Applied Meta-Analysis for Social Science Research

Noel A. Card PhD, is Associate Professor in Educational Psychology at the University of Connecticut. His areas of interest include child and adolescent social development and quantitative research methods. He has received the Society for Research in Child Development Early Career Research Contributions Award and is an elected member of the Society of Multivariate Experimental Psychology.

Card is to be applauded for his thorough discussion of both the fundamentals and recent advances in meta-analysis, and for his use of such friendly, toned-down language. For instance, the graphical presentation of simulation results in order to explain the threat/impact of publication bias will really help readers understand the concept. I really like the author’s discussions of practical matters, which may stimulate readers to investigate new approaches and practices. I will recommend this book to my colleagues in psychology and education who are interested in learning meta-analysis.--Soyeon Ahn, PhD, Research, Measurement, and Evaluation Program, University of MiamiThis book teaches individuals how to do a meta-analysis from start to finish. Readers learn how to search the literature, code studies, statistically combine study results, and write up the results. Card covers topics not included in most textbooks, such as how to retrieve unpublished studies, the creation of a coding manual, effect sizes from multiple regression analysis, publication bias, and multivariate procedures in meta-analysis. I like the 'Practical Matters' sections in the chapters. This is an excellent textbook for a course on meta-analysis, and an excellent manual for anyone wanting to conduct a meta-analysis.--Brad J. Bushman, PhD, Institute for Social Research, University of MichiganThe book is well organized and walks the reader through the concepts in an accessible, logical manner. The definitions and explanations are clear and easy to follow, and the consistent use of examples throughout the chapters is very helpful. The author does a nice job of presenting key considerations, alternative approaches, and the strengths and shortcomings of each option. The reading level and writing style are appropriate for graduate student readers.--Jody A. Worley, PhD, Department of Human Relations, University of Oklahoma-Tulsa- Card's book is written in a nontechnical style that would work well in a graduate course, or it could be used by someone who is just learning meta-analysis. Card takes the reader through the multiple steps in creating a meta-analytic study, such as retrieving studies, creating a coding manual, and using effect size indices, as well as how to write a meta-analytic review paper. His book is peppered with examples of the technique, which makes the topic more accessible and easier to understand. He also provides advice on the utility of meta-analysis software, such as how it can be used and whether it is needed. This book is written like a good recipe that the reader can follow from beginning to end to produce a meta-analytic study ready to submit for publication....Describes how to do meta-analysis without bogging down the reader with equations. Card offers simple explanations of what to do in meta-analysis and provides enough background so that the reader can understand the rationale behind procedures and why the procedures need to be completed. --PsycCRITIQUES, 7/27/2012

• I. The Blueprint: Planning and Preparing a Meta-Analytic Review1. An Introduction to Meta-Analysis1.1 The Need for Research Synthesis in the Social Sciences1.2 Basic Terminology1.3 A Brief History of Meta-Analysis1.4 The Scientific Process of Research Synthesis1.5 An Overview of the Book1.6 Practical Matters: A Note on Software and Information Management1.7 Summary1.8 Recommended Readings2. Questions That Can and Questions That Cannot Be Answered through Meta-Analysis2.1 Identifying Goals and Research Questions for Meta-Analysis2.2 The Limits of Primary Research and the Limits of Meta-Analytic Synthesis2.3 Critiques of Meta-Analysis: When Are They Valid and When Are They Not?2.4 Practical Matters: The Reciprocal Relation between Planning and Conducting a Meta-Analysis2.5 Summary2.6 Recommended Readings3. Searching the Literature3.1 Developing and Articulating a Sampling Frame3.2 Inclusion and Exclusion Criteria3.3 Finding Relevant Literature3.4 Reality Checking: Is My Search Adequate?3.5 Practical Matters: Beginning a Meta-Analytic Database3.6 Summary3.7 Recommended ReadingsII. The Building Blocks: Coding Individual Studies4. Coding Study Characteristics4.1 Identifying Interesting Moderators4.2 Coding Study “Quality”4.3 Evaluating Coding Decisions4.4 Practical Matters: Creating an Organized Protocol for Coding4.5 Summary4.6 Recommended Readings5. Basic Effect Size Computation5.1 The Common Metrics: Correlation, Standardized Mean Difference, and Odds Ratio5.2 Computing r from Commonly Reported Results5.3 Computing g from Commonly Reported Results5.4 Computing o from Commonly Reported Results5.5 Comparisons among r, g, and o5.6 Practical Matters: Using Effect Size Calculators and Meta-Analysis Programs5.7 Summary5.8 Recommended Readings6. Corrections to Effect Sizes6.1 The Controversy of Correction6.2 Artifact Corrections to Consider6.3 Practical Matters: When (and How) to Correct: Conceptual, Methodological, and Disciplinary Considerations6.4 Summary6.5 Recommended Readings7. Advanced and Unique Effect Size Computation7.1 Describing Single Variables7.2 When the Metric Is Meaningful: Raw Difference Scores7.3 Regression Coefficients and Similar Multivariate Effect Sizes7.4 Miscellaneous Effect Sizes7.5 Practical Matters: The Opportunities and Challenges of Meta-Analyzing Unique Effect Sizes7.6 Summary7.7 Recommended ReadingsIII. Putting the Pieces Together: Combining and Comparing Effect Sizes8. Basic Computations: Computing Mean Effect Size and Heterogeneity around This Mean8.1 The Logic of Weighting8.2 Measures of Central Tendency in Effect Sizes8.3 Inferential Testing and Confidence Intervals of Average Effect Sizes8.4 Evaluating Heterogeneity among Effect Sizes8.5 Practical Matters: Nonindependence among Effect Sizes8.6 Summary8.7 Recommended Readings9. Explaining Heterogeneity among Effect Sizes: Moderator Analyses9.1 Categorical Moderators9.2 Continuous Moderators9.3 A General Multiple Regression Framework for Moderation9.4 An Alternative SEM Approach9.5 Practical Matters: The Limits of Interpreting Moderators in Meta-Analysis9.6 Summary9.7 Recommended Readings10. Fixed-, Random-, and Mixed-Effects Models10.1 Differences among Models10.2 Analyses of Random-Effects Models10.3 Mixed-Effects Models10.4 A Structural Equation Modeling Approach to Random- and Mixed-Effects Models10.5 Practical Matters: Which Model Should I Use?10.6 Summary10.7 Recommended Readings11. Publication Bias11.1 The Problem of Publication Bias11.2 Managing Publication Bias11.3 Practical Matters: What Impact Do Sampling Biases Have on Meta-Analytic Conclusions?11.4 Summary11.5 Recommended Readings12. Multivariate Meta-Analytic Models12.1 Meta-Analysis to Obtain Sufficient Statistics12.2 Two Approaches to Multivariate Meta-Analysis12.3 Practical Matters: The Interplay between Meta-Analytic Models and Theory12.4 Summary12.5 Recommended ReadingsIV. The Final Product: Reporting Meta-Analytic Results13. Writing Meta-Analytic Results13.1 Dimensions of Literature Reviews, Revisited13.2 What to Report and Where to Report It13.3 Using Figures and Tables in Reporting Meta-Analyses13.4 Practical Matters: Avoiding Common Problems in Reporting Results of Meta-Analyses13.5 Summary13.6 Recommended ReadingsReferencesAuthor IndexSubject IndexAbout the Author