Abstract
Little is known about what factors contribute to African American youth desisting from offending. Participants were 3,230 moderate- to high-risk adolescents from Washington State who completed a statewide risk assessment to assess the likelihood of recidivism. Participants were screened by juvenile probation officers between 2003 and 2010. Researchers investigated whether youth possessed protective factors and whether developmental change took place after contact with the juvenile justice system. It was hypothesized that having protective factors would decrease the likelihood of recidivism and the impact of each factor would differ by gender. Findings indicate African American youth have protective factors across a range of domains. However, little developmental change occurs after contact with the juvenile justice system. Impulse control, parental supervision, and pro-social peers were important for reducing recidivism. Problem solving was more influential for African American males, while impulse control and parental supervision were more influential for African American females. Implications for practice and policy are discussed.
Introduction
The present longitudinal study focuses on factors that decrease the probability of re-arrest for African American youth in the juvenile justice system. African American youth were chosen particularly because they represent a significant portion of the juvenile justice population (Kempf-Leonard, 2012). Identifying protective factors that can interrupt the path from the juvenile system to the adult system is important because this knowledge can help increase the probability that these young people will lead successful lives. As males and females have unique experiences with the juvenile justice system (Chesney-Lind & Pasko, 2004), gender differences will also be examined to determine whether the influence of each protective factor varies by gender. Identifying youth who are able to optimize their strengths and adapt positively despite a first arrest, is an important issue with implications for the African American family, community, and society as a whole.
Literature Review
Conceptual Framework
The theory of resilience is utilized to determine how moderate- to high-risk African American youth can use their resources to overcome the adversity of being re-arrested. Masten (2001) conceptualizes resilience as having a positive outcome despite threats to positive development. Resilience is conceptualized in this study as not being re-arrested. Protective factors in the domains of cognitive processes, social skills, family, school, and relationships are explored to explain the mechanisms by which African American youth avoid recidivism in the juvenile justice system. Examining resilience utilizing both individual and contextual protective factors is important in advancing the field of resilience research (Masten, 2007). Protective factors are often utilized to understand the mechanisms of resilience.
Protective factors are assets within the individual or context that moderate the impact of adversity when risk is high (Wright & Masten, 2005). The contextual-level factors of particular interest in this study are parenting practices, school status, and relationships. Understanding contextual protective factors are important because the context in which children and youth develop is important for understanding both positive and negative developmental outcomes particularily for ethnic minority youth (Wong & Rowley, 2001).
Specific protective factors were selected based on prior research identified by Wright and Masten (2005) as being important in the resilience process. Risk and protective factors have been examined among juvenile offenders as a whole (Mulvey et al., 2010) and violent offenders versus non-violent offenders (Lodewijks, Ruiter, & Doreleijers, 2010). Scholars have examined risk and protective factors among African American youth regarding schooling (Burchinal, Roberts, Zeisel, & Rowley, 2008), psychopathology (Tandom & Solomon, 2008), low-income status (Gabalda, Thompson, & Kaslow, 2010), perceived racial discrimination (Harris-Britt, Valrie, Kurtz-Costes, & Rowley, 2007), and drug use (Belgrave, Clark, & Nasim, 2009). However, few empirical studies have examined protective factors for African American youth specifically in the context of the juvenile justice system. This study seeks to determine whether protective factors exist within this understudied population. In addition to addressing the conceptual importance of protective factors to reduce recidivism, there are also human capital and economic reasons why decreasing recidivism is important.
Societal Impact
African American youth are overrepresented at every level of the juvenile justice system (Piquero, 2008). The reported number of delinquent acts committed by African American, European American, and Latino youth is comparable (Snyder & Sickmund, 1999). However, African American youth are 1.4 times more likely to be detained for a crime (Snyder & Sickmund, 2006), twice as likely to be re-arrested (Synder, 2006), and more than twice as likely to be transferred to adult prisons after arrest (Juszkiewicz, 2000). Once involved with the juvenile justice system, many continue on to the adult criminal justice system (Johnson, 2004).
Involvement in the adult criminal justice system has devastating effects on African American families and communities because former felons face discrimination in employment, housing, and large-scale disenfranchisement (Alexander, 2010). As a result of this social and economic disadvantage, the consequence of criminal justice involvement makes it difficult, if not impossible, to participate in positive and productive ways within the family system, community, and society at large (Goffman, 2009).
