The emotional distress response rating prompts ("I panic easily") code as more feminine and I am not surprised men were less likely to respond yes. It's much more socially permissable for women to admit to things like panic than men. The social disengagement prompts ("I prefer to be alone") code as more masculine, so it's not surprising to see a reduction in the gender gap there.
Also I don't think this survey is getting at what most people consider the loneliness crisis. What this survey captures well is that mental health and self-reported mental wellbeing is correlated to age, with young people reporting higher negativity than older people.
When I think of the loneliness crisis however I think more of results/things you do in the world. Less a report of emotional status than "do you hang out with friends? How many do you have? How frequently do you socialize?", all of which I have seen convincing data showing men are more physically isolated.
To be fair, that's only descriptive, and this survey is more explanatory. Interesting stuff.
This was my reaction as well. One of the key findings for the male loneliness epidemic thesis is that men are less likely to have any close friends. In general when you ask factual questions about men’s social lives vs. women, you see men are worse off, whereas when you ask feelings questions it’s a mixed picture.
I wish that for these kinds of subjective well-being metrics, studies validated by testing the correlation between real-world outcomes and the subjective metrics for different sub-groups.
For example, correlating subjective self-reported loneliness with number of people you talk to in a week, negative physical health outcomes, or suicidality could help ground these metrics in something less dependent on gendered norms in self-reported feelings. Then, we could more reasonably interpret the gender gap with some notion in how gender differences impact self-reported subjective well-being.
I have no idea if this kind of thing is feasible in surveys, but it's the kind of cross-check we would do in my work to make sure your measured variable actually reflects the underlying process you are trying to study
It's not that surprising that this has become framed as the male problem.
Men are more prone to react to emotional distress with externalising behaviour, so when men are lonely, they lash out and make it everyone else's problem.
Women, on the other hand, tend to react with internalising behaviour, which is much less visible to the outside observer, even though the distress might be just as severe.
It's telling that the term "incel" was coined by a woman, but made infamous by men.
So interesting. I’ve been wondering whether it really matters, practically speaking, which gender is more lonely. Unless the solutions need to be extremely gender-specific (and why would they need to be?), I’m not convinced that the gender of the problem has any important implications at all.
We need to *collect* the gender data, sure, but I’ve started to think using it to *frame* the problem and think about solutions isn’t very helpful. I think it’s just a bad habit we’ve formed—hyper-fixating on demographics in how we talk about stuff.
Maybe “people are way too lonely, and increasingly so” is problem enough, for the discourse, at least.
I've noticed reflections of these survey results on a near-daily basis, in real life.
Young adults, especially, seem to be… terrified. Terrified, I think, is a good way to put it. They're terrified of joining in even the most benign interactions — and yes, I think this is even more true of young adult women.
I'd like to speculate that our online lives may have eroded both our critical reasoning skills, and also our communications skills. Our language skills.
It's hard to overstate the catastrophic effects this is having on society. It seems that we're only now coming to realize the enormity of this disaster, and how hard it will be to reverse.
I'd be very interested to see economic class and family status breakdowns of these results. A friend of mine makes the argument that bottom-half loneliness and lack of family formation are driving right-authoritarian populism:
Thank you for an excellent article using solid data to point out that the loneliness crisis is broader than simply a phenomenon amongst young men.
Yet it's also important to realise that 'the anti-social century', as Derek Thompson calls it in The Atlantic, is not solely caused by the younger generation's retreat into the online world. It's also the product of the erosion of America's social infrastructure. Churches, community centres, youth sports facilities and offices are being hollowed out. In its place, a whole new anti-social infrastructure is emerging to cater to a generation of people retreating to lead increasingly solitary lives. We see it in home-based entertainment, supplied by Netflix, home-based delivery, courtesy of Doordash, and home-based working, enabled by Zoom. Now the question is: will society continue down this path, or will we see a civic renewal which once more strengthens community and social ties?
