A Randomized Impact Evaluation of the No Bully System

Statement of the Problem. The proposed study investigates the impacts of the No Bully program on resolution of bullying incidents; bullying perpetration and victimization; and school safety, peer support, and other indicators of school climate. Bullying affects large numbers of U.S. students in elementary schools and is associated with short and long term harms for victims and bullies. Although prevention is critical, schools also need effective interventions for dealing with bullying once it occurs. The No Bully System is a promising set of interventions designed to activate adult and peer support systems within the school for bullying targets. No Bully trains staff to prevent and interrupt student harassment and bullying and ensure school-wide anti-bullying policies are in place. At its core is the Solution Team, an intervention where a trained adult facilitator (Solution Coach) works with the bullying victim, bully followers, and several pro-social peers on developing solutions for ending the bullying. Although preliminary evidence or program effectiveness is promising, a rigorous evaluation is needed to understand No Bully's impact on bullying, school safety, and school climate. Subjects: The study will be conducted in 24 elementary schools served by Oakland Unified School District. Approximately 6,480 grade 3-5 students will participate in the study. The student population is diverse and largely minority, with 7% White, 33% African-American, 39% Latino and 15% Asian/Pacific-Islander. Approximately 70% of students are eligible for free/reduced-price meals. Partnerships: To conduct the study, WestEd has partnered with the developer of the No Bully System (No Bully), Oakland Unified School District, and Dr. David Farrington of Cambridge University. Research Design, Methods, and Analyses: The study utilizes a cluster randomized experimental design, whereby 24 schools are randomly assigned to an experimental group and a wait-listed control group. Key student outcomes include measures of (a) bullying perpetration and victimization, (b) bully bystander behavior, (c) peer supports, and (d) other indicators of school climate. Self-report survey data will be collected at three time points: in the fall prior to implementation of No Bully and in the spring of the first and second implementation years. The primary hypothesis-testing analyses will involve fitting mixed effects ANCOVA models (HLM or multilevel models). Products, Reports, and Data Archiving: Anticipated products include publications in journals and presentations targeting academic audiences, and dissemination to practitioners and policy makers through WestEd's national centers. Interim/final reports will be completed, and data will be archived for public dissemination. ca/ncf