Why parents aren't to blame for the rise in teen depression

Gen Z teens actually have a better relationship with their parents than previous generations did

A few weeks ago, I posted a rundown of 13 alternative explanations (cross-posted here with a better headline) for why teen depression has increased, showing that none of them fit as well as the upswing in smartphone and social media use.

Since then, I’ve seen another explanation proposed: Abusive and troubled parents. The argument is two-pronged: 1) Parents have become more abusive to teens since 2010 and 2) more parents die from suicide and drug overdoses (“deaths of despair.”) Thus, this argument goes, more teens today are depressed not because they spend more time on social media and less time with friends, but because their relationships with their parents are significantly worse.

In this post, I’ll examine the evidence for this explanation and then present some original data analysis on trends in teens’ relationships with their parents and how the increase in teen depression varies by socioeconomic status, a factor strongly related to drug overdose deaths.

The parental abuse evidence

Sociologist Mike Males recently argued on this platform that “Teens’ depression and anxiety are normal reactions to more troubled, abusive adults.” He followed up this week with “How long will authorities keep ignoring the devastating effects of parents’ soaring addictions and abuse on ‘teenage mental health’?”

Let’s examine abuse first. The reports I could find instead show that rates of child abuse are down in recent years and changed little 1998-2016. If child abuse had increased enough to double teen depression, you’d think there would be some evidence for it in the records we have. Still, most of these child abuse cases involved younger children, and many cases of child abuse are not reported, so perhaps these aren’t the best measures.

Dr. Males presents the figure below to make the case that parental abuse of teens has gotten significantly worse. These are pretty stunning statistics at first glance, so let’s dig into them.

The first thing to notice is that the 2022 data (actually 2021) comes from a different source than the 2008 and 2010 data. If the two datasets had differences in administration or sampling, that could be causing any differences rather than a real change in abuse rates.

That is indeed the case. The 2008/2010 data relies on a phone survey that begins with parents. More than likely, a greater number of abusive parents (vs. non-abusive) did not pick up the phone or allow their children to be interviewed, lowering the rates. In contrast, the 2021 data collection was online and coordinated by teachers at schools, and was thus likely to include more abused teens. The method of data collection (phone vs. online) may also have had an impact — it’s easier to admit abuse in an online survey than to a real person on the phone. These confounds, rather than any true change over time, could easily account for the different rates.

Another crucial issue: Did the two surveys measure abuse in the same way? If not, that also throws doubt on the idea that there has been a true increase.

Sure enough, the questions used in the two surveys are different. The 2008 and 2010 data come from the National Survey of Children’s Exposure to Violence. Here is its question about emotional abuse:

The 2021 data was part of the Adolescent Behaviors and Experiences Survey administered by the CDC. Here is its question about emotional abuse:

These are very different questions. The 2021 question asks about less severe and more ambiguous adult behaviors than the 2008/2010 question. An adult telling a teen they need to do their homework so they don’t fail their classes could be considered a put-down or an insult. So could a parent saying they were disappointed with a teen’s behavior. “Swear at you” could also include parents using swear words at all. Apart from any change over time, it seems likely more teens will agree with this question than the previous one.

In addition, the response choices are also different between the surveys – “yes” vs. “no” for the 2008/2010 question and “never,” “rarely,” “sometimes,” “most of the time,” and “always” for the 2021 question. With “rarely” still counting as “yes” in the graph, that could drive the number upward, too.

Thus, it seems very likely that the huge difference in emotional abuse between 2010 and 2021 is an artifact of the different surveys asking different questions, and not an actual increase over time.

There’s a smaller though still significant increase in reports of physical abuse between 2010 and 2021. Again, though, these are different surveys that were administered and sampled differently. The questions are also different. Here is the 2010 question:

And here is the 2021 question:

These questions are more similar than the two emotional abuse questions, but there are still key differences. The 2010 question specifically excludes spanking, while the 2021 question does not. There are also different response choices, with the 2010 survey using “yes” and “no” and the 2021 survey using “never,” “rarely,” “sometimes,” “most of the time,” and “always.”

Without consistent measurement, it’s not possible to use these surveys to compare trends in emotional or physical abuse over the years. We can’t separate what is due to the differences in the surveys and what is due to time.

