In many testing programs it is assumed that the context or position in which an item is administered does not have a differential effect on examinee responses to the item. Violations of this assumption may bias item response theory estimates of item and person parameters. This study examines the potentially biasing effects of item position. A…

The purpose of this study was to examine the performance of differential item functioning (DIF) assessment in the presence of a multilevel structure that often underlies data from large-scale testing programs. Analyses were conducted using logistic regression (LR), a popular, flexible, and effective tool for DIF detection. Data were simulated…

Although response times on test items are recorded on a natural scale, the scale for some of the parameters in the lognormal response-time model (van der Linden, 2006) is not fixed. As a result, when the model is used to periodically calibrate new items in a testing program, the parameter are not automatically mapped onto a common scale. Several…

This study examined the extent to which log-linear smoothing could improve the accuracy of differential item functioning (DIF) estimates in small samples of examinees. Examinee responses from a certification test were analyzed using White examinees in the reference group and African American examinees in the focal group. Using a simulation…

As an alternative to adaptation, tests may also be developed simultaneously in multiple languages. Although the items on such tests could vary substantially, scores from these tests may be used to make the same types of decisions about different groups of examinees. The ability to make such decisions is contingent upon setting performance…

This paper presents the characteristics, properties and development of the National Reference Scale for reading. This new scale is the result of a reanalysis of the Anchor Test Study data using Rasch model procedures, in an effort to produce equated scores among all reading tests included in that study. (Author/JKS)

Several techniques exist to automatically put together a test meeting a number of specifications. In an item bank, the items are stored with their characteristics. A test is constructed by selecting a set of items that fulfills the specifications set by the test assembler. Test assembly problems are often formulated in terms of a model consisting…

Investigated how well the generalized and Rasch models described item and test performance across a broad range of mixed-item-format test configurations (six tests from two state proficiency testing programs). Evaluating the impact of model assumptions on the predictions of item and test information permitted a delineation of the implications of…

Use of the Rasch model for vertical equating of tests is discussed. Although use of the model is promising, empirical results raise questions about the adequacy of the Rasch model. Latent trait models with more parameters may be necessary. (JKS)

One method to correct for selection bias in comparing Scholastic Aptitude Test (SAT) scores among states is presented, which is a modification of J. J. Heckman's Selection Bias Correction (1976, 1979). Empirical results suggest that sample selection bias is present in SAT score regressions. (SLD)

This book is a collection of nine commissioned papers that consider the educational reform represented by performance assessments, their advantages and disadvantages, and the challenges of successful implementation. Both theoretical and practical issues are addressed. (SLD)

"Assessing Student Performance" sets forth arguments for looking at testing and assessment. The exploration is more about morality and epistemology than about technology and politics. A principal assumption is that mainstream assessment and testing philosophy is flawed. (SLD)

Studied the stability of rater severity over an extended rating period by applying multifaceted Rasch analysis to ratings of 16 raters of writing performances of 8,285 elementary school students. Findings cast doubt on the practice of using a single calibration of rate severity as the basis for adjustment of person measures. (SLD)

Created a developmental scale for the North Carolina End-of-Grade Mathematics Tests using a subset of identical test forms administered to adjacent grade levels with Thurstone scaling and Item Response Theory methods. Discusses differences in patterns produced. (Author/SLD)

The five chapters of this report provide resources that deal with the validity, generalizability, comparability, performance standards, and fairness, equity, and bias of performance assessments. The book is written for experienced educational measurement practitioners, although an extensive familiarity with performance assessment is not required.…