Few studies on the validation of the Center for Epidemiological Studies Depression Scale (CES-D) in Chinese have been conducted in the samples from rural area, whereas the mental health problems among rural Chinese deserve more attention. For instance, the suicide rate in rural China is about 3 times that of the urban rate. This study aimed at assessing reliability and validity of the CES-D in 2 special adult samples in rural China, with data obtained by face-to-face interviews. One sample consisted of the proxy informants for subjects who committed suicide, and the other was of the proxy informants for living controls. For each person who committed suicide and each control, a family member and close friend served as informants. This study focused on informants themselves (not the persons who committed suicide or the living control himself/herself). Suicide informants (n = 781) who had experienced the suicide of a relative or close friend in the past 6 months were compared with the control informants (n = 832) who had no such experience. Internal reliability was satisfactory (Cronbach coefficients were .949 for suicide informants and .880 for control informants). Good concurrent validity and criterion validity were found by analyzing the relations to negative life events and the Scale for Suicide Ideation. Confirmatory factor analyses indicated that 3-factor structure (positive affect, interpersonal problems, depressive mood and somatic symptoms combined) had good fit in rural Chinese. These findings suggested that the CES-D had good reliability and validity when applied in these 2 adult samples in rural China. Factor structure analyses of the CES-D indicated that "somatization" process of expressing depressive symptoms still existed among rural Chinese. (C) 2012 Elsevier Inc. All rights reserved.