This paper focuses on teacher testing issues related to job analysis, test specification development, reliability, and validity. It emphasizes the conceptualization and operational definition of appropriate validity evidence to assess the quality of licensure testing decisions. It is suggested that the process of job, or practice, analysis would be improved by adopting strategies that are more analytic of the settings, activities, and competencies involved in teaching and which focus on cognitive processes that go beyond the recall of facts. Reliability is not sufficient to establish the credibility of a test, but without adequate and appropriate reliability, a test cannot be judged as valid. There is some debate over the necessity of documenting evidence of predictive validity during the validation process, but many educators believe that predictive validity evidence is essential in teacher testing. Others note that micro-level validity studies are also important, and that information must be gathered at the item quality level to insure that the validity decision is correct. Test specifications, the technical qualifications of the people who write the test items, and the documentation of each item as it is developed are all important. It is also important to examine the statistical characteristics of the items. The responsibility for gathering all validity evidence lies in the hands of the contractor for the tests, who must compile the validity evidence, review the evidence provided by the developer or other authorities, summarize the evidence, and establish the validity of the use of the results of the test. (Contains 39 references.) (SLD)