[R-meta] Quick question on metaprop output

Guido Schwarzer @c @end|ng |rom |mb|@un|-|re|burg@de
Mon Mar 2 20:07:19 CET 2020

Am 19.02.20 um 11:58 schrieb Joao Afonso:
> [...]
> Previously I would get the amount of heterogeneity explain by the
> moderator with r^2. However I can't find this parameter in the output
> above. Am I missing something in my code to generate this parameter?

In a subgroup meta-analysis one typically does not report R^2.

However, you can run a meta-regression using metareg() which will call 
internally rma.uni() or rma.glmm() from metafor.


m = metaprop(nlameanimal, ssizeanimal, author, data=prevalence_2020_noout,
              method="inverse", sm="PLOGIT", method.tau="DL", 
              byvar=lcmbi, tau.common=TRUE,prediction = TRUE)


Best wishes, Guido

	[[alternative HTML version deleted]]

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