For 7-years, users in the thread Search for the last Big Secret: Mysteries of Shadow of the Colossus discussed how to find a videogame easter egg that never really existed. I wanted to know why they did so for so long; I hypothesized the vertical design of the PlayStation forum thread was at fault. I used open-source text processing libraries in Python to transform a massive discussion thread into a data set to study how users communicated with each other.

Outcome : By graphing thousands of user interactions I was able to see how the intake of new users in the thread drowned out the discussion of its most constant participants. These users figured out the secret, but the design of the thread didn’t allow them to productively let others know. My findings were peer-reviewed and published in the journal Communication+1.

In 2015, a protest started by a Facebook group forced the corrupt president of Guatemala to resign. Since then, the group has failed to maintain momentum, and I wanted to know why. I used open-source web-scraping libraries in Python to transform the Guatemalan Facebook-based #JusticiaYa activist group into a data-set to study how social media platform design educated its political strategy.

Outcome : Using graphing libraries to facilitate qualitative analysis, I found that the group’s attempts to promote political engagement were undermined by Facebook’s own definition of user engagement. #JusticiaYa had to address its audience in the terms dictated by Facebook’s suggestion algorithm, and did so at the expense of its goals.

First, I reviewed the department’s catalogue of exit interviews to better understand how previous programs had met the needs of different student segments. I used this to develop user-personas to guide my research and expand the student segments the program aimed to target.

I sat in during information sessions to take notes of recurring questions and problems that came organically as students discussed the program and their ability to register. Coding the data and organizing it through affinity diagram workshops with stakeholders, I produced an FAQ page for the program and identified possible paint points to explore in further research

Recruiting students that had demonstrated interest in the program but had not completed their application, I used a TEDW format to develop interview questions that would help flesh out our understanding of pain points in the registration process. Users got some coffee, I got some insight on key issues: 1. Departmental permissions differ by major, so the registration process was too variable. 2. International students were unsure of their work permission while attending university. 3. Information sessions took too long for already hectic graduate student schedules.