Jun
28
Posted by randerson on June 28th, 2023
Posted in: Data Visualization
How do you demonstrate the cross-disciplinary nature of a $100 million-plus research institution with nearly 500 independent projects?
That was the question posed to Jess Newman McDonald as the University of Tennessee Health Science Center (UTHSC) began to update its strategic plan.
For Newman McDonald, turning a dynamic vision into a data visualization was key to creating an answer that would be digestible.
“Data visualization isn’t something I get to do every day, so it’s exciting when I’m able to dive into a project and it really becomes all encompassing,” says Newman McDonald, the research data and scholarly communications lead at UTHSC. “I especially enjoy exploratory visualizations — diving into a dataset without an agenda and just seeing what relationships emerge and what other questions you can ask.”
For her efforts extrapolating publication data as a proxy for identifying research areas of strength at UTHSC, Newman McDonald was awarded first place in the Complex Visualization category at this year’s Network of the National Library of Medicine Data Visualization Challenge, which was hosted by the National Evaluation Center. The winning visualizations can be viewed on Zenodo.
“It was inspiring to witness beginners and advanced data visualization creators challenge themselves — the submitters clearly invested significant time, energy, and creativity into their projects,” says Shanda Hunt, research data education and outreach librarian at the University of Minnesota, who judged entries alongside Chris Belter, program analyst at the National Institute of Child Health and Human Development. “The NNLM challenge is a great mechanism to encourage experimentation with data visualization.”
Tony Ding was on a vacation in New York City when he learned that he and Chen Yang had won the Bright Futures Award for their work visualizing the impact of COVID-19 alongside vaccination data from 2021.
“This achievement means a lot to us because we spent significant time building and perfecting this visualization and genuinely wanted to help improve COVID-19 awareness using our visualizations,” says Ding, a Health Data Science master’s degree student at Harvard University.
Ding and Yang, also a master’s degree student at Harvard, obtained their dataset from the New York Times and Our World in Data. The team chose to create their visualizations using Python’s Altair and Streamlit libraries, with drop-down selection bars, time/date sliders, multi-select widgets, and various types of linked views.
In reviewing 285 randomized clinical trials taking place in Africa, Tyler Nix, associate director of research and informatics at the University of Michigan’s Taubman Health Sciences Library, found just one was truly continent-wide, while many more were mislabeled as African in scope.
Collaborating with Michigan researchers Folafoluwa Olutobi Odetol and Marisa Conte, Nix aimed to illustrate the mischaracterization that can occur with research occurring in Africa.
“A map was the natural solution to convey the number and location, by country, of randomized controlled trials as found in PubMed,” says Nix, who was awarded first place in the Single Visualization category. His map accompanies an original research poster describing the project in more detail.
“Data visualizations are an effective way to translate research findings to broad audiences including policymakers, the media, and the general public,” says Hunt. “I think this is a critical skill for researchers to develop, especially given the trend toward a general distrust in science.”
For Newman McDonald, one goal was to let the dataset speak for itself.
“Exploring relationships within data can tell so many stories that would otherwise be invisible. Showing the interconnectedness of UTHSC research areas made a strong case for both the breadth of our research output as well as the unique intersections growing out of collaborations,” she says. “These collaborations are our strength, and the visualization served to solidify anecdotal evidence emerging from campus conversations as well as highlight quieter connections that could benefit from extra attention.”
Written by Roger Anderson