What the 61% Youth Loneliness Figure Doesn't Show

A single number has done most of the work in the recent public conversation about young people and loneliness: 61%. That figure comes from a 2021 survey by Harvard’s Graduate School of Education, Making Caring Common, which found 61% of adults aged 18 to 25 reported serious loneliness, compared with 36% of American adults overall. It is a real and striking finding, drawn from a national sample, and it has been repeated in enough places that it now functions as shorthand for the state of young adulthood.

But 61% is itself an average across a group of roughly 33 million people. Averages compress variation, and the variation here is worth examining before treating the figure as a description of “young adults” as an undifferentiated category.

What the number is actually built from

The Harvard Making Caring Common survey measured serious loneliness using its own instrument, distinct from the 20-item UCLA Loneliness Scale used in some other work discussed below. Within the same report, two adjacent figures sit next to the 61%: 43% of young adults said their loneliness had increased since the pandemic began, and about half of lonely young adults said no one had taken more than a few minutes recently to ask how they were doing in a way that felt genuine. These are not the same construct as the topline prevalence figure — one is a rate, one is a trend, one is about the quality of contact rather than its absence — and conflating them produces a picture that feels more coherent than the underlying data actually is.

The report also found that 51% of mothers with young children reported serious loneliness, a rate close to the young-adult figure but arrived at through a completely different subgroup definition — parental status, not age. Some of that group overlaps with the 18-25 population; most does not. Reporting the youth figure and the maternal figure side by side, as the source itself does, makes clear that “young people are lonely” and “new mothers are lonely” are two separate findings that happen to produce similar numbers, not one finding restated twice.

The gender question the youth figure cannot answer

The most consequential gap in the 61% statistic is that the Harvard report, as summarized in its public findings, does not break the 18-25 loneliness rate down by gender. This matters because a separate line of research suggests gender may be doing substantial work in the broader friendship and connection story, even if it cannot yet be confirmed within this specific age band.

The 2021 Survey Center on American Life report on friendship, from the American Enterprise Institute, found that 12% of Americans report having no close friends at all, up from 3% in 1990 — a fourfold increase. But that overall figure obscures a sharper divergence by sex: 15% of men report no close friends, versus a much lower share of women, and the collapse in men’s friendship networks over three decades is dramatic on its own terms. In 1990, 55% of men reported having at least six close friends; by 2021 that had fallen to 27%. Men’s rate of complete friendlessness rose roughly fivefold over the same period.

This is a general-population finding, not one restricted to 18-to-25-year-olds, and that limitation should be stated plainly rather than smoothed over. The Survey Center data cannot be used to claim that young men specifically are driving the Harvard youth-loneliness figure. What it does establish is that when broad connection statistics are broken out by gender, men diverge sharply and negatively from the aggregate trend, and that this divergence has been building for three decades, well before the pandemic-era loneliness surveys that produced the 61% figure. Whether that decades-long male friendship decline is concentrated in the youngest cohort, evenly spread across ages, or concentrated in men now in midlife who came of age before it accelerated, is not something the available sources resolve.

That is the honest state of the evidence: a strong signal that gender matters enormously to social connection outcomes generally, and a youth loneliness statistic that does not report gender at all. Putting the two next to each other is suggestive. It is not proof of what is happening inside the 61%.

A second axis: employment and Gen Z

Cigna’s 2020 workplace loneliness survey offers a different subgroup cut, this one for employed young people specifically. It found 73% of workers aged 18 to 22 reported loneliness, and over 80% of employed Gen Z workers reported the same — both markedly higher than the 61% youth figure from Harvard, and higher again than Cigna’s own overall adult figure of 61% (a coincidental match in number, not construct — Cigna’s is a workplace-loneliness measure, not the same instrument as the Harvard survey). Cigna also found that lonely workers missed work twice as often due to illness and five times as often due to stress, which speaks to consequences rather than prevalence.

The gap between the Harvard youth figure and the Cigna employed-Gen-Z figure is itself instructive. Both describe people broadly in the same age range, but Cigna’s sample is restricted to people who are employed, surveyed through a workplace lens, using a workplace-specific instrument. A young person outside the labor force, in education, or unemployed is not represented the same way. Employment status is another axis the single 61% figure averages away, and the direction of the difference — the employed subgroup reporting even higher loneliness than the general youth figure — cuts against any assumption that having a job and colleagues functions as straightforward protection against loneliness for this age group.

Why the averaging matters practically

None of this is an argument that youth loneliness is overstated. If anything, several of the subgroup cuts available — employed Gen Z workers, men across the adult lifespan — suggest particular slices of the youth population may be worse off than the 61% headline implies, not better. The point is narrower: a single prevalence figure for “young adults” flattens differences by gender, employment status, and parental status that the source data itself gestures toward but does not fully cross-tabulate.

For comparison, the AARP Foundation’s 2018 survey of adults 45 and older used the UCLA Loneliness Scale, the standard academic instrument, and reported one in three midlife and older adults as lonely, with social network size and diversity and physical isolation as the strongest predictors. That survey did report loneliness by demographic subgroup, including by living situation and neighborhood contact — 33% of those who had spoken to neighbors were lonely versus 61% of those who never had. It is a model for what the youth literature currently lacks: a widely cited topline figure paired with the subgroup breakdowns needed to know who the average is actually describing.

A study that would strengthen this picture is a single national survey of 18-to-25-year-olds using one consistent instrument, with loneliness reported by gender, employment status, and parental or relationship status simultaneously, rather than as a series of separate reports each cutting the population a different way. Until that exists, the 61% figure should be read as an average whose internal structure is only partially visible.