New evidence from a randomized clinical trial on shutting off social media.

By Alexander Danvers Ph.D.

A new, small-scale clinical trial randomly assigned people to stop using social media for a week and compared them to people who kept using it as normal. Results indicate that being assigned to stop using all social media for a week—including Instagram, Facebook, Twitter, and TikTok—led to an increase in well-being and a decrease in depression and anxiety at the end of that week.

Researchers from the University of Bath in the U.K. recruited a group of 154 individuals through online ads and word of mouth. The people who responded ended up having an average age of around 30 and being majority white, female, and educated*. Ads for the study specified that it was about going off social media; the people who agreed to participate were likely already motivated to try stopping their social media usage. (And a handful of people who initially expressed interest later declined because they weren’t willing to stop using social media if they were assigned to that condition.)

This participant pool means that the results have some limitations. The effect is likely to be generally true of people like those studied—more educated, white, and female, and willing to stop using social media—but may be different for other groups. In particular, it would be important to conduct follow-up research on people who love using social media and don’t necessarily want to leave it, to see if social media might be supporting their mental health. Note that, as I pointed out in a recent post, there is some evidence that social media use is related to worse mental health in adolescent girls in particular.

Can we be sure they really stayed off social media?

The researchers used phone tracking apps to determine how much time participants actually spent on social media. These apps (which participants agreed to use to track their behavior for the study) showed that, overall, people were pretty faithful to the study instructions. They weren’t able to completely stop themselves from using social media, but they dramatically reduce the number of minutes they spent on such apps. For example, participants asked to stop using social media dropped the number of minutes they spent on Instagram a week from 222 minutes to 10 minutes.

They also conducted follow-up analyses indicating that spending fewer minutes on social media was associated with lower depression scores and greater well-being scores, suggesting that it really was the act of not using social media—as opposed to just telling yourself that you were getting off social media—that made a difference.

Follow-up analyses also suggested that social media only improved depression symptoms if the individual already had moderate to high levels of depression. In other words, if you are using social media and not particularly depressed, getting off social media is unlikely to make you feel even healthier.

A Grain of Salt

With only 154 participants, this was a relatively small study. When studying something that might have a large effect, a small study can be appropriate. Yet small studies also have a higher risk of reporting effects that are incorrect in serious ways. The scientific publishing system is biased towards presenting results that find a large effect and tell a compelling story–and biased against presenting results that show no effect or throw cold water on a popular theory. This publication bias (which has been well studied and documented by scientists) actually means that the small studies that do get published and attention tend to be overestimated (or even backward estimates) of what is really happening in the world. Overall, then, this should be treated as an interesting and suggestive piece of evidence, but no definitive proof that getting rid of social media would improve everyone’s mental health.

Research with much larger samples, such as over 1000 participants, would be more convincing. Further, it would be useful to follow up with participants after more than just a week, to see whether a social media break can have an extended effect—or whether people just bounce back to their previous levels of (un)happiness afterward. However, it’s also important to note that, unlike many of the studies of social media I mentioned in a prior post, this study did use the gold standard method for establishing cause-and-effect in research: randomly assigning some people to get the treatment, while allowing others to continue as normal. This is a useful step in exploring the role of social media on mental health and well-being.

*(The authors identify 90 percent of participants as having completed at least their “A levels”, a U.K. education qualification for people planning to attend university. It has been compared to completing a series of Advanced Placement (AP) tests in the U.S.).

Originally published at Psychology Today