Disparate Impacts: Balancing the Need for Safe Schools With Racial Equity in Discipline

Tweet

Policy responses to school shootings have not prevented increases in their rate of occurrence, but restorative justice has the potential to avert bad behavior and school shootings.

Key Points

Introduction

IntroductionSafety and racial/ethnic inequality in school are two recurrent issues on the national policy agenda. In the 1980s, for example, the phrase “zero tolerance” emerged from the “War on Drugs,” as the federal government aimed to “get tough” on drug enforcement. Policies were later enacted in schools to discourage drug abuse and gang activity (Skiba & Peterson, 1999). In the 1990s, the Columbine High School tragedy was preceded by five highly publicized school shootings (Lawrence, 1998), which elicited concerns regarding an “epidemic” of deadly school violence (Muschert, 2007). Subsequently, an array of security measures were implemented in public schools, including school resource officers (SROs), metal detectors, and security cameras (Addington, 2009).

Although the effectiveness of these reforms in improving school safety is questioned with each additional school shooting, much less debate concerns their relation to increased contact with law enforcement in schools and higher rates of disciplinary exclusion for students of color. Research has documented the emergence of the school-to-prison pipeline, whereby excessive suspensions, expulsions, and referrals to law enforcement increase students’ contact with the criminal justice system (Curtis, 2014; Nance, 2016), most notably for students of color (Skiba, Arredondo, & Williams, 2014). Although empirical investigations have begun to reveal the relationships between school safety and racial/ethnic inequality, much is still unknown. We lack models of how school safety policies are effective/ineffective at producing safer schools (Hirschfield, 2018) and how they mitigate or exacerbate social stratification (Welsh & Little, 2018).

This article aims to address those questions by undertaking a critical review of research on disparate school discipline from 1999 (Columbine) to 2018 (Parkland). We frame this review with interests in the multiple ways racial/ethnic disparities in discipline manifest (e.g., between schools, within schools, directly, and indirectly), the policy phase in which these inequalities emerge (i.e., policy formulation and implementation), and the specific policy tools that are used to create racial disparities in school discipline, among them, the presence of law enforcement (e.g., SROs), exclusionary discipline (e.g., suspensions), mandatory sentencing (e.g., zero tolerance consequences), and cultural expression regulations (e.g., dress code). Although we clarify how these policy approaches vary between and within schools (Johnson, 2012), their effects when aggregated create the macro-structural concern of disparate racial and ethnic discipline. Finally, we consider policy alternatives for school safety and conclude that restorative justice (RJ) reforms may function between and within schools to decrease racial inequity.

Conceptualizing Disparities in Discipline

Using available research, we demonstrate how crisis prevention approaches in the formulation and implementation of school safety policies have led to racial inequalities in discipline between and within schools. Specifically, policy formulation refers to a system-level process in which resources are identified to achieve the goals of a specific educational agency (Dunn, 1994); policy implementation refers to the point at which the interpretation and enactment of regulations within schools may vary according to the knowledge and social backgrounds of school personnel and students (Lipsky, 1980). Ultimately, both between-school variation in policy formulation and within-school variation in its implementation directly and indirectly increase racial and ethnic disparities in discipline. This section introduces our conceptual model; the following sections then analyze research relevant to each part.

Schools have employed a variety of safety strategies (e.g., SROs and metal detectors) in response to mass shootings that aim to prevent crises (Muschert, Henry, Bracy, & Peguero, 2014). Although the actual impacts that these crisis prevention strategies have had on mass shootings are equivocal (Price & Khubchandani, 2019), these strategies have a collateral effect: When not being used to thwart the uncommon school shooting, they instead serve to increase the capacity of schools to identify and punish students for less serious offenses. These less serious offenses, like disorderly conduct and disruption of an educational environment, have consequently become more common than assault and weapons violations (Advancement Project, 2018; Theriot, 2009). For example, in the 2010-2011 school year, half of all California suspensions were for “willful defiance” (Watanabe, 2013). As the disruption of a school function is a criminal offense in California and 20 other states, the number of school-based arrest have skyrocketed. From 2005 to 2014, for example, police in San Bernardino, CA, arrested 6,923 minors on streets but more than 30,000 in schools (Ferriss, 2015). These increases are remarkable because national school victimization rates (Butts, 2000), homicides (Robers, Kemp, Rathbun, Morgan, & Snyder, 2014), and teacher reports of threats (Fox & Burstein, 2010) have declined and stabilized since the early 1990s.

