Little significance in political polls about Trump

Leave it to NPR’s political editor Domenico Montanaro to provide a needed dose of reality to the anti-Trump forces gleefully reacting to a poll showing declining popularity of the president.

In Minnesota, only 38 percent approve of the job the president is doing, according to the Marist Poll of Michigan, Wisconsin, and Minnesota. There’s a huge margin-of-error in the Minnesota poll of 4 percent.

“Donald Trump carried or came very close to carrying these three states in 2016. But it’s a very different picture for this fall’s elections,” says Dr. Lee M. Miringoff, Director of the Marist College Institute for Public Opinion.

So it means something, right?

Maybe not, Montanaro appropriately cautioned in his appearance on NewsHour on PBS last evening.

Take Minnesota, for example, just before the November 2016, only 39% of those polled favored Trump against Hillary Clinton. That’s well within the margin of error of this week’s poll.

“I went back and looked at the 2016 exit polls,” Montanaro said. “And you might remember there that some 60 percent said that President Trump didn’t have the temperament to be president, they said he wasn’t qualified to serve as president, and that they would be concerned or scared if he won. And yet, he won.”

“And here we are,” he concluded.

In a world where political polls, like facts, mean nothing.

  • BJ

    In Minnesota, a staggering 38 percent approve of the job the president is doing, according to the Marist Poll of Michigan, Wisconsin, and Minnesota.

  • Brian Simon

    “In a world where political polls, like facts, mean nothing.”

    Which has always been the case.

    The comparison to 2016 is interesting and relevant, of couse; but does it tell the whole story?

    More interesting to me would be a gauge of people who weren’t excited about Clinton & whether they’re more enthused now, checked out entirely, rallying to Trump, or what.

    The POTUS approval rating tells us something; I’d argue that when it hovers near party identification rates, things don’t bode well for that party.

    • The media will get to that story just as soon as they’re finished interviewing the 10 millionth Trump supporter to see how they think he’s doing.

      Any day now.

    • jon

      I read this w/ regard to approval numbers on tuesday… I think it fits well here.

      “33% of Americans approve of how Trump conducted himself, 50% disapprove, and 18% have no opinion. This lines up with many polls about Trump, no matter what the subject. If Trump were to barbeque and eat a live puppy, the poll results would be similar. The actual event being polled doesn’t seem to matter any more. No matter what question the pollster asks, what the respondent hears is: “Do you like Trump?””

  • MrE85

    Consider also that only a small percentage of the eligible voters will vote in the primary, and many thousands more will stay home in November, because the presidential race isn’t on the ballot.

    Is the midterm a referendum on the current administration? Maybe. Maybe not. But as Bob says, don’t count on the polls telling you much until the elections happen.

    • Jerry

      People may be dissatisfied with Trump but loving the job that carpetbagger Jason Lewis is doing. It may be less true than before, but politics is local.

    • plund

      I think if voter registration/vote turnout drives focus on the younger crowd it will help a lot. In the last midterms the turnout for 18-24 yo voters was 20%, and even up to 30-34 yo voters barely got above 30%. And that is Minnesota with a high turnout rate compared to the rest of the country.
      I’m hoping that the Parkland students and any outrage about how low that number is will help drive turnout.

      • MrE85

        Younger voters who don’t vote deserve the country we Olds have given them.

        • plund

          Somewhere in there is a separate discussion about the state of the country we are inheriting, but here is not the place or forum

  • Gary F

    Polls don’t mean much. The sample size and type will always be subject to the biases of the people paying for the poll.

    But it looks good in the headlines and keeps the outrage machine fueled.

    • // s be subject to the biases of the people paying for the poll.

      this is incorrect.

      • Gary F

        So how do you explain The Minnesota Poll?

        • I was going to say you were parroting Bill Cooper’s constant complaining of The Minnesota Poll, which hasnt’ existed, I’d guess, in more than decade in the manner in which Cooper objected.

          There are certainly philosophically aligned polling companies: Rasmussen for the GOP, comes to mind. PPP for the Dems

          but to blanketly says that polls will always be subject to the biasses of the people paying for the poll is simply utter nonsense.

          As a general matter, I it ridiculous that polls are used as predictors in presidential races that are based on an electoral system that we don’t even use.

          • Sonny T

            I don’t want to wade into trouble, but Gary has a point. Polls often seem to be taken and publicized for the benefit (or detriment) of someone.

            Furthermore, the structuring of the poll questions typically contains the pollster’s bias in the form of “How often do you beat your wife?” questions.

            A poll asking if I think “the Russian government has compromising information about President Trump,” is idiotic on its face. It is conspiracy mongering. Certainly this poll had a bias.

            Would we have polled the public asking if Obama was a socialist? Or a secret Muslim? These were identical conspiracies in their day.

          • // A poll asking if I think “the Russian government has compromising information about President Trump,” is idiotic on its face. It is conspiracy mongering. Certainly this poll had a bias.

            There was no such question in the Marist poll.

