Boot Camps for Juvenile Offenders: an Implementation Evaluation of Three Demonstration. Programs Series: NIJ Research Report Published: January 1996 122 pages 265,840 bytes Boot Camps for Juvenile Offenders: An Implementation Evaluation of Three Demonstration Programs A Final Summary Report Presented to The National Institute of Justice By: Blair B. Bourque, Roberta C. Cronin, Frank R. Pearson, Daniel B. Felker, Mei Han, and Sarah M. Hill Supported under Award #92-DD-CX-K043 from the National Institute of Justice, Office of Justice Programs, U.S. Department of Justice. Points of view in this document are those of the authors and do not necessarily represent the official position of the U.S. Department of Justice. ----------------------------------- Table of Contents Chapter 1 Introduction The Boot Camp Movement Why Boot Camps for Juveniles? The OJJDP Demonstration Program The Boot Camp Evaluation The Evaluation Report Chapter 2 The Design and Initial Implementation of the Boot Camps The Boot Camp Program Design The Three Prototype Juvenile Boot Camps Organizational Structure and Management Program Design Chapter 3 Boot Camp Selection Phase OJJDP's Guidelines for Selection Selection Criteria at the Program Sites Selection Process Chapter 4 Boot Camp: The Intensive Training Phase Boot Camp Philosophy and Design Program Capacity and Utilization Facilities Management and Staffing Program Content During Boot Camp Participant Misconduct and Discipline Boot Camp Graduation and Termination Rates Attitudes of Youths Toward the Boot Camp Regimens Implementation Issues Chapter 5 Aftercare Programs Parameters Set by OJJDP Local Aftercare Program Designs The Aftercare Programs in Operation Mechanisms for Monitoring and Disciplining Youths Implementation Issues Chapter 6 Changes During the Boot Camp Phase Evidence of Change During the Boot Camp Phase Evidence of Change During the Aftercare Phase Program Costs Chapter 7 Findings, Conclusions, and Recommendations Key Implementation Findings Conclusions Recommendations Bibliography ------------------------------------------ Chapter 1 Introduction One of the most closely watched developments in corrections over the last few years has been the proliferation of boot camp programs. Characterized by a strong emphasis on military structure, drill, and discipline, these programs offer a new twist on the use of residential programs for convicted criminals. A movement that began with a single 50- bed program in Georgia in 1983 (Parent, 1989), boot camps now operate in more than half the States. In the fall of 1990, the Office of Juvenile Justice and Delinquency Prevention (OJJDP) in the U.S. Department of Justice launched a demonstration program to examine the feasibility, appropriateness, and promise of the boot camp model for juvenile offenders. In 1991, three sites were chosen to participate in the demonstration. At the same time, the American Institutes for Research (AIR) and the Institute for Criminological Research (ICR) at Rutgers University were selected by the National Institute of Justice (NIJ) to conduct an evaluation of these demonstration programs. This final report on the AIR/ICR evaluation presents observations of the first 18 months of the demonstration period. Detailed descriptions of the programs at each site, including the assumptions, rationales, and contexts that determined how each site went about developing a program are presented. Comparing the major components of the camps, the report discusses how well the programs succeeded in the short term-- during the boot camp as well as the subsequent aftercare program. The report ends with an analysis of the promise of boot camps as an effective and appropriate intermediate sanction for juveniles and general considerations for improving their structure and processes. The Boot Camp Movement The correctional system is often characterized as a system in crisis (Byrne, Lurigio, and Petersilia, 1992). In a little over a decade the adult prisoner population has grown a remarkable 150 percent (U.S. General Accounting Office, 1993). Both the rate of prison incarceration and the absolute number of inmates reached their highest levels ever in 1992 (American Correctional Association, 1993). The surge in inmate populations in part reflects a widespread disillusionment with the potential for rehabilitating offenders in the community, a disillusionment that began to take hold in the 1970's (Martinson, 1974; Palmer, 1992). Throughout much of the 1980's, public concern with safety and with giving offenders their "just deserts," not rehabilitation, dominated sentencing policy. To many, incarceration seemed the most natural way to punish offenders and protect society. By 1990, however, the effects of this policy were becoming apparent. Prisons were straining at the seams and correctional costs were spiraling higher, yet crime was not diminishing. Meanwhile, new studies were causing some policymakers and practitioners to reappraise the potential of rehabilitative approaches (Palmer, 1992). In this climate there emerged a vigorous interest in "intermediate sanctions" that could satisfy the public's demand for protection and punishment without further taxing correctional facilities and budgets. Intermediate sanctions are generally defined as sentencing alternatives that fall between the poles of simple probation and incarceration (Morris and Tonry, 1990; Toby, 1982, 1984). The term embraces a wide variety of correctional approaches, including intensive supervision, fines, restitution, and community service, often used in combination. Most intermediate sanctions claim multiple goals: saving money, deterring crime, protecting the public, and rehabilitating offenders (Byrne et al., 1992). Boot camps are a relatively new addition to the intermediate sanctions menu. Billed as "one of the most recent and exciting forms of intermediate sanctions being adopted by the States" (Office of Justice Programs, 1990), they consist of a relatively short period of incarceration in a quasi-military environment, followed by a period of supervision in the community. The first camp, the "Special Alternative Incarceration" program, opened in Georgia in 1983, an outgrowth of discussions between the Commissioner of the Georgia Department of Corrections and a local judge; both were dissatisfied with sentencing options then available (Parent, 1989). Oklahoma and Mississippi soon followed suit, opening paramilitary camps for adult offenders. By 1993, there were at least 59 State and 10 local boot camps for adults operating in 29 States (Cronin, 1994). In addition to their considerable popularity within the correctional system, boot camps have demonstrated extraordinary appeal to the general public. Experts on boot camp programming nationwide note that boot camps are a "natural" for media coverage, which tends to focus on the programs' disciplinary aspects and appeals to "get tough" sentiments. In a culture where many people view military service as a formative experience, the public also seems to intuitively grasp the rehabilitative rationale for the programs (Parent, 1989; MacKenzie and Parent, 1992). Why Boot Camps for Juveniles? The notion of developing boot camps for juvenile offenders evolved logically as the juvenile system increasingly faced pressures similar to those forcing change in the adult system. Although the juvenile justice system nationwide is not as severely burdened as the adult system, it exhibits the same general trends. Juveniles in custody for delinquent offenses increased 35 percent from 1978 to 1989, a period when the youth population of the United States declined by 11 percent (OJJDP, 1992). Thus, satisfactory alternatives to long- term institutionalization are as welcome in the juvenile system as they are in the adult system. Furthermore, age groups served by the juvenile and adult systems overlap. The age of criminal adulthood is 16 or less in some States, and in most States certain categories of juvenile offenders can be tried and sentenced in the adult system. Moreover, a significant proportion of the adult criminal justice system's clientele consists of persons in their teens and early 20's. This is especially true of adult boot camps, which disproportionately recruit offenders who are young and fit enough to tolerate their programs' physical regimen. Thus, if boot camps hold promise as effective interventions for young adults, it is plausible to think that they might be appropriate interventions for older juveniles as well. In some respects, the harsh image of a boot camp regimen appears at odds with a juvenile justice system that, at least in theory, tends to emphasize "rehabilitation" over punishment or public safety. Since the Illinois statute establishing the first juvenile court in 1899, the juvenile justice system in the United States has been influenced strongly by a rehabilitative philosophy (Schlossman, 1983). Delinquents were considered to have lacked proper parental guidance, which the juvenile court judge, acting in loco parentis, would attempt to supply. The juvenile system, including probation officers attached to the juvenile court and what were initially called "training schools," was intended to remedy parental deficiencies in upbringing, the presumed reason for bad behavior in adolescence. Notwithstanding this ideological focus on rehabilitation, the juvenile justice system also has traditionally included discipline. The original training schools were often run in a military style. Custodial institutions for juveniles were often constructed with bars and cells that seemed punitive to the casual observer. The closest approximation to current U.S. boot camp programs--British detention centers--were developed to deal with adolescents rather than with adult offenders. Faced with an upsurge of adolescent crime following World War II, quasi- military "detention centers" were set up in England and Wales under the hypothesis that "a short, sharp shock" to adolescents early in their criminal careers might nip their anti-social tendencies in the bud. When evaluative research produced disappointing recidivism rates for the detention center youths, plans were announced in 1979 to establish two even tougher detention centers: . . . life will be conducted at a brisk tempo. Much greater emphasis will be put on hard and constructive activities, on discipline and tidiness, on self-respect and respect for those in authority. We will introduce on a regular basis drill, parades and inspections. Offenders will have to earn their limited privileges by good behavior . . . (Thornton, et al., 1984). Delinquents were not the only groups singled out for these quasi-military approaches. Military schools have served for many years as an acceptable solution to schooling middle-class adolescents, particularly those seemingly on a troubled path. In any case, the U.S. juvenile justice system has come under increasing attack for its perceived laxity in treating youth and its tendency to emphasize rehabilitation over other correctional goals. Boot camps, in contrast to many other alternatives, offer a particularly attractive package--the chance to pursue rehabilitative goals in an environment that does not appear to coddle delinquents. The OJJDP Demonstration Program In this context, OJJDP initiated a demonstration program to explore whether boot camp programs for adult offenders could be adapted for male juvenile offenders. The demonstration would provide a means to "develop, test, and disseminate information on a prototypical juvenile boot camp as an intermediate sanction program" (OJJDP, 1990). In particular, these programs would identify adaptations to the adult boot camp model to allow boot camps to operate within State or local juvenile justice systems and to make their programming age-appropriate. Despite the growing popularity of boot camp programs in the adult correctional system, the concept had not taken hold in the juvenile system until OJJDP's demonstration program was under way (Toby and Pearson, 1992). In September 1991, OJJDP awarded grants to three groups to develop boot camp programs. Applicants from Cleveland, Ohio; Mobile, Alabama; and Denver, Colorado, were selected through a competitive process. They each received an initial award for 18 months with the potential for a continuance award for an additional 18 months. Grantees spent the first 6 months of the demonstration designing the intervention and then opened their boot camps in March or April 1992. The AIR/ICR evaluation focuses on the demonstration experience during this 18-month period. The Boot Camp Evaluation The OJJDP boot camp demonstration program included a plan for a 2-year evaluation to be sponsored by the National Institute of Justice (NIJ). The evaluation was awarded at about the same time as demonstration sites were selected, in October 1991. During the first year of the evaluation, the Institute for Criminological Research at Rutgers University was the prime grantee, with the American Institutes for Research as its subcontractor. In the second year, the relationship was reversed. The scope of the evaluation of the juvenile boot camps narrowed over the course of the study. Initially, the plan was to conduct both a process and an impact evaluation. However, by the second year it became apparent that sufficient resources would not be available to conduct the impact evaluation, nor to include all of the process measurements envisioned in the evaluation's original design. As a result, the design was scaled back to focus only on process issues. The core of the assessment was a management information system that captured administrative data as offenders progressed through the demonstration program. o At intake, grantees were asked to collect information on participants' demographic characteristics, delinquent history, details of instant (current) offense, disposition of current case, educational history, employment status, living arrangements, and criminality of family members and friends. o At the beginning and end of the 90-day term, boot camp staff were asked to rate participants' performance on a number of dimensions. o At the end of the 90-day term, boot camp staff were asked to report participants' date of graduation or reasons for leaving the program prematurely, any serious infractions of the rules, any honors or awards, and any special services received. o Five months after boot camp graduation, grantees were asked to provide an update on participants' status, including whether they were still participating in boot camp aftercare and, if not, the reason for nonparticipation; their residence and educational placement during aftercare; and any serious disciplinary infractions. At this point, boot camp participants who were still active in the program were asked to rate their experiences, as well as changes in their attitudes, values, behavior, and expectations. o Finally, at the end of the demonstration program in 1993, grantees were asked to provide a status report on all first-year platoons, indicating which youths had completed both boot camp and aftercare, as well as reasons for noncompletion. The grantees designated various staff to assist in obtaining this information and researchers worked with them to fill in gaps and resolve any inconsistencies among data sources. In addition, some sites were able to supplement the basic management information with data they had collected on physical fitness, educational performance, and participants' attitudes. Data collection covered all participants admitted to boot camp in the first year of the program--119 in Cleveland, 76 in Denver, and 122 in Mobile. To supplement these data sources, researchers made an average of five site visits to each program, spoke with participants, and interviewed program staff and representatives of the agencies that constituted each program. Researchers also kept in frequent contact with key program staff by telephone throughout the demonstration. The Evaluation Report The remainder of this report presents the results of the process evaluation of the three boot camp demonstration programs. Chapter 2 describes the basic program design in more detail and provides an overview of the individual programs. Chapter 3 presents each site's selection process (the procedures each program used to choose candidates for participation in the program), the characteristics of program participants, and the characteristics of those who failed to complete the program successfully. Detailed descriptions of the programs' boot camp phase are presented in chapter 4. A discussion of the aftercare programs in each site appears in chapter 5. Chapter 6 offers an analysis of observable changes in boot camp participants during the program period and the program completion and attrition rates. Chapter 7 presents conclusions and considerations for the design of boot camps for juveniles and provides suggestions for further research. Chapter 2. The Design and Initial Implementation of the Boot Camps Chapter 2 describes the overall design of the OJJDP boot camp demonstration program for juvenile offenders and provides a brief overview of the three programs developed within that framework. Chapter 2 also discusses issues that arose during the early stages, as the sites planned their programs and served the first few cohorts of youths. The Boot Camp Program Design The general framework for demonstration programs was set forth in OJJDP's program announcement, which appeared in the Federal Register on July 12, 1990. The purpose of the initiative was to develop and evaluate a juvenile boot camp program. Similar to a typical boot camp program for adults, it would serve adjudicated, nonviolent offenders and emphasize discipline and work. It would also emphasize "treatment," a core concern of virtually all correctional programming for juvenile offenders and a key element in some adult boot camps. OJJDP's agenda for the demonstration programs was ambitious. The programs would: o Serve as a cost-effective sanction. o Promote basic, traditional, moral values and instill a work ethic. o Provide discipline to youths through physical conditioning and teamwork. o Promote literacy and increase academic achievement. o Include activities and resources to reduce drug and alcohol abuse. o Encourage participants to become productive, law-abiding citizens. o Ensure that juvenile offenders are punished and held accountable. OJJDP emphasized that the demonstration programs were to constitute an intermediate sanction (i.e., punishment less severe than long-term institutionalization, but more severe than immediate supervised release). Although the programs were to ensure public safety, they were not to be used for violent or serious repeat offenders deserving long-term institutionalization (Federal Register, 1990). Any public or private organization was eligible to submit a bid for funding, but preference was given to proposals that involved collaborative efforts between public and private agencies and to consortia involving service agencies and local juvenile justice systems. Funds were to be awarded in two 18-month periods. The first award would cover a 6-month planning effort and a 12-month implementation period. The second award would support program operations and allow programs to develop a training manual and a package of technical assistance materials that would explain the methods and approaches used to implement the programs. The program announcement called for applicants to design programs with four phases: selection, intensive training, preparedness, and accountability. Applicants were free to set their own standards for the duration of the program, except that the intensive training or boot camp phase was expected to last at least 90 days. Selection During the first phase, selection, grantees would identify a sufficient number of prospective program participants and establish their eligibility. According to OJJDP, to participate in the demonstration program a youth must: o Be adjudicated by the juvenile court and be awaiting implementation of the court disposition. o Be categorized as at high risk of continuing delinquency or of abusing drugs or alcohol. o Be under the age of 18. o Have no history of mental illness. o Not be considered violent or have a history of involvement in violent crimes. o Not be an escape risk. o Demonstrate motivation to participate in the program. OJJDP offered no specific definition or methodology for determining which prospective participants were violent or an escape risk or for measuring their motivation to participate. Grant applicants were to apply their own criteria for selecting program participants. As part of selection, each grantee would be required to randomly assign eligible youth to either the demonstration program or to a control group, making it possible to conduct a rigorous impact evaluation. Intensive Training This phase of the program, what is usually referred to by the term "boot camp," was to involve rigorous physical conditioning, discipline, activities to instill confidence, self-esteem, teamwork and leadership, and it was to last no less than 90 days. Programs were expected to adapt some portion of a military-style system of dress, drills, courtesy, discipline, and physical training. They were also expected to conduct comprehensive diagnostic assessments, provide participants academic training, teach employment skills, provide participants drug and alcohol assessments and counseling, and provide them other kinds of counseling during the intensive training phase of the program. Programs were given considerable flexibility in the implementation of intensive training. Indeed, as will be discussed in chapter 4, each program also varied in the degree to which it adopted military procedures. The programs also varied the proportion of time devoted to physical training, work, drill and ceremony, education, and other support services. Preparedness During the third phase, preparedness, OJJDP specified that the programs return the participants to a community setting with intensive supervision and aftercare. Each participant was to have an individualized plan and be accountable for accomplishing it. The program staff, with the help of agencies and volunteers in the community, were to provide the various support services that the young offenders need to resist negative influences in their communities, find work, and complete academic programs. The program and agency staff were directed to conduct drug testing, prevention/treatment counseling, and physical conditioning and provide support to participants' families. Participants failing to pursue academic and vocational training or employment, and those not participating in community services and treatment, would be terminated from the program pending a review by the presiding court. Accountability OJJDP envisioned a final phase of aftercare in which youths were to be held accountable for their behavior through a systematic restitution and community service plan. During this stage the services and activities of the preparedness phase would be continued. However, youths would be subject to less intensive monitoring and supervision, including reduced aftercare time commitments. The Underlying Rationale Although each grantee was expected to develop a specific program within the context of local policy, preferences, and resource constraints, the OJJDP program announcement delineated general requirements for development of the programs. The expectation was that the programs would marshal resources--people, physical facilities, counseling and educational materials, agreements with social and community agencies--to systematically implement programming that would positively affect participants' short- and long-term attitudes and behavior. A distinguishing characteristic was the expectation that the intensive training phase of activities would reflect and be conducted in a distinctively military-like setting. Embedded in boot camp theory are the twin themes of discipline and development. The implicit hypothesis is that external discipline fosters the self-discipline needed to engage in, and benefit from, program treatment and development components. The hypothesized logic chain begins with uniformed drill instructors carefully selected and trained to give offenders strict military-like discipline and supervision. Drill instructors who brook no nonsense from program participants and consistently enforce prescribed manners of behavior are instrumental in establishing the boot camp's structured environment. Program participants exposed to this environment will learn self-discipline, resistance to peer pressure, and commitment to traditional values. Over time, these personal changes will enhance feelings of self-worth and reinforce socially desirable behaviors such as seeking and holding jobs, continuing education, and resisting criminal activities. Within the program's disciplined structure, participants are better able to focus on and actively engage in educational (remedial or otherwise) courses, life skills training, and counseling. These, too, enhance feelings of worth and accomplishment, instill an appreciation and propensity for positive behavior, and improve chances for resisting damaging peer pressure and criminal behavior. The Three Prototype Juvenile Boot Camps OJJDP awarded grants to three teams of public and private agencies to develop prototype demonstration programs for juvenile offenders. These teams were: o The Cuyahoga County Court of Common Pleas, Cleveland, Ohio, in association with the North American Family Institute. o The Boys and Girls Clubs of Greater Mobile, Alabama, in association with the Strickland Youth Center of Mobile County Juvenile Court and the University of South Alabama. o The Colorado Division of Youth Services, Denver, Colorado, in association with New Pride, Inc. All three teams cited similar systemic problems driving their decisions to bid. These problems-- rapidly rising numbers of juvenile arrests, increased involvement of youth with drugs, overcrowded juvenile facilities, and the expense of institutionalizing youth--mirrored problems found in most of the country's juvenile justice systems. The demonstration programs were viewed as a possible means to alleviate overcrowding and still provide innovative services. The primary goal of Cleveland's demonstration program was to alleviate severe overcrowding in juvenile correctional facilities while providing adequate supervision and treatment. Although Ohio's Juvenile Code encourages supervision and rehabilitation in a family environment, separating the child from his parents only when necessary for his welfare or in the interests of public safety (Kurtz and Giannelli, 1989), in 1991, the State had the second largest number of children in public custody in the Nation (Bureau of Justice Statistics, 1994). Reducing this population was a major concern in Ohio. By decreasing time in custody, boot camps would lower institutional care costs. Newspaper expos‚s on overcrowding and the resulting diminished ability to provide rehabilitation or recruit enough qualified trained staff increased the State's interest in boot camps and other intermediate sanctions. The Cuyahoga County Juvenile Court was also eager to have correctional alternatives for the increasing numbers of offenders they confronted, particularly alternatives that showed promise in reducing recidivism. Similarly, the Denver program was developed during a period when Colorado's juvenile correctional facilities were under extreme pressure and facing critical public scrutiny. At the time the program developed its proposal to OJJDP, the State's juvenile training schools were estimated to be operating at 120 percent of capacity. Officials were concerned that the practice of diverting juvenile delinquents from secure residential correctional facilities was undermining local officials' efforts to sanction delinquent behavior. They were also concerned that the public believed that youths were being returned to the community without any punishment for their crimes. Interest in the Mobile program stemmed from the increasing number of juveniles arrested in Alabama each year for engaging in some form of criminal behavior. Juvenile court and law enforcement officials in Mobile believed that these youths would have a chance of changing their lives for the better if they could be temporarily removed from their home environment and exposed to a more positive lifestyle. Moreover, elected officials and civic leaders in the area believed that community services and juvenile justice resources could be meshed to meet the needs of troubled youth. The demonstration program seemed a plausible way of accomplishing this goal. Additionally, the Strickland Youth Center had experimented with a 2-week "mini" boot camp in 1991 to explore the feasibility of the model. Organizational Structure and Management As was suggested in the OJJDP program announcement, all three programs included both public and private sector agencies. Public sector involvement was critical in resolving issues that involved the court or probation and correctional agencies. Private sector involvement added skills, resources, and, in some cases, facilities to the program. Together, the partners offered each other a sounding board on policies and procedures and a synergy that proved quite useful to a high- visibility program. In Cleveland, the Cuyahoga County Juvenile Court contracted with Massachusetts-based North American Family Institute (NAFI) to run its program's boot camp and aftercare phases. The juvenile court coordinated all program activities, conducted planning, research and evaluation, and selected participants. The court paid one manager out of grant funds to oversee the selection process. NAFI was responsible for operating both the boot camp and aftercare demonstration program, training staff, and providing technical assistance. Similarly, a public agency, the Colorado Division of Youth Services (DYS), was the prime contractor in Denver. DYS contracted with New Pride, Inc., a private, nonprofit corporation with a long history of providing community-based services to high-risk delinquents, to operate the program. DYS provided two client managers and a project coordinator to oversee program operations, handle selection, and administer case management for boot camp participants throughout their sojourn in the program. New Pride operated both the program's boot camp and aftercare phases. In Mobile, the demonstration program was a partnership of the Strickland Youth Center of the Mobile County Court, the Boys and Girls Club of Greater Mobile (B&GC), and the University of South Alabama. Unlike in Cleveland and Denver, in Mobile the private partner (B&GC) was the prime contractor. However, Mobile County's chief probation officer and the executive director of B&GC served as co-project directors. They jointly supervised the director of the program, who in turn managed both the boot camp and aftercare phases. Strickland Youth Center managed the intake and selection process and dedicated three full- time probation officers to the program. These officers provided case management supervision for youths in boot camp and aftercare. Program Design In designing their programs, the three sites adhered to the broad program parameters set forth in the program announcement. Each of the programs considered selection to be the first stage of the process, screening youths against a set of preordained criteria. Each program implemented a 90-day boot camp on the premises of an existing youth correctional facility. The central feature of their camps was the use of quasi-military structure, discipline, customs, and activities. Each program also developed an aftercare program, although none chose to split the program into the specific preparedness and accountability phases that OJJDP had envisioned. Both Cleveland and Denver did, however, divide aftercare into phases that differed in intensity of surveillance and services. Overall, the sites bought into a general philosophy regarding