Workshop Summary: Addressing Predisposition Revictimization in Cases of Violence Against Women

Wyndham Washington Hotel, Washington, DC
January 31–February 1, 2005

Table of Contents

Opening Comment by Thomas E. Feucht, Acting Assistant Director for Research and Evaluation, National Institute of Justice, U.S. Department of Justice

Introduction and plan for the 1 ½ day workshop

Luncheon Presentation

Day 2

A List of the Workshop Attendees


Opening Comment by Thomas E. Feucht, Acting Assistant Director for Research and Evaluation, National Institute of Justice, U.S. Department of Justice

Thomas Feucht welcomed workshop attendees on behalf of the Director of the National Institute of Justice (NIJ) and the cosponsoring agency, Office on Violence Against Women (OVW). The Office of Justice Programs (OJP) is an umbrella agency to a group of departments with connected service areas, and Feucht noted that staffs are pleased when an issue falls across common missions. There has not been enough research about victimization in the time period of predisposition (postarrest and prior to trial and/or sentencing). Estimates of rearrest of defendants vary. NIJ wants to learn from the people who work day to day with the issues (police, court professionals, victim advocates, researchers) in order to better uncover the various dimensions of the problem and promising directions for programs and research.

NIJ brought this attendee group together to examine strategies, policies and principles currently in place and focus directions for research. The topic is large and varied; Dr. Feucht asked participants to trade ideas and insight in order to try to keep victims safer in the predisposition period. The Institute is interested in long- and short-term answers that may help people wrestling with daily problems in this area. He encouraged a candid conversation to bring out a broad sense of the issues and to indicate a few good, testable research questions.

Back to Top

Introduction and plan for the 1½-day workshop

Spurgeon Kennedy, Operations Deputy Director, Court Services, Supervision, and Treatment, District of Columbia (DC) Pretrial Services Agency

Mr. Kennedy, moderator for the meeting, led mutual introductions among participants. Persons attending the meeting filled a broad spectrum of profiles:

  • Administrators for batterers' intervention programs
  • Victim advocates and community-based program leaders
  • A municipal judge
  • A children's advocate
  • Public defense counsel
  • Public health research staff
  • Researchers
  • Police officials
  • Prosecutors
  • Probation and parole officials for domestic-violence related offenders
  • Federal research staff
  • A pretrial services agency official

Back to Top

Pretrial Misconduct by Domestic Violence-Charged Defendants in the District of Columbia

Mr. Kennedy described a brief review he conducted of pretrial services data in DC. The District's pretrial services agency has automated data sources, good demographic information, and data for assessing the need for substance abuse treatment. Data from neighboring jurisdictions (Prince George's County, Montgomery County) are usually available. Police identification numbers may allow connections for certain individuals, but databases are kept separately due to rules of confidentiality. The separate databases refer to the following: the pretrial services agency database, the U.S. Attorney's database, the court database, the jail database, and the Metropolitan Police Department database. Social and criminal history information is usually available for released defendants.

In his review of pretrial services records, Mr. Kennedy asked whether domestic violence (DV) offenders in DC were rearrested sooner than other kinds of defendants, in particular during the period of pretrial supervision. Based on data from October 2003 to September 2004, people with DV charges were rearrested more quickly. Mr. Kennedy cautioned that the findings should be regarded as "only a snapshot" in a short-term review, not a trend analysis. Records in DC concerning DV-charged defendants in that timeframe revealed:

  • Lower rearrest rates (However, the difference was not great given the rate for domestic violence was 11.8 percent compared to an overall rate of 13.9 percent), but persons who were rearrested were arrested faster. The figures presented by Mr. Kennedy were 123 days for DVcases compared with 223 days for other types of cases.
  • Tendency to commit other DV-related offenses or person crimes (this was often related to drug possession or distribution). (A third of the new arrests were for DV offenses. A high proportion of defendants accused of DV offenses were rearrested for drug charges that may or may not be related to a DV incident).
  • New charges that were not solely with the same victim. (Mr. Kennedy made a point that the victim is unknown therefore none of the statistics offered any insights about whether the "rearrests" involved the same or a different victim).
  • Alcohol, drug, or "anger issues."
  • Lack of social, demographic, or other data that were distinct between groups, except for the frequent relation to drug offenses

