That screen time notification arrives every Sunday morning, quietly delivering its verdict: four hours, up 12% from last week. Most notifications get dismissed without a second thought. But researchers from the University of Cincinnati and Northwestern University spent 11 years and 183,000 survey respondents figuring out exactly where “a lot” becomes “a problem.” The number is 150 minutes. Two and a half hours. Probably less than you think.
What 11 Years of Data Actually Shows
The same pattern appeared every single year — platforms changed, a pandemic arrived, and the association held anyway.
The Media Behaviors and Influence Study tracked roughly 16,700 fresh U.S. adults annually from 2014 to 2024, measuring media use hour-by-hour across 15 variables — age, income, gender, gaming, TV, email, and general browsing. Models correctly classified depression status 72–77% of the time, consistently. Below 150 daily minutes of social media, the link to self-reported depression stayed roughly neutral. Cross that line, and it shifted — clearly and consistently.
By the numbers:
- 183,000+ adults surveyed across 11 annual waves, matched to U.S. Census demographics
- 150 min/day identified as the model’s tipping point
- Each additional hour of daily use associated with ~17% higher odds of self-reported depression
- Models held steady through pandemic years — removing COVID data changed nothing material
- Women showed a stronger association than men, consistent with existing adolescent research
From 2019 onward, social media rose to become the top-ranked single predictor of depression in the models — above occupation, education, TV viewing, and web use. When researchers combined three different importance metrics, personal income held the strongest position overall. Social media ranked right behind it, alongside general internet use and age.
What This Doesn’t Prove – And Why That Still Matters
Correlation sustained across 11 years is worth taking seriously; establishing causation is an entirely different conversation.
The study cannot tell you which came first. New participants each year means no individual tracking — so researchers can’t determine whether heavy use drives depression or whether depressed people scroll more. Both directions remain plausible, and unmeasured factors could influence both behaviors simultaneously.
The study also treats all social media time as equivalent, which matters. Passive scrolling and active messaging likely land differently, emotionally — prior research suggests these behaviors carry distinct psychological weight. Think of the 150-minute figure like a calorie count on a menu: genuinely useful context, not a prescription your doctor wrote.
Gaming occasionally outranked social media year-by-year in the models. News consumption barely registered, reaching significance in only 2 of 11 years.
Researchers describe the 150-minute threshold as a hypothesis for future longitudinal and experimental work — not a clinical cutoff or personal health rule, according to coverage of the npj Mental Health Research paper.
Broader meta-analyses find the association between social media and depression is real but modest. Context, content type, and individual vulnerability matter as much as raw minutes, with some policy discussions now focusing on AI age laws as a potential lever, health research reviews note consistently.
Two and a half hours of daily social media is ordinary — not excessive by most measures — which is precisely what makes this finding worth sitting with. Your screen time app already has your number. The research now offers some context for what that number might mean, even if the full story isn’t written yet.
