Volume 3, Issue 1 • Fall 2013

Table of Contents

Forword

The Impact of Juvenile Mental Health Court on Recidivism Among Youth

Gender-Specific Mental Health Outcomes of a Community-Based Delinquency Intervention

Predicting Recidivism Among Juvenile Delinquents: Comparison of Risk Factors for Male and Female Offenders

Building Connections Between Officers and Baltimore City Youth: Key Components of a Police–Youth Teambuilding Program

Internet-Based Mindfulness Meditation and Self-regulation: A Randomized Trial with Juvenile Justice Involved Youth

Assessing Youth Early in the Juvenile Justice System

A Jury of Your Peers: Recidivism Among Teen Court Participants

Commentary: Place-Based Delinquency Prevention: Issues and Recommendations

Gender-Specific Mental Health Outcomes of a Community-Based Delinquency Intervention

Ashley M. Mayworm and Jill D. Sharkey
University of California, Santa Barbara

Ashley M. Mayworm, Department of Counseling, Clinical, and School Psychology, University of California, Santa Barbara; Jill D. Sharkey, Department of Counseling, Clinical, and School Psychology, University of California, Santa Barbara.

Correspondence concerning this article should be addressed to: Ashley M. Mayworm, Department of Counseling, Clinical, and School Psychology, University of California, Santa Barbara, Santa Barbara, CA 93106; E-mail: amayworm@education.ucsb.edu

Acknowledgments: The authors wish to thank the principal investigators of the original study. Data were originally collected for a Challenge 2 grant evaluation subcontract to Shane R. Jimerson, Michael J. Furlong, and Manuel Casas (PIs) at the University of California, Santa Barbara from the State Board of Corrections. The study sponsors had no role in determining the study design; in the collection, analysis, and interpretation of data; in the writing of the report; or in the decision to submit the paper for publication.

Keywords: juvenile delinquency, community-based delinquency intervention, wrap-around services, female delinquency, gender-specific program, recidivism, internalizing problems, externalizing problems

Abstract

Juvenile delinquency intervention research has recently called for a focus on the specific treatment needs of females. The current study evaluated the gender-specific mental health outcomes for youth involved in a community-based delinquency intervention (NEW VISTAS). Participants included 102 (42% female) delinquency-involved youth and their mothers. After controlling for initial scores, we found no gender differences in post-intervention mental health scores; parent- and self-report for both girls and boys revealed significantly lower mental health concerns at exit. Results suggest that comprehensive and individualized delinquency interventions such as NEW VISTAS are effective in reducing mental health problems for all participants. Considering gender in the delivery of probation services, and specifically addressing mental health concerns, may be related to significantly lower recidivism rates for youth who successfully complete such a program than youth in a historical comparison group.

Introduction

Effective rehabilitation services are critical for preventing negative and promoting positive outcomes for youth involved in delinquency. Girls and boys who engage in delinquent behavior and have subsequent involvement with the juvenile justice system are not only at risk for further criminal offending (Colman, Mitchell-Herzfeld, Kim, & Shady, 2010), but also for serious mental health problems, academic failure (Chesney-Lind & Shelden, 2004), partner violence, risky sexual behavior (Miller, Malone, & Dodge, 2010), and child maltreatment (Colman et al., 2010). Because of the stability of untreated behavior concerns over a lifetime (Dinh, Roosa, Tein, & Lopez, 2002), it is critical for juvenile offenders to receive interventions that effectively promote healthy development and reduce their likelihood of reoffending.

Historically, research on and programming for juvenile offenders have focused almost exclusively on males (Chesney-Lind & Shelden, 2004). In a review of the history of research on female crime, Tracy, Kempf-Leonard, and Abramoske-James (2009) found that female crime was largely ignored in the research literature for much of the 20th century, perhaps because the prevalence and incidence of female criminality was deemed insufficient for examination. Whereas males have traditionally committed (and continue to commit) more crimes than females, the rates of arrest for males and females in the past few decades show changing trends. Uniform Crime Report data for the year 2007 show that although males continue to comprise the vast majority of juvenile arrests in the United States (71% of all arrests), rates of arrest for males have steadily decreased from 1997 to 2007, while rates for females decreased much more gradually and remained relatively stable from 2002 to 2006 (Tracy et al., 2009).

Various researchers and theorists have debated the reason for the increased proportion of females in the juvenile justice system. Although the data seem to indicate a shift in the behavior of girls, many researchers argue that this increase actually reflects a change in the way police and juvenile justice systems are responding to the behavior of girls (Chesney-Lind & Shelden, 2004; Javdani, Sadeh, & Verona, 2011). Girls are more likely to be arrested for less serious crimes, such as status offenses (e.g., running away and curfew/loitering), than boys (Chesney-Lind & Shelden, 2004), but they are also more likely to receive the harshest sanctions in court (e.g., juvenile prison) for status offenses or technical probation violations (Tracy et al., 2009). Girls are placed in correctional facilities at younger ages than boys and are disproportionately placed in residential settings for status offenses; the great majority of boys are placed in residential settings for more serious misdemeanor or felony offenses (Tracy et al., 2009). These findings suggest that juvenile justice systems treat boys and girls differently, even when they commit similar crimes. Although differential treatment may be needed to effectively intervene with juveniles of different genders, there is insufficient evidence to adequately inform practice. Researchers have recently dedicated more attention to gender issues in juvenile delinquency, but further evaluation of delinquency interventions for both males and females is needed. The current study addressed this need by exploring gender-specific outcomes after involvement in a comprehensive, community-based delinquency intervention.

Delinquency Risk and Protective Factors and Trajectories

Involvement in antisocial and criminal behaviors can be predicted by a complex interplay of factors in multiple areas of youths’ lives, including in the individual, family, school, peer, and social/community contexts (Hawkins et al., 2000). The transactional-ecological model of development recognizes the importance of understanding individuals and their behavior as embedded within multiple systems and relationships (Sameroff, 2000). From this perspective, a youth’s difficulties and strengths are a product of the interaction between dynamic, complex aspects of the youth’s environment over time. In addition, risk and resiliency research and theory states that it is not one specific risk factor that determines one’s likelihood to commit a crime or enter the juvenile justice system, but the number, or accumulation, of risks and protections. For example, Whitney, Renner, and Herrenkohl (2010) found that boys and girls who have a high number of risk and low number of protective factors are more likely to engage in delinquency than their peers who are low in risk and high in protective factors. Findings support the need for delinquency interventions that address multiple risk and protective factors, and suggest that the more risks a youth faces, the more comprehensive and multimodal the treatment needs to be.

In general, the risk and protective factors associated with delinquency are similar for boys and girls. For instance, family dysfunction, associating with antisocial peers, and living in disadvantaged neighborhoods have been found to predict both boys’ and girls’ delinquency (Zahn, 2007). However, important distinctions also exist, with particular risk and protective factors for delinquency being more common or having more of an influence for one gender or the other. Findings from the Girls Study Group, which reviewed and summarized more than 1,600 articles and book chapters related to youth delinquency, found that girls’ risk factors for delinquency differ in a number of ways from boys’ (Zahn et al., 2010). For example, one unique risk factor for girls is the combination of early puberty, significant life stressors, and association with delinquent male peers. In addition, although sexual abuse and maltreatment are risk factors for both boys and girls, girls are more likely than boys to experience sexual abuse both in and outside of the home. Furthermore, internalizing mental health problems such as depression and anxiety are more common in girls than boys and may be related to victimization experiences.

