What Counts as Evidence in the Understanding of School Exclusion in England?

Daniels, Harry; Porter, Jill; Thompson, Ian

doi:10.3389/feduc.2022.929912

CONCEPTUAL ANALYSIS article

Front. Educ., 16 June 2022

Sec. Special Educational Needs

Volume 7 - 2022 | https://doi.org/10.3389/feduc.2022.929912

This article is part of the Research Topic The role of evidence in developing effective educational inclusion View all 11 articles

What Counts as Evidence in the Understanding of School Exclusion in England?

$\r\nHarry Daniels$ Harry Daniels^*

Jill Porter $Ian Thompson\r\n$ Ian Thompson

Department of Education, University of Oxford, Oxford, United Kingdom

Exclusion from school can be regarded as a seemingly simple but in fact a rather complex intervention in response to the “wicked problem” of behavior in schools. This manuscript will discuss what counts as evidence that may used to inform policy and judgments on practices of exclusion. The role of evidence, and how this is measured, has long been an issue of contention in educational research. This is particularly true for research that focuses on educational inequality and inclusion or exclusion. In this manuscript we will discuss issues concerning evidence with respect to two aspects of exclusion in England. Firstly, we will focus on questions concerning the scale of the problem, examining both the statistical evidence of official exclusions and data concerning the myriad of ways in which children may experience other forms of exclusion. Taken together, this indicates an under-estimate of the numbers of young people missing an education. We then move to a consideration of the evaluation of means of reducing exclusion, arguing for a shift from an individual to a systemic in context account that recognizes the role of cultural transmission and cultural historical theory.

Introduction

An essential pre-requisite of inclusion is presence or access to education. Indeed, for many developing countries enabling every child to go to school is a shared international goal. UNESCO (2020) describes inclusion as “non-negotiable.” While we would argue that presence on its own does not equate to inclusion, it does provide the possibility or potential for different futures. Conversely, school exclusion is and always has been a consequence of disadvantage and it gives rise to inequalities both social and economic (Daniels and Cole, 2010; Riddell and McCluskey, 2012; Power and Taylor, 2013; Scottish Government, 2017; Thompson, 2017). It is cause for concern worldwide given that exclusion can exacerbate social fragmentation and even give rise to conflict (UNESCO, 2018). O’Donovan et al. (2015) discuss attempts in Australia to reduce exclusion as do Ainscow et al. (2013) more generally. Detailed scrutiny reveals that the political and policy frameworks in a nation state shape the forms which exclusion takes. This is shown clearly by Zancajo (2019) in the case of Chile, Leung et al. (2021) with respect to the exclusion of young mainland Chinese students in Hong Kong Schools, Muderedzi and Ingstad (2011) in Zimbabwe and Bademci et al. (2016) in Turkey. In this manuscript we discuss evidence about exclusion in the English context.

Moll (2000) has shown how children and young people acquire situated competences (skills, values, and knowledge) from what he terms the funds of the community of which they are part through the everyday activities which occur in the particular cultural, social, and historical contexts in which these communities are located. Exclusion from school, in its myriad of different forms, disrupts this process. Our argument aligns with Sen’s (1992) capability and rights based approach to the broader concept of social exclusion. The suggestion is that intervention should ensure that all citizens have equal access to the benefits of participation in society (Sen, 1992). This approach to work on social exclusion, is important in that it points to the dangers of the denial of human rights that is associated with restrictions on participation in society at large. It is also important because it does not suggest that one size fits all – it is not an argument for uniformity and sameness rather it points to the celebration of diversity. This understanding of diversity rests on the assumption that different individuals will make use of opportunities in different ways.

Exclusion from school may be regarded as both a prior and outcome of social exclusion. Young people who are excluded often come from disadvantaging circumstances. Exclusion from school frequently has long term negative consequences in terms of gaining access to the benefits of a position in the mainstream of society. An understanding of the extent of the problem is important in bringing to the attention of public and policy makers a key feature in the processes of social exclusion and marginalization. Exclusion from school can be also regarded as a seemingly simple but, in fact, a rather complex intervention in response to the “wicked problem” of behavior in schools (Armstrong, 2018).

