Tag Archives: data

I’m melting…. – Weeknote #21 – 26th July 2019

Valentine Bridge in Bristol
The view from Mead Reach Bridge looking towards Valentine Bridge in Bristol.

This week, we melted, we had a new Prime Minister, we had a new government and I didn’t go to London.

Monday I was back into the office to do what I initially thought was going to be a demonstration of Jira and Confluence, but in the end turned more into a discussion on how people are using the tools across Jisc.

Had to make a phone call on Monday, something which in work I don’t actually do that often. I make lots of audio conferences and skype calls, but I don’t use the phone as much as I have in other roles. I am part of a telephony project at Jisc and as a result I am now using Teams for making and receiving calls. It was a seamless experience, and it was nice making a call using a sound cancelling headset with microphone, rather than holding a handset or mobile phone to my head! I did feel that it was somewhat odd to use my laptop to dial the number rather than a number pad. A few years back I was looking a telephony and I remember thinking back then that there was a real culture shift needed by organisations moving from traditional PBX (Private Branch Exchange) system to a modern telephony system used through Teams. Even now I think there is still need for a culture shift that isn’t easy for some people to just get and then move on.

This week, eleven years ago I wrote a blog post about the CherryPal mini PC which cost $249.

CherryPal launches $249 cloudy mini PC

CherryPal launches $249 cloudy mini PC

It was funded by advertising…

Today you can buy a better specified Raspberry Pi for under £35 and no advertising.

How things change….

Decided that I would become a Thought Leader and luckily for me, and thank you to Lawrie Phipps for the link, there is a course to do this on LinkedIn Learning…

I’ll let you know how I get on.

I wrote a blog post in response to a tweet I had seen earlier this year about using facial and emotion recognition with gauge the degree of student engagement in a lecture.

This week ten years ago I saw this video from Steve Boneham about something called micro-blogging…

Wonder if it ever caught on….

Crowd
Image by Brian Merrill from Pixabay

Talking about data, read this Guardian article, ‘Anonymised’ data can never be totally anonymous, says study.

“Anonymised” data lies at the core of everything from modern medical research to personalised recommendations and modern AI techniques. Unfortunately, according to a paper, successfully anonymising data is practically impossible for any complex dataset.

The article discusses the how data which has been anonymised data can in a number of methods be deanonymised to identify real people.

This has implications for universities and colleges, who are looking at using deanonymized data for intelligence and informed decision making.

If you think of anonymised data tracking students movement across campus, using wifi, this could be easily deanonymized using attendance data, swipe card data, PC logins, library card data.

Something to think about. The research is published in the journal Nature Communications.

Thursday, I was going to go to London for a meeting with colleagues from the DfE. However due to the heat we decided to have the meeting virtually. Though there are advantages in meeting face to face, the fact we now have the technology to make meetings virtual means that we don’t need to cancel or re-schedule meetings. There are also affordances with virtual meetings, I like using the chat to post relevant links rather than interrupt the flow of the meeting. The fact the links are “live” and saved, means people don’t a) need to copy them down or b) wait until the links are e-mailed to them after the meeting.

microphone
Image by Florian Pircher from Pixabay

I spent some time working on abstracts and proposals for various conferences I am attending in September. Working for an organisation like Jisc, I obviously need to talk about stuff we’re doing at Jisc. I kind of miss the keynotes I was doing ten years ago, when I had a lot more freedom on the topics and subjects I was presenting on. Back then I spent a lot of time talking about the future of learning, which the main thrust was that change is going to happen, but the important part of that journey was people, academics and students. The technology facilitates and provides affordances, but in the end it’s people who will want to change the way they do things and people will need to demonstrate leadership if they want change to happen. For the conferences in September I will mainly be talking about Education 4.0.

Friday I was back in the office in Bristol working on my preparation  for my end of year review. This year has been interesting as I changed roles in March so did not complete my previous objectives and inherited a number of new objectives.

I was reminded of the problems heat can cause this week with this photograph from seven years ago in 2012, my Google Nexus One got so hot I had to put it in the fridge….

my Google Nexus One got so hot I had to put it in the fridge....

My top tweet this week was this one.

 

Learning Analytics – Case Studies and Report

Blackboard

Probably the highest profile technology amongst senior managers and leaders at this time is the use of analytics to support teaching, learning and assessment.

Using the data that institutions gather on a regular basis for the purposes of analysis, looking for patterns is one that has gained traction over the last few years. There are also others who wonder if this analysis of data and patterns is useful and allowing us to make informed decisions about learners.

Jisc have released a new report: Learning Analytics in Higher Education: A review of UK and international practice  (PDF). Drawing on eleven case studies, they examine why institutions are deploying learning analytics, and what the benefits are for learners. They also discuss the main data sources being drawn upon by institutions and the technical architecture required.

The emphasis of the report is on investigating the evidence for learning analytics: what impact it’s having, and to what extent the algorithms can actually predict academic success.

I have always seen analytics as a tool to support and enhance existing decision making and support, that was already in place. The analytics reinforcing an existing view, or bringing to light patterns that were previously hidden.

Analytics in my opinion doesn’t replace good teaching decisions, support and intervention strategies, it helps inform them, so that we can ensure all learners receive the support and advice they need. Which is why I am also pleased to see in the report, that they also look at how institutions are carrying out interventions to attempt to retain students at risk, and provide better support for all students as they progress through their studies.

The interventions arising from analytics are probably the most important aspect of analytics, otherwise why bother?

The main report summarises the case studies.  The full individual case studies are:

  1. Traffic Lights and Interventions: Signals at Purdue University
  2. Analysing use of the VLE at the University of Maryland, Baltimore County
  3. Identifying at-risk students at New York Institute of Technology
  4. Fine-grained analysis of student data at California State University
  5. Transferring predictive models to other institutions from Marist College
  6. Enhancing retention at Edith Cowan University
  7. Early alert at the University of New England
  8. Developing an ‘analytics mind-set’ at the Open University
  9. Predictive analytics at Nottingham Trent University
  10. Analysing social networks at the University of Wollongong
  11. Personalised pathway planning at Open Universities Australia

Using data

Some of you may know that as well as responsibility for ILT (e-learning) at Gloucestershire College I am also in charge of the Libraries.

This case study from the Excellence Gateway is an interesting one.

Tresham College of Further and Higher Education has introduced learning resources loan data into the main College reporting system, which has allowed all staff to see the information. This has encouraged the closer integration of learning resources with the curriculum. As one of a series of innovations this has helped almost double the amount of resources borrowed by learners.

The more you know about how the resources are been used allows you to target the resources you do have better. It also needn’t be an issue of the Learning Resources staff doing all the work, involve practitioners and learners. As Tracey Burrows, Programme Co-ordinator, School of Business and Professional Studies says:

I have used the book loan statistics with my learners and it has been useful to see which books they have used. This has allowed the group to recommend resources to each other and has led to increased book borrowing which has benefited the learners in achieving their objectives.

Of course you can overload users with data and information, so care is needed. I will be looking at this case study to see if we can implement the lessons learned.