Previous research has shown a consistent, albeit weak, negative correlation (r ( -0.20) between statistics anxiety and statistics achievement. Additionally, self-efficacy has been shown to be a consistent predictor of both anxiety and achievement. This study showed that if self-efficacy is assumed to reflect a distribution of confidence, then the relationship between statistics anxiety and statistics achievement can be explained by the differential impact of two features of the self-efficacy distribution. Although only outcome expectancies predict statistics achievement, statistics anxiety is predicted by the interaction between outcome expectancies and outcome uncertainty. It is suggested that these results are indicative of at least two sources (or cognitive interpretations) of statistics anxiety, namely lack of confidence about one's ability and uncertainty in one's performance. The results are discussed in terms of cognitive appraisals of threat and challenge.
The CAUSE Research Group is supported in part by a member initiative grant from the American Statistical Association’s Section on Statistics and Data Science Education