[R-meta] R2 meta-analysis
||@t@ @end|ng |rom dewey@myzen@co@uk
Mon Jan 16 15:13:19 CET 2023
I think my previous reply implies that I do not think meta-analysis of
R^2 is likely, in general, to be informative.
On 15/01/2023 17:57, Catia Oliveira wrote:
> 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.
> Best wishes,
> On Sun, Jan 15, 2023, 2:38 PM Michael Dewey <lists using dewey.myzen.co.uk
> <mailto: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
> <http://www.avg.com/email-signature?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=emailclient> Virus-free.www.avg.com <http://www.avg.com/email-signature?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=emailclient>
More information about the R-sig-meta-analysis