Based on previous research findings, which have found significant relationships between the hypothesized components of self-efficacy judgments for avoiding heavy drinking and actual levels of alcohol consumption in college students, the present study examined the role of anxious affect on self-efficacy for avoiding heavy drinking in college students. Specifically, it was hypothesized that heavy drinking subjects who were either high or low on dispositional social anxiety would report lower levels of self-efficacy for avoiding heavy drinking in response to feelings of social anxiety induced via anticipation of a self-disclosing heterosocial interaction. In addition, it was hypothesized that subjects who were high in dispositional social anxiety would report greater alcohol expectancies of social facilitation and tension reduction than subjects who were low in dispositional social anxiety. Subjects ill = 440) completed measures of alcohol expectancies, social anxiety, and quantity/frequency of alcohol use to identify heavy drinkers who were also either high or low in dispositional social anxiety. Identified subjects (n = 73) were then assigned to either an anxiety inducing or non-anxiety inducing condition and asked to complete measures of state anxiety and self-efficacy for avoiding heavy drinking in positive, negative, and socially anxious situations. Results failed to find a significant main effect for anxiety manipulation on self-efficacy for avoiding heavy drinking, but a main effect of ii dispositional social anxiousness was found on subjects' self-efficacy for avoiding heavy drinking in negative and socially anxious situations. The results of the study also failed to find significant relationships between alcohol expectancies of tension reduction and dispositional social anxiousness, but did find a significant positive relationship between social anxiety and alcohol expectancies of social facilitation. Lastly, exploratory analyses found significant relationships between positive and negative alcohol expectancies and self-efficacy for avoiding heavy drinking in socially anxious situations. Moreover, significant relationships were also found between alcohol expectancies and the frequency of drinking episodes and actual quantity consumed. Results are discussed in terms of the relationship between affect, self-efficacy, and outcome expectancies and implications for alcohol intervention programs with college students.