Specialisation without selection?
By Tony Edwards
Emeritus Professor of Education, University of Newcastle upon Tyne
In its first education White Paper, the new Labour Government declared its intention of ‘modernising comprehensive secondary education’ by overcoming the ‘tendency to uniformity’ which had marked it in the past. It therefore ‘relaunched’ its predecessor’s Specialist Schools Programme through which secondary schools are encouraged to develop ‘their own distinctive identity and expertise’ in one of four ‘specialist subject areas’. By January 1998, 29O of these specialist schools had already been designated in a hundred English Local Authorities — 210 Technology Colleges, 50 Language Colleges, 17 Sports Colleges and 13 Arts Colleges. Sixty had been designated since the May 1997 Election, and Labour’s Secretary of State placed them ‘at the heart of a drive for diversity and excellence within a modernised school system’ which would ‘cater for individual strengths’ rather than offering ‘a bland sameness for all’. He hoped to see their numbers rise to 450, about one in seven maintained secondary schools, by the end of the present Parliament. 
The tangible benefits of the status have not changed since the Election, nor have some conditions for acquiring it. £100,000 of private sector sponsorship has to be secured, a large sum which schools are very differently placed by their location and contacts to obtain. This is then matched by £100,000 capital grant from the Department (DfEE), which also gives £100 per pupil additional recurrent funding for least three years (up to a yearly maximum of £100,000). In return, the school must have a three-year development plan with quantified performance targets for its specialist area, and must ‘build ongoing links with sponsors’. There are however some significant changes since the Election. Lacking its predecessor’s enthusiasm for competition, the Labour Government insists that specialist schools must make their distinctive expertise and resources widely available to ‘local people and neighbouring schools’ rather than to gain market advantage. Selecting up to 10% intake by ‘relevant aptitude’ must not be misused as a means of introducing or restoring partial selection by ‘general academic ability’. And preference is to be given to applications from schools in those areas of socio-economic deprivation designated as Education Action Zones.
The purpose of this Briefing Paper is not to argue for or against the specialist schools policy, but to review evidence relevant to assessing its more likely effects. The schools are still so new that most of the research involved more general investigations of educational choice which had to be applied to this particular sponsoring of diversity. Some is from countries where curriculum diversity emerged earlier, although any ‘lessons from abroad’ have to be applied cautiously with due regard for the different education systems and cultures from which they are taken.
Claim 1. Parents want greater curriculum choice, and enterprising schools respond accordingly.
Greater consumer choice of schooling, a policy objective pursued in many countries since the early 1980s, has seemed to require secondary schools to be different in kind and not merely more and less effective versions of the same model. Demand for ‘schools of faith’ has grown, as has support for schools with a distinctive educational or social philosophy. But while more parents have become inclined to look beyond their local school for a preferable alternative, extensive research into what they are looking for has produced little evidence that they want, or that schools are providing, different kinds of curriculum.
In this country, Conservative Ministers often claimed that the Local Authority ‘monopoly’ of public education had suppressed a ‘natural’ demand for variety which the self-governing grant-maintained schools would therefore lead the way to satisfying. Yet there is no evidence that either parents or schools associate that status with curriculum change. More generally, research into parents’ reasons for choosing one secondary school over another has not yet revealed any substantial demand for departures from a traditional academic curriculum ‘successfully’ taught. Parents may also look for good facilities for science and technology, a good reputation for sport (especially where the child’s view has been actively canvassed), or a good reputation for music or drama. But none of these specific concerns has appeared among the ‘most important’ criteria - a school’s results, its intake, and the likelihood that the child will be ‘happy’ and ‘safe’ there.. While competition for pupils (especially academically able pupils from ‘good homes’) has certainly prompted some energetic marketing, this has tended strongly to focus on maintaining or enhancing a general reputation as a ‘good school’ which is rarely associated by parents with specific organisational or curriculum characteristics. It has certainly not involved targeting particular kinds of consumer with a distinctive form of curriculum.
