Discipline and Participation

Abstract

This study uses the National Longitudinal Survey of Adolescent Health data set to evaluate the long-term influence of school discipline and security on political and civic participation. We find that young adults with a history of school suspension are less likely than others to vote and volunteer in civic activities years later, suggesting that suspension negatively impacts the likelihood that youth engage in future political and civic activities. These findings are consistent with prior theory and research highlighting the long-term negative implications of punitive disciplinary policies and the role schools play in preparing youth to participate in a democratic polity. We conclude that suspension undermines the development of the individual skills and capacities necessary for a democratic society by substituting collaborative problem solving for the exclusion and physical removal of students. The research lends empirical grounds for recommending the reform of school governance and the implementation of more constructive models of discipline.

Keywords

school discipline political and civic participation political socialization

Introduction

Sweeping changes over the past two decades have reshaped the regime of school discipline and security in the United States. Schools of all kinds and in all parts of the country have increasingly adopted harsher, more punitive disciplinary policies, such as zero tolerance and mandatory arrest. They have augmented their use of police, metal detectors, and closed-circuit surveillance (see Casella, 2001; Cornell, 2006; Hirschfield, 2008; Simon, 2007). Though fair and firm discipline is necessary to maintain order in schools (Arum, 2003), research has repeatedly found that the overreliance on exclusionary school punishment (i.e., removing students from schools) and criminal justice–oriented security (e.g., police in schools, drug-sniffing police dogs) has a host of negative effects. These practices increase racial inequality (e.g., Skiba, Michael, Nardo, & Peterson, 2000), alienate students from school (Fine, Burns, Payne, & Torre, 2004; Nolan, 2011), and handicap students’ academic careers (Fabelo et al., 2011) while failing to teach students proper behavior or better protect them from harm (Kupchik, 2010). Some scholars (e.g., Effrat & Schimmel, 2003; Fine et al., 2004; Kupchik, 2010; Nolan, 2011) have suggested that among these other negative effects, harsh school discipline and rigid security practices may socialize students into docility and obedience, whereby they accept authority of adults rather than participate actively in political and civic exchange, though to date these suggestions have not been tested empirically.

During roughly the same period during which we have observed the increasing harshness of school punishment and rigid security, there has been a broad call to renew and reenergize American civic and political life. This is reflected in the substantial theoretical and empirical literatures on civic and political engagement, deliberative democracy, and social capital, among others (e.g., Buss, Redburn, & Guo, 2006; Creighton, 2005; Fuhrman & Lazerson, 2005; Fung & Olin Wright, 2003; Lin, 2002; Schachter & Yang, 2012; Sirianni & Friedland, 2001; Skocpol, 2003; Skocpol & Fiorina, 1999). It is, moreover, reflected concretely in the now established expectation (though the reality often falls short) that governments at all levels involve the public in matters large and small. Indeed the active exchange of information between citizens and government is central for governmental decision making and policy formulation (Bevir, 2006; Catlaw & Sandberg, 2012; Fung, 2006; Hale, 2011). These trends have refocused academic and practical attention on the ways in which citizens are politically socialized and how they develop the capacity to effectively participate in civic and political life (Campbell, 2006; Rawlings & Catlaw, 2011; Rawlings, 2012; Verba, Schlozman, & Brady, 1995). Schools figure prominently in these efforts.

These trends are at cross-purposes: Schools cannot be training grounds for a vibrant democratic polity if they suppress the development of students’ political and civic capacities. Furthermore, there may be grave consequences not only for the lives of individual students but also for the efficacy of the nation’s democratic institutions writ large if schools are playing this negative socializing role. Surprisingly, however, prior research has generally investigated neither the long-term effect of political socialization of schools nor has it tested longitudinally the consequences for students of the contemporary school discipline and security regime.

This study uses the National Longitudinal Survey of Adolescent Health (Add Health) data set to evaluate the long-term influence of school discipline on political and civic participation. Our goal is to examine whether school discipline discourages youth from civic and political participation in their adult years, so that we can better understand the ramifications of contemporary school discipline and security. Our methodological strategy follows directly from and extends a publication in the American Sociological Review, by McFarland and Thomas (2006), which used Add Health data to estimate how youth voluntary associations influence adult political participation. Though we share McFarland and Thomas’s overall analytical strategy, their analysis does not consider school discipline. We hypothesize that increasingly harsh security and discipline practices socialize students into docility and obedience, whereby they accept authority of adults rather than participate actively in political exchanges. We empirically examine this hypothesis by studying the long-term effects of school punishment and security on students’ future levels of democratic participation.

Literature Review

Schooling and Political Socialization

A wealth of research examines the factors that contribute to the likelihood that a young person will become active in political and civic life later in life (Verba et al., 1995). For our purposes, we focus on two broad categories of factors—the family and the school.

First, research establishes the importance of parents: If parents are civically engaged, children are more likely to be as well (Campbell, 2006, Chapter 6; Verba et al., 1995); however, children from homes of a higher socioeconomic status (SES) are more likely to become civically active than children from lower SES homes. Second, formal education plays a critical role in the “second” socialization of young people into society (Durkheim, 1903/1961) and, more particularly, school socialization is an important dimension of producing democratic citizens (Dewey, 1916, 1909/1959; Hartmann, 1946; Hyman, 1959). Early research on this topic focused on the transmission of civic and political knowledge, or the “manifest curriculum,” to students and was pessimistic about the ability of schools to alter students’ political attitudes or to encourage participation (Ehman, 1980, p. 103), though lower SES groups proved to be exceptions (K. Jennings & Jennings, 1968; M. K. Jennings & Niemi, 1974). This general pessimism has been compounded by research that indicates that political knowledge and participation in civil and political life may be in decline in the United States (Putnam, 2000; Skocpol, 2003) and that this decline, in turn, reinforces entrenched patterns of inequality and uneven political participation across social and racial groups (Kahne & Middaugh, 2008; McFarland & Thomas, 2006; Verba et al., 1995). However, other research has tempered these findings (Niemi & Junn, 1998; Youniss & Levine, 2009).

Contemporary scholarship continues to emphasize the importance of civic and political education as well as the work that schools do in preparing (or not preparing) youth for active, engaged participation in a democratic polity (e.g., Apple & Beane, 1999; Battisoni et al., 2003; Gutmann, 1987) and in reproducing the values, habits, and practices of the world beyond the classroom (Bourdieu, 1986). However, researchers have broadened their scope beyond civic and political knowledge to consider Dewey’s (1909/1959) concern for the method of education (see Dobozy, 2007) and the development of individual capacities to participate in public life.

Researchers argue that specific “democratic” capacities and resources (Rawlings, 2012; Verba et al., 1995) should be developed in schools which, in turn, will assist youth in becoming politically engaged later in life and will help to reinvigorate and deepen democracy more generally. In a review, Apple and Beane (1999) identify seven of these capacities, including an appreciation for an open flow of ideas, “faith in the individual and collective capacity of people” to solve problems, critical reflection, “concern for the welfare of others and the ‘common good,’” and a “concern for the dignity and rights of individuals and minorities” (p. 7). The cornerstone of cultivating these capacities is participation—that is, engagement of students in school-related planning, decision making, problem solving, and other activities that affect them at both the classroom and school levels. Exemplary democratic schools tend to view students as active, rights-bearing individuals rather than “objects to be acted upon” (Dobozy, 2007). Other education research tends to support this view: Student engagement with such “democratic practices,” such as school-place decision making, encourages the cultivation of positive school climate which, in turn, enables higher levels of educational achievement and related outcomes (Anderson, 1982, pp. 400-401; see also Effrat & Schimmel, 2003).

For the most part, empirical studies of the effects of democratic capacity building and political socialization have been case studies of individual schools or classrooms (e.g., Angell, 1998; Apple & Beane, 1999) and/or investigations of the effects of various democratic practices on students while students were in school (e.g., Feldman, Pasek, Romer, & Jamieson, 2007). However, empirical examination of the long-term effects of democratic schooling on political and civic participation after graduation from high school has been rare (though see Pasek, Feldman, Romer, & Jamieson, 2008; Zaff, Moore, Papillo, & Williams, 2003).

