Test-item writing efforts typically results in item pools with an undesirable correlational structure between the content attributes of the items and their statistical information. If such pools are used in computerized adaptive testing (CAT), the algorithm may be forced to select items with less than optimal information, that violate the content…

This paper presents a multistage adaptive testing test development paradigm that promises to handle content balancing and other test development needs, psychometric reliability concerns, and item exposure. The bundled multistage adaptive testing (BMAT) framework is a modification of the computer-adaptive sequential testing framework introduced by…

The authors of the "Guidelines," a task force of eight, intend to present an organized list of features to be considered in reporting or evaluating computerized-adaptive assessments. Apart from a few weaknesses, the book is a useful and complete document that will be very helpful to test developers. (SLD)

This study was designed to address two frequent criticisms of the SAT essay--that essay length is the best predictor of scores, and that there is an advantage in using more "sophisticated" examples as opposed to personal experience. The study was based on 2,820 essays from the first three administrations of the new SAT. Each essay was…

Score comparability is important to those who take tests and those who use them. One important concept related to test score comparability is that of "equity," which is defined as existing when examinees are indifferent as to which of two alternate forms of a test they would prefer to take. By their nature, computerized adaptive tests…

The relatively new shadow test approach (STA) to computerized adaptive testing (CAT) proposed by Wim van der Linden is a potentially attractive alternative to the weighted deviation algorithm (WDA) implemented at ETS. However, it has not been evaluated under testing conditions representative of current ETS testing programs. Of interest was whether…

Describes how to construct questions to test all cognitive levels of learning for a course in introductory programming. The framework for learning is based on Bloom's taxonomy. Practical advice is given through examples and by describing a computer system to help deliver quizzes. Includes example questions. (Author/AEF)

Technology today offers many new opportunities for innovation in educational assessment through rich new assessment tasks and potentially powerful scoring, reporting and real-time feedback mechanisms. One potential limitation for realizing the benefits of computer-based assessment in both instructional assessment and large scale testing comes in…

As testing moves from paper-and-pencil administration toward computerized administration, how to present tests on a computer screen becomes an important concern. Of particular concern are tests that contain necessary information that cannot be displayed on screen all at once for an item. Ideally, the method of presentation should not interfere…

A model for constrained computerized adaptive testing is proposed in which the information in the test at the ability estimate is maximized subject to a large variety of possible constraints on the contents of the test. At each item-selection step, a full test is first assembled to have maximum information at the current ability estimate fixing…

Defines computer assisted language testing (CALT), discusses the various processes involved, outlines the advantages and disadvantages, and examines psychometric aspects of computer testing. A table of factors distinguishes between test content and the mechanics of test taking. These factors constitute a table for developing a CALT checklist. (24…

The purpose of this study was to compare the effects of four item selection rules--(1) Fisher information (F), (2) Fisher information with a posterior distribution (FP), (3) Kullback-Leibler information with a posterior distribution (KP), and (4) completely randomized item selection (RN)--with respect to the precision of trait estimation and the…

This study seeks to determine whether item features are related to observed differences in item difficulty (DIF) between computer- and paper-based test delivery media. Examinees responded to 60 quantitative items similar to those found on the GRE general test in either a computer-based or paper-based medium. Thirty-eight percent of the items were…

The use of a variation of linear programing (goal programing) to develop a test blueprint with multiple specification requirements, is the topic of this paper. The computer program STORM was used to develop a table of test specifications for a test that would measure achievement in introductory statistics in the subdomains of frequency…

Three different applications of the testlet concept are presented, and the psychometric models most suitable for each application are described. Difficulties that testlets can help overcome include (1) context effects; (2) item ordering; and (3) content balancing. Implications for test construction are discussed. (SLD)