The objective of the study is to provide a reliable estimate of the pooled prevalence of social anxiety disorder (SAD) and social anxiety symptoms (SAS) among children, adolescents, and young adults (CAYA) in China. Meta-analysis is used to provide pooled-prevalence rate of SAD and SAS. Literature searches were conducted in both English and Chinese databases from the database's inception to April 2019. Eleven studies were identified for SAD, and 17 were included for SAS. The results revealed a pooled prevalence of SAD of 2.1% (95% CI: 1.2–3.8%) with high between-studies heterogeneity (Q = 1,055.2, I2 = 99.1%, p < 0.001). The pooled prevalence estimate of SAS was 23.5% (95% CI: 18.6–29.3%), also with significant heterogeneity (Q = 1,019.3, I2 = 98.4%, p < 0.001). Different diagnostic tools or self-report scales reported significant different prevalence of SAD or SAS. Further analysis stratified by gender, age, sampling methods, economic status, and risk of bias were performed. Limitations include the high level of heterogeneity between studies, inadequate number of the studies, and significant differences in prevalence caused by measurements.