The appropriateness of using Steps 1 and 2 of the United States Medical Licensing Examination (USMLE) to make individual academic decisions and the appropriateness of determining course grades with scores from the National Board of Medical Examiners (NBME) subject tests are examined. Underlying philosophical and psychometric assumptions are…

It is proposed that, despite the fact that no large body of data from the United States Medical Licensing Examination (USMLE) is available, predictive validity of National Board of Medical Examiners test scores supports use of USMLE scores for screening potential new medical residents. (Author/MSE)

A specific standardized test of language proficiency, the Grammatical Analysis of Elicited Language (GAEL), for hearing-impaired children was used to examine language usage between the tester and the child. Results from a two-part test (imitated and prompted) show that communicative competence needs to be incorporated into the test as part of…

Development of an advanced placement examination and certification program that guarantees admission to foreign universities is discussed as a means of promoting international educational exchange and assuring comparability of academic standards. Use of American advanced placement credentials is already common in some foreign countries. (MSE)

Compares commercial and locally developed assessment instruments in terms of assessment purposes, match between instrument and local programs, logistics, institutional acceptance, quality, and student motivation. Recommends identifying information needs, considering costs, designing a measurement plan, reviewing relevant literature, involving…

Many state optometric licensing boards require a scaled or percentage score of 75 on each section. A study of multiple criteria and pass-fail rates suggests that high failure rates are a result not of candidate ability but of arbitrary classification of test content. State boards are encouraged to resolve these problems. (MSE)

This paper provides the basis for understanding score equating through item response theory (IRT). Theoretical justifications and practical advantages of IRT true-score test procedures are discussed. Three steps in the equating process are specified, and a self-test is included. (SLD)

An Australian study investigated multidimensionality in college entrance examination scores. Data from the American College Testing Program (ACT) and Australian Tertiary Entrance scores (which combines year 12 course grades and scores on the Australian Scholastic Aptitude Test) were analyzed using a 4-dimensional model. Results suggest a single…

Discusses gender differences apparent in British public examination entry patterns and outcomes. Questions whether examinations should be designed for equal outcomes and the assumption that assessment techniques may differentially affect performances of males and females. Argues that because other factors such as different experiences and…

The second approach used by L. V. Hedges and L. Friedman (1993) in their reanalysis of Feingold's earlier results is conceptually equivalent and yields results consistent with the original interpretation. The first method yields results that are discrepant. Analyses of variations in tail effect sizes are planned. (SLD)

Feingold's reply illustrates that his steps in characterizing tail effect sizes are not the calculations the authors had imagined. Attempting to reproduce Feingold's calculations, the authors still often find themselves in disagreement with interpretations Feingold has placed in his table. (SLD)

National testing does not produce better education in other countries or spur students to do better work, and it creates barriers for poor and minority students. The paper recommends creating programs to help states and districts develop new ways to evaluate student learning as part of overhauling education. (SM)

The controversial oral English Bagrut examination is seen as a means to prepare students for their written Bagrut exam, specifically to encourage the active use of the English language. (LB)

The emergence of computerized adaptive tests (CATs), which customize assessment for each individual, may mean more efficient testing. A CAT version of the Graduate Record Examination is in production. Critics have many questions about the technique, which is based on item response theory. (MSE)

A survey of 12 constituencies (totaling 2,288 respondents) concerning the National Board of Medical Examiners' (NBME) comprehensive Part I and Part II examinations revealed support for more even distribution of fail rates across examinations, support for a fixed standard, and favorable reaction to setting standards based on review of examination…