Demography: the past 30 years, the present, and the future.

In a population, 30 years approximates the mean length of a generation or the time in which one generation replaces the previous one. In most areas of demographic research, one generation of research replaces the next in a considerably shorter period. A glance at what demographers were doing 30 years ago shows that in most areas, past research no more resembles what demographers do now than the Friedan calculators of that period resemble our current desktop computing environments. In these areas, current research differs from the research of 30 years ago in the theoretical approach, in the methods employed, and in the type of data used. Where great change has occurred, theory, methods, and data appear to have evolved together: each has changed in response to changes in the other areas and then has demanded further developments in those areas. Formal demography is one area that has been characterized by continuity. The questions addressed and the basic techniques employed build on a long heritage, even while steady progress is made in the development of methodology and analytic techniques. Analysis of the 1980s reintroduced the idea of context--the idea that behavior is influenced not only by the characteristics of individuals but also of the environment in which the behavior takes place. In the 1990s, this approach will be incorporated more thoroughly into theoretical developments in all areas of demography as we attempt increasingly to make comparisons across cohorts and time while continuing to model individual behavior. Mortality is the theoretically underdeveloped area of demography which probably will show the greatest change in the next 30 years.