SUMMARY OF MAIN POINTS

—  The link between diversity and choice is subtle and ambiguous: diversity may lead to family perceptions of either increased or reduced choice

—  The available evidence does not indicate a widespread demand among families for between-school diversity

—  There are indications of a "pecking order" of specialisms developing which could reinforce existing hierarchies

—  A pluralist, equitable approach to diversity would require family preferences to be widely spread among available schools

—  New specialist areas might well be broader in scope than existing ones, including some that are not focused on particular subject groupings

—  Further specialist areas should initially be assessed for their strengths and weaknesses as "brands" and subjected to consultation with stakeholders

—  Thorough community consultation in local areas on proposals for diverse secondary schooling would be desirable on grounds both of effectiveness and democratic involvement

—  International data suggests that enhancing within-school diversity may be particularly productive

—  Schemes promoting diversity with collaboration are potentially significant and relevant issues are taken up in the separate paper from the Diversity Pathfinder evaluation team

—  A Supporting Paper giving a brief overview of research and key issues is provided with this Memorandum: members may wish to concentrate on the concluding Section 3 of that paper

  1.  The Supporting Paper offers a brief overview of research and key issues relevant to school choice and diversity in England since the mid 1990s. Some of this will already be familiar to Members from previous sessions but it may nevertheless be useful to have such a short review available. Members may wish to concentrate on the concluding Section 3, headed "Diversity and choice: a new direction in schooling, or a buttressing of the old?"

  2.  Much of the debate in previous sessions concerned comparisons of measurable pupil attainment in different types of school. However the Government has justified the diversity policy in terms not only of raising standards but also of offering more choice. Diversity will create schools with a distinctive ethos and mission, this will provide greater choice and a better fit with individual needs, and therefore, by implication, greater pupil and parent satisfaction. The Supporting Paper suggests that, despite a strong focus on choice and diversity in schools policy over the past decade, the precise connection between them is very little understood and needs much closer attention. The relationship between them appears subtle and ambiguous.

  3.  Just as choice does not necessarily lead to greater diversity, so diversity may not produce perceptions of increased choice. For example it is not self-evident that defining school missions more sharply in terms of subject specialisation will lead to a perception of enhanced choice among families with 10-year-old children. The perception could instead turn out to be one of reduced choice, particularly among the many families in all types of area (not just rural ones) who consider that their realistic choice of schools is very limited. School choice presents significant logistical problems, in terms both of provision and access. Families may also perceive unwelcome pressure to form a judgement about their child's aptitudes at an early age.

  4.  The available evidence does not indicate a widespread demand for between-school diversity, though this could change were a range of school types to be made available. Current diversity policy is heavily focused on one specific form of diversity, namely subject specialisation (bearing in mind that as well as specialist schools a key feature of academies is their subject specialisation). There are also indications of a "pecking order" of specialisms developing which could reinforce existing hierarchies.

  5.  The Government appears to be searching for a form of diversity which is pluralist and equitable in nature. This would need to be one that avoids family choices being concentrated in a small number of popular schools within local areas, ie in which preferences are widely spread among available schools. Given historical and cultural factors in the English context, as well as logistical constraints, this will not be easy to achieve. However one approach might be to specify new specialist areas that are broader in scope than some of those specified currently, including some that are not focused on particular subject groupings. Defining schools on the basis of subject specialisation is clearly not the only way to promote distinct identities.

  6.  The Institute of Public Policy Research proposed in 2001 that secondary schools should be able to specialise in any areas they choose, so long as they can make a strong enough case and their bid can meet stringent criteria. If this suggestion is considered too sharp a break from current policy, the government itself made an interesting proposal in its Green Paper just before the last election. This involved designating some schools as Beacon schools which achieve particular effectiveness "in teaching the skills relevant to the emerging economy, such as creativity or the use of ICT" or which "demonstrate excellence in working with their community" (DfEE, Schools Building on Success, 2001, paragraph 4.17). Since the Beacon programme is now to be phased out, such options could be added to the specialist schools programme. Environmental education, citizenship and social inclusion are among other areas that have been suggested. These are simply examples: any further areas should ideally be identified only after careful analysis of their strengths and weaknesses as "brands" and be subject to consultation with stakeholders.

