Selecting Anchor Items in Differential Item Functioning: A Case Study

Common analyses of Differential Item Functioning (DIF) rely on a number of a priori assumptions about the data of interest. In a paper by Anna Doebler (2018) a new method of DIF detection was proposed which relies on fewer assumptions than the current popular methods. A case study of this method was done using data from the First International Mathematics Study (FIMS). This method was able to provide a more nuanced picture of DIF within the data without requiring assumptions to resolve the identification issue present in common models of DIF