Benchmark

From EvaluationWiki

These terms are used in different ways in different fields, notably surveying, computing, and management. In the evaluation field, the usage varies considerably, often depending on the background from which the user comes. One usage, close to the management use, refers to the practice of comparing a process-for example, the process of selecting students into a graduate program or recruits into an executive training program-to the acknowledged best practice for that process. In this sense, it means 'rating against the best' a type of comparative evaluation. In other cases it is used, as in the computer industry, to mean 'testing on (usually a set of) standard performance tests.' In this sense, it means essentially 'quantitative evaluation on established tests.' That use is readily extended to qualitative testing. Between them these two uses cover almost all uses in the evaluation literature. The arguments for benchmarking are the usual arguments for serious, objective, quasi-external, evaluation: it wakes players up to the fact that others are doing as well or better on similar tasks, it introduces some external validity into a domain of subjective judgment, etc.