The influence of interscholastic athletic participation on delinquency

A premise of contemporary crime prevention programs such as youth athletic leagues and midnight basketball projects is that athletic participation reduces crime. Athletic participation may reduce crime either by way of the socializing effects of participation or by way of the tendency of athletics to channel the time and energy of young adults from crime to more benign and even healthy activities and thus divert potential delinquents from engaging in criminal activity. In this paper, I examine the relationship between athletic participation and the delinquency of young adults by using panel data of over 20,000 high school students from the National Educational Longitudinal Study (NELS). The NELS data permits separate estimates of factors that predict participation in various high school sports programs from estimates of the effect of that participation on delinquency. Specifically, I use an instrumental variable approach to disentangle the self-selection effect from the subsequent effect of athletic participation on delinquency. Moreover, I focus on the effects of particular sports---football, basketball, baseball, swimming, diving, etc. I also transform the standardized regression coefficients into elasticities in order to make comparisons within and across regressions easier. Finally, I address the central policy implications of the findings.