In a published paper would you rather have a bad DAG or no DAG?
-
-
Imo the same question could be phrased as "would you rather see the covariates they adjusted for or just have the phrase "we adjusted for common covariates""
-
I don't see how you can know what to adjust (or not adjust) for without a DAG. I'm often surprised after making a DAG about how my initial guesses are wrong.
End of conversation
New conversation -
-
-
But bad DAG has been used by authors to shield against criticism of their study ("We used a DAG"). I would much rather use the approach described by VanderWeele 2019 (Principles of confounder selection)
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Ohhhh is that what bad DAG means? I thought it was one that didn't (was bad at) represent the thinking
-
I voted no DAG due to ambiguity. One could imagine, for example, something bad enough to obfuscate the actual analysis.
- Show replies
New conversation -
-
-
I agree - the whole point of a DAG is to clarify the causal assumptions the researchers have made in their model(s).
-
But wouldn't a DAG that does this = a good DAG?
- 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.