Endogenous assets-mapping: a new approach to conceptualizing assets in order to understand young people’s capabilities and how these relate to their desired educational outcomes in disadvantaged neighbourhoods

ABSTRACT Asset-based approaches to public service reform suggest a need for policymakers to shift attention from ‘fixing’ the perceived deficits of disadvantaged neighbourhoods, to recognising and building on the resources, or assets, they hold. However, these approaches have also been critiqued for interpreting assets so broadly that they effectively become meaningless, or so narrowly that they perpetuate deficit views. To counter these tendencies, a new conceptually and methodologically robust endogenous assets-mapping approach is proposed. This has been designed to enable nuanced insights into the assets young people living in disadvantaged neighbourhoods may draw upon to pursue positive educational and wider life outcomes. The approach’s utility is illustrated through the case of Ayesha, a 14-year-old student with a difficult relationship with school. The paper concludes that an endogenous assets-mapping approach can help to generate more positive narratives for vulnerable learners living in disadvantaged neighbourhoods than schooling may typically enable.


Introduction
Educational inequalities as young people move through schooling are of long-standing concern in the UK and internationally, with systematic gaps seen between the attainment of the most and least advantaged learners (Hutchinson, Reader, & Akhal, 2020;OECD, 2016). There is also much evidence to suggest that poor educational outcomes are widely concentrated in places that are themselves also characterised by high levels of poverty and other related poor outcomes, not least in health, housing, economic activity and community safety (Department for Education, 2018; Organization for Economic Cooperation and Development, 2016). A common policy response has been to view these 'disadvantaged' places as beset by problems or deficits that need to be 'fixed', and to then treat them as 'containers' into which externally developed, decontextualised interventions can be dropped (Hubbard, Kitchin, & Valentine, 2008). However, such approaches appear limited in their ability to narrow gaps, and can easily disregard the potential value of local resources in supporting efforts to improve outcomes more broadly (Lupton, 2010).
Responding to these limitations in policy, there have been some efforts to explore the distinctive features or resources that disadvantaged places might have, which might support good educational and related outcomes. Webber and Butler (2007), for instance, explored neighbourhoods with higher attainment outcomes than would be anticipated on the basis of their socio-economic factors alone, identifying communities' cultural attitudes to education as important in supporting good outcomes. Furthermore, Riddell (2007) reported how some neighbourhoods create rich learning environments to support young people's outcomes, though she also noted that learning from these was often not transferable to school. In different ways, such studies raise the possibility that if policy could find ways to identify and build upon neighbourhoods' existing resources, or assets, these might present new possibilities for addressing educational inequalities -though whether by supporting or challenging established discourses about valuable educational outcomes is another matter.
Nonetheless, educational research has since done relatively little to follow up and consider such possibilities. In addition to some exploration of community organising as a means of challenging structural inequalities in local education systems (Glickman & Scally, 2008;Shirley, 2009), 'funds of knowledge' (FoK) approaches -that seek to use knowledge held by pupils' families to support curricular learning -have enjoyed some popularity (González et al., 2005;Hayes et al., 2017). However, these approaches have also attracted critical attention, with questions being raised about whether only particular kinds of knowledge held by local residents are being afforded capital within professionally driven school systems (Williams, Tunks, Gonzalez-Carriedo, Faulkenberry, & Middlemiss, 2020). This suggests that approaches that claim to work from an assetsbased stance, by valuing resources -whether tangible or intangible -held within disadvantaged neighbourhoods, may actually also reinforce normative expectations about valuable learning processes and educational outcomes that arguably support existing inequalities. This points to the need for a robust discussion of how assets, held within disadvantaged neighbourhoods that might have the potential to support young people's educational outcomes and wider life chances, should be defined and identified. This paper's purpose is to begin to address this gap. To do so, it draws on community development literatures and specifically the kinds of asset-mapping frameworks these report that develop tools to identify -or map -assets within disadvantaged neighbourhoods in ways that might support design services that can meet local needs more effectively (see, for instance, Kretzmann & McKnight, 1993;Mathie & Cunningham, 2005). In doing so, this paper also explicitly engages