Selection and socialization effects of fraternities and sororities on US college student substance use: a multi-cohort national longitudinal study.

AIMS To examine how membership in fraternities and sororities relates to the prevalence and patterns of substance use in a national sample of full-time US college students. DESIGN Nationally representative probability samples of US high school seniors (modal age 18 years) were followed longitudinally across two follow-up waves during college (modal ages 19/20 and 21/22). SETTING Data were collected via self-administered questionnaires from US high school seniors and college students. PARTICIPANTS The longitudinal sample consisted of 10 cohorts (senior years of 1988-97) made up of 5883 full-time undergraduate students, of whom 58% were women and 17% were active members of fraternities or sororities. FINDINGS Active members of fraternities and sororities had higher levels of heavy episodic drinking, annual marijuana use and current cigarette smoking than non-members at all three waves. Although members of fraternities reported higher levels than non-members of annual illicit drug use other than marijuana, no such differences existed between sorority members and non-members. Heavy episodic drinking and annual marijuana use increased significantly with age among members of fraternities or sororities relative to non-members, but there were no such differential changes for current cigarette use or annual illicit drug use other than marijuana. CONCLUSIONS The present study provides strong evidence that higher rates of substance use among US college students who join fraternities and sororities predate their college attendance, and that membership in a fraternity or sorority is associated with considerably greater than average increases in heavy episodic drinking and annual marijuana use during college. These findings have important implications for prevention and intervention efforts aimed toward college students, especially members of fraternities and sororities.