General Call

We welcome submissions from both research and practice, encompassing different theoretical, methodological, empirical, and technical contributions to the learning analytics field. Learning analytics research draws on many distinct academic fields, including psychology, the learning sciences, education, neuroscience, computer science, and design. We encourage the submission of work conducted in any of these traditions, as long as it is done rigorously. We also welcome research that validates, replicates, and examines the generalizability of previously published findings, as well as examines aspects of the adoption of existing learning analytics methods and approaches. 

Specifically, this year, in line with the conference main theme of Learning Analytics and AI Synergy, we encourage authors to address some of the following topics in their submissions:

• Understanding learning and teaching with AI: Studies that use learning analytics methods and techniques to analyze and understand AI-augmented teaching and/or learning.
• AI-supported educational decision-making: Studies that examine the use of AI for automating analysis of learning-related data to provide insight, predictions, and recommendations that enhance decision-making for teachers, learners, or other educational stakeholders.
• Using AI to empower learning analytics process: Studies that examine and demonstrate how AI methods and technologies can be employed to advance distinct phases of the learning analytics process, from data collection, transformation, and analysis to the interpretation and communication of the analytics results.
• Amplifying real-time learning analytics: Studies that explore how AI facilitates real-time analytics to support adaptive learning environments.
• Scaling learning analytics interventions through AI: Studies that examine how AI could be used to increase the scale of learning analytics-informed educational interventions, including personalisation.
• Expanding the theoretical perspectives and methodological approaches: Studies that introduce novel theoretical frameworks and methodological innovations - including the usage and generation of new types of learning data - to advance the synergy between learning analytics and AI.
• Ensuring ethical and responsible AI use in education: Studies that use learning analytics to examine ethical and responsible deployment of AI technologies in various learning and teaching settings.

Disclaimer: The authors are not required to address the conference theme, and papers for acceptance will not be considered based on their fit to the theme. The purpose of every annual conference theme is to encourage discussions and debate on emerging topics and issues in the field.

Usual topics of interest include, but are not limited to, the following:

Tracing Learning & Teaching:

• Finding evidence of learning: Studies that identify and explain useful data for analyzing, understanding, and enhancing learning and teaching. This may include – but is not limited to – sampling frames; software interfaces for tracing learning and affective events; sensors for recording physiological indicators of learning and other processes; properties of archival data; utterances exchanged with chatbots; artificially intelligent agents for soliciting learners’ on-the-fly accounts of learning processes and affect/emotions; teacher observation methods; and experience sampling methods.
• Assessing student learning: Studies that assess learning progress through the computational analysis of learner actions or artifacts and social interactions with peers and teachers that might support the attainment of intended learning outcomes.
• Analytical and methodological approaches: Studies introducing analytical techniques, methods, and tools for modeling student learning to empower learners or other educational stakeholders and enhance learning. Methods could be based on artificial intelligence, machine learning, natural language processing, social network analysis, text mining, and knowledge tracing, among others, if targeted toward LA-oriented questions. These approaches include the use of multimodal learning analytics as well as flexible approaches to learning and teaching that support learning processes, learner achievements, and learner well-being.
• Technological infrastructures for data storage and sharing: Proposals of technical and methodological procedures to store, share, and preserve learning and teaching traces, taking appropriate privacy-preserving and ethical considerations into account, and involving feedback loops to stakeholders

Understanding Learning & Teaching:

• Data-informed learning theories: Proposals of new learning/teaching theories or revisions to / reinterpretations of / support for existing theories based on or related to large-scale data analysis.
• Insights into specific learning processes: Studies to understand particular aspects of a learning/teaching process through a learning-theory-informed use of data science methods and techniques, including negative results.
• Learning and teaching modeling: Creating mathematical, statistical, or computational models of a learning/teaching process, including its actors and context, when targeted toward learning analytics-oriented questions. Note that this contrasts with work focused on comparing algorithms or prediction models, which may be better targeted toward educational data mining (EDM).
• Systematic and meta reviews: Studies that provide a systematic and methodological synthesis of the available evidence in learning analytics.

Impacting Learning & Teaching:

• Human-centered design processes: Research that documents practices of giving learners, teachers, and other educational stakeholders an active voice in the design process of learning analytics initiatives and enabling technologies.
• Providing decision support and feedback: Studies that evaluate the use and impact of feedback or decision-support systems based on learning analytics (dashboards, early-alert systems, automated messages, etc.). This includes data visualization and data storytelling with learning-related data.
• Data-informed decision-making: Studies that examine how teachers, students or other educational stakeholders come to, work with, and make changes using learning analytics information. It also includes action research and qualitative investigation of how stakeholders understand and utilize data in educational contexts.
• Personalized and adaptive learning: Studies that evaluate the effectiveness and impact of adaptive technologies based on learning analytics. 
• Practical evaluations of learning analytics efforts: Empirical evidence about the effectiveness of learning analytics implementations or educational initiatives guided by learning analytics.

