Explore the vast range of titles available.

In healthcare education, it is important for nursing students to be able to reflect on their performance in high-fidelity clinical simulations in order to develop key skills. Learning Analytics (LA) offers opportunities for data-driven reflection by providing visual representations of educational experiences. While many LA tools rely on data visualisations to communicate insights, these are often difficult for students to interpret, limiting their effectiveness. Despite these challenges, there is limited research exploring alternative and potentially more accessible formats—such as data comics, a narrative visualisation technique that integrates data with the structure of traditional comic strips—to represent and communicate insights from learner data in a more engaging way. This study addresses that gap through a qualitative analysis of nursing students’ perceptions of data comics as reflective tools, focusing on: (i) support for student reflection, (ii) advantages and limitations, and (iii) concerns about their use in healthcare education. Third-year nursing students who participated in a simulation were interviewed and asked to reflect on personalised data comic prototypes generated from their multimodal data using a mix of human input and AI methods. The results indicated that while data comics present an engaging and accessible form of reflective visualisation, considerations need to be made regarding the designs to ensure that they are appropriate for the target audience and do not oversimplify the simulation experience. These findings indicate that data comics should not act as a replacement for conventional visualisations but rather serve as supplementary material to communicate contextual information or aid in interpretation of visualisations.

There is a growing interest in involving students and teachers in the design of human-centered Learning Analytics (LA) systems to align them with authentic learning needs. Yet, limited prior research has explored the implications of integrating both students’ and teachers’ perspectives within a structured co-design process. To address this shortcoming in the literature, we report on a study that examined how undergraduate nursing students and teachers co-designed an AI-powered LA system to support post-debriefing reflection on teamwork and communication in the context of healthcare simulation. This qualitative study, using a co-design approach, examined the design process of an LA system from conceptualization to post-use evaluation. The study addressed two key questions: i ) What tensions emerge from the contrasting perspectives of students and teachers in the co-design an AI-powered LA system? and ii ) How do students and teachers perceive their joint participation in the co-design process? Three key design tension themes emerged from the contrasting perspectives of students and teachers: teaching–learning goals tension , privacy–utility tension , and human-AI guidance preferences tension . The collaborative design process revealed mutual benefits: students valued teachers’ guidance in refining ideas and aligning system goals with learning objectives, while teachers, initially cautious about student involvement, came to see co-design as an opportunity to empower students and deepen their own understanding of responsible data use in practice. These findings contribute to the broader understanding of co-design dynamics in educational technology, underscoring the importance of balanced stakeholder involvement in developing practical, context-aware LA systems.

Teachers’ spatial behaviours in the classroom can strongly influence students’ engagement, motivation and other behaviours that shape their learning. However, classroom teaching behaviour is ephemeral, and has largely remained opaque to computational analysis. Inspired by the notion of Spatial Pedagogy, this paper presents a system called ‘Moodoo’ that automatically tracks and models how teachers make use of the classroom space by analysing indoor positioning traces. We illustrate the potential of the system through an authentic study with seven teachers enacting three distinct learning designs with more than 200 undergraduate students in the context of science education. The system automatically extracts spatial metrics (e.g. teacher-student ratios, frequency of visits to students’ personal spaces, presence in classroom spaces of interest, index of dispersion and entropy), mapping from the teachers’ low-level positioning data to higher-order spatial constructs. We illustrate how these spatial metrics can be used to generate a deeper understanding of how the pedagogical commitments embedded in the learning design, and personal teaching strategies, are reflected in the ways teachers use the learning space to provide support to students.

The ability to communicate, be an effective team or group member and collaborate face-to-face are critical skills for employability in the 21st century workplace. Previous research suggests that learning to collaborate effectively requires practice, awareness of group dynamics and reflection upon past activities. However, although having a teacher closely supervising and providing detailed feedback to each group would be ideal, it may be unrealistic in practice. A promising way to approach this challenge could be to capture behavioural traces from group interactions in order to generate comprehensible and actionable feedback to support team reflection. In this sense, Multimodal Learning Analytics (MMLA) is a promising field, offering the potential to track learners’ activity across digital and collocated contexts, using emerging sensing and pervasive computing technologies. Most of the research in MMLA has been conducted in lab conditions, to help researchers validate learning theories or generate more comprehensive learner models. However, one of the most underexplored aspects of MMLA has been the generation of feedback to support teaching and learning, and moreover, in authentic locations and activities.This thesis reports progress in tackling this challenge by designing and validating computer-based feedback, by means of visual representations and narrative, to support effective, guided reflection using multimodal learning analytics evidence. To achieve this, three contributions are presented. The first contribution is a human-centred design method to translate the informal outputs of codesign sessions with teachers and students, into more meaningful group work constructs with clear MMLA design requirements. The second contribution is a modelling approach to add meaning to low-level multimodal group data based on the characteristics of the context (domain expertise, theory, and the learning design). Finally, the third contribution is an approach for augmenting visual representations with data storytelling elements to facilitate the interpretation of group dynamics insights by educators and students. This thesis is developed in the context of two distinct, collocated group work settings, in the domains of collaborative database design and healthcare simulation. Using a Design-Based Research process, a set of explanatoryinterfaces (i.e. interfaces that communicate insights) was designed and validated with teachers and students. The thesis provides timely and necessary groundwork for researchers and practitioners to design visual representations capable of communicating actionable insights, using multimodal data in complex and authentic collaboration scenarios.

