Item

Abstract

The replication crisis has hit several scientific fields. The most systematic investigation has been done in psychology, which revealed replication rates less than 40% (Open Science Collaboration, 2015). However, the same problem has been well documented in other disciplines, for example preclinical cancer research or economics. It has been argued that one reason for the high prevalence of false-positive findings is the application of "creative" data analysis techniques that allow to present nearly any noise as significant. Researchers who use such techniques, also called "p-hacking" or "questionable research practices", have higher chances of getting things published. What is the consequence? The answer is clear. Everybody should be equipped with these powerful tools of research enhancement. This talk covers the most commonly applied p-hacking tools, and shows which work best to enhance your research output: "If you torture the data long enough, it will confess!". But be careful: recently developed tools allow the detection of p-hacking. The talk also covers some ideas how to overcome the replication crisis.