The aim of this study was to examine the diagnostic validity of four commonly used assessment scales for depression in schizophrenia. The study population consisted of 84 inpatients meeting the DSM-IV criteria for schizophrenia. Depression in the study subjects was defined by the DSM-IV criteria for major depressive episode. The Positive and Negative Syndrome Scale (PANSS) and the Simpson-Angus Rating Scale (SARS) were used to differentiate depression from the negative and extrapyramidal symptom-related depressive phenomena in schizophrenia. The following four depression scales were assessed for their diagnostic validity as measures of depressive disorder in schizophrenia: the Calgary Depression Scale for Schizophrenia (CDSS), the Beck Depression Inventory (BDI), the Hamilton Rating Scale for Depression (HAM-D), and the depression subscale of the PANSS (PANSS-D). Of 84 patients with schizophrenia, 32 were diagnosed as having comorbid depressive disorder. The areas under the Receiver Operating Characteristic (ROC) curves of the CDSS, HAM-D, PANSS-D, and BDI were 0.94, 0.89, 0.90, and 0.81, respectively. The area under the ROC curve of the CDSS was significantly greater than that of the BDI and tended to be more favorable than those for the HAM-D and the PANSS-D. Our study suggests that the CDSS may provide the best assessment for depression in patients with schizophrenia.