Our analysis confirms the findings of previous studies that considerable variation exists in patient perceptions of coercion. That we find such variance even among our sample members who were all legally involuntarily hospitalized supports the findings of previous studies that objective legal status and subjective feelings of coercion are not equivalent. Not surprisingly, the modal score indicated high levels of perceived coercion; but at the same time, an almost equal number of patients perceived little or no coercion. Our distribution on this variable is very similar to that reported by Gardner et al. (1993) whose sample had a majority of legally voluntary patients, a majority without a diagnosis of psychosis, and a much larger proportion with a primary diagnosis of substance abuse. Both samples were distributed across the full range of the perceived coercion scale with marked bimodality, although the proportion in our sample experiencing high coercion was almost 50% larger, and the proportion experiencing low coercion was approximately 50% smaller. As with perceived coercion, considerable variation existed across the full range of the negative pressures and process exclusion scales, indicating that the process of civil commitment does not necessarily involve the stereotypical picture of a resisting mental patient being dragged into a mental institution. Approximately two-fifths of our sample reported little or no negative pressures and little or no process exclusion in their hospital admission. These findings suggest that involuntary admission to a mental hospital can permit patients to feel like they have voice and validation, and can avoid force even in the absence of choice. The challenge is to try to extend to all patients at the time of their admission a demonstration in word and action that they are person with opinions, desires, rights, and dignity, and not just mental patients in an acute crisis. Clinical variables could not account for differences in outcome variables. Neither type of disorder, secondary substance abuse, personality disorder, hostility/suspiciousness, severity of symptoms nor number of recent hospital admissions affected how patients viewed the hospital admission process. This may suggest that some element of the shared characteristic of chronicity or recidivism negates expected differences among clinical groups. Patients who were young, urban, unmarried, and of low education were no more likely to perceive coercion, negative pressures, or process exclusion. This finding was unexpected because they are more likely to resist hospitalization, more likely to be seen as dangerous and in need of force to be hospitalized, and more likely to have had previous negative experiences with authority. Instead, it was white, female, unmarried respondents with more education who were more likely to perceive higher levels of our dependent variables. Such higher status persons have more resources, autonomy, and control in their daily work and family roles. They are the ones more likely to perceive coercion, negative pressures, and process exclusion in hospital admission, possibly because their expectation levels and reference groups make them more aware of the use of coercion and any deprivation in autonomy. The MacArthur study has found that patients who have little voice or validation, and against whom force and threats are used to get them hospitalized, perceive high levels of coercion; that is, procedural inequity and negative pressures predict perceived coercion (Lidz et al., 1995). Our sample, with different measures of these three constructs, found strong positive correlations among the three. When Negative Pressures and Process Exclusion were added to the model predicting perceived coercion, they were highly significant and wiped out the associations of perceived coercion with gender, race, cohabitation, and education. Although those basic and acquired demographic variables predict Negative Pressures and Process Exclusion, they do not predict perceived coercion directly. Rather, as the MacArthur group found, the two procedural variables, using force with patients and including patients in the process of hospital admission, are the most important factors in perceiving coercion.