  7.  The Government is seeking to promote a plural and equitable approach to diversity through schemes that encourage schools with distinct identities to collaborate with one another. Examples include the Diversity Pathfinder initiative, federations and the collegiate structures originally developed in Birmingham. Such schemes are of great interest and potential significance. They are still at an early stage of development and have a long history of institutional separateness to contend with. A paper on behalf of the Diversity Pathfinder evaluation team, of which I am a member, is being submitted to the Committee separately.

  8.  I include a further consideration although I recognise that it probably goes beyond current policy parameters. At present, decisions relating to diversity in particular local areas do not include community consultation, except where there are statutory processes concerning closure or reorganisation. Moving to specialist school status, for example, is entirely a matter for the school staff and governors, the LEA and the DfES. Yet it is arguable that current and prospective parents and local communities should have some ownership of such status changes for particular schools, and also of the overall character of school provision in an area, to enhance their support for them. Recent moves towards extending school diversity in Australia, both in Queensland and Western Australia, included specific arrangements for intensive community consultation. Such consultation could provide an incentive for schools to develop a stronger community focus, identified by Ofsted as the weakest element of specialist schools' work. Moreover, in a period when many other forms of change in local service provision are required to be subject to consultation processes, should such significant alterations in the character of schools be exempted?

  9.  Finally, within-school diversity needs to be carefully considered alongside between-school diversity. In the UK, the variation in student performance in reading is much greater within secondary schools than between them, according to the OECD's Programme for International Student Assessment (PISA). Our level of between-school variation was found to be substantially less than the OECD average, whereas our within-school variation is considerably larger. This suggests that enhancing within-school diversity, that is enabling individual schools to cater more effectively for a diverse client base, may be particularly productive. It draws attention also to the importance of linking other policies, such as that on 14-19 education, closely with that on school diversity.

January 2003

  This paper offers a brief overview of research and key issues relevant to school choice and diversity in England since the mid 1990s.

1.  CHOICE

  The principal elements of the education quasi-market in England introduced by the Conservative government's Education Reform Act of 1988 have frequently been described (for example: OECD, 1994; Whitty et al, 1998; Tomlinson, 2001). There was a considerable extension of parents' rights to choose a state school for their child ("more open enrolment"). Schools became funded by formula based largely on the number of pupils on roll and were required to manage delegated budgets including staff salaries. Crucially these market-based measures were complemented by a strong form of performance regulation, including a national curriculum, frequent testing and the publication of school test and performance tables. A national system of regular inspections controlled by a government agency, the Office for Standards in Education (Ofsted), was instituted in 1993.

  The Labour government first elected in 1997 has retained the essential elements of this system. "The main structures of the quasi-market are still in place—parental choice, open enrolment, funding following pupils, school diversity and publication of league tables" (West and Pennell, 2002, p. 218). It has however made some adaptations. For example, the market emphasis is being enhanced through encouragement for successful and popular schools to expand and to take over weak and "failing" schools (Blair, 2002). On the other hand there is now increased regulation of the school admissions process through a code of practice and an adjudication system. Projects such as "Excellence in Cities" (DfEE, 1999) designed to target resources to areas with high levels of disadvantage have been established. Value added measures are being introduced to school performance tables. There is a strong emphasis, which was not present under the Conservatives, on partnership and the sharing of expertise between schools. Perhaps of particular significance is a much enhanced focus on school diversity, particularly through a large expansion in the number of specialist schools: "This greater diversity is good for pupils and parents and will ensure there is more choice and innovation in the school system" (Morris, 2001, p. 7). The rationale for this emphasis on diversity and innovation may be understood from a brief discussion of one of the major research studies of the operation of the quasi-market under the Conservatives.

