Voters bandage their ballots during the 2022 midterm election. (Atilgan Ozdil/Anadolu Agency/Getty Images)

Pew Research Center conducted this study to understand how Americans voted in 2022 and how their turnout and vote choices differed from 2016. For this analysis, we surveyed U.South. adults online and verified their turnout in the two elections using commercial voter files that aggregate official land turnout records.

We surveyed x,640 U.S. adults online in November 2022 and 4,183 adults in November and Dec 2016. Anybody who took part is a fellow member of Pew Research Middle'due south American Trends Panel (ATP), an online survey panel recruited through national, random sampling of phone numbers or, since 2018, residential addresses. This mode most all U.South. adults have a chance of selection. The surveys are weighted to be representative of the U.Southward. adult population past gender, race, ethnicity, partisan affiliation, education and many other characteristics. Read more most the ATP's methodology. Verification of voter turnout involved matching the panelists to 2 or more commercial voter files. Panelists for whom a tape of voting was located are considered validated voters; all others are presumed non to have voted.

Here are the questions used for this report and its methodology.

Compared with Hillary Clinton's 2-point popular vote advantage over Donald Trump in the 2022 presidential election, the Democratic Political party expanded its margin over the Republican Political party to ix points in votes cast for the U.S. House of Representatives in 2018, a proceeds of vii per centum points. This increased support was sufficient for the Democratic Political party to gain the bulk in the House with a net pickup of 41 seats. Voter turnout as a share of the eligible population was 49%, the highest for a midterm election in 100 years. A new analysis of verified voters from Pew Research Center's American Trends Console examines what 2022 voters and nonvoters did in the 2022 midterm elections and offers a detailed portrait of the demographic composition and vote choices of the 2022 electorate. It provides an update and comparing with findings from our study of the 2022 electorate.

Compared with how Clinton fared in 2016, Democratic candidates for Congress in 2022 made gains from several sources. Among Americans who voted in both elections, Clinton's 2022 voters supported Democrats in 2022 at a slightly higher charge per unit than Trump's voters supported Republican candidates. Slightly more of Clinton'southward than Trump's voters turned out to vote in 2018. In combination, party loyalty, defection and turnout differences among 2022 voters accounted for a niggling less than half of the Democratic gains over Clinton's ii-point margin.

Nonvoters in 2022 who turned out in 2022 voted heavily for Democratic candidates, accounting for about half of the Democratic gains. Additionally, a small share of the gains came from people who voted for 3rd-party candidates in 2016; they favored Democratic candidates over Republican candidates in 2022 by a narrow margin.

Voting patterns in 2022 reflected a great deal of continuity with 2016, though Autonomous candidates in 2022 did amend among a few groups, notably men, young people and secular voters. Voting patterns among several other big groups changed less, including Black voters, voters ages 65 and older, Protestants, regular churchgoers and women.

Given their relatively lower turnout, midterm elections are not necessarily predictive of what will happen in the side by side presidential election, when many more American voters will take role.

This analysis is based on interviews with 10,640 members of Pew Research Eye's American Trends Panel conducted Nov. 7-sixteen, 2018, before long after the general election. It likewise draws on interviews conducted amidst 3,770 of the panelists from Nov. 29 to Dec. 12, 2016, after the general election that year and interviews conducted Aug. 20 to Oct. 28, 2022 among all members of the panel at that time. Researchers attempted to match the panelists to two dissimilar commercial voter files that incorporate official records of voter registration and turnout for 2022 and 2018. For the panelists interviewed in 2016, their 2022 vote history is based on verification with three additional commercial voter files, as described in an earlier report. (For more details, see "Methodology.") This process of verifying voter turnout helps to right for the trend of some people to overreport voting and is mostly regarded as providing a more than accurate picture of the electorate.

Where the 2022 Democratic advantage came from: 2022 nonvoters, higher turnout past Clinton voters, and vote switching

Democratic House candidates gained from 2022 nonvoters and third-party votersMidterm elections consistently experience lower turnout than presidential elections. Yet while the 2022 turnout of 49% did non match turnout in the 2022 presidential election (59%), it was far higher than usual. Midway through President Trump's first term in office, both Democrats and Republicans were energized. A large majority of people who voted in 2022 (76%) also voted in 2018. But somewhat more of Clinton's 2022 voters (78%) than Trump's 2022 voters (74%) turned out in 2018. Overwhelming majorities of both Trump'due south and Clinton'south 2022 voters remained loyal to their respective parties in their 2022 U.S. Business firm vote, though Clinton'due south 2022 voters who turned out in 2022 were slightly more than loyal to Democratic 2022 candidates (96%) than Trump's 2022 voters were to 2022 GOP candidates (93%). Amongst the share who voted for someone other than Trump or Clinton in 2016, 71% voted in 2018. These voters favored Democratic candidates over Republican candidates by a margin of 49% to 37%.

