Political and health authorities who evade the harsh realities girls face and obsess over smartphones and social media are proving incompetent to govern young people.
I have been skeptical of the girls mental health crisis is due to phones and or social media. I have seen quite a bit of the Jon Haidt work and I never was convinced that is was more than correlation. The work that points to this shows all the tell-tale signs of cherry picking and ignoring confounders, there has been a strong skepticism from other scientists about this and like so many flawed observations in psychology this effect seems to be riddled with confounders that have been ignored. It seems to be confirmation bias. Which shows that psychologists are just as sensitive to being fooled by non-financial conflicts of interest as the rest of the scientists are. This seems to be another case of psychologists who have good intentions and are a bit too confident of themselves. I don't think they have any ill-intentions I just think this is just another case of why psychology as a science could do with a good cleanup and more standardised and transparant data gathering. And more ways to control for conscious / unconscious influence by the researchers.
I feel like this issue is a “both / and.” It is, most likely the case that the parental structure is the primary or greatest source of mental health issues in boys and girls, but devices and social media definitely also play a role. To sacrifice one for the other, I believe, would be irresponsible. Both need much attention.
That is true, of course. But the dark side of social media, phones, and Big Tech are also true for ALL ages, not only children and teens.
If we really want to solve the Big Tech problem, for example, here's how to throw the proverbial One Ring into the fire for good, and best of all, without violating anyone's rights or causing an undue moral panic. To wit:
This is right up my alley - I write about developmental trauma. The survey questions point to the same Adverse Childhood Experiences we know since 1998 that are related to negative physical and mental health issues later in life.
Yes. It's not as dramatic as for girls -- and we can discuss why that is -- but here are a few highlights. Boys report only half the frequently poor mental health levels girls do, but they show the same pattern: just 10.5% of boys who are not abused by parents and household adults report poor mental health, compared to 19.0% for occasionally abused boys, 33.3% for sometimes abused boys, and 50.5% for frequently abused boys. One-third of frequently abused boys report suicide attempts, compared to 2.5% of never-abused boys. The patterns are similar as for girls. And the larger, official and media pattern is: why aren't these compelling patterns being talked about?
Gosh. I agree the lack of visibility of these issues is a big problem. I suspect there are a few reasons behind it: the topic elicits a strong emotional reaction, we don’t really have any good solutions for it, and there’s still a tabu when it comes to the parent-child relationship and admitting that something is going terribly wrong there.
Back in the day, Dr. Merton Strommen (statistician, founder of the Search Institute in Minneapolis) did a large meta-study that showed a clear correlation between at-risk behaviors and answering affirmatively the question, "My mom and dad don't get along." The ACE study connects more traumatic experiences with longterm negative health effects. I like Dr. Stommen's observation b/c it's easy to ask any kid--even ones with a low ACE score--and see the struggle. It's common sense. I wonder what the correlation is between this question and time on devices.
The only question on the CDC survey that relates to this issue is whether teens experienced parents or adults in their homes being violent toward each other. Yes, teens in these violent homes spend more time on social media (84% more than daily) than teens whose homes are not violent (78%). Home violence is a much better predictor of teens' poor mental health (r=0.23) than time spent on social media (r=.04). So, the pattern you mention still prevails.
I find the report interesting but your article a little too dismissive of the impact of smartphones and social media on children. Hell, everything they say about the dangers of both on them apply equally to adults (and in fact is one of the reasons I'm on SM less). I'd call this another piece of the pie without ignoring a lot of evidence behind smartphones and SM too.
That is true, of course. But the dark side of social media, phones, and Big Tech are also true for ALL ages, not only children and teens.
If we really want to solve the Big Tech problem, for example, here's how to throw the proverbial One Ring into the fire for good, and best of all, without violating anyone's rights or causing an undue moral panic. To wit:
That’s a good start, for sure, and yeah, I know the mental health problems aren’t just age-specific. Folks have the right to ruin their mental health this way. I don’t spend much time on social media, mostly Substack (which isn’t SM but does have this Notes thing built into it, so it’s got a SM angle!)
Jonathan Haidt encourages bringing back more F2F community which I really like, also maybe more after-school events (presumably, phoneless). I remember when kids were on swim teams, had piano lessons, and sometimes church activities. And bringing back a play-based childhood where they can get into trouble, get hurt, get into fights and learn many skills as a result that today’s Gen Z marshmallows haven’t. I’d like to see similar activities for adults who have lost their social and friendship skills over the last several years (not all of it lockdown-related).
Let me get this straight. You took a CDC survey of at-risk girls who are all in homes with significant alcohol, drug, and sexual abuse. And because social media was not a factor, or even mentioned, alongside the alcohol, drug and sexual abuse, you extrapolated from this that social media is just fine. Is this a joke?
Please read more carefully. I cited all girls in the CDC survey, including those from healthy, somewhat troubled, and severely troubled families and compared their mental health, suicide, and other risks. I also included social media use. These are clearly explained in the table and table notes, as well as the text, and in the CDC's analyses. What I'm having trouble understanding is why parent and household-adult troubles (associated with over 80% of teens' depression and problems) are being ignored, while social media (associated with zero to a tiny fraction of teens' problems) are being wildly hyped.
