Abstract
The Statistical Anxiety Rating Scale has been conceptualized as multidimensional. Three of the six theoretical dimensions, Interpretation Anxiety, Test and Class Anxiety, and Fear of Asking for Help, tap anxiety about statistics; the other three dimensions, Worth of Statistics, Computation Self-concept, and Fear of Statistics Teachers, measure perceptions and attitudes toward statistics. This conceptualization of a second-order factor model consisting of two superordinate factors and six first-order factors has remained empirically untested. In 77 graduate students (15 men, 62 women; M age = 27.6 yr.) enrolled in applied statistics courses in Taiwan, a confirmatory factor analysis of two competing models based on subscale scores yielded results supporting the multidimensional construct validity of the scale. As multidimensionality was previously supported based on item scores, but not subscale scores, the findings are interpreted as partially consistent with previous research testing the theoretical factor structure of the scale.
Get full access to this article
View all access options for this article.