Fox, L. A. (2012). Performance of students with visual impairments on high-stakes tests: A Pennsylvania report card (Publication No. 3532828) [Doctoral dissertation, University of Pittsburgh]. ProQuest Dissertations and Theses Global. https://www.proquest.com/docview/1235335586

University of Pittsburgh (Pittsburgh, PA); ProQuest document ID: 1235335586; also available on University of Pittsburgh webpage at: http://d-scholarship.pitt.edu/13954/1/foxla_etdPitt2012.3.pdf

Accommodations provided naturalistically to the state population on state assessments were examined, including those accommodations for students with visual impairments, students with hearing impairments, and students with learning disabilities.

Test scores in three different years from the entire special education student population of Pennsylvania (U.S.) in grades 3, 8, and 11 formed the control group of approximately 441,000 score sets; the comparison groups included test score sets in three different years from students with visual impairments (VI; n=1,340), students with hearing impairments (HI; n=3,890), and students with learning disabilities (LD; n=251,280). Additional demographic variables were reported. Since three different years of test scores were available, analyses were completed both by-grade and by-cohort. One cohort analysis was examining any changes in accommodations use from year to year for students with different disabilities.

Extant data were studied from the grade-level Pennsylvania System of School Assessment (PSSA) in mathematics and reading for grades 3–8 and grade 11 from three academic years (2005–2006, 2006–2007, and 2007–2008).

Students with visual impairments (VI) scored higher on average for most grade levels in both reading and math than the other two comparison groups of students with disabilities (with HI and with LD). Students with VI generally scored higher in math than reading. In terms of the proportion of students passing math and reading, just over half of the students with VI performed at or above proficiency level in both math and reading. Further, when a subset of students with VI were tracked across years, just under half of them (44%) performed at or above proficiency in reading, and just over half (54%) did so in math, across all three years. On the other hand, almost one-third persisted in scoring below proficiency in math (31%) and reading (31%) across all three years. For students with VI, the most commonly used accommodation in both math and reading for the three test years was large print, followed in popularity by small group, separate room (individual), and extended time. For only some students with VI, there were a few patterns of accommodations bundles in either reading or math, such as large-print and extended-time and magnification and administrator-transcription and sometimes dictated-response. Another bundle used consistently across years for some students was braille and extended-time and separate room (individual) and dictated-response and sometimes braille writer. However, from a longitudinal view of accommodations use, there were very few trends across years in the number or type of accommodations; that is, students typically might be provided specific accommodations in one year but different accommodations in another. Limitations of the study were reported, and future research directions were suggested.