Milwaukee Domestic Violence Experiment, 1987-1989 (ICPSR 9966)
This study represents a modified replication of the Minneapolis Domestic Violence Experiment (SPECIFIC DETERRENT EFFECTS OF ARREST FOR DOMESTIC ASSAULT: MINNEAPOLIS, 1981-1982 [ICPSR 8250]). The Minneapolis study found arrest to be an effective deterrent against repeat domestic violence. The two key purposes of the current study were (1) to examine the possible differences in reactions to arrest, and (2) to compare the effects of short and long incarceration associated with arrest. Research protocol involved 35 patrol officers in four Milwaukee police districts screening domestic violence cases for eligibility, then calling police headquarters to request a randomly-assigned disposition. The three possible randomly assigned dispositions were (1) Code 1, which consisted of arrest and at least one night in jail, unless the suspect posted bond, (2) Code 2, which consisted of arrest and immediate release on recognizance from the booking area at police headquarters, or as soon as possible, and (3) Code 3, which consisted of a standard Miranda-style script warning read by police to both suspect and victim. A battered women's shelter hotline system provided the primary measurement of the frequency of violence by the same suspects both before and after each case leading to a randomized police action. Other forms of measurement included arrests of the suspect both before and after the offense, as well as offenses against the same victim. Initial victim interviews were attempted within one month after the first 900 incidents were compiled. A second victim interview was attempted six months after the incident for all 1,200 cases. Data collected for this study included detailed data on each of the 1,200 randomized events, less detailed data on an additional 854 cases found ineligible, "pipeline" data on the frequency of domestic violence in the four Milwaukee police districts, official measures of prior and subsequent domestic violence for both suspects and victims, interviews of arrested suspects for eligible and ineligible cases, criminal justice system dispositions of the randomized arrests, results of urinalysis tests of drug and alcohol use for some arrestees, and log attempts to obtain interviews from suspects and victims. Demographic variables include victim and suspect age, race, education, employment status, and marital status. Additional information obtained includes victim-offender relationships, alcohol and drug use during incident, substance of conflict, nature of victim injury and medical treatment as reported by police and victims, characteristics of suspects in the Code 1 and 2 arrest groups, victim and suspect reports of who called police, and victim and suspect versions of speed of police response.
The public-use data files in this collection are available for access by the general public. Access does not require affiliation with an ICPSR member institution.
WARNING: Because this study has many datasets, the download all files option has been suppressed, and you will need to download one dataset at a time.
Sherman, Lawrence W., Janell D. Schmidt, and Dennis P. Rogan. MILWAUKEE DOMESTIC VIOLENCE EXPERIMENT, 1987-1989. ICPSR version. Washington, DC: Crime Control Institute [producer], 1990. Ann Arbor, MI: Inter-university Consortium for Political and Social Research [distributor], 1994. http://doi.org/10.3886/ICPSR09966.v1
Persistent URL: http://doi.org/10.3886/ICPSR09966.v1
This study was funded by:
- United States Department of Justice. Office of Justice Programs. National Institute of Justice (86-IJ-CX-K043)
Scope of Study
The Follow-Up Interview Data contain more cases than the Victim Initial Interview Data because the researchers subsequently included some respondents who did not participate in the initial interviews.
Study Purpose: This data collection is patterned after the Minneapolis Domestic Violence Experiment (SPECIFIC DETERRENT EFFECTS OF ARREST FOR DOMESTIC ASSAULT: MINNEAPOLIS, 1981-1982 [ICPSR 8250]), a study testing police responses to domestic violence. That study found that arrest was the most effective of three standard methods police use to reduce domestic violence. The other police methods--attempting to counsel both parties or sending assailants away from home for several hours--were found to be considerably less effective in deterring future domestic violence in the cases examined. The current study represents a modified replication in Milwaukee of the original Minneapolis experiment, with four times as many cases, fewer deviations from random assignment, a majority of Black victims and offenders, and measures of before-and-after differences in offending frequency. The main objectives of the study were (1) to replicate the Minneapolis experiment in a city with higher rates and seriousness of violence, particularly one with a greater ethnic and economic mix, (2) to test the impact of arrest with brief versus overnight jail time, and (3) to measure interaction effects by obtaining a larger sample size. The study focused on repeat domestic violence by the suspect against any victim, the total frequency of repeat violence associated with each of the police actions, and the total count of all violent incidents during the follow-up period.
