Does Stickiness Matter? A Longitudinal Examination of the Stability of Adolescent Peer Groups

• In Event: Peers, Peer Groups, and Delinquency

Abstract

Using data from the Teenage Networks in Schools (TEENS) project, this work provides an empirical test of Warr’s (1993a) “sticky friends” hypothesis. Warr (1993a) argued that, “delinquent friends, once acquired, are not lost in subsequent years” (p. 31). Analyzing data from 9th grade students at a mid-sized Kentucky high school, this study builds on Warr’s work by examining three research questions: (1) are levels of delinquency among peers stable over time?; (2) are the peers themselves stable over time?; and (3) does the stability of peers interact with their delinquency to affect the target respondent’s delinquency? The longitudinal multilevel model for change was the primary analytical tool due to the focus on stability and change over time. Overall, the findings from the analysis support Warr’s (1993a) conclusion that delinquent peers maintain stable relationships with one another, even while controlling for several potential selection effects. Additionally, this analysis adds to the literature by finding that delinquent peers—meaning the actual people themselves and not just their behavioral repertoires—largely remain stable over time. Finally, the results suggest recent friendships may be more influential as a cause of delinquency compared to more stable, long-term friendships.

Author

• Brandon Dulisse, Indiana University of Pennsylvania