Juvenile Institutional Misconduct: Examining the Role of Educational Attainment and Academic Achievement

Abstract

Identifying risk factors of institutional misconduct among juvenile offenders is important in curbing costs and consequences of these infractions. Research shows low educational attainment and poor academic achievement are consistently related to delinquency and crime. Yet, no research has examined whether these educational variables are associated with institutional misconduct among juveniles. Official records and interviews yielded information on the educational characteristics and institutional misconduct of 349 juvenile prisoners in a California facility. Negative binomial regression analysis was employed to examine hypothesized relationships. Results indicate that graduating high school significantly predicts institutional misconduct. Implications for juvenile corrections policies, including the importance of assisting juvenile detainees in obtaining their high school diploma, and directions for future research are provided.

Keywords

youth violence juvenile corrections school violence education juvenile delinquency

Introduction

According to the Bureau of Justice Statistics (BJS, 2017), the number of juveniles in residential placement in 2015 was 48,043, 55% less than the number in 1999. Nationally, the rate of incarceration was 152 per 100,000 youth. Although these statistics indicate that the number of juveniles held in detention facilities has delcined in recent years, this is still an important population to study as juvenile detention has been linked to subsequent offending (MacKenzie & Farrington, 2015; Soyer, 2014) and has a disproportionate impact on racial minorities (Ewert et al., 2014; Laub, 2018). This research studies a sample of incarcerated youth at a secure correctional facility in the state of California with a focus on their educational characteristics and institutional misconduct.

Extensive research has analyzed the relationship between educational characteristics and delinquency/crime (Foley, 2001; Henry et al., 2012; Maddox & Prinz, 2003; Makarios et al., 2017; Robison et al., 2017; Sweeten et al., 2009). Research examining the association between education and institutional misconduct is rare, however, and results are inconsistent (Tewksbury et al., 2014). Tewksbury and colleagues (2014) argue that education as an indicator of prison misconduct should be considered further. The relatively small number of research studies that have examined misconduct among incarcerated youth (DeLisi et al., 2013; MacDonald, 1999; Reid, 2017; Scott, 2018; Trulson, 2007) have not considered educational characteristics as predictors. This study responds to this gap in the research by examining the relationship between educational characteristics and institutional misconduct among a juvenile correctional population.

The two key educational characteristics considered are educational attainment, measured by grade attained and high school graduation status, and academic performance, measured by Grade Point Average (GPA) and standardized test scores. Studies have indicated that grade retention is related to problem behaviors and delinquency (Jimerson & Ferguson, 2007; Katsiyannis et al., 2012). Lower grade attainment is directly linked to dropping out of high school, which is also associated with delinquency and crime (Harlow, 2003; Henry et al., 2012; Sweeten et al., 2009). Research also indicates that poor academic performance is a predictor of delinquency (Henry et al., 2012; Hoffman et al., 2013; Savolainen et al., 2012). This article’s aim is to determine whether the relationship between these two educational factors, educational attainment and academic performance, and offending carries over into secure youth correctional facilities.

Identifying risk factors of institutional misconduct can help reduce related costs and consequences of these infractions, though most of the research in this area has considered incarcerated adults.¹ As Steiner and Wooldredge (2009) point out, rule violations committed by prisoners pose a threat to the safety of inmates and staff. Financial costs related to disciplinary infractions within correctional institutions are also considerable (Tewksbury et al., 2014). In addition, institutional misconduct has been found to predict recidivism and institutional misconduct during subsequent terms of incarceration (Drury & DeLisi, 2010; Tewksbury et al., 2014; Trulson et al., 2011).

Misconduct may also have consequences for the detainee in terms of disciplinary measures that may impose restrictions to services. As the first study to examine the relationship between lower educational attainment and poor academic performance with institutional misconduct among incarcerated youth, results will help to identify youth most at risk for institutional misconduct and suggest measures that may be taken to prevent it. Results of this research may also have an impact on practices and policies related to educational programs within prison as well as programs designed to improve school attachment to prevent delinquency. Thus, this study has important implications for lowering institutional costs, determining educational and treatment needs, and establishing disciplinary practices.

Literature Review

Research consistently demonstrates the lower educational attainment of delinquent youth. Delinquent youth are more likely to be retained, less likely to graduate from high school, and less likely to perform well in school (Harlow, 2003; Kirk & Sampson, 2013; Makarios et al., 2017; Robison et al., 2017; Ward & Williams, 2015). The percentage of incarcerated youth retained at least one grade ranges from 40% to 50% (Foley, 2001; Zabel & Nigro, 2001). Research examining juvenile offenders indicate that between 70% and 85% do not graduate high school (Foley, 2001) and that being involved with the juvenile justice system decreases the odds of graduating high school (Hjalmarsson, 2008; Makarios et al., 2017; Robison et al., 2017; Ward & Williams, 2015). In a study of Chicago public school students, Kirk and Sampson (2013) found that only 26% of arrested students, compared with 64% of students not arrested, graduated high school. Consistent with these findings, Sweeten (2006) discovered that being arrested during high school doubles the likelihood of dropping out.² Using 1997 data from the National Survey of Youth, Makarios et al. (2017) found that juvenile delinquency consistently predicts school failure. Consistent with this finding, Robison et al. (2017) found that juvenile justice contact was one of the best predictors of school dropout among a cohort of Louisiana students.

Although the relationship between education and delinquency is well-founded, causal inferences are more difficult to ascertain (Sweeten et al., 2009). Therefore, due to its direct application to school performance, the current study is also informed by Thornberry’s (1987) interactional approach. Thornberry (1987) suggests that lower school attachment leads to higher rates of delinquency. Research shows that low educational attainment and poor academic achievement are evidence of lower school attachment (Bryan et al., 2012; Lee & Smith-Adcock, 2005). School disengagement literature indicates that students who are disengaged from school are more likely to experience academic failure, school behavioral problems, grade retention, and suspension/exclusion (Fall & Roberts, 2012; Henry et al., 2012; Sweeten et al., 2009). As the following section reveals, these indicators of low school attachment and school disengagement are also consistently linked to a higher likelihood of juvenile justice involvement, thus, lending support to Thornberry’s theory (Henry et al., 2012; Hjalmarsson, 2008; Robison et al., 2017; Sweeten et al., 2009).

