Reflective writing is part of many higher education courses across the globe. It is often considered a challenging task for students as it requires self-regulated learning skills to appropriately plan, timely engage and deeply reflect on learning experiences. Despite an advance in writing analytics and the pervasiveness of human feedback aimed to support student reflections, little is known about how to integrate feedback from humans and analytics to improve students’ learning engagement and performance in reflective writing tasks. This study proposes a personalised behavioural feedback intervention based on students’ writing engagement analytics utilising time-series analysis of digital traces from a ubiquitous online word processing platform. In a semester-long experimental study involving 81 postgraduate students, its impact on learning engagement and performance was studied. The results showed that the intervention cohort engaged statistically significantly more in their reflective writing task after receiving the combined feedback compared to the control cohort which only received human feedback on their reflective writing content. Further analyses revealed that the intervention cohort reflected more regularly at the weekly level, the regularity of weekly reflection led to better performance grades, and the impact on students with low self-regulated learning skills was higher. This study emphasizes the powerful benefits of implementing combined feedback approaches in which the strengths of analytics and human feedback are synthesized to improve student engagement and performance. Further research should explore the long-term sustainability of the observed effects and their validity in other contexts.

Collaboration is argued to be an important skill, not only in schools and higher education contexts but also in the workspace and other aspects of life. However, simply asking students to work together as a group on a task does not guarantee success in collaboration. Effective collaborative learning requires meaningful interactions among individuals in a group. Recent advances in multimodal data collection tools and AI provide unique opportunities to analyze, model and support these interactions. This study proposes an original method to identify group interactions in real-world collaborative learning activities and investigates the variations in interactions of groups with different collaborative learning outcomes. The study was conducted in a 10-week long post-graduate course involving 34 students with data collected from groups’ weekly collaborative learning interactions lasting ~ 60 min per session. The results showed that groups with different levels of shared understanding exhibit significant differences in time spent and maximum duration of referring and following behaviours. Further analysis using process mining techniques revealed that groups with different outcomes exhibit different patterns of group interactions. A loop between students’ referring and following behaviours and resource management behaviours was identified in groups with better collaborative learning outcomes. The study indicates that the nonverbal behaviours studied here, which can be auto-detected with advanced computer vision techniques and multimodal data, have the potential to distinguish groups with different collaborative learning outcomes. Insights generated can also support the practice of collaborative learning for learners and educators. Further research should explore the cross-context validity of the proposed distinctions and explore the approach’s potential to be developed as a real-world, real-time support system for collaborative learning.

Incorporating feedback into the learning process is crucial for learners’ success. Recent advancements in Learning Analytics (LA) and Artificial Intelligence (AI) have introduced a range of feedback generation and delivery opportunities to improve teaching and learning processes, yet questions remain about how learners engage with such tools and their impact. In this study, we investigated postgraduate students’ engagement with analytics feedback in relation to their level of self-regulated learning (SRL) competence and performance in a semester-long (ten-week) course. We specifically focused on the Interactive-Constructive-Active–Passive (ICAP) framework of cognitive engagement. Initially, students were asked to participate in an established SRL questionnaire, based on Zimmerman’s theory of self-regulation to evaluate their SRL competence (N = 39). Throughout the semester, their online behaviour data from Moodle and Google Docs was collected and analyzed to form personalised analytics feedback for each student. We examined how students with different SRL competencies engage with analytics feedback and the impact of this engagement on students’ course performance. Results indicated that students with high SRL competence actively engage with analytics feedback more than students with low SRL competence. However, students' analytics feedback engagement did not significantly affect their course performance. Additionally, we analyzed students’ reflections on the feedback provided to investigate how they perceived it in relation to their learning experiences and performance. Students argued in their reflections that analytics feedback was beneficial in identifying and regulating their online behaviours and providing motivation through objective insights. They also noted limitations in accurately reflecting their behaviours and learning quality, the need for more personalised recommendations and timely feedback, and suggested design improvements to ensure clarity, foster interaction and incorporate tailored, in-depth insights. We conclude the paper with a discussion on future design and research suggestions for ways of monitoring and supporting students’ cognitive engagement with analytics feedback interventions.

Using learning analytics to investigate and support collaborative learning has been explored for many years. Recently, automated approaches with various artificial intelligence approaches have provided promising results for modelling and predicting student engagement and performance in collaborative learning tasks. However, due to the lack of transparency and interpretability caused by the use of “black box” approaches in learning analytics design and implementation, guidance for teaching and learning practice may become a challenge. On the one hand, the black box created by machine learning algorithms and models prevents users from obtaining educationally meaningful learning and teaching suggestions. On the other hand, focusing on group and cohort level analysis only can make it difficult to provide specific support for individual students working in collaborative groups. This paper proposes a transparent approach to automatically detect student's individual engagement in the process of collaboration. The results show that the proposed approach can reflect student's individual engagement and can be used as an indicator to distinguish students with different collaborative learning challenges (cognitive, behavioural and emotional) and learning outcomes. The potential of the proposed collaboration analytics approach for scaffolding collaborative learning practice in face-to-face contexts is discussed and future research suggestions are provided.