Aging encourages people to enhance their friend and family relationships. In general, the elderly tend to have more heterogeneity in relationships as they grow older. They depend on these relationships for instrumental, financial and emotional support. As a result, older adults who have many friends and have close ties with their families are more socially and psychologically well-adjusted than those who are alienated from their networks. Article: An assumption runs throughout the gerontological literature that having friends and active relationships with family members is better than not having them. Since the 1960s, when social gerontologists began debating the relative merits of the disengagement and activity theories, researchers have used the number of friends, the existence of active family relationships, and the amount of contact older adults have with these presumed intimates as indicators of aging well. Recently, however, researchers have recognized that not all personal relationships are good ones and not all social interactions affect older adults positively (e.g., Rook, 1984, 1989). In considering the connection between personal relationships and aging well, it is thus necessary to deconstruct friendship and kin relationships, to examine their myriad dimensions, and to recognize that merely having relationships is not an indication that someone is aging well. The tendency of researchers to assume that all personal relationships are positive is not the only limitation characteristic of the investigations on this topic. In both the family and friendship literatures, samples are often less than adequate, either being representative of very specific subpopulations of older adults or not being representative of any population at all (i.e., snowball, volunteer, or other nonprobability samples). Personal relationship researchers tend to study single respondents rather than pairs of friends or family members. When they do investigate dyads, they often study them in isolation rather than considering them in the context of the family or friendship network. Both literatures are also primarily descriptive rather than theoretically motivated, and, consequently, what is known about relationships is little more than a list of findings of all studies. Each of these personal relationship literatures has additional limitations. For example, researchers have rarely studied friendships longitudinally, making it impossible to examine changes in friendship as people age and to separate out age, period, and cohort effects. Gerontological friendship researchers usually study single-race (almost always Caucasian) or single-sex (usually female) populations; when they do include more than one race or both