Massachusetts Sentencing Commission Recidivism Study 2002
Download original document:
Document text
Document text
This text is machine-read, and may contain errors. Check the original document to verify accuracy.
Comprehensive Recidivism Study A Report to the House Committee on Ways and Means Senate Committee on Ways and Means Joint Committee on Criminal Justice and the Joint Committee on Public Safety Massachusetts Sentencing Commission Honorable Robert A. Mulligan, Chairman June 1, 2002 Massachusetts Sentencing Commission 90 Devonshire Street, Room 2001 Boston, MA 02109 Voice: 617-788-6867 Fax: 617-788-6885 Comprehensive Recidivism Study A Report to the House Committee on Ways and Means Senate Committee on Ways and Means Joint Committee on Criminal Justice and the Joint Committee on Public Safety Massachusetts Sentencing Commission Honorable Robert A. Mulligan, Chairman June 1, 2002 MEMBERSHIP OF THE MASSACHUSETTS SENTENCING COMMISSION Judges Parole Board Honorable Robert A. Mulligan, Chairman Associate Justice, Superior Court Michael J. Pomarole, Esq. Chairman, Parole Board Honorable Margaret R. Hinkle Associate Justice, Superior Court Public Safety James P. Jajuga, Secretary Executive Office of Public Safety Honorable Mark H. Summerville Associate Justice, Boston Municipal Court Victim Witness Board Maria F. Rodriguez, Esq. Chief of the Family Protection Unit Hampden County District Attorney's Office Prosecutors Honorable S. Jane Haggerty Associate Justice, Superior Court Department of Correction Pamela L. Hunt, Esq. Office of the Attorney General David Slade, Esq. Office of the General Counsel Department of Correction Geline W. Williams, Esq. Executive Director MA District Attorneys’ Association Probation John J. O’Brien, Commissioner Office of the Commissioner of Probation Defense Michael J. Traft, Esq. Private Attorney, Boston District Court Liaison Honorable Timothy Gailey Presiding Justice Chelsea District Court William W. Robinson, Esq. Committee for Public Counsel Services Thomas G. Murray, Esq. Private Attorney, Boston (deceased) Staff Francis J. Carney, Jr., Executive Director Linda K. Holt, Research Director Lee M. Kavanagh, Research Analyst Valerie C. Caldwell, Executive Assistant Sheriffs Association Mariellen H. Fidrych Executive Director MA Sheriffs’ Association i ii Acknowledgments This research would not have been possible without the cooperation of administrators and the assistance of researchers and practitioners in several criminal justice agencies and entities. The Commission would like to acknowledge that cooperation and assistance as follows: Franklin County House of Correction Department of Correction Sheriff Frederick B. Macdonald Lt. John Zewski Michael Maloney, Commissioner Rhiana Kohl, Ph.D., Research Director Hollie A. Matthews Hoover, Research Analyst Hampden County House of Correction Sheriff Michael J. Ashe, Jr. J. John Ashe, Deputy Superintendent Martha A. Lyman, Ed.D., Research Director Massachusetts Parole Board Michael J. Pomarole, Chairman Timothy F. App, Executive Director Sallyann V. Sweeney, Director of Support Services Joseph C. Balbo, Systems Analyst Norfolk County House of Correction Sheriff Michael Bellotti George Klier, Assistant Deputy Superintendent Office of the Commissioner of Probation Plymouth County House of Correction John J. O’Brien, Commissioner Sheriff Joseph F. McDonough Lt. Brian C. Case Office of Community Corrections Stephen V. Price, Executive Director Phyllis Buccio-Notaro, Statewide Program Supervisor Vincent Lorenti, Program Specialist John Quinn, Administrative Staff Attorney Sam Bellistri, Regional Manager Pamerson Ifill, Regional Manager Kevin Kearney, Regional Manager John Monahan, Regional Manager Suffolk County House of Correction Sheriff Richard J. Rouse Patrick Bradley, Deputy Sheriff Stefan Lobuglio, Deputy Superintendent Worcester County House of Correction Sheriff John M. Flynn iii Honorable Joseph I. Dever First Justice, Lynn District Court Community Corrections Centers Maureen Richitelli, Manager Barnstable County, Barnstable Honorable Michael C. Lauranzano Associate Justice, Lynn District Court Rick Kassiotis, Jr., Manager Essex County, Lawrence Honorable Herbert H. Hodos First Justice, Greenfield District Court Jack Fitzgerald, Manager Hampden County, Springfield Honorable M. John Schubert, Jr. Acting First Justice, Orange District Court Tim Simons, Manager Hampshire County, Northampton Honorable Robert V. Greco First Justice, Framingham District Court Shawn Murray, Manager Middlesex County, Cambridge Honorable Sydney Hanlon First Justice, Dorchester District Court Richard McDonough, Manager Norfolk County, Quincy Honorable Milton L. Wright, Jr. First Justice, Roxbury District Court Patty Campatelli, Manager Suffolk County, Boston Honorable Robert P. Ziemian First Justice, South Boston District Court Jim Wright, Manager Worcester County, Fitchburg Worcester County, Webster Criminal History Systems Board Barry Lacroix, Executive Director Drug Courts Administrative Office of the Trial Court Honorable Robert A. Cornetta Chairman, Drug Court Working Group Associate Justice, Salem District Court William Marchant Acting Director, Fiscal Affairs Jerome S. Berg, Court Administrator District Court Department Student Interns Paul Q. O'Donnell, Regional Coordinator District Court Region II Amanda E. Wall, College of the Holy Cross Kara Fielder, Northeastern University Honorable Kevin M. Herlihy First Justice, Haverhill District Court iv Executive Summary This report was prepared to meet the legislative requirement in Chapter 177 of the Acts of 2001 which directed the Massachusetts Sentencing Commission to “perform a comprehensive study on the recidivism rate of offenders.” The completed project is the culmination of a collaborative effort of a variety of criminal justice agencies. A total of thirty different criminal justice agencies and programs contributed data to this project or assisted in the compilation of the data and report. The study reports on a one year follow-up of a sample of 3,751 offenders representing drug courts, community corrections centers, houses of correction, Department of Correction, and the Parole Board. The study defined a recidivist as: For offenders released from incarceration: • • an offender with a new arrest (arraignment) in the year following release from a correctional facility; or, an offender with a technical violation of probation or parole that resulted in incarceration in the year following release from a correctional facility; For offenders in community corrections centers or drug court programs: • • an offender with a new arrest (arraignment) in the year following the date of entry into a community corrections center or drug court program; or, an offender with a technical violation of probation or parole that resulted in incarceration in the year following the date of entry into a community corrections center or drug court program. Based on this definition, the one year recidivism rate was estimated to be 49.1% for all samples. The report discusses the method used to measure recidivism and presents data on recidivism rates for each of the criminal justice cohorts. The report provides further information on the nature of the recidivism behavior and explores the relationship between recidivism rates and program involvement, current offense, criminal history, and demographic characteristics. Future research is suggested that would build on the model presented here but address some of the limitations of the present analysis (scope of the sample, length of the followup period, and limited program information). v vi Table of Contents INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 RECIDIVISM RATES - MASSACHUSETTS AND NATIONAL RESEARCH . . . . . . 2 METHOD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 SAMPLE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Sampling Time Frame . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 Originating Criminal Justice Agencies . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 DATA COLLECTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 Recidivism Variables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 Program Involvement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 Offender Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 DATA EXTRACTION AND DATA LINKAGES . . . . . . . . . . . . . . . . . . . . . . . 15 DATA ANALYSIS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 STUDY LIMITATIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 FINDINGS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 RECIDIVISM RATES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 Type of Recidivist . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 Time Until Recidivism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22 New Arrest Offenses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25 PROGRAMS, SECURITY, AND COMMUNITY RELEASE . . . . . . . . . . . . . . 28 Correctional Facilities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28 Community Corrections Centers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30 Drug Courts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30 DEMOGRAPHIC CHARACTERISTICS OF OFFENDERS . . . . . . . . . . . . . . . 33 Gender . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33 Age . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33 Race / Ethnicity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 CURRENT OFFENSE AND CRIMINAL HISTORY . . . . . . . . . . . . . . . . . . . . 38 Current Offense . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38 Type of Offense . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38 Offense Seriousness Level . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 Criminal History . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42 Criminal History Groups . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42 Juvenile Criminal History . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43 Sentencing Grid . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46 vii SUMMARY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48 APPENDIX - METHODOLOGY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51 SAMPLE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51 DATA COLLECTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53 Recidivism Variables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53 New Offenses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54 Incarceration Status . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56 Parole Violation Status . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56 Probation Violation / Other Status . . . . . . . . . . . . . . . . . . . . . . . . 58 Program Involvement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58 Offender Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58 DATA EXTRACTION AND DATA LINKAGES . . . . . . . . . . . . . . . . . . . . . . . 60 DATA ANALYSIS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60 APPENDIX - HOUSE OF CORRECTION SENTENCED OFFENDERS . . . . . . . . . . 66 List of Figures Figure 1. Sentencing Guidelines Grid . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 viii List of Tables Table 1. Originating Criminal Justice Agencies Included in Study Sample . . . . . . . . . . . 6 Table 2. Type of Sentence and Correctional Custody at Time of Release, Correctional Releases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 Table 3. Recidivism Classification by Criminal Justice Cohort . . . . . . . . . . . . . . . . . . . 21 Table 4. New Arrest Status, Parole Supervision Status, and Community Corrections Center Participation Status . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 Table 5. Time Until Recidivism Incident by Criminal Justice Cohort . . . . . . . . . . . . . . 24 Table 6. Offense Seriousness Level of Most Serious New Arrest Offense by Criminal Justice Cohort . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26 Table 7. Type of Offense of Most Serious New Arrest Offense by Criminal Justice Cohort . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 Table 8. Recidivism Status by Security Level at Time of Release and Criminal Justice Cohort, Correctional Releases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 Table 9. Recidivism Status by Selected Program Characteristics, Community Corrections Center Participants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 Table 10. Recidivism Status by Selected Program Characteristics, Drug Court Participants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32 Table 11. Recidivism Status by Gender and Criminal Justice Cohort . . . . . . . . . . . . . . 35 Table 12. Recidivism Status by Age and Criminal Justice Cohort . . . . . . . . . . . . . . . . . 36 Table 13. Recidivism Status by Race / Ethnicity and Criminal Justice Cohort . . . . . . . 37 Table 14. Recidivism Status by Type of Offense (Current Offense) and Criminal Justice Cohort . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40 Table 15. Recidivism Status by Offense Seriousness Level (Current Offense) and Criminal Justice Cohort . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41 Table 16. Recidivism Status by Criminal History and Criminal Justice Cohort . . . . . . 44 Table 17. Recidivism Status by Juvenile Criminal History and Criminal Justice Cohort . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45 Table 18. Recidivism Status by Sentencing Grid Cell and Grid Assignment, All Offenders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47 Table 19. Non-hierarchical Recidivism Classification by Criminal Justice Cohort . . . . 62 Table 20. Alternate Recidivism Measures by Criminal Justice Cohort . . . . . . . . . . . . . 63 Table 21. Conviction Status by Time Until Recidivism and Criminal Justice Cohort . . 65 Table 22. Selected Recidivism Characteristics by Type of Release and Custody at Time of Release, House of Correction Sentenced Offenders . . . . . . . . . . . . . . . . . . . . 67 Table 23. Selected Offender Characteristics by Type of Release and Custody at Time of Release, House of Correction Sentenced Offenders . . . . . . . . . . . . . . . . . . . . . . 69 ix x MASSACHUSETTS SENTENCING COMMISSION COMPREHENSIVE RECIDIVISM STUDY A REPORT TO THE HOUSE COMMITTEE ON WAYS AND MEANS SENATE COMMITTEE ON WAYS AND MEANS JOINT COMMITTEE ON CRIMINAL JUSTICE AND THE JOINT COMMITTEE ON PUBLIC SAFETY INTRODUCTION Chapter 177 of the Acts of 2001 (Chapter 177), the FY2002 budget, directed the Massachusetts Sentencing Commission (Commission) to “perform a comprehensive study on the recidivism rate of offenders.” The legislature established a June 1, 2002 deadline for this study. This report was prepared to meet this legislative mandate. As a measure of the subsequent involvement of offenders in the criminal justice system following release from incarceration or following participation in a criminal justice program, recidivism is a commonly used indicator of the outcome of various criminal justice initiatives. This document reviews previous research on recidivism in Massachusetts, presents the research method used in the present analysis, and discusses the findings from this research initiative. 1 Massachusetts Sentencing Commission RECIDIVISM RATES - MASSACHUSETTS AND NATIONAL RESEARCH Studies of recidivism are both common in the criminal justice literature and diverse in their definition and structure. For purposes of this report, numerous studies were reviewed. The studies reported here concentrate on recent findings from Massachusetts research. One national study is also reviewed. The Department of Correction (DOC) has a long tradition of reporting recidivism rates for offenders released from DOC facilities. The most recent report presented three-year recidivism rates for offenders released in 1995. The DOC defined a recidivist as an offender re-incarcerated for at least 30 days during the three-year follow-up period.1 This definition included those re-incarcerated after conviction of a new offense, as well as those re-incarcerated for a violation of the conditions of their parole or probation. After one year the recidivism rate was 22%; after two years the recidivism rate was 37%; and after three years the recidivism rate was 44%.2 One year recidivism rates were available for prior years and had decreased from 29% for offenders released in 1990 to 22% for offenders released in 1995.3 An analysis of the long term trends in DOC recidivism rates indicated that age at incarceration was the variable most strongly associated with recidivism risk for the DOC release population.4 As the proportion of offenders in the high risk age group (those offenders who began their incarceration at age 25 or younger) decreased, the overall recidivism rate decreased as well. The Hampden County Sheriff’s Department initiated a research program to study the recidivism of offenders released from the correctional facilities in Hampden County with a goal to provide “objective, empirical data necessary for . . . planning, operation and evaluation.”5 A study of the recidivism rates for offenders released in 1998 and 1999 defined a recidivist as an offender re-incarcerated during the follow-up period in a state or 1 Hollie A. M atthews Ho over, The Background Characteristics and Recidivism Rates of Releases from Massachusetts Correctional Institutions During 1995, Massachusetts Department of Correction, May 2000, page 1. 2 Ibid., page 18. 3 Ibid., page 5. 4 Daniel P . LeClair, Recent C hange s in Rates o f Recidivism for Releas es from M assachu setts State Correctional Institutions: A n Exploratory Evaluation o f Possible Causa l Factors, Stonehill College, Fall 1998, page 16. 