By the way, Dr. Twenge's NCVS trends are entirely dependent on the way she groups years. If we look at annual NCVS numbers the way she reports, say, annual teen suicide numbers (which are vastly smaller), the opposite conclusion can be shown for the rate of violent victimizations of teens age 12-17 by relatives:
1.5 in 2011 per 1,000 population age 12-17
1.8 in 2018
3.1 in 2021
3.3 in 2022.
That is, family abuses of teens skyrocketed during the 2011-22 period, exactly the time teens' depression also rose. Subjective presentations of tiny numbers can be shuffled around to show any result desired.
I’m sorry to post three times on Dr. Twenge’s substack challenging my article, but I examine critiques of my work carefully. I want my errors pointed out so I can fix them.
In this case, however, further examination shows the entire premise of Dr. Twenge’s argument is wrong. Her and her colleagues’ such as Jonathan Haidt, and Zach Rausch’s baseless dismissals of parents’ and grownups’ widespread violent and emotional abuses victimizing teenagers, as well as parents’ and caretakers’ depression, addiction, and criminal behaviors, is endangering young people and needs to stop.
First, the NCVS she cites is a useless measure of family violence against teenagers, one I will no longer cite. The reason: five-sixths of the interviews of teens by the NCVS are conducted in the presence of others (primarily parents). Unsurprisingly, these chaperoned teens are vastly less likely to reveal family violence than the one-sixth of teens interviewed in private (see research institute analysis, https://ww2.amstat.org/meetings/proceedings/2015/data/assets/pdf/233894.pdf ). This is why anonymously-conducted surveys like the CDC’s uncover far higher parental and adult violent abuses of teens than the NCVS.
Second, even if we accept Dr. Twenge’s measures, she uses the wrong time period (teens’ depression actually rose from 2010 to 2021 and fell in 2022 and 2023 (it’s also unclear why she graphs the irrelevant 1993-2009 period). She then aggregates the NCVS data into unwieldly 5-year periods that mask the true trend.
Substituting the standard method of determining a trend – in this case, plotting the annual NCVS rates of violent family abuses of 12-17 year-olds and calculating a regression trendline that incorporates all the data – shows that violence by family members victimizing teens rose by a substantial 44% over the 2010-21 period, exactly the time teens reported becoming more depressed.
Finally, even though the 32% of teens (35% of girls, 29% of boys) in the CDC’s definitive 2023 survey who report histories of violent victimization ("hit, beat, kicked, or hurt you physically") by parents and household adults are many times more likely also to report being depressed, sad, suicidal, and self-harming, I don’t argue that family violence is the main factor driving teens’ poor mental health. Parental and adult violence victimizing teens is too closely intertwined with the far larger factor of parental/adult emotional abuse to show up separately.
We already know from extensive analyses of dozens of studies by all sides that while social media use creates problems for a small fraction of teens and adults, social media is a negligible factor in teens’ poor mental health overall. I don’t understand on what grounds Dr. Twenge keeps dismissing parental/adult emotional and violent abuse while insisting the problem is just teens using screen media.
Dr. Twenge is not reporting anything new. I’ve been pointing out for years that violence by and victimizing teenagers has been plummeting for 25 years. (The latest is my review last week of “Social Studies,” which cites both law enforcement and NCVS data.)
The CDC’s 2023 YRB survey (https://www.cdc.gov/yrbs/data/index.html) of 20,000 teenagers is the only data set available that allows comparison of the effects on teens’ mental health of social media use versus their exposure to parents’ and guardians’ emotional abuse, violent abuse, other household violence, drug/alcohol abuse, severe mental health problems, suicidality, and incarceration.
The CDC’s detailed analyses (https://www.cdc.gov/mmwr/volumes/73/su/su7304a5.htm?s_cid=su7304a5_w) of the survey finds that teens’ exposure to parental troubles while growing up are “common, with approximately three in four students (76.1%) experiencing one or more” of these parent-inflicted problems “and approximately one in five students (18.5%) experiencing four or more.”
The most common parent- and adult-inflicted issues teens experienced are “emotional abuse (61.5%), physical abuse (31.8%)” and “poor mental health (28.4%),” the CDC reports. The associations between having abusive and troubled parental “were highest for [teens’] suicide attempts (89.4%), seriously considering attempting suicide (85.4%)” and “persistent feelings of sadness or hopelessness (65.6%).” Girls and LGBTQ teens were the most severely abused and also the most depressed, the CDC noted.
