Title: Creating a New Criminal Justice System for the 21st Century: Findings and Results From State and Local Program Evaluations. Series: Monograph Author: Bureau of Justice Assistance Published: April 2000 Subject: Criminal Justice System, Criminal Justice--general 104 pages 233,000 bytes ---------------------------- Tables are not included in this ASCII plain-text file. To view this document in its entirety, download the Adobe Acrobat graphic file available from this Web site or order a print copy from BJA at 800-688-4252. ---------------------------- U.S. Department of Justice Office of Justice Programs 810 Seventh Street NW. Washington, DC 20531 Janet Reno Attorney General Daniel Marcus Acting Associate Attorney General Mary Lou Leary Acting Assistant Attorney General Nancy E. Gist Director, Bureau of Justice Assistance Office of Justice Programs World Wide Web Home Page www.ojp.usdoj.gov/BJA Bureau of Justice Assistance World Wide Web Home Page www.ojp.usdoj.gov/BJA For grant and funding information contact U.S. Department of Justice Response Center 1-800-421-6770 This document was prepared by Justice Research and Statistics Association, under grant number 95-DD-BX-K011, awarded by the Bureau of Justice Assistance, Office of Justice Programs, U.S. Department of Justice. The opinions, findings, and conclusions or recommendations expressed in this document are those of the authors and do not necessarily represent the official position or policies of the U.S. Department of Justice. ---------------------------- The Bureau of Justice Assistance is a component of the Office of Justice Programs, which also includes the Bureau of Justice Statistics, the National Institute of Justice, the Office of Juvenile Justice and Delinquency Prevention, and the Office for Victims of Crime. ---------------------------- Creating a New Criminal Justice System for the 21st Century: Findings and Results From State and Local Program Evaluations Effective Programs Monograph No. 2 April 2000 NCJ 178936 ---------------------------- Foreword Improving the nation's criminal justice system is the central mission of the Bureau of Justice Assistance (BJA). Building safer, less violent communities is a major challenge all states and local communities are facing. Real progress can be achieved only if we demonstrate and confirm "what works," so we can all profit from the impact of more than 10 years of federal funding through the Edward Byrne Memorial State and Local Law Enforcement Assistance Formula Grant Program. In the year 2000, our knowledge of effective programs will continue to be critical as we solve one of the nation's toughest problems. Underlying our ability to gain reliable knowledge and disseminate information on effective programs and activities is our commitment to evaluating and publicizing those programs and activities as we implement our plans and strategies. Evaluation done right permits program managers to find out what is and is not working, and why. BJA carries the responsibility of building the capacity of state and local governments to design and conduct their own assessments and evaluations of criminal justice programs. BJA's handbooks and technical assistance help to provide a systematic, disciplined framework that focuses on performance and results, both quantitative and qualitative, to be measured and evaluated. This initiative also represents an area of close participation with the National Institute of Justice to build federal, state, and local evaluation systems over the past decade. BJA is pleased to present this monograph, Creating a New Criminal Justice System for the 21st Century: Findings and Results From State and Local Program Evaluations. It is the second in a series of reports to highlight and document approaches and results of evaluations funded at state and local levels. The first report, Improving the Nation's Criminal Justice System: Findings and Results From State and Local Program Evaluations, Effective Programs Monograph No. 1, was published in December 1997. Six demonstration projects affecting many components of the criminal justice system were the focal point of the evaluations presented in the previous monograph. Having been identified as effective, these programs have become models for other states and localities to replicate. This monograph has been produced to make proven state and local programs more accessible to planners and practitioners alike. The document is divided into two parts. Part One provides descriptions and evaluations of programs in seven states: Pennsylvania, Virginia, New Hampshire, Oregon, Illinois, Utah, and Oklahoma. Part Two is a summary of evaluations of state and local multijurisdictional task forces funded by BJA. Future reports in the series will continue to communicate the results of strong federal, state, and local partnerships that are enhancing the role of evaluation, building excellent evaluation systems, and reporting on the impact of efforts to combat crime in America. BJA is proud of its work of identifying effective drug abuse and violent crime prevention programs and communicating lessons learned at state and local levels that can be shared nationally to ensure that the most promising approaches have a broad impact. Nancy E. Gist Director ---------------------------- Acknowledgments This second monograph on effective state and local program evaluation is a product of a cooperative effort by the states and the Bureau of Justice Assistance as part of the State Evaluation Development Program, coordinated by the Justice Research and Statistics Association (JRSA). The State Evaluation Development Program relies on the expertise of the State Planning Agencies to ensure the success of its publications. The following State Planning Agencies contributed their knowledge and time to make this monograph a success: o Pennsylvania Commission on Crime and Delinquency. o Virginia Department of Criminal Justice Services. o New Hampshire Office of the Attorney General. o Oregon Department of State Police, Criminal Justice Services Division. o Illinois Criminal Justice Information Authority. o Utah Commission on Criminal and Juvenile Justice. o Oklahoma District Attorneys Council. The Justice Research and Statistics Association prepared this document under the direction of Joan C. Weiss, Executive Director. JRSA staff who worked on this publication, under the supervision of Michael Connelly, Director of Special Projects, include Kate Ahern, Craig Cussimanio, Nancy Michel, Laura Parisi, Marylinda Stawasz, and Kate Wagner. ---------------------------- Contents Part One: State Program Evaluations o State and Local Evaluations and the Effective Programs Initiative o Pennsylvania: School-Based Probation o Virginia: Detention Center Incarceration Program o New Hampshire: Merrimack County Adult Diversion Program o Oregon: The Multnomah County STOP Drug Diversion Program o Illinois: Homicide and Violent Crime Strike Force in Madison and St. Clair Counties o Utah: The Utah Day Reporting Center--Success With Alternative Incarceration o Oklahoma: Drug Court Program o References Part Two: Multijurisdictional Task Force Evaluation o Multijurisdictional Task Forces: Synthesis of State and Local Evaluation Findings o Multijurisdictional Evaluation Findings o References Appendix A o Identifying Effective Criminal Justice Programs: Guidelines and Criteria for the Nomination of Effective Programs Sources for Further Information ---------------------------- Part One: State Program Evaluations State and Local Evaluations and the Effective Programs Initiative The Bureau of Justice Assistance's (BJA's) evaluation strategy is designed to determine the effectiveness and impact of BJA's grant programs. The goal is to confirm whether performance objectives established by the states are being achieved and, if they are, what critical elements were responsible for success. Hence, the overall goal of the evaluation program is to identify programs of proven effectiveness so that they can be publicized and replicated in other jurisdictions. By building strong assessment and evaluation foundations in the 50 states and 6 territories, BJA can account for the efforts of programs funded by its grants and add to the knowledge of what works and what does not work throughout the criminal justice system. BJA relies on data and the results of research and evaluation to monitor program development under the discretionary program and provide guidance and model programs under the formula and block grant programs. BJA uses evaluation results to guide the formulation of policy and programs within federal, state, and local criminal justice agencies and to ensure that policies and funded programs are based on proven results. The required state annual reports are useful in working toward this goal, but BJA intends to go further to capture both quantitative and qualitative measures of program performance. As part of a continuing effort to provide the criminal justice community with improved access to information on successful programs dealing with problems of drug abuse and/or violent crime, BJA is publishing this second volume highlighting innovative state and local programs. The evaluations, as well as the programs being evaluated, reflect the results of program development and implementation activities funded under BJA's Formula Grant Program to state and local governments and organizations. Lessons learned at state and local levels can be shared nationally to ensure that the most promising approaches have a broad impact. In addition to wanting to know about successful initiatives, justice system planners and managers need to understand the scope and level of effort required for innovative approaches. Readers of this report will observe that state and local agencies are actively funding, implementing, and evaluating a broad range of programs to deal with drug abuse and violent crime. Enhancing state and local assessment and evaluation capabilities is a very high priority. The success of federal, state, and local partnerships yielded many results within the states. The Effective Programs Initiative is designed to develop or enhance drug control and system improvement strategy performance monitoring, measurement, and evaluation capacities in the states and territories. BJA believes that setting standards for evaluation will have major benefits, while informing the public about how federal dollars are spent. Evaluation requires establishing new partnerships among funding agencies, program managers, and evaluators. No longer the sole domain of academic researchers, evaluation is making a real difference in the ability of the criminal justice community to base policy and decisions on accurate and useful information. This is the result of the strong relationships formed between evaluators and practitioners. In response to the Attorney General's charge to "find out what works and spread the word," BJA began this new initiative with the goal of creating a mechanism for enhancing the design, implementation, measurement, evaluation, and dissemination of information on programs in high-priority program areas. The objectives of the Effective Programs Initiative are to: o Enhance the ability of state and local agencies to generate and use evaluation results for strategy development, program improvement, and effective program identification. o Identify and document useful approaches for designing and conducting evaluations at state and local levels. By identifying effective state and local criminal justice programs, practices, and products and disseminating this information to policymakers and practitioners, BJA will help to improve the criminal justice system at the national level. Through this approach, which might be called "leading by example," information on successful programs will be disseminated to the field in a credible and timely fashion. The assessment and evaluation results from the 56 laboratories (50 states and 6 territories) put in place under the Byrne Formula Grant Program are the products of this initiative. ---------------------------- By identifying effective state and local criminal justice programs, practices, and products . . . BJA will help improve the criminal justice system at the national level. ---------------------------- See appendix A, Identifying Effective Criminal Justice Programs: Guidelines and Criteria for the Nomination of Effective Programs, for instructions on submitting potential programs. Once the effective programs have been approved, additional BJA monographs and bulletins will be published. Identifying and promoting sound programs is essential to developing effective strategies at federal, state, and local levels. BJA wants to enhance the criminal justice system in general, while recognizing the many exceptional state and local advances that have been made in combating violent crime and drug abuse through the use of federal funds. The National Institute of Justice (NIJ) is an active participant in BJA's evaluation program. BJA and NIJ develop evaluation guidelines and conduct comprehensive evaluations of selected programs receiving Byrne Discretionary Grant Program and Formula Grant Program funds. BJA worked closely with NIJ to develop the guidelines and criteria for documenting the seven program evaluations presented in this monograph. A program development model that tests innovative state and local ideas in action and then shares the information gained with the broadest audience has proved promising in discovering what works and what does not work at state and local levels. ---------------------------- Pennsylvania: School-Based Probation This summary was adapted from the report School-Based Probation in Pennsylvania, which presents evaluation research conducted by David S. Metzger, Ph.D., Principal Investigator, and Danielle Tobin-Fiore, B.S., Project Coordinator, the University of Pennsylvania Center for Studies of Addiction. School-based probation (SBP) is an approach to supervising youth that shifts the primary location of probation operations to the school environment. The first SBP program was established in 1990 in Lehigh County with pilot funding provided by the Juvenile Court Judges Commission (JCJC). Since its inception, the program has expanded rapidly, supported by grants from the Pennsylvania Commission on Crime and Delinquency (PCCD). By the end of 1995, SBP programs were in place in more than 40 counties. ---------------------------- The SBP program was developed in response to the recognized need for closer communication between probation officers and school staff. ---------------------------- Program Overview Traditionally, juvenile probation officers in Pennsylvania have been based in county offices, often located in the county courthouse. Under this model, juveniles are seen by their probation officers in the county office, at home or in school during periodic visits, or in various other community locations. Consequently, contact and "supervision" most often occur in brief, planned encounters with defined purposes. The SBP program was developed in response to the recognized need for closer communication between probation officers and school staff. Although the shift in location to the school environment is rather simple, it has significant "systems" implications. SBP involves the integration of the juvenile justice system with the educational system at the local level and is believed to enhance both. Goals and Objectives The goals and objectives of the program are to: o Reduce disciplinary referrals in school. o Reduce the frequency and length of detentions. o Improve attendance and academic performance. o Decrease dropout rates. o Reduce recidivism and out-of-home placements resulting from delinquent behavior. Program Activities/Components SBP is a community-focused approach in which a juvenile probation