On 28 March 2022 the Department for Education (DfE) published the white paper ‘Opportunity for all: strong schools with great teachers for your child’, setting out the government’s ambitions to ‘level up’ education across the country and making the case for a fully trust-led system.
Academisation has long been one of the government’s key approaches to raising standards in the areas that need it most. The recently published schools white paper announced the government’s intention that all schools in England will be in a strong multi-academy trust (MAT) or in the process of joining or forming one by 2030.
This pledge, which would see 10,000 primary schools and 700 secondary schools convert to academy-status over the next seven-and-a-half years, will be initially targeted in the 55 ‘Education Investment Areas’ that government has pledged to prioritise with attention and funding. The government argues that these areas are ‘cold spots’ for literacy and numeracy outcomes that have the ‘most urgent need for the benefits that strong trusts can bring’.
DfE’s implementation plan, published at the end of May, confirmed that these areas will be the main focus in the first years of this reform, starting with schools who receive successive Ofsted judgements below Good.
Where these schools are not yet academies, the newly titled Regional Director (formerly Regional Schools Commissioner) will make a final decision on the academy trust they will join. Where an Inadequate or Requires Improvement school is already part of a trust which is judged as unable to deliver the necessary improvements, the academy will be rebrokered to a ‘stronger trust’.
There is growing need for a robust definition of an effective school group
The phrase ‘strong trust’ signals a quiet but significant shift in the government’s case for academisation. Research from across the sector has frequently found that academisation is no silver bullet to school improvement. Our recent report with the UCL Institute of Education focussed specifically on those schools that had consistently received Ofsted ratings below Good and found that academisation brought limited improvements to primary schools, in particular.1
Our earlier studies have shown that there are many excellent high performing multi academy trusts, and there are also as many low performing trusts which are not delivering strong outcomes for their pupils.2 Given this, DfE is now emphasising that it wants to support the high performing ‘strong trusts’ to expand, while creating new powers to tackle underperformance among the others.
These proposals therefore hinge on what exactly the government means by a ‘strong trust’.
We must be able to identify the most effective MATs, especially those most effective at delivering good outcomes for the most vulnerable pupils, so that we can be reassured the right ones are being supported to expand.
This means that the defining detail that lies behind the watchword ‘strong’ is crucial to the success of this reform. What do we mean by a ‘strong’ MAT? Even if we agree on a definition, how can we reliably know which MATs fit that definition?
The white paper sets out five principles of its definition. These cover high quality and inclusive education; school improvement; strategic governance; financial management; and finally, workforce. These five areas are set out with some initial detail on page 49 of the white paper.
We agree that ministers should consider a broad range of measures when it comes to trust effectiveness. Indeed, we have begun work on a major research programme to explore the features of effective school groups, defining effectiveness as spanning four domains: pupil outcomes; pupil inclusion; workforce sustainability; and financial efficiency. This multi-year project will, we hope, culminate in robust and credible measures of trust effectiveness across these four areas, and provide the evidence needed to identify the most effective multi-academy trusts in England. In addition, we will also codify the key practices behind those strong outcomes.
This research is ongoing, and we will be publishing various outputs throughout the rest of 2022. This includes a working paper on how to measure workforce sustainability in school groups, focused on retention of teachers and leaders; the first results measuring the inclusiveness of school groups; and early findings from our survey of operation decisions, strengths and challenges in school groups. You can read more about how your school can contribute to this research here.
Whilst this research is forthcoming, government reforms will already get underway throughout the coming academic year. There is an urgent need to understand:
- the volume and proportion of pupils in academy trusts in Education Investment Areas,
- which trusts are already established in these areas and,
- what their outcomes tell us about which trusts are in the best position to expand and which are most likely to deliver positive improvement for the most disadvantage pupils
In order to explore the strength of trusts in relation to attainment and pupil inclusion, we have used a range of sources including from our existing reports. Some of this data is historic (2017-2019): we use this data from existing reports to give an initial indication of trust performance in Education Investment Areas, whilst we are working on new and updated metrics which will be published over this year and the next. Due to the age of this data and the disruption caused by the COVID-19 pandemic, analysis of trust performance in these areas may not fully reflect the current situation for trusts in England. Analysis of trust size, pupil numbers, and school location reflects the latest Get Information About Schools (GIAS) data as of June 2022.
