Social media is depriving today’s teens of human connection, forcing them to compete with unrealistic, curated lives, and ultimately making them depressed, lonely, and unhappy. So the narrative goes.

But a new study suggests those concerns might be exaggerated.

Research from the Oxford Internet Institute, part of the University of Oxford in England, found that social media has very limited effects on the typical teenager’s well-being. The study, published this week in the Proceedings of the National Academy of Sciences, was based on eight years of survey data on U.K. teenagers.

“We’re not saying there’s no adolescent out there who is positively or negatively affected by social media,” said Amy Orben, coauthor of the study and a psychology researcher at the university. “But taken on average, the grave amount of concern at the moment is overblown.”

While previous research on the subject often captured data on teens’ social media use and well-being at one point in time, this study used data collected over time, allowing researchers to track changes.

If a teen started using social media more, did that increase or decrease well-being? If teens felt better or worse about their life, did that impact their social media use?

Many studies have been unable to answer those questions. They have focused solely on the correlation between social media use and how teens felt without knowing which factor causes the other.

Orben and her colleagues wanted to address that gap. They analyzed data in more than 2,000 ways, testing different options for the plethora of decisions researchers make — from how many years of data to analyze, to how to handle missing data and what factors to control. They posted their analysis online.

“The social media well-being debate is so tense,” Orben said. “This work tries to put all our cards on the table and hope that it helps the discussion progress.”

The study found that to a small degree, increased social media use led to teens reporting lower satisfaction with their lives. The reverse was also true: Lower life satisfaction led to increased social media use. Neither effect was large enough to seem significant in real life, Orben said.

But Melissa Hunt, a University of Pennsylvania psychologist, is skeptical.

Hunt was not involved in the Oxford study but has researched social media’s effects on young people. A study she coauthored last year found that decreasing social media use can lessen depression and loneliness.

“Any conclusions drawn can only be as good as the original data set they are based on,” Hunt wrote in an email. Relying on teens to self-report how much time they spent on social media is likely to result in inaccurate data.

“Even if adolescents wanted to answer the question completely honestly (which is highly unlikely), they probably have no idea how much time they actually spend on social media and wouldn’t remember accurately if they ever did know,” she wrote.

Orben said her team chose to use this information because more precise data is owned by social media companies, like Facebook and Twitter. In the future, she hopes the corporations will share their data with scientists.

Hunt also raised concerns that asking teens to report life satisfaction is not a scientifically valid way of measuring well-being. In contrast, her study used the UCLA Loneliness Scale, which has been tested by various researchers.

Still, Orben and Hunt agreed that more research is needed to inform the debate on social media’s health effects.

Both of their studies showed some indication that social media might be more harmful for certain groups, such as teenage girls or adolescents who are predisposed to depression and anxiety.

Exploring those subtleties more will be key moving forward, Orben said

“Parents would never trust me if I said I know exactly how your kid will react to eating sugar,” she said. It’d depend on whether the child was diabetic and eating it in a chocolate cake, or an athlete eating it in a granola bar after practice.

That same level of nuance needs to be brought to the conversation on social media, she said.

“Social media is a really easy way to explain decreases in adolescent well-being,” Orben said. But in recent years, there have also been large political changes — Brexit in the United Kingdom and the Trump presidency in the United States. There has been economic uncertainty and more pressure on students to start achieving early to ensure lucrative careers.

“We shouldn’t let the social media debate drown out other things that affect adolescents,” Orben said.