Once again, reality has humiliated the polling industry.
Far from the Democratic landslide that the RealClearPolitics polling average and prognosticators like Nate Silver at FiveThirtyEight anticipated, Tuesday’s presidential election was agonizingly tight and Republicans gained seats in Congress and state legislatures.
Independent pollster Richard Baris is also critical of his colleagues.
“They hurt us bad this election,” Baris told The Post. “This industry is dominated by left-wingers. And a big, big problem is they’re trying to profile the voting behavior of people they don’t understand and may even despise.”
Polling has certainly gotten tougher since George Gallup’s day. As recently as 1997, more than a third of Americans routinely agreed to participate in pollsters’ surveys. By last year, that response rate had plummeted to 6 percent, a Pew Research Center study found.
Yet Baris’ Big Data Poll, along with conservative-leaning pollsters like Trafalgar and Susquehanna, managed to come closer to the Election Day results than the leading media outfits did.
Baris’ final poll of Florida gave Trump a 2-point lead — one percentage point shy of the president’s 3-point victory there. In contrast, the New York Times/Siena poll predicted a 3-point win for Biden in the Sunshine State. Quinnipiac found a 6-point Biden edge.
Baris places the blame, in part, on groupthink within the industry.
“I think they bully each other,” he said. “They herd. That’s when you start to mirror other pollsters because you’re afraid that Nate Silver or CNN is going to call you an outlier.”
Another enemy of accurate polling is time.
“You can’t reach a truly representative group of people in a day,” he said. “But many pollsters are under the gun from their media clients. They want a horse race number, and they want it now.”
Baris avoided that pressure by crowd-funding his battleground polls. His Twitter followers and podcast listeners ponied up the cash to conduct them — and chose the states he surveyed.
In return, they received above-average transparency, with access to Baris’ poll questions, crosstab results, and maps showing just where in each state his respondents were found.
“You have to look not just at who you poll, but where you poll,” he said. “You can’t say that a working-class man in Milwaukee has the same opinions as a working-class man in rural Polk County.”
But rural and working-class voters tend to be much more resistant to pollsters’ entreaties — and that, Baris suspects, compounded his competitors’ failure to anticipate Trump’s support this year.
“The way they’re polling, they are reaching voters that skew too urban,” he said. “In that case, your Republican sample will be stacked with the John Kasich Republicans, the Bill Kristol Republicans — and that’s not the Republican Party that gave the presidency to Donald Trump.”
Highly educated voters are often eager to answer a pollster’s call, Baris finds, so it’s easy for time-pressed pollsters to oversample them.
“They are dying to tell you what they think. They want to enlighten you,” he said. “The other people just want to have their dinner and go to bed. It takes more finesse and more time to get to them.”
Baris also designed his poll to uncover secret reservoirs of Trump support — the under-the-radar “shy Trump voters” — by measuring what pollsters call social desirability bias.
In a six-question sequence, he asked whether respondents feel comfortable sharing political opinions with family, friends, neighbors, co-workers, strangers and pollsters.
“You would be shocked at how uncomfortable people said they feel about talking to a pollster,” he said. In Florida, for example, it came to 33 percent of all respondents.
“The most uncomfortable groups were suburban women and black men aged 30 to 65,” Baris said. “Those are the shy Trump voters. And in the exit polls, Trump did unexpectedly well with those groups.”
Longtime Republican pollster Frank Luntz’s focus-groups work this year confirmed Baris’ findings.
“The fact is, Trump people don’t like being interviewed by pollsters,” Luntz said. “They tell me they would never consider talking to a pollster, because that would help the pollster manipulate them, and they are so wary of being manipulated.”
A focus-group setting, Luntz said, makes it possible to build trust with participants — a benefit that gave him insight into undecided voters’ thought processes as Election Day approached.
“It wasn’t until the first debate that I realized . . . they were evaluating the two candidates under a different metric I’ve never seen before,” he said. “It wasn’t, ‘Do I agree with Trump’s agenda or Biden’s agenda?’ It was, ‘Can I tolerate Trump’s persona, which I don’t like, or do I take a risk with Joe Biden’s agenda that I do not know?’ ” he said. “They had completely different factors that they were choosing between with each candidate.”
Pollsters, he believes, should make more of an effort to empathize with these leery voters.
“It requires you to say to them, ‘I respect you, I appreciate you, I value you — and your opinion will have an impact,’ ” he said. “It’s about humility. And too many pollsters approach this with a sense of arrogance.”