Unless there is a reason to believe confounders skew one direction, magically, missed confounders are understood not to have a significant effect on such computed p-values. In fact, that's the point of a p-value: it only asks the question of whether the results happen at random.
-
-
Replying to @EduEngineer @K_Sheldrick
By definition a confounder skews in one direction, otherwise it does not confound the relationship and is not a confounder
3 replies 0 retweets 13 likes -
Replying to @GidMK @K_Sheldrick
A confounder skews in one direction *internally* to a single study, but those wash out at high numbers of studies/coin flips. What you need are global confounders to make the argument.
3 replies 0 retweets 32 likes -
-
Replying to @GidMK @K_Sheldrick
It's provably correct. May I invite you to walk through the math with me on camera?
1 reply 2 retweets 47 likes -
Replying to @EduEngineer @K_Sheldrick
It has very little to do with maths actually. The maths of meta-analysis is really quite simple, it's just a weighted average, if the included studies all make the same errors in terms of confounding than by definition the weighted average will also have this error
5 replies 0 retweets 15 likes -
Replying to @GidMK @K_Sheldrick
"if the included studies all make the same errors in terms of confounding than by definition the weighted average will also have this error" That assumes there is a confounder that is global in bias. Otherwise, it's random. It's a coinflip embedded in a coinflip that...
4 replies 3 retweets 34 likes -
Replying to @EduEngineer @K_Sheldrick
Nonsense. That's just a misunderstanding of confounding
2 replies 0 retweets 1 like -
No, that's not a misunderstanding. Study 1 made mistake A that skewed in direction X. Study 2 made a mistake B that skewed in direction Y. Either X and Y are the same direction (hence local skew becomes the same as global skew) or (assuming X and Y are not orthogonal) they'll
2 replies 0 retweets 1 like -
That might make sense if you had no understanding of confounding in clinical research. A very simple example is to consider the impact of age on studies looking at Covid-19 death rates
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.