Skip to Content
Volume 6, Issue 1 • Spring 2017

Table of Contents

Editor's Note

Facility Operations and Juvenile Recidivism

Neighborhood Risks and Resources Correlated With Successful
Reentry of Youth Returning from Massachusetts Detention Centers

Girls Leaving Detention: Perceptions of Transition to Home After
Incarceration

An Innovative Use of Conjoint Analysis to Understand
Decision-Making by Juvenile Probation Officers

“I’d Prefer an Applicant Who Doesn’t Have a Delinquency History”:
Delinquents in the Labor Market

Gender Comparisons in the Processes and Outcomes of Functional
Family Therapy

Achieving Juvenile Justice Reforms Through Decision-Making Structures: The Case of Georgia

The Benefits of Community and Juvenile Justice Involvement in
Organizational Research

Neighborhood Risks and Resources Correlated With Successful Reentry of Youth Returning from Massachusetts Detention Centers

Lori A. Demeter, Public Policy and Administration, Walden University; Nokuthula Sibanda, Public Policy and Administration, Walden University.

Correspondence concerning this article should be addressed to Lori Demeter, Walden University, 100 Washington Avenue South, Suite 900, Minneapolis, MN 55401. E-mail: lori.demeter@waldenu.edu

Keywords: youth delinquency, recidivism, reoffending youth, recidivism factors, juvenile justice

Abstract

Youth delinquency is a major social problem in the United States, with approximately 29% of youth ages 18 to 21 reoffending within the first year of release in Massachusetts. The purpose of this quantitative, cross-sectional study was to examine whether the level of neighborhood risks, availability of jobs, availability of schooling, and availability of prosocial activities had an effect on recidivism rates. Publicly available data consisting of 347 youth ages 18 to 21 returning from statewide detention centers operated by the Massachusetts Department of Youth Services were analyzed using logistic regression. The results showed that neighborhood resources such as schooling and prosocial activities were statistically related with the rates of reoffending among youth reentering the community following incarceration. These results have important implications: Educators, law enforcement, and the community can benefit by collaborating to provide youth offenders with a special learning community that focuses on educating youth during and after release.

Introduction

Youth delinquency is a major social problem in the United States. According to Aizer and Doyle (2015), incarceration rates for juveniles have increased even faster than those of adults over the last 20 years. In 2014, juvenile courts handled approximately 1 million delinquency cases involving juveniles charged with criminal law violations (Office of Juvenile Justice and Delinquency Prevention, 2015). According to the Justice Policy Institute (2014), each year the United States incurs between $8 billion and $21 billion in long-term costs for the confinement of young people. It is estimated that the United States has a juvenile corrections rate five times higher than the next highest country (Aizer & Doyle, 2013). Further, taxpayers bear the financial burden of treating and incarcerating youth.

Abrams and Freisthler (2010) estimated that 200,000 youth transition back into their neighborhoods each year. Existing studies have focused on individual risk factors, problem behaviors, and negative peer associations of youth to determine the barriers that block a successful integration back to the community (Abrams & Freisthler, 2010; Anthony et al., 2010; Mendel 2011). However, this individual approach has failed to address risks posed by the context of the neighborhood to which they return. That is, little research has been conducted on the risk features of the neighborhood that the juvenile reenters and how these factors contribute to delinquent behavior and patterns of criminal activity. This paper’s research addresses the gap by exploring a neighborhood’s access to resources in mitigating neighborhood risks for reentry youth.

Literature Review

As offending youth return to their communities, they face many challenges that they must overcome to achieve successful reentry. When young people attempt to reintegrate into their communities, they are likely to return to the same situations that played a role in their delinquent behavior. For example, upon their return home, youth may be exposed to contact with delinquent and/or drug-using peers, dysfunctional parents or households, and opportunities for engaging in illegal behavior (Harder, Kalverboer, & Knorth, 2011). Furthermore, juveniles may encounter barriers that make it difficult for them to reintegrate back into the school system. For example, a youth’s reenrollment documentation may be incomplete. Some school district policies require that a youth produce documents that establish residency immunization status. If the detention center does not forward these documents and the youth is unable to provide them, the student may be denied enrollment (Feierman, Levick, & Mody, 2009). Moreover, a youth could experience discrimination within his or her community (Feierman et al., 2009); some members of the community are likely to judge the youth based on his or her previous delinquent behavior. Thus, the youth opts to keep a distance from the community rather than trying to fully reintegrate (Harder et al., 2011).

Given the high costs to society, communities, and the individuals themselves, it is essential to understand what happens to juveniles when they have been released from custody or when they return home after having spent time in a facility. Specifically, how do these youth who have come in contact with the justice system compare in terms of outcome measures such as employment and education?

According to Hartwell, McMackin, Tansi, and Bartlett (2010), data collected in Massachusetts indicated that 29% of youth discharged from the Department of Youth Services (DYS) supervision between the ages of 18 and 21 reoffend within the first year. In addition, research shows that approximately 50% of youth who are released from DYS violate the conditions of their release into the community (Hartwell et al., 2010). Similarly, in New York State, approximately 42% of youth who are released were rearrested within 6 months of their first release, and over 50% were rearrested within 9 months of their release (Hartwell et al., 2010). A study of a large juvenile detention system in the Southwestern U.S. states found that rearrest rates are as high as 85% at 5 years post-release (Abrams & Freisthler, 2010). Research on youth offender recidivism rates tend to show an overall decrease in reconviction after 2 years, indicating that the initial time period post-release is indicative of future arrest and conviction. Therefore, it is critical that attention be given to these initial days and months post-release (Hartwell et al., 2010; Tansi, 2009).

