The goals for this study were to (1) use the National Incident-Based Reporting System (NIBRS) to explore whether changes in economic inequality and racial/ethnic composition over the 2000-2010 decade were associated with changes in the prevalence and nature of violence among and between Whites, Blacks, and Hispanics and (2) to construct more accessible NIBRS cross-sectional and longitudinal databases containing race/ethnic-specific measures of violent victimization, offending, and arrest.
The data were constructed from National Incident-Based Reporting System (NIBRS) extract files and values for missing data were imputed using imputation by chained equations. Users should consult the accompanying documentation for details regarding imputation models.
Ten imputed datasets were created at the incident level for each year under analysis. For each year, the ten imputed datasets were then averaged to take into account the uncertainty associated with the imputations. After averaging the imputations, the data were aggregated to the police agency level and a "months reported correction" was applied to account for the number of months each police agency provided data to NIBRS. Finally, arrests, offending, and victimization counts were aggregated to the census place level to create yearly, census place counts of violence, including measures of inter- and intra-group violence.
For each of the time points provided in this database (2000-2004, 2010-2014), census places were included only if they had at least 500 Whites, Blacks, and Hispanics, all of the agencies within it reported to NIBRS that year, and there were at least two (2) years of valid NIBRS data.
Census-designated places that participated in the National Incident-Based Reporting System (NIBRS) between 2000 and 2014, have a total population of at least 5000, have at least 500 Whites, 500 Blacks, and 500 Hispanics, and all of the agencies within it reported to NIBRS that year, and there were at least two (2) years of valid NIBRS data.
administrative records data
This study contains one Stata dataset that includes 614 cases and 159 variables. The dataset contains administrative variables including data collection years, years of NIBRS coverage, and location codes. A majority of the variables represent administrative data regarding the number of arrests, victimizations, and offenses for each ethnic group examined (White, Black, Hispanic) and by type of crime (violence, homicide, rape, robbery, and aggravated assault). Also included are variables that sort incidents by whether they were inter-racial (White on Black, White on Hispanic, Black on White, Black on Hispanic, Hispanic on White, Hispanic on Black) or intra-racial (White on White, Black on Black, Hispanic on Hispanic) in nature. Census variables were gathered from the National Historical Geographic Information System and include information on population, poverty, unemployment, education, household composition, household mobility, income, percent of population that is young and male, percent of population that is foreign born, and segregation (index of dissimilarity). Finally, the dataset included the number of people per square mile and the number of police officers per 1,000 persons.