Survey of attitudes toward statistics: Factor structure and factorial invariance for women and men

Abstract

White undergraduate students enrolled in introductory statistics courses completed the Survey of Attitudes Toward Statistics (SATS). Student responses were analyzed using confirmatory factor analysis to determine the validity of several alternative models of the structure of attitudes toward statistics. A four‐factor structure composed of Affect, Cognitive Competence, Value, and Difficulty provided the best model fit. Responses of women and men to the SATS were then examined to explore scale invariance across gender. Invariance tests of factor structure, factor‐variable correlations, factor correlations, and variable uniquenesses showed that the four‐factor model fit well for both female and male responses. Only two significant differences in model fit were found: the Value variance was greater for women and the correlation between Affect and Value was greater for men. Generally, the study provided evidence for the appropriateness of a four‐factor structure of the SATS.