Objective: To analyze the status of childhood bullying bystander s behaviors among pupils and the correlation with family function,so as to provide empirical support for educational intervention in campus bullying. Methods: A questionnaire survey using the cluster random sampling method was administered to 1 145 pupils from a primary school in Wushan County within Chongqing City in November 2019, including Pupil Bullying Bystander Behavior Scale and Family Adaptability and Cohesion Evaluation Scale(FACES). The χ 2 test, Spearman related analysis and binary Logistic regression were used to determine the detection rate of different bullying bystander behaviors among pupils and the association with family function. Results: The proportion of bullying promoters, bullying protectors, and bystanders were 4.7%, 92.1%, and 42.1%, respectively. There were statistically significant differences in the three types of bullying bystander s behaviors among the different grades pupils( χ 2=21.45, 7.98, 27.48, P < 0.05 ). There were significant correlation between family function and its dimensions with different types of bullying bystander behavior ( | r |= 0.07-0.20, P <0.05). Binary Logistic regression analysis showed that pupils family function and grade had statistically significant impact on promoting bullying behavior, protecting victim behavior, and bystander behavior ( OR =0.98, 3.33; 1.02, 1.95 ; 0.99, 0.58, 0.41, 0.61, P <0.05). Conclusions The bullying bystanders behaviors of pupils are related to grade in school and family function. Schools should implement educational interventions for pupils of different grades and focus on improving pupils family function to effectively promote the occurrence of positive bystanders behaviors in pupils.