This manuscript will discuss what counts as evidence that may be used to inform policy and judgments on practices of exclusion. The role of evidence, and how this is measured, has long been an issue of contention in educational research. This is particularly true for research that focuses on educational inequality (e.g., Luke et al., 2010) and inclusion or exclusion (e.g., Florian, 2014; Kauffman and Hornby, 2020). We will argue that a practice as important as exclusion from school has proved remarkably difficult to document and calibrate, not least because of difficulties in defining exclusion in a manner that reflects what is happening in schools. There is a need to go beyond a simple interrogation of the validity and reliability of the data. We argue that making sense of “evidence” can only be done with reference to policy, at national and local level. Rather than seeking to examine the evidence base to policy we are contending that the policy, in turn, impacts on the quality of the evidence. This can be viewed as a reversal, or at least a prior, to the government rhetoric of drawing on a range of evidence to inform policy. Evidence can only be evaluated within context, often within competing agendas and perverse incentives. As Slee (2019) argues “policy represents values, choice making and authorization (909).” Absence of policy leads to absence of data and absence of data leads to a lack of accountability systems of the effectiveness of policy. In the context of exclusion, children can simply go missing with their presence, or indeed absence, and their futures unmonitored. The numbers of children missing from registered schooling raises important questions about the notion of a “mainstream” education and its fit for purpose. Here we identify gaps in evidence and reflect on their relationship to policy and the implications for intervention and its evaluation.

Evidence Gaps That Persist Over Time

Over 20 years ago, Vulliamy and Webb (2001) suggested that data on exclusions can be recontextualized and even distorted by schools. In their research they identified differences between the actual and recorded reason(s) for exclusion. In particular they note considerable underestimations of official exclusion figures compared to the number of pupils actually excluded from school. Many of their concerns appear to have persisted over the intervening years and reveal the unintended consequences of policies that have contributed to the unreliability of the data.

The reduction in recorded permanent exclusions in England during 1997–2000 was explained by a number of headteachers resorting to “grey” exclusions in an attempt to avoid financial penalties and to meet ambitious national targets (Daniels and Cole, 2010). These included “managed transfers,” which often do not appear on orthodox means of keeping a record of events (Munn et al., 2000; Osler et al., 2001). A problem of “unofficial” exclusions pointed out by Vulliamy and Webb (2001) is that schools wishing to preserve their position in public accountability systems persuade families to move their child to another school and thus avoid a possible formal exclusion. Then as now this can result in unpopular schools being faced with a high level of demand to make provision available for young people who have become unacceptable in other settings. Also, the Vulliamy and Webb (2001) and Local Government Association (2020) have argued that there is a pressing need for research that provides insight into the relationship between fixed-term and permanent exclusions over time. Whether exclusion is official in that it is recorded and published in national statistics, or it is unofficial, unlawful or not noticed, it can result in a young person missing out on schooling. As Barber (1997) identified in another context, the image that emerges is one where “missing” includes the disappeared who remove themselves from education; the disaffected who for lack of positive educational progression easily move into being amongst the disappeared; and those who are dispossessed because the system has not kept them on its horizon (Barber, 1997, pp. 426–429).

The number of permanent school exclusions (PEX) have risen sharply in England since 2014 prior to the COVID-19 pandemic in stark contrast to the other United Kingdom jurisdictions which under devolution have separate education systems (Cole et al., 2019). The formal data reveals that the likelihood of being excluded is associated with unmet special educational need (SEN) or disability (UNESCO, 2018); as well as psychosocial and mental health difficulties; being of Black-Caribbean heritage; from a low socio-economic background; and being male (Department for Education [DFE], 2020; Strand and Fletcher, 2014). There are no reliable data on those who go missing from education or those subjected to various forms of exclusion within a school or are illegally excluded (Ofsted, 2013; Gazeley et al., 2015; Power and Taylor, 2020). In addition to high levels of PEX and fixed period exclusions (FEX), there has of late been considerable concerns expressed about “hidden” forms of exclusion, including “managed moves,” “off-rolling,” and “internal” exclusions (IPPR, 2017; Education Select Committee, 2018; Ofsted, 2018).

