I can’t always write a nuanced post about something I come across online. So here’s a link to a paper I saw on Reddit that is a good review of the problems of relying on p values to establish the credibility of a study. It focuses on the social sciences, a discipline which is infamous for its obsession with p values. What’s the way we beat this? As always, better statistical education in our universities and ongoing statistical training for researchers.