Be careful what you wish for

.

This extended post is about the selection green paper and the prime ministerial speech preceding it.

I come at this issue from a different position to most.

It is of course essential to ensure that the government’s proposals do not unduly disadvantage the majority of learners.

But it is equally important to consider their impact on the minority that the government expects to benefit directly from increased selection.

For the sake of convenience – and sidestepping the definitional and terminological issues – let us call them ‘disadvantaged high attainers’ or the ‘disadvantaged most able’.

The first section below sets the issue in context and explains my title. I have proposed seven core principles which I invite you to debate and dispute. Consensus would be great but, failing that, discussion helps to expose the ideological fault lines that bedevil any rational discussion of selection, or wider support for the most able.

The middle section critiques the government’s proposals, as set out in the green paper and the associated speech. It highlights important questions and omissions and supplies additional background material that helps to make sense of the text.

The final section presents conclusions based on this analysis, identifying several adjustments that could make the government’s proposals more coherent and potentially more palatable.

It compares the proposals with the core principles and recommends a series of alternative measures to help the disadvantaged most able.

 

Setting the scene

 

Why that title?

The title refers to the scenario I outlined in a previous post, which preceded the speech and green paper

It questioned the very limited initial response to the commitments in ‘Educational Excellence Everywhere’, continuing:

‘So there are grounds to fear half-hearted tokenism, but that would also play directly into the hands of the pro-grammar school lobby.

For want of a convincing alternative we face the very real prospect of more extensive between-school selection, perceived by its supporters as a meritocratic driver of social mobility, despite the lack of evidence to support their case.

We risk substantial sums of taxpayers’ money being diverted towards a fresh tranche of politically totemic institutions, regardless of the substantial deadweight costs, at a time when the wider schools budget is under particularly heavy pressure.

Small groups of disadvantaged learners might benefit, typically at the expense of their lower-attaining peers, but the vast majority of ‘poor bright kids’ will continue to be neglected, suffering from our system-wide confusion over how best to meet their needs.

National excellence gaps between advantaged and disadvantaged high attainers will not be closed, indeed may grow wider, and England will continue to struggle in comparison with leading international competitors.’

For reasons they have not divulged the Morgan-Gibb partnership opted to soft-pedal on their own commitments. They could point to some fine words but little substantive action to date. They could not convincingly argue that they already had the matter in hand.

Then along came the Greening-Gibb partnership, which had no time even to state a position on the white paper before being pushed aboard the runaway selection juggernaut so precipitously launched by Downing Street.

To mix metaphors, they are being railroaded towards a solution that is both politically risky and educationally unsound.

I wished, often, for stronger emphasis on meeting the needs of the most able learners, especially those from disadvantaged backgrounds. Now I have it.

But this new-found interest is not driven primarily by those needs. They are subordinate to the school places crisis, the politically totemic value of grammar schools and, ultimately, the opportunity to distract us from the Brexit impasse.

Prime Minister May wants to distance herself from Cameron and nail her own colours to the mast. Meritocracy and social mobility are the new watchwords, with a particular nod in the direction of those ‘just getting by’ (I have called them the ‘non-advantaged’ below).

She wishes to be seen as the vanquisher of vested interests but, in showing scant regard for the evidence, has succeeded in alienating most of the education sector. Just like Gove she appears to be backing populism over expertise.

For all Greening’s insistence on the need to ‘have the conversation’, I don’t see this ending well for May.

If she compromises heavily on the substance of the green paper the vested interests will be seen to have triumphed once more – and that is even worse than failing to take them on in the first place. If she fails to compromise the educational establishment will become ever more intransigent.

Greening and Gibb are caught between a rock and a hard place, but May will find it difficult to pin the blame on them.

So she too should have been more careful what she wished for.

But what troubles me most of all is that both sides of the selection debate are focused almost exclusively on protecting the majority of learners who will not access new selective places.

The government is fixated on structural solutions that increase the supply of school places. It is overemphasising safeguards that supposedly secure system-wide benefits, because it recognises that its position is otherwise untenable.

The opposition is fixated on segregation, but concerns itself solely with the goats, without a passing thought for the sheep. Unlike old ‘new Labour’ it is not apparently interested in increasing the stock of sheep, or finding an alternative way forward that benefits them as much as the goats.

Neither holds equity and excellence in equilibrium.

Few in the education sector can envision an alternative policy response, beyond expressing a vague desire for more ‘high quality’ comprehensive schools with plenty of ‘excellent teaching’.

There is no credible narrative explaining why wholesale selection is not the best way of meeting the needs of high attaining learners from disadvantaged and non-advantaged backgrounds.

So in this post I want to ask – and begin to answer – the following questions:

• Will the green paper proposals provide the best means of addressing the needs of disadvantaged and non-advantaged high attainers?

• If not, why not?

• What alternative approaches would be preferable?

• What adjustments to May’s proposals would be necessary to minimise their potentially deleterious effect?

.

The disadvantaged most able

The size of the disadvantaged most able population depends critically on one’s definitions of the two variables.

Most of the available evidence relates to old-style KS2, GCSE and A level achievement by FSM-eligible and pupil premium eligible learners (formerly eligible in the case of most A level data).

However it is defined, we know that this population is much smaller than it should be – and that it grows smaller still as it progresses through the key stages.

There are also stubborn and conspicuous ‘excellence gaps’ at each key stage between the performance of advantaged and disadvantaged high attainers. To date these have proven remarkably resistant to the impact of pupil premium.

For example:

• In 2015 only 11,639 disadvantaged (ever 6 FSM and looked after) learners achieved a KS2 fine grade level of 5.5 or above compared to 70,330 non-disadvantaged learners – six times as many.

• In 2015 Only 5,893 disadvantaged (ever 6 FSM and looked after) students managed an average A grade across GCSE English and maths, while 58,995 other students did so – ten times as many.

• Of the 11,700 students who were FSM-eligible in Year 11 and took A levels in 2011, just over 500 (4.3%) achieved AAA+ at A level. In 2014 that percentage was 4.2%; in 2015 it was 4.4%. The corresponding percentages for all other pupils were 9.9% and 9.6% in 2014 and 2015 respectively.

• In 2014 only 31% of disadvantaged pupils with L5a in English went on to achieve A* at GCSE with a further 45% achieving an A grade. However 55% of non-disadvantaged pupils with KS2 L5a went on to achieve A*, while a further 36% achieved an A grade.

• In 2014 only 31% of disadvantaged pupils with L5a. in maths went on to achieve a GCSE A*, while a further 33% achieved grade A. The corresponding percentages for non-disadvantaged pupils were 51% and 32% respectively.

Schools alone cannot be expected to eliminate the negative impact of socio-economic disadvantage on high attainment but, given the will and system-wide support, these excellence gaps could be much reduced.

The critical question is whether increasing between-school selection is the best way of tackling this problem.

.

Earlier work on selection

I have written several previous posts about selection and grammar schools and will try not to repeat myself here.

