Aggressive students often struggle in multiple domains of their school functioning and are at increased risk for high school dropout. Research has identified a variety of warning flags which are strong predictors of high school dropout. While it is known that aggressive students exhibit many of these warning flags, there is little research which identifies the paths aggressive students take towards high school dropout. This study attempts to better characterize the relationship between aggression and dropout using a sample of students (N=685) from two middle schools in an urban, low income school district in the Northeastern United States. This study utilizes survival analysis, a longitudinal data analysis strategy for examining event occurrence. The sample was followed over a course of seven years, spanning from when students start middle school in the sixth grade through their expected on time-graduation year in the twelfth grade. In this study, hazard and survivor functions, key components of survival analysis, were utilized to examine the timing of high school dropout and compare timing of dropout for students characterized as aggressive vs. not aggressive. Discrete time hazard models using logistic regression were analyzed to determine how well aggression as rated by teachers and peers as well as a variety of educational and demographic variables predict high school dropout. Hazard and survivor functions showed that for this sample, aggressive students were at increased risk of dropout through the course of their educational careers. Their pattern of risk, i.e. grades in which they were at a lower or higher risk for dropout, however, was similar to that of non-aggressive students. Peer-rated aggression was related to each of the dropout warning variables as aggressive students were absent more often, had lower GPA's, accumulated more suspensions and were more likely to be retained at some point during their educational careers. Results of discrete time hazard models showed that aggression was no longer a significant predictor of dropout when other educational "warning flag" variables were included in models. A discrete time hazard model including the effects of time, Age for Grade status and Retention status produced the best goodness of fit measures amongst a host of models that were analyzed. In all models that were analyzed, time was a significant predictor, indicating that dropout is not time invariant and a student's grade must be considered when determining the effect of any of the hypothesized predictors on dropout. Implications of these results for schools, specifically regarding effective management of aggressive behavior in schools, are discussed. [The dissertation citations contained here are published with the permission of ProQuest LLC. Further reproduction is prohibited without permission. Copies of dissertations may be obtained by Telephone (800) 1-800-521-0600. Web page: http://www.proquest.com/en-US/products/dissertations/individuals.shtml.]