Good ideas and conversation. No ads, no tracking. Login or Take a Tour!
mk · 4827 days ago · link · · parent · post: What if academics were as dumb as quacks with statistics?
It's funny. I am at my desk and in front of me I have some data that falls exactly into what this article talks about.
To make matters worse, my detected effect between my treatment group and control is about a 7% difference. I've got a 97.9% confidence interval, but my colleague and I were just commenting on how reviewers won't be impressed with any kind of visualization of that data. He joked that the y axis should start at 0.9. Luckily this data isn't a central piece of the study, but this is the kind of stuff that goes on behind a publication.