Friendship group position and substance use

D. Wayne Osgood, Mark E. Feinberg, Lacey N. Wallace, James Moody

Research output: Contribution to journalArticlepeer-review

41 Scopus citations


This paper examines how an adolescent's position relative to cohesive friendship groups in the school-wide social network is associated with alcohol, tobacco, and marijuana use. We extend prior research in this area by refining the categories of group positions, using more extensive friendship information, applying newer analytic methods to identify friendship groups, and making strategic use of control variables to clarify the meaning of differences among group positions. We report secondary analyses of 6th through 9th grade data from the PROSPER study, which include approximately 9500 adolescents each year from 27 school districts and 368 school grade cohort friendship networks. We find that core members of friendship groups were more likely to drink than isolates and liaisons, especially in light of their positive social integration in school, family, and religious contexts. Isolates were more likely to use cigarettes than core members, even controlling for all other factors. Finally, liaisons were more likely to use marijuana than core members.

Original languageEnglish (US)
Pages (from-to)923-933
Number of pages11
JournalAddictive Behaviors
Issue number5
StatePublished - May 2014

All Science Journal Classification (ASJC) codes

  • Medicine (miscellaneous)
  • Clinical Psychology
  • Toxicology
  • Psychiatry and Mental health


Dive into the research topics of 'Friendship group position and substance use'. Together they form a unique fingerprint.

Cite this