World Bank money is funding this driving test. A group of randomly selected people in New Delhi is busy trying to acquire driving licences. The instructions are clear: be honest, one group is told. Another set is gently nudged toward touts swarming the RTO. Some are promised a ‘bonus’ if they get a licence real quick.

Everybody’s experience of bribes and babus is carefully noted and studied.

Then there’s a short driving test.

A group of researchers at the Massachusetts Institute of Technology (MIT) at Cambridge, who are studying the nightmares of ‘‘bureaucratic corruption,’’ will scientifically evaluate this on-going survey and driving test. Are the honest set better drivers?

‘‘We were not satisfied by simply producing high-quality research,’’ MIT professor of economics Abhijit Banerjee told The Indian Express in an e-mail interview.

In June last year, professors of economics Banerjee, Esther Dufflo and Sendhil Mullainathan—with extensive experience studying racial bias in Kenya, USA, India and the Africas, education, health, agriculture and poverty—founded the Poverty Action Lab at MIT.

‘‘We’re convinced,’’ says Banerjee, that policies designed to reduce poverty could be substantially more effective if scientific evidence was available . The lab doesn’t have test tubes, but work is based on rigorous evidence .’’

In sync with researchers at Harvard University, Princeton University, University of California, Berkley, and University of Chicago, random trials are on from Gujarat, Rajasthan and Maharashtra to shanty towns in Kenya and Chicago.

  In black and white  

In another experiment, researchers associated with the Lab recently sent resumes to newspaper ads for jobs in Boston and Chicago. Applications were randomly assigned a black-sounding name (Lakisha Washington, Jamal Jones) or white-sounding name (Emily Walsh, Brendan Baker). The result: White sounding names received 50 percent more call backs than the Black sounding names.

The study’s main result maintained that ‘‘women are politically more active in village councils with a woman leader.’’

In a cluster of non-formal educational centres in Udaipur, teachers—known to attend only 60 per cent of the time—photograph the class at the start and end of the day.

An MIT researcher on this ongoing project checks the camera’s timestamp feature to determine when and for how long each school was open. A bonus awaits teachers with good attendance.

In Bhopal, researchers funded by MIT and University of Chicago are trailing college-age students of upper and lower castes seeking engineering admissions. Tracking their future education and job opportunities, they’ll study and predict the impact of admission policy on future income, social standing.