Investigating the Impact of Missing Data Handling Methods on the Detection of Differential Item Functioning

Abstract

In this study, it is aimed to investigate theimpact of different missing data handling methods on the detection ofDifferential Item Functioning methods (Mantel Haenszel and Standardizationmethods based on Classical Test Theory and Likelihood Ratio Test method basedon Item Response Theory). In this regard, on the data acquired from 1046candidates who entered to Foreign National Student Exam (FNSE) held in year2016 by Mersin University (MEU) and answered Basic Skills subtest, usingdifferent missing data handling methods, differential item functioning analyseswith Mantel Haenszel, Standardization and Likelihood Ratio Test methods areperformed. Basic Skills test consists of 80 multiple choice items. The itemsare all binary scored (1-0) items. Among the participants 523 are female and523 are male. The findings showed that the number of items flagged as DIF haschanged with the used missing data handling methods. The DIF detection methodsbased on Classical Test Theory are more consistent within themselves comparedto DIF detection method based on Item Response Theory, whereas the used missingdata handling methods differentiate the DIF detected items and this differencereaches a significant level for Mantel Haenszel method

Keywords

• differential item functioning dif
• test and item bias
• missing values; imputation of missing data
• mantel haenszel
• likelihood ratio test