9.1 Meta-regression in R

Meta-regressions can be conducted in R using the rma function in metafor. To show the similarity between subgroup analysis and meta-regression, consider the code for our regression-specified subgroup analysis again:

m_dummy <- rma(yi = d, vi = vi, mods = ~ Country, data = df)

This syntax used “country” as a dummy-coded variable. R will always do this for factor variables. It is possible to drop the intercept and estimate the means for all groups instead by specifying ~Country-1, and it is even possible to change the default dummy coding if you want to compare a set of groups with another set of groups: a tutorial.

Continuous variables

Imagine you want to check if the proportion of male participants is associated with effect size. The variable sex contains this information. You can use this predictor in a meta-regression:

m_reg <- rma(yi = d,
             vi = vi,
             mods = ~sex,
             data = df)
m_reg

## 
## Mixed-Effects Model (k = 56; tau^2 estimator: REML)
## 
## tau^2 (estimated amount of residual heterogeneity):     0.0544 (SE = 0.0173)
## tau (square root of estimated tau^2 value):             0.2333
## I^2 (residual heterogeneity / unaccounted variability): 66.50%
## H^2 (unaccounted variability / sampling variability):   2.98
## R^2 (amount of heterogeneity accounted for):            4.53%
## 
## Test for Residual Heterogeneity:
## QE(df = 54) = 149.5878, p-val < .0001
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 2.1607, p-val = 0.1416
## 
## Model Results:
## 
##          estimate      se    zval    pval    ci.lb   ci.ub 
## intrcpt    0.0648  0.1253  0.5168  0.6053  -0.1808  0.3104    
## sex        0.0050  0.0034  1.4699  0.1416  -0.0017  0.0116    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

As you can see from the output, sex was now included as a predictor, but it is not significantly associated with the effect size (\(p=.1416\)).

---