The effects of parental/adult abuses and serious problems on teens’ mental health are so large they leave little room for anything else. However, the CDC did analyze the effects of teens’ “frequent social media use." The CDC analysis associated use of social media several times a day with minuscule effects on teens’ mental health (PR=1.33, d=0.16, R-squared effect size < 1%) and NO effects on suicide attempts.
The CDC’s findings that social media use has negligible effects on teens’ mental health square with multiple meta-analyses of dozens of studies by Ferguson et al, Haidt, Rausch, Stein, Jané, etc., that all agree (amid trivial disputes over method) that social media’s association with teens’ depression is at most 1% to 2% -- that is, nothing. Further, CDC figures show depressed teens in abusive and troubled families use social media more than teens in healthy families, contributing to the false conclusion that social media is the cause of depression.
The near-nothing effects of social media are far too small to explain the increase in teens’ self-reported depression and sadness from 2010 to 2021, and don’t correlate with the rise in teens’ suicide rates from 2010 to 2017 and decline through 2023. Twenge and I also agree that violent victimization of teens is not a likely explanation, nor is parents’ jailings, since these have declined.
So, what does underlie the increase? The connections between teens’ poor mental health and parents’ emotional abuses (d=0.89) and parents' poor mental health (d=0.73) and drug/alcohol abuse (d=0.51) are strong. CDC figures for 2023 show 3 in 10 teens had experienced parents/grownups with drug/alcohol abuse problems, one-third with severe mental health problems, and 35% suffered violent abuses and 62% emotional abuses caused by parents or other household adults.
Whether adults’ emotional abuses victimizing teens have risen is not determinable. However, parents’ poor mental health may well have risen (surveys are contradictory), and drug/alcohol abuse definitely skyrocketed.
In the 25-64 age group, which includes the vast majority of parents, parents’ partners, relatives, teachers, coaches, and other grownups influencing teenagers, drug/alcohol overdose deaths (https://wonder.cdc.gov/mcd.html) surged from 55,000 in 2010 to137,000 in 2021, overdose and self-harm emergencies rose from 670,000 to 1.2 million, and total drug/alcohol hospital ER cases (https://www.samhsa.gov/data/sites/default/files/reports/rpt53161/dawn-national-estimates-2023.pdf) exploded from 2.8 million to 5.1 million.
That is, adults’ drug/alcohol abuse soared during exactly the period that 20% more teens reported becoming depressed, and teens 1.8 times more likely to be depressed in households with drug/alcohol abusing parents and adults (CDC). Yet, Twenge and others insist that couldn’t possibly have anything to do with teens’ depression.
Unfortunately, the universal failure until very recently to survey and analyze the effects of parental abuses and troubles on the levels of and increase in teens’ poor mental health has allowed poorly-grounded speculations to dominate discussion even as the scientific case for blaming social media use for teens’ poor mental health and suicides has collapsed. Standard regression analysis of the CDC's 2023 survey shows social media use is associated with less than 1% of teens’ depression. Parental/adult abuses and troubles accompany perhaps 25%. Other factors led by school problems, lack of sleep, and cyberbullying (all heavily tied to troubled parents; for example, 85% of cyberbullied teens also are emotionally abused by parents/adults) together add another 5% or so.
So, what explains the other two-thirds? No one knows. The CDC still has not surveyed the effects of concern over social issues like climate change, family breakup, divorce, etc., on teens’ mental health.
While I appreciate Twenge’s at least giving attention to this issue, she is so eager to excuse away the slightest hint that parents and adult troubles might be driving teen mental health problems that she renders her own analysis easily refutable. It’s long past time for social-media-blamers to back off their inflammatory claims, legislation, and bans and confront real, though uncomfortable, correlates of teens’ issues.
I’m of the opinion that teen depression and loneliness starts at home - the parent or parents are distressed / depressed by the political climate of the past 8-10 years and that is negatively affecting the children. Of course, there are other factors, but this one is rarely if ever mentioned.
Interesting exploration. I'm most interested in what's behind abuse, sexualization, technology impacts, or any of the other factors affecting biopsychosocial development. Yes, individual factors matter. But, I want to see us go deeper - into the landscape those factors exist within. I have seen a serious decline in people's ability to identify and process emotions, communicate effectively, and connect with each other - all while the world, and life, get more complicated. Those factors, IMO, arise from deficits in our inner landscapes.
