An evidence-based approach to reducing recidivism in court-referred youth.
Despite the juvenile justice system's inherent leniency, juvenile crime is indisputably a serious social problem (Binder, 1988; Latzman, 2008; Steinberg, 2009). Juvenile crime accounted for 16% of all violent crime arrests and 26% of property crime arrests in 2007 (Puzzanchera, 2009). Although cost estimates are not readily available, researchers in the 1990s estimated that reported juvenile crimes cost victims $17.6 billion annually (Freeman, 1996). Furthermore, trajectory analysis of juvenile delinquency has found that juvenile crime is often a doorway to more serious offending and adult criminality (Kadish, Glaser, Calhoun, & Risler, 1999; Van der Geest, Blokland, & Bijleveld, 2009). The total cost to society of a single career criminal, who commenced criminal behaviors during adolescence, is estimated to be $1.5 to $1.8 million (M. Cohen, 1998).
During the 1980s, society witnessed a spike in juvenile crime rates, resulting in harsher sentences, the absorption of many adolescents into the adult justice system, and a shift toward sanction-based programs (Latzman, 2008). Underlying new policy was the philosophy that exposing adolescents to an aversive experience would deter them, and others, from engaging in future criminal behavior (Lipsey & Cullen, 2007). As the juvenile crime rate stabilized since its peak in the early 1990s, the rationale for harsher methods for dealing with juvenile crime has been called into question. According to research, tough love approaches, including intensive supervision, arrest, boot camp, detention, fines, restitution, and drug testing, have limited to negative effects on reoffense rates (Ads, Phipps, Barnoski, & Lieb, 2001). Furthermore, those programs based on incarceration may have a criminogenic effect on adolescents by exposing them to a wider population of delinquents and weakening prosocial relationships (Lipsey & Cullen, 2007).
Few outcome studies have been published in the counseling literature reporting on the work of community agencies with court-referred youth. To date, the majority of studies focused on juvenile corrections have been generated within the disciplines of criminal justice and psychology. Such studies have broadly conceptualized juvenile justice corrections in terms of two theoretically antithetical approaches: sanction based versus rehabilitative (Lipsey & Cullen, 2007). Sanction-based programs aim to reduce recidivism by subjecting youth to aversive programs and policies and by instilling a fear of future consequences. By contrast, rehabilitation programs are designed to bring about lasting changes in behavior by facilitating personal growth, helping youth with social skill development, and subverting negative behaviors (Steinberg, 2009).
The relative value of deterrence-oriented programs over rehabilitation programs is most simply measured by comparing the recidivism rate of youth who were exposed to a particular treatment (Lipsey & Cullen, 2007). Recidivism simply refers to repetition of criminal behavior (Snyder & Sickmund, 2006). However, it is worth noting that recidivism rates are considered a somewhat contentious means of program evaluation because they fail to capture the real reoffense rate and are sensitive to local and state, legal, and bureaucratic differences (Lipsey & Cullen, 2007). Researchers have also debated the relative merits of delineating recidivism by rearrest or reconviction data (Spohn & Holleran, 2002). We used Maltz's (1984) popular recommendation to measure recidivism based on rearrest data. The rationale for this methodology is that the disposition of juvenile crime is frequently diverted outside, or to a lower rung, of the court system and is not, therefore, always reflected in conviction data (Binder, 1988).
Despite these limitations, the goal of all correctional programs is to curtail future criminal activity of program participants; hence, positive outcomes for clients would include an elimination or reduction in crime, as measured by posttreatment recidivism data (Quist & Matshazi, 2000). Indeed, the grant sponsored by the Office of Juvenile Justice and Delinquency Prevention (OJJDP) that supports counseling programs at the community center in the study requires program coordinators to gather data on client recidivism.
A recent meta-analysis synthesizing the results from hundreds of correctional program studies found favorable effect sizes for rehabilitative programs over sanction-based programs for reducing recidivism among juvenile and adult offenders (Lispey & Cullen, 2007). However, rehabilitation cannot be considered the panacea for managing juvenile delinquency, given that positive results have been highly inconsistent with effect sizes ranging from near zero to those that represent a 50% reduction in recidivism and higher (Wilson & Lipsey, 2001). Available data indicate that programs informed by developmental theory, which focus on the acquisition of life skills, yield the greatest effect sizes (Kadish, Glaser, Calhoun, & Ginter, 2001; Latzman, 2008; Lipsey & Cullen, 2007). By contrast, programs informed by traditional psychiatric models of diagnosis and the revisiting of childhood issues have been less successful in curtailing clients from posttreatment offending (Kadish et al., 2001; Latzman, 2008; Lipsey & Cullen, 2007).