Social media are big bad corporations and this seems like the attention grabbing part. The David vs goliath struggle is more appealing to both the media and the researchers pointing it out. Family associated problems are cliché and boring and nobody important cares about those (i am deliberately being facetious here). And maybe it's also an issue of what can be easily controlled and what not? It's easier to stop using social media than it is to solve the problems with families, and until now very few structural reforms help in that front. Nobody actually seems to know how to solve family problems. There are a lot of "experts" just like the "it's all the phones" that make questionable claims about how to solve family issues. But I doubt there is much good and reliable science in that front. But maybe there is and I am just ignorant about it. Who knows.
Thanks Mike. I read your piece again. I apologize for misrepresenting the CDC survey or your analysis. It wasn’t clear to me how you arrived at “well over half of America’s under-16 girls grow up in severely troubled families” from 3200 12-15 year old girls self reporting on a government Youth Risk Behavior survey. I see now those 3200 girls were a subset of a much larger survey, and I’ve spent more of my morning than I would have liked reading through the CDC data documentation. I was also confused by the percentage values in your table columns. Do they reference the entire cohort of 12-15 year old girls, or just the subset within the column? If the former, then I read your table to say 45% of all girls 12-15 are affected by a Parent/Guardian jailed. That strikes me as absurd, and can’t possible be representative of the general population. If its the latter, then the conclusions you derive from comparing the Healthy percentages to the other cohorts are not valid. Either way, a survey that seems to be saying that up to 89% of girls are living in dysfunctional family households, is so extraordinary that we can’t simply take it at face value and then use it as the basis for drawing conclusions that other problems don’t exist.
It happens that I probably agree with you, to a degree. It would be astounding if there was not some significant amount of moral panic infecting our conversation around social media and children. If anyone is directly pinning teen suicide on social media, and ignoring everything else, they are either confused or possibly engaging in dishonest activist rhetoric, and I am deeply sympathetic to pushing back on that. However, it seems equally short-sighted and possibly dishonest to cite the existence of other bad things and draw from that the conclusion that social media is harmless. I have not read Jonathan Haidt as saying that social media is the sole cause of teen mental health issues, but it does seem rather obviously to be a contributor. Given what we know, and don’t know, about how young minds develop, coupled with what we know about how social media exploits several of our weaknesses for maximizing attention, it is simply not an option to sit back and pretend there is nothing to worry about. It’s not obvious that paying attention to the impact that social media has on children is necessarily subverting other efforts to fix other problems. Frankly, I’m not aware that significant resources are being redirected to regulating social media usage in children. Are you suggesting that the excessive hype and noise regarding social media, that is largely happening on social media, is having some kind of malign effect us?
Thanks, I see where the confusion lies. My table is unclear. Its three divisions of parent/adult problems (which are not additive across the table) are as follows:
Healthy family: parents/adults never abused alcohol/drugs, were never depressed/suicidal, were never jailed, were never domestically violent or violently or emotionally abusive toward kids – even one time – and/or were not absent more than occasionally. This is a very tough standard; just 13% of families in the CDC sample meet it. Teens in those families had very, very low rates of problems themselves.
Some problems: families whose adults were occasionally emotionally abusive toward kids and/or sometimes absent. More troubled behaviors by adults show up. Some emotional abuse or parental absence accompanies 20% of parents having been jailed, 44% having been depressed, 8% chronically absent, etc. (the 20% of parents jailed are WITHIN this category, which comprises 30% of the total CDC sample.) Teens raised in these families show considerably more problems on average.
Severely troubled: includes families whose parents/adults have been or are alcohol/drug abusers, depressed/suicidal, jailed, violently abusive toward each other, always or frequently absent, and/or violent or regularly emotionally abusive toward kids. Teens in this category – an alarming 57% of the families in the CDC sample – report more seriously troubled behaviors themselves, indicating their families suffer multiple and frequent rather than past or occasional problems.
This pattern gives me confidence that teens are not grossly exaggerating adult troubles. True, the latter two categories encompass a range of behaviors that might include some one-time problems that never multiplied or recurred. But overall, teens are not reporting one-time instances of a parent/guardian getting drunk, yelling, taking Prozac, throwing a chair, or being AWOL. We are mostly looking at multiple, serious, continuing family problems.
This reality also shows up independently in death and ER statistics (5 million adults ages 25-64 treated in hospital ER for drug/alcohol overdoses in 2022 alone, a crisis that soared during the same 2010-21 period teens became more depressed). Five in six murdered girls ages 12-17 are killed by adults, not peers, and well over half the adult murderers are 25 and older, the FBI reports.
So, I don’t think teens’ reports of adult troubles and abuses (domestic problems the CDC call “common” and associates with 66% of teens’ sadness and 85% of teens’ suicide attempts) can be dismissed as adolescent drama, exaggeration, vengefulness, joking, or snowflakery. If we call them liars, we run into a big problem. These are the same teens who report the high rates of peer cyberbullying, school bullying, depression, sadness, suicide attempts, etc., we have proven eager to believe and run with. If we want teens to be less depressed, we can’t keep picking and choosing the mouse it suits us to challenge while ignoring the elephant.
