Cyberbullying Statistics UK: 2026 Facts, Data & Key Insights
Cyberbullying statistics UK
Understanding who reports online bullying, how it can affect them and whom they tell—with the scope of every figure kept clear.
Evidence checked 16 October 2026.
Ages, definitions and survey populations differ. England and Wales findings are not a combined UK total.

experienced an online bullying behaviour in the previous year
ONS, children aged 10–15 in England and Wales, year ending March 2023.told a parent or guardian about their experience
Among children who experienced online bullying behaviour in the same ONS survey.did not tell anyone
Among children who experienced online bullying behaviour. This is not a percentage of all children.How many children reported online bullying?
ONS estimated that 19.1% of children aged 10–15 in England and Wales experienced at least one of the specified online bullying behaviours in the previous year. That represented an estimated 847,000 children in the year ending March 2023.
This is a dated population estimate for that age range and geography. It is not a UK-wide total, a count of every online incident, or the proportion of children using a particular platform.
The estimate was 22.5% for girls and 16.0% for boys. These are percentages within each group, rather than the proportion of all victims who were girls or boys.
Population prevalence
A percentage of all children in a specified age range answers how common the measured experience was in that survey population.
Conditional percentages
A question asked only of children who experienced bullying has a smaller, selected base. It cannot be presented as a percentage of all children.
Online harms and habits
Messaging habits, harmful-content reports and contact with support services answer different questions. They do not provide interchangeable cyberbullying rates.
Online bullying reported by girls and boys
What proportion of each group experienced at least one specified behaviour?
| Group / measure | percent |
|---|---|
| Girls | 22.5% |
| Boys | 16% |
| All children aged 10–15 | 19.1% |
Population percentages within each group, not the composition of victims. Covers experiences in the previous year. The overall estimate is shown separately in the data table. This is not a combined UK estimate. Lower survey response affected estimate quality. Numerical confidence intervals are not supplied.
Comparisons over time and between surveys
ONS found no statistically significant difference in the overall online-bullying estimate between the years ending March 2020 and March 2023. That does not mean every type of online harm remained unchanged or that no child experienced an increase in harm.
Survey comparisons need a shared definition, age range, geography, recall period and population. A percentage of all children and a percentage of pupils who already reported bullying are not interchangeable.
Changes in communication platforms or access to devices cannot, by themselves, explain a difference in reported bullying. The survey does not establish which platform or personal characteristic caused an individual child’s experience.
Survey coverage and quality
The ONS estimates describe children aged 10–15 living in private households in England and Wales. They are weighted survey estimates, not a complete count of every child’s experience.
ONS cautions that lower response rates after face-to-face interviewing resumed affected the quality of these estimates. National Statistics status was temporarily suspended for the child estimates. This limitation should remain visible when the figures are reused.
The selected workbook tables do not supply numerical confidence intervals. The source does identify a statistically significant girl–boy difference, but that association does not explain why it occurs.
What children said happened online.
The ONS questions describe specific behaviours, rather than relying only on whether a child uses the word “cyberbullying”. In the year ending March 2023, 10.9% reported receiving a nasty message online.
A separate 9.6% reported name-calling, swearing or insults online. That combined category is broader than name-calling alone.
A child may have experienced several of the listed behaviours. The percentages therefore cannot be added to estimate the number of children affected, and the list is not a division of victims into mutually exclusive groups.
The online behaviours children reported
Which specific experiences did children say they had in the previous year?
| Group / measure | percent |
|---|---|
| Nasty messages sent to the child | 10.9% |
| Name-calling, swearing or insults | 9.6% |
| Rumours spread online | 4.5% |
| Deliberate exclusion | 4% |
| Nasty messages posted or passed around | 3.8% |
| Online impersonation | 1.6% |
| Demanded money or possessions | 1.4% |
| Deliberate threats | 1.4% |
Specific behaviours can overlap; do not add them to estimate victims. These percentages describe all children asked, not only children who experienced bullying. Exact valid-answer bases are retained where supplied; similar bases are not treated as identical. Lower survey response affected estimate quality. Numerical confidence intervals are not supplied.
A behaviour category is not an explanation of motive
The survey identifies reported experiences; it does not establish the intention behind every incident or explain why a particular child was targeted.
Bullying may concern appearance, disability, sexuality, race or other characteristics. A question asking bullied pupils why they think it happened has a different denominator from a question about how many pupils with that characteristic experienced bullying.
Similarly, a result from a voluntary charity survey, a gaming community or people contacting a support service needs its own recruitment method and age range. It should not be substituted for a national population estimate.
Emotional effects vary between children.
Bullying can affect confidence, relationships and emotional wellbeing. Online contact may continue beyond school hours, making it harder for someone to get a break.
Children’s experiences differ. The next chart describes reported emotional impact among those who experienced online bullying behaviour; it is not a clinical diagnosis or a measure of every child’s mental health.
Among those answering the emotional-impact question, 22.7% said it affected them a lot, 49.1% a little and 28.2% not at all. The valid-answer sample was 245.
