 99.85 | 1.00 |
| 22.10 | 99.86 | 1.00 |
| 22.20 | 99.87 | 1.00 |
| 22.30 | 99.88 | 1.00 |
| 22.40 | 99.88 | 1.00 |
| 22.50 | 99.89 | 1.00 |
| 22.60 | 99.90 | 1.00 |
| 22.70 | 99.90 | 1.00 |
| 22.80 | 99.91 | 1.00 |
| 22.90 | 99.91 | 1.00 |
| 23 | 99.92 | 1.00 |
| 23.10 | 99.92 | 1.00 |
| 23.20 | 99.93 | 1.00 |
| 23.30 | 99.93 | 1.00 |
| 23.40 | 99.94 | 1.00 |
| 23.50 | 99.94 | 1.00 |
| 23.60 | 99.94 | 1.00 |
| 23.70 | 99.95 | 1.00 |
| 23.80 | 99.95 | 1.00 |
| 23.90 | 99.95 | 1.00 |
| 24 | 99.96 | 1.00 |
| 24.10 | 99.96 | 1.00 |
| 24.20 | 99.96 | 1.00 |
| 24.30 | 99.97 | 1.00 |
| 24.40 | 99.97 | 1.00 |
| 24.50 | 99.97 | 1.00 |
| 24.60 | 99.97 | 1.00 |
| 24.70 | 99.97 | 1.00 |
| 24.80 | 99.97 | 1.00 |
| 24.90 | 99.98 | 1.00 |
| 25 | 99.98 | 1.00 |
| 25.10 | 99.98 | 1.00 |
| 25.20 | 99.98 | 1.00 |
| 25.30 | 99.98 | 1.00 |
| 25.40 | 99.98 | 1.00 |
| 25.50 | 99.98 | 1.00 |
| 25.60 | 99.98 | 1.00 |
| 25.70 | 99.99 | 1.00 |
| 25.80 | 99.99 | 1.00 |
| 25.90 | 99.99 | 1.00 |
| 26 | 99.99 | 1.00 |
| 26.10 | 99.99 | 1.00 |
| 26.20 | 99.99 | 1.00 |
| 26.30 | 99.99 | 1.00 |
| 26.40 | 99.99 | 1.00 |
| 26.50 | 99.99 | 1.00 |
| 26.60 | 99.99 | 1.00 |
| 26.70 | 99.99 | 1 |
| 26.80 | 99.99 | 1 |
| 26.90 | 99.99 | 1 |
| 27 | 99.99 | 1 |
| 27.10 | 100.00 | 1 |
| 27.20 | 100.00 | 1 |
| 27.30 | 100.00 | 1 |
| 27.40 | 100.00 | 1 |
| 27.50 | 100.00 | 1 |
| 27.60 | 100.00 | 1 |
| 27.70 | 100.00 | 1 |
| 27.80 | 100.00 | 1 |
| 27.90 | 100.00 | 1 |
| 28 | 100.00 | 1 |
| 28.10 | 100.00 | 1 |
| 28.20 | 100.00 | 1 |
| 28.30 | 100.00 | 1 |
| 28.40 | 100.00 | 1 |
| 28.50 | 100.00 | 1 |
| 28.60 | 100.00 | 1 |
| 28.70 | 100.00 | 1 |
| 28.80 | 100.00 | 1 |
| 28.90 | 100.00 | 1 |
| 29 | 100.00 | 1 |
| 29.10 | 100.00 | 1 |
| 29.20 | 100.00 | 1 |
| 29.30 | 100.00 | 1 |
| 29.40 | 100.00 | 1 |
| 29.50 | 100.00 | 1 |
| 29.60 | 100.00 | 1 |
| 29.70 | 100.00 | 1 |
| 29.80 | 100.00 | 1 |
| 29.90 | 100.00 | 1 |
| 30 | 100.00 | 1 |
| 30.10 | 100.00 | 1 |
| 30.20 | 100.00 | 1 |
| 30.30 | 100.00 | 1 |
| 30.40 | 100.00 | 1 |
| 30.50 | 100 | 1 |
| 30.60 | 100 | 1 |
| 30.70 | 100 | 1 |
| 30.80 | 100 | 1 |
| 30.90 | 100 | 1 |
| 31 | 100 | 1 |
| 31.10 | 100 | 1 |
| 31.20 | 100 | 1 |
| 31.30 | 100 | 1 |
| 31.40 | 100 | 1 |
| 31.50 | 100 | 1 |
| 31.60 | 100 | 1 |
| 31.70 | 100 | 1 |
| 31.80 | 100 | 1 |
| 31.90 | 100 | 1 |
| 32 | 100 | 1 |
| 32.10 | 100 | 1 |
| 32.20 | 100 | 1 |
| 32.30 | 100 | 1 |
| 32.40 | 100 | 1 |
| 32.50 | 100 | 1 |
| 32.60 | 100 | 1 |
| 32.70 | 100 | 1 |
| 32.80 | 100 | 1 |
| 32.90 | 100 | 1 |
| 33 | 100 | 1 |
| 33.10 | 100 | 1 |
| 33.20 | 100 | 1 |
| 33.30 | 100 | 1 |
| 33.40 | 100 | 1 |
| 33.50 | 100 | 1 |
| 33.60 | 100 | 1 |
| 33.70 | 100 | 1 |
| 33.80 | 100 | 1 |
| 33.90 | 100 | 1 |
| 34 | 100 | 1 |
| 34.10 | 100 | 1 |
| 34.20 | 100 | 1 |
| 34.30 | 100 | 1 |
| 34.40 | 100 | 1 |
| 34.50 | 100 | 1 |
| 34.60 | 100 | 1 |
| 34.70 | 100 | 1 |
| 34.80 | 100 | 1 |
| 34.90 | 100 | 1 |
| 35 | 100 | 1 |
| 35.10 | 100 | 1 |
| 35.20 | 100 | 1 |
| 35.30 | 100 | 1 |
| 35.40 | 100 | 1 |
| 35.50 | 100 | 1 |
| 35.60 | 100 | 1 |
| 35.70 | 100 | 1 |
| 35.80 | 100 | 1 |
| 35.90 | 100 | 1 |
| 36 | 100 | 1 |
| 36.10 | 100 | 1 |
| 36.20 | 100 | 1 |
| 36.30 | 100 | 1 |
| 36.40 | 100 | 1 |
| 36.50 | 100 | 1 |
| 36.60 | 100 | 1 |
| 36.70 | 100 | 1 |
| 36.80 | 100 | 1 |
| 36.90 | 100 | 1 |
| 37 | 100 | 1 |
| 37.10 | 100 | 1 |
| 37.20 | 100 | 1 |
| 37.30 | 100 | 1 |
| 37.40 | 100 | 1 |
| 37.50 | 100 | 1 |
| 37.60 | 100 | 1 |
| 37.70 | 100 | 1 |
| 37.80 | 100 | 1 |
| 37.90 | 100 | 1 |
| 38 | 100 | 1 |
| 38.10 | 100 | 1 |
| 38.20 | 100 | 1 |
| 38.30 | 100 | 1 |
| 38.40 | 100 | 1 |
| 38.50 | 100 | 1 |
| 38.60 | 100 | 1 |
| 38.70 | 100 | 1 |
| 38.80 | 100 | 1 |
| 38.90 | 100 | 1 |
| 39 | 100 | 1 |
| 39.10 | 100 | 1 |
| 39.20 | 100 | 1 |
| 39.30 | 100 | 1 |
| 39.40 | 100 | 1 |
| 39.50 | 100 | 1 |
| 39.60 | 100 | 1 |
| 39.70 | 100 | 1 |
| 39.80 | 100 | 1 |
| 39.90 | 100 | 1 |
| 40 | 100 | 1 |

Views: Bias of the pooled rate (every row); Share of county-years published (every row).

Note: A county-year with a given expected number of deaths is published when its Poisson count is zero or at least 10. The ratio is the average death count in published county-years divided by the average count in all county-years, which is the ratio of the pooled rate to the true rate when counties share a population.

Note: The model assumes a constant rate in each county, independent years and Poisson counts. Real counts vary more than Poisson counts, so the curve gives direction and rough size, and it is not used to correct any estimate in this book.

Source: Calculation from the publication rule of the CDC WONDER county files, 2018-2024, applied to Poisson counts; full counts of the model, no sampling error, equal weight for every county-year

Data: https://vishalsingh.org/zipcode/report/suppressed-counties-in-death-data/figures/bias_curve.json

### Table 7. A suicide rate pooled over published county-years is 0.18 times the panel's rate in counties under 10,000 residents and 1.51 times in counties of 25,000 to 50,000 residents

Death rate pooled over the county-years CDC WONDER publishes, 2019 to 2023, divided by the CDC panel's population-weighted modeled rate for 2019 to 2023 in all counties of each size with a stable panel rate, suicide and drug overdose

| cause | conf_high | conf_low | estimate | n_units | series | size |
| --- | ---: | ---: | ---: | ---: | --- | --- |
| Suicide | 0.419 | 0 | 0.182 | 406 | Pooled over published county-years | Under 10,000 |
| Suicide | n/a | n/a | 0 | 41 | Pooled over counties published every year | Under 10,000 |
| Suicide | 1.07 | 0.496 | 0.781 | 406 | Pooled over published county-years | 10,000 to 24,999 |
| Suicide | 1.55 | 1.47 | 1.51 | 484 | Pooled over published county-years | 25,000 to 49,999 |
| Suicide | 1.76 | 1.39 | 1.56 | 12 | Pooled over counties published every year | 25,000 to 49,999 |
| Suicide | 1.10 | 1.07 | 1.08 | 379 | Pooled over published county-years | 50,000 to 99,999 |
| Suicide | 1.18 | 1.11 | 1.14 | 145 | Pooled over counties published every year | 50,000 to 99,999 |
| Suicide | 1.01 | 1.00 | 1.00 | 325 | Pooled over published county-years | 100,000 to 249,999 |
| Suicide | 1.02 | 1.01 | 1.01 | 304 | Pooled over counties published every year | 100,000 to 249,999 |
| Suicide | 1.00 | 1.00 | 1.00 | 273 | Pooled over published county-years | 250,000 or more |
| Suicide | 1.00 | 1.00 | 1.00 | 273 | Pooled over counties published every year | 250,000 or more |
| Drug overdose | 0.130 | 0 | 0.051 | 454 | Pooled over published county-years | Under 10,000 |
| Drug overdose | n/a | n/a | 0 | 128 | Pooled over counties published every year | Under 10,000 |
| Drug overdose | 1.77 | 1.45 | 1.63 | 673 | Pooled over published county-years | 10,000 to 24,999 |
| Drug overdose | 4.43 | 2.82 | 3.63 | 11 | Pooled over counties published every year | 10,000 to 24,999 |
| Drug overdose | 1.68 | 1.58 | 1.63 | 602 | Pooled over published county-years | 25,000 to 49,999 |
| Drug overdose | 2.29 | 1.92 | 2.09 | 66 | Pooled over counties published every year | 25,000 to 49,999 |
| Drug overdose | 1.20 | 1.16 | 1.18 | 387 | Pooled over published county-years | 50,000 to 99,999 |
| Drug overdose | 1.36 | 1.26 | 1.30 | 227 | Pooled over counties published every year | 50,000 to 99,999 |
| Drug overdose | 1.08 | 1.06 | 1.07 | 325 | Pooled over published county-years | 100,000 to 249,999 |
| Drug overdose | 1.10 | 1.07 | 1.09 | 306 | Pooled over counties published every year | 100,000 to 249,999 |
| Drug overdose | 1.05 | 1.05 | 1.05 | 273 | Pooled over published county-years | 250,000 or more |
| Drug overdose | 1.05 | 1.05 | 1.05 | 273 | Pooled over counties published every year | 250,000 or more |

Views: Suicide (cause = Suicide); Drug overdose (cause = Drug overdose).