boot camps that would permeate both their boot camp and aftercare programs: o The Cuyahoga County Juvenile Court's partner, NAFI, employed the "normative model," which attempts to move youths who have gravitated toward antisocial normative systems back toward healthy functioning. NAFI incorporated only those aspects of military boot camps that were compatible with a safe, structured environment--physical exercise, military customs and courtesies, and group punishment. The program's cornerstone was techniques, such as guided group interaction, that NAFI had employed in other institutions. The military veneer provided the program with straightforward rules, standards, and expectations. NAFI's philosophy permeated both the boot camp and aftercare programs; the military aspects were limited for the most part to the residential boot camp. o The Colorado Division of Youth Services and New Pride saw the boot camp program as an opportunity to diverge dramatically from their existing treatment modalities. By applying a military model that deliberately made boot camp a high-stress experience, they would set high standards for behavior and hold youths accountable to those standards. Unlike counselors in other programs, strict drill instructors would intentionally play a tough role. The Denver program rested on a theory that youths ultimately would perceive the DI's as positive role models and internalize their standards of conduct. Once youths were released from the boot camp, however, the strict military environment would be replaced by a "prep school" that required youths to wear coats and ties and that employed an entirely educational emphasis. o In Mobile, the Strickland Youth Center supported the discipline, regimentation, and emphasis on traditional values inherent in the boot camp model. Mobile planned to employ military components to gain the attention of youths and to increase their focus on activities and goals. Disciplined behavior and exposure to DI role models would ultimately improve the self- discipline and concentration of youths, who would then be more receptive to individualized education and other rehabilitative components. Youths would continue to draw upon their improved self- discipline once they were released to the tutoring and recreational programs of the Boys and Girls Club. In short, all three programs saw the military model as an avenue for creating a secure, structured environment that would permit youths to make other changes. All three programs also viewed drill instructors as positive role models for youths to emulate. Denver and Mobile depended on the external discipline of the military to produce internal controls. Cleveland aimed to improve internal controls via a peer culture that established a positive normative system in each platoon. This system would continue to be nourished during the aftercare period. The next three chapters describe in detail the program implemented at each site. Chapter 3 discusses the selection process and criteria employed by the sites, as well as the types of youths selected for participation. Chapter 4 discusses the three boot camp programs and variations in the degree of militarism they employed, and chapter 5 explores the three starkly different aftercare models implemented by the programs. Chapter 3. Boot Camp Selection Phase According to the framework outlined in the previous chapter, the first phase of the program was participant selection. This chapter describes the screening criteria developed by the programs, the selection processes they employed, and the characteristics of the youths these criteria and processes identified during the first year of operation (April 1992 through March 1993). Because the demonstration programs were conceived to be intermediate sanctions, they were designed to serve midlevel offenders--youths whose criminal behavior warranted some intervention more restrictive than probation but not serious enough to prompt long-term institutionalization. Defining these offenders was a critical question for the programs, in terms of both public support and ability to work successfully with youths. Drawing the line too close to the probation end of the spectrum would raise questions about fairness in sentencing for youths whose criminal records were normally minor enough to warrant a lesser sentence. Net-widening--selecting youths who would otherwise have been placed in less expensive, nonresidential placements--also would reduce the program's potential for lowering costs. On the other hand, if the programs were used simply as an alternative to longer term institutional placements, it would be more difficult to make the case that the programs punished youths and held them accountable for their behaviors. Questions might be raised about whether the programs were shrinking the net by reducing the amount of time youths were incarcerated. A perception that boot camps were diversion programs, providing a less restrictive alternative for youths, might well erode some of the public support enjoyed by the programs. In fact, the three demonstration programs came to different decisions about which youths to define as midlevel offenders. That decisionmaking is reflected in the type of youths who participated in the boot camps. OJJDP's Guidelines for Selection In its program announcement, OJJDP described the appropriate boot camp target population as "adjudicated, nonviolent, juvenile offenders who are under 18 years of age" (Federal Register, 1990). Violence was to be determined on the basis of offense-based criteria, including homicide, rape, armed robbery, aggravated assault, and arson (Summary of Solicitation Workshop Proceedings, 1990). In addition, the programs were to exclude "serious, habitual offenders who ordinarily would be assigned to a correctional institution" (Summary of Solicitation Workshop Proceedings, 1990). However, the target population was to "have a history of offending with prior attempts at intervention that appear to have failed" (Federal Register, 1990), placing them at high risk of further involvement in the juvenile or criminal justice systems. The program announcement and backup application materials did not set objective criteria for "serious, habitual offenders," nor did they clarify whether the programs were to exclude all offenders normally sentenced to correctional facilities, or only violent and habitual offenders who would be so sentenced. In the absence of precise definitions, the programs were able to interpret the guidelines to fit a broad range of offenders. Except for the dictate that candidates have an adjudication and have failed to benefit from previous interventions, the criteria served to exclude offenders who were too serious or violent rather than to define youths who had committed offenses serious enough to warrant inclusion in the target population. Public safety was cited as a reason for excluding violent youths, because the boot camps were designed as minimum security institutions. The ambiguity in threshold criteria indicated that there was less concern about whether boot camps would result in more youths in residential placement than about whether serious and violent offenders would be released prematurely into the community. In addition to exclusion on the basis of youths' criminal records (i.e., serious and violent offenses), programs were to screen out candidates who had histories of mental illness and who were considered escape risks. Programs could elect to serve all juveniles, or they could focus on one of two groups: youths 15 years old and younger or youths 15 to 17 years old. All candidates were to "demonstrate motivation to participate in the program" (Federal Register, 1990). Because they were funded under OJJDP's program development process, the demonstration programs also had research requirements for testing their effects. The guidelines required that youths be randomly assigned to experimental or control conditions. This requirement meant that the programs had to screen and select twice the number of youths they needed to fill their available slots. Selection Criteria at the Program Sites Within the broad screening parameters outlined by OJJDP, the three programs defined their own criteria and procedures for identifying candidates. Only males were eligible for participation. All three programs chose to serve the broadest age range stipulated by OJJDP, rather than focusing only on those above or below age 15. Both Cleveland and Denver set 14 as their minimum age limit, while Mobile selected from among 13-year- olds. As its maximum age limit, Mobile employed the statutory age of juvenile court jurisdiction, 17. In Denver and Cleveland, 18-year-olds could be admitted if they were 17 at the time of referral. Both Cleveland and Mobile met the nonviolent criterion by excluding offenders with offenses on a proscribed list. In Cleveland, this list included murder, manslaughter, rape and other sex offenses, arson, kidnapping, and corruption of a minor. Mobile excluded these same offenders and also those whose offenses were less serious, i.e., assaults 1 and 2, burglary 1, robbery, and shooting into an occupied building or vehicle. Although Denver did not have a set list of exclusionary violent or sex offenses, in practice their offense criteria were similar to Cleveland's. In Mobile and Denver, the exclusions applied to all offenses on a youth's record; in Cleveland they applied only to current offenses. Cleveland did, however, eliminate youths from consideration if they were considered serious, habitual offenders. None of the three sites set lower boundary offense criteria. In all three sites, however, the expectation was that the programs would not serve youths who were before the court for the first time. In fact, Cleveland's criteria implied that only misdemeanor and felony offenses would be eligible. Mobile targeted youths who had failed on probation and whose probation officer recommended them for the program. Denver targeted youths who were committed to the State Department of Youth Services and were awaiting placement at its classification center. Denver also targeted youths under probation supervision by the local juvenile court. Only Cleveland selected participants entirely from youths who would otherwise have had a residential placement either at a Ohio Division of Youth Services (ODYS) facility or at the county-run Youth Development Center (YDC), neither of which accepts status offenders. Thus, on the selection discussed previously, Mobile's criteria fell closest to the net-widening end, admitting youths who might otherwise receive a probation sentence. Denver's criteria defined a midpoint and included both probationers and youths committed to State custody. Cleveland fell closest to the other end of the spectrum, selecting from among the institution-bound only. The special demands of boot camp dictated stringent medical and psychological screening. All three programs required thorough medical screenings to ensure that youths would be able to participate in rigorous physical conditioning. They also attempted to screen out youths with psychological problems that might reduce their tolerance level for a highly structured military- style setting. Youths with serious drug and alcohol addictions (Cleveland specified only those serious enough to require detoxification) were excluded for similar reasons. Because the boot camps would not be located in high-security facilities, youths who were considered escape risks were ruled out as candidates. Finally, Mobile added an intelligence criterion requiring candidates to have a minimum IQ of 60. Although voluntary participation is typical of many adult boot camps, Cleveland was the only site requiring voluntary agreement. Initially, there was concern in Cleveland that too few youths would volunteer for the program because the alternative at ODYS or YDC would only be a few months longer than the "tough" boot camp. Whereas offenders volunteer for boot camp in the adult system because their prison sentence could be reduced by several years, the attraction is less obvious in the juvenile system, where sentence reduction amounts to no more than a few months. Despite these fears, Cleveland remained committed to voluntary participation, believing it to be a powerful motivational tool. Indeed, the concerns proved fruitless, as almost all of the youths offered an opportunity to enter the candidate pool did, in fact, volunteer. Cleveland's voluntary participation requirement went further than OJJDP's stipulation that youths "show a motivation to participate." Denver took a youth's feelings about boot camp into consideration in selection decisions, but it did not necessarily exclude an unwilling youth from eligibility on that basis. In Mobile, there was no voluntary aspect to the assignment process. In part, this policy stemmed from the fact that judges perceived a voluntary requirement as a limit to their judicial discretion. It also stemmed from concerns--similar to those in Cleveland--that the program would not be able to attract enough volunteers. Selection Process The selection process was tailored to the processes already in place in the three juvenile justice systems. Some features, however, were shared across programs. Each camp entered youths in "platoons" or "flights" of 7 to 13. The capacity at both Cleveland and Mobile allowed 10 to 12 youths to enter the program each month. Denver's capacity was smaller, permitting 12 youths to enter the camp every 6 weeks. Each program followed random assignment procedures, requiring the identification of double the number of youths required for an experimental intervention. In addition, the programs first employed less expensive screening strategies, such as reviews of official records, reserving the more expensive medical exams until the pool of candidates had been winnowed to a more manageable level. Following are summaries of the selection processes used in Cleveland, Denver, and Mobile. Cleveland After a juvenile court judge had committed a youth to ODYS or to YDC, a special boot camp program manager conducted a preliminary screening for age and medical history and obtained a release of information. All youths with ODYS or YDC commitments entered the detention center, regardless of whether they were in secure detention during the pretrial stage. At this juncture the judge could override a youth's entry into the candidate pool and send him directly to ODYS or YDC, or a youth could fall out of the pool because he qualified for community corrections. Next, the program manager reviewed the criminal, social, and medical history of youths remaining in the pool and met with each candidate to explain the boot camp program and to obtain a written voluntary participation statement. The program manager also administered mental health and substance abuse screens. Finally, the physician at the detention center conducted a physical exam. All youths who survived this process were then submitted in pairs to the research team for random assignment. The entire screening process took no more than 2 days. Youths who were assigned to the control conditions were generally transported to one of the nine ODYS institutions or to YDC within a few hours. Boot camp youths were held in the detention center or, in a few cases, released to a shelter care facility to await the assembly of the next platoon of youths at the first of the month. During this time the program manager conducted both a parent and a youth orientation to boot camp. Denver In Denver, there were two tracks for entry into boot camp: the Department of Youth Services (DYS) track, accounting for about 56 percent of the candidate pool, and the probation track, accounting for the remainder. The latter involved a less systematic process. The DYS candidate pool consisted of all youths housed at the DYS selection and classification center at that time. Two DYS case managers assigned to the program screened suitable candidates with the assistance of center staff and DYS field officers. The screeners focused on youths with risk scores that qualified them for a medium security facility or community placement. The routine classification process, which involves medical, clinical, educational, and drug and alcohol evaluations at the center, produced most of the information needed to determine whether a youth qualified for the candidate pool. The manager also met with a parent to obtain additional medical information. Finally, case managers met with youths to orient them to the program and to determine their interest in the program. The case managers made the final decision as to whether the youths entered the boot camp pool, and they presented pairs of youths to the research team for random assignment. The other pool of candidates came from individual probation officers in Denver, who could make referrals to the DYS case managers. Youths could be referred before or after sentencing or when they had violated conditions of their probation. Because youths who entered the pool via probation avenues were not part of the DYS system, the DYS case managers did not have easy access to them or to their records, except for court actions recorded in Colorado's statewide computer system. Case managers collected background data from the probation officers and interviewed most youths over the telephone. Youths were required to have medical screenings. Case managers then made final admission decisions on the basis of this information and relayed pairs of youths for random assignment. For youths who had already been sentenced, the judge generally had paved the way for random assignment by specifying boot camp and an alternative placement. Youths already sentenced or who faced a violation of probation at the time of random assignment had to appear before a judge to finalize the disposition. The distinction between youths who entered the program via probation and those who entered via DYS was maintained in the aftercare period, when DYS youths were supervised by DYS case managers and probation youths were assigned to probation officers. Mobile After adjudication, probation officers screened youths to determine if they met the age and offense criteria specified by the program. They also made recommendations to the court regarding which youths should be considered for admission. After this initial screening, the judge ordered youths into the candidate pool or made an alternative disposition. Youths who entered the pool underwent a series of tests: a suicide risk assessment, an escape risk assessment, a drug and alcohol assessment, a psychological examination, and, if the youths did not have an intelligence score on record, an IQ test. Candidates were required to arrange for a medical exam by their own doctor or were provided a medical examination by a local clinic. Finally, the EYC director reviewed the candidate pool and could override assignment decisions. This final review was subject to an appeal by the probation officer to a committee consisting of the boot camp director, the probation officer, the Strickland Youth Center psychometrist, the boot camp drug and alcohol assessor, a boot camp life skills coordinator, and a probation supervisor. Youths selected for the final pool were submitted in pairs for random assignment. Pretransfer Holding youths in detention or at a central classification center simplified the screening process because youths were more accessible and because information is routinely collected at those detention sites. Of the youths who entered the three boot camps during the first year, at least 74 percent in Denver, and slightly fewer than half in the other two sites, had been placed in secure detention prior to their adjudication. Once selected, however, detention was mandatory in Cleveland. All of the youths were held in secure detention during the period from disposition to transfer, approximately 98 percent of them for the entire time. In both Denver and Mobile the percentages of youths in secure detention during all or part of the selection period were 66 percent and 44 percent, respectively. Although it simplifies the selection process, detaining youths adds to the cost of a program and must be factored into cost-effectiveness equations. Because the programs formed platoons on a monthly basis in both Cleveland and Mobile, and every 6 weeks in Denver, it was expected that at least 1 or 2 weeks would elapse between random assignment and intake into the boot camp. Mobile had the shortest time period between assignment and transfer, with 64 percent of youths waiting no more than 9 days. The abbreviated wait, in conjunction with the low detention rate, meant that the * In Mobile, judges do not sentence youths until after selections have been made for boot camp; it is therefore not known to what option a judge would have sentenced youths admitted to boot camp. ** In Denver, probation may refer youths to the program prior to disposition. Youths then appear for disposition. Those youths assigned to the experimental group would have "boot camp" entered on the record, in essence approving the random assignment. Mobile program involved fewer postcommitment detention days, and, therefore, presumably experienced lower costs than the other two programs. As would be expected with the longer period between formation of platoons, Denver had the longest time periods between disposition and transfer, with almost half of the youths waiting 20 or more days. Cleveland fell in the middle with a median wait of about 2 weeks. In Cleveland, the added time in detention was a problem because of overcrowding in the detention center. Once a youth is committed to ODYS or YDC, he generally is transported immediately, rarely waiting longer than 1 or 2 days. Therefore, a median postcommitment time of 13 days for up to 10 youths each month was an added burden on the detention center. The demonstration program had the opposite effect on Denver's detention situation. According to DYS staff, detained boot camp youths, though spending more days on average in detention than youths at either of the other boot camp programs, were likely still detained a shorter period of time than other committed youths in Denver. In fact, Colorado's youth service system is so crowded, and committed youths there wait so long for residential placements, that the shorter preboot camp detention periods may have been an incentive for youths to seek the boot camp alternative. Characteristics of Youths Screened into Boot Camp The programs began selecting youths to participate in boot camp in March 1992, and the first cohort entered facilities in April. During the first year of operation, Cleveland accepted 119 youths, Denver accepted 76, and Mobile accepted 122. The social, educational, and delinquent characteristics of these youths are presented below. The programs' screening criteria yielded an all- male population of predominantly older youths. Over three-quarters of the youths in both Cleveland and Denver were at least 16 years old. Mobile drew a slightly younger recruit class, with an average age of 15.6, compared to 16.5 in both Cleveland and Denver. Mobile was the only program to accept 13-year-olds, and it selected 7 percent of its recruits from this group. Denver set a screening minimum of age 14, but in practice screened out most 14-year-olds because staff believed that programming was not appropriate for such a broad age range (14- to 18-year-olds). About 80 percent of recruits in Cleveland and 64 percent of recruits in Mobile were African American. Except for a few Hispanic youths in Cleveland, the remainder of recruits in both programs were white. Denver's boot camp population was more varied, with 35 percent African American, 35 percent white, 22 percent Hispanic, and 8 percent Native American and other. Criminal Records. Consistent with the program's target criteria, most boot camp youths were not violent or serious, habitual offenders. However, screening rules did vary across sites. Cleveland selected the most serious group of offenders, Mobile selected the least serious group, and Denver fell in the middle. Felony referrals predominate at all three sites, with 72.3 percent of youth in Cleveland, 50.7 in Denver, and 49.2 percent in Mobile entering boot camp on felony charges. These typically were property offenses in Denver and Mobile and, to a lesser extent, in Cleveland. Cleveland differed from the other two sites in that a number of youths entered the program on drug felonies and on felonies against persons. All three programs took in a substantial proportion of probation violators, with no additional offense adjudicated concurrently. About 23 percent of youths in Cleveland, 29 percent in Denver, and 38 percent in Mobile were selected for boot camp on the basis of probation violations. Although the majority of the offenses in all three programs were property crimes, the type of crime differed across the three sites. More than a quarter of instant offenses in Mobile were thefts, but in the other sites thefts represented much smaller proportions. On the other hand, Mobile had fewer auto thefts than Denver or Cleveland. (It is not known how many auto thefts were reported in Cleveland because this offense is charged there as "receiving stolen property" and categorized as "other property.") Cleveland had the highest percentage of youths brought in on robberies (13 percent) and also had a substantial number of drug offenses, the bulk of which were drug trafficking rather than possession or use. The vast majority of instant offenses in Cleveland and Mobile were felonies, accounting for 93.5 percent and 78.9 percent, respectively, of cases reported at those sites. Denver had a much higher percentage of youths brought in on misdemeanor charges (37.7 percent). Three additional indicators of the seriousness of boot camp participants' instant offenses are the degree of injury to the victim, the amount of loss associated with the offense, and whether a gun was involved. As would be expected from the program criteria, most instant offenses did not involve physical injury to the victim. Victims in a few cases required treatment by a doctor or hospitalization, but for the most part injuries were minor. About 15 percent of cases in Cleveland involved use of a gun, however, compared with 7 percent in Denver and Mobile. Property damage or loss was prevalent in Mobile and Denver, where over half of the offenses involved some losses. A relatively small proportion of cases (26.3 percent in Denver, 19.7 percent in Mobile, and 9.9 percent in Cleveland) involved losses of $600 or more. Prior to their instant offense, most boot camp youths had accumulated a delinquency record and were under supervision by the juvenile court. Because Mobile intentionally selected youths who had violated probation, it is not surprising that 73.8 percent of the program's recruits were on probation or parole upon arrest for the boot camp offense. A similar proportion of youths in Cleveland and a slightly lower percentage of youths in Denver were on probation or parole. Youths in the "other" status category had escaped from an institution or otherwise were under some form of court supervision. Only 16.9 percent of the boot camp youths in Cleveland, 9.3 percent in Denver, and 10.7 percent in Mobile were not involved in the juvenile system when they were arrested for their current offense. With respect to prior findings, Cleveland's youths had accumulated the most extensive delinquency records by the time they committed their instant offenses. About 19 percent of the Cleveland youths had three or more felony findings excluding the instant offense and 41 percent of them had two or more felony findings. In Denver and Mobile only about 20 percent of the youths had two or more prior felony findings. Only 24 percent of the Cleveland youths had no prior felony finding before the offense which stimulated program entry, in contrast to 49 percent of the Denver youths and 43 percent of the Mobile youths. Looking at their entire criminal histories at the point of boot camp entry, the differences between youths in Cleveland and those in the other two programs are even more apparent. Ninety-seven percent of Cleveland youths had at least one felony charge on their records. The mean number of felony charges on their records was 3.0, dropping only slightly, to 2.8, for mean felony findings. Denver and Mobile's youths had on average at least one fewer felony finding and fewer felony charges. About 24 percent of youths in Denver and 21 percent in Mobile did not have any felony charges but all youths in Cleveland and all but one in Denver had at least one felony or misdemeanor charge upon entry to boot camp. In Mobile, however, 10 youths whose only delinquent offense was a violation of a court order for a status offense entered the program. On average, youths in Denver initiated their criminal careers at later ages than youths in Mobile and Cleveland. The mean age of first criminal referral ranged from 13.7 years in Mobile to 14.9 years in Denver. These differences may be more apparent than real, however. In Denver, many youths had prior run-ins with Colorado's municipal courts, which handle violations and some misdemeanors. The official State delinquency records for these youths commenced when they contacted a county-level court. The researchers did not have systematic access to municipal court information. Not only had youths in Cleveland accumulated more serious criminal records than youths in Mobile and Denver, but 27 percent had a prior commitment to either the State or county correctional system. Fewer than 5 percent of youths in either Mobile or Denver had State commitments, but both programs had youths who had been confined for less restrictive sentences. In Denver, 45 percent had received a "probation plus detention" sentence prior to boot camp, and in Mobile, 16 percent had served time in Camp Basic, a 2-week precursor to the boot camp program. Only about 16 percent of youths in Mobile and Cleveland and 26 percent in Denver had not received at least a formal probation sentence prior to boot camp. In summary, most youths selected for boot camp had a prior sentence at least as serious as probation, and they were either on probation/parole or had some other system involvement prior to their entry into boot camp. The instant offense was typically a property, drug, or "other" felony involving no injury and relatively small monetary loss. Some youths entered boot camp on a violation of a court order or probation with no other new offense; all of the youths entering on violations had prior criminal offenses, except for 10 youths in Mobile who had been placed on probation for status offenses. According to any of the criteria discussed, recruits in Cleveland had considerably more serious criminal records than youths in either Mobile or Denver. Family and Social Characteristics. The characteristics of the boot camp youths' families reveal a disturbing picture of youths living in broken families that in many cases were already known to the court system. At the time of referral to boot camp, no more than 30 percent of youths in the three programs lived with both parents or a parent and a stepparent. In the two sites where data was available--Cleveland and Mobile--58 percent and 46 percent, respectively, of participants' families were on some form of public assistance. About one-third or more of youths in all three programs had one or more siblings with official delinquency records, and sizeable proportions of the youths' parents were known to have either criminal records or to have been referred for child neglect and abuse. The numbers are slightly lower in Denver, and lower still in Mobile, but Cleveland made a point of collecting this information for use in assessing their candidates' risk, while the other sites did not. The main point is that many of these youths came from families already ensconced in the criminal justice system. Reviews of youths' social background records show that most boot camp youths hang out with peers who also are delinquent. Almost