Conditional Belonging: What Minority Ethnic STEM Students Tell UK Universities

By Student Voice AI

Updated Mar 31, 2026

Belonging is easy to promise and much harder to evidence in student feedback. This open-access paper in Higher Education shows how minority ethnic STEM students can experience belonging as conditional, which gives UK universities a more useful way to interpret comments about inclusion and engagement. Chiu, Wong, Murray, Horsburgh and Copsey-Blake explain what that looks like in practice, and why it matters for teams trying to act on student feedback. [Paper Source]

Context and research question

Universities commonly frame belonging as something they can promote through representation, messaging, and student experience programming. But for many students from underrepresented backgrounds, belonging can feel less like a stable sense of being at home and more like an ongoing negotiation: when do I feel accepted, and under what conditions?

Chiu et al. ask how minority ethnic students experience and construct belonging, and how that belonging can be “conditional” (dependent on context) and “conditioned” (shaped by prior experiences and institutional norms). The study sits within a two-year qualitative project. The wider dataset includes semi-structured interviews with 110 STEM students, conducted online between June and December 2020. This paper draws on 72 interviews with minority ethnic students across two pre-1992 English universities described as having international reputations and “elite” status, using a social constructionist approach and iterative coding in NVivo. For practice, that matters because it shifts the question from “Do students belong?” to “What conditions make belonging more or less possible?”, a framing that complements later work on what ethnic-minority students mean by belonging.

Key findings

The core contribution is conceptual: belonging is not a yes/no state. For these students, it can depend on performance, on who is in the room, and on the perceived risk of being read through stereotypes. That makes the paper especially useful for teams analysing comments, because it points you toward moments and contexts, not just headline scores.

First, the authors describe how institutional elitism and competitiveness can make belonging feel earned rather than granted. When students feel they must continually prove they “deserve” their place, ordinary setbacks, confusion, or academic struggle can start to look like evidence that they do not fit. That can narrow participation and reduce help-seeking.

Second, students’ accounts show how ethnicity shapes academic belonging through everyday interactions. Belonging in seminars, labs, and group work can be undermined by being one of the few minority ethnic students in a space, experiencing exclusionary behaviour, or feeling pressure to represent a whole group. The result is often heightened vigilance. Students assess whether a space is “safe enough” to ask questions, make mistakes, or contribute.

As the authors put it:

"students’ construction and negotiation of belonging can be ‘conditional’ and ‘conditioned’"

Third, the paper highlights tensions in social belonging. Students navigate a familiar double bind: seeking “safe” relationships with people who share experiences can be protective, while also being misread as self-segregation. Belonging is shaped not only by formal teaching, but by who gets included in informal networks, and by whether campus social spaces, norms, and events feel welcoming in practice.

Across these themes, the practical takeaway is consistent: conditional belonging changes how students participate. It affects confidence, willingness to speak, who takes leadership in group work, and whether students interpret institutional support as “for people like me”.

Practical implications

For UK HE teams, the paper points to interventions that are less about slogans and more about daily design:

Treat transition points as belonging-critical moments. Early-stage experiences can lock in assumptions about who “belongs” in a programme or institution. Design induction and early assessment to build confidence, so students are more likely to participate and seek help early, especially in light of later evidence on how welcome week attendance boosts peer belonging.
Make academic spaces actively inclusive. Train staff to recognise and respond to exclusionary dynamics, including microaggressions, and embed anti-racist practice in teaching, assessment, and group work design so students can contribute without self-censoring.
Build safe routes to seek help and report issues. If belonging is conditional, students may avoid help-seeking when they fear being judged. Clear support pathways, personal tutoring and tutor check-ins, and trusted reporting processes for exclusionary behaviour can lower that barrier and surface problems sooner.
Design social inclusion, not just social opportunity. Inclusive social spaces require practical choices about who is centred, how events are structured, and what “normal” looks like. Induction activities and campus events should signal inclusion in the details as well as the messaging.
Use student voice data to detect conditional belonging at scale. Ask open-text questions that invite students to describe when they feel able to participate, who they can be honest with, and what makes a space feel unsafe. Segment findings by student characteristics, then close the loop with visible action so students can see change.

Student Voice Analytics supports this by categorising belonging and inclusion signals in comments, alongside related themes such as assessment fairness, teaching behaviours, and peer dynamics, and benchmarking how experiences differ by cohort and demographic. That gives teams evidence they can use to prioritise interventions and track whether conditions improve over time.

FAQ

Q: How can universities identify “conditional belonging” in student feedback?

A: Look for patterns in when and where students feel comfortable participating, not only whether they report “belonging” in general. Open-text prompts such as “When do you hold back in class?” or “What makes group work feel safe or unsafe?” tend to surface the conditions students are responding to. Analysing those comments by cohort and demographic can reveal whether belonging is stable or contingent for particular groups.

Q: What are the limits of this study’s method for understanding belonging?

A: The study uses qualitative interviews, which are well-suited to understanding lived experience and meaning-making, but they do not estimate prevalence in the way a survey would. The context also matters: interviews were conducted online during 2020, and the institutions are described as elite, which may shape how competitiveness and status are experienced. For practice, the implication is to triangulate: use interviews and focus groups for depth, and student voice survey comments for breadth and monitoring over time.

Q: What does “conditional belonging” change about how we interpret student voice metrics?

A: It suggests that averages can be misleading. A programme can look “fine” on high-level satisfaction scores while specific groups still navigate extra friction: avoiding certain spaces, self-censoring, or disengaging from peer networks. Free-text comments are often where this becomes visible, because students describe concrete situations, who spoke, what was said, and what was assumed, that do not translate into a single scale point.

References

[Paper Source]: Yuan-Li Tiffany Chiu, Billy Wong, Órla Meadhbh Murray, Jo Horsburgh, Meggie Copsey-Blake "‘I deserve to be here’: minority ethnic students and their conditional belonging in UK higher education" DOI: 10.1007/s10734-025-01469-1

Request a walkthrough

Book a free Student Voice Analytics demo

See all-comment coverage, sector benchmarks, and reporting designed for OfS quality and NSS requirements.

All-comment coverage with HE-tuned taxonomy and sentiment.
Versioned outputs with TEF-ready reporting.
Benchmarks and BI-ready exports for boards and Senate.

Prefer email? info@studentvoice.ai

UK-hosted · No public LLM APIs · Same-day turnaround