Criterion-referenced testing is defined and some of its background is discussed. A distinction is made between criterion-referenced scores, norm-referenced scores, cut-off scores, criterion scores, criterion variables, and content-standard scores. The relationship between norm-referenced information and criterion-referenced information is considered. The need for vigorous, empirically-based construct validation studies of criterion-referenced tests is pointed out. The use of criterion-referenced testing in instruction is considered in terms of absolute interpretations and mastery learning. It is seen that whether criterion-referenced testing and/or norm-referenced testing is needed to make instructional decisions depends upon the instructional context within which one operates. It is concluded that, for purposes of instruction and instructional decision-making, there is a need for the integration of measurement knowledge with knowledge about instructional psychology. (Author/TA)