A Bibliometric and Systematic Review of Differential Item Functioning (DIF) in Psychological and Educational Scales: Global Trends, Methods, and Evidence (2010–2023)

الملخص

This study presents a comprehensive bibliometric and systematic review of 34 peer-reviewed articles published between 2010 and 2023 that investigate differential item functioning (DIF) in psychometric and educational assessment instruments.  The study focuses on synthesizing the literature, identifying global trends, methodological approaches, and empirical evidence, with a particular emphasis on ensuring fairness and validity in assessment across diverse populations. The findings reveal a predominance of Item Response Theory (IRT) methods, particularly the Rasch, 2PL, and 3PL models, alongside logistic regression and Mantel-Haenszel procedures. While applied studies dominate the literature, there is a scarcity of theoretical frameworks and limited adoption of advanced techniques such as Bayesian methods and machine learning. The analysis highlights the surge in DIF research after 2020, driven by the digital transformation and the increasing need for cross-cultural testing. The practical implications emphasize the need to integrate DIF analysis into routine test validation procedures and the importance of culturally responsive assessment design. The study recommends future research focus on methodological diversity, cross-cultural validation, and the application of real-time adaptive DIF detection to advance the science and practice of fair assessment