The present dissertation was conducted to gain a better theoretical understanding of the contribution of outcome expectancies to male undergraduate drinking behaviour. Two assumptions important to the construct validity of outcome expectancies were investigated: (1) outcome expectancies are situation-specific and (2) outcome expectancies are different from attitudes toward drinking.;In each of the four studies conducted in the first series of studies, environmental setting (i.e., bar vs. lab) was manipulated and a second variable was manipulated through instructional set. Specifically, expected beverage content (i.e., beer vs. soda pop), social context (i.e., alone, with friends, no explicit instructions), and level of intoxication (i.e., early vs. latter) was manipulated in Experiments 1, 2, and 3 respectively. In Experiments 1 through 3, "weighted" outcome expectancies (i.e., likelihood x valence ratings) were assessed using a modified version of the Alcohol Effects Questionnaire, REQ-Revised (Rohsenow, 1983). Experiment 4 was a partial replication of Experiment 3, however, a modified version of the Alcohol Effects Scale, AES-Revised (Southwick, Steele, Marlatt, & Lindell, 1981) was administered. Evidence for the situational-specificity hypothesis was obtained only in Experiment 4 in which the AES-Revised was employed.;In Experiment 5, the contribution of outcome expectancies (as assessed by both the REQ-Revised and the AES-Revised) to the Theory of Reasoned Action-TRA (Ajzen & Fishbein, 1980) and the Theory of Planned Behaviour-TPB (Ajzen, 1988) was investigated. Overall, the TPB was superior to the TRA in predicting self-report excessive consumption episodes among male undergraduates. Outcome expectancies, as assessed by the AES-Revised, did not make a unique contribution to the TPB. In contrast, weighted expectancies for Enhanced Sexual Functioning (as assessed by the REQ-Revised) improved the prediction of self-report excessive monthly drinking episodes, over and above components of the TPB.;Three tentative conclusions are drawn. First, two types of outcome expectancies may exist: generalized and situation-specific. Secondly, investigations that examine the sole contribution of outcome expectancies to drinking behaviour may be too narrow in their approach. Finally, the TPB is a useful theoretical framework to explore both the construct validity of outcome expectancies and how these processes contribute to drinking behaviour.