SocialLearn Research Blog

Some of the thinking behind SocialLearn is set out in a chapter in the new book edited by a team at the Open University’s Knowledge Media Institute, with contributions from some of the thought leaders on the future of education:

Collaborative Learning 2.0: Open Educational Resources

Current advances and convergence trends in Web 2.0 have changed the way we communicate and collaborate, and as a result, user-controlled communities and user-generated content through Web 2.0 are expected to play an important role for collaborative learning.

Collaborative Learning 2.0: Open Educational Resources offers a collection of the latest research, trends, future development, and case studies within the field. Without solid theoretical foundation and precise guidelines on how to use OER and Web 2.0 for collaborative learning, it would certainly be very difficult to obtain all the benefits that these “user-generated content, resources and tools” promise. The purpose of this handbook is to understand how OERs and Web 2.0 can be deployed successfully to enrich the collaborative learning experience and ensure a positive outcome in terms of user generated knowledge and development of skills.

In Chapter 17, we set out the rationale that motivates the creation of a platform like SocialLearn…

This chapter examines the meaning of “open” in terms of tools, resources, and education, and goes on to explore the association between open approaches to education and the development of online social learning. It considers why this form of learning is emerging so strongly at this point, what its underlying principles are, and how it can be defined. Openness is identified as one of the motivating rationales for a social media space tuned for learning, called SocialLearn, which is currently being trialed at The Open University in the UK. SocialLearn has been designed to support online social learning by helping users to clarify their intention, ground their learning and engage in learning conversations. The emerging design concept and implementation are described here, with a focus on what personalization means in this context, and on how learning analytics could be used to provide different types of recommendation that support learning.

Read the whole chapter as an open eprint on the OU’s Open Research Online server:

Ferguson, R. and Buckingham Shum, S. (2012). Towards a Social Learning Space for Open Educational Resources. In: Okada, A., Connolly, T. and Scott, P. (Eds.), Collaborative Learning 2.0: Open Educational Resources. Hershey, PA: IGI Global, pp. 309–327. Eprint: http://oro.open.ac.uk/33457

This morning on the opening day of the CALRG 2011 Conference, we presented some of the recent thinking we’ve been doing on learning analytics, specifically in a social learning context.

A technical report setting out the line of argument in more detail…

Buckingham Shum, S. and Ferguson, R. (2011). Social Learning Analytics. Available as: Technical Report KMI-11-01, Knowledge Media Institute, The Open University, UK. http://kmi.open.ac.uk/publications/pdf/kmi-11-01.pdf

Abstract: We propose that the design and implementation of effective Social Learning Analytics presents significant challenges and opportunities for both research and enterprise, in three important respects. The first is the challenge of implementing analytics that have pedagogical and ethical integrity, in a context where power and control over data is now of primary importance. The second challenge is that the educational landscape is extraordinarily turbulent at present, in no small part due to technological drivers. Online social learning is emerging as a significant phenomenon for a variety of reasons, which we review, in order to motivate the concept of social learning, and ways of conceiving social learning environments as distinct from other social platforms. This sets the context for the third challenge, namely, to understand different types of Social Learning Analytic, each of which has specific technical and pedagogical challenges. We propose an initial taxonomy of five types. We conclude by considering potential futures for Social Learning Analytics, if the drivers and trends reviewed continue, and the prospect of solutions to some of the concerns that institution-centric learning analytics may provoke.

I recently  attended Alpine Rendez-Vous 2011, meeting people from all over Europe with an interest in technology-enhanced learning (TEL). For most of the time, we were split into eight workshops – my group was exploring ‘Methods and models of next-generation technology enhanced learning‘, examining the roles of assessment and evaluation in learning.

A focus for the workshop was to identify the ‘Grand Challenges’ for future research – constructing a framework for the development of TEL over the next ten years. The ideas were wide-ranging and exciting, for example:  Develop new technology to harness the power of emotions for learning. One challenge was closely related to our work at SocialLearn on learning analytics and recommendations engines: ‘How can learning be assessed in an open, social TEL environment?’

Our current model for the assessment of learning is primarily summative and individual, firmly bound to hierarchical education structures. It is a model that was developed for use when educational technology had only just moved beyond the horn book, knowledge was not abundantly available, groups of learners were taught and examined at the same time in the same physical location, teachers and learners were clearly differentiated and online collaboration and publication were undreamed-of possibilities. As new models of learning have been widely adopted, this model of assessment is no longer fit for purpose. We need new approaches and new models.

Open, social TEL environments have made new models of learning possible. Learners now draw upon many different people and resources, knowledge is dispersed and distributed, individuals may move rapidly between the roles of teacher and learner, and their collaborations extend across time and space. Despite these changes, the group work, the growing archives of activity and the available data about learning networks, there is still a focus on summative assessment of individuals.

TEL environments offer learners and educators a wealth of new resources in the form of the data they record – learners’ demographics, activities, interactions, participation and engagement – but little of this is currently harnessed to support assessment and even less is used to provide formative feedback and to help learners develop their metacognitive skills, their learning dialogue, their skill-sets or their knowledge.

Colleges and universities are beginning to make use of this data, but educational data mining and analytics are often viewed from an institutional point of view – how can we recruit more students? how can we retain students for longer? how can this institution maximise its income? A wordsearch for ‘teaching’ or ‘learning’ in the current literature produces woefully limited results.