Note: The panel is a model, so these ratios compare WONDER's rule with a modeled rate and not with the true rate. The panel's rates for small counties borrow from neighboring counties. WONDER counts and panel rates both cover 2019 to 2023. In counties of 50,000 to 100,000 residents the ratios are 1.09 for suicide and 1.18 for overdose, and in counties of 100,000 to 250,000 residents 1.00 and 1.07.

Note: Counties are all counties with a stable panel rate in every year 2019 to 2023, other than Connecticut's. A county that WONDER never publishes adds nothing to the pooled WONDER rate and stays in the panel rate. Restricting both sides to counties with a published county-year, the ratios for counties of 25,000 to 50,000 residents are 1.34 for suicide and 1.38 for overdose. A series is plotted for a size of county only when it has at least 10 counties; where every complete county has a rate of zero, the ratio is zero and has no interval.

Note: Overdose in the panel is unintentional or undetermined-intent drug overdose and in WONDER drug poisoning of any intent, so the levels differ by definition. In counties of 250,000 or more residents, where every count is published, the overdose ratio is 1.05, which measures that difference.

Source: CDC WONDER underlying cause of death, county files, 2019-2023, and CDC Mapping Injury, Overdose, and Violence county file, 2019-2023; crude rates, modeled estimates, population-weighted

Data: https://vishalsingh.org/zipcode/report/suppressed-counties-in-death-data/figures/naive_vs_panel.json

### Table 8. In every source the counties left without a value are a larger share of counties than of residents, from 55 counties without a premature death rate to 2,881 that WONDER linked birth and infant death does not name

Counties with no value in each source and measure, with the share of counties and the share of residents in them, for WONDER mortality, County Health Rankings, the CDC panel and the counties WONDER names in its birth products

| Source | Measure | Years | Rule that leaves a county out | Counties left out | Share of counties (%) | Share of residents (%) |
| --- | --- | --- | --- | ---: | ---: | ---: |
| CDC WONDER mortality, one county-year | Drug overdose | 2023 | Count of 1 to 9 deaths withheld | 1,426 | 45.4 | 8.6 |
| CDC WONDER mortality, one county-year | Suicide | 2023 | Count of 1 to 9 deaths withheld | 1,792 | 57.0 | 12.0 |
| CDC WONDER mortality, counties published every year | Drug overdose | 2018 to 2023 | Any county-year withheld, or no population | 2,200 | 70.0 | 17.1 |
| CDC WONDER mortality, counties published every year | Suicide | 2018 to 2023 | Any county-year withheld, or no population | 2,397 | 76.3 | 18.9 |
| County Health Rankings, pooled window | Drug overdose | 2019 to 2021 | Fewer than 10 deaths in the window | 1,244 | 39.6 | 4.0 |
| County Health Rankings, pooled window | Suicide | 2017 to 2021 | Fewer than 10 deaths in the window | 716 | 22.8 | 1.3 |
| County Health Rankings, pooled window | Homicide | 2015 to 2021 | Fewer than 10 deaths in the window | 1,767 | 56.2 | 9.7 |
| County Health Rankings, pooled window | Firearm deaths | 2017 to 2021 | Fewer than 10 deaths in the window | 864 | 27.5 | 2.0 |
| County Health Rankings, pooled window | Premature death | 2019 to 2021 | Fewer than 20 deaths in the window | 55 | 1.7 | 0.0 |
| CDC injury and violence panel, modeled | Drug overdose | 2019 to 2023 | No stable rate in some year | 428 | 13.6 | 2.3 |
| CDC injury and violence panel, modeled | Suicide | 2019 to 2023 | No stable rate in some year | 869 | 27.7 | 5.1 |
| CDC WONDER natality, counties named | Births | 2018 to 2024 | Counties under 100,000 residents pooled | 2,564 | 81.6 | 21.0 |
| CDC WONDER linked birth and infant death, counties named | Infant deaths | 2018 to 2022 | Counties under 250,000 residents pooled | 2,881 | 91.7 | 37.1 |

Note: Counties left out are those with no value. The share of counties uses each source's own count of counties (3,142 in WONDER, one more in County Health Rankings). Residents use WONDER's population for the WONDER mortality rows, County Health Rankings' for its rows and 2019-2023 American Community Survey residents for the rest.

Note: County Health Rankings windows are those of its 2024 release for each measure and the floors are the smallest published counts. The CDC panel row counts counties without a stable rate in every year, including Connecticut's 8, which the panel does not carry. Naming thresholds are 100,000 and 250,000 residents.

Source: CDC WONDER mortality, natality and linked birth and infant death, County Health Rankings 2024 and the CDC injury and violence panel, 2018-2024; full counts, crude rates and modeled estimates

Data: https://vishalsingh.org/zipcode/report/suppressed-counties-in-death-data/figures/sources_compared.json

## All estimates

Every fact in facts.json (160), sorted by id.

