Design-Based Research to Enhance Collaboration between Learners and AI in a Middle School Writing Class

• In Event: AERA Roundtable Session Saturday 1:45 pm - Gold 4
  In Roundtable Session: AI, Writing, and Learner Agency in Educational Contexts

Abstract

As AI becomes more embedded in education, supporting students’ ability to collaborate with AI is increasingly vital. This study aims to develop and examine instructional design principles that enhance learner-AI collaboration competencies(LACC). Conducted in a Korean middle school within the context of a Korean language class, the study employed a design-based research approach in which researchers, a teacher, and students worked as a design team over 10 months to co-develop AI-integrated writing lessons. We evaluated changes in students’ AI competencies, writing efficacy, metacognition, and writing interest, alongside interviews with selected students. Finally, three key themes emerged as design principles. Findings show significant improvements in LACC and writing efficacy. These results offer guidance for designing effective AI-supported learning environments.

Authors

• Hyeeun Kim, Seoul National University

• Young Hoan Cho, Seoul National University

• Minju Kang, Seoul National University

• Yeono Son, Seoul National University