Multiple Ways to Detect Differential Item Functioning in SAS

Differential item functioning (DIF), as an assessment tool, has been widely used in quantitative psychology, educational measurement, business management, and insurance and healthcare industries. The purpose of DIF analyses is to detect response differences of items in questionnaires, rating scales, or tests across different subgroups (e.g., gender), while controlling for ability level. There are three general procedures to examine DIF: generalized Mantel-Haenszel (MH), logistic regression, and item response theory (IRT). The SAS system provides flexible and efficient procedures for all these approaches. The goal of this paper is to demonstrate multiple ways to conduct DIF analyses by using different SAS procedures (PROC FREQ, LOGISTIC, GENMOD, GLIMMIX, and NLMIXED).