While some of the causes of school exclusion are well-documented, very little is known about the system-level factors that lie behind the numbers overall. The influence of cultural and historical influences is witnessed in wide variations in official levels of exclusion reported (pre-COVID) across the United Kingdom. Example in 2018–2019 7,894 PEX in England (10 per 10,000), 165 PEX in Wales (4 per 10,000), 33 PEX in Northern Ireland (NI) (1 per 10,000), and only three in Scotland (0 per 10,000).

There are four areas in relation to exclusion which have raised a number of concerns in recent years: Persistent absence, Elective Home Education (EHE), young people who miss out on education in a variety of ways that remain almost invisible, and illegal schools. All four are characterized by a weakness in the evidence that is available for public scrutiny and to policy makers.

Persistent Absence

While there will be a variety of reasons for school non-attendance, this group will include self excluders, including young people who are experiencing difficulties with school work, relationships with others, who are bored, bullied, or have mental health difficulties (Reid, 2012; Thompson et al., 2022). Non-attendance can lead to a self perpetuating cycle as the child falls behind, and relationships with peers weaken. It is perhaps unsurprising that the rate in England is highest amongst young people with SEND (Ofsted, 2021) where over a quarter of children with an Education Health and Care plan fall into the category of persistent absentee children and almost 1 in 5 children who are categorized as SEND support (Ofsted, 2021). This compares to mainstream rates of 11%.

These rates need to be set within policy decisions that frame the measurement. Persistent absence has been a particular concern in recent years in England, with official definitions of measurement changing in 2009/2010 from those who miss 20% or more of school sessions, to a threshold of 15% or more to “ensure that schools take action sooner,” and meaning that some 4,30,000 children now met the bar an increase of 2,46,000 young people (Department for Education [DFE], 2011). Measurement also shifted from 5 half terms to six from 2012/13. Subsequently, the bar was further raised in 2015/16 to those who are absent 10% or more of the time (Department for Education [DFE], 2019). Thus, while persistent absence has been a particular concern, the methodology has been changed to emphasize this.

There are important limitations to the data, and therefore what it can tell us about access to education. Data is based on presence at twice daily registrations, therefore internal absences, and those who disappear off the premises after registration are not recorded, nor are children who miss particular lessons. Special school data are only collected annually, a highly relevant anomaly during the COVID pandemic. Data are only collected for schools that are registered.

As with much official data, it requires a detailed interrogation of the dataset to go beyond headline accounts. For example, while there is reported data for increased persistent absence and SEND, intersectional data requires access to the whole dataset, and is not readily available for public scrutiny. This is well illustrated by the research of Moyse (2021) who, with data provided through a freedom of information request, reveals increasing numbers of girls on the autistic spectrum identified as persistent absentees at state funded secondary schools in England. The girls also now form a much larger proportion of absent autistic pupils. Being able to examine the evidence in detail enables further research to look more closely at the experiences of particular groups who are in effect self-excluding from school. With respect to this identified group further research revealed how persistent absence was associated with unmet needs and a failure to provide appropriate support. In particular their withdrawal reflected their experience of a school environment and ethos that was damaging their mental health, rather than due to a lack of motivation or interest in learning.

There has been a tendency to pathologize the individuals who are persistent absentees (and their families) with policies of parental prosecution (Department for Education [DFE], 2022) that sit uncomfortably alongside failures to find suitable school placements or provision to meet their need (Ofsted, 2020). Official advice focuses on parental responsibilities in finding “ways to improve your child’s attendance” rather than looking at school based factors, including the impact of cultural aspects of the school, school climate and categorization, feeling safe, relationships with staff and peers (Thompson et al., 2022). In the face of prosecution for their child’s persistent absence, parents may opt for elective home education.