Educational Attainment and Offending

Lower grade-level attainment, often measured as being old-for-grade, has been shown to be related to school misconduct, delinquency, and adult criminal behavior. This lower grade-level attainment is often a result of grade retention, which has been linked to higher rates of offending. Using National Survey of Adolescent Health data, Katsiyannis et al. (2012) found that grade retention during elementary school increased the likelihood of being convicted of a violent crime by 1.5 times. Grade retention during middle or high school had an even greater impact, increasing the probability of committing a violent offense as an adult 3.5 times. A meta-analysis of studies examining the impact of grade retention found that retained students have more negative attitudes toward school, lower academic achievement, lower school attendance, and higher rates of problem behavior (Crothers et al., 2010). Similarly, in a longitudinal study comparing student outcomes of retained students and low-achieving promoted students, Jimerson and Ferguson (2007) discovered that retained students displayed more aggressive behaviors, lower academic achievement, lower attendance, and higher high school dropout rates. Combined, these studies demonstrate a clear link between low grade-level attainment and delinquency, which is consistent with interactional theory’s predictions. The current study seeks to determine whether this relationship between attaining lower grade-level and problem behavior carries over to the carceral environment.

Research has consistently demonstrated that graduating high school is associated with delinquency and crime (Foley, 2001; Henry et al., 2012; Maddox & Prinz, 2003; Maguin & Loeber, 1996; Sweeten et al., 2009). Harlow (2003) found that approximately 50% of federal inmates, 68% of state inmates, and 60% of jail inmates did not receive their high school diploma. Using a nationally representative sample, Sweeten et al. (2009) found a large correlation between dropping out of high school and delinquency. Henry et al.’s (2012) research showed the relationship between school disengagement and criminal behavior in early adulthood was mediated by dropping out. The current study examines whether the well-established relationship between not graduating high school and offending also exists within the institutional setting.

Academic Performance and Delinquency

Although academic performance/achievement and educational attainment are highly correlated, they have been assessed as different factors by researchers with low academic performance or achievement measured by GPA and achievement test scores rather than grade level obtained (e.g., Henry et al., 2012). With this consideration, low academic performance has also consistently been considered evidence of low school attachment and linked to delinquency (DeFoe et al., 2013; Henry et al., 2012; Hoffman et al., 2013; Maddox & Prinz, 2003; Sander et al., 2012; Smith-Adcock et al., 2011; Spatz & Bamba, 2000; Sweeten et al., 2009). Furthermore, a number of these studies found that low academic performance and problem behavior are characteristics of students with low levels of attachment to school (Hirschfield & Gasper, 2011; Maddox & Prinz, 2003; McDermott et al., 2001). Maddox and Prinz (2003), in a review of research on school bonding, conclude that higher levels of school bonding are associated with higher self-esteem and academic performance but lower levels of substance use, delinquency, and antisocial behavior. School delinquency studies have revealed that low academic achievement predicts problem behavior in schools (Maddox and Prinz, 2003; Smith-Adcock et al., 2011). Using National Educational Longitudinal Study data, Smith-Adcock and colleagues (2011) found that students with high academic achievement engaged in less misconduct than lower achieving students. In a review of research related to academic performance and delinquency, Maguin and Loeber (1996) found that poor school performance was related to the prevalence, seriousness, and persistence of delinquent offending.

More recent studies have examined whether the relationship between academic performance and delinquency is direct and causal. Using data from 503 boys in the Pittsburgh Youth Study, DeFoe et al. (2013) discovered that, compared with hyperactivity and socioeconomic status (SES), low academic performance had the most direct influence on delinquency. In their analysis of data from the first two waves of the National Longitudinal Study of Adolescent Health, Hoffman et al. (2013) tested an interactional theory³ derived hypothesis of academic achievement and delinquency. Their findings suggest that higher academic achievement measured by GPAs reduces the likelihood of involvement in delinquent activities and that this relationship persists after controlling for proposed common causal factors. These findings again support the hypothesized relationship between school attachment and delinquency proposed by Thornberry (1987).

Although the relationship between educational factors and higher levels of delinquency is well-established and theoretically supported, the impact of academic performance and educational attainment on juvenile institutional misconduct has not been examined. Some research has considered this relationship among adult inmates. A description of the research examining the relationship between education and prison misconduct among adults and other factors related to institutional misconduct among both youth and adults follows.

Factors Related to Institutional Misconduct

The prevalence of prison violence and misconduct has fueled a large body of research examining correlates of institutional misconduct in adult detention facilities (Colvin, 2007; DeLisi et al., 2011; Drury & DeLisi, 2010; Kuanliang et al., 2008; Steiner & Wooldredge, 2009; Tewksbury et al., 2014; Trulson et al., 2011; Walters & Crawford, 2013). According to Wolff and colleagues (2007), the nature of the prison environment fuels violence. Its walls confine “hundreds of thousands of people with antisocial tendencies or behavior in close and frequently overcrowded quarters characterized by material and social deprivation,” resulting in a context of violence and victimization (p. 588). Understanding the causes and correlates of institutional misconduct are essential to successfully addressing behavior issues within correctional facilities.

Although not directly tested, the current study is also informed by the importation (Irwin & Cressey, 1962) and deprivation (Clemmer, 1940; Sykes, 2007) models. The importation model asserts that violent misconduct among incarcerated offenders is associated with individual traits that are brought into the institution; the deprivation model argues that correctional violence happens because of characteristics and dynamics that are unique to the institutional setting (Walters & Crawford, 2013). Factors related to prison misconduct have frequently been categorized as either importation or deprivation variables (Walters & Crawford, 2013). Two importation variables most consistently found to be related to institutional misconduct are age and prior criminal history (Steiner & Wooldredge, 2009; Steiner et al., 2014; Walters & Crawford, 2013). Younger inmates and those with prior criminal history have higher rates of rule violations compared with their counterparts. In addition, street gang affiliation is another well-established covariate of institutional misconduct (DeLisi et al., 2013; Steiner et al., 2014; Walters & Crawford, 2013). A less examined but important covariate of prison misconduct is history of disciplinary infractions during prior incarceration (Drury & DeLisi, 2010; Steiner et al., 2014). Drury and DeLisi (2010) found that inmates with this prior record of misconduct were more likely to engage in violent and nonviolent institutional misconduct. In their review of research on institutional misconduct, Steiner et al. (2014) also found that drug-related and mental health issues were consistently revealed as contributing to inmate misconduct. Additional importation factors linked to prison misconduct but with less consistency include race/ethnicity, education, marital status, employment, and history of abuse (DeLisi et al., 2011; Steiner et al., 2014; Tewksbury et al., 2014; Walters & Crawford, 2013).

