Alcohol Use and Social Interactions among Adolescents Do peer-effects exist within and/or between the majority population and immigrants?

Are adolescents who attend schools with a high level of alcohol use more likely to use alcohol themselves? This paper analyzes peer-effects in adolescent alcohol use based on a survey of 13,337 adolescents in Sweden in 2005. The empirical analysis uses multi-level logistic model to handle non-observable heterogeneity between the schools and the results show that attending a school with a high level of alcohol use is a strong predictor of alcohol use for the individual. However, a positive association is only seen within Swedes and within non-Swedes (1st and 2nd generation immigrants). Between Swedes and non-Swedes there is actually a negative association, i.e. if many Swedes drink in a certain school, alcohol use among non-Swedes is lower (and vice-versa). An exception to these results are schools with a very low share of non-Swedish adolescents, where non-Swedes alcohol use also is positively associated with Swedish peers’ alcohol use.