From Access to Belonging: Rethinking Widening Participation Metrics in UK Higher Education

Posted on by
Class, cultural capital, and social reproduction Method

Over the last two decades, widening participation (WP) has shifted from a niche concern to a central feature of higher education policy. Universities are now expected to demonstrate that they recruit students from “under-represented groups” and close gaps in retention and attainment. These expectations are made visible through a dense layer of numbers: application rates by postcode, continuation rates by demographic group, “good degree” gaps and graduate outcomes.

At first glance, these dashboards tell a reassuring story. Participation among students from lower socio-economic backgrounds has risen in many countries, and in the UK the proportion of young people entering higher education has reached record levels (Archer et al., 2003; Boliver, 2013). In institutional reports and sector briefings, such trends are often presented as evidence that the system is becoming fairer and more meritocratic.

The broader context of massification and marketisation reinforces this policy narrative. As higher education has expanded, institutions have been encouraged to position themselves as “open to all with talent”, often framed through discourses of aspiration, social mobility and choice (Ball, 2012; Burke, 2012). Within this framing, WP becomes proof that universities are doing their part: if more students from disadvantaged backgrounds enter higher education, the system must be working.

Yet sociological work on inequality in education has long warned that access alone does not equal justice. Bourdieu’s analysis of schools and universities emphasised how education systems can incorporate more working-class students while still rewarding the dispositions, cultural styles and networks of the middle classes (Bourdieu & Passeron, 1977). The official story of inclusion can therefore sit alongside the continued devaluation of working-class cultures and knowledges.

More recent research on working-class and first-generation students suggests that they often feel “out of place” in elite institutions even when formal barriers have been lowered (Reay et al., 2010; Bathmaker et al., 2013). Students may gain access to prestigious universities but remain acutely aware that the institution was not designed with them in mind. They navigate curricula, pedagogic practices and campus cultures that are often organised around middle-class assumptions about language, leisure, professionalism and aspiration.

This tension is sharpened when access is coupled with high tuition fees and debt. In systems where students are framed as consumers, access can be reinterpreted as opportunity: those who do not take up university places are seen as lacking aspiration, and those who leave early are imagined as having made poor choices. The structural conditions that shape both entry and exit fade from view.

The key question, then, is not only who gets in, but what happens once students arrive – and how those processes do or do not show up in the data that drive institutional planning. WP dashboards tend to present access and success as linear stages in a pipeline; sociological work suggests that belonging disrupts this neat sequence.

What Student Data Usually Captures

To understand the limits of access-focused WP, it is worth pausing to consider the kinds of indicators that universities routinely collect and report. While details vary across systems, a fairly standard set of measures includes:

  • Recruitment and access: application, offer and enrolment rates by socio-economic background, ethnicity, gender, disability and region, sometimes broken down by subject or tariff band.
  • Continuation and progression: the proportion of students who remain enrolled after their first year and progress to the next level of study, often disaggregated by demographic and subject area.
  • Attainment: degree classification, often summarised through the proportion gaining a “good honours” degree; analysis may focus on awarding gaps between demographic groups.
  • Student voice, engagement and climate: responses to satisfaction, engagement and belonging or climate surveys, participation in co-curricular activities, and occasionally data on attendance or virtual learning environment use, sometimes pooled into multi-institutional datasets.
  • Graduate outcomes: employment or further study six to fifteen months after graduation, sometimes combined into a “positive outcome” or salary indicator.

From a planning or business intelligence perspective, these measures are valuable. They allow institutions to track patterns over time, identify groups with lower continuation or attainment and test whether particular interventions appear to close gaps. They also feed into regulatory frameworks, performance agreements and league tables, making them highly consequential for institutional reputation and funding.

However, the sociological question is not simply whether we have enough numbers. It is how these indicators enact a particular understanding of what matters in higher education – and what is left invisible. The selection of indicators is never neutral. It reflects assumptions about what counts as “success” and which experiences matter enough to measure at all, and it compresses complex trajectories into a small number of categories.

Dashboards are strongest at counting students in and out: who arrives, who persists, who leaves, who attains. They are weaker at capturing how students occupy institutional space: who feels recognised, who self-censors, who avoids certain corridors or offices. Belonging tends to appear only indirectly, through proxies such as survey items on “feeling part of the community” or patterns in retention. This makes it tempting to treat belonging as a “nice to have” factor that might nudge a continuation rate up or down, rather than as a structural issue grounded in classed and racialised power relations.

The temporalities of data also matter. Many indicators are annual or lagged: graduate outcomes, for instance, are captured months after students leave. The everyday work of building or undermining belonging in seminars, corridors and offices unfolds on a different timescale. By the time patterns crystallise in a dashboard, students’ experiences may already have moved on.

