Wishful thinking among voters: Motivational and cognitive influences

About 1,450 voters in the 1993 mayorial election in Jerusalem made predictions about election outcomes and stated their preference for one of the two candidates. Strong wishful thinking effects were found, predictions varying in a linear trend as a function of the direction and intensity of preference. Half of the respondents were promised a substantial monetary reward if their predictions would be accurate (motivational remedy). Knowledge about the results of the most recent polls was tested, and the predictions made by accurate poll respondents (14.7 percent of the sample) were compared to those of non-accurate poll respondents (cognitive remedy). It was hypothesized that both remedies would reduce wishful thinking (i.e. reduce the differences in prediction among groups differing in preference). Significant interaction effects indicated that both remedies reduced wishful thinking somewhat. However, these effects were of very small magnitude compared to the high magnitude of the wishful thinking effects, and the overall intensity of wishful thinking remained unchanged. Differences between level of significance and effect magnitudes were discussed, focusing on implications for theoretical versus applied social research. It was also found that non-accurate poll respondents demonstrated a wishful thinking-like effect in their reported memory of the results of the polls, which were made public just one or two days previously.