Common Mistakes in Meta-Analysis and How to Avoid Them
This book provides a discussion of common mistakes in the following areas:
Choosing a statistical model
Discussing heterogeneity
Interpreting significance tests
Assessing publication bias
Comparing subgroups
The book also provides worked examples and templates for reporting the results correctly.
Reviews of Common Mistakes in Meta-Analysis
Michael Borenstein is the master of accessible and accurate explanations. Thank you, Michael, for drawing on your vast experience mentoring thousands of people around the globe, to produce this book for us. Its value is immense. It is my hope that all scholars undertaking research synthesis will have this book by their side.

– HARRIS COOPER, Chief Editorial Advisor, American Psychological Association (2009−2015); Editor, Psychological Bulletin (2003−2009)
This book should be required reading for anyone conducting or planning to conduct a meta-analysis. By highlighting some of the most common mistakes that occur in meta-analyses, readers will learn how to avoid perpetuating those mistakes and ensure their meta-analyses are conducted and interpreted appropriately. This book should also be of great interest to researchers, policymakers, and practitioners who wish to be informed consumers of meta-analyses.

– EMILY E. TANNER-SMITH, Associate Dean for Research, College of Education, University of Oregon
This book will be tremendously useful to both first time and experienced users of meta-analysis. By alerting us to common pitfalls, it has the potential to improve the use of evidence in science and policy making.

– LARRY HEDGES, Board of Trustees Professor of Statistics and Policy Research, Institute for Policy Research, Northwestern University
How you can use this book:
Among the thousands of meta-analyses that have been published over the past several decades, there are a number of mistakes that appear on a fairly regular basis. This book outlines the most common mistakes, using examples in medicine, epidemiology, education, psychology, criminal justice, and other fields. For each, it explains why it is a mistake, the implications of the mistake, and how to avoid the mistake.

The book is intended primarily for researchers, and so the discussion is conceptual rather than statistical. The examples show the real-world consequences of the mistakes, explaining (for example) how the mistakes can lead to the adoption of interventions that may actually be harmful in some populations.

The book includes a section with examples that show how to report the results of an analysis correctly. These examples can serve as templates for reporting an analysis, while avoiding the mistakes discussed in earlier chapters.

The book’s author is the co-author of the text Introduction to Meta-Analysis, the best-selling text in this field. In the current volume he draws on his experience teaching meta-analysis to thousands of researchers as well as his experience as a reviewer of meta-analyses for numerous journals.
Common mistakes addressed in this volume include the following:
Many researchers believe that the I 2 statistic tells us how much the effect size varies.

In fact, an I 2 value of 10% could correspond to substantial heterogeneity while an I 2 value of 90% could correspond to trivial heterogeneity.
Many readers assume that if the effect is statistically significant, the treatment works in all populations.

In fact, the treatment could be helpful in some populations and harmful in others.
Many meta-analysts use a significance test to choose between the fixed-effect and random-effects models.

In fact, the selection of a model must be based on the goals of the analysis.
Click here to order your copy now: arrow Order Now
Meta-Analysis: Learn More
Join our Mailing List
Please enter a valid email address.
Please select a valid option.