Study: Mobility of workers tracked via their cell phone usage

An academic study charting the daily mobility of people based upon their mobile phone data has raised ethical concerns regarding privacy and nondisclosure.

The study was conducted by Northeastern University and recently published in the journal Nature. Cell phone usage data from a private European mobile phone carrier was used as the primary dataset for the study. Both the carrier and the nation in which the data was gathered were not named.

The information used in the study includes the date, time and coordinates of the phone tower routing the communication for each phone call and text message sent or received by 6 million customers during a six-month period. Individual users' phone numbers were represented by a 26 digit hash code instead of their actual number, and since location was determined by the nearest cell tower, location data is only accurate to an area about 3 square kilometers in size.

Of this 6 million, a random sampling of 100,000 callers' data was used. A further 1,000 users who received location-dependent service updates like weather, traffic, and pollen reports were sampled. These users provided coordinates far more frequently than others, amounting to about one set of location data every 2 hours. None of those being observed, however, were aware that their data was being collected.

In the United States, this sort of surveillance is illegal without first obtaining users' consent. That's assuming the surveillance takes place inside the US and involves "non-United States persons," as US law phrases it.

With this new data, researchers were able to construct what is called a Lévy flight, or a path composed of independent, random steps which are in no way predicted by past actions. The direction, as well as the distance of each step is presumed to be totally random, but when applied to a large number of "walkers," a general pattern can be formed. Interest in this type of diffusion as it pertains to human behavior was renewed in 2006 when it was applied to tracking the path of dollar bills.

The study found that nearly half of those mobile phone users did not leave a circle roughly six miles in diameter in the course of the half-year study, and 83% stayed within a 37 mile circle. At the other end of the spectrum, 3% had a circle that was 200 miles wide, and one percent traveled in a circle that was 621 miles across (roughly the distance from New York City to Cincinnati, Ohio or 10 hours by car according to Google Maps).