The Culture of Control in Schools: How Punitive and Disadvantaged Spaces Impact Race-Specific Suspension Rates

Abstract

Across American societal institutions, a punitive culture of control and surveillance has manifested in a variety of ways, including exponential growth in incarceration rates and school suspension rates over the last four decades. To date, much of the scholarship exploring the relationship between criminal justice outcomes and school-based outcomes has focused primarily on how school punishment is consequential for future involvement in the justice system. What remains unclear, however, is whether an alternative relationship exists. That is, does a culture of control foster an environment where punitiveness in the criminal justice system is mirrored by punitiveness within schools? Drawing on carceral perspectives and place-based stratification theories and analyzing a random sample of Florida middle and high schools combined with school district data, several key findings emerge. Specifically, Black and Hispanic students are more likely to be suspended in places with higher incarceration rates; all students are more likely to be suspended in places with greater concentrated disadvantage; and Black and Hispanic students are significantly more likely to be suspended when attending schools in places with high incarceration rates and greater concentrated disadvantage. These findings highlight the interconnectedness of place and social control in the school setting.

Keywords

school punishment incarceration concentrated disadvantage culture of control

Introduction

The criminalization of American school students has been a consistent theme over the last four decades (Hirschfield, 2008; Morris, 2016; Rios, 2011). In response to real and perceived concerns over school violence, schools and school districts have embraced increased security and surveillance measures and zero-tolerance policies (Simon, 2007). This intensification of disciplinary and safety strategies in schools has coincided with increased office referrals (Rocque & Paternoster, 2011), detentions, suspensions, and expulsions (Gottfredson & Gottfredson, 2001). The Office of Civil Rights (OCR) reports that dating back to the 2000–2001 school year, even as student misconduct rates have declined, at least 3 million of the approximately 50 million students in U.S. schools have experienced exclusionary school punishment annually (U.S. Department of Education, 2018). Notably, students of color have disproportionately borne the brunt of this punitive school environment (Jacobsen et al., 2019; Bell, 2021). Compared to their White counterparts, Black and Hispanic students are more likely to be suspended (Government Accountability Office, 2018; Lehmann et al., 2021; Hughses et al., 2017), more likely to be suspended for their first infraction (Fabelo et al., 2011), more likely to receive office referrals (Rocque & Paternoster, 2011), more likely to be suspended for the same or lesser offenses (U.S. Department of Education, 2018), and less likely to receive restorative discipline (Welch & Payne, 2010). These disparities persist even after controlling for student misconduct (Gregory & Weinstein, 2008).

As scholars have attempted to explore the factors associated with the escalation of and the racial and ethnic disparities in school punishment, one explanation that has received less attention is incarceration. Given the 400% increase in the U.S. incarceration rate over the last 40 years, a present-day prison population of approximately 1.5 million, and the disproportionately stratifying effects of incarceration (Brayne, 2014), it is reasonable to consider whether the intensification in criminal justice punitiveness is indicative of a culture of control that is reflected across other societal contexts, especially schools (Alexander, 2010; Garland, 2001). In other words, is there a culture of control that is locally concentrated to the extent that incarceration and school punishment are associated? To date, existing scholarship has given this possibility less consideration.

Much of the research examining the relationship between schools and incarceration has explored whether school outcomes predict future involvement in the criminal justice system (Mowen & Brent, 2016; Pesta, 2018; Wolf & Kupchik, 2017). Although research supports the notion that school outcomes influence the likelihood of future criminal justice contact, this study considers whether an alternative relationship exists also—the possibility that punishment in the criminal justice system is associated with punishment in schools. That is, are incarceration rates associated with school suspension rates? Consequently, a growing literature has highlighted how the differences in punishment in the justice system and punishment in other institutions have become less distinguishable (Hirschfield, 2008; Wacquant, 2001). Because of this carceral mesh, the major themes of the modern justice system—surveillance, correction, and control—have been expanded and adopted across societal institutions, particularly schools (Fisher et al., 2019; Foucault, 1977; Garland, 2001). In addition, the intensification of punishment and control across the social body has been especially consequential for people of color, who are more likely to become ensnared in this carceral continuum (Wacquant, 2001). Given the increasing pervasiveness of carceral tactics and attitudes and the racialized nature of punishment and control, it is necessary to consider whether punishment in the criminal justice system is related to punishment in schools; and if it is, does such a relationship vary across race and ethnicity? This study attempts to answer these questions by exploring whether greater punitiveness in the judicial system (i.e., incarceration) is associated with escalations in punitiveness in schools (i.e., suspension).

In addition to the potential direct link between incarceration and suspension rates, the current research also considers a possible conditioning effect. Research has demonstrated that concentrated disadvantage is associated with a wide-range of criminological outcomes, including judicial decisions (Rodriguez, 2013), homicides (Kubrin & Weitzer, 2008), and arrests (Parker et al., 2005). In schools, disadvantage has been associated with several negative outcomes—for example, severe discipline (Payne & Welch, 2010) and increases in delinquency and teacher victimization (Gottfredson et al., 2005). Moreover, scholars have found concentrated disadvantage to be particularly salient for people of color, who are more likely to reside in such spaces (Wacquant, 2001; Wilson, 1987). However, less scholarship has explored whether concentrated disadvantage might influence punitive consequences in an already punitive environment; and if so, who is most-impacted by existing in such a space? Drawing on place-based stratification theories, this study explores whether the potential relationship between incarceration rates and school suspension rates is conditioned by the level of concentrated disadvantage in the school district where a school is located. In addition, do the effects of such a relationship, if it exists, vary across students’ race and ethnicity?

Utilizing a sample of 559 middle and high schools embedded within 66 counties, incarceration rates averaged between 2004 and 2010, and suspension rates averaged between 2011 and 2013, this study examines the following research questions: (a) whether county-level incarceration rates are associated with race-specific school suspension rates, (b) whether county-level concentrated disadvantage is related to race-specific school suspension rates, and (c) whether the relationship between county-level incarceration rates and race-specific school suspension rates is moderated by county-level concentrated disadvantage.

Theoretical Framework and Prior Research

Mass Incarceration and the Culture of Control

Mass incarceration has been a constant feature of American society for nearly half a century. Beginning in 1975, the U.S. incarceration rate rose every year, resulting in over 2 million incarcerated individuals by 2003 (Travis et al., 2014). These numbers remained relatively stable until the beginning of the COVID-19 pandemic in early 2020, which saw the release of approximately 8% of the state and federal prison population to reduce overcrowding (Gaines, 2021; Kang-Brown et al., 2021). Nevertheless, the growth in incarceration and the overall metastasis of the carceral state in America has corresponded with larger caseloads, more criminal justice employees, greater expenditures, increased prison constructions, and fewer community sentences (Gottschalk, 2015). Notably, this carceral expansion has disproportionately disadvantaged people of color, fueling what Michelle Alexander (2010) calls “The New Jim Crow.” Specifically, of U.S. residents born in 2001, 1 in 3 Black men, 1 in 6 Latino men, 1 in 18 Black women, and 1 in 45 Latino women can expect to be incarcerated during their lifetime (Bonczar, 2003).

In addition to the rise in punitiveness and the disparate outcomes experienced by people of color in the criminal justice system, the era of mass incarceration has also coincided with a proliferation of control and surveillance mechanisms throughout American life (Garland, 2001; Lopez-Aguado, 2018). Indeed, research has demonstrated that the tentacles of mass incarceration are long, often reaching into various societal contexts, including the workplace (Western, 2006), familial life (Siennick et al., 2014), neighborhoods and communities (Rose & Clear, 1998), and schools (Jacobsen, 2019). Blomberg and Lucken (2010) note this carceral spread, asserting “that American life in the new millennium [is] subject to new and far-reaching levels of control” beyond the justice system (265). This is also consistent with Foucault’s (1977) carceral continuum perspective, which argues that the punitive techniques and strategies that characterize prisons, jails, and the criminal justice system more broadly, have emerged across society. Accordingly, the justice system has not only normalized control and punishment techniques but also modeled how those techniques might be employed across other societal institutions (Foucault, 1977; Garland, 2001; Haggerty & Ericson, 2006). For instance, companies and corporations have developed private security apparatuses intended to neutralize threats of crime (Beck & Willis, 1995). Privately owned and administered public spaces—for example, shopping malls—search, surveil, and restrict the entrance of potential customers (Grabosky, 1992). Similarly, private citizens depend on surveillance mechanisms—security alarms, locks, and keyless entrance devices—to ensure the safety and security of their homes and communities (Blakely & Snyder, 1997; Grabosky, 1992).

