Most rating scales for affective disorders measure either depressive or hypomanic/manic symptoms and there are few scales for hypomania/mania in a self-rating format. We wanted to develop and validate a self-rating scale for comprehensive assessment of depressive, manic/hypomanic and mixed affective states. We developed an 18-item self-rating scale starting with the DSM-IV criteria for depression and mania, with subscales for depression and mania. The scale was evaluated on 61 patients with a diagnosis of affective disorder, predominantly bipolar disorder type I, using Montgomery-Asberg Depression Rating Scale (MADRS), Hypomania Interview Guide-Clinical version (HIGH-C) and Clinical Global Impression scale, modified for bipolar patients (CGI-BP) as reference scales. Internal consistency of the scale measured by Cronbach's alpha was 0.89 for the depression subscale and 0.91 for the mania subscale. Spearman's correlation coefficients (two-tailed) between the depression subscale and MADRS was 0.74 (P<0.01) and between mania subscale and HIGH-C 0.80 (P<0.01). A rotated factor analysis of the scale supported the separation of symptoms in the mania and depression subscale. We established that the self-rating scales sensitivity to identify mixed states, with combined cut-offs on the MADRS and HIGH-C as reference, was 0.90 with a specificity of 0.71. The study shows that the Affective Self Rating Scale is highly correlated with ratings of established interview scales for depression and mania and that it may aid the detection of mixed affective states.