In the overall population, boys have higher levels of risk factors and fewer protections than girls, which may account for boys’ higher rates of involvement in delinquency (Fagan, Van Horn, Hawkins, & Arthur, 2007). However, girls who engage in delinquency typically have more risk factors than boys who exhibit similar behaviors. This phenomenon has been referred to as the “gender paradox,” which states that for disorders or difficulties with an unequal gender ratio, members of the gender with lower prevalence rates (in this example, girls) are more likely to have more serious outcomes and a higher number of risk factors than members of the gender with higher prevalence rates (in this example, boys) (Loeber & Keenan, 1994). In a comparison of risk factors for girls and boys with and without a conviction history, Walrath et al. (2003) demonstrated support for the gender paradox. Both males and females with a conviction history were significantly more likely to report a high number of risk factors than nonconvicted youth. However, females with a conviction history were significantly more likely to report a higher number of individual risk factors than groups of males or nonconvicted females. Considering the relation between gender and differences in type and number of risk factors for delinquency, addressing gender in the design and evaluation of delinquency interventions seems critical and suggests that unequal outcomes may be expected for boys and girls who take part in the same delinquency intervention.

Mental Health and Delinquency

Mental health is a critical factor to address in any delinquency intervention, as one of the most consistent findings in juvenile delinquency research is that the juvenile justice system has an overrepresentation of youth with mental health problems (Graves, Frabutt, & Shelton, 2007). Mental illness is present in about two-thirds of juvenile offenders, which is significantly more than the nonincarcerated population (Cauffman, 2004). Substance abuse, conduct and oppositional disorders, anxiety and depression, and posttraumatic stress disorder (PTSD) are some of the most common mental health concerns for this population (Marston, Russell, Obsuth & Watson, 2012). In a comparison of the behavioral and mental health of males and females in the juvenile justice system, Drerup, Croysdale, and Hoffmann (2008) found that 92% of the males and 97% of females met criteria for at least one mental disorder, while 32% of males and 60% of females met criteria for three or more disorders; others found that 27% of boys and 84% of girls in juvenile justice facilities meet criteria for a diagnosis of mental disorder (Timmons-Mitchell et al., 1997). Mental health problems are of serious concern in both male and female juvenile justice populations, but perhaps are more significant in female delinquency.

Mental health problems are often characterized as falling into one of two categories: externalizing disorders or internalizing disorders. Externalizing disorders are generally defined as those disorders that manifest outwardly, including hyperactivity, aggression, and defiance, whereas internalizing disorders are those with more inward manifestations such as depression, anxiety, and withdrawal (Achenbach, 1991a, 1991b). Both internalizing and externalizing problems have been associated with delinquency in male and female youth. Externalizing problems are common, and often expected, in juvenile justice populations, as many of the behaviors that lead to incarceration are externalizing in nature. In the general population, boys are more likely than girls to exhibit externalizing problems (Rosenfield, Phillips, & White, 2006). However, in the juvenile justice population, females often have as high, if not higher, rates of externalizing problems. Timmons-Mitchell et al. (1997) found that in a sample of incarcerated juvenile delinquents, females scored significantly higher than males on almost all mental health subscales, including externalizing scales such as impulsive propensity, delinquent predisposition, and forceful, oppositional, and unruly behavior. In the Gender and Aggression Project (GAP), Marston et al. (2012) found that 92.9% of the 141 incarcerated adolescent females in their study “met criteria for an externalizing disorder” (p. 107) such as attention deficit hyperactivity disorder (ADHD), conduct disorder (CD), and/or substance abuse. In contrast, in the general population, girls are more likely than boys to exhibit internalizing problems (Rosenfield et al., 2006). This gender difference tends to hold true in juvenile justice populations. Timmons-Mitchell et al. (1997) found girls to have significantly higher scores on internalizing scales such as anxiety, depression, and somatization than boys. In other research, the depression and anxiety/social problems of girls, but not boys, predicted dual involvement in mental health and juvenile justice services (Graves et al., 2007). The GAP project found 33.3% of the female participants “met criteria for an internalizing disorder” (Marston et al., 2012, p. 107) such as major depressive episode (MDE), major depressive disorder (MDD), and/or generalized anxiety disorder (GAD). Although girls in the juvenile justice system tend to have higher rates of internalizing problems than boys, rates for boys are also of concern; for example, Teplin, Abram, McClelland, Dulcan, and Mericle (2002) found that 17.2% of detained male youth met criteria for depression and/or dysthymia.

Research suggests that mental health problems are associated with likelihood to engage in serious delinquency in the future. Postlethwait, Barth, and Guo (2010) explored the ways in which depression, substance use, and parental discipline influenced changes in self-reported youth delinquency over an 18-month period. They found that changes in delinquency varied as a function of level of depression for females and level of substance use for males. More specifically, females with normative levels of depression at baseline had a 16% lower probability of engaging in serious delinquency, whereas females with borderline/clinical levels of depression at baseline had a 204% increase in probability of engaging in serious delinquency. Findings such as these suggest that by addressing mental health concerns, particularly among females, the likelihood of future delinquency and recidivism will decrease.

As Marston et al. (2012) describe, girls with mental health problems who are also involved in delinquency are in double jeopardy—placing them at an increased risk for negative outcomes as they enter adulthood. Numerous mental health professionals have stated that providing mental health services could reduce recidivism (Marston et al., 2012). Nonetheless, a meta-analysis conducted by Foley (2008) found that of the gender-specific intervention articles she reviewed, only one-fourth measured internalizing and/or externalizing behavior change. The study described in this article aims to address this gap in the literature by measuring mental health outcomes among youth involved in a delinquency intervention and attempting to understand the relation between these changes and recidivism rates.

Gender-Specific Delinquency Intervention and NEW VISTAS

In 1992, the 1974 Juvenile Justice and Delinquency Prevention Act was reauthorized, calling for increased research on the specific needs of girls involved in delinquency, as well as the intervention strategies that will best meet those needs (Foley, 2008). These intervention programs, “designed to address the needs unique to the gender of the individual to whom the services are provided,” were titled gender-specific programs (GSPs; Juvenile Justice and Delinquency Prevention Appropriations Authorization, 1992). GSPs refer to programs designed to meet the specific needs of both males and females, although the term is most commonly used to describe female interventions. Research examining effective delinquency interventions for both boys and girls has found that some critical elements of an effective program include: comprehensiveness (addressing multiple risk and protective factors); family involvement, addressing both mental and physical health; and a strengths-based focus (Cooney, Small, & O’Conner, 2008). For girls in particular, programs should address salient factors associated with female involvement in crime, including physical and sexual abuse, social and emotional regulation, relationships with deviant friends and relatives, family problems, running away, substance abuse, teenage pregnancy, and academic failure (Foley, 2008).

Informed by research about risk and protective factors for juvenile delinquency, as well as evidence-based interventions for decreasing delinquency and criminal involvement for males and females, the Santa Barbara County Juvenile Justice Coordinating Council, with funding from a California State Challenge grant and the help of the community and other agencies, created the Neighborhood Enrichment With Vision Involving Services, Treatment, and Supervision (NEW VISTAS) program. NEW VISTAS was a delinquency intervention that utilized a comprehensive service delivery model, which included creating individualized treatment plans for criminally involved youths and their families based on a family-focused, neighborhood-based supervision model. The goals of NEW VISTAS were to increase youth and family “behavioral and emotional strengths, while decreasing recidivism and other negative outcomes,” including mental health concerns (Jimerson et al., 2003, p. 2). Some of the core components of NEW VISTAS were its comprehensive and individualized treatment planning, family focus, provision of substance use and mental and physical health services, and relationship-building opportunities. Intervention elements are summarized here; readers are referred to Jimerson et al. (2003) for more detail.

