Measurement Matters: Comparing Old and New Definitions of Rape in Federal Statistical Reporting

National statistics on the incidence of rape play an important role in the work of policymakers and academics. The Uniform Crime Reports (UCR) have provided some of the most widely used and influential statistics on the incidence of rape across the United States over the past 80 years. The definition of rape used by UCR changed in 2012 to include substantially more types of sexual assault. This article draws on 20 years of data from the National Incident-Based Reporting System to describe the impact this definitional change will have on estimates of the incidence of rape and trends over time. Drawing on time series as well as panel random effects methodologies, we show that 40% of sexual assaults have been excluded by the prior definition and that the magnitude of this error has grown over time. However, the overall trend in rape over time (year-to-year change) was not substantially different when comparing events meeting the prior definition and the subgroups of sexual assault that will now be counted.