The research on barriers to successful transitions into mainstream society has focused on individual risk factors, problem behaviors, and negative peer associations among youth (Abrams & Freisthler, 2010; Anthony et al., 2010). Consequently, this individual approach has failed to address risks posed by the context of the neighborhood to which they return. Research has proven that neighborhood conditions play a role in contributing to delinquent behavior and patterns of criminal activity (Abrams & Freisthler, 2010). Patterns of criminal activity in neighborhoods also may be influenced by factors such as alcohol outlet density, availability of supportive services, or opportunities for youth to engage in prosocial behavior (Abrams & Snyder, 2010). Adequate conceptual and empirical research indicates that neighborhood influences have a more significant role in youth violence above individual risk factors for offending (Abrams & Snyder, 2010). Regardless of existing findings (Abrams & Freisthler, 2010; Abrams & Snyder, 2010; Anthony et al., 2010), theory and interventions on juvenile reentry have failed to acknowledge neighborhood factors as a key source of influence for reducing recidivism rates for juveniles (Abrams & Freisther, 2010).

A limited number of studies have sought to study neighborhood-level factors that affect the reentry experience and outcomes for adult offenders. Mellow, Schlager, and Caplan (2008) sought to understand whether there was a potential match or mismatch with the location of community services and the residences of adult parolees. This study found that the majority of services were located closer to parole district offices rather than the neighborhoods of adult parolees (Abrams & Snyder, 2010).

Little research has specifically focused on youth reentry. Abrams and Freisthler (2010) used archival data from postal codes in Los Angeles County, California, to analyze the associations between the level of neighborhood risks and resources and the success rates of youth returning to communities following incarceration. They concluded that rates of successful reentry were positively associated with neighborhood risks, such as density of off-premise alcohol outlets and level of community violence. Also, Harris and colleagues (2012) conducted a study aimed at investigating the effects of neighborhood and community-based programs and their impact in preventing juvenile recidivism. The authors analyzed data from adjudicated juvenile youth who had been assigned to court-ordered programs by the Family Court of Philadelphia, Pennsylvania. This study found that some types of juvenile offending were likely to be influenced by opportunities, constraints, and pressures present in the youth’s neighborhood. Although these studies contribute to the existing knowledge that neighborhood disadvantages play a significant role in the experiences and outcomes for offenders upon reentry, available research remains sparse, especially research focusing on juvenile offenders.

Problem Statement

Despite the increased theoretical evidence that neighborhood conditions may play a significant role in structuring success for high-risk youth, individual risk factors continue to dominate the focus of community reintegration of incarcerated youth. Research has indicated that when institutional resources that address the needs of community members are made accessible, neighborhood risks decrease. More specifically, neighborhood resources offer reentry youth with support services that can mitigate risk of reoffending, such as programs that provide school and job placement assistance and recreation centers (Abrams & Freisthler, 2010). However, the positive benefits linked to use of social services for reentry youth has not been confirmed empirically (Anthony et al., 2010). Further research is needed to support the notion that access to resources mitigates neighborhood risks for reentry youth.

Purpose of Study

The purpose of this quantitative, cross-sectional correlational study was to narrow the gap in current knowledge regarding youth reentry by examining which neighborhood risks and resources are related with rates of successful reentry of youth returning from statewide detention centers operated by the Massachusetts DYS, including the following objectives:

  1. To examine the relationships between the level of neighborhood risks and rates of reoffending among youth reentering the community following incarceration.
  2. To examine the relationships between environment resources such as availability of jobs, schooling, prosocial activities, and rates of reoffending among youth returning to the community.

Research Questions and Hypotheses

This study evaluated the following research questions and their corresponding hypotheses:

RQ1: To what extent, if any, does a relationship exist between the level of neighborhood risks and the rates of reoffending among youth reentering the community following incarceration?

Ho1: There is no relationship between the level of neighborhood risks and the rates of reoffending among youth reentering the community following incarceration.

Ha1: There is a relationship between the level of neighborhood risks and the rates of reoffending among youth reentering the community following incarceration.

RQ2: To what extent, if any, do relationships exist between availability of jobs, schooling, and prosocial activities and rates of reoffending among youth returning to the community?

Ho2a: There is no relationship between availability of jobs and rates of reoffending among youth returning to the community.

Ha2a: There is a relationship between availability of jobs and rates of reoffending among youth returning to the community.

Ho2b: There is no relationship between availability of schooling and rates of reoffending among youth returning to the community.

Ha2b: There is a relationship between availability of schooling and rates of reoffending among youth returning to the community.

Ho2c: There is no relationship between availability of prosocial activities and rates of reoffending among youth returning to the community.

Ha2c: There is a relationship between availability of prosocial activities and rates of reoffending among youth returning to the community.

Theoretical Framework

This study was based on two theoretical frameworks: collective efficacy and routine activities theory. Collective efficacy theory stems from the hypothesis that “neighborhoods with high levels of social cohesion and community assets are better equipped to contain individual risks for delinquency and youth violence” (Abrams & Snyder, 2010, p. 10). For example, such institutions include but are not limited to libraries, schools and other learning centers, child care, organized social and recreational activities, medical facilities, family support centers, and employment opportunities (Sampson, Morenoff, & Gannon-Rowley, 2002).

Routine activities theory focuses on the circumstances in which offenders commit criminal acts rather than emphasizing the characteristics of the offender (Cohen & Felson, 1979). Thus, an individual’s behavior patterns influence where crimes occur. This study employed these two frameworks to understand whether risks or supports associated with a neighborhood to which a youth must reenter can support or deter successful transition.

Research Method and Design

Statewide data from the Massachusetts DYS were used to measure the constructs considered in this study. The unit of analysis was based on the ZIP Codes to represent the areas in Massachusetts. Each ZIP Code was considered as one sample. Secondary data were used to measure the rates of reoffending among youth as well as the level of neighborhood risks and resources available in each area.

A quantitative correlational research design was used to examine the relationship between the level of neighborhood risks and the rate of reoffending among previously incarcerated youth in Massachusetts. A nonexperimental, cross-sectional quantitative correlational research design was deemed to be appropriate for the study, because the focus was on identifying potential relationships between identified variables (Babbie, 2012). Therefore, the study was not concerned with cause-and-effect relationships between variables. Instead, the focus was on investigating linear relationships between two or more variables (Babbie, 2012). For the purpose of this study, the level of neighborhood risks as well as the rate of reoffending were considered as dichotomous variables. On the other hand, categorical variables, such as race or gender, are variables where the output is not a number or where the number used in the analysis does not align with a value of the variables. The availability of jobs, schooling, and prosocial activities were also considered as categorical variables.

