Disproportionality in Special Needs Education in England

By Dyson, Alan Gallannaugh, Frances

Unlike the United States, England does not have a special education system based on the identification of students as having disabilities of one or another type. Instead, the English system enables help to be provided to students on the basis of assessments of their individual “special educational needs.” The authors consider the implications of this position for the disproportional presence of students from different social groups in the special needs system. They argue that disproportionality is a reality in England, as in the United States, though it cannot be understood simply in relation to racial minorities. Nor, within a non- disability-based system, does it arise principally from the misidentification of students as having disabilities. Instead, it reflects broad educational and social inequalities. Disproportionality research, therefore, needs to concern itself with these inequalities. Keywords: disproportionality; special education; England; ethnicity; class; gender

The issue of disproportionality has attracted considerable attention in recent years in the United States (see, e.g., Artiles, 1998; Donovan & Cross, 2002; Harry & Klinger, 2006; Losen & Orfield, 2002; Skiba et al., 2003). In general terms, this attention has focused on what we might call the “misidentification” of children from different racial groups. The patterns are complex (Parrish, 2002), but there is strong evidence that the proportions of different groups identified as having particular disabilities are different from the proportions in which such groups appear in the school population as a whole. Both over- and underrepresentation in this sense raise questions about the equitability of the special education system, but there is particular concern that minority students are overrepresented in at least some disability categories. Oswald, Coutinho, and Best (2002) offered two broad hypotheses for this phenomenon. One is that minority students are particularly susceptible to disability, not least because of the disadvantaged social and educational conditions many of them experience. The other, in their words, is that “a significant portion of the overrepresentation problem may be a function of inappropriate interpretation of ethnic and cultural differences as disabilities” (p. 2).

A similar concern with the consequences of misidentifying difference as disability led Harry and Klinger (2006) to recommend that

at least for children with high-incidence learning and behaviour difficulties, special education should be reconceptualized as a set of services that are available to children who need them, without the need for a disability label. The current conceptualization of special education as requiring a search for intrinsic disability has not succeeded in adequately serving the needs of students who are performing at the low end of the general-education spectrum. (p. 175)

It is at this point that the experience of the special education system in England (see Note 1) may prove illuminating. In the years following the 1944 Education Act, England had a special education system that, like that of the United States, required children to be identified as having disabilities before they were eligible for services (for a history, see Department of Education and Science, 1978, chap. 2). A medically dominated process of “ascertainment” allocated children with significant difficulties in schooling to 1 of 11 “categories of handicap.” Some of these categories were normative, in the sense that they had clear biological bases and more or less objective diagnostic criteria could be applied to them, but others were nonnormative because their biological bases were uncertain and/or the criteria were less robust. Not surprisingly, in light of current experience in the United States, children from different social and ethnic groups found themselves disproportionately placed in these categories (Croll & Moses, 1985; Sacker, Schoon, & Bartley 2001; Tomlinson, 1982, 1985).

However, in 1978, the report of a committee of inquiry into special education, chaired by Mary (later, Dame Mary) Warnock, was published, and its recommendations continue to form the basis of the special education system in England. The “Warnock report” (Department of Education and Science, 1978), as it is known, proposed precisely the move that Harry and Klinger (2006) advocated: the provision of special education services without the prior need to allocate a child to a specific, or indeed any, disability category. Instead, the report argued, the concept of “special educational needs” (Gulliford, 1971) should be adopted and should be seen in terms not simply of disability but of “all the factors which have a bearing on [the child’s] educational progress” (Department of Education and Science, 1978, [para] 3.6). In consequence, special needs services in England are provided to children who experience any “learning difficulty that calls for special educational provision to be made for them” (Education Act of 1996, [section] 312, as cited in Department for Education and Skills, 2001), regardless of whether that difficulty arises from disability or from some other cause.

This definition, though notoriously circular, shifts attention from medical to educational questions, specifically, what can be done to help a student who is struggling in general education to learn more effectively. There are no tightly defined criteria for deciding who receives services and no prespecification of types of services to meet categories of need. Assessment, therefore, is an educationally focused process aimed not at allocating a child to a disability category but at producing a rounded analysis of the child’s learning characteristics, of the situation in which he or she is expected to learn, and of the modifications, additional support, or alternative provision that might be made. Indeed, assessment and provision in the English system are individualized in a very particular way. It is not simply that children are assessed one by one but that it is their individual characteristics and circumstances that are assessed and that provision is customized to their individual needs and situations.

