Current Research and Statistical Practices in Sport Science and a Need for Change

Current research ideologies in sport science allow for the possibility of investigators producing statistically significant results to help fit the outcome into a predetermined theory. Additionally, under the current Neyman-Pearson statistical structure, some argue that null hypothesis significant testing (NHST) under the frequentist approach is flawed, regardless. For example, a p-value is unable to measure the probability that the studied hypothesis is true, unable to measure the size of an effect or the importance of a result, and unable to provide a good measure of evidence regarding a model or hypothesis. Many of these downfalls are key questions researchers strive to answer following an investigation. Therefore, a shift towards a magnitude-based inference model, and eventually a fully Bayesian framework, is thought to be a better fit from a statistical standpoint and may be an improved way to address biases within the literature. The goal of this article is to shed light on the current research and statistical shortcomings the field of sport science faces today, and offer potential solutions to help guide future research practices.