A Computerized Demonstration of the False Consensus Effect

Abstract

Students in an advanced undergraduate laboratory course on social cognition replicated an experiment on the false consensus effect (Krueger & Clement, 1994). Interacting with a computer program, students viewed 40 statements. For each statement, they indicated whether they personally endorsed it, estimated the proportion of the population that would endorse it, and rated its social desirability. Half the students received feedback on the actual consensus in the population after making each consensus estimate, and the remaining students did not. Students analyzed data using a spreadsheet program. They found the traditional false consensus effect as item endorsers gave higher consensus estimates than did nonendorsers. They also found reliable within-subjects correlations between item endorsements and estimation errors, demonstrating the truly false consensus effect. Students also learned that feedback about the actual consensus does not reduce bias.