| Fact id | Value | Unit | Estimand | Population | Period | Interval | n | Weights | Source | Note |
| --- | ---: | --- | --- | --- | --- | --- | ---: | --- | --- | --- |
| chr_hom_n | 1,767 | counties | count of counties with no homicide rate because the window holds too few deaths | counties in County Health Rankings 2024 | pooled window of the 2024 release |   |   | none (full count) | County Health Rankings |   |
| chr_hom_pop | 9.7% | percent of residents | share of residents living in counties with no homicide rate | residents of the counties in County Health Rankings 2024 | pooled window of the 2024 release |   |   | county population in County Health Rankings | County Health Rankings |   |
| chr_od_n | 1,244 | counties | count of counties with no overdose rate because the window holds too few deaths | counties in County Health Rankings 2024 | pooled window of the 2024 release |   |   | none (full count) | County Health Rankings |   |
| chr_od_pop | 4.0% | percent of residents | share of residents living in counties with no overdose rate | residents of the counties in County Health Rankings 2024 | pooled window of the 2024 release |   |   | county population in County Health Rankings | County Health Rankings |   |
| chr_pmm_floor | 20 | deaths in the window | definition: County Health Rankings leaves out a premature death rate when the window holds fewer deaths than this (checked: the smallest published count is 20) | counties in County Health Rankings 2024 | pooled window of the 2024 release |   |   | none (definition) | County Health Rankings |   |
| chr_pmm_n | 55 | counties | count of counties with no premature death rate because the window holds too few deaths | counties in County Health Rankings 2024 | pooled window of the 2024 release |   |   | none (full count) | County Health Rankings |   |
| chr_su_n | 716 | counties | count of counties with no suicide rate because the window holds too few deaths | counties in County Health Rankings 2024 | pooled window of the 2024 release |   |   | none (full count) | County Health Rankings |   |
| ct_n | 8 | counties | count of Connecticut counties; WONDER has no population for them in 2022 to 2024 | Connecticut's legacy counties in the CDC WONDER county files | 2018 to 2024 |   |   | none (full count) | CDC WONDER underlying cause of death |   |
| cv_od_2 | 2.09 | ratio | the overdose rate pooled over counties published in each year 2019 to 2023, divided by the panel's population-weighted rate in all counties of the class; bootstrap over counties | counties with a stable panel overdose rate in every year 2019 to 2023 and 25,000 to 49,999 | county-years in 2019 to 2023 against annual modeled rates for 2019 to 2023 | 95% CI 1.92 to 2.28 | 602 counties | person-years (numerator); mean WONDER population 2019 to 2023 (panel) | CDC injury and violence mapping |   |
| cv_od_3 | 1.30 | ratio | the overdose rate pooled over counties published in each year 2019 to 2023, divided by the panel's population-weighted rate in all counties of the class; bootstrap over counties | counties with a stable panel overdose rate in every year 2019 to 2023 and 50,000 to 99,999 | county-years in 2019 to 2023 against annual modeled rates for 2019 to 2023 | 95% CI 1.26 to 1.36 | 387 counties | person-years (numerator); mean WONDER population 2019 to 2023 (panel) | CDC injury and violence mapping |   |
| cv_od_4 | 1.09 | ratio | the overdose rate pooled over counties published in each year 2019 to 2023, divided by the panel's population-weighted rate in all counties of the class; bootstrap over counties | counties with a stable panel overdose rate in every year 2019 to 2023 and 100,000 to 249,999 | county-years in 2019 to 2023 against annual modeled rates for 2019 to 2023 | 95% CI 1.07 to 1.10 | 325 counties | person-years (numerator); mean WONDER population 2019 to 2023 (panel) | CDC injury and violence mapping |   |
| lam_tol | 5% | percent | definition: how close to the true rate the pooled rate has to be to count as unbiased | the model of the publication rule | not applicable (definition) |   |   | none (definition) | CDC WONDER underlying cause of death |   |
| lbd_births | 64.0% | percent of live births | share of births in counties that WONDER linked birth and infant death names | live births in the United States | 2018 to 2022 |   |   | none (full count) | CDC WONDER linked birth and infant death |   |
| lbd_named | 263 | county identifiers | count of county identifiers other than the pooled Unidentified Counties rows | county identifiers that CDC WONDER linked birth and infant death lists in every year | 2018 to 2022 |   |   | none (full count) | CDC WONDER linked birth and infant death |   |
| lbd_real | 261 | counties | count of counties with births in every year | counties that CDC WONDER linked birth and infant death names and that have births in every year | 2018 to 2022 |   |   | none (full count) | CDC WONDER linked birth and infant death |   |
| lbd_thr | 250,000 | residents | definition: the population at which WONDER linked birth and infant death names a county (and a boundary between county size classes) | counties that WONDER linked birth and infant death names | 2018 to 2022 |   |   | none (definition) | CDC WONDER linked birth and infant death |   |
| lbd_unnamed | 2,881 | counties | count of counties that WONDER linked birth and infant death does not name, the two empty shells counted as pooled | counties and county equivalents in the CDC WONDER county files | 2018 to 2022 |   |   | none (full count) | CDC WONDER linked birth and infant death |   |
| m_cross | 4.2 | expected deaths a year | least expected number of deaths, before the peak, at which the ratio reaches one | counties that share one expected number of deaths a year under the model of the publication rule | not applicable (definition) |   |   | none (model) | CDC WONDER underlying cause of death |   |
| m_flat | 14.2 | expected deaths a year | least expected number of deaths, after the peak, at which the ratio is within the tolerance of one | counties that share one expected number of deaths a year under the model of the publication rule | not applicable (definition) |   |   | none (model) | CDC WONDER underlying cause of death |   |
| m_peak | 1.8 | ratio of the pooled rate to the true rate | largest value of the ratio of the expected pooled rate over published county-years to the true rate, Poisson counts and a constant rate | counties that share one expected number of deaths a year under the model of the publication rule | not applicable (definition) |   |   | none (model) | CDC WONDER underlying cause of death |   |
| m_peak_at | 5.4 | expected deaths a year | expected number of deaths at which the ratio is largest | counties that share one expected number of deaths a year under the model of the publication rule | not applicable (definition) |   |   | none (model) | CDC WONDER underlying cause of death |   |
| m_pub_peak | 5.3% | percent of county-years | share of county-years that WONDER publishes at the expected number of deaths where the ratio peaks | counties that share one expected number of deaths a year under the model of the publication rule | not applicable (definition) |   |   | none (model) | CDC WONDER underlying cause of death |   |
| map_floor | 20 | pooled deaths | definition: the least number of pooled deaths for a county to be mapped | a county drawn on a map of WONDER counts | not applicable (definition) |   |   | none (definition) | CDC WONDER underlying cause of death |   |
| min_group | 10 | counties | definition: the least number of counties for a group to be plotted | groups of counties plotted with an interval | not applicable (definition) |   |   | none (definition) | CDC WONDER underlying cause of death |   |
| mod_od_comp | 1.1% | percent of county-years that are modeled | share of county-years with a rate that CDC marks as modeled and not direct | county-years of the panel's overdose rates in counties whose WONDER count is published in each year 2019 to 2023 | annual modeled rates for 2019 to 2023, averaged |   | 1,011 counties | none (full count) | CDC injury and violence mapping |   |
| mod_od_other | 70.2% | percent of county-years that are modeled | share of county-years with a rate that CDC marks as modeled and not direct | county-years of the panel's overdose rates in counties whose WONDER count is withheld in at least one of those years | annual modeled rates for 2019 to 2023, averaged |   | 1,703 counties | none (full count) | CDC injury and violence mapping |   |
| mod_su_comp | 0.0% | percent of county-years that are modeled | share of county-years with a rate that CDC marks as modeled and not direct | county-years of the panel's suicide rates in counties whose WONDER count is published in each year 2019 to 2023 | annual modeled rates for 2019 to 2023, averaged |   | 775 counties | none (full count) | CDC injury and violence mapping |   |
| mod_su_other | 72.5% | percent of county-years that are modeled | share of county-years with a rate that CDC marks as modeled and not direct | county-years of the panel's suicide rates in counties whose WONDER count is withheld in at least one of those years | annual modeled rates for 2019 to 2023, averaged |   | 1,498 counties | none (full count) | CDC injury and violence mapping |   |
| n_100k | 606 | counties | count of counties with a mean population at or above the natality naming threshold | counties and county equivalents in the CDC WONDER county files | mean WONDER population 2018 to 2023 |   |   | none (full count) | CDC WONDER underlying cause of death |   |
| n_250k | 277 | counties | count of counties with a mean population at or above the linked birth and infant death naming threshold | counties and county equivalents in the CDC WONDER county files | mean WONDER population 2018 to 2023 |   |   | none (full count) | CDC WONDER underlying cause of death |   |
| n_counties | 3,142 | counties | count of counties in the county files (Valdez-Cordova AK is one county under its old code) | counties and county equivalents in the CDC WONDER county files | 2018 to 2024 |   |   | none (full count) | CDC WONDER underlying cause of death |   |
| n_scope | 3,134 | counties | count of counties in the comparisons of county groups | counties in the CDC WONDER county files other than Connecticut's | 2018 to 2023 |   |   | none (full count) | CDC WONDER underlying cause of death |   |
| nat_births | 79.6% | percent of live births | share of births in counties that WONDER natality names | live births in the United States | 2018 to 2023 |   |   | none (full count) | CDC WONDER natality |   |
| nat_named | 580 | county identifiers | count of county identifiers other than the pooled Unidentified Counties rows | county identifiers that CDC WONDER natality lists in every year | 2018 to 2024 |   |   | none (full count) | CDC WONDER natality |   |
| nat_real | 578 | counties | count of counties with births in every year (the balanced panel) | counties that CDC WONDER natality names and that have births in every year | 2018 to 2024 |   |   | none (full count) | CDC WONDER natality |   |
| nat_thr | 100,000 | residents | definition: the population at which WONDER natality names a county (and a boundary between county size classes) | counties that WONDER natality names | 2018 to 2024 |   |   | none (definition) | CDC WONDER natality |   |
| nat_unnamed | 2,564 | counties | count of counties that WONDER natality does not name, the two empty shells counted as pooled | counties and county equivalents in the CDC WONDER county files | 2018 to 2024 |   |   | none (full count) | CDC WONDER natality |   |
| nchs_ciw | 160% | percent of the estimate | definition: the widest relative width of the 95% interval at which NCHS presents a rate or count (Parker et al. 2023, Table A) | rates and counts that NCHS presents | not applicable (definition) |   |   | none (definition) | Published estimates from cited studies |   |
| nchs_min | 10 | events | definition: the least number of events for NCHS to present a rate or count (Parker et al. 2023, Table A) | rates and counts that NCHS presents | not applicable (definition) |   |   | none (definition) | Published estimates from cited studies |   |
| ns_od_2 | 1.38 | ratio | the overdose rate pooled over county-years WONDER publishes, divided by the panel's population-weighted rate in the same counties (those with a published county-year); bootstrap over counties | counties with a stable panel overdose rate in every year 2019 to 2023, 25,000 to 49,999 and at least one published county-year | county-years in 2019 to 2023 against annual modeled rates for 2019 to 2023 | 95% CI 1.35 to 1.41 | 412 counties | county-years with a published count (numerator); mean WONDER population 2019 to 2023 (panel) | CDC injury and violence mapping |   |
| ns_su_2 | 1.34 | ratio | the suicide rate pooled over county-years WONDER publishes, divided by the panel's population-weighted rate in the same counties (those with a published county-year); bootstrap over counties | counties with a stable panel suicide rate in every year 2019 to 2023, 25,000 to 49,999 and at least one published county-year | county-years in 2019 to 2023 against annual modeled rates for 2019 to 2023 | 95% CI 1.31 to 1.37 | 263 counties | county-years with a published count (numerator); mean WONDER population 2019 to 2023 (panel) | CDC injury and violence mapping |   |
| nv_od_0 | 0.05 | ratio | the overdose rate pooled over county-years WONDER publishes (a county never published adds nothing), divided by the panel's population-weighted rate in all counties of the class; bootstrap over counties | counties with a stable panel overdose rate in every year 2019 to 2023 and under 10,000 | county-years in 2019 to 2023 against annual modeled rates for 2019 to 2023 | 95% CI 0.00 to 0.13 | 454 counties | county-years with a published count (numerator); mean WONDER population 2019 to 2023 (panel) | CDC injury and violence mapping |   |
