Is being popular a risky proposition?

Longitudinal associations between social preference, perceived popularity, and risk behaviors (smoking, alcohol use, and sexual activity) were examined in a sample of high school students. Social preference did not predict any of the risk behaviors assessed, although the interaction between gender and social preference was predictive of sexual behavior. Perceived popularity in Grade 10 was predictive of increased alcohol use and sexual activity in Grade 12 for both boys and girls. A reciprocal relationship was found for boys, in which smoking in Grade 10 also predicted gains in perceived popularity over time. Furthermore, results suggested that gains in perceived popularity may be associated with subsequent losses in social preference over time, lending support for the "cycle of popularity" observed by Eder. Implications are discussed, including the personal impact of risk behaviors on perceived popular teens, as well as the broader influence their behavior may have on their less popular counterparts.