The literature on measures of family decision making preferences, and on the decision making roles patients and their family members prefer to take, is sparse. No work has been done to examine the variables which may affect these role preferences. The purposes of this research were to modify Degner et al.'s (1997b) Control Preferences Scale (CPS) for use with cancer patients and their family members, to describe patient and family member decision making preferences, and to identify differences in preferred decision making roles by selected variables. Using a descriptive correlational design, 61 cancer patient and family member pairs were interviewed twice, 24 hours apart. The CPS was modified to elicit decision making preferences of patients in relation to their family members. Psychometric testing demonstrated that the modified instrument, the Control Preferences Scale - Family (CPS-F), is both unidimensional and reliable. Using this instrument, patients chose highly active decisional roles in relation to their family member. Congruent with this, family members chose passive decisional roles. Family members, however, were prepared to engage in more active roles when they perceived their patient family member as deteriorating cognitively and/or physically. Although higher levels of education were found to be associated with more act decision making preferences for patients in this study, more research is necessary to determine the relationship between demographic variables and family decision making preferences. One of the most striking findings of this study was the movement toward more passive decision making roles by both patients and family members when symptom distress increased. Implications for nursing practice and research are presented, based on the study results.