A Social Network Approach To Study Pro Vs. Anti-Vaxxers’ Mobilization On Facebook In France
Manon Berriche
Despite the strong scientific consensus on the safety and effectiveness of vaccines, lay people’s reluctance, if not refusal, to get vaccinated remains high. In France, for instance, vaccine-safety related sentiment has been found to be among the lowest in the world. Moreover, the new law extending compulsory vaccines for babies to eleven, instead of three, has recently divided the public opinion. In front of these different public health controversies, pro-vaxxers like anti-vaxxers have started to mobilize on social media to argue about this heated topic. As anti-vaccination messages began to thrive on Facebook, fact-checking outlets quickly reacted by frequently publishing correcting information aiming at promoting vaccination.
This study proposes to use social network analysis as a tool to question how Facebook pages and groups shape the audience engagement with vaccination-related information. Our goal was to examine how vaccination-related (mis)information may spur various opinions, as well as strategies of argumentation, counter-argumentation, and evaluation, depending on the online communities in which it has been diffused. We posited that Facebook pages with a varied editorial tone and a large audience will generate comments from both pro and anti-vaxxers, and will make both camps more likely to support their claims with arguments and external sources. Conversely, we also hypothesized that Facebook pages with a more homogenous editorial tone and a smaller audience will generate less diversity of viewpoints among the comments, and Internet users will be less likely to mobilize references and citations to defend their opinion.
To test these hypotheses, we first created a bipartite social network containing, on the one hand, a sample of Facebook pages that have published anti-vaccination content identified as misinformation by the French fact-checking outlet Les Décodeurs, and on the other hand, a sample of publications aiming at correcting misinformation about vaccination. As a result, we obtained a social network showing the Facebook pages that have diffused the most pro or anti-vax (mis)information and how each of them connected to the others depending on the type of information they shared. This network enabled us to define different clusters of Facebook pages having different ideological positions on vaccination. Besides, to better capture the editorial tone of each cluster, we also realized a semantic network based on the description of each Facebook pages. Then, drawing on these first findings, we conducted in-depth comments analysis to examine how the exact same vaccination-related information could spur different reactions from the public depending either on the size of the nodes that have shared it or its ideological coloration within the social network we realized.
Overall, this research underlines how network-based analysis can be relevant to study the way different digital conversational spaces could shape the audience engagement on public health issues, either by fostering or hampering lay people’s ability to argue and counter-argue. Finally, this study also suggests that complementing social network analysis with qualitative methods could be interesting to capture finer argumentative dynamics on social media.← Schedule