A general linear model framework is used to suggest that structure coefficients ought to be interpreted in structural equation modeling confirmatory factor analysis (CFA) studies in which factors are correlated. Two heuristic data sets make the discussion concrete, and two additional studies illustrate the benefits of CFA structure coefficients.…

Structural equation modelling is becoming increasingly popular in education. This article examines and compares a number of alternative assumptions governing nondirected paths in structural equation models without latent variables vis-a-vis a data set on lung ventilation. Some problems with the conventional procedures in path analysis are pointed…

Provides an introduction to the structural equation modeling concepts developed by J. Pearl, discussing the concept he calls "d-separation." Explains how d-separation connects to control variables, partial correlations, causal structuring, and even a potential mistake in regression. (SLD)

Illustrates how commonly available structural equation modeling programs can be used to conduct some basic matrix manipulations and generate multivariate normal data with given means and positive definite covariance matrix. Demonstrates the outlined procedure. (SLD)

Established a new criterion for the identification of recursive linear models in which some errors are correlated. Shows that identification is assured as long as error correlation does not exist between a cause and its direct effect; no restrictions are imposed on errors associated with indirect causes. (SLD)

Some problems of multiple-group factor rotation based on the parallel proportional profiles and confactor rotation of R. B. Cattell are described, and several alternative modeling solutions are proposed. Benefits and limitations of the structural-modeling approach to oblique confactor resolution are examined, and opportunities for research are…

A general approach is proposed to avoid improper solutions in structural equation models. The constrained estimation approach presented, which is based on an adaptation of a globally convergent method for nonlinear programming, has worked well in all trials. (SLD)

Describes how a hierarchical regression analysis may be conducted in structural equation modeling. The main procedure is to perform a Cholesky or triangular decomposition of the intercorrelations among the latest predictors. Provides an example of a hierarchical regression analysis with latent variables. (SLD)

Proposes an extension of the propensity score adjustment method to the analysis of group differences on latent variable models. Uses multiple indicators-multiple causes (MIMIC) structural equation modeling to test hypotheses about treatment group differences. Discusses the role of factorial invariance as it relates to this approach. (SLD)

Using a fictional dialog between Tweedledee and Tweedledum (L. Carroll, 1856), the work of H. W. Marsh and K.-T. Hau (1996) on parsimony is interpreted in several ways. Definitions of "judgment" and "rules" are presented and argued before the author concludes that a main thesis of the Marsh article is that judgment is essential…

Discusses techniques to account for unmeasured third variables in longitudinal designs, introducing a series of less restrictive synchronous common factor models as an extension of the synchronous common factor model. Recommends the use of such models, which can be tested by structural equation modeling, when possible third variables might have…

Discusses the propriety and practical advantages of specifying multivariate time series models in the context of structural equation modeling for time series and longitudinal panel data. For time series data, the multiple indicator model specification improves on classical time series analysis. For panel data, the multiple indicator model…

Outlines a covariance structure analysis approach to the study of parameter trends. Uses the program RAMONA to illustrate the method by fitting a corresponding confirmatory factor analysis model to correlational data from a study involving several psychometric tests and fluid intelligence tasks. (SLD)

Compares two statistical approaches for the analysis of data obtained from married couples. Summarizes a current multilevel (or hierarchical) model that has demonstrated usefulness in marital research and respecifies this model into a more familiar structural equation modeling formulation. (SLD)

Explains why, when one is using a bootstrapping approach for generating empirical standard errors for parameters of interest, the researchers must choose to fix an indicator path rather than the latent variable variance for the empirical standard errors to be generated properly. (SLD)