Computerized adaptive testing is subject to security problems, as the item bank content remains operative over long periods and administration time is flexible for examinees. Spreading the content of a part of the item bank could lead to an overestimation of the examinees' trait level. The most common way of reducing this risk is to impose a…

The possibility of using component analysis for nominal data is discussed. Two nominal scale correlation coefficients are applicable. Tschuprow's coefficient and the J index. The reason is that they satisfy the requirements of a scalar product between normalized vectors in a Euclidean space. Some characteristics of these coefficients are…

Describes the individual chapters of this collection and notes that, although the book lacks consistency in some respects, it contains state-of-the-art reflections on the modeling of change and should stimulate discussion for experts and practitioners. (SLD)

Compared two methods for developing a common metric for the graded response model under item response theory: (1) linking separate calibration runs using equating coefficients from the characteristic curve method; and (2) concurrent calibration using the combined data of the base and target groups. Concurrent calibration yielded consistently,…

Demonstrated that the extra-factor phenomenon (the two-dimensional solution produced when linear factor analysis is applied to a set of unfoldable items) arises because the metric unidimensional unfolding model is equivalent to the unidimensional quadratic factor model and the unidimensional quadratic factor model is not distinguishable from the…

PARAFAC is a generalization of principal components analysis in a factor score matrix and in a factor loadings matrix. How PARAFAC behaves when applied to positive manifold data is examined, and a constrained PARAFAC method is offered for use when PARAFAC does not produce a positive manifold solution. (SLD)

A simplified version of successive intervals scaling is described. Scale values for various datasets obtained with simplified successive intervals scaling are approximately linearly related to those obtained with traditional successive intervals scaling and the method of pair comparisons. The ease of use and some cautions in method use are…

The weighted kappa coefficient is a measure of interrater agreement when the relative seriousness of each possible disagreement can be quantified. This monte carlo study demonstrates the utility of the kappa coefficient for ordinal data. Sample size is also briefly discussed. (Author/JKS)

Extends joint maximum likelihood estimation for the hyperbolic cosine model to the situation in which the units of items are allowed to vary. Describes the four estimation cycles designed to address four important issues of model development and presents results from two sets of simulation studies that show reasonably accurate parameter recovery…

Rasch's psychometric model is presented as a special case of additive conjoint measurement. Two data sets, each of which was analyzed using Rasch scaling methods as well as some of the methods of conjoint measurement, are compared. (Author/JKS)

A test theory model based on the Thurstone judgmental model is described. By restricting various parameters of the model, 3 Rasch models, 2 pseudo-Rasch models, 3 two-parameter models, and a Weber's Law model are derived. (Author/CTM)

Developed a multidimensional Rasch model that is a multidimensional extension of item component models and determined relations to existing multidimensional item response theory models. Applied the model to a questionnaire on students' interest in physics to demonstrate that model parameters can be estimated even when 17 latent dimensions are to…

How longitudinal factor score estimation--the estimation of the evolution of factor scores for individual examinees over time--can profit from the Kalman filter technique is described. The Kalman estimates change more cautiously over time, have lower estimation error variances, and reproduce the LISREL program latent state correlations more…

The effect of scale selection on error in calibrating item and ability parameters was investigated, with particular reference to the standardized mean-squared difference (SMSD) statistic. Through simulation, three scaling methods for selecting the common scale were used to demonstrate their effects on SMSD values. (SLD)

Linear equating methods for the common-item non-equivalent populations design were compared when true-score correlation between the test and anchor was less than unity. Scores from two groups of approximately 300 examinees illustrated three methods: (1) the Tucker equating method; (2) the Angoff-Levine method; and (3) the Congeneric-Levine method.…