And btw this is for a very high impact journal!
-
-
-
And they are running between subject correlations with an N of 24...
Show this thread
End of conversation
New conversation -
-
-
Interpreting fourway interactions is senseless if you haven't also interpreted the (4) threeways, (6) twoways and (4) oneways. That's 15 p-values to interpret. You shouldn't need an authoritative source to see that n=24 is insufficient here. Common sense will do. 1/2
-
And they have all those interactions in 7 different ANOVAs. But I thought common sense would have suggested a) the authors not do it and b) the journal not pass it on to reviewers.
End of conversation
New conversation -
-
-
Thanks that's a great reference. Beyond reviewing, my lab runs up against the challenge of how to properly power within-subject designs all the time.
- 1 more reply
New conversation -
-
Some skepticism probably appropriate, but IMO this isn't obviously a problem. Thoughts: - In 2^k factorials, testing interactions statistically equiv. to testing simple effects: both compare half of obs to other half. No power penalty - Big fx sizes less implausible in RM designs
-
Yes - it is the multiplicity that is more of a problem - rarely do people predict 4-way interactions (though I think I had at least one in my thesis). Also designed experiments have greater power to detect interactions because the factors are usually extremes (+++ vs.---)
End of conversation
New conversation -
-
-
Don’t forget that this will increase the experimentwise error rate
it’s an often overlooked issue in multifactorial anova. I hope the link helps!https://link.springer.com/article/10.3758/s13423-015-0913-5 …Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
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.