Calculating performance measures: A primer for finance officers.

• Author: Williams, W. Anderson
• Position: Forum - Brief Article

As the number of governments using performance measurement increases, it is more important than ever that a wider audience understand the different kinds of measures and how to calculate them. At least to some extent, the usefulness of performance measurement increases in proportion to a government's collective understanding of the art and science thereof. This article is intended to facilitate such understanding by defining the most common measures, explaining the formulas used to calculate them, and illustrating their application.

There are two major categories of performance measures: output-based measures and outcome-based measures. Simply put, output-based measures indicate the amount of service completed or produced, while outcome-based measures indicate the extent to which desired program results have been achieved. This article covers three types of measures: workload, efficiency, and outcome (or effectiveness).

Workload Measures

Workload measures, which are typically the easiest to collect, calculate, and report, are the most common type of performance measure used by local governments. An output-based measure, workload is simply the volume of units produced for a particular service. Examples of workload measures might include the number of repairs completed, the number of job applications processed, or the number of potholes filled. Workload measures alone are not enough to gauge performance; however, they provide a base from which to calculate more useful measures, such as efficiency and outcome measures.

Standardizing workload measures by population allows government managers to compare workloads in one service area with workloads in another service area. For instance, suppose the incidence of rabies has increased in the rural areas of a small county of 47,103 residents. The animal control director hypothesizes that the increase in rabies is the result of fewer registered domestic animals in the rural areas than in the urban areas (rabies vaccination is a prerequisite for registration). To test this hypothesis, it is not enough to simply compare total workloads (i.e., number of animals registered) between the rural and urban areas of the county, since there are more residents, and thus more registered animals, in the urban areas than in the rural areas. Workload statistics alone do not indicate how diligent rural residents have been about registering their pets. Standardizing the number of registered animals by population, however, controls for population and allows for comparison between service areas. The formula for standardizing workloads by population is x/1,000 = total workload/total population.

To compare workload measures by service areas...