There is therefore little evidence of schools responding spontaneously to what they perceive to be a new or emerging consumer demand. What specialisation has occurred has been initiated largely from above, most conspicuously in the creation of city technology colleges (CTCs) in the late 1980s and then through the inducements provided by the 1993 Technology Schools Initiative (TSI) and its subsequent extension through the Specialist Schools Programme to other areas of the curriculum. The CTCs were explicitly intended to stimulate consumer demand for a distinctively modern, high-technology version of secondary education. Yet despite their publicised image as ‘the schools of the future’, the parents who certainly applied in large numbers were more inclined to see them as better resourced with better intakes than neighbouring comprehensives schools, and as combining high-tech facilities with traditional values and a measure of selectiveness. Justifying the continuing priority given to technology in the Conservative Government’s promotion of curriculum diversity, Gillian Shephard argued that it was in the national interest to transform through lively publicity and special funding the image of technical schools as ‘drilling the second-rate for oily rag jobs in factories’. The consequent inducements are illustrated in two research studies, neither of which suggested that the school concerned was already excelling in its ‘chosen’ speciality. An investigation of secondary school recruitment in three ‘local competitive markets’ found only one example of curriculum differentiation, a comprehensive school’s development of technology as a bid for the extra TSI resources. And a grant-maintained girls’ school included in a study of school choice in three London LEAs had received by 1994 £500,000 from the TSI, a further £800,000 for extra science facilities and other ‘plant improvement’, and to have made a further successful bid to the Toyota Technology Fund.
The lack of evidence so far of schools diversifying spontaneously may indicate only a lack of incentives until recently to play to their strengths. Any inclinations to do so have also been held in check by the requirements of a National Curriculum initially constructed in such timetable-consuming form that it left schools no scope for specialising within it. Even the self-consciously innovative CTCs had to offer the full ‘broad and balanced’ version, extra time being created for technology and science through a longer school day and a longer school year. And although the 1988 Education Reform Act made provision for schools to opt-out of the National Curriculum in order to carry out ‘development work’ and ‘experiments’, that escape route was not used. The Dearing review of the statutory requirements then freed (in theory) up to 40% of the secondary curriculum for schools to ‘respond to local circumstances’ or seek a distinctive niche in the local market. This freedom has largely been used to extend pupil choice within a school’s range of non-foundation ‘arts’ and ‘vocational’ subjects rather than to specialise.
It is therefore worth looking to the private sector for examples of diversity because it represents a long-established market not formally bound by the National Curriculum. While the sector has its ‘progressive’ and experimental schools, and although parents pay for private education (especially in small proprietary schools) for many reasons, the dominant reputation and image of private education comes from the academic performance of the market leaders. The Independent Schools Information Service (ISIS) defines their appeal to parents as ‘high academic standards, high expectations of children, a firm disciplinary framework, and smaller class sizes’. A different curriculum has not been part of that appeal except in some ‘progressive’ and some small proprietary schools. Even in countries where the private sector is both more extensive and lacks the peculiarly English association with elite schooling, its main attraction is for those dissatisfied with academic standards and discipline in the available public schools or who seek a distinctive religious or philosophical orientation.
In these countries, but especially in England, the strong association of a traditional-academic curriculum with prestigious schools, selection, and middle-class pupils, has made significant departures from it appear risky. This is the curriculum which has the appeal of familiarity. It is also the curriculum which continues to provide the main route to educational success and consequently enhanced job prospects. In market theory, an enterprising school would identify an unsatisfied consumer demand and position itself to meet it. But it would be a bold school which responded to competition from more academically ‘successful’ neighbours by offering a distinctively ‘modern’ or ‘vocationally relevant’ alternative. As indicated earlier, even CTCs tended to complement their modernity with more traditional virtues. Successive studies of grant-maintained secondary schools in operation have shown a much stronger tendency towards a ‘re-invigorated traditionalism’, doing traditional things better or more visibly, than to doing anything distinctively different. It is not difficult to understand why. In so far as they are more market-oriented than LEA-maintained schools, then the safest ‘product’ to market is traditional academic quality, especially where priority is given to recruiting able pupils and pupils from middle-class homes. If they emphasise part of the curriculum at all, it is unlikely to be technology because this has connotations of vocational training which are not easily reconciled with a high academic reputation. The general conclusion of that research was that the workings of a quasi-market ‘appears to be pulling all schools of whatever type towards conformity’ rather than towards seeking a distinctive market niche.
Claim 2. Specialisation is quite different from selection; recruitment by aptitude is quite different from recruitment by general ability.
The heading of this section embodies a difference more easily defined in principle than in practice. In the 1992 Education White Paper Choice and Diversity for example, specialisation is defined as enabling parents to choose the kind of school which best meets their child’s interests and needs, whereas selection enables schools to do the choosing. But by also expressing its wish to see schools develop ‘a distinctive character’ without being thereby arranged in ‘tiers’ in a single hierarchy of esteem, the Conservative Government accepted in effect that ‘specialisation without selection’ is only possible if parental demand is spread around a range of alternatives.
