Chapter 3.7
Turnout
From 1994 to 2024 turnout among citizens rose with education and family income in every election and was lowest in the youngest age group; in 2024 it was 79.6% for citizens with a bachelor's degree and 39.1% for those without a high school diploma, and about 70% of that gap arose at registration.
DraftFigures and numbers may change as the analysis is rebuilt.what is still being checked
The comparison of CPS turnout rates with the voting-eligible-population turnout series is not yet made because that series was not among the data used.
In the November 2024 Current Population Survey (CPS), 65.3% (95% CI 64.9% to 65.7%)65.3% percentEstimandweighted share who reported voting (Census convention: nonresponse counted as not voting)Interval95% CI 64.9% to 65.7%n72,590WeightsVOSUPPWTPopulationcitizens 18 and olderPeriod2024 presidential electionSourceCPS Voting and Registration Supplement of citizens reported voting, against 66.8%66.8% percentEstimandweighted share who reported voting (Census convention: nonresponse counted as not voting)Interval95% CI 66.4% to 67.1%n81,898WeightsVOSUPPWTPopulationcitizens 18 and olderPeriod2020 presidential electionSourceCPS Voting and Registration Supplement in 2020 and 58.4%58.4% percentEstimandweighted share who reported voting (Census convention: nonresponse counted as not voting)Interval95% CI 58.0% to 58.7%n84,090WeightsVOSUPPWTPopulationcitizens 18 and olderPeriod1996 presidential electionSourceCPS Voting and Registration Supplement in 1996, the first presidential year in which the extract identifies citizens. In 2024 turnout rose with age, education and family income. Citizens aged sixty-five and older reported voting at 74.7%74.7% percentEstimandweighted share who reported voting (Census convention: nonresponse counted as not voting)Interval95% CI 74.0% to 75.4%n20,905WeightsVOSUPPWTPopulationcitizens aged 65 and olderPeriod2024 presidential electionSourceCPS Voting and Registration Supplement and those aged eighteen to twenty-nine at 50.7%50.7% percentEstimandweighted share who reported voting (Census convention: nonresponse counted as not voting)Interval95% CI 49.7% to 51.8%n12,303WeightsVOSUPPWTPopulationcitizens aged 18 to 29Period2024 presidential electionSourceCPS Voting and Registration Supplement. Among citizens twenty-five and older, 79.6%79.6% percentEstimandweighted share who reported voting (Census convention: nonresponse counted as not voting)Interval95% CI 79.0% to 80.2%n25,742WeightsVOSUPPWTPopulationcitizens 25 and older with a bachelor's degree or morePeriod2024 presidential electionSourceCPS Voting and Registration Supplement of those with a bachelor's degree voted, against 39.1%39.1% percentEstimandweighted share who reported voting (Census convention: nonresponse counted as not voting)Interval95% CI 37.4% to 40.8%n4,008WeightsVOSUPPWTPopulationcitizens 25 and older without a high school diplomaPeriod2024 presidential electionSourceCPS Voting and Registration Supplement of those without a high school diploma. The highest third of family income voted at 74.5%74.5% percentEstimandweighted share who reported voting (Census convention: nonresponse counted as not voting)Interval95% CI 73.9% to 75.1%n28,984WeightsVOSUPPWTPopulationcitizens 18 and older in the highest third of family income that yearPeriod2024 presidential electionSourceCPS Voting and Registration Supplement and the lowest third at 54.3%54.3% percentEstimandweighted share who reported voting (Census convention: nonresponse counted as not voting)Interval95% CI 53.6% to 55.0%n26,368WeightsVOSUPPWTPopulationcitizens 18 and older in the lowest third of family income that yearPeriod2024 presidential electionSourceCPS Voting and Registration Supplement. White non-Hispanic citizens voted at 70.5%70.5% percentEstimandweighted share who reported voting (Census convention: nonresponse counted as not voting)Interval95% CI 70.0% to 70.9%n52,237WeightsVOSUPPWTPopulationWhite non-Hispanic citizens 18 and olderPeriod2024 presidential electionSourceCPS Voting and Registration Supplement, Black citizens at 60.1%60.1% percentEstimandweighted share who reported voting (Census convention: nonresponse counted as not voting)Interval95% CI 58.7% to 61.4%n6,636WeightsVOSUPPWTPopulationBlack non-Hispanic citizens 18 and olderPeriod2024 presidential electionSourceCPS Voting and Registration Supplement, Asian citizens at 57.6%57.6% percentEstimandweighted share who reported voting (Census convention: nonresponse counted as not voting)Interval95% CI 55.9% to 59.4%n3,853WeightsVOSUPPWTPopulationAsian or Pacific Islander non-Hispanic citizens 18 and olderPeriod2024 presidential electionSourceCPS Voting and Registration Supplement and Hispanic citizens at 50.6%50.6% percentEstimandweighted share who reported voting (Census convention: nonresponse counted as not voting)Interval95% CI 49.4% to 51.8%n8,064WeightsVOSUPPWTPopulationHispanic citizens 18 and olderPeriod2024 presidential electionSourceCPS Voting and Registration Supplement.
Registration accounts for most of the 2024 gaps by age, education, income and ethnicity, and the share for the White-Black gap depends on how nonresponse is coded. In 2024 the registered share of citizens ran from 51.4%51.4% percentEstimandweighted share registered (voted, or reported being registered; nonresponse counted as not registered)Interval95% CI 49.6% to 53.1%n4,008WeightsVOSUPPWTPopulationcitizens 25 and older without a high school diplomaPeriod2024 presidential electionSourceCPS Voting and Registration Supplement among those without a high school diploma to 84.5%84.5% percentEstimandweighted share registered (voted, or reported being registered; nonresponse counted as not registered)Interval95% CI 84.0% to 85.0%n25,742WeightsVOSUPPWTPopulationcitizens 25 and older with a bachelor's degree or morePeriod2024 presidential electionSourceCPS Voting and Registration Supplement among those with a bachelor's degree, while the share of the registered who voted ran only from 76.1%76.1% percentEstimandweighted share of the registered who votedInterval95% CI 74.0% to 78.2%n2,115WeightsVOSUPPWTPopulationcitizens 25 and older without a high school diploma who were registeredPeriod2024 presidential electionSourceCPS Voting and Registration Supplement to 94.2%94.2% percentEstimandweighted share of the registered who votedInterval95% CI 93.8% to 94.6%n21,949WeightsVOSUPPWTPopulationcitizens 25 and older with a bachelor's degree or more who were registeredPeriod2024 presidential electionSourceCPS Voting and Registration Supplement. About 70% (95% CI 67% to 73%)70% percent of the turnout differenceEstimandshare of the turnout difference that arises at registration (symmetric split of the product of the registered share and turnout among the registered), citizens 25 and older with a bachelor's degree or more minus those without a high school diplomaInterval95% CI 67% to 73%n29,750WeightsVOSUPPWTPopulationcitizens 25 and older with a bachelor's degree or more and without a high school diplomaPeriod2024SourceCPS Voting and Registration SupplementNoteSeeded percentile bootstrap over respondents (500 replicates); it omits the CPS design effect and is too narrow by an unknown factor. The share depends on the order of the split and on how nonresponse is coded. of the education gap in turnout arises at registration, from 62%62% percent of the turnout differenceEstimandlower end of the registration share across the two orders of the split, citizens 25 and older with a bachelor's degree or more minus those without a high school diplomaWeightsVOSUPPWTPopulationcitizens 25 and older with a bachelor's degree or more and without a high school diplomaPeriod2024SourceCPS Voting and Registration Supplement to 77%77% percent of the turnout differenceEstimandupper end of the registration share across the two orders of the split, citizens 25 and older with a bachelor's degree or more minus those without a high school diplomaWeightsVOSUPPWTPopulationcitizens 25 and older with a bachelor's degree or more and without a high school diplomaPeriod2024SourceCPS Voting and Registration Supplement depending on the order of the split (Figure 2).
The data are the Voting and Registration Supplement, a November addition to the Census Bureau's monthly household survey, in the harmonized IPUMS extract. The extract holds 2,494,2182,494,218 respondentsEstimandcount of person recordsWeightsnone (count of records)Populationadults 18 and older with a positive supplement weightPeriod1976 to 2024SourceCPS Voting and Registration Supplement adults from the biennial supplements of 1976 to 2024, of whom 1,422,3111,422,311 respondentsEstimandcount of person recordsWeightsnone (count of records)Populationcitizens 18 and olderPeriod1994 to 2024SourceCPS Voting and Registration Supplement are citizens interviewed from 1994, the first year the extract records citizenship. Respondents say whether they voted, and no report is checked against a record. A person who does not answer counts as not voting, the Census Bureau's convention, and the share who do not answer rose from 10.8%10.8% percentEstimandweighted share who did not answer the voting question (refused, did not know, no response)n96,452WeightsVOSUPPWTPopulationcitizens 18 and olderPeriod2004SourceCPS Voting and Registration Supplement of citizens in 2004 to 17.7%17.7% percentEstimandweighted share who did not answer the voting question (refused, did not know, no response)n88,749WeightsVOSUPPWTPopulationcitizens 18 and olderPeriod2018SourceCPS Voting and Registration Supplement in 2018 and was 14.1%14.1% percentEstimandweighted share who did not answer the voting question (refused, did not know, no response)n72,590WeightsVOSUPPWTPopulationcitizens 18 and olderPeriod2024SourceCPS Voting and Registration Supplement in 2024. Hur and Achen attribute an apparent fall in 2008 relative to 2004 to this coding and to rising nonresponse [1]. In the series here, turnout among citizens was 63.8%63.8% percentEstimandweighted share who reported voting (Census convention: nonresponse counted as not voting)Interval95% CI 63.5% to 64.2%n96,452WeightsVOSUPPWTPopulationcitizens 18 and olderPeriod2004 presidential electionSourceCPS Voting and Registration Supplement in 2004 and 63.6%63.6% percentEstimandweighted share who reported voting (Census convention: nonresponse counted as not voting)Interval95% CI 63.3% to 64.0%n92,360WeightsVOSUPPWTPopulationcitizens 18 and olderPeriod2008 presidential electionSourceCPS Voting and Registration Supplement in 2008. Turnout here is the share of a group who report voting, and the default group is citizens eighteen and older. Wolfinger and Rosenstone's study of who votes analyzed the 1972 and 1974 supplements [2].Estimates are weighted with the supplement person weight, a survey weightSurvey weightsNumbers that scale each respondent so the sample matches the population on known traits such as age, race, education and region. A weighted average counts a respondent with a large weight for more than one with a small weight.. Intervals are 95% confidence intervalsConfidence intervalA range around an estimate that would contain the true value in 95% of repeated samples if the sampling model held. The book reports 95% intervals unless it says otherwise. from the Kish effective sample sizeEffective sample sizeThe number of equally weighted respondents that would give the same precision as the weighted sample. It is smaller than the raw count when the weights vary a lot, and it sets how wide the confidence interval is. [3], which treat the weighted sample as a simple random sample. The extract has no replicate weights, so the intervals leave out the clustering and stratification of the CPS design and are narrower than design-based intervals by an unknown factor.
Citizen turnout in 2018 to 2024 was higher than in any election of its type from 1994 to 2016
Figure 1 plots turnout in every election from 1994. In presidential years it was 58.4%58.4% percentEstimandweighted share who reported voting (Census convention: nonresponse counted as not voting)Interval95% CI 58.0% to 58.7%n84,090WeightsVOSUPPWTPopulationcitizens 18 and olderPeriod1996 presidential electionSourceCPS Voting and Registration Supplement in 1996, 63.8%63.8% percentEstimandweighted share who reported voting (Census convention: nonresponse counted as not voting)Interval95% CI 63.5% to 64.2%n96,452WeightsVOSUPPWTPopulationcitizens 18 and olderPeriod2004 presidential electionSourceCPS Voting and Registration Supplement in 2004 and 63.6%63.6% percentEstimandweighted share who reported voting (Census convention: nonresponse counted as not voting)Interval95% CI 63.3% to 64.0%n92,360WeightsVOSUPPWTPopulationcitizens 18 and olderPeriod2008 presidential electionSourceCPS Voting and Registration Supplement in 2008, then 61.8%61.8% percentEstimandweighted share who reported voting (Census convention: nonresponse counted as not voting)Interval95% CI 61.5% to 62.2%n94,311WeightsVOSUPPWTPopulationcitizens 18 and olderPeriod2012 presidential electionSourceCPS Voting and Registration Supplement in 2012 and 61.4%61.4% percentEstimandweighted share who reported voting (Census convention: nonresponse counted as not voting)Interval95% CI 61.0% to 61.7%n93,794WeightsVOSUPPWTPopulationcitizens 18 and olderPeriod2016 presidential electionSourceCPS Voting and Registration Supplement in 2016, and 66.8%66.8% percentEstimandweighted share who reported voting (Census convention: nonresponse counted as not voting)Interval95% CI 66.4% to 67.1%n81,898WeightsVOSUPPWTPopulationcitizens 18 and olderPeriod2020 presidential electionSourceCPS Voting and Registration Supplement in 2020 and 65.3%65.3% percentEstimandweighted share who reported voting (Census convention: nonresponse counted as not voting)Interval95% CI 64.9% to 65.7%n72,590WeightsVOSUPPWTPopulationcitizens 18 and olderPeriod2024 presidential electionSourceCPS Voting and Registration Supplement in 2024. In midterm years it stayed between 41.9%41.9% percentEstimandweighted share who reported voting (Census convention: nonresponse counted as not voting)Interval95% CI 41.6% to 42.3%n96,267WeightsVOSUPPWTPopulationcitizens 18 and olderPeriod2014 midterm electionSourceCPS Voting and Registration Supplement in 2014 and 48.4%48.4% percentEstimandweighted share who reported voting (Census convention: nonresponse counted as not voting)Interval95% CI 48.0% to 48.8%n95,272WeightsVOSUPPWTPopulationcitizens 18 and olderPeriod1994 midterm electionSourceCPS Voting and Registration Supplement in 1994 until the 2018 midterm, at 53.4%53.4% percentEstimandweighted share who reported voting (Census convention: nonresponse counted as not voting)Interval95% CI 53.1% to 53.8%n88,749WeightsVOSUPPWTPopulationcitizens 18 and olderPeriod2018 midterm electionSourceCPS Voting and Registration Supplement, and the 2022 midterm, at 52.2%52.2% percentEstimandweighted share who reported voting (Census convention: nonresponse counted as not voting)Interval95% CI 51.8% to 52.6%n73,083WeightsVOSUPPWTPopulationcitizens 18 and olderPeriod2022 midterm electionSourceCPS Voting and Registration Supplement. The presidential year exceeded the next midterm by 19.9 (95% CI 19.4 to 20.4)19.9 percentage pointsEstimanddifference in weighted turnout, presidential year minus the next midtermInterval95% CI 19.4 to 20.4n190,578WeightsVOSUPPWTPopulationcitizens 18 and olderPeriod2012 presidential minus 2014 midtermSourceCPS Voting and Registration Supplement points in 2012, by 7.9 (95% CI 7.4 to 8.5)7.9 percentage pointsEstimanddifference in weighted turnout, presidential year minus the next midtermInterval95% CI 7.4 to 8.5n182,543WeightsVOSUPPWTPopulationcitizens 18 and olderPeriod2016 presidential minus 2018 midtermSourceCPS Voting and Registration Supplement in 2016 and by 14.6 (95% CI 14.0 to 15.1)14.6 percentage pointsEstimanddifference in weighted turnout, presidential year minus the next midtermInterval95% CI 14.0 to 15.1n154,981WeightsVOSUPPWTPopulationcitizens 18 and olderPeriod2020 presidential minus 2022 midtermSourceCPS Voting and Registration Supplement in 2020, against 13.1 (95% CI 12.5 to 13.6)13.1 percentage pointsEstimanddifference in weighted turnout, presidential year minus the next midtermInterval95% CI 12.5 to 13.6n168,166WeightsVOSUPPWTPopulationcitizens 18 and olderPeriod1996 presidential minus 1998 midtermSourceCPS Voting and Registration Supplement in 1996.
Figure 1
Turnout among citizens in 2018 to 2024 was higher than in any election of its type from 1994 to 2016
Percent of citizens 18 and older who reported voting, presidential and midterm elections, 1994 to 2024 (the second button counts all residents, 1976 to 2024); 95% intervals from the Kish effective sample size
Show data
| Year | Election type | Universe | Turnout (%) | 95% CI low | 95% CI high | n respondents |
|---|---|---|---|---|---|---|
| 1994 | Midterm years | Citizens 18 and older, 1994 to 2024 | 48.4 | 48.0 | 48.8 | 95,272 |
| 1998 | Midterm years | Citizens 18 and older, 1994 to 2024 | 45.3 | 44.9 | 45.7 | 84,076 |
| 2002 | Midterm years | Citizens 18 and older, 1994 to 2024 | 46.1 | 45.8 | 46.5 | 98,638 |
| 2006 | Midterm years | Citizens 18 and older, 1994 to 2024 | 47.8 | 47.4 | 48.2 | 94,095 |
| 2010 | Midterm years | Citizens 18 and older, 1994 to 2024 | 45.5 | 45.2 | 45.9 | 94,208 |
| 2014 | Midterm years | Citizens 18 and older, 1994 to 2024 | 41.9 | 41.6 | 42.3 | 96,267 |
| 2018 | Midterm years | Citizens 18 and older, 1994 to 2024 | 53.4 | 53.1 | 53.8 | 88,749 |
| 2022 | Midterm years | Citizens 18 and older, 1994 to 2024 | 52.2 | 51.8 | 52.6 | 73,083 |
| 1996 | Presidential years | Citizens 18 and older, 1994 to 2024 | 58.4 | 58.0 | 58.7 | 84,090 |
| 2000 | Presidential years | Citizens 18 and older, 1994 to 2024 | 59.5 | 59.1 | 59.8 | 82,428 |
| 2004 | Presidential years | Citizens 18 and older, 1994 to 2024 | 63.8 | 63.5 | 64.2 | 96,452 |
| 2008 | Presidential years | Citizens 18 and older, 1994 to 2024 | 63.6 | 63.3 | 64.0 | 92,360 |
| 2012 | Presidential years | Citizens 18 and older, 1994 to 2024 | 61.8 | 61.5 | 62.2 | 94,311 |
| 2016 | Presidential years | Citizens 18 and older, 1994 to 2024 | 61.4 | 61.0 | 61.7 | 93,794 |
| 2020 | Presidential years | Citizens 18 and older, 1994 to 2024 | 66.8 | 66.4 | 67.1 | 81,898 |
| 2024 | Presidential years | Citizens 18 and older, 1994 to 2024 | 65.3 | 64.9 | 65.7 | 72,590 |
| 1978 | Midterm years | All residents 18 and older, 1976 to 2024 | 45.9 | 45.6 | 46.2 | 104,870 |
| 1982 | Midterm years | All residents 18 and older, 1976 to 2024 | 48.5 | 48.2 | 48.9 | 112,372 |
| 1986 | Midterm years | All residents 18 and older, 1976 to 2024 | 45.8 | 45.4 | 46.1 | 108,410 |
| 1990 | Midterm years | All residents 18 and older, 1976 to 2024 | 44.5 | 44.1 | 44.8 | 108,987 |
| 1994 | Midterm years | All residents 18 and older, 1976 to 2024 | 45.0 | 44.7 | 45.4 | 101,630 |
| 1998 | Midterm years | All residents 18 and older, 1976 to 2024 | 41.9 | 41.6 | 42.3 | 89,989 |
| 2002 | Midterm years | All residents 18 and older, 1976 to 2024 | 42.3 | 41.9 | 42.6 | 105,682 |
| 2006 | Midterm years | All residents 18 and older, 1976 to 2024 | 43.6 | 43.2 | 43.9 | 101,640 |
| 2010 | Midterm years | All residents 18 and older, 1976 to 2024 | 41.8 | 41.4 | 42.1 | 101,338 |
| 2014 | Midterm years | All residents 18 and older, 1976 to 2024 | 38.5 | 38.1 | 38.8 | 103,323 |
| 2018 | Midterm years | All residents 18 and older, 1976 to 2024 | 49.0 | 48.6 | 49.3 | 95,154 |
| 2022 | Midterm years | All residents 18 and older, 1976 to 2024 | 47.7 | 47.3 | 48.1 | 78,597 |
| 1976 | Presidential years | All residents 18 and older, 1976 to 2024 | 59.2 | 58.8 | 59.5 | 89,090 |
| 1980 | Presidential years | All residents 18 and older, 1976 to 2024 | 59.0 | 58.7 | 59.4 | 123,591 |
| 1984 | Presidential years | All residents 18 and older, 1976 to 2024 | 59.7 | 59.3 | 60.0 | 110,115 |
| 1988 | Presidential years | All residents 18 and older, 1976 to 2024 | 57.1 | 56.8 | 57.5 | 102,945 |
| 1992 | Presidential years | All residents 18 and older, 1976 to 2024 | 60.7 | 60.4 | 61.1 | 105,987 |
| 1996 | Presidential years | All residents 18 and older, 1976 to 2024 | 54.2 | 53.9 | 54.6 | 89,609 |
| 2000 | Presidential years | All residents 18 and older, 1976 to 2024 | 54.4 | 54.1 | 54.8 | 88,884 |
| 2004 | Presidential years | All residents 18 and older, 1976 to 2024 | 58.3 | 57.9 | 58.6 | 103,696 |
| 2008 | Presidential years | All residents 18 and older, 1976 to 2024 | 58.2 | 57.8 | 58.5 | 99,556 |
| 2012 | Presidential years | All residents 18 and older, 1976 to 2024 | 56.5 | 56.2 | 56.9 | 101,568 |
| 2016 | Presidential years | All residents 18 and older, 1976 to 2024 | 56.0 | 55.7 | 56.4 | 100,946 |
| 2020 | Presidential years | All residents 18 and older, 1976 to 2024 | 61.3 | 60.9 | 61.7 | 87,589 |
| 2024 | Presidential years | All residents 18 and older, 1976 to 2024 | 59.3 | 58.9 | 59.7 | 78,650 |
The citizens series starts in 1994 because the extract records citizenship only from that year, and a rate with all adults in the denominator behaves differently. The second button of Figure 1 shows it. Counting noncitizens as not voting, presidential-year turnout was 59.2%59.2% percentEstimandweighted share who reported voting; persons outside the voting-question universe count as not votingInterval95% CI 58.8% to 59.5%n89,090WeightsVOSUPPWTPopulationall residents 18 and olderPeriod1976SourceCPS Voting and Registration Supplement in 1976, 54.2%54.2% percentEstimandweighted share who reported voting; persons outside the voting-question universe count as not votingInterval95% CI 53.9% to 54.6%n89,609WeightsVOSUPPWTPopulationall residents 18 and olderPeriod1996SourceCPS Voting and Registration Supplement in 1996 and 59.3%59.3% percentEstimandweighted share who reported voting; persons outside the voting-question universe count as not votingInterval95% CI 58.9% to 59.7%n78,650WeightsVOSUPPWTPopulationall residents 18 and olderPeriod2024SourceCPS Voting and Registration Supplement in 2024. The two series differ because noncitizens, who count as not voting in the second, were 6.8% (95% CI 6.7% to 7.0%)6.8% percentEstimandweighted share who are not US citizensInterval95% CI 6.7% to 7.0%n101,630WeightsVOSUPPWTPopulationadults 18 and olderPeriod1994SourceCPS Voting and Registration Supplement of adults in 1994 and 9.3% (95% CI 9.1% to 9.5%)9.3% percentEstimandweighted share who are not US citizensInterval95% CI 9.1% to 9.5%n78,650WeightsVOSUPPWTPopulationadults 18 and olderPeriod2024SourceCPS Voting and Registration Supplement in 2024. McDonald and Popkin showed that a denominator which includes people who cannot vote misreads a trend in turnout [4].
Before 1994 the extract has no citizenship item. The supplement did not ask the voting question of 32.2%32.2% percentEstimandweighted share of adults outside the voting-question universeWeightsVOSUPPWTPopulationHispanic adults 18 and olderPeriod1980SourceCPS Voting and Registration Supplement of Hispanic adults in 1980 and 40.2%40.2% percentEstimandweighted share of adults outside the voting-question universeWeightsVOSUPPWTPopulationHispanic adults 18 and olderPeriod1992SourceCPS Voting and Registration Supplement in 1992, against 1.7%1.7% percentEstimandweighted share of adults outside the voting-question universeWeightsVOSUPPWTPopulationWhite non-Hispanic adults 18 and olderPeriod1980SourceCPS Voting and Registration Supplement and 1.9%1.9% percentEstimandweighted share of adults outside the voting-question universeWeightsVOSUPPWTPopulationWhite non-Hispanic adults 18 and olderPeriod1992SourceCPS Voting and Registration Supplement of White non-Hispanic adults, a pattern that matches the noncitizen shares of 1994 (40.8%40.8% percentEstimandweighted share who are not US citizensWeightsVOSUPPWTPopulationHispanic adults 18 and olderPeriod1994SourceCPS Voting and Registration Supplement and 1.8%1.8% percentEstimandweighted share who are not US citizensWeightsVOSUPPWTPopulationWhite non-Hispanic adults 18 and olderPeriod1994SourceCPS Voting and Registration Supplement). The Census Bureau counts these adults as not voting in its total-population series, and so does the all-residents view, which is the only series that runs from 1976 on one definition. The gradients below cover citizens from 1994. The Census Bureau's own tables give a citizen series for earlier years from an item that the extract lacks, and the last section says what that series shows.
