Lifestyle / does-social-media-make-you-lonely
Does Social Media Make You Lonely? What the Research Shows
Social media can make some people lonelier, but for most users the effect is small, and part of the link runs the other way: lonely people tend to use it more. Three experiments in university students found that cutting social media use for two or three weeks reduced loneliness. Three others found no benefit, and one of them found that loneliness went up when people stopped.
The clearest links in the research are with problematic use, the kind that feels compulsive and crowds out other things. Studies of time spent mostly find little or no link with loneliness. How you use it also seems to matter, because direct contact with people you know behaves differently from scrolling a feed. So it helps to look at what social media replaces in your day, as well as how many hours it takes up.
How common is loneliness?
Loneliness is common enough that any link with social media matters. In a June 2025 report, the World Health Organization's Commission on Social Connection estimated that 1 in 6 people worldwide is affected by loneliness, and that it is linked to more than 871,000 deaths a year. That death figure is a modelled estimate, not a count. The same report names digital technology among the possible drivers, citing excessive screen time and negative online interactions.
Younger people report it most. The WHO found that 17% to 21% of people aged 13 to 29 said they felt lonely. In the US, a Gallup survey of 6,289 adults in late 2024 found 20% had felt lonely "a lot of the day yesterday", down from a pandemic peak of 25%. Gallup's May 2025 analysis found the figure was 25% for US men aged 15 to 34, against 17% for all other adults.
In England, the government's Community and Engagement Survey for 2025/26, with more than 175,000 responses, found 7% of people aged 16 and over felt lonely often or always. The figure was 9% among 16 to 24-year-olds and 4% among 65 to 74-year-olds, so the youngest adults report the most frequent loneliness. That pattern on its own says nothing about cause.
What the experiments found
Experiments that ask people to cut back or stop are the best test of cause, and their results disagree. The six below all measured loneliness and show how far the findings diverge.
Three studies where cutting back helped
- Ten minutes per app. In a 2018 trial in the Journal of Social and Clinical Psychology, Melissa Hunt and colleagues at the University of Pennsylvania assigned 143 undergraduates either to limit Facebook, Instagram and Snapchat to 10 minutes each a day for three weeks, or to use them as normal. Use was checked with iPhone battery screenshots. The limited group showed significant falls in loneliness and depression compared with the control group, with the drop in depression largest among those who started out more depressed. The authors described this as a causal link. It is one small trial in iPhone-owning students over three weeks, which is a narrower base than many articles suggest.
- One hour a day. In a 2026 trial in the Journal of Affective Disorders, Gary Goldfield and colleagues recruited 260 Canadian students aged 17 to 25 who had symptoms of anxiety or depression and used social media for at least two hours a day. Of the 219 who finished, those asked to keep to one hour a day cut their use by an average of 78 minutes a day, about half, and their loneliness fell compared with the control group. PsyPost's summary of the study reports the effect as small to moderate, in a mostly female group of psychology volunteers.
- Thirty minutes a day. In a 2023 study in Technology, Mind, and Behavior, Faulhaber, Lee and Gentile at Iowa State University asked half of 230 college students to keep social media to 30 minutes a day for two weeks, with daily reminders. The limited group ended with lower loneliness, anxiety, depression and fear of missing out than the comparison group, including students who sometimes went over the limit. Outcomes were self-reported and the follow-up was short.
Three experiments where it did not
- Up to four weeks off. In a 2019 study in Media Psychology, Jeffrey Hall and colleagues at the University of Kansas randomly assigned 130 people to no change or to one to four weeks off Facebook, Instagram, Twitter and Snapchat, with daily diaries. They found no effect of abstinence on loneliness, wellbeing or quality of day, and the number of weeks off made no difference.
- Loneliness went up. In a 2019 trial in Perspectives in Psychiatric Care, Zahir Vally and Caroline D'Souza randomised 68 people to abstain from social media or carry on as normal. Compared with the control group, the abstainers reported an increase in loneliness. The sample was small.
- Four weeks without Facebook. The largest experiment, published in the American Economic Review in 2020, paid US adults to deactivate Facebook for four weeks before the 2018 midterm elections. Hunt Allcott and colleagues at Stanford and New York University found small but significant gains in life satisfaction, anxiety, depression and happiness, and their working paper puts the overall wellbeing gain at 0.09 standard deviations. Loneliness was not among the clearly significant effects. People who quit spent more time with friends and family, but the number of friends they had seen in person did not change significantly.
