The effects of expectancy and professional identity upon attributions of mental illness

According to Mechanic (1967), the layman has usually assumed that the psychiatric professional is the unbiased expert on who is, or is not, mentally ill. In addition, the psychiatric professional has also usually assumed that his labeling decisions are based upon an unbiased scientific examination of the available information. However, Mechanic, and other sociological observers of the mental illness labeling process, have recently questioned whether these assumptions are legitimate. In fact, these observers have found that the professional’s labeling decisions are apparently heavily biased in favor of mental illness. Mechanic (1967), recounting his experience at two state institutions, reported that once a person is presented for institutionalization, seldom if ever does the psychiatrist dis- agree, although the psychiatrist’s “reasons” for hospitalization may be based on different premises than the public’s. Scheff (1964a, 1964b) found that when psychiatrists work within a court commitment setting, they tended to presume the existence of mental disorder. In a detailed examination of actual court procedures and psychiatrists’ pre- commitment patient interviews, his evidence indicated that the psychiatrists’ investigations appeared biased and presumed the existence of disorder. The psychiatrists frequently chose arbitrary evidence upon which to base their decisions- -evidence that involved elaborate psychiatric straining and interpretation. Often they decided to commit the person in question even where no evidence could be found. Their offhand remarks suggested pre- judgment of the case, and the interviews were conducted with such marked haste and carelessness that they only lasted on the average about ten minutes. As one of the psychiatrist’s states: The petition cases are pretty automatic. If the patient’s own family wants to get rid of him you know there is something wrong (Scheff, 1964a, p. 410). Coffman (1961) has noted that psychiatrists usually seek only that information that is consistent with their opinions through a process he has called “discrediting”. Within the mental hospital with which he was associated, the patient’s records and history were carefully searched to provide justification for the patient’s label, while any evidence of health was usually ignored. On the other hand, Gove (1970) has disputed the veracity of these observations. He has presented evidence that psychiatric professionals do not routinely label people as mentally ill, and do not commit every person they examine to a mental institution. Gove has also questioned the scientific adequacy of the methods employed by Mechanic and Scheff. Unfortunately, Gove’s rebuttal was based upon a review of the literature, and most, if not all, of the studies in his review were not originally directed toward the question of possible professional bias. In the writer’s opinion, an adequate test of the legitimacy, or illegitimacy, of Mechanic’ (1967), Scheff’s (1964a, 1964b), or Goffman’s (1961) observations has not yet been made. If professional bias does exist in the mental illness labeling process, then two primary factors might account for Mechanic’s, Scheff’s, and Goffman’s observations. First, an individual’s work setting may be associated with various situational expectancies that might influence one’s diagnostic judgment. For example, the situational expectancies of the psychiatric professional, operating within his typical work setting, might be biased in favor of the probable appearance of a mentally ill person. If one expects to interview a mentally ill person, then one’s diagnostic judgment of that person’s mental status might be affected. Second, the psychiatric professional’s training and experience might foster an increased inclination to view people as mentally ill--a greater inclination than the non-professional. If the professional is less inclined to tolerate deviance than the non-professional, then the professional’s diagnostic judgments should be more likely to lead to the mental illness label. The primary goal of this dissertation will be to examine these two factors--situational expectancies and tolerance of deviance--and their role in the mental illness labeling process. Bias in the application of the mental illness label may be traceable to the effects of these two factors.This thesis was digitized as part of a project begun in 2014 to increase the number of Duke psychology theses available online. The digitization project was spearheaded by Ciara Healy