At the school level, variation in policy formulation between schools has created learning environments that subject all students to more stringent technologies of surveillance, and subsequently more frequent and severe punishments (Shedd, 2015). These high social-control schools—with high rates of surveillance and punishment relative to the level of disorder and misbehavior—also subject non-offending students to negative, indirect, or collateral effects (Perry & Morris, 2014). An excessive reliance on discipline produces racial disparities when it occurs in schools with higher rates of segregation for Black and Latinx students, which increases their rates of exposure to surveillance and punishments (see Jabbari & Johnson, 2019a). As racial segregation tends to characterize entire metropolitan areas (Johnson, 2017) and segregated schools tend to have higher levels of social control (Jabbari & Johnson, 2019a), segregation severely limits the ability of Black and Latinx families to choose schools with proportionate rates of discipline and less racial disparity.

Exemplifying variation in policy implementation within schools, some schools disproportionately target Black and Latinx students for perceived misbehavior (Barrett, McEachin, Mills, & Valant, 2017) and treat them more harshly than White students for committing the same offenses (Young, Yancey, Betsinger, & Farrell, 2011). This targeting can occur (a) at the level of surveillance, in which minority students may be watched more closely within a given school (see Rios, 2011), (b) at the level of detection, in which minority students may be “caught” more frequently within a given school (Skiba et al., 2011), and (c) at the level of sanction, in which minority students may be punished more harshly within a given school (Young et al., 2011). Whether explicit or implicit, racial bias evidently explains disparities in discipline policy implementation within schools (Riddle & Sinclair, 2019). In fact, Riddle and Sinclair (2019) found the results of 1.6 million implicit bias test-takers were associated with racial disciplinary disparities in the 96,000 schools that serve test-takers’ communities and enroll roughly 32 million students.

There are many policy options that can be adopted to address the aforementioned disparities. Yet, the analysis that follows shows how many of them, such as reducing racial segregation or eliminating zero tolerance policies, have less potential than RJ approaches to curb racial disciplinary disparities. We begin with an overview of crisis prevention policy.

Crisis Prevention and Punishment

In our model, school safety concerns are addressed through crisis prevention strategies; however, in the absence of an emergency, these security measures yield excessive punishment. Although these policies may have stopped some shootings from occurring, they seem ineffective in mitigating the factors that cause them, because a recent study reports that more people have died or been injured in mass school shootings in the past 18 years than in the entire 20th century (Katsiyannis, Whitford, & Ennis, 2018). In 2018, for example, at least 24 mass shootings in K-12 settings had occurred, leaving 114 students killed or injured (Blad & Peele, 2019). Although metrics of school safety other than gun violence should be considered, little evidence shows that increasing punishment improves student behavior or academic performance (Losen & Skiba, 2010). Ironically, stronger evidence suggests that these policies are related to decreased mathematics outcomes and increased dropout rates, especially for racialized students, even after controlling for prior math achievement (Ibrahim & Johnson, 2019; Jabbari & Johnson, 2019b). The next sections describe how racial and ethnic inequalities in discipline arise within schools and between schools, directly and indirectly.

Policy Variation Within Schools

Disciplinary policy variation within schools matters for a few important reasons. First, disciplinary policy may be interpreted inconsistently or applied more harshly, loosely, or discriminately by school personnel according to the circumstances of the situation or student background (Lipsky, 1980). The enactment of varied and subjective definitions of what constitutes “order,” and educator discretion in the application of these definitions, often amounts to nontrivial racial and ethnic disparities when aggregated to the school and system levels. Although order might promote norms that enhance safety, it often extends from educational institutions’ mission in “character and moral education” to socialize youth in accordance with the habits and values (i.e., culture) that are often rewarded with social advancement (i.e., “cultural capital”). Research has extended the concept of cultural capital to include bodily self-representations (e.g., manners, comportment, and dress), with individuals who mirror normative cultural styles being given greater social resources and leniency for deviations from school protocols (Morris, 2005). When crisis prevention resources are applied to students who have been found in violation of cultural norms through these ostensibly subjective practices of cultural discernment, the phenomenon of “cultural policing” is formed and racial and ethnic inequality in discipline emerges. Recent manifestations of cultural policing include suspensions and expulsions for having natural hair (e.g., an afro), “locks/dreads” (i.e., a braided Black hairstyle), or hair extensions (Tate, 2017); “hoodies” (i.e., a shirt with an attached hood; Zacarias, 2019); and saggy pants (Broach, 2015).

