AN OUTCOMES STUDY OF JUVENILE DIVERSION PROGRAMS ON
NON-SERIOUS DELINQUENT AND STATUS OFFENDERS
by
MELODY J. STEWART
Submitted in partial fulfillment of the requirements
For the degree of Doctor of Philosophy
Dissertation Adviser: Dr. Victor K. Groza
Mandel School of Applied Social Sciences
CASE WESTERN RESERVE UNIVERSITY
August, 2008 CASE WESTERN RESERVE UNIVERSITY
SCHOOL OF GRADUATE STUDIES
We hereby approve the dissertation of Melody J. Stewart,
candidate for the Ph.D. degree*.
(signed) Victor K. Groza, Ph.D. (chair of the committee)
Kathleen J. Farkas, Ph.D.
David B. Miller, Ph.D.
Emilia N. McGucken, Ph.D.
(date) May 21, 2008
* We also certify that written approval has been obtained for any proprietary material contained therein
ii
Copyright © 2008 by Melody J. Stewart All rights reserved
iii
Table of Contents
Page
List of Tables ………………………………………………………………….. 4
List of Figures …………………………………………………………………. 6
Acknowledgements and Dedication ………………………………………… 7
Abstract ………………………………………………………………………… 10
Chapter
1. Introduction ……………………………………………………. 12
Overview of Problem …………………………………. 13
Overview of Juvenile Diversion Programs …………. 17
Cuyahoga County Juvenile Diversion Programs …. 22
2. Theoretical Perspectives …………………………………….. 26
Labeling Theory ………………………………………. 26
Restorative Justice …………………………………… 28
Outcomes of Diversion Programs ………………….. 32
Effectiveness of Diversion Programs ………………. 35
Summary and Research Questions ………………… 39
3. Methodology …………………………………………………… 43
Research Design ……………………………………… 44
Sample and Selection Criteria ………………………. 45
Data Collection ………………………………………... 49
1 Protection of Human Subjects ……………………….. 52
Measurement – Variables and Operationalization … 53
Variable Definitions and Coding ……………………… 57
Definition of Key Terminology ………………………… 60
4. Findings …………………………………………………………. 63
Demographic Characteristics …………………………. 63
Legal Characteristics …………………………………... 65
Sample Demographic and Legal Comparisons …….. 66
Program Disposition by Demographic and Legal Characteristics ………………………………………….. 70
The Reintegrative Program …………………………… 75
The Typical Program ………………………………….. 78
Comparison of Diversion Programs …………………. 81
Research Questions and Hypotheses ………………. 86
Programmatic Components and Sanctions ………… 99
Exploratory Research …………………………………. 102
5. Discussion and Conclusion …………………………………… 108
Discussion ……………………………………………… 108
Conclusion ……………………………………………… 114
Implications for Policy, Practice, and Research …… 117
Limitations of Study …………………………………… 120
Appendices …………………………………………………………………….. 125
2 Appendix A …………………………………………….. 125
Appendix B …………………………………………….. 126
Appendix C …………………………………………….. 129
Appendix D …………………………………………….. 132
Appendix E ……………………………………………... 134
Appendix F ……………………………………………… 136
Appendix G ……………………………………………... 138
Endnotes ………………………………………………………………………... 139
References ……………………………………………………………………… 140
3 List of Tables
Tables Page
Table 1. Number of Participants by Diversion Program ……………………… 48
Table 2. Offense Levels and Measurement Codes ………………………….. 54
Table 3. Program Construct Variables and Corresponding Program Components and Sanctions ………………………………………….. 56
Table 4. Percentage of Youth in Each Program Assigned Construct Components/Sanctions.……………………………………………….. 57
Table 5. Sample Demographic Characteristics ………………………………. 64
Table 6. Legal Characteristics of Sample …………………………………….. 65
Table 7. Comparisons of Gender by Other Demographics ………………… 67
Table 8. Comparisons of Offense Type by Demographics ………………… 68
Table 9. Offense Level Means by Demographics …………………………… 69
Table 10. Summary of Program Disposition by Demographic and Legal Characteristics ………………………………………………………. 71
Table 11. Comparison of Program Disposition Using Chi Square on Discrete Data ………………………………………………………………….. 72
Table 12. Summary of Recidivism by Demographic and Legal Characteristics ……………………………………………………… 73
Table 13. Re-offense Levels and Measurement Codes …………………… 74
Table 14. Summary of Offense Escalation by Demographic and Legal Characteristics ……………………………………………………… 75
Table 15. Characteristics of Reintegrative CDP Subgroup ………………. 76
Table 16. Characteristics of Reintegrative CDP by Year …………………. 77
Table 17. Characteristics of Typical CDP by Subgroup …………………... 79
4 Table 18. Characteristics of Typical CDP by Year ………………………… 80
Table 19. Demographic Characteristics of Sample by Program ………… 82
Table 20. Summary of Legal Characteristics of Sample by Program …... 83
Table 21. Offenses Committed by Program ……………………………….. 85
Table 22. Chi Square Analysis of Program Completion by CDP ……….... 87
Table 23. Summary of Logistic Regression Analysis for Variables Predicting Program Completion ………………………………….. 88
Table 24. Summary of Recidivism by Diversion Program ………………… 89
Table 25. Summary of Logistic Regression Analysis for Variables Predicting Recidivism ……………………………………………… 90
Table 26. Analysis of Variance for Days to Re-offense and Diversion Program …………………………………………………. 92
Table 27. Summary of Regression Analysis for Days to Reoffending by Demographic and Legal Characteristics ………………………… 93
Table 28. Summary of Offense Escalation by Diversion Program ………. 94
Table 29. Summary of Logistic Regression Analysis for Variables Predicting Offense Escalation ……………………………………. 95
Table 30. Characteristics of Sample by Offense Type ……………………. 97
Table 31. Logistic Regression Analysis for Programmatic Components Predicting Program Completion …………………………………. 99
Table 32. Logistic Regression Analysis for Programmatic Components Predicting Recidivism …………………………………………….. 100
Table 33. Logistic Regression Analysis for Programmatic Components Predicting Offense Escalation …………………………………… 101
Table 34. Juvenile Responses to Questionnaire ………………………….. 103
Table 35. Parent/Guardian Responses to Questionnaire ……………….. 105
5 List of Figures
Figures Page
Figure 1 …………………………………………………………………. 32
6 Acknowledgements and Dedication
There are many people to thank for their support and encouragement over the years as I completed this doctoral program. First is the chair of my dissertation committee, Dr. Victor Groza, who took me by the hand and would not let got until this dissertation was completed. His patience and guidance meant more to me than I could ever acknowledge on this page, so I just hope he knows how truly grateful I am. To the members of my committee – Dr. Kathleen Farkas,
Dr. David Miller, and Dr. Emilia McGucken – thank you for your willingness to serve on my dissertation committee and for your unwavering patience and support over the years. Dr. Mark Singer, thank you for your vision and your hard work on the Mandel Leadership Fellows Program and to the Mandel family for their belief in the worth of the program. I am honored to be graduating as a
Mandel Fellow.
While I’m acknowledging doctors, I would be remiss if I did not thank a special group of Ph.D.s. First are the members of my cohort for their support and friendship since day one of our doctoral journey. Through all of the papers and projects, career changes and family sorrows, e-mails and get-togethers, we have formed a wonderful bond. If I had to do this all over again (and thank God I don’t) there is not another group of individuals with whom I’d rather travel this road. Dr. Kay Benjamin and Dr. Karen Popovich, thank you for knowing what I needed in order to get going and keep going on the dissertation trail. Your friendship, help, and guidance proved crucial.
7 I greatly appreciate the cooperation and the assistance of those who work
with the Cuyahoga County Community Diversion Programs, particularly Heather
Corcoran, Nancy Stroman, and Doug Stolarski. I could not have completed this
research project without each of you. Thank you, not only for your help with the
project, but for your commitment to the children of our county. To the faculty and
staff of MSASS, thank you for all of your work and your assistance, especially
Helen Menke, doctoral program D.A. And to my friends and family who are sick
of hearing me say, “I can’t. I have to work on the you know what” – ask me now.
Finally, I dedicate the completion of this dissertation and completion of the
program to the memory of three women who played very important roles in my
doctoral journey. Arol Shack, beloved member of the MSASS family, recruited
me to this program. Her help, support, kindness, encouragement, and caring
ways were a mainstay of the Ph.D. program. It is only fitting that I acknowledge her in this dedication. Elizabeth B. Lambright, who was 96 years old when she died, was my personal cheerleader and a source of inspiration to me. I was a doctoral student when we met and a doctoral student when she died. She spent the last years of her life taking great pride in my accomplishments. I regret that she is not here for this one, so I dedicate this work to her memory as well. Last, but certainly not least, anything and everything “good” that I do or that I am I owe to my mother, Ruth E. Stewart. She was, and still is, unquestionably my greatest source of inspiration. To coin a phrase from a current U.S. presidential candidate, “It’s in my DNA. It’s who I am”. My mother was my certainty in a very
8 uncertain world. So, as with all that I accomplish, I dedicate this dissertation and this degree to the loving memory of my mother.
9 An Outcomes Study of Juvenile Diversion Programs on Non-Serious Delinquent and Status Offenders
Abstract
by
MELODY J. STEWART
This study uses a retrospective cohort design to explore the impact of
juvenile diversion programs on non-serious delinquent and status offenders. The study examines the influence of demographic characteristics, legal factors, and diversion program components, on short term program outcomes for two juvenile
Community Diversion Programs in Cuyahoga County, Ohio.
The data for this study are from a non-random sample of juveniles (N =
533) who were referred to one of the two programs between January 1, 2002 and December 31, 2004 as a result of having been charged with committing a delinquent act or status offense. The short term outcomes examined were program completion, recidivism, and offense escalation. The examination period was for one year from the date of program hearing with a one year follow up for those youth who completed the program or had their case dismissed.
Using the perspectives of restorative justice and reintegrative shaming as theoretical guides, this study examines the relationship between diversion programmatic components and outcomes, hypothesizing that the amount of exposure to reintegrative program components would be positively associated with program completion and negatively associated with recidivism and offense escalation.
10 The results of the study did not support the hypotheses. Although one
Community Diversion Program utilized significantly more reintegrative components than the other program utilized, there were no significant differences in program outcomes between the two diversion programs. Regression analyses also failed to show any significant interactions between reintegrative program components and outcomes. Findings did reveal significant interactions between educational components and therapeutic components on program outcomes.
Additionally, the study examined whether the diversion programs impact status offenders differently from delinquent offenders and whether any demographic and/or legal factors influence program outcomes. No differences in impact were found between the two categories of offenders. However, demographic and legal factors were found to interact with program outcomes.
The study suggests that future research on juvenile diversion program components and program outcomes would help to inform decision making with regard to program operations and also serve to help assess the programs’ overall effectiveness.
11 AN OUTCOMES STUDY OF JUVENILE DIVERSION PROGRAMS ON NON-SERIOUS DELINQUENT AND STATUS OFFENDERS
Chapter 1
Introduction
Committing a delinquent act is not an uncommon occurrence in the life of
a juvenile, and most young people who commit delinquent offenses eventually
grow out of such behavior (Cuellar, McReynolds & Wasserman, 2006; Moffitt &
Caspi, 2001; Moffitt, 1993; Arnett, 1992). Research on the influences of
delinquent behavior is often cited in an effort to prevent onset juvenile offending
(Mulvey, Arthur & Reppucci, 1993; Klein, Alexander & Parsons, 1977). However,
for those youth who do offend, some continue to do so throughout their juvenile
years with a portion of those re-offenders committing more serious crimes. The impact of juvenile crime on society is vast in terms of victimization and costs
(Loeber & Farrington, 1998; Heilbrun, Sevin Goldstein & Redding, 2005)
In addition to repeat offending during their juvenile years, some youth continue offending into adulthood (Benda, Corwyn & Toombs, 2001). Many adults incarcerated throughout the United States started offending when they were juveniles (Sherman, Gottfredson, MacKenzie, Eck, Reuter, & Bushway,
1998) and many adult sex offenders report that their sexual offending started during their youth (Zimring, 2004; Martin & Pruett, 1998). Analyzing the potential benefits of saving a youth who is at high-risk for delinquent offending, Cohen
(1998) estimates that a delinquent youth who subsequently continues committing criminal acts into adulthood costs taxpayers approximately $1.3 to $1.5 million in
12 external costs. These costs include lost wages and labor productivity, medical
expenses, and stolen property. He estimates further that eliminating duplication
between crimes committed by those who are both career criminals and heavy
drug users yields an overall monetary value of “saving a high-risk youth” at $1.7
to $2.3 million in external costs.
Even with efforts aimed at prevention, juvenile delinquency remains a
societal problem (Heilbrun et al., 2005). It follows then that efforts aimed toward
stemming the tide of repeat offending is as important as those efforts aimed
toward delinquency prevention.
Juvenile diversion programs have become a popular mechanism used in
an effort to prevent usually non-serious delinquent youth, from subsequent offending and further penetration into the juvenile or criminal justice systems.
The structure of the programs and their operation vary, but the overall goals are essentially the same: namely, to address juvenile deviant behavior informally in the community in an effort to prevent subsequent offending. In addition to there being no set structure or standards of operation for the various types of juvenile diversion programs, very little attention has been given to the programs’ components and how those programmatic components impact the youth who participate in the diversion programs.
Overview of Problem
Juvenile delinquency has long been a great concern in the United States.
Despite the fact that juvenile offending has decreased in recent years (Snyder,
13 2006), delinquency and the consequences thereof remain the subject of study
and debate (Heilbrun et al., 2005).
Juvenile delinquency encompasses criminal acts committed by youth who
have not reached the age of majority. In Ohio and in most states, the age of
majority is eighteen. In addition to committing crimes, juveniles are deemed
delinquent also when they commit a status offense. A status offense is an act
that is prohibited by virtue of the fact that the actor is a child – not yet the age of
majority. A status offender’s conduct violates social regulations moreso than a
criminal law. Examples of status offenses include truancy from school, violating curfew, and running away from home. If the same acts were committed by an adult, they would not be punishable.
Delinquent offenses are usually categorized as non-serious and serious offenses. Non-serious delinquent offenses are status offenses and acts that would be deemed criminal – in most cases a misdemeanor - if committed by an adult but are termed delinquent acts due to the presumed lack of criminal capacity in youths. Some examples of these acts include, trespassing, petty theft, vandalism, and disorderly conduct.
Serious delinquent offenses involve more criminal culpability and are usually felony offenses. In some instances, serious delinquent offenses are severe enough in nature that instead of being adjudicated delinquent for having committed the offense, juveniles are bound over to the criminal court and tried as adults.
14 The FBI Crime Statistic Reports indicate that in the year 2000, an
estimated 2.4 million juvenile arrests were made across the United States for criminal, delinquent and status offenses (FBI Uniform Crime Reports, 2000).1
This figure is down 5% from arrests made in 1999 and 15% lower than those made in 1996 (Snyder, 2002). The year 2004 saw the lowest rate of juvenile offending since 1980 (Snyder, 2006). Still, in 2005, 2.1 million juveniles were arrested representing 3 percent fewer arrests than in 2004 (Butts and Snyder,
2007; FBI Uniform Crime Reports 2005). Males make up the majority of juvenile arrest, however, female arrests have increased over the years while overall juvenile arrests have declined (Snyder, 2006).
In its assessment of access to counsel and quality of representation in delinquency proceedings in Ohio, the American Bar Association noted, “Ohio’s juvenile justice system is overloaded with youth who could be served through less restrictive means” (ABA, 2004, p. 1). This finding was promulgated by the discovery: 1) of large numbers of non-violent and minor offenders held in Ohio’s detention centers; 2) that female offenders do not have the same access to community based services as male offenders; and, 3) that African-American juveniles in the system are significantly overrepresented from the detention phase of the process to the commitment phase (ABA, 2004). The report noted that “[w]hile there is a lack of any consistent and accurate data on the number of youth detained in Ohio’s thirty-five detention centers, data from some centers indicate as many as 70-80% of youth are held on allegations of non-felony conduct” (ABA, 2004, p. 1). The report goes on further to note the following:
15 - Most of Ohio’s community correction facilities, designed to keep
youth who need residential treatment in less restrictive
placements closer to their own communities, do not have beds for
girls.
- Female offenders have treatment needs that are often different
than males; some current programs and services are inadequate
to meet the unique mental health and developmental issues facing
girls.
- African-American youth are overrepresented in detention two to
three times more than their white counterparts.
- Of those youth committed to the Ohio Department of Youth
Services for the 2000-2001 fiscal year 46.7% were African
American, representing four times the corresponding percentage
of African American youth to the general population. (ABA, 2004).
Not surprisingly, some of the report’s recommendations were to: 1) reduce the disproportionate rate of minority juveniles in the system; 2) address the particular needs of females in the system; and 3) reduce the amount of juveniles incarcerated for minor offenses.
Population estimates indicate that there were roughly 1.4 million people who lived in Cuyahoga County in the year 2000. Cuyahoga is the largest of eighty eight counties in Ohio. Of the 1.4 million people living in the county, approximately 348,000 were juveniles age 0 to 17, and 155,600 were between the ages of 10 and 17 (Stahl, Kang & Wilt, 2003). In that same year, nearly
16 4,800 delinquency cases and 700 status offenses were filed in the juvenile court
(Stahl, et. al 2003). The 2006 Annual Report of the Cuyahoga County Juvenile
Court notes that over 13,700 status and delinquency cases filed that year
(Cuyahoga County Juvenile Court 2006 Annual Report). This figure is up 25%
from 2003 when just over 11,000 were filed (Cuyahoga County Juvenile Court
2003 Annual Report). So, although juvenile offending has decreased nationwide,
it remains an apparent problem in Cuyahoga County.
Although juvenile crime has decreased over the years, criminally deviant
youth2 remain a large societal concern and continues to negatively impact communities (Snyder & Sickmund, 1999; Butts and Snyder, 2007). Traditional
methods of, and formal approaches to, addressing juvenile crime and
delinquency have been criticized as ineffective (Davidson, Redner, Amdur, &
Mitchell,1990). Some even believe that penetration into the juvenile justice
system, at least with regard to first time delinquent offenders and those youth
who commit status offenses, perpetuates rather than prevents re-offending
(Minor, Wells, Soderstrom, Bingham & Williamson, 1999; Butts and Harrell,
1998). Due in part to such criticisms, which date back to the 1970s, alternative
approaches to dealing with juvenile delinquency started gaining popularity.
Overview of Juvenile Diversion Programs
In the late 1960s, when a marked increase in juvenile arrests was noted
(Heilbrun et al., 2005), an additional concern was the fact that many children
were being arrested and processed for status offenses – an estimated 25%
17 (Elrod and Ryder, 1999; Laub 2002). Responding to the amount of status
offenses and delinquent cases being formally processed through juvenile courts
nationwide, and coupled with the belief that the courts were not the best way to
address all forms of delinquency, President Lyndon Johnson formed a
Commission on Law Enforcement and the Administration of Justice (President’s
Commission Task Force Report, 1967; Bynum & Thompson, 1999). The
commission recommended that there be alternatives to adjudicating delinquents
through juvenile court. The creation of diversion programs was one such
alternative (Binder & Polan, 1991; Bynum & Thompson, 1999).
Although the formal implementation of diversion programs dates back to the 1960s (Bynum & Thompson, 1999), the diversionary approach has a historical foundation in the origin of the juvenile court over 100 years ago (Feld,
1999). The juvenile court, created in 1899, was designed to divert or remove juveniles from adult courts. This was the beginning of the child-saving movement of juvenile justice.
The child-saving movement of the juvenile court was a collective effort on the part of communities to control delinquent children by offering constructive and
supervised activities as opposed to mere punishment or confinement. Prior to
the movement, offending youth were treated the same as adults. They were tried
in the same courts and were incarcerated in the same jails. As society began to
rethink how juvenile offenders differed from adult offenders, institutions were
established in the United States with the goal of controlling problem youth and rectifying their behavior (Elrod & Ryder, 2005). Juvenile justice reformers argued
18 that children should not be treated the same as adults and that efforts to address juvenile offending should concentrate on rehabilitation as opposed to punishment.
The juvenile court became a judicial-welfare alternative to the adult court system (Feld, 1999). In doing so, the court took on the role as parens patriae, which is the state acting in the best interest of the child. The child saving movement that swept the country was initially met with enthusiastic optimism about the positive impact this juvenile justice reform would have on the lives of troubled youths (Roberts, 2004). This optimism eventually waned when comprehensive rehabilitation services were not provided as planned, juveniles were being confined to facilities that were similar to adult facilities, and juveniles were being denied important constitutional rights (Elrod & Ryder, 2005; Roberts,
2004). Recognizing the shortcomings of the juvenile court as an effective diversionary strategy, juvenile diversion programs emerged as a dominant movement across the United States (Roberts, 2004).
