Speaker
Description
As English language assessment evolves in an increasingly complex and technology-driven landscape, test developers are facing a new generation of challenges—from AI-assisted cheating and deepfakes to designing tasks that are both valid and reflective of real-world communication. At the same time, there is growing pressure to ensure that tests remain fair, reliable, and efficient across diverse learner populations worldwide.
This session explores how these challenges are addressed through innovations in the design of the Enhanced TOEFL iBT (2026). Key areas include the use of multistage adaptive testing to improve measurement precision and test efficiency, principled approaches to item and task design to ensure validity and relevance, and the implementation of advanced at-home security measures, including multi-layered identity verification and live human proctors, supported by advanced AI tools, to ensure both security and test-taker experience.
The session will also discuss how these innovations support fairness across different learner backgrounds while maintaining the integrity of large-scale English assessment. Attendees will gain practical insights into how modern test design balances security, efficiency, and meaningful measurement in today’s global education environment.
