A computerized mastery testing system providing for the computerized implementation of sequential testing in order to reduce test length without sacrificing mastery classification accuracy. The mastery testing system is based on Item Response Theory and Bayesian Decision Theory which are used to qualify collections of test items, administered as a unit, and determine the decision rules regarding examinee's responses thereto. The test item units are randomly and sequentially presented to the examinee by a computer test administrator. The administrator periodically determines, based on previous responses, whether the examinee may be classified as a nonmaster or master or whether more responses are necessary. If more responses are necessary it will present as many additional test item units as required for classification. The method provides for determining the test specifications, creating an item pool, obtaining IRT statistics for each item, determining ability values, assembling items into testlets, verifying the testlets, selecting loss functions and prior probability of mastery, estimating cutscores, packaging the test for administration, randomly and sequentially administering testlets to the examinee until a pass/fail decision can be made.