Comment on LARC learning analytics report 13/03/17 by cthomson

Hi Jeremy, my annotations were an attempt to critically consider the data that was gathered yet not actually used, as well as what was used. So in answer to your question I think context could be measured to some extent without more sophisticated analytics.

For example, if the report had considered my weekly average with the class weekly average (rather than the displayed course average) then it could have acknowledged ‘something’ going on across the class. So the low numbers would look less stark against the class numbers for each week.

Going a step further by combining this with the data that it was one week before the end of the course it could have provided supportive feedback such as “Congratulations, you are nearing the end of course, keep up the good work. You may be finding it difficult to stay motivated at this point but don’t forget to login/join conversations for the last week – it may help with your assignments(exams).”

Essentially, flipping the existing data into supportive encouragement rather than demotivating ….

from Comments for Clare’s EDC blog

Week 9 summary

At the beginning of the week I focused on looking around to get some basics about Learning Analytics (LA), adding links to the HEA and Jisc for example. Overall in general I got the sense that the advantages of LA for the learner was for the institution to be able to provide support and guidance. The proof of this seemed to be increased retention. However, I then explored many examples of algorithms gone wrong and the human impact of this.

I have been a LA sceptic without having an in-depth knowledge on the topic and I was quite surprised that during the tutorial many shared my cynicism and highlighted the need for more qualitative and contextualised analysis.

As education has higher and higher student numbers and fewer teachers LA is yet another way to solve this problem – I tried to show this on my image – but I don’t believe that it will. In any conversation I have ever had with others LA is always framed in terms of surveillance. Essentially, ‘we need to track our content to prove student didn’t engage’.  Every time I ask why.  ‘Proving’ the student has clicked on content is absolutely no proof of engagement. Only a human making contact, face to face or virtual, can deduce this and know if the student is having problems or is simply working at their own pace and timetable. I added my own annotated LARC report to help me frame this.

In addition, there is also the issue of ethics around the collection and analysis of all this ‘big data’ and I tried to highlight this by including the blog post by Lorna Campbell who tried to question a software company on its collection of data policies.

I ended the week by including a Storify and a TAGS Explorer map of the, very informative and enjoyable, Tweetorial as I tried to visualise the conversation as this helps me see past the numbers.

LARC learning analytics report 13/03/17

from Dropbox


It seemed only fitting to include my LARC report for the week after discussions around learning analytics. I have annotated it briefly with some notes but it definitely serves the purpose of highlighting the need for context. Without any context it would baffle an outsider, showing a poor level of social interaction, engagement or attendance.

A quick overview is that it was week 9 of a ten week course and two assignments were looming in my mind. I was focused on working on my learning challenge for the first of these and prioritised this along with reading and planning my writing. None of this was captured in the ‘numbers’. It also didn’t compare the week with my fellow classmates but to the course average which meant there was no direct comparison. This might have served to show some of the context as many of the others would be generating similar reports for the same week.

Leaders and Monitors: The best and the worst of education technology

Last week I attended the Holyrood Connect Learning Through Technology event where I saw a rather jawdropping demonstration of the very best and very worst that education technology has to offer.

from Pocket

I posted this blog post in the Tweetorial as well as embedding it here as it highlights the grave need to call out software providers to consider the ethics around people’s data and stop privileging the surveillance as a selling point. It is yet again a ‘because we can’ function and Lorna Campbell demonstrates that the voice of reason may be a small voice in the dark. This all seems to resonate with the general direction of our discussions this week and the reality of a surveillance society, framed to us a method of keeping us safe.

What happens when algorithms go ‘evil’?

Algorithms are the powerful mathematical tools which shape so much of modern life, from the news which appears in our timelines to the adverts which pop up on our computer.

from Pocket

This short excerpt from the BBC considers large scale data driven algorithms as a parallel with legal systems in that there is no perfect solution and it is a system of smaller parts coming together as one.

Considers the concept of ‘algorithmic discrimination’ and refers to the Microsoft Tay (bot) taken offline after only 16 hours due to rasicist output.

Pinned to #mscedc on Pinterest

Just Pinned to #mscedc: | Sensemaking with Learning Analytics @gsiemens #apereo14 keynote | I’m connecting the dots and hoping the apereo community gets on board with a full scale development of learning analytics open platform as an LMS plug-in.

Week 8 summary

Another week has flown past before I feel I have truly got to grips with it. I am a bit stuck in a ‘catch up one week at the start of the next‘ loop. I really enjoyed looking at and commenting on quite a few ethnographies, but made myself move on mid-week. However, I did add Pocket to IFTTT!

Trying to make sense of algorithms was worrying as I am definitely out of my comfort zone with numbers never mind big numbers. I began by watching some instructive talks and videos.

Despite my focus on my YouTube algorithm exercise the main element that has come through my week is yet again online community. On Twitter I spotted a good article about Google and education and this started a conversation about community and sharing and it turned out to be very circular indeed.

The Tweet from Amanda Taylor re article in the Conversation, author, Ibrar Bhatt brought algorithms and/vs serendipity to life: Amanda in Lancaster University, worked in Queen’s previously, Ibrar wrote article whilst at Lancaster University, now works at Queen’s in a different department from me. When I first retweeted the article I had no idea where Amanda was located or anything about the author so discovering such close network nodes showed me how algorithms are at play without me even realising, as I have no recollection of how I came to follow Amanda in the first place.

Lastly, as the week closes I am again thinking on the paradox of Higher Education’s continual resistance to change whilst simultaneously lauding technological innovations as potentially disruptive. Each time change is slow and minimal with a focus on administrative benefits rather than the learning experience. The virtual learning environment, VLE, is an ever present piece of evidence of this.

There you go Jeremy, proof I act on your feedback – word count under 300!

My video artefact: algorithms

from Dropbox


I have tried to pull all of the different elements of my week into a single video artefact: readings, audio produced in part by algorithms, visuals generated by algorithm alongside the algorithmic generated YouTube recommendations for me. The final curated video displays the human aspects behind the numbers.