Skip to content
Prev 4 / 5632 Next

[R-meta] Multivariate meta-analysis with metafor: Should I adjust sample sizes/variances for multiple groups ('double counting') when combined with multiple endpoints?

Dear James,


thank you so much for your quick reply!

In fact, the distinction between treatment conditions is of interest, 
since my moderators/covariates differ between the different conditions 
within the studies that include multiple groups.

So, I have used the formula(s) from your blogpost to modify the V matrix 
and will run my analysis again using this matrix (with the original 
sample sizes from the control groups).


It seems so simple and logical. However, with my limited math knowledge 
I wouldn't have been able to derive the third formula for myself.

Thanks a lot!


Best,

Emily


.


Am 16.06.2017 um 16:03 schrieb James Pustejovsky: