Social Learning Analytics symposium
— student work-in-progress

A SoLAR Storm event hosted by the Open University SocialLearn Research Project

KMi Podium, Level 4, Berrill Building, Open University, UK [map]

12.30 Lunch
Programme: 2-5pm Tues 26 June [convert timezones]

Tweet on #StormSLA if you’re attending,
so we can see you on TwitterMap

This is the chance to catch up with the research programme on Social Learning Analytics that we are developing.

We’ll start by giving an introduction, and then two of our fabulous SocialLearn Interns will present their work in progress, which moves this programme forward.

This hybrid f-f/webinar is part of SoLAR Storm — the new virtual research lab convened by the Society for Learning Analytics Research, to build research capacity in this new field by networking PhD researchers with each other and the wider community.


12.30 – Lunch

2pm – Social Learning Analytics: Five Approaches

[Tweet comments/questions #StormSLA @r3beccaf @sbskmi]

Rebecca Ferguson and Simon Buckingham Shum, Knowledge Media Institute & Institute of Educational Technology, The Open University, UK

This paper proposes that Social Learning Analytics (SLA) can be usefully thought of as a subset of learning analytics approaches. SLA focuses on how learners build knowledge together in their cultural and social settings. In the context of online social learning, it takes into account both formal and informal educational environments, including networks and communities. The paper introduces the broad rationale for SLA by reviewing some of the key drivers that make social learning so important today. Five forms of SLA are identified, including those which are inherently social, and others which have social dimensions. The paper goes on to describe early work towards implementing these analytics on SocialLearn, an online learning space in use at the UK’s Open University, and the challenges that this is raising. This work takes an iterative approach to analytics, encouraging learners to respond to and help to shape not only the analytics but also their associated recommendations.

Key Reference:

Ferguson, R. and Buckingham Shum, S. (2012). Social Learning Analytics: Five Approaches. Proc. 2nd Int. Conf. Learning Analytics & Knowledge, (29 Apr-2 May, Vancouver, BC). ACM Press: New York. Eprint:

Bios: Simon was Programme Co-Chair for the 2012 Learning Analytics conference, serves on the new Society for Learning Analytics Research, and is a regular invited speaker on the topic including EDUCAUSE and Ascilite. His particular interests are in what learning analytics may be blind to, analytics for informal/social learning, and whether analytics can help build the learning dispositions and capacities needed to cope with complexity and uncertainty — the only things we can be sure the future holds.

Rebecca is a research fellow in the UK Open University’s Institute of Educational Technology, focused on Educational Futures. She works as research lead on the SocialLearn team, developing and researching initiatives to improve pedagogical understanding of learning in online settings, to design analytics to support the assessment of learning in these settings, and to extend the university’s ability to support learning in an open world.

2.45pm – A Self-Training Framework for Automatic Exploratory Discourse Detection

[Tweet comments/questions #StormSLA @zhongyu_wei]

Zhongyu Wei, Department of System Engineering & Engineering Management, The Chinese University of Hong Kong

With the development of online learning platforms such as SocialLearn, learning resources are uploaded to the internet at a dramatically increasing rate, which makes it difficult for individuals to identify information in need. Ferguson and Buckingham Shum (2011) aim at seeking qualitative understanding of context and meaning of the information and identify “exploratory dialogues” to facilitate users to decide if a resource is useful based on sociocultural discourse analysis (Mercer, 2004).

In this project, we extend the previously proposed self-training framework (He and Zhou, 2011) to detect exploratory dialogues from online learning materials automatically. We cast the problem of detection of exploratory dialogues as a binary classification task which classifies a given piece of text as exploratory or non-exploratory. We first train an initial maximum entropy (MaxEnt) classifier based on a small set of manually annotated dataset. The trained classifier is then applied on the large amount of unseen data. Texts classified with high confidence (refer to pseudo-labeled instances) are added into the training data pool for iteratively updating the classifier. Apart from incorporating pseudo-labeled instances directly into the MaxEnt training process, we also explore the use of pseudo-labeled features to constrain the MaxEnt training. Our extensive experiments on the transcribed text from online conferences and the learning paths data downloaded from the SocialLearn platform show that with the self-training framework, the performance of MaxEnt improves significantly. The improvement is more prominent when facing with a smaller number of annotated training instances. The proposed approach will be integrated into the SocialLearn platform for highlighting exploratory discourses in learning paths.

