Studying Statistics Anxiety Requires Sound Statistics: A Comment on Siew, McCartney, and Vitevitch (2019)

Empirical scientists cannot do without statistics. This fact is reflected by the pervasiveness of statistics courses in the curricula of essentially all scientific disciplines. Unfortunately, many students exhibit statistics anxiety, that is, “feelings of anxiety [. . .] when taking a statistics course or doing statistical analyses” (Cruise, Cash, & Bolton, 1985, p. 92). In a recent publication, Siew, McCartney, and Vitevitch (2019) aim to shed new light on this highly relevant topic by using data analysis tools from the field of network science. However, just as with any other statistical model, one has to carefully assess the adequacy and robustness of a network model. In this commentary, we point to a number of shortcomings in the article by Siew et al. (2019) with respect to this goal that question their main conclusions. We explain each problem and suggest ways to address it. We hope that these suggestions help to put future investigation of statistics anxiety using network models on a firm methodological basis.