If Harry and Klinger (2006) were right, such a system should go at least some way toward tackling the issue of disproportionality. Minority students cannot be overidentified as having disabilities if the identification of disability is no longer required. Instead, what we should see is the flexible delivery of services to any and all children who need them. Indeed, it is only fair to point out that this is, to a significant extent, what we do see in the English system. However, this is only part of the story. In the remainder of this article, we argue that disproportionality, though frequently hidden from view and not entirely identical to that in the United States, remains a significant feature of the English system. We bring together some of the scattered evidence about this phenomenon and about its relationship to issues of social and economic marginalization. We will suggest that the problem in England is not the misidentification of minority students as having disabilities but the misleading identification of them as having special educational needs as individuals when the difficulties they experience are systemic and structural in origin. We argue, therefore, that a proper understanding and response to disproportionality in England needs to take into account these systemic and structural factors.

Disproportionality in England

The issue of which social groups find themselves in the special needs system has received some attention in England, though this has been less than in the United States and has not been so tightly focused on issues of race. There has been no comprehensive, national study of disproportionality in all its forms, and only recently have data become available to make possible a national study of ethnic disproportionality (Lindsay, Father, & Strand, 2006). However, there is a range of somewhat scattered studies of the relationship between ethnicity and disability (see, e.g., Ahmed, Darr, Jones, & Nisar, 1998), of the social characteristics of the special needs population (notably Croll & Moses, 1985, 2000), and of issues around particular types of special educational needs and/or particular social groups (see, e.g., Cooper, Upton, & Smith, 1991; Daniels, Hey, Leonard, & Smith, 1999; Derrington, 2005; Troyna & Siraj-Blatchford, 1993). There is also an awareness, high on the current political agenda, of the differential performance of different groups in the education system as a whole (see, e.g., Gillborn & Mirza, 2000).

By and large, these studies have focused on analyzing whatever statistical data have been available on the composition of school and special needs education populations to identify disproportionality. This task has become easier in recent years as the English education system has developed sophisticated and extensive statistical databases. These make it possible, among other things, to characterize the populations of children identified as having special educational needs in terms of a range of demographic variables, including ethnicity, gender, socioeconomic background, and types of need. Both Lindsay et al.’s (2006) and our own (Dyson, Farrell, Polat, Hutcheson, & Gallannaugh, 2004) are recent studies that have made use of these new databases. Nonetheless, there is no coherent, cumulative body of disproportionality research. In the following sections, therefore, we seek to draw together what is known about this issue from the diverse studies that have so far been undertaken. Minority Ethnic Groups

England is a country that has experienced successive waves of immigration, dating back over many centuries. In recent times, the major influxes have been from Ireland, beginning in the 19th century, from the “new commonwealth” (notably, the Indian subcontinent and the West Indies) in the second half of the 20th century, and, in recent years, from troubled regions of the world (such as Somalia) and from Central Europe and Eastern Europe. In consequence, although the large majority of the population identify themselves as White British (see Note 2), there are significant populations that identify themselves and are identified by others as constituting more or less distinct minorities. For the purposes of official statistics, these minorities are described in terms of “ethnicity” rather than “race.” In the 2001 census (reported in National Statistics, 2002), 7.6% of the United Kingdom’s population identified themselves as belonging to minority ethnic groups, with the largest groups being Indian (1.7%), Pakistani (1.3%), Black Caribbean (1.0%), Black African (0.9%), and those of mixed backgrounds (0.8%). However, minority ethnic groups were more likely to live in England, where they made up 9% of the population, and to be regionally concentrated. For instance, 48% of people with minority ethnic backgrounds lived in London, with further concentrations in the industrial towns and cities of the West Midlands, the North West, and Yorkshire.

It is important to bear in mind, however, that ethnic labels such as these have complex meanings in British society. For one thing, the labels that are used both officially and in common usage are based variously on place of origin (e.g., Chinese, Irish, Pakistani), some also refer to color and by implication race (e.g., Black Caribbean, White British), and some even refer to lifestyle (e.g., Traveller). They intersect and interact with other markers of identity in complex ways, so that, for instance, people claiming the same ethnicity may speak different languages, adhere to different religions, and see these as more important ways of defining themselves than ethnicity per se. Indeed, ethnicity in England is a matter of lived identity as much as, if not more than, ascribed grouping. As National Statistics (2002), responsible for analyzing much of the data on ethnicity, pointed out,

The subjective, multi-faceted and changing nature of ethnic identification makes it a particularly difficult piece of information to collect. There is no consensus on what constitutes an “ethnic group.” Membership of any ethnic group is something that is subjectively meaningful to the person concerned and the terminology used to describe ethnic group has changed markedly over time. (p. 4)

Moreover, ethnicity captures only crudely other characteristics, such as language, religion, place (rather than country) of origin, and place of current residence.

In this situation, it is dangerous to make simple binary distinctions between majority and minority groups, much less between White and Black or White and “People of Colour.” Nor is it safe to assume that all minority ethnic groups are socioeconomically disadvantaged, though some-notably the Bangladeshi, Pakistani, and Black groups-do indeed fare badly on a range of social indicators, such as unemployment, income, and health (National Statistics, 2003). This diversity is reflected in the educational performance of children from different ethnic groups and in their representation in the special educational needs system. Table 1 brings together a number of sources of data on the characteristics of children in school by ethnic group (see Note 3).

