Citation

Henry KL, Slater MD, Oetting ER. J. Stud. Alcohol 2005; 66(2): 275-283.

Affiliation

Institute of Behavioral Science, University of Colorado, 910 28th Street, Boulder, Colorado 80303, USA. khenry@colorado.edu

Copyright

(Copyright © 2005, Rutgers Center of Alcohol Studies)

DOI

unavailable

PMID

15957679

Abstract

OBJECTIVE: Friends' substance use, sensation seeking and low perceived harm are well-established risk factors for substance use, but they are often treated as stable factors that affect adolescents' likelihood of substance use. This study instead explores the effects of changes in risk factors for individual adolescents. METHOD: Participants in this study were 1,065 male and female students. The students were in sixth or seventh grade at the initial survey and provided survey data on three additional occasions over a period of 2 years. Random-coefficient models were used to assess the intraindividual variability of friends' alcohol use, perceived harm, and risk taking and their effect on alcohol use. RESULTS: As expected, the overall number of alcohol-using friends is correlated with a student's own alcohol use. In addition, there is a dynamic relationship within student; as friends' alcohol use changes over time, it is accompanied by parallel changes in alcohol use by the individual. Two moderating variables of the effect of friends' use of alcohol were validated: perceived harm of alcohol use and risk taking. The effect of increased exposure to alcohol-using friends is more robust during times when an adolescent also has become less likely to perceive the harmful effects of alcohol use or when an adolescent indicates increased interest in risk-taking behavior. CONCLUSIONS: Although friends' use of alcohol is a salient predictor of an adolescent's own use of alcohol, some types of students are more likely than others to be influenced by friends' behavior, In particular, students' perception of harm and predisposition to risk taking are important moderators of the effect of friends' influence.