Secondary data from Massachusetts were used to measure the rate of reoffending of incarcerated youth who were released from custody in 2008. This was the most recent available sample from the Massachusetts DYS. In addition, secondary data of crime rates and risks were used to measure the level of neighborhood risks and the availability of jobs, schooling, and prosocial activities (Marshall & Rossman, 2006). Because the focus of this study was to examine the relationship between the independent variables of the level of neighborhood risks and resources and the dependent variable of the rate of reoffending, a correlational research design was most appropriate (Bryman, 2012).

Target Population and Sampling

Secondary data were used to measure the rates of reoffending among incarcerated youth in each of the areas as identified through publicly available data for each ZIP Code. Secondary data were also used to measure the level of neighborhood risks and the resources available in each of the areas using crime rates and risks data.

This research study used correlation analysis and independent samples t-tests (Babbie, 2012). Correlation analysis was used for research questions that considered the level of neighborhood risks as the independent variable because both the dependent and the independent variables were continuous in nature (Cozby & Bates, 2011). Independent samples t-tests were used for research questions focused on the resources available in each area, because the independent variable involved two independent groups (Cozby & Bates, 2011).

The minimum sample size of 128 was determined through several factors. The first factor was the effect size, which provides a measure on the strength of the relationship between variables. For the purpose of this study, a medium effect size was used to ensure that the assessment was not too strict nor too lenient (Cozby & Bates, 2011). Another factor considered in the identification of the minimum sample size was the power of the analysis; a standard of 80% power is used for statistical analyses. Moreover, a significance of .05 was used in this study. In the end, 347 cases fit the requirements of the study’s parameters and were included in analysis.

Definition of Variables

The variables considered in this study were defined based on the following:

Data Collection Procedures

After approval was obtained from the Walden University Institutional Review Board (IRB 05-05-14-0262233), a letter of intent to conduct the study was sent to the archival office of the Commonwealth of Massachusetts. Data were obtained through electronically transmitted data from the archival office of the Commonwealth of Massachusetts. Further data on levels of neighborhood risks and resources was obtained from Location Inc., an organization that generates reports on crime rates and risks within an area. Crime data specifically on ZIP Codes from the Commonwealth of Massachusetts were also electronically transmitted.

Data Analysis Procedures

The data collected from participants were entered into the Statistical Package for the Social Sciences (SPSS) 19.0 software. The data gathered were examined through descriptive statistics and inferential statistics. Categorical data were coded using numerical representations to ensure that these could be analyzed through statistical analyses. Descriptive statistics were used to describe the area in Massachusetts that was considered in this study. Descriptive statistics such as measures of central tendency were also used to describe the data gathered for this study. Frequency and percentages were used to describe categorical data, whereas measures of central tendencies such as the mean, standard deviation, and range were used to describe continuous variables such as the rate of reoffending of incarcerated youth and the level of risks and available resources within the area. Later inferential statistics such as the correlational analysis and independent samples t-tests were conducted to assess the relationship between the level of neighborhood risks and the rate of reoffending among incarcerated youth, as well as between the availability of resources such as jobs, schooling, and prosocial activities and the rate of reoffending among incarcerated youth.

To address the first research question, a correlation analysis was considered, because both the independent and the dependent variables were continuous variables (Cozby & Bates, 2011). If a significant correlation existed, considering a significance level of .05, then it could be concluded that there was sufficient evidence to reject the first null hypothesis that was posed in this study. For the second research question, independent samples t-tests were conducted to assess whether the independent variables of availability of jobs, schooling, and prosocial activities could significantly relate to the rate of reoffending among incarcerated youth. The independent samples t-tests determined whether there was a significant difference between the rates of reoffending among incarcerated youth based on the availability of jobs, schooling, and prosocial activities. A significance level of .05 was used for all statistical analyses.

Results

The sample of the study consisted of 347 youth ages 18 to 21 returning from detention centers in the Commonwealth of Massachusetts. The demographic information was summarized using frequency and percentages statistics. The summaries of the demographic information of gender, race, and reconviction rate are shown in Table 1. Table 1 illustrates that most of the 347 youth in the sample were male (325, or 93.7%). Regarding race, almost half or 160 (46.1%) were Race 1 (Caucasian), 86 (24.8%) were Race 2 (African American), and 85 (24.5%) were Race 3 (Hispanic).

Table 1. Frequency and Percentage Summaries of Demographic Information of Gender, Race, and Reconviction Rate

 

Frequency

Percentage

Gender

 

 

F

22

6.3

M

325

93.7

Race

 

 

1

160

46.1

2

86

24.8

3

85

24.5

4

7

2

5

9

2.6

Reconviction

   

No reconviction

204

58.8

Reconviction

143

41.2

In terms of reconviction or reoffending among youth returning to the community, 143 (41.2%) of the 347 youth were reconvicted. The 347 youth came from a total of 101 cities, including Boston (41, or 11.8%), Springfield (38, or 11%), Worcester (38, or 11%), New Bedford (18, or 5.2%), Fall River (14, or 4%), and Brockton (12, or 3.5%).

Table 2 summarizes the descriptive statistics of the continuous measured independent variables of level of neighborhood risks, availability of jobs, availability of schooling, and availability of prosocial activities. The descriptive statistics include the measures of central tendency of mean and standard deviations. The level of neighborhood risk was measured using the total crime index. The total crime index obtained the ratio between the total number of both violent and property crimes per 100,000, with higher values meaning more crimes were committed in a neighborhood. The mean level of neighborhood risk was 27.53, with the level of neighborhood risk among the cities the youth were from ranging from 3.29 to 74.32. In terms of the available resources within the area, the mean values showed greater availability of prosocial activities (M = 14.89) compared with availability of schooling (M = 11.43) and of jobs (M = 6.64). The least resource availability was the number of jobs (M = 6.64).

