Abstract
To test predictions from Kelley's (1967) attribution model, subjects were presented with stimulus items consisting of a behavior description and high or low consensus and distinctiveness information. In Study I 59 subjects chose among four causal alternatives to explain behavior; in Study II free-response explanations were solicited from 49 other subjects. For Study I person attribution was most frequent given low consensus and low distinctiveness information. In Study II person attribution was frequent under low consensus and environmental attribution under high consensus, but no effect was found for distinctiveness information. It is suggested that observers employ consensus information alone when seeking a sufficient cause to explain behavior; when refined causal discriminations are required, observers employ both consensus and distinctiveness information.
Get full access to this article
View all access options for this article.