The homicide group was divided into a psychotic group -- those with clear psychotic symptoms; and conflict group -- those who were not psychotic and who were engaged in interpersonal conflict with the victim; and crime group -- those who committed the homicide in the course of another crime. The larceny comparison group scored significantly higher (more problems) than the homicide group on school adjustment, childhood symptoms, criminal activity, and psychiatric history. Within the homicide group, the psychotic adolescents scored significantly higher than the nonpsychotic (crime and conflict) adolescents on psychiatric history, but scored lower on criminal activity. The crime subgroup scored significantly higher than the conflict subgroup on school adjustment, criminal activity, and substance abuse, but scored lower on stressful life events before the offense. This study draws implications for research on violent behavior and for the future treatment and disposition of juvenile homicide defendants. 6 tables, 31 references.