Comparing disadvantage gaps across EIAs and non-EIAs
To identify EIAs, the DfE has used a 3-year composite measure of pupil attainment at Key Stage 2 and Key Stage 4 to assess the local authorities with the lowest attainment outcomes between 2017 and 2019.
This metric is a clear and widely understandable method of assessing areas in England with lower educational outcomes. However, it is not without limitations, particularly given that even a relatively small local authority can cover a wide range of localised circumstances. It may even result in areas with some of the worst outcomes for disadvantaged pupils not being selected as EIAs. If the low attainment of disadvantaged pupils is masked by significantly higher outcomes for their non-disadvantaged peers, the overall results for the local authority may mean it does not qualify as an EIA. This potential ‘masking’ of disadvantaged pupils with lower attainment risks these pupils falling further behind if their areas miss out on additional funding, a possibility which seems to run counter to the government’s ‘levelling-up’ agenda.
The 55 EIAs are a combination of: the lowest performing local authority areas; local authorities that include one the Department’s existing ‘Opportunity Areas’; and other areas already identified for support. The list of Priority Education Investment Areas is drawn from this group and comprises the very lowest performing areas that also have high levels of deprivation, alongside Opportunity Areas. In the case of the six Opportunity Areas that are local authority districts (as opposed to upper tier local authorities) it is the district that becomes a PEIA rather than the authority as a whole.4
We compare the government’s selection of Education Investment Areas against the data on the disadvantage gap (a measure of the difference in attainment between disadvantaged pupils in a given local authority and all non-disadvantaged pupils nationally) at Key Stage 4.5
Figure 1. Disadvantage GCSE grade gap in English and Maths at KS4 by local authority
Figure 1 shows that while EIAs and PEIAs dominate the local authorities with the largest disadvantage gap, there are a number of notable outliers. For example, Sheffield and Cheshire West and Chester have the 5th and 8th largest disadvantage gaps in England at Key Stage 4, respectively, yet have not been selected as EIAs. Conversely, Luton has been selected as an EIA despite having a smaller disadvantage gap than 54 local authorities not selected as EIAs.
This is of course a comparison of KS4 outcomes alone, whilst EIA selection criteria also cover KS2 outcomes. This analysis nevertheless raises questions as to government plans for closing the disadvantage gap nationally: areas with some of the largest gaps in the country will not be receiving the same focus and investment as others with similar size gaps for disadvantaged pupils.
Academisation in PEIAs
DfE’s case for establishing a fully trust-led system relies partly on the claim that schools benefit from being a part of a large trust of more than 10 schools or 7,500 pupils due to the scale of the trust improving financial stability, providing a well-supported workforce, and driving school improvement. The DfE expects most trusts to be on a trajectory to reach this size.
This raises the question: if improving educational outcomes can be driven by moving more schools into multi-academy trusts in EIAs, how do academisation rates in EIAs compare to the nation as a whole? In this section we will focus on PEIAs, as these areas will receive additional funding and therefore be more likely to targeted for more rapid academisation.
Figures 2.1 and 2.2 show the proportion of pupils already enrolled in a school belonging to a MAT in England in the primary and secondary phases respectively.3 Across England, 36 per cent of primary school pupils are already educated within a multi-academy trust. In PEIAs, this proportion sits at 39 per cent. In the secondary phase, 60 per cent of pupils nationwide are enrolled in a MAT school, compared to 68 per cent in PEIAs. Therefore, a greater than national proportion of pupils in PEIAs attend a school that is already part of a MAT, particularly secondary pupils. Nevertheless, PEIAs themselves vary greatly by this ‘MAT-educated rate’, with Nottingham among the highest in the country at 94 per cent of secondary phase pupils attending MAT schools, whereas Liverpool is among the lowest at just 28 per cent.
The average size of a MAT nationwide stands at 6.4 schools, while this average is much higher for trusts operating in PEIAs at 10.5. Similarly, the average PEIA trust has a higher total number of pupils at 4,700 pupils, compared to the nationwide average total number of pupils for trusts of 2,750. It would appear therefore that, in Priority Education Investment Areas, the average MAT is already operating at DfE’s preferred scale of 10 schools or more.
