Thoughts so far on LAK11

Along with about 400 or so others world-wide, I’ve signed up for the LAK11 (Learning and Knowledge Analytics) MOOC run by George Siemens and colleagues at the Technology Enhanced Knowledge Research Institute (TEKRI) at Athabasca University. We’re now into week 2, and I think I’m just about getting into the swing of things.

When George was in the UK late last year, I managed to catch his presentation at Glasgow Caledonian, and I was intrigued with the concept of learning analytics, and in particular how we can start to use data in meaningful ways for teaching and learning. I wanted to know more about what learning analytics are and so signed up for the course. I’ve also been intrigued by the concept of MOOCs so this seemed liked the ideal opportunity to try one out for myself.

In her overview paper, Tanya Elias provides a useful description: ” Learning analytics is an emerging field in which sophisticated analytic tools are used to improve learning and education. It draws from, and is closely tied to, a series of other fields of study including business intelligence, web analytics, academic analytics, educational data mining, and action analytics.” (Elias, T. (2011) Learning Analytics: Definitions, Processes, Potential)

The course outcomes are:
*Define learning and knowledge analytics
*Map the developments of technologies and practices that influence learning and knowledge analytics as well as developments and trends peripheral to the field.
*Evaluate prominent analytics methods and tools and determine appropriate contexts where the methods would be most effective.
*Describe how “big data” and data-driven decision making differ from traditional decision making and the potential future implications of this transition.
*Design a learning analytics implementation plan at a course level. 
*Evaluate the potential impact of the semantic web and linked data on learning resources and curriculum.
*Detail various elements organizational leaders need to consider to roll out an integrated knowledge and learning analytics model in an organizational setting.
*Describe and evaluate developing trends in learning and knowledge analytics and develop models for their potential impact on teaching, learning, and organizational knowledge

You can check out the full course syllablus here .

The fact that the course is open and non-accredited really appealed to me as, to be honest, I am a bit lazy and not sure if I wanted to commit to to a formal course. The mix of online resources, use of tags, aggregation etc fits right in with my working practices. I blog, I tweet, I’m always picking up bits of useful (and useless) information from my streams – so having a bit of focus for some activity sounded perfect – I’m a self motivated kind of a person aren’t I?

But it’s never that simple is it? Old habits die hard – particularly that nagging feeling of guilt about signing up for a course and not reading all the suggested texts, reading all the forum messages, doing all the suggested activities. Is it just me that suffers from the tensions of trying to be an engaged, self motivated learner and everyday distractions and procrastination? I’ve had some vey circular discussion about myself about why I’m not actually looking at the course material at times.

However, George and the team have been particularly good at reassuring people and emphasising that we need to “let go of traditional boundaries”. With a cohort this large it’s pretty near impossible to keep up with everything so they actively encourage people only to do what they can, and concentrate on what what really interests you. They actively encourage “skim and dive” techniques -skim the all the resources and dive into what catches your eye/interest. If you’ve being thinking about doing one of the MOOCs then I would recommend having a listen to the introductory elluminate session (another great thing about open courses is that all the resources are available to everyone, anytime).

I’ve found the eliminate sessions the most interesting so far. Not because the other resources provided aren’t as engaging – far from it. I think it’s more to do with the synchronous element and actually feeling part of a community. All the speakers so far have been very engaging, as has the chat from participants.

Last week as introduction to Learning Analytics, John Fritz, UMBC gave an overview of the work he’s leading in trying to help students improve performance by giving them access to data about their online activity. They built a BlackBoard building block called Check My Activity (CMA), you can read more about it here. John and colleagues are also now active in trying to use data from their LMS to help teachers design more effective online actives.

This week’s topic is “The Rise of Big Data” and on Tuesday, Ryan Baker from Worcester Polytechnic Institute was in the eliminate hot seat, giving us an introduction to Educational Data Mining (EDM). EDM draws heavily on data mining methodologies, but in the context of educational data. Ryan explained it as a distillation of data for human judgement. In other words making complex data understandable and useful for non information scientists. EDM and Learning Analytics are both growing research areas, and the there are a number of parallels between them. We did have quite a bit of discussion about what the differences were exactly, which boiled down to the fact that both are concerned with the same deep issues, but learning analytics is maybe broader in scope and using more qualitative approaches to data and not so dedicated to data mining methodology as EDM. Ryan gave an overview of the work he has been doing around behaviour modelling from data generated by some of Carnegie Mellon Cognitive Tutor programmes, and how they are using the data to redesign actives to reduce for example students going “off task”. Again you can access the talk from the course moodle site.

Next week I’m hoping to be doing a bit more diving as the topic is Sematinc Web, Linked Data and Intelligent Curriculum. Despite the promise, there really isn’t that much evidence of linked data approaches being used in teaching and learning contexts as we found with the JISC funded SemTech report and more recently when Lorna Campbell and I produced our briefing paper on The Semantic Web, Linked and Open Data. I think that there are many opportunities for using linked data approaches. The Dynamic Learning Maps project at the University of Newcastle is probably the best example I can think of. However, linking data within many institutions is a key problem. The data is invariable not in a standard form, and even when it is there’s a fair bit of house keeping to be done. So finding linkable data to use is still a key challenge and I’m looking forward to finding out what others are doing in this area.

4 thoughts on “Thoughts so far on LAK11

  1. Great post Sheila – you are certainly not alone in feeling that you need to read all the suggested texts and be apart of all the conversations. I think maybe this is the biggest learning point for someone like myself who is new to MOOC – learning that I dont need to do everything I just need to do what suits my needs.

  2. I agree with Mark. It’s quite tough in the beginning, because of the ‘panic’ of not really knowing what we need to know in order to follow the topics… But thanks to previous experience in following George’s teaching in 2008, I’m more confident in ‘leaving behind’ some of the texts/articles recommended to read in the study weeks.

    In fact, I find it more comfortable reading others’ reviews on the texts, and then go into the texts to ‘understand the gist’ myself and compare it with the reviews. At least some of the main ideas are already covered by the reviews, so we don’t really miss much if we skim through and spot other points that interests us.

    One thing for sure, what our eyes spot as interest might be a different angle of understanding compared to others. Interesting! ;D (even though it sounds like we tend to get lost too. LOL)

    – Shazz, Kuala Lumpur