The study explored the validity of the Social Skills Improvement System‐Rating Scales (SSIS‐RS) in a sample of over 4,000 preschool children (aged 3–5 years old) in China. Multiple approaches were used to examine the validity of SSIS‐RS in the Chinese cultural context, including confirmatory factor analysis, convergent and discriminant validity evidence with external measures of learning and developmental outcomes, and internal consistency reliability estimates. The findings supported the SSIS‐RS 7‐factor structure (i.e., communication, cooperation, assertion, engagement, responsibility, empathy, and self‐control) when applied in the Chinese context, and provided strong empirical evidence that SSIS‐RS is a psychometrically sound measure for assessing the social skills of Chinese children. The latent profile analysis was used to provide a descriptive picture of the patterns of different social skills measured by SSIS‐RS among Chinese preschool children. Four classes best described the social skills presentations of children in this study: Lowest Social Skills (14%), Relatively Low Social Skills (41%), Relatively High Social skills (35%), and Highest Social Skills (10%). Implications for intervention practice and future research are discussed.