{
  "alt": "Dot chart with three panels and interval lines. In the panel for the gap the changes are {fact:st_a_1992_1994} points for the ANES from 1992 to 1994, {fact:st_g_1993_1994} for the GSS from 1993 to 1994, {fact:st_a_2008_2012} for the ANES from 2008 to 2012 and {fact:st_g_2008_2010} for the GSS from 2008 to 2010. In the panel for the Republican share the changes are {fact:st_a_r_1992_1994}, {fact:st_g_r_1993_1994}, {fact:st_a_r_2008_2012} and {fact:st_g_r_2008_2010} points, all below zero. In the panel for the Democratic share they are {fact:st_a_d_1992_1994}, {fact:st_g_d_1993_1994}, {fact:st_a_d_2008_2012} and {fact:st_g_d_2008_2010}.",
  "caption": "Every Republican dot in the lower panel is below zero with an interval that excludes zero, in both surveys and around both dates. The Democratic share also fell in the ANES rows, and the gap's interval excludes zero only for ANES 2008 to 2012. Dots are later minus earlier estimates from two separate samples and lines are 95% confidence intervals; the rows give the timing of two measurements and do not test a cause.",
  "data": [
    {
      "comparison": "ANES 1992 to 1994",
      "conf_high": 7.29299,
      "conf_low": -6.21592,
      "estimate": 0.53853,
      "n": 3343,
      "panel": "Change in the gap"
    },
    {
      "comparison": "ANES 1992 to 1994",
      "conf_high": -7.20029,
      "conf_low": -17.0068,
      "estimate": -12.1035,
      "n": 1822,
      "panel": "Change in the Democratic share"
    },
    {
      "comparison": "ANES 1992 to 1994",
      "conf_high": -7.99655,
      "conf_low": -17.2876,
      "estimate": -12.6421,
      "n": 1521,
      "panel": "Change in the Republican share"
    },
    {
      "comparison": "GSS 1993 to 1994",
      "conf_high": 15.9751,
      "conf_low": -1.91764,
      "estimate": 7.02871,
      "n": 2536,
      "panel": "Change in the gap"
    },
    {
      "comparison": "GSS 1993 to 1994",
      "conf_high": 4.66277,
      "conf_low": -7.54363,
      "estimate": -1.44043,
      "n": 1383,
      "panel": "Change in the Democratic share"
    },
    {
      "comparison": "GSS 1993 to 1994",
      "conf_high": -1.92788,
      "conf_low": -15.0104,
      "estimate": -8.46914,
      "n": 1153,
      "panel": "Change in the Republican share"
    },
    {
      "comparison": "ANES 2008 to 2012",
      "conf_high": 18.2047,
      "conf_low": 1.40192,
      "estimate": 9.80332,
      "n": 5674,
      "panel": "Change in the gap"
    },
    {
      "comparison": "ANES 2008 to 2012",
      "conf_high": -0.10576,
      "conf_low": -11.2235,
      "estimate": -5.66464,
      "n": 3469,
      "panel": "Change in the Democratic share"
    },
    {
      "comparison": "ANES 2008 to 2012",
      "conf_high": -9.16855,
      "conf_low": -21.7674,
      "estimate": -15.468,
      "n": 2205,
      "panel": "Change in the Republican share"
    },
    {
      "comparison": "GSS 2008 to 2010",
      "conf_high": 17.1471,
      "conf_low": -2.15868,
      "estimate": 7.49421,
      "n": 2153,
      "panel": "Change in the gap"
    },
    {
      "comparison": "GSS 2008 to 2010",
      "conf_high": 0.21214,
      "conf_low": -12.8702,
      "estimate": -6.32903,
      "n": 1296,
      "panel": "Change in the Democratic share"
    },
    {
      "comparison": "GSS 2008 to 2010",
      "conf_high": -6.72455,
      "conf_low": -20.9219,
      "estimate": -13.8232,
      "n": 857,
      "panel": "Change in the Republican share"
    }
  ],
  "id": "dated_changes",
  "notes": [
    "The ANES added web interviews in 2012, so the 2008 to 2012 row mixes a change of mode with any change of opinion; the mode figure splits it. The GSS pairs are the closest GSS surveys on each side of the event dates.",
    "The 2008 ANES asked the item of a subsample of its respondents, which widens the intervals of the 2008 to 2012 row. The GSS 2010 survey was in the field from March to August, and most of its interviews followed the signing of the ACA in late March."
  ],
  "source": "ANES Cumulative Data File, 1992, 1994, 2008 and 2012, and General Social Survey, 1993, 1994, 2008 and 2010; weighted (weight, VCF0009z; weight_use)",
  "spec": {
    "facet": "panel",
    "hi": "conf_high",
    "lo": "conf_low",
    "n": "n",
    "reference": 0,
    "sort": "data",
    "x": "estimate",
    "xDomain": [
      -25,
      25
    ],
    "xFormat": {
      "digits": 1
    },
    "xLabel": "Change (percentage points)",
    "y": "comparison"
  },
  "subtitle": "Change between the surveys that bracket the Clinton plan (1993 to 1994) and the ACA debate and passage (2009 to 2010) in the party gap and in each party's government-side share; leaners included; 95% intervals",
  "table": {
    "columns": [
      {
        "key": "panel",
        "label": "Panel"
      },
      {
        "key": "comparison",
        "label": "Comparison"
      },
      {
        "format": {
          "digits": 1
        },
        "key": "estimate",
        "label": "Change (points)"
      },
      {
        "format": {
          "digits": 1
        },
        "key": "conf_low",
        "label": "Lower"
      },
      {
        "format": {
          "digits": 1
        },
        "key": "conf_high",
        "label": "Upper"
      },
      {
        "format": {
          "thousands": true
        },
        "key": "n",
        "label": "Respondents"
      }
    ]
  },
  "title": "The Republican share on the government side was lower in the survey after each date than in the survey before it",
  "type": "dot",
  "width": "body"
}