RELATION OF ALCOHOL EXPECTANCIES TO CHANGES IN PROBLEM DRINKING AMONGCOLLEGE-STUDENTS

Objective: To evaluate the relation between alcohol expectancies and p roblem drinking during 3 years of college. Design: Cohort with 3-year follow-up. Setting: Private university campus. Participants: A random sample of 260 students entered a longitudinal study of alcohol use at the beginning of their freshman year. One hundred eighty-four students completed follow-up measures at the end of their junior year. Respond ents were 90% white, with a mean (+/-SD) age of 17.9+/-0.5 years. Inte rventions: None. Main Outcome Measures: A standardized measure of expe ctations and subjective evaluations of outcomes associated with drinki ng, quantity and frequency of alcohol use, and a composite measure of alcohol-related problems. Results: Students were divided into nondrink ing, low-risk, and high-risk groups for problem drinking. A repeated-m easures multivariate analysis of variance indicated significant intera ction effects for risk group by expectancy scale (P<.009) and for expe ctancy scale by time (P<.001). The three risk groups differed signific antly from each other on positive outcome expectations at entry into c ollege and positive expectations and negative outcome evaluations at t he end of the junior year. Students who became problem drinkers during college had significantly higher positive outcome expectation scores at both times and developed less concern for negative outcomes by the end of their junior year. The few initial problem drinkers who moderat ed their drinking during college demonstrated an increased concern for negative outcomes by their junior year. Conclusions: Alcohol expectan cies are associated with differing patterns of alcohol use and are lon gitudinally related to subsequent changes in alcohol use and problem d rinking.