Activating the Digital Community for Disaster Response

Andrew J Schappert — Thu, 31 Oct 2024 18:45:37 +0000

Activating the Digital Community for Disaster Response Andrew J Schappert Thu, 10/31/2024 - 14:45

Who do people listen to during a crisis? In the aftermath of some disasters, misinformation runs rampant. Conspiratorial thinking, scams, and politicization make recovery work difficult, and push vulnerable people away from key resources. Communities that are traditionally wary of authority can be hard to reach.

Schar School professor Qian Hu

As seen immediately following Hurricanes Helene and Milton, rumors about FEMA malfeasance spread like wildfire through X, formerly Twitter. Crisis communications are part of a large social media ecosystem, rather than a one-way information flow from government agencies to residents. Recent research from ��AV’s Qian Hu tackles the challenge of examining this entire system and recommends that government agencies change how they spread the word during an emergency.

“Social media presents challenges for government such as reputation management and monitoring the spread of rumor and misinformation,” said Hu, a professor in the Schar School of Policy and Government. According to its analysis of more than 1 million crisis communication tweets during Hurricane Irma in 2017, Hu’s research team recommends that governments build networks for sharing information with individuals, businesses, and nonprofits before a crisis hits. By doing so, key information may reach constituents more effectively during disasters.

Hu has published regularly using Twitter/X data, including a Public Administration Review study of the politicization of crisis communication. She discussed how this project’s approach is different than previous research.

“Using social media to connect everyone involved in crisis communication is very important,” she said. “We need more studies on how nonprofit organizations, businesses, citizens, and emergent groups communicate during crises on social media.”

Unlike previous studies, Hu’s team cast a wider net to examine tweets from all types of users during and after the storm. Their analysis highlights the role that nongovernmental organizations and even private individuals can play in sharing information and facilitating humanitarian response.

“They employed a full-network approach, which is not used often,” said Jason D. Rivera, associate professor at the John Jay College of Criminal Justice, whose own work examines the experiences of minority and historically marginalized groups in relation to disaster and emergency management. “It’s not often done predominately due to the logistics of dealing with all the potential actors within networks—they looked at over a million tweets.”

By analyzing a diverse and broad collection of Irma-related tweets, rather than focusing solely on governmental communication, Hu’s research demonstrates the substantial role private citizens and nonprofit organizations played in sharing information and combating misinformation during Irma. Nonprofit organizations took center stage in online communication during recovery, highlighting local resources, organizing with victims, and facilitating a broader community-driven recovery effort. Influential citizens like journalists and celebrities employed their large follower base to amplify messages posted by official sources.

Hu’s work suggests that by bringing these diverse voices into the fold ahead of time and vetting their reliability and potential reach, governments can improve the speed at which messages are disseminated. Vital information can be targeted to reach victims faster, even in traditionally marginalized communities. Rivera’s research supports Hu’s recommendations.

“Sometimes we don't have organizations that are trusted by particular populations,” Rivera said. “However, there are individuals who do have the confidence of the community. If those individuals disseminate information through their own personal network channels, these communities can be better provided with the information they need at the most critical times.”

This research highlights a set of untapped nodes that can aid in trustworthy information dissemination, speeding government efforts during time sensitive communications such as hurricane evacuation orders.

“Public agencies should partner with nonprofit organizations to improve communication and address rumors,” Hu said. “They need to include these organizations in emergency planning and social media strategies before disasters hit. These organizations can help spread critical information to local communities and combat misinformation.”

Social media creates vast and emergent networks that are not easily measured and understood. Hu’s work toward understanding the flow and content of information across the whole network, rather than just a narrow slice, is key to developing effective crisis communication strategies. Her research helps readers appreciate that, as disasters like Irma, Helene, and Milton occur more often, it’s time to lean on the community and build a communication coalition, rather than attempting to brave the storm alone.

Qian Hu, Seongho An, Naim Kapucu, Timothy Sellnow, Murat Yuksel, Rebecca Freihaut, Prasun Kanti Dey “Emergency communication networks on Twitter during Hurricane Irma: information flow, influential actors, and top messages” in Disasters.

In This Story

Qian Hu

Learn more about the Schar School

Request Schar School program information

Topics

Schar School of Policy and Government

Schar School Featured Stories

Schar School News for October 2024

Research

Strategic Direction

Schar School Welcomes Thema Monroe-White

Andrew J Schappert — Wed, 04 Sep 2024 15:54:33 +0000

Schar School Welcomes Thema Monroe-White Andrew J Schappert Wed, 09/04/2024 - 11:54

In This Story

Thema Monroe-White

Request Schar School program information

Learn more about the Schar School

‘We’ve talked about the data harm. I want to get at the solutions.’ – Thema Monroe-White

Thema Monroe-White joins ��AV this fall as an associate professor in both the Schar School and the College of Engineering and Computing. She is an artificial intelligence (AI) and policy justice scholar and the recipient of a prestigious $1 million CAREER grant, awarded by the National Science Foundation (NSF) to early-career faculty who are trailblazers within their field.

Monroe-White focuses on the intersection of science, policy, and racial equity, working to combat bias within data science and AI by building a diverse scientific workforce. Her CAREER grant, entitled “Investigating Undergraduate Student Persistence Intentions for a Diverse Data Science Community,” is a huge step toward her goal of developing social inclusion in the scientific community. Monroe-White will investigate which interventions are most effective in encouraging underrepresented students to choose STEM majors, graduate, and subsequently stay in STEM careers. “I want to develop concrete, practical approaches for broadening participation in data science and related fields,” says Monroe-White. “Why do students persist in STEM, and how can we codify those conditions to encourage even more underrepresented students to pursue STEM careers?”

The CAREER grant builds on another NSF-funded project Monroe White brings to Mason, which examines how scientists’ background and experience influence the research topics they choose to pursue. Monroe-White's team has illustrated that a homogeneous workforce in STEM fields results in a less comprehensive exploration of scientific topics. Just as medical research focused on white male subjects until more women and researchers of color achieved authority in the field, Monroe-White's research exposes similar patterns in other fields. Preliminary research focused on scientists in the United States, and the team will now refine their algorithm and apply it to an international dataset.

Monroe-White brings her expertise in AI to George Mason at time when the university is investing in AI research and innovation. The Schar School recently secured two other seven-figure AI-related grants, one from the Department of Defense and another to create the nation’s first Center for AI Innovation for Economic Competitiveness (CAIIEC).

“Thema Monroe-White’s research understands the computational, social, and policy issues that underline artificial intelligence,” said J.P. Singh, Schar School Distinguished University Professor and principal investigator for CAIIEC. “She will greatly help us bridge disciplinary divides and train the next generation of student scholars to address the cutting-edge issues of our times.”

Monroe-White's research and teaching agenda focus both on revealing algorithmic bias and realizing AI’s capacity for positive social transformation. This will be the center of her contributions to George Mason’s new undergraduate program in technology policy, an innovative split between computer science and policy studies. In her fall course, AI: Ethics, Policy, & Society, Monroe-White asks, “Must AI tools reflect reality?” Algorithms learn from the data they are given and reflect the biases embedded in that data.

“We’ve talked about the data harms. I want to get at the solutions,” she said when asked about her goals at George Mason. “How do we round that corner and, instead of stereotyping subordination and omission, how do we get at empowerment and uplift?” AI is a powerful tool – one that has caused harm but also has enormous potential to reverse previous harms. As part of the Schar School, Monroe-White will train the next generation of computer scientists and policy practitioners to build that future.