社交媒體對社會的影響

The social media impacts on society are easy to argue about and surprisingly hard to measure. Roughly 5.79 billion social media user identities existed at the start of April 2026, according to DataReportal's ongoing global tracking — a figure that grew by about 294 million, or 5.4 percent, in a single year, and that now covers more than two-thirds of the world's population on a monthly basis.

One caution before any of that becomes a talking point: those are identities, not people. Duplicate accounts, business profiles and bots all get counted, which is why social media user totals sometimes exceed the population of a country. If you have seen the claim that "more than 6 billion people use social media," it is an overstatement of a number that was already about accounts rather than humans.

This guide walks through what social media demonstrably does well, where it demonstrably causes harm, and where the evidence is genuinely unsettled — especially around mental health, which is the area most often oversold in both directions. It is general information for making informed choices, not medical or clinical advice.

Social media are websites and apps built around user-created content and interaction: you post, other people respond, and the platform decides who sees what. That last part — algorithmic distribution — is what separates a modern feed from a message board. Common examples include:

Facebook
Instagram
Pinterest
X (formerly Twitter)
LinkedIn
TikTok and YouTube

The pattern started long before the word existed. Dial-up bulletin board systems appeared in the late 1970s, Usenet newsgroups followed in 1980, and forums and chat rooms carried the idea through the 1990s. Those systems were text-first and small, but the core mechanic — strangers talking in public, asynchronously — was already there.

The mobile explosion

What changed in the 2000s and 2010s was not the concept but the distribution: cheap smartphones, always-on mobile data, and apps designed around a scrolling feed instead of a page you had to visit deliberately. Today's household names include Facebook, X, Instagram, and TikTok. Growth has slowed from its peak, but it has not stopped.

The numbers behind the headlines

Most arguments about social media collapse the moment someone checks the underlying figure. Here is what a few of the most-quoted claims actually rest on.

Common claim	What the source actually says	Where it comes from
"Over 6 billion people use social media"	About 5.79 billion social media user identities as of early April 2026 — accounts, not verified individuals	DataReportal global social media statistics
"Teens are all on social media"	Up to 95% of young people aged 13–17 report using a social media platform; about a third say they use it "almost constantly"	US Surgeon General's 2023 advisory
"Social media causes teen depression"	The advisory reports that adolescents using social media more than three hours a day faced double the risk of depression and anxiety symptoms — an association, and the advisory explicitly stops short of declaring the platforms safe or proven harmful	US Surgeon General's 2023 advisory
"Everyone gets bullied online"	46% of US teens aged 13–17 had experienced at least one of six cyberbullying behaviours; name-calling was most common at 32%	Pew Research Center, Teens and Cyberbullying 2022
"The 2023 Twitter breach hit 200 million users"	The Have I Been Pwned entry lists 211.5 million accounts, from data dated January 2021 that surfaced publicly in January 2023	Have I Been Pwned breach index

The benefits worth naming

Communication and reach

Social media collapses distance: a group chat with family on another continent, a voice note instead of a scheduled call, a fundraiser organised in minutes. For people in remote areas, or those whose mobility limits in-person contact, that access is not a convenience — it is the main channel.

Information speed and public opinion

News, eyewitness footage and expert commentary travel faster than any newsroom can publish. That speed shapes public opinion and drives civic action — and, as covered below, moves falsehoods just as efficiently.

Tutorials, webinars and subject-specific communities have made a great deal of practical instruction free, and people now learn about global issues from those directly affected rather than only through intermediaries.

Business and marketing opportunities

Online marketplaces and social selling let a small operation reach international customers at a fraction of traditional advertising cost. Brands use the same channels for marketing, customer service and direct sales. Common routes to revenue include:

SEO and ranking of websites for search engines.
Online store and e-commerce business.
Direct selling on platforms such as Pinterest, eBay and Etsy.
Audience-funded income on Facebook, Instagram, TikTok and YouTube through ads, sponsorship and subscriptions.

Social media work is now a normal job category: social media manager, content strategist, creator, paid-ads specialist, community manager and analyst. Companies keep hiring because organic reach has to be earned and paid reach has to be managed. Five skills carry real weight in those roles:

Brand awareness — keeping the brand visible and consistent across platforms.
Audience targeting — using platform insights to reach people most likely to care.
Content production — copy, images and video that suit each platform's format.
Market tracking — trends, social listening data and competitor moves.
Adaptability — new features and networks arrive faster than annual plans absorb them.

