The Campus Was Built for Connection. Why Isn't It Working?
Last month, the University of Michigan announced that starting in fall 2027, first-semester grades for every new student in its largest college will not appear on transcripts or count toward GPA. The stated intention is to provide a buffer for students during one of the most fragile transitions in early adulthood: the leap from high school to college. More specifically, the university expressed that the goal of this policy is to “curb the mental health crisis unfolding among college-aged individuals.”
A week after the announcement, Rose Horowitch's article in The Atlantic highlighted another major shift in campus life, revealing that college students have stopped hanging out. She notes that the share of first-year students who feel isolated from campus life has more than doubled in two decades, and that the share of college seniors spending six or more hours a week socializing face-to-face has fallen from 74 percent to 65 percent since 1994.
The college years are the most socially engineered environment most adults will ever experience. Students live among thousands of same-age peers within walking distance, with shared dining halls, overlapping schedules, and hundreds of weekly opportunities to connect with like-minded individuals via clubs. And yet roughly half of all U.S. college students now report being lonely. This contrast brings into sharper focus something social psychologists have wrestled with since Festinger's classic MIT housing study first linked physical closeness to friendship: yes, proximity creates the opportunity for connection, but it doesn't guarantee belonging.
Closing that gap has typically meant adding more of what already exists: more counseling appointments, more optional workshops, more clubs to join. But those are services. They wait for a student to notice they're struggling and go looking for help. A system works differently. It's built into the structure of campus life itself, in place before anyone knows who will need it.
Research by the behavioral scientists Gregory Walton and Geoffrey Cohen helps explain why that distinction matters so much. They coined the term belonging uncertainty, the quiet, corrosive question of whether you're the only one who feels like they don't fit in. In a randomized trial, helping first-year students understand social struggle as a normal, temporary part of the transition measurably improved academic performance and health outcomes over the following three years, with the gains concentrated among students who had the most reason to doubt they belonged.
That degree of resilience matters financially as much as personally. The number of high school seniors going on to college is shrinking, and since most institutions depend on tuition, every student who quietly withdraws also takes several years of revenue out the door with them. Loneliness is a balance-sheet problem as much as it is a belonging or mental health one.
Most universities have responded to this growing crisis of belonging with one of two levers: more clinical capacity, or more programming. Both matter, but neither is sufficient on its own. Counseling centers cannot scale to meet a problem affecting half of campus, and optional programming has a quieter flaw. As the name suggests, it depends on students opting in, meaning it disproportionately reaches those who are already good at connecting, and fails to reach the ones who need it more.
Michigan's grading pilot avoids that trap by applying the new policy to every first-year student automatically. If covering grades were optional, choosing it would signal a student was struggling, which would defeat the purpose. The policy only works because no one has to ask for it.
That's the real distinction between a system and a service. A service waits for a student to raise a hand. A system is already built in before anyone knows who will need it and who won't. Michigan's transcript policy reflects that design principle directly: it invests before the crisis is visible in any one individual student, rather than after.
For the administrators now rethinking student life, that reframes an important question. They are not simply deciding which existing programs need bigger budgets to do more of the same. They must consider which existing structures, like orientation, advising, housing, the first-semester course schedule, are quietly running on the assumption that belonging will just happen via proximity. And they must decide how much of belonging they're willing to leave to chance, and how much they're ready to build in from the start.
