University of Rochester


Digital public health presents opportunities, challenges

September 19, 2016

Two researchers recently visited the University of Rochester to make the case for harnessing big data mined from social media to aid public and environmental health efforts.

King-wa Fu and Matthew R. Auer presented their insights during a seminar sponsored by the Center for Energy and Environment and the Goergen Institute for Data Science. Fu is associate professor at the University of Hong Kong’s Journalism and Media Studies Center (JMSC), and Auer is professor of environmental studies and dean of faculty and vice president for academic affairs at Bates College.

Fu and Auer shared results from various studies analyzing how social media platforms such as Twitter (or its Chinese equivalent, Weibo) provide useful sources of data for exploring public perceptions of health and environmental issues. Microblogging, in other words, can act as a window into people’s understanding of public issues. 

In one study, Fu and his colleagues found that social media amplified fear and anxiety about Ebola cases in the U.S., despite the negligible risk of infection for Americans. Now imagine a future in which public health professionals are equipped with digital skills: the Twitter data provide public health practitioners with a quantitative indicator of anxiety, anger, or negative emotions in the general public (at least in areas where Twitter has high penetration). Armed with this information, public health professionals can then help to alleviate that anxiety and correctly communicate the risks associated with Ebola or other infectious diseases, such as Zika.

In another case, Fu and Auer used open access data to investigate Weibo censorship in China. After a documentary about air pollution in China went viral, the researchers used Weiboscope (a tool developed by JMSC) to identify posts censored by the Chinese government. These included condemnations of the state and photographs of anti-pollution demonstrations. Notably, they also included public expressions of concern by ordinary citizens, which previous research showed to be generally tolerated by the government.

While applying computational social science research to public and environmental health concerns is laudable, Fu and Auer highlighted the limitations of this approach. First, data collection from Weibo is increasingly restricted. There’s also a concern about the representativeness of the data since a small number of bloggers create most of the posts. In addition, Weibo continues to lose market share to other, newer social media platforms, including private messaging apps like WeChat.

But as long as there is information to discover from the data, researchers like Fu and Auer—as well as those at the University’s Center for Energy and Environment and the Goergen Institute for Data Science—will continue advancing our understanding of the convergence of social media, public health, and environmental health.