Effects of ML-LAF and GRG system in online collaborative learning environments on academic performance, metacognition, participation, and interaction: a repeat measurement method

OpenAlex · İngilizce

Online collaborative learning environment provides many opportunities for the development of twenty-first century learning skills and competencies. Despite the potential of online collaborative learning, there is a deficiency of research investigating the impact of online behaviors of groups and individuals within groups on collaborative processes. In this context, this study develops guidance and feedback approaches to support university students’ learning. The group regulation guidance approach consists of feedback and feedforward mechanisms. The feedback approach (machine learning based learning analytics feedback, ML-LAF) consists of personalized feedback based on the information generated by the clustering of learning analytics in the context of students’ learning analytics/navigation in the online environment with machine learning algorithms. In this context, the aim of this study is to evaluate the feedback system based on learning analytics and the group regulation guidance integrated into this system in online collaborative learning environments. 239 students participated in the study. As a result of the study, it was shown that the combination of both past and future feedback did not have the expected effect on student performance and even decreased performance. Although Group 3 interacted most intensively with the learning materials, it was the group with the lowest metacognition.