Standards for evaluating criterion-referenced tests are presented. Twenty-one standards, grouped in three categories, are discussed. Category one is defined as measurement properties and is comprised of conceptual validity, including description of the domain, test item agreement with objectives, and item representativeness of the objectives; and field test validity, including sensitivity, item uniformity, divergent validity, lack of bias, and consistency of scores. Category two is labeled appropriateness for examinees and is comprised of clarity of instructions; item review; physical format, including layout and legibility; and ease in recording answers. Category three is called practicality, and is composed of adequacy of information about the test; relevance of items of at least two series of teaching materials; flexibility, including multilevel testing of objectives; alternate test forms; clarity of test administration directions; scoring; record keeping; availability of rules to make instructional decisions based on test results; and comparative data on test scores. It is also stated that the test buyer must determine the degree of correspondence between the objectives of a test package and the objectives of the curriculum to be tested. (MH)