Differential Item Functioning of Minnesota Multiphasic Personality Inventory-2 (Depression) Items in among Recruitment Candidates: A Comparison Between the Rasch Model and Mantel-Haenszel

Rahmani Malek Abad, Mahdi; Falsafinejad, Mohammad Reza; Minaei, Asghar; Farrokhi, Noor Ali

doi:10.22059/japr.2024.363263.644703

Differential Item Functioning of Minnesota Multiphasic Personality Inventory-2 (Depression) Items in among Recruitment Candidates: A Comparison Between the Rasch Model and Mantel-Haenszel

Articles in Press

Document Type : Research Paper

Authors

Mahdi Rahmani Malek Abad ¹

Mohammad Reza Falsafinejad ²

Asghar Minaei ³

Noor Ali Farrokhi ⁴

¹ Department of Educational Measurement, Faculty of Psychology and Educational Sciences, Allameh Tabataba’i University,Tehran, Iran.

² Department of Educational Measurement, Faculty of Psychology and Educational Sciences, Allameh Tabataba’i University, Tehran , Iran.

³ Department of Educational Measurement, Faculty of Psychology and Educational Sciences, Allameh Tabataba’i University, Tehran , Iran

⁴ 4. Department of Educational Measurement, Faculty of Psychology and Educational Sciences, Allameh Tabataba’i University, Tehran , Iran.

10.22059/japr.2024.363263.644703

Abstract

The purpose of the present study is to identify the questions of the depression component of the Minnesota Multifaceted Personality Questionnaire-2, which are biased toward a certain group among job candidates based on gender and age. This research is survey research, and the statistical population was all the Recruitment candidates of Isfahan city in 2022 and spring 2023. The sample size was 5997 people (57.31% men and 42.69% women) who were selected by one-step cluster method. The tool used was MMPI-2 questionnaire with 370 questions. The fit of the Items with the model was confirmed using outfit MnSq and infit MnSq, and DIF according to gender and age was checked and compared with Mantel-Haenszel and Rasch model by Winsteps version 3.64. After confirming the fit of the data with the model, Differential Item Functioning was performed. The results showed that more than half of the items (64.9%) were biased against women and young people. Mantel-Haenszel also provided a similar finding to Rasch's model, except for the difference in two items. Using depression component questions at once can cause non-bias, However, selective questions or the use of Harris-Lingoes subscales will cause severe DIF to be heavily biased towards women and young people. Also, the Rasch model has a relative advantage over Mantel-Haenszel in detecting DIF.

Keywords

Depression

Differential Item Functioning

Mantel-Haensze Method

Rasch Model

Recruitment candidates