officer is housed in the school building to provide direct supervision of juvenile offenders on probation. This allows the officer to spend more time on the case and less time traveling and doing intake paperwork. The configuration provides more contact with the offender's family and gives the probation officer more involvement in the student's life and the opportunity to routinely observe the youth in his or her peer group and social environment. Two basic strategies are used by probation officers to manage cases assigned to SBP--single case management and dual case management. In the single case management approach, completion of all work required for an assigned case is the responsibility of the school-based officer. In the dual approach, the responsibility for the case is shared with other probation officers; the division of labor is intended to allow the school-based officer to remain in the school while "nonsupervision" activities are completed by other probation staff. Performance Measures and Evaluation Methods The overall objective of the SPB program evaluation was to build a foundation upon which future experimental studies, designed to objectively evaluate program outcomes, could be constructed. To this end, a series of descriptive studies were completed between January 1996 and July 1997. ---------------------------- The overall objective of the SPB program evaluation was to build a foundation upon which future experimental studies, designed to objectively evaluate program outcomes, could be constructed. ---------------------------- Phase I The first phase of the evaluation focused on the production of a demographic profile of the youth who had been assigned to SBP. Data from existing PCCD SBP reporting forms were linked with the JCJC statistical card database, allowing the profile to include both demographic and arrest data. Descriptions of cases assigned to SBP were derived from SBP reporting forms compiled for youth who completed probation in 1993, 1994, and 1995--4,159 cases from 31 counties. All PCCD-funded SBP programs are expected to forward completed forms to PCCD annually on all youth completing probation during the reporting period. Between 1993 and 1995, 43 counties had been awarded grants to support 5,398 cases. The SBP reporting forms include basic identifying and demographic information (e.g., name, gender, race, date of birth, date of assignment to SBP) as well as performance characteristics such as school attendance, academic performance, in-school and out-of-school suspensions, and enrollment status at the end of SBP. A database was created from all valid forms received from PCCD. To ensure the integrity of the database, several steps were completed in the review of the form and during the entry processes. Prior to data entry, each form was screened for completeness, legibility, and validity (i.e., that values fell within valid ranges). All omissions and notations were marked and, when possible, corrected. Data from these forms were then entered into two separate data files and cross-checked for accuracy. All mismatched entries were identified, inspected, and, when possible, rectified. Approximately 694 forms from 170 cases were not able to be entered into the database. (The majority of these were from one county that submitted forms on a quarterly basis for all active cases.) Forms on 4,159 cases were received; of these, 3,913 (94 percent) were determined to be valid and were entered into the database for subsequent analyses. The breakdown by year was: 555 (14 percent) of the cases were assigned in 1993, 1,982 (51 percent) in 1994, and 1,376 (35 percent) in 1995. A number of cases from counties that did not return valid reporting forms were omitted from the database. Thus, analysis of the data that were included should be interpreted with some caution because of the possibility of selection bias. These data did, however, represent 93 percent of cases and collectively formed the largest existing database on juveniles assigned to SBP; as such, the data provide an opportunity to gain some important insights into the characteristics of the youth assigned to SBP in Pennsylvania. While the descriptive data noted above provide information on the demographic characteristics of the youth served by SBP programs, they do not address the important issue surrounding the nature of the offenses that brought the juveniles into the system--are the charges in cases assigned to SBP programs different from those assigned to other forms of probation? To compare the criminal justice characteristics of SBP cases with the characteristics of the cases assigned to "traditional" probation, data from the SBP reporting forms were matched with data from the "statistical card" database. A form known as the statistical card must be completed for each youth who enters the Juvenile Court System in the Commonwealth of Pennsylvania; it contains information on the youth's involvement with the juvenile court system, including the nature of the offense that brought him or her into the SBP program. The form is completed by the county staff and forwarded to the Center for Juvenile Justice Training and Research for entry into a uniform database that includes important information regarding charges and dispositions for more than 30,000 cases annually. The matching process was complicated by several factors, but most significantly by the lack of a uniform identification number on both the statistical card and the PCCD form. This necessitated scanning by name. Minor differences in spelling or the use of different versions of first names, errors in data entry, and other differences between the two databases further contributed to the difficulties in matching, even by name. Given these challenges involved in achieving matches between the two databases and the size of the statistical card databases, the search was restricted to the 1993 database, the first year of the program. A total of 451 cases in the statistical card database were matched to SBP cases, which represented 81 percent of the total number of 1993 SBP cases with valid PCCD forms. Phase II In the second phase of the evaluation, site visits were conducted to develop an operational understanding of the programs delivering SBP in the Commonwealth of Pennsylvania. The goal of these site visits was to better understand how programs differ and to help identify specific program features that may affect the youth they serve. To accomplish this goal, comprehensive interviews were completed during site visits scheduled between March and July 1996. Each county that had been implementing the program for at least 1 year was scheduled for a site visit. This resulted in visits to 29 (89 percent) counties with PCCD-funded SBP programs operating for more than 1 year. With input from the project's technical consultants and advisory board, semistructured interview guidelines were developed for each of the three respondent groups: probation officers, school administrators, and juveniles assigned to SBP. Each interview was designed to collect both objective information about program operations and subjective data regarding perceptions of program performance. The interviews were conducted by technical consultants and a team of six interviewers; the latter participants were probation officers who were selected on the basis of their experience and training. Prior to the onsite visits, the interviewers participated in a 1-day training session during which each interview item was reviewed and discussed to assess its intent and the method of questioning. In addition to being asked about how they spent their time and what type of case management system they used, probation officers were also asked about how their role was perceived by others in their work environment--school faculty, administration, their cases, the parents and guardians of their cases, and the community at large. Officers were asked to select the role that best described their view of how they were seen by members of these other constituency groups. Officers also assessed their own roles. The interviews concluded with a series of questions regarding the officers' views of the effectiveness of the SBP program in four key areas--academic performance, school attendance, delinquent behavior, and disciplinary referrals (in-school and out-of-school suspensions). School administrators selected for interviews were those who worked most closely with the SBP officers and thus had responsibility for and familiarity with the program in their school. The administrators were asked to assess the range of involvement of the officers, the performance of the program, and the effectiveness of the program in the same four areas as the SBP officers. A total of 111 youth assigned to school-based probation were interviewed during the site visits. Students completing these interviews were randomly selected onsite by the interviewers from a list supplied by the probation officer prior to the visit. To ensure unbiased selection and adequate representation of both genders, interviewers were instructed to interview the third male and the third female on the list from each school visited. Phase III The third phase of the evaluation was devoted to the completion of a case-control study comparing program impact in 75 randomly selected SBP cases with 75 non-SBP cases matched according to age, gender, race, crime, and county of supervision criteria. Rates of rearrest, placements, and cost of placement were used as outcome measures. For those with multiple charges, the most serious charge was used as the basis for matching. This process, although retrospective in nature, was a strategy designed to identify youth who are equivalent in every way except in the type of probation to which they were assigned. Controls were selected from the statistical card database. In selecting counties for this study, several requirements had to be met. First, potential counties must have had an SBP program in operation since 1994 to allow 18 months of followup on each individual selected for the study. Counties also had to have sufficient numbers of cases to select 25 school-based cases and 25 matched controls. Finally, potential counties needed to have documentation accessible to the study staff. Five counties met all criteria, and three counties were included in the study--Erie, Lehigh, and Somerset. For each of the participating counties, data for 18 months from the date of assignment to probation were examined for both cases and controls. This time interval provided an adequate period of observation during which rearrests, probation violations, and placements would be expected to have occurred. These outcomes were assessed through onsite review of case records of the participating counties, with case data documented on structured recording forms developed by the project staff. The primary outcomes measured in this pilot study were related to reinvolvement with the court. The specific events that were monitored for both cases and controls were: (1) arrests for probation violations and new charges and (2) placements made by the courts. Originally, this study had planned to include data from the schools, including attendance reports, behavioral histories, and academic performance records. However, participating schools had a variety of approaches to the collection, retention, and storage of such data with none of these data elements recorded in a consistent manner by all schools. Program Evaluation Findings and Results Phase I The majority (80 percent) of cases, both SBP and non-SBP, were male. The proportion of female cases assigned to SBP increased over the period of observation--from 16.8 percent in 1993 to 22.2 percent in 1995. The racial distribution in SBP and non-SBP cases was not significantly different. When comparing the ages of SBP and non-SBP youth, important differences can be found, with SBP serving younger youth. This difference is most obvious in the 1995 reporting year in which 58 percent of school-based cases were youth aged 13 to 15, while only 40 percent of the non-SBP cases were from this age group. Since both age and gender showed changes over the reporting period, the relationship between these two variables was evaluated. A significant correlation was identified reflecting a higher representation of female cases among younger age groups. Since the increased representation of female cases was not seen in the general population of cases, it is likely that the younger ages of those being served by SBP account for the increased proportion of female cases. While the proportion of youth assigned to SBP from grades 7 to 9 increased during the study cases, the proportion of cases from grades 10 to 12 declined. Data from the statistical card database reveal that these 451 cases had 1,694 allegations of crimes, 25 percent (n=428) of which were against persons, 44 percent (n=746) related to property, and 4.5 percent (n=77) related to drugs. These allegations resulted in 875 (52 percent) substantiated charges, of which 26.6 percent (n=233) were crimes against persons, 42.3 percent (n=370) were property crimes, and 6.3 percent (n=55) were drug-related crimes. In examining the five most common crimes among these cases, theft was the most common substantiated charge (n=79) followed by simple assault (n=72), receiving stolen property (n=65), burglary (n=43), and conspiracy to commit theft (n=34). Summary These data suggest that the program served a diverse population of youth who were somewhat younger than their non-SBP counterparts. The average age of the youth assigned to SBP was just over 15. Given the association between age and gender found in this data set, the SBP cases were also more likely to be female. With respect to the criminal charges that brought these youth into the juvenile court system, there appeared to be few differences between SBP and non-SBP cases. Nearly identical rates of personal, property, drug, and other crimes were found when the juvenile court data for these two groups were compared. In conducting these descriptive analyses, it became apparent that no existing data systems could be used to monitor even the most basic characteristics of the juveniles assigned to SBP. Despite the fact that a significant amount of probation officer time was devoted to the completion of forms documenting the characteristics of the cases assigned to SBP, the forms were not routinely compiled or reviewed. As currently designed, the retrospective completion of performance data regarding behavior, school attendance, and academic performance yields unusable evaluative data at the aggregate level. The validity and reliability of these data are compromised by a variety of problems. In some situations and locations, information required to complete the form was not available. Methods for completing and submitting the forms were not standardized, which resulted in great variations in procedures for completion of the forms. Consequently, a significant amount of evaluative data could not be used. It was recommended that the data collection system be redesigned into a two-part process. The first form would be completed as the youth begin their school-based probation and the second completed at the close of supervision. Both assessments should report on verifiable information for the same intervals. To maximize the value of these data, a numerical identifier common to the statistical card should be included on the form. Phase II School-Based Probation Officer Interviews The 51 SBP officers who were interviewed had an average of 5.6 years (range=0.3 to 25) of probation experience and an average of 1.8 years (range=0.1 to 5.5) of experience as school-based probation officers. Although they reported an average caseload of 26.7 cases (range=6 to 78), 60 percent of these officers also maintained caseloads that were not