This repo contains the materials for the "Meta-Analysis with R" workshop given on March 24, 2023, as part of the 2023 Evidence Synthesis & Meta-Analysis in R Conference. Further details about the workshop can be found here.
The lecture slides are provided here. We loosely followed the slides, but the focus in this workshop was on the application of the methods using R. We started with code_r_bcg.r, which illustrates some of the basic functionality of the metafor package (in particular, the use of the escalc() function for computing various effect size or outcome measures and the rma() function for fitting equal/random-effects model to these values and for conducting meta-regression analyses). This is also the same example as used in Viechtbauer (2010). The exercises.r script provides further examples of this.
Next, we looked at the issue of publication bias and related methods. These are illustrated in code_r_magnesium.r. The exercise script again contains further examples. Due to lack of time, we did not really get into multilevel/multivariate models, but some applications are provided in code_r_ml_mv.r. More examples thereof are again part of the exercises.
Finally, the code_r_other.r script covers various topics, including improved methods for testing / constructing confidence intervals, specialized methods for meta-analyzing 2x2 table data, model diagnostics, and Bayesian models.