The effect of person misfit on classification decisions

The effect of person misfit to an item response theory (IRT) model on a mastery/nonmastery decision was investigated. Also investigated was whether the classification precision can be improved by identifying misfitting respondents using person-fit statistics. A simulation study was conducted to investigate the probability of a correct classification using different estimation methods, person-fit statistics, model violations, test lengths, and sample sizes. In this simulation study, the effect of the presence of misfitting items score patterns on the item parameter estimates was also taken into account. Results show that the effect of the presence of misfitting item score patterns on the classification of nonaberrant simulees was in general small; that is, the classification precision for these simulees hardly suffered. Further, for simulees classified as nonaberrant using a person-fit statistic, the classification decisions were comparable with a priori known nonaberrant simulees. The conclusion is that person-fit statistics can be used for identifying a subsample of respondents where relatively precise mastery/nonmastery decisions can be made. These results were comparable across different person-fit statistics and estimation methods