school based. Those with only school-based cases carried an average caseload of 29, while those with both carried an average of 31. Seventy-three percent of the officers interviewed reported using the single case management model, in which they complete all work related to each of their assigned probation cases. Officers using this management approach spent an average of 66 percent of their time in the school environment compared with 81 percent for those using the dual case management approach, where casework is shared among probation officers. The differences were statistically significant and translated to officers using the dual case management approach being in school an average of 75 percent of a day per week longer than their single-case cohorts. Probation officers reported spending an average of 48 percent of their time in direct case contact and 18 percent in contact with case collaterals, 10 percent in court, 10 percent in travel, and the remainder in training and intake. Although the median percentage of time spent in school was reported to be 70 percent, the percentage ranged from a low of 25 percent to a high of 95 percent. The percentage of time spent in school correlated to the amount of time spent in direct client contact. The range of activities in which SBP officers reported involvement included: visiting the parents of juveniles participating in SBP (94 percent); participating in the disciplinary decisions of assigned cases (84 percent); attending nonacademic school activities (84 percent); giving presentations in classes and monitoring the lunchroom, hallways, and study hall (76 percent); and serving as active participants in the school's Student Assistance Program (68 percent). Seventy-eight percent of the officers interviewed reported that they had developed, or helped to develop, special programs in their schools, including support groups, tutoring services, and mentoring programs. Drug testing was reported to have been used in probation programs by 86 percent of the SBP officers and electronic monitoring, by 79 percent. The majority of officers saw their primary role as advocating, arranging for, and delivering needed services for their cases (75 percent). Responses revealed inconsistencies between the self-defined role and the perceived role of the officer, with others more likely to define the role of the SBP officer as one of police, security, and surveillance. While four areas--academic performance, school attendance, delinquent behavior, and disciplinary referrals--were viewed as being positively affected by the program, the area perceived to have been affected the most was school attendance. Nearly 50 percent viewed the program as extremely effective in this area. ---------------------------- The majority of officers saw their primary role as advocating, arranging for, and delivering needed services for their cases (75 percent). ---------------------------- School Administrator Interviews Fifty-two school administrators were interviewed during site visits. Like the probation officers, the administrators reported a range of officer involvement within their school environment. Administrators described the officers as having full access to school documentation (academic and disciplinary records) for the cases they supervised. The majority of respondents, 85 percent (n=44), indicated that the officers participated in making decisions regarding formal disciplinary actions to be taken with students on probation but were not overly involved. ---------------------------- Consistent with the ratings of the probation officers, school administrators saw school attendance as the area that had been affected the most. ---------------------------- Although ratings in five areas of performance for the school-based program were extremely positive, ranging from a low of 85 percent to a high of 98 percent, it is important to note that the concerns expressed by those who were not satisfied seemed to reflect a desire for the probation officers to fill a policing function. The positive assessments were reflected in the overall ratings of the working relationship with the school-based officer: 87 percent indicated that they had an excellent relationship and 12 percent indicated they had a good relationship. No one indicated a poor working relationship. Eighty-five percent of the administrators believed the program was such an important part of the school environment that it deserved financial support from the school district. Thirty-three percent believed that their school boards would be willing to provide such support. Consistent with the ratings of the probation officers, school administrators saw school attendance as the area that had been affected the most. SBP Case Interviews Of the 111 SBP youth who were interviewed, 67 percent (n=75) were male and 32 percent (n=36) were female; 65 percent (n=72) were Caucasian, 26 percent (n=29) African-American, and 5 percent (n=6) Hispanic. The average age of these students was 15 years, and the median grade level was ninth. These cases had been under supervision for an average of 9.4 months, and 78 percent (n=86) of those interviewed were on probation for the first time. Thirty percent (n=33) had been in some form of out-of-home placement (e.g., foster home, residential center) prior to being assigned to SBP. Frequency of contact with probation officers reported by respondents was found to be significantly associated with the integration of the officer into the school environment. Twenty percent of the youth (n=22) were required to report to their officers daily; overall, the youth reported seeing their probation officer an average of 2.7 times per week. Respondents were also asked a number of questions about the impact the program had had on their behavior. Unlike the probation officers and the school administrators, these respondents reported that the greatest impact of the program had been on their behavior both in and out of school. ---------------------------- [R]espondents reported that the greatest impact of the program had been on their behavior both in and out of school. ---------------------------- Summary Assigning the probation officer to the school allowed much greater opportunity for the establishment of relationships that facilitated supervision and an understanding of the needs of the youth. The percentage of time spent in the school environment may have been the best indicator of this opportunity. More time spent in the school environment was not only a logical prerequisite for building strong working relationships; it was also statistically associated with the amount of direct case contact. Given that this time in school may have been the defining characteristic of the program, it was recommended that a minimum standard be established for the percentage of time an officer must be present in the school environment for the probation program to be considered school based. The presence of the officer in the school was also perceived as directly responsible for improvements in attendance by the youth assigned to SBP. The attainment of this goal was important not only because school attendance is a prerequisite for academic success, but also because school was the primary location of the probation supervision. Thus, a juvenile attending school was exposed to an educational environment as well as the behavioral controls inherent in the frequent contact with their probation officer. ---------------------------- More time spent in the school environment was not only a logical prerequisite for building strong working relationships; it was also statistically associated with the amount of direct case contact. ---------------------------- Phase III The subjects in the case-controlled study had an average age of 14 years at the time of their assignment to school-based probation. Eighty-six percent of the subjects were male (n=129), 54 percent (n=81) were Caucasian, 23 percent (n=34) were African-American, and 19 percent (n=29) were Hispanic. The characteristics of the school-based cases and control sample were statistically equivalent on all of the basic demographic measures. Thirty-two percent (n=48) of the control sample had charges filed during the 18-month study period. There were no significant differences between the school-based probation cases (36 percent) and the controls (28 percent) in the number of individuals who were charged with crimes. For those who had any charges filed against them (n=48), the average number of charges was somewhat lower in the SBP group (1.6) compared with the matched controls (2.1). The group average, including all subjects, for number of charges was 0.55 and 0.53 for SBP and controls, respectively. Although no differences existed between these groups with respect to the number of new charges accrued, significant differences existed in the severity of the charges and the time to first charge. Charges were classified by the following types of offense: (1) probation violations and status offenses and (2) all other charges. Consistent with increased case contact, the SBP group had significantly more charges of probation violation and status offenses than did the non-SBP controls--50 percent versus 18 percent, respectively. Additionally, the time between assignment to probation and the date of the first charge was significantly longer for those assigned to SBP--271 days for SBP cases and 206 days for the controls. Overall, 27.6 percent (n=21) of the SBP cases and 29.7 percent (n=22) of the controls were assigned by the court to some form of placement during the 18-month study period. Placements included detention centers and secure placements, drug and alcohol programs, general residential placements, and a number of less restrictive community-based placements such as foster and group homes. Among those who were placed, SPB cases had a significantly longer period of time until first placement than controls--118 days versus 300 days, respectively. SBP cases also had significantly fewer days in placement than controls: 35.7 days versus 83.8 days. Differences in placements between the two groups resulted in dramatically different costs for placements. Cost of placement was determined using the authorized per diem rate schedule. The average cost of placement for the matched controls in the study was $39,314.86, while that for SBP cases was $17,701.44. To estimate overall program costs, the average cost per individual assigned to each condition was computed, which revealed a significantly lower cost for placing offenders in SBP than in other probation programs ($5,023.38 versus $11,688.20). This estimate could be used to project cost savings for counties adopting school-based probation. The number of placements had a direct impact on the amount of time the subject was in the community without additional supervision. Youth assigned to SBP spent more days in the community than controls who were assigned to traditional probation (448 days versus 400 days). Once placement occurred, the potential for rearrest was also altered. To account for the different rates of placement, rates of charges for new offenses, excluding probation violations and status offenses, were calculated as a function of time in the community. These rates, expressed as number of events per person year in the community, reflect a 43-percent lower rate for SBP cases (0.23 events per year for SBP and 0.40 events per year for controls). Summary The data presented show important differences between a group of randomly selected SBP cases and their matched counterparts who were assigned to more traditional forms of probation. It cannot be stated with certainty that the observed differences in charges and placements are due to the SBP program. The data can only suggest a program effect. Although the case control design is a powerful quasi-experimental approach, it is retrospective in nature and does not involve the random assignment of subjects to experimental and control conditions. Also, these data are derived from only three counties; therefore, the current study cannot address their generalizability to other counties. To confirm these findings, larger prospective studies are required. Despite the limitations of this approach, the findings are very encouraging. In addition to cost savings, reductions in the destructive effects of extended placements and less emphasis on the more restrictive components of the juvenile justice system can be expected. ---------------------------- In addition to cost savings, reductions in the destructive effects of extended placements and less emphasis on the more restrictive components of the juvenile justice system can be expected. ---------------------------- Virginia: Detention Center Incarceration Program This summary was adapted from the report Detention Center Incarceration Program, which presents evaluation research conducted by the staff of the Planning, Research and Certification Unit, Virginia Department of Corrections. The Detention Center Incarceration Program was implemented in the Commonwealth of Virginia in 1995 to provide an effective intermediate sanction for adult probationers. The creation of the program was stimulated by the determination that younger, nonviolent offenders must be assigned to less secure, shorter term programs to reserve the more costly and more secure prison-bed space for violent offenders whose prison terms were expected to lengthen with the abolition of parole. The source of funding was state general revenues. Program Overview The Detention Center Incarceration Program includes two primary components and is patterned after programs in Georgia. The first component is a short residential phase that emphasizes military-style discipline, physical work, and intense educational and treatment services. The second component provides intensive supervision coupled with treatment and transitional services in the community. The philosophy of the program is based on the rationale for the earlier Boot Camp Incarceration Program, which is generally regarded as successful in Virginia. The Detention Center Incarceration Program is designed for newly convicted felony offenders or probation violators who meet the following criteria: convicted of nonviolent felony offenses, determined by the court as needing more security or supervision than that provided by the Diversion Center Incarceration Program (similar to work release), and disqualified by age or physical condition from the Boot Camp Incarceration Program. The offenders are also seen as possibly benefitting from a regimented environment and structured program. Program participation is voluntary; therefore admissions are gender, race, and age neutral. Two features make Virginia's program unique: (1) the strong emphasis on substance-abuse education and treatment in the facility phase and (2) solid community transitional services and followup by probation and parole field staff. The potential for program replication is also a major consideration. Goals and Objectives The goals of the Detention Center Incarceration Program are to safely divert nonviolent felony offenders from long-term incarceration, to avoid the costs of building and operating correctional facilities, and to reduce recidivism. Objectives for achieving these goals include: o Shifting nonviolent offenders and probation violators from long-term incarceration to shorter term incarceration coupled with community supervision. o Reducing the targeted offender group's criminal activity. o Implementing a correctional program that is cost effective when compared with incarceration. ---------------------------- The goals of the Detention Center Incarceration Program are to safely divert nonviolent felony offenders from long-term incarceration, to avoid the costs of building and operating correctional facilities, and to reduce