What Voter Motivation Reveals About 2026

Every September and October before a midterm, the same question resurfaces in polling coverage: could the polls be wrong? This time, there’s a follow up: are Democrats leading by too much? A Change Research national survey of 2,021 registered voters, conducted September 15-24, 2026, illustrates why the challenge inherent in that question. Democrats lead the 2026 congressional generic ballot by 7 points overall. There is, however, a wide range of possible outcomes depending on which voters turn out in November and which way undecided voters move.

With roughly six weeks until Election Day, discussion of what current polling suggests for November has grown. President Trump's approval ratings have declined since his 2024 re-election, and his campaign is traveling to 17 states ahead of the midterms in an effort to drive Republican turnout.

What Polls Measure

A poll is not a precise prediction of a future outcome. It is a tool for understanding the current electoral environment so that campaigns, organizations, and voters can make informed decisions. A useful poll describes a range of possible outcomes. Both a wide Democratic gain, including in districts Kamala Harris lost by double digits, and a modest Democratic gain in a competitive race are visible within the same survey.

Motivation Scores and Vote Choice

The primary challenge in electoral polling is that we do not know who will actually vote. We can make estimates based on past voting behavior and what people tell us about their engagement. In each Compass Poll, we ask respondents to rate their motivation to vote in the November election on a 0-to-10 scale, where 0 means not motivated at all and 10 means extremely motivated. Research has shown this question to be a stronger predictor of actual vote than measures of excitement or straight vote intention. Because people who take polls are already more likely to vote than those who decline to take them, those rating themselves below a 10 on vote motivation tend to view the political environment differently from those whose motivation is a 10.

74% of registered voters rate their motivation as a 10. Among that group, Democrats lead by 13 points. Among the remaining 26% of respondents, Democrats trail by 10 points; 27% of that group supports Democrats, and 37% remain undecided.

2026 Compared to 2014

On the surface, this looks like a familiar midterm pattern. The more you examine the less motivated voter, however, the more this cycle diverges from recent precedent.

The 2014 midterms offer a comparison. During President Obama's second midterm, the most motivated voters disapproved of the President’s performance while less motivated voters approved.

At the time of this survey, 2026 runs in the opposite direction: less motivated voters hold more negative views of President Trump than those most committed to voting.

President Trump's campaign has described the midterm strategy as turning out the "infrequent Trump voter." However, this strategy may be unwise. Those who are less motivated to vote hold more unfavorable views of the President on all issues except the border, in comparison to those who are more committed to voting. This is not a function of less motivated voters leaning Democratic. Few identify as Democrats, and just 23% of them say they voted for Kamala Harris in 2024.

Less Motivated Voters: Economic and Issue Views

Less motivated voters hold negative views of the economy and the cost of living. Roughly a third see their personal finances improving in the next year. They rate the President's performance on most issues we asked about poorly; border security is an exception.

Their views of the Democratic Party are also negative. Republicans hold a spending advantage heading into November, and would need to convince less motivated voters to both turn out and vote Republican despite negative views of both parties' current leadership. Both scenarios remain possible. That is what current polling is showing.

The Range of Outcomes

Democrats hold a 13-point advantage among voters rating their motivation as a 10 in the generic ballot, and Republicans lead by 10 points among less motivated voters, with 37% of that group still undecided. That second group is the key variable. If less engaged voters stay home in November, the electorate looks like the first number. If they turn out in large numbers and break toward Republicans, the race is considerably closer. Where those voters go, and whether they vote at all, is not something that current polling can tell us.

Methodology

Polling was conducted online from September 15-24, 2026. Using Dynamic Online Sampling to attain a representative sample, Change Research polled 2,021 registered voters nationally. The modeled margin of error is ±2.3 percentage points. The 2014 comparison draws from a September 2014 Pew Research Center survey. Because that survey did not include a 0-to-10 motivation scale, the author recoded four questions to identify the most engaged voters in that sample: likelihood of following government and public affairs; amount of thought given to the coming November election; self-reported voting frequency; and enthusiasm relative to prior elections. The 2014 analysis reflects the author's own recoding and any errors are his own. Members of AAPOR's Transparency Initiative disclose all relevant details about our research, with the principle that the public should be able to evaluate and understand research-based findings, in order to instill and restore public confidence in survey results.