Most notably and relevant to the current discussion, the technologies and machinery of the mass incarceration era—uniforms, IDs, metal detectors, zero-tolerance policies, harsh punishments, and so on—have manifested in schools as well, particularly in how schools are surveilled and how students are disciplined (Hirschfield, 2008; Kupchik, 2010; Simon, 2007). For example, by the 2015–16 school year, it was estimated that 57% of American public schools had school resource officers on campus at least 1 day a week, 94% implemented door locks, 93.5% required visitor check-ins, and 80.5% used cameras and closed-circuit televisions to monitor school grounds (Musu-Gillette et al., 2018). In addition, during the 2017–2018 school year, exclusionary discipline resulted in over 11 million days of missed student instruction (Losen and Martinez, 2020). This combination of increased security and surveillance as well as intensified student punishment has fostered what Shedd (2015) calls the “school disciplinary superstructure” and resulted in schools approximating detention centers (Meiners, 2013).

The Symbiosis Between Incarceration and School Punishment

At the intersection of punishment in the criminal justice system and punishment in schools, specific themes emerge that elucidate why and how incarceration may be associated with school punishment.

First, analogous to public concern with the prevalence of crime and maintaining law and order within communities, school stakeholders—particularly parents, teachers, and administrators—are similarly invested in curbing school violence and ensuring a safe learning environment. According to Simon (2007), “few issues keep parents awake [more than] the latest story [about] crime in schools” (208). Teachers and school administrators often prefer clearly defined disciplinary policies that simplify the student punishment process, decrease their responsibility to make subjective decisions, and reduce the likelihood of litigation (Hirschfield, 2008). Moreover, access to federal and state funds is often dependent on how successful teachers and administrators are in promoting and maintaining adequate levels of school safety and security. Such approaches in schools approximate the initiatives and policies that characterize the criminal justice system—for example, determinate sentencing guidelines, get-tough policies, and the tying of state and federal funds to crime control strategies (Alexander, 2010). Irrespective of the context, however, efforts to address real or perceived concerns about crime and safety, both in communities and schools, have reinforced and perpetuated the culture of control in America (Tonry, 1999).

Second, the increasingly centralized approach to security and discipline in schools is comparable to the sentencing guidelines that federal and state criminal justice systems use to inform sentencing decisions. Specifically, schools have embraced not only the goals of maintaining order and discouraging delinquency but also the criminal justice system’s emphasis on control and surveillance intended to identify, prevent, and punish problem behavior. As concerns over school safety intensified due to high-profile school shootings, the Gun-Free Schools Act (1994) extended the reach of zero-tolerance into schools and restricted the ability of teachers and administrators to consider mitigating factors when dispensing student discipline (Hirschfield, 2008). Although initially intended to address concerns over weapons and drugs in schools, this legislation’s scope was eventually broadened to cover lesser instances of student misconduct as well. Thus, the expansion of the punishable behaviors under zero-tolerance policies has amplified the use of punitive student discipline, particularly suspensions and expulsions (Losen & Skiba, 2010).

Third, schools have further approximated a culture of control and surveillance by instituting a number of control strategies with origins in the criminal justice system (Shedd, 2015). Schools regularly use metal detectors at entrances, uniforms, cameras, locker searches, drug-sniffing dogs, and school police who report directly to law enforcement agencies (Kupchik, 2010). Simon (2007) notes that this “merging of the school and penal system has. . . tilt[ed] the administration of schools toward an authoritarian and mechanistic model” (209). Furthermore, Wacquant (2001) argues that this “carceral atmosphere [in] schools” orients and familiarizes students with the tactics of the criminal justice system and produces schools that “operate as institutions of confinement whose primary mission is not to educate but to ensure ‘custody and control’” (108).

Fourth, as the culture of control further infiltrates schools and as zero-tolerance policies have become more pervasive in schools, students of color have disproportionately paid a punishment penalty similar to the imprisonment penalties paid by defendants of color in the criminal justice system (Jordan & Freiburger, 2010; Spohn & Holleran, 2000). Black and Latino students are not only more likely to be severely punished compared to their White counterparts but also more likely to be viewed as threatening or deviant by teachers and administrators (Morris, 2016; Owens & McLanahan, 2020; Marchbanks et al., 2018; Bell, 2021). Furthermore, compared to White students, Black students are less likely to receive warnings for behavioral infractions and more likely to receive more severe nonexclusionary punishments (Wegmann & Smith, 2019). Just as the criminal justice system’s emphasis on control and surveillance has had a disparate effect on people of color (Spohn & Holleran, 2000; Steffensmeier & Demuth, 2000), a similar era of control and surveillance in schools has disadvantaged students of color. In both the criminal justice system and schools, an overarching theme emerges. That is, these institutions have served as mechanisms to “neutralize [those] considered unworthy and unruly” (Wacquant, 2001, p. 108).

Finally, in addition to the parallels between punishment and social control in the criminal justice system and in the school setting, existing scholarship has also highlighted how the effects of mass incarceration and the culture of control often spill over into the school setting (Haskins & Jacobsen, 2017; Jacobsen, 2019; Turney & Haskins, 2014). For instance, Turney and Haskins (2014) find that children of incarcerated parents are significantly more likely to experience grade retention. Similarly, Hagan and Foster (2012) find that paternal imprisonment is associated with lower student grade point average (GPA), lower educational attainment, and reduced likelihood of receiving a college degree. Moreover, the impact of parental incarceration has also been found to be consequential for school punishment in particular. Specifically, Jacobsen (2019) finds that children with an incarcerated father experience greater odds of suspension and expulsion from school. Taken together, the effects of incarceration on school-based outcomes, especially school punishment, along with the expansion of carceral techniques into the school setting imply that punitive approaches in the criminal justice system not only are comparable but also may be empirically associated with the punitive approaches in schools.

The implications flowing from these themes suggest that in places where the criminal justice system employs more punitive responses to crime, schools are likely to embrace similarly punitive approaches to school crime and do so in ways that vary across race and ethnicity. This embrace of the culture of control and the punitiveness that characterizes it gives rise to the following hypotheses:

Hypothesis 1 (H₁): Total school suspension rates will be greater in counties where the incarceration rate is higher.

Hypothesis 2 (H₂): Black and Hispanic suspension rates will be higher in counties where the incarceration rate is higher.

Hypothesis 3 (H₃): White suspension rates will be unchanged in counties where the incarceration rate is higher.

The Role of Concentrated Disadvantage

Across various collectivities, concentrated disadvantage has been linked with a host of criminological outcomes, including violence (Johnson & Kane, 2018; Krivo & Peterson, 2000), social isolation, residential segregation (Wilson, 1987), and legal cynicism (Sampson & Bartusch, 1998). Place-based stratification theories contend that disadvantaged places are high in poverty, residential mobility, and population heterogeneity (Shaw & McKay, 1942). As a result, residents struggle to develop and maintain meaningful social ties (Sampson & Groves, 1989), social capital (Rose & Clear, 1998), collective efficacy, and informal social control (Sampson et al., 1997). In addition to hindering the development of prosocial community processes and increasing the likelihood of adverse outcomes, concentrated disadvantage has also been found to be consequential for the application of formal social control (Kubrin & Weitzer, 2008). For instance, Sampson and Bartusch (1998) conclude that police officers are more likely to be perceived as unresponsive to residents’ issues in disadvantaged communities. Similarly, police misconduct is more pronounced in disadvantaged areas compared to affluent areas (Kane, 2002).

