Applied Meta-analysis for Social Science ResearchOffering pragmatic guidance for planning and conducting a meta-analytic review, this book is written in an engaging, nontechnical style that makes it ideal for graduate course use or self-study. The author shows how to identify questions that can be answered using meta-analysis, retrieve both published and unpublished studies, create a coding manual, use traditional and unique effect size indices, and write a meta-analytic review. An ongoing example illustrates meta-analytic techniques. In addition to the fundamentals, the book discusses more advanced topics, such as artifact correction, random- and mixed-effects models, structural equation representations, and multivariate procedures. User-friendly features include annotated equations; discussions of alternative approaches; and "Practical Matters" sections that give advice on topics not often discussed in other books, such as linking meta-analytic results with theory and the utility of meta-analysis software programs. ? |
Contents
Chapter 1 An Introduction to MetaAnalysis | 3 |
12 Basic Terminology | 4 |
13 A Brief History of MetaAnalysis | 8 |
14 The Scientific Process of Research Synthesis | 9 |
15 An Overview of the Book | 12 |
A Note on Software and Information Management | 13 |
17 Summary | 14 |
Chapter 2 Questions That Can and Questions That Cannot Be Answered through MetaAnalysis | 16 |
76 Summary | 169 |
Part III Putting the Pieces Together Combining and Comparing Effect Sizes | 173 |
Chapter 8 Basic Computations Computing Mean Effect Size and Heterogeneity around This Mean | 175 |
81 The Logic of Weighting | 176 |
82 Measures of Central Tendency in Effect Sizes | 180 |
83 Inferential Testing and Confidence Intervals of Average Effect Sizes | 182 |
84 Evaluating Heterogeneity among Effect Sizes | 184 |
Nonindependence among Effect Sizes | 191 |
21 Identifying Goals and Research Questions for MetaAnalysis | 17 |
22 The Limits of Primary Research and the Limits of MetaAnalytic Synthesis | 19 |
When Are They Valid and When Are They Not? | 23 |
The Reciprocal Relation between Planning and Conducting a MetaAnalysis | 29 |
25 Summary | 31 |
26 Recommended Readings | 32 |
Chapter 3 Searching the Literature | 34 |
32 Inclusion and Exclusion Criteria | 38 |
33 Finding Relevant Literature | 42 |
Is My Search Adequate? | 52 |
Beginning a MetaAnalytic Database | 55 |
36 Summary | 58 |
Part II The Building Blocks Coding Individual Studies | 61 |
Chapter 4 Coding Study Characteristics | 63 |
41 Identifying Interesting Moderators | 64 |
42 Coding Study Quality | 68 |
43 Evaluating Coding Decisions | 73 |
Creating an Organized Protocol for Coding | 77 |
45 Summary | 82 |
Chapter 5 Basic Effect Size Computation | 85 |
52 Computing r from Commonly Reported Results | 96 |
53 Computing g from Commonly Reported Results | 107 |
54 Computing o from Commonly Reported Results | 114 |
55 Comparisons among r g and o | 118 |
Using Effect Size Calculators and MetaAnalysis Programs | 121 |
57 Summary | 122 |
Chapter 6 Corrections to Effect Sizes | 126 |
61 The Controversy of Correction | 127 |
62 Artifact Corrections to Consider | 129 |
When and How to Correct Conceptual Methodological and Disciplinary Considerations | 142 |
64 Summary | 144 |
Chapter 7 Advanced and Unique Effect Size Computation | 147 |
Raw Difference Scores | 154 |
73 Regression Coefficients and Similar Multivariate Effect Sizes | 156 |
74 Miscellaneous Effect Sizes | 161 |
The Opportunities and Challenges of MetaAnalyzing Unique Effect Sizes | 166 |
86 Summary | 195 |
Chapter 9 Explaining Heterogeneity among Effect Sizes Moderator Analyses | 198 |
91 Categorical Moderators | 199 |
92 Continuous Moderators | 207 |
93 A General Multiple Regression Framework for Moderation | 210 |
94 An Alternative SEM Approach | 218 |
The Limits of Interpreting Moderators in MetaAnalysis | 222 |
96 Summary | 226 |
Chapter 10 Fixed Random and MixedEffects Models | 229 |
101 Differences among Models | 230 |
102 Analyses of RandomEffects Models | 234 |
103 MixedEffects Models | 239 |
104 A Structural Equation Modeling Approach to Random and MixedEffects Models | 245 |
Which Model Should I Use? | 250 |
106 Summary | 255 |
Chapter 11 Publication Bias | 257 |
112 Managing Publication Bias | 260 |
What Impact Do Sampling Biases Have on MetaAnalytic Conclusions? | 275 |
114 Summary | 276 |
Chapter 12 Multivariate MetaAnalytic Models | 279 |
121 Metaanalysis to Obtain Sufficient Statistics | 280 |
122 Two Approaches to Multivariate MetaAnalysis | 286 |
The Interplay between MetaAnalytic Models and Theory | 300 |
124 Summary | 305 |
125 Recommended Readings | 306 |
Part IV The Final Product Reporting MetaAnalytic Results | 311 |
Chapter 13 Writing MetaAnalytic Results | 313 |
132 What to Report and Where to Report It | 317 |
133 Using Tables and Figures in Reporting MetaAnalyses | 329 |
Avoiding Common Problems in Reporting Results of MetaAnalyses | 337 |
135 Summary | 342 |
References | 345 |
| 357 | |
| 363 | |
About the Author | 377 |
Other editions - View all
Common terms and phrases
22 studies analysis analytic review ANOVA approach artifact correction associations between-study categorical moderators coded study characteristics coders compute conceptual confidence intervals consider constructs contingency table correlation Crick effect size estimates effects models empirical Equation evaluate example F-ratio failsafe fixed-effects funnel plot goal groups heterogeneity included studies interest internal validity L. V. Hedges literature reviews matrix mean effect sizes measures meta meta-analysis methods moderator analyses Mplus multivariate number of studies obtain odds ratio omnibus test Ostrov peer rejection population effect potential predictors primary studies publication bias random random-effects models regression coefficient relational aggression relevant reliability represents research questions research synthesis Rosenthal sample of studies sample size sample sizes scores significance test specific standard deviation standard error standardized mean difference statistical power statistical significance structural equation modeling studies report study quality t-test tion transformed typically validity variance versus weighted