Frazier and Cochran (1986) describe diversion as refraining from sanctioning a juvenile offender or imposing some sort of rehabilitative sanction instead of subjecting young offenders to traditional punitive sanctions. In more modern terms, diverting a youth generally means addressing a juvenile offense through an alternative to the formal adjudication of juvenile court. Bynum and
Thompson (1999) give a similar definition, referring to diversion as a channeling out of youthful offenders from the juvenile justice system.
19 Juvenile diversion has been described as one of “the most important issues in juvenile delinquency and youth antisocial behavior” (Heilbrun et al,
2005, p. 3). Programs that utilize a diversionary approach are part of a justice system movement toward more community based corrections and have become a widely used response to juvenile offending (Roberts, 2004). More specifically, juvenile diversion programs are a popular alternative to the adjudication through juvenile court of unruly and delinquent youth who have not committed serious or egregious offenses (Potter & Kakar, 2002). The programs help these young offenders avoid the negative effects of formal court adjudication and possibly confinement (Roberts, 2004). Diverting juvenile offenders, instead of formally processing them through the court, is also less costly and provides an opportunity for youth to correct deviant behavior with the help and support of their families and the community (Patrick & Marsh, 2005; Gavazzi, Wasserman,
Partridge, & Sheridan, 1999; Sullivan, Veysey, Hamilton, & Grillo, 2007;
Hamilton, Sullivan, Veysey, & Grillo, 2007).
Diversion programs differ in their structure and operation, target specific populations and are run by various entities (Hamilton et al., 2007; Cuellar et al.,
2006; Campbell & Lerew, 2002; Hassett-Walker, 2002; King, Holmes,
Henderson, & Latessa, 2001; Leiber & Stairs, 1999; Scofield & Davidson, 1995;
Gilbert, 1977). Examples of the different kinds of diversion programs include teen or youth courts, mental health programs, cultural programs, programs that focus on mentoring and skill training, and law-related education (LRE) diversion programs (Fox, Minor, & Pelkey, 1994; Blechman, Maurice, Buecker, & Helberg,
20 2000; King, et al., 2001; Harrison, Maupin & Mays, 2001; Dick, Pence, Jones &
Geertsen, 2004; Cuellar et al., 2006).
Programs are funded by a variety of sources including federal, state, and
local governmental agencies, grants, and private and corporate donations (H.
Corcoran, personal communication February 6, 2005). Some programs charge
families a nominal fee for program participation; however, it is not unusual to
allow a youth to participate in a program even if the family cannot afford the fee
(D. Stolarski, personal communication, February 17, 2005). The majority of
funding is used for staffing but most programs also rely heavily on volunteers (H.
Corcoran, personal communication, January 12, 2005).
The diverse components of diversion programs and their impact on
diverted youth are virtually unobserved. Diversion programs continue to be
created and funded without a systematic examination of their outcomes.
Furthermore, there is little to no guidance on program structure, the use of
personnel, or uniformly accepted ways of how to assess program outcomes.
These points demonstrate the importance of this research from a practice
perspective. Examining outcomes will help shed light on how specific program
components impact status and delinquent offenders.
From a policy perspective, those who oversee and operate diversion
programs can be better informed when conducting cost – benefit analyses. In
this regard, closer attention can be given to whether diversion program funding is
a good expenditure of resources and whether funding should continue at similar
or different levels.
21 Cuyahoga County Juvenile Diversion Program
In 1998, the juvenile division of the Cuyahoga County Court of Common
Pleas established a juvenile diversion program called the Community Diversion
Program (CDP) - as a way of reducing formal court adjudication of juveniles for status and/or first time non-serious delinquent offenses. Stated in the court’s description of the program, the CDP “empowers communities to tailor treatment and sanctions to fit the needs of the individual child and the concerns of the community, as well as offering the child the opportunity to avoid an official
Juvenile Court record” (Cuyahoga County Juvenile Court 2003 Annual Report, p.
10).
There are forty-nine community diversion programs operating in cities throughout Cuyahoga County and under the auspices of the juvenile court.
Funding for the programs is provided by the juvenile court. However, some of the programs also receive funding from (JAIBG) - Juvenile Accountability
Incentive Block Grant. In 1999, nine hundred and nine juveniles were referred to the programs county wide. In 2002, over twenty one hundred youths were referred to the CDPs. But in 2003, the number of juveniles diverted dropped to two thousand fourteen.
The CDPs that are the focus of this study are programs in the cities of
East Cleveland and Maple Heights. According to Corcoran (personal communication, February 6, 2002), the East Cleveland Community Diversion program utilizes more restorative justice/reintegrative program components than any of the other county run programs. The Maple Heights program, according to
22 Corcoran, (personal communication, February 6, 2002), is more typical of how the majority of the county diversion programs operate. Some historical background on the programs follows
The East Cleveland Community Diversion Program
Originally a village incorporated in 1895, East Cleveland is the oldest suburb of Cleveland, Ohio. According to the 2000 census, the population was just over 27,200 and in 2003 the estimated population was 26,200. The population was approximately 25,200 in 2006. (Encyclopedia of Cleveland
History, 2007; U.S. Census Bureau, 2007).
The East Cleveland CDP is currently run through the East Cleveland
Neighborhood Center, Inc. (ECNC). The ECNC was incorporated in 1986 and provides recreational, educational and social services primarily to residents on the northeast side of greater Cleveland as well as to residents in the city of East
Cleveland proper. The program operates by a cooperative effort between the neighborhood center and other community agencies that provide services to youth who come through the program. The program is managed by a coordinator who is a full-time employee of the ECNC. In addition to managing the diversion program, the coordinator of the East Cleveland CDP has other responsibilities with the neighborhood center (N. Stroman, personal communication, February 17, 2005).
Initially in 1999, the East Cleveland police department operated the diversion program with the Chief of Police bearing the primary responsibility for
23 its operation. In 1999, no juveniles came through the diversion program and in
2000, four juveniles were diverted. In 2001, the responsibility for the program was turned over to the neighborhood center and expanded. The ECNC is a not- for-profit organization and a member of the greater Cleveland Neighborhood
Centers Association (NCA). The NCA is a collaborative, comprised of twenty associate members that formed a mother agency of settlement houses – neighborhood centers.
From 2001 to 2006, the program used six volunteer magistrates, five of whom were the hearing magistrates for the subjects of this study.3 The hearings were held on Saturday mornings in the courtroom of the East Cleveland
Municipal Court.
Juveniles who were referred to the program lived primarily in the cities of
East Cleveland, Euclid or a northeast portion of Cleveland close to East
Cleveland’s borders. However, youth from the eastern suburbs of Richmond
Heights, Lyndhurst and Mayfield Heights have also been diverted through the
East Cleveland program. The East Cleveland program has served over 400 juveniles from 2001 to present (N. Stroman, personal communication, April 26,
2007).
The Maple Heights Diversion Program
The city of Maple Heights is a southeast suburb of Cleveland, Ohio. It was incorporated as a village in 1915 and as a city in 1932 (Encyclopedia of
Cleveland History, 2007). According to the 2000 census, the population in Maple
24 Heights was 26,100 and in 2003 the estimated population was approximately
25,300. The population was estimated at just under 24,300 in 2006 (U.S. Census
Bureau, 2007).
The Community Diversion Program in Maple Heights has been operating since 2001. Its coordinator has been the same for the entire time. There was one volunteer magistrate who conducted most of the hearings but the coordinator also conducted a small number of hearings when the volunteer magistrate was unavailable. The hearings are held in a conference room at city hall in Maple
Heights.
The Maple Heights program is operated through its police department and the coordinator is a retired police officer who works for the city in multiple capacities. Youth who are referred to the Maple Heights program live in the city of Maple Heights or in a portion of the city of Cleveland that abuts Maple Heights.
The Maple Heights program also takes juveniles who commit an offense within the city regardless of where they live. The program has served over 500 juveniles since 2001 (D. Stolarski, personal communication, April 18, 2007).
25 Chapter 2
Theoretical Perspectives
Labeling Theory
The theoretical foundation upon which diversion programs are premised is
labeling theory (Klein, 1986; Schur, 1971). The theory espouses the view that if you label a child delinquent, he is stigmatized by the label and likely to act accordingly. Diversion programs, therefore, are alternatives to formally processing juveniles as – and thereby labeling them – delinquent.
Three main goals of diversion – reducing stigma, reducing social control and reducing coercion – reflect the theory’s premise that the unintended negative effects of formal processing through juvenile court can outweigh the benefits of services or assistance that may be provided. Labeling theory, a societal reaction theory, originates from the premise that criminal behavior is a social process – an outcome of specific types of human interaction – not an objective phenomenon (Lemert, 1951, 1972; White & Haines, 2000). The focus of concern for the labeling perspectives “is with the nature of the interaction between
‘offender’, ‘victim’ and criminal justice ‘officials’” (White & Haines, 2000, p.80).
Officials, those in the positions of power, determine what conduct constitutes
criminal or deviant behavior and define it as such. During this labeling process,
stigmatization can occur and the offender may then act in accordance with his or
her label. The stigmatization not only affects how society perceives the offender,
but also how the offender perceives himself. Labeling perspectives, therefore,
26 advocate diverting offenders from the stigmatizing aspects of the criminal justice
system.
Tannenbaum’s Crime and the Community (1938) is acknowledged as
establishing the rudimentary elements of labeling theory (Laub, 2002). In his work, Tannenbaum espouses the view that regardless of the approach one takes in rectifying bad behavior, if it is done in a way that tags, labels, emphasizes, suggests or defines one as bad, the person becomes the very thing he is described as being (Laub, 2002 citing to Tannenbaum, 1938).
Lemert (1951) further developed the concept of labeling theory focusing on three distinct components: societal reaction, primary deviance, and secondary deviance. Societal reaction is the outward expression of disapproval by others of deviant behavior in addition to their attempts to control such behavior. Primary
deviance is the first act of deviant behavior or norm violating. Secondary deviance is deviant behavior that stems from society’s reaction to the primary deviance and is said to derive from the perception of oneself as deviant. Under
Lemert’s analysis, societal reaction and social control are important variables to
examine when theorizing about deviant behavior. With this analysis, Lemert is
credited with being one of the first theorist to offer an alternative to positivist theories of crime (Laub, 2002); that is, a theoretical position different from the belief that deviant behavior can be understood by simply examining traditional ideas about crime.
Lemert’s application of labeling theory was primarily focused on juvenile crime and delinquency during the 1960s and 70s (Laub, 2002). In 1973, Schur
27 extended labeling on secondary deviance in the analysis of the harmful effects of
stigmatizing, calling for radical non-intervention when addressing juvenile
offenses. In other words, Schur’s position is - to the extent possible, young
offenders should be free from official intervention or diverted from formal
adjudication in order to avoid the stigmatization associated with being labeled
delinquent or deviant.
Restorative Justice
Restorative justice is also a philosophical framework under which
diversion programs operate. Restorative justice is a type of informal justice that
focuses more on the harm caused to the victim and the community rather than
the violation of the law committed by the offender (Bazemore & Schiff, 2005).
Matters are resolved when offenders take responsibility and make amends for
the harm done, and the community gives the offender a chance to reintegrate
into the community (Olson & Dzur, 2004). The restorative justice approach
focuses on improved outcomes in addition to being concerned with the problems
surrounding formal adjudication of offenders through juvenile court (Bazemore &
Schiff, 2005).
A restorative justice technique, reintegrative shaming, extends the
concepts of labeling and stigma rooted in labeling theory. This theoretical
extension draws upon existing criminological theories to offer an explanatory
theory based on the concept of shame. It further asserts that not only should
society avoid labeling a youth bad or delinquent and thereby potentially bringing
28 about secondary deviance, we should make sure that negative focus is placed on the offense itself and not on the young offender (Braithwaite, 1989).
Furthermore, just as much emphasis should be placed on reintegrating juveniles into society as there is on emphasizing the wrongfulness of their transgressions
(Braithwaite, 1989; Hay, 2001)
Reintegrative shaming is a method of condemning actions such that the condemnation invokes remorse in the offender. The condemnation is immediately followed up with mechanisms to reintegrate the offender into the community. Reintegrative shaming methods are therefore viewed as positive punishment for offenses committed (James, 1995). This means that stigmatizing experiences are minimized as much as possible. Masters (1997) points out that
“[t]he important contribution of reintegrative shaming is that it illustrates the dangers inherent in any essentially stigmatic encounters, and the crucial need to make any practice predominantly reintegrative” (p. 238).
Braithwaite (1989) insists that condemning or shaming an offender must be done in such a way that it serves to bring the offender back, or closer, to society rather than isolating him. He distinguishes interventions as disintegrative and reintegrative. Interventions that are reintegrative are said to likely reduce reoffending. Disintegrative interventions stigmatize and potentially submerge young offenders into further deviance, clearly an unintended consequence of the interventions. The theory then emphasizes positive ways to deal with offenders in an effort to avoid negative social consequences (White & Haines, 2000).
29 Although the literature demonstrates that reintegrative shaming has been subject to scant empirical testing (Hay, 2001; Zhang, 1995), there is support for use of the theory in studies. In a study that operationalized reintegrative shaming as a combination of deviance disapproval and forgiveness, Zhang and Zhang,
2004 conducted an exploratory study using data from the first two waves of the
National Youth Survey. The researchers “created two product terms with parental and peer disapproval of delinquency and their forgiveness of the delinquent to represent parental and peer reintegrative shaming, respectively, and then examined[d] their effects on involvement in predatory offenses” (p.
447). The study met with mixed results.
The researchers found no effect of peer or parental reintegrative shaming on predatory offending in either wave of the Survey. This result provided no support to the hypothesis key to the study: that reintegrative shaming is a significant predictor of predatory offending. However, the study did find that
parental forgiveness and peer shaming, separately, have significant and negative
effects on the likelihood of youth being involved in predatory offenses in both
waves. The researchers concede that using a secondary data analysis may lack
the sensitivity necessary to capture the subtleties of reintegrative shaming, but
they nonetheless encourage the theoretical exploration of the main effects of
reintegration of the delinquent youth, and shaming of the delinquent act.
In recognition of the negative effects of labeling against which diversion
was intended to protect, Makkai and Braithwaite (1994) note that it may be the
actual intervention themselves – the shaming techniques used - that are sources
30 for the stigmatizing effect, not whether a youth is diverted or formally adjudicated.
They proposed that interventions can have crime-amplifying effects as well as
crime-inhibiting effects. Shaming that is reintegrative, designed to bring about
crime-inhibiting effects, is the response to the juvenile’s deviance that denounces
the wrongdoing and at the same time encourages accepting the youth back into
society. Key to the application of reintegrative shaming is that the focus is on the
actions of the offenders, not on the offenders themselves, and it is those actions
which are the objects of the shaming.
Braithwaite (1989) believed that reintegrative shaming is key to
controlling crime, particularly predatory crime. He also surmised that the positive
effects of reintegrative shaming may be enhanced when there is a mutual
attachment shared between those who do the shaming and those who are
shamed. According to Braithwaite’s theory then, diversion programs that are more reintegrative should meet with more positive outcomes. Additionally, those programs where juveniles seem to develop an attachment to components or personnel of the programs should also yield positive results.
Based on the above analysis, the predatory nature of an offense committed may be important to understanding whether diversion programs impact status offenders the same as delinquent offenders. Generally, reintegrative shaming is thought to be applicable when there is a consensus about the moral wrongfulness of an offense. Delinquent offenses usually fall into this category. When the moral wrongfulness of an offense is in question, reintegrative shaming is thought not to be influential (Hay, 2001). Because
31 status offenses are not predatory by nature and are referable to diversion due
only to the offender being below the age of majority, the moral wrongfulness of these offenses, for example, being truant from school or outside past curfew, arguably may not meet with consensus. Therefore, whether diversion programs impact delinquent offenders and status offenders differently should be examined
(Shelden, 1989).
The theoretical model for this study is represented in Figure 1.
Figure 1.
Outcomes of Diversion Programs
The literature on juvenile diversion is still in the developmental stage.
Most studies are considered dated by social science standards. However, it is
32 clear that historically, the primary reasons for establishing juvenile diversion
programs were to stem the tide of increasing juvenile court caseloads and help
improve the efficiency of the juvenile court (Whitaker, Severy, & Morton, 1984;
Palmer and Lewis, 1980). In addition to the proposition of efficiency, proponents
of juvenile diversion argue that the programs are also more cost effective than
formal juvenile court processing, minimize stigma, reduce the level of coercion inherent in formal adjudication, and reduce subsequent offending (Polk, 1989;
Feld, 1999; Whitehead & Lab, 2001).
Modern day goals of diversion are essentially the same: to reduce costs, stigma, social control, and recidivism in addition to providing services to juveniles
(Roberts, 2004’ Elrod & Ryder, 2005, Heilbrun et al., 2005; H. Corcoran, personal communication, January 12, 2005). The continued use of diversion programs has served to shift juvenile delinquency policy more to the community (Heilbrun et al.,
2005). This shift is in keeping with the “savings movement” discussed in the previous chapter that has been the mantra of the juvenile justice system for over
100 years (Elrod & Ryder, 2005, Roberts, 2004).
Some studies on diversion have demonstrated a reduction of cases
processed through the juvenile courts (Pogrebin, Poole & Regoli, 1984; Latessa,
Travis & Wilson, 1984). However, Ezell (1992) commented that diversion
programs of the 1970s and 1980s met without much overall success in that they
did not appreciably decrease cases formally processed. Ezell urged program
developers to learn and benefit from past mistakes. Specifically, he encouraged
33 researchers to explore what alternatives work best for particular youths and
which type of juveniles are most positively affected by diversion interventions.
Research has not focused much on these aspects of diversion. Programs continue to be developed and implemented without any real theoretical guidance
on program components and structure. Furthermore, sparse attention is paid to how diversion impacts particular segments of the juvenile offending population.
The continued proliferation of diversion programs without focusing specifically on
program impact suggests a level of contentment examining traditional program
outcomes only, namely recidivism.
In 1984, Osgood and Weichselbaum investigated the strengths and weaknesses of diversion programs in accordance with generally articulated goals
of diversion. The authors noted that there are primarily two types of research
evaluating the success of diversion programs: 1) research that focuses on the
impact of programs on youth – impact namely in the form of recidivism, and 2)
research that determines whether diversion programs are indeed true
alternatives to formal processing through the juvenile court. Osgood and
Weichselbaum, (1984) examined the latter in their study. Looking at nine
different diversion programs in seven U.S. metropolitan cities, the authors
concluded that when programs match the theory behind diversion, participating in
a diversion program is a substantially different experience from having a case
formally adjudicated through the juvenile court. Labeling theory was the
theoretical premise for this study. The study found primarily that diversion
programs were better with serving the needs of youth while the justice system
34 focused more on coercion and social control. The authors noted that diversion
programs’ service providers were less likely to view as delinquent, the young
offenders referred to the programs.
What remains unclear about juvenile diversion programs is: How are
programmatic components created and implemented in accordance with a
theoretical framework? How do programs and their components influence
outcomes? In addition, how effective are diversion programs at achieving their
goals? Little attention has been given to specific diversion programming and
interventions employed in an effort to find out what impact, if any, the programs are having on diverted youth and whether some interventions or components are more effective than others.
Effectiveness of Diversion Programs
The level of effectiveness at reducing subsequent delinquent behavior and whether diversion programs are effective at all has been, and continues to be, a topic of great debate. Results have been inconclusive on whether diversion programs help reduce recidivism (Polk, 1989; Ezell, 1992).
There are studies that indicate the success of diversion with regard to recidivism (Thornton, Barrett, & Musolf, 1972; Duxbury, 1973; Forward, Kirby, &
Wilson 1975; Quay & Love, 1977; Palmer, Bohnstedt & Lewis, 1978; Palmer &
Lewis, 1980; Pogrebin, Poole & Regoli, 1984; Davidson, Redner, Blakely,
Mitchell, & Emshoff, 1987). For example, in the study conducted by Pogrebin et
al., (1984), the researchers examined the effects of participating in a juvenile
35 diversion program on recidivism. The study used an experimental designed to
randomly assign youth to participate in a Colorado diversion program versus
being given a lecture and let go. Pogrebin, et al. (1984) concluded that the
program’s services helped to lower the recidivism rate. The recidivism rates of
youth in the diversion program were found to be better than the recidivism rates of juveniles who were not diverted. The program of this study also attempted to provide reintegrative program components to the diverted youth in an effort to increase the self esteem and sense of responsibility of juveniles. However, the impact of the components was not collected or measured.
