Growing reliance on complex constructed response items has generated considerable interest in automated scoring solutions. Many of these solutions are described in the literature; however, relatively few studies have been published that "compare" automated scoring strategies. Here, comparisons are made among five strategies for…

An essential concern in the application of any equating procedure is determining whether tests can be considered equated after the tests have been placed onto a common scale. This article clarifies one equating criterion, the first-order equity property of equating, and develops a new method for evaluating equating that is linked to this…

Prior research indicates that the overall reliability of performance ratings can be improved by using ordinary least squares (OLS) regression to adjust for rater effects. The present investigation extends previous work by evaluating the impact of OLS adjustment on standard errors of measurement ("SEM") at specific score levels. In…

The effects of test item disclosure on resulting examinee equated scores and population passing rates were studied for 5,000 examinees taking a professional licensing examination. Results suggest that the effects of disclosing depended on the nature of the released items. Specific effects on particular examinees are also discussed. (SLD)

Evaluated the equivalence of scores and one-parameter logistic model item difficulty estimates obtained from computer-based and paper-and-pencil forms of a licensure examination taken by 418 examinees. There was no effect of either order or mode of administration on the equivalences. (SLD)

The example of a medical licensure test is used to demonstrate situations in which complex, integrated content must be balanced at the total test level for validity reasons, but items assigned to reportable subscore categories may be used under a multidimensional item response theory adaptive paradigm to improve subscore reliability. (SLD)

A procedure is introduced for determining the effect of testlet nonequivalence on operating characteristics of a testlet-based computerized mastery test (CMT). The procedure, which involves estimating the CMT decision rule twice with testlet likelihoods treated as equivalent or nonequivalent, is demonstrated with testlet pools from the Architect…

Two methods--non-distributional and normal--are derived for estimating measures of pass-fail reliability. Both are based on the Spearman Brown formula and require only a single test administration. Results from a simulation (n=20,000 examinees) and a licensure examination (n=4,828 examinees) illustrate these methods. (SLD)

Effects of numbers of experts (NOEs) and common items (CIs) on the scaling of cutting scores from expert judgments were studied for 11,917 physicians taking 2 forms of a medical specialty examination. Increasing NOEs and CIs reduced error; beyond 5 experts and 25 CIs, error differences were small. (SLD)

An expanded version of the method of content evaluation proposed by S. G. Sireci and K. F. Giesinger (1992) was evaluated with respect to a national licensure examination and a nationally standardized social studies achievement test. Two groups of 15 subject-matter experts rated the similarity and content relevance of the items. (SLD)