Examining Differential Distractor Functioning in Reading Assessments for Students with Disabilities

Abstract

The University of Minnesota is committed to the policy that all persons shall have equal access to its programs, facilities, and employment without regard to race, color, creed, religion, national origin, sex, age, marital status, disability, public assistance status, veteran status, or sexual orientation. This document is available in alternative formats upon request.

A Nested Logit Approach for Investigating Distractors as Causes of Differential Item Functioning

“…In this context, the concept of differential distractor functioning (DDF) is often used to describe conditions where individual distractors display different levels of attractiveness to the studied groups. However, the concept is not used consistently in the psychometric literature, sometimes referring to the differential functioning of a distractor relative to other distractors (e.g., Abedi, Leon, & Kao, 2007; Banks 2009; Green, Crone, & Folk, 1989), and in other places to differential functioning of a distractor relative to all response options, including the correct response (Dorans, Schmitt, & Bleistein, 1992; Penfield, 2008). Although both formulations can be practically useful, they are not equivalent.…”

Differentials of a State Reading Assessment: Item Functioning, Distractor Functioning, and Omission Frequency for Disability Categories

“…As part of the PARA project, Abedi, Leon, and Kao (2007a, 2007b) examined test item characteristics that could cause or signal inappropriate barriers to successful test performance for students with disabilities. They compared the test performance of students with and without disabilities in terms of item bias.…”

“…While the literature of DDF analysis is not very abundant (e.g., Abedi et al, 2007b; Banks, 2006; Green et al, 1989; Marshall, 1983; Middleton & Laitusis, 2007), there is some methodological variation such as the log‐linear approach (Green et al, 1989), the logistic regression approach (Abedi et al, 2007b), and the standardization approach (Dorans, Schmit, & Bleinstein, 1992). The log‐linear and logistic regression approaches specifically focused on DDF using incorrect responses only (e.g., Abedi et al, 2007b; Green et al, 1989). These approaches are appealing, because they enable simple interpretations, and including the correct response in the model sometimes overwhelms subtle DDF (Green et al, 1989).…”

Computer‐adaptive Testing for Students With Disabilities: A Review of the Literature

“…Although improvements have been made in this area, it remains the case that tests are most often made accessible to students with disabilities retrospectively, through the use of accommodations that may affect measurement of the construct and alter the meaning of resulting test scores. Second, a significantly higher proportion of students with learning disabilities fall into the lower tail of the proficiency distribution.Because state tests are usually designed to provide the most accurate measurement in the midrange of the ability scale, where the bulk of students are located, students with proficiencies on the low end of the scale may have no alternative but to guess randomly on many questions (Abedi, Leon, & Kao, 2007;Minnema, Thurlow, Bielinski, & Scott, 2000). Empirical evidence 2 of this was provided by Laitusis, Buzick, Cook, and Stone (2011, p. 293), in which the percentage of students with learning disabilities scoring at chance level on 4th-and 8th-grade mathematics and English-language arts assessments from one state ranged from 12-22%, while the percentage for students without disabilities ranged from 1-3% on the same tests.…”

“…Because state tests are usually designed to provide the most accurate measurement in the midrange of the ability scale, where the bulk of students are located, students with proficiencies on the low end of the scale may have no alternative but to guess randomly on many questions (Abedi, Leon, & Kao, 2007;Minnema, Thurlow, Bielinski, & Scott, 2000). Empirical evidence 2 of this was provided by Laitusis, Buzick, Cook, and Stone (2011, p. 293), in which the percentage of students with learning disabilities scoring at chance level on 4th-and 8th-grade mathematics and English-language arts assessments from one state ranged from 12-22%, while the percentage for students without disabilities ranged from 1-3% on the same tests.…”