Education has been examined less frequently than other importation characteristics (Tewksbury et al., 2014). Research has shown, however, that educational attainment prior to incarceration is negatively related to institutional misconduct in studies of adult correctional populations (Kuanliang et al., 2008; Tewksbury et al., 2014). Using the Test of Adult Basic Education (TABE) to measure educational level, Kuanliang et al. (2008) found that inmates with lower scores on this exam were more likely to commit acts of prison misconduct. Conversely, in their examination of correlates of institutional misconduct, Tewksbury et al. (2014) did not find educational level to be a significant predictor of disciplinary infractions when controlling for other factors. These inconsistent findings warrant further examination.

Deprivation factors identified in the literature as increasing inmate misconduct are harsh punitary factors and other coercive elements (Colvin, 2007; Day et al., 2015). Other deprivation factors examined in the literature on institutional misconduct include crowding, levels of visitation, level of security, and type of conviction (DeLisi et al., 2011; Steiner et al., 2014; Tewksbury et al., 2014). Review of the literature reveals level of security as the only prison characteristic to consistently predict institutional misconduct (Steiner et al., 2014).

Factors Related to Juvenile Institutional Misconduct

Although adult institutional misconduct has been given much attention by criminologists, limited research has focused on misconduct within juvenile facilities. Factors that have been found to be predictors of juvenile institutional misconduct include prior history of delinquency, gang membership, parental income and education, and race (DeLisi et al., 2013; MacDonald, 1999; Reid, 2017; Trulson, 2007). MacDonald (1999) analyzed select importation and deprivation factors on juvenile inmates’ violence and drug use. Importation factors that predicted violent institutional misconduct included length of criminal involvement, prior gang involvement, and history of violence or drugs. He also found that level of security was positively related to violent misconduct. Trulson’s (2007) later examination of a sample (4,684) of state-committed juvenile detainees largely confirmed these results. Inmates committing higher rates of both minor and serious institutional misconduct were males, non-White, those with gang influence, and those with earlier, more extensive, and more serious delinquent histories (Trulson, 2007). Finally, in a study of youth incarcerated with California’s Division of Juvenile Justice (DJJ), Reid (2017) found that gang membership, having an adult court commitment status, and having a close friend in prison increased the likelihood of serious institutional misconduct. No known prior study has specifically examined educational characteristics as a risk factor of institutional misconduct among juveniles. It is this research gap that the current study strives to fill.

Current Study

This study examines whether incarcerated juveniles with lower grade-level attainment and academic performance have more disciplinary infractions compared with other juvenile detainees. Based on the reviewed literature, the following research hypotheses are proposed:

Hypothesis 1: Incarcerated youth who have lower educational attainment, measured by grade attained and high school graduation, will have a higher likelihood of engaging in any institutional misconduct.

Hypothesis 2: Incarcerated youth who have lower educational attainment will have a higher likelihood of engaging in violent misconduct.

Hypothesis 3: Lower achieving incarcerated youth, measured by standardized test scores and GPA, will have a higher probability of any institutional misconduct compared with their higher achieving counterparts.

Hypothesis 4: Lower achieving incarcerated youth will have a higher probability of violent misconduct compared with their higher achieving counterparts.

If these hypotheses are supported, identifying youth with lower educational attainment and low-achieving youth as more at risk of prison misconduct can aid in developing more effective correctional policies, education, intervention, and prevention programs.

Method

Sample

Data for this sample are taken from semi-structured interviews and official records collected through a larger study examining California DJJ incarcerated youth that was conducted in 2010. This larger study analyzed violence and gang issues within these correctional facilities and provided policy suggestions to state correctional administrators. Random sampling procedures generated a representative sample of 349 in-depth semi-structured interviews. The participation rate was 84%. This sample included youth from the five DJJ facilities in California that were in operation at the time.

Variables

For the current study, one key independent variable, highest grade completed, and several control variables were obtained from the interview data. GPA, Standardized Testing and Reporting (STAR) achievement levels, and Disciplinary Decision Making System (DDMS) information were obtained through DJJ official records of the interviewed sample. These official data included youth demographics, facility classifications, and delinquent/criminal history. Descriptions of each of these variables are presented in the following section, and univariate and bivariate statistics are presented in Table 1.

Table 1.

Univariate and Bivariate Analysis.

Nominal variables	%	Bivariate analysis
		Any DDMS incidents		Violent DDMS incidents
		M (SD)	t, F	M (SD)	t, F
Sex (n = 347)			t = 5.49***		t = 1.35
Males	87.3	12.5 (16.5)		2.4 (3.3)
Females	12.7	29.8 (34.5)		3.1 (4.2)
Race (n = 347)			F = 4.12**		F = 6.5***
Latino	59.7	17.3 (21.7)		3.1 (3.9)
African American	28	12.9 (19.6)		1.8 (2.5)
White	7.2	6.4 (13.1)		1.0 (1.5)
Other	5.2	5.4 (6.8)		1.0 (1.4)
Gang member (n = 323)			t = −3.44***		t = −4.71***
Yes	68.7	16.9 (21.2)		2.8 (3.6)
No	31.3	8.6 (17.3)		1.1 (2.0)
Type of offense (n = 349)			F = 3.94**		F = 4.92***
Homicide	11.5	9.1 (14.9)		1.4 (2.2)
Assault	33.5	16.2 (20.9)		3.3 (4.0)
Property crime	28.7	19.4 (24.3)		2.7 (3.4)
Sex violations	16.0	8.2 (13.9)		1.3 (2.5)
Other^a	10.3	12.3 (16.4		2.1 (3.0)
Grad status (n = 347)			t = 3.72***		t = 2.46*
In high school	63.4	17.7 (22.1)		2.8 (3.4)
HS diploma/GED	36.6	9.4 (15.9)		1.9 (3.4)
Continuous variables	M (SD)		$ρ$	$ρ$
Age (n = 333)	18.7 (2.3)		−.05	−.23**
Number of arrests (n = 320)	7.3 (10.6)		.16**	.14*
Length of time served (n = 332)	23 (17.2)		.29**	.06
GPA^b (n = 330)	2.7 (0.51)		−.16**	.08
Highest grade (n = 308)	9.0 (1.6)		−.24**	−.13*
STAR grade level (n = 347)	11.4 (1.3)		−.16**	−.17**

