Studied the meaning of Performance Failure Appraisal Inventory (PFAI; Conroy and others, 2002) by evaluating the comparability of PFAI factor structure over repeated assessments and the stability of the subscales over relatively brief intervals. Results for 356 college students generally show high stability for PFAI scores in long and short forms.…

Used confirmatory factor analysis to evaluate the score integrity of LibQUALl+, an instrument to measure perceptions of library service quality. Results for 60,027 graduate and undergraduate students suggest that the model implied by LibQUAL is reasonable and invariant across independent samples and fits all three major subgroups of library users.…

Monte Carlo methods were used to study the accuracy and utility of estimators of overall error and error due to approximation in structural equation modeling. Effects of sample size, indicator reliabilities, and degree of misspecification were examined. The rescaled noncentrality parameter also was examined. Choosing among competing models is…

Relations of social support and self-esteem to problem behavior were investigated among young adolescents. Three models were evaluated, and the mediated and direct effects model was found best fitting. Social support and self-esteem predicted less involvement in problem behavior, and unique variance in peer self-esteem predicted greater problem…

As a reaction to the article by D. M. Tokar et al. (this issue), discusses the most pressing methodological and theoretical challenges embedded in the study of family relationship factors in the career decision making of college students. Reviews outstanding theoretical and empirical issues characteristic of structural equation modeling, with a…

In his comments regarding Tokar et al.'s article (this issue) Blustein offered several useful suggestions for future research in the area of relationship variables and career development and raised several concerns about the rationale for and use of structural equation modeling in testing their models. In this reply, the authors note points of…

Discusses three research/testing interfaces in second-language (L2) testing: the covariance structure analysis of ex post facto correlational data, the qualitative investigation of test-taking processes, and the development of L2 assessment instruments based on developmental sequences in L2 acquisition. (61 references) (GLR)

The logical flaw of using the methods of path analysis and structural equation modeling to make causative conclusions is demonstrated. A proper evaluation of the role of investigation versus experimentation is cited in the work of W. G. Cochran as explicated by D. R. Rubin (1983). (TJH)

Reports on a study of 101 Canadian librarians that tested individual differences important to enhanced use of the Internet. Analysis of covariance, hierarchical regression, and structural equation modeling provided evidence that technical and procedural knowledge had a significant influence on attitudes toward the Internet and the frequency of…

Suggests that the ability to process narrative information is fundamental to understanding human psychological development. Notes that a culture's system of understanding and interpreting the world is carried mostly by stories and texts. Explores how narrative understanding can be modeled in Fuzzy Trace Theory by using the principles of this…

Alternative models for a large dataset were analyzed by asymptotically distribution-free, generalized least squares, maximum likelihood, and ordinary least squares estimation methods, comparing incremental and nonincremental fit indexes. Incremental methods were quite unstable across estimation methods. This phenomenon is explained. (SLD)

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)

Structural equation modeling was used to examine the relationship among student skills, family climate, and educational opportunities, and self-determination and student action in 237 secondary students with disabilities. Results indicate educational opportunities were related to student self-determination and action taking, with a lack of a…

Used simulation to compare the ability of maximum likelihood (ML) and generalized least-squares (GLS) estimation to provide theoretic fit in models that are parsimonious representations of a true model. The better empirical fit obtained for GLS, compared with ML, was obtained at the cost of lower theoretic fit. (Author/SLD)

Multiple observer ratings of 424 families were obtained across 2 observational task situations using the Iowa Family Interaction Rating Scales. Observer ratings and family member reports were assessed simultaneously through structural equation modeling. Findings support the reliability of the global assessments of warm/supportive marital…