Student ability to succeed in collaborative problem solving (CPS) is increasingly important. However, identifying aspects of CPS that individuals can act on to improve it remains a challenge. This study explores listening behaviour as a lever for effective CPS, given that listening is at once cognitive and social, and that individuals can enact it. Despite its importance in communication, listening is rarely systematically examined in CPS. To address this gap, we propose a framework for identifying listening behaviours in the text of student exchanges, and apply it to analyse patterns of listening behaviours of 34 K-12 students aged 11-14, working in nine groups on CPS activities. By combining content analysis, k-means cluster analysis, and correlation-based coupling network analysis, we identify ten distinct patterns of listening behaviours and their coupling over time, across groups with different success outcomes. We found that listening patterns varied by performance, under the assumption of moderate temporal dependence among interaction segments. Higher performing groups engaged in more counterarguments and constructive listening. Groups with lower CPS successes exhibited two ineffective patterns: questioning with instrumental listening, and counterarguments and questioning without encouraging listening. These findings pose questions about the relationship between listening and learning processes and have implications for multimodal research in learning analytics.

Learner data have become foundational to the feedback and personalisation functions of modern educational technologies. Decisions to adopt and use data-driven educational technologies, in part, depend on learner privacy perceptions related to data sharing. Some privacy theorists posit that these perceptions are shaped by individual characteristics, while others argue that they are driven by contextual details. In learning analytics (LA), little work has systematically examined whether learner characteristics, such as demographics and privacy concerns, predict student perceptions of LA-driven educational technologies and to what extent contextual factors also play a role. To investigate this, we conducted a vignette-based experiment (N = 256) asking students to evaluate acceptability and intended personal use of data-driven educational technologies across systematically manipulated scenarios. Our analysis showed that demographics and privacy concerns predicted adoption-related perceptions, and that individual and contextual factors explained comparable variance. The study quantifies the contributions of individual and contextual factors to adoption-related perceptions showing them to be similar. These findings imply that LA policies need to embed participatory, context-aware processes to help learners interrogate context and revise their data-sharing and adoption decisions accordingly. In addition, LA tools need to offer customisation of data-sharing choices to address diverse dispositions and context-specific choices.

With the rapid advancement of large language models (LLMs), synthetic agents are increasingly proposed as an alternative to human participants in mixed-method studies, yet their reliability for context-sensitive judgement tasks, such as privacy-related data-sharing decisions in learning technologies, remains understudied. In this paper, we assess how closely synthetic agents reproduce human data-sharing judgements in a multi-phase educational experiment, comparing human ratings with synthetic agent responses across experimental phases, conditions, and agent configurations. Using four LLMs (GPT-4.1-mini, Gemini~2.5 Flash, Mistral-Large-Latest, DeepSeek-V3.2), we evaluate alignment through mean absolute error, Jensen--Shannon distance, and Spearman rank correlation. We find that agents preserve the relative ordering of conditions observed in human judgements, while absolute and distributional agreement varies across contexts. Informed agents, conditioned on initial human responses, tend to align more closely with human ratings than uninformed agents, but this advantage is model-dependent and does not uniformly hold across outcomes and conditions. All agents struggle most with group deliberation, where both absolute and distributional divergence are the highest. Our paper suggests that robust methods for simulating variability and context-dependence in synthetic personas remain needed.

Mounting evidence indicates that modern citation practices contribute to inequalities in who receives citations. In response to this evidence, our paper investigates citation practices in learning analytics (LA). We analyse citations in papers published over ten years at the Learning Analytics and Knowledge conference (LAK). Our analysis examines the gender composition of authored and cited papers in LA, estimating various factors that explain why one paper cites another, and if the citation rates differ across different author teams. Results indicate an overall increase in the number of women authors at LAK, while the ratio of men to women remains stable. Citation patterns in LAK are influenced by the seniority of authors, paper age, topic, and team size. We found that LAK papers with women as the last author are under-cited, but papers where the first author is a woman and the last author is a man are over-cited. Author teams with different gender composition also vary in who they over- and under-cite. Upon presenting the empirical results, the paper reflects on the role of mindful citation practices and reviews existing measures proposed to promote diversity in citations.