Note. DDMS = Disciplinary Decision Making System; GED = General Educational Development; HS = high school; GPA = Grade Point Average; STAR = Standardized Testing and Reporting.

Other offenses include technical violation, miscellaneous felonies, and unspecified. ^bGPAs of 0 were excluded.

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

Institutional Misconduct

The dependent variables in this study were two measures of institutional misconduct. These included the number of any DDMS incidents and the number of violent DDMS incidents filed against incarcerated youth in the last year. Descriptive statistics of these two dependent variables are presented in Table 1. The total number of any DDMS incidents filed against respondents in the last year ranged from 0 to 134 with a mean of 14.6 (SD = 20.4). The total number of violent DDMS incidents filed against respondents in the last year ranged from 0 to 20 with a mean of 2.43 (SD = 3.4).

Data Analysis

Univariate analysis of all study variables and bivariate analysis of all independent and control variables with the two measures of institutional misconduct were conducted. A correlational analysis of all study variables was conducted to examine the bivariate relationships among all variables and to determine any potential problems of multicollinearity, of which none were found. After checking for linearity assumptions and transforming variables as deemed appropriate, negative binomial regression analyses were conducted. This analysis was employed due to the nature of the count data and distribution of the DDMS variables.

Univariate and bivariate analysis

Descriptive statistics and bivariate statistical analysis of the four demographic control variables are presented in Table 1. The sample consists of 303 (87.3%) males and just 44 (12.7%) females. Results of t tests indicate sex is significantly related to any institutional misconduct (t = 5.49, p < .001) but not violent misconduct. Females reportedly had a significantly higher number of incidents of any misconduct than males. The age of this sample of juvenile inmates ranged from 12 to 25 with a mean age of 18.7 (SD = 2.3). Spearman’s correlation coefficient (ρ = −.23, p < .01) showed age was moderately and negatively related to violent misconduct but not any misconduct. More than half of the sample (59.7%) were Latino, 28% were African American, 7.2% were White, and about 5% were another ethnicity. Bivariate analysis showed that race/ethnicity was related to both measures of institutional misconduct, with Latinos having the highest number of both any DDMS incidents and violent DDMS incidents. Just more than two thirds (68.7%) of those interviewed were affiliated with a street gang. Street gang membership was significantly related to institutional misconduct with gang-involved youth having a significantly higher number of incidents compared with youth not gang involved.

Descriptive statistics and bivariate analysis results for juvenile justice characteristics are also presented in Table 1. Number of times arrested and length of time served were included because research has shown these factors may impact frequency of institutional misconduct. For example, one study found that prisoners with prior incarceration have lower levels of education compared with juveniles without prior incarceration (Foley, 2001). The total reported number of arrests in this study’s sample ranged from 1 to 100 with a mean of 7.3 (SD = 10.6) and a median of four arrests. Number of arrests was only weakly but positively related to both any DDMS incidents $(ρ = . 16, p < . 01)$ and violent DDMS incidents $(ρ = . 14, p < . 05)$ . The two most common types of offenses for which the sample had been convicted were assault (33.5%) and property crime (28.7%). Type of offense was significantly related to both types of institutional misconduct with property offenders having the highest average number of both any and violent DDMS incidents. A deprivation characteristic related to juvenile justice employed as a control variable based on past research was time spent in the current facility or length of time served. The total number of months the youth had spent in the current facility ranged from less than a week (0.03 months) to 108 months. The mean number of months this group had served was 23 months (SD = 17.2), and 46 of the juveniles in the sample were serving life in prison. Length of time served was moderately and positively related to being involved in any institutional misconduct $(ρ = . 29, p < . 01)$ but not violent misconduct.

Four educational characteristics identified as indicators of low school attachment in the literature were used to operationalize the key independent variables of educational attainment and academic achievement/performance. Measures of educational attainment were the incarcerated youths’ highest grade completed and high school graduation status. Academic achievement was operationalized as GPA and scores on the STAR achievement test.

Univariate and bivariate statistics of these key educational characteristics are presented in Table 1. Of 304 youth responding to the question asking their highest grade completed, only 4.5% had completed 12th grade. About 17% had completed either 11th grade or 10th grade, 25% had completed ninth grade, 22.4% had completed eighth grade, and 14.3% had completed less than eighth grade. The mean highest grade completed was ninth (SD = 1.6). Highest grade completed was weakly and negatively related to both measures of institutional misconduct $(ρ = - . 24, p < . 01, ρ = - . 13, p < . 05)$ . About two thirds (63.4%) were in high school with the other third (34.6%) having attained their high school diploma or GED. Results of the t test indicate that graduation status is related to involvement in any DDMS incidents (t = 3.72, p < .001) and violent DDMS incidents (t = 2.46, p < .05).

The respondents’ average GPA ranged from 1.2 to 4.0 with a mean of 2.7 (SD = .51). Interviewees who had a GPA of 0 were excluded. GPA was weakly and negatively related to any misconduct $(ρ = - . 16, p < . 01)$ but not violent misconduct. Scores on the STAR, a standardized test administered through the California Department of Education from school years 1998–1999 through 2012–2013, were obtained from official DJJ documents for those interviewed. Scores are presented as a grade level, indicating that the student obtained the average score for that grade level. Grade Level 13 is “graduated.” The mean STAR grade level was 11.4 (SD = 1.3). STAR grade level was weakly and negatively related to both any DDMS incidents $(ρ = - . 16, p < . 05)$ and violent DDMS incidents $(ρ = - . 17, p < . 01)$ . Based on significant bivariate results, multivariate analyses were conducted to take the control variables into account and better understand the relationship between educational attainment and misconduct.