Even with that correction, your table has a combined 648 of the 3214 survey respondents (20%) reporting a jailed parent. If your table is accurately reflecting the CDC's data, then the survey is dramatically over-representing the most troubled segment of kids in America. And I'm assuming this is not a one off mistake as the other categories of problem you cite are also wildly inflated. 43% have a depressed/suicidal parent? 39% experienced violent abuse in the home? These numbers are not credible, or at the very least, are not representative of the general population. I'd throw the entire dataset out as worthless.
I'd also toss the numbers on sadness, depression and suicide from this report as being similarly useless. You can't get reliable information from 12-15 year old girls self-reporting their own "sadness." If anyone is citing this data as evidence of a mental health crisis, they should stop. There are plenty of good reasons to restrict children from a 24 hour a day firehose of advertising, misinformation, propaganda and stupidity, without resorting to fear tactics and bulllshit "scientific" surveys.
No. As I said before, the categories are not additive. The Troubled and Some Troubles categories overlap, since they are divided into whether parents/adults also suffered addiction, depression, absence, and jailing in addition to abusive and violent behaviors. I’m experimenting with what types of parental problems most predict teen problems.
Others are welcome to download the CDC survey and divide the categories of parental/adult troubles as they want. They will find the same pattern I did: teens’ mental health, suicide, and other problems escalate sharply along with more parental/adult troubles.
How do you know that it can’t possibly be true that high percentages of teens of 2023 once (or at many points) in their growings-up suffered a drug/alcohol-abusing, jailed, severely depressed/suicidal, and/or absent parent, and/or a domestically violent parent, along with an emotionally abusive parent/household adult? Or, multiple adverse situations?
On what basis do you argue for just tossing out all surveys that find these kinds of numbers as “not credible”? Have you compared CDC survey numbers to corroborating statistics, such as serious-offense arrests, incarcerations, ER treatments, overdose deaths, etc., of parent-aged adults and calculated that those must be thrown out as well?
You claim: “you can’t get reliable information from 12-15 year-old girls.” How do you know? Do you have some basis for branding girls as liars? By what evidence do you know more about girls’ emotional states and lives than they do? (Please don’t cite worthless personal anecdotes or impressions. No one should care about your niche, or mine, in evaluating the experiences of 40 million teenagers.)
You call for restricting teens from “a 24 hour a day firehose of advertising, misinformation, propaganda and stupidity.” Do you mean American life in general? I assume your impression of universal online evils and zero benefits reflects your own internet experience. On what grounds do you assume teens use the internet the way you do?
Personally, I find online life enriching, informative, useful, and a great enhancement complementing offline life – that’s the way I use the internet. If I find a site offensive, the <delete> and <block> tabs are wondrous remedies. Unfortunately, I (and teens) can’t use those great online features offline. Please provide your own scientific analyses as I have, so we can see exactly how you arrive at your sweeping statements.
Your table includes 3,214 girls in the 12-15 year age range. You’ve divided those discretely into 3 family situation groups. Healthy: 430; Some Problems: 947; Severely Troubled: 1837
Some Problems: 20% had incarcerated parent. 20% of 947 = 189
Severely Troubled: 25% had incarcerated parent. 25% x 1837 = 459
According to your table, a total of 648 students had incarcerated parents. As I mentioned in my last email, that’s 20.1% of the total 3214, or one in five students have had an incarcerated parent. That’s a huge claim. The Center on Juvenile and Criminal Justice estimates 7% of American children experience parental incarceration. Your claim is 3 times the CJCJ estimate. Perhaps the CJCJ estimate is wrong? If not, then the data in your table is wrong. Let’s try another data point: Parent/guardian depressed/suicidal.
Some Problems: 44% with depressed/suicidal parent. 44% of 947 = 417
Severely Troubled: 54% with depressed/suicidal parent. 54% x 1837 = 992
Your claim is that a total of 1409 students, or 45% of all students have a depressed or suicidal parent. The National Institute of Mental Health cites only 9.3% of U.S. adults in the 26-49 year age range with past year prevalence of major depressive episode (2021). Allowing for the child to have two chances of having a depressed parent, your claim is still twice the NIMH number. That’s not a small disparity, and in the absence of an explanation, it’s disqualifying.
My basis for tossing out your data is that on the few data points I bothered to check, your numbers are completely out of sync with reality. Until you can account for that, the entire dataset should be treated as not credible. Let’s assume you have not made a mistake in how you’ve extracted data from the CDC survey. We are left with two possible explanations:
1) the survey is over representing students with troubled family situations
2) the survey is not capturing accurate information
If it’s the first explanation, then the survey may still be useful for researchers studying aspects of troubled youth. But you can’t claim its representative of the general population, and you can’t use it for the kind of argument you are making. If it’s the second explanation, then that would be really interesting, and you should be more curious about that. I did not call all 12-15 year old girls “liars.” I said you can’t get reliable information from 12-15 year old girls self-reporting, and that if anyone is citing this as their evidence of a mental health crisis, they should stop. I should have been clearer. If your only source of information is a government survey, administrated by the schools who participated in the survey, and it includes up to 107 questions on every possible dimension of harm or risk that a 12-15 year old might encounter, then you should handle the results carefully, and not pretend you’ve done an accurate evaluation. There are any number of reasons why self-reported results from 12-15 year old kids in that context might not be reliable, and none of them require an accusation of lying.
