{
  "alt": "Dot plot of six slopes with 95% intervals, three for each weighting. All six lie below zero. In each weighting the slope across all counties, the slope within states and the slope between states are close to one another, and the between-states intervals are the widest. The within-state slope is steeper when each county counts once than when counties are weighted by births.",
  "caption": "All six dots lie left of zero, and within each weighting the three slopes are close: the within-state slope differs from the slope across the counties of the same states by {fact:within_minus_pooled_w} points when weighted and {fact:within_minus_pooled_u} points when not, and both intervals include zero. The weighted within-state slope is shallower than the unweighted one by {fact:weight_diff} points. The dots plot least-squares slopes; the within-state slope gives each state its own intercept.",
  "data": [
    {
      "conf_high": -2.71,
      "conf_low": -5.91,
      "estimate": -4.12,
      "n": 49,
      "scope": "Across all counties",
      "weighting": "Weighted by births"
    },
    {
      "conf_high": -2.75,
      "conf_low": -5.53,
      "estimate": -3.88,
      "n": 45,
      "scope": "Within states",
      "weighting": "Weighted by births"
    },
    {
      "conf_high": -1.82,
      "conf_low": -8.33,
      "estimate": -4.71,
      "n": 49,
      "scope": "Between states",
      "weighting": "Weighted by births"
    },
    {
      "conf_high": -4.57,
      "conf_low": -8,
      "estimate": -6.24,
      "n": 49,
      "scope": "Across all counties",
      "weighting": "Each county counts once"
    },
    {
      "conf_high": -5.11,
      "conf_low": -8.2,
      "estimate": -6.63,
      "n": 45,
      "scope": "Within states",
      "weighting": "Each county counts once"
    },
    {
      "conf_high": -2.19,
      "conf_low": -8.69,
      "estimate": -5.29,
      "n": 49,
      "scope": "Between states",
      "weighting": "Each county counts once"
    }
  ],
  "id": "slope_scope",
  "notes": [
    "Large counties are the counties that CDC WONDER names individually: those with {fact:wonder_threshold} 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 {fact:n_named} large counties named in every year from 2018 to 2024, California ({fact:n_ca}) and {fact:n_dropped} counties with a withheld smoking count are left out, which leaves {fact:n_complete}.",
    "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.",
    "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 {fact:n_states_complete} state means, so its interval is the widest; {fact:n_fe_states} states have two or more counties and contribute to the within-state slope.",
    "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",
  "spec": {
    "colors": {
      "Each county counts once": "cat2",
      "Weighted by births": "cat1"
    },
    "group": "weighting",
    "hi": "conf_high",
    "legend": true,
    "lo": "conf_low",
    "n": "n",
    "reference": 0,
    "referenceLabel": "No difference",
    "sort": "data",
    "x": "estimate",
    "xDomain": [
      -10,
      0
    ],
    "xFormat": {
      "digits": 1
    },
    "xLabel": "Percentage points per doubling of county median household income",
    "y": "scope"
  },
  "subtitle": "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",
  "title": "Within a state, a doubling of county income goes with a smoking share {fact:slope_within_w} points lower when counties are weighted by births, and {fact:slope_within_u} points lower when each county counts once",
  "type": "dot",
  "width": "body"
}