Latino/a Student Threat and School Disciplinary Policies and Practices

Abstract

Using a nationally representative sample of approximately 3,500 public schools, this study builds on and extends our knowledge of how ‘‘minority threat’’ manifests within schools. We test whether various disciplinary policies and practices are mobilized in accordance with Latino/a student composition, presumably the result of a group response to perceptions that white racial dominance is jeopardized. We gauge how schools’ Latino/a student populations are associated with the availability and use of several specific types of discipline. We further explore possible moderating influences of school crime and economic disadvantage on punishment. We find that schools with larger percentages of Latino/a students are more likely to favor certain punitive responses and less likely to favor certain mild responses, as predicted by minority threat. The percentage of Latino/a students is also related to greater use of certain disciplinary responses in schools with less crime.

Keywords

Latino/a student composition school discipline minority threat school-to-prison pipeline ethnic disparities

Introduction

Minority threat theory (also referred to as power threat, social threat, and racial threat) suggests that more punitive social control policies and practices will be implemented in places where there are larger percentages of blacks and Latinos/as to manage what is perceived to be a growing threat to the racial and economic dominance of whites (Blalock 1967; Blumer 1958; Liska 1992) as well as to their safety (Crawford, Chiricos, and Kleck 1998). While the proportion of a place that is inhabited by racial or ethnic minorities—the most common proxy for minority threat in prior research—has been associated with harsher criminal justice policies and practices (Crawford et al. 1998; Keen and Jacobs 2009; Stewart et al. 2015), the effects of minority threat are also apparently experienced in schools (Welch and Payne 2010). It is now well established that schools’ reactions to student misbehavior and delinquency have become harsher despite sustained reductions in school-based violations, a phenomenon similar to practices that have been observed in the U.S. criminal justice system, which has imposed stiff penalties despite declining crime rates (Robers et al. 2015). As with criminal justice patterns, school disciplinary policies and practices have disproportionately targeted racial minorities (Lewis and Diamond 2015), making it plausible that students of color are at greater risk for experiencing the school-to-prison pipeline since so many students who are punished in school are subsequently punished in the justice system (Fabelo et al. 2011; Kupchik 2016; Simmons 2017; Wallace et al. 2008).

Minority threat theory suggests that the white majority perceives the growing population of students of color as an ethnic and criminal threat (Shedd 2015). Efforts to maintain the status quo social stratification that has historically privileged whites result in a more authoritarian orientation in schools, increased criminalization of minority youth, harsher disciplinary consequences, and less frequent use of restorative methods (Lewis and Diamond 2015; Rios 2017; Shedd 2015). Research shows that the composition of black students in schools is indeed strongly associated with harsh discipline (Peguero and Shekarkhar 2011; Rocque and Paternoster 2011; Welch and Payne 2010, 2012). Further, there appears to be some association between overall nonwhite student composition and disciplinary practices (Irwin, Davidson, and Hall-Sanchez 2013; Kupchik and Ward 2014; Peguero, Popp, and Shekarkhar 2015). Recently, a limited number of studies has examined the possibility of a similar relationship specifically for Latino/a student populations. Considering that Latinos/as are now the largest minority in the United States—with the Latino/a youth population even larger (Colby and Ortman 2015)—and given the ethnic disparities evident in criminal justice statistics (Bureau of Justice Statistics 2016), it is surprising that so little is known about the association between schools’ ethnic compositions and their disciplinary practices. Understanding more about this relationship is critical in light of the many negative consequences of punitive discipline (see Losen 2011; Losen and Skiba 2010; Rios 2017).

The present study addresses this knowledge gap in several ways. We use a minority threat framework to assess whether and to what extent schools’ Latino/a student composition is associated with social control in the form of disciplinary approaches. Gaining a foundational knowledge of how various disciplinary policies and practices are implemented would help us better understand youth experiences of social control (Ipsa-Landa 2017; Kim, Losen, and Hewitt 2010). We thus use nationally representative school data to separately examine both the availability and use of a wide range of specific types of school discipline, including mild forms, such as privilege loss and community service, as well as more punitive exclusionary forms, like suspension and expulsion. In accordance with minority threat theory, we expect that schools with proportionally more Latino/a students will implement and use more punitive and fewer mild policies and practices. We test for these effects independent of the influence of various student, school, and contextual characteristics that have not been included together in prior research. We further expand existing knowledge about how Latino/a student composition influences disciplinary policies and practices by explicating the potential moderating influences of school crime and economic disadvantage on this relationship. According to minority threat theory, ethnic composition should have a greater influence on social control in contexts that are not already more punitive, that is, contexts where greater discretionary control by the majority group is possible (Keen and Jacobs 2009).

Racial and Ethnic Disparities in Student Punishment

U.S. schools have been so supportive of harsh approaches to identifying and preventing student misbehavior that their policies have been likened to those in the criminal justice system (Simon 2007). Many explanations for the crime control approach to school discipline have focused on the characteristics of students who are most likely to be punished by harsh school policies (Shedd 2015). As with criminal justice punitiveness, one of the strongest and most consistent predictors of discipline is student race and ethnicity: Black and Latino/a students experience more frequent and intense school punishments for the same or lesser offenses than their white peers (Anyon et al. 2014; Beck and Muschkin 2012; Losen et al. 2015; Peguero and Shekarkhar 2011; Rocque 2010; Skiba et al. 2015; Wallace et al. 2008). Several notable studies chronicle the many ways in which minority students are subject to greater scrutiny, surveillance, and social control (Lewis and Diamond 2015; Shedd 2015), mirroring trends seen in the criminal justice system (Welch et al. 2011). Research clearly demonstrates that these racial and ethnic disparities in discipline are not justified by differences in misbehavior or delinquency (Gregory and Weinstein 2008; Skiba et al. 2015; Skiba, Michael, and Nardo 2002). Furthermore, minority students experience harsher school punishment regardless of other influences, such as economic disadvantage (Gregory and Weinstein 2008; Skiba et al. 2002).

Specifically, black and Latino/a students are much more likely than white students to receive office referrals for discipline (Anyon et al. 2014; Rocque and Paternoster 2011) and be referred to law enforcement (Anyon et al. 2014). Compared to white students, minority students are also suspended more often for the same or lesser offenses (U.S. Department of Education 2013). Expulsion, generally the most severe school penalty, is also more frequently assigned for violations by both black students (Gregory and Weinstein 2008; U.S. Department of Education 2013) and Latino/a students (Peguero and Shekarkhar 2011; U.S. Department of Education 2013). Minority students experience corporal punishment more frequently (McClure and May 2014; Wallace et al. 2008) and are less likely to receive the restorative sanction of community service (Skiba et al. 2002; Skiba and Peterson 2000) compared to their white peers. This considerable disparity in punishment for minority students is particularly concerning given the vast overrepresentation of blacks and Latinos/as in criminal justice statistics and the link between involvement in the criminal justice system and earlier schooling experiences (Rios 2017).

Minority Threat Theoretical Framework

Beyond the disparate impact on individual students, there are also school-level differences associated with student race and ethnicity on school culture and approaches to student behavior (Carter 2012). Minority threat, most frequently operationalized in recent research as the racial or ethnic composition of schools, may contribute to the intensification of punitive school discipline. Originating from the “power threat” theoretical work of Blumer (1958) and Blalock (1967), the minority threat thesis posits that social control mechanisms within society will expand and intensify as the proportion of racial and ethnic minorities in an area increases in relation to whites. The intensification of these mechanisms is due to the majority group’s perception that their dominant economic and political social position is jeopardized. Tests for the influence of this threat, also termed social threat (Liska 1992) and racial threat (Crawford et al. 1998), include black crime as another salient source of threat to whites. Similarly, there is evidence of “ethnic threat” and “Hispanic” or “Latino/a threat” on social control (Stewart et al. 2015). Because crime and punishment are so closely linked with race and ethnicity, minority threat theory commonly uses a criminal justice framework; studies have examined the racial and ethnic composition of neighborhoods, cities, counties, states, and even countries (Keen and Jacobs 2009). To this end, minority composition of place has been related to an array of criminal justice outcomes, including higher rates of arrest (Stolzenberg, D’Alessio, and Eitle 2004), expansion of law enforcement capabilities (Kent and Jacobs 2004), police brutality (Holmes 2000), and higher incarceration rates (Keen and Jacobs 2009). Minority threat theory hypothesizes that these social controls will expand along with the relative size of the minority population until a tipping point is reached, at which point whites will no longer have the power to subordinate persons of color.

