In a thermodynamics course, pairs working with a tutorial dialogue agent learned about 1.24 standard deviations more than students working alone, with partner and agent support synergistic rather than redundant¶

Claim · All claims
Evidence · 1 study · review r2 · q2 · i3 large
Currency · newest study 2016 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q2 i3 The review reports that in the Kumar et al. study, working with a partner student and with support of an agent led to a 1.24 standard deviation improvement in learning over a control condition where students worked alone. → Carolyn Penstein Rosé & Oliver Ferschke 2016

Evidence¶

Carolyn Penstein Rosé & Oliver Ferschke 2016¶

Carolyn Penstein Rosé & Oliver Ferschke. (2016). Technology Support for Discussion Based Learning: From Computer Supported Collaborative Learning to the Future of Massive Open Online Courses. Int J Artif Intell Educ. https://doi.org/10.1007/s40593-016-0107-y

q2 · i3 · review · r2

The review reports this result from Kumar and colleagues' evaluation of tutorial dialogue agents as support for pairs working in a power plant design task in a sophomore thermodynamics course, with a printed effect of 1.24 standard deviations (d = 1.24) over working alone. Students with only a partner or only the agent learned approximately 1 standard deviation more than control.

"In that study, working with a partner student and with support of an agent led to a 1.24 standard deviation improvement in learning over a control condition where students worked alone. Students who worked either just with a partner or just with the computer agent learned approximately 1 standard deviation more than the control condition students."

Discussion¶

Related Claims¶

• Peer Discussion Improves Conceptual Understanding — related
• In math problem solving, a tutor-plus-peers LLM arrangement produced the strongest unassisted post-task test accuracy, with accuracy increasing monotonically from Control to Tutor + Peers — related
• Conversational-agent prompts improve student dialogue outcomes at both individual and dyad levels — related
• In math problem solving, a tutor LLM alone significantly improved unassisted test accuracy over no-agent control, while peer-like error agents alone did not reach significance — related
• Using generative AI as a dialogic partner for brainstorming, feedback, and co-construction can support constructive and interactive learning — related
• A synergistic effect between tutor and tutee roles may explain the peer agent's advantage — related