Donald Trump will take the oath of office as the least-favorably viewed president since at least 1977, with only 4 in 10 Americans viewing him positively. That finding, from a new Washington Post-ABC News poll, mirrors the results of a survey from Quinnipiac University, which pegged his favorability at 37 percent.
Trump fired off one of his social-media news releases on Tuesday, shortly after the Post-ABC poll went live. His excuse, as any close Trump observer might have predicted: Polls were and are wrong and rigged.
The same people who did the phony election polls, and were so wrong, are now doing approval rating polls. They are rigged just like before.
— Donald J. Trump (@realDonaldTrump) January 17, 2017
An experienced Twitter user, Trump packed a lot into those 140-odd characters. Most of it was wrong.
As our pollster Scott Clement pointed out at the end of last year, descriptions of 2016 polling as universally wrong are inaccurate. Nationally, polling broadly predicted that Hillary Clinton had more support — and she did, winning the popular vote by nearly 3 million ballots. The final average of polls compiled by RealClearPolitics estimated that Clinton would win by 3.2 points nationally; she won by 2.1. Both that average and the average compiled by Clement in his piece ended up closer to the mark in 2016 than in 2012, when President Obama’s victory was widely underestimated.
Where the polls were wrong was in the states. Pollsters in Wisconsin and Pennsylvania had Hillary Clinton winning those states even late in the cycle, but she didn’t. (Some of those pollsters conducted an autopsy of their misses with our Dave Weigel.) It was narrow wins in those states and Michigan — for a total of fewer than 78,000 votes — that handed Trump his electoral college margin, and the White House.
The Post-ABC and Quinnipiac polls Trump wanted to rebut on Tuesday were national polls. Quinnipiac didn’t poll nationally in the last few weeks of the campaign, but we did. We gave Clinton a four-point edge nationally — 1.9 points higher than it actually was. The margin of error in the poll was plus-or-minus 2.5 points. If the new favorability numbers are similarly wrong, it means that Trump’s viewed positively by 42 percent of Americans — still 16 points worse than any president in 40 years.
Trump was just a “normal” polling error away from winning in November. He is not a normal polling error away from being popular.
— (((Harry Enten))) (@ForecasterEnten) January 17, 2017
More broadly, Trump’s tweet tries to depict the polling as “phony” and “rigged.” This is a broader strategy by the incoming president than applies solely to poll results: He continually seeks to undercut news reports and media coverage as biased or inaccurate because it provides him the opportunity to suggest that accurate corrections or critical coverage are actually part of a sweeping conspiracy to make him look bad. (Since he declared his candidacy, it has proven easier for Trump to try to undercut criticism than to avoid it.)
Claims that polls are rigged or fake are simply nonsense. Read Weigel’s autopsy: Those pollsters who missed state results are clearly troubled by the problems that yielded those bad results. Those polls and ours follow established, proven methodologies for assessing popular opinions, a methodology that is subject to variations and errors (noted in the fine print of the results) but which usually get close to the mark. Unfortunately for pollsters, it’s when they miss that they get the most attention.
In our newest poll, we asked 1,000 Americans for their opinions on a wide range of issues. Those questions were posed by real people, calling landlines and cellphones. (We’ve described our methodology in detail previously.) The replies are compiled and weighted to match the population of the United States using statistically proven tools. And the consistent accuracy of the results — as in our national polling — demonstrates that this system works, even if it’s subject to small margins of error.
But it’s not enough for Trump to say that polling is subject to mistakes, which it is. Instead, he has to posit a grand conspiracy meant to undercut him.
The grand conspiracy is that Americans are skeptical of him as president-elect, more so than they have been of any other recent person in that position. They are skeptical of him now as they were skeptical of him during the campaign. Over the past year-and-a-half, his favorability rating has seldom been at or above 50 percent. Here, too, Trump has a choice: Build a bridge to those who dislike him or simply to wave those people away.
He consistently chooses the latter. Perhaps that’s part of the reason so many Americans remain skeptical of him.