Considering the extent to which racial and ethnic composition is related to harsh criminal justice measures, it is not surprising that the minority threat perspective has also been applied to schools to explain the growing punitiveness of disciplinary policies and practices that impose social control on a disproportionate number of minority students. The implicit connection between race and crime potentially made by teachers and administrators at the local level, as well as government policymakers at many levels, would result in a greater number of harsh punishments and a smaller number of mild alternatives being sanctioned in schools with more minority students in order to control what may be perceived by the white majority as a growing threat to safety and authority (Bekkerman and Gilpin 2014; Irwin et al. 2013; Rocque and Paternoster 2011; Welch and Payne 2010). Unlike minority threat in a criminal justice context, we do not expect a curvilinear relationship based on the relative size of the minority student population because social control is not imposed by other students but rather by adults with the authority to do so. A growing body of research indicates that racial threat may be operating in schools, with high black student body composition associated with more punitive and less mild practices (Anyon et al. 2014; Bekkerman and Gilpin 2014; Kupchik 2009; Mowen and Parker 2014; Payne and Welch 2010; Peguero et al. 2015; Peguero and Shekarkhar 2011; Ramey 2015; Roch and Edwards 2017; Rocque and Paternoster 2011; Servoss and Finn 2014; Welch and Payne 2010, 2012). Schools with a greater percentage of black students more often implement punitive, exclusionary, and zero tolerance disciplinary responses for student misbehavior despite the effects of other individual, classroom, and school influences (Anyon et al. 2014; Irwin et al. 2013; Kupchik and Ward 2014; Mowen and Parker 2014; Ramey 2015; Rocque and Paternoster 2011; Servoss and Finn 2014; Welch and Payne 2010). Moreover, schools with proportionally more black students less often implement mild or restorative forms of discipline, such as referrals to a school counselor, parent conferences, oral reprimands, community service, and restitution (Payne and Welch 2015).

Research testing the minority threat theory demonstrates unequivocally that some school-level race-specific factor is associated with intense discipline regardless of schools’ levels of delinquency and other conditions. Studies have also examined the broader effect of unspecified nonwhite minority student composition on discipline that do not distinguish among specific racial and ethnic minority categories (Irwin et al. 2013; Kupchik and Ward 2014; Peguero and Shekarkhar 2011). This body of research generally indicates that school minority composition is frequently associated with harsher school discipline practices, although it is not as easily predicted and appears less consistent than the relationship between discipline and black school composition. How and to what extent minority threat is evident in Latino/a school contexts is less clear.¹

Latino/a Composition and Punitiveness in Schools

A hypothesized Latino/a threat explanation for disciplinary punitiveness in schools relies on the same tenets as the school-based racial threat thesis. Specifically, schools with proportionally more Latino/a students will more often endorse and use harsher discipline (e.g., suspension and expulsion) and less often endorse and use milder responses (e.g., community service, privilege loss, detention, and probation) because authorities make an implicit association between Latinos/as and delinquency (for more on how forms of discipline are classified as punitive or mild, see Kim et al. 2010; Skiba, Mediratta, and Rausch 2016; Skiba and Rausch 2006; Wallace et al. 2008). That is, to control a student population that is perceived as menacing and a challenge to the power structure that has traditionally subordinated ethnic minorities relative to whites, schools with relatively more Latino/a students will have more intense discipline and less opportunity for students to benefit from mild discipline (Rios 2017). Given that Latinos/as are overrepresented in criminal justice statistics (Bureau of Justice Statistics 2016) and the Trump administration’s renewed focus on Latinos/as in the United States, an exploration of Latino/a threat effects within schools seems timely and necessary.

While we know that the proportion of black and other minority students in schools is associated with more intense and less mild disciplinary approaches, only a few studies have assessed the possibility of a similar Latino/a threat effect in schools. The handful of studies that have examined an effect of Latino/a composition on school discipline (Anyon et al. 2014; Bekkerman and Gilpin 2016; Mowen and Parker 2014; Peguero et al. 2015; Ramey 2015; Servoss and Finn 2014) have produced inconsistent results that appear to depend on the specific type of disciplinary policy or practice analyzed, measurement of the discipline responses, samples examined, and school and community characteristics controlled. For instance, although most of these studies found no relationship between the proportion of Latino/a students and suspensions (Bekkerman and Gilpin 2016; Mowen and Parker 2014; Peguero et al. 2015; Servoss and Finn 2014), one study of Denver public schools found that a greater proportion of Latino/a students was related to more suspensions (Anyon et al. 2014), and another study using national data found it was related to fewer (Ramey 2015). Most of these studies did not find a relationship between Latino/a student composition and removal from school, either through expulsion or disciplinary transfer (Anyon et al. 2014; Bekkerman and Gilpin 2016; Peguero et al. 2015), but one study found that a greater proportion of Latino/a students was related to a decrease in removals (Ramey 2015). Furthermore, Latino/a student composition appears to increase short-term—but not long-term—discipline (Bekkerman and Gilpin 2016). Even results regarding the relationship between Latino/a student composition and milder disciplinary responses are mixed, with one study finding more mild disciplinary measures in schools with a greater percentage of Latino/a students (Anyon et al. 2014) and another finding no relationship (Peguero et al. 2015).

Considered collectively, these inconsistent findings seem to suggest that the minority threat thesis may not be applicable in the context of Latino/a school composition. Yet, there is likely more to learn. As Servoss and Finn (2014:75) note, “The dynamics that produce these varied patterns in high-Hispanic schools remain unclear.” Most of these previous studies have examined the use of only certain disciplinary practices, and none have examined the extent to which Latino/a student composition is related to both the implementation of a range of disciplinary policies and their use in practice. These theoretically and practically distinct outcomes are important to distinguish because the minority threat theory suggests a potentially stronger effect of ethnic composition on the use of discipline than on disciplinary policy since policies are often made with more spatial distance (e.g., at the state or district level) and would therefore be less susceptible to the ethnic dynamics of particular schools. Given the many negative consequences of punitive punishment, the known benefits of restorative sanctions, the widening ethnic disparities observed in schools and beyond, and the fact that the Latino/a population of youth is now the largest in the United States (Colby and Ortman 2015), more comprehensive tests for how the Latino/a composition of schools contributes to school discipline and the presumed school-to-prison pipeline are needed.

The Current Study

This study applies the minority threat perspective in the context of school punitiveness, assessing the impact that Latino/a student composition has on school discipline policies and practices, and it extends previous research in several ways. First, we use a large, nationally representative sample of schools to test for effects on a greater range of specific disciplinary responses. We go beyond previous studies that have used regional samples or measured only a single type (e.g., suspension) or narrow range (e.g., exclusionary) of disciplinary practices. Furthermore, recognizing that some variability in outcomes from related research may be partially attributable to how discipline in schools has been measured, this study is the first to separately explore the extent to which a variety of disciplinary actions are not only available to school administrators as policy options but also to what degree those policy options are indeed used in practice. These two metrics—availability and use—are not equivalent: Schools may have policies that permit certain punishments that are rarely or never used and because policies made at the state or district level may not be as sensitive to individual school demographics. Finally, in accordance with minority threat theory and previous research finding a differential relationship between black student composition and punitive versus mild disciplinary responses, we predict that Latino/a student composition will also have a differential relationship to harsher versus milder discipline. Thus, we test the following hypotheses:

Hypothesis 1: Schools with a greater percentage of Latino/a students are (a) less likely to adopt mild probation, detention, privilege loss, and community service and (b) more likely to adopt harsh in-school suspension, out-of-school suspension, and expulsion as disciplinary policies.

Hypothesis 2: Schools with a greater percentage of Latino/a students are (a) less likely to use mild probation, detention, privilege loss, and community service and (b) more likely to use harsh in-school suspension, out-of-school suspension, and expulsion as disciplinary practices.

In addition, this study goes beyond previous research on Latino/a threat in schools by examining possible contextual relationships, determining whether schools’ Latino/a composition of students is more influential in circumstances in which punitive discipline is not already more likely and school personnel and policymakers are permitted more discretion. Because prior research has found such strong and consistent results regarding the predictive qualities of both economic disadvantage and levels of school crime on school discipline, we examine the possibility that any effects of Latino/a threat will be stronger under conditions in which discipline is not already likely to be harsh (Welch and Payne 2010). Hence, this study tests one final hypothesis:

Hypothesis 3: Latino/a student composition is more influential on the (a) availability and (b) use of probation, detention, privilege loss, community service, in-school suspension, out-of-school suspension, and expulsion in schools with (1) less school crime and (2) a smaller percentage of economically disadvantaged students.

Methods

Data

Data for this study come from the School Survey on Crime and Safety (SSOCS). The U.S. Department of Education’s National Center for Education Statistics uses the Common Core of Data as a sampling frame to create a nationally representative sample of approximately 3,500 public schools, stratified by level, locale, and size. The surveys were administered to different schools starting in the spring of 2000 and completed by principals to provide school-level data on school crime and safety.² This study uses the fully imputed data set from the 2007–2008 survey, which has a weighted response rate of 77.2 percent; the final sample includes data from 2,560 schools. Importantly, that year’s SSOCS data, unlike that of other years and other data sets, allow for the specific measurement of Latino/a student composition rather than combining Latino/a students with other racial and ethnic minorities.

Measures

Dependent variables

We examine both the availability (policies) and the use (practices) of disciplinary responses using the questions “Does your school allow for the use” of a number of actions and if so, “Was the action used this school year?” The binary responses include four relatively mild school sanctions (probation, detention, privilege loss, and community service) and three relatively punitive sanctions (in-school suspension, out-of-school suspension, and expulsion) for a total of 14 dichotomous (yes = 1, no = 0) dependent variables. Descriptive statistics for these measures and the other variables in this study are provided in Table 1.

Table 1.

Descriptive Statistics for Study Measures.

Measure	Mean	SD	Range	N
Discipline outcomes
Availability of
Probation	.669	.471	0–1	2,560
Detention	.821	.384	0–1	2,560
Privilege loss	.951	.216	0–1	2,560
Community service	.379	.485	0–1	2,560
In-school suspension	.862	.345	0–1	2,560
Suspension	.526	.499	0–1	2,560
Expulsion	.598	.490	0–1	2,560
Use of
Probation	.877	.329	0–1	1,713
Detention	.961	.193	0–1	2,101
Privilege loss	.975	.158	0–1	2,435
Community service	.836	.371	0–1	969
In-school suspension	.444	.498	0–1	2,000
Out-of-school suspension	.346	.476	0–1	1,339
Expulsion	.725	.447	0–1	1,839
Percent Latino/a students	15.870	24.321	0–100	2,560
Control variables
School crime	41.114	52.470	0–577	2,560
Community crime (high)	.249	.433	0–1	2,560
Percent free/reduced lunch	43.275	27.478	0–100	2,560
Percent black students	15.580	24.421	0–100	2,560
Percent Asian students	3.81	.081	0–96	2,560
Percent male students	49.286	10.057	0–100	2,560
School size	865.184	621.406	1–4,872	2,560
Grade level
High school (reference)	.366	.482	0–1	2,560
Middle school	.350	.477	0–1	2,560
Primary school	.241	.428	0–1	2,560
Combined school	.043	.202	0–1	2,560
Urbanicity
Urban (reference)	.265	.442	0–1	2,560
Urban fringe	.370	.483	0–1	2,560
Town	.104	.305	0–1	2,560
Rural	.256	.437	0–1	2,560
State percent Latino/a	11.986	11.473	.700–42.100	2,560
South	.371	.483	0–1	2,560

Primary independent variable

Latino/a threat is operationalized by the percentage of Latino/a students in each school (percent Latino/a students), taken from the Common Core of Data. This variable comports with the previously reviewed macro-level minority threat research that uses objective measures of minority composition of place, including schools. On average, Latino/a students comprise 15.87 percent of the sample’s schools.

Control variables

Because previous research suggests there are various influences on a school’s discipline management, we include additional variables to control for their potential effects on the dependent disciplinary responses. Socioeconomic disadvantage, for example, is a strong predictor of the degree to which students are punished in schools (Kupchik and Ward 2014; Ramey 2015); we represent it in this study as the percent of students receiving free or reduced-price lunch (percent students free/reduced-price lunch). In addition, among the strongest predictors of school punitiveness has been the proportion of students who are black (Payne and Welch 2010; Welch and Payne 2010); we control for the effect of black student composition with percent black students. Because some research suggests that stereotypes about Asians as a “model minority” group are associated with a preference for less harsh social controls (Brown 2016), we include percent Asian students as a control variable predicting school discipline. Finally, because boys tend to receive harsher discipline (Ferguson 2001), the analyses include percent male students.

The size of student enrollment (Byrd et al. 2015) and grade levels taught (Kupchik 2010) are also related to the punitiveness of disciplinary responses, with smaller schools and primary schools generally using milder responses. Accordingly, we include school size and grade level, operationalized with dummy variables (yes = 1, no = 0) for primary school and middle school, with high school serving as the reference category.

Two additional circumstances that have been positively associated with school punitiveness and disciplinary practices are the levels of crime and disorder in the school and the surrounding community (Welch and Payne 2010). School Crime is a three-item scale that adds together the number of violent incidents (rape or sexual battery, robbery, weapon possession or use, and attacks or threats of attack), property offense incidents (larceny theft and vandalism), and substance infractions (distribution, use or possession of drugs or alcohol). Community crime is a principal-reported item that indicates whether the community in which the school is located experienced a “high,” “moderate,” or “low” level of crime. Given the frequency distribution, we calculated this variable according to whether a community had high/moderate crime (= 1) or low crime (= 0).

Finally, geographic characteristics may influence the punitive responses available to and used by schools. Although limited by the anonymity of the schools within the data set and our inability to pair schools with location characteristics that are not already included in the data, we are able to include a control for the urbanicity of a school’s surrounding community, which has been associated with increased punitiveness (Kupchik 2010). This measure used categories established in the 2000 U.S. census and is broken into descriptive dummy variables (yes = 1, no = 0) for urban fringe, town, and rural, with urban serving as the reference category. Because the ethnic composition of the state in which each school is located may influence a school’s policies (Holmes et al. 2008), we include the percent of Latino/a residents in each state (state percent Latino/a), taken from the 2010 census. Finally, because southerners tend to endorse more punitive policies and practices (Ramey 2015), we include South as a dichotomous measure constructed from U.S. census data that indicates whether a school is located in the American South (= 1) or not (= 0).

Analytic Strategy

To test this study’s first hypothesis about specific disciplinary policies and practices, we estimated seven binary logistic regression models that predict the availability of probation, detention, privilege loss, community service, in-school suspension, out-of-school suspension, and expulsion as policy options.^3-5 To test the second hypothesis, we estimated seven additional binary logistic regression models that predict the actual use of those same dependent measures in practice. All models include independent interval measures for percent Latino/a students, percent black students, percent Asian students, percent students free/reduced-price lunch, percent male students, school size, School Crime, and state percent Latino/a. We also included dichotomous variables for community crime, primary school, middle school, urban fringe, town, rural, and South, with the reference variables high school and urban omitted. In all models, the Hosmer and Lemeshow χ² test was used to examine the overall fit of the model; well-fitting models result in nonsignificant χ², indicating that the predicted model does not differ significantly from the observed values. Additionally, we used the Wald χ² test to determine the significance of individual parameter estimates (Menard 2002).⁶

The third hypothesis examines the potential interactive effects of Latino/a composition with school crime and student poverty; that is, we investigated the possibility that school crime and economic disadvantage interact with the percent of Latino/a students in schools, thereby exacerbating school punitiveness beyond what each individual variable might contribute. To explore this question, we multiplied School Crime and percent students free/reduced-price lunch by percent Latino/a students. We then regressed the 14 binary discipline responses on these interaction terms in separate analyses to determine whether either interaction had a significant effect. The equations include the variables that comprise each interaction term, and the continuous variables were mean-centered to reduce multicollinearity (Menard 2002). All other control variables were also included.

