As a Beinecke Library Fellow, Sara Doan, Assistant Professor of Experience Architecture at Michigan State University, spent two weeks immersed in the archives of Yale University’s Beinecke Rare Book and Manuscript Library, uncovering materials to inform her forthcoming book, Visualizing Pandemics: A History of Data in Action, that focuses on public health communication and the history of persuasion during disease outbreaks.
More specifically, her research examines the rhetorical role of data visualizations in public health messaging, tracing how charts, graphs, and maps have been used to shape public understanding and influence governmental responses to crises. She is looking at patterns throughout history and the influence made by such factors as credibility, timing, emotional appeals, design, and shifts in scientific theory.
“The central question of the book is how do we persuade people to take action,” Doan said.
Drawing on a range of historical sources and activist materials, Doan’s book, which she hopes to publish in an academic press, examines visual strategies, spanning multiple centuries and epidemics, and the impact made by data visualizations on disease outbreaks.
During her fellowship, she spent time with several archival collections that provided key insights for each chapter of her book. The project begins with Victorian-era public health responses to cholera outbreaks in London.
One of the first uses of data visualization during pandemics occurred in the Victorian period when scholars created graphs to challenge the Miasma theory as the leading cause of cholera. The dominant scientific theory until the 1850s, the Miasma theory suggested that “bad air” or “bad smells” caused disease to spread
Doan’s chapter on this period explores the maps of cholera deaths and early data visualizations, which led to the realization that germs in water were behind the spread of the disease.
“Charts were a huge part of making that scientific paradigm shift happen in Victorian England,” Doan said. “The invention of public health is tied to the invention of charts and graphs and maps.”
“What does persuasion look like in action? What kinds of material conditions need to be met for people to even be persuadable? That’s important because right now misinformation and disinformation and a vacuum around health information and health literacy in general is one of the most pressing issues that we face in the 21st century.”
Doan also explored the 1918 influenza pandemic, a time when the first data visualization classes were taught in the United States. Yale holds the syllabus and course materials from one of those courses.
Doan read through the original research papers of the Yale scientist who discovered the “Influenza germ.” Although this discovery turned out to be false, the research papers provided evidence of data visuals being used in developing more effective vaccines to prevent influenza.
Doan also spent a large portion of her time during her fellowship analyzing the archival collection of ACT UP, an international grassroots political group working to end the AIDS epidemic, and other HIV/AIDS activists of the 1980s and 1990s.
“One of the crown jewels of Yale’s collection is the ACT UP files,” she said. “I looked through a lot of LGBTQ newsletters from the ’80s and ’90s to see how activists were creating maps, charts, and graphs of HIV/AIDS and trying to persuade the government to care about this disease.”
Because the United States government was slow to respond to the HIV/AIDS epidemic in the 1980s and 1990s, activists started creating treatment guides and drug listings to help people treat themselves. From this collection, Doan compiled 130 pages of notes on HIV/AIDS activist data visualizations. She was very moved by these documents, saying, “The most noble human impulse is to create things that save other people.”
The final chapters of Doan’s book focus on the first year of the COVID-19 pandemic when multiple governments were using outdated scientific theories instead of current science to study how the epidemic was spreading and how these information vacuums spurred new forms of data activism.
Doan’s research suggests that ineffective communication strategies often lead to an increase in infections. Antivaxxers and distrust in the government date back to Victorian cholera outbreaks and continue to today with the recent COVID-19 pandemic.
To make an effective data visualization in a pandemic, health communicators need to build trust and combat misinformation; further, data visualizations need to be accessible to all audiences, especially disabled people.
Science and data have persuasive potential, yet these examples show that data alone does not convince governments to take action or people to change behaviors.
“What does persuasion look like in action? What kinds of material conditions need to be met for people to even be persuadable? That’s important because right now misinformation and disinformation and a vacuum around health information and health literacy in general is one of the most pressing issues that we face in the 21st century,” Doan said.
Visualizing Pandemics will conclude with recommendations for how public trust can be built while navigating threats to people and public health.
Designing for Trust
In addition to her historical analysis, Doan brings a designer’s perspective to the study of persuasive data. Her research identifies several strategies that increase the effectiveness of charts and graphs.
“People think that we are persuaded by logic, but we’re actually persuaded by emotion, and by credibility and trust,” she said. “For instance, we might trust a family member who has a questionable approach to a pressing medical situation rather than trust a doctor.”
Doan emphasizes localization and accessibility as key design principles.
“People love to know about local stuff. Whenever there’s a map, people tend to find their house on the map or a place that they know. So, the more we can tap into people caring about where they physically are, that’s great.”
“People love to know about local stuff,” she said. “Whenever there’s a map, people tend to find their house on the map or a place that they know. So, the more we can tap into people caring about where they physically are, that’s great.”
Also crucial to creating persuasive, trustworthy data visualizations are good storytelling and a focus on designing for digital, disability friendly accessibility. The latter is so important to Doan that it is a central feature of her teaching at MSU.
“That’s a joy of my life, because we want the world to have fewer barriers for disabled people,” she said. “And when we lower those barriers for disabled people, everybody wins.”
Building Chart Literacy
Beyond academic circles, Doan wants to empower the public with better tools to read and interpret data.
“I’m writing a handbook chapter right now on misinformation and data visualizations around health and epidemics,” she said. “The first thing: make sure that what the title says about a data visualization matches what the data actually says.”
Doan encourages people to take their time as they explore data visualizations. Charts are excellent because they boil down an argument of a set of information to be read at a glance. However, that can just as easily become a weakness when an audience scans everything quickly and accepts it as fact, especially on social media.
“My work promotes human flourishing. How can we make good decisions? How can we empower people to get the right information? And how can we empower people to have the literacy to know what will help them?”
“Slowing down and being really thoughtful about what you’re sharing is one of the best ways to reduce misinformation,” she said.
One of the overarching goals of all of Doan’s work is to support health literacy and protect public trust.
“My work promotes human flourishing,” she said. “How can we make good decisions? How can we empower people to get the right information? And how can we empower people to have the literacy to know what will help them?”
By Austin Curtis and Bill Hodgkins
