Search

While score comparability across test forms typically relies on common (or randomly equivalent) examinees or items, innovations in item formats, test delivery, and efforts to extend the range of score interpretation may require a special data collection before examinees or items can be used in this way--or may be incompatible with common examinee or item designs altogether. When comparisons are necessary under these nonroutine conditions, forms still must be connected by "something" and this article focuses on these form-invariant connective "somethings." A conceptual framework for thinking about the problem of score comparability in this way is given followed by a description of three classes of connectives. Examples from the history of innovations in testing are given for each class.

An Angoff standard setting study generally yields judgments on a number of items by a number of judges (who may or may not be nested in panels). Variability associated with judges (and possibly panels) contributes error to the resulting cut score. The variability associated with items plays a more complicated role. To the extent that the mean item judgments directly reflect empirical item difficulties, the variability in Angoff judgments over items would not add error to the cut score, but to the extent that the mean item judgments do not correspond to the empirical item difficulties, variability in mean judgments over items would add error to the cut score. In this article, we present two generalizability-theory-based analyses of the proportion of the item variance that contributes to error in the cut score. For one approach, variance components are estimated on the probability (or proportion-correct) scale of the Angoff judgments, and for the other, the judgments are transferred to the theta scale of an item response theory model before estimating the variance components. The two analyses yield somewhat different results but both indicate that it is not appropriate to simply ignore the item variance component in estimating the error variance.

Eye-tracking technology can create a record of the location and duration of visual fixations as a test-taker reads test questions. Although the cognitive process the test-taker is using cannot be directly observed, eye-tracking data can support inferences about these unobserved cognitive processes. This type of information has the potential to support improved test design and to contribute to an overall validity argument for the inferences and uses made based on test scores. Although several authors have referred to the potential usefulness of eye-tracking data, there are relatively few published studies that provide examples of that use. In this paper, we report the results an eye-tracking study designed to evaluate how the presence of the options in multiple-choice questions impacts the way medical students responded to questions designed to evaluate clinical reasoning. Examples of the types of data that can be extracted are presented. We then discuss the implications of these results for evaluating the validity of inferences made based on the type of items used in this study.

In this article, it is shown how item text can be represented by (a) 113 features quantifying the text's linguistic characteristics, (b) 16 measures of the extent to which an information-retrieval-based automatic question-answering system finds an item challenging, and (c) through dense word representations (word embeddings). Using a random forests algorithm, these data then are used to train a prediction model for item response times and predicted response times then are used to assemble test forms. Using empirical data from the United States Medical Licensing Examination, we show that timing demands are more consistent across these specially assembled forms than across forms comprising randomly-selected items. Because an exam's timing conditions affect examinee performance, this result has implications for exam fairness whenever examinees are compared with each other or against a common standard.

Test administrators are appropriately concerned about the potential for time constraints to impact the validity of score interpretations; psychometric efforts to evaluate the impact of speededness date back more than half a century. The widespread move to computerized test delivery has led to the development of new approaches to evaluating how examinees use testing time and to new metrics designed to provide evidence about the extent to which time limits impact performance. Much of the existing research is based on these types of observational metrics; relatively few studies use randomized experiments to evaluate the impact time limits on scores. Of those studies that do report on randomized experiments, none directly compare the experimental results to evidence from observational metrics to evaluate the extent to which these metrics are able to sensitively identify conditions in which time constraints actually impact scores. The present study provides such evidence based on data from a medical licensing examination. The results indicate that these observational metrics are useful but provide an imprecise evaluation of the impact of time constraints on test performance.

Validating performance standards is challenging and complex. Because of the difficulties associated with collecting evidence related to external criteria, validity arguments rely heavily on evidence related to internal criteria--especially evidence that expert judgments are internally consistent. Given its importance, it is somewhat surprising that evidence of this kind has rarely been published in the context of the widely used bookmark standard-setting procedure. In this article we examined the effect of ordered item booklet difficulty on content experts' bookmark judgments. If panelists make internally consistent judgments, their resultant cut scores should be unaffected by the difficulty of their respective booklets. This internal consistency was not observed: the results suggest that substantial systematic differences in the resultant cut scores can arise when the difficulty of the ordered item booklets varies. These findings raise questions about the ability of content experts to make the judgments required by the bookmark procedure.

