Scores on the Prescriptive Reading Inventory, the California Achievement Tests, 1970 Edition, the the Short Form Tests of Academic Aptitude were obtained for black pupils and representative samples of pupils in grades 1-3. These scores were compared in an attempt to asses bias in the Prescriptive Reading Inventory, a criterion-referenced achievement test. Using factor analyses, contingency analyses, and regression analyses, interrelationships among the reading scores on the two tests were compared. It was hypothesized that the norm-referenced test would show more bias as indicated by differential relationships among scores. The regression analyses indicated some bias in all the tests used. The contingency analyses, made of the CRT only, suggested little bias. The factor analyses suggested parts of the CRT were not biased but the remaining parts and the NRT measures were. In all cases, the amount of bias appeared small. (Author)