Second, practitioner views about culture are associated with racial and gendered differences in student outcomes. For example, educators tend to associate students exhibiting Black culture, especially boys’ movement styles, with lower academic outcomes, higher aggression, and the need for special education services (Neal, McCray, Webb-Johnson, & Bridgest, 2003). Second, the perception of Black girls as not “ladylike” is significantly related in research to disciplinary referrals about gender-appropriate dress and manners (Morris, 2005). Third, research has shown that school administrators often perceive Latino boys as threatening and warranting punitive discipline (Morris, 2005). The racial/ethnic divide in discipline is only made greater by the discretion and exemption from cultural policing that school personnel enact for White and Asian students, who may have similar or worse behavior, but who receive less frequent and less severe penalties (Barrett et al., 2017).

Zero tolerance policy has been widely criticized because, in theory, it limits the professional discretion of school personnel in matters of discipline by standardizing responses to violations of school authority and law. However, the occurrence of cultural policing suggests that enhancing professional discretion through the elimination of zero tolerance would not resolve racial/ethnic disparities in discipline and may in fact further aggravate them. Ultimately, the culture of discipline in public schools aligns with a greater social order replicated in policy, which requires stricter adherence to a stereotypical construction of normative child and adolescent behavior from racialized youth than nonracialized students (Wun, 2016). These within-school variations suggest that parents using school choice policies to enroll their children in schools that have fewer punitive policies may not be able to escape the risk of discipline due to cultural policing.

Policy Variation Between Schools

In addition to within-school variation in policy implementation, discipline disparities also extend from between-school variation in policy formulation. Recent studies demonstrate that the majority of the disproportionality in discipline occurs between rather than within schools (Anderson & Ritter, 2017; Skiba et al., 2014). Thus, analyzing how school discipline policies vary between schools is essential in understanding how disparate impacts materialize. The origins of between-school differences in the formulation of discipline policy could consider (a) the built environment (e.g., metal detectors vs. open campuses), (b) school composition (e.g., racial segregation vs. integration), (c) student culture and behavior (e.g., strict vs. permissive regulation of dress and hairstyles), and (d) instruction and pedagogies (e.g., disseminating knowledge vs. exploration and the creation of knowledge), with many of these dimensions reflected in individual schools. Consider, for example, that schools with higher proportions of racial/ethnic minority students tend to have increased levels of surveillance in the built environment (Kupchik & Ward, 2014), more restrictions on dress and hairstyles (DaCosta, 2006; Ibrahim, Barnes, Butler-Barnes & Johnson, 2019), and more regimented and regulated approaches to learning (Goodman, 2013). These heightened regulations of behavior relate to higher dropout rates in predominantly minority schools (Jabbari & Johnson, 2019a).

In addition, although a segregated school’s higher proportion of students of color triggers an increase in the use of disciplinary measures (Blalock, 1967), racially heterogeneous schools may incur demands for increased safety measures from White parents (see Jencks & Mayer, 1990), due to their racial views and pervasive stereotypes of Black and Latino youth. Empirically, an increase in the percentage of Black students within a school directly relates to increased discipline and suspensions and decreased restorative practices, even after adjusting for school levels of disorder and misbehavior (Payne & Welch, 2010; Welch & Payne, 2010).

Although race/ethnicity is not the only mechanism of stratification between schools that informs discipline policies and outcomes, it seems to be a fundamental one. For example, many of the same policy tools used in schools with high proportions of minority students (e.g., increased surveillance, restrictions on dress and hairstyles, and regimented learning) are used inequitably within the distributions of other school-level metrics. For example, when schools are defined according to their socioeconomic status, these policies and practices are most prevalent in schools that are low-income (Mendez, Knoff, & Ferron, 2002). Similarly, schools that are urban (Shedd, 2015), Southern (Smith & Harper, 2015), charter (Losen, Keith, Hodson, & Martinez, 2016), “no-excuse” (Golann, 2015), large (Stewart, 2003), low achieving (Skiba et al., 2014), and have fewer teachers of color (Lindsay & Hart, 2017) tend to rely more heavily on punitive discipline. As most of these dimensions characterize the schools that Black and Latinx students attend, and do not fully account for racial disparities once they are considered (Peguero, Varela, Marchbanks, Blake, & Eason, 2018; Skiba et al., 2014), one could conclude that they merely proxy the pervasiveness of racial disparity in discipline. In sum, between-school variation in racial segregation leads to pronounced racial disparities in discipline, and the pervasiveness of segregation makes it difficult for parents to access less segregated schools where hyper-discipline could be less common.