They include, in chronological order:

• The politics of selection: Grammar schools and disadvantage (November 2014) a magnum opus covering the recent political history of selection up to that point, the evidence base on the socio-economic inclusiveness of grammar schools and initial progress on admissions reform.

• How well do grammar schools perform with disadvantaged students? (December 2014) which explores the achievement of disadvantaged students at different grammar schools on a range of performance measures under the old accountability regime.

• If not grammar schools, what? (October 2015) which ponders how the system might best support high-attaining learners from disadvantaged backgrounds. What combination – of between-school selection strategies; within-school selection and differentiation strategies; and national student programmes – is optimal?’

• Which grammar schools want more pupil premium students? (November 2015) which analysed the 2016/17 admissions arrangements of all 163 grammar schools to establish how many had introduced admissions reforms and in what manner.

• Policy Exchange to the rescue? (February 2016) which dissects the proposal that organisation floated to establish up to 50 ‘super-selective’ schools.

Several other posts propose ways of strengthening support for the disadvantaged most able – I have linked to some of those in the text below.

.

Seven core principles

I want to lay out the core principles upon which this analysis rests. I shall use this as a benchmark against which to assess the new policy framework proposed in the green paper.

You may or may not agree with all of these but, if you disagree, do please consider what you would add, remove or adjust – and exactly why you would do this.

• All learners have an equal right to an education that meets their needs, regardless of background and prior attainment. No learner should be denied access to suitable learning opportunities tailored to their needs.

• Pursuing the twin priorities of raising standards and closing advantaged-disadvantaged attainment gaps means giving relatively greater priority and dedicating relatively more resources to learners from disadvantaged (and non-advantaged) backgrounds, regardless of prior attainment.

• There is no equivalent justification for prioritising or disproportionately resourcing low attainers over high attainers, or vice versa. A disadvantaged (or non-advantaged) high attainer should attract more priority and more resources than an advantaged low attainer.

• Attainment gaps must not be closed by holding back improvement amongst learners from advantaged backgrounds. Both advantaged and disadvantaged/non-advantaged learners should continue to improve, the latter at a relatively faster rate.

• Impact will be maximised through the efficient use of scarce resources. That means minimising deadweight costs, investing in standards over structures so capital costs are avoided and, wherever feasible, targeting support at the learners rather than their schools.

• Education tailored to background and prior attainment is not always most efficiently and effectively provided through universally inclusive learning opportunities (whether or not provided in the same institution). Some opportunities will be restricted to the subsets of learners most likely to benefit. This requires selection in some form.

• Selection operates at many different levels and in many different contexts, including between-school, within-school and within-class selection. None is inherently superior to the others and they may operate independently or in combination. Key considerations when establishing the optimal pattern of selection include need, flexibility, capacity, impact and efficiency.

.

The government’s proposals

This section of the post unpacks the government’s argument and exactly what it wants to do (insofar as it knows at this early point in the policy development process).

The first sub-section deals with the rationale, as revealed in May’s speech. The second examines the specific proposals – foregrounded in May’s speech but set out in much greater detail in the green paper.

.

Fisking May’s speech

Prime Minister Theresa May gave her speech, ‘Britain: the great meritocracy’, on 9 September 2016.

It was accompanied by a press release ‘PM to set out plans for schools that work for everyone’ and there was also an article in the Daily Mail.

.

Twin leitmotifs

The speech has twin leitmotifs:

• Meritocracy as a guiding principle (‘meritocracy’ appears 11 times) and

• Diversity of provision offering individual support for every child (‘every child’ appears 13 times, ‘diverse’ or ‘diversity’ 10 times and ‘individual’ six times)

Meritocracy is initially associated with more extensive support for high attainers, potentially extending beyond between-school selection, but the remainder of the argument is focused exclusively upon the latter.

Here is the critical transition point:

‘There is nothing meritocratic about standing in the way of giving our most academically gifted children the specialist and tailored support that can enable them to fulfil their potential. In a true meritocracy, we should not be apologetic about stretching the most academically able to the very highest standards of excellence.

We already have selection to help achieve this in specialist disciplines like music and sport, giving exceptionally talented young people access to the facilities and training that can help them become world class. I think we should have more of this. But we should also take the same approach to support the most academically gifted too.’

Meritocracy is later associated with diversity and universal impact:

‘A modern, meritocratic education system needs to be much more flexible and agile to respond to the needs of every child.’

This is what May says about meeting individual needs:

‘And I don’t just want to see more new schools, but more good new schools that each in their way contribute to a diversity of provision that caters to the needs and abilities of each individual child, whoever they are and wherever they are from.

Every child should be given the opportunity to develop the crucial academic core. And thanks to our reforms that is increasingly the case. But people understand that every child is different too, with different talents, different interests, different dreams. To help them realise their potential and achieve those dreams we need a school system with the capacity and capability to respond to what they need.’

Her assumption is that support for ‘academically gifted’ learners can only be provided through selection; that diversity can only be provided through choice of school – that it is not feasible for a single institution to provide these things.

This amounts to a denial of the comprehensive principle.

.

Twin educational objectives

There are also twin educational objectives:

• Increasing system capacity to secure a good school place for every child (‘capacity’ appears 12 times and the phrase a ‘good school place’ 10 times). 

• Extending support to, and so promoting social mobility amongst, non-advantaged learners (in families that ‘just get by’) in addition to those from disadvantaged backgrounds.

The former is the top priority. May says the government ‘must deliver a radical increase in the capacity of the school system’.

The only form of tailored provision on the table is between-school selection – because more selective places will contribute to the global total of good school places.

The Conservative manifesto promised a ‘good primary school place available for every child’. May’s speech extends this to ‘a good school place for every child’:

‘Because for too many children, a good school remains out of reach. There are still 1.25 million attending primary and secondary schools in England which are rated by Ofsted as requiring improvement or inadequate. If schools across the north and Midlands had the same average standards as those in the south, nearly 200,000 more children would be attending good schools.’

May argues that government policy must push beyond a straightforward distinction between advantaged and disadvantaged.

This will involve extending support to those ‘just getting by’ on modest family incomes of £16-21K.

‘At the moment there is no way to differentiate between the school experience of children from these families and those from the wealthiest 10%.

Policy has been skewed by the focus only on those in receipt of free school meals, when the reality is that there are thousands of children from ordinary working class families who are being let down by the lack of available good school places.

Putting this right means finding a way to identify these children and measuring their attainment and progress within the school system. That work is underway and is central to my vision of a school system that truly works for everyone.’

But May also commits to retaining pupil premium throughout her time in office (there is already a commitment to retention over the lifetime of this government).

Both objectives carry with them significant resource implications at a time when budgets are tight.

• The capital and revenue costs to build and expand schools 

• The investment in school improvement needed to ensure no school is less than good 

• Targeting additional resources at non-advantaged learners, ideally without reducing per capita funding for disadvantaged and all other learners.

These are not discussed

.

