Stata 18 offers several new features in meta-analysis: multilevel meta-analysis, meta-analysis for prevalence, and, in StataNow™, meta-analysis for correlation. The new commands meta meregress and meta multilevel perform multilevel meta-analysis. meta meregress fits general multilevel meta-analysis models, which include random intercepts and random coefficients, and models assume different covariance patterns between the random effects. meta multilevel provides a convenient way to fit multilevel random-intercepts meta-analysis models, which are common in practice. Many postestimation features, such as test and margins, are also available.

By popular request, the meta suite now supports meta-analysis of one proportion, or prevalence, and of one correlation. All standard meta-analysis features are supported, including forest plots and heterogeneity, subgroups, and cumulative analyses.

Read more about these features and see them in action:

• Multilevel meta-analysis

• Meta-analysis for prevalence

• Meta-analysis for correlations StataNow

• And more

View all the new features in Stata 18.