The first column shows the proportion of children in English schools from different ethnic groups. It confirms that White British children form the large majority in the school system, though the younger age structure of minority populations means that this majority is not as large as in the country as a whole. The second column shows the proportion of children in each ethnic group who are entitled to free school meals. This entitlement provides a crude but widely used indicator of low family income. It is immediately clear that, measured in this way, children from different ethnic groups experience very different levels of poverty. The level of entitlement for Traveller children of Irish heritage, for instance, is between 4 and 5 times that of White British children. Although caution must be exercised in interpreting data for this very small group, other, larger groups-Irish, Pakistani, Bangladeshi, and Black- also appear to experience high levels of poverty. On the other hand, children of Chinese and Indian heritage experience less poverty than their White British peers.

The third column of Table 1 presents a similarly crude but widely used measure of children’s attainment in the school system. At age 16 (in most cases), children are able to take a series of national examinations (the General Certificate of secondary Education examinations and their equivalents). Although there is no formal graduation level, it is the practice to regard high grades in five subjects as an “expected level” that many children should aim for. Again, it is clear that different ethnic groups have very different levels of success in this sense. The Chinese and Indian groups outperform the White British group to a considerable extent, whereas the Traveller and, to a lesser extent, Black Caribbean groups do much less well than their White British peers.

The next set of columns shows the proportion of each group identified as having special educational needs in general education primary (i.e., ages 5 to 11 years) and secondary (i.e., ages 11 to 16 years) schools (see Note 4). A distinction is made between those children who have “statements” and those who do not. In the English system, most children regarded as having special educational needs are identified in the first instance by their teachers in general education schools, who decide that they should be subject to “school action” (whereby the school uses its own resources to meet their needs) or “school action plus” (whereby the school supplements its own resources with those provided by outside specialists). Children whose needs cannot be met in this way (usually those with the most significant needs) may be put forward for a process of formal assessment leading to statements. These are documents with legal force that record the children’s needs and the arrangements to be made to meet those needs. They serve some of the functions of individual education plans in the United States, though detailed planning, target setting, and monitoring are confined to separate, nonlegal documents (confusingly also called individualized education plans). Children with statements need not be placed in special schools, and in practice, some 60% of them remain in general education (National Statistics, 2005).

It is clear from Table 1 that the proportions of children identified as having special educational needs at these various levels differ by ethnic group. Most dramatically, a far greater proportion of Travellers than of all other groups is regarded as having special educational needs both with and without statements. High levels of identification can be seen in other groups too, notably, Black Caribbean children. At the same time, overrepresentation is not an issue with respect to all minority groups. Chinese and Indian children, for instance, experience relatively low levels of identification across both phases of schooling, whether with or without statements.

It is noticeable in these figures that the pattern for children with statements does not correspond precisely to that for students without statements. In general terms, the groups that are most overrepresented in the population without statements are less so in the population with statements (the same holds good for children who are placed in special schools as opposed to those who are educated in general education schools; Department for Education and Skills, 2005b). This may be because of the different procedures for identifying children at different levels. Identification at lower levels is largely a matter of teacher judgment. At the level of a statement, it requires a formal multidisciplinary assessment with significant family input (see Department for Education and Skills, 2001, for a description of the processes involved). It may be that this more formal process serves both to moderate the effects of teacher judgment and to place an onus on the differential capacity of families, and of members of different ethnic groups (Ahmed et al., 1998; Diniz, 1999), to make effective representations.