Negative binomial regression results

Negative binomial regression models were utilized for each dependent variable. Separate models include one for each of the four educational variables and one combined full model. Table 2 shows results of the analyses for any institutional misconduct incidents in the last year.

Table 2.

Binomial Regression Analysis of Any Institutional Misconduct in the Last Year—Models 1 to 5.

Variable	Model 1		Model 2		Model 3		Model 4		Model 5
Variable	IRR	p	IRR	p	IRR	p	IRR	p	IRR	p
Age	0.99	.86	1.02	.59	1.00	.92	0.96	.37	1.04	.42
White	0.37	.01	0.40	.01	0.39	.01	0.37	.00	0.43	.03
Black	1.12	.61	0.91	.61	0.96	.83	1.03	.90	0.99	.96
Other	0.35	.00	0.44	.02	0.42	.01	0.35	.00	0.39	.02
Gang member	2.11	.00	2.15	.00	2.16	.00	2.07	.00	2.11	.00
No. of times arrested	1.01	.48	1.00	.62	1.00	.71	1.00	.78	1.00	.55
Time served	1.00	.93	1.00	.56	1.00	.70	1.00	.92	1.00	.82
Homicide offense	1.03	.96	0.87	.75	0.87	.75	0.97	.95	1.03	.95
Property offense	1.41	.06	1.35	.07	1.41	.04	1.46	.03	1.31	.12
Sex offense	0.71	.20	0.71	.17	0.69	.16	0.71	.18	0.75	.26
Other offense	0.86	.65	0.79	.47	0.72	.31	0.92	.80	0.85	.62
Highest grade	0.91	.06	**	**	**	**	**	**	0.95	.28
Graduation status	**	**	0.50	.00	**	**	**	**	0.54	.01
STAR grade level	**	**	**	**	0.85	.02	**	**	.97	.78
GPA	**	**	**	**	**	**	.91	.55	.89	.49
Wald χ ² (p value)	60.36 (.00)		89.31 (.00)		71.18 (.00)		63.65 (.00)		79.34 (.00)
Pseudo R ²	.03		.03		.03		.03		.03

Note. IRR = incident risk ratio; STAR = Standardized Testing and Reporting; GPA = Grade Point Average.

The first four models reveal similar findings regarding control variables. In these four models, Latino youth and those that identify as street gang members are more likely to have engaged in institutional misconduct in the last year. Regarding educational characteristics, when considered separately, three of the four educational variables were significant predictors of institutional misconduct. Model 1 shows that the association between being old-for-grade and any institutional misconduct participation in the last year is of marginal significance. Model 2 indicates that youth who have graduated from school are at about a 50% decreased risk of having engaged in any institutional misconduct in the last year. Model 3 reveals that youth with a higher STAR grade level reduced that detained juvenile’s risk of having engaged in any institutional misconduct by 15% in the last year. Results of Model 4 indicate that GPA did not have a statistically significant effect on youth’s participation in any type of institutional misconduct in the last year.

Model 5 is the full model including all four educational variables. The results reveal that graduation status is the only educational variable that continues to be significant after controlling for all other study variables. Youth who have graduated from school are at about a 46% decreased risk of having engaged in any institutional misconduct in the last year.

Various control variables are also revealed to be predictors of any institutional misconduct. White youth are at about a 57% decreased risk of having engaged in any institutional misconduct in the last year compared with Latino youth. Youth who identify as street gang members are at 111% greater risk of participating in any institutional misconduct compared with nongang members.

Table 3 presents the five negative binomial regression models for violent misconduct incidents in the last year. The first four models include only one educational variable. These four models indicate that none of these educational variables were significant predictors of likelihood of engaging in institutional misconduct. Control variables that were predictive of higher levels of institutional misconduct included being Latino, identifying as a street gang member, being younger, and being convicted of an assault compared with a sex offense.

Table 3.

Binomial Regression Analysis of Violent Misconduct in the Last Year—Models 6 to 10.

Variable	Model 6		Model 7		Model 8		Model 9		Model 10
Variable	IRR	p	IRR	p	IRR	p	IRR	p	IRR	p
Age	0.88	.00	0.86	.01	0.86	.00	0.86	.00	0.88	.03
White	0.51	.11	0.47	.05	0.47	.05	0.48	.06	0.52	.12
Black	1.12	.58	0.95	.82	0.96	.83	1.03	.89	1.12	.57
Other	0.47	.02	0.43	.00	0.43	.01	0.43	.00	0.47	.03
Gang member	2.29	.00	2.35	.00	2.35	.00	2.47	.00	2.32	.00
No. of times arrested	1.00	.52	1.00	.61	1.00	.61	1.00	.75	1.01	.58
Time served	1.00	.38	1.00	.33	1.00	.33	1.00	.38	1.00	.44
Homicide offense	0.65	.23	0.63	.18	0.63	.18	0.62	.21	0.62	.21
Property offense	0.82	.30	0.92	.64	0.92	.65	0.91	.62	0.80	.22
Sex offense	0.52	.02	0.51	.01	0.51	.01	0.53	.02	0.52	.02
Other offense	0.57	.11	0.56	.09	0.56	.09	0.61	.15	0.58	.11
Highest grade	0.95	.25	**	**	**	**	**	**	0.98	.66
Graduation status	**	**	0.99	.96	**	**	**	**	0.98	.95
STAR grade level	**	**	**	**	1.00	.99	**	**	0.94	.53
GPA	**	**	**	**	**	**	1.11	.48	1.04	.82
Wald χ ² (p value)	62.83 (.00)		58.15 (.00)		58.15 (.00)		56.82 (.00)		59.10 (.00)
Pseudo R ²	.06		.06		.06		.06		.06

Note. IRR = incident risk ratio; STAR = Standardized Testing and Reporting; GPA = Grade Point Average.

