Hot take: “Association” studies are descriptive studies and should be clearly labeled as such. Hotter take: Descriptive results can be stratified but shouldn’t be “adjusted”. Hottest take: Just because a study is descriptive doesn’t mean it’s not important or useful.pic.twitter.com/gFXCjX3DBz
-
-
Replying to @EpiEllie
Agree with 1 and 3, but I'm not sure about the "hotter" take. Surely stratification is a form of adjustment anyway? I think I'd argue that adjustment within a statistical model makes sense, as long as everyone's clear about what that actually means
1 reply 0 retweets 3 likes -
Replying to @GidMK
Adding covariates to a regression model in a descriptive study is about estimating stratified distributions, I agree. Although stratification is a possible method of adjusting it’s also useful in its own right....
1 reply 0 retweets 2 likes -
We could want to know what the causal effect of height is on running speed, adjusting for gender. But we can also ask how running speed is distribtues by height among women.
2 replies 0 retweets 7 likes
Yes absolutely. I feel like it's more about the interpretation of adjustment than about adjusting per se
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.