Validity and Reliability of Criterion-Referenced Measures: Issues and Procedures for Special Educators

Abstract

The controversy over norm-referenced versus criterion-referenced measures is of particular interest to special educators. Over the years, special educators have been encouraged to use CRMs in assessment of learning disorders in order to avoid the biasing and other undesirable features of standardized tests. However, as this article points out, the choice between the two types of measures is not as simple as it might seem. While the reliability of CRMs is generally indisputable, the validity of a measure designed from local curricula is still dependent on the intuitive judgments of teachers. In discussing the advantages and disadvantages of both types of tests, Harris and Wolf emphasize that tests should be selected according to the intended use of the scores, rather than being an all-or-nothing decision. Ultimately, the decision depends on the degree of faith one places in global constructs versus local curricula.