They’re hallowed out because young people aren’t interested in them as much! For example, go to any church, especially mainline Protestant churches, and it will be filled with grey haired people.
I do think "less young people are going to church" isn't *as* strong a sign of people being lonely in general compared to "less young people are doing sports / going to parties / etc". Young people are increasingly secular and it's a lot harder to get an atheist to entertain religion than it is to get someone who's never been interested in sports to consider giving basketball a try (or whatever).
I think the rise in youth loneliness is likely real, but it seems plausible that younger people are always lonelier, and as they get older they become less lonely. It would make sense that younger people have always been lonelier. They're less likely to be married, have just left their homes, and aren't as established.
I may be misunderstanding something, but I don't understand how this data is indicative of a crisis. The loneliness graph shows all genders in all age groups have mean scores above the 0 point. If I understand your coding description correctly, this means that people are tending to respond to questions like "I frequently feel lonely", “I prefer to be alone”, “I keep others at a distance” with the response "Neither agree or nor disagree." You yourself refer to this as a neutral response. If the average for the subgroups is at or more positive than the neutral point on these measures, does this really classify as a "crisis"?
Also, given that we are talking about loneliness, why not show us the mean responses to the "I frequently feel lonely" question directly, rather than obscuring that average in a broader composite that includes a lot of items that aren't really about loneliness?
It would be awesome if people scored higher (and we should work to make that a reality!), but referring to this as evidence of a "loneliness crisis" is clickbait.
Thanks, Lakysha. I see that you're right—I think that I was just focusing too much on the first ten words in each caption. I'm sorry about that. I'll delete my original comment.
The emotional distress response rating prompts ("I panic easily") code as more feminine and I am not surprised men were less likely to respond yes. It's much more socially permissable for women to admit to things like panic than men. The social disengagement prompts ("I prefer to be alone") code as more masculine, so it's not surprising to see a reduction in the gender gap there.
Also I don't think this survey is getting at what most people consider the loneliness crisis. What this survey captures well is that mental health and self-reported mental wellbeing is correlated to age, with young people reporting higher negativity than older people.
When I think of the loneliness crisis however I think more of results/things you do in the world. Less a report of emotional status than "do you hang out with friends? How many do you have? How frequently do you socialize?", all of which I have seen convincing data showing men are more physically isolated.
To be fair, that's only descriptive, and this survey is more explanatory. Interesting stuff.
This was my reaction as well. One of the key findings for the male loneliness epidemic thesis is that men are less likely to have any close friends. In general when you ask factual questions about men’s social lives vs. women, you see men are worse off, whereas when you ask feelings questions it’s a mixed picture.
Source: https://aibm.org/research/male-loneliness-and-isolation-what-the-data-shows/#:~:text=lonely%20right%20now.-,Friendships,%25%2C%20respectively%2C%20in%201990. (Figures 4 and 7)
I wish that for these kinds of subjective well-being metrics, studies validated by testing the correlation between real-world outcomes and the subjective metrics for different sub-groups.
For example, correlating subjective self-reported loneliness with number of people you talk to in a week, negative physical health outcomes, or suicidality could help ground these metrics in something less dependent on gendered norms in self-reported feelings. Then, we could more reasonably interpret the gender gap with some notion in how gender differences impact self-reported subjective well-being.
I have no idea if this kind of thing is feasible in surveys, but it's the kind of cross-check we would do in my work to make sure your measured variable actually reflects the underlying process you are trying to study
It's not that surprising that this has become framed as the male problem.
Men are more prone to react to emotional distress with externalising behaviour, so when men are lonely, they lash out and make it everyone else's problem.
Women, on the other hand, tend to react with internalising behaviour, which is much less visible to the outside observer, even though the distress might be just as severe.
It's telling that the term "incel" was coined by a woman, but made infamous by men.
