{
  "alt": "Dumbbell chart with one row for each of seven traits, joining a dot for the bottom tenth of counties by residual to a dot for the top tenth. The dot for the top tenth lies far to the right for housing built before 1960, somewhat right for the Hispanic share and left for smoking and poverty, and the two dots lie close together for the Black share, ownership and college.",
  "caption": "The open dot is the bottom tenth of counties, below the prediction, and the filled dot is the top tenth. The last column is the top minus bottom difference, and the table under the figure gives it with a 95% interval from draws shared by the two groups: the interval excludes zero for pre-1960 housing, the Hispanic share, poverty and smoking and includes zero for the Black share, ownership and college. Residents of the bottom tenth number {fact:pop_low} and of the top tenth {fact:pop_high}.",
  "data": [
    {
      "bottom": 18.1,
      "diff": 20.1,
      "diff_hi": 29.2,
      "diff_lo": 9.8,
      "hi1": 22.2,
      "hi2": 47.2,
      "lo1": 14.7,
      "lo2": 28.4,
      "n1": 4207,
      "n2": 4512,
      "top": 38.3,
      "trait": "Housing built before 1960 (share of housing units)"
    },
    {
      "bottom": 10.5,
      "diff": 5.2,
      "diff_hi": 9.3,
      "diff_lo": 0.7,
      "hi1": 12.9,
      "hi2": 19.3,
      "lo1": 8.4,
      "lo2": 11.6,
      "n1": 4207,
      "n2": 4512,
      "top": 15.7,
      "trait": "Hispanic residents (share of residents)"
    },
    {
      "bottom": 11.8,
      "diff": -3.2,
      "diff_hi": 0.3,
      "diff_lo": -7.2,
      "hi1": 15,
      "hi2": 10.8,
      "lo1": 9.2,
      "lo2": 6.2,
      "n1": 4207,
      "n2": 4512,
      "top": 8.6,
      "trait": "Black-alone residents (share of residents)"
    },
    {
      "bottom": 15.9,
      "diff": -2.4,
      "diff_hi": -0.9,
      "diff_lo": -3.5,
      "hi1": 16.6,
      "hi2": 14.8,
      "lo1": 15.2,
      "lo2": 12.6,
      "n1": 4207,
      "n2": 4512,
      "top": 13.5,
      "trait": "Adults who currently smoke (crude prevalence, share of adults)"
    },
    {
      "bottom": 70.2,
      "diff": -1.9,
      "diff_hi": 4,
      "diff_lo": -8.4,
      "hi1": 72.8,
      "hi2": 73.2,
      "lo1": 67.4,
      "lo2": 62.3,
      "n1": 4207,
      "n2": 4512,
      "top": 68.2,
      "trait": "Owner-occupied homes (share of households)"
    },
    {
      "bottom": 33.9,
      "diff": 3.6,
      "diff_hi": 9.2,
      "diff_lo": -2.5,
      "hi1": 37.1,
      "hi2": 41.9,
      "lo1": 30.6,
      "lo2": 32.3,
      "n1": 4207,
      "n2": 4512,
      "top": 37.6,
      "trait": "Adults with a bachelor's degree or more (share of adults)"
    },
    {
      "bottom": 12.6,
      "diff": -1.9,
      "diff_hi": -0.2,
      "diff_lo": -3.4,
      "hi1": 13.4,
      "hi2": 12.2,
      "lo1": 11.9,
      "lo2": 9.4,
      "n1": 4207,
      "n2": 4512,
      "top": 10.7,
      "trait": "Residents below the poverty line (share of residents)"
    }
  ],
  "id": "overperformer_traits",
  "notes": [
    "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.",
    "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.",
    "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",
  "spec": {
    "deltaLabel": "Top minus bottom (points)",
    "hi1": "hi1",
    "hi2": "hi2",
    "label1": "Bottom tenth of counties (below the prediction)",
    "label2": "Top tenth of counties (above the prediction)",
    "lo1": "lo1",
    "lo2": "lo2",
    "n1": "n1",
    "n2": "n2",
    "sort": "abs-change",
    "x1": "bottom",
    "x2": "top",
    "xDomain": [
      0,
      80
    ],
    "xFormat": {
      "digits": 1
    },
    "xLabel": "Share (percent)",
    "y": "trait"
  },
  "subtitle": "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",
  "table": {
    "columns": [
      {
        "key": "trait",
        "label": "Trait"
      },
      {
        "format": {
          "digits": 1
        },
        "key": "bottom",
        "label": "Bottom tenth"
      },
      {
        "format": {
          "digits": 1
        },
        "key": "top",
        "label": "Top tenth"
      },
      {
        "format": {
          "digits": 1,
          "signed": true
        },
        "key": "diff",
        "label": "Top minus bottom (points)"
      },
      {
        "format": {
          "digits": 1,
          "signed": true
        },
        "key": "diff_lo",
        "label": "Lower limit of the 95% interval"
      },
      {
        "format": {
          "digits": 1,
          "signed": true
        },
        "key": "diff_hi",
        "label": "Upper limit of the 95% interval"
      }
    ]
  },
  "title": "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",
  "type": "dumbbell",
  "width": "wide"
}