Development of a Revised Version of the Statistical Anxiety Scale

Background: Statistics anxiety is a common problem in students taking statistics courses in the social sciences. It is most widely measured by the statistical anxiety scale. The various adaptations of this instrument have shown certain problems in the replication of its factorial structure and do not have a system to control possible response bias effects. The objective of our study was to propose a short test to measure statistical anxiety that also includes a scale to control social desirability bias. Method: We developed a revised version of the statistical anxiety scale using procedures for controlling response biases and examined its factorial structure using exploratory and confirmatory analysis in a sample of 531 students. Results: The revised version showed a clear four-factor structure in exploratory and confirmatory factor analyses with the expected three content factors plus one social desirability factor. The scales showed no acquiescence effects and moderate social desirability effects, and had a clear relationship with academic success. Conclusions: The revised version of the statistical anxiety scale improves on the psychometric properties of the original version and may overcome the problems detected in some adaptations of the previous version.