Societal Cost of Incarceration
The human capital costs associated with incarceration are formidable. Incarcerated young men and women are often lost to society. Frequently, they are unable to ever become independent and productive members of the community and often struggle to successfully contribute to their families as children, partners, or parents (Goffman, 2009). There is also an economic argument why reducing recidivism is important. The United States currently incarcerates a higher percentage of its population than any other country in the world. The U.S. incarceration rate is now 240% higher than it was in 1980. This increase in incarceration has led to an increase in government spending with approximately US$75 billion on corrections in 2008 (Schmitt, Warner, & Gupta, 2010). All 50 states and the District of Columbia spent on average 2.8 times as much per prisoner as they did per public school student (Children’s Defense Fund, 2008).
Identifying protective factors that decrease the likelihood of African American youth already in the system is critical to reducing recidivism. Luthar and Cicchetti (2000) have noted it is important to identify adolescents’ protective processes within the appropriate social context so as to increase the likelihood of creating effective evidence-based interventions and social policies.
There is, however, little empirical research that has examined risk and protective factors in the United States specifically for African American youth specifically within the context of the juvenile justice system. The present study will address this gap in the literature and identify positive pathways to reduce the rate of recidivism. Research that examines protective factors within-groups among high-risk children is underrepresented within the literature (Vanderbilt-Adriance & Shaw, 2008). Here, we will also examine the role that gender plays in the discontinuity of offending.
Influence of Gender
The impact of the justice system on African American males is significant and is receiving increasing attention in the literature (Alexander, 2010; Goffman, 2009). A less known fact is the number of females in the justice population is steadily increasing (Chesney-Lind & Pasko, 2004). This is especially true for African American girls with arrests steadily increasing since 1980 (Children’s Defense Fund, 2011). The pathway for boys and girls in the juvenile justice system is different from the moment of police contact to the point of case closure (Chesney-Lind & Pasko, 2004).
Girls in the juvenile justice system are much more likely to come from family constellations where sexual, physical, and emotional abuse occur in the home (Le, Arifuku, & Nuñez, 2003). Girls are also more likely to be arrested for simple assaults such as status offenses compared to boys (Zahn et al., 2010). Status offenses are acts that would not be considered crimes if the youth were of legal age. Such acts include running away, being a person in need of supervision, beyond control, truant, or in need of care and protection (Chesney-Lind & Pasko, 2004). Although these studies have examined females generally, few have examined female African American youth particularly. As females come into the system for very different reasons as compared with males, it is important to identify whether protective factors work the same or differently among males and females. The present study seeks to contribute to the literature in an exploratory manner by examining whether these gender differences persist when looking exclusively at African American youth.
The Current Study
Three hypotheses are tested in the present study. First, it is hypothesized at the time of arrest, at least 7% of juvenile offenders will answer positively to having protective factors across the domains of attitudes and behaviors, skills, parenting practices, school status, and community relationships (see Table 1). These protective factors were selected as they have been demonstrated in the resilience literature to aid youth at risk of adversity to have a positive outcome despite the odds (Wright & Masten, 2005). The hypothesis of 7% is based on Born, Chevalier, and Humblet’s (1997) finding that 7% of the total juvenile population were determined to be resilient as not having committed a serious crime or multiple criminal acts over time. As little empirical research has been done regarding protective factors among African American juvenile offenders, we take 7% as a benchmark and seek to ascertain whether this figure is similar among this population specifically.
Protective Factors and Survey Items Selected From the Washington State Juvenile Court Assessment.
Second, it is hypothesized that there will be little change over time for each protective factor in the domains of attitudes and behaviors, skills, parenting practices, school status, and community relationships. As the assumed primary goal of the juvenile justice system is to rehabilitate youth (Scheifer, 1995), it is important to investigate whether rehabilitation is actually occurring within the system. If change within protective factors is taking place, it can be taken as evidence that the system is an effective intervention in rehabilitating troubled youth. However, if little change is taking place within the protective factors of youth, this may point to the need for system reform. Mulvey et al. (2010) examined the self-reported anti-social behavior of youth with court involvement over a 3-year period. The authors found that 14.6% of youth were part of the “desister” group, which started off with high levels of offending that steadily decreased over time. Given this finding, change over time for each factor would have to be greater than 14.6% to be considered significant change.
