A multi-stakeholder analysis of the risks to early school leaving: comparing young peoples’ and educators’ perspectives on five categories of risk

ABSTRACT This paper considers the risks to Early School Leaving (ESL) as perceived by young people who are Early Leavers or at-risk of Early Leaving, and their educators in one region in South England. It draws on 38 interviews and focuses groups with 39 young people and 53 adults working in various roles across 11 educational settings including mainstream and specialist schooling, alternative learning provision and vocational education and training, as well as the local authority. Guided by a holistic framework to examine risks to ESL [Brown et al. (2021. “A Conceptual Framework for Researching the Risks to Early Leaving.” Journal of Education and Work 24 (7–8): 723–739. doi:10.1080/13639080.2021.2003007)], the paper analyses findings according to five different categories of risk: personal challenges; familial circumstances; social relationships; institutional features of school/work and structural factors of policy, economic disadvantage and the educational system. The findings highlight that while educators focused upon structural factors and personal challenges, young people themselves identified social relationships and institutional features of school as the most significant influencers on ESL risk. The contribution of this paper is two-fold; firstly, in presenting a comprehensive analysis of the multiple risk factors to Early Leaving; and secondly, in reflecting on the difference in emphases between young people’s perspectives of the key constituents of risk and those of the educators who support them.


Introduction: early leaving or 'NEEThood' as a UK policy concern
Addressing the failure to make a successful transition into the labour market has been a key issue on the policy agenda in England over the last decade, following concerns about the high numbers of young people who are Not in Employment, Education or Training (NEET).The key policy mechanism through which national levels of NEET young people have been tackled in recent years is the statutory legislation, Raising of the Participation Age of young people in England from 16-18 (DfE 2012).This legislation requires young people in England to participate in education or training until their 18th birthday through full-time study either in school, college or with a training provider (DfE 2016, 16).The obligation to meet this statutory duty has been devolved to local government by allocating them a number of duties including ensuring that there is sufficient education and training provision to meet demand (DfE 2016, 5), providing support for all young people aged 13-19 and those aged 20-25 with Special Educational Needs and Disabilities (SEND) to participate in education and training (DfE 2016, 6); and to provide the tracking mechanisms to report on participation and NEET status so that they can be identified for support.To date, such measures have not led to any meaningful reduction in ESL with recent data showing the number of 16-and 17-year-olds identified as NEET has risen to 4.5% (despite some fluctuations) since the introduction of this legislation, compared with only 4.2% in 2013. 1  Accompanying this has been a concerted policy effort to raise the status of technical education (DfSB 2016), including a new directive enabling young people to choose between an academic or technical route at age 16; the latter having been streamlined to provide a common framework of 15 routes across all technical education (DfSB 2016, 8).These training pathways require young people to choose one of two routes; apprenticeship programmes or a two-year college-based programme aligned to apprenticeships.From September 2020 the newly introduced Vocational Education and Training (VET) pathways comprise T-Level (Technical Level) courses, which are equivalent to the 3-A-levels (Academic Level) (DfE 2019a).Certification from either technical route is nationally recognised at levels 2 and 3. Underpinned by the Technical and Further Education Act 2017, post-18 degree-level apprenticeships have been introduced, with 12 new Institutes of Technology (IoTs) (DfE 2019b) available across England offering higher technical STEM education and training at levels 4 and 5, and a further £120 million to fund another eight IoTs (DfE 2020a).
Concurrent policy measures have aimed to make VET a more appealing option for young people, including through low-interest student loans for Further Education (Johnson 2020a), £1.5 billion government investment in refurbishment for college buildings (Johnson 2020b) and a pledge from the former Prime Minister that: We will also offer an opportunity guarantee, so that every young person has the chance of an apprenticeship or an in-work placement so that they maintain the skills and confidence they need to find the job that is right for them.(Johnson 2020b) Government investment in VET reflects an implicit assumption that NEET is largely the consequence of inadequate and insufficient educational options for those not seeking an academic route.However, this fails to recognise the issue of ESL from a life course perspective (e.g.Alexander, Entwisle, and Kabbani 2001) where factors leading to school drop-out are seen not as singular events, but rather as a 'long process' (Ferguson et al. 2005) and in response to young people's cumulative negative experience throughout their schooling career (Casillas et al. 2012;Lamb et al. 2010).Indeed, reported perceptions of youth training among disadvantaged young people from marginalised groups highlight their continued disengagement and deep dissatisfaction with these programmes, which have been referred to as little more than a 'holding pen' during their post-school years (Shildrick and MacDonald 2007).Furthermore, with Level 2 Maths/English made a condition of funding to raise the status of VET (ESFA 2014), it is questionable how accessible it is for those for whom drop-out is the consequence of low academic attainment; identified as the most significant predictor of ESL (Battin-Pearson et al. 2000).We argue that, by putting all their eggs in one basket, policymakers may be missing the complexity of ESL, which several systematic reviews have identified as an issue that is best understood as complex and multi-faceted (González-Rodríguez, Vieira, and Vidal 2019;Haugan, Frostad, and Mjaavatn 2019) and the consequence of multiple, interrelated problems (Dorn 1996) affecting an increasingly heterogeneous group of young people (Jarjoura 1996).Despite the body of work in recent years elucidating the key predictors of ESL, there has been little disaggregation between those risk factors identified from the perspective of adult educators (e.g.teachers, social workers, school/college leaders, project workers, support staff, policymakers and parents) and those raised by young people who are the object of such interventions, such as Early Leavers and those at risk of ESL.
In aiming to address this gap, this paper reports on the UK findings of a wider European study on Early School Leaving (ESL) in Europe, drawing on interviews and focus groups with 39 young people and 53 adult education professionals who supported them.Thirty of these young people were identified by the adults who supported them, as being 'at risk of becoming NEET,' while nine others were already NEET as they were outside of education and employment at the time of data collection.The following discussion reviews the literature on ESL to elicit the key themes raised in stakeholder accounts and to highlight the different angles on such issues raised by adult stakeholders vis-à-vis young people at risk of, or affected by, ESL.
While US policy tends to focus on school drop-out, ESL is primarily a European term, broadly understood to refer to when young people leave the educational system without achieving formal qualifications (DeWitte et al. 2013;González-Rodríguez, Vieira, and Vidal 2019), or leave before the legal age limit (Kritikos and Ching 2005;Kim et al. 2015).The definition is complicated by the fact that UK policymakers do not recognise the term Early School Leaving, referring instead to the status Not in Employment Education or Training (NEET) to which central and local governments are accountable.Various problems with the use of the term NEET have been highlighted; the term having different meanings 'in everyday speech than in published research' (Holte 2017), the inflexibility of the concept to capture the flux in young adults' movements in and out of education and employment (MacDonald 2011) and in being an 'empty signifier' which lumps together too broad a range of vulnerabilities and characteristics to be useful in understanding and targeting social exclusion (Kleif 2020).Notwithstanding its limitations as a conceptual schema, in their varying roles, educators within this study had an assigned responsibility for reducing NEEThood and did not recognise the term Early School Leaving.Accordingly, our use of these two terms is intended to reflect our engagement with educators' practice-based responses to tackle disengagement and risky transition, to try to prevent ESL, as one facet of NEEThood.Unlike other recent studies that explore NEEThood among young adults in the transition from school to the labour market, or in navigating the labour market (e.g.Parola and Felaco 2020;Gaspani 2018) this study considers the perspectives of young people at risk of NEET while they are of compulsoryschooling age.

