Learning Progression-Guided AI Evaluation of Scientific Models Integrating Writing and Drawing to Support Multi-Modal Knowledge-in-Use

• In Event: AI in Writing and Language Learning: Teachers and Learners

Abstract

Modeling is an essential practice for building a deep science understanding. Modeling assessments should measure the ability to integrate Disciplinary Core Ideas and Crosscutting Concepts, which reflects knowledge-in-use, and requires an open-ended format despite being expensive to score. We also should support learners in demonstrating knowledge-in-use via multiple modalities (e.g., written explanations, drawings). Machine learning has been utilized to score and provide feedback on open-ended Learning Progression (LP)-aligned models and explanations. However, the challenge of aligning multiple modalities for supporting knowledge-in-use in a developmentally appropriate way remains. We advance prior work on using AI to evaluate stand-alone models and explanations and leverage a multi-modal LP for training AI to evaluate LP-aligned assessments for delivering personalized multi-modal LP-aligned feedback.

Authors

• Leonora Kaldaras, Texas Tech University

• Tingting Li, Washington State University

• Kevin Haudek, Michigan State University

• Joseph S. Krajcik, Michigan State University