I also get tremendous value out of the internet. I spent over 20 years in web development making my own minor contributions to it. As a dad with three kids, I am careful and conscientious about how I expose them to the internet, and I’ve had some success in helping them get enormous value out of it. Youtube can be amazing. I think measuring “screen time” is stupid if you are not accounting for what is actually happening on their screen. A computer is just a tool, and kids should learn how to use it well. The problem isn’t the entire internet, although if you are the kind of parent that just abandons your kid to figure out the internet on their own, then it probably is. The problem is specifically social media, and the deliberate manner in which it exploits attention and indiscriminately spreads bad ideas. As an adult, you are equipped perhaps to manage all dimensions of "American life in general." Children are not, but you seem to be saying lets give them exactly the same freedom that you have. No one serious is actually saying that social media is the ONLY cause of an apparent youth mental health crisis. But I cannot for the life of me understand why you are on a mission to obfuscate and deny that there is any problem at all.
Thank you for your relentless advocacy for children. I gasped when I read his words “you can’t get reliable info from 12-15 yr old girls.” The childism ALWAYS rears it’s ugly head
I have to respectfully disagree, based on my personal observations. While they are anecdotal, surveys can also be misleading. If you gave a survey to a bunch of alcoholics how many of them would admit that alcohol use is the cause of their problems? Some, but not as many as it would be true for.
My two teenagers come from a boring, normal family. We’re still married, no abuse, no substances, no infidelity, college grads, decently well-off financially, live in a safe neighborhood, decent schools. Loving extended family that we’re close to. When the kids were young they always had parents to turn to who would help them and listen to them. The house usually looked like a craft store exploded - my husband and kids are super creative and always making things and doing projects. We’re not perfect, and I can remember times we yelled when we shouldn’t have or weren’t as patient as we could have been or my husband and I argued, but overall I think anyone would consider us responsible parents. Both kids were happy and doing well until my oldest started high school, I let her get on social media thinking it would help her fit in, and she fell down a Tumblr rabbit hole that she may never recover from. My youngest had better social connections and never got internet-obsessed (although she does her fair share) and is doing well. I can 100% guarantee you that internet brainwashing is responsible for my daughter going off the rails. She of course would deny it, because if she admitted it she’d have to give up her unhealthy coping mechanism. Just like the alcoholics.
Thanks for comment. I obviously can't comment on your family specifics except to agree with you on two counts. First, anecdotes are not necessarily generalizable, so your family as you characterize it may not be representative of larger trends. Second, just as addicts of all kinds can be in denial, so addicted, abusive, mentally troubled, crime-prone, absent, and/or violent parents and adults can insist their behaviors have nothing to do with their teens' problems -- and these parents are helped with their unhealthy denial by social-media-blaming authorities who pretend such troubled parents don't exist. Third, just as we respect your comments about your family, I think we have to respect the reports by 87% of the 3,200 girls to the CDC survey that their families suffer at least one of these adult problems, and 60% several such problems. One big reason is that these are exactly the girls who report by far the highest levels of depression, suicide attempt, self-harm, cyberbullying and other risks -- and we certainly believe them when they report their risks, don't we? So, while social media use is associated with a small fraction of teens' problems, parental/adult troubles and abuses are associated with over 80% of teens problems, and that seems to me the real issue.
I would also add that I believe our traditional schooling model (in the US at least) that exists in most public and private schools replicates prison-like conditions, is psychologically abusive in many ways, and can have severely damaging effects on children. So I wonder how putting her in the school system impacted Dee’s daughter. That’s another layer of denial our society is NOT ready to address.
Well-said, Mike. And the fact that social media has a dark side does NOT mean that we should throw out the proverbial baby with the bathwater, and ban it or unduly restrict it for people under some arbitrary age limit, which will like do far more harm than good on balance. Just like alcohol Prohibition and it's also illiberal successor, the 21 drinking age.
This is a very disturbing feature of this discussion. Haidt and some commentators reduce the vast, diverse internet, in the words of one, to: “a 24 hour a day firehose of advertising, misinformation, propaganda and stupidity.” Of course, there are terrible corners of the internet, just as there are terrible corners of churches, Boy Scouts, sports, schools, families, businesses, and every aspect of American life -- yet, none of these are the whole of any of these institutions. Teens have told us in no uncertain terms that poor mental health among some of their number overwhelmingly is driven by their parent/family's terrible corners, not by social media. Haidt et al have chosen to ignore teens' reality, to wildly exaggerate online dangers, and to completely ignore vastly greater family troubles. In fact, for teens, online and offline life go together; they are not either-or battlegrounds Haidt et al assume. The biggest difference is that online dangers are infinitely easier to avoid by a mouseclick, while a household, church, athletic, Scout, or physical-world danger is very hard to escape. Teens adapted to modern dualities clearly understand this, while authorities persist in delusional culture-war panics. By crowding out discussion of family troubles, the social-media panickers are an increasing danger to young people.
I have been skeptical of the girls mental health crisis is due to phones and or social media. I have seen quite a bit of the Jon Haidt work and I never was convinced that is was more than correlation. The work that points to this shows all the tell-tale signs of cherry picking and ignoring confounders, there has been a strong skepticism from other scientists about this and like so many flawed observations in psychology this effect seems to be riddled with confounders that have been ignored. It seems to be confirmation bias. Which shows that psychologists are just as sensitive to being fooled by non-financial conflicts of interest as the rest of the scientists are. This seems to be another case of psychologists who have good intentions and are a bit too confident of themselves. I don't think they have any ill-intentions I just think this is just another case of why psychology as a science could do with a good cleanup and more standardised and transparant data gathering. And more ways to control for conscious / unconscious influence by the researchers.
