Grad Life: Bringing the Lab to the Field

Leah Rosenzweig and her work for GOV/LAB is featured in MIT Alumni Association's "Slice of MIT."

(Header image: Leah Rosenzweig conducts an experiment in Mbeya, Tanzania, in November 2015 to understand how citizens evaluate political candidates.)

This post was originally featured in MIT Alumni Association's "Slice of MIT" here.

Discussing strategy with political opposition leaders in Tanzania or interviewing rural Kenyans about their motivations to vote aren’t what you typically think of as laboratories. But for researchers like me at MIT GOV/LAB, rural villages in East Africa, community meetings in the Philippines, and municipal offices in Mexico are our laboratory.

My political science research with GOV/LAB, since its inception in 2013, has taken me to poor rural villages in Tanzania, urban slums in Kenya, and youth hangouts in Uganda. Currently, I am investigating how poor Tanzanian citizens evaluate political candidates and why they vote in elections. Despite the fact that the same political party has ruled the country since 1961, a majority of Tanzanian citizens, some who cannot afford enough food to feed their families, stand in lines for hours to cast their ballots. I want to understand why.

In my second year of the political science PhD program, professor Lily L. Tsai tapped me, along with several other graduate students, to participate in a `lab’ class where we worked with development practitioners and policy makers to co-design research projects. At the time, few political science departments had formally adopted this model. We began examining questions like whether greater civic skills can empower citizens to hold government accountable in the Philippines and how the Liberian government could better encourage citizen compliance during the Ebola crisis? Few academic departments are involved in structured collaboration with NGOs and governments to investigate questions that are both theoretically interesting and have practical policy implications. At GOV/LAB we seek to fill this void.

To bridge the gap between citizens and government, as GOV/LAB seeks to do, we explore elite motivations, government responsiveness in serving its citizens, as well as citizen desires, demands, and actions. These inquiries require constant innovation of tools and techniques for data collection and experimentation. For example, I designed an experiment to investigate whether social pressure explains high voter turnout among poor Tanzanian citizens. I presented participants with hypothetical political candidate profiles using a method known as conjoint analysis. Respondents were shown these candidates either while in the presence of fellow community members or in the privacy of their own home. They then decided if they wanted to vote or abstain. I found that the presence of peers increases the probability that participants will vote in the experiment—preliminary evidence of social pressure to vote.

My research has greatly benefited from the ideas and honest feedback that I get from other lab members. At lab meetings, we share updates on our research, poll the group for help with a sticky design puzzle and offer health-and-safety tips, for example—optimal road safety in the field. Without the precisely controlled environment of a test tube, our explorations into social and political questions often run into unforeseen challenges. We overcome these obstacles through consultation and collaboration. GOV/LAB demonstrates that rigorous scientific research can be accomplished in the absence of white lab coats, beakers, and sterile environments.

To learn more about ongoing GOV/LAB projects and research, check out our website ( We welcome new ideas, partnerships, and feedback.