The Harris Index of Efficiency was used to evaluate the criterion setting (66%) of the Prescriptive Reading Inventory (PRI), a criterion referenced reading test. The subjects, who were involved in a larger experimental study that used the PRI as a criterion measure, were 92 third and fifth grade students. Initial assessment with the Comprehensive Tests of Basic Skills revealed varying student achievement that necessitated the administration of various levels of the PRI on a pre/posttest basis. After these tests, the students were classified as masters/nonmasters using three separate criterion scores (66%, 75%, and 90%). The Harris Index of Efficiency was computed for each criterion score to determine which score proved most efficient at sorting the pupils into two separate groups. The results suggest that the criterion score of 66% was the most efficient cutting point with the PRI; a much less efficient sorting of the subjects was evident when using the 75% and 90% levels. The results also indicate that the Harris Index is an effective descriptive measure that can help to determine at what cutting point a test might yield the most information for instructional purposes. (RL)