Not only are disadvantaged places surveilled and policed differently, they also reflect the culture of control that characterizes the justice system. Wacquant (2001) argues that there is a “deadly symbiosis” between prisons and disadvantaged places. Specifically, the racial and class segregation of the Black inner-city closely resembles the disproportionate racial and class makeup that characterizes American jails and prisons. Impoverished, low-wage working people of color who commonly reside in disadvantaged, structurally- and socially-isolated spaces are the same poor, individuals of color who are overrepresented in prisons and jails (Wacquant, 2001). In addition, residents live in insufficient and unsafe public housing where they are regularly subjected to electronic monitoring, random searches, police shakedowns, curfews, and ID-card checks. Children attend deteriorating, underfunded schools that are often overcrowded and supervised by unqualified teachers and administrators. These factors combined with “on-the-ground extensions of the penal system”—courts, police, and probation and parole officers—have fostered an atmosphere of surveillance and control in disadvantaged spaces comparable to the administration and organization of the criminal justice system (Brayne, 2014; Wacquant, 2001, p. 107).

This culture of control in disadvantaged places is also particularly relevant for people of color. As noted by Wilson (1987), disadvantaged areas are more likely to be populated by socially- and economically-unprotected residents of color, who lack jobs, resources, and conventional social networks. Thus, the adverse effects of disadvantage may be amplified for people of color compared to Whites, who are more likely to reside in areas of greater stability. Even the most disadvantaged areas in which Whites live are demonstrably better than the disadvantaged places where people of color reside (Sampson, 1987). Therefore, formal social control outcomes in disadvantaged contexts are likely to exhibit significant racial and ethnic variation.

The current exploration considers how disadvantage may be associated with a measure of formal social control in schools, specifically race-specific suspension rates. Because schools do not exist in vacuums, contextual factors such as concentrated disadvantage and the intensification of social control therein are likely to be associated with school outcomes—for example, school disorder, violence (Laub & Lauritsen, 1998), suspensions, arrests (Kirk, 2009), and school achievement (Owens, 2010; Wodtke et al., 2011). Specifically, Wodtke and colleagues (2011) find that neighborhood disadvantage significantly reduces graduation rates. Similarly, Thompson (2002) finds neighborhood quality to be associated with achievement test scores. Moreover, Levy (2019) finds that students who live in places characterized by concentrated poverty are significantly less likely to graduate from college. To date, however, there has been less consideration of how concentrated disadvantage might influence school punishment both in punitive places and across various racial and ethnic student populations. This study addresses this shortcoming in the literature in two ways. First, it explores whether disadvantage is associated with suspension rates by considering the following hypotheses:

Hypothesis 4 (H₄): Total school suspension rates will be higher in counties where the level of concentrated disadvantage is greater.

Hypothesis 5 (H₅): Black and Hispanic suspension rates will be higher in counties where the level of concentrated disadvantage is greater.

Hypothesis 6 (H₆): White suspension rates will be unchanged in counties where the level of concentrated disadvantage is greater.

Second, this study explores whether the potential relationship between incarceration and school punishment is moderated by the level of concentrated disadvantage in the counties where schools are located. Exploring this interaction is warranted for several reasons. First, concentrated disadvantage has been found to be especially consequential for school-based outcomes. For example, a meta-analysis conducted by Pusch (2021) finds that, across 75 studies and 30 unique data sets, concentrated disadvantage is a significant predictor of delinquency in schools. Second, high incarceration rates are often concentrated in places where disadvantage is greater (Rose & Clear, 1998), and conversely, spaces characterized by greater concentrated disadvantage are likely to produce intensified social control in what Sampson and Loeffler (2010) describe as a “negative-feedback loop” (p. 29). Thus, the combination of high incarceration and greater disadvantage in a particular area is likely to be consequential for social control outcomes—for instance, school suspensions—in that space. Third, prior research has found that school outcomes, including the administration of school discipline, may be conditioned by disadvantage (Levy et al., 2019; Ramey, 2015; Shedd, 2015). For instance, Ramey (2015) finds that the effects of racial composition on medicalized and criminalized school discipline are conditioned by school- and district-level disadvantage. Similarly, Owens (2010) concludes that neighborhood disadvantage moderates the effect of school racial composition on educational attainment. These findings suggest that disadvantage might moderate the relationship between incarceration and school punishment. Fourth, given that people of color are more concentrated in neighborhoods characterized by high incarceration and greater disadvantage (Wacquant, 2001), an investigation of race-specific suspension rates merits consideration in a potential interaction between incarceration rates and concentrated disadvantage. These reasons give rise to a final hypothesis:

Hypothesis 7 (H₇): The positive effect of county-level incarceration rates on the suspension rates for students of color will be stronger in counties with greater levels of concentrated disadvantage.

Data and Methods

Sample

The data for this study combine 2011–2013 school and school district data from the Florida Department of Education (FDOE) with incarceration data from the Florida Department of Corrections (FDOC), county-level demographic data from the U.S Census and crime data from the Uniform Crime Report. Each academic year, the FDOE gathers comprehensive information from Florida schools and school districts.¹ Across 66 Florida counties (or school districts), 287 public middle schools and 272 public high schools were randomly sampled. To be eligible to be included in the sample, each school had to be opened or founded prior to or for the 2004–2005 academic school year because the analyses include a lagged suspension measure that captures the number of suspensions in each school during that academic year. Thus, the data are arrayed on two levels: schools (level-1) embedded within counties (level-2).

Measures

Dependent variables

This study utilizes the total out-of-school suspension rate and race-specific suspension rates for Black, Hispanic, and White students during the 2011–2012, 2012–2013 school years. Therefore, the four dependent variables are total suspension rate, Black suspension rate, Hispanic suspension rate, and White suspension rate. These measures capture the rate (per 100 students) of suspension for the total student population in addition to Black, Hispanic, and White students, respectively, for each school in the sample. These suspension rate measures were averaged across the two academic years to adjust for possible fluctuations.

Key predictor variables

The two key independent variables, incarceration rate and concentrated disadvantage, are measured at the county-level. Incarceration rate captures the average incarceration rate (per 1,000 residents) in jails and prisons for each Florida county between the years 2004 and 2010. The FDOC provides the yearly incarceration rates for each Florida county. These yearly county incarceration rates were averaged across the 7 years between 2004 and 2010. This approach temporally positions this key predictor variable before the dependent variables and accounts for fluctuations in county incarceration rates over the 7 years included in the measure.

Concentrated disadvantage is a 5-item standardized scale combining the following U.S. Census measures: percent living below the poverty line, percent living on public assistance, percent female-headed households, percent with less than a high school diploma, and percent unemployed. The Cronbach’s alpha for this scale is .63.