Four routes to more good school places 

May sets out four ways in which the supply of school places will be increased

Faith schools will no longer face the requirement that when oversubscribed they must limit to 50% the number of pupils they select on the basis of faith. This is preventing the establishment of new Catholic schools.

The government will consult on new requirements to ensure faith schools are inclusive and their pupils mix with those of other faiths and backgrounds.

Independent schools will face consultation on the introduction of a tougher public benefit test to maintain their charitable status.

May suggests that smaller independent schools with limited capacity can confine themselves to different varieties of direct school-to-school support. But those with the capacity and capability will be asked to sponsor an existing state school, set up a new state school, or fund scholarships at their own institution.

Universities will be required through their access agreements to sponsor a state school (the press notice specifies an underperforming school) or else set up a new free school.

Singleton schools will not be sufficient:

‘And over time we will extend this to the sponsorship or establishment of more than one school, so that in the future we see our universities sponsoring thriving school chains in every town and city in the country.’

The press release expands on the speech:

‘The government intends to set out the new guidance to the independent Director for Fair Access (DfA), with a clear expectation that universities would contribute to school level attainment. This will inform the DfA’s own guidance to higher education institutions on their access agreements, which are conditions of charging the higher rate of fees. The government will consider what further measures are necessary to ensure all universities meet these requirements.’

More selection, to be achieved by:

• relaxing restrictions on the expansion of existing selective schools

• allowing new selective schools to open where parents want them and

• permitting non-selective schools to become selective ‘in the right circumstances and where there is demand’.

In return the government will ensure that all selective schools ‘contribute meaningfully to raising outcomes for all pupils in every part of the system’.

.

The green paper proposals

The green paper ‘Schools that work for everyone’ was published on 12 September, alongside an oral statement to Parliament by Secretary of State Justine Greening. There was also a parallel statement in the Lords and a brief press release.

Greening’s statement pays lip service to meritocracy and ‘building a school system that works for everyone’ before admitting:

‘The various proposals set out today in this consultation document all drive towards one simple goal: increasing the number of good school places for all children.’

She suggests these new proposals build on the commitments in the white paper, without explaining how.

She admits that:

‘This consultation deliberately asks big, open questions about the future of education in this country’

That is code for ‘we haven’t had time to work up detailed proposals or think through all the potential implications’. This green paper couldn’t really be greener if it tried.

The document itself includes an introduction – which makes the broad case for reform, discusses the need for more good places and a new focus on those ‘just getting by’ – followed by separate sections on faith schools, independent schools, universities and selection (though not in that order).

Each of these substantive sections sets out the case for change, an evidence base of sorts and proposals for reform.

Three full months are allowed for the consultation, which closes on 12 December 2016.

The government’s response will be published in spring 2017, but there is otherwise no detail about the anticipated timeline, beyond a reference to influencing universities’ 2018/19 access agreements.

Recent media coverage has suggested this is a movable feast that could be kicked into the long grass if it all proves too difficult.

. 

More good school places

The government’s core problem is that pupil number forecasts predict an undersupply of school places, while the existing stock includes 1.25 million places in schools that are requiring improvement or inadequate according to their most recent Ofsted inspection reports.

The Conservative manifesto promised to invest at least £7bn over this Parliament ‘to provide good school places’. A Conservative government would open at least 500 new free schools accounting for 270,000 new school places. Additionally, all good schools – selective or otherwise – would be allowed to expand.

DfE’s most recent national pupil projections – in SFR 25/2016 (July 2016) – anticipate a 10.3% increase by 2025 in the number of children aged up to 15 attending state-funded schools.

Total numbers are projected to be some 757,000 higher by 2025. The bulk of the expansion will take place in secondary schools, where numbers are projected to increase by 20.6%, implying 567,000 more students than in 2015.

The green paper gives instead the total increase in numbers by 2020. This is 458,000 (174,000 primary and 284,000 secondary).

In January 2016 there were 16,778 primary schools with an average roll of 275 pupils and 3,401 state-funded secondary schools with an average roll of 939 pupils.

Applying these averages to the shortfall implies a need for the equivalent of 630 more average-sized primary schools and 300 more average-sized secondary schools.

Since secondary numbers are projected to increase at the same rate for a further five years, from 2020-2025, this significantly under-estimates the full scale of the problem.

There is no calculation in the green paper showing the anticipated shortfall that these new proposals are expected to meet, either individually or collectively.

It would have been extremely helpful for respondents to have understood the likely orders of magnitude when considering the viability of the proposed reforms and their likely impact.

.

.

The green paper repeats May’s 1.25 million figure. It adds that, at March 2016, the percentage of good and outstanding schools was 86%, compared with 69% in 2010.

But it ignores the big disparity between primary and secondary schools. The most recent Ofsted data shows that 11% of primary schools and 22% of secondary schools are rated neither good nor outstanding.

Nor does it mention the significant regional disparities – with variance between 70% and 89% in the secondary sector – and that the worst-performing local authorities are at 20%.

But it does refer to:

‘65 local authority districts where fewer than 50% of secondary school applicants have a good or outstanding school place available to them within 5km. This means that in 20% of districts fewer than half of secondary school pupils have access to a good school within a reasonable distance of their home.’

The big problem with using Ofsted inspection outcomes as a benchmark of quality is that assessments go down as well as up.

In 2015 Ofsted revealed that:

• Of 3,474 schools judged as requiring improvement at 31 December 2014, 176 had been judged outstanding at their previous inspection and 1,342 had been judged good.

• Of 493 schools judged inadequate at 31 December 2014, 34 had been judged outstanding at their previous inspection and 179 had been judged good.

So, altogether, some 44% of schools had slipped back. The green paper is rather coy about the implications of such recidivism, which is certain to be a bone of contention for all potential sponsors.

Given the scale of this challenge, ministers will be anxious to retain the generous inspection holiday enjoyed by new schools, which ensures most are not inspected until their third year of operation.

There are three inter-related elements to the planned response:

• Radical expansion of the number of good places available

• Providing successful schools with the right incentives to expand

• Securing a ‘diverse school system’ that serves all learners.

The text argues that existing strategies, including MATs and free schools, have made progress in all three directions. But, by implication, current (white paper) policies are insufficient to address the challenge.

So the government is proposing to incentivise, and remove obstacles that prevent, an additional contribution from the four specified sources. The preferred approach is to make the benefits they enjoy conditional on their additional contribution, or else to extend new freedoms with conditions attached.

Local authorities would have been an obvious ‘fifth source’ but they are conspicuous by their absence.

Despite targeting higher attaining learners, no connection is established between these new proposals and the specific support for the most able promised in the white paper.

Although it countenances the creation of more selective 16-19 free schools, the green paper is otherwise almost entirely silent about the implications for the 16-19 sector.

.

Non-advantaged learners

There is only a passing reference to ‘families who are just about managing’. Strangely it gives the overall percentage of pupils eligible for and receiving FSM but not the far bigger percentages entitled to the deprivation element of the pupil premium (25.1% in primary and 29.2% in secondary in 2016-17).