In another example, Davidson et al., (1987) conducted a longitudinal experimental study comparing various delinquency treatments of community- based diversion programs in a Midwest city. The study took place over a four year period and four interventions were compared and contrasted with a placebo group of juveniles and a group of juveniles who received usual treatment from the juvenile court. Recidivism was measured using official filings and self- reports. The study found that the treatment conditions using interventions outside the formal processes of the juvenile court produced lower rates of recidivism than 1) an intervention within the juvenile justice system, 2) the placebo group, and 3) the group processed through the usual formal juvenile court system. The study also found no significant differential effects on delinquency that was self-reported. The authors concluded that the study’s findings support the notions of labeling theory as it appears that interventions
36 conducted outside the formal processes of the juvenile court may beneficially influence recidivism.
On the other hand, some studies indicate no significant positive impact from diversion on recidivism or subsequent delinquent behavior (Venezia &
Anthony, 1978; Haapanen & Rudisill, 1980; Rojeck and Erickson 1982; Rausch
1983). For example, Venezia & Anthony (1978) conducted a study evaluating the impact of youth service bureaus in Wisconsin. The evaluation was based primarily on measuring changes in the areas of delinquency and school difficulty among other measures such as family conflict. With few exceptions, all of the youth bureaus had as their goal to reduce juvenile delinquency. The study found that the projects that had a community change focus had more positive results than those projects that provided direct services to individuals. The study also found that youth involvement in project activities had a beneficial effect.
In another example, Haapanen & Rudisill found no impact of diversionary programs on subsequent deviant behavior. The study by Haapanen & Rudisill
(1980), evaluated nine diversion programs/youth service bureaus in California to assess changes in delinquent behavior of the participating youth compared to youth formally processed through the juvenile court and youth attending local schools. The authors found that the youth service bureaus had no measurable effect on the delinquent behavior of youth. Pre- and post study comparisons revealed that all groups had higher rates of delinquency in six and twelve month periods following intake or pre-test than they had in the period measured beforehand.
37 In addition to the debate on whether diversion programs are effective at helping to mitigate subsequent deviant conduct, there are questions regarding whether diversion programs are actually harmful to youth or bring about negative unintended consequences. In his summary of research on diversion programs,
Polk (1989) noted that empirical data on the impact of diversion programs indicate there are programs that not only are of no benefit in reducing future delinquent behavior, there are programs that may be harmful to the children they are intended to serve. Some studies have found an increase in self-reported delinquency among youth in diversion programs as well as an increase in youth who went through formal adjudication (Elliott, Dunford, & Knowles, 1978; Lincoln,
1976; and Lipsey, Cordray, & Berger 1981). For example, Lincoln (1976) found that a police diversion program seemed to have aggravated, rather than deterred, recidivism.
Also problematic to the implementation of diversion programs has been the issue of net-widening. Net-widening is a term used to describe the concept of including youth in a diversion program who would otherwise have been left alone in the absence of the program. Net-widening has become a major concern of those skeptical about the effectiveness of juvenile diversion programs and some studies have found an increase in the amount of juveniles who are informally processed (Bloomberg, 1979; Palmer & Lewis, 1980; Macallair &
Males, 2004). In a study conducted by Macallair & Males, the researchers found that the city of San Francisco increased its detention bookings, particularly with regard to minority females, when it implemented juvenile diversion programs
38 targeting first time, non-violent offenders. The study revealed that more youth
were subject to significantly higher degrees of confinement and social control
than before the reform programs were implemented.
Recent research on diversion programs is sparse. Kammer, Minor and
Wells (1997) argue that the lack of recent research is partly responsible for the controversy and lack of understanding on diversion programs. More importantly, these authors attribute the mixed findings of studies to vast differences in the programs and methodology, such variation making generalization problematic.
They urge researchers to study the individual content of diversion programs so
that various programming initiatives can be compared and generalizations can be
properly qualified.
Summary and Research Questions
The primary motivation behind diversion programs is the belief that formally processing a youth through the juvenile justice system does not always serve the best interest of the youth or the community. Along with this motivating
factor is the belief that formally adjudicating young offenders will often increase
the likelihood of recidivism (Elrod & Ryder, 1999). To date, the research on
diversion programs has produced mixed results. The above discussion
demonstrates that there are still a number of concerns associated with diversion programs. However, despite the concerns, diversion remains a popular societal method of addressing delinquent conduct. Even researchers who found no substantial evidence for the efficacy of diversion argue that such findings should
39 not be the basis for abandoning the programs (Gensheimer, Mayer, Gottschalk,
& Davidson, 1986). This suggests that more thorough evaluation and basic research is needed.
Efforts to prevent subsequent reoffending is an important goal of diversion but examining specific programmatic components and how they impact diverted youth is important also. Ezell (1992) notes that future research should focus not only on whether diversion programs reduce recidivism, but should also examine which types of youth are more positively affected by certain interventions. He further urged researchers to “improve the descriptions and measurements of the diversion interventions so the programs can be replicated when they are found to be effective with certain youth” (p. 55). Lastly, Ezell (1992) urged researchers to examine the agency context of diversion programs – its administration and characteristics – noting the impact on delivery of services.
Guided by the theoretical philosophy of reintegrative shaming, this discussion leads to the following research questions:
Q1. Is reintegrative shaming positively associated with program completion?
Q2. Is reintegrative shaming negatively associated with recidivism?
Q3. Is reintegrative shaming negatively associated with offense escalation?
Q4. Do diversion programs impact delinquent offenders and status offenders differently?
40 The study examines whether one diversion program subgroup has outcomes that are statistically different from the other subgroup with regard to 1) program completion; 2) recidivism; and 3) offense escalation for those youth who reoffend. Based on prior research and the tenets of reintegrative shaming, this study tests the following hypotheses:
H1. Increased amounts of reintegrative shaming components
in a diversion program will lead to increased rates of program
completion.
H2. Increased amounts of reintegrative shaming components
in a diversion program will lead to decreased recidivism .
H3. Increased amounts of reintegrative shaming components
in a diversion program will lead to decreased offense escalation
for those youth who re-offend.
H4. Diversion programs will impact status offenders and delinquent
offenders differently.
The hypotheses of this study are in keeping with the basic theoretical premise of reintegrative shaming: that there should be a negative relationship between the use of reintegrative shaming and delinquent behavior. This premise
41 can be applied to both a micro level and a macro level of analysis. On a micro
level, those juveniles who are exposed to reintegrative shaming or program
components that are reintegratively shaming in nature should re-offend less than
those juveniles who are exposed to fewer components. Applying the theory on a
macro level, one could argue that communities using high levels of reintegrative shaming techniques should have low aggregate rates of delinquency.
42 Chapter 3
Methodology
Having reviewed the literature on labeling theory, restorative justice, reintegrative shaming, and diversion programs, this chapter discusses the study’s research methodology.
Defining the problem underlying this study, there is limited information on outcomes of juvenile diversion programs. In particular, very little is known about particular program components and what impact, if any, those components have on juveniles and on program outcomes. There have been studies that demonstrate some level of effectiveness for diversion programs (Pogrebin et al,
1984; Davidson et al., 1990; Thornton et al, 1972; Duxbury, 1973; Forward et al,
1975; Palmer & Lewis, 1980) but other studies have shown no effectiveness and warn about potentially harmful effects of diversion (Elliott et al, 1978; Haapanen
& Rudisill, 1980; Rojeck & Erickson, 1982; Rausch 1983; Lincoln 1976).
Examining these issues is important because juvenile diversion continues to be a popular alternative to formal juvenile court adjudication. Diversion programs are prevalent in several forms across the country, yet by and large operate in relative obscurity regarding program impact and outcomes. Also, there appears to be scant focus on how juveniles and their caregivers view the experience of going through a diversion program and even less attention given to how they define program success.
43 Research Design
The research design used for this study is a retrospective cohort design
examining data on short-term outcomes of two diversion programs. The primary
purpose of this type of design is to describe the rates of an outcome and to analyze associations between outcome rates and predictive factors (Kazdin,
2003). This design is well suited for the study because research on juvenile diversion programs is underdeveloped, the sample used for the study was not randomly chosen, and the research questions can be answered with existing data. Short-term outcomes are defined as results which happen in one year or less. The outcomes that are examined in this study are program completion, recidivism, and offense escalation.
The study employs a secondary data analysis in that it uses data previously collected for a purpose other than this study. Archival data, used for this study with permission from the juvenile court, were collected between the years 2002 and 2004 with a one year follow up period. In studies that apply secondary analysis as opposed to being an experimental study, the researcher has no control over the group designation of the study subjects. The subjects were selected for exposure conditions from naturally occurring or previously existing groups and subsequently analyzed for resulting outcomes.
In studies that are experimental in design, the researcher allocates the number of subjects randomly assigned to different comparison groups and regulates the experimental conditions. The subjects are then followed over time and the outcome event of interest is recorded, controlling for potential confounds.
44 Neither participation in a diversion program nor participation in a particular
program was controlled or manipulated for this study.
The research design is also correlational, meaning that the research examines the association between independent or predictor variables and dependent or outcome variables. Correlational studies investigate whether hypothetical relationships exist and any correlation is expressed in terms of strength and direction. Correlation does not infer causality. Prior knowledge development supports the use of this particular design for this research project
(Kammer, Minor and Wells, 1997). Descriptive and inferential data analyses are used and are discussed in more detail in the next chapter. There is also a cross- sectional, qualitative component to this study using questionnaires and interviews. This component will be discussed later in this chapter.
Sample and Selection Criteria
The sample for this study is non-random and comes from two of forty-nine diversion programs in Cuyahoga County, Ohio. The programs are in two inner- ring suburbs of Cleveland and operate under the auspices of the Community
Diversion Program (CDP) of the Cuyahoga County Court of Common Pleas,
Juvenile Division (juvenile court). The study period is from January 1, 2002 to
December 31, 2004 and consists of all juveniles referred to either of the two diversion programs during the study period. There also was a follow up period of one year from the date of program completion. This follow up period ran from
January 1, 2003 to December 31, 2005.
45 The two diversion programs were selected as study programs for several reasons. The primary reason is, after initial conversations with the director of the county’s CDP to explore the study’s feasibility, it was noted that one of the programs operated in a more restorative manner, using more reintegrative shaming program components, than any of the other diversion programs in the county. The second program is viewed by the director as being more typical of how the Community Diversion Programs throughout the county operate. The
CDP director oversees all of the forty-nine programs. In addition, the CDP director stated how both programs had similar referral rates and how both
coordinators submitted timely monthly reports of information required by the
county. She also expressed her sentiment about how cooperative she felt the
program coordinators would be for the study, so the programs were also chosen as a matter of convenience. The programs will hereinafter be referred to as the
Reintegrative CDP and the Typical CDP.
The subjects of the study are juveniles referred to the two programs by the
juvenile court with input from the Cuyahoga County Prosecutor’s intake
department. Although there are no hard and fast rules with regard to which CDP
a juvenile is referred, generally referrals are based on whether a youth lives in, or
commits an offense in, one of the diversion cities or in a nearby neighborhood
with no Community Diversion Program. So for example, if a youth who lives in
city A commits a referable offense in city B, he can be referred to either city’s
CDP – assuming both cities have one.
46 To be eligible for referral to the programs, youth had to have been under the age of 18 at the time of committing an offense, and the infraction had to be a status offense or a non-serious delinquent offense. Participation in a CDP is voluntary and at least one parent or guardian is required to attend the diversion hearing/ conference with the juvenile. Because participation is not mandatory, an additional requirement of eligibility is that juveniles must admit their involvement in the offense. This admission is not a legal admission akin to a guilty plea. It is more aligned with an acknowledgement of having committed deviant behavior and a willingness to discuss the circumstances surrounding the behavior. A youth who denies wrongdoing has his case sent back to the juvenile court for formal charging and processing. These juveniles are not included in this sample.
All CDP hearings or conferences are held in the respective communities and are closed to the public. Victims of crime may appear at a hearing and make a statement if desired. However, this rarely happens (D. Stolarski, personal communication, February 17, 2005; N. Stroman, personal communication,
February 17, 2005). After the diversion hearing, a case is either dismissed or sanctions are imposed. If sanctions are imposed, a juvenile must comply with the sanctions or risk having his case returned to juvenile court for formal processing.
Generally, juveniles referred to a diversion program are first time offenders. However, there are a few instances of repeat offenders in the sample from both programs. Because the total number of repeat offenders is characterized as being “just a handful” or “just a few” over the three year period
47 (H. Corcoran, personal communication, January 12, 2005; D. Stolarski, personal
communication, February 17, 2005; N. Stroman, personal communication,
February 17, 2005) and because repeaters were present in both programs, their
inclusion was not likely to have biased any results.
The overall quality of the data was good. Initially, the sample consisted of
608 cases from both programs, but the final number of cases studied was 533
(see Table 1). After reviewing the cases for missing data, a decision was made
Table 1.
Number of Participants by Diversion Program
Program % Total N
Reintegrative CDP 39 208 Typical CDP 61 325
Total 100 533
to exclude cases that contained missing data in a category used for statistical
analyses. Those categories included date of birth, ethnicity, gender (if not discernable), offense committed, and sanctions imposed. Five percent of the cases (n = 32) was excluded. Additionally, there were some youth who committed offenses at age 17 but were 18 by the time of their hearing. This group, 7% (n = 43) was also excluded from the study due to the fact that they would be 19 during the second data collection period and any information on reoffending would not be collected by the juvenile court. This left a sample size
48 of 533 youth, 61% in the Typical CDP (N = 325), 39% in the Reintegrative CDP
(N = 208), and 88% of the original cases identified for the project.
Data Collection
The final sample for this study has data collected during two time frames.
The first time frame is from the time of a juvenile’s diversion hearing until one
year from the date of hearing. During this time period, demographic and legal
information was collected. Also recorded is whether a juvenile completes the
diversion program. Juveniles are deemed to have completed the programs if
they adhere to the sanctions imposed during their hearings and do not commit a
status or delinquent offense during the one year period.4
The demographic, legal, and programmatic/sanction data are collected in
the normal course of the diversion programs’ administration. As part of the
reporting requirements, each diversion program coordinator submits to the
director of the CDP information on the juveniles who have been referred to the
respective programs. These reports include as demographic information, the juveniles’ date of birth, gender, race, school attended, grade, and the juveniles’ parent(s) or guardian(s).
The legal data collected give information on how the juvenile was referred to the diversion program, the date of referral, the name and level of offense(s) committed, the offense statute(s), and the case disposition. The programmatic/sanction data give information on the program sanctions imposed,
49 the hearing or conference date, and the volunteer magistrate who conducted the
hearing.
Of this reported information, the data used for statistical analyses in this
study are:
Demographic – age at the commission of offense (derived from
date of birth), age at the commissions of re-offense (for re-offenders),
gender, and race.
Legal – offense level and case disposition.
Programmatic – sanctions imposed.
In some instances juveniles were charged with more than one offense. For purposes of this study, only the most serious offense was included as the
variable of study. This was done in order to accurately measure offense
escalation for those youth who re-offended.
The second period of data collection is for one year after program
completion for those who did not reoffend prior to completing the program, or one
year after dismissal for those who had their cases dismissed at the time of
hearing. In this second examination period, official juvenile court filings were
checked to see if any of the subjects re-offended. Information on re-offending
was recorded by date and offense level. These data are collected by the juvenile
court only.
In addition to the secondary data analyses upon which this study relies,
original research was conducted. Questionnaires were sent to participating
families and interviews of diversion staff personnel were conducted in an effort to
50 further explore the nature of diversion. This data collection provided additional insight into program operations and gleaned some impressions of the programs held by participants and staff.
Sending questionnaires to participating families was chosen over other research methods such as focus groups or public hearings due to the confidential nature surrounding juvenile delinquency matters. The questionnaires were sent by each of the diversion program coordinators to 40 families who participated in the respective CDPs during the relevant study period. The families were randomly chosen by the program coordinators and sent two questionnaires: one for the juvenile and one for the parent or guardian, accompanied by a letter explaining the research and assuring them of confidentiality and privacy (see
Appendices A, B and C). Because the packets of information had to be sent out by the programs’ coordinators for purposes of confidentiality, the number of packets chosen (40) was done so as not to overburden the diversion staff. In addition, consent forms were sent to the parents along with an assent form for the youth (see Appendices D and E). Self addressed envelopes with postage pre-paid were provided for the return of each questionnaire separate from the signed consent form.
The instrument asked ten questions requiring responses using a Likert- type scale and one question that required a narrative response. The questions asked of the parents and juveniles are based on the Juvenile Offender Parent
Questionnaire (JOPQ) (Rose, Glaser, Calhoun & Bates, 2004) and examines shame, fear, monitoring and future outlook.
51 Three staff members were interviewed: the director of the county CDP
and the coordinators for each of the study programs. Interviews were both
structured and unstructured. Initial interviews were open-ended and
unstructured, conducted primarily to have questions answered that arose during
the course of the study. The unstructured interviews were with both diversion
program coordinators and the director of the county CDP. Structured interviews
were subsequently conducted with the program coordinators only and written
consent was obtained (see Appendix F). Only the researcher conducted the
interviews. The interview questions are listed in Appendix G. Two unstructured
interviews were conducted with the director, one unstructured interview was
conducted with each program coordinator, and one structured interview was
conducted with each coordinator.
Protection of Human Subjects
Primarily because the data used for this study are archival and were
collected in the CDP’s regular course of operations, the secondary data research
study posed no known risks to the juveniles who represent the sample
population. All data collected from questionnaires and interviews were obtained
in a way that posed no known risks to participants and were obtained with
informed consent (see Appendices B, C, and F). The individuals who were sent
questionnaires were informed by an accompanying letter of the specifics
regarding the research project and made aware that their participation was completely voluntary and could be withdrawn at any time. In addition to this
52 study being approved by the Institutional Review Board, letters of consent and
participation were obtained from authorized staff of the CDP. All research
materials are either in a locked file of the researcher or on file with the university.
When data were electronically sent to the researcher from the juvenile court, all personally identifiable information had been removed from the data with the exception of first name and date of birth. The data were then checked for discrepancies and missing information. After discrepancies were reconciled and all reasonable efforts were made to secure missing information, the first names were removed from the data set.
Measurement – Variables and Operationalization
There are three categories of variables used in this study primarily as independent and/or control variables. They are: 1) demographic, 2) legal, and
3) program component/sanction variables.
Demographic Variables
The demographic variables are age, gender, and race. The age variable used is derived from each juvenile’s date of birth and refers to the juvenile’s age at the time he/she offended. Also an age band variable was created dividing the youth into three categories: those age 16 and above, 14 and 15 year olds, and youth 13 years of age and under. For those youth who reoffended, age at the time of reoffending was used. For an initial descriptive presentation, the specific
53 race designation of the juveniles will be reported and thereafter, race will be designated as black or non-black.
Legal Variables
The legal variables used for the study consist of the type of offense committed by the participants – status offense or delinquent offense; the level of offense committed – ranking offenses by severity from the least severe to the most severe offense committed by the participants; and program disposition – how a juvenile’s referral to the diversion program ended (dismissed, completed program, or did not complete program). Offense type was created as a categorical variable from each juvenile’s designated offense level. Criminal offense levels range in severity from status offense (S) to first degree felony (F1).
Table 2 lists the offense levels from the least serious to the most serious, according to their legal designations by statute, along with their measurement codes.
Table 2.
Offense Levels and Measurement Codes Statutory Offense Level Code Designation S Status Offense 1 MM Minor Misdemeanor 2 M4 Fourth Degree Misdemeanor 3 M3 Third Degree Misdemeanor 4 M2 Second Degree Misdemeanor 5 M1 First Degree Misdemeanor 6 F5 Fifth Degree Felony 7 F4 Fourth Degree Felony 8 F3 Third Degree Felony 9 F2 Second Degree Felony 10 F1 First Degree Felony 11
54 Program Component/Sanction Variables
The last of the three categories is program component/sanction variables.
These variables are constructed from a content analysis of each program
component or sanction applied to the juveniles. Each subject’s applied program
components and/or sanctions were abstracted from the electronic reports sent to
the juvenile court by the respective programs. The data were evaluated and four
variables were constructed in order to measure the components. Program
components/ sanctions were then tallied and grouped into one of the four
construct variables. Those variables are: Punitive, Educational, Therapeutic, and
Reintegrative. Table 3 identifies the program components and sanctions for each construct.
The construction of these variables was crucial for being able to measure the program components and sanctions imposed on each participant. It was also necessary in order to measure the salient component of this study: reintegrative shaming. The assignment of components and sanctions to the punitive, educational, and therapeutic constructs was based on each component’s description and how it they were assigned to a youth (N. Stroman, personal communication, February 17, 2005; D. Stolarski, personal communication,
February 17, 2005). However, the reintegrative shaming construct was guided by literature on restorative justice and reintegrative shaming (Braithwaite, 1989;
Bazemore & Schiff, 2005). The construction of these variables also allowed the examination of how other program components impact program outcomes.