The source does not identify a statistically significant difference between the corresponding online and in-person impact percentages. The chart therefore shows the online responses without presenting them as evidence that one setting is always worse.
Reported emotional impact of online bullying
How much did children say the experience affected them emotionally?
| Group / measure | percent |
|---|---|
| A lot | 22.7% |
| A little | 49.1% |
| Not at all | 28.2% |
Conditional on experiencing online bullying behaviour. These responses are not clinical diagnoses or percentages of all children. They do not establish that online bullying is always more harmful than in-person bullying. Lower survey response affected estimate quality. Numerical confidence intervals are not supplied.
Online bullying should not automatically be described as more harmful than in-person bullying. The circumstances, relationships and individual’s experience matter. Group averages do not establish the seriousness of a particular child’s situation.
Young people contacting support services sometimes describe anxiety, distress, self-harm or suicidal thoughts in connection with bullying. Service users are a selected group, so their reports cannot establish national rates or show that bullying alone caused someone’s difficulties.
A child’s account deserves attention even when it does not match the most common survey response. Statistics can describe patterns, but they cannot decide how much support an individual needs.
Our Mental Health First Aid courses help education and community professionals identify and respond to the mental health impact of cyberbullying in young people.
Whom did children tell?
In the ONS survey, 56.3% of children who experienced online bullying behaviour told a parent or guardian. 18.1% did not tell anyone.
Those figures use the group who experienced online bullying as their base. They do not mean 18.1% of all children experienced undisclosed cyberbullying, nor identify why a child chose not to tell someone.
The next chart describes who children said they told. Where more than one person could be named, the responses overlap and should not be added.
The valid-answer base for this question was 244, distinct from the emotional-impact base. Selected responses also included friends or mutual friends (47.7%), a teacher (35.0%), other family members (19.7%) and another school staff member (16.0%).
Whom children told about online bullying
Who did children say they told about what happened?
| Group / measure | percent |
|---|---|
| Parent or guardian | 56.3% |
| Friends or mutual friends | 47.7% |
| Teacher | 35% |
| Other family members | 19.7% |
| Another school staff member | 16% |
| No one | 18.1% |
Selected disclosure responses, including telling nobody. Children could name more than one person, so support routes overlap. The base is children who experienced online bullying, not all children. Lower survey response affected estimate quality. Numerical confidence intervals are not supplied.
Making it easier to ask for help matters whether the person responsible is a school peer, another acquaintance or a stranger. A young person can seek support from a trusted adult, and harmful behaviour can be reported through the relevant platform and school safeguarding arrangements.
Private messages may be less visible to adults than public posts. That does not mean every platform lacks reporting or safety tools, or that a child should be expected to manage harmful behaviour alone.
School evidence and digital safety.
DfE’s school surveys ask different questions from the ONS household survey. They should not be presented as interchangeable estimates of national cyberbullying prevalence.
In May 2025, 21% of surveyed pupils in Years 7–13 in England reported bullying in the previous year. Among that bullied group, 29% said some bullying happened online and 87% said some happened in person.
Those settings overlap: the same pupil could report both. The 29% is not the proportion of all pupils who were cyberbullied, and it does not identify who was responsible.
More recent evidence asks about digital-safety knowledge.
DfE’s March 2026 wave, collected from 11 March to 7 April 2026 and published on 25 June 2026, asked pupils in Years 7–11 who went online which actions they knew how to take on at least one platform, app or site.
85.8% said they knew how to block people on social media, while 79.4% knew how to report or flag inappropriate content. The chart shows all five selected actions and their published 95% confidence intervals, with a valid unweighted base of 2,318 pupils.
The newer May 2026 wave did not repeat these specific items. March is therefore the latest matching question verified in the current collection, even though other surveys or collection updates have later dates.
Knowing how to use online safety tools
Which actions did pupils say they knew how to take?
| Group / measure | percent | 95% interval |
|---|---|---|
| Block people on social media | 85.8% | 83.4–87.9% |
| Block people in online games | 79.5% | 76.8–82% |
| Report or flag inappropriate content | 79.4% | 76.7–81.9% |
| Limit who can view a profile | 75.5% | 72.5–78.3% |
| Use other safety features | 68.8% | 65.7–71.8% |
Self-reported knowledge on at least one platform, app or site, not actual use, effectiveness or absence of harm. Pupils could know several actions; do not add the percentages. Whiskers show the source’s published 95% confidence intervals.
Knowing how to use a safety tool is different from using it, and using it is different from whether it resolves a problem. A survey question about digital skills or awareness cannot establish that a young person is safe from bullying.
Keep communication habits and abuse statistics distinct.
Online bullying can occur in public posts, private messages, group chats and games. Ofcom’s 2026 research examines how children communicate online, but communication habits, bullying and grooming remain separate measures.
A platform’s popularity does not tell us the proportion of its users who experience bullying. Likewise, a finding about private messaging should not be used to rank platforms by harm without a suitable measure and denominator.
