Using Item Scores and Distractors in Person‐Fit Assessment

Abstract In order to detect a wide range of aberrant behaviors, it can be useful incorporate information beyond the dichotomous item scores. this paper, we extend and person‐fit statistics so that unusual behavior in scores distractors used as indicators aberrance. Through detailed simulations, show new are more powerful than existing detecting several types behavior, they able control Type I error rate instances where model does not exactly fit data. A real data example is also provided demonstrate utility an operational setting.