DIF And Its Detection Methods For Polytomously Scored Items Under The Framework Of Cognitive Diagnosis

On the stage of psychological and educational measurement,the polytomous item has always played a very important role.Compared to the dichotomous item,the polytomous item could provide the investigator with more information.Therefore,the fairness of polytomous item has also been widely concerned.However,in the context of cognitive diagnosis,research on item functional differences is currently based on dichotomous items.Therefore,the main purpose of this study is to introduce several common polytomous DIF detection procedures into polytomous cognitive diagnostic tests to fill the gaps in this field.This study is divided into three parts: study 1 combined the DIF modeling for polytomous items and the characteristics of cognitive diagnosis to define the DIF in the field of polytomous cognitive diagnosis and extended four DIF detection procedures to polytomous cognitive diagnosis.Study 2 explored the performance of several polytomous DIF detection procedures in indetecting the uniform DIF under the estimation of the reduced models through simulation experiments.The selected procedures in this study include the mantel test,the logistic discriminant function analysis(LDFA),the likelihood ratio test(LR test),and Wald test.The third study explored the performance of selected methods under the estimation of the saturation model seq-GDINA through simulation experiments.In addition to the differences in the estimation model,some independent variables also included in the simulation experiment like dif size,sample size,matching variables,and the number of DIF items.The results are as follows:1)Whether in the reduced model or the saturated model,each DIF detection procedures all can effectively detect the uniform DIF in the polytomous cognitive diagnostic tests,and the performance of each procedure is minimally affected by the model.2)The LDFA method and the mantel test with KS as the matching variable have the highest statistical power in detecting the uniform DIF.3)Two model-based DIF tests: Wald test and LR test have better Type I error control than two non-model-based DIF tests: LDFA method and the mantel test method.And within all methods,Wald test has the best Type I error control.4)There are two matching variables for LDFA and the mantel test method.The mantel test and LDFA method with KS as the matching variable have higher power and lower type I error than the total score.5)Among the other experimental conditions,the amount of DIF has the greatest influence on the performance of DIF detection,followed by the size of the group,and finally the number of DIF items.From the method point of view,the mantel test method and LDFA method with total score as the matching variable are most affected by the changes of experimental conditions,while the other methods are less affected by the changes of experimental conditions.