The role of activities and verbal interactions on engineering students' learning outcomes across dyadic and individual conditions

Abstract

Collaborative learning is an instructional method in which two or more people work in small groups to construct knowledge jointly and/or to achieve a common goal (Dillenbourg, 1999; Roschelle, 1992; Webb, Troper, & Fall 1995). Many studies illustrated the cognitive benefits of collaborative learning as comprehension of ideas, retention of knowledge, integration of new and old knowledge, and transfer of knowledge (e.g., Chi & Menekse, 2015; Fischer & Mandl, 2005; Slavin, 1996; Stahl & Hesse, 2009). Even though the value of collaborative learning has been well documented across domains, there is no single definite description in order to explain the learning processes in collaborative learning (e.g., Hausmann, Chi & Roy, 2004; O'Donnell, 2006). In addition, some studies showed, under some conditions, collaboration do not facilitate learning (e.g., Barron, 2003; Dillenbourg et al., 1996; Phelps & Damon, 1989). For example, Lou and colleague's (1996) meta-analysis showed almost 25% of the published studies in collaborative learning showed null or even unpredicted effects when compared to individual learning conditions. Dillenbourg and Hong (2008) argued that the lack of the elaborated explanations, the mismatch in mutual regulations of cognitive processes between group members, low quality of arguments and the nonexistence of negotiation of meanings reduces the effectiveness of collaborative learning. Barron's (2003) work showed students' supportive communication and responsiveness toward proposed ideas significantly affected the learning outcomes of the collaborative group. She found less successful groups did not discuss the proposed ideas or directly rejected them compare to successful groups. We conducted two studies (Study 1 and 2) to explore the learning processes in collaborative settings. Specifically, we investigated the effective dialogue patterns and verbal moves for productive interactions, and how these collaborative interactions are influenced by the instructional materials provided for students to engage. For both studies, we compared the dyads' performance in collaborative learning condition with individual students' learning outcomes in solo condition. While all the analysis for the Study 1 is finalized and reported in this paper, the verbal analysis for the Study 2 is still in progress, therefore we only report the overall learning results for the Study 2.