THE USE OF MULTIPLE INDICATORS TO ESTIMATE CRIME TRENDS IN AMERICAN CITIES

Although the description of crime trends is one of the most basic goal s of criminological research, this task of obvious importance has prov ed to be extremely problematic due to the incompatible definitions and associated measurement errors contained in the most widely available datasets, such as the Uniform Crime Reports and the National Crime Sur vey. Although recent research has attempted to overcome these problems by incorporating the UCR and NCS estimates into a single multiple-ind icator model, this does not resolve the problem for local agencies for which only a single estimate (i.e., the UCR) is generally available. This note argues that the relatively recent availability of calls-for- service data in many jurisdictions provides an opportunity for a multi ple-indicator estimation of these trends at the local level. The appro ach is illustrated with data collected over a 22-month period in Oklah oma City.