Forced to choose between President Biden and former president Donald Trump, she said she would lean toward Biden, then added, “But I just don’t agree with him as president at all.” Her dissatisfaction with Biden includes her belief that he is too old for another term, that he hasn’t delivered on a promise to eliminate her student loan debt and that he supports Israel in the war in the Gaza Strip.
Tamara Etter, 48, lives in Cobb County in suburban Atlanta. She supported Biden in 2020 when she was living in Missouri. She is now a registered voter in Georgia but is looking beyond the two major party candidates. This year she has been attracted to the candidacy of Robert F. Kennedy Jr.
“If Kennedy is on the ballot, 100 percent I’ll vote for him,” she said. “Biden has lost my favor. The way he has been campaigning lately seems a bit pandering.” And, she added, “He is not ‘with it’ mentally.” Only if Kennedy is not on the ballot might she vote for Biden. As for possibly supporting Trump, she said, “No way,” citing his role in the Jan. 6, 2021, attack on the Capitol. “And now he’s a convicted felon.”
Denning and Etter reside in two of the six states likely to decide the outcome of the presidential election, the others being Wisconsin, Michigan, Pennsylvania and Arizona. Trump carried all but Nevada when he won the presidency in 2016. Biden swept all six in 2020.
In a nation where many voters have made up their minds, Denning and Etter are among the voters whose decisions about the presidential race are neither firmly fixed nor whose participation is wholly predictable. As a group, these voters do not exactly fit the description of being undecided. Some lean toward a specific candidate. Some even say they will definitely vote for that candidate. But age or voting history or both leave open the question of how they will vote in November — if they vote at all.
The Washington Post and the Schar School of Policy and Government at George Mason University surveyed 3,513 registered voters in the six key battleground states. The survey was completed in April and May, before a New York jury found Trump guilty on 34 counts in the hush money trial involving an adult-film actress. Of the 3,513 surveyed, 2,255 were classified as “Deciders” — those who fit into one or more categories: They voted in only one of the last two presidential elections; are between ages 18 and 25; registered to vote since 2022; did not definitely plan to vote for either Biden or Trump this year; or switched their support between 2016 and 2020.
They are also classified as Deciders because they will have enormous influence in determining the winner of what are expected to be another round of close contests in the battleground states.
“Definitely”
Biden
15%
“Definitely”
Trump
17%
17% fall
into both categories
Very low-
turnout voters 6%
Voted in 2016
or 2020,
not both
24%
3% of voters are ages 18 to 25 and turned out in 2020
*Includes those who switched or voted for different parties in 2016 and 2020.
Source: April 15-May 30 poll with an error margin of +/- 2.0 percentage points. Numbers may not add to 100 because of rounding.
KEVIN UHRMACHER/THE WASHINGTON POST
“Definitely”
Biden
15%
“Definitely”
Trump
17%
Very
low-
turnout voters 6%
17% fall
into both categories
Voted in 2016
or 2020,
not both
24%
3% of voters are ages 18 to 25 and turned out in 2020
*Includes those who switched or voted for different parties in 2016 and 2020.
Source: April 15-May 30 poll with an error margin of +/- 2.0 percentage points. Numbers may not add to 100 because of rounding.
KEVIN UHRMACHER/THE WASHINGTON POST
In 2020, a shift of about 43,000 votes from Biden to Trump in Arizona, Georgia and Wisconsin would have changed the outcome. As a result, it is common to see suggestions that the 2024 presidential election will not only be decided by just six states but by relatively few voters in those states. While it is broadly true that a fraction of the total electorate will decide the election, the universe of voters whose behavior is not truly predictable is fairly large. By the definitions used in this survey, 61 percent of voters in those six states can be called Deciders. That includes 33 percent who are sporadic voters and 44 percent who are uncommitted to Biden or Trump, with 17 percent fitting both of these categories.
While the size of potentially persuadable voters might seem high, it’s echoed by other data. The Pew Research Center found that 25 percent of 2020 voters had not voted in the 2016 presidential election. And a June Monmouth University national poll found 63 percent of registered voters saying they would definitely vote for Trump or Biden, leaving nearly 37 percent who would “probably” vote for a candidate or were unlikely to support either.
To add more perspective, because there are only six states (seven at most) that will decide the outcome in November, the voters identified in this survey as Deciders represent just 10 percent of American voters overall.
The overall findings of the Post-Schar School poll highlight problems facing Biden as he seeks a second term, in large part because the profile of the Deciders skews more in the direction of his natural coalition than Trump’s — younger and more non-White than the electorate at large, for example. This survey and other public polling this year has shown softness in Biden’s support among younger voters and among some Black and Hispanic voters.
