The false consensus effect and overconfidence: Flaws in judgment or flaws in how we study judgment?

The false consensus effect and the overconfidence in judgment effect are often referred to as established ''facts'' of psychology. At present, the empirical support for these effects is inadequate and logically flawed. We argue that, absent other information, using one's own response to a situation as an observation of size 1 could be an effective use of data and could lead to an increase in the accuracy of predicting others' behavior. In an empirical examination of such use, we find a robust positive correlation between the degree to which people believe that a majority of others are like them and their accuracy in predicting those others' responses, whether this correlation is evaluated within items across people, across items within people, or across items across people. In addition, we show that the finding of overconfidence injudgment follows analytically from the functional relationship used to demonstrate it, a ''finding'' that is easily reversed by considering the inverse relationship. Specifically, we argue that regression effects account for the evidence cited in support of overconfidence. While not definitive, our empirical findings call into the question the acceptance of these two effects-as commonly defined-as facts. (C) 1996 Academic Press, Inc.