College students across the United States more than doubled their rate of voting between the 2014 and 2018 midterm elections, according to a study published Thursday by Tufts University – a dramatic spike in political engagement that could draw unprecedented attention to these voters in next year’s presidential election.

The study found that 40 percent of students who are eligible to vote cast ballots last year, up from 19 percent in 2014.

Census Bureau data has shown that turnout rose in nearly all demographic groups between the two midterm cycles, but it rose most sharply among young adults. The Tufts study shows the turnout spike was particularly stark among college students – an extraordinary level of engagement for voters who typically stay home in nonpresidential elections.

Among all eligible voters, for instance, turnout reached 50 percent in 2018 – less than a 14-point jump since 2014, according to the United States Elections Project.

“It’s really a stunning comparison between college students and the rest of the United States population,” said Nancy Thomas, director of Tufts’ Institute for Democracy & Higher Education, which published the study. “We need to start listening to a constituency that has not been heard very much in the past and that is now making their voices heard.”

While past surveys have documented a spike in voting among young people in 2018, this study focuses on college students. The Tufts researchers amassed records of more than 10 million college students from more than 1,000 institutions, producing a comprehensive portrait of voting patterns by numerous demographic groups.

According to the study, college women voted at higher rates in 2018 than those of men, with black women voting more reliably than any other racial or gender group at 43 percent and Hispanic women logging the greatest gains between 2014 and 2018.

Overall, 36.5 percent of Hispanic college students voted, according to the study – up from just 14 percent in 2014. The rate among Asian students nearly tripled from 9 percent to 26 percent, but more among women than men. Higher engagement, combined with demographic shifts in university enrollment, led to an increase in the vote share of these two groups and a decline in the vote share of white and black students, the study found.

“The overall demographic of who goes to college is changing,” said David Brinker, a senior researcher at Tufts and one of the study’s authors.

The study’s findings mirror a census survey published earlier in the year showing that overall turnout spiked to a 100-year high in the first congressional election since Donald Trump became president. A Washington Post analysis of that data found that turnout increased by 21 percentage points from 2014 to 2018 among citizens ages 18 to 24 who are attending college – similar to the Tufts survey – but that turnout rose by a smaller 12 points among young adults who are not attending college.

The Tufts survey included all college students at the participating institutions, regardless of age.

Increased turnout among college students and young Americans is already drawing the attention of the field of candidates vying for the 2020 Democratic nomination for president.

Surveys show that young voters who turned out last year tended to overwhelmingly oppose Trump and supported such liberal causes as climate change, gun violence, immigration and student debt. In addition, exit polls showed 18- to 24-year-olds favored Democratic House candidates by a more than 2-to-1 margin in 2018.

If they repeat their 2018 performance, these voters could help power a Democratic campaign in battleground states.

Skylar Knight, 22, a graduate of Rollins College in Winter Park, Florida, said Trump’s election pushed his peers to pay attention to politics.

“In the era of normalcy with Barack Obama as president, people liked and disliked him, but it wasn’t so strong and so divided, and people weren’t really engaged in these kinds of conversations that early in life,” said Knight, now attending graduate school at George Washington University. “If there’s any good that’s come from the 2016 election, it’s that it’s really motivated young people to become educated and involved.”

Thomas said the Tufts study indicates that young voters provide an opportunity for both parties. “The Republicans could just as well make policy platforms that appeal to this constituency,” she said.

While one factor driving youth turnout is the heated national political environment, a year-round focus on engagement on campuses is also contributing, Thomas said.

“I know one engineering school, for example, where the voting rates are shockingly high,” she said, “because the president is out front and running this as a civic duty. They talk and talk and talk politics.”

At Rollins, for example, there is a heavy emphasis on civic engagement in a wide array of courses. A student in environmental studies would be required to work on a remediation project in the surrounding community. Someone studying literacy issues would teach young students to read. A counseling class discussing the threat to women in abusive relationships could also spend time in a shelter.

“One of the things we discovered that gets students really into learning is when they can see it play out in their world, and they can see it empowering themselves to make a difference in their world,” said Grant Cornwell, president of Rollins College.

Rollins’s history of high participation by students in elections is “not the consequence of particular get-out-the-vote efforts,” Cornwell said. “That’s a consequence of the way we approach the entire mission of Rollins.”

Overall, the study shows 2018 turnout rates to be slightly higher among students studying education, social sciences, health professions and humanities – and lower among business, science, engineering and math students.

The study revealed a gender voting gap in which women vote at higher rates than men, notably between black men and black women and Asian men and Asian women. Those gaps “present opportunities” for campaigns, Thomas said.

The study also revealed a significant gap between students older than 30 and those under 22, with older students voting at higher rates, but the gap narrowed by five points between 2014 and 2018.

And it found “relative consistency” in voting rates between students attending two-year, four-year, public and private institutions, with women’s colleges voting at the highest rates but all types seeing their rates increase since 2014.

The National Study of Learning, Voting, and Engagement is a project of the Institute for Democracy & Higher Education at Tufts University’s Tisch College of Civic Life.

The study analyzed 2018 voter turnout among more than 10 million students at 1,031 colleges and universities that volunteered to participate across all 50 states and the District of Columbia. Schools provided student data, which was matched to a national database of voter registration and turnout records. The researchers did not receive any information identifying individual students or how they voted.

Roughly one-fifth of students were not matched to voter registration records and were assumed to be non-voters. While some students are not U.S. citizens and thus ineligible to vote, the analysis accounted for the percentage of each institution’s students who are not U.S. citizens.

_ _ _

The Washington Post’s Scott Clement contributed to this report.