Relative Wealth Index for Low and Middle-Income Countries

93 low and middle-income countries are covered by this dataset predicting relative living standards at 2.4km resolution. It was developed by researchers including Guanghua Chi and Joshua Blumenstock, with findings published in the Proceedings of the National Academy of Sciences in January 2022. The dataset is provided by AI for Good at Meta.

Use Cases

• Analyze spatial inequality patterns within a country using the 2.4km resolution wealth index grid.
• Validate national poverty survey results against the satellite imagery and connectivity data-derived index.
• Train models to predict local economic outcomes by correlating the relative wealth index with other geospatial features.
• Target humanitarian aid delivery by identifying low-index geographic clusters within the 93 covered countries.

Strengths

• Covers 93 low and middle-income countries, providing broad geographic scope.
• Provides data at 2.4km spatial resolution, enabling local-level analysis.

Limitations

• Limited to low and middle-income countries, excluding high-income nations.
• Wealth estimates are relative within countries, not absolute or directly comparable across borders.
• Methodology relies on non-traditional data sources like connectivity and satellite imagery, which may introduce specific biases.

Provenance

Source

AI for Good at Meta.

Collection Method

Generated using de-identified connectivity data, satellite imagery, and other non-traditional data sources.

Freshness

Last updated March 2026.

Geography

93 low and middle-income countries.