| nv_od_1 | 1.63 | ratio | the overdose rate pooled over county-years WONDER publishes (a county never published adds nothing), divided by the panel's population-weighted rate in all counties of the class; bootstrap over counties | counties with a stable panel overdose rate in every year 2019 to 2023 and 10,000 to 24,999 | county-years in 2019 to 2023 against annual modeled rates for 2019 to 2023 | 95% CI 1.45 to 1.77 | 673 counties | county-years with a published count (numerator); mean WONDER population 2019 to 2023 (panel) | CDC injury and violence mapping |   |
| nv_od_2 | 1.63 | ratio | the overdose rate pooled over county-years WONDER publishes (a county never published adds nothing), divided by the panel's population-weighted rate in all counties of the class; bootstrap over counties | counties with a stable panel overdose rate in every year 2019 to 2023 and 25,000 to 49,999 | county-years in 2019 to 2023 against annual modeled rates for 2019 to 2023 | 95% CI 1.58 to 1.68 | 602 counties | county-years with a published count (numerator); mean WONDER population 2019 to 2023 (panel) | CDC injury and violence mapping |   |
| nv_od_3 | 1.18 | ratio | the overdose rate pooled over county-years WONDER publishes (a county never published adds nothing), divided by the panel's population-weighted rate in all counties of the class; bootstrap over counties | counties with a stable panel overdose rate in every year 2019 to 2023 and 50,000 to 99,999 | county-years in 2019 to 2023 against annual modeled rates for 2019 to 2023 | 95% CI 1.16 to 1.20 | 387 counties | county-years with a published count (numerator); mean WONDER population 2019 to 2023 (panel) | CDC injury and violence mapping |   |
| nv_od_4 | 1.07 | ratio | the overdose rate pooled over county-years WONDER publishes (a county never published adds nothing), divided by the panel's population-weighted rate in all counties of the class; bootstrap over counties | counties with a stable panel overdose rate in every year 2019 to 2023 and 100,000 to 249,999 | county-years in 2019 to 2023 against annual modeled rates for 2019 to 2023 | 95% CI 1.06 to 1.08 | 325 counties | county-years with a published count (numerator); mean WONDER population 2019 to 2023 (panel) | CDC injury and violence mapping |   |
| nv_od_5 | 1.05 | ratio | the overdose rate pooled over county-years WONDER publishes (a county never published adds nothing), divided by the panel's population-weighted rate in all counties of the class; bootstrap over counties | counties with a stable panel overdose rate in every year 2019 to 2023 and 250,000 or more | county-years in 2019 to 2023 against annual modeled rates for 2019 to 2023 | 95% CI 1.05 to 1.06 | 273 counties | county-years with a published count (numerator); mean WONDER population 2019 to 2023 (panel) | CDC injury and violence mapping |   |
| nv_su_0 | 0.18 | ratio | the suicide rate pooled over county-years WONDER publishes (a county never published adds nothing), divided by the panel's population-weighted rate in all counties of the class; bootstrap over counties | counties with a stable panel suicide rate in every year 2019 to 2023 and under 10,000 | county-years in 2019 to 2023 against annual modeled rates for 2019 to 2023 | 95% CI 0.00 to 0.42 | 406 counties | county-years with a published count (numerator); mean WONDER population 2019 to 2023 (panel) | CDC injury and violence mapping |   |
| nv_su_1 | 0.78 | ratio | the suicide rate pooled over county-years WONDER publishes (a county never published adds nothing), divided by the panel's population-weighted rate in all counties of the class; bootstrap over counties | counties with a stable panel suicide rate in every year 2019 to 2023 and 10,000 to 24,999 | county-years in 2019 to 2023 against annual modeled rates for 2019 to 2023 | 95% CI 0.50 to 1.07 | 406 counties | county-years with a published count (numerator); mean WONDER population 2019 to 2023 (panel) | CDC injury and violence mapping |   |
| nv_su_2 | 1.51 | ratio | the suicide rate pooled over county-years WONDER publishes (a county never published adds nothing), divided by the panel's population-weighted rate in all counties of the class; bootstrap over counties | counties with a stable panel suicide rate in every year 2019 to 2023 and 25,000 to 49,999 | county-years in 2019 to 2023 against annual modeled rates for 2019 to 2023 | 95% CI 1.47 to 1.55 | 484 counties | county-years with a published count (numerator); mean WONDER population 2019 to 2023 (panel) | CDC injury and violence mapping |   |
| nv_su_3 | 1.09 | ratio | the suicide rate pooled over county-years WONDER publishes (a county never published adds nothing), divided by the panel's population-weighted rate in all counties of the class; bootstrap over counties | counties with a stable panel suicide rate in every year 2019 to 2023 and 50,000 to 99,999 | county-years in 2019 to 2023 against annual modeled rates for 2019 to 2023 | 95% CI 1.07 to 1.10 | 379 counties | county-years with a published count (numerator); mean WONDER population 2019 to 2023 (panel) | CDC injury and violence mapping |   |
| nv_su_4 | 1.00 | ratio | the suicide rate pooled over county-years WONDER publishes (a county never published adds nothing), divided by the panel's population-weighted rate in all counties of the class; bootstrap over counties | counties with a stable panel suicide rate in every year 2019 to 2023 and 100,000 to 249,999 | county-years in 2019 to 2023 against annual modeled rates for 2019 to 2023 | 95% CI 1.00 to 1.01 | 325 counties | county-years with a published count (numerator); mean WONDER population 2019 to 2023 (panel) | CDC injury and violence mapping |   |
| od_c_adi | 43.0 | national rank of the Area Deprivation Index, 1 least and 100 most deprived | population-weighted mean of the county values (mean rank); bootstrap over counties | counties whose overdose count WONDER publishes in every year | ADI 2023 | 95% CI 39.9 to 46.0 | 942 counties | mean WONDER population 2018 to 2023 | Area Deprivation Index |   |
| od_c_aian | 1.2% | percent of residents | population-weighted mean of the county values (who are American Indian or Alaska Native); bootstrap over counties | counties whose overdose count WONDER publishes in every year | Census Population Estimates as reported by County Health Rankings 2024 | 95% CI 1.0% to 1.3% | 942 counties | mean WONDER population 2018 to 2023 | County Health Rankings |   |
| od_c_black | 13.4% | percent of residents | population-weighted mean of the county values (who are non-Hispanic Black); bootstrap over counties | counties whose overdose count WONDER publishes in every year | Census Population Estimates as reported by County Health Rankings 2024 | 95% CI 12.1% to 14.7% | 942 counties | mean WONDER population 2018 to 2023 | County Health Rankings |   |
| od_c_hisp | 21.0% | percent of residents | population-weighted mean of the county values (who are Hispanic of any race); bootstrap over counties | counties whose overdose count WONDER publishes in every year | Census Population Estimates as reported by County Health Rankings 2024 | 95% CI 18.2% to 24.0% | 942 counties | mean WONDER population 2018 to 2023 | County Health Rankings |   |
| od_c_med | 134,390 | residents | median county population | counties whose overdose count WONDER publishes in every year | mean WONDER population 2018 to 2023 |   | 942 counties | none (each county counts once) | CDC WONDER underlying cause of death |   |
| od_c_nonmetro | 4.2% | percent of residents | population-weighted mean of the county values (living in nonmetro counties); bootstrap over counties | counties whose overdose count WONDER publishes in every year | 2023 codes | 95% CI 3.4% to 5.1% | 942 counties | mean WONDER population 2018 to 2023 | Rural-Urban Continuum Codes |   |
| od_c_white | 55.3% | percent of residents | population-weighted mean of the county values (who are non-Hispanic white); bootstrap over counties | counties whose overdose count WONDER publishes in every year | Census Population Estimates as reported by County Health Rankings 2024 | 95% CI 52.1% to 58.5% | 942 counties | mean WONDER population 2018 to 2023 | County Health Rankings |   |
| od_n_adi | 70.2 | national rank of the Area Deprivation Index, 1 least and 100 most deprived | population-weighted mean of the county values (mean rank); bootstrap over counties | counties whose overdose count WONDER withholds in every year | ADI 2023 | 95% CI 68.5 to 71.8 | 591 counties | mean WONDER population 2018 to 2023 | Area Deprivation Index |   |
| od_n_aian | 2.1% | percent of residents | population-weighted mean of the county values (who are American Indian or Alaska Native); bootstrap over counties | counties whose overdose count WONDER withholds in every year | Census Population Estimates as reported by County Health Rankings 2024 | 95% CI 1.7% to 2.6% | 591 counties | mean WONDER population 2018 to 2023 | County Health Rankings |   |
| od_n_black | 8.1% | percent of residents | population-weighted mean of the county values (who are non-Hispanic Black); bootstrap over counties | counties whose overdose count WONDER withholds in every year | Census Population Estimates as reported by County Health Rankings 2024 | 95% CI 7.1% to 9.2% | 591 counties | mean WONDER population 2018 to 2023 | County Health Rankings |   |
| od_n_hisp | 10.0% | percent of residents | population-weighted mean of the county values (who are Hispanic of any race); bootstrap over counties | counties whose overdose count WONDER withholds in every year | Census Population Estimates as reported by County Health Rankings 2024 | 95% CI 8.4% to 11.7% | 591 counties | mean WONDER population 2018 to 2023 | County Health Rankings |   |
| od_n_med | 19,761 | residents | median county population | counties whose overdose count WONDER withholds in every year | mean WONDER population 2018 to 2023 |   | 591 counties | none (each county counts once) | CDC WONDER underlying cause of death |   |
| od_n_nonmetro | 75.5% | percent of residents | population-weighted mean of the county values (living in nonmetro counties); bootstrap over counties | counties whose overdose count WONDER withholds in every year | 2023 codes | 95% CI 71.2% to 79.9% | 591 counties | mean WONDER population 2018 to 2023 | Rural-Urban Continuum Codes |   |
| od_n_white | 77.2% | percent of residents | population-weighted mean of the county values (who are non-Hispanic white); bootstrap over counties | counties whose overdose count WONDER withholds in every year | Census Population Estimates as reported by County Health Rankings 2024 | 95% CI 75.3% to 79.1% | 591 counties | mean WONDER population 2018 to 2023 | County Health Rankings |   |
| pan_od_n | 428 | counties | count of counties with no stable panel overdose rate in some year (Connecticut included) | counties and county equivalents in the CDC WONDER county files | annual modeled rates for 2019 to 2023, averaged |   |   | none (full count) | CDC injury and violence mapping |   |
| pan_od_pop | 2.3% | percent of residents | share of residents living in counties with no stable panel overdose rate in some year | residents of the counties in the CDC WONDER county files | annual modeled rates for 2019 to 2023, averaged |   |   | 2019-2023 ACS residents | CDC injury and violence mapping |   |
| reg_su_mw | 31.1% | percent of counties | share of counties whose suicide count WONDER withholds in every year | counties in the Midwest Census region | county-years in 2018 to 2023 |   | 1,055 counties | none (each county counts once) | CDC WONDER underlying cause of death |   |
| reg_su_ne | 12.0% | percent of counties | share of counties whose suicide count WONDER withholds in every year | counties in the Northeast Census region | county-years in 2018 to 2023 |   | 209 counties | none (each county counts once) | CDC WONDER underlying cause of death |   |
| reg_su_south | 32.7% | percent of counties | share of counties whose suicide count WONDER withholds in every year | counties in the South Census region | county-years in 2018 to 2023 |   | 1,422 counties | none (each county counts once) | CDC WONDER underlying cause of death |   |
| reg_su_west | 22.5% | percent of counties | share of counties whose suicide count WONDER withholds in every year | counties in the West Census region | county-years in 2018 to 2023 |   | 448 counties | none (each county counts once) | CDC WONDER underlying cause of death |   |
| rse_floor | 22.4% | percent of the count | relative standard error of a Poisson count at the map floor: 100 divided by the square root of the count | a death count at the map floor | not applicable (definition) |   |   | none (definition) | CDC WONDER underlying cause of death |   |
| rucc_cut | 4 | Rural-Urban Continuum Code | definition: the lowest code of a nonmetro county | counties in the 2023 Rural-Urban Continuum Code | 2023 |   |   | none (definition) | Rural-Urban Continuum Codes |   |
| sc_od_0 | 0.0 | deaths per 100,000 residents | population-weighted mean of the panel's county overdose rates (every county has a rate of zero in every year) | counties with a stable panel overdose rate in every year 2019 to 2023 and under 10,000, whose WONDER count is published in each year 2019 to 2023 | annual modeled rates for 2019 to 2023, averaged |   | 128 counties | mean WONDER population 2019 to 2023 | CDC injury and violence mapping |   |
| sc_od_2 | 52.3 | deaths per 100,000 residents | population-weighted mean of the panel's county overdose rates; bootstrap over counties | counties with a stable panel overdose rate in every year 2019 to 2023 and 25,000 to 49,999, whose WONDER count is published in each year 2019 to 2023 | annual modeled rates for 2019 to 2023, averaged | 95% CI 47.3 to 57.8 | 66 counties | mean WONDER population 2019 to 2023 | CDC injury and violence mapping |   |
