[R-meta] correlational meta-analysis

Viechtbauer, Wolfgang (SP) wo||g@ng@v|echtb@uer @end|ng |rom m@@@tr|chtun|ver@|ty@n|
Tue Jul 27 08:35:51 CEST 2021

Dear Cátia,

As the name implies (variance-covariance matrix), it contains the variances and *covariances* between the correlations. Of course, one can convert this var-cov matrix into a correlation matrix, but for the analysis, we need the covariances. And if by 'inferior' you mean 'smaller', then the answer is no (i.e., the correlation between two correlations r1 and r2 could very well be stronger than r1 and r2 itself).


>-----Original Message-----
>From: R-sig-meta-analysis [mailto:r-sig-meta-analysis-bounces using r-project.org] On
>Behalf Of Cátia Ferreira De Oliveira
>Sent: Tuesday, 27 July, 2021 5:16
>To: R meta
>Subject: [R-meta] correlational meta-analysis
>Dear all,
>I hope you are well.
>When conducting a correlational meta-analysis with multiple effect sizes
>per study, would the variance-covariance matrix represent the correlations
>between those effect sizes, i.e. the correlation of the correlation
>coefficients? If so, is it fair to assume that the correlation of these
>effect sizes has to be necessarily inferior to the original correlations?
>Thank you!
>Best wishes,

More information about the R-sig-meta-analysis mailing list