Shall We Explain? Augmenting Learning from Intelligent Tutoring Systems and Peer Collaboration

Authors:
Robert G. Hausmann;Brett Sande;Kurt Vanlehn
Affiliations:
Pittsburgh Science of Learning Center, University of Pittsburgh, Pittsburgh, 15260-5179;Pittsburgh Science of Learning Center, University of Pittsburgh, Pittsburgh, 15260-5179;Pittsburgh Science of Learning Center, University of Pittsburgh, Pittsburgh, 15260-5179
Venue:
ITS '08 Proceedings of the 9th international conference on Intelligent Tutoring Systems
Year:
2008

Citing 5
Cited 1

Off-task behavior in the cognitive tutor classroom: when students "game the system"

Proceedings of the SIGCHI Conference on Human Factors in Computing Systems
Can Help Seeking Be Tutored? Searching for the Secret Sauce of Metacognitive Tutoring

Proceedings of the 2007 conference on Artificial Intelligence in Education: Building Technology Rich Learning Contexts That Work
Measuring the Effect of Collaboration in an Assessment Environment

Proceedings of the 2007 conference on Artificial Intelligence in Education: Building Technology Rich Learning Contexts That Work
Who Says Three's a Crowd? Using a Cognitive Tutor to Support Peer Tutoring

Proceedings of the 2007 conference on Artificial Intelligence in Education: Building Technology Rich Learning Contexts That Work
Explaining Self-Explaining: A Contrast between Content and Generation

Proceedings of the 2007 conference on Artificial Intelligence in Education: Building Technology Rich Learning Contexts That Work

The effects of domain and collaboration feedback on learning in a collaborative intelligent tutoring system

AIED'11 Proceedings of the 15th international conference on Artificial intelligence in education

Quantified Score

Hi-index	0.00

Visualization

Abstract

Learning outcomes from intelligent tutoring systems (ITSs) tend to be quite strong, usually in the neighborhood of one standard deviation. However, most ITS designers use the learning outcomes from expert human tutoring as the gold standard (i.e., two standard deviations). What can be done, with the current state of the art, to increase learning from an ITS? One method is to modify the learning situation by asking students to use the ITS in pairs. To enhance performance, we drew upon the beneficial effects of structured peer collaboration. The results suggest that the intervention was successful. Pairs of students solved more problems and requested fewer bottom-out hints than individuals. To test the possibility that the effect was due to the best partner in the group directing the problem solving, a nominal groups analysis was conducted. A nominal group is a statistical pairing of the non-interacting individuals' performance. The results from the nominal groups replicated the same pattern of results, but with a reduced magnitude. This suggests that the best member may have contributed to some of the overall success of the pair, but does not completely explain their performance.