Finding Words Associated with DIF: Predicting Differential Item Functioning using LLMs and Explainable AI

TL;DR

This study uses large language models and explainable AI to predict and interpret words linked to differential item functioning in assessments, aiding fairer test design.

Contribution

It introduces a novel approach combining LLMs and XAI to identify words associated with DIF, enhancing understanding and fairness in test items.

Findings

Many DIF-associated words reflect test blueprint sub-domains.

The approach can screen words during item writing for fairness.

It helps review and interpret traditional DIF analysis results.

Abstract

We fine-tuned and compared several encoder-based Transformer large language models (LLM) to predict differential item functioning (DIF) from the item text. We then applied explainable artificial intelligence (XAI) methods to these models to identify specific words associated with DIF. The data included 42,180 items designed for English language arts and mathematics summative state assessments among students in grades 3 to 11. Prediction $R^2$ ranged from .04 to .32 among eight focal and reference group pairs. Our findings suggest that many words associated with DIF reflect minor sub-domains included in the test blueprint by design, rather than construct-irrelevant item content that should be removed from assessments. This may explain why qualitative reviews of DIF items often yield confusing or inconclusive results. Our approach can be used to screen words associated with DIF during the item-writing process for immediate revision, or help review traditional DIF analysis results by highlighting key words in the text. Extensions of this research can enhance the fairness of assessment programs, especially those that lack resources to build high-quality items, and among smaller subpopulations where we do not have sufficient sample sizes for traditional DIF analyses.