Results

Results for the first seven binary logistic regression models shown in Table 2 partially support Hypothesis 1. Percent Latino/a students is significantly and negatively related to the availability of probation (b = −.508, p = .046) and privilege loss (b = −1.435, p = .022); it is significantly and positively related to the availability of out-of-school suspension (b = .242, p = .032). For each 1 percent increase in the percent Latino/a students, the log odds of a school having probation and privilege loss available in its disciplinary policy decrease by .490 and 1.435, respectively, and the log odds of a school having out-of-school suspension available in its disciplinary policy increase by .242. Examining the odds ratio of these parameter estimates shows that for each 1 percent increase in the percent Latino/a students, the odds of a school having probation and privilege loss available (vs. not having them available) decreases by a factor of .602 and .238, respectively, while the odds of a school having out-of-school suspension available (vs. not having it available) increases by a factor of 1.273. Several control variables are significant, with some clear patterns evident. School Crime is positively related to the availability of all disciplinary responses, and primary school is negatively related to all responses except in-school suspension. Percent black students and South are both negatively related to the availability of milder discipline and positively related to the availability of punitive discipline.

Table 2.

Binary Logistic Regression Results for Availability of Individual Disciplinary Policies.

	Probation		Detention		Privilege Loss		Community Service		In-school Suspension		Out-of-school Suspension		Expulsion
Coefficients	b (SE)	e (b)	b (SE)	e (b)	b (SE)	e (b)	b (SE)	e (b)	b (SE)	e (b)	b (SE)	e (b)	b (SE)	e (b)
Constant	.234 (.284)	1.263	2.045** (.370)	12.369	2.454 (.597)	11.640	.245 (.275)	1.278	.697* (.308)	1.996	2.669* (.468)	14.43	1.808* (.305)	4.089
Percent Latino/a students	−.508* (.296)	.602	−.398 (.377)	.672	−1.435* (.627)	.238	.333 (.286)	1.395	−.291 (.329)	.748	.242* (.166)	1.273	.292 (.125)	1.339
School Crime	.004** (.001)	1.004	.004* (.002)	1.004	.005* (.003)	1.005	.002* (.001)	1.002	.003* (.001)	.1.003	.007* (.003)	1.007	.004** (.001)	1.004
Community crime	.313* (.128)	1.368	.240 (.166)	1.272	.416 (269)	1.515	.035 (.119)	1.036	.056 (.137)	1.058	.327 (.212)	1.386	.013 (.132)	1.013
Percent free/reduced lunch	−.001 (.003)	.999	−.001 (.003)	.999	.006 (.005)	1.006	−.005* (.002)	.995	−.001 (.003)	.999	−.003 (.004)	.997	.000 (.003)	1.000
Percent black students	−.007 (.260)	.993	−.793* (.318)	.453	−1.158* (.497)	.314	−.211 (.254)	.810	.386* (.188)	1.472	.719** (.253)	2.053	.350* (.152)	1.419
Percent Asian students	−1.007 (.583)	.365	−.238 (.795)	.788	−.616 (1.301)	.540	−.311 (.549)	.733	−1.30* (.576)	.2872	−.690 (.904)	.502	−1.481* (.591)	.227
Percent male students	.008* (.004)	1.008	.000 (.005)	1.000	−.004 (.009)	.996	.000 (.004)	1.000	−.003 (.005)	1.003	.006 (.007)	1.006	.008* (.005)	1.008
School size	.001* (.001)	1.000	.001* (.001)	1.000	.000 (.000)	1.000	.001 (.001)	1.000	.001 (.001)	1.000	.001 (.001)	1.000	.001 (.001)	1.000
Primary school	−1.03* (.128)	.357	−1.74** (.159)	.175	−.493* (.259)	.611	−.799* (.132)	.450	−014 (.138)	.986	−1.05* (.211)	.351	−1.308* (.142)	.270
Middle school	−.222* (.114)	.801	.011 (.165)	1.011	.136 (.257)	1.146	−.116 (.105)	.891	.159 (.130)	.933	.104 (.219)	1.110	−.425** (.129)	.654
Urban fringe	.148 (.122)	1.159	−.082 (.160)	.922	.738** (.259)	2.092	−.463* (.115)	.629	.023 (.130)	1.023	.204 (.194)	1.226	.277* (.127)	1.320
Town	.235 (.173)	1.265	−.130 (.221)	.878	1.123* (.435)	3.075	−.542* (.168)	.581	.443* (.202)	1.557	.754* (.337)	2.125	.348 (.192)	1.416
Rural	.255* (.145)	1.290	−.189 (.186)	.828	.443 (.284)	1.557	−.558* (.138)	.573	.343* (.162)	1.410	.220 (.235)	1.246	.342* (.154)	1.407
State percent Latino/a	.001 (.005)	1001	.003 (.007)	1.003	.025* (.012)	1.026	.009* (.005)	1.009	.002 (.006)	1.002	.019* (.005)	1.020	.006* (.003)	1.006
South	.008 (.100)	1.008	−.113 (.128)	.893	−.428* (.207)	.652	−.478* (.097)	.620	.351** (.114)	1.421	.317** (.272)	1.362	.355** (.114)	1.426
Hosmer and Lemeshow χ²	9.740		8.320		4.644		20.287		5.369		13.753		4.117
df	8		8		8		8		8		8		8
p value	.284		.403		.795		.009		.718		.188		.846

p < .05. **p < .01.

The binary logistic regression models shown in Table 3 report the findings of the test of Hypothesis 2. As predicted, percent Latino/a students is significantly and negatively related to the use of probation (b = −1.85, p = .044), detention (b = −.490, p = .008), and community service (b = −.683, p = .004). For each 1 percent increase in the percent Latino/a students, the log odds of a school using probation, detention, or community service as a disciplinary practice decrease by 1.85, .490, and .683, respectively. By contrast, percent Latino/a students is significantly and positively related to the use of in-school suspension (b = .940, p = .042), demonstrating that for each 1 percent increase in the percent Latino/a students, the log odds that a school will use in-school suspension as a disciplinary practice increase by .940. Examining the odds ratio of all these parameter estimates shows that for each 1 percent increase in the percentage of Latino/a students, the odds of a school using probation, detention, or community service (vs. not using them) decrease by a factor of .157, .374, and .505, respectively, whereas the odds of a school using in-school suspension (vs. not using in-school suspension) increase by a factor of 2.560. Several control variables are significant, although fewer than in the first set of models. School Crime is positively related to the use of probation and all harsher discipline, while primary school is negatively related to the use of all responses except probation. Finally, percent black students is negatively related to the use of milder responses and positively related to the use of harsher responses.

Table 3.

Binary Logistic Regression Results for Use of Individual Disciplinary Practices.