5 Martha A . Lyman, A Comprehensive Study of Recidivism at the Hampden County House of Correction, Hampden County Sheriff’s Department, February 23, 2001, page 1. 2 Comprehensive Recidivism Study county correctional facility in Massachusetts for either a new offense or violation of release conditions.6 For those offenders released in 1998, the recidivism rate after one year was 12.0% and the recidivism rate after two years was 28%; and, for those offenders released in 1999, the recidivism rate after one year was 21.9%.7 The Hampden County study also reported re-arraignment rates and re-conviction rates. The proportion of offenders with a new arraignment within one year of release was 44.7% and 52.7% for the 1998 and 1999 cohorts, respectively. 8 The proportion of offenders with a new conviction within one year of release was 23.1% and 31.7% for the 1998 and 1999 cohorts, respectively. 9 The study attributed the change in recidivism rates between the 1998 and 1999 cohorts to a change in the definition of recidivism for parole and probation violators and to changes in the policies and practices of other criminal justice agencies. Massachusetts is similar to other jurisdictions in the recidivism rates that have been observed by the DOC and Hampden County. A national study followed offenders released from 11 large states for a period of three years. The study found that 62.5% had been rearrested, 46.8% had been re-convicted, and 41.4% re-incarcerated by the end of the three years.10 This study also found that the rate of re-arrest was highest during the first year following release. These studies indicate some of the important themes found in the recidivism literature: recidivism rates vary by the type of behavior that is considered (arrest, conviction or incarceration); recidivism rates vary by the length of the follow-up period; recidivism rates vary by the composition of the offender population under study; and, recidivism rates vary over time in response to changes in policy and practices in other criminal justice agencies. While the recidivism rates derived in the current analysis will not be directly comparable to these studies due to differences in the definition of recidivism, differences in the population that was considered, and differences in the sampling time period, they serve as a useful and important backdrop for the current study. 6 Ibid., page 3. 7 Ibid., pages 10 and 38. 8 Ibid., page 10. 9 Ibid. 10 Allen J. Be ck and B ernard E . Shipley, Recidivism of Prisoners Released in 1983, Bureau of Justice Statistics, February 1997, page 1. 3 Massachusetts Sentencing Commission METHOD This research project is the result of a collaborative effort of a number of criminal justice agencies. The approach selected for this research was based on the assumption that reliable recidivism information could be collected in a semi-automated fashion by linking a number of data-bases from cooperating criminal justice agencies in the commonwealth: • a sample of offenders released following a period of incarceration and a sample of offenders participating in community based criminal justice programs was provided by the relevant originating criminal justice agency - DOC, Parole Board, sheriffs departments, drug courts, and Office of Community Corrections; • a measure of new arrests was obtained by linking the relevant offender record from the originating criminal justice agency to the CARI (Court Activity Record Information) data-base. New arrests were defined as arrests and arraignments that occurred after the release from incarceration or entry date into the community based program; • a measure of parole violations was obtained by linking the relevant offender record from the originating criminal justice agency to the PATS (Parole’s Automated Tracking System) data-base; and, • a measure of probation violations was obtained by linking the relevant offender record from the originating criminal justice agency to the CARI data-base. This section contains a brief discussion of the sample, the data collection process, the data extraction process, and the data analysis. Additional details on the method are contained in the Appendix. SAMPLE Chapter 177 directed the Commission to include “offenders who have been, or currently are, incarcerated and/or sentenced to the facilities and/or programs of: • • • the Department of Correction, the respective county houses of correction, the Parole Board, 4 Comprehensive Recidivism Study • • those community correction centers established pursuant to chapter 211F that have been fully operational for a period of at least one calendar year, and any drug courts, so-called, operating within the district courts of the commonwealth.” Sampling Time Frame. In order to allow for a reasonable follow-up period and to ensure that the project was completed within the time-frame established by Chapter 177, the sample was defined as offenders released / program participants during the period April 2000 through June 2000. Some of the considerations in the development of the sample were: • The period following release / program participation was one year. A longer follow-up period was not feasible due to the reporting deadline, the newness of some of the programs, and the scope of the sample. While recidivism status was computed at periods of less than one year (e.g. 6 months), the data collection and analysis focused on an initial follow-up period of one year. • Data collection for recidivism was done no earlier than 15 months following release / program participation; and, • A release / program participant cohort of one calendar quarter allowed for a reasonable sample size of 3,751 offenders for the initial study effort. Originating Criminal Justice Agencies. The cooperation of the originating criminal justice agencies was essential to the identification of the release / program participation samples. There were no known data sources except the originating criminal justice agencies that could have provided this information. A total of 3,751 offender records was included in the study sample. Table 1 shows the specific sites participating in the study. 5 Massachusetts Sentencing Commission Table 1. Originating Criminal Justice Agencies Included in Study Sample Release Populations Department of Correction All DOC Facilities Houses of Correction Franklin County House of Correction Hampden County House of Correction Norfolk County House of Correction Plymouth County House of Correction Suffolk County House of Correction Worcester County House of Correction Massachusetts Parole Board All Parolees Community B ased Pro grams Community Corrections Centers Barnstable County Community Corrections Cent er in Barnstable Essex County Community Corrections Center in Lawrence Hampden County Community Corrections Cent er in Springfield Hampshire County Community Corrections Center in Northampton Middlesex County Community Corrections Center in Ca mbridge Norfolk County Community Corrections Center in Quincy Suffolk County Community Corrections Center in Boston Worcester County Community Corrections Center in Fitchburg Worcester County Co mmunity Correct ions Center in Webster Drug Courts Dorchester District Court Framingham District Court Greenfield District Court Haverhill District Court Lynn District Court Orange District Court Roxbury District Court South Boston District Court Department of Correction. The DOC provided a sample of 705 offenders released during the period April 2000 through June 2000. This represented 18.8% of the study sample and included all offenders released via parole and discharge from a DOC facility. Table 2 shows additional details on the type of sentence and gender for the DOC offenders in the study sample. Houses of Correction. Six houses of correction provided samples of offenders released during the period April 2000 through June 2000. The total sample from these six houses of correction was 2,396 representing 63.9% of the study sample. An estimate of the proportion of the house of correction population that was included in this study was derived from a five-year sample of admissions to all houses of correction. The six 6 Comprehensive Recidivism Study participating houses of correction represented 56.3% of all admissions during the five year period and therefore it was estimated that the current study included 56.3% of all releases from houses of correction during the sample period.11 Because the Massachusetts Parole Board provided data on all parolees state-wide, it was assumed that the study sample included all parolees from houses of correction and 56.3% of the discharges. Table 2 shows additional details on the type of sentence and gender for the house of correction offenders in the study sample. Parole Board. The Parole Board provided a state-wide sample of 776 offenders released to parole supervision during the period April 2000 through June 2000. This represented 20.7% of the study sample. This sample included all offenders released to parole supervision from houses of correction and the DOC. This sample included offenders released to parole supervision from the houses of correction participating in the study as well as offenders released to parole supervision from all other houses of correction in Massachusetts. Of the 776 offenders released to parole supervision, 222 were paroled from DOC facilities, 303 were paroled from participating houses of correction, and 251 were paroled from the non-participating houses of correction. Table 2 shows additional details on the type of sentence, gender, and facility at the time of release for the offenders released to parole supervision in the study sample. Community Corrections Centers. Eligible community corrections centers provided samples of offenders who began a community corrections center placement during the period April 2000 through June 2000. The Office of Community Corrections (OCC) was established under G.L. c. 211F. The mission of OCC is “the establishment of intermediate sanctions programs which offer a continuum of sanctions and services for probation, sheriffs, parole and the Department of Correction.”12 The intermediate sanctions are based at the community corrections centers in operation across the state. They are described by OCC as follows: Community Corrections Centers are community based, intensive supervision sites, which deliver bundled sanctions and services, including treatment and education, to high risk offenders via Intermediate Sanction 11 Robert J . Tenaglia, New Court Commitments to Massachusetts County Correctional Facilities During 2000, Massachusetts Department of Correction, September 2001, page 22. 12 Administrative Office of the T rial Court, O ffice of Com munity Corr ections, Overview of the Department, http://www.state.m a.us/courts/ad min/occ/o ccovervie w.html. 7 Massachusetts Sentencing Commission Levels. Among the sanctions delivered at community corrections centers are: • • • • electronic monitoring community service drug & alcohol testing day reporting Among the services provided at community corrections centers are: • • • • • • • substance abuse treatment GED/ABE/ESL or comparable educational component communicable disease prevention education job readiness training and placement referral to Department of Public Health or Department of Mental Health service providers women's services bilingual services Community corrections centers are designed to provide a criminal justice solution for a specific group of offenders. Intermediate Sanction Level III, IV is indicated for those offenders who possess a serious criminal history and are chronic substance abusers . . . Intermediate Sanction Level IV is the most intense level of community based, criminal justice supervision. Sanctions and services required at this level of supervision represent a twenty-four hour restriction upon the liberty of the offender. Level IV participants are required to report to the community corrections center for four to six hours per day, six days per week. Additionally, offenders placed at Intermediate Sanction Level IV are monitored twenty-four hours per day via electronic device, required to submit to the highest category of random drug and alcohol testing, and mandated to attend two four hour community service shifts per week. Intermediate Sanction Level III is an intense level of community-based, criminal justice supervision. Sanctions and services required at this level of supervision represent a daily imposition upon the liberty of the offender. Level III participants are required to report to the community corrections center for one to four hours per day, three to five days per 8 Comprehensive Recidivism Study week. Offenders placed at Intermediate Sanction Level III may be monitored via electronic device. Level III also requires random drug and alcohol testing, and attendance at one four hour community service shift per week.13 Only those community corrections centers in operation by January 2000 were considered eligible for participation in the study and all such centers contributed data. A total of 323 offenders began participation in an eligible community corrections center during the study period. This represented 8.6% of the study sample. Drug Courts. Eligible drug courts provided samples of offenders who began a drug court placement during the period April 2000 through June 2000. The annual report of the Trial Court describes a drug court as: A drug court is not a separate court, but rather a special session of a District Court. The drug court session usually addresses serious drug offenders for whom a special court intervention is likely to be the last stop before incarceration. Drug court typically involves a strenuous regimen of treatment and accountability on the part of the offender, coupled with the strong personal commitment by the offender to take control of his or her life situation and eliminate drug use. Among the key elements to an effective drug court program are intensive probationary and therapeutic programming activities, frequent drug testing, and careful monitoring of progress by the drug court judge.14 Only those drug courts in operation by January 2000 were considered eligible for participation in the study and all such courts contributed data. A total of 76 offenders began participation in these eight drug courts during the study period. This represented 2.0% of the study sample. Of the 76 drug court program participants, 20 or 26.3% began the drug court program under the probation portion of a house of correction / split sentence. 13 Ibid. 14 Massac husetts Supre me Judicia l Court, Annual Report on the State of the Massachusetts Court System Fiscal Yea r 2000, page 98. 9 Massachusetts Sentencing Commission Criminal Justice Cohort In each of the tables, the study sample was divided into the following criminal justice cohorts: Community Based Program Participants: • drug court program participants; • community corrections centers program participants; Offenders released after incarceration: • house of correction sentenced / released via discharge; • house of correction sentenced / released via parole; • state prison sentenced / released via discharge; and, • state prison sentenced / released via parole. The study sample of offenders released after incarceration in a house of correction or the DOC included offenders who had served either a house of correction sentence, which ranges from one day to 2 ½ years, or a state prison sentence, which ranges from one year to life. For purposes of this analysis, recidivism rates were considered by the type of sentence for which the offender was incarcerated (house of correction or state prison) and by the type of release (parole or discharge). As indicated in Table 2, of the 705 offenders released from DOC facilities an estimated that 492 or 69.8% had served a state prison sentence and 213 or 30.2% had served a house of correction sentence prior to release. Many female offenders serving a house of correction sentence served that sentence at a DOC facility and were released from the DOC. Some male offenders serving a house of correction sentence were transferred to the DOC and released from a DOC facility. Of the 213 offenders released from the DOC who had served a house of correction sentence, 165 or 77.5% were female and 48 or 22.5% were male. For purposes of this analysis, those offenders released from a DOC facility after serving a house of correction sentence were included in the house of correction sentence cohort. The statistical appendix contains tables that disaggregate those offenders released after serving a house of correction sentence by the correctional custody at the time of release (DOC or house of correction). As indicated in Table 2, of the 2,396 offenders released from houses of correction an estimated 2,393 or 99.9% had served a house of correction sentence and 3 or 0.1% had served a state prison sentence prior to release. For purposes of this analysis, the three 10 Comprehensive Recidivism Study offenders released from a house of correction after serving a state prison sentence were included in the state prison sentence cohort. The statistical appendix contains tables that disaggregate those offenders released after serving a house of correction sentence by the correctional custody at the time of release (DOC or house of correction). Finally, as indicated in Table 2, there were 251 offenders released by parole from the houses of correction that did not participate in the study. Of these 248 were released by parole after serving a house of correction sentence and 3 were released by parole after serving a state prison sentence. The proportion of offenders released by parole varied by type of sentence and correctional custody at the time of release: • 32.3% of the offenders released from a DOC facility after serving a state prison sentence were paroled and 67.7% were discharged; 29.6% of the offenders released from a DOC facility after serving a house of correction sentence were paroled and 70.4% were discharged; and, 12.5% of the offenders released from houses of correction after serving a house of correction sentence were paroled and 87.5% were discharged. • • It should be noted that offenders with house of correction sentences of less than two months were not eligible for parole. In FY 2000 an estimated 29.3% of all house of correction sentences were for less than two months.15 15 Massac husetts Senten cing Com mission, Survey of Sentencing Practices, November 2001, page 35. 