Our challenge is to make use of new resources and technologies to develop and build on learning analytics – analytics that can make a real difference for learners, producing measurable improvements in areas such as:

• Engagement with learning – supported by appropriate and personalised feedback
• Quality of online learning dialogue
• Engagement with online learning networks
• Enjoyment – due, in part, to development of a students-in-trouble alerting system
• Learners’ and teachers’ awareness of the value of learning analytics and recommendations.

On June 7, I gave an EDUCAUSE webinar, in which I told the story that’s slowly emerging from thinking around Social Learning, Sensemaking Capacity, and Collective Intelligence. I follow through some of the forces that are shaping the social learning contours of the emerging landscape, specifically around the need for sensemaking when confronted with increasingly complex, unfamiliar dilemmas.

Many thanks to the EDUCAUSE-ELI team for the chance to share this work. Some of the videos of tools, which I didn’t get time to show, are blogged here. The slides are below, and on EDUCAUSE [or here as PDF], but the full replay only to EDUCAUSE members for a few months.

In this presentation, I pick up where the OSRG symposium left off, and give a more in depth update on the SocialLearn project to the Knowledge Media Institute.

I set the scene by flagging the history of research around the concept of social learning, and noting some of the tidal forces that many now recognise as shaping the new education landscape. (The latter are not, however, the focus of this talk so detailed treatment awaits another forum.)

I then introduce some of the core building blocks that the project has identified in seeking an answer to the question, what does it mean to design social media tuned for learning/sensemaking? In our experience, this is the question being asked by organisations exploring the potential of collective intelligence and social learning (including but by no means limited to “Educational Institutions”). Are social learning media just regular social media, but used in a particular way through learner/educator intention and activity, or would they have particular affordances, e.g. designed to provoke deeper reflection and learning conversations, beyond the normal rapid-fire information and media exchange of vanilla platforms (i.e. without ‘learning flavouring’ added!).

One of the results of pondering this are some dimensions of the technology design space which researchers and practitioners can flex when seeking to scaffold social learning with software tools, whether f2f or online. Yes, we want the best of social media platforms, plus…

Moreover, this has motivated a number of conceptual building blocks that we are now experimenting with…

…which when pushed in an elevator, I summarise as social+conceptual networks…

Amongst the many pedagogical frameworks that we have drawn on, we note Engeström’s wildfire learning activities with interest: the idea is that the building blocks provide clues as to how to render his proposed wildfire constructs of inquiry, trails, history, consolidation, argument, landmarks, places, and exploration.

In the final part of the talk, I move to a demo of the SocialLearn Space that is being constructed to provide cross-platform widgets for epistemic communities of inquiry, plus the tools to build, from their myriad identities and activities in the cloud, an aggregated learner profile, thus providing the basis for human and machine recommendations of people and resources.

As we noted in our new year update, SocialLearn’s immediate focus is on testing the concepts with communities within the university. PhD students and the wider researcher community are a particular focus right now (although with the blurring of e-learning and e-research/scholarship, by extension, in a course-oriented context we’d see these as tools for authentic student inquiry). However, when we pilot the tools they will be open to the world, so watch this blog for news.

The slide below shows schematically how the SocialLearn Space provides an integrational layer between currently siloed user activity within the university, and extending out to the cloud, providing the four core functions of Profile, UI, Social Graph and Services:

Developers…

Following the Open U’s strategic partnership with Google Apps for Education, we are experimenting with Google Gadgets, OpenSocial and FriendConnect to implement the platform, hosted in a Drupal-based portal. As a learner, we provide you with the tools to manage your learning widgets (Google Gadgets), connect your identities (e.g. sign in with OpenID, Facebook, Google, Yahoo!, LinkedIn, etc), maintain your profile (your learning history, interests and where you publish your learning journal, media, etc), and monitor your peers’ activity, and system recommendations through your personalised homepage.

As we demo in the KMi seminar, we’re dropping Gadgets into a range of Open U platforms as we put it through technical and user experience tests. Here’s one example, with Gadgets running in an experimental version of the Open U’s Cloudworks knowledge sharing environment (these are barebones gadgets – no graphic design work done yet!)…

One of the reasons we’re hosting the SocialLearn team in KMi is to facilitate connections with ongoing semantic web R&D. For instance, from the post-talk discussions, it’s clear that semantic technology could make key contributions to the challenges of merging social graphs from diverse sources, making sense of folksonomic tags, providing good multimedia search results, and recommendations based on content, peer activity and learning intentions. Aspects of the SocialLearn infrastructure will be released open source, with developer events now being planned.

So, you think you have a great learner gadget or computational service?

Prove it! My personal vision for SocialLearn is that it comes to serve as a commons-based innovation space: new end-user tools and backend services can be plugged in and out of a cleanly defined, scaleable architecture, for applied research on a large dataset and user base, enhancing the learner experience with the services and gadgets that add most value — as deemed by the people at the centre of a social media platform tuned for learning: learners and educators. You and your team can doubtless improve on the virtual rack of recommendation engines, provide more effective visualizations of social or conceptual networks, or provide cool connections to the many other tools out there that we want to dnce with. Think social media app store — but yes you got it: tuned for learning/sensemaking (and often, open source).

Sounds fun. But here for me is the kicker: if we can get it right, what we’d be doing collectively is building a suitably anonymised dataset which could become a commons-licensed (non-commercial?) resource for large scale social learning R&D. Every gadget and service is contributing data that can be mined and reasoned about. Haven’t worked the small print on that one through yet!…

Your views welcomed.