Table 2. Descriptive Statistics of Study Variables

 

N

Minimum

Maximum

Mean

Std.
Deviation

Level of neighborhood risk (Total crime index)

347

3.29

74.32

27.53

17.58

Availability of jobs

347

1.00

21.00

6.64

6.07

Availability of schooling

347

1.00

26.00

11.43

7.73

Availability of prosocial activities

347

2.00

37.00

14.89

9.48

A logistical regression model was created to determine the relationships of the independent variables of level of neighborhood risks, availability of jobs, schooling, and prosocial activities, and the dichotomous dependent variable of rate of reoffending among incarcerated youth in Massachusetts. The logistic regression was used, since the dependent variable of rates of reoffending among youth reentering the community following incarceration is a binary variable coded as no reconviction (0) or reconviction (1). The analysis sought to determine whether the independent variables of level of neighborhood risks, availability of jobs, schooling, and prosocial activities predicted whether a youth reoffends following reentry back into the community following a period of incarceration. A level of significance of 0.05 was used in the hypothesis testing.

First, the ratio of the valid cases to independent variables for logistic regression was investigated. The minimum ratio of valid cases (n) to independent variables for logistic regression should be 10 to 1, and the preferred ratio should be 20 to 1. The generated logistic regression model had 347 valid cases and 4 predictor variables (4 independent variables). The ratio of cases to the predictor variables was 86.75 to 1. The ratio satisfied the minimum requirement while also satisfying the preferred ratio of 20 to 1. Therefore, the logistic regression could be conducted since the minimum ratio of valid cases was satisfied.

The first model generated was a null model that did not include independent variables. This model was generated to create a baseline to compare predictor models. Table 3 summarizes the statistics for the equations of the variables not included in the null model. These were the independent variables of level of neighborhood risk (Score [1] = 0.07, p = 0.79), availability of jobs (Score [1] = 0.14, p = 0.71), availability of schooling (Score [1] = 1.25, p = 0.27), and availability of prosocial activities (Score [1] = 0.02, p = 0.88). The probability value of the overall statistics of the regression model, not including the four independent variables, was insignificant (Score [4]= 6.66, p = 0.16), implying that each of the four independent variables did not have any significant effect on the dependent variable when they were included in the null model.

Table 3. Variables Not in the Equation for Null Model

 

Score

df

Sig.

Step 0

Variables

Level of neighborhood risk

0.07

1

0.79

Availability of jobs

0.14

1

0.71

Availability of schooling

1.25

1

0.27

Availability of prosocial activities

0.02

1

0.88

Overall statistics

6.66

4

0.16

The second model generated was the Block 1 logistic regression model and included the entry of the four independent variables: level of neighborhood risks, and availability of jobs, schooling, and prosocial activities. The purpose of the second model was to determine which among the four independent variables significantly influenced the dependent variable of rates of reoffending when included in the model. The results of the overall test for the second model including the control variables are summarized in Table 4. The chi-square test was conducted to test the model to determine the existence of a significant relationship between the independent variables and the dependent variable. The probability value of the chi-square test (χ2 [4] = 6.72, p = 0.15) was greater than 0.05, indicating that the model was insignificant. The results suggested that the overall effects of the four independent variables on the dependent variable were insignificant. That is, results failed to support any effect of the independent variable on the dependent variable.

Table 4. Omnibus Tests of Model Coefficients for Logistic Regression with Independent Variables

 

Chi-square

df

Sig.

Step 1

Step

6.72

4

0.15

Block

6.72

4

0.15

Model

6.72

4

0.15

Table 5 summarizes the accuracy rate for the controlled logistic regression involving the independent variables. The accuracy rate computed by SPSS was 58.5%. Therefore, only 58.5% of the influences of the independent variables on the dependent variable were captured in the model.

Table 5. Classification Accuracy Rate for Controlled Logistic Regression with Independent Variables

 

Observed

Predicted

 

Reconviction

Percentage Correct

 

No Reconviction

Reconviction

Step 1

Reconviction

No Reconviction

182

22

89.2

Reconviction

122

21

14.7

Overall percentage

   

58.5

Note. The cut value is .500.

Table 6 summarizes the results of the significance of the logistic regression and the coefficients of the variables in the equation of the logistic regression. The analysis of this statistic determined the influence of the independent variables of level of neighborhood risks and availability of jobs, schooling, and prosocial activities on the dependent variable of rates of reoffending among youth reentering the community following incarceration. The coefficients, standard errors, Wald test statistic with associated degrees of freedom, and p values, as well as the exponentiated coefficient (also known as an odds ratio), are enumerated in Table 6. The relationship between the independent and the dependent variables is stronger when the deviation of the odds is farther from one (Cozby & Bates, 2011). A level of significance of 0.05 was used in the statistical testing. Statistical significance of the statistics would mean the rejection of Null Hypothesis 1, that there is no relationship between the level of neighborhood risks and the rates of reoffending among youth reentering the community following incarceration; Null Hypothesis 2a, that there is no relationship between availability of jobs and rates of reoffending among youth returning to the community; Null Hypothesis 2b, that there is no relationship between availability of schooling and rates of reoffending among youth returning to the community; and Null Hypothesis 2c, that there is no relationship between availability of prosocial activities and rates of reoffending among youth returning to the community. This would then suggest that there was a statistically significant relationship between the independent variables and the dependent variable.

Table 6. Variables in the Equation for Controlled Logistic Regression with Independent Variables

 

B

SE

Wald

df

Sig.

Exp(B)

Step 1a

Level of neighborhood risk

0.01

0.01

0.42

1

0.52

1.01

Availability of jobs

0.02

0.04

0.34

1

0.56

1.02

Availability of schooling

0.07

0.03

5.35

1

0.02*

1.07

Availability of prosocial social activities

−0.07

0.03

4.70

1

0.03*

0.93

Constant

−0.45

0.27

2.87

1

0.09

0.64

Note. Variable(s) entered on Step 1: Level of Neighborhood Risk, Availability of Jobs, Availability of Schooling, Availability of Prosocial Activities.

*Significant at level of significance of 0.05.

