Last month, MIT GOV/LAB practitioner-in-residence, Anisha Singh, presented her research on the importance of context when implementing behavioral science tools in the Global South. Currently the Director for Research and Innovation at the Busara Center for Behavioral Economics in Nairobi, Kenya, Anisha has extensive experience conducting behavioral science research and has also taught a course on behavioral experiments for international development at the University of Chicago.
While behavioral science offers great potential in aiding social science research, it comes with its own sets of challenges – only 0.002% of academic research comes from Africa, while the bulk comes from ‘WEIRD’ (western, educated, industrialized, rich, and democratic) countries. This is also problematic because behavioral science findings in WEIRD countries may not be applicable in developing countries which face a completely different context. “A lot of behavioral science research in the US is done with students, and then borrowed to low-income individuals in the Global South. We need to understand how citizens in each context perceive and interact with their institutions, and use that to inform applications of behavioral science,” Anisha said during her presentation.
Busara works with researchers, and civil society organizations to advance and apply behavioral science tools for poverty alleviation. “We started as a small lab, and now have a compiled database of over 133,000 participants, and have collaborated with 100+ practitioners, on 90 papers, and 3000 citations,” Anisha stated while talking about Busara’s reach. Busara is currently spread over five offices, and has 200 partners in 20 countries. Anisha mentioned that lab experiments come-in handy when primary and secondary data sources may not be viable. To illustrate, she spoke about a novel voting game that helps assess whether people vote for incumbents, even when they learn of their performance.
Lab experiments for governance questions
Elaborating on lab experiments, Anisha used the example of a study on participatory decision making in Kenya which was led by Tara Grillos, Associate Professor in the Department of Political Science at Purdue University, and later published in the British Journal of Political Science. The experiment aimed at understanding whether participation in decision making by citizens and stakeholders improves democratic governance. While there is substantial empirical work on procedures for decision making, the researchers sought to assess whether participatory decision making actually leads to better outcomes. Using lab experiments enabled the researchers to create groups that mimic real-world settings. The groups were then asked to participate in a public goods game involving multiple tasks, and were given payoffs depending on their performance. The groups were then asked to vote on which tasks they would choose to participate in again. “People in the group that deliberated before decision making saw an 11% higher payoff,” Anisha stated about the experiment. “Additionally, fairness in any experiment matters, and people should think if they were treated fairly,” she added.
In the discussion, attendees asked about the scope and limits of using experiments, sampling bias, and involving political stakeholders as they deliberated on the caveats of using behavioral science methodology for political science research.
At MIT GOV/LAB, Anisha is compiling a guide on learnings from applying behavioral science to solve development challenges. You can read more about Anisha’s work here.
Header: Image by Nina Stock from Pixabay.