55 Table 3.
Program Construct Variables and Corresponding Program Components and Sanctions
Construct Program Component/Sanction
Punitive - Community Service/Labor - House Arrest - Telephone – television – electronic games restrictions - Home monitoring - No contact with certain individuals - Early Curfew - Chores - Other restrictions (ex. loss of driving privileges, no social outings, etc.)
Educational - Tutoring - Raise grades - Attend school regularly - Book Club - Gang Free Program - Y.E.S. (shoplifting) program - Academic testing - Obtain GED
Therapeutic - Anger management - Drug/Alcohol Assessment - Counseling - Straight Talk Program - PACT (Parents and Children Together) Program - Referral to Bellfaire or Euclid House - Journaling
Reintegrative - Letters of apology - Essays/Written Assignments - Referral to Big Brothers/Sisters Program - Rites of Passage Program - Neighborhood Center Activities - Mentoring - Job/Career Shadowing
56 Table 4 reveals what percentage of youth in each subgroup received at least one
program component or sanction categorized under the program construct
variables. Juveniles were assigned program components or sanctions in one,
two, three, or all four of the categories.
Table 4.
Percentage of Youth in Each Program Assigned Construct Components/Sanctions
Diversion Program Construct Variables
Punitive Educational Therapeutic Reintegrative Reintegrative CDP 60 52 31 62
Typical CDP 76 20 24 10
Dependent Variables
The dependent variables that are the basis for the study’s hypotheses are
program completion – whether a youth completed the diversion program if the
case was not dismissed after hearing; recidivism – whether a juvenile reoffended
while participating in the diversion program or during the follow up period; and
offense escalation – whether those youth who reoffended committed an offense
more serious than the one for which they were initially referred to the CDP.
Variable Definitions and Coding
The following is a list of the variables used in this study, described and presented with their measurement levels and codes.
57 Independent Variables
Demographic (Sample N = 533)
- AGE AT OFFENSE = a juvenile’s age at the time the offense is
committed (interval).
- AGE RANGE AT OFFENSE = a banding of the variable age
(ordinal partitioned) (1 = juveniles age 13 and under; 2 =
juveniles age 14 and 15; 3 = juveniles 16 and above).
- AGE AT RE-OFFENSE = a juvenile’s age at the time he/she
reoffends (interval).
- GENDER = female or male (categorical) and recoded
dichotomously (nominal) (0, 1).
- ETHNICITY = African American, Caucasian, Asian, Hispanic
and other (originally categorical) recoded dichotomously
(nominal) non-black/black (0, 1).
Legal (Sample N = 533)
- OFFENSE TYPE = delinquent offense or status offense
(categorical) coded dichotomously (nominal) (0, 1).
- OFFENSE LEVEL = degree of offense committed ranging from
status offense to first degree felony, coded 1 through 11
(ordinal).
58 - PROGRAM DISPOSTION = whether a juvenile completed a
diversion program or had his case dismissed – (categorical)
(Dismissed, Completed Program, Did Not Complete).
Program Components/Sanctions (Sample N = 533)
- Punitive = sanctions imposed that are classified as punitive
(ratio).
- Educational = sanctions imposed that are classified as
educational (ratio).
- Therapeutic = sanctions imposed that are classified as
therapeutic (ratio).
- Reintegrative = sanctions imposed that are classified as
reintegrative (ratio).
Dependent Variables
Program Outcomes
- COMPLETED PROGRAM (Sample N = 455) = whether a
juvenile whose case was not dismissed, completes a diversion
program (categorical) coded No or Yes and recoded (0, 1).
Youth who had their cases dismissed were excluded from the
analysis.
- RECIDIVISM (Sample N = 533) = whether a juvenile commits
another status or delinquent offense within a year of the
59 diversion hearing or during the follow up period of one year after
program completion (categorical) coded No or Yes and recoded
(0, 1).
- ESCALATION (Sample N = 82) = an increase in the level of
delinquent offense committed by a juvenile who reoffended
(categorical) coded No or Yes and recoded (0, 1).
Other Variables (Sample N = 533)
- DIVERSION PROGRAM = One of the two Community Diversion
Programs, Reintegrative CDP or Typical CDP (categorical)
recoded (0, 1)(nominal).
- YEAR = One of the years relative to the study period (2002,
2003, or 2004) (categorical).
Definitions of Key Terminology
The following key terms are used throughout this study.
- Disposition – how a diversion case is disposed of, or no longer
under the jurisdiction of the CDP.
- Educational sanctions – requirements imposed, instructions given
or actions taken to encourage or facilitate learning and academic success.
- Escalation – being officially charged with an offense that is a
higher level offense than the one for which the juvenile was initially
diverted.
60 - Formally processed – to be officially charged with an offense and
adjudicated in juvenile court.
- Net widening – processing juveniles through a diversion program who likely would not have been processed at all: formally or informally.
- Program completion – adhering to and completing imposed diversion program sanctions without having reoffend for one year from the date of diversion hearing.
- Punitive sanctions – prohibitions from participating in entertaining or enjoyable activities and/or tasks imposed as a penalty or discipline.
- Recidivism or re-offending – being officially charged with having committed one or more status or delinquent offenses within one year after a diversion hearing or during the one year evaluation period for juveniles who complete the program.
- Reintegrative sanctions – activities designed to foster remorse and awareness of the impact of offending, encourage thinking about future goals, and boost self esteem.
- Short term outcomes - results stemming from the diversion programs which happen in one year or less.
- Therapeutic sanctions – referral to a program or activities designed to assist with diagnosis and treatment of some potential psychological and/or physiological condition.
61 The next chapter sets forth the findings and analysis of the data discussed above. If the hypotheses are supported, participants in the Reintegrative CDP should display better program outcomes than the participants in the Typical CDP.
Additionally the analysis should be able to shed light on the impact the programmatic components have on the participants.
62 Chapter 4
Findings
In this chapter, sample characteristics and comparisons will be given along with the results of statistical analyses addressing the study’s research
questions and hypotheses.
Demographic Characteristics
The total sample in this study consisted of 533 juveniles; 39% (n = 208)
from the Reintegrative CDP and 61% (n = 325) from the Typical CDP. Sixty five
percent (n = 345) were males and 35% (n = 188) females. Almost 84% (n = 445)
of the juveniles were African American (black) and 15% (n = 70) were white.
Less than 2% (n = 9) of the youth were designated Asian, Hispanic or Other. The youth ranged in age from 7.36 to 17.94 at the time of offense with an average age of 14.66 years (SD = 2.05). Approximately 30% (n = 159) of the juveniles were ages 16 and above, 39% (n = 205) were ages 14 and 15, and 32% (n =
169) were age 13 and under.
The majority of referred juveniles 97% (n = 517) attended public or alternative school with a slight majority 55% (n = 294), being high school students. There were a handful of offenders, less than 1% (n = 3), who had either dropped out of school or had graduated from high school.
“Mother” was the primary designation listed for custodial parent, (77%, n
= 408). Only 7% (n = 38) of youth had both parents listed as custodian. The category of other includes father, grandparents, other relatives, caseworker,
63 guardian, and the actual listings of “other”. Table 5 displays the demographic characteristics of the sample.
Table 5.
Sample Demographic Characteristics (N = 533) % or Variables SD N Mean
Mean Age 14.66 (2.05) 533
% Age Range 16 and above 30 159 14 and 15 39 205 13 and below 32 169
Gender Percent Male 65 345
Race Black 84 445 White 15 70 Other < 2 9
School Public or Alternative 97 517 Private or Home Schooled 2 10 Graduated or Dropped Out < 1 3 Missing Data < 1 3
Custodial Parent Mother 77 408 Both Parents 7 38 Other 9 51 Missing Data 7 36
64 Legal Characteristics
A majority of diversion referrals, 76% of the reported data (n = 378), were initiated by local law enforcement agencies (see Table 6).
Table 6.
Legal Characteristics of Sample (N = 533)
Variable % N
Referral % Law Enforcement 76 378
Offense Type % Delinquent 88 471
Offense Level Status Offense 12 62 Minor Misdemeanor 4 19 4th Degree Misdemeanor 31 165 3rd Degree Misdemeanor 2 10 2nd Degree Misdemeanor 8 40 1st Degree Misdemeanor 44 234 5th Degree Felony 1 3
Offense Description Assault 15 78* Criminal Damaging 6 32 Criminal Trespass 6 33 Curfew Violationª 5 27 Disorderly Conduct 26 139* Theft 9 49 Unrulyª 4 23* Otherb 28 152
Program Disposition Dismissed 15 78 Completed 75 402 Did Not Complete 10 53 Note. * Primary offenses committed
65 ª Indicates status offense.
b Includes aggravated menacing, arson, breaking and entering, carrying a concealed weapon, criminal mischief, domestic violence, endangering a minor, facsimile firearms, failure to report a crime, falsification, gambling, inciting to violence, inducing panic, making false alarms, menacing, misconduct at an emergency, obstruction of official business, possession of criminal tools, possession of drug paraphernalia, possession of marijuana, public indecency, receiving stolen property, resisting arrest, rioting, telephone harassment, truancy, unauthorized use of a motor vehicle, and vandalism.
The remaining reported referrals were listed as being made by parents or
guardians and by other sources such as school officials and citizens.
Approximately 88% (n = 471) of the cases referred to the diversion
programs were referred for delinquent offenses and 12% (n = 62) were referred for status offenses. Disorderly conduct 26% (n = 139) and assault 15% (n = 78) were the primary delinquencies reported for the sample, and unruliness was the most reported status offense (4%, n = 23). Fifteen percent (n = 78) of the cases were dismissed after the diversion hearing.
Sample Demographic and Legal Comparisons
A factor to consider when deciding which statistical tests to use is whether the study sample is normally distributed. Because the subjects for this study were not randomly drawn, there could be no assumption that individual observations were normally distributed. However, because of the central limit theorem, the distribution of the sample means will be normally distributed when the sample size is 30 or more, even if the sample is non-normally distributed
66 (Hair, Black, Babin, Anderson, & Tatham, 2006). This means that both parametric and non-parametric tests could be used in this study.
Analyses of demographic characteristics revealed some significant associations. Females were an average age of 15.09 (SD = 1.58) while males were younger, 14.43 (SD = 2.23) (see Table 7). An independent samples t – test revealed that the two groups are significantly different. Females were significantly older than males, t (495.138) = -3.960, p = .000. The Levene’s Test for Equality of Variances was significant (F = 20.897, p = .000) meaning that the two variances are not assumed equal.
Table 7.
Comparisons of Gender by Other Demographics
Variable Males (N=345) Females (N =188 )
Mean SD N Mean SD N
Age* 14.43 2.23 345 15.09 1.58 188
% % Race Black 65 287 35 158 Non-Black 66 58 34 30 Note. *p < .001
Comparing the age of black youth (M = 14.52, SD = 2.07) to non-black youth (M
= 15.37, SD = 1.75) revealed that black youth were significantly younger, t (531) = 3.599, p = .000.
67 Cross tabulation procedures were conducted to examine the relationship
between categorical variables. As part of the procedures, a chi square analysis was used to determine whether any relationships between the two categorical variables, gender and race, were statistically significant. No significant relationship was found.
Cross-tabulation analysis was performed to examine significant
associations between categorical demographic and legal variables. There were
no statistically significant relationships between the type of offense committed and gender or race (see Table 8). However a t-test showed that there was a
Table 8.
Comparisons of Offense Type by Demographics
Status Offenses (N=62) Delinquencies (N =471 ) TOTAL Variable % SD N % SD N % SD N
Gender Male 10 36 90 309 65 345 Female 14 26 86 162 35 188
Race Black 11 50 89 395 84 445 Non-Black 14 12 86 76 16 88
Mean Mean Mean Age* 15.77 1.20 62 14.52 2.09 471 14.66 2.05533 Note. *p < .001
68 significant difference in the type of offense committed by age t(531) = -4.605, p =
.000. The average age for status offenders was 15.77 (SD = 1.20) while the delinquent offenders’ average age, 14.52 (SD = 2.09), was significantly younger.
Analyses of offense severity levels by demographic characteristics revealed significant differences with regard to age range and race (see Table 9).
The youngest offenders, those age 13 and under, had the highest mean offense level 4.66 (SD = 1.57) within all subgroups. Their mean level of offending was significantly higher than those youth age 14 and above (F = 8.33, df 2, p = .000).
Also, offenders whose race was other than black had a mean offense level of
4.59 (SD = 1.87) compared to black offenders, 4.17 (SD = 1.80). This difference
was also significant (F = 3.916, df 1, p = .048).
Table 9.
Offense Level Means by Demographic Variables
Variable Mean SD N F DF P
Age Range* 8.33 2 .000 16 and above 3.85 1.93 159 15 and 14 4.20 1.86 205 13 and below 4.66 1.57 169
Gender .7251 .395 Male 4.29 1.76 345 Female 4.15 1.92 188
Race** 3.911 .048 Black 4.17 1.80 445 Non-Black 4.59 1.87 88 Note. *p < .001, **p < .05
69 In order to explore whether collapsing the variable “age” into the band of
“age range”, thereby transforming a continuous variable into a categorical variable, resulted in a loss of precision that affected the measure of association
between offense level and age, the Pearson correlation coefficient (r) was used
to more precisely determine the degree and direction of the relationship between
the two continuous variables. Using the Pearson correlation coefficient assumes
1) that the any correlated variables are normally distributed, and 2) that the relationship between the correlated variables approximates a linear relationship.
The correlation matrix supported a conclusion that there is a weak negative relationship between offense severity level and age (r = -.201, p = .000,
2-tailed). Consistent with the analysis of offense level by age range, the Pearson r indicates that an average decrease in age is related to an average increase in offense severity.
Program Disposition by Demographic and Legal Characteristics
Table 10 displays the characteristics of the sample by program disposition. There was no significance difference in the average age of offenders regardless of the resolution or disposition of their cases – whether a case was dismissed, the juvenile completed the program, or the juvenile did not complete the program. However, there was a significant difference in the mean offense level by program disposition (F = 3.780, df 2, p = .023). Youth who had their cases dismissed had a mean offense level that was significantly higher than those youth who continued through the diversion programs.
70 Table 10.
Summary of Program Disposition by Demographic and Legal Characteristics
Variable (N=533)
Did Not Dismissed Completed Complete
Mean SD Mean SD Mean SD Age 14.44 2.4814.69 2.02 14.76 1.50
Offense Level* 4.56 1.77 4.25 1.83 3.68 1.74
% N % N % N Gender Male 9 46 49 260 7 39 Female 6 32 27 142 3 14
Race Black 12 62 64 341 8 42 Non Black 3 16 11 61 2 11 Note. *p < .05
Chi square analyses did not reveal any significant associations between program
disposition and any of the categorical demographic or legal variables. Table 11 displays a contingency table of the variables along with their corresponding p values.
71 Table 11.
Comparison of Program Disposition Using Chi Square on Discrete Data
Variable Dismissed Completed Did Not P Value N Complete
Gender .228 Males 13.3 75.4 11.3 345 Females 17.0 75.5 7.5 188
Race .346 Black 14.0 76.6 9.4 445 Non-Black 18.2 69.3 12.5 88
Offense Type .400 Status Offense 9.7 77.4 12.9 62 Delinquency 15.3 75.2 9.6 471
Table 12 compares the sample characteristics of those youth who reoffended to those who did not reoffend. Two variables were found to have a significant relationship to recidivism: gender (χ2 = 3.963, df1, p = .047) and age range (χ2 = 11.269, df 2, p = .004). Males were more likely to reoffend than females. Although a t-test showed that there was no statistically significant difference in the average age of those who reoffended and those who did not, chi square analysis of age range revealed that younger juveniles, those age 13 and under, were more likely to reoffend than juveniles age 14 and over.
72 Table 12.
Summary of Recidivism by Demographic and Legal Characteristics
Variable (N=533)
Reoffended Did Not Reoffend Mean SD Mean SD P Value Age 14.67 1.45 14.66 2.14
Offense Level 4.11 1.74 4.26 1.84
% N % N Gender* .047 Male 17.7 61 82.3 284 Female 11.2 21 88.8 167
Race .637 Black 15.1 67 84.9 378 Non Black 17.0 15 83.0 73
Age Range** .004 16 and above 8.2 13 91.8 146 15 and 14 21.0 43 79.0 162 13 and below 15.4 26 84.6 143
Offense Type .565 Status 12.9 8 87.1 54 Delinquent 15.7 74 84.3 397 Note. *p < .05, **p < .01
Of the 82 youth who reoffended during the study period, more than half,
51.2% (n = 42), committed a higher level offense than the one for which they were diverted. The highest level offense committed was a second degree felony
(F2). Table 13 displays the levels of re-offenses along with their measurement codes.
73 Table 13.
Re-offense Levels and Measurement Codes Measurement Re-offense Level N Code
Status Offense 1 6 Minor Misdemeanor 2 13 Fourth Degree Misdemeanor 3 6 Third Degree Misdemeanor 4 3 Second Degree Misdemeanor 5 4 First Degree Misdemeanor 6 31 Fifth Degree Felony 7 4 Fourth Degree Felony 8 4 Third Degree Felony 9 4 Second Degree Felony 10 7
Total 82 (Did Not Reoffend) 429
Total 533
In Table 14, sample characteristics of offending youth by offense escalation are given. The difference in the mean offense level was significant between those youth who escalated their offending conduct and those who did not (F = 8.169, df 1, p = .005). Youth who did not escalate had a higher mean offense level. In addition to comparing age at the time of offense, for this examination, the average age when juveniles re-offended was also analyzed.
Neither mean was significant. There were no other variables significantly associated with offense escalation.
74 Table 14.
Summary of Offense Escalation by Demographic and Legal Characteristics
Variable (N= 82)
Escalated Did Not Escalate Mean SD Mean SD P Value Age at offense 14.59 1.45 14.74 1.46 Age at re-offense 15.61 1.44 15.79 1.28
Offense Level* 3.60 1.57 4.65 1.72 .005
% N % N Gender .902 Male 50.8 31 49.2 30 Female 52.4 11 47.6 10
Race .856 Black 50.7 34 49.3 33 Non Black 53.3 8 46.7 7
Offense Type .502 Status 62.5 5 37.5 3 Delinquent 50.0 37 50 37
Age Range .571 16 and above 41.2 7 58.8 10 15 and 14 51.3 20 48.7 19 13 and below 57.7 15 42.3 11 Note. *p < .01
The Reintegrative Program
The sample subgroup from the Reintegrative Community Diversion
Program (CDP) contained 208 youth. The majority of youth are black 86% (n =
179) which is consistent with the demographic composition of the area, and the majority are male 60% (n = 125). The average age of the juvenile offenders was
14.24 (SD = 2.20) with the youngest offender being 7.48 and the oldest offender
75 being 17.74. The proportion of juveniles charged with having committed a
delinquent offense is 85% (n = 176) compared to the 15% (n = 32) who were
status offenders (see Table 15).
Table 15.
Characteristics of Reintegrative CDP Subgroup
Variable Reintegrative CDP (N=208)
Mean SD N
Age 14.24 (2.20) 208
% Gender (% Male) 60 125 Race (% Black) 86 179 Offense Type (% Delinquent) 85 176
In 2002 there were 64 juveniles who were referred to the Reintegrative
CDP. In 2003, there was a 25% increase (n = 16) in participants from the
previous year and in 2004, the number of participants dropped back to 64. Table
16 lists the demographic characteristics of the Reintegrative subgroup by year.
Males and black youth made up the majority of the subgroup for all three years.
The average age of female offenders in 2002 was 15.13 (SD =
1.57) but the average age for male offenders was 13.77 (SD 2.82). The difference in age was significant t (62) = -2.232, p = .029). The overall average
age was 14.32 (SD = 2.47). The youngest offender that year was 7.48 and the
oldest was 17.42. The oldest male and female were nearly identical in age but
76 the youngest female referred in 2002 was 11.5 while the youngest male was
7.48.
Table 16.
Characteristics of Reintegrative CDP by Year
Variable 2002 2003 2004
% N % N % N Demographics Gender Male 59 38 65 52 55 35 Female 41 26 35 28 45 29
Race Black 89 57 80 64 91 58 White 6 4 15 12 8 5 Other 5 3 5 4 1 1
Mean* SD Mean SD Mean SD Age Male 13.77 2.82 13.86 2.44 14.19 1.94 Female 15.13 1.57 14.29 2.04 14.76 1.44 All 14.32 2.47 14.01 2.30 14.45 1.74 Note. *Significant age difference between males and females in 2002, p < .05
Half of the youth were in high school when they offended. However, some offenders who were still in elementary school began offending while in the second grade.
