Westat Researchers Discuss Human–AI Collaboration in Assessment and Education at 2026 AIME-Con
September 22, 2026
Westat staff will present at the National Council on Measurement in Education (NCME)’s Artificial Intelligence in Measurement and Education Conference (AIME-Con), held October 5 to 7 in Pittsburgh, Pennsylvania. The conference brings together experts in educational measurement, psychometrics, artificial intelligence (AI), natural language processing (NLP), and learning analytics to explore how AI is transforming assessment and education. Conference attendees will share insights and stay at the forefront of emerging research and best practices in AI-supported measurement, assessment, and learning.
Tuesday, October 6, 3:45–5:15 pm (ET)
Session: Transforming Assessment Data into Educational Action: Human–AI Collaboration in Educational Measurement
Westat experts are presenting the following papers at this session, which was organized by moderator Judy H. Tang.
Human–AI Collaboration in Educational Measurement: Transforming Assessment Data into Educational Action
Laura C. Egan and Judy H. Tang
The presenters will discuss a framework for the responsible use of AI tools in educational measurement, showing how AI tools can support human expertise to strengthen assessment systems and maintain reliability, validity, fairness, transparency, and accountability. The framework organizes AI applications into four key functions: evidence generation, evidence processing, evidence interpretation, and educational action. Drawing on examples from large-scale assessment programs, the presenters will illustrate how AI tools can amplify data analysis, scoring, reporting, and personalized feedback while preserving meaningful human oversight. The framework provides researchers and practitioners with practical guidance for implementing trustworthy, human-centered approaches to human–AI collaboration and safeguarding the quality and integrity of educational measurement.
Human–Machine Learning for Large-Scale Qualitative Coding in Educational Measurement
Judy H. Tang, Tom Krenzke, Jin Hui Xu, and Karen Lo
The presenters will discuss research that explores how machine learning (ML) models can support human expertise to improve the efficiency and accuracy of large-scale educational data coding. In this study, researchers used a nationally representative sample of high school transcript data. They applied AI-supported recommendations based on previously coded course data to assist human coders in assigning standardized course classifications. This human–ML framework streamlined the coding process while maintaining expert oversight to support coding quality, consistency, and reliability.