Detecting Differential Item Functioning: Item Response Theory Methods Versus the Mantel-Haenszel Procedure

This Monte Carlo study assessed Type I error in differential item functioning analyses using Lord’s chi-square (LC), Likelihood Ratio Test (LRT), and Mantel-Haenszel (MH) procedure. Two research interests were investigated: response theory (IRT) model specification LC the LRT continuity correction MH enhances literature by investigating correct incorrect model-data fit comparing those results to There three fixed factors (number of test items, IRT parameter estimation method, equating) four varied (IRT used generate data data, sample size, impact). The findings suggested procedure without is best based on rate.