Which grammar schools want more pupil premium students?


I wanted to find out how many of our 163 grammar schools give priority to disadvantaged pupils in their admissions arrangements for academic year 2016/17, and by what means.

This analysis was prompted by a comment made by Secretary of State Nicky Morgan during the Commons debate on her statement of 19 October 2015. The statement explained her decision to approve the expansion of the Weald of Kent Grammar School to a new site in Sevenoaks.

She said:

‘The admissions code, which was changed by this Government, specifically allows grammar schools to give priority to children who are eligible for the pupil premium in their admission arrangements. Half of the grammar school sector has introduced, or intends to consult on adopting, that admissions priority, and I would like more of them to go further.’ (Hansard 19 October 2015, Column 683)

How many schools have actually included such a provision in their AY2016/17 admission arrangements (as opposed to merely thinking about it) and what progress has been made since the issue was first raised by the previous Coalition Government?


The School Admissions Code

School admissions in 2016/17 and subsequent years must comply with the Schools Admissions Code (December 2014).

It says that:

  • Admission arrangements must include oversubscription criteria – the criteria against which places will be allocated when there are more applications than places – and the order in which they will be applied. (Para 1.6)
  • Only grammar schools can select their entire intake on the basis of ‘high academic ability’. (Para 1.18)
  • But where selective admission arrangements are not based exclusively on the highest scores in a selection test, the oversubscription criteria must give priority to all looked after children and previously looked after children who meet ‘the pre-set standards of the ability test’. (Para 1.20)
  • Oversubscription criteria may give priority to children eligible for the early years pupil premium, the pupil premium and the service premium. (Para 1.39A)

The latter provision was introduced following a 2014 consultation exercise. The Government response explains that academies could already apply it by virtue of their funding agreements: the amendment extends the same freedom to LA-maintained institutions.

It adds:

‘We are clear that admission authorities can use this priority flexibly and do not have to give admission priority to every pupil premium recipient. For example:

  • they could choose to adopt the pupil premium priority for a proportion of their places, to mirror local pupil premium levels
  • they can use it to give priority with reference to other oversubscription criteria, such as limiting it to pupil premium children living within the school’s catchment area.’

Further guidance was issued (also in December 2014). This explains that:

‘Ministers want to see high-performing schools admit a proportion of disadvantaged children that reflects the demographics of their local area.’

It offers additional suggestions for ways in which a provision might be shaped:

‘ Schools can:

  • choose from which group or groups (Early Years (EYPP), Pupil Premium or Service Premium recipients) to give priority…
  • …specify a number or percentage of their published admission number. For example, this can be representative of the number of disadvantaged children resident in the school’s local area; or they can prioritise a certain percentage of local eligible children;
  • limit priority to specific eligible sub-groups. For example, restrict the admissions priority to children currently in receipt of Free School Meals; or children in a catchment area;
  • decide the ranking given to the priority (after looked after and previously looked after children);
  • choose to give higher priority to EYPP/PP/SP eligible children of that faith than those not of that faith, if they have a faith designation.’


How many schools already have such provision?

In June 2014, David Laws, Coalition Minister of State for Schools, confirmed that:

‘32 grammar schools have implemented an admissions priority for pupils eligible for free school meals this year’

In November 2014, a reply to a Parliamentary Question also said 32 schools had already prioritised disadvantaged learners in their admissions, while a further 65 intended to consult on doing so.

There is a reference in the May 2014 edition of the Newsletter of the Grammar School Heads’ Association (GSHA):

‘A significant number of schools 38 have either adopted an FSM priority or consulted about doing so in the last admissions round. A further 59 are considering doing so in the next admissions round.’

In a spring 2015 newsletter, the Association says only that:

‘Grammar Schools have shown a more robust response to the issue than other sectors and many that have not taken action intend to do so for next year.’

So it seems that:

  • 32 schools had included some provision in their AY2015/16 admissions policies
  • 6 schools had consulted on introducing some provision for AY2015/16 and had not gone ahead (but might be expected to consult again in AY2016/17)
  • 59 more schools intended to consult on such a provision in the AY2016/17 admissions round or subsequently.

How many grammar schools make provision in their AY2016/17 admissions policies?

My survey shows that 55 grammar schools (34% of the total) will make some sort of provision for disadvantaged applicants within their admissions policies for AY2016/17.

A full list is included at Annex A below

Looked at from one perspective, this is a significant improvement on the 20% of grammar schools doing so in their AY2015/16 admissions arrangements. Altogether, 23 more schools have taken this step in AY2016/17.

On the other hand there has been relatively little movement amongst the schools intending to consult on such a change. Of the 65 schools assumed to have been contemplating introducing some provision, only 35% have actually gone ahead. This despite explicit support from the previous Government, as well as encouragement from the Sutton Trust.

