Pollsters ask the public all sorts of questions about their political beliefs. But what does the public actually think? Is there any reason to assume that people are truthfully answering pollsters’ questions?
That may seem like a strange question. Why would people lie to pollsters? I’m not sure, but there is evidence that they lie about their beliefs. Reason Magazine has an excellent article by ronald bailey, discussing the tribal nature of views expressed on issues with political implications:
A 2015 study in the Quarterly Journal of Political Science attempted to distinguish partisan cheerleading from genuine partisan divergence. Northwestern University political scientist John Bullock and his colleagues found that offering small payments for correct and “don’t know” answers to politically significant questions narrowed the party gap between Republicans and Democrats by about 80 percent.
“To the extent that factual beliefs are determined by partisanship, paying partisans for the correct answer should not affect their answers to factual questions. But it does,” they state. “We find that even modest payments significantly reduce the observed differences between Democrats and Republicans, suggesting that Democrats and Republicans do not have wildly different views on many important facts.”
The article cites another academic study that reported some truly amazing results after people were shown images of the modest crowd at Trump’s inauguration and the large crowd at Obama’s inauguration:
But do partisans really see other things? Perhaps they mainly cheer for their team rather than assert actual beliefs. This is the thesis that University of Nottingham philosopher Michael Hannon explored in a 2020 paper for Political Epistemology. He points to a survey of nearly 1,400 Americans conducted in January 2017. The researchers showed half of the respondents photos, simply labeled A and B, of the crowds on the National Mall during the inauguration of Barack Obama in 2009 and the inauguration of Donald Trump in 2017. They were asked which photo represented the crowd represents for each President. Forty-one percent of Trump voters said the photo with the larger crowd was of Trump’s inauguration, which was actually the picture of Obama’s inauguration. Only 8 percent of Hillary Clinton voters picked the wrong photo. The researchers argue that it’s likely that Trump voters chose the photo with the larger crowd to express their party allegiance and show their support for him.
More revealingly, the researchers asked the other half of respondents which photo depicted the larger crowd. One answer was clearly correct. But Trump voters were seven times more likely (15 percent) than Clinton voters (2 percent) to claim that the much less populous Trump inauguration photo had more people. Remarkably, 26 percent of college-educated Trump voters answered incorrectly. “When a Republican says there are more people in Trump’s inaugural photo, they’re not really contradicting those who say otherwise. They’re just cheerleaders,” argues Hannon. “People simply make claims about factual issues to signal their affiliation with a particular ideological community.”
Former Econlog blogger Bryan Caplan occasionally bets people on certain factual issues because he thinks people have less incentive to engage in wishful thinking when it comes to money. These academic studies support Bryan’s contention that people don’t always believe what they say they believe.
Robin Hanson has argued that some policy decisions should be guided by prediction markets, and I have strongly advocated using the NGDP futures markets to guide monetary policy. Public policy is likely to be more effective when based on views that prove costly when wrong.
hp I wrote about it in a recent post this story:
In 2006, lawmakers passed legislation banning almost all abortions, which Gov. Mike Rounds signed into law. It sparked a brutal campaign that became the dominant issue in a busy election year that saw a governor’s race and 10 other voting issues at stake. Voters rejected the ban by 56% to 44%.
Anti-abortion activists decided to start another run in 2008 and gather enough signatures to put abortion back on the ballot. The main difference between the two measures was that the 2008 effort included exceptions for rape and maternal health. Opponents thought the lack of exceptions in 2006 had doomed their efforts.
You were wrong. The 2008 vote was almost identical to that of 2006, with 55% opposing the measure.
I suspect they were wrong because they took poll results seriously, which suggest a wide range of views on abortion. If you give people 4 or 5 options to choose from, the answers will be spread across those options. People don’t like to sound extreme or unreasonable. But in a binary ebb and flow, it turns out that people are simply pro-life or pro-choice, with very little in between.
PPS. North Dakota had a similar referendum with a similar result.
Comments are closed.