with some of the dilemmas these assets-frameworks and mapping processes reveal, which are often unacknowledged in the literature. Two related issues in this regard are: (i) who has the power to define what counts as an asset within mapping processes; and (ii) how the potentially varied purposes for which assets might valuably be used is mapped. A further issue is how broadly assets can be defined before the concept becomes so all-encompassing as to be rendered effectively meaningless. Fuller, Guy and Pletsch (2002, p. 5), for example, suggest that the assets held within disadvantaged neighbourhoods can range from '150 year old trees' to efforts to 'beautify the main street' to foodbanks. Indeed, in fields where assetsmapping processes are already well established, some of the dangers of poorly theorised notions of assets have already been seen. For instance, the International Association of Community Development (IACD) reported that assets are an example of the 'plethora of concepts' that government and decision makers use, 'with some degree of abandon without taking on the real and challenging demands which each of them involves if they are to be effective.' (IACD, 2012cited MacLeod & Emejulu, 2014 MacLeod and Emejulu (2014) further note that, far from being empowering, a policy focus on assets can also be used to overlook basic needs and structural inequalities experienced by people living in disadvantaged neighbourhoods. They suggest that by arguing that local residents already have access to assets in their immediate areas that they can use to improve their outcomes, policy makers can place the onus for improvement on residents and blame them if they fail to succeed.
With these challenges in mind, in the following sections, we present what we have termed an endogenous assets-mapping approach, and demonstrate its empirical use and value. Our approach has been designed for use with young people aged 11-16 living in disadvantaged neighbourhoods, specifically to identify the assets they value and use in their pursuit of positive educational outcomes and wider life chances. We use the term 'approach' here deliberately. In doing so, we seek to clarify that we are not claiming to have developed a new theory or model of assets-based education, nor are we presenting a methodological toolkit for identifying assets. Rather, we are presenting a flexible heuristic framework -what Sen (1993) refers to as a framework of thought -for surfacing young people's understandings of assets, in relation to educational outcomes and wider life chances. This was developed in the first phase of a three-year empirical study with the central research question: 'What locally-accessible assets in disadvantaged neighbourhoods, beyond internal school and family processes, do young people identify, value and use, in support of positive educational outcomes and wider life chances?' Findings from the study's later phases are reported in detail elsewhere and explore professionals' views of the neighbourhood assets available to young people, the assets young people identify and would ideally like to access, and the implications for the design of local strategies that address educational inequalities. Our focus in this article is specifically on providing a detailed explanation of, and justification for, our elaborated conceptual framework underpinned by a participatory research design, which seeks to overcome some of the established criticisms of current assets-mapping processes, as outlined earlier.
In what follows, our endogenous assets-mapping approach is presented in three parts. First, the approach's underpinning conceptual framework is detailed. This draws on Kretzmann and McKnight's (1993) Assets Based Community Development (ABCD) approach, and elaborates this with reference to capability scholars' work as an additional lens to explore how assets are not only defined, but also valued and used, by young people locally. Second, the methods used to support the empirical application of our endogenous assets-mapping approach are outlined. A specific focus here is on the involvement of pupil co-researchers in an effort to address some of the challenges related to power and decision-making already noted, by enabling them to apply their local knowledge and expertise to our elaborated conceptual framework in ways that support rich, endogenous conceptualisations of assets and the thoughtful, locally relevant design of local mapping processes that explore value and use. Third, to illustrate the kinds of rich empirical insights it is possible to generate, the case of Ayesha (pseudonym), a 14-year-old student with a difficult relationship with schooling, is briefly explored. This illustrative case helps to demonstrate: (i) how our endogenous assets mapping approach is able to surface individualised accounts of the ways that particular young people living in a disadvantaged neighbourhood identify, value and use assets; (ii) the extent to which our approach might be used in ways that enable conceptually derived understandings of assets to be empirically elaborated; and (iii) how the contextually rich, empirical understandings of assets generated by our approach might valuably inform efforts to improve outcomes for young people living in disadvantaged neighbourhoods.

