Young Male DV Cases 10 by Administrator

CJA

NEW YORK CITY CRIMINAL JUSTICE AGENCY, INC. NEW YORK CITY CRIMINAL USTICE AGENCY

Jerome E. McElroy Executive Director

YOUNG MALE DOMESTIC VIOLENCE OFFENDERS IN NEW YORK CITY

Richard R. Peterson, Ph.D. Project Director and Director, Research Department

FINAL REPORT

January 2010 52 Duane Street, New York, NY 10007

(646) 213-2500

YOUNG MALE DOMESTIC VIOLENCE OFFENDERS IN NEW YORK CITY

Richard R. Peterson, Ph.D. Project Director and Director, Research Department Research Assistance: Steve Mardenfeld Senior Data Analyst Raymond P. Caligiure Graphics and Production Specialist Information Systems Programming: Wayne Nehwadowich Senior Programmer/Analyst Administrative Support: Annie Su Administrative Associate

January 2010

ď&#x192;Ł 2010 NYC Criminal Justice Agency, Inc.

ACKNOWLEDGEMENTS

This report could not have been completed without the assistance of colleagues at CJA, Judges, members of the District Attorneysâ&#x20AC;&#x2122; offices and others. The author appreciates the work of Marian Gewirtz and Elyse J. Revere, who developed the third quarter 2005 dataset, and of Steve Mardenfeld, who prepared the data for analysis and checked the final draft of the manuscript. The author also thanks Annie Su who prepared the figures and appendices, Raymond P. Caligiure who created several tables and David J. Hauser who checked the final manuscript. The author extends special thanks to Wayne Nehwadowich who programmed the case processing data. The author also acknowledges other colleagues at CJA who provided advice, information, and editorial suggestions: Jerome E. McElroy, Executive Director of CJA, and Mari Curbelo, Esq., Geraldine Ferrara, Esq., Peter Kiers, Dr. Mary Phillips, Frank Sergi, Dr. Qudsia Siddiqi and Dr. Freda F. Solomon. The author would also like to thank Deirdre Bialo-Padin, Esq., Abena Darkeh, Esq., Hon. Matthew J. Dâ&#x20AC;&#x2122;Emic, Prof. Jo Dixon, Peter Glick, Esq., Scott Kessler, Esq., Karen Kleinberg, Esq., Ms. Sharon Lastique, Hon. John M. Leventhal, Wanda Lucibello, Esq., Hon. Deborah Stevens Modica, Audrey Moore, Esq. and Hon. Esther M. Morgenstern for their assistance with the research. The methodology, findings, and conclusions of the study, as well as any errors, omissions and misinterpretations are the responsibility of the author.

TABLE OF CONTENTS

Introduction ........................................................................................................... 1 A. Review of the Literature .................................................................................... 2 B. Research Plan................................................................................................... 4

II.

Methodology.......................................................................................................... 5 A. B. C. D. E. F.

III.

A Portrait of Young Male Domestic Violence Offenders ................................. 13 A. B. C. D.

IV.

Offender and Case Characteristics ................................................................. 14 Case Dispositions, Sentence Outcomes, and Length of Jail Sentences ......... 25 Pretrial Misconduct.......................................................................................... 29 Summary and Discussion of Findings ............................................................. 34

Explaining Differences in Case Outcomes and Pretrial Misconduct Between Young Male DV Offenders and Other Offenders .............................................. 37 A. B. C. D. E.

Overview of the CJA Database and the Third Quarter 2005 Dataset ................ 5 Identifying Domestic Violence Cases ................................................................ 6 Identifying Re-arrests for Domestic Violence Offenses ..................................... 9 Selection of the Crimes Against Persons and Property Subsample ................ 10 Using a Defendant-Based Data File ................................................................ 11 Plan of Analysis............................................................................................... 11

Conviction ....................................................................................................... 38 Failure to Appear............................................................................................. 39 Re-arrest for a New DV Offense...................................................................... 42 Re-arrest for a New Non-DV Offense.............................................................. 43 Summary and Discussion of Findings ............................................................. 46

Predicting Pretrial Misconduct Among Young Male DV Offenders ................ 49 A. Failure to Appear............................................................................................. 49 B. Pretrial Re-arrest for a New DV Offense ......................................................... 50 C. Summary and Discussion of Findings ............................................................. 54

Table of Contents Continues on Next Page

TABLE OF CONTENTS, CONTINUED VI.

Conclusion .......................................................................................................... 57 A. Major Findings................................................................................................. 57 B. Discussion ....................................................................................................... 58 C. Conclusion....................................................................................................... 62

VII. References .......................................................................................................... 65 Appendix A: Statistical Methods............................................................................... 69 1. Logistic Regression Analysis........................................................................... 69 2. Correcting for Selection Bias ........................................................................... 72 Appendix B: Distribution of Variables for Regression Models .............................. 75 Appendix C: Coding of Variables for Regression Models ...................................... 79

LIST OF TABLES

Table 3-1

Offender’s Demographic Characteristics by Age for DV and Non-DV Offenders ................................................................................................... 14

Table 3-2

Offender-Victim Relationship by Age for DV Offenders ............................. 16

Table 3-3

Offender’s Criminal History by Age for DV and Non-DV Offenders ........... 17

Table 3-4

Outcome of Prior DV Arrest by Age for DV Offenders ............................... 18

Table 3-5

Community Ties by Age for DV and Non-DV Offenders ............................ 20

Table 3-6

Case Filing Outcome by Age for DV and Non-DV Offenders..................... 21

Table 3-7

Arrest and Arraignment Charge Characteristics by Age for DV and Non-DV Offenders ................................................................................................... 22

Table 3-8

Case Processing Characteristics by Age for DV and Non-DV Offenders .. 24

Table 4-1

Differences in Case Outcomes and Pretrial Misconduct Between Young Male DV Offenders and Other Offenders ....................................... 37

Table 4-2

Differences in Odds of Conviction Between Young Male DV Offenders and Young Male Non-DV Offenders................ 39

Table 4-3

Differences in Odds of Failure to Appear Between Young Male DV Offenders and Older Male DV Offenders ......................... 40

Table 4-4

Differences in Odds of Failure to Appear Between Young Male DV Offenders and Young Male Non-DV Offenders................ 41

Table 4-5

Differences in Odds of Re-arrest for a New DV Offense Between Young Male DV Offenders and Young Male Non-DV Offenders................ 43

Table 4-6

Differences in Odds of Re-arrest for a New Non-DV Offense Between Young Male DV Offenders and Older Male DV Offenders ......................... 44

Table 4-7

Differences in Odds of Re-arrest for a New Non-DV Offense Between Young Male DV Offenders and Young Male Non-DV Offenders................ 45

List of Tables Continues on Next Page

iii

LIST OF TABLES, CONTINUED

Table 5-1

Logistic Regression Model Predicting Likelihood of Failure to Appear Among Released Defendants .................................................................... 50

Table 5-2

Logistic Regression Model Predicting Likelihood of Pretrial Re-arrest for a New DV Offense Among Released Defendants ........................................ 51

Table 5-3

Logistic Regression Model Predicting Likelihood of Pretrial Re-arrest for a New DV Offense Among Released Defendants Who Were Arrested for a DV Offense in the Prior Two Years............................................................ 52

Table 5-4

Logistic Regression Model Predicting Likelihood of Pretrial Re-arrest for a New DV Offense Among Released Defendants Not Arrested for a DV Offense in the Prior Two Years.................................................................. 53

Table 5-5

Pretrial Re-arrest Rate for New DV Offenses Among Released Defendants by Engagement in Full-Time Activity and History of DV Arrests in Prior Two Years ......................................................................................................... 56

LIST OF FIGURES

FIGURE 3-1

Case Dispositions in Criminal Court ..................................................... 26

FIGURE 3-2

Sentence Outcomes in Criminal Court ................................................. 27

FIGURE 3-3

Length of Jail Sentence in Criminal Court ............................................ 29

FIGURE 3-4

Failure to Appear Rates by Age for DV and Non-DV Defendants Who Were Ever Released.................................................................... 31

FIGURE 3-5A Pretrial Re-arrest Rates for Any New Offenses by Age for DV and Non-DV Defendants Who Were Ever Released ...................... 33 FIGURE 3-5B Pretrial Re-arrest Rates for New DV and Non-DV Offenses by Age for DV and Non-DV Defendants Who Were Ever Released ...................... 33

YOUNG MALE DOMESTIC VIOLENCE OFFENDERS IN NEW YORK CITY I. INTRODUCTION The criminal justice system has devoted considerable resources to addressing domestic violence in the United States over the last two decades. Domestic violence incidents are now being taken more seriously as a result of more frequent arrests, vigorous prosecution and harsher sentencing laws. Federal legislation, notably the Violence Against Women Act (VAWA) of 1994 (re-authorized in 2000 and 2006), encouraged new efforts to combat domestic violence. Under this Act, the Department of Justice has provided funding and technical assistance to state and local law enforcement agencies. New York State legislation required mandatory arrest for certain family offenses, enhanced penalties for violating an order of protection, and criminalized stalking. New York City created specialized police units, prosecution bureaus, and court parts to focus more attention on domestic violence cases. While much has been done to address domestic violence, it is not clear that the new policies and practices have been effective at “specific deterrence,” that is, deterring offenders from committing new offenses.1 The consensus of recent studies is that arrest has a relatively small deterrent effect on domestic violence offenders (Maxwell et al. 2002). Prosecution, conviction and jail sentences are generally ineffective (Peterson 2003). There is little evidence that specialized prosecution units, specialized domestic violence courts (Aos, Miller and Drake 2006, Peterson 2004), and batterer intervention programs (Aos et al. 2006, Feder and Wilson 2005) deter offenders. Perhaps these findings should not be viewed as surprising, since there are many other types of crime where criminal justice interventions have not been found to deter offenders. Nevertheless, these findings suggest that a new focus is needed to determine whether “specific deterrence” is possible in domestic violence cases (Peterson 2008a). One area that may be promising is early intervention. Most evaluations of criminal justice efforts to deter domestic violence have focused not on early intervention but on the general population of offenders. These studies almost universally find that offenders with more extensive histories of domestic violence are more likely to recidivate. The best predictor of recidivism is the offender’s history of committing domestic violence. This suggests that once an offender has an extensive domestic violence history, intervention is unlikely to be successful. But it also raises the question: Would criminal justice interventions at an early age prevent offenders from developing an extensive history of domestic violence? A focus on young offenders might provide some evidence to identify an effective early intervention. This report examines criminal court case processing of young male domestic violence offenders in New York City. It compares and contrasts these offenders with 1

Lerman (1992:225) argues that “specific deterrence of a particular offender is not the only goal.” She argues that general deterrence (deterring other potential offenders) and establishing community standards against domestic violence are also important goals. The current study does not address these goals because they are beyond the scope of the available data.

2 older male domestic violence offenders as well as young male offenders charged with offenses that do not involve domestic violence. It also begins to address the question of whether early intervention can deter offenders by examining the impact of criminal justice interventions on pretrial re-arrests for new DV offenses. A. Review of the Literature Several studies of the criminal justice system have examined the connection between age and domestic violence. Most of these studies include data on both male and female offenders, so they do not directly address how age influences offending patterns among male DV offenders. However, in most studies, a large majority of DV offenders are male, so it is likely that the findings summarized below can be generalized to the population of male DV offenders. Research generally shows that age influences the likelihood of recidivism and pretrial misconduct among DV offenders. Older DV defendants (40 and over) are less likely to fail to make scheduled court appearances than younger DV defendants (Peterson 2006). DV defendants 40 and over are also less likely to be re-arrested for any offense during the pretrial period (Peterson 2008b), and less likely to be re-arrested for a new DV offense during the pretrial period (Peterson 2006). Older DV defendants also are less likely to be re-arrested for a new DV offense following case disposition (Murphy et al. 1998, Peterson 2003). These findings suggest that recidivism and pretrial misconduct may be more likely among DV defendants under 40. However, the findings do not enable us to make predictions for DV defendants under 25. Research on whether a defendantâ&#x20AC;&#x2122;s age influences the processing of DV cases yields mixed results. Prior CJA research shows that age had inconsistent effects on the likelihood of conviction or incarceration in DV cases in New York City, and had no effect on the length of jail sentences (Peterson 2001). These findings do not enable us to make any clear predictions about the processing of young malesâ&#x20AC;&#x2122; DV cases. Studies that describe the population of young DV offenders are rare. We have located two studies of specialized courts that focus on young DV offenders and provide descriptive information. One of these studies is an evaluation of the Brooklyn Youthful Offender Domestic Violence Court (Cissner 2005). This court was designed to handle cases of intimate partner violence committed by teenage offenders. The evaluation includes information about 279 defendants with 360 cases during the period from January 2004 to March 2005. Defendants in the Youthful Offender DV Court were between the ages of 16 and 19, and the average age was 18. About 88% of the defendants were male, so most of the findings can be generalized to the population of young male DV offenders. In the Brooklyn Youthful Offender DV Court, about one-third of the offenders had a non-sealed prior arrest, and the average number of prior arrests was 1.8 (Cissner 2005). Almost half of the defendants were enrolled in school at the time of the arrest, and only one in seven defendants had completed high school. One quarter of the

3 defendants reported drug use, and one third reported alcohol use. Almost half of the defendants had a child in common with the complaining witness. About two thirds of the defendants were charged with assault, and one sixth were charged with criminal contempt (violating an order of protection). Data on the processing of cases in the Youthful Offender DV Court shows that 42% of the cases were adjourned in contemplation of dismissal, usually with the condition that the defendant successfully complete a teen-oriented batterer intervention program (Cissner 2005). About 28% of the cases were disposed with a guilty plea, including 11% that were sentenced to jail. The median jail time was 30 days. Bench warrants were issued in about 10% of the cases. Data on pretrial re-arrests or post-disposition re-arrests are not yet available. A study of the Santa Clara Juvenile Domestic and Family Violence Court also provides some data on teenage offenders (Sagatun-Edwards et al., 2003). Unlike the Brooklyn court, this court handles cases of family violence (e.g., violence against parents, siblings or other relatives) as well as partner violence, and it hears only cases of offenders under the age of 18 who are processed as juvenile (rather than adult) offenders. The evaluation provides information about 127 juveniles processed between April 1999 and June 2001. The median age of the offenders was 16, and about three quarters of the offenders were male. About 60% of the juveniles in the Santa Clara study had a record of prior juvenile delinquency. One fifth of the juveniles were diagnosed with a mental illness (most of those with this diagnosis were charged with family violence rather than partner violence). Over 60% had been abused as a child, and over half the juveniles had parents who had a DV history. About half of the parents had a criminal record and half had a substance abuse history. About two thirds of the juveniles were charged with partner violence, and one third were charged with family violence. Most of the offenders were ordered to participate in a batterer intervention program and a substance abuse program. Recidivism data indicated that over one half had committed new offenses, including over one quarter who had committed new domestic violence or family violence offenses. A control group had a similar rate of new offenses, so these results indicate that the court had no impact on recidivism. The data on the Brooklyn Youthful Offender DV Court and the Santa Clara Juvenile Domestic and Family Violence Court provide valuable information about an understudied group. Nevertheless, both court studies reported information about male and female offenders together. Furthermore, the sample sizes were relatively small. Finally, the information was limited to teenage offenders. More information is needed if we are to understand the dynamics of domestic violence among young male DV offenders.

4 The lack of information on young male DV offenders, and the special relevance of this group for prevention efforts, are important reasons for undertaking the current study. To aid in prevention efforts, information is needed specifically about young male DV offenders. Female offenders differ significantly from male offenders, since they are more often involved in dual arrests (where both parties are arrested in a domestic violence incident). Female DV offenders also differ from male DV offenders in terms of their criminal histories and their patterns of offending and re-offending. Furthermore, information is needed for a larger sample of young male offenders involved in all types of domestic violence. Finally, early intervention efforts might focus on both teens and young adults, so research is needed on young male DV offenders that includes both groups. B. Research Plan The current study is designed to address three research questions: 1) What are the characteristics of young male DV offenders? 2) How do young male DV offenders compare to older male DV offenders and to young male Non-DV offenders? 3) What factors influence pretrial misconduct among young male DV offenders? These questions will be addressed through an analysis of data on New York City arrestees drawn from the New York City Criminal Justice Agency, Inc. database. As described in Chapter 2, the dataset includes offenders arrested in the third quarter of 2005. The dataset includes information from the CJA interview, the arrest report, and case processing information. Chapter 3 provides an overview of the characteristics of young male DV offenders, and compares them to older offenders and Non-DV offenders. Chapter 4 examines models that attempt to account for differences between young male DV offenders and other offenders. Chapter 5 develops models to examine the predictors of pretrial misconduct among young male DV offenders. The report concludes with a summary of the findings, and a discussion of their implications.

5 II. METHODOLOGY A. Overview of the CJA Database and the Third Quarter 2005 Dataset The data for this study were drawn primarily from the CJA database. This database contains information about the arrest, case processing, and case outcomes of most New York City arrestees. The CJA database includes data from three sources: CJAâ&#x20AC;&#x2122;s pre-arraignment interview,2 the New York City Police Departmentâ&#x20AC;&#x2122;s On-Line Booking System (OLBS) database, and the New York State Office of Court Administration (OCA).3 Information concerning demographic characteristics and the community ties of the defendants is taken from the CJA pre-arraignment interview. Information about the arrests is based on the OLBS data. Detailed Criminal Court and Supreme Court processing and outcome data on each of the arrests are drawn from the OCA data. This report is based on analyses of the Third Quarter 2005 Dataset, which includes data collected on a three-month cohort of arrests made from July 1, 2005 to September 30, 2005. The dataset includes information on 80,123 arrests. The district attorney elected to bring charges and assigned a docket number for 74,416 of these arrests. The remaining 5,707 arrests were declined for prosecution by the district attorney, and were not docketed. For cases that had multiple dockets, case-processing information in this study is based on the docket that had the most severe arraignment charge (based on Penal Law severity) in Criminal Court.4 When the most severe arraignment charges on two or more dockets are of equal Penal Law severity, the top charge is determined according to guidelines developed by OCA. These guidelines provide a consistent set of rules for determining which of two arraignment charges of equal severity will be identified as the top arraignment charge.

CJA conducts pre-arraignment interviews to measure the defendantâ&#x20AC;&#x2122;s community ties, which serve as the basis for making a recommendation as to whether or not the defendant should be released on recognizance at his or her first court appearance. Defendants who are arrested on a bench warrant, given a Desk Appearance Ticket (DAT), or who are held for arraignment on prostitution charges in the downtown Manhattan Criminal Court are not interviewed by CJA. CJA collects arrest and Criminal Court information for all arrestees, and arrestees were included in the Third Quarter 2005 Dataset whether or not they were interviewed by CJA. 3 DCJS, OCA, and the NYPD are not responsible for the methods or conclusions of this report. 4 New York State Penal Law categorizes most offenses according to their severity. The most serious crimes are A felonies, followed by felonies classified as being of severity B through E. Misdemeanors are less severe than felonies, and are classified as A or B misdemeanors or Unclassified misdemeanors. (A misdemeanors are more severe than B misdemeanors, and Unclassified misdemeanors are less severe than B misdemeanors.) Violations are less severe than misdemeanors, and are not considered crimes, although they can result in jail sentences. No distinctions of severity are made within the category of violations.