Model 10, as presented in Table 3, is the full model analyzing predictors of violent institutional misconduct in the last year. No educational variables were revealed to have significance. Older youth are at about a 12% decreased risk of engaging in violent incidents in the last year. Youth who identify as a race other than Latino, Black, or White are at about a 54% decreased risk of violent incident involvement in the last year compared with Latino youth. Youth who report being a street gang member are about 132% more likely to have participated in violent incidents in the last year compared with nongang members. Youth committed for a sex offense are at about a 48% decreased risk of violent incident involvement in the last year compared with youth committed for an assault.

Summary

The only educational variable found to have a significant effect on any type of institutional misconduct in the last year when controlling for other educational variables was graduation/GED status. When controlling for all other variables, those who had graduated from high school had less than half the risk of committing institutional misconduct compared with those who had not graduated high school. No educational variables, however, were significantly related to juvenile detainees’ likelihood of being involved in violent misconduct in the last year. Results support the research hypothesis that those with higher educational attainment as measured by graduation/GED status are involved in less institutional misconduct than those with lower educational attainment. Findings, thus, provide only limited support that juvenile inmates who are disengaged from school are more likely to commit institutional misconduct.

Discussion

Theoretical and Policy Implications

Descriptive statistics of this sample related to educational attainment are very informative. They show that a large portion of the young offenders in the sample had low educational attainment and academic achievement. For instance, 36.4% had not completed ninth grade but only 3.7% were under 15, the age a student is typically expected to complete ninth grade. In addition, 32.6% were age 20 or older but only 4.5% had completed 12th grade. These statistics indicate that a large portion of these young offenders had low grade-level attainment, and many had not completed high school by age 18 or 19. Standardized test scores suggest that many of these youth did not reach their academic potential in school. Although the large majority (80.6%) had not completed 11th grade, the majority (73.6%) scored at Grade 11 on the STAR achievement test, and more than one fourth (27.4%) scored at Grade 13. These educational characteristics suggest low levels of school attachment among these juvenile offenders and emphasize the need to implement programs in schools and the community to increase levels of school attachment among at-risk youth.

Having graduated high school or obtained a GED was the only educational variable to have a significant association with institutional misconduct in the last year in the full model. This finding is arguably of practical significance. This finding suggests that graduation is crucial in decreasing the likelihood of negative behavior among incarcerated youth. Other educational correlates appear to have less of an effect on the behavior of youth. This indicates that the diploma or GED may be an indicator of success for these youth that brings with it a desire to maintain positive behavior, as they do not want to jeopardize this success. Misconduct could also negatively affect goals related to release that may seem more attainable once they have obtained their diploma or GED. Because graduation is also an indicator of school attachment, this finding is arguably consistent with Thornberry’s interactional theory, and supports the idea that youth with higher levels of school attachment will engage in less misconduct. Other educational variables such as a standardized test score, GPA, and even grade level may not be the best measures of school attachment, or school attachment may not be a significant factor within the institution. Alternatively, it is possible that the relationship between school attachment and institutional misconduct is indirect until they obtain their diploma or GED. The results highlighted other additional factors worth consideration, including gang involvement and age when examining education and the likelihood of misconduct among incarcerated youth.

Results clearly demonstrate the importance of gang membership as a predictor of both any institutional misconduct and violent misconduct. In all 10 multivariate analyses, gang members had a higher probability of engaging in institutional misconduct compared with those with no gang affiliation. It is possible that gang membership has a negative impact on a variety of things in a youth’s life. Not only does it increase the likelihood of crime (see Decker et al., 2013, for a review), but it might also have a negative impact on educational success among incarcerated youth due to increases in the likelihood of gang-related violent misconduct while in school (see Scott, 2018). This is consistent with Thornberry’s (1987) interactional approach, which suggests that youth with lower levels of school attachment will have attachment to antisocial peers (also see Jang, 1999). More work needs to be conducted on the potential relationship between gang membership and academic achievement, and, specifically, how this influences misconduct within youth correctional settings. It is also possible that youth were gang-involved prior to incarceration, which increases the likelihood of misconduct while incarcerated (DeLisi et al., 2004; Tasca et al., 2010), reducing the likelihood of academic success. These findings again suggest the need to implement programs designed to improve school attachment for at-risk youth, including targeted programs for those that are gang-involved.

It is also important to examine age in relation to education because the younger the youth, the less likely they are to have obtained a diploma/GED. It is possible that younger youth are more likely to engage in violence because they are less mature. For example, younger incarcerated offenders may be experiencing deprivation that results in anxiety (Gover et al., 2000), which could increase the likelihood of engaging in misconduct, and decrease the probability of academic success. Thus, changing the learning environment to a space where they are not likely to interact with potential rivals could increase their likelihood of receiving a diploma/GED, and thereby decrease the likelihood of misconduct.

Limitations

There are various limitations to this study that need to be addressed. The data are cross-sectional, making it impossible to infer causation. In addition, interviews with these youth were conducted at a time in the California DJJ when major changes were occurring. Facilities were closing, and various youth were being sent to the county level, leaving the state facilities for only the most serious youth offenders. Furthermore, the values of the pseudo-R² in the multivariate analyses are a little low. Future analyses will need to be conducted to determine the relationship between education and facility misconduct among incarcerated youth. Research will also need to be conducted in different jurisdictions to confirm the generalizability of these findings.

Conclusion

Achieving a diploma or GED is significantly associated with a reduced likelihood of misconduct among incarcerated youth. For these youth, academic performance, measured by GPA and STAR test score, has little impact on their behavior. These findings suggest that developing strategies to increase the likelihood of graduating high school and improving school attachment is crucial to reducing misconduct among incarcerated youth. The significant association found between misconduct and gang membership and race suggests a complex relationship between individual youth characteristics and institutional misconduct. It is possible that reducing the likelihood of gang involvement and developing effective racial integration strategies would increase the probability of academic success among incarcerated youth, significantly reduce the likelihood of misconduct, and, ultimately, aid in the success of incarcerated youth upon release.