The conceptual framework
Conceptually, the endogenous assets-mapping approach we propose has three core components. First, it is concerned to recognise the assets young people themselves identify as accessible within their neighbourhoods. To do so, it begins by building on the pre-existing Assets-Based Community Development (ABCD) mapping framework developed by Kretzmann and McKnight (1993). Second, to address some of the core weaknesses in the ABCD framework, we draw on the work of capability scholars to explore how assets-approaches might take account of how young people actually value and use the assets they identify, and the outcomes they choose to pursue in doing so. Third, it considers what counts as valuable educational outcomes. Each element is now addressed in detail, starting with the justification for building on ABCD.

Building on ABCD
We begin by locating ABCD within the wider field of asset-mapping processes, justifying our decision to build on this in our approach. Asset-mapping processes span many disciplines, varying in the emphases they place on different kinds of assets and the assumptions they make about how assets may be used to improve outcomes. For example, in addition to community organising and FoK in education, these range from the 'sustainable livelihoods' approach commonly found in global development research and concerned with the conversion of assets into life chances (Park & Allaby, 2017), to public health's interest in using community resources to support social prescribing (Pescheny, Randhawa, & Pappas, 2018). Within this broad spectrum, ABCD appears particularly pertinent to our concerns, given its distinct focus on improving outcomes in high-poverty neighbourhoods in developed nations. It involves residents in a process of 'mapping' neighbourhood assets to create an inventory of local institutions, associations and individuals, which residents and professionals alike are then expected to draw upon to support the achievement of improved outcomes. As Kretzmann and McKnight (1993, p. 5) explain, it is intended to 'lead toward the development of policies and activities based on the capacities, skills and assets of lower income people and their communities'.
Over the last 20 years, ABCD has been popularised internationally in community development research by the US-based ABCD Institute (https://resources.depaul.edu/ abcd-institute/Pages/default.aspx), and has gained traction across Canada, Australia and Europe in particular. It has also been shown to be adaptable to addressing specific facets of community development in high-poverty neighbourhoods, for instance, building resilience (Mathie & Peters, 2014) and promoting social and economic inclusion (Foot & Hopkins, 2010;MacLeod & Emejulu, 2014). Notably, there are also signs that it has started to gain some purchase in education, specifically within the international community schools movement, where it has been advocated for schools' use by the combined US-based Institution for Educational Leadership and Coalition for Community Schools in their 2018 'Community School Standards'. To date, however, it remains almost entirely absent from scholarly literature in education, and in the rare instances when it has been applied, it has been adopted largely uncritically (Butterfield, Yeneabat, & Moxley, 2016;Forrester, Kurth, Vincent, & Oliver, 2020). In taking ABCD as our starting point for exploring neighbourhood assets that can support positive educational outcomes, we will now address this limitation by critically considering its utility in educational research.

Mapping assets using ABCD
As already noted, ABCD uses a framework for mapping assets within disadvantaged neighbourhoods that assumes that assets fit into three broad categories -individual, associational and institutional -that are then broken down into more detail. Individual assets are divided into three dimensions. First, every resident is considered to be talented or skilled in some way, with possible assets ranging from being caring, to having artistic flair, to formal employability-related skills. Second, individuals' 'community skills' are considered, including 'the kinds of community work the person has participated in and . . . what kind of work they would be willing to do in the future' (Kretzmann & McKnight, 1993, p. 15). Third, individuals' entrepreneurial activity within their neighbourhoods is considered, with this seen as integral to involving residents in creating local economic opportunities (McKnight, 2005). As regards associational assets, ABCD aims to identify resident-led local associations, informal networks and professionally led services. As Kretzmann and McKnight (1993, p. 109) explain, 'For purposes of local community building, an association is usually a group of local citizens joined together with a vision of a common goal.' Finally, institutional assets refer to visible, physical places and structures, which typically house formal organisations and professional services, for example, hospitals, schools, churches, libraries and shops. Kretzmann and McKnight also note that while the tangible nature of institutions makes them easy to identify, these can be extremely complex to understand fully as assets. This is because 'in order to accomplish their purposes, they bring together people with different skills, buildings, equipment, budgets, and relationships with other institutions. Institutions, then, are themselves collections of assets' (Kretzmann & McKnight, 1993, p. 171).
To implement this mapping framework, Kretzmann and McKnight suggest that local residents and professionals should undertake this jointly, often by walking around and observing the neighbourhood together. They suggest this is essential to enable dialogue and so guard against professionals imposing their own understandings of what constitutes a valuable asset on residents. In seeking not to limit the assets residents can identify, ABCD perhaps remains open to the charge that assets can be defined so broadly as to be potentially all-encompassing. Set against this, however, it does provide a clear framework to focus attention on particular kinds of assets and, hence, it supports the systematic identification of a comprehensive -though not all-encompassing -range of assets, while, importantly, avoiding prescribing a list of assets that disadvantaged neighbourhoods should have or must develop. Indeed, reviewing the available scholarly evidence on studies employing ABCD-style mapping exercises, there is evidence to suggest that using its broad categories to identify assets can generate 'a large inventory of resources that can be used to help strengthen the community ties and improve quality of life for residents of the neighborhood' (Jasek-Rysdahl, 2001, p. 314).
From a professional and service perspective, the dialogic nature of joint mapping activities has also been reported to have the potential to enrich, and perhaps even to steer, professional practices and processes, by enabling them to work from the basis of a coproduced understanding of assets and needs within local communities (Foot & Hopkins, 2010). On the rare occasions when ABCD mapping approaches have been used in educational research, children and young people have also been shown to engage easily with its broad categories. In Forrester et al.'s study (2020), for example, young people used ABCD as a guiding framework to identify the range of ways in which their school might function as a neighbourhood asset.