Control Variables

Consistent with prior research, this study controls for a number of school- and county-level factors that have been associated with school-based outcomes, particularly punishment (Gottfredson, 2001). Controlling for the following factors accounts for a host of variables that may confound the association between the key predictor variables and the dependent variables in this study. At the county-level, the following control variables are included: percent male, high school dropout rate, crime rate, teacher experience, school resource officers, and racial/ethnic heterogeneity. Percent male is the percentage of male residents in each county. High school dropout rate represents the percentage of students who dropped out of school across each county. The crime rate represents the number of index crimes per 100,000 people in each county. Teacher experience is the average experience (in years) of the teachers in each county. School resource officers is the total number of law enforcement officers in the middle and high schools within each county. Racial/ethnic heterogeneity is captured using the Herfindahl–Hirschman Index. This index captures the concentration of racial and ethnic groups within each county. Higher scores on this scale indicate more racial and ethnic diversity. The Herfindahl–Hirschman Index is calculated using the following formula:

Herfindahl–Hirschman Index = 1 – (percent Black)² + (percent Hispanic)² + (percent White)²

At the school-level, control variables include: percent Black and Hispanic, school size, percent school poverty, student–teacher ratio, school misconduct, school grade, middle school, and prior suspensions. Percent Black and Hispanic captures the percentage of Black and Hispanic students in each school. School size reflects the enrollment size of the school. Percent school poverty represents the percentage of students in each school who are eligible for free and/or reduced lunch. Student–teacher ratio is the number of students per every one teacher. School misconduct captures instances of school crime, including assaults, theft, weapon possession, and drug offenses. A composite measure of school crime was created and reflects the rate of school misconduct per 100 students. The alpha coefficient for the scale is .85. Higher scores on the school misconduct scale indicate higher levels of school-related misbehavior. School grade is a measure of academic achievement that captures the academic standing of each school in the sample. Schools are annually awarded a school grade (A through F) based on students’ performance on the Florida State Assessment (FSA) exams. Higher scores indicate high academically performing schools. Middle school is a dichotomous measure in which middle schools are coded as “1” and high schools are coded as “0.” The prior suspensions measure is the total number of suspensions for each school during the 2004–2005 school year.² This lagged variable is intended to capture a school’s history of punitiveness.³

Analytic Strategy

This study uses multilevel modeling techniques to examine the effects of incarceration rates and concentrated disadvantage on race-specific school suspensions. Multilevel modeling is customary for estimating contextual effects when one unit of analysis is clustered within a second unit of analysis (Raudenbush & Bryk, 2002). The 559 schools in the sample are nested within 66 school districts (or counties). Multilevel models recognize that schools within a particular school district may be more similar to one another than to schools in another school district and, therefore, may not constitute independent observations. Failure to account for nonindependence of observations can result in standard errors that are biased downward, increasing the chance of reaching biased conclusions (Raudenbush & Bryk, 2002).

In addition, two dependent variables, the Black and Hispanic suspension rates, show evidence of overdispersion. That is, the standard deviations for the Black and Hispanic suspension rate variables are higher than their means (Table 1). The appropriate approach to modeling such data is to estimate a negative binomial model. To ensure that negative binomial regression is the most appropriate statistical method for model estimation, tests for overdispersion were conducted, assessing the hypothesis that the mean dispersion parameter is equal to 0 (H₀: K = 0) against the alternative hypothesis that the mean dispersion parameter is greater than 0 (H_a: K > 0). In both instances, evidence suggests that the null hypothesis can be rejected, which assumes that the variance and the mean are equal (K = 0). Negative binomial regression, therefore, is the most appropriate approach. In the total suspension and White suspension models, evidence of overdispersion was not found; therefore, multilevel Poisson regression is utilized.⁴

Table 1.

Descriptive Statistics.

Variable	M	SD	Minimum	Maximum	N
Dependent variables
Total suspension rate	12.02	7.40	0	42.86	559
Black suspension rate	23.05	26.49	0	123.15	559
Hispanic suspension rate	13.81	16.32	0	121.14	559
White suspension rate	11.14	7.01	0	302.80	559
Key predictor variables
Incarceration rate	8.22	3.63	1.39	19.73	66
Concentrated disadvantage	0.07	.01	0.04	0.12	66
Control variables
County-level
Percent male	0.49	.02	.48	0.63	66
High school dropout rate	2.11	1.15	0.23	4.90	66
Crime rate	4,137.92	1,203.92	570.20	6,098.87	66
Teacher experience	12.50	1.30	9.27	17.36	66
School resource officers	28.44	25.13	0	78	66
Racial/ethnic heterogeneity	1.49	0.17	.81	1.82	66
School-level
Percent Black and Hispanic	.21	.19	0	.98	559
School size	1,313.95	671.38	120	4,192	559
Percent school poverty	0.53	0.20	.06	1	559
Student-teacher ratio	17.07	3.04	2.82	45	559
Student misconduct	0.01	0.10	0	0.39	559
School grade	4.1	.96	1	5	559
Middle school	0.51	—	0	1	559
Prior suspensions	202.4	144.51	0	892	559

Results

Direct Effects of Incarceration and Concentrated Disadvantage on Suspension Rates

Total suspension rates

Table 2 presents the Poisson regression analysis of the direct effects of county incarceration rates and concentrated disadvantage on the total suspension rate. Models 1 and 2, respectively, present the independent effects of the county incarceration rate and concentrated disadvantage on the total suspension rate. In Model 1, the effect of the incarceration rate on the total suspension rate is insignificant, suggesting that the likelihood of suspension among the general student population is not affected by the county incarceration rate. In Model 2, the effect of concentrated disadvantage on the total suspension rate is positive and significant (b = .28), indicating that the overall suspension rate is higher in schools located in counties where concentrated disadvantage is greater. Model 3 presents the effects of both the incarceration rate and concentrated disadvantage on the total suspension rate in a full regression model. These results are consistent with the effects illustrated in Models 1 and 2. The county incarceration rate does not significantly affect the total suspension rate and concentrated disadvantage (b = .26) is positively and significantly associated with the total suspension rate.

Table 2.

Multilevel Poisson Regressions of Direct Effects of County Incarceration Rates and Concentrated Disadvantage on Total Suspension Rates.

	Model 1		Model 2		Model 3
Variable	B	SE	B	SE	B	SE
Key county predictors
Incarceration rate	.01	.10	—	—	.04	.10
Concentrated disadvantage	—	—	.28*	.10	.26*	.10
County-level controls
Percent male	−.57*	.20	−.49*	.21	−.50*	.21
High school dropout rate	−.07^†	.04	−.05	.04	−.06	.04
Crime rate	.08^†	.04	.11*	.04	.10*	.04
Teacher experience	.02	.03	.03	.03	.03	.03
Racial/ethnic heterogeneity	.62*	.31	.38	.32	.40	.33
School resource officers	−.20*	.04	–.18*	.03	−.17*	.03
School-level controls
Percent Black and Hispanic	.19	.19	.09	.19	.10	.19
School size	.51*	.05	.52*	.05	.52*	.05
Percent school poverty	.07*	.02	.09*	.02	.09*	.02
Student–teacher ratio	.14^†	.08	.12	.84	.12	.84
School crime rate	.46*	.18	.42*	.18	.42*	.18
School grade	−.12*	.03	−.12*	.03	−.12*	.03
Middle school	.20*	.06	.18*	.06	.18*	.06
Prior suspensions	.14*	.02	.10*	.01	.10*	.02
Log likelihood	−3,062.61		−3,058.49		−3,058.44
Chi²	599.47*		605.49*		605.19*

Note. N = 66 counties, 559 schools.

†

p ≤ .10. * p ≤ .05.

While these analyses explore the effects of the key independent variables on the total suspension rate, it is possible that the effects of incarceration and disadvantage on suspension rates vary across race and ethnicity. To test for these race- and ethnicity-specific effects, the analysis will now explore the effects of incarceration and disadvantage on school suspension rates in race-specific models.

Black suspension rates

Table 3 presents the negative binomial regression analysis of the direct effects of county incarceration rates and concentrated disadvantage on Black suspension rates. In Model 1, the effect of the county incarceration rate on the Black suspension rate is positive and significant (b = .36), suggesting that Black students are more likely to be suspended when attending school in counties characterized by higher incarceration rates. In Model 2, the effect of concentrated disadvantage on Black suspension rates is also positive and significant (b = .34), indicating that Black students are suspended at a higher rate in counties where concentrated disadvantage is greater. Model 3 presents the effects of both the county incarceration rate and concentrated disadvantage on the Black suspension rate in a full regression model. The effects of both county incarceration rates (b = .46) and concentrated disadvantage (b = .42) on the Black suspension rate are positive and significant, net of the control variables included in the model. Controlling for theoretically-relevant variables, a one-unit increase in the county incarceration rate yields a 58% increase in the Black suspension rate and a one-unit increase in the level of concentrated disadvantage corresponds with a 52% increase in the Black suspension rate.⁵

Table 3.

