Implementing School-Based Restorative Justice to Improve Population Health

Background:

Over the past decade, school-based restorative justice (RJ) has gained national prominence as a relational approach to improving school climate, responding to misconduct, and interrupting the school-to-prison pipeline. Surprisingly, despite its simultaneous positive impact on factors associated with health--academic success, school safety, connectedness, and social support--most scholars and practitioners do not associate RJ with population health. In the United States, this is especially important because the school-to-prison pipeline disproportionately impacts African-Americans, Latinos, Native Americans, and youth with disabilities, who also experience health inequities. In this study, we use ecosocial theory to conceptualize school-based RJ as a structural population health intervention. We hypothesize that school-based RJ, if implemented with fidelity, simultaneously improves educational and health outcomes.

Methods:

We tested the hypothesis by using 1) hierarchical linear modeling, 2) 2013/2014 California School Climate, Health, and Learning Survey System data, and 3) one districts’ publicly available RJ implementation data. The study sample included 6992 students nested in 32 middle and high schools in one California school district with a district-wide RJ policy. School-level predictors were: 1) use of the whole school RJ approach (Yes/No), 2) level of RJ implementation (beginning, intermediate, or advanced), and 3) use of peer RJ (Yes/No). School-level control variables were: 1) school size (number of enrolled students) and 2) the cumulative percentages of African American, Latino, and students who received free and reduced-price meal. Student-level predictors were: 1) grade, 2) ethnicity, 3) sex and 4) parental education. The dependent variables were: 1) mental health (missed school in the past 30 days due to feeling very sad, hopeless, anxious, stressed, or angry [Yes/No]), 2) physical health (missed school in the past 30 days due to feeling physically sick, including problems with breathing or teeth [Yes/No]), and 3) academic achievement (self-reported average grade for the past 12 months [e.g., "A’s and B’s"; "mostly B’s"]). After determining the between-school variance in outcomes and estimating intraclass correlation coefficients, we constructed sequential 2-level theoretically driven models for each dependent variable.

Results:

Compared to schools without RJ, attending a school which used peer RJ reduced the odds of mental health-related absences by 36%. The whole school approach to RJ implementation was also protective, reducing the odds of physical health-related absences by 32%. Finally, the use of peer RJ predicted an increase in the average grade by 0.64 points (e.g., approximately translated to change from "mostly Cs" to "Cs and Bs"), compared to schools not using RJ.

Conclusions:

Our findings indicate that school-based RJ shows promise as a structural intervention that can contribute to improving educational and, ultimately, population health outcomes. Literature suggests that higher academic performance and lower absenteeism predict high school graduation, which in turn predicts better future health outcomes. No cause and effect can be evaluated given the cross-sectional nature of the data. However, the findings point to an exciting avenue for structural interventions by revealing the central role that relational approaches and youth engagement can play in transforming schools as structures that impact health.