Politics

Add as a preferred source on Google

Marist Poll Finds Democrats Ahead by Eight Points on U.S. House Ballot

The survey put Democratic candidates at 52% and Republicans at 44%, with a ±3.8-point margin of error. AAPOR's review found a seven-point average error in 2024 House polls.

Updated 2 min read
A voter casts a paper ballot at a neighborhood polling station in Washington, D.C. / EconoTimes.ai
A voter casts a paper ballot at a neighborhood polling station in Washington, D.C. / EconoTimes.ai

The U.S. midterm election will turn on 435 House district races, while a national poll's 52% Democratic preference offers a broad read on voters but no answer on who will control the chamber.

Released Oct. 27, the NPR/PBS News/Marist Poll surveyed 1,289 registered voters from Oct. 19 to Oct. 20 by phone, text or online. Among likely voters, 52% favored the Democratic candidate for Congress in their district, compared with 44% for the Republican. The poll reported a margin of error of ±3.8 percentage points.

Election Day is Tuesday, Nov. 3. The Federal Election Commission lists that date as the next regularly scheduled federal general election; USAGov says all 435 House seats and about one-third of the Senate are contested every two years. The Associated Press says a nationwide House ballot question does not track outcomes in individual districts, the races that determine the chamber's makeup. The Marist result therefore does not show which party will control the House.

Past performance offers context, not a verdict on this year's polling. The American Association for Public Opinion Research (AAPOR) assessed 611 surveys whose interviews concluded shortly before the 2024 election. It calculated average absolute error of 3.3 percentage points in the two-party margin, rising to 7 points for House surveys. The review also found that polling generally understated Republican support. Those results describe 2024 performance and do not establish how accurate 2026 polls will be.

The margin printed beside a poll captures sampling uncertainty, AAPOR says, but leaves out other errors, including nonresponse bias and turnout models that misidentify voters. The association urges readers to examine how survey participants were selected, contacted and weighted, along with how pollsters define likely voters. The reported margin does not measure every source of polling uncertainty.

In an NPR pre-election Q&A, Domenico Montanaro, the network's senior political editor and correspondent, said a Democratic majority in either chamber would likely bring more investigations and oversight of President Donald Trump's administration. Legislation could stall without compromise, he said. “Polling is one tool. It’s not everything,” Montanaro said. Voters will decide control of Congress on Nov. 3.

Related articles

View all