May’s references to gross incomes of up to £21,000 per year would suggest that she is interested in support for the lowest 40% of families by household income. Some learners receiving pupil premium will have household income that exceeds this, by virtue of the ‘ever 6’ provision.

This will be expensive. We are not told whether the distinction will feature in the national funding formula.

The argument is that, because those ‘just getting by’ cannot be distinguished from others, including the wealthiest:

‘This distorts policy to focus at a cliff edge whereas the reality is that there are children from ordinary, working families with otherwise similar educational prospects not getting the support they need.’

The green paper invites advice about how best to identify this cohort, to measure their attainment and progress. No reference is made to any existing work undertaken to date.

There is an obvious risk that, by promoting new emphasis on non-advantaged learners, schools will be provided with a ready excuse to shift focus away from disadvantaged learners.

Selective schools in particular might use this as an escape clause, given the comparative scarcity of disadvantaged candidates. Such behaviour is not unknown in access to higher education, where progress with POLAR quintile two can help to mask very limited progress with quintile one.

Any fair access requirements (see below) will need to be worded carefully to avoid this scenario.

HE institutions can confidently expect pressure to introduce this distinction into their work on widening participation and fair access, although that sector is still getting to grips with pupil premium.

.

Faith schools

This category is largely irrelevant to my argument, so I will confine this commentary to the substance of the proposals.

The green paper says:

‘This consultation therefore proposes that we replace the ‘cap’ for faith free schools – including for existing schools – with a series of strengthened safeguards to promote inclusivity, thereby allowing free schools with up to 100% faith-based admissions.’

Yet the proposals appear only to specify detailed safeguards for ‘new faith free schools’, namely:

• Prove there is demand from parents of other faiths ‘through local consultation and signatures’.

• Establish twinning arrangements with other schools not of their faith.

• Consider establishing mixed-faith MATs, including sponsoring underperforming non-faith schools.

• Consider including someone of a different faith or no faith on the school’s governing body.

These would be enforced through guidance, the application process and monitoring visits. Non-compliance would result in change of status to a non-faith school.

Respondents are invited to suggest alternative measures, monitoring arrangements, ways of holding schools to account and potential sanctions.

It is important to consider the possibility of new selective faith free schools.

.

Independent schools

The proposal to introduce a tougher public benefit test is not unfamiliar. We have been round these houses before, indeed the issue has been festering away in the background over the last five years.

It is extremely unlikely that the powerful independent sector will comply fully with the government’s ambitious expectations: we can expect protracted negotiations and, if necessary, legal challenge.

The green paper puts the number of independent schools at 2,300, of which almost 50% have fewer than 150 pupils.

About 1,300 are registered as charities, so only that proportion – below 60% – would be caught by changes to public benefit rules.

Although many independent schools have partnership arrangements (including all but 45 of those with the Independent Schools Council) these are highly variable.

Eight schools sponsor 11 academies between them. Mention is made of three sponsored free schools, of which two are selective 16-19 institutions. The difficulties experience by the likes of Wellington and Dulwich College are diplomatically overlooked.

Different expectations are set out in relation to schools with and without ‘capacity and capability’, but there are no suggested criteria on which to rest this distinction.

According to the most recent ISC census, some 260 ISC schools have 600 or more pupils. That might be a reasonable guide as to capacity, if not capability.

Those with such capacity and capability should:

• Sponsor academies (in the plural) or set up a new free school, with the government meeting the capital and revenue costs. These would be expected to become at least good ‘within a certain number of years’, or

• Offer ‘a certain proportion of places as fully funded bursaries to those who are insufficiently wealthy to pay fees’. Presumably this includes non-advantaged as well as disadvantaged learners. The proportion should be ‘considerably higher than that offered currently at most independent schools’, but no other quantification is supplied. The text is somewhat ambiguous but it seems that these costs would not be met by the government.

The most recent ISC census reveals that only 5,629 pupils at its schools pay no fees at all, by virtue of a 100% bursary or scholarship. Since over 518,000 pupils attend ISC schools, that is slightly over 1%.

Would the government expect the proportion of fully-funded bursaries to be broadly similar to the percentage of disadvantaged and non-advantaged pupils admitted to grammar schools (see below)? Are we talking 10% or 20%, or more? No light is shed.

On the face of it independent schools are much more likely to choose the first option than the second. It seems improbable therefore that this will float the Sutton Trust’s long-cherished but misguided Open Access Scheme.

This is mentioned briefly in May’s speech. It would denude many state schools (including grammar schools) of their most able learners, but is presumably acceptable because it would reduce the capacity required in the state system.

That said, the government is not apparently interested in other methods of growing the independent sector, to take the strain off the state sector.

The Trust itself argues that its scheme should be preferred to ‘creating new untried free schools’.

The path of least resistance might well be further selective 16-19 free schools or, provided that the restrictions are withdrawn, new selective 11-19 free schools. Some of the former direct grant grammar schools may also be tempted back to the maintained sector.

Schools without capacity and capability will select one or more from a menu of options including:

• Direct school-to-school support of various kinds

• Supporting teaching in minority subjects

• Senior leaders serving as MAT directors

• Providing access to expertise and facilities

• Providing sixth form scholarships, supporting teaching or supporting university applications.

Benchmarks are proposed for schools according to their ‘size and capacity’ (capability has disappeared). Legislation will be considered to give these a statutory basis and to remove charitable status from those schools not meeting their benchmarks.

Respondents are asked to suggest other potential expectations of schools in each category and to comment on the planned benchmarks and legislation.

.

Universities

The green paper argues that criticism of universities for failing to widen access has been unfair hitherto, since they have had little involvement in or control over school-level attainment.

But they should be focused precisely on raising standards and attainment in the school sector because this is ‘where they can make the most difference’.

Unfortunately there is no justification for why this should take the form of increasing the supply of good school places.

Strangely, the evidence section gives only a single example – the KCL Maths School, itself a 16-19 institution. There is no information about the progression of its students – especially its disadvantaged and putative non-advantaged students – to selective universities.

HEFCE has published a list, based on data available at December 2015, of those HE institutions sponsoring schools.

This shows that 12 of the 20 English Russell Group universities are engaged in such activity – as are some 70 HEIs in total if UTCs are included in the calculation.

However the green paper says ‘this level of direct involvement is far from the rule’.

It proposes that, ‘as a condition of charging fees’ all HEIs should:

• Establish a new state school, with capital and revenue costs met by the government, or

• Sponsor an academy (it does not specify an under-performing academy).

Such schools would be expected to be good or outstanding within an unspecified number of years.

Universities would also be expected to extend such support to an unspecified number of other schools after an unspecified period, presumably by operating as MATs.

One commentator has calculated that, of the academies presently sponsored by HEIs, fewer than 60% for which reports are published are rated good or outstanding.

The history of university sponsorship is not unblemished. Bournemouth, Chester, Chichester, Christchurch Canterbury, Hull, Liverpool, Nottingham, Teeside, UCA and UCL have all been reported as experiencing difficulties at one time or another.

