A Comparison of Single and Multi-Test Working Memory Assessments in Predicting Academic Achievement in Children

Abstract

Children assessed as having low working memory capacity have also been shown to perform more poorly than their same-aged peers in measures of academic achievement. Early detection of working memory problems is, therefore, an important first step in reducing the impact of a working memory deficit on the development of academic skills. In this study, we compared a single-test assessment, the Working Memory Power Test for Children (WMPT) and a multi-test assessment, the Automated Working Memory Assessment (AWMA), in their ability to predict academic achievement in reading, numeracy, and spelling. A total of 132 Australian school children (mean age 9 years, 9 months) participated in the research. Strong positive correlations between the WMPT and AWMA total scores were found, indicating good convergent validity of the single and multi-test measures. WMPT scores correlated with each of the four AWMA subtests designed to assess verbal and visuospatial short-term and working memory. WMPT and AWMA scores separately predicted performance on Word Reading, Numerical Operations, and Spelling. Compared with either measure alone, the WMPT and the AWMA in combination predicted more of the variance in Word Reading and Numerical Operations, but not in Spelling. Theoretical and practical implications of these findings are discussed.