What is the correlation between screen time and adolescent well-being? It’s an important question for which we have limited and often contradictory answers. A story in Wired this week all but let our phones off the hook for any negative consequences — but I wonder if the study fully understands the way we live now.
Much of the research to date on the potential effects of screen time on well-being relies on self-reported data from massive surveys, Wired reports. The sheer amount of data available means that creative scientists can make nearly any argument about correlations, leading to widespread confusion about the truth.
In the latest issue of Nature Human Behavior, researchers Andrew Przybylski and Amy Orben apply a novel statistical method to this problem: a tool called specification curve analysis that lets them evaluate many thousands of possible correlations simultaneously. Here’s Wired’s Robbie Gonzalez on their findings:
The result was a series of visualizations that map the wide gamut of potential effects researchers could detect in the three repositories, and they reveal several important things: One, that small changes in analytical approach can lead to dramatically different findings along that spectrum. Two, that the correlation between technology use and well-being is negative. And three, that this correlation is very, very small, explaining — at most — 0.4 percent of the variation in adolescent well-being.
To put it in perspective, the researchers compared the link between technology use and adolescent well-being to that of other factors examined by the large-scale data sets. "Using technology is about as associated with well-being as eating potatoes,” Przybylski says. In other words: hardly at all. By the same logic, bullying had an effect size four times greater than screen use. Smoking cigarettes? 18 times. Conversely, getting enough sleep and eating breakfast were positively associated with adolescent well-being at a magnitude 44 and 30 times that of technology use, respectively.
What’s immediately striking about these findings is the way in which the researchers separate bullying from "screen use.” The data collected pertains to children who were born in the year 2000 or later. While surely bullying continues to happen offline, it has a major online component as well. (An anti-bullying initiative in Australia found that 84 percent of children who had been bullied offline had been bullied online as well.) Bullying and screen use are linked, in other words — but this paper has no means of separating them.
I searched the study for more information about the data it evaluated, and was surprised at how fusty the whole thing feels. The categories of screen use data collected include whether the child owns a computer, plays "weekday electronic games,” watches "weekday TV,” or "uses the internet at home.” These categories might be useful in longitudinal studies comparing a 1980s childhood to a 2010s childhood, but they hardly capture the rich variety of screen time a young person today might encounter. (There is a category for "hours of social media use,” which feels as relevant today as ever. )
But for the most part, these are pre-smartphone data categories being applied to a post-smartphone world. It seems fair to question whether any analysis of the data, no matter how statistically rigorous, can reflect the individual or collective effects of screens on our psychology or behavior.
Last month, writing about the Yellow Vest protests in France, Max Read lamented "theoretical discussions that imply the possibility of some counter-historical ‘control world’ without Facebook.” Facebook’s existence has exerted influence over us for more than a decade now, he argued, making it all but impossible to conceive a present-day world in which it never happened.
I wonder if smartphones aren’t like that, too. The majority of phones shipped globally have been smartphones — which is to say, phones that have internet access — since 2013. Trying to quantify a singular "effect” of smartphones on well-being, in a world where they are ubiquitous, strikes me as naive. Smartphones changed the way we meet romantic partners, send nudes, navigate through cities, buy drugs and alcohol, and spend time with our friends, to name just a few of their consequences. Good luck reducing all that to a variable.
In some ways, smartphones seem to have been quite good for young people — teen pregnancy rates are down, for example, and along with the rate of adolescent drug abuse. A popular theory here holds that teens are spending more time connecting with friends at home using screens and less time out in the world making mischief. At the same time, online bullying remains a scourge of every platform. (Did you know 42 percent of teens report being bullied on Instagram?) And if you believe that social platforms have contributed to the spread of hate speech, or the rise of far-right populism, or the radicalization of young minds, how do you factor that into a statistical account of well-being?
Wired’s headline, echoing Przybylski, reads, "Screens might be as bad for mental health as ... potatoes.” And yet reading the study, I’m far less certain. The data certainly seem to indicate what Przybylski suggests. But I’m not sure this data really answers the question we want it to — or whether the researchers even asked the right question to begin with.