I feel like this issue is a “both / and.” It is, most likely the case that the parental structure is the primary or greatest source of mental health issues in boys and girls, but devices and social media definitely also play a role. To sacrifice one for the other, I believe, would be irresponsible. Both need much attention.
Thanks for comment, I get it a lot -- but I feel it's an oranges-apples comparison. I'll talk about why next.
That is true, of course. But the dark side of social media, phones, and Big Tech are also true for ALL ages, not only children and teens.
If we really want to solve the Big Tech problem, for example, here's how to throw the proverbial One Ring into the fire for good, and best of all, without violating anyone's rights or causing an undue moral panic. To wit:
https://21debunked.blogspot.com/2024/01/how-to-solve-big-tech-problem-without.html
Well-said as usual, Mike. The kids aren't alright BECAUSE the ADULTS aren't alright. Shout it from the rooftops!
This is right up my alley - I write about developmental trauma. The survey questions point to the same Adverse Childhood Experiences we know since 1998 that are related to negative physical and mental health issues later in life.
Just curious - is there similar data for boys?
Yes. It's not as dramatic as for girls -- and we can discuss why that is -- but here are a few highlights. Boys report only half the frequently poor mental health levels girls do, but they show the same pattern: just 10.5% of boys who are not abused by parents and household adults report poor mental health, compared to 19.0% for occasionally abused boys, 33.3% for sometimes abused boys, and 50.5% for frequently abused boys. One-third of frequently abused boys report suicide attempts, compared to 2.5% of never-abused boys. The patterns are similar as for girls. And the larger, official and media pattern is: why aren't these compelling patterns being talked about?
Gosh. I agree the lack of visibility of these issues is a big problem. I suspect there are a few reasons behind it: the topic elicits a strong emotional reaction, we don’t really have any good solutions for it, and there’s still a tabu when it comes to the parent-child relationship and admitting that something is going terribly wrong there.
Back in the day, Dr. Merton Strommen (statistician, founder of the Search Institute in Minneapolis) did a large meta-study that showed a clear correlation between at-risk behaviors and answering affirmatively the question, "My mom and dad don't get along." The ACE study connects more traumatic experiences with longterm negative health effects. I like Dr. Stommen's observation b/c it's easy to ask any kid--even ones with a low ACE score--and see the struggle. It's common sense. I wonder what the correlation is between this question and time on devices.
The only question on the CDC survey that relates to this issue is whether teens experienced parents or adults in their homes being violent toward each other. Yes, teens in these violent homes spend more time on social media (84% more than daily) than teens whose homes are not violent (78%). Home violence is a much better predictor of teens' poor mental health (r=0.23) than time spent on social media (r=.04). So, the pattern you mention still prevails.
Amazing work as ever from Mike Males right here, analyzing the true sources of teen's mental health issues (hint: it's not the phones).
Amen
If we really want to solve the Big Tech problem, here's how to throw the proverbial One Ring into the fire for good.
https://21debunked.blogspot.com/2024/01/how-to-solve-big-tech-problem-without.html
I find the report interesting but your article a little too dismissive of the impact of smartphones and social media on children. Hell, everything they say about the dangers of both on them apply equally to adults (and in fact is one of the reasons I'm on SM less). I'd call this another piece of the pie without ignoring a lot of evidence behind smartphones and SM too.
That is true, of course. But the dark side of social media, phones, and Big Tech are also true for ALL ages, not only children and teens.
If we really want to solve the Big Tech problem, for example, here's how to throw the proverbial One Ring into the fire for good, and best of all, without violating anyone's rights or causing an undue moral panic. To wit:
https://21debunked.blogspot.com/2024/01/how-to-solve-big-tech-problem-without.html
That’s a good start, for sure, and yeah, I know the mental health problems aren’t just age-specific. Folks have the right to ruin their mental health this way. I don’t spend much time on social media, mostly Substack (which isn’t SM but does have this Notes thing built into it, so it’s got a SM angle!)
Jonathan Haidt encourages bringing back more F2F community which I really like, also maybe more after-school events (presumably, phoneless). I remember when kids were on swim teams, had piano lessons, and sometimes church activities. And bringing back a play-based childhood where they can get into trouble, get hurt, get into fights and learn many skills as a result that today’s Gen Z marshmallows haven’t. I’d like to see similar activities for adults who have lost their social and friendship skills over the last several years (not all of it lockdown-related).
Let me get this straight. You took a CDC survey of at-risk girls who are all in homes with significant alcohol, drug, and sexual abuse. And because social media was not a factor, or even mentioned, alongside the alcohol, drug and sexual abuse, you extrapolated from this that social media is just fine. Is this a joke?
Please read more carefully. I cited all girls in the CDC survey, including those from healthy, somewhat troubled, and severely troubled families and compared their mental health, suicide, and other risks. I also included social media use. These are clearly explained in the table and table notes, as well as the text, and in the CDC's analyses. What I'm having trouble understanding is why parent and household-adult troubles (associated with over 80% of teens' depression and problems) are being ignored, while social media (associated with zero to a tiny fraction of teens' problems) are being wildly hyped.
