Source URL: View this document on the ASHA website
This was one of the first of many articles publishing research demonstrating the severe limitations of using commercially available child language tests when assessing children for speech and language disability.
Background: The authors of this article reviewed thirty preschool articulation and language tests for ten psychometric measures considered necessary for norm-referenced tests, including measures of validity, reliability, and accuracy.
Methodology and Participants: The criteria used in this review are a selected sample of a larger number of important psychometric criteria. Adherence to more numerous and, in some cases, stricter guidelines is commonly considered necessary for a well-developed norm-referenced test.
Conclusion: Just half of the tests surveyed met two of the ten criteria, and four of 30 reviewed tests met four of the criteria. Only three tests met over four criteria (the TOLD, ITPA, and PPVT-II). The authors found a significant lack of empirical evidence of validity and reliability in determining the presence of language disorders. Psychometric measures that were most frequently unmet included description of the sample, evidence of concurrent validity, test-retest reliability, predictive validity, and inter-examiner reliability. The latter two measures were not met by any of the reviewed tests. The study demonstrated the grave state of norm-referenced tests at the time of this review. Nearly all of the tests lacked even the most basic information about validity, reliability, and accuracy regarding the tests’ reported purpose of diagnosing a child with a speech and language disability. The authors concluded that tests that do not provide evidence of acceptable standardization samples, validity, reliability, and lack of significant bias should not be considered good diagnostic tools.
Relevance to the Field: This was one of the first of many articles publishing research demonstrating the severe limitations of using commercially available child language tests when assessing children for speech and language disability. It also highlighted the need for clinicians to review the materials they use, personally, when performing evaluations for evidence of validity, reliability, and accuracy. Clinicians, as the consumers of these tests, can drive change by ensuring that the testing materials they purchase meet certain levels of validity, reliability, and accuracy and are free of bias against the populations with which they work. Since the vast majority of tests reviewed lacked information regarding validity (and continue to lack validity today), this study provided further support against using standardized tests as a way to determine eligibility for speech and language services.
McCauley, R. J., & Swisher, L. (1984). Psychometric review of language and articulation tests for preschool children. Journal of Speech and Hearing Disorders, 49, 34-42.