December 5th, 2017, 1 pm
GESIS, Cologne, Conference room East
The reproducibility of research findings is a core criterion of science. Large replication projects in psychology, medicine, or economics, however, revealed in the recent years that only 20-40% of all findings could be replicated. This is not surprising when one considers that the current incentive structure favors long publication lists that contain new and surprising findings, which usually are based on very small samples and are statistically "just significant".
On the one hand, this led to a replication crisis: "Which result can we trust at all?". On the other hand, many stakeholders have decisively moved towards "Open Science" in the recent years, with the goal to make research more transparent, more reproducible, and more trustworthy. Also research funders and scientific organizations, such as the DFG and the EU Research Council, go into this direction.
In my talk, I will give a quick overview about the current state of the replication/credibility crisis and show, why open science must be one answer to this crisis. This has implications for research practices, teaching, hiring decisions, and journal guidelines.
- download presentation (8.15 MB)
About the speaker
Felix Schönbrodt obtained his PhD in psychology in 2010 at the Humboldt-University Berlin and received his habilitation 2014 at the Ludwig-Maximilians-University Munich. His research interests include implicit and explicit motives, quantitative methods, data visualization, and issues revolving open science and the replicability of research. One special focus is to provide statistical packages in R and interactive statistical web apps which can be used for teaching and for an enhanced understanding and usage of quantitative methods (http://www.shinyapps.org). Felix Schönbrodt is an initiator of the "Commitment to Research Transparency" (http://www.researchtransparency.org).