There is a special unit in the jurisdiction for defendants with mental health problems, and of these type of offenders, a larger percentage than expected have been involved in domestic violence. Mr. Kennedy noted that alcohol and substance abuse in connection with domestic violence "is off the charts" (very high). Pretrial services in DC has authority to test for drug and alcohol use during pretrial and probation. About 47 percent of arrested defendants test positive for drug use, even higher when the question's response includes "positive at some time during supervision."

In presenting this information, Mr. Kennedy's main points were that DC has certain advantages in providing information on domestic violence cases. These include: (1) special statutes identifying offenses as being DV offenses; (2) a comprehensive database of offense categories, defendant characteristics, and court events; and (3) the ability to generate statistical reports. He also mentioned some flaws or shortcomings of DC as far as measuring revictimization: (1) unless the offense is a DV-specific offense, it is impossible to discern (and hence measure) if an assault or a property crime or any other crime involved a DV-related victim and defendant; and (2) the pretrial services database is not integrated with other criminal justice agency databases although there is some access to those databases. Another flaw, which is not specific to DV, is that only new arrests are captured.

Back to Top

Open Discussion

Participants discussed other experiences with rearrest statistics. Predisposition rearrest information from DC was provided as an example: It was known that 4 percent of persons who are arrested for a DV offense and get out on release are arrested for a new DV offense. This of course does not answer the question of whether non-DV offenses were nonetheless DV-related (e.g., property and drug offenses). Definitions of the arrest charge vary and are sometimes unclear. For example, DC has a charge defined as "violation of condition of release," which may or may not relate to an additional domestic violence charge. People were interested in finding misconduct that is "predictive" of rearrest and whether cases relating to domestic violence remain in the system for a shorter time.

Among defendants rearrested in DC, about one-third involved arrests relating to new domestic violence charges, such as assault, violation of a civil protection order (PO), attempted threat, or destruction of property. According to information from the Metropolitan Police Department and the U.S. Attorney's Office, the rearrests mostly happened in a short time frame. In addition, 37.8 percent of felony arrests involved drug possession or distribution charges and 35.8 percent of the cases listed "assault" as the top charge.

Several people asked about the victims' influence or input to the justice system and whether victim information was recorded and used in handling the offender. One man asked if "victims' laws" were applied to domestic violence cases or whether prosecutors tended to overlook that. There may be homicides (or other crimes) that relate to domestic violence charges, but that is not recorded in a clear way. It is a significant gap in "the filing process" to leave out victims' information. Some speakers felt that victim interviewing was beginning to occur more often. One participant, a public defender, noted that cases are sometimes "overcharged" and then changed. For example, a person brought in under felony charges might have the charge downgraded. (However, this may or may not be an example of overcharging. Downgrading is often used as a device of plea bargaining.) Mr. Kennedy noted that, in DC, "new person crimes," which are recorded in a rearrest, do not necessarily include victim information. Annually, in about 20 to 23 percent of DC domestic violence cases, U.S. Attorneys decide to "no-paper" the charges, in which case the defendant may leave on the same day.

NIJ staff suggested that the discussants turn to issues that will help researchers to understand the extent of the problem in a variety of jurisdictions, not only DC. Another participant noted that a recent risk assessment study showed (by self-report) that 31 percent of women who experience intimate partner violence (IPV) have been reassaulted. This is four or five times higher than official survey statistics. (This statement does not address other jurisdictions but it does address the fact that "rearrest" information is only one measure or definition of revictimization. Obviously another definition is actual "reoffenses," which cannot be measured by rearrest statistics but rather by other means, such as self-reports.) In response to NIJ's suggestion, participants commented on the issue of what constitutes a domestic violence offense. It was pointed out that cases of homicide increasingly involve offenders who had no prior contact with the criminal justice system. In addition, there may be other kinds of offenses (theft, vandalism) that actually relate to domestic violence.

