Purpose: Peer aggression and bullying occur frequently in schools with negative consequences at multiple levels (student, classroom, school), yet many existing school-based programs that try to prevent peer aggression and bullying are limited in scope. In order for a peer aggression and bullying prevention program to be effective, it must move beyond a focus on the aggressor or bully and consider broader classroom and school climate issues that may foster such behaviors. Furthermore, such a prevention program must consider relational aggression (e.g., starting rumors, threatening to withdraw friendships) along with physical aggression, as well as the less structured settings such as the playground and lunchroom where most aggression and bullying incidents occur in elementary schools. To address these concerns, the researchers will develop the Partner for Prevention (P4P) program to address peer bullying and aggression at multiple levels—the individual, peer group, and school-wide—in under-resourced school districts.

Project Activities: The researchers will use focus groups, semi-structured interviews, and participant observations in the first 2 project years to develop, adapt, and enhance P4P in an iterative manner using a community-based participatory research approach. In the final project year, the team will conduct a small-scale efficacy study to estimate effect sizes and intra-class correlations for primary outcomes to determine the promise of the program for elementary school students in low-income communities. Data on acceptability, feasibility, and implementation fidelity will be collected throughout the project.

Products: The products of this project will be the fully developed Partner for Prevention (P4P) program that integrates a classroom program, playground and lunchroom strategies, teacher and school staff coaching and consultation, and community outreach to engage parents in efforts to prevent aggression and bullying in elementary schools. Peer-reviewed publications will also be produced.

Structured Abstract

Setting: This project will be conducted in four elementary schools in a large urban center in Pennsylvania.

Sample: Approximately 190 students and 26 teachers in third-, fourth- and fifth-grade classrooms, 20 recess supervisors, 4 administrators, and 10 parents and/or community members will participate. The vast majority of youth (> 95%) attending these urban high-risk schools are African American.

Intervention: P4P will consist of four integrated components that address aggression and bullying at the individual, peer group, and school-wide level: (1) a classroom-based prevention program that teaches youth problem-solving strategies, sympathy and perspective taking skills, and strategies for bystanders of bullying; (2) playground and lunchroom consultation that addresses behavioral challenges during the lunch-recess period, with the goal of helping school staff to feel empowered and to develop specific rules to prevent peer bullying and aggression in this setting; (3) teacher training and coaching that supports teachers with classroom behavior management and student engagement strategies while reinforcing bullying prevention strategies; and (4) community outreach activities that are intended to build more positive community-school relationships, provide resources for parents, and help parents to promote the prevention of peer bullying and aggression at school.

Research Design and Methods: The first 2 years of the project will be devoted to intervention development. The researchers will use a community-based participatory research process incorporating a range of qualitative data collection techniques (freelisting, focus groups, participant observations, and semi-structured interviews). In Year 1, students, teachers, recess supervisors, administrators, and community/parent participants in two schools will be engaged in the development of each of the four intervention components. In Year 2, the intervention will be implemented and refined based on participant data collected through observations, focus groups and semi-structured interviews. In Year 3, a pilot study will be conducted at two additional schools to determine the potential promise of the integrated P4P Program through estimation of both effect sizes and intra-class correlations for primary outcomes (problem solving, sympathy, aggressive behaviors, school and classroom climate, and academic competency) and to determine the acceptability, feasibility, and implementation fidelity of the program.

Control Condition: Researchers will collect control data from youth in eight randomly selected classrooms across two schools receiving no intervention in Year 2 and will compare that to youth from eight randomly selected classrooms within the same two schools that are participating in the intervention in Year 3 as part of the pilot implementation of P4P.

Key Measures: The researchers will collect data in four primary outcome domains: problem solving and sympathy (e.g., the Cartoon-Based Hostile Attributional Bias Measure, the Peer Sympathy Scale); aggressive behavior (e.g., Children's Social Behavior Questionnaire, school records); peer bullying (e.g., the Illinois Bully Scale, the School Climate Bully Scale); climate (e.g., the Academic Competence Evaluation Scales, the Student-Teacher Relationship Scale, the Playground and Lunchroom Climate Questionnaire); and parental involvement in school (Parent-Teacher Involvement Questionnaire). Researcher-developed measures of implementation fidelity and user acceptability will also be used.

Data Analytic Strategy: During the development phase of this project, freelist data will be analyzed using a consensus procedure. Focus-groups, semi-structured interviews and participant observations will be analyzed using modified grounded theory. Data collected during the pilot study will be described through summary and descriptive statistics (e.g., means, standard deviations, ranges, etc.). Parametric statistical tests, such as, the paired t-test and non-parametric statistical tests such as, the Wilcoxon signed-rank test will be used as needed. A four-level longitudinal model will be used to test potential intervention effects. A variety of approaches (e.g., analysis of covariance or a comparison of percent change scores using an independent t-test) will be considered to deal with potential confounds (e.g., principal change, school implements a new program) given that the control and intervention conditions happen in different years.