House of Commons - Education and Skills

Select Committee on Education and Skills Memoranda

Memorandum by Martin Johnson

Research Fellow, IPPR

February 2003

Institute for Public Policy Research

30-32 Southampton Street London WC2E 7RA

tel 0207 470 6100 fax 0207 470 6111

m.johnson@ippr.org www.ippr.org

Registered Charity 800065

Not Choice But Champion

A New Look at Secondary Admissions in London

by Martin Johnson, IPPR

Introduction

This paper discusses the process by which pupils at age eleven are admitted to secondary schools in London. It forms a major strand of the IPPR project, Schooling in London, because it is clear that the process requires radical improvements which other agencies are not addressing. Although primary admissions is also an issue in a few places, such as Camden, the problems of secondary admissions are more general, more longstanding, and ultimately more important for the performance of London schools and London pupils.

Whatever the principles of the present system, its operation is far less problematic across the country. In small town and rural England, parental preference has generally been an empty entitlement:

'The main factor influencing parental choice appears to be the happiness of the child; for most parents, that would seem to be related to social well-being as much as academic success and this derives from factors such as proximity to the secondary school, the choice of school the child's friends make and the quality of relationships within the school. Educational choice, therefore, is perhaps far less an issue than the architects of the 1980 Act might have envisaged or wished. In a rural area where each market or seaside town of any size has its own secondary school, there is little pressing need for schools to market themselves aggressively in competition with other schools, especially when there are plenty of children to go round. Travelling to the next town to a school similar to the one in your own seems inconvenient and pointless.' (Birch 2001)

Difficulties arise, generally speaking, in a small number of circumstances.

Areas affected by the presence of grammar schools. Whatever the arguments about selection at 11+, in a few places where a single super-selective school takes a tiny percentage of pupils from a wide area the overall transfer scheme is unlikely to be seriously adversely affected. However, in other areas where grammar schools select a significant proportion of pupils, all of the well-known disadvantages of selective education are played out in the transfer process.

Areas where the bulk of schools are their own admissions authorities. Where most schools are foundation schools, the variety of procedures, timetables, and selection criteria makes the process very complex.

There is considerable overlap between these circumstances within many of the England's conurbations. London's unique position makes the problems of transfer particularly pressing, and a necessary focus for the IPPR.

London's Unique Position

London has 405 secondary schools, over 10% of the whole of England. Many of them are within walking distance of each other. They are of almost every kind known in this country: there are boys schools, girls schools, mixed schools, 11-16, 12-16, 11-18, community, foundation, and voluntary - pick from CofE, Catholic, Jewish, Islamic. There are city technology colleges, specialist schools, beacon schools, city academies; a nautical school, a performing arts college, a school with a music specialism. The apparent choice for parents is incomparably larger than anywhere else. The reality of choice is different, since popular schools choose their pupils, or in some cases their parents. In each locality, there is a hierarchy, which has been reinforced by the quasi-market reforms in place for a decade, and given further impetus by the more recent introduction of yet more labels for the purpose of differentiation (Johnson 1999). Recently, this hierarchy, called a ladder, seemed to become government policy.

Those unsatisfied by that degree of choice have the largest private sector in the country. There are no authoritative statistics to show the proportion of London's pupils who attend private secondary schools, but the DfES calculates it to be 10.1% (13.3% in inner London), against a national average of 6.8%. Some cite this as signifying a crisis in London education but the situation is hardly surprising. Those patronising private schools in England are very largely from the higher social classes, who are disproportionately resident in the London area. The motivations of those who choose private education are discussed below, but their decisions may be based on social and cultural dispositions rather than an objective evaluation of the qualities of local maintained schools. This hardly forms the basis for a negative assessment of London's schools.

The variety of schools is more than matched by the range and complexity of London's socio-economic structure. The London area is one of the wealthiest places in the world. It has a concentration of millionaires and million pound homes. One of its boroughs, Tower Hamlets, is also in a league of its own - for child poverty, with a rate of 73.5%, 12% higher than the next borough (Hackney) and 15% higher than the poorest non-London authorities (Liverpool and Knowsley)(Bradshaw 2002 Table 1.8). On the index of additional educational need calculated for primary school funding purposes, nine of the top ten and fourteen of the top twenty authorities are London Boroughs. A relevant factor is the very close juxtaposition of locations of wealth and poverty, so that not only postal districts but electoral wards and even postcodes can be very varied in their social composition.

England London Inner Outer

% of secondary school pupils known to be eligible for FSM 16.9 27.6 44.2 19.7

% of households with a gross domestic income over £750 per week 16 24

% of secondary school pupils with EAL 8.0 29.2 42.3 23.3

Table 1: Social Characteristics of Londoners

Sources: Statistics of Education: Schools in England 2001, table 14; Office for National Statistics: Region in Figures, Winter 2001; DfES

There are over 52,500 homeless households, with 8,000 of them in bed and breakfast accommodation. More than one in four London children live in households where no-one works, compared with 18% nationally. Unemployment in inner London is 7.2%, compared with 5.2% nationally. Londoners comprise 25% of England's problem drug users.

In addition to extremes of wealth and poverty, London has an unequalled ethnic diversity. This contains long-established groups from all over the world, current refugees from poverty and persecution, and misfits from Britain and the world. In many of its schools over a hundred languages are spoken. Inner London's population is particularly mobile. In England as a whole, 5.6% of admissions are non-standard (pupils do not start in a school at the 'normal' time), but in inner London no fewer than 14.2% are non-standard, and in parts of the city where transience is concentrated the proportion is much higher. A forthcoming report by Save the Children (Welborn forthcoming) investigates the numbers of children in London who have no school place. Although responses from boroughs were incomplete, a number reported between 25 and 55 pupils each in this position, with the situation changing daily. The main reasons were mobility, refugee and asylum seekers, parental choice, and imbalance in places. Although the majority of boroughs feel pressure on places, there are often places available in schools which parents are unwilling to take up.

The pressure of places is partly due to rising secondary school rolls, in line with a national trend. Nationally, numbers are due to peak in 2004. The London Plan envisages much more substantial growth in the capital, sufficient to require an additional 130 schools, over the next fifteen years. The question of how London's 33 local education authorities will cope remains to be answered. Some are amongst the smallest in the country. They have a variety of policies with regard to admissions.

