Fairness and Test Adaptations for Students With Functional Impairments

Abstract

Fairness is maximized under the condition that the evidence is most compelling for those individuals taking the test. With that assumption in mind, the authors address three considerations for research and practice in test adaptations and fairness, with an added consideration about future implications of these points. First, the authors question the paradigm whereby the definition of adaptations as accommodations versus modifications depends on whether a change was made to the testing occasion (accommodation) versus to the test itself (modification). They suggest using reliability and validity evidence to identify an adaptation as a successful accommodation versus a modification. Second, they question differential boost as a pervasive framework for gathering supporting evidence. They recommend prioritizing more direct studies of reliability and validity evidence across subgroups and conditions to determine differences or improvements in measurement. Third, they assert that fairness studies that broadly define groups by disability status are problematic, given the heterogeneity of those groups. They recommend that research on test adaptations identify groups based on empirically measured functional impairments, which are more likely than disability status to indicate which test takers need which adaptations. The authors conclude by addressing several misconceptions that highlight gaps between theory, research, and practice caused by misunderstandings about evidence for selecting adaptations.  

Keywords:  accommodations; adaptations; fairness; modifications; testing