Effects of linguistic complexity and accommodations on estimates of ability for students with learning disabilities

Abstract

Many students with learning disabilities (SLD) participate in standardized assessments using test accommodations such as extended time, having the test items read aloud, or taking the test in a separate setting. Yet there are also aspects of the test items themselves, particularly the language demand, which may contribute to the effects of test accommodations. This study entailed an analysis of linguistic complexity (LC) and accommodation use for SLD in grade four on 2005 National Assessment of Educational Progress (NAEP) reading and mathematics items. The purpose of this study was to investigate (a) the effects of test item LC on reading and mathematics item difficulties for SLD; (b) the impact of accommodations (presentation, response, setting, or timing) on estimates of student ability, after controlling for LC effects; and (c) the impact of differential facet functioning (DFF), a person–by-item-descriptor interaction, on estimates of student ability, after controlling for LC and accommodations' effects. For both reading and mathematics, the higher an item's LC, the more difficult it was for SLD. After controlling for differences due to accommodations, LC was not a significant predictor of mathematics items' difficulties, but it remained a significant predictor for reading items. There was no effect of accommodations on mathematics item performance, but for reading items, students who received presentation and setting accommodations scored lower than those who did not. No significant LC-by-accommodation interactions were found for either subject area, indicating that the effect of LC did not depend on the type of accommodation received.