11 Massachusetts Sentencing Commission Table 2. Type of Sentence and Correctional Custody at Time of Release, Correctional Releases Custody a t Time of Relea se and Gender Released from DO C Facility Female Male Sub-Total House of Correction Sentence State Prison Sentence Discharge Parole Total Discharge Parole Total Total 127 23 150 38 25 63 165 48 213 14 319 333 9 150 159 23 469 492 188 517 705 132 1961 2,093 43 257 300 175 2,218 2,393 0 0 0 0 3 3 0 3 3 175 2,221 2,396 0 0 0 22 226 248 22 226 248 0 0 0 1 2 3 1 2 3 23 228 251 2,243 611 2,854 333 165 498 3,352 Released from Participating House of Correction Female Male Sub-Total Released by Parole from Other House of Correction Female Male Sub-Total Total 12 Comprehensive Recidivism Study DATA COLLECTION The data that was collected can be classified in three general areas: • • • recidivism - those measures of the criminal justice involvement by the offender in the period following release / program participation; program involvement - those measures (e.g. security level, intermediate sanction level) that described the criminal justice program in which the offender was involved prior to release or during program participation; and, offender background - those demographic, offense, and criminal history variables that describe the offender and that the research literature suggested were related to the risk of re-offending. Recidivism Variables There are many different measures of recidivism. The length of the follow-up period may vary. In the period following release / program participation, recidivism can include measures of new criminal activity as well as technical violations of supervised release. In terms of new criminal activity, recidivism can be defined from the presence of a new arrest or new conviction, to a new sentence to incarceration. In general, recidivism rates will vary according to the length of the follow-up period and the definition of recidivism used. The study defined a recidivist as: For offenders released from incarceration: • • an offender with a new arrest (arraignment) in the year following release from a correctional facility; or, an offender with a technical violation of probation or parole that resulted in incarceration in the year following release from a correctional facility; For offenders in community corrections centers or drug court programs: • • an offender with a new arrest (arraignment) in the year following the date of entry into a community corrections center or drug court program; or, an offender with a technical violation of probation or parole that resulted in incarceration in the year following the date of entry into a community corrections center or drug court program. 13 Massachusetts Sentencing Commission For all offenders in the sample new arrest information was collected from the CARI database, probation violation information was collected from the CARI data-base, and parole violation information was collected from the Parole Board file. Those offenders under parole or probation supervision may be returned to custody as the result of a new arrest or some other technical violation of the conditions of parole or probation. Technical violations can range from curfew violations, failure to attend and complete required programs, or failing a drug test. Recidivism rates for offenders under parole or probation supervision include returns to custody for technical violations in addition to new arrests that occurred during the follow-up period. While considered as part of the overall recidivism rates, the ability to return an offender to custody on a technical violation before an arrest for a new crime is an important tool to promote public safety. Program Involvement Chapter 177 directed that the study consider the relationship between criminal justice programs and recidivism rates: . . . said study shall include, but not be limited to, information regarding recidivism rates, by program and facility, including an analysis detailing the effect of pre-release, post-release, diversionary and intermediate sanction rehabilitative/supervisory programs on said recidivism rates . . . Many of the originating criminal justice agencies provided programs and facilities at a variety of security levels. Where available, this information was included in the analysis. Offender Background It was important to consider offender background variables in relation to observed recidivism rates. The study included demographic, current offense, and criminal history information. The study did not develop a comparison sample or control group. It was not assumed that these were comparable offender groups with respect to recidivism risk. Throughout the analysis, comparisons were made among the populations from the various originating criminal justice agencies that indicate the substantial differences in these populations that were related to the observed differences in recidivism rates. The demographic characteristics of the population that were collected included: gender, race / ethnicity, and age. The source of offender demographic information was the CARI data-base. 14 Comprehensive Recidivism Study The current offense and criminal history associated with the offender’s placement in the originating criminal justice agency were collected from the adult and juvenile case records in the CARI data-base. The case records were identified through a link via the docket number where available. DATA EXTRACTION AND DATA LINKAGES. Originating criminal justice agencies were asked to provide a minimum set of data elements in order to minimize the disruption of the research to their on-going operations. The data elements included the offender’s name and date of release or date of program entry. The criminal justice agency also provided either the offender’s date of birth or the offender’s PCF number. If the participating agency provided the offender’s date of birth, the name and date of birth were used to extract the PCF number from the CARI database. The PCF number was then used to establish the link between the various data sets. The minimum set of required data elements was sufficient to establish a link with the CARI data-base and to establish a starting point for the follow-up period. In all cases, the resulting case extract was validated in order to ensure that a unique name, date of birth, and PCF number match was made. Of all cases submitted by participating agencies, only five cases or 0.1% were excluded from the sample because they were not identified in the CARI data-base. This method resulted in a very complete sample for the recidivism study. DATA ANALYSIS Some discussion of how recidivism rates were calculated, how the sample total was estimated, and how the sentencing guidelines grid was used as a framework for some sections of this analysis, may be helpful to the interpretation of the statistical information. Recidivism Rates. In most of the tables in the report, three columns are presented: the total number of cases, the total number of recidivists, and the recidivism rate. The recidivism rate is expressed as the percentage of offenders classified as recidivists of the total population of offenders. These columns are repeated for each of the criminal justice cohorts in the study. Sample Totals. In each of the tables a sample total has been calculated. No weights were used in calculating the total. This total is only representative of the sample used for 15 Massachusetts Sentencing Commission this particular study. Because not every criminal justice program was included in the analysis, the total is not representative of all offenders in the criminal justice system. In particular, not all offenders discharged from houses of correction during the sample period were included in the study sample. Further, only those drug courts and community corrections centers established prior to January 2000 were considered eligible for the study. A sample total was also calculated for the house of correction and state prison sentenced cohorts. No adjustment was made to the house of correction portion of the sample in estimating totals, even though this sample includes all offenders paroled from the houses of correction and an estimated 56.3% of the offenders discharged from the houses of correction. Sentencing Guidelines Grid. The proposed sentencing guidelines are in the form of a sentencing guidelines grid which has been used as a framework for some parts of the analysis of the recidivism data. The vertical axis of the grid is comprised of nine levels of offense seriousness, ranging from level 1, the lowest level of offense seriousness, to level 9, the highest level of offense seriousness. The horizontal axis of the grid is comprised of five categories of defendant criminal history, ranging from criminal history category A, “No/Minor Record” to criminal history category E, “Serious Violent Record”. The resulting grid, comprising a total of forty-five cells, is also divided into three sentencing zones: the “incarceration zone”; the “discretionary zone”; and, the “intermediate sanction” zone. The sentencing guidelines grid also contains a brief description of Level III and Level IV intermediate sanctions which are provided by the community corrections centers. The sentencing guidelines grid is shown in Figure 1. A detailed description of proposed sentencing guidelines can be found in the Report to the General Court.16 The definition of the criminal history groups is contained in the Appendix. STUDY LIMITATIONS There are certain limitations in the design of this research that should be acknowledged before presenting the findings. There was no possibility of utilizing an experimental design whereby offenders could be randomly assigned to treatment and control groups 16 Massac husetts Senten cing Com mission, Report to the G eneral Court , April 1996. 16 Comprehensive Recidivism Study Figure 1. Sentencing Guidelines Grid Level Illustrative Offense 9 Murder 8 Sentence Range Life Life Life Life Life Rape of Child with Force Aggravated Rape Armed Burglary 96 - 144 Mos. 108 - 162 Mos. 120 - 180 Mos. 144 - 216 Mos. 204 - 306 Mos. 7 Armed Robbery (Gun) Rape Mayhem 60 - 90 Mos. 68 - 102 Mos. 84 - 126 Mos. 108 - 162 Mos. 160 - 240 Mos. 6 Manslaughter (Invol) Armed Robbery (No gun) A&B DW (Sign. injury) 40 - 60 Mos. 45 - 67 Mos. 50 - 75 Mos. 60 - 90 Mos. 80 - 120 Mos. 5 Unarmed Rob bery Stalking ( Viol. of Orde r) Unarmed Burg lary Larceny ($50,000+) 12 - 36 Mos. IS-IV IS-III IS-II 24 - 36 Mos. IS-IV IS-III IS-II 36 - 54 Mos. 48 - 72 Mos. 60 - 90 Mos. 4 Larceny From a Person A&B DW (Mod. injury) B&E (Dwelling) Larceny ($10,000-$50,000) 0 - 24 Mos. IS-IV IS-III IS-II 3 - 30 Mos. IS-IV IS-III IS-II 6 - 30 Mos. IS-IV IS-III IS-II 20 - 30 Mos. 24 - 36 Mos. 3 A&B DW (No/minor injury) B&E (Not dwelling) Larceny ($250 to $10,000) 0 - 12 Mos. IS-IV IS-III IS-II IS-I 0 - 15 Mos. IS-IV IS-III IS-II IS-I 0 - 18 Mos. IS-IV IS-III IS-II IS-I 0 - 24 Mos. IS-IV IS-III IS-II 6 - 24 Mos. IS-IV IS-III IS-II 2 Assault Larceny Under $250 0 - 6 Mos. 0 - 6 Mos. IS-III IS-II IS-I IS-III IS-II IS-I 0 - 9 Mos. IS-IV IS-III IS-II IS-I 0 - 12 Mos. IS-IV IS-III IS-II IS-I IS-III IS-II IS-I C Serious Record 0 - 3 Mos. IS-IV IS-III IS-II IS-I D Violent or Repetitive 0 - 6 Mos. IS-IV IS-III IS-II IS-I E Serious Violent IS-III IS-II IS-I 1 Operate After Suspension Disorderly Conduct Vandalism Criminal History Scale IS-II IS-I A No/Minor Record IS-III IS-II IS-I B Moderate Record Sentencing Zone Intermediate Sanction Level Incarceration Zone Discretionary Zone (Incarceration/Intermediate Sanctions) IS-IV IS-III IS-II IS-I 24-Hour Restriction Daily Accountability Standard Supervision Financial Accountability Intermediate Sanction Zone The numbers in each cell represent the range from which the judge selects the m aximum sentenc e (Not More Than); The minimum sentence (Not Less Than) is 2/3rds of the maximum sentence and constitutes the initial parole eligibility date. 17 Massachusetts Sentencing Commission (e.g., for drug court or community corrections participants) or of matching offenders participating in certain programs with similar non-participating offenders for comparison purposes in the follow-up period. This is significant because the research did indicate some important differences in the characteristics of offenders in the various criminal justice cohorts - including differences on factors that are highly associated with recidivism risk. Therefore, it would not be valid to simply compare overall recidivism rates across the criminal justice cohorts as a measure of the relative effectiveness of these agencies or initiatives. This research was not designed to be an evaluation of the programs or agencies involved. Rather, the goal was to document outcomes on one dimension - recidivism - of the many dimensions that could be studied regarding these programs or agencies. Time and resource constraints also led to other limitations. It was not possible to compile detailed program participation data on offenders in the various cohorts so that the relationship between the degree of program involvement and recidivism could be more fully explored. For some cohorts - e.g., drug courts - the sample size was fairly small. The one year follow-up period was relatively brief and the use of arrest as the primary indicator of recidivism poses some problems (some are found not guilty or have their charges dismissed). While this research has its limitations, the fact that some thirty criminal justice agencies and programs could work together to meet the mandate of the Legislature and produce in a timely fashion a comprehensive recidivism study which adopts a uniform definition of recidivism and systematically applies it to offenders in different jurisdictions is a noteworthy accomplishment. FINDINGS The analysis of recidivism covers four major areas. First, the overall findings on recidivism rates are presented. Second, recidivism rates are considered in the context of program related variables. This is done separately for the major program groups in the study: correctional facilities (DOC and the houses of correction, including paroles and discharges), community corrections centers, and drug courts. Third, recidivism rates are considered in relation to the demographic characteristics of the population. Some comparisons of the demographic composition of the various criminal justice cohorts are provided, especially as they are related to recidivism rates. Finally, recidivism rates are considered in relation to the offender’s current offense, criminal history and placement on the proposed sentencing guidelines grid. 18 Comprehensive Recidivism Study RECIDIVISM RATES The definition of a recidivist used for this analysis was: For offenders released from incarceration: • • an offender with a new arrest (arraignment) in the year following release from a correctional facility; or, an offender with a technical violation of probation or parole that resulted in incarceration in the year following release from a correctional facility; For offenders in community corrections centers or drug court programs: • • an offender with a new arrest (arraignment) in the year following the date of entry into a community corrections center or drug court program; or, an offender with a technical violation of probation or parole that resulted in incarceration in the year following the date of entry into a community corrections center or drug court program. In this section, the overall recidivism rates are discussed along with a discussion of the timing of the recidivist behavior and a description of that behavior. Type of Recidivist The overall recidivism rate was estimated to be 49.1% within one year of release or program participation. Of the 3,751 offenders in the sample, 1,841 or 49.1% met the definition of a recidivist. Table 3 shows the manner in which offenders were classified as recidivists. The most common reason for classifying an offender as a recidivist was one or more new arrests / arraignments during the one year follow-up period. Of the 3,751 offenders in the sample, an estimated 1,668 or 44.5% had one or more new arrests during the follow-up period. The proportion of offenders with one or more new arrests was lower for the community based programs: 38.2% of the drug court participants had one or more new arrests and 39.3% of the community corrections center participants had one or more new arrests. Of the incarcerated offenders, the proportion of offenders with one or more new arrests was lower among those released via parole than those offenders who were discharged. Of the offenders released after serving a house of correction sentence, 49.8% of those discharged had one or more new arrests compared to 38.5% of those released by parole; and, of the 19 Massachusetts Sentencing Commission offenders released after serving a state prison sentence, 36.3% of those discharged had one or more new arrests compared to 23.6% of those released by parole. Other offenders had no new arrests but were classified as a recidivist solely on the basis of a technical violation of parole or probation. Of all offenders in the sample, 173 or 4.6% were classified as recidivists based on a technical violation. Because all offenders released by parole and all community corrections center and drug court program participants were under supervision for at least part of the follow-up period, they were expected to be subject to a higher rate of violation activity. As indicated in Table 3, the proportion of offenders classified as recidivists based on technical violations was higher for the community based programs: 11.8% of the drug court participants were classified as a recidivist based on a technical violation and 12.7% of the community corrections center participants were classified as a recidivist based on a technical violation. Of the incarcerated offenders, the proportion of offenders classified as recidivists based on a technical violation was higher among those released via parole than those offenders who were discharged. Of the offenders released after serving a house of correction sentence, 1.7% of those discharged were technical violators compared to 8.7% of those released by parole; and, of the offenders released after serving a state prison sentence, 0.0% of those discharged were technical violators compared to 18.8% of those released by parole. The nature of the new charges for which the offender was arraigned during the follow-up period was also considered. The first characteristic of the new offense behavior that was considered was whether the new charge was a felony or misdemeanor. In Massachusetts, a felony is any offense for which the offender can be sentenced to state prison and misdemeanors are all other offenses. Table 3 shows the felony / misdemeanor breakdown associated with the most serious new offense from among all of the charges for which an offender was arrested / arraigned during the one year follow-up period. Of the 3,751 offenders in the sample, for 784 or 20.9% the most serious new offense was a felony and for 884 or 23.6% the most serious new offense was a misdemeanor. The proportion of offenders whose most serious new arrest offense was classified as a felony varied by criminal justice cohort. For example, of the community corrections center participants, 46 or 14.2% had a felony as the most serious new offense. 