Retention work needs belonging evidence, not just a single score
Updated Apr 22, 2026
Retention work usually starts too late. By the time continuation figures arrive, many students have already spent weeks or months feeling less connected, less supported, or less sure they belong. That is why Lan Thi Nguyen, Matthew McDonald, Thao Thi Thu Nguyen, Carmel Foley, Rebecca Kechen Dong and Simon Darcy's Journal of Further and Higher Education paper, "The implications of sense of belonging for student retention in higher education: a systematic literature review and research agenda", matters for UK universities using student voice to understand who is connecting, who is drifting, and which students may be at greatest risk of leaving.
Context and research question
Belonging now sits near the centre of higher education retention work. Universities link it to continuation, engagement, wellbeing, and student success, yet they do not always define it precisely. In practice, belonging can mean peer connection, academic fit, institutional recognition, confidence in navigating systems, or some blend of all four, which echoes recent work on what new students need to feel they belong and stay well. That ambiguity matters because retention strategies become vague when the evidence base is vague too.
Nguyen and colleagues tackle that problem through a systematic literature review of 66 original empirical studies on belonging and student retention in higher education. Their question is straightforward and useful for UK Student Experience, Planning, and Market Insights teams: what does the research actually show about how belonging relates to retention, and what evidence is still missing if universities want to act on it well?
Key findings
The first finding is that the literature does not ignore belonging, but uses it unevenly. The review identifies a substantial body of work linking belonging to student retention, yet the field remains conceptually scattered. Studies ask related questions, but they often define belonging differently, measure it differently, and connect it to retention through different lenses. For institutions, that is a practical warning: a term that appears throughout strategy can still carry too many meanings to guide action well.
The authors describe a "fragmented state of knowledge on the role of belonging in student retention". That matters because fragmented evidence often leads to over-confident interventions. Universities may treat a small shift in one belonging item as strong retention evidence, even when the wider literature suggests the construct is broader, more contested, and more context-dependent than that.
The second finding is that belonging needs deeper theoretical treatment, not just wider use. The paper argues that research on retention often treats belonging as an obvious good without being clear enough about what kind of belonging is actually under discussion. Is the issue academic recognition, social connection, institutional fit, identity safety, or practical inclusion in everyday university life? If those ideas are blurred together, the evidence becomes too thin to guide action, much as universities need to validate belonging measures before they benchmark them.
A third finding is that belonging should be understood as multidimensional and interdisciplinary. The review calls for approaches that draw together psychological, social, cultural, and institutional perspectives rather than reducing retention to a single motivational or behavioural problem. For UK higher education teams, that matters because continuation risk rarely belongs to one office alone. It often sits across teaching practice, assessment, communication, student support, induction, peer networks, and institutional culture.
The fourth finding is methodological, and it matters for equity work. Nguyen and colleagues argue that future research needs designs that capture students' subjective experiences and intersectionality more convincingly. In practical terms, that means universities should expect belonging to look different across mature students, commuter students, first-generation students, disabled students, and minoritised groups, not because those students are inherently less connected, but because the conditions around participation are different, much as our review of conditional belonging for minority ethnic STEM students also shows. Average results can therefore hide the very mechanisms retention work needs to see.
The review's most useful contribution is that it reframes belonging as evidence that needs interpretation, not a KPI that explains itself. That is a stronger starting point for institutions. It pushes teams to ask what kind of belonging they are measuring, whose experience is being captured, and how that evidence should shape retention decisions in practice.
Practical implications
For UK universities, the first implication is to define what the belonging evidence is meant to do. If the goal is earlier retention risk detection, one annual question is not enough. Teams need repeated light-touch listening at key transition points, such as induction, first assessment, pre-winter break, and re-enrolment, which is also why tracking student belonging over time matters. The benefit is earlier intervention, before disconnection hardens into withdrawal.
Second, institutions should pair belonging measures with open-text prompts that explain the pattern. A score can show that connection is weaker for one group or at one point in the year, but it rarely shows why. Add prompts such as "What has made it easier or harder to feel part of your course this term?" and analyse the responses systematically. A defensible method such as our NSS open-text analysis methodology helps teams distinguish whether the issue is peer culture, assessment pressure, timetable friction, staff responsiveness, or something else. That turns retention work into a clearer operational agenda, rather than a vague concern.
Third, universities should segment belonging evidence much more deliberately. Whole-cohort averages can make an institution look stable while one group is struggling with access, confidence, safety, or recognition. Review patterns by student group, mode, stage, and transition point, then check whether the same themes recur in qualitative comments. Student Voice Analytics fits naturally here because it helps universities compare belonging-related comments at scale with one reproducible method. That makes continuation work more specific and less anecdotal.
Finally, retention teams should treat belonging as a cross-functional evidence problem. Planning teams may hold the continuation figures, student support teams may see help-seeking behaviour, and academic teams may hear the course-level concerns first. Belonging evidence becomes more useful when those signals are read together rather than in isolation, following the same logic as benchmarking and triangulating student survey evidence. The practical benefit is not just better reporting. It is faster, more targeted action for the students most likely to feel the gap first.
FAQ
Q: How can a university use this paper without launching another large survey?
A: Start with the feedback routes you already have. Add one belonging item and one open-text prompt to existing pulse surveys, induction check-ins, personal tutoring reviews, or student support follow-ups. Then use a simple governance process, such as the student comment analysis governance checklist, so teams know who reviews the evidence, how it is segmented, and when it should trigger action.
Q: What are the methodological limits of this paper?
A: This is a systematic literature review and research agenda, not a new longitudinal dataset or a sector benchmark. Its strength is synthesis: it shows where the evidence on belonging and retention is strong enough to use, where it is conceptually thin, and where future work needs to go deeper. Universities should use it to sharpen their own listening strategy, not as a substitute for local evidence.
Q: What does this change about student voice practice more broadly?
A: It shifts student voice from a retrospective exercise to an earlier warning system. If belonging is part of retention, then comments about isolation, confusion, not fitting in, or weak support should be treated as continuation evidence, not soft background context. That gives universities a better chance of acting while students are still deciding whether they can remain connected.
References
[Paper Source]: Lan Thi Nguyen, Matthew McDonald, Thao Thi Thu Nguyen, Carmel Foley, Rebecca Kechen Dong and Simon Darcy "The implications of sense of belonging for student retention in higher education: a systematic literature review and research agenda" DOI: 10.1080/0309877X.2025.2585161
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.
UK-hosted · No public LLM APIs · Same-day turnaround