Previous research has consistently demonstrated a relationship between gang membership and involvement in illegal substances. In addition, researchers have noted that gang members are frequently more heavily involved in drug sales, which often lead to increases in violent behaviors. Most of this research, however, is either cross-sectional or ethnographic in nature, and therefore does not reveal the causal nature of these relationships. This research attempts to establish the temporal ordering of these relationships while controlling for a variety of relevant variables and to determine whether the relationships between drug involvement and violence differ for gang members versus nongang members. The findings indicate that gang membership is weakly associated with drug involvement, including both usage and sales. This involvement, however, does not appear to be related to assaults. Results suggest that gang membership is not determinative of drug involvement among a national random sample of youth. (Contains 5 tables, 1 figure and 8 notes.)