PURPOSE: To present a range of tests for the assessment of reading skills, and to investigate its reliability in measuring the performance of children from eight to 11 years and 11 months old. METHODS: It was created an instrument containing 12 test items organized into four assessment skills: Knowledge of Letters and Phonographemic Correspondence, Decoding of Isolated Items, Text-Reading Fluency, Reading Comprehension. For the reliability study, 100 children (64 girls) from public schools, with ages ranging from eight years to 11 years and 11 months, were selected. Twenty of these children (12 girls) participated in the applicability study, which resulted in the version of the Reading Scale used in the present study to assess the remaining 80 children (52 girls). The responses obtained were analyzed and scored by item, which allowed the calculation of a score for each assessment skill and the raw score. Data were statistically analyzed to obtain the Cronbach's alpha coefficient, used to calculate the instrument's reliability. Additionally, a study of correlations between items (Pearson's correlation coefficient) was carried out. A 0.05 level of significance was adopted. RESULTS: The internal consistency analysis obtained α=0.866 for the Reading Scale. Correlations between the items were observed in some scales, ranging from weak to strong, and corroborated the alpha coefficient values. CONCLUSION: The Reading Scale was a reliable instrument to measure the reading performance of the subjects.
Reading; Evaluation/methods; Speech therapy; Learning; Primary education