Tindal, G., Heath, B., Hollenbeck, K., Almond, P., & Harniss, M. (1998). Accommodating students with disabilities on large-scale tests: An experimental study . Exceptional Children , 64 (4), 439–450. https://doi.org/10.1177/001440299806400401

Reading and math multiple-choice tests were completed either by bubbling an answer sheet or directly marking the booklet. Math tests were orally read in their entirety by the test administrator (read-aloud), including the general directions (for filling out the forms and taking the test), each specific problem, and all options for multiple choice problems. The reading of the math test was standardized.

A total of 481 students in grade 4, from 22 classrooms in seven buildings, were included as participants. The average age was 10.3 with a range from 9–12 years old. Most students were White, 3.7% were Hispanic, and very few other ethnicities were represented. Most students (97%) reported English as their first language. A total of 78 students were being served in special education (171 different IEPs).

The dependent variable was the percent correct on the statewide math and reading test (total correct possible was 30 per format).

General education students performed significantly higher than special education students in reading and in math. For both tests, performance was not higher when students were allowed to mark the booklet directly than when they had to use a separate bubble sheet. Students in special education with IEPs in reading or math performed significantly higher when the math test was read by teachers, rather than when they read the test themselves. In contrast, the performance of the 10 lowest achievement-ranked students in general education revealed no improvements when teachers orally read the math test over that achieved when students silently read the math test.