	Probation		Detention		Privilege Loss		Community Service		In-school Suspension		Out-of-school Suspension		Expulsion
Coefficients	b (SE)	e (b)	b (SE)	e (b)	b (SE)	e (b)	b (SE)	e (b)	b (SE)	e (b)	b (SE)	e (b)	b (SE)	e (b)
Constant	.943* (.555)	2.568	2.874* (.932)	8.634	2.442** (.956)	7.992	2.037** (.645)	7.885	2.546* (.643)	12.75	.621 (.501)	1.861	−.265 (.372)	.767
Percent Latino/a students	−1.85* (.999)	.157	−.490* (.249)	.374	−.919 (.833)	.399	−.683** (.238)	.505	−.048 (.284)	.953	.940* (.508)	2.560	−.229 (.403)	.795
School Crime	.006* (.003)	1.006	−.001 (.003)	.999	.004 (.005)	1.004	.004 (.003)	1.004	.019** (.005)	1.020	.009** (.004)	1.009	.008** (.002)	1.008
Community crime	.396 (.272)	1.487	−.318 (.352)	.744	.426 (.412)	1.531	−.098 (.279)	.907	−.053 (.278)	.945	−.080 (.231)	.923	.197 (.158)	1.217
Percent free/reduced lunch	.006 (.005)	1.006	.010 (.007)	1.008	−.001 (.007)	.999	.005 (.005)	1.005	.007 (.005)	1.007	.005 (.004)	1.005	.006 (.003)	1.006
Percent black students	−1.151* (.219)	.316	−974* (.455)	.063	−.155 (.817)	.857	−.377* (.214)	.686	.530* (.215)	1.699	1.630* (.516)	5.103	.646* (.325)	1.907
Percent Asian students	.702 (.550)	2.018	.861 (1.835)	3.690	−1.549 (1.450)	.213	.090 (1.284)	1.095	−.253 (1.277)	.777	1.161 (1.159)	3.192	−1.069 (.875)	.344
Percent male students	−.001 (.008)	.999	−.013 (.014)	.987	−.012 (.014)	.988	−.007 (.010)	.993	−.003 (.009)	.997	.002 (.007)	1.002	−.006 (.006)	.994
School size	.001 (.000)	1.001	.001 (.000)	1.001	.001 (.000)	1.001	.000 (.000)	1.000	.001 (.001)	1.00	.002 (.000)	1.002	.001 (.001)	1.001
Primary school	.847 (.633)	2.333	−2.779* (.448)	.795	−.653* (.354)	.521	−1.501** (.281)	.223	−1.522* (.271)	.218	−1.674* (.209)	.187	−2.055* (.193)	.128
Middle school	−.113 (.206)	.893	−.199 (.684)	1.373	.381 (.380)	1.464	.902 (.673)	2.465	−.426 (.273)	.653	−.303 (.217)	.739	−.524** (.124)	.592
Urban fringe	−.322 (.260)	.725	−.345 (.340)	.721	−.412 (.392)	.663	−.421 (.258)	.656	−.165 (.271)	.848	.178 (.221)	1.195	−.036 (.161)	.965
Town	−.012 (.338)	.988	−.499 (.448)	.600	−.373 (.529)	.688	−.164 (.385)	.849	−.462 (.353)	.630	.113 (.284)	1.120	.248 (.209)	1.281
Rural	−.207 (.288)	.813	−.018 (.008)	.982	−.426 (.452)	.653	−.209 (.315)	.812	−.467 (.307)	.627	.250 (.249)	1.284	−.040 (.183)	.961
State percent Latino/a	−.010 (.009)	.990	.018 (.014)		.000 (.014)	1.000	.003 (.010)	1.00	−.007 (.010)	.993	.116* (.442)	1.136	.224* (.127)	1.996
South	.060 (.186)	1.062	−.075 (.272)	.969	−.164 (.300)	.849	.179 (.565)	1.196	.504* (.217)	1.584	−.081 (.171)	.922	−.265 (.372)	.799
Hosmer and Lemeshow χ²	19.727		6.780		10.132		9.610		6.270		20.941		31.904
df	8		8		8		8		8		8		8
p value	.011		.561		.256		.293		.617		.007		.001

p < .05. **p < .01.

Table 4 shows the last set of binary logistic regression results. These examine the potential conditioning effects of school crime and the economic composition of the student body by regressing all 14 of the binary disciplinary outcomes on interactions between percent Latino/a students and two school context variables: School Crime and the percent students free/reduced-price lunch. Only the models with significant interaction terms are shown. Hypothesis 3 predicts moderated relationships between these variables such that a higher percentage of Latino/a students would show a greater association with discipline in schools with less crime and in schools with a smaller percentage of economically disadvantaged students. As demonstrated, the interaction between percent Latino/a students and School Crime significantly affects four of the disciplinary outcomes: use of probation (b = −.220, p = .002), privilege loss (b = −.200, p = .049), in-school suspension (b = −.280, p = .014), and out-of-school suspension (b = −.190, p = .023). These significant negative interaction terms illustrate that the effect of student Latino/a composition on discipline practices depends on the level of school crime. Figures 1 through 4 show these interactions plotted, with School Crime split at the median. Percent Latino/a students increases the log odds of the use all four disciplinary outcomes in both low- and high-crime schools. However, the slope of this relationship is steeper in low-crime schools, indicating that percent Latino/a students is more strongly associated with the use of certain disciplinary outcomes in schools with less crime, thus supporting Hypothesis 3. It is noteworthy that the interaction between Latino/a composition and School Crime is not significantly associated with the availability of any of these responses but only the use of these practices. By contrast, none of the models investigating the interaction between percent Latino/a students and percent students free/reduced-price lunch were statistically significant.

Table 4.

Binary Logistic Regression Results for Individual Disciplinary Responses on Significant Interaction Terms.^a

	Probation Use		Privilege Loss Use		In-school Suspension Use		Out-of-school Suspension Use
Coefficients	b (SE)	e (b)	b (SE)	e (b)	b (SE)	e (b)	b (SE)	e (b)
Constant	1.410* (.595)	4.095	2.547* (.987)	8.598	2.381** (.687)	7.992	1.505** (.543)	4.506
Percent Latino/a × school crime	−.022** (.007)	.978	−.020* (.010)	.981	−.028* (.011)	.972	−.019* (.008)	.981
Mean-centered percent Latino/a	.679 (.647)	1.971	−.949 (.854)	.387	−1.034 (.642)	.355	.293 (.513)	1.341
Mean-centered School Crime	.009* (.004)	1.009	.009 (.006)	1.009	.024 (.006)	1.024	.012** (.004)	1.012
Community crime	.420 (.276)	1.522	.427 (.419)	1.533	−.034 (.287)	.967	−.063 (.233)	.939
Percent free/reduced lunch	.006 (.005)	1.006	−.002 (.007)	.998	.007 (.005)	1.007	.005 (.004)	1.005
Percent black students	−1.943* (1.019)	.143	−.192 (.815)	.825	−.445 (.584)	.641	1.587* (.516)	4.889
Percent Asian students	.633 (.551)	1.883	−1.627 (1.456)	.197	−.328 (1.272)	.720	1.084 (1.158)	2.957
Percent male students	−.001 (.008)	1.000	−.011 (.014)	.989	−.002 (.009)	.998	.002 (.007)	1.002
School size	.001** (.000)	1.001	.000 (.000)	1.000	.000 (.000)	1.000	.002** (.000)	1.002
Primary school	−1.108* (.221)	.330	−.582 (.355)	.559	−1.462* (.271)	.232	−1.638** (.210)	.194
Middle school	−.092 (.207)	.912	.410 (.380)	1.507	.394 (.273)	.674	−.284 (.218)	.753
Urban fringe	−.306 (.261)	.736	−.403 (.393)	.668	−.165 (.271)	.848	.180 (.221)	1.197
Town	−.012 (.340)	.988	−.368 (.531)	.692	−.473 (.354)	.623	.101 (.285)	1.106
Rural	−.170 (.291)	.844	−.368 (.454)	.692	−.455 (.307)	.634	.264 (.249)	1.303
State percent Latino/a	−.010 (.009)	.990	−.001 (.014)	.999	−.007 (.010)	.993	−.018* (.008)	.983
South	.059 (.187)	1.060	−.171 (.300)	.843	.522* (.215)	1.685	−.085 (.172)	.918
Hosmer and Lemeshow χ²	12.738		9.820		6.269		10.512
df	8		8		8		8
p value	.121		.278		.617		.272

p < .05. **p < .01,

Only models with significant interaction terms are shown.

Figure 1.

Interaction plot of simple slopes for impact of percent Latino/a students on probation use in low- and high-crime schools.

Figure 2.

Interaction plot of simple slopes for impact of percent Latino/a students on privilege loss use in low- and high-crime schools.

Figure 3.

Interaction plot of simple slopes for impact of percent Latino/a Students on in-school suspension use in low- and high-crime schools.

Figure 4.

Interaction plot of simple slopes for impact of percent Latino/a students on suspension use in low- and high-crime schools.

