Verified by 
Key points
- Last month’s CDC report shows a rise of mood disorders in teens—particularly in teen girls.
- Research shows correlations between mood disorders and social media use—but with so many variables, it's not clear if there's a causal link.
- A new report published this week by The Female Lead shows that measures used in most research are unreliable.
- Observing teen girls’ own efforts to manage social media harms themselves yields more useful results.
The question of what is driving increasing rates of depression in teens—particularly teen girls—is more urgent today than ever.
As part of the Youth Risk Behavior Survey, the U.S. Centers for Disease Control and Prevention (CDC) published a report in February 2023 showing a rise in teen girls’ mood disorders. Sixty percent of girls reported continuing sadness and hopelessness and nearly a quarter revealed that they had a suicide plan as a means of escape if their feelings overwhelmed them.
As politicians and psychologists ponder the causes behind this alarming trend—one not limited to the U.S. but also apparent in Britain and elsewhere around the world—the potentially corrosive influence of social media on teen girls’ mental health is once again placed in the inquisitorial limelight.
Measuring Mood in Teen Girls is Problematic
Having just written a report for education charity The Female Lead on our research exploring the possibility of improving social media health—i.e. the use of social media in ways that foster rather than undermine girls’ wellbeing—I am struck by how much is missing from mainstream arguments linking social media use to low mood in teens.
First, there is heavy reliance on correlations: the much-cited work of Jean Twenge, for example, showed that by 2014 over 80 percent of teen girls were using social media every day, compared to fewer than 60 percent in 2009. Jonathan Haidt, an experimental psychologist making significant contributions to understanding the negative impacts of modern-day “protections” of teen wellbeing, notes in his review of published research on social media and teens that 55 studies show significant correlations between social media use (usually measured in terms of time spent on social media) and depression or anxiety, compared to only 11 studies reporting no significant link between them. This ignores publication bias, whereby research that does not show a correlation is less likely to be published. But this is not the only problem.
Intervention is Better Than Hand-Wringing
There are significant issues with most studies on teen girls’ mood and social media use, whether or not these studies report any correlation between them.
Our recent study set out to assess the replicability of our 2019 project where we found that tweaking girls’ social media use by inviting them to add at least five positive female role models—profiles carefully picked in accord with each girl’s specific interests—dramatically changed their social media. In response to this intervention, they used social media more positively to explore goals and interests. Many of the girls participating in this exercise also went on to “clean up their social media feed” (Hence our use of “Disrupt Your Feed” as the title for the project).
In the recent follow-up project with teen girls between the ages of 13 and 17, we also wanted to explore links between social media use and mood. But the challenges we had in assessing teens’ moods lead me to question how robustly anyone can explore these links.
Mismatched Measures
We found that girls’ moods vacillated markedly, from week to week, day to day, and even within a single day. We found that their use of and response to social media was dependent on their mood before they looked at social media. When their mood was “up,” those glossy images of others’ glamorous lives could lift their spirits. When their mood was “down” they were reminded of their deficits.
Even more surprising, for me, was that the Life Satisfaction Scale, often taken as a proxy for girls’ well-being, bore no relationship to teen girls’ moods. Girls could rate their overall mood very low yet have a strong positive result on the Life Satisfaction Scale because they saw that the general conditions of their lives were objectively speaking close to ideal, especially as they compared their situation to those in the midst of war, or unable to heat their homes and access food.
Bring Teens’ Agency Into the Picture
My primary concern is that the heated debates, as currently formed, leave out one of the most interesting findings of our new project. Teen girls themselves, usually presented as hapless victims of, in Michelle Goldberg’s words, “unaccountable corporate behemoths,” were in fact working hard to gain control over their social media feeds.
They noticed how their more positive use of social media influenced the algorithms that governed their feed. They noticed that they could influence their feed by dismissing some threads or profiles using any “this doesn’t interest me” option available. They noticed that passive scrolling left them dissatisfied and depleted, whereas active searching for posts of interest could be inspiring.
This does not mean that they nailed the problem. They knew that higher powers were working to magnify their less productive interests, or “foibles” as they called them. But if policymakers and educators have a real interest in improving social media health, and if those corporate behemoths really want to serve their customers, then our study shows they should start with what teen girls can teach them about their own successes and frustrations in maintaining their mental health while using social media.