20 Comprehensive Recidivism Study Table 3. Recidivism Classification by Criminal Justice Cohort Sample Total Drug Courts House of Correction Sentence Community Corrections Centers Discharge Parole State Prison Sentence Sample Total Discharge Parole Sample Total % Type of Recidivist N % N % N % N % N % N % N % N % N Recidivism Rate 1,841 49.1% 38 50.0% 168 52.0% 1,156 51.5% 288 47.1% 1,444 50.6% 121 36.3% 70 42.4% 191 38.4% One or More Arrests 1,668 44.5% 29 38.2% 127 39.3% 1,117 49.8% 235 38.5% 1,352 47.4% 121 36.3% 39 23.6% 160 32.1% felony 784 20.9% 15 19.7% 46 14.2% 530 23.6% 104 17.0% 634 22.2% 67 20.1% 22 13.3% 89 17.9% misdemeanor 884 23.6% 14 18.4% 81 25.1% 587 26.2% 131 21.4% 718 25.2% 54 16.2% 17 10.3% 71 14.3% 173 4.6% 9 11.8% 41 12.7% 39 1.7% 53 8.7% 92 3.2% 0 0.0% 31 18.8% 31 6.2% Non-Recidivist 1,910 50.9% 38 50.0% 155 48.0% 1,087 48.5% 323 52.9% 1,410 49.4% 212 63.7% 95 57.6% 307 61.6% Total 3,751 100.0% 76 100.0% 323 100.0% 2,243 100.0% 611 100.0% 2,854 100.0% 333 100.0% 165 100.0% 498 100.0% Most serious: No Arrests / Technical 21 Massachusetts Sentencing Commission Time Until Recidivism The behavior that was associated with the offender being classified as a recidivist could occur at any point during the one year follow-up period. Offenders were considered as recidivists if there was any arrest in the one year follow-up period regardless of the parole or program status at the time of the new arrest. For many offenders released to parole, especially those released after serving a house of correction sentence, the period of parole supervision terminated before the end of the one year follow-up period. Similarly, for offenders in community corrections center programs, where program cycles typically run for 12 weeks, program participation was expected to be completed before the end of the one year follow-up period. For this reason, Table 4 shows the relationship between new arrests, parole supervision status, and participation status for the parole and community corrections center cohorts. Of those offenders paroled with a house of correction sentence, 235 (38.5%) had one or more new arrests. However, the new arrest occurred after the end of the parole supervision for 160 offenders. Of those paroled with a state prison sentence, 39 (23.6%) had one or more new arrests and five occurred after the end of the period of parole supervision. Of the 127 offenders participating in a community corrections center program with one or more new arrests, 27 occurred prior to program completion / termination, 31 occurred after program completion / termination, and the completion / termination date was not available for 69 offenders. Table 5 shows the time until the recidivism behavior by criminal justice cohort. The recidivism rates that are shown in Table 5 can be interpreted as interim or cumulative recidivism rates. This data demonstrates the relationship between the length of the follow-up period and estimated recidivism rates; that is, longer follow-up periods will be associated with higher recidivism rates. Comparisons between rates might also change over time. For example, when comparing the offenders in the community corrections center programs with the offenders released from houses of correction, the recidivism rate for the community corrections center programs was higher initially but the difference between the rates declined over the follow-up period. 22 Comprehensive Recidivism Study Table 4. New Arrest Status, Parole Supervision Status, and Community Corrections Center Participation Status Community Corrections Centers Time of New Arrest House Parole Sample State Parole Sample N % N % N % No New Arrests 196 60.7% 376 61.5% 126 76.4% New Arrest - before end of parole / program 27 8.4% 75 12.3% 34 20.6% New Arrest - after end of parole / program 31 9.6% 160 26.2% 5 3.0% New Arrest - completion date not provi ded 69 21.4% 0 0.0% 0 0.0% Sub-Total 127 39.3% 235 38.5% 39 23.6% Total 323 100.0% 611 100.0% 165 100.0% 23 Massachusetts Sentencing Commission Table 5. Time Until Recidivism Incident by Criminal Justice Cohort Non-Recidivists Cumulative Recidivism Rate New Recidivists Cumulative Recidivism Rate New Recidivists Cumulative Recidivism Rate New Recidivists Cumulative Recidivism Rate New Recidivists Cumulative Recidivism Rate New Recidivists Cumulative Recidivism Rate Sample Total (n = 498 ) New Recidivists Parole (n = 165 ) Cumulative Recidivism Rate Discharge (n = 333 ) New Recidivists Sample Total (n = 2854 ) Cumulative Recidivism Rate Parole (n = 611 ) New Recidivists Less than 1 month 1 to 2 months 2 to 3 months 3 to 4 months 4 to 5 months 5 to 6 months 6 to 7 months 7 to 8 months 8 to 9 months 9 to 10 months 10 to 11 months 11 to 12 months Discharge (n = 2243 ) Cumulative Recidivism Rate Time Until Recidivism Incident Drug Courts (n =76) State Prison Sentence House of Correction Sentence New Recidivists Sample Total (n =3751) Community Corrections Centers (n =323) 223 219 215 187 196 148 128 116 116 103 88 102 5.9% 11.8% 17.5% 22.5% 27.7% 31.7% 35.1% 38.2% 41.3% 44.0% 46.4% 49.1% 4 9 7 2 3 3 2 2 3 1 0 2 5.3% 17.1% 26.3% 28.9% 32.9% 36.8% 39.5% 42.1% 46.1% 47.4% 47.4% 50.0% 27 23 19 21 14 15 10 7 11 9 8 4 8.4% 15.5% 21.4% 27.9% 32.2% 36.8% 39.9% 42.1% 45.5% 48.3% 50.8% 52.0% 147 133 138 113 131 98 80 69 70 67 43 67 6.6% 12.5% 18.6% 23.7% 29.5% 33.9% 37.4% 40.5% 43.6% 46.6% 48.6% 51.5% 31 38 26 29 25 19 19 24 21 14 24 18 5.1% 11.3% 15.5% 20.3% 24.4% 27.5% 30.6% 34.5% 38.0% 40.3% 44.2% 47.1% 178 171 164 142 156 117 99 93 91 81 67 85 6.2% 12.2% 18.0% 23.0% 28.4% 32.5% 36.0% 39.2% 42.4% 45.3% 47.6% 50.6% 8 11 17 12 15 9 11 11 7 7 7 6 2.4% 5.7% 10.8% 14.4% 18.9% 21.6% 24.9% 28.2% 30.3% 32.4% 34.5% 36.3% 6 5 9 10 7 4 6 3 4 5 7 4 3.6% 6.7% 12.1% 18.2% 22.4% 24.8% 28.5% 30.3% 32.7% 35.8% 40.0% 42.4% 14 16 26 22 22 13 17 14 11 12 14 10 2.8% 6.0% 11.2% 15.7% 20.1% 22.7% 26.1% 28.9% 31.1% 33.5% 36.3% 38.4% 1,910 38 155 1,087 323 24 1,410 212 95 307 Comprehensive Recidivism Study New Arrest Offenses Further characteristics of the new charges that occurred during the follow-up period were also considered. In this section the offense seriousness level and the type of offense are studied. The offense seriousness level of all new arrest offenses was considered according to the ranking of offenses proposed by the Sentencing Commission. Table 6 shows the level of the most seriousness new arrest offense by criminal justice cohort. For all offenders with one or more new arrests, the most serious new arrests were classified at level 4 through 9 for 21.0% of the offenders and at level 1 through 3 for 78.8% of the offenders. The seriousness of the new arrests varied by criminal justice cohort. For example, of the offenders with one or more new arrests, only 10.3% of the drug court program participants with one or more new arrests had a new arrest at offense seriousness level 4 or above. In contrast, of the offenders released after serving a state prison sentence with one or more new arrests, 30.6% had a new arrest at offense seriousness level 4 or above. For those offenders with one or more new arrests, the most serious offense was classified by type of offense. As shown in Table 7, for all offenders with one or more new arrests, 462 or 27.7% were charged with an offense against the person or sex offense as the most serious new arrest offense and 1,206 or 72.3% were charged with some other offense (property, drug, motor vehicle, etc.) as the most serious new offense. 25 Massachusetts Sentencing Commission Table 6. Offense Seriousness Level of Most Serious New Arrest Offense by Criminal Justice Cohort Sample Total Drug Courts House of Correction Sentence Community Corrections Centers Discharge Parole State Prison Sentence Sample Total Discharge Parole Sample Total Level of Most Serious Offense N % N % N % N % N % N % N % N % N 9 1 0.1% 0 0.0% 0 0.0% 1 0.1% 0 0.0% 1 0.1% 0 0.0% 0 0.0% 0 0.0% 8 6 0.4% 0 0.0% 0 0.0% 4 0.4% 0 0.0% 4 0.3% 1 0.8% 1 2.6% 2 1.3% 7 7 0.4% 0 0.0% 0 0.0% 6 0.5% 0 0.0% 6 0.4% 0 0.0% 1 2.6% 1 0.6% 6 31 1.9% 0 0.0% 4 3.1% 23 2.1% 1 0.4% 24 1.8% 2 1.7% 1 2.6% 3 1.9% % 5 73 4.4% 0 0.0% 4 3.1% 47 4.2% 13 5.5% 60 4.4% 8 6.6% 1 2.6% 9 5.6% 4 232 13.9% 3 10.3% 15 11.8% 163 14.6% 17 7.2% 180 13.3% 26 21.5% 8 20.5% 34 21.3% 3 652 39.1% 11 37.9% 47 37.0% 449 40.2% 90 38.3% 539 39.9% 41 33.9% 14 35.9% 55 34.4% 2 407 24.4% 10 34.5% 38 29.9% 257 23.0% 67 28.5% 324 24.0% 27 22.3% 8 20.5% 35 21.9% 1 256 15.3% 5 17.2% 19 15.0% 166 14.9% 45 19.1% 211 15.6% 16 13.2% 5 12.8% 21 13.1% Not Assigned 3 0.2% 0 0.0% 0 0.0% 1 0.1% 2 0.9% 3 0.2% 0 0.0% 0 0.0% 0 0.0% Total 1,668 100.0% 29 100.0% 127 100.0% 1,117 100.0% 235 100.0% 1,352 100.0% 121 100.0% 39 100.0% 160 100.0% % Level 4+ 21.0% 10.3% 18.1% 21.8% 13.2% 20.3% 30.6% 30.8% 30.6% % Level 1-3 78.8% 89.7% 81.9% 78.1% 86.0% 79.4% 69.4% 69.2% 69.4% 26 Comprehensive Recidivism Study Table 7. Type of Offense of Most Serious New Arrest Offense by Criminal Justice Cohort Sample Total Type of Offense N % Person 439 26.3% Sex 23 1.4% Drug Courts N Community Corrections Centers State Prison Sentence House of Correction Sentence Discharge Parole Sample Total Discharge Parole Sample Total % N % N % N % N % N % N % N % 2 6.9% 31 24.4% 326 29.2% 43 18.3% 369 27.3% 29 24.0% 8 20.5% 37 23.1% 0 0.0% 1 0.8% 17 1.5% 3 1.3% 20 1.5% 1 0.8% 1 2.6% 2 1.3% Property 426 25.5% 15 51.7% 26 20.5% 267 23.9% 77 32.8% 344 25.4% 30 24.8% 11 28.2% 41 25.6% Drug 313 18.8% 5 17.2% 26 20.5% 206 18.4% 40 17.0% 246 18.2% 27 22.3% 9 23.1% 36 22.5% Motor Vehicle 294 17.6% 7 24.1% 29 22.8% 183 16.4% 44 18.7% 227 16.8% 23 19.0% 8 20.5% 31 19.4% Other 173 10.4% 0 0.0% 14 11.0% 118 10.6% 28 11.9% 146 10.8% 11 9.1% 2 5.1% 13 8.1% Total 1668 100.0% 29 100.0% 127 100.0% 1117 100.0% 235 100.0% 1352 100.0% 121 100.0% 39 100.0% 160 100.0% 27 Massachusetts Sentencing Commission PROGRAMS, SECURITY, AND COMMUNITY RELEASE Exposure to the criminal justice programs varied for the offenders in the study sample. The community based program samples included offenders who dropped out of the program after a brief stay as well as offenders who successfully completed the program after many weeks or months of participation. The correctional release samples included offenders incarcerated for varying lengths of time in a wide range of security settings. The relationship between recidivism rates and the nature of the criminal justice placement are considered in this section. For those offenders released from the DOC or houses of correction, the security level at the time of release was considered. For community corrections centers, the intermediate sanction level, program completion status, and length of time in the program were considered. For drug courts, the completion status and length of time in the program were considered. Correctional Facilities For many offenders released from correctional facilities, the security level of the offender at the time of release was available. All security levels were defined by the contributing criminal justice agency and may not be comparable across the various criminal justice cohorts. As indicated in Table 8, the recidivism rates of those offenders released from lower security facilities were less than the recidivism rates of those offenders released from higher security facilities. For all offenders, it was estimated that the recidivism rate for releases from lower security facilities was 39.9% and the recidivism rate for releases from higher security facilities was 53.1%. This pattern held for each of the correctional cohorts. 28 Comprehensive Recidivism Study Table 8. Recidivism Status by Security Level at Time of Release and Criminal Justice Cohort, Correctional Releases House of Correction Sentence State Prison Sentence Sample Total Total Recidivists Recidivism Rate Total Recidivists Recidivism Rate 129 Recidivism Rate 50.0% Recidivists 1 Sample Total Total 2 Parole Recidivism Rate 48.0% Discharge Recidivists Total 61 Recidivism Rate 127 Recidivists 45.5% Sample Total Total Recidivists 80 Parole Recidivism Rate Total 176 Recidivism Rate Maximum Recidivists Security Level Total Discharge 62 48.1% 46 18 39.1% 1 0 0.0% 47 18 38.3% Medium 1046 569 54.4% 707 412 58.3% 97 55 56.7% 804 467 58.1% 180 69 38.3% 62 33 53.2% 242 102 42.1% Medium/Minimum 82 44 53.7% 67 35 52.2% 15 9 60.0% 82 44 53.7% 0 0 N.A. 0 0 N.A. 0 0 N.A. Sub-Total Secure 1304 693 53.1% 901 508 56.4% 114 65 57.0% 1015 573 56.5% 226 87 38.5% 63 33 52.4% 289 120 41.5% Minimum 475 194 40.8% 297 125 42.1% 90 41 45.6% 166 42.9% 57 20 35.1% 31 8 25.8% 88 28 31.8% 387 Minimum/Pre-release 113 41 36.3% 32 11 34.4% 13 6 46.2% 45 17 37.8% 37 11 29.7% 31 13 41.9% 68 24 35.3% Pre-release 176 70 39.8% 76 34 44.7% 53 21 39.6% 129 55 42.6% 13 3 23.1% 34 12 35.3% 47 15 31.9% Sub-Total Lower 764 305 39.9% 405 170 42.0% 156 68 43.6% 561 238 42.4% 107 34 31.8% 96 33 34.4% 203 67 33.0% Not Available 1284 637 49.6% 937 478 51.0% 341 155 45.5% 1278 633 49.5% 0 0 N.A. 6 4 66.7% 6 4 66.7% 3352 1635 48.8% 2243 1156 51.5% 611 288 47.1% 2854 1444 50.6% 333 121 36.3% 165 70 42.4% 498 191 38.4% Total 29 Massachusetts Sentencing Commission Community Corrections Centers Table 9 shows recidivism status by selected program characteristics of community corrections center participants. Because all of these data elements were not required, the proportion of unknowns varies extensively. However, the data does point to some interesting preliminary trends. For example, the recidivism rate for program completers (39.5%) was lower than the recidivism rate for those who terminated the program (68.3%) - some of whom were terminated due to a new arrest or other violation - and the recidivism rate was lower for those who were in the program longer. Further research should place more emphasis on collecting information on the program level and the program completion / termination information for this population. Drug Courts Table 10 shows recidivism status by selected program characteristics of drug court participants, completion status and length of time in the program. Many of the drug court program participants were still in the program at the end of the one year follow-up period. Offenders in the “other” category included inappropriate referrals, medical problems, and probation transfers. As was the case with the community corrections centers cohort, the recidivism rate was much lower for those who completed the program (21.4%) as well as for those who were still in the program at the end of the one year follow-up period (16.7%). In light of these findings, further research should place more emphasis on the distinction between program completers and those offenders who terminated the drug court program. 