Once again it seems that selective schools might be the path of least resistance, or selective specialist institutions, though there is no explicit encouragement to complete the network of 16-19 maths schools touted by the previous Coalition government.

Universities will need to be wary of opposition from their staff, not least within their education faculties.

The government plans new guidance to the Director for Fair Access:

‘with a clear expectation that universities would contribute to school-level attainment as a condition of charging higher fees, and that we want them to do so by sponsoring academies or establishing new free schools’.

This would inform the DFA’s guidance on 2018/19 access agreements. It is implied, but not explicitly stated, that the DFA would reject access agreements which did not comply.

But there is also a threat that:

‘Beyond this guidance, we will consider what further measures, including potential legislation in a future Parliamentary session, are necessary to require sponsorship of a school as the specific means by which universities contribute to raising attainment and widening participation, where the DFA does not currently have the power to do so.’

This suggests that there will be no attempt to include provisions in the present Higher Education Bill.

There is no intention to permit universities to substitute an alternative means of raising attainment, but they may consider as additional activities:

• Membership of a governing body or trust board

• Supporting curriculum design, pupil mentoring or providing other educational support and

• Providing human resources, teaching capacity or financial support

Respondents are invited to suggest other ways in which universities might help to raise school-level attainment, but they are also asked to describe ‘the best way to ensure that all universities sponsor schools as a condition of higher fees’.

To date there has been conspicuous silence from OFFA, usually so assiduous in welcoming government initiatives.

All the indications are that the imperative to increase places trumps the need to raise attainment.

There is no argument from here that universities should be much more directly engaged in raising school attainment. I do not disagree with May that:

‘We need to go to the root of the problem, which is that there are not enough students from disadvantaged backgrounds and from ordinary families fulfilling their potential with the grades to get into the best universities.

So I want our universities to do more to help us to improve the quality of schools so that more students of all backgrounds have the grades, the subjects, and the confidence, to apply to top universities and to be successful in their exams in the first place.’

But I am far from convinced that this is most effectively achieved by sponsoring or establishing schools, whether individually or in chains.

Highly selective universities with national reach should be using alternative methods to target and directly support learners in all our schools, rather than providing a ‘talent pipeline’ for those attending a handful of institutions.

HEFCE, which is funding a National Collaborative Outreach Programme (NCOP) says:

‘With regard to the Green Paper consultation, the [HEFCE] Board noted the importance of wide-ranging collaborative partnerships across HE, further education and schools to support the progression of students alongside the measures to improve attainment set out in the Green Paper.’

The Vice-Chancellor of the University of Oxford has commented that sponsoring schools would be a distraction from its core mission.

I have suggested that one university at least could play its part by hosting a national research centre for high attainers and high attainment.

I have also shown how the demand and supply sides of the market could be managed, so securing a coherent package of tailored support to help disadvantaged high attainers progress to the most demanding HE courses and the most selective universities. Disadvantaged learners in schools and colleges need to be able to draw together opportunities offered by different universities and a variety of third sector providers to create such coherence.

.

Increasing selection

The document explains that there are presently 163 grammar schools educating about 5% of pupils in the state secondary sector, but they are confined to 36 local authorities of which 10 are deemed fully selective. Together these schools account for some 166,000 school places.

There are also some 45 bilateral and partially-selective schools as well as several selective 16-19 institutions.

The green paper tells us that:

‘In January 2016, 2.5% of pupils in selective schools were eligible for free school meals, compared to 13.2% for all state-funded schools.’

This suggests that the total FSM population in existing grammar schools is around 4,150. The annual intake will be some 600-800 FSM pupils.

We are not told what proportion of the grammar school population attracts the deprivation element of the pupil premium. But a recent PQ reply confirmed that, across England, 8.5% of grammar school pupils are presently in receipt of pupil premium (not just the deprivation element). There is substantial regional variation, from 12.5% in the East Midlands to 5.6% in the East of England.

In total 29.2% of pupils in secondary year groups are eligible for the deprivation element of the pupil premium in 2016-17. So pupil premium students are heavily under-represented in grammar schools, but not quite as heavily as FSM-eligible pupils.

There is no indication of the scale of change from these baselines that the government would find acceptable.

Are we aiming to increase the FSM percentage to 13.2% or thereabouts, which would multiply the annual intake fivefold?

Clearly there comes a point, even at national level, where there would be too many grammar schools.

According to the projections, the system will need some 3 million state-funded secondary places (11-15) by 2020. If the system as a whole is pegged at an overall national selection rate of 20%, that would imply 600,000 selective places, roughly a fourfold increase on the current position net of those in partially selective schools.

One imagines that would not be feasible and that a doubling to 300,000 places might be the best the government could hope for.

Given that some grammar schools are far more selective than others, it is open to question what selection rate the government would deem permissible for a local authority or similar area. Would they stay at 20% or could they go as high as 50%, or even beyond?

Current legislation prevents the establishment of new selective schools (though selective 16-19 institutions are permissible) or existing schools changing character to become selective.

Expansion is permissible under certain conditions but, according to May’s speech, ‘the funding necessary for expansion has not been consistently available’.

The removal of the present restrictions is justified on the grounds that:

• 99% of existing grammar schools are rated good or outstanding, 80% of them outstanding

• the attainment gap is allegedly ‘reduced to almost zero for children in selective schools’ (according to May’s speech) and

• the schools themselves want to expand.

The first seems a slight exaggeration. At the last count, 133 of 163 are rated outstanding and 26 are rated good (97.5%).

The second is also incorrect, as I demonstrated in this earlier post from December 2014. On the basis of the data then available I found that:

‘ While there is a relatively large group of consistently high performers, roughly one in five grammar schools is a cause for concern on at least one of the three [headline] measures. Approximately one in ten is performing no more than satisfactorily across all three…

A handful of grammar schools are recording significant negative gaps between the performance of disadvantaged students and their peers.’

We have no published information about what proportion of the 163 grammar schools want to expand.

The argument is that relaxing these restrictions will create more good school places and can be made conditional on selective schools simultaneously ‘supporting the creation of good new places in non-selective schools locally’.

‘We believe that these proposals will make grammar schools engines of academic and social achievement for all pupils, whatever their background, wherever they are from and whatever their ability.’

The evidence presented is selective (pardon my pun). Just one 2004 study is cited in support of the claim that grammar schools can be particularly beneficial for low income pupils while analysis of performance table data is eschewed.

It is acknowledged that some studies:

‘…suggest that there may be an association with poorer educational consequences for those pupils not attending selective schools in areas where selection is allowed’

A single study (by the Sutton Trust in 2008) is cited as counter-evidence. The Trust itself has challenged this interpretation of its research.

Alternative summaries of the evidence base can be gleaned from:

• This education datalab post, which references some of my earlier work

• These briefings by education datalab and the IFS.

Strangely there is no entry for selection in the EEF Toolkit.

