The study found that both violence victimization and perpetration correlated with depressive symptoms; however, when examined simultaneously as predictors, only violence victimization emerged as significant. In addition, once exposure to community violence was entered into the model, it emerged as the most robust predictor of depressive symptoms. This suggests that chronic exposure to violence in the community, as well as individual incidences of violence victimization and perpetration are linked to depression. Additional risk factors for depressive symptoms included female gender and the use of multiple substances. Possible explanations for these findings are offered for future research. Study participants were 115 adolescents ages 12-17. The majority of participants were African-American (n=68), the remainder were White, American Indian, Asian, and "other." Violence victimization, perpetration, and substance use were assessed with items from the SAGE questionnaire. The six-item modified depression scale was used to assess the frequency with which participants experienced depressive symptoms, e.g., felt very sad, felt hopeless about the future, slept a lot more or a lot less than usual, and had difficulty concentrating on schoolwork in the past 30 days. 2 tables and 10 references