* Purpose of the Study
The counseling community is entrenched in the world of juvenile offender rehabilitation, yet peer-reviewed publications tying treatment to offender outcomes remain scarce. Furthermore, the effect of counseling programs on offender rehabilitation has been inconsistent (Lipsey & Cullen, 2007). The current study augments available research by describing and reporting on a psychoeducational program that has been used with a predominantly Latino/a sample of juvenile delinquents. In addition, the study contributes to the field of counseling by addressing a dearth in the literature and establishing a precedent for program evaluation and publication by community service providers. The current inquiry was conducted to evaluate the effectiveness of a life-skills-oriented psychoeducational program on participant recidivism. We hypothesized that youth who complete the program at a university-operated community center would have a lower rate of recidivism than would a matched sample of youth who participated in community probationary programs.
The study used a sample of 120 predominantly Latino/a youth who had been court referred to attend counseling programs at a university community center. Table I provides demographic information related to the treatment and control groups with respect to sex, age, ethnicity, and status of first offense. Youth were eligible for the study if they had successfully completed the center's counseling program at some juncture within a 4-year period from 2004 to 2008 and their parents had signed an agreement permitting the use of their child's data for ongoing research. The data in this study were gathered ex post facto, and therefore obtaining a randomly assigned control group was not feasible. However, we obtained a locally derived control sample from a juvenile justice database representing adolescents from the same geographic area and juvenile court system consisting of six different juvenile judges who serve the county. The control group consisted of 120 youth who had participated in community-based probationary programs but did not receive programming at the center.
Nonprobability sampling was used to select members of the control group, because each participant was selected with respect to matching the demographic characteristics of the treatment group. Matching variables consisted of sex, age, ethnicity, and status of first offense. Initial offenses were recorded as misdemeanors (e.g., truancy, running way, theft less $1,000) and felonies (property damage, assault, etc.). Variables were selected on the basis of controlling for cultural variation and type of offense, given that an inequality of offense status between the treatment and control groups could influence the results. Other factors of interest, such as educational/ability level, were not available in the juvenile justice database, could not be considered for the study, and present a limitation to the results. Chi-square tests for association were conducted to identify equivalence between the treatment and control groups. No significant differences were noted among the demographic variables (see Table 1). Despite some missing data, no significant differences were found across both groups with respect to first offense type.
The university community center that houses the counseling programs is located in the heart of an impoverished inner-city neighborhood in the southern United States. Counseling programs have been offered at the center since 2004. Their inception followed the securing of a substantial grant from the OJJDP. Graduate and doctoral students from a university counseling program conduct the programs. A tri-tiered supervision arrangement provides the linkage between services to clients and university oversight. The doctoral students develop, coordinate, and oversee the provision of services by master's students. Doctoral students also serve as site supervisors to the master's students. Counselor educator professors from the university in turn supervise the doctoral students.
The youth in the treatment group participated in a psychoeducational counseling group with a life skills emphasis. The group was denoted as life skills oriented because training and acquisition of prosocial behaviors was embedded into weekly sessions (Kadish et al., 2001). The program operated in 7-week cycles during which time clients met once weekly for 2-hour group counseling sessions. The program used a psychoeducational model that distributed group time between didactic presentations, application opportunities, and group process (Furr, 2000). Following other successful life skills programs, the interventions used within the group were multisystemic, encompassing modeling, role playing, verbal feedback, reinforcement, and education (Kadish et al., 2001). The content of group sessions was organized around several analogous life skills including identifying feelings, triggers to anger and other feelings/emotions, healthy coping skills, stress management, healthy communication, familial patterns, building self-esteem, and substance abuse. The delivery of these topics was contingent on client need, as evidenced through formal and informal assessment measures given during intake.