Figure 2.1. Proportion of primary phase pupils in school belonging to a MAT by local authority, England 2022
Figure 2.2. Proportion of secondary phase pupils in schools belonging to a MAT by local authority, England 2022
Figure 2.3. Proportion of secondary pupils in LA maintained schools and academy trusts by trust size (Priority Education Investment Areas)
In summary, academisation is well underway in most Priority Education Investment Areas, and more secondary pupils in these areas attend a school in a MAT than the national average. However, most pupils who attend academies are in trusts of fewer than ten schools. Furthermore, there remain some PEIAs with lower levels of academisation. In Liverpool and Tameside, more than 50 per cent of secondary pupils attend an LA-maintained school, and Liverpool, Rochdale and Tameside currently have no large trusts operating in their areas. This snapshot reflects the national picture in micro-scale: a mixed and fragmented system. It seems apt that the government’s implementation plan states that Regional Directors will work with each local authority using Area Based Commissioning to address local needs, given the different states of academisation in each area. Capacity is only one question, however, and quality is another question which we address in the next section.
Trust outcomes in EIAs
If outcomes are to improve for pupils in EIAs by their schools joining larger stronger MATs, it becomes crucial to identify which MATs are already operating at scale in EIAs, and how they are currently performing.
As outlined above, we are currently developing robust and up-to-date metrics of trust performance.
In the meantime, we turn to our historic publications for initial indications of trust performance in EIAs. Our existing publications allow us to explore school improvement and a measure of pupil inclusion known as ‘unexplained exits’.
This section draws specifically on two publications: ‘School Performance in Academy Chains and Local Authorities – 2017’2, published 2018, and ‘Unexplained pupil exits from schools’6, published 2019. Both reports use data from the National Pupil Database and School Census for pupils finishing KS4 in 2016/17.
These are existing published measures, but for the first time we link trust-level outcomes together to explore how multiple measures of trust effectiveness can help us identify the most effective MATs.
To assess school improvement, we use a contextualised improvement measure of Key Stage 4 attainment that accounts for the range of pupil-level characteristics that are known to impact attainment.
For pupil inclusion, our measure is the average termly rate of unexplained exits. By unexplained exits, we mean exits from a school to either another school, alternative provision or an unknown destination, where those exits do not appear to be driven by families or a formal exclusion. Our research in this area has found that larger MATs (those with at least ten schools with secondary pupils) all have above average rates of unexplained exits. Note that, unlike the KS4 improvement measure, this measure does not control for underlying pupil characteristics such as deprivation. This is because to contextualise this measure of exclusion would be to expect a different (higher) level of exclusion for disadvantaged pupils as the norm, which does not fit with principles of inclusion.
Data used in this section is from academic year 2016/17 and trust membership and performance may now be substantially different. We therefore use these figures as indicators only. DfE’s metric for selecting EIAs tracks attainment from 2017-2019, and so trust performance in 2016/17 will have contributed to the selection of these areas. In addition, while we use these metrics to understand trust performance within Education Investment Areas, it should be noted that trust metrics also include results from schools located outsides of these areas. Our rationale is that trusts already working in education investment areas are those most likely to expand. Assessing the performance of all their schools, particularly when this performance is contextualised for the characteristics of the pupils attending those schools, is a useful insight into the overall effectiveness of these trusts.
Figure 3.1. Key Stage 4 performance of academy trusts operating in Priority Education Investment Areas by local authority
Figure 3.1 presents 2017 KS4 contextualised improvement for all trusts currently working in each of the Priority Investment Areas for which data is available. Each point represents a trust (some trusts are operating in multiple local authorities), with the vertical lines to illustrate the range of KS4 contextualised improvement. The figure shows that the majority of trusts that are working in PEIAs did in fact improve their Key Stage 4 attainment results between 2015 and 2017. However, not all trusts operating in PEIAs succeeded in improving attainment during this time – eight of the 18 local authority PEIAs contain trusts with a negative contextualised improvement measure.
