Content analysis of computer-mediated collaborative mathematical problem solving

This study provides a preliminary look into the process of mathematical problem solving within small groups in an online environment, and distinguishes the behaviors of individuals and groups involved in effective collaborative problem solving via computermediated communication. The subjects of this study included four groups of three middle school teachers enrolled in a mathematical content course in which the primary format was group problem solving with communications mediated by computer. Data was collected by observing and performing a content analysis on transcripts of online problem-solving protocols, and coding messages along four dimensions (a) participative, (b) social, (c) interactive, and (d) heuristic episode using a framework developed by the researcher. Problem-solving success was measured by group achievement on the assigned problem-solving task as determined by an analytical rubric. Several important conclusions emerged concerning computer-mediated collaborative problem solving. The primary observations related to the ways in which groups communicated and participated. The data indicated that the most effective group not only participated more, but also socialized less, interacted more, with more task-specific content and more interactions related to solving the problem. Additionally, this group collaborated throughout the problem-solving process in a recursive nature, working to understand the problem, develop a plan, execute that plan, and verify the results, and revisiting each of these heuristic stages. This study also illuminated several issues related to the use of technology for collaborative problem solving. Evidence pointed to the choice between asynchronous and synchronous modes for communication as the primary determinant in the amount of participation and the amount and type of interactivity amongst members. While asynchronous messages tended to contain more lengthy messages with multiple units of meaning, the messages were few, and the communication between group members was highly non-interactive and appeared to do little to advance group problem-solving efforts. In contrast, the synchronous mode was much more interactive in nature and appeared to promote much greater collaboration. In conclusion the author includes several suggestions for promoting effective collaboration and successful problem-solving groups; they include communication requirements and guidelines, and considerations for group composition and task characteristics. Additionally, the author suggests several ideas for future research.