There is strong research support for that proposition. An OECD survey of the effects of choice in six countries concluded that where demand is heavily concentrated on a dominant model of the ‘good school’, then the schools which most closely resemble that model will be over-chosen and so be able to pick the more desirable applicants. The alternative situation of ‘a range of preferences matched by a variety of school types’ is therefore a ‘happier ideal’. But it is also ‘an ideal more rarely fulfilled’. Even in Germany, where the relatively high status of the technically-oriented Realschule has been cited to justify apparently similar diversity in England, the academic and selective Gymnasien have continued to dominate the market. Other cross-national evidence also indicates strongly that ‘as long as schools tend to be judged on a scale of academic excellence’, parental choice reinforces ‘the existing hierarchy of schools based on academic test results and social class’.
From that perspective, the Labour Government’s support for specialist schools can be defended (at least in large urban areas) as a strategy for avoiding that effect by dispersing demand around a range of alternatives so that parents will have different reasons for making different choices. A frequently cited example of how this can be done successfully is the American Magnet schools. These were designed primarily to resist the segregated intakes of neighbourhood schools by using the ‘magnetic’ pull of a specialised curriculum (or a distinctive cultural ethos or distinctive educational philosophy) to attract students voluntarily across racial and social-class lines. Whether they recruited by ability or simply by interest, they would be less socially selective than either the ‘best’ urban neighbourhood schools or high schools in the more affluent suburbs. Arguing in 1993 against the Conservative education reforms because they were ‘encouraging the tendency of markets to redistribute resources from the weak to the strong’, Michael Barber (now heading the Labour Government’s Standards and Effectiveness Unit) cited the success of Magnet schools and other ‘alternative curriculum’ programmes in attracting a wide non-selective social and ethnic mix and involving parents more actively in schools which they had positively chosen. His argument that specialisation was therefore an altogether ‘different issue’ from the harmful practice of selection by ‘overall academic ability’ was made in very similar terms to those later employed in the 1997 White Paper.
Evidence suggests that they are not so easily separated. Several American school systems have ‘magnetised’ all their schools in a complementary division of labour between different specialisms which might provide a model for Labour’s vision of ‘families of schools’ in large urban areas sharing their expertise. But Magnet schools are normally a selected set of schools within the public system. Claims that their intakes are less socially selective than those of either suburban or inner-city high schools have to be taken with other evidence that they tend to attract relatively high proportions of academically-motivated students to the disadvantage of other schools, either because they are seen as being generally ‘different’ and ‘better’ or because of the attractions of the more ‘academic’ specialisms which some of them offer. Reference to such high profile ‘speciality schools’ as the Bronx High School of Science or the Boston Latin School are especially inappropriate because these schools select by ability rather than by interest. A survey of the rapid spread of Magnet schools to most large urban districts reported that nearly 20% of the schools studied were ‘highly’, ‘very’ or ‘moderately’ selective, and that many others had been able to take advantage of their image of catering for abler, better motivated students to shape their intakes so as to confirm it.
It is clearly possible to develop and to market a curriculum speciality without using it either to select applicants openly or to engage in covert selection. Many Magnet schools avoid selectiveness, and some are required to ‘select’ by ballot when there are too many applicants. In this country, a specialist school may simply be the neighbourhood school which has decided to emphasise a curriculum strength. But where it offers one specialism among the several available to parents and children in that locality, what does it do if demand exceeds supply? Believing in selection, whether by ability or aptitude, the Conservative Government had a ready answer. Wishing to restore grammar schools, it would also have allowed specialist schools to select up to 30% of their intake (and grant-maintained schools up to 50%) if its last (1996) Education Bill had become law. The matter is much more complicated for Labour Ministers because while they support specialisation, they also reject selection by ‘general academic ability’. Any secondary school is to be allowed to select not more than 10% of its intake, but the Clause in the School Standards and Framework Bill (Clause 93) which ‘permits’ limited selection by ‘aptitude for particular subjects’ also excludes new testing of ‘ability or . . . to elicit any aptitude other than for the subject or subjects in question’. In the Standing Committee discussion of that Clause, the Conservative ‘Shadow’ Minister (Stephen Dorrell) dismissed the distinction between testing ability and testing aptitude as ‘an extraordinarily rotten floorboard’ on which to stand; the Liberal Democrat Don Foster accepted the concept of aptitude, but claimed that ‘no-one has yet devised a means of assessing it’; and the Minister (Stephen Byers) both provided a working definition of the concept and insisted that schools like Emmanuel CTC in Gateshead had shown how it could be assessed. Ability, he argued, is ‘what the child has already achieved’. Aptitude is ‘a natural talent and interest that a child has in a specific subject’; it is ‘potential which might flourish and blossom if the child is exposed to particular types of education’.
