More Americans voted in 2020 than in any presidential election in 120 years. About 67% of eligible voters cast ballots this year, but that still means a third did not.

That amounts to about 80 million people who stayed home.

To better understand what motivates these nonvoters, NPR and the Medill School of Journalism commissioned Ipsos to conduct a survey of U.S. adults who didn't vote this year. The Medill school's graduate students did deep dives into various aspects of the survey here.

Nonvoters' reasons for not voting include:

  • not being registered to vote (29%)
  • not being interested in politics (23%)
  • not liking the candidates (20%)
  • a feeling their vote wouldn't have made a difference (16%)
  • being undecided on whom to vote for (10%)

They are disengaged, disaffected and don't believe politics can make a difference in their lives. They are also more likely to be Latino, younger, make less money and have lower levels of education than voters.
A lack of engagement overall

Difficulty voting doesn't appear to be a major reason why they don't vote. Three-quarters said they think it's at least somewhat easy to vote.

It's more that these voters feel a sense of alienation and apathy. They are generally detached from the news and pessimistic about politics, the survey found.

Politics is simply not the way to make change, they said. Two-thirds of nonvoters agree, for example, that voting has little to do with the way that real decisions are made in this country; they are 21 points more likely to say so than people who voted.

A majority also said they believe it makes no difference who is elected presidentand that things will go on just as they did before. Nonvoters were 29 points more likely to say that than people who voted. ( Read more about why they didn't vote, in their own words.)

These 80 million Americans are also less engaged in their communities and have less confidence even in their local governments. They're also less likely to volunteer or to be civically engaged — doing things like sending letters to the news media and elected officials or participate in marches, protests and demonstrations.

Nonvoters are also more likely than voters to say that traditional parties and politicians don't care about people like me, the mainstream media is more interested in making money than telling the truth, the American economy is rigged to advantage the rich and powerful, success in life is pretty much determined by forces outside our control; and to feel that most issues discussed in Washington don't affect them personally.

They are also less likely to say that elections in this country are free and fair for all; to have a favorable opinion of the news media; to watch, read or listen to news every day; or to discuss politics with friends or family. If family and friends vote, that makes people more likely to vote, the survey found.

Nonvoters also mostly shunned news of the presidential election. For example, only 38% said they followed stories about the presidential and congressional campaigns in 2020 at least fairly closely, compared to 79% of people who voted.

And they were far less likely to have watched the debates or conventions. Less than a third watched the presidential debates, compared to almost two-thirds of voters. What's more, two-thirds of nonvoters said they didn't watch the presidential or vice presidential debates or either the Republican or Democratic conventions.

Campaigns, which have sophisticated data programs showing which are the highest-propensity voters, don't seem to view this group as worth much effort. Less than a quarter of nonvoters said campaigns had reached out to them, offered information or asked them to vote. That's compared to almost half of voters.

Not much, it seems, would motivate them to vote, either. Asked what they think would most encourage people to vote, the top answer was none or nothing (35%). That was followed by cleaning up government (27%), having more candidates to choose from (20%), being automatically registered to vote (16%) and making Election Day a national holiday (15%).

Making Election Day a national holiday, incidentally, was the top answer for voters (42%), showing the disconnect between the groups.

Who are nonvoters?

There are strong socioeconomic correlations between voting and not voting.

Nonvoters tend to make less money, have lower levels of education, be less likely to own their home or be married.

While only 21% of voters made $50,000 a year or less, 43% of nonvoters did. Just over a quarter of voters had a high school degree or less, but a majority (52%) of nonvoters did. Sixty percent of voters said they were married, but just 44% of voters did, which impacts economic power in a society when dual-income households are becoming increasingly necessary in more expensive parts of the country.

Nonvoters were also more likely to be young and Latino. Thirty-five percent of nonvoters in this survey were between the ages of 18 to 34, compared to 24% of voters. And notably, a quarter of nonvoters were Hispanic, compared to just 7% of voters.

Latinos tend to lean Democratic, and voter-registration efforts with Latinos in most presidential years are key for Democrats. That was hampered in this year of coronavirus, and Democrat Joe Biden wound up underperforming in some heavily Latino areas, like South Texas and South Florida.

Latinos are also more likely than other groups to say they are not interested in politics and most haven't voted in any other recent elections.

The top reason Latinos give for why they don't vote is that they don't care much about politics, but another is that they have never been registered.

Only 52% of Latinos surveyed overall said they were registered. That compares to 80% of white respondents and 78% of Black Americans.

But that 52% jumps to a whopping three-quarters among Latinos who did not vote.

Overall, 70% of nonvoters said they were not registered to vote.

SeparatorNPR and the Medill School of Journalism, Media, Integrated Marketing Communication commissioned Ipsos to conduct a survey of U.S. adults who voted and did not vote in the 2020 presidential election. The survey of 1,843 adults included 1,103 nonvoters and 740 voters. The survey was conducted using Ipsos' online KnowledgePanel from Nov. 4 to 13.

Overall, the survey has a +/-2.9 percentage point margin of error for all adults; for nonvoters, it is +/- 3.5 points; for voters, it is +/- 3.9 percentage point margin.

The Medill School of Journalism's stories, produced by graduate students in its Washington program, are available at

Copyright 2020 NPR. To see more, visit