Objective To explore the reliability and validity of Chinese version of the Self-Evaluation of Negative Symptoms（ SNS）， so as to provide an effective and practical new tool for the assessment of negative symptoms in schizophrenia. Methods Totally 120 inpatients with schizophrenia meeting the diagnosis criteria of Diagnostic and Statistical Manual of Mental Disorders， Fourth Education（ DSM-Ⅳ） were recruited and tested with MATRICS Consensus Cognitive Battery（ MCCB）. 79 patients with relatively good cognitive function（ total score ≥ 40） were evaluated by SNS， Brief Negative Symptom Scale（ BNSS）， Positive and Negative Symptom Assessment Scale（ PANSS） and Calgary Depression Scale（ CDSS）， to evaluate the construct validity， relevance validity and internal consistency reliability of the scale. 30 inpatients of them were randomly selected and retested after 2 weeks. Results The exploratory factor analysis revealed a two-factor（ impairment of subjective experience and impairment of observable behavior） solution that accounted for 72.56% of the variance. Validity analysis showed that the total score of SNS was positively correlated with BNSS， PANSS negative symptoms and PANSS（ r=0.36，0.26，0.30，P ＜ 0.05）， and each dimension score was positively correlated with BNSS total score（ r=0.25-0.37， all P＜0.05）. The Cronbach's alpha for the SNS total score was 0.88 and for 5 subscales were between 0.58-0.76. The test-retest reliability coefficient for the SNS total score was 0.79（ P＜0.001） and for 5 subscales were between 0.34-0.82（ P＜ 0.01）. Conclusions The Chinese version of the SNS has good psychometric characteristics and can be used to quantitatively evaluate the negative symptoms of schizophrenia patients with certain cognitive function.