Footnotes

Declaration of Conflicting Interests

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

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was conducted with the support of funding from the California Department of Corrections and Rehabilitation (CDCR), Agreement #5600001351.The opinions expressed herein represent those of the authors and do not necessarily represent the position of the CDCR.

ORCID iDs

Ronda Shepherd Engstrom

Daniel Scott

Notes

Author Biographies

Ronda Shepherd Engstrom, MA, is a doctoral candidate and teaching associate in the Department of Criminology and Criminal Justice at Indiana University of Pennsylvania. Her research interests include the importance of helping at-risk youth obtain educational success, foster youth and criminal offending, and crossover youth. She has presented her research at the American Society of Criminology, American Criminal Justice Society, and the Northeast American Criminal Justice Society’s annual conferences.

Daniel Scott, PhD, is an assistant professor of Criminology and Criminal Justice justice at the University of West Florida. His research interests are in juvenile delinquency with a focus on youth and gang violence, juvenile justice and corrections, gang involvement, policies and programs, criminological theory, and mixed methods. His current research examines youth correctional violence, and regional differences in gang polices/programs. His work has appeared in Deviant Behavior, Criminal Justice Education, and Journal of Ethnicity in Criminal Justice.

References

Bryan

Moore-Thomas

Gaenzle

Kim

Lin

(2012). The effects of school bonding on high school seniors’ academic achievement. Journal of Counseling & Development, 90(4), 467–480.

Bureau of Justice Statistics. (2017). OJJDP Statistical Briefing Book. Retrieved from https://www.ojjdp.gov/ojstatbb/corrections/qa08201.asp?qaDate=2015.

Taft

D.R.

Clemmer

(1940). “The Prison Community.” Journal of Criminal Law and Criminology, (1931-1951) 31(4), 479-481.

Colvin

(2007). Applying differential coercion and social support theory to prison organization: The case of the penitentiary of New Mexico. The Prison Journal, 87, 367–387.

Crothers

L. M.

Schreiber

J. B.

Schmitt

A. J.

Bell

G. R.

Blasik

Comstock

L. A.

Greisler

M. J.

Keener

King

J. M.

Lipinski

(2010). A preliminary study of bully and victim behavior in old-for-grade students: Another potential hidden cost of grade retention or delayed school entry. Journal of Applied School Psychology, 26(4), 327–338.

Day

J. C.

Brauer

J.R.

Butler

H. D.

(2015). Coercion and social support behind bars: Testing an integrated theory of misconduct and resistance in US prisons. Criminal Justice and Behavior, 42(2), 133–155.

Decker

S. H.

Chris

David

C. P.

(2013). “What Do We Know About Gangs and Gang Members and Where Do We Go From Here?” Justice Quarterly, 30(3), 369–402.

DeFoe

I. N.

Farrington

D. P.

Loeber

(2013). Disentangling the relationship between delinquency and hyperactivity, low achievement, depression, and low socioeconomic status: Analysis of repeated longitudinal data. Journal of Criminal Justice, 41, 100–107.

DeLisi

Berg

M. T.

Hochstetler

(2004). Gang members, career criminals and prison violence: Further specification of the importation model of inmate behavior. Criminal Justice Studies, 17(4), 369–383.

10.

DeLisi

Spruill

J. O.

Peters

D. J.

Caudill

J. W.

Trulson

C. R.

(2013). “Half in, half out:” Gang families, gang affiliation, and gang misconduct. American Journal of Criminal Justice, 38, 602–615.

11.

DeLisi

Trulson

C. R.

Marquart

J. W.

Drury

A. J.

Kosloski

A. E.

(2011). Inside the prison black box: Toward a life course importation model of inmate behavior. International Journal of Offender Therapy and Comparative Criminology, 55(8), 1186–1207.

12.

Drury

DeLisi

(2010). The past is prologue: Prior adjustment to prison and institutional misconduct. The Prison Journal, 90(3), 331–352.

13.

Ewert

Sykes

B. I.

Pettit

(2014). The degree of disadvantage: Incarceration and inequality in education. The Annals of the American Academy of Political and Social Science, 651(1), 24–43.

14.

Fall

Roberts

(2012). High school dropouts: Interactions between social context, self-perception, school engagement and student dropout. Journal of Adolescence, 35, 787–798.

15.

Foley

R. M.

(2001). Academic characteristics of incarcerated youth and correctional educational programs. Journal of Emotional and Behavioral Disorders, 9(4), 248–259.

16.

Gover

A. R.

Mackenzie

D. L.

Armstrong

G. S.

(2000). Importation and deprivation explanations of juveniles’ adjustment to correctional facilities. International Journal of Offender Therapy and Comparative Criminology, 44(4), 450–467.

17.

Harlow

(2003). Education and correctional populations (Bureau of Justice Statistics Special Report). U.S. Department of Justice.

18.

Henry

K. L.

Knight

K. E.

Thornberry

T. P.

(2012). School disengagement as a predictor of dropout, delinquency, and problem substance use during adolescence and early adulthood. Journal of Youth and Adolescence, 41, 156–166.

19.

Hirschfield

P. J.

Gasper

(2011). The relationship between school engagement and delinquency in late childhood and early adolescence. Journal of Youth and Adolescence, 40(1), 3–22.

20.

Hjalmarsson. (2008). Criminal justice involvement and high school completion. Journal of Urban Economics, 63, 613–630.

21.

Hoffman

J. P.

Erickson

L. D.

Spence

K. R.

(2013). Modeling the association between academic achievement and delinquency: An application of interactional theory. Criminology, 51(3), 629–660.

22.

Irwin

Cressey

D. R.

(1962). Thieves, convicts and the inmate culture. Social Problems, 10, 142–155.

23.

Jang

S. J.

(1999). Age-varying effects of family, school, and peers on delinquency: A multilevel modeling test of interactional theory. Criminology, 37(3), 643–686.

24.

Jimerson

S. R.

Ferguson

(2007). A longitudinal study of grade retention: Academic and behavioral outcomes of retained students through adolescence. School Psychology Quarterly, 22(3), 314–339.

25.

Katsiyannis

Thompson

M. P.

Barrett

D. E.

Kingree

J. B.

(2012). School predictors of violent criminality in adulthood: Findings from a nationally representative longitudinal study. Remedial and Special Education, 34(4), 205–214.

