Differential Item Functioning

Abstract

Differential item functioning (DIF) occurs when individuals at the same level on the trait(s) or construct(s) being assessed but from different subpopulations have unequal probabilities of attaining a given score on the item. Methods for investigating DIF have been developed for both dichotomously and polytomously scored items. These methods may be classified by whether they condition on an unobserved or observed variable. Item response theory, logistic regression, Mantel–Haenszel (MH), and SIBTEST procedures for dichotomously scored responses and their extensions to polytomous responses are currently the most widely used methods for detecting DIF in practice. Current research interest focuses on sophisticated latent variable models that may facilitate a better understanding of the causes of DIF.