Abstract:
For an enterprise tasked with discovering truth, science has a startling lack of quality assurance. In general, there are stronger safeguards to ensure that food is safe to eat than to ensure that scientific articles contain truths rather than falsehoods. While reforms in the last decade such as data sharing, preregistration, and replication have improved transparency, they still rely on traditional quality assurance mechanisms and on the assumption of researcher diligence and good faith. Other industries do not merely trust but verify. This talk argues that preserving the integrity and credibility of the scientific record requires abandoning the assumption that intentional fraud is very rare and, more importantly, confronting the threats posed by honest error, negligence, and recklessness. Suggestions for innovative, parallel systems of verification are presented, and their requirements and feasibility are discussed.
Speaker:
Ian Hussey is Senior Lecturer in Digitalisation of Psychology at the University of Bern. His research focuses on quality management in science, (forensic) meta-science, and measurement practices. He co-leadsERЯOR: A bug bounty program for science (https://error.reviews).
----------
Other talks from ReproducibiliTea Basel
Want to see what else is coming up? Check out our complete semester program
Can't wait? Dive into our past seminar recordings
