Putting evidence into policymaking: RCTs as a tool for decision-making

In India, the Abdul Latif Jameel Poverty Action Lab, a network of researchers who run randomized control trials based at the Massachusetts Institute of Technology, is working with the Tamil Nadu government to integrate findings from RCTs into the policymaking design phase — a collaborative approach which evolved from J-PAL’s existing evaluation programs there.

The venture signals the first time the government is systematically integrating RCTs in evaluating the impact of programs across departments and designing central policies, a departure from the traditional role of RCTs as tests for individual program evaluations.

“This is cutting edge. Five years ago, we were partnering more with [nongovernmental organizations], and now governments are investing in centralizing RCTs as a tool to learn what works in social programs,” John Floretta, deputy director at J-PAL South Asia, told Devex from India.

Formally launched in November 2014, the Tamil Nadu Innovation Initiative enables various departments to work with J-PAL researchers in designing new evidence-based programs to fit the government’s most pressing policy priorities, and test the policies through RCTs to measure success.

A $25 million state fund finances the initiative and other innovative projects for social solutions.

Before they can design a program, however, departments and researchers first have to submit a joint proposal for RCT evaluations to the initiative. Five studies have so far been approved, including measuring the impact of integrating preschool education programs into local community centers, behavioral interventions to reduce open defecation, and a rice fortification program to reduce anemia, Floretta said.

The findings of the RCT evaluations will also help a steering committee determine whether programs should continue, be scaled up, redesigned or discontinued, he said.

Through the collaboration, the government gains expertise from J-PAL, enabling it to build in-house capacity on measurement, large-scale data collection and outcome-based monitoring.

The program is unique, Floretta further explained, because it’s the first case J-PAL has had where RCTs are integrated centrally within a state government as a means of evidence-based policy across departments during planning, instead of “one-off” trials with separate partners. This enables easier decision-making and action, he noted.

The full integration of RCTs from the beginning into the design phase of social programs is also “very powerful, a huge opportunity for development policies informed by evidence,” Floretta concluded.

Read more international development news online, and subscribe to The Development Newswire to receive the latest from the world’s leading donors and decision-makers — emailed to you FREE every business day.