Magnetic schools
The distribution problem that sits at the heart of the SEND crisis
The debate about special educational needs provision in English schools often centres around the level of need: the number of pupils identified as having special educational needs and the extent of their needs. This is rightly a concern as the proportion of children identified as having special educational needs has steadily increased over the last decade (from 1.2m in 2015/16 to 1.7m in 2024/25). In the same period, the proportion of children with an Education, Health and Care Plan (EHCP) has increased from 2.8% to 5.3% and continues to rise at an alarming rate.
However, what is far less talked about is how need is distributed across the system.
Over half of children with an EHCP are now educated in a mainstream school. This proportion has increased steadily in recent years. Almost all SEN Support pupils attend mainstream schools. What is less clear is whether the increase in these numbers represents a real increase in need or simply an increase in recognition or a lowering of thresholds. Either way, there is an expectation that mainstream schools must do more to meet children’s needs. The weight of this hangs heavy on schools.
What may be more surprising to those working in schools is the way these needs are spread across the system. According to a recent NFER report, the distribution of children with special educational needs is far from even across schools. The top quartile of primary schools with proportionally the highest EHCPs have six times the average of those schools in the lowest quartile. This equates to an average of 17 pupils with EHCPs in the highest incidence schools compared to 3 pupils in the lowest. In secondary schools, the difference is a multiple of five; an average of 54 versus 14 EHCPs respectively. These differences have huge implications for schools and the children requiring additional support.
The pattern is the same when we look more broadly at ‘any SEND’ (EHCP + SEN Support), although the difference between schools in the top and bottom quartiles falls to a multiple of two. However, the schools with high levels of EHCPs are not always the schools with high levels of SEN Support. This suggests that some of the inequalities in EHCP rates may be due to how inclined schools are to pursue statutory assessments. More work is needed to understand the data. What we can be sure of is that the distribution of needs, and the distribution of statutory entitlements, is far from even across mainstream schools.
Why do some schools attract children with SEND?
High-SEND schools are divided by the NFER report into two categories: high-EHCP schools (HES) and high-any-SEND schools (HASS). Neither is a homogenous group, but schools within each group share certain characteristics.
High-any-SEND schools are more likely to have high levels of deprivation, low attainment and lower Ofsted grades, particularly for secondary schools. We may speculate that there may be higher incidence of special educational needs in the local area and/or that social needs are to some extent being labelled and managed through a SEN designation. We know that the use of SEND Support as a category varies widely between schools and regions. Regardless, these schools are likely to be experiencing considerable pressure, often without the additional funding and specialist support that EHCPs might bring.
High-EHCP schools are more likely to exist in an area of high EHCP rates, serve more deprived communities (particularly in the case of primary schools), and have low levels of attainment, but not necessarily a lower Ofsted grade. HES in the secondary sector are also more likely to be relatively small. A final significant feature is that these schools are more likely to have an SEN Unit or resourced provision.
There appear to be three key factors at play that may explain SEN clustering:
Demographic: we might expect to see higher rates of SEND in more deprived area.
Categorisation: there are inconsistencies in identification and how needs are categorised and supported.
Push and pull factors: some schools appear to attract a higher proportion of children with special educational needs than other schools.
Those schools that appear to attract children with additional needs have become known as ‘magnet schools’. What makes a school magnetic?
Reputation is a key factor in parental choice. Parental networks, particularly those for parents with children with additional needs, are strong and word-of-mouth is powerful. The more inclusive schools are in their practices, from how they engage with parents, to the quality of their teaching, to the effectiveness of support that is put in place, the stronger their reputation will become.
How schools portray themselves is also important. Headteachers who stand up on open evening and paint the picture of an inclusive school, who open up the SEND department and resource base for parents to visit, and who ensure staff are available to discuss individual pupils and how their needs can be met, will signal that those with additional needs are welcome.
We can see in the data that schools with SEN Units or resource provision also attract children with EHCPs into mainstream provision, as if the presence of this specialist provision signals that this is a school that is ‘good at SEND’. In secondary schools in particular, size and culture also matter. Parents will be more attracted to schools where they feel their child will be ‘known’, and which is calm and friendly.