The quasi-experimental nature of the research design precluded us from empirically establishing the treatment conditions received by the control group. Furthermore, the control group participants' status as both minors and adjudicated persons afforded them double legal protection, and investigating them beyond information provided on the OJJDP database was prohibited. The database indicated that the youth were on probation, and thus were subject to measures stipulated by the local adjudicating judge. Data were gathered on youth who had been adjudicated by local judges between 2004 and 2008. During this time frame, youth on probation incurred curfews, fines, community service hours, and court-assigned monitoring. However, specific programs assigned under the terms of probation were not delineated by the database. Decisions to refer to either counseling services or sanction-based probationary programs are left to the discretion of the juvenile court judges. Thus, depending on the judge, a juvenile may or may not be referred for counseling services. Juveniles who are not referred may still obtain services. An additional limitation of this study is that the extent to which adolescents obtained counseling services outside of the university community center is not known.
A posttest-only control group design was used to evaluate the difference between the treatment and control groups across recidivism rates. In an effort to ameliorate some of the concerns presented by using recidivism as an outcome variable, we followed the OJJDP recommendation to analyze recidivism along time increments no longer than 2 years from the original registered offense (Synder & Sickmund, 2006). The usage of a locally derived control group also regulated for differences that ensue when recidivism rates of youth across both regions and states are compared (Spohn & Holleran, 2002). Data related to recidivism at 3-, 6-, 12-, 18-, and 24-month intervals were collected for the treatment group and control group. SPSS 17.0 was used to identify differences in recidivism rate for court-referred youth who received counseling services versus a control group. A chi-square test of association was conducted to evaluate the relationship between court-referred adolescents who received a community-based intervention and those who did not, with respect to recidivism over a 2-year time period. An a priori power analysis was conducted to determine an adequate sample size for the study. Given an alpha level of .05 and a moderate effect size, a sample size of 143 was necessary to have adequate power (.80) in the study (J. Cohen, 1988). Thus, the sample size of 240 was sufficient.
A statistically significant effect was noted in recidivism rates across the treatment and control groups, [chi square](5) = 14.64, p = .012, [PHI] = .25, indicative of a small to moderate effect size. Table 2 presents recidivism rates for the treatment and control groups. For youth who received counseling services (i.e., treatment group), 60% did not reoffend during the 24-month period when data were gathered. For youth who received counseling services and reoffended, one half of the youth who reoffended (20% of the treatment group) did so within 3 months. Youth who did not reoffend after 3 months were not likely to reoffend within 2 years after completing the program. Only 20% of the treatment group reoffended after 3 months. For the control group, approximately 46% of the youth did not reoffend. For the 54% of the control group youth who did reoffend, approximately 42% of them reoffended within 1 year. Dimitrov (2009) indicated that standardized residuals greater than 2.0 in absolute value are major contributors to statistical significance of the chi-square value. Although none of the standardized residuals (see Table 2) were noted as major contributors (which is not a surprise given the small to moderate effect size), a pattern was noted between the treatment and control groups and recidivism rates. Adolescents in the treatment group were more likely to avoid reoffending within a 2-year time period or reoffend within 3 months. Adolescents in the control group were more likely to reoffend between 6 months and 24 months.
Court-referred youth, primarily of Latino/a ethnicity, were significantly less likely to reoffend if provided counseling services. Three out of five court-referred youth did not reoffend when they received a community-based intervention, and the likelihood of maintaining progress after participating in a life skills group for 7 weeks increased if clients passed the 3-month threshold of not reoffending.
In an era in which the efficacy of counseling services has been questioned (Thomason, 2010), particularly for ethnically diverse youth (Ngo et al., 2008), the counseling program shows promise in decreasing reoffending behavior. Because the control group did not receive a referral for services, a life-skills- oriented psychoeducational program may be effective for early intervention of offending youth. Given the ethnic identities of the participants in this study, life skills approaches may work for Latino/a clients in community settings who have been referred for status offenses and misdemeanors. However, more research is needed to establish whether this type of program works for offenders charged with more serious offenses. Future research should also aim to establish experimental conditions by regulating the treatment received by control groups.
Despite the inherent limitations of quasi-experimental research, the strength of this study was in design. Preliminary analyses indicated equivalence across age, ethnicity, sex, and status of first offense. Statistical tests demonstrated group equivalence between the two groups. A decrease in recidivism was noted in court-referred youth who obtained counseling services compared with those who did not receive these services, denoting clear evidence that the life skills program was effective. The majority of participants (87%) were Latino/a, lending credibility of the program's effectiveness with a minority group.
