This book is a valuable adjunct to the extant literature on specification, estimation, and identification. My overall impression is that this volume is indispensable for those wishing to keep current with this fast-moving field. I recommend that this book be used as a supplementary text in a graduate-level course in structural equation modeling. This book … provides students with the necessary literature for a broad understanding of structural equation modeling. –Structural Equation Modeling
This book is worth its weight in gold! Drawing on the expertise of key researchers in the field, Bollen and Long provide readers with a comprehensive review of the critical issues, as well as innovative approaches that address these issues in the fitting, estimating, and testing of structural equation models. The book is an absolute ‘must’ for all researchers interested in conducting sound structural equation modeling applications.
–Barbara M. Byrne, Department of Psychology, University of Ottawa, Ontario
This collection of papers, so nicely written for and edited by Professors Bollen and Long, presents the ‘state of the art’ in significance testing and goodness-of-fit indices for structural equation models. The coverage of topics is almost as impressive as the set of authors–nearly all the methodological leaders in this important and quite active research area have helped make this volume an immediate classic. It should be used as a text in graduate-level courses on structural equation models to augment the standard textbooks. The editors are to be praised for lending their own expertise and for taking the time to put this excellent collection together.
–Stanley Wasserman, Departments of Psychology and Statistics, University of Illinois, Urbana-Champaign What is the role of fit measures when respecifying a model? Should the means of the sampling distributions of a fit index be unrelated to the size of the sample? Is it better to estimate the statistical power of the chi-square test than to turn to fit indices? Aimed at exploring these and other related questions, this group of well-known scholars examines the methods of testing structural equation models (SEMS) with and without measurement error–as estimated by such programs as EQS, LISREL, and CALIS. Highly integrated and valuable, this book is a must for every researcher’s shelf, particularly with coverage like: testing structural equation models, multifaceted conceptions of fit, Monte Carlo evaluations of goodness of fit indices, specification tests for the linear regression model, bootstrapping goodness of fit measures, bayesian model selection, alternative ways of assessing model fit, power evaluations, goodness of fit with categorical and other non-normal variables, new covariance structure model improvement statistics, and nonpositive definite matrices.