Analysis of CSCL Chats for Cognitive Assessment and Individual Participations

Abstract

Computer Supported Collaborative Learning (CSCL) provides the possibility of learning through collaborative interaction, and the social construction of knowledge through the utilization of information technology (IT). It contains several tools that allow learners to participate and interact in a virtual environment. Chat is one of the most effective tools in CSCL owing to the accessibility and ease of its use by learners. It is also one of the best methods used in the learning discussions. However, there are a few tools that help teachers evaluate students' chats. Due to this difficulty, chats are rarely used in the formal learning context. To remedy this problem, this paper provides a model that enables teachers to automatically assess cognition and individual participations in chats. With regards to the cognitive assessment, a test will be utilized. The test contains a set of questions that students were asked to pose about the threads they will discuss. Then, these questions were put in a database to be answered by the students after discussions. Accordingly, the cognitive assessment of the students will be evaluated through results obtained from the test. As for the assessment of individual participations, chats will be analyzed using the Trausan-Matu’s polyphonic model based on Bakhtin’s ideas, in order to obtain the most frequent words for each participant and their inter-relations. Obtaining results in the form of graphics and statistical tables helps teachers to know the results of cognitive assessment and relate them to the individual participations of each student. The aim is to reach rational findings that link the individual participations, namely the students’ behavior in their dialogs within the chat, and the results of cognitive assessment resulting from the test given to the students at the end of the chat. In sum, the system offers a general perception of the assessment that is more effective and able to measure the level of collaborative learning in CSCL chats. From these results, teachers can easily assess the learning dialogs, in order to assess the students' cognition of the threads discussed. By doing so, this would help teachers to develop learning strategies that are capable of reaching an effective collaborative learning dialog.