Most conceptions of power refer to the relative ability of persons to achieve individual ends in interpersonal contexts in which their ends conflict. Within groups, such as the family, which are characterized by functionally disuse and ongoing relations, there are serious methodological problems with determining actors' ends. This has led to the use of a variety of substitute procedures that allow one to talk of power without assessing the particular ends of specific persons. This paper critically examines these substitute procedures and suggests their inadequacy for the study of families and similar small groups. It is suggested that an alternative to power must be found in order to proceed with the work for which that concept has been used. The sketch of one alternative is presented and its implications discussed.