Learning analytics is the application of big data techniques such as machine based learning and data mining to help learners and institutions meet their goals.
Everytime a learner touches something digital, they leave a digital fingerprint. By looking at changes in these, can improve retention, achievement, employability, and help deliver personalised learning
Learning analytics has been requested as a new national shared service by the sector.
This is the proposed architecture. Shared multi tenanted data warehouse at the centre.
Will be a staff dashboard, and also will present the data back to students through an app.
JISC have been out to tender and filled all slots with commercial partners except student consent service. Going to build that themselves.
The data model is consistent with the HEDIP landscape and HESA data service.
Commercial partners are Unicom, Marist, Blackboard, Tribal, Therapy, Box and HT2. Half are open source.
Staff dashboard will be delivered through Tribal insight. Will show how a student digital fingerprint changes. Eg if a student stops attending lectures, borrowing books, but spending more on thier cashless card at midnight in the bar it might trigger an alarm for some sort of human intervention. Should help with retention, student well being.
Will be an app which feeds back data to the students so for example they can set targets. Can even share targets for a bit of competition.
Are some ethical questions. JISC have produced a code of practice to help with issues such as informed consent.
Also is a guide produced by NUS
Working with universities to build the system which they hope will be running before Christmas
Posted with Blogsy
No comments:
Post a Comment