Violent Offending by Drug Users: Longitudinal Arrest Histories of Adults Arrested in Washington, DC, 1985-1986 (ICPSR 6254)
Principal Investigator(s): Cohen, Jacqueline, Carnegie Mellon University, H. John Heinz III School of Public Policy and Management
This data collection effort examined the influence of drug use on three key aspects of offenders' criminal careers in violence: participation, frequency of offending, and termination rate. A random sample of arrestees was taken from those arrested in Washington, DC, during the period July 1, 1985, to June 6, 1986. The sample was stratified to overrepresent groups other than Black males. Drug use was determined by urinalysis results at the time of arrest, as contrasted with previous studies that relied on self-reports of drug use. The research addresses the following questions: (1) Does drug use have an influence on participation in violent criminal activity? (2) Does drug use influence the frequency of violent offending? (3) Is there a difference in the types and rates of violent offending between drug-using offenders who use stimulants and those who use depressants? Variables regarding arrests include date of arrest, drug test result, charges filed, disposition date, disposition type, and sentence length imposed. Demographic variables include race, sex, birthdate, and place of birth.
These data are freely available.
Cohen, Jacqueline. VIOLENT OFFENDING BY DRUG USERS: LONGITUDINAL ARREST HISTORIES OF ADULTS ARRESTED IN WASHINGTON, DC, 1985-1986. ICPSR version. Pittsburgh, PA: Carnegie Mellon University, H. John Heinz III School of Public Policy and Management [producer], 1992. Ann Arbor MI: Inter-university Consortium for Political and Social Research [distributor], 1995. http://doi.org/10.3886/ICPSR06254.v1
Persistent URL: http://doi.org/10.3886/ICPSR06254.v1
This study was funded by:
- United States Department of Justice. Office of Justice Programs. National Institute of Justice (88-IJ-CX-0037)
Scope of Study
Date of Collection:
Unit of Observation: The units of observation are individual offenders and arrests. Since the data file is hierarchical, the unit of observation is dependent upon how the records are read from the data file.
Universe: Adult arrestees in Washington, DC, from the period July 1, 1985, to June 30, 1986.
Data Types: event/transaction data
Data Collection Notes:
The dataset is hierarchical and contains three record types. Record type 1, comprised of 13 variables and 1,368 records, contains stable personal information. Each unique individual represented in the data has a single record type 1. Record type 2, comprised of 49 variables and 7,117 records, contains changing personal information at the time of arrest. Record type 3, comprised of 78 variables and 7,117 records, contains arrest and court information. Each occurrence of record type 1 has at least one associated record type 2. Each occurrence of record type 2 has exactly one associated record type 3.
Study Purpose: The study was conducted to examine the relationship between drug use and violent criminal behaviors. The research addresses the following questions: (1) Does drug use have an influence on participation in violent criminal activity? (2) Does drug use influence the frequency of offending for violent crimes? (3) Is there a difference in the types and rates of violent offending between drug-using offenders who use stimulants and those who use depressants? This research may impact criminal justice decisions by delineating the role of drug use in different aspects of violent offending. The research suggests that drug use may be an aggravating factor among drug-using violent offenders who commit violent crimes at higher frequencies and over more extended time periods than do nonusing violent offenders.
Study Design: Longitudinal arrest data for a sample of arrestees were collected to characterize the violent offending patterns of drug users and nonusers. The sample included adults arrested in Washington, DC, on any charge from July 1, 1985, to June 30, 1986. Data were collected from case files maintained by the Washington, DC, Pretrial Services Agency for 1,368 arrestees and 7,117 arrests. Previous and subsequent arrests of the same individual were linked together to form an arrest history. Arrest histories date back to 1943 and continue through 1990. A major emphasis of the study was to refine measures of offending in order to better distinguish different aspects of violent offending by drug users and nonusers. Offending by drug users and nonusers, as determined by outcomes of urinalysis at the time of arrest, was characterized in terms of several distinct aspects of individual offending, particularly: (1) participation in violence estimated from the fraction of arrestees with any arrests for violent offenses, (2) frequency of violence reflected in the annual rate of violent criminal activity by these active offenders, and (3) termination rates of ending involvement in violent activities.
Sample: Stratified random sample of adult arrestees.
case files of the Washington, DC, Pretrial Services Agency
Description of Variables: The data are contained in three different record types. Record type 1 contains identifying information that remains constant over an individual's lifetime (e.g., race, sex, birthdate, place of birth). Each individual has one record type 1. Record type 2 contains personal information of a changing nature at each arrest (e.g., marital status, employment, substance abuse information, probation and parole information). There is a record type 2 for each arrest. Consequently, each record type 1 has at least one corresponding record type 2. Record type 3 contains arrest and court information for each arrest (e.g., charges, warrant information, case disposition, and sentencing). Each record type 2 has one, and only one, corresponding record type 3.
Response Rates: Not applicable.
Presence of Common Scales: None.
- Standardized missing values.
- Checked for undocumented or out-of-range codes.
Original ICPSR Release: 1996-01-22
- Citations exports are provided above.
Export Study-level metadata (does not include variable-level metadata)