Voters in 2022 who did not vote in 2022 were a modest group (about 11% of all 2022 voters) but an important part of why the Democratic Party made gains. Among the 2022 nonvoters who voted in 2018, Democratic House candidates led Republican House candidates past a more than than a two-to-1 (68% to 29%) margin.

Of everyone eligible by citizenship and historic period to vote in 2018, 44% voted in both the 2022 and 2022 elections; 36% voted in neither; 14% were drop-off voters (voting in 2022 merely non in 2018) and a small share (half-dozen%) were new voters – voting in 2022 but non in 2016.

Few defections from party amalgamation

Equally they did in 2016, Republicans and Democrats voted almost unanimously for House candidates of their own party in 2018. Among those who practise not initially identify with either party (including leaners, members of third parties and "pure" independents), Democratic candidates picked up 13 percent points of support in 2022 over Clinton'due south levels. Autonomous candidates also fabricated gains among Republicans and leaners who describe themselves as moderate or liberal (from viii% for Clinton to 15% for Autonomous House candidates).

Party loyalty remained strong in 2022 midterm vote

Democrats did meliorate in 2022 than 2022 among men, young voters

Among well-nigh groups, voting patterns in 2022 were generally similar to those in 2016, albeit with most reflecting somewhat greater support for Autonomous candidates for the U.S. Firm compared with Hillary Clinton. Men, young people and secular voters were notably more than supportive of Autonomous candidates in 2022 than these groups had been in 2016.

Democratic gains among men resulted in some narrowing of the gender gap. In the 2022 election, Donald Trump won men by eleven points (52% to 41%) and Hillary Clinton won women by 15 (54% to 39%), for a deviation of 26 points. In 2018, women supported Democratic candidates by a like margin (18 points, 58% to xl%) but the GOP advantage among men vanished (fifty% voted Autonomous, 48% Republican). Trump carried White men by 30 points in 2022 (62% to 32%), a Republican reward that shrank to merely 12 points in 2022 (55% to 43%).

Democrats fared much better among men in 2022 than in 2016, narrowing the gender gap

Much as the gender gap shrank from 2022 to 2018, so did the marriage gap. Married voters in 2022 voted for Trump past a 55% to 39% margin but supported GOP Firm candidates in 2022 by only a six-signal margin, 52% to 46%. Unmarried voters were strongly Democratic in both years (58% to 34% for Clinton in 2022 and 64% to 33% for Democratic House candidates in 2018). Much of the decline in the matrimony gap came from men. Trump won married men by a 30-point margin in 2016, simply this group backed GOP House candidates by 12 points in 2018. Married women were evenly divided between the parties in both elections. Amidst unmarried voters, women were more supportive of Democratic candidates in 2022 than they had been of Hillary Clinton in 2016.

Immature voters ages eighteen-29 were solid supporters of Clinton in 2016, but equally a grouping were even more Democratic in 2018. In 2016, voters ages eighteen-29 voted for Clinton over Trump by a 58% to 28% margin, with 14% casting votes for 3rd party candidates. In 2018, this group's votes went 72% for Democratic candidates and 23% for Republican candidates. Young voters, even so, were significantly underrepresented in the electorate due to depression turnout (as they ordinarily are). In 2018, they made upward 11% of all voters, significantly below their 21% share of the voting eligible population. Nonetheless, 37% of young voters in 2022 had not voted in the 2022 election, a far higher share than in any other age group.

Young voters supported Democratic candidates in 2022 at higher rates than Clinton in 2016, but continued to lag in turnout

By contrast, older voters continued to be the Republican Political party'due south near loyal age group. Trump carried voters ages 65 and older past a 9-point margin in 2016; Republican candidates for the House won this group by 6 points in 2022 (52% to 46%). Older voters were nearly one-3rd of all voters in 2022 (31%), most three times the share of those ages 18-29, despite making up almost the same overall share of the voting eligible population.