| sc_od_3 | 35.8 | deaths per 100,000 residents | population-weighted mean of the panel's county overdose rates; bootstrap over counties | counties with a stable panel overdose rate in every year 2019 to 2023 and 50,000 to 99,999, whose WONDER count is published in each year 2019 to 2023 | annual modeled rates for 2019 to 2023, averaged | 95% CI 33.5 to 38.5 | 227 counties | mean WONDER population 2019 to 2023 | CDC injury and violence mapping |   |
| sc_od_4 | 27.6 | deaths per 100,000 residents | population-weighted mean of the panel's county overdose rates; bootstrap over counties | counties with a stable panel overdose rate in every year 2019 to 2023 and 100,000 to 249,999, whose WONDER count is published in each year 2019 to 2023 | annual modeled rates for 2019 to 2023, averaged | 95% CI 26.2 to 29.2 | 306 counties | mean WONDER population 2019 to 2023 | CDC injury and violence mapping |   |
| sc_su_0 | 0.0 | deaths per 100,000 residents | population-weighted mean of the panel's county suicide rates (every county has a rate of zero in every year) | counties with a stable panel suicide rate in every year 2019 to 2023 and under 10,000, whose WONDER count is published in each year 2019 to 2023 | annual modeled rates for 2019 to 2023, averaged |   | 41 counties | mean WONDER population 2019 to 2023 | CDC injury and violence mapping |   |
| sc_su_2 | 30.9 | deaths per 100,000 residents | population-weighted mean of the panel's county suicide rates; bootstrap over counties | counties with a stable panel suicide rate in every year 2019 to 2023 and 25,000 to 49,999, whose WONDER count is published in each year 2019 to 2023 | annual modeled rates for 2019 to 2023, averaged | 95% CI 27.5 to 34.9 | 12 counties | mean WONDER population 2019 to 2023 | CDC injury and violence mapping |   |
| sc_su_3 | 21.6 | deaths per 100,000 residents | population-weighted mean of the panel's county suicide rates; bootstrap over counties | counties with a stable panel suicide rate in every year 2019 to 2023 and 50,000 to 99,999, whose WONDER count is published in each year 2019 to 2023 | annual modeled rates for 2019 to 2023, averaged | 95% CI 20.7 to 22.5 | 145 counties | mean WONDER population 2019 to 2023 | CDC injury and violence mapping |   |
| sc_su_4 | 17.0 | deaths per 100,000 residents | population-weighted mean of the panel's county suicide rates; bootstrap over counties | counties with a stable panel suicide rate in every year 2019 to 2023 and 100,000 to 249,999, whose WONDER count is published in each year 2019 to 2023 | annual modeled rates for 2019 to 2023, averaged | 95% CI 16.4 to 17.5 | 304 counties | mean WONDER population 2019 to 2023 | CDC injury and violence mapping |   |
| shell_lbd_n | 2 | counties | count of empty-shell counties (Hinds MS and Mahoning OH) | counties that WONDER linked birth and infant death lists with no births of their own | 2018 to 2022 |   |   | none (full count) | CDC WONDER linked birth and infant death |   |
| shell_n | 2 | counties | count of empty-shell counties (Cape May NJ and Portsmouth VA) | counties that WONDER natality lists with no births of their own | 2018 to 2024 |   |   | none (full count) | CDC WONDER natality |   |
| so_od_0 | 13.0 | deaths per 100,000 residents | population-weighted mean of the panel's county overdose rates; bootstrap over counties | counties with a stable panel overdose rate in every year 2019 to 2023 and under 10,000, whose WONDER count is withheld in at least one of those years, including counties never published | annual modeled rates for 2019 to 2023, averaged | 95% CI 11.7 to 14.3 | 326 counties | mean WONDER population 2019 to 2023 | CDC injury and violence mapping |   |
| so_od_2 | 22.3 | deaths per 100,000 residents | population-weighted mean of the panel's county overdose rates; bootstrap over counties | counties with a stable panel overdose rate in every year 2019 to 2023 and 25,000 to 49,999, whose WONDER count is withheld in at least one of those years, including counties never published | annual modeled rates for 2019 to 2023, averaged | 95% CI 21.5 to 23.2 | 536 counties | mean WONDER population 2019 to 2023 | CDC injury and violence mapping |   |
| so_od_3 | 17.5 | deaths per 100,000 residents | population-weighted mean of the panel's county overdose rates; bootstrap over counties | counties with a stable panel overdose rate in every year 2019 to 2023 and 50,000 to 99,999, whose WONDER count is withheld in at least one of those years, including counties never published | annual modeled rates for 2019 to 2023, averaged | 95% CI 16.5 to 18.5 | 160 counties | mean WONDER population 2019 to 2023 | CDC injury and violence mapping |   |
| so_od_4 | 11.3 | deaths per 100,000 residents | population-weighted mean of the panel's county overdose rates; bootstrap over counties | counties with a stable panel overdose rate in every year 2019 to 2023 and 100,000 to 249,999, whose WONDER count is withheld in at least one of those years, including counties never published | annual modeled rates for 2019 to 2023, averaged | 95% CI 10.2 to 12.6 | 19 counties | mean WONDER population 2019 to 2023 | CDC injury and violence mapping |   |
| so_su_0 | 14.2 | deaths per 100,000 residents | population-weighted mean of the panel's county suicide rates; bootstrap over counties | counties with a stable panel suicide rate in every year 2019 to 2023 and under 10,000, whose WONDER count is withheld in at least one of those years, including counties never published | annual modeled rates for 2019 to 2023, averaged | 95% CI 13.2 to 15.4 | 365 counties | mean WONDER population 2019 to 2023 | CDC injury and violence mapping |   |
| so_su_2 | 19.4 | deaths per 100,000 residents | population-weighted mean of the panel's county suicide rates; bootstrap over counties | counties with a stable panel suicide rate in every year 2019 to 2023 and 25,000 to 49,999, whose WONDER count is withheld in at least one of those years, including counties never published | annual modeled rates for 2019 to 2023, averaged | 95% CI 19.0 to 19.8 | 472 counties | mean WONDER population 2019 to 2023 | CDC injury and violence mapping |   |
| so_su_3 | 17.0 | deaths per 100,000 residents | population-weighted mean of the panel's county suicide rates; bootstrap over counties | counties with a stable panel suicide rate in every year 2019 to 2023 and 50,000 to 99,999, whose WONDER count is withheld in at least one of those years, including counties never published | annual modeled rates for 2019 to 2023, averaged | 95% CI 16.6 to 17.4 | 234 counties | mean WONDER population 2019 to 2023 | CDC injury and violence mapping |   |
| so_su_4 | 12.4 | deaths per 100,000 residents | population-weighted mean of the panel's county suicide rates; bootstrap over counties | counties with a stable panel suicide rate in every year 2019 to 2023 and 100,000 to 249,999, whose WONDER count is withheld in at least one of those years, including counties never published | annual modeled rates for 2019 to 2023, averaged | 95% CI 11.3 to 13.7 | 21 counties | mean WONDER population 2019 to 2023 | CDC injury and violence mapping |   |
| state_cells | 357 | state-years | count of state-years for each kind of death | the 50 states and DC in the WONDER state files, for one kind of death | 2018 to 2024 |   |   | none (full count) | CDC WONDER underlying cause of death |   |
| state_withheld | 0 | state-years | count of state-years whose death count WONDER withholds | the 50 states and DC in the WONDER state files, for the three kinds of death | 2018 to 2024 |   | 1,071 state-years | none (full count) | CDC WONDER underlying cause of death |   |
| sz_c1 | 10,000 | residents | definition: a boundary between county size classes | counties in the size classes of this chapter | mean WONDER population 2018 to 2023 |   |   | none (definition) | CDC WONDER underlying cause of death |   |
| sz_c2 | 25,000 | residents | definition: a boundary between county size classes | counties in the size classes of this chapter | mean WONDER population 2018 to 2023 |   |   | none (definition) | CDC WONDER underlying cause of death |   |
| sz_c3 | 50,000 | residents | definition: a boundary between county size classes | counties in the size classes of this chapter | mean WONDER population 2018 to 2023 |   |   | none (definition) | CDC WONDER underlying cause of death |   |
| sz_wh_alc_0 | 54.9% | percent of county-years | share of county-years whose alcohol-induced count WONDER withholds | county-years of counties in the size class under 10,000 | county-years in 2018 to 2023 |   | 4,374 county-years | none (full count) | CDC WONDER underlying cause of death |   |
| sz_wh_alc_1 | 84.3% | percent of county-years | share of county-years whose alcohol-induced count WONDER withholds | county-years of counties in the size class 10,000 to 24,999 | county-years in 2018 to 2023 |   | 4,794 county-years | none (full count) | CDC WONDER underlying cause of death |   |
| sz_wh_alc_4 | 9.7% | percent of county-years | share of county-years whose alcohol-induced count WONDER withholds | county-years of counties in the size class 100,000 to 249,999 | county-years in 2018 to 2023 |   | 1,974 county-years | none (full count) | CDC WONDER underlying cause of death |   |
| sz_wh_alc_5 | 0.1% | percent of county-years | share of county-years whose alcohol-induced count WONDER withholds | county-years of counties in the size class 250,000 or more | county-years in 2018 to 2023 |   | 1,662 county-years | none (full count) | CDC WONDER underlying cause of death |   |
| sz_wh_od_0 | 45.2% | percent of county-years | share of county-years whose drug overdose count WONDER withholds | county-years of counties in the size class under 10,000 | county-years in 2018 to 2023 |   | 4,374 county-years | none (full count) | CDC WONDER underlying cause of death |   |
| sz_wh_od_1 | 82.3% | percent of county-years | share of county-years whose drug overdose count WONDER withholds | county-years of counties in the size class 10,000 to 24,999 | county-years in 2018 to 2023 |   | 4,794 county-years | none (full count) | CDC WONDER underlying cause of death |   |
| sz_wh_od_4 | 2.5% | percent of county-years | share of county-years whose drug overdose count WONDER withholds | county-years of counties in the size class 100,000 to 249,999 | county-years in 2018 to 2023 |   | 1,974 county-years | none (full count) | CDC WONDER underlying cause of death |   |
| sz_wh_od_5 | 0.0% | percent of county-years | share of county-years whose drug overdose count WONDER withholds | county-years of counties in the size class 250,000 or more | county-years in 2018 to 2023 |   | 1,662 county-years | none (full count) | CDC WONDER underlying cause of death |   |
| sz_wh_su_0 | 61.3% | percent of county-years | share of county-years whose suicide count WONDER withholds | county-years of counties in the size class under 10,000 | county-years in 2018 to 2023 |   | 4,374 county-years | none (full count) | CDC WONDER underlying cause of death |   |
| sz_wh_su_1 | 93.1% | percent of county-years | share of county-years whose suicide count WONDER withholds | county-years of counties in the size class 10,000 to 24,999 | county-years in 2018 to 2023 |   | 4,794 county-years | none (full count) | CDC WONDER underlying cause of death |   |
| sz_wh_su_4 | 1.5% | percent of county-years | share of county-years whose suicide count WONDER withholds | county-years of counties in the size class 100,000 to 249,999 | county-years in 2018 to 2023 |   | 1,974 county-years | none (full count) | CDC WONDER underlying cause of death |   |
| sz_wh_su_5 | 0.0% | percent of county-years | share of county-years whose suicide count WONDER withholds | county-years of counties in the size class 250,000 or more | county-years in 2018 to 2023 |   | 1,662 county-years | none (full count) | CDC WONDER underlying cause of death |   |
| sz_zero_alc_0 | 44.8% | percent of county-years | share of county-years whose alcohol-induced count WONDER publishes as zero | county-years of counties under the smallest boundary | county-years in 2018 to 2023 |   | 4,374 county-years | none (full count) | CDC WONDER underlying cause of death |   |
| sz_zero_od_0 | 54.7% | percent of county-years | share of county-years whose drug overdose count WONDER publishes as zero | county-years of counties under the smallest boundary | county-years in 2018 to 2023 |   | 4,374 county-years | none (full count) | CDC WONDER underlying cause of death |   |
| sz_zero_su_0 | 38.5% | percent of county-years | share of county-years whose suicide count WONDER publishes as zero | county-years of counties under the smallest boundary | county-years in 2018 to 2023 |   | 4,374 county-years | none (full count) | CDC WONDER underlying cause of death |   |
| w_1019_alc | 43.8% | percent of published county-years with 10 or more deaths | share of those county-years with fewer deaths than the map floor | county-years whose alcohol-induced count WONDER publishes at the floor or above | county-years in 2018 to 2023 |   | 5,224 county-years | none (full count) | CDC WONDER underlying cause of death |   |
| w_1019_od | 35.4% | percent of published county-years with 10 or more deaths | share of those county-years with fewer deaths than the map floor | county-years whose drug overdose count WONDER publishes at the floor or above | county-years in 2018 to 2023 |   | 6,987 county-years | none (full count) | CDC WONDER underlying cause of death |   |
| w_1019_su | 44.8% | percent of published county-years with 10 or more deaths | share of those county-years with fewer deaths than the map floor | county-years whose suicide count WONDER publishes at the floor or above | county-years in 2018 to 2023 |   | 6,037 county-years | none (full count) | CDC WONDER underlying cause of death |   |
| w_cell_max | 9 | deaths in a county-year | definition: the largest county-year count that WONDER withholds | county-years in the CDC WONDER county files, each kind of death counted separately | not applicable (definition) |   |   | none (definition) | CDC WONDER underlying cause of death |   |
