ICT, Datafication, Covid-19, Social Listening and AI Technology in Development
Social Listening, Misinformation and Vaccine Hesitancy

Social Listening, Misinformation and Vaccine Hesitancy

Serious issues with social media have been discussed and in particular during the COVID pandemic that is still keeping the world in its grip. On this blog specifically, my colleagues have addressed for example the challenges with AI, algorithms and data manipulation, misinformation and data responsibility, privacy and ethics. But big data from social media also provides opportunities, by listening to the voice and opinions of people worldwide and this publicly available information can be crucial in responding to a crisis like the COVID pandemic, and its linked misinformation and vaccine hesitancy.

“Alongside public health measures, vaccination is key to controlling this pandemic”

WHO Director-General  Dr Tedros Adhanom Ghebreyesus, opening remarks at the media briefing on COVID-19 – 14 May 2021.

Hou et al (2021) sought to assess COVID vaccine hesitancy, confidence and public engagement of social media users in Sao Paolo, New York, London, Mumbai and Beijing. Using the meltwater tool, they collected individual twitter posts (no news sites etc.) between June and July 2020, a time where the actual vaccines were entering their clinical trials. Over 7000 tweets were manually coded by a framework based on the literature for vaccine hesitancy. Hou et al found that “among social media users, 36.4% (571/1568) in New York, 51.3% (738/1440) in London, 67.3% (144/214) in São Paulo, 69.8% (726/1040) in Mumbai, and 76.8% (2128/2770) in Beijing indicated that they intended to accept a COVID-19 vaccination. With a high perceived risk of getting COVID-19, more tweeters in New York and London expressed a lack of confidence in vaccine safety, distrust in governments and experts, and widespread misinformation or rumors. Tweeters from Mumbai, Sao Paulo, and Beijing worried more about vaccine production and supply, whereas tweeters from New York and London had more concerns about vaccine distribution and inequity. Negative tweets expressing lack of vaccine confidence and misinformation or rumors had more followers and attracted more public engagement online (2021; 1).”

The study recommends that health organizations should carefully listen to the concerns and sentiment of their communities and that vaccination campaigns should include those insights in their communication so that they can maximize their effectiveness and impact. As shown above, the sentiment in Mumbai is different from that in London, suggesting that different approaches should be developed. Studies like these reveal sentiments and allow organizations to change their education and awareness campaigns rapidly, and tailor it to the needs of the public/ community. This is particularly needed since tweets regarding vaccine confidence and misinformation generate more engagement (Hou et al, 2021). For organizations concerned with pre- or debunking misinformation, the social listening approach appears to be an effective, low-cost, and accessible method for that. Tapping into that wealth of information, allows organizations to tailor make their campaigns for their area/region.

The strength of this method in yielding relevant information is further reinforced by the Vaccination Demand Observatory, which is a partnership between Yale Institute for Global Health, the Public Goods Projects (PGP) and UNICEF, and one of those initiatives that have developed a real-time data dashboard that gives insights, advice and trends based on social listening. Access to the dashboard can be requested if you work in the public health sector, so that practitioners all over the world can have access to their advice, expertise, and publications. The VDO ‘launched in Fall 2020 to provide equitable, action-oriented, and customized guidance to social-listening informed vaccine communications. Spearheaded by UNICEF, we aim to build a sustainable network of country ‘infodemic managers’ supporting national immunization programs and their networks of community-based organizations’. Not only did they develop this highly useful dashboard, they also developed a Vaccine Misinformation Management Field Guide, which you can access here, and work together with Facebook to analyse the social engagement of different posts in different countries, using their insights for impact program to maximize visibility (Khan, 2021). A very insightful webinar on how the VDO adopts social listening comes from Angus Thomson, the Senior Social Scientist from UNICEF:

 

The world benefits from a vaccinated population, and in the words of the WHO director, is the only way of controlling the current situation. Social listening can be a valuable tool to assist in making aware communities. It allows us to learn, reinvent, and respond with our communications in a very short period. Social listening has also been applied in other Crisis situations, see for example the  https://www.globalfloodmonitor.org/ from de Bruijn et al (2019), which uses real-time data from twitter to compile a database of worldwide floodings, through something they refer to as using humans as sensors. Or learn from the outcomes of the research of Stewart & Wilson (2015), who studied the role of social media during a hurricane in the US, and based on it, introduce the potential STREMII model, a model to incorporate social listening in any organisation’s (crisis) communications.

The STREMII (STREAM-ee) model of social media crisis communication (Stewart & Wilson, 2015).

 

Social listening is not something to overlook in research, development, policy and communications – it rather should be thought of as a crucial element in any organisation’s strategy to innovate. An extremely helpful keynote speech covering the research process by Dr. Alan Shaw:

 

If the above triggered your curiosity and you are interested in exploring a free monitoring tool for your own organization, you may want to check out tools like Mention or Talkwalker Alerts. There are also paid plans that would offer you access to more/advanced features.

Do you have experience with, or more examples of social listening during the COVID pandemic? Please do share in the comments below!

Bib:

de Bruijn J, de  Moel H, Jongman B, de Ruiter M, Wagemaker J, & Aerts J. 2019. A global database of historic and real-time flood events based on social media. Scientific Data, 6:311. https://doi.org/10.1038/s41597-019-0326-9

Hou Z, Tong Y, Du F, Lu L, Zhao S, Yu K, Piatek SJ, Larson HJ, Lin L. 2021. Assessing COVID-19 Vaccine Hesitancy, Confidence, and Public Engagement: A Global Social Listening Study. J Med Internet Res 2021; 23 (6). doi: 10.2196/27632

Khan, K. 2021. Building Vaccine confidence amid Covid-19 with UNICEF. https://about.fb.com/news/2021/04/building-vaccine-confidence-amid-covid-19-with-unicef/

Stewart, M.C, Gail Wilson, B. 2015. The dynamic role of social media during Hurricane #Sandy: An introduction of the STREMII model to weather the storm of the crisis lifecycle.  Computers in Human Behavior 54. https://doi.org/10.1016/j.chb.2015.07.009

https://www.globalfloodmonitor.org/

https://vaccinationdemandobservatory.org/about