This study aimed to test the diathesis-stress and mediation components of the hopelessness model of depression in an academic context. University students completed measures of attributional style and depression at the beginning of the academic year and again at the end of the academic year. In addition, approximately 4 months after the initial testing session, students were asked for their particular attributions for their mid-semester examination performance and completed the depression measure again. A synchronous, but not prospective, relationship was found between attributional style and depression. However, the diathesis-stress component of the hopelessness model received strong support as there was a significant interaction between attributional style and grade satisfaction in the prediction of students' more enduring depressive reaction. Here, unlike previous studies, the longer term depressive reaction was assessed several months following the experience of a negative life event. This result provides support for the model's contention that a depressogenic attributional style is a vulnerability factor for the development of an enduring depressive reaction. In contrast, no support whatsoever was received for the model's claim that depressive reactions are mediated by particular attributions.