In 2024 turnout rose with age, education and family income, and registration accounts for most of the gaps
Figure 2 shows three rates for the same groups: the share who voted, the share who were registered, and the share of the registered who voted. A citizen counts as registered if the person voted or said they were registered. The differences in turnout in 2024 were 23.9 (95% CI 22.7 to 25.2)23.9 percentage pointsEstimanddifference in weighted turnout: citizens 65 and older minus citizens 18 to 29Interval95% CI 22.7 to 25.2n33,208WeightsVOSUPPWTPopulationcitizens aged 65 and older and citizens aged 18 to 29Period2024SourceCPS Voting and Registration Supplement points between citizens aged sixty-five and older and those aged eighteen to twenty-nine, 40.5 (95% CI 38.7 to 42.4)40.5 percentage pointsEstimanddifference in weighted turnout: citizens 25 and older with a bachelor's degree or more minus those without a high school diplomaInterval95% CI 38.7 to 42.4n29,750WeightsVOSUPPWTPopulationcitizens 25 and older with a bachelor's degree or more and without a high school diplomaPeriod2024SourceCPS Voting and Registration Supplement between a bachelor's degree and no high school diploma, 20.2 (95% CI 19.3 to 21.1)20.2 percentage pointsEstimanddifference in weighted turnout: highest minus lowest third of family incomeInterval95% CI 19.3 to 21.1n55,352WeightsVOSUPPWTPopulationcitizens in the highest and lowest thirds of family incomePeriod2024SourceCPS Voting and Registration Supplement between the highest and lowest third of family income, 10.4 (95% CI 9.0 to 11.8)10.4 percentage pointsEstimanddifference in weighted turnout: White non-Hispanic minus Black non-Hispanic citizensInterval95% CI 9.0 to 11.8n58,873WeightsVOSUPPWTPopulationWhite and Black non-Hispanic citizensPeriod2024SourceCPS Voting and Registration Supplement between White and Black citizens and 19.9 (95% CI 18.6 to 21.2)19.9 percentage pointsEstimanddifference in weighted turnout: White non-Hispanic minus Hispanic citizensInterval95% CI 18.6 to 21.2n60,301WeightsVOSUPPWTPopulationWhite non-Hispanic and Hispanic citizensPeriod2024SourceCPS Voting and Registration Supplement between White and Hispanic citizens. The education difference is the largest of the five.
Figure 2
In 2024 turnout rose with age, education and family income, and Hispanic citizens had the lowest turnout of the racial and ethnic groups
Percent of citizens who voted, who were registered, and who voted among the registered, by age, race and ethnicity, education and family income, 2024 (buttons show other elections); 95% intervals from the Kish effective sample size
Voted (share of citizens)
Registered (share of citizens)
Voted (share of the registered)
Percent of the group
Show data
| Year | Measure | Dimension | Group | Percent | 95% CI low | 95% CI high | n respondents |
|---|---|---|---|---|---|---|---|
| 2024 | Voted (share of citizens) | All citizens | All citizens | 65.3 | 64.9 | 65.7 | 72,590 |
| 2024 | Voted (share of citizens) | Age | Ages 18 to 29 | 50.7 | 49.7 | 51.8 | 12,303 |
| 2024 | Voted (share of citizens) | Age | Ages 30 to 44 | 61.9 | 61.0 | 62.7 | 16,754 |
| 2024 | Voted (share of citizens) | Age | Ages 45 to 64 | 70.0 | 69.3 | 70.6 | 22,628 |
| 2024 | Voted (share of citizens) | Age | Ages 65 and older | 74.7 | 74.0 | 75.4 | 20,905 |
| 2024 | Voted (share of citizens) | Race and ethnicity | White | 70.5 | 70.0 | 70.9 | 52,237 |
| 2024 | Voted (share of citizens) | Race and ethnicity | Black | 60.1 | 58.7 | 61.4 | 6,636 |
| 2024 | Voted (share of citizens) | Race and ethnicity | Hispanic | 50.6 | 49.4 | 51.8 | 8,064 |
| 2024 | Voted (share of citizens) | Race and ethnicity | Asian | 57.6 | 55.9 | 59.4 | 3,853 |
| 2024 | Voted (share of citizens) | Race and ethnicity | Other or multiple races | 57.3 | 54.5 | 60.2 | 1,800 |
| 2024 | Voted (share of citizens) | Education (25 and older) | Less than high school | 39.1 | 37.4 | 40.8 | 4,008 |
| 2024 | Voted (share of citizens) | Education (25 and older) | High school graduate | 55.3 | 54.4 | 56.1 | 18,383 |
| 2024 | Voted (share of citizens) | Education (25 and older) | Some college | 69.2 | 68.4 | 70.0 | 17,338 |
| 2024 | Voted (share of citizens) | Education (25 and older) | Bachelor's degree or more | 79.6 | 79.0 | 80.2 | 25,742 |
| 2024 | Voted (share of citizens) | Family income | Lowest third | 54.3 | 53.6 | 55.0 | 26,368 |
| 2024 | Voted (share of citizens) | Family income | Middle third | 67.2 | 66.6 | 67.8 | 35,210 |
| 2024 | Voted (share of citizens) | Family income | Highest third | 74.5 | 73.9 | 75.1 | 28,984 |
| 2024 | Registered (share of citizens) | All citizens | All citizens | 73.6 | 73.3 | 74.0 | 72,590 |
| 2024 | Registered (share of citizens) | Age | Ages 18 to 29 | 61.7 | 60.7 | 62.7 | 12,303 |
| 2024 | Registered (share of citizens) | Age | Ages 30 to 44 | 71.9 | 71.2 | 72.7 | 16,754 |
| 2024 | Registered (share of citizens) | Age | Ages 45 to 64 | 77.2 | 76.6 | 77.8 | 22,628 |
| 2024 | Registered (share of citizens) | Age | Ages 65 and older | 80.3 | 79.7 | 80.9 | 20,905 |
| 2024 | Registered (share of citizens) | Race and ethnicity | White | 77.7 | 77.2 | 78.1 | 52,237 |
| 2024 | Registered (share of citizens) | Race and ethnicity | Black | 70.3 | 69.1 | 71.6 | 6,636 |
| 2024 | Registered (share of citizens) | Race and ethnicity | Hispanic | 61.5 | 60.3 | 62.6 | 8,064 |
| 2024 | Registered (share of citizens) | Race and ethnicity | Asian | 66.2 | 64.5 | 67.8 | 3,853 |
| 2024 | Registered (share of citizens) | Race and ethnicity | Other or multiple races | 68.9 | 66.2 | 71.5 | 1,800 |
| 2024 | Registered (share of citizens) | Education (25 and older) | Less than high school | 51.4 | 49.6 | 53.1 | 4,008 |
| 2024 | Registered (share of citizens) | Education (25 and older) | High school graduate | 65.8 | 65.0 | 66.6 | 18,383 |
| 2024 | Registered (share of citizens) | Education (25 and older) | Some college | 78.2 | 77.5 | 78.9 | 17,338 |
| 2024 | Registered (share of citizens) | Education (25 and older) | Bachelor's degree or more | 84.5 | 84.0 | 85.0 | 25,742 |
| 2024 | Registered (share of citizens) | Family income | Lowest third | 65.1 | 64.4 | 65.8 | 26,368 |
| 2024 | Registered (share of citizens) | Family income | Middle third | 74.8 | 74.2 | 75.4 | 35,210 |
| 2024 | Registered (share of citizens) | Family income | Highest third | 81.0 | 80.4 | 81.5 | 28,984 |
| 2024 | Voted (share of the registered) | All citizens | All citizens | 88.8 | 88.5 | 89.1 | 54,231 |
| 2024 | Voted (share of the registered) | Age | Ages 18 to 29 | 82.2 | 81.2 | 83.2 | 7,627 |
| 2024 | Voted (share of the registered) | Age | Ages 30 to 44 | 86.0 | 85.3 | 86.7 | 12,233 |
| 2024 | Voted (share of the registered) | Age | Ages 45 to 64 | 90.6 | 90.1 | 91.1 | 17,491 |
| 2024 | Voted (share of the registered) | Age | Ages 65 and older | 93.0 | 92.6 | 93.5 | 16,880 |
| 2024 | Voted (share of the registered) | Race and ethnicity | White | 90.7 | 90.4 | 91.1 | 40,754 |
| 2024 | Voted (share of the registered) | Race and ethnicity | Black | 85.4 | 84.3 | 86.5 | 4,698 |
| 2024 | Voted (share of the registered) | Race and ethnicity | Hispanic | 82.3 | 81.1 | 83.4 | 5,000 |
| 2024 | Voted (share of the registered) | Race and ethnicity | Asian | 87.1 | 85.6 | 88.6 | 2,546 |
| 2024 | Voted (share of the registered) | Race and ethnicity | Other or multiple races | 83.2 | 80.7 | 85.8 | 1,233 |
| 2024 | Voted (share of the registered) | Education (25 and older) | Less than high school | 76.1 | 74.0 | 78.2 | 2,115 |
| 2024 | Voted (share of the registered) | Education (25 and older) | High school graduate | 84.0 | 83.3 | 84.7 | 12,334 |
| 2024 | Voted (share of the registered) | Education (25 and older) | Some college | 88.5 | 87.9 | 89.1 | 13,658 |
| 2024 | Voted (share of the registered) | Education (25 and older) | Bachelor's degree or more | 94.2 | 93.8 | 94.6 | 21,949 |
| 2024 | Voted (share of the registered) | Family income | Lowest third | 83.4 | 82.8 | 84.1 | 17,698 |
| 2024 | Voted (share of the registered) | Family income | Middle third | 89.8 | 89.4 | 90.3 | 26,950 |
| 2024 | Voted (share of the registered) | Family income | Highest third | 92.0 | 91.6 | 92.5 | 23,573 |
| 2022 | Voted (share of citizens) | All citizens | All citizens | 52.2 | 51.8 | 52.6 | 73,083 |
| 2022 | Voted (share of citizens) | Age | Ages 18 to 29 | 31.2 | 30.3 | 32.1 | 12,305 |
| 2022 | Voted (share of citizens) | Age | Ages 30 to 44 | 46.9 | 46.1 | 47.8 | 16,909 |
| 2022 | Voted (share of citizens) | Age | Ages 45 to 64 | 58.3 | 57.5 | 59.0 | 23,378 |
| 2022 | Voted (share of citizens) | Age | Ages 65 and older | 66.8 | 66.1 | 67.6 | 20,491 |
| 2022 | Voted (share of citizens) | Race and ethnicity | White | 57.6 | 57.2 | 58.1 | 52,569 |
| 2022 | Voted (share of citizens) | Race and ethnicity | Black | 45.6 | 44.3 | 47.0 | 7,247 |
| 2022 | Voted (share of citizens) | Race and ethnicity | Hispanic | 37.9 | 36.7 | 39.0 | 7,741 |
| 2022 | Voted (share of citizens) | Race and ethnicity | Asian | 40.2 | 38.5 | 42.0 | 3,755 |
| 2022 | Voted (share of citizens) | Race and ethnicity | Other or multiple races | 42.9 | 40.0 | 45.7 | 1,771 |
| 2022 | Voted (share of citizens) | Education (25 and older) | Less than high school | 29.2 | 27.7 | 30.8 | 4,251 |
| 2022 | Voted (share of citizens) | Education (25 and older) | High school graduate | 43.9 | 43.1 | 44.7 | 18,919 |
| 2022 | Voted (share of citizens) | Education (25 and older) | Some college | 55.8 | 55.0 | 56.6 | 17,609 |
| 2022 | Voted (share of citizens) | Education (25 and older) | Bachelor's degree or more | 67.7 | 67.0 | 68.3 | 25,233 |
| 2022 | Voted (share of citizens) | Family income | Lowest third | 41.8 | 41.1 | 42.5 | 29,857 |
| 2022 | Voted (share of citizens) | Family income | Middle third | 53.3 | 52.7 | 54.0 | 35,033 |
| 2022 | Voted (share of citizens) | Family income | Highest third | 61.5 | 60.8 | 62.2 | 25,708 |
| 2022 | Registered (share of citizens) | All citizens | All citizens | 69.1 | 68.7 | 69.5 | 73,083 |
| 2022 | Registered (share of citizens) | Age | Ages 18 to 29 | 53.5 | 52.5 | 54.5 | 12,305 |
| 2022 | Registered (share of citizens) | Age | Ages 30 to 44 | 68.0 | 67.2 | 68.8 | 16,909 |
| 2022 | Registered (share of citizens) | Age | Ages 45 to 64 | 73.4 | 72.8 | 74.0 | 23,378 |
| 2022 | Registered (share of citizens) | Age | Ages 65 and older | 77.4 | 76.7 | 78.0 | 20,491 |
| 2022 | Registered (share of citizens) | Race and ethnicity | White | 73.2 | 72.8 | 73.6 | 52,569 |
| 2022 | Registered (share of citizens) | Race and ethnicity | Black | 64.6 | 63.4 | 65.9 | 7,247 |
| 2022 | Registered (share of citizens) | Race and ethnicity | Hispanic | 57.8 | 56.6 | 59.0 | 7,741 |
| 2022 | Registered (share of citizens) | Race and ethnicity | Asian | 59.8 | 58.0 | 61.5 | 3,755 |
| 2022 | Registered (share of citizens) | Race and ethnicity | Other or multiple races | 63.7 | 60.9 | 66.4 | 1,771 |
| 2022 | Registered (share of citizens) | Education (25 and older) | Less than high school | 48.7 | 47.0 | 50.4 | 4,251 |
| 2022 | Registered (share of citizens) | Education (25 and older) | High school graduate | 62.2 | 61.4 | 63.0 | 18,919 |
| 2022 | Registered (share of citizens) | Education (25 and older) | Some college | 74.7 | 73.9 | 75.4 | 17,609 |
| 2022 | Registered (share of citizens) | Education (25 and older) | Bachelor's degree or more | 80.3 | 79.8 | 80.9 | 25,233 |
| 2022 | Registered (share of citizens) | Family income | Lowest third | 60.2 | 59.5 | 60.8 | 29,857 |
| 2022 | Registered (share of citizens) | Family income | Middle third | 70.9 | 70.2 | 71.5 | 35,033 |
| 2022 | Registered (share of citizens) | Family income | Highest third | 76.3 | 75.7 | 76.9 | 25,708 |
| 2022 | Voted (share of the registered) | All citizens | All citizens | 75.5 | 75.1 | 76.0 | 51,451 |
| 2022 | Voted (share of the registered) | Age | Ages 18 to 29 | 58.3 | 56.9 | 59.6 | 6,584 |
| 2022 | Voted (share of the registered) | Age | Ages 30 to 44 | 69.0 | 68.0 | 70.0 | 11,601 |
| 2022 | Voted (share of the registered) | Age | Ages 45 to 64 | 79.4 | 78.7 | 80.1 | 17,266 |
| 2022 | Voted (share of the registered) | Age | Ages 65 and older | 86.4 | 85.8 | 87.0 | 16,000 |
| 2022 | Voted (share of the registered) | Race and ethnicity | White | 78.7 | 78.3 | 79.2 | 38,831 |
| 2022 | Voted (share of the registered) | Race and ethnicity | Black | 70.6 | 69.1 | 72.1 | 4,840 |
| 2022 | Voted (share of the registered) | Race and ethnicity | Hispanic | 65.5 | 64.0 | 67.0 | 4,424 |
| 2022 | Voted (share of the registered) | Race and ethnicity | Asian | 67.3 | 65.1 | 69.5 | 2,232 |
| 2022 | Voted (share of the registered) | Race and ethnicity | Other or multiple races | 67.3 | 63.9 | 70.7 | 1,124 |
| 2022 | Voted (share of the registered) | Education (25 and older) | Less than high school | 60.0 | 57.6 | 62.4 | 2,106 |
| 2022 | Voted (share of the registered) | Education (25 and older) | High school graduate | 70.6 | 69.7 | 71.6 | 12,050 |
| 2022 | Voted (share of the registered) | Education (25 and older) | Some college | 74.7 | 73.9 | 75.6 | 13,339 |
| 2022 | Voted (share of the registered) | Education (25 and older) | Bachelor's degree or more | 84.2 | 83.7 | 84.8 | 20,514 |
| 2022 | Voted (share of the registered) | Family income | Lowest third | 69.5 | 68.7 | 70.3 | 18,956 |
| 2022 | Voted (share of the registered) | Family income | Middle third | 75.3 | 74.6 | 75.9 | 25,614 |
| 2022 | Voted (share of the registered) | Family income | Highest third | 80.5 | 79.9 | 81.2 | 19,807 |
| 2020 | Voted (share of citizens) | All citizens | All citizens | 66.8 | 66.4 | 67.1 | 81,898 |
| 2020 | Voted (share of citizens) | Age | Ages 18 to 29 | 54.1 | 53.2 | 55.0 | 14,347 |
| 2020 | Voted (share of citizens) | Age | Ages 30 to 44 | 64.4 | 63.6 | 65.1 | 19,144 |
| 2020 | Voted (share of citizens) | Age | Ages 45 to 64 | 71.0 | 70.4 | 71.6 | 27,161 |
| 2020 | Voted (share of citizens) | Age | Ages 65 and older | 74.5 | 73.8 | 75.1 | 21,246 |
| 2020 | Voted (share of citizens) | Race and ethnicity | White | 70.9 | 70.5 | 71.4 | 59,555 |
| 2020 | Voted (share of citizens) | Race and ethnicity | Black | 62.8 | 61.6 | 64.0 | 8,419 |
| 2020 | Voted (share of citizens) | Race and ethnicity | Hispanic | 53.7 | 52.6 | 54.9 | 8,163 |
| 2020 | Voted (share of citizens) | Race and ethnicity | Asian | 59.3 | 57.6 | 61.1 | 3,860 |
| 2020 | Voted (share of citizens) | Race and ethnicity | Other or multiple races | 58.7 | 55.9 | 61.5 | 1,901 |
| 2020 | Voted (share of citizens) | Education (25 and older) | Less than high school | 43.7 | 42.1 | 45.2 | 5,000 |
| 2020 | Voted (share of citizens) | Education (25 and older) | High school graduate | 58.0 | 57.3 | 58.8 | 21,412 |
| 2020 | Voted (share of citizens) | Education (25 and older) | Some college | 71.1 | 70.4 | 71.8 | 20,091 |
| 2020 | Voted (share of citizens) | Education (25 and older) | Bachelor's degree or more | 80.0 | 79.4 | 80.5 | 27,130 |
| 2020 | Voted (share of citizens) | Family income | Lowest third | 56.1 | 55.5 | 56.8 | 30,319 |
| 2020 | Voted (share of citizens) | Family income | Middle third | 67.7 | 67.1 | 68.3 | 33,174 |
| 2020 | Voted (share of citizens) | Family income | Highest third | 76.5 | 75.9 | 77.1 | 36,140 |
| 2020 | Registered (share of citizens) | All citizens | All citizens | 72.7 | 72.3 | 73.0 | 81,898 |
| 2020 | Registered (share of citizens) | Age | Ages 18 to 29 | 62.5 | 61.6 | 63.4 | 14,347 |
| 2020 | Registered (share of citizens) | Age | Ages 30 to 44 | 71.5 | 70.8 | 72.2 | 19,144 |
| 2020 | Registered (share of citizens) | Age | Ages 45 to 64 | 75.7 | 75.1 | 76.3 | 27,161 |
| 2020 | Registered (share of citizens) | Age | Ages 65 and older | 78.6 | 77.9 | 79.2 | 21,246 |
| 2020 | Registered (share of citizens) | Race and ethnicity | White | 76.5 | 76.1 | 76.9 | 59,555 |
| 2020 | Registered (share of citizens) | Race and ethnicity | Black | 69.1 | 68.0 | 70.2 | 8,419 |
| 2020 | Registered (share of citizens) | Race and ethnicity | Hispanic | 61.1 | 60.0 | 62.3 | 8,163 |
| 2020 | Registered (share of citizens) | Race and ethnicity | Asian | 63.5 | 61.8 | 65.2 | 3,860 |
| 2020 | Registered (share of citizens) | Race and ethnicity | Other or multiple races | 68.7 | 66.1 | 71.3 | 1,901 |
| 2020 | Registered (share of citizens) | Education (25 and older) | Less than high school | 53.1 | 51.6 | 54.7 | 5,000 |
| 2020 | Registered (share of citizens) | Education (25 and older) | High school graduate | 65.5 | 64.8 | 66.3 | 21,412 |
| 2020 | Registered (share of citizens) | Education (25 and older) | Some college | 77.0 | 76.4 | 77.7 | 20,091 |
| 2020 | Registered (share of citizens) | Education (25 and older) | Bachelor's degree or more | 83.1 | 82.6 | 83.6 | 27,130 |
| 2020 | Registered (share of citizens) | Family income | Lowest third | 64.3 | 63.7 | 65.0 | 30,319 |
| 2020 | Registered (share of citizens) | Family income | Middle third | 73.4 | 72.8 | 73.9 | 33,174 |
| 2020 | Registered (share of citizens) | Family income | Highest third | 80.3 | 79.8 | 80.9 | 36,140 |
| 2020 | Voted (share of the registered) | All citizens | All citizens | 91.9 | 91.6 | 92.1 | 59,993 |
| 2020 | Voted (share of the registered) | Age | Ages 18 to 29 | 86.6 | 85.8 | 87.4 | 8,900 |
| 2020 | Voted (share of the registered) | Age | Ages 30 to 44 | 90.0 | 89.4 | 90.6 | 13,745 |
| 2020 | Voted (share of the registered) | Age | Ages 45 to 64 | 93.7 | 93.4 | 94.1 | 20,600 |
| 2020 | Voted (share of the registered) | Age | Ages 65 and older | 94.8 | 94.4 | 95.2 | 16,748 |
| 2020 | Voted (share of the registered) | Race and ethnicity | White | 92.8 | 92.5 | 93.0 | 45,350 |
| 2020 | Voted (share of the registered) | Race and ethnicity | Black | 90.9 | 90.0 | 91.7 | 5,894 |
| 2020 | Voted (share of the registered) | Race and ethnicity | Hispanic | 87.9 | 87.0 | 88.9 | 4,970 |
| 2020 | Voted (share of the registered) | Race and ethnicity | Asian | 93.5 | 92.4 | 94.6 | 2,476 |
| 2020 | Voted (share of the registered) | Race and ethnicity | Other or multiple races | 85.5 | 83.1 | 87.8 | 1,303 |
| 2020 | Voted (share of the registered) | Education (25 and older) | Less than high school | 82.2 | 80.6 | 83.9 | 2,686 |
| 2020 | Voted (share of the registered) | Education (25 and older) | High school graduate | 88.5 | 87.9 | 89.1 | 14,188 |
| 2020 | Voted (share of the registered) | Education (25 and older) | Some college | 92.3 | 91.8 | 92.7 | 15,568 |
| 2020 | Voted (share of the registered) | Education (25 and older) | Bachelor's degree or more | 96.2 | 95.9 | 96.5 | 22,666 |
| 2020 | Voted (share of the registered) | Family income | Lowest third | 87.2 | 86.7 | 87.8 | 19,967 |
| 2020 | Voted (share of the registered) | Family income | Middle third | 92.3 | 91.9 | 92.7 | 24,506 |
| 2020 | Voted (share of the registered) | Family income | Highest third | 95.2 | 94.9 | 95.5 | 28,675 |
| 2018 | Voted (share of citizens) | All citizens | All citizens | 53.4 | 53.1 | 53.8 | 88,749 |
| 2018 | Voted (share of citizens) | Age | Ages 18 to 29 | 35.6 | 34.7 | 36.4 | 15,758 |
| 2018 | Voted (share of citizens) | Age | Ages 30 to 44 | 48.8 | 48.0 | 49.6 | 20,821 |
| 2018 | Voted (share of citizens) | Age | Ages 45 to 64 | 59.5 | 58.8 | 60.1 | 30,528 |
| 2018 | Voted (share of citizens) | Age | Ages 65 and older | 66.1 | 65.4 | 66.8 | 21,642 |
| 2018 | Voted (share of citizens) | Race and ethnicity | White | 57.5 | 57.0 | 57.9 | 64,983 |
| 2018 | Voted (share of citizens) | Race and ethnicity | Black | 51.4 | 50.2 | 52.6 | 8,773 |
| 2018 | Voted (share of citizens) | Race and ethnicity | Hispanic | 40.4 | 39.3 | 41.5 | 8,745 |
| 2018 | Voted (share of citizens) | Race and ethnicity | Asian | 40.4 | 38.7 | 42.1 | 4,100 |
| 2018 | Voted (share of citizens) | Race and ethnicity | Other or multiple races | 45.5 | 42.9 | 48.2 | 2,148 |
| 2018 | Voted (share of citizens) | Education (25 and older) | Less than high school | 30.0 | 28.7 | 31.3 | 6,074 |
| 2018 | Voted (share of citizens) | Education (25 and older) | High school graduate | 44.9 | 44.2 | 45.6 | 23,004 |
| 2018 | Voted (share of citizens) | Education (25 and older) | Some college | 57.7 | 56.9 | 58.4 | 22,009 |
| 2018 | Voted (share of citizens) | Education (25 and older) | Bachelor's degree or more | 69.5 | 68.9 | 70.1 | 28,734 |
| 2018 | Voted (share of citizens) | Family income | Lowest third | 43.3 | 42.6 | 43.9 | 34,941 |
| 2018 | Voted (share of citizens) | Family income | Middle third | 54.1 | 53.5 | 54.7 | 35,374 |
| 2018 | Voted (share of citizens) | Family income | Highest third | 62.9 | 62.3 | 63.5 | 36,649 |