6 Brooklyn, Manhattan, Queens and Staten Island have a two-tiered court system for handling criminal cases. In these four counties, the Criminal Courts only have trial jurisdiction over cases having a most serious charge of misdemeanor or lesser severity. Most defendants charged with felonies are first arraigned in Criminal Court. Cases sustained at the felony level must be brought for prosecution in Supreme Court. In felony cases where the DA decides not to prosecute the case in Supreme Court (or the Grand Jury fails to return an indictment), the case may be disposed in Criminal Court by dismissal or by a plea to a reduced charge less severe than a felony, or by a transfer to another court’s jurisdiction (e.g., Family Court).5 In the Bronx, as in the other boroughs, all criminal cases are first arraigned in Criminal Court. However, all criminal cases that are not disposed at arraignment, whether they are felony or misdemeanor cases, are transferred to Supreme Court in the Bronx for subsequent appearances. As a result, Supreme Court cases in the Bronx include not only cases sustained as felonies, but also all other criminal cases that were not disposed at arraignment. In other counties, the cases not sustained as felonies would have remained in Criminal Court. To compare cases from the Bronx to cases from other counties, we attempted to determine whether cases processed in Bronx Supreme Court would have been processed in Criminal Court had they been processed in another county. We classified certain cases that were transferred to Bronx Supreme Court as “criminal-court equivalents.” Bronx Supreme Court cases that were not sustained as felonies were considered “criminal-court equivalents,” while those that were sustained as felonies were considered to be “true” Supreme Court cases. The cases selected for inclusion in the analyses in this report include only cases that reached a final disposition in Criminal Court, or that were classified as “criminal-court equivalents” in the Bronx. In the remainder of this report, we refer to these cases as Criminal Court cases, whether they were disposed in Criminal Court or were criminal-court equivalent cases disposed in Bronx Supreme Court. The overwhelming majority (about 94%) of domestic violence cases citywide were disposed in Criminal Court. The Third Quarter 2005 Dataset includes case processing information in Criminal Court through final disposition (and sentencing, if there was a conviction), or until March 5, 2007. Case processing information in Supreme Court is included through final disposition (and sentencing, if there was a conviction), or until April 17, 2007. Information about any final dispositions in Criminal Court or Supreme Court beyond these cutoff dates was not included in the dataset. B. Identifying Domestic Violence Cases Social scientific and legal definitions of domestic violence have changed over the last 30 years (Peterson 2001). In New York State, the statutory definition of domestic 5

The Family Courts have concurrent jurisdiction over certain domestic violence cases (Aldrich and Domonkos 2000). Some DV cases are heard only in Criminal Court, some are heard in both Criminal Court and Family Court, and others are heard only in Family Court. We do not have access to data on DV cases that are heard only in Family Court, and our report draws no conclusions about these cases.

7 violence approximates what has come to be known in the social scientific literature as “family violence.” During the time period covered in the current study (third quarter of 2005) New York State’s Criminal Procedure Law (CPL) §530.11, defined family offenses as offenses committed against a member of the same family or household, where “family or household” was defined as: (1) persons related by consanguinity or affinity; (2) persons legally married to each other; (3) persons who were formerly married, and; (4) persons who have a child in common, whether or not they have ever been married or lived together. During the time period covered in the current study (third quarter of 2005), New York State’s statutory definition of domestic violence excluded unmarried partners, unless they had a child in common. However, the New York City Police Department (NYPD) operated with an expanded definition of domestic violence that included individuals who were not married, but who were cohabiting or had previously lived together. This NYPD definition of “family” expanded on New York State law by including “common-law” marriages, same-sex couples, and registered New York City domestic partners (NYPD 2000). By citywide agreement, the DA’s offices and the Criminal Courts in all five boroughs also used this expanded definition to identify DV cases, whether or not the relationship between the victim and defendant met the New York State statutory requirements contained in CPL §530.11. New state legislation expanded the definition of “family or household” in 2008.6 To identify domestic violence cases, assistant district attorneys (ADAs) use information collected by the police about the relationship between the victim and the defendant, if any. The ADAs also often ask victims about their relationship with the defendant. When this information indicates that the offender-victim relationship meets the NYPD expanded definition of domestic violence, the case is identified as a DV case. In all five boroughs DV case files are then given beige “backs” (special color-coded back sheets) to distinguish them from other case files. At Criminal Court arraignment, court clerks assign an arraignment hearing type of “DV” to domestic violence cases, and this designation is entered in OCA’s computerized court records. At the time the defendants in the third quarter of 2005 were arrested, cases identified at arraignment as DV cases were processed in different ways depending on the borough. In all boroughs, most cases with a DV hearing type were sent to a specialized domestic violence part for post-arraignment appearances. However, there were exceptions. In Brooklyn and the Bronx, cases with a DV hearing type that involved physical or sexual abuse of children or other types of non-intimate partner violence (e.g., violence between siblings) were not sent to the specialized domestic violence parts. Finally, some cases that did not have a DV hearing type at arraignment were 6

In 2008, New York State amended the statutory definition of “family or household” to include current and former intimate partners, whether or not they had ever lived together. This statutory change incorporates all the relationships formerly included in NYPD’s expanded definition of family or household. It also includes other relationships, such as couples who are dating or have dated and have never lived together. The new legislation became effective on July 21, 2008.

8 also sent to the specialized DV parts, presumably because information that these cases involved domestic violence became available only after arraignment. In this study, we identified “domestic violence” cases by relying on the court’s identification of these cases. We used information about both hearing type and court part, since not all DV cases were assigned a DV hearing type. We identified cases as domestic violence cases if the Office of Court Administration reported that: (1) the case had a domestic violence hearing type at Criminal Court arraignment, and/or, (2) the case had one or more appearances in a specialized domestic violence part.7 Using these criteria, we identified 6,298 docketed Criminal Court domestic violence cases in the Third Quarter 2005 Dataset, about 8.5% of the total sample of 74,416 cases disposed in Criminal Court. About 66% of the 6,298 DV cases had both a DV hearing type at arraignment and at least one appearance in a specialized DV part. An additional 15% of cases had a DV hearing type at arraignment but no appearance in a specialized DV part. These included cases that were disposed at arraignment as well as DV cases that were sent to Non-DV parts. Finally, 19% of the cases had at least one appearance in a specialized domestic violence part, but did not have a domestic violence hearing type at Criminal Court arraignment. For arrests that were declined for prosecution, we used NYPD’s information about the nature of the offender-victim relationship to identify DV cases. There were 1,125 cases declined for prosecution in the Third Quarter 2005 Dataset where the offender-victim victim relationship was a relationship included in the expanded definition of domestic violence described above. These 1,125 cases represented about 20% of the total sample of 5,707 cases declined for prosecution. As noted in a previous report (Peterson 2003), the measure identifying DV cases does have some limitations. First, there may be instances where a docketed DV case was not identified as such in court records (i.e., it did not receive a DV hearing type at Criminal Court arraignment and did not appear in a specialized domestic violence part) or where a DV case declined for prosecution did not have information indicating that the offender-victim relationship was a relationship included in the expanded definition of domestic violence. The measure did not identify these as DV cases but instead categorized them as Non-DV cases. In the current study, this limitation affected our analyses in two ways. First, it reduced the sample size of DV cases on which we report. Nevertheless, we have an adequate sample size for our analyses, and were able to 7

In the third quarter of 2005, the specialized domestic violence Criminal Court parts were DV1 and DV2 in Brooklyn, AP-4, Q-IDV, and T-DV in Queens, D and JURY13 (for DV jury trials) in Manhattan and AP2-DV in Staten Island. (Although AP2-DV in Staten Island was identified as a separate court part in our data, it was actually a specialized DV calendar. DV cases on this calendar were heard in an all-purpose part two days a week. We identified cases as DV cases if they had one or more appearances on this calendar.) The specialized domestic violence Supreme Court Parts were DV, 4 and IDV in Brooklyn, K4 and Q-IDV in Queens, R-IDV in Staten Island, and 26, DV, and IDV-SC in the Bronx. There was no specialized domestic violence Supreme Court in Manhattan.

9 draw valid conclusions about DV cases from our sample. Second, we slightly overestimated the number of Non-DV cases in the sample. However, the number of DV cases misidentified as Non-DV cases is likely to be a very small proportion of the total number of Non-DV cases. C. Identifying Re-arrests for Domestic Violence Offenses We measured recidivism in this study by examining re-arrests for new offenses during the pretrial period (i.e., between arrest and case disposition). We classify each re-arrest as a DV re-arrest or a Non-DV re-arrest. Our analyses use three measures of pretrial re-arrest. One measure indicates whether the defendant had at least one rearrest during the pretrial period for any new offense. Another measure indicates whether the defendant had at least one re-arrest during the pretrial period for a new DV offense. A third measure indicates whether the defendant had at least one re-arrest during the pretrial period for a new Non-DV offense. Unfortunately, re-arrest rates are likely to underestimate recidivism. New DV offenses may not lead to re-arrest, since many victims do not call the police when a new offense occurs, and police may not make an arrest even when they are called. To overcome the limitations of re-arrest, some studies measure recidivism using interviews with the victim. Interviewers can learn about incidents that did not result in calls to the police and the re-arrest of the defendant. Rates of recidivism based on victim interviews are generally higher than rates based on re-arrest. Victim interviews also have weaknesses, however. It is often very difficult to reach victims and to complete interviews with them. Furthermore, measures based on victim interviews ignore the possibility that the defendant has re-offended with a new victim. Because both types of data have strengths and weaknesses, we would have preferred to measure recidivism using both victim interviews and re-arrest data. We used re-arrest data for practical reasonsâ&#x20AC;&#x201D;it was the only measure available to us. Although re-arrest may underestimate recidivism, it has two advantages over victim interviews. Data are potentially available for all defendants, not just those for whom victim interviews were completed. In addition, it measures recidivism against new victims as well as against the same victim. One study of domestic violence incidents in New York State found that limiting a measure of recidivism to new incidents with the same victim reduced estimates of recidivism by 15% to 20% (Frisch et al. 2001). This suggests that a significant number of domestic violence defendants move on to new victims. Our measures of re-arrest also are affected by the problems described above regarding the identification of DV arrests. To the extent that the courts fail to identify DV arrests as such, our measures of re-arrest underestimated the actual number of DV rearrests and overestimated the actual number of Non-DV re-arrests. To create our re-arrest measures, we collected data on all re-arrests and then determined whether each re-arrest was for a DV case or a Non-DV case. This determination was made using the criteria described in Section II above. That is, we counted a re-arrest as a

10 domestic violence case if either: (1) the case had a DV hearing type at arraignment, or (2) the case had any appearances in a specialized DV part. For re-arrests that were declined for prosecution, we used NYPD’s information about the nature of the offendervictim relationship to identify DV cases. D. Selection of the Crimes Against Persons and Property Subsample The current study used data from the Crimes Against Persons and Property (CAPP) Subsample described in a previous report (Peterson 2003).8 This subsample of cases was selected so that we could examine DV cases and comparable Non-DV cases across a wide range of charges. We began by selecting cases for this subsample where there was an alleged attempt to cause injury or where an overt threat of injury was made (Weis 1989). We used the most severe arraignment charge (based on Penal Law severity) to determine the nature of the offense, since this charge determines how the case is handled in the court system. We did not use the most severe arrest charge, which reflects charging decisions made by the police. We initially selected all cases that had a top (i.e., most severe) arraignment charge from any of the following New York State Penal Law articles: PL 120 (Assault), PL 130 (Sex Offense), PL 160 (Robbery), PL 260 (Crimes Against Children), or PL 265 (Weapons). Unfortunately, we were not able to include cases disposed in Criminal Court that had top arraignment charges from PL 125 (Homicide), PL 150 (Arson), and PL 135 (Kidnapping).9 Cases charged with offenses in these Penal Law articles were excluded since there were too few domestic violence cases with top arraignment charges in each of these Penal Law articles for reliable multivariate analysis. Recognizing that domestic violence often includes offenses that result in financial and psychological harm, rather than just physical harm, we also selected cases if they had a top arraignment charge from one of the following Penal Law articles: PL 140 (Burglary), PL 145 (Criminal Mischief), PL 155 (Larceny), PL 205 (Escape and Resisting Arrest), PL 215 (Criminal Contempt),10 and PL 240 (Public Order Offenses). Within each Penal Law article, we selected only those cases that had charges that could plausibly include elements comparable to those found in domestic violence cases. For example, cases with a top arraignment charge of Assault in the Third Degree (PL §120.00) were included in the subsample. However, cases of Gang Assault in the First Degree (PL §120.07) or Gang Assault in the Second Degree (PL §120.06) were excluded, since it is unlikely that a domestic violence case would include a gang assault charge. Similarly, in PL 240, prostitution charges (PL §240.37) were excluded. 8

In one of our reports (Peterson 2001), we also discussed results for an Assaults Subsample, which included all cases where the top arraignment charge was assault (PL 120). We used the Assaults Subsample to provide a more focused comparison of DV cases to similar Non-DV cases. In this study, we focus primarily on DV cases, and make only a few comparisons of DV to Non-DV cases. We therefore use only the Crimes Against Persons and Property Subsample for our analyses. This subsample includes information about the full range of DV offenses, including assaults as well as violations of orders of protection, crimes against children, etc. 9 Most, but not all, of these cases were sustained as felonies and disposed in Supreme Court. 10 Penal Law article 215 includes violations of orders of protection.

11 We then narrowed the subsamples further to identify an appropriate group of cases for the analysis. First, we limited the sample to Summary Arrests (i.e., cases in which the defendant was held in custody pending Criminal Court arraignment), excluding cases where the defendant was issued a Desk Appearance Ticket (DAT) and released by the arresting officer. DATâ&#x20AC;&#x2122;s are rarely issued in DV cases. We also excluded cases with juvenile defendants (under age 16)11, cases that did not reach a final disposition by the cutoff date, and cases that were missing data on the defendantâ&#x20AC;&#x2122;s criminal history or sex. We also excluded cases that were disposed in Criminal Court on Vehicle and Traffic Law (VTL) or Administrative Code (AC) charges. After these exclusions, the Crimes Against Persons and Property Subsample contained 26,323 cases, including 6,579 DV cases. The number of docketed cases was 23,775, including 5,559 DV cases. E. Using a Defendant-Based Data File Some of the analyses presented in this study use a defendant-based data file that includes information on all defendants who were arrested in the third quarter of 2005. While the vast majority of defendants had only one case in the case-based data file, some defendants were arrested two or more times. For most defendants who had multiple arrests, we included information in the defendant-based data file about the first arrest, i.e., the arrest that occurred earliest. However, for defendants whose first arrest was not for a DV case but who had a subsequent DV arrest, we selected the first DV arrest, not the first arrest. This procedure enabled us to identify all the defendants who had at least one DV arrest. The defendant-based file contained 24,520 cases, including 6,200 DV cases. The number of docketed cases in the defendant-based file was 22,116 cases, including 5,204 DV cases. F. Plan of Analysis Chapter 3 uses the combined Third Quarter 2005 Dataset to describe the characteristics of young male DV defendants and to compare them to older male DV defendants and to young male Non-DV defendants. Chapter 4 examines whether, and to what extent, differences between young male DV offenders and other offenders can be explained. Chapter 5 examines the factors that influence pretrial misconduct among young male DV offenders. Chapter 6 summarizes the results of the study.

With only a few exceptions, the cases of juvenile offenders (under age 16) are heard in Family Court, not Criminal Court. We have no access to data on case processing in Family Court.

[This page intentionally left blank]

13 III. A PORTRAIT OF YOUNG MALE DOMESTIC VIOLENCE OFFENDERS This chapter describes the characteristics of young male DV offenders in New York City. To place the findings in context, young male DV offenders are also compared to older male DV offenders and to young male Non-DV offenders. Although data are presented for older male Non-DV offenders as well, they are discussed only in a few instances. The data presented here are drawn from the case-based Crimes Against Persons and Property Subsample described in Chapter 2. The analyses are restricted to male offenders (N = 21,425) in the subsample. For purposes of comparison, male offenders were divided into 4 categories based on age and type of offense. Offenders age 16-24 were classified as “young” and offenders 25 and over were classified as “older.” Each age group was further divided into those who were charged in the third quarter of 2005 with a DV offense and those who were charged with a Non-DV offense (as defined in Chapter 2). The following table lists the number of cases in each category. DISTRIBUTION OF CASES IN CASE-BASED FILE BY AGE AND TYPE OF OFFENSE Category Number of Cases Percent of Cases Young Male DV Offenders 1,280 6% Older Male DV Offenders 4,144 19% Young Male Non-DV Offenders 6,384 30% Older Male Non-DV Offenders 9,617 45% Total 21,425 100% To determine whether differences between young male DV offenders and other offenders are real differences, or are due to chance alone, we use tests of statistical significance. Tests of Statistical Significance Statistical significance tests assess the likelihood that the percentage differences observed in the sample could have occurred by chance alone. The tests take into account the size of the sample and the magnitude of the differences observed. Larger percentage differences and percentage differences based on larger samples are more likely to be statistically significant. In this report, following standard convention, significance levels less than .05 were considered statistically significant. This means that the statistically significant differences found in this study had less than a 5% chance of being due to chance alone. Stronger significance levels (less than .01 and less than .001) indicate that the differences found were even less likely to be due to chance alone (less than a 1%, or a 0.1% chance, respectively).

14 A. Offender and Case Characteristics We begin by examining differences between young male DV offenders and other offenders in terms of offender and case characteristics. These characteristics will be used in subsequent sections of the report as predictors of case outcomes and pretrial misconduct. The discussion highlights areas where there were differences between young male DV offenders and other offenders. Offender’s Demographic Characteristics Two demographic variables were included in our analyses: ethnicity and age. Young male DV offenders were slightly more likely to be Non-Hispanic Black or Hispanic, and slightly less likely to be Non-Hispanic White, than older male DV offenders (see Table 3-1). However, the ethnicity of young male DV offenders was similar to that of young male Non-DV offenders. TABLE 3-1 OFFENDER’S DEMOGRAPHIC CHARACTERISTICS BY AGE FOR DV AND NON-DV OFFENDERS OFFENDER’S DEMOGRAPHIC CHARACTERISTICS ETHNICITY Non-Hispanic Black Non-Hispanic White Hispanic Other, non-Hispanic Total, all cases (N of cases)

Young Male DV Offenders 48% 6 38 8 100% (1,280)

Older Male DV Offenders

Young Male Non-DV Offenders

Older Male Non-DV Offenders

***

43% 14 33 11 100%1 (4,144)

48% 9 34 9 100% (6,384)

45% 13 30 11 100%1 (9,617)

***

AGE CATEGORY Age 16-20 Age 21-24 Age 25-29 Age 30-39 Age 40 and older Total, all cases (N of cases)

41% 59 0 0 0 100% (1,280)

0% 0 24 40 36 100% (4,144)

60% 40 0 0 0 100% (6,384)

0% 0 23 35 43 100%1 (9,617)

Mean Age

20.9

37.4***

19.8***

38.2***

Percentages do not sum to 100% due to rounding. Differences between this group and young male DV offenders are statistically significant at p < .01. *** Differences between this group and young male DV offenders are statistically significant at p < .001. **

15 We also examined the age distribution of offenders within each age group. Almost 60% of young male DV offenders were age 21-24 (vs. 16-20). The distribution of young male Non-DV offenders was reversed—60% were younger, age 16-20, and only 40% were age 21-24. This suggests that among young men under 25, the likelihood of an arrest for a DV offense increases as the men approach 25. At the youngest ages (under 21), Non-DV offenses are more likely. We also examined the distribution of age within the older age group. Among men age 25 and over, DV offenders tended to be younger than Non-DV offenders. For example, only 36% of DV offenders were 40 or older, compared to 43% of Non-DV offenders. Taken together, these findings suggest that DV offending is more concentrated in the ages from 21 to 39, whereas Non-DV offending is more widely distributed, and is more common than DV offending at both younger and older ages. Offender-Victim Relationship There were statistically significant differences between young male and older male DV offenders in the type of offender-victim relationship (see Table 3-2). Young male DV offenders were less likely to be married to the victim, and more likely to be in boyfriend-girlfriend and “other” family relationships (usually sibling or child-parent12), than older male DV offenders. These differences are not particularly surprising, since they reflect the age differences between the two groups. Young men are less likely to be married than older men. Young men are also probably more likely to be living with parents and siblings, and therefore more likely to commit offenses against these family members. The youngest DV offenders (those aged 16-20) were most likely to commit offenses against a parent or sibling (data not shown). One possibility, which we were not able to explore with our data, is that such offenses may be a precursor of later domestic violence against intimate partners. Offender’s Criminal History Young male DV offenders were slightly less likely (57% vs. 60%) than older male DV offenders to have prior criminal convictions or open cases at the time of their arrest (Table 3-3). They were much less likely to have prior misdemeanor convictions (24% vs. 42%), prior felony convictions (18% vs. 33%), and any prior bench warrants (21% vs. 31%). These patterns are not surprising, since young offenders would be expected to have a less extensive criminal history than older offenders. Young male DV offenders had more extensive criminal histories than young male Non-DV offenders. Young male DV offenders were more likely to have prior criminal convictions or open cases than young male Non-DV offenders (57% vs. 50%). They also were more likely to have prior misdemeanor convictions (24% vs. 20%), prior 12

Because we did not have information about the age of the victim, we were unable to identify cases of child abuse or elder abuse using the NYPD data. For example, when the offender was identified as the parent of the victim, we do not know if the child was a minor under the age of 18, or an adult. Similarly, when the offender was identified as the son or daughter of the victim, we do not know if the victim was elderly or not.