Belonging as a Sociological Problem

Belonging can sound like a soft concept, more at home in wellbeing campaigns than in sociological analysis. Yet recent work suggests that it is helpful to distinguish at least three dimensions: belonging as psychological need-satisfaction (feeling accepted and valued), belonging as relational and participatory (being embedded in networks and practices) and belonging as structural and institutional (having one’s histories, knowledges and futures recognised as legitimate). These dimensions are analytically distinct but intertwined in practice, and research on student experience has developed a robust body of work linking them to both identity and structure. They overlap with themes identified in recent reviews – such as connectedness, feeling safe, being valued and inclusion – even as this article places particular emphasis on structural and institutional recognition. The aim here is not to offer a full review of that literature but to sketch a framework that is adequate for rethinking WP metrics in context. In what follows, discussions of surveys and engagement data speak primarily to need-satisfaction and relational belonging, while analysis of habitus and institutional practices speaks more directly to structural and institutional belonging.

Tinto’s (1993) model of student departure, for instance, emphasises academic and social integration as key to persistence: students are more likely to stay when they feel connected to peers, staff and the intellectual life of the institution. Subsequent studies have extended this, showing that a sense of belonging is shaped by curriculum, pedagogic relationships and perceptions of institutional fairness (Thomas, 2012; Meehan & Howells, 2018). In parallel, Strayhorn (2012) conceptualises belonging as a basic human need in educational contexts, linked to motivation, engagement and persistence.

Bourdieu’s framework adds a deeper structural layer. Habitus – the embodied dispositions shaped by one’s upbringing – affects how at ease or awkward students feel in different institutional fields (Bourdieu, 1984). Cultural and social capital influence whether students know the “rules of the game”, from email etiquette and office-hour norms to which societies to join and how to navigate internships. Symbolic capital determines whose ways of speaking and being are recognised as legitimate and whose are marked as deficient.

Building on this, Reay and colleagues have introduced the idea of institutional habitus: the taken-for-granted values, assumptions and practices of a university or school (Reay et al., 2001). Institutional habitus shapes which students feel that the institution was “made for people like us” and which feel like guests with a temporary invitation. It is visible in everything from prospectus imagery and campus architecture to expectations about participation in seminars and extra-curricular life.

Belonging, in this sense, is not just about whether students feel comfortable. It is about whether the institution’s structures and culture recognise their histories, knowledges and future aspirations as legitimate. It involves processes of recognition and misrecognition (Skeggs, 1997), where some identities and trajectories are celebrated while others are tolerated, pathologised or rendered invisible. When belonging is framed this way, it becomes clear that it cannot be reduced to a single survey item.

Ahmed’s (2012) work on diversity policies is helpful here. She argues that institutions often talk about diversity as something they “have” or “do”, while leaving underlying power relations largely intact. Belonging can be drawn into similar institutional narratives: it becomes an outcome to be achieved through generic community-building initiatives, rather than a lens for interrogating whose presence requires adjustment from the institution and whose is assumed as the norm.

When Belonging Disappears into Metrics

How, then, does belonging appear – or disappear – in student data? Consider four examples.

Continuation as a proxy for belonging

Low continuation rates among particular groups often trigger concern about “lack of belonging”. This is not unreasonable: feeling out of place can certainly contribute to withdrawal (Yorke & Longden, 2004). Yet attrition is a blunt indicator. Students may leave for many reasons – financial pressures, caring responsibilities, mental health or course structure – that intersect with but are not reducible to belonging.

Moreover, continuation data usually treat withdrawal as a single endpoint. They rarely capture the cumulative micro-experiences that shape the decision to leave: the seminar where a working-class student’s examples are dismissed as “anecdotal”, the seminar where a racialised student’s experiences are treated as illustrative “case studies” rather than as knowledge, the meeting where a disabled student’s access needs are framed as burdensome or “special”, the office-hour conversation in which a first-generation student feels foolish for asking about postgraduate routes, the careers fair that seems designed for others. These experiences matter sociologically, but they leave only faint traces in the administrative record, perhaps appearing as a generic “personal reasons” code. These micro-experiences are patterned differently across intersections of class, race, gender and disability, producing distinct routes out of the institution that aggregate continuation data cannot distinguish.

The focus on continuation can also encourage deficit framings. If a particular group has lower continuation rates, the temptation is to ask what those students lack – resilience, academic preparedness, engagement – rather than what the institution is failing to provide in terms of recognition, support and flexibility.

