A14(b) - Perceptions, Political Assessment and Affect on Social and Traditional Media Digital Ecology Research Challenge Panel 2
Date: Jun 3 | Time: 03:45pm to 05:15pm | Location:
Chair/Président/Présidente : Mireille Lalancette (Université du Québec à Trois-Rivières)
Discussant/Commentateur/Commentatrice : Thierry Giasson (Université Laval)
Discussant/Commentateur/Commentatrice : Stéphanie Yates (Université du Québec à Montréal)
Session Abstract: This second panel will present some of the studies realized as part of the Digital Ecology Research Challenge. The project is a collaborative effort which aims to support research and innovative exploration into the ways in which digital media impact the 2019 Federal Election in Canada. Social media, online news, web, and survey data at a large scale were collected and used by the different researchers in order to answer their questions. This project has been made possible in part by the Government of Canada through a History Fund Grant. The project aimed at documenting different aspects of the election. The Digital Ecosystem Research Challenge is a collaborative effort which aims to support research and innovative exploration into the ways in which digital media impact the 2019 Federal Election in Canada. This panel will feature 3 selected and founded projects. Panelists will discuss issues related to citizens’ perceptions of public opinions polls, moral rhetoric and political assessment of politicians on social media, affect and propaganda using a cross-platform study of Twitter, Facebook public pages, Reddit and YouTube comments.
The Quality of Online Political News and Citizens' Perceptions of Public Opinion Polls: Jean-François Daoust (University of Edinburgh), Frédérick Bastien (Université de Montréal)
Abstract: Citizens get a growing proportion of their news from online sources (Digital News Media, 2019). Whereas the “broadcast news regime” (Williams and Delli Carpini, 2011) was dominated by news organizations whose credibility and professionalism provided some guarantees about the quality of information, the digital ecosystem increases citizens’ responsibilities in this process. In times of elections, this responsibility appears to be particularly important.
We aim to improve our understanding of the Canadian digital ecosystem in two ways. First, we analyze the average causal effect of the credibility of established news organizations when they provide political news compared to ordinary citizens who can now easily access the digital public sphere and share information. Second, we examine whether the presence of different methodological information included in pre-elections polls forecasts affects citizens’ understanding of the state of the race. We rely on a six-treatment experiment embedded in the 2019 Canadian Election Study.
This study entails important implications about the conditions under which citizens trust and understand (or not) pre-election polls results. It is also very fruitful for more ‘practical’ issues regarding the potential harmful consequences of election polls and its regulation by policy makers – especially given that restrictions are on the rise around the world (Frankovic et al., 2018).
Not my Prime Minister! Moral Rhetoric and Political Assessment Through the Lenses of Social Media: Lalancette Mireille (Université du Québec à Trois-Rivières), Tania Gosselin (Université du Québec à Montréal), Simon Fitzbay (Université du Québec à Trois-Rivières)
Abstract: As Canadians increasingly rely on online content for information and entertainment, the election campaign is a key time for political candidates to reach citizens through digital social media on the one hand, and for citizens to discuss political issues on the other. This research project aims to better understand their content and influence through quantitative and qualitative big data analysis of data from traditional, social media (Twitter, Reddit and political memes). By adopting the perspective offered by the Moral Foundations Theory, our study will 1) identify the uses of moral and emotional arguments by political candidates and citizens; 2) assess the impacts of this moral rhetoric on the justification of political preferences. This research is innovative because almost all studies on the moral character of rhetoric focus on the United States. By looking at Canada, it will be possible to observe its mobilization in a less polarized political-media system. In addition, social media content analyses related to the policy frequently include Twitter and less often Reddit. The inclusion of Reddit in the corpus allows the analysis to be extended to a forum that has received less attention. Finally, while studies on the links between media content and public opinion are quite common, they are less common in comparing media content with political attitudes or preferences expressed in the form of answers to open-ended questions.
Affect, Elections, and Social Media: Understanding the Emotional Dimensions of Propaganda: Megan Boler (University of Toronto)
Abstract: This paper shares the mixed-methods findings from our past year of funded research, "Affect, Elections, and Social Media: Understanding the Emotional Dimensions of Propaganda." Using mixed-methods of sentiment analysis and discourse analysis, this study aims to explore: RQ#1: What emotions, feeling rules, and deep stories are associated with expressions related to racial and national identity and belonging, in the context of the 2019 Canadian election? RQ#2: What are political functions of these emotional expressions within social media communications, (particularly in terms of how such emotions amplify circulation of stories, foment intensity of debate)? RQ#2a: How do the above functions differ, depending on political perspective/camp? RQ#2b: How do these functions differ, depending on the social media platform in question?
Our research tracked the presence of "deep stories" and "feeling rules" (Hochschild, 1987, 2016) across social media surrounding the 2019 Canadian election.
Using affective computing, critical discourse analysis, and semi-structured interviews, the study examines content produced by lone individuals and individuals linked to grassroots social movements. This cross-platform study examines Twitter, Facebook public pages, Reddit and YouTube comments, focusing specifically on which emotions are expressed and how emotions “stick” and circulate in relation to narratives of racial and national identity. The study compares which emotional expressions characterize political messaging on the right, center, and left of the political spectrum. The research aims to contribute a critical theoretical approach grounded in the social sciences and humanities.