A Psychometric Evaluation of a State Testing Program: Accommodated versus Non-Accommodated Students

Abstract

Federal legislation such as "No Child Left Behind" mandated that students with disabilities be included in accountability standards, creating an important responsibility to fairly assess all students, even those with disabilities. Consequently, a sense of urgency was placed on the entire educational system to ensure that these students had a fair chance at being adequately tested, to ensure equity and access. Alternate assessments and accommodations are a part of access for these students in the testing realm. This study, which centered around fairness, focused on psychometrically comparing three subject exams administered to eighth grade students who received accommodations as opposed to those who did not. Covariance matrices, dimensionality, factor structure, and item functioning were all compared across groups to examine invariance and show that the use of accommodations did not affect the validity or fairness of the testing program. Analyses revealed that for each of the tests, there was a significant difference in the covariance matrices, but unidimensionality held across groups implying that the dimensionality structure was the same for both groups. The factor structure was tested only for the math exam and the one-factor structure held for accommodated and non-accommodated students, again confirming unidimensionality. Despite consistent dimensions being measured for both groups, DIF did manifest but without actual test items it could not be attributed to bias. [The dissertation citations contained here are published with the permission of ProQuest LLC. Further reproduction is prohibited without permission. Copies of dissertations may be obtained by Telephone [phone]. Web page: http://www.proquest.com/en-US/products/dissertations/individuals.shtml.]