With the U.S. midterm elections approaching on November 3, 2026, our latest survey examines Americans’ voting intentions, confidence in the electoral process and views of President Donald Trump’s job performance.
Conducted from October 23 to 25, 2026, the survey provides a snapshot of public opinion ahead of the upcoming elections.
Key Highlights
-
Among decided voters, 52% intend to vote for the Democratic candidate, compared with 45% for the Republican candidate.
-
48% of Americans believe the upcoming midterm elections will be free and democratic, while 25% do not and 26% are unsure or prefer not to answer.
-
68% of Americans say they would accept the election results even if the political party they support were defeated.
-
33% approve of Donald Trump’s job performance as president, while 58% disapprove.
How Do Americans View the 2026 Midterm Elections?
Views on the upcoming elections vary by political affiliation. When asked whether they believe the elections will be free and democratic, 60% of Democrats, 57% of Republicans and 38% of independents said yes.
Willingness to accept an election result in which their preferred party loses is also reported across political affiliations, with 76% of Republicans, 73% of Democrats and 63% of independents saying they would accept the outcome.
How Do Americans Assess President Trump's Performance?
Presidential job approval differs considerably by political affiliation. Among Republican respondents, 77% approve of Trump’s job performance, compared with 7% of Democrats and 24% of independents.
Across the individual policy areas measured in the survey, satisfaction with the president’s performance ranges from 24% to 39%.
For more findings and detailed breakdowns of Americans’ views on U.S. politics, download our full report.
Methodology
This online survey was conducted from October 23 to 25, 2026, among 1,037 Americans aged 18 or older, recruited through our LEO online panel.
Results were weighted according to age, gender, region, education, ethnicity and household size to reflect the U.S. adult population.
Because this was a non-probability online survey, no conventional margin of error applies. For comparison purposes, a probability sample of this size would yield a margin of error no greater than ±3.09 percentage points, 19 times out of 20.