A significant exception is McFarland and Thomas (2006), who examined the ways in which extracurricular activities affected civic and political involvement later in life (e.g., voting, involvement in presidential campaigns, volunteering in community or civic organization). They found compelling evidence that “involvement in politically salient youth voluntary associations has significant, positive returns on adult political participation seven to twelve years later” (p. 412). Interestingly, classes in government and civics did not have such effects, a finding in line both with prior scholarship and with contemporary concern for the “hidden curriculum” (e.g., Giroux & Purpel, 1983), or the nonconscious learning children do in the school above and beyond the explicit transmission of knowledge (see also Ehman, 1980).

Although some researchers explore the difference between political and civil activities (Zukin, Keeter, Andolina, Jenkins, & Delli Carpini, 2006), they generally do not distinguish between civil and political effects (Rawlings, 2012) and so do not investigate whether particular school activities are more likely to encourage one over the other. This is likely because both political and civic behavior tend to be conditioned by similar familial factors and that the primary thrust of research has been on the development of individual capacities to engage in “public life” defined broadly. Indeed, even research that teases out factors contributing to civic versus political behavior tends not to test the effects of various school activities on future practice. Torney-Purta and Amadeo (2003), for example, found differences among predictors of voting and volunteering among adolescents. Their cross-sectional analysis, however, addressed only the impact of civic knowledge and not the broader palate of school-place activities on civic versus political behavior. A second example is Campbell (2006), who explored the impact of community political diversity on the long-term political and civic involvement of youth. Though his research affirms the importance of schools in cultivating “civic norms” of good citizenship, his studies focus more on the explicit normative content of school curricula rather than, again, the impact of ostensibly “nonpolitical” school-place practices on later civic and political involvement.

In summary, relevant research on the effects of schooling on the political socialization of youth has focused on classroom-level attributes, such as the characteristics of the teacher and instructional materials, or school-level attributes, such as student participation in school governance and extracurricular activities, as well as school climate and organization (e.g., school size, religiosity, demographic composition of schools/classrooms; Ehman, 1980). It has tended, moreover, to deal with political and civil behavior together. This body of prior research is consistent in demonstrating a direct positive effect of participatory, democratic, and inclusive schooling on future political and civic engagement. Participatory, democratic, and inclusive school environments teach students how to be active participants in democratic institutions, just as parents who participate in civic society teach their children to do so as well.

School Discipline

Since the early 1990s, schools across the United States have tightened their security practices and increased the punishments they give to students (see Cornell, 2006; Dinkes, Kemp, & Baum, 2009; Kupchik & Monahan, 2006). It is now common to find armed police officers, drug-sniffing dogs, surveillance cameras, and zero-tolerance policies in all types of schools and all areas of the United States. Existing research documents several problems with these new school discipline and security practices, including the increasing marginalization of poor students and youth of color (e.g., Noguera, 2003; Skiba et al., 2000), unnecessary denial of future educational opportunities due to suspension and expulsion (e.g., American Psychological Association Zero Tolerance Task Force, 2008; Fabelo et al., 2011), and increases in the numbers of students who are formally prosecuted in the juvenile and criminal justice systems (known as the “school-to-prison pipeline”; for example, Kim, Losen, & Hewitt, 2010; Na & Gottfredson, 2013; Wald & Losen, 2003). This body of research consistently finds large discrepancies in punishment rates between White youth and youth of color, where African American and Hispanic American students are far more likely than Whites to be punished, even when controlling for self-reported rates of misbehavior (American Psychological Association Zero Tolerance Task Force, 2008).

In Homeroom Security, Kupchik (2010) describes how the primary mission of school discipline is to assert the school’s authority: to enforce the rules for the sake of the rules themselves, not for the betterment of students. That is, the harsh punishments and tight security we now see in schools create conditions whereby students are disempowered and treated as objects to be acted on—exactly the opposite of what scholars propose for a democratic, participatory, and inclusive education (Lyons & Drew, 2006). Students are socialized to believe that they are powerless in the face of a rigid discipline system and that they are potential criminals rather than citizens who deserve respect (see Fine et al., 2004; Nolan, 2011). In sum, they are taught that their only option is to comply with the school’s authority with neither complaint nor ability to shape their environment; in this sense, school discipline offers lessons that are the antithesis of what prior research finds can directly and positively influence future democratic participation.

Based on the prior research on how schools can influence future democratic participation, we fear that the contemporary school discipline regime is preparing students for disengaged political and civic futures and that the lessons of compliance and obedience translate to a lack of participation once they become young adults. Moreover, because school discipline is disproportionately applied to youth of color, we are concerned that the effects of school discipline are particularly harmful to youth of color and their future civic and political engagement. As Fine et al. (2004) find,

. . . poor and working-class youth and youth of color in California’s most disadvantaged schools are being educated away from these “obligations of citizenship” and toward civic alienation. They are learning that their needs are irrelevant to policy makers and government leaders. (p. 2212)

Recent research by Godsay, Kawashima-Ginsberg, Kiesa, and Levine (2012) on working-class “non-college youth” reinforces this conclusion. Drawing on data collected from 20 focus groups with noncollege youth (ages 18-29) in four cities, their study found that these former students’ recollections of schools’ efforts to develop political and civic capacities were “overwhelmingly and sometimes scathingly critical” (p. 34). Former students described their schools as largely distrusting, disempowering environments.

Despite the importance of these questions for educational policy, the existing research has failed to consider how school discipline and security may be shaping future democratic participation and civic life in the United States. We address this gap in the research by empirically examining the long-term effects of the new school discipline regime in two ways. The first is at the individual level, as it considers students’ individual experiences with school discipline. Here we examine whether a history of suspension—the most common form of school punishment—relates to students’ future civic roles.

We hypothesize that: 1) students who have been suspended are less likely to vote and volunteer than others in future years; and 2) this effect is most pronounced for youth of color. Our second test is a school-level test that looks at school practices, hypothesizing that individuals who attend schools with rigid security mechanisms, such as police officers, metal detectors, surveillance cameras, and harsh punishment policies, are less likely to vote and volunteer than others in the future.

Method

Using data from the Add Health survey, we evaluate the long-term influence of school discipline and security on civic participation. The Add Health data include interviews of youth, school administrators, and parents during the 1994-1995 school year, when a nationally representative sample of adolescents were in Grades 7 to 12; the study follows up with them multiple times, through a fourth wave of interviews, completed in 2007-2008.

The Add Health study is compiled by the University of North Carolina (UNC) Population Center and funded by a number of agencies (including the National Science Foundation [NSF], National Institute of Mental Health [NIMH], Centers for Disease Control and Prevention [CDC], and National Institutes of Health [NIH]). It is a longitudinal, nationally representative sample of adolescents who were in Grades 7 to 12 in the 1994-1995 school year. It includes a cluster sample of 80 high schools selected from a sampling frame of 26,666, and their feeder schools. Within these schools, 90,118 students completed in-school questionnaires, and an administrator from each school completed an administrator questionnaire. Of these students, 20,745 were randomly selected to complete in-home interviews at multiple times (the fourth wave of data is now complete), as were their parents. The UNC Population Center provides wave-specific sampling weights that adjust for unequal probability of selection, thus offering a nationally representative view of adolescents’ experiences.

We use all cases of adolescents who have complete data from in-school Wave 1 questionnaires, Wave 1 and Wave 2 school administrator interviews, in-home Wave 1 questionnaires, in-home Wave 1 parental questionnaires, in-home Wave 3 questionnaires, and in-home Wave 4 questionnaires (n = 9,006 for Wave 3; n = 7,361 for Wave 4). The Wave 1 data were collected in 1994-1995, Wave 2 data in 1996, Wave 3 in 2001-2002, and Wave 4 data in 2007-2008.

