Abedi, J., Kao, J. C., Leon, S., Sullivan, L., Herman, J. L., Pope, R., Nambiar, V., & Mastergeorge, A. M. (2008). Exploring factors that affect the accessibility of reading comprehension assessments for students with disabilities: A study of segmented text (CRESST Report No. 746). University of California, Los Angeles, National Center for Research on Evaluation, Standards, and Student Testing. https://cresst.org/publications/cresst-publication-3117/

The impact on accessibility of converting reading comprehension passages into shorter “segments” or “chunks” was examined.

A total of 738 grade 8 students participated, from 10 public schools in California (U.S.), including two in the northern part, and the remaining in the southern part of the state. Participants included 117 students (15.9%) with disabilities. Additional demographic data were reported, including ethnicity, sex, and language learner status (30% were English language learners). [Note: about 68% (n=79) of the students with disabilities were English language learners.]

A reading comprehension test was the primary instrument used. Secondary instruments included student background questions and student motivation scale, which were printed inside the reading test booklet. There was also a student background questionnaire for teachers to complete.

"The results of the segmenting study indicated that: (a) segmenting did not affect reading performance of students without disabilities; suggesting that it does not compromise the validity of reading assessment; (b) segmenting did not affect reading performance of students with disabilities; (c) the segmented version had a higher reliability for students with disabilities without affecting the reliability for students without disabilities; and (d) no trends were observed with student motivation, general emotions and moods with respect to the segmented assessment. The study also introduced the idea of incorporating some commonly used accommodations for students with disabilities, such as test breaks, into the assessment" [from Abstract]. Limitations of the study were reported, and future research possibilities were suggested.