[R-meta] R package to use slopes as effect size in a Meta Analysis

[Viechtbauer, Wolfgang (SP)]

Dear Norman,

metafor (not metaphor, despite what your spell checker wants it to be) can do this. Just supply the slopes via argument 'yi' and the corresponding sampling variances via argument 'vi' (or if you have the standard errors, via argument 'sei').

For this to make sense, the outcome variable that was used in the regression models must have been the exact same variable across the studies (with the same units). The same goes for the predictor variable corresponding to the slope you are interested in. Then there is also the issue of other predictors / covariates in the models. Strictly speaking, those should also be exactly the same across studies, since the regression slope of interest is 'adjusted' for those other predictors, so if they differ, then the regression slope reflects a different parameter across studies.

Best,
Wolfgang

>-----Original Message-----
>From: R-sig-meta-analysis [mailto:r-sig-meta-analysis-bounces using r-project.org]
>On Behalf Of Norman DAURELLE
>Sent: Friday, 22 May, 2020 9:34
>To: r-sig-meta-analysis using r-project.org
>Subject: [R-meta] R package to use slopes as effect size in a Meta Analysis
>
>Hello,
>
>I have been searching for an R package, possibly similar to the package
>metaphor,
>that would allow me to use regression slopes between two variables, and not
>correlations,
>as the effect sizes of interest.
>
>May you please tell me if you know of a way to do this using R ?Thank
>you,Norman.