According to the CDC, no single group is categorically exempt from bullying, but the students at the lowest risk are those with strong peer support and secure social standing—often students who are not perceived as different in appearance, identity, or social skill, and who report high levels of school connectedness. The CDC's surveillance work identifies sexual and gender minority youth, students with disabilities, and socially isolated youth as bearing the highest risk. The lowest-risk students are, in practice, well-connected peers who fit dominant social norms and have at least one trusted adult at school.
That answer sounds simple, but the mechanism behind it is not. Bullying risk is not a list of protected labels—it is the product of visibility, power imbalance, and social context. Children who differ in ways peers notice, or who lack allies to buffer aggression, get targeted more often. Children who blend in and have backup get targeted less. This explainer unpacks why that happens, what the CDC's data actually show and do not show, and where the common "who is not at risk" assumption breaks down.
The mechanism: why visibility and power predict targeting
Bullying is not random aggression. The CDC defines it around three features: unwanted aggressive behavior, a real or perceived power imbalance, and repetition or high likelihood of repetition (CDC, 2024). Each element points to the same mechanism. Bullies select targets who are less able to retaliate, less able to avoid the situation, or less likely to be defended by bystanders.
Visibility is the first filter. A child who stands out—by body size, clothing, disability status, accent, gender expression, or social awkwardness—is easier to mark. That does not make difference the cause of bullying. It makes difference a perceptual cue that aggressive peers use when choosing targets. The cause sits in the aggressor and the social environment, not in the victim.
The second filter is power. Power here means social leverage rather than physical strength. A student without friends, without status, and without an adult ally has less social capital to spend on defense. A well-connected student can mobilize peers, and the threat of social consequences raises the aggressor's cost. This is why school connectedness shows up repeatedly as a protective factor in CDC data.
The third filter is repetition. Bullying requires a pattern or a credible threat of one. Isolated conflict between equals is not bullying under CDC criteria. This distinction matters clinically because it changes which interventions help. Conflict resolution fits mutual disputes; bullying requires adult intervention aimed at the aggressor and systemic supports for the target.
Plain version: kids get bullied when they are easy to notice and hard to defend, and the safest kids are the ones who are neither conspicuous nor alone.
Quantified context: what CDC surveillance shows
CDC surveillance is the reference standard here, and its findings are consistent across years. In the CDC's Youth Risk Behavior Survey data, roughly 1 in 5 high school students reported being bullied on school property in the past 12 months, with cyberbullying somewhat lower. For middle school, other CDC-supported surveys put in-school bullying higher, often in the range of 1 in 4 to 1 in 3, because middle grades concentrate the transition years when peer hierarchies re-form.
Risk gradients are steep for identity and belonging factors. Sexual and gender minority students report bullying victimization at rates well above their heterosexual and cisgender peers, with some data indicating nearly double the risk for certain groups. Students with disabilities, including those in special education settings, are disproportionately represented in victimization data. Students who report low school connectedness and few close friends show elevated rates as well.
The inverse is just as informative. Students who report high family support, at least one supportive adult at school, and participation in structured activities show lower victimization rates. That protective pattern does not mean these students never get bullied. It means their odds are meaningfully lower, and their recovery is better.
Rates are also stable in character over time. Bullying victimization peaks in elementary and middle school, declines across high school, and changes form rather than disappearing—physical bullying falls, relational and cyber forms persist. Any claim that one group is "safe" needs to account for this age gradient.
Edge cases: where the lowest-risk assumption fails
The question assumes a clean ranking. Real data are messier, and four edge cases matter.
Popular students face relational aggression rather than exclusion. High-status kids rarely get pushed in hallways, but they are targets for rumor campaigns and social sabotage, which are more common among peers who compete for the same status. Students who move between schools, grades, or friend groups have a temporary buffering gap. A popular student who transfers can become isolated in a single semester. Students with protective traits in one domain may be exposed in another. A confident athlete may be targeted for a disability, an accent, or a family characteristic they cannot hide. Bullying risk is not fixed. Classroom composition, teacher turnover, and school climate changes can shift a low-risk student into a higher-risk position within a year.
The professional caveat is blunt: CDC categories are population descriptors, not individual predictions. A student in a "low-risk" demographic can still be victimized severely, and a student in a high-risk group may not be. No CDC report provides a group that is exempt, and reading the data as an exemption list misuses it.
Misconceptions: what "not highest risk" does not mean
The most common misconception is that low risk means no risk. CDC surveillance groups describe averages. Any individual student's experience depends on their classroom, their friends, and their school's response. Averages hide the tails.