recidivism. ---------------------------- Program Activities/Components Detention Center Incarceration Program activities include 20 weeks of military-style management and supervision, physical labor in organized public works projects, counseling, remedial education, substance-abuse testing and treatment, and community reentry services. Nonviolent felony offenders may volunteer for the program in lieu of imprisonment at the time of initial sentencing or a probation violation. Their acceptance depends on an evaluation and recommendations made by probation/parole officers to the court. If accepted, an offender receives a suspended prison sentence and is placed on probation supervision with a special condition to complete the program. Transportation to the facility is provided by local correctional facility and center staff. Physical examinations are conducted immediately upon arrival. Platoons are formed at intake, and the introduction to military-style discipline and general orders begins. The average day begins at 0550 and concludes at 2200 hours. Daily activities include personal hygiene and area cleanup, military drill, supervised work details at nearby state prison or public service sites, individual and group treatment and education, and brief moments of free time. The environment is spartan; there is no television; and visitors are permitted only on alternate weekends. A graduation ceremony to which friends and family are invited concludes the residential phase of the program. As graduation day approaches, the program probation and parole officers begin developing transitional plans with the detainee's assigned community probation/parole officer. The transition plans address living arrangements, employment, payment of court-ordered fines, costs, restitution or community service, and substance-abuse or other treatment services. Upon return to the community, the detainee is assigned to intensive supervision for a period of time specified by the sentencing judge. The supervision is supplemented as needed with treatment or transitional services from community or contractual service providers. Once the detainee has successfully completed intensive supervision, he or she is placed on a regular supervision caseload for at least 1 year. There are currently three detention centers for men, with a fourth to be opened in 1999, and one center for women. Additional centers are in the planning stages. A staff of 45 include the superintendent, 5 probation and parole officers and counselors, 31 correctional officers, and other maintenance and administrative workers. Their activities include general management and direction; information dissemination to the public and education for field staff and judges; provision of individual and group assessment, treatment, and rehabilitative services; transportation; and other maintenance services. The program also provides training for staff to meet the basic and specialized requirements established by the Department of Corrections (DOC) for their respective positions. Additionally, staff must participate in training in the proper use of military-style maneuvers and discipline. Care must be taken to ensure that the forceful disciplinary approach does not degenerate into detainee abuse. Monitoring and evaluation procedures and instruments include an offender database in each center, consolidation of data into a monthly management summary report, program reviews and audits, and evaluation of program outcomes (requested by the executive staff), as needed. Performance Measures and Evaluation Methods The purpose of the report Detention Center Incarceration Program was to address the following questions: o What is the offender profile of detention center participants? o What are the program outcome rates (i.e., successful completion versus dropout)? What is the average length of stay for offenders completing the program versus dropouts? o What percentage of detention center participants commit new offenses following release or have a subsequent probation/parole revocation within the 12 months following release? o How many program participants were diverted from prison or were included in the program as a result of net widening? o Is the program cost effective compared with prison incarceration? To analyze the issues, the Planning, Research and Certification Unit at the Department of Corrections collected profile and recidivism information on all offenders who cycled through the program during calendar year 1996 and used the following methodologies to complete this research project: o Review of existing community corrections programs. o Analysis of detention center participant and completion data compiled and maintained by detention center staff. o Development of a computer program to merge data obtained from detention center staff with data of all probationers and parolees maintained in DOC's automated database. o Development of a survey instrument for use by probation and parole officers to compile recidivism and probation/parole revocation data. o Review of budget reports prepared by the DOC Budget Unit to obtain per capita costs for prisoners and detainees. The sample consisted of all offenders (n=425) who cycled through the Southampton Detention Center, Nottoway Detention Center, and Southampton Detention Center for Women during calendar year 1996. Completed survey instruments were received for 87 percent of this cohort, or 368 of the 425 offenders. The survey instrument collected the following data elements: type of obligation (parole, probation, or both), referral source, location prior to program admission, reconviction data, and sanction and revocation data. In addition, a question was included to determine whether the offender was sanctioned to the Detention Center Incarceration Program as a last resort prior to beginning the revocation process. To ensure consistent coding and reliable data, an instructional booklet listing available codes was disseminated to the probation and parole officers responsible for completing the survey instrument. ---------------------------- The sample consisted of all offenders . . . who cycled through the Southampton Detention Center, Nottoway Detention Center, and Southampton Detention Center for Women during calendar year 1996. ---------------------------- The first phase of data collection began with compiling the offender's name, identification number, supervising probation/parole officer, admission date, exit date, and reason for admission from detention center staff. These data were then merged with the probation and parole automated database to extract demographic variables, including date of birth, gender, race, marital status, number of children, educational level, and most serious offense. The second phase of data collection, which began in June 1997 and lasted for 1 month, concentrated on the completion of survey instruments by the supervising probation and parole officers. Data collected during the first phase of the project were included and were supplemented with information provided by the officers or the researchers. Once completed, the survey instruments were faxed or mailed back and entered into a database for analysis and interpretation. After all data were entered into the database, file edits were run to ensure data consistency. The data analysis phase of the project conducted by the Planning, Research and Certification Unit focused on completing offender profiles and assessing program outcomes for offenders who successfully completed the Detention Center Incarceration Program and for those who dropped out. Program outcome was cross-tabulated with gender, age at admission, race/ethnicity, and obligation type. Additional analysis compared the cost-effectiveness of the program with that of prison incarceration. Characteristics of the study population were as follows: o Almost all detention center participants were probationers referred by circuit court judges. o The vast majority of program participants were single African-American males with no children. o The average (mean) age at time of admission to the Detention Center Incarceration Program was 27 years. The youngest participant was 17 at admission, and the oldest was 57. o Just over 25 percent were high school graduates or had received a general equivalency diploma (GED). o Almost half (48 percent) of the detention center participants were drug offenders; 68 percent of this group were convicted for drug distribution. o The second most prevalent offense was a probation/parole violation, followed by burglary. o The average (mean) sentence was 5 years. ---------------------------- The data analysis phase of the project conducted by the Planning, Research and Certification Unit focused on completing offender profiles and assessing program outcomes for offenders who successfully completed the Detention Center Incarceration Program and for those who dropped out. ---------------------------- Program Evaluation Findings and Results Program Outcome Rates Nearly 8 out of 10 offenders (78 percent) admitted to the program successfully completed it. The mean length of stay for this group was 134 days; the average length of stay for dropouts was 41 days. There were no absconders. For those who dropped out of the program, the following reasons were recorded: medical/psychological (42 percent), disciplinary infractions (29 percent), voluntary withdrawal (18 percent), and other (11 percent). Parolees exhibited lower success rates than probationers: 78 percent of probationers successfully completed the program compared with only 43 percent of parolees. However, the vast majority of detention center participants were probationers; only seven parolees were in the program, and they were under concurrent probation supervision to meet the required eligibility criterion. When program outcome was measured by race/ethnicity, little difference was found in the success rates for African-American and Caucasian offenders with success rates of 76 and 80 percent, respectively. ---------------------------- Parolees exhibited lower success rates than probationers: 78 percent of probationers successfully completed the program compared with only 43 percent of parolees. ---------------------------- Release Outcome Rates The Detention Center Incarceration Program was effective in reducing reconviction and subsequent probation/parole revocation rates for those who completed the program--only 15 percent (42 out of 286) had a subsequent probation or parole revocation. Dropouts had a rate of 63 percent. Only 3 percent of the offenders (9 out of 286) who successfully completed the program had been reconvicted for new offenses (2.4 percent felony conviction and 0.6 percent for misdemeanor conviction). Offenders who dropped out of the program exhibited a slightly higher rate of felony reconviction (4 percent). For those who had successfully completed the program and subsequently received a probation/parole revocation, almost 70 percent (29 out of 42) of this group were assigned for technical violations rather than new crimes. Prison Diversion/Cost-Effectiveness Alternative placement programs are most effective when participating offenders would, absent these programs, be incarcerated in prison at a much higher cost. To assess whether the offenders studied were true prison diversions or were in the program as a result of net widening, a question was included on the survey instrument to determine whether the offender was assigned to the program as a last resort to prison incarceration. Probation and parole officers completing the survey instrument indicated that 128 (35 percent) of those assigned to the Detention Center Incarceration Program were prison diversions. Since probation/parole revocation was the most serious sanction for 57 of the 368 participants and these offenders accounted for 45 percent of the 128 offenders identified as prison diversions, it is reasonable to expect that, at minimum, these offenders were assigned to the program as a last resort to prison incarceration. To address the question of whether detention centers are cost effective, the per capita cost for detention center placement was compared with the per capita cost of prison incarceration. Data prepared by DOC's Budget Unit in November 1996 for the fiscal year ending June 30, 1996, indicated that the per capita cost for detention center placement was $14,186, compared with $16,590 for prison incarceration (a cost differential of $2,404). Diverting 128 offenders from prison resulted in a cost savings of $307,712. ---------------------------- Alternative placement programs are most effective when participating offenders would, absent these programs, be incarcerated in prison at a much higher cost. ---------------------------- The findings of this research study suggest that the Detention Center Incarceration Program was successful in meeting its intended goals and objectives. The program's strong discipline and physical requirements, focused treatment services, and period of intensive probation supervision in the community served to divert offenders from prison incarceration, thus resulting in cost savings to the Commonwealth of Virginia, reduced felony and misdemeanor reconvictions, and fewer subsequent probation and parole revocations. Offenders who cycled through the program in 1996 have outcome rates (successful completion versus dropout) that are similar to those for boot camp participants during the same time period. Seventy percent of the boot camp participants successfully completed their program, compared with 78 percent of the detention center participants. Furthermore, subsequent probation/parole revocation rates are comparable for detention and boot camp participants. Sixteen percent of those successfully completing the boot camp program had a subsequent probation/parole revocation hearing compared with 15 percent for the detention center program. However, for those who dropped out of the boot camp program, 51 percent had no subsequent probation/parole revocation, compared with only 28 percent for detention center dropouts. ---------------------------- Overall, the results of these evaluations suggest that this intensive, short-term program is an effective and safe alternative for selected offenders. ---------------------------- In 1998, the Department's Planning, Research and Certification Unit conducted a followup study of the 1996 participant cohort and found that 80 percent of the participating offenders had not returned to jail or prison. Overall, the results of these evaluations to date suggest that this intensive, short-term program is an effective and safe alternative for selected offenders. ---------------------------- New Hampshire: Merrimack County Adult Diversion Program This summary was adapted from Evaluating the Merrimack County Adult Diversion Program: Final Report, which presents evaluation research conducted by D. Alan Henry and Spurgeon Kennedy of the Pretrial Services Resource Center in Washington, D.C. Under an agreement dated December 12, 1996, the Pretrial Services Resources Center (PSRC) agreed to conduct process and outcome evaluations of the Merrimack County (New Hampshire) Adult Diversion Program. The process evaluation would assess: o Whether the program's operations and procedures comply with standards for pretrial diversion adopted by the National Association of Pretrial Services Agencies (NAPSA), the National District Attorneys Association, and the American Bar Association. o Whether the program's structure and services address the specific problems that contributed to its participants' criminal behavior. o How other criminal justice officials, program clients, and program alumni view the diversion program. The outcome evaluation would determine: o Rates of recidivism and defendant compliance with conditions of diversion. o Performance of program defendants compared with similar defendants remaining in existing case processing. Program Overview The Merrimack County Adult Diversion Program was established in 1992 under grant number 209-619-02 from the Bureau of Justice Assistance, Office of Justice Programs, U.S. Department of Justice. The diversion program was originally part of the County Attorney's Office but was moved in 1994 from Concord to its current location in Boscawen with the Department of Corrections (DOC). The program's full-time staff consist of a director, programs coordinator, and receptionist. A DOC employee serves as the program's substance-abuse counselor, and a contracted social worker assesses potential diversion clients. In addition, an advisory board helps establish the policy direction and actively reviews client participation. The program accepts defendants aged 17 years and older who are charged with nonviolent felony offenses and have no history of violent criminal behavior. Consistent with accepted standards, there are no restrictions on program participation except for current and past criminal history and a defendant's receptiveness to the program. A guilty plea is not required for diversion participation. ---------------------------- The Diversion Offender Profile was created by Merrimack County police agencies and the county attorney during the program's development phase to identify nonviolent, nonhabitual defendants eligible for referral. ---------------------------- Diversion program placement is a three-step process: 1. Local arresting agencies determine initial eligibility using a Diversion Offender Profile. Defendants identified as eligible for diversion are referred to the Merrimack County attorney. 