Multilevel Negative Binomial Regressions of Direct Effects of County Incarceration Rates and Concentrated Disadvantage on Black Suspension Rates.

	Model 1		Model 2		Model 3
Variable	B	SE	B	SE	B	SE
Key county predictors
Incarceration rate	.36*	.15	—	—	.46*	.13
Concentrated disadvantage	—	—	.34*	.12	.42*	.12
County-level controls
Percent male	−.92*	.29	−.65*	.29	−.75*	.30
High school dropout rate	−.16*	.05	−.13*	.05	−.15*	.05
Crime rate	−.16*	.07	−.07	.07	−.13*	.07
Teacher experience	.09*	.04	.10*	.04	.09*	.04
School resource officers	−.10*	.04	−.19*	.04	−.15*	.04
Racial/ethnic heterogeneity	−.22	.41	−.81*	.41	−.45	.42
School-level controls
Percent Black and Hispanic	.17*	.03	.14*	.03	.17*	.03
School size	.22*	.07	.19*	.07	.24*	.07
Percent school poverty	.02	.03	.02	.03	.03	.03
Student–teacher ratio	.10	.11	.05	.12	.06	.12
Student misconduct	.30	.24	.19	.25	.21	.24
School grade	−.11*	.04	−.13*	.04	−.11	.04
Middle school	.17*	.08	.10	.08	.13	.08
Prior suspensions	.20*	.02	.19*	.02	.17*	.02
Log likelihood	−2,588.48		−2,588.92		−2,585.36
Chi²	406.89*		408.97*		431.13*

Note. N = 66 counties, 559 schools.

†

p ≤ .10. * p ≤ .05.

Hispanic suspension rates

Table 4 presents the negative binomial regression analysis of the direct effects of county incarceration rates and concentrated disadvantage on Hispanic suspension rates. In Model 1, the effect of the county incarceration rate on the Hispanic suspension rate is positive and significant (b = .26), suggesting that Hispanic students are more likely to be suspended when attending school in counties characterized by higher incarceration rates. In Model 2, the effect of concentrated disadvantage on Hispanic suspension rates is also positive and significant (b = .43), indicating that Hispanic students are suspended at a higher rate in counties where concentrated disadvantage is greater. Model 3 then presents the effects of both the county incarceration rate and concentrated disadvantage on the Hispanic suspension rate in one model. The effects of both county incarceration rates (b = .25) and concentrated disadvantage (b = .46) on the Hispanic suspension rate are positive and significant. For every one-unit increase in the county incarceration rate, the Hispanic suspension rate increases by 28% and for every one-unit increase in the level of concentrated disadvantage, the Hispanic suspension rate increases by 58%. These effects hold while controlling for theoretically-relevant control variables.

Table 4.

Multilevel Negative Binomial Regressions of Direct Effects of County Incarceration Rates and Concentrated Disadvantage on Hispanic Suspension Rates.

	Model 1		Model 2		Model 3
Variable	B	SE	B	SE	B	SE
Key county predictors
Incarceration rate	.26*	.11	—	—	.25^†	.14
Concentrated disadvantage	—	—	.43*	.14	.46*	.14
County-level controls
Percent male	−.41	.33	−.06	.34	−.12	.34
High school dropout rate	−.02	.06	−.03	.06	−.02	.06
Crime rate	−.04	.07	.05	.07	.03	.08
Teacher experience	−.05	.05	−.04	.05	−.05	.05
School resource officers	−.11*	.04	−.16*	.04	−.14*	.04
Racial/ethnic heterogeneity	.13*	.05	.09^†	.04	.10*	.05
School-level controls
Percent Black and Hispanic	.94*	.26	.80*	.26	.83*	.26
School size	.64*	.06	.63*	.06	.64*	.06
Percent school poverty	.07*	.03	.08*	.03	.09*	.03
Student–teacher ratio	.09	.11	.04	.12	.06	.12
Student misconduct	.63*	.26	.60*	.26	.61*	.26
School grade	.01	.30	.01	.40	.09	.40
Middle school	.27*	.08	.24*	.08	.24*	.08
Prior suspensions	.90*	.20	.95*	.21	.88*	.22
Log likelihood	−2,175.44		−2,171.25		−2,169.66
Chi²	318.48*		331.89*		332.73*

Note. N = 66 counties, 559 schools.

†

p ≤ .10. * p ≤ .05.

White suspension rates

Table 5 presents Poisson regression analysis of the direct effects of county incarceration rates and concentrated disadvantage on White suspension rates. In Model 1, the effect of the county incarceration rate on the White suspension rate is insignificant, suggesting that the county incarceration rate is not associated with the likelihood of White students being suspended. In Model 2, the effect of concentrated disadvantage on White suspension rates is positive and significant (b = .22), indicating that White students are suspended at a higher rate in counties where concentrated disadvantage is greater. Model 3 then presents the effects of both the county incarceration rate and concentrated disadvantage on the White suspension rate in a single regression model. While the effect of the county incarceration rate remains insignificant, the effect of concentrated disadvantage (b = .23) on the White suspension rate remains positive and significant in Model 3. For every one-unit increase in the level of concentrated disadvantage, the White suspension rate increases by 26%.

Table 5.

Multilevel Poisson Regressions of Direct Effects of County Incarceration Rates and Concentrated Disadvantage on White Suspension Rates.

	Model 1		Model 2		Model 3
Variable	B	SE	B	SE	B	SE
Key county predictors
Incarceration rate	−.02	.10	—	—	.04	.10
Concentrated disadvantage	—	—	.22*	.09	.23*	.09
County-level controls
Percent male	−.55*	.21	−.44*	.22	−.45*	.22
High school dropout rate	−.07^†	.04	−.06	.04	−.06	.04
Crime rate	.08^†	.05	.11*	.04	.10*	.05
Teacher experience	−.03	.03	.01	.03	.01	.03
School resource officers	−.17*	.03	−.18*	.03	−.17*	.03
Racial/ethnic heterogeneity	−.01	.37	−.23	.37	−.21	.37
School-level controls
Percent Black and Hispanic	−.29*	.02	−.29*	.02	−.29*	.02
School size	.65*	.05	.65*	.05	.65*	.05
Percent school poverty	.15*	.02	.16*	.02	.16*	.02
Student–teacher ratio	.13	.09	.11	.09	.11	.09
Student misconduct	.55*	.19	.50*	.20	.49*	.19
School grade	−.11*	.03	−.11*	.03	−.11*	.03
Middle school	.15*	.06	.14*	.06	.14*	06
Prior suspensions	.11*	.01	.11*	.01	.11*	.01
Log likelihood	−2,500.79		−2,497.85		−2,497.77
Chi²	707.45*		713.92*		713.99*

Note. N = 66 counties, 559 schools.

†

p ≤ .10. * p ≤ .05.

Interaction Effects of Incarceration and Concentrated Disadvantage on Suspension Rates

Table 6 presents the results of the negative binomial and Poisson regression analyses of the interactive effects of county incarceration rates and concentrated disadvantage on race-specific suspension rates.⁶ For the White suspension rate, the interaction effect between the incarceration rate and concentrated disadvantage is statistically insignificant. These findings suggest that the level of concentrated disadvantage in the county does not moderate the effect of incarceration rates on the likelihood of suspension for White students. However, for the total suspension rate (b = .30), the Black suspension rate (b = .59), and the Hispanic suspension rate (b = .31), the interaction effect is positive and significant, indicating that the effect of the incarceration rate on the total, Black, and Hispanic suspension rates is amplified in counties with greater concentrated disadvantage. In other words, students of color are more likely to be suspended when attending schools located in counties characterized by both high incarceration and high concentrated disadvantage. The total suspension rate increases by 35% when the school is located in a county where incarceration rates and concentrated disadvantage are high. The likelihood of Black students being suspended increases by 80% when the school is located in a county where incarceration rates and concentrated disadvantage are high. And the likelihood of Hispanic students being suspended increases by 36% when the school is located in a county with high incarceration and high concentrated disadvantage. Given that the significance of the interactive effect on the total suspension rate is likely being driven by the effect of the Black and Hispanic suspension rates, two central themes emerge. First, students of color are vulnerable to being suspended when attending schools in counties marked by high incarceration rates. Second, the likelihood of suspension for students of color escalates when they attend schools that are located in areas of greater concentrated disadvantage.

