Resource Review: Meta-Analysis Textbook

The book Doing Meta-Analysis with R focuses on providing a guide for applied researchers to conduct meta-analyses using the programming language R. The authors in the preface explain that the intention of the book is to not only teach readers which meta-analysis techniques to apply in certain scenarios but also why the techniques should be applied. The book also has an online version which I found helpful to reference while using packages and code from the book. There is also a companion R package called {dmetar} that provides a number of extra functions to assist in the coding for meta-analysis projects.

The main content about meta-analysis starts in Chapter 3 with discussing effect sizes. I found Chapters 3 to 9 most helpful when working on multiple meta-analyses for proportions and odd ratios. The format of the chapters was helpful by providing examples of code in gray boxes throughout each of the chapters as reference code along with output from the code. The chapters of “Forest Plots in R”, “Subgroup Analysis in R” and “Meta-regression in R” provided extensive information and code for creating forest plots and conducting subgroup and regression analysis in R, topics which are not explained well, with minimal examples, in most meta-analysis books.

At the end of every chapter there is a Questions & Answers section along with a Summary containing bullet points for the main concepts explained within the chapter. These sections can be useful for instructors that are using this book as a resource for teaching meta-analysis in a course. Also, I found the chapter titled “What is Publication Bias?” to be very helpful with explaining the consequences of publication bias to collaborating non-statisticians in an informative and clear manner.

The one issue that I found was that there were a few warning issues in R when using the {meta} package mentioned in Chapter 4 about “Effect Size Pooling in R”. The warning issues came from options within the metabin function that were deprecated already. This was frustrating when completing the coding of a meta-analysis, given that the function examples in a book printed in 2022 were already not up to date with respect to options.

The last part of the book focuses on advanced methods such as multilevel meta-analysis, structural equation modeling meta-analysis, network meta-analysis, and Bayesian meta-analysis. There are also chapters on a few helpful tools such as power analysis, risk of bias plots, reporting and producibility in R, and effect size calculation and conversions. The last section of the book is informative and helpful if you are looking for a book about general methods and analysis code for running a meta-analysis. If you are wanting more advanced learning about meta-analysis with theory behind the complex concepts in the last section, there are resources and references provided in those chapters. Overall, I would recommend this book if you are interested in a teaching resource for teaching about meta-analysis techniques, or conducting and interpreting meta-analysis methods using R as your primary programming language.

Online Hands-on Guide link: https://bookdown.org/MathiasHarrer/Doing_Meta_Analysis_in_R/

Citation: Harrer, M., Cuijpers, P., Furukawa, T., & Ebert, D. (2021). Doing Meta-Analysis with R: A Hands-On Guide (1st ed.). Chapman and Hall/CRC. https://doi.org/10.1201/9781003107347