Pitfalls of p-value hypothesis testing: How statistical methods affect pychological research
Eric-Jan Wagenmakers: Pitfalls of p-value hypothesis testing: How statistical methods affect pychological research
In cognitive psychology, the practice of p-value null-hypothesis testing is as widespread as ever. Despite its popularity, or perhaps because of it, most cognitive psychologists are not aware of the statistical peculiarities of the p-value procedure. For instance, p-values are based on data that were never observed, and these hypothetical data are themselves influenced by subjective intentions. In the first part of this presentation, practical examples are used to highlight the theoretical problems of p-value null-hypothesis testing. In the second part, a Bayesian alternative is proposed and discussed. It is argued that when cognitive psychologists start to adopt Bayesian methods of inference this will alter and improve the way they do science in fundamental ways.