Research on the Quality of Collaboration in Project-Based Learning Based on Group Awareness

Project-based learning (PBL) is an important form of collaborative learning that has a significant positive impact on student capacity development. However, data generated during long periods of PBL are difficult to record in real time, and there is a dearth of specific empirical research on the relationship between the quality of collaboration and the effectiveness of collaboration. In this study, we employ text mining to measure and analyze process data from a college physics course that utilizes project-based learning at a university. Under the group awareness theory, we construct a project-based learning system and extract six multi-dimensional awareness data (including numbers of text, cognitive awareness, emotional awareness, behavioral awareness, social awareness of group members, and group leaders' cognitive awareness of the project). These data are then utilized to build a multiple regression model, which enables the analysis of collaboration quality within collaborative groups. The results show that the group awareness information automatically processed by text mining can reflect the quality of collaboration, and the quality of collaboration can determine about 50% of the effectiveness of collaboration. The important factors affecting the collaboration quality of project-based learning were group leaders' cognitive awareness of the project and the cognitive awareness of group members. Therefore, in order to improve the quality of collaboration in PBL, teachers should select responsible group leaders and encourage meaningful communication between group members, thereby fostering high-quality and sustainable collaboration development.