by Geraint Evans (SOCSI)
Geraint is an eLearning Officer within the SOCSI eLearning team at the Cardiff School of Social Sciences. He is the eLearning lead for the Continuing Professional Education and Learning (CPEL) programmes commissioned by the Care Council for Wales.
What can we learn from learning analytics? A case study based on an analysis of student use of video recordings (overview)
This session was one of the more interesting from the conference, and had two main angles which were of interest to me:
- Learning Analytics – an area of growing importance and something we should possibly doing more of.
- Lecture Capture with Panopto – A hot topic in Cardiff at the moment and this talk had some useful insights.
This project was about measuring and collecting data at Imperial College from 18 undergraduate modules across year 1 and 2. The idea was to get ‘actionable insights leading to a change in practice’. The project came across a number of challenges including ethics, privacy, flexibility and ease of use.
There were some interesting findings from the project, including:
- Use varies considerably across different students and modules;
- No direct trend or correlation of use of lecture capture with grades;
- There was more engagement with key lectures;
- No significant difference in use for students with specific learning difficulties;
- In terms of ‘required’ lectures (i.e. ones they are asked to view before going to the face to face session) there was a correlation between use and grades. i.e. students who viewed these did well;
- How much students watch varies by degree. There is also an interesting distinction between learning period and revision period;
- If recordings are released late (because of technical difficulties) they are watched much less;
- Spacing out lectures allows students to use the VLE (Virtual Learning Environment) to consolidate learning from one lecture before the next.
There were a number of actionable insights including advice for students, advice for lecturers, and giving all lecturers training in how to use the analytics tools.
Going forward there are plans to use this approach for other learning resources such as Blackboard and PeerMark. Also it was noted that the project raised a few new questions and that further investigation, particularly qualitative, would be useful to understood more fully exactly what is happening, particularly with regard how students are actually using the VLE.
This all got me thinking that we could be doing more as learning technologists to explicitly use learning analytics projects to make an impact in our schools and colleges, and lecture capture may be a great starting point for this, given the statistics that Panopto can produce. It could also be a way to challenge some myths and change perceptions. Someone also mentioned the idea of learning technology as a ‘Trojan Horse’ to get people thinking about their teaching, and learning analytics could be an extension of this.