Approval ratings normally provide the president with a real-time report card. That is, if the president believes the polls. President-elect Donald Trump might not.
On Tuesday Trump tweeted, “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.”
The tweet came hours after the release of a new poll from CNN/ORC which found that Trump has an approval rating of just 40%. President Barack Obama took office in 2009 with an 84% approval rating.
When contacted by CNN on Tuesday, former pollsters for Obama, George W. Bush and Bill Clinton expressed concern about Trump’s dismissive comment.
A fourth former pollster, Fred Steeper, who worked for both Bush and his father, said “I faced plenty of skepticism about poll results over 40 years of polling.”
Skepticism is one thing, but Trump is flirting with outright denial of unfavorable polls, recalling the unfounded claims about “crooked polls” he made during the campaign.
The subject seems to touch a nerve: just how popular is he? The data indicates he is much less popular than past presidents-elect were. But by casting doubt on the data, Trump is giving his supporters license to do the same.
The Drudge Report’s misleading headline on Tuesday was “BITTER MEDIA PLAYS WITH APPROVAL POLLS.”
Contrary to Trump’s assertions, professional pollsters do not “rig” polls. While imperfect to be sure, polling is a science, and experts work hard to reflect the country’s views as accurately as possible.
Some pre-election polls were far off the mark, especially in swing states, for reasons that are now being studied by experts. These state polls caused many analysts to anticipate a Hillary Clinton victory on election night. But most national polls correctly showed Clinton winning the popular vote by a relatively small margin.
Nationally, the polls “were roughly as accurate as they’ve been since ’68,” FiveThirtyEight founder Nate Silver wrote on Twitter.
News outlets like CNN conduct presidential approval rating polls, and so do private companies that work for the president’s team. Such polls, charting approval and disapproval for a period of years, have been critically important to past presidents.
The Clinton administration was famous for its frequent polling.
“It is a measure of how much politicians will defer to you,” Clinton’s longtime polling advisor Stan Greenberg said. “But it is not a public discussion. As president, you want the focus to be on your mandate, not the process.”
Greenberg said Tuesday’s tweet by Trump “shows how much he can be baited by bad news — and distracted from governing. Welcome to the NFL.”
Clinton’s successor, George W. Bush, “never dismissed polls as rigged, or that they weren’t factual,” Bush’s former pollster and strategist Matthew Dowd said. “He would often, though, say that he was going to do what was right regardless of the polls. His response might be, the polls will change if i do what is right.”
Bush “went out of his way post-2000 election to reach out and demonstrate he would be president for all,” Dowd said. “It is why his poll numbers rose between Election Day and inauguration, which is normal for a president. The opposite has happened for Trump.”
Bush took the oath of office with a 61% approval rating.
On Tuesday, Trump supporters tried to argue that CNN’s survey and another recent poll by ABC and The Washington Post are “oversampling” Democrats, perhaps intentionally to skew the poll results, which is not true.
The misunderstanding seems to lie in the difference between the partisan leanings of registered voters and those of all American adults.
CNN and ABC’s approval rating polls sought to measure the attitudes of all adults. Registered voters are slightly more Democratic than Republican, and that Democratic advantage widens when looking at the bigger pool of all adults.
The results of telephone polls of the adult population tend to line up pretty closely with other U.S. government statistics about the makeup of the population that can be used as a marker for poll accuracy.
So some of the claims about “rigging” simply fall apart under scrutiny.
Greenberg said that his research firm’s election year polling “did find that most of his supporters think the media rigged the polls because they thought it would hurt Trump.”
If Trump’s own skepticism is boomeranging back at him from supporters, it could be “a dangerous echo chamber for him,” Greenberg said.
Steeper, a Republican polling veteran, said Trump could raise some legitimate points about the limits of approval rating polling.
“I had many clients skeptical of poll results including presidents,” he said. The skepticism was usually for a good reason. The result could be transitory and would change with new information or the unfolding of events. That could be a point Trump could make now.”
News coverage also makes a difference, he said.
Cornell Belcher, a pollster for both of Obama’s presidential campaigns, had a different reaction to Trump’s claim about “rigged” polls.
“The fact that you are asking me a question about polls being rigged is the madness, and big tent carnival foolishness of the game Trump plays so well. I’m not going to dignify the game he’s playing by responding to the absurdity,” Belcher wrote in an email.
“The real question here is given how poorly he has behaved, how outside the boundaries of normal presidential behavior we see here since the election, and the fact that less than a plurality of voters supported him in the first place, of course his approval numbers are bad. Why would they not be in the tank? If, in fact, a majority of Americans approved of this unpresidential behavior in the polls — then I would have reason to be suspicious of the polls.”
