Reflections on the Evolving Ecosystem of Academic Data Science
July 24, 2023
Hello everyone! My name is Kristina Riemer. I’m an ecologist turned data scientist, and I currently lead the CCT Data Science team at the University of Arizona. We are a small group of data scientists and software engineers who help researchers implement data science tools and skills in their work. We engage with researchers through a variety of programming, including workshops, short courses, data science incubator projects, and longer-term collaborations. Most of the people we work with are in life sciences domains, including environmental science, nutritional sciences, and ecology.
I attended the ADSA Data Science Leadership Summit earlier this year to interact with and learn from other leaders in the area of academic data science. I was in particular looking for ideas about how different groups are integrated in their institutions, how they recruit and manage group members, and how different funding mechanisms work. It was really useful to see how some units that have been established for much longer, such as the University of Virginia’s School of Data Science and the University of Washington’s eScience Institute, have developed over time. I had further opportunities to interact with folks through the ADSA Career Development Network program, which brings together early-career folks who are working in the area of academic data science.
There were some notable aspects of other academic data science programs that stood out to me because they were new to me or differ a bit from how our group functions. While it is becoming common for academic institutions to have master’s degrees in data science, I learned that some institutions recently started to offer PhDs in data science! While we primarily use data science as a tool to improve life sciences research, some folks who get these higher degrees are advancing research in data science itself, but also a lot of those PhD programs include a concentration in a domain area like civil engineering or epidemiology.
One of the summit sessions was focused on lots of ongoing work to integrate data science training through K-12 education and also at community colleges. Due to the increasing amounts of data available in every field, improving data literacy and ability at all levels is crucial.
This is an area I do not know much about, so it was fascinating to hear all of these leaders talk about the substantial progress they’ve made and also some of the challenges they have dealt with. The Data Science for Everyone initiative was a great one making strides nationally in increasing K-12 data skills. There was also a lot of discussion about the (mostly logistical) difficulties of aligning community college curriculums with four-year institutions in order to improve the pipeline of data science education.
An unexpected takeaway of mine from some of the earlier summit sessions was how much of the work in data science at higher education institutions is being driven by faculty. Most of the data science institutes or schools of data science that the presenters represented seemed to work with or be composed of faculty members, as opposed to long-term staff. As our group is composed entirely of staff, some of our challenges differ from those of these units that are primarily faculty. We are focused on research support instead of developing our own research programs, so metrics for advancement and career progression are balanced differently. Consequently, there was a really interesting session on the challenges of hiring faculty specifically that helped inform how integrating faculty can require certain considerations. This also meant that there was not as much discussion of one major area that I am increasingly interested in, which is potential career paths in academic data science beyond faculty positions. This is a burgeoning area of interest that is being covered through other venues.
Overall, the ADSA Data Science Leadership Summit was a useful and fun event! Because it was a smaller and invitation-based conference, it was more straightforward to get to know lots of folks doing cool work. The summit was hosted by Boston University, which has a beautiful campus along the Charles River, with most of the events held at the very modern and new Center for Computing and Data Sciences building. Boston was a great place to go in late spring to escape the already oppressive heat of the south. The summit organizers also included optional visits to some of the local museums. I greatly enjoyed the brief break to explore the Museum of Science; the exhibit on illusions was probably my favorite!
I want to give a huge thank you to the folks at ADSA for organizing this event, and building such great community for academic data science. These types of initiatives really help this field grow and mature as we get changes to brainstorm and connect with each other. I’m looking forward to future interactions and great ideas!