Third, it is hypothesized that protective factors at the time of the first assessment will decrease the likelihood of recidivism and the influence of each factor will differ significantly by gender. As females tend to come from family constellations with more sexual, physical, and emotional abuse (Chesney-Lind & Pasko, 2004; Le et al., 2003), it is hypothesized that factors that aid females in exiting the juvenile justice system will differ from those of males. Taking an appropriate social context into account, in this case race and gender, increases the likelihood of creating effective evidence-based interventions and social policies to improve the outcomes for at-risk groups (Luthar & Cicchetti, 2000).
Method
Data
Data for this study are derived from the Washington State Juvenile Court Assessment (WSJCA). The WSJCA is a modified version of Baird’s 1984 Wisconsin Risk Scale (WSIPP, 2004). It is utilized to determine placement security (minimum to maximum) and level of community supervision (low to intensive; Washington State Institute for Public Policy [WSSIP], 1999) for adolescent offenders in Washington State. The assessment is comprised of self-reported risk and protective factors and categorizes youth into a moderate- to high- or low-risk group. If assessed as moderate to high risk, the young person consequently takes the full WSJCA, which consists of 127 items. The full assessment identifies risk and protective factors across 12 domains. For the purpose of this study, attitudes and behaviors, skills, parenting practices, school status, and relationships were assessed in relation to recidivism. Protective factors were selected utilizing literature that focused on fostering resilience in the context of adversity through a developmental and ecological lens (Wright & Masten, 2005).
Participants
The sample consists of 3,230 moderate- to high-risk African American adolescent offenders who completed the full WSJCA between 2003 and 2010. Of the sample, 2,358 (73%) were males and 872 (27%) were females. The average age at the first assessment was 15.63 for males and 15.58 for females. These youth were taken out of the larger sample of all Washington State juvenile offenders (n = 25,447) who completed the full WSJCA. The dependent measure was a subsequent official report of subsequent arrest, either as a juvenile or adult.
Measures
In the present study, we focus on dynamic protective factors that have the potential for change. Each protective factor and survey question is displayed in Table 1. All questions in the survey were asked at an ordinal level of measurement. During the analysis process, we re-coded each variable into a dichotomous measure. A value of “1” is associated with having a protective factor, and a value of “0” is associated with lacking the protective factor or having it inconsistently. For example, one protective factor is parental supervision. If youth respond they have good supervision, this item is re-coded so that a value of “1” is associated with good parental supervision. If youth respond they receive sporadic or inadequate supervision, they receive a value of “0.” Arrest was coded as a dichotomous variable. A value of “1” indicated the youth received one subsequent arrest as a juvenile or adult.
Analytic Strategy
The first research question examines whether youth possess protective factors at entry into the juvenile justice system. Frequencies at Time 1 were examined across each protective factor to establish a baseline among youth. For example, whether a youth possessed the skill of dealing with others at the time of the first assessment will be determined. This information will determine whether youth have protective factors that can be built on through policy or intervention practices.
The second research question proposes little change over time within each protective factor. If little change is taking place within the protective factors of youth, this may point to the necessity of system reform within the juvenile justice system. Change-over-time variables were created between the completion of the first and second full-scale assessment to determine change. The frequencies of each variable were examined to determine whether any change had taken place as a result of contact with the juvenile justice system. One example of a change-over-time variable is whether the youth is close to school staff and whether this relationship changes over time. This item is broken down into four categories: (a) Youth did not have a close relationship with school staff at the first assessment, however did at the second assessment. Having a new close relationship with school staff at the second assessment signifies protection increased. (b) Youth who had a close relationship with school staff at Time 1 and Time 2 signifies protection maintained. (c) Youth who did not have a close relationship with school staff at Time 1 and Time 2 signifies risk maintained. (d) Youth who had a close relationship with school staff at Time 1 but not Time 2 signifies risk increased.
The third research question proposes having protective factors across individual and contextual domains will decrease recidivism, and the influence of each protective factor will differ by gender. The risk assessment is administered at the discretion of the judge for each individual youth after arrest depending on the unique situation of the youth. For example, all youth do not complete an assessment every 6 months. Therefore, there is no definite time period for when each individual re-takes the assessment. Consequently, the risk of being re-arrested will vary substantially between youth. The time period of observation is from 2003 to 2010. Therefore, a youth initially arrested in 2003 has a much higher risk of being re-arrested than a youth initially arrested in 2008 because of the difference in observation time. Survival analysis was used to adjust for these time discrepancies in the data. The calculated risk period is the number of days between the completed assessment and the final day of observation (January 22, 2010). Interactions with gender were created for each variable to determine the relative influence of each variable as compared with males as males were used as the reference group. Both these findings were determined through using an odds ratio. Age at time of arrest, felony status, special education status, and use of drugs and alcohol were used as controls. Each control variable was coded dichotomously in the direction of a protective factor. For example, in the case of special education status, not having a special education status would result in a value of “1.” IBM SPSS Statistics 19 was used to conduct the analyses.
