Submitted to: Alcohol
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 8/2/2004
Publication Date: 1/1/2005
Citation: Power, T.G., Stewart, C.D., Hughes, S.O., Arbona, C. 2005. Predicting patterns of adolescent alcohol use: a longitudinal study. Journal of Studies on Alcohol. 66(1):74-81. Interpretive Summary: Most studies looking at alcohol use in teenagers have focused on how much and how often teenagers drink. In our research, we focused on other aspects of children's lives, including their relationships with family members and friends. A total of 1,253 students in grades 9 through 12 participated in the research study. Students completed questionnaires four times over a two-year period, with 743 students completing the four time points. Results showed six patterns of drinking: one problem drinking group (heavy alcohol use), two high-risk groups (drinking on dates and outdoor drinking), and three normal groups (light alcohol use, drinking with family members, and moderate alcohol use at parties). When looking at alcohol use over time in our sample, the data showed that emotional distress among teenagers predicted movement into problem drinking. These results are consistent with other research showing that movement into problem drinking can be predicted by parental and social factors.
Technical Abstract: Because most studies of adolescent alcohol use have focused primarily on the frequency and quantity of consumption, we know little about how adolescent drinking patterns change during the high school years. The purpose of this article is to provide such data, as well as to identify some of the individual, family, social life and community predictors of changes in drinking patterns over time. A sample of 1,253 students in grades 9 through 12 (57% female) in a large metropolitan school district participated. Three ethnicities were represented: African American, European American, and Mexican American. Students completed questionnaires every 6 months for a 2-year period (n = 743 at Time 4). Cluster analyses of the drinking variables yielded one problem-drinking group (heavy, multiple-context drinking), two high-risk groups (i.e., date and outdoor drinking) and three normative groups (light, family/parent, moderate/ party drinking). The most predictable patterns of longitudinal changes in adolescent drinking were consistent with the following progression (or the reverse): abstainer --> normative drinker --> high-risk drinker problem drinker. Paternal attitudes toward adolescent drinking and peer involvement in antisocial behavior predicted movement into normative drinking; social activity with peers predicted movement into high-risk drinking; and emotional distress predicted the transition into problem drinking. These findings are consistent with the multistage social learning model, demonstrating that the predictors of adolescent alcohol use vary across different levels of adolescent alcohol involvement.