logoalt Hacker News

oofbey • today at 3:17 PM • 1 reply • view on HN

Sounds like they noticed a few extreme outliers were pulling the difference down and hacked it out. Methodologically that’s really sus to do this after the study is done. But generally I think it’s okay to exclude extreme outliers - we generally assume Gaussian distributions, but the real world happens and messes things up. Having a few kids that are 5 or 6 sigma out pulling your averages - I can believe the rules are different for them. But this should have been decided ahead of time.


Replies

rahimnathwani • today at 3:21 PM

Just to be clear, I can't say for sure that the preregistration didn't include the extreme value exclusion. I haven't been able to find a link[0], but I will email the authors to ask for it.

[0] https://osf.io/search?tab=3&search=burland