Detection of Person Misfit in Computerized Adaptive Testing with Polytomous Items (CT-00-03)
by Rob R. Meijer and Edith M. L. A. van Krimpen-Stoop, University of Twente, Enschede, The Netherlands
Item scores that do not fit an assumed item response theory model may cause the latent trait value to be inaccurately estimated. For computerized adaptive tests (CAT), several person-fit statistics for detecting nonfitting item score patterns for dichotomously scored tests have been proposed. Both for paper-and-pencil (P&P) tests and CAT, the detection of person misfit with polytomous items is hardly explored. In this study, the theoretical and empirical distributions of a person-fit statistic for polytomous items are investigated both for P&P tests and CAT. Results showed that the distribution of this statistic was close to the standard normal distribution, for both P&P tests and CAT.