Where Polls Got It Wrong
AUDIE CORNISH, HOST:
In the run up to the 2020 election, many polls show Joe Biden and Democrats with a significant lead, both nationally and in multiple swing states. As NPR's Danielle Kurtzleben reports, pollsters are now confronting the question of how far off polls were and what caused it.
DANIELLE KURTZLEBEN, BYLINE: Pollsters arguably have even more reason than everyone else to obsess over vote counts right now, and here's why.
PATRICK MURRAY: Until we count all the votes and know exactly how far the polls were off, then we won't really be able to get at why they were off. We need to know the how much before we'll know the why.
KURTZLEBEN: That's Patrick Murray, director of the Monmouth University Polling Institute. This year, polls appear to have been off substantially, both nationally and in some states. That's different from 2016, when national polls were very close to the final presidential result. For now, pollsters like Murray do have some educated guesses as to what caused polls to be off. Here's one important one.
CLIFF ZUKIN: It's safe to say that we don't have enough Republicans in our samples.
KURTZLEBEN: Cliff Zukin is a retired professor of political science with 40 years of election polling experience. The question then is what caused that? It may be the complicated math of polling, Zukin says. Pollsters try to figure out who is likely to vote and what the electorate will look like. So, for example, college graduates tend to respond to polls more than noncollege graduates. So pollsters do math called waiting to try to account for that. Long story short, that math may have been off this year.
ZUKIN: In 2016, that's where the state polls were off, was in how they waited and not having enough less-than-college-educated people and especially males in the sample. You know, it's certainly possible that this time we might find that rural voters were undercounted.
KURTZLEBEN: But this is where it gets more complicated. Pollsters attempted to adjust for this after 2016, explains Courtney Kennedy, director of survey research at the Pew Research Center. They looked at how they weighted their numbers, and they tried hard to make sure they got enough Republican respondents. And yet...
COURTNEY KENNEDY: And yet we had the results that we did. So what does that mean? That means that the Republicans taking the polls may not have been good proxies for all of the Republicans.
KURTZLEBEN: But the fact is, polls don't seem perennially wrong. For example, they did a good job in the 2018 midterms. And so, Murray says, there may just be something about Trump.
MURRAY: We seem to have this phenomenon that whenever Donald Trump is on the ballot and that's who we're asking about, that polling just seems to be off.
KURTZLEBEN: Could it be shy Trump voters? There wasn't good evidence of this in 2016, and experts NPR spoke to are split about 2020. Answering if and why Trump throws polls off is vitally important to pollsters. Kennedy explains.
KENNEDY: If it's really because of his unique ability to turn out voters who are not easily modeled, not easily identified, not easily reachable, is it that or is it something more long lasting and something more fundamental to how surveys are being done these days?
KURTZLEBEN: If it's the latter, that's scary. It means pollsters have to ferret out that fundamental problem and fix it. On the other hand, if it's something about Trump...
KENNEDY: In the future, when there's no Donald Trump on the ballot, then maybe we go back to more normal times.
KURTZLEBEN: But that's not a satisfying answer. And, of course, another politician who confuses the polls could always come along.
Danielle Kurtzleben, NPR News.
(SOUNDBITE OF MUSIC) Transcript provided by NPR, Copyright NPR.