26.

Kirk

D. S.

Sampson

R. J.

(2013). Juvenile arrest and collateral educational damage in the transition to adulthood. Sociology of Education, 86(1), 36–62.

27.

Kuanliang

Sorensen

J. R.

Cunningham

M. D.

(2008). Juvenile inmates in an adult prison system: Rates of disciplinary misconduct and violence. Criminal Justice and Behavior, 35, 1186–1201.

28.

Laub

J. H.

(2018). Reducing justice system inequality: Introducing the issue. The Future of Children, 28(1), 3–10.

29.

Lee

S. M.

Smith-Adcock

(2005). A model of girls’ school delinquency: School bonding and reputation. Professional School Counseling, 9(1), 78–87.

30.

MacDonald

J. M.

(1999). Violence and drug use in juvenile institutions. Journal of Criminal Justice, 27, 33–44.

31.

MacKenzie

D. L.

Farrington

D. P.

(2015). Preventing future offending of delinquents and offenders: What have we learned from experiments and meta-analyses? Journal of Experimental Criminology, 11, 565–595.

32.

Maddox

S. J.

Prinz

R. J.

(2003). School bonding in children and adolescents: Conceptualization, assessment, and associated variables. Clinical Child & Family Psychology Review, 6(1), 31–49.

33.

Maguin

Loeber

(1996). Academic performance and delinquency. Crime and Justice, 20, 145–264.

34.

Makarios

Cullen

Piquero

(2017). Adolescent criminal behavior, population heterogeneity, and cumulative disadvantage: Untangling the relationship between adolescent delinquency and negative outcomes in emerging adulthood. Crime & Delinquency, 63(6), 683–707.

35.

McDermott

P. A.

Mordell

Stoltzfus

J. C.

(2001). The organization of student performance in American schools: Discipline, motivation, verbal learning, and nonverbal learning. Journal of Educational Psychology, 93(1), 65–76.

36.

Reid

S. E.

(2017). Friendship group composition and juvenile institutional misconduct. International Journal of Offender Therapy and Comparative Criminology, 61(2), 191–209.

37.

Robison

Jaggers

Rhodes

Blackmon

B. J.

Church

(2017). Correlates of educational success: Predictors of school dropout and graduation for urban students in the Deep South. Children and Youth Services Review, 73, 37–46.

38.

Sander

J. B.

Patall

E. A.

Amoscato

L. A.

Fisher

A. L.

Funk

(2012). A meta-analysis of the effect of juvenile delinquency interventions on academic outcomes. Children and Youth Services Review, 34, 1695–1708.

39.

Savolainen

Hughes

Mason

Hurtig

Ebeling

Moilanen

. . . Taanila

(2012). Antisocial propensity, adolescent school outcomes, and the risk of criminal conviction. Journal of Research on Adolescence, 22(1), 54–64.

40.

Scott

(2018). A comparison of gang- and non-gang-related violent incidents from the incarcerated youth perspective. Deviant Behavior, 39(10), 1336–1356.

41.

Smith-Adcock

Lee

S. M.

Kerpelman

Majuta

Young

C. B.

(2011). Reputation enhancement and school delinquency: A prospective study using the National Educational Longitudinal Survey. Youth & Society, 45(2), 201–220.

42.

Soyer

(2014). The imagination of desistance. British Journal of Criminology, 54(1), 91–108.

43.

Spatz

Bamba

(2000). Education and delinquency: Summary of a workshop. National Academies Press.

44.

Steiner

Wooldredge

(2009). Implications of different outcome measures for an understanding of inmate misconduct. Crime & Delinquency, 59(8), 1234–1262.

45.

Steiner

Butler

H. D.

Ellison

J. M.

(2014). Causes and correlates of prison inmate misconduct: A systematic review of the evidence. Journal of Criminal Justice, 42, 462–470.

46.

Sweeten

(2006). Who will graduate? Disruption of high school education by arrest and court involvement. Justice Quarterly, 23(4), 462–480.

47.

Sweeten

Bushway

S. D.

Paternoster

(2009). Does dropping out of school mean dropping into delinquency? Criminology, 47(1), 47–91.

48.

Sykes

G. M.

(2007). The society of captives: A study of a maximum security prison. Princeton University Press.

49.

Tasca

Griffin

M. L.

Rodriguez

(2010). The effect of importation and deprivation factors on violent misconduct: An examination of Black and Latino youth in prison. Youth Violence and Juvenile Justice, 8(3), 234–249.

50.

Tewksbury

Connor

D. P.

Denney

A. S.

(2014). Disciplinary infractions behind bars: An exploration of importation and deprivation theories. Criminal Justice Review, 39(2), 201–218.

51.

Thornberry

T. P.

(1987). Toward an interaction theory of delinquency. Criminology, 25, 863–891.

52.

Trulson

C. R.

(2007). Determinants of disruption: Institutional misconduct among state-committed delinquents. Youth Violence and Juvenile Justice, 5, 7–34.

53.

Trulson

C. R.

DeLisi

Marquart

J. W.

(2011). Institutional misconduct, delinquent background, and rearrest frequency among serious and violent delinquent offenders. Crime & Delinquency, 57(5), 709–731.

54.

Tyler

T. R.

(2015). Why trust matters with juveniles. American Journal of Orthopsychiatry, 85(6, Suppl.), S93–S99.

55.

Walters

G. D.

Crawford

(2013). In and out of prison: Do important factors predict all forms of misconduct or just serious ones. Journal of Criminal Justice, 41, 407–413.

56.

Ward

Williams

(2015). Does juvenile delinquency reduce educational attainment? Journal of Empirical Legal Studies, 12(4), 716–756.

57.

Wolff

Blitz

C. L.

Shi

Siegel

Bachman

(2007). Physical violence inside prisons: Rates of victimization. Criminal Justice and Behavior, 34(5), 588–599.

58.

Zabel

R. H.

Nigro

F. A.

(2001). The influence of special education experience and gender of juvenile offenders on academic achievement scores in reading, language, and mathematics. Behavioral Disorders, 26(2), 164–172.