16 felony convictions (18% vs. 16%) and any prior bench warrants (21% vs. 19%). One possible explanation of these differences is that they may reflect the age differences between the two groups of young males. As noted above, young male DV offenders were older than young male Non-DV offenders, and thus had more time to accumulate TABLE 3-2 OFFENDER-VICTIM RELATIONSHIP BY AGE FOR DV OFFENDERS OFFENDER-VICTIM RELATIONSHIP*** Boyfriend-Girlfriend Married or Divorced Common-law Spouse Other family relationship1 Missing Total, all cases (N of cases) 1 ***

Young Male DV Offenders

Older Male DV Offenders

37% 7 20

29% 29 24

100% (1,280)

100% (4,144)

Includes parent-child, sibling, grandparent-grandchild and other family relationships. Differences between young male DV offenders and older male DV offenders were statistically significant at p < .001.

a criminal history. A closer examination, however, reveals that at every age, starting at age 16, young male DV offenders have a more extensive criminal history than young male Non-DV offenders (data not shown). This suggests that young male DV offenders are more active early in their criminal careers, and that age differences are not the primary reason for their more extensive criminal history. We also examined each offenderâ&#x20AC;&#x2122;s arrest history during the 2 years prior to his arrest in the third quarter of 2005. We included both docketed arrests and arrests that were declined for prosecution during that 2-year period.13 Unlike the measures discussed previously (prior criminal convictions or open cases), these measures include disposed cases that did not result in a criminal conviction. They therefore serve as a measure of relatively recent arrest activity. Over three fifths of young male DV offenders (62%) were arrested in the 2 years prior to their arrest in the third quarter of 2005. This is considerably higher than the percentage of older male DV offenders (47%) and somewhat higher than the percentage of young male Non-DV offenders (57%). 13

We included arrests that were declined for prosecution because District Attorneys offices in some boroughs routinely declined to prosecute certain types of cases, including a significant proportion of domestic violence cases.

17 TABLE 3-3 OFFENDER’S CRIMINAL HISTORY BY AGE FOR DV AND NON-DV OFFENDERS OFFENDER’S CRIMINAL HISTORY

Young Male DV Offenders

Older Male DV Offenders

Young Male Non-DV Offenders

Older Male Non-DV Offenders

PRIOR CRIMINAL HISTORY Any prior criminal convictions or open cases

57%

60%***

50%***

68%***

Any prior misdemeanor convictions

24%

42%***

20%**

56%***

Any prior felony convictions

18%

33%***

16%

45%***

Any prior bench warrants

21%

31%***

19%*

45%***

(N of cases)

(1,280)

(4,144)

(6,384)

(9,617)

ARREST HISTORY IN PRIOR 2 YEARS Any arrests in prior 2 years

62%

47%***

Any DV arrests in prior 2 years

29%

27%

Any Non-DV arrests in prior 2 years

51%

32%***

(N of cases)

(1,280)

(4,144)

57%**

57%***

6%***

8%***

56%**

55%*

(6,384)

(9,617)

Differences between this group and young male DV offenders are statistically significant at p < .05. ** Differences between this group and young male DV offenders are statistically significant at p < .01. *** Differences between this group and young male DV offenders are statistically significant at p < .001.

A closer look at the type of prior arrests helps us to understand the reasons for these differences. We examined whether each offender had any DV arrests in the prior 2 years, as well as whether he had any Non-DV arrests during that period (some offenders had both types of arrests). Male DV offenders, whether young or old, were

18 about equally likely to have DV arrests in the prior 2 years (29% and 27%, respectively). However, young male DV offenders were much more likely than older male DV offenders to have Non-DV arrests in the prior 2 years (51% vs. 32%). This indicates that the higher arrest rate in the prior 2 years for younger vs. older male DV offenders is due to their greater likelihood of being arrested for Non-DV offenses. The comparisons with young male Non-DV offenders show a different pattern. Young male DV offenders were much more likely than young male Non-DV offenders to have been arrested for DV offenses in the prior 2 years (29% vs. 6%, respectively). Furthermore, young male DV offenders were actually slightly less likely than young male Non-DV offenders to have been arrested for Non-DV offenses (51% vs. 56%). This indicates that the higher arrest rate in the prior 2 years for young male DV vs. NonDV offenders is due to their greater likelihood of being arrested for DV offenses. For DV offenders who had a prior DV arrest, we also examined the outcome of that arrest (see Table 3-4). The most notable difference between young male DV offenders and older DV offenders is that the younger offenders were less likely to have been convicted and sentenced to jail (8% vs. 14%). Younger DV offenders were slightly more likely to have received an adjournment in contemplation of dismissal (12% vs. 8%). TABLE 3-4 OUTCOME OF PRIOR DV ARREST BY AGE FOR DV OFFENDERS1

OUTCOME OF PRIOR DV ARREST** Declined for Prosecution Dismissed ACD Convicted, no jail sentence Convicted, with jail sentence Not disposed Total, all cases (N of cases) 1

Young Male Older Male DV DV Offenders Offenders 10% 42 12 27 8 1 100% (370)

8% 43 8 25 14 2 100% (1,133)

Data presented only for cases in which the defendant had a domestic violence arrest in the prior two years. Differences between young male DV offenders and older male DV offenders were statistically significant at p < .01.

19 Community Ties Next, we consider several community ties variables (Table 3-5). Young male DV offenders had weaker community ties than older male DV offenders on only one of our measures. Young male DV offenders were slightly less likely to be engaged in full-time activity (employment, school, and/or training program) than older male DV offenders (58% vs. 62%). However, young male DV offenders had stronger community ties on two measures—they were considerably more likely to live with someone (73% vs. 57%) and to expect someone at arraignment (46% vs. 38%). Nearly equal proportions of young and older male DV offenders lived at their current address for more than one year (73% and 72%), had a telephone or cell phone (73% and 71%), and lived in the New York City area (98% and 97%). Young male DV offenders had weaker community ties than young male Non-DV offenders on two of our measures. They were less likely to be engaged in full-time activity (58% vs. 62%) and less likely to live at their current address for more than one year (73% vs. 77%). However, young male DV offenders had slightly stronger community ties on two measures—living with someone (73% vs. 70%) and living in the New York City area (98% vs. 95%). Young male DV and Non-DV offenders were about equally likely to expect someone at arraignment (46% and 44%) and to have a telephone or cell phone (73% and 72%). In addition to the information about the community ties items, we also examined the CJA release recommendation (see Table 3-5). This recommendation is based on several of the community ties items discussed above (full-time activity, telephone/cell phone, New York City area address, expect someone at arraignment) as well as on information about prior bench warrants and open cases. The release recommendation was coded in four categories: 1) recommended or moderate risk, 2) not recommended due to high risk for FTA, 3) no recommendation due to outstanding bench warrant and 4) other or missing. The “other” category includes defendants whose NYSID’s were unavailable or whose interviews were incomplete or conducted for information only (i.e., charged with homicide or attempted homicide). The “missing” category includes defendants who were not interviewed by CJA.14 Young male DV offenders were more likely to be recommended for release (63% vs. 55%) than older male DV offenders (see Table 3-5). However, young male DV offenders were about as likely to receive a recommendation for release as young male Non-DV offenders. This suggests that young offenders generally have stronger community ties.

CJA interviews about 96% of the defendants held for arraignment in New York City, as well as about 78% of offenders whose cases were declined for prosecution. Using information from the interviews, CJA assesses the strength of the defendant’s New York area community ties and provides a release recommendation to the court at the time of arraignment. For further information about the CJA interview and CJA’s release recommendation, see Siddiqi (2004).

20 TABLE 3-5 COMMUNITY TIES BY AGE FOR DV AND NON-DV OFFENDERS COMMUNITY TIES

Young Male DV Offenders

Older Male DV Offenders

Young Male Non-DV Offenders

Older Male Non-DV Offenders

COMMUNITY TIES ITEMS Engaged in full-time activity

58%

62%*

50%***

At current address more than one year

73%

72%

77%**

76%**

Lives with someone

73%

57%***

70%*

42%***

Expects someone at arraignment

46%

38%***

44%

34%***

Has a telephone or cell phone

73%

71%

72%

56%***

Lives in New York City area

98%

97%

95%***

91%***

(N of cases)

(1,280)

(4,144)

RELEASE RECOMMENDATION Recommended or moderate risk No recommendation: High Risk for FTA Open Bench Warrant At Time of Arrest Other or missing Total, all cases (N of cases) *

(6,384)

***

(9,617) ***

63%

55%

65%

37%

100%1 (1,280)

100% (4,144)

100% (6,384)

100% (9,617)

21 Case Filing Outcome The arrests of young male DV offenders were more likely to be declined for prosecution (DPâ&#x20AC;&#x2122;d) than the arrests of older male DV offenders (17% vs. 13%, see Table 3-6). The DP rate was even lower for young male Non-DV offenders (9%). DP rates vary by borough, and reflect case screening policies in each borough (see Peterson 2002 and Peterson and Dixon 2005 for a discussion of case screening policies). In the third quarter of 2005, only three of the five boroughs declined prosecution of significant numbers of DV cases: the Bronx, Brooklyn and Staten Island. A closer look at the data by borough reveals that the only borough where there is a difference in DP rates between young and older male DV offenders is Brooklyn (data not shown). The reason for this difference is not clear. It could be that young male DV offenders were less likely to have a prior DV history, or that injuries to the victim were less likely in the cases of young males. Because we do not have sufficient data to determine the reasons, the explanation for these differences remains unresolved. TABLE 3-6 CASE FILING OUTCOME BY AGE FOR DV AND NON-DV OFFENDERS CASE FILING OUTOME Declined for Prosecution (N of cases) ***

Young Male DV Offenders

Older Male DV Offenders

Young Male Non-DV Offenders

Older Male NonDV Offenders

17%

13%***

9%***

6%***

(1,280)

(4,144)

(6,384)

(9,617)

Differences between this group and young male DV offenders were statistically significant at p < .001.

Arrest and Arraignment Charge Characteristics The average number of arrest charges did not vary much, ranging from 1.7 to 1.9 for the four categories of offenders (see Table 3-7).

TABLE 3-7 ARREST AND ARRAIGNMENT CHARGE CHARACTERISTICS BY AGE FOR DV AND NON-DV OFFENDERS ARREST AND ARRAIGNMENT CHARGE CHARACTERISTICS MEAN NUMBER OF ARREST CHARGES (N of cases) ARRAIGNMENT CHARGE PENAL LAW ARTICLE1 Assault (PL 120) Criminal Contempt (PL 215) Harassment (PL 240) Crimes Against Children (PL 260) Other Total, all cases (N of cases)

Young Male DV Offenders

Older Male DV Offenders

Young Male Non-DV Offenders

Older Male Non-DV Offenders

1.75

1.69*

1.93***

1.87***

(1,280)

(4,144)

(6,384)

(9,617)

***

62%

24%

26%

100%2 (1,061)

100%2

100%

100%2

(3,599)

(5,782)

(9,006)

***

ARRAIGNMENT CHARGE SEVERITY1 Violation

Misdemeanor

87%

68%

78%

Felony

12%

24%

16%

Total, all cases

100%

100%2

(1,061)

(3,599)

(5,782)

(N of cases) 1

(9,006)

Data presented only for docketed cases. Percentages do not sum to 100% due to rounding. * Differences between this group and young male DV offenders are statistically significant at p < .05. *** Differences between this group and young male DV offenders are statistically significant at p < .001. 2

23 We next considered the arraignment charge Penal Law article. Note that the number of cases for the arraignment variables is lower than the number of cases discussed previously, because arraignment information is available only for docketed cases. (Demographic, criminal history and community ties measures are generally available for cases declined for prosecution as well as for docketed cases.) There was very little difference between younger and older male DV offenders in the type of arraignment charges filed (see Table 3-7). Most DV offenders were arraigned on assault charges (62%). Younger male DV offenders were slightly less likely to be charged with criminal contempt (violating an order of protection) or harassment, and slightly more likely to face “other” charges. Non-DV offenders (both young males and older males) were less likely to be charged with assault and more likely to be charged with “other” offenses. “Other” offenses were primarily robbery, larceny, and weapons charges. To measure the seriousness of the charges against the defendant, we categorized each case by severity of the top arraignment charge: whether the case was charged as a violation, misdemeanor or felony. There were no differences in charge severity between young male DV offenders and older male DV offenders (Table 3-7). However, young male Non-DV offenders were twice as likely to be charged with felonies as young male DV offenders (24% vs. 12%). It is worth noting that we are examining only those cases ultimately disposed as misdemeanors or violations in Criminal Court. Although very few DV cases were sustained as felonies and disposed in Supreme Court, as noted in Chapter 2, more Non-DV cases were disposed in Supreme Court. This suggests that the gap in the percentage arraigned on felony charges between young male Non-DV offenders and DV offenders of any age would be even larger if Supreme Court cases were included. Case Processing Characteristics DV cases were rarely disposed at arraignment (Table 3-8). This reflects the policy of District Attorney’s offices in New York City. ADA’s are generally not permitted to agree to dispositions at arraignment in DV cases.15 The policy is to gather more information on these cases and to keep them active as a means of preventing the defendant from committing further acts of domestic violence. In contrast, a considerable proportion of Non-DV cases were disposed at arraignment (34% for younger Non-DV offenders, and 40% for older Non-DV offenders).

The few DV cases that we identified as disposed at arraignment may have been disposed at that time due to error (e.g., failure to realize the case was a DV case) or due to special circumstances involved in the case. It is also possible that the court mistakenly identified these cases as DV cases (i.e., gave the cases a DV hearing type at arraignment), when they were not in fact DV cases.

TABLE 3-8 CASE PROCESSING CHARACTERISTICS BY AGE FOR DV AND NON-DV OFFENDERS CASE PROCESSING CHARACTERISTICS CASE DISPOSED AT ARRAIGNMENT1

Young Male DV Offenders 2%

Older Male DV Offenders

Young Male Non-DV Offenders

Older Male Non-DV Offenders

1%**

34%***

40%***

(3,599)

(5,782)

(9,006)

89%

87%**

77%***

(3,560)

(3,799)

(5,417)

(N of cases)

(1,061)

DEFENDANT EVER RELEASED2

90%

(N of cases)

(1,036)

RELEASED ON BAIL (vs. ROR)3

14%

16%

17%*

18%*

(N of cases)

(937)

(3,163)

(3,319)

(4,146)

MEAN NUMBER OF WEEKS FROM ARRAIGNMENT TO DISPOSITION2

(N of cases)

(1,036)

(3,560)

(3,799)

(5,417)

Data presented only for docketed cases. Data presented only for cases that were not disposed at arraignment. 3 Data presented only for cases in which the defendant was released. * Differences between this group and young male DV offenders are statistically significant at p < .05. ** Differences between this group and young male DV offenders are statistically significant at p < .01. *** Differences between this group and young male DV offenders are statistically significant at p < .001. 2

Defendants were released prior to case disposition in about 90% of DV cases not disposed at arraignment (including DV cases of both young males and older males). Release was also common among young male Non-DV offenders (87%). Young male DV offenders were slightly less likely to be released on bail (14%) than young male Non-DV offenders (17%). There were no differences in case processing time among the four groups. Cases that were not disposed at arraignment took an average of 16-17 weeks to reach a disposition (Table 3-8).

B. Case Dispositions, Sentence Outcomes, and Length of Jail Sentences We now turn our attention to reviewing data on three major case outcomes: case dispositions, sentence outcomes, and length of jail sentences. Young male DV offenders whose cases were docketed were only slightly less likely than older male DV offenders to have their cases end in conviction (35% vs. 37%) (see Figure 3-1). The two groups were equally likely to receive an ACD (10%) and over half of DV defendants in each group had their cases dismissed. Conviction rates were considerably higher for young male Non-DV offenders (50%) and even higher for older male Non-DV offenders (62%). Young male Non-DV offenders were more likely than older male Non-DV offenders to have their cases disposed with an ACD (22% vs. 14%). About 28% of young male Non-DV defendants had their cases dismissed, only slightly higher than the 24% of older male Non-DV defendants. CASE DISPOSITIONS IN CRIMINAL COURT In New York State, cases disposed in Criminal Court can result in one of several final dispositions: a plea of guilty, a finding of guilty after trial, an acquittal after trial, a dismissal, or an adjournment in contemplation of dismissal (ACD). In this report, convictions are defined to include pleas of guilty and findings of guilty after trial, including pleas or findings of guilty to a violation. (Although violations are not considered crimes under New York State Penal Law, they can result in a jail sentence.) Acquittals, dismissals and ACDs are categorized in this report as nonconvictions. Although ACDs are not convictions, they sometimes have conditions attached. In DV cases, an ACD may require completion of a batterer intervention program.1 Defendants who fail to comply may have their cases restored to the calendar. The use of ACDs is one of a variety of strategies employed to achieve the goal of monitoring defendants in DV cases. When these conditions are not fulfilled, or if the defendant is re-arrested within 6 months (12 months in the case of a family offense), the case can be re-opened and restored to the calendar for another, possibly more severe, disposition. However ACDs do not result in a conviction unless the defendant violates the conditions of the ACD and subsequently is returned to court and convicted. 1

In cases where the victim did not sign a supporting deposition, a defendant whose case was disposed with an ACD generally could not be required by the court to complete a batterer intervention program.

FIGURE 3-1 CASE DISPOSITIONS IN CRIMINAL COURT Prosecuted Cases YOUNG MALE DV OFFENDERS (N = 1,061)

OLDER MALE DV OFFENDERS (N = 3,599)

Convicted 35%

Convicted 37% Dismissed 53%

Dismissed 55% ACD 10%

ACD 10%

YOUNG MALE NON-DV OFFENDERS ***

OLDER MALE NON-DV OFFENDERS ***

(N=5,782)

(N = 9,006)

Dismissed 28%

Convicted 50%

Convicted 62% ACD 22%

***

Dismissed 24%

ACD 14%

Differences between this group and young male DV offenders are statistically significant at p < .001.

Next, we considered sentence outcomes for those cases that ended in a conviction (see Figure 3-2). There were no differences in sentence outcomes between young male DV offenders and older male DV offenders who were convicted. About 25% received a jail sentence, 70% received a conditional discharge and 5% received other sentences. (Jail sentences include both “time served” sentences and definite sentences, i.e., sentences for a specified number of days.) Convicted young male NonDV offenders also received similar sentences, except that they were somewhat less likely to receive a conditional discharge (63%) and more likely to receive an “other” sentence (11%). Older male Non-DV convicted defendants were more likely to receive a jail sentence than any other group—half were sentenced to jail.

FIGURE 3-2 SENTENCE OUTCOMES IN CRIMINAL COURT Convicted Defendants YOUNG MALE DV OFFENDERS

OLDER MALE DV OFFENDERS

(N=376)

(N=1,338)

Jail 25%

Other Sentence 5%

Conditional Discharge 70%

YOUNG MALE NON-DV OFFENDERS **

Other Sentence 5%

OLDER MALE NON-DV OFFENDERS

Jail 26%

(N = 5,617)

(N=2,900)

Other Sentence 11%

Conditional Discharge 70%

Conditional Discharge 63%

Jail 50%

Conditional Discharge 43%

Other Sentence 7%

Differences between this group and young male DV offenders are statistically significant at p < .01.

