I recently met up with Mark Stanley, the Managing Director of Sydney-based Literatu, whose NAPLAN Explorer enables teachers to deep dive into analysing NAPLAN assessment data and to identify trends in learning. It’s one thing to make sense of NAPLAN reports, but another journey altogether to improve learning with data.
Mark’s a ’Big Data in Education’ explorer — he thinks non-stop about how schools can get the most out of linking together all of their data, to help students improve. As schools move teaching, testing and attendance data onto electronic files, the sheer volume of data they produce is rocketing. But here’s the challenge Mark’s working on: how do we use that data to create the learning insights that actually add value?
Practical insights that target teaching
Mark’s starting point is that teachers are hungry for good data. The Teachers Know Best report from the Bill and Melinda Gates Foundation reports that 78% of teachers are convinced that data will help them engage students based on where they are educationally. If a teacher can see at a glance which specific sub-skills each student struggles with, then every interaction can be targeted at the point of need.
As Mark said, “What we really want is a proper longitudinal view of individual students and multi-year school performance. The NAPLAN data gives teachers a great baseline, and every school can request the detailed data at skill and question level. The trick is to turn the data into a practical, day-to-day resource that teachers can use to target their teaching.
Three wins of longitudinal data
Literatu’s tool, NAPLAN Explorer, is an interactive dashboard built alongside our Power BI service. It provides a visual way of analysing NAPLAN data — and it’s well worth clicking on this link just to get a feel for what visualisation can look like.
For example, a teacher that is new to a class can instantly get a six-year view of each student’s NAPLAN performance, broken down into 24 different sub-skills — so not just question areas, but real skills that teachers can relate to, like ‘paragraphing’. This helps the teacher to:
- Identify students’ problem areas. If a student is underperforming in one area, say numeracy, the tools can isolate which specific sub-skill is responsible, e.g. algebra. What’s more, a teacher can see how long this sub-skill has challenged the student.
- Observe trends & forecast performance. By comparing each student’s performance against their class cohort, teachers and parents can predict likely outcomes. This means all stakeholders know what skills to target in order to achieve the best outcome for each student.
- Match curriculum inputs with performance outputs. With longitudinal data, schools can clearly see how different parts of their curriculum delivery are performing. This means they can identify which year and subject groups have the greatest scope for improvement.
Real-time data in day-to-day teaching
What gets Mark really excited, though, is how you can use NAPLAN data as a base for real-time targeted teaching. What he’s doing in his work with schools – to design something beyond NAPLAN Explorer – is to combine the Years 3, 5, 7 and 9 end-of-year data with all sorts of class and school data sets: spelling tests, gradebooks, attendance records and so on.
The point here is that Power BI makes it relatively easy to pull these different strings of data together — and anyone with mid-range Excel skills should be able to do it.
Once you have the data you can use the Q&A function — which, to be honest, is my favourite part of Power BI. It works like a normal web search engine, but for your own data, so you can ask questions in plain English, just like you do with Google. This is the critical step, because data only starts to become valuable to classroom teachers when it can deliver answers to specific, detailed questions. So, a teacher could type in:
“Show the top 10 students at risk of underperforming in maths in Year 9?”
“Which high attendance students are underperforming?”
“Which year groups show the biggest improvement in spelling?”
Note that Power BI allows you to define conditions and language descriptors around your data, to make it easier for you to make it relevant to teachers, leaders, students (or even parents?)
For Mark, one of the biggest upsides of cross-matching data is that it generates solid evidence for teachers, faculties and principals. With precise data they can confidently identify the contributing factors behind underperformance whether it’s attendance, behaviour — or just basic interest in subject matter.
It also means that data is not just being used for diagnostics, it’s being used to answer the really high-value, prescriptive questions, like: “How do I make something happen.” So the clear wins for schools are:
- Solution monitoring. By isolating root causes, schools can target resources. They can also gather the evidence week-by-week to see if remedial actions are working.
- Personalised teaching. As Mark says, ‘data is inclusive.’ With data-based diagnostics, teachers and parents are more likely to collaborate on the specific actions needed for change, and have a less emotion-driven debate.
- Classroom efficiency. All the time that teachers are over-burdened with delivering standard content, anything that helps target interventions will increase their impact.
Do schools have the data to do this?
While NAPLAN Explorer is ahead of what most schools will have today, the Big Data revolution is increasingly reaching education, and it’s better for schools and teachers to be in a position of leading the change, rather than being victims of it. As Mark points out, companies are already using business intelligence solutions for real-time analysis of everything from supply chains, to customer purchasing to employee performance. Insights mean fast reactions. Mark reckons schools should prepare now for the change that’s around the corner when everybody will be talking about how to improve learning with data.
“This isn’t about using today’s buzz phrase – ‘Big Data’ – in education,” he says. “The challenge is to start capturing data today, so that even if it’s another year before you start crunching it, you will have the raw material to light up every child’s development. Real-time, child-centric data that is instantly visualised – and which you can probe yourself to answer specific questions – is incredibly powerful. Who doesn’t want that?”