The goal of this study was to explore the drugs-violence link among youth in two settings which presented similar rates of youth involvement in drugs and property crime, but lower levels of violence in one setting. The question was whether alcohol and illicit drug use were independently associated with violent behavior among a cross-section of the United States and Ontario student population. Two surveys, the U.S. Monitoring the Future Study (MTF) and the Ontario Student Drug Use Survey (OSDUS), were employed to students 12- to 18-years-old. There were similarities in the models between U.S. grade 12 and Ontario grade 11 students. The students had very similar profiles in terms of the prevalence of licit and illicit drug use. There was less violence in Ontario based on both self-report and official arrest statistics, but the models show that drug use and violence were highly connected in the respective student populations. There were also similarities in the models in Ontario regardless of the age of the students. Youth likely to engage in binge drinking and cannabis and cocaine use were also likely to engage in violent behavior. Cannabis use was a strong predictor of violence among the general student population, even when controlling for binge alcohol drinking and cocaine use, as well as being a male. The analyses supported the robustness of the drugs-violence link for past year cannabis and cocaine use, and binge drinking among student population in the two countries, but the data confirmed only that these behaviors were correlated. This correlation indicated that the relationship between drug use and violence among youth was spurious in that both were related because they shared common causes or predictors such as antisocial personality, impulsivity, or poor relations with parents. This calls for further exploration of the social and contextual factors that may lead to violent outcomes. 4 tables, 58 references