Chapter 3.1

The two coalitions

Voters with a four-year degree rose from 19.2% to 43.4% of Democratic voters between the elections of 1976 to 1988 and those of 2012 to 2024, and from 25.2% to 31.4% of Republican voters, and a group's share of a party's voters can move opposite to its share voting for the party.

Author: Vishal Singh
Affiliation: NYU Stern
Published: 3 October 2026

DraftFigures and numbers may change as the analysis is rebuilt.what is still being checked

The ANES and the CES disagree on the degree-holding share of Republican voters after 2008; the chapter reports both and has not resolved the difference.

In the American National Election Studies (ANES), voters with a four-year degree made up 19.2% (95% CI 17.5% to 20.9%) of Democratic voters in the presidential elections of 1976 to 1988 and 43.4% (95% CI 42.0% to 44.8%) of those in 2012 to 2024. Among Republican voters the share went from 25.2% (95% CI 23.5% to 26.9%) to 31.4% (95% CI 29.9% to 32.8%). White non-Hispanic voters went from 72.7% (95% CI 70.8% to 74.6%) to 58.0% (95% CI 56.6% to 59.4%) of Democratic voters and from 94.8% (95% CI 93.9% to 95.7%) to 85.9% (95% CI 84.8% to 87.0%) of Republican voters, so both coalitions became more diverse and the Democratic coalition changed by the larger amount.

Those are shares of a party's voters. They answer a different question from the one about how likely a group is to vote for the party, and a change in one does not identify a change in the other. White voters were 39.4% (95% CI 37.8% to 40.9%) Democratic in 1976 to 1988 and 44.4% (95% CI 43.2% to 45.6%) in 2012 to 2024, a rise, while their share of Democratic voters fell by 14.7 points. The chapter keeps two denominators apart throughout: the share of a group that votes Democratic, where the group is the denominator, and the share of a party's voters who belong to the group, where the party is the denominator. It also keeps apart four populations that are easy to blur: eligible adults, voters, party identifiers and nonvoters.

Three surveys carry the evidence. The ANES Cumulative Data File has presidential vote and demographics for every election from 1952 to 2024, with 34,423 respondents in 19 elections who reported voting for the Democratic or the Republican candidate. The Cooperative Election Study (CES), an online panel, has 188,145 such respondents from 2008 to 2024 and finer religious and racial categories. The Current Population Survey (CPS) voting supplement records no vote choice, and the chapter uses it for one purpose: the make-up of citizens aged 18 or older, who are the eligible adults here, and of the voters among them, in the presidential elections from 1996 to 2024.Every interval is a 95% confidence interval. Shares use the Kish effective sample size [1] and treat the weighted sample as a simple random sample, which leaves out the stages of each sample design. The CES is an opt-in panel and the CPS extract carries no replicate weights, so their intervals are narrower than a design-based interval would be. The terms of the two decompositions come from a seeded bootstrap, described in the methods note. A voter in the ANES and the CES is a respondent who reported voting for one of the two major-party candidates; third-party voters and nonvoters are outside that frame, which is why a party's group shares sum to a whole. A party identifier is a respondent who calls themselves a Democrat or a Republican and counts leaners.

In the 2024 CES, 43.5% (95% CI 42.7% to 44.4%) of white voters supported the Democratic candidate, and white voters made up 64.3% (95% CI 63.3% to 65.3%) of Democratic voters. Black voters supported the Democratic candidate at 81.9% (95% CI 79.8% to 84.0%), the highest rate of any racial or ethnic group, and made up 15.6% (95% CI 14.8% to 16.3%) of Democratic voters. The first group leans Republican and is a majority of the coalition because it is 72.3% of all voters. The second is overwhelmingly Democratic and a small part of the coalition because it is 9.3% of all voters. Figure 1 shows both numbers for every group in the CES.

Figure 1

In 2024 white voters leaned Republican and still made up most of the Democratic coalition

Two shares for each group of two-party presidential voters in 2024: the Democratic share of the group's voters and the group's share of all Democratic voters, percent; 95% intervals from the Kish effective sample size

Democratic share of the group's voters
Group's share of Democratic voters

Education

Race and ethnicity

Religion

Age

Gender

Percent of two-party voters

Show data

Data for In 2024 white voters leaned Republican and still made up most of the Democratic coalition
Dimension	Group	Share	Estimate (%)	95% CI low	95% CI high	n voters in the denominator	Group's share of all voters (%)	Group's share of Republican voters (%)
Education	Postgraduate degree	Democratic share of the group's voters	61.9	60.3	63.4	6,565	16.0	12.0
Education	Postgraduate degree	Group's share of Democratic voters	20.3	19.4	21.1	23,517	16.0	12.0
Education	College graduate	Democratic share of the group's voters	52.6	51.2	53.9	10,429	25.8	24.0
Education	College graduate	Group's share of Democratic voters	27.7	26.7	28.6	23,517	25.8	24.0
Education	Some college	Democratic share of the group's voters	47.2	45.8	48.5	14,759	31.1	32.2
Education	Some college	Group's share of Democratic voters	30.0	29.0	30.9	23,517	31.1	32.2
Education	High school or less	Democratic share of the group's voters	40.0	38.3	41.6	10,255	27.1	31.8
Education	High school or less	Group's share of Democratic voters	22.1	21.2	22.9	23,517	27.1	31.8
Race and ethnicity	White	Democratic share of the group's voters	43.5	42.7	44.4	31,857	72.3	80.0
Race and ethnicity	White	Group's share of Democratic voters	64.3	63.3	65.3	23,517	72.3	80.0
Race and ethnicity	Black	Democratic share of the group's voters	81.9	79.8	84.0	3,753	9.3	3.3
Race and ethnicity	Black	Group's share of Democratic voters	15.6	14.8	16.3	23,517	9.3	3.3
Race and ethnicity	Hispanic	Democratic share of the group's voters	54.7	51.9	57.6	3,564	10.4	9.2
Race and ethnicity	Hispanic	Group's share of Democratic voters	11.6	11.0	12.3	23,517	10.4	9.2
Race and ethnicity	Asian	Democratic share of the group's voters	63.0	58.6	67.4	1,065	3.7	2.6
Race and ethnicity	Asian	Group's share of Democratic voters	4.7	4.3	5.1	23,517	3.7	2.6
Race and ethnicity	Other	Democratic share of the group's voters	43.0	39.1	47.0	1,769	4.3	4.9
Race and ethnicity	Other	Group's share of Democratic voters	3.8	3.4	4.2	23,517	4.3	4.9
Religion	Protestant	Democratic share of the group's voters	32.9	31.8	34.1	14,201	36.9	48.5
Religion	Protestant	Group's share of Democratic voters	24.8	23.9	25.7	23,506	36.9	48.5
Religion	Catholic	Democratic share of the group's voters	43.3	41.6	45.0	7,973	18.5	20.6
Religion	Catholic	Group's share of Democratic voters	16.4	15.6	17.1	23,506	18.5	20.6
Religion	Jewish	Democratic share of the group's voters	68.0	64.0	72.1	1,219	2.7	1.7
Religion	Jewish	Group's share of Democratic voters	3.7	3.3	4.1	23,506	2.7	1.7
Religion	Other religion	Democratic share of the group's voters	45.7	43.1	48.2	4,287	10.9	11.5
Religion	Other religion	Group's share of Democratic voters	10.1	9.5	10.7	23,506	10.9	11.5
Religion	No religion	Democratic share of the group's voters	70.9	69.7	72.2	14,311	31.0	17.7
Religion	No religion	Group's share of Democratic voters	45.0	43.9	46.0	23,506	31.0	17.7
Age	18 to 29	Democratic share of the group's voters	60.6	57.6	63.6	3,082	11.7	9.1
Age	18 to 29	Group's share of Democratic voters	14.5	13.8	15.3	23,517	11.7	9.1
Age	30 to 44	Democratic share of the group's voters	55.1	53.5	56.8	7,073	23.1	20.3
Age	30 to 44	Group's share of Democratic voters	26.1	25.1	27.0	23,517	23.1	20.3
Age	45 to 64	Democratic share of the group's voters	43.9	42.8	45.1	16,848	35.3	38.8
Age	45 to 64	Group's share of Democratic voters	31.7	30.7	32.7	23,517	35.3	38.8
Age	65 and over	Democratic share of the group's voters	45.5	44.4	46.7	15,005	29.8	31.8
Age	65 and over	Group's share of Democratic voters	27.7	26.8	28.6	23,517	29.8	31.8
Gender	Women	Democratic share of the group's voters	50.8	49.8	51.8	21,759	50.7	48.7
Gender	Women	Group's share of Democratic voters	52.9	51.8	53.9	23,300	50.7	48.7
Gender	Men	Democratic share of the group's voters	46.6	45.5	47.8	19,998	49.3	51.3
Gender	Men	Group's share of Democratic voters	47.1	46.1	48.2	23,300	49.3	51.3

The two dots in a row answer different questions and can sit far apart. 81.9% of Black voters voted Democratic, and Black voters were 15.6% of Democratic voters; 43.5% of white voters voted Democratic, and white voters were 64.3% of Democratic voters. The first dot divides by the group's voters and the second by the party's voters, and the group's share of all voters links them. The table gives that share and the group's share of Republican voters.Note: The CES is an opt-in online panel, so its intervals are model based and do not cover how respondents were recruited.Source: Cooperative Election Study cumulative file, 2024; weighted (weight_cumulative); two-party presidential voters

For a group $g$, the share of Democratic voters who belong to it is

$c_D(g) = s(g)\,\frac{d(g)}{D},$

where $s(g)$ is the group's share of all voters, $d(g)$ is the share of the group's voters who vote Democratic and $D$ is the Democratic share of all voters. The ratio $d(g)/D$ is the group's relative lean. The share of Republican voters who belong to the group is the same product with Republican shares in place of Democratic ones. Axelrod's account of the party coalitions from 1952 to 1968 used the same arithmetic, breaking a group's contribution to a party's vote into its size, its turnout and its loyalty [2]. A group can become a larger part of a coalition because it grew, because it leans further toward the party, or both, and its share of the coalition can fall while its Democratic share rises.

A rise in the Democratic share among degree holders therefore says nothing by itself about how many Democratic voters hold degrees, and a rise in the degree holders' share of Democratic voters says nothing about whether any voter changed party.

Eligible adults, voters, party identifiers and nonvoters differ in make-up

Eligible adults are citizens aged 18 or older. Voters are the eligible adults who voted, party identifiers are people who identify with a party whether or not they vote, and nonvoters are eligible adults who did not vote.The definition of eligible adults here excludes noncitizens, whom the CPS identifies from 1994, and keeps everyone else, including adults who are barred from voting for other reasons [3]. The CPS counts a respondent who gave no turnout answer as a nonvoter, the Census convention, which Hur and Achen examine [4]. Turnout is self-reported in all three surveys. In the CPS, 37.1% (95% CI 36.7% to 37.5%) of eligible adults held a four-year degree in 2024 and 44.9% (95% CI 44.4% to 45.4%) of voters did. In every presidential election from 1996 to 2024 voters were more likely than eligible adults to hold a degree and less likely to be in the youngest age band, ages 18 to 29 (Figure 2). The degree-holding share of eligible adults rose from 22.3% (95% CI 22.0% to 22.6%) in 1996 to 37.1% (95% CI 36.7% to 37.5%) in 2024 and that of voters from 29.4% (95% CI 29.0% to 29.8%) to 44.9% (95% CI 44.4% to 45.4%), a rise of 15.5 (95% CI 14.8 to 16.2) points. A gap between voters and eligible adults at one date is a snapshot and does not split that rise, so the chapter splits it. Had degree holders and other adults kept their 1996 turnout rates, degree holders would have been 46.1% (95% CI 45.5% to 46.7%) of 2024 voters. On that counterfactual the growth of the group among eligible adults accounts for +16.7 (95% CI 16.1 to 17.3) points of the rise and the change in relative turnout for −1.2 (95% CI −1.6 to −0.8) points. Chapter 3.7, Turnout reports turnout rates by group.

Figure 2

In every presidential election since 1996 voters were more likely than eligible adults to hold a four-year degree and less likely to be in the youngest age band

Share of eligible adults (citizens aged 18 or older) and share of voters in each group, percent, by presidential election; CPS voting supplement; 95% intervals from the Kish effective sample size

