Russell Hobby, general secretary of heads’ union the NAHT, writes: 

The pupil premium may be remembered as the coalition government’s most valuable legacy. Let’s leave aside that it really should be built on the foundation of a fair national funding formula to work properly. In a spirit of generosity we will focus on the determination to break the link between disadvantage and educational underachievement, matched by the reallocation of resources.

The pupil premium is driven by registration for free school meals (FSM). I want to look here at whether free school meals work as a measure of disadvantage; for the whole edifice rests on this. There are deeper issues about how we connect poverty with achievement.

Many headteachers say that FSM does not perfectly capture poverty. There are families who are not eligible who are struggling to get by. More fundamentally, poverty itself is a proxy. Poverty does not inevitably cause educational underachievement. It is caused by things that also harm education - family dysfunction, tragedy, low aspirations, lack of knowledge about the ways to work the system, addiction, long term sickness and neglect. We all know of people who have excelled educationally from the poorest of backgrounds and of families who prize education above all else as a route out of poverty.

FSM is a proxy of a proxy, which is bad enough. It also enables a subtle form of covert selection. Having a high FSM intake triggers funding and can make demonstrating progress easier - if you get the “right” poor pupils. You want those families who, although poor, do prize education. You make it clear, for example, that you will expect a lot of parents and place high demands on them. You suggest that children with special needs might be better served by the school down the road. It happens. And it shouldn’t.

FSM’s virtue is its simplicity. The time has come for a debate on whether we can find a practical replacement that gets closer to the underlying causes of disadvantage.

What if we replaced FSM with a measure of the prior attainment of pupils? For example using key stage 2 results for secondary and the forthcoming reception baseline for primary. This would target resources at children falling behind whatever the reasons. It puts a lot of pressure on the measure chosen as the baseline, of course, particularly if conducted in house like that in reception, but has a simplicity to it.

As I understand it, and this is hearsay but the idea is interesting, the Danes go one step further back and use parental attainment as their proxy. There is a neat connection to social mobility here.

Or we could pursue more sophisticated measures. IDACI, the deprivation index, is one system. I know of an academy in Manchester that targets interventions using a sophisticated set of health measures derived from NHS work.

As I say, we need a debate on this because every solution has pros and cons, and with sophistication comes complexity. However, if we are serious about breaking the link between educational success and disadvantage we need to get serious about defining disadvantage as it manifests itself in education.

This blog first appeared on the NAHT website.