Sustainability assessments have gained considerable attention in the last decades with an increasing presence for decision-making, especially when multidisciplinary perspectives, environmental, economic, and social criteria are considered for process improvements and policy-making. However, evaluating sustainability is not straightforward, and the demand for quantifying such impacts for processes and products is critical to achieving sustainable development goals. This study focuses on quantifying and communicating the sustainability of cellulose-based materials by evaluating three different sustainability tools. First of all, a streamlined LCA for potential waste cotton application in an early stage of technology readiness. Second, a process-based simulation to generate life cycle inventory data was used to analyze impact assessments of manufacturing processes. Thirdly, a multi-criteria decision analysis tool was applied herein to select preferable alternatives integrating the three dimensions of sustainability: environment, economic and social.A literature review for end-of-life waste cotton highlighted a wide range of applications, conventional (e.g., regenerated cellulose) and non-conventional, such as biofuels and composites. This review shed light on potential new markets for waste cotton applications. In addition, further efforts to understand the environmental impacts/reductions of switching to waste cotton shows that there is a potential for mitigating climate change in products that can benefit from using waste cotton as raw material. Therefore, the results from this study illustrate that applying streamlined LCAs to analyze the use of waste cotton (considered free of environmental burdens) to reduce/replace virgin raw materials can reduce global warming potential. The three potential products studied were (1) pulp and paper applications and dissolving pulp, (2) chemicals such as succinic acid, and (3) insulation materials using waste cotton composites.In the next study, data gaps in the life cycle inventory of dissolving pulp production are addressed by applying a process-based simulation to quantify detailed mass and energy balances critical to carrying out any life-cycle impact analysis. Dissolving pulp is an interesting case in which the end-products applications are diverse. Dissolving pulp grades (e.g., acetates and viscose) can be produced from different biomass such as hardwoods and softwoods. The results show that the biomass feedstock directly affects the environmental impacts of dissolving pulp production. For instance, hardwood acetate grade has a higher global warming potential (1,010 kg CO2 eq./ADmt) than softwood acetate (860 kg CO2 eq./ADmt ). Nonetheless, hardwood acetate has lower environmental impacts in other categories related to ecosystems and human health. Additionally, a hotspot analysis identified that on-site emissions and chemicals are the main contributors to the environmental impacts across the grades of pulp studied herein. Furthermore, the results and life cycle inventory data generated in this research provide critical information to support future sustainability assessment for end-products derived from dissolving pulp. Finally, a methodology integrating the three pillars of sustainability using multi-criteria decision analysis to understand the preference of products/alternatives (e.g., bio-based hemp, wood, and regenerated cellulose vs. fossil-based polypropylene) for decision making is modeled. In this case, a model for stochastic multi-attribute analysis using pairwise outranking normalization is built, evaluated, and applied to wet wipes production. In addition, an indicator for litter accumulation based on the water and soil biodegradability of materials is proposed. The study determined the preference among different raw materials; wood pulp, hemp, viscose, and polypropylene for wet-wipes production. The results showed that wood pulp has the highest probability of ranking first 74%, followed by hemp with 19%, viscose with 3%, and polypropylene with 5%, respectively. The inclusion of a litter score caused the nonbiodegradable polypropylene to have lower preferability relative to when a litter score was not included, as expected. The results from this research shed light on feedstock selection, considering all the aspects of sustainability to aid decision-making with multi-variable criteria.In summary, the assessment tools applied to cellulose-based materials from this research provide an insightful evaluation to effectively quantify environmental impacts at an early stage of development, a more robust detailed unit-operation based modeling process for more granularity, and overall integrated sustainability analysis that is holistic and objective, providing insights that can enhance future decision-making by stakeholders and policymakers.

Data storytelling (DS) is rapidly gaining attention as an approach that integrates data, visuals, and narratives to create data stories that can help a particular audience to comprehend the key messages underscored by the data with enhanced efficiency and effectiveness. It has been posited that DS can be especially advantageous for audiences with limited visualisation literacy, by presenting the data clearly and concisely. However, empirical studies confirming whether data stories indeed provide these benefits over conventional data visualisations are scarce. To bridge this gap, we conducted a study with 103 participants to determine whether DS indeed improve both efficiency and effectiveness in tasks related to information retrieval and insights comprehension. Our findings suggest that data stories do improve the efficiency of comprehension tasks, as well as the effectiveness of comprehension tasks that involve a single insight compared with conventional visualisations. Interestingly, these benefits were not associated with participants' visualisation literacy.

