Access to high-quality and timely information about extreme poverty is required to develop and target appropriate policies, strategies, and programmes to tackle it, and to monitor progress towards achieving the Sustainable Development Goals. Despite advances over the last 20 to 30 years, this information remains slow to emerge, is often available only at a high cost, and is frequently only available at high levels of spatial aggregation. Traditional approaches to poverty measurement have relied on household surveys or census data, which are costly and therefore collected infrequently.
New sources of data – including remote sensing data, data from the data exhaust, online data, crowd-sourced data, and mobile phone survey data – and novel statistical techniques have the potential to enhance the toolkit of approaches available for measuring and investigating extreme poverty. This paper is one of a series produced by the Data and Evidence to End Extreme Poverty (DEEP) research programme to explore how innovation in data collection, data processing, and data analysis might, with further development, provide solutions to ‘pinch points’ in policymaking and policy management for poverty reduction. The focus of this paper is on exploring the suite of different data sources that can be used for measuring and investigating poverty. Our focus is on sources that can be used to help measure extreme poverty directly, as well as related attributes, such as determinants, proxies, or correlates of poverty. We adopt a consistent assessment framework to describe and compare different approaches and understand their attributes and limitations.
We conclude by discussing to what extent new technologies can enable better insights into the incidence, distribution, and severity of extreme poverty over space and time, and where the most promising avenues for adopting these methods or investing in further research lie.
Available at: https://doi.org/10.55158/DEEPWP2