          • Sonny T

            It’s a direct quote from yesterday’s post, “What’s on NPR News”:

            10 a.m.- 1A with Joshua Johnson
            The Russia investigation update. A poll released this week says 51 percent of American voters think “the Russian government has compromising information about President Trump,” according to Politico. But in the same poll, 70 percent of Republican voters don’t think the Kremlin has compromising information on him.

            Guests: Carol Leonnig, national investig

            Let’s not bury ourselves in this, I assume you have bigger fish with today’s posts. My point is clear I hope. Doesn’t mean everyone hhas to agree

          • Ah, I thought you were talking about the marist poll.

          • jon

            Those happen to be with in the margin of error for trumps approval ratings… and every other poll about trump…..

            50% despise him, 38% love him, 10% undecided…
            70-80% of republicans love him 90+% of democrats hate him…

            The numbers haven’t moved much in months, and they don’t change regardless of the question…How trump is handling, family separation at the border, same results, how trump is handling dressing himself, same results, how trump is handling trade, same results.

            It’s the kind of thing you’d expect from a political system that is build “us v.s them” rather than policy….

    • Sonny T

      well stated

  • Mike Worcester

    Too many years ago to be certain when it was (and believe me, I tried to find video of this), Jeff Greenfield essentially said “stop taking so many polls, stop reading so many polls, stop reporting on so many polls.” It made sense then (the early 1990s), it makes even more sense now.

    Now I tend to lean on just about anything done by the PEW Center as they are quite thorough in their work and methodolgy. I’ll take their conclusions just about any day.

  • lindblomeagles

    Right now, Trump is the least of our problems. There’s enough things going wrong with Trump’s administration (revelations about infidelity, a poor showing with Vladimir Putin, a trade war that has brought about another buyout – this time for farmers, moves to remove Mueller and Rosenstein, which is a sure sign Trump’s campaign benefitted from Russian attacks) that winning a second term in office is probably going to be very difficult for him. The greater concern has to be the makeup in Congress and the Governor’s race here in Minnesota. There are a lot of unknowns on this year’s ticket, and one gigantic loser (Tim Pawlenty). That last time Pawlenty was Governor, job growth stalled, deficits were the norm, school budgets got slashed, and Tim spent the vast majority of his time running for President of the United States. We definitely need to steer the state clear of him. Meanwhile, I don’t know what Republicans in Congress are doing, but it sure seems like they made more noise when Obama was President and Hillary Clinton was running for President. Neither one of these two individuals was anything like the odor of funk coming from Trump and his team of handlers. That this Congress sits quietly when there are all kinds of signs they need to do more to find out what Trump is doing AND what’s good for the country is damning evidence that changes at Washington MUST happen.

    • Frank

      I understand why you’re so upset. The economy just passed 4%…It’s wonderful.

      But if all your information comes from leftist sources, you’re going to think the wheels are coming off…

      Expand your horizons. Talk to other people, not just leftists. And for God’s sake, read news from more places than MPR, the Strib, LA Times or Mother Jones.

      They don’t have anything on Trump; there is no There, there. I know that’s disappointing, but there it is. Give it up.

      Meantime, we’re all doing pretty damn good. We’re gonna be fine.

      • I’m glad the economy is chugging along but to be fair, no legitimate economist is saying it’s expanding at 4%

        2 1/2 seems about right. Like everything else in the economy, short term data is volatile

        • Frank

          First quarter was 2.2%. Second was 4.1

          I’ve read some are suggesting people were spending their tax cuts last quarter, and that may be so. But the economy has been growing steadily since the first quarter of 2016.

          The tax cut boost won’t be as large longer term, but it’s sustainable. If Trump follows through with his infrastructure rhetoric, steady 4% is within reach.

          Our standard of living is on the rise; that’s wonderful news.

          • The economy has been growing for 10 years. We’re almost two years in since the infrastructure idea was floated and there was no follow through, so that’s probably dead, especially with an enormous deficit.

            A lot will depend on what corporations do with their tax breaks. Seeing more stock buybacks than capital investment.

      • lindblomeagles

        Frank, ahem, I don’t read the Strib, the L.A. Times, OR Mother Jones. I do read MPR, the New York Times, and the Washington Post for comparison sake.

        There’s a there there. That’s WHY Rudy Gulianni is always on Fox and Friends putting out Trump’s fires. It’s why Trump has been trying to secretly replace Mueller, and last week, Rod Rosenstein.

        As for the economy, you really don’t want very fast growth because markets correct themselves, and really fast growths often trigger bubbles that burst and really drag the economy down. You want slow, steady growth, a mix of savings and spending, and peace among nations. That’s why the economy did so well between World War II and roughly 1970.

  • plund

    Is it bad that I really want to be polled? In 2016 two survey peoplecalled, but had maxed out on my demographics. But it was interesting hearing my aunt give a poll person a piece of her mind about the quality of a survey.

    • I’ve been called a few times. I don’t participate unless they identify the polling firm. Most of the time, though, it’s push polls.

      • jon

        I do the same… and recently it seems like “we are calling from [garbled]” is doing a lot of the polling…. every time they give the name the phone cuts out…