Wisconsin School Violence and Bullying Prevention Study

Violence within schools is of great concern. Instances range from pushing/shoving to shootings and homicides. In addition to the immediate implications, school violence, both as victim and perpetrator, holds a relationship with current and future violent acts outside of the school environment. These implications illustrate the need to understand ways that schools can prevent instances of violence for the safety and well-being of students, faculty, and the greater community.

The Wisconsin Department of Public Instruction (DPI) proposes a three-year study to the US Department of Justice, National Institute of Justice. The proposed study researches the effectiveness of a Positive Behavior Interventions and Supports (PBIS) and Bullying Prevention program regarding outcomes of school safety and climate among students in grades six through eight. The anticipated project sample is 50 middle schools, representing approximately 22,000 racially and economically diverse students. The Wisconsin DPI will partner with schools, academic institutions, and several community health partners in this endeavor.

This study will utilize a matched case-control methodology, in which schools implementing PBIS, but without an active comprehensive bullying prevention program, as of December 2014, will be assigned to experimental or control groups. Experimental schools will implement a bullying prevention program beginning in September 2015. Control schools will delay implementation until September 2017, after data collection ends. The hypotheses involve improved metrics of school safety among experimental schools, relative to control schools. These metrics are: bullying (rate and incident), exclusionary discipline for violent acts and all reasons, and perceptions of school climate. Annual collection of the outcomes will be used to support the hypotheses or null hypotheses.

Statistically, hierarchical generalized linear modeling will be used to account for clustering of students and data non-normality. The study is designed to maximize internal and external validity. Internal validity is addressed through uniform data collection procedures for all metrics. In addition, schools are matched at the district level before random assignment as experimental or control. External validity is addressed through study design and its effectiveness-trial format, which will enable program replication to a greater degree than previously available through existing efficacy-type trials.

The anticipated global outcomes from the project involve generalizable knowledge on methods by which schools can reduce violence on their grounds and, indirectly, reduce violence by students outside of the school environment, school year and their school careers. Methods by which this knowledge will be shared include presentations at national conferences and publications in peer-reviewed journals. ca/ncf