Demographic and Health surveys (DHS) data regarding the rate of literacy in people age 15-49 were used to predict high spatial resolution gender disaggregated maps, at unsampled locations in Nigeria, Kenya, Tanzania and Bangladesh, using predictive modelling techniques. Geostatistical models (machine learning and Bayesian geostatistical techniques) were used to produce detailed maps of literacy, taking advantage of the fact that literacy shows strong correlations with environmental and sociological factors, many of which are available nowadays as gridded spatial datasets. The outputs consist of high-resolution maps (1x1 km) of the proportions of women and men that are literate, together with estimates of mapping uncertainty. Quantifying the impact of gender disaggregated literacy rate in low- or medium-low- income countries is crucial to adequately inform policy- makers and decision- makers in order to promote any initiative that aims to make advances towards increasing the number of literate people and achieving gender equality.
Andy Tatem, Claudio Bosco
WorldPop, Geography and Environment University of Southampton A.J.Tatem@soton.ac.uk, C.Bosco@soton.ac.uk (023) 8059 2636
UN Foundation, Data2x program
Bill and Melinda Gates Foundation