  A substantial longitudinal (1991-96) project—the Parental and School Choice Interaction (PASCI) study (Woods et al, 1998)—contained three inter-related sets of findings of particular relevance to subsequent policy.

  First, the study noted a tendency for schools to "privilege" the academic aspects of their provision as a response to more market-like conditions. This appeared to be less a reflection of parental preferences, since most parents do not emphasise the academic over and above personal and social factors, than of the policy environment which provides strong incentives in this direction through, for example, the published performance tables accentuating academic performance. Second, the tendency for schools in England to appeal to a broad grouping of potential parents and pupils rather than to differentiate themselves sharply in order to focus on a specific niche, noted in the OECD's (1994) report on school choice in six countries, was confirmed. This tendency towards homogenisation arose both from central prescriptions such as the national curriculum and also from market incentives promoted by per capita funding and more open enrolment. Third, and closely connected to both the previous points, there was little evidence that the competitive arrangements established in England in the 1990s had encouraged innovation within the system. Where innovation did take place it was running counter to the centralising trends of policy, and there were indications of it being curbed sometimes by a reluctance to appear to step outside the dominant model of the high status school.

  These are of course broad generalisations drawn from the detailed study and need to be understood as such. We will return later to the issues raised by the findings.

  In the later 1990s the English research on choice became increasingly quantitative, including attempts to probe the connection between the competitive system and educational outcomes. The PASCI study had already found that the most consistent improvement in exam pass rates over a four-year period took place in the least competitive of its three case study areas, which was in a semi-rural location. However, a later study based on a more quantitative methodology found some evidence of a link between degrees of competition in local areas and rates of examination improvement over time (Levacic, 2001). In a sample of over 300 schools, a statistical association was found between heads perceiving that they were in competition with at least five other schools and performance in the "headline" performance measure of five or more grade A* to C in the General Certificate of Secondary Education (GCSE) examination. The author suggests that "this is due both to greater stimulus to improve and maintain the school's position in the local hierarchy and to more opportunities for co-operation and emulation related to product quality" (ibid, p. 40). However, as the author indicates, the finding must be interpreted with caution. First, another key indicator—the degree of competition as perceived by the head—was not found to be associated with performance improvement. Second, it relates to only one performance measure: the limitations of this particular measure as an indicator of the achievement of all pupils in a school have been widely recognised, despite the political significance that has been accorded to it.

  Gorard and his associates pursued a different issue through quantitative analysis: whether choice and competition increases polarisation. Analysing data for every state-funded school in England and Wales over a twelve-year period, they found that overall segregation in terms of poverty had declined between 1989 and 2001: although it began to rise after 1997, in 2001 it remained below the 1989 level (Gorard et al, 2002a). They attributed this finding to three sets of factors:

—  local social geography, such as the pattern of local housing

—  school organisation at a local level, including closures and mergers of schools (which tend to decrease local segregation) and selection and school diversity (where higher levels of segregation tend to be found) and

—  school admission systems.

  With regard to the latter, the authors' data suggest that local education authorities (LEAs) which use catchment-area based systems, and LEAs in which a large proportion of schools are their own admission authorities (such as voluntary-aided and foundation schools) have higher levels of segregation. One of the authors' overall conclusions is that "Choice policies do not appear to have either the clear benefits their advocates had hoped or the dangers of segregation their opponents feared" (ibid, p. 36).

  This study has generated a bitter academic and methodological dispute. For example Gibson and Asthana (2000) published data indicating that, within local markets, initially high-ranking schools have been drawing to themselves the most advantaged pupils and improving their GCSE performance fastest. They claim this gives solid support to the thesis that competitive markets in schooling promote social polarisation. Noden (2000) criticised the Gorard et al study for using an inappropriate measure of segregation and proposed an alternative. Using his alternative as well as Gorard's measure he concluded that there had been a slight increase in social segregation between 1994 and 1999.

