Every generation of parents faces its own new challenges. For the current crop of moms and dads — and in the midst of a pandemic, when screen time has in many cases become a necessity, rather than a choice — questions surrounding technology and adolescent mental health are among the most pressing and bewildering.
Are all screens created equal? And how much screen time is too much? Is social media uniquely dangerous? Different studies seem to yield different answers. While one group of experts issues warnings, another argues that concerns are overblown. Competing op-eds alarm or pacify.
This expert debate is set against a backdrop of what most consider to be historically steep increases in adolescent depression, anxiety, and suicide, even before the pandemic. “It’s unequivocally the case that we have one of the worst mental health situations among our youth that has been documented,” says Mary Helen Immordino-Yang, EdD, a professor of psychology and education at the University of Southern California.
That situation is showing no signs of improving. According to a 2019 report from the U.S. Department of Health and Human Services (HHS), suicide rates among Americans aged 10 to 24 increased by 56% between 2007 and 2017. Some of the most pronounced increases occurred among the very young; suicide rates nearly tripled during that time period in kids ages 10 to 14.
Trends in youth depression and anxiety are on similar trajectories, and young women seem to be experiencing these challenges more than young men. According to federal data collected from across the country, more than one in five adolescent girls experienced a major depressive episode at some point during 2018, the most recent year for which data are available. That represents an 84% increase during the past decade. Rates of depression among young women have risen steadily since 2009, which suggests the cause or causes of their struggles are the result of persistent phenomena.
“Is it coincidence that all this happened at the same time as the social media and smartphone revolutions? Could be,” Immordino-Yang says. “But even if it is, we know those two things are happening in the same people.” She’s referring to the evidence — which at this point is ample — linking some aspects of heavy screen use to an elevated risk for mental health problems among youths. One of many recent examples: a 2019 study in JAMA Pediatrics found that an adolescent’s risk for depression symptoms rises with each daily hour spent on social media.
But correlation is not causation, as every scientist knows. And some experts have posited that young people with underlying challenges may just be more likely to spend large amounts of time on tech — meaning heavy use is a byproduct of an underlying issue, not a cause or contributing factor.
“A large observational study from Canada just found that early mental health problems for girls predicts later social media usage,” says Candice Odgers, PhD, a professor of psychological science at the University of California, Irvine.
Recently, some of the loudest voices on either side of the screen-time debate have started challenging one another’s findings.
She’s referring to a 2019 study in Clinical Psychological Science. That same study also failed to find a connection between youth social media use and the later emergence of depression. “Even if there is an association [between kids and social media], I think it’s more likely that the arrow goes in the other direction — so kids struggling have different viewing patterns,” Odgers says. Some of her own recent work has concluded that the evidence linking daily digital technology use to adolescent mental-health struggles is mostly weak and conflicting.
Anyone who is raising, teaching, counseling, or coaching kids would breathe a deep sigh of relief if it turned out that popular new forms of digital technology are harmless. That’s especially true right now, at a time when schools have been forced online and parents are desperate to keep kids occupied while they complete work or chores. But concerns surrounding kids and screens extend beyond the associations with mental illness; questions abound about the cognitive and behavioral effects of media multitasking, online pornography, social comparison, and other elements of tech use. Some experts who have studied the effects of screens on young people say that a Pollyanna-ish view is not supported by the evidence.
Recently, some of the loudest voices on either side of the screen-time debate have started challenging one another’s findings. A close look at one of these exchanges suggests that expert agreement may not come anytime soon, but that parent concerns about digital media are justified — if not yet proved.
Last year, in the journal Nature Human Behaviour, two psychologists at the University of Oxford published an influential study that found a “small” association between digital technology use and lower well-being among adolescents.
The study relied on large-scale survey data collected from kids and parents in the U.S. and U.K. In an effort to contextualize technology’s real-world effects among young people, the study looked at how some other common activities included in the surveys — such as listening to music or eating potatoes — stacked up to tech. “The association of well-being with regularly eating potatoes was nearly as negative as the association with technology use,” the authors of that study wrote. At the time of publication, these conclusions — and especially the potato comparison — garnered a lot of media attention. The take-home message was unambiguous: despite widespread concern among parents, there’s little reason to fret about kids and screens.
In the context of Covid-19 and nationwide stay-at-home orders, this message is surely music to the ears of cooped-up parents. And in an April New York Times op-ed titled “Don’t Freak Out About Quarantine Screen Time,” Andrew Przybylski, PhD, co-author of the above study and director of research at the Oxford Internet Institute, made this point more explicitly. “The evidence linking screens to harm is, in reality, paper thin,” he and a coauthor wrote. His op-ed made no mention of social media — arguably the screen-based activity that parents worry about most, and that has garnered the most scientific scrutiny — but it did highlight the possible benefits of video-game play.
Oxford’s Przybylski is among the most outspoken critics of the “screens are harmful” narrative that has gained traction in recent years. Almost any high-profile piece that has lately challenged this narrative has highlighted his work or his words — or both. The home page of his research project at Oxford not only calls into question the assumption that new digital technologies are imperiling young people, but it also casts doubt on the idea that the well-being of adolescents is in decline.
Przybylski’s views are contested. In a pair of recently published critiques, a group of U.S. researchers assert that some of his work is methodologically flawed and also misleading in the way it presents its findings. In one of those critiques, published April 17 in Nature Human Behaviour, these researchers claim that, in his influential potato-comparison study, Przybylski and his co-author engineered their analysis and summarized their findings in ways that minimize the negative effects of certain forms of digital technology.
“They have taken datasets that show a clear relationship between heavy social media use and poor mental health outcomes for girls, and they made a series of decisions in the analysis that obscure that relationship,” says Jonathan Haidt, PhD, co-author of those critiques and a social psychologist at New York University’s Stern School of Business.
In a published reply to these criticisms, Przybylski and his co-author, the University of Cambridge research fellow Amy Orben, PhD, say that they “reject” these claims. On the question of downplaying the effects of social media, Orben says that, at the time of her study’s writing, which was roughly two-and-a-half years ago, the kids-screens debate was centered around smartphones, not social media, and that is why she and her co-author did not focus on social media specifically. (Asked to respond to criticisms of his study, Przybylski referred to his published reply.)
These sorts of expert back-and-forths are fairly common in scholarly journals. But a close look at the arguments both sides make helps reveal just how tricky it can be to shake out hard-and-fast answers from a big bundle of data.
“Lies, damned lies, and statistics”
Jean Twenge, PhD, is a professor of psychology at San Diego State University. In books and peer-reviewed papers, Twenge has made the case that smartphones, social media, and other aspects of today’s popular screen-based technologies are contributing to the widely documented increases in youth depression, anxiety, and suicide. “I think there is an increasing amount of evidence that the ways young people are using technology leads to mental health issues and poor psychological well-being,” she says.
In many ways, Twenge is Przybylski’s antipode in the kids-screens debate. She is also, along with Haidt, one of the authors of the two recent critiques of Przybylski’s work. “Everyone can start with the same numbers, but the way you analyze those can lead to very different conclusions,” she says. She argues that some of Przybylski’s own study data show negative relationships between kids and aspects of tech use, but that his work tends to parse and present these data in ways that obscure any harms.
Przybylski directs some of the same criticisms at Twenge’s work. “There are three kinds of lies,” he says, quoting a line often attributed to the former U.K prime minister Benjamin Disraeli. “Lies, damned lies, and statistics.” He says this in reference to a paper of Twenge’s, published in 2018, that found that a young person’s psychological well-being tends to decline as hours of daily screen time increase. He says that she and her co-authors “cherry picked” outcome measures in an effort to make their results appear significant.
In the introduction to his potato-comparison study, he and his coauthor Orben make a similar argument — namely, that much of the research that has linked screen-based activities to declines in youth mental health are influenced by researcher bias and data manipulation. In what they describe as an attempt to eliminate this sort of jiggering, they employ a complex method of statistical modeling called specification curve analysis (SCA).
In a nutshell, SCA runs a large number of analyses based on a range of variables and outcomes and then comes up with cumulative effects based on the pooled results. When done properly, the theory is that SCA eliminates the temptation for researchers to “cherry pick” variables and analyses that produce the largest effect sizes. But others say that SCA itself leaves plenty of room for researchers to steer the data in one direction or another.
“Their paper introduced this statistical technique, SCA, which looks very impressive because it reports tens of thousands of analyses, and some people writing about it seemed to treat it as the decisive paper,” says NYU’s Haidt. “But when we read the paper, a few things jumped out at us right away.”
