worth adding that significant variables aren’t automatically good for understanding either http://jakewestfall.org/publications/ivy.pdf …https://twitter.com/russpoldrack/status/664496223395557376 …
@talyarkoni Good that someone wrote about that problem. I had also noticed it before, but never wrote anything about it.
-
-
@KirkegaardEmil epidemiologists call it “residual confounding" (see our paper for refs), but hasn’t been discussed in psychology@CookieSci -
@talyarkoni@CookieSci The implication for discriminant validity analysis is probably one of the reasons for rampant construct proliferation -
@KirkegaardEmil@CookieSci no doubt. “we need this new measure because it captures error differently from all the other existing measures." -
@talyarkoni@CookieSci That was one thing that bothered me about your paper. The subjective temperature ratings errors would correlate IRL. -
@KirkegaardEmil@talyarkoni autocorr over time? (1) plausible but not a given (2) would adding this complication really help understanding?? -
@CookieSci@talyarkoni I meant just regular cor bwtn "subjectively perceived heat levels" & "estimate the average temperature for each day". -
@KirkegaardEmil@CookieSci sure; correlated error is a huge can of worms in its own right. but you can’t do everything in one paper :)
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.