Democrats held a 12-point advantage over Republicans when registered voters were asked which party they wanted to control the U.S. House, according to a Quinnipiac University national poll released Tuesday. The survey found 51% preferred Democratic control, 39% preferred Republican control and 9% gave no answer.
That is a measure of national preference, not a forecast of individual House races or seat totals. Quinnipiac surveyed 1,032 registered voters Sept. 24-27 and reported a margin of sampling error of plus or minus 3.8 percentage points, including the design effect. Its Sept. 10 poll showed a narrower 49%-38% Democratic lead; the difference between the two polls is too small on its own to establish a clear shift.
Independents favor Democrats while disapproving of both parties
Among independents, 54% wanted Democrats to control the House and 30% wanted Republicans to control it. Another 17% did not offer an opinion. The independent subgroup figures have a larger sampling uncertainty than the full-sample result, although Quinnipiac did not specify that subgroup's margin of error in its release.
The poll also found broad dissatisfaction with both congressional parties. Twenty-six percent approved of congressional Republicans' job performance and 67% disapproved. Congressional Democrats drew 22% approval and 69% disapproval. Even among their own party's voters, 49% approved of congressional Democrats, compared with 64% of Republican voters who approved of congressional Republicans.
On the seven issues Quinnipiac tested, voters more often picked Democrats than Republicans to do a better job. The largest gaps were on health care, 55% to 31%; the cost of living, 49% to 36%; government corruption, 45% to 33%; and artificial intelligence, 42% to 29%. The margins were narrower on the economy, 48% to 42%, and immigration, 47% to 44%.
Those responses describe perceptions of the parties, not evaluations of specific policy proposals. Some respondents volunteered that there was no difference between the parties or did not answer, so the two party percentages do not always sum to 100.
Trump approval and economic ratings
President Donald Trump's overall job approval was 34%, with 61% disapproving. Quinnipiac said the disapproval figure matched a high last recorded in its January 2021 survey. Its question on Trump's handling of the economy yielded 31% approval and 65% disapproval, which the pollster described as a low across his two terms.
On foreign policy, 34% approved and 62% disapproved. His handling of the situation with Iran drew 28% approval and 66% disapproval, while 25% approved and 58% disapproved of his handling of artificial intelligence. Thirteen to 17% of respondents did not give an opinion on the questions about China policy and AI, respectively, making those figures less directly comparable with questions that had fewer undecided respondents.
The party preference question showed a sharp gender divide: 62% of women preferred Democrats to control the House and 31% preferred Republicans, while men split 48% for Republicans and 41% for Democrats. As with the independent figures, subgroup estimates carry greater uncertainty than the overall poll.
Voters want tighter AI rules and resist data centers
The survey also tested attitudes about AI. Seventy-one percent said they would be more likely to vote for a candidate who supports stricter guardrails on AI development, while 16% favored a candidate who supports expanding development. That preference crossed party lines: 52% of Republicans, 86% of Democrats and 73% of independents chose stricter guardrails.
When asked to choose between AI safety and remaining at the forefront of innovation, 79% chose safety and 15% chose innovation. The wording presented those as competing priorities; the result does not establish which specific regulations voters would support.
Opposition to a data center in one's own community reached 73%, against 21% support. That included 55% opposition among Republicans, 85% among Democrats and 76% among independents. The question concerned a hypothetical local project and did not assess any particular proposed facility or its economic and energy effects.
Quinnipiac said it used a probability-based sample and live interviewers calling landline and cell phones. The survey reflects opinion during the Sept. 24-27 field period, about five weeks before Election Day. Polling cannot predict turnout, district-level outcomes or how undecided voters will ultimately vote.
FingerLakes1.com is the region’s leading all-digital news publication. The company was founded in 1998 and has been keeping residents informed for more than two decades. Have a lead? Send it to news@fingerlakes1.com.