Men without a college degree are roughly twice as likely as college graduates to report having no close friends at all, according to an analysis published by the American Institute for Boys and Men. That gap, split along educational lines rather than gender lines, is now sharper than the gap between men and women. In 1990, friendship counts barely differed by class. By 2021, education had become the starker divide in America’s social networks, according to the same analysis of long-running survey data. The story of the “male loneliness epidemic” may be less about being male and more about who has the money, time, and built-in social infrastructure that come with a diploma.
The gender numbers alone do show a real gap, just a narrowing one. In 2021, 15% of men and 10% of women reported having zero close friends, up sharply from 3% and 2% in 1990. By 2023, according to Pew Research Center’s survey on social connections, that gap had closed to 9% of men and 8% of women. Men and women, Pew found, are about equally likely to say they have at least one close friend, and both groups report roughly five close friends on average.
What Pew’s data does show a persistent gender split in is upkeep, not existence. Women with close friends communicate with them far more often, by text, phone, video chat, and social media, several times a week, compared with men. Men who have friends tend to keep them at arm’s length, contact-wise.
Class, not gender, is doing the sorting
The AIBM analysis, authored by researchers Isaac Bledsoe and Ben Smith and published in August 2025, argues that educational attainment now predicts friendlessness more reliably than sex does. The gap between college graduates and non-graduates in the size of their social networks has widened more since 1990 than the gap between men and women has. That lines up with a simple structural explanation: college creates a dense, self-renewing pool of same-age peers, alumni networks, and shared institutional identity that persists for decades after graduation. Without that on-ramp, adults have to build a social circle almost entirely through work, neighborhood, and family, a slower and less reliable process with far more points of failure.
Age compounds the problem for everyone, but especially for men without that early social scaffolding. Isolation climbs steadily as men get older, and men 65 and up have more than double the odds of being socially isolated compared with younger men, per the AIBM research. Married men also lean disproportionately on a single relationship to meet their social needs: 85% cite a spouse as a support source, compared with 72% of women, who tend to spread that reliance across a wider circle of friends and family rather than concentrating it in one person.
None of this makes gender irrelevant. Men are more likely than women to say they don’t feel meaningfully part of any group or community, and more likely to say their place in the world doesn’t feel relevant, a distinct flavor of disconnection from simply lacking someone to call. But the data increasingly suggests that treating “male loneliness” as a purely gendered problem misses where the actual damage concentrates: among men, and to a lesser degree women, who left the education system without the built-in social infrastructure a degree tends to provide automatically.
Income tells a related, though not identical, story. College attendance and household income are tightly correlated in the U.S., which means the “education gap” in friendship overlaps heavily with a straightforward money gap: people with less disposable income have less capacity to travel for a wedding, cover the cost of a group dinner, or take unpaid time off to visit a friend going through a hard stretch. Every one of those small transactions is also a maintenance ritual for a friendship, and pricing people out of the ritual, even without anyone intending to, tends to price them out of the relationship over time.
What that means for fixing it
If class is doing more of the sorting than gender, then interventions built around “getting men to open up emotionally” are aiming at the wrong variable for a large share of the people actually affected. Community colleges, trade programs, churches, sports leagues, and workplaces outside the four-year-degree world may be the more useful places to rebuild the kind of recurring, low-effort social contact a college campus generates automatically for its students. The loneliest man in America, statistically speaking, isn’t defined mainly by his gender. He’s defined by the institutions he never had access to in the first place.