Dependent Variables

To assess political and civic participation as adults, we look at voting and volunteering behaviors at two different time periods, using data from both Wave 3 (collected in 2001-2002) and Wave 4 (collected in 2008-2009); our inclusion of measures of both voting and volunteering follows precedent in prior research, particularly McFarland and Thomas (2006). The Wave 3 data are collected when some of the respondents are just old enough to vote, as they are aged 18 to 26 years at that time; thus, these data allow us to view early adult behaviors. The Wave 4 data offer a different view, as the respondents are aged 25 to 33 years and no longer going through the final stages of transition to adulthood.

There are two Wave 4 variables that relate to our research question. Each comes from an ordinal-scaled question. One asks “How often do you usually vote in local or statewide elections?” and is coded from 1 (never) to 4 (always). The second asks “In the past 12 months, about how many hours did you spend on volunteer or community service work?” and is coded from 1 (0 hours) to 6 (160 hours or more).

We selected Wave 3 variables that mirror the available Wave 4 variables; thus, we have three dependent variables from the Wave 3 data. Each of them is dichotomous, where a value of 1 indicates a response of “yes” (“no” responses = 0) to each of the following questions: “Are you registered to vote?” “Did you vote in the most recent presidential election?” and “During the last 12 months did you perform any unpaid volunteer or community service work?”

Independent Variables

To test our hypotheses at both the individual and school levels, we include variables for school security and discipline measured at each level of aggregation. Our individual-level variable is dichotomous, indicating whether each respondent had ever been suspended from school (by Wave 1).

We include several school-level variables that indicate schools’ security policies and punishment responses to different misbehaviors. The following dichotomous variables are derived from the Wave 1 school administrator surveys (variable names are in parentheses):

Whether students in any grade may not leave school grounds (closed campus);

Whether students in any grade must obey a dress code (dress code);

Whether a student who is caught on a first offense of cheating is suspended or expelled (cheating punishment);

Whether a student who is caught on a first offense of fighting is suspended or expelled (fighting punishment);

Whether a student who is caught on a first offense of “verbally abusing a teacher” is suspended or expelled (verbal punishment);

Whether a student who is caught on a first offense of smoking is suspended or expelled (smoking punishment).

We also include several indicators from the Wave 2 school administrator survey:

Whether there is a security officer or police officer on duty during school hours (officer);

Whether students walk through metal detectors as they enter the building (metal detectors);

Whether the school has surveillance cameras (surveillance);

Whether students are prohibited from wearing “certain colors,” or whether “bandanas or other gang paraphernalia” are prohibited (antigang rules).

Because our research questions include consideration of racial disparities in the effects of punishment, we include both main effects and interaction terms for race/ethnicity. The main effects are a series of dichotomous variables indicating a respondent’s self-identified racial/ethnic group, with categories for Hispanic, Black, American Indian, Asian, and Other (White youth are excluded as a contrast category). We include two interaction terms as well: Black respondent × Ever suspended, and Hispanic respondent × Ever suspended. These variables test whether the effects of suspension on future civic participation differ for Black and Hispanic youth compared with other youth; we use only these two racial/ethnic categories because they include the youth who have been found in prior research to suffer most from disproportionate school discipline.

One of the most significant challenges to our analyses is the need to factor out the underlying propensity of students to participate in civic life, regardless of school discipline and security. To reduce the potential influence of such confounding factors, we include many independent variables that control for factors found by prior research to shape civic participation. These include the respondent’s age at Wave 1 interview, the respondent’s sex (coded female = 1), whether the respondent’s primary language is not English, parental education level (coded for the highest level either parent reached, with 1 = less than high school and 5 = graduate education), respondent’s grades (measured as the mean, scored where 1 = A and 4 = F, of English, history/social studies, math, and science), whether the respondent does not live with his or her mother, whether the respondent does not live with his or her father, and whether the respondent expresses an interest in going to college. The models include type of area (variables for suburban and rural, with a contrast of urban), the school’s average attendance (measured ordinally, from 1 = 95% or more to 5 = 75%-79%), the school’s average class size, and whether it is a public school. We include a measure of how often a respondent attended religious services in the last 12 months, ranging from 1 (never) to 4 (once a week or more). We also include a measure of the respondents’ parents’ civic participation, measured as the sum of the following activities in which his or her parents report participating: parent/teacher organization, military veterans’ organization, labor union, sports/bowling team, and civic or other social organization.

We compute indices intended to control for characteristics and perceptions of respondents that may shape their community involvement; each index was formed after exploratory factor analysis, and each has moderate to high reliability (Cronbach’s alpha values are reported below). One measures the extent to which each respondent sees his or her school as a community; this has a Cronbach’s alpha of .7618, and is the mean (from 1 = strongly agree to 5 = strongly disagree) of responses about whether the respondent feels close to people at the school, feels like a part of the school, is happy to be at his or her school, and feels safe at his or her school. We also include respondents’ answers to whether they feel that teachers at the school “treat students fairly,” recoded so that 1 = strongly disagree and 5 = strongly agree.

Another variable measures autonomy from respondents’ parents, using the sum of the number of the following the respondents report being able to decide on: a weekend curfew, who to “hang around with,” what to wear, how much TV to watch, which TV programs to watch, when to go to bed on week nights, and what to eat (Cronbach’s α = .9431). Following McFarland and Thomas (2006), we include a variable measuring the range of discussions respondents have with their parents, measured as the sum of the following topics each respondent reports discussing with either his or her mother or father (each measured separately): Someone he or she is dating or a party attended, personal problems, and school work or grades. Another index measures low self-esteem by taking the mean response to several statements about respondents’ feelings toward themselves where 1 = strongly agree and 5 = strongly disagree: “you have a lot of good qualities,” “you are physically fit,” “you have a lot to be proud of,” “you like yourself just the way you are,” “you feel like you are doing everything just about right,” “you feel socially accepted,” and “you feel loved and wanted” (Cronbach’s α = .8474).

Furthermore, we created an index that sums the number of statements indicating neighborhood bonds to which respondents agreed (Cronbach’s α = .6017): knowing most people in the neighborhood, stopping to talk on the street with a neighbor, that people in the neighborhood look out for each other, using a recreation center in the neighborhood, feeling safe there, being happy living there, and being unhappy if he or she had to leave the neighborhood. In earlier analyses, we included a variable for whether the respondent’s parent reports that the respondent has ever been diagnosed with a learning disability or is in special education classes. This variable caused a problem with multicollinearity with our variable for suspension, which relates directly to our hypotheses. As a result, we remove it from the current analyses, though the connection between learning disabilities, suspension, and civic participation remains important to consider in future work.

We also use several variables to control for respondent drug use and delinquency to account for the fact that students who are suspended may be different than other students in ways that are likely to also affect their future democratic participation. These variables include the natural logarithm of the number of times the respondent reports that he or she has used each of the following variables (with a different variable for each substance): marijuana, cocaine, inhalants, and other drugs. We created a delinquency index, computed as the mean ordinal responses (along a scale of 0 = never to 3 = 5 or more times) indicating the frequency of respondents committing each of 14 different offenses and misbehaviors over the past 12 months (Cronbach’s α = .8314): graffiti, damage to property, lying to parents, theft from a store, fighting, injuring someone badly, car theft, theft (more than US$50), burglary, threat with a weapon, selling drugs, petty theft (less than US$50), group fight, and creating a public disturbance.

Following the primary results of McFarland and Thomas (2006), we include a series of dichotomous variables indicating respondents’ membership in varying student activities: honors society, student council, future farmers of America, performing arts, news or yearbook, academic clubs, sports teams, and other clubs.

One complication to our analysis is the potential mediating effect of incarceration: Students who are suspended in school are at elevated risk of future incarceration (Fabelo et al., 2011), and those who are incarcerated are often disenfranchised, unable to vote years after their incarceration (Manza & Uggen, 2008). To better measure the direct effect of school discipline on future democratic participation, we control for incarceration. For Wave 3 analyses, we include a Wave 3 variable measuring whether respondents had ever been incarcerated based on an adult conviction; for Wave 4 analyses, we use the most similar variable, which considers whether respondents had been incarcerated for at least a year after age 18. We present descriptive statistics for all variables in Table 1.

Table 1.

Descriptive Statistics for Variables in Multivariate Models.

