Workplace bullying is associated with substantial risks to mental and occupational health, making reliable measurement important for both research and workplace monitoring. A new Swedish study evaluates whether a single survey question can provide a useful measure of exposure to workplace bullying.
Large population studies often have limited space for extensive questionnaires, creating a need for brief measures that can still capture complex working conditions. In the Swedish Longitudinal Occupational Survey of Health, SLOSH, workplace bullying has traditionally been assessed using a single question about exposure to personal persecution through unkind words or behaviours from colleagues or supervisors. The study examined how well this question reflects established concepts and measures of workplace bullying.
The researchers used a mixed methods approach based on two SLOSH samples from 2008 and 2023. Qualitative descriptions of experiences of personal persecution were analysed alongside quantitative comparisons with established measures of workplace bullying. The researchers also examined whether the single item could distinguish bullying from other forms of workplace mistreatment and predict later mental health symptoms.
The single item showed moderate agreement with established workplace bullying measures and captured a broad range of negative behaviours, including personal devaluation, social exclusion and actions affecting work performance and professional standing. It was also able to distinguish workplace bullying from workplace violence to some extent.
Exposure identified through the measure was associated with substantially higher odds of subsequent symptoms of depression and anxiety. However, the question appeared to capture person related negative acts better than work related behaviours and may also include some less severe negative interactions. Overall, the findings support the measure as a useful brief indicator of workplace bullying, while suggesting that factors such as frequency and duration of exposure should also be considered.
Read more: 10.1007/s42380-026-00377-x