Possible grand challenges

In the future, learning analytics…

Visions abstracted from the two videos made for LAK25.

·      make education better, more just, and more rewarding

·      increase understanding of the private and public benefits of education

·      increase understanding of how education contributes to our communities

·      are implemented in school (K–12) settings

·      are available in easy-to-use toolboxes

·      bring stakeholders together

·      are trusted by stakeholders

·      make effective use of (generative) AI

·      support academic success

·      are used by students and educators to support reflection

·      support real-time adjustments of learning and teaching

·      offer a more personalised approach to learning

·      support collaborative learning

·      clearly demonstrate the value they add

·      improve the student experience

·      are used to target the provision of social services and help

·      suggest strategies to make student life more enjoyable

·      emphasise human interaction

·      support care for the whole student, including mental / physical wellbeing

·      support responses to the climate crisis

·      are protected against cyber-crime attacks

·      are aligned with international data standards, making them interoperable

·      are designed to protect student data – including from political misuse

Grand challenges: what the literature suggests

These have been rephrased as visions to work towards and they have been grouped for clarity. References to the original challenges are provided.

Pedagogical challenges

·      Learning analytics harness rich, long-term data for self-knowledge and shared understanding [1]

·      Learning analytics act as a catalyst for learning and for designing learning [2]

·      Learning analytics are used to advance personalised learning [3]

·      Learning analytics provide invaluable support to self-regulated learning [4]

Research challenges

·      Learning analytics are part of a larger research endeavour to understand if and when technology supports learning [1]

·      Learning analytics are used with remote learning and AI to scale present capabilities [5]

·      Learning analytics are used to reflect on and reimagine the learning experiences of students [5]

Infrastructure challenges

·      Learning analytics are part of learning environments that feel safe, trusted, scrutable and controllable [1]

·      The privacy and ethical aspects of using learning analytics are known and acted upon [2]

·      We know what infrastructure is needed to make a start with learning analytics [2]

·      We know how to extract knowledge from rich data sets and to integrate these understandings into a coherent picture of students, campuses, instructors and curricular designs [6]

·      Learning analytics are interoperable and their findings are transferable [7]

·      We know how to collect data that is sufficiently granular to be meaningful yet is completely unidentifiable [8]

·      New techniques can capture multimodal learning data [9]

·      Learning analytics are elements of smart classrooms that seamlessly fit into teaching and learning [1]

Technical challenges

·      Learning analytics are built into long-term personal digital learning companions [1]

·      Learning analytics make predictions rapidly and with high reliability [10]

·      Learning analytics identify who needs which interventions, clearly improving the success of the students who receive interventions [7]

·      The rationale behind learning analytics interventions is sufficiently clear that it can be queried and its trustworthiness assessed [7]

·      Learning analytics are reliably transferred between settings [7]

·      Learning analytics solve generalisable problems [7]

Classroom challenges

·      We know what requirements a dashboard for teachers must meet [2]

·      Learning analytics support student progression in line with national expectations [9]

·      Learning analytics are used to assess mastery of content and competences and to make recommendations for teacher practice in response to these assessments [9]

·      Learning analytics augment classroom teachers’ awareness and memory [1]

·      Learning analytics make it possible to find the right open learning resources [2]

Specific challenges

·      Learning analytics identify which student in a group is the dominant domain expert [10]

·      Learning analytics identify which problems will be solved in/correctly by a group [10]

References

[1] Judy Kay. 2022. Grand Challenges for Pervasive Technology to Transform Pervasive Education. IEEE Pervasive Computing 21, 3.

[2] Jocelyn Manderveld, Alan Berg, Robert Schuwer, and Hendrik Drachsler. 2015. Grand Challenges Learning Analytics and Open Online Onderwijs.

[3] National Academy of Engineering. 2008. Grand challenges for engineering in the 21st century from http://www.engineeringchallenges.org

[4] Ido Roll, and Philip H Winne. 2015. Understanding, evaluating, and supporting self-regulated learning using learning analytics. Journal of Learning Analytics 2, 1, https://doi.org/ https://doi.org/10.18608/jla.2015.21.2

[5] Chris Dede, and William Lidwell. 2023. Developing a next-generation model for massive digital learning. Education Sciences 13, 8,

[6] Christopher J Dede, Andrew Dean Ho, and Piotr Mitros. 2016. Big data analysis in higher education: Promises and pitfalls. Educause Review

[7] Ryan Baker. 2019. Challenges for the Future of Educational Data Mining: The Baker Learning Analytics Prizes. Journal of Educational Data Mining 11

[8] Petri Ihantola, Arto Vihavainen, Alireza Ahadi, Matthew Butler, Jürgen Börstler, Stephen H Edwards, Essi Isohanni, Ari Korhonen, Andrew Petersen, and Kelly Rivers. 2015. Educational data mining and learning analytics in programming: Literature review and case studies. Proceedings of the 2015 ITiCSE on Working Group Reports

[9] R Pea, and D Jacks 2014. Building the field of learning analytics for personalized learning at scale. Graduate School of Education, Stanford University. https://ed.stanford.edu/sites/default/files/law_report_complete_09-02-2014.pdf

[10] Sharon Oviatt, Adrienne Cohen, and Nadir Weibel 2013. Multimodal learning analytics: Description of math data corpus for ICMI grand challenge workshop. https://dl.acm.org/doi/10.1145/2522848.2533790