While research cannot resolve the conceptual confusion, it does not support the Minister’s apparent confidence that the practical problem can be solved. Indeed, as working examples of selection (and self-selection) by interest as well as by aptitude, the CTCs illustrate some of the problems. These schools were required by the contracts with government to recruit intakes representative of their catchment areas by ability and social background. But they were also required to select by ‘general aptitude’ (from evidence of attainment and progress at Primary school) and by a ‘readiness for’ and ‘likelihood to benefit from’ that ‘type’ of education as demonstrated by tests and at interview. Their parents also had to demonstrate at interview their commitment to the CTCs’ intended ethos of ‘enterprise, self-reliance and responsibility’. The assessment both of pupil aptitude and family suitability was therefore unusually overt. It afflicted some senior CTCs’ staff with severe doubts about the practicality of the process, and brought suggestions from within the CTC ‘movement’ that random selection would be fairer where there were too many ‘interested’ applicants for the school to admit.
Some specialist schools, such as the Royal Ballet School, and the Chetham’s and Yehudi Menuhin Music Schools, are of course well practised in assessing performing skills at the point of entry. But their example is largely irrelevant to assessing less tangible aptitudes or ‘natural talents’ supposedly separable from the prior effects of home background, schooling, and special tuition. In relation to the science and technology specialism of the CTCs, two reports commissioned by the Department for Education and Employment were inconclusive about the validity of aptitude testing. One produced no evidence of a consistent relationship between pupils’ apparent aptitude at entry and their subsequent examination performance in that subject area. The other reported very different views of whether and how ability and aptitude could be separated in practice, with two CTCs refusing to use aptitude tests because applicants were too young to be categorised in that way and/or because the tests could not eliminate effects of prior learning and parental encouragement. CTCs generally showed more confidence in interviews as a way of identifying ‘interest’. These have been rejected by the Labour Government except where used by church schools to confirm the appropriate religious affiliation, presumably because they are known to favour children from socially advantaged and ‘educated’ homes.
Among the more general findings from research into secondary school recruitment are that the proportion of parents who get their first choice has not obviously risen since the late 1980s; that the less transparent the procedures and the less overt the criteria, the greater the unfairness; that there are considerable social class differences in the types of school preferred, reflecting processes of self-selection through which some parents aware of the popularity of a school may fall back on a more ‘realistic’ choice; and that greater parental choice has not reduced and may have increased, the social segregation of school intakes. A main conclusion from an extensive survey of comprehensive secondary education was that ‘specialisation is unlikely to be achieved without selection in a fiercely competitive market’ such as operated in many urban areas. It takes this review back to the difficulty of separating the two processes.
Even the appearance of being partly selective seems both to increase the attractiveness of a school to ambitious, ‘educated’ and confident parents and to skew the intake in their direction by discouraging some parents from applying at all. This has been evident in the studies of recruitment to CTCs cited earlier, and from the relatively low proportions of children from low income families in many grant-maintained schools compared with LEA-maintained schools nearby. As noted earlier in relation to Magnet schools, the appearance of being different from the ‘ordinary’ can also help recruitment. The 10% limit on selection by aptitude is justified by the Government as sufficient to support diversity but low enough not to change the ‘character’ of a school. It is certainly too low in itself to transform the fortunes of an unpopular school. But if the specialism permeates the curriculum, which is what it is hoped will happen, then its effects on recruitment may extend well beyond the formal limit.
The potential marketing disadvantages of a technical orientation (other than in high-tech versions) are avoided because Technology Colleges also ‘specialise’ in science and mathematics. But the Arts and perhaps the Language Colleges may attract more than their ‘share’ of girls, who are regarded by some schools seeking to improve their position in the performance tables as a valuable resource. Recent research into the effects of parental choice in London indicated that ‘unusual talent’ in (for example) music and dance was being used by some schools to increase the proportion of academically able children from middle-class families. Such partial selection may fulfil the Magnet objective of making a school more truly comprehensive in ability and family background than if it recruited entirely or mainly from its neighbourhood. But it may also enable over-subscribed schools to favour applicants thought likely to enhance future results and unlikely to be a drain on resources. Contemporary French research found that secondary schools wishing to improve the academic quality of their intake tended to display that intention by emphasising modern languages or mathematics, or by introducing special options in dead or more exotic languages; comparative research in Israel and the United States supports predictions that low-status parents may be channelled towards ‘zone’ (neighbourhood) schools which have no specialised curriculum offerings and poor resources.