One participant emphasized the quality of first response to a victim. If this response is not good (from the part of police, advocates, or the judge), then the victim will not turn again to the system for protection against victimization. Another speaker, who was a prosecutor, noted that failure to include victims' information during the filing process is "a big gap in service." During the process of securing a protective order or anywhere else that the victim first contacts criminal justice actors, quality of response makes an impact. A participant mentioned that cross-jurisdictional treatment has been extremely varied. These have different dynamics and should be broken out to consider what "access points" to the criminal justice system play a role (there is some contact related to offending which is not yet an "arrest") for different kinds of victims (elder abuse, child victim, etc.). There may be significant differences between urban and rural jurisdictions. The participant from the judiciary asked about research to uncover information on "victims who choose to stay away"; what acts occur in the justice system (court, probation, police officers, etc.) to discourage victims from further contact or participation in programs?

Participants discussed definitions of victimization and family violence that took forms other than between intimate partners. Definitions of victimization must be carefully framed. Stalking, telephone threats from a jail, even flowers sent to a workplace (from a batterer) could be forms of victimization. One participant noted the need to augment many existing databases with supplemental interviewing. Additions could include, for example, how often the Immigration and Naturalization Service (INS) has been involved in cases of revictimization. Many administrations are experiencing shrinking budgets, sometimes after spending millions on programs that "do not work" according to research. Cost-benefit analysis could show at the front end what financial and public safety costs are involved in revictimization and its impact on children and families.

Participants continued to raise issues that reflected the challenge in obtaining a clear understanding of the problem of predisposition revictimization. A participant asked at what point do reoffenses come to the attention of the criminal justice system? Are child custody circumstances bringing the victim into contact with the perpetrator? To what extent do subsequent arrests happen when victims and offenders have been together again (are they usually married when this happens)? Interviewing persons who have been victimized is often very difficult. In telephone interviews involving the National Crime Victim Survey (NCVS), about 26 percent of women who were victimized said that they never reported the incident. In this survey, performed primarily by telephone (2/3 of all respondents), about 30 percent of those who were victimized by their intimate partners also said they were repeat victims. Rearrest does not seem to be a sufficient indicator of the extent of predisposition revictimization, although it is the most readily accessible data source, but locating and contacting victims has many obstacles including confidentiality requirements. Attendees discussed female perpetrators of domestic violence and abused women who are in the criminal justice system. Most women (78 percent) who have been incarcerated for committing domestic violence feel that they have themselves been abused by their partners. Participants who were from the corrections field said that they often use "victims' manuals" to determine how to assess service needs for such women. Another participant recommended considering the dimensions of racial and cultural norms in the patterns of violence. In definitions of "victimization," is a woman who has been "pulled into the system" for her own reaction no longer a victim? What kind of treatment should be offered to her? Many cross-connections exist to drug dependency, mental health problems, or alcohol dependence. Participants noted that research would benefit from taking a closer look at the connection between the civil and criminal court processes. Do protection orders from the civil system show up in or affect the criminal process? What "early release" policies work best to protect victims and their children?

On the offenders' side, less than 20 percent of defendants are willing to come back and talk about what is effective (for breaking a pattern of abuse). Definitions of violence are also disputed ("push," "shove," "slap," etc.) How can mixed perceptions of what constitutes violence be resolved in order to reduce victimization? There is concern that the criminal justice system is discouraging repeat victims from coming to get help. Is the arrestee a batterer, or a batterer and victim? A representative from the victims' advocacy community said that 40 to 50 percent of abused women return repeatedly to the shelter because when they go home, they are victimized again. Women often do not want to report this; the man may be the sole provider, and there is fear that there will be bad consequences for the children. There are many implications for the juvenile justice system as well. Research has shown that kids charged with violent crime often had a much higher exposure to violence in the home.

