Discusses automatic thesaurus construction and characterizes the statistical behavior of terms by using an inference network. Highlights include low-frequency terms and data sparseness, Bayesian networks, collocation maps and term similarity, constructing a thesaurus from a collocation map, and experiments with test collections. (Author/LRW)

Explores the robustness of conclusions from a statistical model against variations in model choice with an illustration from G. Box and G. Tiao (1973). Suggests that simultaneous consideration of a class of models for the same data is sometimes superior to analyzing the data under one model and demonstrates advantages to Adaptive Bayesian…

Formulates optimal sequential rules for mastery testing using an approach derived from Bayesian sequential decision theory to consider both threshold and linear loss structures. Adopts the binomial probability distribution as the psychometric model. Provides an empirical example for concept-learning in medicine. (SLD)

Presents a fully Bayesian analysis for the Probability Matrix Decomposition (PMD) model using the Gibbs sampler. Identifies the advantages of this approach and illustrates the approach by applying the PMD model to opinions of respondents from different countries concerning the possibility of contracting AIDS in a specific situation. (SLD)

For complex educational assessments, there is an increasing use of item families, which are groups of related items. Calibration or scoring in an assessment involving item families requires models that can take into account the dependence structure inherent among the items that belong to the same item family. This article extends earlier works in…

D. Trafimow presented an analysis of null hypothesis significance testing (NHST) using Bayes's theorem. Among other points, he concluded that NHST is logically invalid, but that logically valid Bayesian analyses are often not possible. The latter conclusion reflects a fundamental misunderstanding of the nature of Bayesian inference. This view…

This paper comments on the response offered by Trafimow on Lee and Wagenmakers comments on Trafimow's original article. It seems our comment should have made it clear that the objective Bayesian approach we advocate views probabilities neither as relative frequencies nor as belief states, but as degrees of plausibility assigned to propositions in…

A scheme is described for locally Bayesian parameter updating in models structured as successions of component functions. The essential idea is to back-propagate the target data to interior modules, such that an interior component's target is the input to the next component that maximizes the probability of the next component's target. Each layer…

The "Facilities Performance Indicators Survey" ("FPI") supersedes and builds upon the two major surveys APPA conducted in the past: the Comparative Costs and Staffing (CCAS) survey and the Strategic Assessment Model (SAM). The "FPI" covers all the materials collected in CCAS and SAM, along with some select new data points and improved survey…

This study examines how high school students were affected by having an openly gay teacher and compares its results to a previous study (Rofes, 2000). The author's lesbian, gay, or bisexual students experienced a sense of relief that they could finally feel comfortable about themselves, as well as feeling happy that others in the school were…

This study examined the accuracy of a curriculum-based assessment for use during the eligibility process for Individuals with Disabilities Education Improvement Act (IDEA) services. The study extended a previous investigation in which performance scores of children without disabilities on the Assessment, Evaluation, and Programming System (AEPS)…

See TM 501 201-3 and EJ 060 883 for related articles. (MLP)

A computer program for two-group nonparametric discriminant analysis is presented. Based on Bayes' Theorem for probability revision, the statistical rationale for this program uses the calculation of maximum likelihood estimates of group membership. The program compares the Bayesian procedure to the standard Linear Discriminant Function.…

A system of automatic indexing based on Baye's theorem is described briefly. (18 references) (Author)