Put together, cutting back seems to help some people, especially heavy users who are already struggling, as in the 2026 trial. For others it makes no difference, or removes contact they valued. All three studies that found a benefit limited use rather than stopping it, and all were in students. A broad review of the field points the same way. A meta-analysis led by Jeffrey Hancock at Stanford, covering 226 studies and 275,728 people from 2006 to 2018, found that social media use had small links with anxiety and depression "but not loneliness". More than three-quarters of those studies were snapshots taken at a single point in time.
If you want to try a break and see what changes, the wider trials are covered in does a digital detox actually work.
Lonely people use social media more
Part of the link runs from loneliness to social media, not the other way round. When researchers follow the same people over time, they find effects in both directions.
- Within days. In a 2013 study in PLOS ONE, Ethan Kross and colleagues at the University of Michigan texted 82 young adults five times a day for two weeks. Facebook use predicted later drops in mood and life satisfaction, and people used Facebook more over time when they felt lonelier. Talking to people directly, by phone or face to face, did not predict the same drop.
- Over months and years. A 2025 meta-analysis in the Journal of Medical Internet Research by Jia Yuin Fam and Niko Männikkö pooled 26 longitudinal studies of 24,798 people. Loneliness predicted later problematic media use, and problematic use predicted later loneliness, with similar small effects each way. The authors found signs of publication bias and called for more studies.
The best-known survey result fits this picture. In a 2017 study in the American Journal of Preventive Medicine, Brian Primack and colleagues at the University of Pittsburgh surveyed a nationally representative sample of 1,787 US adults aged 19 to 32. Those in the top quarter for time on social media had twice the odds of feeling socially isolated as those in the bottom quarter. It was a single survey at one point in time, and the authors said the direction of the link was still to be worked out. It is often quoted as proof that social media causes isolation. It cannot show that.
The Facebook trial makes the same point in numbers. Allcott's team found the experimental effect on wellbeing was much smaller than the gap they saw between heavy and light users at the start, and said the difference was consistent with reverse causality, such as lonely or depressed people spending more time on Facebook.
What social media is replacing
The more convincing case against social media is about displacement, meaning what you are not doing while you scroll. In the US, time with friends in person has fallen sharply over two decades.
Average time spent with friends in person, US
Show the numbers
| 2003 | 60 min |
|---|---|
| 2019 | 34 min |
| 2020 | 20 min |
A 2023 analysis in SSM Population Health by Viji Kannan and Peter Veazie used the American Time Use Survey, a government diary study, covering 219,368 people. Average time spent with friends in person fell from 60 minutes a day in 2003 to 34 minutes in 2019, then to 20 minutes in 2020. People aged 15 to 24 saw the steepest fall. Time spent alone rose from 285 to 333 minutes a day over the same period.
The study does not show that phones caused this. The decline began in the 2000s, well before the pandemic, the survey records only one diary day per person, and the authors say the 2020 figures probably overstate how much time people spent together because data collection paused that spring. But it gives a plausible mechanism. If a feed fills the evening that used to go on seeing people, loneliness could rise even if no single post makes you feel worse.
That is also why the same app can help one person and harm another. Social media that leads to a meeting, a phone call or a reply from a friend is doing something different from social media that replaces them. For couples, phubbing is the version of this that happens in the same room.
Does how you use it matter?
Probably, but not in the simple way often claimed. For years the standard advice was that "active" use, such as posting and messaging, is fine and "passive" scrolling is harmful. A 2022 review in New Media & Society by Patti Valkenburg and colleagues at the University of Amsterdam found most studies did not support that split, and argued that both categories are too broad. Neither Allcott's Facebook trial nor Hancock's meta-analysis found a reliable difference between active and passive users.
Two narrower distinctions hold up better:
- Problematic use versus time. The longitudinal links with loneliness are clearest for use that feels compulsive or hard to stop, as in the 2025 meta-analysis above. Studies of time spent, such as Hancock's, mostly find little or no link with loneliness.
- Bad experiences versus good ones. In a 2019 follow-up, Primack's team surveyed 1,178 students and found that negative experiences on social media were linked to feeling more isolated, while positive experiences did not offset them.
The comparison trap covers one of the commonest negative experiences: measuring your life against other people's best moments.
Can social media reduce loneliness?
