Estimating Fertility Rates from Multiple Data Sources by Mother's Educational Attainment

Dilek Yildiz , Wittgenstein Centre for Demography and Global Human Capital
Arkadiusz Wisniowski, University of Manchester
Afua Durowaa-Boateng, Vienna Institute of Demography

The main source of detailed fertility data in many developing countries is sample surveys. However, data on past fertility rates by level of education can be difficult to estimate from sample surveys due to small number of observations in the higher educational attainment groups, biases, and measurement and sampling errors. Moreover, they may not be consistent across sources, countries, and time. Previous researchers documented the inconsistencies and accounted for quality issues in some of the demographic indicators. The uncertainty in the estimates that have been in part quantified have not yet been used in multistate fertility rates (past and future) i.e. by educational attainment. We propose a Bayesian hierarchical model to reconstruct fertility rates by educational attainment using the full birth history module of Demographic and Health Surveys and censuses whenever available.

See extended abstract

 Presented in Session 47. Educational attainment and fertility