It should be noted that this improvement is calculated across all schools in the trust, and not just those located in EIAs. Whilst we do not currently have data to isolate how trusts have driven improvement specifically within these investment areas, this trust-wide data provides insight into the overall effectiveness of the trusts which are most likely to be the first to expand into these areas. The improvement measure is contextualised for underlying pupil characteristics, meaning we are accounting for the some of the key differences between the different local areas these trusts are operating across.
This therefore gives an indication which Priority Education Investment Areas already have local trusts which are delivering strong school improvement (for example Knowsley, Middlesbrough and Doncaster), and which areas may not yet have strong candidates for trust expansion (such as Hartlepool and Rochdale, though there may now be some high performing trusts in these areas which are not included in our data).
Figure 3.2. Key Stage 4 performance against average termly exit rate per cent of trusts in England
Figure 3.2 maps the Key Stage 4 attainment results of all trusts in England where data is available against their average termly unexplained exit rate. The quadrant lines signify the national average: 0.0 for contextual improvement, and 1 per cent rate of termly unexplained exits.
Overall, there appears to be a weak positive correlation between the termly rate of unexplained exits and contextualised improvement at KS4. This is consistent with the fact that KS4 performance is measured on those pupils who are in the trust at the end of KS4, and excludes the attainment of pupils who have exited the school, whether due to permanent exclusion or an unexplained exit, and we know these pupils tend to be lower achieving. The majority of trusts have average termly unexplained exit rates equal to the national average. EIA trusts appear to be roughly in step with trusts nationwide, with few outliers in either attainment or exit rates.
There are four trusts, all of which operate in EIAs, which have above-average improvement and below-average termly unexplained exit rates. Based on these outcomes, these trusts appear to be the better candidates for discussing expansion or sharing best practice with other trusts. Nevertheless, these are just two measures of performance, and a greater breadth of information is needed before making conclusions about high performance or inclusivity. For example, it is important to looked at both unexplained exits and a trust’s rate of permanent exclusions, as those with lower rates of unexplained exits may still have higher rates of permanent exclusions.
Conversely, there are five trusts, three of which operate in an EIA, with above-average improvement but which also have higher-than-average termly rates. By just looking at pupil outcomes these may appear to be high-performing trusts, but coupling these outcomes with their unexplained exits data raises questions around inclusion that should be addressed if there is possibility of expanding these trusts.
Overall, this small snapshot of data on trust performance demonstrates the importance of looking across multiple measures, and particularly looking beyond pupil attainment to consider pupil inclusion and unexplained exits. This is exactly what DfE intends to do, and they must ensure they use robust measures and quality evidence to put their definition of a strong trust into practice, and particularly that they deliver on the commitment to valuing inclusion as well as attainment.
Given the strong presence of large MATs already operating in PEIAs, and the high level of academisation already underway in these areas, policy makers must be particularly discerning in choosing which MATs to expand across England. In addition, these decisions must also carefully consider the levels of disadvantage and deprivation in these schools and the areas they operate to ensure funding is available for the areas most in need of support.
To be clear, we would not suggest that these metrics should be used for accountability, or as a blunt instrument for decision making. They are objective evidence which should inform discussion among policymakers, identifying areas where trusts might support others, areas where trusts need to demonstrate improvement before expanding, and areas where trusts need stronger government support for improvement.
The reforms contained in the white paper will undoubtedly leave their mark on England’s school system, as MATs expand and merge to become even more prominent players in improving educational standards. It is by no means a given that MAT expansion will raise educational standards, but a deep understanding of which are the best MATs and what sets them apart from the rest will improve the chances of delivering on the white paper’s promises.
- Education Policy Institute, UCL Institute of Education, “‘Stuck’ schools: Can below good Ofsted inspections prevent sustainable improvement?”, published June 2022.
- Education Policy Institute, ‘School Performance in Academy Chains and Local Authorities – 2017’, published 2018.
- Get Information About Schools, Downloaded May 2022, percentage of all pupils (maintained mainstream) attending schools flagged as supported by a multi academy trust.
- Department for Education, ‘Priority Education Investment Areas – selection methodology’, published 2022.
- Education Policy Institute, ‘Covid-19 and Disadvantage gaps in England 2020’, published 2022.
- Education Policy Institute, ‘Unexplained pupil exits from schools’, published 2019.