  From a smaller-scale study of the secondary school transfer process in London, Noden et al (1998) found that middle-class families gained access to significantly higher scoring schools in terms of GCSE passes. There was little evidence that this was due to where they lived ("selection by mortgage"), but appeared to be because they could afford to travel further in order to flee low-scoring inner city schools and because some schools had adopted admissions policies favouring more privileged applicants. More recently a government-sponsored survey of parents' experience of school choice drew attention to the role of cultural capital as a resource for promoting access to desired schooling (Flatley et al, 2001). Better-educated mothers were much more likely than others to say they knew how pupil allocations to popular schools were carried out. Owner-occupiers and mothers of white ethnic origin were also particularly likely to assert that they understood the technicalities of the allocation process. This study also indicated that parents in London were least likely to be offered a place in the school they most wanted (nearly 70% compared with 85% nationally). London parents were also less likely to apply to their nearest school than those living in other areas (including other urban areas) and they were the least satisfied with the outcome of the application process.

  From this necessarily brief and selective review of the substantial body of research on school choice in England, some general points might be made. There are evident methodological difficulties involved in investigating the effects of such a complex set of policy developments. These difficulties are rendered more acute when other reforms, some of which were intended in part to counteract the impact of marketisation, were being introduced at the same time, and when the changes themselves were and remain the subject of intense ideological debate. This cluster of factors may explain why the research results do not point unequivocally in one direction, for example over the question of polarisation. However an alternative explanation may be that even policy changes that appear radical when they are first proposed and implemented may have a much more limited impact than expected because of deep-rooted social and geographical factors and because of coterminous trends and forces that operate to reduce their effect. For example, Gorard et al, (2002b) found no evidence of the predicted school "spirals of decline", attributing this finding to school rolls being higher than they would otherwise have been because of a rising school population and school closures and mergers during the period in question. Despite his criticisms of the Gorard methodology, Noden makes a similar general point: "The sustained population loss from some declining urban areas, and in particular the loss of more advantaged families, may be of greater importance to changes in the social mix of local schools than any `within-LEA' quasi-market effects" (Noden, 2000, p. 383). Subsequent research and analysis have tended to confirm Gewirtz et al's assessment in their pioneering study carried out in the early 1990s: "The diversity of local settings and the particularity of their politics, social geographies and histories make it difficult to generalise about market forces in education" (Gewirtz et al, 1995, p. 57).

2.  DIVERSITY

  The Labour government has put great emphasis on an enhancement of school diversity, arguing that "each school should have its own ethos and sense of mission" (DfES, 2002, p. 17) to combat the excessive uniformity which they claim the existing comprehensive system developed since the 1960s has promoted. However the research referred to above suggests that the reforms initiated by the 1988 Act were particularly strong drivers of uniformity and homogenisation.

  This greater diversity is being achieved in large measure through a major extension of the Conservatives' "experiment in specialisation" with plans for at least 2,000 specialist schools by 2006 (DfES, 2002a)—there were about 570 in 2001—when they would constitute more than half of all secondary schools. Specialist school status is to become available to all schools that can submit convincing applications. Secretary of State Charles Clarke has said that ". . . Specialist schools lie at the heart of our drive to raise standards and offer more choice in secondary schools" (DfES, 2000b). Specialisms that schools could previously bid for covered technology, languages, sport and the arts and in 2001 the list was extended to include engineering, science, "business and enterprise" and "mathematics and computing". These schools have to set and meet targets in the specialist area and raise business sponsorship for a relevant project: they receive additional government grants, including an element for co-operation and sharing of expertise with other schools.

