Abstract
Pillow, West, and Reich, using path analysis in 1991, were unable to confirm the causal model predicted from the reformulated learned helplessness theory of Abramson, Seligman, and Teasdale which links the dimensions of attributional style with self-esteem and depression. Because their failure to confirm the model may have been due to their using normal subjects instead of psychiatric patients, the model was retested in the present study on psychiatric patients, many of whom had been diagnosed as depressed. Although the Abramson, et al. model was once again not confirmed, neither was the alternative model proposed by Pillow, et al. The model that fitted the data best in this study differed from both of these models and indicated that all three attributional dimensions affect depression solely through the mediation of self-esteem.