What if every factor we can identify is simply riding a current of underdeveloped skills across the lifespan? What if the "real" issue is that parents didn't learn how to handle their emotions or meet their own needs, so they can't teach their kids how to talk about feelings, actively explore their identities, or relate skillfully to others? No one in our culture is actively teaching life skills - to parents or kids.
I suggest that we start exploring the data on contributing factors to identify the skills that would alter those factors and start TRAINING those skills en masse. I envision an organization dedicated to human flourishing; one that applies developmental science through activities, classes, and events that train life skills and influence outcomes. I'd start it myself if I had the resources. So many of us are getting further away from being "well" - it seems like it's time to start upping our skills in parenting, communicating, connecting, and listening.
What if we stopped assuming that mental health declines are related to major factors and started working with the landscape they live in - applying findings from developmental science to actively improve individual and social functioning? I dream...
I don't know when, exactly, it started but there has been a dramatic increase in the sexualization of children that has accelerated in the last couple of decades. Children are now being taught they have a right to experience sexual pleasure whenever and with whomever they want and that they have a right to consent, whatever their age, despite laws to the contrary. Simultaneously, they are being taught, beginning at 3 and 4, the doctor only guessed that they were a boy or a girl based on their physical appearance. They won't really know until they discover their sex sometime as they grow older and begin experimenting. Uncertainty, even about something as fundamental as your sex, is destructive to healthy childhood development.
And yet, rape and sexual violence by and victimizing teens has plummeted over the last 30 years (FBI, NCVS); pregnancies and births by teen girls (especially younger girls, and especially those caused by older men) have plummeted by 80%; and abortions have also fallen sharply among teens -- all to the lowest levels ever recorded. Teens' sexual activity has also fallen, as have their drug/alcohol use and crime in general. We've been hearing that teens are more sexualized at ever-younger ages for at least a century, and perhaps longer. Whatever the case, Gen Z suffers far fewer problems in that area than documented in any previous generation.
By the way, Dr. Twenge's NCVS trends are entirely dependent on the way she groups years. If we look at annual NCVS numbers the way she reports, say, annual teen suicide numbers (which are vastly smaller), the opposite conclusion can be shown for the rate of violent victimizations of teens age 12-17 by relatives:
1.5 in 2011 per 1,000 population age 12-17
1.8 in 2018
3.1 in 2021
3.3 in 2022.
That is, family abuses of teens skyrocketed during the 2011-22 period, exactly the time teens' depression also rose. Subjective presentations of tiny numbers can be shuffled around to show any result desired.
I’m sorry to post three times on Dr. Twenge’s substack challenging my article, but I examine critiques of my work carefully. I want my errors pointed out so I can fix them.
In this case, however, further examination shows the entire premise of Dr. Twenge’s argument is wrong. Her and her colleagues’ such as Jonathan Haidt, and Zach Rausch’s baseless dismissals of parents’ and grownups’ widespread violent and emotional abuses victimizing teenagers, as well as parents’ and caretakers’ depression, addiction, and criminal behaviors, is endangering young people and needs to stop.
First, the NCVS she cites is a useless measure of family violence against teenagers, one I will no longer cite. The reason: five-sixths of the interviews of teens by the NCVS are conducted in the presence of others (primarily parents). Unsurprisingly, these chaperoned teens are vastly less likely to reveal family violence than the one-sixth of teens interviewed in private (see research institute analysis, https://ww2.amstat.org/meetings/proceedings/2015/data/assets/pdf/233894.pdf ). This is why anonymously-conducted surveys like the CDC’s uncover far higher parental and adult violent abuses of teens than the NCVS.
Second, even if we accept Dr. Twenge’s measures, she uses the wrong time period (teens’ depression actually rose from 2010 to 2021 and fell in 2022 and 2023 (it’s also unclear why she graphs the irrelevant 1993-2009 period). She then aggregates the NCVS data into unwieldly 5-year periods that mask the true trend.
Substituting the standard method of determining a trend – in this case, plotting the annual NCVS rates of violent family abuses of 12-17 year-olds and calculating a regression trendline that incorporates all the data – shows that violence by family members victimizing teens rose by a substantial 44% over the 2010-21 period, exactly the time teens reported becoming more depressed.