Back to Top

Luncheon Presentation

Jolanta Juszkiewicz, Ph.D., Deputy Director, Pretrial Services Resource Center, Washington, DC: Problems and Challenges with Revictimization Research

Dr. Juszkiewicz noted that statistics can shed light on this complex topic and that the justice system has come a long way in a short time relative to domestic violence. Large statistical systems, though limited, serve as beacons to find the best research methodology for analyzing key problems and challenges. However, people must ask the right questions and look for real-world measures that are practical, which apply to chosen operational definitions and terms, such as "rearrest," "violence," and "revictimization."

She spoke about the difficulties of resolving records problems arising in various jurisdictions. Unlike electronic databases that can store virtually an infinite amount of information, if one must resort to paper files there are some obstacles that must be overcome. The foremost issue is access. In some jurisdictions, such as Cook County, Illinois (the Chicago metropolitan area) where the volume of cases is great, prosecutor offices resort to shredding much of the information, so timing of research is of the essence. (Chicago is part of Cook County but there are many other municipalities besides Chicago that are part of Cook County.) For example, in Cook County, the courts have more than 50,000 misdemeanor cases each month connected to domestic violence. In other jurisdictions, like New York State, a case defined as a "nonconviction" is legally sealed within 5 days and primary data are no longer available. So research planners have to sort out available information in view of confidentiality rules and carefully examine how operations are defined and interpreted.

If revictimization is defined as persons "again falling victim to their aggressor" during the predisposition time period, the search through law enforcement records for evidence of such an event remains complex. Even if one wanted to continue to rely on arrest information, which is clearly not the best measure of reoffense, there remain certain issues. If one seeks detailed information about an arrest, relying on law enforcement records may prove daunting. For example, one county may have more than 40 law enforcement agencies. Resorting to other sources of information may not be much of an improvement. For example, State criminal history repositories often have missing information and by the nature of their mandate do not collect what is considered by the DV researcher to be valuable information. Location of records generally depends on whether the offense was classed as a felony or misdemeanor. Felonies, for example, are bound over to central court. Logs from 911 calls are not easily used (this example was given as an alternative measure to arrests, namely reported incidents), prosecutors' "charge papers" do not match geographic descriptors used by county administration or the police, and often there is no knowledge in the neighborhood of the violent character of the offender-victim relationship.

The pilot project attempts to establish the local broad-based need relative to the problem of revictimization in the predisposition time frame. Access to open cases has not been allowed, so the project uses a huge paper-record storage facility, Iron Mountain. Access to open files for research purposes is limited, sometimes not available at all. On the other hand, there may be problems with closed files. In some instances, closed cases are sent to a storage facility offsite. For example, in this pilot project, Iron Mountain is a company that was used to store files offsite for a rental fee. Access to these files is sometimes problematic because it may take a long time for the storage company to find the appropriate file; there may be a cost involved. Among 26 jurisdictions that agreed and have been screened for participation, the study is using "incident-based" key questions to connect information. The study entails collecting two types of information, incident-based information and case processing information. These two types of information will be connected to show how a particular incident became a case and the outcome of the case. Finally, 16 jurisdictions were chosen in 8 different States (Arizona, California, Florida, Georgia, Indiana, Ohio, Tennessee, and Texas). Definitions in the study have been compiled from varied statutes driving the procedures in these places, with some adjustments. (The study used an overarching definition of domestic violence so that all the disparate definitions would be captured by the study.) For example, the research project wants to consider violence in a "dating relationship" as a domestic offense, but some jurisdictions have not supported this definition. Sometimes a charge of "property damage" could actually be intimidation, depending on the domestic context of the offense.