The limitations of ABCD
While ABCD's categories allow a wide range of assets to be systematically identified, it does also have some notable limitations. For instance, it is not entirely clear on what warrants its categorisations of assets rest, and to what extent these are research-informed, driven by professionally led improvement concerns, or derived from local residents' endogenous understandings. What we can say, however, is that ABCD's internationally widespread use indicates that as a heuristic device, its categories have, at the very least, strong face validity, and do not unduly limit residents' opportunities to identify assets within their particular contexts.
A further limitation to ABCD is that it focuses squarely on the mapping of assets, simply assuming that these can and will be used to support local reforms to improve outcomes. However, whether people actually value or use the assets identified, and for what purposes, does not appear to be considered. Neither are the range of structural and individual factors that might influence this, positively or negatively. This points to a simple assumption at the heart of ABCD: that the process of mapping assets -of acknowledging their existence and pinpointing their location -will be enough to catalyse activities to ensure the use of assets to improve outcomes. Yet even if, for example, young people identify a local library as an institutional asset, it does not follow that they will engage with its resources and improve their reading and so too their educational outcomes. In this sense, it seems that ABCD may operate with a tacit view of communities as homogeneous. For instance, while there is an understanding that assets are to some extent subjective, the final neighbourhood-level inventory has no clear process to capture differences in residents' world-views or lived experiences. Furthermore, if the inventories of neighbourhood assets which result from ABCD are not in some way connected to the wider social dynamics in which these neighbourhoods and their residents are embedded, there is a risk of overlooking the ways in which the dynamic interplay of societal structures, as well as individual and collective agency, might enable or constrain engagement with assets.
Given these limitations, simply importing ABCD into studies seeking to understand the neighbourhood assets available to young people would be insufficient. In developing our endogenous assets-mapping approach we have therefore sought to elaborate ABCD by drawing on understandings of assets from sustainable livelihoods approaches (SLA). Although less concerned with mapping assets -for which ABCD arguably has greater utility -SLAs place greater emphasis on exploring the capabilities the use of assets might enable in individuals, and the freedoms individuals require for this (Bebbington, 1999). Bringing this to bear can valuably support the consideration of how the assets identified by ABCD processes might be valued and used by (different) young people.

A capability approach to understanding assets
In common with ABCD, the capability approach rejects a deficit-oriented stance towards people and places experiencing poverty. Instead, it considers poverty and disadvantage in terms of the lives people can actually lead, and the freedoms they actually have (Sen, 2001). Applied to the mapping of assets, this suggests neighbourhood assets, as identified through ABCD, are not important in and of themselves; rather, their importance lies in what they can enable people to 'be' or 'do', in order to achieve outcomes they have reason to value. Thus, it is not enough simply to locate assets as ABCD does. As Sen (2001, p. 80) explains: 'Given interpersonal diversity, related to such factors as age, gender, inborn talents, disabilities and illnesses, the commodity holdings can actually tell us rather little about the nature of the lives that the respective people can lead.' To understand this more fully, it is important to engage with the concepts of 'capabilities' and 'functionings', which Sen (1987, p. 36) defines as follows: 'a functioning is an achievement, whereas a capability is the ability to achieve.' Sen suggests capabilities can also be understood as a type of instrumental freedom, by enabling individuals to 'achieve alternative functioning combinations (or less formally put, the freedom to achieve various lifestyles) ' (2001, p. 75). Applied to our endogenous assets-mapping approach, these concepts invite consideration of the: (i) educational outcomes and wider life chances (i.e. functionings) young people might value and want to achieve; (ii) opportunities they might have to help them do so (i.e. capabilities); and (iii) resources (i.e. assets) they might draw upon to access these opportunities and achieve their ambitions. They therefore extend ABCD by allowing connections to be made between the assets identified and the pursuit of outcomes.
In addition, rather than simply assuming, as in ABCD, that the identification of assets will lead to their use and then to positive outcomes, a focus on capabilities and functionings contends that neighbourhoods can hold assets that young people identify, but do not then value or use. For instance, they might identify a library as an institutional asset, but not engage with it. To reflect this, in our endogenous mapping approach, we term assets that are identified and not used as 'potential' assets, while, by contrast, those assets that young people identify, and then value and use to enhance their capabilities to achieve desired functionings, we term 'realised' assets. This distinction is important in demonstrating that even if ABCD suggests a neighbourhood is asset-rich, it may make little difference to the opportunities and outcomes actually valued and realised by young people. The emphasis on capabilities and functionings as being individual traits also presents a possible means to overcome ABCD's tendency to homogenise communities. It has been important for our endogenous-assets mapping approach to recognise that one person's potential asset may be another person's realised asset; the same assets may be realised in different ways by different people, possibly also in pursuit of different ends; and that different assets may be used by different people to pursue the same goals.
Alongside this nuanced and fine-grained understanding of individuals' value and use of assets, it has been equally important not to lose sight of ABCD's neighbourhood-level focus. This is likely to be particularly important in considering the freedoms which young people have 'to achieve whatever the person, as a responsible agent, decides he or she should achieve' (Sen, 1985, p. 204), and the extents to which their choices and the attendant use of assets are freely made, or are in some way constrained by structural factors within their environments. As Sen reflects: 'The freedom of agency that we individually have is inescapably qualified and constrained by the social, political and economic opportunities that are available to us. There is a deep complementarity between individual agency and social arrangements ' (2001, p. xi). This hints at the complex spatial dynamics at work in local neighbourhood contexts, where the numerous characteristics of place, and the challenges and opportunities arising from these, shape the people who live there, while they, in turn, also simultaneously shape the place (Lefebvre, 1991). Nussbaum (2000, p. 31) reflects that in such complex ways, local contexts 'affect the inner lives of people: what they hope for, what they love, what they fear, as well as what they are able to do.' An endogenous assets-mapping approach must therefore also locate the exploration of assets within the spatial dynamics of disadvantaged neighbourhoods, and how these may constrain or enhance the choices available to young people, as well as influence what they see as valuable. It also draws attention to the importance of maintaining an understanding of the individual within the neighbourhood, rather than -as ABCD tends towards -treating neighbourhoods as homogeneous.

