Objective To test the reliability and validity of the Chinese version Psychiatric Anger Rating Scale（ PARS）. Methods With the consent of the original author， English version PARS was translated into Chinese. PARS contains 41 items. A total of 125 psychosis from Shanghai Mental Health Center were collected to evaluate with PARS. Cronbach's α was applied to measure internal consistency reliability， Guttman Split-Half to measure Split-Half reliability. The correlation coefficient between each item and total score was calculated. Exploratory factor was used to analyze and test the construct validity of PARS. Pearson correlation analysis was used to explore associated validity between PARS and Positive and Negative Symptom Scale （PANSS）. Results Cronbach's α was 0.89. Guttman Split-Half was 0.88. Correlation coefficient between each item and total score was from 0.19 to 0.72（ P＜0.05）. Exploratory factor analysis yielded 9 factors， accounting for 67% of the total variance. There was a significant correlation between PARS and item-G14 of the PANSS（ lack of anger control）（ r=0.54，P＜0.05）. Conclusions Chinese version PARS has a good reliability and validity， which can be used to study anger symptoms in Chinese psychotics.