Best Friends' Drinking Behavior in Middle School: A Moderator of the Influence of Peer Group Norms on Young Adolescent Substance Use

Early onset drinking behavior in adolescence is associated with increased risk for the development of subsequent substance use issues in high school and adulthood. In the study of the predictors of adolescent drinking, peer consumption is one of the strongest correlates of adolescent alcohol use. Moreover, because drinking is a social activity in adolescence, adolescents’ willingness to drink generally depends on the normative context within which they are. In fact, adolescents who perceive greater peer acceptance for drinking behaviors are more likely to engage in alcohol-related activities (Perkins et al., 1999). Also, in a preliminary study, we showed that young adolescents who perceive alcohol use as normative within the peer group display higher levels of alcohol use 2 years later. This study investigates best friends’ substance use as a moderator of the extent to which young adolescents are influenced by perceived substance use norms among their peer group. We hypothesize that best friends’ substance use will exacerbate the impact of perceiving substance use as normative on students’ own consumption levels over time. Moderation analyses are conducted using the PROCESS macro in SPSS v. 21.

We recruited 1,278 participants in Oregon, who filled out paper-and-pencil questionnaires in class in Grades 6, 7 and 8. The Perception of Peer Group Norms Questionnaire is used to assess participants’ perceptions of peer substance use norms. Student substance use is assessed by teacher ratings on the Teacher Risk Perception Questionnaire (Soberman, 1994). Participants also identified their 3 best friends, allowing for the direct use of best friends’ substance use ratings as attributed by teachers. Friends’ substance use in Grade 7 is tested as a moderator of the relation between student perception of substance use norms in Grade 6 and student use in Grade 8.

Friends’ substance use significantly moderates the relation between perceiving substance use as normative within the peer group in Grade 6 and student’s increase in substance use by Grade 8 (b = .44, SE = .17, 95%, CI [.1135, .7691]), p < 0.01). In other words, having best friends who use substances more amplifies the predictive effect of perceiving substance use as normative on individual use over time. These results demonstrate the complexity of the underlying processes involved in the influence of social norms on adolescent behavior. They also serve to extend theory and understanding of best friend influence behavior in young adolescence. In all, providing young adolescents with the tools and opportunities to form friendships with peers who do not use substances would complement prevention programs that aim to reduce substance use among adolescents by targeting norm perception.