Meta-Analytic Structural Equation Modelling (eBook)

Suzanne Jak (Autor)

eBook Download: PDF

2015 | 1st ed. 2015
VIII, 88 Seiten
Springer International Publishing (Verlag)
978-3-319-27174-3 (ISBN)

Infos

This book explains how to employ MASEM, the combination of meta-analysis (MA) and structural equation modelling (SEM). It shows how by using MASEM, a single model can be tested to explain the relationships between a set of variables in several studies. 

This book gives an introduction to MASEM, with a focus on the state of the art approach: the two stage approach of Cheung and Cheung & Chan. Both, the fixed and the random approach to MASEM are illustrated with two applications to real data. All steps that have to be taken to perform the analyses are discussed extensively. All data and syntax files are available online, so that readers can imitate all analyses. By using SEM for meta-analysis, this book shows how to benefit from all available information from all available studies, even if few or none of the studies report about all relationships that feature in the full model of interest.

Dr. Suzanne Jak is a researcher in the Faculty of Social and Behavioural Sciences at the University of Utrecht, Netherlands. She also works as a lecturer in Methods and Statistics at the department of Education of the University of Amsterdam.

Inhalt

Preface 6
Contents 8
1 Introduction to Meta-Analysis and Structural Equation Modeling 10
Abstract 10
1.1 What Is Meta-Analysis? 10
1.1.1 Issues in Meta-Analysis 11
1.1.2 Statistical Analysis 12
1.2 What Is SEM? 13
1.2.1 Path Analysis 13
1.2.2 Model Fit 16
1.2.3 Factor Analysis 18
1.3 Why Should You Combine SEM and MA? 21
References 22
2 Methods for Meta-Analytic Structural Equation Modeling 24
Abstract 24
2.1 Introduction 24
2.2 Univariate Methods 25
2.3 Multivariate Methods 26
2.3.1 The GLS Method 26
2.3.2 Two Stage Structural Equation Modeling (TSSEM) 28
References 31
3 Heterogeneity 33
Abstract 33
3.1 Introduction 33
3.2 Testing the Significance of Heterogeneity 34
3.3 The Size of the Heterogeneity 35
3.4 Random Effects Analysis or Explaining Heterogeneity 36
3.4.1 Random Effects MASEM 36
3.4.2 Subgroup Analysis 38
References 39
4 Issues in Meta-Analytic Structural Equation Modeling 40
Abstract 40
4.1 Software to Conduct MASEM 40
4.2 Fit-Indices in TSSEM 42
4.3 Missing Correlations in TSSEM 42
4.4 The ML-Approach to MASEM 43
References 44
5 Fitting a Path Model with the Two-Stage Approach 46
Abstract 46
5.1 Introduction 46
5.2 Preparing the Data 47
5.3 Fixed Effects Analysis 49
5.4 Random Effects Analysis 52
5.5 Random Effects Subgroup Analysis 60
References 63
6 Fitting a Factor Model with the Two-Stage Approach 64
Abstract 64
6.1 Introduction 64
6.2 Preparing the Data 65
6.3 Fixed Effects Analysis 66
6.4 Random Effects Analysis 68
References 76
Appendix A Model Implied Covariance Matrix of the Example Path Model 77
Appendix B Fitting a Path Model to a Covariance Matrix with OpenMx 78
Appendix C Model Implied Covariance Matrix of the Example Factor Model 86
Appendix D Fitting a Factor Model to a Covariance Matrix with OpenMx 87

Produkt-Details

• Erscheint lt. Verlag: 28.11.2015
• Reihe/Serie: SpringerBriefs in Research Synthesis and Meta-Analysis
• Reihe/Serie: SpringerBriefs in Research Synthesis and Meta-Analysis
• Zusatzinfo: VIII, 88 p. 5 illus. in color.
• Verlagsort: Cham
• Sprache: englisch
• Themenwelt: Mathematik / Informatik ► Mathematik ► Statistik
• Themenwelt: Mathematik / Informatik ► Mathematik ► Wahrscheinlichkeit / Kombinatorik
• Themenwelt: Sozialwissenschaften ► Politik / Verwaltung
• Themenwelt: Technik
• Themenwelt: Wirtschaft
• Schlagworte: MASEM • multivariate methods • random effects analysis • SEM • univariate methods
• ISBN-10: 3-319-27174-1 / 3319271741
• ISBN-13: 978-3-319-27174-3 / 9783319271743