Although the English system no longer allocates children to categories of handicap to make provision, it has recently (in 2004) begun to collect data on the broad types of special education needs children are regarded as having. Again, there are interesting differences by ethnic group, though the patterns are complex (Department for Education and Skills, 2005b). The clearest picture emerges from analyses carried out by Lindsay et al. (2006). They calculated the odds of minority ethnic group members’ being identified as having special educational needs of a range of types (in this case, at the higher levels of school action plus and with a statement) compared with those of White British children (see Note 5). Their analyses revealed the patterns of ethnic disproportionality in England to be complex and nuanced and deserve a fuller account than is possible in this article. All that is possible here is to present some indicative data. So, Table 2 presents data for a sample of the ethnic groups and types of special educational need analyzed by Lindsay et al. for which patterns of representation are strikingly different. We focus here on special educational needs as a whole and on three types of need: moderate learning difficulties (MLD); behavioral, emotional, and social difficulties (BESD); and visual impairment (VI; see Department for Education and Skills, 2005a, for definitions of each type). These types contrast in interesting ways. VI is a “normative” disability in which professional judgment is mediated by more or less objective diagnostic criteria and achievement in school is a relatively minor factor. The other two types lack such clear criteria, and identification is dependent more on professional judgment. In the case of MLD, achievement is also likely to come into play, whereas BESD are, by definition, about children’s social functioning, not least in relation to their teachers. The odds are presented in an unadjusted form (i.e., by simply comparing relative rates of identification between groups) and as they have been adjusted by Lindsay et al. for differences between groups in terms of gender, age, and socioeconomic disadvantage. It is clear that the patterns of identification are different for each of these types of need. In the case of the normative VI category, there are no significant differences between the minority groups and their White British comparators. The exception is in the case of children of Pakistani children, for whom susceptibility to such impairment may possibly be heightened by a relatively high proportion of consanguineous marriages in this community (Lindsay et al., 2006). However, there are no such factors to explain the differential odds ratios with respect to MLD and BESD. The odds for Traveller children here, as with identification in general, are dramatically high. Even leaving aside this relatively small group, however, we see how Indian and Chinese children are relatively underrepresented in both types, whereas Black Caribbean children are overrepresented. These odds change between the two types of need. Groups that are less likely than White British children to be identified as having MLD are represented even less in the BESD type, whereas the Black Caribbean group is represented even more. The odds shrink when they are adjusted for age, gender, and poverty, but the pattern of differences remains. The odds for special needs as a whole broadly reflect those for MLD and BESD, which may not be surprising given that these are the two largest types of need and that the other large types-speech language and communication needs and specific learning difficulties-are equally nonnormative (National Statistics, 2005).

Ethnicity and Other Factors

Although we have been concerned so far with the identification of special educational needs in relation to ethnicity alone, it seems clear from Table 1 that other factors associated with ethnicity are also at work. In particular, those ethnic groups that experience most poverty and/or achieve least well also seem to be most likely to be identified as having special educational needs. The differences between Lindsay et al.’s (2006) adjusted and unadjusted odds ratios in Table 2 confirm this impression. By and large, when background factors are taken into account, the odds of ethnic minority children being identified as having special educational needs decrease relative to those of White British children. Put another way, those other factors increase the chances of children being identified as having special educational needs, and some of the differences in levels of identification between ethnic groups are explicable in terms of those factors.

As part of their analysis, Lindsay et al. (2006) portioned out the contribution of a series of factors to the likelihood of an individual’s being identified as having special educational needs. They found that ethnicity explained only 0.5% of the variance once other factors are taken into account (Lindsay et al., 2006, pp. 45ff). Because, then, ethnicity is a rather weak predictor of the likelihood of identification at individual level, other background factors deserve examination in their own right. It is to this task that we now turn.

Other Forms of Disproportionality

Table 3 brings together data from recent government sources (National Statistics, 2005, 2006), which show how factors other than ethnicity are related to the identification of children as having special educational needs. One of these factors is age or, more accurately, school year. As an illustration, the first two rows show the proportion of children in general education schools by their special educational needs status at two points in time: Year 1 (the start of primary school) and Year 7 (the 1st year of secondary school). They indicate that proportionately more older children are identified as having special educational needs than younger children. This is particularly true for children with statements. In fact, the pattern is slightly more complex than this would suggest. Increasing proportions of children are identified through the primary and early secondary years, but after that point, the rate begins to decline again (National Statistics, 2005). The fact remains, however, that rates of identification vary with school year group, and the special needs education system consequently has a “bulge” of children in the late primary and early secondary years.

The second pair of columns examines the relationship between poverty (as indicated by entitlement to free schools meals) and identification for children in general education schools. Children living in poverty are approximately twice as likely as their more affluent peers to be identified as having special educational needs, whether with or without statements. Similarly, the third pair of columns indicates that gender is related to identification: Boys in general education schools are approximately twice as likely as girls (a little more with statements, a little less without statements) to be identified as having special educational needs.

Nor are they the only other forms of disproportionality. Our own analyses suggest that month of birth is also a factor (Dyson et al., 2004). Children in England usually progress through their schooling in a year cohort, so some may be nearly a full year younger than others in their classes. Among 5- to 7-year-olds, the youngest children in the year cohort are nearly twice as likely to be identified as the oldest (30.9% to 16.1%), and this effect holds (albeit in diminished form) throughout the school years. Similarly, children whose behavior is troublesome to their teachers are particularly likely to be identified as having special educational needs (Galloway, Armstrong, & Tomlinson 1994; Rees, Farrell, & Rees, 2003). There may be some interactions with poverty here, as Croll (2002) noted that “discipline problems” identified by schools correlate highly with poor home background. There may also be interactions with ethnicity (Gillborn & Gipps, 1996; Tennant, 2004), with certain minority ethnic groups particularly likely to find themselves excluded from school for disciplinary reasons (Department for Education and Skills, 2005b).