Finally, even though the 32% of teens (35% of girls, 29% of boys) in the CDC’s definitive 2023 survey who report histories of violent victimization ("hit, beat, kicked, or hurt you physically") by parents and household adults are many times more likely also to report being depressed, sad, suicidal, and self-harming, I don’t argue that family violence is the main factor driving teens’ poor mental health. Parental and adult violence victimizing teens is too closely intertwined with the far larger factor of parental/adult emotional abuse to show up separately.
We already know from extensive analyses of dozens of studies by all sides that while social media use creates problems for a small fraction of teens and adults, social media is a negligible factor in teens’ poor mental health overall. I don’t understand on what grounds Dr. Twenge keeps dismissing parental/adult emotional and violent abuse while insisting the problem is just teens using screen media.
Dr. Twenge is not reporting anything new. I’ve been pointing out for years that violence by and victimizing teenagers has been plummeting for 25 years. (The latest is my review last week of “Social Studies,” which cites both law enforcement and NCVS data.)
The CDC’s 2023 YRB survey (https://www.cdc.gov/yrbs/data/index.html) of 20,000 teenagers is the only data set available that allows comparison of the effects on teens’ mental health of social media use versus their exposure to parents’ and guardians’ emotional abuse, violent abuse, other household violence, drug/alcohol abuse, severe mental health problems, suicidality, and incarceration.
The CDC’s detailed analyses (https://www.cdc.gov/mmwr/volumes/73/su/su7304a5.htm?s_cid=su7304a5_w) of the survey finds that teens’ exposure to parental troubles while growing up are “common, with approximately three in four students (76.1%) experiencing one or more” of these parent-inflicted problems “and approximately one in five students (18.5%) experiencing four or more.”
The most common parent- and adult-inflicted issues teens experienced are “emotional abuse (61.5%), physical abuse (31.8%)” and “poor mental health (28.4%),” the CDC reports. The associations between having abusive and troubled parental “were highest for [teens’] suicide attempts (89.4%), seriously considering attempting suicide (85.4%)” and “persistent feelings of sadness or hopelessness (65.6%).” Girls and LGBTQ teens were the most severely abused and also the most depressed, the CDC noted.
The effects of parental/adult abuses and serious problems on teens’ mental health are so large they leave little room for anything else. However, the CDC did analyze the effects of teens’ “frequent social media use." The CDC analysis associated use of social media several times a day with minuscule effects on teens’ mental health (PR=1.33, d=0.16, R-squared effect size < 1%) and NO effects on suicide attempts.
The CDC’s findings that social media use has negligible effects on teens’ mental health square with multiple meta-analyses of dozens of studies by Ferguson et al, Haidt, Rausch, Stein, Jané, etc., that all agree (amid trivial disputes over method) that social media’s association with teens’ depression is at most 1% to 2% -- that is, nothing. Further, CDC figures show depressed teens in abusive and troubled families use social media more than teens in healthy families, contributing to the false conclusion that social media is the cause of depression.
The near-nothing effects of social media are far too small to explain the increase in teens’ self-reported depression and sadness from 2010 to 2021, and don’t correlate with the rise in teens’ suicide rates from 2010 to 2017 and decline through 2023. Twenge and I also agree that violent victimization of teens is not a likely explanation, nor is parents’ jailings, since these have declined.
So, what does underlie the increase? The connections between teens’ poor mental health and parents’ emotional abuses (d=0.89) and parents' poor mental health (d=0.73) and drug/alcohol abuse (d=0.51) are strong. CDC figures for 2023 show 3 in 10 teens had experienced parents/grownups with drug/alcohol abuse problems, one-third with severe mental health problems, and 35% suffered violent abuses and 62% emotional abuses caused by parents or other household adults.
Whether adults’ emotional abuses victimizing teens have risen is not determinable. However, parents’ poor mental health may well have risen (surveys are contradictory), and drug/alcohol abuse definitely skyrocketed.
In the 25-64 age group, which includes the vast majority of parents, parents’ partners, relatives, teachers, coaches, and other grownups influencing teenagers, drug/alcohol overdose deaths (https://wonder.cdc.gov/mcd.html) surged from 55,000 in 2010 to137,000 in 2021, overdose and self-harm emergencies rose from 670,000 to 1.2 million, and total drug/alcohol hospital ER cases (https://www.samhsa.gov/data/sites/default/files/reports/rpt53161/dawn-national-estimates-2023.pdf) exploded from 2.8 million to 5.1 million.