Satisfaction surveys and the problem of voice

National and institutional surveys often include items on feeling part of a learning community or being able to “bring your whole self” to campus. These can provide useful comparative snapshots, and in some cases multi-item scales are explicitly designed to capture several facets of belonging or climate rather than a single global score, and systematic reviews now recommend multidimensional instruments for the complexity of higher education contexts. However, even carefully designed instruments are shaped by response bias, cultural differences in how people use Likert scales and small-numbers issues once data are disaggregated, and psychometric refinements cannot, on their own, resolve questions about power and recognition or fully capture structural and institutional dimensions of belonging.

From a critical perspective, the survey format can also individualise belonging. Students are asked to rate their own sense of inclusion, but not to evaluate the institution’s structures of recognition or the classed and racialised norms that circulate in lecture theatres and corridors. A strongly positive score from a majority group can mask sharp divides in experience for a minority group whose data are suppressed for confidentiality. In addition, open-text comments – where critiques of belonging often surface – are typically summarised or coded in ways that strip out their situated, affective texture.

Engagement analytics and the temptation to pathologise

Learning analytics systems increasingly track log-ins, clicks, submission patterns and attendance. These data are often used to flag “at risk” students, or more descriptively to map patterns of engagement across cohorts. While such tools – especially when used in “human-in-the-loop” or justice-oriented ways, for example as prompts for supportive conversations rather than automatic sanctions – can help staff intervene early, they also risk pathologising students whose patterns do not match institutional expectations of engagement.

For example, a commuter student working long hours may have patchy access records but strong offline study habits, while a student who “lurks” in online forums may learn effectively without visible participation. Without a sociological understanding of how class, disability, race and caring responsibilities shape time and space, engagement metrics can slide into moral judgements about who is “committed”. The algorithmic translation of complex lives into risk scores can obscure these dynamics.

Complaints, conduct and hidden signals of misfit

Finally, complaints data, disciplinary records and informal reports of “unprofessional” behaviour often contain signals about belonging that rarely make their way into mainstream dashboards. Students who challenge racist or classist comments may be framed as difficult; those who resist expected forms of deference may be labelled disruptive. Conversely, some students simply disengage rather than complain, leaving little trace beyond a quiet absence.

Across these examples, belonging is both everywhere and nowhere in the data. It is inferred from patterns, but rarely measured directly in ways that capture power and recognition. At the same time, there is a growing body of quantitative work that attempts to operationalise belonging and climate through survey scales and multi-institutional datasets, explicitly acknowledging their limits. These studies remind us that numbers can be used not only to police performance but also to interrogate it, if they are designed and interpreted within a critical framework. Numbers point to unevenness, but they do not in themselves explain how that unevenness feels or how it is sustained; for that, they must be brought into conversation with qualitative accounts and critical theory.

Bourdieu, Institutional Habitus and the Reproduction of Inequality

Placing student data within Bourdieu’s theory of practice helps clarify why access-focused WP may leave deep inequalities intact.

Fields such as universities are structured spaces where different actors compete for valued resources – grades, references, placements – using the capital they possess (Bourdieu, 1986). When more working-class or minoritised students enter these fields, the rules of the game do not automatically change. Instead, those rules may be reasserted through subtle expectations about conduct, dress, language and aspiration. These expectations are often experienced not as explicit rules but as feelings of comfort or discomfort, ease or strain.

Students bring habitus formed in particular classed, racialised and gendered contexts. For some, the university field resonates with existing dispositions: the kinds of cultural references used in lectures feel familiar; networking events mirror family social life; time and space on campus are easily available. For others, the field demands constant adjustment. They may learn to “play the game” but at the cost of feeling divided between home and university worlds, or between their own values and those they perceive as dominant (Reay et al., 2010; Reay, 2018).

In addition, institutional habitus means that universities develop their own logics of worth. A campus that celebrates study-abroad, extensive extra-curricular involvement and unpaid internships as markers of the “ideal student” is implicitly privileging those with sufficient economic and social capital to participate (Bathmaker et al., 2013). Those who cannot join in may appear in data as less engaged, less ambitious or even less resilient. In this way, institutional habitus shapes how different forms of capital are recognised and converted into institutional rewards.

These dynamics are rarely visible in dashboards. A working-class commuter student who juggles paid work, caring responsibilities and study may “count” as a success if they progress each year and attain a good degree. Yet the data will not show the extent of the labour required to remain in the field, or the emotional costs of negotiating a sense of belonging in spaces where they feel under constant scrutiny.

When WP success is defined through access statistics alone, these dynamics remain hidden. Institutions can show that they have diversified their intake while leaving the underlying structures of recognition largely untouched. The risk is that WP becomes a story about numbers rather than about transforming the social relations that structure higher education.

