Abstract
Validation is an important step in the developmental process of any reading test. Yet traditional approaches to reliability and validity assessment are not relevant for criterion-referenced reading tests. A more relevant approach defines reliability in terms of the consistency of decision-making across repeated test administrations and validity in terms of the accuracy of decision-making between an administration of a reading test and an administration of a criterion measure. A technique for measuring this type of reliability and validity is described in this paper along with an example of how it was used with a criterion-referenced reading battery.
