Schools are complicated, multi-faceted and evolving institutions.
However, they are frequently modelled as if they were much simpler, to convince ourselves that we have a measure of control. We talk about ethos, learning goals, KPIs, development plans and evidence-based initiatives, as if each one of them precisely related to some aspect of the school.
Such ‘levers’, we imagine, can be manipulated to make some desired change to the way the school runs. Given a combination of such levers, the institution can be effectively steered, like some proud luxury liner deftly navigating around the bathtub of government regulations.
This level of control is illusory. Any initiative will interact with other factors, internal and external, to bring about change in a less than predictable manner. Of course, it is still better to attempt navigation using these rather loose levers than to settle for drifting according to the tide.
In any school, we have financial data, recruitment data, surveys and appraisal information along with baseline and value-added data. Each of these instruments tries to capture some aspect of a complex human environment in a starkly numerical format, which is simultaneously a strength and a weakness.
When badly used, data can re-enforce the idea that we have greater control than is in fact the case.
Consequently, one of the most crucial goals of data analysis is to develop an understanding of its limitations. Those of us who work with data in education have a particular responsibility to explain the level to which it can be trusted and the dangers of placing too much faith in it.
(Some of our colleagues will always object to any attempt to measure human characteristics, such as creativity and talent, and so be suspicious of data in all its forms. Others will point to the use of averages and trends and object that there is no such thing as an average pupil.)
It is important to remember, data is data – it is the humans analysing the data that make the judgements and give it meaning.
When looking at data in schools, the answer is to look at it alongside holistic, longitudinal, and contextual information and ask the question: ‘Do you recognise the pupil/department in this data?’
If you do, then the data is valuable as it confirms your impression and promotes trust to your judgement. If not, then that should trigger a conversation about the validity of the data in that context.
Whatever the accuracy of the data, the conversation is the important result.
Correctly and non-confrontationally framed, a proper conversation can only lead to greater understanding of the data and the human side of the people involved. The worst thing we can do is focus on a single aspect of data and use that as a driver for change.
A successful school will always have a strong ethos and vision, maintained by the leadership team. To use data really successfully, you need a good quality whole-school data system with effective management:
If all of this is in place, which is no easy task, then the use of data can have a real impact on student achievement and school performance.
As we emerge from these extraordinary pandemic times, we will hopefully find a moment to take pause and reflect about the best features of our schools as buildings filled with human relationships. As a great deal can be achieved remotely, we need redouble our efforts to support and develop those education goals that can only be achieved in person.
A broader perspective on data and its integration into management approaches works to help us focus on the human side, not to dehumanizingly reduce things to pure numbers.