Elective Home Education

The Association of Directors of Children’s Services (ADCS, 2021) survey reported in 2021 that 94,258 young people were thought to be supposedly receiving Elective Home Education (EHE) in 2020/21 across 124 Local Authorities (LA) in England responding to their nationwide survey. They estimated that 1,15,542 children and young people were being home educated across all 152 LAs in England during the previous academic year. This provides evidence of a rise of about 34% from the 2019/20 academic year (ADCS, 2021). Somewhat worryingly they stated that LAs are very concerned about the lack of power they have to ensure that all EHE children are safeguarded and receiving a suitable and meaningful education. They also reported a number of significant concerns including: the absence of reliable comprehensive data on the children concerned; the safety and appropriateness of the schooling environment, safeguarding children against harm or exploitation; the impact of EHE on obtaining qualifications; and the credentials and qualifications of some of the tutors being employed. It was also suggested that given that only 7% of local authorities are at all certain that they have knowledge of all the EHE children and that the data available underestimates the true extent of the phenomenon (The Office of the Children’s Commissioner [OCC], 2019).

Whilst the Office of the Children’s Commissioner claimed that the evidence suggests a significant increase in children being home educated they cautioned that there cannot be complete certainty on the numbers due to the lack of any formal registration. This absence of regulation marks England as an outlier in comparison with other European states in which EHE is legal (The Office of the Children’s Commissioner [OCC], 2019). Seabrook et al. (2021) report from one of the Athenaeum Club’s Topical Discussion Groups which identified a pressing need for a robust supportive and protective framework for Elective Home Education. Overall this widespread activity proceeds in the absence of evidence of relevance and its consequences for young people. There are signs that attempts will be made to rectify this situation (Foster, 2019). Savage (2021) reports that the chief inspector of Ofsted (the government education inspection body in England), “Amanda Spielman, has warned that ministers have “no handle” on who the missing children are or where they are. She said school absences had led to significant fall in the numbers of referrals to social care, potentially putting more children at risk of abuse. It was further stated that Spielman has called for an official register of children who are not attending school (Savage, 2021).”

Illegal Schools

Some parents have made claims that they are home educating their children, when in fact they are sending them to unregistered and illegal schools (or “tuition centers”). In these situations they may be offered a poor education with worrying standards of welfare and hygiene. Illegal schools operate beyond the gaze of the legal frameworks that have been put in place to safeguard children. This settings lack definition or even identification and function in a way that is not open to any form of inspection. Since setting up a specialist taskforce in 2016, Ofsted has identified 439 schools which are possibly operating illegally (The Office of the Children’s Commissioner [OCC], 2019).

Missing Children

Nationally, there is a distinct lack of any detailed, reliable data concerning the extent to which children are missing out on extended periods of formal, full-time education (Parish et al., 2020). Estimates vary considerably from Feuchtwang (2018) who suggests that nearly 50,000 children in England have been missing from education to Parish et al. (2020) who state that their best estimate developed in their Local Government Association study is that in 2018/19, more than a quarter of a million children in England may have missed out on education. This equates to around 2% of the school age population. There is also considerable variation in the operational definition of missing education. This, in turn, conditions the nature and extent of the data that are gathered.

The statutory guidance Children Missing Education (Department for Education [DFE], 2016) sets out the key principles to enable local authorities in England to implement their legal duty under section 436A of the Education Act 1996 to make arrangements to identify, as far as it is possible to do so, children missing education. It states:

“All children, regardless of their circumstances, are entitled to an efficient, full time education which is suitable to their age, ability, aptitude and any special educational needs they may have. Children missing education are children of compulsory school age who are not registered pupils at a school and are not receiving suitable education otherwise than at a school (Department for Education [DFE], 2016, p. 5).”

Interestingly, ofsted uses a broader definition of pupils missing from education in their inspections, which includes those on a school roll but on unsuitable part-time timetables or unlawfully excluded (Ofsted, 2016).

Data on children missing education are not collected in a systematic way at national level. This means that there are no reliable figures for the whole of England (National Children’s Bureau, 2014). This is a cause for concern given that careers, schools, and local authorities all have responsibilities in preventing children missing education, which is set out in national guidance and procedures (Department for Education [DFE], 2016).

