Decision Accuracy of Commonly Used Dyslexia Screeners Among Students Who Are Potentially at Risk for Reading Difficulties

Abstract

Students with dyslexia demonstrate reading difficulty in early literacy skills (e.g., phonemic awareness, word recognition, decoding), and administering screeners is a necessary step to implement effective intervention. There are several commonly used reading screeners, but the decision accuracy and predictive value between them varies. In the current study, scores on two different reading screeners, the Shaywitz DyslexiaScreen (SDS) and Dynamic Indicators of Basic Early Literacy Skills Next (DIBELS Next) were compared for 115 U.S. Grades K–3 students with specific reading deficits using the Phonological Awareness Composite of the Comprehensive Test of Phonological Processing as the criterion. Tests of sensitivity, specificity, and post-test probabilities were used to evaluate the decision accuracy of the data. Results suggested that the decision accuracy for DIBELS Next (78%) was better than SDS (45%), and both sensitivity (DIBELS Next = 90%, SDS = 35%) and positive post-test probability (DIBELS Next = 71%, SDS = 42%) favored DIBELS Next. Thus, the DIBELS Next measures demonstrated acceptable decision accuracy in identifying students with low phonological awareness, but the SDS did not.