Variable	n	M	SD	%	Minimum	Maximum
Voted	9,622			35.25	0	1
Registered to vote	9,622			57.59	0	1
Volunteered	9,622			23.62	0	1
Voting frequency	7,820	2.37	1.17		1	4
Volunteer frequency	7,816	1.66	1.13		1	6
Ever suspended	9,612			25.31	0	1
Verbal punishment	9,622			39.72	0	1
Cheating punishment	9,622			2.94	0	1
Fighting punishment	9,622			71.94	0	1
Smoking punishment	9,622			45.17	0	1
Dress code	9,622			11.32	0	1
Closed campus	9,622			9.56	0	1
Officer	9,622			44.61	0	1
Metal detectors	9,622			22.71	0	1
Surveillance	9,622			8.97	0	1
Antigang rules	9,622			91.52	0	1
Age	9,617	14.83	1.57		11	20
Female	9,622			51.96	0	1
Foreign language	9,622			11.26	0	1
Hispanic	9,622			16.25	0	1
Black	9,622			22.19	0	1
American Indian	9,622			2.48	0	1
Asian American	9,622			6.80	0	1
Other race/ethnicity	9,622			4.65	0	1
Does not live with mother	9,622			4.28	0	1
Does not live with father	9,622			28.00	0	1
Parent education level	9,622	2.82	1.35		0	5
Grades	9,430	2.20	0.76		1	4
School community	9,522	2.23	0.78		1	5
Teacher fairness	9,520	3.50	1.06		1	5
Marijuana use (ln)	9,377	0.53	1.17		0	6.86
Cocaine use (ln)	9,529	0.03	0.28		0	6.55
Inhalant use (ln)	9,527	0.09	0.42		0	6.40
Other drug use (ln)	9,500	0.13	0.60		0	6.69
Delinquency scale	9,583	0.29	0.36		0	3
Wants college	9,591	4.49	0.98		1	5
Autonomy from parents	9,622	4.99	1.62		0	7
Discussions with parents	9,622	2.13	1.52		0	6
Low self-esteem	9,604	1.92	0.59		1	5
Neighborhood bonds	9,622	5.46	1.44		0	7
Parents’ civic participation	9,622	0.68	0.90		0	5
Honor society	9,622			9.45	0	1
Student council	9,622			8.06	0	1
Future farmers	9,622			2.11	0	1
Performing arts	9,622			26.76	0	1
News/yearbook	9,622			10.58	0	1
Academic clubs	9,622			20.12	0	1
Sports teams	9,622			58.69	0	1
Other clubs	9,622			17.16	0	1
Religious service attendance	9,461	2.69	1.39		0	4
Public school	9,622			83.78	0	1
Suburban	9,622			50.67	0	1
Rural	9,622			19.73	0	1
Average attendance	9,622	1.98	0.89		1	5
Average class size	9,622	26.57	5.82		10	38
Prior incarceration—Wave 3	7,619			0.60	0	1
Prior incarceration—Wave 4	7,858			1.90	0	1

A second complication is the possibility that suspension is a mediating mechanism rather than a direct contributor to future political and civic participation. That is, students who are more deviant than others are at greater risk of suspension, and it is possible that suspension mediates the effect of deviance on participation. In other words, delinquent behavior triggers suspension, which, in turn, may contribute to political and civic disengagement. This poses not only an important methodological concern but also an alternative conceptual basis for empirical inquiry with potentially distinctive policy implications.

Though the mediating effects of suspension are not a concern raised by the prior research that guides our theoretical framework and research questions here, it is clearly an important methodological issue for our analyses. To directly address this issue, we perform several robustness checks to supplement our analyses, including propensity score modeling, structural equation modeling (SEM), and a Baron and Kenny (1986) mediation test. The results of each of these, which we describe briefly below, broadly support our modeling strategy by finding that suspension is not a consistent mediator of deviance’s effect on suspension, nor is the effect of suspension due to measured differences among youth who are suspended versus not suspended. As such, we conclude below that evidence of suppressive effects of suspension on future political and civic engagement of youth—particularly when combined with myriad other negative effects of suspension—advise against the use of suspension and in favor of alternative disciplinary tools when addressing delinquent or deviant school-place behavior.

Analytic Strategy

To analyze the data, we compute a series of multilevel models, as is appropriate and commonly done when analyzing data at multiple units of analysis (here, data on students nested within schools; see Rabe-Hesketh & Skrondal, 2008; Raudenbush & Bryk, 2002). All analyses are done using Stata 12.1, using the provided wave-specific sample weights. Because the Wave 3 dependent variables are dichotomous, we use random-intercept logistic regression models to predict whether respondents are registered to vote, voted in the previous election, or volunteered recently, as measured in Wave 3. The Wave 4 variables are measured differently, along ordinal scales. To accommodate this level of measurement, we used random-intercept ordinal logistic models, using Stata’s gllamm command with an ologit link. Each model estimates a random intercept for each sampled school. Wave 3 outcomes are reported in Table 2 and Wave 4 outcomes in Table 3. For each outcome, we compute models with and without interaction terms; the interaction terms indicate whether the impact of school suspension significantly differs for Black or Hispanic youth relative to others. All results listed in Tables 2 and 3 refer to the log odds of each outcome. Negative results suggest that an independent variable is associated with lower likelihood of voting, volunteering, and so on, whereas positive results suggest greater likelihood of each outcome.

Table 2.

Random-Intercept Logistic Regression of Wave 3 Civic Participation on School Discipline and Security Indicators and Control Variables (n = 9,006), Log Odds Reported.

	Voted		Registered to vote		Volunteered
		Full model		Full model		Full model
Ever suspended	−0.13*	−0.05	−0.09	−0.06	−0.19*	−0.20
Verbal punishment	−0.10	−0.10	−0.10	−0.10	−0.09	−0.09
Cheating punishment	−0.42*	−0.42*	−0.23	−0.23	−0.29	−0.29
Fighting punishment	0.02	0.02	0.08	0.08	0.10	0.09
Smoking punishment	0.07	0.07	0.02	0.01	0.17*	0.17*
Dress code	−0.06	−0.06	−0.17	−0.17	−0.13	−0.13
Closed campus	0.24	0.24	0.03	0.03	0.22	0.23
Officer	−0.03	−0.03	−0.07	−0.07	0.05	0.05
Metal detectors	0.12	0.12	0.10	0.10	−0.13	−0.13
Surveillance	0.10	0.10	0.10	0.10	−0.12	−0.12
Antigang rules	0.01	0.01	0.03	0.02	−0.08	−0.08
Black × Suspended		−0.19		−0.22		−0.07
Hispanic × Suspended		−0.05		0.19		0.22
Age	0.13***	0.13***	0.09***	0.09***	−0.03	−0.03
Female	0.05	0.06	0.09	0.09	0.03	0.03
Foreign language	−0.53***	−0.53***	−0.56***	−0.56***	0.06	0.06
Hispanic	0.12	0.12	0.07	0.01	−0.05	−0.09
Black	0.28***	0.34***	0.19**	0.27**	0.00	0.03
American Indian	0.21	0.20	0.28	0.28	0.25	0.26
Asian American	−0.53***	−0.53***	−0.31**	−0.31**	0.01	0.01
Other race/ethnicity	−0.30*	−0.30*	−0.40**	−0.40**	−0.27	−0.27
Does not live with mother	−0.13	−0.13	−0.10	−0.11	−0.05	−0.05
Does not live with father	−0.05	−0.05	−0.06	−0.06	−0.01	−0.01
Parent education level	0.14***	0.14***	0.09***	0.09***	0.18***	0.18***
Grades	−0.20***	−0.20***	−0.14***	−0.12***	−0.43***	−0.43***
School community	0.02	0.02	−0.00	−0.01	0.01	0.01
Teacher fairness	0.02	0.02	−0.02	−0.02	−0.03	−0.03
Marijuana use (ln)	0.03	0.03	0.04	0.04	−0.06	−0.06
Cocaine use (ln)	−0.04	−0.04	0.01	−0.00	−0.43	−0.43
Inhalant use (ln)	0.03	0.03	0.02	0.02	0.03	0.02
Other drug use (ln)	−0.05	−0.06	−0.00	−0.01	0.00	−0.00
Delinquency scale	−0.10	−0.10	0.01	0.01	−0.02	−0.02
Wants college	0.10**	0.10**	0.09***	0.09***	0.07	0.07
Autonomy from parents	−0.01	−0.01	−0.01	−0.01	−0.01	−0.01
Discussions with parents	0.03*	0.03*	0.02	0.02	0.03	0.03
Low self-esteem	−0.04	−0.04	−0.06	−0.06	−0.15**	−0.15**
Neighborhood bonds	0.06**	0.06**	0.05**	0.05**	0.04	0.04
Parents’ civic participation	0.09**	0.09**	0.11***	0.11***	0.11***	0.11***
Honor society	0.20*	0.20*	0.30**	0.30**	0.20*	0.20*
Student council	0.01	0.01	0.13	0.13	0.09	0.09
Future farmers	0.01	0.01	−0.14	−0.13	−0.08	−0.08
Performing arts	0.18**	0.18	0.15**	0.15**	0.32***	0.32***
News/yearbook	−0.00	−0.00	−0.04	−0.04	0.08	0.07
Academic clubs	0.07	0.07	0.05	0.05	−0.12	−0.11
Sports teams	−0.03	−0.03	−0.01	−0.01	0.19**	0.19**
Other clubs	0.16**	0.17**	0.13	0.13	0.23***	0.23**
Religious service attendance	0.11***	0.11***	0.09***	0.09***	0.09***	0.09***
Public school	−0.30**	−0.30**	−0.11	−0.10	−0.10	−0.10
Suburban	0.10	0.11	0.11	0.11	−0.07	−0.07
Rural	0.17	0.17	0.30*	0.30*	−0.00	−0.00
Average attendance	−0.07	−0.07	−0.06	−0.05	−0.10	−0.09
Average class size	0.02*	0.02*	0.01	0.01	−0.00	−0.00
Prior incarceration	−0.72	−0.71	−0.22	−0.21	0.71	0.71
Constant	−3.93***	−3.95***	−1.87***	−1.90***	−0.78	−0.79
Random intercept (SD)	0.21	0.21	0.24	0.24	0.15	0.15
Log likelihood	−5,513.252	−5,512.109	−5,812.217	−5,808.935	−4,542.469	−4,541.556