The rapid expansion of Learning Analytics (LA) and Artificial Intelligence in Education (AIED) offers new scalable, data-intensive systems but also raises concerns about data privacy and agency. Excluding stakeholders -- like students and teachers -- from the design process can potentially lead to mistrust and inadequately aligned tools. Despite a shift towards human-centred design in recent LA and AIED research, there remain gaps in our understanding of the importance of human control, safety, reliability, and trustworthiness in the design and implementation of these systems. We conducted a systematic literature review to explore these concerns and gaps. We analysed 108 papers to provide insights about i) the current state of human-centred LA/AIED research; ii) the extent to which educational stakeholders have contributed to the design process of human-centred LA/AIED systems; iii) the current balance between human control and computer automation of such systems; and iv) the extent to which safety, reliability and trustworthiness have been considered in the literature. Results indicate some consideration of human control in LA/AIED system design, but limited end-user involvement in actual design. Based on these findings, we recommend: 1) carefully balancing stakeholders' involvement in designing and deploying LA/AIED systems throughout all design phases, 2) actively involving target end-users, especially students, to delineate the balance between human control and automation, and 3) exploring safety, reliability, and trustworthiness as principles in future human-centred LA/AIED systems.

The goal of this research is to assess the environmental impacts of Peracetic Acid (PAA) used as a bleaching agent for bleached kraft pulp production. Pulp bleaching is one of the largest sources of wastewater generation in the pulp and paper industry. PAA is a promising bleaching agent that can be used in the Total Chlorine Free (TCF) bleaching technology to eliminate AOX in effluents, to reduce water consumption and wastewater discharge. Although PAA is considered as an environmentally friendly bleaching agent, few studies have quantified the overall environmental impacts of PAA production and its application as a bleaching agent in TCF compared to the traditional Elemental Chlorine Free (ECF) bleaching technology. In this study, a comparative Life Cycle Assessment (LCA) of kraft pulp production was performed to compare the environmental impacts of pulp bleached by TCF using PAA and by ECF technology using ClO2. The environmental impacts of PAA produced from two different pathways (i.e., acetic acid pathway and triacetin pathway) were assessed. Life Cycle Inventory (LCI) data was collected from lab experiments using PAA (achieving an average pulp ISO brightness of 84%), process simulation tools (e.g. WinGEMS, Aspen Plus) and LCI databases including Ecoinvent. Nine environmental impacts categories were included in the Life Cycle Impact Assessment analysis (LCIA) and a hotspot analysis was performed to identify the major life cycle stages contributing to the environmental impacts results. Based on the results, compared to a traditional ECF technology, TCF bleaching using PAA produced from the triacetin pathway shows lower environmental impacts in three categories and higher environmental impacts in six categories. Moreover, the TCF bleaching using PAA from the triacetin pathway eliminates AOX from effluents, reduces energy consumption (14%), water consumption (14%) wastewater discharge (16%) and make-up chemicals such as sodium hydroxide (18%) and sodium sulfate (13%). Finally, the hotspot analysis showed that the upstream environmental impacts of the bleaching chemicals were the major processes driving the environmental impacts.

Healthcare simulations help learners develop teamwork and clinical skills in a risk-free setting, promoting reflection on real-world practices through structured debriefs. However, despite video's potential, it is hard to use, leaving a gap in providing concise, data-driven summaries for supporting effective debriefing. Addressing this, we present TeamVision, an AI-powered multimodal learning analytics (MMLA) system that captures voice presence, automated transcriptions, body rotation, and positioning data, offering educators a dashboard to guide debriefs immediately after simulations. We conducted an in-the-wild study with 56 teams (221 students) and recorded debriefs led by six teachers using TeamVision. Follow-up interviews with 15 students and five teachers explored perceptions of its usefulness, accuracy, and trustworthiness. This paper examines: i) how TeamVision was used in debriefing, ii) what educators found valuable and challenging, and iii) perceptions of its effectiveness. Results suggest TeamVision enables flexible debriefing and highlights the challenges and implications of using AI-powered systems in healthcare simulation.

Research shows that dialogue, the interactive process through which participants articulate their thinking, plays a central role in constructing shared understanding, coordinating action, and shaping learning outcomes in teams. Analysing dialogue content has been central to advancing team learning theory and informing the design of computer-supported collaborative learning environments, yet this progress has depended on labour-intensive qualitative coding. LLMs offer new possibilities for automating and enhancing the dialogue layer within emerging multimodal learning analytics approaches, with recent studies showing that they can approximate human coding through few-shot prompting. However, prior work has focused on replicating human coding accuracy for research purposes, rather than addressing a more educationally consequential question: how can we design prompts that allow an LLM to label team dialogue accurately and fast enough to be useful in real settings, such as in-person healthcare simulations, where results must be returned quickly and computational cost and sustainability also matter? This paper investigates how prompt design and batching strategies can be optimised to balance coding accuracy, processing time, and environmental impact in team-based healthcare simulation debriefing. Using a dataset of 11,647 utterances coded across 6 dialogue constructs, we compared 4 prompt designs across varying batch sizes, evaluating coding performance, processing time, and energy consumption, as well as the trade-offs between these metrics. Results indicate that increasing batch size improves speed and reduces energy use, but negatively impacts coding performance. Beyond demonstrating the feasibility of LLM-based qualitative analysis, this study offers practical guidance for scaling dialogue analytics in contexts where timeliness, privacy, and sustainability are critical.