After identifying and discussing the errors occasioned by the use of county-level Uniform Crime Reports (UCR) data in a particular study, this article recommends that until improved methods of imputing county-level crime data are developed, tested, and implemented, they should not be used, especially in studies upon which policy is to be based.
The authors first describe the relevant aspects of the FBI's Uniform Crime Reporting Program, followed by a description of the characteristics of the data used in the study entitled, "More Guns, Less Crime" (Lott, 1998, 2000). This study used county-level crime data in concluding that the imposition of "right-to-carry" gun laws reduced the incidence of homicide and other violent crimes. Because these findings challenged the intuition of many researchers, they have been subjected to unusually close scrutiny (e.g., Ayres and Donohue, 1999; Black and Nagin, 1998; and Zimring and Hawkins, 1998). So many problems were found with the county-level crime data sets used in the Lott study, that its analyses must be questioned. The county-level crime data used by Lott et al. were the data sets aggregated to the county level and archived at the National Archive of Criminal Justice Data (NACJD). They were based on the "Crime by County" data file provided by the FBI's Uniform Crime Reporting Program. The NACJD data sets have a number of features that militate against their use in analyses of the type conducted in the Lott study. An analysis of these data found serious gaps; for example, over 50 percent of the county-level crime data points in three States (Connecticut, Indiana, and Mississippi) were found to be missing crime data from more than 30 percent of their populations; and in another 13 States more than 20 percent of the data points contained coverage gaps of this magnitude. Moreover, those States with the greatest gaps in crime reporting coverage tended to be States that had laws that permitted the carrying of concealed weapons. This article recommends that under current circumstances, county-level UCR data should not be used in research, particularly research intended to examine policy impacts. 3 tables, 8 figures, and 17 references