Does the combination of a variety of schools and a heterogeneous school population produce a satisfying mix or will it in the future? There is no evidence that the proportion going to private sector education is growing so perhaps the system is working for some. However, around 70% of pupils are placed in the secondary school of first preference compared with 85% nationally (Williams et al. 2001). While the national average percentage of appeals on admissions in 2000/01 was 10.3%, the inner London average was 18.8% and the outer London 21.0% (DfES 2002). It may well be that satisfaction is even lower than these figures indicate. Not only do they obscure cases in which the desired school was not a stated preference because of the small likelihood of success, they do not measure unhappiness about the whole process. This is found drawn out and stressful by the knowledgeable, and confusing and alienating by those who have not learnt how to play the system (Templeton and Hood 2002).

Though the London Plan projection is open to debate on its likelihood as well as its desirability, it seems certain that London will need some additional secondary schools, and planning is needed not only with regard to their location, but also to their admissions policies. This paper argues that admissions policies cannot be considered at school level, but must be reviewed more collectively. The opening of new schools alone argues for a fresh look at London admissions.

Are London Schools Good Schools?

Since most parents profess to want to choose a good school for their children, this question might be thought to be central to the admissions problem. It is often claimed that London has too many bad schools, and this explains their unpopularity, thus enhancing the hierarchy referred to earlier.

Much of the discourse on good and bad schools is based on loose thinking and misleading data. Whatever the GCSE results for a school tell us, what they certainly cannot do is offer any evidence as to how any individual child might perform in that school. School effectiveness research over a number of years shows that around 80% of attainment differences are due to individual, home and background factors, rather than the performance of the school. Or, to put it another way, by far the best predictor of a school's performance is the nature of its intake. Indeed, a technique commonly used by schools to improve their results is to attract a 'better' intake (Lupton 2002a). Yet, we hear frequently that a grammar school is 'good' because it achieves a high percentage of A-C GCSEs, despite the fact that its intake was selected on the basis of being able to achieve just that.

Many conclude that value added data will solve this problem, and was published for the first time for 2002, showing progress between KS2 and KS3 and between KS3 and GCSE/GNVQ. However, there are criticisms for the approach being used as anything more than a very broad guide. The use of small populations of pupils within schools means that, whilst outlying schools can be identified, the more subtle differences between most schools cannot be fairly judged (Goldstein 2001). The DfES itself points out:

'at KS3 to GCSE/GNVQ when comparing schools with cohorts of about 50 pupils, differences of up to 4.0 should not be regarded as statistically significant, while for schools with about 100 pupils, differences up to 2.8 should not be regarded as significant.' (DfES 2001)

Value added scores are constructed so that the mean improvement is 100. The problem is that most schools fall within the range 97-103, and therefore cannot be differentiated. Even as a broad guide, this data may be considered crude in not taking full account of the individual pupil's prior achievement or a school's ability to deal with the individual needs and abilities pupils. As yet there has been no solution to the methodological problems of pupil mobility, which constitutes a major factor for schools in some parts of London. These factors are crucial and without taking them into account the statistics do not provide the kind of evidence needed to allow sophisticated parents to choose the best school for their individual child. (Goldstein & Sammons 1997; Thomas & Mortimore 1996) Finally, there is potential for manipulation of results. Could the value-added measures prove a demotivator in KS3 for more competitive schools, determined to show improvement?

Policymakers are being forced back to the formerly well understood fact that the bulk of the difference in achievement between schools in England is due to the pupils and their backgrounds. They must also accept that the remaining element of difference which is due to the school is very complex and cannot be captured by the simple 'value-added' statistics now being generated. Recent attention to simple achievement statistics has reinforced a popular discourse of 'good school' - 'bad school' which fails to understand either the very substantial differences in the size of the task between schools with different intakes, or the mix of strengths and weaknesses, some persistent, some transitory, found in all schools.

Unfortunately, this misleading discourse has been reinforced by the Ofsted inspection process, which despite detailed evaluations of strengths and weaknesses ends with a simple conclusion. These simple conclusions, piled together, create a misunderstanding, for it is now part of the discourse that 'bad schools' are found disproportionately in areas of social deprivation, including inner London, which indeed has 13.1% 'bad schools' compared with 4.1% in the county authorities. However, recent work (Lupton 2002b) suggests that the largest variations in Ofsted measurement between schools in deprived and non-deprived areas are 'school-climate' variables, which seem to many to measure the social context of the school much more than its performance.

In London, official support for this popular discourse has exacerbated the trend towards market distortion. Parental choice is exercised with high levels of ignorance about the product range. Diversity in a market only works when consumers make diverse choices, but the strong tendency for 'consumers' of London secondary schools to choose the same schools creates excess demand at one end of the graph and inadequate demand at the other, without any mechanism tending towards equilibrium. The experience of London secondary schools places in question the whole concept of the quasi-market as a means for whole system improvement.

Principle: The Government must lead a drive to alter the tone of political and popular discourse about secondary schools in London.

London's Schools in Context

	England	London	Inner	Outer
% of secondary school pupils known to be eligible for FSM	16.9	27.6	44.2	19.7
Vacancy rate in secondary schools	1.3	2.9	3.2	2.6
% of teachers without QTS in all schools	1.3	3.3
% of secondary school teachers under 30 years old	16	19
% per annum teacher turnover rate (all schools)	11	15
% of lessons rated as excellent or very good	15	19	20
% of lessons rated as poor or unsatisfactory	7	11	12
Schools' match of teaching and support staff to curriculum needs: % rated poor or unsatisfactory	16		42
Average secondary class size	21.9	22.1	22.5	21.9
% of 15 year olds achieving GCSE/GNVQ 5+ A*-C	48.6	46.7	38.2	50.5
% of 15 year olds achieving GCSE/GNVQ 5+ A*-C in schools within 21-35% FSM band	34.1	41.7	47.4	39.7
% of 15 year olds achieving GCSE/GNVQ 5+ A*-C in schools with 21% and over FSM (bands 5-7)	31	35.3	34.7	35.8
% of 17 year olds in education (f/t or p/t)	66	69	65	71
% passing advanced vocational qualifications	79.4	79.3	79.3	79.3

Table 2: Some Characteristics of London Schools

Source: DfES

Table 2 shows that achievement at Year 11 in London is lower in raw terms than the national average. There is a dramatic distinction between inner and outer London, with the latter doing better than the country as a whole. As argued above, this tells us much about the social composition of the respective areas' schools, but little about their quality. More useful are the figures relating to free school meals eligibility. Looking at schools in England with more than 20% of pupils eligible (national average 16.9%), 31% of pupils gain good GCSE results. In London 35.3% of that group achieve that level, and in inner London 34.7%. The evidence relating to schools with moderately high levels of eligibility is even more impressive. Where eligibility is between 21% and 35%, 34.1% of pupils attain the standard, but in London the figure is 41.7%, and in inner London no less than 47.4%. Since over a quarter of London pupils, and almost a half of inner London pupils, are eligible for free school meals, it could be argued that these figures are an indication of good service to Londoners, and certainly better than average.

