(2023-03-08) ZviM The Kids Are Not Okay

Zvi Mowshowitz: The Kids are Not Okay. Teen Suicide attempts are up. Depressive episodes are way up... Spoiler alert, I’m going to blame the smartphones and social media.

One child in ten attempted suicide this past year, and it is steadily increasing? Yikes.

male adolescents die by suicide at a rate five times greater than that of female adolescents, although suicide attempts by females are three times as frequent as those by males... method of attempted suicide for males is typically that of firearm... Females have more parasuicides.

The conflation of suicide rates with forced sex here seems at best highly misleading

Also, can we zoom out a bit? On a historical graph, the suicide rate does not look all that high (scale is suicides per 100,000 children, per year)? (Prev peak 1987)

The kids are not okay. The kids in the 1990s were, by some of these graphs, even more not okay

We also know that the rate of major depressive episodes among US adolescents increased by more than 52 percent between 2005 and 2017. I do think some of that is changing norms on what we call such an episode. I doubt that is all (or more than all) of it.

They are happy to report several things. Adolescents are engaging in less ‘risky sexual behavior,’ which means less sexual behavior period, both at all and with four or more partners. There is less substance (alcohol, marijuana, illicit drug, misused prescription drug) use. Overall: So less fun. Sounds depressing. Less sexual activity is flat out called ‘improvement’ by the CDC. I am going to come out and say that the optimal amount of teenage sexual activity, and the optimal amount of teenage substance use, are importantly not zero.

They also talk glowingly about ‘parental monitoring,’ defined as parents or other adults in their family knowing where students are going and who they are with, as ‘another key protective factor for adolescent health and well-being.’ While I certainly would agree with the correlation with short term physical safety, and that this leads to less sexual activity and substance use, I would centrally say that considering this a key protective factor is one of the prime suspects for reasons kids are depressed so often.

They also observe a rise in ‘experiences of violence.’ When I saw that I noted I was outright calling BS on any real effect. To the extent that ‘experiences of violence’ are rising, it is about what people are told to report experiencing changing, not about any increase in real physical violence

Also, seriously, no increase in electronic bullying since the introduction of the smart phone, are you kidding me? How is this even real? How did we do it? Huge if true.

There are a bunch of other ‘are you seriously saying this trend went this way during the pandemic?’ stats and I will spare you the rest of them.

On to their depression stats, which match other sources.

I am curious about the 13% or more who experienced persistent feelings of sadness or hopelessness without poor teen mental health

This New York Magazine article says that ‘No, Teen Suicide Isn’t Rising Because Life Got Objectively Worse.’

Life isn’t ‘objectively worse,’ in this view, because the economy has improved as has our social safety net.

I would suggest this misunderstands what it means for things to be ‘objectively worse.’

I would assert that the existence of social media can absolutely make things ‘objectively’ worse if it works like its critics fear

To the extent one disagrees with this, the word ‘objective’ is being used to dismiss concerns other than access to material goods as not ‘objective.’

That does not tell us what is causing the kids to not be okay. It does make it less likely that we should be looking mostly or exclusively at social media, or at wokeness, or fear of climate change, since none of those can explain the early 1990s. That would require the two peaks have distinct causes.

Children Versus Adults

To the extent that their happiness with things in general changed, it does not correspond at all to when the kids were relatively not okay.

School Daze

A reasonable suspect for the core problem would be American schools. Remember that suicide rates go up during school, go down during summer vacation. When there were school closings, suicide rates declined. (schooling)

The problem with the school hypothesis is that while it explains kids not being okay in general, it does not explain the changes over time

I can tell a story for why school was especially toxic during the pandemic. ‘Remote learning’ was a new level of dystopian nightmare. That still does not fit the graphs

Thus I would say that school lays the foundation for children to be miserable. I would say school directly causes children to be miserable. It still does not seem to explain why things recently got so much worse

Liberal Politics

The story this data is telling is something like:
Liberal children are more often depressed than conservative children.
Since 2005, depression among children has risen across the board.
Since 2005, something has made liberal girls in particular more depressed.

Causation between politics and depression is not obvious here. All of these stories seem plausible:
If you are depressed, that tends to make you liberal. Change is needed.
If you are liberal, that tends to make you depressed. Change is needed.
If you are liberal, you tend to identify as depressed more often.

A left-wing theory by authors of a paper on the subject of why this is happening is that it is all because bad political actors are doing bad things, which are very depressing if you are a good person who understands

To me and to Matt Yglesias, this sounds like a story of political discourse among and directed to young people making them depressed.

This is instead a story about how these bad things became central and constant parts of the discourse that young people felt socially obligated to discuss and endorse. As the authors say, they became ‘unavoidable.’

The youth both had to endorse that these things were depressing and terrible and unacceptable, focusing carefully on the current thing of the week, and signal that this depressed them, and also express the belief that these things were getting worse.

regardless of the truth the claims subject to these social cascades. (sociogenic)

Regardless of the extent to which these issues are central, I highly endorse not catastrophizing, and not encouraging others to catastrophize.