Discussion

The results of this study partially support the minority threat theory in relation to Latino/a student composition and school discipline, going substantially beyond what prior research has found. Hypotheses 1 and 2 are partially supported: Schools are more likely to adopt and use punitive out-of-school suspension when a disproportionately high number of Latino/a students is enrolled. Furthermore, schools with proportionally more Latino/a students are less apt to adopt the mild disciplinary policies of privilege loss and probation, and they are less apt to impose probation and detention in practice. Thus, a partial Latino/a threat effect is apparent for both the implementation of certain specific policies and use of certain specific practices, a finding not previously established. The Latino/a threat effect does not appear any less applicable for policy implementation than for use. Hypothesis 3 is partially supported as well: The findings demonstrate that levels of school crime and delinquency moderate the relationship between Latino/a student composition and the use of two mild practices (privilege loss and probation) and two punitive practices (in-school suspension and out-of-school suspension). We found no moderating effects of school crime or student economic disadvantage on policy implementation.

Hypotheses 1 and 2 were supported for many disciplinary responses, but it is important to recognize that significant relationships were not consistently established across all types. For instance, the percentage of Latino/a students was not related to the availability of detention, community service, in-school suspension, or expulsion; nor was it related to the use of privilege loss, in-school suspension, or expulsion. This suggests that a more nuanced examination of each type of disciplinary response could be useful. For example, the lack of significance for both the availability and use of expulsion could be due to the extreme punitive nature of this response and a corresponding insusceptibility to ethnicity-driven social control. Additionally, the lack of a negative relationship between percentage of Latino/a students and some of the milder responses—such as detention, community service, and privilege loss—may be the result of schools responding to student misbehavior with multiple types of discipline at once, adding milder methods to punitive ones to increase overall social control (Ipsa-Landa 2017). It is possible that when any form of discipline is used, schools may become even harsher in predominantly Latino/a schools simply because of the number of disciplinary measures (including mild and restorative ones) implemented in them, thereby obscuring relationships between mild discipline and Latino/a school composition.

Among this study’s unique contributions, we found that schools with a higher percentage of Latino/a students are likelier to both have the policy option of sanctioning rule violators with out-of-school suspension as well as use out-of-school suspension in practice. This is a relatively severe exclusionary tactic: Students are removed from the learning environment and suffer the consequences of that lost opportunity. On the other hand, neither the availability nor the use of the other exclusionary punishments (expulsion and in-school suspension) were related to Latino/a school composition. Because expulsion is the most severe punishment, its adoption and use might be reserved for only the most serious cases, regardless of student ethnicity. In contrast, because in-school suspension is moderate compared to the other punitive measures, it might be used both as a developmental consequence in some schools yet as a harsh tactic in others, thereby negating any statistically significant association with Latino/a composition.

This study’s findings in relation to the control variables provide additional support for the Latino/a threat thesis. The proportion of a state that is Latino/a is related to increased implementation of harsh policies permitting out-of-school suspension and expulsion but also to an increase in the milder policies of privilege loss and community service; together, this could intensify overall punitiveness (Ipsa-Landa 2017). Schools in states with more Latinos/as are also more likely to use exclusionary out-of-school suspension and expulsion in practice but not restorative discipline. These findings are not entirely surprising given the role of certain statewide school policies: State policymakers may approve harsher school-level policies and practices when their states have disproportionately more Latino/a residents. Using this logic, additional ethnic contexts at other levels of government, or within school districts and communities, likely affect school discipline and indicate a minority threat effect. Because it is adults who ultimately enact school policies and impose school punishments, it is important to consider the size of schools’ Latino/a populations relative to the racial and ethnic composition of those with the authority to impose social control in schools.

This study is the first to explore how levels of school-based delinquency and student economic disadvantage might confound the influence of Latino/a student composition on disciplinary policies and practices. Hypothesis 3 was confirmed: School crime moderates the relationship between Latino/a student composition and discipline such that Latino/a composition has a stronger effect in schools where crime is not already independently driving more punitive disciplinary approaches. The Latino/a threat effect was larger in schools with less school crime when it came to the use of multiple disciplinary practices but not policies. This suggests that school officials (but not policymakers) might use more discretion in assigning discipline in schools with more Latino/a students when school delinquency is lower and punishments may not already be as severe as they would be in high crime schools. In lower crime schools, Latino/a threat might have more opportunity to manifest. It is possible that the differential results for policy and practice in the context of school crime mean that minority threat is less influential when social control and school crime are abstract, as they often are for state-level policymakers. We found no similar moderating effect of student socioeconomic disadvantage, suggesting that the association between Latino/a student composition and disciplinary approach is unrelated to students’ economic characteristics.

Overall, this study’s findings substantially extend what we know about the minority threat framework in relation to Latino/a students. U.S. schools with relatively more Latino/a students have harsher policies and use harsher practices even when controlling for student-, school-, community-, and state-level characteristics that had not been previously examined together. Furthermore, schools are more likely to magnify disciplinary harshness in response to larger Latino/a student populations if the schools are not already using harsh measures that correspond to high levels of school offending. This suggests that at some level, discretion is responsible for these ethnic disparities. This finding corresponds with previous minority threat research that reveals ethnic and racial threat with regard to both Latinos/as and blacks in both the criminal justice system and schools: Punitive measures proliferate and mild or restorative alternatives are withheld in places where there are more persons of color. The exception here pertains to the size of the Asian student body: Schools with proportionally more Asian students enrolled are less apt to endorse harsh disciplinary policies (in-school suspension and expulsion), as would be expected by those who stereotype individuals of Asian descent as “model minorities” (Brown 2016). An effect contrary to minority threat appears to be operating in the context of Asian student composition that is worth further consideration.

Our research contributes to a body of work demonstrating that regardless of other influences, institutions with disproportionately high numbers of Latinos/as favor harsher social control mechanisms. This suggests that various forms of social control are imposed to subordinate those who are perceived to threaten the dominant white majority. Prior research indicates this is true for Latinos/as in the criminal justice system, and this study’s results show it is also apparent in educational institutions. This phenomenon is likely operating in other institutions and contexts and could be evidence of a broader neo-liberalization process occurring within society (Simmons 2017).

Another important conclusion that can be made about the results of this study is that the inconsistencies evident with regard to Latino/a composition do not apply to racial composition: The proportion of black students is uniformly associated with the increased availability and use of all three types of exclusionary punishment. Schools’ composition of black students is also more consistently associated with decreased adoption and use of restorative discipline, with only policies for community service and probation and the practice of privilege loss being unassociated with it. Moreover, the effect of black school composition is consistently stronger than the effect of Latino/a school composition. These comparative findings suggest that a high percentage of black students in schools may present a more powerful threat, requiring more exclusionary methods, than that presented by a high enrollment of Latino/a students. These findings also indicate that the benefits of restorative discipline are even less accessible in schools with more black students than in schools with more Latino/a students, and the privileges of restorative methods are most available to students at predominantly white schools.

Along with expanding our understanding of the minority threat effect and showing its relevance for Latinos/as in a particular institutional setting, our results also have significant implications for students and school discipline. Together with state policymakers, school districts and administrators must actively and thoughtfully reevaluate school policies and practices to ensure that discipline is not imposed in a discriminatory manner. By uniformly favoring preventive and restorative disciplinary options, schools will better serve students and enhance their educational and future success (Losen 2011; Payne and Welch 2015). Encouraging programming that aims to reduce school authorities’ and policymakers’ biases, which may be facilitating Latino/a threat, could also weaken the effect of Latino/a school composition on punitive discipline. Finally, eliminating the structural forces that sustain the micro-processes that give rise to inequality should be the ultimate goal.

Study Limitations and Future Research

The implications of this study are potentially constrained by certain limitations. The secondary data analyzed here come with the typical limitations inherent to this type of research. In particular, this study’s measures were selected according to what was available in this data set; for example, data regarding local community- or district-level ethnic composition were unavailable. We could particularly benefit from future research that examines how discipline is associated with the size of the Latino/a student population relative to the racial and ethnic composition of the community and those with the power to shape disciplinary processes. As with previous research, we could only presume that a white majority was most responsible for imposing control. Future examinations of how community ethnic and racial context is associated with school discipline should also examine the possibility of a tipping point effect on discipline at high levels of Latino/a composition within communities.

Future research should also explore how individual disciplinary policies and practices are used in conjunction to intensify the social control of students. This study contributes to our understanding of how these individual policies and measures are applied in school ethnic contexts, but learning to what extent and under what conditions these individual policies and practices coexist could illuminate more about how Latino/a threat may operate in school settings. It is possible that in schools with proportionally more Latino/a students, mild responses are not used with a restorative objective but instead are used in conjunction with punitive punishments to make them even more controlling.

