Examining Arrest Probabilities Among Female and Male Offenders Using Incident Based Arrest Reports (NIBRS) for Violent Crime

Anne L. Stahl, University of Pittsburgh
Phyllis Coontz, University of Pittsburgh

ABSTRACT
This paper examines the gender-crime gap by analyzing the probability of arrest for female and male offenders arrested for violent crime using NIBRS data. For several decades there has been an ongoing debate in criminology (as well as the popular press) about changes in the nature and extent of female criminality. The data typically used for examining this debate based on UCR arrest data. One of the biggest shortcomings with UCR data has been that they cannot be disaggregated simultaneously by age, sex, and race, and thus cannot be used to predict the probability of arrest among specific offenders over time. The FBI's NIBRS system represents a sharp improvement over UCR limitations. As an incident-based reporting system, NIBRS data are collected on each single crime occurrence within 22 offense categories and are include detailed information about each crime. The addition of more comprehensive information, particularly information about the sex, age, and race of the offender for each incident allows for the calculation of arrest probabilities for specified groups arrested for specific crimes. Data are now available from 21 states (representing some 900 originating agencies) reporting consistently from 1996 through 2000. Although this 21-state sample is not sufficient to produce estimates of national trends, the data set does contain detailed records representing a census of aggravated assault and robbery incidents known to this set of law enforcement agencies and can potentially be valuable in addressing the ongoing debate in criminology about the convergence of crime rates for females and males.

In preliminary analyses, offer some surprising findings. For example, the strongest predictor of arrest among females for robbery is offenders age--rather than for example offender's race, weapon, victim's race. The strongest predictor of arrest among males for robbery is whether there was an injury to the victim. To our knowledge, NIBRS data have not been used to examine the gender-crime gap, and our analyses raise a number of interesting issues about relevance of age for females.

(Return to Program Resources)

Updated 05/20/2006