Comparing Robust Linking and Regularized Estimation for Linking Two Groups in the 1PL and 2PL Models in the Presence of Sparse Uniform Differential Item Functioning

In the social sciences, performance of two groups is frequently compared based on a cognitive test involving binary items. Item response models are often utilized for comparing groups. However, presence differential item functioning (DIF) can impact group comparisons. order to avoid biased estimation groups, appropriate statistical methods handling required. This article compares performance-regularized and several robust linking approaches in three simulation studies that address one-parameter logistic (1PL) two-parameter (2PL) models, respectively. It turned out at least as effective regularized approach most conditions studies.