Positive educational outcomes and wider life chances
Finally, given our specific interest in the use of assets to support positive educational outcomes and wider life chances, we have also considered what is meant by these terms. Conscious of critiques that asset-mapping processes can, in practice, be formulated by professionals to support professionally determined agendas, we have sought to allow young people the freedom to define the capabilities and functionings they have reason to value. Nonetheless, there are still some a priori considerations that have informed our approach in this respect.
First, policy discussions about how to improve poor outcomes in high-poverty neighbourhoods have, as Kerr and Ainscow (2017, p. 9) reflect, tended to 'become fixated on a narrow range of measured attainments, confusing these with the purposes of education.' We have therefore sought to recognise and allow for 'the oft-neglected distinction between education and schooling' (Lawson, 2016, p. 10), the latter being focused more strongly on school-centric processes and academic attainment, while the former is not bound by the constraints of the school system. Indeed, in considering the capabilities and functionings 'human beings require for a flourishing life' (Nussbaum, 2011, p. 125), the capability approach points to a much broader conceptualisation of the purposes of education. Walker (2006), for instance, formulates these around: autonomy; knowledge; social relations; respect and recognition; aspiration; voice; bodily integrity and health; and emotional integrity and emotions. Although such lists are contested, they nonetheless illustrate the importance of being open to such broad understandings, and not simply conflating education and wider life chances with attainment and progression to further and higher education.
We have also deliberately focused on the achievement of what we refer to as 'positive' outcomes, explicitly taking the stance that some assets, and the capabilities they may be used to enable, may place young people at risk of significant harm -even if young people themselves relate these assets to the achievement of desired functionings. This decision was especially important given that our study was embedded within a secondary school and so was itself limited by the safeguarding norms adhered to and required by the school. Within such parameters, we were guided by Sen's notion of 'reasonableness', a principle within the capability approach (see Sen, 1985), that invites individuals to submit what they value to reasoned scrutiny, and only those values that are agreed upon through a collective process can be said to be reasonable. Engaging with both senior school staff and the group of pupil co-researchers working on this project in a collective, dialogic process, we therefore decided that it was 'reasonable' to restrict our focus in this approach to what we collectively understand as 'legitimate' assets and activities. These are legal, would not reasonably be considered to pose a risk of significant harm (British Educational Research Association, 2018) and they also directly support young people's educational outcomes (as a key aim of this study). Of course, this decision is not without potential limitations, and there can be no doubt that in restricting our focus to 'positive' or 'legitimate' assets as defined earlier, we may have missed part of the young people's stories and/or in effect, restricted their voices. However, we did take steps to address this potential limitation, and engage in a process of collective reasoning as to what might constitute a 'positive' asset, through the inclusion of pupils as researchers in this study, working alongside university researchers. We will discuss the characteristics of this participatory process, as well as notions of co-design further, in later sections of this paper.