  Diversity has also been promoted by providing encouragement for schools supported by the churches and other faith groups. A few Muslim, Sikh and Greek Orthodox schools have been brought inside the state system and are funded as "voluntary aided" schools on the same basis as Church of England, Roman Catholic and Jewish schools. The government proposed changing the capital funding arrangements to make it easier to establish new schools of this type. This became a highly controversial proposal prompting fears of increased racial segregation and the teaching of contentious religious doctrines such as creationism (Branigan, 2002). While stressing the need for faith-based schools to be "inclusive" (DfES, 2001), the government removed this feature of the diversity policy from relevant official documents (for example DfES, 2002a).

  Given the salience of the specialist school model in current policy it is worth reviewing some relevant research. West et al (2000) undertook a survey of existing specialist schools funded by the government. By far the most common reason cited for seeking specialist school status (by 51% of the headteachers responding) was the additional money it would bring from sponsors and the government. More than half the heads (53%) said that the specialism chosen for the bid was not the school's strongest teaching area. These two responses might suggest a predominantly tactical approach to the opportunity of specialist school status rather than a strategy born out of educational conviction. In terms of the requirement to benefit other schools, work with feeder primary schools was the most common form of collaboration (as would be expected in a competitive environment). With respect to other secondary schools, links tended to be with more distant schools such as other specialist schools, those with common sponsors or schools in other countries. In a parallel government-funded study based on case studies of twelve specialist schools, Yeomans et al (2000) reported that across all their schools the weakest links were with neighbouring secondary schools. An evaluation by Ofsted (2001) concluded that specialist schools were weak in sharing resources and expertise with local schools and the wider community. This raises policy implications which will be discussed later.

  The West et al research indicated that specialist schools' GCSE performances have improved more than those of other schools, and a number of other benefits were reported by those involved with the schools. In addition, studies by Jesson (2001) for the Technology Colleges Trust (which exists to develop specialist schooling) using value added methodology indicated that schools specialising in technology and languages added more "value" in terms of helping pupils make progress towards GCSE than did non-specialist schools. Those specialising in arts or sport did less well: they produced value added GCSE scores almost identical with those of non-specialist schools. These findings clearly strengthened the government's confidence in pressing ahead with extending the programme. However, as both reports acknowledge, there could be a variety of reasons for the superior performance of some of these schools. The bidding process may identify improving schools that would have made these improvements in any case, and the additional resources which inclusion in the programme brings are very likely to have a positive influence on performance. Further, such studies are of limited value as a guide to national policy unless they cover not just these schools' own performances but also how the schools have affected the performances of other schools in their localities. The Jesson research has also been heavily criticised on technical grounds by a respected expert in school performance analysis who maintained that this purported evidence for the success of specialist schools "does not stand up to close examination" (Goldstein, 2002).

  Further research using more sophisticated value-added methodology (Schagen et al, 2002) indicated that specialist schools produced only a slight peformance advantage over non-specialists and this advantage was attributable entirely to two of the four existing forms of specialism, technology and languages. This study also provided some tentative evidence that specialist schools might be succeeding at the expense of neighbouring non-specialist schools. It also reported that LEAs with a high proportion of specialist schools (20% or over) did not perform as well as those with a low proportion. "There was thus no evidence to support the suggestion that an increase in the number of specialist schools would yield improvements in overall performance results" (ibid, p. 45). The finding that (for whatever reason) specialist schools performed only slightly better than non-specialists was supported by a government statistical study. It observed that "Differences in average progress were small compared to the spread of outcomes for pupils with similar prior attainment" (National Statistics, 2002, p. 33).

  The Schagen et al (2002) study was also one of the very few to examine the performance of faith-based schools. Church of England schools were found overall to perform marginally better, but Roman Catholic schools no better or worse, than non-religious schools. (However the very small number of Jewish schools had significantly better results than either Christian or non-religious schools). The authors concluded that they had not found any clear evidence to support the view that, if these schools created a specially supportive and well-ordered environment, it provided a climate that led to high achievement.