Content analysis (CA) is a method frequently used in the learning sciences and so increasingly applied in learning analytics (LA). Despite this ubiquity, CA is a subtle method, with many complexities and decision points affecting the outcomes it generates. Although appearing to be a neutral quantitative approach, coding CA constructs requires an attention to decision making and context that aligns it with a more subjective, qualitative interpretation of data. Despite these challenges, we increasingly see the labels in CA-derived datasets used as training sets for machine learning (ML) methods in LA. However, the scarcity of widely shareable datasets means research groups usually work independently to generate labelled data, with few attempts made to compare practice and results across groups. A risk is emerging that different groups are coding constructs in different ways, leading to results that will not prove replicable. We report on two replication studies using a previously reported construct. A failure to achieve high inter-rater reliability suggests that coding of this scheme is not currently replicable across different research groups. We point to potential dangers in this result for those who would use ML to automate the detection of various educationally relevant constructs in LA.

Learning analytics approaches to profiling students based on their study behaviour remain limited in how they integrate temporality and change. To advance this area of work, the current study examines profiles of change in student study behaviour in a blended undergraduate engineering course. The study is conceptualised through complex dynamical systems theory and its applications in psychological and cognitive science research. Students were profiled based on the changes in their behaviour as observed in clickstream data. Measure of entropy in the recurrence of student behaviour was used to indicate the change of a student state, consistent with the evidence from cognitive sciences. Student trajectories of weekly entropy values were clustered to identify distinct profiles. Three patterns were identified: stable weekly study, steep changes in weekly study, and moderate changes in weekly study. The students with steep changes in their weekly study activity had lower exam grades and showed destabilisation of weekly behaviour earlier in the course. The study investigated the relationships between these profiles of change, student performance, and other approaches to learner profiling, such as self-reported measures of self-regulated learning, and profiles based on the sequences of learning actions.

Over the past few decades, universities and adult learning environments have undergone significant changes. Class sizes have grown. Many students opt to work, limiting their on-campus engagement. Class attendance is hybrid. Study programs become more learner-customised modifying the notion of a cohort. Learner-machine interactions are not only possible but also easy to scale. These developments are changing the communal nature of learning and the opportunities for relational processes among learners. However, promoting relational processes is crucial for student well-being, academic achievement, and social capital. In this talk, I will argue for the need to design and support relational processes and present directions for future work in this area. Drawing on my research in different learning environments, I will focus on the evidence about the presence of relational processes at scale, as well as on how teacher and student behaviour affect them. First, I will discuss relational processes in large online groups in MOOCs on edX, Coursera, and Twitter. Then, I will describe the patterns of relational processes in university online discussions in a cross-institutional study and explain how teacher decisions may affect these patterns. Finally, I will discuss the role of student social learning strategies by drawing on a study of mobile communication among adult learners. Throughout these examples, I will reflect on the implications for pedagogy, institutional effort required to modify instructor behaviour, and interventions needed to support student social learning strategies that can spark relational processes.

Peer effects, an influence that peers can have on one’s learning and development, have been shown to affect student achievement and attitudes. A large-scale analysis of social influences in digital online interactions showed that students interact in online university forums with peers of similar performance. Mechanisms driving this observed similarity remain unclear. To shed light as to why similar peers interact online, the current study examined the role of organizing factors in the formation of similarity patterns in online university forums, using four-years of forum interaction data of a university cohort. In the study, experiments randomized the timing of student activity, relationship between student activity levels within specific courses, and relationship between student activity and performance. Analysis suggests that similarity between students interacting online is shaped by implications of the course design on individual student behaviour, less so by social processes of selection. Social selection may drive observed similarity in later years of student experience, but its role is relatively small compared to other factors. The results highlight the need to consider what social influences are enacted by the course design and technological scaffolding of learner behaviour in online interactions, towards diversifying student social influences.

Asynchronous online discussions are broadly used to support peer interaction in online and hybrid courses. In this paper, we argue that the analysis of online peer interactions would benefit from the focus on relational events that are temporal and occur due to a range of factors. To demonstrate the possibility, we applied Relational Event Modeling (REM) to a dataset from online discussions in seven online classes. Informed by a conceptual model of social interaction in online discussions, this modeling included (a) a learner attribute capturing aspects of temporal participation, (b) social dynamics factors such as preferential attachment and reciprocity, and (c) turn-by-turn sequential patterns. Results showed that learner activity and familiarity from recent interactions affected their propensity to form ties. Turn-by-turn sequential patterns, that capture individual posting in bursts, explain how two-star network patterns form. Since two-star network patterns could further facilitate small group formation in the network, we expected the models to also capture communication in triads (i.e. triadic closure). Yet, models, devoid of the content of exchanges, did not capture the social dynamics well, and failed to predict patterns for communication across triads. By bringing in discourse features, future work can investigate the role of knowledge building behaviours in triadic closure of digital networks. This study contributes fresh insights into social interaction in online discussions, calls for attention to micro-level temporal patterns, and motivates future work to scaffold learner participation in similar contexts.

