Group differences have practical implications in analysing data from achievement tests or questionnaires. For example, whether two persons from different demographic groups, such as gender or race, with the same shopping preferences have different shopping habits on one aspect helps store managers better design their displays. Shopping habits and shopping preferences can be measured, respectively, by items and some latent factor in a questionnaire, and the different shopping habits observed on an item are called differential item functioning (DIF). In the current study, we develop a model that accounts for between-group differences, DIF, latent factors, and missing item response data simultaneously by expanding a one-group item response tree model into a multiple-group model. Different from most of the present DIF studies where one has to iteratively select anchor items and detect DIF items, we achieve DIF detection and parameter estimation simultaneously by properly reparameterizing model parameters and applying some spike-and-slab priors (Ishwaran and Rao 2005a; Rockova and George 2018) in Bayesian estimation. Simulation studies are conducted to illustrate the validation of the proposed estimation procedure and the efficiency of DIF detection. The proposed method is further applied to a real dataset for illustration.