Title of article
Detecting Academic Field-based Differential Item Functioning and Differential Distractor Functioning in the Islamic Azad University EPT Employing the Rasch Model
Author/Authors
Jafaripour ، Sarallah English Department - Islamic Azad University, Najafabad Branch , Tabatabaei ، Omid English Department - Islamic Azad University, Najafabad Branch , Salehi ، Hadi English Department - Islamic Azad University, Najafabad Branch , Vahid Dastjerdi ، Hossein English Department - Islamic Azad University, Najafabad Branch
From page
740
To page
765
Abstract
This study aimed to explore Differential Item Functioning (DIF) and Differential Distractor Functioning (DDF) based upon academic fields in the Islamic Azad University English Proficiency Test (IAUEPT). Applying the Rasch model, DIF and DDF were subjected to analysis among PhD candidates belonging to different fields. The 1069 participants were broken down into Social and Human Disciplines (SHD) and Non-Social and Human Disciplines (N-SHD) groups. Findings suggested minimal academic field-related DIF, with only two out of 100 items displaying such effects. Likewise, academic field-specific DDF analysis highlighted disparities in the number of items: one in Choice A, three in Choice B, four in Choice C, and three in Choice D. The results have important implications for individuals involved in the development of high-stakes proficiency tests. Through identifying potential biases in such tests, it becomes possible to promote fairness and equality for all testees. Additionally, the identification of academic field-based DIF can provide insights into potential gaps in the curriculum, revealing areas where learners from different fields of study may face disadvantages.
Keywords
Differential Distractor Functioning (DDF) , Differential Item Functioning (DIF) , English Proficiency Test (EPT) , Rasch Model , Test Bias
Journal title
Research in English Language Pedagogy (RELP)
Journal title
Research in English Language Pedagogy (RELP)
Record number
2771412
Link To Document