The research study will experimentally evaluate the impact of PBS on early adolescent development through a randomized control trial involving 36 middle schools. The impact of PBS on school staff discipline practices and student behavior will be evaluated. The study will examine whether the likely reductions in negative behavior in school are accompanied by reductions in peer harassment and victimization, peer rejection, deviant peer formation, and the development of antisocial behavior, substance use, high risk sexual behavior, and depression.

The specific aims for the project are as follows:

1. Evaluate the effects of the PBS intervention in a group-randomized trial in which middle schools are assigned to receive systematic training and consultation in the implementation of PBS or a one-day workshop on PBS (the control group).

   1. The PBS intervention group is expected to improve school behavior management practices compared to the control group.
   2. School climate is expected to improve in the PBS intervention group compared to the control group.
   3. Problem behaviors and deviant peer group formation are expected to decrease for students in the PBS intervention group compared to those in the control group.
   4. The development of problem behavior in later adolescence is expected to be lessened for the PBS intervention students compared to those in the control group.

   The effects of the intervention will be assessed through staff, student, and observer reports of discipline practices in schools. The investigators will assess changes in the frequency of in-school problem behaviors as measured by student reports, office discipline referrals, and out-of-school suspensions. Measures of student and teacher perception of the quality of school climate will include student reports of harassment and teacher reports of burnout. Formation of deviant peer groups will be assessed through student reports of their peer affiliation. Finally, adolescents' development of problems will be assessed over as many as five years through annual surveys of antisocial behavior, substance use, high risk sexual behavior, and depression.

2. Assess the differential impact of intervention fidelity on targeted outcomes. No real-world implementation of a preventive intervention can be expected to be implemented with perfect fidelity in every school. The study will:

   1. Assess the degree of fidelity of implementation of all PBS intervention aspects.
   2. Examine the relationship of fidelity to changes in each of the targeted outcomes: discipline practices, in-school problem behavior, peer relations, deviant peer group formation, and adolescent problems.
   3. Provide further information about the specific facets of PBS associated with particular outcomes.
   4. To the extent that fidelity is related to outcome, above the simple relationship between condition assignment and outcome, it will strengthen the investigators confidence in the efficacy of PBS practices.

3. Test a model of the influence of middle school discipline practices on the development of adolescent problems. The model, which is specified in Figure 1 and the following pages, posits that:

   1. Inadequate behavior management practices in middle schools contribute to aggressive and disruptive behavior among students in those schools.
   2. Schools with high levels of aggression and harassment make the formation of deviant peer groups more likely.
   3. Escalating aggressive behavior, harassment, and deviant peer group formation then contribute to the development of a range of problems in both school and non-school settings including antisocial behavior, substance use, depression, and high risk sexual behavior.

   The model will be tested using state of the art hierarchical linear modeling (e.g., Singer & Willett, 2003).

4. Assess the maintenance of the PBS intervention after termination of training and technical assistance. It is unlikely that effective behavior management will be implemented and maintained, unless the school-as an organization-has adopted a system that is supported by all staff and used in every setting (Gottfredson, Gottfredson & Czeh, 2000; Sprague, Walker, Golly et al., 2001). Research is needed, however, on the extent to which the PBS system, once implemented is maintained. The present project will provide evidence about the degree to which PBS is maintained and identify the school characteristics that are associated with its maintenance.

5. Assess the relationship of Office Discipline Referral data to adolescents' reports of their engagement in diverse problems. Office Discipline Referrals (ODR) have been widely used to evaluate PBS interventions. PBS typically produces substantial reductions in ODR's (e.g., Metzler, Biglan, et al. 2001; Taylor-Green et al., 1997; Sprague, Walker, Golly, White & Myers, 2001). However, little is known about the relationship between the ODR measure and other measures of adolescent problems (Irvin, Tobin, Sprague, Sugai, & Vincent, 2004). The proposed project will:

   1. Assess the validity of the ODR by examining both the concurrent and longitudinal relationships of the ODR with student self-report measures of problems.
   2. Examine the significance of reductions in ODR's for PBS intervention schools and students.

Inclusion Criteria:

• All students and teachers in the school

Exclusion Criteria:

• Any students or teachers who decline consent (passive or active)