The root mean square error of approximation (RMSEA) is a popular fit index in structural equation modeling (SEM). Typically, RMSEA is computed using the normal theory maximum likelihood (ML) fit function. Under nonnormality, the uncorrected sample estimate of the ML RMSEA tends to be inflated. Two robust corrections to the sample ML RMSEA have…

The comparative format used in ranking and paired comparisons tasks can significantly reduce the impact of uniform response biases typically associated with rating scales. Thurstone's (1927, 1931) model provides a powerful framework for modeling comparative data such as paired comparisons and rankings. Although Thurstonian models are generally…

Demonstrates the use of the well-known Bayes factor in the Bayesian literature for hypothesis testing and model comparison in general two-level structural equation models. Shows that the proposed method is flexible and can be applied to situations with a wide variety of nonnested models. (SLD)

Manifest variables in covariance structure analysis are often combined to form parcels for use as indicators in a measurement model. The purpose of the present study was to evaluate four empirical algorithms for creating such parcels, focusing on the effects of dimensionality on accuracy of parameter estimation and model fit. Results suggest that…

The main objective of this article is to investigate the empirical performances of the Bayesian approach in analyzing structural equation models with small sample sizes. The traditional maximum likelihood (ML) is also included for comparison. In the context of a confirmatory factor analysis model and a structural equation model, simulation studies…

Suppose P to the subpower of 1 and P to the subpower of 2 are two competing discrimination procedures. To compare the relative discriminatory power of both procedures, test statistics are suggested for the hypothesis that P to the subpower of 1 and P to the subpower of 2 performed no better than random assignment versus the alternative that P to…

Six methods for fitting path models with weighted composites of variables replacing latent variables (of which five are easily implemented with conventional computer software) are introduced and related to "soft" modeling by Partial Least Squares. Criteria for comparing their performance are devised, and some evaluative remarks are…

This article examines whether relationships between individual characteristics and HIV status can be identified when self-report data are used as a proxy for HIV serotest results. The analyses use data obtained from HIV serotests and face-to-face interviews with 7,256 out-of-treatment drug users in ten sites from 1992 to 1998. Relationships…

Two recent indices of fit, the Relative Noncentrality Index (RNI) (R. P. McDonald and H. W. Marsh, 1990) and the Comparative Fit Index (P. M. Bentler, 1990), are shown to be algebraically equivalent in most applications, although one condition in which the RNI may be advantageous for model comparison is identified. (SLD)

Compared the fit of several Item Response Theory (IRT) models to two personality assessment instruments using data from 13,059 individuals responding to one instrument and 1,770 individuals responding to the other. Two- and three-parameter logistic models fit some scales reasonably well, but not others, and the graded response model generally did…

Factor analysis, image analysis, and principal component analysis are compared with respect to the factor patterns they would produce under various conditions. The general conclusion that is reached is that the three methods produce results that are equivalent. (Author/JKS)

Use of confirmatory factor analysis (CFA) with nonzero latent means in testing six different measurement models from classical test theory is discussed. Implications of the six models for observed mean and covariance structures are described, and three examples of the use of CFA in testing the models are presented. (SLD)

To assess problems in hypothesis testing and model comparisons based on normed indices caused by latent class models with sparse contingency tables, a simulation was carried out investigating the distributions of the likelihood ratio statistic, the Pearson statistic chi-square, and a new goodness of fit statistic. (SLD)