A person's history of phone communication can be used to infer aspects of his or her socioeconomic status, a new study suggests.
The study, conducted by researchers from the University of Washington, focused on Rwanda. It reveals how mobile phone metrics can be a source of "big data" in resource-constrained regions.
Collecting data on basic economics quantities -- such as wealth and income -- is challenging in developing countries, making reliable quantitative data scarce.
In much of Africa, for instance, national statistics on economic production may be off by as much as 50 per cent, previous research suggests. Yet insights into the geographic distribution of poverty and wealth are critical for policymakers and others.
In their study published in the journal of Science, the researchers say that they have devised a way to estimate the distribution of wealth and poverty in an area by studying metadata from calls and texts made on mobile phones. Such metadata contains information about the time, location and nature of the "mobile phone events" but not their content.
In developing or war-ravaged countries, where government censuses are few and far between, gathering data for public services or policymaking can be difficult, dangerous or near-impossible. Big data is, after all, mainly a First World opportunity.
The study sought to develop a new approach to measuring how poverty and wealth are distributed in developing countries. The researchers took advantage of the ubiquity of mobile phones in Rwanda, and the fact that mobile phone data captures information about the structure of a person's social network, his patterns of travel, his histories of data use and expenditure, and more.
'Signature' of wealth
By combining data about individual mobile phone users' calls from an anonymised call database containing billions of interactions with information from a follow-up phone survey on basic welfare indicators involving more than 850 respondents, the researchers developed a model that maps poverty and wealth of individual phone users at very high resolution (greater than that achieved with satellites).
They used the model to predict wealth throughout Rwanda, showing that their predictions agreed with detailed, boots-on-the-ground surveys of the Rwandan population.
The researchers also conducted telephone interviews with 1,000 mobile phone owners chosen at random. Questions were designed to learn where the individuals fell on the socioeconomic ladder and what the "signature" of wealth is in the metadata -- that is, what mobile phone habits are particular to the relatively wealthy.
"For those thousand people, we know roughly whether they're rich or poor," said study lead author Joshua Blumenstock, assistant professor in the UW Information School.