Kuti, L. M. (2011). Accommodations for English language learners with disabilities in federally-mandated statewide English language proficiency assessment (Publication No. 3468551) [Doctoral dissertation, Virginia Commonwealth University]. ProQuest Dissertations and Theses Global. https://www.proquest.com/docview/888048530

Virginia Commonwealth University (Richmond, VA); ProQuest document ID: 888048530

Accommodations that were provided during Virginia's English language proficiency test included (in decreasing order of frequency): modified test directions, modified timing/scheduling, other practices for reducing test anxiety, modified presentation format, scribe, computer assistance, audio amplification, braille, magnification or low vision aid, and large print in test booklet.

Students in grades 3 through 12 who were English language learners (ELLs) who also had disabilities—as identified by the Virginia (U.S.) Department of Education—contributed data to this study. The statewide population of ELLs numbered 52,517 in the 2009–2010 school year. Those ELLs who had disabilities were 7,002 in all; numbers of students in specific disability categories were also reported. Eight (8) educators from one district, purposively sampled due to its representativeness of Virginia's population, were interviewed for the qualitative part of the study.

Extant data were analyzed from Virginia’s English language proficiency (ELP) assessment, called the Assessing Comprehension and Communication in English Test (ACCESS), which was created by the World-Class Instructional Design and Assessment (WIDA) Consortium. This test has five content areas: language arts, mathematics, science, social studies, and social and instructional language. The assessment requires testing in four domains: reading, writing, listening, and speaking. Additional information for qualitative analysis was drawn from interviews with educational professionals at the levels of state, district, and school.

Students with disabilities who completed the ACCESS for ELLs testing had various disabilities, the most common of which were learning disabilities (66%), speech/language impairment (14%), and other health impairment (8%). These ELLs with disabilities were provided 11 different accommodations. The most commonly provided accommodations included modified test directions, modified timing, other accommodation (practices for reducing test anxiety), and modified presentation format. Statistically significant relationships were found between the four most commonly-occurring disabilities and three of the four most commonly-provided accommodations. Further, educators reported that those accommodations were often linked with those high-incidence disabilities, and offered specific reasons for those associations. ELLs with disabilities who used accommodations, on average, scored lower on the ELP assessment than those ELLs with disabilities not using accommodations. Two exceptions to this tendency were: the ELLs with other health impairments who used modified test directions scored better than those who did not, and the ELLs with emotional disabilities who used other approved accommodations scored better than those who did not. Additional factors involved with this relationship noted by quantitative data were the grade level of students and the amount of time they spent in the ELL programs, and five other factors identified by educators such as familiarity with using specific accommodations throughout the year. Finally, findings for the fourth research question included that educators reported challenges pertaining to the policy limitations about the ELP test assessing the four domains of reading, writing, listening, and speaking for students with visual impairments and students with hearing impairments, as well as reading disabilities and significant cognitive disabilities. Other issues discussed included test administration issues such as fidelity to guidelines and actual use of accommodations provided. Limitations of the study were reported, and future research directions were suggested.