Comprehensive and individualized treatment. The theoretical underpinnings of the NEW VISTAS program were based on the ecological-transactional model of development, thus targeting intervention to environmental influences in multiple levels of each youth’s ecology. Numerous researchers have discussed the importance of implementing comprehensive, multidimensional interventions for delinquency, rather than those that focus on single risk factors (Hipwell & Loeber, 2006). Those interventions that have utilized a multicontextual approach have been found to be more effective than more narrowly focused programs (Henggeler, Schoenwald, Borduin, Rowland, & Cunningham, 1998). Multimodal interventions for juvenile offenders, such as Multisystemic Therapy (Ogden & Hagen, 2009) and Multidimensional Treatment Foster Care (Leve, Chamberlain, Smith, & Harold, 2012), have found significantly less recidivism for program participants than for youth in control groups.

The NEW VISTAS comprehensive service-delivery model was accomplished through interagency collaboration between agencies and schools in the community. An interagency team provided treatment supervision, case planning, and management and also tracked, supported, and monitored the progress of participating families. Youths were referred to NEW VISTAS through Probation, Truancy, Parole, or Child Welfare Services, and once identified, received an individualized, comprehensive treatment plan to meet the diverse needs of the population, including being effective for clients of different acculturation levels and both males and females. This focus not only on comprehensiveness, but individualization, supports research findings that interventions should be tailored to youth’s specific needs; when adolescents are placed in treatment that they do not need (e.g., group drug treatment when they are not abusing drugs) they may actually have more problems after intervention (Cécile & Born, 2009).

NEW VISTAS provided youth with some choice in determining their treatment in order to avoid iatrogenic effects, as well as to encourage their active engagement in treatment (Jimerson et al., 2003). Matthews and Hubbard (2008) suggest that allowing youth to have personal agency in their treatment helps build strong therapeutic relationships and results in better treatment outcomes. Bloom, Owen, Deschenes, and Rosenbaum (2002) also recommended small, community-based programs for females involved in juvenile justice, which foster one-on-one relationships. One way NEW VISTAS addressed this intervention need was by assigning smaller than typical caseloads to probation officers to try and promote more supportive, one-on-one relationships (Jimerson et al., 2003).

Some of the aspects of NEW VISTAS that meet the multimodal model of treatment delivery include the focus on providing visits and supervision at home (86% of the girls and 83% of the boys), with families (79% of the girls, 76% of the boys), and in the schools (98% of the girls, 78% of the boys). The collaboration with schools, in addition to families, is essential, as both girls and boys involved in juvenile justice frequently experience difficulties in school. For example, juvenile offenders frequently require special education services, or have been diagnosed with a learning disability, at some point in their lives (Kataoka et al., 2001); they often drop out of school (Jenson & Howard, 1998), have low educational achievement (Murray & Farrington, 2010), and low school bonding (Payne, 2008). NEW VISTAS attempted to address these school-related factors through school visits and supervision, as well as tutoring and mentoring programs.

Family-focused intervention. Another core component of NEW VISTAS was to involve family members in the treatment program to counteract the family-based risk factors associated with delinquency; research has found that family risk is significantly, positively, and similarly related to delinquency for boys and girls in middle and high school (Fagan, Lee, Antaramian, & Hawkins, 2011). To address family factors, in addition to receiving home visits and family case management and supervision, many NEW VISTAS families received family counseling or conferencing (33% of the girls, 27% of the boys). Utilizing family therapy in delinquency and substance use interventions for male and female adolescents has been supported by numerous studies, which have found multiple forms of family therapy (e.g., Brief Strategic Family Therapy, Functional Family Therapy) to be effective in reducing these harmful behaviors (Baldwin, Christian, Berkeljon, Shadish, & Bean, 2012). Furthermore, Garcia and Lane (2012) suggested that relationship strain, or negative and painful stimuli associated with relationships with family members, peers, and others, is highly associated with delinquency in girls. Research findings suggest that interventions for girls should target relationships, particularly familial relationships.

Many parents also received individual counseling and/or parent training through NEW VISTAS (24% of the girls, 44% of the boys), based on their needs. For boys, parent training has been found more effective in reducing criminal offending than juvenile justice treatment as usual (Bank, Marlowe, Reid, Patterson, & Weinrott, 1991). However, few (if any) studies have specifically analyzed the effect of parent training on reducing delinquency among female juvenile offenders. Providing individual counseling for parents of youth involved in delinquency is supported by research showing that some of the risk factors for youth delinquency are parental incarceration or criminal offending and substance use/mental health concerns (Aaron & Dallaire, 2010).

Drug and alcohol abuse intervention. There is also ample evidence that substance use is related to delinquency among youth. For example, Johansson and Kempf-Leonard (2009) found that individual-level risk factors predicting serious, violent, and chronic offending by males and females were similar, with mental health problems, running away, and substance abuse problems among some of the most important. NEW VISTAS integrated substance use prevention and treatment into its service-delivery model. Available services included alcohol abuse treatment for youth (43% of the girls, 51% of the boys) and/or counseling (19% of the girls, 20% of the boys), drug abuse treatment for youth (43% of the girls, 51% of the boys) and/or counseling (17% of the girls, 20% of the boys), and substance use treatment/counseling for parents. Families were separated into different substance use service tracks based on their degree of impairment (i.e., moderate or severe), so that individual family needs could be met (Jimerson et al., 2003). Numerous researchers have recommended integrating treatment for substance abuse into a comprehensive, individualized intervention for both female delinquency (Cauffman, 2008) and youth delinquency in general (Chassin, 2008).

Mental and physical health intervention. As mentioned previously, mental health problems are a particularly important risk factor to address in delinquency intervention. One of the specific goals of NEW VISTAS was to reduce internalizing and externalizing problems among youth participating in the program by providing youth counseling (57% of the girls, 44% of the boys) and anger management (21% of the girls, 19% of the boys). Part of the NEW VISTAS model was to match staff and clients appropriately, respect cultural strengths and barriers, and provide gender-specific treatment (Jimerson et al., 2003). All of these goals were integrated into mental health treatment. Matthews and Hubbard (2008) discuss the importance of allowing the option of same-gender counselor matches in gender-specific delinquency interventions, which may help to promote a therapeutic alliance and improve the therapeutic relationship. Research on gender-specific interventions has also pointed out the importance of recognizing within-gender differences, such as cultural differences, and utilizing interventions appropriate for youth of different cultural, ethnic, and linguistic backgrounds (Vincent, Grisso, Terry, & Banks, 2008). NEW VISTAS had bicultural and bilingual staff, focused on neighborhood-based service providers, and individualized mental health treatment with these needs in mind.

Recreational and peer-related intervention. The NEW VISTAS program also provided additional services that have been suggested in the delinquency intervention literature, including probation supervision and a variety of recreational and educational activities. Research has indicated that involvement in structured activities decreases youth delinquency, while involvement in unstructured, social activities, such as hanging out with friends, increases delinquency among males and females (Novak & Crawford, 2010). Thus, recreational programming, including tutoring, mentoring, peer counseling, vocational and life skills training, community service, leadership training, and afterschool recreation were available to youth. The majority of these services were available in all-boy and all-girl settings. In addition to providing structured environments for youth, these services aimed to foster positive relationships, as “creating caring relationships” was one of the core components of the NEW VISTAS model (Jimerson et al., 2003). Research has emphasized the importance of fostering positive peer and adult relationships for juvenile justice involved youth, especially girls, as positive relationships are protective against delinquency (Bloom et al., 2002; Garcia & Lane, 2012; Matthews & Hubbard, 2008). It is important that youth are provided with opportunities to build relationships with prosocial peers, as research has found that those who have a greater number of delinquent peers are more likely to engage in delinquency and violence themselves (Worthen, 2012); this relation is particularly strong for girls (Fagan & Wright, 2012). In addition, male youth who lack positivity in friendships are more likely to engage in delinquency (Worthen, 2012). Programs such as peer tutoring and community service provide real-world settings to cultivate these essential positive relationships. In addition, in order to facilitate involvement in NEW VISTAS programs, a large percentage of the participants (57% of the girls, 41% of the boys) also received transportation to their services, which otherwise may have been difficult to attend.

