Reading strategy use scale: an analysis using the rasch rating scale model

Abstract

Although reading strategy use can be measured using different methods, self-report measures are particularly cost-effective in educational settings. The main goal of this study was to investigate the psychometric properties of the items of the Reading Strategy Use Scale (RSU), using the Rasch Rating Scale Model. Although the RSU has already been studied using classical analyses, some issues remain regarding the suitability of the response scale and the items’ validity. A total of 179 students at- tending the fifth and sixth grades participated in this study. The results supported a unidimensional structure and local independence of the items. The seven-category response scale showed inadequate functioning. Rather, the results suggested that a five-category scale was more appropriate. Rasch reliability indicators were very high. There was no meaningful differential item functioning as a function of gender. A test of differences indicated that girls use reading strategies more frequently than boys. A positive correlation between the RSU scores and a reading comprehension measure was found. The findings support RSU as a robust self-reporting instrument to measure the frequency of use of reading strategies and highlight the usefulness of Rasch analysis in developing robust educational measures.