National Evaluation of the Arrest Policies Program Under the Violence Against Women Act (VAWA), 1996-2000 (ICPSR 3795)
Principal Investigator(s): Archer, Cassandra, Institute for Law and Justice; DuPree, Cheron, Institute for Law and Justice
This study was undertaken to evaluate the impact of the Arrest Policies Program, funded by the Violence Against Women Office (VAWO), on criminal justice system changes and offender accountability, and victim safety and well-being. Through convenience sampling, six project sites were chosen to participate in the study. Part 1, Case Tracking Data, contains quantitative data collected from criminal justice agencies on arrests, prosecution filings, criminal case disposition, convictions, and sentences imposed for intimate partner violence cases involving a male offender and female offender. Data for Part 2, Victim Interview Data, were collected from in-depth personal interviews with domestic violence victims/survivors (1) to learn more about victim experiences with and perceptions of the criminal justice response, and (2) to obtain victim perceptions about how the arrest and/or prosecution of their batterers affected their safety and well-being. The survey instrument covered a wide range of topics including severity and history of domestic violence, social support networks, perceptions of police response, satisfaction with the criminal justice process and the sentence, experiences in court, and satisfaction with prosecutors, victim services provider advocates, and probation officers.
One or more files in this study are not available for download due to special restrictions ; consult the restrictions note to learn more. You can apply online for access to the data. A login is required to apply for access.
Access to these data is restricted. Users interested in obtaining these data must complete a Restricted Data Use Agreement, specify the reasons for the request, and obtain IRB approval or notice of exemption for their research.
Archer, Cassandra, and Cheron DuPree. NATIONAL EVALUATION OF THE ARREST POLICIES PROGRAM UNDER THE VIOLENCE AGAINST WOMEN ACT (VAWA), 1996-2000. ICPSR version. Washington, DC: Institute for Law and Justice [producer], 2003. Ann Arbor, MI: Inter-university Consortium for Political and Social Research [distributor], 2003. doi:10.3886/ICPSR03795.v1
Persistent URL: http://doi.org/10.3886/ICPSR03795.v1
This study was funded by:
- United States Department of Justice. Office of Justice Programs. National Institute of Justice (98-WE-VX-0012)
Scope of Study
Subject Terms: arrests, case processing, convictions (law), criminal justice system, disposition (legal), domestic violence, offenders, police response, probation officers, prosecution, process evaluation, social support, victim services, victims, victim safety
Smallest Geographic Unit: none
Geographic Coverage: United States
Date of Collection:
Unit of Observation: Part 1: court cases. Part 2: individuals.
Universe: Part 1: Intimate partner violence cases involving a male offender and female victim from any of the six pre-selected study sites. Part 2: Female victims/survivors of an intimate partner violence incident (from any of six pre-selected study sites) where the victims/survivors experienced the arrest and prosecution of their batterers.
Data Types: administrative records data, and survey data
Data Collection Notes:
The user guide, codebook, and data collection instrument are provided by ICPSR as Portable Document Format (PDF) files. The PDF file format was developed by Adobe Systems Incorporated and can be accessed using PDF reader software, such as the Adobe Acrobat Reader. Information on how to obtain a copy of the Acrobat Reader is provided on the ICPSR Web site.
Study Purpose: The Institute for Law and Justice (ILJ), through a grant from the National Institute of Justice (NIJ) and funding from the Violence Against Women Office (VAWO), conducted a national evaluation of the Grants to Encourage Arrest Policies Program (more recently known as the Grants to Encourage Arrest Policies and Enforcement of Protection Orders Program), which was funded under the Violence Against Women Act of 1994. Under this act and its reauthorization in 2000, the VAWO provided grants for the purpose of establishing or enforcing policies favoring arrest and prosecution of persons committing domestic violence. Development of this program was based on testimony at congressional hearings about the pervasiveness and seriousness of domestic violence, and on research showing that arrest of batterers can deter future domestic violence by defendants. This study sought to evaluate the impact of the Arrest Policies Program on criminal justice system changes and offender accountability, as well as victim safety and well-being.
