Journal of Interpersonal Violence, Ahead of Print.
Self-reported measures of school bullying can be divided into two subtypes. Definition-based measures present a bullying definition followed by one question about being bullied and one question about bullying others, while behavior-based measures avoid using terms like “bully” and “bullying,” do not provide an explicit bullying definition, include items describing specific bullying behaviors, and respondents are asked to rate how often they have engaged in or have been a target of each behavior. The current study aimed to compare bullying perpetration and victimization prevalence rates between a definition-based scale and a behavior-based scale. The current study was part of a 4-year longitudinal research project, where students in Sweden completed an annual web-based survey at five waves starting with the school year of 2015 to 2016 (Wave 1; approximately age = 10.5 years) and ending in the school year of 2019 to 2020 (Wave 5; approximately age = 14.5 years). Because they responded to both measurement conditions, the study controlled for their possible individual differences. In this study, data from 1,469 to 1,715 students were analyzed. Findings revealed that the behavior-based scale displayed higher bullying perpetration and victimization prevalence than the definition-based scale. The behavior-based scales used in this study offer researchers and practitioners a self-report bullying measurement that includes power imbalance, concrete, and specific negative behaviors, and the ability to estimate repetition, but without using bullying terminology. Still, pros and cons of both approaches can be further discussed, and both definition-based and behavior-based self-report measures are vulnerable to a number of biases while they provide estimates or approximations—not exact pictures—of bullying prevalence.