| w_cells | 18,852 | county-years | count of county-years for each kind of death | counties and county equivalents in the CDC WONDER county files | 2018 to 2023 |   |   | none (full count) | CDC WONDER underlying cause of death |   |
| w_comp_alc | 644 | counties | count of counties with no withheld county-year (complete counties) | counties whose alcohol-induced count WONDER publishes in every year 2018 to 2023, other than Connecticut's | county-years in 2018 to 2023 |   |   | none (full count) | CDC WONDER underlying cause of death |   |
| w_comp_od | 942 | counties | count of counties with no withheld county-year (complete counties) | counties whose drug overdose count WONDER publishes in every year 2018 to 2023, other than Connecticut's | county-years in 2018 to 2023 |   |   | none (full count) | CDC WONDER underlying cause of death |   |
| w_comp_pop_alc | 77.9% | percent of residents | share of residents living in counties whose alcohol-induced count is published in every year | residents of counties other than Connecticut's | county-years in 2018 to 2023 |   |   | mean WONDER population 2018 to 2023 | CDC WONDER underlying cause of death |   |
| w_comp_pop_od | 83.8% | percent of residents | share of residents living in counties whose drug overdose count is published in every year | residents of counties other than Connecticut's | county-years in 2018 to 2023 |   |   | mean WONDER population 2018 to 2023 | CDC WONDER underlying cause of death |   |
| w_comp_pop_su | 82.0% | percent of residents | share of residents living in counties whose suicide count is published in every year | residents of counties other than Connecticut's | county-years in 2018 to 2023 |   |   | mean WONDER population 2018 to 2023 | CDC WONDER underlying cause of death |   |
| w_comp_su | 745 | counties | count of counties with no withheld county-year (complete counties) | counties whose suicide count WONDER publishes in every year 2018 to 2023, other than Connecticut's | county-years in 2018 to 2023 |   |   | none (full count) | CDC WONDER underlying cause of death |   |
| w_floor | 10 | deaths in a county-year | definition: WONDER withholds a county-year count above zero and below this number | county-years in the CDC WONDER county files, each kind of death counted separately | not applicable (definition) |   |   | none (definition) | CDC WONDER underlying cause of death |   |
| w_max_alc | 29.1% | percent of deaths in the county files | largest share of the county files' alcohol-induced deaths that the withheld cells could hold at the largest withheld count each; the public limit and the published cells only | deaths in county-years in the CDC WONDER county files | county-years in 2018 to 2023 |   |   | none (bound from the public limit) | CDC WONDER underlying cause of death |   |
| w_max_od | 13.4% | percent of deaths in the county files | largest share of the county files' drug overdose deaths that the withheld cells could hold at the largest withheld count each; the public limit and the published cells only | deaths in county-years in the CDC WONDER county files | county-years in 2018 to 2023 |   |   | none (bound from the public limit) | CDC WONDER underlying cause of death |   |
| w_max_su | 28.4% | percent of deaths in the county files | largest share of the county files' suicide deaths that the withheld cells could hold at the largest withheld count each; the public limit and the published cells only | deaths in county-years in the CDC WONDER county files | county-years in 2018 to 2023 |   |   | none (bound from the public limit) | CDC WONDER underlying cause of death |   |
| w_never_alc | 782 | counties | count of counties with every county-year withheld | counties whose alcohol-induced count WONDER withholds in every year 2018 to 2023, other than Connecticut's | county-years in 2018 to 2023 |   |   | none (full count) | CDC WONDER underlying cause of death |   |
| w_never_od | 591 | counties | count of counties with every county-year withheld | counties whose drug overdose count WONDER withholds in every year 2018 to 2023, other than Connecticut's | county-years in 2018 to 2023 |   |   | none (full count) | CDC WONDER underlying cause of death |   |
| w_never_pop_alc | 6.4% | percent of residents | share of residents living in counties whose alcohol-induced count is withheld in every year | residents of counties other than Connecticut's | county-years in 2018 to 2023 |   |   | mean WONDER population 2018 to 2023 | CDC WONDER underlying cause of death |   |
| w_never_pop_od | 4.1% | percent of residents | share of residents living in counties whose drug overdose count is withheld in every year | residents of counties other than Connecticut's | county-years in 2018 to 2023 |   |   | mean WONDER population 2018 to 2023 | CDC WONDER underlying cause of death |   |
| w_never_pop_su | 5.8% | percent of residents | share of residents living in counties whose suicide count is withheld in every year | residents of counties other than Connecticut's | county-years in 2018 to 2023 |   |   | mean WONDER population 2018 to 2023 | CDC WONDER underlying cause of death |   |
| w_never_su | 919 | counties | count of counties with every county-year withheld | counties whose suicide count WONDER withholds in every year 2018 to 2023, other than Connecticut's | county-years in 2018 to 2023 |   |   | none (full count) | CDC WONDER underlying cause of death |   |
| w_nowh_alc | 652 | counties | count of counties shaded at zero withheld years on the map (complete counties plus Connecticut's, which have no population in 2022 and 2023) | counties with no withheld alcohol-induced county-year 2018 to 2023, Connecticut's included | county-years in 2018 to 2023 |   |   | none (full count) | CDC WONDER underlying cause of death |   |
| w_nowh_od | 950 | counties | count of counties shaded at zero withheld years on the map (complete counties plus Connecticut's, which have no population in 2022 and 2023) | counties with no withheld drug overdose county-year 2018 to 2023, Connecticut's included | county-years in 2018 to 2023 |   |   | none (full count) | CDC WONDER underlying cause of death |   |
| w_nowh_su | 752 | counties | count of counties shaded at zero withheld years on the map (complete counties plus Connecticut's, which have no population in 2022 and 2023) | counties with no withheld suicide county-year 2018 to 2023, Connecticut's included | county-years in 2018 to 2023 |   |   | none (full count) | CDC WONDER underlying cause of death |   |
| w_pop_alc | 14.8% | percent of residents | share of the summed county populations that sit in county-years whose alcohol-induced count is withheld (WONDER publishes the population) | residents counted in the county-years of the WONDER county files | county-years in 2018 to 2023 |   |   | none (full count) | CDC WONDER underlying cause of death |   |
| w_pop_od | 10.1% | percent of residents | share of the summed county populations that sit in county-years whose drug overdose count is withheld (WONDER publishes the population) | residents counted in the county-years of the WONDER county files | county-years in 2018 to 2023 |   |   | none (full count) | CDC WONDER underlying cause of death |   |
| w_pop_su | 12.2% | percent of residents | share of the summed county populations that sit in county-years whose suicide count is withheld (WONDER publishes the population) | residents counted in the county-years of the WONDER county files | county-years in 2018 to 2023 |   |   | none (full count) | CDC WONDER underlying cause of death |   |
| w_some_alc | 1,708 | counties | count of counties with some county-years withheld | counties whose alcohol-induced count WONDER withholds in some but not all years 2018 to 2023, other than Connecticut's | county-years in 2018 to 2023 |   |   | none (full count) | CDC WONDER underlying cause of death |   |
| w_some_od | 1,601 | counties | count of counties with some county-years withheld | counties whose drug overdose count WONDER withholds in some but not all years 2018 to 2023, other than Connecticut's | county-years in 2018 to 2023 |   |   | none (full count) | CDC WONDER underlying cause of death |   |
| w_some_pop_od | 12.1% | percent of residents | share of residents living in counties whose overdose count is withheld in some but not all years | residents of counties other than Connecticut's | county-years in 2018 to 2023 |   |   | mean WONDER population 2018 to 2023 | CDC WONDER underlying cause of death |   |
| w_some_sc_od | 51.1% | percent of counties | share of counties whose overdose count is withheld in some but not all years | counties other than Connecticut's | county-years in 2018 to 2023 |   |   | none (each county counts once) | CDC WONDER underlying cause of death |   |
| w_some_su | 1,470 | counties | count of counties with some county-years withheld | counties whose suicide count WONDER withholds in some but not all years 2018 to 2023, other than Connecticut's | county-years in 2018 to 2023 |   |   | none (full count) | CDC WONDER underlying cause of death |   |
| w_wh_alc | 57.9% | percent of county-years | share of county-years whose alcohol-induced count WONDER withholds | county-years in the CDC WONDER county files, each kind of death counted separately | county-years in 2018 to 2023 |   | 18,852 county-years | none (full count) | CDC WONDER underlying cause of death |   |
| w_wh_od | 47.1% | percent of county-years | share of county-years whose drug overdose count WONDER withholds | county-years in the CDC WONDER county files, each kind of death counted separately | county-years in 2018 to 2023 |   | 18,852 county-years | none (full count) | CDC WONDER underlying cause of death |   |
| w_wh_su | 57.5% | percent of county-years | share of county-years whose suicide count WONDER withholds | county-years in the CDC WONDER county files, each kind of death counted separately | county-years in 2018 to 2023 |   | 18,852 county-years | none (full count) | CDC WONDER underlying cause of death |   |
| w_zero_alc | 34.2% | percent of published county-years | share of published county-years with a count of zero | county-years whose alcohol-induced count WONDER publishes | county-years in 2018 to 2023 |   | 7,943 county-years | none (full count) | CDC WONDER underlying cause of death |   |
| w_zero_od | 30.0% | percent of published county-years | share of published county-years with a count of zero | county-years whose drug overdose count WONDER publishes | county-years in 2018 to 2023 |   | 9,982 county-years | none (full count) | CDC WONDER underlying cause of death |   |
| w_zero_su | 24.7% | percent of published county-years | share of published county-years with a count of zero | county-years whose suicide count WONDER publishes | county-years in 2018 to 2023 |   | 8,017 county-years | none (full count) | CDC WONDER underlying cause of death |   |
| yr_alc_pop_2018 | 17.4% | percent of residents | share of residents in county-years whose alcohol-induced count is withheld, in 2018 | residents counted in the county-years of the WONDER county files | county-years in 2018 to 2024 |   |   | none (full count) | CDC WONDER underlying cause of death |   |
| yr_alc_pop_min | 12.5% | percent of residents | share of residents in county-years whose alcohol-induced count is withheld, in its lowest year | residents counted in the county-years of the WONDER county files | county-years in 2018 to 2024 |   |   | none (full count) | CDC WONDER underlying cause of death |   |
| yr_od_pop_2018 | 12.3% | percent of residents | share of residents in county-years whose drug overdose count is withheld, in 2018 | residents counted in the county-years of the WONDER county files | county-years in 2018 to 2024 |   |   | none (full count) | CDC WONDER underlying cause of death |   |
| yr_od_pop_min | 8.6% | percent of residents | share of residents in county-years whose drug overdose count is withheld, in its lowest year | residents counted in the county-years of the WONDER county files | county-years in 2018 to 2024 |   |   | none (full count) | CDC WONDER underlying cause of death |   |
| yr_pop_max | 17.4% | percent of residents | largest share of residents in withheld county-years in any year and for any of the three kinds of death | residents counted in the county-years of the WONDER county files | county-years in 2018 to 2024 |   |   | none (full count) | CDC WONDER underlying cause of death |   |
| yr_pop_min | 8.6% | percent of residents | smallest share of residents in withheld county-years in any year and for any of the three kinds of death | residents counted in the county-years of the WONDER county files | county-years in 2018 to 2024 |   |   | none (full count) | CDC WONDER underlying cause of death |   |
| yr_share_max | 59.9% | percent of county-years | largest share of county-years withheld in any year and for any of the three kinds of death | county-years in the CDC WONDER county files, each kind of death counted separately | county-years in 2018 to 2024 |   |   | none (full count) | CDC WONDER underlying cause of death |   |
| yr_share_min | 44.6% | percent of county-years | smallest share of county-years withheld in any year and for any of the three kinds of death | county-years in the CDC WONDER county files, each kind of death counted separately | county-years in 2018 to 2024 |   |   | none (full count) | CDC WONDER underlying cause of death |   |
| yr_su_pop_max | 12.7% | percent of residents | share of residents in county-years whose suicide count is withheld, in its highest year | residents counted in the county-years of the WONDER county files | county-years in 2018 to 2024 |   |   | none (full count) | CDC WONDER underlying cause of death |   |
| yr_su_pop_min | 11.8% | percent of residents | share of residents in county-years whose suicide count is withheld, in its lowest year | residents counted in the county-years of the WONDER county files | county-years in 2018 to 2024 |   |   | none (full count) | CDC WONDER underlying cause of death |   |