30 Comprehensive Recidivism Study Table 9. Recidivism Status by Selected Program Characteristics, Community Corrections Center Participants Community Corrections Centers Program Characteristic Total Recidivists Recidivism Rate Beginning Level Level IV Level III Not Available Total 81 192 50 323 40 97 31 168 49.4% 50.5% 62.0% 52.0% Referral Source Probation Sheriff Parole Not Available Total 212 24 1 86 323 123 12 0 33 168 58.0% 50.0% 0.0% 38.4% 52.0% Completion Type Completion Termination Not Available Total 43 41 239 323 17 28 123 168 39.5% 68.3% 51.5% 52.0% Change in Level Level IV / no change Level IV / Level III Level III / no change Not Available Total 11 6 33 273 323 4 1 17 146 168 36.4% 16.7% 51.5% 53.5% 52.0% Time in Program 1 Month or Less 2 to 4 Months 5 Months or More Not Available Total 47 74 18 184 323 30 41 7 90 168 63.8% 55.4% 38.9% 48.9% 52.0% 31 Massachusetts Sentencing Commission Table 10. Recidivism Status by Selected Program Characteristics, Drug Court Participants Drug Courts Program Characteristic Total Recidivists Recidivism Rate Completion Type Completion 14 3 21.4% Still in Program at End of Year 18 3 16.7% Termination 20 16 80.0% Other Type 4 3 75.0% Not Available 20 13 65.0% Total 76 38 50.0% 4 2 50.0% Time in Program 1 Month or Less 2 to 4 Months 7 6 85.7% 5 Months or More 20 10 50.0% Still in Program at End of Year 16 2 12.5% Not Available 29 18 62.1% Total 76 38 50.0% 32 Comprehensive Recidivism Study DEMOGRAPHIC CHARACTERISTICS OF OFFENDERS The demographic characteristics of the offenders available for the analysis included: gender, age, and race / ethnicity. Gender Table 11 shows recidivism status by gender and criminal justice cohort. Overall, there were 446 (11.9%) female offenders and 3,305 (88.1%) male offenders in the sample. The proportion of female offenders varied across the various criminal justice cohorts: • • • • 30.3% of the drug court program participants were female; 12.7% of the offenders released after serving a house of correction sentence were female; 11.5% of the community corrections center program participants were female; and, 4.8% of the offenders released after serving a state prison sentence were female. Overall, the recidivism rate for female offenders was slightly lower than that of male offenders (47.3% and 49.3%, respectively). This pattern held for all of the criminal justice cohorts. Further research could consider some of the specialized community corrections centers for female offenders that have been developed by the Office of Community Corrections but were not in operation early enough to be included in this study. Age Age was calculated at the time of release from a correctional facility or the time the offender began the community based program. The age of offenders in the sample ranged from 17 to 73 years. The mean age was 32.0 years and the median age was 31 years. The age of offenders varied by criminal justice cohort. To the extent that the particular criminal justice agency services those younger “high-risk” offenders, it was expected that the corresponding recidivism rate would be higher. Those offenders in drug courts and those offenders released from the DOC with state prison sentences were the oldest on average, while those offenders participating in community corrections centers were the youngest on average: 33 Massachusetts Sentencing Commission • • • • 57.6% of the community corrections center program participants were 29 years or younger; 45.1% of offenders released after serving a house of correction sentence were 29 years or younger; 34.1% of offenders released after serving a state prison sentence were 29 years or younger; and, 23.7% of the drug court program participants were 29 years or younger. Table 12 shows recidivism rates by age. Younger offenders were more likely to be classified as recidivists than older offenders. The recidivism rate for offenders under 20 years of age was 61.5% while the recidivism rate for offenders age 50 to 59 was 25.5%. For all offenders age 29 or younger, the recidivism rate was 55.2% and for all offenders age 30 and older, the recidivism rate was 44.2%. This pattern held for all of the criminal justice cohorts with the exception of the drug court program participants. Race / Ethnicity Table 13 shows recidivism rates by race / ethnicity. Overall, the sample was 53.3% white and 46.7% racial / ethnic minorities. Race / ethnicity was not available for 1.1% of the sample. The racial / ethnic composition of the population varied by criminal justice cohort: • • • • 62.7% of the offenders released after serving a state prison sentence were racial / ethnic minorities; 44.8% of the offenders released after serving a house of correction sentence were racial / ethnic minorities; 41.0% of the community corrections center program participants were racial / ethnic minorities; and, 35.6% of the drug court program participants were racial / ethnic minorities. Recidivism rates were lower for whites than for racial / ethnic minorities across each of the cohorts. 34 Comprehensive Recidivism Study Table 11. Recidivism Status by Gender and Criminal Justice Cohort House of Correction Sentence Recidivists Recidivism Rate Total Recidivists Recidivism Rate Total Recidivists Recidivism Rate 117 36.7% 155 66 42.6% 474 183 38.6% 47.1% 2854 1444 50.6% 333 121 36.3% 165 70 42.4% 498 191 38.4% 23 11 47.8% 37 15 40.5% 259 125 48.3% 103 52 50.5% 362 Ma le 3,305 1,630 49.3% 53 27 50.9% 286 153 53.5% 1,984 1031 52.0% 508 236 Total 3,751 1,841 49.1% 76 38 50.0% 323 168 52.0% 2,243 1156 51.5% 611 288 % Fe male 11.9% Total Total 319 Recidivism Rate 1267 50.8% Recidivists 46.5% 2,492 Total 33.3% Recidivism Rate 8 Recidivists 24 Total 40.0% Recidivism Rate 4 Recidivists 10 Total 28.6% Recidivism Rate 4 Recidivists 14 Total 48.9% 47.3% Fema le 446 30.3% 11.5% 11.5% 16.9% 35 Sample Total Paro le 177 211 Gender Discharge Total Sample Total Recidivism Rate Paro le Recidivism Rate Discharge Recidivists Community Corrections Centers Drug Co urts Recidivists Sample Total State Prison Sentence 12.7% 4.2% 6.1% 4.8% Massachusetts Sentencing Commission Table 12. Recidivism Status by Age and Criminal Justice Cohort 68 40 58.8% 223 132 59.2% 3 2 66.7% 2 1 50.0% 5 3 60.0% 59.2% 132 67 50.8% 558 319 57.2% 35 18 51.4% 22 10 45.5% 57 28 49.1% Total R ecidivists 59.4% 252 Total 92 426 Total R ecidivism R ate 155 51.8% R ecidivists 70.2% 44 Total 40 85 R ecidivists 57 22.2% Total 100.0% R ecidivists 1 2 R ecidivists 1 9 Total 61.5% 55.4% R ecidivists 176 R ecidivists R ecidivism R ate Sample Total 393 Total R ecidivism R ate Paro le 286 Total R ecidivism R ate Discharge 709 R ecidivists R ecidivism R ate State Prison Sentence Sample Total 20 to 24 Total R ecidivism R ate Paro le R ecidivism R ate Discharge R ecidivism R ate House of Correction Sentence Commu nity Corrections C enters R ecidivists Drug Co urts Under 20 Age at Release or Program R ecidivism R ate Sample Total 25 to 29 665 348 52.3% 8 4 50.0% 44 28 63.6% 385 214 55.6% 120 58 48.3% 505 272 53.9% 71 27 38.0% 37 17 45.9% 108 44 40.7% 30 to 34 634 327 51.6% 23 13 56.5% 49 26 53.1% 370 200 54.1% 91 45 49.5% 461 245 53.1% 65 23 35.4% 36 20 55.6% 101 43 42.6% 35 to 39 640 308 48.1% 10 8 80.0% 36 17 47.2% 409 202 49.4% 97 50 51.5% 506 252 49.8% 64 23 35.9% 24 8 33.3% 88 31 35.2% 40 to 49 645 248 38.4% 22 9 40.9% 40 10 25.0% 399 169 42.4% 82 24 29.3% 481 193 40.1% 71 24 33.8% 31 12 38.7% 102 36 35.3% 50 to 59 145 37 25.5% 3 1 33.3% 10 3 30.0% 85 24 28.2% 16 3 18.8% 101 27 26.7% 20 4 20.0% 11 2 18.2% 31 6 19.4% 60 to 69 23 4 17.4% 0 0 N.A . 1 0 0.0% 13 3 23.1% 5 1 20.0% 18 4 22.2% 2 0 0.0% 2 0 0.0% 4 0 0.0% 70 to 79 4 0 0.0% 0 0 N.A . 1 0 0.0% 1 0 0.0% 0 0 N.A. 1 0 0.0% 2 0 0.0% 0 0 N.A. 2 0 0.0% Total 3751 1841 49.1% 76 38 50.0% 323 168 52.0% 2243 1156 51.5% 611 288 47.1% 2854 1444 50.6% 333 121 36.3% 165 70 42.4% 498 191 38.4% 29 and Under 1660 917 55.2% 18 7 38.9% 186 112 60.2% 966 558 57.8% 320 165 51.6% 1286 723 56.2% 109 47 43.1% 61 28 45.9% 170 75 44.1% 30 and Older 2091 924 44.2% 58 31 53.4% 137 56 40.9% 1277 598 46.8% 291 123 42.3% 1568 721 46.0% 224 74 33.0% 104 42 40.4% 328 116 35.4% ` % 29 and Younger Mean Age Median Age 44.3% 32.0 Ye ars 31 Yea rs 23.7% 35.0 Ye ars 33 Yea rs 57.6% 29.1 Ye ars 28 Yea rs 43.1% 32.2 Ye ars 32 Yea rs 52.4% 30.4 Ye ars 29 Yea rs 36 ` 45.1% 31.8 Ye ars 31 Yea rs 32.7% 35.0 Ye ars 34 Yea rs 37.0% 34.0 Ye ars 32 Yea rs 34.1% 34.6 Ye ars 33 Yea rs Comprehensive Recidivism Study Table 13. Recidivism Status by Race / Ethnicity and Criminal Justice Cohort House of Correction Sentence Community Corrections Centers Total R ecidivists R ecidivism R ate Total R ecidivists R ecidivists R ecidivism R ate Total R ecidivists R ecidivism R ate Total R ecidivists R ecidivism R ate Total R ecidivists R ecidivism R ate 1,976 879 44.5% 47 22 46.8% 190 78 41.1% 1,139 532 46.7% 414 186 44.9% 1,553 718 46.2% 122 36 29.5% 64 25 39.1% 186 61 32.8% Black 885 488 55.1% 17 11 64.7% 47 27 57.4% 565 330 58.4% 91 45 49.5% 656 375 57.2% 113 49 43.4% 52 26 50.0% 165 75 45.5% Hispa nic 776 431 55.5% 8 4 50.0% 83 61 73.5% 455 264 58.0% 94 53 56.4% 549 317 57.7% 93 32 34.4% 43 17 39.5% 136 49 36.0% Other 71 29 40.8% 1 0 0.0% 2 2 100.0% 45 17 37.8% 12 4 33.3% 57 21 36.8% 5 4 80.0% 6 2 33.3% 11 6 54.5% U n k n ow n 43 14 32.6% 3 1 33.3% 1 0 0.0% 39 13 33.3% 0 0 N.A. 39 13 33.3% 0 0 N.A. 0 0 N.A. 0 0 N.A. Total 3,751 1,841 49.1% 76 38 50.0% 323 168 52.0% 611 288 333 121 36.3% 165 70 42.4% 498 191 38.4% % W hite 53.3% 64.4% 59.0% 51.7% 67.8% 55.2% 36.6% 38.8% 37.3% % M inority 46.7% 35.6% 41.0% 48.3% 32.2% 44.8% 63.4% 61.2% 62.7% % Missing 1.1% 3.9% 0.3% 1.7% 0.0% 1.4% 0.0% 0.0% 0.0% 2,243 1,156 51.5% 37 Total R ecidivism R ate White Race / Ethnicity R ecidivism R ate R ecidivists Sample Total Total Paro le R ecidivism R ate Discharge R ecidivists Sample Total Total Paro le R ecidivism R ate Discharge State Prison Sentence R ecidivists Drug Co urts Total Sample Total 47.1% 2854 1444 50.6% Massachusetts Sentencing Commission CURRENT OFFENSE AND CRIMINAL HISTORY In this section the current offense and criminal history of the offender are discussed in relation to observed recidivism rates. Both of these variables, current offense and criminal history, were measured at the time of the current placement. For correctional populations, current offense refers to that offense for which the offender had been convicted and on which the offender was now being released. For the community-based programs, current offense refers to that offense for which the offender was being supervised at the time of the placement. For all populations, criminal history refers to criminal activity that occurred prior to the current offense, excluding the current offense, and excluding any new arrest behavior. Current Offense Type of Offense Current offense was available for 2,675 (71.3%) of the study sample. The sample included 732 (27.4%) person offenders, 77 (2.9%) sex offenders, 737 (27.6%) drug offenders, 586 (21.9%) property offenders, 373 (13.9%) motor vehicle offenders, and 170 (6.4%) other (public order, prostitution, and weapons) offenders. The nature of current offense varied by criminal justice cohort: • • • • • 42.4% of the offenders released after serving a state prison sentence were person offenders; 10.4% of the offenders released after serving a state prison sentence were sex offenders; 26.8% of the community corrections center program participants were property offenders; 52.6% of the drug court program participants were drug offenders; and, 17.5% of the offenders released after serving a house of correction sentence were motor vehicle offenders. Table 14 shows the current offense and recidivism status by criminal justice cohort. Of the major offense categories, recidivism rates were lowest for sex offenders (20.8%) and highest for property offenders (56.5%). These patterns were generally consistent across criminal justice cohorts. 38 Comprehensive Recidivism Study Offense Seriousness Level Table 15 shows current offense classified according to the offense seriousness level proposed by the commission, recidivism status and criminal justice cohort. There was an inverse relationship between offense seriousness level and recidivism rates. In general, the more serious the current offense, the lower the recidivism rate of the offender. The recidivism rate for offenders with a current offense level of 4 or above was 40.2% and the recidivism rate for offenders with a current offense level of 1 through 3 was 51.7%. 39 Massachusetts Sentencing Commission Table 14. Recidivism Status by Type of Offense (Current Offense) and Criminal Justice Cohort R ecidivists R ecidivism R ate Total R ecidivists R ecidivism R ate Total R ecidivists R ecidivism R ate Total R ecidivists R ecidivism R ate 45.8% Total 11 Sample Total R ecidivism R ate 24 Paro le R ecidivists R ecidivism R ate 81.8% Discharge Total R ecidivists 9 State Prison Sentence Sample Total R ecidivism R ate Total 11 Paro le R ecidivists R ecidivism R ate 50.0% Discharge Total R ecidivists 366 Total 732 House of Correction Sentence Commu nity Corrections C enters Drug Co urts R ecidivism R ate Person R ecidivists Governing Offense Total Sample Total 373 204 54.7% 114 58 50.9% 487 262 53.8% 139 54 38.8% 71 30 42.3% 210 84 40.0% Sex 77 16 20.8% 0 0 N.A. 0 0 N.A. 22 4 18.2% 3 2 66.7% 25 6 24.0% 45 9 20.0% 7 1 14.3% 52 10 19.2% Property 586 331 56.5% 15 7 46.7% 34 23 67.6% 287 160 55.7% 159 91 57.2% 446 251 56.3% 62 32 51.6% 29 18 62.1% 91 50 54.9% Drug 737 344 46.7% 40 14 35.0% 50 27 54.0% 333 180 54.1% 196 90 45.9% 529 270 51.0% 66 17 25.8% 52 16 30.8% 118 33 28.0% Mo tor V ehicle 373 137 36.7% 4 3 75.0% 14 7 50.0% 236 93 39.4% 110 30 27.3% 346 123 35.5% 8 3 37.5% 1 1 100.0% 9 4 44.4% Other 170 99 58.2% 6 5 83.3% 5 4 80.0% 112 63 56.3% 29 17 58.6% 141 80 56.7% 13 6 46.2% 5 4 80.0% 18 10 55.6% Not A vailable 1076 548 50.9% 0 0 N.A. 196 96 49.0% 880 452 51.4% 0 0 N.A. 880 452 51.4% 0 0 N.A. 0 0 N.A. 0 0 N.A. Total 3751 1841 49.1% 76 38 50.0% 323 168 52.0% 2243 611 288 333 121 36.3% 165 70 42.4% 498 191 38.4% % Person % Sex 27.4% 14.5% 18.9% 1156 51.5% 27.4% 18.7% 47.1% 2854 1444 50.6% 24.7% 41.7% 43.0% 42.2% 2.9% 0.0% 0.0% 1.6% 0.5% 1.3% 13.5% 4.2% 10.4% % Prop erty 21.9% 19.7% 26.8% 21.1% 26.0% 22.6% 18.6% 17.6% 18.3% % Drug 27.6% 52.6% 39.4% 24.4% 32.1% 26.8% 19.8% 31.5% 23.7% % M Vehic le 13.9% 5.3% 11.0% 17.3% 18.0% 17.5% 2.4% 0.6% 1.8% 6.4% 7.9% 3.9% 8.2% 4.7% 7.1% 3.9% 3.0% 3.6% % Other 40 Comprehensive Recidivism Study Table 15. Recidivism Status by Offense Seriousness Level (Current Offense) and Criminal Justice Cohort House of Correction Sentence Total R ecidivists R ecidivism R ate Total R ecidivists R ecidivism R ate Total R ecidivists R ecidivism R ate Total R ecidivists R ecidivism R ate 26.3% R ecidivism R ate 15.4% 10 R ecidivists 4 38 Total 26 R ecidivism R ate 0.0% R ecidivists 0 Total 4 R ecidivism R ate R ecidivism R ate Sample Total R ecidivists Paro le R ecidivists Discharge Total Sample Total R ecidivism R ate Paro le R ecidivists Discharge State Prison Sentence Total Commu nity Corrections C enters Drug Co urts Total Sample Total 4 0 0.0% 0 0 N.A. 0 0 N.A. 0 0 N.A. 0 0 N.A. 0 0 N.A. 0 0 N.A. 4 0 0.0% 26 4 15.4% 0 0 N.A. 0 0 N.A. 0 0 N.A. 0 0 N.A. 0 0 N.A. 20 3 15.0% 6 1 16.7% 38 10 26.3% 0 0 N.A. 0 0 N.A. 0 0 N.A. 0 0 N.A. 0 0 N.A. 30 6 20.0% 8 4 50.0% 6 137 50 36.5% 0 0 N.A. 3 0 0.0% 0 0 N.A. 2 0 0.0% 2 0 0.0% 88 34 38.6% 44 16 36.4% 132 50 37.9% 5 116 49 42.2% 0 0 N.A. 4 3 75.0% 23 10 43.5% 3 2 66.7% 26 12 46.2% 61 22 36.1% 25 12 48.0% 86 34 39.5% 4 447 196 43.8% 10 3 30.0% 17 10 58.8% 187 91 48.7% 143 60 42.0% 330 151 45.8% 50 18 36.0% 40 14 35.0% 90 32 35.6% 3 1025 537 52.4% 28 17 60.7% 46 29 63.0% 579 303 52.3% 255 129 50.6% 834 432 51.8% 79 36 45.6% 38 23 60.5% 117 59 50.4% 2 644 316 49.1% 37 18 48.6% 46 23 50.0% 405 202 49.9% 155 72 46.5% 560 274 48.9% 1 1 100.0% 0 0 N.A. 1 1 100.0% Governing Offense Level 9 8 7 1 221 124 56.1% 1 0 0.0% 9 5 55.6% 161 95 59.0% 50 24 48.0% 211 119 56.4% 0 0 N.A. 0 0 N.A. 0 0 N.A. Not Assigned 17 7 41.2% 0 0 N.A. 2 2 100.0% 8 3 37.5% 3 1 33.3% 11 4 36.4% 4 1 25.0% 0 0 N.A. 4 1 25.0% Not A vailable 1076 548 50.9% 0 0 N.A. 196 96 49.0% 880 452 51.4% 0 0 N.A. 880 452 51.4% 0 0 N.A. 0 0 N.A. 0 0 N.A. Total 3751 1841 49.1% 76 38 50.0% 323 168 52.0% 2243 1156 51.5% 611 288 47.1% 2854 1444 50.6% 333 121 36.3% 165 70 42.4% 498 191 38.4% 41 Massachusetts Sentencing Commission Criminal History Offenders were assigned to a criminal history group consistent with the definitions proposed by the commission. The criminal history groups are defined in the Appendix. The relationship between juvenile criminal history and recidivism rates was also considered. Criminal History Groups Based on the data extraction and linkage procedures discussed in the methodology section, offenders were assigned to a criminal history group as proposed by the Commission. This assignment was made for 2,334 or 62.2% of the study sample. Of the offenders in the sample, 55.7% were classified in criminal history groups A or B and 44.3% were classified in criminal history groups C, D, or E. The proportion of offenders in the more serious criminal history groups varied by criminal justice cohort: • • • • 51.7% of the offenders released after serving a state prison sentence were classified in criminal history groups C, D, or E; 46.1% of the drug court program participants were classified in criminal history groups C, D, or E; 43.7% of the offenders released after serving a house of correction sentence were classified in criminal history groups C, D, or E; and 22.0% of the community corrections center program participants were classified in criminal history groups C, D, or E. Table 16 shows recidivism rates of offenders by criminal history group and criminal justice cohort. Recidivism rates varied consistently by criminal history group. Those offenders in the less serious criminal history groups had lower recidivism rates and those offenders in the more serious criminal history groups had higher recidivism rates. For example, offenders classified as criminal history group A, no / minor record, had a recidivism rate of 36.5%. Offenders assigned to criminal history group D, violent or repetitive record, had a recidivism rate of 60.8%. These patterns held for all of the criminal justice cohorts. 