Evaluating the Multiple-Group Mean and Covariance Structure Analysis model for the detection of Differential Item Functioning in polytomous ordered items

Résumé 0

"We use simulated data to evaluate the adequacy of the Multiple-Group (MG) extension of the ConfirmatoryFactor Analysis (CFA) with Mean and Covariance Structure (MACS) model for the detection ofd i fferential item functioning (DIF) in polytomous graded response items. Two factors are varied to generatethe DIF conditions: type of DIF (uniform and non-uniform) and size of DIF (low, medium andhigh). For each condition, 100 samples (N= 800) are generated according to the Graded Response model.Ten items with 5-response categories are considered, and only one item is manipulated to show DIF.The data are analyzed by means of LISREL 8. Results show that the DIF detection procedure analyzedreasonably maintained control of its Type I error under non-uniform DIF conditions, but showed somed i fficulties under uniform-DIF conditions. With respect to power, the procedure was satisfactory for detectingmedium and high uniform DIF, and lacked power for detecting non-uniform DIF."