The Current Study

The current research literature on delinquency interventions lacks studies with adequate numbers of females in their samples, as well as studies that specifically evaluate gender effects (Hipwell & Loeber, 2006). In addition, delinquency interventions often focus on evaluating only changes in recidivism rates, rather than also measuring changes in mental health concerns, such as internalizing and externalizing problems (Foley, 2008). As is clear from the research literature on risk factors for delinquency, mental health problems are common concerns in the juvenile justice population and have been found to predict delinquency (Postlethwait et al., 2010). The current study aimed to address these gaps in the literature by measuring how participants’ internalizing and externalizing symptoms changed, by gender, after completion of the NEW VISTAS program. A comprehensive report of the NEW VISTAS program (Jimerson et al., 2003) found that 6-month recidivism rates for youth who successfully completed the program (19%) were significantly lower than a historical comparison group (39%), although gender differences were not examined. We hypothesize that similar reductions in mental health symptoms will be found. Five research questions are examined: (a) Do males and females differ at intake on levels of internalizing and externalizing problems?; (b) Are there gender differences in youth-reported internalizing and externalizing symptoms at exit from NEW VISTAS, after controlling for initial levels?; (c) Are there gender differences in mother-reported youth internalizing and externalizing symptoms at exit, after controlling for initial levels?; (d) Do participants’ internalizing and externalizing scores significantly decrease from intake to exit?; (e) Are changes in symptomatology related to recidivism rates, and does this differ by gender?

Method

Participants

Participants included 102 juveniles who were enrolled in, and successfully completed, the community-based NEW VISTAS program between the years 1999 and 2003. Following guidelines described by Dattalo (2008), we ran a power analysis using the G*Power 3.1 program (Faul, Erdfelder, Lang, & Buchner, 2007). Results indicated that a MANCOVA (with 2 groups, 2 covariates, and 2 outcomes; numerator df = 4) conducted with 102 participants and power of .80 would be able to detect a small to medium effect size (f2 = .12). Participants were selected to participate in NEW VISTAS based on referrals from juvenile probation (n = 90) and truancy (n = 12). Participants were enrolled in the program for anywhere between 86 days to 1,013 days, with an average of 317 days (M = 317.18, SD = 192.23). As shown in Table 1, participants were relatively balanced across genders and most were Latino. In addition, the majority (79.5%) were in high school (9th-12th grade). Almost one-half (48.0%) of the participants had been suspended or expelled from school in the 12 months prior to program entry, and almost one-fourth (24.5%) identified themselves as being in a gang. Overall, participants are representative of the juvenile justice population in Santa Barbara County, except that the proportion of females is larger in our sample, as researchers oversampled females for the program evaluation. It was not possible to have an experimental control group for this study, but archival data from the Santa Barbara County Probation Department for the 1994–1998 time period was used to gather information about 104 youths who had previously successfully completed probation services. This historical comparison group was matched with the NEW VISTAS participants by gender, ethnicity, age, and severity of offense.

Attrition. Table 1 compares participants who successfully completed the program and were included in the current analyses with those who dropped out through no fault of their own (e.g., moved; n = 34) and those who dropped out because they failed the program (n = 12). Participants who dropped out were similar to those who completed the program in age and ethnicity; participants who dropped out for reasons beyond their control were more likely to be male than participants who completed the program. Drop-out rates for males (7.6% of males) and females (8.7% of females) were similar. Participants who failed the program had higher levels of internalizing and externalizing problems at intake than the group who completed the program, although analysis of variance (ANOVA) comparing intake levels of internalizing and externalizing problems between the groups revealed no statistically significant differences.

Table 1. Comparison of Demographics and Variables of Interest for Study Participants Versus Dropped Participants

  Study Participants Mean (SD) Dropped Out- No Fault Mean (SD) Dropped Out- Failed Mean (SD)

n

102

34

12

Sex

 

 

 

    

Male

57.80%

73.50%

58.30%

    

Female

42.20%

26.50%

41.70%

Age

16.24 (1.04)

16.30 (1.14)

16.37 (1.45)

Ethnicity

 

 

 

    

Latino

77.50%

76.50%

75.00%

    

Anglo

19.60%

17.60%

25.00%

    

 Black

1.00%

0%

0%

    

American Indian

2.00%

0%

0%

    

  Other

0%

5.90%

0%

Internalizing Pretest (Youth Report)

51.49 (11.15)

51.50 (11.36)

56.67 (10.05)

Externalizing Pretest (Youth Report)

51.48 (11.76)

51.91 (12.07)

56.33 (13.40)

Internalizing Pretest (Mother Report)

58.42 (12.00)

55.71 (12.76)

60.67 (9.20)

Externalizing Pretest (Mother Report)

57.52 (12.31)

57.59 (11.69)

62.08 (8.32)


Note. Internalizing and externalizing scores reflect Standardized Scores (M = 50, SD = 10).

Measures

Child Behavior Checklist (CBCL). Mothers completed the CBCL, a parent-report measure of child behavior problems, at intake and exit. The youth self–report (YSR) and CBCL were designed as complementary multi-informant measures of child behavior and are some of the most widely used measures of child behavior problems in psychological research (Achenbach, 1999). The CBCL can be used to rate children ages 4 to 18 years and takes about 15 minutes to complete. Measures were available in English or Spanish. In this study, two broadband behavior scales from the CBCL (Internalizing and Externalizing Problems) were used, for a total of 63 items. To measure Internalizing Problems, the 9-item Withdrawn Behavior (e.g., sad, sulks), 9-item Somatic Complaints (e.g., dizzy), and 13-item Anxious/Depressed (e.g. lonely) subscales were combined. Externalizing Problems were measured by combining the 11-item Delinquent Behavior (e.g., runs away) and 21-item Aggressive Behavior (e.g. starts fights) subscales. Participants responded to each item on a three-point Likert scale (0 = “not true”, 1 = “somewhat or sometimes true”, 2 = “very true or often true”). We then transformed raw scores for both scales into age-standardized scores (M =50, SD =10). Standard scores below 60 are considered to be in the normal range, scores ranging from 60 to 63 are in the borderline range, and scores above 63 are in the clinical range. Achenbach (1991b) reported adequate test-retest reliability, criterion validity, and concurrent validity. In the current study, internal consistencies for both the Internalizing and Externalizing scales were excellent (α = 0.89 and 0.93, respectively).

Youth Self Report. To measure internalizing and externalizing problems, juvenile participants completed five of eight subscales of the YSR (Achenbach, 1991a) at both intake and exit. The YSR is a standardized, self-report measure that is appropriate for children between the ages of 11 and 18. Participants responded to each item on a three-point Likert scale (0 = “not true”, 1 = “somewhat or sometimes true”, 2 = “very true or often true”). To measure internalizing symptoms, we combined the 7-item Withdrawn Behavior (e.g., “I would rather be alone than with others”), 9-item Somatic Complaints (e.g., “I feel dizzy”), and 15-item Anxious/ Depressed (e.g., “I feel that I have to be perfect”) subscales to create an overall Internalizing cluster. To measure externalizing symptoms, we combined the 11-item Delinquent Behavior (e.g., “I lie or cheat”) and 19-item Aggressive Behavior (e.g., “I argue a lot”) subscales to create an overall Externalizing cluster. We transformed raw scores into age-standardized scores in the same way we did for the CBCL. The YSR has been found to have good test-retest reliability and construct validity (Achenbach, 1991a). In the current study, the internal consistencies of both scales were excellent (α = 0.90).

Recidivism. Recidivism was measured by the number of arrests within 6 months after exiting the program. Data were collected by the probation department and transmitted to the researchers at the end of the project. Number of arrests was dichotomized into 1 = had at least one arrest and 0 = had no arrests.