The varying emphases of stakeholder perspectives on the risks to ESL
The absence of young people's voices from research into the risks to ESL is an identified area of concern (Downes 2013;Holte 2017;Santos et al. 2020).Early leavers have assumed the societal perception of being 'problems' that need to be 'treated' (Ross and Leathwood 2013, 328) and construed as '"culturally deprived", unintelligent, unskilled, unadjusted … and delinquent' (De Witte et al. 2013, 16).Most studies on the risks to ESL are quantitative interrogations of large-scale data sets from national studies (De Witte et al. 2013, 16-17;Holte 2017) where data from young people and adult stakeholders tend to be conflated.Self-report questionnaires are the most common method for adult perspectives (see González-Rodríguez, Vieira, and Vidal 2019, 225), while studies exploring young people's perspectives on the risks to ESL are predominantly qualitative (e.g.Santos et al. 2020;Taylor 2002;Bridgeland, Dilulio, and Morrison 2006;Hodgson 2007;Bickerstaff 2010;Rouse 2019).Although important practical and methodological reasons may explain this contrast in methodological approach (the difficulty of constructing a representative sample of NEET young people to include notoriously difficult to access groups such as young parents for instance), this nevertheless risks further restricting the influence of young people's perspectives on the risks to ESL in terms of policy responses.Comparing these two bodies of literature, it is clear that young people's accounts highlight different risks and reveal assumptions inherent in adults' perspectives, which we will now outline.Disaggregating young people's views on the risks to ESL from adults' perspectives may offer new lenses on existing categories of risk as well as clarifying which risks are most pressing for young people themselves, enabling a better understanding of how best to mitigate those risks.
Low academic achievement is identified as 'the strongest predictor,' of ESL, mediating the impact of all other factors (Battin-Pearson et al. 2000, 579).However, while 'academic failure' is often seen as a cause in itself by adults, young people's testimonials nuance this assumption in pointing to schools' 'excessive emphasis on academic achievement' (Santos et al. 2020), a personal sense of hopelessness (Hodgson 2007, 56;Rouse 2019, 83;Doll, Eslami, and Walters 2013, 9), feeling too far behind to catch up (Bridgeland, Dilulio and Morrison 2006, 3) and reaching a point at which retrieving education is not perceived to be possible: 'my grades were so low they couldn't actually rescue me' (John in Hodgson 2007, 53).This suggests that it may be the perception that the education system has no other pathways or support, which may explain the link between academic failure or underachievement and drop-out for young people.Lörinc et al. (2020) argue for the importance of an ecological approach to better understand the ways that the choices and outcomes for NEET individuals are constrained.They show how NEET young people have 'internalised neoliberal individualistic discourses,' which frame educational failure or low achievement as an individual factor when it should be more appropriately viewed as 'fundamentally social and structural in nature' (425).Similarly, recent quantitative research draws attention to ways in which NEET risk factors previously viewed as characteristic of the individual are impacted by other arenas of risk.Simões et al. (2021) highlight an association between key aspects of hope (agency and the ability to plan future pathways) and the quality of the individual's social relationships and support.The study demonstrates that lower education levels, lack of financial resources and NEEThood are 'associated with lower agency prospects' (178) and draws attention to the ways in which the lack of institutional support for young people's pathways in the Italian context hits those without informal familial support the hardest.
While young people cite feeling unsafe and a lack of physical comfort as key factors in their decision to leave school, these factors are not mentioned in research from adults' perspectives.Instead, they reference the responses to causes rather than the experiences that lead to them.Truancy is an important risk factor, signalling declining school engagement, given that absentees have a far higher likelihood of drop-out (Cabus and De Witte 2015).Research from adult perspectives calls for more to be done 'to keep kids in school every day' (Henry 2007, 33).In contrast, a major US survey of high-school dropouts found a lack of order or safety to be key problems (Bridgeland, Dilulio, and Morrison 2006, 9).Feeling unsafe because of dirty, overcrowded buildings (Bickerstaff 2010), bullying (Bickerstaff 2010) and a lack of adult support to tackle it (Hodgson 2007), daily harassment (Hattam and Smyth 2003) and disruption and lack of discipline (Bickerstaff 2010;Bridgeland 2006) are recurring themes in early leavers' accounts.For these young people, truancy may be a solution to problems that were not successfully resolved within the school.
Behavioural descriptors are frequently cited as risk factors for researchers using lenses such as 'deviance' and 'bonding to antisocial peers,' see Battin-Pearson et al. (2000, 568); 'maladjustment' and 'delinquency' (Henry 2007); problem behaviour (Rumberger 2004); poor self-control (Blue and Cook 2004) and a lack of resilience (Lessard et al. 2014).De Witte et al. (2013) argue that such stereotyping has grown from statistical approaches to characterising risk 'as an aggregate of combined probabilities to an array of separate risk factors, the overall average of which is thought to represent the typical dropout' (16).In stark contrast, early leavers themselves cite a lack of discipline and order in school as key factors in their decision to leave (Lörinc et al. 2020, 418;Bridgeland 2006, 3).One young person described her daily school life as 'walking into a lion's den' (Bickerstaff 2010, 39), while another early leaver described how she 'felt really explosive' (Juliana in Santos et al. 2020), becoming aggressive in response to feeling unsafe.Santos et al. (2020) highlight the key role played by 'institutional conditions' and experiences of school-level processes in young people's account of dropping out.This challenges assumptions around intrinsic factors in early leaving; that early leavers are badly behaved and antisocial, and attests that some feel distressed and unable to learn in the school environment.Other narrative research with early leavers suggests that it is the rigidity of the school system itself that contributes to and builds deviant identities where 'school processes and acceptable student identities are centred on a normative core' (Hodgson 2007, 47), relegating those who do not fit to the margins.
Lastly, the importance of the relational dynamic between teachers and students has been identified (Blue and Cook 2004) and implicated in young people's resilience to risk (Lessard et al. 2014).Young people's perspectives signal that students' sense of 'belonging' (Lee and Breen 2007) is dependent on the 'relational web' (Clandinin et al. 2013) around them, while feeling 'uncared for' and experiencing a 'lack of teacher support' can 'become crucial in the decision to leave school early' (Santos et al. 2020).Difficult interactions, characterised by shouting and communication failures with authoritarian teachers (Hattam and Smyth 2003), are experienced by students as part of the 'fabric of school' (Hodgson 2007, 47).Downes (2013) points to 'diametric structures of communication' in schools as leading to 'authoritarian, fear-based relations which drive students out, with students describing teachers who shout, 'roar' and how 'very harsh teachers usually make me stay out of school' (student age 16 in Downes 2013).
From this perspective, early leaving can be seen as a tactical and agentic choice (Hattam and Smyth 2003;Bickerstaff 2010) where qualitative research into young people's accounts offers invaluable insight into the incremental encounters that emphasise the process of leaving school.This situates the improvisational nature of their daily interactions as young people navigate the education system and forge an identity and life for themselves.Young people's focus on the immediacy of their lived experience of school as fluid and evolving contrasts with research from adults' perspectives which tends to view ESL from a longer-term and established trajectory, as the inevitable consequence of bad choices and poor behaviour.

