I spent a few months reading the research on this before I did anything about my own situation, mostly as a way of not doing anything about my own situation.

What I found was that the numbers are real, that they are quoted more confidently than they deserve, and that one of the best studies says something close to the opposite of the popular version. That last part is the useful bit, so I have written it down properly.

Where the word comes from

The phrase entered general use through the US Surgeon General's May 2023 advisory, Our Epidemic of Loneliness and Isolation. That document is about the whole population, not men specifically, and it is largely concerned with the health consequences of social disconnection rather than with which sex has it worse.

Almost everything you read about lonely men borrows its urgency from that title and then attaches it to a set of statistics from elsewhere. It is worth keeping the two separate.

I try to avoid the word in my own writing, partly because it is used loosely and partly because it does something unhelpful to the reader: it turns an ordinary, fixable social problem into a public health condition you are a case of. But it is the phrase people search, so here is what sits underneath it.

The figure everyone quotes

The Survey Center on American Life published The State of American Friendship: Change, Challenges, and Loss in 2021, using its May 2021 American Perspectives Survey. Two findings did most of the travelling:

  • 15 per cent of American men reported having no close friends.
  • 27 per cent reported six or more close friends, against 55 per cent in 1990. The no-close-friends figure in 1990 was 3 per cent.

Both are real. The 2021 half is a single well-conducted survey and I would take it at face value.

The 1990 half needs a caveat that it almost never gets. That baseline comes from a Gallup poll, a different study with a different instrument, thirty-one years earlier. Comparing them gives you a genuine signal, but "fivefold increase" and "cut in half" are tidier claims than two non-identical surveys can strictly support. Every article I read reported those multipliers as though they came out of one longitudinal dataset. They do not.

The direction is well corroborated elsewhere. The precision is decorative.

The finding that complicates the story

Here is the part that changed how I think about this.

Pew Research Center published Men, Women and Social Connections in January 2025, based on a survey of 6,204 American adults conducted in September 2024. It set out to test the popular claim directly, and found:

  • Men do not report feeling lonely or isolated more often than women. Roughly one in six of both say they feel that way all or most of the time.
  • Men do not report having fewer close friends than women.
  • What does differ is use. Men who have close friends contact them less often. Men are less likely to turn to their network for emotional support, and they draw on a narrower part of it when they do.

That is a different problem from the one the slogan describes, and I think it is a more accurate one. It is not that men have nobody. It is that the machinery for using the people they have has stopped running - fewer calls, fewer approaches, a smaller set of people considered askable.

Both things can be true at once, and I think they are: a real minority of men have genuinely empty inner circles, and a much larger group have people they never actually contact.

What I take from it

The problem is behavioural more often than it is structural. For most men in this position the fix is not acquiring new people. It is contacting the ones already there, more often, for less reason. That is a smaller and much cheaper intervention than the framing usually implies.

The 15 per cent is a floor, not a description of the average man. If you have nobody at all, you are in a real and sizeable minority. If you have three people you never speak to, you are in a much larger one, and the surveys mostly count you as fine.

Sex comparisons are a distraction. Whether men have it worse than women has no bearing on what any individual man should do on Tuesday. It gets the clicks and it is the least actionable question in the field.

Be suspicious of confident multipliers. Including mine. When I use the 15 per cent figure I try to say where it came from and what its limits are, because this subject is full of numbers that have been photocopied so many times the source is gone.

What the data does not tell you

It does not tell you whether what you have is a social problem or a medical one. Loneliness and depression overlap substantially, and no survey will sort your case. If the flatness is constant, present even in company, or has taken the enjoyment out of things you used to like, that belongs with a doctor rather than with an article. I write about the social side because it is the side I worked on. I am not qualified for the other one and will not pretend to be.

It also does not tell you anything about causation over your own decade. The reason my circle emptied was a relocation, a team folding, a chat going quiet and about ten years of not paying attention. No dataset was going to hand me that. I had to go back through it myself.

The practical read

If you want one sentence out of all of it: the evidence supports a real and under-discussed problem, does not support men being lonelier than women, and points at contact frequency rather than headcount as the thing that has actually changed.

Which means the first useful move is almost never "make new friends". It is to pick someone already in your life and contact them without a reason. I have set out what that actually looks like, including the rate at which it goes nowhere.

If you want the experiential version rather than the numbers, what this actually feels like from inside is here.

Sources