all youths in Cleveland and Mobile and about 78 percent in Denver reported having delinquent friends. Discipline problems at home or at school also were prevalent among these youths, particularly in Cleveland and Denver. Almost half of the youths in Denver had "major" drug or alcohol use problems. Major drug problems were reported less frequently for youths in Cleveland and Mobile, possibly because more youths with such problems were screened out. Gang problems surfaced most frequently among youths in Denver, with 42 percent suspected of major gang involvement. About 30 percent of youths in Cleveland and 26 percent in Mobile had strong gang links. At the point of boot camp entry, the vast majority of youths in Cleveland and Denver were either not enrolled in school or were enrolled but attending very sporadically. Proportionately fewer of the youths in Mobile exhibited school dropout or severe attendance patterns, perhaps because they were somewhat younger. In addition to data from social records, in Cleveland self-report data were available from a youth survey administered at the detention center before selection for the program. Because these surveys were administered by a boot camp staff member at a time when the youths may have been trying to cast themselves in the best light possible to gain entry into the boot camp program, limited significance can be given to the results. However, the surveys do provide some insight into how the youths perceived themselves, or how they may have wanted to portray themselves. Many survey items asked youths to indicate their agreement or disagreement with statements on a 5- point Likert scale. More than 50 percent of the youths reported a "positive attitude" toward themselves on four of the five statements. However, fewer than half were satisfied with themselves. A number of youths reported difficulty controlling their tempers and saying things without stopping to think. On the other hand, 92 percent responded in the direction indicating internal locus of control by agreeing that a person can pretty well make whatever he wants out of his life. Most youths perceived that they had some control over their destiny. Responses to the statements concerning peer relationships were generally positive. Youths generally felt close to their friends and believed they fit in well with them. They also agreed that it is easy to have friends if you try to be friendly. Boot camp youths also perceived that their peers generally disapprove of criminal behavior, particularly if it involves breaking into a building or using force. However, fewer than half reported that their friends would disapprove of their hitting or threatening someone or of their selling hard drugs. A percentage of youths reported that they had engaged in various delinquent and other negative behaviors in the 3 months prior to boot camp admission. Generally, the more serious the crime, the fewer the number of youth admitting having committed it. More than half the youths reported that they had been rowdy in a public place, hit or threatened to hit other students in the last 3 months, or taken part in a fight where a group of their friends were pitted against another group. On the other hand, none reported that he had raped someone, and few reported using force to rob students or adults. An interesting finding from the survey was the frequency of reported sexual activity. All but 5 percent of the youths reported having had sex in the 3 months prior to admittance to the camps, and 45 percent reported having had sex on 10 or more occasions. The majority reported using alcohol, and about 38 percent reported using marijuana. None of the youths reported using drugs such as crack, cocaine, or heroin. Taken at face value, the surveys described a pool of youths with fairly good self-esteem and strong peer relationships, most of whom did not admit to criminal activity of a serious nature during the 3 months prior to detention. However, a number of the youths did admit to minor criminal behavior and fighting among peers, as well as alcohol use. The vast majority also claimed frequent sexual activity. Problems In the Selection Process Three types of selection problems emerged as the programs implemented their designs: screening procedures inadequate to detect all medical, behavioral, and psychological problems; uncertainty about the appropriateness of the screening criteria, particularly with regard to age and criminal history; and problems attaining adequate cohort sizes. (A fourth problem, deviations from random assignment, is described in Appendix A.) Inability of Screening Procedures To Detect Problems Soon after program implementation, some flaws in the screening process became apparent. Almost immediately, program staff found that some medical problems had eluded detection in the screening stage. This problem caused concern not only because physical training was an integral part of the program, but also because the program could potentially be held liable in the case of injury. In Cleveland, the program supplemented the detention physical with a more thorough physical by the physician at the Youth Development Center. In Denver, case managers attempted to supplement medical information on participants' health problems with medical histories obtained from parents. Behavioral problems that did not respond to the boot camp environment also became apparent fairly early in the program. In response, the programs tightened screening by reviewing the characteristics of some of the early program failures and alerting the screeners to those problems. Mobile took a second look at its screen for psychosis after some particularly bizarre animalistic behavior on the part of one youth. Uncertainty About the Appropriateness of the Screening Criteria Both Cleveland and Denver expressed some dissatisfaction with the blanket exclusion of youths who had committed "violent" crimes, and no site adhered to this requirement 100 percent. In Denver, staff observed that most youths in the juvenile system had some history of violence, in part because gang involvement was so prevalent, even though this violent experience might not appear on their official records. In any case, staff felt that youths presumed to have a violent history did no worse than others in boot camp, and some performed extremely well. Denver did not relax the criterion until the end of the second year, and then only occasionally. In actuality, however, the programs included offenders with violent instant offenses, including youths who had committed aggravated assaults or robberies. Fifteen percent of the instant offenses in Cleveland included use of a gun. Cleveland took youths who had committed violent offenses as long as those offenses were not their instant offense. Another area of uncertainty was deciding what ages were appropriate for the programs. As noted above, none of the programs elected to serve a more narrow age population as suggested by OJJDP, but chose instead to serve all juveniles ages 13 to 17 (in Mobile) or 14 to 17 (in Cleveland and Denver). Denver, after a couple of months of experience, avoided 14-year-olds and instead sought older youths believed to be more socially mature. On the other hand, older youths were a problem because many of them were inappropriate for regular school placements, and because aftercare programs had been designed with the assumption that youths would be supervised at school during the day. All three programs had a number of 18-year-old youths in boot camp who turned 18 after the offense was filed (date age calculated for screening purposes), but before transport to boot camp. Problems Attaining Adequate Caseload Sizes Except for some initial tinkering with the timing of selection for a new cohort, neither Cleveland nor Mobile had any major difficulties obtaining enough cases, even though the research design required them to identify two youths for every program slot. It is notable that in Cleveland participation was adequate despite the fact that the youths had to volunteer for the program. In Denver, however, the program was plagued by a lack of participants, and it never reached capacity. Certainly the random assignment procedures exacerbated this problem. Overall, except for the random assignment requirements, the programs appeared satisfied with the selection criteria and process. Although Denver and Cleveland believed they could handle youthful violent offenders in their programs, staff admitted that it was difficult to predict which types of youth would succeed or fail in the boot camp environment. Chapter 5 attempts to answer that question empirically. Chapter 4. Boot Camp: The Intensive Training Phase A 90-day period of residence in a boot camp, known as the "intensive training phase," was the heart of the demonstration program. In this chapter, the key features of the boot camp phase are described. Unless otherwise noted, descriptions are based on the first year of boot camp operations, from April 1992 through March 1993. Boot Camp Philosophy and Design According to OJJDP's program announcement, the intensive training phase was expected to offer a crime- and drug-free environment in which to change the behavior and attitudes of participants. To achieve these changes, the program would expose youths to military-like routine, drills and discipline, and rigorous physical conditioning. The program would also include rehabilitative components such as education, counseling, development of work- and life-coping skills, drug and alcohol abuse programming, and family involvement. Upon graduation from boot camp, each youth would return to the community with an individualized plan for continuing and reinforcing the progress he had made (Federal Register, 1990). Although OJJDP dictated the framework for the boot camp phase, it was up to the demonstration sites to flesh out a program that incorporated the required elements, was philosophically coherent, and was appropriate for juveniles. Much of the initial framework was completed during the proposal development stage. Because none of the grantees had previous experience with boot camp programs for juveniles or adults, all looked closely at the available literature and visited at least one boot camp for adults. The grantees also struggled to translate the broad vision of a "character-building" program into a plan for day- to-day operations. As a result, each boot camp developed a rationale and perspective that was distinctive. o Cleveland's program designers at juvenile court steered away from what they perceived to be the negative, degrading, and excessively punitive aspects of some boot camps. Instead, they elected the treatment approach, augmented by an atmosphere of military discipline, that the North American Family Institute (NAFI) had implemented in other locations. NAFI's approach, based on the "normative model," assumes that individuals assimilate group norms in order to belong. Because delinquents have gravitated toward antisocial normative systems, treatment must move them back toward healthy, prosocial values and functioning. NAFI aims to create a safe, comfortable environment in which staff can set clear expectations, establish a positive peer culture, and persuade youths to accept different values and behaviors. A counseling technique called guided group interaction is a crucial part of the NAFI approach. For Cleveland's program, NAFI would incorporate only those aspects of military boot camps compatible with the normative model. In general, this meant regimentation and structure, physical exercise, and military customs and courtesies--elements that would create a safe, secure environment for participants. o The Denver grantees, the Colorado Division of Youth Services and New Pride, Inc., had lengthy experience in providing a wide variety of programs for delinquent juveniles in residential and nonresidential settings. They saw the boot camp phase as a chance to depart radically from previous treatment models--that is, to commit fully to the military model. Although their boot camp would include some traditional rehabilitative programming, that would not be its primary emphasis. Thus, the staff would deliberately try to make boot camp a high-stress experience in which profound life changes could take root. Drill instructors (DI's) would be expected to play the "tough" role, unlike counselors that youths might encounter in other types of programs. Strict military-style discipline would teach socially acceptable behavior and the consequences of deviance. Meanwhile, DI's would provide the positive role models of leadership and integrity that many youths were lacking. Ultimately, as in Cleveland, the hope was that youths would come to identify and internalize these positive standards of conduct. Program staff also felt that the toughness of the program would provide another benefit: Unlike the usual DYS parolee, a youth who graduated from boot camp would have an accomplishment to be proud of. In short, Denver's boot camp would equip its graduates with the self- esteem, self-discipline, and positive values to profit from the educational and treatment programs available during aftercare. o In Mobile, the military model was familiar and readily acceptable to program planners and the wider community. As home to a number of military bases and to many military careerists, veterans, and retirees, Alabama is generally sympathetic to program models that stress discipline and regimentation and uphold traditional values. Mobile, in addition, had experimented with a 2- week boot camp prior to receiving the OJJDP demonstration grant. The driving concept behind Mobile's demonstration program was that unfocused and inattentive youths cannot benefit from rehabilitative and treatment regimens. The military model was viewed as an effective means to gain the attention of participants and get them to focus on specific activities and aims. In time, disciplined behavior and exposure to DI role models would improve self-discipline and the ability to concentrate. Youths then would be ready to respond to individualized education, challenging activities requiring teamwork, and counseling. Moreover, part of the boot camp experience would include an opportunity to learn respect for the environment and take part in outdoor activities to help maintain it--an emphasis giving rise to the program's name, the Environmental Youth Corps. In all three programs, these basic orientations remained constant throughout the demonstration and accounted for some differences in program emphasis. As discussed below, despite differences in emphasis, the boot camps shared many features. In chapter 5, the authors illustrate how there was much more cross-site variation in aftercare. Program Capacity and Utilization Program capacity, which ranged from 24 participants in Denver to 30 in Cleveland and 52 in Mobile, was dictated mainly by the amount of the OJJDP award and the size of the building(s) available to the boot camps. The boot camps were designed to admit "platoons"-- groups of youths who would enter boot camp together and graduate together. To fully use its facility capacity, Cleveland intended to start a new platoon of 10 youths every month. Similarly, Denver planned on a platoon of 12 every 6 weeks, and Mobile planned on a platoon of 13 every 3 weeks. Although Cleveland was generally able to meet its targets, platoons in Denver and Mobile were smaller than expected, averaging 10 participants each. Also, the period between platoons in Denver and Mobile was, on average, 7 to 10 days longer than expected. In any case, the Mobile facility would have been extremely cramped if it had admitted platoons according to the original plan. In fact, during the second year of Mobile's program, official capacity was limited to 32 by order of the fire marshal. All three programs accepted their initial cohort in early to mid-April 1992, about 6 months after the OJJDP award. Cleveland and Mobile admitted 12 platoons during the first full year, and Denver took in eight. Facilities All three programs modified existing facilities to serve as boot camps. The Cleveland program, called Camp Roulston, used two cottages at a county- operated residential center for delinquents, situated about 30 miles outside the city. Denver's Camp Foxfire occupied a building on the grounds of a DYS youth services center on the outskirts of Denver. The Mobile program used a separate building and two trailers at the county court's Strickland Youth Center. Although each program had its own building or buildings, the location enabled the demonstration programs to share some facilities and services with other programs, such as classrooms in Cleveland and medical services, cafeteria services, and recreational areas at all three sites. Through constant supervision and careful scheduling, the programs avoided commingling boot camp participants with youths in other programs. At times, however, boot camp participants were within sight and sound of others. Staff reported that participants occasionally were taunted by youths from other programs, but generally this kind of harassment was minor. In fact, in Cleveland, youths from other programs looked favorably upon boot camp and even requested that some of its procedures be adapted for them. Similarly, in Denver, after adjacent youths had complained that the boot camp's early morning exercise drills were waking them up, the camp director noticed that some of these youths had begun running in the morning. The boot camps depended on staff supervision, rather than fences, for security. There were no fences in Cleveland, and in Denver and Mobile, portions of the boot camp grounds were fenced but program activities were conducted beyond those boundaries. Management and Staffing At each site, the boot camp was operated by the private nongovernmental partner in the demonstration. This was the North American Family Institute (NAFI) in Cleveland, New Pride, Inc., in Denver, and the Boys and Girls Clubs of Greater Mobile in Mobile. The Cleveland boot camp was staffed entirely by NAFI employees. Although private employees predominated at both the Denver and Mobile boot camps, these camps also included staff employed by their public agency partners, the Colorado Division of Youth Services (DYS) and the County Court, respectively. The staffing level at each site included not only personnel who were stationed at the boot camp and had day-to-day responsibility for the program there; each boot camp received additional administrative and management support, as well as assistance with diagnostic assessments, from the agencies participating in the demonstration. Cleveland had the most generous staffing support, with 22 staff members for the intensive training phase, including three teachers who were provided under subcontract. This level of staffing amounted to one staff person for every 1.4 beds. Denver had 12 staff and Mobile had 21, translating into one staff member for every two beds in Denver and one for every 2.4 beds in Mobile. In all three boot camps, drill instructors (DI's) made up the majority of the staff, and it was in this position that differences in staffing across sites were most apparent. Cleveland had more DI's than Mobile and Denver put together. The latter sites used security guards to supplement or replace the DI force after lights out, while Cleveland used part-time DI's to relieve full- timers when necessary. As a result, Denver normally assigned only one DI to each of the day shifts. At night, one DI slept in the facility while a security guard kept watch. Mobile averaged three DI's per day shift and two DI's or security guards at night. Cleveland normally had five DI's on the day shifts and four at night. In contrast to Cleveland, both Denver and Mobile struggled to cover all shifts while allowing sufficient sick leave and vacation time for DI's. The need to take pressure off the DI staff was one reason why these two sites sometimes stretched out the interval between new platoons. (Inability to find sufficient qualified applicants was another reason, especially in Denver.) As a result of the slower than intended pace of intake in Denver and Mobile, as well as early terminations from all three programs, facilities were never filled to capacity. As a result, ratios of participants to total staff were variable and consistently lower than ratios of beds to staff. Staffing levels for positions other than drill instructors were more consistent across sites. Mobile had two to three client managers and Denver had two, compared to Cleveland's one. However, the functions of Cleveland's client manager were narrower. Called a human resources manager, he was primarily responsible for youths currently in the camp and was employed by NAFI. In contrast, client managers in both Denver and Mobile had broader functions and were government employees. In Denver, the client managers worked for DYS and had responsibility for screening boot camp candidates and managing boot camp graduates, as well as working with current boot camp residents. Mobile's client managers, who were county probation officers, had similar responsibilities for youths in and out of the camp. At all three sites, the client managers were assigned exclusively to the demonstration programs. Consistent with a strong emphasis on educational progress during boot camp, Mobile employed four educational staff to deliver academic and life skills curriculums, in contrast to Denver's two. Cleveland contracted with an alternative school, obtaining three teachers for its education program. This circumvented the need to hire teachers directly, an action which had threatened to antagonize the local teacher's union. Overall supervision of each boot camp was provided by an onsite director, employed by the private partner in the demonstration. In Denver and Mobile, the director did not supervise the client managers at boot camp because they were government employees and under the supervision of their respective agencies. The Denver director's duties were limited to the boot camp, but directors in Mobile and Cleveland had broader responsibilities. Mobile's director was also responsible for supervising the demonstration's aftercare coordinator. In Cleveland, responsibility for directing boot camp and aftercare was combined in a single position, with the director maintaining offices at both the boot camp and aftercare locations. Cleveland had begun operations with separate directors for boot camp and aftercare but combined these functions within a few months under a central director to provide more continuity of programming. DI Qualifications and Training At all three sites, the positions of drill instructor and camp director were new to the juvenile correctional system. Each site wanted DI's who were able to relate to youths, but in other respects they took differing approaches. Cleveland deliberately chose a mix of staff representing both military and counseling backgrounds "to provide checks and balances." Whatever their backgrounds, staff had to be flexible enough to adapt to the broader program demands. Denver and Mobile, on the other hand, specifically chose people with military experience. (Only one DI in Denver did not have a military or ROTC background.) They wanted people who could "play the role" of a DI, offering a model of good physical condition and consistent, fair leadership. In fact, Denver specifically avoided hiring DI's with counseling or therapeutic backgrounds for fear that they would have difficulty adhering to the military model. The camp commander in Denver was ex-military and also had law enforcement experience. In Cleveland, neither of the first two directors had military experience. One had a background in child care and the other had a background in corrections. The third director had served in the military, but his primary work experience was in corrections. In Mobile, the initial director was retired military with experience in adult correctional boot camps. The second director was a social scientist and evaluation specialist, and the third director was retired military with expertise in basic skills education. Each site took a distinctive approach to training its boot camp team. In Cleveland, the original staff received several days of training from NAFI before the boot camp opened. NAFI also offered 8 hours per month of inservice training, and senior staff received additional training at NAFI headquarters in Massachusetts. Replacement staff received a 3-day orientation at another NAFI facility, as well as onsite orientation and one- on-one monitoring and support. Denver sent its commander and initial DI team for a 3-week stint at Quantico, Virginia, where they attended Marine DI training specially designed for correctional officers. This session was supplemented by some local training on the juvenile justice system. One DI was trained on the job, having come on board after the initial training. Mobile's initial team underwent 3 weeks of intensive, onsite training designed by the camp director. Subsequent hires were trained on the job. Staff Turnover Staff turnover at the three boot camps varied dramatically. During the first full year of boot camp operations, the period in which systematic data were collected for participating youths, both Cleveland and Mobile changed boot camp directors. In Cleveland, the change resulted from philosophical differences between NAFI and the original director. In Mobile, local policymakers were unhappy with what they saw as an overly militaristic approach by the first director. He was replaced by a social scientist from the University of South Alabama who had been involved in designing the original program. Cleveland and Mobile also experienced relatively high turnover in other positions. Some staff were dismissed for disobeying program rules. Others quit, often for better pay. In Mobile, some DI's also left because they were unhappy with the change of directors, and two other staff were laid off because of downsizing necessitated by a pared- down continuation award from OJJDP. One of Mobile's DI's was let go after a child molestation charge came to light; although the charge had nothing to do with his boot camp performance, it would have disqualified him for hiring if known at the time of his job application. By the end of the first year at Mobile's boot camp, only three of the nine original DI's and one of the five original teachers remained. In contrast, Denver's turnover was low in both the first and second years of operations. The original camp director remained with the boot camp for the first 18 months, with the senior drill instructor stepping in for the final 6 months. Otherwise, the program lost just two staff. In the first year a DI was fired for not adhering to the program's disciplinary guidelines, and in the second year one teacher was terminated as a result of cuts in OJJDP funding. Denver staff attribute the low turnover among DI's in part to competitive salaries and a tight job market resulting from the closing of a local military base. Program Content During Boot Camp All three boot camps provided recruits a mixture of regimentation and discipline plus rehabilitative programming in a spartan environment. Physical Environment Mobile housed its recruits in barracks with bunk beds, and Denver assigned its junior, or newest, platoon to beds in the day room. All youths in Cleveland's camp had individual rooms, as did members of Denver's senior platoon. Despite the differences in sleeping arrangements, boot camp settings were spartan at all locations. Facilities contained limited, extremely functional furniture, and sleeping rooms were devoid of personal touches. Youths stored uniforms and their few belongings in foot lockers or a single bureau. Televisions and radios were generally off limits, unless granted as a special privilege. Visiting privileges were strictly limited, and home visits were not allowed. Telephone access also was severely restricted. Even at the site with the most generous telephone policy, youths could call home only once a week. Military Regimentation and Discipline Following through on the military theme, all programs accorded a central role to the drill instructor. Drill instructors were the primary supervisors of boot camp participants and were responsible for keeping them on a demanding, tightly controlled schedule that provided little or no personal time. Drill instructors used military titles (e.g., sergeant, lieutenant), and DI's and youths wore military-style uniforms. Platoons earned hats or special insignia to differentiate their rank in the program. DI's also trained youths to use military customs and courtesies ("Yes, Sir," "Recruit Jones requests permission to...," etc.), accustomed them to standing at attention, and taught them to march in drill formation. There were routine inspections of living quarters and personal hygiene, and mail was monitored. In addition to enforcing discipline, these techniques served to prevent contraband from getting into the facilities and to discourage gang insignia, such as self-administered tattoos. All programs were harshest or most restrictive in the first days and weeks while military rules and protocol were being learned. Programs were split into three 30-day periods, with attainment of the next phase signified by a change in shirt color or hat, as well as increased privileges. Participants could earn the right to visitation or phone calls, for example. Senior recruits were expected to be models for newer recruits. Like their military counterparts, these boot camps ended with a public graduation ceremony attended by parents and friends of the participants as well as local dignitaries, including judges, the local chief of police, and others. Typically, the ceremonies featured marching drills and accorded some special recognition to superior performers. A youth might serve as platoon leader or flag- bearer, for example, or make a brief speech. Despite these similarities, there were variations in the way the military model was implemented at the three sites. Verbal intimidation and yelling were a standard part of the DI style in Denver and Mobile. In fact, both Denver and Mobile used a high-confrontation intake ritual, complete with military haircuts, to establish control initially. Cleveland's intake was briefer and involved less verbal intimidation, and staff there did not yell or use "in-your-face" techniques. In contrast to Denver and Mobile, Cleveland also lacked a "brig" or isolation room for miscreants and placed less emphasis on drilling. Although none of the programs reproduced the rigors and regimentation of a real military boot camp, the Denver camp came closest. Cleveland fell at the other extreme, with Mobile in the middle. In effect, Cleveland, the site that selected the youths with the most serious records, also had the least rigorous boot camp. Mobile and Denver, on the other hand, selected less serious youths and employed more military emphasis. Emulating military models, all three boot camps required youths to routinely maintain their living areas. This requirement included cleaning and polishing private rooms and common spaces and yard work. In Mobile, some cleaning tasks, such as cleaning the administration or classroom buildings, were treated as a privilege because they involved little supervision. Generally, the boot camp staff saw these work activities as one method of teaching good work habits and daily living skills. Extra work also was used as a punishment at all sites. In any event, the result was that the camps looked extremely clean and orderly. Physical Training Physical fitness and conditioning activities were a daily part of the program at all sites. Youths participated in early morning runs and calisthenics, as well as team sports such as football or basketball. Over time, Mobile shifted from individual conditioning toward more group sports. Both Cleveland and Denver made it a practice to test the fitness levels of new platoons in order to establish benchmarks and to protect against injuries from too intense