Sense of Belonging in Higher Education
The phrase “evidence-based” is used a lot in higher education. It’s also misused a lot — attached to programs that are well-intentioned but not actually grounded in rigorous research.
This post is for practitioners who want to cut through that. For retention coordinators, institutional effectiveness directors, student success deans, and faculty development leads who need to know: of all the things we could do to improve students’ Sense of Belonging, what does the research actually support?
The short answer:
More than you might expect. The field of belonging research in higher education has matured considerably over the past decade, and several intervention types now have enough evidence behind them to justify institutional investment.
Here’s what that evidence actually looks like.
The Research Foundation: Why Belonging Interventions Work
The theoretical basis for belonging interventions draws from multiple research traditions. Social belonging theory, as developed in the work of Gregory Walton and Geoffrey Cohen at Stanford, holds that belonging uncertainty — not knowing whether you fit in or whether others want you there — is cognitively costly. It consumes mental resources that would otherwise go toward learning.
In their landmark 2011 study published in Science, Walton and Cohen found that a single one-hour belonging intervention reduced the GPA gap between Black and white students at a selective university by 52% over three years.
The intervention involved having students read stories from older students describing how their own sense of not fitting in was temporary and normal — and then write an essay explaining how that experience resonated with their own.
That’s not a curriculum overhaul. That’s one hour.
The implication is significant: belonging uncertainty is not just a background psychological condition. It is an active academic handicap, and relatively targeted interventions can reduce its impact dramatically.
Intervention Type 1: Belonging Affirmations and Uncertainty Reduction
The Walton-Cohen intervention has been replicated and extended across institutional contexts, including community colleges and Hispanic-Serving Institutions (HSIs). A 2020 study published in Proceedings of the National Academy of Sciences found that belonging affirmation interventions had significant positive effects on persistence among first-generation college students, with effects that persisted across semesters.
What makes these interventions work: they normalize the experience of uncertainty. Students — especially those from underrepresented groups — often interpret their own struggles as evidence that they don’t belong. Belonging affirmations reframe those struggles as universal and transient.
Implementation consideration: These interventions are most effective early in a student’s enrollment and when delivered in a context that feels credible — from peers who have successfully navigated the same experience, not from administrative communications.
Intervention Type 2: Social Belonging Courses and Structured Reflection
Several institutions have moved beyond one-time affirmation activities toward semester-length courses built around belonging and identity development. These aren’t generic “college success” courses. They’re explicitly designed to help students process their academic identity, build peer relationships, and develop the metacognitive skills that support persistence.
The University of Texas at Austin’s “University Leadership Network” program, studied by researchers at the UT-Austin Population Research Center, produced statistically significant improvements in first-generation student retention over four years. Students who participated graduated at rates comparable to continuing-generation students — a gap the institution had struggled to close for years.
The belonging mechanism: Regular structured peer interaction, combined with explicit content on identity and belonging, reduced isolation and built the social integration that Tinto’s model identifies as a key persistence factor.
Intervention Type 3: Instructor-Level Belonging Signals
One of the most scalable — and underutilized — belonging interventions is improving what instructors do in the classroom. The research here is compelling.
A 2022 study in the Journal of Educational Psychology found that instructor-level belonging signals — using students’ names, acknowledging course difficulty without judgment, expressing genuine interest in students’ success — had a measurable effect on students’ sense of belonging in the class and their likelihood of seeking help when struggling.
This is where empathy training becomes an evidence-based intervention, not just a professional development preference. When instructors are trained to recognize signs of disengagement, respond with curiosity rather than judgment, and communicate genuine care for student success, the classroom environment changes in ways students notice and respond to.
Empathy training programs designed for higher education contexts — such as those incorporating structured perspective-taking, active listening practice, and recognition of micro-exclusions — have documented effects on classroom climate and student mental health indicators including sense of isolation and academic anxiety.
Intervention Type 4: Peer Mentorship With Structured Belonging Protocols
Peer mentorship is common in higher education. Evidence-based peer mentorship is less common.
The difference is in structure. Programs that pair incoming students with near-peers (one or two years ahead, from similar backgrounds) and give those mentors explicit training in belonging-facilitation — how to normalize struggle, how to invite connection, how to avoid advice-giving that inadvertently signals deficit — show significantly stronger outcomes than programs that simply pair students and hope for the best.
A study from the Journal of College Student Retention, Research, and Practice examining structured peer mentorship at a large public university found that first-generation students in the structured program showed 14-percentage-point higher second-year retention than those in the control group. The key differentiator was mentor training: mentors were taught to share their own belonging struggles, not just provide logistical support.
Intervention Type 5: Inclusive Design of Physical and Digital Spaces
Campus belonging is not only relational. Physical space — how welcoming, safe, and inclusive the campus environment feels — is also a belonging variable.
Research from the ACPA (American College Personnel Association) has examined how campus facilities, digital platforms, and spatial design communicate institutional belonging signals. Campuses where cultural centers, multilingual signage, and accessible design are visible and prominent demonstrate institutional commitment to belonging in ways that students notice.
In the digital realm, the design of course management systems, portal interfaces, and advising platforms can either reduce or reinforce belonging uncertainty. Institutions investing in inclusive UX design for student-facing digital tools are beginning to see this as a belonging intervention in its own right.
Intervention Type 6: Early Alert With Human Follow-Through
Early alert systems have become standard infrastructure at most institutions. What makes them an evidence-based belonging intervention is the quality of the human response they trigger.
Studies of early alert effectiveness consistently find that the alert itself is less important than what happens after it. Students who receive a personal outreach — a call or an in-person meeting, not an automated email — from someone who knows their name and expresses genuine concern report significantly higher belonging scores following that contact.
The National Academic Advising Association (NACADA) has published practice guidelines emphasizing that early alert programs function as belonging interventions when advisors are trained to lead with connection before problem-solving.
What Doesn’t Have Strong Evidence
In the interest of accuracy: some commonly implemented programs have weaker evidence than their popularity suggests.
- Mandatory diversity training without follow-up shows inconsistent effects on belonging; one-time sessions are rarely sufficient
- Orientation week programming in isolation rarely produces lasting belonging effects, though it can be effective when it’s part of a sustained first-year experience
- Mentorship matching without mentor training shows minimal belonging effects and sometimes increases belonging anxiety
This doesn’t mean these programs are worthless — it means the evidence for their impact on belonging, specifically, is mixed without additional structural support.
Building a Portfolio of Interventions
The strongest belonging ecosystems in higher education don’t rely on a single intervention. They layer multiple approaches: early affirmation activities, faculty and staff empathy development, structured peer mentorship, learning communities, and inclusive physical and digital environments.
What connects these layers is a shared institutional commitment to student engagement as a relational, not transactional, practice — and the assessment infrastructure to know whether what you’re doing is actually working.
The evidence is there. The question is whether institutions will build the structures to act on it.
Empathable designs evidence-informed empathy training and professional development programs for higher education institutions focused on improving Sense of Belonging and student success outcomes. Contact us to learn more.