This report describes 12 studies dealing with the knowledge monitoring component of metacognition. It is assumed that knowledge monitoring is basic to other metacognitive activities, such as evaluating learning, selecting appropriate strategies, or planning, because distinguishing between what students know and do not know ought to be a prerequisite for these other higher-level activities. The 12 studies, 10 in the verbal domain and 2 in mathematics, used various versions of a knowledge monitoring assessment (KMA) that evaluates the discrepancy between students' estimates of their knowledge in a domain and their demonstrated knowledge in that domain on a multiple-choice test. The results provide a good deal of support for the construct validity of the KMA and suggest that it has considerable generalizability over different types of content and varying student populations. Since the KMA may be group or computer administered and is objectively scored, it has substantial advantages over other means of evaluating metacognition. Suggestions for further research using the procedure are made.