Network analysis has become a major approach for analysing social learning, used to capture learner positioning in online forum networks. LA research investigated the association between positioning in forum networks with academic performance and discourse quality, the latter two serving as proxies for learning. However, the research findings have been inconsistent, in part due to the discrepancies in the adopted approaches to network construction. Yet, it is still unclear how online forum networks should be modelled to assure that the learners' network positioning is properly captured. To address this gap, the current study explored if some existing approaches to network construction may complement each other and thus offer richer insights. In particular, we hypothesised that the post-reply learner network could represent interpersonal positioning, whereas the network based on co-participation in discussion threads could encapsulate intergroup positioning. The study used learner social interaction data from a large edX MOOC forum to examine the relationship between these two kinds of network positioning. The results suggest that intergroup and interpersonal positioning may capture different aspects of social learning, potentially related to different learning outcomes. We find that although interpersonal and intergroup positioning indicators covary, these measures are not congruent for some 37% of forum posters. Network coevolution analysis also reveals an interdependent relationship between the intergroup and interpersonal centrality in a forum network. Co-occurrence of learners in a discussion thread prior to direct exchanges is predictive of a direct post-reply interaction at a later stage of the course, and vice-versa, suggesting that intergroup positioning is a precursor of direct communication. The study contributes to the discussion around the definition of learner forum positioning in learning analytics, and validated approaches towards measuring it.

The mission of learning analytics (LA) is to improve learner experiences using the insights from digitally collected learner data. While some areas of LA are maturing, this is not consistent across all LA specialisations. For instance, LA for social learning lack validated approaches to account for the effects of cross-course variability in learner behavior. Although the associations between network structure and learning outcomes have been examined in the context of online forums, it remains unclear whether such associations represent bona fide social effects, or merely reflect heterogeneity in individual posting behavior, leading to seemingly complex but artefactual social network structures. We argue that to start addressing this issue, posting activity should be explicitly included and modelled in forum network representations. To gain insight to what extent learner degree and edge weight are merely derivatives of learner activity, we construct random models that control for the level of posting and post properties, such as popularity and thread hierarchy level. Analysis of forum networks in twenty online courses presented in this paper demonstrates that individual posting behavior is highly predictive of both the breadth (degree) and frequency (strength) in forum communication networks. This implies that, in the context of forum-based modelling, degree and frequency may not reflect the social dynamics. However, results suggest that clustering of the network structure is not a derivative of individual posting behaviour. Hence, weighted local clustering coefficient may be a better proxy for social relationships. The empirical results are relevant to scientists interested in social interactions and learner networks in digital learning, and more generally to researchers interested in deriving informative social network models from online forums.

Learning Analytics (LA) studies the learning process in order to optimize learning opportunities for students. Although LA has quickly risen to prominence, there remain questions regarding the impact LA has made to date. To evaluate the extent that LA has impacted our understanding of learning and produced insights that have been translated to mainstream practice or contributed to theory, we reviewed the research published in 2011-2018 LAK conferences and Journal of Learning Analytics. The reviewed studies were coded according to five dimensions: study focus, data types, purpose, institutional setting, and scale of research and implementation. The coding and subsequent epistemic network analysis indicates that while LA research has developed in the areas of focus and sophistication of analyses, the impact on practice, theory and frameworks have been limited. We hypothesize that this finding is due to a continuing predominance of small-scale techno-centric exploratory studies that to date have not fully accounted for the multi-disciplinarity that comprises education. For the field to reach its potential in understanding and optimizing learning and learning environments, there must be a purposeful shift to move from exploratory models to more holistic and integrative systems-level research. This necessitates greater effort applied to understanding the research cycles that emerge when multiple knowledge domains coalesce into new fields of research.