In 2003, the male to female ratio was 65% (n = 52) to 35% (n = 28). Black youth were 80% (n = 64) of the subgroup, followed by 15% (n = 12) white and the remaining 5% (n = 4) were listed as Asian or other. The average age for female offenders was slightly younger than the previous year at 14.29 (SD = 2.04) while
77 the average age for male offenders was about the same as the year before,
13.86 (SD = 2.44). The overall average age was 14.01 (SD = 2.30) and the
youngest offender was 7.81 while the oldest was 17.74. Females ranged in age
from 9.61 to 17.02 and males from 7.81 to 17.74. As in 2002, half of the youth were in high school when they offended. Once again, juvenile offending started
as early as the second grade.
In 2004, 90% (n = 58) of the juveniles were black and the remaining 10%
were 5 white and 1 Asian. The average age of female offenders was14.76 (SD
1.44) and the average age for male offenders was 14.19 (SD = 1.94). The
overall average age was 14.5 (SD = 1.74). For 2004, a slight majority of youth
55% (n = 35) were in high school when they offended and offending began later for this group – fourth grade. The age range for males was 10.27 to 17.07 and for females, it was 11.93 to 17.62.
The Typical Program
The Typical CDP subgroup consisted of 325 juveniles. Similar to the
Reintegrative CDP, the majority of youth were black (82%, n = 266) and males
68% (n = 220). At the time the various offenses were committed, the average age of a Typical CDP juvenile offender was 14.93 (SD = 1.89) with the youngest
offender being 7.36 and the oldest 17.79. The proportion of juveniles committing
a delinquent offense compared to those who were status offenders was 91% (n =
295) and 9% (n = 30) respectively (see Table 17).
78 Table 17.
Characteristics of Typical CDP Subgroup Typical CDP (N=325) Variable Mean SD N Age 14.93 (1.89) 325
% Gender (% Male) 68 220 Race (% Black) 82 266 Offense Type (% Delinquent) 91 295
Ninety seven juveniles were referred to the Typical CDP in 2002, 108 in
2003 for an 11% increase, and 2004 saw yet another 11% increase with 120
referrals (see Table 18). For 2002, 74% (n = 72) were male and 26% (n = 25)
were female. Seventy five percent (n = 77) of the youth were black and 25% (n =
24) white. The average age of female offenders was 15.98 (SD = 1.22) and the
average age for male offenders was 14.94 (SD = 1.95). The difference in age
was significant t(95) = -2.497, p = .014). The overall average age was 15.21 (SD
= 1.84) and ages ranged from 7.3 to 17.64. The age range for females was
12.84 to 17.48 while the range for males was quite a bit larger, primarily due to the lower range being 7.36 while the upper range extended a bit farther at 17.64.
Unlike the 2002 Reintegrative CDP subgroup, more than half of the offenders,
63% (n = 61) were in high school when they offended. Similar to the
Reintegrative CDP, elementary school offending started in the second grade.
Of the 108 juveniles referred to the Typical CDP in 2003, 61% (n = 66)
were male and 39% (n = 42) female. Black youth were 84% (n = 91) of the
subgroup while the remaining 16% (n = 17) were white with the exception
79 Table 18.
Characteristics of Typical CDP by Year
Variable 2002 2003 2004
% N % N % N Demographics Gender Male 74 72 61 66 68 82 Female 26 25 39 42 32 38
Race Black 75 73 84 91 85 102 White 25 24 15 16 15 18 Other -- -- 1 1 -- --
Mean* SD Mean SD Mean* SD Age Male 14.94 1.95 14.56 2.19 14.64 2.00 Female 15.98 1.22 15.03 1.30 15.38 1.52 All 15.21 1.84 14.75 1.90 14.88 1.90 Note. -- Indicates none reported
*Significant age difference between males and females in 2002
and 2004, p < .05
of one Hispanic youth. The average age of female offenders was slightly older
than the previous year 15.03 (SD = 1.3) while the average age for male offenders
was slightly higher than the year before 14.56 (SD = 2.19). The overall average
age was 14.75 (SD = 1.90) and the youngest and oldest juveniles were 9.58 and
17.62 respectively. Female offenders ranged in age from 12.21 to 17.54. Males
narrowed their range from 9.58 to 17.62 years old. Similar to the Reintegrative
80 CDP subgroup in 2003, half of the youth were in high school when they offended.
However, juvenile offending did not start until the fourth grade.
Of the 120 juveniles referred to the Typical CDP in 2004, 68% (n = 82) were male and 32% (n = 38) female. The total was a 24% increase from 2002.
Of the total, 85% (n = 102) were black and 15% (n = 18) white. The average age of female offenders was 15.38 (SD = 1.52) and the average age for males was just over 14.64 (SD = 2.0). The difference in age was significant t(118) = -1.987, p = .049). The overall average was 14.88 (SD = 1.9) while the overall range was
9.9 to 17.79. In 2004, younger males narrowed the age gap between themselves and their younger female counterparts even more than the prior year. The age range for females was again 12 to 17.5, consistent with the two previous years
(12.12 to 17.46) but the age range for males started at age 9.96 and extended to
17.79. The majority of juveniles 60% (n = 72) were in high school when they offended and offending began at grade 5.
Comparison of Diversion Programs
Table 19 provides demographic data of the program participants by program site.
The difference in the age mean was significant between the two diversion programs t(531) = 3.840, p = .000). Youth in the Reintegrative CDP (M = 14.24,
SD = 2.20) were significantly younger than youth in the Typical CDP (M = 14.93,
SD = 1.89). There were no significant differences between programs with regard to gender and race. However, there were significant differences between gender and race within each diversion program: Reintegrative CDP – gender (χ2 = 8.48,
81 df 1, p = .004); race – (χ2 = 108.17, df 1, p = .000), Typical CDP – gender (χ2=
40.69, df 1, p = .000); race (χ2 = 131.84, df 1, p = .000). There were significantly
more males than females, and black youth than non-black youth in both CDPs.
Table 19.
Demographic Characteristics of Sample by Program
Variable Reintegrative CDP Typical CDP TOTAL (N=208) (N=325) (N=533)
Mean SD N Mean SD N Mean SD N
Age* 14.24 2.20 208 14.931.89 325 14.66 2.05 533
% % % Gender* Male 60 125 68 220 65 345 Female 40 83 32 105 35 188
Race* Black 86 179 82 266 83 445 Non-Black 14 29 18 59 17 88 Note. *p < .05
Table 20 compares the two programs by legal variables. There were
significant differences between the two programs regarding all three legal
variables. Cross-tabulation analyses revealed that the type of offense committed
and diversion program were significantly related (χ2 = 4.673, df 1, p = .031). Of the 62 youth who committed status offenses, 15% (n = 32) were in the
Reintegrative CDP and 9% (n = 30) were in the Typical CDP. Looking at the amount of youth who committed delinquent offenses, 85% (n = 176) of delinquent
82 offenders were from the Reintegrative CDP and 91% (n = 295) were from the
Typical CDP. There is a fairly equal number of status offenders in both programs, but in general, the Typical CDP is more likely to have delinquent offenders than the Reintegrative CDP.
Table 20.
Summary of Legal Characteristics of Sample by Program
Reintegrative Typical Both Variable CDP CDP Programs
% Total N % Total N % Total N Offense Type* (Delinquent) 85 176 91 295 88 471
Offense Level** Status 15 32 9 30 12 62 MM – M3 19 39 48 155 36 194 M2 – F5 66 137 43 140 52 277 Mean SD MeanSD Mean SD 4.59 2.01 4.02 1.65 4.24 1.82
Disposition Dismissed 62 48 39 30 15 78 Completed Program 35 140 65 262 75 402 Did Not Complete 38 20 62 33 10 53
Note. *p <.05, **p < .001
83 An analysis of variance revealed that CDP offense levels differed
significantly (F = 12.56, df 1, p = .000). The mean offense level for the
Reintegrative CDP was 4.59 (SD = 2.01) and the Typical CDP mean offense
level was 4.02 (SD = 1.65). Youth in the Reintegrative CDP, on average,
committed more severe offenses.
The primary delinquent offenses committed by juveniles in the
Reintegrative CDP were assault 26% (n = 53) and a theft offense (theft, petty theft, or shoplifting) 16% (n = 33), while the status offenders were referred mostly for unruliness 11% (n = 23) and curfew violations 10% (n = 21). For the Typical
CDP, disorderly conduct 38% (n = 123) led all delinquent offenses committed by
this subgroup with criminal damaging 9% (n = 28) and assault 8% (n = 25)
following a distant second. Being incorrigible at home 4% (n = 12) was the
primary status offense committed by the subgroup (see Table 21).
84 Table 21.
Offenses Committed Listed by Program
Offenses Reintegrative CDP Typical CDP TOTAL (N=208) (N=325)
% N % N % N
Aggravated Menacing 6 13 2 6 4 19 Assault 26 53* 8 25* 15 78 Criminal Damaging 3 6 9 28* 6 32 Criminal Trespass 7 14 6 19 6 33 Curfew Violationª 10 21* 2 6 5 27 Disorderly Conduct 8 16 38 123* 26 139 Incorrigible at Homea -- -- 4 12* 2 12 Prohibition (Under 21) -- -- 4 14 3 14 Theftb 16 33* 5 17 9 50 Unrulyª 11 23* <1 1 4 24 Other Crimes Against Persons 5 10 1 4 3 14 Against Property 2 5 1 3 1 8 General Offenses 5 11 19 62 14 73 Status Offenses 1 3 1 5 1 8 Note: * Primary offenses committed in CDP
-- Indicates none reported
ª Indicates status offense.
b Includes theft, petty theft, and shoplifting
85 Research Questions and Hypotheses
The first three research questions for this study, framed by application of restorative justice principles, examine whether certain programmatic components are associated with program outcomes. Research questions and hypotheses 1 through 3 are related to the influence of higher amounts of reintegrative shaming components on program outcomes. Table 4 displayed in chapter 3, shows that the Reintegrative CDP had a higher percentage of youth exposed to reintegrative program components and sanctions than the Typical CDP.
To establish, statistically, which one of the two diversion programs was more reintegrative in nature, a correlation was run between the diversion program variables and the reintegrative program component/sanction. Diversion program was made a dichotomous variable with the Typical CDP being assigned
0 and the Reintegrative CDP being assigned 1. The correlation between the
Reintegrative CDP and the reintegrative program components/sanctions was significant r (.561), p < .01(2-tailed).
Research question 4 examines whether the diversion programs impact status offenders differently from delinquent offenders. Specifically, this question examines whether there are any significant differences in programmatic outcomes of status offenders compared to those of delinquent offenders.
Analyses will also explore how diversion programmatic components impact the subgroups.
86 Research Question One
The first research question asks whether completion of a CDP, the initial goal of diversion programs, is positively associated with reintegrative shaming components. Only those cases designated “yes” or “no” for program completion were analyzed (n = 455). Juveniles who had cases dismissed after their diversion hearing were excluded from the analysis. It was hypothesized that increased amounts of reintegrative shaming components in a diversion program would increase the likelihood of completing the program. In other words, participation in the Reintegrative CDP was expected to yield a significantly higher rate of program completion than participation in the Typical CDP. Cross- tabulation and regression analyses were performed.
The hypothesis was unsupported in the cross-tabulation analyses. There was no significant relationship between program completion and diversion programs (see Table 22). Youth from the Reintegrative CDP and youth from the
Typical CDP had program completion rates that were almost identical: 88% (n =
140) and 89% (n = 262) respectively.
Table 22.
Chi Square Analysis of Program Completion by CDP (N = 455) Did Not Variable Completed P value Complete % N % N Program .677 Reintegrative 87.5 140 12.5 20 Typical 88.8 262 11.2 33
87 In the regression analysis, demographic and legal variables were included
in the model as control variables. Adding the variables to the model improved the R2, however the predictor variable, diversion program, still was not
significantly related to program completion even after model improvement. The
only variable that was a significant predictor of whether a juvenile completes the
diversion program was one of the legal variables, offense level (B=.326, p = .005)
(see Table 23). The results indicate an expected increase of .326 in the log-odds
of program completion, holding all other predictor variables constant, for every
one unit increase in offense level. The test of the overall model was not
statistically significant, (χ2 =11.53, p =.073).
Table 23.
Summary of Logistic Regression Analysis for Variables Predicting Program
Completion (n = 455)
Odds Ratio Predictor B SEB P Value (eB)
Diversion Program .412 .333 .217 1.510
Demographic Variables Gender (Male) .525 .338 .120 1.691 Race (Black) -.598 .388 .123 .550 Age .003 .081 .974 1.003
Legal Variables Offense Type (Status) -.914 .587 .120 .401 Offense Level* .326 .116 .005 1.386 Note. *p < .01
88 Research Question Two
The second research question asks whether reintegrative shaming is negatively associated with recidivism. To answer this research question, the original sample used for this study (n = 533) was analyzed because youth who had their cases dismissed at the hearing phase were nonetheless evaluated for reoffending. It was hypothesized that increased amounts of reintegrative shaming components in a diversion program would decrease the likelihood of juvenile reoffending. It was expected that participation in the Reintegrative CDP would contribute to lower instances of reoffending compared to participation in the Typical CDP. Cross-tabulation and logistic regression analyses were conducted.
The hypothesis was not supported by the data analyses. There was no significant relationship between diversion program and recidivism. Similar to the comparison of program completion by diversion program, the Reintegrative CDP and the Typical CDP had almost identical rates of recidivism: 15% (n = 31) and
16% (n = 51) respectively (see Table 24).
Table 24.
Summary of Recidivism by Diversion Program Did Not Variable Reoffended P Value Reoffend % N % N Program .806
85.1 177 14.9 31 Reintegrative Typical 84.3 274 15.7 51
89 Demographic and legal variables were entered into a regression model, including the legal variable, program disposition, in addition to the legal variables offense type and offense level code. Because the program disposition variable has three categories, (completed, did not complete, and dismissed) a multinomial regression analysis was performed. The reference category is “completed” and the recidivism variable was dummy coded 1 = reoffended. No predictor variables bore any significant association with recidivism (see Table 25).
Table 25.
Summary of Logistic Regression Analysis for Variables Predicting Recidivism
(n = 533)
Odds Ratio Predictor B SEB P Value (eB)
Demographic Variables Gender (Male) -.643 .453 .156 .526 Race (Black) -.274 .565 .628 .761 Age .035 .091 .703 1.035
Legal Variables Offense Type (Status) 17.439 3733.316 .996 4E+007
Offense Level .102 .138 .459 1.107 Program Disposition (Completed) (Dismissed) .327 .482 .497 1.387 (Did Not Complete) 39.641 5942.061 .995 2E+017 Note. Program Disposition included 3 categories (Completed, Did Not
Complete, and Dismissed). Completed is the reference category.
90 To further explore any possible significant interactions between the
diversion programs and recidivism, the amount of days until youth re-offended
was examined. The subsample for this analysis included only those youth who
re-offended (n = 82). This is the same subgroup that was used in the analyses of
research question three. Thirty eight percent of youth (n = 31) were in the
Reintegrative CDP and the other 62% (n = 51) were in the Typical CDP.
The majority of youth from the subgroup, 71% (n = 58) reoffended before the one year requirement for program completion – the first of two evaluation periods for recidivism. Seventy seven percent (n = 24) of youth in the
Reintegrative CDP re-offended before program completion and 67% (n = 34) of youth in the Typical CDP re-offended before completing the program.
Using a comparison of means test, the predictor variable diversion program was analyzed with the outcome variable, days until re-offense. For youth in the Reintegrative CDP, the average amount of days between diversion hearing and re-offending was 213.42 (SD = 192). For the Typical CDP, the average amount was 299.80 (SD = 173). The Reintegrative CDP average was significantly lower than the average for the Typical CDP (F = 4.40, df1, p = .039).
Juveniles in the Reintegrative CDP were likely to reoffend an average of approximately three months earlier than juveniles from the Typical CDP (see
Table 26), and youth from both programs, on average, reoffended less than a year after diversion hearing.
91 Table 26
Analysis of Variance for Days to Re-offense and Diversion Program
Sum of Mean Variable Squares df Square F Sig. Days Between Hearing Between and Re-offense by Groups 143876.656 1 143876.656 4.400 .039 Diversion Program Within Groups 2615825.588 80 32697.820 Total 2759702.244 81
. A linear regression analysis was conducted on demographic and legal
variables and the dependent variable, days to re-offense. The overall model was statistically significant (F = 4575.137, p = .000), but age was the only variable
found to be a significant predictor of days to reoffending (see Table 27). The age
of juveniles at the time of offense (t = -2.751, p = .008), at the time of hearing (t =
-45.219, p = .000), and at the time of re-offense (t = 181.08, p = .000) were all
found to be significant predictors.
The unstandardized coefficients for the three age categories are all quite a
bit larger than their standard errors. This means that the coefficients are different
from zero and any observed differences are not likely by chance. The
coefficients (standardized and unstandardized) for age at offense and age at
hearing are negative, while the coefficients for age at re-offense is positive. This
tells us that, when other independent variables are held constant, the average
amount of days to reoffending is predicted to decrease by .162 when the age at
offense variable goes up by one unit, decrease by 2.636 when the age at hearing
variable goes up by one unit, and increase by 2.662 when the age at time of re-
offense goes up by one unit.
92 Table 27
Summary of Regression Analysis for Days to Reoffending by Demographic and Legal
Characteristics
Unstandardized Standardized Variable Coefficients Coefficients t Sig. B Std. Error Beta (Constant) .412 14.022 .029 .977 Age at Offense* -20.505 7.455 -.162 -2.751 .008 Age at Hearing** -339.721 7.513 -2.636 -45.219 .000 Offense Level -.346 .759 -.003 -.456 .650 Offense Type (Status = 1) -1.167 4.073 -.002 -.287 .775 Age at Re-offense** 360.093 1.989 2.662 181.087 .000 Re-offense Level .358 .531 .005 .674 .502 Escalation (Yes = 1) -1.468 2.871 -.004 -.511 .611 Race (Black = 1) -.264 2.465 -.001 -.107 .915 Gender (Male = 1) -1.046 2.178 -.002 -.480 .632 Note. *p < .01, **p <.001
93 Research Question Three
The third research question of this study examines whether reintegrative shaming is negatively associated with offense escalation. For this research question, only those youth who reoffended were included in the analyses (n =
82). The hypothesis derived from this research question is, re-offending juveniles who participated in the Reintegrative CDP would have lower instances of offense escalation than juveniles who participated in the Typical CDP. Cross
– tabulation and logistic regression analyses were conducted to test this hypothesis.
The analyses did not support the hypothesis. There was no significant association between offense escalation and diversion program (see Table 28).
Table 28.
Summary of Offense Escalation by Diversion Program Did Not Escalate Escalated P Value Variable % N % N Program .392 Reintegrative 55 17 45 14 Typical 45 23 55 28
Likewise, the logistic regression analyses revealed no significant associations of categorical predictors to offense escalation. However, an analysis of variance revealed a significant relationship between offense level and escalation (F =
8.169, df1, p = .005). For re-offenders who did not escalate their offending, the mean offense level for their initial offense was 4.65 (SD 1.76): higher than re-
94 offenders who did escalate their offending: 3.60 (SD = 1.57). Table 29 shows that the difference was statistically significant (B = -.524, df 1, p = .005).
Table 29.
Summary of Logistic Regression Analysis for Variables Predicting Offense
Escalation (n = 82)
P Odds Ratio Predictor B SEB Value (eB)
Demographic Variables Gender (Male) .237 .576 .680 1.268 Race (Black) .431 .658 .513 1.538 Age -.362 .398 .363 .696
Legal Variables Offense Type (Status) 1.417 1.018 .164 4.124
Offense Level* -.524 .186 .005 .592 Program Disposition (Completed) (1) -1.318 1.351 .329 .268 (2) .024 .570 .967 1.024 Note: Program Disposition included 3 categories (Completed, Did Not
Complete, and Dismissed). Completed is the reference category.
*p < .01
95 Research Question Four
The fourth and final research question examines whether there are any
differences of statistical significance between status offenders and delinquent
offenders with regard to program outcomes. Because the conduct of status
offenders is deemed criminal by virtue of the fact that the offenders are below the
age of majority as opposed to their conduct being predatory in nature, diversion
programs are theorized to have less of an impact on status offenders compared
to delinquent offenders. Therefore, it was hypothesized that the diversion
programs would impact status offenders differently from delinquent offenders.
Data analyses did not support the hypothesis.
Table 30 displays characteristics of the sample by offense type (status
offenders versus delinquent defenders). Cross-tabulation analysis and a t-test
were conducted to determine any significant associations between offense type
and demographic and legal variables. Age was the only variable found to be
significantly associated with offense type t(531) = -4.60, p = .000). Status
offenders had a mean age of 15.77 (SD = 1.20) and were significantly older than
delinquent offenders with a mean age of 14.52 (SD = 2.09). A split file
examination of the sample characteristics by program revealed that gender was
significantly associated with offense type in the Typical CDP. Males in the
Typical CDP were more likely to be delinquent offenders than status offenders
(χ2 = 4.730, df 1, p = .030). As was the case with the total sample, status offenders were significantly older than delinquent offenders in both programs:
96 Reintegrative CDP t (206) = -3.32, p = .004); Typical CDP t (323) = - 3.77, p =
.001).
