National Archive of Criminal Justice Data
This dataset is maintained and distributed by the National Archive of Criminal Justice Data (NACJD), the criminal justice archive within ICPSR. NACJD is primarily sponsored by three agencies within the U.S. Department of Justice: the Bureau of Justice Statistics, the National Institute of Justice, and the Office of Juvenile Justice and Delinquency Prevention .
Intercity Variation in Youth Homicide, Robbery, and Assault, 1984-2006 [United States] (ICPSR 30981)
Principal Investigator(s): Strom, Kevin, RTI International; Browne, Angela, Vera Institute of Justice
Summary: The research team collected data on homicide, robbery, and assault offending from 1984-2006 for youth 13 to 24 years of age in 91 of the 100 largest cities in the United States (based on the 1980 Census) from various existing data sources. Data on youth homicide perpetration were acquired from the Supplementary Homicide Reports (SHR) and data on nonlethal youth violence (robbery and assault) were obtained from the Uniform Crime Reports (UCR). Annual homicide, robbery, and assault arrest rates ... (more info)
This data is freely available.
Strom, Kevin, and Angela Browne. Intercity Variation in Youth Homicide, Robbery, and Assault, 1984-2006 [United States]. ICPSR30981-v1. Ann Arbor, MI: Inter-university Consortium for Political and Social Research [distributor], 2012-09-20. doi:10.3886/ICPSR30981.v1
Persistent URL: http://dx.doi.org/10.3886/ICPSR30981.v1
This survey was funded by:
- United States Department of Justice. Office of Justice Programs. National Institute of Justice (2007-IJ-CX-0025)
Scope of Study
Summary: The research team collected data on homicide, robbery, and assault offending from 1984-2006 for youth 13 to 24 years of age in 91 of the 100 largest cities in the United States (based on the 1980 Census) from various existing data sources. Data on youth homicide perpetration were acquired from the Supplementary Homicide Reports (SHR) and data on nonlethal youth violence (robbery and assault) were obtained from the Uniform Crime Reports (UCR). Annual homicide, robbery, and assault arrest rates per 100,000 age-specific populations (i.e., 13 to 17 and 18 to 24 year olds) were calculated by year for each city in the study. Data on city characteristics were derived from several sources including the County and City Data Books, SHR, and the Vital Statistics Multiple Cause of Death File. The research team constructed a dataset representing lethal and nonlethal offending at the city level for 91 cities over the 23-year period from 1984 to 2006, resulting in 2,093 city year observations.
Subject Terms: age, assault, crime patterns, crime statistics, drug related crimes, firearms, gang violence, gangs, homicide, juvenile crime, juveniles, robbery, trends, violence, violent crime, violent crime statistics, youths
Smallest Geographic Unit: city
Geographic Coverage: United States
Date of Collection:
Unit of Observation: city-by-year
Universe: All youth between the ages of 13 to 24 in the 100 most populous central-cities in the United States from 1984 to 2006.
Data Types: aggregate data
Data Collection Notes:
Detailed information about the Uniform Crime Reporting Program (UCR), including the Supplementary Homicide Reports (SHR), is available through the Uniform Crime Reporting Program Resources Guide.
Users should refer to the project's final technical report (Browne and Strom, 2010; NCJ 232622) for additional information on the study methodology, missing data, and imputation procedures.
Study Purpose: The purpose of this study was to estimate temporal trends in youth violence rates variation across 91 of the 100 largest cities in the United States from 1984-2006, and to model city-specific explanatory predictors influencing these trends.
In order to estimate trends in homicide offending for youth 13 to 24 years of age in 91 of the 100 largest cities in the United States from 1984-2006, data for youth homicide were acquired from the Supplementary Homicide Report (SHR), a component of the FBI?s Uniform Crime Reporting Program (UCR). Measures of youth arrests for the nonlethal violent crimes of robbery and assault were acquired from UCR city arrest data for the same time period. Annual homicide, robbery, and assault arrest rates per 100,000 age-specific (i.e., 13 to 17 and 18 to 24 year olds) population were calculated by year for each city in the study. Annual homicide rates were calculated through a conventional procedure: annual incidents in a specific city, divided by the age-specific population of that city, multiplied by 100,000. Partial reporting during the time period resulted in dropping 9 cities from the homicide data and 10 cities from the robbery and assault data. Data on city-level characteristics including measures of structural disadvantage, drug market activities, gang presence-activity, and firearm availability were derived from the County and City Data Books, SHR, and the Vital Statistics Multiple Cause of Death File, respectively.
Missing data came from two sources; failure to report in homicide and some of the Census collections, and lack of data for specific years, mainly in Census data, between major data collection points like the Decennial Census and the Mid-decade estimates from Census related sources. Missing data in the homicide measures were addressed using an Iterative Chain equation procedure to conduct Multiple Imputation. Variables from the original source used in the multiple imputation procedure included age of victim, race, ethnicity, gender, seven available measures of homicide circumstances, and city population size. Extrapolation methods were used to adjust for missing data in the robberies and assaults by age, and in the census and economic data sources. To estimate a missing year between two reported values, the missing year was estimated to be mid-way between the two observed years on either side of the missing year. Longer gaps involved further averaging and allocating according to the number of years missing; these estimates amount to maximum likelihood estimates of the missing years or in the case of the robberies and assaults, months as well.
Sample: The initial sample consisted of the 100 largest cities in the United States based on the 1980 Census; however, several cities were dropped due to missing data problems, resulting in a sample of 91 cities for the homicide data and 90 cities for the nonlethal violence data. If a city had 10 or more consecutive years of missing data, the researchers eliminated it from the final dataset. The 91 cities were measured over the course of 23 years from 1984 to 2006, resulting in 2,093 total observations.
Mode of Data Collection: record abstracts
UNIFORM CRIME REPORTS [UNITED STATES]: SUPPLEMENTARY HOMICIDE REPORTS, 1984-2006 [Annual Data Files]
UNIFORM CRIME REPORTING PROGRAM DATA [UNITED STATES]: ARRESTS BY AGE, SEX, AND RACE, 1984-2006 [Annual Data Files]
United States Census of the Population, 1980, 1990, 2000
United States Economic Census, previously known as the Census of Business and Industry, 1982, 1987, 1992, 1997, 2002, 2007
City and County Data Book Series, 1987, 1996, 2006
American Community Survey, 2001-2006
National Center for Health Statistics, Division of Vital Statistics, Multiple Cause of Death file
Description of Variables: The study contains a total of 39 variables including city name, year, crime rate variables, and city characteristics variables. Crime rate variables include imputed and non-imputed homicide rate variables for juveniles aged 13 to 17, young adults aged 18 to 24, and adults aged 25 and over. Other crime variables include the number of imputed and non-imputed homicides as well as the robbery rate and assault rate for juveniles and young adults. City characteristics variables include population, poverty rates, percentage of African Americans, percentage of female-headed households, percentage of residents unemployed, percentage of residents receiving public assistance, home-ownership rates, gang presence and activity, and alcohol outlet density.
Response Rates: Not applicable.
Presence of Common Scales: One scale was used: The FAC1_1 "REGRESSION BASED FACTOR SCORE INCLUDING POVERTY AFROAM FEMHH PUBAST UNEMP" variable is a regression based factor score based on a principal components factor analysis of five of the variables.
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: 2012-09-20
Use any of the notification links to add this study to your RSS feed; you will then receive notification if the study is substantively updated.
- 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.