Protective Factors Against Peer Aggression: An Exploration of the Interconnection between School Connectedness, School Climate, and Social Norms

Background/Purpose: Youth violence is one of the leading causes of death and non-fatal injuries among adolescents in the United States (David-Ferdon et al., 2016). While contextual frameworks highlight the importance of the environment on behavior, youth violence prevention approaches often target individual characteristics. The Promoting Protective Policy research project focuses on policy level improvements to foster safe, inclusive school environments intolerant of violence.

Although contextual elements contribute to the school environment differently, they do not exist independent of one another. Therefore, interventions targeting a specific element are not siloed and are likely to impact others. This presentation explores the associations between student perceptions of harmful peer behavior and school connectedness, school climate, and social norms individually and when combined.

Methods: This study uses data from three middle schools and four high schools in Michigan. Cross sectional survey data was collected from approximately 50 randomly selected eighth or tenth graders from each school, N=389.

The survey consisted of 14 sets of questions measuring perception of the school environment. Independent variables in this study included school connectedness, a ten-item scale; school climate, an eight-item scale; and social norms, a nine-item scale. These questions were rated on Likert scales, lower scores indicating more negative perceptions of the variable and higher scores indicating more positive perceptions. The dependent variable, a five-item scale assessing perception of harmful peer behavior, was scored with lower score indicating the specific behavior was perceived to occur less often while higher scores indicated greater occurrence.

Linear regressions were run to assess variance in the dependent variable associated with each independent variable. To examine variance in the dependent variable associated with each pair of independent variables, bivariate regressions were conducted. Then the multivariate regression was run to assess variance in the dependent variable associated with the three independent variables together.

Results: Significant associations with harmful peer behavior were found for each independent variable alone, when paired, and as a group of three. Greatest variance in peer behavior was seen in the model including all three independent variables. However, the group of independent variables did not account for greater variance compared to the summed variance of each independent variable alone. While all associations with the dependent variable were significant in the single and bivariate regression models, school connectedness was no longer significantly associated with harmful peer behavior in the multivariate model.

Conclusion and Implications: These results may suggest that the observable attributes of school connectedness are also attributes of school climate and/or social norms. This highlights the importance of understanding how environmental elements function in the context of one another. Interventions require time, money, and effort, so determining the most effective and efficient approach to achieving the desired outcome can help reduce strain on limited resources.