Building Structural Equation Models in Social Science

by Alex Liu

This book is based on a collection of author's lecture notes used for teaching structural equation modeling to Ph.D. candidates in the Paul Merage School of Business, University of California at Irvine, from 2003 to 2005. During re-writing lecture notes for this book, the author made it clear to make this book a friendly and step-by-step model-building guide to anyone who plans to use structural equation modeling for high quality empirical research. In other words, this book is not written for people who just want to think about or to talk about structural equation modeling. The book is for these students and practioners who need to use structural equations to analyze some real datasets and to produce high quality research reports. To avoid some of the widely spread confusion in social science empirical research fields, a RM4Es framework has been used to summarize all the main concepts and methods of structural equation modeling. To state briefly, RM4Es consists of (e1) equations for model representations, (e2) estimation methods for coefficients calculation, (e3) evaluation methods for model assessment, and (e4) explanation for results interpretations. Feedback from author's teaching and training has continuously confirmed that the RM4Es framework is indeed a very effective tool in helping students to gain a clear understanding of structural equation modeling. Throughout the book, many real research examples have been used to demonstrate all the important structural equation modeling techniques presented in this book and to keep students updated of the most current applications. Many of the examples are taken from articles published in leading journals such as the American Economic Review and the American Political Science Review. Proper computing is always an important part of any high quality structural equation modeling project. For this book, LISREL and R are selected as the computing tools, because LISREL is still the most widely used software for structural equation modeling and R is available at free of charge. Step by step LISREL or R implementation of all the important ideas and techniques are included either in the book content or in the appendix. By using these LISREL and R programs together with a few datasets included in the book, the readers of the book should be able to test all the important structural equation modeling techniques and methods to ensure a full mastering of the modern structural equation modeling covered in this book. This book may be used as a textbook for a graduate seminar course or as a training manual for some intensive sessions. Students who completed the course or training should be able to use LISREL and R to analyze their own datasets to produce high quality structural equation models for their research.

75 pages, Kindle Edition

First published June 6, 2009