[R-meta] integrating SMD, ZCOR and ZPCOR
li@t@ @ending from dewey@myzen@co@uk
Tue Aug 28 13:22:30 CEST 2018
If I were using different effect size types I would regard it as almost
obligatory to include a categorical moderator to differentiate them.
However in this case they are so different that I would begin to doubt
the desirability of doing one overall analysis.
On 28/08/2018 10:07, Wolf, Katrin wrote:
> Dear all,
> With the "escalc" function I calculated different effect sizes depending on the existing data (SMD, ZCOR, ZPCOR). For obtaining overall effects, can I treat the resulting yi in the same way? I mean, can I just integrate them in my random effect model by the "rma.uni" function without specifying what kind of effect size it is?
> And how to interpret resulting overall effect sizes? Can these be interpreted like Cohen's d?
> Kind regards,
More information about the R-sig-meta-analysis