| 2018 | Registered (share of citizens) | All citizens | All citizens | 66.9 | 66.5 | 67.2 | 88,749 |
| 2018 | Registered (share of citizens) | Age | Ages 18 to 29 | 52.9 | 52.0 | 53.8 | 15,758 |
| 2018 | Registered (share of citizens) | Age | Ages 30 to 44 | 64.3 | 63.6 | 65.1 | 20,821 |
| 2018 | Registered (share of citizens) | Age | Ages 45 to 64 | 71.4 | 70.8 | 72.0 | 30,528 |
| 2018 | Registered (share of citizens) | Age | Ages 65 and older | 76.0 | 75.3 | 76.6 | 21,642 |
| 2018 | Registered (share of citizens) | Race and ethnicity | White | 71.0 | 70.6 | 71.4 | 64,983 |
| 2018 | Registered (share of citizens) | Race and ethnicity | Black | 64.5 | 63.3 | 65.6 | 8,773 |
| 2018 | Registered (share of citizens) | Race and ethnicity | Hispanic | 53.7 | 52.6 | 54.9 | 8,745 |
| 2018 | Registered (share of citizens) | Race and ethnicity | Asian | 52.6 | 50.8 | 54.3 | 4,100 |
| 2018 | Registered (share of citizens) | Race and ethnicity | Other or multiple races | 62.0 | 59.4 | 64.5 | 2,148 |
| 2018 | Registered (share of citizens) | Education (25 and older) | Less than high school | 47.2 | 45.8 | 48.6 | 6,074 |
| 2018 | Registered (share of citizens) | Education (25 and older) | High school graduate | 60.3 | 59.5 | 61.0 | 23,004 |
| 2018 | Registered (share of citizens) | Education (25 and older) | Some college | 72.2 | 71.5 | 72.9 | 22,009 |
| 2018 | Registered (share of citizens) | Education (25 and older) | Bachelor's degree or more | 78.7 | 78.2 | 79.2 | 28,734 |
| 2018 | Registered (share of citizens) | Family income | Lowest third | 59.1 | 58.4 | 59.7 | 34,941 |
| 2018 | Registered (share of citizens) | Family income | Middle third | 67.2 | 66.6 | 67.8 | 35,374 |
| 2018 | Registered (share of citizens) | Family income | Highest third | 74.4 | 73.9 | 75.0 | 36,649 |
| 2018 | Voted (share of the registered) | All citizens | All citizens | 79.9 | 79.5 | 80.3 | 60,349 |
| 2018 | Voted (share of the registered) | Age | Ages 18 to 29 | 67.2 | 66.0 | 68.3 | 8,298 |
| 2018 | Voted (share of the registered) | Age | Ages 30 to 44 | 75.9 | 75.0 | 76.7 | 13,571 |
| 2018 | Voted (share of the registered) | Age | Ages 45 to 64 | 83.3 | 82.7 | 83.8 | 21,897 |
| 2018 | Voted (share of the registered) | Age | Ages 65 and older | 87.1 | 86.5 | 87.6 | 16,583 |
| 2018 | Voted (share of the registered) | Race and ethnicity | White | 80.9 | 80.5 | 81.3 | 46,399 |
| 2018 | Voted (share of the registered) | Race and ethnicity | Black | 79.7 | 78.5 | 80.9 | 5,775 |
| 2018 | Voted (share of the registered) | Race and ethnicity | Hispanic | 75.2 | 73.8 | 76.5 | 4,687 |
| 2018 | Voted (share of the registered) | Race and ethnicity | Asian | 76.8 | 74.8 | 78.8 | 2,151 |
| 2018 | Voted (share of the registered) | Race and ethnicity | Other or multiple races | 73.5 | 70.6 | 76.4 | 1,337 |
| 2018 | Voted (share of the registered) | Education (25 and older) | Less than high school | 63.6 | 61.6 | 65.6 | 2,930 |
| 2018 | Voted (share of the registered) | Education (25 and older) | High school graduate | 74.5 | 73.7 | 75.3 | 14,108 |
| 2018 | Voted (share of the registered) | Education (25 and older) | Some college | 79.9 | 79.2 | 80.6 | 16,064 |
| 2018 | Voted (share of the registered) | Education (25 and older) | Bachelor's degree or more | 88.3 | 87.8 | 88.8 | 22,907 |
| 2018 | Voted (share of the registered) | Family income | Lowest third | 73.3 | 72.5 | 74.0 | 21,328 |
| 2018 | Voted (share of the registered) | Family income | Middle third | 80.5 | 80.0 | 81.1 | 24,401 |
| 2018 | Voted (share of the registered) | Family income | Highest third | 84.6 | 84.0 | 85.1 | 27,344 |
| 2016 | Voted (share of citizens) | All citizens | All citizens | 61.4 | 61.0 | 61.7 | 93,794 |
| 2016 | Voted (share of citizens) | Age | Ages 18 to 29 | 46.1 | 45.3 | 47.0 | 17,431 |
| 2016 | Voted (share of citizens) | Age | Ages 30 to 44 | 58.8 | 58.0 | 59.5 | 21,923 |
| 2016 | Voted (share of citizens) | Age | Ages 45 to 64 | 66.6 | 66.1 | 67.2 | 33,170 |
| 2016 | Voted (share of citizens) | Age | Ages 65 and older | 70.9 | 70.2 | 71.6 | 21,270 |
| 2016 | Voted (share of citizens) | Race and ethnicity | White | 65.3 | 64.9 | 65.7 | 69,164 |
| 2016 | Voted (share of citizens) | Race and ethnicity | Black | 59.6 | 58.5 | 60.7 | 9,556 |
| 2016 | Voted (share of citizens) | Race and ethnicity | Hispanic | 47.6 | 46.4 | 48.7 | 8,703 |
| 2016 | Voted (share of citizens) | Race and ethnicity | Asian | 48.9 | 47.2 | 50.7 | 4,079 |
| 2016 | Voted (share of citizens) | Race and ethnicity | Other or multiple races | 50.2 | 47.6 | 52.7 | 2,292 |
| 2016 | Voted (share of citizens) | Education (25 and older) | Less than high school | 37.3 | 36.0 | 38.6 | 6,908 |
| 2016 | Voted (share of citizens) | Education (25 and older) | High school graduate | 53.9 | 53.2 | 54.6 | 24,702 |
| 2016 | Voted (share of citizens) | Education (25 and older) | Some college | 66.0 | 65.3 | 66.7 | 23,525 |
| 2016 | Voted (share of citizens) | Education (25 and older) | Bachelor's degree or more | 77.1 | 76.6 | 77.7 | 28,684 |
| 2016 | Voted (share of citizens) | Family income | Lowest third | 50.2 | 49.6 | 50.8 | 32,737 |
| 2016 | Voted (share of citizens) | Family income | Middle third | 62.5 | 61.9 | 63.1 | 42,544 |
| 2016 | Voted (share of citizens) | Family income | Highest third | 71.4 | 70.9 | 72.0 | 35,668 |
| 2016 | Registered (share of citizens) | All citizens | All citizens | 70.3 | 70.0 | 70.7 | 93,794 |
| 2016 | Registered (share of citizens) | Age | Ages 18 to 29 | 58.5 | 57.6 | 59.3 | 17,431 |
| 2016 | Registered (share of citizens) | Age | Ages 30 to 44 | 68.6 | 67.9 | 69.3 | 21,923 |
| 2016 | Registered (share of citizens) | Age | Ages 45 to 64 | 74.0 | 73.5 | 74.5 | 33,170 |
| 2016 | Registered (share of citizens) | Age | Ages 65 and older | 78.0 | 77.4 | 78.7 | 21,270 |
| 2016 | Registered (share of citizens) | Race and ethnicity | White | 73.9 | 73.5 | 74.3 | 69,164 |
| 2016 | Registered (share of citizens) | Race and ethnicity | Black | 69.6 | 68.5 | 70.6 | 9,556 |
| 2016 | Registered (share of citizens) | Race and ethnicity | Hispanic | 57.3 | 56.1 | 58.4 | 8,703 |
| 2016 | Registered (share of citizens) | Race and ethnicity | Asian | 56.5 | 54.8 | 58.3 | 4,079 |
| 2016 | Registered (share of citizens) | Race and ethnicity | Other or multiple races | 63.0 | 60.5 | 65.5 | 2,292 |
| 2016 | Registered (share of citizens) | Education (25 and older) | Less than high school | 49.2 | 47.8 | 50.5 | 6,908 |
| 2016 | Registered (share of citizens) | Education (25 and older) | High school graduate | 64.6 | 64.0 | 65.3 | 24,702 |
| 2016 | Registered (share of citizens) | Education (25 and older) | Some college | 75.0 | 74.4 | 75.6 | 23,525 |
| 2016 | Registered (share of citizens) | Education (25 and older) | Bachelor's degree or more | 82.5 | 82.0 | 83.0 | 28,684 |
| 2016 | Registered (share of citizens) | Family income | Lowest third | 61.9 | 61.2 | 62.5 | 32,737 |
| 2016 | Registered (share of citizens) | Family income | Middle third | 71.1 | 70.5 | 71.6 | 42,544 |
| 2016 | Registered (share of citizens) | Family income | Highest third | 78.1 | 77.6 | 78.6 | 35,668 |
| 2016 | Voted (share of the registered) | All citizens | All citizens | 87.3 | 87.0 | 87.6 | 66,881 |
| 2016 | Voted (share of the registered) | Age | Ages 18 to 29 | 78.9 | 78.0 | 79.8 | 10,241 |
| 2016 | Voted (share of the registered) | Age | Ages 30 to 44 | 85.6 | 85.0 | 86.3 | 15,171 |
| 2016 | Voted (share of the registered) | Age | Ages 45 to 64 | 90.0 | 89.6 | 90.4 | 24,684 |
| 2016 | Voted (share of the registered) | Age | Ages 65 and older | 90.9 | 90.4 | 91.3 | 16,785 |
| 2016 | Voted (share of the registered) | Race and ethnicity | White | 88.3 | 88.0 | 88.7 | 51,372 |
| 2016 | Voted (share of the registered) | Race and ethnicity | Black | 85.6 | 84.7 | 86.6 | 6,810 |
| 2016 | Voted (share of the registered) | Race and ethnicity | Hispanic | 83.1 | 81.9 | 84.2 | 5,015 |
| 2016 | Voted (share of the registered) | Race and ethnicity | Asian | 86.6 | 85.0 | 88.2 | 2,262 |
| 2016 | Voted (share of the registered) | Race and ethnicity | Other or multiple races | 79.6 | 77.0 | 82.2 | 1,422 |
| 2016 | Voted (share of the registered) | Education (25 and older) | Less than high school | 75.9 | 74.3 | 77.5 | 3,498 |
| 2016 | Voted (share of the registered) | Education (25 and older) | High school graduate | 83.4 | 82.7 | 84.0 | 16,166 |
| 2016 | Voted (share of the registered) | Education (25 and older) | Some college | 88.0 | 87.5 | 88.6 | 17,825 |
| 2016 | Voted (share of the registered) | Education (25 and older) | Bachelor's degree or more | 93.5 | 93.2 | 93.9 | 23,854 |
| 2016 | Voted (share of the registered) | Family income | Lowest third | 81.2 | 80.6 | 81.8 | 20,718 |
| 2016 | Voted (share of the registered) | Family income | Middle third | 87.9 | 87.5 | 88.4 | 30,683 |
| 2016 | Voted (share of the registered) | Family income | Highest third | 91.5 | 91.1 | 91.9 | 27,965 |
| 2012 | Voted (share of citizens) | All citizens | All citizens | 61.8 | 61.5 | 62.2 | 94,311 |
| 2012 | Voted (share of citizens) | Age | Ages 18 to 29 | 45.0 | 44.2 | 45.9 | 17,998 |
| 2012 | Voted (share of citizens) | Age | Ages 30 to 44 | 59.5 | 58.8 | 60.3 | 22,680 |
| 2012 | Voted (share of citizens) | Age | Ages 45 to 64 | 67.9 | 67.3 | 68.4 | 34,382 |
| 2012 | Voted (share of citizens) | Age | Ages 65 and older | 72.0 | 71.3 | 72.7 | 19,251 |
| 2012 | Voted (share of citizens) | Race and ethnicity | White | 64.1 | 63.7 | 64.5 | 71,460 |
| 2012 | Voted (share of citizens) | Race and ethnicity | Black | 66.6 | 65.6 | 67.7 | 9,232 |
| 2012 | Voted (share of citizens) | Race and ethnicity | Hispanic | 48.0 | 46.7 | 49.2 | 7,812 |
| 2012 | Voted (share of citizens) | Race and ethnicity | Asian | 47.1 | 45.2 | 49.0 | 3,640 |
| 2012 | Voted (share of citizens) | Race and ethnicity | Other or multiple races | 55.3 | 52.8 | 57.9 | 2,167 |
| 2012 | Voted (share of citizens) | Education (25 and older) | Less than high school | 41.2 | 40.0 | 42.5 | 7,512 |
| 2012 | Voted (share of citizens) | Education (25 and older) | High school graduate | 56.0 | 55.3 | 56.7 | 25,733 |
| 2012 | Voted (share of citizens) | Education (25 and older) | Some college | 67.6 | 66.9 | 68.3 | 23,322 |
| 2012 | Voted (share of citizens) | Education (25 and older) | Bachelor's degree or more | 77.8 | 77.3 | 78.4 | 27,231 |
| 2012 | Voted (share of citizens) | Family income | Lowest third | 51.6 | 51.0 | 52.3 | 31,102 |
| 2012 | Voted (share of citizens) | Family income | Middle third | 62.1 | 61.5 | 62.7 | 37,036 |
| 2012 | Voted (share of citizens) | Family income | Highest third | 71.7 | 71.1 | 72.3 | 41,910 |
| 2012 | Registered (share of citizens) | All citizens | All citizens | 71.2 | 70.9 | 71.5 | 94,311 |
| 2012 | Registered (share of citizens) | Age | Ages 18 to 29 | 57.8 | 57.0 | 58.7 | 17,998 |
| 2012 | Registered (share of citizens) | Age | Ages 30 to 44 | 70.2 | 69.5 | 70.9 | 22,680 |
| 2012 | Registered (share of citizens) | Age | Ages 45 to 64 | 75.4 | 74.9 | 75.9 | 34,382 |
| 2012 | Registered (share of citizens) | Age | Ages 65 and older | 79.4 | 78.8 | 80.1 | 19,251 |
| 2012 | Registered (share of citizens) | Race and ethnicity | White | 73.7 | 73.4 | 74.1 | 71,460 |
| 2012 | Registered (share of citizens) | Race and ethnicity | Black | 73.2 | 72.2 | 74.2 | 9,232 |
| 2012 | Registered (share of citizens) | Race and ethnicity | Hispanic | 58.7 | 57.5 | 59.9 | 7,812 |
| 2012 | Registered (share of citizens) | Race and ethnicity | Asian | 56.1 | 54.3 | 58.0 | 3,640 |
| 2012 | Registered (share of citizens) | Race and ethnicity | Other or multiple races | 67.9 | 65.5 | 70.3 | 2,167 |
| 2012 | Registered (share of citizens) | Education (25 and older) | Less than high school | 54.5 | 53.2 | 55.7 | 7,512 |
| 2012 | Registered (share of citizens) | Education (25 and older) | High school graduate | 66.9 | 66.2 | 67.5 | 25,733 |
| 2012 | Registered (share of citizens) | Education (25 and older) | Some college | 76.9 | 76.2 | 77.5 | 23,322 |
| 2012 | Registered (share of citizens) | Education (25 and older) | Bachelor's degree or more | 83.4 | 82.9 | 83.9 | 27,231 |
| 2012 | Registered (share of citizens) | Family income | Lowest third | 63.3 | 62.7 | 63.9 | 31,102 |
| 2012 | Registered (share of citizens) | Family income | Middle third | 71.3 | 70.8 | 71.9 | 37,036 |
| 2012 | Registered (share of citizens) | Family income | Highest third | 78.9 | 78.4 | 79.5 | 41,910 |
| 2012 | Voted (share of the registered) | All citizens | All citizens | 86.8 | 86.5 | 87.1 | 68,167 |
| 2012 | Voted (share of the registered) | Age | Ages 18 to 29 | 77.9 | 77.0 | 78.8 | 10,464 |
| 2012 | Voted (share of the registered) | Age | Ages 30 to 44 | 84.8 | 84.1 | 85.4 | 16,044 |
| 2012 | Voted (share of the registered) | Age | Ages 45 to 64 | 90.0 | 89.6 | 90.4 | 26,196 |
| 2012 | Voted (share of the registered) | Age | Ages 65 and older | 90.6 | 90.1 | 91.2 | 15,463 |
| 2012 | Voted (share of the registered) | Race and ethnicity | White | 87.0 | 86.7 | 87.3 | 53,258 |
| 2012 | Voted (share of the registered) | Race and ethnicity | Black | 91.0 | 90.3 | 91.8 | 6,847 |
| 2012 | Voted (share of the registered) | Race and ethnicity | Hispanic | 81.7 | 80.5 | 82.9 | 4,547 |
| 2012 | Voted (share of the registered) | Race and ethnicity | Asian | 83.9 | 82.1 | 85.7 | 2,054 |
| 2012 | Voted (share of the registered) | Race and ethnicity | Other or multiple races | 81.5 | 79.1 | 84.0 | 1,461 |
| 2012 | Voted (share of the registered) | Education (25 and older) | Less than high school | 75.7 | 74.3 | 77.2 | 4,129 |
| 2012 | Voted (share of the registered) | Education (25 and older) | High school graduate | 83.8 | 83.1 | 84.4 | 17,415 |
| 2012 | Voted (share of the registered) | Education (25 and older) | Some college | 87.9 | 87.4 | 88.5 | 18,059 |
| 2012 | Voted (share of the registered) | Education (25 and older) | Bachelor's degree or more | 93.3 | 92.9 | 93.7 | 22,925 |
| 2012 | Voted (share of the registered) | Family income | Lowest third | 81.5 | 80.9 | 82.1 | 20,020 |
| 2012 | Voted (share of the registered) | Family income | Middle third | 87.0 | 86.6 | 87.5 | 26,583 |
| 2012 | Voted (share of the registered) | Family income | Highest third | 90.8 | 90.4 | 91.2 | 32,994 |
| 2008 | Voted (share of citizens) | All citizens | All citizens | 63.6 | 63.3 | 64.0 | 92,360 |
| 2008 | Voted (share of citizens) | Age | Ages 18 to 29 | 51.1 | 50.2 | 51.9 | 17,867 |
| 2008 | Voted (share of citizens) | Age | Ages 30 to 44 | 61.8 | 61.1 | 62.5 | 23,057 |
| 2008 | Voted (share of citizens) | Age | Ages 45 to 64 | 69.2 | 68.6 | 69.7 | 34,266 |
| 2008 | Voted (share of citizens) | Age | Ages 65 and older | 70.3 | 69.5 | 71.1 | 17,170 |
| 2008 | Voted (share of citizens) | Race and ethnicity | White | 66.1 | 65.7 | 66.5 | 71,497 |
| 2008 | Voted (share of citizens) | Race and ethnicity | Black | 65.2 | 64.1 | 66.3 | 8,607 |
| 2008 | Voted (share of citizens) | Race and ethnicity | Hispanic | 49.9 | 48.6 | 51.1 | 6,925 |
| 2008 | Voted (share of citizens) | Race and ethnicity | Asian | 47.2 | 45.3 | 49.2 | 3,349 |
| 2008 | Voted (share of citizens) | Race and ethnicity | Other or multiple races | 57.5 | 54.7 | 60.2 | 1,982 |
| 2008 | Voted (share of citizens) | Education (25 and older) | Less than high school | 41.6 | 40.4 | 42.8 | 8,467 |
| 2008 | Voted (share of citizens) | Education (25 and older) | High school graduate | 57.0 | 56.3 | 57.7 | 26,066 |
| 2008 | Voted (share of citizens) | Education (25 and older) | Some college | 70.6 | 69.9 | 71.3 | 22,758 |
| 2008 | Voted (share of citizens) | Education (25 and older) | Bachelor's degree or more | 79.3 | 78.7 | 79.9 | 24,625 |
| 2008 | Voted (share of citizens) | Family income | Lowest third | 57.0 | 56.3 | 57.7 | 27,450 |
| 2008 | Voted (share of citizens) | Family income | Middle third | 68.7 | 68.0 | 69.3 | 26,322 |
| 2008 | Voted (share of citizens) | Family income | Highest third | 77.9 | 77.3 | 78.5 | 33,247 |
| 2008 | Registered (share of citizens) | All citizens | All citizens | 71.0 | 70.7 | 71.3 | 92,360 |
| 2008 | Registered (share of citizens) | Age | Ages 18 to 29 | 61.1 | 60.3 | 61.9 | 17,867 |
| 2008 | Registered (share of citizens) | Age | Ages 30 to 44 | 69.4 | 68.7 | 70.1 | 23,057 |
| 2008 | Registered (share of citizens) | Age | Ages 45 to 64 | 74.9 | 74.4 | 75.4 | 34,266 |
| 2008 | Registered (share of citizens) | Age | Ages 65 and older | 77.4 | 76.7 | 78.1 | 17,170 |
| 2008 | Registered (share of citizens) | Race and ethnicity | White | 73.5 | 73.1 | 73.9 | 71,497 |
| 2008 | Registered (share of citizens) | Race and ethnicity | Black | 70.1 | 69.1 | 71.2 | 8,607 |
| 2008 | Registered (share of citizens) | Race and ethnicity | Hispanic | 59.4 | 58.2 | 60.7 | 6,925 |
| 2008 | Registered (share of citizens) | Race and ethnicity | Asian | 55.0 | 53.0 | 56.9 | 3,349 |
| 2008 | Registered (share of citizens) | Race and ethnicity | Other or multiple races | 67.7 | 65.1 | 70.2 | 1,982 |
| 2008 | Registered (share of citizens) | Education (25 and older) | Less than high school | 53.3 | 52.1 | 54.5 | 8,467 |
| 2008 | Registered (share of citizens) | Education (25 and older) | High school graduate | 66.1 | 65.4 | 66.7 | 26,066 |
| 2008 | Registered (share of citizens) | Education (25 and older) | Some college | 77.1 | 76.4 | 77.7 | 22,758 |
| 2008 | Registered (share of citizens) | Education (25 and older) | Bachelor's degree or more | 83.0 | 82.4 | 83.5 | 24,625 |
| 2008 | Registered (share of citizens) | Family income | Lowest third | 67.6 | 66.9 | 68.3 | 27,450 |
| 2008 | Registered (share of citizens) | Family income | Middle third | 76.5 | 75.9 | 77.1 | 26,322 |
| 2008 | Registered (share of citizens) | Family income | Highest third | 83.4 | 82.8 | 83.9 | 33,247 |
| 2008 | Voted (share of the registered) | All citizens | All citizens | 89.6 | 89.4 | 89.9 | 66,700 |
| 2008 | Voted (share of the registered) | Age | Ages 18 to 29 | 83.5 | 82.8 | 84.3 | 10,951 |
| 2008 | Voted (share of the registered) | Age | Ages 30 to 44 | 89.1 | 88.5 | 89.6 | 16,222 |
| 2008 | Voted (share of the registered) | Age | Ages 45 to 64 | 92.4 | 92.0 | 92.7 | 26,021 |
| 2008 | Voted (share of the registered) | Age | Ages 65 and older | 90.8 | 90.3 | 91.4 | 13,506 |
| 2008 | Voted (share of the registered) | Race and ethnicity | White | 90.0 | 89.7 | 90.2 | 53,301 |
| 2008 | Voted (share of the registered) | Race and ethnicity | Black | 93.0 | 92.2 | 93.7 | 6,119 |
| 2008 | Voted (share of the registered) | Race and ethnicity | Hispanic | 84.0 | 82.8 | 85.2 | 4,090 |
| 2008 | Voted (share of the registered) | Race and ethnicity | Asian | 85.9 | 84.1 | 87.7 | 1,883 |
| 2008 | Voted (share of the registered) | Race and ethnicity | Other or multiple races | 84.9 | 82.5 | 87.3 | 1,307 |
| 2008 | Voted (share of the registered) | Education (25 and older) | Less than high school | 78.1 | 76.7 | 79.4 | 4,586 |
| 2008 | Voted (share of the registered) | Education (25 and older) | High school graduate | 86.3 | 85.8 | 86.9 | 17,531 |
| 2008 | Voted (share of the registered) | Education (25 and older) | Some college | 91.6 | 91.2 | 92.1 | 17,689 |
| 2008 | Voted (share of the registered) | Education (25 and older) | Bachelor's degree or more | 95.6 | 95.3 | 95.9 | 20,803 |
| 2008 | Voted (share of the registered) | Family income | Lowest third | 84.4 | 83.7 | 85.0 | 18,906 |
| 2008 | Voted (share of the registered) | Family income | Middle third | 89.8 | 89.3 | 90.3 | 20,261 |
| 2008 | Voted (share of the registered) | Family income | Highest third | 93.5 | 93.1 | 93.9 | 27,640 |