Procedure

Participants were referred to the NEW VISTAS program through the truancy and juvenile probation offices between 1999 and 2003. A Research Advisory Panel, including researchers, police officers, school district personnel, and mental health professionals developed and reviewed the program evaluation protocol. All participants in NEW VISTAS provided data required by the Board of Corrections for the evaluation; a sample of one in two females and one in three males was recruited for a more intensive set of local evaluation measures. Data were collected from the involved families at program intake and exit by the team leaders, who were probation staff responsible for case planning, negotiating services, and family supervision. The forms could be filled out in Spanish or English and were available in interview or questionnaire format. When the data packets were complete they were transferred to researchers who scanned the forms and downloaded the data into the Statistical Package for the Social Sciences (SPSS) software program. Recidivism data on the intervention and historical comparison group were collected by the probation department and sent to the researchers to be analyzed. To adhere to the procedures established to protect the participants, neither names nor identification numbers were included with these data; therefore they were matched with the intervention group using date of birth.

Design and Data Analysis

The current study employed a pretest-posttest design to clarify the association between treatment provision and mental health, and a posttest-only design with a historical control group to examine the association between treatment provision and recidivism. We ran a series of independent samples t-tests to assess whether there were significant differences in both youth- and mother-reported youth internalizing and externalizing symptoms at intake. We made a Bonferroni correction to adjust the p value from 0.05 to 0.0125 to accommodate multiple t-tests. To determine whether gender differences exist for posttest internalizing and externalizing mental health symptoms, after controlling for scores at intake, we conducted two multivariate analysis of covariance (MANCOVA) tests (one for youth-reported symptoms and one for mother-reported youth symptoms) on two dependent variables associated with mental health: post-intervention internalizing and externalizing scores. We adjusted for two covariates: intake internalizing scores and intake externalizing scores. The independent variable was gender. This method is recommended over repeated measures ANOVA when using a pretest-posttest design and when groups are not randomly assigned (Jennings, 1988). The covariates were moderately correlated with each other and with all of the dependent variables (range .312 to .671), which also support the use of MANCOVA rather than multiple ANCOVA tests. All assumptions of MANCOVA were met for both the youth-reported and mother-reported analyses. Finally, descriptive analysis depicted the association between improvement in mental health and recidivism for those participants whose data we could match. All analyses were run using SPSS 20 statistical software.

Results

Research Question A: Gender Differences at Intake and Exit

Results of t-tests indicate that males and females did not differ significantly at intake on youth-reported internalizing, t(100) = -1.91, p = .059, mother-reported internalizing, t(100) = 0.60, p = .549, or mother-reported externalizing problems, t(100) = 0.01, p = .996, at intake. However, females had higher youth-reported externalizing behaviors than males, t(100) = -3.02, p = .003, at intake. Males and females did not differ significantly at exit on any of the scales before controlling for initial levels: youth-reported internalizing, t(100) = -0.59, p = .558, youth-reported externalizing, t(100) = -1.15, p = .254, mother-reported youth internalizing, t(100) = -0.29, p = .772, and mother-reported youth externalizing, t(100) = 0.25, p = .803. Table 2 provides means and standard deviations.

Table 2. Intake and Exit Standard Score Means and Standard Deviations by Gender

    Internalizing Externalizing
Gender n Intake (SD) Exit (SD) Intake (SD) Exit (SD)

Youth Self-Report (YSR)

102

34

 

12

 

Male

59

49.71 (11.52)

47.22 (11.39)

48.59 (12.46)

45.49 (11.45)

Female

43

53.93 (10.26)

48.58 (11.78)

55.44 (9.51)

48.23 (12.56)

Total

102

51.49 (11.15)

47.79 (11.52)

51.48 (11.76)

46.65 (11.95)

Mother Report of Youth (CBCL)

 

 

 

 

 

Male

59

59.03 (12.38)

52.78 (11.66)

57.53 (12.94)

51.41 (12.50)

Female

43

57.58 (11.55)

53.49 (12.76)

57.51 (11.54)

50.79 (11.94)

Total

102

58.42 (12.00)

53.08 (12.08)

57.52 (12.31)

51.15 (12.21)

Research Questions B & C: Gender Differences at Exit When Controlling for Initial Levels

Results of the MANCOVA revealed that the main effect of gender on youth self-reported internalizing and externalizing scores at exit was not significant, after controlling for intake mental health scores, F(1, 97) = 0.46, p = .634. There were similar findings for the MANCOVA conducted on mother-reported youth mental health, F(1,97) = 0.75, p = .475. This indicates that males’ and females’ self-reported and mother-reported posttest internalizing and externalizing mean scores did not differ significantly from each other, after controlling for initial levels.

Research Question D: Pre-Post Change by Gender

Results of MANCOVA also revealed significant main effects for youth self-reported internalizing, F(1, 97) = 17.73, p < .001 (partial eta squared = .15), and externalizing behavior, F(1,97) = 12.16, p = .001 (partial eta squared = .11), as well as mother-reported internalizing, F(1,97) = 59.65, p < .001 (partial eta squared = .38), and externalizing behavior, F(1,97) = 11.82, p = .001 (partial eta squared = .11); participants’ internalizing and externalizing scores from both informants were significantly lower at exit than intake. See Figure 1 for a plot of male and female youth-reported and mother-reported youth scores at intake and exit.

Figure 1. Youth self-reported and mother-reported youths’ internalizing and externalizing standard score means at intake and exit

Youth self-reported and mother-reported youths’ internalizing and externalizing standard score means at intake and exit

Change in internalizing and externalizing scores from intake to exit can also be described in terms of categorical change (whether scores fell in the normative, borderline, or clinical range). The percentage of youth whose internalizing and externalizing scores were in the normative range increased from intake to exit for all measures: from 76.5% to 85.3% for youth-reported internalizing, 76.5% to 87.3% for youth-reported externalizing, 52.0% to 70.6% for mother-reported internalizing, and 51.0% to 77.5% for mother-reported externalizing. Percentages of youth in borderline and clinical ranges decreased from intake to exit for all measures. The percentage of youth in the clinical range for youth- and mother-reported internalizing problems dropped from 12.7% to 8.8%, and 40.2% to 24.5%, respectively; youth- and mother-reported clinical range externalizing scores dropped from 14.7% to 4.9%, and 34.3% to 16.7%, respectively. Figure 2 depicts changes in participants’ mental health categories from intake to exit.

Figure 2. Number of youths who were in either the normative, borderline, or clinical range at intake and the range they were in at exit, for youths’ self-reports and mother-reports of youths’ externalizing and internalizing symptoms.

Number of youths who were in either the normative, borderline, or clinical range at intake and the range they were in at exit, for youths’ self-reports and mother-reports of youths’ externalizing and internalizing symptoms.

Research Question E: Relation Between Change in Symptoms and Recidivism

Unfortunately, 6-month recidivism data were missing or unable to be matched with mental health data for 52 (51%) of the participants. Participants with unmatched follow-up arrest data were more likely to be male (75% versus 42%) and to have had an arrest during the intervention (66% versus 30%) than those with matched follow-up arrest data. Because of the small sample size, high percentage of missing data, and some notable differences between participants with and without follow-up data, we did not run statistical analyses. However, descriptive data suggest that recidivism rates at 6-month follow-up were very low for both males (81.0% did not recidivate, n = 17 ) and females (96.6% did not recidivate, n = 28). These rates of recidivism are lower than in the historical comparison group, of which only 64.2% of males (n = 18) and 55.5% of females (n = 10) did not recidivate during the 6-month follow-up period. In addition, in the NEW VISTAS sample, the magnitude of symptom change does appear to be related to whether a participant recidivated (see Figure 3).