Creativity, expression and community

Anyone can publish — writing, video, art, music, memes — and people find others with the same rare interest or craft, building communities that outlast the platform they started on. For isolated or marginalised people in particular, that connection is the clearest documented upside.

This is where public discussion runs furthest ahead of the research, so it is worth being precise.

What the major reviews report

The US Surgeon General's 2023 advisory on social media and youth mental health gathered the strongest available evidence. It reports that up to 95% of 13–17-year-olds use a social media platform, that teenagers average around three and a half hours a day on these apps, that those using them more than three hours a day showed double the risk of depression and anxiety symptoms, and that 46% of 13–17-year-olds said social media makes them feel worse about their body image.

Why "social media causes depression" overstates it

The advisory's own conclusion is carefully worded: the evidence is not sufficient to conclude that social media is safe for children and adolescents. That is an honest statement about a gap in knowledge — and it is not the same as concluding the platforms cause harm. Several limitations sit behind that caution:

Association is not causation. Much of the research is cross-sectional: it measures use and mood at the same moment, so it cannot show which came first.
Reverse causation is plausible. Young people who are already anxious, isolated or depressed may retreat into their phones, rather than the phone producing the state.
Average effects are small. Across large datasets, the typical correlation between screen time and wellbeing is modest, and researchers analysing the same data have reached different conclusions depending on the controls they chose.
"Social media use" is one label for many behaviours. Messaging a close friend, watching video passively and being harassed in a comment section are not one exposure, but studies often measure them as if they were.

The defensible position is the uncomfortable one: heavy use is reliably associated with worse mental health outcomes in adolescents, the direction and size of any causal effect remain contested among researchers, and the absence of proof of harm is not proof of safety.

Who appears to be most affected

Effects are not evenly distributed. Pew's 2022 data found older teenage girls (15–17) reported the highest cyberbullying rate at 54%, against 46% of teens overall. Risk concentrates among young people already experiencing difficulties, and around specific content — appearance comparison, self-harm material, harassment — rather than around minutes of screen time in the abstract. If you are worried about yourself or a young person in your care, that is a conversation for a qualified clinician, not an article.

The real risks, stated accurately

Privacy, data exposure and doxxing

Oversharing location, granting excessive app permissions and reusing credentials all widen your exposure. Two distinct problems get confused here. A data breach is an attacker obtaining data from a service. Doxxing is publishing someone's private identifying information — home address, workplace, phone number — without consent, usually to invite harassment, and it is frequently assembled from what the target posted publicly rather than stolen at all.

Misinformation and fake news

Speed is the mechanism: a claim reaches millions before anyone checks it, and corrections travel less widely than the original. The consequences include eroded trust in institutions, panic during emergencies and lasting reputational damage. Before resharing, check whether an image is genuinely from the event it claims — a reverse image search will often show the same photo circulating years earlier under a different story.

Cyberbullying and harassment

Distance and pseudonymity lower the cost of cruelty. Pew's 2022 survey of US teens found 46% had experienced at least one of six cyberbullying behaviours: name-calling (32%), false rumours (22%), unsolicited explicit images (17%), persistent questioning about their location and activities (15%), physical threats (10%) and having explicit images of them shared without consent (7%). Some 28% had experienced more than one type. The emotional and psychological damage is well documented, and harassment that starts online frequently does not stay online.

Compulsive use and lost time

Infinite scroll, autoplay and notification schedules are engineered to remove natural stopping points, so an intended five minutes becomes forty — the pattern usually called doomscrolling. Note that "social media addiction" is not a formally recognised clinical diagnosis in the way substance use disorders are; problematic use is real and worth addressing, but the label is looser than it sounds.

Children, kidfluencers and exploitation

Young users face grooming attempts, age-inappropriate content and commercial pressure when family accounts monetise a child's image. The "kidfluencer" economy raises unresolved questions about consent, labour and who controls the money, and several jurisdictions have begun legislating on exactly that point.

Scams, phishing and account takeover

Fraud on social platforms is industrial: fake stores, impersonated support accounts, romance scams and phishing links that harvest logins. Two habits blunt most of it. Resolve any shortened link with a URL unshortener before you click, so you can see the real destination. Then check how old the site is — a "10-year-old brand" whose domain was registered three weeks ago is not a brand, and a domain age checker or a WHOIS lookup settles it in seconds.