There is also a kind of “historical” disproportionality, in the sense that the proportions of children and types of need identified change over time in ways that may not be related to changes in children’s characteristics. Croll and Moses (2000, 2003), for instance, found an overall 38.8% increase in the incidence of identified needs between 1981 and 1998 with, as we have seen, different patterns in groups identified as having different types of special educational need. More recently, a survey of local authority personnel by the Audit Commission (2002) reported significant perceived increases in the number of children with autistic spectrum disorders, speech and communication difficulties, and profound and multiple learning difficulties, together with perceived decreases in MLD and specific learning difficulties. The same survey found significant variations in the proportions of children with statements between local authorities (as did Lindsay et al., 2006) and between schools.

Disproportionality in England: Toward an Explanation

It is clear that the special needs education system in England is characterized by various forms of disproportionality. These may be different in some ways from those in the U.S. systems, but they are just as marked. As we saw, in the U.S. context, Oswald et al. (2002) proposed that disproportionality may be explained by real differences in the incidence of disability (perhaps related to disadvantage) between social groups and/or by the “inappropriate interpretation” of ethnic and cultural difference as disability. Certainly, there is some evidence from England to support both of these hypotheses. For instance, the high incidence of VI among Pakistani-origin children may well have a genetic origin, while the overrepresentation of children from poor families may point to a relationship between poverty and increased risks of disability. Likewise, the overrepresentation of boys, children living in poverty, and children from certain ethnic groups may indicate that there are distinct (albeit complex) gender, class, and ethnic cultures and that these generate behaviors in school that are interpreted by professionals as indicating disability. Supporting evidence here might be the differences between rates of identification when teacher judgment operates freely and when it is moderated by formal assessment, the residual effects of ethnicity, even when other factors are taken into account, and the variations over time in the proportions of children identified and in the types of special educational needs they are regarded as having. Explanations based on notions of misidentification, however, have to be modified to take account of the fact that children do not have to be identified as having disabilities to fall within the purview of the English special needs education system. As we have seen, all that is necessary is that a child is regarded as having a “difficulty in learning” such that additional or different provision might prove helpful. To some extent, the modification that is required is minor. Instead of focusing the explanation on the details of identification procedures, it is possible to see such procedures as specific examples of what Minow called “the ways in which institutions construct and utilize difference to justify and enforce exclusions” (as quoted in Artiles, 2003, p. 194).

Indeed, there is a substantial body of critical scholarship in England that examines how particular student characteristics are constructed negatively by schools, how such constructions lead to reduced impoverished experiences and reduced opportunities, and how students become alienated by schools they feel reject them (see, e.g., Benjamin, Nind, Hall, Collins, & Sheehy, 2003; Booth, 2003; Corbett, 1996; Corbett & Slee, 2000; Cremin & Thomas, 2005, among many others). Much of this literature argues that there is, as Thomas and Loxley (2001) suggested, an “imperative to seek homogeneity in institutional life and [a] corresponding imperative to delineate and differentiate those who differ from the norm” (p. 87).

Because the norm, arguably, is set in terms of those groups that schools find it easiest to deal with (Armstrong, 2005), it is no surprise that other groups-those from nondominant cultures or whose behavior challenges their teachers, for instance-are disproportionately likely to be identified as deviant and, specifically, as “having special educational needs.” From this perspective, the disability-based identification procedures in the United States and the more open processes of individual assessment in England simply offer alternative media through which common processes can operate. This, of course, casts doubt on suggestions, such as those made by Harry and Klinger (2006), that changes in the identification procedures of special education might be enough in themselves to counter disproportionality. On the contrary, only changes in the deeper-lying processes of construction are likely to have any real impact in this regard.

However, the origins of disproportionality may lie even deeper than this construction of difference explanation allows. We say this because the patterns of disproportionality do not reflect simply the arbitrary constructions of teachers or the institutional imperatives of schools but also underlying patterns of social and educational inequality. For instance, disproportionality seems to be related to underlying differences in the achievements of groups of children. Children with low achievement are much more likely to be identified than are their higher achieving peers. At age 16, 63.3% of students not identified as having special educational needs achieve the expected levels in examinations, but only 17% of those identified without statements and 7.1% of those with statements (National Statistics, 2006). Because there is no presumption about the presence of disabilities that cause this low achievement, it seems that schools and teachers see low achievement itself as an important criterion for identification.

Indeed, other forms of disproportionality, which at first sight seem likely to embody teacher constructions of differences, appear in fact to be mediated by achievement. This is true of socioeconomic background (Sacker et al., 2001); ethnicity (Bhattacharyya, Ison, & Blair, 2003; Department for Education and Skills, 2005b); gender, because boys are lower achievers than girls in the English system (Gillborn & Mirza, 2000); and of our own puzzling finding about the overrepresentation of young-for-cohort children, given that such children are lower achievers than their older peers (Dyson et al., 2004). Similarly, disproportionality seems to reflect deeper social divisions and inequalities. Ethnic, gender, and socioeconomic groups not only experience different educational and special educational outcomes, but in an unequal society, they also experience different social outcomes in terms of health, employment, income, and so on (see, e.g., Blanden, Gregg, & Machin, 2005; National Statistics, 2003, 2004).

