No Evidence that Social Media Affects Mental Health, Zuckerberg Says (Patreon)

[This is a transcript with links to references.]

Last week, the US senate had a hearing  on the dangers of social media in preparation of a legislation to improve child safety online. In this hearing, Meta CEO Mark Zuckerberg  claimed that it has not been scientifically proved that social media causes mental health problems in adolescents.

This upset a lot of people who think that the link is obvious. But I am afraid Zuckerberg is right. Let’s have a look.

 Contrary to what some headlines have claimed, we don’t have a global mental health crisis. I discussed this at length in an earlier episode.  By and large mental health globally is remarkably stable.

However, the mental health of one demographic group has been suffering in the past decades, that’s adolescent girls in some countries, and to a lesser extent also boys. This is evident in data from the USA, the UK, Australia,  and some other countries where self-reports also correlate with self-harm. In other countries, such as Sweden, this mental health trend is primarily found in self-reports from girls, difficult to interpret, and does not correlate with self-harm statistics.

Still, many put the blame on the increased use of social media, and that includes the affected young people themselves. It’s a topic I care a lot about because I have two children who just got their first smartphones.

 The most vocal critic of social media use for children has probably been the American psychologist Jonathan Haidt. His major argument is that there isn’t any other hypothesis, which is also what string theorists said about the existence of strings, and generally not how science works.  

Now, look, I’m not saying that Haidt is wrong. I don’t know what’s going on. But I do know that the studies on the topic have been inconclusive. I don’t expect you to believe what I say, but Haidt has been criticized by his own colleagues for jumping to conclusions. And again it’s not that they say he’s wrong. They point out that to the extent that studies have found any influence of social media on young people’s mental health, it’s been a small effect.

For example, a study from 2019 followed about 13 thousand adolescents in the UK between the ages of 10 and 15, though not all of them participated in all parts of the study. They were asked how many hours a day they were interacting with friends through social media and to rate their well-being, but they were also asked about their life at home. The correlation between social media use and life satisfaction was very small and, in many cases, almost compatible with zero.

A similar conclusion was reached in 2020 by a group of researchers who followeda group of 500 adolescents in the United States for eight years from age 13 to 20. They found that more time spent on social media was not associated with more mental health issues. The authors conclude basically that research should move on.

However, then there’s this study from 2022 by researchers from the UK and the Netherlands. They looked at the mental health of over 20 thousand kids in the UK in the age range of 10 to 15. They found that for girls between the ages of 11 and 13 increases in social media use *predicted a decrease in life satisfaction ratings one year later. Something similar happened to boys but at ages 14 and 15. So this seems to show that there is both an effect, and the causation indeed goes from social media use to decreases in life-satisfaction and not the other way round.

Then again there’s a recent study by researchers from the UK published in the journal Nature Mental Health. These researchers followed over twelve thousand British teenagers. They found that while social media does have a negative impact on mental health, it’s one of the least influential factors. The authors say that it makes no sense to claim social media use is the main reason of adolescent mental health problems and that and other factors, such as bullying, or lack of family support should be the focus of attention instead.

So as you see it’s quite a mixed bag. Quite possibly the reason is similar to the issue with the supposed problem of political polarization and echo chambers, which Jonathan Haidt has also previously been going on about and that I discussed at length in yet another episode.

Upon closer inspection it turns out that while these problems of polarization and echo chambers do exist, their strength and prevalence depends on the medium and the country and on exactly what question you ask. Basically, the issue that psychology and sociology are very context dependent. Not every field of science is as nice as physics where you have universal laws and Einsteins whose theories still hold up a century later.

My reading of the literature on the subject of social media and mental health is that some psychologists are a little wary and worry that blaming social media for whatever problem children might paper over other issues.

And just so we’re on the same page, it’s not like I’m saying social media is no problem for children, or anyone really, or that law which the senate is working on isn’t needed. Actually, much of the senate hearing focused on other issues, such as privacy concerns or children being able to buy products they shouldn’t get their hands on. Or being exposed to content that’s unsuitable for their age such as deep existential problems posed by quantum mechanics. It’s a wild world out there, so stay safe.