The premise of this book is that we have long studied natural systems using a limiting paradigm, one that is based on the univariate model of statistics. The univariate model, as exemplified by ANOVA and multiple regression, is ideally designed for the study of individual responses and single processes. Understanding systems requires a different approach, one in which we can propose, evaluate, and draw our interpretations from multivariate hypotheses about the interactions among parts. Structural equation modeling (SEM) affords us with a means to do this.

We should realize that as with most methodologies, SEM is neither perfect nor complete. Innovations in the development of statistical capabilities as well as innovations in application using SEM continue to expand the utility of this approach.

Table of Contents

Part I: A Beginning

Chapter 1. Introduction

Chapter 2. An Example Model

Part II: Basic Principles of Structural Equation Modeling

Chapter 3. The Anatomy of Models I: Observed Variable Models

Chapter 4. The Anatomy of Models II: Latent Variables

Chapter 5. Principles of Estimation and Model Assessment

Part III: Advanced Topics

Chapter 6. Composite Variables and Their Uses

Chapter 7. Additional Techniques for Complex Situations

Part IV: Applications and Illustrations

Chapter 8. Model Evaluation in Practice

Chapter 9. Multivariate Experiments

Chapter 10. The Systematic Use of SEM: An Example

Chapter 11. Cautions and Recommendations

Part V: The Implications of SEM for the Study of Natural Systems

Chapter 12. How can SEM contribute to scientific advancement?

Chapter 13. Frontiers in the application of SEM

Appendix I: Example Analyses

References

Index