Counterfactual reasoning as a framework for attribution theories.

This model is used as a framework for understanding the use and success of Kelley's analysis of variance model, and is also used to provide an understanding of the fundamental attribution error and actor-observer attributional differences.

Abstract

It is proposed that all of the major attribution theories are based on the use of counterfactual information (which examines whether the event would have occured if the proposed cause had not occurred). However, because counterfactual information is difficult to obtain, 2 proxies are often substituted for it. A model based on these proxies ― covariation data and information regarding alternative explanations ― is proposed. This model is used as a framework for understanding the use and success of Kelley's analysis of variance model; Jones and Davis' correspondent inference theory; Weiner's theory for attributions of success and failure; Jaspars, Hewstone, and Fincham's natural logic model; and Hilton and Slugoski's abnormal conditions focus model. The model is also used to provide an understanding of the fundamental attribution error and actor-observer attributional differences