In meta-analysis, it is important to not only focus on the pooled effect, but also on the heterogeneity in our data. The pooled effect size alone does not capture that some studies' effects may differ substantially from the point estimate.

This chapter provides an introduction into the concept of heterogeneity in meta-analysis, and how it is measured. Using a hands-on simulation example, the behavior of Cochran's Q is explained. Limitations of Q and the Q-test are also discussed, as well as alternative ways to quantify between-study heterogeneity. This includes I 2, H 2, τ https://s3-euw1-ap-pe-df-pch-content-public-p.s3.eu-west-1.amazonaws.com/9781003107347/de07b140-3dcb-4e86-8547-42a1b8d32683/content/math0_2.tif"/> and prediction intervals.

The second part of the chapter deals with outliers and influential cases. It provides several hands-on R examples covering methods to identify such studies.