The three reforms proposed are:

• Supporting the expansion of existing grammar schools, with an annual £50m budget to meet cost (presumably both capital and recurrent) on the basis of estimates rather than retrospectively. Satellite provision would be included. No reference is made to any limits on this mode of expansion. Could a grammar school potentially operate a handful of satellites across a region? Grammar school expansion will not come cheap. The expansion of one Kent grammar school was recently approved at a cost of £11m.

• Permitting new schools, both wholly and partially selective. The latter could select by ability or aptitude. (A footnote suggests that, according to the OSA, the former means ‘attainment’; the latter ‘potential to attain’. This is clearly nonsense since most grammar schools do not presently select on attainment.) This opens up the possibility of selective specialist 11-18 institutions with a focus on STEM, or languages, or arts, or music. More schools might become bilateral, with a distinct grammar department. There is also scope to introduce super-selective schools that would cream off from other run-of-the-mill grammar schools as well as from comprehensives.

• Permitting non-selective schools to change their character. It appears they could only become wholly selective, but (for some unknown reason) not partially selective. Unspecified measures will be considered to ‘preserve school diversity’ in affected areas.

New and expanding selective schools will need to meet specified conditions, drawing from a suggested ‘menu of options’:

• Taking an (unspecified) proportion of students from ‘lower income households’.

• Providing opportunities to join the school at 14 and 16 as well as at age 11. (The speech also mentions taking in pupils from a sponsored non-selective school for particular subjects or specialisms, or accepting them full-time if that is justified by their performance.)

• Establishing a new non-selective secondary school, the capital and revenue costs met by the government.

• Establishing a primary feeder school ‘in an area with higher density of lower income households’ the capital and revenue costs again paid by the government.

• Partnering with an existing non-selective school within a MAT or sponsoring an under-performing non-selective academy.

New schools or partner schools would again be expected to become good or outstanding within an (unspecified) number of years.

Schools would be held to account through requirements to publish details on their websites, monitoring by RSCs and ‘existing data collections’.

Sanctions – to be imposed where schools are not meeting expectations, or do not deliver additional good or outstanding places – might include: ‘removing any additional funding streams’ for new pupils or programmes; removing the right to select by ability, either temporarily or permanently; and preventing further expansion.

There is no threat of legislation – other than in relation to fair access (see below) – as there is with universities and independent schools. This is surely an oversight.

No reference is made to the possibility of transferring the school to different sponsors.

There will be some element of planning through the free schools application process. Geography and ‘the level of pre-existing selection’ are identified as important factors. There are hints that local demand and need for additional good places will feature in the criteria for approval. This is all extremely sketchy.

A system in which high parental demand is thwarted by a range of planning considerations is doomed to create unhelpful tensions. As is one in which parental opposition is over-ridden.

Managing this level of complexity will present substantial additional challenge for both the centre and the regional tier. Lawyers will have a field day.

.

Provisions applying to all existing selective schools

As for existing selective schools – presumably those that expand as well as those that choose not do so – the government intends to:

• Encourage MATs to create a real or virtual location for their most able learners, who might then benefit from a ‘more challenging and targeted curriculum’. This would not be deemed selection since the learners are identified as ‘most able’ after being admitted. (Presumably MATs will be free to determine how they define ‘most able’ for these purposes, since there is no agreed national definition.) It is not clear whether learners would attend these centres full-time or only for regular ‘pull-out’ sessions, but one assumes the latter. One also assumes that this might also operate at primary level. MATs would presumably need to topslice school budgets, including pupil premium where appropriate, to meet the cost. This is the only reference in the green paper to provision for the most able which need not depend on providing additional school places. It is not immediately clear how this would impact on the accountability arrangements for participating schools. There is no clear benefit beyond the confines of the MAT.

• Require outreach activity, including work with local primary schools to encourage disadvantaged pupils to apply. This will involve identifying the individuals most likely to benefit, which might indicate further support for the most able. A range of support may be provided, including financial assistance with transport and uniform costs.

• Secure fair admissions and access:

‘We therefore propose to require all selective schools to have in place strategies to ensure fair access. Legislation would require selective schools to prioritise the admission of, or set aside a number of specific places for, pupils of lower household income in their oversubscription criteria.’

It is unclear why this last would not extend to a lower entrance requirement for disadvantaged applicants, although the Grammar School Heads’ Association has recently noted (p15):

‘The DfE team acknowledge that the legal position on this is complex and not entirely clear and is something they are still looking at in respect of the Code and possible changes. That said, where schools have introduced this, the OSA has not upheld objections to it and the view of the DfE is that schools wishing to pursue this route should continue to do so.’

Contextualised admissions are now firmly established in the HE sector, backed up by evidence that students admitted with lower entry grades are fully capable of achieving similar or better outcomes.

The interim report from Scotland’s Commission on Widening Access says:

‘There is an increasingly compelling evidence base which shows that pupils who achieve modestly lower grades in more challenging circumstances consistently operate to the same, or an even better, academic standard than their more advantaged, higher attaining peers. This suggests that the applicant pool may be unnecessarily, and unfairly, narrowed by an over reliance on pure attainment, measured in terms of grades, as the primary measure of academic ability.

For example, a study at the University of Bristol showed that over the course of a 3-year degree students from state schools with lower attainment caught up and then academically surpassed their higher attaining peers from private schools…

…These findings are backed by more practical evidence. The Scottish universities who have used contextual indicators to lower entry tariffs for students with lower attainment from disadvantaged backgrounds report little or no evidence of a drop off in academic standards…’

Both the speech and the green paper mention the impact of tutoring, the speech in these terms:

‘While there is no such thing as a tutor-proof test, many selective schools are already employing much smarter tests that assess the true potential of every child. So new grammars will be able to select in a fair and meritocratic way, not on the ability of parents to pay.’

As education datalab has shown, there are serious questions about the degree of tutor proofing achieved by the best-known of these tests.

It is unclear why there are no proposals to guarantee in-school preparation for all potential applicants from disadvantaged backgrounds.

There is huge variance in the selection instruments used by different grammar schools, which plays into the hands of the tutoring industry and indeed the ‘sharp-elbowed middle classes’.

There is no obvious reason why schools should not adopt a standard national selection process and common tests.

Another option might be to move to a position where all selective schools select on prior attainment, supplemented if they wish by near-universal CATs scores. This might be combined with intensive government support – backed up by grammar schools themselves – to close KS2 excellence gaps.

There is a final note to the effect that the National Funding Formula will reward schools with a ‘higher proportion of lower attaining pupils and those from less wealthy households’. There is no recognition that provision for higher attaining learners might also impose additional costs.

Respondents are asked a series of questions including:

‘What is the right proportion of children from lower income households for new selective schools to admit?’

And:

‘How can we best ensure that new and expanding selective schools and existing non-selective schools becoming selective are located in the areas that need good school places the most?’

And finally:

‘Should the conditions we intend to apply to new or expanding selective schools also apply to existing selective schools?’

It is not clear why all grammar schools should not be expected to satisfy those conditions that stop short of establishing and/or sponsoring schools.