## Definitions

- Complete county: A county whose death count is published in every year of a window, possibly as zero, so its rate over the window is exact. Counties that are complete are selected by their counts, so as a group they are not typical of counties their size.
- Suppression: The withholding of a number because the count behind it is too small to publish without risking the identification of a person. A suppressed value is missing, not zero.
- Crude rate: Deaths or births divided by the population, with no adjustment for age. A place with older residents has a higher crude death rate even if health at each age is the same.
- Area Deprivation Index (ADI): A score that ranks neighborhoods by income, education, employment and housing conditions drawn from the American Community Survey. In this book a higher national percentile means a more deprived neighborhood.
- Ecological inference: Drawing conclusions about individuals from data about places or groups. A pattern across counties need not hold for the people inside them.
- Modeled estimate: A figure produced by a statistical model instead of a direct count or measurement, usually for small areas where direct data are thin, borrowing strength from surveys, neighboring areas or other data. Neighboring areas can share errors, so single extremes call for caution.

## Methods

This chapter examines which county-years CDC WONDER withholds in its death counts for drug poisoning of any intent, alcohol-induced causes and suicide, 2018 to 2024, across 3,142 counties. A withheld cell is left out of both the numerator and the denominator of a rate and nothing is imputed, and county groups are formed on whether WONDER publishes the count, which depends on the death count itself, so every comparison between groups is a comparison between selected sets of counties. The comparison rate is CDC's modeled county panel for 2019 to 2023, which is mostly modeled in the counties WONDER withholds, so a gap between a WONDER rate and the panel rate sets a direct rate against a model. Connecticut's counties are left out of the group comparisons, and rates are crude, not age-adjusted.

### Which death counts and years are used

County-year counts come from the consolidated county files of CDC WONDER underlying cause of death for 2018 to 2024: drug poisoning of any intent, alcohol-induced causes and suicide, 3,142 counties by year. WONDER publishes a count of zero and counts of 10 or more and withholds counts above zero and below 10; the county population is published for every cell. The county statistics use 2018 to 2023, the window of the chapter on overdose, alcohol and suicide, and the by-year figure uses all seven years, which are final in the June 2026 pull. North Carolina's 2023 overdose counts in these files predate the state's amended death certificates and are low.

A default WONDER query omits county-years with no deaths, so zeros appear only because the pulls asked for them (Show Zero Values).

Explains Table 1.

### How withheld cells and complete counties are handled

A withheld cell is counted and left out of both the numerator and the denominator of a rate. Nothing is imputed, derived from a total or bounded cell by cell; the one bound printed is the largest share of the county files' deaths that withheld cells could hold at 9 each, computed from the public limit and the published cells.

A complete county has a published count and a population in every year 2018 to 2023. The comparisons with the panel use 2019 to 2023 on both sides, the panel's years (its 2024 values are provisional), and call a county complete there when its count is published in each of those years. Connecticut's 8 counties have deaths but no WONDER population in 2022 to 2024 and are left out of the group comparisons and of every county rate.

### County size, groups and characteristics

County size is the mean WONDER population over 2018 to 2023, in classes bounded at 10,000, 25,000, 50,000, 100,000 and 250,000 residents, and the results move with these boundaries.

County groups are formed on whether WONDER publishes the count, which depends on the death count itself, so every comparison between groups is a comparison between selected sets of counties.

County deprivation is the mean of tract Area Deprivation Index national ranks weighted by 2019-2023 tract population (Connecticut tracts assigned to the eight legacy counties), county type is the Rural-Urban Continuum Code (nonmetro from code 4), and race and ethnicity shares are County Health Rankings' figures from Census Population Estimates.

Explains Table 2 and Table 4.

### Counties named in the birth products

Counties named by WONDER natality and linked birth and infant death are the identifiers other than the pooled Unidentified Counties rows, and identifiers with no births of their own, which WONDER has moved into the pooled rows, count as pooled; they (2 in natality, Cape May NJ and Portsmouth VA, and 2 in linked birth and infant death, Hinds MS and Mahoning OH) are set to missing.

Explains Table 8.

### The comparison rate and the bias calculation

The comparison rate is the mean of CDC's annual modeled county rates for 2019 to 2023 from the injury and violence panel, kept only when CDC marks the rate stable in every year, weighted by mean county population over the same years; the panel is modeled where counts are small, and it is mostly modeled in the counties WONDER withholds. The rate pooled over published county-years is deaths over person-years in the published cells only, and a county never published adds nothing to it while it stays in the panel rate.

Bias here means a difference from the rate in a systematic direction. The calculation of the bias of that rate treats a county's yearly deaths as Poisson counts with a constant rate and is a calculation from the rule, not an estimate.

The counts of counties without a value in other sources use County Health Rankings' numerators (a county has no rate when the window holds fewer than 10 deaths, or 20 for premature death) and the panel's stable-rate flag.

Explains Table 5, Table 6 and Table 7.

### Uncertainty, maps and rate type

Means across counties are population-weighted with a 95% interval from a bootstrap that resamples whole counties (2,000 replicates, seed 12345); ratios and pooled rates use shared draws.

Maps of WONDER counts would need at least 20 pooled deaths; the map here counts withheld years and needs no floor. Rates are crude, not age-adjusted.

Explains Table 3.

## What these data cannot show

The comparisons of complete counties with the rest use panel rates that are modeled for the counties WONDER withholds. They show the direction of the selection and cannot give the true rate of a withheld county. The two sources also define overdose differently. The panel counts unintentional and undetermined-intent drug poisonings, and WONDER counts drug poisoning of any intent, so levels differ a little by definition.

The calculation of the bias treats a county's deaths as Poisson counts with a constant rate and independent years. Real counts vary more than that, and rates move over time, so the curve gives direction and rough size and does not correct any estimate in this book. The sizes of county and their boundaries are choices, and the figures move with them.