Results
In the following section, we begin with basic descriptive analyses. The average time at risk of arrest was 5 years (1,827 days) for boys and 5.16 years (1,886 days) for girls. Youth demographics were examined across both sexes. With regard to offense type and likelihood of recidivism, males were statistically more likely to have a felony referral at 69% compared with females at 49%, χ2(1, 3230) = 111.4, p < .01. Females were more likely to have a misdemeanor referral at 59% compared with males at 50%, χ2(1, 3230) = 25.6, p < .01. Males were also more likely to have been re-arrested after the first assessment at 70% as compared with females at 61%, χ2(1, 3230) = 25.4, p < .01.
Protective Factors at Entry Into the Juvenile Justice System
Using the 7% benchmark as an indicator of whether African American youth possess protective factors, findings are mixed. African American youth do possess protective factors, though they differ widely across domains (see Table 2). With regard to attitudes and behaviors, 73% of youth had optimism, 94% belief in self-control, and 79% had positive attitudes toward pro-social rules at the time of the first assessment. Youth had relative low skills at entry into the system. Four percent had problem-solving skills, 27% had the ability to deal with others, and 9% had the ability to control impulsive behavior. Forty-four percent had good parental supervision while 26% had parental control. With regard to school status, 51% were close with a teacher, coach, or staff, while 28% were involved in structured activities. Seventy-four percent had pro-social friends, 10% were resistant to anti-social friends, and 84% felt support from extended family members. Little variation is observed by gender across each domain.
Protective Factors at Entry Into the Juvenile Justice System.
Change-Over-Time Protective Factors
Little change was observed over time as a result of contact with the juvenile justice system (see Table 3). In the area of attitudes and behaviors, the majority of youth responded at optimal levels for each protective factor. Findings were mixed at the contextual level. Few youth improved or declined on protective factors between the first and second assessments. The majority maintained their level of protection or risk. There is little variation by gender.
Descriptive Statistics for Change-Over-Time Protective Factors.
Relationship Between Protective Factors and Recidivism
A wide range of covariates were explored to build a parsimonious model. Those variables that did not contribute to the overall prediction were dropped if they did not significantly alter the estimates remaining in the model. Results from the Cox Regression are displayed in Table 4. The table includes the coefficient and standard error for each independent variable as well as the hazard ratio, Exp(β). In the main effect portion of the model, a hazard ratio less than 1 indicates the decreased likelihood of recidivism. If 1 is subtracted by the hazard ratio and multiplied by 100, the resultant is equal to the percentage change in the hazard of re-arrest. The main effects model indicates that age is important in calculating the probability of recidivism. For each increase in age, youth decreased their probability of being re-arrested by 11%. Youth without a special education reduced their probability of being re-arrested by 31%. With regard to protective factors, control of impulsive behavior decreased the probability of re-arrest by 48%. Parental supervision reduced the probability of recidivism by 27% while having pro-social friends reduced the odds of being re-arrested by 22%. Felony referral status and alcohol and drug use did not have a significant impact on recidivism. In the domain of attitudes and behaviors, optimism, belief in self-control, and attitudes toward pro-social rules did not make a significant difference for recidivism. Problem solving, dealing with others, and having parental control also did not significantly affect the probability of re-arrest. Protective factors related to school status, being resistant to anti-social peers, and having support from extended family did not have a significant impact on re-arrest.
Cox Regression: Impact of Protective Factors on Re-Arrest and Influence of Sex.
Indicates significance at the p < .05 level.
Indicates significance at the p < .01 level.
Influence of Gender
The interaction portion of the model demonstrates the influence of gender for each variable. Males were used as the reference group. A hazard ratio greater than 1 signifies the importance of the protective factor for females. A hazard ratio of less than 1 signifies the importance of the protective factor for males. Felony referral, alcohol and drug use, and problem solving had a greater influence for males than females. Impulse control, special education status, and parental supervision had a greater influence for females than males. Age did not have a greater impact by gender. At the individual level, optimism, belief in self-control, and attitudes toward pro-social rules did not differ by gender. Dealing with others, parental control, and protective factors within the domains of school status and relationships also did not significantly differ by gender.
