The other way around: Effects of causal attributions on estimates of consensus, distinctiveness, and consistency

Abstract

It was hypothesized that the attributed cause of a given person's behavior will affect inferences about its generalizability over persons (consensus), stimuli (distinctiveness), and circumstances (consistency). Moreover, these effects were expected to parallel the effects of consensus, distinctiveness, and consistency information on causal attributions. Experiment 1 provided support for these predictions but also showed that attribution affected consensus judgments less than it affected judgments of distinctiveness and consistency, particularly when consensus was not the first characteristic estimated. Using a different set of stimulus materials and a different manipulation of attribution, Experiments 2 and 3 provided further evidence for the effects of attribution on inferences of consensus information. Experiment 3 indicated that the false consensus effect-actors' tendency to assume that the majority of people share their behavior-may be due to actors' tendency to attribute their behavior to situational factors. Implications of the present studies for biased estimates of consensus and the use of consensus and attribution as mediating variables are discussed. © 1979.