What unites these Deciders? First, they are even more sour on the major candidates than the broader electorate. About 7 in 10 Deciders are not satisfied with the choice of candidates, compared with about 4 in 10 among voters who have clearly made up their minds. About 4 in 10 Deciders say no candidate truly represents their views well, compared with about 1 in 10 among those who are not Deciders.
“You get the sense from the polling numbers that people sense the urgency of this election,” said Justin Gest, a professor of government at the Schar School. “They feel the electricity and the stakes behind it, and they want to care. But the parties have nominated two candidates that deflate the public.”
The lack of enthusiasm for the two major candidates is evident in other ways. Only 6 percent of Deciders say they would be “enthusiastic” if Biden were reelected while 15 percent say they would be “enthusiastic” with a Trump victory. More say they would feel at least “satisfied” with a Kennedy victory than say the same for Biden or Trump.
The survey did not ask people to pit Biden versus Trump in a hypothetical ballot test of the kind that many polls do. Instead, respondents were asked how likely they were to support each of the candidates separately. The results find 44 percent of Deciders say they are considering voting for Trump, compared with 38 percent saying the same for Biden. If Kennedy makes the ballot as an independent, 39 percent say they would consider voting for him. A far-smaller 10 percent of Deciders say they would “definitely” vote for Kennedy, which is closer to polls that ask a traditional vote choice question.
At the same time, 61 percent said they would definitely or probably not vote for Kennedy, compared with 59 percent who said that of Biden and 54 percent who said that of Trump.
These percentages add up to more than 100 because respondents were asked about each candidate individually, not forced to choose in an either-or question. Among registered voters overall in these states, 48 percent said they would definitely or probably support Trump, 41 percent for Biden and 29 percent for Kennedy.
Among Deciders, both Biden and Trump are viewed negatively in assessments of how they handled their presidencies, but Biden’s disapproval (66 percent) is notably higher than Trump’s (53 percent). Among non-Deciders in the six battleground states, approval of Trump’s handling of the office is positive (55 percent approve to 45 percent disapprove) while Biden’s is negative (44 percent approve to 56 percent disapprove).
One factor in assessing how these Deciders might act in the fall election is their view of the current state of politics and their interest in the election. Although some say they will definitely vote for either Biden or Trump, their voting history leaves open the question of whether they will follow through. The fact that they are less focused on the election underscores that these voters are different from those who are firmly committed and have a history of regular voting.
Almost 8 in 10 Deciders say they feel “worn out by the amount of politics news there is these days,” compared with about 7 in 10 non-Deciders. Four in 10 say they have given “quite a lot” of thought to the election, considerably smaller than the roughly 3 in 4 voters who are locked in supporting Trump or Biden. And while 60 percent of Deciders say they are either “extremely” or “very” motivated to vote in November, that’s far lower than the more than 9 in 10 (96 percent) of locked-in Biden and Trump voters who are so motivated.
“There’s a sense that there’s something important happening but nobody is particularly motivated to do anything about it,” said Mark J. Rozell, dean of the Schar School. “That is an opportunity for either candidate to find something to mobilize the American public.”
Among those non-Deciders, 65 percent say their vote matters “a great deal,” compared with 38 percent of Deciders. On a related question, 48 percent of Deciders said they care “a great deal” about who wins the presidential election, compared with 84 percent who are not identified as Deciders.
Denning, for example, said she would feel badly if Trump were to win the election, but is nonetheless ambivalent about what to do. “I either am voting for Biden or not at all,” she said. “I’m not voting for Trump.” But she added, “I don’t think it’s going to matter either way who I vote for. I am just one person.”
That points to one big wild card this year — the question of how many people will vote. Between 2016 and 2020, overall turnout jumped from about 137 million to about 158 million. Both parties saw a substantial increase in total votes for president, but the growth among Democrats was about 4 million more than the growth among Republicans. Analysts have said that Trump has motivated both his own supporters as well as those opposed to him.
There was an opposite pattern in the two most recent midterm elections. Turnout declined from about 112 million people in 2018 to 108 million in 2022. The big difference was a sharp decline — about 9 million — in the number of people who cast ballots for Democrats even as Republican turnout increased.
In addition, there’s significant churn in which voters turn out from election to election. The L2 voter database shows 28 percent of people who voted in 2020 in the six states in the Post-Schar School poll have no record of voting in 2016.
In the Post-Schar School poll, the Deciders are younger than non-Deciders, in part because of the definitions used to delineate among different voters. All voters in the 18- to 25-year-old age group are included because, while they are registered, they have had the opportunity to vote in only one presidential election.
