Investigating the Effect of Missing Data imputation Method and its Percentage on the Detection of Differential Item Functioning for Items of Test Using Raju's Method

Abstract

The aim of this study was to investigate the effect of the method of compensation for missing data and its percentage in the detection of the differential performance of the test paragraphs using the Rajo method according to the three-parameter response model using the generated data. The design of the study included different conditions based on the study variables. The differential performance in the test is 10%, 20% and 40%. The differential performance of the paragraph is: (low level of 0.5 and high level of 1.5). Accordingly, the study was designed from (6) different groups with different performance differentials and differential performance of the paragraph, as well as a reference group that does not include any differential performance clause, which is the basis of comparison, resulting in (7) sets Study.