
Social Media Children
Social Media: The Moral Panic
COMMENTARY
Published Aug. 22, 2023 • Read time: 9m
A recent study found that using social media was unrelated to long-term mental health problems in Norwegian youth. One well-intentioned scholar suggested that doomsayers proclaiming that social media is causing suicide and other mental health problems in youth would need to “reckon” with this new study. To which I replied, in effect, they would do no such thing. As happens during technology moral panics, evidence that calls the panic into question would simply be ignored. Indeed, as far as I am aware, this new study has gotten zero news media or political attention.
So it is with moral panics, and there’s little question that we’re in one regarding social media. Advocates for the panic claim that social media has caused a decline in youth mental well-being, beginning somewhere around 2009-2015 (the exact date, of course, being flexible depending upon convenience). Like many such moral panics, this one repeats easily predictable cycles. These include a lack of curiosity regarding evidence calling the panic into question, selective interpretation of societal data, selective news and political attention, scary but vacuous statements from professional guilds such as the American Psychological Association, and a few “big names” who, though well-intentioned, end up carrying the pail for the panic (and holding the bag once it is over). Below, I illustrate the evidence documenting that, once again, we’re blundered into a moral panic cycle.
Youth Studies
One of the features of moral panics is advocates for the panic claim that research consistently supports the panic when in fact it does not. We saw this as a feature of the video game violence moral panic. We can see that narrative once again, regarding social media. For instance, one mental health website claimed, “However, multiple studies have found a strong link between heavy social media and an increased risk for depression, anxiety, loneliness, self-harm, and even suicidal thoughts.” Unfortunately, that’s not at all an accurate portrayal of the evidence. Last year, a large group of media psychologists and I published a meta-analysis of studies of screen use, including social media, and mental health. Overall, we found that the evidence was inconsistent at best, and weak overall. Better studies were less likely to find evidence for correlations. As with the recent Norwegian study, the best studies do not find that social media use leads to later mental health problems for youth.
Part of the miscommunication is a quirk in social science research wherein large samples can produce “statistically significant” results that are in fact no more than statistical noise. This is a widespread problem in psychology that researchers have been slow to grapple with, but it does misinform social media research. For instance, researchers have found that the tiny correlations between social media use and youth mental wellness are no larger than those for eating potatoes or wearing eyeglasses on teen mental health. Yet we don’t run around warning parents of the dangers of letting their kids eat potatoes or wear glasses.
Experimental Studies
If correlational and longitudinal studies have been unable to support beliefs that social media is associated with decreased well-being, what about experimental studies? Some commentators have been enamored of experimental studies linking social media to reduced mental health.
Such studies (mostly done with adults, not teens), are pretty basic. People are randomly assigned to either avoid or reduce using social media for some length of time (several weeks or a month) or are in a control group which is not asked to change their behavior. However, these studies are not convincing for several reasons.
First, like the correlational and longitudinal studies, results have been inconsistent. Second, for some that do find effects, the effects are once again so small that they are within the “noise” or “margin of error” for social science research. Third, all that aside, they suffer from a rather glaring flaw: It’s obvious to the participants how they are expected to respond. Once asked to reduce social media time, most respondents can guess that the researchers expect them to report feeling better after the time has elapsed. Unfortunately, it’s well known that if people know they’re being observed and can guess what they’re expected to do, they’ll usually do it even if it’s not their normal response. This renders the results invalid and means these studies tell us nothing.
Some studies also appear to have non-random drop-out. This means people assigned to reduce social media time may not have enjoyed the experience and might have dropped out of the study more often than did people in the control group. Likely, people experiencing bad outcomes were more likely to drop out. Unfortunately, this means their data is lost and without it, any results are spurious.
Cross-National Comparisons
Much of the argument on social media and mental health has focused on teens in the United States, who have experienced increased suicide rates over the past decade. Unfortunately, this argument ignores data from other countries where the data do not demonstrate a similar pattern, even when those countries adopted widespread social media.
For example, data from European countries suggest that a majority of countries experienced a decrease in teen suicides from 2011-2020. The United Kingdom similarly showed no rise in teen suicides during the social media age. Of course, outcomes vary country by country, but advocates for the social media panic tend to cherry-pick outcomes rather than look at the larger picture.