Writing the turnout rate as the registered share times the share of the registered who vote splits each difference into two parts, and the methods note gives the formula. The split depends on which part is changed first, so the text reports the symmetric midpoint, the range across the two orders for education, and a bootstrap interval that omits the design effect. In 2024 the registered share differed more across groups than the share of the registered who vote. For education the registered shares differed by 33.2 (95% CI 31.3 to 35.0)33.2 percentage pointsEstimanddifference in the weighted registered share: citizens 25 and older with a bachelor's degree or more minus those without a high school diplomaInterval95% CI 31.3 to 35.0n29,750WeightsVOSUPPWTPopulationcitizens 25 and older with a bachelor's degree or more and without a high school diplomaPeriod2024SourceCPS Voting and Registration Supplement points and the shares of the registered who voted by 18.1 (95% CI 16.0 to 20.2)18.1 percentage pointsEstimanddifference in the weighted share of the registered who voted: citizens 25 and older with a bachelor's degree or more minus those without a high school diplomaInterval95% CI 16.0 to 20.2n24,064WeightsVOSUPPWTPopulationcitizens 25 and older with a bachelor's degree or more and without a high school diploma who were registeredPeriod2024SourceCPS Voting and Registration Supplement. The registration part of the turnout difference was 68% (95% CI 65% to 71%)68% percent of the turnout differenceEstimandshare of the turnout difference that arises at registration (symmetric split of the product of the registered share and turnout among the registered), citizens 65 and older minus citizens 18 to 29Interval95% CI 65% to 71%n33,208WeightsVOSUPPWTPopulationcitizens aged 65 and older and citizens aged 18 to 29Period2024SourceCPS Voting and Registration SupplementNoteSeeded percentile bootstrap over respondents (500 replicates); it omits the CPS design effect and is too narrow by an unknown factor. The share depends on the order of the split and on how nonresponse is coded. for age, 70% (95% CI 67% to 73%)70% percent of the turnout differenceEstimandshare of the turnout difference that arises at registration (symmetric split of the product of the registered share and turnout among the registered), citizens 25 and older with a bachelor's degree or more minus those without a high school diplomaInterval95% CI 67% to 73%n29,750WeightsVOSUPPWTPopulationcitizens 25 and older with a bachelor's degree or more and without a high school diplomaPeriod2024SourceCPS Voting and Registration SupplementNoteSeeded percentile bootstrap over respondents (500 replicates); it omits the CPS design effect and is too narrow by an unknown factor. The share depends on the order of the split and on how nonresponse is coded. for education, 69% (95% CI 67% to 71%)69% percent of the turnout differenceEstimandshare of the turnout difference that arises at registration (symmetric split of the product of the registered share and turnout among the registered), highest minus lowest third of family incomeInterval95% CI 67% to 71%n55,352WeightsVOSUPPWTPopulationcitizens in the highest and lowest thirds of family incomePeriod2024SourceCPS Voting and Registration SupplementNoteSeeded percentile bootstrap over respondents (500 replicates); it omits the CPS design effect and is too narrow by an unknown factor. The share depends on the order of the split and on how nonresponse is coded. for income, 62% (95% CI 55% to 68%)62% percent of the turnout differenceEstimandshare of the turnout difference that arises at registration (symmetric split of the product of the registered share and turnout among the registered), White non-Hispanic minus Black non-Hispanic citizensInterval95% CI 55% to 68%n58,873WeightsVOSUPPWTPopulationWhite and Black non-Hispanic citizensPeriod2024SourceCPS Voting and Registration SupplementNoteSeeded percentile bootstrap over respondents (500 replicates); it omits the CPS design effect and is too narrow by an unknown factor. The share depends on the order of the split and on how nonresponse is coded. for White and Black citizens and 70% (95% CI 67% to 74%)70% percent of the turnout differenceEstimandshare of the turnout difference that arises at registration (symmetric split of the product of the registered share and turnout among the registered), White non-Hispanic minus Hispanic citizensInterval95% CI 67% to 74%n60,301WeightsVOSUPPWTPopulationWhite non-Hispanic and Hispanic citizensPeriod2024SourceCPS Voting and Registration SupplementNoteSeeded percentile bootstrap over respondents (500 replicates); it omits the CPS design effect and is too narrow by an unknown factor. The share depends on the order of the split and on how nonresponse is coded. for White and Hispanic citizens. In the 2020 presidential election it ran from 72%72% percent of the turnout differenceEstimandshare of the turnout difference that arises at registration (symmetric split of the product of the registered share and turnout among the registered), citizens 65 and older minus citizens 18 to 29Interval95% CI 69% to 74%n35,593WeightsVOSUPPWTPopulationcitizens aged 65 and older and citizens aged 18 to 29Period2020SourceCPS Voting and Registration SupplementNoteSeeded percentile bootstrap over respondents (500 replicates); it omits the CPS design effect and is too narrow by an unknown factor. The share depends on the order of the split and on how nonresponse is coded. for age to 83%83% percent of the turnout differenceEstimandshare of the turnout difference that arises at registration (symmetric split of the product of the registered share and turnout among the registered), White non-Hispanic minus Black non-Hispanic citizensInterval95% CI 76% to 91%n67,974WeightsVOSUPPWTPopulationWhite and Black non-Hispanic citizensPeriod2020SourceCPS Voting and Registration SupplementNoteSeeded percentile bootstrap over respondents (500 replicates); it omits the CPS design effect and is too narrow by an unknown factor. The share depends on the order of the split and on how nonresponse is coded. for White and Black citizens. In the 2022 midterm it ran from 48%48% percent of the turnout differenceEstimandshare of the turnout difference that arises at registration (symmetric split of the product of the registered share and turnout among the registered), citizens 65 and older minus citizens 18 to 29Interval95% CI 46% to 50%n32,796WeightsVOSUPPWTPopulationcitizens aged 65 and older and citizens aged 18 to 29Period2022SourceCPS Voting and Registration SupplementNoteSeeded percentile bootstrap over respondents (500 replicates); it omits the CPS design effect and is too narrow by an unknown factor. The share depends on the order of the split and on how nonresponse is coded. for age to 62%62% percent of the turnout differenceEstimandshare of the turnout difference that arises at registration (symmetric split of the product of the registered share and turnout among the registered), highest minus lowest third of family incomeInterval95% CI 59% to 64%n55,565WeightsVOSUPPWTPopulationcitizens in the highest and lowest thirds of family incomePeriod2022SourceCPS Voting and Registration SupplementNoteSeeded percentile bootstrap over respondents (500 replicates); it omits the CPS design effect and is too narrow by an unknown factor. The share depends on the order of the split and on how nonresponse is coded. for income, nearer an even split, because voting among the registered differed more across groups: the age difference among the registered was 28.2 (95% CI 26.7 to 29.6)28.2 percentage pointsEstimanddifference in the weighted share of the registered who voted: citizens 65 and older minus citizens 18 to 29Interval95% CI 26.7 to 29.6n22,584WeightsVOSUPPWTPopulationcitizens aged 65 and older and citizens aged 18 to 29 who were registeredPeriod2022SourceCPS Voting and Registration Supplement points in 2022 against 10.8 (95% CI 9.7 to 11.9)10.8 percentage pointsEstimanddifference in the weighted share of the registered who voted: citizens 65 and older minus citizens 18 to 29Interval95% CI 9.7 to 11.9n24,507WeightsVOSUPPWTPopulationcitizens aged 65 and older and citizens aged 18 to 29 who were registeredPeriod2024SourceCPS Voting and Registration Supplement in 2024.
A person who does not answer the voting question counts as not voting, and as not registered unless the person reported being registered, and nonresponse differs by group. In 2024 it was 18.0%18.0% percentEstimandweighted share who did not answer the voting question (refused, did not know, no response)Interval95% CI 17.3% to 18.8%n12,303WeightsVOSUPPWTPopulationcitizens aged 18 to 29Period2024 presidential electionSourceCPS Voting and Registration Supplement of citizens aged eighteen to twenty-nine against 12.1%12.1% percentEstimandweighted share who did not answer the voting question (refused, did not know, no response)Interval95% CI 11.6% to 12.6%n20,905WeightsVOSUPPWTPopulationcitizens aged 65 and olderPeriod2024 presidential electionSourceCPS Voting and Registration Supplement of those sixty-five and older, and 17.6%17.6% percentEstimandweighted share who did not answer the voting question (refused, did not know, no response)Interval95% CI 16.5% to 18.6%n6,636WeightsVOSUPPWTPopulationBlack non-Hispanic citizens 18 and olderPeriod2024 presidential electionSourceCPS Voting and Registration Supplement of Black and 17.0%17.0% percentEstimandweighted share who did not answer the voting question (refused, did not know, no response)Interval95% CI 16.1% to 17.9%n8,064WeightsVOSUPPWTPopulationHispanic citizens 18 and olderPeriod2024 presidential electionSourceCPS Voting and Registration Supplement of Hispanic citizens against 12.6%12.6% percentEstimandweighted share who did not answer the voting question (refused, did not know, no response)Interval95% CI 12.2% to 12.9%n52,237WeightsVOSUPPWTPopulationWhite non-Hispanic citizens 18 and olderPeriod2024 presidential electionSourceCPS Voting and Registration Supplement of White citizens. Part of the registration component therefore reflects who answers the survey. Among respondents only, the registration part of the White-Black difference was 44%44% percent of the turnout differenceEstimandshare of the turnout difference that arises at registration among respondents only (nonrespondents dropped from the registered share and from turnout), White non-Hispanic minus Black non-Hispanic citizensn51,535WeightsVOSUPPWTPopulationWhite and Black non-Hispanic citizens who answered the voting questionPeriod2024SourceCPS Voting and Registration SupplementNoteA point estimate without an interval. in 2024, 61%61% percent of the turnout differenceEstimandshare of the turnout difference that arises at registration among respondents only (nonrespondents dropped from the registered share and from turnout), White non-Hispanic minus Black non-Hispanic citizensn57,946WeightsVOSUPPWTPopulationWhite and Black non-Hispanic citizens who answered the voting questionPeriod2020SourceCPS Voting and Registration SupplementNoteA point estimate without an interval. in 2020 and 38%38% percent of the turnout differenceEstimandshare of the turnout difference that arises at registration among respondents only (nonrespondents dropped from the registered share and from turnout), White non-Hispanic minus Black non-Hispanic citizensn50,856WeightsVOSUPPWTPopulationWhite and Black non-Hispanic citizens who answered the voting questionPeriod2022SourceCPS Voting and Registration SupplementNoteA point estimate without an interval. in 2022, against 62%62% percent of the turnout differenceEstimandshare of the turnout difference that arises at registration (symmetric split of the product of the registered share and turnout among the registered), White non-Hispanic minus Black non-Hispanic citizensInterval95% CI 55% to 68%n58,873WeightsVOSUPPWTPopulationWhite and Black non-Hispanic citizensPeriod2024SourceCPS Voting and Registration SupplementNoteSeeded percentile bootstrap over respondents (500 replicates); it omits the CPS design effect and is too narrow by an unknown factor. The share depends on the order of the split and on how nonresponse is coded., 83%83% percent of the turnout differenceEstimandshare of the turnout difference that arises at registration (symmetric split of the product of the registered share and turnout among the registered), White non-Hispanic minus Black non-Hispanic citizensInterval95% CI 76% to 91%n67,974WeightsVOSUPPWTPopulationWhite and Black non-Hispanic citizensPeriod2020SourceCPS Voting and Registration SupplementNoteSeeded percentile bootstrap over respondents (500 replicates); it omits the CPS design effect and is too narrow by an unknown factor. The share depends on the order of the split and on how nonresponse is coded. and 53%53% percent of the turnout differenceEstimandshare of the turnout difference that arises at registration (symmetric split of the product of the registered share and turnout among the registered), White non-Hispanic minus Black non-Hispanic citizensInterval95% CI 47% to 60%n59,816WeightsVOSUPPWTPopulationWhite and Black non-Hispanic citizensPeriod2022SourceCPS Voting and Registration SupplementNoteSeeded percentile bootstrap over respondents (500 replicates); it omits the CPS design effect and is too narrow by an unknown factor. The share depends on the order of the split and on how nonresponse is coded. under the Census convention. It is the only 2024 share that falls below an even split; among respondents the other four stay above it, at 67%67% percent of the turnout differenceEstimandshare of the turnout difference that arises at registration among respondents only (nonrespondents dropped from the registered share and from turnout), citizens 25 and older with a bachelor's degree or more minus those without a high school diploman26,343WeightsVOSUPPWTPopulationcitizens 25 and older with a bachelor's degree or more and without a high school diploma who answered the voting questionPeriod2024SourceCPS Voting and Registration SupplementNoteA point estimate without an interval. for education, 65%65% percent of the turnout differenceEstimandshare of the turnout difference that arises at registration among respondents only (nonrespondents dropped from the registered share and from turnout), citizens 65 and older minus citizens 18 to 29n28,718WeightsVOSUPPWTPopulationcitizens aged 65 and older and citizens aged 18 to 29 who answered the voting questionPeriod2024SourceCPS Voting and Registration SupplementNoteA point estimate without an interval. for age, 65%65% percent of the turnout differenceEstimandshare of the turnout difference that arises at registration among respondents only (nonrespondents dropped from the registered share and from turnout), highest minus lowest third of family incomen48,044WeightsVOSUPPWTPopulationcitizens in the highest and lowest thirds of family income who answered the voting questionPeriod2024SourceCPS Voting and Registration SupplementNoteA point estimate without an interval. for income and 67%67% percent of the turnout differenceEstimandshare of the turnout difference that arises at registration among respondents only (nonrespondents dropped from the registered share and from turnout), White non-Hispanic minus Hispanic citizensn52,800WeightsVOSUPPWTPopulationWhite non-Hispanic and Hispanic citizens who answered the voting questionPeriod2024SourceCPS Voting and Registration SupplementNoteA point estimate without an interval. for ethnicity. Dropping the nonrespondents also changes some turnout differences. The White minus Black difference was 7.7 (95% CI 6.3 to 9.1)7.7 percentage pointsEstimanddifference in the weighted share who reported voting among respondents (self-report convention): White non-Hispanic minus Black non-Hispanic citizensInterval95% CI 6.3 to 9.1n51,535WeightsVOSUPPWTPopulationcitizens 18 and older who answered the voting questionPeriod2024SourceCPS Voting and Registration Supplement points among respondents against 10.410.4 percentage pointsEstimanddifference in weighted turnout: White non-Hispanic minus Black non-Hispanic citizensInterval95% CI 9.0 to 11.8n58,873WeightsVOSUPPWTPopulationWhite and Black non-Hispanic citizensPeriod2024SourceCPS Voting and Registration Supplement under the Census convention, and the education difference was 42.9 (95% CI 40.9 to 44.9)42.9 percentage pointsEstimanddifference in the weighted share who reported voting among respondents (self-report convention): bachelor's degree or more minus no high school diplomaInterval95% CI 40.9 to 44.9n26,343WeightsVOSUPPWTPopulationcitizens 25 and older who answered the voting questionPeriod2024SourceCPS Voting and Registration Supplement against 40.540.5 percentage pointsEstimanddifference in weighted turnout: citizens 25 and older with a bachelor's degree or more minus those without a high school diplomaInterval95% CI 38.7 to 42.4n29,750WeightsVOSUPPWTPopulationcitizens 25 and older with a bachelor's degree or more and without a high school diplomaPeriod2024SourceCPS Voting and Registration Supplement. The age, Hispanic and income differences were 23.023.0 percentage pointsEstimanddifference in the weighted share who reported voting among respondents (self-report convention): citizens 65 and older minus citizens 18 to 29Interval95% CI 21.8 to 24.3n28,718WeightsVOSUPPWTPopulationcitizens 18 and older who answered the voting questionPeriod2024SourceCPS Voting and Registration Supplement, 19.619.6 percentage pointsEstimanddifference in the weighted share who reported voting among respondents (self-report convention): White non-Hispanic minus Hispanic citizensInterval95% CI 18.3 to 21.0n52,800WeightsVOSUPPWTPopulationcitizens 18 and older who answered the voting questionPeriod2024SourceCPS Voting and Registration Supplement and 20.720.7 percentage pointsEstimanddifference in the weighted share who reported voting among respondents (self-report convention): highest minus lowest third of family incomeInterval95% CI 19.8 to 21.6n48,044WeightsVOSUPPWTPopulationcitizens 18 and older who answered the voting questionPeriod2024SourceCPS Voting and Registration Supplement, close to their Census-convention values.
Age: the oldest group voted most and the youngest least in every election, and the gap narrowed from 1996 to 2008
Figure 3 follows the four age groups. In presidential years the gap between citizens aged sixty-five and older and those aged eighteen to twenty-nine fell from 29.5 (95% CI 28.4 to 30.7)29.5 percentage pointsEstimanddifference in weighted turnout: citizens 65 and older minus citizens 18 to 29Interval95% CI 28.4 to 30.7n32,310WeightsVOSUPPWTPopulationcitizens 18 and olderPeriod1996 presidential electionSourceCPS Voting and Registration Supplement points in 1996 to 19.3 (95% CI 18.1 to 20.4)19.3 percentage pointsEstimanddifference in weighted turnout: citizens 65 and older minus citizens 18 to 29Interval95% CI 18.1 to 20.4n35,037WeightsVOSUPPWTPopulationcitizens 18 and olderPeriod2008 presidential electionSourceCPS Voting and Registration Supplement in 2008, a change of −10.3 (95% CI −11.9 to −8.7)−10.3 percentage pointsEstimandchange in the turnout gap between the two years (difference of two differences), gap = citizens 65 and older minus citizens 18 to 29Interval95% CI −11.9 to −8.7n67,347WeightsVOSUPPWTPopulationcitizens 18 and olderPeriod1996 to 2008SourceCPS Voting and Registration Supplement points. It rose to 27.027.0 percentage pointsEstimanddifference in weighted turnout: citizens 65 and older minus citizens 18 to 29Interval95% CI 25.9 to 28.1n37,249WeightsVOSUPPWTPopulationcitizens 18 and olderPeriod2012 presidential electionSourceCPS Voting and Registration Supplement in 2012, was 20.4 (95% CI 19.2 to 21.5)20.4 percentage pointsEstimanddifference in weighted turnout: citizens 65 and older minus citizens 18 to 29Interval95% CI 19.2 to 21.5n35,593WeightsVOSUPPWTPopulationcitizens 18 and olderPeriod2020 presidential electionSourceCPS Voting and Registration Supplement in 2020 and was 23.9 (95% CI 22.7 to 25.2)23.9 percentage pointsEstimanddifference in weighted turnout: citizens 65 and older minus citizens 18 to 29Interval95% CI 22.7 to 25.2n33,208WeightsVOSUPPWTPopulationcitizens 18 and olderPeriod2024 presidential electionSourceCPS Voting and Registration Supplement in 2024. The narrowing between 1996 and 2008 came from the youngest group, whose turnout rose from 39.6%39.6% percentEstimandweighted share who reported voting (Census convention: nonresponse counted as not voting)Interval95% CI 38.8% to 40.4%n17,234WeightsVOSUPPWTPopulationcitizens aged 18 to 29Period1996 presidential electionSourceCPS Voting and Registration Supplement to 51.1%51.1% percentEstimandweighted share who reported voting (Census convention: nonresponse counted as not voting)Interval95% CI 50.2% to 51.9%n17,867WeightsVOSUPPWTPopulationcitizens aged 18 to 29Period2008 presidential electionSourceCPS Voting and Registration Supplement, while the oldest group's moved from 69.1%69.1% percentEstimandweighted share who reported voting (Census convention: nonresponse counted as not voting)Interval95% CI 68.3% to 70.0%n15,076WeightsVOSUPPWTPopulationcitizens aged 65 and olderPeriod1996 presidential electionSourceCPS Voting and Registration Supplement to 70.3%70.3% percentEstimandweighted share who reported voting (Census convention: nonresponse counted as not voting)Interval95% CI 69.5% to 71.1%n17,170WeightsVOSUPPWTPopulationcitizens aged 65 and olderPeriod2008 presidential electionSourceCPS Voting and Registration Supplement.