Because of the difficulty of controlling for extraneous variables, this study was exploratory. The dearth of extant research with this population probably stems from (a) research with a doubly protected population (i.e., minors and adjudicated), (b) lack of support from external bodies (i.e., juvenile judges), (c) challenges with statistical control, and (d) measurement issues with recidivism. Recidivism in general is not a normally distributed variable and may be difficult to track because of the transient nature of the population. Despite these shortcomings, the focus on counseling outcomes, efficacy, and accountability is important, and this study provides some support for accountability in the practice of counseling.
Implications for Counseling
Counselors in school or community-based programs may rely on group counseling services (Balkin & Leddick, 2005; Corey & Corey, 2006). Life skill programs may be developed to enhance or fulfill a guidance curriculum or youth intervention program. Additionally, the 7-week program used for this study can be easily implemented into existing programs and curricula and is brief in nature so as to provide ample opportunity for the group to be recurring and involve a new set of participants. Such an approach may have the advantage of reaching many youth in a short period of time, thereby providing a beneficial, cost-effective intervention.
The design for this study fits the model for control of extraneous variables, while providing evidence-based data. Such a model, with the cooperation of the juvenile system, may be implemented and easily explained to stakeholders. In an era of accountability, the role of the counselor as both a prevention/intervention specialist and a program evaluator is pertinent (Erford, 2008). Despite the positive direction of the results, in order to construct an efficacious model for juvenile offender counseling, other community counseling programs need to be evaluating their programs and reporting results. Given the inconsistent outcomes counseling programs have demonstrated with respect to client recidivism, further research is imperative. In the future, we aim to refine services by incorporating a qualitative component in the program evaluation process. This may help refine current programming by isolating critical features of the counseling program that clients perceived as most helpful.
Aos, S., Phipps, P., Barnoski, R., & Lieb, R. (2001). The comparable costs and benefits of programs to reduce crime. Olympia, WA: Washington State Institute for Public Policy.
Balkin, R. S., & Leddick, G. R. (2005). Advanced group training for school counselors. In G. R. Waltz & R. K. Yep (Eds.), VISTAS: Compelling perspectives on counseling 2005 (pp. 211-214). Alexandria, VA: American Counseling Association.
Binder, A. (1988). Juvenile delinquency. Annual Review of Psychology, 39, 253-282. doi:10.1146/annurev.ps.39.020188.001345
Cohen, J. (1988). Statistical power analysis for the behavioral sciences. Hillsdale, N J: Erlbaum.
Cohen, M. (1998). The monetary value of saving a high-risk youth. Journal of Quantitative Criminology, 14, 5-33.
Corey, M. S., & Corey, G. (2006). Groups: Process and practice (7th ed.). Belmont, CA: Thomson Brooks/Cole.
Dimitrov, D. M. (2009). Quantitative research in education: Intermediate and advanced methods. Oceanside, NY: Whittier.
Erford, B. T. (2008). Research and evaluation in counseling. Boston, MA: Houghton Mifflin.
Freeman, R. B. (1996). Why do so many young American men commit crime and what might we do about it? Journal of Economic Perspectives, 10, 25-42.
Furr, S. (2000). Structuring the group experience: A format for designing psychoeducational groups. The Journal for Specialists in Group Work, 25, 29-49.
Granville, D. (2007). Risk factors for juvenile delinquency among Latino/a youth (Unpublished doctoral dissertation). Adelphi University, Garden City, NY.
Kadish, T., Glaser, B. A., Calhoun, G. B., & Ginter, E. (2001). Identifying the developmental strengths of juvenile offenders: Assessing four life-skills dimensions. Journal of Addictions & Offender Counseling, 21, 85-95.
Kadish, T. E., Glaser, B. A., Calhoun, G. B., & Risler, E. A. (1999). Counseling juvenile offenders: A program evaluation. Journal of Addictions & Offender Counseling, 19, 88-95.
Latzman, T. L. (2008). An evaluation of Project Peer: A juvenile delinquency treatment program (Unpublished doctoral dissertation). Pace University, New York, NY.
Lipsey, M. W., & Cullen, F. T. (2007). The effectiveness of correctional rehabilitation: A review of systematic reviews. Annual Review of Law and Social Science, 3, 297-320.
Maltz, M. (1984). Recidivism. Orlando, FL: Academic Press.