Summarising the core conceptual elements of an endogenous assets-mapping approach
In summary, the endogenous assets-mapping approach we have developed firstly maintains and builds on ABCD's broad yet detailed categories of assets. These give the approach sufficient structure and focus, while still allowing scope for young people's endogenous understandings to shape the identification of assets. Secondly, our approach has significantly extended and elaborated the insights ABCD can generate, inviting consideration of whether the assets identified by ABCD-style mapping are potential or realised, and the capabilities (opportunities) and functionings (outcomes arising from these opportunities) assets may be used to support. This also helps to counter ABCD's tendency to decontextualise and homogenise communities, by drawing attention to the differing configurations of capabilities and functionings that may arise through individuals' use of assets and the contextual factors influencing this. Thirdly, our approach notes the distinction between education, understood broadly, and a narrower academicattainment agenda, inviting consideration of the former, while not precluding the latter.

Utilising an endogenous assets-mapping approach in practice
To demonstrate the empirical utility of our conceptual approach, we now explain how this has informed asset-mapping activities with young people (aged 11-15) living and attending school in Hollyburgh, identified against national monitoring data as a highly disadvantaged neighbourhood in England.

Research design
In practice, one of the authors (Forbes) worked as an embedded researcher in Hollyburgh's only secondary school over a two-year period. During this time, she worked with a group of 10 pupil co-researchers (Fielding, 2001), aged 13-14, to operationalise the endogenous assets-mapping approach. This study was granted ethical approval, by the School of Education, Environment and Development Ethics Committee at the University of Manchester. In line with this approval, informed consent was obtained from all coresearchers and their parents. The decision to work with pupil co-researchers, following Fielding's (2001) notion of students as co-researchers, was deliberate, explicitly recognising them as uniquely positioned to help access and interpret their peers' understandings of assets, and to allow them some co-ownership of the study. Given that participatory research with children can take varied forms (see Groundwater-Smith, Dockett, & Bottrell, 2015), we were specifically engaged in participatory consultation (see Lansdown, 2005, p. 15). This is where young people are involved in co-developing research processes based upon an initial adult-defined agenda -namely the identification of assets young people value and use to support their education and life chances. The co-researchers were all self-nominated, responding to a 'job description' that included a willingness to work on a research study and to direct research activities with their peers, as essential requirements. Although the final group of pupil co-researchers had only one male pupil, it was otherwise diverse, including three pupils with English as an Additional Language, three receiving pupil premium funding (state funding for pupils from low-income backgrounds), and pupils from White British, White European, Chinese, Jamaican and Nigerian backgrounds. This diversity supported the participation of a full range of young people and helped to counter any risks of unintentionally treating them, or the assets they identified, homogeneously.

Developing the participatory approach
To help co-design the study, the co-researchers took part in six two-hour training sessions that were timetabled into the school day and thus, given the status of a robust academic learning activity. Although these sessions were professionally instigated by the embedded researcher -given the academic level of the content covered, including research design, methods training and so forth -they were also designed to enable the young people to discuss and explore the concepts of assets, and positive educational outcomes and life chances in an open-ended way. The aim here was for the pupil coresearchers and university researchers to develop a shared understanding of these concepts, to collectively explore the 'reasonableness' of these, as well as how best to explore them empirically with a wider sample of young people. In this sense, both the pupil co-researchers and university researchers were engaged in a form of boundary practice, engaging in ongoing dialogic processes that allowed them mutually to value and synthesise their different knowledge.
Through this, we began collectively to build 'radical collegiality' between the pupil coresearchers and university researchers, understood as 'a collegiality constitutive of a professionalism commensurate with the move towards a more dialogic form of democracy' (Fielding, 2001, p. 130). A practical example of this is how we -the pupil coresearchers and university researchers -co-produced a definition of the term 'asset' to ensure its accessibility to young people in Hollyburgh, and which could then be adopted in data collection activities. To do this, the embedded researcher facilitated a dialogic process with the co-researchers during which they collectively explored ABCD, capabilities, positive educational outcomes and wider life chances from their different perspectives. This resulted in a definition of assets, developed by the pupil co-researchers through a professionally instigated and facilitated process, that broadly reflects ABCD's categories of individual, associational and institutional assets, and basic notions of capabilities and functionings, in relation to their understandings of what would constitute positive educational outcomes and wider life chances. Their agreed definition, which was then used through data collection activities with other young people, and in other phases of the project with diverse stakeholders, was as follows: An asset is a service, place or person that enables a young person or people to achieve qualifications, develop skills, talents and personal attributes to support their personal, professional and financial aspirations and wellbeing.
Educational outcomes refer to the achievement of recognised qualifications, the development of personal skills and qualities, as well as positive social and emotional wellbeing. Wider life chances refer to the achievement of personal, professional and financial aspirations and wellbeing.
By professional, we mean secure employment in a meaningful, enjoyable job, based on one's own values. By financial, we mean the economic means to support oneself and lead a meaningful, enjoyable life, based on one's own values. By personal, we include social, emotional and physical health.

