Given the wide individual differences among any group of students and since measurements are always accompanied by errors, the question of how tests should be used in assessing the quality of an educational program is considered. The ways in which educators have dealt with this problem are reviewed, from the systematic examinations of the question beginning in the 1930s through the grasp of the problem today. A better understanding of the evaluation process as proposed in 1951 by W. Cook is presented. Since that time educational measurement has been characterized by alternating emphasis on the development of excellence and a focus on minimum essentials. There currently is no general agreement on reference standards, and no simple answer to the meaning behind test results has been found. Five exhibits present sample data illustrating the effects of individual differences among students' test scores. A personal perspective is provided throughout the discussion. (SLD)