The quality of demographic data on older Africans

<b>Background</b>: Developing appropriate and equitable policies for older people in Africa requires accurate and reliable data. It is unclear whether existing data can accurately assess older African population structures, let alone provide the detailed information needed to inform policy decision-making. <b>Objective</b>: To evaluate the quality of nationally representative data on older Africans through examining the accuracy of age data collected from different sources. <b>Methods</b>: To measure the accuracy of age reporting overall we calculate Whipple's Index, and a modified Whipple's Index for older adults, using the single year age-sex distributions from (a) the household roster of 17 Demographic and Health Surveys (DHS), (b) the censuses of 12 of these countries, and (c) the Living Standards Measurement Study (LSMS) for Ethiopia and Niger. We compare reported sex ratios by age. <b>Results</b>: The quality of age data is very poor for most countries outside Southern Africa, especially for older adults. In some Sahelian countries DHS surveys appear to omit a considerable proportion of older women. Data on population structure of older people by age and sex produced by the DHS and the census are inconsistent and contradictory. <b>Conclusions</b>: Different field methodological approaches generate contradictory data on older Africans. With the exception of Southern Africa, it is impossible to assess accurately the basic demographic structure of the older population. The data available are so problematic that any conclusions about age-related health and welfare and their evolution over time and space are potentially compromised. This has ramifications for policymakers and practitioners who demand, fund, and depend on large-scale demographic data sources. <b>Contribution</b>: The paper highlights a number of problems with data on older Africans, beyond the well-known issues of age heaping. In doing so it contributes to general understanding of the limitations of existing demographic data for any detailed analysis of the situation or characteristics of older Africans. The heterogeneity of data quality and data problems for older Africans across the continent suggests that considerable care should be taken in (a) drawing conclusions from comparative studies using internationally standardised data sets and (b) analyses which combine different sources of data. Particular data problems with surveys in the Sahelian countries could be addressed in future data collection exercises