Resumen

Do victims of crime increase or decrease participation in social groups? Despite being an interesting question, the literature on this issue is almost non-existent (Bateson, 2012). Our results, based on the Chilean victimization survey, show that victims of crime are 7% more likely to participate in social groups than non victims. Moreover, participation is higher in companionship than instrumental groups and for victims of property crimes. This is consis-tent with the predictions of the stress-bu er hypothesis (Cohen and Wills, 1985). Finally, sensitivity analysis reveals that the use of victimization surveys is more robust than opinion polls to the presence of unobservables.