Routledge Handbook of Digital Language Assessment

Ben Naismith is a Staff Assessment Scientist at Duolingo where he works on research and development of the Duolingo English Test.Alina A. von Davier is the Chief of Assessment at Duolingo. She has provided thought leadership to the Research and Development area for the Duolingo English Test for the past nine years.Jill Burstein is a Principal Assessment Scientist at Duolingo, where she leads validity, efficacy, and responsible AI research for the Duolingo English Test.Geoffrey T. LaFlair is a Principal Assessment Scientist at Duolingo where he leads Assessment Research and Development for the Duolingo English Test.

• List of figures iList of tables iiList of contributors iiiAcknowledgements ivForewordBurr Settles1. Introduction: Digital Language Assessment in the Age of AccelerationsBen NaismithPart I: Test Theory2. Paradigm Shift: A Dynamic Ecosystem for AI-Driven Language AssessmentJill Burstein, Alina A. von Davier3. The Empirical Void: Re-evaluating Trait-Based Tasks in Language TestingGeoffrey T. LaFlair, Xun Yan4. Opportunity to Test: Geographical Access as a Neglected Aspect of Test ValidityRamsey L. Cardwell, William C. M. Belzak5. Evolving Perspectives on Reading Assessment: Theory, Practice, and the Case of the Duolingo English TestXinyi Ma, Xun YanPart II: Test Development6. Compositional Automatic Item Generation: Developing Scalable, Valid, and Innovative AssessmentsYigal Attali, Andrew Runge, Yena Park, Sarah Goodwin, Geoffrey T. LaFlairThe Duolingo English Test Interactive Writing Task: A Conceptual and Evidence-based Innovation in Large-scale Writing Assessment. Mya Poe, Norbert Elliot7. Digital-first Content Development for Test FairnessYena Park, Jacqueline Church, Andrew Runge, Phoebe Mulcaire8. The DET Platform for Item eXperimentation, Iteration and Evaluation (PIXIE)Xiaowan Zhang, Andrew Runge, Yigal Attali, Yena Park, Geoffrey T. LaFlairPart III: Test Scoring9. Applications of Large Language Models in Automated Writing EvaluationNitin Madnani, Phoebe Mulcaire, Kai-Ling Lo10. Salient Pronunciation Features for L2 Oral Proficiency and Intelligibility: Implications for the DET’s Automated Scoring AlgorithmsMasha Kostromitina, Okim Kang11. Maintaining the Meaning of Test Scores in a High-Volume Evolving AssessmentSteven W. Nydick, Manqian Liao, Siyuan M. Chen, J.R. Lockwood12. Calibration, Scoring, and Administration of a High-stakes Computerized Adaptive TestJames Sharpnack, Steven W. Nydick, J.R. Lockwood, Alexander Tsigler, Alina A. von DavierPart IV: Test Infrastructure and Quality Assurance13. Software Architecture for a High-Stakes, On-demand, Digital-First Language AssessmentKevin Yancey14. Securing the DET: Principles and Practices for a Secure Remote High-Stakes TestMizra B. Baig, Chenhao Niu, Yong-Siang Shih, Liangyu Chen15. Measuring Accuracy in Remote Proctoring DecisionsWilliam C. M. Belzak, Yigal Attali16. Driving Innovation: Technical Program Management in AI-Enabled Digital Language AssessmentNeha Jain, Alina A. von DavierPart V: Test Impact17. Investigating Duolingo English Test Washback: A Focus on Test Preparation ActivitiesDaniel R. Isbell, Ann Tai Choe, Daniel Holden, Ayano Kawasaki, Jieun Kim, Yoonseo Kim, Maggie McGehee, Hitoshi Nishizawa, Leeseul Park, Alexander F. Tang18. Breaking Barriers: How the Duolingo English Test Expands Access to Global EducationEmma McLeavey-Weeder, Adam Shuck19. Views on Malian English Pedagogy and Assessment and the DET Language Assessment Tasks: Finding Alignments and Opportunities Through Focus-group ResearchKadidja Koné, Paula Winke20. Building Trust in a Digital Test Through Marketing, Messaging, and Stakeholder EngagementSophie Wodzak, Ramsey L. Cardwell, Masha KostromitinaAfterwordInnovations in Digital-First Adaptive Testing: The Duolingo English Test and the Future of Personalized AssessmentAlina A. von Davier, Stephen G. Sireci

'Ten years after the Duolingo English Test began reshaping access to English proficiency testing, this handbook captures both the ambition and the evidence behind that work. It shows how digital assessment can be rigorous, fair, scalable, and deeply human.'- Luis von Ahn, Co-Founder and CEO at Duolingo'A rigorous and timely contribution to digital assessment science. This handbook tackles AI-driven language testing where it matters: validity, fairness, security, and access. Grounded in evidence from the Duolingo English Test, it sets a new benchmark for responsible, scalable assessment design.'- Virginia Dignum, Professor of Responsible AI, Umeå University; author of The AI Paradox (Princeton University Press, 2026)