Withheld counts cannot be recovered. A withheld cell is known only to lie above zero and below the limit, and the bound printed here is a ceiling set by the public limit and the published cells, and it says nothing about any one cell. The group comparisons describe counties and their residents, and County Health Rankings' shares of residents come from Census population estimates with that source's categories. Connecticut's counties are left out of every county comparison because WONDER has no population for them in 2022 and 2023. North Carolina's 2023 overdose counts predate the state's amended death certificates and are low, which enters the 2018 to 2023 window of the overdose counts. A pull that left out WONDER's option to show zero values would turn county-years with no deaths into absent rows, and the below-one ratios in the smallest counties depend on keeping them. The chapter covers three kinds of death. Kinds of death with more or fewer deaths would move the sizes of county at which the rule bites, which the Poisson calculation implies and these data do not test.

## References

1. Klein, Richard J.; Proctor, Suzanne E.; Boudreault, Manon A.; Turczyn, Kathleen M. (2002). Healthy People 2010 criteria for data suppression. Healthy People 2010 Statistical Notes, No. 24 (National Center for Health Statistics). https://www.cdc.gov/nchs/data/statnt/statnt24.pdf
2. Parker, Jennifer D.; Talih, Makram; Irimata, Katherine E.; Zhang, Guangyu; Branum, Amy M.; Davis, Danielle; Das, Barnali; Hamilton, Brady E.; Kochanek, Kenneth D.; McCarty, Frances; Strashny, Alexander (2023). National Center for Health Statistics data presentation standards for rates and counts. Vital and Health Statistics, Series 2, No. 200 (National Center for Health Statistics). https://doi.org/10.15620/cdc:124368
3. Gelman, Andrew; Price, Phillip N. (1999). All maps of parameter estimates are misleading. Statistics in Medicine 18(23): 3221-3234. https://doi.org/10.1002/(sici)1097-0258(19991215)18:23<3221::aid-sim312>3.0.co;2-m
4. Brillinger, David R. (1986). The natural variability of vital rates and associated statistics. Biometrics 42(4): 693-734. https://doi.org/10.2307/2530689
5. Quick, Harrison (2019). Estimating county-level mortality rates using highly censored data from CDC WONDER. Preventing Chronic Disease 16: 180441. https://doi.org/10.5888/pcd16.180441

## Provenance

- Pipeline version: 0.1.0
- Build script: book2/chapters/suppressed-counties-in-death-data/build.py (sha256 0be035700ce4c13573b11f2a73071da7befd9033763e6a0fc6ab4d79c61a6312)
- Built on: 2026-10-07
- Site commit: c2b1ff3309cb4a10833f884da91ac0057a66ad0a

Inputs:

| File | Size (bytes) | SHA-256, first 1 MB | SHA-256, last 1 MB |
| --- | ---: | --- | --- |
| data/analytic/tract_atlas.parquet | 39,659,925 | 29d5e4d943552d54aeb6ba92aadc939d9bd15bdea553c78357dbd516757be627 | 1e10d1ecf722340f23ee2212d4999a90250f85b65fa89b39e32db06d9472e22e |
| data/staged/births/wonder_natality_county_total_2018_2024.parquet | 41,774 | e85d351fa72c022a05415d332185113b6badcaff1119da1d6f45c76ffe16d86d | e85d351fa72c022a05415d332185113b6badcaff1119da1d6f45c76ffe16d86d |
| data/staged/crosswalks/ct_tract2020_to_old_county.parquet | 15,842 | 961ef39f31596ef65fca2c67d5f2f63e533c5dc9b49745d3bb6437844e4b831d | 961ef39f31596ef65fca2c67d5f2f63e533c5dc9b49745d3bb6437844e4b831d |
| data/staged/crosswalks/ers_rucc_county_2023.parquet | 67,278 | ad01d36a53664a2bfafbcbebb0f5a921f5b81122ffa80cf96333db23b41c191e | ad01d36a53664a2bfafbcbebb0f5a921f5b81122ffa80cf96333db23b41c191e |
| data/staged/health/chr_county_2024.parquet | 8,776,933 | 619795f7ecfec8aff10694b649eaee5e421dbeffbe4be47af4161b51ceefa5cd | c69357f1bcf630407bb9716cb445618838999558c019aaaf77d4681b6bf45c56 |
| data/staged/mortality/injury_violence_county.parquet | 722,427 | d0dfd060a3ab9eaa3a87b35d1b24b5a1e9f21af3513c10367edb651c46f40403 | d0dfd060a3ab9eaa3a87b35d1b24b5a1e9f21af3513c10367edb651c46f40403 |
| data/staged/mortality/wonder_lbd_county_total_2018_2022.parquet | 29,903 | 8e0f2604f2be86d6d86c2d206fa4486b2479411c88a6bb9b1595f3d945a4e5b8 | 8e0f2604f2be86d6d86c2d206fa4486b2479411c88a6bb9b1595f3d945a4e5b8 |
| data/staged/mortality/wonder_ucd_alcohol_county_2018_2024.parquet | 289,631 | 478b6b5f8ffa061bcca1df84162d1e169babeb7832bc9818379af9be382a778f | 478b6b5f8ffa061bcca1df84162d1e169babeb7832bc9818379af9be382a778f |
| data/staged/mortality/wonder_ucd_alcohol_state_2018_2024.parquet | 16,162 | 94c821fe0bce60f99f2a300acf8e929b857a8b0f780f5634cadce0698fd665a4 | 94c821fe0bce60f99f2a300acf8e929b857a8b0f780f5634cadce0698fd665a4 |
| data/staged/mortality/wonder_ucd_drug_county_2018_2024.parquet | 308,988 | b0e5cb6dfb44fc1ce53a192faf74cb84277bb5d601a3fc32b65078a7ec03bbf1 | b0e5cb6dfb44fc1ce53a192faf74cb84277bb5d601a3fc32b65078a7ec03bbf1 |
| data/staged/mortality/wonder_ucd_drug_state_2018_2024.parquet | 17,336 | 72a7ecbb0c49a8ba9b90b22f6fad1a6d19c025c256c72799a7acf2369ebabe44 | 72a7ecbb0c49a8ba9b90b22f6fad1a6d19c025c256c72799a7acf2369ebabe44 |
| data/staged/mortality/wonder_ucd_suicide_county_2018_2024.parquet | 290,192 | 448558540a3f42296acf7f66a945f7c6794fd6f4dc9c7032a18a3d105d0541f1 | 448558540a3f42296acf7f66a945f7c6794fd6f4dc9c7032a18a3d105d0541f1 |
| data/staged/mortality/wonder_ucd_suicide_state_2018_2024.parquet | 15,948 | 7588fe51ddda48e57e88852172442b4d2bf64689882e600ebf342668ff79ea1a | 7588fe51ddda48e57e88852172442b4d2bf64689882e600ebf342668ff79ea1a |

Outputs:

| File | SHA-256 |
| --- | --- |
| facts.json | 2e4793e07040c1c46d5b7699ee19be04371b9757c5df30872eb87ef5b0dc8f39 |
| figures/bias_curve.json | a058216f6e2b6b5bc15da34d50d7e44b29bda0ec28b8456fed309097a4738dea |
| figures/naive_vs_panel.json | d368724bd9fbb6b522cf8544f47a314d998cb08c8d09a2b71bbf0c5b3e56c3dc |
| figures/panel_rate_by_status.json | e1d62c22a520b95c1489df6a62a10eb98c2ef49e9bafe7e71e77afa6d166a991 |
| figures/sources_compared.json | ea933eb62096a0d8e18de37c54c6d166e9ba9dd5d43ec120a7de4af0963e4c4b |
| figures/who_lives_in_withheld.json | ef6a8d0f6dcb1a8aa9a9b2d5ba52925bd1a2abc70a569cd15080c520e13a3fac |
| figures/withheld_by_size.json | 14ce2750e0ae465a7193793dfcf22424888e120fca45b4e11c571e182267350f |
| figures/withheld_by_year.json | 30a7742899dbc27a521636508b42c93ae47a85401970c0ba963e29aeefc63609 |
| figures/years_withheld_map.json | 4346f780624c01c337080414444743d98d39e52470293c6675b5c0b909de58b2 |
| meta.json | ccad6a25ac8b151d1ac0a4b407ecfe5996be8b9d9ec7bfd03113484d8b028f59 |
| public/data/suppressed-counties-in-death-data/years_withheld_alc.json | a66b8f4800b86bbfc8fe318c0b7fe686b96b2df67f955abe816b8446b7679c0b |
| public/data/suppressed-counties-in-death-data/years_withheld_od.json | b125033cac1d13d5a54c95df96056d7fdf7bbfe27dfd36f202d1506c336840b6 |
| public/data/suppressed-counties-in-death-data/years_withheld_su.json | 23fa30ccf142b86ff89da916a984d43dcea603b34984863f953cbbdb34f43d39 |
| refs.json | a574352100fa16cf945a0afa913026618387e09606acaa91381fe0a3feed8a00 |

## Machine formats

- Report (HTML): https://vishalsingh.org/zipcode/suppressed-counties-in-death-data/report
- Report (Markdown): https://vishalsingh.org/zipcode/suppressed-counties-in-death-data/report.md
- Every estimate (JSON): https://vishalsingh.org/zipcode/report/suppressed-counties-in-death-data/facts.json
- Table 1, figure "withheld_by_year" (JSON): https://vishalsingh.org/zipcode/report/suppressed-counties-in-death-data/figures/withheld_by_year.json
- Table 2, figure "withheld_by_size" (JSON): https://vishalsingh.org/zipcode/report/suppressed-counties-in-death-data/figures/withheld_by_size.json
- Table 3, figure "years_withheld_map" (JSON): https://vishalsingh.org/zipcode/report/suppressed-counties-in-death-data/figures/years_withheld_map.json
- Table 4, figure "who_lives_in_withheld" (JSON): https://vishalsingh.org/zipcode/report/suppressed-counties-in-death-data/figures/who_lives_in_withheld.json
- Table 5, figure "panel_rate_by_status" (JSON): https://vishalsingh.org/zipcode/report/suppressed-counties-in-death-data/figures/panel_rate_by_status.json
- Table 6, figure "bias_curve" (JSON): https://vishalsingh.org/zipcode/report/suppressed-counties-in-death-data/figures/bias_curve.json
- Table 7, figure "naive_vs_panel" (JSON): https://vishalsingh.org/zipcode/report/suppressed-counties-in-death-data/figures/naive_vs_panel.json
- Table 8, figure "sources_compared" (JSON): https://vishalsingh.org/zipcode/report/suppressed-counties-in-death-data/figures/sources_compared.json
- Payload of "years_withheld_map" (JSON): https://vishalsingh.org/zipcode/data/suppressed-counties-in-death-data/years_withheld_od.json
- Index of every report: https://vishalsingh.org/zipcode/llms.txt; all reports in one file: https://vishalsingh.org/zipcode/llms-full.txt