p < .05. **p < .01. ***p < .001.

Table 3.

Random-Intercept Ordinal Logistic Regression of Wave 4 Civic Participation on School Discipline and Security Indicators and Control Variables, Log Odds Reported (n = 7,361).

	Voting frequency		Volunteer frequency
		Full model		Full model
Ever suspended	−0.11	−0.21*	0.06	−0.04
Verbal punishment	−0.11	−0.11	−0.18*	−0.18*
Cheating punishment	0.14	0.14	0.15	0.15
Fighting punishment	0.06	0.06	−0.06	−0.06
Smoking punishment	0.13	0.13	−0.00	0.00
Dress code	−0.12	−0.12	0.06	0.05
Closed campus	0.18	0.19	−0.12	−0.11
Officer	−0.17	−0.17	0.17	0.17
Metal detectors	0.34**	0.34**	0.00	−0.00
Surveillance	0.28*	0.27*	0.10	0.10
Antigang rules	−0.27*	−0.27*	−0.05	−0.05
Black × Suspended		0.09		0.15
Hispanic × Suspended		0.40**		0.30
Age	0.09***	0.09***	0.02	0.02
Female	0.13**	0.13**	0.08	0.08
Foreign language	−0.34**	−0.34**	−0.29*	−0.30*
Hispanic	0.02	−0.08	−0.32**	−0.38***
Black	0.51***	0.50***	−0.18*	−0.21*
American Indian	0.17	0.18	0.00	0.01
Asian American	−0.70***	−0.70***	−0.18	−0.18
Other race/ethnicity	−0.18	−0.18	0.14	0.14
Does not live with mother	−0.04	−0.05	0.13	0.13
Does not live with father	0.01	0.01	−0.03	−0.03
Parent education level	0.17***	0.17***	0.10***	0.10***
Grades	−0.15***	−0.14***	−0.28***	−0.28***
School community	0.02	0.02	−0.05	−0.05
Teacher fairness	−0.01	−0.01	−0.01	−0.01
Marijuana use (ln)	0.01	0.01	−0.01	−0.00
Cocaine use (ln)	0.02	0.01	−0.17	−0.16
Inhalant use (ln)	−0.01	−0.01	0.01	0.01
Other drug use (ln)	0.07	0.07	0.05	0.05
Delinquency scale	−0.02	−0.03	0.05	0.04
Wants college	0.16***	0.16***	0.07*	0.07*
Autonomy from parents	0.02	0.02	0.01	0.01
Discussions with parents	0.03*	0.03*	0.08***	0.08***
Low self-esteem	−0.08	−0.08	−0.03	−0.03
Neighborhood bonds	0.03	0.03	0.04*	0.04*
Parents’ civic participation	0.10***	0.10***	0.10***	0.10***
Honor society	0.17*	0.17*	0.27**	0.27**
Student council	−0.01	−0.01	0.24**	0.24**
Future farmers	0.10	0.09	0.08	0.08
Performing arts	0.18**	0.18**	0.19**	0.19**
News/yearbook	0.11	0.11	−0.03	−0.03
Academic clubs	0.04	0.04	0.06	0.06
Sports teams	0.02	0.02	0.08	0.08
Other clubs	0.15*	0.14*	0.30***	0.29***
Religious service attendance	0.12***	0.12***	0.13***	0.13***
Public school	−0.25*	−0.25*	−0.04	−0.04
Suburban	0.01	0.01	−0.04	−0.04
Rural	0.13	0.13	−0.13	−0.13
Average attendance	0.01	0.01	−0.07	−0.07
Average class size	0.01	0.01	−0.00	−0.00
Prior incarceration	−1.30***	−1.30***	−0.60*	−0.60*
Constant 1	2.05***	2.03***	1.24**	1.23**
Constant 2	3.28***	3.26***	2.67***	2.66***
Constant 3	4.14***	4.12***	3.28***	3.26***
Constant 4			3.93***	3.92***
Constant 5			4.61***	4.59***
Random intercept (variance)	0.07	0.07	0.05	0.05
Log likelihood	−9,526.410	−9,523.061	−7,670.389	−7,668.941

p < .05. **p < .01. ***p < .001.

Results

Beginning with Wave 3 outcomes, shown in Table 2, we see that being suspended in school decreases the log odds of respondents having voted or having volunteered while a young adult; expressed otherwise, the odds of a student voting years later are 12% lower among those suspended, and the odds of a student volunteering are 18% lower. In Figure 1, we show the predicted probability of each of these three outcomes as a function of suspension while holding other variables constant; here the black bars represent the predicted probability for youth who were not suspended, overall, and the gray bars for youth who were suspended. These results are modest, as they do not remain once we add interaction terms for Suspension × Black, or Suspension × Hispanic; some reduction in strength of effect is to be expected, due to the inevitable multicollinearity that comes along with interaction terms. With regard to the school-level measures of security and discipline, we see that schools’ punitive responses to cheating reduces the likelihood of having voted, but that punitive responses to smoking increases the likelihood of volunteering. No other measure of school discipline or security shapes future civic participation based on these Wave 3 outcome measures.

Figure 1.

Estimated probability of Wave 3 outcomes, by suspension and race/ethnicity.

The results for race and ethnicity suggest that Black young adults have higher log odds of voting and being registered to vote. However, as shown by the interaction terms, the effect of suspension does not significantly vary for Black respondents or Hispanic respondents, relative to its effect on others. Figure 1 also illustrates the predicted probability of each Wave 3 outcome specifically for Black youth and for Hispanic youth who have been suspended. Though there is variation across these predicted probabilities, the differences are not statistically significant.

