A randomized control trial was conducted to evaluate the impact of the Rural Early Adolescent Learning Program (Project REAL) on teachers' awareness of peer groups at the beginning of the 1st year of middle school. Two schools were randomly assigned to the intervention condition and 2 to the control condition. Thirty-nine teachers and 466 students (252 girls, 214 boys) participated in the study. As part of Project REAL, teachers were taught about social dynamics and early adolescent peer group processes. Social cognitive mapping procedures were used to assess and compare students' and teachers' perceptions of 6th grade peer groups. As compared with teachers in the control condition, intervention teachers were more likely to accurately identify peer groups. In addition, when the analyses were restricted to students who were involved in bullying, intervention teachers were more likely to accurately identify peer group membership. However, there were no significant differences between intervention and control teachers for the accurate peer group membership identification of specific bullying involvement subtypes (i.e., bullies, victims, bully victims). (Contains 2 tables.)