Exploring the Value of Incident-Based Crime Data for Calculating Local Crime Rates: Problems and Prospects

• In Event: National Crime Statistics Exchange (NCS-X): Increasing Use of Incident-Based Data to Strengthen Law Enforcement Decision-Making

Abstract

Researchers, law enforcement, and policy makers have criticized the FBI’s index crime classification system on a number of fronts. The index crimes are broad and heterogeneous, and offer little information on how the nature of crime within these broad offense classes varies across local jurisdictions. This presentation uses data from the National Incident Based Reporting System (NIBRS) to illustrate an alternative model for measuring crime rates that emphasizes the knowledge about local crime that is gained from an attribute-based crime classification system. The presentation illustrates the additional knowledge that can be gleaned about local crime rates by measuring rates of gun violence, intimate partner violence, public violence, and stranger violence. An ancillary goal of such an exercise is to demonstrate a framework for making more meaningful comparisons of crime rates across jurisdictions that adjusts rates of crime for the severity and composition of criminal incidents. The presentation concludes with a discussion of the methodological challenges inherent in such a task, chief of which are addressing agency non-response and missing information about the attributes of crimes, and a brief overview of various imputation strategies currently being explored by the Bureau of Justice Statistics to address missing data in NIBRS.

Author

• Kimberly Martin, Bureau of Justice Statistics, USDOJ