A second misconception is that bullying is about victim characteristics at all. Research consistently finds that perpetrator characteristics and school climate predict bullying behavior better than victim traits do. Framing risk around what makes a child "different" can slide into victim-blaming. The CDC's own guidance emphasizes prevention through school-wide climate, supervision, and bystander training.
A third misconception is that protective factors are permanent. School connectedness is measured at a point in time. It can be built or lost. A student who reports strong connectedness in the fall can report none in the spring after a friend group dissolves.
A fourth misconception is that cyberbullying follows the same risk map as in-person bullying. Digital environments change visibility rules. Students who are not targeted at school can be targeted online by students from other schools, and the repetition threshold is easier to cross because content persists. The overlap is substantial but not identical.
Comparison of risk contexts
| Context | Typical victimization pattern | Strongest protective factor | Key limitation of the data |
|---|---|---|---|
| General high school population | About 1 in 5 report in-school bullying | At least one supportive adult at school | Self-report and recall bias |
| Sexual and gender minority students | Higher rates than peers, often near double | Inclusive school policies and staff training | Group sizes vary by district |
| Students with disabilities | Disproportionate victimization, especially in segregated settings | Integrated settings with trained staff | Disability definitions differ across surveys |
| Socially isolated students | Elevated risk tied to low peer support | Structured group activities and mentoring | Isolation is hard to measure |
| High-status popular students | Lower physical risk, higher relational risk | Diverse friend networks | Relational bullying is underreported |
| Students with high school connectedness | Lowest reported victimization | Family support plus school belonging | Ceiling effects in survey items |
The table makes the core point: the "not highest risk" group is defined by protective conditions, not by identity labels. Connection is the lever.
Practical actions and thresholds
For parents and educators, the actionable question is not "which group is safe" but "does this child have buffers." Use concrete thresholds to guide attention.
Treat two or more bullying incidents involving the same child within a school year as a pattern requiring a formal response, not a one-off conversation. Check for at least one named supportive adult at school. A child who cannot name one is in a higher-risk position regardless of popularity. Screen for school connectedness at least once per year in middle school, where risk peaks. Watch for sudden changes in friend groups, lunch seating, or online activity, which often precede reported victimization. If a child reports victimization, document dates and observers, and request the school's anti-bullying policy in writing. For cyberbullying, preserve screenshots with timestamps before reporting.
Timeline of what happens when bullying is unaddressed
| Time frame | Typical trajectory | What changes it |
|---|---|---|
| First incident | Conflict or targeted aggression, often unreported | Bystander intervention, adult notice |
| 1 to 4 weeks | Repetition establishes pattern; target disengages | School response, parental report |
| 1 to 3 months | Avoidance behaviors, absenteeism, anxiety symptoms | Counseling, safety planning, monitoring |
| 1 school year | Chronic victimization linked to depression and school disengagement | Sustained climate intervention |
| 1 to 3 years | Persistent effects on mental health and academic trajectory | Early and consistent intervention |
| Long term | Elevated risk of anxiety, depression, and lower school attainment | Protective relationships and timely support |
Plain version: the first month matters most, because repetition is what turns conflict into bullying and what makes the mental health effects stick.
Limitations of the CDC framing
CDC data have real boundaries that readers should hold in mind. The surveys rely on self-report, so underreporting is likely, especially for relational and cyber forms. Question wording and recall windows differ across instruments, which makes direct comparisons across years imprecise. Some groups, including sexual and gender minority students and students with disabilities, are measured inconsistently across state and local surveys.
The data also describe associations, not causes. Lower victimization among connected students does not prove connectedness caused the difference, though the mechanism is plausible and consistent with prevention research. And no CDC dataset can identify a group as exempt, because the lowest-risk category in any table still contains victimized students.
The most defensible reading is this. According to the CDC, the group not at the highest risk of being bullied is the group with strong social support and low visibility as a target—well-connected students with at least one trusted adult and stable peer relationships. That is a description of protective conditions, not a safe harbor. The practical takeaway is to build those conditions for every child, because the students who most need them are often the ones least likely to be counted in the low-risk column.
This article is general educational information about bullying risk and prevention, not a diagnosis, treatment plan, or legal opinion. It does not promise any outcome for any individual child. If a child is in immediate danger or at risk of self-harm, contact local emergency services or a crisis line right away, and consult a qualified local professional for individual guidance.