Where Stats and Rights Thrive Together

summer interns at HRDAG whiteboard, 2018I first heard about HRDAG my sophomore year of undergrad from a professor in my university’s political science department. At the time I knew that I wanted to combine my love for math and programming with my interest in studying world issues. HRDAG provided me with the perfect example of a combination of those two seemingly disparate passions. I bookmarked their website and visited it frequently, especially in the moments when I was feeling uninspired by my coursework. I looked forward to one day reaching out to Patrick to learn more about how I might get involved with the organization, but I didn’t realize at the time how soon that opportunity would come.

If I’m being honest, I wasn’t completely certain what to expect when I first arrived in San Francisco, but like Erin said, it quickly became clear to me that mentorship is deeply ingrained in HRDAG’s culture. No one expected me to have all of the answers, and everyone was willing to help in any way that they could. Alongside Erin, I too spent my first few weeks getting acquainted with Patrick’s paradigms for principled data processing, which helped me to better understand the practice of reproducible research. I learned make basics, became well versed in version control, and got into the habit of regularly writing tests for the software I was writing. Gus was also a crucial part of this process—he taught me all of the cool vim tricks I know and was always there to troubleshoot when I got stuck, often preventing me from, as we would say, “running the lawn mower over my toes.”

Aside from learning about HRDAG’s practice of reproducible research, I spent a majority of this time working towards a recalculation, with the inclusion of a new dataset, of HRDAG’s Multiple Systems Estimation (MSE) estimates for the number of casualties that occurred in Guatemala from 1960-1996 during the internal conflict. This project came with a lot of existing code, so when I was first put on it, I spent a full day drawing a diagram of the project repository on the whiteboard; it would be the first of many diagrams, charts, and lists that I would put onto the whiteboard during my time in San Francisco. The first part of the project seemed straightforward—process the data so that it’s ready to be matched—especially because there was code to work off of already. As I would soon find out, things are never as simple as they originally seem… . Language encoding issues emerged as a recurring nightmare, record filters failed hard, fast, and often, and natural cross-dataset standardization was hard to come by. The office whiteboard and I really bonded during this process, and Erin and Gus were always there for brainstorming sessions, emotional support, and snack breaks. In my coursework I’m all too accustomed to well-behaved datasets that are ready for model fitting right out of the box. The Guatemala data is anything but that and presented me with challenges I had never even needed to think about in the past. Sometimes this meant that the lawn mower came hurtling towards my toes, but it also allowed for real growth as a data problem solver.

Of course, there are so many things to celebrate about spending the summer in San Francisco with HRDAG in addition to the engaging technical work. The most important being the people; the door opening in the morning was always my favorite part of the morning. Everyone was so kind right from my very first correspondences. When I arrived to the office for the first time I felt welcomed immediately, and attending retreat at the beginning of the summer only solidified this feeling. As the summer went on I found myself often thinking about how lucky I am to have collaborators like those at HRDAG. Everyone I had the pleasure of interacting with enriched my summer in some way. A special thanks is due to Erin, Gus, Megan, Kristian, Patrick, and Suzanne, whom I had the pleasure to see, chat, and collaborate with regularly.  

One of the highlights of the summer was attending the Joint Statistical Meetings (JSM) in Vancouver with Erin, Megan, Kristian, and Patrick. It was my first time attending a conference devoted entirely to statistics, and it was encouraging to see statisticians from so many different disciplines and professions together celebrating the field. Throughout the conference I had the opportunity to attend a number of engaging talks and presentations on a variety of topics including: novel methods for imputing missing data, best practices for reproducible data analysis, and advances in record linkage. I was also excited to sit in on the meeting of the ASA Committee on Scientific Freedom and Human Rights (chaired by Megan), celebrate Patrick’s receipt of the Karl E. Peace Award for Outstanding Statistical Contributions for the Betterment of Society, and catch up with other members of the HRDAG team. Outside of conference activities I had some time to explore downtown Vancouver and the food scene. I’m still thinking about the ice cream that Erin and I got on our last night there…

When I first started studying statistics I worried that I would never fit in, that I might one day need to choose between being interested in mathematics or social issues. At HRDAG, however, these dual interests were celebrated and nurtured and I emerged with a new perspective on how these passions can be combined to address grave instances of injustice with rigorous science. I returned to campus for my last year of my undergraduate program both humbled and inspired by my time in San Francisco and thankful that I’ll be able to continue working with the team over the next year. As I consider my future plans, I’m constantly thinking about new and innovative ways to continue work in the realm of human rights statistics, building on the foundation I laid this summer, and I feel so fortunate to have such a wonderful group of mentors and friends helping me along the way.