Measuring Change in City Murder Clearance Rates with Incomplete Information

Abstract Objective Missing information is a complicating factor in measuring murder clearance rates at the city level. This paper contributes to the literature by providing estimates of the direction and magnitude of change scores for murder clearance rates for 67 large American cities. Methods Partial identification methods are used to study the changes in murder clearance statistics over time. These methods emphasize interval rather than point identification. The width of the intervals are directly proportional to the fraction of cases with missing information. Results The analysis reveals strong evidence of decline in murder clearance rates from the mid-1960’s to the mid-1990’s. There is much more uncertainty about the direction and magnitude of change between the mid-1990’s and late 2010’s. Conclusions Uncertainty about key parameters related to clearance rates are often conveyed in terms of sampling error. The analysis described in this paper is meant to highlight the additional ambiguity arising from missing information. Partial identification methods can be a useful complement to existing methods of studying clearance rates over time and comparing clearance rates across jurisdictions.