Error Detection

I learned about granularity testing from James Heathers' post on Medium after he preprinted his founding paper on granularity testing, and immediately realized its potential. I saw that they didn't have a web application for their test so I went ahead and made one for them. I then extended their test to standard deviations and test statistics, reporting on a property of variances that may have never been shown before. These tools were instrumental in uncovering the many errors in publications by Brian Wansink, which has grown into a scandal large enough that Cornell has launched not one, but two investigations into his work and there have been numerous calls for him to step down from his endowed chair (Wansink's longtime collaborator Collin Payne no longer has his faculty position, and one of Wansink's longtime postdocs is now a yoga teacher). Identifying errors in papers is a controversial area, but my collaborators and I have been receiving a lot of positive publicity recently. I am continuing to develop additional tools to identify errors, and am somewhat of a leader in this field given how small the field is. As evidence of this, I get invited to review related publications despite having no academic affiliation.
Related Publications
The GRIMMER test: A method for testing the validity of reported measures of variability PeerJ Preprints
Statistical heartburn: An attempt to digest four pizza publications from the Cornell Food and Brand Lab PeerJ Preprints
Statistical infarction: A postmortem of the Cornell Food and Brand Lab pizza publications PeerJ Preprints
Related Posts
GRIM test and scientific utopia Posted June 4th, 2016 in Blog
My big discovery Posted September 11th, 2016 in Blog
The GRIMMER test Posted September 11th, 2016 in Research
GRIMMEST: Granularity of test statistics Posted December 27th, 2016 in Research