OBJECTIVE The main objectives of this study were to determine: (1) the relative roles of heavy episodic drinking (HED), drinking frequency and drinking volume in explaining alcohol-related aggression and (2) whether drinking context variables (i.e. usual drinking locations, typical drinking companions and extent of peer drinking) confound or modify the relationship between HED and alcohol-related aggression or whether they predict alcohol-related aggression independently. METHODS A secondary analysis of the US National Longitudinal Survey of Youth was conducted. Alcohol-related aggression (denoted fights after drinking) was measured based on self-reports of arguments or fights that occurred during or after drinking in the previous 12 months. METHODS A composite sample of drinkers, ages 17-21, from the 1994, 1996 and 1998 Young Adult surveys (n = 738) was used. RESULTS Frequency of drinking and drinking volume largely confounded the association between HED and fights after drinking. Usually drinking in public locations away from home versus private locations was found to be significantly associated with a greater likelihood of fights after drinking among females. Among males, usual drinking location modified the relationship between drinking frequency and alcohol-related aggression, with the greatest risk of aggression for males who drank frequently and usually drank in public locations away from home. CONCLUSIONS Programs designed to reduce drinking frequency in this population and to increase the safety of drinking locations in public places away from home may prove to be beneficial in reducing alcohol-related aggression.