Responsible AI for Test Equity and Quality: The Duolingo English Test as a Case Study

TL;DR

This paper discusses how Responsible AI practices are essential for ensuring fairness and quality in AI-based assessments, using the Duolingo English Test as a detailed case study.

Contribution

It introduces RAI standards tailored for high-stakes language assessments and demonstrates their application in promoting test fairness and validity.

Findings

Development of RAI standards for the Duolingo English Test

Implementation of practices addressing fairness, privacy, and transparency

Enhanced test equity and quality through RAI principles

Abstract

Artificial intelligence (AI) creates opportunities for assessments, such as efficiencies for item generation and scoring of spoken and written responses. At the same time, it poses risks (such as bias in AI-generated item content). Responsible AI (RAI) practices aim to mitigate risks associated with AI. This chapter addresses the critical role of RAI practices in achieving test quality (appropriateness of test score inferences), and test equity (fairness to all test takers). To illustrate, the chapter presents a case study using the Duolingo English Test (DET), an AI-powered, high-stakes English language assessment. The chapter discusses the DET RAI standards, their development and their relationship to domain-agnostic RAI principles. Further, it provides examples of specific RAI practices, showing how these practices meaningfully address the ethical principles of validity and reliability, fairness, privacy and security, and transparency and accountability standards to ensure test equity and quality.