An official website of the United States government
Poverty maps derived from satellite imagery are increasingly used to inform high-stakes policy decisions, such as the allocation of humanitarian aid and the distribution of government resources. Such poverty maps are typically constructed by training machine learning algorithms on a relatively modest amount of “ground truth” data from surveys, and then predicting poverty levels in areas where imagery exists but surveys do not. Using survey and satellite data from ten countries, this paper investigates disparities in representation, systematic biases in prediction errors, and fairness concerns in satellite-based poverty mapping across urban and rural lines, and shows how these phenomena affect the validity of policies based on predicted maps. Our findings highlight the importance of careful error and bias analysis before using satellite-based poverty maps in real-world policy decisions.
more »
« less
Microestimates of wealth for all low- and middle-income countries
Many critical policy decisions, from strategic investments to the allocation of humanitarian aid, rely on data about the geographic distribution of wealth and poverty. Yet many poverty maps are out of date or exist only at very coarse levels of granularity. Here we develop microestimates of the relative wealth and poverty of the populated surface of all 135 low- and middle-income countries (LMICs) at 2.4 km resolution. The estimates are built by applying machine-learning algorithms to vast and heterogeneous data from satellites, mobile phone networks, and topographic maps, as well as aggregated and deidentified connectivity data from Facebook. We train and calibrate the estimates using nationally representative household survey data from 56 LMICs and then validate their accuracy using four independent sources of household survey data from 18 countries. We also provide confidence intervals for each microestimate to facilitate responsible downstream use. These estimates are provided free for public use in the hope that they enable targeted policy response to the COVID-19 pandemic, provide the foundation for insights into the causes and consequences of economic development and growth, and promote responsible policymaking in support of sustainable development.
more »
« less
- Award ID(s):
- 1942702
- PAR ID:
- 10418102
- Date Published:
- Journal Name:
- Proceedings of the National Academy of Sciences
- Volume:
- 119
- Issue:
- 3
- ISSN:
- 0027-8424
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Poverty maps derived from satellite imagery are increasingly used to inform high-stakes policy decisions, such as the allocation of humanitarian aid and the distribution of government resources. Such poverty maps are typically constructed by training machine learning algorithms on a relatively modest amount of “ground truth” data from surveys, and then predicting poverty levels in areas where imagery exists but surveys do not. Using survey and satellite data from ten countries, this paper investigates disparities in representation, systematic biases in prediction errors, and fairness concerns in satellite-based poverty mapping across urban and rural lines, and shows how these phenomena affect the validity of policies based on predicted maps. Our findings highlight the importance of careful error and bias analysis before using satellite-based poverty maps in real-world policy decisions.more » « less
-
null (Ed.)Recent papers demonstrate that non-traditional data, from mobile phones and other digital sensors, can be used to roughly estimate the wealth of individual subscribers. This paper asks a question more directly relevant to development policy: Can non-traditional data be used to more efficiently target development aid? By combining rich survey data from a "big push" anti-poverty program in Afghanistan with detailed mobile phone logs from program beneficiaries, we study the extent to which machine learning methods can accurately differentiate ultra-poor households eligible for program benefits from other households deemed ineligible. We show that supervised learning methods leveraging mobile phone data can identify ultra-poor households as accurately as standard survey-based measures of poverty, including consumption and wealth; and that combining survey-based measures with mobile phone data produces classifications more accurate than those based on a single data source. We discuss the implications and limitations of these methods for targeting extreme poverty in marginalized populations.more » « less
-
Abstract Social scientists have increasingly used asset‐based wealth scores, like the Demographic and Health Survey (DHS) wealth index, to assess economic disparities. However, current indices primarily capture wealth in globalized market economies, thus ignoring other forms of prosperity, such as success in agricultural activities. Using a simple extension to the standard estimation of the DHS wealth index, we describe procedures for estimating an agricultural wealth index (AWI) that complements market‐based wealth indices by capturing household success in agricultural activities. We apply this procedure to household data from 129 DHS surveys from over 40 countries with sufficient land and livestock data to estimate a reliable and consistent AWI. We assess the construct validity of the AWI using benchmarks of growth in both adults and children. This alternative measure of wealth provides new opportunities for understanding the causes and consequences of wealth inequality, and how success along different dimensions of wealth creates different social opportunities and constraints for health and well‐being.more » « less
-
Data from surveys administered after Hurricane Sandy provide a wealth of information that can be used to develop models of evacuation decision-making. We use a model based on survey data for predicting whether or not a family will evacuate. The model uses 26 features for each household including its neighborhood characteristics. We augment a 1.7 million node household-level synthetic social network of Miami, Florida with public data for the requisite model features so that our population is consistent with the survey-based model. Results show that household features that drive hurricane evacuations dominate the effects of specifying large numbers of families as \early evacuators" in a contagion process, and also dominate effects of peer influence to evacuate. There is a strong network-based evacuation suppression effect from the fear of looting. We also study spatial factors affecting evacuation rates as well as policy interventions to encourage evacuation.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Journal Article:
- https://doi.org/10.1073/pnas.2113658119
