A total of 979 Chinese college students were recruited in order to examine the psychometric properties of the Chinese version of the South Oaks Gambling Screen (C-SOGS). The C-SOGS was internally consistent and correlated with gambling-related problems and negative mental health status. Results of likelihood ratios (LRs) and receiver operating characteristic (ROC) curve supported the C-SOGS as a valid screening instrument for probable pathological gambling. Using the DSM-IV criteria for pathological gambling, the conventional cut score of 5 on the C-SOGS showed satisfactory sensitivity, specificity and false negative rates. This cut score identified 85% of the true positives and 95% of the true negatives. In general, the C-SOGS overestimated the number of pathological gamblers relative to DSM-IV criteria. Future research on Chinese gambling should build on current empirical work on the SOGS to refine the C-SOGS for use with Chinese populations.