2. The county attorney, after discussing the referral with the arresting agency, determines the strength of the case and whether diversion from conventional case processing is appropriate. If charges are filed and diversion appears suitable, the charging documents are completed, the defendant's case is placed on a suspended calendar, and the defendant is referred to the program for a final assessment. 3. The diversion program conducts an assessment that includes an investigative interview with the defendant, a substance-abuse evaluation to determine current drug use and appropriate monitoring or treatment, and a social work evaluation to identify potential clients whose mental or emotional problems make them unsuitable for diversion and determine an appropriate alternative program. The Diversion Offender Profile was created by Merrimack County police agencies and the county attorney during the program's development phase to identify nonviolent, nonhabitual defendants eligible for referral. According to the profile, defendants are ineligible if they meet any of the following criteria: o Are charged with a violent, sex-related, or drug-trafficking crime. o Have a previous conviction resulting in an incarceration of more than 7 days. o Have already participated in diversion. o Are charged with an offense that "represents a major threat to society or requires general deterrent." o Are charged with a crime whose conviction "would justify" a prison sentence. o Have a prior person, property, or drug-related felony conviction. The program has an automated management information system containing data on all referrals, accepted clients, rejected defendants, and program outcomes (successful completion or termination). Goals and Objectives According to its mission statement, the diversion program seeks to: Divert nonviolent offenders from the criminal court docket into a rigorous, comprehensive, community-based rehabilitation program. . . . To assist in this self-rehabilitation, the program provides opportunities for treatment, community service, restitution, counseling, and education. Program Activities/Components Diversion program assessment takes 138 days on average and uses two main screening instruments--substance-abuse and social work assessments. The substance-abuse assessment, performed by a DOC substance-abuse counselor, includes evaluations based on a drug-abuse test and an alcohol- screening test. Based on the results of these tests, the counselor recommends placement in one of three supervision groups: o Level One--for participants assessed as not having substance abuse-related disorders. Clients under this level of supervision attend informational and educational workshops on substance abuse. o Level Two--for participants assessed as having substance-abuse problems. Besides the workshops mentioned above, these clients are required to attend 90-minute group sessions and submit to regular substance-abuse testing. o Level Three--for participants whose substance-abuse problems require a level of inpatient treatment. Upon successful inpatient treatment, these clients are referred to Level Two supervision. The social work assessment gauges a defendant's current and prior mental, emotional, and behavioral conditions. It is performed by a licensed social worker and is extensive, focusing on a defendant's self-perception, problems that may contribute to negative behavior, and past and present mental and emotional problems and treatment. The program director and program coordinator make the final decision to accept or reject a defendant based on the results of the initial social work assessment, a drug-abuse test, and an alcohol test. Once a defendant is accepted into the Merrimack County Adult Diversion Program, he or she and his or her defense attorney meet with the program director or a designated staff member for a program review during which the defendant's conditions and responsibilities under direct supervision are presented to the defendant as a written diversion contract. After thorough review, the document is signed by the program director, the defendant, defense counsel, and the county attorney. Program supervision includes four mandatory conditions: o 600 hours of community service work. o A tour of the New Hampshire State Prison for Men located in Concord. o Restitution payments, if ordered by the county attorney. o 10 hours of substance-abuse classes to be completed within 6 months of program acceptance. Specific diversion program conditions are categorized under the life skills component. This consists of educational classes and 2-hour workshops, and clients must complete a total of 30 hours. Programs are selected through joint consultation between a client and the program coordinator and include topics such as conflict resolution, stress management, domestic violence, job search skills, parenting skills, wellness, communication skills, and AIDS/sexually transmitted diseases. Classes are open to a client's family members and partners to enhance the classes' rehabilitative effects. Performance Measures and Evaluation Methods Process Evaluation The process evaluation was based on PSRC's onsite inspection of the diversion program using a checklist of diversion program policies and procedures drawn from relevant criminal justice standards and onsite interviews with the program director, staff, and advisory board; county DOC administration; county attorney; defense bar; and program clients and alumni. Diversion program procedures also were compared with those of 20 similar programs surveyed by PSRC. The evaluation focused specifically on: o Client eligibility and enrollment. o Supervision services and conditions. o Organizational structure. ---------------------------- The evaluation focused specifically on: o Client eligibility and enrollment. o Supervision services and conditions. o Organizational structure. ---------------------------- Client Eligibility and Enrollment Criminal justice standards recommend that eligibility criteria for pretrial diversion be stated in writing. They should include all defendants who might benefit from diversion and should ensure that program placement is not denied to a client because of race, gender, sexual preference, economic status, disability, or inability to pay restitution or potential program fees. In addition, criteria should not require a guilty plea and the defendant's participation should be voluntary. Supervision Services and Diversion Conditions Diversion standards suggest that diversion conditions address specific client needs, particularly addressing behavior that may lead to criminal activity. These conditions should be specific, achievable, and as minimally restrictive as possible to achieve the goals of rehabilitation and deterrence. An appropriate diversion service plan also may include voluntary community service, restitution, and drug testing as general conditions. Diversion standards also recommend that supervision include clear time limits for program completion and a mechanism for reviewing and possibly revising conditions as needed to achieve contract goals. NAPSA standards also encourage supervision time that is neither longer nor more costly than necessary to achieve the goals of rehabilitation and deterrence. Most diversion programs surveyed had supervision times of 6 months to 1 year. Finally, diversion standards suggest that programs not release client-based information without the client's knowledge or consent and have in place written agreements with the courts, prosecutors, and service providers detailing the type of information the program will release and under what circumstances. Organizational Structure NAPSA standards suggest that diversion programs have a written mission statement or statement of goals and objectives, an information system that allows program information to be recorded and managed, and a staff sufficient in size and experience to help clients meet program requirements. Outcome Evaluation The outcome evaluation sought to identify differences in recidivism rates among defendants completing the diversion program successfully, those terminated from the program, and those who withdrew during the assessment phase and to determine if these differences could be explained wholly or in part by program participation. A quasi-experimental design was used due to problems implementing an experimental design--that is, a smaller sample size and a shorter duration. This method constructed comparison groups (rather than control groups) from program records of defendants who closely matched diversion participants. The groups included defendants who either failed or withdrew during assessment and clients terminated from the program. These groups were then matched against successful program participants by the variables studied, with differences attributed in some degree to diversion program participation. The design allowed PSRC to consider all defendants referred to or accepted by the program as comparison or program group participants. In most cases, the evaluation also allowed more time to track behavior after program development. ---------------------------- The outcome evaluation sought to identify differences in recidivism rates among defendants completing the diversion program successfully, those terminated from the program, and those who withdrew during the assessment phase and to determine if these differences could be explained wholly or in part by program participation. ---------------------------- Group One included successful program participants; Group Two, defendants who dropped out during assessment; and Group Three, program terminations. Each group was originally screened through the Diversion Offender Profile, which ensured similarity in categories such as type of charge, prior criminal history, prior diversion program participation, and circumstances regarding the most recent offense. To further evaluate similarities, PSRC compared each group by characteristics found in the database, including age, gender, race, professional status, marital status, and educational level. Due to the small number of cases in the overall sample, PSRC evaluated all defendants and clients considered for diversion. Because sampling was not used, measures of association could not be applied; differences between the groups were therefore presented in percentages. Program Evaluation Findings and Results Process Evaluation PSRC found that the Merrimack County Adult Diversion Program met the guidelines for client eligibility and enrollment by having no restrictions on program participation except current and past criminal history and a defendant's receptiveness to the program. Client Eligibility and Enrollment The evaluation suggested that the profile categories "offense represents a major threat to society or requires general deterrent" and "prison-bound offender" required prosecutorial input. From interviews with police officials and the prosecutor, PSRC could not determine whether these categories were ever used by arresting agencies and, if so, whether the agencies defined these categories consistently. If they did not, it was possible that similarly situated arrestees in separate police districts were not being screened for diversion in the same way. Other profile categories seemed inconsistent or vague. For example, while the profile excluded defendants with prior jail incarcerations of more than 7 days, past sentences such as probation, fines, and restitution were not considered restrictions. Additionally, there were no time limits on the prior jail time restriction. For example, a defendant who served a 30-day jail sentence 10 years ago apparently would score the same as a defendant who served a similar sentence within the past 3 months. ---------------------------- PSRC found that the Merrimack County Adult Diversion Program met the guidelines for client eligibility and enrollment by having no restrictions on program participation except current and past criminal history and a defendant's receptiveness to the program. ---------------------------- PSRC recommended that the Diversion Offender Profile be used by arresting agencies to gauge only a defendant's current and past criminal behavior and that the following restrictions be used to ensure that exclusions to diversion eligibility were specific and less restrictive: o No current violent, sex-related, or drug-trafficking offense. o No prior felony convictions. o No prior misdemeanor convictions within a specified period of time (for example, the past 2 years). o No prior diversion participation. o No pending criminal charges or current probation or parole status. The 138 days needed for assessment was longer than the 1- to 3-month assessment period noted by most survey diversion programs. According to data from 160 reviews, the diversion program offered 110 defendants contracts after the screening period. For cases for which substance-abuse evaluation data were available (n=104 cases), 50 screened defendants (48.1 percent) were placed into Level One supervision and 53 (50.9 percent) into Level Two. This suggests that a significant number of screened defendants have substance abuse-related issues that can be addressed through counseling and/or regular drug and alcohol testing. After assessments of 20 defendants who failed during this period for reasons other than rearrest or withdrawal were reviewed, it was found that only three exhibited mental or emotional problems, making them inappropriate for placement. This finding suggested that most defendants who had been assessed would not require extensive social and psychological screening to determine needed services. PSRC recommended that the diversion program consider replacing the substance-abuse assessment with a procedure based on a defendant's current and prior drug use and results from drug and alcohol tests. Placement into drug monitoring or treatment would depend on a defendant's admitted drug or alcohol usage, test results, and the circumstances of the current charge. PSRC also recommended that the diversion program consider discontinuing the social work assessment since the county's population eligible