The implication is that although the identification of children as having special educational needs may result most immediately from the construction of difference at the school and teacher levels, that construction is itself a response to educational and social inequalities. It follows that a proper understanding of disproportionality, capable of generating effective means of combating it, requires an analysis not only of processes of construction but also of the underlying processes and structures through which social and educational inequality are produced. Viewed from this perspective, the current identification procedures in special needs education, even if they do not lead to widespread misidentification, are nonetheless profoundly misleading. This is because of the high level of individualization that we noted earlier. Practitioners are encouraged (Department for Education and Skills, 2001) to identify the particular problems of each individual child and to put together individual packages of provision to overcome those difficulties. In this way, the extent to which these individual problems may be linked to educational and social outcomes for whole groups, and so to social and educational inequalities, is disguised. Likewise, the need for interventions at the level of schools’ practice, the organization of the education system, or the structuring of society is concealed by the focus on what can be done for this child now. Following Wright Mills (1959), we can see how, in this way, the English system reconstructs the “public issues” of the failures of schooling as the “private troubles” of individual children.

To some extent, special education research in England, particularly as it has refocused itself around issues in inclusion, has avoided this trap. Although it has paid relatively little attention to disproportionality, its focus on the construction of difference in schools has in fact served to surface some of the school-level practices and ways of thinking out of which disproportionality can arise. Most obviously, the Index for Inclusion (Booth & Ainscow, 2002), distributed by the government to every school in England, spawning many local variants, and widely adopted in other countries, takes the form of

a set of materials designed to support the inclusive development of the cultures, policies and practices of schools. It is about making schools reflect and be responsive to all aspects of diversity. It is thus concerned with reducing all forms of personal and institutional discrimination. (Booth, 2003, p. 11)

It is a moot point, however, whether this focus on processes of construction at the school and teacher levels engages adequately with the sorts of social and educational inequalities we have outlined above (Dyson & Gallannaugh, 2007). In some ways this is puzzling, given that many of the early critical analyses in England arose out of a “radical structuralist view” (Tomlinson, 1995), which sought to understand special education in terms of the deep-lying social structures and processes out of which it was produced. Not surprisingly, these early analyses took some interest in the differential positions of different ethnic and social groups in special education (see, e.g., Barton, 1986; Tomlinson 1982, 1989).

It may be, therefore, that a renewed interest in disproportionality, stimulated by the availability of the new national data sets, might now serve to reconnect inclusion researchers in England with these issues. It is our contention that just as practice needs to connect the particular difficulties of individual students with wider patterns of group disadvantage and social inequality, so researchers need to find a way of reconnecting their interest in teacher- and school-level processes of construction with wider social processes. As Artiles (2003) argued, all of us who take a critical interest in special education now need to “transcend the traditional individualistic perspective and infuse a social justice dimension so that the improvement of educational experiences and life opportunities for historically marginalized students are of central importance” (pp. 194-195).


1. In the United Kingdom, education is a devolved responsibility of the administrations of its constituent parts: England, Northern Ireland, Scotland, and Wales. Although the education systems within these administrations are similar, there are important differences, and we confine ourselves here, therefore, to England. Some of the census and other population data on which we draw relates to the United Kingdom as a whole or to Great Britain (the United Kingdom excluding Northern Ireland), but the patterns hold good for England, which accounts for over 80% of the United Kingdom’s population. By the same token, we acknowledge that our references to “the U.S. system” disguise the significant differences between education systems in the constituent states.

2. The ethnic groupings used in this article are those used in national statistics. They rely on people’s self-identification and constrain that identification into a limited number of categories, which inevitably are arbitrary to some extent. see National Statistics (2001) for a discussion of some of the issues raised by these categorizations.

3. Here and elsewhere in this article, we present tables that are derived from multiple sources. Although these tables are useful for seeing at a glance some of the key forms of disproportionality in the English system, caution should of course be exercised in interpreting data collated in this way. Readers should note that Table 1 has been simplified by omitting data on some of the “mixed” and “other” groups where there is no single ethnicity or where small ethnic groups have been combined for statistical purposes. Readers are also advised to pursue the original sources for fuller analyses. 4. Data on the small proportion of children (just over 1%) placed in special schools are reported separately in official statistics, though the pattern of ethnicity is similar to that in general education schools (see Department for Education and Skills, 2005b).

5. Lindsay et al. explained that these odds ratios “tell us how much more (or how much less) likely an outcome is for one group relative to a comparator group. For ethnicity, these ratios contrast the odds for each ethnic group relative to the White British majority group” (p. 23).


Ahmed, W., Darr, A., Jones, L., & Nisar, G. (1998). Deafness and ethnicity: Services, policy and politics. Bristol, UK: Policy Press.