Study Design: Experimental procedures involved 35 patrol officers in four of Milwaukee, Wisconsin's six police districts screening all misdemeanor domestic battery cases for eligibility, then calling the Crime Control Institute office at police headquarters to request a randomly-assigned disposition. The eligibility requirements included probable cause that a misdemeanor battery had taken place between two adult cohabitants or former cohabitants or parents of the same child, where a single suspect was present and none of the following conditions were found: (a) a valid restraining order in effect, (b) an outstanding warrant for the suspect's arrest, (c) serious bodily injury or threat of more violence, and (d) suspect's assault on a police officer. The method of random assignment was to pre-number the dispositions as Codes 1, 2, or 3, and arrange the sequence of those three dispositions in a computer- generated order in sealed envelopes with the sequence numbers marked on the outside. Crime Control Institute staff opened the envelopes when police called, telling police the correct disposition and recording the case in the caselog. Code 1 consisted of arrest and at least one night in jail, unless the suspect posted $250 bond. Code 2 involved arrest and immediate release on recognizance from the booking area at police headquarters, preferably within two hours, or as soon as possible. Code 3 consisted of a standard Miranda-style script warning read by police to both suspect and victim, telling them that no arrest would be made unless police had to return to the home that evening. The primary measurement of recidivism was calls recorded by a Milwaukee battered women's shelter hotline system. Additional measures of prior offending and recidivism were obtained from the Milwaukee Police Department's file on individual arrest histories and victim interviews. Initial victim interviews were attempted within one month after the first 900 incidents were compiled. At some point six months after the randomized incident, follow-up interviews were attempted for all 1,200 cases. Interviews with arrested eligible suspects (Codes 1 and 2) and ineligible suspects were conducted in the police department's lockup cells prior to booking. To explore the connection between domestic violence and alcohol or drug abuse, urine sampling was undertaken for the last 300 eligible and ineligible domestic violence arrestees.
Sample: Calls received by the Milwaukee Police regarding misdemeanor domestic assault were screened by police officers to establish eligibility for the experiment. Eligible calls were referred to the Crime Control Institute staff, who randomly assigned one of three treatments. Selection of cases continued until 1,200 eligible cases were obtained.
personal interviews and police records
Description of Variables: Victim interviews collected information regarding the nature of the domestic assault incident such as reason for the argument, extent of violence, threats, and property damage, and incidents of domestic assault subsequent to the presenting incident. Additional information was collected regarding victim-offender relationships, alcohol and drug use during the incident, nature of victim injury and medical treatment as reported by police and victims, characteristics of suspects in the two arrest groups, victim and suspect reports of who called police, and victim and suspect versions of speed of police response. Demographic variables include victim and suspect age, race, education, employment status, and marital status.
Response Rates: A total of 705 initial interviews were conducted from the 1,200 eligible domestic battery incidents, or 59 percent. Six-month follow-up interviews were conducted with 921 respondents from the original 1,200 incidents, or 77 percent. There are more "follow-up" interviews than initial interviews because the researchers included some individuals who were not initially interviewed.
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: 1995-03-16
- 2006-01-12 All files were removed from dataset 16 and flagged as study-level files, so that they will accompany all downloads.
- 2006-01-12 All files were removed from dataset 8 and flagged as study-level files, so that they will accompany all downloads.
- 2005-11-04 On 2005-03-14 new files were added to one or more datasets. These files included additional setup files as well as one or more of the following: SAS program, SAS transport, SPSS portable, and Stata system files. The metadata record was revised 2005-11-04 to reflect these additions.
- List all ~28 citations associated with this study
Most Recent Publications
- Citations exports are provided above.
Export Study-level metadata (does not include variable-level metadata)
If you're looking for collection-level metadata rather than an individual metadata record, please visit our Metadata Records page.