A missing link in the fertility literature? Evidence of household structure effects on fertility from Ghana Demographic and Health Surveys

Magdalena Benza Fiocco, San Diego State University
John R. Weeks, San Diego State University

The goal of this study is to explore the relationship between fertility levels and household structure in Ghana, focusing on whether a woman's level of reproduction is related to the type of household in which she lives. The analysis is based on the Ghana Demographic and Health Surveys for the years 1993, 1998 and 2003. Household structure is approximated through a series of dummy variables that define the organization of the family unit, such as the couple's composition (cohabitation, polygamy and monogamy) and the management of extended family with a special focus on practice of fosterage. The characteristics of these household arrangements are systematically linked to two major structural determinants of reproductive behavior in west-Africa: ethnicity and religion. Fertility is defined for this analysis as a set of dummy variables based on women having (1) at least one live birth in the last year and (2) at least one in the last five years. A series of logistic regressions are run to estimate the degree of correlation between the two sets of variables controlling for age, education, income, parity, knowledge and the use of contraceptive. Results for 1993, 1998 and 2003 are compared for the purpose of examining regional and temporal patterns in this correlation. Results indicate that whereas the number of household members has a positive relationship to fertility levels some specific type of family composition are negatively correlated to reproduction.