Figure 3
Turnout was highest in the oldest and lowest in the youngest age group in every election, and the gap between them fell from 29.5 points in 1996 to 19.3 in 2008
Percent of citizens who reported voting, by age group, presidential years 1996 to 2024 (the second button shows midterm years, 1994 to 2022); 95% intervals from the Kish effective sample size
Reported voting (% of citizens in the group)
Show data
| Year | Election type | Group | Turnout (%) | 95% CI low | 95% CI high | n respondents |
|---|---|---|---|---|---|---|
| 1996 | Presidential | Ages 18 to 29 | 39.6 | 38.8 | 40.4 | 17,234 |
| 2000 | Presidential | Ages 18 to 29 | 40.2 | 39.3 | 41.0 | 15,924 |
| 2004 | Presidential | Ages 18 to 29 | 49.0 | 48.1 | 49.8 | 18,315 |
| 2008 | Presidential | Ages 18 to 29 | 51.1 | 50.2 | 51.9 | 17,867 |
| 2012 | Presidential | Ages 18 to 29 | 45.0 | 44.2 | 45.9 | 17,998 |
| 2016 | Presidential | Ages 18 to 29 | 46.1 | 45.3 | 47.0 | 17,431 |
| 2020 | Presidential | Ages 18 to 29 | 54.1 | 53.2 | 55.0 | 14,347 |
| 2024 | Presidential | Ages 18 to 29 | 50.7 | 49.7 | 51.8 | 12,303 |
| 1996 | Presidential | Ages 30 to 44 | 56.9 | 56.2 | 57.5 | 27,397 |
| 2000 | Presidential | Ages 30 to 44 | 58.4 | 57.7 | 59.0 | 25,111 |
| 2004 | Presidential | Ages 30 to 44 | 62.4 | 61.8 | 63.1 | 26,906 |
| 2008 | Presidential | Ages 30 to 44 | 61.8 | 61.1 | 62.5 | 23,057 |
| 2012 | Presidential | Ages 30 to 44 | 59.5 | 58.8 | 60.3 | 22,680 |
| 2016 | Presidential | Ages 30 to 44 | 58.8 | 58.0 | 59.5 | 21,923 |
| 2020 | Presidential | Ages 30 to 44 | 64.4 | 63.6 | 65.1 | 19,144 |
| 2024 | Presidential | Ages 30 to 44 | 61.9 | 61.0 | 62.7 | 16,754 |
| 1996 | Presidential | Ages 45 to 64 | 68.2 | 67.5 | 68.8 | 24,383 |
| 2000 | Presidential | Ages 45 to 64 | 67.8 | 67.1 | 68.4 | 26,634 |
| 2004 | Presidential | Ages 45 to 64 | 70.4 | 69.9 | 71.0 | 34,070 |
| 2008 | Presidential | Ages 45 to 64 | 69.2 | 68.6 | 69.7 | 34,266 |
| 2012 | Presidential | Ages 45 to 64 | 67.9 | 67.3 | 68.4 | 34,382 |
| 2016 | Presidential | Ages 45 to 64 | 66.6 | 66.1 | 67.2 | 33,170 |
| 2020 | Presidential | Ages 45 to 64 | 71.0 | 70.4 | 71.6 | 27,161 |
| 2024 | Presidential | Ages 45 to 64 | 70.0 | 69.3 | 70.6 | 22,628 |
| 1996 | Presidential | Ages 65 and older | 69.1 | 68.3 | 70.0 | 15,076 |
| 2000 | Presidential | Ages 65 and older | 69.6 | 68.8 | 70.4 | 14,759 |
| 2004 | Presidential | Ages 65 and older | 71.0 | 70.2 | 71.8 | 17,161 |
| 2008 | Presidential | Ages 65 and older | 70.3 | 69.5 | 71.1 | 17,170 |
| 2012 | Presidential | Ages 65 and older | 72.0 | 71.3 | 72.7 | 19,251 |
| 2016 | Presidential | Ages 65 and older | 70.9 | 70.2 | 71.6 | 21,270 |
| 2020 | Presidential | Ages 65 and older | 74.5 | 73.8 | 75.1 | 21,246 |
| 2024 | Presidential | Ages 65 and older | 74.7 | 74.0 | 75.4 | 20,905 |
| 1994 | Midterm | Ages 18 to 29 | 26.2 | 25.5 | 26.9 | 20,018 |
| 1998 | Midterm | Ages 18 to 29 | 22.3 | 21.6 | 23.0 | 16,580 |
| 2002 | Midterm | Ages 18 to 29 | 22.4 | 21.7 | 23.1 | 18,524 |
| 2006 | Midterm | Ages 18 to 29 | 25.5 | 24.8 | 26.2 | 17,898 |
| 2010 | Midterm | Ages 18 to 29 | 23.9 | 23.2 | 24.7 | 18,176 |
| 2014 | Midterm | Ages 18 to 29 | 19.9 | 19.2 | 20.5 | 18,215 |
| 2018 | Midterm | Ages 18 to 29 | 35.6 | 34.7 | 36.4 | 15,758 |
| 2022 | Midterm | Ages 18 to 29 | 31.2 | 30.3 | 32.1 | 12,305 |
| 1994 | Midterm | Ages 30 to 44 | 46.5 | 45.8 | 47.1 | 30,988 |
| 1998 | Midterm | Ages 30 to 44 | 41.8 | 41.2 | 42.5 | 26,396 |
| 2002 | Midterm | Ages 30 to 44 | 42.0 | 41.3 | 42.6 | 28,955 |
| 2006 | Midterm | Ages 30 to 44 | 43.0 | 42.3 | 43.7 | 24,768 |
| 2010 | Midterm | Ages 30 to 44 | 40.4 | 39.6 | 41.1 | 22,958 |
| 2014 | Midterm | Ages 30 to 44 | 35.6 | 34.9 | 36.3 | 22,709 |
| 2018 | Midterm | Ages 30 to 44 | 48.8 | 48.0 | 49.6 | 20,821 |
| 2022 | Midterm | Ages 30 to 44 | 46.9 | 46.1 | 47.8 | 16,909 |
| 1994 | Midterm | Ages 45 to 64 | 59.9 | 59.2 | 60.5 | 26,899 |
| 1998 | Midterm | Ages 45 to 64 | 56.5 | 55.8 | 57.1 | 25,933 |
| 2002 | Midterm | Ages 45 to 64 | 56.1 | 55.5 | 56.7 | 33,729 |
| 2006 | Midterm | Ages 45 to 64 | 57.6 | 57.0 | 58.1 | 34,368 |
| 2010 | Midterm | Ages 45 to 64 | 54.4 | 53.8 | 55.0 | 35,039 |
| 2014 | Midterm | Ages 45 to 64 | 49.6 | 49.0 | 50.2 | 34,444 |
| 2018 | Midterm | Ages 45 to 64 | 59.5 | 58.8 | 60.1 | 30,528 |
| 2022 | Midterm | Ages 45 to 64 | 58.3 | 57.5 | 59.0 | 23,378 |
| 1994 | Midterm | Ages 65 and older | 63.3 | 62.5 | 64.1 | 17,367 |
| 1998 | Midterm | Ages 65 and older | 61.3 | 60.5 | 62.2 | 15,167 |
| 2002 | Midterm | Ages 65 and older | 62.7 | 61.9 | 63.5 | 17,430 |
| 2006 | Midterm | Ages 65 and older | 62.5 | 61.7 | 63.4 | 17,061 |
| 2010 | Midterm | Ages 65 and older | 60.8 | 60.0 | 61.6 | 18,035 |
| 2014 | Midterm | Ages 65 and older | 59.4 | 58.6 | 60.2 | 20,899 |
| 2018 | Midterm | Ages 65 and older | 66.1 | 65.4 | 66.8 | 21,642 |
| 2022 | Midterm | Ages 65 and older | 66.8 | 66.1 | 67.6 | 20,491 |
In midterm years the gap was 37.137.1 percentage pointsEstimanddifference in weighted turnout: citizens 65 and older minus citizens 18 to 29Interval95% CI 36.0 to 38.2n37,385WeightsVOSUPPWTPopulationcitizens 18 and olderPeriod1994 midterm electionSourceCPS Voting and Registration Supplement points in 1994, 39.5 (95% CI 38.5 to 40.5)39.5 percentage pointsEstimanddifference in weighted turnout: citizens 65 and older minus citizens 18 to 29Interval95% CI 38.5 to 40.5n39,114WeightsVOSUPPWTPopulationcitizens 18 and olderPeriod2014 midterm electionSourceCPS Voting and Registration Supplement in 2014 and 30.6 (95% CI 29.4 to 31.7)30.6 percentage pointsEstimanddifference in weighted turnout: citizens 65 and older minus citizens 18 to 29Interval95% CI 29.4 to 31.7n37,400WeightsVOSUPPWTPopulationcitizens 18 and olderPeriod2018 midterm electionSourceCPS Voting and Registration Supplement in 2018, a change of −9.0 (95% CI −10.5 to −7.5)−9.0 percentage pointsEstimandchange in the turnout gap between the two years (difference of two differences), gap = citizens 65 and older minus citizens 18 to 29Interval95% CI −10.5 to −7.5n76,514WeightsVOSUPPWTPopulationcitizens 18 and olderPeriod2014 to 2018SourceCPS Voting and Registration Supplement points from 2014, and it was 35.735.7 percentage pointsEstimanddifference in weighted turnout: citizens 65 and older minus citizens 18 to 29Interval95% CI 34.5 to 36.8n32,796WeightsVOSUPPWTPopulationcitizens 18 and olderPeriod2022 midterm electionSourceCPS Voting and Registration Supplement in 2022, up by +5.1 (95% CI 3.5 to 6.7)+5.1 percentage pointsEstimandchange in the turnout gap between the two years (difference of two differences), gap = citizens 65 and older minus citizens 18 to 29Interval95% CI 3.5 to 6.7n70,196WeightsVOSUPPWTPopulationcitizens 18 and olderPeriod2018 to 2022SourceCPS Voting and Registration Supplement points from 2018. A cross-section cannot say whether the narrowing means that young adults vote more than earlier young adults did, because each survey interviews different people and the age groups differ in other ways too. Chapter 3.6, Age and generation separates age and birth cohortCohortPeople born in the same period, such as the same decade. Comparing cohorts shows whether a difference comes from when people were born, as distinct from their age or the year they were surveyed. where the data allow it, and panel studies that follow the same young adults over time are the usual tool for how voting habits form [5].
Race and ethnicity: Black turnout moved from above White turnout in 2012 to below it in 2024
Figure 4 compares four groups with White non-Hispanic citizens. The White minus Black difference was 7.7 (95% CI 6.5 to 9.0)7.7 percentage pointsEstimanddifference in weighted turnout: White non-Hispanic minus Black non-Hispanic citizensInterval95% CI 6.5 to 9.0n76,814WeightsVOSUPPWTPopulationcitizens 18 and olderPeriod1996 presidential electionSourceCPS Voting and Registration Supplement points in 1996 and 0.9 (95% CI −0.3 to 2.1)0.9 percentage pointsEstimanddifference in weighted turnout: White non-Hispanic minus Black non-Hispanic citizensInterval95% CI −0.3 to 2.1n80,104WeightsVOSUPPWTPopulationcitizens 18 and olderPeriod2008 presidential electionSourceCPS Voting and Registration Supplement in 2008, an interval that includes zero. In 2012 Black turnout was above White turnout, and the difference was −2.5 (95% CI −3.6 to −1.4)−2.5 percentage pointsEstimanddifference in weighted turnout: White non-Hispanic minus Black non-Hispanic citizensInterval95% CI −3.6 to −1.4n80,692WeightsVOSUPPWTPopulationcitizens 18 and olderPeriod2012 presidential electionSourceCPS Voting and Registration Supplement points. It was then 5.75.7 percentage pointsEstimanddifference in weighted turnout: White non-Hispanic minus Black non-Hispanic citizensInterval95% CI 4.5 to 6.9n78,720WeightsVOSUPPWTPopulationcitizens 18 and olderPeriod2016 presidential electionSourceCPS Voting and Registration Supplement in 2016, 8.1 (95% CI 6.9 to 9.4)8.1 percentage pointsEstimanddifference in weighted turnout: White non-Hispanic minus Black non-Hispanic citizensInterval95% CI 6.9 to 9.4n67,974WeightsVOSUPPWTPopulationcitizens 18 and olderPeriod2020 presidential electionSourceCPS Voting and Registration Supplement in 2020 and 10.4 (95% CI 9.0 to 11.8)10.4 percentage pointsEstimanddifference in weighted turnout: White non-Hispanic minus Black non-Hispanic citizensInterval95% CI 9.0 to 11.8n58,873WeightsVOSUPPWTPopulationcitizens 18 and olderPeriod2024 presidential electionSourceCPS Voting and Registration Supplement in 2024, a change of +12.9 (95% CI 11.1 to 14.7)+12.9 percentage pointsEstimandchange in the turnout gap between the two years (difference of two differences), gap = White non-Hispanic minus Black non-Hispanic citizensInterval95% CI 11.1 to 14.7n139,565WeightsVOSUPPWTPopulationcitizens 18 and olderPeriod2012 to 2024SourceCPS Voting and Registration Supplement points from 2012. The White minus Hispanic difference was 16.816.8 percentage pointsEstimanddifference in weighted turnout: White non-Hispanic minus Hispanic citizensInterval95% CI 15.1 to 18.4n73,087WeightsVOSUPPWTPopulationcitizens 18 and olderPeriod1996 presidential electionSourceCPS Voting and Registration Supplement points in 1996 and 19.9 (95% CI 18.6 to 21.2)19.9 percentage pointsEstimanddifference in weighted turnout: White non-Hispanic minus Hispanic citizensInterval95% CI 18.6 to 21.2n60,301WeightsVOSUPPWTPopulationcitizens 18 and olderPeriod2024 presidential electionSourceCPS Voting and Registration Supplement in 2024, a change of +3.1 (95% CI 1.0 to 5.2)+3.1 percentage pointsEstimandchange in the turnout gap between the two years (difference of two differences), gap = White non-Hispanic minus Hispanic citizensInterval95% CI 1.0 to 5.2n133,388WeightsVOSUPPWTPopulationcitizens 18 and olderPeriod1996 to 2024SourceCPS Voting and Registration Supplement points. The White minus Asian difference went from 22.5 (95% CI 20.3 to 24.7)22.5 percentage pointsEstimanddifference in weighted turnout: White non-Hispanic minus Asian or Pacific Islander non-Hispanic citizensInterval95% CI 20.3 to 24.7n79,118WeightsVOSUPPWTPopulationcitizens 18 and olderPeriod2004 presidential electionSourceCPS Voting and Registration Supplement points in 2004 to 11.6 (95% CI 9.8 to 13.4)11.6 percentage pointsEstimanddifference in weighted turnout: White non-Hispanic minus Asian or Pacific Islander non-Hispanic citizensInterval95% CI 9.8 to 13.4n63,415WeightsVOSUPPWTPopulationcitizens 18 and olderPeriod2020 presidential electionSourceCPS Voting and Registration Supplement in 2020 and 12.9 (95% CI 11.0 to 14.7)12.9 percentage pointsEstimanddifference in weighted turnout: White non-Hispanic minus Asian or Pacific Islander non-Hispanic citizensInterval95% CI 11.0 to 14.7n56,090WeightsVOSUPPWTPopulationcitizens 18 and olderPeriod2024 presidential electionSourceCPS Voting and Registration Supplement in 2024, a change of −9.6 (95% CI −12.5 to −6.8)−9.6 percentage pointsEstimandchange in the turnout gap between the two years (difference of two differences), gap = White non-Hispanic minus Asian or Pacific Islander non-Hispanic citizensInterval95% CI −12.5 to −6.8n135,208WeightsVOSUPPWTPopulationcitizens 18 and olderPeriod2004 to 2024SourceCPS Voting and Registration Supplement points since 2004.
Figure 4
Black turnout went from above White turnout in 2012 to below it in 2024, and Hispanic turnout stayed below White turnout in every election
Percent of citizens who reported voting, by race and ethnicity, presidential years 1996 to 2024 (other buttons show midterm years and all residents, noncitizens counted as not voting); 95% intervals from the Kish effective sample size
Show data
| Year | Election type | Universe | Group | Turnout (%) | 95% CI low | 95% CI high | n respondents |
|---|---|---|---|---|---|---|---|
| 1996 | Presidential | Presidential Citizens | White | 60.7 | 60.3 | 61.1 | 68,779 |
| 2000 | Presidential | Presidential Citizens | White | 62.0 | 61.6 | 62.4 | 65,881 |
| 2004 | Presidential | Presidential Citizens | White | 67.2 | 66.8 | 67.6 | 76,209 |
| 2008 | Presidential | Presidential Citizens | White | 66.1 | 65.7 | 66.5 | 71,497 |
| 2012 | Presidential | Presidential Citizens | White | 64.1 | 63.7 | 64.5 | 71,460 |
| 2016 | Presidential | Presidential Citizens | White | 65.3 | 64.9 | 65.7 | 69,164 |
| 2020 | Presidential | Presidential Citizens | White | 70.9 | 70.5 | 71.4 | 59,555 |
| 2024 | Presidential | Presidential Citizens | White | 70.5 | 70.0 | 70.9 | 52,237 |
| 1996 | Presidential | Presidential Citizens | Black | 53.0 | 51.8 | 54.2 | 8,035 |
| 2000 | Presidential | Presidential Citizens | Black | 57.1 | 55.9 | 58.2 | 8,285 |
| 2004 | Presidential | Presidential Citizens | Black | 60.3 | 59.2 | 61.5 | 8,882 |
| 2008 | Presidential | Presidential Citizens | Black | 65.2 | 64.1 | 66.3 | 8,607 |
| 2012 | Presidential | Presidential Citizens | Black | 66.6 | 65.6 | 67.7 | 9,232 |
| 2016 | Presidential | Presidential Citizens | Black | 59.6 | 58.5 | 60.7 | 9,556 |
| 2020 | Presidential | Presidential Citizens | Black | 62.8 | 61.6 | 64.0 | 8,419 |
| 2024 | Presidential | Presidential Citizens | Black | 60.1 | 58.7 | 61.4 | 6,636 |
| 1996 | Presidential | Presidential Citizens | Hispanic | 44.0 | 42.4 | 45.6 | 4,308 |
| 2000 | Presidential | Presidential Citizens | Hispanic | 44.3 | 42.8 | 45.8 | 5,100 |
| 2004 | Presidential | Presidential Citizens | Hispanic | 47.2 | 45.8 | 48.5 | 6,089 |
| 2008 | Presidential | Presidential Citizens | Hispanic | 49.9 | 48.6 | 51.1 | 6,925 |
| 2012 | Presidential | Presidential Citizens | Hispanic | 48.0 | 46.7 | 49.2 | 7,812 |
| 2016 | Presidential | Presidential Citizens | Hispanic | 47.6 | 46.4 | 48.7 | 8,703 |
| 2020 | Presidential | Presidential Citizens | Hispanic | 53.7 | 52.6 | 54.9 | 8,163 |
| 2024 | Presidential | Presidential Citizens | Hispanic | 50.6 | 49.4 | 51.8 | 8,064 |
| 1996 | Presidential | Presidential Citizens | Asian | 45.8 | 43.3 | 48.3 | 1,950 |
| 2000 | Presidential | Presidential Citizens | Asian | 43.2 | 40.8 | 45.6 | 2,174 |
| 2004 | Presidential | Presidential Citizens | Asian | 44.7 | 42.6 | 46.9 | 2,909 |
| 2008 | Presidential | Presidential Citizens | Asian | 47.2 | 45.3 | 49.2 | 3,349 |
| 2012 | Presidential | Presidential Citizens | Asian | 47.1 | 45.2 | 49.0 | 3,640 |
| 2016 | Presidential | Presidential Citizens | Asian | 48.9 | 47.2 | 50.7 | 4,079 |
| 2020 | Presidential | Presidential Citizens | Asian | 59.3 | 57.6 | 61.1 | 3,860 |
| 2024 | Presidential | Presidential Citizens | Asian | 57.6 | 55.9 | 59.4 | 3,853 |
| 1994 | Midterm | Midterm Citizens | White | 51.0 | 50.6 | 51.4 | 78,425 |
| 1998 | Midterm | Midterm Citizens | White | 47.4 | 47.0 | 47.8 | 68,228 |
| 2002 | Midterm | Midterm Citizens | White | 49.1 | 48.7 | 49.5 | 79,499 |
| 2006 | Midterm | Midterm Citizens | White | 51.6 | 51.2 | 52.0 | 73,383 |
| 2010 | Midterm | Midterm Citizens | White | 48.6 | 48.2 | 49.0 | 72,006 |
| 2014 | Midterm | Midterm Citizens | White | 45.8 | 45.3 | 46.2 | 71,723 |
| 2018 | Midterm | Midterm Citizens | White | 57.5 | 57.0 | 57.9 | 64,983 |
| 2022 | Midterm | Midterm Citizens | White | 57.6 | 57.2 | 58.1 | 52,569 |
| 1994 | Midterm | Midterm Citizens | Black | 39.0 | 37.9 | 40.1 | 9,512 |
| 1998 | Midterm | Midterm Citizens | Black | 41.8 | 40.7 | 43.0 | 7,993 |
| 2002 | Midterm | Midterm Citizens | Black | 42.7 | 41.6 | 43.8 | 9,310 |
| 2006 | Midterm | Midterm Citizens | Black | 41.3 | 40.1 | 42.4 | 8,671 |
| 2010 | Midterm | Midterm Citizens | Black | 44.0 | 42.9 | 45.1 | 9,097 |
| 2014 | Midterm | Midterm Citizens | Black | 40.6 | 39.5 | 41.7 | 9,912 |
| 2018 | Midterm | Midterm Citizens | Black | 51.4 | 50.2 | 52.6 | 8,773 |
| 2022 | Midterm | Midterm Citizens | Black | 45.6 | 44.3 | 47.0 | 7,247 |
| 1994 | Midterm | Midterm Citizens | Hispanic | 34.4 | 32.8 | 35.9 | 4,343 |
| 1998 | Midterm | Midterm Citizens | Hispanic | 32.8 | 31.4 | 34.3 | 4,707 |
| 2002 | Midterm | Midterm Citizens | Hispanic | 30.4 | 29.1 | 31.7 | 5,872 |
| 2006 | Midterm | Midterm Citizens | Hispanic | 32.3 | 31.1 | 33.5 | 6,558 |
| 2010 | Midterm | Midterm Citizens | Hispanic | 31.2 | 30.1 | 32.4 | 7,522 |
| 2014 | Midterm | Midterm Citizens | Hispanic | 27.0 | 26.0 | 28.0 | 8,243 |
| 2018 | Midterm | Midterm Citizens | Hispanic | 40.4 | 39.3 | 41.5 | 8,745 |
| 2022 | Midterm | Midterm Citizens | Hispanic | 37.9 | 36.7 | 39.0 | 7,741 |
| 1994 | Midterm | Midterm Citizens | Asian | 40.0 | 37.4 | 42.5 | 1,820 |
| 1998 | Midterm | Midterm Citizens | Asian | 32.3 | 30.1 | 34.6 | 2,163 |
| 2002 | Midterm | Midterm Citizens | Asian | 31.0 | 29.0 | 33.1 | 2,779 |
| 2006 | Midterm | Midterm Citizens | Asian | 32.0 | 30.1 | 33.9 | 3,189 |
| 2010 | Midterm | Midterm Citizens | Asian | 31.3 | 29.6 | 33.1 | 3,592 |
| 2014 | Midterm | Midterm Citizens | Asian | 27.6 | 26.0 | 29.1 | 4,076 |
| 2018 | Midterm | Midterm Citizens | Asian | 40.4 | 38.7 | 42.1 | 4,100 |
| 2022 | Midterm | Midterm Citizens | Asian | 40.2 | 38.5 | 42.0 | 3,755 |
| 1996 | Presidential | Presidential All residents | White | 59.6 | 59.2 | 60.0 | 70,039 |
| 2000 | Presidential | Presidential All residents | White | 60.6 | 60.2 | 61.1 | 67,287 |
| 2004 | Presidential | Presidential All residents | White | 65.8 | 65.4 | 66.1 | 77,710 |
| 2008 | Presidential | Presidential All residents | White | 64.8 | 64.4 | 65.2 | 72,783 |
| 2012 | Presidential | Presidential All residents | White | 63.0 | 62.6 | 63.4 | 72,615 |
| 2016 | Presidential | Presidential All residents | White | 64.1 | 63.7 | 64.5 | 70,249 |
| 2020 | Presidential | Presidential All residents | White | 69.8 | 69.3 | 70.2 | 60,390 |
| 2024 | Presidential | Presidential All residents | White | 69.2 | 68.8 | 69.7 | 53,001 |
| 1996 | Presidential | Presidential All residents | Black | 50.9 | 49.7 | 52.1 | 8,374 |
| 2000 | Presidential | Presidential All residents | Black | 54.2 | 53.1 | 55.4 | 8,698 |
| 2004 | Presidential | Presidential All residents | Black | 57.2 | 56.1 | 58.3 | 9,322 |
| 2008 | Presidential | Presidential All residents | Black | 61.7 | 60.6 | 62.8 | 9,100 |
| 2012 | Presidential | Presidential All residents | Black | 63.5 | 62.4 | 64.5 | 9,696 |
| 2016 | Presidential | Presidential All residents | Black | 56.5 | 55.4 | 57.6 | 10,002 |
| 2020 | Presidential | Presidential All residents | Black | 59.5 | 58.4 | 60.7 | 8,817 |
| 2024 | Presidential | Presidential All residents | Black | 56.1 | 54.8 | 57.5 | 7,056 |
| 1996 | Presidential | Presidential All residents | Hispanic | 26.7 | 25.6 | 27.9 | 6,969 |
| 2000 | Presidential | Presidential All residents | Hispanic | 26.7 | 25.7 | 27.7 | 8,382 |
| 2004 | Presidential | Presidential All residents | Hispanic | 28.0 | 27.0 | 28.9 | 10,174 |
| 2008 | Presidential | Presidential All residents | Hispanic | 31.6 | 30.6 | 32.5 | 10,871 |
| 2012 | Presidential | Presidential All residents | Hispanic | 31.8 | 30.9 | 32.7 | 11,846 |
| 2016 | Presidential | Presidential All residents | Hispanic | 32.5 | 31.6 | 33.4 | 12,622 |
| 2020 | Presidential | Presidential All residents | Hispanic | 38.8 | 37.8 | 39.7 | 11,259 |
| 2024 | Presidential | Presidential All residents | Hispanic | 35.2 | 34.3 | 36.2 | 11,556 |
| 1996 | Presidential | Presidential All residents | Asian | 25.9 | 24.2 | 27.6 | 3,200 |
| 2000 | Presidential | Presidential All residents | Asian | 25.1 | 23.5 | 26.7 | 3,515 |
| 2004 | Presidential | Presidential All residents | Asian | 30.5 | 28.8 | 32.1 | 4,104 |
| 2008 | Presidential | Presidential All residents | Asian | 32.1 | 30.6 | 33.6 | 4,779 |
| 2012 | Presidential | Presidential All residents | Asian | 31.8 | 30.3 | 33.2 | 5,217 |
| 2016 | Presidential | Presidential All residents | Asian | 34.1 | 32.7 | 35.5 | 5,752 |
| 2020 | Presidential | Presidential All residents | Asian | 42.7 | 41.2 | 44.2 | 5,206 |
| 2024 | Presidential | Presidential All residents | Asian | 41.9 | 40.4 | 43.4 | 5,207 |
| 1994 | Midterm | Midterm All residents | White | 50.1 | 49.7 | 50.5 | 79,878 |
| 1998 | Midterm | Midterm All residents | White | 46.5 | 46.1 | 46.9 | 69,529 |
| 2002 | Midterm | Midterm All residents | White | 48.0 | 47.6 | 48.4 | 81,060 |
| 2006 | Midterm | Midterm All residents | White | 50.5 | 50.1 | 50.9 | 74,748 |
| 2010 | Midterm | Midterm All residents | White | 47.8 | 47.4 | 48.2 | 73,169 |
| 2014 | Midterm | Midterm All residents | White | 45.0 | 44.6 | 45.4 | 72,831 |
| 2018 | Midterm | Midterm All residents | White | 56.5 | 56.1 | 56.9 | 65,927 |
| 2022 | Midterm | Midterm All residents | White | 56.6 | 56.1 | 57.1 | 53,341 |
| 1994 | Midterm | Midterm All residents | Black | 37.4 | 36.3 | 38.5 | 9,945 |
| 1998 | Midterm | Midterm All residents | Black | 40.0 | 38.9 | 41.2 | 8,365 |
| 2002 | Midterm | Midterm All residents | Black | 40.4 | 39.3 | 41.5 | 9,803 |
| 2006 | Midterm | Midterm All residents | Black | 39.2 | 38.1 | 40.3 | 9,119 |
| 2010 | Midterm | Midterm All residents | Black | 41.6 | 40.5 | 42.7 | 9,650 |
| 2014 | Midterm | Midterm All residents | Black | 38.6 | 37.5 | 39.7 | 10,387 |
| 2018 | Midterm | Midterm All residents | Black | 48.8 | 47.6 | 50.0 | 9,190 |
| 2022 | Midterm | Midterm All residents | Black | 43.4 | 42.2 | 44.7 | 7,574 |
| 1994 | Midterm | Midterm All residents | Hispanic | 20.2 | 19.1 | 21.2 | 7,222 |
| 1998 | Midterm | Midterm All residents | Hispanic | 20.0 | 19.1 | 21.0 | 7,656 |
| 2002 | Midterm | Midterm All residents | Hispanic | 18.9 | 18.0 | 19.7 | 9,416 |
| 2006 | Midterm | Midterm All residents | Hispanic | 19.3 | 18.5 | 20.1 | 10,912 |
| 2010 | Midterm | Midterm All residents | Hispanic | 20.5 | 19.7 | 21.3 | 11,500 |
| 2014 | Midterm | Midterm All residents | Hispanic | 18.4 | 17.7 | 19.2 | 12,156 |
| 2018 | Midterm | Midterm All residents | Hispanic | 28.5 | 27.6 | 29.4 | 12,237 |
| 2022 | Midterm | Midterm All residents | Hispanic | 26.7 | 25.8 | 27.6 | 10,928 |
| 1994 | Midterm | Midterm All residents | Asian | 21.8 | 20.2 | 23.4 | 3,258 |
| 1998 | Midterm | Midterm All residents | Asian | 19.2 | 17.7 | 20.6 | 3,439 |
| 2002 | Midterm | Midterm All residents | Asian | 19.2 | 17.9 | 20.6 | 4,207 |
| 2006 | Midterm | Midterm All residents | Asian | 21.7 | 20.3 | 23.0 | 4,533 |
| 2010 | Midterm | Midterm All residents | Asian | 21.8 | 20.5 | 23.2 | 5,006 |
| 2014 | Midterm | Midterm All residents | Asian | 19.5 | 18.4 | 20.7 | 5,606 |
| 2018 | Midterm | Midterm All residents | Asian | 28.8 | 27.5 | 30.2 | 5,621 |
| 2022 | Midterm | Midterm All residents | Asian | 29.7 | 28.3 | 31.1 | 4,957 |
The buttons for all residents count noncitizens as not voting. Noncitizens were 30.3%30.3% percentEstimandweighted share who are not US citizensWeightsVOSUPPWTPopulationHispanic adults 18 and olderPeriod2024SourceCPS Voting and Registration Supplement of Hispanic adults and 27.3%27.3% percentEstimandweighted share who are not US citizensWeightsVOSUPPWTPopulationAsian or Pacific Islander non-Hispanic adults 18 and olderPeriod2024SourceCPS Voting and Registration Supplement of Asian adults in 2024, against 1.8%1.8% percentEstimandweighted share who are not US citizensWeightsVOSUPPWTPopulationWhite non-Hispanic adults 18 and olderPeriod2024SourceCPS Voting and Registration Supplement of White non-Hispanic adults, so the denominator moves the Hispanic and Asian lines far more than the White line. The White minus Hispanic difference was 34.0 (95% CI 32.9 to 35.0)34.0 percentage pointsEstimanddifference in weighted turnout, White non-Hispanic minus Hispanic residents (noncitizens counted as not voting)Interval95% CI 32.9 to 35.0n64,557WeightsVOSUPPWTPopulationall residents 18 and olderPeriod2024SourceCPS Voting and Registration Supplement points among all residents in 2024 against 19.919.9 percentage pointsEstimanddifference in weighted turnout: White non-Hispanic minus Hispanic citizensInterval95% CI 18.6 to 21.2n60,301WeightsVOSUPPWTPopulationcitizens 18 and olderPeriod2024 presidential electionSourceCPS Voting and Registration Supplement among citizens. Chapter 3.4, Hispanic, Black and Asian voters (planned) takes up how the three groups voted. The race groups are single-race and non-Hispanic, except for the Hispanic group of any race, and the survey allowed respondents to report more than one race from 2003.