A newer Ofcom measure uses a different definition
Ofcom’s report published on 21 May 2026 provides a newer, differently defined UK measure. Among 8–17-year-olds who opted to answer its question, 20% reported nasty or hurtful behaviour through an online or digital channel in the previous year.
The question was answered by 1,112 of the 1,695 children asked. Fieldwork ran from November 2025 to February 2026. The digital category includes phone calls, so it is broader than strictly internet bullying.
The optional question, age range, definition and survey design differ from ONS. This figure is therefore kept separate: it is not evidence of a change from the 19.1% ONS estimate or a replacement for the dedicated ONS series.
Ofcom’s newer October 2026 gaming publication draws together existing research. It does not supply a new comparable child-bullying prevalence wave.
AI-generated abuse is a distinct safeguarding concern.
The Internet Watch Foundation’s 2025 report documents AI-generated child sexual abuse imagery. Such imagery can cause harm even when fabricated, but IWF’s detected-content and report counts are not a survey of UK children or a national measure of cyberbullying prevalence.
Image-based abuse can overlap with bullying while also raising separate safeguarding and legal issues. The existence of AI-generated material does not provide a valid multiplier for the ONS bullying estimate.
Law depends on the conduct, image and jurisdiction.
Laws covering indecent images of children can apply to computer-generated images. Separately, offences in England and Wales covering the creation or requesting of purported intimate images of adults without consent commenced on 6 February 2026 under the Data (Use and Access) Act 2025. These offences include statutory requirements about consent and reasonable belief in consent.
A Bill alone does not create an offence, and different conduct may engage different laws. This adult-image offence should not be described as the only legal protection relevant to a child experiencing online abuse.
Reading cyberbullying statistics carefully.
How common is cyberbullying in the UK?
There is no single figure here covering every age and all four UK nations. ONS estimated 19.1% for children aged 10–15 in England and Wales in the year ending March 2023. Other surveys use different groups and questions.
Why use 2023 ONS figures in a 2026 review?
They remain the latest dedicated ONS bullying and online-experiences dataset verified at this review. The redesigned children’s survey published in September 2026 does not yet replace that bullying module.
Does 29% mean almost three in ten pupils were cyberbullied?
Not in the cited May 2025 DfE result. Its 29% applies only to pupils in Years 7–13 who reported bullying, not all pupils. Online and in-person settings overlap.
Can the reported behaviours be added together?
No. A child may report more than one behaviour. Adding those percentages would count some children more than once.
Does online bullying always cause more harm than in-person bullying?
No. These surveys do not justify that general claim. Harm depends on the circumstances and the child’s experience, and group-level statistics cannot determine individual support needs.
Do AI-abuse reports show how common cyberbullying is?
No. Reports of detected abusive content and representative survey estimates have different units and populations. IWF’s findings document a safeguarding concern, not a national cyberbullying prevalence rate.
Methods, uncertainty and reuse
This guide presents published evidence visually; it is not a new survey. The age range, geography, dates and question base are kept with each dataset.
For conditional questions, the denominator is the group asked that question, such as children who experienced online bullying. It is not automatically all children or the whole survey sample.
Published confidence intervals and item bases are included where supplied. A reference base described as “similar” is not copied across rows as an exact denominator. Missing precision is not invented.
All charts start at zero. Keep the wording, period, population and caveats when reusing them, and do not add overlapping categories or interpret an association as proof of causation.
Cyberbullying charts.
With the context that matters.
Download SVG and PNG charts, the underlying CSV data and methodology. Keep the age group, survey question, geography, period and limitations with every figure.
Mental Health First Aid Course (2026). Cyberbullying Statistics UK. Evidence checked 16 October 2026. https://www.mentalhealthfirstaidcourse.co.uk/blog/cyberbullying-statistics-uk
Sources
Primary evidence and methodology. Checked 16 October 2026.
- ONS: bullying and online experiences, year ending March 2023
- ONS: detailed child-bullying tables
- ONS: redesigned children’s survey and planned bullying module
- DfE: March 2026 online safety knowledge
- DfE: current survey collection and publication history
- DfE: May 2025 bullying settings
- DfE: preventing and tackling bullying
- DfE: advice for parents and carers on cyberbullying
- IWF: Annual Data and Insights Report 2025
- IWF: AI-generated child sexual abuse material, 2025 observations
- CPS: indecent and prohibited images of children
- CPS: communications and intimate-image offences
- Ministry of Justice: commencement of adult intimate-image offences
- Mental Health Innovations: dated bullying research and service context
- Ofcom: Children and Parents Media Use and Attitudes Report 2026
- Ofcom: Children and Parents Media Literacy Tracker 2025/26 technical report
- Ofcom: statistical release calendar 2026
- Ofcom: Growing Up Gaming summary announcement
Written by Mental Health Experts. This guide was produced by the team at Mental Health First Aid Course. We publish evidence-based data resources to help schools, parents, and community professionals understand the scale of cyberbullying and its mental health consequences.