Lin, P. Y., & Lin, Y. C. (2013). The impact of setting accommodation on large-scale assessment of English language learners with and without learning disabilities: Balanced vs. unbalanced data in latent class analyses . Journal of Studies in Education , 3 (2), 1–19. https://doi.org/10.5296/jse.v3i2.3371

A setting accommodation for grade 6 students with learning disabilities in math was explored.

Grade 6 students in Ontario, Canada who took the Junior Assessment of Reading, Writing, and Mathematics in 2005-2006 participated. In the unbalanced dataset, 114,034 students participated; in the balanced dataset, 1,843 students participated. Students’ test results were excluded if they took the French version of the exam, were enrolled in a French immersion program, were exempt from one or more portions of the exam, or were missing any necessary information. Students were only included if they received no accommodations or only setting accommodations. 45% of students spoke languages other than English.

Six multiple choice items that measure number sense and numeration from the math portion of the Junior Assessment of Reading, Writing, and Mathematics were used as outcome measures. Latent class analyses were used to determine whether examinees with different characteristics showed a distinct pattern of item responses. ANOVA analyses were used to assess the effect of English language learner status, learning disabilities, and a setting accommodation on students’ math outcomes.

Students with different characteristics had similar response patterns, with some exceptions. Students without a learning disability who had an accommodation performed significantly worse on the math items than did their peers who took the test in standard testing procedures. Students with learning disabilities had no significant difference in performance when comparing a setting accommodation with standard testing procedures. Home language was not a significant indicator of the effect of the setting accommodation on math performance.