Remember the 2016 presidential election polls just before the vote?
The American Association for Public Opinion Research concluded Hillary Clinton’s “likelihood of winning the presidency was about 90 percent, with estimates ranging from 71 to over 99 percent.”
Now, President Trump is recalling that failed forecast in his response to a Fox News poll that finds a new high of 51 percent wants Trump impeached and removed from office.
“From the day I announced I was running for President, I have NEVER had a good @FoxNews Poll,” he wrote Thursday on Twitter. “Whoever their Pollster is, they suck. But @FoxNews is also much different than it used to be in the good old days.”
Fox News said that “since July, support for impeachment increased among voters of all stripes: up 11 points among Democrats, 5 points among Republicans and 3 among independents.”
“Support also went up among some of Trump’s key constituencies, including white evangelical Christians (+5 points), white men without a college degree (+8), and rural whites (+10).”
The report, which was the top headline on the popular Drudge Report, said that “among voters in swing counties (where Hillary Clinton and Trump were within 10 points in 2016), support for impeachment increased to 52 percent, up from 42 percent in July.”
Fox pointed out that “a lot has happened” on impeachment, with House Democrats ramping up their “impeachment inquiry.”
But just a day earlier, a Zogby poll found that in most of the nation’s demographic groups, a plurality expects President Trump to be re-elected in 2020.
“Nearly half of voters (47%) think President Trump will win re-election in 2020, while 41% say no, and it’s not just his base!” the report said.
Men by a 55% to 29% percent margin say so. So do young Millennials by 43% to 38%. And Gen X by 49% to 30%. And older voters 47% to 32%. Independents by 36% to 29%. Suburban voters, 42% to 34%. Large city voters, 51% to 32%. Union voters, 58% to 23%. Middle income, 56% to 32%. Upper income voters, 59% to 26%.
“Even in the demographic groups in which he’s not winning — women, Hispanics and moderates — the differential was no more than a single percentage point,” the report said.
“The public’s sentiment toward Trump is a mixed bag. Although, a plurality might support impeachment of Trump, his approval numbers are still very high. We currently have Trump’s job approval at 50% approve and 48% disapprove-October 3, 2019,” the report said.
The analysts at FiveThirtyEight suggested Trump might have some grounds for complaint, arguing the media focus on the impeachment drive itself could throw off polls.
“The impeachment story is blowing up. It’s a high-stakes moment — for President Trump, for Democrats and for pollsters. It’s also a scary moment for polling. Yes, people who follow politics are now intensely interested in whether the latest developments might shift public opinion about Trump and impeachment. But when news is exceptionally big, a growing body of evidence suggests it can throw off the accuracy of polling itself,” FiveThirtyEight said.
“The problem comes from what pollsters call ‘differential nonresponse bias.’ The idea behind this complex-sounding term is fairly straightforward: If partisans on one side of a political question respond to a survey more readily than partisans on the other side, you can get a polling error. The results in your poll won’t match the real-world opinion you’re trying to measure — instead, the poll will be skewed by how willing some people are to respond to a survey.”
The site pointed out that pollsters try to correct for such issues but aren’t always successful.
Especially, when it comes to “the highest of high-profile news events.”
“We have evidence of several brief episodes of nonresponse error, and all three came in the wake of big news stories near the conclusion of an election — in 2012, 2016 and 2018.”
In 2012, what appeared to be challenger Mitt Romney “chipping away” at President Barack Obama’s poll lead was “caused by a change in what kinds of people were responding to surveys,” it said.
Then in 2016, it was discovered Trump’s supporters “were less likely than Hillary Clinton supporters to participate in panel reinterviews conducted just after the release of the ‘Access Hollywood’ tape,” the report said.
In 2018, following the nationally televised congressional hearing in which Christine Blasey Ford leveled accusations of sexual assault against Brett Kavanaugh, “the survey completion rate jumped 3 percentage points among those who expressed approval of Trump while remaining essentially flat among those who expressed disapproval,” the report said.
The composition of the sampled groups changed, even after demographic weighting.
“What do these phantom trends have in common? Each came about a month before a national election, which may speak to the power of election campaigns in heightening awareness of political news. But more importantly, each involved the kind of story that FiveThirtyEight editor-in-chief Nate Silver recently described, in the context of handling ‘outlier’ polls, as ‘spectacular, blockbuster news events that dominate the news cycle for a week or more,’ the sort of story that happens only once or twice in an election cycle,” FiveThirtyEight said.
When polls suddenly show significant shifts on an issue such as impeachment, FiveThirtyEight asked, “will it be because Americans are genuinely warming to the idea or simply because Democrats are becoming more likely to take surveys than Republicans?”
The AAPOR defending its work in 2016, claiming that “national polls were generally correct by historical standards.”
They showed Clinton with a 3 percentage point lead, and she won the “popular vote by 2 percentage points.” However, the nation doesn’t elect by popular vote, but by the Electoral College, where Trump was a huge winner.
The pollsters blamed “state-level polls” that “showed a competitive, uncertain contest.”
The state polls showed Trump was one state away from winning the election.
In Pennsylvania, Michigan and Wisconsin, however, the polls forecast a seventh consecutive win for the Democratic candidate.
The pollsters said 13% of voters in those states decided in the last week. And pollsters failed to adjust for an overrepresentation in of voters with “higher education levels” who were “more likely to support Clinton.”
And there simply were many voters who “did not reveal themselves as Trump voters until after the election and they outnumbered late-revealing Clinton voters.”
The reported explained: “About those predictions that Clinton was 90 percent likely to win. … However well-intentioned these predictions may have been, they helped crystalize the belief that Clinton was a shoo-in for president, with unknown consequences for turnout. While a similar criticism can be leveled against polls – i.e., they can indicate an election is uncompetitive, perhaps reducing some people’s motivation to vote – polls and forecasting models are not one and the same. As the late pollster Andrew Kohut once noted (2006), ‘I’m not a handicapper, I’m a measurer. There’s a difference.’ Pollsters and astute poll reporters are often careful to describe their findings as a snapshot in time, measuring public opinion when they are fielded … Forecasting models do something different – they attempt to predict a future event. As the 2016 election proved, that can be a fraught exercise, and the net benefit to the country is unclear.”
Source: Why Fox’s ‘51% want impeachment’ poll is probably wrong