12.07.2015 Views

Deliverable 4.4 - INSEAD CALT

Deliverable 4.4 - INSEAD CALT

Deliverable 4.4 - INSEAD CALT

SHOW MORE
SHOW LESS

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

AtGentive IST-4-027529-STP - Attentive Agents for Collaborative LearnersAvg # of MC-Intervention Vs Paper quality1412Avg # of MC-Intervention1086420bad medium goodPaper qualityFigure 1 – MC-Interventions received by children who produced papers that were graded asbad, medium, and good.A detailed analysis of the children's behaviour on the logs has lead us to the conclusionthat no significant correlation between the quality of paper and Meta-Cognitiveinterventions is shown because some of the students who finally produced a bad paperkept swapping between activity windows, which has generated many meta-cognitiveinterventions. It appears therefore that the system was unable to help appropriately acertain number of students who kept moving around the application windows receivingmeta-cognitive interventions that didn't really help them improving their performance.Figure 2 shows that the same trend cannot be found in the case of Cognitive andMotivational interventions, in fact many children didn't receive any interventions of thesetypes, and others, who have received the most cognitive and motivational interventions,have produced papers of worse quality.Avg # of M-Intervention Vs Paper QualityAvg # of C-Interventions Vs Paper Quality43Avg # of M-Intervention3.532.521.510.5Avg # of C-Interventions2.521.510.500bad medium goodPaper QualitybadPaper QualitymediumgoodFigure 2 – Motivational (left) and Cognitive (right) Interventions received by children whoproduced papers that were graded as bad, medium, and good.The large number of Cognitive and Motivational interventions received by some of thechildren who produced poor papers (and the negative correlation between the number ofCognitive interventions and the quality of paper) could be explained by the fact that suchinterventions were addressed to students who were clicking on the question mark buttonand on the unhappy button. We hypothesized that children who had been previouslyDel <strong>4.4</strong>: AtGentive Final Evaluation Report – Appendix C page 6

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!