Special populations, such as the aged or handicapped, were also discussed. One difficult dimension of the pilot project has been finding special populations to include them in the study. Even where a local jurisdiction has a dedicated domestic violence prosecution unit, many things are not defined similarly between jurisdictions. That is, a DV prosecution unit in one jurisdiction may handle different cases than a DV unit in another jurisdiction. For example, in one jurisdiction the DV unit may handle only cases involving intimate partner offenses, whereas a DV unit in another jurisdiction handles cases involving all family members who are related by blood or marriage. Examples were given of some jurisdictions where there was a DV unit, a child abuse unit, and an elder abuse unit. For the pilot study the research had to identify all of the DV cases defined broadly to include offenses involving persons who are related through blood, marriage, or affinity. To do so, all of the DV unit cases would be studied, as well as those child abuse and elder abuse cases that involved persons who were related as defined above. The difficulty is that the child abuse and elder abuse units handle all child abuse and elder abuse cases, not only those cases involving related individuals.

In conclusion, Ms. Juszkiewicz emphasized investing significant project resources in careful planning (as much as 60 percent). In such a difficult undertaking, research can shed light on key questions, but the researchers need to go out to the various people and programs in specific locales for a complete understanding of the terms, processes, and procedures used in identifying, tracking, prosecuting, and adjudicating DV cases.

Back to Top

Open Discussion: Responding to the Problem

One participant noted that, throughout the variety of settings in which a response to predisposition violence may be needed, the demographics of the offender must also be kept under consideration. There can be mental health diagnoses such as posttraumatic stress disorder (PTSD), other ethnic influences, and immigration consequences (for family members or the offender). Gathering data for a better response involves cooperative design among different agencies at local, State, and Federal levels.

The victim's view of the relationship and perspective on the criminal justice system should be clarified. How do they view the system's first responses, the reoffending person, or the effect of protection orders? How many "actors" in the justice system have come in contact with the victim? Mr. Kennedy asked the group whether agencies that were responsible for diagnostics during the predisposition time period keep or have access to background data on the offender. Could this uncover a need for monitoring or a stay-away order? Only a few jurisdictions represented at the meeting said that they had this kind of casework. The judge at the meeting noted that many jurisdictions are cautious about defendants' rights and prefer to keep only cryptic information concerning an offender prior to any conviction. Mr. Kennedy noted that a significant number of offenders in the DC jurisdiction have less supervision after they are sentenced than they had during pretrial.

Several people spoke about law enforcement's role relative to victims. An advocates' group representative demonstrated some of the "tools" they give officers during rollcall training in order to prepare them to brief domestic violence victims concerning safety for themselves, counseling, and leaving an abusive relationship. Police officers are a good source of information concerning victims to whom they have responded. "Being trained" on domestic violence topics is not the same thing as applying what they have heard. They often need to apply what they have learned in such training. Sometimes officers use discretion to call domestic disputes "disorderly conduct" to avoid mandatory arrest requirements.

A State police representative noted that law enforcement training budgets have decreased in almost all areas. There are many catches on Federal spending, but agencies do recognize the need for greater focus on domestic and youth violence. The participant from Michigan noted that police in his area do mini-risk assessments in connection with recommendations for bond or bail. This has been very successful. Also, he suggested looking to shelters for information about women who have been revictimized. The shelters generally offer safety planning and help women with court hearings, although information on specific individuals is confidential.

Mr. Kennedy asked the group: What happens when bond has been paid? Participants responded that money is no guarantee that behavior will be controlled. Nonprofit or advocacy organizations might have some answers for this question. Many victims call jails or prosecutors to find help. They may have no place to go during the predisposition time period. Organizations would benefit from having the needed information ready and knowing what to tell a victim.

Drug testing of offenders provides a lot of opportunities for supervision. Should victims be alerted if the offender or probationer fails to show up at treatment? Mr. Kennedy said that the DC pretrial program allows increase of bail, assignment to a judicial oversight program, drug monitoring, or other measures if the risk is considered to warrant the action.