conditioning. Because of concern about the wide variations in fitness levels, Cleveland did not perform this test until the 1-month mark, and Denver limited the running test to one mile. Challenge or Adventure Programming "Challenge" or "adventure" programs typically involve a series of stressful, physically demanding tasks that require group cooperation and problem solving. The tasks are of graduated difficulty and are carried out under the supervision of specially trained staff. All three sites originally proposed to make this type of programming an integral part of the boot camp experience. The reality was somewhat different. Cleveland and Mobile purchased ROPES courses, which included a standard curriculum, one or more obstacle courses, and instructor training. In Cleveland, construction of the courses was delayed as staff attempted to resolve issues of location with the host facility and to obtain construction funds. An indoor course was ready by the second quarter, but an outdoor course was not available until the second year. Challenge activities mainly took place on weekends. In Mobile, challenge and adventure programming, including outings like overnight camping and canoeing, were intended to be a weekly part of the program. Although the program did purchase the ROPES course and some mountain bikes for use on trails at a Boys and Girls Clubs' campground, these activities were not regularly scheduled. A serious problem, not encountered in Cleveland, was staff turnover among the DI's who had been certified as ROPES trainers. With only one ROPES- certified instructor left, use of Mobile's course eventually dwindled to just 10 days out of the boot camp's 90-day period. Denver never implemented a formal challenge program, but the boot camp did construct and use a military-style obstacle course as part of its physical conditioning activities. Education Each boot camp provided several hours of educational programming on weekdays in compliance with State regulations as well as OJJDP guidelines. The typical youth entered boot camp with a record of school failure; many were far behind grade level and some had already dropped out of school. Thus, all the programs emphasized basic skills and remedial education in reading, writing, and mathematics. The specific arrangements for providing the education program varied. After a few months of using teachers hired part time from the host institution, Cleveland subcontracted with an alternative school to run the educational portion of its program. This arrangement was intended to provide a certified program, while freeing the boot camp of the union strictures that would arise from hiring teachers directly. The Denver and Mobile programs hired their own instructors. Mobile's instructors used a packaged curriculum known as the PACE Learning System, which tests a youth's educational level upon intake and then provides an individualized self- paced program of instruction geared to that level. Teachers worked with students individually as they progressed through the curriculum. Although Cleveland's and Denver's programs did not use self-paced instruction, their instructors gave a good deal of individual attention to students because of the great variation in academic skills. Life Skills In addition to basic education, each boot camp offered a "life skills" curriculum, which might be best described as a hybrid of education and group counseling. Life skills curriculum modules varied somewhat but typically incorporated such topics as drugs and alcohol, AIDS, sexual behavior, risk taking, conflict resolution, values clarification, responsible decisionmaking, and goal setting. The curriculums combined techniques like classroom discussion, self-assessments, group exercises, and written assignments. Cleveland's life skills curriculum, designed by the program director, was delivered by the drill instructors. In Denver and Mobile, instructors were hired to deliver the life skills curriculum. The Denver instructor was one of the boot camp's two certified teachers; because of budget cuts, however, she was laid off at the end of the first year, leaving the remaining teacher to cover both academic and life skills instruction. In Mobile, the life skills teacher (there were several over the course of the program) typically brought some experience in counseling and criminal justice. Mobile had originally planned to integrate the life skills and ROPES curriculums under a single instructor. This arrangement never worked out, however, in part because of personnel problems related to finding and retaining a person with the right combination of skills, and in part because of scheduling difficulties. At a more fundamental level, each camp's entire regimen was designed to improve basic life skills such as grooming and personal hygiene, maintenance of surroundings, comportment, and time management through daily instruction and practice. Substance Abuse Education and Counseling All programs tried to avoid admitting youths in need of drug treatment, although many youths were assumed to be drug-involved at some level. As reported in chapter 3, background records of the youths admitted to boot camp often indicated a problem with drugs or alcohol, especially in Cleveland and Denver. As a result, although none of the boot camps provided a separate and distinct substance abuse education or counseling program, all incorporated relevant materials in their life skills curriculum. In Cleveland, some drug counseling also occurred in the context of its guided group interaction program, described below. Other Counseling or Therapy To some extent, staff in all three camps considered every component of their programs to be a form of counseling. However, only Cleveland offered a formal counseling program beyond what was embedded in the life skills or ROPES curriculums. A major focus of Cleveland's program was guided group interaction, a counseling approach designed to foster a positive peer culture. One-hour sessions led by trained drill instructors were held daily. Following explicit written rules of interaction, youths discussed feelings and problem behaviors and attempted to develop acceptable responses. In addition, DI's, probation officers, client managers, or other staff at each site counseled participants informally from time to time on specific behavior problems or a particular issue that was troubling a youth. Family Involvement In contrast to standard practices at many juvenile institutions, family participation and visitation was not encouraged during the program's boot camp phase. The camps generally permitted family visits only in the latter half of the program, and even then visits were severely limited. In Denver, for instance, parents typically visited just once before graduation. However, parents could call case managers for information about their sons, and case managers consulted them during planning for their sons' release from boot camp. Cleveland also held a formal orientation session for each family as youths entered the camp to establish a relationship with the parents and to inform them about the program. Mobile was unique in that it offered an 8-week parenting class at the boot camp, conducted by the program's senior probation officer. Participation was not required, but a major incentive for parents to attend the class was the opportunity to see, but not speak to, their children. DI's marched the platoons through the classroom at some point during each session. Community Service OJJDP did not require that the intensive training phase include community service, but both Mobile and Cleveland did so anyway. In keeping with the program's name, the Environmental Youth Corps, Mobile's youths participated in park, schoolyard, and beach cleanups. Some of Mobile's work activities, such as helping to build the bike trail at the Boys and Girls Clubs of Greater Mobile's camp, also had a community service element that served the broader community. Cleveland incorporated at least two community service activities during boot camp, including landscaping work and helping to bag food for the homeless. Denver did not include community service among its boot camp activities. Case Management Each boot camp had client managers--a NAFI social worker in Cleveland, two DYS client managers in Denver, and two to three county probation officers in Mobile. In each case, the client manager was responsible for assessing each youth's needs, preparing a workplan for each youth at intake, updating the plan according to his progress and experiences during the boot camp phase, and arranging for his transition to the community. The workplans were designed to be relatively uniform for all participants, but they needed to be much more individualized for the postrelease phase. Some youths might not be able to return to their own homes, for example, and others might require continuing drug counseling. Also, educational goals might vary at this point. Whether the boot camp case manager's responsibility ended at graduation depended on the site. In Mobile, it did not; the probation officers continued as case managers during aftercare. In Cleveland, responsibility shifted to an aftercare case manager. In Denver, the situation depended on the youth's legal status. Case managers retained responsibility for youths who had been committed to DYS but relinquished it to probation officers for youths referred to the program by the juvenile court. Program Duration All three boot camps were designed to last about 90 days. During the first year, the average graduate in Cleveland spent 92 days in boot camp, and in Denver the average stay was 91 days. Time to graduation in Mobile was longer--108 days on average--primarily because the program often held youths back for disciplinary infractions or unsatisfactory progress. However, youths whose performance was exemplary would finish the program in around 90 days. Denver did not use "setbacks"; instead, in rare instances expelled youths were permitted to repeat the entire boot camp program. In exceptional cases, youths also might remain over 90 days in boot camp because case managers needed extra time to find them a place to live in the community. In Denver, this process never involved more than a few days, but in Mobile, at least one youth remained in boot camp for months beyond his scheduled release date because a community placement could not be found. Programming Mix Time was allocated among the main activities at the three camps. Official schedules, which were not always strictly followed and tended to differ at different stages in the programs. Also, program schedules from the three sites did not always use the same terminology or method of breaking down activities. All three programs had 15«- to 16-hour days, beginning with reveille no later than 5 to 6 a.m. and ending with lights out by 9 to 10 p.m. Daytime activities, however, differed somewhat at each site. Looking first at the activities normally associated with military boot camps--military training and drill, work, and physical conditioning--Cleveland devoted 3¬ hours to them each day, compared with 5¬ hours in Mobile and 8 hours in Denver. Conversely, Denver devoted the least time each day to education and life skills programs, 4 hours, while Mobile devoted 7 hours and Cleveland scheduled 5ţ hours. Cleveland also regularly scheduled an hour of counseling beyond what was offered in its life skills curriculum. The camps' program schedules confirmed the researchers' opinion, acquired through observation and interviews, that Denver provided the most militaristic setting and Cleveland the least, and that Mobile placed the heaviest emphasis on educational achievement. Participant Misconduct and Discipline Sanctions All three boot camps had a written policy governing sanctions for participant misconduct. Overall, there were two broad levels of sanctions: o Summary punishment, or sanctions that could be imposed on the spot, at the discretion of the drill instructor. o Sanctions that could be imposed only with the review or approval of some higher authority. Summary punishment or on-the-spot sanctions for misconduct were reserved primarily for less serious breaches of rules. In Denver and Mobile, this form of misconduct might result in a stern reprimand, extra exercise, such as pushups, or extra work detail. In Denver, DI's could place a youth in the brig for up to 24 hours without review, although it is not known how often summary "sentences" as long as 24 hours actually occurred. (In practice, the distinction between summary punishment and reviewed sanctions was blurred in Denver. If the camp commandant was present when brig time was imposed, he could review the punishment and modify it if he saw fit.) Staff concluded that some youths preferred brig time to the rigors of the boot camp regimen. More serious breaches of discipline were referred up the chain of command--to the camp commander in Denver or to the head drill instructor or the camp director in Mobile--for imposition of a sanction. In Mobile, the most serious infractions were heard by a disciplinary board composed of the camp director, the head DI, the academic and the life skills coordinators, and the probation supervisor. In these instances, the disciplinary board heard testimony from witnesses and dismissed the charges or imposed punishment accordingly. Punishments in Denver and Mobile could involve several days in the brig (or isolation room), although most punishments were lighter. In general, this sanctioning system was patterned after military procedures. However, Mobile's disciplinary system, under its first director, was distinctive in that it often used meaningless work as a form of punishment. Cleveland's system of sanctions, adapted from procedures in other nonmilitary programs run by NAFI, was more elaborate than either Denver's or Mobile's. It involved an explicit progression of punishments. At the low end of the spectrum, sanctions ranged from "suggestions" to "directives" (orders) to "confrontation." At the upper end, sanctions included client contracts and physical control or restraint, the latter used when a youth was out of control and perceived to be a danger to himself or others. Intermediate sanctions included separation for counseling, "incentive exercise" (e.g., pushups), and work detail. As in the other two sites, Cleveland's DI's could handle less serious misconduct or misconduct requiring immediate intervention on their own, while a board was convened for the most serious infractions. Minor infractions were often dealt with during the daily guided group interaction session. Youths were rarely observed doing incentive exercise in Cleveland, although elsewhere it was common for DI's to impose extra exercise as punishment. The disciplinary systems all borrowed one important element from the military model. DI's could punish an individual or the entire platoon, even for the misconduct of just one member. One improperly made bed might mean that everyone had to remake their beds, or one youth's misconduct might cost the whole group some privilege. Group rewards and punishments were expected to help produce peer pressure for compliance and instill a sense of individual responsibility to the group. In Mobile, the whole platoon could be put before the disciplinary board. In addition to specifying how and when different levels of sanctions might be applied, all sites had rules setting limits on DI behavior. For example, DI's were forbidden to use vulgar, profane, or humiliating language, including ethnic epithets, and physical contact with recruits was limited to certain prescribed situations. Departures from the approved standards were not observed at any of the camps during onsite visits. However, staff reported that there were lapses, and all three sites dismissed at least one staff person for failing to follow approved disciplinary procedures. In Mobile, for instance, a staff member was disciplined for breaking the rules regarding use of the barrack's TV monitor, which taped the barrack's activities 24 hours a day. In Denver a DI was dismissed for violating personal contact rules after he taped a youth's mouth shut. Setbacks In addition to their regular policies for sanctioning misbehavior, Cleveland and Mobile had a provision for delaying a youth's graduation from boot camp. In both sites, youths earned points daily and platoons earned points weekly that determined their privileges and attainment of levels. In Cleveland, a negative review (or a serious infraction) could cause a youth to be held back an extra 30 days, in effect demoting him to the platoon that entered after his own. This demotion could occur just once, however. Mobile's system was more elaborate, and setbacks were used more frequently. Although the system allowed setbacks up to 30 days, the usual setback was for a week, and a recruit could win back days for good behavior and rejoin his original platoon. Denver had no setback system, although DI's rated each youth weekly, and there was a formal evaluation of behavior every 45 days. The youth, his case manager, the camp commander, and the youth's lead DI all participated in the evaluation meeting. Although Denver did not allow setbacks, three youths (3.9 percent of the camp's first-year participants) repeated the program after having been expelled. Termination The most serious sanction a youth could receive was termination from the program. Termination could occur as a result of a single act, such as escape from the facility or assault on a staff member, or it could occur as a result of a continuing pattern of misconduct. Programs initially did not have explicit policies governing termination, and to a large extent, this remained at the discretion of staff. However, after its first escapes, Cleveland adopted a firm policy that youths who were absent from the camp for over 48 hours would not be permitted to return. Youths absent for over 24 hours would not be readmitted unless they had voluntarily returned to camp. The single infraction most likely to lead to termination in Denver and Mobile was escape. In Cleveland, it was assault against staff, with escape ranking second. At every site, however, some escapees and assaulters were returned to the program. Mobile was the most likely to retain escapees. However, attempted escape in Mobile resulted in an automatic sentence of 72 hours in isolation. Youths might leave boot camp prematurely for other than disciplinary reasons, most notably because of a physical or psychological problem. In theory, youths at all three sites could also quit the programs, although entry to boot camp had been voluntary only in Cleveland. All sites discouraged quitting. For instance, Cleveland youths had to complete a long series of steps in order to quit. In any case, staff reported that in practice, youths typically "quit" by running away or by becoming so uncooperative and disruptive that the boot camp expelled them. In Denver, staff believed that at least three youths faked suicide attempts to prompt their removal from the program. Consequences of Termination Presumably, if the consequences of termination were predictable and adverse, youths would be more likely to comply with the boot camps' regimens. And, indeed, in Cleveland the consequences of termination were predictable and adverse. Youths who were expelled from boot camp were returned to either DYS or the county's residential program to serve their backup sentence, which was always longer than the 90-day boot camp regimen. In Denver and Mobile, the situation was not so clear cut. In Denver, youths who had been committed to DYS were returned to an assessment center for reassessment and reassignment. This process likely involved a waiting period in a detention facility and eventual transfer to another secure residential placement. But youths referred by the Denver probation department and all youths in Mobile were returned to court for imposition of a new penalty. This policy could result in a more severe punishment, but the outcome depended on the probation officer's recommendation and, ultimately, the judge's discretion. Infractions and Sanctions: Experience of the First-Year Platoons As part of the evaluation, systematic information was collected about the infractions committed by youths in boot camp and the sanctions applied. The scope of this research was limited to serious rule infractions, however, because all sites required staff to document them. It should be noted that sites may have differed in the completeness of their documentation and the way they defined a "serious" infraction. There are some indications, for instance, that Denver kept less complete records of disciplinary actions, potentially deflating its overall infraction rate. As a result, only the most serious of the infractions across sites--escape, attempted escape, assaults against staff, and other assaults or fighting--are discussed and compared. These infractions were more consistently documented at every site than infractions such as disobedience. The number of serious infractions per youth in Mobile averaged 3.4, versus 2.6 in Cleveland and 1.9 in Denver. Whereas all sites experienced escapes, Mobile had by far the largest number, 34, versus 6 in Cleveland and 8 in Denver. Mobile also had more attempted escapes, although here the number was much closer to that experienced by the other sites. Escape and attempted escape rates in Mobile appeared to be well above the norm in juvenile facilities nationwide, according to data reported in Conditions of Confinement, a nationwide survey of juvenile detention and corrections facilities conducted in 1991 (Parent, Lieter, Kennedy, Livens, Wentworth, and Wilcox, 1994). That study reported an overall escape and attempted escape rate of 2.46 per 100 juveniles in facilities of all types, a rate of 2.72 for training schools, and a rate of 4.77 for ranches. Mobile's escape and attempted escape rate were estimated to be about 8 per 100. Denver's rate, at 4.6, also appeared somewhat above average, unless ranches are used as the standard. Cleveland's rate, 1.8, was below average. The methods used to calculate rates differ somewhat from those used in the Conditions of Confinement study, however, so it would be unwise to attach too much significance to precise numbers. As for assaults against staff, 20 were reported for the first-year platoons. Seventeen of them occurred in Cleveland and three in Mobile. No assaults against staff were reported in Denver. Infractions involving assaults or fighting among the youths were considerably more common than assaults against staff. Again, this behavior was least frequent in Denver. Overall, the most common serious infractions at the three camps were assaults or fighting among participants, disobedience or insubordination, and "loss of bearing," a term that covers a variety of behaviors deemed inappropriate for a boot camp participant, such as swearing, inappropriate dress, or roughhousing. Another way to look at serious infractions is to consider what proportion of participants committed each type of infraction. First, the proportion of youths who escaped from boot camp varied widely-- from almost 1 in 4 in Mobile to 1 in 9 in Denver to about 1 in 15 in Cleveland. These differences cannot be explained by the relatively minor differences in the physical security of the three facilities. Instead, there may have been differences across sites in the nature of supervision or in the propensities of youths to run away. The distance from one's home turf may have discouraged some potential escapees, for instance. Mobile's program is located near the center of the city. In contrast, Cleveland's boot camp is 30 miles away, and Denver's program, although not in quite as remote a location as Cleveland's, is still several miles from downtown Denver. Differences in escapes may also reflect the relative onerousness of the boot camp regimen or the punishments imposed for attempting to escape. Unfortunately, these factors are difficult to disentangle. For example, Cleveland, with the lowest escape rates, combined the least militaristic program with potentially the most adverse consequences for escape. Youths in Denver were relatively unlikely to have been charged with assaultive infractions, whereas nearly two-thirds of youths in Mobile had an assault or fight on their boot camp records. In Cleveland, over 40 percent had been involved in such an incident. As noted before, no assaults against staff were recorded in Denver. In Mobile, 2.5 percent of the youths assaulted staff, and in Cleveland, 13.8 percent did so. Each site had distinctive preferences for certain types of sanctions. Over half of Mobile's participants served time in the camp's isolation room or were assigned extra duty as a punishment for a serious rule infraction. Loss of privileges and letters of apology were the only other sanctions meted out with any regularity in Mobile. In Denver, too, the majority of youths in first- year platoons served time in the brig, with extra duty the only other sanction meted out to more than 10 percent of the participants. Cleveland's pattern was quite different. The only sanction experienced by a majority of participants was counseling or "redirection," a program- specific term that describes other kinds of verbal responses to misbehavior. Restraint, which refers to the practice of physically holding or constraining the movements of a youth who is out of control, and loss of privileges were the next most common responses to misbehavior. Few youths in Cleveland spent time in isolation because Cleveland had no "brig." It is tempting to speculate about the relationships between styles of discipline and the incidence of serious infractions, such as assaults and fighting reported above. However, no persuasive hypotheses can be offered. Although Cleveland, for instance, relied less than the other sites on traditional forms of military discipline and had the highest rate of assaults against staff, other factors may have been more significant. As reported in chapter 3, youths in Cleveland had more extensive criminal records than did participants in Mobile or Denver. Also, while commenting on the low rates of assault in Denver, the Denver camp commander noted that his staff tended to be older than those at the other boot camps and, therefore, may have had an easier time establishing and maintaining authority. Of the youths in first-year platoons set back for poor performance or specific misconduct, about 8 percent of youths were set back in Cleveland, whereas 31 percent were set back in Mobile. All of those set back in Cleveland eventually graduated. In contrast, 7 (18.4 percent) of the 38 setbacks in Mobile failed to graduate. As noted earlier, Denver did not use setbacks as a sanction. Boot Camp Graduation and Termination Rates The programs' most serious sanction is removal from boot camp. Of the youths who successfully completed boot camp and entered an aftercare program, the Cleveland program achieved the highest graduation rate, with 94.1 percent of its first-year participants completing boot camp. Completion rates in Denver and Mobile were lower, at 80.3 percent and 82 percent, respectively. The majority of terminations in Cleveland and Mobile occurred because of attempted escape. The next most common reason for termination was noncompliance with the program. Mobile also lost three participants because of physical or psychological problems. Reasons for termination from the Denver program were more varied. Of the 15 youths who failed to graduate, 6 were disruptive or otherwise noncompliant, 3 escaped, 5 were removed for medical reasons, and 1 was removed for a combination of medical problems and poor performance. Three of the five youths removed from that program for medical reasons were involved in suicide attempts, as was the youth removed for both medical problems and poor performance. The timing of removals from boot camp was also examined to determine if there were any trends. In particular, it was important to know if there were more program removals during the first 4 months of program operation. In contrast, Denver's termination rate was relatively constant throughout most of the first year, declining only slightly toward the end. To determine if there were any periods during which youths were at particular risk of dropping out, dropout was examined as a function of time in the boot camp. In both Cleveland and Denver, the majority of youths who were going to drop out did so within the camp's first 2 weeks. These early dropouts could be considered to be a type of participant selection error that was identified soon after transfer. In Mobile, however, the first 2 weeks of the program accounted for just 27 percent of boot camp losses, and a relatively large number of Mobile youths dropped out after the 10-week mark. Some of these late dropouts were youths who had been recycled. Attitudes of Youths Toward the Boot Camp Regimens In light of the preceding description of the boot camps' structure and discipline, it would be useful to know how the participants assessed the programs. Two measures of the attitudes of participants toward the regimens are available: o Informal interviews with youths belonging to the first or second platoon at each site conducted by evaluation staff during site visits. o Rating form responses from a number of youths who had reached the eighth month in the demonstration program and were participating in aftercare at that time. In both surveys, unfortunately, the respondents were a small and nonrepresentative sample of all youths who participated in the camps. The reactions of these youths are included, however, because they may be of interest to some readers. Results were difficult to interpret given the limited samples involved. Many of those participants most unhappy or most unsuited to boot camp were no longer available to express their opinions when these data were collected. In the interviews, several themes were common across sites: o Youths had anticipated that boot camp would be tough, and most found it to be a challenging experience. In Cleveland, however, several youths were disappointed that the program was not more intense ("like a Marine [Corps] boot camp"), although in Denver some youths described the first few days as the worst of their lives. o Most youths found it hard to adjust to boot camp rules and discipline at first, but they viewed boot camp as a challenge and were proud to be making it through. o Youths varied in their attitudes toward the drill instructors but reported that most were fair and consistent. Some youths specifically mentioned liking the level of personal attention from drill instructors. o Youths at every site particularly liked the programs' physical conditioning and drilling activities. In Cleveland, they also liked the guided group interaction program. However, youths also complained that they were sometimes treated inconsistently because staff were still developing some of the rules and procedures. Other unpopular aspects of the regimen, especially in Denver and Mobile, were cleaning up the barracks and "getting yelled at." In a survey asking participants to rate their agreement or disagreement with five dimensions of the camps, the majority of respondents rated camp rules as clear and saw drill instructors as fair, consistent, and helpful. Youths in Cleveland were somewhat more likely than youths at the other two sites to feel that DI's cared about them personally. On the other hand, proportionally more youths in Denver felt that their boot camp was like a real military boot camp; about 87 percent felt this way, compared with 58 percent in Cleveland and Mobile, confirming the evaluation team's impression that Denver was the most "military" of the three camps. Implementation Issues Each site gave considerable thought to the purposes and design of its boot camp program during the proposal development stage and the 6- month planning phase that followed the grant award. As a result, all programs opened on schedule in 1992 with a full complement of staff and a