One might wonder, upon digesting any sort of political discourse these days, whether the late senator and sociologist Daniel Patrick Moynihan, famous for saying “everyone is entitled to his own opinion, but not to his own facts,” is turning in his grave. Yes, whether it’s the latest Republican debates or a coffeehouse conversation with your hippie friend Barack, it feels a bit like truth has taken a backseat to, well, something else. But there is hope for the truth: Two new studies published in the Quarterly Journal of Political Science suggest that people are more knowledgable—or at least more aware of their ignorance—than they may at first appear.
Americans are famous for not knowing much about politics while at the same time spouting off dubious information with a discernible political slant. For example, only 30 percent of “strong Democrats”—those who identify most firmly with the Democratic Party—surveyed in 1988 reported that unemployment had gone down under President Reagan, according to study authors Markus Prior, Gaurav Sood, and Kabir Khanna. Republicans were much closer to the truth: Unemployment had indeed gone down, and 80 percent of “strong Republicans” said it was so.
In other words, if you ask a factual question about politics, you should expect different answers from Democrats and Republicans. What’s less well understood is whether their answers diverge because they actually believe what they’re saying. Indeed, partisan-sounding assertions “may reflect the expressive value of making statements that portray one’s party in a favorable light…. Partisan divergence in surveys may therefore measure the joy of partisan ‘cheerleading’ rather than sincere differences in beliefs about the truth,” political scientists John Bullock, Alan Gerber, Seth Hill, and Greg Huber write in one of the new studies.
A surprising number of people are likely aware of their own ignorance.
If that’s true—if people know the truth but say something else as a sort of “partisan cheerleading”—there ought to be an easy way to find out: Pay people to tell the truth, and see if that brings Democrats’ and Republicans’ answers closer together. That’s precisely what Bullock and his colleagues did. In the first of two experiments, they asked 419 Democrats and Republicans to take a short quiz on political facts—for example, how many soldiers had been killed in Iraq up to that point? (The original survey was conducted in 2008.) Most importantly, half the participants had an incentive to answer questions correctly: for every correct response, they would get one entry in a drawing for a $200 gift card.
That incentive had a striking effect. Though Democrats and Republicans in the gift-card group still gave different answers to the questions, the divergence in the answers was less than half that of those in the control group, which had no incentive to answer questions correctly.
A second experiment confirmed those results and also showed that people were willing to acknowledge their ignorance in exchange for money. This time, 795 people participated: some were paid for correct responses, a second group was paid for correct responses and for each time they stated they didn’t know the answer, and a control group was given no such incentives.
Simply paying people for correct answers closed the gap between Democrats and Republicans by 60 percent—and paying people for their “don’t know” responses cut the gap by another 20 percent. All told, 48 percent of participants responded with “don’t know,” which means a surprising number of people are likely aware of their own ignorance.
Given those results, you might think that, well, money talks. But Prior, Sood, and Khanna reached a different, rather surprising conclusion: Appealing to Democrats and Republicans to answer questions as best they can improves their answers almost as much as a couple bucks does. Similar to Bullock’s team, Prior and his colleagues offered one-third of their 1,002 participants two dollars for each correct answer on a five-item quiz about economic facts, while one-third simply answered the questions without any incentives to get the answers right. The remainder saw a novel appeal that read, in part:
As you probably know, the government gathers a lot of statistical information about the economy. We are interested in learning whether this information is finding its way to the general public…. In order for your answers to be most helpful to us, it is really important that you answer these questions as accurately as you can.
In other words, the only incentive those participants had to get the answers right—as opposed to engaging in a little cheerleading—was the joy they’d get out of helping researchers.
Remarkably, that was enough to make quite a difference. While actual money had a slightly larger effect, simply appealing to people to try to get the answers right in the name of science and public policy substantially reduced the differences between Democrats’ and Republicans’ answers—and improved their accuracy overall as well.
“The demonstration that citizens have the capacity to perceive reality in a less partisan manner than previously thought is important in and of itself,” Prior, Sood, and Khanna write. That you don’t need money to make a difference could be the icing on the cake—or at least the silver lining in a world where truth seems to matter less and less.