The third case outcome we examined was length of jail sentence. Jail sentences for young male DV offenders were shorter than those for older male DV offenders (39 days vs. 53 days; see Figure 3-3). Young male DV offendersâ&#x20AC;&#x2122; sentences were longer than those for young male Non-DV offenders (39 vs. 32 days), but since the sample sizes on which these averages are based are quite small, the difference is not statistically significant.

28 Measuring Length of Jail Sentences We measured length of sentence by determining how many days the defendant actually spent in jail for the sentence in the case. For “time served” sentences, we used information about release status to measure the amount of time the defendant was incarcerated between arrest and final disposition. For definite sentences, we used the number of days of jail imposed by the court. We then subtracted one-third of the length of the definite sentence to account for the time allowance that most defendants receive for “good behavior,” as provided by New York State Penal Law §70.30(4b). For example, a 30-day definite sentence was coded as 20 days in jail, after allowing for a 10-day reduction in the sentence. However, if the definite sentence was imposed after the defendant had already served more than two-thirds of the sentence, we used the actual time served as the sentence. For example, if a defendant who had been held for 25 days received a 30-day sentence, the sentence was coded as 25 days to indicate the actual time the defendant served.

FIGURE 3-3 LENGTH OF JAIL SENTENCE IN CRIMINAL COURT Defendants Sentenced to Jail Mean Number of Days Sentenced to Jail Incarcerated DV Defendants 75 50

25 0

YOUNG MALE DV DEFENDANTS (N = 96)

OLDER MALE DV DEFENDANTS (N = 337)

Mean Number of Days Sentenced to Jail Incarcerated NON-DV Defendants 75

YOUNG MALE NON-DV DEFENDANTS

OLDER MALE NON-DV DEFENDANTS

50 25 0

(N = 768)

(N = 2,806)

Differences between this group and young male DV offenders were statistically significant at p < .05.

C. Pretrial Misconduct For the analyses of pretrial misconduct, we used a defendant-based file, that is, a file which included only the first case for each defendant (see discussion in Chapter 2, Section E). As shown in the table below, each category is slightly smaller than it was before, because we are counting the number of defendants, not the number of cases. However, the distribution of cases among the categories is essentially unchanged.

CASES IN DEFENDANT-BASED FILE BY AGE AND TYPE OF OFFENSE Category Number of Cases Percent of Cases Young Male DV Offenders 1,195 6% Older Male DV Offenders 3,891 19% Young Male Non-DV Offenders 5,922 30% Older Male Non-DV Offenders 8,849 45% Total 19,857 100% We examined two types of pretrial misconductâ&#x20AC;&#x201D;failure to appear and pretrial rearrest for a new offense. Starting with failure to appear, we found that young male DV offenders have higher FTA rates than older male DV offenders (17% vs. 12%; see Figure 3-4). They also have a slightly higher rate than young male Non-DV offenders (17% vs. 14%). These differences are statistically significant.

FIGURE 3-4 FAILURE TO APPEAR RATES BY AGE FOR DV AND NON-DV DEFENDANTS WHO WERE EVER RELEASED

20%

17% 12%

***

10%

0% YOUNG MALE DV OFFENDERS

OLDER MALE DV OFFENDERS

(N = 868)

(N = 2,960)

20% 14%

14%

10%

0% YOUNG MALE NON-DV OFFENDERS (N = 3,082)

OLDER MALE NON-DV OFFENDERS (N = 3,928)

Differences between this group and young male DV offenders were statistically significant at p < .05. *** Differences between this group and young male DV offenders were statistically significant at p < .001.

Next, we examined pretrial re-arrests. About 28% of young male DV defendants were re-arrested during the pretrial period for any new offense (see Figure 3-5A). This rate was about 10 percentage points higher than the rate for older male DV defendants (18%), 5 percentage points higher than the rate for young male Non-DV offenders (23%) and 10 percentage points higher than the rate for older male Non-DV offenders (18%). These results indicate that young male DV defendants pose the highest risk of pretrial re-arrest of any of the four categories of defendants considered here. What types of offenses were defendants re-arrested for during the pretrial period? About 14% of young male DV defendants were re-arrested for new DV offenses, and 17% were re-arrested for new Non-DV offenses (see Figure 3-5B). (These categories overlapâ&#x20AC;&#x201D;about 3% were re-arrested for both DV and Non-DV

32 offenses.) Among older male DV defendants, 12% were re-arrested for new DV offenses and 8% were re-arrested for new Non-DV offenses. Young and older male DV defendants were about equally likely to be re-arrested for a new DV offense during the pretrial period (14% and 12%, respectively). However, young male DV defendants were much more likely to be re-arrested during the pretrial period for a new Non-DV offense (17% vs. 8%). Comparisons of young male DV offenders to young male Non-DV offenders reveal a different pattern. Young male DV offenders were much more likely than young male Non-DV offenders to commit a new DV offense (14% vs. 1%), and somewhat less likely to commit a new Non-DV offense (17% vs. 22%). Taken together, these findings show that young male DV defendants have a unique re-arrest pattern during the pretrial period. They are much more likely than older male DV defendants to be re-arrested for a Non-DV offense, and they are much more likely than young male Non-DV defendants to be re-arrested for a DV offense. The pattern of pretrial re-arrests is consistent with the pattern reported earlier regarding arrest histories in the prior two years. Young male DV offenders were more likely to have been arrested for Non-DV offenses in the prior 2 years than older male DV offenders. However, they were more likely than young male Non-DV offenders to have been arrested for DV offenses in the prior 2 years. These patterns suggest that many young male DV offenders are not â&#x20AC;&#x153;specializingâ&#x20AC;? in DV or Non-DV offensesâ&#x20AC;&#x201D;rather they appear to be engaging in both types of offending.

FIGURE 3-5A PRETRIAL RE-ARREST RATES FOR ANY NEW OFFENSES BY AGE FOR DV AND NON-DV DEFENDANTS WHO WERE EVER RELEASED 28%

30%

23%

***

18%

20%

18%

10% 0% YOUNG MALE DV OFFENDERS

OLDER MALE DV OFFENDERS

YOUNG MALE NON-DV OFFENDERS

(N = 2,960)

(N = 3,082)

(N = 868)

OLDER MALE NON-DV OFFENDERS (N = 3,928)

FIGURE 3-5B PRETRIAL RE-ARREST RATES FOR NEW DV AND NON-DV OFFENSES BY AGE FOR DV AND NON-DV DEFENDANTS WHO WERE EVER RELEASED

22%** 8%***

17%

25%

17%

20% 15%

14%

10%

12%

Pretrial Re-arrest Rates for New Non-DV Offenses

1%***

Pretrial Re-arrest Rates for New DV Offenses

YOUNG MALE DV OFFENDERS

OLDER MALE DV OFFENDERS

YOUNG MALE NON-DV OFFENDERS

OLDER MALE NON-DV OFFENDERS

(N = 868)

(N = 2,960)

(N = 3,082)

(N = 3,928)

Differences between this group and young male DV offenders were statistically significant at p < .01. *** Differences between this group and young male DV offenders were statistically significant at p < .001.

34 D. Summary and Discussion of Findings Many young male DV offenders are engaged in active criminal careers. Nearly three fifths had prior criminal convictions or open cases at the time of their arrest. Moreover, many of them were not arrested for an isolated incident of domestic violence. Nearly 30% had been arrested for a domestic violence incident in the prior two years. In terms of community ties, the typical young male DV offender lived in the New York City area, resided at his current address for over a year, currently lived with someone and had a telephone or cell phone. Nearly three fifths were engaged in full-time activity (employment, school, and/or training). The most common charge on the current case was assault, and over half the victims were intimate partners (usually girlfriends or common-law spouses, since few were married to the victim). Over one quarter of the victims were â&#x20AC;&#x153;otherâ&#x20AC;? family members, usually the offenderâ&#x20AC;&#x2122;s parent or sibling. About one sixth of the cases of young male DV offenders were declined for prosecution. Among the cases that were docketed, the most typical arraignment charge was misdemeanor assault. Almost all the offenders were released pending disposition, almost always by release on recognizance (ROR) rather than bail. Their cases took an average of 17 weeks to reach a disposition. Over half the cases ended in a dismissal, while one third ended in a conviction. The remainder was adjourned in contemplation of dismissal. Among those convicted, 70% received a conditional discharge, while 25% were sentenced to jail, and 5% received other sentences (e.g., fines). Those sentenced to jail spent an average of 39 days in jail (including pretrial detention time, if any). Young male DV offenders engaged in significant amounts of pretrial misconduct. About one sixth failed to appear for at least one court appearance, and over one quarter were re-arrested during the pretrial period, half of them for a new DV offense. Our review of offender and case characteristics, case outcomes, and pretrial misconduct revealed a number of differences between young male DV offenders and other offenders. Young male DV offenders were less likely to be married to their victims than older male DV offenders, and more likely to have victimized a girlfriend or other family member. Young male DV offenders generally had less extensive criminal histories than older male DV offenders. Given the age differences between the two groups, this is not a surprising finding, since younger offenders have not had as much time to accumulate a criminal history. What was somewhat surprising, however, is that young male DV offenders had a more extensive criminal history than young male Non-DV offenders. This difference did not appear to reflect age differences between these two groups; instead our analyses revealed that young male DV offenders had more active criminal careers than young male Non-DV offenders even at the youngest ages. Data on arrest history in the two years prior to the current case also suggest that young male DV offenders were more active in their criminal careers than other offenders. Young male DV offenders were more likely to have been arrested in the prior two years than older male DV offenders or young male Non-DV offenders. A closer

35 look at the prior arrests reveals that young male DV offenders were more likely than older male DV offenders to have been arrested for Non-DV offenses. On the other hand, young male DV offenders were more likely than young male Non-DV offenders to have been arrested for DV offenses. In their most recent prior DV arrest, young male DV offenders had somewhat more lenient outcomes than older male DV offenders. Young male DV offenders were slightly more likely to have had their prior arrest declined for prosecution or to have received an ACD than older male DV offenders. Moreover, they were less likely to be convicted and sentenced to jail. Differences among the groups in terms of community ties produced no clear pattern. In general, young male DV offenders were less likely than others to be engaged in full-time activity. On the other hand, they were more likely to live with someone and to expect someone at arraignment. Young male DV offenders were more likely to be recommended for ROR than older male DV offenders, but were not more likely to be recommended for ROR than young male Non-DV offenders. On their current case, young male DV offenders were more likely than older male DV offenders or young male Non-DV offenders to have the case declined for prosecution. When cases were prosecuted, arraignment charges differed slightly for young and older male DV offenders. Young male DV offenders were slightly less likely to be charged with criminal contempt, and slightly more likely to be charged with crimes against children. There were few case processing differences between young male DV offenders and older male DV offenders. Case dispositions were also similar for young male DV offenders and older male DV offenders. Slightly over one half of all prosecuted DV cases were dismissed, slightly over one third ended in a conviction, and 10% were adjourned in contemplation of dismissal (ACD). However, young male Non-DV offenders were more likely than either group of DV offenders to have their cases end in conviction or an ACD, and less likely to have their cases dismissed. Among those who were convicted, sentence outcomes were fairly similar for all three groups. About two thirds received a conditional discharge, one quarter received a jail sentence and the remainder received some other sentence, such as a fine. Among those who received a jail sentence, young male DV offenders received sentences about two weeks shorter than older male DV offenders. We examined two measures of pretrial misconduct: failure to appear (FTA) and pretrial re-arrest. Young male DV offenders were considerably more likely to FTA than older male DV offenders (17% vs. 12%). They were also more likely to FTA than young male Non-DV offenders (17% vs. 14%). Similar patterns prevail when examining rates of pretrial re-arrest. Young male DV offenders were considerably more likely to be rearrested during the pretrial period than older male DV offenders (28% vs. 18%), and somewhat more likely to be re-arrested than young male Non-DV offenders (28% vs. 23%). While their overall pretrial re-arrest rate was higher than for either of these other groups, the pattern of re-arrests was different. Young male DV offenders were more

36 likely than older male DV offenders to be re-arrested for Non-DV offenses, while they were more likely than young male Non-DV offenders to be re-arrested for DV offenses. Overall, our findings suggest that young male DV offenders stood out from other offenders primarily because they were especially active in their criminal careers. Considering their age, they had fairly extensive criminal and arrest histories, and they continued their misconduct during the pretrial period. It also appears that their DV arrests were somewhat less likely to be prosecuted.

IV. EXPLAINING DIFFERENCES IN CASE OUTCOMES AND PRETRIAL MISCONDUCT BETWEEN YOUNG MALE DV OFFENDERS AND OTHER OFFENDERS In this chapter, we look more closely at some of the key differences between young male DV offenders and other offenders. Specifically, we focus on case outcomes (disposition and sentencing) and pretrial misconduct (FTA and re-arrest). We examine differences either between young male DV offenders and older male DV offenders, or between young male DV offenders and young male Non-DV offenders. As shown in Chapter 3, young male DV offenders stood out on some of the key measures of case outcomes and pretrial misconduct (see summary in Table 4-1). TABLE 4-1 DIFFERENCES IN CASE OUTCOMES AND PRETRIAL MISCONDUCT BETWEEN YOUNG MALE DV OFFENDERS AND OTHER OFFENDERS Young vs. Older Male DV Offenders

Young Male DV vs. NonDV Offenders

CASE OUTCOMES Conviction

-0-

-15%***

Type of sentence

-0-

Failure to Appear

+5%***

+3%*

Re-arrest for a New DV offense

-0-

Re-arrest for a New Non-DV Offense

+9%***

PRETRIAL MISCONDUCT +13%*** -5%**

- 0 - Indicates no statistically significant difference between young male DV offenders and the comparison group * The percentage difference between the two groups was statistically significant at p < .05. ** The percentage difference between the two groups was statistically significant at p < .01. *** The percentage difference between the two groups was statistically significant at p < .001.

A brief review of the results identifies several differences to be explored. First, young male DV offenders were less likely than young male Non-DV offenders to have their cases end in conviction. Second, young male DV offenders were more likely than older male DV offenders to FTA. Third, young male DV offenders were more likely than young male Non-DV offenders to FTA. Fourth, young male DV offenders were more likely than young male Non-DV offenders to be re-arrested for a DV offense during the pretrial period. Fifth, young male DV offenders were more likely than older male DV

38 offenders to be re-arrested for a Non-DV offense during the pretrial period. Finally, young male DV offenders were less likely than young male Non-DV offenders to be rearrested for a Non-DV offense during the pretrial period. (Note that we found no differences in sentence outcomes between young male DV offenders and other offenders. Consequently, we do not consider sentence outcomes in this chapter.) This chapter examines each of the six differences, and develops a model to attempt to account for the differences. These models are based on prior research (Peterson 2001, 2004, 2006). We will not show the details of each model in this chapter, but rather highlight the modelsâ&#x20AC;&#x2122; ability to account for the differences of interest. The models we use are logistic regression models, since each of the outcomes we measure has only two categories. (For a description of logistic regression analysis and a discussion of how to interpret regression results, see Appendix A.) A. Conviction Young male DV offenders were less likely to be convicted than young male NonDV offenders. As shown in Table 4-1, their conviction rate was 15% lower. As our previous research has shown (Peterson 2001), conviction rates in DV cases are generally lower than conviction rates in Non-DV cases. The difference between young male DV offenders and young male Non-DV offenders probably reflects this general pattern, rather than any specific differences in conviction rates among young offenders. Based on previous research (Peterson 2001), we developed a logistic regression model to predict the likelihood of conviction for young male offenders, both DV and NonDV. The model used the case-based file for the Third Quarter 2005 dataset, including only young male offenders whose cases were docketed. The model included a variable indicating whether the offender was a young male DV offender or a young male Non-DV offender. We first assessed the effect of this variable without any other predictors in the model, and then assessed its effect after including all the predictors of conviction. Since logistic regression uses odds ratios (rather than percentages) to assess the effect of variables, we will compare the odds ratios for the DV variable before and after the predictors were included. (See discussion in Appendix A of logistic regression models and of the interpretation of odds ratios.) Results of our analysis are presented in Table 4-2. As shown there, the odds ratio was initially .54, indicating that the odds of conviction of a young male DV offender were only half as large as the odds of conviction of a young male Non-DV offender. (Note, as explained in Appendix A, that an odds ratio of 1.0 would indicate no difference between the two groups.) Once the predictors of conviction were added to the model, the odds ratio was .62. Because the odds ratio moved closer to 1.0, this change indicates that the model explained a small portion of the difference between young male DV and Non-DV offenders. However, the difference between young male DV offenders and young male Non-DV offenders remained statistically significant.

39 TABLE 4-2 DIFFERENCES IN ODDS OF CONVICTION BETWEEN YOUNG MALE DV OFFENDERS AND YOUNG MALE NON-DV OFFENDERS Outcome Measure

Odds Ratio for Young Male DV Offenders vs. Young Male Non-DV Offenders Before predictors were added

Conviction (N of cases) *** 1

0.54*** (6,843)

After predictors1 were added 0.62*** (6,843)

The odds ratio for the difference between young male DV offenders and young male Non-DV offenders was statistically significant at p < .001. The predictors were: whether the defendant had any prior criminal convictions or open cases, the number of prior misdemeanor convictions, disposition of prior DV case, if any, number of arrest charges, arraignment charge, arraignment charge severity, whether the severity of the top charge was increased or decreased between arrest and arraignment, whether the defendant was ever released, type of release, number of weeks between arraignment and disposition, borough, and ethnicity.

We examined the results of the model to determine whether any particular variables were influential in accounting for the difference between young male DV and Non-DV offenders (data not shown). Arraignment charge and whether the defendant was ever released seemed to be the most important variables. Young male DV offenders were more likely to be charged with assault than young male Non-DV offenders, and the conviction rate for assaults was lower than for other charges. Similarly, young male DV offenders were more likely to be released than young male Non-DV offenders, and defendants who were released were less likely to be convicted. Of course, what is most notable about the findings is that most of the difference between young male DV and Non-DV offenders remains unexplained. B. Failure to Appear Young male DV offenders were more likely to fail to appear than older male DV offenders. Their failure to appear rate was 5% higher (see Table 4-1). Based on previous research (Peterson 2006), we developed a logistic regression model to predict the likelihood of FTA for male DV offenders, both young and old. The model used the defendant-based Third Quarter 2005 dataset, including only DV offenders whose cases were docketed, and who were released at some time between arraignment and disposition. We began by including two control variables in both models (i.e., both the “before” and “after” models). These control variables are included to allow us to more accurately estimate the impact of other variables in the models. One was a control variable that measured time at risk, because the risk of failing to appear increases as the length of pretrial release increases. The other was a control variable to measure selection bias related to the likelihood of release (see Appendix B, section 2, for a

40 discussion of selection bias). We then included a variable indicating whether the offender was a young male DV offender or an older male DV offender. We first assessed the effect of this variable without any other predictors in the model, and then assessed its effect after including all the predictors of failure to appear. With only the control variables included in the model, the odds ratio for young male DV offenders was initially 1.74, and was statistically significant (see Table 4-3). This indicates that the odds of failure to appear for a young male DV offender were 1.74 times as large as the odds of failure to appear for an older male DV offender. Once the predictors of failure to appear were added to the model, the odds ratio was reduced to 1.27 and was statistically insignificant. This indicates that the model explained the difference between young male DV offenders and older male DV offenders. TABLE 4-3 DIFFERENCES IN ODDS OF FAILURE TO APPEAR BETWEEN YOUNG MALE DV OFFENDERS AND OLDER MALE DV OFFENDERS Outcome Measure

Odds Ratio for Young Male DV Offenders vs. Older Male DV Offenders1 Before predictors were added

Failure to Appear (N of cases) *** 1 2

1.74*** (3,828)

After predictors2 were added 1.27 (3,828)

The odds ratio for the difference between young male DV offenders and older male DV offenders was statistically significant at p < .001. The models for both odds ratios control for time at risk and selection bias. The predictors were: whether the defendant had any prior bench warrants, whether the defendant had been arrested for a Non-DV offense in the prior two years, whether the defendant was employed, in school, and/or a training program, whether the defendant had lived at his current address for less than 1 year, type of release, victim-offender relationship, defendantâ&#x20AC;&#x2122;s ethnicity and borough.

