A researcher who published a study a few weeks before the presidential election stating the country’s population is more conservative than most believe isn’t as surprised as most that Donald Trump was elected president.
“I am a little shocked, yes, but not completely,” said Paul D. Tieger, who published the study Born Liberal or Conservative, which suggests that many people like Trump because many people “are like Trump.”
When Tieger released his study, he said those who support Trump have inborn personality types — which influence their perceptions and values — “more similar to his than to Hillary Clinton’s.
“People like the fact that he speaks his mind,” Tieger said on Thursday. “That is the most common comment you hear.”
Trump won the presidency by winning 279 Electoral College votes, 7 more than the 270 needed. Clinton won 228. Clinton actually won more votes than Trump, nationwide, getting 47.7 percent to Trump’s 47.5 percent. The rest of the votes went to minor party candidates.
In the days leading up to the election, most national polls showed Clinton having anywhere from a 3- to 6-percent lead.
Even in states that went for Clinton, like Connecticut, “more than 40 percent (actually 41 percent) voted for Trump,” Tieger said. “That’s a lot of votes.”
Asked why so many of the national polls got the election wrong, Tieger had a simple answer: “They didn’t ask the right questions.”
He said they ask gender and ethnicity questions when they should be asking about their values and how they see the world.
“Asking more questions about who they are as people, what their background is. That’s what polling often doesn’t do,” Tieger said. And those types of more personal questions, Tieger added, translates into more accurate polling results.
One recognized pollster who bristled at the notion that the polls called the presidential election wrong was Peter A. Brown, assistant director of the well-known Quinnipiac University Polling Institute.
Brown said the media should shoulder some of the blame for saying polls got it wrong.
“Let me ask you a question,” Brown said. “If you do a poll and say a candidate is going to win an election by 2 points and she loses by 2 points and the margin of error is 4 points, was the poll wrong?
“Let me answer that question for you,” continued Brown, “the answer is ‘no’.”
Brown said too many times the news media ignore the margin of error explanation in press releases that are put out on polling data — or puts that disclaimer as the last paragraph of the story as an afterthought.
Brown added that another complicating factor is “increasing number of national polls done these days and the myriad of methodologies used” to do those polls.
For instance, he explained, some polls use computerized phone call interviewing, which he said is much cheaper in cost but also provides less accurate results.
The Quinnipiac Poll does all its interviews with live phone calls. Brown added another complicating factor in polling these days is that many who are called only have cell phones, and are on data payment plans, meaning they aren’t as likely to spend as much time answering survey questions.
But Tieger insists there is more at play, saying his study showed that personality type — and not demographics — may be the X factor that decided the election.
“People who are predisposed (by their personality type) to be conservative represent about 75 percent of the U.S. population, compared with only about 25 percent for those who are more liberal,” Tieger said. “Although by no means a traditional conservative, Trump has a much bigger pool who are naturally hard-wired to receive his message.”
The goal of his study, Tieger said, was “to determine if there are inborn personality type characteristics that influence voters’ behavior, and how they may impact the 2016 presidential election.”
Tieger said his study showed the link between liberalism, conservatism, and educational level: the more education respondents indicated, the more they were likely to be liberal; the less education, the more conservative.
Of the 2,700 who participated in the study, 11.8 percent had a high school degree or equivalent, 20.5 percent had some college but no degree, 9.6 percent an associate degree, 33.4 percent a bachelor degree, and 23.7 percent a graduate degree.
Tieger said traditional pollsters should look to the results of his study and learn.
“Polls which rely on demographic factors in its analysis can yield inaccurate results,” Tieger said, “because type often trumps (no pun intended) demographics. For example, two same-sex fraternal twins born moments apart, raised by the same parents in the same home, will almost always have very different personality types, which affect their attitudes, values, talents . . . and voting behavior.
“But if pollsters overemphasize the importance of demographics, they are likely to inaccurately predict voter behavior,” Tieger said.