evidoria

← Back to browse

Good practice Imported

ChatTutor — University of Copenhagen's Two-Experiment Test of a Learning-Theory-Based AI Tutor Against ChatGPT

Denmark · Copenhagen · See the Denmark profile · See the Copenhagen profile

Top 56% 47/100 · Ask Evidence Copilot about this practice

Two experiments (175 university, 234 high-school students) found a learning-theory-based chatbot beat ChatGPT on conceptual knowledge, trust and enjoyment; gains vs control were mixed and one author has a commercial interest.

ChatTutor — University of Copenhagen's Two-Experiment Test of a Learning-Theory-Based AI Tutor Against ChatGPT

Details

Promoter
University of Copenhagen (Department of Psychology) / ChatTutor
Period
2024-2025
Keywords
higher education, secondary education, AI tutoring, generative learning

Description

University of Copenhagen researchers tested ChatTutor, a chatbot that prompts students to explain concepts and gives feedback on gaps instead of giving answers, in two experiments (175 university and 234 high-school students) against ChatGPT and a control.
Study 1 was pre-registered, run in a real cognitive psychology course; Study 2 replicated it in high schools. ChatTutor beat ChatGPT on immediate conceptual knowledge in both studies and on trust and enjoyment; at the 4-week follow-up it beat the teaching-as-usual control in Study 1 (M=7.26 vs 5.93, p=0.001) but not ChatGPT.
Caveats: the intervention was a ~15-minute follow-up activity, attrition at follow-up was substantial (124 of 175 in Study 1), self-efficacy differences were not significant, the immediate difference vs control was non-significant in Study 1, and one author has a commercial interest in ChatTutor. The study was not externally funded.

Read the full analysis: https://news.ku.dk/all_news/2025/09/ai-tool-built-for-learners-supports-learning-better-than-chatgpt/

Implementation

Implementation detail (cost, timeline, staffing, conditions for success) is not yet available for this practice.

Do you run this practice? Claim it — verified implementers get a public contact pathway and can propose corrections.

Data sources

Where this practice's information was retrieved from, and when.

Attachments

Similar practices you may find useful