Predicting North Carolina third grade end-of-grade test of reading comprehension scores from first, second, and third grade variables

A significant body of research exists on the development of early literacy skills and their relationship to the development of literacy as a whole. Phonological awareness, orthographic processing, rapid automatized naming, phonological memory, and receptive vocabulary have all been shown to be predictive of early reading outcome measures. What is unknown, however, is whether or not these variables are predictive of the North Carolina third grade End of Grade Test of Reading Comprehension, a major outcome measure used in North Carolina to determine student, teacher, school, and district academic performance. This study addressed the following research questions: 1) Do the variables identified through a review of the literature as being predictors of reading achievement in first, second, and third grade contribute to scores on the reading EOG? 2) Does growth on measures of these variables, identified through a review of the literature as being predictors of reading achievement in first, second, and third grade contribute to scores on the reading EOG? Linear regressions were conducted on data consisting of 111 children. After controlling for age, IQ, and inherent characteristics of the data set, five predictor variables in grades one, two, and three were found to be significantly predictive of the outcome measure. A significant portion of variance was accounted for by receptive vocabulary at time point one; phonological awareness, RAN and receptive vocabulary at time point two; and orthographic processing and RAN at time point three. Further examination indicated phonological awareness and RAN were most predictive and time point two and orthographic processing was most predictive at time point three. Phonological memory was never significantly predictive at any given time point but contributed the most to outcome measure prediction at time point two. Findings suggest scores on the reading EOG are predicted by a child's development on skills key to the development of early literacy. An individual's literacy skills can be used to estimate later performance on this high stakes test of reading ability. The results of this study suggest students can be screened for potential EOG failure and interventions can be implemented to remediate key skills. This study also suggests a model for the evaluation of other high stakes outcome measures