Table 6.

Multilevel Negative Binomial and Poisson Regressions of Interaction Effects of County Incarceration Rates and Disadvantage on Black, Hispanic, and White Suspension Rates.

	Total suspensions		Black suspensions		Hispanic suspensions		White suspensions
Variable	B	SE	B	SE	B	SE	B	SE
Key county predictors
Incarceration rate	−.22	.20	−.39	.29	−.16	.30	−.14	.20
Concentrated disadvantage	.54*	.13	.35*	.17	.59*	.16	.32*	.13
Interaction
Incarceration rate * Concentrated disadvantage	.30*	.13	.59*	.18	.31^†	.18	.14	.13
Log likelihood	−3,054.22		−2,573.65		−2,164.38		−2,497.17
Chi²	611.04*		448.64*		346.13*		716.56*

Note. All control variables are included in the regression models displayed in Table 6. They are not shown here for viewing ease. N = 66 counties, 559 schools.

†

p ≤ .10. * p ≤ .05.

The interaction effects between incarceration and concentrated disadvantage on the total, Black, and Hispanic suspension rates are further illustrated in Figures 1, 2, and 3. In counties where the incarceration rate and concentrated disadvantage (both measured as 1SD below the mean) are lower, the probability of suspension for Black and Hispanic students (and the total student population) is lower as well. However, in counties where the incarceration rate and concentrated disadvantage (both measured as 1SD above the mean) are higher, the probability of suspension was also higher for Black and Hispanic students (and the total student population).

Figure 1.

Predicted probability of total suspensions at low and high levels of incarceration and concentrated disadvantage.

Figure 2.

Predicted probability of Black suspensions at low and high levels of incarceration and concentrated disadvantage.

Figure 3.

Predicted probability of Hispanic suspensions at low and high levels of incarceration.

Discussion

Over the last four decades, control, surveillance, and punishment have become endemic across American society. This culture of control has spread through numerous societal institutions, including the judicial system and schools. What has remained unclear is whether places characterized by punitiveness in the criminal justice system exhibit comparable punitiveness in schools. This study attempts to fill this void by exploring seven hypotheses: (a) total school suspension rates will be higher in counties with higher incarceration rates; (b) Black and Hispanic suspension rates will be higher in counties with higher incarceration rates, while (c) White suspension rates will be unchanged in counties with higher incarceration rates; (d) total school suspension rates will be higher in counties with greater concentrated disadvantage; (e) Black and Hispanic suspension rates will be higher in counties with greater concentrated disadvantage, while (f) White suspension rates will be unchanged in counties with greater concentrated disadvantage; and (g) the positive relationship between county incarceration rates and school suspension rates for students of color will be moderated by the level of county concentrated disadvantage. The results provide considerable support for these hypotheses.

Regarding Hypothesis 1, the incarceration rate was not significantly associated with the total suspension rate. Regarding Hypothesis 2, Black and Hispanic students were significantly more likely to be suspended when attending schools located in high incarceration rate counties. In addition, consistent with Hypothesis 3, this relationship was not significant for White students, indicating that the expansion of carceral techniques and strategies is especially consequential for students of color. This is consistent with existing research that highlights that people of color are typically perceived as more blameworthy and as a result, experience intensified formal social control (Spohn & Holleran, 2000). As defendants of color often pay an imprisonment penalty in the criminal justice system, students of color pay a suspension penalty, particularly when attending school in punitive places.

Regarding Hypotheses 4, 5, and 6, the analyses demonstrated substantial support for place-based stratification theories. It was hypothesized that the total, Black, and Hispanic suspension rates would be higher in spaces with greater concentrated disadvantage. Contrary to Hypothesis 6, this positive and significant effect of concentrated disadvantage extended to White suspension rates also. Specifically, White suspension rates were higher in counties with greater concentrated disadvantage. Even though Blacks and Hispanics are often disproportionately concentrated in disadvantaged places, the results here suggest that residing in disadvantaged places is detrimental across race and ethnicity (Kantor & Brenzel, 1993).

Finally, the analyses illustrated support for Hypothesis 7. The relationship between county incarceration rates and suspension rates was significantly conditioned by concentrated disadvantage for the total student population and for Black and Hispanic students. Given that the interaction effect was not significant for the White suspension rate, the implication is that the suspension rates of students of color are driving the significant interactive effects for the total suspension rate. In other words, in comparison to their White counterparts, students of color who attend school in places characterized by high incarceration and greater concentrated disadvantage are significantly more vulnerable to being suspended.

There are several potential explanations for these key findings. First, because people of color are more likely to be concentrated in disadvantaged places, the positive relationship between incarceration rates and suspension rates may be amplified for students of color in those areas. The vulnerability of Black and Hispanic students to suspension may be exacerbated as the levels of punitiveness and disadvantage accumulate in a place. In addition, the imposition of social control has been found to be more prevalent and more severe in disadvantaged contexts, so it follows logically that this emphasis on social control is likely to be intensified in schools that are located in disadvantaged spaces.

Existing research has also advanced a second explanation for the intensification of school punishment for students of color attending school in places characterized by high incarceration rates and high concentrated disadvantage (Oliver & Shapiro, 2006). Specifically, the proliferation of exclusionary school punishment in disadvantaged places might be due to the lack of legal recourses and social capital available to the people of color who are typically concentrated there. Students of color and their families might not have the financial or social status to support serious legal challenges to what may be perceived as the disproportionate application of severe school discipline (Oliver & Shapiro, 2006). This fact may further discourage Black and Hispanic students and their families residing in these disadvantaged spaces from contesting school punishment decisions.

In sum, this study contributes to the existing literature by elucidating several ways in which the effects of place are consequential for school punishment outcomes. First, this study illustrates that in places where the criminal justice system exhibits greater punitiveness, similar punitive approaches may manifest in other local institutions, especially schools. This implies that in certain locales, a culture of control, bolstered by enhanced security and surveillance, has the potential to foster harsh punishment in the criminal justice system and a comparable approach to student discipline in schools. More specifically, students who attend school in areas where punitive approaches have been embraced in the criminal justice system are more likely to experience severe school discipline.

Second, the findings suggest that the relationship between concentrated disadvantage and school punishment may be invariant across race and ethnicity. Specifically, schools located in disadvantaged places exhibit harsh school discipline for all students, regardless of race or ethnicity. Although concentrated disadvantage is most often associated with negative outcomes for people of color in particular, this study indicates that, for school punishment, it may be detrimental for all students who attend school in impoverished, disorganized spaces.

Third, this study also highlights the contextual conditions under which the disparate punishment outcomes that students of color often experience in schools may arise. In particular, this study demonstrates that Black and Hispanic students may be vulnerable to an accumulation of disadvantage that increases their likelihood of experiencing severe punishment when they attend school in a punitive environment that is also characterized by greater concentrated disadvantage. This is especially noteworthy given the fact that people of color are more likely to populate areas where there is both a pervasive culture of control and a concentration of poverty and disorder. Just as these factors are consequential for the policing of people of color by the criminal justice system, this study suggests that students of color are similarly susceptible to severe discipline in schools that are located in such areas.

Future Research

Beyond the contributions of this study, avenues for future research remain. First, this study considers a novel approach to the relationship between punishment in the school context and the criminal justice system. However, a more comprehensive examination of the relationship between punishment outcomes in both the school and the criminal justice system is warranted. Existing research suggests that this relationship may operate in both directions. Punishment outcomes in schools may influence criminal justice outcomes (termed the “school-to-prison pipeline”) and criminal justice outcomes may be associated with school punishment outcomes. Because of the data used here, this study is unable to test for the presence of two-way causation. Future research should employ methodological techniques that allow for consideration of a two-way relationship and thus, a broader understanding of the link between school punishment and judicial punishment.

