The study found that including violent-media-preferences scores added significantly to the prediction of both violence and general aggression. Even for those lowest in other risk factors, a preference for violent media was predictive of violent behavior and general aggression. This finding is consistent with earlier research which showed that even low-aggressive individuals were affected by media violence (Eron et al., 1972). These findings indicate the importance of including exposure to violent media into developmental models of risk for aggression. One noted limitation in this study was the inability to measure a variety of key personal and contextual influences on violent and aggressive behavior. Future research should focus on the role played by exposure to media violence in the emergence and maintenance of, as well as increase in, violent behavior over time in at-risk and high-risk youth populations. The sample completed extensive individual interviews, and their parents/guardians as well as teachers/staff provided data through telephone or mailed surveys. Youth indicated their three favorite television shows, movies/films, and video/computer games during childhood (7 or 8 years old) and since being “a teenager.” All titles listed by youth in the three media categories were subsequently coded by reliable independent raters in determining the extent to which they contained visible interpersonal violence, using a five-point rating scheme. Youth completed the 22-item Delinquency Scale (Elliott and Huizinga, 1983). Instruments also measured serious physical aggression, trait aggressiveness, callous-unemotional traits, academic skills, psychopathology, exposure to neighborhood violence, and exposure to low-level aggression. 2 tables and 65 references