@ImHardcory @EPoe187 Can you run some more participants on this part of the study? This p value is way too suspicious. Just add another 300 and update results.pic.twitter.com/d3lTlR7ekm
-
So the effect is reliable, just that one study is a bit suspect with the p? If so, I would say if the pattern of results is reliable then that argument is quite fine and in some ways more telling. Admittedly I am blindly contributing to the discussion lol
One of the betas in the model below has p = .034 as well. I recommend getting all main results to p < .01 at least, and preferably p < .005 per new NHST standard paper. Here's some easy attack points I would go for as critic.pic.twitter.com/XE2x7z4EhD
-
-
That’s not how one should run studies. We preregistered and report all data. Just because critics will attack doesn’t mean we should use sloppy methods. The important thing is the overall pattern of findings, not a p here or there
-
I don't agree. Your study will only have impact if it won't be attacked easily. Even if you pre-reg and don't cheat, findings with suspicious p values will always be seen as suspicious and providing only greyzone tier evidence. But it's easy to run more Ss so why not do it?
-
Well, I am happy to if they ask : ) At this point, it has already been submitted.
End of conversation
New conversation -
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.