Association for Public Policy Analysis & Management

The active patrolling of schools by police officers has expanded rapidly over the past several decades, in various efforts to promote proactive community policing, enforce zero-tolerance policies, and combat the wave of juvenile violence during the 1990s. During this same time period, the probability of a secondary student experiencing at least one suspension had increased by approximately around 40%, with a particularly dramatic increase for African-Americans of 140% (Losen & Martinez, 2013). Perhaps because of the common trend, there is now a close link between SROs and exclusionary discipline in both research and policy discussions about school discipline (Executive office of the President, 2016, Fisher & Hennessy, 2016, Gottfredson & Na, 2011).

Unfortunately, until recently, there has been very little strong empirical research documenting the impact of SROs on either positive (violence reduction) or negative school outcomes (Fisher & Hennessy, 2016; James & McCallion, 2013; Steinberg & Lacoe, 2017). A few new rigorous studies map the impact of SROs onto measurable outcomes, primarily using federal COPS grants as a proxy measure of SRO placement (Owens, 2017; Weisburst, 2019; Zhang, 2019). Together, these studies generally find that SROs increase school safety through deterring violent and other offenses, but also lead to increases in suspensions for minor offenses and long-term adverse impacts on eventual college enrollment and graduation.

In the current research, we build off of these studies by carefully isolating the separate effects of SROs on student disciplinary behavior – versus administrator disciplinary response, and by estimating long-term SRO impacts on criminal conviction and incarceration. We do so with a panel dataset of middle schools in North Carolina during the years 2005-2009. Our research design makes use of within-district variation over time in student exposure to school resource officers, facilitated by the fact that the state of North Carolina invested heavily in SROs in middle schools through a graduated rollout during the observed time period of interest.

We find that school resource officers do improve safety at their assigned middle schools, reducing the incidence of violent offenses by nearly 20 percent. In terms of mechanisms, it appears that these crime reduction effects may be created in part by increasing the chances that a student caught doing one of these offenses is referred to law enforcement by a whopping 127%. The presence of an SRO also appears to increase the chance of short-term out-of-school suspension (OSS) by 6% (19% in high-minority schools) and the chance of long-term OSS or expulsion or transfer to an alternative school by 11% (52% in high-minority schools). So although the largest impact is on referrals to law enforcement, there does appear to be a “trickle-down” effect whereby school administrators elevate their response to these serious offenses with increased use of suspensions. Despite this evidence that SROs alter student and administrator behaviors within the school, we do not detect any long-term impacts of SROs on high school graduation or criminal conviction rates.