Families & young people
Screen time and digital wellbeing: what the evidence actually supports
Total screen time is a weak measure of harm. What matters more is what the time is spent doing, when it happens, and what it displaces — plus the changes that genuinely help.
"Screen time" is the metric everyone reaches for because it is the one that is easy to count. It is also, on its own, one of the least informative numbers you can collect about a person's relationship with their devices.
Two teenagers with identical daily totals can be having completely different experiences: one video-calling friends and editing a film, the other scrolling passively through content selected by strangers at one in the morning. The number is the same. Nothing else is.
This guide is about the measures that carry more signal, and the changes that tend to help.
Three questions that beat the total
What is the time being spent doing? The most useful distinction is between active and passive use. Active use — messaging people you know, calling family, creating something, playing a game with friends — behaves quite differently from passive consumption of an algorithmically-chosen feed. Research in this area is genuinely contested and often correlational, but the active/passive split holds up better than raw duration.
When is it happening? Displacement is where much of the real harm sits, and the largest single displacement is sleep. Sleep loss has well-established effects on mood, concentration, immune function, and impulse control — and a great deal of what gets attributed to screens generally is downstream of not sleeping enough. Time that displaces sleep is categorically worse than the same time on a Sunday afternoon.
How does it end? Notice the difference between stopping because you are finished and stopping because something interrupted you. Feeling unable to stop, or consistently spending far longer than intended, is a more meaningful signal than any total. So is how someone feels afterwards: satisfied and connected, or flat and slightly worse than before.
If you want one thing to watch instead of the number, watch the after.
What the design is doing
None of this is an accident, and it helps to name the mechanisms rather than treating self-control as the only variable.
- Infinite feeds remove the natural stopping point that a finite page or an episode ending used to provide.
- Variable rewards — sometimes the next item is great, usually it is not — produce the same persistence pattern as slot machines. Unpredictability is the active ingredient.
- Autoplay makes continuing the default and stopping the decision.
- Social obligation mechanics — streaks, read receipts, activity indicators — convert a leisure activity into a commitment with a penalty for absence.
- Notifications manufacture re-entry points throughout the day, most of which were not urgent.
Understanding this matters because it reframes the problem. Someone struggling to put a device down is not displaying a character defect; they are interacting with a system built by skilled people to produce exactly that outcome.
Changes that tend to work
In rough order of effect for effort:
Get devices out of the bedroom overnight. This single change addresses the sleep displacement problem, the late-night-scrolling problem, and the wake-up-and-immediately- scroll problem simultaneously. Buy a cheap alarm clock so the phone is not needed. It is the highest-leverage change available to most households.
Turn off notifications from anything that is not a person. Not all notifications — messages from actual humans are usually worth having. What is worth removing is the entire class of notification generated to bring you back: recommendations, reminders that something is happening, activity summaries. Most people find their phone becomes noticeably calmer within a day.
Remove the two or three worst apps from the home screen. Not deleted — moved somewhere that requires deliberate searching. Friction defeats absent-minded opening, which is where a surprising share of the total accumulates.
Create device-free contexts, not device-free hours. Meals, the car, the first thirty minutes after school or work. A rule tied to a situation is easier to keep than one tied to a clock, and it produces the conversation time that everything else depends on.
Make the boundaries symmetrical. Rules that apply to children and not to adults do not survive contact with adolescence, and they undermine the credibility of everything else you have said.
For parents specifically
Parental controls are useful and limited, and being clear about which is which prevents disappointment.
They are good at: age-appropriate content filtering for younger children, providing visibility into what is installed, and enforcing device-off periods.
They are poor at: stopping a determined and technically capable teenager, and substituting for a conversation.
The evidence on what actually protects adolescents points consistently at the relationship rather than the software. A teenager who believes they can tell you about something frightening without losing their phone will tell you. One who believes disclosure means confiscation will handle it alone, which is the outcome you least want.
That has a direct practical implication: separate the response to something bad happened to you from the response to you broke a rule. If those two produce the same consequence, you have taught them not to report.
The age-appropriate conversation structure is covered in more depth in A parent's guide to teenagers and social media.
Signals worth paying attention to
Ordinary heavy use is not the same as a problem. The things that warrant a closer look are changes rather than levels:
- Withdrawal from activities and friendships that were previously enjoyed offline
- A marked shift in mood specifically following use
- Secrecy that is new, particularly about who they are talking to
- Sleep disruption that persists
- Declining performance or attendance
- Distress when access is unavailable that is out of proportion
Any of these individually can be adolescence. Several together, sustained over weeks, is worth a conversation — and if it persists, worth involving a professional. Framing this as a wellbeing question rather than a technology question usually gets further.
For adults, honestly
Most of the advice aimed at families applies unchanged to the person delivering it. The adult version of the problem tends to be work-shaped — the inbox that is never closed, the notification that arrives at nine in the evening and is answered because it seems easier than leaving it — and it is often defended as necessity in a way that a teenager's use is not.
The same three questions apply. What was that time spent doing, what did it displace, and how did it end.