Two zero-one linear programing models for constructing tests using classical item and test parameters are given. These models are useful, for instance, when classical test theory must serve as an interface between an item response theory-based item banking system and a test constructor unfamiliar with the underlying theory. (TJH)

Reviews optimal test-assembly literature and introduces the contributions to this special issue. Discusses four approaches to computerized test assembly: (1) heuristic-based test assembly; (2) 0-1 linear programming; (3) network-flow programming; and (4) an optimal design approach. Contains a bibliography of 90 sources on test assembly.…

Applied optimal design methods to the item-bank design of adaptive testing for continuous testing situations using a version of the weighted-deviations model (M. Stocking and L. Swanson, 1993) in a simulation. Independent and overlapping item banks used items more efficiently than did a large item bank. (SLD)

Describes a procedure for automated-test-forms assembly based on Classical Test Theory (CTT). The procedure uses stratified random-content sampling and test-form preequating to ensure both content and psychometric equivalence in generating virtually unlimited parallel forms. Extends the usefulness of CTT in automated test construction. (Author/SLD)

Ensuring item and pool security in a continuous testing environment is explored through a new method of controlling exposure rate of items conditional on ability level in computerized testing. Properties of this conditional control on exposure rate, when used in conjunction with a particular adaptive testing algorithm, are explored using simulated…

Describes a study that was conducted to track the effect of candidate response patterns on a computerized adaptive test. The effect of altering responses on estimated candidate ability, test tailoring, and test precision across segments of adaptive tests and groups of candidates is examined. (Author/LRW)

The three-parameter logistic (3-PL) model is described and a derivation of the 3-PL observed information function is presented for a single binary response from one examinee with known item parameters. Formulas are presented for the probability of negative information and for the expected information (always nonnegative). (SLD)

Showed how Taylor approximation can be used to generate a linear approximation to a logistic item characteristic curve and a linear ability estimator. Demonstrated how, for a specific simulation, this could result in the special case of a Robbins-Monro item selection procedure for adaptive testing. (SLD)

Administrative office technology assessments were delivered via the Internet to 141 vocational students. Compared to 61 who took a paper test, scores for the Internet group were significantly higher; 62% preferred the Internet format. There were few problems with accessing, downloading, taking, and submitting the test. (SK)

Presents a constrained computerized adaptive testing (CAT) algorithm that can be used to equate CAT number-correct scores to a reference test. Used an item bank from the Law School Admission Test to compare results of the algorithm with those for equipercentile observed-score equating. Discusses advantages of the approach. (SLD)

Studied strategy development in a computerized task developed from the block design task of S. Kohs (1923). Results for 90 subjects aged 12, 17, and adult show 3 strategies: global, analytic, and "synthetic." Discusses some age differences and implications for recent work in cognitive and developmental psychology. (SLD)

Considers computerized adaptive testing from the perspective of graphical modeling (GM). GM provides methods for making inferences about multifaceted skills and knowledge and for extracting data from complex performances. Provides examples from language-proficiency assessment. (SLD)

Evaluates a method for item selection in adaptive testing that is based on Kullback-Leibler information (KLI) (T. Cover and J. Thomas, 1991). Simulation study results show that testing algorithms using KLI-based item selection perform better than or as well as those using Fisher information item selection. (SLD)

Brothen and Wambach (2001) found that under certain conditions students do not benefit from online resources. They awarded points toward the course grade for high performance on small sets of practice questions and concluded student strategies in completing these quizzes were not beneficial to course performance. We investigated student use of…

The author breaks down the debate over high-stakes testing into three politically charged issues: how test scores are used as a means to reform schools; whether test designs and passing scores truly reflect students' abilities; and how public perception of schools can be unfairly based on test results alone.