Answers are grounded in this paper and cite their sources.
Licensure and certification decisions are generally based on a chain of inference from the results of a practice analysis to test specifications, to a test, to examinee performance on the test, to a pass-fail decision. This article focuses on the first two steps in this chain of inference: (a) the design of practice analyses and (b) the translation of practice-analysis results into test specifications. The approach taken is to develop a tentative model of practice prior to conducting the practice analysis and to use this practice model to structure both the data collection and the interpretation of results. The practice model provides a preliminary version of the test specifications that is then empirically tested and refined in the practice-analysis study. If the practice model is supported by the data, the translation of practice-analysis results into test specifications can be simple and straightforward.
No takes yet. Share an insight, caveat, or question.
Michael T. Kane (1997) studied this question.
Synapse has enriched 5 related papers. Consider them for comparative context: