This study identified correlations between perceived family support and hopelessness in patients admitted to Neuro-surgical Intensive Care Units. The purpose was to enhance theoretical understanding of the relationships of these two variables. The subjects of this study were 51 patients admitted to N-ICU, at three general hospitals in Seoul. Data were collected by researcher in structured interviews from Aug. 12 to Oct. 13, 1992. The research tools were parts of the Moos Family Environment Scale and the Beck hopelessness Scale. The general characteristic data were analyzed for frequency and percentage ; the hypothesis was tested by the pearson product Moment Correlation coefficient. After normality tests by using Kolmogorov-Sminorvtest, and T-test, ANOVA, and Mann-Whitney U test, Kruskal-Wallis test were used on the Family support and the Hopelessness about general characteristics. The results of the above analysis were as follows; 1) The average family support score for the group was 63.61 (tool average 51) and item average was 3.74(tool item average 3) ; the family support score of this sample was higher than average. The average family cohesion score of family support was 35.25 (tool average 27) and item average was 3.91 (tool item average 3). The average family expression score of family support was 28.35 (tool average 24) and item average was 3.57 (tool average 3). In this sample, perceived family expression was lower than family cohesion. 2) The average hopelessness score was 45.99 (tool average 60) and item average was 2.29 9tool item average 3) ; the hopelessness score of this sample was low in comparison to the average. 3) The hypothesis in this study was supported. The main hypothesis that the higher the perceived family support level, the lower the level of the hopelessness, was supported(r=-.3688 p=.004). The sub-hypothesis that the higher the perceived family expression level, the lower the level of hopelessness, was supported(r=-.3068 p=0.014). 4) General characteristics of the objects related to family support were 'economic status'(p=.025) and 'helping person'(p=.044); the higher the economic status, the greater the family support. When the patient identified the helping person as a spouse, family support was rated more highly. The only general characteristic related to family cohesion was 'helping person'(p=.041). No general characteristics were related to family expression. 5) The one general characteristic related to hopelessness was 'education'(p=.002) ; the higher their education, the lower their hopelessness. For these ICU patients, were related perceived family support and hopelessness, and family expression level was low in comparison to family cohesion level. The perceived family support of these seriously ill patients in situational crisis may have influenced the patient's emotional reaction of hopelessness. This study concluded that nurses in the ICU confirm the family support of the patient, and involve the family as the most intimate support systems in the care of the patient to help reduce the patient's hopelessness.