Indirect Effects

What remains largely unacknowledged and worthy of policymakers’ attention are the unintended consequences of surveillance and punishment for non-offending students. For example, with the technologies of surveillance and exclusion in place—often under the guise of “safety”—a culture of control can dominate a learning environment (Nolan, 2011). No-excuse schools, for example, can cause children to “monitor themselves, hold back their opinions, and defer to authority” (Golann, 2015, p. 103), which limits important college and labor market skills, such as taking initiative, asserting one’s needs, and negotiating with authority (Golann, 2015). High schools with higher levels of exclusionary discipline not only lower the achievement of disciplined and non-disciplined students alike (Perry & Morris, 2014, p. 1071) but also their college attendance (Jabbari & Johnson, 2019a).

Indirect discipline effects also magnify racial and ethnic disparities in discipline. For example, criminological research shows that distrust of authority arises in Black youth vicariously as they see or hear about the troubling experiences of other African Americans with state agents (Brunson, 2007). The awareness of racial bias in school discipline may encourage future disciplinary infractions—ultimately leading to lower college enrollment (Yeager, Purdie-Vaughns, Hooper, & Cohen, 2017). In both high social control settings and environments with discipline disparities, the threat of undeserved punishment can increase anxiety (Kupchik, 2010), which can, in turn, undermine the moral authority of schools (Arum, 2003), affecting both well and poorly behaved students alike (Perry & Morris, 2014).

Policy Alternatives

Whether between or within schools, discipline policies have both direct and indirect effects that oftentimes disproportionately affect minority students (Ibrahim & Johnson, 2019; Jabbari & Johnson, 2019b). As a result, interventions must address the tendency for discipline policies to vary in both their formulation between schools and their implementation within schools. One alternative would alter how policies stemming from crisis prevention strategies are formulated and implemented, with an intent to reduce punishments in schools with high rates of surveillance and disciplinary sanctions.

The Philadelphia schools, for example, recently banned out-of-school suspensions (OSS) for low-level offenses and reduced the length of OSS for more serious offenses. These reforms, however, did not (a) receive compliance from the majority of schools, (b) reduce the number of suspensions for low-level conduct, (c) improve achievement for previously suspended students, and, ultimately, (d) increased racial disparities in discipline (Steinberg & Lacoe, 2017). Even alternatives to OSS, such as in-school-suspensions (ISS), have negative direct and indirect effects that rival those of OSS (Cholewa, Hull, Babcock, & Smith, 2018; Ibrahim & Johnson, 2019; Jabbari & Johnson, 2019a). Although Philadelphia’s approach might be consistent with disciplinary reform, it suffered from incomplete implementation and unintended policy outcomes (i.e., increased ISS).

Another policy option calls on schools with disproportionate rates of discipline to implement implicit bias training for their personnel. Although racial bias may drive discipline disparities both between and within schools, explicit and implicit bias training, alone, as a way to effectively curb school discipline has yet to be validated (Marcucci, 2019). However, at least two studies have found mandatory implicit training may increase employee resentment toward the groups it is supposed to ease (Dobbin & Kalev, 2016; Legault, Gutsell, & Inzlicht, 2011).

A third policy option prioritizes RJ to attain both safer schools and greater racial/ethnic equity. RJ is an alternative approach to the creation of safe and equitable communities. It centers on repairing harm if a wrongdoing occurs. Howard Zehr (2015) outlines three central questions of a restorative philosophy: “Who has been hurt? What are their needs? Who has the obligation to address the needs, to put right the harms, to restore relationships?” In contrast, normative, punitive approaches to justice ask, “What rules were broken? Who did it? What do they deserve?” (p. 91). Although crisis prevention attempts to address a narrow definition of school safety, restorative approaches understand school safety holistically (Zehr, 2015; see also Marcucci, 2016).

When applied in schools, RJ is a spectrum of practices, from informal restorative conversations to structured restorative curricula (Amstut & Mullet, 2015). The most well-known restorative practice is the circle process, which is central to both preventive and reactive uses of RJ. Talking circles, or community circles, are preventive practices that educators can use to build up relationships and community within school environments. Just as exclusionary discipline sets the tone of the school culture, so can RJ. In this case, RJ practitioners have codified practices to proactively build up a positive school community. Harm circles, however, occur after a wrongdoing has taken place. Both the offender and the victim, as well as other impacted community members, come together in a circle to discuss the event and decide on an appropriate response.

