Poverty Mapped in Near Real-Time Using Mobile Phone and Satellite Data
The new approach one-ups conventional methods that typically rely on out-dated census figures.
When it comes to understanding the distribution of poverty in a country, experts are often in the dark. Typically, locating and accurately measuring the three billion people globally that live on less than $2.50 a day depends heavily on census data, which in most low-income countries is unavailable or out-of-date.
But a new study published today in The Journal of the Royal Society Interface shows that combining mobile phone data with satellite data can create high-resolution maps in near real-time that could produce a more accurate and on-going picture of where poor people live.
"If we're going to tackle poverty, we need ways of mapping and measuring it," Jessica Steele, research fellow at the University of Southampton and lead author of the new study, told Seeker.
The researchers collected three main chunks of information: remote sensing data from satellites, call detail records from Grameenphone, the leading telecommunications service provider in Bangladesh, and three national household surveys taken across Bangladesh.
Satellite data helped reveal the physical details of the land related to human living conditions. For example, vegetation can indicate farming practices and night-time lights in various regions can be linked to economic prosperity and activity. The cell phone data, such as number of text messages and monthly credit consumption, show where people are using their phones as well as link to a household's financial status.
The surveys provided concrete information about the financial status of tens of thousands of people.
Next, the researchers used the satellite and mobile phone data to create maps showing where the poorest people lived. The survey data helped confirm that the maps were accurate.
The researchers found there was a strong correlation between poverty and satellite and mobile data and based on their analysis, confirmed that the predictive power of their mapping technique was quite strong. Using that information, they were able to generate poverty maps for regions where little information or data existed before.
Accurately monitoring poverty with near real-time data could have significant advantages for targeting and measuring the success of aid interventions, as well as potentially improving factors related to poverty, such as reducing child mortality. It could also bring the world another inch closer toward ending world poverty.
"In 2016, the United Nations pledged to '...end poverty in all forms and dimensions by 2030' as part of its Sustainable Development Goals," Steele said. "One of the steps towards achieving this involves targeting the most vulnerable in society."
Steele added that she and her team are now expanding their work to other countries.
"We hope the findings can be utilized to track poverty more effectively in the future and provide detailed and accurate information to better inform governments and relief organizations," Steele said.
High resolution maps from the study showing the distribution and levels of poverty in Bangladesh can be found at World Pop.
WATCH VIDEO: Is Poverty Sexist?