This bestselling text provides a balance between the technical and practical aspects of structural equation modeling (SEM). Using clear and accessible language, Rex B. Kline covers core techniques, potential pitfalls, and applications across the behavioral and social sciences. Some more advanced topics are also covered, including estimation of interactive effects of latent variables and multilevel SEM. The companion Web page offers downloadable syntax, data, and output files for each detailed example for EQS, LISREL, and Mplus, allowing readers to view the results of the same analysis generated by three different computer tools. Critical Acclaim: "Kline provides a text that is accessible for graduate students, practitioners, and researchers who are not intimately familiar with SEM techniques. In addition, he effortlessly summarizes current information that researchers who already use SEM should have….A major strength of the book is the individual chapter examples with explanation of the values provided from a variety of statistical analysis packages."
—James B. Schreiber, Center for Advancing the Study of Teaching and Learning, Duquesne University "The coverage is excellent and the writing style is friendly and direct, with a subtle humor that I find refreshing. I especially like the new topic boxes in the third edition, most of which discuss issues that I have had to address separately in lectures."
—Jacob Marszalek, Division of Counseling and Educational Psychology, University of Missouri–Kansas City "This is now the #1 book I will recommend to students and substantive researchers (who are not quantitative specialists) who want to learn SEM! Compared to most SEM books that I have seen, this one strikes a better balance between accessibility and breadth. In the third edition, Kline not only has updated the material, but has substantially improved it. "—Noel A. Card, Division of Family Studies and Human Development, University of Arizona "In the third edition, Kline has improved the pedagogical value of his book relative to prior editions and to other SEM books. The Web page featuring complete computer syntax and data for the examples is very helpful. Other new material further supports a reader’s understanding of SEM."—Craig Wells, School of Education, University of Massachusetts–Amherst "Of all the introductory SEM texts, this one is the most interesting to read. Anyone who has taken a course in basic algebra or introductory statistics will be able to understand the ideas and work through the exercises, and those who work their way through the book will have a good foundation in SEM and will be able to use it effectively."—David F. Gillespie, George Warren Brown School of Social Work, Washington University in St. Louis I. Concepts and Tools
1. Introduction
The Book's Website
Pedagogical Approach
Getting Ready to Learn about SEM
Characteristics of SEM
Widespread Enthusiasm, but with a Cautionary Tale
Family History and a Reminder about Context
Extended Latent Variable Families
Plan of the Book
Summary
2. Fundamental Concepts
Multiple Regression
Partial Correlation and Part Correlation
Other Bivariate Correlations
Logistic Regression
Statistical Tests
TOPIC BOX 2.1. The "Big Five" Misinterpretations of Statistical Significance
Bootstrapping
Summary
Recommended Readings
Exercises
3. Data Preparation
Forms of Input Data
Positive Definiteness
TOPIC BOX 3.1. Causes of Nonpositive Definiteness and Solutions
Data Screening
Selecting Good Measures and Reporting about Them
Summary
Recommended Readings
Exercises
4. Computer Tools
Ease of Use, Not Suspension of Judgment
Human-Computer Interaction
TOPIC BOX 4.1. Graphical Isn't Always Better
Core SEM Programs and Book Website Resources
Other Computer Tools
Summary
Recommended Readings
II. Core Techniques
5. Specification
Steps of SEM
Model Diagram Symbols
Specification Concepts
Path Analysis Models
CFA Models
Structural Regression Models
Exploratory SEM
Summary
Recommended Readings
Exercises
6. Identification
General Requirements
Unique Estimates
Rule for Recursive Structural Models
Rules for Nonrecursive Structural Models
Rules for Standard CFA Models
Rules for Nonstandard CFA Models
Rules for SR Models
A Healthy Perspective on Identification
Empirical Underidentification
Managing Identification Problems
Summary
Recommended Readings
Exercises
APPENDIX 6.A. Evaluation of the Rank Condition
7. Estimation
Maximum Likelihood Estimation
TOPIC BOX 7.1. Two-Stage Least Squares Estimation
Detailed Example
Brief Example with a Start Value Problem
Fitting Models to Correlation Matrices
Alternative Estimators
A Healthy Perspective on Estimation
Summary
Recommended Readings
Exercises
APPENDIX 7.A. Start Value Suggestions for Structural Models
APPENDIX 7.B. Effect Decomposition in Nonrecursive Models and the Equilibrium Assumption
8. Hypothesis Testing
Eyes on the Prize
State of Practice, State of Mind
A Healthy Perspective on Fit Statistics
Types of Fit Statistics and "Golden Rules"
Model Chi-Square
Approximate Fit Indexes
Visual Summaries of Fit
Recommended Approach to Model Fit Evaluation
Detailed Example
Testing Hierarchical Models
Comparing Nonhierarchical Models
Power Analysis
Equivalent and Near-Equivalent Models
Summary
Recommended Readings
Exercises
9. Measurement Models and Confirmatory Factor Analysis
Naming and Reification Fallacies
Estimation of CFA Models
Detailed Example
Respecification of Measurement Models
Special Topics and Tests
TOPIC BOX 9.1. Reliability of Construct Measurement
Items as Indicators and Other Methods for Analyzing Items
Estimated Factor Scores
Equivalent CFA Models
Hierarchical CFA Models
Models for Multitrait–Multimethod Data
Measurement Invariance and Multiple-Sample CFA
Summary
Recommended Readings
Exercises
APPENDIX 9.A. Start Value Suggestions for Measurement Models
APPENDIX 9.B. Constraint Interaction in Measurement Models
10. Structural Regression Models
Analyzing SR Models
Estimation of SR Models
Detailed Example
Equivalent SR Models
Single Indicators in Partially Latent SR Models
Cause Indicators and Formative Measurement
TOPIC BOX 10.1. Partial Least Squares Path Modeling
Invariance Testing of SR Models
Reporting Results of SEM Analyses
Summary
Recommended Readings
Exercises
APPENDIX 10.A. Constraint Interaction in SR Models
III. Advanced Techniques, Avoiding Mistakes
11. Mean Structures and Latent Growth Models
Logic of Mean Structures
Identification of Mean Structures
Estimation of Mean Structures
Latent Growth Models
Structured Means in Measurement Models
MIMIC Models as an Alternative to Multiple-Sample Analysis
Summary
Recommended Readings
12. Interaction Effects and Multilevel SEM
Interaction Effects of Observed Variables
Interaction Effects in Path Models
Mediation and Moderation Together
Interactive Effects of Latent Variables
Estimation with the Kenny-Judd Method
Alternative Estimation Methods
Rationale of Multilevel Analysis
Basic Multilevel Techniques
Convergence of SEM and MLM
Multilevel SEM
Summary
Recommended Readings
13. How to Fool Yourself with SEM
Tripping at the Starting Line: Specification
Improper Care and Feeding: Data
Checking Critical Judgment at the Door: Analysis and Respecification
The Garden Path: Interpretation
Summary
Recommended Readings
•Suggested Answers to Exercises About the Author: Rex B. Kline, Department of Psychology, Concordia University, Montreal, Quebec, Canada |
