{
  "alt": "Four panels, one per pollutant, each with five dots and intervals for five sets of controls around a line at zero. Diesel runs from {fact:diesel_raw} with no controls to {fact:diesel_ctrl} with race, income and education, {fact:diesel_dens} with density added, {fact:joint_diesel} with all four exposures in one model, an interval that includes zero, and {fact:diesel_fe} within county. Traffic proximity runs from {fact:traffic_raw} to {fact:traffic_dens} with density, keeps {fact:joint_traffic} in the joint model and is {fact:traffic_fe} within county. Nitrogen dioxide runs from {fact:no2_raw} to {fact:no2_dens} with density and {fact:no2_fe} within county, and its interval includes zero in the joint model. Fine particles run from {fact:pm25_raw} to {fact:pm25_dens} with density, and their intervals include zero once controls are added.",
  "caption": "Diesel, traffic proximity and nitrogen dioxide have intervals above zero in every row except the joint model, and fine particles have an interval that includes zero once controls are added. In the joint row only traffic proximity has an interval above zero. Each panel is one pollutant and each dot is the coefficient from a vote-weighted regression of the tract margin on the standardized exposure, with the controls named on the row; in the joint row the coefficient is that pollutant's in a model with all four. One standard deviation of diesel exposure is {fact:diesel_sd} micrograms per cubic meter.",
  "data": [
    {
      "conf_high": 27.9854,
      "conf_low": 21.5224,
      "estimate": 24.7539,
      "n": 81134,
      "pollutant": "Diesel particulate matter",
      "specification": "No controls"
    },
    {
      "conf_high": 12.5632,
      "conf_low": 8.10701,
      "estimate": 10.3351,
      "n": 81134,
      "pollutant": "Diesel particulate matter",
      "specification": "Race, income, education"
    },
    {
      "conf_high": 7.21005,
      "conf_low": 2.59994,
      "estimate": 4.90499,
      "n": 81134,
      "pollutant": "Diesel particulate matter",
      "specification": "Plus density"
    },
    {
      "conf_high": 4.6391,
      "conf_low": -2.05104,
      "estimate": 1.29403,
      "n": 81134,
      "pollutant": "Diesel particulate matter",
      "specification": "All four together"
    },
    {
      "conf_high": 8.88794,
      "conf_low": 5.26034,
      "estimate": 7.07414,
      "n": 81134,
      "pollutant": "Diesel particulate matter",
      "specification": "County fixed effects"
    },
    {
      "conf_high": 30.2238,
      "conf_low": 19.1857,
      "estimate": 24.7048,
      "n": 81134,
      "pollutant": "Traffic proximity",
      "specification": "No controls"
    },
    {
      "conf_high": 15.3484,
      "conf_low": 9.04161,
      "estimate": 12.195,
      "n": 81134,
      "pollutant": "Traffic proximity",
      "specification": "Race, income, education"
    },
    {
      "conf_high": 11.2904,
      "conf_low": 6.08499,
      "estimate": 8.68768,
      "n": 81134,
      "pollutant": "Traffic proximity",
      "specification": "Plus density"
    },
    {
      "conf_high": 11.7204,
      "conf_low": 5.5867,
      "estimate": 8.65356,
      "n": 81134,
      "pollutant": "Traffic proximity",
      "specification": "All four together"
    },
    {
      "conf_high": 8.56034,
      "conf_low": 4.63502,
      "estimate": 6.59768,
      "n": 81134,
      "pollutant": "Traffic proximity",
      "specification": "County fixed effects"
    },
    {
      "conf_high": 26.2829,
      "conf_low": 22.3458,
      "estimate": 24.3144,
      "n": 81134,
      "pollutant": "Nitrogen dioxide",
      "specification": "No controls"
    },
    {
      "conf_high": 11.8589,
      "conf_low": 8.29078,
      "estimate": 10.0748,
      "n": 81134,
      "pollutant": "Nitrogen dioxide",
      "specification": "Race, income, education"
    },
    {
      "conf_high": 5.80191,
      "conf_low": 1.4632,
      "estimate": 3.63256,
      "n": 81134,
      "pollutant": "Nitrogen dioxide",
      "specification": "Plus density"
    },
    {
      "conf_high": 1.84955,
      "conf_low": -3.40346,
      "estimate": -0.776959,
      "n": 81134,
      "pollutant": "Nitrogen dioxide",
      "specification": "All four together"
    },
    {
      "conf_high": 7.80071,
      "conf_low": 5.16336,
      "estimate": 6.48204,
      "n": 81134,
      "pollutant": "Nitrogen dioxide",
      "specification": "County fixed effects"
    },
    {
      "conf_high": 12.2421,
      "conf_low": 5.26456,
      "estimate": 8.75334,
      "n": 81134,
      "pollutant": "Fine particles (PM2.5)",
      "specification": "No controls"
    },
    {
      "conf_high": 3.0851,
      "conf_low": -1.36804,
      "estimate": 0.858529,
      "n": 81134,
      "pollutant": "Fine particles (PM2.5)",
      "specification": "Race, income, education"
    },
    {
      "conf_high": 1.35929,
      "conf_low": -2.80037,
      "estimate": -0.720538,
      "n": 81134,
      "pollutant": "Fine particles (PM2.5)",
      "specification": "Plus density"
    },
    {
      "conf_high": 0.308902,
      "conf_low": -3.75886,
      "estimate": -1.72498,
      "n": 81134,
      "pollutant": "Fine particles (PM2.5)",
      "specification": "All four together"
    },
    {
      "conf_high": 2.52827,
      "conf_low": -2.42791,
      "estimate": 0.0501788,
      "n": 81134,
      "pollutant": "Fine particles (PM2.5)",
      "specification": "County fixed effects"
    }
  ],
  "id": "pollutant_ladder",
  "notes": [
    "Controls are the white, Black and Hispanic shares of residents, the college share and median household income; density is the log of people per square kilometer. The joint row enters all four exposures in one model with the controls and density.",
    "The county fixed effects row is a different comparison and not a further control: it uses the same controls and density and compares tracts only within their own county.",
    "Intervals treat counties as the units that vary. They do not include error in the allocation of precinct votes to tracts or in the modeled pollution values."
  ],
  "source": "Precinct votes allocated to census tracts (Fekrazad 2025), 2020 presidential election; EPA EJScreen version 2.3 (2024 release, 2020 air data); American Community Survey tract data (vintage not recorded in the atlas); vote-weighted, full counts of tracts, no sampling weights",
  "spec": {
    "colors": {
      "All four together": "total",
      "County fixed effects": "cat4",
      "No controls": "muted",
      "Plus density": "cat2",
      "Race, income, education": "cat1"
    },
    "facet": "pollutant",
    "group": "specification",
    "hi": "conf_high",
    "lo": "conf_low",
    "n": "n",
    "reference": 0,
    "sort": "data",
    "x": "estimate",
    "xDomain": [
      -6,
      32
    ],
    "xFormat": {
      "digits": 0,
      "signed": true
    },
    "xLabel": "Margin per SD of exposure (points)",
    "y": "specification"
  },
  "subtitle": "Difference in the Democratic two-party margin (percentage points) per standard deviation of tract exposure, by pollutant and set of controls; {fact:n_tracts} tracts; 95% intervals clustered by county",
  "title": "Controls shrink the Democratic gap for every pollutant, and with all four exposures in one model traffic proximity keeps {fact:joint_traffic} points while diesel falls to {fact:joint_diesel}",
  "type": "dot",
  "width": "wide"
}