The primary objective of this thesis is to develop a model that predicts the threat of drug use at the local command level. The model is developed from two surveys: (a) the National Household Survey on Drug Abuse (U.S. Department of Health and Human Services), and (b) the 1995 Department of Defense Survey of Health Related Behaviors Among Military Personnel. This predicting technique is applied to specific Navy commands from data obtained from the Defense Manpower Data Center (West) on each command's demographic profile. The results show that a model can be developed to predict drug use at the local command level based on the underlying civilian drug use propensity. The sex of an individual is the most important predictor for civilians. The education level and the age of the individual are the most important predictors for the military. Race and sex do not have an impact on drug use among military members. The model could be used by local commanders to determine the potential threat of drug use at the command. Commands should test at a monthly test rate relative to the magnitude of this threat. The model should be revalidated periodically as demographic and locational factors change.