At the same time, there is a number of unpopular schools with very low achievement. As should be inferred from above, their intakes are overwhelmingly disadvantaged. In a city where teacher recruitment and retention is a major inhibition to success, these schools find it impossible to employ sufficient staff.

Does School Composition Matter?

These figures also begin to cast light on the proposition that the social mix of the intake itself is an independent variable affecting overall performance. It can be inferred that London schools with moderately high levels of FSM eligibility are much more successful at producing high-achieving pupils than schools with levels over 35%. The major international study by the Programme for International Student Assessment (PISA) has produced further evidence (OECD 2001).

It is worth repeating the headline PISA results. They showed UK 15 year olds as amongst the highest achievers within OECD nations in standard literacy, maths and science tests. They also showed that the UK has one of the strongest relationships between pupils' socio-economic status and test scores. Of the twelve countries scoring above average for literacy, six also manage low levels of inequality in scores, while the UK is one of three with above average inequality

However, the study also found that the average socio-economic make-up of a school's intake has a stronger relationship with performance than students' socio-economic status itself. In other words, the pupil mix is a vital factor.

'In almost all countries, and for all students, there appears to be a clear advantage in attending a school whose students are, on average, from more advantaged family backgrounds.' (OECD 2001 p.198)

A number of school-level factors which affect pupil performance are found to be strongly interrelated to the average socio-economic status of pupils. For example, with regard to learning environment, in the UK in particular schools with a high concentration of pupils from lower socio-economic groups have significantly worse disciplinary climates (OECD, 2001:Tbl 8.5 & 8.5a). A 1995 review of research on effective schools identified 'home-school partnership - parental involvement with their children's learning', 'a learning environment - an orderly atmosphere' and 'high expectations all round' amongst the top ten key ingredients for effective schools (Sammons et al. 1995). PISA found a strong link between pupil performance and the use, as opposed to the availability of educational resources and equipment.

As stressed in the ippr publication Parents Exist OK!?, schools as a whole benefit from parental involvement, but there are practical time issues for underprivileged or single parents as well issues of confidence in school involvement from parents with weak basic skills or lack of qualifications (Hallgarten 2000). Expectations of parents and teachers must clearly play a role but, more tellingly in this context, there is evidence to suggest that peer group influence is a major factor in achievement.

Schools with a high density of poorer children face greater obstacles in achieving all of these things.

What are the implications for a rational admissions policy? Clearly, individual pupils benefit from attending a school with a socially advantaged intake, and equally clearly, not all pupils can have that access. Adherents of a selective system (Schagen & Schagen 2002, Shuttleworth and Daly 2000) suggest that the 'grammar school effect' can lead to improving the achievement of the 'borderline' pupil, and by implication the greatest good to the greatest number. Schagen & Schagen's work is unconvincing, being based on performance in higher level KS3 tests which few comprehensive schools enter. However, London aspiring parents would seem to be behaving rationally by having their children huddle together. The question is, how many of them need to huddle together in order to achieve the required effect? What degree of social segregation in schools is desirable, from the point of view of maximising overall achievement?

UK schools are socially segregated, but at a level below the OECD average (Smith and Gorard 2002). London schools are also socially segregated, but the degree varies greatly, due to a number of factors including the presence of grammar schools in some areas, including counties adjacent to outer London, and the admissions policies of individual LEAs (Taylor and Gorard 2003). A recent report by the Education Network quantified segregation more simply: 'Of 405 schools in 32 London boroughs (the City of London has no secondary school), 21 (5%) have a proportion of pupils eligible for free school meals that is double that of their LEA, or higher; in 67 (16.5%) the proportion is half that of their LEA, or less. Of the latter group, 21 schools are selective; 18 are church schools which interview; 11 are church schools which don't interview. This leaves 17, where a mixture of circumstances play a part - including location in an advantaged area.' (TEN 2003)

Principle: Achieving a socially mixed intake should be a policy ambition.

The grammar school effect should not be underrated, because many of them select from a very wide catchment area across a number of LEAs. There has been no broad consensus from research emanating form the last thirty years on the question of selective or comprehensive education (Crook et al 1999). However, Jesson looked at GCSE level results in value-added terms and concluded that that pupils of average ability do better in the non-selective system and in selective areas GCSE performance in the majority of schools (which are non-selective) is actually depressed. That is, selective systems result in lower educational performance (Jesson 2001).

The Secretary of State recently asked the 36 education authorities which still have grammar schools to review their admissions procedures. He told the Select Committee on Education, 'selection regimes inhibited educational opportunities' for thousands of young people '...I would hope and believe that the authorities will look at their own practice from the point of view of education standards.'

Recommendation: No more selection by ability or aptitude

The argument against selection remains as strong as when Labour in opposition supported its abolition. The PISA study provides contemporary international evidence that selective school systems are comparatively less successful. In London grammar schools particularly distort the admissions system, increasing the inner to outer movement, creating stress and uncertainty for year 6 pupils and families, attacking the esteem of those who fail to be selected. Some London grammar schools are particularly separatist and elitist institutions which will be unable to comply with the new collaboration agenda.

It is important to stress that this is not a proposal for the closure of schools; it is for a change in their character. Many London grammar schools have a long history, but they are histories of change to adjust to changing circumstances.

At the same time as this selection is ended, it would be appropriate to end the right of specialist schools to select part of their intake. This has been an unpopular provision, taken up by very few, but those few have disproportionate effects in their own catchment areas.