Claim 3. Specialised schools are more effective; greater curriculum diversity will therefore help to raise educational standards.
The proposition that specialist schools are likely to be especially effective is partly an application of more general findings about the benefits to schools of defining their core business clearly and recruiting students (and staff) accordingly. But moving from that general proposition to the benefits of a particular curriculum focus is complicated. Although those located in Education Action Zones seem likely to be given unusual freedom to depart from or even to ‘re-design’ the National Curriculum, specialist schools have to enhance their chosen subject area without undermining other subjects to make room for it. They would therefore seem to be extending rather than reducing their task. Evidence of the beneficial effects of more clearly defined purposes comes mainly from schools which represent ‘communities of values’ with parents who share the same religious or philosophical beliefs, or from schools enabled by highly selected intakes to maintain a pervasive academic orientation. It is not obvious that a specialised curriculum would provide a similar sense of common purpose. Curriculum differences have received very little attention in school effectiveness research, the strong tendency to concentrate on school organisation and leadership reflecting the lack until recently of substantial curriculum differences to be investigated. A typical conclusion, that curriculum content made no difference to pupil achievement, is the more significant for coming from research in the Netherlands where parental choice has been unusually extensive and the private sector unusually large. Indeed, the international comparisons often favoured by advocates of a more open market in schooling tend to show the higher-achieving systems to be those with a consistently common curriculum.
Yet belief in the motivating effects of different kinds of learning goes back a long way. Thus the tripartite system of ‘secondary education for all’ constructed after 1944 was justified on the grounds that children did best when the curriculum was shaped by a pervasive (rather than subject-specific) emphasis, and that the types of mind suited to ‘academic’ or ‘technical’ or ‘practical’ modes of learning could be identified quite accurately as early as the age of 11. Scepticism about the validity of such early assessment was reinforced by objections to restricting access to a full ‘liberal’ education because much of it was assumed to be too difficult for most children and irrelevant to their future employment. Comprehensive secondary education was intended to override those assumptions. The breadth of the National Curriculum then challenged directly both social-class and gender stereotyping of subject ‘choice’. The Specialist Schools Programme represents a move away from ‘uniformity’, with beneficial effects expected to extend well beyond the specialism itself. Yet it is surely too soon to claim the speciality ‘spills over into other curriculum areas, making the school attractive to teachers of all subjects’ and creating unusual opportunities for independent learning, or to determine whether specialist schools are being ‘genuinely encouraged to adopt different educational models rather than merely competing for funds attached to labels such as “technology colleges”’. It is certainly too early to identify a distinctive specialised ‘style of education’, a phrase used to justify opening up membership of the City Technology Colleges Trust to ‘affiliated schools’ which had adopted or wished to adopt the same educational ‘style’. A small minority of Magnet schools offer a distinctive pedagogy, but they are not generally known for innovations in teaching and learning.
They have prompted, however, many comparisons of their performance with unspecialised ‘zone’ (neighbourhood) schools. For reasons noted earlier, the evidence is inconclusive. In school systems which are partly ‘magnetised’, student achievements in Magnet schools are generally somewhat better. But some Magnet programmes are school-wide, some are available only to selected students, some are integrated through the curriculum and some are set apart as an addition. As indicated earlier, some Magnet schools are openly selective and many others attract an unusually high proportion of able applicants from ‘educated’ families Their apparently greater effectiveness could be the result of a capacity to select pupils from whom academic success is easier to obtain.
The CTCs may appear to offer a more relevant example of the effects of specialisation because their intakes are supposed to be representative of the areas from which they recruit. In the 1997 school performance tables, almost all CTCs were well above the average for their LEA. But large differences between them (for example, from 34% to 89% in the proportion of GCSE entrants achieving 5 or more A-C passes) suggest considerable differences in their intakes and illustrate the difficulty of generalising about the causal effects of specialisation itself. If this is the case for relatively well-established CTCs, it is even more misleading to cite the numbers of much newer specialist schools among the ‘best performing’ and ‘most improved’ comprehensive schools of 1997 if the implication is that specialisation had caused their relative success. For those which were already ‘good’ or ‘improving’, their recently acquired status is as yet more a reward for past achievement; for those which were not, it is too soon to establish an explanation for their improvement.