According to both the GSHA and the PQ reply, a total of 97 schools (60% of all grammar schools) had already acted or were considering doing so. But that leaves a substantial rump of at least 66 schools that have not been contemplating any such reform and show no sign of changing their minds.

As matters stand then, two-thirds of grammar schools (108 in all) have either retained the status quo (66 schools), or have failed to put a proposal out for consultation despite saying they would do so or, having done so, have been unable to persuade their consultees to support change (42 schools).

The geographical distribution of reforming schools is markedly uneven, with some local authority areas leading the way and others dragging their heels.

  • Whereas 12 of 13 schools in Buckinghamshire, six of eight in Birmingham and all five Warwickshire grammar schools have introduced a provision of this nature…
  • …None of the 15 Lincolnshire grammar schools or the six grammar schools in each of Medway and the Wirral has done so. All but three of Kent’s 32 grammar schools have made no provision.

The table below shows the full distribution of the 55 with provision of some kind in their AY2016/17 admissions policies.

Almost 42% of this group is accounted for by the schools in the three authorities named in the first bullet point above. Twenty-one local authority areas feature, meaning that 15 of the local authorities with grammar schools (42%) are not represented.


Buckinghamshire 12
Birmingham 6
Warwickshire 5
Gloucestershire 3
Kent 3
Sutton 3
Torbay 3
Bournemouth 2
Poole 2
Reading 2
Slough 2
Trafford 2
Wiltshire 2
Plymouth 1
Devon 1
Enfield 1
Essex 1
Lancashire 1
Medway 1
North Yorkshire 1
Telford and Wrekin 1


How are disadvantaged learners given priority?

A quantitative analysis takes us only so far. It is important to distinguish the different ways in which disadvantaged applicants are given priority, and to assess their comparative impact.

What proportion of schools have included in their AY2016/17 admissions policies a provision that will make a significant difference, as opposed to a largely meaningless gesture?


Pupil premium or FSM?

It is surprising to discover that a significant minority of the sample retain provisions relating to those who are eligible for and claiming free school meals (FSM) at a specified date, rather than to those in receipt of the pupil premium, having been eligible for and claiming FSM at any point in the previous six years (‘ever 6’ FSM).

Only 33 of the 55 (60%) opt for a pupil premium measure while 22 (40%) continue to specify FSM.

I could find only one School that made use of both simultaneously: Sir Henry Floyd School has FSM as its second oversubscription criterion and pupil premium as its third.

Interestingly, in a recent Determination on the admission arrangements of The Royal Grammar School High Wycombe (November 2015), the Office of the Schools Adjudicator considers whether a FSM criterion conflicts with paragraph 9f of the Admissions Code ‘which prohibits  giving priority to children based on parents’ financial status’ since ‘eligibility for free school meals is based on parental income’.

The Determination concludes that, as a consequence of the December 2014 guidance, such a provision is permitted, but does not further discuss the apparent conflict between these two parts of the Code.



A provision based on the pupil premium is likely to be significantly more generous. The reply to a recent Parliamentary Question gives the 2015/16 national eligibility rate for the deprivation element of the pupil premium in the primary sector as 22.2%.

By comparison, the January 2015 national FSM rate for state-funded nursery and primary schools – according to SFR16/2015: Schools, Pupils and their Characteristics, January 2015 – is 16.5%.

We do not have access to local FSM and pupil premium rates amongst the populations served by grammar schools, but these national figures suggest that many would better serve disadvantaged learners by switching from FSM to a pupil premium measure.


This chart summarises the outcomes of the analysis below.


GS admissions chart revised



Oversubscription criteria

By far the most common approach, adopted by 36 of the sample (65%), is to cite either eligibility for FSM (17 schools or 31%) or for pupil premium (19 schools or 34%) as an oversubscription criterion, to be applied when deciding which learners who have achieved the 11+ cut-off score should be offered a place.

  • In 24 of the 55 (44%) either FSM or pupil premium is the second oversubscription criterion. Priority for looked after children is almost invariably the first.
  • In seven more either FSM or pupil premium is the third oversubscription criterion; in three cases it is the fourth and in three more the fifth.

When priority for disadvantaged learners appears lower down the pecking order, it may be preceded by a catchment-based criterion, provision for siblings and for those with exceptional medical or social needs.

Occasionally priority for disadvantaged learners within the catchment area may feature as one criterion, with priority for out-of-catchment disadvantaged learners appearing lower down.

Generally speaking, the lower the place in the oversubscription criteria, the less significant the effect is likely to be on the intake. In the six cases where disadvantage features at only the fourth or fifth position, the impact is likely to be negligible.

However, other factors can influence the impact of an oversubscription criterion.