Study Design: At the time this evaluation began, VAWO had awarded 130 Arrest Policies Program grants. With guidance from a national advisory board, ILJ developed site selection criteria and selected 20 projects/sites to participate in a program process evaluation. Six projects were later selected from among the 20 to be included in a more intensive evaluation of the Arrest Policies Program on criminal justice system changes and offender accountability (Part 1), and victim safety and well-being (Part 2). With respect to criminal justice system change and offender accountability, the evaluation plan called for collection of quantitative data on arrests, prosecution filings, criminal case disposition, convictions, and sentences imposed for intimate partner violence cases involving a male offender and female offender. While the methodology was intended to gather comprehensive and comparable statistics from all sites, data availability and reliability varied. For example, only one site could provide arrest statistics on domestic violence for years prior to 1996. Since most grantees began operations in 1997, this offered just one year of pre-grant data across all sites, although most projects were able to provide data for the years 1996 through 1999. Cross-site comparisons were also hampered by missing data in some records and the lack of automation at all sites As a result, researchers were not able to make valid cross-site comparisons from available statistics on the various measures of interest. To address the problem of inconsistent official data across sites, ILJ selected samples of cases from two time periods at all six sites. The sampling plan called for random selection at each site from 100 misdemeanor cases during a specified time period in 1996 (pre-grant), and 100 cases during the same time period in 1999 (post-grant). The cases were to be identified from police arrest reports and were restricted to intimate partner violence cases involving a male offender and female victim, the most common form of domestic violence. Because not all six sites had specific provisions in the law for designating domestic violence as felonies distinct from other felonies, only misdemeanors were included in the sample. The selection process also required that each offender be included only once. Although all six sites followed the data collection protocols to the extent possible, circumstances at the individual sites led to variations at three sites in how cases were selected. In one city, the small number of cases annually required a change in the number of cases sampled to 50 for each year. At two other sites, police arrest information was not available, and cases had to be selected from prosecutor files. After the samples were selected, the local coordinators, who were experienced researchers and had received training specific to this project from ILJ research staff, reviewed criminal justice agency case files to determine the case status at each stage of criminal justice processing. A research protocol was provided to guide the local coordinators in supervising the collection of agency data. ILJ staff closely monitored the work of the local coordinators through site visits as well as regular contact via e-mail, phone, fax, and mail. The data collection forms were adjusted to allow for site variations in charging codes, as well as disposition and sentencing options. While there were minor variations in each of the sites' data reflecting these local differences, a core set of data elements was captured. Information collected included incident date, offender and victim demographics, use of drugs, alcohol, or weapons by either party, primary arrest and filing charges, level of charges, disposition of case and date of disposition, sentencing for those convicted, and probation violations. Cases in Part 1 were weighted at each site based on the ratio of the sample to the universe for the time period of the sample. For example, in one county, all of the cases were selected into the sample, resulting in a weight of 1 for each case. In another county, every third case was selected in the specified time period, so each case received a weight of 3. Part 1 is comprised of 1,185 cases. The weights range from 1 to 5. The weighting procedures give a weighted total of 2,552 cases. Data for Part 2 were collected from in-depth personal interviews with domestic violence victims/survivors (1) to learn more about victim experiences with and perceptions of the criminal justice response, and (2) to obtain victim perceptions about how the arrest and/or prosecution of their batterers affected their safety and well-being. The evaluation design called for completion of 30 interviews with victims/survivors at each site. Fifteen of these 30 women were to be selected to participate in a "long version" of the interview. This group of women experienced the arrest and prosecution of their batterers. The remaining 15 women were to be selected for a "short version" of the interview. In these cases, the batterer was arrested, but the case was either declined for prosecution or initially dismissed by the prosecutor. With the level of prosecutorial effort as the primary difference between the two groups, the researchers sought to explore the effect of the prosecution on the victims' safety and well-being. Therefore, data used in Part 2 are the responses of those victims/survivors who were administered the "long version" of the interview. The recruitment of victims/survivors for interviews was challenging at all sites. The key variable, prosecution effort, required that victims/survivors be first identified from prosecutors' files. In addition, the cases selected had to be closed to assure prosecutors that the interviews would not interfere with any criminal cases. With assistance from victim witness staff, local research coordinators used a variety of techniques to recruit victims/survivors. Across sites, the greatest difficulty was locating, contacting, and recruiting victims whose cases had not been fully prosecuted. Furthermore, attempts to locate correct addresses or phone numbers were not always successful. Because of these problems, local coordinators were not always able to obtain the desired number of victims for interviews. Despite recruitment difficulties, a total of 62 interviews were conducted across the six sites. Each interview, lasting approximately 90 minutes, was conducted using a structured questionnaire that included both closed- and open-ended items, administered by trained ILJ research staff. The multi-item questions covered a wide range of topics, including (1) severity and history of domestic violence, (2) social support networks, (3) perceptions of police response, (4) satisfaction with the criminal justice process and the sentence imposed, (5) experiences in court, and (6) satisfaction with prosecutors, victim services providers, advocates, and probation officers.
