We use the method of situational decomposition in an attempt to identify the causal effect of alcohol intoxication on adolescent sexual intercourse and contraception use. This approach assumes that the relationship between frequency of alcohol intoxication and behavior "while sober" is entirely spurious. With this in mind, the "total association" between drinking and behavior can be decomposed into spurious and "non-spurious" (i.e., causal) components. The larger the magnitude of the total association relative to the spurious association, the larger the influence of alcohol on behavior. This method was employed to analyze data from the first wave of National Longitudinal Study of Adolescent to Adult Health (Add Health) in-home interview (N = 20,743). The results suggest that almost all of the relationship between alcohol and sexual intercourse is spurious. On the contrary, alcohol intoxication was found to reduce the use of contraceptives during sexual intercourse, especially among boys.