Learning Analytics and Educational Data Mining
I’ve been thinking a bit about how some of our work on Learning Analytics compares with what folks in the Educational Data Mining community are doing. Might be interesting for some of you…
In my view, Learning Analytics is about collecting traces that learners leave behind and using those traces to improve learning. Educational Data Minging can process the traces algorithmically and point out patterns or compute indicators. My personal interest is more in using the traces in order to empower learners to be ‘better learners’.
My team focuses on building dashboards that visualize the traces in ways that help learners or teachers to steer the learning process. I like this approach because it focuses on helping people rather than on automating the process. It is inspired by a ‘modest computing’ approach where the technology is used to support what we want people to be good at (being aware of what is going on, making decisions, …) by leveraging what computers are good at (repetitive, boring tasks…).
Of course, capturing meaningful learning traces is something that both we and the EDM community struggle with. Translating those traces into visual representations and feedback that support learning is another challenge: the danger of presenting meaningless eye candy or networks that confuse rather than help is all too real.
Both our work and that of the EDM community is also difficult to evaluate: we can (and do!) evaluate usability and usefulness, but assessing real learning impact is hard – both on a practical, logistical level (as it requires longitudinal studies) as well as on a more methodological level (as impact is ‘messy’ and it is difficult to isolate the effect of the intervention that we want to evaluate).
In both these areas, we may be able to make better progress by exchanging our experiences. There is also an opportunity to combine both approaches: for instance, we can use visualization techniques to help people understand what data mining algorithms come up with and why. In that way, work on visualization can help to increase understanding of and trust in what the EDM community achieves.
Does this sound right to you? Do you have another view on how (educational) data mining and (learning) analytics relate to one another?
16 Comments
Leave a CommentTrackbacks
- Rastros significativos « Vapor de hojalata
- Semaine 1 : | Learning Analytics
- Learning analytics « Sylvia's Blog
- “Big things have small beginnings ..” | @svencharleer #hci #kul
- Engaging students in Learning Analytics | @svencharleer #hci
- Rastros significativos de aprendizaje | Bianka Hajdu
- Learning Analytics | Innohunt
Erik Duval's Weblog 
“Learning Analytics is about collecting traces that learners leave behind and using those traces to improve learning. Educational Data Minging can process the traces algorithmically and point out patterns or compute indicators.”
I believe that these two sentences clarify very well these two different terms.
Hello Erik
A group in Change11 Mooc are curating a blog calendar called One Change a Day. We would like your permission to reblog this post as it would be of itnerest to our audience. By the way I really enjoyed your ‘visit’ to Change11…… Regards Liz
Not sure why you would want to reblog it – can you not just refer to my blog post?
In any case, I’m happy you think it’s interesting 😉
Perfect summary. I had similar thoughts of the importance of the role of the learner as an active participant in learning analytics vs a more passive recipient of algorithmic prescriptives or data in EDM.
I doubt that the old methodology of watching the traces of learning directly or observing graphs leads to new insights about learning.
Using IT, huge amounts of data can be collected and processed. But it is not only the amount of data. The granularity is the new quality. With conventional tools, we could not record the learning activities in such detail. This was the beginning of Educational Data Mining (EDM). And this gave a new and massiv accent to learning analytics. I think that IT will lead to a completely new and expanded view of learning. IT is used for collecting data, but also for processing the collected data. This is more than an individudal brain could do. This ends perhaps in constructing new categories, new concepts, new paradigms of learning.
Learning Analytics is the process of analyzing data to improve learning. It is first an individual aspect of trying to better understand how the learning process works. This is not only for prediction success. The old pedagogical ideas, how can we empower students to get better learners or the connectivistic view, how can we engage and participate students in groups are quite another questions. But all could/should belong to Learning Analytics.
“Learning analytics” as improving learning is just one perspective to EDM. Another but not really an alternate perspective could be “Psychology of Learning” “Billing and fee system for educational services”, “teacher earnings”, evaluation, quality management etc.
http://en.wikipedia.org/wiki/Learning_analytics
http://en.wikipedia.org/wiki/Educational_data_mining
http://en.wikipedia.org/wiki/Psychology_of_learning
Reblogged this on One Change a Day.
Hi Erik, interesting point of view on the learning analytics.
What is the place of teachers in your vision? Sometimes, it is useful to collect traces of all parties involved in the learning process: students and teachers. Or you consider teachers also being learners, as they themselves learn in the process?
In my view, teachers can learn a lot from the traces they and their students leave behind too, indeed. And I do believe that teachers always are among those who learn most…
So should we also implicitly think about teachers as learners while reading your definition:
“In my view, Learning Analytics is about collecting traces that learners leave behind and using those traces to improve learning.”
or it’s better to write explicitly “… that learners __and teachers__ leave behind …”?