“…We repeated these multivariate analyses among subgroups of persons with severe, moderately serious, and substance‐related disorders, respectively, to check the possibility that biased labeling may be contingent on the individual’s degree of impairment (not shown, available on request). Rushing (1978; Rushing and Esco, 1977) has argued that when disturbed individuals’ symptoms are severe and obvious, their social status will play little role in psychiatric professionals’ judgments, but when their symptoms are mild or moderate in seriousness, their relative status will sway professionals’ categorizations. Our analyses showed that among respondents with severe disorders, status variables had few, scattered, and unsystematic influences on hospitalization, consistent with Rushing’s argument as well as with Scheff’s (1964:452) proposition that serious, persistent, and visible deviant behaviors will result in labeling.…”