The purpose of the present paper was to examine the reliability and validity of the Japanese version of the Social Adjustment Scale-Self Report (SAS-SR) and to present its normative data. The SAS-SR was administered to a random sample of all the employees of a large general hospital, together with the General Health Questionnaire (n = 363). It was also administered to a representative subset of first-visit patients at 33 psychiatric hospitals and clinics from all over Japan, along with the semistructured psychiatric interview to ascertain the patients' diagnoses (n = 1581). For the internal consistency reliability of the subscales and the overall scale of the SAS-SR, Cronbach's alpha was between 0.61 and 0.73. The Pearson product-moment correlations between the subscale and overall scale scores with the GHQ score were mostly >0.3. The scores were statistically significantly and substantively different between the normal sample and the patient samples, and were also meaningful, differentiating between various diagnostic subgroups. The reference ranges of the SAS-SR scores for mentally healthy subjects were calculated as 95% prediction intervals; for example, 1.22-2.22 for the overall score. The Japanese version of the SAS-SR has good reliability and satisfactory validity. The present study provided reference ranges for its scores in order to increase their interpretability. With its ease of administration and its rich subscales, the scale promises to offer a psychometrically sound measure with which to assess social adjustment in people with various psychiatric disorders.