Data for Solving
Global Challenges

World Data Lab specializes in creating economic and demographic models that help organizations make data-driven decisions. Our global team of data scientists, economists, demographers, and developers are passionate about combining new data sources with groundbreaking methods to push the frontiers of what is knowable.

Granular models of income for every country—based on peer-reviewed methodologies

Leveraging data science to support reaching Sustainable Development Goal 1

The world's first global life expectancy model and custom spatial demography models: population density + age structure

Subnational Poverty Mapping

To eradicate extreme poverty worldwide, country-level data is only the first step. Decision-makers need to know which regions, districts, and villages are home to those with the greatest needs—and which areas will likely still experience poverty in five years' time. As a result, World Data Lab has rolled-out a series of subnational poverty piloting projects in select countries worldwide.

Methodology: Integrated Luminosity Downscaling

This model estimates poverty and income in all of Pakistan’s districts. Under this approach, PSLM income and consumption surveys are combined (consumption surveys are province-based, income surveys are district-based). District-based surveys are ultimately used to distribute consumption across districts and missing areas in the surveys are filled using nighttime lights based on the relationship between economic activity (in our case consumption, which is correlated to output) and luminosity at night.

Following this preparatory work, Beta-Lorenz curves for each district are calculated and incorporated into a convergence model based on consumption growth and share of people with secondary education. The national growth rate of the share of people with secondary education is assumed for all districts in the survey, and national distribution parameters (and national education shares) are used for missing districts.

Read the paper

Putting Our Data
To Work
Pakistan Subnational Poverty Mode In partnership with the International Growth Centre, we developed a subnational poverty model for Pakistan incorporating multiple data sources and nighttime imagery. Results from this model will ultimately be published on the World Poverty Clock to inform decision-making and research.
use clients image
Kenya Subnational Poverty Mode n partnership with the NOMIS Foundation, we developed a subnational poverty model for Kenya incorporating multiple data sources and nighttime imagery. The results from this model are featured on the World Poverty Clock to inform decision-making and research.
use clients image
Let's Eliminate Extreme
Poverty Together

We seek partnerships with like-minded development organizations interested in developing consistent and credible subnational poverty estimates and forecasts for specific countries where pockets of extreme poverty persist.