Younger voters often turn out in lower percentages than older voters and, beyond that, young voters are far less committed to one of the major parties than are older voters. Biden’s campaign is keenly aware of the need to find ways to motivate these younger voters, who in recent campaigns have strongly backed Democratic candidates.
Deciders are more likely to be Black or Hispanic than non-Deciders. One-third of Deciders are non-White compared with about a fifth of other voters.
Deciders are more likely than non-Deciders to live in urban areas and are more likely to have no religion. They are twice as likely as non-Deciders (17 percent to 9 percent) to get their political news from TikTok and somewhat more likely to turn to YouTube for political news.
Overall, Deciders are relatively liberal on issues like abortion and immigration. More than 6 in 10 support the idea of offering most undocumented immigrants the opportunity to apply for legal status, rather than deporting them, as Trump has said he would do in a second term. That’s a higher percentage than among non-Deciders. Still, 6 in 10 Deciders say people who have immigrated illegally have made the communities in which immigrants live worse, about the same as non-Deciders.
More than 7 in 10 Deciders oppose the Supreme Court decision ending the constitutional right to an abortion. That’s less than locked-in Biden supporters (97 percent) yet higher than all non-Deciders (63 percent). Deciders are 12 percentage points more likely to say hard work and determination are no guarantee of success for most people, 49 percent vs. 37 percent for non-Deciders.
This pattern is in line with Deciders having lower incomes, as voters with household incomes under $50,000 are more likely to doubt that hard work guarantees success than those with higher incomes.
Deciders are more likely to rate the economy negatively than non-Deciders. Three-quarters of Deciders say the U.S. economy overall is “not-so-good” or “poor” (73 percent), compared with 64 percent of non-Deciders. And while 63 percent of Deciders rate their local economy negatively, 57 percent of non-Deciders say the same. Deciders split 50-50 on whether their personal finances are good or bad, while 63 percent of non-Deciders rate their finances positively.
Deciders say their top issues this year, those rated as extremely important, are the economy, threats to democracy in the United States and crime. On all three, more trust Trump to deal with them over Biden.
About 6 in 10 rate the economy as extremely important and trust Trump over Biden by more than 2-to-1. More than half (56 percent) say threats to democracy in this country represent an extremely important issue — lower than the 7 in 10 among non-Deciders — and by a nine percentage-point margin (38 percent to 29 percent), Deciders say they trust Trump over Biden. About half (48 percent) rate crime as an extremely important issue and trust Trump over Biden by about a 2-to-1 margin.
Three other issues are seen as extremely important by about 4 in 10 Deciders: racism, abortion and immigration. By double digits, they trust Biden over Trump on racism and abortion but trust Trump by a wider margin on immigration (49 percent to 21 percent).
Four other issues rate slightly lower on their scale of importance to Deciders: gun control, climate change, the Israel-Gaza war and the war between Russia and Ukraine. Only on climate change do they trust Biden more to handle the issue rather than Trump.
Jason Burnett, 42, of Janesville, Wis., said, “I don’t see a candidate who will actually bring the country together,” but he will probably vote for Trump to “get the country back on track to be the superpower it once was.” The Army veteran criticized Biden’s handling of covid-19, immigration and the withdrawal from Afghanistan, and said he is “too old and too incompetent to lead our country.”
Among all Deciders, 53 percent said they believe Trump accomplished either “a great deal” or “a good amount” as president, while 31 percent of that group said the same of Biden’s accomplishments.
Andrew Moran, 36, who lives in Avella, Pa., said he thinks the country was in better shape when Trump was president. “Overall, I didn’t feel the way I feel now,” he said. “It felt better then.”
But for Moran, neither of the two major candidates holds much appeal. He called Biden “a lifetime politician” and said the president’s age is a major issue to him. Of Trump, he said, “He’s noisy and pretty obnoxious. He’s a Republican. I generally vote that way, but he’s kind of an embarrassment, too. His values are not aligned with what a good person is. He cheats on his wife. His character isn’t there.”
Today, he said he plans to support Kennedy. The independent candidate’s message and approach “align with my values,” he said, adding, “I think Kennedy needs to be on the debate stage.”
The first debate is scheduled for June 27 in Atlanta, hosted by CNN. The network has not said whether anyone other than Trump or Biden has qualified to participate.
The Deciders poll was conducted by The Post and George Mason University’s Schar School of Policy and Government April 15 to May 30, 2024, among a stratified random sample of 3,513 registered voters in Arizona, Georgia, Michigan, Nevada, Pennsylvania and Wisconsin. The sample was drawn from the L2 database of registered voters in each state, with voters contacted by mail, phone (including calling and texting) and email to complete the survey online or by phone. Results for the overall sample of voters in key states have a margin of sampling error of plus or minus two percentage points.