In their report for the Local Government Association Parish et al. (2020) employ the following definition: “any child of statutory school age who is missing out on a formal, full-time education.” By “formal,” they mean an education that is well-structured, contains significant taught input, pursues learning goals that are appropriate to a child or young person’s age and ability and which supports them to access their next stage in education, learning or employment. By full-time they mean an education for at least 18 h per week. They conclude that:

“Children missing education can be found in a variety of both formal and informal education settings, they can be found at home receiving different forms of educational input or none at all, they can be found in employment and they can be simply unknown to those providing services in the community. This complexity helps to explain why the numbers of children missing out on their entitlement to education might be routinely underestimated and why it has historically been a challenge to construct legislation and guidance that ensures that no children miss out on the education which is their right, by law (Parish et al., 2020, p. 18).”

The differences in the numbers arrived at by their three approaches to estimation attest to the lack of precision in definition a general weakness of the evidence on prevalence (see Figure 1).

FIGURE 1

Figure 1. LGA estimates of children missing from education.

Done and Knowler (2020a,b) have recently discussed “offrolling”- a practice which involves young people being removed from schools in ways which are variously referred to as unofficial, unlawful, or illegal and which appears to be prevalent in both England and Australia (Done et al., 2021). The evidence on offrolling is not strong. Ofsted (2019b) raised concerns about 5,800 pupils with special educational needs and disabilities (SEND) who they found had left school between Years 10 and 11 (January 2017 to January 2018), however, they state that a significant proportion “may have been off rolled” (2019b, p. 53). When all Year 10 pupils were considered, it was found that 19,000 had left school during this period and 9,700 of these remained unaccounted for (2019b, p. 50). The report speculates that schools may have been “gaming” the accountability systems (Ofsted, 2019a: 50).

It is almost unsurprising to note that the Education Policy Institute [EPI], 2019 demonstrated that the pupils most likely to be off rolled were: those who had previously undergone an official permanent exclusion (1 in 3) or fixed term exclusion (1 in 5); pupils in contact with the social care system (1 in 3); those with a high number of authorized absences (approximately 2 in 5 of whom in the 2017 cohort had experienced at least one unexplained exit); pupils eligible for free school meals (1 in 7); those from black ethnic backgrounds (1 in 8); and those in the lowest prior attainment quartile (1 in 8). The disadvantaged and the marginalized were most likely to be further distanced from education through processes which remain invisible.

So what can be said about what counts as evidence as to the scale of the problem of exclusion? If the definition of exclusion is restricted to official permanent exclusion and suspension then the Statistical First Release data released by the DfE provide a strong evidence base, with particular groups over-represented. However, if the definition of exclusion is widened to include the wide range of unofficial practices that lead to young people missing out on education then the evidence base is weak, and its access limits public scrutiny of the intersection between groups of individuals who may share common unfulfilled needs. The true extent of the latter remains uncertain although the evidence that is available suggests that the scale of this problem is worryingly large. In consequence, these limitations impact on the availability of measures to evaluate the effectiveness of strategies to prevent exclusion.

What Counts as Evidence on the Prevention of Exclusion

What counts as evidence in the prevention of school exclusion is contested. Qualitative research designs are often critiqued on the grounds of narrow contextualization and lack of generalizability. For example, in a DfE commissioned independent literature review produced as part of the evidence base for the United Kingdom government’s Timpson Review of School Exclusions (Timpson, 2019) the authors note that:

“Much of the literature focusing on preventative initiatives and approaches is based on qualitative evidence, which is limited in terms of its applicability beyond the circumstances in which the study was carried out and the purposive nature of the sample design. As a consequence, the evidence on the impact of these initiatives is limited (Graham et al., 2019, p. 43).”