Ngo, V., Langley, A., Kataoka, S. H., Nadeem, E., Escudero, P., & Stein, B. D. (2008). Providing evidence-based practice to ethnically diverse youths: Examples from the Cognitive Behavioral Intervention for Trauma in Schools (CBITS) program. Journal of the American Academy of Child and Adolescent Psychiatry, 47, 858-862. doi: 10.1097/CHI.0b013e3181799f19.
Nissen, L. (2006). Building strength-based philosophy to life in juvenile justice. Reclaiming Children and Youth, 15, 40-46.
Puzzanchera, C. (2009). Juvenile arrests 2007. Washington, DC: Office of Juvenile Justice and Delinquency Prevention.
Quist, R. M., & Matshazi, D. G. M. (2000). The Child and Adolescent Functional Assessment Scale (CAFAS): A dynamic predictor of juvenile recidivism. Adolescence, 35, 181-192.
Snyder, H. N., & Sickmund, M. (2006). Juvenile offenders and victims: 2006 national report. Washington, DC: Office of Juvenile Justice and Delinquency Prevention.
Spohn, C., & Holleran, D. (2002). The effect of imprisonment on recidivism of felony offenders: A focus on drug offenders. Criminology, 40, 329-355.
Steinberg, L. (2009). Adolescent development and juvenile justice. Annual Review of Clinical Psychology, 5, 458-485.
Thomason, T. (2010). The trend toward evidence-based practice and the future of psychotherapy. American Journal of Psychotherapy, 64, 29-38.
Van der Geest, V., Blokland, A., & Bijleveld, C. (2009). Delinquent development in a sample of high-risk youth: Shape, content, and predictors of delinquent trajectories from age 12 to 32. Journal of Research in Crime and Delinquency, 46, 111-144.
Wilson, D. B., & Lipsey, M. W. (2001). The role of method in treatment effectiveness research: Evidence from meta-analysis. Psychological Methods, 6, 413-429. doi: 10.1037/1082-989X.6.4.413
Chloe Lancaster, Department of Counseling and Educational Psychology & Research, University of Memphis; Richard S. Balkin and Alexandra Valarezo, Department of Counseling and Educational Psychology, and Roberto Garcia, Antonio E. Garcia Arts and Education Center, Texas A&M University-Corpus Christi. The Office of Juvenile Justice and Delinquency Prevention funded this study. Correspondence concerning this article should be addressed to Chloe Lancaster, Department of Counseling and Educational Psychology & Research, University of Memphis, 304 D Ball Hall, Memphis, TN 38152 (e-mail: Clncstr2@memphis.edu).
TABLE 1 Descriptive Statistics and Corresponding Tests of Significance for Treatment and Control Groups Treatment Control Group Group Variable f % f % [chi square] P Sex 0.60 0.438 Male 55 45.8 61 50.8 Female 65 54.2 59 49.2 Ethnicity 3.61 0.307 Latino/a 100 83.3 110 91.7 European American 10 8.3 8 6.7 African American 3 2.5 1 0.8 Missing 7 5.8 1 0.8 First offense 0.22 0.640 Felony 7 5.8 11 9.2 Misdemeanor 87 72.5 108 90.0 Missing 26 21.7 1 0.8 Age 0.158 Note. For age, treatment group m = 14.38 and SD = 1.32; control group M = 14.14 and SD = 1.23; t = 1.42. Percentages may not total 100 because of rounding. TABLE 2 Recidivism Rates for Treatment and Control Groups Reoffence Treatment Control Interval Group Group f % SR f % SR No reoffenses 72 60.0 1.1 55 45.8 -1.1 3 months 24 20.0 1.0 15 12.5 -1.0 6 months 6 5.0 -0.9 11 9.2 0.9 12 months 9 7.5 -1.8 24 20.0 1.8 18 months 6 5.0 -1.0 12 10.0 1.0 24 months 3 2.5 0.0 3 2.5 0.0 Note. SR = standardized residual.
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||Assessment & Diagnosis|
|Author:||Lancaster, Chloe; Balkin, Richard S.; Garcia, Roberto; Valarezo, Alexandra|
|Publication:||Journal of Counseling and Development|
|Date:||Sep 22, 2011|
|Previous Article:||The authorship determination process in student-faculty collaborative research.|
|Next Article:||Courtland Lee: a global advocate for counseling.|