Abstract:
The effects of curriculum-test overlap on standardized reading test performance (WRAT-R, K-TEA, and WRMT-R) were assessed with 181 first- and second-grade students receiving reading instruction from the MacMillan-R (Smith & Arnold, 1986) reading curriculum. Two methods of quantifying curriculum-test overlap were used to predict the patterns of students’ performance across the three tests and to compare their relative power for prediction. Results were consistent with other findings (Good & Salvia, 1988), examining different grade levels, tests, and curricula, indicating a strong and systematic curriculum bias which could confound educational decisions based on test scores. Neither method of quantifying curriculum-test overlap consistently outperformed the other indicating both are suited for use by researchers and assessment practitioners.
Additional information
Notes on contributors
Paul F. Bell
Paul F. Bell, PhD, received his doctorate from the University of Cincinnati in 1990. He is currently Director of Behavioral Sciences at the Family Practice Residency Program, The Medical Center, Beaver, PA, Inc.
Francis E. Lentz
Francis E. Lentz, Jr., PhD, is Associate Professor of School Psychology at the University of Cincinnati. His research interests include behavioral assessment and intervention strategies for school-based problems.
Janet L. Graden
Janet L. Graden, PhD, is Professor and Director, School Psychology Program, University of Cincinnati.