人口分析の環(<特集>小林和正教授退官記念号 : 東南アジアの人口)

Abstract

How to disaggregate demographic variables is a basic question in studying determinants and consequences of population trends, which are the central subject of demography. In studying determinants of population trends, attention is usually directed to how the vital or migration rates concerned are affected by what non-demographic variables. The usual way of dealing with vital events or migration in formal demography is to assume that such events occur 'suddenly' with a certain probability of occurrence. Any event, be it birth, death, or migration, however, is produced in reality after the completion of a certain set of steps. For example, a live birth can occur at the completion of a necessary set of intrauterine developmental steps beginning with conception. Conception may be further traced back to pre-pregnancy sexual behavior since marriage or the last birth. Variables involved in such a context are called 'intermediate' variables in demographic literature. The study of intermediate variables has recently advanced most in the field of fertility. A similar approach should now be applied to mortality, nuptiality, and migration research. What is important is to isolate intermediate variables involved and to devise effective ways to manipulate them quantitatively for study purposes. Interdisciplinary interests may be drawn more to the socio-economic consequences of population trends. From the demographic standpoint, however, what is essential is to investigate the nondemographic variables involved in the linkage between one demographic change and the next.