Objective: To evaluate the efficacy of ultraviolet light in influenza B prevention and control in schools in Yantian District of Shenzhen in winter, and to provide a reference for preventing and controlling influenza B in schools. Methods: Multistage stratified cluster sampling was conducted. Ultraviolet irradiance was detected in 98 classes of 11 schools. The irradiance of different distances under each ultraviolet lamp were measured. Correlational analysis was conducted with influenza cases reported by Shenzhen Student Health Surveillance System. Results: A total of 473 ultraviolet lamps were detected, the qualified rate was 53.91%. The effective distance from the lamp corresponding power supply to efficient irradiance was (1.43±0.54)m. The total effective irradiance volume per unit time of each class was(13.65±32.66)m3. Totally 599 influenza cases in 74 classes were reported, with incidence rate of 13.76%. Fifty-eight classes reported influenza clustering, accounting for 59.18%. Average duration of the epidemic was(14.5±25.25)d. Spearman correlation analysis showed that the proportion of irradiance volume per unit time in classroom volume was negatively correlated with class influenza incidence(r=-0.32, P=0.00), but not significant with duration of influenza epidemic(P=0.78). Covariance analysis showed that the incidence of influenza and the duration of influenza epidemic in classes was not significantly correlated with the proportion of irradiance volume per class per unit time in classroom volume (P>0.05). Conclusion Ultraviolet light might not effectively reduce the incidence rate of influenza B and the duration of influenza epidemic in schools in Yantian District of Shenzhen over 2017-2018 winter.