“Learning Analytics is the measurement, collection, analysis and reporting of data about learners and their contexts, for purposes of understanding and optimizing learning and the environments in which it occurs.” — from SoLAR (Society for Learning Analytics Research, www.solaresearch.org)
Open Learning Analytics, an integrated & modularized platform, was proposed by SoLAR to build an open platform approach to integrate heterogeneous learning analytics techniques.
Why Open Learning Analytics platform?
“The history of technology adoption in education suggests a consistent and challenging model: important ideas and innovation ideas developed piecemeal and in isolation, resulting in a fragmentation and confusion for end users who are most in need of efficient solutions. Our proposed integrated learning analytics platform attempts to circumvent the piecemeal process of educational innovation by providing an open infrastructure for researchers, educators, and learners to develop new technologies and methods.” — Simon Buckingham Shum, Professor of Learning Informatics at Open University UK
Simon Buckingham Shum gave more arguments about it:
The scientific argument:
“Analytics need to be broad-based, multi-sourced, contextual and integrated in order to mitigate the limitations of any single dataset or algorithm.“
The economic innovation argument:
“The wealth of networks lies in orchestrating mass innovation via open platforms.“
Three critical beliefs underpin the proposal:
- Openness of process, algorithms, and technologies is important for innovation and meeting the varying contexts of implementation.
- Modularized integration: core analytic tools (or engines) include: adaptation, learning, interventions, and dashboards. The learning analytics platform is an open architecture, enabling researchers to develop their own tools and methods to be integrated with the platform.
- Reduction of inevitable fragmentation by providing an integrated, expandable, open technology that researchers and content producers can use in data mining, analytics, and adaptive content development. Educators, learners, and administrators benefit from modularized functionality: with customizable and extendable core analytics, intervention, and content tools to meet needs of learners and educators (particularly in identifying at-risk students). Administrators benefit from integrated tools that track learning-related activity and then influence resource allocation across multiple tools and spaces of learning. Learners will benefit from having timely and relevant feedback on their performance, as well as content, activity, and social network recommendations to improve and guide their learning.
How can such a platform be delivered, which as emphasized, enables them to compare and contrast tools, and datasets, from diverse sources? Fundamentally, we require an open platform with standards for adding new “plugins” . As long as developers of analytics, recommender services, visual user interfaces, and intervention strategies, comply with these standards, their work can become part of this ecosystem.
The open and extensible learning analytics platform aims to grow an ecosystem of stakeholders and tools around this. The concept of the platform and the community around it are depicted as below:
The intent for this project is to create the architecture for researcher and practitioner innovation in education through access to tools that provide insight into the teaching and learning process. The following goals are emphasized:
- Development of common language for data exchange.
- Analytics engine: transparency in algorithms so learners and educators are aware of data being gathered and researchers can customize analytics methods to reflect the needs of different contexts (schools, circumstances, policy or administrative priorities).
- Dashboards and reporting tools to visualize information and provide real-time information to learners, educators, administrators, and researchers.
- Open repository of anonymized data for training and research development
- Connect to and amplify the existing research being conducted by EDUCAUSE, the International Educational Data Mining Society, Next Generation Learning Challenge, and related analytics initiatives such as the EU dataTEL initiative.
xAPI will bring this dream true
xAPI is a standard protocol using REST architecture style for writing and querying all learning activity data. It can act like a glue for learning analytics across diversified data sources. In a recent interview on the Open Learning Analytics summit with Josh Baron, Baron pointed out that open standards like xAPI and Learning Record Store(LRS) can bring data from diversified sources into a single repository. Because after xAPI data are sent to any one LRS, they can be retrieved and aggregated by another LRS for analytics and reporting purposes. A data common language like xAPI makes the Open Learning Analytics picture possible.
Baron also emphasized the importance of open data sets and open data models:
If our data models remain proprietary to vendors, and we can’t peek into them, that creates a barrier to doing research to further the field, and it even creates practical problems for the institution in terms of understanding how systems work — imagine trying to answer a student’s question about why they received an alert if the institution doesn’t have full visibility into to the data model! A lack of openness is just not going to cut it for institutions in the future: Being able to understand how these models work will become very important to institutions that will be making important decisions based on how they work.
In the future, institutional repositories built on open standards will eventually facilitate a lot of learning analytics that, for now, simply aren’t taking place at all.
Open Learning Analytics from SoLAR (OpenLearningAnalytics, pdf file)
The presentation of Professor Simon Buckingham Shum :