3 ms·
> How does "more samples" help control for confounders rather than just confirm a bias? I think you're correct that randomising patient assignments doesn't con
by JumpCrisscross 1y ago
> How does "more samples" help control for confounders rather than just confirm a bias?
I think you're correct that randomising patient assignments doesn't control for provider-side confounders. Curious if the study also randomised nursing assignments.