Second, this study focuses primarily on the relationship between incarceration rates and school suspension rates. Future research should consider how other criminal justice-related factors—for example, allocation of funding, law enforcement presence, or sentence length—may be associated with other school punishment outcomes—for instance, expulsions, detentions, and office referrals. Exploration of these alternative measures of criminal justice punitiveness and school disciplinary measures may provide a clearer picture of how school punishment outcomes may be associated with punitiveness in the judicial system.

Third, additional scholarship should explore whether these results are consistent across various student populations and geographic locales. More specifically, future research should consider how the effects of the culture of control in the criminal justice system and the school context impact outcomes for male and female students, elementary students, students with disabilities, and across other states and nationally. Such approaches could provide a more comprehensive understanding of the culture of control and its consequences for particular students and in additional locales.

Conclusion

As the culture of control expands across places and institutions, it is important to consider its costs and implications for the populations most adversely affected. This study highlights the symbiotic relationship between punitiveness and disadvantage and its salience for Black and Hispanic students in the school setting. Just as increased social control in the criminal justice system is likely to affect defendants of color, increased surveillance, control, and punishment are likely to impact students of color disproportionately. Furthermore, the far-reaching consequences of severe school punishment—for instance, academic failure, arrest, gang involvement, and other negative life outcomes—underscore the need for continued investigations around what factors are associated with the continued creep of the culture of control into American schools (Peguero et al., 2021; Pesta, 2018; Widdowson et al., 2021).

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Notes

Author Biography

Cresean Hughes is an assistant professor in the Department of Sociology and Criminal Justice at the University of Delaware. His research interests include race, ethnicity, punishment, and social control, with a particular focus on disparities in school discipline and other youth justice outcomes. His work has been published in the Journal of Research in Crime and Delinquency, Race and Justice, Social Science Research, and Criminal Justice and Behavior.

References

Alexander

(2010). The New Jim Crow: Mass incarceration in the age of colorblindness. New Press.

Beck

Willis

(1995). Crime and security: Managing the risk to safe shopping. Perpetuity Press.

Bell

(2021). Suspended. Johns Hopkins University Press.

Blakely

E. J.

Snyder

M. G.

(1997). Fortress America: Gated communities in the United States. Brookings Institution Press.

Blomberg

T. G.

Lucken

(2010). American penology: A history of control. Transaction Publishers.

Bonczar

(2003). The Prevalence of Imprisonment in the U.S. Population, 1974-2001. U.S. Department of Justice. Bureau of Justice Statistics.

Brayne

(2014). Surveillance and system avoidance: Criminal justice contact and institutional attachment. American Sociological Review, 79, 367–391.

Fabelo

Thompson

M. D.

Plotkin

Carmichael

Marchbanks

M. P.

Booth

E. A.

(2011). Breaking schools’ rules: A statewide study of how school discipline relates to students’ success and juvenile justice involvement. Council of State Governments Justice Center.

Fisher

B. W.

Higgins

E. M.

Homer

E. M.

(2019). School crime and punishment and the implementation of security cameras: Findings from a national longitudinal study. Justice Quarterly, 38, 22–46.

10.

Foucault

(1977). Discipline and punish: The Birth of the prison. Vintage.

11.

Gaines

(2021, September 12). Thousands were released from prison because of COVID: Will they have to return? NBC News. https://www.nbcnews.com/news/nbcblk/thousands-released-prison-covid-will-return-rcna1977

12.

Garland

(2001). The culture of control. University of Chicago Press.

13.

Gottfredson

D. C.

(2001). Schools and delinquency. Cambridge University Press.

14.

Gottfredson

G. D.

Gottfredson

D. C.

(2001). What schools do to prevent problem behavior and promote safe environments. Journal of Educational and Psychological Consultation, 12, 313–344.

15.

Gottfredson

G. D.

Gottfredson

D. C.

Payne

A. A.

Gottfredson

N. C.

(2005). School climate predictors of school disorder: Results from a national study of delinquency prevention in schools. Journal of Research in Crime and Delinquency, 42, 412–444.

16.

Gottschalk

(2015). Caught: The prison state and the lockdown of American politics. Princeton University Press.

17.

Government Accountability Office. (2018). K-12 education: Discipline disparities for Black students, boys, and students with disabilities.

18.

Grabosky

P. N.

(1992). Law enforcement and the citizen: Non-governmental participants in crime prevention and control. Policing and Society, 2, 249–271.

19.

Gregory

Weinstein

R. S.

(2008). The discipline gap and African Americans: Defiance or cooperation in the high school classroom. Journal of School Psychology, 46, 455–475.

20.

Hagan

Foster

(2012). Intergenerational educational effects of mass imprisonment in America. Sociology of Education, 85, 259–286.

21.

Haggerty

K. D.

Ericson

R. V.

(2006). The new politics of surveillance and visibility. University of Toronto Press.

22.

Haskins

A. R.

Jacobsen

W. C.

(2017). Schools as surveilling institutions? Paternal incarceration, system avoidance, and parental involvement in schooling. American Sociological Review, 82, 657–684.

23.

Hirschfield

P. J.

(2008). Preparing for prison? The criminalization of school discipline in the USA. Theoretical Criminology, 12, 79–101.

24.

Hughes

Warren

P. Y.

Stewart

E. A.

Tomaskovic-Devey

Mears

D. P.

(2017). Racial threat, intergroup contact, and school punishment. Journal of Research in Crime and Delinquency, 54, 583–616.

25.

Jacobsen

W. C.

(2019). The intergenerational stability of punishment: Parental incarceration and suspension or expulsion in elementary school. Journal of Research in Crime and Delinquency, 56, 651–693.

26.

Jacobsen

W. C.

Pace

G. T.

Ramirez

N. G.

(2019). Punishment and inequality at an early age: Exclusionary discipline in elementary school. Social Forces, 73, 973–998.

27.

Johnson

L. T.

Kane

R. J.

(2018). Deserts of disadvantage: The diffuse effects of structural disadvantage on violence in urban communities. Crime & Delinquency, 64, 143–165.

28.

Jordan

K. L.

Freiburger

T. L.

(2010). Examining the impact of race and ethnicity on the sentencing of juveniles in adult court. Criminal Justice Policy Review, 21, 185–201.

29.

Kane

R. J.

(2002). The social ecology of police misconduct. Criminology, 40, 867–896.

30.

Kang-Brown

Montagnet

Heiss

(2021). People in jail and prison in 2020. Vera Institute of Justice.

31.

Kantor

Brenzel

(1993). Urban education and the “truly disadvantaged”: The historical roots of the contemporary crisis, 1945–1990. Teachers College Record, 94, 279–314.

32.

Kirk

D. S.

(2009). Unraveling the contextual effects on student suspension and juvenile arrest: The independent and interdependent influences of school, neighborhood, and family social controls. Criminology, 47, 479–520.

33.

Krivo

L. J.

Peterson

R. D.

(2000). The structural context of homicide: Accounting for racial differences in process. American Sociological Review, 65, 547–559.

34.

Kubrin

C. E.

Weitzer

(2008). Retaliatory homicide: Concentrated disadvantage and neighborhood culture. Social Problems, 50, 157–180.

35.

Kupchik

(2010). Homeroom security: School discipline in an age of fear. New York University Press.

36.

Laub

J. H.

Lauritsen

J. L.

(1998). The interdependence of school violence with neighborhood and family conditions. In Elliott

D.S.

Hamburg

B. A.

Williams

K. R.

(Eds.), Violence in American schools: A new perspective (pp. 127–155). Cambridge University Press.

37.

