Computerized adaptive testing (CAT), while well-grounded in psychometric theory, has had few large-scale applications for high-stakes, secure tests in the past. This is now changing as the cost of computing has declined rapidly. As is always true where theory is translated into practice, many practical issues arise. This paper discusses a number…

In this study some alternative item selection criteria for adaptive testing are proposed. These criteria take into account the uncertainty of the ability estimates. A general weighted information criterion is suggested of which the usual maximum information criterion and the suggested alternative criteria are special cases. A simulation study was…

Previous studies have shown that, when administered a self-adapted test, a few examinees will choose item difficulty levels that are not well-matched to their proficiencies, resulting in high standard errors of proficiency estimation. This study investigated whether the previously observed effects of a self-adapted test--lower anxiety and higher…

This document provides printed instructions for teachers to use with an IBM-compatible microcomputer to construct tests and then have the computer give the tests, grade them, and print the test results. Computerized tests constructed in this way may contain true-false questions, multiple-choice questions, or a combination of both. The questions…

Expert systems can be used to aid decisionmaking. A computerized adaptive test is one kind of expert system, although not commonly recognized as such. A new approach, termed EXSPRT, was devised that combines expert systems reasoning and sequential probability ratio test stopping rules. Two versions of EXSPRT were developed, one with random…

To investigate the psychometric feasibility of replacing a paper-and-pencil licensing examination with a computer-administered test, a validity study was conducted. The computer-administered test (Cadm) was a common set of items for all test takers, distinct from computerized adaptive testing, in which test takers receive items appropriate to…

In self-adapted testing, examinees are allowed to choose the difficulty of each item to be presented immediately before attempting it. Previous research has demonstrated that self-adapted testing leads to better performance than do fixed-order tests and is preferred by examinees. The present study examined the strategies that 29 college students…

The effects of item position on item statistics were studied in a large set of data from tests of word knowledge (WK) and arithmetic reasoning (AR). Position effects on item response theory (IRT) parameter estimates and classical item statistics were also investigated. Data were collected as part of a project to refine the Army's Computerized…

The process of writing test items is analyzed, and a blueprint is presented for an authoring system for test item writing to reduce invalidity and to structure the process of item writing. The developmental methodology is introduced, and the first steps in the process are reported. A historical review traces the advances made in the field and the…

Four discussions of test construction based on item response theory (IRT) are presented. The first discussion, "Test Design as Model Building in Mathematical Programming" (T. J. J. M. Theunissen), presents test design as a decision process under certainty. A natural way of modeling this process leads to mathematical programming. General…

The answer-until-correct procedure has made comparatively little impact on the field of educational testing due to the absence of a sound theoretical base for turning the response data into measures. Three new latent trait models are described. They differ in their complexity, though each is designed to yield a single parameter to measure student…

An earlier study was extended and replicated to examine the feasibility of generating multiple-choice test questions by transforming sentences from prose instructional material. In the first study, a computer-based algorithm was used to analyze prose subject matter and to identify high-information words. Sentences containing selected words were…

Discusses the use of a bank of about 9,000 test items in a computer-assisted instructional program at Adelaide University (South Australia). Describes the program and outlines the steps in producing an instructional program. (TW)

Multidimensional item response data were created. The strength of a general factor, the number of common factors, the distribution of items loadingon common factors, and the number of items in simulated tests were manipulated. LOGIST effectively recovered both item and trait parameters in nearly all of the experimental conditions. (Author/JAZ)

The purpose of this study was to compare and evaluate three online pretest item calibration/scaling methods in terms of item parameter recovery when the item responses to the pretest items in the pool would be sparse. The three methods considered were the marginal maximum likelihood estimate with one EM cycle (OEM) method, the marginal maximum…