The PISA study points out that it is difficult to deduce policy developments from its data, because of the specific cultural and organisational features of education systems in each country. It is for policymakers to be aware of the importance of school composition effects, and to optimise them for everyone. For London, it is imperative to move away from a system in which they work primarily to the advantage of the advantaged. Policy should have the following aims:

maximising the number of middle-class students who stay within state secondary schools
ensuring that all secondary schools have a critical mass of advantaged pupils
discovering more about the necessary dimensions of a critical mass, and the mechanisms by which it has effects on performance

The Government's Approach

Legislation since the 1988 Education Reform Act, together with interpretations by the courts, has created a very complex system. The complexity arises largely from the devolution to a large number of individual schools of the right to establish their own admissions procedures. The 1944 settlement allowed this to voluntary aided (largely, but not only, Church of England and Catholic) schools, but its extension to grant-maintained (now foundation) schools created substantial administrative difficulty. London has 219 secondary admissions authorities. This particularly affects some outer London boroughs, such as Barnet, Hillingdon, and Bromley, which has 17 secondary schools and 17 admissions authorities since only one of the 17 has remained a community school.

The other main factor within current law is the overwhelming priority of parental preference. This is a central requirement of the quasi-market approach to school standards adopted by governments since 1988. One result in London is multiple applications by parents, increasing the bureaucratic load. Barnet secondary schools had 3,559 available places for 3,400 pupils in the year six cohort in 2002, and 10,186 applications for them.

The 1998 Education Framework and Standards Act recognised the need for regulation of school admissions, and introduced a Code of Practice. It established an office of Adjudicator for cases of disputes and complaints. The 2002 Education Act increased the duties on admissions authorities in respect of co-ordination, and a revised and toughened Code of Practice came into effect in January 2003.

Three of the Government's aims are:

to maximise parental satisfaction with the system, including satisfying parental preference. For parents the procedure will be much more simple; they will complete a single form, providing for at least three preferences. Local authorities must produce a scheme for co-ordination of all schools within their areas which should culminate on 1 March each year when every year 6 parent will receive a letter with a single place offer.

to apply pressure towards compliance with equal opportunities principles. A number of practices, particularly certain oversubscription criteria and interviewing prospective parents, have been found by adjudicators to be likely to be in contravention of the Race Relations Act - although their decisions have not always been consistent.

The revised Code specifically bans interviews from 2005, currently permitted for the purpose of ascertaining religious affiliation where that is an oversubscription criterion. The Churches are content with such a ban, although some schools claim that they cannot trust priests to submit honest certification.

The avoidance of marginalisation of pupils with vulnerability or challenging behaviour. For example, it recommends giving priority to looked-after children in over-subscription criteria. In the context, this is a nod towards the general problem of unbalanced intakes which is a problem in city schools, and in London particularly. However, the provision is weak in that the admissions forum is given responsibility to promote agreement between all schools but its power is limited to a requirement on all admissions authorities to have regard to its decisions.

Other Aims for an Admissions Policy

The statement in the Code that all schools should contribute to accommodating vulnerable and difficult children opens the door to a much larger discussion. Since the evidence described above is that an improvement in overall performance would result from more socially balanced intakes in our cities, it must be appropriate for this to be one aim of admissions procedures.

Education is a public service, and the public as represented by the state at local or national levels has an interest in admissions, insofar as the system affects the overall nature of schools. The Code is one expression of that interest, but the local state should also have a role.

Many will assume that the Autumn tradition (or, in some cases trauma) of inspection of schools and their documents by year six pupils and their parents results in a preference freely negotiated between the two. The assumption may be misplaced. Earlier research suggests that in perhaps half the cases, children make the decision (Walford 1994), with the tendency stronger within the working class (West 1991), and a further 30% of decisions are indeed joint.

It is becoming clear that London children sometimes want very different outcomes from their parents (Templeton and Hood 2002). Templeton discovered that in primary schools whose pupils dispersed to a large number of secondaries, 'the process of changing schools has become unacceptably damaging to them' (p26).

Few of these children found that factors key to all pupils had been satisfied: moving on with friends, to a familiar school, a local school, a good school. There is a case for an independent 'child-centred' factor in secondary school admissions. It should reflect the child's wishes, but also a judgement on the child's needs, in which the primary school should play a major role. It cannot be right that the child who is at the centre of the process should formally have no part to play and no rights.

The current admissions system does not reflect a balance of the interests listed above. It is based on a simple market concept in which the parent is treated as the consumer. Indeed the 1998 Act reinforced this by giving admissions authorities an overriding duty to comply with parents' declared preferences. The interests of social justice are not served by such a limited perspective. State education is not an individual consumption good, but a public service provided to meet a variety of aims, only some of which relate to individual learners. Others are social and political, relating to social order, social cohesion, the transmission of democratic and liberal values, and so on. A school admissions system should reflect that variety of aims.

It must be recognised that there are tensions between and even within the aims listed above. Within a pattern of very popular and very unpopular schools, where the popular select their pupils, choice means little to many parents in London. A very substantial proportion of admissions is non-standard, that is at times other than the beginning of Year Seven, when there may be no local school with vacancies, or the available school is undersubscribed and already suffers disproportionately from pupil turnover. Parental satisfaction with transfer arrangements is significantly lower in London than in the country generally (Williams et al 2001).

Principle: The Government must accept the principle of compulsion on schools in improving the fairness of the system.

The conclusion must be that whether the aim is to produce a procedure which pleases parents, or one which balances a number of interests, another look is needed at admissions in London. It could be argued that recent legislative changes, and the accompanying revision of the Code of Practice, is that other look.

Does the Revised Code Resolve London's Problems?

Co-ordination within an LEA resolves only a part of the problem. An estimated 60,000 London pupils cross LEA areas, so simplification of the process requires co-ordination between LEAs. A proposal to require this was abandoned following strong representations that it was a very large administrative task requiring substantial additional resources, and more time to establish than proposed. An informal pilot in a number of south London authorities is revealing software sharing problems, for example. Some respondents doubted the theoretical practicability because neighbouring authorities had oversubscription criteria which were incompatible from a co-ordination point of view.

Recommendation: London local government must agree a single admissions system for the city.