It is important to cultivate the skill of not letting such things bring you down, and to encourage a discourse and culture that helps others not be brought down rather than reinforcing such failure modes. Whereas, as far as I can tell, liberal discourse explicitly reinforces not doing that.

Then there is the tendency to medicate children every time they get out of line or pose any sort of problem, or are given any kind of label that needs fixing – you can imagine why worries about this happening could make one paranoid and unhappy. Also the drugs themselves often make kids unhappy.

Phones and Social Media

The Social Media Hypothesis (SMH) is both common and common sense. It is easy to see why we might expect social media to be (1) damaging to everyone, (2) especially damaging to teenagers, (3) even more damaging to girls and (4) extremely difficult to escape even when you know about the problem.

The timing also matches quite well, although it obviously can’t explain the 90s

Richard Hanania considers the SMH. He notes that given his other views he has strong motivations to reject the SMH and instead blame anything else and especially to blame wokeness. Despite this, he is convinced: After looking at various kinds of evidence, however, I have changed my mind.

Johnathan Haidt notes that his larger story is the transition from play based childhood to phone based childhood. In brief, it’s the transition from a play-based childhood involving a lot of risky unsupervised play, which is essential for overcoming fear and fragility, to a phone-based childhood which blocks normal human development by taking time away from sleep, play, and in-person socializing, as well as causing addiction and drowning kids in social comparisons they can’t win. (zero-sum game)

He also frames the social media, I think correctly, as primarily about network effects rather than individuals or dose-response effects. The existence of social media transforms the social landscape. As an adult, one can mitigate this by choosing one’s friends and colleagues. As a young student in school, you have no chance.

The theory here was that this hits girls much harder than boys because girls are far more vulnerable to social comparison, which social media forces on them even more and worse than traditional media.

What I am confident is making this worse is the transition away from seeing your followers or friends stuff and towards algorithmic feeds. When I post something on Twitter that the algorithm does not care for, maybe 10% of my followers will see it. When I post something that catches fire, sky is the limit. Most of my views on Twitter come from a handful of posts – which means your likes and retweets actually matter a lot for effective visibility, and are appreciated.

Noah Smith agrees: It’s probably the phones. Here’s his fresh variation of everyone’s favorite graphs.

Once again, something quite terrible happened around 2011 or 2012.

And look, it’s the reverse version. (graph of teens spending time with friends, dropping steadily since 2000) (thread with some stats/charts I pulled)

It is noteworthy that the 1990s did not have especially low numbers here, things slowly got worse until finally the bottom fell out.

Noah doesn’t consider it that meaningful to differentiate the phones versus social media, since to have anything like its full effects social media requires smartphones. It does still point to very different best responses now.

A consequence of social media and phones that needs more attention is the destruction of privacy and the expansion of the permanent record. I don't think this is that big an issue until you're 17, so doesn't explain the data.

Is This Core Case for the Social Media Hypothesis Convincing? Is It The Phones? Mostly, yes.

I am convinced that the central problem here is likely a combination of phones, social media and the resulting physical social isolation.

There are a lot of different angles of evidence gathered to support the phone hypothesis. Except for the need to explain the 1990s, and discounting the misleading alarmist stuff, they all point in the same direction – the phones.

There is little question that something very terrible for teenagers happened around 2012. That rules out an economic cause. That rules out some weird political shift, other than the shift in discourse that came from the rise of phones and social media. The only two things suggested that could possibly match the timing at all are a cultural shift (e.g. towards some form of social-media-and-being-online-reinforced wokeness.)

What About Other Causes?

Tablets are a serious problem for younger children. I know of a number of examples of children who are very very attached to their tablets. If you are not careful, they will lose interest in everything else in favor of a bunch of optimized dreck.

Computers and video games are the screens that came online in the 1990s. Could that have been what happened then? In the early-mid 1980s, you played video games in an arcade. The arcade was social

A common pattern when dismissing the dangers of new tech is to say ‘remember how everyone said television would rot our brains and destroy our communities? You know, like Socrates worried about books?’ Books turned out to be good, but television (TV)? Have we considered that perhaps, while there is also plenty of good television out there that enhances our lives and culture, those warnings were one hundred percent right? And then it pretty much happened? I think it pretty much did happen.

The social isolation problem, and lack of community, predates smart phones and even widespread video games and computers. Remember Bowling Alone? I do think that phenomenon was real and important then, and has been massively amplified now. Its other causes count too.

What about if all this wasn’t about actual economic problems, or other problems, but the newfound perception of those problems? That could alter the timing and brings us back to the smart phones.

We could no longer maintain our mismatches between our rhetoric and reality. Kids could no longer be gently introduced to the realities of the world over time.

