Meta Analysis - Correlation Coefficients with only M and SD

Dason

Ambassador to the humans
#21
This is the stated assumption:

"In applying this conversion assume that a continuous variable was dichotomized to
create the treatment and control groups."
Which doesn't directly state it but this seems to imply that the observations are independent. So I don't think the situation that the OP is in meets these assumptions.
 
#22
Hey everybody,

Thanks a lot for the discussion... sorry that I'm a little late getting back into the game.

From what I gleamed, I'm not totally sure that I 100% got across what we are looking for. We are looking for a single correlation coefficient which correlates VAR 1 vs VAR 2 across all the time points (with only M and SD).

(Var 1 time 1 + Var 1 time 2 + Var 1 time 3) correlated with (Var 2 time 1 + Var 2 time 2 + Var 2 time 3)

In other words, trying to somehow bring together the different time points while correlating VAR 1 and VAR 2.

I've been skimming a lot of resources, but nothing seems to come close to what we are looking form. Perhaps this is the end of line :(

Thanks again for all your help!