Rethinking Student Data Through Belonging

None of this means that student data are useless. On the contrary, careful analysis can highlight patterns that merit further investigation and action. The challenge is to approach data with a sociologically informed sense of what they can – and cannot – show.

A more critical approach might involve several shifts.

Looking for patterns within groups, not just between them

WP dashboards often compare outcomes for predefined categories: for example, students from lower versus higher participation neighbourhoods. This can obscure the diversity of experiences within those categories and the intersecting effects of class, race, gender, disability and age. Disaggregating where numbers allow, while respecting confidentiality, can reveal pockets of exclusion masked by overall improvements.

A focus on intersectionality is particularly important. For example, the experiences of working-class white women, middle-class Black men and mature care-experienced students are unlikely to be captured by broad categories such as “low socio-economic status” or “BAME”. Where sample sizes are too small for robust statistical analysis, qualitative work and targeted engagement can help surface patterns that would otherwise remain invisible.

Triangulating quantitative and qualitative evidence

Qualitative research, focus groups and narrative accounts are sometimes positioned as softer forms of evidence compared with “hard” numbers. Yet they are crucial for interpreting what patterns in continuation or attainment actually mean for students’ lived experiences of belonging (Burke, 2012). A spike in withdrawals from a particular course might look like an issue of academic difficulty in the data; interviews could reveal that students feel the curriculum invalidates their backgrounds or that assessment practices reward particular cultural styles.

Triangulation does not mean giving equal weight to every type of evidence in every context. Rather, it involves using different forms of data to ask better questions and to test emerging interpretations. For planning and BI teams, this might mean building structured opportunities for student narratives and staff observations to feed into the design and interpretation of dashboards, rather than treating qualitative insights as an afterthought.

Treating indicators as questions, not verdicts

Rather than treating a retention gap as proof that a particular group “lacks resilience”, planners can approach it as a sociological question: what aspects of institutional habitus are misaligned with these students’ habitus and capitals? What assumptions are built into teaching, assessment and support services? Data then serve as starting points for inquiry, not end points for judgement.

This might involve convening cross-functional groups – including students – to explore specific patterns revealed by dashboards. The aim would not be to explain away gaps but to understand how they are produced through everyday practices and structures. This process can itself surface issues of belonging, as students articulate where they feel recognised and where they experience symbolic violence.

Making space for measures of recognition and respect

Some institutions have experimented with indicators that go beyond participation counts: for example, monitoring staff diversity in leadership roles, tracking the inclusion of diverse authors in reading lists or capturing student perceptions of fairness in assessment. Such measures are imperfect, but they at least acknowledge that belonging is tied to recognition of knowledge and identity, not only to staying enrolled.

From a Bourdieusian perspective, these indicators can be read as attempts to shift the distribution of symbolic capital: to broaden whose voices are considered authoritative and whose experiences are centred. They are unlikely to be sufficient on their own – symbolic gestures can easily slide into box-ticking – but they create openings for more substantive conversations about whose habitus the institution is organised around.

Working with students as partners in interpreting data

Rethinking data through belonging also involves considering who is invited into conversations about evidence. Student representatives are often consulted on survey design or asked to comment on high-level findings, but they may have limited access to the analytical process itself.

Approaches that treat students as partners or co-researchers can shift this dynamic. For example, students might work alongside staff to interpret dashboard patterns, co-design local surveys or lead inquiries into particular aspects of belonging in their programmes. Such practices do not automatically equalise power, but they can disrupt the assumption that data interpretation is a purely technical task reserved for experts.

These shifts do not require abandoning dashboards. They require reimagining them as tools embedded in a broader critical conversation about power, habitus and institutional change.

Implications for Universities and for Sociological Work

For universities, taking belonging seriously means resisting the temptation to declare WP success once access targets are met. It involves asking more uncomfortable questions: Who feels entitled to occupy lecture theatres and common rooms? Whose accents, clothing and examples are treated as normal? Who hesitates before emailing a tutor or entering a departmental office? Whose knowledges appear on reading lists, and whose remain footnotes or case studies?

From an institutional perspective, this may require rethinking policies and practices that are often treated as neutral. Recruitment imagery, scholarship schemes, attendance requirements, placement expectations, the timing of office hours and the design of induction activities can all either foster or undermine belonging, depending on how they align with students’ lives and capitals.