Support for Republican candidates amongst Blackness voters in 2022 was minimal (92% Democratic vs. 6% Republican in 2018, like to the 91% to 6% margin for Clinton in 2016). Republicans had more support amongst Hispanic than Black voters, but there were still lopsided majorities for Autonomous candidates (72% vs. 25% in 2022 and 66% for Clinton and 28% for Trump in 2016). There were as well few Asian American voters in the sample to yield a reliable estimate, but among Asian and other voters of color collectively the 2022 vote was 67% Democratic and thirty% Republican. White voters backed GOP candidates over Democrats by 6 points in 2022 (52% to 46%), though this represents a narrowing of Trump's 15-point margin over Clinton among White voters.

Geography remained a potent correlate of vote choice in 2018, with urban voters breaking Democratic past virtually a 3-to-one margin (73% to 25%), similar to their split in 2022 (70% Clinton, 24% Trump). Republicans had nearly a two-to-one reward over the Democrats with rural voters in both presidential voting and in 2018. Meanwhile, the Democrats fabricated gains among suburban voters. While Trump and Clinton had roughly divided the suburban vote in 2022 down the centre (47% Trump, 45% Clinton), Autonomous House candidates won the suburban vote by seven percentage points two years later (52% to 45%).

Voters of colour more often than not voted Autonomous regardless of where they lived, though Republican candidates received 37% of the votes of suburban Hispanics and 12% support among rural Black voters. White urban voters supported Democratic candidates by a roughly ii-to-one margin (64% to 34%) while rural White adults were a near mirror image (64% Republican, 33% Democratic). Suburban White voters, who favored Trump by 16 points in 2016, were more than divided in 2022 (51% Republican, 47% Democratic).

Already a strong Democratic grouping, those unaffiliated with a religious tradition became more than so

In 2018, voters were highly politically polarized by religious affiliation and attendance at worship services, as they accept been for many years in the U.South. Solid majorities of Protestants supported Republican candidates in 2018, while Catholics were more divided and the less religious were strongly Democratic in their votes.

White evangelicals remained loyal to Republican candidates, but Democrats made gains among secular voters

The Republican Party's well-nigh supportive demographic group (other than voters who identify as Republican or who are conservative) were White evangelical Protestants (81% voted Republican and 17% voted Autonomous). This margin was very like to 2022 (77% Trump vs. sixteen% Clinton). A sizable majority of White Catholics as well supported Republicans (59% to 39%), with White non-evangelical Protestants close behind (55% to 42%).

Unaffiliated voters – and especially atheists and agnostics – were even more supportive of Democratic candidates in 2022 than they had been of Hillary Clinton, with at to the lowest degree some of the change coming from those who had supported Gary Johnson or Jill Stein in 2016. The margins amid voters who describe their religious amalgamation as "nothing in particular" were fairly similar in 2022 and 2018. Atheists (7% of voters in 2018) supported Democratic candidates by an overwhelming 88% to 9% margin, rivaling Black back up for the Democrats. Agnostics (also 7% of voters) were not far behind, supporting Autonomous candidates by a 79% to eighteen% margin.

The solid back up for Democratic candidates among the unaffiliated is also reflected in voting patterns by attendance at worship services. Among those who nourish a few times a year or less often, 61% voted Democratic and 37% voted Republican. In 2016, this group voted 54% to 38% for Clinton. By dissimilarity, voters who attend services monthly or more often voted 58% to twoscore% Republican in 2018. Two years earlier, they voted for Trump by a 58% to 37% margin.

Democrats made minor gains in 2022 amidst non-college White voters

Perhaps the most important political tendency reflected in the 2022 issue was the continued movement of working-course White voters toward the GOP. Hillary Clinton lost White voters who did not accept a college degree by a wide 36 percentage points (64% for Trump vs. 28% for Clinton). Merely in 2018, Autonomous candidates managed to narrow the gap somewhat, losing this group by 61% to 36%, a 25-betoken margin.

At the same time, the Democratic Political party maintained a wide margin amidst higher-educated White adults. In 2016, Democrats won this group by 17 points (55% to 38%) and in 2022 past a nearly identical eighteen-point margin (58% to 40%).

2018 electorate highly polarized by education among White voters

Democratic candidates in 2022 did well among both the highest- and lowest- income voters. Voters reporting almanac family incomes of $150,000 or higher voted for Democratic candidates by a 59% to 39% margin. At the other farthermost, those with incomes below $30,000 voted 62% to 34% Democratic. Even amidst White low-income voters, Democratic and Republican candidates battled to a tie (48% each). Amid White voters with incomes between $30,000 and $74,999, Republican candidates had a 54% to 44% majority.

