Estimating peer effects in adolescent smoking behavior: a longitudinal analysis.

Purpose: To empirically quantify the role of peer social networks in explaining smoking behavior among adolescents. We follow adolescents over time to examine whether the role of adolescent peers persists in affecting individual behaviors as adolescents transition into adulthood.

Methods: Using longitudinal data of a nationally representative sample of adolescents we use a multivariate structural model with school-level fixed effects to account for the problems of contextual effects, correlated effects, and peer selections to purge the potential biases from the estimates of peer influence. Our peer group measures are drawn not only from the nomination of close friends but also from classmates. Smoking prevalence among the peer groups were constructed using the peers' own reports of their cigarette consumption.

Results: Controlling for parent level characteristics and other demographic parameters, we find that a 10% increase in the proportion of classmates who smoke will increase the likelihood of smoking by more than 3%. Also, an increase in smoking rates among an individual's close friends by 10% will increase the likelihood of smoking by 5%. We find evidence to show that the influence of close friends from adolescence years continue to have an impact on smoking propensities even when a transition into adulthood is made.

Conclusions: Our findings support the literature reporting that peer effects are important determinants of smoking even after controlling for potential biases in the data and that these effects persist into adulthood. Effective policy aimed at reducing smoking rates among adolescents would consider these long-lasting peer effects.