Bullying — what works, what doesn't

Dan Olweus, working in Bergen from the 1970s onward, established the modern field of bullying research. He defined bullying as repeated negative actions over time, characterized by power imbalance and intent to harm — distinguishing it from one-off conflicts or fights between equals. His Norwegian national trial in the early 1980s of the Olweus Bullying Prevention Program produced reductions in bullying of 50% in the original cohort, a result that has been partially replicated but rarely matched. The program's distinctive features were its whole-school approach, sustained duration, adult supervision of hot spots, clear rules, and specific incident protocols. The American adaptations have generally produced smaller effects than the original Norwegian studies, in part because schools implement only some components, in part because cultural context matters.

Three features distinguish bullying from other negative peer interactions: repetition (it happens more than once, often patterned), power imbalance (the target cannot easily defend themselves due to size, social status, numbers, or other factors), and intent (the behavior is meant to harm, not accidental). The definition matters because programs designed for bullying don't work well for other types of conflict, and vice versa. A one-time fight between two kids of roughly equal status is not bullying; it requires different intervention. Ongoing harassment of a single child by a group, or by one socially powerful child, is bullying and requires the specific tools the bullying literature has developed.

Bullying is a social ecology phenomenon, not an individual one. It occurs in the relationship between the kid doing the bullying, the targeted kid, the bystanders, the broader peer group, the adult supervisory structure, and the institutional culture. Trying to fix it by intervening only with the bully misses that bullying behavior is often reinforced by social rewards from peers — laughter, attention, status. Trying to fix it by intervening only with the target puts the burden on the wrong person and ignores the dynamics maintaining the behavior. Whole-school approaches change the entire system: the rules, the supervision, the bystander norms, the consequences, the language adults use, the curriculum, the parent engagement. Meta-analyses by Ttofi and Farrington and others consistently find that the programs with the most components implemented for the longest duration produce the largest effects.

In most bullying incidents, only a small share of students participate as either bully or target. The majority watch. Their behavior — whether they laugh, walk away, intervene, get an adult, support the target later — shapes whether the bullying behavior is reinforced or extinguished. Bystander training, which gives this majority of students a clear repertoire of actions they can take without putting themselves at risk, is one of the higher-leverage components of effective programs. The training emphasizes that bystanders don't need to confront the bully directly to help — they can refuse to laugh, walk away with the target, report to an adult, support the target privately afterward. Multiple small actions by many bystanders change norms more than dramatic intervention by a few.

Susan Swearer and colleagues have argued that bullying must be understood as occurring within nested contexts — individual, family, peer group, school, community, culture. Interventions that ignore the nesting fail. A child who is bullying may be replicating dynamics from home, responding to peer-group hierarchy, or expressing distress they cannot otherwise communicate. A child being bullied may be vulnerable for reasons rooted in family circumstance, identity, neurodivergence, or social skills. The targets of bullying are not random; they cluster around kids who are different in ways the local peer group has decided to penalize. Effective intervention pays attention to all the layers, not just the immediate incident.

Zero-tolerance policies — automatic severe punishment for any bullying incident — emerged in the 1990s and 2000s and have been studied substantially. They suppress reporting because students don't want peers permanently labeled or expelled over incidents that could have been addressed less severely. They produce racial disparities in enforcement. They remove discretion from teachers and administrators who often know the context. And they don't reduce bullying. The American Psychological Association reviewed the evidence in 2008 and concluded that zero-tolerance had failed to deliver on its stated goals. Most states have moved away from it in formal policy, but practice often lags. Effective programs use graduated, context-sensitive responses tied to behavior change, not blanket severe sanctions.

Peer mediation, in which a trained student mediator helps two parties work through a conflict, has evidence for ordinary peer disputes between equals. It is the wrong tool for bullying. Bullying involves power imbalance and intentional harm; mediation assumes a misunderstanding to be resolved. Putting the targeted child in a room with the bully under the framing of "let's work this out" tends to retraumatize the target, gives the bully a platform, and signals to both that the institution misunderstands the dynamic. Schools that use peer mediation for bullying are doing harm with good intentions. The category distinction is important.

Practical sequence: validate the child's account without immediately demanding solutions ("I believe you, this is real, we are going to handle this"); document specifics in writing — dates, locations, what was said or done, who was present, any digital evidence; review the school's anti-bullying policy and identify the specific provisions that apply; request a meeting with the appropriate administrator with the documentation and the policy citation in hand; ask for specific actions and a specific timeline for follow-up; escalate to the district if the school does not respond. Avoid: telling the child to "just ignore it," telling them to "stand up to the bully" without giving them tools that work for power imbalances, contacting the other family directly (this usually escalates), and confronting the bully personally. Schools have authority and protocols. Insist on their use.

