Applied Structural Equation Modelling for Researchers and Practitioners: Using R and Stata for Behavioural Research

✍ Scribed by Indranarain Ramlall

• Publisher: Emerald Group
• Year: 2017
• Tongue: English
• Leaves: 152
• Category: Library

✦ Synopsis

During the last two decades, structural equation modelling (SEM) has emerged as a powerful multivariate data analysis tool in social science research settings, especially in the fields of sociology, psychology, and education. Social science researchers and students benefit greatly from acquiring knowledge and skills in SEM, since the methods can provide a bridge between the theoretical and empirical aspects of behavioural research. Ramlall explains in a rigorous, concise, and practical manner all the vital components embedded in structural equation modelling (SEM). Focusing on R and Stata to implement and perform various structural equation models, Ramlall examines the types, benefits, and drawbacks of SEM, delving into model specifications and identifications, fit evaluations, and path diagrams.

✦ Table of Contents

Front Cover
Applied Structural Equation Modelling for Researchers and Practitioners
Copyright Page
Dedication
Contents
Preface
1 Definition of SEM
1.1. Introduction
1.2. Regression and SEM
1.3. Data Pre-Processing in SEM
1.4. Confirmatory Factor Analysis versus Exploratory Factor Analysis
1.5. Measurement and Structural Models
1.6. Software for SEM
1.7. Non-Experimental Data (Absence of Treatments and Controls)
1.8. Theoretical Construct
1.9. Conclusion
1.10. Key Elements to Retain Whenever Working with SEM
2 Types of SEM
2.1. Multilevel SEM
2.2. Non-Linear SEM
2.3. Bayesian SEM
2.4. Non-Parametric SEM
3 Benefits of SEM
4 Drawbacks of SEM
5 Steps in Structural Equation Modelling
5.1. Specification
5.2. Identification
5.3. Estimation (Focus Implied Covariance Matrix)
5.4. Testing Fit (Fitting Sample Covariance Matrix and Implied Covariance Matrix)
5.5. Re-Specification
6 Model Specification: Path Diagram in SEM
6.1. Introduction
6.2. Benefits of Path Analysis
6.3. Drawback of Path Analysis
6.4. Solution to Latent Construct
6.5. Concepts under Path Analysis
6.6. Assumptions under Path Diagram
6.7. From Path Analysis to Equation Analysis: Deriving Covariances from Path Analysis
6.8. Simple and Compound Paths
6.9. SEM Analysis Based on Correlation Coefficients and Visual Display
6.10. Minimizing Residuals between Observed and Estimated Covariances under SEM
6.11. Saturated versus Independence models
6.12. Sample Size under SEM
7 Model Identification
7.1. Introduction
7.2. Types of Parameters in SEM
7.3. Types of Identification
7.4. Examples of Model Identification Explained
7.4.1. Under-Identified Model
7.4.2. Overidentified Model (The One to Be Used)
7.5. Model Identification: Both Measurement and Structural Equations
8 Model Estimation
8.1. Introduction
9 Model Fit Evaluation
9.1. Introduction
9.2. Types of Fit Evaluation
9.3. Absolute Fit
9.4. Relative (Comparative Fit)
9.5. Incremental Fit
9.6. Model Comparison and Model Parsimony
(a) Model Comparison
(i) Tucker-Lewis Index (TLI)
(ii) Normed Fit Index and Comparative Fit Index
(b) Model Parsimony
(1) Parsimony Normed Fit Index
(2) Akaike Information Criterion (AIC)
9.7. Nested Models
9.8. Types of Model-Fit Criterion
9.9. Parameter Fit
Recommended Steps for Parameter Estimates
10 Model Modification
10.1. Introduction
11 Model Cross-Validation
12 Parameter Testing
13 Reduced-Form Version of SEM
14 Multiple Indicators Multiple Causes Model of SEM
15 Practical Issues to Consider when Implementing SEM
15.1. Introduction
(a) Specify the SEM Model Based on the Theoretical Foundation
(b) Model Identification Based on the Specified Model
(c) Data Pre-Processing: Sample Size, Outliers, Normality and Missing Data
(d) Data Normalization
(e) Variance-Covariance Matrix: Ensure Positive Definiteness of Covariance Matrix
(f) Model Fit Evaluation-Based on the Structural Equation
Relationship Values under Measurement Model
(g) Parameter Estimation
(i) Sound Model Fit Is Ensured
(ii) Sound Statistical Significance of the Parameters
(iii) Sound Magnitude and Direction of the Parameter Estimates
(h) Mediation Analysis
(i) Cross-Validation
(j) Model Modification
16 Review Questions
17 Enlightening Questions on SEM
17.1. Questions and Answers on SEM
18 Applied Structural Equation Modelling Using R
19 Applied Structural Equation Modelling using STATA
19.1 Steps in Structural Equation Modelling Using STATA
19.1.1 Data Preprocessing to Remove Outliers
19.1.2 Model Building
19.1.3 Model Identification
19.1.4 Estimation
19.1.5 Post-Estimation Analysis
19.1.5.1 Modified Indices
19.1.5.2 Goodness-of-Fit Statistics
19.1.5.3 Decompose Total Effects into Direct Effects and Indirect Effects
19.1.5.4 Model Fit
19.1.5.5 Check Stability Index of the Model
Appendix
A.1 Matrix Operations
A.2 Determinant of a Matrix
A.3 Matrix of Minors
A.4 Matrix of Cofactors
A.5 Transpose of a Matrix
A.6 Inverse of a Matrix
A.7 Matrix Formulation of SEM
Bibliography
About the Author