The Center for Information and Research on Civic Learning and Engagement (CIRCLE)—the preeminent youth research organization at Tufts University—this morning released an analysis of 2016 exit poll data in historical context. As in the poll that we conducted before the election, we continue to see big differences among young people depending on their race and ethnicity.
This is part of a continued series by CIRCLE of national electoral results. Stay tuned for more demographic analysis for the presidential election and down-ballot races.
Youth presidential Vote Choice
Our pre-election poll of Millennials ages 18-34 had Clinton 49% vs. Trump 28%, a 21-point preference for the Democratic candidate. The National Exit Poll suggests that the actual split in the election was 55% for Clinton to 37% for Trump (an 18-point gap) among youth aged 18-29.
Throughout the election season, our analysis has emphasized the demographic and ideological heterogeneity of Millennials, contradicting facile generalizations that characterize them as the “Obama generation.” In particular, there are regularly stark differences in ideology and issue positions among Millennials of different races, genders, and socioeconomic status.
Nationally, compared to 2012, youth support for Republicans remained relatively constant among Whites (-3 points), African Americans (+1 point), and Latinos (+1 point). The national exit poll suggests that there were more youth in 2016 who supported a third-party candidate, did not vote for a presidential candidate, or specifically chose not to answer this poll question. White youth have been markedly less likely than youth of color to support Democratic candidates. This year, 43% of young Whites voted for Clinton, while Black youth supported her by almost double that margin (83%). However, it’s worth noting that young African Americans were notably less likely to support Clinton in 2016 (83%) than Obama in 2012 (91%).
Twenty-eight states had exit polls during the 2016 general election—primarily the most populous and most competitive states. Two of those states didn’t report presidential vote choice by age. Among the remaining 26 states, Trump won young voters in four states (Kentucky, Missouri, South Carolina, and Utah) and two saw a tie in youth support between the candidates (Indiana and Iowa), while . The youth support margins varied state-by-state, sometimes wildly. In South Carolina the exit poll estimates that youth support for Trump was 43% and 42% for Clinton; in Kentucky, Trump garnered 59% of youth votes while Clinton received 34%. Secretary Clinton won young voters in the other 20 states, and in ten of them by over 20 points.
Youth Affect Presidential Competitiveness in Key States
Our analysis of margin of victory in relation to estimated youth turnout and choice revealed that young voters played an important role in keeping some key races very close. If it were not for young voters’ support for Hillary Clinton, electoral college votes in Michigan, New Hampshire, and Nevada would have gone more decisively to Donald Trump. Michigan and New Hampshire are currently too close to call in part because of the edge that young voters gave Clinton in those states.
- In Nevada, Hillary Clinton won by an estimated 26,000 votes. Youth decisively favored Clinton in Nevada (52% to 35%), and this resulted in a net gain of 35,000 votes for Clinton. If young people had voted in equal number for Donald Trump and Hillary Clinton, Clinton’s victory would have been by an even narrower margin of less than 10,000 votes.
- In Michigan, which has not been called yet, young voters’ decisive support has played a major role in keeping the race close. Currently, Donald Trump is leading Hillary Clinton by about 12,000 votes, but youth favored Hillary Clinton by over 232,000 votes (57% to 34%). If young voters’ support for her was even slightly less decisive, Donald Trump would have won the state handily.
- In New Hampshire, where the vote is also too close to call with just about 1,400 votes separating Hillary Clinton and Donald Trump, young voters also gave the Democratic candidate a boost. Youth gave Clinton a net of 12,000 by favoring her 49% to 41%, and if young people had chosen instead to split their votes equally between Trump and Clinton, she would currently be behind by almost 5,000 votes. In New Hampshire, young voters were the only group that gave a significant edge to Clinton.
“In 2016, young voters were a substantial voting bloc and they influenced the outcome, although a majority of them ended up on the losing side of the presidential race,” said CIRCLE director Kei Kawashima-Ginsberg. “Young voters supported Hillary Clinton and other Democratic candidates more than any other age group did. However, they are a heterogeneous generation, and their choices differed greatly depending on their own race, their state, and other factors.”
Youth Voter Turnout
“Turnout” is the percentage of age-eligible citizens who voted, and we define “youth voter turnout” in this instance is the percentage of eligible 18 to 29-year-olds who voted. CIRCLE’s preliminary turnout estimate uses data from the National Exit Polls conducted by Edison/Mitofsky, the number of ballots cast in the United States (aggregated from data provided by local election officials), and current Census data on the number of young citizens in the United States.
As of this writing (noon on November 9, 2016), an estimated 23.7 million young voters participated in the 2016 presidential election, which is 50% of citizens aged 18-29 in the United States. We estimate that 13 million youth voted for Secretary Clinton and almost 9 million youth voted for Donald Trump. An additional 2 million young people either voted for third-party candidates or chose not to vote for any of the Presidential candidates on the ballot.
CIRCLE estimates youth voting after elections based on several variables, including the total number of ballots counted and the exit polls. These variables are subject to change in the hours and days after an election. In several states, less than 95% of precincts have reported their votes. Also, in some past years, the National Exit Polls (NEP) conducted by Edison Research have adjusted their statistics in the first few days after an election.
While we produced a preliminary national turnout estimate of 50% based on best-available data, it is vital to note that there can be dramatic differences in youth voter turnout by state.
By another measure of youth participation, the proportion of all votes cast by young people, in most of the 28 states where exit polls were conducted youth made up an equal or larger share of the electorate than senior citizens: higher in 14 states, and tied in three.
Note: No national official count of voters by age is immediately available after an election. Therefore, any statistic on youth voter turnout is an estimate based on survey data. Like any survey, the National Exit Poll uses methods that may introduce sampling bias. In past years, our estimates of youth turnout from the National Exit Poll (shown above) have produced a trend that very closely tracks the turnout trend in the Census Current Population Survey (CPS), which is the other major source for estimating youth turnout once it is released in the Spring. These estimates diverged slightly in 2012, although both showed a decrease. A third source of youth voting data is national voter files, which are based on aggregated state files available in the new year. Estimates based on this source tend to be underestimates, since not all states keep comprehensive age data. A lack of reliable data for years before 2012 also means that national trends are not currently available.