As we did in the conviction model, we examined the results of this model to determine which variables were influential in accounting for the difference between young male DV offenders and older male DV offenders (data not shown). Whether the defendant had been arrested for non-DV offenses in the prior two years and victimoffender relationship seemed to be the most important variables. Young male DV offenders were more likely to have been arrested for non-DV offenses in the prior two years, and the FTA rate was higher for those who had such arrests. Similarly, young male DV offenders were less likely than older male DV offenders to have committed an offense against a wife, and defendants whose victims were wives were less likely to FTA.

Young male DV offenders were also more likely to fail to appear than young male Non-DV offenders. Their failure to appear rate was 3% higher (see Table 4-1). Our previous research reported that DV offenders overall (including male and female offenders of all ages) had a slightly lower FTA rate than Non-DV offenders (Peterson 2006). Thus, the higher rate for young male DV offenders reported here does not reflect general patterns for differences in FTA rates between DV and Non-DV offenders. Based on previous research (Peterson 2006), we developed a logistic regression model to predict the likelihood of FTA for young male offenders, both DV and Non-DV. The model used the defendant-based Third Quarter 2005 dataset, including only young male offenders whose cases were docketed, and who were released at some time between arraignment and disposition. TABLE 4-4 DIFFERENCES IN ODDS OF FAILURE TO APPEAR BETWEEN YOUNG MALE DV OFFENDERS AND YOUNG MALE NON-DV OFFENDERS Outcome Measure

Odds Ratio for Young Male DV Offenders vs. Young Male Non-DV Offenders1 Before predictors were added

Failure to Appear (N of cases)

1.41** (3,950)

After predictors2 were added 1.30* (3,950)

The odds ratio for the difference between young male DV offenders and young male Non-DV offenders was statistically significant at p < .05. ** The odds ratio for the difference between young male DV offenders and young male Non-DV offenders was statistically significant at p < .01. 1 The models for both odds ratios control for time at risk and selection bias. 2 The predictors were: whether the defendant had any prior bench warrants, whether the defendant had been arrested for a Non-DV offense in the prior two years, whether the defendant had any prior felony convictions, whether the defendant was employed, in school, and/or a training program, whether the defendant lived with someone, whether the defendant expected someone at arraignment, whether the defendant had a telephone or cell phone, whether the defendant was released at arraignment, and borough.

As in our previous analysis of FTA, we began by including two control variables in both models (i.e., both the “before” and “after” models). One was a control variable that measured time at risk, because the risk of failing to appear increases as the length of pretrial release increases. The other was a control variable to measure selection bias related to the likelihood of release. We then included a variable indicating whether the offender was a young male DV offender or a young male Non-DV offender. We first

42 assessed the effect of this variable without any other predictors in the model, and then assessed its effect after including all the predictors of failure to appear. Before predictors were added to the model, the odds ratio for young male DV offenders was 1.41, and was statistically significant. Adding the predictors to the model decreased the odds ratio only slightly, and it remained statistically significant. These findings indicate that the model accounted for very little of the difference in FTA rates between young male DV offenders and young male Non-DV offenders. The only variable that appeared to have an influence (data not shown) was whether the defendant had any prior bench warrants. Young male DV offenders were more likely than young male Non-DV offenders to have a prior bench warrant, and those with a prior bench warrant were more likely to fail to appear. C. Re-arrest for a New DV Offense Young male DV offenders were much more likely than young male Non-DV offenders to be re-arrested for a new DV offense during the pretrial period. Their pretrial re-arrest rate was 13% higher (see Table 4-1). This finding is generally consistent with our previous research comparing all DV and Non-DV offenders (including males and females of all ages; Peterson 2006). However, the difference was not quite as large in our previous research (about 8%; see Peterson 2006, Fig. 10B, p. 28), so the larger difference among young male offenders may indicate that the risk of re-arrest for a new DV offense is affected by both age and gender. Based on previous research (Peterson 2006), we developed a logistic regression model to predict the likelihood of pretrial re-arrest for a new DV offense for young male offenders, both DV and Non-DV. The model used the defendant-based Third Quarter 2005 dataset, including only young male offenders whose cases were docketed, and who were released at some time between arraignment and disposition. We began by including two control variables in both models (i.e., both the “before” and “after” models). One was a control variable that measured time at risk, because the risk of failing to appear increases as the length of pretrial release increases. The other was a control variable to measure selection bias related to the likelihood of release. We then included a variable indicating whether the offender was a young male DV offender or a young male Non-DV offender. We first assessed the effect of this variable without any other predictors in the model, and then assessed its effect after including all the predictors of pretrial re-arrest for a new DV offense. Before predictors were added to the model, the odds ratio for young male DV offenders was 17.8 and was statistically significant (see Table 4-5). Adding the predictors to the model decreased the odds ratio only slightly, to 14.0, and it remained statistically significant. The only variable that appeared to have any influence in reducing the odds ratio was whether the defendant had any history of DV arrests in the two years prior to the arrest in the third quarter of 2005 (data not shown). Young male DV offenders were much more likely to have a prior DV arrest, and those with a prior

43 DV arrest were much more likely to be re-arrested for a new DV offense during the pretrial period. TABLE 4-5 DIFFERENCES IN ODDS OF RE-ARREST FOR A NEW DV OFFENSE BETWEEN YOUNG MALE DV OFFENDERS AND YOUNG MALE NON-DV OFFENDERS Outcome Measure

*** 1 2

Odds Ratio for Young Male DV Offenders vs. Young Male Non-DV Offenders1 Before predictors were added

After predictors2 were added

Re-arrest for A New DV Offense

17.8***

14.0***

(N of cases)

(3,950)

The odds ratio for the difference between young male DV offenders and young male Non-DV offenders was statistically significant at p < .001. The models for both odds ratios control for time at risk and selection bias. The predictors were: whether the defendant had any prior criminal convictions or open cases, whether the defendant had been arrested for a DV offense in the prior two years, whether the defendant was employed, in school, and/or a training program, type of release, whether the defendant was released at arraignment, and defendant’s ethnicity.

D. Re-arrest for a New Non-DV Offense Young male DV offenders were much more likely than older male DV offenders to be re-arrested for a new Non-DV offense during the pretrial period. Their pretrial rearrest rate was 9% higher (see Table 4-1). To examine the reasons for this difference, we developed a model to explain the likelihood of re-arrest for a new Non-DV offense. The model used the defendant-based Third Quarter 2005 Dataset, including only male DV offenders whose cases were docketed, and who were released at some time between arraignment and disposition. We began by including two control variables in both models (i.e., both the “before” and “after” models). One was a control variable that measured time at risk, because the risk of failing to appear increases as the length of pretrial release increases. The other was a control variable to measure selection bias related to the likelihood of release. We then included a variable indicating whether the offender was a young male DV offender or an older male DV offender. We first assessed the effect of this variable without any other predictors in the model, and then assessed its effect after including all the predictors of pretrial re-arrest for a new Non-DV offense.

44 Before predictor variables were added to the model, the odds ratio for young male DV offenders was 2.9, and was statistically significant (see Table 4-6). After the predictor variables were added to the model, the odds ratio declined to 2.1, but remained statistically significant. Which variables in the model accounted for the decline? Only one variable seems to have an influence: whether the defendant was arrested for a Non-DV offense in the two years prior to the arrest in the third quarter of 2005 (data not shown). Young male DV offenders were much more likely than older male DV offenders to have had such an arrest, and those who had such an arrest were much more likely to be re-arrested for a new Non-DV offense during the pretrial period. TABLE 4-6 DIFFERENCES IN ODDS OF RE-ARREST FOR A NEW NON-DV OFFENSE BETWEEN YOUNG MALE DV OFFENDERS AND OLDER MALE DV OFFENDERS Outcome Measure

Odds Ratio for Young Male DV Offenders vs. Older Male DV Offenders1 Before predictors were added

Re-arrest for A New Non-DV Offense (N of cases) *** 1 2

After predictors2 were added

2.9***

2.1***

(3,828)

The odds ratio for the difference between young male DV offenders and older male DV offenders was statistically significant at p < .001. The models for both odds ratios control for time at risk and selection bias. The predictors were: whether the defendant had any prior criminal convictions or open cases, number of prior misdemeanor convictions, whether the defendant had any open cases, whether the defendant had been arrested for a Non-DV offense in the prior two years, type of release, victim-offender relationship, and borough.

Finally, young male DV offenders were less likely than young male Non-DV offenders to be re-arrested for a new Non-DV offense during the pretrial period. The difference was about 5% (see Table 4-1). This pattern is similar to the pattern we found in our previous research on all DV offenders (male and female, young and old; Peterson 2006). So the difference between young male DV offenders and young male Non-DV offenders is probably a reflection of the more general pattern for all DV offenders, and not peculiar to young male offenders. Nevertheless, we developed a model to explain the likelihood of re-arrest for a new Non-DV offense to account for this difference. The model used the defendant-based Third Quarter 2005 Dataset, including only young male offenders whose cases were docketed, and who were released at some time between arraignment and disposition.

45 We began by including two control variables in both models (i.e., both the “before” and “after” models). One was a control variable that measured time at risk, because the risk of failing to appear increases as the length of pretrial release increases. The other was a control variable to measure selection bias related to the likelihood of release. We then included a variable indicating whether the offender was a young male DV offender or a young male Non-DV offender. We first assessed the effect of this variable without any other predictors in the model, and then assessed its effect after including all the predictors of pretrial re-arrest for a new Non-DV offense. Before predictor variables were added to the model, the odds ratio was 0.7, and was statistically significant (see Table 4-7). (Recall that when an outcome is less likely, the odds ratio is below 1.0.) Adding predictors to the model produced no change in the odds ratio—it remained a statistically significant 0.7. This indicates that the model was unable to account for any of the difference between young male DV offenders and young male Non-DV offenders in the likelihood of pretrial re-arrest for a new Non-DV offense.

TABLE 4-7 DIFFERENCES IN ODDS OF RE-ARREST FOR A NEW NON-DV OFFENSE BETWEEN YOUNG MALE DV OFFENDERS AND YOUNG MALE NON-DV OFFENDERS Outcome Measure

Odds Ratio for Young Male DV Offenders vs. Young Male Non-DV Offenders1 Before predictors were added

Re-arrest for A New Non-DV Offense (N of cases) ** 1 2

0.7** (3,950)

After predictors2 were added 0.7** (3,950)

The odds ratio for the difference between young male DV offenders and young male Non-DV offenders was statistically significant at p < .01. The models for both odds ratios control for time at risk and selection bias. The predictors were: whether the defendant had any prior criminal convictions or open cases, number of prior misdemeanor convictions, whether the defendant had been arrested for a Non-DV offense in the prior two years, arraignment charge, whether the defendant was employed, in school, and/or a training program, whether the defendant had a telephone or cell phone, defendant’s ethnicity and borough.

46 E. Summary and Discussion of Findings This chapter has focused on explaining differences in case outcomes and pretrial misconduct between young male DV offenders and other offenders. In each case, we began by determining what the initial difference was, and placing it in the context of prior research, where available. We then developed a model to explain the outcome of interest, and determined whether that model could account for the difference. When the difference was at least partially explained by the model, we reported which variables seemed to be responsible for explaining the difference. Only one difference in case outcomes was found: young male DV offenders were less likely to be convicted than young male Non-DV offenders. Our model predicting the likelihood of conviction accounted for only a small portion of the difference, and the difference remained statistically significant even after the predictors were included. The model’s ability to account for a portion of the difference was due to two variables: arraignment charge and whether the defendant was ever released. Young male DV offenders were more likely to be charged with assault, and conviction was less likely for assaults, whether committed by DV offenders or Non-DV offenders. Similarly, young male DV offenders were more likely to be released, and conviction was less likely for those who were released. While these explanations are illuminating, it is noteworthy that most of the difference between young male DV offenders and young male Non-DV offenders in the likelihood of conviction is unexplained. Since prior research has found that conviction rates are lower in DV cases than Non-DV cases, regardless of the defendant’s gender or age, the reasons for the lower conviction rate of young male DV offenders are probably not specifically related to the age of the offenders. Rather, the low conviction rate is probably due to the factors generally identified in research comparing DV and Non-DV offenders of all ages: victims’ reluctance to participate in the prosecution, mandatory arrest laws for DV offenses, and prosecutorial policies for filing charges in DV cases (Peterson 2001). Young male DV offenders were more likely than older male DV offenders to fail to appear. Our model was able to explain this difference, and it was no longer statistically significant. Three variables accounted for the difference: whether the defendant had any prior bench warrants, whether the defendant had been arrested for Non-DV offenses in the prior 2 years, and victim-offender relationship. Young male DV offenders were more likely to have a prior bench warrant, and to have been arrested for Non-DV offenses in the prior 2 years, and FTA rates were higher for those with a prior bench warrant or any prior arrests for Non-DV offenses. Young male DV offenders were also less likely to be married than older male DV offenders, and FTA rates were lower for those who were married. It is notable that these variables had substantial effects, and this is the only model presented in this chapter that successfully accounted for a difference between young male DV offenders and other offenders. Young male DV offenders were also more likely to FTA than young male Non-DV offenders. While the initial difference in FTA rates was smaller than the initial difference between young male DV offenders and older male DV offenders, our model was less

47 successful in accounting for the difference. Only one variable, whether the defendant had any prior bench warrants, was able to account for the difference in FTA rates. Young male DV offenders were more likely to have a prior bench warrant than young male Non-DV offenders, and those with a prior bench warrant were more likely to FTA. Young male DV offenders were more likely than young male Non-DV offenders to be re-arrested for a new DV offense during the pretrial period. This difference was large (13%, as reported in Table 4-1) and our model explained only a small portion of the difference. The only variable that helped to account for the difference was whether the defendant had been arrested for a DV offense in the 2 years prior to the arrest in the third quarter of 2005. Young male DV offenders were much more likely to have had such an arrest, and those with a prior DV arrest were more likely to have a pretrial rearrest for a new DV offense. Nevertheless, our most important finding from this model is that we were largely unable to account for the difference in likelihood of re-arrest for a new DV offense. Young male DV offenders were more likely than older male DV offenders to be re-arrested for a new Non-DV offense. Our model was able to explain only a portion of this difference. The only variable that accounted for a portion of the difference was whether the defendant had been arrested for a Non-DV offense in the 2 years prior to the arrest in the third quarter of 2005. Older male DV offenders were less likely to have such an arrest, and those without such an arrest were less likely to be re-arrested for a new Non-DV offense. Again, however, our model was largely unable to account for the difference in likelihood of re-arrest for a new DV offense. Finally, young male DV offenders were less likely than young male Non-DV offenders to be re-arrested for a new Non-DV offense. Our model was unable to account for any of the difference in re-arrest rates for the two groups. After the predictor variables were added to the model, the difference remained as strong as it had been. In reviewing the overall pattern of findings, two things stand out. First, although we examined six differences between young male DV offenders and other offenders, our models were only able to account fully for one of these differences. It is fair to say that our models were largely unsuccessful in accounting for differences between young male DV offenders and other offenders. In some cases, these unexplained differences reflect more general differences between DV and Non-DV offenders (i.e., conviction, rearrest for a new DV offense, re-arrest for a new Non-DV offense). However, the unexplained difference between young and older male DV offenders for re-arrests for a new Non-DV offense does appear to be related to age-specific differences. Furthermore, the difference between young male DV and Non-DV offenders in FTA rates also appears to be related to age-specific differences between young male DV and Non-DV offenders. A second notable pattern concerns our modelsâ&#x20AC;&#x2122; ability to explain differences in pretrial misconduct. Many of the explanatory variables we discussed are not particularly illuminating. Specifically, in many cases the explanatory variable is nothing more than a

48 measure of whether the defendant had a prior history of committing the same type of misconduct. This was true in both models of FTA, in the model of DV re-arrest, and one of two models of Non-DV re-arrest. This â&#x20AC;&#x153;explanationâ&#x20AC;? amounts to saying that those who commit a type of misconduct are likely to repeat it. Nevertheless, the knowledge that defendants are likely to repeat misconduct may be a useful guide for prosecutors and judges in determining how to address concerns about misconduct in DV cases.

49 V. PREDICTING PRETRIAL MISCONDUCT AMONG YOUNG MALE DV OFFENDERS In this chapter, we focus exclusively on young male DV offenders. Our goal is to identify factors that predict pretrial misconduct. We will examine two types of pretrial misconduct that are of particular interest for this group: failure to appear and pretrial rearrest for a new DV offense. A. Failure to Appear Our model of failure to appear used the standard set of variables considered for the FTA models in Chapter 4 (see Tables 4-3 and 4-4). The model used the defendantbased Third Quarter 2005 Dataset, including only young DV male offenders whose cases were docketed, and who were released at some time between arraignment and disposition. The model includes control variables to take into account the effects of time at risk and selection bias (see discussion in Appendix A, Section 2). Very few variables were statistically significant predictors of FTA for young male DV offenders (see Table 5-1). The first variable to enter the model was not surprising: whether the defendant had any prior bench warrants. The odds of FTA were over 2 times greater for young male DV offenders who had a prior bench warrant than for those who did not have any prior bench warrants. As indicated by the size of the standardized beta (0.48), this was the strongest predictor in the model with the exception of time at risk. The next variable to enter the model was borough. Young male DV offenders whose cases were docketed in Staten Island were more likely to FTA than those in other boroughs. The odds of FTA were over 4 times greater in Staten Island than in Brooklyn. Finally, young male DV offenders who were 16-20 years old were more likely to FTA than those who were 21-24 years old. Their odds of FTA were 1.55 times greater. Even within this narrowly defined age group of young male DV offenders age 16-24, the youngest offenders were more likely to FTA. Overall, the model explained only 14% of the variation in failure to appear (see Nagelkerke R2). This indicates that most of the variation in failure to appear is unexplained, and reflects to some degree the limited number of variables that entered the model. Although many additional variables were considered for inclusion, none of them had a statistically significant effect on likelihood of failure to appear. The findings from this model indicate that to reduce the FTA rate of young male DV offenders, courts should pay more attention to prior warrant history when making release decisions. Although it is not clear why the FTA rate is higher in Staten Island, the findings from this model also suggest that release policies for young male DV offenders in that borough may need some adjustment. Finally, even among young male DV offenders, the youngest offenders have a higher risk of FTA. Since the court is not permitted to take age into account when making release decisions, it may be difficult to address this problem.

