Conversational-agent prompts improve student dialogue outcomes at both individual and dyad levels¶

Claim · All claims
Evidence · 1 study · causal r? · q2
Currency · newest study 2016 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q2 i? In an experimental study with pre- and post-test treatment and control groups, prompts by conversational agents led to improved outcomes for students at the individual level and at the level of the dyad as a whole. → Sten Ludvigsen 2016

Evidence¶

Sten Ludvigsen 2016¶

Sten Ludvigsen, Ulrike Cress, Nancy Law, Carolyn P. Rosé, & Gerry Stahl. (2016). Collaboration scripts and scaffolding. International Journal of Computer-Supported Collaborative Learning, 11, 381–385. https://doi.org/10.1007/s11412-016-9247-1

q2 · i? · causal · r?

The editors report the experimental study by Tegos, Demetriadis, Papadopoulos, and Weinberger with second-year computer science students in a human-computer interaction course, using pre- and post-test treatment and control groups. They state the prompts "lead to improved outcomes for the students"; no effect sizes are printed.

"The results show that the prompt by the conversational agents lead to improved outcomes for the students, not only at the individual level but also at the level of the dyad as a whole."

Discussion¶

Related Claims¶

• In a 2x2 experiment on jigsaw dyads, the implicit gaze awareness tool affected dialogue processes much more than the explicit pedagogical conversational agent, which compensated when gaze awareness was absent — related
• Pre-written prompts increase Justifications and Alternative ideas in children's pair-programming talk and shift the role of Simple questions — related
• Teaching a ChatGPT agent improves students' self-regulated learning, specifically self-efficacy and use of cognitive strategies — related
• Students who teach a ChatGPT teachable agent in natural language achieve greater knowledge gains than students who learn the same material from online videos — 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