Eligible adults
Voters

Show data

Data for In every presidential election since 1996 voters were more likely than eligible adults to hold a four-year degree and less likely to be in the youngest age band
Election	Dimension	Group	Eligible adults (%)	95% CI low	95% CI high	n eligible	Voters (%)	95% CI low	95% CI high	n voters	Voters minus eligible (points)	95% CI low of difference	95% CI high of difference
1996	Education	Less than high school	16.2	15.9	16.4	84,090	10.7	10.4	11.0	50,025	−5.4	−5.7	−5.2
1996	Education	High school graduate	34.4	34.1	34.8	84,090	30.5	30.0	30.9	50,025	−3.9	−4.2	−3.6
1996	Education	Some college	27.1	26.8	27.5	84,090	29.4	28.9	29.8	50,025	+2.2	+1.9	+2.5
1996	Education	Four-year degree or more	22.3	22.0	22.6	84,090	29.4	29.0	29.8	50,025	+7.1	+6.9	+7.4
1996	Race and ethnicity	White non-Hispanic	79.2	78.9	79.6	84,090	82.5	82.1	82.8	50,025	+3.2	+2.9	+3.5
1996	Race and ethnicity	Black non-Hispanic	11.7	11.5	11.9	84,090	10.6	10.3	10.9	50,025	−1.1	−1.3	−0.8
1996	Race and ethnicity	Hispanic	6.2	6.0	6.4	84,090	4.7	4.5	4.9	50,025	−1.5	−1.7	−1.4
1996	Race and ethnicity	Asian or Pacific Islander	2.1	2.0	2.2	84,090	1.6	1.5	1.7	50,025	−0.4	−0.5	−0.4
1996	Race and ethnicity	Other or multiple races	0.8	0.7	0.8	84,090	0.6	0.5	0.7	50,025	−0.2	−0.2	−0.1
1996	Age	18 to 29	22.0	21.7	22.3	84,090	14.9	14.6	15.2	50,025	−7.1	−7.4	−6.8
1996	Age	30 to 44	32.6	32.3	33.0	84,090	31.8	31.3	32.3	50,025	−0.8	−1.1	−0.5
1996	Age	45 to 64	28.2	27.9	28.6	84,090	33.0	32.5	33.4	50,025	+4.7	+4.5	+5.0
1996	Age	65 and over	17.2	16.9	17.4	84,090	20.3	19.9	20.7	50,025	+3.2	+2.9	+3.4
1996	Gender	Women	52.3	52.0	52.7	84,090	53.4	52.9	53.9	50,025	+1.1	+0.8	+1.4
1996	Gender	Men	47.7	47.3	48.0	84,090	46.6	46.1	47.1	50,025	−1.1	−1.4	−0.8
2000	Education	Less than high school	14.3	14.1	14.6	82,428	9.3	9.0	9.5	50,010	−5.1	−5.3	−4.8
2000	Education	High school graduate	33.6	33.2	33.9	82,428	29.7	29.2	30.1	50,010	−3.9	−4.2	−3.6
2000	Education	Some college	28.2	27.8	28.5	82,428	29.9	29.5	30.4	50,010	+1.8	+1.5	+2.0
2000	Education	Four-year degree or more	23.9	23.6	24.3	82,428	31.2	30.7	31.6	50,010	+7.2	+7.0	+7.5
2000	Race and ethnicity	White non-Hispanic	77.3	77.0	77.6	82,428	80.6	80.2	81.0	50,010	+3.3	+3.0	+3.6
2000	Race and ethnicity	Black non-Hispanic	11.7	11.4	11.9	82,428	11.2	10.9	11.5	50,010	−0.5	−0.7	−0.3
2000	Race and ethnicity	Hispanic	7.4	7.2	7.6	82,428	5.5	5.3	5.7	50,010	−1.9	−2.1	−1.7
2000	Race and ethnicity	Asian or Pacific Islander	2.8	2.6	2.9	82,428	2.0	1.9	2.1	50,010	−0.8	−0.9	−0.6
2000	Race and ethnicity	Other or multiple races	0.9	0.8	1.0	82,428	0.7	0.6	0.8	50,010	−0.2	−0.2	−0.1
2000	Age	18 to 29	20.8	20.5	21.1	82,428	14.1	13.7	14.4	50,010	−6.8	−7.0	−6.5
2000	Age	30 to 44	30.4	30.1	30.8	82,428	29.8	29.4	30.3	50,010	−0.6	−0.9	−0.3
2000	Age	45 to 64	31.5	31.1	31.8	82,428	35.8	35.4	36.3	50,010	+4.4	+4.1	+4.7
2000	Age	65 and over	17.3	17.0	17.6	82,428	20.3	19.9	20.6	50,010	+2.9	+2.7	+3.2
2000	Gender	Women	52.5	52.1	52.9	82,428	53.6	53.1	54.1	50,010	+1.1	+0.8	+1.5
2000	Gender	Men	47.5	47.1	47.9	82,428	46.4	45.9	46.9	50,010	−1.1	−1.5	−0.8
2004	Education	Less than high school	13.0	12.8	13.3	96,452	8.1	7.8	8.3	63,052	−5.0	−5.2	−4.8
2004	Education	High school graduate	32.3	32.0	32.7	96,452	28.5	28.1	29.0	63,052	−3.8	−4.0	−3.5
2004	Education	Some college	28.7	28.3	29.0	96,452	31.0	30.5	31.4	63,052	+2.3	+2.0	+2.5
2004	Education	Four-year degree or more	26.0	25.6	26.3	96,452	32.4	32.0	32.9	63,052	+6.5	+6.2	+6.7
2004	Race and ethnicity	White non-Hispanic	75.2	74.9	75.5	96,452	79.2	78.8	79.6	63,052	+4.0	+3.7	+4.2
2004	Race and ethnicity	Black non-Hispanic	11.6	11.4	11.8	96,452	11.0	10.7	11.3	63,052	−0.6	−0.8	−0.4
2004	Race and ethnicity	Hispanic	8.2	8.0	8.4	96,452	6.0	5.8	6.2	63,052	−2.1	−2.3	−2.0
2004	Race and ethnicity	Asian or Pacific Islander	3.3	3.2	3.5	96,452	2.3	2.2	2.5	63,052	−1.0	−1.1	−0.9
2004	Race and ethnicity	Other or multiple races	1.7	1.6	1.8	96,452	1.5	1.4	1.6	63,052	−0.2	−0.3	−0.2
2004	Age	18 to 29	20.9	20.6	21.2	96,452	16.0	15.7	16.3	63,052	−4.8	−5.1	−4.6
2004	Age	30 to 44	27.9	27.6	28.3	96,452	27.3	26.9	27.7	63,052	−0.6	−0.9	−0.4
2004	Age	45 to 64	34.1	33.8	34.4	96,452	37.6	37.2	38.1	63,052	+3.5	+3.3	+3.8
2004	Age	65 and over	17.1	16.8	17.4	96,452	19.0	18.7	19.4	63,052	+1.9	+1.7	+2.1
2004	Gender	Women	52.2	51.8	52.6	96,452	53.5	53.1	54.0	63,052	+1.3	+1.0	+1.6
2004	Gender	Men	47.8	47.4	48.2	96,452	46.5	46.0	46.9	63,052	−1.3	−1.6	−1.0
2008	Education	Less than high school	11.2	10.9	11.4	92,360	6.9	6.7	7.1	59,975	−4.3	−4.5	−4.1
2008	Education	High school graduate	31.7	31.4	32.1	92,360	27.3	26.9	27.8	59,975	−4.4	−4.6	−4.1
2008	Education	Some college	29.6	29.3	29.9	92,360	31.6	31.2	32.1	59,975	+2.0	+1.8	+2.3
2008	Education	Four-year degree or more	27.5	27.2	27.9	92,360	34.1	33.7	34.6	59,975	+6.6	+6.4	+6.8
2008	Race and ethnicity	White non-Hispanic	73.4	73.1	73.8	92,360	76.3	75.9	76.7	59,975	+2.9	+2.6	+3.1
2008	Race and ethnicity	Black non-Hispanic	11.8	11.6	12.0	92,360	12.1	11.8	12.4	59,975	+0.3	+0.1	+0.5
2008	Race and ethnicity	Hispanic	9.5	9.3	9.7	92,360	7.4	7.2	7.7	59,975	−2.0	−2.2	−1.9
2008	Race and ethnicity	Asian or Pacific Islander	3.6	3.5	3.7	92,360	2.7	2.5	2.8	59,975	−0.9	−1.0	−0.8
2008	Race and ethnicity	Other or multiple races	1.7	1.6	1.8	92,360	1.5	1.4	1.6	59,975	−0.2	−0.2	−0.1
2008	Age	18 to 29	21.3	21.0	21.6	92,360	17.1	16.7	17.4	59,975	−4.2	−4.5	−4.0
2008	Age	30 to 44	25.5	25.2	25.8	92,360	24.8	24.4	25.2	59,975	−0.7	−1.0	−0.5
2008	Age	45 to 64	35.6	35.3	36.0	92,360	38.7	38.2	39.1	59,975	+3.1	+2.8	+3.3
2008	Age	65 and over	17.6	17.3	17.9	92,360	19.5	19.1	19.8	59,975	+1.8	+1.6	+2.0
2008	Gender	Women	52.0	51.7	52.4	92,360	53.7	53.2	54.1	59,975	+1.6	+1.4	+1.9
2008	Gender	Men	48.0	47.6	48.3	92,360	46.3	45.9	46.8	59,975	−1.6	−1.9	−1.4
2012	Education	Less than high school	10.2	9.9	10.4	94,311	6.2	6.0	6.5	59,597	−3.9	−4.1	−3.7
2012	Education	High school graduate	30.4	30.1	30.7	94,311	25.9	25.5	26.3	59,597	−4.5	−4.8	−4.3
2012	Education	Some college	30.1	29.8	30.5	94,311	31.3	30.9	31.7	59,597	+1.2	+0.9	+1.4
2012	Education	Four-year degree or more	29.3	29.0	29.7	94,311	36.6	36.2	37.0	59,597	+7.3	+7.0	+7.5
2012	Race and ethnicity	White non-Hispanic	71.1	70.7	71.4	94,311	73.7	73.3	74.1	59,597	+2.7	+2.4	+3.0
2012	Race and ethnicity	Black non-Hispanic	12.0	11.7	12.2	94,311	12.9	12.6	13.2	59,597	+0.9	+0.7	+1.1
2012	Race and ethnicity	Hispanic	10.8	10.6	11.1	94,311	8.4	8.2	8.7	59,597	−2.4	−2.6	−2.2
2012	Race and ethnicity	Asian or Pacific Islander	4.0	3.9	4.2	94,311	3.1	2.9	3.2	59,597	−1.0	−1.1	−0.8
2012	Race and ethnicity	Other or multiple races	2.1	2.0	2.2	94,311	1.9	1.7	2.0	59,597	−0.2	−0.3	−0.1
2012	Age	18 to 29	21.2	20.9	21.5	94,311	15.4	15.1	15.8	59,597	−5.8	−6.0	−5.5
2012	Age	30 to 44	24.0	23.7	24.3	94,311	23.1	22.7	23.5	59,597	−0.9	−1.1	−0.6
2012	Age	45 to 64	35.6	35.3	36.0	94,311	39.1	38.7	39.6	59,597	+3.5	+3.2	+3.8
2012	Age	65 and over	19.1	18.9	19.4	94,311	22.3	21.9	22.7	59,597	+3.2	+2.9	+3.4
2012	Gender	Women	52.1	51.7	52.5	94,311	53.7	53.2	54.2	59,597	+1.6	+1.3	+1.9
2012	Gender	Men	47.9	47.5	48.3	94,311	46.3	45.8	46.8	59,597	−1.6	−1.9	−1.3
2016	Education	Less than high school	9.1	8.9	9.3	93,794	5.1	4.9	5.3	58,533	−4.0	−4.2	−3.8
2016	Education	High school graduate	29.2	28.9	29.6	93,794	24.6	24.2	25.0	58,533	−4.7	−5.0	−4.4
2016	Education	Some college	29.8	29.5	30.2	93,794	30.8	30.3	31.2	58,533	+0.9	+0.7	+1.2
2016	Education	Four-year degree or more	31.9	31.5	32.2	93,794	39.6	39.2	40.1	58,533	+7.8	+7.5	+8.0
2016	Race and ethnicity	White non-Hispanic	68.9	68.6	69.3	93,794	73.3	72.9	73.7	58,533	+4.4	+4.1	+4.7
2016	Race and ethnicity	Black non-Hispanic	12.3	12.0	12.5	93,794	11.9	11.6	12.2	58,533	−0.4	−0.6	−0.2
2016	Race and ethnicity	Hispanic	11.9	11.7	12.1	93,794	9.2	9.0	9.5	58,533	−2.7	−2.9	−2.5
2016	Race and ethnicity	Asian or Pacific Islander	4.7	4.6	4.9	93,794	3.8	3.6	3.9	58,533	−1.0	−1.1	−0.8
2016	Race and ethnicity	Other or multiple races	2.2	2.1	2.3	93,794	1.8	1.6	1.9	58,533	−0.4	−0.5	−0.3
2016	Age	18 to 29	20.9	20.6	21.2	93,794	15.7	15.4	16.1	58,533	−5.2	−5.5	−4.9
2016	Age	30 to 44	23.5	23.2	23.8	93,794	22.5	22.1	22.9	58,533	−1.0	−1.3	−0.8
2016	Age	45 to 64	34.6	34.3	35.0	93,794	37.6	37.1	38.0	58,533	+3.0	+2.7	+3.2
2016	Age	65 and over	21.0	20.7	21.3	93,794	24.2	23.8	24.6	58,533	+3.2	+3.0	+3.5
2016	Gender	Women	52.0	51.6	52.4	93,794	53.6	53.2	54.1	58,533	+1.6	+1.3	+1.9
2016	Gender	Men	48.0	47.6	48.4	93,794	46.4	45.9	46.8	58,533	−1.6	−1.9	−1.3
2020	Education	Less than high school	7.6	7.4	7.8	81,898	4.6	4.4	4.8	55,170	−3.0	−3.2	−2.8
2020	Education	High school graduate	29.2	28.8	29.5	81,898	24.2	23.8	24.6	55,170	−4.9	−5.2	−4.7
2020	Education	Some college	28.5	28.2	28.9	81,898	29.7	29.3	30.2	55,170	+1.2	+1.0	+1.4
2020	Education	Four-year degree or more	34.7	34.4	35.1	81,898	41.5	41.0	41.9	55,170	+6.7	+6.5	+7.0
2020	Race and ethnicity	White non-Hispanic	66.9	66.5	67.2	81,898	71.0	70.6	71.5	55,170	+4.2	+3.9	+4.5
2020	Race and ethnicity	Black non-Hispanic	12.4	12.2	12.7	81,898	11.7	11.4	12.0	55,170	−0.7	−0.9	−0.5
2020	Race and ethnicity	Hispanic	13.2	13.0	13.5	81,898	10.6	10.4	10.9	55,170	−2.6	−2.8	−2.4
2020	Race and ethnicity	Asian or Pacific Islander	5.1	4.9	5.2	81,898	4.5	4.3	4.7	55,170	−0.6	−0.7	−0.4
2020	Race and ethnicity	Other or multiple races	2.4	2.3	2.5	81,898	2.1	2.0	2.3	55,170	−0.3	−0.4	−0.2
2020	Age	18 to 29	20.3	20.0	20.6	81,898	16.5	16.1	16.8	55,170	−3.9	−4.1	−3.6
2020	Age	30 to 44	24.0	23.7	24.4	81,898	23.2	22.8	23.6	55,170	−0.9	−1.1	−0.6
2020	Age	45 to 64	32.6	32.3	33.0	81,898	34.7	34.2	35.1	55,170	+2.1	+1.8	+2.3
2020	Age	65 and over	23.0	22.7	23.4	81,898	25.7	25.3	26.1	55,170	+2.7	+2.4	+2.9
2020	Gender	Women	51.9	51.5	52.3	81,898	53.1	52.7	53.6	55,170	+1.3	+1.0	+1.5
2020	Gender	Men	48.1	47.7	48.5	81,898	46.9	46.4	47.3	55,170	−1.3	−1.5	−1.0
2024	Education	Less than high school	6.9	6.7	7.1	72,590	3.8	3.7	4.0	48,387	−3.1	−3.2	−2.9
2024	Education	High school graduate	28.6	28.3	29.0	72,590	23.0	22.6	23.5	48,387	−5.6	−5.9	−5.3
2024	Education	Some college	27.4	27.0	27.8	72,590	28.2	27.8	28.7	48,387	+0.8	+0.6	+1.1
2024	Education	Four-year degree or more	37.1	36.7	37.5	72,590	44.9	44.4	45.4	48,387	+7.8	+7.5	+8.1
2024	Race and ethnicity	White non-Hispanic	65.5	65.1	65.9	72,590	70.6	70.1	71.1	48,387	+5.1	+4.8	+5.4
2024	Race and ethnicity	Black non-Hispanic	12.5	12.2	12.8	72,590	11.5	11.1	11.8	48,387	−1.0	−1.2	−0.8
2024	Race and ethnicity	Hispanic	13.9	13.6	14.2	72,590	10.7	10.4	11.1	48,387	−3.1	−3.4	−2.9
2024	Race and ethnicity	Asian or Pacific Islander	5.6	5.4	5.8	72,590	4.9	4.7	5.2	48,387	−0.7	−0.8	−0.5
2024	Race and ethnicity	Other or multiple races	2.6	2.5	2.7	72,590	2.3	2.1	2.4	48,387	−0.3	−0.4	−0.2
2024	Age	18 to 29	19.6	19.3	20.0	72,590	15.3	14.9	15.6	48,387	−4.4	−4.7	−4.1
2024	Age	30 to 44	24.7	24.4	25.1	72,590	23.4	23.0	23.8	48,387	−1.3	−1.6	−1.0
2024	Age	45 to 64	31.1	30.7	31.5	72,590	33.3	32.8	33.8	48,387	+2.2	+1.9	+2.5
2024	Age	65 and over	24.6	24.2	24.9	72,590	28.1	27.6	28.5	48,387	+3.5	+3.3	+3.8
2024	Gender	Women	51.6	51.2	52.0	72,590	52.8	52.3	53.3	48,387	+1.2	+0.9	+1.5
2024	Gender	Men	48.4	48.0	48.8	72,590	47.2	46.7	47.7	48,387	−1.2	−1.5	−0.9

