This study explores the differences in students' verbal behavior sequences between high-performing (HP) and low-performing (LP) groups in a computer-supported collaborative learning (CSCL) environment. Employing quantitative content analysis and Lag Sequential Analysis (LSA), this study analyzed the verbal interactions of these two groups. The findings reveal that HP groups frequently engaged in cycles of negotiation, clarity-seeking, and task coordination, leading to effective collaboration and problem-solving. In contrast, LP groups exhibited fragmented problem-solving approaches and frequent off-task behaviors. These insights highlight the importance of structured support and focused task management in enhancing collaborative learning outcomes. These findings suggest that educators should foster learning environments that promote continuous critical evaluation and seamless coordination to improve group performance.

This study explores students' verbal interaction dynamics in two computer-supported collaborative learning (CSCL) environments: immediate collaboration and individual preparation (IP) followed by group collaboration. Although verbal interactions are not always central to all CSCL designs, they are critical in contexts that emphasize face-to-face or synchronous communication, where they facilitate negotiation, idea sharing, and collaborative knowledge construction. By applying content analysis and lag sequential analysis (LSA), this study examined the verbal interaction behavioral sequences of students in both conditions to understand how IP influences collaborative dynamics. The findings highlight the crucial role of IP in enhancing collaborative dynamics, suggesting that well-structured preparatory activities can significantly improve group interaction efficiency. This research contributes valuable insights for refining CSCL instructional strategies, emphasizing the need to balance structured preparation with opportunities for spontaneous interaction to optimize collaborative learning outcomes. By managing distractions and maintaining task focus, educators can create more effective collaborative learning environments.

Abstract: Peer feedback is a pedagogical strategy for peer learning. Despite recent indications of Large Language Models (LLMs) ' potential for content analysis, there is limited empirical exploration of their application in supporting the peer feedback process. This study enhances the analytical approach to peer feedback activities by employing state-of-the-art LLMs for automated peer feedback feature detection. This research critically compares three models—GPT-3.5 Turbo, Gemini 1.0 Pro, and Claude 3 Sonnet— to evaluate their effectiveness in automated peer feedback feature detection. The study involved 69 engineering students from a Singapore university participating in peer feedback activities on the online platform Miro. A total of 535 peer feedback instances were collected and human-coded for eleven features, resulting in a dataset of 5,885 labeled samples. These features included various cognitive and affective dimensions, elaboration, and specificity. The results indicate that GPT-3.5 Turbo is the most effective model, offering the best combination of performance and cost-effectiveness. Gemini 1.0 Pro also presents a viable option with its higher throughput and larger context window, making it particularly suitable for educational contexts with smaller sample sizes. Conversely, Claude 3 Sonnet, despite its larger context window, is less competitive due to higher costs and lower performance, and its lack of support for training and fine-tuning with researchers' data weakens its learning capabilities. This research contributes to the fields of Al in education and peer feedback by exploring the use of LLMs for automated analysis. It highlights the feasibility of employing and fine-tuning existing LLMs to support pedagogical design and evaluations from a process-oriented perspective.

This study investigated (the extent to which levels of certainty impacted the argumentative knowledge construction in individual work and group work. Argumentative knowledge construction has been characterized into simple claims, grounds, qualifiers, counterarguments, and integrated replies to illustrate the components of argumentation and nature of resolving conflicts in argumentation where certainty levels have been divided into uncertain, neutral, and certain. Findings showed that individual and group work differed significantly in terms of levels of certainty for simple arguments and counterarguments. Study implications were discussed.

Through collaborative argumentation, students gain in-depth understanding of learning content when they build on one another’s knowledge. Although individual preparation (IP) is found to be effective to foster collaborative learning, the mechanism of how IP influence the knowledge construction behavior is underexplored. This study investigated how IP influenced secondary school students in relation to knowledge construction behavioral patterns when participating in online collaborative argumentation activities. 20 students participated in two computer-supported collaborative argumentation lessons with one group with IP, and the other group without. Screen video recordings of students constructing arguments in groups during two lessons were collected and analyzed. Epistemic Network Analysis was conducted to examine students’ knowledge construction behaviors in the two lessons with and without IP. The results show that there were significant impact on students’ knowledge construction characteristics between the two lessons. Students who did not go through the IP phase tended to exhibit behaviors related to ideas refinement more than the students who went through the IP phase. The implications of how to design and implement effective knowledge construction are discussed.

This study compares the high- and low-performing groups’ knowledge co- construction process in the context of computer-supported collaborative argumentation from epistemic, argument, and social perspectives. Product analysis, lag sequential analysis, Sankey diagram visualization, and social network analysis were used to analyze groups’ written argumentation artefacts, on-screen behaviors, and online interactions. Results show that the high-performing group students demonstrated a higher level of engagement and cognitive elaboration than the low-performing group. The high-performing group was more competent in integrating various argumentation elements than the low-performing group. And the students in the high-performing group tended to contribute equally to their group work. The implications of the findings in designing and implementing knowledge co-construction activities are discussed.