42 Comprehensive Recidivism Study Juvenile Criminal History Juvenile criminal history was available for all offenders in the sample. For purposes of this analysis all offenders were classified into two categories: (1) no prior juvenile adjudications; or, (2) one or more prior juvenile adjudications. Overall, 23.6% of the offenders had one or more prior juvenile adjudications. The proportion of offenders with one or more prior juvenile adjudications ranged from 17.1% of the drug court program participants to 26.1% of the offenders released after serving a state prison sentence. Those offenders with one or more prior juvenile adjudications were younger than those offenders with no prior juvenile adjudications, 25.7 years on average compared with 34.0 years. Table 17 shows recidivism rates by this juvenile criminal history indicator. The recidivism rate for offenders with one or more prior juvenile adjudications was 60.8%, much higher than the rate for offenders with no prior juvenile adjudications, 45.5%. This pattern held for all of the criminal justice cohorts, with the exception of the drug court program participants. 43 Massachusetts Sentencing Commission Table 16. Recidivism Status by Criminal History and Criminal Justice Cohort House of Correction Sentence Commu nity Corrections C enters State Prison Sentence R ecidivists R ecidivism R ate Total R ecidivists R ecidivism R ate Total R ecidivists R ecidivism R ate Total R ecidivists R ecidivism R ate Total R ecidivists R ecidivism R ate Total R ecidivists R ecidivism R ate 222 36.5% 16 5 31.3% 55 26 47.3% 337 131 38.9% 65 27 41.5% 402 158 39.3% 79 17 21.5% 57 16 28.1% 136 33 24.3% Group B Mode rate Reco rd 691 336 48.6% 25 14 56.0% 44 28 63.6% 404 200 49.5% 114 59 51.8% 518 259 50.0% 72 23 31.9% 32 12 37.5% 104 35 33.7% G r o up C Serious Re cord 526 271 51.5% 12 7 58.3% 20 14 70.0% 321 183 57.0% 67 30 44.8% 388 213 54.9% 75 22 29.3% 31 15 48.4% 106 37 34.9% G r o up D Violent or Repetitive 505 307 60.8% 23 12 52.2% 8 4 50.0% 298 188 63.1% 27 18 66.7% 325 206 63.4% 105 58 55.2% 44 27 61.4% 149 85 57.0% Group E Serious Violent 3 66.7% 0 0 N.A. 0 0 N.A. 1 1 100.0% 0 0 N.A. 1 1 100.0% 2 1 50.0% 0 0 N.A. 2 1 50.0% Not Available or Youthful Offender 1,417 703 49.6% 0 0 N.A. 196 96 49.0% 882 453 51.4% 338 154 45.6% 1,220 607 49.8% 0 0 N.A. 1 0 0.0% 1 0 0.0% Total 3,751 1,841 49.1% 76 38 50.0% 323 168 52.0% 2,243 1,156 51.5% 611 288 47.1% 2854 1444 50.6% 333 121 36.3% 165 70 42.4% 498 191 38.4% % C, D , or E 44.3% R ecidivists 609 Total Total Sample Total R ecidivism R ate Paro le R ecidivists Discharge Total Sample Total R ecidivism R ate Paro le R ecidivists Discharge Total Drug Co urts G roup A No/M inor Record Crimina l History Group 2 R ecidivism R ate Sample Total 46.1% 22.0% 45.6% 34.4% 44 43.7% 54.7% 45.7% 51.7% Comprehensive Recidivism Study Table 17. Recidivism Status by Juvenile Criminal History and Criminal Justice Cohort House of Correction Sentence State Prison Sentence Total R ecidivists R ecidivism R ate Total R ecidivists R ecidivism R ate Total R ecidivists R ecidivism R ate 477 207 43.4% 2,195 1,032 47.0% 237 74 31.2% 131 48 36.6% 368 122 33.2% 60.8% 13 5 38.5% 83 52 62.7% 331 63.0% 134 81 60.4% 96 47 49.0% 34 22 64.7% 130 69 53.1% Total 3,751 1,841 49.1% 76 38 50.0% 323 168 52.0% 2,243 1,156 51.5% 611 288 47.1% 2854 333 121 36.3% 165 70 42.4% 498 191 38.4% % One or More 23.6% 885 538 17.1% 25.7% 525 23.4% 21.9% 45 659 412 R ecidivism R ate 48.0% R ecidivists R ecidivists 825 Total Total 48.3% 1,718 One or M ore R ecidivism R ate R ecidivism R ate 116 2,866 1,303 45.5% Total 240 R ecidivists 52.4% Total R ecidivists Sample Total R ecidivists Paro le Total Discharge R ecidivism R ate Sample Total R ecidivists Paro le Total Discharge 33 None R ecidivism R ate Commu nity Corrections C enters Drug Co urts 63 Juven ile Adjudications R ecidivism R ate Sample Total 62.5% 1444 50.6% 23.1% 28.8% 20.6% 26.1% Massachusetts Sentencing Commission Sentencing Grid The current offense and criminal history were combined to place offenders on the sentencing guidelines grid proposed by the commission. Those offenders convicted of OUI offenses or mandatory firearms offenses were not assigned to the grid because the proposed sentencing guidelines grid would not apply to these offenders. Table 18 shows the number of offenders assigned to each grid cell, the number of recidivists in each grid cell, and the resulting recidivism rate. The recidivism rate was higher at the lower offense seriousness levels and the recidivism rate was higher for those offenders in the more serious criminal history groups. For any particular offense level, the recidivism rates are generally higher for those offenders in more serious criminal history groups and for any particular criminal history category, the recidivism rates are generally higher for those offenders in the less serious offense levels. The recidivism rate for OUI offenders was 24.3%, lower than the overall rate of 49.1%, suggesting that further analysis of the specialized programs for this population may be merited. 46 Comprehensive Recidivism Study Table 18. Recidivism Status by Sentencing Grid Cell and Grid Assignment, All Offenders ALL DEFENDANTS RECIDIVISM RATE RECIDIVISTS A B C D E N A B C D E N 9 4 0 0 0 0 4 9 0 0 0 0 0 0 8 8 6 10 2 0 26 8 2 0 1 1 0 7 17 7 7 7 0 38 7 6 0 3 1 6 48 27 30 29 0 134 6 11 11 10 5 29 29 21 32 0 111 5 5 11 4 110 97 83 54 0 344 4 34 3 174 231 181 239 1 826 3 2 116 167 114 74 0 471 1 35 69 55 37 0 N .A. 6 5 3 3 Total 547 638 504 477 Total A B C D E 9 0.0% N.A. N.A. N.A. N.A. 4 8 25.0% 0.0% 10.0% 50.0% N.A. 0 10 7 35.3% 0.0% 42.9% 14.3% N.A. 18 0 50 6 22.9% 40.7% 33.3% 62.1% N.A. 8 21 0 45 5 17.2% 37.9% 38.1% 65.6% N.A. 46 41 32 0 153 4 30.9% 47.4% 49.4% 59.3% N.A. 76 121 106 150 1 454 3 43.7% 52.4% 58.6% 62.8% 100.0% 2 57 94 62 46 0 259 2 49.1% 56.3% 54.4% 62.2% N.A. 196 1 17 37 32 24 0 110 1 48.6% 53.6% 58.2% 64.9% N.A. 0 17 N.A. 3 1 3 0 0 7 N.A. 50.0% 20.0% 100.0% 0.0% N.A. 1 2,167 Total 211 321 266 293 1 1,092 Total 38.6% 50.3% 52.8% 61.4% 100.0% Sentencing Zone Assig ned to Sente ncing G rid Incarceration Zone Discretionary Zone Intermediate Sanction Zone Not Assigned Sub-Total N 309 1,566 275 17 2,167 Sentencing Zone Assig ned to Sente ncing G rid Incarceration Zone Discretionary Zone Intermediate Sanction Zone Not Assigned Sub-Total Not A ssigne d to G rid OUI Offenses Mandatory Firearms Offenses Not Available or Youthful Offender 148 19 1417 Not A ssigne d to G rid OUI Offenses Mandatory Firearms Offenses Not Available or Youthful Offender Total 3,751 Total 47 N 125 817 143 7 1,092 36 10 703 1,841 Sentencing Zone Assig ned to Sente ncing G rid Incarceration Zone Discretionary Zone Intermediate Sanction Zone Not Assigned Sub-Total 0.0% 15.4% 26.3% 37.3% 40.5% 44.5% 55.0% 55.0% 56.1% 41.2% 50.4% Recidivism 40.5% 52.2% 52.0% 41.2% 50.4% Not A ssigne d to G rid OUI Offenses Mandatory Firearms Offenses Not Available or Youthful Offender 24.3% 52.6% 49.6% Total 49.1% Massachusetts Sentencing Commission SUMMARY A main purpose of this report has been to meet the legislative mandate set forth in Chapter 177 by compiling recidivism statistics from a variety of criminal justice and correctional entities using a consistent definition of recidivism. Among the highlights are: The project developed a reliable measure of the recidivism rates across a range of criminal justice cohorts. Based on the cooperation of the originating criminal justice agencies, a reliable measure of the recidivism rates was achieved as directed by Chapter 177 in the time frame provided by the legislature. Data was contributed from a variety of different criminal justice agencies and programs and integrated in a manner that yielded both reliable results and a basis for some comparisons of the populations and associated recidivism rates across criminal justice cohorts. This project demonstrated the utility and feasibility of conducting research across traditional criminal justice agency lines. This research project was the result of a collaborative effort among thirty different criminal justice agencies and programs. The project succeeded without any additional resources devoted to its completion. The ability of the various agencies to collect data independently and to contribute unique agency data to the development of a comprehensive research project was shown to be feasible. This model is one that could be applied to other research topics or to the expanded study of recidivism. It was important to consider the nature of the recidivism behavior in addition to the overall recidivism rates. The analysis demonstrated that it is important to consider the nature of the recidivism behavior in addition to comparing the overall recidivism rates. The analysis provided an examination of the nature of the recidivism behavior (new arrest or technical violation), the timing of the recidivism behavior, and the seriousness of that behavior. As suggested by the data collected for this study, offenders are subject to varying levels of supervision that can lead to technical violations. Even if the overall recidivism rates were the same, an argument can be made that there is a public safety benefit if the recidivism classification resulted 48 Comprehensive Recidivism Study from a technical violation rather than a new arrest. That is, a swift and certain response to a violation of the conditions of probation or parole may be viewed as a preventive step that contributes to public safety by intervening in the life of a supervised offender who may be in danger of slipping back into a pattern of criminal behavior. It was important to consider offender background information along with recidivism information. The study demonstrated both the strong relationship between certain offender characteristics and recidivism rates and the large differences in the composition of the offender population across criminal justice cohorts. In particular, the age of the offender and the criminal history of the offender were shown to be strongly related to the probability of recidivism across all of the criminal justice cohorts in the study. It is important to recognize that differences in the age and criminal history characteristics of the offender population in each of the criminal justice cohorts are related to observed recidivism rates. Additional data collection on program involvement and supervision variables should be encouraged in any future studies of recidivism. Future research should include program involvement data in order to shed more light on the relative effectiveness of different programs and on what types of programs are more (or less) effective with what types of offenders. Some of the offenders discharged from correctional facilities might have been under probation supervision during the follow-up period. Future research should also include the probation supervision status of the discharged offenders. Longer follow-up periods should be encouraged in any future studies of recidivism. The time frames available for completion of this project allowed for a one year follow-up. Longer follow-up periods may be interesting and useful for the comparisons to be made. For example, at the end of the one year follow-up period the recidivism rates for the community corrections centers and houses of correction were very close. However, the interim rates were different, suggesting that longer term recidivism rates may look different for these or other populations. 49 Massachusetts Sentencing Commission Replication of the model developed for the study is feasible and should be encouraged. The methodology developed in the process of completing this study could be used to provide a valuable basis to support on-going research and monitoring of recidivism rates. The method provided a foundation for routine updates of this initial effort. Although on-going studies are not required by the current legislative mandate, continued comprehensive recidivism studies should be encouraged. This model could be expanded to accommodate new criminal justice programs such as additional community corrections centers, drug courts, or additional houses of correction. Extensions to other probation supervised offender populations could be considered as well. The model should allow for longer term follow-up periods than were used in this study. Further, the model should be able to sustain routine follow-up studies on succeeding time periods. Much has been written in recent years about the importance of partnerships in criminal justice. Community policing and community corrections are example of recent initiatives that have emphasized cooperation and coordination among criminal justice agencies to achieve common goals. This research endeavor has underscored the value of such collaboration among criminal justice researchers to produce a comprehensive study that goes beyond the purview of each of the participating entities. In demonstrating that such a collaborative research process is feasible, this study has not only produced useful data on recidivism across a spectrum of the criminal justice system, but has also established a model that could be applied for future system-oriented research. 50 Comprehensive Recidivism Study APPENDIX - METHODOLOGY This appendix contains additional information of the method used to select the sample for this analysis and the definition of the variables collected for the analysis. SAMPLE All samples were submitted by the originating criminal justice agencies and were assumed to be complete. The data extraction and data linkage model provided for some validation between samples, e.g. the parole board sample was linked to parolees from all contributing correctional agencies. All validations indicated that the samples were substantially complete and the links between the various data sets were valid. Multiple Releases / Admissions. In all samples, the data were examined for cases with two or more program admissions or two or more releases during the sample period. It was possible that an offender could be released more than once during the three month sampling period. Where the case was determined to be a valid duplicate admission or release, both admissions or releases were included in the sample and subject to routine data extraction and linkage routines. It was also possible that offenders would be included as part of the sample of more than one criminal justice agency. A total of 22 cases were identified where the offender had two valid releases from the same facility or program admissions during the sampling period. In addition, a total of 38 cases were identified where the offender had two valid releases or program admissions from different facilities over the sampling period. Release Status. Offenders who were released by court or released to another correctional placement (transfer, pre-trial placement, or forthwith sentence) were not included in the analysis. Offenders released from a correctional facility from a “week-end” sentence were included only once in the sample. The most recent release among multiple weekend releases was selected. Those offenders released to parole from a correctional facility following a period of incarceration as the result of a 15-day parole detainer were excluded from the sample. For those offenders who were sentenced to a drug court or a community corrections center a s part of the probation supervision on a split sentence, the one year follow-up period began when the offender entered the program following release from the incarcerated portion of the split sentence. Conviction Status. It was not assumed that offenders participating in all of the originating criminal justice agencies had been convicted at the time of program participation. In particular, it was possible for a offender to be referred to a community 51 Massachusetts Sentencing Commission corrections center or drug court program following a disposition of general continuance or CWOF (Continued without a Finding). While a disposition of general continuance or CWOF is not considered a conviction, offenders are generally under probation supervision and may be required to attend a drug court or community corrections center program as part of the conditions of that probation supervision. Offenders in drug court or community corrections center samples were included in the analysis even if not convicted of the current offense. The release samples from the DOC and houses of correction included only those offenders released following a conviction and excluded any releases from pre-trial detention status. Length of Incarceration. Most recently, the DOC has defined a recidivist to be “any release who is re-incarcerated . . . for at least 30 days within three years of their date of release.”17 Because of the definition of a recidivist, the DOC includes “only inmates released off of a commitment of at least 30 days”18 in the analysis. This definition was not used as a sampling criterion, i.e. all inmates released during the sampling time frame were included. Juvenile Offenders / Youthful Offenders. The study sample included adult offenders. Those offenders adjudicated delinquent who were participating in the specialized community corrections centers or drug courts for juvenile offenders were not included in the study. Those offenders adjudicated youthful offenders, sentenced as adults, and released from an adult correction facility were included in the study. Four youthful offenders were identified and included in the study sample. Out-of-State and Federal Offenders. Some of the offenders in the release / program participation cohorts were originally sentenced in another state or federal court and subsequently transferred to Massachusetts. Not all of those offenders were identified in the CARI data-base. A total of 28 federal and out-of-state cases were identified and excluded from the final sample. Mortality Information. Any offenders who died within the follow-up period were excluded from the analysis. For each offender in the study sample, the identification 17 Hoover, op. cit., page 1. 18 Ibid., page 32. 