This study was designed to investigate the construction of a new measure for assessing reading performance which would prove to be a more sensitive and useful instrument for measuring the achievement growth induced by reading programs than the usual standardized tests. This report outlines the procedures followed in generating the model and constructing the tests, and presents some preliminary data collected in the course of field testing the new instrument. The hypothesis that the criterion-referenced tests would prove to be more sensitive measures of reading improvement over a lengthy period of time when compared to norm-referenced tests was partially confirmed. The results of the study are presented in narrative and table form. (RB)