Many psychology students experience statistics anxiety, which negatively affects academic performance and necessitates the application of new methods to understand the underlying nature of statistics anxiety. In the present analysis we use the tools of network science, a set of mathematical techniques to examine the relationships between entities in a system, to create networks from the items in the Statistical Anxiety Rating Scale (STARS), a widely used instrument to measure statistics anxiety. Separate networks were constructed and analyzed for students who were high or low in statistics anxiety. The items in the STARS were the nodes in the network, with connections placed between nodes if responses to those 2 items in the STARS were correlated. The network analysis revealed differences in the overall structure of high-and low-anxiety networks and in which nodes were most "important" in the 2 groups, suggesting that different interventions might be required depending on the level of statistics anxiety experienced by the student. Furthermore, our results suggest that the most effective interventions for students with very high levels of statistics anxiety would need to specifically address the (incorrect) perception that strong mathematical ability is a necessary prerequisite for excelling in statistics. Centrally located nodes in the network may help instructors appropriately target classroom interventions to help all students succeed regardless of the level of statistical anxiety experienced by students.

No abstract

No abstract

No abstract