The results for the control variables mirror results from prior research. In sum, we find that respondents whose parents are more educated, whose parents participate in civic activities, who attain high grades in school, who speak English as their primary language, who desire to go to college, who have high neighborhood bonds, who participate in honor society, performing arts, and other school clubs, who attend religious services frequently, and who are older have higher log odds of at least two of the measured civic participation outcomes. For the sake of brevity, given our long list of control variables, we refrain from a full discussion of the results for these variables. All results are listed in Tables 2 and 3.

Table 3 lists the results for the regressions of Wave 4 outcomes. Again, being suspended has a long-term effect on voting, though not on volunteering. Here we see an effect that approaches but does not reach statistical significance (p = .064) in the first model predicting voting frequency, and a statistically significant effect once we introduce the interaction terms, where the odds of a respondent being one level higher on the ordinal voting frequency measure are 18.6% lower if he or she was suspended. We also find some unexpected results for the school-level indicators of security and discipline. Rather than having a suppressive effect, as expected, the presence of metal detectors and surveillance cameras is associated with greater log odds of voting frequently. The presence of antigang rules is negatively related to the log odds of voting frequently.

Regarding race and ethnicity, Black and Hispanic respondents are less likely to volunteer frequently than Whites, and Black respondents are more likely to vote frequently than Whites. We find that the interaction between Hispanic and being suspended is positive for voting frequency, meaning that Hispanic respondents who were suspended as youth are more likely than others to vote frequently, which contradicts our hypotheses.

Results for our control variables are similar to our results in Wave 3, with several expected relationships observed. Respondents are more likely to vote and volunteer frequently if they speak English as their primary language, have parents with high levels of education and who participate in civic activities, attain high grades, want to go to college, discuss a wide range of topics with their parents, are members of honor society, performing arts clubs, or other clubs, and attend religious services frequently.

Taken together, the results from our analyses at both waves display a pattern that supports one of our hypotheses, but not others. We find that youth who are suspended at school have lower odds of future civic and political participation while controlling for several alternative explanations. The results thus suggest that firsthand experiences with school discipline—being suspended—have a suppressive effect on future civic and political participation. We observe these results in both waves, suggesting that the effect lasts beyond the young adult years and can shape long-term behaviors well into adulthood. Though the effect of suspension is not significant in a majority of our models, the fact that we do find this effect in multiple models illustrates an important influence, as predicted by the literature, and suggests the need for further analyses. At the same time, we do not find the expected relationships for school-level security and discipline. Here there are few significant results and some that run contrary to our expectations. We also find no evidence that the effect of suspension varies significantly by race or ethnicity.

Robustness Check

As we state above, we extended our analyses with a series of tests to confirm the robustness of our results. Our concern was that the primary causal factor was actually respondents’ deviance, with suspension acting as a mediator between deviance and future political and civic participation. To consider this possibility, we first performed propensity score matching analysis to determine whether differences between suspended youth and not suspended youth (including deviance) affect our results. Using caliper matching (caliper = 0.01) with Stata’s psmatch2 command, with all independent variables other than suspension as predictors of the propensity for suspension, we found significant differences in voting, volunteering, and volunteer frequency across matched pairs, with suspended youth less likely to participate in each activity, which confirms our regression results. We further probed this possibility through the use of SEM¹ and Baron and Kenny (1986) mediation tests. Here we used five measures of youth deviance—delinquency, marijuana use, cocaine use, inhalant use, and other drug use—in exploring whether their effects on future civic participation are mediated by suspension. These analyses (results available on request) demonstrated that in our five models, there are only 2 (of a possible 25, given the five independent variables in each of five models) observed mediating effects, which is approximately what one would expect from chance alone using an α = .05. We find that the effect of delinquency on voting is slightly mediated by suspension (indirect effect = –.01) and that the effect of marijuana use on volunteering is slightly mediated by suspension (indirect effect = −0.003). The Baron and Kenny test found that the proportions of the effects of delinquency on voting and of marijuana use on volunteering that are due to suspension are relatively small (.184 and .122, respectively).

Finally, we computed additional structural equation models to determine whether bonds to teachers, participation in extracurricular activities, delinquent activity, and subsequent incarceration mediate the relationship between suspension and future civic and political participation. Again we found no evidence of substantial mediation. The direct effect of suspension on each dependent variable remained statistically significant in these models despite the presence of potential mediators. Again this robustness test confirms and corroborates our regression models.

Discussion

By finding that a history of suspension is related to decreased odds of future civic participation, our research extends previous findings in a new direction. The education research literature is clear that overreliance on suspension is an ineffective, counterproductive practice in terms of enhancing school safety and improving important educational and related outcomes. Not only does it fail to advance these stated ambitions but suspension is also associated with a range of negative outcomes, such as higher rates of dropping out and diminished academic achievement (Fabelo et al., 2011; Skiba et al., 2006). It also fails to reduce the likelihood of future disciplinary action (Way, 2011). Yet to our knowledge, no prior research has considered how suspension shapes students’ future civic participation. Thus, though we do not find a suppressive effect of suspension in all of our models, these results still add an important and previously overlooked element to the literature by uncovering an additional negative consequence of school suspension, observed at multiple points in time: both during young adult years and several years later.

Following prior research, we theorize that the observed negative effect of suspension is because suspension short-circuits dialogue and student involvement; it removes a student from the school rather than responding constructively and therapeutically to problematic behavior. Research on suspension finds that it is administered in ways that alienate students from the school and from the school’s authority structure, leading them to view school staff as unfair, arbitrary, and uncaring (Kupchik, 2010; Lyons & Drew, 2006). We interpret the results of our analysis to suggest that this practice teaches students a lesson about authority and their powerlessness relative to governing bodies. To the extent that students learn this lesson and apply it to their future roles as citizens, they may be less likely to vote and volunteer because they see little opportunity to actively shape governance or community life. School punishment, thus, may socialize students into cynicism, disengagement, and apathy.

Our results are unsurprising because prior research finds that (a) civic participation is taught through inclusive educational climates that encourage participation and (b) suspension tends to reduce student participation and contribute to nonparticipatory, noninclusive school climates. Though unsurprising and modest, these results are important, for this is the first empirical effort of which we are aware to test the long-term effects of school suspension on civic and political participation. Indeed, our results discouragingly suggest that schools’ recent shift toward vigorous enforcement of harsh discipline may be detrimental to the nation’s long-term civic and political health. Our analyses strengthen existing calls to reduce school’s reliance on suspension, and invest instead on evidence-based practices such as positive behavioral supports, inclusive social climates, and behavioral counseling in schools (see Losen & Martinez, 2013).

Contrary to our expectations, however, our study provides little evidence regarding school-level effects of security and discipline on future civic participation. That is, attending a school with police or security officers, metal detectors, harsh punishment policies, and other criminal justice–oriented practices has little to no effect on the likelihood of voting and volunteering in the future. On one hand, it may be the case that students are relatively unfazed by school discipline and security efforts. Recent research finds that many students appreciate having rigid security and tough punishment policies (Kupchik, 2010). This appreciation may mean that these policies have little long-term effect on the behaviors of most students and that their individual experiences with school discipline are what matters instead. On the other hand, the lack of results may be due to limitations in our measures of school security. The variables to which we are limited are somewhat vague, which may hide actual long-term suppressive effects of security on civic and political participation. For example, despite drastic differences between security guards (who are employed by and report to schools, have no arrest power, and usually do not carry weapons) and police officers stationed in schools, the two are measured together by a single question in the Add Health interviews. Future research should use more specific and careful measurements of school security to better test whether school-level discipline and security shape long-term civic participation.

We also find it surprising that the effects of school suspension are not experienced more acutely among Black and Hispanic respondents, as indicated by nonsignificance of our interaction terms. It is important to keep in mind that despite this result, Black and Hispanic youth do still suffer the negative consequences of school suspension at disproportionately high rates. We find school suspension to have an overall negative effect on civic and political participation, even while controlling for race/ethnicity; because Black and Hispanic youth are far more likely than White youth to be suspended, they bear the brunt of this overall negative effect far more often than do White students. In other words, the effects of school suspension are felt by all youth at somewhat equal intensity, though Black and Hispanic youth are far more often exposed to this effect.