50 TABLE 5-1 LOGISTIC REGRESSION MODEL PREDICTING LIKELIHOOD OF FAILURE TO APPEAR AMONG RELEASED DEFENDANTS 1 YOUNG MALE DV OFFENDERS, CRIMES AGAINST PERSONS AND PROPERTY SUBSAMPLE Third Quarter 2005 Dataset 2

INDEPENDENT VARIABLES

Standardized

Odds

ď ˘

Ratio

CONTROL VARIABLES SELECTION BIAS CORRECTION: LIKELIHOOD OF RELEASE TIME AT RISK

-0.30 ** 0.56 ***

0.08 1.01

0.48 ***

2.19

DEFENDANT'S CRIMINAL HISTORY ANY PRIOR BENCH WARRANTS BOROUGH Reference Category: Brooklyn Manhattan Queens Staten Island Bronx

-0.08 0.21 0.30 ** -0.01

0.84 1.54 4.21 0.99

0.27 *

1.55

DEFENDANT'S DEMOGRAPHIC CHARACTERISTICS AGE: Reference Category: Age 21-24 Age 16-20 2

Nagelkerke R (N of cases)

.14 *** (868)

NOTES 1 2

See text for a description of the dataset and the subsample. See Appendix C for information about the measurement and coding of the variables. * Statistically significant at p < .05 ** Statistically significant at p < .01 *** Statistically significant at p < .001

B. Pretrial Re-arrest for a New DV Offense Our model of pretrial re-arrest for a new DV offense used the standard set of variables considered in previous research (Peterson 2006). The model used the defendant-based Third Quarter 2005 Dataset, including only young DV male offenders whose cases were docketed, and who were released at some time between arraignment and disposition. The model controls for time at risk and for selection bias. The model shows that arrest history in the prior 2 years had a statistically significant effect on the likelihood of pretrial re-arrest for a new DV offense (see Table 5-

51 2). Young male DV defendants who had a history of arrests for DV offenses in the prior two years were more likely to be re-arrested for a new DV offense. Their odds of rearrest were 2.35 times greater than for defendants who had no DV arrests in the prior two years. Other than time at risk, this variable was the strongest predictor of pretrial re-arrest for a new DV offense. Defendants who were not employed, in school, and/or in a training program were also more likely to be re-arrested for a new DV offense during the pretrial period. This finding suggests that those not engaged in full-time activity have weaker community ties and are more likely to re-offend. Finally, young male DV offenders who were released on bail were less likely to be re-arrested for a new DV offense. The reasons for this finding are not clear. Although it could be argued that bail was more effective than ROR at preventing pretrial re-arrest for a new DV offense, it may be that bail is associated with other unmeasured characteristics (e.g., the defendant’s financial resources) that accounted for this effect. TABLE 5-2 LOGISTIC REGRESSION MODEL PREDICTING LIKELIHOOD OF PRETRIAL RE-ARREST FOR A NEW DV OFFENSE AMONG RELEASED DEFENDANTS YOUNG MALE DV OFFENDERS, CRIMES AGAINST PERSONS AND PROPERTY SUBSAMPLE1 Third Quarter 2005 Dataset 2

INDEPENDENT VARIABLES

Standardized

Odds



Ratio

CONTROL VARIABLES SELECTION BIAS CORRECTION: LIKELIHOOD OF RELEASE TIME AT RISK

0.08 0.73 ***

2.16 1.01

0.41 ***

2.35

0.30 *

1.70

-0.30 *

0.46

DEFENDANT'S CRIMINAL HISTORY ANY DV ARRESTS IN 2 PRIOR YEARS DEFENDANT'S COMMUNITY TIES NOT EMPLOYED, IN SCHOOL, OR IN A TRAINING PROGRAM RELEASE CHARACTERISTICS RELEASED ON BAIL 2

Nagelkerke R (N of cases)

.16 *** (868)

NOTES 1 2

52 The model accounted for about 16% of the variation in pretrial re-arrest for a new DV offense (see Nagelkerke R2). Most of the variation in pretrial re-arrest for a new DV offense remains unexplained. Only three variables had statistically significant effects, although many more were considered for inclusion in the model. To further explore the factors influencing pretrial re-arrest for a new DV offense, we divided young male DV offenders into two groups: those who had a history of arrests for DV offenses in the prior two years, and those who did not. As indicated in the findings from the model in Table 5-2, a history of prior DV arrests was an important predictor of pretrial re-arrest for a new DV offense. In our subsequent analyses, we attempted to determine whether the factors predicting re-arrest were different for those with a history of DV arrests in the prior 2 years and for those without. Among those with a history of DV arrests in the prior 2 years, two variables were significant predictors of pretrial re-arrest for a new DV offense (Table 5-3). By far TABLE 5-3 LOGISTIC REGRESSION MODEL PREDICTING LIKELIHOOD OF PRETRIAL RE-ARREST FOR A NEW DV OFFENSE AMONG RELEASED DEFENDANTS WHO WERE ARRESTED FOR A DV OFFENSE IN THE PRIOR TWO YEARS YOUNG MALE DV OFFENDERS, CRIMES AGAINST PERSONS AND PROPERTY SUBSAMPLE1 Third Quarter 2005 Dataset INDEPENDENT VARIABLES

Standardized

Odds

ď ˘

Ratio

CONTROL VARIABLES SELECTION BIAS CORRECTION: LIKELIHOOD OF RELEASE TIME AT RISK

0.12 0.54 ***

3.28 1.01

ANY PRIOR CRIMINAL CONVICTIONS OR OPEN CASES DEFENDANT CONVICTED IN PRIOR DV CASE

0.70 * 0.34 *

11.97 2.40

Nagelkerke R2 (N of cases)

.20 *** (209)

DEFENDANT'S CRIMINAL HISTORY

NOTES 1 2

the most important predictor was whether the defendant had any prior criminal convictions (in DV or Non-DV cases) or open cases. Among young male DV offenders with a DV arrest in the prior two years, the odds of pretrial re-arrest for a new DV offense were almost 12 times greater for those who had prior criminal convictions or open cases than for those who did not. One other variable also had an influence:

53 whether or not the defendant had been convicted in the DV case prior to the current one. The odds of re-arrest for a new DV offense were 2.4 times greater for young male DV offenders who had been convicted in their prior DV case than for those whose cases had been dismissed or had been disposed with an ACD. Taken together, these two variables account for about 12% of the variation in pretrial re-arrest for a new DV offense (data not shown). Among those who did not have a history of DV arrests in the prior 2 years, two different variables were significant predictors (see Table 5-4). The two variables were whether the defendant was employed, in school, and/or in a training program full time, and whether the defendant was released on bail. These two variables were also significant in the model presented in Table 5-2, and their effects were in the same direction. Defendants who were not engaged in full-time activity were more likely to be re-arrested for a new DV offense during the pretrial period, while those released on bail (vs. ROR) were less likely to be re-arrested. TABLE 5-4 LOGISTIC REGRESSION MODEL PREDICTING LIKELIHOOD OF PRETRIAL RE-ARREST FOR A NEW DV OFFENSE AMONG RELEASED DEFENDANTS NOT ARRESTED FOR A DV OFFENSE IN THE PRIOR TWO YEARS YOUNG MALE DV OFFENDERS, CRIMES AGAINST PERSONS AND PROPERTY SUBSAMPLE1 Third Quarter 2005 Dataset 2

INDEPENDENT VARIABLES

Standardized

Odds

ď ˘

Ratio

CONTROL VARIABLES SELECTION BIAS CORRECTION: LIKELIHOOD OF RELEASE TIME AT RISK

0.18 0.64 ***

14.30 1.01

0.35 **

2.17

DEFENDANT'S COMMUNITY TIES NOT EMPLOYED, IN SCHOOL, OR IN A TRAINING PROGRAM RELEASE CHARACTERISTICS RELEASED ON BAIL

-0.50 *

Nagelkerke R2 (N of cases)

.17 *** (659)

0.18

NOTES 1 2

Taken together, the findings from the models predicting pretrial re-arrest for a new DV offense suggest that the factors influencing this type of pretrial misconduct are

54 different for young male DV offenders who have a recent history of DV arrests and those who do not. Specifically, those who had DV arrests in the prior 2 years were more likely to be re-arrested for a new DV offense during the pretrial period if they had prior criminal convictions or open cases, and if their most recent DV arrest in the prior 2 years had ended in conviction. For these defendants, these two measures of criminal history are strong predictors of pretrial re-arrest for a new DV offense. On the other hand, young male DV offenders who had no DV arrests in the prior 2 years were more likely to be re-arrested during the pretrial period if they were not engaged in full time activity (work, school, and/or a training program), and if they were released on recognizance rather than on bail. For these defendants, criminal history is not a predictor of this type of pretrial misconduct. Instead, one measure of community ties (full-time activity) and type of release are the most important factors. C. Summary and Discussion of Findings This chapter has developed models predicting pretrial misconduct among young male DV offenders. Specifically, we have examined models predicting pretrial failure to appear and pretrial re-arrest for a new DV offense. Both models found that, aside from time at risk, the strongest predictor of pretrial misconduct was a history of prior misconduct of the same type. Young male DV offenders who had prior bench warrants were more likely to fail to appear. Similarly, young male DV offenders who had any DV arrests in the prior 2 years were more likely to be re-arrested for a new DV offense during the pretrial period. As we noted in Chapter 4, it may be useful to know that defendants are likely to repeat prior misconduct. Prosecutors and judges may use this information to determine how to handle particular cases. However, these findings are not as helpful if the goal is to develop an early intervention strategy. Our model predicting failure to appear found two additional predictors of FTA: borough and age. Young male DV defendants in Staten Island were more likely to fail to appear, and special attention may be needed in this area. We also found that even among our sample of young male DV defendants age 16-24, the youngest defendants, those age 16-20 were more likely to FTA. This finding suggests that age is especially important, and that the youngest defendants are at the highest risk of this type of misconduct. While this may provide some guidance suggesting how prosecutors and judges might respond to young male DV offenders, it is not particularly illuminating in suggesting how to intervene with these defendants. Our model predicting pretrial re-arrest for a new DV offense also identified two additional predictors: whether the defendant was employed, in school, and/or in a training program full time, and whether the defendant was released on bail. Young male DV defendants who were not engaged in full-time activity (employment, school, and/or training) were more likely to be re-arrested for a new DV offense during the pretrial period. This finding suggests that full-time involvement in employment, education and/or training may be an important factor inhibiting young males from

55 committing new DV offenses. We also found that defendants released on bail are less likely to be re-arrested for a new DV offense during the pretrial period. However, as noted in a previous report, it is unclear whether it is bail or the financial circumstances of the defendant that explain the lower re-arrest rate (Peterson 2006). Moreover, it seems unlikely that judges would set bail more often in DV cases, for a variety of reasons (Peterson 2006). To further explore the factors associated with pretrial re-arrest for a new DV offense for young male DV offenders, we conducted separate analyses for those with a history of DV arrests in the prior 2 years, and for those with no DV arrests in the prior 2 years. Among those with a history of DV arrests, we found that those with prior criminal convictions (in DV or Non-DV cases) or open cases were more likely to be re-arrested, and that those who were convicted in their prior DV case were more likely to be rearrested. These findings suggest that among those with a history of DV arrests in the prior 2 years, those with a criminal record, particularly those convicted in their prior DV case, pose a higher risk of re-arrest for a new DV offense. Among those who did not have a DV arrest in the prior 2 years, the results mirrored those for all young male DV offenders. (This is not particularly surprising, since over three quarters of the young male DV offenders were in this group.) Those engaged in full-time activity and those released on bail were less likely to be rearrested for a new DV offense. Taken together these findings suggest that there may be an opportunity for early intervention with young male DV offenders who are not engaged in full-time activity. Among those who did not have a history of DV arrests, young male DV offenders who were not employed, in school, and/or in a training program had a rearrest rate of 16% for new DV offenses during the pretrial period (see Table 5-5, next page). However, among those who did not have a history of DV arrests, young male DV offenders who were employed, in school, and/or in a training program had a rearrest rate of 8%. A closer look (data not shown) reveals that it is employment, rather than school or training programs, that is responsible for the difference in re-arrest rates. If these defendants could be placed in effective employment programs, their re-arrest rate for new DV offenses might be reduced by half. Our analyses indicate that such programs should target those who do not have a history of DV arrests in the prior 2 years. The targeted group (young male DV offenders who do not have a history of DV arrests in the prior 2 years and who were not employed, in school, and/or in a training program full time) constitutes 30% (258) of the 868 young male DV offenders in the third quarter of 2005. This may represent a significant opportunity for early intervention. Of course, defendants targeted for employment programs would be likely to require additional schooling, training and referral services, and it might take considerable time for them to become stably employed. The benefits of such an intervention would not be seen in the short-term, and certainly not during the typical 4month pretrial period. We anticipate that the benefits would be seen in the long-term, however, after the disposition of the current case.

56 Among young male DV offenders who did have a history of DV arrests in the prior 2 years, our models do not suggest that employment, school or training programs would reduce the pretrial re-arrest rate for new DV offenses. Their re-arrest rate was 23% whether or not they were engaged in full-time activity. These offenders constitute almost 25% of all young male DV offenders (209 of the 868), and they appear to be chronic offenders who have considerable contact with the criminal justice system. However, there is little evidence, either in this study, or in previous research, that the penalties imposed by the criminal justice system have a deterrent effect on these offenders. TABLE 5-5 PRETRIAL RE-ARREST RATE FOR NEW DV OFFENSES AMONG RELEASED DEFENDANTS BY ENGAGEMENT IN FULL-TIME ACTIVITY AND HISTORY OF DV ARRESTS IN PRIOR TWO YEARS Young Male DV Offenders, Crimes Against Persons and Property Subsample1 Third Quarter 2005 Dataset PRETRIAL RE-ARREST RATE FOR NEW DV OFFENSES ENGAGED IN FULL-TIME ACTIVITY Re-arrested for a new DV offense during the pretrial period (N of cases) NOT ENGAGED IN FULL-TIME ACTIVITY Re-arrested for a new DV offense during the pretrial period (N of cases) 1

No DV Arrests in Prior Two Years

Had DV Arrests in Prior Two Years

23%

(401)

(95)

16%

23%

(258)

(114)

See Chapter 2 for a description of the dataset and the subsample.

Interestingly, engagement in full-time activity does not have a similar effect on older male DV offenders. Among older male DV offenders with no DV arrests in the prior two years, 11% of those not engaged in full-time activity were re-arrested in the pretrial period, compared to 9% of those who were engaged in full-time activity (data not shown). This suggests that employment programs would have little impact on re-arrest rates of older male DV offenders. It also emphasizes the importance of intervening with DV offenders early on, while they are relatively young. It is at this stage that there is a potential to influence both short-term and long-term outcomes.

VI. CONCLUSION A. Major Findings This study addressed three questions using data from offenders arrested in New York City in the third quarter of 2005: 1) What were the characteristics of young male DV offenders? 2) How did young male DV offenders compare to older male DV offenders and to young male Non-DV offenders? 3) What factors influenced pretrial misconduct among young male DV offenders? The answers to these questions can be summarized as follows. First, most young male DV offenders had active criminal careers, and many had a history of domestic violence arrests in the prior two years. Although they generally had strong community ties, over 40% were not engaged in full-time activity (work, school and/or training). The typical young male DV offender was charged with an assault against an intimate partner (usually a girlfriend or common-law spouse) or against his parent or sibling. District Attorneys declined to prosecute about one sixth of the cases against young male DV offenders. Among the cases that were prosecuted, most defendants were charged with misdemeanor assault, and released on recognizance prior to disposition. About one third of the cases ended in a conviction, and one quarter of those convicted were sentenced to jail. During the pretrial period, many young male DV offenders engaged in pretrial misconduct, either missing a court appearance and/or being re-arrested. Second, young male DV offenders had more extensive criminal histories than young male Non-DV offenders. They were also more likely to have a recent arrest history than both young male Non-DV offenders and older male DV offenders. Young male DV offenders were more likely than young male Non-DV offenders to have been arrested for DV offenses in the prior 2 years. They were also more likely than older male DV offenders to have been arrested for Non-DV offenses in the prior two years. In their current case (in the third quarter 2005), young male DV offenders were less likely than older male DV offenders to be married to the victim, and more likely to have victimized a girlfriend or other family member (usually his parent or sibling). Young male DV offenders were less likely than others to be engaged in full-time activity. There were only a few case processing differences. Young male DV offenders were more likely to have their cases declined for prosecution than older male DV offenders or young male Non-DV offenders. Case dispositions were similar for young male DV offenders and older male DV offenders, however, young male non-DV offenders were more likely to have their cases end in conviction or an ACD and less likely to have their

58 cases dismissed. Young male DV offenders who were sentenced to jail received shorter sentences than older male DV offenders. Young male DV offenders were more likely to engage in pretrial misconduct (FTA and pretrial re-arrest) than older male DV offenders or young male Non-DV offenders. We developed models to attempt to account for differences in case outcomes and pretrial misconduct between young male DV offenders and other offenders. Our models were generally unsuccessful in explaining the differences. Third, our models found only a few factors that predicted which young male DV offenders engaged in pretrial misconduct. The strongest predictor of FTA was a history of FTA in previous cases. Surprisingly, we also found that even among young male DV offenders, age mattered: the youngest of the young male DV offenders, those age 1620, were more likely to FTA than those age 21-24. The strongest predictor of pretrial rearrest for a new DV offense was whether the defendant had any DV arrests in the prior two years. In addition, those engaged in full-time activity were less likely to be rearrested for a new DV offense, as were those released on bail (rather than ROR). Further exploration of these findings showed that among young male DV defendants who did not have a history of DV arrests in the prior 2 years, those engaged in full-time activity had a substantially lower pretrial re-arrest rate (8% vs. 16%, as reported in Table 5-5) than those who were not engaged in full-time activity. This suggests that employment programs might be effective at reducing the re-arrest rate for new DV offenses for some young male DV offenders. This is a potential area for early intervention. B. Discussion The picture of young male DV offenders provided in this report addresses issues about the role of age in understanding domestic violence. We have explored three areas where age may play a role: 1) criminal activity and community ties, 2) the response of the criminal justice system, and 3) pretrial misconduct. Comparisons of young male DV offenders to older male DV offenders reveal several differences in criminal activity and community ties that are related to age. Young male DV offenders are more likely to commit offenses against girlfriends, parents and siblings. They are also more likely to have been arrested for Non-DV offenses in the prior 2 years. Young male DV offenders are less likely to be engaged in full-time activity. These findings suggest that young male DV offenders spend more time engaged in crime and less time in work or school than older male DV offenders. They are unlikely to be married, and they commit DV offenses against those with whom they are most likely to have active relationships. Our findings suggest that the criminal justice system generally responds to young male DV offenders and older male DV offenders in similar ways. Case dispositions and types of sentences were the same, regardless of age. There were two areas where it appeared that young male DV offenders received more lenient treatment. Young male DV offenders were slightly more likely to have their cases declined for

59 prosecution than older male DV offenders. Also, among the small numbers of DV offenders who received jail sentences, the jail terms of young offenders were about two weeks shorter than those of older offenders. These findings suggest that although age does not generally affect criminal justice outcomes, to the extent that age does have an effect, young male DV offenders are treated somewhat more leniently. Age also affects the type and amount of pretrial misconduct committed by DV offenders. Young male DV offenders are more likely to fail to appear than older male DV offenders. They are also more likely to be re-arrested during the pretrial period. However, this difference emerges because young male DV offenders are more likely than older male DV offenders to be re-arrested for new Non-DV offenses. Their pretrial re-arrest rates for new DV offenses are similar. These findings suggest that although age affects overall rates of misconduct by DV offenders, it is not associated with the likelihood of pretrial re-arrest for a new DV offense.16 Age plays an important role in understanding domestic violence. Arrests of young males for domestic violence do not appear to be isolated events among young males with clean records. Instead, young male DV offenders generally have a criminal record and a history of recent arrests. Many young men are arrested for domestic violence offenses as part of a broader pattern of criminal behavior. Moreover, they are often re-arrested during the pretrial period. Considering age differences between the groups, young male DV offenders (average age: 21) have fairly extensive criminal and arrest histories when compared to older male DV offenders (average age: 37), who have had a much longer period to accumulate criminal and arrest histories. This suggests that early intervention may be more difficult than might have been expected. Even at young ages, many male DV offenders already have extensive criminal and arrest histories. As we found repeatedly in our analyses, these histories were the strongest predictors of misconduct. However, as we also noted, knowing that past misconduct predicts future misconduct does not help to identify a way to intervene and break the cycle. Because high rates of prior misconduct limit the number of young male DV offenders for whom early intervention is possible, we focused on those who did not have a record of DV arrests in the prior 2 years. We found that within this group, those who were engaged in full-time activity (primarily employment) had lower rates of pretrial rearrest for a new DV offense than those who were not engaged in full-time activity. This finding suggests that employment provides young men with a tie to the community that reduces the likelihood of re-arrest. On the basis of this finding, we have suggested that an employment program for young male DV offenders who have not been arrested for a DV offense in the prior 2 years might reduce re-arrest rates for new DV offenses. Over two thirds of young male DV offenders would be eligible for such a program. 16

This finding initially seemed surprising, since previous research (Peterson 2006, 2008b) found that older defendants were less likely to be re-arrested for a new DV offense during the pretrial period. However, a closer look at previous research shows that the re-arrest rate did not decline until age 40, whereas older male DV offenders are defined here as those 25 and older.

