Braille Reading Accuracy of Students Who Are Visually Impaired: The Effects of Gender, Age at Vision Loss, and Level of Education

Introduction: The present study assesses the performance of students who are visually impaired (that is, those who are blind or have low vision) in braille reading accuracy and examines potential correlations among the error categories on the basis of gender, age at loss of vision, and level of education. Methods: Twenty-one visually impaired Greek school-aged children participated in the present study. The students who participated were enrolled in different educational settings; that is, special schools and mainstream educational settings. The research tool was a subset (three subscales) of a standardized instrument (Test A, Padeliadu & Antoniou, 2008) that evaluates reading accuracy in Greek. All interactions between researchers and students were video-taped, and the analysis of the obtained data was focused on phonological and nonphonological-type errors. Results: Significant differences in performance were found between male and female participants-t(19) = 2.12, p < .05-as well as between students who attained primary and secondary education: t(19) = 1.96, p congruent to .05. The average number of errors in the three subscales correlated very highly, signifying that performance was very similar. Positive correlation was found between replacement and subtraction types of error (p < .05), and replacement and recognition (p < .001), and the total number of errors was positively correlated with replacement (p < .001), subtraction (p = .001), and recognition errors (p < .001). Male participants made more replacement errors: t(19) = 2.09, p congruent to .05; participants in secondary education made significantly fewer errors of recognition: t(19) < 2.49, p < .05; and students who were congenitally blind made significantly more errors of addition: t(19) = 1.96, p congruent to .05. Regarding the recognition type of error, there was a significant interaction effect between grade and age at loss of vision: F(3/17) = 3.09, p = .05. Discussion: Participants did not benefit exceptionally from semantic information, and it is unclear whether a higher school level leads to the improvement of braille reading accuracy. "Reading the entire word" seems the most effective decoding strategy. Nevertheless, further research is needed to obtain relevant data from longitudinal studies. Implications for practitioners: Listing and analyzing braille reading errors systematically may reveal error patterns. Based on these patterns, teachers would be able to differentiate their instruction to improve students' performances in braille speed and accuracy