Education: the gap between a bachelor's degree and no diploma was wider in 2024 than in 1996
Education is measured for citizens twenty-five and older, and Figure 5 shows four levels. The order of the levels is the same in every election. In presidential years the gap between a bachelor's degree and no high school diploma was 36.1 (95% CI 34.9 to 37.3)36.1 percentage pointsEstimanddifference in weighted turnout: citizens 25 and older with a bachelor's degree minus those without a high school diplomaInterval95% CI 34.9 to 37.3n29,768WeightsVOSUPPWTPopulationcitizens 25 and olderPeriod1996 presidential electionSourceCPS Voting and Registration Supplement points in 1996, 37.737.7 percentage pointsEstimanddifference in weighted turnout: citizens 25 and older with a bachelor's degree minus those without a high school diplomaInterval95% CI 36.4 to 39.0n33,092WeightsVOSUPPWTPopulationcitizens 25 and olderPeriod2008 presidential electionSourceCPS Voting and Registration Supplement in 2008, 36.3 (95% CI 34.6 to 37.9)36.3 percentage pointsEstimanddifference in weighted turnout: citizens 25 and older with a bachelor's degree minus those without a high school diplomaInterval95% CI 34.6 to 37.9n32,130WeightsVOSUPPWTPopulationcitizens 25 and olderPeriod2020 presidential electionSourceCPS Voting and Registration Supplement in 2020 and 40.5 (95% CI 38.7 to 42.4)40.5 percentage pointsEstimanddifference in weighted turnout: citizens 25 and older with a bachelor's degree minus those without a high school diplomaInterval95% CI 38.7 to 42.4n29,750WeightsVOSUPPWTPopulationcitizens 25 and olderPeriod2024 presidential electionSourceCPS Voting and Registration Supplement in 2024. The change from 1996 to 2024 was +4.5 (95% CI 2.3 to 6.6)+4.5 percentage pointsEstimandchange in the turnout gap between the two years (difference of two differences), gap = bachelor's degree or more minus no high school diplomaInterval95% CI 2.3 to 6.6n59,518WeightsVOSUPPWTPopulationcitizens 25 and olderPeriod1996 to 2024SourceCPS Voting and Registration Supplement points, and the change from 2020 to 2024 was +4.3 (95% CI 1.8 to 6.7)+4.3 percentage pointsEstimandchange in the turnout gap between the two years (difference of two differences), gap = bachelor's degree or more minus no high school diplomaInterval95% CI 1.8 to 6.7n61,880WeightsVOSUPPWTPopulationcitizens 25 and olderPeriod2020 to 2024SourceCPS Voting and Registration Supplement points. The gap between a bachelor's degree and a high school diploma changed less, from 22.9 (95% CI 21.9 to 23.8)22.9 percentage pointsEstimanddifference in weighted turnout: citizens 25 and older with a bachelor's degree minus those with a high school diploma onlyInterval95% CI 21.9 to 23.8n43,901WeightsVOSUPPWTPopulationcitizens 25 and olderPeriod1996 presidential electionSourceCPS Voting and Registration Supplement points in 1996 to 24.3 (95% CI 23.3 to 25.3)24.3 percentage pointsEstimanddifference in weighted turnout: citizens 25 and older with a bachelor's degree minus those with a high school diploma onlyInterval95% CI 23.3 to 25.3n44,125WeightsVOSUPPWTPopulationcitizens 25 and olderPeriod2024 presidential electionSourceCPS Voting and Registration Supplement in 2024, a change of +1.4 (95% CI 0.1 to 2.8)+1.4 percentage pointsEstimandchange in the turnout gap between the two years (difference of two differences), gap = bachelor's degree or more minus high school diploma onlyInterval95% CI 0.1 to 2.8n88,026WeightsVOSUPPWTPopulationcitizens 25 and olderPeriod1996 to 2024SourceCPS Voting and Registration Supplement points.
Figure 5
The turnout gap between citizens with a bachelor's degree and those without a high school diploma was wider in 2024 than in 1996
Percent of citizens 25 and older who reported voting, by educational attainment, presidential years 1996 to 2024 (the second button shows midterm years); 95% intervals from the Kish effective sample size
Reported voting (% of citizens 25 and older in the group)
Show data
| Year | Election type | Group | Turnout (%) | 95% CI low | 95% CI high | n respondents |
|---|---|---|---|---|---|---|
| 1996 | Presidential | Less than high school | 41.9 | 41.0 | 42.9 | 11,692 |
| 2000 | Presidential | Less than high school | 42.0 | 41.0 | 43.1 | 9,858 |
| 2004 | Presidential | Less than high school | 42.2 | 41.1 | 43.3 | 10,233 |
| 2008 | Presidential | Less than high school | 41.6 | 40.4 | 42.8 | 8,467 |
| 2012 | Presidential | Less than high school | 41.2 | 40.0 | 42.5 | 7,512 |
| 2016 | Presidential | Less than high school | 37.3 | 36.0 | 38.6 | 6,908 |
| 2020 | Presidential | Less than high school | 43.7 | 42.1 | 45.2 | 5,000 |
| 2024 | Presidential | Less than high school | 39.1 | 37.4 | 40.8 | 4,008 |
| 1996 | Presidential | High school graduate | 55.1 | 54.5 | 55.8 | 25,825 |
| 2000 | Presidential | High school graduate | 56.0 | 55.3 | 56.7 | 24,610 |
| 2004 | Presidential | High school graduate | 59.0 | 58.3 | 59.6 | 28,012 |
| 2008 | Presidential | High school graduate | 57.0 | 56.3 | 57.7 | 26,066 |
| 2012 | Presidential | High school graduate | 56.0 | 55.3 | 56.7 | 25,733 |
| 2016 | Presidential | High school graduate | 53.9 | 53.2 | 54.6 | 24,702 |
| 2020 | Presidential | High school graduate | 58.0 | 57.3 | 58.8 | 21,412 |
| 2024 | Presidential | High school graduate | 55.3 | 54.4 | 56.1 | 18,383 |
| 1996 | Presidential | Some college | 67.0 | 66.3 | 67.8 | 18,879 |
| 2000 | Presidential | Some college | 67.3 | 66.6 | 68.0 | 19,364 |
| 2004 | Presidential | Some college | 71.6 | 70.9 | 72.3 | 22,991 |
| 2008 | Presidential | Some college | 70.6 | 69.9 | 71.3 | 22,758 |
| 2012 | Presidential | Some college | 67.6 | 66.9 | 68.3 | 23,322 |
| 2016 | Presidential | Some college | 66.0 | 65.3 | 66.7 | 23,525 |
| 2020 | Presidential | Some college | 71.1 | 70.4 | 71.8 | 20,091 |
| 2024 | Presidential | Some college | 69.2 | 68.4 | 70.0 | 17,338 |
| 1996 | Presidential | Bachelor's degree or more | 78.0 | 77.4 | 78.7 | 18,076 |
| 2000 | Presidential | Bachelor's degree or more | 78.1 | 77.5 | 78.8 | 19,070 |
| 2004 | Presidential | Bachelor's degree or more | 80.3 | 79.7 | 80.9 | 24,193 |
| 2008 | Presidential | Bachelor's degree or more | 79.3 | 78.7 | 79.9 | 24,625 |
| 2012 | Presidential | Bachelor's degree or more | 77.8 | 77.3 | 78.4 | 27,231 |
| 2016 | Presidential | Bachelor's degree or more | 77.1 | 76.6 | 77.7 | 28,684 |
| 2020 | Presidential | Bachelor's degree or more | 80.0 | 79.4 | 80.5 | 27,130 |
| 2024 | Presidential | Bachelor's degree or more | 79.6 | 79.0 | 80.2 | 25,742 |
| 1994 | Midterm | Less than high school | 34.1 | 33.2 | 34.9 | 14,134 |
| 1998 | Midterm | Less than high school | 33.4 | 32.4 | 34.4 | 10,963 |
| 2002 | Midterm | Less than high school | 31.0 | 30.0 | 32.0 | 10,837 |
| 2006 | Midterm | Less than high school | 30.4 | 29.3 | 31.4 | 9,376 |
| 2010 | Midterm | Less than high school | 28.3 | 27.2 | 29.5 | 8,070 |
| 2014 | Midterm | Less than high school | 25.1 | 24.0 | 26.2 | 7,697 |
| 2018 | Midterm | Less than high school | 30.0 | 28.7 | 31.3 | 6,074 |
| 2022 | Midterm | Less than high school | 29.2 | 27.7 | 30.8 | 4,251 |
| 1994 | Midterm | High school graduate | 46.7 | 46.1 | 47.4 | 29,451 |
| 1998 | Midterm | High school graduate | 42.7 | 42.0 | 43.4 | 25,556 |
| 2002 | Midterm | High school graduate | 43.4 | 42.7 | 44.0 | 29,364 |
| 2006 | Midterm | High school graduate | 43.9 | 43.3 | 44.6 | 27,085 |
| 2010 | Midterm | High school graduate | 41.3 | 40.6 | 42.0 | 26,354 |
| 2014 | Midterm | High school graduate | 37.2 | 36.6 | 37.9 | 25,791 |
| 2018 | Midterm | High school graduate | 44.9 | 44.2 | 45.6 | 23,004 |
| 2022 | Midterm | High school graduate | 43.9 | 43.1 | 44.7 | 18,919 |
| 1994 | Midterm | Some college | 56.9 | 56.2 | 57.7 | 21,201 |
| 1998 | Midterm | Some college | 53.6 | 52.8 | 54.4 | 19,236 |
| 2002 | Midterm | Some college | 53.2 | 52.5 | 54.0 | 23,691 |
| 2006 | Midterm | Some college | 54.4 | 53.7 | 55.2 | 22,620 |
| 2010 | Midterm | Some college | 50.7 | 50.0 | 51.4 | 23,320 |
| 2014 | Midterm | Some college | 46.5 | 45.8 | 47.2 | 23,955 |
| 2018 | Midterm | Some college | 57.7 | 56.9 | 58.4 | 22,009 |
| 2022 | Midterm | Some college | 55.8 | 55.0 | 56.6 | 17,609 |
| 1994 | Midterm | Bachelor's degree or more | 68.9 | 68.1 | 69.6 | 19,319 |
| 1998 | Midterm | Bachelor's degree or more | 62.5 | 61.8 | 63.3 | 18,893 |
| 2002 | Midterm | Bachelor's degree or more | 64.5 | 63.8 | 65.2 | 23,619 |
| 2006 | Midterm | Bachelor's degree or more | 65.3 | 64.7 | 66.0 | 24,389 |
| 2010 | Midterm | Bachelor's degree or more | 62.3 | 61.6 | 63.0 | 25,857 |
| 2014 | Midterm | Bachelor's degree or more | 57.6 | 57.0 | 58.3 | 28,265 |
| 2018 | Midterm | Bachelor's degree or more | 69.5 | 68.9 | 70.1 | 28,734 |
| 2022 | Midterm | Bachelor's degree or more | 67.7 | 67.0 | 68.3 | 25,233 |
In midterm years the gap was 29.129.1 percentage pointsEstimanddifference in weighted turnout: citizens 25 and older with a bachelor's degree minus those without a high school diplomaInterval95% CI 27.9 to 30.4n29,856WeightsVOSUPPWTPopulationcitizens 25 and olderPeriod1998 midterm electionSourceCPS Voting and Registration Supplement points in 1998, 32.532.5 percentage pointsEstimanddifference in weighted turnout: citizens 25 and older with a bachelor's degree minus those without a high school diplomaInterval95% CI 31.3 to 33.8n35,962WeightsVOSUPPWTPopulationcitizens 25 and olderPeriod2014 midterm electionSourceCPS Voting and Registration Supplement in 2014 and 38.4 (95% CI 36.7 to 40.1)38.4 percentage pointsEstimanddifference in weighted turnout: citizens 25 and older with a bachelor's degree minus those without a high school diplomaInterval95% CI 36.7 to 40.1n29,484WeightsVOSUPPWTPopulationcitizens 25 and olderPeriod2022 midterm electionSourceCPS Voting and Registration Supplement in 2022. Chapter 3.2, The diploma divide covers the issues on which graduates and non-graduates differ.
Family income: turnout rose with income, and the gap changed little from 2012 to 2024
Family income is reported in brackets that differ across decades: fourteen through 2002 and sixteen from 2004, with different dollar limits. A fixed dollar scale would compare different parts of the distribution, so the groups here are within-year thirds. In each year the brackets are laid out along the distribution of citizens, a bracket that straddles a cut-off is split in proportion, and each group holds the same weighted number of citizens. Figure 6 shows the result. Turnout rose with income in every election.
Figure 6
Turnout rose with family income in every election, and the highest third voted at rates about the same distance above the lowest third in 2012 and 2024
Percent of citizens who reported voting, by within-year third of family income, presidential years 1996 to 2024 (the second button shows midterm years); 95% intervals from the Kish effective sample size
Reported voting (% of citizens in the group)
Show data
| Year | Election type | Group | Turnout (%) | 95% CI low | 95% CI high | n respondents |
|---|---|---|---|---|---|---|
| 1996 | Presidential | Lowest third of family income | 48.0 | 47.3 | 48.7 | 25,451 |
| 2000 | Presidential | Lowest third of family income | 49.5 | 48.8 | 50.2 | 27,793 |
| 2004 | Presidential | Lowest third of family income | 54.9 | 54.2 | 55.6 | 28,685 |
| 2008 | Presidential | Lowest third of family income | 57.0 | 56.3 | 57.7 | 27,450 |
| 2012 | Presidential | Lowest third of family income | 51.6 | 51.0 | 52.3 | 31,102 |
| 2016 | Presidential | Lowest third of family income | 50.2 | 49.6 | 50.8 | 32,737 |
| 2020 | Presidential | Lowest third of family income | 56.1 | 55.5 | 56.8 | 30,319 |
| 2024 | Presidential | Lowest third of family income | 54.3 | 53.6 | 55.0 | 26,368 |
| 1996 | Presidential | Middle third | 59.2 | 58.5 | 59.9 | 37,994 |
| 2000 | Presidential | Middle third | 63.0 | 62.3 | 63.7 | 29,923 |
| 2004 | Presidential | Middle third | 67.9 | 67.3 | 68.5 | 34,409 |
| 2008 | Presidential | Middle third | 68.7 | 68.0 | 69.3 | 26,322 |
| 2012 | Presidential | Middle third | 62.1 | 61.5 | 62.7 | 37,036 |
| 2016 | Presidential | Middle third | 62.5 | 61.9 | 63.1 | 42,544 |
| 2020 | Presidential | Middle third | 67.7 | 67.1 | 68.3 | 33,174 |
| 2024 | Presidential | Middle third | 67.2 | 66.6 | 67.8 | 35,210 |
| 1996 | Presidential | Highest third | 71.8 | 71.2 | 72.4 | 24,604 |
| 2000 | Presidential | Highest third | 72.7 | 72.1 | 73.4 | 22,500 |
| 2004 | Presidential | Highest third | 78.2 | 77.6 | 78.7 | 30,412 |
| 2008 | Presidential | Highest third | 77.9 | 77.3 | 78.5 | 33,247 |
| 2012 | Presidential | Highest third | 71.7 | 71.1 | 72.3 | 41,910 |
| 2016 | Presidential | Highest third | 71.4 | 70.9 | 72.0 | 35,668 |
| 2020 | Presidential | Highest third | 76.5 | 75.9 | 77.1 | 36,140 |
| 2024 | Presidential | Highest third | 74.5 | 73.9 | 75.1 | 28,984 |
| 1994 | Midterm | Lowest third of family income | 37.6 | 37.0 | 38.2 | 32,705 |
| 1998 | Midterm | Lowest third of family income | 36.9 | 36.3 | 37.6 | 27,133 |
| 2002 | Midterm | Lowest third of family income | 37.9 | 37.3 | 38.6 | 30,713 |
| 2006 | Midterm | Lowest third of family income | 39.6 | 38.9 | 40.3 | 25,387 |
| 2010 | Midterm | Lowest third of family income | 35.8 | 35.2 | 36.4 | 32,675 |
| 2014 | Midterm | Lowest third of family income | 32.7 | 32.1 | 33.2 | 35,868 |
| 2018 | Midterm | Lowest third of family income | 43.3 | 42.6 | 43.9 | 34,941 |
| 2022 | Midterm | Lowest third of family income | 41.8 | 41.1 | 42.5 | 29,857 |
| 1994 | Midterm | Middle third | 48.9 | 48.3 | 49.5 | 36,492 |
| 1998 | Midterm | Middle third | 46.5 | 45.8 | 47.1 | 31,286 |
| 2002 | Midterm | Middle third | 48.2 | 47.5 | 48.8 | 35,262 |
| 2006 | Midterm | Middle third | 51.9 | 51.2 | 52.6 | 31,974 |
| 2010 | Midterm | Middle third | 45.7 | 45.1 | 46.3 | 37,807 |
| 2014 | Midterm | Middle third | 42.9 | 42.3 | 43.4 | 43,489 |
| 2018 | Midterm | Middle third | 54.1 | 53.5 | 54.7 | 35,374 |
| 2022 | Midterm | Middle third | 53.3 | 52.7 | 54.0 | 35,033 |
| 1994 | Midterm | Highest third | 60.8 | 60.2 | 61.5 | 34,235 |
| 1998 | Midterm | Highest third | 56.0 | 55.3 | 56.7 | 27,870 |
| 2002 | Midterm | Highest third | 57.3 | 56.7 | 58.0 | 29,617 |
| 2006 | Midterm | Highest third | 60.2 | 59.6 | 60.9 | 32,257 |
| 2010 | Midterm | Highest third | 55.1 | 54.5 | 55.7 | 39,511 |
| 2014 | Midterm | Highest third | 50.3 | 49.7 | 50.9 | 33,852 |
| 2018 | Midterm | Highest third | 62.9 | 62.3 | 63.5 | 36,649 |
| 2022 | Midterm | Highest third | 61.5 | 60.8 | 62.2 | 25,708 |
The highest third voted 23.8 (95% CI 22.9 to 24.7)23.8 percentage pointsEstimanddifference in weighted turnout: highest minus lowest third of family incomeInterval95% CI 22.9 to 24.7n50,055WeightsVOSUPPWTPopulationcitizens 18 and olderPeriod1996 presidential electionSourceCPS Voting and Registration Supplement points above the lowest in 1996, 20.9 (95% CI 20.0 to 21.8)20.9 percentage pointsEstimanddifference in weighted turnout: highest minus lowest third of family incomeInterval95% CI 20.0 to 21.8n60,697WeightsVOSUPPWTPopulationcitizens 18 and olderPeriod2008 presidential electionSourceCPS Voting and Registration Supplement in 2008, 20.1 (95% CI 19.2 to 20.9)20.1 percentage pointsEstimanddifference in weighted turnout: highest minus lowest third of family incomeInterval95% CI 19.2 to 20.9n73,012WeightsVOSUPPWTPopulationcitizens 18 and olderPeriod2012 presidential electionSourceCPS Voting and Registration Supplement in 2012 and 20.2 (95% CI 19.3 to 21.1)20.2 percentage pointsEstimanddifference in weighted turnout: highest minus lowest third of family incomeInterval95% CI 19.3 to 21.1n55,352WeightsVOSUPPWTPopulationcitizens 18 and olderPeriod2024 presidential electionSourceCPS Voting and Registration Supplement in 2024. The line breaks between 2008 and 2010 because the treatment of missing income changes. Through 2008 citizens who did not report a bracket are outside the thirds. The share who did not report rose from 11.4%11.4% percentEstimandweighted share who did not report a family-income bracketn84,090WeightsVOSUPPWTPopulationcitizens 18 and olderPeriod1996SourceCPS Voting and Registration Supplement of citizens in 1996 to 20.6%20.6% percentEstimandweighted share who did not report a family-income bracketn92,360WeightsVOSUPPWTPopulationcitizens 18 and olderPeriod2008SourceCPS Voting and Registration Supplement in 2008, and in 2008 they voted at 47.3%47.3% percentEstimandweighted share who reported voting (Census convention: nonresponse counted as not voting)n18,033WeightsVOSUPPWTPopulationcitizens 18 and older who did not report family incomePeriod2008SourceCPS Voting and Registration Supplement against 67.9%67.9% percentEstimandweighted share who reported voting (Census convention: nonresponse counted as not voting)n74,327WeightsVOSUPPWTPopulationcitizens 18 and older who reported family incomePeriod2008SourceCPS Voting and Registration Supplement for citizens who reported income. From 2010 the Census Bureau allocates missing family income, so the thirds include allocated values. The two segments are not comparable in level, and the change within the first, from 1996 to 2008, was −2.9 (95% CI −4.2 to −1.6)−2.9 percentage pointsEstimandchange in the turnout gap between the two years (difference of two differences), gap = highest minus lowest third of family incomeInterval95% CI −4.2 to −1.6n110,752WeightsVOSUPPWTPopulationcitizens 18 and olderPeriod1996 to 2008SourceCPS Voting and Registration Supplement points, while the change within the second, from 2012 to 2024, was +0.2 (95% CI −1.1 to 1.4)+0.2 percentage pointsEstimandchange in the turnout gap between the two years (difference of two differences), gap = highest minus lowest third of family incomeInterval95% CI −1.1 to 1.4n128,364WeightsVOSUPPWTPopulationcitizens 18 and olderPeriod2012 to 2024SourceCPS Voting and Registration Supplement points, an interval that includes zero. In midterm years the gap was 19.119.1 percentage pointsEstimanddifference in weighted turnout: highest minus lowest third of family incomeInterval95% CI 18.1 to 20.0n55,003WeightsVOSUPPWTPopulationcitizens 18 and olderPeriod1998 midterm electionSourceCPS Voting and Registration Supplement points in 1998, 17.717.7 percentage pointsEstimanddifference in weighted turnout: highest minus lowest third of family incomeInterval95% CI 16.8 to 18.5n69,720WeightsVOSUPPWTPopulationcitizens 18 and olderPeriod2014 midterm electionSourceCPS Voting and Registration Supplement in 2014, 19.719.7 percentage pointsEstimanddifference in weighted turnout: highest minus lowest third of family incomeInterval95% CI 18.8 to 20.5n71,590WeightsVOSUPPWTPopulationcitizens 18 and olderPeriod2018 midterm electionSourceCPS Voting and Registration Supplement in 2018 and 19.719.7 percentage pointsEstimanddifference in weighted turnout: highest minus lowest third of family incomeInterval95% CI 18.7 to 20.7n55,565WeightsVOSUPPWTPopulationcitizens 18 and olderPeriod2022 midterm electionSourceCPS Voting and Registration Supplement in 2022. Chapter 3.3, The income ladder takes up how income relates to the choice of party among those who vote.
