Cause of erroneous presidential polling data unclear, survey group says
An American Association for Public Opinion Research review released Monday found that pollsters overstated Biden's lead over former President Donald Trump by 3.9% nationwide and 4.3% on the state level. Photo by Jim Lo Scalzo/UPI | License Photo
July 19 (UPI) -- A national survey of over 2,000 polls nationwide has found the 2020 presidential polls to be the least accurate in 40 years while the state polls were the worst in at least two decades.
An American Association for Public Opinion Research review released Monday found that pollsters overstated Biden's lead over former President Donald Trump by 3.9% nationwide and 4.3% on the state level.
Polling numbers for Biden also were off by about 1%. The organization was unable to readily explain the disparity that occurred across a variety of polls.
"Identifying conclusively why polls overstated the Democratic-Republican margin relative to the certified vote appears to be impossible with the available data," the report states, according to the Washington Post.
In all, the association election task force reviewed 2,858 polls to reach its determination.
"We could rule some things out, but it's hard to prove beyond a certainty what happened," said Vanderbilt University professor Josh Clinton, who chaired the association's 2020 election task force.
"Based on what we know about polling, what we know about politics, we have some good prime suspects as to what may be going on."
The organization said the most likely answer is a segment of the Republican voting base is less likely to engage in polling.
"It seems plausible to the task force that, perhaps, the Republicans who are participating in our polls are different from those who are supporting Republican candidates who aren't participating in our polls," Clinton said. "But how do you prove that?"
The polls saw the worst performances since 1980.
In a visit to Philadelphia last week, Biden called allegations of election fraud a "big lie" and struck out against states that have passed sweeping election laws.
An American Association for Public Opinion Research review released Monday found that pollsters overstated Biden's lead over former President Donald Trump by 3.9% nationwide and 4.3% on the state level. Photo by Jim Lo Scalzo/UPI | License Photo
July 19 (UPI) -- A national survey of over 2,000 polls nationwide has found the 2020 presidential polls to be the least accurate in 40 years while the state polls were the worst in at least two decades.
An American Association for Public Opinion Research review released Monday found that pollsters overstated Biden's lead over former President Donald Trump by 3.9% nationwide and 4.3% on the state level.
Polling numbers for Biden also were off by about 1%. The organization was unable to readily explain the disparity that occurred across a variety of polls.
"Identifying conclusively why polls overstated the Democratic-Republican margin relative to the certified vote appears to be impossible with the available data," the report states, according to the Washington Post.
In all, the association election task force reviewed 2,858 polls to reach its determination.
"We could rule some things out, but it's hard to prove beyond a certainty what happened," said Vanderbilt University professor Josh Clinton, who chaired the association's 2020 election task force.
"Based on what we know about polling, what we know about politics, we have some good prime suspects as to what may be going on."
The organization said the most likely answer is a segment of the Republican voting base is less likely to engage in polling.
"It seems plausible to the task force that, perhaps, the Republicans who are participating in our polls are different from those who are supporting Republican candidates who aren't participating in our polls," Clinton said. "But how do you prove that?"
The polls saw the worst performances since 1980.
In a visit to Philadelphia last week, Biden called allegations of election fraud a "big lie" and struck out against states that have passed sweeping election laws.
Finding will alarm Democrats aiming to hold on to their narrow control of the US House and Senate in 2022
Political polls regarding US elections in 2020 overstated Democratic support “across the board”, US political scientists found, while understating support for Republicans and Donald Trump.
‘It didn’t matter what type of poll you were doing, whether you’re interviewing by phone or internet or whatever.’ Photograph: Erik S Lesser/EPA
Martin Pengelly in New York@MartinPengelly
Mon 19 Jul 2021
The finding, which will alarm Democrats aiming to hold on to their narrow control of the US House and Senate in 2022, is contained in a new study by the American Association for Public Opinion Research.
Josh Clinton, a Vanderbilt University professor and AAPOR taskforce member, told the Washington Post: “There was a systematic error that was found in terms of the overstatement for Democratic support across the board.
“It didn’t matter what type of poll you were doing, whether you’re interviewing by phone or internet or whatever. And it didn’t matter what type of race, whether Trump was on the ballot or was not on the ballot.”
Polls were better at predicting support for Joe Biden against Trump in the presidential election than for Democrats in state elections, the study said.
In 2020, polling pointed to Democratic gains in the House, only for Republicans to eat into the majority which made Nancy Pelosi speaker in 2018.
Parties which hold the White House often lose seats in midterm elections. Republicans in Washington are duly bullish about their chances of retaking the House next year, boosted by GOP-run state governments implementing laws meant to restrict voting among communities likely to vote Democratic and to make it easier to overturn results.
Speaking at a conservative conference earlier this year, the House minority leader, Kevin McCarthy, said: “We’re going to get the majority back … I would bet my house.”
The AAPOR study found that polls were more accurate in predicting a popular-vote win for Biden, a contest he eventually won by more than 7m ballots.
The electoral college result was 306-232, the same margin by which Donald Trump beat Hillary Clinton in 2016, when the Republican lost the popular vote by “only” 2.8m.
On Monday, the AAPOR website was down for maintenance. As quoted by the Post, its study said: “That the polls overstated Biden’s support more in whiter, more rural, and less densely populated states is suggestive (but not conclusive) that the polling error resulted from too few Trump supporters responding to polls.
“A larger polling error was found in states with more Trump supporters.”
Josh Clinton, the Vanderbilt professor, said: “It’s possible that if President Trump is no longer on the ticket or if it’s a midterm election where we know that the electorate differs in the presidential election, that the issue will kind of self-resolve itself.
“But if the polls do well in 2022, then we don’t know if the issue is solved or whether it’s just a phenomenon that’s unique to presidential elections, with particular candidates who are making appeals about ‘Don’t trust the news, don’t trust the polls’ that kind of results in taking polls becoming a political act.”
No comments:
Post a Comment