The extent to which log-linear smoothing could improve the accuracy of common-item equating by the chained equipercentile method in small samples of examinees was investigated with responses from a 100-item test and 93,283 examinees. Smoothing reduced the sample size required for a given degree of accuracy. (SLD)

A conceptual framework and heuristic model for considering the existence, magnitude, and consequences of context effects are presented through an extension of some generalizability theory concepts. Context effects are often misunderstood, and current measurement models have serious limitations for examining them. Their importance needs to be…

Criteria used in the equating process proposed in the literature are reviewed. The discussion begins by examining how equating is defined. The controversy over the best criterion, the utility of some, and whether a criterion is needed at all means that much work needs to be done in this area. (SLD)

This report synthesizes the research to date addressing the construct comparability of the SAT Reasoning Test and prior SAT I: Reasoning Test and the series of research studies addressing the equatability and subpopulation invariance of the SAT and SAT I.

This study addresses the sample error and linking bias that occur with small and unrepresentative samples in a non-equivalent groups anchor test (NEAT) design. We propose a linking method called the "synthetic function," which is a weighted average of the identity function (the trivial equating function for forms that are known to be…

Traditionally, error in equating observed scores on two versions of a test is defined as the difference between the transformations that equate the quantiles of their distributions in the sample and population of test takers. But it is argued that if the goal of equating is to adjust the scores of test takers on one version of the test to make…

The Board of Registry (BOR) certifies medical technologists and other laboratory personnel. The BOR has studied adaptive testing for over 6 years and now administers all 17 BOR certification examinations using computerized adaptive testing (CAT). This paper presents an overview of the major research efforts from 1989 to the present related to test…

The use of two alternative item response theory (IRT) estimation models in the scaling and equating of the Test of English as a Foreign Language (TOEFL) was explored; and item scaling and test equating results based on these models were compared with results based on the three-parameter (3PL) model currently being used with the TOEFL. Models were…

An attempt was made to determine which item response theory (IRT) equating method results in the least amount of equating error or "scale drift" when equating scores across one or more test forms. An internal anchor test design was employed with five different test forms, each consisting of 30 items, 10 in common with the base test and 5…

Four multidimensional data configurations and one unidimensional data configuration were simulated for three differences in mean difficulty between two tests to be equated. Two chi-square statistics, Q1 and Q2, were examined for their ability to detect multidimensionality. Results indicated that Q1 did not discriminate between any of the…

The Anchor Test Study Manual was reviewed with the practitioner in mind. It represents an effort to equate and standardize eight commonly used elementary reading tests. Possibilities and limitations in using the manual are discussed. (BJG)

In an article published in the spring 2003 issue of "Harvard Educational Review", Roy Freedle stated that the SAT® is both culturally and statistically biased. Freedle proposed a solution to this bias, which involved using a half-test made up of the most difficult items culled from complete SAT examination. His claims, which garnered…

The results obtained from two basic types of pre-equatings of tests were compared: the item response theory (IRT) pre-equating and section pre-equating (SPE). The simulated data were generated from a modified three-parameter logistic model with a constant guessing parameter. Responses of two replication samples of 3000 examinees on two 72-item…

Uses of the variety of scores generated by standardized achievement tests are discussed. Desirable characteristics of scales, raw score scales, percent of correct items, percentile ranks, grade equivalents, normal curve equivalents, and scale scores are considered. The various meanings and purposes of each type of score are discussed. It is…

Much previously published material for estimating the reliability of classification has been based on the assumption that a test consists of a known number of equally weighted items. The test score is the number of those items answered correctly. These methods cannot be used with classifications based on weighted composite scores, especially if…