Detecting DIF in 2PL Multistage Assessments

The detection of differential item functioning is crucial for the psychometric evaluation of multistage tests. This paper discusses five approaches presented in the literature: Logistic regression, SIBTEST, analytical score-based tests, Bootstrap score-based tests and permutation score-based tests. We further compare these approaches with respect to their Type I error rate and their power in a simulation study that is inspired by educational large-scale assessments. We furter present an application to an empirical dataset. We find that all tests show a Type I error rate close to the nominal alpha level. All tests are shown to be sensitive against uniform and non-uniform DIF effect, with the score-based tests showing the highest power.