Tailor feedback strategies by question type: discourage MCQ trial-and-error and scaffold open-ended revision¶

Strategy · All strategies
Evidence · no claims cited

Description¶

Based on the divergent engagement patterns by question type, the authors recommend differentiating feedback design: "For MCQs, systems should discourage superficial trial-and-error behaviors and focus on guiding conceptual understanding. For OEQs, structured, targeted scaffolding can foster deeper engagement and higher-quality revisions." They also suggest learning analytics consider multimodal indicators such as revision quality to capture meaningful engagement.

Design Implications¶

Context¶

Requirements¶

• Feedback generation that can condition on question type and on the learner's specific answer

Constraints¶

• Evidence comes from a single-session crowdsourced study with one topic; the OEQ submission trend was not statistically significant

Target Learners¶

• online learners completing practice activities with MCQs and open-ended questions

Target Learning Goals¶

• conceptual understanding in multiple-choice practice and higher-quality revision of open-ended responses

Related Strategies¶

• Discourage mental offloading to AI by balancing guidance depth and frequency with application opportunities
• Integrate AI-driven scaffolding with teacher-led offloading and refine feedback granularity to prevent over-scaffolding

Examples¶

-

Key Sources¶

• Chloe Qianhui Zhao, Jie Cao, Jionghao Lin, and Kenneth R. Koedinger. (2026). LLM-based Multimodal Feedback Produces Equivalent Learning and Better Student Perceptions than Educator Feedback. LAK 2026. https://arxiv.org/abs/2601.15280