All SROs are not created equal: Heterogeneity in school policing

• In Event: School Resource Allocation

Abstract

This paper examines the role of School Resource Officers (SRO), who are sworn law enforcement officers that work within school grounds to promote safety, in contributing to the school-to-prison pipeline. Past work has shown that SROs increase discipline and arrest rates in schools, especially for lower-level violations or school code of conduct violations, and students of color are disproportionately affected (Gottfredson et al, 2020; Owens, 2017; Sorensen et al, 2023; Weisburst, 2019). However, the quantitative literature does not account for how SRO programs interact with unique school- and state-level contexts and what impact these interactions have on program effects.

In this paper, we examine how the impact of SRO programs on student disciplinary and academic outcomes differs by school characteristics using a staggered difference-in-differences design. We use the Department of Justice’s COPS Hiring Program (CHP) grant award mechanism as plausibly exogenous variation in SRO hiring to account for the fact that school districts that are interested in hiring SROs may differ systematically from districts that do not. To measure school-level student outcomes, we use the Civil Rights Data Collection (CRDC). Compared to schools that never received the CHP grant, we find an increase in the number of students who received exclusionary discipline (expulsions) but a decrease in formal law enforcement action (school-based arrests, referrals to law enforcement) in schools that were awarded the grant. We also find that male students, students of color, and disabled students were disproportionately impacted. 

We then use both classical statistical bounds and machine learning methods to measure by how much these effects vary across schools and which school characteristics best explain these differences. Drawing from a review of the qualitative literature, we consider the following characteristics: whether the grant is funding a new SRO program; whether the school is in a state that requires SROs to receive training specific to working in a school; presence of a school counselor; school-level racial composition; and state laws that may inform a school’s discipline policies (state laws on mandatory expulsion; laws on using restorative justice in student discipline). We also use topic modeling, a machine learning technique, to capture thematic differences across schools in how they planned to implement their SRO programs using novel CHP grant narrative data. We find that schools that used the CHP grant to fund a new SRO program had a smaller increase in expulsions and a larger decrease in referrals relative to schools used the grant to expand an existing SRO program. We also find that effects varied based on how schools planned to implement their SRO programs. 

Our findings complement existing work by providing insight on how the effects of a school’s SRO program on its students may be context dependent, specifically exploring the interaction between the larger state policy environment and the implementation of a local school district-level policy. By identifying factors that magnify or dampen the impact of SRO hiring on student disciplinary and academic outcomes, our work can inform policymakers whose aim is to dismantle the school-to-prison pipeline. 

Author

• Hanna Han, University of Wisconsin-Madison
  Presenting Author