52 Comprehensive Recidivism Study records and adult arraignment records in the CARI data-base were evaluated for occurrences of the phrase “DECEAS.” A total of 31 offender records were identified with this phrase. The date of death was taken from the Social Security Death Index (SSDI). Of the 31 identified offenders: 19 were excluded from the analysis because the date of death was verified and the death occurred within the one year follow-up period; 8 were included in the analysis because the date of death was verified and occurred after the one year follow-up period; and, 4 were included in the analysis because there was no verifying information in the SSDI. Whereabouts Unknown. An offender could be “whereabouts unknown” for some or all of the follow-up period. that period of time for which the offender was “whereabouts unknown” should be accounted for in the analysis. There were five parolees who were classified as “whereabouts unknown” during the recidivism follow-up period and remained in that classification until the end of that period. For the purposes of this analysis, these offenders have been classified as recidivists because the status of these offenders could not be determined for the entire follow-up period and they were counted in the category, “no new arrest / technical violation.” FY2002 Data Requirement. Chapter 177 contained a provision requiring that the study contain data from FY2002. Because of the reporting deadline and the desire to have a reasonable follow-up period, the study used data from time periods prior to FY2002. While some data from FY2002 was collected during the research process, the focus of the study was on data sets earlier than FY 2002. DATA COLLECTION Recidivism Variables Data collection for new arrest information was from the CARI database. As such, recidivism was measured as subsequent arrest / arraignment in Massachusetts and did not systematically capture criminal activity that occurred in other states. Data collection for parole and probation violations was from the PATS file and CARI database. Of primary interest was identifying those technical violations of supervised release that result in a period of incarceration. Those incarcerations resulting from 15-day parole detainers were not considered as indications of recidivism. In a recent study of recidivism by the DOC, of the 1,504 offenders defined as recidivists, 19.8% were re- 53 Massachusetts Sentencing Commission incarcerated technical violators and 80.2% were re-incarcerated following a new arrest.19 For those cases in drug courts and community corrections centers where there was no conviction of the current offense prior to the program admission, that is the offender was under probation supervision with a disposition of a CWOF or a general continuance, a violation that resulted in a guilty verdict being entered and a period of incarceration was treated as a probation violation. A number of offenders were arraigned for offenses prior to the correctional release or program participation and convicted of those offenses during the follow-up period. Those offenders who had an incarceration of less than 15 days resulting from a conviction of an offense that was arraigned prior to the correctional release or program participation were treated as non-recidivists and included in the sample. Those offenders who had an incarceration of more than 15 days resulting from a conviction of an offense arraigned prior to the correctional release or program participation and had no other new arrests or technical violations of parole or probation were excluded from the sample because they were not at liberty for the entire follow-up period. A total of 23 offenders were excluded for this reason. New Offenses. Six variables were included in the recidivism data base that describe the offenses for which offenders were charged during the follow-up period. The first three were selected from those charges that were arraigned during the first arraignment event following release or program participation: • • • most serious charge; most serious charge resulting in a conviction; and, most serious charge resulting in a sentence of incarceration. The second three variables were selected from all charges that were arraigned during the one year period following release or program participation: • • • 19 most serious charge; most serious charge resulting in a conviction; and, most serious charge resulting in a sentence of incarceration. Ibid., page 2. 54 Comprehensive Recidivism Study In all instances, the ranking that has been proposed by the Commission was used to assign a seriousness level to an offense. For purposes of the statistical presentation only two variables were selected: the offense level of the most serious charge among all new charges and the offense level of the most serious charge resulting in a conviction. The type of offense was assigned based on the most serious charge among all new charges. Routine editing was performed on offenses that were arraigned in the district court so that all offense levels were set consistent with district court jurisdiction. For example, an arraignment charge of robbery in the district court was assumed to be an arraignment charge of larceny person. If an offender was arraigned in district court during the followup period and indicted in the superior court after the follow-up period, the seriousness level of the new offense may be under-stated. The disposition and conviction status of all new offenses were determined based on the standard methodology developed by the Commission for use in the Survey of Sentencing Practices.20 If a defendant was arraigned on a new charge during the follow-up period and the case resulted in a conviction after the follow-up period, for purposes of this analysis, that charge was not counted as a conviction. Only new arraignments that occurred during the one year follow-up period and case processing activity that occurred during the one year follow-up period were included in the analysis of convictions. Type of Offense. All new offenses were assigned to one of six general type of offense categories consistent with the method used in the Survey of Sentencing Practices.21 The six categories are: • • • • • person offenses - includes murder, robberies, assault & battery; sex offenses - includes rapes, indecent assault & battery; property offenses - includes larceny, burglary, breaking & entering; drug offenses - includes possession and distribution of controlled substances; motor vehicle offenses - includes driving with licence suspended and vehicular homicide; and, other offenses - includes weapons, public disorder, and other offenses such as prostitution, disorderly person, dangerous weapons. • 20 Massac husetts Senten cing Com mission, Survey of Sentencing Practices, FY 2000, November 2001. 21 Ibid. 55 Massachusetts Sentencing Commission Incarceration Status. A number of variables was included in the recidivism data base that indicate if the new offense resulted in a sentence to incarceration during the one year follow-up period. The first three were selected from those charges that were arraigned during the first arraignment event following release or program participation: • • • number of charges resulting in a sentence of incarceration; first date an incarceration sentence was imposed; and, most serious charge resulting in a sentence of incarceration; The next three were selected from all charges that were arraigned during the one year period following release or program participation: • • • number of charges resulting in a sentence of incarceration; most recent date an incarceration sentence was imposed; and, most serious charge resulting in a sentence of incarceration; In all instances, the ranking that was developed by the Commission was used to assign a seriousness level to an offense. In most cases these variables were set based on an initial sentence to incarceration. For those offenders who had no other sentence of incarceration from any new offense behavior in the follow-up period but who did have one or more sentences to probation followed by a violation of probation that resulted in an incarceration for a new offense that occurred during the follow-up period, the recidivism incarceration variables were set based on the incarcerated violation of probation information. For purposes of the statistical presentation only one variable was selected: the number of charges resulting in a sentence of incarceration. Parole Violation Status. The Massachusetts Parole Board provided information on all offenders released to parole supervision during the three-month time period under study. For each of those offenders information on all administrative activity was available for the year following release from the correctional facility. The administrative chronology contained information on all detainers, warrants, and board decisions during the followup period. If the Parole Officer believes that the parolee has lapsed or is about to lapse into criminal ways or has violated the conditions of his parole and cannot remain in the community, the Parole Officer with the consent of a parole supervisor will issue a warrant for temporary custody, also known as a 15-day detainer. The 15-day detainer authorizes the parolee to 56 Comprehensive Recidivism Study be detained for a maximum of 15 days during which time the Board will conduct a preliminary revocation hearing. The parolee may postpone the preliminary revocation hearing to obtain legal representation or the attendance of witnesses. A second detainer may issue if the Board grants a postponement. If the parolee is being held on the 15-day detainer, the preliminary parole revocation hearing will be held by a Hearing Examiner during this time and the Parole Board, after considering the recommendation of the Hearing Examiner, will vote whether to provisionally revoke parole. If the Board votes to provisionally revoke parole, a parole violation warrant, also known as a warrant for permanent custody will be issued. The warrant for permanent custody is served when the parolee is physically detained under the Board's warrant. Upon service of the warrant for permanent custody, a final revocation hearing will be scheduled within 60 days. • If the Board provisionally revokes the parole, the parolee remains in custody until the Board conducts a final revocation hearing and makes a final decision regarding revocation. • If the Board does not provisionally revoke parole, the parolee is returned to the community under parole supervision. At the final revocation hearing, the Board members determine whether the parolee violated the terms and conditions of parole and whether the parolee will be re-released on parole supervision. • If the Board does not find by a preponderance of the evidence that the parolee violated a condition of parole, supervision will resume within 24 hours unless more time is necessary to notify victims, notify the District Attorney, approve the home plan, etc. The Board may modify or add conditions to parole. • Where the Board finds by a preponderance of the evidence that the parolee did violate the parole condition(s), it will 'affirm' the provisional parole revocation. The Board may either set a reparole date pending the fulfillment of certain conditions or the Board may vote to deny reparole, thus returning the parolee to an inmate status. Many of the offenders in the study were on parole at the end of the one year follow-up period. Other offenders successfully completed their parole during the one year followup period. An individual was considered to be a parole violator if a detainer was issued before the end of the one year follow-up period and if parole was subsequently revoked. 57 Massachusetts Sentencing Commission An individual was also considered to be a parole violator if a detainer was issued before the end of the one year follow-up period and if the offender ended the parole supervision while in custody and the period between the detainer and the end of the parole supervision was more than 15 days. An individual was not considered a parole violator if a detainer was issued but the parole was not revoked. Information on new arrests was taken from the CARI database. It was assumed that offenders who were classified as parole violators but who had no new arrests were technical parole violators. Probation Violation / Other Status. An offender was considered a probation violator if the adult criminal record included one or more charges that were arraigned prior to the release from a correctional facility or placement in a community based program which resulted in a violation of probation and a sentence to incarceration during the one year follow-up period. For example, it is not unusual for a person to be placed on probation subsequent to a period of incarceration as part of a split sentence or a from and after probation. Such persons would be counted as recidivists if they violated probation and were incarcerated during the one year follow-up period. Those violations of probation that resulted in a non-incarceration disposition (e.g. probation extended or probation conditions modified) were not used to classify the offender as a probation violator. If an offender had multiple charges that resulted in an incarceration as a probation violator, the earliest return as a probation violator was used. Offenders placed in a residential facility (e.g. detox or in-patient substance abuse facility) during the one year follow up period were not considered recidivists. Program Involvement For those correctional facilities that did not provide type of release, the parole board information was used to classify the offender as a parolee or a dischargee. For parolees, the security level of release was derived from the corresponding correctional record where that was available. Those offenders for whom the Parole Board’s interest in the case was closed on the same day as the release from the correctional institution were included in the category of discharges. For these offenders there was no period of parole supervision following the release. Offender Background Demographic Information. The principal source of offender demographic information was the CARI data-base. Because the CARI data-base did not contain the variable race 58 Comprehensive Recidivism Study for all offenders, supplemental data collection was done by the DOC and Parole Board to improve the quality of information available for the analysis. Current Offense and Criminal History. An optional data element that could be submitted by the originating criminal justice agency was the docket number(s) associated with the current placement. If the docket number(s) were submitted this allowed for the current offense and criminal history to be extracted from the record. An individual could have one or more docket numbers associated with the criminal justice placement at the time of the program admission or correctional release. All docket numbers submitted as part of the originating criminal justice agencies sample were inspected and linked to the CARI data-base. The docket number associated with the earliest arraignment date was used to set the parameters for the current offense. For parolees, the current offense and criminal history were derived from the corresponding correctional record where that was available. For youthful offenders, the current offense was contained in the juvenile portion of the CARI data-base and was assigned an offense seriousness level. Youthful offenders were not assigned to one of the five criminal history groups and the juvenile criminal history indicator excluded the adjudication for the current offense. All current offenses were assigned to an offense seriousness level and type of offense category in a manner consistent with the method established in the Survey of Sentencing Practices. Each defendant was assigned to a criminal history group in accordance with the five criminal history groups proposed by the Commission: E Serious Violent Record Two or more prior convictions for offenses in level 7 through 9 D Violent or Repetitive Record One prior conviction for offenses in levels 7 through 9, or Two or more prior convictions for offenses in levels 5 or 6, or Six or more prior convictions in levels 3, 4, 5 or 6 C Serious Record One prior conviction for offenses at levels 5 or 6, or Three to five prior convictions for offenses in levels 3 or 4 59 Massachusetts Sentencing Commission B Moderate Record One or two prior convictions for offenses in 3 or 4, or Six or more prior convictions for offenses in levels 1 or 2 A No/Minor Record One to five prior convictions for offenses in levels 1 or 2, or No prior convictions of any kind DATA EXTRACTION AND DATA LINKAGES The Commission used the computing facilities of the Criminal History Systems Board for data extraction and data linkage procedures relative to the CARI data-base. DATA ANALYSIS Missing Information. In all tables, cases with missing information have been reported. In the narrative, cases with missing information were excluded prior to the calculation of percentages that describe the characteristics of the population. State Prison Sentence. The category of state prison sentence included offenders with a life sentence, a state prison sentence, a state prison / split sentence, and a reformatory sentence. Non-Hierarchical Recidivism Classification. The definition of recidivism used for this analysis was based on the presence of one or more new arrests or re-incarceration as a parole or probation violator. Many offenders with parole or probation violations also had one or more new arrests. Table 19 shows the number of offenders classified in each category separately. Alternate Definitions of Recidivism.. The definition of recidivism used for this analysis relied on the presence of one or more new arrests or a re-incarceration due to a violation of parole or probation during the one year follow-up period. An offender with a new arrest was considered to be a recidivist even if that new charge had not reached disposition at the end of the one year follow-up period or if the new charge resulted in a non-conviction disposition (e.g., dismissed or CWOF) during the one year follow-up period. In this section, alternate recidivism definitions are considered. Two alternate recidivism rates were estimated: 60 Comprehensive Recidivism Study • • First, alternate recidivism rates were estimated if only those with charges resulting in a conviction during the one year follow-up period were considered to be recidivists; and, Second, alternate recidivism rates were estimated if only those with charges resulting in a sentence to incarceration were considered to be recidivists. In developing these alternate estimates the conviction decision or the sentence to incarceration had to occur within the one year follow-up period for the offender to be considered a recidivist. Table 20 shows estimated recidivism rates using these alternate definitions. In all cases, those offenders who were returned as the result of a parole or probation violation were included as recidivists. Offenders who had an incarceration resulting from some other offense were included in the conviction and incarceration category. Overall, the recidivism rate, based on arrest / arraignment and including technical violators, was estimated to be 49.1%. If only arrests that resulted in a conviction during the one year follow-up period were considered, the estimated recidivism rate would be 30.3%. If only arrests that resulted in a conviction and a sentence of incarceration during the one year follow-up period were considered, the estimated recidivism rate would be 20.7%. 