The Averaged-Away Young Adult

The Surgeon General’s 2023 advisory put a single number in front of the American public: roughly half of adults report experiencing loneliness. That figure travelled widely, and it is accurate as a national average. It is also the kind of number that erases the group most affected by it.

Break the same underlying pattern down by age and a different picture appears, consistently, across nearly every survey that bothers to report age separately.

What the age breakdowns actually show

The Harvard Graduate School of Education’s Making Caring Common project found 36% of Americans reporting serious loneliness in its 2021 survey — already a sobering figure on its own. But among adults aged 18 to 25, the rate was 61%. That is not a modest gap. It means a survey that reports “36% of Americans are lonely” is describing a national average built from a subgroup at 61% and other subgroups considerably lower, and the average obscures both ends.

Gallup’s 2023 global survey, conducted with Meta across 142 countries, found the same shape internationally. Overall, 24% of people worldwide reported feeling very or fairly lonely. Adults aged 19 to 29 reported loneliness at 27%, while adults 65 and older reported it at 17% — the lowest of any age band Gallup tracked. The direction of the gradient is consistent: loneliness declines with age, at least until very late life, rather than rising as popular assumption about isolated older people might suggest.

Cigna’s 2025 Loneliness in America report, based on fieldwork from mid-2024, found 57% of Americans overall reporting loneliness, with Gen Z and Millennials reporting more loneliness than Gen X or Baby Boomers — despite Gen Z and Millennials being, by a wide margin, the most digitally connected cohorts ever surveyed. That detail matters for anyone tempted to treat smartphone use as a sufficient explanation either way; the relationship between technological connectedness and felt loneliness does not run in the direction a simple substitution story would predict.

None of this is a US-specific artifact. A 2026 eight-country study reported that nearly half of young adults report loneliness, with the loneliness-depression-anxiety association holding after adjusting for demographics and prior diagnoses. The pattern generalises across national contexts with very different welfare systems, housing markets, and cultural norms around independence — which argues against explanations rooted in any single country’s institutions.

Why the averaging matters, not just as arithmetic

There is a methodological reason to care about this beyond fairness to young adults as a subgroup. Julianne Holt-Lunstad’s 2015 meta-analysis in Perspectives on Psychological Science found that social deficits were more predictive of early death in samples averaging under 65 than in older samples. If the health consequences of loneliness are themselves age-dependent, then an age-blind prevalence figure is not just imprecise description — it risks misallocating the analytic weight assigned to the problem. A policy briefing that treats “half of adults are lonely” as evenly distributed risk is describing a different disease than one that recognizes a concentrated, comparatively younger burden.

The workplace data reinforces the pattern from another angle. Gallup’s 2024 State of the Global Workplace report found loneliness higher among employees under 35 than older colleagues, and higher among fully remote workers (25%) than hybrid (21%) or fully on-site staff (16%). Younger workers are also more likely to be in remote or hybrid roles early in their careers, before they have built the workplace social ties that come with tenure — a plausible mechanism, though the surveys available do not test it directly.

The American Enterprise Institute’s 2021 survey on friendship adds a structural layer often missing from prevalence figures alone: the share of Americans reporting no close friends rose from 3% in 1990 to 12% in 2021, and the share of men with six or more close friends fell from 55% to 27% over the same period. That decline began among people who are now the older adults reporting comparatively lower loneliness — meaning today’s older cohort came of age with denser friendship networks than today’s young adults are building. If network size correlates with felt loneliness later in life, as AARP’s research on adults 45 and older has repeatedly found, then the age gradient observed today may not hold for the cohort now in its twenties once it reaches 65.

What this changes about how the numbers should be reported

None of the sources reviewed here offer a randomised design that would establish why young adults report more loneliness than older adults; the evidence is cross-sectional and descriptive, consistent across many populations but silent on mechanism. What the age breakdowns do establish, repeatedly and across independent survey instruments, is that a single national loneliness percentage is not a neutral summary statistic. It is an average across groups with meaningfully different rates, and the group furthest above that average is consistently the youngest adults surveyed — not, as intuition about isolated elderly people might suggest, the oldest. Any report that publishes a topline loneliness figure without an age breakdown alongside it is withholding the information a reader would need to know who, specifically, the number is about.