ABSTRACT
This chapter considers a questionnaire on depression in which respondents are asked to indicate whether they experienced certain emotions or performed certain behaviors. It discusses basic models from item response theory (IRT) to investigate differential item functioning (DIF) among different groups taking the test. The investigation of DIF is concerned with the investigation of measurement invariance in a rather analytical way. The chapter discusses IRT models where the relationship between a positive response on the item and the latent variable is strictly monotonously increasing. Characteristic for IRT models is that they model the relationship between the latent variable and the response behavior at the level of the item responses themselves, and not at the level of the aggregated test score. Various specialized software packages exist for estimating IRT models and doing DIF research. The chapter discusses how various item response theory techniques can be applied to detect DIF.
