The Biggest Screen-Time Study Ever Done Just Came Out. Here's What It Actually Says

On September 28, 2026, JAMA Pediatrics published what is probably the largest screen-time study ever assembled: an "umbrella review" — a study of studies — pulling together 23 separate meta-analyses covering 2,817,252 children and teenagers across 64 countries. Lead author Elizabeth Al-Jbouri at the University of Calgary and her team weren't trying to run a new experiment. They were trying to answer a simpler question: when you combine everything researchers have already found, what does screen time actually do to kids?

If you've seen this story shared already, you've probably seen one of two reactions: "see, I told you screens are destroying this generation" or "it's fine, studies always say this and nothing changes." Both reactions skip the part that actually matters for a parent trying to make a decision tonight — the size of the effect.

What the numbers actually say

Here's the breakdown, in the same units researchers use (correlation coefficients, roughly "how strong is the link," where 0 = no link and 1 = a perfect link):

To translate "r = 0.10 to 0.15" into something concrete: researchers studying effect sizes generally treat anything under 0.1 as tiny, 0.1–0.3 as small, and above 0.3 as starting to matter at an individual level. Most of these numbers sit in the small-to-moderate range — real, worth paying attention to, but nowhere near "screens are the dominant factor in how your child turns out."

Two outside reviewers quoted in coverage of the study — Professor Esther van Sluijs at Loughborough University and Tatiana Íñiguez Berrozpe at the University of Zaragoza — both made the same point from different angles: the effect sizes here are small (researchers estimate screen time alone accounts for roughly 1% of the variation in these outcomes), and because every underlying study was observational, none of this proves screens cause the problems. A kid who's already anxious might reach for a screen more, not the other way around.

Where the signal is strongest

If you only remember one part of this review, make it this: the clearest, most consistent links were with emotional and social outcomes — peer-relationship problems (r≈0.28) and general emotional difficulties (r≈0.30) — not academics. Reading, math, and science scores showed no significant relationship at all.

That's the opposite of a lot of parenting advice, which tends to frame screens mainly as an academic threat ("it's rotting their focus, their grades will drop"). This data points somewhere else: screens seem to matter more for how kids feel and get along with other kids than for what they can calculate or read.

What this means for tonight's homework-vs-tablet argument

This isn't a license to stop paying attention — a moderate link to emotional difficulties and peer problems is not nothing. But it does change what's worth watching for:

A few other things worth knowing before you make any household rule off the back of this study:

• It's a review of existing studies from many countries, many app types, and many years — it can't tell you what's happening with your specific kid and their specific apps today.
• "No significant link to test scores" doesn't mean screens are harmless for learning — it means this particular body of research didn't find a reliable one, which is itself useful to know given how much "screens ruin grades" messaging is out there.
• Small effect sizes at a population level can still matter a lot for an individual child who's vulnerable in a specific way (already anxious, already socially isolated) — population averages hide individual variation.

Where CubLearn fits into this

We built CubLearn around the part of this research that actually has a clear lever: the type of engagement, not just the clock. Reading and practice sessions on CubLearn are built to be short, focused, and built around real content (not an infinite feed), specifically because the emotional/social outcomes in this review track more with passive, open-ended screen use than with short, bounded, content-driven sessions. We'd rather be the 20 minutes you don't have to feel guilty about than pretend any app is "zero risk."

The takeaway

The biggest screen-time study ever done is good news and bad news at once: the effects are real, but they're smaller than the scariest headlines claim, and they show up more in friendships and feelings than in grades. That's a much more useful thing to actually act on than a vague sense of guilt every time your kid asks for five more minutes.