Social media are big bad corporations and this seems like the attention grabbing part. The David vs goliath struggle is more appealing to both the media and the researchers pointing it out. Family associated problems are cliché and boring and nobody important cares about those (i am deliberately being facetious here). And maybe it's also an issue of what can be easily controlled and what not? It's easier to stop using social media than it is to solve the problems with families, and until now very few structural reforms help in that front. Nobody actually seems to know how to solve family problems. There are a lot of "experts" just like the "it's all the phones" that make questionable claims about how to solve family issues. But I doubt there is much good and reliable science in that front. But maybe there is and I am just ignorant about it. Who knows.
Thanks Mike. I read your piece again. I apologize for misrepresenting the CDC survey or your analysis. It wasn’t clear to me how you arrived at “well over half of America’s under-16 girls grow up in severely troubled families” from 3200 12-15 year old girls self reporting on a government Youth Risk Behavior survey. I see now those 3200 girls were a subset of a much larger survey, and I’ve spent more of my morning than I would have liked reading through the CDC data documentation. I was also confused by the percentage values in your table columns. Do they reference the entire cohort of 12-15 year old girls, or just the subset within the column? If the former, then I read your table to say 45% of all girls 12-15 are affected by a Parent/Guardian jailed. That strikes me as absurd, and can’t possible be representative of the general population. If its the latter, then the conclusions you derive from comparing the Healthy percentages to the other cohorts are not valid. Either way, a survey that seems to be saying that up to 89% of girls are living in dysfunctional family households, is so extraordinary that we can’t simply take it at face value and then use it as the basis for drawing conclusions that other problems don’t exist.
It happens that I probably agree with you, to a degree. It would be astounding if there was not some significant amount of moral panic infecting our conversation around social media and children. If anyone is directly pinning teen suicide on social media, and ignoring everything else, they are either confused or possibly engaging in dishonest activist rhetoric, and I am deeply sympathetic to pushing back on that. However, it seems equally short-sighted and possibly dishonest to cite the existence of other bad things and draw from that the conclusion that social media is harmless. I have not read Jonathan Haidt as saying that social media is the sole cause of teen mental health issues, but it does seem rather obviously to be a contributor. Given what we know, and don’t know, about how young minds develop, coupled with what we know about how social media exploits several of our weaknesses for maximizing attention, it is simply not an option to sit back and pretend there is nothing to worry about. It’s not obvious that paying attention to the impact that social media has on children is necessarily subverting other efforts to fix other problems. Frankly, I’m not aware that significant resources are being redirected to regulating social media usage in children. Are you suggesting that the excessive hype and noise regarding social media, that is largely happening on social media, is having some kind of malign effect us?
Thanks, I see where the confusion lies. My table is unclear. Its three divisions of parent/adult problems (which are not additive across the table) are as follows:
Healthy family: parents/adults never abused alcohol/drugs, were never depressed/suicidal, were never jailed, were never domestically violent or violently or emotionally abusive toward kids – even one time – and/or were not absent more than occasionally. This is a very tough standard; just 13% of families in the CDC sample meet it. Teens in those families had very, very low rates of problems themselves.
Some problems: families whose adults were occasionally emotionally abusive toward kids and/or sometimes absent. More troubled behaviors by adults show up. Some emotional abuse or parental absence accompanies 20% of parents having been jailed, 44% having been depressed, 8% chronically absent, etc. (the 20% of parents jailed are WITHIN this category, which comprises 30% of the total CDC sample.) Teens raised in these families show considerably more problems on average.
Severely troubled: includes families whose parents/adults have been or are alcohol/drug abusers, depressed/suicidal, jailed, violently abusive toward each other, always or frequently absent, and/or violent or regularly emotionally abusive toward kids. Teens in this category – an alarming 57% of the families in the CDC sample – report more seriously troubled behaviors themselves, indicating their families suffer multiple and frequent rather than past or occasional problems.
This pattern gives me confidence that teens are not grossly exaggerating adult troubles. True, the latter two categories encompass a range of behaviors that might include some one-time problems that never multiplied or recurred. But overall, teens are not reporting one-time instances of a parent/guardian getting drunk, yelling, taking Prozac, throwing a chair, or being AWOL. We are mostly looking at multiple, serious, continuing family problems.
This reality also shows up independently in death and ER statistics (5 million adults ages 25-64 treated in hospital ER for drug/alcohol overdoses in 2022 alone, a crisis that soared during the same 2010-21 period teens became more depressed). Five in six murdered girls ages 12-17 are killed by adults, not peers, and well over half the adult murderers are 25 and older, the FBI reports.
So, I don’t think teens’ reports of adult troubles and abuses (domestic problems the CDC call “common” and associates with 66% of teens’ sadness and 85% of teens’ suicide attempts) can be dismissed as adolescent drama, exaggeration, vengefulness, joking, or snowflakery. If we call them liars, we run into a big problem. These are the same teens who report the high rates of peer cyberbullying, school bullying, depression, sadness, suicide attempts, etc., we have proven eager to believe and run with. If we want teens to be less depressed, we can’t keep picking and choosing the mouse it suits us to challenge while ignoring the elephant.