Armstrong, D. (2005). Reinventing “inclusion”: New Labour and the cultural politics of special education. Oxford Review of Education, 31, 119-134.

Artiles, A. J. (1998). The dilemma of difference: Enriching the disproportionality discourse with theory and context. Journal of Special Education, 32, 32-36.

Artiles, A. J. (2003). Special education’s changing identity: Paradoxes and dilemmas in views of culture and space. Harvard Educational Review, 73, 164-202.

Audit Commission. (2002). Special educational needs: A mainstream issue. London: Author.

Barton, L. (1986). The politics of special educational needs. Disability, Handicap and Society, 1, 273-289.

Benjamin, S., Nind, M., Hall, K., Collins, J., & Sheehy, K. (2003). Moments of inclusion and exclusion: Pupils negotiating classroom contexts. British Journal of Sociology of Education, 24, 547-558.

Bhattacharyya, G., Ison, L., & Blair, M. (2003). Minority ethnic attainment and participation in education and training: The evidence (Research Topic Paper RTP 01-03). London: Department for Education and Skills.

Blanden, J., Gregg, P., & Machin, S. (2005). Intergenerational mobility in Europe and North America. London: Centre for Economic Performance, London School of Economics for the Sutton Trust.

Booth, T. (2003). Inclusion and exclusion in the city: Concepts and contexts. In P. Potts (Ed.), Inclusion in the city: Selection, schooling and community (pp. 1-14). London: Routledge.

Booth, T., & Ainscow, M. (2002). Index for inclusion: Developing learning and participation in schools. Bristol, UK: Centre for Studies on Inclusive Education.

Cooper, P., Upton, G., & Smith, C. (1991). Ethnic minority and gender distribution among staff and pupils in facilities for pupils with emotional and behavioural difficulties in England and Wales. British Journal of Sociology of Education, 12, 77-94.

Corbett, J. (1996). Bad-mouthing: The language of special needs. London: Palmer.

Corbett, J., & Slee, R. (2000). An international conversation on inclusive education. In F. Armstrong, D. Armstrong, & L. Barton (Eds.), Inclusive education: Policy contexts and comparative perspectives (pp. 133-146). London: David Fulton.

Cremin, H., & Thomas, G. (2005). Maintaining underclasses via contrastive judgement: Can inclusive education ever happen? British Journal of Educational Studies, 53, 431^446.

Croll, P. (2002). Social deprivation, school-level achievement and special educational needs. Educational Research, 44, 43-53.

Croll, P., & Moses, D. (1985). One in five-The assessment and incidence of special educational needs. London: Routledge Kegan Paul.

Croll, P., & Moses, D. (2000). Special needs in the primary school: One in five? London: Cassell.

Croll, P., & Moses, D. (2003). Special educational needs across two decades: Survey evidence from English primary schools. British Educational Research Journal, 29, 731-747.

Daniels, H., Hey, V., Leonard, D., & Smith, M. (1999). Issues of equity in special needs education from a gender perspective. British Journal of Special Education, 26, 184-194.

Derrington, C. (2005). Perceptions of behaviour and patterns of exclusion: Gypsy Traveller students in English secondary schools. Journal of Research in Special Educational Needs, 5, 55-61.

Department of Education and Science. (1978). Special educational needs: Report of the committee of enquiry into the education of handicapped children and young people (the Wamock report). London: Her Majesty’s Stationery Office.

Department for Education and Skills. (2001 ). Special educational needs code of practice. London: Author.

Department for Education and Skills. (2005a). Data collection by type of special educational need. London: Author.

Department for Education and Skills. (2005b). Ethnicity and education: The evidence on minority ethnic pupils (Research Topic Paper RTP01-05). London: Author.

Diniz, F. A. (1999). Race and special educational needs in the 1990s. British Journal of Special Education, 26, 213-217.

Donovan, M. S., & Cross, C. T. (Eds.). (2002). Minority students in special and gifted education. Washington, DC: National Academy Press.

Dyson, A., Farrell, P., Polat, R, Hutcheson, G., & Gallannaugh, F. (2004). Inclusion and pupil achievement (Research Report RR578). London: Department for Education and Skills.

Dyson, A., & Gallannaugh, F. (2007). National policy and the development of inclusive school practices: A case study. Cambridge Journal of Education, 37(4), 473-488.

Galloway, D., Armstrong, D., & Tomlinson, S. (1994). Special educational needs: Whose problem? London: Longman.

Gillborn, D., & Gipps, C. (1996). Recent research on the achievements of ethnic minority pupils. London: Her Majesty’s Stationery Office.

Gillborn, D., & Mirza, H. S. (2000). Educational inequality: Mapping race, class and gender. London: Office for Standards in Education.

Gulliford, R. (1971). Special educational needs. London: Routledge Kegan Paul.