Data collection
Once we had a shared understanding of key concepts, the pupil co-researchers designed three research activities, which were developed into a series of parallel focus groups managed and led in practice by the pupil co-researchers, supported by the embedded researcher and school staff. The first was a photo-elicitation exercise (Miles & Howes, 2014) inviting participants to share their responses to photographs of tangible assets of Hollyburgh, in order to surface and capture the young people's perceptions of Hollyburgh and the challenges and opportunities their neighbourhood environment might present. The second was a 'visual mapping exercise' (Amsden & Vanwynsberghe, 2005) where each participant drew and annotated a personal 'map' of the assets in Hollyburgh that they valued and how they used these. To help ensure these provided robust and comparable data, the co-researchers firstly shared and explained their understanding of key concepts of assets, educational outcomes, as underpinned by the notion of 'reasonableness'. They subsequently used a series of ABCD-inspired questions to prompt participants to think about the people, services and places they might engage with for different purposes, asking, for instance: 'where do you go in your spare time?', 'who do you spend it with?', before asking a series of capability-inspired questions to support the process of annotating the maps, for instance: 'what opportunities does . . . provide?', 'what do you hope to achieve through this opportunity?' The final task involved a group discussion, guided by the co-researchers, about the sorts of assets that might support participants' future plans. In total, 225 young people, drawn from Years 7-10 (aged 11-15) took part in these focus groups, and all activities were done in the same order to ensure comparable data across the dataset. In addition, the co-researchers also developed personal assets maps and were interviewed individually about these. This was to ensure that the individual nature of capabilities and functionings could be considered in greater depth than the small group activities might allow, and as a further guard against treating Hollyburgh's young people as a homogeneous group.

Data analysis
Data were analysed thematically. Due to reduced time available to the co-researchers in the school timetable to work on the project, the embedded researcher did the initial work of transcription, digitising and some first-cycle coding to analyse data in relation to concepts from the literature. This created the foundations for the co-researchers to engage in second cycle coding, i.e. a less formalised grounded analysis, incorporating more open descriptive coding, within a discussion group format. Although the data analysis was a professionally instigated process, the young people were active participants in shaping final understandings.

Findings
To provide some indication of the kinds of rich, nuanced insights generated by the endogenous assets-mapping approach, we now report the illustrative case of Ayesha (pseudonym). We focus on Ayesha as her story illustrates both individual distinctiveness (her own particular blend of individual capabilities and agency interacting with school and neighbourhood structures), as well as illuminating some of the wider patterns in the data. On the one hand, her behaviour in school, manifesting as defiance and sometimes physical violence towards staff, appears to exemplify deficit discourses around young people in disadvantaged neighbourhoods who are disengaged from schooling and whose poor behaviour places them at risk of exclusion. Ayesha was, for instance, removed from a co-researcher session for physically threatening a teacher on her way there. On the other hand, her self-reported use of assets outside school suggested that she was also independently developing a very different self-curated narrative about herself as a learner who was achieving positive outcomes.
Despite her behaviour in school, Ayesha identified school as an asset for the opportunities it offers to gain qualifications. She explained her thinking thus: 'Education is the most important. If you don't have that, you're not going to go anywhere or be anyone. If you're not good at school, you're literally not going to go anywhere in life.' However, the tensions she experienced with school staff and peers, whom she believed held negative attitudes towards her, and her defiant responses to them, appeared to be undermining her ability to convert the capabilities on offer into valued functionings. School therefore appeared more of a potential than realised asset.
Outside school, however, Ayesha appeared to be engaged in a process of identifying and using assets, often quite strategically, to curate an alternative, positive future trajectory. Central to this, she identified her home as a realised asset providing a safe space, in which she is not judged negatively or subject to the anti-social behaviour she feels characterises Hollyburgh, where, she said, 'there are always issues, like fighting on the streets.' Ayesha then identified social media as a realised asset within the home, explaining how she has developed an international network of on-line friends who provide practical (for instance, sending resources) and emotional support for her ambitions to study to become a fashion designer: Big kids like me; they make clothes that are really ugly. So, I want to make something that's nice that all sizes can wear. I've made friends with other people with similar interests in fashion design from all over the world. I've some in America that I would love to visit.
Indeed, when interviewed, Ayesha was designing and making her own prom dress with her network's support.
Outside school and her home, Ayesha was acting in ways that were aligned with Hollyburgh's weak employment infrastructure and emphasis on micro-level entrepreneurship (see Gherhes, Williams, Vorley, & Vasconcelos, 2016) as a route to economic activity. Building on her knowledge of micro-businesses in the community, including her parents' very small-scale local Jamaican food take-away service run from their home, Ayesha talked about how she had been able to set up and now manages a hairdressing salon in an empty shop, where she works at weekends. This was allowing her to access various capabilities to develop business, management and interpersonal skills, which might also be converted into future functionings relating to her desired career. She explained her motivation as: 'Cos I want to get somewhere in life, I constantly try my best in everything that I do. I deal with the money, the employees, I have five . . . ' While Ayesha is atypical in some respects -in running her own micro-business aged 14 and her behaviour in school -in others, she reflects some common themes in the wider dataset. Pupils commonly expressed concerns that they would be judged negatively by professionals as a result of living in Hollyburgh, which they frequently described as 'dirty', 'rude' and 'violent', and that this might prevent them achieving their desired capabilities and functionings, regardless of the potential assets available. Like Ayesha, the majority responded by seeking out assets which they might use autonomously to curate more positive identities. While these were mainly focused upon young people's current wellbeing, there were a small number who, like Ayesha, were exploring future careers, including an aspiring journalist who was publishing fan fiction online.

