Structural Equation Models A Review With Applications to Environmental Epidemiology

• Published in: Journal of the American Statistical Association Vol. 100; no. 472; pp. 1443 - 1455
• Main Authors: Sánchez, Brisa N, Budtz-Jørgensen, Esben, Ryan, Louise M, Hu, Howard
• Format: Journal Article
• Language: English
• Published: Alexandria, VA Taylor & Francis 01.12.2005; American Statistical Association; Taylor & Francis Ltd
• Subjects: Applications; Biological markers; Biology, psychology, social sciences; Blood; Bones; Cord blood; Correlated outcomes; Covariance; Data analysis; Environmental science; Epidemiology; Exact sciences and technology; General topics; Latent variable; Linear equations; Lisrel model; Mathematical foundations; Mathematical independent variables; Mathematical models; Mathematics; Measurement error; Measurement errors; Medical sciences; Methodology; Modeling; Multiple testing; Multivariate analysis; Multivariate data; Parametric models; Probability and statistics; Review Article; Sciences and techniques of general use; Statistical discrepancies; Statistical methods; Statistical models; Statistics; Structural equation models; Multivariate data; Error estimation; LISREL model; Psychometrics; Latent variable model; Multivariate analysis; Epidemiology; Implementation; Statistical method; Statistical regression; Latent variable; Structural model; Correlated outcomes; Multiple testing; Measurement method; Application; Measurement error
• ISSN: 0162-1459; 1537-274X
• DOI: 10.1198/016214505000001005

Structural equation models (SEMs) have been discussed extensively in the psychometrics and quantitative behavioral sciences literature. However, many statisticians and researchers in other areas of application are relatively unfamiliar with their implementation. Here we review some of the SEM literature and describe basic methods, using examples from environmental epidemiology. We make connections to recent work on latent variable models for multivariate outcomes and to measurement error methods, and discuss advantages and disadvantages of SEMs compared with traditional regressions. We give a detailed example in which two models fit the same data well, yet one is physiologically implausible. This underscores the critical role of subject matter knowledge in the successful implementation of SEMs. A brief discussion on open research areas is included.