The Government’s expectation is that these schools will provide local and regional ‘centres of achievement and excellence in their specialist subjects’, and be a focal point for revitalising education in areas of ‘social disadvantage’. This is why every new Education Action Zone is intended to have at least one. There are loud echoes here of the ‘beacons of excellence’ role highlighted in the CTCs’ mission. Certainly conditions are more favourable for specialist schools to perform that function. The CTC programme aroused strong objections nationally because of their favourable funding, and some CTCs faced additional hostility from neighbouring schools threatened by the appearance of a new and apparently favoured competitor in inner-city and other urban areas still experiencing falling rolls and the closure or merging of established schools. Schools in the Specialist Programme are redesignated not newly created or rebuilt, although their additional funding may cause some resentment. It is the Government’s intention that they will use their special expertise for the benefit of the local community and not to secure the market advantage which their distinctiveness and extra resources may bring. Yet the potential is there for another level in a hierarchy of schools. From this wider perspective, the Magnet model is important but complicated. As noted earlier, evidence of their generally greater effectiveness has to be placed in the context of somewhat favoured intakes, the effects on recruitment and standards elsewhere, and indications that those worst served by the pre-Magnetised system are under-represented among the beneficiaries of the new. Since the association of high school performance with socially and academically ‘balanced’ intakes is strong, the claim that ‘diversity and specialism will boost standards’  can only be properly tested by looking beyond the specialist schools to the local and national systems in which they are embedded.
Education policy cannot be determined by research because the priorities it represents are so largely a matter of value-judgements. Even evidence thought relevant to assessing a policy’s success, as well as its interpretation, is likely to appear differently from different value positions. Like its predecessor, the present Government claims that a standard, uniform curriculum ignores both the real differences in children’s aptitudes and interests, and the increasing cultural heterogeneity of modern societies. This Briefing Paper has not presented a counter-case for uniformity. But its conclusions, summarised at the outset, indicate grounds for caution about the effectiveness and fairness of a policy which successive Governments have vigorously promoted.
I am grateful to RISE Trustees for comments on an earlier draft of this Briefing Paper, and to Frank Coffield, Patrick Eavis, David Halpin, Sally Power, Geoffrey Walford, Tony Webster and Geoff Whitty who also responded critically to that version. Ronald Eunson, Caroline Gipps, David Hopkins and Chris Whetton replied helpfully to specific queries. None of the above is responsible for the use I made of their advice, nor is RISE for the contents of the Paper.
The review is organised around some of the main claims made in support of what has appeared to successive Governments to be an evidently good thing. These are: (i) that parents want greater curriculum choice; (ii) that specialisation is quite different from selection; (iii) that diversity through specialisation will raise educational standards in and beyond the specialist schools themselves. Although it is too soon for a decisive assessment of the validity of those claims, the weight of evidence supports the following conclusions :-
References and recommended reading:
 Department for Education and Employment, Excellence in Schools, London: HMSO 1997. David Blunkett’s enthusiasm for specialist schools was expressed in speeches to the Technology Colleges Trust, 5 November 1997 and at the Trust’s annual dinner 25 February 1998; the quoted comments are taken from the DfEE press releases.
 This is true of all eight national case studies reported in a special issue (edited by Geoff Walford) of the Oxford Studies in Comparative Education 6(1), 1996.
 J. Fitz, D. Halpin & S. Power, Grant-maintained Schools: Education in the Marketplace, London: Kogan Page 1993; for later findings, see note 12.
 For summaries and examples of substantial research, see M. Adler, Looking backwards to the future: parental choice and education policy, British Journal of Educational Research 23,1997, pp.297-313; F. Echols & J. Willms, Reasons for school choice in Scotland, Journal of Education Policy 10(2) 1995, pp.143-156; S.Gorard, Three steps to ‘heaven’? The family and school choice, Educational Review 48(3) 1996.
 C. Bagley, P. Wood & R. Glatter, Scanning the market: school strategies for discovering parental perspectives, Educational Management and Administration 24(2) 1996, pp.125-138; D. Bridges & T. McClaughlin (eds) Education and the Market-place, Lewes: Falmer Press 1994; S. Gewirtz, S. Ball & R. Bowe, Markets, Choice and Equity in Education, Buckingham: Open University Press 1995, especially pages 121-154; D. Smedley, Marketing secondary schools: some lessons from the research on parental choice, Educational Management and Administration 23(2) 1995, pp.96-103. The significance of a school’s general reputation, or what this Israeli researcher calls ‘an aura of distinction’, is explored in M. Tatar, The prestige of the high school as viewed by parents, British Journal of Sociology of Education 16(1) 1995, pp.93-108.
