ABSTRACT

This paper explores the formation of a learning community facilitated by custom collaborative learning software. Drawing on research in group cognition, knowledge building discourse, and learning analytics, we conducted a mixed-methods field study involving an asynchronous online discussion consisting of 259 messages posted by 50 participants. The cluster analysis results provide evidence that the recommender system within the software can support the formation of a learning community with a small peripheral cluster. Regarding knowledge building discourse, we identified the distinct roles of central, intermediate (i.e., middle of three clusters), and peripheral clusters within a learning community. Furthermore, we found that message lexical complexity does not correlate to the stages of knowledge building. Overall, this study contributes to the group cognition theory to deepen our understanding about collaboration to construct new knowledge in online discussions. Moreover, we add a much-needed text mining perspective to the qualitative interaction analysis model.

摘要

本文探讨了由自定义协作学习软件促进的学习社区的形成。借鉴对群体认知、知识构建话语和学习分析的研究，我们进行了一项混合方法实地研究，涉及由 50 名参与者发布的 259 条消息组成的异步在线讨论。聚类分析结果证明，软件中的推荐系统可以支持形成具有小型外围集群的学习社区。关于知识构建话语，我们确定了学习社区中中心、中间（即三个集群的中间）和外围集群的不同角色。此外，我们发现消息词汇复杂性与知识构建的阶段无关。全面的，这项研究有助于群体认知理论加深我们对合作的理解，以在在线讨论中构建新知识。此外，我们在定性交互分析模型中添加了急需的文本挖掘视角。