In 2024, 44.9% of voters held a four-year degree against 37.1% of eligible adults, a difference of 7.8 points. The ring is the group's share of eligible adults and the dot is its share of voters, and a dot to the right of its ring means the group is over-represented among voters. The buttons switch the election.Note: Voters are respondents who said they voted, with the Census convention that a missing turnout answer counts as not voting. The CPS extract has no replicate weights, so the intervals ignore the survey design and are narrower than design-based intervals.Note: The difference and its interval in the table use a Taylor linearization of the voter share minus the eligible share.Source: CPS November voting supplement, 1996 to 2024; weighted (VOSUPPWT); citizens aged 18 or older

Party identifiers are a third population. Figure 3 compares the 2024 ANES populations, with party identifiers in the lower rows. Degree holders were 50.0% (95% CI 47.0% to 53.0%) of Democratic voters and 43.3% (95% CI 40.5% to 46.1%) of Democratic identifiers, against 27.8% (95% CI 24.8% to 30.9%) of Republican voters and 27.2% (95% CI 24.7% to 29.8%) of Republican identifiers. Among nonvoters the share was 15.7% (95% CI 12.3% to 19.2%), and among independents with no party lean it was 18.2% (95% CI 12.5% to 23.9%). Democratic voters are more likely than Democratic identifiers to hold a degree, and Republican voters and identifiers differ little. A statement about Democrats can refer to any of the Democratic rows, so the chapter says which one it means: voters, except in this figure.

Figure 3

Democratic voters are more likely than Democratic identifiers to hold a four-year degree, and nonvoters are less likely than voters

Share of each ANES population in four groups, percent, 2024 (the buttons show 2020); 95% intervals from the Kish effective sample size

Four-year degree or more

White non-Hispanic

Women

Age 65 and over

Share of the population in the row (%)

Show data

Data for Democratic voters are more likely than Democratic identifiers to hold a four-year degree, and nonvoters are less likely than voters
Election	Group	Population	Share (%)	95% CI low	95% CI high	n
2024	Four-year degree or more	All respondents	34.4	32.6	36.2	5,465
2024	White non-Hispanic	All respondents	64.3	62.4	66.1	5,454
2024	Women	All respondents	51.6	49.7	53.5	5,475
2024	Age 65 and over	All respondents	23.5	21.8	25.1	5,242
2024	Four-year degree or more	Nonvoters	15.7	12.3	19.2	829
2024	White non-Hispanic	Nonvoters	53.8	49.0	58.5	829
2024	Women	Nonvoters	51.5	46.7	56.2	832
2024	Age 65 and over	Nonvoters	10.4	7.5	13.3	814
2024	Four-year degree or more	Voters	39.8	37.7	42.0	4,083
2024	White non-Hispanic	Voters	67.1	65.1	69.2	4,076
2024	Women	Voters	51.5	49.3	53.7	4,091
2024	Age 65 and over	Voters	26.8	24.8	28.8	3,916
2024	Four-year degree or more	Democratic voters	50.0	47.0	53.0	2,178
2024	White non-Hispanic	Democratic voters	57.3	54.3	60.3	2,178
2024	Women	Democratic voters	55.6	52.6	58.6	2,181
2024	Age 65 and over	Democratic voters	28.6	25.8	31.4	2,104
2024	Four-year degree or more	Republican voters	27.8	24.8	30.9	1,725
2024	White non-Hispanic	Republican voters	79.5	76.7	82.2	1,719
2024	Women	Republican voters	47.3	43.9	50.7	1,730
2024	Age 65 and over	Republican voters	25.7	22.7	28.7	1,648
2024	Four-year degree or more	Democratic identifiers	43.3	40.5	46.1	2,619
2024	White non-Hispanic	Democratic identifiers	53.6	50.8	56.4	2,621
2024	Women	Democratic identifiers	55.9	53.2	58.7	2,627
2024	Age 65 and over	Democratic identifiers	25.1	22.6	27.5	2,543
2024	Four-year degree or more	Republican identifiers	27.2	24.7	29.8	2,436
2024	White non-Hispanic	Republican identifiers	77.7	75.3	80.1	2,428
2024	Women	Republican identifiers	46.3	43.5	49.2	2,440
2024	Age 65 and over	Republican identifiers	23.3	20.8	25.8	2,316
2024	Four-year degree or more	Independents with no party lean	18.2	12.5	23.9	378
2024	White non-Hispanic	Independents with no party lean	53.1	45.7	60.5	375
2024	Women	Independents with no party lean	52.3	44.9	59.7	375
2024	Age 65 and over	Independents with no party lean	12.4	7.3	17.4	356
2020	Four-year degree or more	All respondents	36.7	35.2	38.1	8,149
2020	White non-Hispanic	All respondents	65.7	64.3	67.1	8,178
2020	Women	All respondents	51.8	50.3	53.3	8,213
2020	Age 65 and over	All respondents	22.1	20.8	23.3	7,932
2020	Four-year degree or more	Nonvoters	17.2	14.1	20.3	1,022
2020	White non-Hispanic	Nonvoters	53.3	49.2	57.4	1,027
2020	Women	Nonvoters	49.8	45.8	53.9	1,031
2020	Age 65 and over	Nonvoters	9.6	7.1	12.0	1,001
2020	Four-year degree or more	Voters	40.5	38.8	42.2	6,311
2020	White non-Hispanic	Voters	69.4	67.8	71.0	6,343
2020	Women	Voters	52.6	50.9	54.3	6,366
2020	Age 65 and over	Voters	22.4	20.9	23.8	6,155
2020	Four-year degree or more	Democratic voters	47.1	44.7	49.5	3,461
2020	White non-Hispanic	Democratic voters	58.1	55.8	60.5	3,481
2020	Women	Democratic voters	54.4	52.0	56.8	3,490
2020	Age 65 and over	Democratic voters	21.5	19.5	23.5	3,410
2020	Four-year degree or more	Republican voters	32.0	29.5	34.5	2,526
2020	White non-Hispanic	Republican voters	85.8	84.0	87.7	2,549
2020	Women	Republican voters	50.3	47.6	53.0	2,553
2020	Age 65 and over	Republican voters	24.8	22.4	27.1	2,448
2020	Four-year degree or more	Democratic identifiers	42.0	39.8	44.2	3,787
2020	White non-Hispanic	Democratic identifiers	52.8	50.6	55.0	3,804
2020	Women	Democratic identifiers	56.5	54.3	58.7	3,816
2020	Age 65 and over	Democratic identifiers	21.4	19.6	23.2	3,732
2020	Four-year degree or more	Republican identifiers	33.6	31.6	35.7	3,394
2020	White non-Hispanic	Republican identifiers	82.7	81.0	84.4	3,409
2020	Women	Republican identifiers	47.6	45.4	49.8	3,425
2020	Age 65 and over	Republican identifiers	24.5	22.6	26.4	3,289
2020	Four-year degree or more	Independents with no party lean	26.4	22.5	30.3	943
2020	White non-Hispanic	Independents with no party lean	55.3	50.9	59.6	946
2020	Women	Independents with no party lean	48.2	43.8	52.6	949
2020	Age 65 and over	Independents with no party lean	16.1	12.8	19.4	897

In the 2024 default view, 50.0% of Democratic voters held a four-year degree against 43.3% of Democratic identifiers, while Republican voters (27.8%) and Republican identifiers (27.2%) differ little. Nonvoters held a degree at 15.7%. Respondents are adult citizens, so the top row stands in for eligible adults, and voters and nonvoters are self-reported.Note: Party identifiers include leaners (seven-point party identification, 1 to 3 Democratic, 5 to 7 Republican); independents with no party lean are the midpoint.Note: A Democratic voter or identifier appears in more than one row, because the rows are overlapping populations.Source: ANES Cumulative Data File, 2020 and 2024; weighted (weight for all respondents and party identifiers, weight_postelection for voters and nonvoters)

Degree holders and the highest income third became larger parts of the Democratic coalition

In 1976 the Republican coalition had the larger share of voters with a four-year degree, 24.2% (95% CI 20.9% to 27.6%) of Republican voters against 13.8% (95% CI 11.1% to 16.4%) of Democratic voters, and the larger share of voters in the highest third of the income distribution, 49.4% (95% CI 45.3% to 53.4%) against 33.5% (95% CI 29.8% to 37.3%). By 2024 both differences had reversed: 50.0% (95% CI 47.0% to 53.0%) of Democratic voters held a degree against 27.8% (95% CI 24.8% to 30.9%) of Republican voters, and 49.7% (95% CI 46.5% to 52.8%) against 39.9% (95% CI 36.4% to 43.4%) were in the highest income third (Figure 4). The income measure is relative: the ANES assigns each respondent to a percentile group of household income within the year's sample, and the chapter uses thirds. The thirds are not constant shares of the electorate, and the highest third was 38.6% (95% CI 37.1% to 40.0%) of voters in 1976 to 1988 and 42.9% (95% CI 41.8% to 43.9%) in 2012 to 2024. Chapter 3.3, The income ladder follows the vote by dollar bracket in the CES.

Figure 4

The Republican coalition once had the larger shares of degree holders and of high-income voters, and by 2024 the Democratic coalition did

Share of each party's two-party presidential voters who belong to the group, percent, by election, 1952 to 2024; 95% intervals from the Kish effective sample size

Share of the party's voters (%)