That is, adults’ drug/alcohol abuse soared during exactly the period that 20% more teens reported becoming depressed, and teens 1.8 times more likely to be depressed in households with drug/alcohol abusing parents and adults (CDC). Yet, Twenge and others insist that couldn’t possibly have anything to do with teens’ depression.
Unfortunately, the universal failure until very recently to survey and analyze the effects of parental abuses and troubles on the levels of and increase in teens’ poor mental health has allowed poorly-grounded speculations to dominate discussion even as the scientific case for blaming social media use for teens’ poor mental health and suicides has collapsed. Standard regression analysis of the CDC's 2023 survey shows social media use is associated with less than 1% of teens’ depression. Parental/adult abuses and troubles accompany perhaps 25%. Other factors led by school problems, lack of sleep, and cyberbullying (all heavily tied to troubled parents; for example, 85% of cyberbullied teens also are emotionally abused by parents/adults) together add another 5% or so.
So, what explains the other two-thirds? No one knows. The CDC still has not surveyed the effects of concern over social issues like climate change, family breakup, divorce, etc., on teens’ mental health.
While I appreciate Twenge’s at least giving attention to this issue, she is so eager to excuse away the slightest hint that parents and adult troubles might be driving teen mental health problems that she renders her own analysis easily refutable. It’s long past time for social-media-blamers to back off their inflammatory claims, legislation, and bans and confront real, though uncomfortable, correlates of teens’ issues.
Jean, have you looked into the relationship between educational technology and mental health? I feel as if this is often over looked and there seem to be some trends that line up. I recently wrote about my thoughts here: https://restorechildhood.substack.com/p/the-edtech-paradox-screens-in-schools
I’m of the opinion that teen depression and loneliness starts at home - the parent or parents are distressed / depressed by the political climate of the past 8-10 years and that is negatively affecting the children. Of course, there are other factors, but this one is rarely if ever mentioned.
Interesting exploration. I'm most interested in what's behind abuse, sexualization, technology impacts, or any of the other factors affecting biopsychosocial development. Yes, individual factors matter. But, I want to see us go deeper - into the landscape those factors exist within. I have seen a serious decline in people's ability to identify and process emotions, communicate effectively, and connect with each other - all while the world, and life, get more complicated. Those factors, IMO, arise from deficits in our inner landscapes.
What if every factor we can identify is simply riding a current of underdeveloped skills across the lifespan? What if the "real" issue is that parents didn't learn how to handle their emotions or meet their own needs, so they can't teach their kids how to talk about feelings, actively explore their identities, or relate skillfully to others? No one in our culture is actively teaching life skills - to parents or kids.
I suggest that we start exploring the data on contributing factors to identify the skills that would alter those factors and start TRAINING those skills en masse. I envision an organization dedicated to human flourishing; one that applies developmental science through activities, classes, and events that train life skills and influence outcomes. I'd start it myself if I had the resources. So many of us are getting further away from being "well" - it seems like it's time to start upping our skills in parenting, communicating, connecting, and listening.
What if we stopped assuming that mental health declines are related to major factors and started working with the landscape they live in - applying findings from developmental science to actively improve individual and social functioning? I dream...
I don't know when, exactly, it started but there has been a dramatic increase in the sexualization of children that has accelerated in the last couple of decades. Children are now being taught they have a right to experience sexual pleasure whenever and with whomever they want and that they have a right to consent, whatever their age, despite laws to the contrary. Simultaneously, they are being taught, beginning at 3 and 4, the doctor only guessed that they were a boy or a girl based on their physical appearance. They won't really know until they discover their sex sometime as they grow older and begin experimenting. Uncertainty, even about something as fundamental as your sex, is destructive to healthy childhood development.
And yet, rape and sexual violence by and victimizing teens has plummeted over the last 30 years (FBI, NCVS); pregnancies and births by teen girls (especially younger girls, and especially those caused by older men) have plummeted by 80%; and abortions have also fallen sharply among teens -- all to the lowest levels ever recorded. Teens' sexual activity has also fallen, as have their drug/alcohol use and crime in general. We've been hearing that teens are more sexualized at ever-younger ages for at least a century, and perhaps longer. Whatever the case, Gen Z suffers far fewer problems in that area than documented in any previous generation.