Language：English   Publisher：Proceedings 25th IEEE International Conference on Advanced Learning Technologies Icalt 2025  

Enhancing peer interactions is a crucial topic in collaborative learning, where both the quantity and quality of interaction are emphasized. Despite its importance, the social and emotional aspects of collaborative learning, such as peer interactions, remain underexplored. This study applied learning analytics to examine how learner similarity in learning behavior, engagement, and knowledge influences the quantity and quality of peer interaction within the peer help system. Three aspects were analyzed using logistic regression and ordinal logistic regression: whether learners reply to questions, whether they evaluate the help received, and the evaluation scores given by askers to helpers. The results reveal that knowledge similarity has a significant negative impact on whether learners reply to questions, while no factors were found to significantly impact whether learners evaluate the help received. However, knowledge similarity showed a significant positive effect on evaluation scores. These findings provide insights into the complexity of learner similarity in shaping peer interactions and offer implications for designing methods that balance homogeneity and heterogeneity to enhance peer interactions in computer-supported collaborative learning environments.

DOI： 10.1109/ICALT64023.2025.00029

Scopus

Language：English   Publisher：Australasian Journal of Educational Technology  

Collaborative learning in tertiary education faces challenges such as limited teacher intervention and effective student pairing. This study addresses these issues by proposing a data-driven peer recommendation approach enhanced with learner profile visualisation. The system dynamically matches students based on evolving learning profiles, using an open learner model to improve transparency and decision-making. Implemented in a Japanese university, a pilot study in an academic reading course showed that peer feedback improved report scores, with visualisation aiding in selecting suitable peer reviewers. Comparisons across three recommendation rounds suggested that integrating recursive data accumulation strengthened personalised peer recommendations and encouraged greater participation. By demonstrating the workflow of peer learning implementation, this research also highlights the broader potential of data-driven systems to support collaborative learning in higher education. Implications for practice or policy: Peer review activities with clear criteria can help students revise and improve writing, even within limited rounds. Data-driven analytics and recursive evidence accumulation can enable personalised peer recommendations while ensuring inclusivity by mitigating algorithmic biases. Visualised reviewer profiles via open learner models may enhance perceived feedback quality and student agency, though more controlled validation is needed. Policymakers could support flipped learning models that leverage data-driven personalised recommendations to enhance learner autonomy and peer collaboration.

DOI： 10.14742/ajet.10411

Web of Science

Scopus

Language：English   Publisher：Eye Tracking Research and Applications Symposium ETRA  

Eye tracking provides a marker of attention. In the educational context, such behavior can be harnessed to understand learning behaviors. However, a technology framework that captures and utilizes such multimodal indicators in educational activities is lacking. This paper presents LA-ReflecT, a platform integrating multimodal data for micro-learning activities. Teachers can author learning tasks and enable tracking eye fixation behaviors. A web camera-based eye-tracking function captures the gaze data while attempting the learning task. Learners can control the settings to stop or pause recording. We present data-driven services such as visualizing gaze attention heatmap and genetic algorithm-based group formation. A classroom study with 41 students illustrates using the proposed framework in an authentic context. Data collected is analyzed to answer an initial research question regarding the correlation between the heterogeneity of the click and gaze patterns in a learning task. The work is open for a demo.

DOI： 10.1145/3715669.3726825

Web of Science

Scopus

Language：English   Publisher：Technology Knowledge and Learning  

This paper explores co-design in Japanese education for deploying data-driven educational technology and practice. Although there is a growing emphasis on data to inform educational decision-making and personalize learning experiences, challenges such as data interoperability and inconsistency with teaching goals prevent practitioners from participating. Co-design, characterized by involving various stakeholders, is instrumental in addressing the evolving needs of technology deployment. Japan's educational context aligns with co-design implementation, with a learning and evidence analytics infrastructure facilitating data collection and analysis. From the Japanese co-design practice of educational technologies, the paper highlights a 6-phase co-design framework: motivate, pilot, implement, refine, evaluate, and maintain. The practices focus on data-driven learning strategies, technology interventions, and across-context dashboards, covering assorted learning contexts in Japan. By advocating for a co-design culture and data-driven approaches to enhance education in Japan, we offer insights for education practitioners, policymakers, researchers, and industry developers.

DOI： 10.1007/s10758-024-09759-w

Web of Science

Scopus

Language：English   Publisher：Technology Knowledge and Learning  

This article brings together literature and perspectives that have evolved during the last decade on issues of social and emotional aspects of learning in a digital age. This topic points to some core challenges and worries of contemporary social developments within digitalized societies, and ways of perceiving future developments of how we conceptualize learning and education within and beyond formal schooling to better provide for ways of engaging young learners. The aim is to address some emerging issues on the importance of digital social and emotional skills (D-SEL) relevant for our understanding of learning and education in contemporary and future societies. We use developments in selected countries (Norway, Czech Republic, USA and Japan) as examples to discuss how social and emotional skills have entered educational systems. The findings show that not only knowledge in a cognitive sense is important for human life, but also people’s approach to life and their ability to adapt to changes as digital social and emotional ways of learning.

