Lewis, D., Green, D. R., & Miller, L. (1999, June). Using differential item functioning analyses to assess the validity of testing accommodated students with disabilities [Paper presentation]. Council of Chief State School Officers (CCSSO) National Conference on Large-Scale Assessment, Snowbird, UT, United States.

Students were grouped according to the conditions in which they took the test: extra time (extended-time) and read aloud (read-aloud), extra time alone, read aloud (which includes students who received the read aloud accommodation with or without additional accommodations), and extra time (which includes students who received the extra time accommodation with or without additional accommodations). There were not enough students who took the test with only the read aloud accommodation to allow for analysis of this accommodation alone.

Grade 10 students receiving special education services in Indiana (U.S.) were included in the study. Data were analyzed further for specific groups of students: those with identified intellectual disabilities and learning disabilities. Altogether, test data from over 3,000 students was analyzed.

Results of the 1997 administration of the Indiana Statewide Testing for Educational Progress Graduation Qualifying Exam (ISTEP+ GQE) were analyzed for statistical bias (differential item functioning) for students with special needs. English/Language Arts and Math test sections were analyzed. There were multiple choice and constructed-response items on both sections of the test (state assessment extant data).

Relatively few items exhibited differential item functioning (DIF) for students under the accommodated conditions. There were more items exhibiting DIF for English language arts items than for math items. The number of DIF items ranged from 1-3 for math items (depending on disability group), and 1-5 for language arts items. The authors point out that this study did not examine how items would function for students without disabilities who might receive accommodations.