60 While we believe such a program would be worth considering, we note several cautions. First, employment itself may not be the reason for the lower rate of pretrial rearrest for a new DV offense. Employment might instead measure broader social class background. If this is the case, young male DV offenders who come from working-class (as opposed to impoverished) backgrounds may be more likely to have a job, and less likely to be re-arrested. If their working-class backgrounds influence both their employment prospects and re-arrest rates, employment by itself may not influence rearrest rates. That would suggest that providing employment for impoverished defendants might not reduce their re-arrest rate, since they do not have a working-class background. We do not have a good measure of social class background that would let us directly explore this question. It is interesting, however, that two other measures that may be associated with social class (having a telephone or cell phone, living at current address more than one year) do not have the same effect on re-arrest rates that employment does. This lends some support to the argument that employment itself, and not social class background, is associated with lower re-arrest rates. Second, our models predicting the likelihood of pretrial re-arrest for a new DV offense among young male DV offenders explained less than one fifth of the variance. We were only able to identify a limited number of variables that predicted the likelihood of re-arrest. Because of these limitations, recommendations based on these findings are likely to have modest effects. Third, the effect of employment on the re-arrest rate for new DV offenses has only been examined in the short-term, during the pretrial period. To verify that employment would be an effective deterrent, re-arrests should be examined over a longer period of time. This would enable us to verify that the re-arrest rate for those engaged in full-time activity remains substantially lower in the long-term. It would also enable us to identify additional predictor variables, since there would be more variation in the re-arrest rate over a longer period. Furthermore, if additional predictor variables are found, they could be used to identify a more narrowly defined target group for an employment program. Finally, it may be difficult to establish an effective employment program. We know very little about which employment programs are effective at reducing recidivism, especially for young offenders (Bloom 2006, Buck 2000, Lipsey et al. 2000). Some recent programs appear promising (Aos et al. 2006, Buck 2000) and several rigorous studies are under way to evaluate new programs (Bloom 2006, Bloom 2009, Bloom et al. 2007, Redcross et al. 2009). Early results of these studies are promising, showing slightly lower re-arrest rates for those participating in employment programs (Redcross et al. 2009, Lattimore 2009). Once these studies are completed, they may provide strong evidence about the effectiveness of employment programs for criminal offenders and some guidance about how to design effective programs. However, until these studies are completed, establishing effective employment programs for offenders will involve guesswork. Moreover, the task is even more difficult because we are not aware of any employment programs specifically designed for young male DV offenders. New

61 employment programs for young male DV offenders would have to be designed on the basis of the experience with employment programs for criminal offenders in general. In spite of these cautionary notes, we believe that establishing employment programs for young male DV offenders would be worth considering. Other approaches, such as batterer intervention programs and most criminal justice interventions (arrest, prosecution, conviction and jail sentences), have been tried and have limited or no effect on DV recidivism (Aos et al. 2006, Maxwell et al. 2002, Peterson 2003). Considering the high cost of interventions that are currently in use but ineffective, a pilot program to test the effectiveness of employment programs for young male DV offenders could be a worthwhile investment. Prevention of domestic violence is a relatively new field. Early intervention to address domestic violence has only recently become a focus of attention (Carter 2005, Edleson 2000, Kahn and Paluzzi 2006, Rosewater 2003). Young women age 16-24 experience the highest rates of domestic violence, and young men in the same age group have high rates of domestic violence offending (Carter 2005). Domestic violence among teens and young adults may be a precursor to domestic violence later in life. Intervening early has the potential to significantly reduce future rates of domestic violence. Yet only a few early intervention programs have been established, primarily public education programs and school-based educational programs (Carter 2005). Moreover, there is little research available on the long-term success of these programs (Cissner 2005). Testing the effectiveness of employment programs would be a new direction in the field of domestic violence prevention. In addition to our finding that employment seems to inhibit pretrial re-arrest for new DV offenses, we also found that young male DV defendants released on bail (vs. ROR) were less likely to be re-arrested for a new DV offense during the pretrial period. It is not clear whether other unmeasured characteristics (e.g., the defendantâ&#x20AC;&#x2122;s financial resources) accounted for this effect, or whether bail was more effective than ROR at preventing pretrial re-arrest for these defendants. Regardless of the reason, this finding is not likely to be useful for developing an early intervention program. As discussed in previous reports (Peterson 2006, 2008b), it would be difficult in practice for the courts to set bail rather than release the defendant on recognizance. Judges are likely to be reluctant to set bail more often, or to set higher bail, because defendants in misdemeanor DV cases are unlikely to be convicted, or if convicted, unlikely to be sentenced to jail. Changes in bail may also hurt victims financially if they are dependent on the defendantâ&#x20AC;&#x2122;s income. Setting bail more often at arraignment would cause the detention of numerous defendants who would not fail to appear if released and would increase costs to the criminal justice system for detaining and transporting defendants. Changing release decisions to prevent re-arrest is further complicated by New York State law, which does not allow for preventive detention of defendants based on the potential risk of re-offending. Given the legal restrictions and the limitations suggested by our research, what actions can the courts take to address concerns about risk of pretrial re-arrest in DV

62 cases? As discussed in our previous report, the use of supervised release programs for both the general population of defendants, and for DV defendants in particular, has grown more common over the past decade (Clark and Henry 2003). While it is beyond the scope of the current study to develop a plan for a supervised release program for DV defendants in New York City, we proposed some general guidelines for such a program in our previous reports (Peterson 2006, 2008b). C. Conclusion This study was undertaken with the goal of learning whether there is a potential for early intervention with young male DV offenders that might deter them from committing future acts of domestic violence. While previous research suggests that few interventions or criminal justice sanctions have a deterrent effect, most of that research was conducted on the general population of domestic violence offenders. The current study focused on young male DV offenders to determine whether there are opportunities to intervene early in an offenderâ&#x20AC;&#x2122;s criminal career in ways that might deter future acts of domestic violence. Pretrial misconduct, including failure to appear and re-arrest, is a frequent problem among young male DV defendants in New York City. Furthermore, such pretrial misconduct often includes a re-arrest for a new DV offense. What can be done to reduce the rates of pretrial re-arrest for new DV offenses? We have proposed establishing an employment program for young male DV offenders. Our research suggests that employment is a key factor inhibiting young male DV offenders from committing new DV offenses during the pretrial period. While no employment program is likely to be successful in providing stable employment to targeted defendants within the short pretrial period, an employment program may have effects over the long term. If this assumption is correct, post-disposition recidivism may be reduced. More importantly, an employment program for young male DV offenders has the potential to reduce recidivism over a long period of time, since the peak years for DV offending are ages 21 to 39. Reducing re-arrests for new DV offenses during the pretrial period among young male DV offenders is likely to be difficult. Changes in bail-setting practices significant enough to reduce pretrial misconduct among DV defendants are unlikely. Setting conditions of release, particularly by requiring supervised release for some DV defendants, may be an effective way to address concerns about both FTA and pretrial re-arrest for new DV offenses. This would require development and implementation of a supervised release program. Further research is needed to explore issues raised in this report. The effect of employment on post-disposition recidivism should be examined to ensure that employment has long-term effects. Such research could identify additional predictor variables, which may provide more information about which defendants should be targeted for employment. Research is also needed on early patterns of offending

63 among the youngest DV offenders, age 16-20. One intriguing possibility discussed in chapter 3 is that young males age 16-18 who commit acts of domestic violence against a parent or sibling may later be likely to commit acts of domestic violence against intimate partners. Testing this hypothesis would require a larger sample of very young DV offenders. Research on post-disposition recidivism and on domestic violence against parents and siblings might yield further insights for early intervention to stop the cycle of domestic violence.

[This page intentionally left blank]

65 VII. REFERENCES Aldrich, Liberty, and Julie A. Domonkos. 2000. “Navigating the NYPD: A Guide to New York City Police Department Policies and Procedures for Family Court Attorneys Handling Order of Protection Cases.” Pp. 279-292 in Julie A. Domonkos and Jill Laurie Goodman, Lawyer’s Manual on Domestic Violence: Representing the Victim. New York: Supreme Court of the State of New York, Appellate Division, First Department. Aos, Steve, Marna Miller, and Elizabeth Drake. 2006. Evidence-Based Public Policy Options to Reduce Future Prison Construction, Criminal Justice Costs, and Crime Rates. Olympia: Washington State Institute for Public Policy. Berk, Richard A. 1983. “An Introduction to Sample Selection Bias in Sociological Data.” American Sociological Review 48:386-398. Bloom, Dan. 2006. “Employment Focused Programs for Ex-Prisonsers: What Have We Learned, What Are We Learning, and Where Should We Go From Here?” Working Paper. New York: MDRC. Bloom, Dan. 2009. Transitional Jobs Reentry Demonstration: Testing Strategies to Help Former Prisoners Find and Keep Jobs and Stay Out of Prison. Chicago: The Joyce Foundation. Bloom, Dan, Cindy Redcross, JoAnn Hsueh, Sarah Rich and Vanessa Martin. 2007. “Four Strategies to Overcome Barriers to Employment: An Introduction to the Enhanced Services for the Hard-to-Employ Demonstration and Evaluation Project.” Working Paper. New York: MDRC. Buck, Maria L. 2000. Getting Back to Work: Employment Programs for Ex-Offenders. Philadelphia, PA: Public/Private Ventures. Carter, Janet. 2005. Domestic Violence, Child Abuse, and Youth Violence: Strategies for Prevention and Early Intervention. San Francisco: Family Violence Prevention Fund. Cissner, Amanda. 2005. Process Evaluation of the Brooklyn Youthful Offender Domestic Violence Court. New York: Center for Court Innovation. Clark, John, and D. Alan Henry. 2003. Pretrial Services Programming at the Start of the 21st Century: A Survey of Pretrial Programs. Washington, D.C.: U.S. Department of Justice. Edleson, Jeffrey L. 2000. “Primary Prevention and Adult Domestic Violence.” Paper presented at the meeting of the Collaborative Violence Prevention Initiative, San Francisco, February 17-18.

Feder, Lynette and David B. Wilson. 2005. "Meta-Analytic Review of Court-Mandated Batterer Intervention Programs: Can Courts Affect Abusers' Behavior?" Journal of Experimental Criminology 1:239-262. Frisch, Lisa A., Melissa I. Mackey, Donna Hall, and Alissa Pollitz Worden. 2001. Family Protection and Domestic Violence Intervention Act of 1994: Evaluation of the Mandatory Arrest Provisions (Final Report to the Governor and Legislature). Albany, NY: Division of Criminal Justice Services. Heckman, James. 1979. “Sample Selection Bias as a Specification Error.” Econometrica 47:153-161. Kahn, Abby and Pat Paluzzi. 2006. Boys Will Be Boys: Understanding the Impact of Child Maltreatment and Family Violence on the Sexual, Reproductive, and Parenting Behaviors of Young Men. Washington, DC: Healthy Teen Network. Lattimore, Pamela K. 2009. Impact of SVORI Reentry Program Participation. Presentation at the American Probation and Parole Association Training Institute, Anaheim, CA, August 25. Lerman, Lisa G. 1992. “The Decontextualization of Domestic Violence.” The Journal of Criminal law and Criminology 83:217-240. Lipsey, Mark W., David B. Wilson, and Lynn Cothern. 2000. Effective Intervention for Serious Juvenile Offenders. Washington, DC: Office of Juvenile Justice and Delinquency Prevention. Maxwell, Christopher D., Joel H. Garner and Jeffrey A. Fagan. 2002. “The Preventive Effects of Arrest on Intimate Partner Violence: Research, Policy and Theory.” Criminology and Public Policy 2:51-79. Menard, Scott. 1995. Applied Logistic Regression Analysis. Sage Publications: Thousand Oaks, CA. Mohr, Lawrence B. 1990. Understanding Significance Testing. Sage Publications: Thousand Oaks, CA. Murphy, Christopher M., Peter M. Musser, and Kenneth I. Maton. 1998. “Coordinated Community Intervention for Domestic Abusers: Intervention System Involvement and Criminal Recidivism.” Journal of Family Violence 13:263-284. NYPD. 2000. Patrol Guide. New York: New York City Police Department. Peterson, Richard R. 1989. Women, Work and Divorce. Albany, NY: State University of New York Press.

Peterson, Richard R. 2001. Comparing the Processing of Domestic Violence Cases to Non-Domestic Violence Cases in New York City Criminal Courts. New York: NYC Criminal Justice Agency, Inc. Peterson, Richard R. 2002. Cross-Borough Differences in the Processing of Domestic Violence Cases in New York City Criminal Courts. New York: NYC Criminal Justice Agency. Peterson, Richard R. 2003. The Impact of Case Processing on Re-Arrests among Domestic Violence Offenders in New York City. New York: NYC Criminal Justice Agency, Inc. Peterson, Richard R. 2004. The Impact of Manhattan’s Specialized Domestic Violence Court. New York: NYC Criminal Justice Agency. Peterson, Richard R. and Jo Dixon. 2005. “Court Oversight and Conviction under Mandatory and Nonmandatory Domestic Case Filing Policies.” Criminology and Public Policy 4:301-324. Peterson, Richard R. 2006. Pretrial Failure to Appear and Pretrial Re-Arrest Among Domestic Violence Defendants in New York City. New York: NYC Criminal Justice Agency, Inc. Peterson, Richard R. 2008a. “Reducing Intimate Partner Violence: Moving Beyond Criminal Justice Interventions.” Criminology and Public Policy 7:537-545. Peterson, Richard R. 2008b. Predicting Pretrial Misconduct Among Domestic Violence Defendants in New York City. New York: NYC Criminal Justice Agency. Redcross, Cindy, Dan Bloom, Gilda Azurdia, Janine Zweig and Nancy Pindus. 2009. Transitional Jobs for Ex-Prisoners: Implementation, Two-Year Impacts, and Costs of the Center for Employment Opportunities (CEO) Prisoner Reentry Program. New York: MDRC. Rosewater, Ann. 2003. Promoting Prevention, Targeting Teens: An Emerging Agenda to Reduce Domestic Violence. San Francisco: Family Violence Prevention Fund. Sagatun-Edwards, Inger, Eugene M. Hyman, Tracy LaFontaine, and Erin NelsonSerrano. 2003. “The Santa Clara County Juvenile Domestic and Family Violence Court.” Journal of the Center for Families, Children, and the Courts 4:91-114. Siddiqi, Qudsia. 2004. CJA’s New Release-Recommendation System. (Research Brief #5). New York: NYC Criminal Justice Agency, Inc.

SPSS, Inc. 1999. SPSS Regression Models 9.0. Chicago: SPSS, Inc. Weis, Joseph G. 1989. “Family Violence Research Methodology and Design.” Pp. 117-162 in Lloyd Ohlin and Michael Tonry (eds.), Crime and Justice: A Review of Research, Volume 11: Family Violence. Chicago: University of Chicago.

69 APPENDIX A: STATISTICAL METHODS 1. Logistic Regression Analysis This report used logistic regression analysis to predict the likelihood of conviction, the likelihood of failure to appear, and the likelihood of pretrial re-arrest. Logistic regression analysis is a statistical technique that is used when the outcome to be explained (i.e., the dependent variable) has two categories. In our analyses in Chapters 4 and 5, all cases were coded on our dependent variable in one of two categories. In the models of conviction, cases were classified as “not convicted” (dismissed cases, ACD cases and cases that ended in acquittal after trial were coded 0) or “convicted” (pleas of guilty and findings of guilty after trial were coded 1). In the models of failure to appear, cases were coded as “did not fail to appear” (coded 0) or “failed to appear for at least one scheduled court appearance” (coded 1). In the models of re-arrest, cases were coded as “re-arrested during the pretrial period” (coded 0) or “not re-arrested during the pretrial period” (coded 1). In both chapters, the predictions were made on the basis of information we have about a variety of defendant and case characteristics (i.e., the independent variables). Logistic regression techniques provide several ways of evaluating the effect of these independent variables. The current study examined three statistical measures to evaluate the effect of the independent variables on a dependent variable. First, we report the statistical significance of each independent variable. Statistical significance takes into account the size of the sample as well as the magnitude of the effect of the independent variable. Taking this information into account, statistical significance assesses the probability that the effect observed in the sample could have occurred by chance alone. In this report, following standard convention, significance levels of .05 or less were treated as statistically significant. In other words, when an effect has a 5% or less probability of having occurred by chance, we conclude that the independent variable is a statistically significant predictor of the likelihood of the outcome being considered. One weakness of using statistical significance to measure the effect of an independent variable is that when sample sizes are large (e.g., more than several thousand cases), many independent variables have statistically significant effects even when the magnitude of their effects is small. For example, in a very large sample, we may find that having prior criminal convictions or open cases has a statistically significant effect on the likelihood of pretrial re-arrest, even though the odds of pretrial re-arrest for those with prior criminal convictions or open cases are only 1.01 times larger than for those without prior criminal convictions or open cases. In this hypothetical example, we can say that the effect of having prior criminal convictions or open cases is unlikely to be due to chance. However, it is also clear that knowing whether or not a defendant had prior criminal convictions or open cases does not explain much of the variation in likelihood of pretrial re-arrest.

70 Our second statistical measure used to evaluate the effect of the independent variables is the odds ratio. The odds ratio supplements information about statistical significance by evaluating the magnitude of the effect of the independent variable. Specifically, it tells us how much the odds of an outcome (e.g., pretrial re-arrest) change, for each one-unit increase in the independent variable. If an independent variable is coded in two categories (e.g., 0 and 1), then the odds ratio tells us how the odds of the outcome change when cases are coded 1 on the independent variable (vs. cases coded 0). An odds ratio greater than one indicates an increase in the likelihood of the outcome occurring, while an odds ratio less than one indicates a decrease in the likelihood of the outcome occurring. An odds ratio of 1 indicates that the odds of an outcome occurring are not affected by the independent variable. To return to our previous example, if the odds ratio for the effect of having prior criminal convictions or open cases on the likelihood of pretrial re-arrest was 1.12, this would mean that in cases where the defendant had prior criminal convictions or open cases, the odds of pretrial re-arrest are 1.12 times greater than in cases where the defendant did not have prior criminal convictions or open cases. In contrast, if we examined the impact of whether the defendant was released on recognizance while the case was pending, we might find an odds ratio less than 1. For example, if the odds ratio was .83, this would mean that in cases where the defendant was released on recognizance, the odds of pretrial re-arrest were only .83 times as large as the odds when the defendant was released on bail. To simplify interpretation of odds ratios less than 1, it is common to examine the inverse of the odds ratio (1 divided by the odds ratio). When this is done, the interpretation of the effect of the independent variable is reversed. For example, if the odds ratio for being released on recognizance is .83, we can take the inverse of the odds ratio, 1.20 (1 divided by .83), and say that in cases where the defendant was released on bail, the odds of pretrial re-arrest were 1.20 times greater than in cases where the defendant was released on recognizance. Finally, if the odds ratio was 1.00, this would mean that whether the defendant was released on recognizance or on bail had no impact on the odds of pretrial re-arrest. (These examples are hypothetical and do not necessarily reflect our expectations about the findings.) In the analyses presented in this report, results are presented for independent variables coded in three different waysâ&#x20AC;&#x201D;categorical variables that have two categories, categorical variables that have more than two categories, and continuous variables that measure the quantity of a defendant or case characteristic (e.g., the number of prior felony convictions for the defendant). When a categorical independent variable has two categories, the odds ratio measures the change in the odds when cases are in one category vs. the other (e.g., defendant had prior criminal convictions or open cases vs. did not have prior criminal convictions or open cases). When a categorical independent variable has more than two categories, one of the categories is chosen as a reference category, and the odds ratios measure the effect of being in each of the other categories vs. being in the reference category (e.g., defendants in Manhattan are compared to defendants in Brooklyn, which is used as the reference category). Finally, when the independent variable is continuous, the odds ratio measures the change in the

71 odds associated with an increase of one unit on the scale of the independent variable (e.g., for number of arrest charges, the odds ratio measures the effect of having one additional arrest charge). Our third statistical measure used to assess the effect of the independent variables is the standardized beta coefficient (Menard 1995). The standardized beta (ď &#x201A;ď&#x20AC;Š coefficient takes into account not only the change in the likelihood of the outcome associated with a change in the independent variable, but also the distribution of the cases among the categories of the independent variable. Being in one category of an independent variable may have a large effect on the likelihood of an outcome (and therefore the variable may have a large odds ratio), but if there are relatively few cases in that category, the variable will not help to explain much of the variation in the likelihood of the outcome. For example, a defendant charged with a sex offense might have a high probability of pretrial re-arrest, and this variable would have a high odds ratio. However, if only a small number of defendants in the sample are charged with a sex offense, this variable would not be able to explain much of the variation in likelihood of pretrial re-arrest. Standardized betas measure this overall effect of the independent variable on the dependent variable. Standardized betas vary from -1 to +1; values closer to zero indicate that the effect of the independent variable is relatively small, while values closer to +1 or -1 indicate that the effect of the independent variable is relatively strong. There are no commonly accepted absolute standards to determine whether a standardized beta is strong or weak. Consequently, we discuss the relative strength of variables, describing some as stronger or weaker than others. In the current study, we used all three of the measures discussed above to evaluate the effect of the independent variables. We used the statistical significance level to distinguish those independent variables that had a detectable17 effect on the dependent variable from those that did not. We used the odds ratio to evaluate the size of the effect of the independent variable and we used the standardized beta to evaluate the ability of the independent variable to account for variation in the dependent variable. The models we discuss include a large number of predictors of the dependent variable. In these models, the measures of the effect of each independent variable (statistical significance, odds ratio, and standardized beta) evaluate the effect of that independent variable after controlling for the effects of all the other independent variables in the model. These effects represent the net effect of a given independent variable after the effect of all the other independent variables have been taken into account. This net effect differs from the total effect of the independent variable, which is the effect of the independent variable when it is used as the only predictor of the dependent variable.