for diversion does not exhibit emotional or mental problems requiring such a lengthy screening process. Client Supervision and Diversion Conditions PSRC found that none of the four mandatory diversion conditions addressed specific client needs. Rather, as expressed in interviews with criminal justice personnel, these conditions are meant to have an overall rehabilitative effect and serve as community service and restitution to allow the clients to "repay" the community during their time in diversion. While mandating these conditions runs counter to the diversion standards' recommendation for voluntary participation, each condition appears designed to meet diversion goals. However, PSRC cautioned that diversion, as a sanction, was not intended as punishment and, therefore, its conditions should not carry a potentially punitive effect, which may be the case with an across-the-board community service requirement. Most surveyed diversion programs that include a community service requirement limit the condition to a 50-hour maximum. PSRC recommended that the diversion program drop its policy of a uniform 600-hour community service condition and adopt a "sliding scale," with hours based on a client's overall supervision plan, employment status, and schedule, and limit the maximum community service hours to a total more in line with current diversion program practices nationwide. The specialized courses under the life skills component appear to provide an excellent avenue for addressing specific client behavior that may lead to future criminal behavior and meet the standards' requirement for establishing conditions suited to client needs. Diversion program time is open-ended, depending on the client's completion of diversion classes, community service, and restitution payments. Data supplied by the program show that the median supervision time was 660 days for clients successfully completing the program and 224 days for clients terminated for noncompliance. The bulk of supervision time is spent meeting community service obligations. PSRC recommended that the diversion program limit supervision to 1 year (since most diversion conditions could be completed within that period) and extensions be made on a case-by-case basis. The program releases client data only to agencies providing services to clients, such as substance-abuse treatment or testing facilities; however, no formal written agreements exist regarding release of such data. To avoid the possibility of problems and to be consistent with standards, PSRC recommended that the program enter into written memoranda of understanding with service providers outlining appropriate release and use of program data. Organizational Structure The program has a clear mission statement and an automated management information system. However, PSRC believes the staffing levels make maintaining adequate supervision and the current level of services problematic and expansion impractical. Accordingly, PSRC recommended that the program use its existing budget allocation for contracted social work to hire a part-time programs coordinator. PSRC also noted in several onsite interviews that the Boscawen location was inconvenient. Program data showed that nearly a third of the clients are from Concord or surrounding townships, and most of the service providers are in the Concord area. Given Concord's location within the county, easier access, and proximity to most program clients and services, PSRC recommended that county officials move the diversion program back to the Concord area. Conclusion The Merrimack County Adult Diversion Program is an established and well-respected component of the county's efforts to address and deter criminal conduct. According to criminal justice officials as well as current and past clients, the program has a definite and positive effect on defendant behavior. With certain revisions to its policies and expansion to other appropriate defendant populations (such as misdemeanor and juvenile offenders), the program will continue to maintain its current high level of service to its clients and the criminal justice system. Outcome Evaluation In evaluating variables that favorably affected recidivism rates, PSRC found the following: o Recidivism: There is a notable difference in recidivism rates by group; there were no rearrests in Group One after program completion as compared with the rearrest of 22.7 percent of Group Two cases and 24.1 percent of Group Three cases. o Recidivism and Diversion Participation Points: Most clients or defendants involved with the diversion program were arrested before 1995. In both Groups Two and Three, recidivism occurred within 1 year of program or assessment termination. o Recidivism and Age: Younger clients and defendants were rearrested more often than older: 22.7 percent of clients and defendants between ages 18 and 25 were rearrested, compared with 11.1 percent of those between ages 26 and 34, 5.9 percent of those between ages 35 and 40, and none of those over 40. o Program Compliance and Recidivism Rates: Of the 67 program clients in the sample, 56.7 percent (38) successfully completed supervision and 10.4 percent were rearrested after program termination. From an initial database of 160, 111 defendants were assigned to an "evaluation group" resulting in 38 diversion program clients successfully completing supervision; 44 defendants from 1 of 2 "comparison groups" dropped from consideration following the program's initial assessment; and 29 program clients terminated from supervision. ---------------------------- There is a notable difference in recidivism rates by group; there were no rearrests in Group One after program completion as compared with the rearrest of 22.7 percent of Group Two cases and 24.1 percent of Group Three cases. ---------------------------- Conclusion The outcome evaluation of the Merrimack County Adult Diversion Program yielded the following findings: o There is a marked difference in recidivism rates between persons successfully completing diversion and other similar defendants not participating in or having been terminated from diversion. o The relationship between diversion completion/recidivism weakens when variables such as age, educational level, and marital status are used as controls. o Although successful diversion program participation may have some effect on the likelihood of future recidivism, given the impact of the control variables, PSRC cannot state conclusively that participation is the reason for the differences between diversion clients and similar defendants. ---------------------------- There is a marked difference in recidivism rates between persons successfully completing diversion and other similar defendants not participating in or having been terminated from diversion. ---------------------------- These findings are tempered by the limitations of the research design and the data sample. The quasi-experimental design relied heavily on crafting comparison groups that closely matched the evaluation group. While Groups Two and Three were similar enough in many respects to Group One, the differences in educational levels and marital status leave room to reasonably question whether these groups were similar enough to compare. Despite these limitations, certain definite conclusions about diversion in Merrimack County were reached following the process and outcome reviews and included: o A well-defined defendant subgroup exists in Merrimack County that is suitable for diversion placement. County officials managed to identify a stable and well-defined defendant population eligible for diversion consideration. The relatively low failure rate for those accepted into the program, defined through recidivism, and the willingness of most clients to comply with supervision appear to justify the use of diversion for this population. o The diversion program has established itself as an effective and needed component in the county's continuum of sanctions to address and deter criminal behavior. In their onsite interviews with PSRC, county criminal justice officials stated that the diversion program had become an effective sanction for eligible defendants. The combination of supervision, life skills and services, and community service work gave participants the means to address circumstances in their lives that might lead to future criminal behavior while "paying back" the community for the privilege of participating in diversion. Most interviewees also believed the defendants eligible for diversion benefited more from their program participation than they would have through other more expensive sanctions such as incarceration or probation. o The diversion program operates in compliance with nationally recognized diversion standards. With certain noted exceptions, the Merrimack County Adult Diversion Program conforms to national criminal justice standards. It has written guidelines that identify eligible defendants and that determine each client's eligibility and specific program needs. A defendant's decision to participate in diversion is voluntary and is made with the advice of defense counsel and after a thorough review of general and specific program requirements. An admission of guilt is not a program requirement; moreover, successful program participation results in dismissal of charges. Most conditions and services are specific, reasonable, and, with certain exceptions, geared to a client's needs. Finally, the program regularly reviews each client's progress and has specific in-house sanctions to address minor rule infractions. o The diversion program is in a good position to apply services that may help reduce future recidivism. The outcome evaluation appeared to show that educational level has a controlling effect on, though not a direct relationship to, reduced recidivism. Defendants with higher educational levels who are subject to criminal justice supervision may be less involved in future crime. This finding suggests that future supervision efforts should include education components, such as GED instruction, or encourage continued school enrollment. As the sanction applied earliest in case processing, the diversion program is in an excellent position to promote education to its participants. ---------------------------- Oregon: The Multnomah County STOP Drug Diversion Program This summary was adapted from the report An Outcome Program Evaluation of the Multnomah County STOP Drug Diversion Program, which presents evaluation research conducted by Michael Finigan, Ph.D., Northwest Professional Consortium, developed under technical assistance grant SJI-96-06X-T-A-174 from the State Justice Institute. The STOP (Sanction-Treatment-Opportunity-Progress) Drug Diversion Program was initiated in 1991 to reduce the increasing backlog of drug cases in Multnomah County and to encourage treatment for those with first- offense drug charges. Program Overview The Multnomah County STOP program includes the following components: o Court oversight and active judicial case management. o Immediate access to a dedicated treatment resource. o Drug testing. o A range of intermediate sanctions. In 1995, STOP added a series of enhancements to expand its target population and to provide additional access to health, mental health, family intervention, resource coordination, and aftercare services. Goals and Objectives STOP program goals were to reduce substance abuse by improving treatment outcomes and to reduce recidivism by improving program impact. Objectives for the first goal included the diversion and treatment of up to 700 clients from the drug court docket in 1994 and 1995, successful program completion by 65 percent of program participants, and the delivery of drug-free babies borne by all pregnant women in the program. Objectives for the second goal included a recidivism rate of no more than 15 percent, a conviction rate of no more than 10 percent within 1 year following program completion, partnership and funding with an outside institution to complete process and outcome evaluations, 1-year followup on participants by program staff, and opportunities for program staff to provide technical assistance to other Oregon jurisdictions wishing to implement similar programs. ---------------------------- STOP program goals were to reduce substance abuse by improving treatment outcomes and to reduce recidivism by improving program impacts. ---------------------------- Program Activities/Components Judicial Process After review by the district attorney, defendants are offered the opportunity for drug court diversion at their first appearance before the court; they are eligible only if the charge is for possession of a controlled substance (PCS) (not distribution or manufacturing), preferably in small quantities. If there are additional nondrug criminal charges, individuals are eligible provided conditions of probation do not interfere with participation in the program. Between 900 and 1,100 cases per year are first-time appointments from district court to petition for the STOP program, with from 400 to 700 actually being admitted and participating in part of the program. The STOP program is voluntary; when defendants petition to enter the STOP track, they agree that if they fail the program, they will be tried solely on the basis of the police report. This "stipulated facts" trial is brief with a swift and sure sanction resulting from program failure. One negative side of this process is that some of the "front-end" costs savings expected by a diversion program are lost since legal representation and judicial time are required throughout the process. Nonetheless, it clearly adds a powerful incentive to remain compliant in the program, because a violation of any rule involves, at minimum, an appearance before the judge who oversees the program. ---------------------------- About 46 percent of admitted cases graduate from the program. ---------------------------- Treatment Treatment is provided by InAct, Inc., a private, nonprofit agency that provides outpatient, multiphased intervention, including group counseling sessions and acupuncture treatments. STOP clients are required to engage in group and individual counseling sessions at InAct every weekday during the initial phase of the program and once or twice a week during the final phases. The court provides active case management involving monthly status hearings at which time the court reviews drug test results and treatment progress. Other specialized services are also available, including women's services, Hispanic services, and a literacy program. The level of intervention and accountability for STOP clients far exceeds what is typical in outpatient treatment. About 46 percent of admitted cases graduate from the program. Performance Measures and Evaluation Methods The evaluation had two major goals: o Assess whether program participants have positive outcomes particularly focusing on outcomes related to criminal recidivism (e.g., lower subsequent arrests and convictions and more compliant probation or parole supervision). o Assess the ratio of the program's cost to the avoided costs resulting from the program's positive outcomes (if any). The study sample included all cases--those who graduated and those who did not complete the program but received partial treatment; incompletion could be due to either drug test failure or failure to appear (FTA) at a status hearing. Because the period from 1991 to 1992 was the early implementation phase, a period often premature for reviewing an innovative program such as this, the sample was taken from the 1994 to 1995 period to study results of the fully implemented program. The sample included 150 subjects who were diverted and graduated and 150 subjects who were diverted but did not graduate, which included those who received a great deal of treatment and those