formal set of operating procedures. Two years later, programs in Cleveland and Mobile continued to operate, while Denver's program was about to be replaced by a larger boot camp program, authorized by the State legislature and operated under State auspices. In general, program staff and other juvenile justice personnel were pleased with the accomplishments of the intensive training phase, and the programs were well received by the public. Boot camp implementation was not trouble-free, however. Several problem areas emerged during the demonstration experience. Staffing Although the sites varied in their vision of what the boot camp phase should entail, all attempted to strike a delicate balance between the military model and the more traditional supportive and rehabilitative services model. At the level of program or camp director, it was especially important to have someone who could understand both the military and rehabilitative dimensions. The staff, too, had to strike the right balance. Moreover, the staff in Cleveland had to understand and integrate the therapeutic model that NAFI adapted to the boot camp setting. In practice, achieving this balance was not easy. Personnel with military backgrounds sometimes found it hard to adjust to working with youths who were younger, more defiant, and less accustomed to structure than the average military recruit. Those without military experience, on the other hand, lacked knowledge of military procedures and skills like marching, or a military orientation toward leadership and esprit de corps. At a more fundamental level, some staff probably mistrusted the military model and favored more traditional rehabilitative approaches. As a result, all three sites had to replace one or more staff who did not perform to expectations or was not philosophically in tune with the site's vision of how a boot camp should operate. In addition, both Cleveland and Mobile changed boot camp directors in the first year. Staff Turnover Two of the three sites, Cleveland and Mobile, saw substantial staff turnover as a result of resignations and dismissals. One reason may have been the camps' workload, which was intense and demanding, especially for drill instructors. In Mobile, the program was sometimes short-staffed, and DI's had to work overtime. In Cleveland, DI's were trained and expected to perform as counselors as well as DI's, and administrators noted that a certain amount of staff burnout was inevitable. Pay levels may also have been a factor in both Cleveland and Mobile, as staff often left for higher paying opportunities. Personnel turnover was costly in many ways. It sometimes left programs short-staffed, necessitated more on-the-job training, and when boot camp leadership changed, it required staff to adjust to new philosophies and procedures. Sometimes the impact of staff turnover was more specific. In Mobile, when turnover depleted most of the staff who had attended special training in the ROPES curriculum, on-the-job training was not deemed a satisfactory substitute. The boot camp responded by reducing use of the ROPES course even though it had been a core component in the original program design. Unfortunately, the researchers can only speculate about how these changes were experienced by boot camp recruits. On a personal level, it is likely that they welcomed some staff departures and regretted others. (In one instance, youths wrote letters of support for a dismissed staff member.) But more generally, turnover presumably exposed youths to staff with less experience and, on occasion, to new procedures and philosophies. Discipline Programs also struggled to find appropriate and effective disciplinary procedures that were consistent with a military model. o In Denver, the camp initially made heavy use of time in the brig, or isolation cell, as a punishment for serious infractions. The original brig was closet-sized, but the program began using a larger room when visitors from OJJDP objected. Denver staff gradually realized, however, that some youths actually sought brig time as a way to escape participating in the daily boot camp regimen. As a result, they made less use of the brig as time went on and substituted more extra- duty assignments. Staff also tried to make the brig less hospitable by piping in loud music or periodically waking up "prisoners." Mobile prevented youths from sleeping in the isolation room during the day. o In Mobile, the first boot camp director often ordered misbehaving youths to perform meaningless work such as digging big holes and filling them up. Under the next director, youths were more often ordered to perform constructive work such as planting shrubs around the perimeter fence. Termination Termination policy was a particularly difficult area for the camps. Staff recognized that they had to be realistic about their expectations for performance, especially during a youth's first few weeks in boot camp. Otherwise, there would be an unacceptably high level of program terminations; if standards were too exacting at first, youths would find it easy to "quit" the program simply by being noncompliant. The sites grappled with this problem in a variety of ways. o In Denver, late in the first year, four youths appeared to attempt suicide. No injuries resulted and Denver staff believed that at least three or four of the youths faked the incidents in order to be removed from the program. However, even staged incidents can sometimes result in accidental injury or death. Rather than risk another attempt, staff consulted with mental health experts and immediately terminated all but one of the youths involved. The fourth youth was eventually removed as well. Cleveland also experienced two or three suicide attempts. However, both attempts were treated as not serious and the youths were allowed to stay. There is no evidence that the sites viewed the suicide attempts as grounds for altering either their selection criteria and screening or the boot camp regimen itself. o Cleveland and Mobile developed a "setback" system that provided an alternative to expulsion for some youths who performed poorly. As noted earlier, however, Mobile used the option much more frequently than Cleveland. Mobile also used setbacks as a punishment for escape, but Cleveland did not. In both sites, expulsions declined over the life of the program. o After the first escape attempt, Cleveland took a hard line to deter further episodes. Absent youths who did not return within 48 hours were expelled from boot camp and transferred to the State Division of Youth Services for placement in a State juvenile facility. With the exception of the Cleveland policy on escapes, none of the programs had hard and fast ground rules for expulsion. They preferred to make decisions on a case-by-case basis. It is not known whether participating youths viewed these decisions as fair and consistent. In Denver, however, a youth who had been expelled from boot camp was allowed to restart the program after filing a grievance questioning the grounds for his expulsion. Other Issues Aside from these problems, a number of other issues surfaced at one or more sites: o Budget cuts. All of the programs suffered from unexpected cutbacks in OJJDP funds going into the continuation phase, which coincided with the beginning of the second full year of boot camp operations. Cleveland succeeded in filling the gaps from other funding sources, but both Denver and Mobile had to scale back their programming. Denver dropped one of the camp's two teachers, a specialist in the program's life skills curriculum, thereby weakening what was already an area of lesser emphasis at this site. Mobile laid off two drill instructors. o Training new staff. All of the programs trained their initial staff as a group before opening camp. But high turnover among staff in Cleveland and Mobile forced those programs to train replacements as they came on board. In Mobile, this training method was a particular problem because the boot camp was frequently understaffed and there was no formal one-on-one training regimen in place. Cleveland's boot camp operator, NAFI, had an advantage in that it had a more generous staffing pattern and operated a network of other programs where new personnel could be sent for offsite training. o Changing expectations on the part of new entrants. At every site, the first platoons entered boot camp relatively ignorant of the regimen they would face. As word spread to new recruits through the grapevine, however, this situation changed quite quickly and had, according to staff, mostly negative consequences. Youths were less likely to be thrown off balance by the camps' intake rituals and were sometimes less responsive to disciplinary methods. o Gang rivalries. All three sites accepted a substantial proportion of gang-involved youths, and inevitably, there were times when youths from rival gangs entered camps at the same time and were placed in the same platoon. Camp uniforms and rules that forbid personalization of clothing and tattooing kept gang-identified insignia to a minimum, and, in general, staff felt the situation was manageable under the close supervision and control afforded by the boot camp setting. None of the programs attempted to put rival gang members in separate platoons. o Educational programming. For a brief period, Cleveland ran its educational program using teachers from the host institution, a county-run residential facility. Because the two facilities had such diverse philosophies and behavioral expectations for participating youths, it was difficult to integrate their teachers. The program eventually hired a subcontractor to provide schooling. o Coexisting with the host institution. Each site had to cooperate with its host institution, particularly in scheduling the use of the indoor facilities and the grounds, which led to some tension in the startup phase, especially in Cleveland. Other adjustments were needed, too. Because Cleveland and Denver relied on the medical services of the host institution, they had to sensitize medical personnel to the possibility that youths would fabricate illnesses to avoid program participation. In Cleveland, the portion sizes at the host institution's cafeteria were skimpy for boot camp youths and had to be increased. o Planning for release. All three sites had occasional problems locating suitable placements in the community for boot camp graduates. To avoid this problem, the Cleveland program stopped accepting youths who did not have a postrelease placement lined up at intake. Chapter 5. Aftercare Programs When youths graduate from boot camp, they are released into a community aftercare program. During this crucial reentry phase, aftercare programs are to provide the supervision and support required to successfully reintegrate them into the community. This chapter describes the parameters the Office of Juvenile Justice and Delinquency Prevention (OJJDP) set for the aftercare designs and the aftercare programs developed by each demonstration site over the first 2 years of operation. The description begins with an overview of the three aftercare program models. The models are then discussed in terms of their underlying philosophy, their management and staffing structure, their content, and their mechanisms for monitoring and disciplining youths. The chapter concludes with discussions of implementation issues and the evolution of aftercare programs. Parameters Set by OJJDP One of the distinguishing features of the OJJDP program was the requirement that aftercare be an integral part of grantees' program designs. This requirement reflected the increasing body of research and opinion focusing on the central role that aftercare programming plays in the successful rehabilitation of youths. According to OJJDP's program announcement, aftercare was to be case-management driven, with individualized performance workplans developed and updated to guide delivery of services. The program was to draw upon existing employment, education, drug testing, and treatment resources in the community to assist in delivering services. Aftercare also was to build on the discipline and character-building processes initiated during the intensive training stage at boot camp. The program announcement described two separate aftercare phases--preparedness and accountability- -distinguished primarily by the amount of responsibility youths were to be asked to assume. The transition from preparedness to accountability was also to be marked by a movement away from external controls on a youth's behavior to internal, self-imposed controls. The goals for the preparedness phase sound remarkably similar to those for the intensive training phase: "To provide discipline, treatment and work experience, educational services, and character development" (Federal Register, 1990). Youths were to continue their participation in "rigorous physical conditioning pursuant to their performance workplan" and to "gain experience in highly structured settings" (Federal Register, 1990). In the final phase--accountability--the program was to provide youths support for self-discipline, work experience, educational services, and drug resistance skills. Participants would be required to continue in a physical conditioning program and to make restitution to victims. Youths who did not pursue academic and vocational opportunities, or who did not participate in treatment and community service activities, would be terminated from the program. Local Aftercare Program Designs How did the programs read these Federal requirements? Initially, aftercare did not receive the same attention from program developers as did boot camp, in part because they had the benefit of 3 months to plan aftercare services before the first youth graduated. The original proposals had sketched broad outlines for the aftercare component of the program; designs were finalized in April to June 1992. The programs also benefited from the fact that the aftercare approach outlined by OJJDP was not so different from the approach probation and parole agencies were already employing in the three jurisdictions. Emphasizing case management and referral to community resources were not new ideas. All that was required was to package these strategies and apply them to the population of youths who were graduating from boot camp. The programs departed from Federal guidelines in that they did not divide aftercare into preparedness and accountability phases, the latter of which was to emphasize restitution. Because none of the programs had restitution components, it did not make sense to maintain a distinction on this basis. The programs did, however, decrease supervision levels over the course of the aftercare program. Cleveland had three phases-- intensive, stepdown 1, and stepdown 2--and each level permitted greater independence. In Denver there were two stages of supervision, Wyatt Academy and probation. In Mobile there were no formal phases, and decisions about when to relax monitoring requirements were made on an individual basis. The Aftercare Programs in Operation Each jurisdiction configured its aftercare program differently. Because the programs were so different, a brief overview of each is given before their organizational and programmatic dimensions are compared. Cleveland The North American Family Institute (NAFI) operated a centralized aftercare program for boot camp graduates at City Center, a downtown location that served as the hub of operations for its staff of 15 case managers, coordinators, counselors, and teachers. NAFI operated Cleveland's aftercare program, giving the company commander responsibility for oversight of both boot camp and aftercare components. As did the boot camp, the program employed group counseling techniques such as guided group interaction to encourage the development of a positive peer culture. A full array of services was offered, including assistance in preparing for and obtaining jobs, substance abuse counseling, family counseling, tutoring, challenge activities, recreational and social events, and referrals to local self-help and professional agencies. The program added a transitional academy during its second year of operation so that youths could build on the educational progress made at boot camp and have more flexibility about when they would reenter the regular school system. Youths moved through three progressively decreasing levels of supervision, starting with mandatory daily attendance at City Center from 8:15 a.m. to 5 p.m. By the fourth month after release from boot camp, most youths had earned the privilege of attending City Center only three evenings a week. The program also employed a vocational placement coordinator who trained youths in how to participate in an employment interview and arranged for their local employment. Denver In Denver, an alternative school run by New Pride, Inc., Wyatt Academy, was the core of the aftercare program. For the first 6 months after release, boot camp graduates were assigned to the school, which was located on the top floor of a downtown office building. Three teachers taught a curriculum that included math, language arts, and social studies. An aftercare coordinator, who acted as assistant principal of Wyatt, taught physical education and leadership courses adapted from an Army curriculum. The aftercare coordinator also served as a liaison with the different agency staff involved with each youth. After school, youths had access to other services such as drug and alcohol counseling at New Pride, Inc., as indicated by their individualized plans. Case management and supervision of youths committed to the Division of Youth Services (DYS) were the responsibility of the DYS case managers assigned to Denver's boot camp; for probation youths, those same functions were the responsibility of the probation officers who originally referred the youths to the program. When a youth completed Wyatt Academy, the probation officer or case manager continued to supervise him until his court-ordered sentence was complete. At this stage, requirements for boot camp youths were the same as those for any other youths on probation. Mobile Seven Boys and Girls Clubs in the Mobile, Alabama, area had primary responsibility for aftercare programming for Mobile's graduates. Each youth was assigned to a club close to his residence and was expected to participate in that club's activities. The clubs provided youths with afterschool and weekend educational and recreational services such as tutoring, life skills courses, social skills classes, basketball and other sports, and community service opportunities. Overall management of boot camp graduates was assigned to a single aftercare coordinator who was responsible for ensuring compliance with a performance workplan specifying each youth's aftercare activities. All boot camp graduates were placed on probation status with the Mobile Juvenile Court and were required to report twice a month to a probation officer. In addition, youths required to make restitution were supervised by a restitution coordinator from the Strickland Youth Center. The directors of each club monitored attendance and participation in required activities and reported absences and other violations to the aftercare coordinator and to the participants' probation officers. However, with services to youths dispersed across a number of locations and fragmented among different agencies, it proved to be extremely difficult to deliver all required services at all seven Boys and Girls Clubs sites and, at the same time, monitor each youth's attendance or progress. Therefore, in September 1993, the aftercare program was revamped to provide more focus and control. The revised approach included a weekly core aftercare session located at a single Boys and Girls Club site, school and home visits by probation officers, weekday offsite aftercare events, and weekend aftercare events. The structural characteristics, locations, philosophies, content, and monitoring systems of these three aftercare programs are compared below. Structure of the Aftercare Programs As was the case in the boot camp phase, the private nongovernmental partners--New Pride, Inc., NAFI, and Boys and Girls Clubs of Greater Mobile-- operated the aftercare programs. Cleveland had a simple, centralized management structure. A few months into the program, the program assigned responsibility for the boot camp and aftercare components to a company commander who had offices at both the aftercare and boot camp locations. With central control of both boot camp and aftercare, it was relatively easy to implement common policies and procedures, to train staff jointly, and to develop joint programs and curriculums. Of the three programs, Cleveland also had the largest staff, all of whom were NAFI employees. Staff consisted of an aftercare coordinator, a family services coordinator, a vocational services coordinator, a substance abuse coordinator, four case managers, and two driver/counselors. In addition, when the program opened the transitional academy as an alternative to regular schools, it added a principal and three teachers. The staffing pattern allowed the Cleveland aftercare program to assume responsibility for the case management functions as well as the educational, recreational, and counseling activities. In contrast, Denver had the most complex management structure and the fewest staff. Although New Pride, Inc., operated both the boot camp and aftercare components, there was no central coordinator to manage both programs. Staffing at the aftercare program, Wyatt Academy, consisted solely of an aftercare coordinator, a principal, and three teachers. Their primary responsibility was to provide youths alternative schooling for the first 6 months after graduation from boot camp. Responsibility for case management was dispersed among a number of individuals and agencies. The aftercare coordinator was responsible for supervising youths each day at school and for following up on absentees. However, legal responsibility for youths who had been sentenced to DYS rested with the DYS case managers. These managers in turn sometimes delegated responsibility for youths in out-of-home placements, known as "proctor care," to the proctor care providers. Responsibility for probation youths ultimately rested with the individual probation officers who had referred the youths to the program. This structure was awkward and in some instances made it difficult to take prompt, effective action on problems such as truancy. Mobile's structure was almost the reverse of Denver's. Mobile had no central location for aftercare, and youths were dispersed among seven Boys and Girls Clubs in greater Mobile. In effect, the youths were mainstreamed into the educational and recreational programs already available to other youths in the community. There were no staff at those clubs who were exclusively dedicated to boot camp graduates, nor were there activities exclusively designed for them. The only staff person dedicated to aftercare was an aftercare coordinator, whose primary responsibility was to act as a liaison between the Boys and Girls Clubs and the boot camp. The probation officers assigned at boot camp retained responsibility for case management throughout the aftercare period. This decentralized management structure made adequate supervision of Mobile's graduates extremely difficult during the first year, a problem exacerbated by a coordinator who did not aggressively monitor attendance. The distribution of youths across seven different locations contributed to a deterioration of the camaraderie and esprit de corps that the program had been trying to build during the intensive training boot camp phase. Not only was there no integration of boot camp activities and policies into the aftercare phase, but the boot camp peer group was also broken up. During the second year of operation, the Boys and Girls Clubs of Greater Mobile attempted to rectify these problems by restructuring the aftercare program around a core curriculum. Boot camp graduates were required to attend a weekly aftercare session held at one designated club. This session was conducted by staff, probation officers, a team of Boys and Girls Clubs of Greater Mobile staff specifically assigned to the boot camp aftercare program, community organizations, and volunteers. In addition, youths were required to participate in monthly community service projects. Location Denver and Cleveland located their aftercare programs in downtown central locations. In Mobile, youths were dispersed geographically according to the Boys and Girls Clubs to which they were assigned. Youths in Mobile also had access to the resources of the County Court's Strickland Youth Center and to a Boys and Girls Clubs beach property for recreational and community service activities. In Cleveland, staff had difficulty finding a location that met the program's needs for space and access to recreational facilities. The program had to make two temporary moves before the City Center facility, located in a deserted industrial complex, became available. Cleveland's staff considered this sparsely populated area an ideal location because their facility was beyond the city's gang turfs. Moreover, the offices were near public transportation and quite spacious. The only drawback was that the complex had no outdoor space for physical training or recreational sports and activities. Denver's Wyatt Academy is located on one floor of an office building in downtown Denver about two blocks away from New Pride, Inc.'s main center. The academy's offices were intended to be temporary quarters while New Pride, Inc., sought funding for a much larger version of the school that would serve trouble-prone youths from a variety of referral sources and operate in a renovated school building. The program never moved from its original location. Consequently, it lacked recreational facilities, and meals had to be brought in because the space had no kitchen. On the positive side, however, the academy was close to New Pride, Inc., and to public transportation routes. Aftercare Philosophy Each of the programs applied a different theoretical framework to the activities and structure of aftercare. The rationale underlying the programs in Cleveland and Denver called for a segregated, intense program to ease the transition from boot camp to the community. In Mobile, however, the theory was that boot camp graduates could be mainstreamed into the supportive environment of the Boys and Girls Clubs of Greater Mobile. None of the programs formally documented the "aftercare" philosophy versus the "boot camp" philosophy. The philosophical descriptions below were gleaned from interviews with project staff and observations of the programs. The underpinning of both the boot camp and the aftercare approach in Cleveland was the normative model. Essentially, program efforts were directed toward using the peer group to develop a positive normative system that would replace a negative, antisocial normative system. The normative model could be applied in residential or nonresidential settings, using techniques such as guided group interaction. Guided group interaction sessions were an important aspect of Cleveland's aftercare program, although they were not held on a daily basis as they had been during boot camp. Whether or not their aftercare program had military overtones was irrelevant to Cleveland's approach. The aftercare services there were initially designed with few military elements and were not well integrated with the program's intensive training boot camp phase. Over time, however, an attempt was made to make the transition between boot camp and aftercare less abrupt by incorporating additional military elements into aftercare services and by developing joint programs. In contrast, the Denver program intentionally applied a rationale for its aftercare program that was different from the rationale applied to its boot camp. Wyatt Academy was to provide participants a protected, supportive, educational environment in which youths could practice new behaviors and reinforce the values they had learned in boot camp. The transition from the military boot camp to a serious school was symbolized by a change from the military-style uniform to a "prep school" uniform consisting of blazer and tie. Boot camp graduation ceremonies, where youths wore these new uniforms, also served to emphasize that expectations for the youths had changed. The primary focus was now to be on education. Youths would, however, continue to use courtesies such as "Yes, Sir" and "No, Sir" and attend physical education and leadership training courses based on military models. Aftercare in Mobile represented a total shift in focus from the external controls on behavior employed at boot camp to a youth development model. The assumption was that the programs and philosophy of the Boys and Girls Clubs of Greater Mobile would help to sustain the positive attitudes and behavior that youths had acquired during the intensive training boot camp phase. An individually tailored program would be developed for each youth, emphasizing education, social skills, or work, depending on the specific needs and objectives identified in a detailed "Administrative Transfer Plan" developed at boot camp. Aftercare Content Upon graduation from boot camp, youths in Cleveland and Denver were subject to strict surveillance and supervision requirements, including daily attendance at a day treatment center. In Cleveland, supervision was gradually reduced as youths exhibited responsibility for controlling their behavior. Youths were released after a minimum aftercare period of 6 months. In Denver, however, the same attendance requirements were maintained for 6 months, at which time youths were released to regular probation supervision for the duration of their sentences. Mobile's requirements were less stringent, with youths attending Boys and Girls Clubs programs for 9 months. Cleveland defined different levels of supervision and the approximate length of time youths were expected to stay at each level. Cleveland's supervision was most intense at the beginning, requiring daily attendance at City Center from 8:15 a.m. to 5 p.m. for 1 month and attendance at the transitional academy. Next, youths progressed to Stepdown 1, which gave them weekends off and permitted attendance at their regular school. Finally, in Stepdown 2, the program was individualized, but it typically included a requirement that youths attend City Center functions 3 nights a week. Denver's aftercare requirements were less stringent; youths there were expected to attend Wyatt Academy from 9 a.m. to 3:15 p.m., Monday through Friday, and to attend afterschool drug counseling or other programs as required by their case manager. After 120 school days, youths were released from the demonstration program to the sole supervision of their regular probation officer or DYS case manager. Mobile did not define specific categories of supervision. However, the case managers responsible for monitoring boot camp graduates' progress and attendance gradually reduced the number of contacts over the course of the aftercare period. The sites not only varied in the amount and length of supervision provided, but they also differed in their emphasis on types of services. Denver's primary focus was educational intervention through programs at Wyatt, supplemented as needed with substance abuse education and treatment at New Pride, Inc. Cleveland also emphasized education, adding a transitional academy in the second year of operation, but the centerpiece of the program was counseling, borrowing extensively from the guided group interaction approach. Cleveland also had the broadest array of direct services, offering substance abuse education and groups, family counseling, and vocational training. The emphasis was less clear in Mobile, possibly because staff stressed the importance of tailoring the program to the specific needs of youths. In Mobile, youths focused on education, social skills, or work, depending on their