ABSTRACT
Effect sizes are the core elements of each meta-analysis. To perform a meta-analysis, we need at least an estimate of a study's effect size, and its standard error.
This chapter provides a general discussion of the different ways the term “effect size” can be defined, and explains the concept of a standard error. Furthermore, commonly used effect size metrics and outcome measures are covered in detail, including means, proportions, correlations, standardized mean differences, and ratios. Transformations that are sometimes needed before pooling are also described.
Several effect size correction methods are presented, as well as strategies to deal with effects sizes that are not independent. The chapter also includes information on how study data has to be prepared in order to make the pooling using R functions easier in later steps.
