[R-meta] multiple correlation meta-analysis as in Cheung & Chan (2002)

[Mike Cheung]

Dear Catia,

I am afraid I did not get your questions, as something was wrong here.

The study you mentioned neither used multiple correlations as the effect
sizes nor Hunter and Schmidt's approach. It used correlation matrices as
inputs to fit a customized meta-analytic structural equation modeling
(MASEM) to handle interactions.

The complete data and R code are available at https://osf.io/3w2k7/ for
reproducibility.

Best,
Mike

On Thu, Jan 12, 2023 at 10:47 PM Catia Oliveira <catia.oliveira using york.ac.uk>
wrote:

> Hello,
>
> I am interested in running a similar study to that conducted by Mike
> Cheung (The role of perceived behavioral control in predicting human
> behavior: A
> meta-analytic review of studies on the theory of planned behavior)
> where the multiple correlations estimates (square root of R2) were
> tracked to determine whether a set of predictors have explanatory
> power when one has been accounted for. This will be done across
> datasets, so I would have to pool all the multiple correlations
> estimates using the same variables (e.g. all studies with 2 predictors
> X ~ A + B and all studies with 3 predictors X ~ A + B + C). Could I
> still use the Hunter and Schmidt approach used by Mike Cheung to
> aggregate the multiple correlations estimates? If so, how can I then
> compare the size of the R2 based on the number of predictors (e.g., is
> the multiple correlations estimates for 3 predictors sig higher than
> for 2)?
>
> Thank you,
>
> Catia
>
> _______________________________________________
> R-sig-meta-analysis mailing list @ R-sig-meta-analysis using r-project.org
> To manage your subscription to this mailing list, go to:
> https://stat.ethz.ch/mailman/listinfo/r-sig-meta-analysis
>

	[[alternative HTML version deleted]]