How data can inform teaching

A user's guide

---

A recent article in the The Hechinger Report argued there is no strong evidence that analysing data leads to improvements in teaching. That’s a sobering null result, given the amount of time that teachers and school leaders typically spend poring over technicolor spreadsheets and bar-graphs.

In addition, there have recently been calls on Twitter by successful school leaders to focus on formative assessment in the classroom and let the data look after itself.

And the entire neoliberal education project was based upon the assumption that if strong external measure were in place, teachers would work out the best ways to improve performance against those measures. Yet, grand projects such as America’s ‘No Child Left Behind’ policy have not led to large gains in performance. Is it time for a rethink?

Perhaps. However, as a data user, I am convinced that there are more and less effective ways of using it. If we want data to have an impact on teaching then this is what I suggest.

Link data to research evidence

The first problem we have when using data is that while it can tell us that something is wrong, it cannot generally tell us how to fix it — I will discuss one exception below.

It is worth bearing in mind the ‘politician’s syllogism’:

1. We must do something.

2. This is something.

3. Therefore, we must do this.

In the face of disappointing data, it is tempting to reach for something — anything — in response. If that something is not based upon evidence then why would we expect it to make a difference? If you are a reader of this Substack then you know that I think the evidence supports aexplicit teaching. However, there is still a lot of detail to unpack within this model and so looking for relevant research — such as reading summaries of research on retrieval practice — adds value.