Structural and Proximate Determinants of Fertility in Ethiopia

Abstract

Introduction: Sub-Saharan Africa is still experiencing a high fertility rate with projections of population growth into the near future. To slow down the exponential increase in population and lower fertility rate, determinants of fertility need to be identified. While previous research identified many determinants, there are limited publications on the effect of ethnicity on fertility for Ethiopia even though it was found to be significant in some sub-Saharan countries. Objectives: The main objective of this research was to determine if ethnicity exerts an independent effect on fertility in Ethiopia. Method: Data for 7025 married women from Determinants of Fertility in Ethiopia: A National Survey that was conducted between December 2016 and January 2017 was used. The dependent variable was the number of live births per women. The explanatory variables can be divided into two categories – socio-demographic and fertility decision variables. Under socio-demographic variables are region, woreda type, womans age, education status,religion, womans occupation, cohabitation status, age of first marriage, number of husbands, type of marriage, husbands age, husbands education status, and husbands occupation. Fertility decision variables include ever had abortion, heard of family planning, contraception use, intention to use family planning, ideal child number, have sex preference, family planning decision by, decision to stop family planning by, decision to use contraception, and number of children decided by. Data analysis was done using R. Generalized linear regression quasi-Poisson method was used to analyze the data. Results: The results showed that women of different ethnicities experience different number of live births and differ in socio-demographic and fertility decision characteristics. Controlling for other variables, ethnicity still produced differences in the number of live births per women for select ethnicities. Determinants of fertility differed for individual ethnicities but three were consistently found in women of all ethnic groups except in those from Wolayata – womans age, age when first married, and ideal child number. Conclusion: Our results provide tentative support for ethnicity as a determinant of fertility directly and indirectly. For certain ethnic groups, ethnicity directly affects the number of live births while for others ethnicity may affect fertility by influencing socio-demographic and fertility decision variables. Implementing policy to raise the initial age of marriage, lower the desired fertility, and tailoring fertility policy to target Tigraway are methods that could lower fertility in Ethiopia.