Net of age, education and income, the White-Black difference shrinks and the Asian-White difference does not
The groups overlap: older citizens differ from younger ones in education, and White citizens differ from Black and Hispanic citizens in education and income. Figure 7 compares each difference with one characteristic in a linear probability modelLinear probability modelA regression of a yes-or-no outcome on predictors, fitted by least squares so that each coefficient is a change in the probability of yes. The book weights it and reports robust standard errors. against the same difference with age, race and ethnicity, education and family income in the model together. The models cover citizens twenty-five and older.
Figure 7
Net of age, education and family income, the White-Black turnout difference was 3.4 points in 2024 and the Asian-White difference did not shrink
Difference in turnout between the higher-turnout and lower-turnout group, percentage points, citizens 25 and older, 2024 (buttons show other elections); one characteristic alone against all four together; 95% intervals from standard errors clustered by state
- One characteristic alone
- All four together
Show data
| Year | Contrast | One characteristic alone (points) | 95% CI low | 95% CI high | All four together (points) | 95% CI low (adjusted) | 95% CI high (adjusted) | n respondents |
|---|---|---|---|---|---|---|---|---|
| 2024 | Ages 65 and older minus ages 25 to 44 | +14.5 | 13.2 | 15.7 | +18.2 | 17.0 | 19.5 | 65,471 |
| 2024 | White minus Black | +9.7 | 7.6 | 11.8 | +3.4 | 1.6 | 5.3 | 65,471 |
| 2024 | White minus Hispanic | +18.8 | 16.5 | 21.2 | +10.5 | 8.1 | 12.9 | 65,471 |
| 2024 | White minus Asian | +13.1 | 11.4 | 14.8 | +14.6 | 13.0 | 16.1 | 65,471 |
| 2024 | Bachelor's degree minus less than high school | +40.5 | 36.5 | 44.6 | +34.5 | 30.8 | 38.2 | 65,471 |
| 2024 | Bachelor's degree minus high school graduate | +24.3 | 22.7 | 26.0 | +20.8 | 19.1 | 22.4 | 65,471 |
| 2024 | Highest minus lowest third of family income | +20.3 | 18.6 | 22.0 | +13.6 | 12.3 | 14.9 | 65,471 |
| 2020 | Ages 65 and older minus ages 25 to 44 | +11.9 | 10.1 | 13.7 | +15.5 | 13.8 | 17.1 | 73,633 |
| 2020 | White minus Black | +7.4 | 5.2 | 9.7 | +0.9 | −1.0 | 2.9 | 73,633 |
| 2020 | White minus Hispanic | +15.9 | 14.4 | 17.5 | +7.6 | 6.0 | 9.1 | 73,633 |
| 2020 | White minus Asian | +12.5 | 9.9 | 15.2 | +14.1 | 11.8 | 16.4 | 73,633 |
| 2020 | Bachelor's degree minus less than high school | +36.3 | 33.6 | 38.9 | +30.8 | 27.8 | 33.8 | 73,633 |
| 2020 | Bachelor's degree minus high school graduate | +21.9 | 20.1 | 23.8 | +18.7 | 16.9 | 20.5 | 73,633 |
| 2020 | Highest minus lowest third of family income | +20.1 | 18.7 | 21.5 | +13.2 | 12.0 | 14.4 | 73,633 |
| 2016 | Ages 65 and older minus ages 25 to 44 | +14.4 | 12.9 | 16.0 | +19.2 | 18.0 | 20.5 | 83,819 |
| 2016 | White minus Black | +5.0 | 2.5 | 7.6 | −2.6 | −5.2 | −0.1 | 83,819 |
| 2016 | White minus Hispanic | +16.7 | 13.6 | 19.8 | +7.2 | 4.9 | 9.5 | 83,819 |
| 2016 | White minus Asian | +17.1 | 13.9 | 20.2 | +18.1 | 15.7 | 20.4 | 83,819 |
| 2016 | Bachelor's degree minus less than high school | +39.8 | 37.4 | 42.2 | +35.0 | 32.2 | 37.9 | 83,819 |
| 2016 | Bachelor's degree minus high school graduate | +23.3 | 21.9 | 24.6 | +20.6 | 19.3 | 21.8 | 83,819 |
| 2016 | Highest minus lowest third of family income | +21.6 | 20.5 | 22.7 | +13.9 | 12.9 | 14.8 | 83,819 |
| 2022 | Ages 65 and older minus ages 25 to 44 | +22.6 | 20.8 | 24.4 | +26.9 | 25.2 | 28.6 | 66,012 |
| 2022 | White minus Black | +11.6 | 9.1 | 14.0 | +3.9 | 1.3 | 6.6 | 66,012 |
| 2022 | White minus Hispanic | +18.7 | 16.3 | 21.1 | +8.8 | 6.2 | 11.5 | 66,012 |
| 2022 | White minus Asian | +17.9 | 14.7 | 21.1 | +19.4 | 17.0 | 21.7 | 66,012 |
| 2022 | Bachelor's degree minus less than high school | +38.4 | 35.4 | 41.5 | +34.1 | 31.1 | 37.0 | 66,012 |
| 2022 | Bachelor's degree minus high school graduate | +23.8 | 21.9 | 25.7 | +21.0 | 19.3 | 22.7 | 66,012 |
| 2022 | Highest minus lowest third of family income | +19.9 | 18.2 | 21.7 | +13.9 | 12.6 | 15.2 | 66,012 |
In 2024 the White minus Black difference fell from 9.7 (95% CI 7.6 to 11.8)9.7 percentage pointsEstimanddifference in turnout from a weighted linear probability model with this characteristic alone: White minus BlackInterval95% CI 7.6 to 11.8n65,471WeightsVOSUPPWTPopulationcitizens 25 and olderPeriod2024 presidential electionSourceCPS Voting and Registration SupplementNoteInterval from standard errors clustered by state (51 clusters), a normal approximation. The size is the number of persons; the models use more rows because a person whose income bracket straddles a cut-off enters as two rows with fractional weights. points to 3.4 (95% CI 1.6 to 5.3)3.4 percentage pointsEstimanddifference in turnout from a weighted linear probability model with age, race and ethnicity, education and family income together: White minus BlackInterval95% CI 1.6 to 5.3n65,471WeightsVOSUPPWTPopulationcitizens 25 and olderPeriod2024 presidential electionSourceCPS Voting and Registration SupplementNoteInterval from standard errors clustered by state (51 clusters), a normal approximation. The size is the number of persons., and the White minus Hispanic difference from 18.8 (95% CI 16.5 to 21.2)18.8 percentage pointsEstimanddifference in turnout from a weighted linear probability model with this characteristic alone: White minus HispanicInterval95% CI 16.5 to 21.2n65,471WeightsVOSUPPWTPopulationcitizens 25 and olderPeriod2024 presidential electionSourceCPS Voting and Registration SupplementNoteInterval from standard errors clustered by state (51 clusters), a normal approximation. The size is the number of persons; the models use more rows because a person whose income bracket straddles a cut-off enters as two rows with fractional weights. to 10.5 (95% CI 8.1 to 12.9)10.5 percentage pointsEstimanddifference in turnout from a weighted linear probability model with age, race and ethnicity, education and family income together: White minus HispanicInterval95% CI 8.1 to 12.9n65,471WeightsVOSUPPWTPopulationcitizens 25 and olderPeriod2024 presidential electionSourceCPS Voting and Registration SupplementNoteInterval from standard errors clustered by state (51 clusters), a normal approximation. The size is the number of persons.. In 2020 the adjusted White minus Black difference was 0.9 (95% CI −1.0 to 2.9)0.9 percentage pointsEstimanddifference in turnout from a weighted linear probability model with age, race and ethnicity, education and family income together: White minus BlackInterval95% CI −1.0 to 2.9n73,633WeightsVOSUPPWTPopulationcitizens 25 and olderPeriod2020 presidential electionSourceCPS Voting and Registration SupplementNoteInterval from standard errors clustered by state (51 clusters), a normal approximation. The size is the number of persons. points, an interval that includes zero, and in 2016 it was −2.6 (95% CI −5.2 to −0.1)−2.6 percentage pointsEstimanddifference in turnout from a weighted linear probability model with age, race and ethnicity, education and family income together: White minus BlackInterval95% CI −5.2 to −0.1n83,819WeightsVOSUPPWTPopulationcitizens 25 and olderPeriod2016 presidential electionSourceCPS Voting and Registration SupplementNoteInterval from standard errors clustered by state (51 clusters), a normal approximation. The size is the number of persons., with Black citizens above White citizens in the point estimate. The education and income differences shrank proportionally less: the bachelor's degree minus no diploma difference went from 40.5 (95% CI 36.5 to 44.6)40.5 percentage pointsEstimanddifference in turnout from a weighted linear probability model with this characteristic alone: Bachelor's degree minus less than high schoolInterval95% CI 36.5 to 44.6n65,471WeightsVOSUPPWTPopulationcitizens 25 and olderPeriod2024 presidential electionSourceCPS Voting and Registration SupplementNoteInterval from standard errors clustered by state (51 clusters), a normal approximation. The size is the number of persons; the models use more rows because a person whose income bracket straddles a cut-off enters as two rows with fractional weights. points to 34.5 (95% CI 30.8 to 38.2)34.5 percentage pointsEstimanddifference in turnout from a weighted linear probability model with age, race and ethnicity, education and family income together: Bachelor's degree minus less than high schoolInterval95% CI 30.8 to 38.2n65,471WeightsVOSUPPWTPopulationcitizens 25 and olderPeriod2024 presidential electionSourceCPS Voting and Registration SupplementNoteInterval from standard errors clustered by state (51 clusters), a normal approximation. The size is the number of persons., and the highest minus lowest third of income from 20.3 (95% CI 18.6 to 22.0)20.3 percentage pointsEstimanddifference in turnout from a weighted linear probability model with this characteristic alone: Highest minus lowest third of family incomeInterval95% CI 18.6 to 22.0n65,471WeightsVOSUPPWTPopulationcitizens 25 and olderPeriod2024 presidential electionSourceCPS Voting and Registration SupplementNoteInterval from standard errors clustered by state (51 clusters), a normal approximation. The size is the number of persons; the models use more rows because a person whose income bracket straddles a cut-off enters as two rows with fractional weights. to 13.6 (95% CI 12.3 to 14.9)13.6 percentage pointsEstimanddifference in turnout from a weighted linear probability model with age, race and ethnicity, education and family income together: Highest minus lowest third of family incomeInterval95% CI 12.3 to 14.9n65,471WeightsVOSUPPWTPopulationcitizens 25 and olderPeriod2024 presidential electionSourceCPS Voting and Registration SupplementNoteInterval from standard errors clustered by state (51 clusters), a normal approximation. The size is the number of persons.. The White minus Asian difference was 13.1 (95% CI 11.4 to 14.8)13.1 percentage pointsEstimanddifference in turnout from a weighted linear probability model with this characteristic alone: White minus AsianInterval95% CI 11.4 to 14.8n65,471WeightsVOSUPPWTPopulationcitizens 25 and olderPeriod2024 presidential electionSourceCPS Voting and Registration SupplementNoteInterval from standard errors clustered by state (51 clusters), a normal approximation. The size is the number of persons; the models use more rows because a person whose income bracket straddles a cut-off enters as two rows with fractional weights. alone and 14.6 (95% CI 13.0 to 16.1)14.6 percentage pointsEstimanddifference in turnout from a weighted linear probability model with age, race and ethnicity, education and family income together: White minus AsianInterval95% CI 13.0 to 16.1n65,471WeightsVOSUPPWTPopulationcitizens 25 and olderPeriod2024 presidential electionSourceCPS Voting and Registration SupplementNoteInterval from standard errors clustered by state (51 clusters), a normal approximation. The size is the number of persons. with the other characteristics, and the difference between citizens sixty-five and older and those aged twenty-five to forty-four grew from 14.5 (95% CI 13.2 to 15.7)14.5 percentage pointsEstimanddifference in turnout from a weighted linear probability model with this characteristic alone: Ages 65 and older minus ages 25 to 44Interval95% CI 13.2 to 15.7n65,471WeightsVOSUPPWTPopulationcitizens 25 and olderPeriod2024 presidential electionSourceCPS Voting and Registration SupplementNoteInterval from standard errors clustered by state (51 clusters), a normal approximation. The size is the number of persons; the models use more rows because a person whose income bracket straddles a cut-off enters as two rows with fractional weights. to 18.2 (95% CI 17.0 to 19.5)18.2 percentage pointsEstimanddifference in turnout from a weighted linear probability model with age, race and ethnicity, education and family income together: Ages 65 and older minus ages 25 to 44Interval95% CI 17.0 to 19.5n65,471WeightsVOSUPPWTPopulationcitizens 25 and olderPeriod2024 presidential electionSourceCPS Voting and Registration SupplementNoteInterval from standard errors clustered by state (51 clusters), a normal approximation. The size is the number of persons.. These are associations in a descriptive model, and they do not show that education or income changes anyone's probability of voting.
By 2020 and 2024 a majority of voters used mail or voted in person before Election Day
The supplement asks every person who reports voting how they voted, in person or by mail, from 1996, and from 2004 whether they voted on Election Day or before it. Figure 8 plots the shares. In 2004, 79.0%79.0% percentEstimandweighted share of all persons who reported voting, by method (in person on election day)Interval95% CI 78.7% to 79.4%n63,052WeightsVOSUPPWTPopulationpersons who votedPeriod2004 presidential electionSourceCPS Voting and Registration Supplement of voters voted in person on Election Day, 12.9%12.9% percentEstimandweighted share of all persons who reported voting, by method (by mail)Interval95% CI 12.6% to 13.2%n63,052WeightsVOSUPPWTPopulationpersons who votedPeriod2004 presidential electionSourceCPS Voting and Registration Supplement by mail and 7.8%7.8% percentEstimandweighted share of all persons who reported voting, by method (in person before election day)Interval95% CI 7.5% to 8.0%n63,052WeightsVOSUPPWTPopulationpersons who votedPeriod2004 presidential electionSourceCPS Voting and Registration Supplement in person before Election Day. By 2016 the shares were 59.5%59.5% percentEstimandweighted share of all persons who reported voting, by method (in person on election day)Interval95% CI 59.1% to 60.0%n58,533WeightsVOSUPPWTPopulationpersons who votedPeriod2016 presidential electionSourceCPS Voting and Registration Supplement, 20.9%20.9% percentEstimandweighted share of all persons who reported voting, by method (by mail)Interval95% CI 20.5% to 21.3%n58,533WeightsVOSUPPWTPopulationpersons who votedPeriod2016 presidential electionSourceCPS Voting and Registration Supplement and 18.9%18.9% percentEstimandweighted share of all persons who reported voting, by method (in person before election day)Interval95% CI 18.6% to 19.3%n58,533WeightsVOSUPPWTPopulationpersons who votedPeriod2016 presidential electionSourceCPS Voting and Registration Supplement. In 2020 only 30.4%30.4% percentEstimandweighted share of all persons who reported voting, by method (in person on election day)Interval95% CI 29.9% to 30.8%n55,170WeightsVOSUPPWTPopulationpersons who votedPeriod2020 presidential electionSourceCPS Voting and Registration Supplement voted in person on Election Day, while 43.0%43.0% percentEstimandweighted share of all persons who reported voting, by method (by mail)Interval95% CI 42.5% to 43.4%n55,170WeightsVOSUPPWTPopulationpersons who votedPeriod2020 presidential electionSourceCPS Voting and Registration Supplement voted by mail and 26.0%26.0% percentEstimandweighted share of all persons who reported voting, by method (in person before election day)Interval95% CI 25.6% to 26.4%n55,170WeightsVOSUPPWTPopulationpersons who votedPeriod2020 presidential electionSourceCPS Voting and Registration Supplement in person before Election Day, and mail or early in-person voting together accounted for 69.0%69.0% percentEstimandweighted share of all persons who reported voting, by mail or in person before Election DayInterval95% CI 68.5% to 69.4%n55,170WeightsVOSUPPWTPopulationpersons who votedPeriod2020 presidential electionSourceCPS Voting and Registration Supplement. In 2024 the shares were 39.6%39.6% percentEstimandweighted share of all persons who reported voting, by method (in person on election day)Interval95% CI 39.1% to 40.1%n48,387WeightsVOSUPPWTPopulationpersons who votedPeriod2024 presidential electionSourceCPS Voting and Registration Supplement, 29.0%29.0% percentEstimandweighted share of all persons who reported voting, by method (by mail)Interval95% CI 28.5% to 29.4%n48,387WeightsVOSUPPWTPopulationpersons who votedPeriod2024 presidential electionSourceCPS Voting and Registration Supplement and 30.7%30.7% percentEstimandweighted share of all persons who reported voting, by method (in person before election day)Interval95% CI 30.3% to 31.2%n48,387WeightsVOSUPPWTPopulationpersons who votedPeriod2024 presidential electionSourceCPS Voting and Registration Supplement, and mail or early in-person voting together accounted for 59.7%59.7% percentEstimandweighted share of all persons who reported voting, by mail or in person before Election DayInterval95% CI 59.2% to 60.2%n48,387WeightsVOSUPPWTPopulationpersons who votedPeriod2024 presidential electionSourceCPS Voting and Registration Supplement.
Figure 8
Voting by mail rose to 43.0% of voters in 2020 and fell to 29.0% in 2024, while voting in person before Election Day reached 30.7%
Percent of persons who reported voting, by method and timing, presidential years 1996 to 2024 (the second button shows midterm years); 95% intervals from the Kish effective sample size
Show data
| Year | Election type | Method | Share of voters (%) | 95% CI low | 95% CI high | n respondents |
|---|---|---|---|---|---|---|
| 2004 | Presidential | In person on Election Day | 79.0 | 78.7 | 79.4 | 63,052 |
| 2008 | Presidential | In person on Election Day | 68.9 | 68.5 | 69.3 | 59,975 |
| 2012 | Presidential | In person on Election Day | 66.8 | 66.4 | 67.2 | 59,597 |
| 2016 | Presidential | In person on Election Day | 59.5 | 59.1 | 60.0 | 58,533 |
| 2020 | Presidential | In person on Election Day | 30.4 | 29.9 | 30.8 | 55,170 |
| 2024 | Presidential | In person on Election Day | 39.6 | 39.1 | 40.1 | 48,387 |
| 2004 | Presidential | In person before Election Day | 7.8 | 7.5 | 8.0 | 63,052 |
| 2008 | Presidential | In person before Election Day | 14.2 | 13.9 | 14.6 | 59,975 |
| 2012 | Presidential | In person before Election Day | 14.2 | 13.9 | 14.5 | 59,597 |
| 2016 | Presidential | In person before Election Day | 18.9 | 18.6 | 19.3 | 58,533 |
| 2020 | Presidential | In person before Election Day | 26.0 | 25.6 | 26.4 | 55,170 |
| 2024 | Presidential | In person before Election Day | 30.7 | 30.3 | 31.2 | 48,387 |
| 1996 | Presidential | By mail | 7.8 | 7.6 | 8.1 | 50,025 |
| 2000 | Presidential | By mail | 10.1 | 9.8 | 10.4 | 50,010 |
| 2004 | Presidential | By mail | 12.9 | 12.6 | 13.2 | 63,052 |
| 2008 | Presidential | By mail | 16.4 | 16.0 | 16.7 | 59,975 |
| 2012 | Presidential | By mail | 18.5 | 18.1 | 18.8 | 59,597 |
| 2016 | Presidential | By mail | 20.9 | 20.5 | 21.3 | 58,533 |
| 2020 | Presidential | By mail | 43.0 | 42.5 | 43.4 | 55,170 |
| 2024 | Presidential | By mail | 29.0 | 28.5 | 29.4 | 48,387 |
| 2006 | Midterm | In person on Election Day | 80.1 | 79.7 | 80.5 | 47,175 |
| 2010 | Midterm | In person on Election Day | 73.0 | 72.5 | 73.5 | 44,802 |
| 2014 | Midterm | In person on Election Day | 68.4 | 67.9 | 68.9 | 42,559 |
| 2018 | Midterm | In person on Election Day | 59.6 | 59.1 | 60.1 | 48,337 |
| 2022 | Midterm | In person on Election Day | 49.5 | 48.9 | 50.1 | 39,280 |
| 2006 | Midterm | In person before Election Day | 5.8 | 5.6 | 6.0 | 47,175 |
| 2010 | Midterm | In person before Election Day | 8.3 | 8.0 | 8.6 | 44,802 |
| 2014 | Midterm | In person before Election Day | 10.2 | 9.9 | 10.5 | 42,559 |
| 2018 | Midterm | In person before Election Day | 16.6 | 16.2 | 17.0 | 48,337 |
| 2022 | Midterm | In person before Election Day | 18.0 | 17.6 | 18.5 | 39,280 |
| 1998 | Midterm | By mail | 8.4 | 8.1 | 8.7 | 39,512 |
| 2002 | Midterm | By mail | 10.6 | 10.3 | 10.9 | 47,897 |
| 2006 | Midterm | By mail | 13.7 | 13.4 | 14.1 | 47,175 |
| 2010 | Midterm | By mail | 18.1 | 17.7 | 18.5 | 44,802 |
| 2014 | Midterm | By mail | 20.8 | 20.4 | 21.3 | 42,559 |
| 2018 | Midterm | By mail | 23.1 | 22.7 | 23.6 | 48,337 |
| 2022 | Midterm | By mail | 31.8 | 31.2 | 32.3 | 39,280 |
From 2020 to 2024 the mail share fell from 43.0%43.0% percentEstimandweighted share of all persons who reported voting, by method (by mail)Interval95% CI 42.5% to 43.4%n55,170WeightsVOSUPPWTPopulationpersons who votedPeriod2020 presidential electionSourceCPS Voting and Registration Supplement to 29.0%29.0% percentEstimandweighted share of all persons who reported voting, by method (by mail)Interval95% CI 28.5% to 29.4%n48,387WeightsVOSUPPWTPopulationpersons who votedPeriod2024 presidential electionSourceCPS Voting and Registration Supplement, the early in-person share rose from 26.0%26.0% percentEstimandweighted share of all persons who reported voting, by method (in person before election day)Interval95% CI 25.6% to 26.4%n55,170WeightsVOSUPPWTPopulationpersons who votedPeriod2020 presidential electionSourceCPS Voting and Registration Supplement to 30.7%30.7% percentEstimandweighted share of all persons who reported voting, by method (in person before election day)Interval95% CI 30.3% to 31.2%n48,387WeightsVOSUPPWTPopulationpersons who votedPeriod2024 presidential electionSourceCPS Voting and Registration Supplement and the Election Day share rose from 30.4%30.4% percentEstimandweighted share of all persons who reported voting, by method (in person on election day)Interval95% CI 29.9% to 30.8%n55,170WeightsVOSUPPWTPopulationpersons who votedPeriod2020 presidential electionSourceCPS Voting and Registration Supplement to 39.6%39.6% percentEstimandweighted share of all persons who reported voting, by method (in person on election day)Interval95% CI 39.1% to 40.1%n48,387WeightsVOSUPPWTPopulationpersons who votedPeriod2024 presidential electionSourceCPS Voting and Registration Supplement. Election Day voting in 2024 was still below its 2016 level of 59.5%59.5% percentEstimandweighted share of all persons who reported voting, by method (in person on election day)Interval95% CI 59.1% to 60.0%n58,533WeightsVOSUPPWTPopulationpersons who votedPeriod2016 presidential electionSourceCPS Voting and Registration Supplement. In the 2022 midterm 31.8%31.8% percentEstimandweighted share of all persons who reported voting, by method (by mail)Interval95% CI 31.2% to 32.3%n39,280WeightsVOSUPPWTPopulationpersons who votedPeriod2022 midterm electionSourceCPS Voting and Registration Supplement of voters used the mail, against 23.1%23.1% percentEstimandweighted share of all persons who reported voting, by method (by mail)Interval95% CI 22.7% to 23.6%n48,337WeightsVOSUPPWTPopulationpersons who votedPeriod2018 midterm electionSourceCPS Voting and Registration Supplement in 2018, and 18.0%18.0% percentEstimandweighted share of all persons who reported voting, by method (in person before election day)Interval95% CI 17.6% to 18.5%n39,280WeightsVOSUPPWTPopulationpersons who votedPeriod2022 midterm electionSourceCPS Voting and Registration Supplement voted in person before Election Day. Older voters used the mail more: 53.6%53.6% percentEstimandweighted share of all persons who reported voting, by mailInterval95% CI 52.7% to 54.5%n15,856WeightsVOSUPPWTPopulationpersons aged 65 and older who votedPeriod2020 presidential electionSourceCPS Voting and Registration Supplement of voters aged sixty-five and older voted by mail in 2020 against 39.9%39.9% percentEstimandweighted share of all persons who reported voting, by mailInterval95% CI 38.2% to 41.6%n4,190WeightsVOSUPPWTPopulationpersons aged 18 to 24 who votedPeriod2020 presidential electionSourceCPS Voting and Registration Supplement of those aged eighteen to twenty-four, and 36.8%36.8% percentEstimandweighted share of all persons who reported voting, by mailInterval95% CI 35.9% to 37.6%n15,714WeightsVOSUPPWTPopulationpersons aged 65 and older who votedPeriod2024 presidential electionSourceCPS Voting and Registration Supplement against 26.3%26.3% percentEstimandweighted share of all persons who reported voting, by mailInterval95% CI 24.7% to 28.0%n3,419WeightsVOSUPPWTPopulationpersons aged 18 to 24 who votedPeriod2024 presidential electionSourceCPS Voting and Registration Supplement in 2024.