Figure 3. Change in symptom scores (Pre – Post) on the CBCL (youth self-report) and YSR (mother report) for youths with no arrests and at least one arrest at 6-month follow-up. A positive score indicates better mental health (lower levels of reported symptoms at exit).

 Change in symptom scores (Pre – Post) on the CBCL (youth self-report) and YSR (mother report) for youths with no arrests and at least one arrest at 6-month follow-up. A positive score indicates better mental health (lower levels of reported symptoms at exit).

Discussion

The current study aimed to understand the gender-specific mental health and recidivism outcomes for youth who participated in a comprehensive, individualized, and neighborhood-based delinquency intervention with a focus on gender-specific treatment. Results of the study demonstrated that males and females had similar internalizing and externalizing symptoms at exit from NEW VISTAS after controlling for initial levels, which were significantly lower than their scores at intake. Results suggest that participants in the NEW VISTAS program had a significant decrease in mental health symptoms, with males and females showing similar changes. In addition to improvements in mental health, there was a lower recidivism rate for participants in NEW VISTAS when compared to a historical control group; NEW VISTAS participants’ change in mental health symptom scores appears to be related to arrest rates at 6-month follow-up.

At intake we found that boys and girls were similar on measures of internalizing and externalizing symptoms. The finding that girls did not have significantly higher internalizing scores at intake is surprising, considering previous research that has found girls involved with the juvenile justice system typically have higher rates of depression and other internalizing problems than boys (Teplin et al., 2002; Timmons-Mitchell et al., 1997). This is especially surprising considering the fact that girls engaged in delinquency often have traumatic experiences that result in higher rates of PTSD and anxiety-related disorders (Zahn et al., 2010). This finding may reflect a difference in our sample of youth as compared to other studies, particularly because we have a larger Latino population than most other studies (e.g., Teplin et al., 2002; Timmons-Mitchell et al., 1997). Girls and boys did not differ significantly on mother-reported externalizing symptoms, with boys and girls having almost identical intake scores. However, girls had significantly higher self-reported externalizing scores at intake than boys. In the general population, boys typically demonstrate more externalizing problems than girls (Rosenfield et al., 2006); however, many studies have found girls to have higher rates of all mental health problems, including externalizing problems, compared to boys (Timmons-Mitchell et al., 1997). Our finding may be explained by the “gender paradox” in youth delinquency, which states that because girls are less likely to be involved in juvenile delinquency, those girls who do engage in it are likely to experience higher levels of related risk factors, particularly comorbid mental health problems and disruptive behavior (Loeber & Keenan, 1994).

At exit, there were no significant differences between boys and girls on any of the measures of mental health after controlling for mental health scores at intake. As a group, participants’ internalizing and externalizing scores decreased significantly after involvement in the program. These findings align with previous research on gender-specific interventions for delinquency, which show that programs that address the multitude of risk factors and ecological contexts that influence youth involvement in delinquency are the most effective (Henggeler et al., 1998; Mullis, Cornille, Mullis, & Huber, 2004). However, because we had the power to detect an effect size of f2 = .12, it is possible that if a real, but very small effect of gender on outcomes existed, it may not have been detected in this study; a larger sample size may reveal these differences. Nonetheless, findings are supported by other studies of multimodal interventions (i.e., Multisystemic Therapy and Multimodel Treatment Foster Care), which have found that these individualized and comprehensive delinquency interventions are effective for both boys and girls (Leve et al., 2012; Ogden & Hagen, 2009). This study adds to this current literature, as it evaluated the gender-specific outcomes of a different multimodal intervention, NEW VISTAS, and focused on mental health symptoms as the outcome of interest. Few studies have measured changes in mental health after involvement in a delinquency intervention, despite the preponderance of evidence for its important relation to juvenile delinquency (Foley, 2008). Because both males and females experienced decreased mental health symptoms, and at similar rates, the current study provides evidence for the use of comprehensive, community-based programming for the reduction of mental health problems in girls and boys.

Kazdin (2003) emphasized the importance of measuring not only statistically significant changes in adolescent behavior or mental health after an intervention, but also clinical significance (i.e., did the adolescent’s behavior move into the nonclinical range?). Results of descriptive analyses revealed that the majority of the youths’ internalizing and externalizing symptoms were in the normative range after exiting from NEW VISTAS, and most youth moved to a less severe range or stayed in the same range (Figure 2). The proportion of youth with borderline and clinical levels of mental health concerns at exit mirrored what is expected in the general population rather than the higher rates that occur in the juvenile justice population. There were a few participants whose mental health rating moved from either the normative range to the borderline range or the normative or borderline range to the clinical range. It is unclear why this occurred, and findings could indicate iatrogenic effects. Cécile and Born (2009) discussed the risk for iatrogenic effects in youth delinquency interventions, particularly group settings where youths can learn or reinforce further negative behavior because of their association with other deviant peers. Their review of the literature found that interventions that involve the family and target multiple factors that promote delinquency, including substance use and deviant peers, show the greatest success in reducing delinquency. Considering this previous research, it is surprising that some youths’ behavior appear to have worsened after involvement in a comprehensive program such as NEW VISTAS. It is likely that these findings simply reflect that these youth were already on a negative trajectory and that the intervention failed to stop or only mitigated this negative course. Differences between youth whose symptoms increased and those whose did not should be explored in the future.

In addition to mental health outcomes, this study explored the possible relation between change in mental health symptoms and recidivism. A review of the literature conducted by McLean and Ransford (2004) found that one of the most important types of programs for reducing recidivism among parolees is mental health treatment, but that it is rarely provided. In the current study, missing 6-month follow-up recidivism data precluded statistical comparisons; however, descriptive analyses of the available 6-month follow-up data suggest that the great majority of participants did not recidivate. In fact, of participants in this study, only 19% of males and 3% of females reoffended in the 6 months after successfully completing NEW VISTAS, which is much lower than the historical control group reoffense rate of 39%. Descriptive results pictured in Figure 3 indicate that with the exception of self-reported externalizing behaviors, participants with decreases in reported internalizing and externalizing symptoms had no arrest at 6-month follow-up, while all participants with increases in these symptoms had an arrest. These findings must be cautiously interpreted because of the large percentage of missing data, small sample size, and potential differences between those with and without follow-up data. Future studies that are able to analyze the relation between changes in mental health symptoms and recidivism are needed to further elucidate this association.

Implications and Future Directions

The current study points to the potential for a comprehensive, individualized, and culturally- and gender-sensitive program to significantly reduce mental health symptoms for both males and females on probation. These findings are an important addition to the research literature on effective delinquency programming for female youth, as females have traditionally been neglected in delinquency intervention studies (Chesney-Lind & Shelden, 2004; Hipwell & Loeber, 2006). Zahn et al. (2010) described a great need for further research on what effective treatment for females involved in the juvenile justice system looks like. The current study addresses this gap in the literature, although additional research must continue to strive to adequately understand the intervention needs of youths of different genders.

As this study relied on data gathered through the evaluation of a comprehensive community initiative, there were several limitations that need to be addressed in more controlled research. First, there was lack of specificity regarding the intervention programs that participants received. Probation contracted with several providers to implement services, including mental health treatment. Although probation partnered with agencies to select and implement programs that met NEW VISTAS program criteria, it is unclear to what degree each intervention was implemented with fidelity. In the future it will be critical for studies to isolate which specific aspects of the intervention are most and least helpful for youth of different genders. Future research should evaluate the gender-specific effectiveness of intervention models, such as NEW VISTAS, using randomized, control group studies with intervention fidelity checks to allow outcomes to be attributed to the treatment. In addition, human subjects protection for research with juveniles on probation and in the juvenile justice system, which required the sealing of juvenile records, yielded missing follow-up data. Such constraints to community-based research need to be identified and addressed proactively in future research. Finally, some participants were not included in the current analyses because of failure to complete the program, either because they moved or were unable to successfully participate. Even though dropped participants were not significantly different from the study sample, findings should only be generalized to youth who are able to complete a comprehensive delinquency intervention such as NEW VISTAS, since research has found that families with multiple risk factors, including mental health concerns, are more likely to drop out of treatment prematurely (Hipwell & Loeber, 2006).

