Article Preview
TopIntroduction
In the teaching and learning processes developed in Computer-Supported Collaborative Learning (CSCL) environments, the concept of group is paramount. A learning group is a structure made up of people who interact to achieve certain learning objectives through their participation (Souto, 1993). People participation is related to communication, information processing, coordination, interpersonal relationships and motivation. All these factors are very important in the learning process (Meier, Spada, & Rummel, 2007) because they influence cognitive processes of learners (Dillenbourg, 1999). Unfortunately, forming a group does not guarantee neither the success of the collaborative learning nor that the group works in a coordinated and efficient way (Lazareva, 2015; Matteucci et al., 2010).
CSCL systems provides communication, coordination and collaboration tools that ease group learning by removing geographical and temporal limitations for group members. However, in collaborative learning situations groups are exposed to several factors that can affect the social dynamic which in turn can have a negative influence in learning process (Järvenoja, Volet, & Järvelä, 2013). Two of these factors are the collaborative skills that group members display and the roles that they can play.
Soller (2001) argues that students which learn in groups manifest collaborative learning skills such as encourage each other to ask questions, explain and justify their opinions, articulate their reasoning, and elaborate and reflect upon their knowledge. Benne and Sheats (1948) point out that the quality and the amount of group production is a responsibility of all group members through the adoption of roles. Each role is characterized by manifesting interaction patterns during the communication process (Benne & Sheats, 1948). A role is related to a set of behaviors and attitudes that a person displays in their interactions (R. Turner, 2001).
Without the appropriated collaborative skills and roles, learners cannot have a dynamic of work that enable them to make a coordinated work and reach a successful collaborative learning. It is known that a group needs a balanced combination of roles to achieve a coordinated work (Belbin, 2010). Balanced combination means that group members play all possible roles and roles are played by the different team members (Alberola, Del Val, Sanchez-Anguix, Palomares, & Teruel, 2016; Belbin, 2010).
Taking into account that roles are manifested as behaviors and that these behaviors can be materialized as collaborative skills, it would be desirable to define an explicit connection between collaborative skills and group roles. The log of students interactions stored by CSCL systems is a rich data source to explore relations between collaborative skills and roles, yet manual analysis of interaction logs is prohibitive when there are a lot of students and messages exchanged between them (Borges & Baranauskas, 2003; Chen, 2006; Constantino-González, Suthers, & Escamilla De Los Santos, 2003; Dönmez, Rosé, Stegmann, Weinberger, & Fischer, 2005; Rosé et al., 2008). Datamining provides techniques for automatic discovering of patterns when analyzing large volumes of data.
This paper proposes the use of dataminig when analyzing log of students’ interactions in CSCL systems to explore relations between collaborative skills and the roles played by group members. In this study collaborative skills defined by Soller (2001) and group role classification made by Belbin (2001) were considered.