Several people noted the need for research on the use and effectiveness of protection orders. Some areas enforce "no-contact orders" very seriously, with civil penalties, increased cash bail, or house arrest. Other areas have had conflicts of interest in bail-bond decisions or failure to communicate thoroughly with the offender about the meaning of the order. Perpetrators have to understand the meaning of the restraining order and receive information on services that are available to them. An attendee from Dallas said their agency has a team for responding to PO violators (Violation of Emergency Protective Order), and team members perform random checks. When the subject is out, they ask if he has returned or contacted the victim. Officers can issue an arrest warrant if the "no-contact order" is violated. This option is valuable in cases where the victim is too intimidated to call.

Though some thought that the day-to-day response of law enforcement to victims of domestic violence was not adequate to ensure victim safety, other areas (Los Angeles law enforcement agencies were cited) would go so far as to sometimes help an abused woman move if necessary. Most speakers agreed that a broader civil system of support was needed. Coercive power from the court can sometimes be used creatively to drive participation of other social partners. Another speaker noted that the woman's wishes must remain part of balancing the prosecution goals for holding offenders accountable and victims' needs. Her participation in decisions is critical. Women have been arrested for violating their own "stay away" orders, for example when children's custody is involved.

The group discussed circumstances of sexual assault by strangers and trafficking in persons across State lines for prostitution. How should this kind of criminal activity be "upgraded" to reflect victimization events that occur in that context? Many women who are in the criminal justice system have been abused and then fall into a series of increasing criminal activity (such as bad check writing, drug or alcohol abuse, theft). The partner who originally abused the woman might be continuing to cause intimidation by appearing at court hearings or threatening witnesses.

In defining victimization, stalking, technology-aided stalking, and identity theft should be considered among forms of intimidation. Some kinds of intimidation are very difficult to note (such as phone calls made from a jail by the offender). One participant quoted disturbing results from a national women's study, saying that threat or intimidation often causes a woman to drop sexual assault charges. Additionally, even the recognition of an act as "sexual assault" may depend on very specific descriptions of particular acts against the person's will. Research needs to look into the overlap with the child abuse and neglect field. Cases of incest have grown alarmingly in one participant's jurisdiction. Sometimes a woman is trying to get children out of abusive situations. In another scenario, women may come across State lines when they have been told about "modeling jobs" and then have few resources for help in circumstances of sexual coercion or other abuse. An attendee noted that battered women have been charged with parental kidnapping at an increased rate. Immigration concerns might play a role.

Sexual abuse at campuses and universities warrants more study. A University of Kentucky study indicated that about one out of four women on campuses reported being raped by "someone they knew." Rape continues to be underreported at most institutions, although the Cleary Act has mandated standardized training (widely implemented) against sexual assault and abuse. Another disturbing trend is the growing number of intimate homicides by perpetrators who have had no previous contact with the criminal justice system at all. Neither have they had any contact with related court or social agencies. Is the profile of domestic violence-related homicides changing? Speakers noted that at least half of this kind of homicide occurred with criminals who had no prior crime history or only some record of "small crimes."

One person asked about law enforcement's postarrest role during the predisposition period. One person noted that simply documenting the danger and pointing the victim to available resources might be the best response. Officers often do things informally, such as revisiting the neighbors or contacting community-based advocates. Another person mentioned offenders coming out of batterers' programs need to "leave with something" that will convince them to apply what they have learned in treatment. This might mean tools that influence "how they define themselves," to inhibit further abuse. One man, experienced with children's issues in courts, said that innovative partnership programs, perhaps faith-based, are needed to deal with offenders and motivate them to choose to apply what they have learned on behavior changes. For example, the Dorchester program (Massachusetts) on fatherhood, bringing the child's perspective to the offender, has helped people understand the impact of their conduct.

One person suggested research on civil restraining orders and child protective orders; data may be captured from those sources, which would support more coordinated response to members of the same family. The judge noted that general attitudes in a community are often reflected by the way the bench handles cases. A number of people supported the idea of gathering information on services outside the criminal justice programs that enhance victim safety and reduce reoffending. This could include activities such as programs that put advocates in shelters or other places where victims go (animal shelters temporarily helping their pets, for example). It could focus on education about the social context of domestic violence. Criminal justice system responses must reflect an effort to help, rather than actions that are "revictimizing." Several people emphasized recording the victim's perceptions.