Introducing a conceptual framework for analysing the risks to ESL
As the literature review has indicated, while adults are more likely to point to psychological risk factors, young people themselves identify social and environmental factors.Our challenge was, therefore, to employ a conceptual schema that could account for the interactions between micro level contextual factors through to macro level policy and cultural factors (Grace, Hodge, and McMahon 2017).In one of the most widely used and compelling attempts to do this, Bronfenbrenner's (1979) 'ecological systems' theory ' (1979) distinguished between the different key contexts of social influence including the micro, meso, exo and macro layers of the social world.This has been applied in the varying domains of health, social work, youth criminal justice and education.In integrating all of these influences, an ecological model has the advantage of theorising forces at work in the individual, family, community and cultural arenas in shaping children's development, as well as accounting for the interaction between the forces operating at different and nested spheres (micro, meso, exo, macro) in the child's daily life (Sidebotham 2001, 3).In line with others who have taken an ecological perspective (Bronfenbrenner 1979) to conceptualising the risks to ESL (e.g.Lörinc et al. 2020, Haugan, Frostad, andMjaavatn 2019), we employ a holistic framework of five discrete but interrelated categories of risk developed (see Brown et al. 2021) through the process of this research.This includes: a) 'Personal challenges' which the child is born with, acquires or experiences; b) 'Family circumstances', such as; living in a low-income or workless household, and familial support and value of education; c) 'Social relationships', which refer to relational challenges brought about through all types of relationships outside of the family; d) 'Institutional features of the school and work-place' such as spatial aspects, organisational policies, institutional norms, expectations and available resources; and finally e) 'Structural factors' which refer to those forces furthest removed from young peoples' daily lives, in operating outside of the agency of young people, their families, and the individual institutions that support them.These include economic challenges, national policies and challenges of the educational system.
The advantage of applying this classificatory model is that it enables consideration of the varying arenas that affect how young people navigate their education and work trajectories, including their daily lives at home, school and in the community, as well as a consideration of the mediating influences of individual factors and the broader social, policy and structural influences.An ecological approach, however, is not without limitations.It has been argued that the complexity of the model has led to static interpretations of the different influencers on children's development (Sidebotham 2001).We, therefore, urge following at life course perspective (e.g.Alexander, Entwisle, and Kabbani 2001) in considering how different 'spheres' of the framework may exert more or less influence at different times over the child's development, e.g.'social relationship' factors may become more prominent as s/he moves into adolescence.It is notable that 'structural factors' include a number of foci with processes that could warrant a disaggregated analysis.We combine them here for the key purpose of emphasising a category of risks that individuals can do little to circumvent, by highlighting those forces constraining young people's pathways that are outside of individual control at the local level.See Theory Development and Data Analysis for further explanation of how this model was developed.

Methods
The findings presented here are taken from the first phase of the UK sample of a threeyear European-funded research project, which aimed to understand and intervene on ESL across five European nations.The UK project aimed to understand Early Leaving' in the local region, through conducting interviews and focus groups with 39 young people (identified as either NEET or at risk of NEEThood) and 53 educators supporting them in various capacities, including teachers, school leadership teams, career's advisors, tutors, mentors, educational support officers and local authority service leads.The UK sample included 11 settings, 3 academy schools, 2 specialist schools for children with SEND; two educational projects serving Early Leavers and those at risk of Early Leaving; one 'virtual school' offering educational support for children in care, and the local authority, which offered employment-bed training for NEET young people.