So interesting. I’ve been wondering whether it really matters, practically speaking, which gender is more lonely. Unless the solutions need to be extremely gender-specific (and why would they need to be?), I’m not convinced that the gender of the problem has any important implications at all.
We need to *collect* the gender data, sure, but I’ve started to think using it to *frame* the problem and think about solutions isn’t very helpful. I think it’s just a bad habit we’ve formed—hyper-fixating on demographics in how we talk about stuff.
Maybe “people are way too lonely, and increasingly so” is problem enough, for the discourse, at least.
I've noticed reflections of these survey results on a near-daily basis, in real life.
Young adults, especially, seem to be… terrified. Terrified, I think, is a good way to put it. They're terrified of joining in even the most benign interactions — and yes, I think this is even more true of young adult women.
I'd like to speculate that our online lives may have eroded both our critical reasoning skills, and also our communications skills. Our language skills.
It's hard to overstate the catastrophic effects this is having on society. It seems that we're only now coming to realize the enormity of this disaster, and how hard it will be to reverse.
I'd be very interested to see economic class and family status breakdowns of these results. A friend of mine makes the argument that bottom-half loneliness and lack of family formation are driving right-authoritarian populism:
https://cathyreisenwitz.substack.com/p/bottom-half-folks-voted-for-trump
and it's worth examining whether these data are consistent with that hypothesis.
Thank you for an excellent article using solid data to point out that the loneliness crisis is broader than simply a phenomenon amongst young men.
Yet it's also important to realise that 'the anti-social century', as Derek Thompson calls it in The Atlantic, is not solely caused by the younger generation's retreat into the online world. It's also the product of the erosion of America's social infrastructure. Churches, community centres, youth sports facilities and offices are being hollowed out. In its place, a whole new anti-social infrastructure is emerging to cater to a generation of people retreating to lead increasingly solitary lives. We see it in home-based entertainment, supplied by Netflix, home-based delivery, courtesy of Doordash, and home-based working, enabled by Zoom. Now the question is: will society continue down this path, or will we see a civic renewal which once more strengthens community and social ties?
They’re hallowed out because young people aren’t interested in them as much! For example, go to any church, especially mainline Protestant churches, and it will be filled with grey haired people.
I do think "less young people are going to church" isn't *as* strong a sign of people being lonely in general compared to "less young people are doing sports / going to parties / etc". Young people are increasingly secular and it's a lot harder to get an atheist to entertain religion than it is to get someone who's never been interested in sports to consider giving basketball a try (or whatever).
Do we have any historical data on this?
I think the rise in youth loneliness is likely real, but it seems plausible that younger people are always lonelier, and as they get older they become less lonely. It would make sense that younger people have always been lonelier. They're less likely to be married, have just left their homes, and aren't as established.
I think it's the reverse.
When I was 20 i had many friends and went out literally Every weekend.
Now i'm almost fifty and have a lot less friends. And i'm mainly home with my family.
I'm certainly not lonely, but I have a much smaller group of friends
I may be misunderstanding something, but I don't understand how this data is indicative of a crisis. The loneliness graph shows all genders in all age groups have mean scores above the 0 point. If I understand your coding description correctly, this means that people are tending to respond to questions like "I frequently feel lonely", “I prefer to be alone”, “I keep others at a distance” with the response "Neither agree or nor disagree." You yourself refer to this as a neutral response. If the average for the subgroups is at or more positive than the neutral point on these measures, does this really classify as a "crisis"?
Also, given that we are talking about loneliness, why not show us the mean responses to the "I frequently feel lonely" question directly, rather than obscuring that average in a broader composite that includes a lot of items that aren't really about loneliness?
It would be awesome if people scored higher (and we should work to make that a reality!), but referring to this as evidence of a "loneliness crisis" is clickbait.
hmm the captions are all slightly different; can you show me what you see?
Thanks, Lakysha. I see that you're right—I think that I was just focusing too much on the first ten words in each caption. I'm sorry about that. I'll delete my original comment.