Marriage in Korea III. Age at Marriage, Family Planning Practices, and Other Variables as Correlates or Fertility

Data from this study support the View that the following factors are not sufficiently important in Korea to invalidate the relationships observed between age of marriage and fertility: (1) Premarital pregnancy and common law marriage, (2) shortening of birth intervals in late marriages, (3) adverse effects of very eary marriage in reproductive capacity, and (4) postponement of first pregnancy among early marriages. Thirteen variables which were considered to be potential predictors of fertility were studied to determine their influence on three indices of fertility. Age of marriage and family planning praetice are the strongest predictors and account for about 10% and 7% of the total variance, respectively. Seven other factors each account for an intermediate amount of variability; these are ideal number of children, rural versus urban study area, education, aspiration for daughter, index of exposure to mass media, economic index of respondent's home at survey, and residence before marriage. The remaining variables have no consistently significant relationship to fertility. Most of the relationships appear to be stable and consistent over time; others appear to be changing. The latter group include those variables which are associated with modernization indices of family planning practice, mass media exposure. and aspiration for daughters. Thus, the index of family planning practice is of limited significance for the $40{\sim}49$ age group but is the most important variable for the $20{\sim}29$ year women. The relationship is a direct one for the two age groups between 30 and 49 years which suggests that these groups already had high fertility when family planning services became available and that this high fertility then became an inducement to acccept contraception. The pattern of relationship is not yet clear for the $20{\sim}29$ year group. Similar interactions are observed for the other indices of modernity and are discussed. The thirteen variables together can account for a maximum of about 40% of the variance in the number of live births in the age group $30{\sim}39$, and for lesser amounts of variance in other age and fertility groupings.