Despite the importance of our findings, there are a number of limitations to our analyses that should be addressed by future research. Above we refer to the vague measurements of school-level discipline and security. Another data limitation is that we are predicting a very narrow range of political and civic participation variables: voting, registering to vote, and volunteering. Future analyses that consider other types of civic and political participation, such as participating in social and professional networks, being engaged in political life (e.g., going to political demonstrations, donating money to political causes, etc.), and building social networks with fellow members of one’s community more broadly, would greatly enhance our understanding of the long-term ramifications of school discipline and security. Analyses, in particular, should build from the literature on volunteering (Wilson, 2000) to explore the relationship between specific school-place activities and various forms of volunteering to see how consistent the effects of suspension and other disciplinary policies are. Furthermore, by considering only conventional forms of political engagement such as voting, our analyses are unable to measure less formal methods of political engagement and expressions of social capital that are more commonly found in marginalized communities, such as low-income African American and Hispanic communities (see Suttles, 1968).

A final data limitation is that our measures of school discipline and security come from Wave 1 and Wave 2 data, collected during 1994-1996, early in the chronology of the buildup of school discipline and security. Although we have sufficient variation in students’ experiences to model their effects, we are mindful of the fact that results may be somewhat different if measures were collected in 2014, when harsh punishments, rigid rules, and criminal justice–oriented security measures are more commonplace across the United States and in light of the potentially unique patterns of political and civic behavior among the “DotNet” generation (Zukin et al., 2006). These limitations, though, are fairly minor; though they may reduce the clarity of our results, the fact that we have such findings using a large, nationally representative, longitudinal database leaves us confident in our results and conclusions.

This study also raises important questions for future research. First, though the prior research suggests such a direct effect, it is necessary to continue to probe these results by considering other mediating effects that we are not able to include here. In particular, qualitative research strategies would be especially helpful for uncovering how former students who had experienced school punishment perceive and approach civic engagement. Second, notwithstanding the complex etiology of deviance and delinquency (Loeber, Burke, & Pardini, 2009) and the many salient structural factors of school context that may contribute to both delinquency and political and civic behavior (Anderson, 1982; Ehman, 1980; Zimmerman & Rees, 2014), it is important for future research to focus on the factors that precipitate suspension and how these factors shape future political and civic participation. Though we explore the possibility that deviance shapes participation indirectly through suspension (and do not find compelling evidence that this is the case), our ability to test for such effects is limited by the available data. Further exploration into this question would help to flesh out the varying ways in which school punishment and climate can shape students’ future roles as citizens.

In sum, the results of our analyses make an important and substantive contribution to the literatures on school discipline and on civic and political participation, despite the fact that our hypotheses for racial/ethnic interaction and school-level effects are not supported. In the first empirical test of the long-term effects of suspension on civic participation, we find that being suspended is associated with reduced odds of voting both in young adult years and beyond, and on volunteering while a young adult. When joined with existing evidence of the negative or null effects of suspension on youth, this research counsels against the use of suspension and in favor of alternative disciplinary tools when addressing delinquent and deviant school-place behavior. More fundamentally, this study adds political and civic participation to the agenda of issues scholars, policy makers, and communities should consider as they debate how best to govern schools today.

Footnotes

Acknowledgements

This research uses data from Add Health, a program project designed by J. Richard Udry, Peter S. Bearman, and Kathleen Mullan Harris, and funded by a grant P01-HD31921 from the Eunice Kennedy Shriver National Institute of Child Health and Human Development, with cooperative funding from 17 other agencies. Special acknowledgment is due to Ronald R. Rindfuss and Barbara Entwisle for assistance in the original design. Persons interested in obtaining Data Files from Add Health should contact Add Health, The University of North Carolina at Chapel Hill, Carolina Population Center, 123 W. Franklin Stree, Chapel Hill, NC 27516-2524 (addhealth@unc.edu). No direct support was received from grant P01-HD31921 for this analysis.

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) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported, in part, by a National Research Foundation of Korea Grant funded by the Korean Government (MEST; NRF-2011-220–B00027) and Arizona State University’s New American University Exemplar Fund.

Notes

Author Biographies

Aaron Kupchik is a Professor and the Director of Graduate Studies in the Department of Sociology and Criminal Justice at the University of Delaware. He is author of Homeroom Security: School discipline in an age of fear (NYU Press, 2010) and Judging Juveniles: Prosecuting adolescents in adult and juvenile courts (NYU Press, 2006).

Thomas J. Catlaw is Associate Professor and Frank and June Sackton Chair in Public Administration in the School of Public Affairs at Arizona State University in Phoenix. He is the author of Fabricating the People: Politics and administration in the biopolitical state (University of Alabama Press, 2007) and Theories of Public Organization (with Robert Denhardt, Cengage, 2014).

References

American Psychological Association Zero Tolerance Task Force (2008). Are zero tolerance policies effective in the schools? An evidentiary review and recommendations. American Psychologist, 63, 852-862

Anderson

C. S.

(1982). The search for school climate: A review of the research. Review of Educational Research, 52, 368-420.

Angell

A. V.

(1998). Practicing democracy at school: A qualitative analysis of an elementary class council. Theory and Research in Social Education, 26(2), 149-172.

Apple

M. W.

Beane

J. A.

(1999). Democratic schools: Lessons from the chalk face. Buckingham, UK: Open University Press.

Arum

(2003). Judging school discipline: The crisis of moral authority. Cambridge, MA: Harvard University Press.

Baron

R. M.

Kenny

D. A.

(1986). The moderator-mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology, 51, 1173-1182.

Battisoni

Berman

S. H.

Billig

S. H.

Both

Fickling Brainard

Brown

. . . Youniss

(2003). The civic mission of schools. New York: Carnegie Corporation of New York/CIRCLE.

Bevir

(2006). Democratic governance: Systems and radical perspectives. Public Administration Review, 66, 426-436.

Bourdieu

(1986). The forms of capital. In Richardson

J. G.

(Ed.), Handbook of theory and research for the sociology of education (pp. 241-258). New York, NY: Greenwood Press.

10.

Buss

T. F.

Redburn

F. S.

Guo

(Eds.). (2006). Modernizing democracy: Innovations in citizen participation. Armonk, NY: M.E. Sharpe.

11.

Campbell

D. E.

(2006). Why we vote: How schools and communities shape our civic life. Princeton, NJ: Princeton University Press.

12.

Casella

(2001). “Being down”: Challenging violence in urban schools. New York, NY: Teachers College Press.

13.

Catlaw

T. J.

Sandberg

(2012). “Dangerous government”: Info-liberalism, active citizenship, and the Open Government Directive. Administration & Society, 6(3), 223-254.

14.

Cornell

D. G.

(2006). School violence: Fear versus facts. New York, NY: Taylor & Francis.

15.

Creighton

J. L.

(2005). The public participation handbook: Making better decisions through citizen involvement. San Francisco, CA: Jossey Bass.

16.

Dewey

(1916). Democracy and education. New York, NY: Macmillan.

17.

Dewey

(1959). Moral principles in education. New York, NY: Philosophical Library. (Original work published 1909)

18.

Dinkes

Kemp

Baum

(2009). Indicators of school crime and safety: 2009 (NCES 2010–012/NCJ 228478). Washington, DC: National Center for Education Statistics, Institute of Education Sciences, U.S. Department of Education, and Bureau of Justice Statistics, Office of Justice Programs, U.S. Department of Justice.

19.

Dobozy

(2007). Effective learning of civic skills: Democratic schools succeed in nurturing capacities of students. Education Studies, 33, 115-128.

20.

Durkheim

(1961). Moral education: A study in the theory and application of the sociology of education ( Schnurer

Wilson

E. K.

Trans.). Glenco, IL: The Free Press. (Original work published 1903)

21.

Effrat

Schimmel

(2003). Walking the democratic talk: Introduction to a special issue on collaborative rule-making as preparation for democratic citizenship. American Secondary Education, 31, 3-15.

22.

