May 2, 2024

Data Visualizations: When More COVID-19 Data Doesn’t Equal More Understanding

As data visualizations increased to prominence early in the pandemic, Lee and her associates set out to understand how they were being deployed throughout the social media universe. “A preliminary hypothesis was that if we had more data visualizations, from information gathered in a systematic way, then individuals would be much better notified,” says Lee. Lee states this computational technique gave them a broad view of Covid-19 data visualizations. The combined techniques research “advances our understanding of data visualizations in forming public perception of science and politics,” states Jevin West, an information scientist at the University of Washington, who was not involved with the research. Lee describes the teams research study as a first action in making sense of the role of information and visualizations in these wider arguments.

MIT researchers discovered that Covid-19 skeptics on Twitter and Facebook– far from being “information illiterate”– frequently use sophisticated data visualization strategies to argue against public health precautions like mask requireds. Credit: Jose-Luis Olivares, MIT
Social network users share charts and charts– frequently with the exact same underlying data– to advocate opposing approaches to the pandemic.
Considering that the start of the Covid-19 pandemic, charts and charts have helped communicate info about infection deaths, vaccinations, and rates. In some cases, such visualizations can encourage habits that lower virus transmission, like using a mask. The pandemic has been hailed as the breakthrough moment for data visualization.
But brand-new findings recommend a more complex picture. A research study from MIT demonstrates how coronavirus doubters have marshaled data visualizations online to refute public health orthodoxy about the advantages of mask requireds. Such “counter-visualizations” are typically quite advanced, utilizing datasets from main sources and advanced visualization approaches.

The researchers combed through numerous thousands of social media posts and discovered that coronavirus skeptics often release counter-visualizations alongside the exact same “follow-the-data” rhetoric as public health professionals, yet the skeptics argue for significantly various policies. The scientists conclude that information visualizations arent enough to convey the seriousness of the Covid-19 pandemic, since even the clearest charts can be interpreted through a variety of belief systems.
This figure reveals a network visualization of Twitter users appearing in the research. Color encodes neighborhood and nodes are sized by their degree of connectedness. Credit: Courtesy of the scientists
” A great deal of individuals think about metrics like infection rates as objective,” states Crystal Lee. “But theyre plainly not, based upon how much dispute there is on how to think of the pandemic. Thats why we state data visualizations have ended up being a battlefield.”
The research will exist at the ACM Conference on Human Factors in Computing Systems in May. Lee is the research studys lead author and a PhD student in MITs History, Anthropology, Science, Technology, and Society (HASTS) program and MITs Computer Science and Artificial Intelligence Laboratory (CSAIL), in addition to a fellow at Harvard Universitys Berkman Klein Center for Internet and Society. Co-authors include Graham Jones, a Margaret MacVicar Faculty Fellow in Anthropology; Arvind Satyanarayan, the NBX Career Development Assistant Professor in the Department of Electrical Engineering and Computer Science and CSAIL; Tanya Yang, an MIT undergrad; and Gabrielle Inchoco, a Wellesley College undergrad.
As data visualizations rose to prominence early in the pandemic, Lee and her colleagues set out to comprehend how they were being deployed throughout the social networks universe. “A preliminary hypothesis was that if we had more information visualizations, from data gathered in a methodical method, then individuals would be better informed,” says Lee. To check that hypothesis, her group mixed computational techniques with innovative ethnographic approaches.
With those tweets, the scientists produced a network chart to discover out “whos retweeting whom and who likes whom,” says Lee. The scientists found that antimask groups were producing and sharing data visualizations as much as, if not more than, other groups.
And those visualizations werent sloppy. “They are practically indistinguishable from those shared by traditional sources,” states Satyanarayan. “They are typically simply as polished as graphs you would expect to come across in information journalism or public health control panels.”
” Its an extremely striking finding,” states Lee. “It shows that characterizing antimask groups as data-illiterate or not engaging with the data, is empirically false.”
Lee states this computational approach gave them a broad view of Covid-19 information visualizations. “What is really interesting about this quantitative work is that were doing this analysis at a substantial scale. Theres no chance I might have checked out half a million tweets.”
“I think it misses out on a lot of the granularity of the conversations that individuals are having,” says Lee. For that, the researchers turned to a more traditional anthropology research study method– with an internet-age twist.
Lees team followed and examined discussions about data visualizations in antimask Facebook groups– a practice they called “deep hiding,” an online version of the ethnographic technique called “deep hanging out.” Lee says “understanding a culture requires you to observe the everyday informal goings-on– not simply the big official events. Deep hiding is a method to transpose these traditional ethnography approaches to digital age.”
Antimaskers on Facebook werent shunning information. Rather, they went over how different kinds of information were gathered and why. In reaction, members of the group would typically develop their own counter-visualizations, even instructing each other in data visualization techniques.
” Ive been to livestreams where people screen share and look at the information portal from the state of Georgia,” says Lee. “Then theyll talk about how to download the data and import it into Excel.”
Jones states the antimask groups “concept of science is not listening passively as experts at a location like MIT tell everyone else what to believe.” He adds that this kind of behavior marks a new turn for an old cultural existing. “Antimaskers usage of data literacy reflects ingrained American worths of self-reliance and anti-expertise that date back to the founding of the nation, however their online activities push those values into new arenas of public life.”
He adds that “making sense of these complicated dynamics would have been difficult” without Lees “visionary leadership in masterminding an interdisciplinary cooperation that spanned SHASS and CSAIL.”
The mixed approaches research study “advances our understanding of information visualizations in forming public perception of science and politics,” states Jevin West, an information researcher at the University of Washington, who was not included with the research. Data visualizations “bring a veneer of objectivity and clinical accuracy. However as this paper shows, data visualizations can be used successfully on opposite sides of a problem,” he says. “It underscores the intricacy of the issue– that it is not enough to simply teach media literacy. It needs a more nuanced sociopolitical understanding of those producing and interpreting information graphics.”
Combining computational and anthropological insights led the scientists to a more nuanced understanding of information literacy. Lee says their study exposes that, compared to public health orthodoxy, “antimaskers see the pandemic in a different way, using information that is quite comparable. I still believe information analysis is essential. Its certainly not the salve that I believed it was in terms of convincing people who believe that the clinical facility is not reliable.” Lee says their findings point to “a bigger rift in how we think of science and expertise in the U.S.” That very same rift goes through concerns like climate change and vaccination, where similar dynamics frequently play out in social media discussions.
To make these outcomes available to the public, Lee and her partner, CSAIL PhD trainee Jonathan Zong, led a group of 7 MIT undergraduate scientists to develop an interactive narrative where readers can explore the visualizations and discussions for themselves.
Lee explains the groups research as a primary step in making sense of the role of information and visualizations in these wider disputes. “Data visualization is not unbiased. Its not absolute. It remains in truth a exceptionally social and political venture. We need to be attentive to how individuals analyze them outside of the clinical establishment.”
Referral: “Viral Visualizations: How Coronavirus Skeptics Use Orthodox Data Practices to Promote Unorthodox Science Online” by Crystal Lee, Tanya Yang, Gabrielle D Inchoco, Graham M. Jones and Arvind Satyanarayan, 7 May 2021, CHI 21: Proceedings of the 2021 CHI Conference on Human Factors in Computing Systems.DOI: 10.1145/ 3411764.3445211.
This research study was moneyed, in part, by the National Science Foundation and the Social Science Research Council.