3.  DIVERSITY AND CHOICE: A NEW DIRECTION IN SCHOOLING, OR A BUTTRESSING OF THE OLD?

  One of the government's key principles in 1997, as set out in its White Paper Excellence in Schools, was "The focus will be on standards, not structures" (DfEE, 1997, p. 5). By 2002 it appears they have discovered the attractions of significant structural change (Morris, 2002). A major question is whether their present policy stance in the area of choice and diversity turns out to be a radical and visionary approach or a reinforcement of old and deep-rooted divisions. The government has sought to combine a major extension in diversity with an equivalent growth in collaborative practice between schools, even though the central characteristics of, and incentives relating to, the competitive system are still in place and "greater consumer choice" still features in policy statements (Morris, 2001, p. 7). In addition, there is a strongly articulated objective of enhancing equality of opportunity and also a new focus on reducing the "achievement gap" (DfES, 2001). It will be interesting to see whether diversity, collaboration and equality can all be significantly enhanced or whether the inevitable tensions between these distinct objectives will result in one or two of them becoming dominant. The research on specialist schools discussed earlier indicates that competition and partnership can make uneasy bedfellows.

  Will the new diversity be built on a competitive or a genuinely pluralistic model? As the 1994 OECD report stated: "Unlike some other nationalities, the English are used to the concept that routes to academic success may lie in centres of academic excellence rather than comprehensive neighbourhood schools" (OECD, 1994, p. 64) and that this familiarity derived from both the "public" and the grammar school traditions. Like the City Technology Colleges (CTCs) and grant-maintained (GM) schools before them, specialist schools experienced a significant increase in their popularity following designation, and the majority of headteachers in the West et al study (2000) attributed this, at least in part, to their new status.

  The policy for specialist and other new types of school was originally presented as "modernising" or "overhauling" the comprehensive system. However, its many critics, conscious of the specific social and cultural context of English secondary education, regard it as signifying the death of that system and its replacement by a two-tier structure of "winning" and "losing" schools and communities. More recently reference has been made (for example in Tony Blair's speech to the 2002 Labour Party conference) to a "post-comprehensive era" but retaining the comprehensive principle of equality of opportunity.

  The prospects for achieving a pluralistic rather than an hierarchic/competitive form of diversity seem to depend on at least two key factors. First, in terms of supply, the models so far developed are relatively limited in number and fall far short of the possible range (see the typology of school diversity proposed in Glatter et al, 1997, p. 8). The policy is heavily dependent on the specialist school model (curricular diversity in terms of the typology) and this dependence has been accentuated by the recent government reticence over faith schools. For example the new "Academies"—publicly-funded independent schools sponsored by private and voluntary bodies and established in areas of disadvantage—are also required to have curricular specialisation. A more creative approach to developing contrasting types of school would be needed, and this would imply a greater willingness to relax central controls particularly in the area of performance regulation.

  With regard to demand, there is a critical issue concerning the relationship between diversity and choice. The two terms have now been linked in policy discourse for a decade, since the Conservative government's 1992 White Paper, Choice and Diversity: a new framework for schools (DfE, 1992): the specialist schools policy has been explicitly presented in terms of enhancing choice (Blair, 2002). However the precise connection between them is very little understood and despite the significant quantity of research on choice and the quasi-market little attention has been given to this particular topic. The limited empirical evidence available suggests that, apart from preferences among relatively small proportions of parents for specific forms of religious education or for single-sex schooling there is no widespread demand for school diversity (see for example Woods et al, 1998). Parents generally appear to be looking for a school which will deliver the "standard product" best, whether or not it carries a "badge" of distinctiveness. Of course this could change once a range of school types is made available, but Walford's judgement of some years ago that the (then) government's diversity policy ". . . has been largely generated by the government itself, and has not been the result of pressure from parents" (Walford, 1996, p. 145) still holds true.