DOI： 10.1007/s10758-024-09775-w

Web of Science

Scopus

Language：English   Publisher：Journal of Computers in Education  

Reading is fundamental in language learning, and active reading strategies using annotations prove beneficial. In parallel, collaborative learning is also an instructive practice for active reading in flipped classrooms. The advance of information infrastructures with increasing learning log data facilitates technical innovations to scaffold reading activities and collaborative learning. This research proposes an innovative approach to incorporate annotation features from learning logs for algorithmic group formation and evaluates the effects of the grouping in authentic English classes in a Japanese junior high school context. Students from the marker-based heterogeneous groups performed better in the vocabulary recognition quiz and the after-class summary writing assignment. Besides, they also had higher self-perception of their group work engagement in the survey. Results indicate that group formation using the reading annotation behaviors can affect the learning outcome and group work engagement in the active reading language learning context. Meanwhile, the study also proposes opportunities for further group formation implementations using assorted data in the data-driven environment.

DOI： 10.1007/s40692-023-00274-y

Web of Science

Scopus

Language：English  

Web of Science

Language：English  

Web of Science

Language：English  

Web of Science

Language：English  

Web of Science

Language：English   Publisher：Lecture Notes in Computer Science Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics  

Group formation plays a crucial role in designing collaborative learning activities, as the success of the group largely relies on the makeup of its members. While numerous algorithms exist, many group formation systems tend to adopt a single grouping strategy, such as either heterogeneous or homogeneous grouping, limiting their ability to address diverse student characteristics simultaneously. In this paper, we propose an integrated approach utilizing a mixed genetic algorithm within a data-driven learning platform, which considers both homogeneous and heterogeneous characteristics concurrently. Through an exploratory implementation in a university course, we examined the algorithm’s performance using authentic log data under various grouping strategies in classroom settings. We also highlight the potential of interpretability in group formation results, particularly through the composition panel, enabling teachers to make informed interventions and thereby enhancing overall class performance.

DOI： 10.1007/978-3-031-67998-8_16

Web of Science

Scopus

Language：English   Publisher：Institute of Electrical and Electronics Engineers (IEEE)  

Iterative team-based learning (TBL) is a common educational strategy for collaborative learning that involves sequential phases of individual and group learning activities. The advent of digital learning platforms, with the accumulation of learning log data, presents an opportunity to leverage data-driven techniques to enhance TBL practices. However, applying data-driven approaches in iterative TBL scenarios has received limited exploration in existing literature. Through a review of initial studies in this domain, data-driven iterative TBL emerges as a promising area. To explore this topic, we introduce a novel framework, drawing from the GLOBE framework for group learning, aimed at integrating data-driven designs into iterative TBL settings. This framework is proposed to guide data and activity design within iterative TBL, comprising four phases of group learning activity workflow and three essential steps of data flow. Additionally, we present two authentic instances supported by empirical evidence, offering insights into how educators can implement data-driven designs across different phases of TBL. Within the data-driven environment, we also uncover potential impacts and challenges of data-driven iterative TBL, to identify avenues for future research that can further expand our understanding of the possibilities in this domain.

DOI： 10.1109/access.2024.3393421

Web of Science

Scopus

CiNii Research

Language：English  

Web of Science

Language：English   Publisher：31st International Conference on Computers in Education Icce 2023 Proceedings  

With the development of information technology tools and the Internet, computer-supported collaborative learning has become increasingly accessible and promising. Peer help is a popular practice of collaborative learning. In this paper, we propose a way to realize intelligently mediated peer help. We obtain open learning activity data from an integrated learning platform called LEAF for modeling. First, we create a network-based knowledge model. Then, we construct learner models associated with the knowledge model. Based on the knowledge and learner modeling, we propose a method to find problems of a learner based on the order of closeness centrality of knowledge nodes. Also, the system recommends potential peer helpers who can help with these problems. We present a scenario of physics learning at the high school level to explain the practical use of this method which is aimed to enhance learners' initiative during peer help.

Scopus

Language：English   Publisher：31st International Conference on Computers in Education Icce 2023 Proceedings  

This study explored the possibility of teaching analytics utilizing daily learning log data recorded in xAPI format for class activity visualization. A junior high school reading activity designed to utilize several ICT tools distributed among multiple platforms was visualized. The visualized class activity was shown to a learning designer of the reading activity unit and then asked what she realized when revising the class activity design. As a result, we found that the integrated learning logs processed from the xAPI could visualize the differences in the actual activities compared with the lesson plan and how active the learners were. In addition, when the learning designer saw this visualization, she expressed her desire to change the activity design to reduce the transition of activities during class. Based on these results, we concluded that the daily learning log data recorded in the xAPI could visualize the time and content of the activities performed by the learners. The results show the possibility of capturing and visualizing the progress of a class by cross-pla|tform analysis using xAPI instead of using multiple special sensors.