It will also be important for future studies evaluating gender-specific delinquency interventions to focus not only on changes in youth problems, such as mental health symptoms and recidivism, but also on strengths. Positive psychology points to the importance in understanding not only whether problems decrease during and after intervention implementation, but whether positive attributes that promote resiliency and well-being are also developed (Matthews & Hubbard, 2008). Agencies, communities, and schools working with youth involved in delinquency and/or the juvenile justice system should consider the use of collaborative approaches to intervention, such as NEW VISTAS, which recognize that the causes and facilitating factors for delinquency are complex and multifaceted.

About the Authors

Ashley M. Mayworm, MEd, is a doctoral candidate in the Counseling, Clinical and School Psychology Department at the University of California, Santa Barbara.

Jill D. Sharkey, PhD, is a full-time faculty member and the school psychology program coordinator in the Department of Counseling, Clinical, and School Psychology in the Gevitz Graduate School of Education at the University of California, Santa Barbara.

References

Aaron, L., & Dallaire, D. H. (2010). Parental incarceration and multiple risk experiences: Effects on family dynamics and children’s delinquency. Journal of Youth Adolescence, 39, 1471–1484. doi: 2048/10.1007/s10964-009-9458-0

Achenbach, T. M. (1991a). Integrative guide for the 1991 CBCL/4-18, YSR, and TRF profiles. Burlington, VT: University of Vermont, Department of Psychiatry.

Achenbach, T. M. (1991b). Manual for the Child Behavior Checklist/4-18 and 1991 Profile. Burlington, VT: University of Vermont, Department of Psychiatry.

Achenbach, T. M. (1999). The Child Behavior Checklist and related instruments. In M. E. Maruish (Ed.), The use of psychological testing for treatment planning and outcomes assessment (pp. 429–466). Mahwah, NJ: Lawrence Erlbaum Associates, Inc.

Baldwin, S. A., Christian, S., Berkeljon, A., Shadish, W. R., & Bean, R. (2012). The effects of family therapies for adolescent delinquency and substance abuse: A meta-analysis. Journal of Marital and Family Therapy, 38, 281–304. doi: 10.111/j.1752-0606.2011.00248.x

Bank, L., Marlowe, J. H., Reid, J. B., Patterson, G. R., & Weinrott, M. R. (1991). A comparative evaluation of parent-training interventions for families of chronic delinquents. Journal of Abnormal Child Psychology, 19, 15–33. doi: 2048/10.1007/BF00910562

Bloom, B., Owen, B., Deschenes, E. P., & Rosenbaum, J. (2002). Moving toward justice for female juvenile offenders in the new millennium. Journal of Contemporary Criminal Justice, 18, 37–56. doi: 2048/10.1177/1043986202018001004

Cauffman, E. (2004). A statewide screening of mental health symptoms among juvenile offenders in detention. Journal of the American Academy of Child and Adolescent Psychiatry, 43, 430–439. doi: 2048/10.1097/00004583-200404000-00009

Cauffman, E. (2008). Understanding the female offender. The Future of Children, 18, 119–142. doi: 2048/10.1353/foc.0.0015

Cécile, M., & Born, M. (2009). Intervention in juvenile delinquency: Danger of iatrogenic effects. Children and Youth Services Review, 31, 1217–1221. doi:10.1016/j.childyouth.2009.05.015

Chassin, L. (2008). Juvenile justice and substance use. The Future of Children, 18, 165–183. doi: 2048/10.1353/foc.0.0017

Chesney-Lind, M., & Shelden, R. G. (2004). Girls, delinquency, and juvenile justice (3rd ed.). Belmont, CA: Wadsworth/Thomson Learning.

Colman, R. A., Mitchell-Herzfeld, S., Kim, D. H., & Shady, T. A. (2010). From delinquency to the perpetration of child maltreatment: Examining the early adult criminal justice and child welfare involvement of youth released from juvenile justice facilities. Children and Youth Services Review, 32, 1410–1417. doi: 2048/10.1016/j.childyouth.2010.06.010

Cooney, S. M., Small, S. A., & O’Conner, C. (2008). Girls in the juvenile justice system: Toward effective gender-responsive programming. What works, Wisconsin research to practice series, 7. Madison, WI: University of Wisconsin–Madison/Extension.

Dattalo, P. (2008). Determining sample size: Balancing power, precision, and practicality. New York: Oxford University Press.

Dinh, K. T., Roosa, M. W., Tein, J., & Lopez, V. A. (2002). The relationship between acculturation and problem behavior proneness in a Hispanic youth sample: A longitudinal mediation model. Journal of Abnormal Child Psychology, 30, 295–309. doi: 2048/10.1023/A:1015111014775

Drerup, L. C., Croysdale, A., & Hoffmann, N. G. (2008). Patterns of behavioral health conditions among adolescents in a juvenile justice system. Professional Psychology: Research and Practice, 39, 122–128. doi: 10.1037/0735-7028.39.2.122

Fagan, A. A., Lee, V. H., Antaramian, S., & Hawkins, J. D. (2011). How do families matter? Age and gender differences in family influences on delinquency and drug use. Youth Violence and Juvenile Justice, 9, 150–170. doi: 10.1177/1541204010377748

Fagan, A. A., Van Horn, M. L., Hawkins, J. D., & Arthur, M. W. (2007). Gender similarities and differences in the association between risk and protective factors and self-reported serious delinquency. Prevention Science, 8, 115–124. doi:10.1007/s11121-006-0062-1

Fagan, A. A., & Wright, E. M. (2012). The effects of neighborhood context on youth violence and delinquency: Does gender matter? Youth Violence and Juvenile Justice, 10, 64–82. doi: 10.1177/1541204011422086

Faul, F., Erdfelder, E., Lang, A. G., & Buchner, A. (2007). G*Power 3: A flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behavior Research Methods, 39, 175–191.

Foley, A. (2008). The current state of gender-specific delinquency programming. Journal of Criminal Justice, 36, 262–269. doi: 2048/10.1016/j.jcrimjus.2008.04.007

Garcia, C. A., & Lane, J. (2012). Dealing with the fall-out: Identifying and addressing the role that relationship strain plays in the lives of girls in the juvenile justice system. Journal of Criminal Justice, 40, 259–267. doi:10.1016/j.jcrimjus.2012.02.005

Graves, K. N., Frabutt, J. M., & Shelton, T. L. (2007). Factors associated with mental health and juvenile justice involvement among children with severe emotional disturbance. Youth Violence and Juvenile Justice, 5, 147–167. doi: 10.1177/1541204006292870

Hawkins, J. D., Herrenkohl, T. I., Farrington, D. P., Brewer, D., Catalano, R. F., Harachi, T. W., et al. (2000). Predictors of youth violence. Washington, DC: U.S. Department of Justice, Office of Justice Programs, Office of Juvenile Justice and Delinquency Prevention.

Henggeler, S. W., Schoenwald, S. K., Borduin, C. M., Rowland, M. D., & Cunningham, P. B. (1998). Multisystemic treatment of antisocial behavior in children and adolescents. New York: Guilford.

Hipwell, A., & Loeber, R. (2006). Do we know which interventions are effective for disruptive and delinquent girls? Clinical Child and Family Psychology Review, 9, 221–255. doi:10.1007/s10567-006-0012-2.

