Non diagnostic information and the effects of context evaluation in the moderation of False Consensus 1The author expresses his appreciation to Stephan Kraitsowits and anonymous reviewers for their comments on earlier versions of this article.

Abstract

The hypothesis that non-diagnostic information about consensus may reduce False Consensus (FC) was tested in a 2 × 2 × 2 between-subjects design. Eighty nine participants were told that they were to be assessed on their judgement expertise in impression formation. First, half of the participants were led to believe that, during a vigilance task they had performed, they had been exposed to consensus information about the opinion of their peers towards a specific target person. The other half received no information. Subjects had then to complete two questionnaires. An attitude questionnaire assessed subjects' attitude towards the target (favourable or not favourable), deduced from their predictions of the target's answers. In a consensus questionnaire the subjects had to make percentage estimates. The order for completing the two questionnaires was counterbalanced. Finally, subjects were asked to indicate their personal confidence as to their consensus estimates. Results provided direct evidence for a decrease in FC when subjects were informed about their exposure to consensus information, and an increase in FC when they were not informed. Decrease in FC is linked to situational characteristics which cause subjects to question the reliability of their estimates and to take into account data that are not solely related to the self.