[R-meta] R2 meta-analysis
c@t|@@o||ve|r@ @end|ng |rom york@@c@uk
Sun Jan 15 18:57:04 CET 2023
That's a very good question. Do you know of a good way of aggregating R
squared values? I have seen some methods for example those available in the
mice package in R for pooling R squared values when one does multiple
imputation, but usually in those scenarios sample characteristics stay very
similar, would it still be appropriate for aggregating R squared values
from different studies?
Thank you again.
On Sun, Jan 15, 2023, 2:38 PM Michael Dewey <lists using dewey.myzen.co.uk> wrote:
> Dear Catia
> How would you know whether to give it a positive or a negative sign?
> On 14/01/2023 19:29, Catia Oliveira wrote:
> > Hello,
> > I hope you had a lovely start to the new year.
> > Is it possible to analyse the square root of the R2 from multiple
> > studies in the same way as one would a bivariate correlation?
> > Best wishes,
> > Catia
[[alternative HTML version deleted]]
More information about the R-sig-meta-analysis