Democratic voters
Republican voters

Show data

Data for The Republican coalition once had the larger shares of degree holders and of high-income voters, and by 2024 the Democratic coalition did
Group	Party's voters	Election	Share (%)	95% CI low	95% CI high	n party voters	n in group
Four-year degree or more	Democratic voters	1956	7.7	5.4	10.0	507	39
Four-year degree or more	Democratic voters	1960	9.2	6.1	12.4	444	35
Four-year degree or more	Democratic voters	1964	9.5	7.4	11.6	746	71
Four-year degree or more	Democratic voters	1968	14.3	10.9	17.6	421	60
Four-year degree or more	Democratic voters	1972	19.1	15.8	22.3	566	108
Four-year degree or more	Democratic voters	1976	13.8	11.1	16.4	669	92
Four-year degree or more	Democratic voters	1980	16.8	13.0	20.6	381	64
Four-year degree or more	Democratic voters	1984	22.1	18.7	25.5	571	126
Four-year degree or more	Democratic voters	1988	24.8	21.2	28.4	552	137
Four-year degree or more	Democratic voters	1992	26.2	23.1	29.3	778	208
Four-year degree or more	Democratic voters	1996	24.2	20.4	28.1	599	179
Four-year degree or more	Democratic voters	2000	29.3	25.2	33.4	590	222
Four-year degree or more	Democratic voters	2004	31.9	26.8	36.9	399	143
Four-year degree or more	Democratic voters	2008	30.9	27.2	34.5	1,020	244
Four-year degree or more	Democratic voters	2012	33.3	30.7	35.9	2,469	841
Four-year degree or more	Democratic voters	2016	43.0	39.8	46.3	1,341	691
Four-year degree or more	Democratic voters	2020	47.1	44.7	49.5	3,461	1,962
Four-year degree or more	Democratic voters	2024	50.0	47.0	53.0	2,178	1,316
Four-year degree or more	Republican voters	1952	9.8	7.5	12.0	686	67
Four-year degree or more	Republican voters	1956	11.3	9.0	13.5	753	85
Four-year degree or more	Republican voters	1960	14.3	10.5	18.1	452	65
Four-year degree or more	Republican voters	1964	19.7	15.6	23.8	361	71
Four-year degree or more	Republican voters	1968	20.4	16.9	24.0	489	100
Four-year degree or more	Republican voters	1972	16.4	14.1	18.6	1,020	167
Four-year degree or more	Republican voters	1976	24.2	20.9	27.6	650	158
Four-year degree or more	Republican voters	1980	24.1	20.4	27.9	493	119
Four-year degree or more	Republican voters	1984	22.5	19.6	25.4	797	179
Four-year degree or more	Republican voters	1988	30.2	26.6	33.8	623	188
Four-year degree or more	Republican voters	1992	34.8	30.8	38.9	545	193
Four-year degree or more	Republican voters	1996	34.9	29.9	40.0	433	181
Four-year degree or more	Republican voters	2000	31.6	27.2	35.9	528	207
Four-year degree or more	Republican voters	2004	30.7	25.8	35.5	412	146
Four-year degree or more	Republican voters	2008	36.4	31.6	41.1	513	156
Four-year degree or more	Republican voters	2012	34.0	31.1	36.9	1,675	630
Four-year degree or more	Republican voters	2016	31.7	28.6	34.8	1,230	479
Four-year degree or more	Republican voters	2020	32.0	29.5	34.5	2,526	1,016
Four-year degree or more	Republican voters	2024	27.8	24.8	30.9	1,725	691
White non-Hispanic	Democratic voters	1952	91.7	89.2	94.1	492	451
White non-Hispanic	Democratic voters	1956	93.7	91.6	95.8	508	476
White non-Hispanic	Democratic voters	1960	92.4	89.5	95.3	444	413
White non-Hispanic	Democratic voters	1964	87.4	85.1	89.8	748	654
White non-Hispanic	Democratic voters	1968	79.3	75.5	83.2	421	334
White non-Hispanic	Democratic voters	1972	76.0	72.4	79.5	566	430
White non-Hispanic	Democratic voters	1976	81.6	78.6	84.6	661	542
White non-Hispanic	Democratic voters	1980	70.2	65.6	74.8	383	269
White non-Hispanic	Democratic voters	1984	71.4	67.7	75.1	574	410
White non-Hispanic	Democratic voters	1988	66.5	62.6	70.4	561	373
White non-Hispanic	Democratic voters	1992	69.7	66.5	72.9	792	551
White non-Hispanic	Democratic voters	1996	68.5	64.3	72.6	598	429
White non-Hispanic	Democratic voters	2000	67.9	63.7	72.2	586	419
White non-Hispanic	Democratic voters	2004	63.0	57.8	68.3	390	247
White non-Hispanic	Democratic voters	2008	58.9	55.0	62.8	1,019	352
White non-Hispanic	Democratic voters	2012	58.2	55.5	60.9	2,491	1,121
White non-Hispanic	Democratic voters	2016	58.3	55.1	61.5	1,343	840
White non-Hispanic	Democratic voters	2020	58.1	55.8	60.5	3,481	2,381
White non-Hispanic	Democratic voters	2024	57.3	54.3	60.3	2,178	1,521
White non-Hispanic	Republican voters	1952	98.5	97.6	99.4	686	676
White non-Hispanic	Republican voters	1956	97.6	96.5	98.7	755	737
White non-Hispanic	Republican voters	1960	96.9	95.1	98.8	451	440
White non-Hispanic	Republican voters	1964	100.0	98.9	100.0	360	360
White non-Hispanic	Republican voters	1968	98.4	97.2	99.5	490	482
White non-Hispanic	Republican voters	1972	97.1	96.0	98.1	1,021	991
White non-Hispanic	Republican voters	1976	97.9	96.8	99.0	650	637
White non-Hispanic	Republican voters	1980	95.7	94.0	97.5	494	473
White non-Hispanic	Republican voters	1984	93.1	91.3	94.9	796	741
White non-Hispanic	Republican voters	1988	92.9	90.8	94.9	630	585
White non-Hispanic	Republican voters	1992	90.1	87.6	92.6	561	505
White non-Hispanic	Republican voters	1996	93.5	90.9	96.1	433	404
White non-Hispanic	Republican voters	2000	88.6	85.7	91.6	527	472
White non-Hispanic	Republican voters	2004	86.3	82.7	90.0	410	354
White non-Hispanic	Republican voters	2008	92.6	90.0	95.3	511	431
White non-Hispanic	Republican voters	2012	89.5	87.6	91.4	1,690	1,434
White non-Hispanic	Republican voters	2016	88.6	86.5	90.8	1,232	1,105
White non-Hispanic	Republican voters	2020	85.8	84.0	87.7	2,549	2,225
White non-Hispanic	Republican voters	2024	79.5	76.7	82.2	1,719	1,470
Protestant	Democratic voters	1952	60.2	55.8	64.5	492	296
Protestant	Democratic voters	1956	62.0	57.8	66.2	511	317
Protestant	Democratic voters	1960	53.8	48.3	59.3	445	235
Protestant	Democratic voters	1964	63.6	60.2	67.0	750	477
Protestant	Democratic voters	1968	58.9	54.2	63.6	416	245
Protestant	Democratic voters	1972	58.9	54.9	63.0	565	333
Protestant	Democratic voters	1976	60.7	56.9	64.5	669	410
Protestant	Democratic voters	1980	65.3	60.5	70.1	380	248
Protestant	Democratic voters	1984	54.2	50.1	58.3	574	311
Protestant	Democratic voters	1988	57.5	53.5	61.6	563	324
Protestant	Democratic voters	1992	51.4	47.9	54.9	789	406
Protestant	Democratic voters	1996	49.7	45.2	54.2	599	301
Protestant	Democratic voters	2000	49.8	45.3	54.4	586	290
Protestant	Democratic voters	2004	47.3	41.9	52.7	397	191
Protestant	Democratic voters	2008	47.6	43.6	51.5	1,023	544
Protestant	Democratic voters	2012	38.0	35.3	40.7	2,491	1,066
Protestant	Democratic voters	2016	37.8	34.6	41.0	1,349	492
Protestant	Democratic voters	2020	32.5	30.3	34.7	3,509	1,118
Protestant	Republican voters	1952	76.1	72.9	79.3	685	521
Protestant	Republican voters	1956	75.3	72.2	78.4	753	567
Protestant	Republican voters	1960	90.4	87.2	93.6	452	407
Protestant	Republican voters	1964	79.2	75.0	83.4	361	286
Protestant	Republican voters	1968	80.1	76.5	83.6	487	390
Protestant	Republican voters	1972	71.9	69.2	74.7	1,019	733
Protestant	Republican voters	1976	72.7	69.2	76.2	651	477
Protestant	Republican voters	1980	67.8	63.7	72.0	491	333
Protestant	Republican voters	1984	65.6	62.3	68.9	797	523
Protestant	Republican voters	1988	69.8	66.3	73.4	630	440
Protestant	Republican voters	1992	66.4	62.5	70.3	562	374
Protestant	Republican voters	1996	61.7	56.6	66.8	434	271
Protestant	Republican voters	2000	56.5	51.8	61.1	529	301
Protestant	Republican voters	2004	54.3	49.0	59.5	411	222
Protestant	Republican voters	2008	62.8	58.0	67.6	514	301
Protestant	Republican voters	2012	52.6	49.5	55.7	1,692	879
Protestant	Republican voters	2016	56.0	52.7	59.3	1,238	707
Protestant	Republican voters	2020	48.0	45.3	50.6	2,567	1,295
Age 65 and over	Democratic voters	1952	11.2	8.4	14.0	490	55
Age 65 and over	Democratic voters	1956	12.9	10.0	15.9	502	65
Age 65 and over	Democratic voters	1960	11.9	8.3	15.4	446	58
Age 65 and over	Democratic voters	1964	11.7	9.4	14.1	749	88
Age 65 and over	Democratic voters	1968	16.4	12.9	20.0	420	69
Age 65 and over	Democratic voters	1972	13.3	10.5	16.1	564	75
Age 65 and over	Democratic voters	1976	17.5	14.6	20.4	666	118
Age 65 and over	Democratic voters	1980	21.1	17.1	25.2	383	81
Age 65 and over	Democratic voters	1984	17.5	14.4	20.7	570	100
Age 65 and over	Democratic voters	1988	19.9	16.6	23.2	563	112
Age 65 and over	Democratic voters	1992	23.0	20.1	26.0	793	178
Age 65 and over	Democratic voters	1996	21.0	17.3	24.6	598	146
Age 65 and over	Democratic voters	2000	20.2	16.5	23.8	589	120
Age 65 and over	Democratic voters	2004	18.3	14.1	22.5	399	76
Age 65 and over	Democratic voters	2008	15.7	12.8	18.6	1,014	169
Age 65 and over	Democratic voters	2012	19.5	17.3	21.7	2,469	507
Age 65 and over	Democratic voters	2016	19.7	17.0	22.3	1,321	320
Age 65 and over	Democratic voters	2020	21.5	19.5	23.5	3,410	1,029
Age 65 and over	Democratic voters	2024	28.6	25.8	31.4	2,104	767
Age 65 and over	Republican voters	1952	14.1	11.4	16.7	676	95
Age 65 and over	Republican voters	1956	11.2	8.9	13.4	744	83
Age 65 and over	Republican voters	1960	17.9	13.7	22.0	452	90
Age 65 and over	Republican voters	1964	19.7	15.6	23.8	361	71
Age 65 and over	Republican voters	1968	17.2	13.9	20.6	488	84
Age 65 and over	Republican voters	1972	15.8	13.5	18.0	1,015	160
Age 65 and over	Republican voters	1976	18.8	15.7	21.8	648	127
Age 65 and over	Republican voters	1980	19.1	15.6	22.6	492	94
Age 65 and over	Republican voters	1984	16.8	14.2	19.4	793	133
Age 65 and over	Republican voters	1988	19.3	16.3	22.4	631	122
Age 65 and over	Republican voters	1992	21.3	17.9	24.7	564	118
Age 65 and over	Republican voters	1996	19.1	15.0	23.3	434	102
Age 65 and over	Republican voters	2000	18.6	14.9	22.3	528	108
Age 65 and over	Republican voters	2004	16.7	12.8	20.6	412	70
Age 65 and over	Republican voters	2008	23.7	19.4	28.0	504	125
Age 65 and over	Republican voters	2012	25.5	22.8	28.2	1,679	476
Age 65 and over	Republican voters	2016	25.6	22.6	28.5	1,201	359
Age 65 and over	Republican voters	2020	24.8	22.4	27.1	2,448	854
Age 65 and over	Republican voters	2024	25.7	22.7	28.7	1,648	623
Women	Democratic voters	1952	49.8	45.4	54.2	494	246
Women	Democratic voters	1956	47.7	43.4	52.1	511	244
Women	Democratic voters	1960	49.4	43.9	54.8	446	220
Women	Democratic voters	1964	55.7	52.2	59.3	750	418
Women	Democratic voters	1968	58.2	53.5	62.9	421	245
Women	Democratic voters	1972	59.9	55.9	63.9	566	339
Women	Democratic voters	1976	56.1	52.3	60.0	670	378
Women	Democratic voters	1980	59.3	54.3	64.2	383	227
Women	Democratic voters	1984	60.9	56.9	64.9	575	350
Women	Democratic voters	1988	58.8	54.7	62.9	563	331
Women	Democratic voters	1992	58.5	55.1	61.9	793	462
Women	Democratic voters	1996	60.4	56.0	64.8	600	360
Women	Democratic voters	2000	60.0	55.5	64.4	590	357
Women	Democratic voters	2004	55.5	50.2	60.9	399	229
Women	Democratic voters	2008	59.6	55.7	63.4	1,025	631
Women	Democratic voters	2012	55.1	52.4	57.8	2,496	1,355
Women	Democratic voters	2016	56.8	53.6	60.1	1,328	777
Women	Democratic voters	2020	54.4	52.0	56.8	3,490	2,003
Women	Democratic voters	2024	55.6	52.6	58.6	2,181	1,256
Women	Republican voters	1952	52.0	48.2	55.7	687	357
Women	Republican voters	1956	54.2	50.6	57.7	755	409
Women	Republican voters	1960	54.7	49.3	60.1	452	249
Women	Republican voters	1964	51.5	46.4	56.7	361	186
Women	Republican voters	1968	56.3	51.9	60.7	490	276
Women	Republican voters	1972	52.8	49.7	55.9	1,021	539
Women	Republican voters	1976	55.9	52.0	59.8	652	361
Women	Republican voters	1980	51.6	47.2	56.0	494	255
Women	Republican voters	1984	53.2	49.7	56.6	801	426
Women	Republican voters	1988	52.4	48.5	56.3	632	331
Women	Republican voters	1992	52.1	48.0	56.3	564	293
Women	Republican voters	1996	45.2	40.0	50.4	434	204
Women	Republican voters	2000	51.2	46.5	55.9	530	266
Women	Republican voters	2004	48.5	43.2	53.7	412	208
Women	Republican voters	2008	54.6	49.6	59.5	514	282
Women	Republican voters	2012	49.9	46.8	53.0	1,692	818
Women	Republican voters	2016	50.7	47.4	54.0	1,230	610
Women	Republican voters	2020	50.3	47.6	53.0	2,553	1,311
Women	Republican voters	2024	47.3	43.9	50.7	1,730	825
Highest third of income	Democratic voters	1952	43.4	39.0	47.8	488	212
Highest third of income	Democratic voters	1956	32.9	28.7	37.0	496	163
Highest third of income	Democratic voters	1960	41.8	36.4	47.2	443	191
Highest third of income	Democratic voters	1964	35.7	32.2	39.2	725	259
Highest third of income	Democratic voters	1968	35.8	31.1	40.4	411	147
Highest third of income	Democratic voters	1972	33.2	29.2	37.1	552	183
Highest third of income	Democratic voters	1976	33.5	29.8	37.3	626	217
Highest third of income	Democratic voters	1980	28.1	23.4	32.9	345	97
Highest third of income	Democratic voters	1984	28.7	24.8	32.6	526	151
Highest third of income	Democratic voters	1988	32.3	28.2	36.3	514	166
Highest third of income	Democratic voters	1992	34.0	30.6	37.4	742	256
Highest third of income	Democratic voters	1996	31.6	27.2	36.0	549	164
Highest third of income	Democratic voters	2000	34.0	29.4	38.6	504	164
Highest third of income	Democratic voters	2004	32.7	27.4	38.0	364	111
Highest third of income	Democratic voters	2008	24.9	21.4	28.4	958	180
Highest third of income	Democratic voters	2012	31.7	29.1	34.3	2,429	677
Highest third of income	Democratic voters	2016	42.2	38.9	45.4	1,330	536
Highest third of income	Democratic voters	2020	46.7	44.3	49.1	3,451	1,240
Highest third of income	Democratic voters	2024	49.7	46.5	52.8	2,026	817
Highest third of income	Republican voters	1952	50.1	46.4	53.9	674	338
Highest third of income	Republican voters	1956	37.9	34.4	41.5	720	273
Highest third of income	Republican voters	1960	44.3	38.9	49.7	450	208
Highest third of income	Republican voters	1964	51.4	46.2	56.7	348	179
Highest third of income	Republican voters	1968	38.2	33.8	42.5	482	184
Highest third of income	Republican voters	1972	41.0	38.0	44.1	994	408
Highest third of income	Republican voters	1976	49.4	45.3	53.4	615	308
Highest third of income	Republican voters	1980	39.0	34.5	43.5	449	175
Highest third of income	Republican voters	1984	46.1	42.5	49.8	726	335
Highest third of income	Republican voters	1988	46.2	42.2	50.3	586	271
Highest third of income	Republican voters	1992	47.5	43.2	51.7	525	251
Highest third of income	Republican voters	1996	45.2	39.7	50.6	394	182
Highest third of income	Republican voters	2000	40.9	35.9	46.0	438	183
Highest third of income	Republican voters	2004	44.3	38.8	49.9	365	150
Highest third of income	Republican voters	2008	42.8	37.7	47.8	487	183
Highest third of income	Republican voters	2012	39.5	36.4	42.5	1,650	632
Highest third of income	Republican voters	2016	45.0	41.6	48.3	1,210	509
Highest third of income	Republican voters	2020	48.6	45.9	51.3	2,494	857
Highest third of income	Republican voters	2024	39.9	36.4	43.4	1,554	493

In 1976 the Republican line is above the Democratic line in the degree panel (24.2% against 13.8%) and in the income panel (49.4% against 33.5%); by 2024 the order has reversed (50.0% against 27.8%, and 49.7% against 39.9%). Each panel plots the share of Democratic voters and of Republican voters in one group. Income is relative: the respondent's household income rank within the year's sample, in thirds. Bands are confidence intervals.Note: Cells need at least 100 voters in the party, a Kish effective sample of at least 100 and at least 30 respondents in the group and party; the Democratic degree cell in 1952 is dropped.Note: The Hispanic respondents of the ANES are identified from 1966, so before then they are counted as white. The Protestant panel ends in 2020 because the 2024 religion codes of the cumulative file are unusable.Note: Intervals are model based (Kish effective sample size, normal approximation) and ignore clustering in the sample design.Source: ANES Cumulative Data File, presidential years 1952 to 2024; weighted (weight_postelection where the file has one, else weight); two-party presidential voters

The trend figure uses single elections. For comparisons between two periods, the chapter pools four elections at each end, 1976 to 1988 and 2012 to 2024, so that small groups have enough respondents, and it gives each election the same weight. By that comparison the degree-holding share of Democratic voters rose by 24.2 (95% CI 22.0 to 26.4) points and that of Republican voters by 6.2 (95% CI 4.0 to 8.5) points, a smaller rise (Figure 5, education view). The highest income third went from 30.9% (95% CI 28.8% to 32.9%) to 42.6% (95% CI 41.1% to 44.0%) of Democratic voters and from 45.1% (95% CI 43.0% to 47.1%) to 43.3% (95% CI 41.7% to 44.8%) of Republican voters. The change in the highest third's share of Democratic voters splits into a group-size term of +3.9 (95% CI 2.3 to 5.5) points and a relative-lean term of +7.8 (95% CI 6.0 to 9.6) points, and for Republican voters into +4.7 (95% CI 2.7 to 6.7) and −6.5 (95% CI −8.2 to −4.8) (Figure 7, income views). The CES, which starts in 2008, shows the same direction for Democratic voters: the degree-holding share went from 31.2% (95% CI 30.2% to 32.3%) in 2008 to 47.9% (95% CI 46.9% to 49.0%) in 2024. Whether the Republican share rose or fell since 2008 depends on the survey, a disagreement taken up near the end.

Figure 5

Four-year degree holders rose from 19.2% to 43.4% of Democratic voters and from 25.2% to 31.4% of Republican voters

Make-up of each party's two-party presidential voters, percent, in 1976 to 1988 and 2012 to 2024 (the buttons switch the dimension); pooled elections; 95% intervals from the Kish effective sample size

No four-year degree
Four-year degree or more

Show data

Data for Four-year degree holders rose from 19.2% to 43.4% of Democratic voters and from 25.2% to 31.4% of Republican voters
Dimension	Party and period	Group	Share of the party's voters (%)	95% CI low	95% CI high	n party voters	n in group
Education	Democratic voters, 1976 to 1988	No four-year degree	80.8	79.1	82.5	2,173	1,754
Education	Democratic voters, 1976 to 1988	Four-year degree or more	19.2	17.5	20.9	2,173	419
Education	Democratic voters, 2012 to 2024	No four-year degree	56.6	55.2	58.0	9,449	4,639
Education	Democratic voters, 2012 to 2024	Four-year degree or more	43.4	42.0	44.8	9,449	4,810
Education	Republican voters, 1976 to 1988	No four-year degree	74.8	73.1	76.5	2,563	1,919
Education	Republican voters, 1976 to 1988	Four-year degree or more	25.2	23.5	26.9	2,563	644
Education	Republican voters, 2012 to 2024	No four-year degree	68.6	67.2	70.1	7,156	4,340
Education	Republican voters, 2012 to 2024	Four-year degree or more	31.4	29.9	32.8	7,156	2,816
Income	Democratic voters, 1976 to 1988	Lowest third	33.5	31.4	35.6	2,011	664
Income	Democratic voters, 1976 to 1988	Middle third	35.7	33.5	37.8	2,011	716
Income	Democratic voters, 1976 to 1988	Highest third	30.9	28.8	32.9	2,011	631
Income	Democratic voters, 2012 to 2024	Lowest third	24.9	23.6	26.2	9,236	2,855
Income	Democratic voters, 2012 to 2024	Middle third	32.5	31.2	33.9	9,236	3,111
Income	Democratic voters, 2012 to 2024	Highest third	42.6	41.1	44.0	9,236	3,270
Income	Republican voters, 1976 to 1988	Lowest third	20.3	18.6	21.9	2,376	474
Income	Republican voters, 1976 to 1988	Middle third	34.6	32.7	36.6	2,376	813
Income	Republican voters, 1976 to 1988	Highest third	45.1	43.0	47.1	2,376	1,089
Income	Republican voters, 2012 to 2024	Lowest third	21.1	19.8	22.4	6,908	1,911
Income	Republican voters, 2012 to 2024	Middle third	35.6	34.1	37.2	6,908	2,506
Income	Republican voters, 2012 to 2024	Highest third	43.3	41.7	44.8	6,908	2,491
Race and ethnicity	Democratic voters, 1976 to 1988	White non-Hispanic	72.7	70.8	74.6	2,179	1,594
Race and ethnicity	Democratic voters, 1976 to 1988	Black non-Hispanic	20.0	18.3	21.7	2,179	425
Race and ethnicity	Democratic voters, 1976 to 1988	Hispanic or another race	7.3	6.2	8.4	2,179	160
Race and ethnicity	Democratic voters, 2012 to 2024	White non-Hispanic	58.0	56.6	59.4	9,493	5,863
Race and ethnicity	Democratic voters, 2012 to 2024	Black non-Hispanic	19.9	18.7	21.0	9,493	1,778
Race and ethnicity	Democratic voters, 2012 to 2024	Hispanic or another race	22.1	20.9	23.3	9,493	1,852
Race and ethnicity	Republican voters, 1976 to 1988	White non-Hispanic	94.8	93.9	95.7	2,570	2,436
Race and ethnicity	Republican voters, 1976 to 1988	Black non-Hispanic	1.3	0.9	1.8	2,570	33
Race and ethnicity	Republican voters, 1976 to 1988	Hispanic or another race	3.9	3.1	4.6	2,570	101
Race and ethnicity	Republican voters, 2012 to 2024	White non-Hispanic	85.9	84.8	87.0	7,190	6,234
Race and ethnicity	Republican voters, 2012 to 2024	Black non-Hispanic	1.6	1.2	2.0	7,190	82
Race and ethnicity	Republican voters, 2012 to 2024	Hispanic or another race	12.5	11.5	13.5	7,190	874
Religion	Democratic voters, 1976 to 1988	Protestant	59.5	57.4	61.6	2,186	1,293
Religion	Democratic voters, 1976 to 1988	Catholic	28.0	26.0	29.9	2,186	619
Religion	Democratic voters, 1976 to 1988	Jewish, another religion or none	12.6	11.1	14.0	2,186	274
Religion	Democratic voters, 2012 to 2020	Protestant	36.0	34.5	37.6	7,349	2,676
Religion	Democratic voters, 2012 to 2020	Catholic	21.8	20.4	23.1	7,349	1,571
Religion	Democratic voters, 2012 to 2020	Jewish, another religion or none	42.2	40.6	43.8	7,349	3,102
Religion	Republican voters, 1976 to 1988	Protestant	68.8	67.0	70.7	2,569	1,773
Religion	Republican voters, 1976 to 1988	Catholic	23.1	21.5	24.8	2,569	595
Religion	Republican voters, 1976 to 1988	Jewish, another religion or none	8.0	7.0	9.1	2,569	201
Religion	Republican voters, 2012 to 2020	Protestant	52.3	50.5	54.0	5,497	2,881
Religion	Republican voters, 2012 to 2020	Catholic	22.5	21.0	24.0	5,497	1,258
Religion	Republican voters, 2012 to 2020	Jewish, another religion or none	25.2	23.7	26.7	5,497	1,358
Age	Democratic voters, 1976 to 1988	18 to 29	19.2	17.6	20.9	2,182	409
Age	Democratic voters, 1976 to 1988	30 to 44	30.6	28.6	32.6	2,182	675
Age	Democratic voters, 1976 to 1988	45 to 64	31.2	29.2	33.1	2,182	687
Age	Democratic voters, 1976 to 1988	65 and over	19.0	17.3	20.7	2,182	411
Age	Democratic voters, 2012 to 2024	18 to 29	18.5	17.4	19.6	9,304	1,125
Age	Democratic voters, 2012 to 2024	30 to 44	24.6	23.4	25.9	9,304	2,255
Age	Democratic voters, 2012 to 2024	45 to 64	34.6	33.2	35.9	9,304	3,301
Age	Democratic voters, 2012 to 2024	65 and over	22.3	21.1	23.5	9,304	2,623
Age	Republican voters, 1976 to 1988	18 to 29	18.9	17.4	20.5	2,564	472
Age	Republican voters, 1976 to 1988	30 to 44	33.0	31.2	34.9	2,564	851
Age	Republican voters, 1976 to 1988	45 to 64	29.6	27.8	31.4	2,564	765
Age	Republican voters, 1976 to 1988	65 and over	18.5	16.9	20.0	2,564	476
Age	Republican voters, 2012 to 2024	18 to 29	13.1	12.0	14.2	6,976	548
Age	Republican voters, 2012 to 2024	30 to 44	21.5	20.2	22.8	6,976	1,431
Age	Republican voters, 2012 to 2024	45 to 64	40.0	38.5	41.6	6,976	2,685
Age	Republican voters, 2012 to 2024	65 and over	25.4	24.0	26.8	6,976	2,312
Gender	Democratic voters, 1976 to 1988	Women	58.6	56.5	60.7	2,191	1,286
Gender	Democratic voters, 1976 to 1988	Men	41.4	39.3	43.5	2,191	905
Gender	Democratic voters, 2012 to 2024	Women	55.5	54.0	56.9	9,495	5,391
Gender	Democratic voters, 2012 to 2024	Men	44.5	43.1	46.0	9,495	4,104
Gender	Republican voters, 1976 to 1988	Women	53.2	51.2	55.2	2,579	1,373
Gender	Republican voters, 1976 to 1988	Men	46.8	44.8	48.8	2,579	1,206
Gender	Republican voters, 2012 to 2024	Women	49.6	48.0	51.1	7,205	3,564
Gender	Republican voters, 2012 to 2024	Men	50.4	48.9	52.0	7,205	3,641

Each bar shows who is in one party's coalition in one period, and the segments in a bar sum to the whole of the party's voters. In the default view, the four-year degree segment grows in both coalitions and by more in the Democratic one. Segments are shares of the party's voters, not shares of the group voting for the party. The tooltip and the table give the interval for each segment.Note: Each period pools four elections, 1976, 1980, 1984 and 1988 against 2012, 2016, 2020 and 2024, and every election has the same total weight. The religion view ends in 2020 because the 2024 religion codes of the cumulative file are unusable.Note: Intervals are model based (Kish effective sample size, normal approximation) and ignore clustering in the sample design.Note: Stacked bars carry no whiskers; the interval and the number of voters in the party's denominator are in the tooltip and the table.Source: ANES Cumulative Data File, 1976 to 1988 and 2012 to 2024; weighted (weight_postelection where the file has one, else weight; each election has the same total weight); two-party presidential voters

Both coalitions became less white, and both became less Protestant

White non-Hispanic voters were 72.7% (95% CI 70.8% to 74.6%) of Democratic voters and 94.8% (95% CI 93.9% to 95.7%) of Republican voters in 1976 to 1988, and 58.0% (95% CI 56.6% to 59.4%) and 85.9% (95% CI 84.8% to 87.0%) in 2012 to 2024. The groups that grew were Hispanic voters and voters of other races, who went from 7.3% (95% CI 6.2% to 8.4%) to 22.1% (95% CI 20.9% to 23.3%) of Democratic voters and from 3.9% (95% CI 3.1% to 4.6%) to 12.5% (95% CI 11.5% to 13.5%) of Republican voters. Black voters were 20.0% (95% CI 18.3% to 21.7%) of Democratic voters in the first period and 19.9% (95% CI 18.7% to 21.0%) in the second, and a small share of Republican voters in both. Chapter 3.4, Hispanic, Black and Asian voters (planned) follows how those groups voted. This chapter uses only their size in each coalition.

Protestants were 59.5% (95% CI 57.4% to 61.6%) of Democratic voters in 1976 to 1988 and 36.0% (95% CI 34.5% to 37.6%) in 2012 to 2020, and 68.8% (95% CI 67.0% to 70.7%) and 52.3% (95% CI 50.5% to 54.0%) of Republican voters. The ANES cumulative file cannot separate voters with no religion from those with a religion outside the three it names, and its 2024 religion codes are unusable, so the later period ends in 2020 and the last category is Jewish, another religion or none. The CES separates them. Voters with no religion went from 35.7% (95% CI 34.7% to 36.8%) of Democratic voters in 2012 to 45.0% (95% CI 43.9% to 46.0%) in 2024, and from 14.5% (95% CI 13.7% to 15.2%) to 17.7% (95% CI 16.8% to 18.5%) of Republican voters. Chapter 9.1, The god gap in America covers the gap between frequent and infrequent attenders, and Chapter 9.2, The rise of the nones (planned) covers the religiously unaffiliated.

