Principal Investigator: Dan Olweus

Level of Intervention: Universal

Target Population: Elementary and middle schools students, teachers and parents

References: Olweus (1991, 1993, 1994)

Theory (Risk & Protective Factors Targeted):
The Bullying Intervention program is based on a child-rearing model applied in a school setting to reduce low level aggression and conflict. The program targets aggressive behavior, and favorable school, community, and family attitudes toward aggression.

Description of Intervention:
The program was implemented as part of a national campaign against bullying in Norway from 1983 to 1985. It is a school-focused anti-bullying initiative based on awareness and cognitive skill-building. Components of the intervention included a 32-page informational booklet on bullying, bullies and victims provided to all schools, a folder of information and recommendations about children involved as bullies or victims provided to parents, a video of vignettes about bullying available to be shown in classrooms, and a school questionnaire to assess the level of bully/victim problems school-wide and serve as a catalyst for school-wide discussion.

Research Subjects:
The study examined 2500 students in grades 4-7 (the equivalent of grades 5-8 in U.S. schools) in 42 Norwegian schools. The sample was assigned to 4 age-equivalent cohorts of 600-700 each. The sample was demographically, ethnically, and socioeconomically diverse. Students ranged in age from 11 to 14, with approximately equal number of boys and girls.

Research Design:
Because the study was conducted as part of a nationwide campaign, a randomized trial was not possible. Instead, a quasi-experimental staggered cohort design was used in which four adjacent cohorts of students were followed over a two and a half year period. Data were collected at three time periods (pretest, 8 months, and 20 months), such that some cohorts served both as intervention and control (baseline) groups, in different comparisons. Program effectiveness was assessed using an extended version of the Olweus Bully/Victim Questionnaire which provided both self-report bully/vitimization data and classroom-aggregate peer reports of the level of bullying, as well as a 23-item self report questionnaire about antisocial behavior (Olweus, 1989).

Significant reductions in bullying, aggressive and antisocial behavior were found at 8 and 20 months based both on student self reports and peer reports, including a 50 % reduction in the percentage of students who reported being bullied or bullying others. The author also reported an increase in satisfaction with school life (another possible indicator of schoolwide change in levels of aggression). Generally, the changes were equally substantial for boys and girls. Students reported significant improvements with respect to the climate of order and discipline in the classroom, more positive social relationships, and a more positive attitude toward schoolwork and school.

Strengths & Limitations:
The Intervention Campaign Against Bully/Victim Problems demonstrated substantial reductions in bullying and low-level aggression through a relatively simple intervention which focused primarily on educating schools, students, and families about the problem of bullying. The reported effects of the intervention were found across teacher, student self report and peer report measures, and can be expected to be generalizable given the very large and diverse population of the study. Potential bias due to attrition was investigated and was not indicated by the data analysis. Analysis of program fidelity at the classroom level demonstrated a clear dosage-response relationship. Classrooms that implemented the three core components of the intervention (including the establishment of classroom rules against bullying) to a greater degree than other classes experienced a greater decrease in bully/victim problems. This too lends to the argument that the effects were indeed a result of the intervention.

Unfortunately, the published (English language) literature on the program does not provide detail on statistical significance or effect sizes for the findings reported, nor does the study examine the role of each of the program components individually in producing the desirable effects. Large scale, independent replications of the program have been conducted in England, Germany, and the United States. The two foreign replications generally supported the findings reported here, although the studies both suffered from methodological flaws (Whitney, et. al, 1994; Hanewinkel & Knaack, 1997). The United States replication produced ambiguous results, finding a significant reduction on self-reported bullying, but no self-reported reduction in being bullied (Melton, et. al, 1998).

back to text     back to table of contents     external links to program     contact investigator