Table 30.
Characteristics of Sample by Offense Type
Variable Status Offenders Delinquent Offenders (N= 62) (N= 471)
Mean SD N Mean SD N
Age* 15.77 1.20 62 14.52 2.09 471
% % Gender Male 10.4 36 89.6 309 Female 13.8 26 86.2 162
Race Black 11.2 50 88.8 395 Non-Black 13.6 12 86.4 76
Disposition Dismissed 7.7 6 92.3 72 Completed 11.9 48 88.1 354 Program Did Not 15.1 8 84.9 45 Complete Note. * p < .01
As noted in previous tables, logistic regression analyses revealed that whether a juvenile was a status offender or a delinquent offender was not a significant predictor of program completion (see Table 23), recidivism (see Table
25), or offense escalation (see Table 29) for the sample population. An analysis of the same by diversion program found that the only significant difference
97 between status offenders and delinquent offenders with regard to program
outcomes was seen in the Typical CDP. Offense type was a significant predictor
of program completion (B = -1.586, df 1, p = .049). Delinquent offenders in the
Typical CDP were more likely to complete the program than status offenders in the Typical CDP. Chi Square analyses did not reveal any significant differences in the program outcomes of status offenders in the Reintegrative CDP compared to status offenders in the Typical CDP.
98 Programmatic Components and Sanctions
Because the study’s hypotheses were not supported by the data analysis,
an exploration of the impact of programmatic components/sanctions on program
outcomes seemed warranted.
To assess the impact of the punitive, educational, therapeutic, and
reintegrative programmatic components/sanctions on the outcome, program
completion, juveniles whose cases were dismissed were removed from the
analysis since those juveniles were not sanctioned (n = 455). Logistic regression
analyses revealed that the educational variable was found to be a significant
predictor of program completion (B = -.593, df 1, p = .007) (see Table 31).
Table 31.
Logistic Regression Analysis for Programmatic Components Predicting Program
Completion (n = 455) eB B P Predictor SEB Wald Odds Value Ratio Punitive .169 .218 .604 .437 1.184 Educational* -.593 .220 7.297 .007 .553 Therapeutic -.259 .226 1.312 .252 .772 Reintegrative .243 .213 1.296 .255 1.275
Constant 2.142
χ2 8.363
df 4 Note. *p < .01
99 The more educational sanctions given to a juvenile, the less likely he was to
complete the program. No other significant associations were found.
The strength of the educational program component comes primarily from the Typical CDP. For exploratory purposes, a comparison of the two programs using the split file analysis revealed that the educational variable was a significant predictor of program completion for the Typical CDP in the same direction as the overall model (B = -1.340, df 1, p = .001). There were no significant predictors found for the Reintegrative CDP.
Logistic regression analysis of the four program predictors and recidivism again revealed the educational program component to be the sole significant predictor variable (see Table 32). The more educational program sanctions given to youth, the more likely they were to reoffend.
Table 32.
Logistic Regression Analysis for Programmatic Components Predicting
Recidivism (N = 455) eB B P Predictor SEB Wald Odds Value Ratio Punitive .046 .170 .073 .787 1.047 Educational* .513 .196 6.839 .009 1.670 Therapeutic .364 .194 3.537 .060 1.440 Reintegrative -.182 .182 .999 .318 .834
Constant -2.033
χ2 11.350
df 4 Note. *p < .01
100 A split file analysis revealed that the educational program component was
a significant predictor of recidivism for the Typical CDP only (B = 1.333, df1, p =
.000. The therapeutic component was a predictor of recidivism in the
Reintegrative CDP at the .10 level (B = .459, df1, p = .084). No programmatic components were found to be statistically significant predictors for offense escalation (see Table 33).
Table 33.
Logistic Regression Analysis for Programmatic Components Predicting
Offense Escalation (N = 82) eB P Predictor B SEB Wald Odds Value Ratio Punitive .362 .356 1.033 .309 1.436 Educational .291 .368 6.28 .428 1.338 Therapeutic -.142 .360 .154 .694 .868 Reintegrative .159 .318 .250 .617 1.172
Constant -.497
χ2 2.612
df 4
101 Exploratory Research
Forty survey packets were sent to families randomly chosen from each of the two diversion programs. The decision to mail only 40 packets was made so
as not to overburden the diversion staff members who mailed the surveys.
Therefore, no power analysis was conducted to determine an appropriate sample
size. This part of the research was exploratory, extending the use of
administrative data to collect original data.
Only six youth questionnaires and six parent/guardian questionnaires
were returned. The low response rate, 6 out of 40 or 15 %, affected the quality of the survey results and the potential for non response bias significantly impacts the accuracy of survey estimates (Fogliani, 1999). Therefore, only the frequency of responses will be reported.
Each survey questionnaire contained ten questions using a Likert scale format for responses. The possible responses were: 1 – “strongly disagree”,
2 – “disagree”, 3 – “not sure”, 4 – “agree”, and 5 – “strongly agree”.
Table 34 gives a summary of the questions asked of juveniles along with the frequency of their responses. The complete questionnaire is in Appendix B.
Of those who responded, the majority of juveniles indicated that they were happy for the opportunity to participate in the diversion program, appreciated the wrongfulness of their actions, felt the program was helpful, and had a relatively positive outlook for the future. However, the responses indicate that some youth were uncertain about the helpfulness of the programs.
102 Table 34.
Juvenile Responses to Questionnaire (N = 6) % Cumulative Response
Strongly Not Strongly Questions Disagree Agree Disagree Sure Agree 1. Glad to be able to participate. 33 33 33 2. The CDP was good for me. 17 33 50 3. Recognition of wrongfulness of 17 50 33 offense. 4. Negative feelings about offense. 33 33 33 5. Nervous/scared about hearing. 50 50 6. Recognition of consequence 33 67 after CDP. 7. Better person since CDP 33 33 33 experience. 8. No reoffending since CDP. 50 17 33 9. Positive future outlook. 33 67 10. Future better decision making. 17 17 67
Overall, the respondents were in the greatest agreement about a positive outlook for the future. The patterns that appear to emerge from this important, albeit small, sample are that 1) juveniles understand the nature of their offending
and 2), in spite of having committed an offense, they feel they are still capable of
doing good.
In addition to the ten scaled-response questions, the questionnaire
contained an eleventh question that enabled respondents to include anything
they might what to express about the program. All responses are included in their original language with the exception of omitting names. The question reads:
Use the space below to write anything else you would like to say about the community diversion program, your hearing, the things you had to do after your hearing or anything else you would like to say.
103 Only two of the six respondents answered question 11, and one of the two indicated that he/she would rather comment via e-mail. Although not completely clear, this young person ironically seemed to have some specific reservations about expressing his/her thoughts, feelings or experiences in written form. The reservations appear to have something to do with the department of children and family services. His/her comment states:
I like to write you e-mail, this case child and family service I no longer write things do. I can get a better ideal my story.
Because the statement is not well written and somewhat difficult to understand, it is not clear who is “you” to which the writer directs his/her comments: the researcher or the diversion program coordinator. I, the researcher, interpret the comment to read:
I would like to e-mail you my response to this question as opposed to writing a response on paper. The reason is, because of my experiences with, or because of what I think or know about, the department of children and family services, I no longer write things down. If I can e-mail you instead of writing my response on the paper you provided, I can give you a better idea of the circumstances surrounding my case.
This interpretation, although purely speculative, is reasonable.
The second youth response states:
It did help me realize how MANY people are affected by just one little thing you do. I was very happy that the diversion program was available for me, so that the 1 stupid thing I did didn't stay on or even make me a record. It is valuable, as where it teaches you the effects, and gives you bigger perspective of what you're doing instead of just getting probation or community service to where you gain no knowledge of why it was wrong instead of just being like "Oh" well I was wrong. Put the picture with the words. I thank you for allowing me to be part of this study. Because I do feel the D. program helps many youth. It did Me!
104 This one response has no generalizing or representative value of course, but it at least gives an idea of what someone who participated in the program thinks and feels about the program in his/her own words.
Table 35 displays the responses to the parent/guardian questionnaire.
The complete questionnaire is in Appendix C. The results show that parents were very happy about the opportunity for their children to participate in the diversion program. They had a great appreciation for the wrongfulness of their children’s actions but did not feel responsible for those actions. Their assessment about the helpfulness of the program was neutral to positive, and the parents had a positive outlook about their children’s future.
Table 35.
Parent/Guardian Responses to Questionnaire (N = 6) % Cumulative Response
Strongly Not Strongly Questions Disagree Agree Disagree Sure Agree 1. Glad child could participate in CDP. 100 2. The CDP was good for my child. 100 3. Recognition of wrongfulness of 33 67 offense. 4. Sense of Responsibility for actions. 67 17 17 5. Nervous/scared at hearing. 33 17 17 33 6. Recognition of consequence after 17 33 50 CDP. 7. Better person since CDP experience. 17 50 33 8. No reoffending since CDP. 33 17 50 9. Positive future outlook for child. 17 33 50 10. Future better decision making by 17 50 33 child.
105 Overall, parents were in the greatest agreement with regard to their feelings on the opportunity for their children to participate in the diversion program and agreed least about their level of nervousness during the diversion hearing. This indicates that although parents overwhelmingly felt good about the juveniles being given a chance to correct their behavior outside the formality of the juvenile court, some were still nervous or apprehensive about the informal diversion hearing.
Again, similar to the youth questionnaire, in addition to having the ten scaled-response questions, the questionnaire contained an eleventh question giving parents/guardians the opportunity to include anything they might what to express about the program. The question reads:
Use the space below to write anything else you would like to say about the community diversion program, your child’s hearing, the things he/she had to do after the hearing or anything else you would like to say.
Of the six respondents, four answered question 11. Their comments are noted below in their original language. However, identifying information has been edited.
- I believe this program effects everyone involved positively. The hard work and dedication of [the diversion program coordinator] can not be put into words. I know [he/she] not only cares about these kids [she/he] loves them and nurtures them. The time and conversations and concern [he/she] gives these kids and their families is overwhelming. [She/he] has become a dear family friend!!! I love [him/her] and so do my children, [she/he] has been their for my daughter even when I have given up (several times!!!) This program is wonderful because of the caring adults that show the children involved there is always a 2nd chance. to do
106 what’s right instead of what’s easy. Please let [the diversion program coordinator] know [he/she] is a blessing. - My daughter was invited to [the diversion program’s] awards banquet for mention on her essay against shoplifting, as was I. Judges, councilpeople & other community leaders were present. The experience gave [child’s name] quite a boost in self esteem, as did the woman from the diversion program that came to speak to us, she was kind, understanding & had good words for my daughter.
- I know this is a wonderful program. I never knew there was so many dedicated and caring people in this serious area. I count it a privilege and a grave opportunity to meet such wonderful people, in spite of any because of this situation that I was engaged. I can never find adequate words to express my thanks, so I will very simply say Thank You !! (Signed by parent).
- I'm sure the program gave my child something to think about when he decide to do something wrong.
Although from a small segment of the study population, these comments uniformly show that parents who responded to the questionnaire were satisfied with the CDP in which their child participated.
107 Chapter 5
Discussion and Conclusion
Discussion
This study examined the association of programmatic components,
demographic, and legal factors on outcomes for two of forty-nine juvenile diversion programs in Cuyahoga County. The outcomes assessed were program completion, recidivism (reoffending), and offense escalation. Using labeling theory and restorative justice as a theoretical framework, specifically this study explored whether a diversion program that utilizes more reintegrative shaming, a restorative justice technique, had better outcomes compared to a diversion program that utilized fewer reintegrative shaming components. Additionally, this study examined whether the diversion programs had a different impact on status offenders versus delinquent offenders.
Statistical analyses did not support any of the study’s hypotheses. There were no significant differences in the outcome variables – program completion, recidivism, or offense escalation – between the two diversion programs.
Regression analyses also failed to reveal any significant interaction between reintegrative shaming program components and outcomes. Finally, the programs were not found to impact status offenders and delinquent offenders differently. Statistical analyses did, however, provide some interesting findings.
108 Program Disposition
Juveniles referred to the diversion programs had their cases disposed of in one of three ways, cases were dismissed, juveniles completed the program or juveniles did not complete the program. For the sample, there were no demographic factors found to have any significant interaction with program disposition, and offense level was the only legal factor found to interact significantly with program disposition. Youth who had their cases dismissed had a mean offense level that was significantly higher than those youth who continued through the programs. This finding suggests that the more culpable offenders arguably were given the lightest treatment. It is doubtful that this was so.
As is the case with juvenile court adjudications and adult criminal proceedings, the offenses assigned to juveniles in the diversion programs are charges or allegations that the juvenile has committed a certain crime. For a juvenile to have a case dismissed at the diversion hearing, the magistrate likely felt that the youth was overcharged, there was insufficient information to find the youth responsible for the offense, or the hearing magistrate may have decided that whatever transpired at the hearing, for example, a stern lecture, awareness of restitution having already been made, adequate parental intervention, and/or a great display of remorse, was sufficient to dispose of a case by dismissal.
109 Program Completion
Although the hypothesis with regard to program completion was not supported, logistic regression did reveal that offense level was a significant predictor of whether those you who did not have their cases dismissed would complete the diversion program. The results indicate that the greater the offense committed by a juvenile referred to a CDP, the greater likelihood of completing the program. Additionally, whether a juvenile was a status offender or a delinquent offender was a significant predictor for the Typical CDP but not for the
Reintegrative CDP. Delinquent offenders in the Typical CDP were more likely to complete the diversion program than status offenders.
Recidivism
Two demographic variables were found to have a significant association to whether a youth re-offended, they are gender and age range at the time of offense. Males were more likely to reoffend than females and analysis also revealed that younger offenders, those 13 years of age and younger, were more likely to reoffend than those juveniles in the program who were 14 and older.
These findings suggests that a lack of sophistication and experience play a role in reoffending and maybe younger offenders, particularly males, ought to be exposed to program interventions that are different from those imposed on older offenders. At a minimum, the findings suggest that the gender and age appropriateness of program components should be evaluated.
110 Offense Escalation
The legal variable, offense level, was the only variable shown to have any
significant interaction with offense escalation. Re-offenders who did not escalate
their delinquent conduct had a higher mean offense level for their initially referred offense than those re-offenders who did escalate their offending conduct. The
implication of this analysis is, the more severe the offense initially committed by a
juvenile referred to a CDP, the less likely the youth will commit a more severe
offense if he reoffends. This analysis, however, is made with great caution for
several reasons. First of all, the juveniles that are generally eligible for referral to
a diversion program are first time, non-serious offenders. This means that
juveniles charged with a felony offense are not likely to be referred to a diversion
program. They are more likely to be formally adjudicated in the juvenile division
of the Common Pleas court or bound over to the general division of the Common
Pleas court to be tried as adults depending on the severity of the crime.
A second reason for caution with this analysis is that the number of
offense levels, eleven, is limited and range from status offense (S) to first degree
felony (F1). In this study, the offense levels are coded as interval variables
assigned values of one through eleven. Obviously the more severe the initial
offense committed, meaning the higher the assigned value, the fewer offense
levels left to exceed. For instance, a youth who committed a first degree
misdemeanor (M1, assigned the value 6), has only 5 higher offense levels that
are possible to commit in order to escalate his offending conduct. It follows that
111 a youth referred to a CDP for committing a status offense (S, assigned the value
of 1) has ten, or twice the amount of possible offense levels to exceed.
A final reason for caution with regard to the results is that there are some
property offenses that are assigned severity levels based on the value of an item.
Therefore whether a subsequent offender is deemed to have escalated his
delinquent conduct may be more dependent on the object of the offense as
opposed to the offender’s actions or his culpability. For instance, a diverted
youth who initially steals an item valued at $450.00 would be charged with theft as a first degree misdemeanor (M1, assigned the value 6). If he reoffends by committing another theft but the item is valued at $500.00, he would be charged with theft as a fifth degree felony (F5, assigned the value 7). By criminal offense level standards, this juvenile has escalated his offending. However, as the above example demonstrates, in some instances, the demarcation of offense levels may just be a distinction without a meaningful difference.
Programmatic Components
Of the four categories of programmatic components and sanctions – punitive, educational, therapeutic, and reintegrative – the educational component was found to be the sole predictor of program outcomes for the sample population. It was a significant predictor of the outcomes, program completion and recidivism.
The educational component was found to have a significant negative relationship to program completion. Oddly enough, the results imply that the
112 more educational sanctions imposed on a juvenile the less likely the juvenile will
complete the program. There can be multiple reasons for this finding. Arguably, youth who are sanctioned with an educational program component are likely not
doing well academically. Imposing educational sanctions in the context of a
diversion program may not only be unproductive, but doing so may even
increase possible academic and educational frustration.
Analysis revealed that the educational component was a significant
predictor of recidivism in that the more educational components imposed the
more likely a youth will reoffend. The same analysis discussed above regarding
the impact of this component to program completion can be applied here. The impact of educational components on program completion and recidivism
suggested by the analyses presents an area ripe for further and more in depth research.
Exploring the impact of the programmatic components on the respective
diversion programs revealed that the educational component was a significant
predictor of recidivism in the Typical program and the therapeutic component was a significant predictor of recidivism in the Reintegrative program. As the impact of the educational program component was discussed above, only the therapeutic predictor will be discussed, and although the therapeutic component
was found to be a significant predictor of recidivism for the Reintegrative CDP at the .10 level only, the implications are worth discussing.
Analysis of the relationship between the therapeutic component and the
Reintegrative CDP reveals that, because the outcome variable, recidivism, was
113 coded as “yes” = 1, the analysis means that the more therapeutic components imposed on juveniles in the Reintegrative program, the more likely they were to reoffend. The majority of therapeutic sanctions imposed were for drug and/or alcohol related issues. This finding suggests that these juvenile offenders who used or abused substances were more likely to reoffend than referred juveniles who were not known to be substance users or abusers. This analysis would not be novel as it is consistent with the literature examining alcohol or substance
abuse and repeat offending (Niarhos & Routh, 1992; Myner, Santman,
Cappelletty & Perlmutter, 1998; Harrison, 2001; Langan & Levin, 2002; Taxman,
Perdoni, & Harrison, 2007). Other therapeutic components included counseling for behavioral problems and anger management. Again, the imposition of these program components indicates that some juveniles may have social, psychological or mental health issues that need to be addressed. Not surprisingly, such issues play a key role in delinquent behavior for some juveniles
(Grisso, 2007).
Conclusion
The hypotheses formulated for this study were not supported by the data.
Although the Reintegrative CDP utilized more reintegrative program components than the Typical CDP, there were no significant differences between the two programs with regard to program outcomes. In addition, reintegrative program components were not significantly associated with any program outcomes.
Comparing two or more diversion programs with only one utilizing reintegrative
114 shaming components and the other(s) using none would produce a stronger model for study under the theoretical framework. Although the hypotheses were not supported, the findings do reveal some useful information regarding the study’s research questions.
The above discussion suggests that some diversion program components do significantly impact programmatic outcomes. The educational program component was a significant predictor of program completion and recidivism for the sample and the therapeutic program component was a significant predictor of recidivism for the Reintegrative CDP. A program that would systematically evaluate each of its programmatic components could find support for discontinuing those components or sanctions found to have no relationship to program outcomes or those found to have a negative impact on program outcomes. Additionally, resources could be better directed to those programmatic components found to have a positive impact on program outcomes.
The demographic characteristics of gender and age were found to be significant predictors of recidivism only. This indicates that the impact of diversion programs, as it relates to reoffending, is not the same for males and females or for all youth regardless of age. Although race was not found to be a predictor of any program outcome, the sample was fairly homogenous since the majority of juveniles in both diversion programs were black. A more diverse sample might yield different results.
115 The level of offense initially committed by a juvenile was the only legal
variable revealed to be a predictor of program outcomes. Also interesting was,
whether a juvenile had his case dismissed and therefore not subject to any program sanctions or components, had no significant impact on recidivism or offense escalation. This finding, or lack thereof, would be of interest to those who espouse the view that diversion programs only, or primarily, serve to widen the net of juveniles subjected to some form of organized intervention, regardless of its informal nature compared to that of official adjudication through the juvenile court. On the other hand, one could argue that the focus here should not center on whether there is net widening – that is, referring a youth to a diversion program instead of taking no action, formally or informally – but that the focus should center on whether simply acknowledging a juvenile’s delinquent behavior and taking no further action is just as effective as imposing sanctions and having a youth go through a diversion program. This is an experimental study that would appear to be well worth conducting.