It is clear that the complexity of co-ordinating borough schemes would be immense. However, the argument that the Council Tax payers of Borough X are entitled to their own unique oversubscription criteria is weak when compared with the parental stress and system stress of current arrangements. The 33 borough and corporation admission authorities must agree a common scheme. The mechanism by which they can do this is beyond this paper but not beyond the combined skills of London local government.

Even this improvement would not address the basic weakness in any system built exclusively on parental preference: preferred schools are oversubscribed, and then select, while undersubscribed schools by definition are not preferred but may be the only ones available.

The 2002 Act and the Code has provision for ensuring good practice by admissions authorities. Apart from the interview ban, however, it relies on malpractice such as dubious oversubscription criteria described in the Code being challenged by other admissions authorities. LEAs are given a strong push to do this, but it is hardly conducive to progress within the co-ordination without compulsion model promoted by the Code. The Borough of Bromley has recently shown what is possible by complaining successfully about five of its 'own' schools, but at what price to local relationships?

The weakness is personified in the role of the admissions forum. The Code expects all parties to be decent chaps, play the game and share the load. It ignores the effect of a decade of quasi-market measures and the lauding of entrepreneurial schools, which has created a cadre of very independent, very high achieving (because in practice highly selective) London secondary schools which are formally or informally outside the local community of schools. A prime example is the City Technology Colleges, which are to be invited to be part of co-ordinated arrangements, but whose refusal is final. Neither are the grammar schools to be coerced into sharing the load.

In key areas where current inequality of distribution creates difficulty - pupils with very challenging behaviour, asylum seekers, other non-standard admissions - the Code expresses pious hopes that the forum will develop not just majority but unanimous agreement on co-ordination. In London, it is important that all secondary schools offer places to a proportion of applicants who are likely to be mobile. In combination with measures on waiting lists, this would ensure in the most popular schools some casual vacancies become available to high-need pupils. The Code is silent on this.

By 2005, all maintained schools within an LEA will be required to comply with the single co-ordinated scheme, and will no longer be permitted to interview applicant parents or pupils as part of a selection process. The only remaining privilege for voluntary aided and foundation schools will be the right to fix their own admissions criteria, within tightened requirements. It would be a small step in practice to hand over responsibility for admissions to all local schools to the LEA. This would be a significant change in principle for faith schools, and it would be necessary to build in safeguards about their intakes. It seems clear that school admissions is an issue for the whole community, one which should be subject to democratic accountability.

Recommendation: In 2005 the LEA should become the admissions authority for all maintained schools in its area, including Voluntary Aided, foundation, CTCs, and academies.

A benefit of the new Code will be a reduction of form-filling by parents. Set against that is the gamble to be taken by grammar school minded parents, who must decide whether to make it one of their preferences before knowing whether their child has passed the entrance exam. Organisations campaigning for the full use of the provisions for complaint against malpractice can see some improvement in leverage. As suggested above, the requirement for clear and objective criteria enables bold admissions authorities to secure the end of dubious practices by others. Hence the combination of an end to selection and the rigorous implementation of the Code could produce a major reduction in the segregation index in some of the more segregated boroughs. Some argue that it would be sufficient to resolve the bulk of the problems of admissions to London schools.

However, the Code will not address all the issues.. It gives the child no place in the process, gives neither community nor state strong rights to express their interests, does not deal effectively with non-standard admissions, and does not in itself attack exaggerated pecking orders. The Code was not designed to address these problems; it deals with administrative difficulties produced by the status quo. Only new thinking about parental preference will enable these problems to be faced at last.

Parental 'Choice' - Solution or Problem?

Open admissions, with parental choice as it is popularly and misleadingly known, is a central element of the market model which has been employed since 1988. Although successive governments have claimed that this model is responsible for a remarkable increase in pupil performance, it would be extremely difficult or impossible to produce any evidence in support, not least because of the large number of other systemic changes taking place simultaneously.

One factor in the adherence to this idea was the rise of the school improvement movement, in reaction to earlier over-determinist views of pupil performance. Academic findings on school improvement were measured - the quality of the school does make a difference - but were sometimes misinterpreted to suggest that schools make all the difference. This misapprehension leads to the market distortions discussed earlier, and to the rhetoric of 'good school' 'bad school' which continues to dominate political discourse. This ignores the findings that almost all schools are a mixture of strengths and weaknesses. As argued earlier, the kinds of published data on which many London parents rely are unreliable, particularly as a predictor of their own child's' future success. To repeat, at least 80% of the variation in pupil achievement is due to factors outside the school, so parents should be made aware that the choice of school for their children is far less important than they might imagine.

Apart from academic achievement, there are other possible motivations for choosing popular schools. There may be a greater range of extra-curricular activities available, because of an existing sufficiency of motivated pupils to create demand. Increasing the availability of extra-curricular experience is indeed necessary across London schools as a whole, and we return to this below.

However, there may be another much stronger motivation. It could be expressed benignly as the 'birds of a feather' habit. The less charitable would express it as the 'fear of the estate yobs' attitude. There is a widespread feeling, though not supported by research evidence, that many inner London parents avoid their local schools in case their children pick up the wrong accents, the wrong manners, or the wrong attitudes. This may be understandable. It is an ignoble sentiment which cannot be a basis for social policy. It also underestimates the capacity of young people in a multi-cultural environment to build a large repertoire of behaviour and to select appropriately according to context.

Parental choice seems to be so embedded that any reduction in its primacy seems politically impossible. For a government with the will, however, there is an alternative case. To recap:

nationally, choice is not a substantial issue;
in London, a complex market operates, but the large numbers of very popular and very unpopular schools leads to selection by the schools and parental dissatisfaction;
the school attended is much less important as a determinant of pupil performance than many parents apparently assume;
basing admissions solely on parental preference ignores other important stakeholders, including the community and the state, and not least the child;
an exaggerated pecking order produced by unfettered parental choice produces intakes so unbalanced as to reduce overall pupil performance and threaten the achievement of national targets;
there are ways of ensuring all pupils have access to a wider range of opportunities;
snobbery is no basis for an education policy.

The argument is not that parents should have no say in which school their child attends. It is that parental preference should be one of a number of factors to be taken into account. The challenge is to devise a system for London which does that and improves overall parental satisfaction with their children's educational opportunities. The challenge can be met if government is prepared to take the argument to Londoners.