particular circumstances. Services emphasized there were those available at Boys and Girls Clubs--tutoring, case management, life skills training, recreation, and community service. Although youths at all three sites were expected to leave boot camp with individualized treatment plans and to take advantage of appropriate agencies in the community, few were actually referred to outside agencies, at least in the first months after graduation. A determination of appropriate custody arrangements was usually made in conjunction with development of the treatment plan. Researchers learned of the living arrangements for the small sample of youths for whom staff filled out aftercare ratings forms. Almost one-third of the sample in Denver was in foster care, with most of the remaining youths residing with at least one parent. Cleveland did not have any aftercare youths in foster care, but one-quarter of the youths resided with either a grandparent or a friend. In Mobile, however, all but 10 percent of its graduates resided with at least one parent. Described below is each component of the aftercare programs: education; physical fitness; military drill and discipline; counseling; drug and alcohol counseling, education, treatment, and testing; family counseling; vocational training and job preparation; and community service and restitution. Education Upon graduation from boot camp, youths in Denver entered Wyatt Academy. Graduates in Cleveland and Mobile initially were mainstreamed into the regular public school system. Because Cleveland's aftercare program encountered problems locating appropriate school placements for a number of youths and because youths had adjustment difficulties from entering school programs midstream, the program initiated a transitional academy in the summer of 1993. Thereafter, following the procedure in Denver, youths in Cleveland entered the transitional academy directly from boot camp. However, although Denver youths remained in the academy for a full 6 months, after about 3 months Cleveland made a decision to either mainstream the youths into a regular educational setting or continue their studies at the academy. GED candidates generally remained in the academy until they obtained their diploma. Cleveland's transitional academy was run by the Phoenix School, the same organization that operated the boot camp's educational program. The Phoenix School operates other alternative education programs in the State, serving similar populations of youths. Staff included two academic instructors, a half-time GED coordinator, and a half-time academic counselor. Courses focused on both academic achievement and behavior and included English, mathematics, history or citizenship, literature, science, youth leadership, and physical education. Unlike Cleveland's academy, Denver's Wyatt Academy was deliberately segregated from other New Pride aftercare services. The separate location, the requirement that youths wear blazers and ties, and the academy's heavy emphasis on basic subjects were intended to impart the message that school was the youths' primary focus for the next half- year or more. Wyatt was accredited by the State Education Department, permitting the program to obtain State educational funds for the students it taught. Staff included a principal and three teachers who specialized in math, social studies, and language arts. In addition, the aftercare coordinator taught physical education and a leadership training class. Wyatt's traditional daily curriculum included two periods of English, two periods of mathematics, and two periods of social studies. Classes were small and the instruction highly individualized, particularly in mathematics, a subject in which students' proficiencies varied considerably. School began at 9 a.m. and ended at 3:15 p.m., with the last period of the day designated for physical training and a leadership course. Youths were usually assigned daily homework assignments. Wyatt attempted to control attendance through both a positive incentive system and a disciplinary system. Youths who had perfect attendance all week or whose behavior was designated as exemplary earned the right to leave an hour and a half early on Friday, skipping the physical and leadership training. Absenteeism was punished by requiring a youth to make up each day of unexcused absence with 2 days of additional time in school. Denver's main objectives for youths were improving their educational performance and developing longer term educational goals, leaving Wyatt for another alternative school, or in the case of those who had completed their GED's, entering a community college or vocational training. Returning to a public school was also a possibility, but staff viewed this as an unrealistic plan for most youths. Youths could also stay at Wyatt beyond their required term, as did one youth. Mobile did not operate a transitional school, and the program confronted many of the same difficulties finding appropriate school placements initially experienced in Cleveland. Often, the schools that youths in Mobile had attended prior to boot camp were deemed inappropriate. Boot camp staff believed that officials at these schools often viewed boot camp youths negatively. Aftercare staff also were concerned about the effects of peer pressure and expectations, including gang membership. Staff helped youths finalize educational plans and accompanied them on a visit to their assigned school prior to graduation. The program described the youth's educational progress while at boot camp to school officials, and attempted to alleviate any concerns the school might have had about the youth's conduct. Most boot camp graduates were enrolled in regular schools, but truancy was a serious problem. As a result, the program attempted to increase the frequency with which staff checked on how youths were progressing in school. Physical Fitness Denver was the only program to continue daily physical training activities during the aftercare phase. Weather permitting, the last period of each school day was devoted to calisthenics and runs. Both Mobile and Cleveland incorporated sports into their aftercare programs, but did not offer intensive physical training. For instance, Cleveland's weekly schedule usually included 2 or 3 evenings of basketball or other games at nearby recreational facilities. In Mobile sports and games were offered at the Boys and Girls Clubs of Greater Mobile. Weekend activities at both locations included challenge activities. Military Drill and Discipline The aftercare programs incorporated very little that was military in nature except for military- style courtesies and manners and some military titles. In Denver, for example, youths were required to refer to each other as "Cadet" or "Mister," and faculty were addressed as "Sir" or "Ma'am." In Cleveland, some staff bore civilian titles, but the aftercare counselor/drivers wore uniforms and were referred to as "drill sergeants" by the youths. In fact, the counselor/drivers rotated assignments as drill sergeants in the boot camp. In response to youths' requests, additional military components, such as marching and drilling, were gradually added to the aftercare programs in Cleveland and Mobile. The aftercare programs in Denver and Cleveland incorporated more aspects of military leadership training into their curriculums. The last period at Wyatt was a combination of physical training and leadership training adapted by the aftercare coordinator, a retired Command Sergeant Major, from courses in the Sergeants Major Leadership Academy. These courses included effective listening, communication skills, character building, courtesy, and discipline. Cleveland's "Youth Leadership Curriculum" was developed by the third director of the program, a member of the Marine Corps reserves. Counseling The Cleveland and Mobile programs conducted values clarification exercises as part of their aftercare leadership or life skills courses, and all three programs referred youths exhibiting serious emotional problems for professional counseling. However, only Cleveland made counseling the linchpin of its aftercare program. During the first intensive phase of aftercare, youths in that program continued the daily guided group interaction or confrontational group sessions initiated in boot camp. Essentially, the program used the peer group to discuss issues confronted by youths and developed concrete, socially acceptable means of dealing with them. Drug and Alcohol Counseling, Education, Treatment, and Testing Some substance abuse education and counseling were incorporated into the leadership training and life skills curriculum at all three sites. In addition, both Denver and Cleveland had separate substance abuse program components. Cleveland's substance abuse program was under the direction of a substance abuse coordinator. City Center activities in this domain included: o The AMER-I-CAN program, a group counseling, self-help program designed to increase self- confidence and achievement among youths. It included an emphasis on prevention as well as cessation of substance abuse. All aftercare youths attended, regardless of their involvement in substance abuse. o The RAAM (Resocialization of African American Males) program, an education and group counseling program directed primarily at drug traffickers, although other youths also attended. The RAAM sessions were held weekly at City Center. o Referral to other community agencies, including Alcoholics Anonymous and Narcotics Anonymous. Denver referred boot camp youths with substance abuse problems to Out-patient Abuse Therapy (OPAT), a treatment program operated by New Pride, Inc. Among the therapeutic techniques OPAT employed were rational emotive therapy, relapse prevention therapy, and positive peer pressure. Drug testing was conducted infrequently at all three sites. In Cleveland, urine tests were occasionally administered when a case manager suspected that a youth had used drugs. In Denver, youths were tested infrequently at Wyatt Academy, but the New Pride, Inc., OPAT program routinely tested youths referred there for treatment. In Mobile, few youths were tested during the first year. However, under the new aftercare design, a few participants were selected randomly each week to undergo urine tests. Family Counseling Although all three of the demonstration programs recognized that dysfunctional families were a serious problem for many participating youths, the programmatic emphasis was on the individual and resolving his problems, rather than on the family. Only in Cleveland was a staff person specifically designated for family issues. This coordinator visited youths and their families at home and incorporated family goals into each youth's performance workplan. She also attempted to link families requiring assistance to community services. For example, the aftercare program linked several clients' families to the Home Energy Assistance Program, which helped heat those families' homes. She also referred them to support groups that focused on improving parenting skills. In Denver, youths who had been placed in proctor care were eligible for family counseling through the proctor care agency. Vocational Training and Job Preparation In the long run, the boot camps hoped to increase the work ethic among youths as well as their employability. By design, Denver focused on improving youths' long-term employability by concentrating on academic achievement. Wyatt Academy staff discouraged youths from holding jobs during aftercare, arguing that work would interfere with their attention to school. Cleveland, on the other hand, had a full-time vocational service coordinator responsible for locating employment for graduates, preparing them to obtain and hold a job (e.g., filling out applications and role-playing interviews), and monitoring their performance. However, the program fell short of its goal of having 25 percent of the youths employed during the first year. Mobile incorporated some job skills training into the life skills curriculum, but made no concerted effort in this area. In Mobile, youths showed a great deal of interest in an automobile mechanics training program that was in the planning stage when this study was conducted. Community Service and Restitution Although OJJDP's program announcement stipulated that the programs require restitution to victims and the grant applications had claimed restitution as a program objective, in practice none of the programs had a restitution component. In Cleveland, the juvenile court generally does not order restitution in conjunction with an institutional sentence. In fact, during the first year none of the youths at each camp had restitution orders. Among the small aftercare sample for which information was available, 5 of the 43 youths in Denver and 5 of the 27 youths in Mobile participated in court-ordered restitution. The boot camp program did not become involved in providing assistance with or monitoring restitution payments in those cases. Instead, restitution was handled through existing court routines. All three programs had some community service activity, although only in Mobile was there a routine, coordinated effort. There, service activities were scheduled almost every weekend. These activities ranged from working in nursing homes to helping out at community events such as "fun runs." Over 86 percent of the youths in the aftercare sample in Mobile participated in community service activities. The percentage of youths participating in community service was considerably lower in the other two sites: 17 percent in Cleveland, and 5 percent in Denver. Cleveland required just one community service activity for youths in aftercare, and in Denver the aftercare coordinator sometimes "persuaded" cadets who had misbehaved during the school week to "volunteer" to work on a community service project over the weekend. Summary In summary, the programs' aftercare components varied considerably more in their activities and structures than did their boot camps. They shared an individualized, case management approach to aftercare that included substance abuse prevention activities and life skills training. Mobile implemented a decentralized aftercare program, dispersing the boot camp youths among seven Boys and Girls Clubs of Greater Mobile and immediately mainstreaming them into "normal" environments. Both Denver and Cleveland centralized their efforts, focusing activities during the first few months around transitional schools serving only boot camp graduates. The transition to regular systems and activities was, therefore, more gradual at both of these sites. Finally, Cleveland offered the broadest array of services and employed the largest staff, more than twice the number employed in Denver. Mobile assigned only one staff person exclusively to aftercare activities. Mechanisms for Monitoring and Disciplining Youths Discipline during aftercare was a far more complex and difficult issue than discipline during boot camp. While youths were under the tight supervision and control of boot camp, their misbehavior was obvious and the discipline straightforward. Moreover, once the program spelled out the rules, youths who were unable to comply with them were simply expelled, in many cases to serve a sentence in a less desirable environment. In contrast, during aftercare a great deal of a youth's time was spent without the direct supervision of the aftercare staff, making it difficult to monitor behavior problems. Once a problem was brought to the attention of aftercare staff, what should be done? How much latitude should a youth be given before being expelled from the program? What should be tolerated as "normal" backsliding that is expected from youths who had already shown themselves capable of getting through boot camp? These considerations aside, programs were aware that a high expulsion rate would cast serious doubts on the viability of the boot camp model. The most pervasive disciplinary problem in aftercare for all three programs was absenteeism. In both Denver and Cleveland, the fact that youths were required to attend the program center each day meant that absenteeism was discovered fairly quickly. In Mobile, knowledge of a youth's absenteeism depended on the frequency of checks made by the aftercare coordinator. During the first several months, these checks were not very systematic. Once a problem was noted, however, the programs had different policies for handling youths: o In Cleveland, youths who were absent for a few days were brought before an attendance review board consisting of the youth's case manager, a teacher, a student, the family services coordinator, the aftercare coordinator, and the commander of the program. This board reviewed the problem, formulated a plan to improve attendance, and imposed a punishment, usually extra weekend hours at City Center. Youths who continued to miss activities could be brought in for a second or third review but were eventually subjected to a more serious review by the commander's board. At this juncture, a youth could be placed on "SPIN" status, meaning the program was attempting to engage him, but he was in danger of being classified as absent from the program and subject to a violation of a court order on that basis. o In Denver, the aftercare coordinator attempted to contact youths (and their families) who were absent for more than 1 or 2 days. If this contact did not prompt absent youths to show up at Wyatt, the coordinator would notify their probation officers or DYS case managers. There was some disincentive for youths to return to Wyatt if their absences were unexcused because their time in the academy would be extended by double the number of days they had missed. Ultimately, continued absenteeism resulted in expulsion from the program. Any additional consequences, such as return to court as a probation violator or transfer to another placement, were the province of DYS case managers or probation officers. o In Mobile, youths who failed to attend the Boys and Girls Clubs or school were first counseled by probation officers. Then, if attendance problems continued, youths were either sent back to boot camp for "recharging" for 1 or 2 months or were charged with a probation violation and brought back before the court. Strategies for handling other behavior problems were decided on a case-by-case basis, and some policies changed over time. In Cleveland, for instance, early adjustment problems prompted the program to try sending youths back to boot camp for a weekend refresher course. Threatening of staff and disruptive behavior resulted in some expulsions, but more commonly, youths simply did not show up for aftercare activities. Committing a new crime generally was another basis for expulsion from the program. As a practical matter, it often meant that youths were held in detention pending disposition of new charges and, as a result, could not participate in the aftercare program. Eventually, most youths brought before the court were committed to State institutions. However, for misdemeanors and local infractions, aftercare staff might attempt to retain the youths and sometimes even acted as advocates on their behalf in court. Aftercare completion, enrollment, and dropout rates for youths in the three aftercare programs were calculated on the basis of the status of youths in February 1994, about 10 months after the last of the cohorts in the sample had entered the program. Because the program was designed to last 12 months, all of the sites still had some youths enrolled, ranging from 2 youths in Denver to 6 in Cleveland and 23 in Mobile. At this juncture, Mobile had graduated almost half of the demonstration youths and Cleveland had graduated about 45 percent. In Denver, a smaller proportion, 26 percent, had graduated. Termination rates from the program ranged from 70.5 percent in Denver to 50 percent in Cleveland and 28.3 percent in Mobile. The reasons for termination varied somewhat by site. By far the most common reason for termination in Cleveland and Mobile was arrest for a new offense, whereas slightly fewer youths in Denver were terminated for new offenses than for simply going "AWOL"--the acronym that indicates a youth had simply ceased attending the program and was considered to have dropped out but was not known to have committed a new crime. AWOL terminations were less common in Cleveland and rare in Mobile. All sites also terminated some youths for failure to comply with other program requirements. Cleveland was the only site where participants died while in aftercare; the two fatalities there were reportedly caused by gang-related incidents. It is difficult to interpret these attrition patterns. However, it is interesting that Cleveland and Denver lost similar proportions to new offenses--33 percent and 29.5 percent, respectively--compared to 20.2 percent in Mobile. This order was expected, given the prior delinquent histories of youths at the three sites. As discussed in chapter 3, Cleveland youths had the most serious criminal histories and Mobile youths the least serious, with Denver youths falling between the two but closer to Cleveland. The remaining differences in attrition were largely attributable to differences in the AWOL percentages, as well as to the fact that Mobile had many youths classified as still active in the program. The AWOL differences may partly reflect differing responses to absenteeism or differing monitoring procedures. In Mobile, for instance, where youths reported to seven different Boys and Girls Clubs, attendance requirements were variable and only loosely monitored during some portions of the demonstration. As long as a youth continued to report regularly to his probation officer, he may well have escaped sanctions for poor attendance at his club site. In Denver, part of the explanation for the high proportion of AWOL terminations may lie in the unique structure of that program, which split aftercare responsibilities between Wyatt Academy staff and various case managers, including two DYS staff based at the boot camp and local probation officers who had referred probationers to the program. Program staff at Wyatt Academy were able to flag absenteeism but were ultimately dependent on these case managers and probation officers to sanction persistent truants. However, Wyatt staff felt that they did not always get aggressive backup from case managers and probation officers, in part because the structure was cumbersome and in part because some case managers were less enthusiastic about the aftercare program than the boot camp itself. In any case, if a youth did not reappear, Wyatt staff eventually listed him as withdrawn from the program. Unlike the boot camp attrition, which peaked in the first 2 weeks, the majority of aftercare attrition occurred after 3 months. It is not known if this pattern says more about the point at which youths stopped participating in the program or the point at which sites classified them as dropouts. However, the pattern does indicate that the programs cannot relax if youths make it through the first few months after release from boot camp because attrition occurs throughout the aftercare period. Implementation Issues The programs' aftercare components grappled with a number of implementation problems over their first 2 years of operation. Because these programs were the first demonstrations of juvenile boot camps, no aftercare models existed to emulate. Adult boot camp programs provided little guidance because little was known about boot camp aftercare for adults, and many of the issues and requirements confronting juveniles, such as custody arrangements or compulsory education, are irrelevant to adult programs. In the absence of models, the programs confronted a number of key issues. Tailoring Juvenile Aftercare to Boot Camp Graduates All three programs combined elements of traditional juvenile aftercare--case management, individualized treatment plans, and regular reporting--with service delivery in new, nontraditional settings. In Cleveland and Denver, day centers were created, and in Mobile, boot camp graduates were distributed to seven existing Boys and Girls Clubs. Aside from delivering services in innovative settings, it was an open question as to whether or how much programs needed to tailor aftercare programming to the boot camp population. There were several dimensions to the question: o After the strict regimentation and close supervision of boot camp, what degree of external control and supervision is appropriate during aftercare? The Cleveland and Mobile programs moved from less supervision to tighter controls over the course of this evaluation. Both required refresher courses at boot camp for youths showing signs of regressing. Cleveland also developed a separate, midcourse educational program to eliminate some of the problems youths had in adjusting to normal school environments. Many staff felt that the transition to aftercare needed to be more gradual, with more external controls extended into aftercare. o How important is it to maintain the platoon as a group and to build aftercare activities around that group? Cleveland was the only program that attempted to continue building group cohesion during the aftercare period. Cleveland's aftercare approach relied on group interaction techniques to foster the development of a positive peer culture. Although Denver kept platoons intact simply by having a central location, there was no concerted effort during aftercare to build on the sense of responsibility to the group that youths had developed in boot camp. In Mobile, original platoons did not even congregate in aftercare. However, late in the program, Mobile did implement a requirement that all youths attend one session per week at the same Boys and Girls Clubs. o To what extent should military policies, procedures, and customs be integrated into the aftercare program? Except for titles and customs, the program incorporated very little that was military in nature. In that sense, they were virtually indistinguishable from non-boot camp aftercare programs. Except in Denver, even the physical training conducted during boot camp was discontinued. Some youths expressed interest in continuing some of the military drills, and youths in Cleveland conducted sporadic drill practices. Sanctions A second set of implementation issues concerned how and where to draw the line on behavior problems in aftercare. If it was difficult to develop consistent sanctions in boot camp, an environment that limited the set of problems to be addressed, it was even harder during aftercare, where staff sometimes did not even have a consensus about what behavior should be considered a problem. o What constitutes a behavior problem in aftercare, and at what point does a problem warrant intervention? A new arrest was an obvious problem, but more commonly the programs faced behaviors like absenteeism that fell in a gray area between acceptable teenage limit testing and a real problem. After how many absences does the program shift from attempting to lure a youth back into the program to taking an action that probably ends his tenure in the program? What messages does it impart to other youths when a youth has been absent for 10 days and then shows up again with no obvious consequences? For the most part, responses to these issues developed on an ad hoc basis as individual cases occurred. Only at Denver's Wyatt Academy were the consequences of absenteeism predictable. o What sanctions should be applied to problem behaviors? Does early punishment result in additional absenteeism? Conversely, do youths learn that they can get away with skipping school or aftercare programs as long as they show up every now and then? Some of the programs used incentives to reward attendance and other positive behavior. But for serious behavior problems, the only sanction that sent a strong message without pushing the youths out of the program was to send them back to boot camp to be refreshed or recycled. Both Cleveland and Mobile made limited use of this strategy. Part of the programs' reluctance to develop concrete policy in these areas may lie in the fact that they had invested considerable effort in each youth by the time he entered aftercare and did not want to give up on him so late in the process. Organization and Management Structure Lack of clarity in case management responsibilities and procedures and in lines of authority were issues in Mobile and Denver. In both sites, case management involved transfer of information and responsibility from one agency to another. Lines of authority and responsibility were much clearer in Cleveland, which employed a management structure in which one organization controlled both program and case management functions. o Who monitors day-to-day supervision of youths? Who has the responsibility to locate absent youths and attempt to re-engage them in the program? When these responsibilities are divided among different staff in agencies or locations, it is difficult to monitor the activities of youths and to intervene early when problems first arise. In Mobile, boot camp case managers, who were county probation officers, continued to be responsible for youths in aftercare. However, the aftercare coordinator employed by the Boys and Girls Clubs of Greater Mobile was responsible for monitoring youths at the seven club locations. There were no clear-cut policies delineating how or when information would be shared among these parties and who had responsibility to take definitive action if problems arose. In Denver, the situation was even more complex with case management responsibility assigned to DYS case managers, probation officers, or proctor care providers, depending on the youth's initial sentence and aftercare residence. As a result, Wyatt Academy's aftercare coordinator might have to contact multiple individuals at those agencies to discuss a youth's behavior problems in school. At a deeper level, these structures involved a fragmentation of responsibility for decisions about a youth's status in the program and, in some cases, ambiguity about his aftercare goals. If a serious behavior problem is observed, who or which agency should impose sanctions? More generally, who should judge whether the program was meeting the youth's needs? In both Denver and Mobile, not all of the staff shared the same notions of their own and others' responsibilities. And, if there was a problem with a policy or implementation of that policy in another agency, it was unclear what avenues they could use to address the problem. In Denver, for instance, there was some evidence that Wyatt staff and case managers sent youths mixed messages about the importance of complying with the Wyatt Academy program and disagreed about whether youths should get jobs. There was no formal mechanism for resolving these differences. Resources A final issue for Denver and Mobile was that the resources available to those programs were inadequate to implement the full scope of planned activities. When grant allocations were cut during the second award period, the still-evolving aftercare programs were particularly hard hit. Many improvements that the sites had been considering were set aside. Denver, for instance, had originally proposed to establish a systematic community service component. Early on, program staff also spoke of making a more concerted effort to work with families while youths were in aftercare. Neither of these plans was ever carried out. In contrast, Cleveland incorporated different assumptions about funding levels into their original proposal and had backup plans that enabled the site to more adequately fund its program. By the close of the evaluation period, the boot camps had worked through many of their problems and settled into a routine. In contrast, the aftercare programs were still struggling, although they already were quite different from those in place when the first youths graduated from boot camp. Disappointed by first-year attrition rates, program staff were virtually unanimous in