There are particular features of our system that facilitate magnetism:
the ability for parents to ‘shop around’ and choose which school their child should attend
a funding system that directs income towards schools according to the number on roll, incentivising schools to fill up places from beyond their local area where possible and, where necessary, commission transportation from out-of-catchment areas
the right for parents of children with an EHCP to name the school their child should attend (which trumps all other criteria and the limits of a published admissions number).
The inclusion penalty
The challenge for schools is that being inclusive comes at a very high cost. The funding system does not adequately resource schools who have disproportionately high incidence of special educational needs. The more pupils with EHCPs on roll, the harder it becomes to meet each child’s needs and to fulfil statutory requirements. Those drawn to the school with additional needs but without the funding that an EHCP brings add further pressure on resources and push schools towards seeking more statutory assessments. Furthermore, performance measures are punitive towards schools that have higher levels of need. These disincentives mean that we are almost entirely reliant on the moral purpose of school leaders to want to be, and to be seen as, a truly inclusive school. Inevitably, some pull up the drawbridge and let others carry the weight.
Does it matter that SEND clusters in this way? Clearly it matters if schools are not resourced to meet the needs of children and are punished for being inclusive. If we solved this problem, we would likely see a more even distribution of need across the system as schools would fear less the goldrush of being seen as ‘good at SEND’. But this doesn’t answer the question of whether clustering is in itself undesirable. Would a world in which children with special educational needs were distributed evenly across mainstream schools be a good thing? Is it even possible?
What distribution is optimal?
The first issue to consider is demographics. If special educational needs are not distributed evenly across the population then it will not be possible to achieve an even distribution between schools. We’ll put this matter to one side because it opens a can of worms about the relations between poverty and learning needs.
The second issue is consistency in identification and response. We should seek to reduce the inconsistencies between schools and regions so that a child receives a similar level of support wherever they are in the country. Doing so would ensure we understand better what the actual distribution of need is.
The third issue is local clustering fuelled by the magnetic schools effect. Reducing this effect is desirable for a number of reasons. First, children would be able to attend a school in their local community. Second, the need to travel would be reduced, saving cost and time. Third, children would benefit socially from being in a diverse and inclusive community. Fourth, the school funding system could be simplified and perverse incentives could be reduced.
However, there is a counter argument which is that there are benefits in having specialist provision within mainstream schools. For example, a centre for speech and language difficulties located within a school may mean that the school attracts more children needing this support (thereby creating a cluster) but that the quality of support improves as specialist support can be provided. There is a world in which such provision would not skew the overall distribution of need, and that is a world in which all mainstream schools provided some form of specialist provision.
On the whole, a more even distribution of needs across schools is desirable. But the policies required to achieve this may not be popular or straightforward.
Policy instruments
For this last part of this post, I will focus on the problem of magnetic schools. The policy aim here would be to create the conditions within which all schools are incentivised to be inclusive and to be seen to be inclusive. Achieving this would require changes in funding, accountability, and admissions.
Funding would be needed to achieve two things. First, to ensure schools invest in their universal provision so that it can meet the variety of predictable needs that a comprehensive cohort would bring. Second, to ensure that schools are fully funded to meet exceptional needs. There must be no financial penalty for being truly inclusive.
Inclusion would need to be placed at the centre of a shared conception of school effectiveness. In other words, an effective school is by definition an inclusive school, and an inclusive school must be effective for all. This belief must then be reflected in school accountability: Ofsted, performance tables, and so on.
Once the school system shifts towards being consistently more inclusive, the right for parents of children with an EHCP to name a mainstream school could be limited to the need to access specialist provision in a mainstream setting.
The level of need across the country matters, but so too does the distribution. If we want a fairer, more effective, education system for children with special educational needs, incentivising all schools to be inclusive is a good place to start.
In Australia, we have higher incidence of diagnosis in wealthier areas so higher numbers for what might be less intensive support like anxiety. Diagnosis is expensive here.