The purpose of this study was to survey the hospital staff's perception of hospice and death and thereby, suggest the ways to help them have a mature attitude towards and a better understanding of death through an effective education on hospice. For this purpose, this study was designed to provide some data useful for the hospital staff not experienced in facing the dying patients to handle the desperate situation skillfully and engage themselves more effectively in their hospice services. For this study, the researcher conducted a questionnaire survey for the staff of 'C' hospital in Seoul about their hospice philosophy, attitudes towards hospice and perception of death for the period from February, 2006 to March, 2006. A total of 751 subjects responded effectively to the survey. The data collected were analyzed using the SF55 11.0 for ANOVA and T-test in order to test the relationships among subjects' perception of death, their demographic variables, their health condition, their hospice philosophy formed from their experiences of patients' death and hospice services and their attitudes towards hospice. Besides, the correlations among their hospice philosophy, perception of death and attitudes towards hospice were tested. The results of this study can be summarized as follows; First, as a result of testing the relationships among subjects' experiences of health consulting. their experiences of patients' death and hospice and their hospice philosophy, it was confirmed that their experiences of health and death consulting and their experiences of having been educated were relevant. Second, it was found that such variables as health condition, death, experience of hospice and attitude towards hospice were not significantly correlated with each other. Third, as a consequence of testing the relationships among health condition, death, experience of hospice and perception of death, it was disclosed that only the physical health condition was significant. Fourth, it was confirmed that subjects' hospice philosophy, perception of death and attitudes towards hospice were at the usual level on average. Fifth, hospice philosophy, perception of death and attitudes towards hospice were found correlated significantly with each other. Sixth, as a result of the stepwise variable adjustment for such variables as hospice philosophy, perception of death, attitudes towards hospice, it was found that the adjusted r-square value was 0.347 when departments, experience of having been requested by dying patient for consulting, experience of having been educated on hospice, religion and marital status were set as independent variables. And the estimated value of each variable was found significant. Seventh, as a consequence of conducting the multiple regression analysis by setting 'religion' as significant independent variable, it was found that the estimated value of physical health condition was not significant statistically. This, as a result of the simple regression analysis for 'religion' only, its explanatory power was found .197, while its adjusted r-square value was 0.20. Eight, it was found that subjects' attitude towards hospice was significantly correlated with such variables of experience of patient's or relative's death, experience of having been requested for consulting about death, gender, marital status and departments. As a result of the multiple regression analysis and the subsequent stepwise adjustment for this variable, it was found that only 'experience of having been requested for consulting' had some explanatory power: its adjusted r-square value was 0.089. As discussed above, this study tested the correlations among various variables including hospital staff's attitude towards hospice and perception of death and thereby, provided for the data useful for their education on hospice. This study may be significant in that it proved that it would be essential to educate hospital staff on hospice for more effective care of hospice patients and their family members at hospital.