How 61% Became the Number Everyone Cites About Young Adults and Loneliness

In February 2021, Harvard’s Making Caring Common project published a survey with a startling headline number: 61% of Americans aged 18 to 25 reported serious loneliness, against 36% of the adult population overall. The finding travelled fast. It has since surfaced in federal policy documents and international reports as shorthand for a generational crisis. It is worth tracing exactly how it moved, and what happened to its context along the way.

The original finding

The Harvard Making Caring Common report drew on an online survey of roughly 950 American adults, fielded during a specific and unusual moment: the winter of 2020-21, amid pandemic closures, remote instruction, and restricted in-person contact. Within that sample, 43% of young adults said their loneliness had increased since the pandemic began, and about half of lonely young respondents said no one had spent more than a few minutes recently asking how they were doing in a way that felt genuine. These are precise, useful numbers about a bounded period. The 61% figure was never presented by its authors as a permanent trait of an age cohort — it was a snapshot of young adults living through an acute disruption to in-person contact.

That distinction did not survive contact with press coverage. Headlines converted “61% of young adults reported serious loneliness during a pandemic winter” into “young adults are the loneliest generation,” a claim about identity rather than circumstance. The conversion is subtle but consequential: it turns a time-bound finding into a durable trait, which is a much bigger claim than the data support.

Into the policy document

The 2023 Surgeon General’s advisory on loneliness and isolation, issued by Vivek Murthy, cites the broader finding that about half of U.S. adults experience loneliness, and it foregrounds young adults as a population of particular concern. The advisory’s six-pillar national strategy names strengthening social infrastructure for youth and young adults as a distinct priority, an emphasis for which the Harvard figure functions as a load-bearing data point even though the advisory draws on a wider evidence base as well. This is where the number stops being a survey result and becomes a policy premise. Once a figure is embedded in a strategy document, it tends to get repeated without the caveats that accompanied it originally — the small sample, the single country, the pandemic-specific timing.

Into the international report

The migration continues. The World Health Organization’s June 2025 report from its Commission on Social Connection, co-chaired by Murthy, states that loneliness rates are highest among young people globally. That claim is true and well supported — but the WHO report’s own headline statistic comes from a different instrument entirely: Gallup’s 2023 survey of roughly 1,000 respondents per country across 142 countries, which found 24% of people worldwide report feeling very or fairly lonely, rising to 27% among adults aged 19-29 and falling to 17% among those 65 and older. Twenty-seven percent and sixty-one percent are both, in some sense, “the young-adult loneliness number.” They are not measuring the same thing.

Gallup’s question asked about feeling lonely “yesterday,” a single-day recall embedded in a broader wellbeing module, administered face-to-face or by phone across a global, probability-based sample. Harvard’s survey used a different loneliness measure administered online to a US-only convenience sample during a period of extraordinary disruption. Neither is wrong. They answer different questions, at different times, to different populations, using different recall windows. Citing them interchangeably — as “young adults are the loneliest group, somewhere between a quarter and two-thirds of them” — flattens a methodological gap into a rounding error.

A third figure, from a third instrument

Cigna’s loneliness index adds a third data point that rarely gets reconciled with either. Cigna’s 2020 report found 73% of workers aged 18-22 reported loneliness, rising above 80% among employed Gen Z respondents specifically — a figure restricted to people who hold jobs, surveyed through a workplace-oriented instrument, which is a narrower and more specific population than “young adults” as a category. Cigna’s 2025 follow-up, using fieldwork from mid-2024, found 57% of Americans overall report loneliness, with Gen Z and Millennials reporting more than Gen X and Baby Boomers despite greater technological connectedness. That framing — young, online, and lonely simultaneously — has become its own load-bearing claim in policy and press writing, distinct from either the Harvard or Gallup figures, and drawn from yet another sampling frame and question wording.

None of these four surveys — Harvard’s, Gallup’s, and Cigna’s two waves — used the same instrument, the same population, or the same reference period. All four get cited, sometimes in the same paragraph, as evidence for a single proposition: young adults are disproportionately lonely. The proposition may well be true. The evidence assembled for it is not additive; a reader cannot combine a 27% global figure with a 61% US pandemic figure with a 73-80% employed-Gen-Z figure and arrive at anything meaningful. Each is a legitimate number describing a different, narrower thing.

What gets lost in the compression

Something else disappears in this game of telephone: the distinction between loneliness, a subjective state, and the structural conditions that produce it. The American Enterprise Institute’s 2021 survey on American friendship found that 12% of Americans report having no close friends at all, up from 3% in 1990, and that the share of men reporting six or more close friends fell from 55% to 27% over three decades. That is a network-structure finding, not a feelings-based one, and it points toward a genuinely long-running decline rather than a pandemic-era spike. It rarely appears alongside the loneliness percentages in policy summaries, even though it may be doing more explanatory work than any of them. A young adult with a shrinking friend network and a young adult reporting loneliness “yesterday” on a Gallup phone survey may or may not be the same phenomenon measured two different ways.

A 2023 methodological review published in BMC Public Health identified inconsistent measurement as one of the central barriers to comparing findings across the loneliness literature, and the young-adult figures are a clean illustration of the problem it describes. The issue is not that any individual survey is unsound. It is that headline figures get lifted out of their instrument, their sampling frame, and their reference period, and then get stacked into policy documents and press coverage as though they were fungible.

What would settle it

A more defensible claim about young-adult loneliness would come from a single instrument, fielded with consistent wording, to comparable samples, at repeated intervals — the way AARP has tracked adults 45 and older using the 20-item UCLA Loneliness Scale since 2018, producing a genuinely comparable time series. No equivalent exists yet for adults under 25 at a global scale. Until one does, the honest response to “how lonely are young adults, really” is that the range currently on offer — roughly a quarter to three-quarters, depending on which survey is quoted — reflects differences in method as much as differences in experience, and the second-hand citation trail obscures rather than resolves that gap.