 .

Summary and conclusions

.

Problems with the government’s proposals

This analysis suggests that:

• The government’s proposals are over-complex and over-ambitious. They are trying to kill at least five birds – radically increasing the supply of school places; effectively eliminating the possibility of schools being rated less than ‘good’; increasing the supply of quality MAT sponsors; promoting meritocracy and social mobility by challenging vested interests; and introducing support for non-advantaged learners – with a single stone. They plan to do all this alongside the existing white paper agenda. They are biting off much more than they can chew. But this also means that solutions for each of these elements must fit with the package as a whole, so policy choices are unnecessarily constrained. It would be far preferable to develop parallel stand-alone strategies and to ensure that they complement each other as far as possible.

• The cost is almost certainly excessive. There has been no published estimate of the total cost of this reform package, which must include: the capital and recurrent costs of new schools and the expansion of existing schools; intensive school improvement effort to push 100% of schools to ‘good’ with zero backsliding; and presumably additional targeted support for non-advantaged learners, ideally not at the expense of the current per capita support for disadvantaged and all other learners. This at a time when the schools budget is already under serious pressure, a national funding formula is imminent and the full spectrum of white paper reforms is waiting in the wings. All we know is that £50m per year is to be allocated to the expansion of existing grammar schools, but that will not go very far. We need some handle on the estimated budget, broken down as far as possible, to judge whether this reform package is viable and to take a view on whether it offers value for money. If there is an opportunity cost we need to know what it is. 

• There is no timeline and little quantification. We do not know when each outcome is supposed to be achieved and there are no interim objectives. The government needs to be clearer about what it can achieve during its lifetime and what will be left for the next government to take forward. There is no substantive quantification of the desired effect for any of the elements in the package, so we cannot determine the true scale of the government’s ambition. There is a significant risk that all the fine words will not be matched by commensurate action; that much of this is empty political posturing; that the real impact will be distinctly underwhelming. 

• The proposals are a denial of the comprehensive principle because they refuse to accept the possibility that every learner’s individual needs can be met within a single institution, relying where necessary on varieties of within-school selection. It is unnecessarily prescriptive to insist that learners’ needs must be met within a single type of institution, be it a selective school for the most able or a comprehensive school for all learners. On the other hand, the system does not need a bewildering diversity of school types and specialisms with ubiquitous between-school selection. There is a trade-off between choice and diversity on one hand and system-wide coherence on the other. 

• Increasing the stock of school places should be addressed immediately and separately as a system-wide planning issue. The scale of expansion required within a decade is sufficiently challenging in itself. It is much more likely to be achieved if the waters are not muddied by more contentious issues. What matters is the number of additional places needed, where they are needed and how they will be supplied. The badge worn by the supplier or attached to bundles of places is a second order issue. Introducing selection into the mix is so contentious that it will almost certainly be an unhelpful brake on progress. 

• An implicit target of 100% of schools rated good or better is impossible to achieve certainly within the lifetime of this government, regardless of the progress made towards full academisation. There are obvious priorities: secondary over primary and schools within a subset of underperforming local authorities and chains. The white paper outlined a policy for addressing these priorities that we are still waiting to see implemented. So certain is it that a proportion of schools will always slip back from their good or outstanding rating that policies must be adjusted to account for this. 

• Support for the non-disadvantaged is also risky. The cliff-edge argument against FSM is strong, but the government is failing to acknowledge that it has been partly addressed already through ‘ever 6’ pupil premium. Because the overall budget seems unlikely to increase, this may well involve reducing per capita support for disadvantaged learners or, if they are protected, for all other learners. The national funding formula principles will need adjustment, presumably without a further round of consultation. The formula itself could be radically different. This also gives selective schools an opportunity to game the system by concentrating on admission of the non-disadvantaged at the expense of the disadvantaged. 

• Fully-funded places in independent schools are a non-starter and short-sighted policy that does nothing to strengthen the state sector. There is apparently little interest in other methods of growing the independent sector, but there is scope for further thinking. For example, the government might explore the possibility of public-private partnership to introduce a tranche of low-cost non-selective independent schools, with fees pegged at the cost of a state school place and free places for all learners attracting the pupil premium. 

• The independent sector will resist public benefit reform beyond a certain point. It is far from clear that the government would win this battle, but failure, or even a weak compromise, would immediately destroy the government’s credibility as an enemy of vested interests. It might be better to focus exclusively on the obstacles that are preventing more independent schools from acting as state school sponsors and find ways of removing them. If the government is determined to legislate it might consider alternatives, such as regulating and/or taxing independent school fees, with favourable offsets for state school sponsors.

• School sponsorship is not the only or even the best way for universities to help raise school-level attainment. The contribution to widening participation and fair access is limited and there is substantial deadweight (because many students are advantaged and/or incomers from the independent sector, and/or would have succeeded anyway). All the evidence points to the value long-term collaborative partnerships that directly address the differing needs of disadvantaged learners. Compulsory state school sponsorship may be perceived as an infringement of institutional autonomy. There is also an opportunity cost. As with independent schools, the government might do better to investigate the obstacles preventing the remaining universities from acting as school sponsors and find more constructive ways round them.

• Neither universities nor independent schools have an unblemished record as state school sponsors. It is highly likely that a proportion will always be less than good, some having slid back to that position, but there is no clear statement of the likely consequences. Denial of charitable status and capacity to increase university fees are nuclear options, the latter cutting directly across the newly-invented TEF. Both independent schools and universities will demand safeguards which will undermine the government’s position. Given this situation there is no obvious reason why local authorities should be prevented from playing a bigger role. 

• None of the three justifications for increased selection is completely secure, especially the assertion that gaps between advantaged and disadvantaged high attainers are negligible in grammar schools. We do not know what proportion of existing grammar schools want to expand further, nor do we have information about the match between their locations and those where additional places are most urgently required. There is no guarantee that expanded selective schools will maintain their quality across satellite and partner institutions. The evidence base cited in the green paper is thin and weak, which reinforces the case for an independent university-based research centre.

• The devil will be in the detail of the planning process. Managing this process will be incredibly complex and difficult. The market cannot be left unregulated. It will be impossible to reconcile completely the pattern of parental demand with the incidence of need, or the willingness of existing schools to expand. Additional layers of complexity will be introduced by decisions over what constitutes an acceptable selection rate and whether super-selective schools can introduce a second tier of selection. Secondary school admissions will be in a constant state of flux, a situation that will further benefit the middle classes. It seems doubtful that Regional School Commissioners would cope, at least not without significant additional resource. This further increases the cost to the taxpayer. 

• Reform of existing grammar school admissions is the path of least resistance. Much more clarity is required about where expectations are likely to be pitched, and the consequences of non-compliance. Selective schools can learn from the experience of contextualised admissions to universities. More radical thinking is necessary. There is scope to introduce a single standard 11+ assessment with a compulsory lower pass mark for disadvantaged applicants. Given the slippery nature of ability and increasing disagreement over its very existence, let alone its assessment, we should consider the benefits of all selective schools shifting to attainment-based admissions, perhaps utilising the new KS2 higher standard. This would have to be backed up by intensive government support to close primary excellence gaps.