Finally, the study revealed that there were no statistically significant differences in program outcomes between status offenders and delinquent offenders for the sample, but being a delinquent offender was a significant predictor of program completion for youth in the Typical CDP. There were no programmatic components that impacted status offenders differently from delinquent offenders.
116 Implications for Policy, Practice, and Research
There were high expectations for diversion programs during their inception
in the late 1960s. However, time has demonstrated that the programs are not a
panacea for juvenile offending. There have been some noted successes with
diversion programs, but there have been some noted failures. Many criticisms of
diversion still abound, namely: that they widen the net of the juvenile court,
bringing in more youth under the auspices of the court than there would be without the programs. Additionally, problems such as lack of adequate funding, no systematic tracking of youth, and having overworked diversion staff who must divide their time and energy with other work responsibilities, no doubt have an impact on the operations of the programs.
Further study conducted on the outcomes of specific programmatic components would inform decision making regarding the operation of diversion programs and assess their overall effectiveness. Additional study may also establish or support some consistent standards of operation. A research study proposing to randomly assigning juveniles to either a diversion program or the formal processing of juvenile court would pose ethical problems. However, to test the hypothesis that certain types of program components, like reintegrative shaming, will improve program outcomes, an experimental study could be conducted by randomly assigning juveniles from the study population to two or more diversion programs with one having well defined reintegrative shaming components. The subjects could then be followed over time to examine outcomes.
117 To enhance the understanding of how a youth’s background and the
various types of program components and sanctions impact program outcomes,
more comprehensive and detailed data collection is needed (Hogan & Campbell,
2005). This data collection should included the juveniles’ family socio-economic
status; information on parent/guardian’s educational attainment, alcohol or drug
use, prior or current incarcerations, and physical and mental health; familial and
peer influences; whether the juvenile uses or abuses substances; the juvenile’s
educational performance; and specific information on the diversion program
components and sanctions assigned to youth so that the components/sanctions
can be properly measured.
As the foundation upon which diversion programs are premised, labeling
theory and its assumptions remain ripe for theoretical application and research.
Future studies might explore the fact that, because juvenile offending is classified as status or delinquent offending, and because the majority of youth who are referred to diversion programs are done so for delinquent offending, the negative effects of being labeled may still be associated with informal processing via diversion. Also, although the study’s hypotheses were not supported by this research study, the absence of support should not result in a lack of confidence in reintegrative shaming as a theoretical guide. Questions still remain about how best to empirically test the theory (Hay, 2001, Zhang, 1995) and there is little guidance on how to define and measure reintegrative shaming components
(Zhang, & Zhang, 2004). Instead, future research should examine ways to empirically tease out reintegrative shaming techniques in diversion programs and
118 focus on factors that might serve to prevent or constrain operating programs in
accordance with the theoretical guide.
Diversion program completion rates and the more commonly relied upon
recidivism rates are statistically sound measures of programmatic success. But
future research should focus also on how juveniles and their parents or
guardians define program success. For these stakeholders, program success
may be defined in broader terms: for instance, better academic performance;
increased self esteem; better decision making processes; increased respect for
oneself and for others; and obtaining a greater appreciation of how deviant
behavior impacts society.
To say that, understanding the significant relationships between education
and educational diversion program components to delinquency is important,
would understate what may be one of the most obvious paths to preventing and
curbing deviant behavior. While many studies have presented laudable
associations between education, school bonds and delinquency (Frease, 1973;
Pink, 1984; Niarhos & Routh, 1992; Hsing, 1995; Soares, 2004; Sprott, Jenkins &
Doob, 2005), utilizing educational program components, like the ones characterized in this study, may require a more critical analysis of potential
unintended consequences.
Mental health and alcohol and/or substance abuse issues continue to play a major role in criminal offending and recidivism (Hamilton et al., 2007; Sullivan et al., 2007; Prinz & Kerns, 2003). The impact is seen in juvenile populations as well as in adult populations (Niarhos & Routh, 1992; Harrison, 2001; Langan &
119 Levin, 2002; Taxman et al., 2007). Early diagnoses and treatment would likely
go a long way in delinquency prevention, but systems – institutions,
governmental agencies, and programs – should have in place adequate
resources and mechanisms to minimize the impact of these issues on delinquent
offending.
Noted in the introduction of this study are statistics relative to the
nationwide decrease of juvenile crime over the years. Having the ability to
evaluate any causal associations of diversion programs to the decrease in
juvenile offending would certainly justify, if not mandate, their continued operation
regardless of structure or lack of knowledge of the impact of programmatic
components. It appears, however, that at a basic level, diversion programs will
remain a popular alternative to the formal adjudication of youth through the
juvenile justice system. As such, there is a responsibility to make programs as
effective as possible and policy makers must commit to allocating a sufficient amount of resources for effective programming – resources no doubt saved from diverting youth from the formal adjudication of juvenile court.
Limitations of Study
Research studies, particularly those that do not use an experimental
design, have certain limitations. This research project was no exception.
Although a secondary data analysis using archival data is cost-effective, the
study was constrained by the data collection and fraught with methodological difficulties. There were important variables that could not be assessed because
120 data were missing, unreliable, or not collected. These variables include custodial
head of household, highest educational level attained in household, religious
affiliations, peer group influences, cultural influences, and socio-economic status.
For example, “mother” was listed as the custodial head of household for the
majority of cases. This would suggest that, at the time juveniles were referred to
the diversion programs, the majority of them lived in single parent, female
headed households. Although this conclusion is supported by the diversion
program coordinators (N. Stroman, personal communication, February 17, 2005;
D. Stolarski, personal communication, February 17, 2005), a statistical analysis
of the data provided would be unreliable due to the fact that the juvenile court
collects information on a juvenile’s mother only, unless the mother is not the
custodial parent or one of the custodial parents. This means that any child in a
two-parent home is not noted. Although there are instances where the
Reintegrative CDP data note both parents as custodial, in keeping with the data collection process of the juvenile court5 the Typical CDP data contain no
instances of a two parent home. Also, because data on recidivism and offense
escalation were collected only by official filings in the juvenile court, the figures
are likely understated.
Another limitation with regard to the quality of the data is the fact that there
was no control over data collection, so the study had to rely on existing record
information where measurement of program exposure variables, particularly
reintegrative shaming, was based solely on report information and subsequently
a content analysis. As was the case in the study by Zhang and Zhang (2004),
121 this procedure of data collection in a study using secondary analyses likely lacks the sensitivity necessary to capture the subtleties of reintegrative shaming in a diversion program. Furthermore, there were aspects of the Reintegrative CDP that could not be measured. For instance, the Reintegrative CDP would hold a banquet each year to recognize those youth who completed the program. This event could certainly be characterized as a salient reintegrative shaming program component. However, no official records were kept regarding the event. The probable inability to capture subtle aspects of reintegrative shaming with this type of study likewise may be true for other program components. There also was no mechanism to evaluate the accuracy of the original record data.
Although the Reintegrative CDP utilized more reintegrative shaming programmatic components, the Typical CDP was not completely devoid of reintegrative shaming components. This fact raises questions about the internal validity of the study.
The study design for this project was not experimental and did not include a randomly selected control group. Therefore, no causal pathways could be measured. This poses a threat to external validity, so generalizing the findings outside of the two subgroups would be inappropriate.
Although the study had the benefit of assessing recidivism and offense escalation for up to a year following program completion, collecting data over a longer period of time would make the study more robust. Also, when evaluating offense escalation for those who reoffended, the sample size decreased
122 dramatically. Because of this, there may not have been enough statistical power
in the data analysis of this program outcome.
Assessing recidivism, and consequently offense escalation, was limited to
juvenile reoffending only. This limitation impacted the study in a couple of ways:
1) juveniles who were age eighteen and over when their diversion hearings were
held had to be excluded from the sample; and 2) the effects of diversion, if any,
on older teenagers could not be fully assessed.
Finally, the results of the questionnaires shed some light on juvenile and
parental attitudes, impressions, and feelings about: 1) the diversion programs; 2)
the juvenile’s offending; and 3) the juvenile’s future outlook. However, the questionnaires were brief and did not adequately flush out these impressions.
The instruments were not piloted or tested before being distributed to the study
population. Also, they were sent to a relatively small subgroup of the sample, there was no follow-up mailing, the recipients were selected by the respective diversion program coordinators, and the response rate was low. Furthermore, if parents were present when juveniles completed the questionnaire, or if parents had access to the questionnaires before being sealed and mailed, parental presence may have influenced one or more responses, particularly since the questionnaire asks about misconduct. All of these conditions mean that the data derived from the responses could reflect possible biases, for instance, non- response bias and intentional or unintentional selection bias.
Limitations notwithstanding, this study sheds light on areas of juvenile diversion programming that are ripe for detailed exploration and analysis. The
123 relative ease by which the programs operate is likely a key factor to their popularity. As diversion programs continue to operate in northeast Ohio and throughout the United States, concerted efforts should be devoted to developing modalities that are grounded in theory with programmatic components whose implementation and impact can be accurately measured. However, these recommendations should be implemented in a way that does not overburden the service providers and does not so greatly increase the cost of operating the programs that a cost benefit analysis would prove the implementations detrimental. Lastly, a systematic way of evaluating diversion programs should be established before they begin to operate. Hopefully this study assists in improving upon the positive work of these programs.
124 Appendix A
TITLE OF STUDY: An Outcome Study on the Effectiveness of Juvenile Diversion Programs on Recidivism and Offense Escalation for Non-Serious Delinquent and Status Offenders
INVESTIGATOR: Melody J. Stewart PHONE: (216) 534-2893 E-MAIL: [email protected]
RE: Study Participation DATE: September 2005
Dear Parent/Guardian:
The East Cleveland/Maple Heights Community Diversion Program has agreed to participate in a research project for my dissertation. For the research, I am looking at two juvenile diversion programs in Cuyahoga County in an effort to see if certain program components help to reduce repeat offending and/or help to prevent committing more serious offenses. In addition to looking at this information, I am hoping to get some feedback from parents/guardians and juveniles who are or have been a part of the community diversion program. This letter is to request your participation in this study.
Should you agree to be a part of this study, your input will be invaluable to gaining a more comprehensive view of the (East Cleveland or Maple Heights) program. Participation will involve completing the attached Consent Forms for you and your child, answering the enclosed questionnaire, and mailing both documents back in the self-addressed stamped envelopes. Your participation is completely voluntary and is separate from any activity with the Community Diversion Program. Your responses will be kept confidential and any information used in the study will be reported anonymously.
If you have any questions about the study or protecting you as a participant, please contact me at the telephone number or e-mail address listed above.
Thank you in advance for your consideration.
Sincerely,
Melody J. Stewart
125 Appendix B
QUESTIONNAIRE (Child)
For the questions below, please circle the number that best describes your feelings about each question.
1. I am glad I was able to participate in the juvenile diversion program.
1 2 3 4 5
Strongly disagree Disagree Not sure Agree Strongly agree
2. The community diversion program was a good program for me.
1 2 3 4 5
Strongly disagree Disagree Not sure Agree Strongly agree
3. When I did the thing or things that got me referred to the diversion
program, I knew that what I was doing was wrong.
1 2 3 4 5
Strongly disagree Disagree Not sure Agree Strongly agree
4. I feel bad or ashamed about the things I did that got me sent to the
diversion program.
1 2 3 4 5
Strongly disagree Disagree Not sure Agree Strongly agree
126 5. I was scared or nervous at my diversion hearing.
1 2 3 4 5
Strongly disagree Disagree Not sure Agree Strongly agree
6. I did not know how much my actions hurt other people until going through
the diversion program.
1 2 3 4 5
Strongly disagree Disagree Not sure Agree Strongly agree
7. Going through the diversion program has helped me be a better person.
1 2 3 4 5
Strongly disagree Disagree Not sure Agree Strongly agree
8. Since going through the diversion program, I have not done anything that
might get me sent to court.
1 2 3 4 5
Strongly disagree Disagree Not sure Agree Strongly agree
9. I feel like I can do good things in the future.
1 2 3 4 5
Strongly disagree Disagree Not sure Agree Strongly agree
127 10. In the future, I think I will make better decisions about my conduct and
behavior.
1 2 3 4 5
Strongly disagree Disagree Not sure Agree Strongly agree
11. Use the space below to write anything else you would like to say about the
community diversion program, your hearing, the things you had to do after
your hearing or anything else you would like to say.
______
______
______
______
______
______
______
______
______
______
______
______
______
______
______
______
128 Appendix C
QUESTIONNAIRE (Parent/Guardian)
For the questions below, please circle the number that best describes your feelings about each question.
1. I am glad my child was able to participate in the juvenile diversion
program.
1 2 3 4 5
Strongly disagree Disagree Not sure Agree Strongly agree
2. The community diversion program was a good program for my child.
1 2 3 4 5
Strongly disagree Disagree Not sure Agree Strongly agree
3. When my child committed the act or acts that got him/her referred to the
diversion program, he/she knew that the actions were wrong.
1 2 3 4 5
Strongly disagree Disagree Not sure Agree Strongly agree
4. I feel bad or ashamed about the things I did that got my child sent to the
diversion program.
1 2 3 4 5
Strongly disagree Disagree Not sure Agree Strongly agree
129 5. I was scared or nervous at my child’s diversion hearing.
1 2 3 4 5
Strongly disagree Disagree Not sure Agree Strongly agree
6. I don’t think my child knew how much his/her actions hurt other people
until going through the diversion program.
1 2 3 4 5
Strongly disagree Disagree Not sure Agree Strongly agree
7. I think that going through the diversion program has helped my child be a
better person.
1 2 3 4 5
Strongly disagree Disagree Not sure Agree Strongly agree
8. Since going through the diversion program, I don’t think my child has done
anything that might get him/her sent to court.
1 2 3 4 5
Strongly disagree Disagree Not sure Agree Strongly agree
9. I feel like my child will do good things in the future.
1 2 3 4 5
Strongly disagree Disagree Not sure Agree Strongly agree
130 10. I think my child will make better decisions about his/her conduct and
behavior.
1 2 3 4 5
Strongly disagree Disagree Not sure Agree Strongly agree
11. Use the space below to write anything else you would like to say about the
community diversion program, your child’s hearing, the things he/she had
to do after the hearing or anything else you would like to say.
______
______
______
______
______
______
______
______
______
______
______
______
______
______
______
131 Appendix D
CASE WESTERN RESERVE UNIVERSITY Mandel School of Applied Social Sciences Consent Form (Parent/Guardian)
You are being asked to participate in a research study about the effectiveness of juvenile diversion programs. You were selected as a possible participant because you have a child who has participated in the (East Cleveland/Maple Heights) Community Diversion Program. Please read this form and feel free to contact me with any questions that you may have before agreeing to be in the research.
My name is Melody J. Stewart, and I am a doctoral student at Case Western Reserve University conducting this study. The purpose of this research is to examine the effectiveness of juvenile diversion programs.
If you agree to be a participant in this research, I would ask you to complete the enclosed questionnaire. This research has no foreseeable risks to you. The benefit of participation is that I will be able to get the thoughts and impressions of individuals like you who have personally been involved with a juvenile diversion program.
The records of this research will be kept private. In any sort of report I might publish, I will not include any information that will make it possible to identify a participant. Research records will be kept in a locked file, and access will be limited to the faculty members of my dissertation committee, the University review board responsible for protecting human participants, regulatory agencies and me.
Your participation is voluntary. If you choose not to participate, it will not affect your current or future relations with the University, the diversion program or any other entity or organization. There is no penalty or loss of benefits for not participating.
If you have any questions, you may contact me at (216) 534-2893 or by e-mail at [email protected]. If I cannot be reached, or if you would like to talk to someone other than the me, the researcher, about: 1) concerns regarding this study, 2) research participant rights, 3) research-related injuries, or 4) other human subjects issues, please contact Case Western Reserve University’s Institutional Review Board at (216) 368-6925 or write: Case Western Reserve University, Institutional Review Board, 10900 Euclid Avenue, Cleveland, Ohio 44106-7230.
You will be given a copy of this form for your records.
132 STATEMENT OF CONSENT
I have read the above information. At this time, I have no questions or I have received answers to the questions I asked. I consent to participate in this research. (I am at least 18 years of age.)
Name of Participant (Print) ______
Signature of Participant ______Date
Name of Person Obtaining Consent (Print) ______
Signature of Person Obtaining Consent ______Date
133 Appendix E
CASE WESTERN RESERVE UNIVERSITY Mandel School of Applied Social Sciences Child Assent Form
You are being asked to participate in a research study about the effectiveness of juvenile diversion programs. You were selected because you participated in the (East Cleveland/Maple Heights) Community Diversion Program. Please read this form and feel free to contact me with any questions that you may have before agreeing to be in the research.
My name is Melody J. Stewart, and I am a doctoral student at Case Western Reserve University conducting this study. The purpose of this research is to examine the effectiveness of juvenile diversion programs.
If you agree to be a participant in this research, I would ask you to complete the enclosed questionnaire. This research has no foreseeable risks to you. The benefit of participation is that I will be able to get the thoughts and impressions of individuals like you who have personally been involved with a juvenile diversion program.
The records of this research will be kept private. In any sort of report I might publish, I will not include any information that will make it possible to identify a participant. Research records will be kept in a locked file, and access will be limited to the faculty members of my dissertation committee, the University review board responsible for protecting human participants, regulatory agencies and me.
Your participation is voluntary. If you choose not to participate, it will not affect your current or future relations with the University, the diversion program or any other entity or organization. There is no penalty or loss of benefits for not participating.
If you have any questions, you may contact me at (216) 534-2893 or by e-mail at [email protected]. If I cannot be reached, or if you would like to talk to someone other than the me, the researcher, about: 1) concerns regarding this study, 2) research participant rights, 3) research-related injuries, or 4) other human subjects issues, please contact Case Western Reserve University’s Institutional Review Board at (216) 368-6925 or write: Case Western Reserve University, Institutional Review Board, 10900 Euclid Avenue, Cleveland, Ohio 44106-7230.
You will be given a copy of this form for your records.
134 STATEMENT OF CONSENT
I have read the above information. At this time, I have no questions or I have received answers to the questions I asked. I agree to participate in this research.
Name of Participant (Print) ______
Signature of Participant ______Date
Name of Parent/Guardian (Print) ______
Signature of Parent/Guardian ______Date
Name of Person Obtaining Consent (Print) ______
Signature of Person Obtaining Consent ______Date
135 Appendix F
CASE WESTERN RESERVE UNIVERSITY Mandel School of Applied Social Sciences Consent Form (Diversion Staff)
You are being asked to participate in a research study about the effectiveness of juvenile diversion programs. You were selected as a possible participant because you do work with the Cuyahoga County Community Diversion Program. Please read this form and feel free to contact me with any questions that you may have before agreeing to be in the research.
My name is Melody J. Stewart, and I am a doctoral student at Case Western Reserve University conducting this study. The purpose of this research is to examine the effectiveness of juvenile diversion programs.
If you agree to be a participant in this research, I would ask you participate in an interview conducted by me. This research has no foreseeable risks to you. The benefit of participation is that I will be able to get the thoughts and impressions of individuals like you who have personally been involved with a juvenile diversion program.
The records of this research will be kept private. In any sort of report I might publish, I will not include any information that will make it possible to identify a participant. Research records will be kept in a locked file, and access will be limited to the faculty members of my dissertation committee, the University review board responsible for protecting human participants, regulatory agencies and me.
Your participation is voluntary. If you choose not to participate, it will not affect your current or future relations with the University, the diversion program or any other entity or organization. There is no penalty or loss of benefits for not participating.
If you have any questions, you may contact me at (216) 534-2893 or by e-mail at [email protected]. If I cannot be reached, or if you would like to talk to someone other than the me, the researcher, about: 1) concerns regarding this study, 2) research participant rights, 3) research-related injuries, or 4) other human subjects issues, please contact Case Western Reserve University’s Institutional Review Board at (216) 368-6925 or write: Case Western Reserve University, Institutional Review Board, 10900 Euclid Avenue, Cleveland, Ohio 44106-7230.
You will be given a copy of this form for your records.
136
STATEMENT OF CONSENT
I have read the above information. At this time, I have no questions or I have received answers to the questions I asked. I consent to participate in this research. (I am at least 18 years of age.)
Name of Participant (Print) ______
Signature of Participant ______Date
Name of Person Obtaining Consent (Print) ______
Signature of Person Obtaining Consent ______Date
137 Appendix G
Questions for Interviews
Diversion Program Personnel
1. How long have you worked with the Community Diversion Program?
2. What is your background, training, and level of education?
3. What are your feelings/impressions about diversion programs?
4. What are your feelings/impressions about your community diversion
program in particular?
5. What is the goal of your diversion program?
6. What is it like for you coordinating the program?
7. In your view, what is or what are some of the more positive attributes, if
any, of the program?
