Assessing collaboration from real computer supported collaborative learning sessions

Considering that collaboration is one of the main goals to be achieved in Computer Supported Collaborative Learning (CSCL) systems, one of the most important and complex task in these systems is to assess the collaboration among participants. In the present work, collaborations take place by interacting through dialogs that contain actions performed by participants organized by dyads. Such actions are transformed into graphical representations called Networks of Concepts (NoC), which synthesizes the knowledge being learned. Here we propose how to measure quantitative collaborations. Quantitative collaborations are measured through the analysis of evidences derived from actions taking place within collaborative learning sessions.1