Exclusive: Pollster Explains His Finding That 10 Percent of Trump Supporters Lie on Surveys

Find what drives you at Beaman Auto!
  •  
  •  
  •  
  •  

The lead researcher of the study: “Are Election 2020 Poll Respondents Honest About Their Vote?” told Star News Network there are twice as many “shy voters” among the supporters of President Donald J. Trump than among supporters of former vice president Joseph J. Biden Jr.

“The term ‘shy voter’ has been floating around, so we use it because that is the term, which is closest to how people talk about this,” said Leib Litman, CloudResearch’s Co-CEO and chief research officer. In the United Kingdom, conservative voters lying to pollsters is so common they regularly referred to at “shy Tories.”

Among the results, Litman and his team found that 11.7 percent of Republicans would not report their true opinions about their preferred presidential candidate to telephone pollsters.

This is more than twice the 5.4 percent of Democrats, who responded that they would not give their real preference to telephone pollsters, the study found. Among Independents, 10.5 percent said they were shy about giving their actual preferences to telephone pollsters.

After inquiring by party affiliation, the team asked the same questions of Trump and Biden supporters. In this round, 10.1 percent of Trump supporters said they would be untruthful to phone pollsters roughly twice the 5.1 percent of Biden supporters who said they would be untruthful in a phone survey.

The six leading reasons respondents gave for misleading phone pollsters were:

  • Lack of trust in phone polls as truly being anonymous;
  • Apprehension to associate their phone numbers with recorded responses;
  • Fear that their responses will become public in some manner;
  • Fear of reprisal and related detrimental impact to their financial, social and family lives should their political opinions become publicly known.
  • General dislike of phone polls; and
  • Malicious intent to mislead polls due to general distrust of media and political pundits—although this was a sentiment expressed by only a few of the respondents.

Signs black voters among shy Trump supporters

Litman said as part of his analysis, he and his team members looked to see if there were specific demographics where the shy Trump voters where hiding.

“We looked in the obvious places, like white voters without a college degree, we did see a slightly elevated tendencies there, but not anything that would really jump out at us,” he said.

“We also found, interestingly enough, with African-Americans there was a slightly elevated effect, but again, nothing that we say conclusively,” he said.

Litman, who in addition to his work for CloudResearch, is a professor at New York City’s Lander College for Men, said the main limitation to breaking down the results by race, gender, age and region, is that each subset becomes a significantly smaller sample, which increases the margin of error. “We would really have to over-sample these groups in order to look at them more closely. Once we get that granular, we just can’t say with any level of certainty.”

There is reason to conduct more research into shy black Trump supporters with a larger sample, he said. “There is a lot more to be done and people should be interested looking into the demographics of this effect.”

Challenges polling shy voters

Litman said going into the survey, he suspected that respondents would be more honest to automated calls, rather than live interviewer calls, because they did wanted to impress or not disappoint the human questioner.

It surprised him that this was not confirmed by the study, he said.

“We actually did not find that that’s the case, because people in some ways were more concerned about the automated system, because the call was being recorded and then put in a database, so people were concerned about that,” he said.

Litman said he was not surprised that other pollsters have not studied the shy voters before, because of the very nature of the question at stake.

“It is a little paradoxical to ask people who lie on a poll whether they lie on a different poll,” he said. “It is almost circular.”

The pollster, who earned his doctorate in experimental psychology from the City University of New York, said, “The reason why we approached this issue, though, is because we don’t look at people as liars necessarily, we think people have genuine concerns about anonymity and they, because of those concerns are reluctant to tell people outright, who they are going to vote for.”

One of the main takeaways for political professionals is that there  is not enough information about what kind of magnitude this problem is, he said.

“People should definitely look closely at this issue, because this issue is not something that people are currently correcting for to the extent that it does exist–people need to understand it, measure and correct for it,” he said. “They have no problem telling people about they’re concerned about anonymity.”

2020 pollsters haunted by 2016 failures, Bradley-Effect at play with Trump

Litman’s report comes on the heels of the Aug. 16 article “‘Hidden’ Trump Voters Exist. But How Much Impact Will They Have?” by New York Times reporter Jeremy W. Peters. In the article, Peters spoke to pollsters and political operatives about the phenomenon that still haunts those who predicted former first lady Hillary Clinton would beat the New York City developer in 2016.

David Winston, a GOP pollster, said to Peters that shy Trump voters exist.

“The idea that people lie, it’s an interesting theory, and it’s not like it’s completely off-the-wall,” said David Winston, a pollster who works with congressional Republicans. “But it’s obviously a very complicated thing to try to prove because what do you do? Ask them, ‘Are you lying?’”

Mr. Winston said that many proponents of the theory about hidden Trump voters rely on what is known as the Bradley effect, named after Tom Bradley, the former mayor of Los Angeles who lost the 1982 California governor’s race despite polling consistently ahead of his white opponent.

The so-called Bradley Effect is another one of the ghosts that continue to haunt U.S. political polling. Democrat Bradley was running against Republican C. George Deukmejian Jr. Pundits said the polls were wrong, because white voters were either afraid to admit they were not voting for the black candidate or they were trying impress the poll taker.

In 2016, pollsters knew there was an aversion to admitting one’s support for Trump, but in the mainstream polls, this was dismissed.

In his role as a consultant and crafter of the 2016 Breitbart-Gravis polls, legendary pollster Patrick H. Caddell included attitudinal questions about the economy’s direction or bringing more Syrian refugees into the U.S. When Caddell then cross-tabbed responses of Clinton supporters with Clinton’s policies, he determined that at least a fifth of Clinton’s supporters agreed with Trump and would vote for him.

CloudResearch debunked Center for Disease Control report of rampant bleach drinking

“We conduct a lot of different polls at CloudResearch and we’ve actually explored a number of issues over that last few months that were of issue to us,” he said.

“Just as an example, the last issue we studied—there was a report from the CDC that people were drinking bleach in very high numbers—something like 10 percent of people were drinking bleach and household cleaner and stuff like that—and we suspected that that result was due almost entirely due to biased methodology,” he said.

“We re-ran that study and we found almost nobody was doing that,” he said.

How the shy Trump voter study was conducted

This CloudResearch Study: Are Election 2020 Poll Respondents Honest About Their Vote? study was conducted Aug. 19 through Aug. 27 with online responses from two different sets of 1,000 registered voters, with the sample population evenly divided by Republicans, Democrats and Independents. The results presented are from the first set of respondents. The second set of respondents was matched to Bureau of Census data for current party affiliation percentages, but the results of the second set closely aligned with the first set.

Related posts

Comments