Approach on Triangulating Mortality Data to Strengthen Household Surveys for Mortality Associated Research

Chido Chinogurei, University of Cape Town

In sub-Saharan Africa where the vital registration data on mortality may not be fully complete, alternative sources for mortality data may help to improve our knowledge of trends and changes in mortality patterns close to real-time. Apart from the census and the vital registration, other sources often lack the scale of mortality data required to produce reliable estimates for mortality. On the other hand, household surveys provide a valuable source of data to explore various research in diverse fields of interest. However, despite their ability to explore a variety of research questions they are not that reliable when it comes to mortality indicators. South Africa records all the death that occur in the population from death notification forms which feeds into the vital registration records for deaths. This remains the most reliable information for mortality data in South Africa. Other research has looked at ways to triangulate the census and vital registration data to get mortality estimates at lower levels of the population3-5. In this paper we seek to triangulate data sources in order to create survey weights for deaths reported by households in the survey that can be used for mortality related research.

Keywords: Linked data sets , Data and Methods, Civil Registration and Vital Statistics

See extended abstract.