Title: My Summer Internship: Gaining New Perspectives on Data in Development

I am writing this post after spending two weeks in the rural Rulindo district of Rwanda, where I supported the rollout of a baseline livelihoods survey to assess the impact of a commercial ginger cultivation project. Data we collected will be used to measure how growing ginger for export affects the incomes and livelihoods of Rulindo smallholder farmers and their families. The survey design was one of my main deliverables as the Social Enterprise Evaluation Intern at DAI, which implements the USAID-funded Nguriza Nshore project. This experience has been the most rewarding of my summer internship at DAI so far—but also the most challenging. Over the past months, I have been deeply immersed in the many small, necessary details of planning for the survey’s implementation. However, now that the data collection in Rulindo is complete, I find myself with space to step back and reflect more broadly on what I learned and observed during the process.

While Rulindo is located just two hours north of the capital Kigali by car, the contrast between the two regions makes the distance between them appear much greater. Reaching the project site entails turning off the well-maintained highway outside Kigali city limits onto a rutted dirt road that climbs steeply into the mountains. Kigali’s orderly neighborhoods and solid infrastructure give way to small, dusty communities scattered among terraced subsistence plots where access to electricity and clean water are far from universal. Rulindo’s poverty rate exceeds the national average, and the farmers we surveyed are among the district’s poorest. Many of those we interviewed lack the means to save money, eat more than once per day, or send their children to school. In Kigali, I have noticed that it is common to see Rwanda characterized as a definitive development success story. Without a doubt, the country has come a very long way in the years since the 1994 genocide, and it is clear that development is a priority. However, I found it increasingly challenging to reconcile Rwanda’s narrative of development success with the difficult living conditions that I observed while working in Rulindo.

Because the livelihoods survey will need to be repeated in coming years, my team tried to build sustainability into its design. Before data collection began, we selected a cohort of ten youth from the same Rulindo communities as the farmers we were interviewing and paired them with experienced enumerators from Kigali. Our plan was for the youth to interview the project farmers with the guidance of the experienced enumerators, helping to generate local data collection expertise for future surveys and build professional skills. We then developed a two-day training on data collection ethics, interview best practices, and mobile survey tool navigation—to be delivered in Kinyarwanda since the Rulindo youth were most comfortable using their native language. Since my Kinyarwanda knowledge is limited to very basic greetings and food items, I felt a little out of place during the first day of the training. Because of my limited communication abilities, my colleagues translated the training segments I delivered to Kinyarwanda and then translated the enumerators’ questions and comments back to English, a cumbersome process that made me question how my presence as an outsider was impacting the group dynamic.

However, this feeling quickly faded as the Rulindo youth, experienced enumerators, and our DAI team got to know each other better. To bridge our different backgrounds, my team and I strove to create an open and inclusive environment for the Rulindo youth to learn about good data collection practices while they educated us about local context. I drew upon the catalogue of icebreakers and team building activities that I learned while conducting workshops in my previous job to help facilitate group communication. After two weeks, while we had successfully collected the data we had set out to obtain, I was most pleased that the experienced enumerators had grown into their roles as supportive mentors, that youth data collectors were confidently leading interviews and managing complex data collection protocols, and that the enumerators from Rulindo and Kigali had coalesced into a unified, cohesive team.

After training concluded, we began implementing the survey. During the data collection period, each day began with a pre-dawn moto commute to DAI’s Kigali office, followed by a mad dash to sync the tablets used to record survey responses, print out paper surveys and forms for the enumerators, and double check the list of farmers to be interviewed. After a bumpy drive from Kigali to Rulindo, our team would aim to interview forty farmers each day before debriefing and heading back to Kigali in the evening. In Rulindo, my role focused predominantly on coordinating logistics, supporting enumerators with survey questions and the mobile data collection tool, and ensuring the quality of the data being collected.

Participating in the data collection process from the initial research design to interviews and analysis has shifted my perspective on data in development. Until now, much of the data I have used has been collected by others. While in concept I knew that each data point represented the experience of a real individual or household, in practice I have found that it can be easy to lose sight of the people behind the data while entering Stata commands or debating which type of regression is most appropriate. However, participating in the interview process makes it impossible to forget that behind each data point is a human being. In reviewing the data we collected, I find myself remembering the difficulties we faced in trying to neatly capture the details of a farmer’s experience in a survey question response. While we often assume that data reflects reality, working directly with survey respondents has reminded me that reality is messy and often does not fit perfectly into multiple choice categories. To better capture this complexity, my team also conducted qualitative interviews and rapid rural appraisal focus groups to supplement the quantitative survey and to explore community power dynamics. As we begin to analyze the data we collected, the formal and informal conversations we had held with farmers over the past weeks will provide critical insight into local challenges and motivations, helping us to accurately interpret the quantitative survey results.

Looking back with only several weeks remaining in Rwanda, I am grateful for all the people who have made my experience here a positive one. From my DAI colleagues who patiently answered my numerous questions about how to structure a survey to the Rulindo youth who welcomed me into their community and the moto drivers who helped me find my destination even when I did not know the right Kigali landmarks, I am deeply thankful to all those who went out of their way to support me as I slowly adapted to living and working in a new cultural context. When I pack my suitcase in a few weeks and head back to D.C., I hope to also bring with me some of the optimism, creative problem-solving skills, and openness to new ideas that I have encountered while in Rwanda. With any luck I will be back soon.