Project ALERT is a substance use prevention curriculum for middle grades. The program is based on the social influence model. It has previously been formally evaluated; the current research is a second evaluation ten years after the first. The purpose of the study is to determine if implementation of Project ALERT was related to level of initiation of drug use among middle school students and/or level of use among previous experimenters. The study utilized a pretest-postest model with a control group for comparison. A broad spectrum of socioeconomic status and racial and ethnic groups were represented in the study sample. A treatment group of schools received the Project ALERT curriculum which was implemented faithfully, according to the design of the program. Control group schools were allowed to continue any drug education or information programs they might have had. Trained data collectors administered questionnaires before and after delivery of the seventh grade curriculum and at least fifteen months after the completion of the eighth grade booster curriculum. The questionnaire solicited information on whether, how often, and how much students used alcohol, cigarettes, marijuana and inhalants, and on psychosocial variables related to drug use. Logical regression was applied to analyze the data at the student level. Preliminary results indicated that implementation of the curriculum is related to decreased initiation of drug use, amount and frequency of drug use among previous experimenters as well as decreased intent to use among treatment group members. Additional data are currently being collected and analyzed with final conclusions expected in late fall 2003.Keyword(s): middle school issues, research