The results showed that the Wald statistic for the two independent variables of availability of schooling (Wald [1] = 5.35, p = 0.02) and availability of prosocial activities (Wald [1] = 4.70, p = 0.03) were significant. The results suggested that the availability of schooling and prosocial activities significantly influenced the dependent variable of rates of reoffending among youth reentering the community following incarceration, as the p-value was less than the level of significance value of 0.05. The results supported the rejection of Null Hypothesis 2b, that there is no relationship between availability of schooling and rates of reoffending among youth returning to the community; and Null Hypothesis 2c, that there is no relationship between availability of prosocial activities and rates of reoffending among youth returning to the community. On the other hand, the independent variable of level of neighborhood risk (Wald [1] = 5.35, p = 0.02) was not significantly related to the rates of reoffending among youth returning to the community. With this result, the null hypothesis for research question one (there is no relationship between the level of neighborhood risks and the rates of reoffending among youth reentering the community following incarceration), was not rejected. In addition, Null Hypotheses 2a (there is no relationship between availability of schooling and rates of reoffending among youth returning to the community) was also not rejected.

The coefficient of the odds ratio statistic of Exp(B) of the significant independent variables of availability of schooling and prosocial activities were investigated to determine changes in the log odds of the dependent variable for a one-unit increase in the availability of schooling and prosocial activities. The Exp(B) coefficient for availability of schooling was 1.07, which implied that a one-unit increase in availability of schooling increased the odds for the youth to be reconvicted (versus not being reconvicted) by 0.01 or 1.0%. The Exp(B) coefficient for availability of prosocial activities was 0.93, which implied that a one-unit increase in availability of schooling decreased the odds for the youth to be reconvicted (versus not being reconvicted) by 0.07 or 7.0%. The significant finding meant that the youth had a higher probability of being reconvicted if there were higher availability of schooling, since the Exp(B) coefficient was a positive value, and lesser availability of prosocial activities, since the Exp(B) coefficient was a negative value.

SPSS computed the accuracy rate for the controlled logistic regression with independent variables as 58.5%. The third model tested the significance of the logistic regression and the coefficients of the variables in the equation of the logistic regression. Based on the Wald’s statistic, availability of schooling (Wald [1] = 5.35, p = 0.02) and availability of prosocial activities (Wald [1] = 4.70, p = 0.03) were significant, meaning that both influenced the reoffending rate among youth entering the community after incarceration. However, the Wald’s statistic failed to reject the null hypothesis of Research Question 1, or the level of neighborhood risk (Wald [1] = 5.35, p = 0.02). A null hypothesis for Research Question 2 was also not rejected, nor was the influence of availability of jobs on reoffending rates. Therefore, the level of neighborhood risk and the availability of jobs do not influence reoffending rates.

The coefficient of the odds ratio statistic of Exp(B) of the significant independent variables of availability of schooling (1.07) and prosocial activities (0.93) implied that a one-unit increase in availability of schooling increased the odds for the youth to be reconvicted (versus not being reconvicted) by 0.01 or 1.0%, whereas a one-unit increase in availability of prosocial activities decreased the odds for the youth to be reconvicted (versus not being reconvicted) by 0.07 or 7.0%. The significant finding meant that the youth have higher probability of being reconvicted if there is higher availability of schooling due to a positive Exp(B) coefficient, and lesser availability of prosocial activities due to a negative Exp(B) coefficient.

Discussion

The findings of the study offer insight regarding how neighborhood risks and availability of resources such as jobs, schooling, and prosocial activities influenced the rate of reoffending by incarcerated youth in Massachusetts. The results show that the level of neighborhood risks do not affect recidivism, contrary to the findings of Anthony and colleagues (2010) that youth returning to an urban neighborhood face higher recidivism rates due to increased crime rates. Meanwhile, the availability of jobs do not have a significant effect on reoffending rates, which is consistent with the general observation that employment is not a factor in recividism. On the other hand, the availability of schooling increases the likelihood that juveniles would commit a crime during the integration period. Conversely, the presence of prosocial activities decreases the chances that juveniles reoffend. These results reveal the stark reality that the kind of community that is sought for the reintegration of juveniles affects whether a juvenile will be reconvicted, similar to theories and studies in current literature (Abrams & Freisthler, 2010; Abrams & Snyder; 2010). Decreasing the recidivism rate benefits the youth, because young people who have been sentenced to adult correctional facilities face a higher chance of physical and sexual assault while in prison as well as increasing recidivism rates (Carmichael, 2010).

Reintegrating juvenile delinquents back into educational institutions poses numerous challenges, as noted by current research. Some of these problems may help explain the inverse relationship between the option of schooling and the chances of reoffending. Sedlak and Bruce (2010) and Abrams and Synder (2010) blamed educational neglect, learning disabilities, and poor school records as the culprits for an unsuccessful reintegration. The 1992 Juvenile Justice Delinquency Prevention Act mandated that detained juveniles should receive proper educational opportunities, but 75% of facilities housing juveniles violated regulations that offer educational opportunities to these individuals (Braithwaite et al., 2010). One such program was the Individualized Education Program (IEP), which sought to address academic needs of youth while incarcerated. However, it was argued that the proper transfer process may not have been communicated to the juveniles upon release. It was possible that at the onset, the juveniles did receive adequate education to enable them to keep up with their peers who were not incarcerated. However, the juveniles may not have received the appropriate support during their incarceration, making it difficult for them to transition back into the educational system.

It was also possible that the juveniles did receive education, but it was not on par with the quality of education their peers received. The confines of the prison cell would also make it difficult for these juveniles to grow maturely without proper guidance, thus making it hard for them to have the emotional stability to deal with the challenges of the educational system outside the cell. In fact, Hatcher, Maschi, Rosato, and Schwalbe (2008) discovered that youth with serious emotional disturbance represent around 5% of a school population, making it difficult for the educational system to coordinate educational services with youth involved with the juvenile justice system.