Scopus

Language：English   Publisher：31st International Conference on Computers in Education Icce 2023 Proceedings  

Peer evaluation of individual or group work is often adopted in team-based learning design. However, some raters may not take the evaluation process seriously and exhibit behaviors such as using the same score, rushing through evaluations, or not evaluating during the presentation. This study investigates the issue of unserious peer evaluation in group presentations, focusing on their behavior patterns. Using evaluation behavior analysis indicators, we identified unserious raters who exhibited low reliability in the peer evaluation process. Further, we conducted a preliminary analysis to detect unserious raters based on learner model data available before the peer evaluation process. This information can assist teachers in providing personalized prompts and interventions prior to the peer evaluation process, thus enhancing the evaluation quality of these students with timely prompts to them.

Scopus

Language：English   Publisher：Educational Technology and Society  

Data-driven platforms with rich data and learning analytics applications provide immense opportunities to support collaborative learning such as algorithmic group formation systems based on learning logs. However, teachers can still get overwhelmed since they have to manually set the parameters to create groups and it takes time to understand the meaning of each indicator. Therefore, it is imperative to explore predictive indicators for algorithmic group formation to release teachers from the dilemma with explainable group formation indicators and recommended settings based on group work purposes. Employing learning logs of group work from a reading-based university course, this study examines how learner indicators from different dimensions before the group work connect to the subsequent group work processes and consequences attributes through correlation analysis. Results find that the reading engagement and previous peer ratings can reveal individual achievement of the group work, and a homogeneous grouping strategy based on reading annotations and previous group work experience can predict desirable group performance for this learning context. In addition, it also proposes the potential of automatic group formation with recommended parameter settings that leverage the results of predictive indicators.

DOI： 10.30191/ETS.202310_26(4).0006

Web of Science

Scopus

Language：English   Publisher：The Japan Association of College English Teachers  

Language teaching has a rich research history. A research discipline of Computer-Assisted Language Learning (CALL) has focused on technology integration in that practice. However, integrating the learning logs to create a data-driven workflow for the teachers and students is still limited. We design a technology framework called LEAF (Learning and Evidence Analytics Framework) to integrate daily evidence-based educational practices. The datadriven learning tools integrated into LEAF were implemented in actual live classrooms across multiple universities and schools within Japan. In this paper, we discussed the teaching and learning practices with the LEAF system in a language learning context and its impact. We highlight how to use LEAF for active reading, recommendation-based vocabulary learning, self-directed language learning approaches, and group work for language learning and teaching, specifically in English classes. Moving ahead, we aim to have an evidence-driven approach where the technology can continuously support and update best practices by analyzing the log data gathered continuously from the real-world educational setting.

DOI： 10.50943/jacetselectedpapers.9.0_15

CiNii Research

Language：English   Publisher：30th International Conference on Computers in Education Conference Icce 2022 Proceedings  

Learning Analytics research provides findings and analytical methods. However, those have not been frequently shared and reused in other studies. The objective of this study is to clarify the possibilities and challenges in applying a behavioral analysis method to other system contexts. This study uses the Evaluation Behavior Analysis (EBA) as an example of an analytical method, and compared the applicability of the method and reproducibility of findings in two different studies: the original dataset (Study A), and which applied EBA (Study B). Not all methods were able to be applied as the data and activity were different. However, the reproducibility was far higher than we expected. This research contributes to expanding the reuse of analysis methods for small-scale systems such as EBA, which have not usually been reused, and further development in Learning Analytics.

Scopus

Language：English   Publisher：30th International Conference on Computers in Education Conference Icce 2022 Proceedings  

Group work activities can promote the interpersonal skills of learners. To support the teachers in facilitating such activities, we suggested a learning analytics-enhanced technology framework, Group Learning Orchestration Based on Evidence (GLOBE) with data-driven approaches. We designed and implemented a group formation system using genetic algorithms to form groups using learning log data. Even if there is no existing data, we presented a paradigm of continuous data-driven support for the whole group learning process, incorporating the peer and teacher evaluation results as input to subsequent groupings. Further, utilizing accumulated group learning evidence in such an ecosystem, we aim to explore predictive group formation indicators which can lead to automatic group formation based on teachers' purpose in different contexts for desirable performance in subsequent group learning phases.

