Also, not controlling for a covariate on which groups of interest differ makes a statistical analyses stronger? 
-
-
I don't think that's the main point of the article, just the third word ;-). I was referring mostly to the 'anomalous results' section
-
Sean: if you are interested in 1) ADHD ~ brain volume when 2) IQ ~ brain volume 3) ADHD ~ IQ you better add IQ in your model.
-
Sure, present that too, and interpret accordingly. A lot of people interested in dev psych think IQ should always be controlled for. I don't
-
May be subtle, but 'controllling for' has a sligthly diff meaning than "including in the model" - in the latter u r interested in its effect
-
Yeah that's true. I'm not the anti IQ in the model guy. Check out the preprint I just posted. Just think the article I shared is interesting
-
with "control for" I mean add it to model so you can 1) interpret coefficient & 2) it's partialled out of the other predictors (e.g. ADHD)
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.