What drives movement in the city? Many things. A new study, however, found that a fifth of the movement in urban areas is social in nature, a finding that is evident in various cities.

In a study published in the journal Interface, researchers turned to anonymous phone data offering information that allowed for people's social networks and locations to be reconstructed. By piecing together the information they had, the researchers were able to isolate networks as social- or work-oriented.

Jameson Toole from the Massachusetts Institute of Technology, one of the authors for the study, explained that they were able to determine how much of a city's movement was social in nature by layering mobile and social data over the other.

By coming up with a way to quantify how much of urban movement is affected by social activity, researchers are of the belief that they have taken a step towards the creation of a new analytical aid that will be highly useful for policymakers and planners.

"If you are trying to estimate movement in a city and you don't include the social component, your estimates are going to be off by about 20 percent," stated Toole.

The study used data from three big cities in South America and Europe. By analyzing call locations, contact times and call networks, the researchers uncovered that there are essentially three social networks at play in cities: distant acquaintances, work colleagues and social companions. After quantifying activity from these networks, they concluded that primarily social movement represents up to 30 percent of total movement in the city.

In the study, the researchers detailed as well the building of an urban social movement model called "GeoSim." It goes beyond previous urban mobility models by including a layer related to choices involving social activity. The GeoSim model fits data from the study better and may also be compared against other data in the future.

Experts say the study sheds new light on research on urban mobility, using two different kinds of data to establish relationship between two individuals. According to Esteban Moro from the Charles III University of Madrid, the study offers a quantitative grasp of how individuals manage tasks, time and interactions with a geographical context.

The study received funding support from the National Science Foundation, MIT's Center for Complex Engineering Systems and the Accenture-MIT Alliance in Business Analytics. Other authors include Marta C. González, Christian M. Schneider, and Carlos Herrera-Yaqüe.