Even with that correction, your table has a combined 648 of the 3214 survey respondents (20%) reporting a jailed parent. If your table is accurately reflecting the CDC's data, then the survey is dramatically over-representing the most troubled segment of kids in America. And I'm assuming this is not a one off mistake as the other categories of problem you cite are also wildly inflated. 43% have a depressed/suicidal parent? 39% experienced violent abuse in the home? These numbers are not credible, or at the very least, are not representative of the general population. I'd throw the entire dataset out as worthless.
I'd also toss the numbers on sadness, depression and suicide from this report as being similarly useless. You can't get reliable information from 12-15 year old girls self-reporting their own "sadness." If anyone is citing this data as evidence of a mental health crisis, they should stop. There are plenty of good reasons to restrict children from a 24 hour a day firehose of advertising, misinformation, propaganda and stupidity, without resorting to fear tactics and bulllshit "scientific" surveys.
No. As I said before, the categories are not additive. The Troubled and Some Troubles categories overlap, since they are divided into whether parents/adults also suffered addiction, depression, absence, and jailing in addition to abusive and violent behaviors. I’m experimenting with what types of parental problems most predict teen problems.
Others are welcome to download the CDC survey and divide the categories of parental/adult troubles as they want. They will find the same pattern I did: teens’ mental health, suicide, and other problems escalate sharply along with more parental/adult troubles.
How do you know that it can’t possibly be true that high percentages of teens of 2023 once (or at many points) in their growings-up suffered a drug/alcohol-abusing, jailed, severely depressed/suicidal, and/or absent parent, and/or a domestically violent parent, along with an emotionally abusive parent/household adult? Or, multiple adverse situations?
On what basis do you argue for just tossing out all surveys that find these kinds of numbers as “not credible”? Have you compared CDC survey numbers to corroborating statistics, such as serious-offense arrests, incarcerations, ER treatments, overdose deaths, etc., of parent-aged adults and calculated that those must be thrown out as well?
You claim: “you can’t get reliable information from 12-15 year-old girls.” How do you know? Do you have some basis for branding girls as liars? By what evidence do you know more about girls’ emotional states and lives than they do? (Please don’t cite worthless personal anecdotes or impressions. No one should care about your niche, or mine, in evaluating the experiences of 40 million teenagers.)
You call for restricting teens from “a 24 hour a day firehose of advertising, misinformation, propaganda and stupidity.” Do you mean American life in general? I assume your impression of universal online evils and zero benefits reflects your own internet experience. On what grounds do you assume teens use the internet the way you do?
Personally, I find online life enriching, informative, useful, and a great enhancement complementing offline life – that’s the way I use the internet. If I find a site offensive, the <delete> and <block> tabs are wondrous remedies. Unfortunately, I (and teens) can’t use those great online features offline. Please provide your own scientific analyses as I have, so we can see exactly how you arrive at your sweeping statements.
Well-said. And his response to this seems to be a resounding.......
(crickets)
(And maybe some cicadas too)
Your table includes 3,214 girls in the 12-15 year age range. You’ve divided those discretely into 3 family situation groups. Healthy: 430; Some Problems: 947; Severely Troubled: 1837
Some Problems: 20% had incarcerated parent. 20% of 947 = 189
Severely Troubled: 25% had incarcerated parent. 25% x 1837 = 459
According to your table, a total of 648 students had incarcerated parents. As I mentioned in my last email, that’s 20.1% of the total 3214, or one in five students have had an incarcerated parent. That’s a huge claim. The Center on Juvenile and Criminal Justice estimates 7% of American children experience parental incarceration. Your claim is 3 times the CJCJ estimate. Perhaps the CJCJ estimate is wrong? If not, then the data in your table is wrong. Let’s try another data point: Parent/guardian depressed/suicidal.
Some Problems: 44% with depressed/suicidal parent. 44% of 947 = 417
Severely Troubled: 54% with depressed/suicidal parent. 54% x 1837 = 992
Your claim is that a total of 1409 students, or 45% of all students have a depressed or suicidal parent. The National Institute of Mental Health cites only 9.3% of U.S. adults in the 26-49 year age range with past year prevalence of major depressive episode (2021). Allowing for the child to have two chances of having a depressed parent, your claim is still twice the NIMH number. That’s not a small disparity, and in the absence of an explanation, it’s disqualifying.
My basis for tossing out your data is that on the few data points I bothered to check, your numbers are completely out of sync with reality. Until you can account for that, the entire dataset should be treated as not credible. Let’s assume you have not made a mistake in how you’ve extracted data from the CDC survey. We are left with two possible explanations:
1) the survey is over representing students with troubled family situations
2) the survey is not capturing accurate information
If it’s the first explanation, then the survey may still be useful for researchers studying aspects of troubled youth. But you can’t claim its representative of the general population, and you can’t use it for the kind of argument you are making. If it’s the second explanation, then that would be really interesting, and you should be more curious about that. I did not call all 12-15 year old girls “liars.” I said you can’t get reliable information from 12-15 year old girls self-reporting, and that if anyone is citing this as their evidence of a mental health crisis, they should stop. I should have been clearer. If your only source of information is a government survey, administrated by the schools who participated in the survey, and it includes up to 107 questions on every possible dimension of harm or risk that a 12-15 year old might encounter, then you should handle the results carefully, and not pretend you’ve done an accurate evaluation. There are any number of reasons why self-reported results from 12-15 year old kids in that context might not be reliable, and none of them require an accusation of lying.
