GRIM test and scientific utopia
I recently experienced science at its absolute finest, which could only happen in the digital age outside of the constraints of academia.
It all started when I perused my emails after waking up and scrolled through my email from Medium which listed the most recent science stories. I saw The GRIM test by James Heathers and clicked just to see a little bit more of the text. I next saw that they had preprinted their paper at PeerJ Preprints, which is where I preprint my papers. I also checked PrePubMed to make sure the GRIM test got indexed correctly (it is, but someone has the last name Grim, so you have to search “grim test”). I next finally found out what GRIM stands for: Granularity-Related Inconsistency of Means. Damn, that sounds complicated, I might need some coffee for this. I also kept seeing the word “Likert”, what the heck is that?
I’m someone who likes to learn by example so I quickly scrolled down until I saw numbers and calculations. In Python the Grim test for two decimal places is round(round(mean*size,0)/size,2). Wait, that’s it? That’s the whole test? Surely there’s more. I can tell if someone made a mistake in a publication or possibly committed fraud by typing in the mean and sample size into that equation? This is a GAME CHANGER!!!
I quickly tried to figure out who these authors were. I saw that the corresponding author is on Twitter so I went to his timeline. He had tweeted about how many times the GRIM calculator had been downloaded. Wait, what calculator? Geez, I guess I need to actually read the paper now (at least I finally learned the origin of drinking someone’s Kool-Aid). It turns out that their calculator was just an excel file available at the OSF. They really didn’t have an online version of this?
So I tweeted Nicholas Brown, @sTeamTraen, and asked him if he wanted a website for his test. He replied that I was free to make a web version of the test. The next day I added the test to PrePubMed.
As it turns out Nicholas will be in Charlottesville next week for a psychology conference where he will discuss the GRIM test, and that’s where I happen to be living. I’ll meet up with him to discuss his research and if there is anything else I can help with.
This is how science should go in the 21st century. Researchers should post their results as soon as they are ready so that anyone who might benefit or contribute will see them. And someone who can significantly help the project should do so without worrying about any sort of authorship or acknowledgment. But that’s a scientific utopia, and most scientists unfortunately are stuck in the Dante’s inferno that is academia.