These restorative practices have been gaining momentum in American schools since the 1990s (Winn, 2018). Recently, major urban districts (Oakland Unified School District, Denver Public Schools, and Los Angeles Unified School District) have adopted some form of restorative initiative. In a few urban areas, restorative initiatives in schools have even been facilitated by local police, rather than school or district officials. For example, the Philadelphia Police Department spearheaded a collaboration with the local school district to implement a restorative initiative called Philadelphia Police School Diversion Program (City of Philadelphia, 2019). This program suggests that school-to-prison pipeline solutions may rest outside of educational policy as well.

Although complex, investigations into school-based restorative approaches indicate that restorative practices could both reduce overall rates of discipline and mitigate underlying racial disparities. In the first-ever randomized control trial of RJ, the RAND corporation (2018) and Pittsburgh Public Schools found that schools adopting restorative practices reduced the number of days suspended by 36% in the study time frame (compared with 18% reduction in the control group of schools, which was using alternative disciplinary reforms). In addition, racial discipline disparities, particularly for African American students, were reduced in schools that adopted restorative practices. In restorative schools, teachers reported a more positive climate overall, perhaps mitigating some of the indirect impacts of punitive environments. These school-level findings are supported by student-level findings (Anyon et al., 2016), showing that when students received more restorative interventions in the first semester, their chances of receiving a disciplinary referral the second semester were lower, regardless of student race. Restorative approaches may be disparity-mitigating, as well. Rehabilitative approaches to school discipline, including RJ, were more robust against teacher implicit bias (Marcucci, 2019). Regardless of school policy, teachers’ implicit bias influenced their punitive disciplinary behaviors more than their rehabilitative disciplinary behaviors. Policy variation within schools (i.e., which results from racial bias) may be mitigated with restorative initiatives. Restorative practice could produce more equitable schools, without compromising school safety.

Policymakers can use their position to prioritize restorative practices. First, policymakers can mandate the use of suspensions for only the most extreme misbehaviors. This will encourage schools and educators to use other tools—namely, restorative practices—to address more everyday, mundane issues of student misbehavior and school safety. A number of states have begun limiting schools’ use of OSS. For example, Illinois’s Public Act 99-0456 limits OSS over 3 days to those students who pose a threat to the safety of the school and bans the use of zero tolerance except when required by federal law. Policymakers can consider supporting similar legislation in their jurisdiction. Suspension bans should target unnecessary in-school suspensions as well.

Second, policymakers can support funding that will offer educators the training to implement and utilize restorative practices well. One of the main criticisms of the Illinois legislation is that it removes a primary tool of classroom management without offering educators alternatives. If policymakers couple suspension-limiting legislation with funding for training and professional development, school safety policies can become both more effective and more racially equitable.

Finally, policymakers can earmark research funds to continue to explore the impact of restorative initiatives in school communities. Although the initial evidence is exciting, additional research can show how to optimize implementation in certain communities, as well as the specific mechanisms that would make restorative practices produce positive student and school outcomes, particularly around school safety.

Conclusion

Moments of extreme school violence violate the basic assumption that students should be safe in schools. Tragic shootings in Columbine, Newtown, and Parkland are moments that foment school safety policies. However, the school safety policies that implemented after these moments (e.g., law enforcement, zero tolerance, metal detectors) do not avoid these tragic events. School shootings, while rare, still occur despite these crisis prevention policies, which inadvertently lead to harsher punishment of racialized students, exacerbating racial inequity in American schools. The current model showcases how school safety policies exacerbate racial inequity via between- and within-school policy variation. It also highlights how the detrimental effects of crisis prevention policies extend beyond the misbehaving student to impact their peers, as entire school communities become punitive environments focusing on social control over academic learning. These indirect effects of crisis prevention policies, therefore, contribute as significantly to racial inequity as the direct effects on disciplined students.

Given the shortcomings of the current crisis prevention approach to school safety, policymakers must consider alternative approaches. RJ provides an approach to school safety that could encourage investment in school communities, allow developmentally appropriate reactions to normative child and adolescent behavior, lower interpersonal and intergroup tensions, and, ultimately, promote safer school environments. In addition, restorative approaches could begin to mitigate the racial disparities in school discipline. The article outlines steps that policymakers could take to protect and encourage RJ in schools, including professional development funding, legislation that prohibits long suspensions for nonviolent offenses, and funding for relevant research.

Racial equity and safe schools are not in opposition. Racial equity is not a societal luxury, so it is not a reasonable sacrifice for safer schools. Instead, policies should promote safer schools and more equitable outcomes for students of all races, allowing children and youth to become contributing citizens to American democracy. These aims are bedrock for sustainable democracies and thriving economies.