Statistics, Department of


Date of this Version



J Stud Alcohol Drugs. 2015 Mar; 76(2): 267–277.


Copyright 2015 Center of Alcohol Studies, Rutgers, The State University of New Jersey. Used by permission.


Objective: There has been an increase in the use of social network analysis in studies of peer socialization effects on adolescent substance use. Some researchers argue that social network analyses provide more accurate measures of peer substance use, that the alternate strategy of assessing perceptions of friends’ drug use is biased, and that perceptions of peer use and actual peer use represent different constructs. However, there has been little research directly comparing the two effects, and little is known about the extent to which the measures differ in the magnitude of their influence on adolescent substance use, as well as how these two effects may be redundant or separate constructs. Method: Using Waves I and II of the National Longitudinal Study of Adolescent Health (Add Health) saturated subsample, we directly compared effects of perception of friends’ use (PFU) and actual friends’ use (AFU) on alcohol, cigarette, and marijuana initiation and persistence of use 1 year later. We also examined potential moderating effects of friendship quality and individual use on the relationship between perceived and actual friends’ substance use and outcomes. Results: Results indicated that, overall, PFU effects were larger than AFU effects; however, these effects did not significantly differ in magnitude for most models. In addition, interaction effects differed for different substances and usage outcomes, indicating the meaning of PFU and AFU constructs (and thus, different types of peer socialization) may change based on substance and type of use. Conclusions: These results highlight the multifaceted nature of peer influence on substance use and the importance of assessing multiple aspects of peer socialization while accounting for distinct contexts related to specific substances and use outcomes.