Case studies worth knowing

Cambridge Analytica and the FTC settlement. On 24 July 2019 the US Federal Trade Commission imposed a $5 billion penalty on Facebook, which it described as the largest ever imposed for violating consumers' privacy. The FTC simultaneously brought separate actions against Cambridge Analytica, its former chief executive Alexander Nix, and app developer Aleksandr Kogan, alleging they used false and deceptive tactics to harvest personal information from millions of Facebook users.

The Twitter data incident. The Have I Been Pwned breach index lists a "Twitter (200M)" entry covering 211.5 million accounts, with the underlying data dated January 2021 and the dataset added to the index on 5 January 2023. It is usually described as a 2023 breach; it is more accurately a 2021 dataset that became public in 2023.

Blocking and age limits. Governments in numerous countries have restricted or blocked individual platforms, and several have moved to set minimum ages for social media accounts — a measure of how much political weight these platforms now carry.

Audit your privacy settings. Review who can see your posts, friends list and location on every platform, and repeat after major app updates — defaults change without notice.
Turn on two-factor authentication everywhere. Prefer an authenticator app or hardware key over SMS, which is vulnerable to SIM-swap attacks.
Use a unique password per platform. Generate them rather than inventing them — a password generator produces something you will never reuse by accident, and a password strength test tells you whether an existing one is worth keeping.
Check whether your credentials have already leaked. Run your current passwords through a pwned password checker; anything that appears in a known breach corpus should be changed today, on every account where you used it.
Verify before you share. Trace the original source, check the date, and reverse-search any image carrying the emotional weight of the claim.
Inspect links before clicking. Read the actual domain rather than the display text, and treat urgency as a warning sign rather than a reason to hurry.
Set boundaries you can keep. Fixed windows for checking feeds, notifications off for anything non-essential and phones out of the bedroom beat an aspirational daily limit. Notice how you feel after a session, not only how long it lasted.
Review your footprint periodically. Search your name, see what is publicly visible, and remove old posts exposing your address, workplace or routine.

If you would like a fuller treatment of the habits side of this, we have written separately about using social media without losing your sanity.

For parents and organisations

For children and teenagers, supervised access, age-appropriate settings and an ongoing conversation about what they encounter do more than a blanket ban, which tends to push use underground. Agree rules on sharing images, meeting people met online, and what happens if something goes wrong — the response to a bad experience matters more than the prohibition that failed to prevent it.

For organisations, publish a clear social media policy, train staff to recognise phishing and impersonation, restrict who holds account credentials, and rehearse what you will do when a post goes wrong. Reputation management is far cheaper before the incident than after it.

Frequently asked questions

DataReportal recorded roughly 5.79 billion social media user identities at the start of April 2026, growing by about 294 million year on year. Because those are accounts rather than verified individuals, the number of real users is lower — claims of "6 billion people" are not supported.

The honest answer is that the evidence shows a consistent association, not a proven cause. The US Surgeon General's 2023 advisory reports that adolescents using social media more than three hours a day faced double the risk of depression and anxiety symptoms, while concluding only that the platforms cannot be judged sufficiently safe for young people. Researchers continue to disagree about direction and effect size. If you are concerned about your own or a child's mental health, speak to a qualified professional.

Problematic, compulsive use is real and worth taking seriously, but it is not a formally recognised clinical disorder in the way substance use disorders are. Treat it as a habit-design problem with practical remedies — notification control, fixed check-in windows, friction between you and the app.

What is doxxing, exactly?

Doxxing is publishing someone's private identifying information — home address, workplace, phone number, family details — without their consent, typically to expose them to harassment. It is often assembled from information the target posted publicly over the years, which is why periodically auditing your own visible footprint is worth the hour.

Check the destination of the link rather than its label, look up how long the domain has existed, and be suspicious of urgency, pressure to move to a private channel, and payment by irreversible methods. A store claiming years of history on a domain registered last month is the clearest single signal you will get.

Neither verdict survives contact with the evidence. The properties that make these platforms useful — speed, reach, low barriers to publishing — are the same ones that make misinformation, harassment and fraud efficient. What varies is how they are designed, regulated and used.

The bottom line

Social media connects people across distances, supports trade and learning, and gives anyone a route to an audience. It also carries genuine risk: data exposure, harassment, fraud, and evidence that heavy adolescent use tracks with worse wellbeing even though the causal picture remains unresolved.

The useful response is neither panic nor dismissal. Know what the numbers actually say, secure your accounts, check what you share before you share it, and notice how the time you spend there leaves you feeling — the one variable in all of this that you fully control.

What social media is, and how it got this big

A short history of social media