The harder conversation for many parents. The first step is hearing the report without immediate defense. Most parents' instinct is to disbelieve, minimize, or counterattack. This protects no one. Better: ask what happened, listen to the child's account, hold them accountable for the behavior, work on understanding what is driving it. Children who bully often have their own distress — being bullied themselves elsewhere, social anxiety expressed as preemptive aggression, learned behavior from family, lack of empathy skills. The intervention works on the behavior and the drivers simultaneously. Public shame backfires. Calm, sustained accountability paired with sustained relationship is what changes behavior. The goal is not punishment; it is behavior change with support.

Bullying targets are not random. LGBTQ+ youth, especially trans youth, experience bullying at much higher rates than peers. Students with disabilities, particularly visible disabilities or neurodivergence, are targeted disproportionately. Racial, religious, and ethnic minorities in predominantly homogeneous environments face elevated risk. Body size, family socioeconomic status, and perceived difference of any kind raise risk. Effective programs explicitly address identity-based bullying rather than treating all incidents as generic. Pretending bullying is identity-neutral leaves the most vulnerable students least protected.

Longitudinal studies — including work by Olweus following his original cohorts decades later — show measurable adult costs of childhood bullying experience. Targets show elevated rates of depression, anxiety, suicidal behavior, lower self-esteem, more difficulty in adult relationships, worse health outcomes. Children who persistently bully others without intervention show higher rates of antisocial behavior, criminal involvement, substance use, and poor adult relationships. Children who are both bullied and bully others — the bully-victims — have the worst outcomes of any group. The notion that bullying is character-building is empirically wrong. The damage compounds over decades for some kids and is not erased by adulthood.

A short list. Does the school have a specific anti-bullying policy that names bullying, identifies reporting procedures, and specifies response protocols? Is there a designated person responsible for receiving reports and following up? Is there staff training on bullying prevention, and how often? Are unstructured spaces (cafeterias, hallways, playgrounds, buses) actively supervised? Is there bystander education built into the curriculum? Is there an evidence-based program in use, and what is its evaluation? How does the school handle cyberbullying that involves its students? What is the protocol when a student is identified as repeatedly bullying others? If the answers are vague or absent, the school is not running an effective program, regardless of what the website says.

Bullying prevention research is mature. The Olweus tradition, the Finnish KiVa program, the Steps to Respect program, and others have strong evidence bases. Implementation is the rate-limiting step. Schools cycle through one-off campaigns, posters, and assemblies because they are cheap and visible. Sustained multi-year program implementation with fidelity is expensive and unflashy and works. Where parent communities organize to demand actual programs with actual evaluation, change happens. Where they don't, schools default to performance over substance. This is not a knowledge problem at this point. It is a will problem, and the lever for changing it is collective parent insistence informed by what the evidence actually says works.

Olweus, Dan. Bullying at School: What We Know and What We Can Do. Oxford: Blackwell, 1993.

Olweus, Dan. "School Bullying: Development and Some Important Challenges." Annual Review of Clinical Psychology 9 (2013): 751–780.

Ttofi, Maria M., and David P. Farrington. "Effectiveness of School-Based Programs to Reduce Bullying: A Systematic and Meta-Analytic Review." Journal of Experimental Criminology 7, no. 1 (2011): 27–56.

Swearer, Susan M., and Shelley Hymel. "Understanding the Psychology of Bullying: Moving Toward a Social-Ecological Diathesis-Stress Model." American Psychologist 70, no. 4 (2015): 344–353.

Espelage, Dorothy L., and Susan M. Swearer, eds. Bullying in North American Schools. 2nd ed. New York: Routledge, 2010.

Salmivalli, Christina, Antti Kärnä, and Elisa Poskiparta. "Counteracting Bullying in Finland: The KiVa Program and Its Effects on Different Forms of Being Bullied." International Journal of Behavioral Development 35, no. 5 (2011): 405–411.

American Psychological Association Zero Tolerance Task Force. "Are Zero Tolerance Policies Effective in the Schools? An Evidentiary Review and Recommendations." American Psychologist 63, no. 9 (2008): 852–862.

Hinduja, Sameer, and Justin W. Patchin. Bullying Beyond the Schoolyard: Preventing and Responding to Cyberbullying. 2nd ed. Thousand Oaks, CA: Corwin, 2015.

Copeland, William E., Dieter Wolke, Adrian Angold, and E. Jane Costello. "Adult Psychiatric Outcomes of Bullying and Being Bullied by Peers in Childhood and Adolescence." JAMA Psychiatry 70, no. 4 (2013): 419–426.

Kowalski, Robin M., Gary W. Giumetti, Amber N. Schroeder, and Micah R. Lattanner. "Bullying in the Digital Age: A Critical Review and Meta-Analysis of Cyberbullying Research Among Youth." Psychological Bulletin 140, no. 4 (2014): 1073–1137.

Limber, Susan P., Dan Olweus, Weijun Wang, Marlene Masiello, and Jane Riese. "Evaluation of the Olweus Bullying Prevention Program: A Large Scale Study of U.S. Students in Grades 3–11." Journal of School Psychology 69 (2018): 56–72.

Bradshaw, Catherine P. "Translating Research to Practice in Bullying Prevention." American Psychologist 70, no. 4 (2015): 322–332.