when i was an escort, there was a slight (r=0.2) negative correlation (p=0.1) between the likelihood that I had an orgasm and the physical attractiveness of my client.
-
-
i like the idea of playing around with narratives explaining these things, but yeah the p-hacky stuff is what i had in mind heuristic: if you have a ton of data and notice patterns between some variables, then mayyybe you're just the kind of person who is good at data/patterns
-
This is true! I usually am much more careful around p-hacking, but I have an emotional attachment to this - it's what I recorded the data to find out in the first place and is the first correlation I checked
End of conversation
New conversation -
-
-
With 4 comparisons and a p threshold of 0.1, you would have 1-(1-0.1)^4 ~ 35% probability of finding a significant correlation if there is no effect.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.