61 Massachusetts Sentencing Commission Table 19. Non-hierarchical Recidivism Classification by Criminal Justice Cohort Sample Total Drug Co urts Type of Recidivist Commu nity Corrections C enters House of Correction Sentence Discharge Paro le State Prison Sentence Sample Total Discharge Paro le Sample Total N % N % N % N % N % N % N % N % N % None 2,083 55.5% 47 61.8% 196 60.7% 1,126 50.2% 376 61.5% 1,502 52.6% 212 63.7% 126 76.4% 338 67.9% One or mo re 1,668 44.5% 29 38.2% 127 39.3% 1,117 49.8% 235 38.5% 1,352 47.4% 121 36.3% 39 23.6% 160 32.1% Total 3,751 100.0% 76 100.0% 323 100.0% 2,243 100.0% 611 100.0% 2,854 100.0% 333 100.0% 165 100.0% 498 100.0% 3,569 95.1% 76 100.0% 322 99.7% 2,243 100.0% 487 79.7% 2,730 95.7% 333 100.0% 108 65.5% 441 88.6% New Arrest Parole Violation R eturn None One or mo re 182 4.9% 0 0.0% 1 0.3% 0 0.0% 124 20.3% 124 4.3% 0 0.0% 57 34.5% 57 11.4% Total 3,751 100.0% 76 100.0% 323 100.0% 2,243 100.0% 611 100.0% 2,854 100.0% 333 100.0% 165 100.0% 498 100.0% 3,438 91.7% 54 71.1% 240 74.3% 2,073 92.4% 584 95.6% 2,657 93.1% 327 98.2% 160 97.0% 487 97.8% Probation V iolation Return None One or mo re 313 8.3% 22 28.9% 83 25.7% 170 7.6% 27 4.4% 197 6.9% 6 1.8% 5 3.0% 11 2.2% Total 3,751 100.0% 76 100.0% 323 100.0% 2,243 100.0% 611 100.0% 2,854 100.0% 333 100.0% 165 100.0% 498 100.0% 62 Comprehensive Recidivism Study Table 20. Alternate Recidivism Measures by Criminal Justice Cohort House of Correction Sentence R ecidivists R ecidivism R ate R ecidivists R ecidivism R ate R ecidivists R ecidivism R ate R ecidivists R ecidivism R ate R ecidivists R ecidivism R ate Sample Total (n = 498) R ecidivism R ate Paro le (n = 165) R ecidivists Discharge (n = 333) R ecidivism R ate Sample Total (n = 2854) R ecidivists Paro le (n = 611) R ecidivism R ate Discharge ( n = 2243) State Prison Sentence R ecidivists Commu nity Corrections Centers (n =323) R ecidivism R ate Drug Co urts (n =76) R ecidivists Sample Total (n =3751) One or M ore Arrests 1841 49.1% 38 50.0% 168 52.0% 1156 51.5% 288 47.1% 1444 50.6% 121 36.3% 70 42.4% 191 38.4% One or M ore Convictions 1136 30.3% 28 36.8% 116 35.9% 673 30.0% 206 33.7% 879 30.8% 52 15.6% 61 37.0% 113 22.7% One or M ore Incarcerations 777 20.7% 23 30.3% 89 27.6% 419 18.7% 164 26.8% 583 20.4% 25 7.5% 57 34.5% 82 16.5% Time Until Recidivism Incident Note: Offenders incarcerated as a result of a violation of parole or probation or some other offense were included in all categories. 63 Massachusetts Sentencing Commission The estimated recidivism rates (i.e. any new arrest, only convicted behavior, or only incarcerated behavior) were sensitive to the time during the follow-up period that the recidivism behavior occurred. Table 21 shows the relationships between these variables. As expected, the probability of conviction within the one year follow-up period was related to the time during the follow-up period when the recidivism behavior occurred. For those offenders classified as a recidivist in the first month, 78.9% had one or more convictions within the one year follow-up period. In contrast, for those offenders classified as a recidivist in month eight, 44.8% had one or more convictions within the one year follow-up period. In general, because a disposition decision in any case takes some time, the use of a conviction definition biased recidivism rate estimates towards those populations with recidivism behavior occurring more quickly. Because parole and probation violators were included in all of the alternate recidivism measures, the relation between the two measures was close for the parole and community based program cohorts. 64 Comprehensive Recidivism Study Table 21. Conviction Status by Time Until Recidivism and Criminal Justice Cohort House of Correction Sentence 96.8% 178 140 78.7% 8 5 62.5% % 30 All C onvictions 31 All Arrests 110 74.8% % 147 All C onvictions 77.8% Sample Total All Arrests All Arrests 21 % % 27 Paro le All C onvictions All C onvictions 100.0% Discharge All Arrests All Arrests 4 % % 4 All Arrests All C onvictions 78.9% % All Arrests 176 All C onvictions % 223 Sample Total All Arrests All C onvictions Less than 1 m onth Paro le % Time U ntil Recidivism All Arrests Discharge State Prison Sentence All C onvictions Commu nity Corrections C enters Drug Co urts All C onvictions Sample Total 6 6 100.0% 14 11 78.6% 1 to 2 months 219 157 71.7% 9 7 77.8% 23 19 82.6% 133 83 62.4% 38 34 89.5% 171 117 68.4% 11 9 81.8% 5 5 100.0% 16 14 87.5% 2 to 3 months 215 163 75.8% 7 3 42.9% 19 18 94.7% 138 100 72.5% 26 22 84.6% 164 122 74.4% 17 12 70.6% 9 9 100.0% 26 21 80.8% 3 to 4 months 187 135 72.2% 2 2 100.0% 21 16 76.2% 113 79 69.9% 29 22 75.9% 142 101 71.1% 12 6 50.0% 10 10 100.0% 22 16 72.7% 4 to 5 months 196 128 65.3% 3 2 66.7% 14 10 71.4% 131 85 64.9% 25 20 80.0% 156 105 67.3% 15 4 26.7% 7 6 85.7% 22 10 45.5% 5 to 6 months 148 99 66.9% 3 3 100.0% 15 9 60.0% 98 63 64.3% 19 17 89.5% 117 80 9 3 33.3% 4 4 100.0% 13 7 53.8% 68.4% 6 to 7 months 128 77 60.2% 2 2 100.0% 10 8 80.0% 80 46 57.5% 19 11 57.9% 99 57 57.6% 11 5 45.5% 6 5 83.3% 17 10 58.8% 7 to 8 months 116 61 52.6% 2 1 50.0% 7 6 85.7% 69 33 47.8% 24 15 62.5% 93 48 51.6% 11 3 27.3% 3 3 100.0% 14 6 42.9% 8 to 9 months 116 52 44.8% 3 3 100.0% 11 4 36.4% 70 31 44.3% 21 9 42.9% 91 40 44.0% 7 3 42.9% 4 2 50.0% 11 5 45.5% 9 to 10 months 103 43 41.7% 1 1 100.0% 9 2 22.2% 67 24 35.8% 14 10 71.4% 81 34 42.0% 7 2 28.6% 5 4 80.0% 12 6 50.0% 10 to 11 months 88 32 36.4% 0 0 N. A . 8 3 37.5% 43 16 37.2% 24 9 37.5% 67 25 37.3% 7 0 0.0% 7 5 71.4% 14 5 35.7% 11 to 12 months 102 13 12.7% 2 0 0.0% 4 0 0.0% 67 3 4.5% 18 7 38.9% 85 10 11.8% 6 0 0.0% 4 2 50.0% 10 2 20.0% 1,841 1136 61.7% 38 28 73.7% 168 288 206 71.5% 1444 121 52 43.0% 70 61 87.1% 191 113 59.2% Total 116 69.0% 1,156 673 58.2% 65 879 60.9% Massachusetts Sentencing Commission APPENDIX - HOUSE OF CORRECTION SENTENCED OFFENDERS For reference purposes, Table 22 shows recidivism rates and associated offender characteristics for house of correction sentenced offenders by type of release and the correctional jurisdiction of release: • • • • house of correction house of correction house of correction house of correction sentence / discharged / released from county facility; sentence / paroled / released from county facility; sentence / discharged / released from DOC facility; sentence / paroled / released from DOC facility. 66 Comprehensive Recidivism Study Table 22. Selected Recidivism Characteristics by Type of Release and Custody at Time of Release, House of Correction Sentenced Offenders House Sentence / Released from House Recidivism Char acter istic Discharge Paro le H o u se S e nt e nc e / Re l ea s e d f r om D O C Sample Total Discharge Paro le Sample Total N % N % N % N % N % N % Recidiv ism Ra te 1,082 51.7% 254 46.4% 1,336 50.6% 74 49.3% 34 54.0% 108 50.7% One or M ore Arrests 1,049 50.1% 206 37.6% 1,255 47.5% 68 45.3% 29 46.0% 97 45.5% Type of Recidivist Most serious: felony 507 24.2% 97 17.7% 604 22.9% 23 15.3% 7 11.1% 30 14.1% misdemeanor 542 25.9% 109 19.9% 651 24.6% 45 30.0% 22 34.9% 67 31.5% No Arrests / Technical 33 1.6% 48 8.8% 81 3.1% 6 4.0% 5 7.9% 11 5.2% Non-Recidivist 1,011 48.3% 294 53.6% 1,305 49.4% 76 50.7% 29 46.0% 105 49.3% Total 2,093 100.0% 548 100.0% 2,641 100.0% 150 100.0% 63 100.0% 213 100.0% New Cases 134 Cum Rate 6.4% New Cases 29 Cum Rate 5.3% New Cases 163 Cum Rate 6.2% New Cases 13 Cum Rate 8.7% New Cases 2 Cum Rate 3.2% New Cases 15 Cum Rate 7.0% Time Until Recidivism Incident Less than 1 m onth 1 to 2 months 125 12.4% 35 11.7% 160 12.2% 8 14.0% 3 7.9% 11 12.2% 2 to 3 months 134 18.8% 19 15.1% 153 18.0% 4 16.7% 7 19.0% 11 17.4% 3 to 4 months 102 23.7% 25 19.7% 127 22.8% 11 24.0% 4 25.4% 15 24.4% 4 to 5 months 122 29.5% 22 23.7% 144 28.3% 9 30.0% 3 30.2% 12 30.0% 5 to 6 months 94 34.0% 14 26.3% 108 32.4% 4 32.7% 5 38.1% 9 34.3% 6 to 7 months 75 37.6% 17 29.4% 92 35.9% 5 36.0% 2 41.3% 7 37.6% 7 to 8 months 65 40.7% 22 33.4% 87 39.2% 4 38.7% 2 44.4% 6 40.4% 8 to 9 months 67 43.9% 19 36.9% 86 42.4% 3 40.7% 2 47.6% 5 42.7% 9 to 10 months 59 46.7% 12 39.1% 71 45.1% 8 46.0% 2 50.8% 10 47.4% 10 to 11 months 42 48.7% 23 43.2% 65 47.6% 1 46.7% 1 52.4% 2 48.4% 11 to 12 months 63 51.7% 17 46.4% 80 50.6% 4 49.3% 1 54.0% 5 50.7% 67 Massachusetts Sentencing Commission House Sentence / Released from House Recidivism Char acter istic Discharge N % Paro le N H o u se S e nt e nc e / Re l ea s e d f r om D O C Sample Total % N % Discharge N % Paro le N Sample Total % N % Offense Seriousness Level, Most Serious New Arrest Offense 9 1 0.0% 0 0.0% 1 0.0% 0 0.0% 0 0.0% 0 0.0% 8 4 0.2% 0 0.0% 4 0.2% 0 0.0% 0 0.0% 0 0.0% 7 6 0.3% 0 0.0% 6 0.2% 0 0.0% 0 0.0% 0 0.0% 6 23 1.1% 1 0.2% 24 0.9% 0 0.0% 0 0.0% 0 0.0% 5 46 2.2% 13 2.4% 59 2.2% 1 0.7% 0 0.0% 1 0.5% 4 157 7.5% 16 2.9% 173 6.6% 6 4.0% 1 1.6% 7 3.3% 3 419 20.0% 84 15.3% 503 19.0% 30 20.0% 6 9.5% 36 16.9% 2 239 11.4% 52 9.5% 291 11.0% 18 12.0% 15 23.8% 33 15.5% 1 153 7.3% 38 6.9% 191 7.2% 13 8.7% 7 11.1% 20 9.4% Not Assigned 1 0.0% 2 0.4% 3 0.1% 0 0.0% 0 0.0% 0 0.0% Total 1,049 50.1% 206 37.6% 1,255 47.5% 68 45.3% 29 46.0% 97 45.5% Type of Offense, M ost Serious New Arrest Offense Person 304 29.0% 41 19.9% 345 27.5% 22 32.4% 2 6.9% 24 24.7% Sex 17 1.6% 3 1.5% 20 1.6% 0 0.0% 0 0.0% 0 0.0% Property 249 23.7% 68 33.0% 317 25.3% 18 26.5% 9 31.0% 27 27.8% Drug 198 18.9% 33 16.0% 231 18.4% 8 11.8% 7 24.1% 15 15.5% Mo tor V ehicle 172 16.4% 39 18.9% 211 16.8% 11 16.2% 5 17.2% 16 16.5% Other 109 10.4% 22 10.7% 131 10.4% 9 13.2% 6 20.7% 15 15.5% Total 1,049 100.0% 206 100.0% 1,255 100.0% 68 100.0% 29 100.0% 97 100.0% 68 Comprehensive Recidivism Study Table 23. Selected Offender Characteristics by Type of Release and Custody at Time of Release, House of Correction Sentenced Offenders House Sentence / Released from House H o u se S e nt e nc e / Re l ea s e d f r om D O C R ecidivism R ate Total R ecidivists R ecidivism R ate Total R ecidivists R ecidivism R ate Total R ecidivists R ecidivism R ate Total R ecidivists R ecidivism R ate Sample Total R ecidivists Paro le Total Discharge R ecidivism R ate Sample Total R ecidivists Paro le Total Discharge Maximum Medium Medium/M inimum Sub-Total Se cure 127 597 67 791 61 352 35 448 48.0% 59.0% 52.2% 56.6% 2 67 15 84 1 38 9 48 50.0% 56.7% 60.0% 57.1% 129 664 82 875 62 390 44 496 48.1% 58.7% 53.7% 56.7% 0 110 0 110 0 60 0 60 N.A. 54.5% N.A. 54.5% 0 30 0 30 0 17 0 17 N.A. 56.7% N.A. 56.7% 0 140 0 140 0 77 0 77 N.A. 55.0% N.A. 55.0% Minimum Minimum/Pre-release Pre-release Sub-Total Lower 293 0 72 365 125 0 31 156 42.7% N.A. 43.1% 42.7% 72 0 51 123 31 0 20 51 43.1% N.A. 39.2% 41.5% 365 0 123 488 156 0 51 207 42.7% N.A. 41.5% 42.4% 4 32 4 40 0 11 3 14 0.0% 34.4% 75.0% 35.0% 18 13 2 33 10 6 1 17 55.6% 46.2% 50.0% 51.5% 22 45 6 73 10 17 4 31 45.5% 37.8% 66.7% 42.5% Not Provided 937 478 51.0% 341 155 45.5% 1278 633 49.5% 0 0 N.A. 0 0 N.A. 0 0 N.A. Total 2093 1082 51.7% 548 254 46.4% 2641 1336 50.6% 150 74 49.3% 63 34 54.0% 213 108 50.7% Gender Fema le Ma le Total 132 1961 2093 58 1024 1082 43.9% 52.2% 51.7% 65 483 548 32 222 254 49.2% 46.0% 46.4% 197 2444 2641 90 1246 1336 45.7% 51.0% 50.6% 127 23 150 67 7 74 52.8% 30.4% 49.3% 38 25 63 20 14 34 52.6% 56.0% 54.0% 165 48 213 87 21 108 52.7% 43.8% 50.7% Offend er Ch aract eristic Security Level at Release 69 Massachusetts Sentencing Commission House Sentence / Released from House H o u se S e nt e nc e / Re l ea s e d f r om D O C R ecidivism R ate Total R ecidivists R ecidivism R ate Total R ecidivists R ecidivism R ate Total R ecidivists R ecidivism R ate Total R ecidivists R ecidivism R ate Sample Total R ecidivists Paro le Total Discharge R ecidivism R ate Sample Total R ecidivists Paro le Total Discharge Age at Release Under 20 20 to 24 25 to 29 30 to 34 35 to 39 40 to 49 50 to 59 60 to 69 70 to 79 Total 149 413 358 336 375 374 74 13 1 2093 88 243 197 183 185 160 23 3 0 1082 59.1% 58.8% 55.0% 54.5% 49.3% 42.8% 31.1% 23.1% 0.0% 51.7% 65 119 108 81 83 72 16 4 0 548 37 60 51 41 39 22 3 1 0 254 56.9% 50.4% 47.2% 50.6% 47.0% 30.6% 18.8% 25.0% N.A. 46.4% 214 532 466 417 458 446 90 17 1 2641 125 303 248 224 224 182 26 4 0 1336 58.4% 57.0% 53.2% 53.7% 48.9% 40.8% 28.9% 23.5% 0.0% 50.6% 6 13 27 34 34 25 11 0 0 150 4 9 17 17 17 9 1 0 0 74 66.7% 69.2% 63.0% 50.0% 50.0% 36.0% 9.1% N.A. N.A. 49.3% 3 13 12 10 14 10 1 0 0 63 3 7 7 4 11 2 0 0 0 34 100.0% 53.8% 58.3% 40.0% 78.6% 20.0% 0.0% N.A. N.A. 54.0% 9 26 39 44 48 35 12 0 0 213 7 16 24 21 28 11 1 0 0 108 77.8% 61.5% 61.5% 47.7% 58.3% 31.4% 8.3% N.A. N.A. 50.7% Race / E thnicity White Black Hispa nic Other U n k n ow n Total 1029 544 438 43 39 2093 478 323 252 16 13 1082 46.5% 59.4% 57.5% 37.2% 33.3% 51.7% 361 87 90 10 0 548 159 41 51 3 0 254 44.0% 47.1% 56.7% 30.0% N.A. 46.4% 1390 631 528 53 39 2641 637 364 303 19 13 1336 45.8% 57.7% 57.4% 35.8% 33.3% 50.6% 110 21 17 2 0 150 54 7 12 1 0 74 49.1% 33.3% 70.6% 50.0% N.A. 49.3% 53 4 4 2 0 63 27 4 2 1 0 34 50.9% 100.0% 50.0% 50.0% N.A. 54.0% 163 25 21 4 0 213 81 11 14 2 0 108 49.7% 44.0% 66.7% 50.0% N.A. 50.7% Type of Offense (Current Offense) Person Sex Property Drug Mo tor V ehicle Other Not A vailable Total 338 22 251 307 198 97 880 2093 184 4 143 165 80 54 452 1082 54.7% 18.2% 55.7% 54.1% 39.4% 56.3% 51.4% 51.7% 108 3 137 176 102 22 0 548 54 2 77 80 29 12 0 254 50.9% 66.7% 57.2% 45.9% 27.3% 58.6% N.A. 46.4% 446 25 388 483 300 119 880 2641 238 6 220 245 109 66 452 1336 53.8% 24.0% 56.3% 51.0% 35.5% 56.7% 51.4% 50.6% 35 0 36 26 38 15 0 150 20 0 17 15 13 9 0 74 38.8% 20.0% 51.6% 25.8% 37.5% 46.2% N.A. 49.3% 6 0 22 20 8 7 0 63 4 0 14 10 1 5 0 34 42.3% 14.3% 62.1% 30.8% 100.0% 80.0% N.A. 54.0% 41 0 58 46 46 22 0 213 24 0 31 25 14 14 0 108 40.0% 19.2% 54.9% 28.0% 44.4% 55.6% N.A. 50.7% Offend er Ch aract eristic 70 Comprehensive Recidivism Study House Sentence / Released from House H o u se S e nt e nc e / Re l ea s e d f r om D O C R ecidivism R ate Total R ecidivists R ecidivism R ate Total R ecidivists R ecidivism R ate Total R ecidivists R ecidivism R ate Total R ecidivists R ecidivism R ate Sample Total R ecidivists Paro le Total Discharge R ecidivism R ate Sample Total R ecidivists Paro le Total Discharge Offense Seriousness Level (Current Offense) 9 8 7 6 5 4 3 2 1 Not Assigned Not A vailable Total 0 0 0 0 23 173 505 356 151 5 880 2093 0 0 0 0 10 84 268 178 88 2 452 1082 N.A. N.A. N.A. N.A. 43.5% 48.6% 53.1% 50.0% 58.3% 40.0% 51.4% 51.7% 0 0 0 2 3 132 231 133 45 2 0 548 0 0 0 0 2 56 113 60 22 1 0 254 N.A. N.A. N.A. 0.0% 66.7% 42.4% 48.9% 45.1% 48.9% 50.0% N.A. 46.4% 0 0 0 2 26 305 736 489 196 7 880 2641 0 0 0 0 12 140 381 238 110 3 452 1336 N.A. N.A. N.A. 0.0% 46.2% 45.9% 51.8% 48.7% 56.1% 42.9% 51.4% 50.6% 0 0 0 0 0 14 74 49 10 3 0 150 0 0 0 0 0 7 35 24 7 1 0 74 N.A. N.A. N.A. N.A. N.A. 50.0% 47.3% 49.0% 70.0% 33.3% N.A. 49.3% 0 0 0 0 0 11 24 22 5 1 0 63 0 0 0 0 0 4 16 12 2 0 0 34 N.A. N.A. N.A. N.A. N.A. 36.4% 66.7% 54.5% 40.0% 0.0% N.A. 54.0% 0 0 0 0 0 25 98 71 15 4 0 213 0 0 0 0 0 11 51 36 9 1 0 108 N.A. N.A. N.A. N.A. N.A. 44.0% 52.0% 50.7% 60.0% 25.0% N.A. 50.7% Criminal History Group Group A - No/M inor Record Group B - Mode rate Reco rd Group C - Serious Re cord Group D - V iolent or Repetitive Group E - Serious V iolent Not Assigned or Youthful Offender Total 284 342 299 285 1 882 2093 110 167 170 181 1 453 1082 38.7% 48.8% 56.9% 63.5% 100.0% 51.4% 51.7% 50 87 52 21 0 338 548 19 43 25 13 0 154 254 38.0% 49.4% 48.1% 61.9% N.A. 45.6% 46.4% 334 429 351 306 1 1220 2641 129 210 195 194 1 607 1336 38.6% 49.0% 55.6% 63.4% 100.0% 49.8% 50.6% 53 62 22 13 0 0 150 21 33 13 7 0 0 74 39.6% 53.2% 59.1% 53.8% N.A. N.A. 49.3% 15 27 15 6 0 0 63 8 16 5 5 0 0 34 53.3% 59.3% 33.3% 83.3% N.A. N.A. 54.0% 68 89 37 19 0 0 213 29 49 18 12 0 0 108 42.6% 55.1% 48.6% 63.2% N.A. N.A. 50.7% Juvenile Criminal History No Prior Adjudications One or M ore Adjudications Total 1579 514 2093 757 325 1082 47.9% 63.2% 51.7% 423 125 548 179 75 254 42.3% 60.0% 46.4% 2002 639 2641 936 400 1336 46.8% 62.6% 50.6% 139 11 150 68 6 74 48.9% 54.5% 49.3% 54 9 63 28 6 34 51.9% 66.7% 54.0% 193 20 213 96 12 108 49.7% 60.0% 50.7% Offend er Ch aract eristic 71