singling out aftercare as the demonstration element most in need of further improvement. Chapter 6. Changes During the Boot Camp Phase The rationale underlying the demonstration programs discussed in this report is that the intensive training experiences during boot camp produce changes in participants' knowledge and skills, attitudes and values, and behavior. As indicated by the rationale presented in chapter 2, these changes in turn result in increased self- sufficiency in the community and decreased crime and delinquency, all of which would ultimately reduce the costs of delinquency to society in general. At the same time, youths are held accountable to victims and the community through restitution and participation in community service activities. In Cleveland, an additional objective was sought: to divert youths from longer term institutional placements and, in so doing, achieve cost savings. In theory, many of the changes expected for participants were to have occurred by the time they graduated from boot camp. There was an assumption that military structure and discipline, in conjunction with the boot camps' other elements, would alter the youths' skills, attitudes, and behavior. Support during the aftercare period would then enable youths to consolidate what they had learned in boot camp and to gradually test their new skills in their own environment. This chapter discusses the very limited data available from the implementation evaluation. These data evaluate the progress of youths in three areas: improvement in knowledge and skills, improvement in attitudes and values, and improvement in behavior. Measures of improvement included physical fitness tests, educational tests, and staff ratings of individuals on seven categories of performance, supplemented by anecdotal evidence on the effectiveness of the boot camp obtained during interviews with youths and onsite staff. While evidence of positive changes in skill, attitude, and behavior at the conclusion of the 3-month camps does not necessarily ensure that the program will produce the long-term impacts it desires, the absence of any positive changes would certainly raise concerns about the validity of the concept or its implementation. The focus of this chapter, therefore, is on any changes that occurred in the short term. In the absence of appropriate comparison data and lacking adequate followup periods, the longer term changes that the programs hope to stimulate are not addressed. This chapter also reports program completion rates during the evaluation period, describing the reasons for program attrition at selected attrition points. It then compares the characteristics of youths who graduated to those who did not. Finally, although cost-effectiveness analyses are beyond the scope of this evaluation, the operating costs of the three boot camp and aftercare models are described, and operating costs per bed are estimated. Evidence of Change During the Boot Camp Phase Some data were available to describe changes observed in youths who participated in the intensive training or boot camp phase of the demonstration programs. The following areas were assessed: literacy and educational performance, physical fitness, attitudes and values, and behavior while at boot camp. Test scores in Cleveland. When the Phoenix School took over the educational program at Cleveland's boot camp in September 1992, all youths were routinely given the Wide Range Achievement Test- Revised (WRAT-R) on entry into boot camp and just before graduation. The WRAT-R is a standardized, nontimed exam testing reading, spelling, and math skills. It is used to determine academic achievement levels from age 5 to adulthood. Data were available for youths entering boot camp from cohorts 5 through 12. On entry to boot camp, testing of 76 Cleveland youths in reading revealed that roughly one-third tested at elementary school level (grade 5 or below), about one-third tested at middle school level (grades 6 to 9), and about one-third tested at high school level (grades 10 to 12). Spelling and math scores were lower, with over half of the youths testing at elementary level in spelling, and three-quarters testing at grade 7 or below in math. The appropriateness of grade levels according to individual ages was not considered because of the wide range of acceptable ages for any given grade. Recalling that the mean age of entry into the boot camp program is 16.5 years in Cleveland, however, an average grade level of 7 in reading and 6 in spelling and math means that, by any standard, a considerable number of youths were below grade level. Posttest administration of the WRAT-R permitted an assessment of changes in grade level at the end of boot camp. There were statistically significant increases in reading, spelling, and math as measured by the WRAT-R. Over half of the youths increased by at least one grade level in reading and math, and 46 percent increased by that amount in spelling. A sizeable proportion of youths increased their standing by at least two grade levels--36.8 percent in reading, 25 percent in spelling, and 28.9 percent in math. For a 3-month educational program, these changes are very encouraging, particularly for youths with a history of academic failure. There also was a group, however, who did not exhibit grade-level changes during the boot camp period. Thirty-three percent of youths in the camps did not improve in reading, 39.5 percent did not improve in spelling, and 32.9 percent did not improve in math. Overall, the odds that these changes could be explained by chance alone are less than one in a thousand. Test scores in Mobile. Mobile's educational program was based on the PACE Learning System, an individually prescribed instructional curriculum that encompasses grade levels 2 through 12. The goal of the program was to return recruits to the public school system with both improved abilities and increased confidence or to prepare youths for the high school equivalency examination (GED). After entering boot camp, each youth took the Test of Adult Basic Education (TABE) to determine reading, math, language, and spelling achievement levels. Pretest TABE scores were converted to grade equivalents using norms for juvenile offenders. Teachers used scores on the TABE to design the appropriate self-paced instructional package. For youths who scored high on the TABE, PACE instructional materials were supplemented with grade-appropriate curriculum materials from the Mobile County School system. Youths took the TABE again on completion of the educational plan or at the conclusion of the intensive boot camp training period. Of the grade levels, as determined by TABE results, of boot camp youths at the time they entered the demonstration program, over one-half were at grade level 6 or below, and one-quarter were at grade levels 7 or 8. The mean pretest grade level was 5.9--considerably lower than that expected for youths whose average age was 15.5. During their time in boot camp, two-thirds of the youths were able to improve their total TABE grade-level equivalents by at least one grade level. Twenty-nine percent of the youths increased their total scores by two grades or more. Youths were more likely to improve their reading and language scores than their math and spelling scores. The mean grade-level changes on the four subtests--reading, math, language, and spelling-- were all 1 year or more. As with the WRAT-R, changes for each of the subtests and for the total score were statistically significant. As was the case in Cleveland, youths who had initially scored at the low end of the distribution did not make gains as great as those who had initially scored higher. These mean grade-level increases of about 1 year in Cleveland and more than 1 year in Mobile are consistent with other assessments of correctional educational programs that report pre-to-post increases in achievement scores (Cohen and Filipczak, 1971; Mayer and Hoffman, 1982; Ball, Parker, and Saunders, 1982; McAfee, 1984). What is striking about the grade-level increases in Cleveland and Mobile is that they occurred over only a 3-month educational program, a shorter pre- to-postinterval than is typically employed. Youths' assessment of educational programs. In addition to examining the test scores discussed above, a small sample of youths in aftercare were asked whether the boot camps' educational programs had improved their reading, writing, and math skills. This sample was limited to youths who were still active in the fifth month of aftercare. Of the 19 respondents in Cleveland, 73.6 percent agreed or strongly agreed that the educational program there had stimulated improvements. Similarly, 78.8 percent of the 33 respondents in Mobile agreed with the statement. However, in Denver only 33.4 percent of respondents perceived that the educational program had improved their reading, writing, and math skills. These results are not particularly surprising in light of the observation made in chapter 4 that Mobile placed the most emphasis on education and spent the most time per day on educational activities during boot camp, whereas Denver placed the least emphasis during boot camp on this aspect of the program. Physical Fitness Only the Cleveland program provided enough scores on physical fitness tests to report on this aspect of the camps' regimens. The Cleveland program tested muscular strength, endurance, and cardiorespiratory endurance at three 30-day intervals. Because there were concerns about the fitness levels of youths on entry to boot camp and fears that injuries would result if the fitness program were too aggressive, no testing was conducted until the youths had about 1 month of preconditioning. The test consisted of repetitions of pushups in 1 minute, repetitions of situps in 2 minutes, and a timed 1«-mile run. The physical fitness test was similar to the Army Physical Fitness Test, but with requirements scaled back for a younger age group. When the pretest and posttest mean number of pushups and situps, the mean times for the 1«-mile run, and changes in these scores are considered, the pretest scores are quite high. For instance, the mean score for situps at the pretest point would earn a youth a high score (78) on the Army Physical Fitness Test, based on standards for the 17- to 21-year-old range. This suggests that either the average youth had very quickly improved his conditioning since point of entry, or staff concerns about the effects of a sedentary lifestyle proved to be unwarranted for most youths. By the 90-day mark, youths' conditioning had improved still further. They performed an average of 12.1 more pushups in 1 minute, 9.9 more situps in 2 minutes, and ran the 1« miles in 37 fewer seconds than at the 30-day mark. Paired t- tests indicated that these differences were statistically significant at the .05 level or higher. There was anecdotal evidence from all three sites to bolster these test data on improvements in physical conditioning. For instance, several youths in Cleveland and Mobile reportedly shed a perceptible amount of weight as they progressed through the fitness program. In Denver, some youths were so proud of their physical shape that they attempted to send photographs home for relatives and girlfriends. Whether because of the physical conditioning, a drug-free environment, or improved nutrition, a number of youths reported in interviews that they felt better while at boot camp. Attitudes and Values Staff at all three sites rated youths on eight attitudinal and behavioral dimensions: respect for authority, self-discipline and control, responsibility, integrity, teamwork, personal appearance and bearing, social behavior, and work ethic. Instructions called for the ratings to be completed by the drill instructor working most closely with each participant and for ratings to be made at the end of week 2 and week 12, when youths completed the boot camp phase of the program. The ratings were made on a 7-point scale, with a rating of 1 indicating little respect for authority, and a rating of 7 indicating great respect for authority. Because staff did not routinely rate youths until the first few cohorts had gone through, particularly in Denver and Mobile, ratings were not available for all youths. A total of 134 complete sets of pre- and postratings were collected, the bulk of which were from Cleveland. Because there were different raters at each site, it is not appropriate to conclude that youths at one site exhibited more positive or negative behaviors and attitudes than youths at another site. What these figures do reveal, however, is that just before graduation from boot camp, staff on average rated youths above the midpoint in every category of behavior and that the ratings were fairly consistent across the eight categories. Staff perceived that the youths generally followed orders and acted respectfully; exerted control over their behavior and stayed out of trouble; kept commitments and maintained property, clothing, and equipment; told the truth about personal matters and admitted mistakes; contributed to team efforts; kept themselves properly and neatly groomed and maintained military bearing; were polite and considerate and used appropriate manners; and persevered in completing assignments as well as volunteering for extra tasks. On average, staff rated youths slightly higher in terms of their personal appearance and bearing and their work ethic than on their respect for authority and self- discipline. When the average change in staff ratings of boot camp participants by site during the 10 weeks between the initial and final ratings are presented, because of differences in rating styles, one cannot conclude that one program had more impact on these behaviors than did another program. In fact, in Cleveland there may have been a ceiling effect in the ratings. The average initial rating on the eight behavioral dimensions was 4.8 in Cleveland, whereas it was only 3.1 in Denver and 2.6 in Mobile. On a 7-point scale, the initial scores recorded by Cleveland staff left less room to indicate any gains exhibited by the participants in that boot camp. In fact, Cleveland staff ratings did not show much improvement over the course of boot camp, with all gains being less than 1 point on the 5-point scale. However, all but three of those changes were significant at the .05 level. The Denver and Mobile ratings showed substantial improvements during 10 weeks of boot camp, and all of these changes were statistically significant. On average, these improvements were uniform across the eight dimensions rated. From the perspective of drill instructors who were working most closely with the youths, there had been considerable positive changes in participants' attitudes and behaviors by the week of graduation. The patterns of change were somewhat different in the two sites. In Denver, staff indicated that youths changed most in their work ethic and personal appearance and least in terms of their teamwork. Youths' respect for authority changed most in the opinion of Mobile staff, and least in integrity and social behavior. Frequency of Behavior Requiring Disciplinary Action Another indicator of whether or not the youths' behavior improved during the boot camp period is the frequency of disciplinary infractions. The percentage of disciplinary infractions that occurred in the first, second, and third months of boot camp show that Cleveland's infractions show a pattern that would be expected if behavior were improving with time in the program: close to half of the infractions occurred in the first month and less than 22 percent in the third month. The relationship was nonmonotonic in Mobile and Denver, although at both sites there were fewer infractions in the last month than in the first month. With the possible exception of Cleveland, these differences are too small to draw any conclusions about improvements in behavior. Evidence of Change During the Aftercare Phase During this evaluation only two types of data were available to assess the progress of youths once they left boot camp: ratings of changes at approximately the eighth month (typically the fifth month in aftercare) and logs describing each youth's status in the program or attrition from it through the 12-month period of expected program participation. Youths' Ratings of Changes Nineteen youths from Cleveland, 15 from Denver, and 33 from Mobile rated changes on 14 dimensions of their behavior. They considered changes that had occurred from before boot camp to 4 months into the aftercare phase. Simple 3-point scales were used that asked whether a youth was engaging in a behavior more, the same, or less, or whether an area was better, the same, or worse. Considering the percentage of youths at each site rating, each dimension in a positive or negative direction. The most striking aspect of the youths' assessment of their progress in almost every area is that the majority reported doing better in aftercare than they had before boot camp. The vast majority claimed to have reduced their criminal activities and drug and alcohol use since entering boot camp. Although youths were promised that their responses were confidential, the respondents may, of course, have had a tendency to select socially desirable responses. Nonetheless, the pattern of responses is interesting and there is no evidence that the average youth consistently selected the "correct" answer to every item. There were some areas, however, where less than half of the youths claimed improvements. A smaller percentage of Denver youths reported that they were getting along better with other people (27 percent) and working better with others (40 percent). And at all three sites slightly less than half of the respondents indicated that they were trying to do school assignments and getting along better in school since entering the boot camp program. This response is somewhat surprising given that youths in Cleveland and Mobile appeared to have achieved educational gains at the conclusion of the boot camp phase. Also of interest was that relatively small percentages of youths at each site reported having changed their peer group since boot camp. Youths reported that they still were socializing with the same friends they had spent time with before boot camp. Although not much weight can be attached to these results because of the small sample sizes and possible response biases, it appears that the average youth perceived the program as having changed his behavior for the better. Program Attrition Chapters 4 and 5 described attrition patterns during the programs' boot camp and aftercare phases. About 10 months after the last of the cohorts in this study's sample had entered the program, 42 percent of the youths in Cleveland, 21 percent in Denver, and 40 percent in Mobile had graduated from the aftercare program. Another 19 percent of the youths in Mobile were still enrolled in the aftercare program, as were smaller percentages of youths in Cleveland and Denver. Over three-quarters of the youths in Denver did not complete the program; the bulk of these losses occurred during the aftercare period. The attrition rate was considerably lower in Cleveland at 53 percent, with nearly all losses occurring during the aftercare period. Mobile had the lowest overall attrition rate, losing slightly more in the aftercare period than during boot camp. To determine if program startup difficulties were a factor in program attrition, the attrition rates for the first few cohorts of youths to enter the program (four in Cleveland and Mobile and three in Denver) were compared with those for the last few cohorts. In both Cleveland and Mobile, the first four cohorts accounted for a higher proportion of program losses than did the last four cohorts. The changes both of these programs made in their aftercare programs--Cleveland implemented a transitional school and Mobile restructured aftercare to include more activities exclusively for boot camp youths--may have begun to reap some benefits. In Denver, however, there were just as many dropouts among the last few cohorts as among the first three. The primary reasons for program attrition varied across sites. Program terminations for medical reasons or for "failure to comply" accounted for over half of Mobile's losses and about one-quarter of Denver's losses. In Cleveland, these categories accounted for only 15.9 percent of the program's losses (i.e., 53 percent of Cleveland's youths did not graduate, and 15.9 percent of this group were terminated for "failure to comply.") Rather, 61.9 percent of the nongraduating group were terminated due to arrests for new charges and 18.9 percent due to what the program referred to as "AWOL"-- escape from boot camp or nonattendance in aftercare. Denver lost a smaller proportion to new crimes, but it had a considerably larger proportion of youths classified as "AWOL" with no known new offenses. Cleveland's pattern of attrition reflects the fact that very few program losses occurred during the boot camp phase when most terminations due to failure to comply or medical reasons took place. Cleveland did not terminate any youths for medical reasons and only 8.4 percent for failure to comply with the program. Close to one-third of Cleveland youths were terminated from the program because of arrests for new offenses, and another 10.1 percent were classified as "AWOL," with no known new offenses. Two youths were removed from the rolls when they were killed in incidents reportedly related to gang activity. In contrast, about one-fifth of participants in Denver were terminated because of medical or psychological problems or failure to comply with the program. Denver also lost a sizeable proportion of youths to new arrests during the program period. In addition, a significant proportion of youths, 31.6 percent, were classified as "AWOL." Similarly, Mobile terminated 22 percent of its entrants for failure to comply. The program reported few terminations for medical and psychological problems, however. About 11 percent of youths in Mobile were terminated because of arrests for new crimes during the program period, and 6.6 percent were designated as "AWOL." Overall, 11 youths were charged with violent crimes, 2 of which were murder. Receiving stolen property (which in Cleveland includes auto theft) and selling or possessing drugs accounted for most of the other new charges. In Denver, 13 youths were charged with property offenses, and 3 were charged with violent crimes, 2 of which were robberies. Of the 14 youths in Mobile arrested for new offenses, all but 2 were charged with property offenses. One youth was charged with rape and one with robbery. The relationship between program attrition and the criminal and social characteristics of the youths at intake was also examined. In Cleveland, there was a statistically significant relationship between the seriousness of a youth's prior sentence and completion of boot camp (x2 = 14.3, df = 3, p < .01). Of the 32 youths who had a prior institutional sentence to either ODYS or YDC, only about 22 percent graduated from boot camp or were still in the program. Of the youths who had been placed on intensive probation, slightly more than half dropped out of the program. On the other hand, the majority of youths whose most serious prior sentence was probation completed the program successfully. Boot camp staff in Cleveland suspected that youths with prior incarcerations did not perform as well in boot camp. Some surmised that they were less afraid of the boot camp alternative, in this case ODYS, because they had already spent time there. Other staff believed that a prior incarceration was just another indication that a youth had penetrated too far into the criminal justice system to easily reverse course. The fact that none of the indicators of the seriousness of the youths' instant or prior offenses was related to dropout tends to support the former interpretation--that it was the previous experience of incarceration rather than the seriousness of the criminal career that was associated with program losses. The other factor related to dropout in Cleveland was gang involvement, with youths exhibiting strong gang affiliation more likely to drop out than those with minor or no gang affiliation (x2 = 6.7, df = 2, p < = .05). In Denver, the same relationship between program dropout and prior incarceration emerged, although the cell sizes were too small to test for statistical significance. Of the Denver youths who had been incarcerated (as part of a probation plus detention sentence or Department of Institutions commitment), only 10.8 percent completed or were still enrolled in the program, compared to 30.8 percent of youths with no prior commitments. As in Denver, the association between program dropout and previous incarceration emerged in Mobile, although there were too few youths with previous incarcerations to make statistical significance levels meaningful. Of the six youths who had been incarcerated, five did not complete the program. Other variables related to dropout in Mobile were discipline problems at home or school and a history of fighting. Also, youths who were involved in drug sales were more likely to drop out. As was the case for Cleveland, however, the variables characterizing the seriousness or length of the criminal history were unrelated to program completion. Program Costs Over the course of the demonstration, Cleveland received approximately $1.4 million in OJJDP funds, and Denver and Mobile each received about $1.5 million. All sites were awarded two separate grants, each covering 18 months. Because of budget limitations at OJJDP, the second award was smaller than expected, which forced the programs to cut costs or find money from other sources. In contrast to the other sites, Cleveland heavily subsidized its demonstration with nonfederal funds. In part, this was possible because Cleveland's program qualified for a State subsidy designed to encourage localities to use more correctional options in the community. It was difficult to examine program costs in detail due to different budget procedures and time frames across sites and limited information on nonfederal and inkind contributions to the demonstration programs. Caution should be taken in the following areas when interpreting these figures: o It was necessary to allocate some budget line items (including administrative costs, training costs, and a number of other expenses) between the boot camp and aftercare phases. A 60 percent boot camp/40 percent aftercare split was used. o Denver's aftercare costs did not include the cost of teachers, an expense underwritten by the school system. o Budgets included travel to meetings, data collection, and some other administrative costs that probably would not have been incurred if the sites had not been part of a demonstration program. o Costs excluded some inkind contributions if the sites did not attach a specific dollar value to them. Denver, for example, did not estimate the value of their boot camp space. Personnel costs accounted for the largest share of the program budgets, both in boot camp and aftercare. Allowing for the crudity of these estimates, it is still interesting that the costs of boot camp were quite similar across sites, ranging from a high of $75 per day in Cleveland to a low of $66 in Mobile. Costs of aftercare were more diverse for several reasons, including the intrinsic differences in the three programs. Mobile's aftercare, which did not include an educational component or a special program center for boot camp graduates, was by far the cheapest. In contrast, Cleveland's aftercare program was the most comprehensive and therefore the most costly. Differences between the cost of aftercare in Denver and Cleveland are overstated, however, because the cost of teachers' salaries in Denver are not included. Note that the various costs estimates (per slot, per day, per participant) are affected by assumptions about program capacity, occupancy rates, and length of stay. The estimates shown assume full occupancy of the programs. Other jurisdictions would need to consider these assumptions carefully in forecasting their own cost estimates. Using full occupancy as the cost standard, however, all programs believed that their boot camp residential programs operated at a lower daily cost than alternative facilities. Ohio estimated the average daily cost of placement in its State facilities at $99, whereas Denver's higher security facilities averaged $138 per day, and its community residential facilities averaged $92. This chapter has presented some preliminary evidence that youths who entered the demonstration programs during the first year achieved physical fitness, educational, and attitudinal gains during the 3-month boot camp phase. Most youths completed the boot camp phase, but attrition during aftercare was a problem for the programs. There was a strong, statistically significant relationship between an incarceration prior to boot camp and program attrition. In the next chapter, the implications of these findings are discussed. Chapter 7. Findings, Conclusions, and Recommendations The Office of Juvenile Justice and Delinquency Prevention initiated the boot camp demonstration program to develop boot camp models for the juvenile system and to test the feasibility and appropriateness of their implementation. In September 1991, three groups received awards to develop and implement boot camp programs as intermediate sanctions: o Cuyahoga County Court of Common Pleas, Cleveland, Ohio, working in association with the North American Family Institute. o Colorado Division of Youth Services, Denver, Colorado, in association with New Pride, Inc. o Boys and Girls Clubs of Greater Mobile, Alabama, in association with the Strickland Youth Center of the Mobile County Juvenile Court and the University of South Alabama. Simultaneously, the National Institute of Justice sponsored this evaluation of the demonstration programs. The evaluation covered program implementation, focusing on the experiences of youths who entered the programs during the first year of operation from April 1992 through March 1993. The goal of the evaluation was to provide systematic information about program design and implementation to the three demonstration sites, to OJJDP, and to jurisdictions considering the development of juvenile boot camps. Impact analyses were beyond the scope of this evaluation. Key Implementation Findings The key findings from this implementation evaluation are highlighted below. Program Design Despite their limited experience with boot camps and a dearth of juvenile programs to model, all three sites met the challenge of implementing the boot camp demonstration. They put together working public-private partnerships, developed and refined a coherent program rationale, and opened their program doors on schedule 6 months after the grant award. Each program incorporated a structured selection process, a 90-day residential boot camp phase, and an aftercare phase of 6 to 9 months. All programs shared the assumption that the boot camp environment would give youths a unique opportunity to make positive changes in their attitudes, values, skills, and behaviors. These improvements would in turn provide them with a solid foundation for educational, work, and treatment activities when youths returned to the community. Each of the sites adopted a slightly different philosophy for its program. Cleveland applied the "normative model," attempting to develop prosocial normative systems during the boot camp and aftercare phases. Military customs and procedures were adapted to the underlying treatment philosophy of Cleveland's private partner, the North American Family Institute. Both Denver and Mobile operated with the philosophy that highly structured and stressful military experience, in concert with positive role models, would promote internal changes in youths. As a result, they would be more receptive to education and treatment in the community. Participants All programs targeted male delinquents, ages 13 and up in