Evidence of stable standard setting results over panels or occasions is an important part of the validity argument for an established cut score. Unfortunately, due to the high cost of convening multiple panels of content experts, standards often are based on the recommendation from a single panel of judges. This approach implicitly assumes that the variability across panels will be modest, but little evidence is available to support this assertion. This article examines the stability of Angoff standard setting results across panels. Data were collected for six independent standard setting exercises, with three panels participating in each exercise. The results show that although in some cases the panel effect is negligible, for four of the six data sets the panel facet represented a large portion of the overall error variance. Ignoring the often hidden panel/occasion facet can result in artificially optimistic estimates of the cut score stability. Results based on a single panel should not be viewed as a reasonable estimate of the results that would be found over multiple panels. Instead, the variability seen in a single panel can best be viewed as a lower bound of the expected variability when the exercise is replicated.

The present study examined the long-term usefulness of estimated parameters used to adjust the scores from a performance assessment to account for differences in rater stringency. Ratings from four components of the USMLE[R] Step 2 Clinical Skills Examination data were analyzed. A generalizability-theory framework was used to examine the extent to which rater-related sources of error could be eliminated through statistical adjustment. Particular attention was given to the stability of these estimated parameters over time. The results suggest that rater stringency estimates obtained at a point in time and then used to adjust ratings over a period of months may substantially decrease in usefulness. In some cases, over several months, the use of these adjustments may become counterproductive. Additionally, it is hypothesized that the rate of deterioration in the usefulness of estimated parameters may be a function of the characteristics of the scale.

Although the Angoff procedure is among the most widely used standard setting procedures for tests comprising multiple-choice items, research has shown that subject matter experts have considerable difficulty accurately making the required judgments in the absence of examinee performance data. Some authors have viewed the need to provide performance data as a fatal flaw for the procedure; others have considered it appropriate for experts to integrate performance data into their judgments but have been concerned that experts may rely too heavily on the data. There have, however, been relatively few studies examining how experts use the data. This article reports on two studies that examine how experts modify their judgments after reviewing data. In both studies, data for some items were accurate and data for other items had been manipulated. Judges in both studies substantially modified their judgments whether the data were accurate or not.

Although multivariate generalizability theory was developed more than 30 years ago, little published research utilizing this framework exists and most of what does exist examines tests built from tables of specifications. In this context, it is assumed that the universe scores from levels of the fixed multivariate facet will be correlated, but the error terms will be uncorrelated because subscores result from mutually exclusive sets of test items. This paper reports on an application in which multiple subscores are derived from each task completed by the examinee. In this context, both universe scores and errors may be correlated across levels of the fixed multivariate facet. The data described come from the United States Medical Licensing Examination[R] Step 2 Clinical Skills Examination. In this test, each examinee interacts with a series of standardized patients and each interaction results in four component scores. The paper focuses on the application of multivariate generalizability theory in this context and on the practical interpretation of the resulting estimated variance and covariance components.

Used generalizability theory to assess the impact of using independent, randomly equivalent groups of experts to develop scoring algorithms for computer simulation tasks designed to measure physicians' patient management skills. Results with three groups of four medical school faculty members each suggest that the impact of the expert group may be significant, but that it can be controlled with appropriate test development strategies. (SLD)

Two studies focused on aspects of the rating process in performance assessment. The first, which involved 15 raters and about 400 medical students, made the "committee" facet of raters working in groups explicit, and the second, which involved about 200 medical students and four raters, made the "rating-occasion" facet explicit. (SLD)

Research on automated scoring is extended by comparing alternative automated systems for scoring a computer simulation of physicians' patient management skills. A regression-based system is more highly correlated with experts' evaluations than a system that uses complex rules to map performances into score levels, but both approaches are feasible. (SLD)

Compared the results of differential item functioning (DIF) analysis with matching based on the total test score, matching based on subtest scores, or multivariate matching using multiple subtest scores. Results using 2,000 responses from medical students suggest that matching on multiple subtest scores may be superior to the other methods. (SLD)