In addition to the issue of research design is the question of how we evaluate the impact of an intervention. Valdebenito et al.’s (2018) Campbell Collaboration Systematic Review entitled “School-Based Interventions for Reducing Disciplinary School Exclusion” reports only on studies that have used randomized controlled trials (RCTs). RCTs are considered by many to be the best methodological design for isolating confounding factors and producing an accurate measure of intervention effects. In this case the measure of impact is the use of school exclusionary sanctions, as formally reported by the school. Given the earlier discussion about the limitations to the reporting of these data this raises questions about the accuracy and narrowness of the measures. Amongst the boundaries set around the inclusion of studies is the exclusion of special schools and studies where the intervention was specific to special needs. Of the 37 studies in the review, carried out across nine countries, the majority (73%) focused on changing some aspect of the pupils’ skills or behavior, example, social skill training, anger management. The remaining 27% reported intervention changing some aspect of the school or teacher.

The Valdebenito et al. (2018) report evidences a limited number of short term weak effects:

“The analyses reported in previous chapters suggest that school-based interventions are capable of producing a small and significant (SMD = 0.30; 95% CI, 0.20–0.41; p < 0.001) drop in exclusion rates. It means that those participating in interventions are less likely to be excluded than those allocated to control/placebo groups. These results are based on measures of impact collected on average, 6 months after treatment. When the impact was tested in the long-term (i.e., 12 or more months after treatment), the effect of interventions was not sustained. In fact, the impact of school-based programmes showed a substantial reduction (50%), and was no longer statistically significant (Valdebenito et al., 2018, p. 84).”

Even then the authors of the report lament the quality of the small number of publications that they drew on in that they are lacking a considerable amount of information for judging the quality of the procedures carried out (Valdebenito et al., 2018).

Goldacre (2013) is one of many advocates of RCT designs as a means of providing “gold standard” evidence about what works in education. This advice drives a lot the activity funded by the Education Endowment Fund (EEF) which seeks to generate new evidence of “what works” to improve teaching and learning. The EEF has been awarded significant and sustained endowment funding by the Department for Education since its foundation in 2011 and it has tended to favor RCTS when funding educational research.

Tensions between advocates of small scale, in depth qualitative research and large scale quantitative experimental designs have been witnessed in what have become almost tribal skirmishes in the battle for evidence.

“Evidence matters in the ongoing struggle for more equitable and just education. But there is no direct link between “fact” and norm, between science and policy. To address questions of equity requires rich, interpretive, and evolving sciences, not a narrow technical approach that invites capture by particular doctrinal and generic approaches to systems reform, public policy, and institutional governance (Luke et al., 2010, p. xv).”

These debates have taken a slightly different turn in recent years. Cartwright and Hardie (2012) have highlighted some of the common misuses and abuses of RCT methods in social and medical sciences. In particular they suggest that the importance of context is often ignored and this can lead to the simplistic adoption of a “what works” approach to policy making. A central concern in this series of arguments is that RCTs may provide evidence about what worked “there and then” but will not necessarily provide evidence about what will work “here and now.” A similar argument is promoted by advocates of realist RCTs. Realist researchers focus not merely on what works, but on what works for whom and under what conditions:

“Randomized trials of complex public health interventions generally aim to identify what works, accrediting specific intervention “products” as effective. This approach often fails to give sufficient consideration to how intervention components interact with each other and with local context. “Realists” argue that trials misunderstand the scientific method, offer only a “successionist” approach to causation, which brackets out the complexity of social causation, and fail to ask which interventions work, for whom and under what circumstances (Bonell et al., 2012, p. 2299).”

Marchal et al. (2013) question the assumptions made by Bonell et al. that research from a realist paradigm can adapt RCT research designs that come from a mentally positivist position.

These are issues which confront researchers across the social sciences and medicine. Skivington et al. (2021) report on the replacement of the United Kingdom Medical Research Council’s widely used guidance for developing and evaluating complex interventions with a new framework, which takes into consideration recent developments in research design and methods and the need to maximize the efficiency, use, and impact of research. Skivington et al. (2021) argue that there is a need to step beyond questions of efficacy and effectiveness by employing a broader range and combination of research perspectives and methods. Among their plea for new forms of questioning are:

“Will this effective intervention reproduce the effects found in the trial when implemented here? and how are the intervention effects mediated by different settings and contexts? (Skivington et al., 2021, p. 3)”

The mixing of methods and methodologies raises a tension between what is taken as fact and how it is aligned with the meanings associated with a supposed fact and the status accorded to each perspective.

