{
  "alt": "Line chart of three series across six pooled decades. The observed cross-pressured share falls from {fact:obs_1970} in the 1970s to {fact:obs_2020} in the 2020s. The series with cohorts frozen at 1970s rates stays flat and ends at {fact:cfa_2020}. The series frozen at entry rates stays level through the 1990s and then falls to {fact:cfb_2020}, above the observed line.",
  "caption": "The observed line falls from {fact:obs_1970} to {fact:obs_2020}, the line with cohorts frozen at their entry rates (a cohort's rate in its first well-observed decade) ends at {fact:cfb_2020}, and the line frozen at 1970s rates stays near its start. The mix of cohorts accounts for {fact:share_b} of the fall under the entry-rate version and {fact:share_a} of it under the 1970s-rate version, which is within sampling error of zero. Both versions apply the observed cohort mix of each decade to fixed cohort rates and are accounting exercises, not fitted models.",
  "data": [
    {
      "band_high": 19.6498,
      "band_low": 17.2727,
      "conf_high": 19.6498,
      "conf_low": 17.2727,
      "estimate": 18.4578,
      "n": 5059,
      "period": "1972-1978",
      "series": "Observed",
      "year": 1975
    },
    {
      "band_high": 19.2837,
      "band_low": 17.4066,
      "conf_high": 19.2837,
      "conf_low": 17.4066,
      "estimate": 18.3712,
      "n": 5808,
      "period": "1980-1988",
      "series": "Observed",
      "year": 1984
    },
    {
      "band_high": 17.0888,
      "band_low": 15.227,
      "conf_high": 17.0888,
      "conf_low": 15.227,
      "estimate": 16.1891,
      "n": 6268,
      "period": "1990-1998",
      "series": "Observed",
      "year": 1994
    },
    {
      "band_high": 14.1713,
      "band_low": 11.7591,
      "conf_high": 14.1713,
      "conf_low": 11.7591,
      "estimate": 12.9332,
      "n": 4034,
      "period": "2000-2008",
      "series": "Observed",
      "year": 2004
    },
    {
      "band_high": 9.7239,
      "band_low": 8.02016,
      "conf_high": 9.7239,
      "conf_low": 8.02016,
      "estimate": 8.83387,
      "n": 7451,
      "period": "2012-2016",
      "series": "Observed",
      "year": 2014
    },
    {
      "band_high": 6.28912,
      "band_low": 5.01138,
      "conf_high": 6.28912,
      "conf_low": 5.01138,
      "estimate": 5.62179,
      "n": 10489,
      "period": "2020-2024",
      "series": "Observed",
      "year": 2022
    },
    {
      "band_high": null,
      "band_low": null,
      "conf_high": 19.4677,
      "conf_low": 17.0513,
      "estimate": 18.275,
      "n": 5059,
      "period": "1972-1978",
      "series": "1970s rates",
      "year": 1975
    },
    {
      "band_high": null,
      "band_low": null,
      "conf_high": 19.4062,
      "conf_low": 16.9426,
      "estimate": 18.1756,
      "n": 5808,
      "period": "1980-1988",
      "series": "1970s rates",
      "year": 1984
    },
    {
      "band_high": null,
      "band_low": null,
      "conf_high": 19.4064,
      "conf_low": 16.9618,
      "estimate": 18.201,
      "n": 6268,
      "period": "1990-1998",
      "series": "1970s rates",
      "year": 1994
    },
    {
      "band_high": null,
      "band_low": null,
      "conf_high": 19.4753,
      "conf_low": 17.0229,
      "estimate": 18.2578,
      "n": 4034,
      "period": "2000-2008",
      "series": "1970s rates",
      "year": 2004
    },
    {
      "band_high": null,
      "band_low": null,
      "conf_high": 19.4271,
      "conf_low": 17.0366,
      "estimate": 18.2725,
      "n": 7451,
      "period": "2012-2016",
      "series": "1970s rates",
      "year": 2014
    },
    {
      "band_high": null,
      "band_low": null,
      "conf_high": 19.5075,
      "conf_low": 17.0989,
      "estimate": 18.3277,
      "n": 10489,
      "period": "2020-2024",
      "series": "1970s rates",
      "year": 2022
    },
    {
      "band_high": 19.4706,
      "band_low": 17.0756,
      "conf_high": 19.4706,
      "conf_low": 17.0756,
      "estimate": 18.2939,
      "n": 5059,
      "period": "1972-1978",
      "series": "Entry rates",
      "year": 1975
    },