Ehman

L. H.

(1980). The American school in the political socialization process. Review of Educational Research, 50, 99-119.

23.

Fabelo

Thompson

Plotkin

Carmichal

Marchbanks

Booth

(2011). Breaking schools’ rules: A statewide study of how school discipline relates to students’ success and juvenile justice involvement. New York, NY: Council of State Governments Justice Center and Public Policy Research Institute.

24.

Feldman

Pasek

Romer

Jamieson

K. H.

(2007). Identifying best practices in civic education: Lessons from the Student Voices program. American Journal of Education, 114, 75-100.

25.

Fine

Burns

Payne

Y. A.

Torre

M. E.

(2004). Civics lessons: The color and class of betrayal. Teachers College Record, 106, 2193-2223.

26.

Fuhrman

Lazerson

(Eds.). (2005). The public schools (Institutions of American Democracy). New York, NY: Oxford University Press.

27.

Fung

(2006). Varieties of participation in complex governance. Public Administration Review, 66, 66-75.

28.

Fung

Olin Wright

(2003). Deepening democracy: Institutional innovations in empowered participatory governance. London, England: Verso.

29.

Giroux

H. A.

Purpel

D. E.

(Eds.). (1983). The hidden curriculum and moral education. Berkeley, CA: McCutchan Publishing.

30.

Godsay

Kawashima-Ginsberg

Kiesa

Levine

(2012). “That’s not democracy”: How out-of-school youth engage in civic life and what stands in their way. Medford, MA: The Center for Information & Research on Civic Learning and Engagement.

31.

Gutmann

(1987). Democratic education. Princeton, NJ: Princeton University Press.

32.

Hale

(2011). How information matters: Networks and public policy innovation. Washington, DC: Georgetown University Press.

33.

Hartmann

G. W.

(1946). Interrelations of education and democracy. Review of Education Research, 16, 81-93.

34.

Hirschfield

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

35.

Hyman

H. H.

(1959). Political socialization: A study in the psychology of political behavior. Glencoe, IL: The Free Press.

36.

Jennings

M. K.

(1968). Political socialization and the high school civics curriculum. American Political Science Review, 62, 862-867.

37.

Jennings

M. K.

Niemi

(1974). The political character of adolescence. Princeton, NJ: Princeton University Press.

38.

Kahne

Middaugh

(2008, February). Democracy for some: The civic opportunity gap in high school (CIRCLE Working Paper 59). Medford, MA: The Center for Information & Research on Civic Learning and Engagement.

39.

Kim

C. Y.

Losen

D. J.

Hewitt

D. T.

(2010). The school-to-prison pipeline: Structuring legal reform. New York: New York University Press.

40.

Kupchik

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

41.

Kupchik

Monahan

(2006). The new American school: Preparation for post-industrial discipline. British Journal of Sociology of Education, 27, 617-632.

42.

Lin

(Ed.). (2002). Social capital. Cambridge, UK: Cambridge University Press.

43.

Loeber

Burke

J. D.

Pardini

D. A.

(2009). Development and etiology of disruptive and delinquent behavior. Annual Review of Clinical Psychology, 5, 291-310.

44.

Losen

D. J.

Martinez

T. E.

(2013). Out of school and off-track: The overuse of suspensions in American middle and high schools. Los Angeles, CA: Center for Civil Rights Remedies, The Civil Rights Project.

45.

Lyons

Drew

(2006). Punishing schools: Fear and citizenship in American public education. Ann Arbor: University of Michigan Press.

46.

Manza

Uggen

(2008). Locked out: Felon disenfranchisement and American democracy. New York, NY: Oxford University Press.

47.

McFarland

D. A.

Thomas

R. J.

(2006). Bowling young: How youth voluntary associations influence adult political participation. American Journal of Sociology, 71, 401-425.

48.

Gottfredson

D. C.

(2013). Police officers in schools: Effects on school crime and the processing of offending behaviors. Justice Quarterly, 30, 619-650. doi:10.1080/07418825.2011.615754

49.

Niemi

Junn

(1998). Civic education: What makes students learn. New Haven, CT: Yale University Press.

50.

Noguera

(2003). City schools and the American dream: Reclaiming the promise of public education. New York, NY: Teachers College Press.

51.

Nolan

(2011). Police in the hallways: Discipline in an urban high school. Minneapolis: University of Minnesota Press.

52.

Pasek

Feldman

Romer

Jamieson

K. H.

(2008). Schools as incubators of democratic participation: Building long-term efficacy with civic education. Applied Development Science, 12, 26-37.

53.

Putnam

R. D.

(2000). Bowling alone: The collapse and revival of American community. New York, NY: Simon & Schuster.

54.

Rabe-Hesketh

Skrondal

(2008). Multilevel and longitudinal modeling using Stata (2nd ed.). College Station, TX: Stata Press.

55.

Raudenbush

S. W.

Bryk

A. S.

(2002). Hierarchical linear models: Applications and data analysis methods (2nd ed.). Thousand Oaks, CA: Sage.

56.

Rawlings

K. C.

(2012). Attending Tocqueville’s school: Examining the intrapersonal, political, and civic effects of nonprofit board participation. Administrative Theory & Praxis, 34, 320-356.

57.

Rawlings

K. C.

Catlaw

T. J.

(2011). “Democracy as a way of life”: Rethinking the place and practices of public administration. In King

C. S.

(Ed.), Government is us 2.0 (2nd ed., pp. 31-58). Armonk, NY: M.E. Sharpe.

58.

Schachter

H. L.

Yang

(Eds.). (2012). The state of citizen participation in America. Charlotte, NC: Information Age Publishing.

59.

Simon

(2007). Governing through Crime: How the war on crime transformed American democracy and created a culture of fear. New York, NY: Oxford University Press.

60.

Sirianni

Friedland

(2001). Civic innovation in America: Community empowerment, public policy, and the movement for civic renewal. Berkeley: University of California Press.

61.

Skiba

R. J.

Michael

R. S.

Nardo

A. C.

Peterson

(2000). The color of discipline: Sources of racial and gender disproportionality in school punishment (Research Report SRS1). Bloomington, IN: Indiana Education Policy Center.

62.

Skocpol

(2003). Diminished democracy: From membership to management in American civic life. Norman: University of Oklahoma Press.

63.

Skocpol

Fiorina

M. P.

(Eds.). (1999). Civic engagement in American democracy. Washington, DC: The Brookings Institution.

64.

Suttles

G. D.

(1968). The social order of the slum: Ethnicity and territory in the inner city. Chicago, IL: University of Chicago Press.

65.

Torney-Purta

Amadeo

J.-A.

(2003). A cross-national analysis of political and civic involvement among adolescents. PS: Political Science & Politics, 36, 269-274.

66.

Verba

Schlozman

K. L.

Brady

(1995). Voice and equality: Civic voluntarism in American politics. Cambridge, MA: Harvard University Press.

67.

Wald

Losen

D. J.

(Eds.). (2003). Deconstructing the school-to-prison pipeline: New directions for youth development. San Francisco, CA: Jossey-Bass.

68.

Way

S. M.

(2011). School discipline and disruptive classroom behavior: The moderating effects of student perceptions. The Sociological Quarterly, 52, 346-375.

69.

Wilson

(2000). Volunteering. Annual Review of Sociology, 26, 215-240.

70.

Youniss

Levine

(Eds.). (2009). Engaging young people in civic life. Nashville, TN: Vanderbilt University Press.

71.

Zaff

J. F.

Moore

K. A.

Papillo

A. R.

Williams

(2003). Implications of extra-curricular activity participation during adolescence on positive outcomes. Journal of Adolescent Research, 18, 599-630.

72.

Zimmerman

G. M.

Rees

(2014). Do school disciplinary policies have positive social impacts? Examining the attenuating effects of school policies on the relationship between personal and peer delinquency. Journal of Criminal Justice, 42, 54-65.

73.

Zukin

Keeter

Andolina

Jenkins

Delli Carpini

M. X.

(2006). A new engagement? Political participation, civic life, and the changing American citizen. Oxford, UK: Oxford University Press.