For some groups, it appears to. A 2025 meta-analysis in the Journals of Gerontology by Ruoxuan Chen and colleagues pooled 19 reports covering 24,877 adults over 50. More social media use went with slightly less loneliness (r = -0.06), and the link was stronger in developing regions (r = -0.24). These are correlations, and they do not show that social media made older people less lonely. They do suggest that for people whose friends and family live far away, staying in touch online may help more than it harms.
Evidence that video calls can reduce loneliness is weak. A 2020 Cochrane rapid review found that video calls for older adults may make little or no difference to loneliness, and rated the evidence as very uncertain.
How to use social media without feeling lonelier
No trial has tested this exact set of steps, so treat it as a starting point based on the studies above.
- Notice what you open it for. If you reach for a feed when you feel low or alone, that is the pattern the longitudinal research links to loneliness. A reply to a friend is a different use from 40 minutes of scrolling.
- Limit use rather than quitting outright. The studies that found a benefit capped time rather than banning social media. None tested feeds and messaging separately, but the 2013 study found direct contact did not predict the same drop in mood, so limiting feed apps while keeping messages to friends is a reasonable choice.
- Turn some scrolling time into contact. Use the time you save for a call, a voice note or a plan to meet. Displacement is the most plausible way feeds make people lonelier, so put something social back in its place.
- Unfollow accounts that leave you feeling worse. Negative experiences mattered more than positive ones in the 2019 survey.
- Measure for a fortnight. Compare your use and how you feel against a normal week using the free screen time tracker. If nothing changes after cutting back, the app may not be the main factor.
To set a daily limit on feed apps:
- iPhone — Settings > Screen Time > App Limits > Add Limit, expand Social, tick only the feed apps (for example Instagram or TikTok), then tap Next and choose a time. Ticking the whole Social category would also limit messaging apps.
- Android (Pixel) — Settings > Digital Wellbeing & parental controls > Dashboard, choose the app and set an App timer.
- Android (Samsung) — Settings > Digital Wellbeing and parental controls > App timers, then choose the app.
How to limit TikTok, Instagram and YouTube has app-by-app settings. If you decide to leave a platform completely, how to quit social media covers keeping in touch with the people who were only reachable there.
What this means for teenagers
Most of the experiments above used students and adults, not children. For teenagers, the research on social media and mental health finds small average effects with large differences between individuals, as covered in how much of teen mental health is about screens. The Vally result, in adults, suggests a sudden ban can remove contact some people value. For a teenager whose friendships happen mostly online, that risk may be higher, so swapping some feed time for time with friends is a more realistic aim than zero.
Common questions
Is social media the main cause of the loneliness epidemic?
The evidence does not support that. In the US, time with friends was already falling in the 2000s. Loneliness also predicts heavier use, and the experiments that cut social media found small effects or none. The WHO's 2025 report treats digital technology as one possible driver among several, alongside factors such as poor health, low income and living alone.
Does deleting social media make you less lonely?
Sometimes. The studies that reduced loneliness limited use rather than stopping it, and all studied students. Complete abstinence for up to four weeks made no difference in one experiment, and loneliness rose in another. Limiting use is better supported than quitting outright.
Why do I feel lonely after scrolling?
The research points to three likely reasons: comparing yourself with other people's highlights, bad experiences such as arguments or being left out, and time spent scrolling that would otherwise go on contact with people. Feeling low may also be what led you to open the app, which is the reverse direction found in the longitudinal studies.
Is messaging friends better than scrolling?
Probably. Talking to people directly did not predict a drop in mood in the 2013 experience-sampling study, while Facebook use did. The broader "active versus passive" split is less clear-cut than it was once thought to be, so judge by whether the contact leaves you feeling closer to someone.
When to get help
Cutting back on feeds is low-risk and worth trying for a few weeks. If loneliness has lasted for months, if it comes with low mood, poor sleep or loss of interest in things you used to enjoy, or if you feel you have no one to talk to, speak to your GP. Loneliness is not a medical condition, but it often comes alongside depression and anxiety, which can be treated, and a GP can point you to local groups and support. The NHS Every Mind Matters guide to loneliness has practical steps. If you are struggling to cope, Samaritans is free to call on 116 123, day or night.
sources for this page
- Social connection linked to improved health and reduced risk of early deathWorld Health Organization (Commission on Social Connection) · 2025 · The 871,000 annual deaths figure is a modelled estimate of deaths linked to loneliness, not a count.
- Daily Loneliness Afflicts One in Five in U.S.Gallup · 2024 · 6,289 US adults, 27 August to 4 September 2024 · Single-day self-report; Gallup has no pre-pandemic baseline for this measure.
