In light of the widespread use of competency testing, the authors consider that it is important to determine ways of developing and using competency testing to insure that it achieves its full potential. The paper, in three parts, introduces a model for the development and validation of competency tests, reviews several methods for setting standards or competency levels, and makes suggestions for future research and development. Firstly, definitions of competency testing, criterion referenced tests, and standards are provided. The twelve step development and validation model introduced incorporates: competency selection; test specification; writing and editing test items; determining content validity; further editing; test assembly; standard setting; test administration; collection of reliability, validity, and norm data; preparation of users and technical manuals; periodic collection of additional information. The standard setting models considered are continuum models, of which the major assumption is that mastery is a continuously distributed ability. These models are further subdivided, for descriptive and comparative purposes, into judgmental, empirical, and combination models. The characteristics of the nineteen models thus categorized are then discussed. The development of guidelines for competency test development and further work on the moral and technical issues involved in standard setting are recommended. (AEF)