Overview of the region and key groups
Due to the complexity of risks to ESL, it is necessary to consider the local factors as well as the structural and national factors which may lead to NEEThood.Our research, therefore, focused on one region in Southern England that we named Windy county.The region was chosen for its distinctive profile of NEET young people.Following national government austerity measures, the local authority team responsible for identifying and supporting NEET young people was replaced with key individuals responsible only for tracking the number of 16/17-year-olds to remain in education or training, as part of the September guarantee (DfE 2014).While overall employment within the county remained above the national average, the number of 16-and 17-year-olds known to be participating in Education or Training had dropped in recent years.At the time of the research, only 89% of young people were in education or training, which placed the county in the 5th quintile in England.Furthermore, of the remaining young people, only 2% of them were identified as 'known' Early Leavers (for whom data was available), with the remaining 9%+ assigned to the category of 'destinations and activity is not known'.This was surprising given that the economy was broadly stable with above national average levels of employment in manufacturing and public administration due to the number of military establishments within the county, and agricultural work reflecting the largely rural geography of the county.The long and dispersed shape of the region and relatively poor transportation links contributed to poor access to educational provision (particularly in VET).Provision to support Early Leavers was, therefore, largely dispersed and mainly fixed-term project funded.
The young people identified within their settings as being NEET or at risk of NEEThood fell under a number of different, and at times, overlapping groups that have been found to be of a higher risk of ESL.This included young people from Traveller/Roma/Gypsy families (European Union Agency for Fundamental Rights 2018), in care (Dickens and Marx 2020), military service families, with SEND (Batini et al. 2017), with Social Emotional, or Behavioural Difficulties (BESD) (Holmes, Murphy, and Mayhew 2019), young carers (Sempik and Becker 2014), experiencing mental health difficulties (Rodwell et al. 2018), and from low-income families (Schoon 2014).We were unable to speak with young offenders or young parents (especially young mothers), arguably one of the groups most at risk of Early Leaving (Maguire and McKay 2017) but also the hardest to access.

Sampling approach
Participants were recruited using a snow-balling sampling strategy (Taylor et al 2015, 52).Having secured support from the local authority, the research team was guided by service leaders to approach educators and professionals at settings with a reputation for their productive work with disengaged young people locally.These professionals then identified young people in their setting where markers of vulnerability or disengagement flagged them up as being at risk of NEEThood.While there are benefits in researching the perspectives of educators with demonstrable success in working with vulnerable young people, this inevitably excludes the perspectives of those with less sensitivity or motivation to understand the risks to ESL and may not be a representative reflection of all educators who work in this area.Information sheets and consent/assent forms were provided to young people and their parents/carers.The educators then invited the young people to come and speak with us.To avoid stigmatisation, the invitation was attributed to young people's status as working in a setting identified as supporting young people's futures.As all young people had been working with these educators already, there was an established rapport, which leant to the rationale that they had been identified because they had relevant experiences to share.The sample, therefore, reflects those who responded and wanted to participate.Two research team members carried out a total of 30 individual/paired interviews and focus groups with 53 adults and 8 focus groups with 39 young people at the settings, all of which were audiorecorded and are included in the data analysis.While focus groups lack the opportunity for deeper individualised probing and reflection, educators universally advised us that the gains of offering a 'spring board' for discussion would outweigh the greater risk of nonparticipation, especially among the more reticent young people who had previous negative experiences with unfamiliar adults, and often needed to 'warm up' into conversations.Furthermore, the issues of ESL and NEEThood were introduced at the start of the focus group for greater transparency.Therefore, while participants were assured at the outset of each focus group that there was no obligation to contribute a response, we believed that young people would feel less expectation to speak in a group setting.Adults were given the option to speak with us on an individual, paired or group basis, and as such we were mindful to consider any possible correlations between factors raised and interview type, but none emerged.As we were reliant on educators scheduling group discussions, we could not offer this choice to our young person participants.

Theory development and data analysis
An initial framework of risk factors was developed from a review of the literature into ESL across the UK and Europe at the start of the project.Each of the five project teams undertook a literature search into the risk factors to ESL in their respective nations using the major Education and Sociology databases available in each country.These reports were shared for analysis by the UK team (the authors of this paper) responsible for developing the theoretical framework and compiling the research tools.In analysing the literature reports, the UK team identified and listed all factors raised, before looking across the set of issues to group them according to the different arenas of the young person's life.Accordingly, five emergent categories of risk were developed: Individual, Family, Social, Institutional and Structural.Issues from the literature were assigned to one of the five categories.This early conceptual model was subsequently shared with the four other national project teams, both to agree the categories themselves and the assignment of issues to them.
The categories resulting from this initial analysis of the literature were used as 'orienting concepts' (Layder 2014, 6-7) to guide the data analysis while allowing an exploratory approach which could remain alert to new issues arising in the data.Existing risk frameworks were therefore engaged with 'as one of many possible accounts,' while ensuring that the accounts of our participants could inspire 'new understandings of the categories' and even suggest new ones (Holte 2017).As a consequence of the 'looseness' of the orientating concepts, the risk categories were subsequently developed and transformed through the analytic process.Where originally we had, in following the literature, framed the categories in emphasising the agency (and therefore accountability) of social actors, the iterative process of data analysis highlighted the circumstantial and contextual challenges faced in these socialising arenas (see Brown et al. 2021).For example, the category of risk 'Individual factors' (e.g.see González-Rodríguez, Vieira, and Vidal 2019) was revised to 'Personal Challenges' and 'Family Factors' (e.g.see De Witte et al. 2013) was adjusted to 'Family Circumstances'.In light of these findings that highlight the distinction between adult and young person perspectives on the risks to EL, we emphasise the importance of theoretical categories which do not conflate challenge with choice, particularly with regards to attributions made to young people.Furthermore, while some issues classified as 'structural factors' could be grouped under broad themes identified from the literature, 'i.e.Education Policy', we also coded them in the terms used by educators which reflected both national and regional specificities 'i.e.lack of alternative provision available'.
Each transcript was coded by two or more of the four-person research team, who discussed their interpretations and collaborated to identify themes emerging from the analytic process.The significance of the issues raised was considered in two ways.Firstly, in order to determine any broad patterns between young people's and educators accounts we analysed the percentage of total discussion time across the dataset devoted to the respective risk categories according to particular stakeholder groups.To move beyond a descriptive overview, the team engaged in a final analysis of the coding schema to elicit the generative mechanisms (Haig 1987) which underpinned the ways that risks were produced, mediated or compounded.
In presenting the findings, the tables that follow within the findings section show the top 10 risk factors, ordered according to the percentage of the discussion dedicated to each issue, which was then corroborated against the issues that participants named as being the most significant.

Findings
Our findings highlighted key differences in the relative importance attributed by adults and young people to the various categories of risk.While for young people, the most discussed categories were Institutional features of school and work (37%) and Social relationships (27%), for educators Structural factors (40%) and Personal challenges (23%) predominated (see Table 1).The proportion of the dataset attributed to discussing Family circumstances was comparable for both young people (11%) and adults (8%).It was the fourth most discussed category (of five) for both young people and adults.
We now focus on a more in-depth consideration of the issues within our data according to the five categories of risk outlined by Brown et al. (2021).In clustering the various risk factors to ESL in this way, we are better able to consider the patterning of responses according to the various categories emphasised by the two key participant groups in our study: educators and young people.