8. In your view, what is or what are some of the more negative attributes, if
any, of the program?
9. If you could make changes to your community diversion program, what
changes, if any, would you make and why?
10. Is there anything else you would like to say or add to this interview?
138
1 For delinquent and status offenses, technically juveniles are not arrested. They are taken into custody.
2 Butts and Snyder (2007) note that youth crime and juvenile crime are not the same. Juveniles are generally below the age of 18. Crime categorized as youth crime encompasses young adults between the ages of 18 and 24. However, for purposes of this study, the terms “juvenile”, “youth” and “young offenders” will be used interchangeably to refer to offenders under the age of 18.
3 The author of this dissertation served as one of the volunteer magistrates from 2002 to 2004.
4 Individuals under the age of 18 commit status or delinquent offenses regardless of the infraction or level of criminality. However, for more severe offenses (felonies), a juvenile may be bound over to the adult court system and prosecuted for the crime as opposed to adjudicated a delinquent.
5 According to Heather Corcoran, director of the juvenile court’s Community Diversion Program, this practice of only noting the mother’s name on juvenile court records is being discontinued.
139 References
American Bar Association (2004). Justice cut short: An assessment of access to counsel and quality of representation in delinquency proceedings in
Ohio. Retrieved May 3, 2004 from, http://www.abanet.org/crimjust/juvjus/ohtalkingpoint.html
Arnett, J. (1992). Reckless behavior in adolescence: A developmental perspective. Developmental Review, 12, 339-373.
Bazemore, G. &Schiff, M. (2005). Juvenile justice reform and restorative justice: Building theory and policy from practice. Devon: Willan Publishing.
Benda, B. B., Corwyn, R. F., & Toombs, N. J. (2001). Recidivism among
adolescent serious offenders: Prediction of entry into the correctional system for adults. Criminal Justice and Behavior, 28(5), 588-613.
Binder, A. & Polan, S. (1991). The Kennedy-Johnson Years, social theory
and federal policy in the control of juvenile delinquency. Crime & Delinquency,
37(2), 242 - 261.
Blechman, E. A., Maurice, A., Buecker, B., & Helberg, C. (2000). Can
mentoring or skill training reduce recidivism? Observational study with propensity
analysis. Prevention Science, 1(3), 139-155.
Bloomberg, T. (1979). Diversion from juvenile court: A Review of the
evidence. In F. L. Fause and P. Brantingham (Eds.), Juvenile justice philosophy:
Readings, cases and comments (2nd ed.) (pp. 39-68). St. Paul: West Publishing
Company.
140
Braithwaite, J. (1989). Crime, shame, and reintegration. New York:
Cambridge University Press.
Butts, J. & Harrell, A. (1998). Delinquents of criminals: Policy options for
young offenders. Washington D.C.: Urban Institute.
Butts, J. and Snyder, H. (2007, spring). Where are juvenile crime trends
headed? Juvenile and family justice today, 16-20. Retrieved August 19, 2007
from, http://www.jbutts.com/pdfs/whereheaded.pdf
Bynum, J.; and Thompson, W. (1999). Juvenile delinquency: A sociological approach. Needham Heights, MA: Allyn and Bacon.
Campbell, J.S., Lerew, C. (2002). Juvenile sex offenders in diversion.
Sexual Abuse: A Journal of Research and Treatment, 14(1), 1-17.
Cohen, M. (1998). The monetary value of saving high-risk youth. Journal of Quantitative Criminology, 14(1), 5-33.
Cuellar, A. E., McReynolds, L. S., & Wasserman, G. A. (2006) A cure for crime: Can mental health treatment diversion reduce crime among youth?
Journal of Policy Analysis and Management, 25(1), 197-214.
Cuyahoga county juvenile court 2003 annual report. Retrieved on January
15, 2005 from,
http://www.cuyahoga.oh.us/juvenile/annual_report/Statistical_Info.htm
Cuyahoga county juvenile court 2006 annual report. Retrieved on April 26,
2007 from, http://www.cuyahoga.oh.us/juvenile/annual_report/Statistical_Info.htm
141
Davidson, W. S., Redner, R., Blakely, C. H., Mitchell, C. M., & Emshoff, J.
G. (1987). Diversion of juvenile offenders: An experimental comparison. Journal of Consulting and Clinical Psychology 55(1), 68-75.
Davidson, W., Redner, R., Admur, R.L., & Mitchell, C.M. (1990).
Alternative treatment for troubled youth: The case of diversion from the justice system. New York: Plenum.
Dick, A. J., Pence, D. J., Jones, R. M., & Geertsen, H. R. (2004). The need for theory in assessing peer courts. American Behavioral Scientist, 47(11),
1448-1461.
Duxbury, E. (1973). Evaluation of youth service bureaus. Sacramento:
California Youth Authority.
Elliott, D.S., Dunford, F.W., & Knowles, B. (1978). Diversion: A study of alternative processing practices: An overview of initial study findings. Boulder:
Behavioral Research Institute.
Elrod, P. & Ryder, R. S. (1999). Juvenile Justice. Gaithersburg: Aspen
Publishers, Inc.
Elrod, P. & Ryder, R. S. (2005). Juvenile Justice: A social, historical, and legal perspective. Sudbury, MA: Jones and Bartlett Publishers.
Encyclopedia of Cleveland History. Retrieved on November 16, 2007 from http://ech.cwru.edu/ech-cgi/article.pl?id=EC
Ezell, M. (1992). Juvenile diversion: The ongoing search for alternatives.
In I. Schwartz (Ed.), Juvenile justice and public policy: Toward a national agenda
(pp. 45-58). New York: Lexington Books.
142
Federal Bureau of Investigation. (2000) Crime in the United States.
Washington: U.S. Department of Justice.
Federal Bureau of Investigation Uniform Crime Reports, retrieved January
15, 2005 from, http://www.fbi.gov/ucr/cius_03/pdf/03sec4.pdf
Feld, B.C. (1999). The transformation of the juvenile court – part II: Race and the “Crack Down” on youth crime, Minnesota Law Review, 84, 327-395.
Fogliani, M. (1999). Low response rates and their effects on survey results. Methodology Advisory Committee paper, November 1999 meeting, retrieved on March 30, 2008 from http://www.nss.gov.au/nss/home.NSF/75427d7291fa0145ca2571340022a2ad/4f c144d438726fbbca2571ab00247118/$FILE/Low%20Response%20Rates.pdf
Forward, J., Kirby, M. & Wilson, K. (1975). Volunteer intervention with
court-diverted juveniles. Denver: Partner’s court diversion program.
Fox, J. W., Minor, K. I., & Pelkey, W. L. (1994). The relationship between
law-related education diversion and juvenile offenders’ social-and self-
perceptions. American Journal of Criminal Justice, 19(1), 61-77.
Frazier, C.E. & Cochran, J.K. (1986). Official intervention, diversion from
the juvenile justice system, and dynamics of human services work: Effects of a
reform goal based on labeling theory. Crime & Delinquency, 32(2), 157-176.
Frease, D. (1973). Delinquency, social class, and the schools. Sociology
and Social Research, 57, 443-459.
Gavazzi, S. M., Wasserman, D., Partridge, C. & Sheridan, S. (2000). The
growing up fast diversion program: An example of juvenile justice program
143
development for outcome evaluation. Aggression and Violent Behavior, 5(2),
159-175.
Gensheimer, L., Mayer, J., Gottschalk, R., & Davidson, W. (1986).
Diverting youth from the juvenile justice system: A meta-analysis of intervention
efficacy. In S.J. Apter & A.P. Goldstein (Eds.), Youth violence: Programs and
prospects, (pp. 39-57). New York: Pergamon Press.
Gilbert, G.R. (1977). Alternate routes: A diversion project in the juvenile
justice system. Evaluation Quarterly, 1(2), 301-318.
Grisso, T. (November, 2007). Do childhood mental disorders cause adult crime? The American Journal of Psychiatry, 164; 1625-1627.
Haapanen, R. & Rudisill, D. (1980). The evaluation of youth service bureaus (Final Report). Sacramento: California Youth Authority.
Hair, J., Black, B., Babin, B., Anderson, R., & Tatham, R. (2006).
Multivariate Data Analysis (6th ed.). New Jersey: Prentice Hall.
Hamilton, Z. K., Sullivan, C. J., Veysey, B. M. & Grillo, M. (2007). Diverting
multi-problem youth from juvenile justice: Investigating the importance of
community influence on placement and recidivism. Behavioral Sciences and the
Law, 25, 137-158.
Harrison, L. (2001). The revolving prison door for drug-involved offenders:
Challenges and opportunities. Crime & Delinquency, 47(3), 462-485.
Harrison, P., Maupin, J. R., Mays, G. L. (2001). Teen court: An
examination of processes and Outcomes. Crime & Delinquency, 47(2), 243-264.
144
Hassett-Walker, C. (2002). Juvenile conference committees issues in
assessing a diversionary court program. Journal of Criminal Justice, 30, 107-119.
Hay, C. (2001). An exploratory test of Braithwaite’s reintegrative shaming
theory. Journal of Research in Crime and Delinquency, 38(2), 132-53.
Heilbrun, K., Sevin Goldstein, N., & Redding, R. (2005). Juvenile
delinquency. New York: Oxford University Press.
Hogan, M.J. & Campbell, J.S. (2005). Contrasting juvenile and program- level impacts on diversion service programs. Youth Violence and Juvenile
Justice, 3(1), 41-58.
Hsing, Y. (1995). Impacts of welfare participation, school dropouts, and police force on illegal drug arrests. International Journal of Social Economics,
22(8), 16-21.
James, A. (1995). Probation values for the 1990s and beyond. The
Howard Journal, 34, 326-43.
Kammer, J. J., Minor, K. I. & Wells, J. B. (1997). An outcome study of the diversion plus program for juvenile offenders. Federal Probation, (61), 51-56.
Kazdin, A. (2003). Research design in clinical psychology (4th ed.) Boston:
Allyn & Bacon.
Kerlinger, F.N. (1992). Foundations of behavioral research (3rd ed.) Fort
Worth: Harcourt Brace.
King, W. R., Holmes, S. T., Henderson, M. L., & Latessa, E. J. (2001). The
community corrections partnership: Examining the long-term effects of youth
145
participation in an afrocentric diversion program. Crime & Delinquency, 47(4),
558-572.
Klein, M.W. (1986). Labeling theory and delinquency policy: An
experimental test. Criminal Justice and Behavior, 13 (1), 47-79.
Klein, N. C., Alexander, J. F., & Parsons, B. V. (1977). Impact of family
systems intervention on recidivism and sibling delinquency: A model of primary
prevention and program evaluation. Journal of Consulting & Clinical Psychology,
45(3), 469-474.
Langan, P., Levin, D., (2002). Recidivism of prisoners released in 1994.
U.S. Department of Justice, Bureau of Justice Statistics. Washington D.C.
Latessa, E. J., Travis, L. F. III, Wilson, G. P. (1984). Juvenile diversion:
Factors related to decision making and outcome. In S. H. Decker (Ed.), Juvenile
justice policy: Analyzing trends and outcomes (pp. 145-165). Beverly Hills: Sage.
Laub, J.H. (2002) A century of delinquency research and delinquency
theory. In M.K. Rosenheim, F.E. Zimring, D.S. Tanenhaus, and B. Dohrn (Eds.),
A century of juvenile justice (pp. 179-205). Chicago: The University of Chicago
Press.
Leiber M.J. & Stairs, J.M. (1999). Race, contexts, and the use of intake diversion. Journal of Research in Crime and Delinquency, 6(1), 56-86.
Lemert, E.M. (1951). Social Pathology. New York: McGraw-Hill.
Lemert, E.M. (1972). Human Deviance, Social Problems, and Social
Control (2nd ed). Englewood Cliffs: Prentice-Hall.
146
Lincoln, S.B. (1976). Juvenile referral and recidivism. In R.M. Carter &
M.W. Klein (Eds.), Back on the Street: Diversion of juvenile offenders (pp. 23-46)
Englewood Cliffs: Prentice-Hall.
Lipsey, M.W., Cordray, D.S., & Berger, D.E. (1981). Evaluation of a juvenile diversion program: Using multiple lines of evidence. Evaluation Review,
5(3), 283-306.
Loeber, R., & Farrington, D. R. (Eds.). (1998). Serious offender and violent offenders: Risk factors and successful interventions. Thousand Oaks, CA: Sage.
Macallair, D. & Males, M. (2004). A failure of good intentions: An analysis of juvenile justice reform in San Francisco during the 1990s, Review of Policy
Research, 21(1), 63-78.
Makkai, T. & Braithwaite, J. (1994). Reintegrative shaming and compliance with regulatory standards. Criminology, 32, 361-385.
Martin, E. & Pruett, M.K., (1998). The juvenile sex offender and the juvenile justice system. American Criminal Law Review, 35(2), 279-332.
Masters, G. (1997). Values for probation, society and beyond. The
Howard Journal, 36(3), 237-247.
Minor, K.; Wells, J.; Soderstrom, I.; Bingham, R.; & Williamson, D. (1999).
Sentence Completion and recidivism among juveniles referred to teen courts.
Crime & Delinquency, 45(4), 467-480.
Moffitt, T.E. (1993). Adolescence-limited and life-course-persistent antisocial behavior: A developmental taxonomy. Psychological Review, 100, 674-
701.
147
Moffitt, T.E. & Caspi, A. (2001). Childhood predictors differentiate life- course persistent and adolescence-limited antisocial pathways among males and females. Development and Psychopathology, 13, 355-375.
Mulvey, E. P., Arthur, M. W., & Reppucci, N. D. (1993). The prevention and treatment of juvenile delinquency: A review of the research. Clinical
Psychology Review, 13, 133-167.
Myner, J., Santman, J., Cappelletty, G., Perlmutter, B. (1998). Variables related to recidivism among juvenile offenders. International Journal of Offender
Therapy and Comparative Criminology, 42(1), 65-80.
Niarhos, F. & Routh, D. (1992). The role of clinical assessment in the juvenile court: Predictors of juvenile dispositions and recidivism. The Journal of
Clinical Child Psychology, 21, 151-159.
Olson, S. & Dzur, A. (2004). Revisiting informal justice: Restorative justice and democratic professionalism. Law & Society, 38(1), 139-176.
Osgood, D.W., & Weichselbaum, H.F. (1984). Juvenile diversion: When practice matches theory. Journal of Research in Crime and Delinquency, 21(1),
33-56.
Palmer, T. & Lewis, R. (1980). A differential approach to juvenile diversion. Journal of Research in Crime and Delinquency, 17: 209-277.
Palmer, T., Bohnstedt, M., & Lewis, R. (1978). The evaluation of juvenile diversion (Final report). Sacramento: California Youth Authority.
Patrick, S & Marsh, R. (2005). Juvenile diversion: Results of a 3-year experimental study. Criinal Justice Policy Review, 16(1), 59-73.
148
Pink, W. (July 1984). Schools, youth, and justice. Crime & Delinquency,
30(3), 439-461.
Pogrebin, M.R., Poole, E.D., & Regoli, R.M. (1984). Constructing and implementing a model juvenile diversion program. Youth & Society, 15(3), 305-
324.
Polk, K. (1989). Juvenile diversion: A look at the record. Crime &
Delinquency, 30(4), 648-859.
Potter, R. & Kakar, S. (2002). The diversion decision-making process from the juvenile court practitioners’ perspective: Results of a survey. Journal of contemporary criminal justice, 18(1), 20-36.
President’s Commission on Law Enforcement and Administration of
Justice Task Force Report: Juvenile Delinquency and Youth Crime (1967).
Washington D.C.: U.S. Government Printing Office.
Prinz, R. J. & Kerns, S. E. (2003). Early substance use by juvenile offenders. Child Psychiatry and Human Development, 33(4), 263-277.
Quay, H. and Love, C. (1977). The effects of a juvenile diversion program on rearrests. Criminal Justice and Behavior, 4, 377-396.
Rausch, S. (1983). Court processing versus diversion of status offenders:
A test of deterrence and labeling theories. Journal of Research in Crime and
Delinquency, 20, 39-54.
Roberts, A.R. (2004). The emergence of the juvenile court and probation services. In A.R. Roberts (Ed.). Juvenile justice sourcebook. pp. 163-182. New
York, NY: Oxford University Press
149
Rojek, D.G., & Erickson, M.L. (1982). Reforming the juvenile justice system: The diversion of status offenders. Law and Society Review, 16(2), 241-
264.
Rose, C., Glaser, B., Calhoun, G., & Bates, J. (2004). Assessing the parents of juvenile offenders: A preliminary validation study of the Juvenile
Offender Parent Questionnaire. Journal of Child and Family Behavior Therapy,
26(1), 25-43.
Schur, E. (1971) Labeling deviant behavior. New York: Harper & Row.
Schur, E. (1973). Radical non-intervention: Rethinking the delinquency problem. Englewood Cliffs: Prentice-Hall.
Scofield, M. G. & Davidson II, W. S. (1995). A community perspective:
Organizational and management implications to creating alternatives to juvenile court. In B. Glick and A.P. Goldstein (Eds.). Managing delinquency programs that work. pp 53-75. Laurel, MD: American Correctional Association.
Shelden, R. (1989). “Do status offenders get worse? Some clarifications on the question of escalation.” Crime & Delinquency 35, 202-216.
Sherman, L. W., Gottfredson, D., MacKenzie, D., Eck, J., Reuter, P., &
Bushway, S. (1998). Preventing crime: What works, what doesn’t, what’s promising: A report to the United States Congress. Washington, D.C.: National
Institute of Justice.
Snyder, H. (2002, November). Juvenile Arrests 2000. Office of Juvenile
Justice and Delinquency Prevention Juvenile Justice Bulletin, Washington D.C.:
U.S. Department of Justice, Office of Justice Programs.
150
Snyder, H. (2006, December). Juvenile arrests 2004. Office of Juvenile
Justice and Delinquency Prevention Juvenile Justice Bulletin, Washington DC:
U.S. Department of Justice, Office of Justice Programs.
Snyder, H. and Sickmund, M. (1999). Juvenile offenders and victims: 1999
national report, Washington D.C.: Office of Juvenile Justice and Delinquency
Prevention.
Snyder, H., Puzzanchera, C., Kang, W. (2003) "Easy Access to FBI Arrest
Statistics 1994-2001" Online. Retrieved August 4, 2005 from,
http://ojjdp.ncjrs.org/ojstatbb/ezaucr/
Soares, R. (2004). Development, crime and punishment: Accounting for
the international differences in crime rates. Journal of Development Economics,
73, 155-184.
Sprott, J.B., Jenkins, J.M., & Doob, A.N. (2005). The importance of school:
Protecting at-risk youth from early offending. Youth Violence and Juvenile
Justice, 3(1), 59-77.
Stahl, A. Kang, W., and Wilt, D. (2003). "Easy Access to State and County
Juvenile Court Case Counts, 2000". Retrieved on August 4, 2005 from,
http://www.ojjdp.ncjrs.org/ojstatbb/ezaco/
Sullivan, C., Veysey, B., Hamilton, Z., & Grillo, M. (2007, October).
Reducing out-of-community placement and recidivism: Diversion of delinquent
youth with mental health and substance use problems from the justice system.
151
International Journal of Offender Therapy and Comparative Criminology, 51(5),
555-577.
Tannenbaum, F. (1938). Crime and the community. New York: Ginn and
Co.
Taxman, F., Perdoni, M., Harrison, L. (April 2007). Drug treatment services for adult offenders: The state of the state. Journal of Substance Abuse
Treatment. 32(3), 239-254.
Thornton, W., Barrett, E. & Musolf, L. (1972). The Sacramento county probation department 601 diversion project. Sacramento: Sacramento County
Probation Department.
U.S. Census Bureau State and County QuickFacts. Retrieved on
November 16, 2007 from http://quickfacts.census.gov/qfd/states/39/3923380lk.html.
Venezia, P. & Anthony, D. (1978). A program level evaluation of
Wisconsin’s Youth Service Bureaus. Tucson: Associates for Youth Development.
Whitaker, J.M., Severy, M.J., & Morton, D.S. (1984). A comprehensive community-based youth diversion program. Child Welfare, 63(2), 175-181.
White, R. & Haines, F. (2000). Crime and Criminology (2nd ed.). South
Melbourne: Oxford University Press.
Whitehead, J. T. & Lab, S. P. (2001). Juvenile justice: An introduction
(3rd.). Cincinnati, OH: Anderson Publishing Company.
Zhang, L. & Zhang, S. (2004). Reintegrative shaming and predatory delinquency. Journal of Research in Crime and Delinquency, 41(4), 433-453.
152
Zhang, S. X. (1995). Measuring shaming in an ethnic context. British
Journal of Criminology, 35, 248-62.
Zimring, F. (2004). An American travesty: Legal responses to adolescent sexual offending
153