For planning and business intelligence teams, the implications are similarly significant. Belonging cannot be reduced to a single metric, but it can be brought into view through the kinds of questions analysts ask, the way data are sliced and the narratives that accompany dashboards. Analysts can foreground uncertainty – for example, annotating charts where samples are small, explicitly marking “interpret with caution”, or pairing each quantitative slide with a brief qualitative vignette – and resist deficit framings that attribute gaps solely to student characteristics. They must do so, however, within organisational and regulatory environments that privilege comparability, standardisation and league-table visibility, leaving limited room to redesign indicators from first principles. The challenge is therefore not simply technical but political: working incrementally on what is measured and how it is interpreted, while navigating external demands for simple, headline indicators.

Within the constraints of national KPIs, for instance, a university might retain standard continuation and attainment charts for regulatory reporting but add a complementary “belonging panel” that combines disaggregated survey items, carefully caveated commentary on data quality and a rotating set of short student and staff narratives, overseen by a mixed staff–student working group. Such a panel would not solve structural inequality, but it would make questions of belonging more visible in routine governance. In this way, the panel would hold together indicators of need-satisfaction and relational belonging (survey items, narratives of connection) with glimpses of structural and institutional belonging (whose knowledge is cited, whose critiques are heard), rather than collapsing them into a single score.

For sociologists, the task is twofold. First, we can continue to theorise belonging in ways that connect micro-experiences with macro-structures, drawing on concepts such as habitus, capital, symbolic violence and institutional habitus. Secondly, we can engage with planning and business intelligence colleagues to shape how data are collected, interpreted and acted upon. This might mean collaborating on mixed-methods projects, co-designing survey items that attend to power or offering critical readings of dashboards.

Such collaborations do not mean abandoning critique. Rather, they take seriously the fact that decisions about which indicators to monitor, how to disaggregate them and what counts as improvement are themselves political. As Ball (2012) argues, contemporary universities operate within a performativity regime where metrics and targets shape behaviour. Sociological engagement can help ensure that belonging is not reduced to another box-ticked compliance exercise, nor to a branding slogan detached from everyday practice.

Belonging, then, is not an optional extra once participation targets are met. It is a central part of what it means to create a more just higher education system. Treating it as such demands conceptual clarity and a measure of methodological humility. It asks for a willingness to recognise the limits of existing data, to listen to the stories that do not fit easily into dashboards and to see WP as an ongoing, contested process rather than a completed project.

References

  • Ahmed, S. (2012). On being included: Racism and diversity in institutional life. Duke University Press.
  • Archer, L., Hutchings, M., & Ross, A. (2003). Higher education and social class: Issues of exclusion and inclusion. Routledge.
  • Ball, S. J. (2012). Global education Inc.: New policy networks and the neo-liberal imaginary. Routledge.
  • Bathmaker, A.-M., Ingram, N., & Waller, R. (2013). Higher education, social class and the mobilisation of capitals: Recognising and playing the game. British Journal of Sociology of Education, 34(5–6), 723–743.
  • Boliver, V. (2013). How fair is access to more prestigious UK universities? British Journal of Sociology, 64(2), 344–364.
  • Bourdieu, P. (1984). Distinction: A social critique of the judgement of taste. Routledge.
  • Bourdieu, P. (1986). The forms of capital. In J. G. Richardson (Ed.), Handbook of theory and research for the sociology of education (pp. 241–258). Greenwood.
  • Bourdieu, P., & Passeron, J.-C. (1977). Reproduction in education, society and culture. Sage.
  • Burke, P. J. (2012). The right to higher education: Beyond widening participation. Routledge.
  • Meehan, C., & Howells, K. (2018). In search of the feeling of belonging in higher education: Undergraduate students’ transition into higher education. Journal of Further and Higher Education, 42(7), 962–977.
  • Reay, D. (2018). Miseducation: Inequality, education and the working classes. Policy Press.
  • Reay, D., Crozier, G., & Clayton, J. (2010). Fitting in or standing out? Working-class students in UK higher education. British Educational Research Journal, 36(1), 107–124.
  • Reay, D., David, M. E., & Ball, S. J. (2001). Making a difference? Institutional habitus and working-class students. Sociological Research Online, 5(4).
  • Skeggs, B. (1997). Formations of class and gender. Sage.
  • Strayhorn, T. L. (2012). College students’ sense of belonging: A key to educational success. Routledge.
  • Thomas, L. (2012). Building student engagement and belonging in higher education at a time of change. What Works? Student Retention & Success programme report.
  • Tinto, V. (1993). Leaving college: Rethinking the causes and cures of student attrition (2nd ed.). University of Chicago Press.
  • Yorke, M., & Longden, B. (2004). Retention and student success in higher education. Open University Press.

Cite this article

Select a style and copy the citation into your document.