This continues our series of student reflections and analysis authored by our research team.
Reflections from a first-semester team member
Emma Lovejoy
I came on board with tPP in August 2019, and I think the project is amazing. The issues we are investigating, the information we’re accessing, and the things our data can be used for in the future, all really exciting things. As a brand-new coder though, there have been challenges with developing confidence in my work.
The biggest hurdle has been cultivating any sense of ownership: I don’t have an academic background that’s very applicable, and picking up the mechanics of coding and information management didn’t come naturally to me. The process of doing so didn’t leave a lot of time or space for getting a working understanding of the data we already have.
Even after months looking at our spreadsheets, I don’t know how clearly I could talk about the data as it looks today. That disconnect, while minor in the abstract, in practice has caused my work to come slower than it could, since every decision I come across (from coding to excluding cases) feels like something I’m not qualified to do. It’s imposter syndrome, in a word, but I think there are ways to help new members prevent and overcome it in years to come.
In our team’s closing discussions we debated a restructuring of the roles for future student participants as a possible the first step. Having smaller student groups with distinct goals and responsibilities, such as finding cases, updating existing cases, coding from scratch, and even document management with no data entry, would allow people to self-select roles that compliment their skill sets. I really think allowing specialization like this without first requiring a certain volume of coding would help students to approach the project with confidence, rather than having to ‘wait and see’ if they find something to do that they enjoy and are good at. While there is tremendous value in having diverse skills and proficiencies, being able to gravitate toward things that are instinctively interesting might provide a more accessible starting place.
A consistent, perhaps self-paced training program has also come up in brainstorming for the future. There’s definitely something to be said for everyone receiving the same orienting information when they come on board, and creating a digital program would allow more hands-on practice before actually touching a new case. I think it’s likely that extra practice at whatever task one’s focus will be on would greatly reduce unnecessary mistakes that come of second guessing, and to some degree increase productivity for the same reason: choices are faster and easier to make when you’re confident in your understanding.
I’m sure at this point someone has decided I’m just complaining, and that if I’d spent more time with the data or the manual or the codebook I wouldn’t have struggled with these things. And that may well be true for people with a background in data science. But, coming to it as a new skill to learn, and primed in a much more narrative style of research, it’s taken me longer to get accustomed to the simultaneous speed and specificity with which we need to address each case. I think the changes I’ve mentioned, as well as any number of ideas that have and will come up in the brainstorming process, will help to streamline the work of tPP as it continues to grow, and new people continue to come on board.