Partisanship has long been known to affect people's attitudes and votes, but political scientists increasingly suggest that it may also affect people's beliefs about purely factual matters. For example, Republicans seem more likely than Democrats to believe that the deficit rose during the Clinton administration; Democrats seem more likely than Republicans to believe that inflation rose under Reagan. What remains unclear is whether partisan patterns in responses to factual questions actually reflect differing beliefs among partisans or instead reflect a desire to voice support for one party or opposition to another. We report results from a 2x2 survey experiment designed to shed light on this question. All subjects were asked a set of factual questions about politics. Some received financial incentives to answer correctly. Others were told that their answers would be scored and reported back to them. And others were exposed to neither or both of these treatments. We find consistent partisan response patterns across all four conditions, which constitutes the strongest evidence to date that such patterns reflect sincere differences in factual beliefs