We would also benefit from future research that examines to what extent individual-level decisions by school personnel are responsible for disparate outcomes in schools with proportionally more Latino/a students. Individual teacher, administrator, and policymaker race and ethnicity do not necessarily mitigate bias and discrimination against students (Bradshaw et al. 2010; Pigott and Cowen 2000), but controlling for authorities’ ethnicity may help differentiate what portion of punitiveness in schools is attributable to macro-level Latino/a threat versus individual bias against Latino/a students. Further, it may be helpful for subsequent analyses to disaggregate the Latino/a student population according to nationality of origin, generational immigration status, and English language learners.

Another line of inquiry should address the fact that the racial threat perspective does not hold up in relation to Asian student composition in these analyses. The current findings support the “model minority” hypothesis and warrant more exploration and theorizing. An examination of how policymakers’ political affiliations affect school policies and practices would also be worthwhile, as would research examining the effect of Latino/a threat in the context of punitive environments and security technology. This study is also limited by the cross-sectional nature of the data, which precludes making assumptions about causality. Future researchers may benefit from collecting longitudinal data that will allow these conclusions.

Conclusion

To date, this research is the most comprehensive examination of the dynamics shaping Latino/a disparities in school punishments. Using a nationally representative sample and measuring many different dimensions of school discipline—all while controlling for the effects of other theoretically and empirically cogent influences—we were able to build on the work of prior researchers, expanding our understanding of how minority threat operates in relation to Latinos/as in schools. Our findings generally support the minority threat theory, with results indicating that Latino/a student composition is related to the adoption of myriad disciplinary policies and practices. In addition, we found that school crime has a moderating influence such that the relationship between Latino/a student composition and punitive discipline is stronger in schools with less crime. Our results also indicate that although a Latino/a threat effect does exist in schools, the association between student composition and discipline is not nearly as consistent or strong as it is for schools with proportionally more black students. This suggests the minority threat theory may be most applicable in the context of the size of the black population.

Significant problems are posed by the development and implementation of school policies and practices that are based—either intentionally or inadvertently—on the size of minority populations and implicit assumptions about Latino/a youth and delinquency, which is what this study indicates is occurring. This phenomenon not only reinforces prevailing perceptions of Latino/a individuals as delinquents and criminals, but it strengthens the unfortunate school-to-prison pipeline that Latinos/as disproportionately eventually endure.

Footnotes

Notes

Author Biographies

Kelly Welch is an associate professor in the Department of Sociology and Criminology at Villanova University. Her research interests include racial and ethnic justice, criminal stereotypes, sociology of punishment, and the school-to-prison pipeline.

Allison Ann Payne is a professor in the Department of Sociology and Criminology and associate director of the Honors Program at Villanova University. Her research interests include juvenile delinquency, school violence and disorder, school-based delinquency prevention, and criminological theory.

References

Anyon

Yolanda

Jenson

Jeffrey M.

Altschul

Inna

Farrar

Jordan

McQueen

Jeanette

Greer

Eldridge

Downing

Barbara

Simmons

John

. 2014. “The Persistent Effect of Race and the Promise of Alternatives to Suspension in School Discipline Outcomes.” Children and Youth Services Review 44:379–86.

Beck

Audrey N.

Muschkin

Clara G.

2012. “The Enduring Impact of Race: Understanding Disparities in Student Disciplinary Infractions and Achievement.” Sociological Perspectives 55(4):637–62.

Bekkerman

Anton

Gilpin

Gregory

. 2014. “On Understanding Inconsistent Disciplinary Behaviour in Schools.” Applied Economics Letters 22(10):772–76.

Bekkerman

Anton

Gilpin

Gregory

. 2016. “Crime and Punishment: The Role of Student Body Characteristics in Schools’ Disciplinary Behaviours.” Applied Economics 48(15):1402–15.

Blalock

Hubert M.

1967. Towards a Theory of Minority Group Relations. New York: Capricorn.

Blumer

Herbert

. 1958. “Race Prejudice as a Sense of Group Position.” The Pacific Sociological Review 1(1):3–7.

Bradshaw

Catherine P.

Mitchell

Mary M.

O’Brennan

Lindsey M.

Leaf

Philip J.

2010. “Multilevel Exploration of Factors Contributing to the Overrepresentation of Black Students in Office Disciplinary Referrals.” Journal of Educational Psychology 102(2):508–20.

Brown

Jessica A.

2016. “Running on Fear: Immigration, Race and Crime Framings in Contemporary GOP Presidential Debate Discourse.” Critical Criminology 24(3):315–31.

Bureau of Justice Statistics. 2016. “Total Correctional Population.” Retrieved January 26, 2018 (https://www.bjs.gov/index.cfm?ty=tp&tid=11).

10.

Byrd

Kaitland M.

Kahle

Lindsay L.

Peguero

Anthony A.

Popp

Ann Marie

. 2015. “Social Control and Intersectionality: A Multilevel Analysis of School Misconduct, Location, Race, Ethnicity, and Sex.” Sociological Spectrum 35:109–35.

11.

Cameron

A. Colin

Trivedi

Pravin K.

2005. Microeconometrics: Methods and Applications. Cambridge, UK: Cambridge University Press.

12.

Carter

Prudence L.

2012. Stubborn Roots: Race, Culture, and Inequality in U.S. and South African Schools. New York: Oxford University Press.

13.

Colby

Sandra L.

Ortman

Jennifer M.

2015. “Projections of the Size and Composition of the U.S. Population: 2014 to 2060: Current Population Reports.” Retrieved January 26, 2018 (https://www.census.gov/content/dam/Census/library/publications/2015/demo/p25-1143.pdf).

14.

Crawford

Charles

Chiricos

Ted

Kleck

Gary

. 1998. “Race, Racial Threat, and Sentencing of Habitual Offenders.” Criminology 36(3):481–512.

15.

Fabelo

Tony

Thompson

Michael D.

Plotkin

Martha

Carmichael

Dottie

Marchbanks, III

Miner P.

Booth

Eric A.

2011. Breaking Schools’ Rules: A Statewide Study of How School Discipline Relates to Students’ Success and Juvenile Justice Involvement. New York: Council of State Governments Justice Center.

16.

Ferguson

Ann Arnett

. 2001. Bad Boys: Public Schools in the Making of Black Masculinity. Ann Arbor, MI: The University of Michigan Press.

17.

Gregory

Anne

Weinstein

Rhona S.

2008. “The Discipline Gap and African Americans: Defiance or Cooperation in the High School Classroom.” Journal of School Psychology 46(4):455–75.

18.

Hagan

John

Shedd

Carla

Payne

Monica R.

2005. “Race, Ethnicity, and Youth Perceptions of Criminal Injustice.” American Sociological Review 70:381–407.

19.

Holmes

Malcolm D.

2000. “Minority Threat and Police Brutality: Determinants of Civil Rights Criminal Complaints in U.S. Municipalities.” Criminology 38(2):343–67.

20.

Holmes

Malcolm D.

Smith

Brad W.

Freng

Adrienne B.

Munoz

Ed A.

2008. “Minority Threat, Crime Control, and Police Resource Allocation in the Southwestern United States.” Crime and Delinquency 54:128–52.

21.

Ipsa-Landa

Simone

. 2017. “Racial and Gender Inequality and School Discipline: Toward a More Comprehensive View of School Policy.” Social Currents 4(6):511–17.

22.

Irwin

Katherine

Davidson

Janet

Hall- Sanchez

Amanda

. 2013. “The Race to Punish in American Schools: Class and Race Predictors of Punitive School- crime Control.” Critical Criminology 21(1):47–71.

23.

Keen

Bradley

Jacobs

David

. 2009. “Racial Threat, Partisan Politics, and Racial Disparities in Prison Admissions: A Panel Analysis.” Criminology 47(1):209–38.

24.

Kent

Stephanie L.

Jacobs

David

. 2004. “Social Divisions and Coercive Control in Advanced Societies: Law Enforcement Strength in Eleven Nations from 1975 to 1994.” Social Problems 51(3):343–61.

25.

Kim

Catherine Y.

