A program named EXACON is presented which performs exact cellwise analyses of two-way contingency tables. One-tailed probabilities are computed for the observed frequency of each cell according to two different models. Even though exact probabilities are computed, EXACON does not demand much computer time even for fairly large samples. (Author/LMO)

Formally states the Thick Point Theorem and explicates the proof. Explores the key point for use of the Theorem. (GEA)

Presents a mathematical model of a decision maker subject to goal uncertainty which shows that a dominant course of action will always emerge. The theory of multivalued functions is used to express goal uncertainty, and a specific example is described to provide insight into how the procedure works in practice. (Author/EM)

Presents notations and definitions necessary to identify the concepts and relationships that are important in modelling information retrieval objects and processes in the context of vector spaces. Earlier work on the use of vector models is evaluated in terms of the concepts introduced and certain problems are identified. (Author/EM)

An alternative for analyzing responses to Likert Scales is proposed, using additive conjoint measurement. It assumes that subjects can report their attitudes toward stimuli in rank order. Neither within-subject nor between-subject distributional assumptions are made. Nevertheless, interval level stimulus values and response category boundaries are…

Preparing this article posthumously from Choppin's presentation notes, the author used the historical development of thermometry to suggest some lessons for educational measurement: (1) mathematical models are important; (2) models can be useful long before their underlying processes are understood; and (3) since there are no true models, there…

The axioms of a three-set Guttman simplex model are presented and the effects of relaxing the axioms for one of the three sets are examined. This model can be used to define longitudinal developmental scales. (NSF)

Subjected sequences of violent and nonviolent offenses to log-linear analyses of the stabilities and magnitudes of their transition probabilities. Results were seen to support previous research in which nonviolent criminality emerged as more fundamental than violence in potential for pattern development. (LLL)

A definition of test validity as the ratio of a covariance term to a variance term, analogous to the classical definition of test reliability, is proposed. When error scores on distinct tests are uncorrelated, the proposed definition coincides with the usual one, but it remains meaningful when error scores are correlated. (Author/BW)

This study investigated statistical methods for identifying errors in Bayesian networks (BN) with latent variables, as found in intelligent cognitive assessments. BN, commonly used in artificial intelligence systems, are promising mechanisms for scoring constructed-response examinations. The success of an intelligent assessment or tutoring system…

This paper describes and evaluates the use of decision theory as a tool for classifying examinees based on their item response patterns. Decision theory, developed by A. Wald (1947) and now widely used in engineering, agriculture, and computing, provides a simple model for the analysis of categorical data. Measurement decision theory requires only…

This paper explores the application of a mathematical optimization model to the problem of optimal enrollments. The general model, which can be applied to any institution, seeks to enroll the "best" class of students (as defined by the institution) subject to constraints imposed on the institution (e.g., capacity, quality). Topics…

This paper outlines two types of discriminant analysis, predictive discriminant analysis (PDA) and descriptive discriminant analysis (DDA). Important differences between PDA and DDA are introduced and discussed using a heuristic data set, specifically indicating the portions of the Statistical Package for the Social Sciences (SPSS) output relevant…

A Monte Carlo simulation approach was taken to investigate the effect of sample size on a variety of latent growth models. A fully balanced experimental design was implemented, with samples drawn from multivariate normal populations specified to represent 12 unique growth models. The models varied factorially by crossing number of time points,…

A group of 21 students was tested under a time limit considerably shorter than should have been allowed. This report describes a tryout of a method for estimating the power'' scores that would have been obtained if the students had had enough time to finish. (Author/CB)