Cowen et al. (2017) make a related point arguing that as ’the same policy or intervention will have different effects in different populations that have different support factors or different distributions of support factors, it matters in new settings which support factors are present and in what proportions (p. 269).” From this perspective one size cannot possibly fit all.

Joyce and Cartwright (2018) suggest the need for a reconsideration of the division of labor between practitioners and researchers in the production of evidence. They acknowledge that practitioners are well positioned to identify support factors or the distribution of support factors. They bring insights into the ways in which interventions worked in particular situations and to identify causal components. They argue that:

“Although educators within local contexts are in the best epistemic position to secure evidence for some of these premises, researchers can help. However, they further suggest that researchers can consider which aspects of the arrangements in study settings and features of individuals affect the outcome (Joyce and Cartwright, 2018, p. 17).”

Also, they can identify intermediate steps observed during the study that indicate success. Taken together these two perspectives enhance the possibilities for reliable predictions. A primary need is that researchers “must be explicit in describing the local contexts in which interventions are studied” (Cruz et al., 2021, p. 421). Here lies a major challenge for the development of this field. The move from crude models of context to more sophisticated accounts of the ways in which context is actively made rather than surrounds interventions requires considerable theoretical and methodological development (Cole, 1996). Recourse to developments in the sociology of pedagogy, cultural transmission and cultural historical theory may prove fruitful.

Conclusion

An examination of the role of evidence in developing effective inclusion must incorporate evidence on exclusion, a school practice that violates students’ fundamental rights and, in turn, impacts negatively on their futures. We draw attention here to the longstanding difficulties of calibrating the extent of exclusionary practices. Absence of data results in a lack of accountability systems that serve to safeguard students. In addition to the formal systems of school suspension and expulsion are a raft of practices that indicate that the scale of exclusion is much larger. These include managed moves, off-rolling, internal exclusions, and children who effectively have part-time provision, either through restricted timetables or being sent home early. Evidence on the scale of these practices can best be described as fragile. Data reviewed here indicates that in England some 2% of the population are missing education, not receiving their education in a registered school. Taking both informal and formal systems together indicates a wide-spread failure of schools to provide for the diversity of students.

The relationship between policy and data collection is further witnessed in responses that target the individual rather than the system and demonstrates the pathologizing of students and their families. The formal collection of data indicate that particular groups of students are disproportionately represented but access to the data set prohibits a full public understanding of the intersection between these groups. This in turn limits our understanding of their experiences in school and the ways in which we can effectively remove barriers to participation. Reviews of large scale studies indicate the restricted impact of interventions that target the individual. We have also shown how constrained are the methods that are used to provide evidence for policy makers on the implementation of interventions. In part this reflects a policy reliance on decontextualized data (such as statistical data and RCTs) or the idea that one size fits all. Decontextualized data can lead to policies that ignore the context. Using flawed formal data to measure impact only serves to further marginalize students. New methodologies are needed that shift the evidence base from an individual to a systemic in context account, that recognizes the role of cultural transmission and cultural historical theory. To close this manuscript we return to the text we referred to at the start.

Given the complexity of both teaching and educational innovation, any neglect of research into the processes of change in naturalistic settings will not only lead to a restricted awareness of a project’s impact but also to a failure to understand what certain apparent outcomes actually mean. Vulliamy and Webb (2001) p. 368.

Exclusion from school should be understood as being a multifaceted cultural and historical phenomenon or process and a complex intervention. To reiterate Sen (1992) different individuals will make use of different opportunities in different ways. Evidence on understanding school exclusion, and what might work to prevent exclusion, needs to be framed in equally complex ways that attend to both contextual and societal practice and their cultural historical origins.

Author Contributions

HD produced the first draft which was subsequently developed by JP and IT. All authors contributed to the article and approved the submitted version.

Funding

The article submitted was written with funding from an ESRC Large Grant - Reference: ES/S015744/1.

Conflict of Interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Publisher’s Note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

References

ADCS (2021). Elective Home Education Survey 2021. London: The Association of Directors of Children’s Services Ltd.