Examining the Predictive Validity of Universal Screening Measures of Reading in Grades 4-5

Abstract

Universal screening of reading is an increasingly common practice within schools (Silverman et al., 2021), and it’s often considered a core component of a multi-tiered system of supports (MTSS) framework (Compton et al., 2006; Jenkins et al., 2007). Universal screening of reading can be used to identify students that may benefit from additional supports, assess key academic skills, allocate resources effectively, and predict performance on state assessments (Ardoin & Christ, 2008; January & Klingbeil, 2020; Jenkins et al., 2007; Missall et al., 2021). Universal screening of reading measures vary in terms of length, content, and psychometric quality. The current study compared the predictive validity of three universal screening measures of reading: prior year Smarter Balanced Assessment – English Language Arts (SBA-ELA), iReady Reading Diagnostic, and Dynamic Indicators of Basic Early Literacy – Oral Reading Fluency (DIBELS-ORF). Additionally, the moderating effects of grade level and learning assistance on predictive utility were tested. The results indicated that all three measures significantly and substantially predicted end-of-year SBA-ELA scores, but that iReady Reading was the strongest predictor – albeit only slightly better than previous year SBA-ELA scores. In contrast, DIBELS-ORF did not contribute meaningfully to explaining variation in end-of-year literacy scores beyond the other two measures. Limitations and implications for data-based decision making and universal screening for reading are discussed. Keywords: universal screening of reading, data-based decision making, iReady Reading, DIBELS-ORF, SBA-ELA, multi-tiered system of supports