Fertility,development, and family planning, 1970–1980: An analysis of cases weighted by population

A series of related studies (Freedman and Berelson, 1976; Mauldin and Berelson, 1978; and Tsui and Bogue, 1978) have presented empirical findings based on multiple regression analysis which indicated that family planning program effort (FP), as measured by an index developed by Lapham and Mauldin (1972), was the single most important predictor of (or influence on) fertility reduction in less-developed countries (LDCs). The basic results have been confirmed repeatedly. A more extensive data set was used to extend the analysis to a comparison of results of corss-sectional models circa 1970 and 1980. The study builds upon the results of past studies yet differs from them in several ways. All the variables in the present study were measured at 2 points in time: circa 1970 and circa 1980, allowing a comparison between cross-sectional models for 1970 and 1980. Among the cases included in this multivariate analysis was China, a country usually excluded for lack of data. The analysis was extend to 85 countries. Cases were weighted by population, having the effect of increasing the impact of larger countries such as India and China on the outcome of the analysis. Total fertility rate (TFR) was used as an indicator of fertility. For 1970, family planning program effort had the strongest direct influence on fertility (a result consistent with previous studies). Life expectancy at birth was the other direct influence. The direct influence of life expectancy at birth was less than that of family planning, but the total influence was greater. After life expectancy and family planning, school enrollment and relative educational status of women had the strongest indirect and total influences. The other variables all had a positive influence on fertility. When the total variance attributable was considered, directly and indirectly to each of the independent variables, urbanization, carlorie supply, and per capita gross national product all accounted for less than 5% of the variance in fertility, all of it indirect. Life expectancy, family planning, and school enrollment each explained (directly plus indirectly) more than 10% of the variance in fertility. The pattern differed somewhat for 1980. Calorie supply, per capita gross national product, and relative educational status of women had no influence, direct or indirect on fertility. Also for 1980, life expectancy had a stronger direct influence on fertility than family planning. Overall, life expectancy at birth, family planning program effort, and total school enrollment emerged as the principal influences on fertility.