Science AMA Series. PLOS Science Wednesdays: I’m Jelte Wicherts, here to talk about a data simulation study published in PLOS ONE that showed why publishing everything is more effective than just publishing significant findings — AMA! : science

"Hi Reddit, My name is Jelte Wicherts and I am an associate professor at Tilburg University in The Netherlands, where I lead the Human Factors in Statistics group. Our research focuses on how researchers in the social sciences deal with the analysis of their data, errors with statistics, publication bias, and reproducibility. I recently published a study titled Why Publishing Everything Is More Effective than Selective Publishing of Statistically Significant Results in PLOS ONE. In this article, we argue that the best way forward in science is to publish all results rather than just those that are significant. We used simulation, a data fabrication technique that is ethically acceptable because it lets us see what happens in particular circumstances, to show that publishing all relevant results leads to much better estimates of effects in the literature. Publishing these results in PLOS ONE aligns closely with the journal’s guidelines to not focus exclusively on novel or important results, but instead publishes sound research regardless of whether results are significant or not. I will be answering your questions on May 27 at 1pm EST — AMA! Don’t forget to follow me on Twitter @JelteWicherts."