Due to sampling error, and limitations of logistic regression techniques, it is possible that some independent variables that do affect the dependent variable are found to be statistically insignificant in our particular sample of cases. See Mohr (1990) for a further discussion of these issues.

72 To evaluate the overall ability of all the independent variables in the logistic regression model to predict the dependent variable, we used a statistical measure called Nagelkerke R2 (SPSS, Inc. 1999). This measure varies from 0 to +1. It can be roughly interpreted as indicating what proportion of the variation in the dependent variable is explained by all the independent variables in the model (see Menard 1995 for a full discussion of the R2 statistic in logistic regression models). Low values of R2 (closer to 0) indicate that the model as a whole is relatively weak in accounting for variation in the dependent variable. High values (closer to +1) indicate that the model as a whole is very successful in accounting for variation in the dependent variable. 2. Correcting for Selection Bias Our models predicting likelihood of pretrial failure to appear and of pretrial rearrest included a control variable to correct for selection bias. Selection bias is a problem that arises in statistical analysis when the group of cases that could have ended up in the sample is restricted to a selected set of respondents (Berk 1983). When selection bias occurs, the statistical estimates of the effects of the independent variables may be biased. These estimates may overstate, or understate, the influence of an independent variable. If problems of selection bias are not addressed, the interpretation of the results may be misleading. In our analysis, the models predicting failure to appear and pretrial re-arrest include only cases that were released at some point between arraignment and disposition. The variables that influence whether a defendant was released also may influence whether the defendant fails to appear and/or is re-arrested during the pretrial period. For example, having prior criminal convictions or open cases may affect both the likelihood of release and the likelihood of pretrial re-arrest. If a model predicting the likelihood of pretrial re-arrest among those released does not control for selection bias, the estimate of the effect of having prior criminal convictions or open cases will be biased. Part of its effect on likelihood of pretrial re-arrest will actually be due to its influence on the likelihood of release, i.e., on the likelihood that the case ended up in the sample of released cases. The remainder of its effect, if any, will be due to its influence on the likelihood of pretrial re-arrest. To control for this kind of selection bias, we included in the models a control variable that measures the predicted probability of release. This predicted probability of release was created using the logistic regression model shown in Table A-1. To avoid statistical problems18 the predicted probability of release used as a correction for selection bias was created using a somewhat different set of independent variables than was used in the models presented in Chapters 4 and 5. The predicted probability of release can theoretically vary from a low of 0.00 to a high of 1.00. Of course, the models predicting likelihood of pretrial failure to appear or pretrial re-arrest included only 18

When the predicted probability of release is included in a model predicting an outcome, such as pretrial re-arrest, it is important that the predicted probability of release not be highly correlated with other variables in the model. This problem, known as multicollinearity, is particularly likely if the same set of independent variables is used in both the release and outcome models. For this reason, the model creating the predicted probability of release uses some different independent variables than those used in the analyses presented in this report.

73 TABLE A-1 LOGISTIC REGRESSION MODEL PREDICTING LIKELIHOOD OF PRETRIAL RELEASE MALE OFFENDERS, CRIMES AGAINST PERSONS AND PROPERTY SUBSAMPLE1 Third Quarter 2005 Dataset INDEPENDENT VARIABLES

Standardized

Odds



Ratio

ARRAIGNMENT CHARGE TYPE Reference Category: Assault (PL 120) Criminal Contempt (PL 215) Harassment (PL 240) Crimes Against Children (PL 260) Other

-0.05 ** 0.05 * 0.00 -0.26 ***

0.74 1.51 0.97 0.47

-0.29 -0.25 -0.13 -0.04 0.15

*** *** *** * ***

0.44 0.47 0.67 0.87 1.55

-0.13 *** -0.09 ***

0.83 0.78

0.41 *** -0.12 *** -0.05 *

3.24 0.46 0.75

0.06 * -0.02 0.05 *

1.28 0.95 1.30

-0.08 ** -0.08 ** -0.09 **

0.78 0.76 0.74

DEFENDANT'S CRIMINAL HISTORY ANY PRIOR CRIMINAL CONVICTIONS OR OPEN CASES ANY PRIOR MISDEMEANOR CONVICTIONS ANY PRIOR FELONY CONVICTIONS ANY OPEN CASES AT TIME OF ARREST DEFENDANT HAD 2 OR MORE BENCH WARRANTS AT TIME OF ARREST CHARGE CHARACTERISTICS NUMBER OF ARREST CHARGES ARREST CHARGE IS A FELONY RELEASE RECOMMENDATION Reference Category: No Recommendation (Weak NYC Ties) Recommended or Moderate Risk Open Bench Warrant at Time of Arrest Missing DEFENDANT'S DEMOGRAPHIC CHARACTERISTICS ETHNICITY: Reference Category: Non-Hispanic Black Non-Hispanic White Hispanic Other AGE: Reference Category: Age 16-20 Age 21-29 Age 30-39 Age 40 and over Nagelkerke R2 (N of cases) NOTES 1 2

0.28 *** (12,826)

those cases where the defendant actually was released. The predicted probability of release for released cases was skewed toward the higher end of the scale. The mean predicted probability for the released cases in our analyses was .87 (data not shown). Nevertheless, even among cases where the defendant was released, there was significant variation: the predicted probability of release ranged from .24 to .99 (data not shown). It is this variation that enables the predicted probability of release to correct for selection bias. Among released cases, those with a low predicted probability of release are more similar to those who were not released, while those with a high predicted probability of release are more representative of those actually in the sample. For example, the influence of the predicted probability of release on the likelihood of failure to appear controls for the influence of variables that affect both release and failure to appear. When the predicted probability of release is included as a control variable in the model predicting the likelihood of failure to appear, the estimates of the effects of the other independent variables in the model are more accurate.19 Similarly, the influence of the predicted probability of release on the likelihood of pretrial re-arrest controls for the influence of variables that affect both release and pretrial re-arrest. Including the predicted probability of release as a control variable in the model predicting failure to appear and/or pretrial re-arrest makes the estimates of the effects of the other independent variables in the model more accurate. To return to the example discussed above, whether the defendant had prior criminal convictions or open cases may influence both the likelihood of release and the likelihood of pretrial re-arrest. In a model predicting the likelihood of pretrial re-arrest, the estimate of the effect of prior criminal convictions or open cases is more accurate because the model controls for the influence of prior criminal convictions or open cases on the likelihood of release.20 As a result, we have greater confidence in our estimates of the effects of this and other independent variables, as well as in our interpretation of the results of the model. Although we included the predicted probability of release as a control variable in the model, we did not discuss the impact of this variable since its primary purpose was to enable us to estimate accurately, and to interpret, the effects of the other variables.

See Heckman (1979) and Peterson (1989) for a more detailed discussion of selection bias and corrections for it. 20 The extent to which the estimates are more accurate depends on the ability of the model predicting the probability of release to explain a significant portion of the variation in likelihood of release. Our model was reasonably successful at explaining variation in likelihood of release. The model correcting for selection bias accounted for approximately 28% of the variation in likelihood of release.

75 APPENDIX B: DISTRIBUTION OF VARIABLES FOR REGRESSION MODELS (Third Quarter 2005 Dataset, Released Young Male Defendants in Defendant-Based File, N = 868)

VARIABLES AND DISTRIBUTION DEPENDENT VARIABLES DEFENDANT EVER FAILED TO APPEAR Defendant never failed to appear Defendant failed to appear for at least one court appearance Total, all cases DEFENDANT EVER RE-ARRESTED FOR A NEW DV OFFENSE PRIOR TO CASE DISPOSITION Defendant not re-arrested for a new DV offense prior to case disposition Defendant was re-arrested for a new DV offense prior to case disposition Total, all cases

83% 17% 100%

86% 14% 100%

CONTROL VARIABLES MEAN PREDICTED PROBABILITY OF RELEASE MEAN NUMBER OF DAYS AT RISK OF PRETRIAL FAILURE TO APPEAR

0.92 113

MEAN NUMBER OF DAYS AT RISK OF PRETRIAL RE-ARREST

121

DEFENDANT'S CRIMINAL HISTORY ANY PRIOR CRIMINAL CONVICTIONS OR OPEN CASES No Yes Total

46% 54% 100%

ANY PRIOR MISDEMEANOR CONVICTIONS No Yes Total

82% 18% 100%

ANY PRIOR FELONY CONVICTIONS No Yes Total

87% 13% 100%

NUMBER OF PRIOR MISDEMEANOR CONVICTIONS None 1 2 3 4 5 or more Total

83% 10% 3% 1% 1% 1% 100%1

NUMBER OF PRIOR FELONY CONVICTIONS None 1 2 or more Total

89% 9% 2% 100%

NOTE Percentages do not sum to 100% due to rounding.

Table Continues on Next Page

76 APPENDIX B: DISTRIBUTION OF VARIABLES FOR REGRESSION MODELS (continued) VARIABLES AND DISTRIBUTION NUMBER OF OPEN CASES None 1 2 or more Total

80% 16% 4% 100%

DEFENDANT HAD PRIOR BENCH WARRANTS AT TIME OF ARREST Defendant did not have prior bench warrants at time of arrest Defendant had prior bench warrants at time of arrest Total

62% 38% 100%

DEFENDANT'S ARREST HISTORY IN PRIOR TWO YEARS ANY ARRESTS IN PRIOR TWO YEARS No Yes Total

42% 58% 100%

ANY DV ARRESTS IN PRIOR TWO YEARS No Yes Total

76% 24% 100%

ANY NON-DV ARRESTS IN PRIOR TWO YEARS No Yes Total

51% 49% 100%

DEFENDANT CONVICTED IN DV CASE IN PRIOR TWO YEARS No prior DV case or not convicted in DV case in prior two years Convicted in DV case in prior two years Total

91% 9% 100%

RELEASE RECOMMENDATION: Recommended or Moderate Risk Not recommended: High Risk for FTA Open Bench Warrant At Time of Arrest Other or Missing Total DEFENDANT'S COMMUNITY TIES

70% 23% 4% 3% 100%

ENGAGED IN FULL-TIME ACTIVITY Employed, in school, or in a training program full time Not employed in school or in a training program full time Total

57% 43% 100%

AT CURRENT ADDRESS 1 YEAR OR LESS Defendant at current address more than 1 year Defendant at current address 1 year or less Total

71% 29% 100%

LIVES WITH SOMEONE No Yes Total

22% 78% 100% Table Continues on Next Page

77 APPENDIX B: DISTRIBUTION OF VARIABLES FOR REGRESSION MODELS (continued) VARIABLES AND DISTRIBUTION DOES NOT EXPECT ANYONE AT ARRAIGNMENT Expects someone at arraignment Does not expect anyone at arraignment Total

42% 58% 100%

HAS NO TELEPHONE OR CELL PHONE Has a telephone or cell phone Has no telephone or cell phone Total

77% 23% 100%

LIVES OUTSIDE NEW YORK CITY AREA Lives in New York City area Lives outside New York City area Total

98% 2% 100%

CHARGE CHARACTERISTICS MEAN NUMBER OF ARREST CHARGES

1.81

ARRAIGNMENT CHARGE TYPE Assault (PL 120) Criminal contempt (PL 215) Other Total

67% 11% 22% 100%

ARRAIGNMENT CHARGE SEVERITY B Felony C Felony D Felony E Felony A Misdemeanor B Misdemeanor Violation Total

1% 1% 5% 3% 85% 4% 1% 100%

RELEASE CHARACTERISTICS STAGE OF RELEASE Released at arraignment Released after arraignment Total

79% 21% 100%

TYPE OF RELEASE Released on recognizance Released on bail Total

86% 14% 100%

CASE PROCESSING TIME MEAN NUMBER OF WEEKS FROM ARRAIGNMENT TO DISPOSITION Table Continues on Next Page

13.40

78 APPENDIX B: DISTRIBUTION OF VARIABLES FOR REGRESSION MODELS (continued) VARIABLES AND DISTRIBUTION DEFENDANT'S DEMOGRAPHIC CHARACTERISTICS BOROUGH Brooklyn Manhattan Queens Staten Island Bronx Total

36% 18% 20% 3% 23% 100%

ETHNICITY Non-Hispanic Black Non-Hispanic White Hispanic Other Total

49% 6% 41% 4% 100%

AGE Age 16-20 Age 21-24 Total

41% 59% 100%

OFFENDER-VICTIM RELATIONSHIP Boyfriend-girlfriend Married Common-law marriage Other relationship Missing Total NOTE Percentages do not sum to 100% due to rounding.

37% 8% 20% 25% 11% 100%1

APPENDIX C: CODING OF VARIABLES FOR REGRESSION MODELS

CODING SCHEME

VARIABLES DEPENDENT VARIABLES

DEFENDANT FAILED TO APPEAR DEFENDANT RE-ARRESTED FOR A NEW DV OFFENSE PRIOR TO CASE DISPOSITION

Ever failed to appear = 1, Never failed to appear = 0 Ever re-arrested = 1, Never re-arrested = 0

CONTROL VARIABLES SELECTION BIAS CORRECTION: LIKELIHOOD OF RELEASE

Continuous, ranges from 0.00 to 1.00

TIME AT RISK OF PRETRIAL MISCONDUCT

Number of days

DEFENDANT'S CRIMINAL HISTORY ANY PRIOR CRIMINAL CONVICTIONS OR OPEN CASES

Any prior criminal convictions or open cases = 1, All other categories = 0

ANY PRIOR MISDEMEANOR CONVICTIONS

Any prior misdemeanor convictions = 1, All other categories = 0

ANY PRIOR FELONY CONVICTIONS

Any prior felony convictions = 1, All other categories = 0

NUMBER OF PRIOR MISDEMEANOR CONVICTIONS

None = 0, One = 1, Two = 2, Three = 3, Four = 4, Five = 5, Six = 6, Seven = 7 or more

NUMBER OF PRIOR FELONY CONVICTIONS

None = 0, One = 1, Two = 2, Three = 3 or more

NUMBER OF OPEN CASES

None = 0, One = 1, Two = 2 or more

DEFENDANT HAD PRIOR BENCH WARRANTS AT TIME OF ARREST

Defendant had prior bench warrants at time of arrest = 1, All other categories = 0

DEFENDANT'S ARREST HISTORY IN PRIOR TWO YEARS ANY ARRESTS IN PRIOR TWO YEARS

Any arrests in prior 2 years = 1, All other categories = 0

ANY DV ARRESTS IN PRIOR TWO YEARS

Any DV arrests in prior 2 years = 1, All other categories =0

ANY NON-DV ARRESTS IN PRIOR TWO YEARS

Any Non-DV arrests in prior 2 years = 1, All other categories = 0 Defendant convicted in DV case in prior two years = 1, All other categories = 0

DEFENDANT CONVICTED IN DV CASE IN PRIOR TWO YEARS RELEASE RECOMMENDATION: Reference Category: Not Recommended: High Risk of FTA Recommended or Moderate Risk Open Bench Warrant At Time of Arrest Other or Missing

Not Recommended: High Risk of FTA: Reference Category Recommended or moderate risk = 1, All other categories = 0 Open bench warrant = 1, All other categories = 0 Other or missing = 1, All other categories = 0

DEFENDANT'S COMMUNITY TIES ENGAGED IN FULL-TIME ACTIVITY

Employed, in school and/or in training program full time = 1, All other categories = 0

AT CURRENT ADDRESS 1 YEAR OR LESS

At current address 1 year or less = 1, All other categories = 0

LIVES WITH SOMEONE

Lives with someone = 1, All other categories = 0

DOES NOT EXPECT ANYONE AT ARRAIGNMENT

Does not expect anyone at arraignment = 1, All other categories = 0

HAS NO TELEPHONE OR CELL PHONE

Has no telephone or cell phone = 1, All other categories = 0

LIVES OUTSIDE NEW YORK CITY AREA

Lives outside New York City area = 1, All other categories = 0

Table Continues on Next Page

APPENDIX C: CODING OF VARIABLES FOR REGRESSION MODELS (continued)

CODING SCHEME

VARIABLES CHARGE CHARACTERISTICS

NUMBER OF ARREST CHARGES ARRAIGNMENT CHARGE TYPE Reference Category: Assault (PL 120) Criminal Contempt (PL 215) Other

Number of charges, ranges from 1 to 4

ARRAIGNMENT CHARGE SEVERITY

A Felony = 1, B Felony = 2, C Felony = 3, D Felony = 4, E Felony = 5, A Misdemeanor = 6, B Misdemeanor = 7, Unclassified Misdemeanor = 8, Violation = 9

Assault: Reference Category Criminal Contempt = 1, All other categories = 0 Other = 1, All other categories = 0

RELEASE CHARACTERISTICS STAGE OF RELEASE

Released at arraignment = 1, Released after arraignment = 0

TYPE OF RELEASE CASE PROCESSING TIME NUMBER OF WEEKS FROM ARRAIGNMENT TO DISPOSITION DEFENDANT'S DEMOGRAPHIC CHARACTERISTICS

Bail = 1, ROR = 0 Number of weeks, ranges from 0 to 84

BOROUGH Reference Category: Brooklyn Manhattan Queens Staten Island Bronx

Brooklyn: Reference Category Manhattan = 1, All other categories = 0 Queens = 1, All other categories = 0 Staten Island = 1, All other categories = 0 Bronx = 1, All other categories = 0

ETHNICITY: Reference Category: Non-Hispanic Black Non-Hispanic White Hispanic Other

Non-Hispanic Black: Reference Category Non-Hispanic White = 1, All other categories = 0 Hispanic = 1, All other categories = 0 Other = 1, All other categories = 0

AGE: Reference Category: Age 16-20 Age 21-24

Age 16-20: Reference Category Age 21-24 = 1, All other categories = 0

OFFENDER-VICTIM RELATIONSHIP Reference Category: Married Boyfriend-girlfriend Common-law marriage Other relationship Missing

NOTE 1

See text for a description of the variables in the models.

Married: Reference Category Boyfriend-girlfriend = 1, All other categories = 0 Common-law marriage = 1, All other categories = 0 Other relationship = 1, All other categories = 0 Missing = 1, All other categories = 0