Bond criticized the base-free measure of change proposed by Tucker, Damarin, and Messick by pointing to an incorrect derivation which is here viewed instead as a correct derivation entailing an inadequately specified assumption. Bond's revision leads to negatively biased estimates, whereas the original approach leads to unbiased estimates.…

Presents minimization and maximization models for parallel test construction under constraints. The minimization model constructs weakly and strongly parallel tests of minimum length, while the maximization model constructs weakly and strongly parallel tests with maximum test reliability. (Author/SLD)

Rubin and Thayer recently presented equations to implement maximum likelihood estimation in factor analysis via the EM algorithm. It is argued here that the advantages of using the EM algorithm remain to be demonstrated. (Author/JKS)

The authors respond to a criticism of their earlier article concerning the use of the EM algorithm in maximum likelihood factor analysis. Also included are the comments made by the reviewers of this article. (JKS)

Polynomial algorithms are presented that are used to solve selected problems in test theory, and computational results from sample problems with several hundred decision variables are provided that demonstrate the benefits of these algorithms. The algorithms are based on optimization theory in networks (graphs). (SLD)

The Mantel-Haenszel (MH) procedure attempts to identify and quantify differential item performance (item bias). This paper summarizes the MH statistics, and identifies the parameters they estimate. An equivalent procedure based on the Rasch model is described. The theoretical properties of the two approaches are compared and shown to require the…

A network-flow model is formulated for constructing parallel tests based on classical test theory while using test reliability as the criterion. Practitioners can specify a test-difficulty distribution for values of item difficulties as well as test-composition requirements. An empirical study illustrates the reliability of generated tests. (SLD)

A continuum of item-writing methods is proposed ranging from informal-subjective methods to algorithmic-objective methods. Examples of techniques include objective-based item writing, amplified objectives, item forms, facet design, domain-referenced concept testing, and computerized techniques. (Author/CP)

Two optimalization models for the construction of tests with a maximal value of coefficient alpha are given. Both models have a linear form and can be solved by using a branch-and-bound algorithm. The first model assumes an item bank calibrated under the Rasch model and can be used, for instance, when classical test theory has to serve as an…

This report summarizes the results of research designed to study the psychometric and technological feasibility of adaptive testing to assess spatial ability. Data was collected from high school students on two types of spatial items: three-dimensional cubes and hidden figure items. The analysis of the three-dimensional cubes focused on the fit of…

Using data from 135 sixth-, seventh-, and eighth-graders between 11 and 15 years old attending a middle school in a suburban Southwest Ohio school district, two hypothesized models of the factor structures for the Coopersmith Self-Esteem Inventory were tested. One model represents the original Coopersmith factor structure, and the other model is…

Recent experience with the Computerized Adaptive Test (CAT) has raised a number of concerns about its practical applications. The concerns are principally involved with the concept of having the computer construct the test from a precalibrated item pool, and substituting statistical characteristics for the test developer's skills. Problems with…