First and foremost, he and Twenge say that, although the focus of their paper is “digital technology use,” Orben and Przybylski chose to pool together all forms of screen-based media, including television. Haidt says this “greatly dilutes the effect sizes” of social media and the other forms of digital tech that most concern today’s parents. “They merge all forms of device use together, even though most kinds of device use are harmless, and they merge boys and girls together, even though the correlations are consistently larger for girls,” he says. “They do this in several other papers as well.”
On a more technical note, he says they chose to control for certain variables — meaning they adjusted their analyses to account for differences between groups — in ways that are “hard to defend.” An example: He says that Orben and Przybylski controlled for young people’s reported happiness at school. “That’s obviously related to the outcome variable — poor mental health,” he says. “If you throw it in as a control . . . then of course it reduces the association between social media use and poor mental health.”
Other researchers not affiliated with either group say some of these criticisms hold water. “I really do question the results of the study because I don’t think that the methodology is appropriate,” says Brian Primack, MD, PhD, dean of the College of Education and Health Professions at the University of Arkansas, who in the past has studied kids and new digital technologies. Rather than clarify the picture, he says that in this case throwing in too many variables and combining male-female data obscures the truths that a simpler, more-focused analysis could reveal.
David Funder, PhD, a distinguished professor of psychology at the University of California, Riverside, has written about the challenges and pitfalls of presenting effect sizes in psychological research. He also says that he sees some problems with Orben and Przybylski’s paper. His main criticism is that the way their study chooses to present effect sizes — as “r-squared” — is misleading. “It adds no information at all. It’s just presenting an effect in a way that always makes the effect look small,” he explains. “Their bottom-line is that there is a negative effect, but that it’s not big enough to worry about, but these quote-unquote small effect sizes can have large effects in the long run, or when applied to large numbers of people.” He also reinforces Haidt’s point about control variables. “They are obscuring what’s going on by making those adjustments, and I don’t know why they’d do that,” he says. “If you control for too many things, you know you’re going to lower the size of the effect.”
In reply, Orben says that she and Przybylski re-ran some of their analyses as Twenge and Haidt’s critique requested, and the results “clearly don’t change.”
UC-Irvine’s Odgers, who reviewed Orben and Przybylski’s study before its original publication, says she considers it one of the better analyses to date on young people and screens. “This team does some of the most careful work in the field — they follow the best practices in registering their hypotheses ahead of time and they make their data and analysis code openly available for others to check,” she says. “If other researchers worked in this type of open and transparent way, we would have less misinformation and unstable findings causing confusion about this topic.”
Odgers says that there is “a small association between depression symptoms and social media use, and you see it pop up more consistently for girls.” But she says the biggest predictors of depression among young people remain a family history of mental disorder, socioeconomic hardship, and other factors that have nothing to do with technology. “Can social media play a role [in adolescent depression]? Probably. But I think the rise in depression is more a reflection of things happening in society than social media driving it,” she says.
NYU’s Haidt does not agree. “There is a giant wave of depression and anxiety that swept through Gen Z, especially the girls, beginning around 2012, which is right after American teens started using social media platforms every day,” he says. He also points to the research, including a 2018 study from the University of Pennsylvania, that has found young people who cut back on social media experience reductions in depression. “Until someone else can explain what happened to Gen Z around 2012, not just in the U.S. but in the U.K. and Canada too, I think parents should be wary of researchers who tell them not to worry about their kids spending four hours a day on social media.”
With so much expert disagreement, it’s little wonder that parents — as well as public health officials and policymakers — are confused about how best to handle young people and their interactions with screens.
Since video chat is in real time and you can see the other person’s face, it’s preferable to the more curated and less interactive world of social media.
Consensus, of a kind
While experts are often at loggerheads when it comes to kids and technology, there are places in which their views overlap.
Cambridge’s Orben says that, when it comes to kids and new digital technologies, “it’s complicated, and we should not oversimplify.” At least there, SDSU’s Twenge seems to agree. The question of kids and screens “is much more complicated than saying one thing is completely okay and one thing isn’t,” she says.
While watching TV or playing video games isn’t the best way that kids — and especially young children — can spend huge chunks of their time, Twenge says that most of her concerns center on the risks of smartphones and “heavy” social media use, which she defines as more than two or three hours a day. “I think most people are wondering whether heavy users of these technologies have worse outcomes,” she says. “And the published data show that they do.”