Oversubscription criteria with limited places

Several policies place limits on the proportion of disadvantaged students that may be admitted. For example:

  • Bournemouth School and Bournemouth School for Girls each specify pupil premium eligibility within catchment as their third oversubscription criterion following looked after children and those who scored highest in the tests. Bournemouth School sets aside 120 places, against a PAN of 150, for the highest scorers, while Bournemouth Girls sets aside 133 places against a PAN of 166. Hence pupil premium places cannot exceed 20% at either school.
  • The Latymer School cites pupil premium eligibility as its third oversubscription criterion, but restricts this to a maximum 20 applicants from a defined catchment (the PAN is 186). The proportion of pupil premium places cannot exceed 11%.
  • Upton Court Grammar School identifies pupil premium eligibility as its third oversubscription criterion after looked after children and those with Education and Health Care plans. But a maximum of 15 places is available, against a PAN of 145. The proportion of pupil premium places cannot exceed 10%.
  • Wallington County Grammar School (PAN 135) has FSM eligibility as the third oversubscription criterion following looked after children and highest scorers. Only 10 FSM places are available. The proportion of pupil premium places is 9%.
  • Wallington High School for Girls (PAN 210) has eligibility for pupil premium as fourth oversubscription criterion following looked after children, special medical or social reasons and 100 high scorers. But only 15 places are available for those within a defined catchment area. This means the proportion of pupil premium places is 7%.
  • Nonsuch High School for Girls has pupil premium or service premium eligibility as its second oversubscription criterion after looked after children, but restricted to 10 places (PAN 210) for those who live within the catchment area. The proportion of pupil premium places will be less than 5%.
  • Urmston Grammar School (PAN 150) has pupil premium eligibility as its third oversubscription criterion, after looked after children and the top 20 test scores, but only three places are available, plus ties for the final place. Unless there are ties, the proportion of pupil premium places is a mere 2%.

By restricting the number of places available, schools can manipulate the impact of an oversubscription criterion, regardless of its position in the hierarchy. In most cases the effect will be to hold the admissions rate for disadvantaged learners well below the incidence of disadvantage in the area.


Lower qualifying scores

A handful of schools specify lower cut-off scores for disadvantaged candidates in their selection tests:

  • Four of the five Warwickshire grammar schools (King Edward VI, Lawrence Sheriff, Rugby and Stratford) include pupil premium eligibility as their second oversubscription criterion, restricting this to a maximum of 10 candidates. Given their PANs, this limits the proportion of places to 12% (King Edward VI) or 8% (Lawrence Sheriff, Rugby and Stratford). However, these places are reserved for children whose scores in the selection test are between one and ten marks lower than the cut-off. In the case of Lawrence Sheriff eligibility for pupil premium is also used as a tie-break for other categories of admission.
  • Churston Ferrers (PAN 145) defines a ‘borderzone’ of candidates who have not achieved a CEM test score ranking them in the top 100 and have not achieved a minimum 50% in maths and English, but have succeeded in two of these three elements and achieved 40% or higher in the third or have a combined score equal to or exceeding that of the 140th ranked candidate. SEN, looked after and FSM-eligible children, as well as children of staff within the ‘borderzone’ will be given priority, in that order.


Reserved places

A further small group of schools, which may or may not have common cut-off scores, reserve a specific number of places for disadvantaged candidates:

  • The five schools in Birmingham’s King Edward VI Foundation – Aston, Camp Hill Boys, Camp Hill Girls, Five Ways and Handsworth – each set a lower qualifying score for candidates eligible for the pupil premium. This qualifies them for a specified number of places set aside for them, equivalent to 20% of each school’s PAN (25% at Aston).
  • Borden Grammar School (PAN 120) reserves up to 15% of places (so 18) for students eligible for the pupil premium. If more than this number applies, the places are allocated according to oversubscription criteria which favour looked after children, then siblings, then those with special health and social issues, then those in specific catchment areas.



A few schools confine themselves to using disadvantage as a tie-breaker:

  • Devonport High School describes pupil or service premium as its second oversubscription criterion, after looked after children, but these only apply to applicants on the waiting list.
  • The only Catholic grammar school appearing amongst the 55 is Loreto, which uses pupil premium eligibility as a tie-breaker in each of seven religious admissions criteria.

It is hard to see this form of provision as anything other than a marginal gesture.



Of the 55 schools that give some sort of priority to the admission of those from disadvantaged backgrounds, no more than 45 (28% of all grammar schools) have provisions likely to have a significant impact.

Even among these 45, there is huge variation in impact. All schools place some form of cap on the proportion of disadvantaged students they will admit, ranging from as low as 2% of PAN to as high as 25%.

Although we do not know the rates of disadvantage for the populations served by each school, the national figures suggest that almost all schools are holding their admission rates below the local incidence of disadvantage.