Discussion
There is little research devoted to identifying protective factors among African American youth. The vast majority of research conducted on African American youth involves the identification of problem behaviors and race-comparative studies (McLoyd, 1998; Wong & Rowley, 2001). This article sought to address the critical gap in the literature and advance knowledge focused specifically on African American boys and girls in the juvenile justice system utilizing resilience theory. Identifying the strengths of African American youth can be beneficial in creating evidenced-based policies and practices to improve developmental outcomes for youth.
Protective Factors at Entry Into the Juvenile Justice System
It is evident that African American adolescents do have protective factors at entry into the juvenile justice system. Youth reported high levels of positive attitudes and behaviors at the time of entry, but low levels of skills, effective parenting, and school engagement. The results were mixed in the area of relationships. Both males and females were similar regarding strengths and weaknesses at the time of the first assessment. There may have been little difference between genders in this study because of the high-risk nature of the sample. Had all juvenile offenders arrested in the state of Washington been included in this sample, perhaps there may have been more variation across genders.
With regard to other protective factors, there were areas where the majority of youth responded in the affirmative and there appeared to be little room for improvement. Youth responded quite positively to having optimism, belief in self-control, positive attitudes toward pro-social rules, and having support from extended family. Perhaps these areas of strength can be built on to improve other areas where youth are weak within a certain context.
Change-Over-Time Protective Factors
Consistent with the second hypothesis, little change took place over time in the developmental trends of African American youth after coming into contact with the juvenile justice system. Overwhelmingly, youth remained stable in either the positive or negative direction across each domain. As one of the assumed major goals of the juvenile justice system is the rehabilitation of youth (Scheifer, 1995), one would expect there would be dramatic change taking place in the development of youth as a result of contact with the system. If change were taking place, this would highlight the efficiency of the selected intervention (i.e. imprisonment, probation, etc.) on rehabilitating troubled youth. Utilizing resilience theory, perhaps a more effective approach to rehabilitating troubled youth may be used. For example, enrolling youth in programs that build on the specific protective factors important in aiding the decrease of recidivism would be more effective. This study identified protective factors at both the individual and contextual levels that can be utilized in targeted interventions to specifically decrease the likelihood of recidivism, which are detailed below.
Relationship Between Protective Factors and Recidivism
Having the ability to control impulsive behavior had the greatest probability of reducing recidivism for all youth. Impulse control decreased the probability of recidivism by 48%. However, 85% of youth continued to lack this skill at the time of the second assessment according to our change-over-time variable. An intervention targeting the control of impulsive behavior may be a more effective intervention for juvenile offenders than imprisonment or probation.
The Promoting Alternative Thinking Strategies (PATHS) curriculum is one approach used to change impulse control among youth with deficiencies in executive functioning (Greenberg, 2006). The PATHS curriculum is a preventive neuroscience-based intervention that seeks to change cognitive processes related to inhibitory control, planning, problem solving, and emotional regulation. Students who used the PATHS curriculum improved significantly in the area of inhibitory control and teacher reported externalizing and internalizing behavior at 1-year follow-up.
Youth also demonstrated resilience at the contextual level. Having parental supervision and pro-social peers made recidivism less likely. This finding is supported in previous research where both parenting and peers were found to play a major role in explaining outcomes for serious juvenile offending (Chung & Steinberg, 2006). Multisystemic therapy (MST) is another evidence-based alternative intervention that addresses the issues of parental supervision and peers among delinquent populations. MST significantly decreases the likelihood of youth being re-arrested within 4 years of treatment and can reduce the seriousness of an offense if a crime is later committed (Borduin et al., 1995). This study also found that MST improved family cohesion, adaptability, and decreased family conflict and hostility. MST produced a significant change in addressing the underlying developmental needs of youth. These are only two of many evidence-based interventions that may decrease recidivism for youth.
Influence of Gender
Alcohol, drug use, and felony referral status were significantly important for males with regard to recidivism. There is an abundance of literature on the deleterious effects of drug policy on the overrepresentation of African American males in the justice system. These policies include increased police surveillance in low-income and outdoor spaces that are disproportionately African American (Beckett, Nyrop, & Pfingst, 2006; Kakar, 2006), differential court processing that increases the likelihood of imprisonment for African American youth (Kempf-Leonard, 2007), and the 100-to-1 disparity in the sentencing of crack versus powder cocaine implemented in the 1980s and mandatory minimum sentencing for simple drug possession (King & Mauer, 2006). Although the crack to powder cocaine disparity has been reduced in 2010 to 18:1 and mandatory minimum sentences for simple drug possession offenses were eliminated (Berman, 2011), African American males continue to be overrepresented in the juvenile justice system.
