Structural Equation Modeling

***Winner of the 2008 Ziegel Prize for outstanding new book ofthe year*** Structural equation modeling (SEM) is a powerful multivariatemethod allowing the evaluation of a series of simultaneoushypotheses about the impacts of latent and manifest variables onother variables, taking measurement errors into account. As SEMshave grown in popularity in recent years, new models andstatistical methods have been developed for more accurate analysisof more complex data. A Bayesian approach to SEMs allows the use ofprior information resulting in improved parameter estimates, latentvariable estimates, and statistics for model comparison, as well asoffering more reliable results for smaller samples. Structural Equation Modeling introduces the Bayesianapproach to SEMs, including the selection of prior distributionsand data augmentation, and offers an overview of thesubject's recent advances. * Demonstrates how to utilize powerful statistical computingtools, including the Gibbs sampler, the Metropolis-Hastingalgorithm, bridge sampling and path sampling to obtain the Bayesianresults. * Discusses the Bayes factor and Deviance Information Criterion(DIC) for model comparison. * Includes coverage of complex models, including SEMs withordered categorical variables, and dichotomous variables, nonlinearSEMs, two-level SEMs, multisample SEMs, mixtures of SEMs, SEMs withmissing data, SEMs with variables from an exponential family ofdistributions, and some of their combinations. * Illustrates the methodology through simulation studies andexamples with real data from business management, education,psychology, public health and sociology. * Demonstrates the application of the freely available softwareWinBUGS via a supplementary website featuring computer code anddata sets. Structural Equation Modeling: A Bayesian Approach is amulti-disciplinary text ideal for researchers and students in manyareas, including: statistics, biostatistics, business, education,medicine, psychology, public health and social science.