The validity of the equipercentile hypothesis of the Title I Evaluation and Reporting System (TIERS) norm-referenced evaluation model was examined. The California Achievement Test, Reading, was administered as a pretest and posttest to 3,224 seventh and ninth grade students. The equipercentile hypothesis predicts that the posttest percentile status would be the same as the pretest percentile status for students not receiving special education programs. Students' gains at 10 different achievement levels were evaluated employing the norm-referenced model. The findings contradicted the equipercentile hypothesis. There was a clear pattern of large gains for students not receiving any special educational instruction. (Author/CM)