A multivariate analysis of the contribution of environmental and personality components to privacy satisfaction in a sample of mental hospital residents



Journal Title

Journal ISSN

Volume Title



This investigation examined the relationships among selected setting variables in a large psychiatric institution, a personality attribute regarding privacy needs and privacy satisfaction of psychiatric inpatients. It was hypothesized that a patient's privacy satisfactions would be significantly related to these sets of variables. In addition, possible relationships between social and demographic variables and privacy satisfaction were explored. One hundred and twenty recently admitted male and female psychiatric inpatients, residing in New Hampshire Hospital, Concord, New Hampshire, were interviewed over a period of three months. Patients resided in four residential buildings which, because of architectural differences, were assumed to show a variable influence on privacy attainment and consequently privacy satisfactions of residents. In order to control for possible differential patient management practices by ward staff, each of the admission wards in the target residential buildings of interest was evaluated prior to commencement of the main investigation. Instruments used were the Residential Management Survey and the Characteristics of the Treatment Environment. The selected wards included in the study were statistically comparable in terms of treatment orientation and patient management practices. A subject's privacy needs were assessed by the Privacy Needs Scale (PNS), a twenty-item instrument sampling situations and settings reflecting diverse privacy behaviors and privacy situations. This instrument represents a modified and abbreviated form of the PPS originally developed by Marshall (1970). Other measures assessed the degree of spatial movement (Home-range measures), duration of residency on a locked ward and an objective description of the hospital environment in terms of privacy opportunities (Privacy Questionnaire). Reliability measures were obtained by interviewing ward staff on several parallel instruments. A subject's privacy satisfaction during hospitalization, the predicted variable, was measured by a 39-item questionnaire, the Privacy Rating Form. This instrument inquired about issues related to actual or perceived privacy opportunities at the hospital. Dimensions of interest reflected satisfaction with control overt movement, space, solitude and interpersonal disclosure. A total of 30 predictor variables were correlated with the subjects' privacy satisfaction ratings. Predictor variables with a correlation at or below the .05 level were included in a multiple regression analysis. The results show a highly significant correlation between positive descriptions of the hospital environment and privacy satisfaction. Hospital residential buildings accounted for a total of four percent of unique variance in the predicted variable. A discriminant analysis failed to predict above chance levels single or combined residential buildings from a subject's privacy satisfaction score. Variables measuring the restrictiveness of the hospital experience (duration of residency on a locked ward, Home-range measures) did not contribute significantly in understanding privacy satisfaction ratings. A subject's privacy needs (PNS) were significantly and inversely related to privacy satisfaction ratings and accounted for about four percent of unexplained variance. A subject's age resulted in positive and significant correlations but failed to account significantly for privacy satisfaction in the multiple regression analysis The largest and most significant amount of variance (about 27 percent) was accounted for by the subject's legal status (ccxnmitment-type). Voluntary patients obtained significantly larger privacy satisfaction ratings, reflecting a more positive perception of the hospital environment than for involuntary patients. This finding remained unchanged when duration of residency on a locked ward was statistically controlled. The unexpected and small contribution of hospital setting variables and the subjects' privacy needs suggests that the hospital setting variables and personality variables singularly or combined are inadequate measures of privacy satisfactions. These results were discussed within the framework of Kelvin's (1974) theory linking privacy to interpersonal power. Subsequent discussions related these findings to the consistency-inconsistency issue in personality research. Finally, practical implications of these results on patient management practices in psychiatric hospitals and therapeutic issues were discussed.