Substance abuse treatment may be more effective in decreasing drug use. Long-term court ordered substance abuse treatment has been found to reduce the rate of recidivism among high-risk offenders to that of low-risk offenders if treatment is sustained and not ended prematurely (Evans, Huang, & Hser, 2011). Substance abuse treatment may also be beneficial in improving problem-solving skills, which this study shows is particularly important for recidivism for African American males.
For African American females, impulse control and having parental supervision had a huge impact on recidivism. Understanding the relationship between impulse control and gender is important in understanding the relationship between impulse control and recidivism for girls. Hipwell and Loeber (2006) acknowledge being aware of the important role of gender is key in developing effective interventions for at-risk youth. This study shows this relationship is also important for African American girls.
In relation to parental supervision, girls in the juvenile justice system are more likely to incur status offenses from running away from home or being a child without supervision (Le et al., 2003). Relational aggression is more prevalent among adolescent females than males (Crick & Grotpeter, 1995). Perhaps this relational aggression leads to higher levels of conflict between the adolescent and parent. Higher levels of conflict then leads to running away or a lack of parental supervision, which increases the likelihood for re-arrest. This hypothetical situation would support this study’s finding that former female offenders who do have adequate parental supervision are less likely to be re-arrested. As a result, perhaps an intervention can focus specifically on enhancing parenting skills among youth at risk for delinquency or re-offending. This intervention would use parental supervision as part of the larger theoretical framework of resilience among at-risk youth to decrease recidivism.
Limitations and Future Directions
While providing unique new insights into factors influencing recidivism among African American boys and girls, there are still limitations to this research that must be acknowlegded. It is important to remember this sample is composed of moderate- to high-risk African American youth. As a result, it is not reflective of the normative developmental trajectories of African American youth who do not fit this criterion. What may be a protective factor among this population might not be a protective factor among a lower risk juvenile justice population or a sample from the general population.
Furthermore, the time period for which youth were expected to change may not have been sufficient to capture sustained change over time. In the event that youth were in the process of making sustained change, the observation period may have been too brief to observe it. It would be beneficial to examine within-group change over a longer period of time than the 7 years allowed by this study. Likewise, examining the severity of crime committed may play a major role in recidivism. It is likely that serious offenders differ in important ways from low-level offenders. More research will have to be conducted in this area.
Conclusion
African American youth in the juvenile justice system do have protective factors. However, little developmental progress is being made in light of these protective factors. Increasing impulse control, parental supervision, and contact with pro-social peers may reduce the probability of recidivism for African American youth. The ability to control impulsive behavior reduced the probability of recidivism by 48%; however, 84% of youth continued to lack this skill after contact with the justice system. Problem solving was found to be more important for African American males as compared with females, but only 3% of males possessed this skill after contact with the system. Though these numbers are staggering, they give clear indication of how resilience theory through the use of protective factors can be used to target recidivism in the future.
Utilizing resilience theory to inform policy and practices may give these troubled youth a chance at living healthy and productive lives. The results of this study continue to support the notion that context matters in relation to high-risk ethnic minority offenders. Addressing the developmental needs of today’s youth by investing in effective interventions and public policy options that focus on strengthening protective factors may be the most direct route in reducing recidivism rates.
Increased communication between researchers who are familiar with developmental science regarding the process of resilience and protective factors and policy makers who can set the agenda on such issues is critical. Informed by resilience theory, targeted evidence-based interventions utilizing the protective factors identified above have the potential to decrease recidivism and improve the lives of African American youth, their families, the community, and society as a whole. Reducing recidivism may also save millions of dollars as the cost of incarceration is skyrocketing (Welsh & Farrington, 2011). Incarceration is currently competing with funding for education, public health, public safety, parks and recreation, and delinquency prevention programs (Wexler, 2010). Targeted evidence-based interventions informed by developmental science may be the most effective and cost-saving manner to rehabilitate troubled youth.
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 work was funded by a grant from the John D. and Catherine T. MacArthur Foundation as part of the Models for Change: Systems Reform in Juvenile Justice research initiative.