A computationally efficient three-stage estimator of thresholds and covariance structure parameters is prepared for analysis of structural equation models with polytomous variables. The method is based on partition maximum likelihood and generalized least squares estimation. An analysis of questionnaire responses of 739 young adults illustrates…

Using an approach to the analysis of missing data, this study investigated developmental trends in alcohol, marijuana, and cigarette use among 750 adolescents across 5 years using multiple-group latent growth modeling. Latent variable structural equation modeling and missing data approaches to studying developmental change are explored. (SLD)

A study at the University of Tennessee Knoxville used mixed-effect structural equation models incorporating latent variables as an alternative to conventional methods of analyzing college students' (n=722) first-year-to-senior academic gains. Results indicate, contrary to previous analysis, that coursework and student characteristics interact to…

A two-stage approach based on the rationale of maximum likelihood and generalized least-squares methods is developed to analyze the general structural equation model for continuous and polytomous variables. Some illustrative examples and a small simulation study (50 replications) are reported. (SLD)

Three psychophysical auditory processing tests, varying in emphasis on speed of processing and pitch discrimination, and tests of verbal and nonverbal ability were administered to 108 13-year olds. A structural equation model suggests that information processing speed and pitch discrimination ability are significantly associated with nonverbal and…

The nature of the attitude-behavior relation was investigated through the use of structural equation models in a cross-lagged panel design involving 254 undergraduates reacting to 2 kinds of goal-directed behaviors. Results suggest that intentions, and to a lesser extent attitudes, are dependent on behavior and not the reverse. (SLD)

A study integrated the major propositions of 2 theories of college persistence (Tinto's and Bean's) and used the resulting framework to survey a population of 466 first-year college students at 1 university. Findings supported most of the hypothesized links between the models and revealed a complex role for environmental factors in retention.…

Examined relation between fathers' and children's communicative skills and child behavior problems early and late in the Head Start school year. Found that the structural models for externalizing and internalizing behavior confirmed the hypothesis that father communication was linked to child communication skills and child communication was linked…

Statistical methods related to Rasch modeling, social network analysis, and latent variable structural equation modeling for experimental designs are critiqued with regard to three important issues: (1) measuring constructs; (2) modeling processes of change; (3) operationalizing social relations. An argument is made for adequate conceptualization…

Three proposed models were tested to explore the inter-relationships affecting student academic performance and nonacademic factors using structural equation modeling. Results with 147 undergraduates suggest that a Bio-Model and a Family Mediated Model serve to predict academic performance moderately well. (SLD)

An alternative two-stage least squares (2SLS) estimator of the parameters in LISREL type models is proposed and contrasted with existing estimators. The new 2SLS estimator allows observed and latent variables to originate from nonnormal distributions, is consistent, has a known asymptotic covariance matrix, and can be estimated with standard…

Provides an introduction to structural equation modelling (SEM) for language research, including: general objectives of SEM applications relevant to language assessment; methodology and statistical assumptions about data that must be met; commonly-used SEM steps and concepts; application matters, with sample models; and recent critical discussions…

Studied the mechanisms by which social competence may be associated with substance use during early adolescence in a sample of 1,568 rural junior high school students. Structural equation modeling indicated that social competence had a direct protective association with substance use. Discusses mediating factors. (SLD)

Examined relations between mothers' expressed positive and negative emotion and 55- to 79-month-olds' regulation, social competence, and adjustment. Structural equation modeling revealed unique effects of positive and negative maternal expressed emotion on children's regulation, and the relations of maternal expressed emotion to children's…

This article describes a SAS macro to assess model fit of structural equation models by employing a test of the model-implied vanishing tetrads. Use of this test has been limited in the past, in part due to the lack of software that fully automates the test in a user-friendly way. The current SAS macro provides a straightforward method for…