The juveniles may also be discriminated against when trying to reintegrate with schools. Previously incarcerated youth would bear the stigma of being potential criminals and are thought to be more educationally deficient than other youth. This educational deficiency among detained youth may significantly affect delinquent behavior. Theriot (2009) offered a recommendation on how to address this concern through increased scrutiny of special education services offered through juvenile detention facilities.

Another challenge that these youth face in reintegration with the educational system stems from the schools themselves. As explained by Goldkind (2011), since these youth are generally sent back to the same educational institutions they attended before incarceration, the schools may be apprehensive about reenrolling students who have returned from mandated placements. However, there is some merit in why schools may not consider the reenrollment of these students: Negative experiences from reenrolling students, ensuring the safety of current students, and the educational gap between the two groups may negatively hurt the school’s performance (Goldkind, 2011).

Given this analysis, current educational leadership should evaluate the kind of school environment that juvenile delinquents are placed back into. Since the results of this study show that going back to school increases the likelihood that a youth will reoffend, the school environment may not be conducive to helping previously incarcerated youth to get back on the right track. Therefore, it is up to educational leadership to help create a school environment that is both accepting and supportive of these youth to bridge the educational gap and to aid these students in maturing as individuals. These environments are extremely important for juveniles who have been diagnosed with mental health disorders. The literature shows that a majority of youth in detention centers have mental health issues (Grande et al., 2012). Hence, providing mental health treatment to these juveniles would increase the chances of a smoother transition back to the educational system.

An offshoot of the hardships in integrating back into the educational system is that roughly 20% of youth who have been detained do not earn their GED or high school diploma (Osgood, Foster, & Courtney, 2010). The lack of this arguably basic requirement for employment makes the job opportunities available for these individuals very dim. Fewer job prospects may increase the likelihood of committing crime to meet basic costs of living. Despite this theory, the availability of jobs did not have a significant impact on the recidivism rates of these youth. This is particularly interesting, because this finding goes against the argument that unemployment would push people to a life of crime. A possible explanation is that having a job is not one of the goals of these youth, since they know they need to finish their education first before thinking about getting a job. Alternatively, perhaps they do not bother to look for a job because the majority of employers hire more skilled and formally educated peers. Abrams and Snyder (2010) theorized that youth with minimal work skills and little prior work history have trouble obtaining employment. In addition, as with the dilemma that schools face in reenrolling delinquents, employers may be apprehensive in hiring previously detained youth, which could add to the apathy of these youth regarding employment. Legal barriers are also present that prohibit the employment of ex-offenders. For example, a majority of U.S. states also allow employers the full right to deny employment to applicants who have a criminal record (Spjeldnes & Goodkind, 2009). This scenario poses a challenge for educators to put more focus on employment alongside education and social support services, as mentioned by Harder and colleagues (2011). In addition, youth experience a more successful transition when education is linked with community-based social service agencies other than mental health services or parole (Harder et al., 2011).

Prosocial activities were shown to decrease the likelihood of reoffending among youth. This finding is similar to current studies that argue that programs are successful when they prevent youth from engaging in delinquent behavior (David-Ferdon & Simon, 2012; Greenwood, 2008). These examples include community-based programs; school-based programs; home-visiting programs that focus on engagement, establishment, and maintenance of new patterns of family behavior; treatment of youth with serious clinical problems; collaborative planning; and problem solving. These types of programs engage youth with the community that they are trying to integrate with and make them feel part of the community. Programs such as functional family therapy, multisystemic therapy, and school violence prevention and maintenance programs have been found to be successful in decreasing criminal behavior by improving family functioning and decreasing the association with deviant peers, thereby creating positive outcomes for juvenile offenders. Positive relationships between students and their peers, teachers, and families can be critical assets in promoting youth’s well-being and preventing school violence. For instance, many school-based violence prevention programs improve the student body’s social skills and problem-solving abilities, which can result in more positive peer and student-teacher relationships throughout the school. Some school-based programs also help students know how to appropriately and safely intervene to stop an escalating violent episode between peers (David-Ferdon & Simon, 2012; Greenwood, 2008; Henggeler & Schoenwald, 2011). These programs may also be applied to the juveniles from Massachusetts described in this study. However, it should be noted that programs that focused on the individual offender have not been successful (Greenwood, 2008), perhaps because they do not offer the necessary social stimulation for youth to interact with individuals from their neighborhood.

In relation to the theoretical construct, the findings support the idea of Abrams and Snyder (2010) that neighborhoods with high levels of cohesion and community assets can decrease individual risks with regard to delinquency and youth violence. This study only investigated the effects of availability of jobs, schooling, and prosocial activities as neighborhood risks. However, these factors are far from the only ones that should be considered when assessing the quality of the neighborhood that a juvenile should be introduced to after incarceration. Such factors as noted in the literature include density of off-premise alcohol outlets and level of community violence (Abrams & Freisthler, 2010; Anthony et al., 2010).

Limitations of the Study

Several study limitations should be noted. The first limitation regards the applicability of the results. Since the study only considered data from Massachusetts, results cannot be generalized to a greater population, especially for those with different racial or ethnic compositions. The results of the study will only be generalizable to the population group of incarcerated youth within the Commonwealth of Massachusetts.

The second limitation was on the accuracy of ZIP Codes in identifying the immediate neighborhood of the offender. It was possible that the participants have already transferred to another ZIP Code without even knowing or identifying the proper authorities of the transfer. It is assumed that all data received were accurate; the researcher was not the one who collected the data personally from the samples, but the data were obtained using secondary data collection. Using a cross-sectional design also presented a limitation for the study. Cross-sectional research is commonly used to collect self-reported data from a particular group or population at the same time or within close proximity (Cozby & Bates, 2011). However, the data examined were from one period, not a longitudinal examination; therefore, it was not possible to evaluate potential trends or changes due to the fluctuating availability of programs. The third limitation was on the amount of available resources. Since the data obtained would be from the social service directory for each study area, the data might not capture all the available resources for the area. It was assumed that all data obtained from the social service directory were complete and accurate, given that the researcher was not the one who originally collected the data.