who received little; random selection was used for both the graduate sample and the nongraduate sample. The groups did not differ significantly in gender, age, or race/ethnicity composition. Both groups were approximately 72 percent male, the average participant age was 33, and the proportion of nonwhites was 23 percent. Hispanic clients were excluded due to a preponderance of Immigration and Naturalization Service (INS) holds. Because the groups differed, however, in the number of prior arrests, this category was used as a control variable. The mean number of prior arrests for all program participants was 2.5; however, graduates averaged 1.6 prior arrests, while nongraduates averaged 3.4. The selection of a comparison group is a critical element in a nonrandomized research design. In this case, the best comparison group for the STOP program graduates was a contemporaneous sample of arrestees who were eligible for the program but were not able to participate. These individuals needed to meet the following criteria: a charge for PCSI or PCSII (preferably not with large amounts of the drug in possession) and eligibility but not entrance into the STOP program. Figures given by the Metropolitan Public Defender's Office indicated that approximately 2,400 individuals were arrested each year with drug offenses as the primary charge and 4,400 were arrested with drug offenses as the secondary charge. Because 400 to 500 defendants, out of an estimated 1,100 first-time appointments from district court to petition for the STOP program, were actually admitted to the program, the comparison group was rather large. The reasons for nonentry into STOP included withdrawal, client refusal (choosing trial instead), court denial, and prior bench warrants. These reasons had the potential to introduce a bias that may have made these clients unfit to be part of a comparison group; to offset this bias, a representative matched sample approach was used. County and state databases were used to compile a pool of eligible clients who were randomly selected to be representative of program participants based on the following: gender, age, race/ethnicity, and prior criminal history. This effort produced a sample of 150 individuals with backgrounds similar to those who had entered the program. There were no statistically significant differences between the two groups on these critical control variables. Data were collected on the three study groups (graduates, nongraduates, and comparison group) on treatment outcomes for 2 years before and 2 years after respective criterion dates. The criterion date for graduates was the date the client graduated from the STOP program; for nongraduates, the date the client left the program; and for the comparison group, the date the client received the STOP appointment. Data were collected using various sources including but not limited to Law Enforcement Data System (LEDS), Client Process Monitoring System (CPMS), and Adult and Family Services. ---------------------------- Data were collected on the three study groups (graduates, nongraduates, and comparison group) on treatment outcomes for 2 years before and 2 years after the respective criterion dates. ---------------------------- The key outcome variables (performance measures) focused on the following societal outcomes: o Subsequent arrests. o Subsequent convictions. o Subsequent incarcerations. o Types of crime committed. o Supervision experiences. o Use of public assistance resources, including food stamps. Supervision focused on providing a positive reentry of the criminal into the community. Signs of positive adjustment included finding employment, receiving training and education, receiving substance-abuse treatment and counseling, finding housing, and using other services to develop a noncriminal lifestyle. To assist in the analysis of these positive adjustment outcome measures, a scale of positive adjustment for parolees was designed, adapted from Latessa and Vita (1988). Data collection staff used this scale to measure each client's adjustment in becoming a productive member of society after examining the client's supervision officer file. ---------------------------- Supervision focused on providing a positive reentry of the criminal into the community. ---------------------------- Outcome data were also used to assess the relative cost/benefits of the STOP program to the taxpayer. This study focused on "avoided costs" or costs to taxpayers had the participants not received treatment. The cost-to-taxpayers approach measured the costs related to untreated substance abuse that was paid by taxpaying citizens. Avoided costs were assessed for Multnomah County taxpayers and Oregon taxpayers and included criminal justice system costs (police protection from crime, adjudication, jail, supervision), victim losses, theft losses, health-care services, and public assistance. The primary analysis strategy was to examine each outcome measure for all three sample groups and analyze a covariance model with prior arrests as the chief covariate. If statistical significance was gained for the model, individual comparisons (e.g., graduates versus eligible candidates) were tested for significance. This method was preferred since a series of bivariate tests can occasionally produce spurious statistical significance. ---------------------------- Substance abuse is one of the strongest motivating factors for an individual to continue criminal activity. ---------------------------- Program Evaluation Findings and Results Substance abuse is one of the strongest motivating factors for an individual to continue criminal activity. A study of substance use, abuse, and dependence in the Multnomah County Justice Center found that approximately two-thirds of those arrested were recent users of drugs (based on urinalysis) and that about half could be classified as clinically dependent on either alcohol or drugs. Drug use played a major role in the life of this population. Reducing substance abuse should have had a positive effect on reducing criminal recidivism. Subsequent New Arrests Participants Versus Comparison Group: Those who participated in STOP had 61 percent fewer subsequent arrests over a 2-year period than the control comparison group. Program Graduates Versus Program Nongraduates: Program graduates had 49 percent fewer subsequent new arrests than nongraduates over a 2-year period. Amount of Program Completed: For those who completed less than one-third of the program, the rate of subsequent arrests was more than twice that of those who completed at least a third of the program (139 rearrests versus 62 rearrests per hundred participants). Participating in a substantial portion of the program had positive effects on recidivism. ---------------------------- Completing and graduating from the STOP program had the most positive effect on reducing recidivism. ---------------------------- Program Graduates Versus Comparison Group: Program graduates had 76 percent fewer total subsequent arrests than the comparison group over a 2-year period. Completing and graduating from the STOP program had the most positive effect on reducing recidivism. Subsequent Serious Arrests (Felony Type A and B) Program Participants Versus Comparison Group: Clients who participated in the STOP program had 64 percent fewer subsequent felony arrests than the comparison group over a 2-year period. Program Graduates Versus Program Nongraduates: Program graduates had 56 percent fewer subsequent Class A and B felony arrests over a 2-year period than program participants who did not graduate. Program Graduates Versus Comparison Group: STOP graduates had 80 percent fewer total subsequent felony arrests than the comparison group clients over a 2-year period. Subsequent Convictions In Oregon, new arrests are a better measure of recidivism than convictions or incarcerations for two reasons: (1) most new arrests reported in LEDS are felony arrests and most lead to some kind of conviction and (2) actual adjudication outcomes are dependent on variables that are difficult to interpret as outcome measures. A combination of plea bargaining, jail release or transfer programs, and prior bench warrant issues complicate sentencing and incarceration in any given case. Nonetheless, it is interesting to examine the rates of conviction among the three groups. Program Participants Versus Comparison Group: There was a 57-percent difference between the two groups in total convictions over a 2-year period, with those participating in STOP having fewer total convictions than the comparison group. Program Graduates Versus Program Nongraduates: There was a 51-percent difference between the two groups in total subsequent convictions over a 2-year period, with a conviction rate of 59 per 100 nongraduates compared with 29 per 100 graduates. Program Graduates Versus Comparison Group: Program graduates were convicted at a rate of 29 new convictions per 100 participants in a 2-year period compared with 111 per 100 for the comparison group. This was a 74-percent difference between graduates and the comparison group in total subsequent convictions in a 2-year period. Subsequent Drug Arrests Since a drug-related arrest was the chief criterion for eligibility in the STOP Drug Court Diversion Program, and since drug treatment was a central part of the diversion, examining the rate of subsequent arrests for drug-related crimes was of particular interest. All subsequent drug-related arrests collected from LEDS for each group were felonies. Program Participants Versus Comparison Group: There was a 72-percent difference between participants and the comparison group in total subsequent drug-related arrests over a 2-year period (22 new drug-related arrests per 100 participants compared with 78 per 100 of the comparison group). Program Graduates Versus Program Nongraduates: There was a 56-percent difference between graduates and nongraduates in total subsequent drug-related arrests over a 2-year period (12 rearrests per 100 graduates and 27 rearrests per 100 of nongraduates). Program Graduates Versus Comparison Group: There was an 85-percent difference between the two groups in total subsequent drug-related arrests over a 2-year period (12 rearrests per 100 graduates compared with 78 rearrests per 100 of the comparison group). A substantial difference between program participants and the comparison group existed in the rate of drug-related arrests in the 2-year period after the program. The difference in drug-related arrests between groups was greater than the difference in total new arrests. The STOP program appeared to have its greatest effect on reducing subsequent drug arrests. Subsequent Property Crime Arrests Program Participants Versus Comparison Group: There was a 58-percent difference between program participants and the comparison group in total subsequent felony property arrests over a 2-year period. STOP participants had a rate of 13 new arrests per every 100 participants after leaving the program compared with 31 per 100 of the comparison group. STOP program participants had a total of 2 new subsequent serious property crime arrests per 100 participants compared with 6 per 100 of the comparison group. Program Graduates Versus Program Nongraduates: Program graduates were rearrested at a rate of 7 new felony property crime arrests per 100 participants compared with 10 per 100 of nongraduates. This was a 30-percent difference between the two groups over a 2-year period. STOP program graduates had a total of 2 new subsequent felony property crime arrests per 100 participants compared with 3 per 100 nongraduates. Program Graduates Versus Comparison Group: Graduates were rearrested at a rate of 7 new felony property crime arrests per 100 compared with 29 per 100 of the comparison group who were eligible but did not participate in the program. This is a 76-percent difference over a 2-year period. STOP program graduates had a total of 2 new subsequent felony property crime arrests per 100 participants compared with 6 per 100 of the comparison group. A substantial difference in the rate of all property crime arrests existed in the 2-year period after the program for program participants compared with the comparison group. Graduates, in particular, had far fewer subsequent property crime arrests than the comparison group. The percentage differential in property crime arrest was similar to the difference in total new arrests for all comparisons. Subsequent Personal Crime Arrests Program Participants Versus Comparison Group: STOP participants had subsequent felony (Class A and B) arrests at a rate of 1 per 100 participants in a 2-year period compared with 4 per 100 of the comparison group. This was a 75-percent difference over a 2-year period. Program Graduates Versus Program Nongraduates: No graduates were rearrested for serious personal crimes. But the rearrest rate was 2 per 100 in the nongraduate group. This represented a 100-percent difference over a 2-year period. Program Graduates Versus Comparison Group: No graduates were rearrested but the rearrest rate was 4 per 100 of the comparison group. This was a 100-percent difference in total subsequent serious personal crime arrests over a 2-year period. A substantial difference in the rate of all personal crime arrests existed in the 2-year period after the program for participants compared with the comparison group. Graduates, in particular, had far fewer subsequent personal crime arrests than the members in the comparison group. The percentage differential in total personal crime arrests was similar to the difference in total new arrests for all comparisons; however, the percentage difference in serious personal crime arrest was greater than the difference in total new arrests for all comparisons. Subsequent Parole Violation Arrests Program Participants Versus Comparison Group: STOP participants had subsequent probation or parole violation arrests at a rate of 2 new arrests per 100 participants compared with 10 per 100 of the comparison group. This was an 80-percent difference over a 2-year period. Program Graduates Versus Program Nongraduates: There was no difference between the two groups in a 2-year period. Rearrests for new probation or parole violations were 2 per 100 graduates compared with 2 per 100 nongraduates. Program Graduates Versus Comparison Group: Graduates were rearrested at a rate of 2 new probation or parole violations per 100 whereas 10 per 100 of the comparison group were rearrested for violations, representing an 80-percent difference over a 2-year period. A substantial difference between program participants and the comparison group existed in the rate of parole or probation violation arrests in the 2-year period after the program. Graduates and nongraduates, however, showed no difference in subsequent parole or probation violation arrests. Positive Adjustment Score of 1 or More Program Participants Versus Comparison Group: More than half of STOP participants for whom supervision records were available scored at least 1 point on the Positive Adjustment Scale, a set of indicators measuring factors such as employment, enrollment in school, standard of living, participation in self-improvement programs, and other important variables of positive community adjustment and successful supervision. Just under 40 percent of the comparison group individuals scored as well. Positive Adjustment Score of 4 or More Program Participants Versus Comparison Group: Twenty-seven percent of program participants for whom supervision records were available scored 4 or more points compared with 10 percent of the comparison group. Cost Analysis The STOP Drug Diversion Program saved the Multnomah County Criminal Justice System $2,476,795 per cohort during the 2 years of data collection after participation in the program. The study estimated