Scotopic sensitivity/Irlen syndrome and reading in college level students

Abstract

This study investigated whether a recommended treatment for Scotopic Sensitivity/Irlen Syndrome (SS/IS), colored overlays, was more effective in improving reading disabled college students' reading scores than either reading skills instruction or no treatment. It was hypothesized that subjects using colored overlays would perform significantly better on posttest measures of Reading Rate and Reading Comprehension than would subjects receiving an alternative treatment or no treatment. To test these hypotheses, 30 subjects were selected who demonstrated low reading abilities on the Nelson-Denny Reading Test and who also demonstrated symptomatology of SS/IS in the severe range as measured by the Pre-Assessment for Scotopic Sensitivity. Subjects were then divided into three treatment groups of 10 subjects each: Group I - colored overlays, Group II - reading instruction, and Group III - no treatment. After receiving treatment, subjects were posttested on the Nelson-Denny for Reading Rate and Reading Comprehension. Pretest to posttest changes on these two dependent variables for the three treatment groups were analyzed for significance using Repeated Measures Analyses of Variance (ANOVAS). Significant Time by Group interaction effects were expected for both variables. Contrary to expectations, the ANOVA testing for the Reading Rate variable did not indicate a significant Time by Group interaction. However, results of the ANOVA performed for the Reading Comprehension variable, as predicted, did show a significant Time by Group interaction. In other words, there was significant differential improvement over time in Reading Comprehension related to group membership. An examination of group means for Reading Comprehension showed that Group I (colored overlays) improved more than Group II or Group III. Nonetheless, post hoc testing utilizing the Sheffé Test failed to detect significant posttest differences in pairwise comparisons of the three groups. It was hypothesized that the small number of subjects per group and the conservative parameters of the test were responsible for this failure. Due to the statistical outcomes obtained in this study, it was suggested that similar studies in the future utilize more appropriate measures of reading rate and increase the numbers of subjects.