Young and Lower-Income Voters Abandon Republicans in 2026 Midterms
The AP Voter Poll reveals a dramatic demographic reversal, with under-30s, Hispanic voters, and moderates swinging sharply toward Democratic candidates.
By Allison Rojas ·
A sharp demographic realignment defined the 2026 midterm elections, as young voters, Hispanic citizens, and lower-income households abandoned the Republican Party in favor of Democratic candidates. The AP Voter Poll, which surveyed more than 100,000 voters across all 50 states, indicates that the gains President Donald Trump made with these groups in 2024 have largely evaporated. This shift reversed the traditional Democratic coalition's trajectory, with key demographic blocks moving left in a manner that helped Democrats secure victories in several contested states.
The most pronounced change occurred among voters under 30. In 2024, this group was roughly split between Trump and Vice President Kamala Harris. This year, about 7 in 10 young voters supported the Democratic candidate in their local U.S. House race, while only about 3 in 10 voted for the Republican. Approximately half of these young voters described themselves as angry about the direction of the country, a sentiment shared by only about one-third of voters aged 65 and older. About 6 in 10 under-30 voters said they cast their ballot to express opposition to Trump. This sentiment was visible in Senate races in Texas, Kansas, Iowa, Ohio, and North Carolina, where Democrats picked up seats.
Hispanic voters also reversed their 2024 trend. In 2024, about 4 in 10 Hispanic voters backed Trump. This year, only about one-third supported the Republican candidate in local House races, while roughly 6 in 10 voted for the Democrat. In the Texas Senate race, Democrat James Talarico won about 6 in 10 Hispanic voters, outpacing Republican Ken Paxton who received support from only about 3 in 10. In Florida, Republican Ashley Moody won reelection, but she secured support from less than half of Hispanic voters, a lower share than Trump received in 2024.
Black voters, particularly men, also moved back toward the Democratic Party. About 9 in 10 Black voters supported the Democratic House candidate, up from roughly 8 in 10 who backed Harris in 2024. Black men shifted even more dramatically; only about 1 in 10 backed a Republican House candidate this year, down from roughly one-quarter who supported Trump in 2024. In Georgia, about 9 in 10 Black men voted for Keisha Lance Bottoms, who became the first Black woman elected as governor. In Florida, Republican Byron Donalds won the governorship, but he received support from only about one-quarter of Black men and about 1 in 10 Black women.
Moderate voters played a decisive role in flipping Senate seats. Roughly two-thirds of moderates supported the Democratic House candidate this year, up from about 6 in 10 who backed Harris in 2024. About 8 in 10 moderates said Congress was not doing enough to check the presidency. In Kansas, Democrat Adam Hamilton won about 7 in 10 moderate voters, compared to roughly half for Harris. In Iowa, Democrat Josh Turek won about two-thirds of moderates, and in North Carolina, Democrat Roy Cooper won about three-quarters.
Economic anxiety drove lower-income voters to the Democratic side. More than half of voters with a 2025 household income under $50,000 voted for the Democratic House candidate, a group that had been evenly split between Trump and Harris in 2024. About 6 in 10 of these lower-income voters said the cost of living was crucial to their vote, compared to about half of voters earning $50,000 or more. These voters expressed higher levels of concern about food, gas, and utility prices than their higher-income counterparts.
The survey, conducted by SSRS for The Associated Press and partner news organizations, blends probability-based and non-probability-based interviews. Data was collected from October 19 to November 3 in Alaska and October 22 to November 3 in all other states. The margin of sampling error for the full sample is plus or minus 0.6 percentage points.