Faith-based assistance or nonprofit organizations that help immigrants may reach young foreign-born women better. Immigrants who do not trust the (government) criminal justice system will not report abuse. Families may need to be educated and talk together about power and control issues, in a different way. An inexpensive place such as a community center, which has affiliated groups such as Al-Anon and health programs, could fill this role and allow family members to be together in a place where violence cannot occur. Women who have enrolled in detoxification centers for help with substance abuse problems might be available for cross training or information relative to domestic violence. Outreach should also extend to workplaces. Employee assistance services can promote (for example) tolerance for missing some work in order to leave an abusive relationship.

A majority of revictimization cases relate to custody issues. [Civil] family courts have information on custody and visitation, which can be valuable for framing strategies to stop violence. People of all educational levels have been recorded with problems of violence in the home. Thorough approaches, which address "mindset," are necessary to success. Dorchester's "Close to Home" project uses focus groups among neighbors, friends, and block policing volunteers to help women go to court and leave domestic abusers. In a Chicago educational program discussed by attendees, about 1,500 people came together on one weekend. Another good educational program targets middle school boys with information and training on gender and conflict resolution. Community "witness" programs have been helpful to encourage men who are completing batterers' treatment. The community supports the man's pledge and works out definitions of accountability. Community or volunteer meetings reduce "denial" about domestic violence and can be a source of knowledge. One dentists' program mentioned by participants encourages reporting of suspect events, such as a woman who has had teeth knocked out.

NIJ staff returned to particular questions:

  • To what extent is predisposition violence a problem? There are conflicting numbers inside and outside the system.
  • What is a good term for predisposition revictimization; is it "reoffense", "intimidation", "stalking"?
  • Does some victimization go unnoticed hence unmeasured (e.g. victimization after the end of a case)?
  • What does the "community role" mean?
  • What civil justice mechanisms might be helpful for improving criminal justice responses?

Back to Top

Day 2

Mary Louise Kelley, Evaluation Specialist, Office on Violence Against Women, U.S. Department of Justice

Through their partnership, NIJ and OVW want to particularly examine this "predisposition" time-phase to identify gaps in service for victims. OVW discretionary projects reach out to victims during both pretrial and probation periods, follow-up on police reports, and help to link victims to social services. The agency explores what people are successfully finding in the research, and Ms. Kelley noted that staff want to look beyond the justice system as well to use creative community solutions for diverse clients. The leadership looks to research for options and knowledge concerning assessment, monitoring strategies, batterer accountability, and survivor assistance.

Back to Top

Breakout Sessions

Small groups at the conference worked together to assemble key questions for Federal research. NIJ directed the group to frame the discussion into several general areas for research planning:

  • Scope. What revictimization offenses should be considered? What actions qualify as victimization? Some examples in the discussion included stalking, human trafficking (prostitution), and sexual assault. Is there a range in predisposition periods for these offenses?
  • Coordination, both inside and outside justice agencies. What traditional kinds of supervision or monitoring exist to maintain victim safety? Who is missing from this? Aside from law enforcement and case monitoring, what role does the criminal justice system have in responding to predisposition revictimization? Do any behaviors in the justice system make victims feel "revictimized" or reluctant to report or return?
  • Enhancement of women's safety. Are some things, which were intended to help, actually hindering safety?
  • Data for research. What are the "points of access" to the justice system or to social assistance that could help define revictimization and yield numbers on current practices? Do the sources "talk to each other"? What agency actors and current practices need to be examined? Some currently reported quantities are contradictory, sometimes due to different measuring methods in different jurisdictions. NIJ particularly needs to be able to collect quantifiable information.
  • Quantifying victims' experiences. What (besides arrests) may be used for this? Some considerations include vulnerable, connected groups, such as children; access to economic and transportation resources; and rules about housing and assistance with housing or childcare. Federal definitions of "permanent housing" have caused some women to return to an abuser's house rather than stay safer with family or friends in "temporary housing" because the latter caused denial of benefits.
  • Special populations. The elderly who live in assisted living facilities or persons with handicaps can easily be overlooked. Are there services that reach these groups that would allow reports showing prevalence of victimization?
  • Cultural and ethnic differences. Are there different considerations in finding and treating problems of revictimization among women of different cultures? Is there agreement on when to consider a relationship to be "violent" and abusive? Is there a "typology" relative to offenders or victims?
  • Demands on the community or other providers. What is feasible to ask of a community?

