types of drug misuse. Relatedly, our drug-related aggression measure excluded marijuana-related aggression. This was of less concern because marijuana misuse has been linked to lesser aggression (Myerscough & Taylor, 1985), yet this important and relatively frequent form of drug misuse and its relations to aggression and aggression types should be investigated in more detail. Third, our findings were correlational and experimental work is needed to examine the causal influence of alcohol and drug misuse in affecting aggressive behavior. Fourth, our measures were self-reported and retrospective, which are likely subject to biases and memory errors. Future research employing other measures that are less susceptible to bias should be conducted. Self-reports were also scored dichotomously, inherently eliminating much of the naturally-occurring variability in aggressive behavior. The future use of continuous measures that better capture the varying degrees of the various aggression phenotypes that exist in the real-world would alleviate such issues. Fifth, we had higher-than-desirable rates of missing data for several variables. These missing data were due to screening procedures that excluded individuals without considerable histories of substance misuse, as alcohol and drug naive individuals would have not had the opportunities to perpetrate alcohol-related and drug-related aggression. Yet 7,442 participants provided full