Javdani, S., Sadeh, N., & Verona, E. (2011). Gendered social forces: A review of the impact of institutionalized factors on women and girls’ criminal justice trajectories. Psychology, Public Policy, and Law, 17, 161–211. doi: 10.1037/a0021957

Jennings, E. (1988). Models for pretest-posttest data: Repeated measures ANOVA revisited. Journal of Educational Statistics, 13, 273–280. doi: 2048/10.2307/1164655

Jenson, J. M., & Howard, M. O. (1998). Correlates of gang involvement among juvenile probationers. Journal of Gang Research, 5, 7–15.

Jimerson, S. R., O’Brien, K. M., Furlong, M. J., Sharkey, J., Sia, A., & Graydon, K. (2003). Analysis of a four-year longitudinal study of a neighborhood-based, family-focused, intervention program for at-risk adolescents: Examining behaviors, relationships, and numerous individual, family, neighborhood, and juvenile justice outcomes. Santa Barbara, CA: Santa Barbara Probation Department. Available online at http://www.education.ucsb.edu/newvistas

Johansson, P., & Kempf-Leonard, K. (2009). A gender-specific pathway to serious, violent, and chronic offending?: Exploring Howell’s risk factors for serious delinquency. Crime & Delinquency, 55, 216–240. doi: 10.1177/0011128708330652

Juvenile Justice and Delinquency Prevention Appropriations Authorization, Pub. L. 102–586, 1, 106 Stat. 4982 (1992).

Kataoka, S. H., Zima, B. T., Dupre, D. A., Moreno, K. A., Yang, X., & McCracken, J. T. (2001). Mental health problems and service use among female juvenile offenders: Their relationship to criminal history. Journal of American Academy of Child and Adolescent Psychiatry, 40, 549–555. doi:2048/10.1097/00004583-200105000-00014

Kazdin, A. (2003). Clinical significance: Measuring whether interventions make a difference. In A. Kazdin (Ed.), Methodological issues & strategies in clinical research (3rd ed., pp. 691–710). Washington, DC: American Psychological Association.

Leve, L. D., Chamberlain, P., Smith, D. K., & Harold, G. T. (2012). Multidimensional Treatment Foster Care as an intervention for juvenile justice girls in out-of-home care. In S. Miller, L. D. Leve, & P. K. Kerig (Eds.), Delinquent girls: Contexts, relationships, and adaptation (pp. 147–160). New York: Springer Science+Business Media, LLC.

Loeber, R., & Keenan, K. (1994). Interaction between conduct disorder and its comorbid conditions: Effects of age and gender. Clinical Psychology Review, 14, 497–523. doi: 10.1016/0272-7358(94)90015-9

Marston, E. G., Russell, M. A., Obsuth, I., & Watson, G. K. (2012). Dealing with double jeopardy: Mental health disorders among girls in the juvenile justice system. In S. Miller, L. D. Leve, & P. K. Kerig (Eds.), Delinquent girls: Contexts, relationships, and adaptation (pp. 105–118). New York: Springer Science+Business Media, LLC.

Matthews, B., & Hubbard, D. J. (2008). Moving ahead: Five essential elements for working effectively with girls. Journal of Criminal Justice, 36, 494–502. doi: 2048/10.1016/j.jcrimjus.2008.09.011

McLean, L., & Ransford, C. (2004). Current strategies for reducing recidivism. Retrieved from the Center for Impact Research Web site: http://www.impactresearch.org/documents/recidivismexecutivesummary.pdf

Miller, S., Malone, P. S., & Dodge, K. A. (2010). Developmental trajectories of boys’ and girls’ delinquency: Sex differences and links to later adolescent outcomes. Journal of Abnormal Child Psychology, 38, 1021–1032. doi: 2048/10.1007/s10802-010-9430-1

Mullis, R. L., Cornille, T. A., Mullis, A. K., & Huber, J. (2004). Female juvenile offending: A review of characteristics and contexts. Journal of Child and Family Studies, 13, 205–218. doi: 2048/10.1023/B:JCFS.0000015708.71295.2a

Murray, J., & Farrington, D. P. (2010). Risk factors for conduct disorder and delinquency: Key findings from longitudinal studies. The Canadian Journal of Psychiatry, 55, 633–642.

Novak, K. B., & Crawford, L. A. (2010). Routine activities as determinants of gender differences in delinquency. Journal of Criminal Justice, 38, 913–920. doi: 10.1016/j.jcrimjus.2010.06.008

Ogden, T., & Hagen, K. A. (2009). What works for whom? Gender differences in intake characteristics and treatment outcomes following Multisystemic Therapy. Journal of Adolescence, 32, 1425–1435. doi:10.1016/j.adolescence.2009.06.006

Payne, A. A. (2008). A multilevel analysis of the relationships among communal school organization, student bonding, and delinquency. Journal of Research in Crime and Delinquency, 45, 429–455. doi: 10.1177/0022427808322621

Postlethwait, A. W., Barth, R. P., & Guo, S. (2010). Gender variation in delinquent behavior changes of child welfare-involved youth. Children and Youth Services Review, 32, 318–324. doi: 10.1016/j.childyouth.2009.09.012

Rosenfield, S., Phillips, J., & White, H. (2006). Gender, race, and the self in mental health and crime. Social Problems, 53, 161–185. doi: 2048/10.1525/sp.2006.53.2.161

Sameroff, A. J. (2000). Developmental systems and psychopathology. Development and Psychopathology, 12, 297–312.

Teplin, L. A., Abram, K. M., McClelland, G. M., Duncan, M. K., & Mericle, A. A. (2002). Psychiatric disorders in youth in juvenile detention. Archives of General Psychiatry, 59, 1133–1143.

Timmons-Mitchell, J., Brown, C., Schulz, S. C., Webster, S. E., Underwood, L. A., & Semple, W. E. (1997). Comparing the mental health needs of female and male incarcerated juvenile delinquents. Behavioral Sciences & the Law, 15, 195–202. doi:10.1002/(SICI)1099-0798(199721)15:2<195::AID-BSL269>3.0.CO;2-8

Tracy, P. E., Kempf-Leonard, K., & Abramoske-James, S. (2009). Gender differences in delinquency and juvenile justice processing: Evidence from national data. Crime & Delinquency, 55, 171–215. doi: 10.1177/0011128708330628

Vincent, G. M., Grisso, T., Terry, A., & Banks, S. (2008). Sex and race differences in mental health symptoms in juvenile justice: The MAYSI-2 national meta-analysis. Journal of the American Academy of Child & Adolescent Psychiatry, 47, 282–290. doi: 10.1097/CHI.0b013e318160d516

Walrath, C., Ybarra, M., Holden, E. W., Manteuffel, B., Santiago, R., & Leaf, P. (2003). Female offenders referred for community-based mental health service as compared to other service-referred youth: Correlates of conviction. Journal of Adolescence, 26, 45–61. doi: 2048/10.1016/S0140-1971(02)00113-6

Whitney, S. D., Renner, L. M., & Herrenkohl, T. I. (2010). Gender differences in risk and promotive classifications associated with adolescent delinquency. The Journal of Genetic Psychology: Research and Theory on Human Development, 171, 116–138. doi: 10.1080/00221320903548092

Worthen, M. G. F. (2012). Gender differences in delinquency in early, middle, and late adolescence: An exploration of parent and friend relationships. Deviant Behavior, 33, 282–307. doi: 10.1080/01639625.2011.573421

Zahn, M. A. (2007). The causes of girls’ delinquency and their program implications. Family Court Review, 45. 456–465. doi: 10.1111/j.1744-1617.2007.00161.

Zahn, M. A., Agnew, R., Fishbein, D., Miller, S., Winn, D., Dakoff, G., et al. (2010). Causes and correlates of girls’ delinquency. Bulletin: Girls Study Group Series. Retrieved from Office of Juvenile Justice and Delinquency Prevention Web site: https://www.ncjrs.gov/pdffiles1/ojjdp/226358.pdf

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