Implementing Change in Learning & Teaching:

• Ethical issues around learning analytics: Analysis of issues and approaches to the lawful and ethical capture, protection, and use of educational data traces, including data privacy; tackling unintended bias and value judgements in selecting data and algorithms; perspectives and methods that empower stakeholders.
• Equity, fairness, and transparency in learning analytics: Consideration of how certain practices of data collection, analysis, and subsequent action impact particular populations and affect human well-being, specifically groups that have been previously disadvantaged. Discussions of how learning analytics may impact (positively or negatively) social change and transformative social justice.
• Learning analytics adoption: Discussions and evaluations of strategies to promote and embed learning analytics initiatives in educational institutions and learning organizations. Studies that examine organizational change processes and professional development practices that support impactful learning analytics use.
• Learning analytics strategies for scalability: Discussions and evaluations of strategies for scaling the capture and analysis of information in useful and ethical ways at the program, institution or national level; critical reflections on organizational structures that promote analytics innovation and impact in an institution.
• Strategic planning of learning & teaching: Consideration and application of trustworthy learning analytics in all phases of strategic planning, including needs analysis for change, decision-making, implementation and monitoring, as well as evaluation of strategic decisions, taking into account the complexity and context of the strategic decision-making domain.

Out of Scope for Learning Analytics

• Work focused solely on aspects of methodology and data manipulation that does not discuss, work towards, or include closing the loop to stakeholders, for example, by empowering learners, teachers, or other educational stakeholders.
• Work without any data or evidence that relates to learners, educators, or learning processes.
• Work introducing analytical techniques, methods, and tools without connecting to learners or learning processes.
• Work focused on comparing different approaches to predictive modelling or algorithm accuracy will likely be a better fit for educational data mining (EDM).
• Work that evaluates the effectiveness and impact of adaptive/personalised learning technologies that are not explicitly grounded in learning analytics.
• Classic learning sciences research is better suited for the International Society of the Learning Sciences (ISLS) or similar venues in the learning sciences. This includes quantitative, qualitative and mixed-methods studies that use data related to education and learning but do not make strong connections to learning analytics in their research questions or findings.
• Work that focuses on comparisons between departments, institutions, regions or countries, for example management dashboards or work using PISA data, is usually classed as academic analytics. 
• Studies involving artificial intelligence (AI) may be in or out of scope, depending on the context and intention of the work. For example, the contribution of work on how AI is deployed in primary classrooms might be a better fit for AI-Ed, but if the contribution is on how to use analytics in evaluating the effectiveness of AI-supported learning environments is in scope for learning analytics.
• Work without a sound theoretical foundation in learning theory or that does not use sound methodologies when collecting and analysing data. 
• Studies that do not situate themselves in relation to previous work in LA. This might involve building on, extending, supporting, or revising that work, or identifying a gap that should be filled. A paper that does not mention ‘learning analytics’ in its text or references is highly likely to be out of scope.

The conference has three distinct submission types, which are described below. Please see the submission guidelines page (coming soon) for information on paper format and other technical details for each track.

LAK26 will use a double-blind peer review process for all submissions except demos, leadership academy, and the doctoral consortium (which each require elements that prevent blinding).

To continue to strengthen the review process for both authors and reviewers, LAK26 will have a rebuttal phase for full and short research papers in which authors will be given seven days to flag any inaccuracies, omissions, or errors in the reviews in a maximum of 500 words. Please note that rebuttals are optional and should not be treated as responses to reviewer comments for revisions. Therefore, we urge authors to prepare and submit a rebuttal if and only if they note inaccuracies, misunderstandings, or other kinds of errors in the reviewers’ comments. In such cases, the authors are requested to list inaccuracies, omissions, and errors in reviewer comments first, and then reply to specific errors. When writing the reply, authors should keep in mind that papers are being evaluated as submitted and thus, the response should not propose new results or restructuring of the presentation.

Meta-reviewers, senior members of the research community, make final recommendations for paper acceptance or rejection with justification to the program committee chairs after the rebuttal phase is concluded. Acceptance decisions are ultimately taken by the program committee chairs based on all available information from the review process in combination with the constraints of the allowable space in the conference program.

Accepted full and short research papers will be included in the LAK26 conference proceedings published and archived by ACM (pending ACM approval). Other types of submissions (posters, demos, workshops, tutorials, practitioner reports and doctoral consortium) will be included in the open access LAK companion proceedings, published on SoLAR’s website. Please note at least one of the authors of each accepted submission must register for the conference by the Early Bird deadline in order for the paper to be included in the ACM or LAK Companion Proceedings.