Learning analytics

The easiest way to start thinking about learning analytics is to draw a comparison with what happens in retail. Big chain stores around the world are getting really good at recommending products to consumers. One in particular done some incredible stuff around this — they know that if a woman of a certain age starts buying unscented lotion, a bigger handbag and multivitamins, there’s an extremely high likelihood that she’s pregnant, and would like more reviews and recommendations related to maternity products and babies. Now you might think that this is about moving baby products onto consumers, but, if you think about the fact that one of our goals is to get the perfect, most engaging learning opportunity in front of a student at just the right moment, this kind of big data analysis, or learning analytics, becomes very powerful.

Let me give you another example: Amazon’s ‘people who bought this also bought...’, or their ‘recommendations for you…’ section is basically learning analytics. This kind of thinking has been brought into education to help raise literacy levels. The New Zealand library management and cataloguing software Koha has analytics built into it: if a student rates a book 5 stars, the software is able to look across all borrowers and say ‘other people who rated these books 5 stars also rated this book 5 stars — would you like to read it? We’ll reserve it and text you when it’s available.’ You can choose to limit your analytics to you or your school only, or vastly increase the accuracy of your predictions by looking across all other users who have rated items.

So what? — The implications

We need to ask what data we’re gathering about our students and their progress through learning. If we’re completing tasks in a range of different online spaces, how do we bring all of that disparate data about a learner and make it whole again — make a complete picture of this child.

Another implication for us is the challenge to use that data once it’s gathered. There’s a great saying about data: it needs to be useful and used. It must be relevant, reliable and meaningful, but it’s pointless to gather data if we’re going to use it. What are your teaching as inquiry processes like in your school? How well is data used when making decisions about what needs to be learnt next and how students might best learn it? Are you drawing on the rich data you have about your students?

Some of the ethical implications for us centre around data sovereignty and privacy, the real power of learning analytics is unlocked when you’re able to work with large data sets — which means sharing data across schools. How are you going to ensure you deal fairly with students and other schools when sharing data? If you’re contributing to national-level data collection, have you thought through the implication around who has access to it, how student rights are managed?

If we can start to make use of learning analytics to get the right learning activity into those student's hands, and maximise the engagement and motivation they have for that learning activity, we’ve got a really powerful model for personalising learning for every student.

Examples and links

• University of Bristol: Learning analytics and technology enhanced assessment (PDF, 348 KB)
• Educause Library: Learning Analytics
• DML Central: Big Data