the avoided costs to the taxpayers of Oregon to be $10,223,532 over 2 years. Every taxpayer dollar spent on cohorts of clients who participated in the program produced $2.50 in cost savings to the taxpayers of Multnomah County. The ratio of benefit to the Oregon taxpayer was $10 saved for every $1 spent. Advantages of the STOP Drug Diversion Program The STOP Drug Diversion Program in Multnomah County offers advantageous features in both its court management and its substance-abuse treatment procedures. In court management: o A judge overseeing the process, creating a continuity in the judicial process. o A "stipulated facts" trial, allowing for swift and sure punishment for program failure. o Careful court monitoring with frequent drug testing and frequent appearances before the judge. In substance-abuse treatment: o A single provider, ensuring consistency on the treatment side. o A multiphase, multiaspect 12-month treatment program. o Frequent drug testing. o Court-enforced attendance and progress. Summary This study had certain limitations. Because it was retrospective, random assignment to treatment and control conditions was not possible (and probably not possible even if it had been a contemporaneous sample). Therefore, some unmeasured (and possibly unmeasurable) differences may have existed among the three groups that affected the study outcomes. Nonetheless, the comparison group was matched on all critical variables, removing possible sources of bias. This study found that program participants, especially graduates, had significantly fewer subsequent arrests and convictions, particularly felony arrests, and had lower rates of drug-related arrests--both of which suggest that the program may have had an effect on lowering participant involvement with drugs. This lower rate of recidivism, in turn, affects public safety and public costs. Fewer arrests and convictions lessen the pressure on the criminal justice system, particularly the jail system, and can lead to more reasonable loads on the current system, fewer future expenditures, or both. Lower rates of recidivism can also result in savings to taxpayers in terms of costs for dealing with new offenses, treating victims, and improving safety. ---------------------------- This study found that program participants, especially graduates, had significantly fewer subsequent arrests and convictions. ---------------------------- Illinois: Homicide and Violent Crime Strike Force in Madison and St. Clair Counties This summary was adapted from the report An Evaluation of the Homicide and Violent Crime Strike Force Program in Madison and St. Clair Counties, which presents evaluation research conducted by Richard Schmitz, J.D., and Pinky S. Wassenberg, Ph.D., J.D., Center for Legal Studies, University of Illinois at Springfield. Madison and St. Clair Counties experienced severe levels of violent crime during the late 1980s and early 1990s. The cause of the influx of crime may have been due to educational inadequacies, loss of industry, and the lack of funding to local law enforcement. In response to this epidemic, the Homicide and Violent Crime Strike Force (hereinafter referred to as the "task force") was created by the Illinois State Police (ISP) Department and the Illinois Attorney General's (AG's) Office. --------------------------- The task force was designed as a joint venture between ISP and the Illinois AG's Office. ---------------------------- Program Overview The task force was designed as a joint venture between ISP and the Illinois AG's Office. The ISP program component was made up of one supervisor and four case agents who were experienced homicide investigators with strong community ties to the selected counties. The AG's Office program component, made up of two attorneys, one investigator, and one secretary who served both agencies, was designed to assist local prosecutors in the prosecution of task force cases and give legal advice to the State Police component. It was anticipated that task force attorneys would occasionally take a lead role in the prosecution of task force cases. Resources and personnel levels were consistent throughout the program, but there were changes in task force operating procedures and interagency relationships. These are discussed in detail under Program Activities/Components on page 51. This study focuses on St. Clair and Madison Counties in Illinois, which are located in the southwestern section of the state and share a border with Missouri. Their combined population in 1992 was approximately 516,000. More than two-thirds of all persons residing in St. Clair County are white, as are 90 percent of those in Madison County; during the past 30 years, both counties experienced increases in nonwhite populations. Both counties report more female residents than male: 52.2 percent in St. Clair and 52.1 percent in Madison. Less than 15 percent of the population in either county possess a bachelor's or higher degree. According to the Regional Economic Information System, St. Clair County residents reported a 1994 per capita personal income (PCPI) of $18,452, compared with $20,530 reported by Madison County residents. This placed St. Clair County 52nd (out of 102) in the state for income level and represented approximately 78.2 percent of the state average ($23,611) and approximately 85 percent of the national average ($21,696). Madison County was ranked 22nd in the state, and its PCPI was 87 and 94.6 percent of state and national averages, respectively. Of all St. Clair families, 13.9 percent reported an income below the poverty level; 43.4 percent of all single female head-of-household families lived in poverty. In Madison County, fewer families (8.5 percent) reported an income below the poverty level with 32.9 percent of households headed by females falling into that category. In both counties, the majority of residents were employed in wholesale and retail trade industries: 22.5 percent for St. Clair and 21.9 percent for Madison. In St. Clair, 14.0 and 10.7 percent were employed in manufacturing and health care, respectively; in Madison, the percentages for manufacturing and health care were 21.3 and 8.6 percent, respectively. Two indicators are commonly used to report levels of crime and subsequent police response: the number of crimes known to law enforcement as having occurred within a particular jurisdiction and the number of arrests made. Both of these indicators were considered for the jurisdictional area covered by the Homicide and Violent Crime Strike Force. During 1991, 25,504 serious crimes were known to police working in the 2 counties as having occurred within their jurisdictions. Of these, 14.2 percent were violent and 85.8 percent were property related. From 1982 to 1995, a sizable increase in the number of serious crimes known to law enforcement was observed in St. Clair County (40.6 percent), while Madison County experienced a decline. During the same period, the majority of violent crimes that occurred within these two counties occurred in St. Clair County; on average, less than 20 percent of all violent crimes known to police over this time period occurred in Madison County. Based on Illinois Uniform Crime Reports (IUCR) data, law enforcement agencies within the 2 counties arrested an average of 5,073 individuals each year for serious crimes between 1982 and 1995. The majority of individuals arrested each year were from St. Clair County. Similar findings were revealed when only violent crime arrests were considered--that is, while the 2 counties averaged 1,163 arrests involving violent crimes each year, the majority of such arrests originated in St. Clair County. Madison County reported a higher incidence of arrests for criminal sexual assault while St. Clair County reported higher numbers of arrests for murder, robbery, and aggravated assault. During the years 1992 through 1996, the territory covered by the Homicide and Violent Crime Strike Force weathered congressional redistricting, loss of industries (plant closings and the flood of 1993) in the region, political scandals, and acknowledgment of educational inadequacies. In 1994, the East St. Louis (ESL) School District 189 was taken over by the state of Illinois after it was revealed that the district was operating $2.7 million in debt and holding classes in unsafe, dilapidated buildings. However, it also received millions of dollars in revenue generated by gambling in the most impoverished city, East St. Louis, and a new railway system that linked St. Clair County with downtown St. Louis and the St. Louis airport (Metro-Link Rail). Each of these factors may have affected the ease with which the task force was able to interact with communities and their political leaders. Also during these years, other law enforcement initiatives operating throughout the area may have affected the impact of the task force in realizing its goals. Goals and Objectives The program goals for the Illinois State Police component were to: o Select experienced homicide investigators who also had significant ties to the communities in which the task force operated. o Obtain the involvement of local law enforcement officers, especially those of the ESL Police Department. The program goals for the Illinois Attorney General's Office were to: o Assist local prosecutors in the prosecution of task force cases. o Give legal advice to the State Police component. o Take a lead role as requested in the prosecution of task force cases. The objectives of the evaluation of the Homicide and Violent Crime Strike Force in St. Clair and Madison Counties included: o Documenting and examining the original goals of the task force, its initial operating procedures, practices, organizational structure, and resource allocation, as well as its internal and external relationships. o Documenting and examining changes in the structure, procedures, practices, resources, and relationships that occurred over time. o Documenting and examining the impact of the task force on cases, law enforcement, prosecutors, the judicial system, and the communities in which it operated. ---------------------------- The main activity of the task force was investigating homicides and other violent crimes (which had already been investigated by other agencies in the two counties) by combining the resources of ISP and the AG's Office. ---------------------------- Program Activities/Components The main activity of the task force was investigating homicides and other violent crimes (which had already been investigated by other agencies in the two counties) by combining the resources of ISP and the AG's Office. The AG's Office investigator was the primary source for the selection of cases for the task force. The investigator reviewed unsolved homicide and violent crime cases and then selected cases to be moved directly to the ISP squad leader. The perceived solvabililty of a case was the greatest determinant in whether the case was accepted by the task force. As the task force evolved, cases were also referred by local law enforcement agencies. The task force, however, did engage in other activities. For example, it investigated four shootings by local police officers. In addition, investigators from the task force rendered assistance to ISP, federal law enforcement agencies, and local police departments. Because this assistance did not entail extensive involvement of the task force, neither the task force nor the evaluation counted these "assist cases" as task force cases. The task force experienced relative stability in both resources and personnel. Two ISP case agents and one AG's Office attorney transferred out of the task force. One of the case agents and the attorney left during the first year of operations. The other case agent left in 1995. There were no other personnel changes. Resources were also stable and generally deemed adequate to accomplish the job. As the two components of the task force adjusted to and more clearly defined their roles in the internal operations of the task force, operating procedures changed. The original design called for joint decisionmaking between both components regarding whether to initiate investigation, proceed further with a case, or ask for an arrest warrant and prosecution of a case. During the first year of task force operations, the process moved away from attorney participation in these decisions in all cases to consultation of the attorneys by the ISP component on an as-needed basis. Within the first year of operations, an understanding was reached whereby the ISP component controlled task force activities up to the point of arrest and the AG's Office component controlled task force activities after arrest. The task force was unable to realize its goal of local law enforcement participation from the ESL Police Department. At the time of task force initiation, the department was understaffed and coping with the demands of high violent crime rates. On the one hand, the city was financially unable to contribute resources to replace any officers who might join the task force; on the other hand, the task force was not able to pay any local officers who wanted to participate in task force activities and was thus only able to obtain the services of one local officer from the Alton Police Department. He was recalled after 9 months. Performance Measures and Evaluation Methods The evaluation included four parts. First, the evaluation team evaluated the process through which the task force was designed and implemented. This evaluation was divided into two sections--one describing the initiation and design of the task force and one describing the evolution of the task force since its inception. Second, the team examined the impact of the task force on the cases handled, law enforcement and the judicial process, and the larger community. Third, by combining the information gained in the implementation and impact evaluations, the team assessed whether the task force was a viable approach to the investigation and prosecution of homicides and violent crimes. Fourth, the evaluation team made recommendations for the future development of the task force in Madison and St. Clair Counties and for those interested in starting similar task forces elsewhere. A variety of strategies was used to obtain the information needed to describe the operating procedures and practices of the task force. Two sources of information were central to the evaluation: task force program documents maintained by the Illinois Criminal Justice Information Authority (ICJIA) and interviews with task force participants. The program documents obtained from ICJIA included task force grant applications, correspondence between ICJIA and the task force, and the task force's monthly data reports. UCR data were also used in the evaluation. Additional information was obtained from personal interviews with ISP and AG's Office administrators, task force personnel, and other agencies or individuals involved in the creation and development of the program. These interviews were based on written protocols developed by the evaluation team in conjunction with ICJIA. The impact portion of the report relied on quantitative and qualitative data collected to describe the effectiveness of the task force program. This three-part analysis investigated the impact of the task force on the cases chosen for investigation; law enforcement, prosecutors, and the courts; and the broader community. The goals of the case analysis were to describe the cases handled by the task force and to examine the movement of those cases through critical decision points in the criminal justice process. In assessing the impact of the task force on both criminal justice staff and the broader community, evaluation project staff interviewed task force personnel, local law enforcement personnel, prosecutors, judges, defense atto