Structural Equation Modeling: Concepts, Issues, and Applications

by Rick H. Hoyle (editor)

This largely nontechnical volume reviews some of the major issues facing researchers who wish to use structural equation modeling. Individual chapters present recent developments on specification, estimation and testing, statistical power, software comparisons and analyzing multitrait/multimethod data. Numerous examples of applications are given and attention is paid to the underlying philosophy of structural equation modeling and to writing up results from structural equation modeling analyses.

312 pages, Paperback

First published February 1, 1995

About the author

Rick H. Hoyle, Ph.D., is Professor of Psychology and Neuroscience at Duke University. He is a Fellow of the American Psychological Association (Divisions 5, Evaluation, Measurement, and Statistics, and 9, Society for the Psychological Study of Social Issues) and a Fellow and Charter Member of the Association for Psychological Science. Dr. Hoyle has served as Associate Editor of the Journal of Personality and Social Psychology, Journal of Personality, and Self and Identity, and Editor of Journal of Social Issues. Among his book projects are, Selfhood: Identity, Esteem, Regulation (coauthored with Michael Kernis, Mark Leary, and Mark Baldwin) and the Handbook of Individual Differences in Social Behavior (co-edited with Mark Leary).

Reviews

[Steven Peterson · Rating 4 out of 5]

A fine work on structural equation modeling (SEM). This is a technique that allows one to develop path models coupled with confirmatory factor analysis (in its full and most useful form) to predict phenomena. This book has some nice essays in it, and I have used this as one tool by which to master SEM.

Now, there are a number of software packages that allow one to use this technique. My personal choice? AMOS (Analysis of Moment Structures for those who care!). I have analyzed a number of data sets with this program, and it provides satisfactory analysis.

Some advantages of SEM? There are a variety of "measures of fit," showing how well one's model describes the data. This is a substantial advantage over other prediction techniques, like regression, which have some fit statistics--but nothing like SEM.

Among the most useful chapters in this collection of essay: Hoyle's Introduction to SEM; Chou and Bentler on tests in SEM; Hu and Bentler's excellent essay on model fit and its evaluation (one of my most cited references when I use SEM in research); and so on.

There are several works that do a nice job outlining SEM. This volume was published in 1995, so it is a decade and a half old. Still, it is a solid work.