Mobile, and 14 and up in Cleveland and Denver. In keeping with OJJDP guidelines, all sites excluded some types of violent offenders. In this respect, Mobile was the most exclusionary and Cleveland the least, allowing entry to some youths whose previous offense was violent but ruling out offenders whose current offense was violent. All programs screened candidates for medical and psychological suitability. Only Cleveland's program was voluntary, and only Cleveland exclusively targeted youths who would otherwise be confined in a State or county facility. During the first year, 119 youths entered Cleveland's boot camp, 76 entered the Denver program, and 122 entered Mobile's camp. Judging from their official records, youths who entered the Cleveland program had previously been more involved with the juvenile justice system and had committed more serious offenses than youths entering the Denver and Mobile programs. In addition, over one-fourth of Cleveland's youths had a prior commitment to State or county facilities. In Denver and Mobile, less than 5 percent of participants had prior State commitments, although many Denver youths had served short sentences in detention facilities, and one out of six Mobile youths had attended a 2- week mini-boot camp. Many participants had background characteristics associated with youths at risk, such as single- parent families, low income, poor school attendance, delinquent siblings or peers, gang involvement, drug or alcohol use, and a record of disciplinary problems at home or school. Participants in Cleveland and Denver, on average, tended to be older than those in Mobile (16.5 years versus 15.6 years). Boot Camp All three sites developed 90-day residential programs that exposed youths to military-like routine, discipline, and physical conditioning, as well as to more traditional rehabilitative components such as remedial education, life skills education and counseling, and substance abuse education. Cleveland and Mobile also offered physical challenge programming. Other key features of the programs included: o A platoon structure in which groups of 10 to 13 youths entered together and were expected to graduate together. o Spartan facilities located on the grounds of an existing institution, but with separate programming. o Onsite drill instructors (DI's), teachers, and case managers. o Intentional selection of some staff with military backgrounds. o An initial staff training program. o Military-style uniforms for youths and DI's and use of military jargon, customs, and courtesies. o A tightly programmed day, starting at 5:30 or 6 a.m. and ending with lights out by 9 or 10 p.m. o Summary (on-the-spot) punishment for some breaches of rules, for example, 25 pushups for inappropriate language. o A progression of sanctions, culminating in a board or commander's authority to remove youths from the program. o A public graduation ceremony. Although they had many common characteristics, the boot camps differed in a number of ways that reflected their distinctive philosophies and rationales. Cleveland's program, modeled after other residential programs operated by the same organization but with a "military overlay," placed the least emphasis on military structure, regimentation, and physical exercise. It also was the only site to make therapeutic counseling a central part of its program. Denver, in contrast, deliberately avoided traditional treatment modalities and created the most military-like setting. Compared with the other sites, Denver spent the most time on military drill, fitness, and work and devoted the least time to remedial education and life skills curriculum. Mobile placed moderate emphasis on military aspects, and, of the three sites, allocated the most programming time to education. Most participants committed at least one infraction serious enough to be recorded for the evaluation team. These infractions typically involved disobedience or disrespect, improper language or conduct, or a fight with another youth. The most serious infractions, escape and assaults on staff, were less common. There were 34 escapes attempted in Mobile, 6 in Cleveland, and 8 in Denver. Cleveland reported 17 assaults on staff and Mobile 3; none were reported in Denver. In keeping with their greater emphasis on military aspects, Denver and Mobile often ordered brig or isolation time for misconduct, while Cleveland did not. The sanction most commonly applied in Cleveland was counseling. Boot camp completion rates for youths admitted during the first program year were high, ranging from 94 percent in Cleveland to 82 percent in Mobile and 80 percent in Denver. The majority of youths who left the camps were terminated for noncompliance or escape. In Denver, however, a third of its dropouts left for medical reasons. In Cleveland and Denver, the majority of the boot camp dropouts occurred within the first 2 weeks. Interviews with a small sample of youths from early platoons suggested that they liked the physical conditioning and drill elements of boot camp, but they found it hard to adjust to boot camp rules and discipline, especially during the first month of the program. Implementation of the boot camps proceeded relatively smoothly at all three sites. However, two sites, Cleveland and Mobile, experienced high staff turnover. Staff attributed the turnover to staff burnout from the intensity of the program and to low salaries. In general, program staff felt that it was difficult to achieve the right balance between military orientation and sensitivity to the special characteristics of juvenile delinquents. They also found it difficult to train replacement staff adequately without disrupting program operations. In addition, boot camps struggled to find appropriate and effective disciplinary measures and a clear-cut termination policy. Both Cleveland and Mobile developed a "setback" system, which usually forced youths being disciplined to graduate later than expected. Mobile used this sanction for almost a third of first-year participants, while Cleveland used it for only 8 percent of its youths. Aftercare The aftercare components of the three demonstration programs were more diverse than the boot camps. Denver's 6-month program was the shortest, and Mobile's 9-month program was the longest. Cleveland's youths could be released at 6 months or kept in the program for up to 8 months. Youths lived at home or in other approved placements and were required to attend the aftercare programs. Cleveland and Denver created aftercare centers specifically for boot camp graduates, whereas Mobile "mainstreamed" boot camp graduates to seven local Boys and Girls Clubs. Denver's center was an academy-style school, with nonacademic services such as drug counseling obtained by referral to other programs. In contrast, Cleveland's center was the hub of most counseling and supportive services other than school. Mobile's youths were expected to participate in existing afterschool and evening programs at their assigned Boys and Girls Clubs. For the most part, aftercare represented an abrupt break from the military discipline and regimentation of the boot camp phase. Most of the military elements were abandoned, with the exception of a few military-style courtesies and titles, the inclusion of some military leadership training materials in the curriculum used in Cleveland and Denver, and continuation of regular physical training in Denver. In general, Cleveland provided the most comprehensive aftercare services. As in boot camp, counseling was a core element of the program. Unlike the other two programs, Cleveland employed a staff member to work with families and a full- time vocational services counselor. All three programs organized community service activities but none had a specific restitution component. Case managers monitored restitution payments for youths who had restitution orders in Denver and Mobile. Restitution was generally not ordered for youths committed in Cleveland. During the aftercare phase, arrangements for day- to-day supervision and case management varied in complexity. In Cleveland, both functions were handled by the aftercare center staff. In Denver and Mobile, youths were supervised daily by the staff at Wyatt Academy and the Boys and Girls Clubs of Greater Mobile, respectively, but case management usually remained the responsibility of a youth's probation officer or Department of Youth Services case manager. This split in responsibility made it more difficult to identify youths who were not complying with program requirements and to take remedial action. The most persistent disciplinary problem during aftercare was absenteeism. Programs struggled to find appropriate incentives for good attendance, but solutions were not readily available to address absenteeism. Programs also developed policies and procedures for handling other problem behaviors. Threatening staff, extremely disruptive behavior, and committing a serious new crime generally were considered grounds for expulsion. Despite high completion rates during the residential boot camp phase, many youths did not complete aftercare. Of the 112 youths entering the aftercare program in Cleveland, 33 percent were arrested for a new offense before completing the program and 17 percent were terminated for other reasons, including 2 deaths. In Denver, 70 percent failed to complete the program; about half of these youths dropped out, and about half were terminated because they were arrested for a new offense. Mobile reported fewer losses during aftercare; 28 of the youths who entered aftercare there were terminated. However, Mobile's statistics may not be directly comparable. Because its aftercare program was so decentralized, Mobile did not identify persistent absentees as reliably as did the other sites. Most of the youths who did leave Mobile's aftercare were arrested for new offenses. All programs experimented with their aftercare programs and made constant adjustments. Both Cleveland and Mobile significantly modified aftercare in the second year. Because many first- year youths had trouble rejoining regular school programs in midstream, Cleveland added an alternative school to its center to serve as a transitional point for youths who wanted to return to a regular school. Because there was no other opportunity to keep the boot camp peer group together after graduation, Mobile added a weekly session and one monthly activity that brought boot camp graduates together in aftercare. Short-term Effects Consistent with the program rationales, there were indications that youths improved their educational performance, physical fitness, and behavior during boot camp. In Cleveland and Mobile, where boot camp staff tested educational achievement of participants upon entry and at graduation, the average youth gained one grade level or more during the 90-day residential phase. In Cleveland, the only site reporting sufficient data on changes in physical fitness, youths also made significant gains on fitness tests. Anecdotal data and observations suggest that physical gains also occurred at the other sites. Drill instructor ratings of participant performance in several areas, including respect for authority, self- discipline, team work, and personal appearance, also improved substantially over the course of Denver and Mobile's boot camps. Ratings improved less in Cleveland, although they moved in the desired direction. Youths who remained in the program for at least 8 months (3 months of boot camp phase and 5 months of aftercare) reported positive changes in their attitudes and behaviors. These findings were based on a small sample, however, and despite promises of confidentiality, responses may have been biased toward socially desirable answers. Costs Only crude estimates of program operating costs were possible. However, these estimates showed that daily costs for the residential phase of the program, assuming full occupancy, ranged from $66 in Mobile to $75 in Cleveland. The sites indicated that these costs are lower than the daily costs of confinement in State or local facilities. Aftercare costs differed much more, reflecting greater diversity in the aftercare programs. In short, the three grantees developed viable boot camp programs that fell within the broad parameters laid out by OJJDP. There was variation across sites in terms of the characteristics of participating youths, the type of military environment, the service emphasis, and the aftercare program structure and services. At the conclusion of the boot camp phase, empirical and anecdotal evidence of positive changes in the skills, behaviors, and attitudes of participants was found. During aftercare, program completion became a problem, prompting considerable revisions to programming and structure in Cleveland and Mobile. Mostly because of heavy attrition in aftercare, overall program attrition amounted to 53 percent in Cleveland, 76 percent in Denver, and 41 percent in Mobile. Conclusions Can the boot camp model be adapted to the juvenile justice system? The experiences of the three demonstration programs indicate that residential boot camps can be implemented in the juvenile justice system. Adult boot camp models require adaptations to fit the requirements of the juvenile system, including additional educational programming and some altering of disciplinary procedures. The programs did not always foresee the corrections required, but by the conclusion of the evaluation period, each had adjusted adult boot camp designs to fit its juvenile environments. The programs were developed with the assumption that boot camp should stimulate changes in youths' attitudes, skills, and behavior, and that those changes would, in turn, increase self-sufficiency in the community and decrease crime and delinquency. As part of the implementation evaluation, it is important to note several important observations about the front end of this logic chain. The first criterion for measuring the success of the programs is whether youths stayed in the program long enough to accrue any personal benefits. At all three sites, the majority of youths successfully graduated from boot camp. Cleveland graduated 95 percent of its entrants, and both Denver and Mobile graduated 80 percent or more. A second indicator of successful program implementation is whether the boot camp stimulated changes in attitudes and behaviors. Both empirical and anecdotal evidence exists that first-year boot camp youths made gains in physical fitness, education, and attitude over the 3-month period, at least at some sites. In the absence of a control or comparison group, these changes cannot be attributed to the boot camps. Nor can judgments be made about the significance of the observed changes. The fact that youths in both Cleveland and Mobile on average increased their educational grade level by over one year may, indeed, reflect the experience of youths in other institutions. In fact, other studies have documented educational achievements of this magnitude, although perhaps over a longer period of institutionalization. What can be said is that youths who participated in boot camp did not appear to be worse off in any measure of educational or behavioral performance at the conclusion of their training. As they were poised to leave boot camps, the youths whom we interviewed expressed confidence that they had made considerable changes in their lives and that their behavioral trajectory would be different from what it had been prior to boot camp. What appeared to be a promising prognosis at the conclusion of boot camp disintegrated during aftercare. All three programs were plagued by high attrition rates for noncompliance, absenteeism, and new arrests during the aftercare period. No other indicators of progress were observed during this phase that would help pinpoint where the problems lay. In all fairness to the programs, aftercare was particularly affected by unexpected cuts in Federal support, especially in Denver and Mobile, where reductions resulted in programs far less comprehensive than originally planned. The programs expressed considerable concern about making aftercare services more effective and were still seeking expert advice and adjusting this component when the evaluation came to a close. However, at this juncture it does not appear that the demonstration programs solved the problem that typically plagues residential correctional programs: inmates who appear to thrive in the institutional environment but falter when they return home. This evaluation cannot draw conclusions about the long-term impact of the program. The study did not track postprogram recidivism, and it collected data on inprogram recidivism only to the extent that a new arrest prompted termination from the program. Of youths entering the boot camps during the first year, 32.8 percent in Cleveland, 25 percent in Denver, and 11.5 percent in Mobile were removed for new arrests. Without knowing what the arrest rates would have been for a control group of comparable youths, it is difficult to interpret what these attrition rates mean. However, it is possible that more information on these issues may be forthcoming in 1996, under the auspices of research funded by OJJDP. Neither can the programs' impact on correctional crowding or cost savings be assessed without more information about recidivism and the costs of alternative placements. Juvenile programs cannot be expected to have the same potential to reduce costs as adult boot camps. Adult boot camps generally serve as a 4- to 6-month alternative to a much longer sentence, measured in years rather than months. Juvenile sentences are much shorter, thereby reducing the savings in days and costs that juvenile camps can produce. In fact, Cleveland was the only program that set reductions in institutional crowding as a goal. But because the Cleveland program was so small in relation to the size of Ohio's juvenile correctional population, the effects are likely to be marginal. Did the boot camp programs select appropriate youths? As conceived by OJJDP, the boot camp program was to have been an intermediate sanction, designed for intermediate offenders who did not merit long-term institutionalization but were considered too serious for diversion or a probation sentence. During the first year of implementation, Mobile probably erred on the side of including too many youths for whom the less restrictive and costly probation sanctions would have been appropriate, including some youths whose most serious offense was a violation of probation for a status offense. Cleveland and Denver, on the other hand, may have erred too far on the side of including serious offenders. Cleveland's youths averaged 2.8 felony convictions, and 27 percent had been sentenced to an institution before boot camp. Denver's program also had a number of youths who had previously been incarcerated. The selection criteria and procedures generated a sufficient number of eligible youths, and the programs made few changes to the criteria. A few medical problems that surfaced in the first few cohorts, however, led to greater attention to the camps' medical and psychological screenings. Judging by boot camp completion rates, these criteria successfully screened youths who could withstand the boot camp regimen. It is not so clear whether the criteria successfully identified youths who would benefit most from the program. In particular, the finding that youths with prior incarceration experience were less likely to complete the boot camp and aftercare components raises questions about the appropriateness of this group for boot camps. Staff in Cleveland recognized this pattern and expressed concern about the ability of the program to deal with those with prior incarcerations. Given the brevity of the observation period and the shortcomings of aftercare programs during their early implementation months, not enough information is available at this juncture to translate this finding into a recommendation. However, it deserves additional research and experimentation on the part of the programs, particularly in light of staff uneasiness about their ability to work with youths with longer criminal histories. Did the programs' military overlay contribute to their effectiveness? The differential effects of the military aspects of the program cannot be disentangled from other components. However, it is known that most program staff perceived the military aspects as important, if not essential. An analogy to the military boot camp is helpful in assessing the viability and limits of the military model for corrections. In the military, boot camp is the initiation process that attempts to mold a group of individuals into a cohesive team that will respond without question to authority. Ultimately, the conformity and respect established during boot camp may come into play in life- threatening situations. When a young person leaves a military boot camp, he remains in the service for a minimum of 2 years. During this time, the same structure and rules of behavior that guided his behavior in boot camp are in place. In addition, that person has a job, clothing, and a place to live. For many individuals, the military becomes a family of sorts, and a source of structure and support. Contrast this model with a correctional boot camp. In a correctional camp, the focus is less on building a team and more on changing an individual's behavior and teaching cooperation. Unlike the military, when a youth leaves a correctional boot camp, he leaves its structure and support and way of life and returns to an environment that has already failed him. Because of these differences, there are clearly limits to what a correctional boot camp can be expected to accomplish. The demonstration programs employed a military framework in tandem with strong educational and skill-building components. Rather than serving as a good in and of itself, this structure establishes the discipline and control that enable youths to focus on constructive changes and accomplishments. The appropriate analogy may be a military school rather than a military boot camp, where the military context serves as a setting in which to offer educational services. This is not to say that the military aspects of the program are unimportant. The fact that these "boot camps" evoked images of a tough, macho environment may have been essential for acceptance, because some youths may be more accepting of "treatment" when it is delivered from within a military structure. Recommendations Boot camp programs need to clarify the cause and effect rationales they are using to explain expectations for changing participant behavior. Imparting a clear understanding of the underpinnings of a program is critical to its smooth operation, particularly when the idea is novel, as is the concept of a boot camp for juveniles. The Cleveland program carefully delineated a rationale describing how different program activities would affect youths and what they hoped to accomplish. Having a well-developed rationale was an enormous advantage for both Cleveland's program directors and staff. In making day-to-day decisions about how to react to the multitude of behaviors and questions encountered in a new program for juvenile offenders, it is very useful to be able to step back and ask, "What are we trying to do?" In addition, the process of describing how the program will work may identify inconsistencies in the chain of logic. Boot camp programs should carefully define and select target populations in light of their goals for rehabilitation, recidivism, cost containment, punishment, and other important areas. OJJDP intended the demonstration to target intermediate offenders. Given the costs of these programs, it would make sense that they be reserved for youths who have already failed in less costly placements such as probation. Whether boot camps should exclude high-end or violent offenders is an open question, and an issue for which these programs can give little guidance. However, preliminary indications are that offenders with previous incarceration experience do not perform well in demonstration settings. High-end offenders would undoubtedly impose additional burdens that these programs did not face, including the need for additional security and political opposition from those who see boot camp placement as insufficiently punitive. Because aftercare was the period during which most attrition occurred, this part of the demonstration program clearly needs to be improved and perhaps restructured. During the evaluation, a number of changes were made to the programs' aftercare phases, and their effects are as yet unknown. Presented below are recommendations for improving the programs' aftercare services, based primarily on reports from boot camp staff and assessments made by correctional and court officials. o Programs need to identify strategies for smoothing the transition from boot camp to aftercare. Aftercare designs should expressly build on the structure, discipline, and learning experiences of the programs' boot camps. o The transition from a closely monitored boot camp to aftercare should be less abrupt, permitting youths to adjust more slowly to the community. There was general consensus among program staff that the requirements for participation in aftercare should be stiffer and that youths should be more closely monitored, particularly in the period immediately following release. The development of the transitional academy in Cleveland, the revisions to Mobile's aftercare program, and the decision to recycle youths to boot camp from aftercare were, in part, an expression of that consensus. Closer supervision not only reduces the time youths have to get into trouble, it also provides more program exposure and more treatment. A number of staff mentioned that a halfway house might help in this respect. o It is imperative that programs develop clear sets of disciplinary rules, expectations for behavior, and graduated sanctions for both boot camp and aftercare. Although the programs had tightened and refined their rules and procedures for boot camp, they still had not developed hard and fast rules about acceptable behavior in aftercare. What to do about nonattendance was a particularly thorny issue. Aftercare staff wanted youths to follow the rules and believed that some sanctions should be applied for misconduct and absences, but they feared that harsher sanctions might drive youths into further nonattendance. These issues were never satisfactorily resolved. In programs where multiple agencies may be involved in monitoring youths, the responsibilities of each agency need to be spelled out in detail. If youths are to be held accountable and if positive changes begun in boot camp are to be reinforced, there must be a clear understanding of who is responsible for the youths and when sanctions should be applied to keep them focused on the program. Delineating responsibilities is particularly important if the aftercare program is drawing on the capabilities of a variety of agencies, a situation in which staff are especially susceptible to a "pass-the- buck" mentality. In light of the high staff turnover rates experienced by two of the programs, there is a continuous need for staff training. Intensive, offsite training is frequently too expensive and may not be available exactly when needed. Particularly for very specific training, such as certification as a challenge course instructor, turnover can become a problem if the staff capable of conducting that training leave the program. New programs must anticipate high staff turnover rates and build into workplans training at frequent intervals. The boot camp concept warrants additional attention and research. A comparison of boot camp and control youths would yield important insights into how boot camps perform relative to other sanctions. Moreover, not enough is known about boot camps and their effectiveness to support the endorsement of a single model for wide implementation. A series of studies that systematically looks at the effects of variations in boot camp and aftercare design would be highly desirable. The demonstration programs evaluated in this study demonstrated that boot camps can be implemented in the juvenile justice system. Whether those boot camps work is still unknown. That question merits further research. ---------------------------------- Bibliography American Correctional Association. (1993). Directory: Juvenile and adult correctional departments, institutions, agencies and paroling authorities. Laurel, MD: Author. Buros, O.K. (1993). Mental measurements yearbook. Highland Park, NJ: The Gryphon Press. Buros Institute of Mental Measurement. (1992). The eleventh mental measurements yearbook. Jack J. Kramer and Jane Close Conoley (Eds.). Lincoln, NE: University of Nebraska Press. Byrne, J.M., A.J. Lurigio, and J. Petersilia. (1992). Smart sentencing: The emergence of intermediate sanctions. Newbury Park, CA: Sage Publications. Bureau of Justice Statistics (1994). Sourcebook of criminal justice statistics--1993. Washington, DC: U.S. Department of Justice. California Youth Authority. (1992). LEAD program. Sacramento, CA: Author. Cronin, R. (1994). Boot camps for adult and juvenile offenders: Overview and update. Washington, DC: U.S. Department of Justice. Federal Register, Vol. 55, No. 134, July 12, 1990, pp. 28,718-28,726. Washington, DC: U.S. Government Printing Office. Kurtz, W., and Giannelli, P. (1989). Ohio juvenile law (2nd Edition). Ohio: Banks-Baldwin Law Publishing Company. MacKenzie, D.L., and D. Parent. (1992). Boot camp prisons for young offenders. In James M. Byrne, Arthur J. Lurigio, and Joan Petersilia (Eds.), Smart sentencing: The emergence of intermediate sanctions. Newbury Park, CA: Sage Publications. Martinson, R. (1974). What works? Questions and answers about prison reform. The Public Interest, 35. Morris, N., and M. Tonry. (1990). Between prison and probation: Intermediate punishment in a national sentencing system. New York: Oxford University Press. Office of Justice Programs. (1990). Survey of intermediate sanctions. Washington, DC: National Institute of Justice. Office of Juvenile Justice and Delinquency Prevention. (1990). Boot camps for juvenile offenders: Constructive intervention and early support--package of information. Washington, DC: U.S. Department of Justice. Office of Juvenile Justice and Delinquency Prevention. (1992). National juvenile custody trends, 1978-1989. Washington, DC: U.S. Department of Justice. Ohio Revised Code, December 3, 1986. Palmer, T. (1992). The re-emergence of correctional intervention. Newbury Park, CA: Sage Publications. Parent, D. (1989). Shock incarceration: An overview of existing programs. Washington, DC: U.S. Department of Justice. Parent, D.G., V. Lieter, S. Kennedy, L. Livens, D. Wentworth, and S. Wilcox. (1994). Conditions of confinement: Juvenile detention and corrections facilities. Research Report. Washington, DC: Office of Juvenile Justice and Delinquency Prevention. Scholossman, S.L. (1983). Juvenile justice: History and philosophy. In S.H. Kadish (Ed.), Encyclopedia of crime and justice. Vol. 3, pp. 961-969. New York, NY: Free Press. Summary of solicitation workshop proceedings, October 1990. Thornton, D., et al. (1984). Tougher regimes in detention centers: Report of an evaluation by the young offender psychology unit. London: Her Majesty's Stationery Office. Toby, J. (1982). The crisis in our prisons and what we can do about it. New York Times: May 2. ______ (1984). A higher price for lesser crimes: Punish nonviolent offenders with weekend public work. Los Angeles Times: February 24. Toby, J., and Pearson, F.S. (1992). Appendix 1: Juvenile boot camps, 1992. In Institute for Criminological Research and American Institutes for Research, Boot camps for juvenile offenders: Constructive intervention and early support-- implementation evaluation. New Brunswick, NJ: Rutgers University. U.S. General Accounting Office. (1993). Prison boot camps: Short-term prison costs reduced, but long-term impact uncertain. Washington, DC: Author.