The need for better validation of professional licensing and certification examinations is discussed in Part I. Predictive criterion-related validation studies are difficult to develop, time consuming, impractical and expensive. Since training and knowledge variables are not necessarily sufficient to predict job proficiency, it appears logical to look to performance testing for solutions to the problems of validating and revising licensure exams more effectively. Four areas in which there are differences in practice across differing licensure examinations: (1) the type of test, (2) the type of controls involved in the testing process, (3) the kind of evidence utilized in the entry decision, and (4) the range of authority and the degree to which the examination provides a right to practice are discussed in Part II. Differences in practices across these four areas appear to be directed toward differential control of three primary sources of malpractice--absence of entry level professional skills, absence of adequate behavior patterns to cope with the pressures of professional practice and absence of adequate updating of professional practice. (MV)