Can accurate knowledge reduce wishful thinking in voters' predictions of election outcomes?

About 3, 000 Israeli voters were asked to predict the outcomes of the 1992 general election and to state their political preference. Political science students were found to possess more accurate knowledge than education students about some outcomes of the previous (1988) election, but the predictions made by both groups varied as a function of their preferences, indicating a wishful thinking effect. Wishful thinking effects of the same magnitude were found for groups differing in the accuracy of their knowledge about the outcomes of the previous election and for respondents who had been provided with partial or full base-rate information about the outcomes of the previous election. Thus, accurate knowledge did not reduce the effects of wishful preferences on predictions. Respondents' predictions differed from the results of public opinion polls published at the same time in the Israeli printed media. The results were more compatible with a motivational than with a purely cognitive interpretation.