2020 polls poorly underestimated support for GOP candidates, exam results, and no one knows why



[ad_1]

Joe Biden, Donald Trump

Joe Biden, Donald Trump Angela Weiss, Mandel Ngan / AFP / Getty Images

The 2020 polls have been staggered by an “unusual magnitude”, missing national results by the biggest margin in 40 years and getting the most wrong in state inquiries in at least 20 years, according to a new study released Monday by the American Association for Public Opinion Research (AAPOR).

“A systematic error has been found in terms of overstating Democratic support at all levels,” said Vanderbilt University political scientist Josh Clinton, chairman of the 19-member task force. “It doesn’t matter what type of poll you do, whether you interview over the phone or the Internet or whatever. And it doesn’t matter what type of race you are, whether or not President Trump is on the ballot.”

The task force looked at 2,858 presidential polls and found they were down 3.9 percentage points nationally and 4.3 percent at the state level. The numbers for President Biden were pretty accurate, about a point higher than his final vote count, but “Trump’s support was underestimated by 3.3 points on average,” Politics reports. “The polls of the Senate and Governor races were even further apart: 6 points on average.”

The 66 national polls conducted over the past two weeks accurately predicted Biden would win, the report said, but only 66% of Senate polls correctly predicted the winner.

The task force determined that pollsters largely corrected the errors in the 2016 poll, including the sub-sampling of voters without a college degree. “It’s hard to prove beyond a certain degree what happened,” Clinton said, but “we have good prime suspects as to what might be.”

The most likely theory is probably that pro-Trump Republicans have specifically refused to speak to pollsters, while “self-identified Republicans who choose to answer polls are more likely to support Democrats,” the report argued, skewing the sample of GOP voters.

The task force ruled out a significant “Trump voter timid” effect and found that the predicted composition of the electorate was largely accurate. “To conclusively identify why the polls overestimated the Democratic-Republican margin over certified voting seems impossible with the data available,” the report revealed. That leaves pollsters with little guidance for 2022 and 2024. “We’ll have to wait and see what happens – which is not a particularly reassuring position,” Clinton said. “But I think that’s the honest answer.” You can read the full report on AAPOR.

You may also like

Fox News host Tucker Carlson appears outraged that reporters investigate his allegations of NSA spying

Doctors unlikely to persuade COVID-19 vaccine hesitation, poll finds

Tennis player Coco Gauff to miss Olympics after testing positive for COVID-19

[ad_2]

Source link