Alternative Procedures

There are a number of suggested alternatives to the present arrangements.

lottery All year six pupils would be allocated to secondary schools by lot. The theoretical outcome would be that all schools within an area had intakes with identical characteristics. It would be possible to make real comparisons about the performance of schools, and increase the likelihood of socially balanced intakes.

There are some drawbacks to this solution. What could be the unit area of a lottery? If it were a borough, it could lead to many difficult journeys to school. If the unit were smaller, say a group of wards, school populations might not reflect the overall population of the borough. More basically, it ignores the rights of parents and children to express a preference. In particular, it assumes a single model of school, and would mean the end of faith-based schools as presently operated and single sex schools. It is entirely contrary to the Government's diversity agenda.

A lottery system operates in a community in Israel, and it could be effective in certain cultural and social conditions. However, such conditions do not apply in London in 2003. Allocation by lottery is not practical politics.

catchment areas Catchment areas are widely used, and the new Code of Practice makes it clear that it is acceptable as an oversubscription criterion. Under the Code, a map showing the areas should be freely available and in the admissions prospectus. There are many advantages of a catchment system. It encourages primary-secondary links, the transfer of friendship groups, shorter journeys to school. On the other hand, a catchment system may reduce the balance of the intake with respect to the whole borough, let alone the city. It also has well-known effects on housing markets, being similar to the proximity criterion which is currently ubiquitous. It is particularly prone to deception, by means of false claims about the address of a pupil. In principle, it would be possible for catchment boundaries to be drawn to maximise balanced intakes, and to be redrawn continually to compensate for the operation of the housing market, but this would be contentious practice for an admissions forum to undertake.

banding Four London boroughs currently use banding mechanisms. All year six pupils take a test and allocated to one of (usually) four groups according to attainment. All schools in the borough should then recruit proportionately from each band, so ensuring balanced intakes across the borough. This practice was operated by the Inner London Education Authority. The admissions Code allows existing schemes to continue, but prevents new schemes of this kind from being introduced. The banding it permits applies only at individual school level with regard to its applicants.

The effectiveness of this system has been reduced by the ban on leaving places unfilled in one band if there are vacancies in other bands. Another problem in the past was the ability of popular schools to select within each band, so that the social composition of the school was more favoured. Since the interview was a mechanism for this, it would be interesting to see the effects of the interview ban on this tendency.

Although there are mixed views as to the present effectiveness of banding, in a number of boroughs it is still seen as useful. Boroughs which continued its use achieve better balance of intakes than those which did not but it does not seem to answer all the needs of an admissions system as listed above.

Recommendation: Consideration should be given to the possible roles of catchment areas and banding within the single London wide procedure recommended above.

the all-through school There is another alternative which is considerably more radical. The difficulties of the transfer at 11+ could be eliminated by abolishing the transfer. An all through school (at least 3-14) would possibly transfer parental anxiety to a different time in their child's life, or possibly be a means of changing parental thinking as recommended above.

It must be said that a pan-London change of this sort is impracticable for many reasons, including premises and equipment and staffing problems. It also flies in the face of orthodoxy about the optimum size of secondary schools. On the other hand, many children find secondary school too large, and secondary schooling lacking in small-scale humanity. Some of the highest achieving countries in the PISA study, notably Scandinavia, have all-through systems, although there is no evidence as to a causal connection.
London will need many new schools in the next decade. It would be worth experimenting with an extended age range in some places, to enable a more informed judgement on the relative advantages and disadvantages.

Recommendation: When new schools are being planned, consideration should be given to piloting Year 9 as the age of transfer, so that pupils would attend the same school from 3-14.

The New Collaboration

Without recognising it, the Government itself has begun to develop mechanisms which could provide a radical solution to the secondary admissions problem. Its root lies in the developing agenda of diversity and collaboration.

Early descriptions of secondary school diversity seemed embedded within the market model. Competition would spur schools at the bottom of the ladder to climb it and join their neighbours. Yet almost simultaneously came the recognition that those at the bottom, particularly in our cities, were faced with such challenging circumstances that they could not be expected to make the climb unaided. Collaboration between schools is the new theme.

The soft end of collaboration is the partnership with neighbours expected of specialist schools and more urgently expected of Advanced schools which will replace Beacons. However, the Government is aware that collaboration has been a weak feature of previous programmes. Even when the 2002 Education Act was drafted, provision was included for collaboration by the development of schools sharing a single governing body or jointly working through a contract, becoming a federation. This is the hard, formally established end of collaboration.

There are varieties of emphasis about the gains from collaboration, but they all recognise that all schools have both strengths and weaknesses. One is the benefit to a number of schools of the best leadership. A more bottom up model foresees the best practice being spread by means of working links between staff at all levels.

While the Government is very enthusiastic about promoting collaboration, there are a number of difficulties.

There is tension between the concepts of collaborative schools and a market in schools. As observed above, in parts of the country where admissions is not an issue, schools have not been in fierce competition for pupils and collaboration already takes place, and has much scope for development. In London, however, whereas schools in challenging circumstances can immediately see the benefits of joint working, oversubscribed schools which have benefited from the market will be much less enthusiastic.
The Government insists that the development of federations should be on a voluntaristic basis. It offers financial inducements in encouragement, notably the substantial Leadership Incentive Grant.
There remains an undercurrent of the proposition that schools with low pupil attainment are 'bad' schools, and that the main thrust of joint working would be to improve their practice. The collaboration policy as presently conceived does not address the problem of unbalanced intakes in London.

If these issues can be addressed, the concept of secondary school federations could make a substantial contribution to increasing opportunity and achievement, particularly for more disadvantaged pupils, in London.

Recommendation:. London secondary schools should be encouraged to join together in federations.

In this voluntary phase, the curriculum and extra-curriculum and school improvement benefits would be demonstrated. In London, federations would consist of perhaps six schools, a mixture containing perhaps a faith school, mixed and single gender schools, over- and undersubscribed schools. In some cases, a federation would cross borough boundaries. Where appropriate, a local independent school should be invited to join the federation; a refusal should be taken as evidence of a lack of charity.

A variation of the concept is strongly supported by the Commissioner for London Schools, Tim Brighouse, who has piloted the 'collegiate' in Birmingham. The collegiate depends on a group of schools fully committed to being part of a strong community, and each responsible for the whole community. The commitment would appear in all aspects of the schools' work..

