[R] meta.summaries and se's of effect sizes

Thomas Lumley tlumley at u.washington.edu
Mon Oct 18 18:01:43 CEST 2004

On Mon, 18 Oct 2004, Ingmar Visser wrote:

> Hi All,
> I would like to use meta.summaries from package rmeta to do a meta-analysis.
> I have available effect sizes as r's (which could be easily transformed to
> effect sizes in terms of d's).
> My problem is that I'm not sure what the se's of these r's should be ...

I'm not sure even what r-values and d-values are in this context.

meta.summaries takes a number of independent estimates of the same 
quantity together with standard errors and combines them to get a single 
estimate. It doesn't care what the quantity is.


More information about the R-help mailing list