Computational model of engagement with stigmatised sentiment: COVID and general vaccine discourse on social media

Abstract

The growth rate of new social media users continues to surpass new Internet users and new unique mobile phone subscribers and this trend remains consistent over the past 5 years (2019–2023). The most frequently visited types of websites or apps worldwide are chat and messaging, closely followed by social networks and this trend has also remained relatively constant. The dominating role of social media, especially as a source for information seeking, is staggering, particularly during the COVID-19 pandemic. However, the research in Keller et al. (J Mec Internet Res 16:e8, 2014) indicates that not many experts consider social media as a tool for sharing their expertise or for integrating social media into their research efforts. This is a troubling fact, especially considering that stigmatised health narrative are fueled in the face of uncertainty and spread very quickly among the lay population. The latter contributes to the spread of misinformation and, consequently, fosters hesitancy about preventive measures such as vaccines. This research presents new evidence on engagement with stigmatised vaccine discourse on Facebook (Meta), Twitter (X), YouTube and Reddit. Engagement with health-related sentiment can be an important indicator of perceptions regarding preventive measures. The current research can draw the attention of public health experts to the connection between stigmatised discourse and engagement in health discussions, as well as the potential impact of other linguistic features on engagement. It can also guide health authorities and medical professionals in developing effective communication strategies for the general public. Moreover, there are only a handful of studies discussing engagement with socially shared health-care discussions. The research focuses on examining engagement with stigmatised sentiment in vaccination discussions during and prior to the COVID-19 pandemic, using a cross-sectional approach. The study is based on primary data from social media domains, supplemented by secondary data analyses of literature related to the topic. To achieve the research goal, the study employs a multi-method design primarily based on quantitative methods of content analyses, such as Linguistic Inquiry and Word Count (LIWC) (Pennebaker et al. in Linguistic inquiry and word count: Liwc 2001, vol 71. Lawrence Erlbaum Associates, Mahway, 2001) to extract general language features and analyse stigmatised sentiment as the result of previous research findings in Straton et al. (Stigma annotation scheme and stigmatized language detection in health-care discussions on social media. In: Proceedings of The 12th Language Resources and Evaluation Conference (LREC, 2020), pp 1178–1190 (European Language Resources Association, 2020), Straton (Appl Intell, pp 1–26, 2022). Unsupervised K-means clustering methods, along with F-score and Z-score measures, are applied to draw insights from the features. The goal is to explore the phenomena of engagement with socially shared health information, investigate whether stigma can affect the engagement, and identify any other language features that may have an impact. Highly engaging general vaccine discussions before the pandemic appear to be more planned and less emotional. They exhibit reasoning and differentiation typical of more complex discussion sections, incorporating references to authority and family concerns simultaneously. The engaging messages convey stigmatised sentiment, likely shared either with the intent to deceive, or are based on a genuinely biased belief. Some elements of stigmatised discussions indicate deception, while others are more frequent in truthful statements. Engaging content in COVID vaccine discussions, similar to general vaccine content, lacks emotional elements and exhibits high linguistic complexity. However, there seems to be no connection between engagement and stigmatised sentiment. This difference is likely due to the removal of public anti-vaccination pages during the pandemic where anti-vaccination groups moved to discussion section of posts that try to disprove vaccine conspiracy theories. Mixing polarised groups will result in different feedback and engagement with the content. Almost half of the features in COVID and general vaccine discussions have a similar impact on engagement. However, further exploration of the findings suggests that there are distinct differences. Understanding the differences is very important to prevent generalisation errors in the conclusions drawn from the data. Engagement with written text on social media should be measured within a narrow scope: limited regional, demographic and temporal scope. Otherwise, interpretation of the findings risks on being inconsistent.