- Younger Men in the U.S. Among the Loneliest in WestGallup · 2025 · Gallup World Poll, 2023 and 2024 pooled · Self-reported loneliness on the previous day; subgroup estimates carry wider margins of error.
- Community and Engagement Survey 2025/26: loneliness reportDepartment for Culture, Media and Sport (gov.uk) · 2026 · More than 175,000 responses, adults in England, October 2025 to March 2026 · England only; measures how often people feel lonely, not why.
- No More FOMO: Limiting Social Media Decreases Loneliness and DepressionJournal of Social and Clinical Psychology · 2018 · 143 undergraduates, 3 weeks · iPhone users at one university; small sample and short follow-up.
- Reducing social media use decreases loneliness regardless of gender or level of social comparisons in youth with anxiety and depression: A randomized controlled trialJournal of Affective Disorders · 2026 · 260 students aged 17 to 25 recruited, 219 analysed, 3 weeks · Mostly female psychology volunteers with anxiety or depression symptoms; details here are from the abstract and press coverage.
- Limiting social media to one hour a day reduces loneliness in distressed individualsPsyPost · 2026 · Science journalism summarising the Goldfield trial; used for the 78-minute reduction and effect-size description.
- The effect of self-monitoring limited social media use on psychological well-beingTechnology, Mind, and Behavior · 2023 · 230 college students, 2 weeks · Short follow-up and self-reported outcomes; not confirmed from the abstract whether assignment was random.
- Experimentally Manipulating Social Media Abstinence: Results of a Four-week Diary StudyMedia Psychology · 2019 · 130 community and student participants, up to 4 weeks · Abstinence was self-reported; people could still view content without logging in.
- Abstinence from social media use, subjective well-being, stress, and lonelinessPerspectives in Psychiatric Care · 2019 · 68 participants · Small sample; found loneliness increased among abstainers.
- The Welfare Effects of Social MediaAmerican Economic Review · 2020 · US Facebook users paid to deactivate for 4 weeks in 2018 · Facebook only, during an election period; participants were paid to take part.
- The Welfare Effects of Social Media (NBER Working Paper 25514)National Bureau of Economic Research · 2019 · Working-paper version of the AER study; source of the 0.09 SD wellbeing figure and the reverse-causality discussion.
- Social Media and Psychological Well-BeingThe Psychology of Technology (American Psychological Association) · 2022 · 226 studies, 275,728 participants, 2006 to 2018 · More than three-quarters of included studies were cross-sectional.
- Facebook use predicts declines in subjective well-being in young adultsPLOS ONE · 2013 · 82 young adults, texted 5 times a day for 2 weeks · Small sample and Facebook only.
- Loneliness and Problematic Media Use: Meta-Analysis of Longitudinal StudiesJournal of Medical Internet Research · 2025 · 26 studies, 24,798 people · Covers problematic media use rather than time spent; the authors found signs of publication bias.
- Social Media Use and Perceived Social Isolation Among Young Adults in the U.S.American Journal of Preventive Medicine · 2017 · 1,787 US adults aged 19 to 32 · Cross-sectional survey from 2014; cannot show direction.
- Positive and Negative Experiences on Social Media and Perceived Social IsolationAmerican Journal of Health Promotion · 2019 · 1,178 students aged 18 to 30 · Cross-sectional survey; cannot show direction.
- US trends in social isolation, social engagement, and companionship, nationally and by age, sex, race/ethnicity, family income, and work hours, 2003-2020SSM Population Health · 2023 · 219,368 American Time Use Survey respondents · One diary day per person; descriptive, does not identify causes; 2020 data collection was paused for 52 days.
- The associations of active and passive social media use with well-being: A critical scoping reviewNew Media & Society · 2022 · Scoping review rather than a meta-analysis.
- Lonely Older Adults in the Era of Social Media: A Meta-Analytic ReviewThe Journals of Gerontology: Series B · 2025 · 19 reports, 24,877 adults over 50 · Pooled correlations; cannot show that social media reduced loneliness.
- Video calls for reducing social isolation and loneliness in older people: a rapid reviewCochrane Database of Systematic Reviews · 2020 · Very uncertain evidence from three small trials at the 3-month point.
- LonelinessNHS Every Mind Matters · 2025 · Health service guidance rather than a study.
- Contact SamaritansSamaritans · 2026 · Support service, not a study.