  A pluralistic approach to diversity would require "a relatively even spread of choices" (OECD, 1994, p. 42) so as to avoid the situation where some schools widely seen as the `best' are heavily over-subscribed and there is a "concentration of the most disadvantaged pupils in the least popular schools" (DfEE, 2001, p. 87). Such a spread of choices would be more likely to happen "if parents have diverse `frames of reference' placing different values on aspects of educational attainments" (Adnett and Davies, 2002, p. 202). Historical and cultural factors militate against such a development in the English context (Edwards and Whitty, 1997).

  An important set of issues centre on availability and illustrate the intimate connection between supply and demand. A striking small-scale research conducted in the early 1990s in a single medium-sized English town demonstrated the logistical difficulties involved in increasing diversity (Brain and Klein, 1994). Parents were surveyed about their preferences among the more restricted range of school types available at that time, principally single sex/co-educational and church-linked/non-denominational. The authors calculated that almost twice as many secondary schools would need to be provided in the town in order to meet all the parents' preferences. They also pointed out that if the choice menu had been extended to cover different curricular specialisms and a wider range of faith-based options (as are now being offered) the logistical problem would have been considerably exacerbated.

  Parental perceptions of availability are also a significant factor. The PASCI study conducted large-scale parental surveys in three contrasting areas of England. In a semi-rural area, only around one in four parents thought they had a realistic choice between three or more schools: the figure went up to just over half in a medium-sized town. Even in a heavily urbanised area a substantial proportion of parents—varying between a third and a quarter across the three years that the survey was conducted—considered that their realistic choice was limited to one or two schools (Woods et al, 1998).

  This raises the possibility that increased diversity may reduce rather than enhance parents' perception of the extent of choice open to them. For example, in the case of specialist schools it is an open question whether a choice between a small number of schools emphasising particular subject specialisms will be perceived as a more or a less attractive menu of options than was available previously under a more generalist system. For some parents and pupils, where the latter's specific talents and strengths are already clearly evident by the age of 10, or who are attracted not so much by the particular subject specialism as by the sense of "special-ness" it confers, the offer may be welcome (assuming that they can gain admission). For others the particular mix of specialisms available may be perceived as unappealing or may provide an additional source of anxiety in appearing to require an early judgement about a child's aptitudes.

  The next phase of research should examine such issues by focusing directly on the relationship between choice and diversity. This discussion also indicates that neither choice nor diversity is an end in itself. They are both means intended to contribute to wider goals, such as enhancing parent and pupil satisfaction over school allocations and achieving a good fit between the school allocated, the child's educational and social needs and the family's preferences in an equitable manner. Whether and how policy and practice over school admissions contributes to such goals has rarely been examined and requires close attention.

REFERENCES

  Adnett, N and Davies, P (2002) Education as a positional good: implications for market-based reforms of state schooling, British Journal of Educational Studies, 50, 189-205.

  Blair, T (2002) The Courage of our Convictions: why reform of the public services is the route to social justice (London, The Fabian Society).

  Brain, J and Klein, R (1994) Parental Choice: myth or reality? Bath Social Policy Papers No. 21 (Bath, University of Bath, Centre for the Analysis of Social Policy).

  Branigan, T (2002) Top school's creationists preach value of biblical story over evolution, The Guardian, 9 March.

  DfE (Department for Education/Welsh Office) (1992) Choice and Diversity: a new framework for schools (London, HMSO)

  DfEE (1999) Excellence in Cities (London, Department for Education and Employment)

  DfEE (Department for Education and Employment) (2001) Schools Building on Success (London, The Stationery Office).

  DfES (Department for Education and Skills) (2001) Schools Achieving Success (London, The Stationery Office).

  DfES (2002a) Investment for Reform (London, Department for Education and Skills).

  DfES (2002b) "Clarke gives green light for all schools to become specialist", Press Notice 2002/0228, London, Department for Education and Skills.

  Edwards, T and Whitty, G (1997) Marketing quality: traditional and modern versions of educational excellence, in R Glatter, P A Woods and C Bagley (eds.), Choice and Diversity in Schooling: perspectives and prospects (London, Routledge).

