The BDN Editorial Board operates independently from the newsroom, and does not set policies or contribute to reporting or editing articles elsewhere in the newspaper or on bangordailynews.com.
Similar to polling misses in 2016, Democratic presidential candidate Joe Biden held large polling advantages nationally and in swing states that have not materialized. Here in Maine, every independent and public poll from 2020 in the U.S. Senate race had Democrat Sara Gideon leading Republican Sen. Susan Collins, yet Collins won decisively.
It has plenty of people questioning the future of political polling — even pollsters themselves.
“The political polling profession is done,” Republican pollster Frank Luntz told Axios. “It is devastating for my industry.”
Strong words, and strong self-criticism that his industry deserves. But to borrow from Mark Twain, we have to think that predictions of its demise have been exaggerated. At least for now.
As University of Maine associate political science professor Rob Glover pointed out in an interview with the BDN editorial board, what’s the alternative?
We don’t see any obvious, proven replacements for measuring public opinion over the course of a campaign. Standing outside a grocery store and asking people who they’re going to vote for can get a reporter some good quotes, but it’s not as if colorful anecdotes can provide more consistent, precise windows into overall voter preference.
Glover said that this “seems kind of like a moment of reckoning for polling,” especially at the state level, with a need for an “internal accounting” of how polls are conducted. That could include taking a hard look at how polling data is collected and how different forms of bias factor into answers from respondents. But that is supposedly what happened after the 2016 election. We share Glover’s concern that we’re now having a similar conversation in 2020. The industry can only have so many reckonings before public trust slips away.
“I think there’s going to be demand for changes that make these more reliable,” Glover added.
The future of polling depends on internal adjustments, but there’s also an external imperative from those of us who engage with polling data — namely the media — to reassess how we analyze and amplify it.
“I have never been a big fan of election polling, because I don’t really see what the purpose of it is for citizens,” UMaine political science professor Amy Fried, a BDN columnist, told the editorial board. She said that “too much time and too much space” is given to polling related to political campaigns rather than focusing on other aspects of a contest like policy differences between candidates.
“I say give them less attention,” Fried said about political polls. That’s an adjustment that falls to us in the media, and to the general public.
Patience has been paramount in the presidential election. That applies to the much-needed assessment of the polling industry as well. It will take time to study the 2020 polls, both those that whiffed and those that performed well. And as more votes are counted, some of the polls showing Biden up big nationally in the popular vote may not prove as off base as they seemed earlier in the week.
“If this projection holds up – and it’s clearly trending this way – what we’re witnessing is a slow-motion landslide in the national vote % that’s in the Reagan over Carter territory,” UMaine journalism and communications associate professor Michael Socolow tweeted Thursday morning. “That’s the story. Not that the polls were ‘wrong.’”
No matter the outcome in the presidential race, we know that polling was off in certain swing states and in statewide races like Maine’s U.S. Senate contest. Dan Shea, a government professor at Colby College who supervised four polls during this election cycle, spoke with both the BDN and Portland Press Herald after the election.
“We missed the mark, but I think it’s important to remember, a poll is a snapshot of a particular point in time, campaigns evolve and turnout matters,” Shea told the Press Herald, in reference to a survey released in late October that had Gideon up 3 points (Collins won by roughly 9 points). “I think our poll, as well as others across the country, failed to capture the passion of rural voters.”
The same can and should be extended to members of the media as well. Despite years, if not decades, of trying to better reach rural America, our industry is failing to capture and understand the realities of a big portion of the electorate. Pollsters are in serious need of reflection and adjustment, surely. But they aren’t the only ones.