machine learning

A new adjusted and harmonised satellite nighttime-lights series for 1992–2023 tracks local development in the Global South more accurately than the off-the-shelf data – especially in panels and at fine spatial resolution.

We develop a new method to analyse open-ended qualitative interviews with large samples, and apply it to interviews with Rohingya refugees and Bangladeshi hosts on parent’s aspirations for children, revealing dimensions of aspiration that standard su...

Although machine learning models using mobile phone data can make poverty targeting faster and more cost-effective, traditional survey-based methods remain more accurate. The optimal approach therefore depends on striking the right balance between co...

In settings where reliable data on poverty is difficult to come by, non-traditional data sources such as mobile phone metadata has the potential to fill data gaps. New research on a cash transfer programme in Togo reveals that mobile phone data enabl...

How can we estimate state presence in areas where direct measurements are lacking? New research offers a solution to measuring state presence using machine learning techniques.

When governments harness public budget data and machine learning, they can better predict the incidence of local government corruption.

The introduction of mass public transportation systems in Lima, Peru, connected neighbourhoods, reduced commuting times, and increased access to college

Governments can use satellites and machine learning technology to reduce illegal activity