International Journal of Computer-Supported Collaborative Learning

Effective group formation is an indispensable yet challenging aspect of classroom-based collaborative learning. While existing group formation algorithms show promising computational performance in controlled settings, their practical impact on diverse, real-world classrooms remains underexplored. This paper presents a mixed genetic algorithm integrated into a data-driven learning platform designed to accommodate both homogeneous and heterogeneous student characteristics simultaneously. Implemented in a senior high school EFL classroom, the approach leverages active reading marker logs for data-driven grouping. It incorporates a WordCloud tool to enhance educators’ and learners’ understanding of group composition. Empirical results indicate that this system improves vocabulary learning, and the marker-based grouping strategies positively influence group learning dynamics. These findings underscore the algorithm’s practical relevance and highlight the benefits of interpretable, adaptive group formation methods for authentic educational contexts.

Understanding individual participation is critical for uncovering how individuals learn in collaborative learning as well as for providing personalized support to scaffold team success and individual gains. Modelling individual participation requires a process-oriented method rather than an outcome-focused approach. There is a need for a theoretical framework guiding the collection and analysis of process data for gauging individual participation in collaborative learning. This systematic review synthesizes theoretical aspects and analytical methods for modelling individual participation using process data in collaborative learning. It analyzes 66 studies published between 2005 and 2024, identified through the PRISMA process. Based on the analytical results, we propose a new theoretical framework, COMPAS, consisting of six components to model the multi-faceted and multi-level nature of individual participation in collaborative learning processes: Cognitive interactions, Coordinative interactions, Metacognitive interactions, Passive participation, solo Active participation, and Socio-emotional interactions. The six forms of individual participation were studied using various forms of collaborative learning process data—including oral conversational data, textual input data, log data, and non-verbal physical data—with analytical methods primarily involving descriptive analysis, content analysis, network analysis, and clustering. The synthesized factors influencing individual participation reflect a bi-directional relationship between individual participation and group performance in collaborative learning. This study contributes a new theoretical framework for understanding different forms of individual participation in collaborative learning, as well as highlights the need for and importance of multimodal process data in collaborative learning analytics.

This study qualitatively develops further understandings regarding knowledge and identity construction within computer-supported collaborative learning (CSCL) research by applying discourse analysis and the dialogical self theory (DST) to investigate the role of interpersonal and intrapersonal voices in facilitating knowledge construction. We analyzed and compared the audio recordings of ten students separated into two groups of five (group A and group B) as they engaged in dialogue to construct knowledge for a learning task on physics in a CSCL environment. We divided the dialogue of each group into dialogues by identifying their discourse functions (DF) on the basis of interactional events related to knowledge construction. We then grounded the I-positions of I–it, I–me, I–you and I–we at the utterance level so that we could visualize and describe them within the dialogues that were relevant during knowledge construction in each group. Results showed that the process of knowledge construction for collaborative learning related to DF as well as their use of I-positions. Group A, who failed the learning task, often employed interpersonal voices (I–it/I–me) in dialogues that focused on reviewing directions and strategies at the individual level rather than collective clarification and elaboration or empirical evaluation of knowledge. Group B, who succeeded with the learning task, employed both interpersonal (I–it/I–me) and intrapersonal voices (I–you/I–we) with intrapersonal voices peaking in usage during collective empirical evaluation of knowledge. Our findings underscore prior research that CSCL involves not only knowledge but also identity negotiation as well as demonstrating that DST can aid this exploration.

The development of immersive virtual reality (IVR) hardware and software has accelerated in recent years. The conceptual vocabulary has, however, not received the same amount of attention, especially in the context of collaborative learning settings. Existing concepts such as immersion, presence and interactivity focus predominantly on the individual user’s experience, neglecting the social and collaborative dimensions of learning supported by IVR. This limitation is particularly evident in the context of collaborative 360-degree virtual reality (360VR), in which learners must interact with each other to understand the recorded activity rather than interact directly with the environment itself. This paper challenges the current conceptualisations of IVR learning processes and proposes a new conceptual framework: Activity-based collaborative virtual reality (ABC-VR). ABC-VR builds on activity theory and is aligned with a dual sense of activity – the ‘original’ activity recorded with 360-degree video cameras and the activity produced by learners collaboratively and interactively immersing themselves in an ABC-VR session. In this dual sense of activity, ABC-VR is oriented to activities and the ways learners inhabit activities from educational, social and work practices, where actions and operations occur in a situated context shaped by the learners’ embodied actions and use of material resources. In ABC-VR, learners can question each other and imagine future actions – but they can also figuratively reason, imagine and reflect with the participants in the original activity. Through this paper, we unfold a conceptualisation of collaborative 360VR that can support the design of pedagogical activities and future research on IVR in collaborative learning activities.