  Flatley, J, Connolly, H, Higgins, V, Williams, J Coldron, J, Stephenson, K, Logie, A and Smith, N (2001) Parents' Experiences of the Processes of Choosing a Secondary School, Research Report RR278 (London, Department of Education and Skills).

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

  Glatter, R, Woods, P A and Bagley, C (1997) Diversity, differentiation and heirarchy: school choice and parental preferences, in R. Glatter, P A Woods and C Bagley (eds.) Choice and Diversity in Schooling: perspectives and prospects (London, Routledge).

  Gibson, A and Asthana, S (2000) What's in a number? Commentary on Gorard and Fitz's "Investigating the determinants of segregation between schools" Research Papers in Education, 15, pp. 133-154.

  Goldstein, H (2002) Selective versions of reality The Times Educational Supplement 25 January.

  Gorard, S, Taylor, C and Fitz, J (2002a) Markets in public policy: the case of the United Kingdom Education Reform Act 1988, International Studies in Sociology of Education, 12, pp. 23-41.

  Gorard, S, Taylor, C and Fitz, J (2002b) Does school choice lead to "spirals of decline"? Journal of Educational Policy, 17, 367-384.

  Jesson, D in association with Taylor, C (2001) Educational Outcomes and Value Added Analysis of Specialist Schools for the Year 2000 (London, Technology Colleges Trust).

  Levacic, R (2001) An Analysis of Competition and its Impact on Secondary School Examination Performance in England Occasional Paper No. 34 (New York, Columbia University, Teachers College, National Center for the Study of Privatization in Education)

  Morris, E (2001) Professionalism and Trust: the future of teachers and teaching, a speech by the Rt. Hon. Estelle Morris, Secretary of State for Education and Skills, to the Social Market Foundation, 12 November (London, Department for Education and Skills).

  Morris, E (2002) Social Market Foundation Speech 24 June (London, Department for Education and Skills)

  National Statistics (2002b) Statistics of Education: pupil progress in secondary schools by school type in England: 2001 Issue No. 05/02, June 2002 [available on the internet only at www.dfes.gov.uk/statistics: downloaded July 2002]

  Noden, P (2000) Rediscovering the impact of marketisation: dimensions of social segregation in England's secondary schools, 1994-99, British Journal of Sociology of Education, 21, 371-390.

  Noden, P, West, A, David, M and Edge, A (1998) Choice and destinations at transfer to secondary schools in London, Journal of Educational Policy, 13, pp. 221-236.

  OECD (1994) Schooli: a Matter of Choice (Paris, Organisation for Economic Co-operation and Development, Centre for Educational Research and Innovation).

  Ofsted (2001) Specialist schools: an evaluation of progress (London, Office for Standards in Education).

  Schagen, S, Davies, D, Rudd, R and Schagen, I (2002) The Impact of Specialist and Faith Schools on Performance (Slough, National Foundation for Educational Research)

  Tomlinson, S (2001) Education in a Post-Welfare Society (Buckingham, Open University Press).

  Walford, G (1996) Diversity and choice in school education: an alternative view, Oxford Review of Education, 22, 143-154.

  West, A, Noden, P, Kleinman, M and Whitehead, C (2000) Examining the Impact of the Specialist Schools Programme, Research Report RR196 (London, Department for Education and Employment).

  West, A and Pennell, H (2002) How New is New Labour? The quasi-market and English schools 1997-2001, British Journal of Educational Studies, 50, pp. 206-224.

  Whitty, G, Power, S and Halpin, D (1998) Devolution and Choice in Education (Buckingham, Open University Press).

  Woods, P A, Bagley, C and Glatter, R (1998) School Choice and Competition: markets in the public interest? (London, Routledge).

  Yeomans, D, Higham, J and Sharp, P (2000) The Impact of the Specialist Schools Programme: Case Studies, Research Report RR197 (London, Department for Education and Employment).