Date of Award
Doctor of Philosophy
Arthur C Graesser
Leah C. Windsor
Communication, collaboration and the social co-construction of knowledge are now considered critical 21st century skills and have taken a principal role in recent theoretical and technological developments in education research. The overall objective of this dissertation was to investigate collaborative learning to gain insight on why some groups are more successful than others. In such discussions, group members naturally assume different roles. These roles emerge through participants’ interactions without any prior instruction or assignment. Different combinations of these roles can produce characteristically different group outcomes, being either less or more productive towards collective goals. However, there has been little research on how to automatically identify these roles and fuse the quality of the process of collaborative interactions with the learning outcome. A major goal of this dissertation is to develop a group communication analysis (GCA) framework, a novel methodology that applies automated computational linguistic techniques to the sequential interactions of online group communication. The GCA involves computing six distinct measures of participant discourse interaction and behavioral patterns and then clustering participants based on their profiles across these measures. The GCA was applied to several large collaborative learning datasets, and identified roles that exhibit distinct patterns in behavioral engagement style (i.e., active or passive, leading or following), contribution characteristics (i.e., providing new information or echoing given material), and social orientation. Through bootstrapping and replication analysis, the roles were found to generalize both within and across different collaborative interaction datasets, indicating that these roles are robust constructs. A multilevel analysis shows that the social roles are predictive of success, both for individual team members and for the overall group. Furthermore, the presence of specific roles within a team produce characteristically different outcomes; leading to specific hypotheses as to optimal group composition. Ideally, the developed analytical tools and findings of this dissertation will contribute to our understanding of how individuals learn together as a group and thereby advance the learning and discourse sciences. More broadly, GCA provides a framework to explore the intra- and inter-personal patterns indicative of the participants’ roles and the sociocognitive processes related to successful collaboration.
dissertation or thesis originally submitted to the local University of Memphis Electronic Theses & dissertation (ETD) Repository.
Dowell, Nia Marcia Maria, "A Computational Linguistic Analysis of Learners Discourse in Computer-Mediated Group Learning Environments" (2017). Electronic Theses and Dissertations. 1639.