Back to Top

Breakout Group, Open Discussion

The breakout session speakers noted that very many ideas had circulated. Participants thought that getting numerical data for rigorous research might often be a problem. However, rigorous research in multiple jurisdictions would give the Federal sponsor agencies a sense of policies and practices that bear on revictimization during the predisposition period. They discussed and considered the following priority areas for research:

  • Children and finances are large considerations for victims of intimate partner violence. What effect does a protection order have on a woman's ability to financially support her children? Could comparisons be made between jurisdictions that have supportive measures for victims and those that do not?
  • Are victims asked routinely whether there are safety considerations? What would be a good form of risk assessment that includes victim input?
  • Can research look beyond arrests in defining "predisposition" or followup? What effect do variances in the range of predisposition periods have on revictimization?
  • What is the design of protection orders in jurisdictions? Are there two-tiered or three-tiered systems? For example, a civil protection order against stalking might actually have more provisions than the criminal proceeding. What is the significance of "temporary" versus "permanent" POs? Is a victim notified when a PO has been served on a defendant (this could be a dangerous circumstance)?
  • How is revictimization defined in the victim's perspective; and does this match somehow with "chargeable offenses"?
  • Are there specific programs that explain bail and restraining order conditions for defendants (outside of the hearing itself)? Would that lower revictimization rates?
  • Are there effective measures to deal with fear and a sense of helplessness? How would they fit into criminal justice definitions of revictimization?
  • What are the reasons that victims do not come back for help? What alternate systems do they use to stay safe?
  • Is a "color blind" court desirable, or is there a need for one that recognizes distinctions between ethnic groups, differences in vulnerabilities and strengths that need to be acknowledged? What difference does this make in terms of victim safety?
  • Are there intra-system communications in the justice system such that the victim's perspective is taken into account, a communications model between agencies that understands their needs?
  • What coordinated community responses exist? Could there be a pyramid or baseline of services that include what is available from the criminal justice system, faith-based organizations, the community, and other "middle areas." Victims often come to community or faith-based organization facilities; how do people there evaluate and respond to risk?
  • How do police officers and victims themselves assess the probability of reassault?
  • Use cost-benefit analysis to see if resources can be directed to those victims who are at greatest risk. Put resources where greater harm is done.
  • Use training in schools as a front-end intervention and part of coordinated community response.
  • Can researchers take arrest as an "incident-based" starting point and follow whatever occurs to see what is working? This might include faith-based organization activities, criminal justice system activities, or other activities.
  • Hard-to-reach groups, like immigrant or military communities, need particular strategies. Fear of deportation and economic dependence complicate abusive relationships. Look into the effects of the USA PATRIOT Act on immigrant women: economic dependence and the partner's deportation are also issues.
Angela Moore Parmley, Chief, Violence and Victimization Research Division, Office of Research and Evaluation, NIJ, warmly thanked all the participants for their time and work. She noted that meeting proceedings would be circulated for corrections or additions so that NIJ can capture the essence of this event. Priorities for the short term and agency activities will grow from the use of this information. She expressed hope that attendees had used this chance to make valuable connections, gather information, and draw ideas for improved practice in their own jurisdictions, so that services for women, children, men, and families are made better and victimization is curbed.

Back to Top

Date Created: November 28, 2007