Approaches to learning: factor invariance across gender

The aim of this study was to investigate the factorial invariance of a 30-item version of the Revised Approaches to Studying Inventory (RASI) across male (N=137) and female (N=171) university students. Confirmatory factor analysis was used to test the fit of the data obtained on the 12 scale scores for male and female samples to the three (higher-order) factor model. The hypothesised three-factor model fit for both samples. However, the model fit for the female sample was better than that of the male sample. A range of progressively restrictive models are tested to establish whether the RASI is equivalent across gender. Factor loadings, error variances, and factor covariances were not invariant, indicating that test constructs may be differentially indicated across groups. These findings, and implications for educational researchers, are discussed.