Scopus

Language：English   Publisher：30th International Conference on Computers in Education Conference Icce 2022 Proceedings  

Collaborative learning in online context has been in educational practice and also gained attention during the emergency remote teaching due to the pandemic. This study follows the Group Learning Orchestration Based on Evidence (GLOBE) framework that considers four phases of technology-supported group work and proposes data-driven services for each phase. In this paper, we present an analysis tool named "GWpulse'' that processes learning log data and shares the analysis results as an attribute of the learner model considered in GLOBE and also presents it in the group work dashboard to facilitate group awareness. In the current version GWpulse considered logs of forum activity generated during group activities in the learning management system. Apart from including that data in the learner model, it also visualizes the activity data for the orchestration, evaluation and reflection phases. GWpulse dashboard displays the basic statistics of their students' forum activities such as time interval and the number of posts, as well as a novel indicator of "Assistance Needed Level" calculated using sentiment analysis method that classifies textual statements into positive and negative.

Scopus

Language：English   Publisher：30th International Conference on Computers in Education Conference Icce 2022 Proceedings  

Using digital systems to group students according to their indicators provides opportunities for better group work implementation. However, how these indicators can affect group work performance remains unclear. Teachers tend to feel confused about which indicators should be considered when creating groups using learning log data. Capitalized on the data-driven environment under GLOBE, we conducted a preliminary study to explore predictive indicators for algorithmic group formation in a reading-based group learning context. This study presented our effort to explore the key factors that correlated to a desirable group work via factor analysis and correlation analysis. We found that reading engagement and previous peer rating scores suggest a higher potential to predict desirable group work performance in the reading-based online group work, which aims to help teachers set appropriately in future student model data-based group formation.

Scopus

Language：English   Publisher：Informa UK Limited  

In-class group work activities are found to promote the interpersonal skills of learners. To support the teachers in facilitating such activities, we designed a learning analytics-enhanced technology framework, Group Learning Orchestration Based on Evidence (GLOBE) using data-driven approaches. In this study, we implemented the algorithmic group formation and group work evaluation systems in a Japanese junior high school context. Data from a series of 12 collaborative learning activities were used to validate the difference in the measured heterogeneity of the formed homogeneous and heterogeneous groups compared to random grouping. Further, the peer rating and self-perception of the group work were compared for comparative reading and idea exchange tasks. We found algorithmically formed groups, considering the learner model data, either heterogeneously or homogeneously performed better than random grouping. Specifically, students in groups created by the homogeneous algorithm received higher peer ratings and more positive self-perception of group work in the idea exchange group tasks. We did not find significant differences in the comparative reading tasks. Along with the empirical findings, this work presents a paradigm of continuous data-driven group learning support by incorporating the peer and teacher evaluation scores as an input to the subsequent algorithmic grouping.

DOI： 10.1080/10494820.2022.2121730

Web of Science

Scopus

CiNii Research

Language：English   Publisher：Library and Information Service  

[Purpose/Significance] From the perspective of knowledge organization, this study explores the influence of two different organization forms on the reading effect of learning materials, and provides references for the research and application of knowledge representation in subsequent reading materials. [Method/Process] Using a between-group experiment design, 26 subjects participated two types of reading tests, a reading search task and a memory task, respectively. The differences were compared by reading time, the number of correct questions and subjective preference of subjects. [Result/Conclusion] The results show that there is no significant difference in the performance of the subjects in the search task, but in the memory task, the reading effect of tree-structure organization form is significantly better than traditional linear paragraphs. In addition, the merits of the tree-structured materials in the evaluation of subjects’ subjective preferences are more, such as clear logic, easy to remember, and so on.

DOI： 10.13266/j.issn.0252-3116.2022.08.002

Scopus

Language：English   Publisher：Proceedings of International Conference of the Learning Sciences Icls  

Conducting group work often needs careful scripting and alignment to the learning objectives for a particular class. However, there are challenges such as anonymity among participants when the group work goes online. To facilitate the orchestration of such group work, a learning analytics (LA) enhanced approach is described in this work. Potential for future research related to cultivating peer evaluation skills based on the platform is also highlighted.

Scopus

Language：English   Publisher：Lecture Notes in Computer Science Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics  

Evaluation plays a substantial role in group work implementation and peer evaluation gets prevalent with increasing flipped learning scenarios and online evaluation platforms. The accuracy of peer evaluation remains contingent in group work practice thus eliciting relevant studies on grader reliability. In this study, we present a data-driven approach to solving this issue utilizing learner models. On the one hand, we use previous learning logs to estimate and visualize the grader reliability in group work evaluation sessions as “peer evaluation potential”, which is used to align peer rating accuracy. On the other hand, leveraging reliability indicators created in the current session, learner models can be updated with new dimensions for subsequent usage. In addition, a case study in a high school English class was presented to examine this data-driven workflow and the results suggest the estimated peer evaluation potential correlates with the deviation from average peer judgment. Further potentials to cultivate peer evaluation-related capabilities are proposed as well.

DOI： 10.1007/978-3-031-20218-6_20

Web of Science

Scopus