Re-weighting South African national Household survey data to create a consistent series over time: A cross entropy estimation approach

Abstract

In the absence of established longitudinal panel surveys in South African, national cross-sectional household survey data are frequently used to analyse change. When these data are stacked side-byside, however, inconsistencies both in time trends and between household and person level data are found. This study uses a new set of weights calibrated to the ASSA 2003 model totals using a cross entropy estimation approach. This approach is favoured because the calculated weights are similar to the initial sample weights (and hence retain the survey design benefits) but match to a series of age-sex-race and province marginal totals that are consistent over time. The weights are publicly available for a fourteen year period between 1994 and 2007. This is a SALDRU/DataFirst Working Paper