‘A feeding ground for misinformation’: Pitt researchers study vaccine hesitancy on Twitter


Photo courtesy of Beth Hoffman

Beth Hoffman is the project coordinator for “Twitter and COVID-19 vaccine hesitancy,” a project funded with a $117,000 grant from the Richard King Mellon Foundation studying how vaccine misinformation spreads on social media.

By Punya Bhasin, Staff Writer

The COVID-19 vaccines brought along with them not only a light at the end of the tunnel, but another topic for spreading misinformation about vaccines and COVID-19 on social media. Twitter now includes a “COVID-19 misleading information policy” section under its general user guidelines, but Pitt researchers hope to further detangle the web of lies in order to create effective public health messaging systems.

Pitt researchers at the Center for Research on Behavioral Health Care, Media and Technology received a $117,000 grant in January from the Richard King Mellon Foundation for their project, “Twitter and COVID-19 vaccine hesitancy.” This grant will fund the project through 2021 and will be used for research materials and partnerships with the data analytics company, VISIMO. Dr. Jaime Sidani, an assistant professor of medicine who studies the intersection between media and health, leads the project.

Beth Hoffman, a Ph.D. student in the Behavioral Community Health Science Department and the project coordinator for this research, said the aim is to determine how misinformation surrounding COVID-19 spreads on Twitter, which she termed “a feeding ground for misinformation.”

“We have seen that many of these popular social media platforms have enabled people to create large social groups within the platform,” Hoffman said. “They spread conspiracy theories, misinformation and illegitimate sources, which tends to be the root cause of the large amount of hesitancy and paranoia surrounding COVID-19 and the vaccine.”

Sidani said the goal of this project is to understand how hesitancy around the COVID-19 vaccine is spread by searching through tweets and analyzing how the tweets are broadcast to their target audiences.

We are looking at Twitter data to get an in-depth understanding of the reasons people are hesitant to accept a COVID-19 vaccine,” Sidani said. “We also want to see how these messages spread through social networks on Twitter and whether we can work with different communities to develop messages to increase vaccine acceptance.”

Hoffman said the team started its research in January after members saw an increase in misinformation about COVID-19 being spread online.

“We were noticing that in news reports, and in national polls that were being conducted related to the COVID-19 vaccine, that the numbers of people who said that they would take a vaccine were starting to go down,” Hoffman said.

Hoffman said with her experience in the field studying media and health along with Dr. Sidani, she felt as though they should investigate the impacts social media has on public health messages.

“We decided it would be valuable to see over the course of the year what people were talking about on social media with regards to the COVID-19 vaccine and see what people’s concerns are, and also how these concerns are changing over time,” Hoffman said.

Riley Wolynn, a sophomore health services major, works as an undergraduate student researcher for the project and is responsible for looking through Twitter posts and determining what people are posting about. She works with various other team members to take the findings and organize them into broader themes. Wolynn said the team hopes to gain more information about how these hesitations are conceived.

“We are hoping to find out who is talking about COVID vaccines, how they are feeling about getting a COVID-19 vaccine and the reasons they might be hesitant to get one,” Wolynn said.

Hoffman said the team identifies certain keywords that are commonly found in posts about vaccine hesitancy and COVID-19 misinformation, and quantifies these occurrences in a codebook.

“We’ve been going through the scientific literature as well as news reports to see what are some themes that people have already identified and then our undergraduate student research assistants can quantify how many times a keyword was mentioned,” Hoffman said.

Hoffman said she hopes the research will be able to effectively quantify the number of times certain phrases and keywords were mentioned in order to better visualize what concerns exist behind the hesitancy for vaccines.

“Through this quantitative analysis we can say, for example, 10% of tweets mentioned a conspiracy theory, and here were the conspiracy theories mentioned, and 10% of tweets mentioned concerns about safety, and these were the concerns,” Hoffman said.

According to Hoffman, the methodology used for this research is a social network analysis consisting of researchers constructing “webs” showcasing what specific information is spread, and which audiences are spreading and receiving this information.

“If somebody was going to do a social network analysis of a tweet that I sent, they would put me in the middle, and then anybody who retweeted my tweet would be connected to me and then maybe some of their followers who retweeted that tweet will be connected to them,” Hoffman said. “And if you can imagine that we do that for all the tweets we see, we start to build these networks.”

Wolynn said a large part of this research is not only to understand how misinformation and hesitancy spreads, but also to determine how public health messages can be spread more effectively.

“It is important to understand that public health research is vital in addressing misinformation and disinformation. We must understand the tactics of anti-vaccine activists if we want to keep them from continuing to spread false information,” Wolynn said.

Hoffman said by understanding the tactics used on Twitter to spread information to large platforms, the team can utilize those strategies to help effectively spread credible information.

“Our research will allow us to systematically look at what these health care providers are posting,” Hoffman said. “If it’s being retweeted or liked and how it’s being received by the intended audiences and as we go on with our research we will take what we see online, and use that information in conjunction with community partners to develop effective vaccine messaging.”

Dr. Sidani hopes this research helps the public health community to better understand the reasons for hesitancy surrounding vaccines and other movements and create more effective messaging to the public.

“I think it’s important to have in-depth analyses of why people are hesitant to receive this vaccine because this understanding will be important for increasing acceptance of the COVID-19 vaccine and it may have implications for other vaccines and future vaccines,” Sidani said. “My hope is that we can find ways to develop community-driven messages to help with vaccine acceptance and decide best how to facilitate these messages through social networks on Twitter.”

Wolynn likens the spread of misinformation to the virus itself and hopes people understand how dangerous the spread of misinformation can be.

“Misinformation is just like the virus,” Wolynn said. “It spreads uncontrollably, rapidly and globally, and has proven to be deadly.”