"adjustments". Or magic, as more rigorous epistemologists would admit. Mate, before you bloat yourself up like a smug pufferfish, note that most of us ignore ALL crappy epidemiology, whether it confirms our biases or not.
No, that's the point of correcting for confounders! To put it another way, correcting for baseline results reduces the likelihood that people's characteristics at the start of the study caused the results
-
-
That's for confounders that aren't proxies for mediators. Do you think I've misinterpreted
@yudapearl 's meaning, or that he's wrong, or that this isn't an instance of it?Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
They corrected for baseline & 3rd visit and forecasted the rest from there, excluding all data from people who died, got metabolic disease during the data collection. That's cheating & not adjusting for confounders
@GidMK -
I could have misunderstood, but I read it as they didn't change the quartiles for those diagnosed (also questionable), not that they excluded them.
- Show replies
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.