Concluding discussion
Our premise at the start of this paper was that exploring the assets disadvantaged neighbourhoods hold, and which young people may use to pursue positive educational outcomes and wider life chances, could open new possibilities for intervening in spatial concentrations of poor outcomes. To support this, we have presented the development of a conceptually and methodologically robust endogenous assets-mapping approach that addresses some of the critiques that we note with regard to existing assets-mapping processes, and notably ABCD. While developed partly in situ in Hollyburgh, the approach is transferable to diverse contexts nationally and internationally. Indeed, mapping assets across diverse contexts appears an important future task. Were this to reveal some core assets which young people similarly value and use across multiple contexts, it might point to ways in which policy makers could strengthen particular assets through universal approaches, while also indicating where locally bespoke and individual supports might prove most effective. If, for example, the 'Community School Standards' cited earlier were to adopt an endogenous assets-mapping approach rather than the more simplistic ABCD, this could provide a ready vehicle for such a wide-scale exploration.
The rich story of Ayesha's use of assets that our endogenous assets-mapping approach has elicited points to a number of challenges and opportunities for policy makers and practitioners to consider. For example, Ayesha's case suggests that young people with a clear sense of desired longer-term destinations might be able, purposefully and autonomously, to seek out very particular forms of assets to help pursue their interests. It would seem that some young people, potentially including those at risk of exclusion, might be starting creatively to develop their own social mobility strategies outside school, albeit without discretely naming or conceiving of them as such. What Ayesha's case also strikingly shows is that narrowly conceived discourses of school success can be so powerful in these young people's lives that they may not fully recognise what their selfcurated strategies and use of assets are enabling them to do. Ayesha, for instance, appears unable to connect her in-school and out-of-school experiences; while outside school she is highly entrepreneurial, she still believes she will, in her own words, fail 'to get anywhere in life' if she is 'not good in school'. This appears to speak to broader arguments that a socially just education system must find ways to valorise vocational knowledge and opportunities (see Francis, Mills, & Lupton, 2017) -and using an endogenous assetsmapping approach to generate insights from students like Ayesha, and conceptualise these through the input of pupil co-researchers, could help to inform what this might look like. It is important to note that Ayesha's use of assets was facilitated by clear career goals and role models in her family and neighbourhood contexts. For students without these, highquality, personalised careers education, underpinned by strong relational processes that might help to counter difficult relationships with school, could prove an important first step.
However, policy cannot simply assume that young people will have sufficient access to assets, or the ability to realise these, to meet all their learning needs, and secure positive post-school destinations. Policy makers may well have to find suitable ways to help build a neighbourhood's assets-base, but without tacitly imposing deficit narratives on young people and the places where they live. An endogenous assets-mapping approach could help policy makers and professionals to understand how best to approach this, and to resist the urge to reshape endogenously identified assets to meet their own professional agendas.
To conclude, the newly developed endogenous assets-mapping approach advanced in this paper has the potential to create a point of departure in the ways that policy makers and professionals view and act upon young people in high-poverty neighbourhoods. It could help to shift their gaze from a deficit stance, underpinned by a narrow view of education as schooling, towards a broader and more richly contextualised understanding of education, and of how locally available assets may be conceptualised and drawn upon by young people to support a diversity of positive outcomes.

Disclosure statement
There are no relevant financial or non-financial competing interests to report.