Will it take data vigilantes to restore some order in the House of Science? With the replication crisis showing no sign of letting up, some committed scientists have taken it upon themselves to find ways to sniff out cases of egregious fraud. As it turns out, identifying scientific misbehavior is surprisingly easy!
Our guest is a full-time research scientist, author/consultant at Northeastern University in Boston in a Computational Behavioral Science lab. James Heathers completed his undergraduate work in Psychology and Industrial relations from the University of Sydney and obtained his doctorate degree on the topic of methodological improvements in heart rate variability at the same institution in 2015.
He and a couple of his colleagues have captured the limelight after exposing problems in the work of a world-famous nutrition researcher, which led to the retraction of 5 papers. These “data thugs” have since designed a couple of tools that can identify suspicious data through a simple analysis of descriptive statistics.
Brian Wansink. The Grad Student Who Never Said “No” (from the WayBack Machine internet archives)
James Heathers. Introducing SPRITE and the Case of the Carthorse Child
Adam Marcus and Ivan Oransky. Meet the Data-Thugs Out to Expose Shoddy and Questionable Research (Blog post in Science, Feb 2018)
Tom Bartlett. “I want to Burn Things to the Ground”: Are the foot soldiers behind psychology’s replication crisis saving science — or destroying it? (Article in The Chronicles of Higher Education, September 2018)
Ep. 57 Neither Fisher Nor Bayes: The Limits of Statistical Inference (with Michael Acree)
WATCH ON YOUTUBE:
Watch the episode on our YouTube channel