Another limitation is that this study only considered youth who had been reconvicted of new crimes, rather than those youth who returned to detention centers for technical violations. This caused the actual recidivism rate to be lower and thus limits the applicability of this study’s results. Similarly, the usual criteria for a recidivism study is a minimum of 2 years post-release. Since this study only looked at 1 year post-release, this might affect the reliability of the results.

Lastly, the reconviction rate found in this study is likely lower due to some youth aging out of the DYS system and therefore not being accounted for during the year of follow-up.

Implications

The results of this study offer insight into the issue of reoffending youth and support further exploration of issues that potentially affect the reoffending rates of juveniles who have been released from incarceration. This study also suggests that educators improve the quality of the schools in which the returning juveniles are placed. This would address the problem of increasing reoffending rates among youth due to unavailability of adequate schooling.

It is also recommended that educators and law enforcement consider placing these students in a special learning community that educates, guides, and supports them without the confinement of a detention center. This would allow a more personalized and collaborative exchange between youth and their teachers, amplifying the likelihood that youth would relate to positive role models. To facilitate a better transition process, educators should endeavor to have a more freely flowing exchange of information and communication between schools and detention centers. This would alleviate the challenges related to processing the academic records of the juvenile delinquent. An example would be a dedicated cell-to-classroom coordinator (CCC) who focuses on a seamless handling of educational system reintroductions. The CCC would be tasked with gathering educational data about the youth and matching the youth’s skills and competence to the right grade level.

Since the results showed that neighborhood schooling and prosocial activities available were significantly related to the rates of reoffending among youth reentering the community following incarceration, policies should be considered that increase the number of available adequate schools and the number of prosocial activities. Although some of this may fall within the purview of educators and law enforcement, there is a responsibility of legislators to further explore, enact, and fund such policies as well as lower recidivism rates to benefit society as a whole. Safer communities may result, and tax burdens may actually be alleviated by a lesser need to house, rehabilitate, and reintroduce former delinquents back into society.

Recommendations

The scope and limitations of the study have been focused on youth returning to their neighborhoods following incarceration in Massachusetts. Given that this study is a preliminary exploration, it would be insightful for future researchers to widen the scope of the study, analyze individuals from other states, or change the composition of the participant groups to contribute to knowledge on the factors that influence youth to become reoffenders. As a result, the researchers would like to recommend the following activities.

Building on the theoretical construct of this study, it would be important to understand the neighborhoods of juveniles in other states, since it is highly likely that there are significant differences among the various state environments. The analysis may also be extended to include how demographics coupled with neighborhood resources play a role in discouraging reoffending. This would allow a better allocation of resources toward programs that would suit a juvenile in a specific kind of neighborhood.

Another recommendation is to gather firsthand information from reoffenders on the factors that led them to incarceration after being reintroduced to their communities. Such a study would offer excellent insights into why juveniles are led to reoffend. Particular focus should be given on the quality of the neighborhood that the juvenile is put into, to further solidify or refute the results of the present study.

A final recommendation is to consider analyzing other factors related to neighborhood risks and their influence on the likelihood to reoffend. The study presented supplementary empirical research on the introductory understanding of neighborhood risks and juvenile delinquent reintegration. The seminal work done by Abrams and Freisthler (2010) already gave several examples of other possible factors. Further research is recommended to determine other neighborhood risks that may derail a successful reintegration process.

Conclusion

The results of the study support the finding that juvenile delinquents who have been released from incarceration in Massachusetts are more likely to reoffend due to issues related to availability of schooling, and less likely to reoffend due to availability of prosocial activities. These findings conclude that specific neighborhood risks are vital to the understanding of youth recidivism rates. A successful reintegration of youth poses numerous benefits to the individual and to society. Therefore, people in positions of influence over juvenile delinquents and the policies that guide them should consider developing and enforcing policies that increase the number of youth-centered prosocial activities in the community. Future research is recommended to examine a larger group within different geographic boundaries, include qualitative data analysis, and consider studying other neighborhood risk factors.

About the Authors

Lori A. Demeter, PhD, has taught over 50 courses (both traditional and online) in criminal justice, organizational behavior, public policy, and public administration. With nearly 20 years of experience working in all levels of government, she enjoys the field of public policy and particularly juvenile justice, because the consequences of such work produce far-reaching effects for generations to come.

Nokuthula Sibanda, PhD, serves as the deputy division director of Adolescent Services for Community Teamwork in Lowell, Massachusetts. She is also a technical assistance coach for Youthbuild USA, which provides education, vocational training, and supportive services to young men and women who have dropped out of high school and are at risk for entering the criminal justice system.

References

Abrams, L. S., & Freisthler, B. (2010). A spatial analysis of risks and resources for reentry youth in Los Angeles County. Journal of the Society for Social Work and Research, 1(1), 41–55. doi:10.5243/jsswr.2010.4

Abrams, L. S., & Snyder, S. (2010). Youth offender reentry: Models for intervention and directions for future inquiry. Children and Youth Services Review, 32(12), 1787–1795. doi:10.1016/j.childyouth.2010.07.023

Aizer, A., & Doyle, J. J. (2015). Juvenile incarceration, human capital and future crime: Evidence from randomly-assigned judges. Quarterly Journal of Economics, 130(2), 759–803.

Anthony, K. E., Samples, D. M., Kervor, D., Ituarte, S., Lee, C., & Austin, J. M. (2010). Coming back home: The reintegration of formerly incarcerated youth with service implications. Children and Youth Services Review, 32, 1271–1277. doi:10.1016/j.childyouth.2010.04.018

Babbie, E. (2012). The practice of social research (13th ed.). New York, NY: Wadsworth.

Braithwaite, R., De La Rosa, M., Holliday, C. R., Toldson, A. I., & Woodson, M. K. (2010). Academic potential among African American adolescents in juvenile detention centers: Implications for reentry to school. Journal of Offender Rehabilitation, 49, 551–570.

Bryman, A. (2012). Social research methods (4th ed.). New York, NY: Oxford University Press.