Adult Suicide
Speaking of the larger picture, part of the problem with our discussions of teen suicide is that they neglect the larger picture of suicide across all age groups. Teen suicides don’t happen in isolation but are part of a larger pattern of suicide increases across all age groups. In fact, suicide has risen in the U.S. almost universally. The two groups of individuals with the highest suicide rates (and greatest recent increases) don’t remotely include teens (who actually still have remarkably low suicide rates). Instead, young American Indian men and middle-aged white men have the highest suicide rates. Middle-aged men commit suicide roughly 3-5 times more often than teen girls, according to Centers for Disease Control data.
RCP
By ignoring this, we’re engaged in a classic “blind men and the elephant” phenomenon. As we obsess over teen girls, we focus on something relatively unique to them (social media) and fail to consider that suicides between older adults and their youth are connected. One of the best predictors of teen suicide is experiencing suicide in the family, such as a parent. In this sense, the social media panic has arguably done outright harm by distracting us from a more holistic look at suicide.
Gender Dysphoria
One interesting hypothesis that has been raised is the argument that the explosion of teens identifying as LGBT can be attributed to social contagion through social media. This hypothesis has most famously been raised by journalist Abigail Shrier in her book “Irreversible Damage.”
In a general sense, I agree the rapid rise in LGBT identification hints at a social cause, whether status-chasing by mainly straight kids, or cross-over between certain mental health conditions such as borderline personality disorder and autism with gender dysphoria. However, it’s less clear that social media (as opposed to more direct encouragement by parents, teachers, and peers) is a primary cause.
I find the hypothesis intriguing but unfortunately, as far as I can see (from a search of research databases), there is no data on this one way or another. I would be excited to see some (particularly using transparent, open-science principles), but at present I think people need to be very cautious advancing this theory without data. On the other hand, I can’t help but wonder if a research community that has been overzealous in connecting social media to mental health generally has considered it taboo to extend that concern to gender dysphoria specifically. Either way, we simply need more data.
Other Explanations
One final argument I sometimes hear is, well if the rise in teen suicides isn’t due to social media, what else can explain it? Of course, the answer is that lots of things have changed over the past decade or so. Income inequality has increased, as has our political partisanship. The K-12 education system has changed with an increased emphasis on divisive, paranoid, and anti-American belief systems such as Critical Race Theory and progressive gender ideology. Fatherlessness has increased (it turns out fathers are actually rather useful for kids).
It's not hard to find other correlations, in fact. For instance, here is the correlation between teen girl suicide and an increased obsession among educators regarding oppression, racism, sexism, homophobia, etc. Suicide data was obtained from the CDC, whereas the educational trends data was obtained by searching an educational research database for the subject terms noted above.
RCP
And here is the correlation between teen girl suicide and political polarization:
RCP
Data on political polarization was provided by the Pew Research Center. Both correlations are quite strong. Of course, all this data is correlational, and we can’t make causal attributions. Nor are these anything like an exhaustive list of potential correlates. Personally, I suspect that the United States is experiencing something I call “the blob” … a wave of negativity (some reasonable, some irrational) affecting all Americans, not just teen girls. But that will require far more rigorous testing than what we’ve allowed to pass for the social media panic.
Concluding Thoughts
One remarkable thing about the social media panic is that so much of it takes place on social media, where advocates of the panic blast out their latest concerns, policy efforts, or book publicity. If these concerns were sincere, it would be rather like anti-heroin advocates meeting in an opium den, and it calls into question how seriously people think social media use is harmful.
The issue of suicide (all suicide) is an important one. Unfortunately, by hyper-focusing on teen girls, we may have misdiagnosed the issue. Regarding social media effects, we need to insist on far more rigorous evidence than what we’ve allowed to pass thus far and stop cherry-picking data. But given the poor showing thus far for this hypothesis, we need to start considering other possible causes as well. To do so, we’ll need to zoom out and consider all suicides among all groups holistically, rather than continuing to stumble about like blind people poking one increasingly frustrated elephant.
RCP Today
Top stories on RealClearPolitics
The Takeaway
Polling intelligence and analysis delivered to your inbox
World News
Featuring the latest world news
Policy
Featuring the latest policy news
Support Our Mission
As part of RealClearPolitics, authors aim to provide balanced, non-partisan analysis by presenting top reporting and commentary from all sides of the political spectrum.
Trusted by policymakers and citizens alike since 2000






