Reframing the debate
"The findings of this study underscore the need to reframe the issue of testing accommodations as one that is relevant to all students, and not only to students with disabilities," writes Elbaum. In designing assessment systems, educators should incorporate testing accommodations such as oral presentation of items on a mathematics test for general education students as well as students with LDs, the article states.

Do the results of the study mean the oral accommodation is not fair or invalid? Elbaum asks. One view on accommodations is that to be valid they should improve the performance of students with disabilities while having no effect on students without disabilities.

For example, allowing students with motor difficulties to dictate their answers to a scribe addresses the students' disability but would not be expected to improve the performance of students without motor impairments.

Click here to subscribe to Educational Research Newsletter 

In this study, student were tested with  two 30-item mathematics tests with items drawn from practice materials from statewide tests. The two tests were administered in one class period--one in read-aloud and one in standard condition.

Rather than proving that oral accommodation is not valid or fair, the researcher proposes that the results of the study bolster an argument by S. Sireci (2005) that testing accommodations may simply allow more precise measurement of students' abilities.

More precise measurement
"If the primary objective of implementing testing accommodations is not that of 'closing the gap', but rather that of achieving a more precise measurement of students' abilities in an area such as mathematics, then it would follow that the oral accommodation ought to be regularly offered to all students," Elbaum writes.

Besides showing that general education students can benefit from read-aloud accommodation in mathematics testing, the study also showed that not all students with LDs benefit from this accommodation. Approximately 8% of students with disabilities performed significantly less well with a read-aloud accommodation for reasons that are not clear, the researcher says.

A previous study found that general education students "became impatient with the time needed to finish reading the items aloud and reported that they disliked the pacing of the test in the accommodated condition." In this study, students in the read-aloud condition were encouraged to follow along with the test administrator, who read each item aloud twice.

Not appropriate for all LD students
Students with LDs should only be assigned to an accommodated testing condition based on prior evidence indicating the student is likely to benefit from it, the researcher notes.

Though blanket assignment of students to an oral accommodation might improve group performance, the author notes, for a small minority of students the testing accommodation would introduce another variance in results.

One of the purposes of the study was to show the effect of accommodations with an older set of students--many such studies have been done with elementary students. When the results of this study were combined with those of 13 previously published studies, Elbaum says the impact of oral accommodations on students' mathematics performance was not the same for elementary and secondary students.

Among elementary students, "the accommodation boost for elementary students is clearly of greater magnitude for students with LD than it is for students without LD."

Among secondary students, the read-aloud accommodation may have helped improve performance because of fewer errors owing to failure to encode or to make careful distinctions among response choices.

"Effects of an Oral Testing Accommodation on the Mathematics Performance of Secondary Students With and Without Learning Disabilities", by Batya Elbaum, The Journal of Special Education, Volume 40, Number 4, 2007, pp. 218-229.

Published in ERN February 2007 Volume 20 Number 2