Global population datasets systematically underrepresent rural population

Main implications of the results. The findings from this study hold significant implications for a wide array of research and policy fields that consider rural areas and their populations, including disaster preparedness, public health planning, environmental conservation, and, ultimately, sustainable development. We assessed the accuracy of global gridded population datasets specifically in rural areas around the globe using reported human resettlement numbers from over 300 dam projects, which provide multi-national reference data fully independent from population censuses. We found a significant and systematic tendency for all datasets to underestimate rural population, with biases ranging from −53% (WorldPop) to −85% (GHS-POP). This is remarkable, as countless studies have employed these datasets without questioning their accuracy in the rural domain, and the systematic underrepresentation of rural population directly propagated into their results. It implies that the results of suc ... Read full article.