Objective To explore current situations and influencing factors of cycling violations among secondary school students in Beijing,in order to provide scientific basis for developing interventions of reducing secondary school students' cycling violations. Methods The study was conducted from March to May in 2014.An anonymous questionnaire survey was conducted among 32 810 secondary students selected by stratified cluster sampling method from 16 districts of Beijing.The two-step clustering method was used to classify the research objects with cycling violations into four types of risk level groups.Multivariate Logistic regression was used to analyze the correlation between four cycling risk levels and demographic characteristics and other unintentional injury behaviors. Results The cycling rate of secondary school students in Beijing was 80.0%,of which 37.2% reported having cycling violations.The reported rate of boys' violations was higher than that of girls(P<0.05),but there was no significant difference on the reporting rate of cycling violation between high school students and middle school students(P＝0.626).Also,the reporting rate of cycling violation in rural students was higher than that in urban students(P<0.05).The lower the mother's education level was,the higher the rate of student's violation would be.The reported rate of violations among students living in single-parent or reconstituted or intergenerational family was higher than students living in nuclear or extended family.Students who had other unintentional injury behaviors reported higher rate of cycling violations than those without such behaviors(P<0.05).Male,high school students,reconstituted family were more likely to have higher risk level of violations than girls,middle school students and students from other types of family.Rural students' reporting rates of various risk level behaviors were generally higher than those of urban students.The reporting rate of risk level-4 violation was the highest in students whose maternal education was high school/technical school/secondary school. The reporting rate of risk level-4 violation of students with other unintentional injury behaviors was higher than that of students without such behaviors(P<0.05).Logistic regression analysis showed that male(OR=2.75,95% CI:2.51-3.01),rural area(OR=1.41,95% CI:1.30-1.54),high maternal education,and reconstituted family(OR=1.51,95%CI:1.10-2.10),walking violation(OR=2.47,95% CI:2.19-2.78),swimming(OR=1.29,95% CI:1.12-1.49) and skating in non-safe places(OR=1.80,95% CI:1.57-2.07) were risk factors of high risk level of cycling. Conclusions Risk cycling behaviors among adolescents are closely associated with gender,school paragraph,dwelling place,maternal education,family type and other unintentional injury behaviors.When intervening on adolescent cycling violations,multi-level comprehensive intervention should be considered from five perspectives,including intrapersonal,interpersonal,institutional,community and public policy.