How much of that is economic reality versus social reality or other problems one can debate.

I strongly don’t believe in lying to kids, that still doesn’t mean dumping the weight of the world on them all at once at age seven, or even twelve, or ideally even seventeen. At this point, we kind of do exactly that.

Looping back to the economic and competition aspects, kids nowadays can very clearly see how rough it is going to be out there for them. That, in turn, makes their real physical situation worse. (Precariat)

If kids are mostly going around being kids and then colleges judge them, and then employers judge them on where they went to college and what they studied, and there are lots of consequences for messing up at every step, and the losers suffer a lot.

If that is all happening and it is common knowledge, that is so much worse.

Everyone is suddenly spending their adolescence in a rigid thankless contest

Then the distribution of final economic outcomes doesn’t change and everyone has had less fun, is less of an actual person and is in worse mental health. Moloch triumphs. This seems really, really bad.

If all of this is happening, it is common knowledge, except that people think things are even worse than they are, that’s even worse than an accurate perception. Parents and kids are convinced you absolutely have to stay on the straight and narrow and go to the best possible schools and all that, or you’re finished. And, frankly, that’s not true.

A similar pattern on the personal level is what happens if we now more reliably learn about, understand and process information about bad things that have happened to us

adjustments made in response are often harmful. One way that happens is that we calibrate our understanding of a problem based on the group that couldn’t hide the problem, and then tell the mild cases that could have muddled through that they have this terrible problem, and how bad it must be. It often goes poorly. The correct level of walking things off is not zero.

Compare this to the ‘revolt of the public’ theory. What if this revolt is that much more widespread, for most of the same reasons?

Perhaps We’re Spouting All Sorts of Obvious Nonsense?

What Is To Be Done about Ubiquitous Phone Use and Social Media?

The inadequate equilibrium with social media problem is obvious. Even for an individual, turning off one’s phone or deleting one’s accounts is hard. Once all your peer group’s social coordination is on social media, what are you going to do about it?

As an obvious parallel, at times in the past, smoking was a key part of socialization. If you wanted to hang out with the cool people, you had to smoke. So lots of people smoked, which meant lots of other people smoked, and everyone was colder and sicker and poorer and worse off. It is at least reasonable to propose a regulation to shift the norm away from that, where restricting people’s choices makes everyone better off. If it’s kids, the case is that much stronger.

Externalities are a classic market failure that, if big enough and important enough, justify intervention to fix them, ideally in the form of a tax but alternatives can easily be superior to nothing.

Would I support stricter age restrictions on social media or smartphone use? I am loathe to have the government come in and start restricting our ability to communicate.

If we can’t reduce usage, one idea as hinted at above is perhaps to make evidence drawn from children’s social media and electronic communications broadly inadmissible. Make it hide auto-delete or at least hide from everyone else by default after a while, viewable only by friends, illegal to consider in any school disciplinary action or admissions process or job interview. (privacy)

Social media and phone use are about to crash head first into the problem of ubiquitous and rapidly advancing AI. If it wasn’t for AI, I’d say I expect the social media issues are now about as bad as they are going to get and should improve as we adapt to the new world, and that the culture should start shifting soon to get people to look up from their phones more often. Instead, things are about to get super weird, in ways that are very hard to predict.

We can also help this along by improving alternatives to phone use. If children aren’t allowed to go places without adults knowing, or worse adults driving them and coming along and watching them, what do you think they are going to do all day?

allow and encourage free range childhood (free-range kids)

What is To Be Done About Information Overload?

the unfortunate answer is, essentially, ‘not much.’

What About Other Economic and Social Problems?

divided into perception versus expectations versus reality

For fixing perception, there are the places where the problem is that perception is wrong and the places where perception is right but causing problems.

Where it is wrong and doing harm, the ideal answer is obvious, but also sometimes one must actually ‘fix’ the problem due to such perceptions alone.

Where it is right but doing harm, that is tricker. Going back to lying is not going to work.

That brings us to the problem of rising expectations. In many cases, our expectations for economic or social progress have become completely divorced from physical reality and how humans and civilization work. (Golden Age)

Then what? Lowered expectations? Hopefully via better understanding of the problems, but also perhaps when that isn’t working some ordinary despair and willingness to compromise, realizing that things are actually pretty good, considering?

Conclusion and First Step

The good news is that for most of us, ubiquitous use of smart phones and social media is transparently terrible for us.

You can profitably be the change you want to see in the world here. My recommendation for adults (including myself, some of this is aspirational)is that you do the following, and insist kids do the same:

Don’t ever passively use social media on your phone. No scrolling, ever

Never scroll. Be present. In case of boredom, see the approved uses list.

Actively fine uses of a smartphone include: Maps and directions, phone calls and video calls, reading e-books, playing music and podcasts and audio books, quickly looking up relevant information, storing tickets or otherwise showing others info.

…and make concerted efforts to see people in person as often as possible.