Play Live Radio
Next Up:
0:00
0:00
0:00 0:00
Available On Air Stations

UW Professor Can Tell The Rich From The Poor Using Cell Phone Metadata

Adam Jones
/
Wikimedia Commons
A woman in southern Rwanda checks her cell.

Researchers at the University of Washington say they can use phone records to help humanitarian efforts in developing countries. The key is the different cell phone habits of wealthier and poorer people.

 

The secret lies in cell phone metadata – the stuff that, until late last month, the National Security Agency was able to collect in order to spot suspicious patterns. But Joshua Blumenstock has found very different patterns in the data. The UW Information School assistant professor has found a way to identify a person's socioeconomic status just by looking at who, when and where they call, and how long they talk.

 

 

Blumenstock and his team began with a traditional phone survey of about 1,000 Rwandans, in which they determined people's wealth and income. Then they took those people's phone metadata, fed it into a computer, and allowed “machine learning” to find patterns that correlate with wealth or poverty.

 

Blumenstock says they found a number of signatures in the data – poorer people tend to receive more calls than they make, for example. People who call at certain regular times tend to be wealthier.

 

And it’s not just who’s rich or poor – the researchers say they can make more specific predictions.

 

“We look at whether we can predict whether the household has electricity or not, or whether the household owns a motorcycle,” Blumenstock said.

 

This could be an important tool because governments and aid groups need to know where the neediest people are in order to help them. Many developing countries can’t or don’t spend the money to do a traditional census.

 

“It's potentially an alternative to nothing. In a lot of countries you just don't have any data,” he said.

 

Blumenstock says this method might someday be used to gather information about public opinion or mental health status. His findings are published in the journal Science.

 

Gabriel Spitzer is a former KNKX reporter, producer and host who covered science and health and worked on the show Sound Effect.