Preparatory material: LAK11 paper introducing Exploratory Discourse Learning Analytics: [paper] [replay]

Key References:

R. Ferguson and S.Buckingham Shum. 2011. Learning analytics to identify exploratory dialogue within synchronous text chat. In Proceedings of the 1st International Conference on Learning Analytics and Knowledge, pp. 99–103. ACM. Eprint:

N. Mercer. 2004. Sociocultural discourse analysis. JAL, 1:137-168.

Y. He and D. Zhou. 2011. Self-training from labeled features for sentiment analysis. Information Processing & Management, 47(4):606–616.

Bio: Zhongyu WEI is a Ph.D. student in the Department of System Engineering & Engineering Management at The Chinese University of Hong Kong, under the supervision of Prof. Kam-Fai Wong. He received both M.Eng. and B.Eng. degrees in Computer Science and Engineering from Harbin Institute of Technology, China. His research focuses on social network analysis, information retrieval and machine learning. He is now visiting Knowledge Media Institute as an intern, working together with Dr. Simon Buckingham Shum, Dr. Yulan He and Dr. Rebecca Ferguson.

3.30pm – Prototyping Learning Power Modelling in SocialLearn

[Tweet comments/questions #StormSLA @shaofutw]

Shaofu Huang, Graduate School of Education, University of Bristol

Many educational experts argue that good learners in the 21st century will need a strong set of skills and attitudes to learning. To measure this, research of Effective Lifelong Learning Inventory (ELLI) has identified the “power” of an effective learner with seven dimensions [i]. The seven dimensions structure was designed to assess a learner’s learning power as well as to open up a window for coaching conversation which stimulates further development of the learner’s learning [ii].

Now being used by learners at many levels of education and in many countries, ELLI models learning power according to what learners think about themselves. EnquiryBlogger, a blogging platform for learners developing since 2010, enables users to link their online activities with learning power dimensions [iii]. What we are interested in this study, however, is whether and how we can model learning power according to what learners do in SocialLearn.

This study looked like a transplanting process to me at the beginning: from learner self-reporting to automated platform reporting. It still is, but I have now expanded my understanding of what “transplanting” means: it involves not only different languages, but also different ways of quality assurance and different interests. I will first introduce the learning power dimensions developed with ELLI as a background, then focus on the method and research design of this study and its inter-discipline nature.

Preparatory material: LAK12 paper introducing Dispositional Learning Analytics: [paper] [replay]

Key References:

[i] Ruth Deakin Crick, Patricia Broadfoot, and Guy Claxton, ‘Developing an Effective Lifelong Learning Inventory: The ELLI Project’, Assessment in Education: Principles, Policy & Practice 11 (2004): 247–272.

[ii] Ruth Deakin Crick, ‘Learning How to Learn: The Dynamic Assessment of Learning Power’, Curriculum Journal 18 (2007): 135–153.

[iii] Rebecca Ferguson, Simon Buckingham Shum, and Ruth Deakin Crick, ‘EnquiryBlogger: Using Widgets to Support Awareness and Reflection in a PLE Setting’ (presented at the PLE Conference 2011, Southampton, UK, 2011),

Bio: Shaofu Huang is a PhD student in the Graduate School of Education, University of Bristol. His broad research interests include authentic pedagogy, participatory learning and systems thinking. He has six year experience of teaching, learning design and teacher training before starting his postgraduate study here in the UK. He has been involved in the recent development of Learning Warehouse and many ELLI data analysis projects since 2010. Shaofu is now working as a SocialLearn intern with Simon Buckingham Shum and Rebecca Ferguson.

4.15 – Tea break

4.40pm – Open Discussion

5pm – Close