A general model for two-level multivariate data, with responses possibly missing at random, is described. The model combines regressions on fixed explanatory variables with structured residual covariance matrices. The likelihood function is reduced to a form enabling computational methods for estimating the model to be devised. (Author)

In experiments with paired comparisons, judges are occasionally allowed to express indifference between alternatives. For the analysis of such data, models for paired comparisons with ties are needed. Several models with tie parameters are reviewed. All are extensions of the basic models of L. L. Thurstone (1927) and R. A. Bradley and M. E. Terry…

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…

An efficient procedure is described for specifying and estimating parameters of general mixed models that contain both hierarchical and crossed random factors. The approach, using a model formulated for purely hierarchical structured data, is illustrated for the continuous response linear model with extensions to generalized linear, nonlinear, and…

Discusses two estimators of the variance of random effects models often used in meta-analysis. Provides mathematical proof that the second, less intuitively appealing, estimator is more efficient when the variance of the random effects model is small. (SLD)

The issue of person fit has received an increasing amount of attention by researchers in the past few years. Several studies have focused on the issue of how nonmodel-fitting responses affect the accuracy of ability estimates (e.g. R. Meijer and S. Nering, in press; Reise, 1995). The purpose of this study was to examine the effects that…

Describes the development and evaluation of the statistical models used to produce indirect state estimates for schools, teachers, students, and graduates from the Private School Survey of the National Center for Education Statistics for the 1991-92 and 1993-94 school years. (SLD)

The buffer model proposed by Atkinson and Shiffrin gives a valid measure of primary-memory capacity (i.e., the estimate is close to the true value), a property lacking in both the original Waugh and Norman method and Watkins' modifications of that method. (Author/PN)

Describes four approaches to factor analysis of ordinal variables that take proper account of ordinality and compared three of these approaches with respect to parameter estimates and fit using generated data and an empirical data set. Focuses on how to test the model and how to measure model fit. (SLD)

A method is proposed to equate different sets of items administered to different groups of individuals using the Rasch model. A Rasch equating model was formulated to describe one common Rasch scale in different groups with different but overlapping sets of items. The item parameters can then be estimated simultaneously, avoiding different…

Studied the performance of rank (percentile) estimators in a basic two-stage hierarchical model capturing essential features of more complicated models commonly used to estimate effects. Simulation results highlight the need to assess whether even optimal percentile estimators perform sufficiently well to be used to evaluate teachers or schools.…

Investigated the data demands associated with the marginal maximum likelihood (MML) expected a posterior (EAP) methodology and the precision of the resulting parameter estimates when data fit the underlying model through simulation. Also studied the extent to which a misspecified prior distribution would affect the item and person parameter…

Residuals for checking model fit in the polytomous Rasch model are examined. Comparisons are made between using counts for all response patterns and using item totals for score groups in the construction of the residuals. Using item totals for residuals based on score group totals was compared with using the estimated item parameters. (SLD)

Suggests a probabilistic method to compute the term relationships from relevance information, which complements the studies on a nonprobabilistic technique called pseudo-classification. A quadratic ranking function is derived by incorporating the term-by-term relationships. Procedures for estimating the required parameters are provided by…

Develops an item-parameter estimation procedure using an EM algorithm for the continuous-response model (F. Samejima, 1973). Examines the usefulness of this model using simulated data. Results show that the procedure performs well in estimating items and theta parameters. (SLD)