*Warning: This is not a post about GDPR!

With changes to the data protection regulations, people are becoming more aware of their rights relating to how organisations store and process their data. Data, relating to an individual is anything form names, email address and phone numbers to shopping habits, internet browsing times and recent Netflix binges.

Thankfully this blog is not about GDPR. But it is about the type of data we collect and use as teachers, about students, and whether we are collecting the right type of data. One of the phrases that I heard when teaching was 'measure what you treasure'. In other words, measure the important things, the things that really make a difference, because if these are the focal point for measurement, then they become the focal point for improvement.

So back to GDPR, or rather, back to the data that companies want to hold on us. Let's take advertising, with Google being the king. Many companies are interested in our details so that they can target adverts to us. Making the adverts more specific works as they don't waste money advertising to people who are not interested and their adverts end up in front of people who might actually want to see them. Personally, I like it. I see adverts for things I like and see less of the ones I don't. And who know's when machine learning improves, maybe they will be able to show me an advert for something I need, but don't realise I need it yet!

How does this link to schools, teachers and students? Let's keep exploring. Advertisers are not really interested in my name, it doesn't tell them anything about me. Maybe it gives someone a crude tool for estimating gender, but it's probably not that accurate.  Age, hobbies and internet browsing times, however, are more valuable as they are things the advertisers can act upon. Even more useful would be the websites I visit, the products I typically buy and the average amount I spend on each product. For the purpose of this post, let's call this general and specific data. General data being things which, on the surface, are useful, like name, but in reality don't tell us much and specific data is the valuable stuff. It's the things which advertisers can act upon to make adverts more relevant.

In terms of the data we track relating to pupils, I wonder how much of it is specific? How much of it can the teachers really act upon, alongside the student, to bring about improvements in learning? Learning being the small, often invisible, incremental changes that a child experiences as they acquire more knowledge. How much of what we measure in schools is able to pinpoint the specific lesson in a specific subject that a child is struggling on at a specific point in time? And how much of it is general? Specific enough to place a child in a particular sub-group, but not specific enough to pinpoint exactly why they are there. For instance, a child making below average progress in maths might be labelled as someone who needs intervention, but, without further exploration, we don't know which element they are struggling on or why they are finding it hard. Teachers marking books is the often the point where people argue this more specific data is picked up, but I fear we are missing vast amounts of data by relying on book marking aloe. It is far too crude a tool to use when talking about whether a student has actually learned something.

In my experience, data such as target grades, predicted grades and even 'working at' grades all fall into the category of general data. Sure, it's kind of useful to be able to see, at a glance, how well a student is performing, but having this information doesn't help the teacher to help the student. The teacher needs more specific data. They need to know exactly what the student is struggling on, or finding too easy, so that, just like the advertisers, they can target their intervention ad make it relevant. The only difference is that where this increased efficiency saves advertisers money, it saves educators time. Time which they don't always have to diagnose the specific thing a child is struggling on and help them with it.

And just like Google isn't manually inputting data relating to people's interests, behaviours and habits, teachers shouldn't be inputting data relating to students knowledge and understanding. The margin for error is too large. Teachers, for all of their training, professionalism and effort, just don't have the data stream coming in from each pupil. As with the advertising, the more data the algorithm has, the more accurate it can target the adverts to the audience. The more information the teacher has, the better their judgement. The fact is though, they don't have enough time to get this information, and I don't see them being given any more to get it either. Relying on book-marking and limited one-to-one interactions with students just don't provide enough data for it to be meaningful. 

We track everything that students do on TheEverLearner.com. This doesn't take in to account the things the student does in the classroom, granted, but anyone who has studied our work on Classroom 21 and Education Reimagined will know that with the Classroom 21 model, teachers have more time to gain insights from students. Our pioneers, who work with the Classroom 21 model regularly, report to us that they spend less and less time at the front of the classroom (because students take tutorials online which are relevant and specific to them). This means the teacher is no longer the 'deliverer of information' from the front and has time to circulate and discuss with students what they are working on, struggling with and succeeding at. This enriches the data stream that they teacher has because they glean specific data, which they can act upon, about individuals. They also have a rich data stream relating to the tutorials and questions on the site which exists alongside the interaction between student and teacher.

We know we have a long way to go until we have a full solution, and we constantly drive to improve, but we also know is that when we equip teachers with lots of rich data about students' learning and empower them to spend time with students one-to-one in the classroom, they can unpick this data on a human level and begin to understand the student more as an individual. This way teachers can target interactions with students more accurately and, ultimately, be more effective in helping them learn.

Data is often a dirty word in schools, but I think it's not really the data that people don't like. It's the cumbersome processing and laborious reporting with little or no impact which makes people mad. If we can use data which is self-generated and presented beautifully and increase the amount of personal, discussion-derived data that teachers elicit from interactions with pupils, we might be in a position to be truly data-centric and enjoy all of the benefits that this brings.