Abstract : The PLATO Air Force Base Computer-Based Education (PLATO AFB CBE) project at Chanute adopted the mastery learning technique in their 34 lessons and set the mastery criterion at 80% correct on the end of lesson test. They used the performance result of each criterion-referenced test (CRT) in two different ways: (1) for assessing the individual performance, and (2) for evaluation, or more precisely within Chanute's context, lesson evaluation. Adoption of a criterion-referenced testing approach to evaluation raises two measurement issues that have relatively less importance in norm-reference testing. The issues are: (1) definition of mastery, and (2) a priori standards. These issues still remain unsolved, but are receiving increasing attention. One purpose of this paper is to examine the appropriateness of the use of CRTs as a means of controlling an individual student's advancement to the next level of instruction or retainment in the current unit of instruction in the PLATO AFB CBE Program at Chanute. A second purpose is to turn the focus from the aspect of individual assessment to that of program evaluation, which requires the establishment of a criterion rate for validation of a lesson, so that a lesson would be considered validated if the percentage of failure rate at the end of the lesson was less than the criterion.