Lehmann

P. S.

Azimi

A. M.

Fortney

Alaniz

(2021). Racial and ethnic disparities in school discipline: The interactive effects of gender and parental educational attainment. Crime and Delinquency. Advance online publication. https://doi.org/10.1177/00111287211029879

38.

Levy

B. L.

(2019). Heterogeneous impacts of concentrated poverty during adolescence on college outcomes. Social Forces, 98, 147–182.

39.

Levy

B. L.

Owens

Sampson

R. J.

(2019). The varying effects of neighborhood disadvantage on college graduation: Moderating and mediating mechanisms. Sociology of Education, 92, 269–292.

40.

Long

J. S.

(1997). Regression models for categorical and limited dependent variables. SAGE.

41.

Lopez-Aguado

(2018). Stick together and come back home: Racial sorting and the spillover carceral identity. University of California Press.

42.

Losen

D. J.

Martinez

(2020). Lost opportunities: How disparate school discipline continues to drive differences in the opportunity to learn. The Civil Rights Project, The University of California.

43.

Losen

D. J.

Skiba

R. J.

(2010). Suspended education: Urban middle schools in crisis. The Civil Rights Project, The University of California, Los Angeles.

44.

Marchbanks

M.P.

III Peguero

A. A.

Varela

K. S.

Blake

J. J.

Eason

J. M

. (2018). School strictness and disproportionate minority contact: Investigating racial and ethnic disparities with the “School-to-Prison Pipeline.” Youth Violence and Juvenile Justice, 16, 241–259.

45.

Meiners

(2013). Schooling the carceral state: Challenging the school-to-prison pipeline. In Scott

(Ed.), Why prison? (pp. 261–277). Cambridge University Press.

46.

Morris

M. W.

(2016). Pushout: The criminalization of Black girls in schools. The New Press.

47.

Mowen

Brent

(2016). School discipline as a turning point: The cumulative effect of suspension on arrest. Journal of Research in Crime and Delinquency, 53, 628–653.

48.

Musu-Gillette

Zhang

Wang

Zhang

Kemp

Diliberti

Oudekerk

(2018). Indicators of school crime and safety: 2017. U.S. Department of Education. National Center for Education Statistics.

49.

Oliver

M. L.

Shapiro

(2006). Black wealth/White wealth: A new perspective on racial inequality. Routledge.

50.

Owens

(2010). Neighborhoods and schools as competing and reinforcing contexts for educational attainment. Sociology of Education, 83, 287–311.

51.

Owens

McLanahan

S. S.

(2020). Unpacking the drivers of racial disparities in school suspension and expulsion. Social Forces, 98, 1548–1577.

52.

Parker

K. F.

Stults

B. J.

Rice

S. K.

(2005). Racial threat, concentrated disadvantage and social control: Considering the macro-level sources of variation in arrests. Criminology, 43, 1111–1134.

53.

Payne

A. A.

Welch

(2010). Modeling the effects of racial threat on punitive and restorative school discipline practices. Criminology, 48, 1019–1062.

54.

Peguero

A. A.

Varela

K. S.

Marchbanks

M. P.

III Blake

Eason

J. M.

(2021). School punishment and education: Racial/ethnic disparities with grade retention and the role of urbanicity. Urban Education, 56, 228–260.

55.

Pesta

(2018). Labeling and the differential impact of school discipline on negative life outcomes: Assessing ethno-racial variation in the school-to-prison pipeline. Crime and Delinquency, 64, 1489–1512.

56.

Pusch

(2021). Macro-level criminological perspectives and school delinquency: A meta-analysis. Youth and Society. Advance online publication. https://doi.org/10.1177/0044118X211022387

57.

Ramey

D. M.

(2015). The social structure of criminalized and medicalized school discipline. Sociology of Education, 88, 181–201.

58.

Raudenbush

S. W.

Bryk

A. S.

(2002). Hierarchical linear models: Applications and data analysis methods. SAGE.

59.

Rios

(2011). Punished: Policing the lives of Black and Latino boys. New York University Press.

60.

Rocque

Paternoster

(2011). Understanding the antecedents of the school-to-jail link: The relationship between race and school discipline. The Journal of Criminal Law and Criminology, 101, 633–665.

61.

Rodriguez

(2013). Concentrated disadvantage and the incarceration of youth. Journal of Research in Crime and Delinquency, 50189–50215.

62.

Rose

D. R.

Clear

T. R.

(1998). Incarceration, social capital, and crime: Implications for social disorganization theory. Criminology, 36, 441–480.

63.

Sampson

R. J.

(1987). Urban Black violence: The effect of male joblessness and family disruption. American Journal of Sociology, 93, 348–382.

64.

Sampson

R. J.

Bartusch

D. J.

(1998). Legal cynicism and (subcultural) tolerance of deviance: The neighborhood context of racial difference. Law & Society Review, 32, 777–804.

65.

Sampson

R. J.

Groves

W. B.

(1989). Community structure and crime: Testing social disorganization theory. American Journal of Sociology, 94, 774–802.

66.

Sampson

R. J.

Loeffler

(2010). Punishment’s place: The local concentration of mass incarceration. Daedalus, 139, 20–31.

67.

Sampson

R. J.

Raudenbush

S. W.

Earls

(1997). Neighborhoods and violent crime: A multilevel study of collective efficacy. Science, 277, 918–924.

68.

Shaw

C. R.

McKay

(1942). Juvenile delinquency and urban areas. University of Chicago Press.

69.

Shedd

(2015). Unequal city: Race, schools, and perceptions of injustice. Russell Sage Foundation.

70.

Siennick

S. E.

Stewart

E. A.

Staff

(2014). Explaining the association between incarceration and divorce. Criminology, 52, 371–398.

71.

Simon

(2007). Governing through crime. Oxford University Press.

72.

Spohn

Holleran

(2000). The imprisonment penalty paid by young, unemployed Black and Hispanic male offenders. Criminology, 38, 281–306.

73.

Steffensmeier

Demuth

(2000). Ethnicity and sentencing outcomes in US federal courts: Who is punished more harshly?". American Sociological Review, 65, 705–729.

74.

Thompson

F. T.

(2002). Student achievement, selected environment characteristics, and neighborhood type. The Urban Review, 34, 277–292.

75.

Tonry

(1999). Why are US incarceration rates so high? Crime and Delinquency, 45, 419–437.

76.

Travis

Western

Redburn

F. S.

(2014). The growth of incarceration in the United States: Exploring causes and consequences. National Academies Press.

77.

Turney

Haskins

A. R.

(2014). Falling behind? Children’s early grade retention after parental incarceration. Sociology of Education, 87, 241–258.

78.

U.S. Department of Education. (2018). U.S. civil rights data collection: First look report.

79.

Wacquant

(2001). Deadly symbiosis: When ghetto and prison meet and mesh. Punishment & Society, 3, 95–134.

80.

Wegmann

K. M.

Smith

(2019). Examining racial/ethnic disparities in school discipline in the context of student-reported behavior infractions. Children and Youth Services Review, 103, 18–27.

81.

Welch

Payne

A. A.

(2010). Racial threat and punitive school discipline. Social Problems, 5, 25–48.

82.

Western

(2006). Punishment and inequality in America. Russell Sage Foundation.

83.

Widdowson

A. O.

Garduno

L. S.

Fisher

B. W.

(2021). The school-to-gang pipeline: Examing the impact of school suspension on joining a gang for the first time. Crime and Delinquency, 67, 997–1021.

84.

Wilson

W. J.

(1987). The truly disadvantaged: The inner city, the underclass, and public policy. University of Chicago.

85.

Wodtke

G. T.

Harding

D. J.

Elwert

(2011). Neighborhood effects in temporal perspective: The impact of long-term exposure to concentrated disadvantage on high school graduation. American Sociological Review, 76, 713–736.

86.

Wolf

K. C.

Kupchik

(2017). School suspensions and adverse experiences in adulthood. Justice Quarterly, 34, 407–430.