Political Behaviour/Sociology

F19 - Media Coverage of Politics II

Date: Jun 1 | Time: 01:30pm to 03:00pm | Location: Classroom - CL 434 Room ID:15743

Chair/Président/Présidente : Jason Roy (Wilfrid Laurier University)

Discussant/Commentateur/Commentatrice : David Moscrop (Simon Fraser University)

Audiences Perceptions of House of Cards. The Role of Transportation and Identification: Alexandra Manoliu (Université de Montréal)
Abstract: A fictional narrative can persuade its viewers/readers through two main mechanisms: transportation and identification. Transportation into a narrative world is a complex psychological process, “a melding of cognitive, emotional and imagery in a story” (Green & Sestir 2017:1). Transportation is usually linked to identification, which can be described as the state where “a reader or audience member becomes one with the character in a story or a play” (Green et al 2002: 61). Could House of Cards create the experience of transportation and identification for its viewers? And if yes, what are their effects upon people's perceptions about the impact of the series on their lives? I am interested in seeing if after watching it, people have the impression it impacted them in a positive way, making them more knowledgeable and capable of understanding politics, or by increasing their interest on a topic they were not paying attention to before. The data I rely on was gathered through an online survey (via Reddit), immediately after the official release date of season 5 of House of Cards. Results seem to indicate that people are transported into the fictional world of politics and also identify with the main (negative) character- Frank Underwood. Those transported believe they understand better politics, they have gained real political information and have a higher interest in politics because of the series. Identification has no impact at all, while the perceived plausibility of the series (a dimension of realism) also affect those 3 dimensions.

Making Sense of Web Data for Public Opinion Research: Ludovic Rheault (University of Toronto)
Abstract: The web contains an ever-expanding quantity of textual data offering researchers a golden opportunity to study the evolution of public opinion in democracies. Yet making sense of such large collections of unstructured data for substantive research involves sensitive methodological steps. Minimally, these steps comprise the accurate recognition of political issues of interest, and the tracking of meaningful opinions (stances) on each of these issues. The present study relies on a feed of real-time data streamed from the Twitter microblogging platform during six consecutive months starting in 2017, filtered by language, geographical area and keywords. The dataset comprises several hundred million tweets and counting on the topic of politics written by Canadian and American users. I use neural network classifiers trained on a large corpus of annotated news articles to group these tweets by subtopic, and implement a new method of natural language processing to identify clusters of opinions on each issue, based on multi-word expressions and named entities. The study presents this methodology and uses the data to test a number of hypotheses explaining the response of public opinion to exogenous political events during the year, in particular reactions to terrorist attacks.

Separate but Equal? How the French- and English- Language Media Cover Contentious Electoral Issues: Randy Besco (University of Toronto), Marc-André Bodet (Université Laval), Erin Tolley (University of Toronto)
Abstract: Separate but Equal? How the French- and English- Language Media Cover Contentious Electoral Issues Linguistic diversity, cultural duality, and the relationship between Quebec and the rest of Canada have been longstanding preoccupations of Canadian political science. Despite this, the two solitudes prevail, with English Canadian academics, in particular, often bracketing off Quebec and French Canada from their analyses. This tendency is pervasive in studies of the media, which recognize there are likely to be differences between English- and French-language coverage, but which mostly focus on either one or the other. This paper aims to bridge that divide. To do so, we use automated text analysis to track media coverage of three contentious issues in the English- and French-language media during the 2015 federal election. We focus on the Syrian refugee crisis, the debate over the niqab, and the criticism surrounding so-called barbaric cultural practices, issues that encompass political and cultural debates that are germane to understandings of “identity” in both Quebec and the rest of Canada. Our paper provides empirical insights into differences in the media landscape, as well as a baseline for understanding the discursive environment in which these issues emerged on the political landscape.