    {
      "band_high": 19.6833,
      "band_low": 17.49,
      "conf_high": 19.6833,
      "conf_low": 17.49,
      "estimate": 18.6088,
      "n": 5808,
      "period": "1980-1988",
      "series": "Entry rates",
      "year": 1984
    },
    {
      "band_high": 20.0148,
      "band_low": 17.6139,
      "conf_high": 20.0148,
      "conf_low": 17.6139,
      "estimate": 18.8397,
      "n": 6268,
      "period": "1990-1998",
      "series": "Entry rates",
      "year": 1994
    },
    {
      "band_high": 19.0817,
      "band_low": 16.0689,
      "conf_high": 19.0817,
      "conf_low": 16.0689,
      "estimate": 17.5421,
      "n": 4034,
      "period": "2000-2008",
      "series": "Entry rates",
      "year": 2004
    },
    {
      "band_high": 17.9049,
      "band_low": 14.6236,
      "conf_high": 17.9049,
      "conf_low": 14.6236,
      "estimate": 16.2012,
      "n": 7451,
      "period": "2012-2016",
      "series": "Entry rates",
      "year": 2014
    },
    {
      "band_high": 16.7337,
      "band_low": 13.4929,
      "conf_high": 16.7337,
      "conf_low": 13.4929,
      "estimate": 15.0551,
      "n": 10489,
      "period": "2020-2024",
      "series": "Entry rates",
      "year": 2022
    }
  ],
  "id": "frozen_cohorts",
  "notes": [
    "Placed partisans with no birth cohort in the file are left out of all three series so that the accounting is exact.",
    "Cohorts not observed with {fact:floor_n} or more placed partisans in the 1970s take the mean 1970s cohort rate of {fact:avg70} in the first counterfactual.",
    "The within-cohort part of the fall is the observed fall minus the mix part, so it includes a baseline residual: the frozen rates applied to the 1970s mix predict a 1970s share {fact:resid_a} points (1970s rates) and {fact:resid_b} points (entry rates) below the observed one.",
    "The bootstrap re-draws respondents with replacement within each pooled decade many times and re-estimates every quantity each time; the intervals are the 2.5th and 97.5th percentiles.",
    "Bands are drawn for the observed and entry-rate lines; the interval of the 1970s-rate line is in the data table. Differences are computed from unrounded values, so a printed difference can differ in the last digit from the difference of the printed shares."
  ],
  "source": "ANES Cumulative Data File, 1972-2024; weighted (VCF0009z); placed partisans, leaners included; intervals from a bootstrap over respondents within pooled decades (1,000 replicates, seed 12345)",
  "spec": {
    "colors": {
      "1970s rates": "muted",
      "Entry rates": "accent",
      "Observed": "highlight"
    },
    "dashed": [
      "1970s rates",
      "Entry rates"
    ],
    "hi": "band_high",
    "lo": "band_low",
    "n": "n",
    "series": "series",
    "x": "year",
    "xDomain": [
      1970,
      2026
    ],
    "xLabel": "Survey years pooled by decade (plotted at the midpoint)",
    "y": "estimate",
    "yDomain": [
      0,
      25
    ],
    "yFormat": {
      "digits": 1,
      "suffix": "%"
    },
    "yLabel": "Share of placed partisans (%)"
  },
  "subtitle": "Cross-pressured share of placed partisans, observed and under two frozen-rate counterfactuals, ANES survey years pooled by decade, 1972-2024, with 95% bootstrap intervals",
  "table": {
    "columns": [
      {
        "key": "series",
        "label": "Series"
      },
      {
        "key": "period",
        "label": "Survey years pooled"
      },
      {
        "format": {
          "digits": 1
        },
        "key": "estimate",
        "label": "Share (%)"
      },
      {
        "format": {
          "digits": 1
        },
        "key": "conf_low",
        "label": "Interval low"
      },
      {
        "format": {
          "digits": 1
        },
        "key": "conf_high",
        "label": "Interval high"
      },
      {
        "key": "n",
        "label": "Placed partisans"
      }
    ]
  },
  "title": "Freezing cohort rates predicts a 2020s cross-pressured share of {fact:cfb_2020} to {fact:cfa_2020}, against {fact:obs_2020} observed",
  "type": "line",
  "width": "wide"
}