The influence of Socio-biological Factors upon the Birth Rate in Japan.

Abstract

If the birth rate of various prefectures in Japan be compared, we see that it is not uniform all over the country. In the northeastern part the rate tends to be high, whereas in the western and the urban districts it is comparatively low.
The object of the present investigation is to elucidate statistically the relationship between birth rate and socio-biological factors by geographical districts. For this purpose the method of partial correlation is appLed. The factors taken into consideration and their notations are as follows:
p=Population
b=Birth rate= Number of live births/ Population ×1000
d=Density of population=log (population/ Area of land in km²)
u=Urbanization= Population in cities/ Total population ×100
s=Per cent of unmarried women
Number of unmarried women aged 15-44/ Total population of women aged 15-44 ×100
w=Average income= Total amount of income (¥) / Population
a=Average age of females at marriage
c=Age composition of population =Population of 20-44 years of age/ Totat population ×100
m=Infantile motality= Number of deaths under one year of age/ Number of births ×100
h= Number of marriages recorded/ Number of women aged 15 and over ×1000
e=Prevalence of female educatien
=Number of girls attending high schools/ Population of unmarried girls aged 15-19 ×100
The data wilized are those of the year 1925, from the national census carried out in this year.
1. Density of population (d) and birth rate (b).
Pearl reported that there exists a negative net correlation between birth rate and density of population in 132 cities in the United States. I also found that 97 cities in Japan show a significan negative correlation between d and b,
Not only cities but also prefectures show a similar negative correlation,
These results indicate that more densely populated districts generally have a lower birth rate. But here arises a question: D'oes density of population itself exert such a repressive influence on birth rate? Or is not the correlation rather of secondary nature due to some other factors? In order to solve question, the coefficients of partial correlation were computed, and it was found that among the various factors studied, the average age of females at marriage (a) and the average income per capita (w) are the most important factors which bring b and d into close relationship, as the following ccfficients show,
These coefficicients are obviously insignificant, and we see that the previous significant negative correlation of zero-order between b and d is reduced to insignificance by eliminating the influence of w and a. Such a result is obtained because the average income (w) andthe marriage age (a) stand, as will be discussed later, in very intimate relationship with the birth rate (b) on one side and with density of population (d) on the other side, as the following oefficients prove,
From this we may conclude that it is a fact that comparatively densely populated prefectures in Japan generally have a definitely lower birth rate, but this is not due to the repressive influence of density itself. In the more densely populated districts the average income per capita (w) is relatively high, and people enter into matrimony at a higher age. Such factors as high income and late mar. riage are, as we shall see later negatively correlated with birth rate. We may, therefore, conclude that the original negative corrlation between birth rate and density of population is of secondary nature.