 G. Whitty, T. Edwards & S. Gewirtz, Specialisation and Choice in Urban Education: the city technology college experiment. Routledge 1993, pp.82-89; G. Walford, Choice of school at the first CTC, Educational Studies 17(10) 1991, pp.65-75
 Speech to the City Technology Colleges Trust, 5 November 1995; the objectives identified for specialist schools in that speech are very similar to those stated by David Blunkett on the comparable occasion two years later (see note 1).
 R. Glatter, P. Woods & C. Bagley, Diversity, differentiation and hierarchy: school choice and parental preferences, in R. Glatter et al (eds), Choice and Diversity in Schooling, London, Routledge, 1997; S. Ball, Good school/bad school: paradox and fabrication, British Journal of Sociology of Education 18(3) 1997, pp.317-336.
 J. Tooley, Loopholes in the National Curriculum? The genesis and uses of Sections 16 and 17 of the 1988 Education Reform Act, Education and Law 8(4) 1996, pp.319-330.
 ISIS, Annual Census 1994, London: ISIS. This academic emphasis is examined in T. Edwards & G. Whitty, Marketing quality: traditional and modern versions of educational excellence, in R. Glatter et al, Choice and Diversity in Schooling, London: Routledge, 1997. Diversity within the private sector in Wales is described by S. Gorard, School Choice in an Established Market, Aldershot: Ashgate, although a significantly different curriculum was only found in some very small proprietary schools. Despite its title, there is no reference to curriculum innovation in T. Tapper, Fee-paying Schools and Educational Change in Britain, London: Woburn Press 1997; this concentrates largely on leading boys’ schools.
 A. van Zanten, Market trends in the French school system, Oxford Studies in Comparative Education 6(1) 1996, pp.63-75.
 J. Fitz, D. Halpin & S, Power, Between a rock and a hard place: diversity, institutional identity and grant-maintained schools, Oxford Review of Education 23(1) 1997, pp.17-30; D. Halpin, Fragmentation into different types of school: diversifying into the past? in R. Pring and G. Walford (eds), Re-affirming the Comprehensive Ideal, London: Falmer Press 1997.
 D. Hirsch, A Matter of Choice, Paris: OECD Centre for Educational Research and Innovation 1994. The pulling power of schools with strong academic records is documented in H. Lauder et al, Trading Places: the nature of choice in educational markets in New Zealand, Wellington, Ministry of Education 1995; the findings are summarised in S. Waslander & M. Trupp, Choice, competition and segregation: an empirical analysis of a New Zealand secondary school market 1990-1993, Journal of Education Policy 10(1) 1995, pp.1-26.
 M. Weiss & B. Steinert, Germany: competitive inequality in educational quasi-markets, Oxford Studies in Comparative Education 6(1) 1996, pp.77-94.
 G. Whitty, Creating quasi-markets in education: a review of recent research on parental choice and school autonomy in three countries, Review of Research in Education 22 1997, pp.3-47. British evidence is examined in G. Whitty, S. Power & D. Halpin, Devolution and Choice in Education: the School, the State and the Market, Buckingham: Open University Press.
 Hopes for the desegregating effects of Magnet Schools are discussed and ‘tested’ in M. Metz, Different by Design, New York & London: Routledge 1986; in her contribution to Choice and Control in American Education (see note 18); and by J. Coleman, Some points on choice in education, Sociology of Education 65, 1992, pp.260-262.
 Michael Barber, Raising Standards in Deprived Urban Areas, National Commission on Education Briefing Paper No.18, 1993. As the text makes evident, Barber was a prominent co-author of Labour’s 1997 White Paper.
 R. Blank, Educational effects of magnet schools, and G, Orfied, Do we know anything worth knowing about the educational effects of Magnet schools?, in W. Clune & J. Witte (eds), Choice and Control in American Education (Volume 2), Basingstoke: Falmer Press 1990. At a time when Education Secretary Kenneth Baker was extolling their effectiveness, similar doubts about their relatively advantaged intakes were expressed in the HMI Report Teaching and Learning in New York City Schools, London, HMSO 1990 (see also note 31).
 Parliamentary Debates, Standing Committee A, School Standards and Framework Bill (Eighteenth Sitting, Tuesday 24 February 1998), pp.652 &657.