Trends in teens’ relationships with their parents: New data analysis

It would be better to find a survey with consistent sampling, administration, and questions. One example is Monitoring the Future (MtF), which has surveyed teens via their schools since 1976 — I relied on it quite a bit in writing Generations. All responses in MtF are anonymous (key since much of the survey asks about drug and alcohol use) and it is administered in person at schools. Although MtF does not have questions that specifically ask about parental abuse, it does have several questions very relevant to the parent-teen relationship. For example, teens are asked how often they fight or argue with their parents. That should certainly have some overlap with abuse.

What does that show? The opposite: Teens in recent years are actually less likely to report frequently fighting with their parents compared with teens in previous eras (see Figure 1).

Figure 1: Percent of U.S. 12th graders who argued or had a fight with their parents 5 or more times in the last year, 1976-2022. Source: Monitoring the Future

Maybe teens with abusive parents have given up and don’t even try to argue. Even if so, abuse would still be reflected in questions about satisfaction with the parent-teen relationship. If abuse has increased in recent years, more teens should be dissatisfied with how they get along with their parents.

Instead, teens’ satisfaction with their parents has increased slightly since 2012 (see Figure 2). That’s even more noteworthy because teens’ satisfaction with their lives as a whole has declined significantly over the same time. So teens have become more satisfied with their parents even as the rest of their lives became less satisfactory.

Figure 2: U.S. 12th graders’ satisfaction with “the way you get along with your parents” and with “your life as a whole these days,” 1976-2022. Source: Monitoring the Future.

All of this argues against the idea that parental abuse is behind the increase in teen depression.

The “troubled parents” and drug overdose evidence

Now let’s turn to the question of troubled parents. Dr. Males includes this graph:

These are striking trends, but note that this says “parent-age,” not “parents.” From this data, we have no way of knowing how many of these adults actually have teens at home, and if their trends are the same or different. Perhaps the trend is driven primarily – or even exclusively – by those without teens at home.

We’ve also seen this before: This is the sharp increase in “deaths of despair” that Case and Deaton (2015) documented. They found that deaths of despair increased much more among those without a college education compared to those with a 4-year college degree. In recent data, the difference is stark: From 2018 to 2021, drug overdose deaths increased 32.1 per 100,000 for those with high school diplomas but only 1.5 deaths per 100,000 for those with a four-year college degree. Overall drug overdose rates in 2021 were 82.6 per 100,000 among those with a high school diploma and 9.0 for those with a four-year college degree – 9 times less.

Thus, if increasing levels of drug addiction among parents is a cause of the increase in teen depression, the rise in depression should be much more pronounced among teens whose parents do not have a four-year college degree (lower SES). The rise in depression will be less – and possibly non-existent – among those whose parents have four-year college degrees (higher SES), given that opioid overdose deaths are 9x less common in this group, and increased 16x less.

That’s not what happens. Instead, the increase in depression is virtually identical among those whose mothers have a college degree vs. those whose mothers do not (see Figure 3). The lower SES teens do have higher levels of depression overall – but they always have, and the gap between the two does not grow larger during the opioid overdose era after 2010.

Figure 3: Percent of U.S. 12th graders with high levels of depressive symptoms, by mother’s level of education, 1976-2021. Source: Monitoring the Future.

Final thoughts

It is difficult to conclude from this data that the rise in teen depression is due to issues with their parents. The data Dr. Males presents on parental abuse can’t be compared over time due to significant differences in sampling, survey administration, and measurement. Instead, in a survey asking the same questions using the same methods over decades, teens are more satisfied with their parents and fight with them less – trends inconsistent with an increase in parental abuse.

Suicide and drug overdose deaths have soared among middle-aged people, primarily among those without a college degree. Yet the increase in teen depression is just as large among those whose parents have a college degree, a group much less impacted by the recent increases in drug overdoses.

Of course, we already know what changed in the lives of both groups of teens: They started spending more time on social media and smartphones, less time with friends in person, and less time sleeping. The way they spent their time outside of school completely changed. That is a much more likely explanation for the increase in teen depression in a generation that actually gets along with their parents better than previous generations ever did.

What about the idea that parents are responsible for the rise in teen depression because they failed to keep their kids off social media? Considering social media apps do not require parental permission to open an account and do not reliably verify age, that seems like an unfair accusation. Social media companies have poured billions into making their products as attractive as possible, especially for kids and teens, and it’s tough for parents to fight this.

That said, there are a few things parents can do, like putting off giving your kids a smartphone (a phone with internet access) as long as possible and doing what you can to keep them off social media until they are at least 16. More on that another time — I think a whole post on practical solutions is warranted, and I plan to do that soon. For now, I’d love to hear your thoughts on whether we can blame parents for the rise in teen depression.