The Resource Advances in meta-analysis, Terri D. Piggott

Advances in meta-analysis, Terri D. Piggott

Label
Advances in meta-analysis
Title
Advances in meta-analysis
Statement of responsibility
Terri D. Piggott
Creator
Subject
Language
eng
Summary
Meta-analysis is used increasingly in the social sciences to synthesize research results. As both primary research and the questions addressed by meta-analysis have been grown more complex, meta-analysis techniques have evolved to address these issues. This book covers a number of advances in meta-analysis that are not covered in detail in many introductory books on research synthesis. More specifically, this book discusses the planning of a meta-analysis for complex questions, computing power for tests in meta-analysis, handling missing data in meta-analysis, integrating individual data into a traditional meta-analysis and generalizing from the results of a meta-analysis. For each topic, a fully annotated example is provided with sample computer programs for the major statistical packages. This book assumes a familiarity with basic meta-analytic techniques. The goal of the book is to provide researchers with advanced strategies for strengthening the planning, conduct and interpretations of meta-analysis with complex data
Member of
Cataloging source
NhCcYBP
Dewey number
300.15195
Illustrations
illustrations
Index
index present
LC call number
R853.M48
LC item number
P54 2012
Literary form
non fiction
Nature of contents
bibliography
Series statement
Statistics for social and behavioral sciences
Label
Advances in meta-analysis, Terri D. Piggott
Publication
Note
DOI 10.1007/978-1-4614-2278-5
Bibliography note
Includes bibliographical references and index
http://library.link/vocab/branchCode
  • net
Contents
  • Contents note continued: 10.3.Having an a Priori Plan for the Meta-analysis -- 10.4.Carefully and Thoroughly Interpret the Results of Meta-analysis -- References -- 11.Data Appendix -- 11.1.Sirin (2005) Meta-analysis on the Association Between Measures of Socioeconomic Status and Academic Achievement -- 11.2.Hackshaw et al. (1997) Meta-analysis on Exposure to Passive Smoking and Lung Cancer -- 11.3.Eagly et al. (2003) Meta-analysis on Gender Differences in Transformational Leadership -- References
  • Contents note continued: 3.7.Interpretation of Moderator Analyses -- References -- 4.Power Analysis for the Mean Effect Size -- 4.1.Background -- 4.2.Fundamentals of Power Analysis -- 4.3.Test of the Mean Effect Size in the Fixed Effects Model -- 4.3.1.Z-Test for the Mean Effect Size in the Fixed Effects Model -- 4.3.2.The Power of the Test of the Mean Effect Size in Fixed Effects Models -- 4.3.3.Deciding on Values for Parameters to Compute Power -- 4.3.4.Example: Computing the Power of the Test of the Mean -- 4.3.5.Example: Computing the Number of Studies Needed to Detect an Important Fixed Effects Mean -- 4.3.6.Example: Computing the Detectable Fixed Effects Mean in a Meta-analysis -- 4.4.Test of the Mean Effect Size in the Random Effects Model -- 4.4.1.The Power of the Test of the Mean Effect Size in Random Effects Models -- 4.4.2.Positing a Value for x2 for Power Computations in the Random Effects Model --
  • Contents note continued: 4.4.3.Example: Estimating the Power of the Random Effects Mean -- 4.4.4.Example: Computing the Number of Studies Needed to Detect an Important Random Effect Mean -- 4.4.5.Example: Computing the Detectable Random Effects Mean in a Meta-analysis -- References -- 5.Power for the Test of Homogeneity in Fixed and Random Effects Models -- 5.1.Background -- 5.2.The Test of Homogeneity of Effect Sizes in a Fixed Effects Model -- 5.2.1.The Power of the Test of Homogeneity in a Fixed Effects Model -- 5.2.2.Choosing Values for the Parameters Needed to Compute Power of the Homogeneity Test in Fixed Effects Models -- 5.2.3.Example: Estimating the Power of the Test of Homogeneity in Fixed Effects Models -- 5.3.The Test of the Significance of the Variance Component in Random Effects Models -- 5.3.1.Power of the Test of the Significance of the Variance Component in Random Effects Models --
  • Contents note continued: 5.3.2.Choosing Values for the Parameters Needed to Compute the Variance Component in Random Effects Models -- 5.3.3.Example: Computing Power for Values of x2, the Variance Component -- References -- 6.Power Analysis for Categorical Moderator Models of Effect Size -- 6.1.Background -- 6.2.Categorical Models of Effect Size: Fixed Effects One-Way ANOVA Models -- 6.2.1.Tests in a Fixed Effects One-Way ANOVA Model -- 6.2.2.Power of the Test of Between-Group Homogeneity, QB, in Fixed Effects Models -- 6.2.3.Choosing Parameters for the Power of QB in Fixed Effects Models -- 6.2.4.Example: Power of the Test of Between-Group Homogeneity in Fixed Effects Models -- 6.2.5.Power of the Test of Within-Group Homogeneity, QW, in Fixed Effects Models -- 6.2.6.Choosing Parameters for the Test of QW in Fixed Effects Models -- 6.2.7.Example: Power of the Test of Within-Group Homogeneity in Fixed Effects Models --