Carmichael, T. (2010). Sentencing disparities for juvenile offenders sentenced to adult prisons: An individual and context analysis. Journal of Criminal Justice, 38, 747–757. doi:10.1016/j.jcrimjus.2010.05.001

Cohen, L., & Felson, M. (1979). Social change and crime rate trends: A routine activity approach. American Sociological Review, 44(4), 588-608. Retrieved from http://www.jstor.org/stable/2094589. Accessed February 9, 2017.

Cozby, P., & Bates, S. (2011). Methods in behavioral research (11th ed.). New York, NY: McGraw-Hill.

David-Ferdon, C., & Simon, T. (2012). Striving To Reduce Youth Violence Everywhere (STRYVE): The Centers for Disease Control and Prevention’s national initiative to prevent youth violence foundational resource. Atlanta, GA: Centers for Disease Control and Prevention.

Feierman, J., Levick, M., & Mody, A. (2009). The school-to-prison pipeline and back: Obstacle and remedies for the re-enrollment of adjudicated youth. New York Law School Law Review, 54, 1116–1129. Retrieved from http://www.nylslawreview.com/wp-content/uploads/sites/16/2013/11/54-4.Feierman-Levick-Mody.pdf. Accessed February 2, 2017.

Goldkind, L. (2011). A leadership opportunity for school social workers: Bridging the gaps in school reentry for juvenile justice system youths. Children & Schools, 33(4), 229–239. doi:10.1093/cs/33.4.229

Grande, T. L., Hallman, J., Rutledge, B., Caldwell, K., Upton, B., Underwood, L. A., & Rehfuss, M. (2012). Examining mental health symptoms in male and female incarcerated juveniles. Behavioral Sciences & the Law, 30(3), 365–369.

Greenwood, P. (2008). Prevention and intervention programs for juvenile offenders. The Future of Children, 18(2),185–210. Retrieved from http://www.futureofchildren.org/publications/docs/18_02_09.pdf. Accessed February 2, 2017.

Harder, T. A., Kalverboer, E. M., & Knorth, J. E. (2011). Transition secured? A follow-up study of adolescents who have left secure residential care. Children and Youth Services Review, 33, 2482–2488. doi:10.1016/j.childyouth.2011.08.022

Harris, P., Mennis, J., Obradovic, Z., Izenman, A., Grunwald, H., Lockwood, B., . . . Chisholm, L. (2012). Investigating the simultaneous effects of individual, program, and neighborhood attributes on juvenile recividism using GIS and spatial data mining. [U.S. Department of Justice Research Report]. Philadelphia, PA: Temple University. Retrieved from https://www.ncjrs.gov/pdffiles1/nij/grants/237986.pdf. Accessed February 9, 2017.

Hartwell, S., McMackin, R., Tansi, R., & Bartlett, N. (2010). "I grew up too fast for my age": Postdischarge issues and experiences of male juvenile offenders. Journal of Offender Rehabilitation, 49(7), 495–515.

Hatcher, S. S., Maschi, T., Rosato, S. N., & Schwalbe, S. C. (2008). Mapping the social service pathways of youth to and through the juvenile justice system: A comprehensive review. Children and Youth Services Review, 30, 1376–1385. doi:10.1016/j.childyouth.2008.04.006

Henggeler, S.W., & Schoenwald, S. K. (2011). Evidence-based interventions for juvenile offenders and juvenile justice policies that support them. Sharing Child Youth Development Knowledge, 25(1), 1–28. Retrieved from http://www.mtfc.com/2011_EB_Interventions_for_Juv_Offenders.pdf. Accessed February 2, 2017.

Justice Policy Institute. (2014). Sticker shock: Calculating the full price tag for youth incarceration. Washington, DC: Justice Policy Institute. Retrieved from http://www.justicepolicy.org/uploads/justicepolicy/documents/sticker_shock_final_v2.pdf. Accessed February 2, 2017.

Marshall, C. & Rossman, B. (2006). Designing qualitative research (4th ed.). Thousand Oaks, CA: Sage.

Mellow, J., Schlager, M., & Caplan, J. (2008). Using GIS to evaluate post-release prisoner services in Newark, NJ. Journal of Criminal Justice, 36, 416–425.

Mendel, R. A. (2011). No place for kids: The case for reducing juvenile incarceration. Baltimore, MD: Annie E. Casey Foundation. Retrieved from http://www.voxeu.org/article/what-long-term-impact-incarcerating-juveniles. Accessed February 2, 2017.

Office of Juvenile Justice and Delinquency Prevention. (2015). OJJDP Statistical Briefing Book. Washington, DC: U.S. Department of Justice, Office of Justice Programs, Office of Juvenile Justice and Delinquency Prevention. Retrieved from http://www.ojjdp.gov/ojstatbb/crime/qa05101.asp?qaDate=2014. Accessed February 2, 2017.

Osgood, D. W., Foster, E. M., & Courtney, M. E. (2010). Vulnerable populations and the transition to adulthood. The Future of Children, 20(1), 209–229.

Sampson, R., Morenoff, J., & Gannon-Rowley, J. (2002). Assessing “neighborhood effects”: Social processes and new directions in research. Annual Review of Sociology, 28, 443–478.

Sedlak, A. J., & Bruce, C. (2010). Youth’s characteristics and backgrounds: Findings from the Survey of Youth in Residential Placement. Washington, DC: U.S. Department of Justice, Office of Justice Programs, Office of Juvenile Justice and Delinquency Prevention. Retrieved from https://www.ncjrs.gov/pdffiles1/ojjdp/227730.pdf. Accessed February 2, 2017.

Spjeldnes, S., & Goodkind, S. (2009). Gender differences and offender reentry: A review of the literature. Journal of Offender Rehabilitation, 48(4), 314–335. doi:10.1080/10509670902850812

Tansi, R. (2009). Juvenile recidivism report for clients discharged during 2005. Boston, MA: Massachusetts Department of Youth Services.

Theriot, M. T. (2009). School resource officers and the criminalization of student behavior. Journal of Criminal Justice, 37, 280–287.

 

OJJDP Home | About OJJDP | E-News | Topics | Funding
Programs | State Contacts | Publications | Statistics | Events