She says that pairing social media with the attention-diverting, habit-forming, always-with-you qualities of a smartphone is an especially fraught combination. She also says that if young people are spending many hours a day online, then they don’t have as much time for sleep, talking face-to-face with friends, or engaging in other activities that support healthy development.
“During times of stay-at-home orders, teens might not have the option of getting together with their friends in person. But they can video chat,” she says. “Since video chat is in real time and you can see the other person’s face, it’s preferable to the more curated and less interactive world of social media.”
Haidt says that, in the midst of Covid-19, parents shouldn’t feel guilty about leaning on screens to help them occupy their kids and pass the time. “I’m very pro-technology, and I recently bought an Xbox for my 13-year-old son so he can play with his friends during lockdown,” he says. “I think the research is becoming increasingly clear that screen time is not the problem. Social media is.” He defines social media as any platforms or apps “where kids create content that is commented on or rated by others,” which tends to fuel social comparison and judgment, and also shaming and bullying.
While he and other experts who study kids and screens debate one another, those who work directly with young people say the impact tech is having is apparent and, at least in some cases, detrimental.
“Smartphones and social media — this is certainly where they live and interact,” says Jill Cook, assistant director of the American School Counselor Association. “One of the problems is that these platforms allow young people to engage in activities they probably wouldn’t be able to face-to-face.” As an example, she says some students set up fake profiles and harass one another in ways they likely wouldn’t, or couldn’t, if they lacked the anonymity of a screen. “We’ve seen students bully other students into harming themselves,” she says.
The rise in mental health challenges among teens is something she and other school counselors have noticed and discussed, and she says that tech often appears to be at the center of the issue. “With social media, you’re comparing yourself to whatever that perfect image is that people are putting out there,” she says. “That can be very hard, and we have to help kids navigate that in a healthy way.”
“I think it’s critical for parents to be thinking about social media,” says U. of Arkansas’s Primack. He points out that kids who are disadvantaged or struggling with health issues may find helpful support communities on social media, and so the lesson shouldn’t be that any and all use is harmful. But if kids are spending many hours a day on social media, he says, any benefits may be outweighed by the risks. “I’d suggest that people reflect on their experiences with these platforms and adjust their use accordingly,” he says.
He also says that, at this moment, parents can direct their kids toward “positive screen-related activities,” which include educational and entertaining YouTube channels. He mentions Life Noggin, CrashCourse, and TED Ed as some examples. That doesn’t mean parents should park kids in front of screens all day; when it comes to most forms of digital tech, most experts say moderation is key. Unfortunately, moderation doesn’t seem to be the status quo.
While it’s hard to know exactly how any young person is spending his or her time, a 2018 report from the Pew Research Center found that 45% of teens say they’re online “almost constantly” — up from 24% just a few years earlier — and that social media sites make up nearly all of the online platforms where they spend the most time. Meanwhile, a 2019 report from Common Sense Media found that, among 8-to-12-year-olds, average daily screen time is now up to nearly five hours, which doesn’t include time spent on schoolwork. Among teens, that figure jumps to more than seven hours per day. The Covid-19 pandemic and stay-at-home orders are certain to augment those figures.
“Is tech bad or not? I don’t think there’s any straight answer to that,” says USC’s Immordino-Yang. “That’s like asking, ‘Is food good or bad?’ It depends on what and when and how you’re consuming it.” But just as the data on obesity-related disease indicate that many Americans aren’t eating properly or in moderation, young people may be over-consuming some unhealthy forms of technology. Building screen-free hours and activities into a young person’s day may help reestablish a sense of balance that has been lost.
“It’s really important to understand that a young person’s brain and social and cognitive and psychological development is incredibly plastic and incredibly experience-dependent,” Immordino-Yang says. “That means the way young people use their brains will impact how they grow and develop.” Knowing this, she says it’s unreasonable to think that relocating so much of a young person’s time and social interaction into a digital space would not have an impact on how they think and feel. And at a population level, that relocation seems to be having some negative consequences.
“You can slice and dice the data up in different ways, but let’s get real,” she says. “We need to think about shifting the ways in which kids use these technologies in order to leverage the benefits and attenuate the risks.”