Different numbers mean different things to different people

In this blog I am reflecting on how analytics, such as those generated by Google Analytics could be useful in education, specifically, what would be most useful to the different people.  As always within education I want to start with the learner. 

What can numbers tell me about how I learn and about how other learn?
The first part of this question might sound logical but why should a learner care about how others learn as well. I think there are two reasons they should. Firstly, the truth is that often there is an element of competition - maybe not openly amongst peers but certain in the job market and it is therefore helpful to have a sense of where you are in the cohort. Secondly, knowing how a first class student works can provide insight for other students. The latter is something we often lament in neuroscience - when trying to understand how, for example, the brain learns, we look at what happens when it goes wrong. Now this can be truly insightful but it is strange that we do not consider looking at the brains of those who have it mastered! I think there are a number of analytics that could help a learner address these questions:

• User explorer - i.e. the individual learner behaviour on the VLE for example
• Cohort analysis - where the cohort could be defined as your whole class e.g. every one taking a set module or every one taking a set module who performed similarly to you on the pre-requisite course etc.
• User Flow - this would be very helpful if shown for an individual learner and then for a high performing cohort. This would enable the learner to see how the strongest learners navigate their learning and could give them tips for changing their behaviour.

What analytics can I use enhance my teaching?
I think there is lots of information that a teacher can use from analytics to enhance their practice but I also think that with a class of 200, for example, individual level data will be conflicting and unhelpful so I think for the educator designing material/learning activities group level data is key. For example, the following could help:

• Demographic data including location: This can be helpful because it tells you a little bit more about your cohort and diversity but it is also a way to make your learning more inclusive. Take Psychology for example; the traditional student in this discipline is the white middle class female, but the field is becoming increasingly diverse and having an awareness of this enables educators to choose relevant and appropriate examples in their teaching. This is particularly important in a self-reflective discipline.
• Cohort analysis: This can be helpful to allow comparisons before and after specific changes to a module but also to see if any particular cohorts require extra support e.g. those with or without a specific A-level subject may require extra help in the first year with certain modules.
• User behaviour: For example, looking at how often people are logging and for how long can be helpful if the VLE will be used for specific time limited activities.

Following on from these group level analytics, if an individual student requires additional help then it is useful to have information on them specifically, similar to the kind of information that the learner themselves may find helpful.

What about TEL support staff?
We were asked as part of the activity to consider administrators as well, but in my current role, administrators have little to do with the design of the programme or student support beyond processing attendance and dealing with mitigating circumstances around assessment so instead I chose to consider the role of TEL support staff. For this group there is some potentially useful data that could inform high-level design of the online learning resources we offer, such as:

• Mobile devices and technology: It is very helpful to know what kind of devices learners are using to access learning resources. This gives a sense of what minimum requirements must be met and also whether we are spending too much resource on, for example, creating a better mobile app, when most students log in from a laptop.
• Benchmark data: Knowing how we compare to other universities could be very helpful. 

I think the key with analytics is that all data should be available to any role but that it is sensible to first provide the relevant data to specific individuals. If they then wish to delve a little deeper then it may be appropriate to share, for example, data you would normal reserve for TEL support staff with the educator. Of course, this would probably not be necessary if teaching and learning was co-constructed by all three of these key roles.