Studies about nurses' stressors and social support behaviors are limited. This study explored differences between Intensive Care Units (ICUs) and wards in regard to Jordanian nurses' job stressors and social support behaviors as well as predictors of the two concepts. A quantitative research design using a survey method was used. The Nursing Stress Scale (NSS) (Gray-Toft & Anderson 1981) and the Inventory of Social Supportive Behaviors (ISSB) (Barrera, Sandler & Ramsay 1981) were used to collect data from a convenience sample of 228 nurses who were working in 12 ICUs and 235 nurses who were working in nine wards of 13 hospitals, with a total response rate of 66.2%. Stressors in ICUs were higher than those in wards. The ICUs scored higher than wards in 'conflict with physicians' subscale of NSS. The ICUs scored higher than wards in 'emotional support' and 'tangible assistance' subscales of ISSB. Shift worked, model of nursing care, and level of education predicted nurses' job stressors in ICUs and wards. 'Model of nursing care' was a shared predictor of social support behaviors in ICUs and wards. High job stressors and low social support behaviors were evidenced in Jordan. Job stressors were higher in ICUs than those in wards, thus more social support behaviors should be provided to nurses in ICUs. Nurses' stressors should be assessed and managed. In all settings in general and in ICUs in particular, nurse managers should use various social support behaviors to buffer the influence of job stressors on nurses.