 G. Whitty, T. Edwards & S. Gewirtz, Specialisation and Choice in Urban Education: the city technology college experiment. Routledge 1993, pp.65-81.
 M. Coffey & C. Whetton, Aptitude Tests of Technology; an investigation of aptitude and its relationship with GCSE scores, London: DfEE 1996; M. Barber & B. McCallum, City Technology Colleges and Technology Colleges: review of admissions procedures at Year 7, London: DfEE 1996. Both the technical complexities and the need to take account of the consequences of particular approaches to testing are discussed in C. Gipps & P. Murphy, A Fair Test? Assessment, Achievement and Equity, Buckingham: Open University Press 1994.
 Evidence from which these conclusions are drawn include the Audit Commission, Trading Places: the supply and allocation of school places, London: Audit Commission 1996; A. West, M. David, J. Hailes & J. Ribbens, Parents and the process of choosing secondary schools: implications for schools, Educational Management and Administration 234(1) 1995, pp.28-38; A. West, H. Pennell & P. Noden, Admissions to Secondary School: towards a national policy? London School of Economics and Political Science, Centre for Educational Research 1997; D. Reay & S. Ball, Spoilt for choice? The working-classes and educational markets, Oxford Review of Education 23(1) 1997, pp.89-101; J. Willms, School choice and community segregation: findings from Scotland, in A. Kerckhoff (ed), Generating Social Stratification: towards a new research agenda, Boulder Colarado: Westview Press 1996.
 A. Kerckhoff, K. Fogelman & J. Manlove, Going Comprehensive in England and Wales, London: Woburn Press 1996, p.44.
 C. Benn & C. Chitty, Thirty Years On: Is Comprehensive Education Well, or Struggling to Survive? London: David Fulton 1996, pp.142-3.
 S. Gewirtz, S. Ball & R. Bowe, Markets, Choice and Equity in Education, Buckingham: Open University Press 1995; S. Ball & S. Gewirtz, Girls in the education market: choice, competition and complexity, Gender in Education 9(2) 1997, pp.207-222.
 A. van Zanten, Market trends in the French school system, Oxford Studies in Comparative Education 6(1) 1996, pp.63-75. E. Goldring, Parental involvement and school choice: Israel and the United States, in R. Glatter et al (eds), see note 7.
 J. Coleman, Choice, community and future schools in W. Clune and J. Witte (eds), Choice and Control in American Education (Volume 1), Basingstoke: Falmer 1990. A similar view that cohesive schools are more likely in schools serving distinct cultural sub-communities is expressed by D. Hargreaves, The Mosaic of Learning: schools and teachers for the next century, London: DEMOS 1994 and Diversity and choice in school education: a modified libertarian approach, Oxford Review of Education 22(3), 1996, pp.131-41 and 155-6.
 See the chapter by Creemers in D. Reynolds & B. Creemers (eds), School Effectiveness: research, policy and practice, London: Cassell 1992. Curriculum does not appear in the index of D. Hopkins, M. West and M. Ainscow, Improving the Quality of Education for All, London: David Fulton 1996.
 A. Green, Educational achievement in centralised and decentralised systems, in Green, Education, Globalisation and the Nation State, Basingstoke: Macmillan 1997.
 D. Hargreaves, A road to the Learning Society, School Leadership and Management 17(10) 1997, pp.9-21; D. Hirsch, Policies for school choice: what can Britain learn from abroad? in R. Glatter et al (eds) Choice and Diversity in Schooling, London: Routledge 1997.
 Problems of ‘unclear evidence and premature assessments’ are explored by J. Henig, Rethinking School Choice: the limits of the market metaphor, Princeton New Jersey: Princeton University Press 1994, pp.128-145. See also D. Moore and S. Davenport, School choice: the new improved sorting machine, in W. Boyd & H. Walberg (eds), Choice in Education, Berkeley California, McCutchan; the chapters by Moore, Bennett, and Alves & Willie, in W. Clune & J. Witte, Choice and Control in American Education, Basingstoke, Falmer Press 1990; V. Lee, R. Croninger & J. Smith, Parental choice of schools and social stratification in education: the paradox of Detroit, Educational Evaluation and Policy Analysis 16(4), pp.434-457.
 These figures of successful specialist schools are included in a ‘note to editors’ in a DfEE News 25 February 1998.
 ‘Diversity and specialism will boost standards - Blunkett’ is the heading of DfEE News 5 November 1997, reporting the Secretary of State’s speech to the Technology Colleges Trust (see note 1).