.

Comparison with the core principles 

How do the government’s proposals, especially its plans for increased selection, stand up against my seven suggested core principles?

There is no inherent reason why they cannot observe the first four principles and simultaneously pursue increased selection.

• All learners have an equal right to an education that meets their needs, regardless of background and prior attainment. No learner should be denied access to suitable tailored learning opportunities.

• Pursuing the twin priorities of raising standards and closing advantaged-disadvantaged attainment gaps implies giving relatively greater priority and relatively more resources to learners from disadvantaged (and non-advantaged) backgrounds, regardless of prior attainment.

• There is no separate stand-alone justification for prioritising low attainers over high attainers, or vice versa. A disadvantaged (or non-advantaged) high attainer should attract more priority and more resources than an advantaged low attainer.

• Gaps should not be closed by holding back improvement amongst learners from advantaged backgrounds. Both advantaged and disadvantaged/non-advantaged should continue to improve, the latter at a faster rate.

Purist defenders of the comprehensive principle might want to argue that the first is fundamentally inconsistent with selection, but that is to deny the capacity of parallel comprehensive schools (or indeed secondary moderns) to operate effectively alongside grammar schools.

In my experience it is the most ardent exponents of the comprehensive principle who have greatest difficulty with the second and third principles above.

Even the former pupil premium champion is not immune:

‘Sir John Dunford said the proposal that selective schools could expand on the condition they take a proportion of pupils from lower-income households would push a “disproportionate” amount of pupil premium cash to the most able poorer students – while damaging the prospects of disadvantaged children “whose needs are greatest”.’

.

.

Diverting more pupil premium funding into selective schools should help to address the excellence gaps that exist in some 20% of them.

The problems begin to arise with the last three principles, where the difficulties presented by the government’s ‘five birds with a single stone’ approach begin to emerge:

• Resources are scarce; impact can be maximised through efficient use of resources. That means minimising deadweight costs, investing in standards not structures and, wherever feasible, targeting support at the learner rather than the school.

If support for disadvantaged and non-advantaged high attainers is judged as a stand-alone policy, then these proposals are inefficient. There is deadweight in investing in more selective places because the substantial majority will be taken up by learners from advantaged backgrounds

The government can counter that it needs to provide the additional places anyway, to address the under-supply. The question again arises whether a commitment to increased selective places will not compromise the wider objective.

• Education tailored to background and prior attainment is not always most efficiently and effectively provided through universally inclusive learning opportunities. Some opportunities will be designed for the subsets of learners most likely to benefit. This requires selection in some form.

• Selection operates at many different levels, including between-school, within-school and within-class selection. These may operate independently or simultaneously. Key considerations when establishing the optimal combination include need, capacity, efficiency and flexibility.

The penultimate principle is a reminder that, even in the most comprehensive of settings, much within-school selection takes place. That is not in itself a justification for more between-school selection but, as the final principle states, there are several other options.

If the government is really serious about supporting disadvantaged and non-advantaged high attainers it should adopt a more holistic approach, reaching into all schools, whether selective or comprehensive.

If it relies exclusively on additional between-school selection, the government will be offering a one-dimensional solution to a much more complex and far-reaching issue.

.

What else to do?

A more nuanced approach would include:

• Admissions and outreach reform in existing grammar schools taking in some of the more radical approaches outlined above. All grammar schools should work with all their feeder primary schools to prepare all disadvantaged and non-advantaged learners who wish to apply for entry. Existing grammars should be able to expand provided that they comply fully with these new obligations. Further generalist or specialist 16-19 selective schools could be introduced within the free schools programme, provided there is demand from sponsors.

• Developing, trialling and evaluating the concept of a MAT centre for the most able. This should be a real or virtual pull-out centre for learners who remain on the rolls of their own schools, rather than a separate location which serves as an alternative full-time school. Through the EEF the government should fund the evaluation of several different models to establish the most effective. Such centres could play a role in facilitating increased pupil transfer into grammar schools after the initial selection process.

• Government acknowledgement of excellence gaps and a national commitment to closing them. This should feature prominently in any eventual life chances strategy. The government should report annually on progress in closing excellence gaps, revising and enhancing its strategy if progress is too slow.

• Core principles and guidance for all schools – primary as well as secondary – on effective institution-wide provision for disadvantaged and non-advantaged high attainers. (The white paper promised something similar as part of the ITT core content framework but has so far failed to deliver.) This should include more substantive guidance about effective use of the pupil premium and how this issue should be addressed in pupil premium reviews.

• Fulfilling the white paper commitment to ‘investigate, fund and evaluate approaches to help the brightest students in state schools to fulfil their potential’  with priority given to those from disadvantaged and non-advantaged backgrounds.

• Sharper accountability arrangements. Ofsted needs to revisit its approach to the most able which relies on old-style national curriculum levels, is secondary-specific and overly broad (taking in almost half of the cohort). The secondary performance tables should include a ‘higher standard in Attainment 8’ measure requiring an overall point score of, say, 70 (adjusted as appropriate during the awkward transition period to universal grades 9-1) with separate reporting for disadvantaged learners. This would complement the higher standard now included in the primary tables. The secondary disadvantaged attainment gap index should be based on Attainment 8 performance rather than only English and maths. The associated ‘bar chart and odds ratio’ methodology should be applied more widely, with analysis at regional, local authority and even school level. All quality thresholds – whether for MATs, Achieving Excellence Areas, pupil premium reviews, teaching school alliances, coasting schools etc. – should have an excellence gaps component.

• Further EEF-funded development work to establish effective within-school selection and grouping strategies, learning from international experience and pushing beyond the current study of setting to explore more innovative and hybrid approaches. The EEF might also publish a high attainer variant of its toolkit since the existing version tends to mask differential impact (see for example the entry on ‘setting or streaming’).

• Far stronger commitment to securing fair access to the most selective HE courses and institutions. There is much to commend the recommendations of the Scottish Commission on Widening Access, since accepted in full by the Scottish government. As a last resort, the Director for Fair Access should be able to impose institutional access targets on institutions that consistently fail to meet their own. A national system of access thresholds might provide the basis for imposed targets.

• Targeted support for all disadvantaged high attainers from 11-19 to increase progression to the most selective HE courses and institutions. This would be funded through an annual £50m pupil premium topslice and matched by contributions from universities’ fair access budgets. (For further detail see the presentation embedded here.)

• An independent national research centre located in a top university. (For further detail see the final section of my previous post.)

If it pursued that ten point plan, I venture to suggest that the government would not need to introduce new grammar schools or extend selection to schools that are not presently selective. The additional cost would be minimal.

It could avoid stirring up a political hornet’s nest and alienating much of the education sector. That would make it much easier, rather than more difficult, to generate more good new school places and new MAT sponsors.

 

.

TD

September 2016