Verified by 
Social media researchers are as perplexed as the average parent about the conflicting and often hyperbolic messages being delivered about the effects of Facebook and the like on psychological well-being. It’s easy to find studies showing positive effects (larger networks, more social support and so on) and even easier to find studies showing negative effects (increases in depression or loneliness). Sometimes the contradictory papers are based on the same data set.
A few years ago, psychologist Jeff Hancock of Stanford University’s Social Media Lab set out to investigate what was going on. He and his colleagues embarked on an ambitious analysis that would assess all of the work—and I do mean all—that had been done in the field since the first paper was published on the subject of social media and well-being in 2006. Hancock announced his findings this week at the annual meeting of the International Communication Association in Washington, D.C. And the results should help to ease our collective minds.
“Using social media is essentially a tradeoff,” Hancock says. “You get very small advantages for your well-being that come with very small costs.”
One source of confusion in this area is the concept of statistical significance, which scientists use to describe something that happens more often than chance would predict. In both directions, positive and negative, the effects Hancock found were statistically significant, but not necessarily of practical significance. With all effects combined, the amount of variation in individual well-being that could be attributed to technology use was “essentially zero,” Hancock says. (To be specific, it was 0.01 on a scale in which 0.2 is considered a small effect.)
These results align with a study published earlier this year by Andrew Przybylski and Amy Orben from Oxford University. Using a more rigorous statistical technique than others have employed, they analyzed data on more than 350,000 adolescents and showed persuasively that, for the general population, technology use has a nearly negligible relationship to adolescent well-being. For context, they included comparisons with other factors that were included in the data. They found that eating potatoes is correlated with well-being to nearly the same degree as technology use, and wearing glasses even more so. Yet no one is saying teenagers should stop eating potatoes or wearing glasses.
All of this work reveals the pitfalls of the science of social media so far. Some studies have not been adequately designed. They may be unintentionally biased by what the researchers expect to find. For instance, if surveys use the word “addiction” in questions rather than more neutral language, the results are far more likely to show a negative effect.
Importantly nearly all work in this field so far shows only correlations, which is not the same as showing cause. It’s quite possible that the problems with well-being come first and drive people to turn to social media. “It could be a two-way street,” Hancock says.
Hancock’s study was a meta-analysis. Such research combines data from many studies of the same subject in order to try to get a better sense of the bigger picture. The paper, which is under review for publication, ultimately included 226 studies done over 12 years, and more than 275,000 participants.
To unpack the findings a little more, he found that research into the effects on well-being generally fall into six categories: depression, anxiety, loneliness, eudemonic happiness (finding meaning in life), hedonic happiness (enjoyment in the moment), and relationships. There are significant but small negative connections to anxiety and depression, though not loneliness. At the same time, there are significant but small positive links to life satisfaction and relationships. None of the effects rose above 0.2, which is considered small. The positive association with relationships, out of all six categories, is the largest (0.19).
So if the positive wins out anywhere, it may be when it comes to relationships, especially in extending social networks. And for older adults, additional online connections are strongly positive. “For older people, social media is a real avenue of connection, of relational well-being,” says Hancock. Those older adults who use social media report more support from both their grown children and their friends.
There are several important takeaways from Hancock’s work. “We’re asking the wrong question if we’re interested in wellbeing,” he says. Research to date has been largely centered on time spent online: both duration and frequency of use. But that focus misses the very important questions of what exactly people are doing online and with whom. Asking homework questions or Skyping with faraway relatives could be positive uses. Looking at pornography or inflammatory political content might be worse for a person’s well-being. Content, context and individual variation seem to be far more important than time.
This means that the science of social media needs to change. “We need new approaches to understanding media and our use of it,” Hancock says. “What Orben and Przybylski have shown is that we may not know exactly what the effects are because our research designs have been quite weak so far, but even with our gold standard measures it was well-being driving social media use and not the other way around.”
It’s also worth pointing out that journalists should be more careful in their coverage. Hancock finished his presentation in Washington by saying, “all the hyperbole in the media is adding additional stress and anxiety for parents and others.”
Copyright: Lydia Denworth, 2019
If you like what you read here and want to receive future posts in your inbox sign up here.
References
Orben, Amy, and Andrew K. Przybylski. "The association between adolescent well-being and digital technology use." Nature Human Behaviour 3.2 (2019): 173.
Orben, Amy, Tobias Dienlin, and Andrew K. Przybylski. "Social media’s enduring effect on adolescent life satisfaction." Proceedings of the National Academy of Sciences 116.21 (2019): 10226-10228.
