This report summarizes the results from two studies. The first assessed the comparability of scores derived from linear computer-based (CBT) and computer adaptive (CAT) versions of the three Graduate Record Examinations (GRE) General Test measures. A verbal CAT was taken by 1,507, a quantitative CAT by 1,354, and an analytical CAT by 995…

Test developers continue to struggle with the technical and logistical problems inherent in assessing achievement across different languages. Many testing programs offer separate language versions of a test to evaluate the achievement of examinees in different language groups. However, comparison of individuals who took different language versions…

When an examination consists in whole or part of constructed response test items, it is common practice to allow the examinee to choose a subset of the constructed response questions from a larger pool. It is sometimes argued that, if choice were not allowed, the limitations on domain coverage forced by the small number of items might unfairly…

The purpose of this study was to determine which equating models are appropriate for the Alabama High School Graduation Examination (AHSGE) by equating two previously administered fall forms for each subject area of the AHSGE and determining whether differences exist in the test score distributions or passing scores resulting from the equating…

The advent of computers in educational and psychological measurement has lead to the need for algorithms for optimal assembly of tests from item banks. This paper reviews the literature on optimal test assembly and introduces the contributions to this report on the topic. Four different approaches to computerized test assembly are discussed:…

A modified regression procedure for finding a common metric for item response theory (IRT) parameter estimates and for total test scores is proposed and compared to the chi-square method of equating. The chi-square method was chosen after literature review for the currently used scaling procedures. The modification used with the regression…

This paper discusses the assumptions required by the item response theory (IRT) true-score equating method (with Stocking & Lord, 1983; scaling approach), which is commonly used in the nonequivalent groups with an anchor data-collection design. More precisely, this paper investigates the assumptions made at each step by the IRT approach to…

In October 1981, the Graduate Record Examinations (GRE) Program introduced a new version of the General Test (GT) that differed from the previous version in three major ways. The GT was altered to: reduce the verbal measure's speededness and allow the addition of several quantitative items; delete two item types from the analytical measure; and…

The research leading to the decisions to revise the Graduate Record Examination Aptitude Test (GRE) (beginning in October 1981) is reviewed. The issues discussed include the format of the test (the timing of each section and the number of sections, the content of the sections--especially the analytical section), the scoring procedure for the GRE,…

Open testing and test disclosure legislation has fostered a renewed interest in the problems and issues associated with test equating. This paper focuses on a specific application of test equating, that of professional and occupational testing for certification and licensure. Test equating is the process of making test scores from different forms…

An alternative is proposed to the legally required uniform procedures and criteria for reporting the results of Elementary Secondary Education Act Title I projects. It is suggested that Congress is unlikely to respond to nationwide impact data on the effectiveness of Title I even when such data are available. It is further proposed that the RMC…

As a result of using the Rasch model in building item banks in reading and mathematics, we were able to design two new testing programs for our district. The first was a grade level competency-based evaluation program designed to identify our district's low achievement-level students as well as the precompetency goals on which they needed help.…

Rasch model statistics and domain-referenced test development methods were used in the Virginia Basic Learning Skills (BLS) testing programs and the Graduation Competencies testing program in order to increase the fairness and effectiveness of those programs. Mastery of each of the 83 objectives in reading and mathematics for grades one through…

PRIME is a library of 44 batch-oriented computer routines: 20 major package programs, which use 12 statistical utility routines, and 12 other utility routines for input/output and data manipulation. This manual contains a general description of data preparation and coding, standard control cards, input deck arrangement, standard options, and…

The Rasch model was used to compare the grade equivalent scores which students of equal ability would have obtained, had they taken different levels of the Iowa Tests of Basic Skills (ITBS). Raw scores on items common to levels 9-14 of the ITBS (vocabulary subtest) were obtained from 6,000 students in Dallas, Texas. Rasch scales were used to…