In his June 2014 speech, former Minister David Laws set an aspirational target:

‘My ambition is that all selective schools should aim for the same proportion of children on free school meals in their schools as in their local area.’

His officials had calculated (in an unpublished analysis) that:

‘This would mean an additional 3,500 free school meal pupils in selective schools every year, or an additional 35,000 pupils over 10 years.’

Laws added:

‘We in the Department for Education will fully support any school that chooses to change its admissions criteria in this way – in fact, I want to see all grammar schools give preference to pupil premium pupils over the next few years.’

We are still a long way from achieving this position.

Morgan’s statement that she would ‘like more [grammar schools] to go further’ is positive but guarded. It does not amount to a ringing endorsement of Coalition policy.

If the Conservative Government was really serious about speeding up progress it might use the upcoming consultation and revision of the Admissions Code as an opportunity to tighten up the existing rules.

For example, it might require schools to:

  • replace all remaining FSM-related criteria with pupil premium equivalents
  • introduce some form of provision in the next admissions round where they have not already done so
  • give automatic priority to pupil premium admissions alongside looked after children
  • admit disadvantaged students in line with the local incidence of pupil premium eligibility.

The Government might also ensure that approval of any new selective schools (exclusively 16-19 academies and free schools), proposals for annexes on the Weald of Kent model and increases in selective schools’ PANs should all be conditional on the agreement of strict targets, designed to achieve representative pupil premium admission rates within a maximum of three years.

For this analysis suggests that the impact of a laissez faire approach will be too slow, too patchy and too limited.



November 2015


Note: This is a slightly amended version of the text first published on 4 November 2015


Annex A : The 55 grammar schools making some provision in their admissions policies

Adams’ Grammar School

Alcester Grammar School

Aylesbury Grammar School

Aylesbury High School

Beaconsfield High School

Bishop Wordsworth’s Grammar School

Borden Grammar School

Bournemouth School

Bournemouth School for Girls

Burnham Grammar School

Chesham High School

Churston Ferrers Grammar School

Colchester County High School for Girls

Colyton Grammar School

Cranbrook School

Devonport High School for Boys

Dr Challoner’s Grammar School

Dr Challoner’s High School

John Hampden Grammar School

Kendrick School

King Edward VI Aston School

King Edward VI Camp Hill School for Boys

King Edward VI Camp Hill School for Girls

King Edward VI Five Ways School

King Edward VI Handsworth School

King Edward VI School

Lancaster Girls’ Grammar School

Langley Grammar School

Lawrence Sheriff School

Loreto Grammar School

Marling School

Nonsuch High School for Girls

Parkstone Grammar School

Poole Grammar School

Queen Elizabeth’s Grammar School

Rainham Mark Grammar School

Reading School

Royal Latin School

Rugby High School

Sir Henry Floyd Grammar School

Sir William Borlase’s Grammar School

Skipton Girls’ High School

Slough Grammar School (Upton Court)

South Wilts Grammar School for Girls

Stratford-Upon-Avon Grammar School for Girls

Stroud High School

Sutton Coldfield Grammar School for Girls

The Crypt School

The Latymer School

The Royal Grammar School, High Wycombe

Torquay Boys’ Grammar School

Torquay Girls Grammar School

Urmston Grammar School

Wallington County Grammar School

Wallington High School for Girls

6 thoughts on “Which grammar schools want more pupil premium students?

  1. A tad misleading – for example if a PP student achieved the required standard at Reading they would get in no matter how many achieved a higher standard. In reality they are classed exactly the same as a looked after or previous looked after child. Issue is about drive and achievement at primary; motivation and engagement of parents and lack of support for primary work


    1. Not really. I identify Reading as one of the 55 schools that gave some priority to disadvantaged applicants in their 2015/16 admission arrangements.
      The purpose of my post was to establish what proportion of schools made such provision – and to compare the ways in which they did so.

      I may in due course revisit with reference to 2016/17 admission arrangements. I can see from this the arrangements at Reading which are broadly (but not exactly) as you describe.


  2. I’d like to ask Reading School how many FSM children have actually benefitted from this policy but they tend to refuse my requests for information. If you read it, their admissions criteria explains they make up the pass mark on the fly. Admittedly an FSM child whose score would have previously placed them on the waiting list will now get a place but given the pass mark is set *after* the tests with the express purpose of minimising this waiting list it’s hard to see this as anything other than a token gesture.


    1. I can’t see why that information should not be published. The Reading School admission arrangements seem to me good in the sense that there is no artificial limit placed on the number of FSM learners that can be admitted (provided they achieve the cut-off).

      I’m less impressed that they give themselves carte blanche over where and how the cut-off should be set.

      I’d also question why they can’t extend the criterion to include all learners eligible for the pupil premium, rather than only those eligible for FSM at the time they take the test.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s