Datasets for semi-automated Rasch analysis with DIF

We aim to incorporate the differential item functioning (DIF) assessment into the procedure. For the implementation, we parameterize DIF effect sizes for each item and consider DIF items as split items based on the DIF-inducing covariates (test participants' background information).

This implementation leads to a new criterion called in-plus-out-of-questionnaire log likelihood with differential item functioning (IPOQ-LL-DIF). The datasets presented here are used in empirical studies. The simulated datasets are generated randomly and independently to illustrate the characteristics of the semi-automated method.