We derive formulas for the differential item functioning (DIF) measures that two routinely used DIF statistics are designed to estimate. The DIF measures that match on observed scores are compared to DIF measures based on an unobserved ability (theta or true score) for items that are described by either the one-parameter logistic (1PL) or…

The Mantel-Haenszel delta difference (MH D-DIF) and the standardized proportion difference (STD P-DIF) are two observed-score methods that have been used to assess differential item functioning (DIF) at Educational Testing Service since the early 1990s. Latentvariable approaches to assessing measurement invariance at the item level have been…

Quantitative fairness procedures have been developed and modified by ETS staff over the past several decades. ETS has been a leader in fairness assessment, and its efforts are reviewed in this report. The first section deals with differential prediction and differential validity procedures that examine whether test scores predict a criterion, such…

The Mantel-Haenszel (MH) procedure (Mantel and Haenszel) is a popular method for estimating and testing a common two-factor association parameter in a 2 x 2 x K table. Holland and Holland and Thayer described how to use the procedure to detect differential item functioning (DIF) for tests with dichotomously scored items. Wang, Bradlow, Wainer, and…

In this study, the accuracies of four strategies were compared for estimating conditional differential item functioning (DIF), including raw data, logistic regression, log-linear models, and kernel smoothing. Real data simulations were used to evaluate the estimation strategies across six items, DIF and No DIF situations, and four sample size…

In this study, differential item functioning (DIF) methods utilizing 14 different matching variables were applied to assess DIF in the constructed-response (CR) items from 6 forms of 3 mixed-format tests. Results suggested that the methods might produce distinct patterns of DIF results for different tests and testing programs, in that the DIF…

This study uses historical data to explore the consistency of SAT® I: Reasoning Test score conversions and to examine trends in scaled score means. During the period from April 1995 to December 2003, both Verbal (V) and Math (M) means display substantial seasonality, and a slight increasing trend for both is observed. SAT Math means increase more…

In many practical settings, essentially the same differential item functioning (DIF) procedures have been in use since the late 1980s. Since then, examinee populations have become more heterogeneous, and tests have included more polytomously scored items. This paper summarizes and classifies new DIF methods and procedures that have appeared since…

The sample invariant properties of five anchor test equating methods are addressed. Equating results across two sampling conditions--representative sampling and new-form matched sampling--are compared for Tucker and Levine equally reliable linear equating, item response theory true-score equating, and two equipercentile methods. (SLD)

The application of the Mantel-Haenszel test statistic (and other popular DIF-detection methods) to determine DIF requires large samples, but test administrators often need to detect DIF with small samples. There is no universally agreed upon statistical approach for performing DIF analysis with small samples; hence there is substantial scope of…

A procedure for checking the score equivalence of nearly identical editions of a test is described and illustrated with Scholastic Aptitude Test data. The procedure uses the standard error of equating and uses graphical representation of score conversion deviations from the identity function in standard error units. (SLD)

This paper describes an approach to item analysis that is based on the estimation of a set of response curves for each item. The response curves show, at a glance, the difficulty and the discriminating power of the item and the popularity of each distractor, at any level of the criterion variable (e.g., total score). The curves are estimated by…

Statistical procedures for detecting differential item functioning (DIF) are often used as an initial step to screen items for construct irrelevant variance. This research applies a DIF dissection method and a two-way classification scheme to SAT Reasoning Test™ verbal section data and explores the effects of deleting sizable DIF items on reported…

The equating methods and sampling designs used in the empirical studies in this special issue on the use of matched samples for test equating are described. Four requisites for equating are listed, and the invariance of equating functions requisite is identified as the focus of this issue. (SLD)

A new approach to assessing unexpected differential item performance (item bias or item fairness) is introduced and applied to the item responses of different subpopulations of Scholastic Aptitude Test (SAT) takers. The essential features of the standardization approach are described. The primary goal of the standardization approach is to control…