This article “Verhaltensforschung geht ins Feld Ein Kurs in Kenia bringt den Globalen Süden und den Norden ins Gespräch” was written for the Berlin Social Science Center quarterly magazine (WZB- Mitteilungen) and is available online in German. An English language version is below.
Behavioral research goes into the field. A course in Kenya brings the Global South and the North into conversation
Summary: In early 2020, Busara Center for Behavioral Economics and the MIT Governance Lab (MIT GOV/LAB) brought together 12 scholars from U.S.- and East African universities to participate in a novel behavioral science course. This initiative was developed and launched while Kelly Zhang was a visitor at the Institutions and Political Inequality (IPI) research unit at the WZB. It covered best practices in the design and implementation of lab-in-the field experiments at the cross-section of economics, political science, and psychology. The goal of the initiative is to produce new knowledge and train the next generation of scholars in cutting-edge interdisciplinary methods and policy-relevant research questions. It also aimed to ground research questions within the local Kenyan context and to encourage active dialogue between Global North and Global South scholars in the development of these research designs.
Humans by nature are not perfect. We can be indecisive and irrational. We make mistakes.
We can also be empathetic, kind, and caring. Take a moment to reflect:
When were you last asked to donate to charity? Did you donate? How much was too much?
When did you last have an extra beer, cigarette, or choose vegetarian? Who was around?
When did you last intervene to help a stranger? Was there a time you chose not to intervene?
When did you last read a news article that changed your opinion? Why or why not?
Human behavior is complex. The decisions, big and small, we make every day are influenced by a complex set of interwoven social, economic, and political factors —how we were raised, individual personalities, community norms, the specific context where and when a decision is made. In recent years, behavioral science has emerged as a leading way to better understand determinants of human behavior, with lab-in-the-field experiments leading as a gold standard.
A critical question is how we generalize about human behavior and how precisely context matters in measuring our actions. For example, how do we consider differences between Berlin and Nairobi when we set out to measure behavioral trends in these two global cities? Currently most of the scientific evidence comes from U.S. and European contexts, and comparatively little research is conducted in Kenya, East Africa, or the African continent at large. There is a “WEIRD” bias of existing behavioral science research, that is the disproportionate amount of data coming from western, education, industrialized, rich and democratic contexts.
To start addressing this knowledge gap, we launched a new course in Kenya. The Behavioral Science in the Field course is a collaboration between MIT Governance Lab and the Busara Center for Behavioral Economics, to train graduate students from the U.S. and local universities in East Africa in cutting-edge behavioral science research. Conducted in Kenya, the course is structured as an intensive deep dive into interdisciplinary behavioral science, with three tracks in economics, political science, and social psychology, and training in a complete research cycle. (The full course syllabus is available online).
Why behavioral science in the Kenyan context?
While technology companies and businesses excel at using behavioral science to advance marketing and sales, the international development field and public sector lag behind. There is now a growing opportunity to apply behavioral science to improve development outcomes, like poverty reduction and improving access to basic services like water, electricity, education, and health.
Lab-in-the-field experiments are one behavioral science method where participants are recruited to play a set of games and then prompted to answer a series of questions. The games, usually viewed and played on a portable tablet, are designed to mimic real life scenarios and correspond to decisions that participants would undertake in their everyday lives. The questions that follow, combined with observation of how participants play the games, give researchers a measure of actual behavior as well as corresponding attitudes.
Let’s give a concrete example. A recent study conducted in Kenya wanted to better understand how and why ethnicity biases behavior, that is the reason underlying why respondents might behave differently when confronted with a task given by someone of their same ethnicity or someone from a different group (Blum, Hazlett, Posner 2020). While many studies have focused on this question, traditional research methods are often unable to accurately measure bias, either because people often don’t openly admit their own biases in an effort to be likeable (i.e., social desirability bias) or because they may not themselves be aware of them. Yet, understanding these underlying biases and how they arise (e.g., is bias due to lack of altruism, fear, mistrust, or actual dislike) is critical to inform efforts seeking to reduce discrimination , and even prevent ethnic conflict. The study therefore used a behavioral lab experiment to document how changing ethnicity impacts individual behaviors and attitudes towards a different group. It used behavioral games embedded within the local context and unobtrusively varied ethnicity in order to document participants’ hidden biases.
The study was additionally exciting because it was conducted by political scientists who combined theory from social psychology and games from economics to examine a well-studied topic from an innovative and novel approach. The Behavioral Science in the Field course was designed precisely to enable graduate students to produce this type of research, while being grounded deeply in the local Kenyan context.
Ingredients for success: motivation and mentorship
The course is the brain-child of Kelly Zhang, MIT GOV/LAB research scientist, and Chaning Jang, Busara’s chief executive officer. They envisioned designing the field practicum course that they wish they had been able to take in graduate school. Typical courses cover one or more aspects of theory, methods, and research design, but this course includes all aspects of research design to implementation in a single go.
To encourage innovative thinking beyond disciplinary boundaries, the course was designed and taught by instructors from across disciplines for students from a diversity of backgrounds. To create a cohesive cohort, we emphasized peer-feedback and equal exchange in seminars and workshops throughout. Mentorship from leading experts in the field, including WZB’s Macartan Humphreys and Alexandra Scacco as guest lecturers and reviewers, were elemental in setting the bar for expectations and providing support for high-quality research questions and designs.
A central part of this initiative was to invest in East African scholars so that they could develop original research questions alongside scholars from the U.S. universities. Similarly, another aim was to embed these researchers in the field to build empathy and recognize and respect the perspective of the research subject in the research process. Both of these were critical for building productive and long-lasting relationships between disciplines and cultures for collaboration, as well as innovative research designs that were relevant to the local context.
Designing novel behavioral games
The course took place over four intensive weeks, which included both classroom time and fieldwork. The first cohort had twelve participants (nearly half women), six from US doctoral programs and six from East Africa, including Kenya, Uganda, and Tanzania. While purposely designed with a demanding and rigorous pace, the course feedback was overwhelmingly positive and after an exhausting sprint, participants produced innovative research designs that addressed timely and relevant questions for the Kenyan context.
For example, Laura Barasa, professor at the University of Nairobi, designed her behavioral game around the rising issue of online youth gambling in Kenya. She wanted to understand what factors increased risky gambling behavior and what interventions, including social pressure and warning messages, could positively influence youth behavior.
Another participant, George Kinyanjui, originally from Kenya and completing his doctoral studies at the University of Capetown, is investigating the motivations behind community giving to cover medical costs. In Kenya, communal giving is an important social norm and it is common for individuals to donate money to cover healthcare costs. Even though insurance is usually a cheaper option, less than 20 percent of people have formal insurance.
Stuart Russell and Nicole Wilson, American doctoral candidates at MIT, were most interested in accountability in politics, especially in cases of corruption. They wanted to know whether or not information on a politicians’ good or bad performance influences how Kenyan’s vote, and specifically if they decide to keep someone in office or vote them out.
Peter Babyenda, originally from Uganda and currently a doctoral student at the University of Nairobi, focused his research on how climate change information can influence farmers to make medium and long-term changes to their agricultural inputs.
Other research projects covered a variety of policy relevant issues, for example, trust in mobile-money agents (M-PESA), how ethnic bias plays a role in trading networks, when people value freedom of choice over sound advice, what motivates people to fight or flee in the case of violent threat, how to inspire critical thinking in youth, the role of emotions in reducing corruption. We look forward to sharing the results from our first cohort of scholars once data collection and analysis is complete.
(Course participants conducting field work in Machakos County, Kenya)