Voters in the oldest age band, ages 65 and over, rose from 19.0% (95% CI 17.3% to 20.7%) to 22.3% (95% CI 21.1% to 23.5%) of Democratic voters and from 18.5% (95% CI 16.9% to 20.0%) to 25.4% (95% CI 24.0% to 26.8%) of Republican voters, in step with the aging of the electorate: the CPS puts that band at 15.9% (95% CI 15.7% to 16.0%) of adults in 1976 to 1988 and 21.9% (95% CI 21.8% to 22.1%) in 2012 to 2024. In 1976 to 1988 the two coalitions had a similar age profile, with the youngest band at 19.2% (95% CI 17.6% to 20.9%) of Democratic voters and 18.9% (95% CI 17.4% to 20.5%) of Republican voters. In 2012 to 2024 the shares were 18.5% (95% CI 17.4% to 19.6%) and 13.1% (95% CI 12.0% to 14.2%), so the Republican coalition had become the older of the two. Chapter 3.6, Age and generation separates age from birth cohort.

Women were 58.6% (95% CI 56.5% to 60.7%) of Democratic voters and 53.2% (95% CI 51.2% to 55.2%) of Republican voters in 1976 to 1988 and 55.5% (95% CI 54.0% to 56.9%) and 49.6% (95% CI 48.0% to 51.1%) in 2012 to 2024. Their share of both coalitions fell by about the same amount even though the Democratic share among women rose from 48.4% (95% CI 46.4% to 50.3%) to 56.9% (95% CI 55.5% to 58.4%), because women were a smaller share of voters in the ANES's later elections, 55.7% (95% CI 54.2% to 57.1%) then and 52.8% (95% CI 51.7% to 53.8%) now. The CPS, over the same two windows, shows women at 53.1% (95% CI 52.9% to 53.3%) of voters in 1976 to 1988 and 53.3% (95% CI 53.1% to 53.6%) in 2012 to 2024, with no change beyond sampling error, so the ANES figure for the earlier window is above the CPS figure and the ANES fall belongs to that window. Chapter 3.5, The gender gap in attitudes and votes covers the vote gap between women and men.

Figure 6 turns to the other denominator, the Democratic share of each group's voters, for the same two periods. Among degree holders the share rose from 39.2% (95% CI 36.2% to 42.2%) to 62.0% (95% CI 60.6% to 63.5%), a rise of 22.8 (95% CI 19.5 to 26.1) points, and the rise is larger than that of every other group once sampling error is allowed for. Among the highest income third the share went from 36.7% (95% CI 34.3% to 39.0%) to 54.0% (95% CI 52.3% to 55.7%) and among the youngest age band from 46.4% (95% CI 43.0% to 49.7%) to 62.7% (95% CI 59.5% to 65.8%). The lowest income third and Black voters kept nearly the same Democratic share at high levels, 58.3% and 58.5% for the lowest third and 92.8% and 93.7% for Black voters. Education below a degree is not split further, because the cumulative file defines its lower groups differently after 2016. The Democratic share of all voters rose over these years from 45.9% (95% CI 44.5% to 47.4%) to 54.1% (95% CI 53.1% to 55.2%), so a group whose share rose by less than that, as white voters' did, moved toward the Democratic candidate by less than the electorate. The later elections also lean Democratic against the certified counts, as the last section shows.

Figure 6

The Democratic share rose more among four-year degree holders than in any other group

Democratic share of each group's two-party presidential voters, percent, 1976 to 1988 and 2012 to 2024; pooled elections; 95% intervals from the Kish effective sample size

1976 to 1988
2012 to 2024

Show data

Data for The Democratic share rose more among four-year degree holders than in any other group
Dimension	Group	Earlier period (%)	95% CI low earlier	95% CI high earlier	n earlier	Later period (%)	95% CI low later	95% CI high later	n later	Change (points)
Education	No four-year degree	47.8	46.1	49.4	3,673	49.3	47.9	50.8	8,979	+1.5
Education	Four-year degree or more	39.2	36.2	42.2	1,063	62.0	60.6	63.5	7,626	+22.8
Income (relative thirds)	Lowest third	58.3	55.3	61.2	1,138	58.5	56.4	60.6	4,766	+0.2
Income (relative thirds)	Middle third	46.5	44.0	49.1	1,529	52.2	50.3	54.0	5,617	+5.6
Income (relative thirds)	Highest third	36.7	34.3	39.0	1,720	54.0	52.3	55.7	5,761	+17.3
Race and ethnicity	White non-Hispanic	39.4	37.8	40.9	4,030	44.4	43.2	45.6	12,097	+5.0
Race and ethnicity	Black non-Hispanic	92.8	90.4	95.2	458	93.7	92.0	95.4	1,860	+0.9
Race and ethnicity	Hispanic	61.7	55.0	68.5	204	72.2	68.9	75.4	1,616	+10.4
Religion (later period 2012 to 2020)	Protestant	42.4	40.6	44.2	3,066	44.7	42.9	46.5	5,557	+2.4
Religion (later period 2012 to 2020)	Catholic	50.7	47.8	53.6	1,214	53.2	50.7	55.7	2,829	+2.5
Religion (later period 2012 to 2020)	Jewish, another religion or none	57.1	52.5	61.6	475	66.3	64.3	68.2	4,460	+9.2
Age	18 to 29	46.4	43.0	49.7	881	62.7	59.5	65.8	1,673	+16.3
Age	30 to 44	44.1	41.5	46.6	1,526	57.8	55.6	59.9	3,686	+13.7
Age	45 to 64	47.3	44.6	49.9	1,452	50.7	49.0	52.4	5,986	+3.4
Age	65 and over	46.7	43.3	50.0	887	51.1	49.2	53.1	4,935	+4.5
Gender	Women	48.4	46.4	50.3	2,659	56.9	55.5	58.4	8,955	+8.6
Gender	Men	42.9	40.7	45.0	2,111	51.0	49.5	52.6	7,745	+8.2

Among degree holders the share rose by 22.8 points, from 39.2% to 62.0%, and no other group's rise is as large once sampling error is allowed for. Among the highest income third the share went from 36.7% to 54.0% and among the youngest age band from 46.4% to 62.7%. The ring is the earlier period and the dot the later one; the denominator is the group's voters. The Democratic share of all voters rose from 45.9% to 54.1% in these surveys, so a group that rose by less than that moved less than the electorate did.Note: Education is split into degree and no degree only: the cumulative file defines its lower education groups differently after 2016. The other-races group has too few respondents in the earlier period to publish (fewer than 100) and has no row. Hispanic respondents are identified in the ANES from 1966.Note: The ANES late-period elections lean Democratic relative to the certified counts (see the last section), so the later dots are higher than a count-based benchmark would give. Intervals are model based (Kish effective sample size, normal approximation) and ignore clustering in the sample design.Note: Each period pools four elections, 1976, 1980, 1984 and 1988 against 2012, 2016, 2020 and 2024, and every election has the same total weight. The religion rows compare 1976 to 1988 with 2012 to 2020.Source: ANES Cumulative Data File, 1976 to 1988 and 2012 to 2024; weighted (weight_postelection where the file has one, else weight; each election has the same total weight); two-party presidential voters

Figure 7 splits the change in a coalition share into the two terms of the product and starts with degree holders. Their share of all voters rose from 22.4% (95% CI 21.2% to 23.6%) to 37.9% (95% CI 36.9% to 38.9%). Their Democratic share was 39.2% (95% CI 36.2% to 42.2%) against 45.9% (95% CI 44.4% to 47.3%) for all voters in 1976 to 1988 and 62.0% (95% CI 60.6% to 63.5%) against 54.1% (95% CI 53.1% to 55.2%) in 2012 to 2024. The degree-holding share of Democratic voters rose by 24.2 (95% CI 22.0 to 26.4) points, and the group-size term is +15.5 (95% CI 14.0 to 17.1) points and the relative-lean term +8.8 (95% CI 7.1 to 10.6). Among Republican voters the group-size term is +15.1 (95% CI 13.6 to 16.7), close to the Democratic one, and the lean term is −8.9 (95% CI −10.6 to −7.4).The terms use the symmetric split of a product: the change in group size times the average relative lean, plus the change in relative lean times the average group size. The split is exact, and its two terms do not depend on which factor is moved first. Kitagawa's standardization of a difference between two rates is built the same way [5]. Intervals come from a bootstrap that resamples respondents within each election. The group-size term is about the same in both coalitions, and the relative-lean term has opposite signs, positive for Democratic voters and negative for Republican voters. The CES gives the same signs for 2008 to 2024, with a group-size term of +13.1 (95% CI 12.0 to 14.1) points for Democratic voters and +10.8 (95% CI 9.8 to 11.7) for Republican voters and lean terms of +3.5 (95% CI 2.5 to 4.5) and −3.0 (95% CI −4.0 to −1.9).

Figure 7

Among Democratic voters the degree-holding share rose through group size (+15.5 points) and relative lean (+8.8 points)

Change in a group's share of a party's two-party presidential voters from 1976 to 1988 to 2012 to 2024, split into group size and relative lean, percent and percentage points; ANES; the buttons switch group and party

Show data

Data for Among Democratic voters the degree-holding share rose through group size (+15.5 points) and relative lean (+8.8 points)
Group	Party's voters	Step	Value	95% CI low	95% CI high	n
Degree holders	Democratic voters	Share of Democratic voters, 1976 to 1988	19.2	17.5	20.9	2,173
Degree holders	Democratic voters	Group size	15.5	14.0	17.1	21,554
Degree holders	Democratic voters	Relative lean	8.8	7.1	10.6	21,554
Degree holders	Democratic voters	Share of Democratic voters, 2012 to 2024	43.4	42.0	44.8	9,449
Degree holders	Republican voters	Share of Republican voters, 1976 to 1988	25.2	23.5	26.9	2,563
Degree holders	Republican voters	Group size	15.1	13.6	16.7	21,554
Degree holders	Republican voters	Relative lean	−8.9	−10.6	−7.4	21,554
Degree holders	Republican voters	Share of Republican voters, 2012 to 2024	31.4	29.9	32.8	7,156
White non-Hispanic voters	Democratic voters	Share of Democratic voters, 1976 to 1988	72.7	70.8	74.6	2,179
White non-Hispanic voters	Democratic voters	Group size	−11.6	−12.8	−10.4	21,554
White non-Hispanic voters	Democratic voters	Relative lean	−3.1	−4.6	−1.5	21,554
White non-Hispanic voters	Democratic voters	Share of Democratic voters, 2012 to 2024	58.0	56.6	59.4	9,493
White non-Hispanic voters	Republican voters	Share of Republican voters, 1976 to 1988	94.8	93.9	95.7	2,570
White non-Hispanic voters	Republican voters	Group size	−16.2	−17.8	−14.5	21,554
White non-Hispanic voters	Republican voters	Relative lean	7.3	5.7	8.9	21,554
White non-Hispanic voters	Republican voters	Share of Republican voters, 2012 to 2024	85.9	84.8	87.0	7,190
Highest income third	Democratic voters	Share of Democratic voters, 1976 to 1988	30.9	28.8	32.9	2,011
Highest income third	Democratic voters	Group size	3.9	2.3	5.5	21,554
Highest income third	Democratic voters	Relative lean	7.8	6.0	9.6	21,554
Highest income third	Democratic voters	Share of Democratic voters, 2012 to 2024	42.6	41.1	44.0	9,236
Highest income third	Republican voters	Share of Republican voters, 1976 to 1988	45.1	43.0	47.1	2,376
Highest income third	Republican voters	Group size	4.7	2.7	6.7	21,554
Highest income third	Republican voters	Relative lean	−6.5	−8.2	−4.8	21,554
Highest income third	Republican voters	Share of Republican voters, 2012 to 2024	43.3	41.7	44.8	6,908

The first bar is the group's share of the party's voters in 1976 to 1988 and the last bar is its share in 2012 to 2024. The two steps between them are the change from the group's share of all voters (group size) and the change from the group's Democratic share relative to the Democratic share of all voters (relative lean); they add exactly to the change. The default view is degree holders among Democratic voters, and the buttons show degree holders among Republican voters and the same split for white voters and for the highest income third.Note: The start and end bars carry Kish intervals. The two steps carry percentile intervals from a bootstrap that resamples respondents within each election (400 replicates, seed 12345).Note: The split is symmetric: the change in group size times the average relative lean, plus the change in relative lean times the average group size. Relative lean is the group's Democratic (or Republican) share divided by the party's share of all voters.Source: ANES Cumulative Data File, 1976 to 1988 and 2012 to 2024; weighted (weight_postelection where the file has one, else weight; each election has the same total weight); two-party presidential voters

The split for white voters runs the other way. White non-Hispanic voters fell from 84.7% (95% CI 83.6% to 85.7%) to 70.8% (95% CI 69.8% to 71.7%) of all voters, which gives a group-size term of −11.6 (95% CI −12.8 to −10.4) points for Democratic voters. Their Democratic share rose, from 39.4% (95% CI 37.8% to 40.9%) to 44.4% (95% CI 43.2% to 45.6%), but by less than the Democratic share of all voters, so their relative lean fell and that term is −3.1 (95% CI −4.6 to −1.5) points. Among Republican voters the group-size term is −16.2 (95% CI −17.8 to −14.5) points and the lean term is +7.3 (95% CI 5.7 to 8.9). A group's share of all voters is its share of adults weighted by its turnout, so the CPS bears on the group-size term. Over the same two windows, 1976 to 1988 and 2012 to 2024, the degree-holding share rose by 18.9 (95% CI 18.6 to 19.2) points among voters and by 17.0 (95% CI 16.8 to 17.2) points among adults aged 18 or older, so growth of the group among adults accounts for most of the rise among voters. Before 1994 the CPS does not record citizenship, so the earlier window counts all residents and the later window counts citizens.

The Democratic share of all voters rose from 45.9% (95% CI 44.5% to 47.4%) to 54.1% (95% CI 53.1% to 55.2%) between the two periods, a rise of 8.2 (95% CI 6.4 to 10.0) points. Figure 8 splits that change using cells defined by sex, three racial and ethnic groups and four age bands. The split is an accounting identity under those cells: it divides the change into two parts that sum to it and does not say why either part moved. The mix term, the part that comes from a different make-up of the voters across cells, is +2.9 (95% CI 2.1 to 3.8) points, and the within-cell term is +5.3 (95% CI 3.6 to 6.8) points. The within-cell term is not a measure of persuasion. It includes voters who changed their vote, new voters who replaced those who died, different people voting in the same cell from one election to the next, and changes in how the surveys sampled each cell. Repeated cross-sections cannot separate these.

The within-cell term also contains survey lean. The ANES share minus the certified count averaged +0.2 (95% CI −1.3 to 1.6) points over 1976 to 1988 and +3.0 (95% CI 1.9 to 4.0) over 2012 to 2024, a movement of 2.8 (95% CI 1.0 to 4.6) points, while the certified Democratic share rose by 5.4 points between the two sets of elections and the ANES share by 8.2 (95% CI 6.4 to 10.0). As much as 2.8 points of the within-cell term could therefore come from the surveys' lean and not from a change in how people voted.