I also get tremendous value out of the internet. I spent over 20 years in web development making my own minor contributions to it. As a dad with three kids, I am careful and conscientious about how I expose them to the internet, and I’ve had some success in helping them get enormous value out of it. Youtube can be amazing. I think measuring “screen time” is stupid if you are not accounting for what is actually happening on their screen. A computer is just a tool, and kids should learn how to use it well. The problem isn’t the entire internet, although if you are the kind of parent that just abandons your kid to figure out the internet on their own, then it probably is. The problem is specifically social media, and the deliberate manner in which it exploits attention and indiscriminately spreads bad ideas. As an adult, you are equipped perhaps to manage all dimensions of "American life in general." Children are not, but you seem to be saying lets give them exactly the same freedom that you have. No one serious is actually saying that social media is the ONLY cause of an apparent youth mental health crisis. But I cannot for the life of me understand why you are on a mission to obfuscate and deny that there is any problem at all.
Thank you for your relentless advocacy for children. I gasped when I read his words “you can’t get reliable info from 12-15 yr old girls.” The childism ALWAYS rears it’s ugly head
I have to respectfully disagree, based on my personal observations. While they are anecdotal, surveys can also be misleading. If you gave a survey to a bunch of alcoholics how many of them would admit that alcohol use is the cause of their problems? Some, but not as many as it would be true for.
My two teenagers come from a boring, normal family. We’re still married, no abuse, no substances, no infidelity, college grads, decently well-off financially, live in a safe neighborhood, decent schools. Loving extended family that we’re close to. When the kids were young they always had parents to turn to who would help them and listen to them. The house usually looked like a craft store exploded - my husband and kids are super creative and always making things and doing projects. We’re not perfect, and I can remember times we yelled when we shouldn’t have or weren’t as patient as we could have been or my husband and I argued, but overall I think anyone would consider us responsible parents. Both kids were happy and doing well until my oldest started high school, I let her get on social media thinking it would help her fit in, and she fell down a Tumblr rabbit hole that she may never recover from. My youngest had better social connections and never got internet-obsessed (although she does her fair share) and is doing well. I can 100% guarantee you that internet brainwashing is responsible for my daughter going off the rails. She of course would deny it, because if she admitted it she’d have to give up her unhealthy coping mechanism. Just like the alcoholics.
Thanks for comment. I obviously can't comment on your family specifics except to agree with you on two counts. First, anecdotes are not necessarily generalizable, so your family as you characterize it may not be representative of larger trends. Second, just as addicts of all kinds can be in denial, so addicted, abusive, mentally troubled, crime-prone, absent, and/or violent parents and adults can insist their behaviors have nothing to do with their teens' problems -- and these parents are helped with their unhealthy denial by social-media-blaming authorities who pretend such troubled parents don't exist. Third, just as we respect your comments about your family, I think we have to respect the reports by 87% of the 3,200 girls to the CDC survey that their families suffer at least one of these adult problems, and 60% several such problems. One big reason is that these are exactly the girls who report by far the highest levels of depression, suicide attempt, self-harm, cyberbullying and other risks -- and we certainly believe them when they report their risks, don't we? So, while social media use is associated with a small fraction of teens' problems, parental/adult troubles and abuses are associated with over 80% of teens problems, and that seems to me the real issue.
I would also add that I believe our traditional schooling model (in the US at least) that exists in most public and private schools replicates prison-like conditions, is psychologically abusive in many ways, and can have severely damaging effects on children. So I wonder how putting her in the school system impacted Dee’s daughter. That’s another layer of denial our society is NOT ready to address.
Amen. Prison-like schooling is indeed quite harmful. We need to move towards the Finnish education model, yesterday.
Well-said, Mike. And the fact that social media has a dark side does NOT mean that we should throw out the proverbial baby with the bathwater, and ban it or unduly restrict it for people under some arbitrary age limit, which will like do far more harm than good on balance. Just like alcohol Prohibition and it's also illiberal successor, the 21 drinking age.
This is a very disturbing feature of this discussion. Haidt and some commentators reduce the vast, diverse internet, in the words of one, to: “a 24 hour a day firehose of advertising, misinformation, propaganda and stupidity.” Of course, there are terrible corners of the internet, just as there are terrible corners of churches, Boy Scouts, sports, schools, families, businesses, and every aspect of American life -- yet, none of these are the whole of any of these institutions. Teens have told us in no uncertain terms that poor mental health among some of their number overwhelmingly is driven by their parent/family's terrible corners, not by social media. Haidt et al have chosen to ignore teens' reality, to wildly exaggerate online dangers, and to completely ignore vastly greater family troubles. In fact, for teens, online and offline life go together; they are not either-or battlegrounds Haidt et al assume. The biggest difference is that online dangers are infinitely easier to avoid by a mouseclick, while a household, church, athletic, Scout, or physical-world danger is very hard to escape. Teens adapted to modern dualities clearly understand this, while authorities persist in delusional culture-war panics. By crowding out discussion of family troubles, the social-media panickers are an increasing danger to young people.
And as you have noted years ago, a teen is actually statistically safer online unsupervised than in church unsupervised.
Well-said, Mike.