We need to be more discerning about disadvantage

Of all the many dividing lines that bisect England’s education system – north versus south, boys versus girls, city versus countryside, religious versus secular – by far the most pervasive and significant is the one that separates children from poor families and their more affluent peers.

On average, poorer children tend to show lower attainment at every stage of their school careers. They are also more likely to attend schools with lower Ofsted ratings, to have fallen behind during the pandemic and go to on to less desirable post-school destinations. Indeed, some other well-known educational discrepancies, such as the north-south divide, are in large part just the poverty gap by another name.

This importance makes it all the more surprising that the way we think about and measure educational disadvantage is rather simplistic – overly so in our view. The most common measure for schools is the proportion of children who are eligible for the pupil premium (roughly equivalent to the proportion eligible for free school meals).

Pupil premium payments to schools are important in mitigating some of the educational disparities that would otherwise affect poorer children, but as a school-level indicator of disadvantage the Pupil Premium measure suffers from some major shortcomings:

• It’s a threshold metric. In other words, any given child is either eligible or not. This isn’t a good way to represent poverty, which exists on a continuum.
• Where a family’s income increases to above the threshold, their child’s eligibility expires after six years. Yet the attainment of these children typically continues to be below that of their more affluent peers.
• It focuses on income deprivation. This is certainly important, but there are many other forms of deprivation: health, crime, housing, environment and so on. These are often tracked in official statistics and many of them have an impact on education, or at least on schools, yet we tend not to talk about them.

We have explored this issue in a new analysis. For example, we identified schools at which exactly 20 per cent of children were eligible for the pupil premium (about average for England). Despite the fact that these schools were literally indistinguishable on this measure, their local neighbourhoods and catchment areas display widely varying income profiles.

What’s more, these same schools showed very varied characteristics in terms of other deprivation measures. For some, local crime rates were high, for others they were low. Some were located in areas with good health outcomes, others in places where these were much worse. Levels of environmental deprivation varied all the way from terrible to great.

In other words, it’s not possible to rank schools on a simple one-dimensional list from best-off to worst-off without losing a lot of important information about the true social context in which each one exists.

But does any of this matter? We think it does. Educating children in areas with poor health outcomes is (or at least should be) different from doing so in areas where health is generally good. Ditto for crime, the environment, participation in higher education, social mobility, transport links and so on. Each of these tells us something different, and all are distinct from the school’s pupil premium measure.

As an example, a school we are supporting in the East Midlands is undergoing a major review of its sixth form provision. Ostensibly, pupil premium levels are not unusual, but attracting students to stay on and follow suitable courses has proved very difficult.

However, analysis of local rates of participation in higher education showed very low historical levels in the neighbourhoods around the school. It wasn’t income that was the key source of deprivation, but a long-established lack of educational ambition and opportunity, and an insufficient understanding of higher education’s benefits. The school decided that in order to break this trend it needed to convince parents. Pupil premium alone gave no insight into this problem.

Disadvantage is not all or nothing. It isn’t even one thing. Children, schools and communities exist in a wide range of local contexts with different strengths and challenges. We need to take these into account. We have now begun to do so in SchoolDash’s profile pages, which are used to provide summary information about schools, and we plan to build on this further in future versions and forthcoming analyses.

After all, the information exists so why no use it? Whether a school leader, parent or policymaker, we encourage you to do so too.