Figure 8

The mix of voters accounts for +2.9 points of the change in the Democratic share and change within groups of the same sex, race and age for +5.3 points

Change in the Democratic share of two-party presidential voters from 1976 to 1988 to 2012 to 2024, split into the mix of voters across sex, race and age cells and change within cells, percent and percentage points; ANES; the buttons switch the voters

Show data

Data for The mix of voters accounts for +2.9 points of the change in the Democratic share and change within groups of the same sex, race and age for +5.3 points
Voters	Step	Value	95% CI low	95% CI high	n
All two-party voters	Democratic share, 1976 to 1988	45.9	44.5	47.4	4,770
All two-party voters	Change in the mix of voters	2.9	2.1	3.8	21,554
All two-party voters	Change within groups of the same sex, race and age	5.3	3.6	6.8	21,554
All two-party voters	Democratic share, 2012 to 2024	54.1	53.1	55.2	16,784
Voters with a four-year degree	Democratic share, 1976 to 1988	39.2	36.2	42.2	1,063
Voters with a four-year degree	Change in the mix of voters	1.1	−0.1	2.9	8,689
Voters with a four-year degree	Change within groups of the same sex, race and age	21.7	18.0	24.8	8,689
Voters with a four-year degree	Democratic share, 2012 to 2024	62.0	60.6	63.5	7,626
Voters without a four-year degree	Democratic share, 1976 to 1988	47.8	46.1	49.4	3,673
Voters without a four-year degree	Change in the mix of voters	4.7	3.5	5.9	12,652
Voters without a four-year degree	Change within groups of the same sex, race and age	−3.2	−5.4	−1.5	12,652
Voters without a four-year degree	Democratic share, 2012 to 2024	49.3	47.9	50.8	8,979

The first bar is the Democratic share in 1976 to 1988 and the last bar is the share in 2012 to 2024. The first step is the part of the change that comes from a different mix of voters across cells of sex, race and ethnicity (three groups) and age (four bands), and the second step is the part that comes from a different Democratic share inside the cells. The split is an accounting identity under these cells. The second step is not a measure of persuasion: it also holds replacement of voters, a different set of people voting, survey lean and sampling change. The buttons restrict the voters by degree.Note: The first and last bars carry Kish intervals and the two steps carry bootstrap intervals. The cells combine sex, race and ethnicity (white non-Hispanic, Black non-Hispanic, Hispanic or another race) and four age bands; a respondent with a missing value forms a cell of its own. A cell with no respondents in one period takes the other period's Democratic share, so it adds nothing to the within-cell term.Note: Intervals come from a bootstrap that resamples respondents within each election (400 replicates, seed 12345). The split is symmetric between the mix and the within-cell change.Source: ANES Cumulative Data File, 1976 to 1988 and 2012 to 2024; weighted (weight_postelection where the file has one, else weight; each election has the same total weight); two-party presidential voters

Inside a narrower group the split can look different. Among degree holders the Democratic share rose from 39.2% (95% CI 36.2% to 42.2%) to 62.0% (95% CI 60.6% to 63.5%), the mix term is +1.1 (95% CI −0.1 to 2.9) points and the within-cell term +21.7 (95% CI 18.0 to 24.8), so the sex, race and age make-up of graduates accounts for little of the rise. That does not make the rise persuasion, and the cells do not hold the group's own size fixed: degree holders went from 22.4% to 37.9% of voters. Among voters without a degree the Democratic share barely moved, from 47.8% (95% CI 46.1% to 49.4%) to 49.3% (95% CI 47.9% to 50.8%), and the flat total hides a mix term of +4.7 (95% CI 3.5 to 5.9) points and a within-cell term of −3.2 (95% CI −5.4 to −1.5) points.

Turnout enters in two places. The voters here reported voting, so a group's Democratic share can move because more or fewer of its Democratic-leaning members voted, and a group's share of voters moves with its share of eligible adults and its turnout. The surveys hold no vote preference for nonvoters, so the chapter cannot say how the coalitions would look if they had voted. Chapter 1.4, Cohort change in partisan sorting separates switching from replacement for ideology by following birth cohorts, which these vote questions are too thin to support.

The surveys differ on levels, and the differences reach the coalition shares

Self-reported vote can lean toward one party, and the surveys here do not lean equally. In 2024 the ANES gives a Democratic share of the two-party vote of 54.5% (95% CI 52.3% to 56.7%) and the CES gives 49.0% (95% CI 48.2% to 49.7%), against 49.2% in the certified count. The ANES share minus the count averages +3.0 (95% CI 1.9 to 4.0) points over the four elections from 2012 to 2024 and +0.2 (95% CI −1.3 to 1.6) points over those from 1976 to 1988, so the ANES changes in vote shares between the two periods are larger than the change in the count. Comparisons of survey reports with voter records are the usual check on self-reported turnout and vote [6], and neither is available for these files.

The composition of voters differs as well (Figure 9). In 2024 the CPS gives 44.9% (95% CI 44.4% to 45.4%) of voters with a degree, the ANES 39.9% (95% CI 37.7% to 42.1%) and the CES 41.8% (95% CI 41.1% to 42.5%). The two surveys disagree most about Republican voters: the ANES degree-holding share fell from 36.4% (95% CI 31.6% to 41.1%) in 2008 to 27.8% (95% CI 24.8% to 30.9%) in 2024, a change of −8.5 (95% CI −14.2 to −2.8) points, while the CES share rose from 28.2% (95% CI 27.2% to 29.1%) to 35.9% (95% CI 34.9% to 37.0%), a change of +7.8 (95% CI 6.4 to 9.2) points. The ANES mode series show where the fall sits. The 2008 baseline is face to face only. In the web samples the Republican share went from 31.9% (95% CI 28.6% to 35.2%) in 2012 to 30.8% (95% CI 25.9% to 35.7%) in 2024, a change of −1.1 (95% CI −7.0 to 4.8) points whose interval includes zero, and in the face-to-face samples from 37.9% (95% CI 32.2% to 43.6%) to 25.1% (95% CI 18.2% to 31.9%), a change of −12.8 (95% CI −21.8 to −3.9) points. The fall in the all-mode ANES series is therefore not present in the web samples, which like the CES are collected online, and those samples do not show the CES's rise either. The within-education vote differs too. In 2024 degree holders voted Democratic at 68.2% (95% CI 65.4% to 71.1%) in the ANES and 56.1% (95% CI 55.1% to 57.2%) in the CES, and voters without a degree at 45.3% (95% CI 42.2% to 48.5%) and 43.8% (95% CI 42.8% to 44.9%). If the 2024 ANES voters had the CPS education mix and kept their within-education vote shares, degree holders would be 55.1% (95% CI 53.4% to 57.2%) of Democratic voters and 32.1% (95% CI 29.8% to 34.1%) of Republican voters, against 50.0% and 27.8% unadjusted. The surveys also differ on race: in 2024 white non-Hispanic voters were 72.3% (95% CI 71.6% to 73.0%) of voters in the CES, 70.6% (95% CI 70.1% to 71.1%) in the CPS and 67.4% (95% CI 65.2% to 69.5%) in the ANES, and Black voters were 9.3% (95% CI 8.9% to 9.7%), 11.5% (95% CI 11.1% to 11.8%) and 11.2% (95% CI 9.8% to 12.7%).

Figure 9

The ANES shows the degree-holding share of Republican voters falling after 2008, the CES shows it rising, and the ANES web samples show no change

Share of voters with a four-year degree, percent, by presidential election: all voters (CPS, ANES, CES), Democratic voters and Republican voters (ANES, ANES by mode, CES); the buttons show the white and Black shares of voters; 95% intervals from the Kish effective sample size

Share with a four-year degree (%)

CPS voting supplement
ANES
ANES, face to face
ANES, web
CES

Show data

Data for The ANES shows the degree-holding share of Republican voters falling after 2008, the CES shows it rising, and the ANES web samples show no change
Measure	Voters	Survey	Election	Share (%)	95% CI low	95% CI high	n
Four-year degree	All voters	CPS voting supplement	1996	29.4	29.0	29.8	50,025
Four-year degree	All voters	CPS voting supplement	2000	31.2	30.7	31.6	50,010
Four-year degree	All voters	CPS voting supplement	2004	32.4	32.0	32.9	63,052
Four-year degree	All voters	CPS voting supplement	2008	34.1	33.7	34.6	59,975
Four-year degree	All voters	CPS voting supplement	2012	36.6	36.2	37.0	59,597
Four-year degree	All voters	CPS voting supplement	2016	39.6	39.2	40.1	58,533
Four-year degree	All voters	CPS voting supplement	2020	41.5	41.0	41.9	55,170
Four-year degree	All voters	CPS voting supplement	2024	44.9	44.4	45.4	48,387
Four-year degree	All voters	ANES	1996	28.7	25.6	31.8	1,032
Four-year degree	All voters	ANES	2000	30.4	27.4	33.4	1,118
Four-year degree	All voters	ANES	2004	31.3	27.8	34.8	811
Four-year degree	All voters	ANES	2008	33.3	30.4	36.3	1,533
Four-year degree	All voters	ANES	2012	33.6	31.7	35.6	4,144
Four-year degree	All voters	ANES	2016	37.7	35.4	39.9	2,571
Four-year degree	All voters	ANES	2020	40.4	38.7	42.2	5,987
Four-year degree	All voters	ANES	2024	39.9	37.7	42.1	3,903
Four-year degree	All voters	ANES, face to face	2008	33.3	30.3	36.2	1,533
Four-year degree	All voters	ANES, face to face	2012	35.9	32.3	39.5	1,327
Four-year degree	All voters	ANES, face to face	2016	39.9	35.5	44.3	745
Four-year degree	All voters	ANES, face to face	2020	n/a	n/a	n/a	n/a
Four-year degree	All voters	ANES, face to face	2024	41.1	36.0	46.2	688
Four-year degree	All voters	ANES, web	2012	32.4	30.1	34.7	2,817
Four-year degree	All voters	ANES, web	2016	37.1	34.5	39.8	1,846
Four-year degree	All voters	ANES, web	2020	n/a	n/a	n/a	n/a
Four-year degree	All voters	ANES, web	2024	42.5	39.0	46.0	1,357
Four-year degree	All voters	CES	2008	29.8	29.1	30.5	23,585
Four-year degree	All voters	CES	2012	30.7	30.0	31.4	37,789
Four-year degree	All voters	CES	2016	30.9	30.3	31.5	40,891
Four-year degree	All voters	CES	2020	38.8	38.2	39.5	43,872
Four-year degree	All voters	CES	2024	41.8	41.1	42.5	42,008
Four-year degree	Democratic voters	ANES	1996	24.2	20.4	28.1	599
Four-year degree	Democratic voters	ANES	2000	29.3	25.2	33.4	590
Four-year degree	Democratic voters	ANES	2004	31.9	26.8	36.9	399
Four-year degree	Democratic voters	ANES	2008	30.9	27.2	34.5	1,020
Four-year degree	Democratic voters	ANES	2012	33.3	30.7	35.9	2,469
Four-year degree	Democratic voters	ANES	2016	43.0	39.8	46.3	1,341
Four-year degree	Democratic voters	ANES	2020	47.1	44.7	49.5	3,461
Four-year degree	Democratic voters	ANES	2024	50.0	47.0	53.0	2,178
Four-year degree	Democratic voters	ANES, face to face	2008	30.8	27.2	34.5	1,020
Four-year degree	Democratic voters	ANES, face to face	2012	34.1	29.6	38.6	915
Four-year degree	Democratic voters	ANES, face to face	2016	47.1	40.9	53.4	385
Four-year degree	Democratic voters	ANES, face to face	2020	n/a	n/a	n/a	n/a
Four-year degree	Democratic voters	ANES, face to face	2024	54.7	47.8	61.7	387
Four-year degree	Democratic voters	ANES, web	2012	32.9	29.7	36.1	1,554
Four-year degree	Democratic voters	ANES, web	2016	41.7	38.0	45.4	970
Four-year degree	Democratic voters	ANES, web	2020	n/a	n/a	n/a	n/a
Four-year degree	Democratic voters	ANES, web	2024	51.0	46.2	55.8	786
Four-year degree	Democratic voters	CES	2008	31.2	30.2	32.3	11,519
Four-year degree	Democratic voters	CES	2012	32.9	31.9	34.0	19,048
Four-year degree	Democratic voters	CES	2016	36.1	35.2	37.0	22,136
Four-year degree	Democratic voters	CES	2020	45.8	44.9	46.6	26,179
Four-year degree	Democratic voters	CES	2024	47.9	46.9	49.0	23,517
Four-year degree	Republican voters	ANES	1996	34.9	29.9	40.0	433
Four-year degree	Republican voters	ANES	2000	31.6	27.2	35.9	528
Four-year degree	Republican voters	ANES	2004	30.7	25.8	35.5	412
Four-year degree	Republican voters	ANES	2008	36.4	31.6	41.1	513
Four-year degree	Republican voters	ANES	2012	34.0	31.1	36.9	1,675
Four-year degree	Republican voters	ANES	2016	31.7	28.6	34.8	1,230
Four-year degree	Republican voters	ANES	2020	32.0	29.5	34.5	2,526
Four-year degree	Republican voters	ANES	2024	27.8	24.8	30.9	1,725
Four-year degree	Republican voters	ANES, face to face	2008	36.2	31.4	41.0	513
Four-year degree	Republican voters	ANES, face to face	2012	37.9	32.2	43.6	412
Four-year degree	Republican voters	ANES, face to face	2016	31.8	25.8	37.8	360
Four-year degree	Republican voters	ANES, face to face	2020	n/a	n/a	n/a	n/a
Four-year degree	Republican voters	ANES, face to face	2024	25.1	18.2	31.9	301
Four-year degree	Republican voters	ANES, web	2012	31.9	28.6	35.2	1,263
Four-year degree	Republican voters	ANES, web	2016	32.0	28.4	35.6	876
Four-year degree	Republican voters	ANES, web	2020	n/a	n/a	n/a	n/a
Four-year degree	Republican voters	ANES, web	2024	30.8	25.9	35.7	571
Four-year degree	Republican voters	CES	2008	28.2	27.2	29.1	12,066
Four-year degree	Republican voters	CES	2012	28.1	27.2	29.1	18,741
Four-year degree	Republican voters	CES	2016	25.6	24.8	26.4	18,755
Four-year degree	Republican voters	CES	2020	31.2	30.4	32.0	17,693
Four-year degree	Republican voters	CES	2024	35.9	34.9	37.0	18,491
White non-Hispanic voters	All voters	CPS voting supplement	1996	82.5	82.1	82.8	50,025
White non-Hispanic voters	All voters	CPS voting supplement	2000	80.6	80.2	81.0	50,010
White non-Hispanic voters	All voters	CPS voting supplement	2004	79.2	78.8	79.6	63,052
White non-Hispanic voters	All voters	CPS voting supplement	2008	76.3	75.9	76.7	59,975
White non-Hispanic voters	All voters	CPS voting supplement	2012	73.7	73.3	74.1	59,597
White non-Hispanic voters	All voters	CPS voting supplement	2016	73.3	72.9	73.7	58,533
White non-Hispanic voters	All voters	CPS voting supplement	2020	71.0	70.6	71.5	55,170
White non-Hispanic voters	All voters	CPS voting supplement	2024	70.6	70.1	71.1	48,387
White non-Hispanic voters	All voters	ANES	1996	78.9	76.1	81.6	1,031
White non-Hispanic voters	All voters	ANES	2000	77.8	75.1	80.6	1,113
White non-Hispanic voters	All voters	ANES	2004	74.9	71.6	78.1	800
White non-Hispanic voters	All voters	ANES	2008	74.1	71.4	76.9	1,530
White non-Hispanic voters	All voters	ANES	2012	72.7	70.9	74.5	4,181
White non-Hispanic voters	All voters	ANES	2016	72.6	70.6	74.7	2,575
White non-Hispanic voters	All voters	ANES	2020	70.4	68.8	72.0	6,030
White non-Hispanic voters	All voters	ANES	2024	67.4	65.2	69.5	3,897
White non-Hispanic voters	All voters	CES	2008	76.5	75.8	77.2	23,585
White non-Hispanic voters	All voters	CES	2012	78.6	78.0	79.3	37,789
White non-Hispanic voters	All voters	CES	2016	76.7	76.2	77.3	40,891
White non-Hispanic voters	All voters	CES	2020	73.1	72.6	73.7	43,872
White non-Hispanic voters	All voters	CES	2024	72.3	71.6	73.0	42,008
Black voters	All voters	CPS voting supplement	1996	10.6	10.3	10.9	50,025
Black voters	All voters	CPS voting supplement	2000	11.2	10.9	11.5	50,010
Black voters	All voters	CPS voting supplement	2004	11.0	10.7	11.3	63,052
Black voters	All voters	CPS voting supplement	2008	12.1	11.8	12.4	59,975
Black voters	All voters	CPS voting supplement	2012	12.9	12.6	13.2	59,597
Black voters	All voters	CPS voting supplement	2016	11.9	11.6	12.2	58,533
Black voters	All voters	CPS voting supplement	2020	11.7	11.4	12.0	55,170
Black voters	All voters	CPS voting supplement	2024	11.5	11.1	11.8	48,387
Black voters	All voters	ANES	1996	9.9	7.9	12.0	1,031
Black voters	All voters	ANES	2000	11.2	9.2	13.3	1,113
Black voters	All voters	ANES	2004	13.7	11.1	16.3	800
Black voters	All voters	ANES	2008	12.1	10.0	14.1	1,530
Black voters	All voters	ANES	2012	13.1	11.7	14.4	4,181
Black voters	All voters	ANES	2016	11.2	9.7	12.6	2,575
Black voters	All voters	ANES	2020	10.5	9.4	11.6	6,030
Black voters	All voters	ANES	2024	11.2	9.8	12.7	3,897
Black voters	All voters	CES	2008	11.6	11.1	12.1	23,585
Black voters	All voters	CES	2012	8.6	8.2	9.1	37,789
Black voters	All voters	CES	2016	9.6	9.2	10.0	40,891
Black voters	All voters	CES	2020	10.6	10.2	10.9	43,872
Black voters	All voters	CES	2024	9.3	8.9	9.7	42,008

