Visual limitations can negatively affect academic achievement. Prior research conducted in international contexts revealed that the reading performance of students with low vision is poorer than that of their sighted peers. No study has compared the reading performance of Turkish students with low vision with that of their peers. Language features are among the factors that affect reading comprehension and development. In this context, it is necessary to determine whether there is a difference in the reading skills of Turkish students with low vision compared with their sighted peers. This study aimed to compare the reading skills (reading comprehension, reading rate, types of reading error, and total reading error percentages) of 3rd-, 4th-, and 5th -grade students with low vision and normally sighted students according to their grade levels and to examine whether the reading skills of both groups differed as they moved on to the next grade. A causalcomparative research model was applied to compare the reading skills of the two groups, and a correlational model was used to determine whether reading skills differed according to grade level. The participants were 3rd-, 4th-, and 5th -grade students from eight different cities studying at general education primary and secondary schools (n=87) and students with low vision (n=78) studying at primary and secondary schools for the visually impaired. The reading performance of students with low vision and normally sighted students was assessed using the Nonformal Reading Inventory developed by Karasu (2011). The reading rate of students with low vision was found to be low at all grade levels, and their total reading errors were high compared to normally sighted students. Regarding reading comprehension and types of reading error, different findings were obtained depending on grade level. As students moved to the next grade, their reading comprehension levels increased. When normal -sighted students moved to the next grade, their reading rates increased gradually. The findings are discussed in light of the related literature. Suggestions for further research and implementation are provided.