All teaching staff in a curriculum area would share responsibility for spreading best practice, and schools' timetables would be co-ordinated to permit this in practice.
Curriculum offers, particularly at 14-19, would be co-ordinated, with all pupils being able to take advantage of specialisms at other schools.
Extra-curricular offers would be co-ordinated. Potentially, this would produce a greater range of opportunities than found in the most elite school. A parent with a musically talented child could send the child to any school in the federation, knowing that she would have access to the federation orchestra, the federation big band, the federation steel band, and more.
Shared responsibility for pupils with special educational needs.

It is important to recognise that collaboration must involve sharing in key respects. Schools must share the load with respect to pupils with Statements of SEN, and with the very large numbers in London who need support but are not statemented. The recent Audit Commission report argued that many schools, with more than half an eye on league tables, try to avoid taking such responsibility. Schools which truly integrate pupils with special needs, who involve the whole school in their education, are fulfilling a vital role in developing amongst their pupils, particularly the more advantaged, a better understanding of the varieties within society, a greater tolerance of those varieties, and making a contribution to a more compassionate society. Indeed, this is one of the social outcomes which can only be achieved by means of a comparatively unsegregated intake.

Achieving this balance, and rebuilding the comprehensive commitment, is a necessary aim of an admissions policy. However, this commitment cannot flourish unless the pressures towards competition rather than collaboration are addressed. An important component of the market structure is school level data on pupil attainment. In a federation with truly shared responsibilities, pupil outcomes at federation level are the object of attention.

Recommendation: When a group of secondary schools forms a federation, target setting and the publication of pupil attainment data should be only at federation level.

An argument in favour of this proposition is that high achieving schools have relatively little scope for further improvement in results, but could make a large contribution to improvements across the federation, and by this mechanism would have the incentive so to do. A counter argument is that schools with low achievement could hide behind a federation outcome, but of course this is simply another example of the 'bad school' fallacy, and ignores the collective pressure for improvement which would drive a federation.

Federating schools may itself, in time, reduce the popularity differentials between the constituent schools, but more needs to be done to enforce balanced intakes. Schools within a federation must collaborate, not compete, on admissions.

Recommendation: When a federation is formed, admissions would be to it rather than to its constituent schools. Within a federation, admissions to the schools would be on the basis of parental preference, child's preference, child's need, and community interests.

Parents would state a preference for the federation. The preference should carry the same weight as within the present system. It would be beneficial if parents could be guaranteed a place in their home federation. Pan-London criteria for oversubscription should apply, with proximity given high priority. At the end of this stage, every child would have a guaranteed place within a federation.

Parents would also state a preference for a school within the federation (excluding any independent school). However, the preference would be weighted against two other interests:

the wishes and needs of the pupil, negotiated as necessary between the federation, the primary school, the pupil and the parent;
the community interest, such as a requirement for balanced intakes, which could be stated in the form of a policy of the admissions forum.

Every child would be allocated a place within one of the schools within the federation.

For example, parents with a well-founded commitment to faith-based education would be entitled to have that preference given due weight. However, the child's view on the matter would also carry some weight, as would their views, and those of the primary school, on a preference to keep a friendship group together. So would the overall importance of the faith school having an intake in balance with the federation as a whole, including its proportion of pupils with SEN, who are difficult or disaffected, or who may be transient. This should not be taken as a requirement to meet quotas rigidly.

For some imaginary case studies exemplifying how the process might work, see the Annexe.

Recommendation: In the longer term, all London secondary schools should form federations.

Conclusion

London's future depends amongst other things on maintaining a sustainable social mix, and moreover a mix of groups who all feel a commitment to London as a community. There is a tendency for some of the more advantaged strata to occupy a different place, and to use different facilities, than the mass of Londoners. This is of course a universal feature of capital cities, and as previously discussed explains the greater take-up of independent education in London than in the country generally.

The task for policymakers is to create the conditions which encourage commitment, in particular to public services. Social justice demands that this is not done at the expense of quality provision for all. The evidence is that London schools offer a better quality of provision than the national average, allowing for the perennial difficulty of retaining sufficient high-quality staff in inner London, and allowing for the high levels of deprivation amongst London children. London local government, together with London Challenge, must ensure that the quality is maximised and available to all. Levels of achievement in real terms must be raised significantly if national targets and important social goals of reducing the links between class, race and attainment are to be reached, and those currently achieving least must raise their achievement disproportionately.

Collaboration between schools is an important tool for that maximisation, and our radical proposals on secondary admissions would both further develop collaboration and send a message to Londoners that a high quality educational offer really can be available to all.

Then there remains a task for the Government, supported by all parties involved in London schools. There is an ethical appeal to the upper strata of London society. The advantaged have a particular responsibility to be part of society and to help build it. In London, the advantaged have a duty to open their minds to the real quality of education, to realise that it is their self-interest to use local provision, and in everyone's interest to play a part in continuing improvement of local provision. By embracing the local federation, they can build and strengthen it. By strengthening local schools, they can make a vital contribution to civil society.

Annex

Case Studies: Transfer to Federation

Kaleda

lives on the edge of an East London borough. On transfer, she applies to her home federation, which consists of four schools in Hackney, and two in Tower Hamlets, and is given a place. Kaleda's main objective is to attend a school with her three closest friends, all of whom attended the same primary school. According to school reports, this group of friends were a positive influence on each other's learning. Kaleda's parents also desire this, but in addition are enthusiastic that Kaleda can develop her aptitude for both her Urdu and Spanish, both of which were taught at Primary school. Kaleda is placed in a mixed Community school a short bus ride or cycle away from her home, as are her friends. In addition, she is guaranteed an additional four hours a week of language tuition for both Key Stages Three and Four at the specialist language school, which is also part of the Federation.

Freddy

attended three different schools in his borough, and during his primary years was one permanently excluded and received fixed term exclusions on another five occasions. At risk of truancy and disaffection, both Freddy's parents and the other agencies involved in his care agree that Freddy should attend a school as close as possible to his home, so that good home-school relationships can be fostered. This school is both popular and high achieving, and Freddy would previously not have been in the catchment area. The Federation allocates Freddy a place at this school, together with some time at a local learning support unit, and some hours out of school working in a voluntary organisation. A learning mentor is assigned to Freddy so that his movements can be co-ordinated, and that links are fostered between home and school. The Headteacher of his new school understands that, if problems do arise and an exclusion is necessary, another school in the federation will have the commitment and spare capacity to take on the challenge of this challenging pupil.

