(PhysOrg.com) -- If your political candidate of choice falls behind in the polls, will you lose faith in his ability to win? Probably not. A new study from Northwestern University suggests that people tend to believe that their preferred candidate will win an election, no matter what the polls predict.
The study was published Feb. 21 in the Proceedings of the National Academy of Sciences journal.
People thought their preferred candidate had a higher chance of winning, in every election, no matter in which state they live, no matter who was running, no matter which political party, said Charles F. Manski, co-author of the paper. This is one of the strongest empirical regularities I've ever seen.
Manski is the Board of Trustees Professor in Economics in the Weinberg College of Arts and Sciences and a Faculty Fellow in the Institute for Policy Research at Northwestern. Adeline Delavande from the Institute for Social and Economic Research at the University of Essex is a co-author of the paper.
It appears that Americans, despite having access to the same publicly available information, nevertheless inhabit disparate perceptual worlds, Manski said.
For example, someone who strongly supports a Democrat candidate over a Republican will give a 20 to 30 percent higher chance, on average, that the Democrat would win the election than would someone who strongly supports the Republican.
Manski said that the findings are consistent with a psychological phenomenon called the false consensus effect, in which people project their own preferences onto others.
Previous studies on the false consensus effect have focused on topics in which there exists little or no public information about social preferences. In presidential and statewide elections there is considerable common knowledge of social preferences from poll data, Manski said.
The authors used data from the American Life Panel, an online survey that has been administered to several thousand adult Americans by the RAND Corporation, a policy research nonprofit. They analyzed survey responses collected around the 2008 U.S. presidential election and 2010 senatorial and gubernatorial state elections.
Regardless of gender and level of schooling, people tended to favor the electoral chances of their preferred candidate. The effect was equally strong among white and black respondents. Also, when individuals changed their candidate preferences over time, their expectations of election outcomes changed similarly.
Many researchers are trying to understand why people vote and how they vote, Manski said. The false consensus effect might influence voting behavior.
Explore further: New 'Surveyman' software promises to revolutionize survey design and accuracy