The Republican panel is where the surveys part: from 2008 to 2024 the ANES share changed by −8.5 points and the CES share by +7.8 points. The dashed lines split the ANES by interview mode: the 2008 baseline is face to face only, and between 2012 and 2024 the web share changed by −1.1 points and the face-to-face share by −12.8 points. In 2024 the CPS gives 44.9% of all voters with a degree, the ANES 39.9% and the CES 41.8%. The CPS counts all voters, including third-party voters, and the ANES and the CES count two-party voters.Note: The ANES 2012, 2016 and 2024 samples mix interview modes and the 2020 study has no mode weights, so the mode lines have no 2020 point; the 2008 study was face to face only. The all-mode ANES line includes panel and paper respondents in 2024.Note: The CES is an opt-in panel and its intervals do not cover how respondents were recruited; the CPS intervals ignore the survey design. Intervals are model based (Kish effective sample size, normal approximation) and ignore clustering in the sample design.Source: CPS November voting supplement 1996 to 2024 (weighted, VOSUPPWT); ANES Cumulative Data File 1996 to 2024 (weighted; mode series use weight_ftf and weight_web); CES cumulative file 2008 to 2024 (weighted, weight_cumulative)

In the 2024 ANES, degree holders were 54.7% (95% CI 47.8% to 61.7%) of Democratic voters among face-to-face respondents and 51.0% (95% CI 46.2% to 55.8%) among web respondents, and the two intervals overlap. Religion shows a gap too: the 2024 ANES Time Series puts voters with no religion at 34.2% (95% CI 31.1% to 37.2%) of Democratic voters and the CES at 45.0% (95% CI 43.9% to 46.0%). Both surveys show a rise in the degree-holding share of Democratic voters, they disagree about the Republican share, and the levels in either survey are best read as a range.

What these data cannot show

The surveys interview different people in each election, so the chapter compares groups and not individuals. It cannot say whether a voter changed party, how much of a change comes from older voters leaving the electorate and younger ones entering it, or how much comes from different people turning out. The group terms of the decompositions are arithmetic and describe a split of numbers, not a mechanism, and the within-cell term in particular absorbs every change that the cells do not hold fixed. The split is also sensitive to the groups chosen: degree holders were a smaller share of voters in the first period than in the second.

Vote and turnout are self-reports. The ANES leans Democratic in its late elections, the CES is an opt-in panel whose weights match it to the adult population and not to voters, and the CPS has no vote choice, so no survey here measures a party's voters exactly. Nonvoters have no vote preference in these data, third-party voters are outside the two-party frame, and the CES file used here has no validated turnout. The ANES samples of 2012, 2016 and 2024 mix interview modes and the 2020 study has no mode weights, so a change between those elections can include a change of mode. Intervals are model based and leave out the sample designs; the CPS intervals in particular are narrower than the survey's own design-based errors.

Several groups are coarse. Education below a degree is not analysed, because the ANES cumulative file defines its lower groups differently after 2016. The ANES identifies Hispanic respondents from 1966 and has few of them before the 1990s, so the early race cells pool Hispanic voters with voters of other races for the coalition shares and drop the other-races group for the vote shares. The relative income thirds do not describe dollar income, and religion in the ANES cumulative file stops at 2020 with four categories. The two periods, 1976 to 1988 and 2012 to 2024, were chosen to give each group enough respondents, and a different choice of years would give different numbers. The chapter does not examine why any group's vote changed, how the parties' candidates, issues or campaigns affected it, or how coalitions differ below the national level.

Data sources

ANES Cumulative Data FileCumulative Data File, presidential years 1952 to 2024Years: 1948-2024 (33 study years; even years only after 2004; no 2006, 2010 or 2014 study) · Weights: weight_postelection where available, else VCF0009z · License: Free download after accepting the ANES terms of use; cite the release.
ANES 2024 Time Series2024 Time Series, summary religion code, one comparison with the CESYears: 2024 (pre-election and post-election interviews) · Weights: V240107b · License: Free download after accepting the ANES terms of use; cite the release.
Cooperative Election StudyCumulative Common Content, presidential years 2008 to 2024Years: 2006-2024 (every year; self-reported turnout is available in the 10 even years) · Weights: weight_cumulative · License: See the provider's terms of use.
CPS Voting and Registration SupplementNovember Voting and Registration Supplement, presidential years 1996 to 2024 (and 1976 to 1988 for one comparison), adults aged 18 or olderYears: 1976-2024 (25 November samples, biennial, even years) · Weights: VOSUPPWT · License: IPUMS CPS conditions of use: a free extract after registration, no fee may be charged for the data, IPUMS must be cited, and publications are to be added to the IPUMS bibliography. The underlying Census Bureau data are public domain.
State presidential returnsState presidential returns 1976 to 2024, certified national two-party shareYears: 1976-2024 (13 elections) · Weights: none (full counts) · License: See the provider's terms of use.

All data sources

Methods

Sample: ANES Cumulative Data File respondents in the 19 presidential elections from 1952 to 2024 who reported voting for the Democratic or the Republican candidate (34,423 respondents with a positive vote weight); CES cumulative file respondents in 2008, 2012, 2016, 2020 and 2024 who reported such a vote (188,145); CPS November voting supplement respondents in the presidential-election years 1996 to 2024 who were citizens aged 18 or older with a positive supplement weight and a turnout answer (eligible adults); for the comparison of the two pooled windows, adults aged 18 or older with a turnout answer in 1976, 1980, 1984, 1988, 2012, 2016, 2020 and 2024 (citizenship is recorded from 1994, so the earlier window counts all residents and the later one citizens). Third-party voters and nonvoters are outside the ANES and CES frame. A coalition is a party's two-party voters. Weights: ANES weight_postelection in the elections where the file has one (1996, 2000, 2004, 2008, 2016, 2020, 2024), otherwise weight (VCF0009z), the rule used by the gender-gap chapter; CES weight_cumulative; CPS VOSUPPWT. A pooled period (1976, 1980, 1984 and 1988; 2012, 2016, 2020 and 2024) divides each election's weights by that election's total vote weight, so every election counts equally. Groups: four-year degree or more against less (ANES college_grad; CES education_4cat College or Post-grad; education below a degree is not split in the ANES because the cumulative file defines its lower groups differently after 2016); relative income thirds (the ANES percentile income groups regrouped into 0 to 33, 34 to 67 and 68 to 100); race and ethnicity from the ANES four-category variable (Hispanic respondents identified from 1966; the other-races group has too few respondents to publish in the earlier period) and the CES five-category variable; religion from the ANES cumulative four-category variable (Protestant, Catholic, Jewish, other or none; the 2024 values are unusable and are not used) and the CES five-category variable (2012 to 2024; the CES religion categories are not comparable between 2008 and 2012, when the Protestant share jumps and the other-religion share falls); age in four bands built from age in years; sex from the survey's two-category variable. Definitions: for a group g, c(g) is the group's share of a party's voters, d(g) the Democratic share of the group's voters, s(g) the group's share of all voters and D the Democratic share of all voters, so that c_D(g) = s(g) d(g) / D and c_R(g) = s(g) (1 - d(g)) / (1 - D). A change in c between two periods splits exactly into a group-size term, (s1 - s0) times the mean of the two relative leans, and a relative-lean term, (r1 - r0) times the mean of the two group shares, with r the relative lean. The vote decomposition splits the change in the Democratic share into a mix term, the sum over cells of sex, race and age (a missing value is its own category) of the change in the cell's weight share times the mean of the two Democratic shares, and a within-cell term, the sum of the change in the cell's Democratic share times the mean of the two weight shares; a cell empty in one period takes the other period's Democratic share. Intervals: 95%. Shares and differences of shares from independent periods use the Kish effective sample size and a normal approximation, with standard errors combined in quadrature; the terms of the two decompositions and the CPS-adjusted degree shares come from a percentile bootstrap that resamples respondents within each election (400 replicates, seed 12345); the CPS voter-minus-eligible difference uses a Taylor linearization with the CPS weights, and the CPS counterfactual (2024 voters with 1996 turnout rates for degree holders and other adults) a bootstrap within year. All intervals are model based. The CES is an opt-in panel and the CPS extract has no replicate weights, so those intervals leave out how respondents were recruited and the survey design. Suppression: a published share needs at least 100 voters in its denominator, a Kish effective sample of at least 100 and at least 30 respondents in the group and party; the other-races vote share in the earlier period (57 respondents) and the 1952 Democratic degree share are dropped. Survey mode: the ANES 2012, 2016 and 2024 samples mix modes and the 2020 study has no mode weights; the text and the survey-check figure report face-to-face and web estimates (2008 is face to face only). Benchmark: the certified two-party share from the MIT Election Data and Science Lab state returns, full counts.

References

Kish, Leslie (1965). Survey Sampling. John Wiley and Sons.
Axelrod, Robert (1972). Where the Votes Come From: An Analysis of Electoral Coalitions, 1952-1968. American Political Science Review 66(1): 11-20. doi:10.2307/1959275
McDonald, Michael P.; Popkin, Samuel L. (2001). The Myth of the Vanishing Voter. American Political Science Review 95(4): 963-974. doi:10.1017/S0003055400400134
Hur, Aram; Achen, Christopher H. (2013). Coding Voter Turnout Responses in the Current Population Survey. Public Opinion Quarterly 77(4): 985-993. doi:10.1093/poq/nft042
Kitagawa, Evelyn M. (1955). Components of a Difference Between Two Rates. Journal of the American Statistical Association 50(272): 1168-1194. doi:10.1080/01621459.1955.10501299
Ansolabehere, Stephen; Hersh, Eitan (2012). Validation: What Big Data Reveal About Survey Misreporting and the Real Electorate. Political Analysis 20(4): 437-459. doi:10.1093/pan/mps023

How to cite

@misc{singh2026thetwocoalitions,
  author = {Singh, Vishal},
  title = {The two coalitions},
  year = {2026},
  url = {https://vishalsingh.org/politics/the-two-coalitions},
  note = {In The Public Mind. Published 3 October 2026.}
}

A group's Democratic share and its share of Democratic voters are different numbers

Eligible adults, voters, party identifiers and nonvoters differ in make-up

Degree holders and the highest income third became larger parts of the Democratic coalition

Both coalitions became less white, and both became less Protestant

The oldest age band grew in both coalitions, and women's share fell in both

The Democratic share rose more among degree holders than in any other group

A coalition share moves with group size and with relative lean

A change in a group's vote share does not identify persuasion

The surveys differ on levels, and the differences reach the coalition shares

What these data cannot show

Data sources

Methods

References

How to cite