Despite strong interest in learning analytics (LA), adoption at a large-scale organizational level continues to be problematic. This may in part be due to the lack of acknowledgement of existing conceptual LA models to operationalize how key adoption dimensions interact to inform the realities of the implementation process. This paper proposes the framing of LA adoption in complexity leadership theory (CLT) to study the overarching system dynamics. The framing is empirically validated in a study analysing interviews with senior staff in Australian universities (n=32). The results were coded for several adoption dimensions including leadership, governance, staff development, and culture. The coded data were then analysed with latent class analysis. The results identified two classes of universities that either i) followed an instrumental approach to adoption - typically top-down leadership, large scale project with high technology focus yet demonstrating limited staff uptake; or ii) were characterized as emergent innovators - bottom up, strong consultation process, but with subsequent challenges in communicating and scaling up innovations. The results suggest there is a need to broaden the focus of research in LA adoption models to move on from small-scale course/program levels to a more holistic and complex organizational level.

There has been a growing trend in higher education towards increased use and adoption of Massive Open Online Courses (MOOCs). Despite this interest in learning at scale, limited work has compared MOOC activity across subsequent course offerings. In this study, we explore forum activity in ten iterations of the same MOOC. Our results suggest that participation in MOOC forums has changed over the past four years of delivery. First, overall participation in MOOC forums have decreased. Second, in later iterations cohorts of more committed forum users start to resemble formal online courses in size (67>n>36). However, despite the smaller groups of learners that should find it easier to form connections with one another, our analysis did not reveal the expected increase in the quality of social activity. Instead, MOOC forums evolved into smaller on-task question and answer (Q&A) spaces, not capitalizing on the opportunities for social learning. We discuss practical and research implications of such changes.

Video materials have become an integral part of university learning and teaching practice. While empirical research concerning the use of videos for educational purposes has increased, the literature lacks an overview of the specific effects of videos on diverse learning outcomes. To address such a gap, this paper presents preliminary results of a large-scale systematic review of peer-reviewed empirical studies published from 2007-2017. The study synthesizes the trends observed through the analysis of 178 papers selected from the screening of 2531 abstracts. The findings summarize the effects of manipulating video presentation, content and tasks on learning outcomes, such as recall, transfer, academic achievement, among others. The study points out the gap between large-scale analysis of fine-grained data on video interaction and experimental findings reliant on established psychological instruments. Narrowing this gap is suggested as the future direction for the research on video-based learning.

In this poster, we present the results of the study which examined the relationship between student differences in their use of the available technology and their perceived levels of cognitive presence within the MOOC context. The cognitive presence is a construct used to measure the level of practical inquiry in the Communities of Inquiry model. Our results revealed the existence of three clusters based on student technology use. The clusters significantly differed in terms of their levels of cognitive presence, most notably they differed on the levels of problem resolution.

The facilitation of interpersonal relationships within a respectful learning climate is an important aspect of teaching practice. However, in large-scale online contexts, such as MOOCs, the number of learners and highly asynchronous nature militates against the development of a sense of belonging and dyadic trust. Given these challenges, instead of conventional instruments that reflect learners' affective perceptions, we suggest a set of indicators that can be used to evaluate social activity in relation to the participation structure. These group-level indicators can then help teachers to gain insights into the evolution of social activity shaped by their facilitation choices. For this study, group-level indicators were derived from measuring information exchange activity between the returning MOOC posters. By conceptualizing this group as an identity-based community, we can apply exponential random graph modelling to explain the network's structure through the configurations of direct reciprocity, triadic-level exchange, and the effect of participants demonstrating super-posting behavior. The findings provide novel insights into network amplification, and highlight the differences between the courses with different facilitation strategies. Direct reciprocation was characteristic of non-facilitated groups. Exchange at the level of triads was more prominent in highly facilitated online communities with instructor's involvement. Super-posting activity was less pronounced in networks with higher triadic exchange, and more pronounced in networks with higher direct reciprocity.

Research in formal education has repeatedly offered evidence of the importance of social interactions for student learning. However, it remains unclear whether the development of such interpersonal relationships has the same influence on learning in the context of large-scale open online learning. For instance, in MOOCs group members frequently change and the volume of interactions can quickly amass to chaos, therefore impeding an individual's propensity to foster meaningful relationships. This paper examined a MOOC for its potential to develop social processes. As it is exceedingly difficult to establish a relationship with somebody who seldom accesses a MOOC discussion, we singled out a cohort defined by its participants' regularity of forum presence. The study, analysed this 'cohort' and its development, in comparison to the entire MOOC learner network. Mixed methods of social network analysis (SNA), content analysis and statistical network modelling, revealed the potential for unfolding social processes among a more persistent group of learners in the MOOC setting.