Aleesha

is a high achieving pupil who lives in the middle of a South London borough. She takes the entrance exam for a Kent Grammar school, and is accepted. In addition, however, she applies to the borough Federation, and, once accepted, applies to a school in the North of the borough which used to be highly selective, through interviews and other covert means, but now has a far more balanced intake. She is allocated a place at this school and decides not to go to Kent. She spends the time she saves by not commuting to Kent on a gifted and talented programme which is located in what used to be one of the most unpopular schools in the borough.

Jose

is a Colombian refugee, who arrives in London in March, with little previous experience of schooling. For the rest of the year, he is allocated a school as close as possible to his bed and breakfast accommodation. He spends half his time there, and the other half at a centre dedicated to supporting the learning of newly arrived refugees, on the borders of and funded by three different LEAs. In the Summer, he moves elsewhere to permanent accommodation, where he chooses a Catholic school in his home Federation, but still attends the same Centre as before for two days a week. Even though he lives in another borough, all involved in his development agree that his learning will benefit from the continuity that this Centre can offer him.

Amy

and her parents are very enthusiastic about going to one particular school in a Federation five miles from her home. This school has a reputation for high standards of achievement and discipline. She is admitted to this federation, but does not gain a place at her chosen school, rejected through the Federation's combination of a banding system and catchment area to determine admissions. Her parents do not wish to consider the alternative allocation, although this school is far from failing and, due to the new admissions system, has a similarly mixed intake to every other school in the Federation. Amy's appeal is unsuccessful, so Amy's parents find her a place at a nearby independent school. This school is part of the same Federation, working with some of the most disaffected pupils in the area.

References

Birch D. (2001) The impact of parental choice and open enrolment on liaison between primary and secondary schools unpublished M.Ed. dissertation, Exeter University

Bradshaw J. (2002) The Well-being of Children in the UK London: Save the Children

Crook D., Power S., WhittyG. (1999)The Grammar School Question: a review of research on comprehensive and selective education, IoE: London

DfES (2001) Performance Tables 2001: Value Added Information KS3 to GCSE/GNVQ Technical Annex

DfES (2002) Admissions Appeals for Maintained Primary and Secondary Schools in England 2000/01 Statistical First Release

Gewirtz S., Ball S. and Bowe R. (1995). Markets, Choice and Equity in Education. Buckingham: Open University Press

Goldstein H. (2001) The 2001 Education White Paper and evidence based policy: a commentary, IoE: London

Goldstein, H. and Sammons, P. (1997). The influence of secondary and junior schools on sixteen-year examination performance: a cross-classified multilevel analysis. School effectiveness and school improvement. 8: 219-230.)

Jesson D (2001) Selective Systems of Education - blueprint for lower standards? at:

http://www.york.ac.uk/depts/econ/rc/21pbrf3.pdf

Hallgarten J. (2000) Parents Exist, OK!?, London: ippr

Johnson M. (1999) Failing School, Failing City John Carpenter

Lupton R. (2002a) Case Study: Southside Grange School unpublished Ph.D. dissertation

Lupton R. (2002b) School Quality, Free School Meals And Area Deprivation : Reading Between The Lines

Paper presented at the LSE Research Laboratory All-Centre Event. July 3^rd 2002.

OECD (2001) Knowledge and Skills for Life Paris: OECD

Sammons P., Hillman J., and Mortimore P. (1995) Key Characteristics of effective schools: a review of school effectiveness research, IoE and Ofsted: London

Schagen I. and Schagen S. (2002) Analysis of the Impact of Selection on Pupil's Performance, NfER

Shuttleworth I. and Daly P. (2000) The Pattern of Performance at GCSE Research Paper SEL 3.1 Bangor: DENI

Smith E. and Gorard S. (2002) What does PISA tell us about equity in education systems? Occasional Paper 54 Cardiff University School of Social Sciences

Taylor C. and Gorard S. (2003) Secondary School Admissions in London London: ippr

Templeton J. and Hood S. (2002) Changing schools London: Office of the Children's Rights Commissioner for London

TEN (2003) Education Network media release, 31 January 2003

Thomas S. and Mortimore P. (1996) 'Comparison of value-added models for secondary-school effectiveness' in Research Papers in Education 11(1) pp5-33, Routledge: London

Walford G. (1994) Choice and Equity in Education. London: Cassell.

Welborn S. (forthcoming) Whose choice is it anyway? A study of the availability of secondary school places in London London: Save the Children

West (1991). Choosing a Secondary School: Parents of Junior School Children Educational Research Vol 33 no.1. NFER-Nelson

Williams J, Coldon J, Stephenson K, Logie A and Smith N (2001) Parents Experiences of the Process of Choosing a Secondary School Research Report 278 Sheffield Hallam University DfES

About the Author

Martin Johnson is Research Fellow in Education at the IPPR. He had over thirty years experience as a teacher, mainly in inner London, specialising in working with secondary pupils with behaviour difficulties. He is the author of 'Failing School, Failing City', an account of teaching in the most difficult secondary schools. He was also President of the NASUWT in 2000. Prior to working on 'Schooling in London', he collaborated with Joe Hallgarten on the project 'The Future of the Teaching Profession'.

Martin can be contacted at m.johnson@ippr.org

About the Project

Schooling in London was initiated in Spring 2002. Its proposals were announced at a conference on 5 February 2003. Joe Hallgarten and Jodie Reed worked with Martin Johnson on the project.

The following project papers are on the ippr website:

Not Choice But Champion by Martin Johnson

Secondary School Admissions in London by Chris Taylor and Stephen Gorard

Education Funding - Fair Enough? by Martin Johnson

Education Funding Formula: Increasing the AEN Unit Cost by Nicola Morton

Schooling in London, An Overview by Martin Johnson

Schooling in London is supported by the London Development Agency, PricewaterhouseCoopers in association with the London Partnership, the Random House Group Ltd. and Select Education, and in association with BBC London