Latent Variable Models: An Introduction to Factor, Path, and Structural Analysis

by John C. Loehlin

This book provides an introduction to a rapidly-growing area in the social and behavioral sciences -- the modeling of systems in which one or more variables are hypothesized, but not directly observed. Providing a conceptually unified treatment of modeling of this type -- exploratory and confirmatory factor analysis, path analysis, and structural equation analysis -- it is intended to introduce these techniques to individuals who have had some exposure to statistical methods in general, but are beginners in this particular area. Using an inductive and informal approach, it emphasizes the use of path diagrams and a variety of concrete examples, and keeps the mathematics largely intuitive. Examples are drawn from a variety of fields, including psychometrics, sociology, psychology, education and behavior genetics. Although some introductory material is provided for "LISREL, EQS," and "CALIS," and for exploratory factor analysis programs in "SAS, SPSS," and "BMPD," the book is not closely tied to any one computer program or statistical package.

Genres: Research

304 pages, Paperback

First published January 1, 1987

Reviews

[Steven Peterson · Rating 4 out of 5]

This is an accessible survey of structural equation models. This is a technique that, in its fullest version, combines factor analysis with path analysis. It uses maximum likelihood estimation. This is a very potent analytical tool that can provide insights to the dynamics of models. I am a political scientist; I have used SEM to assess what factors affect political phenomena.

This volume? It is readable. Some works feature scores of equations and become difficult for those who are not mathematically well attuned. In the process of reading the book, people will get some sense of different software packages (e.g., LISREL, EQS, or AMOS--the latter of which is the program that I use). There are a small handful of books on SEM that have been particularly useful to me. And this is one of that small number of key works that I depend upon in understanding this tool.