Losen

Daniel J.

Hewitt

Damon T.

2010. The School-to-prison Pipeline. New York: New York University Press.

26.

Kupchik

Aaron

. 2009. “Things Are Tough All over: Race, Ethnicity, Class and School Discipline.” Punishment and Society 11(3):291–317.

27.

Kupchik

Aaron

. 2010. Homeroom Security: School Discipline in an Age of Fear. New York: New York University Press.

28.

Kupchik

Aaron

. 2016. The Real School Safety Problem: The Long-term Consequences of Harsh School Punishment. Oakland: University of California Press.

29.

Kupchik

Aaron

Ward

Geoff

. 2014. “Race, Poverty, and Exclusionary School Security: An Empirical Analysis of U.S. Elementary, Middle, and High Schools.” Youth Violence and Juvenile Justice 12(4):332–54.

30.

Lewis

Amanda E.

Diamond

John B.

2015. Despite the Best Intentions: How Racial Inequality Thrives in Good Schools. Oxford: Oxford University Press.

31.

Liska

Allen E.

1992. Social Threat and Social Control. Albany: SUNY Press.

32.

Losen

Daniel

. 2011. Discipline Policies, Successful Schools, and Racial Justice. Boulder, CO: National Education Policy Center.

33.

Losen

Daniel

Hodson

Cheri

Jongyeon

Martinez

Tia E.

2015. “Disturbing Inequities: Exploring the Relationship between Racial Disparities in Special Education Identification and Discipline.” Pp. 89–106 in Closing the School Discipline Gap, edited by Losen

D. J.

New York: Teachers College Press.

34.

Losen

Daniel

Skiba

Russell J.

2010. Suspended Education: Urban Middle Schools in Crisis. Los Angeles: The Civil Rights Project at UCLA, Center for Civil Rights Remedies.

35.

McClure

Timothy

May

David C.

2014. “Corporal Punishment in the United States.” Pp. 97–118 in School Safety in the United States: A Reasoned Look at the Rhetoric, edited by May

David C.

Durham, NC: Carolina Academic Press.

36.

Menard

Scott

. 2002. Applied Logistic Regression Analysis. Thousand Oaks, CA: Sage.

37.

Mowen

Thomas J.

Parker

Karen

. 2014. “Minority Threat and School Security: Assessing the Impact of Black and Hispanic Student Representation on School Security Measures.” Security Journal 9:504–22.

38.

Payne

Allison Ann

Welch

Kelly

. 2010. “Racial Threat in Schools: Modeling the Influence of Punitive and Restorative Discipline Practices.” Criminology 48(4):1019–62.

39.

Payne

Allison Ann

Welch

Kelly

. 2015. “Restorative Justice in Schools: The Influence of Race on Restorative Discipline.” Youth & Society 47(4):539–64.

40.

Peguero

Anthony A.

Popp

Ann Marie

Shekarkhar

Zahra

. 2015. “Breaking Stereotypes and School Punishment: Family Socioeconomic Status, Test Scores, Academic and Sport Activities, Backlash, and Racial and Ethnic Discipline Disparities.” Journal of Ethnicity in Criminal Justice 13(1):59–86.

41.

Peguero

Anthony A.

Shekarkhar

Zahra

. 2011. “Latino/a Student Misbehavior and School Punishment.” Hispanic Journal of Behavioral Sciences 33(1):54–70.

42.

Pigott

Rowen L.

Cowen

Emory L.

2000. “Teacher Race, Child Race, Racial Congruence, and Teacher Ratings of Children’s School Adjustment.” Journal of School Psychology 38:177–95.

43.

Ramey

David M.

2015. “The Social Structure of Criminalized and Medicalized School Discipline.” Sociology of Education 88(3):181–201.

44.

Rios

Victor M.

2017. Human Targets: Schools Police, and the Criminalization of Latino Youth. Chicago: University of Chicago Press.

45.

Robers

Simone

Zhang

Anlan

Morgan

Rachel E.

Musu-Gillette

Lauren

. 2015. Indicators of School Crime and Safety: 2014. Washington, DC: National Center for Education Statistics.

46.

Roch

Christine H.

Edwards

Jason

. 2017. “Representative Bureaucracy and School Discipline: The Influence of Schools’ Racial Contexts.” American Review of Public Administration 47(1):58–78.

47.

Rocque

Michael

. 2010. “Office Discipline and Student Behavior: Does Race Matter?” American Journal of Education 116(4):557–81.

48.

Rocque

Michael

Paternoster

Raymond

. 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(2):633–65.

49.

Servoss

Timothy J.

Finn

Jeremy D.

2014. “School Security: For Whom and with What Results?” Leadership and Policy in Schools 13(1):61–92.

50.

Shedd

Carla

. 2015. Unequal City: Race, Schools, and Perceptions of Injustice. New York: Russell Sage Foundation.

51.

Simmons

Lizbet

. 2017. The Prison School: Educational Inequality and School Discipline in the Age of Mass Incarceration. Oakland, CA: University of California Press.

52.

Simon

Jonathan

. 2007. Governing through Crime: How the War on Crime Transformed American Democracy and Created a Culture of Fear. New York, NY: Oxford University Press.

53.

Skiba

Russel J.

Chung

Choong-Guen

Trachok

Megan

Baker

Timberly

Sheya

Adam

Hughes

2015. “Where Should We Intervene? Contributions of Behavior, Student, and School Characteristics to Out-of-school Suspension.” Pp. 132–146 in Closing the School Discipline Gap, edited by Losen

D. J.

New York: Teachers College Press.

54.

Skiba

Russel J.

Mediratta

Kavitha

Rausch

M. Karega

. 2016. Inequality in School Discipline. New York: Palgrave Macmillan.

55.

Skiba

Russell J.

Michael

Robert S.

Nardo

Abra Carroll

. 2002. “The Color of Discipline: Source of Racial and Gender Disproportionality in School Discipline.” The Urban Review 34(4):317–42.

56.

Skiba

Russell J.

Peterson

Reece L.

2000. “School Discipline at a Crossroads: From Zero Tolerance to Early Response.” Exceptional Children 66(3):335–47.

57.

Skiba

Russell J.

Rausch

M. Karenga

. 2006. School Disciplinary Systems: Alternatives to Suspension and Expulsion. Children’s Needs III: Development, Prevention, and Intervention. Bethesda, MD: National Association of School Psychologists.

58.

Stewart

Eric A.

Martinez, Jr.

Ramiro

Baumer

Eric P.

Gertz

Marc

. 2015. “The Social Context of Latino/a Threat and Punitive Latino/a Sentiment.” Social Problems 62(1):68–92.

59.

Stolzenberg

Lisa

D’Alessio

Stewart

Eitle

David

. 2004. “A Multilevel Test of Racial Threat Theory.” Criminology 42(3):673–98.

60.

Thompson

Mary Lou

Myers

J. E.

Kriebel

1998. “Prevalence Odds Ratio or Prevalence Ratio in the Analysis of Cross Sectional Data: What Is to Be Done?” Occupational and Environmental Medicine 55:272–77.

61.

U.S. Department of Education. 2013. “The Transformed Civil Rights Data Collection.” Retrieved January 26, 2018 (http://www2.ed.gov/about/offices/list/ocr/docs/crdc-2012-data-summary.pdf).

62.

Wallace

John M.

Jr. Goodkind

Sara

Wallace

Cynthia M.

Bachman

Jerald G.

2008. “Racial, Ethnic, and Gender Differences in School Discipline among U.S. High School Students: 1991–2005.” Negro Educational Review 59(1–2):47–62.

63.

Welch

Kelly

Payne

Allison Ann

. 2010. “Racial Threat and Punitive School Discipline.” Social Problems 57(1):25–48.

64.

Welch

Kelly

Payne

Allison Ann

. 2012. “Exclusionary School Punishment: The Effect of Racial Threat on Expulsion and Suspension.” Youth Violence and Juvenile Justice 10(2):155–71.

65.

Welch

Kelly

Payne

Allison Ann

Chiricos

Ted

Gertz

Marc

. 2011. “The Typification of Hispanics as Criminals and Support for Punitive Crime Control Policies.” Social Science Research 40:822–40.