Democratic candidates had wide advantages among the highest- and lowest-income voters

The parties' coalitions, 2022 vs. 2016

Democratic voters in 2022 were younger, much more racially diverse than Republican votersPeople who voted for Democratic vs. Republican candidates for the House in 2022 were quite unlike demographically, in ways consistent with previous elections including 2016. The Republican coalition is more likely to be older, male, White, somewhat less educated and Protestant or Catholic.

In 2016, men made upwards only 39% of Hillary Clinton's voters. This share grew to 45% for Autonomous Firm candidates in 2018. But other than a slight increase in the share of Republican voters ages 65 and older, there was little change in the respective historic period profiles of the two parties' voters. Nearly half of those who voted for Democratic candidates were under 50 years of age, compared with nigh a third (32%) of Republican voters.

Non-Hispanic White adults fabricated up nearly nine-in-ten Republican voters (88%), compared with just two-thirds (65%) of Democratic voters. Only 1% of voters who chose Republican House candidates were Black (16% of Democratic voters were Black). Hispanics were 11% of the Autonomous voter coalition, compared with 5% for the Republican coalition.

Half of Democratic voters in 2022 were college graduates, compared with 35% of Republican votersHalf of Democratic voters in 2022 had a 4-year higher degree or more, compared with 35% of Republican voters. Voters with postgraduate degrees made up nearly a quarter (24%) of the Democratic electorate, compared with 13% among Republican voters. Combining this with the racial profile of the parties' supporters, 57% of GOP voters were White adults with no college caste, compared with 28% amidst Democratic voters.

Protestants made upwards a majority of those voting Republican in 2018, simply as they did in 2016. Overall, 57% of GOP House voters were Protestant, compared with just a 3rd (32%) of Democratic voters. Catholics made upward a slightly higher share of Republican voters as well (22% vs. 16% of Democratic voters). Voters who were unaffiliated with any religious tradition (atheists, agnostics and those who draw themselves as "nothing in particular") brand upward 42% of Democratic voters but just 15% of Republican voters.

Religiously unaffiliated were a bigger share of Democratic voters in 2022 than 2016

The demographic contour of voters and nonvoters is very dissimilar

As in 2016, 2022 nonvoters were younger and more racially diverse than votersThe roughly half of Americans who voted in 2022 differ from the voting-eligible adult population in some key respects. There were sizeable, if familiar, demographic and political differences in who did and did not turn out.

Compared with citizens who did not vote, voters were older, more likely to be college educated, amend off financially, more likely to be White Protestants or Catholics and more Republican in party affiliation and candidate preference. These differences are regular features of U.S. elections, as a comparing with voters and nonvoters in 2022 makes clear.

Compared with 2022 voters, nonvoters preferred Democratic candidates by a wider margin, but many declined to express a preferenceAll citizen panelists – whether voters or nonvoters – were asked which U.South. Firm candidate they supported in the full general election. Nonvoters tend to limited more uncertainty almost the option, attributable in big role to the fact that many of them pay little attention to politics. But amongst those who did express a preference, Democratic candidates led Republican candidates by fourteen percentage points (44% to 30%) a larger margin than among voters (9 points, 53% to 44%).

Demographically, the dissimilarity between voters and nonvoters is most stark on historic period, race, education and income. Voters in 2022 were considerably older than nonvoters: 31% of voters but just 10% of nonvoters were ages 65 and older. At the other end of the age spectrum, just 11% of voters were under 30 years of age; 30% of nonvoters fell into this category. These gaps are quite similar to those seen in 2016.

As in 2016, voters in 2022 were more affluent and more highly educated than nonvoters

Similarly, three-quarters of voters (75%) were non-Hispanic White adults, while 60% of nonvoters were White. Hispanics, in detail, were underrepresented equally voters. Only 8% of 2022 voters were Hispanic. Among the voting-eligible nonvoters, 17% were Hispanic. Blackness adults were 9% of voters merely 14% of nonvoters.

Voters tend to be more highly educated and more affluent than nonvoters. 1-quarter of voters had only a loftier school education, but 47% of nonvoters did so. More than four-in-ten voters (43%) were college graduates, compared with but 19% of nonvoters. The differences by income were similarly substantial. Just 17% of voters had almanac family unit incomes of less than $30,000. Among nonvoters, 40% did then.

White Protestants and White Catholics make upwards nearly half of all voters (46%) but only 32% of nonvoters. People who describe their religious affiliation as "nothing in particular" are underrepresented among voters, constituting 28% of all nonvoters but just 16% of voters.