Structural factors
While 'structural factors' emerged as the most significant risk category to ESL based on it being the most discussed category across educators and young people combined (see Table 1), this masks a significant discrepancy between their perspectives on the most significant risk category.While 'structural factors' was the most discussed category amongst educators (40%), for young people, it was the least discussed category (7%).Nevertheless, 27 young people mentioned educational policy during discussions.Furthermore, while, educators discussed 28 different risks within this category, young people mentioned only 3 risks (see Table 2).Educational policy was discussed both by young people and educators.The requirements for post-16 Maths/English and participating in education/training until 18 were seen by adults to encourage young people to 'disengage,' 'stop attending' (LA Support Service Officer) and as something which 'probably puts more young people off' pursuing educational and training pathways (LA Support Service Leader).
Young people themselves expressed that they felt pushed to pursue core subjects associated with the EBacc (DfE 2015) that did not interest them and made them anxious about passing exams: We had to pick either geography or history […]  They described the considerable exam pressure and stress they felt to achieve certain targets and grades, particularly in core subjects like English and Maths: A lot of people may not know how to deal with the pressure and they'll fall into leaving school and not actually doing their GCSEs.(James,14,Dusky Academy) When I am working, she is like, this set is like grade 8 or this work is grade 6. (Catherine,14, Cloudy Academy) Critiquing the school's emphasis on measuring outcomes, young people felt teachers were concerned with the school's 'Ofsted and […] want[ing] the school to get outstanding' (Sarah,13,Cloudy Academy), above focusing on students 'learning it to learn it' (Imogen,14,Cloudy Academy).Educators also identified the stress of 'getting students through' exams and felt the pressure of performative benchmarks had 'forced [them] down a pursuit of excellence road' (NEET Project Manager).
Educators alluded to another response to the pressure to meet these performance benchmarks, which may impact those young people at greatest risk, the 'off-rolling of children in Year 10/11 when it comes up to them taking their GCSEs.'They described students being given 'alternative packages and alternative provision […] because schools just say, well, we can't meet their needs, but then […] they don't want it on their league tables' (LA Support Service Leader).
The top risk factor raised by educators was 'lack of funding.'They discussed the deleterious impacts of the 'rapidly decreasing budget' and 'depleting resources' in schools.They also described cuts to youth services, and a precarious and insecure landscape for at risk young people and those with additional needs needing alternative 'Accessibility to education' was identified as a key risk by educators, with inadequate, expensive transport causing families who live in 'out of the way places' to feel 'they're cut off', 'isolated', 'can't go anywhere' and 'can't access much' (Kite Academy Careers Advisor).

Institutional factors
For young people, institutional factors of school and work were the most discussed category (37%), whereas for educators, it was the third most discussed category (16%) (see Table 1).
Educators viewed traditional schooling as inappropriate for at-risk learners who 'don't learn constructively in that sort of environment' (LA Support Service Officer).Large class sizes were seen as challenging for these learners who 'generally don't function in classroom settings in groups over about ten (Misty SEND School Leader) and 'could really do with a more personalised approach' (Dapple Academy Curriculum Lead).Students echoed this, complaining about mainstream classrooms where 'there are 30 other kids and you are asking for help […] [teachers] have to go around all the other students and then go to you' (Adam, Dusky SEND School).
While young people lamented the lack of help in class, educators emphasised the limited time and resources available to support learners' diverse needs and expressed frustration over feeling forced to prioritise teaching the 'teachable' to the detriment of those at risk.This also resulted in a lack of guidance available for transitions from school to alternative or further education/training: You are stuck between a rock and a hard place.So, do the 29 kids matter, or does the one child matter?[…] You know that the one child is struggling everywhere, but there isn't a safety net there to catch this child.What do you do for this child?Where should they go?(Dapple Academy Curriculum Lead) Young people complained about feeling physically and socially uncomfortable in hot, busy classrooms, which caused frustration and led to them wanting to 'walk out of class' (Alex,12, Misty SEND school).They also spoke about feeling unsafe outside the classroom due to a lack of staff available to notice and deal with other children's challenging behaviours.
Greg: I've got to say that sometimes I don't really feel safe when boys around me kick off.
Interviewer: So, that's a big issue for you then feeling safe is it, Greg?Greg: It's an important aspect for schools.All schools need to have protection; you need to make sure boys don't get in to fights … in [X] school they literally just have a circle of people … .where two people are having a fight and then everyone.. no one did anything … One kid was just beating up another kid and someone videoed it and posted it on Facebook.(Greg,12,Misty School) In addition, both young people and educators discussed stringent school rules and behavioural policies, which often led to at-risk learners being 'sent straight away to an isolation room' often 'for six periods' (Dapple Academy Pastoral Lead) so that teachers 'can concentrate on the ones that do want to or can learn' (LA Support Service Officer).Daniel (14,Dapple Academy) found the behavioural sanctions confusing and unsupportive of students' learning needs: I didn't understand anything that was going on because everyone was getting isolations every day and I was, 'This is insane!' […] even though people go to isolation, they don't learn anything in there because […] every time you go in there, you're doing the same work and it's the same book that you'll get given.
Another student in a SEND school noted being removed to an inappropriate learning space in her former academy school: they kept me in a corridor for seven months and I did not do any work […] I just did colouring in and stuff.(Juliet,15,Dusky SEND School) While educators clearly acknowledged that such internal exclusion meant individuals needs weren't adequately met, they saw such 'internal exclusions' as preferable to excluding students from school, which could put them at further risk of 'offending behaviour', 'antisocial behaviour' and 'kids being stabbed' (LA Support Service Officer), especially given that at-risk learners were seen as cut off from future learning opportunities due to poor transport impacting attendance (Table 3).

Social relationships factors
Risks coded as social relationships issues emerged as the second most significant category for young people (27%), while for adults they were the least discussed area (7%) (see Table 1).Despite this disparity, there were similarities in that difficult relationships in school, and peer group challenges were, respectively, the second and third most discussed issues for both young people and adults (see Table 4).Those that had disengaged pointed to the lack of support from teachers and school leaders who they didn't like talking to and who they felt didn't listen to them.They reported being told to 'just get on with it' (Ben,18,NEET) and felt there was 'no support available'(May,19,NEET) and 'when you stuck up for yourself you get expelled' (Jake,18, NEET).
In contrast, adults focused more on difficulties in peer relationships in school, particularly navigating friendships; I think their ability to make relationships is not always that great … , they haven't necessarily had good role models in, particularly in.. anything to do with reparation, they think that once something has gone wrong, you walk away from it.[Violet School Officer] This was seen to connect with a more generalised social anxiety; I think peer anxiety is huge at the moment … they're frightened of people of their own age.They are really conscious about who those people are, and what they might be thinking about them and how they might or might not get on with them.[Dapple Academy Pastoral Lead] Such challenges were seen to have knock-on social implications in 'coping with large groups' (NEET mentor) in turn heightening vulnerability to negative peer influences, either by joining gangs or turning to social media or online gaming communities in place of real-world socialising.
The most significant difference in stakeholders' perspectives was the attention given to the issue of bullying, which was the most discussed issue for young people but mentioned much less frequently by adults.Several young people cited bullying as the reason they left or moved school, while others still struggled with it.
There's two bullies in my year and they just don't stop, … it just gets annoying sometimes because you want to come to school to learn and they just make it hard for you.(Daniel,12,Dapple Academy) Those already disadvantaged were seen to be most vulnerable, particularly anyone who 'looks different in clothing and shoes' (Bob,13,Dusky SEND School) or 'if you've done something or you are a certain way … have a disability … people just aren't able to accept you' (Alex,12,Misty SEND School).
While both educators and young people discussed students' fear and anxiety, adults framed this as a self-consciousness or social anxiety, while young people talked of feeling worn down by relentless intimidation.

Familial circumstances factors
Risks coded as familial circumstances emerged as the second least significant category for both young people (10%) and adults (14%) (see Table 1).While both referred to low parental value of education and low family support, young people acutely felt that the loss of a family member and having to support the family were predominant risk factors, factors not referenced by the adults.
Young people voiced their discontent with a rigid school system which was not equipped to deal with or support their emotions when processing the loss of a family member: If you skip a lesson, you have to go in front of the whole of your year and say why you skipped.[…] I skipped because […] my mum told me that my granddad passed away and I was really angry so I wanted to walk out of school and see my mum.(Carlos, 13, Dapple Academy) Students indicated that supporting family members (either as carers or financially, through work) were significant risk factors for ESL: I reckon it could, if people drop out, it could also be for that reason -[…] because she was a carer; she had to look after her parents.(Alex,12,Misty SEND School) Your parents can't afford stuff[…] maybe if they [young people] are like 14 or 15 and they might have to go and work somewhere like little so they can afford money and then that means that they will have to, they will be too tired to come to school, or they may not attend.(Juliet,11,Dusky SEND School) A further risk factor both adults and young people converged on were the different ways in which low family support could affect remaining in education.Young people suggested that; When [young people] lose connection with their mums and dads […] there is nothing holding them back from making their own choices saying 'I'm not doing school, I'm going to sit around at my mum's house and play games all day' (John,Misty SEND School).
Adults conceptualised 'low family support' in terms of limited parental skills: [some parents] have no real skill in how to be a parent and therefore it breaks down and their children become difficult to manage (Dapple Academy School Leader).
While some educators conceded that parents want to support their children, but struggle to do so effectively, others took a more moralistic view of 'low family support' which compounded what adults felt was the largest family circumstances risk factor: parental values.
As in, parents haven't got jobs, so the young person doesn't have an aspiration to get a job, so they don't get a job.They have children, those children see that parents and grandparents haven't had a job, so they're like, 'well, what's the point in me getting a job, I can just have benefits'.(NEET Mentor) Where young people highlighted specific challenges they face due to low support from home, such as caring responsibilities, pressure to earn money, lack of guidance or losing emotional connection to home, educators tended to individualise the issues, referring to more generalised consequences for the young person, such as becoming 'difficult to manage,' having low aspirations or poor parenting skills (Table 5).

Personal challenges factors
'Personal challenges' was the third most significant category for young people (18%) and the second most significant for adults (22%) (see Table 1).Both groups highlighted the barriers of mental health issues, and low aspiration and motivation.While adults felt disengagement was the cause of some of those issues, young people indicated that irregular school transitions led to those problems (see Table 6).
Adults and young people agreed that mental health plays a significant role in ESL: Mental health in students and in parents as well has been a massive factor and … we're seeing […] more and more students suffering with mental health that is having a direct impact on education which … in many cases is becoming disastrous for education and further education and their employment.(Kite Academy School Leader) Educators raised concerns about mental health challenges without official diagnoses and their role in ESL: Lots of sort of low-level mental health stuff, you know, paranoid behaviours and bad experiences through school where they might have been bullied or … just didn't fit in … I think that's a huge problem for young people that I work with, who aren't meeting services to be supported, so they haven't technically got a mental health problem that is being addressed by a service, so they're just winging it.(LA Support Service Officer) Adults and young people also described mental health problems in terms of struggles with motivation, aspiration and expectations as a consequence of experiencing NEEThood: I saw loads of people on Facebook, and they say about like life, as soon as you finish GCSE people think it is going to be like life is going to go to be really great, but they just say it goes downhill […] I think the money and trying to find a place to live and all that stuff and trying to do well in college.(Andrew,13,Dusky SEND School) Table 5. Key risks prominent in the familial circumstances category.
The saddest thing I see are young people that have been NEET for six months, twelve months, and the longer they spend in that situation of not getting up until 11.30am, 12 o'clock, not really socialising, not really engaging with people … The harder it is for those guys to turn it around.(NEET Programme Leader) Mental Health struggles and being NEET generated disengagement, in adults' perspectives.However, young people's accounts drew attention to the role of socioeconomic and logistical factors, 'the money' and 'trying to find a place to live,' in this process of disengagement.
While adults lamented the influence of NEET peer groups as something they found difficult to counter, young peoples' discussion attributed difficulties with motivation and engagement with irregular transition, to things going 'downhill' upon leaving school.Personal challenges around mental health, motivation and disengagement were frequently discussed by both stakeholder groups.While adults discussed it as a challenge they were struggling, without adequate support, to address, young people referred to the personal challenges they experienced with lacking support in areas such as transition, finances, finding a job and a home.

Discussion
While the data revealed clear parallels between the issues raised by young people and the educators that supported them, applying our conceptual framework to both young people's and educators' accounts during analysis revealed that each group attributed these issues or risks to different domains.While young people focused on institutional and social factors, the adults supporting them focused rather on structural factors and personal challenges.An important implication of an ecological approach is that it is necessary to move beyond the identification of individual factors in order to consider interactions between and within the nested spheres of influence (Sidebotham 2001, 106).A key advantage of our holistic framework was that it enabled us to consider the constraining pressures of systemic factors and the interrelation between the risk domains to see the commonalities between the issues, but the contrast in the ways young people and educators understood and explained these.We, therefore, identified three discreet but interlinking generative mechanisms (Haig 1987); Firstly, the pressure of performative benchmarks in schooling integrated risks raised under 'structural' (e.g.emphasis on measuring outcomes) 'institutional' (e.g.high stakes testing and examinations) and social relationships (e.g.pressure to achieve) categories.Secondly, we identified the inability of mainstream schooling to meet young people's diverse needs in integrating risks raised under 'institutional' (e.g.institutional rigidity) and 'personal challenges' (e.g.irregular school transition) categories.Thirdly, difficult relationships in school integrated risks raised under 'social relationships' (e.g.bullying) 'family' (e.g.dysfunctional relationships at home) and 'personal challenges' (e.g.difficulty in trusting others).
Of course, the educators interviewed were motivated by a passion to support young people, and therefore may be more keyed into the risks young people experience than some of the other adults that young people had encountered over their educational careers.The following discussion explores both the alignment between stakeholders' perspectives and their different lenses and emphases given for each of these generative mechanisms.

The pressure of performative benchmarks
The pressure of the performative benchmarks associated with a neoliberal approach to educational policymaking has been well documented within the sociology of education (Reay and Wiliam 1999;Harlen and Deakin Crick 2002) since national assessment tests became linked to expected outcomes for schools, children and more recently, to teachers.However, unlike teachers in Booher-Jennings' study (2008), who attributed children's achievements to effort and behaviour, educators in this study voiced frustration at their lack of agency to resist policy demands that compromised their pedagogy.Ball (2003) termed this 'value's schizophrenia': the catch-22 situation experienced by educators who feel their values are fundamentally at odds with the pressures of performative demands.This echoes the impotence expressed by educators in this study, who lamented that the policy demands linked to core subjects both pushed them towards advocating the academic route while also compromising young people's completion of the vocational pathway.Educators also drew attention to a lack of affordable, regular transport as an additional locally-specific barrier, which further limited young people's access to suitable provision.As they were unaware of the external forces limiting teachers' agency to mediate or resist policy demands (see also Silfer, Sjorberg, and Bagger 2016), young people could be forgiven for seeing themselves through the eyes of teachers as merely statistics reflecting school success or failure, and as pawns of the educational system, where their final outcomes were more important that their current achievements.

The inability of mainstream schooling to meet young people's diverse needs
While the constraints of the neoliberal policymaking were most evident in adults' discussion of the risks to Early Leaving, the effects of such policies upon the daily experience of schooling dominated young people's discussions.Underpinning both stakeholders' perspectives on the institutional barriers to learning was a perception of the fundamental inability of mainstream (and particularly academy) schooling to meet children's diverse learning needs.While adults pointed to contextual factors such as a lack of resources and time, limiting their capacity and opportunity to engage with students' additional support needs, young people pointed to a consequential negative experience of being at school.Young people in this study claimed they wanted to learn but were deterred or prevented from doing so by unresolved bullying, poorly managed behaviour (Bickerstaff 2010) and having to study subjects they did not enjoy.This challenges the stigmatising narratives of deviance that characterise adult perspectives of Early Leavers (Battin-Pearson et al. 2000;Henry 2007).Indeed, young people highlighted features of the school environment itself, which they experienced as emotionally, and socially alienating, unsafe or physically uncomfortable, echoing Bridgeland et al. (2006).The rigidity of institutional structures was particularly evident in discussions of perceived deficiencies in meeting the needs of students with mental health difficulties; a concern shared evenly between young people and adults.Schools have been afforded greater responsibilities to support children's mental health in recent years (see DoH/DfE 2017).However, our findings support policy critiques that have questioned the framing of mental health as an educational problem (Brown and Carr 2019).The conflation of educational success and mental resilience in policy leads students to identify schooling narratives that equate well-being difficulties in school with individual weakness (Brown and Dixon 2020).Educators recognised the impact of constraints on attending to learners' diverse needs in terms of 'disengagement' from learning, which has been construed as one of the most significant personal risk factors to ESL (Henry 2007).While young people also expressed a sense of disengagement, for them it was the consequence of their needs not being met and the attendant assault to their motivation, aspirations and expectations; particularly the impact of social exclusion or the sense of alienation incurred through irregular schooling transitions.

Difficult relationships in school
Both young people and their educators highlighted difficult relationships in school as a significant risk factor in early leaving, echoing calls for greater attention to the 'the social dimension surrounding the dropout process' (Haugan, Frostad, andMjaavatn 2019, 1274).Young people in the study emphasised the central role of bullying in their decisions to leave education.The educators, however, focused on problems within peer relationships, inferring a more egalitarian dynamic.While they framed it differently, both emphasised fraught and difficult social interactions between young people as a major risk factor.Adults perceived these to be peer group challenges resulting from students' difficulty in navigating friendships.This issue has been signalled in the recent statutory requirement to teach Relationship Education (DfE 2020b).However, here the importance of friendships is secondary to an emphasis on sexual relationships.Furthermore, the individualistic focus on building individual competencies, for instance, to 'know what a healthy relationship looks like' (25) and cultivate 'resilience and character in the individual' (DfE 2020b) is in contrast with the reciprocal emphasis, highlighted in this data, on the challenges of doing relationships (see Brown and Donnelly 2022).Given that student-teacher communication failure is also implicated as a risk to ESL in the literature (Hattam and Smyth 2003;Blue and Cook 2004), the peer trouble teachers discuss may rather be a subtle symptom of an exclusionary and competitive culture of schooling, which staff dismiss as peer discordance.Not recognising an inherent power imbalance, staff may see these conflicts and difficulties as reciprocal, while young people see themselves as the victims of bullying.
Furthermore, these difficult relationships in school may be further fuelled by perceived pressures associated with a focus on performance, cited by educators in this study as resulting in a lack of time and a push towards advocating academic routes.The young people's sense that school staff cared primarily about their grades echoes findings in the literature that early leavers find interactions with teachers unsupportive (Hodgson 2007) and not meaningful (Bickerstaff 2010).Taken together, the perspectives of adults and young people suggested the neglect of relationship-building in schools and the marginalisation of young people's social needs.With educators focusing on academics and attainment, young people were left to navigate this on their own, without reporting difficulties to teachers, who they perceive to not be concerned with their social and emotional well-being.Adults in the study lamented a lack of role models at home for effective relationships and resolving conflict.Notably, young people's accounts of school experiences with teachers shouting and of being removed to exclusion rooms, raised questions about how much role modelling of doing relationships and dealing with conflict they experience in school.Overall, young people's experiences of feeling unwanted, uncared for and unsupported in education appeared to be a significant risk factor that the adults who support them did not seem to recognise as a factor within their control.

