Improving College Student Retention Through Connection
Improving College Student Retention Through Connection: A Leadership Guide for 2026
Retention is the number that keeps university presidents up at night.
Not because they don’t care about learning. Because retention is where learning, belonging, mental health, institutional sustainability, and mission all intersect — and where the gap between what universities say they value and what they actually build becomes visible.
This piece is for leaders: provosts, deans, VP-level student affairs professionals, and department chairs who understand that improving retention is not the job of one office. It is a campus-wide cultural and structural project. And at the center of that project, whether leaders name it or not, is connection.
The Connection-Retention Link: What the Data Shows
The correlation between social connection and student persistence is one of the most replicated findings in higher education research.
Dr. Alexander Astin’s Input-Environment-Outcome (I-E-O) model, foundational in student development research, consistently identified peer interaction and faculty-student interaction as among the most powerful environmental factors influencing student outcomes — including persistence. His findings, published across decades of longitudinal research at UCLA’s Higher Education Research Institute, established that connection is not peripheral to academic success. It is central to it.
More recent data reinforces this. Gallup’s “Great Jobs Great Lives” study, which surveyed thousands of college graduates about the relationship between their college experience and long-term wellbeing, found that students who reported at least one professor who cared about them as a person, one mentor who encouraged their goals, and one professor who made them excited about learning were far more likely to thrive — and to report that their college education was worth the cost.
Three relational variables. That’s what separated the students who described their education as transformative from those who described it as merely credentialing.
Why Connection Breaks Down — and Where
Connection breaks down at predictable places. Knowing where makes it possible to intervene.
Scale. Large institutions create anonymity by default. Students in lecture halls of 300 are invisible to each other and to their instructors. Advisors with caseloads of 500+ cannot provide meaningful connection. Without structural countermeasures, scale systematically erodes the conditions for belonging.
Commuter and part-time enrollment patterns. Students who come to campus, attend class, and leave have far fewer organic opportunities for connection. They are among the highest-risk populations for attrition, and the standard campus community-building model — which assumes residential life and co-curricular involvement — often doesn’t reach them at all.
Identity-based belonging threats. Students from underrepresented groups experience belonging uncertainty as an active cognitive load — a persistent background question of whether this institution was built for people like them. This is not a perception problem. Campuses that haven’t actively designed for belonging across identity groups are less belonging-affirming for those students. The data on this from the Review of Higher Education and annual HERI surveys is consistent and significant.
The transition moments. First year, first semester, first weeks after a major life disruption (health issue, financial shock, family crisis) — these are the moments when connection is most fragile and students are most vulnerable to disengagement. Institutions that don’t have warm-transfer protocols at these moments lose students who might otherwise have stayed.
What Connection-Centered Retention Looks Like in Practice
Redesign Advising Around Relationship, Not Transactions
The most effective advising models in higher education right now aren’t built around information transfer. They’re built around relational continuity. The same advisor. Over multiple semesters. With enough caseload reduction that they can actually know their students.
NACADA’s research on intrusive advising — proactive, relationship-based outreach that doesn’t wait for students to come to advisors — consistently shows stronger retention outcomes than reactive, appointment-based models. The key factor is the relationship. Students who trust their advisor will respond to outreach. Students who don’t will ignore it.
Invest in Faculty as Belonging Architects
Faculty have more student contact than anyone else on campus. They are also, in most institutions, the least developed in terms of relational skill.
This is not a criticism of faculty. Most faculty were trained as disciplinary experts, not as community builders. But the research is clear that instructor behavior shapes belonging. And belonging can be taught.
Empathy training for faculty — real empathy training, not a two-hour diversity workshop — builds the capacities that matter: recognizing distress signals, responding without judgment, creating classrooms where struggle is normalized and help-seeking feels safe.
Institutions that treat faculty development in relational skills as a legitimate professional investment — with time, support, and follow-up — report improvements in both student mental health indicators and course completion rates. The University of Indiana Bloomington’s faculty learning communities model, which integrates belonging and inclusion skills into sustained peer learning across semesters, is one example of this done with real institutional commitment.
Build Connection Into Academic Structures
The single most effective structural intervention for retention through connection isn’t a program. It’s a design choice: give the same group of students shared academic experiences over time.
Learning communities. Cohort-based programs. Linked courses. Collaborative pedagogies. These create connection as a byproduct of being in class together, repeatedly, with the same people. They don’t require students to opt into community building outside of class — they build connection through the structure of learning itself.
A meta-analysis published in Community College Review found that students in learning communities showed significantly higher retention rates, GPA, and credit completion than comparable students not in learning communities, with effect sizes that were consistent across institution types and student populations.
Address Student Mental Health as a Connection Problem (and vice versa)
Student mental health and connection are bidirectionally linked. Loneliness produces anxiety and depression. Anxiety and depression produce withdrawal. Withdrawal erodes connection. The cycle compounds.
Institutions that break this cycle don’t do it by scaling up counseling services alone — though adequate mental health support is essential. They break it by building enough connection into the fabric of the campus experience that fewer students arrive at crisis.
This is a campus culture argument, not just a clinical one. When students feel known, supported, and genuinely part of something, their baseline mental health is better. When their baseline mental health is better, they are more likely to seek help before crisis, more likely to persist through difficulty, and more likely to complete.
The American Foundation for Suicide Prevention’s “More Than Sad” college initiative, the JED Foundation’s campus program, and active research from the Student Experience in the Research University (SERU) Consortium all point to the same conclusion: social connection is a mental health protective factor that campuses can actively build or passively neglect.
Sense of Belonging as a Leadership Priority
One of the most important shifts in higher education right now is the move to treat Sense of Belonging not as a student affairs concept but as a leadership imperative.
When belonging is owned only by student affairs, it stays underfunded and siloed. When academic affairs, enrollment management, facilities, IT, and the provost’s office all understand belonging as a shared strategic priority — and when it shows up in budget conversations, curriculum design, hiring criteria, and institutional reporting — it becomes possible to actually change the campus culture.
This is the work of leadership. Not managing belonging programs. Building belonging infrastructure.
A Framework for Getting Started
For leaders who want to move from intention to action:
Assess. What does your institution actually know about the connection experiences of different student populations? Where are the gaps in your data?
Identify the friction points. Where does connection structurally fail? Scale? Advising caseloads? Faculty development? Commuter students? Transfer students?
Choose one structural intervention. Not a program. A structural change: caseload reduction, cohort learning design, a faculty development initiative with real depth and follow-up.
Measure. Define belonging as an outcome variable and track it alongside retention data. Use the data to make the next decision.
Commit publicly. Belonging as a campus priority becomes real when it’s in the strategic plan, in the president’s communications, and in the criteria by which academic and student affairs leaders are evaluated.
The connection-retention relationship is not a theory. It is the evidence base for one of the most important investments higher education institutions can make in 2026.
The campuses seeing the strongest retention outcomes are not the ones with the most programs. They are the ones where connection is the operating logic — woven into advising, coursework, faculty practice, physical space, and leadership accountability.
That’s what it takes. And it is worth doing.
Empathable partners with higher education institutions to build the empathy skills and relational cultures that support student belonging, mental health, and retention. If your institution is ready to treat connection as infrastructure, let’s talk.