Harry, B., & Klinger, J. (2006). Why are so many minority students in special education? Understanding race and disability in schools. New York: Teachers College Press.

Lindsay, G., Pather, S., & Strand, S. (2006). Special educational needs and ethnicity: Issues of over- and wider-representation (Research Report RR757). London: Department for Education and Skills.

Losen, D. J., & Orfield, G. (Eds.). (2002). Racial inequality in special education. Cambridge, MA: Harvard Education Press.

National Statistics. (2001). The classification of ethnic groups. Retrieved April 18, 2006, from http://www.statistics.gov.uk/ about/ Classifications/ns_ethnic_classification.asp

National Statistics. (2002). Social focus in brief: Ethnicity 2002. Retrieved April 18, 2006, from http://www.statistics.gov.uk/ downloads/theme_social/social_focus_in_brief/ethnicity/ ethnicity.pdf

National Statistics. (2003). Census 2001-Ethnicity and religion in England and Wales. Retrieved April 16, 2006, from http:// www.statistics.gov.uk/census2001/profiles/commentaries/ ethnicity.asp

National Statistics. (2004). Focus on gender. Retrieved April 17, 2006, from http://www.statistics.gov.uk/downloads/theme_ compendia/ fog2004/Gender.pdf

National Statistics. (2005). Special educational needs in England, January 2005 (SFR 24/2005). Retrieved April 17, 2006, from http://www.dfes.gov.uk/rsgateway/DB/SFR/s000584/SFR242005.pdf

National Statistics. (2006). National curriculum assessment, GCSE and equivalent attainment and post-16 attainment by pupil characteristics, in England 2005. Retrieved August 17, 2006, from http://www.dfes.gov.uk/rsgateway/DB/SFR/s000640/ index.shtml

Oswald, D. P., Coutinho, M., & Best, A. M. (2002). Community and school predictors of overrepresentation of minority children in special education. In D. J. Losen & G. Orfield (Eds.), Racial inequality in special education (pp. 1-13). Cambridge, MA: Harvard Education Press.

Parrish, T. (2002). Racial disparities in the identification, funding and provision of special education. In D. J. Losen & G. Orfield (Eds.), Racial inequality in special education (pp. 15-37). Cambridge, MA: Harvard Education Press.

Rees, C., Farrell, P., & Rees, P. (2003). Coping with complexity: How do educational psychologists assess students with emotional and behavioural difficulties? Educational Psychology in Practice, 19, 35- 47.

Sacker, A., Schoon, I., & Hartley, M. (2001). Sources of bias in special needs provision in mainstream primary schools: Evidence from two British cohort studies. European Journal of Special Needs Education, 16, 259-276.

Skiba, R. J., Simmons, A. B., Ritter, S., Kohler, K., Henderson, M., & Wu, T. (2003). The context of minority dispmportionality: Local perspectives on special education referral. Bloomington: Indiana Education Policy Center, Indiana University.

Tennant, G. (2004). Differential classroom interactions by ethnicity. Emotional and Behavioural Difficulties, 9, 191-204.

Thomas, G., & Loxley, A. (2001). Deconstructing special education and constructing inclusion. Buckingham, UK: Open University Press.

Tomlinson, S. (1982). A sociology of special education. London: Routledge Kegan Paul.

Tomlinson, S. (1985). The expansion of special education. Oxford Review of Education, 11, 157-165.

Tomlinson, S. (1989). Asian pupils and special issues. British Journal of Special Education, 16, 119-122.

Tomlinson, S. (1995). The radical structuralist view of special education and disability : Unpopular perspectives on their origins and development. In T. M. Skrtic (Ed.), Disability and democracy: Reconstructing (special) education for postmodemity (pp. 122-134). New York: Teachers College Press.

Troyna, B., & Siraj-Blatchford, I. (1993). Providing support or denying access? The experiences of students designated as “ESL” and “SN” in a multi-ethnic secondary school. Educational Review, 45, 3- 11. Wright Mills, C. (1959). The sociological imagination. Oxford, UK: Oxford University Press.

Alan Dyson

Frances Gallannaugh

University of Manchester

Alan Dyson is a professor of education at the University of Manchester, where he also codirects the Centre for Equity in Education. His research interests are in the field of urban education and inclusive education. He is a coauthor (with MeI Ainscow and Tony Booth) of Improving Schools, Developing Inclusion (Routledge, 2006).

Frances Gallannaugh is a research associate in the Centre for Equity in Education at the University of Manchester. Her research is in the field of inclusive and urban education. She is a coauthor (with Mairead Dunne, Sara Humphreys, Judy Sebba, Alan Dyson, and Daniel Muijs) of Effective Teaching and Learning for Pupils in Low Attaining Groups (Department for Children, Schools and Families, 2007).

Copyright PRO-ED Journals May 2008

(c) 2008 Journal of Special Education, The. Provided by ProQuest Information and Learning. All rights Reserved.