Conclusion
A better understanding of the ways in which young people and adults experience, perceive and explain the risks to ESL may reveal where adults overlook critical factors for young people.It is important that both stakeholder groups are on the same page in order for effective interventions and support strategies to be imagined and implemented.The ecological framework of five risk categories applied in this research considered the contextual, structural and psycho-social influences that frame how young people navigate their educational, life and work pathways.Underpinning this broad spectrum of risk factors, were a range of findings that revealed a current schooling system that is fundamentally not fit for purpose.A broad overview of the findings highlighted educators' overwhelming emphasis upon the structural risks arising from a focus on performance, and the limited funding or alternative options for learners who don't meet benchmark expectations, as well as the stringent requirements to remain in education/training.While adults' focus was on educational policy, young people emphasised the institutional manifestation of structural factors, which shape the daily experience of schooling.These risks were indicative of the rigidity of schooling structures and behavioural systems that led to educational exclusion for learners with diverse needs.
Young people's testimonials also highlighted the emphasis they put on social factors, a unique element from our framework not disaggregated in other models.These are experienced by learners as inescapably part-and-parcel of school life andpart of the 'fabric of school' (Hodgson 2007, 47).Adults also pointed to the interpersonal factors, but their emphasis was in line with policy narratives (e.g.DfE 2020b) regarding young people's individual orientations to relationships (see Brown and Donnelly 2022) as opposed to the ways that young people are caught within fraught relationships with teachers (Hattam and Smyth 2003) and friendship 'binds,' which they wrestle with but cannot overcome (Brown 2014).
This prompts reflection on the reasons behind the distinct shifts in emphasis between adult and young person perspectives.The first explanation points to stakeholder's various experiential framings.In being more cognisant of, and subject to, policy and resourcing strictures, educators may well point to structural factors.While young people, as the objects of educational systems are best placed to reflect on the institutional constraints.A second argument relates to the auspices under which each category of risk falls, as well as the agency that stakeholders perceive themselves to have within each of the five arenas of risk.Doll, Eslami, and Walters (2013) argue that different stakeholders may 'intuitively' report 'views of school problems that are in their best interests either politically, or with their peers' (10).This may explain why students point to institutional management while educators point to regional and national governance as those areas over which they have the least control and accountability for failure.A more sympathetic analysis is that the limits to agency are difficult to acknowledge, especially in the face of considerable effort to navigate such challenges.A final explanation is that theories of development are limited by societal assumptions that construe adulthood/childhood through dichotomous terms as 'beings' versus 'becomings' (see Prout 2011).This argument suggests that adults may be locked into conceptions of young people's pathways towards the final destination of childhood, while young people themselves are caught in the eye of the storm, and struggle to see beyond their immediate present challenges.This may explain why what young people perceived as social relationships-related causal factors to ESL, educators construed in terms of personal challenges, such as disengagement and low motivation, factors that young people implied were the outcomes or amplifiers of other risks.
In acknowledging the limits of qualitative case study research, it is notable that this study's findings highlighted the regional and contextual factors which may evade large-scale national or qualitative reviews (De Witte et al. 2013).This study was conducted in an area in which, while NEET figures were high, employment levels were above average.The structural challenges perceived by young people within less buoyant local economies may well moderate their perceptions of the key risks to NEEThood (Shildrick and MacDonald 2007).Further research could therefore profitably probe deeper the distinctions between stakeholder perspectives of risk in other regions on an issue that remains a key European challenge.The overwhelming emphasis in findings from this study highlight that ESL requires preventative as opposed to purely compensatory interventions, which calls into question the current English policy direction that approaches the issue of ESL as one of only strengthening training pathways.It also suggests a need for further research to address the trenchant failure to unpick the alignment and divergence in stakeholder perspectives and which integrates the child's voice, seeing young people as vital informants in understanding and tackling ESL.

Disclosure statement
No potential conflict of interest was reported by the author(s).

Table 1 .
Relative percentage contribution for each risk category from young people and adults.bold reflect the most discussed category by; young people, adults, and overall.
and […] they are not really my interests and … I just don't think I am going to pass them.[…]I think you should pick what you want, because if you like the subject you are more likely to pass it because you are going to enjoy how you learn it.(Jane,13,Cloudy Academy)

Table 2 .
Most prominent risks raised in the structural factors category.

Table 3 .
Key risks prominent in the institutional factors category.

Table 4 .
Key risks prominent in the social relationships category.It is notable, however, that young people focused upon teacher relationships;You sort of get a vibe off some teachers on whether they like your class [or not] … they raise their voice, shout a lot, you sort of get that they don't like you at all.(Cleo,16, Cloudy  Academy) The social groups we have got[in Windy County][…] will go round together and that obviously impacts on it [disengagement] because … if the other person isn't doing anything and they're like 'why should I go and work?' or 'why should I engage in something like that?' (LA Support Service Lead)[…] it could be completely different to another person trying to move school … I think moving schools could help, but also maybe they are facing another whole school to like try and face it, It's just horrible and if you move away and then you try and start a new thing, its like so much goes on.(Jane,13,Cloudy Academy)