Sample: Through convenience sampling, six project sites were chosen by the ILJ to participate in the study. Researchers believed that the six sites would allow for inclusion of different types of projects and project environments. They also determined that six was the maximum number of sites that could be examined in depth within the evaluation budget. The six sites were chosen from 20 sites which were already participating in a process evaluation of the Arrest Policies Program. Criteria had been established earlier in the project to assess each of the 20 projects/sites in terms of data availability and access to victims/survivors. Other factors were also considered: type of project leadership (e.g., police, prosecutor), focus of the grant-funded efforts, jurisdiction size, geographic representation, and access to samples of domestic violence victims who were members of racial and ethnic minorities. The final selection of sites was made by the ILJ research staff in consultation with members of the national advisory board, the federal technical assistance providers, the National Institute of Justice, and the Violence Against Women Office. For Part 1, a random selection of 100 misdemeanor cases from a specified time period in 1996 (pre-grant), and 100 cases from the same time period in 1999 (post-grant) was chosen. The cases were to be identified from police arrest reports and were restricted to intimate partner violence cases involving a male offender and female victim. Because not all six sites had specific provisions in the law for designating domestic violence as felonies distinct from other felonies, only misdemeanors were included in the sample. The selection process also required that each offender be included only once. For Part 2, the research design called for 15 women from each of the six sites to be chosen to complete a "long version" of the interview. These women experienced the arrest and prosecution of their batterers.
For Part 1, administrative records data were collected from police incident and arrest reports, prosecutors' offices, and criminal justice agency case files. For Part 2, data were collected from personal interviews conducted with victims/survivors.
Description of Variables: Variables in Part 1 include year of incident, type of weapon used, whether children were present, primary arrest and filing charges, any prior domestic violence arrests, whether the offender or victim was injured, days to disposition, type of plea, disposition outcome, whether there was a reduction of charges, and whether the offender was sentenced to jail, probation, batterer intervention, drug/alcohol treatment, or to pay a fine. Demographic variables for Part 1 include the race and age of both the offender and victim. Demographic variables for Part 2 include the race/ethnicity, age, and occupation of the victim and offender. Other variables include domestic violence history (e.g., whether victim was pushed, grabbed, slapped, bit, kicked, burned, stabbed, shot, or beat up), whether the domestic violence incident resulted in bruises, burns, bites, black eyes, scarring or permanent, whether anyone knew about the violence and provided support to the victim (e.g., a relative, friend, neighbor, religious leader, doctor/nurse, prosecuting attorney, therapist, or the police), whether the victim wanted the abuser arrested, the number of officers involved, what sorts of actions the police took (e.g., gave written information about the legal system, took photos of injuries, helped victim leave premises), whether there were any witnesses to the incident, reasons why the victim may have wanted charges dropped, the victim's interaction with individuals from the prosecutor's office, the victim's involvement with the offender's hearing/trial, the victim's level of satisfaction with various aspects of the criminal justice system, and the victim's experience with victim advocates, judges, counseling services, and probation officers.
Response Rates: Part 1: Not applicable. Part 2: Not available.
Presence of Common Scales: Several Likert-type scales were used in Part 2.
Extent of Processing: ICPSR data undergo a confidentiality review and are altered when necessary to limit the risk of disclosure. ICPSR also routinely creates ready-to-go data files along with setups in the major statistical software formats as well as standard codebooks to accompany the data. In addition to these procedures, ICPSR performed the following processing steps for this data collection:
- Standardized missing values.
- Checked for undocumented or out-of-range codes.
Original ICPSR Release: 2003-10-30
- 2006-03-30 File UG3795.ALL.PDF was removed from any previous datasets and flagged as a study-level file, so that it will accompany all downloads.
- 2006-03-30 File CQ3795.ALL.PDF was removed from any previous datasets and flagged as a study-level file, so that it will accompany all downloads.
- List all ~21 citations associated with this study
Most Recent Publications
- Citations exports are provided above.
Export Study-level metadata (does not include variable-level metadata)