The Census Bureau's report on 2020 describes the rise as a response to changes in election procedures during the pandemic [6]. The Election Assistance Commission's tabulation of administrative records agrees with the survey on the mail and Election Day shares for 2020 [7], but the Commission publishes higher early in-person shares than the survey in both 2020 and 2024 [8]. Its percentages are computed state by state and sum to more than the whole, and its counts of early in-person ballots agree closely with the survey's weighted counts and show the same rise from 2020 to 2024, so the survey's split between early and Election Day voting in 2024 should be read as approximate (the methods note gives the figures). The checks on these items are a tabulation of codes and universes by year, which shows that only persons who reported voting have an answer in every supplement, and a match of the 2018 to 2024 shares to the Census Bureau's published tables.
The CPS count of reported voters exceeded the ballots counted through 2000 and was close to them from 2004
Surveys of reported turnout are known to run above official turnout, and over-reporting is more common among some groups than others [9]. The rates here are not benchmarked to the voting-eligible population, the usual denominator for official turnout, because that series was not available. Figure 9 compares counts and sets the weighted number of adults who reported voting against the ballots counted for president in each presidential year.
Figure 9
The CPS count of reported voters exceeded the ballots counted for president by 6.0% to 11.6% through 2000 and was within 3% of them from 2004 to 2024
Millions of persons: ballots counted for president against the CPS weighted number of adults 18 and older who reported voting, presidential years 1976 to 2024; the CPS interval is the 95% interval from the Kish effective sample size
- Ballots counted for president
- CPS reported voters
Show data
| Election | Ballots counted for president (millions) | CPS reported voters (millions) | 95% CI low (millions) | 95% CI high (millions) | n respondents |
|---|---|---|---|---|---|
| 1976 | 81.6 | 86.7 | 86.2 | 87.2 | 89,090 |
| 1980 | 86.5 | 94.8 | 94.3 | 95.3 | 123,591 |
| 1984 | 92.7 | 101.9 | 101.3 | 102.4 | 110,115 |
| 1988 | 91.6 | 102.2 | 101.6 | 102.9 | 102,945 |
| 1992 | 104.6 | 113.4 | 112.8 | 114.0 | 105,987 |
| 1996 | 96.4 | 105.0 | 104.3 | 105.7 | 89,609 |
| 2000 | 105.6 | 111.9 | 111.2 | 112.7 | 88,884 |
| 2004 | 122.3 | 125.7 | 125.0 | 126.5 | 103,696 |
| 2008 | 131.4 | 131.1 | 130.4 | 131.9 | 99,556 |
| 2012 | 129.1 | 132.9 | 132.1 | 133.8 | 101,568 |
| 2016 | 136.8 | 137.5 | 136.7 | 138.4 | 100,946 |
| 2020 | 158.5 | 154.6 | 153.7 | 155.6 | 87,589 |
| 2024 | 155.4 | 154.3 | 153.3 | 155.3 | 78,650 |
Relative to the count of ballots for president, the CPS count was 6.0%6.0% percent of ballots counted for presidentEstimandsmallest excess of the CPS count over the certified presidential totalWeightsVOSUPPWTPopulationvoters in the CPS estimate, all residents 18 and olderPeriod1976 to 2000 presidential electionsSourceCPS Voting and Registration Supplement to 11.6%11.6% percent of ballots counted for presidentEstimandlargest excess of the CPS count over the certified presidential totalWeightsVOSUPPWTPopulationvoters in the CPS estimate, all residents 18 and olderPeriod1976 to 2000 presidential electionsSourceCPS Voting and Registration Supplement higher in the seven elections from 1976 to 2000. It was +2.8% (95% CI 2.2% to 3.4%)+2.8% percent of ballots counted for presidentEstimandCPS weighted count of persons who reported voting, relative to the certified presidential totalInterval95% CI 2.2% to 3.4%n103,696WeightsVOSUPPWTPopulationvoters in the CPS estimate, all residents 18 and olderPeriod2004SourceCPS Voting and Registration SupplementNoteThe interval treats the adult population total as fixed and uses the Kish interval on the share who voted. in 2004 and within 3%3% percent of ballots counted for presidentEstimandlargest absolute difference between the CPS count of reported voters and the certified presidential totalWeightsVOSUPPWTPopulationvoters in the CPS estimate, all residents 18 and olderPeriod2004 to 2024 presidential electionsSourceCPS Voting and Registration Supplement of the count in every election from 2004 to 2024, with −2.5% (95% CI −3.0% to −1.9%)−2.5% percent of ballots counted for presidentEstimandCPS weighted count of persons who reported voting, relative to the certified presidential totalInterval95% CI −3.0% to −1.9%n87,589WeightsVOSUPPWTPopulationvoters in the CPS estimate, all residents 18 and olderPeriod2020SourceCPS Voting and Registration SupplementNoteThe interval treats the adult population total as fixed and uses the Kish interval on the share who voted. in 2020, below the count. Two things push in opposite directions. Reporting voting when one did not raises the count. Counting a nonrespondent as not voting lowers it, and nonresponse rose from 10.8%10.8% percentEstimandweighted share who did not answer the voting question (refused, did not know, no response)n96,452WeightsVOSUPPWTPopulationcitizens 18 and olderPeriod2004SourceCPS Voting and Registration Supplement of citizens in 2004 to 13.8%13.8% percentEstimandweighted share who did not answer the voting question (refused, did not know, no response)n92,360WeightsVOSUPPWTPopulationcitizens 18 and olderPeriod2008SourceCPS Voting and Registration Supplement in 2008 and 15.7%15.7% percentEstimandweighted share who did not answer the voting question (refused, did not know, no response)n81,898WeightsVOSUPPWTPopulationcitizens 18 and olderPeriod2020SourceCPS Voting and Registration Supplement in 2020. That rise is consistent with the narrowing of the count, but the data do not test it. Dropping the nonrespondents gives turnout of 79.2%79.2% percentEstimandweighted share who reported voting; nonresponse excluded (self-report convention)WeightsVOSUPPWTPopulationcitizens 18 and older who answered the voting questionPeriod2020SourceCPS Voting and Registration Supplement in 2020 against 66.8%66.8% percentEstimandweighted share who reported voting (Census convention: nonresponse counted as not voting)Interval95% CI 66.4% to 67.1%n81,898WeightsVOSUPPWTPopulationcitizens 18 and olderPeriod2020 presidential electionSourceCPS Voting and Registration Supplement under the Census convention, and 76.1%76.1% percentEstimandweighted share who reported voting; nonresponse excluded (self-report convention)WeightsVOSUPPWTPopulationcitizens 18 and older who answered the voting questionPeriod2024SourceCPS Voting and Registration Supplement in 2024 against 65.3%65.3% percentEstimandweighted share who reported voting (Census convention: nonresponse counted as not voting)Interval95% CI 64.9% to 65.7%n72,590WeightsVOSUPPWTPopulationcitizens 18 and olderPeriod2024 presidential electionSourceCPS Voting and Registration Supplement. The ANES, a different survey of a different sample, gives 77.3% (95% CI 75.9% to 78.6%)77.3% percentEstimandweighted share who reported votingInterval95% CI 75.9% to 78.6%n7,446Weightsweight_postelection (VCF9999)PopulationANES respondents who answered the turnout item and have a post-election weightPeriod2020SourceANES Cumulative Data File for 2020 and 78.9% (95% CI 77.3% to 80.5%)78.9% percentEstimandweighted share who reported votingInterval95% CI 77.3% to 80.5%n4,954Weightsweight_postelection (VCF9999)PopulationANES respondents who answered the turnout item and have a post-election weightPeriod2024SourceANES Cumulative Data File for 2024 with its post-election weight. Both are above the CPS rate under the Census convention, and the ANES is below the respondents-only CPS rate in 2020 and above it in 2024. The count comparison does not test the CPS denominator, the civilian noninstitutional population.
What these data cannot show
The CPS reports what respondents say, and the extract has no validated vote. The validated-vote study cited above found over-reporting more common among more educated respondents and found that using reported instead of verified voting inflates the effect of variables tied to over-reporting [9]. The education gradient and the registered shares of the better educated may therefore be overstated here by an amount the data do not give. Nonresponse is also tied to the characteristics studied, and the Census convention's treatment of it changes some differences: the White minus Black difference in 2024 is smaller among respondents than under the convention.
The supplements are repeated cross-sections. They interview different people each time, so they cannot say whether a group's turnout changed because its members changed their behavior or because its composition changed, for example as the share of citizens with a bachelor's degree grew, and a gap between age groups mixes age, period and birth cohort. Chapter 3.1, The two coalitions takes up who the voters are compared with all eligible adults, a different question from the turnout rates here.
The headline comparison starts in 1994, and citizen turnout was higher earlier on the Census Bureau's own series: its historical table of reported voting gives 67.7%67.7% percentEstimandpublished weighted share who reported voting (Census Bureau Table A-1), nonresponse counted as not votingWeightsVOSUPPWT, as published by the Census BureauPopulationcitizens 18 and older, as the Census Bureau defines them from the supplement's own citizenship itemPeriod1992 presidential electionSourceCPS Voting and Registration SupplementNoteA published figure with no interval; the extract has no citizenship item before 1994 and cannot reproduce it. for 1992, above the 66.8%66.8% percentEstimandweighted share who reported voting (Census convention: nonresponse counted as not voting)Interval95% CI 66.4% to 67.1%n81,898WeightsVOSUPPWTPopulationcitizens 18 and olderPeriod2020 presidential electionSourceCPS Voting and Registration Supplement of 2020 and the 65.3%65.3% percentEstimandweighted share who reported voting (Census convention: nonresponse counted as not voting)Interval95% CI 64.9% to 65.7%n72,590WeightsVOSUPPWTPopulationcitizens 18 and olderPeriod2024 presidential electionSourceCPS Voting and Registration Supplement of 2024, from a citizenship item that the extract lacks. The finding that 2018 to 2024 exceeded earlier elections of their type therefore covers 1994 to 2016 only. Citizens are also not the same as eligible voters. The citizen count includes citizens who cannot vote because of a criminal conviction or residence rules and leaves out citizens in institutions and abroad, and the extract has no measure of either. The CPS identifies roughly three hundred large counties, which supports no county or small-area estimate, and no state estimates are made, because state rates would need the benchmark that was not available here.
The intervals omit the CPS design effect, so every interval is too narrow by an unknown factor, and a result whose interval lies close to zero should be read with that in mind: the change in the gap between a bachelor's degree and a high school diploma from 1996 to 2024, +1.4 (95% CI 0.1 to 2.8)+1.4 percentage pointsEstimandchange in the turnout gap between the two years (difference of two differences), gap = bachelor's degree or more minus high school diploma onlyInterval95% CI 0.1 to 2.8n88,026WeightsVOSUPPWTPopulationcitizens 25 and olderPeriod1996 to 2024SourceCPS Voting and Registration Supplement, is an example. Family income is a householder-family bracket with allocated values from 2010, so it measures neither wealth nor individual earnings, and the thirds before and after 2010 are not on one footing. The race groups are broad, with all Asian citizens in one group and Hispanic citizens of every origin in another. The adjusted differences describe a model with four characteristics and do not identify why any group votes at a different rate. The method and timing items are self-reports. They match the Census Bureau tables, but the administrative tabulation gives higher early in-person shares, so the split between early and Election Day voting should be read as approximate.
Data sources
- CPS Voting and Registration SupplementIPUMS-CPS November Voting and Registration Supplement, 25 biennial supplements 1976-2024; citizens from 1994Years: 1976-2024 (25 November samples, biennial, even years) · Weights: VOSUPPWT (supplement person weight) · 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 1976-2024, used only for the certified national total of ballots counted for presidentYears: 1976-2024 (13 elections) · Weights: none (full counts) · License: See the provider's terms of use.
- ANES Cumulative Data FileCumulative Data File 1976-2024, self-reported turnout in presidential years, as a comparison of levelsYears: 1948-2024 (33 study years; even years only after 2004; no 2006, 2010 or 2014 study) · Weights: weight_postelection (VCF9999) · License: Free download after accepting the ANES terms of use; cite the release.
Methods
Sample: the IPUMS-CPS extract of the Census Bureau's November Voting and Registration Supplement, 25 biennial supplements from 1976 to 2024. The analysis uses adults 18 and older with a positive supplement weight (2,494,218 person records), and from 1994, the first year the extract records citizenship, the citizens among them (1,422,311). Weight: VOSUPPWT, the supplement person weight. Outcome: voted_census, the Census Bureau convention, in which a respondent who refused, did not know or did not respond counts as not voting; from 2022 citizens whom IPUMS codes not in universe on the voting item but who carry a positive weight (the supplement nonrespondents) also count as not voting. The share of citizens who did not answer rose from 6.2% in 1994 to 17.7% in 2018 and was 14.1% in 2024. Two universes: citizens 18 and older, 1994 to 2024 (the headline), and all residents 18 and older, 1976 to 2024, in which every adult outside the voting-question universe counts as not voting, the Census total-population convention. From 1978 to 1992 the voting question was not asked of 3.5% to 6.8% of adults (40.2% of Hispanic adults in 1992, against 1.9% of White non-Hispanic adults), a pattern that matches the noncitizen shares from 1994 (in 1976 the supplement asked every adult but a handful). The extract has no citizenship item before 1994, so it can give no citizen series for 1976 to 1992. Published-figure match: the loader asserts that weighted turnout of citizens equals the published Census figure to one decimal in each of 2008, 2010, 2012, 2014, 2016, 2018, 2020, 2022 and 2024. Further checks against the Census Bureau tables of the same supplement (Table A-1, voting and registration by age, race and ethnicity for all residents and citizens; Table A-2, voting by education; Table 14, method of voting, 2018, 2020, 2022 and 2024): 314 of 314 percentages from 1996 to 2024 (1996 and 1998, then 2002 to 2024) match to the printed decimal, the largest difference being 0.05 points; in 1994 17 of 19 match (the largest difference is 0.4), in 2000 3 of 19 (largest 0.8) and in 1976 to 1992 58 of 115 (largest 0.6); all 176 percentages and all 44 totals of Table 14 match. Registration follows the Census convention: registered if the person voted or said they were registered, and everyone else, including a person who did not answer the voting question and did not report being registered, counts as not registered. Age groups are 18 to 29, 30 to 44, 45 to 64 and 65 and older. Race and ethnicity is the extract's five-group variable: Hispanic of any race, and White, Black, Asian or Pacific Islander, and another or multiple races, all non-Hispanic; respondents could report more than one race from 2003. Education is the extract's four levels for adults 25 and older. Family income is the householder's family income in the brackets of the basic monthly survey, which differ across decades (14 brackets to 2003, 16 from 2004). Income groups are within-year thirds: the brackets are laid out along the weighted distribution of citizens with a bracket, and a bracket that straddles a cut-off is split between the two groups in proportion, so each third holds one third of weighted citizens in every year. Up to 2008 citizens who did not report a bracket (9.1% of citizens in 1994, 11.4% in 1996 and 20.6% in 2008) are outside the thirds; from January 2010 the Census Bureau allocates missing family income (IPUMS notes an allocation rate of about 20 percent), so the thirds from 2010 include allocated values and the series is broken between 2008 and 2010. Intervals: 95%. Shares and differences between independent groups use the Kish effective sample size and a normal approximation, combining standard errors in quadrature; changes in a gap combine four independent estimates. These intervals treat the weighted sample as a simple random sample. The extract has no strata, primary sampling units or replicate weights, so they omit the clustering and stratification of the CPS design and are narrower than design-based intervals by an unknown factor; the generalized variance parameters in the Census Bureau's source-and-accuracy statement were not applied. The adjusted differences use weighted linear probability models fitted separately for each election with standard errors clustered by state (51 clusters, the states and the District of Columbia) and normal critical values; they leave out the within-state design effect. Suppression: a published cell needs at least 100 respondents and a Kish effective sample of at least 100, and 0 were suppressed; the smallest published cell has 985 respondents. Method and timing: VOTEHOW (1996 onward) and VOTEWHEN (2004 onward) are asked only of persons who reported voting (no person who did not vote has an answer in any of the 25 supplements); codes are in person or by mail, and on Election Day or before it, with refused, don't know and no response under one percent of voters. Shares use all persons who voted in the denominator, with don't know or refused as a fourth category, as Table 14 does; in person before Election Day is in person and before, and by mail is by mail whatever the timing. For 2020 the supplement gives 43.0% by mail and 30.4% in person on Election Day, against 43.1% and 30.5% in the Election Assistance Commission's administrative tabulation; for 2024 the supplement gives 30.7% in person before Election Day and 39.6% in person on Election Day against 35.2% and 37.4% there. Benchmarks: ballots counted for president are the sum of the state totals in the MEDSL returns (full counts); the CPS count is the weighted number of adults who reported voting, and its interval takes the adult total as fixed. The voting-eligible-population turnout series was not available, so no benchmark of the turnout rates is applied. ANES self-reported turnout, a post-election item, uses the post-election weight (weight_postelection, VCF9999), and the CES is not used because the universe of its self-reported item changes across years. Survey mode: the extract has no interview-mode variable, so no mode split is made; the CPS is interviewer-administered, by telephone and in person. The ANES 2020 study was collected online, by telephone or by video and has no mode weights, and the 2024 study mixes face-to-face and web samples, so the ANES values are shown only as levels of a different instrument and not as a trend. The Census Bureau's statement that 69.4 percent of 2020 voters used a nontraditional method (mail, or in person before Election Day) uses the voters with a reported method as the denominator, and the same denominator gives 69.4% here. The all-residents series matches the total-population figures of Table A-1 within 0.05 points in every year from 1994 except 2000, where it is 0.27 points below, and sits 0.2 to 0.6 points below them in 1984 to 1992; the 1976 value is on a slightly different footing because the supplement asked every adult that year. Table A-1 also publishes a citizen series from 1978 (64.0% in 1980 and 67.7% in 1992, the latter above the 66.8% of 2020 and the 65.3% of 2024 in this extract); the extract has no citizenship item before 1994 and cannot reproduce it. Registration split: with V the turnout rate, R the registered share and T the share of the registered who voted, V = R x T, and for a higher-turnout group a and a lower-turnout group b, V_a - V_b = (R_a - R_b)(T_a + T_b)/2 + (T_a - T_b)(R_a + R_b)/2 exactly. The first term is the part that arises at registration and the second the part among the registered. The split depends on the order in which R and T are changed: changing R first at the lower group's T, or at the higher group's T, gives the two ends of a range (for education in 2024, 62% to 77%), and the split used in the text is the symmetric one, the midpoint. Its interval is a seeded percentile bootstrap over respondents (500 replicates, seed 12345; for income over the stacked rows) and omits the design effect, so it is too narrow by an unknown factor. The share also depends on how nonresponse is coded: among respondents only (nonrespondents dropped from the registered share as well as from turnout) the White minus Black share in 2024 is 44% against 62% under the Census convention, and the education share is 67% against 70%. Administrative comparison for 2024: the Election Assistance Commission reports early in-person shares of 30.6% for 2020 and 35.2% for 2024 against 26.0% and 30.7% in the supplement, but its percentages are computed state by state with casewise deletion and sum to 104.2% in 2020 and 102.9% in 2024, so they are not comparable with shares that sum to 100%. Counts compare better: the Commission reports 41,415,710 early in-person ballots for 2020 (41,266,229 in its 2020 report) and 48,872,528 for 2024, an increase of 18.0%, and the weighted counts of CPS voters who voted in person before Election Day are 40.2 million and 47.4 million, an increase of 17.9%.
References
- 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
- Wolfinger, Raymond E.; Rosenstone, Steven J. (1980). Who Votes?. Yale University Press.
- Kish, Leslie (1965). Survey Sampling. John Wiley and Sons.
- 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
- Plutzer, Eric (2002). Becoming a Habitual Voter: Inertia, Resources, and Growth in Young Adulthood. American Political Science Review 96(1): 41-56. doi:10.1017/S0003055402004227
- Fabina, Jacob; Scherer, Zach (2022). Voting and Registration in the Election of November 2020. U.S. Census Bureau, Current Population Reports P20-585. www.census.gov/library/publications/2022/demo/p20-585.html
- U.S. Election Assistance Commission (2021). Election Administration and Voting Survey 2020 Comprehensive Report. A Report from the U.S. Election Assistance Commission to the 117th Congress. www.eac.gov/sites/default/files/document_library/files/2020_EAVS_Report_Final_508c.pdf
- U.S. Election Assistance Commission (2025). Election Administration and Voting Survey 2024 Comprehensive Report. A Report from the U.S. Election Assistance Commission to the 119th Congress. www.eac.gov/sites/default/files/2025-07/2024_EAVS_Report_508.pdf
- Bernstein, Robert; Chadha, Anita; Montjoy, Robert (2001). Overreporting Voting: Why It Happens and Why It Matters. Public Opinion Quarterly 65(1): 22-44. doi:10.1086/320036
How to cite
@misc{singh2026turnout,
author = {Singh, Vishal},
title = {Turnout},
year = {2026},
url = {https://vishalsingh.org/politics/turnout},
note = {In The Public Mind. Published 3 October 2026.}
}