Objective  To evaluate reliability and validity of Olweus Bully/Victim Questionnaire (OBVQ) Chinese version for sibling bullying for the development of instrument for studies on sibling bullying among Chinese children and adolescents.

  Methods  We recruited 1 559 students aged 8 – 18 years in 6 primary and high schools in 2 cities of Hunan province using stratified multistage cluster sampling; then we conducted a survey on sibling bullying among the students with OBVQ Chinese version for sibling bullying and Juvenile Victimization Questionnaire (JVQ) between May and June 2018. The reliability and validity of OBVQ for sibling bullying were evaluated.

  Results  The correlation coefficients between 12 items and their domains in OBVQ for sibling bullying ranged from 0.425 to 0.697 (P < 0.001 for all), indicating a good item discrimination of the scale. The exploratory analysis on the scale resulted in a goodness of fit (GFI) of 0.911 and a root mean square error of approximation (RMSEA) of 0.103, demonstrating a good structural validity of the scale. The correlation coefficient for victimization domain between OBVQ (sibling version) and JVQ was 0.453 (P < 0.001). For the OBVQ (sibling version), the Cronbach′s coefficients alpha of the internal consistency reliability for the whole scale, victimization domain, and perpetration domain were 0.832, 0.764, and 0.777; those of the split-half reliability were 0.675, 0.729, and 0.761; and those of the test-retest reliability were 0.841, 0.867, and 0.891, respectively.

  Conclusion  OBVQ (sibling version) is of good reliability and validity and is suitable for surveys on sibling bullying among Chinese children and adolescents.