  • Contents note continued: 6.3.Categorical Models of Effect Size: Random Effects One-Way ANOVA Models -- 6.3.1.Power of Test of Between-Group Homogeneity in the Random Effects Model -- 6.3.2.Choosing Parameters for the Test of Between-Group Homogeneity in Random Effects Models -- 6.3.3.Example: Power of the Test of Between-Group Homogeneity in Random Effects Models -- 6.4.Linear Models of Effect Size (Meta-regression) -- References -- 7.Missing Data in Meta-analysis: Strategies and Approaches -- 7.1.Background -- 7.2.Missing Studies in a Meta-analysis -- 7.2.1.Identification of Publication Bias -- 7.2.2.Assessing the Sensitivity of Results to Publication Bias -- 7.3.Missing Effect Sizes in a Meta-analysis -- 7.4.Missing Moderators in Effect Size Models -- 7.5.Theoretical Basis for Missing Data Methods -- 7.5.1.Multivariate Normality in Meta-analysis -- 7.5.2.Missing Data Mechanisms or Reasons for Missing Data -- 7.6.Commonly Used Methods for Missing Data in Meta-analysis --
  • Contents note continued: 7.6.1.Complete-Case Analysis -- 7.6.2.Available Case Analysis or Pairwise Deletion -- 7.6.3.Single Value Imputation with the Complete Case Mean -- 7.6.4.Single Value Imputation Using Regression Techniques -- 7.7.Model-Based Methods for Missing Data in Meta-analysis -- 7.7.1.Maximum-Likelihood Methods for Missing Data Using the EM Algorithm -- 7.7.2.Multiple Imputation for Multivariate Normal Data -- References -- 8.Including Individual Participant Data in Meta-analysis -- 8.1.Background -- 8.2.The Potential for IPD Meta-analysis -- 8.3.The Two-Stage Method for a Mix of IPD and AD -- 8.3.1.Simple Random Effects Models with Aggregated Data -- 8.3.2.Two-Stage Estimation with Both Individual Level and Aggregated Data -- 8.4.The One-Stage Method for a Mix of IPD and AD -- 8.4.1.IPD Model for the Standardized Mean Difference -- 8.4.2.IPD Model for the Correlation -- 8.4.3.Model for the One-Stage Method with Both IPD and AD --
  • Contents note continued: 8.5.Effect Size Models with Moderators Using a Mix of IPD and AD -- 8.5.1.Two-Stage Methods for Meta-regression with a Mix of IPD and AD -- 8.5.2.One-Stage Method for Meta-regression with a Mix of IPD and AD -- 8.5.3.Meta-regression for IPD Data Only -- 8.5.4.One-Stage Meta-regression with a Mix of IPD and AD -- References -- 9.Generalizations from Meta-analysis -- 9.1.Background -- 9.1.1.The Preventive Health Services (2009) Report on Breast Cancer Screening -- 9.1.2.The National Reading Panel's Meta-analysis on Learning to Read -- 9.2.Principles of Generalized Causal Inference -- 9.2.1.Surface Similarity -- 9.2.2.Ruling Out Irrelevancies -- 9.2.3.Making Discriminations -- 9.2.4.Interpolation and Extrapolation -- 9.2.5.Causal Explanation -- 9.3.Suggestions for Generalizing from a Meta-analysis -- References -- 10.Recommendations for Producing a High Quality Meta-analysis -- 10.1.Background -- 10.2.Understanding the Research Problem --
  • Machine generated contents note: 1.Introduction -- 1.1.Background -- 1.2.Planning a Systematic Review -- 1.3.Analyzing Complex Data from a Meta-analysis -- 1.4.Interpreting Results from a Meta-analysis -- 1.5.What Do Readers Need to Know to Use This Book? -- References -- 2.Review of Effect Sizes -- 2.1.Background -- 2.2.Introduction to Notation and Basic Meta-analysis -- 2.3.The Random Effects Mean and Variance -- 2.4.Common Effect Sizes Used in Examples -- 2.4.1.Standardized Mean Difference -- 2.4.2.Correlation Coefficient -- 2.4.3.Log Odds Ratio -- References -- 3.Planning a Meta-analysis in a Systematic Review -- 3.1.Background -- 3.2.Deciding on Important Moderators of Effect Size -- 3.3.Choosing Among Fixed, Random and Mixed Effects Models -- 3.4.Computing the Variance Component in Random and Mixed Models -- 3.4.1.Example -- 3.5.Confounding of Moderators in Effect Size Models -- 3.5.1.Example -- 3.6.Conducting a Meta-Regression -- 3.6.1.Example --
Control code
000048906284
Dimensions
24 cm
Extent
xiii, 155 p.
Isbn
9781461422778
Lccn
2011945854
Other physical details
ill.
http://library.link/vocab/recordID
.b27157982
System control number
  • (OCoLC)760976283
  • springer1461422779

Library Locations

    • Deakin University Library - Geelong Waurn Ponds CampusBorrow it
      75 Pigdons Road, Waurn Ponds, Victoria, 3216, AU
      -38.195656 144.304955
Processing Feedback ...