Which social media apps harm teens the most?
Instagram and TikTok. In a national survey of 472 US teens, those two apps carried the highest rates of almost every harm the researchers measured, and YouTube ran several times lower on the same questions. Among teens using TikTok, 42.2% said the app was disrupting their sleep, against 10.2% of teens who named text messaging.
Most of the argument about teens and phones treats the internet as one big place. Researchers behind this survey did something different. They asked teens aged 13 to 17 what had happened to them in the past four weeks, then asked which of 20 specific apps it happened on. The same harm, measured the same way, on one app after another. That is how you find out whether “screen time” is the problem or whether the app itself is.
What the data show
Bad experiences online were common, and they were far more common online than off. Teens reported seeing violent imagery online at 10.7% compared with 1.6% offline. Unwanted sexual images ran 13.4% online versus 2.0% offline. Content encouraging self-harm came in at 9.7% online versus 5.2% offline. Across both worlds combined, 44.7% had witnessed harassment and 29.8% had been targeted by it directly. More than half of the teens, 51.5%, said technology was disrupting their sleep.
The app-by-app breakdown is where the ranking appears. Teens answered the same harm questions about each app they used, so the comparison is same teens, same four weeks, different apps.
Seeing violent imagery was reported by 9.1% of teens using Instagram, against 1.9% of teens using YouTube, close to five times the rate. Feeling worse about yourself because of other people’s posts was most common on Instagram at 16.6% and TikTok at 13.4%, against 4.1% on YouTube. Sleep took the biggest hit on TikTok, where 42.2% of teens using the app blamed it for disrupted sleep, followed by Instagram at 27.8%, while text messaging sat at 10.2%. The overall 51.5% sleep figure is not spread evenly across the phone. It concentrates on a short list of apps.
One number moved the other way. Direct harassment was reported at similar rates offline and online, 18.5% versus 16.4%. Bullying did not migrate to the internet so much as it added a second address.
Dr. Kumar’s Take
The useful idea here is not “social media is bad.” It is that the label on the app predicts the risk. A teen scrolling YouTube and a teen scrolling TikTok are both “on their phone” by any parental control app or screen time report, and a fivefold gap in violent imagery between those two apps says those two hours are not the same hour. Video feeds that push an endless personalized stream behaved differently from ones people mostly search and watch.
The sleep numbers carry the most weight for me, because sleep is the part of adolescent health where damage compounds fastest. More than half of these teens, 51.5%, reported technology breaking their sleep, and the apps were not equally responsible. When 42.2% of the teens using one app say it is costing them sleep, that is a public health number, not a parenting quibble.
This is a pilot survey, and it is worth saying what that limits. Everything here is what teens reported about themselves, with no way to confirm it. It is a snapshot, not a follow-up over time, so it cannot show that an app caused a harm rather than that teens prone to a harm gravitate to that app. And 472 adolescents is a modest sample once you split it 20 ways.
How strong is the evidence?
The survey ran from February to December 2025 through the Understanding America Study, a probability-based national panel, which is a real strength. Probability sampling means households were selected at random rather than recruited from people who volunteered to answer online questionnaires, so the results reflect US teens better than the typical internet poll does.
The design is cross-sectional, meaning it captures one moment rather than following teens forward. That makes it good at measuring how common something is and poor at proving what causes what. The authors present it as baseline measurement, the kind of routine monitoring public health already does for smoking and seat belts, and that framing is the right one. You cannot regulate or advise on something nobody has counted.
Why the platform matters more than the screen time
Screen time as a single number tells a parent almost nothing. Two hours of watching soccer highlights and two hours of an algorithmic feed serving self-harm content are the same number and different risks. This survey puts figures on that gap at the app level: 9.1% versus 1.9% for violent imagery, 16.6% versus 4.1% for feeling worse about yourself, 42.2% versus 10.2% for lost sleep. Which app is open turns out to be a sharper question than how many minutes it stays open, and sharper still after bedtime.
Practical Takeaways
- Ask which apps your teen actually uses rather than how many hours they spend, since Instagram and TikTok carried the highest rates of nearly every harm measured while YouTube ran several times lower.
- Treat overnight phone access as its own problem, given that 42.2% of teens using TikTok said the app was disrupting their sleep, the highest rate of any platform in this survey.
- Ask directly about unwanted sexual images and content encouraging self-harm, both of which teens reported far more often online than offline, because they are unlikely to volunteer it.
- Remember that harassment happens at similar rates offline and online, so taking the phone away does not remove the problem on its own.
Related Studies and Research
- Is vaping harmful for kids and teens?
- Do HIV shots every 8 weeks beat daily pills for teens?
- Can tai chi or mindfulness ease cancer fatigue and pain?
- Exercise timed to your body clock lowers blood pressure more
FAQs
Is YouTube safe for teens?
Safer than Instagram and TikTok on these measures, which is not the same as safe. YouTube came in at 1.9% for seeing violent imagery and 4.1% for feeling worse about yourself, both several times below the Instagram and TikTok rates, but neither is zero. The likely difference is how the content reaches a teen. A platform where people mostly search for something specific behaves differently from one that decides what you see next. Treat app choice as one lever, not a substitute for supervising.
How do I know if my teen’s phone is wrecking their sleep?
The tell is usually not the complaint, it is the pattern. Teens rarely say “my phone is keeping me awake,” but they will fall asleep in class, resist mornings, or spend weekends catching up on sleep. In this survey, more than half of teens said technology was disrupting their sleep, so the base rate is high enough that it is worth asking about even with no obvious symptoms. Charging the phone outside the bedroom removes the decision from a tired teenager at midnight.
Does this study prove Instagram and TikTok cause harm?
No. It is a cross-sectional survey, which counts what teens report at one point in time, so it cannot separate the app causing the harm from the harm and the app simply traveling together. A teen already struggling with body image may spend more time on image-heavy feeds. What the survey does establish is that the risks are unevenly distributed, and that is a real finding on its own. It gives regulators and researchers a baseline to test against instead of arguing from anecdote.
Bottom Line
Instagram and TikTok topped nearly every harm this national survey of 472 US teens measured, with YouTube several times lower on the same questions. Violent imagery reached 9.1% of Instagram users against 1.9% of YouTube users, and 42.2% of TikTok users said the app was costing them sleep. The practical message for families is that which app matters more than how many minutes, and that the hours after bedtime are where the damage concentrates.
