F21(a) - New Methods for Studying Political Behavior
Date: Jun 6 | Time: 03:15pm to 04:45pm | Location: SWING 307
Chair/Président/Présidente : Eric Merkley (University of British Columbia)
Discussant/Commentateur/Commentatrice : Alessandro Nai (University of Amsterdam)
More than Meets the Eye: Insights and Opportunities in the Study of Visuals and Digital Political Communication: Mireille Lalancette (Université du Québec à Trois-Rivieres), Vincent Raynauld (Emerson College) Abstract: Uses of still and moving-image content for political communication, mobilization, and organizing in the social mediascape has intensified and diversified over the last decade. On one hand, established political elites have used digital media tools such Instagram and Snapchat to reach out to and engage with citizens and news media outlets. On the other hand, members of the public have deployed novel strategies on these platforms to express themselves and be active politically, such as by criticizing politicians through political Internet memes. In a context where most of scholarly works on digital political communication focuses on textual data, researchers are faced with multifaceted and ongoing challenges. Among them include finding existing or developing new theoretical frameworks to study visuals’ role in digital political communication. They are also looking for methodological instruments to harvest, store, and examine visual content. Finally, they are facing issues relating to big data, including the exhaustive collection and analysis of large volumes of political information. This paper offers a snapshot of this evolving area of research through an overview of academic works on digital political communication and visuals. Specifically this paper provides the following: 1) theoretical overview of conceptual approaches; 2) categorization of types of research; 3) pinpointing trending research questions and methodologies; 4) identification of research opportunities within this understudied field. This paper is of interest to students and researchers who want to know more about digital political communication and visuals or to orient their work around visual topics.
Little Big Data: Geospatial Analysis of Voter Behaviour Using Census Data: Doug Munroe (Quest University Canada) Abstract: The vast majority of our insight into the behaviour of voters is derived from surveys like the Canadian Election Study or the Local Parliament Project. Surveys impose two limitations: First, our ability to draw statistically significant conclusions about any sub-population scales with sample size. Second, as the number of items in a survey increases, the response rate – and, therefore, sample size - decreases. Overcoming the compounding effects of these two limitations is costly. An alternative method is to use a Geographic Information System (GIS) to combine electoral results with census data. This produces a miniature version of “big data” on Canadian voters, with ample statistical precision and over two thousand variables, at no cost. How does this geospatial approach compare to survey methods? This paper explains the application of geospatial analysis to electoral results and census data, and assesses its performance as a method by contrasting it with the Canadian Election Study in testing some common propositions about demographics and voter behaviour in the 2015 federal election.
Polling in Modern Times: The Canadian Polling Industry in the 2000s: Christopher Adams (University of Manitoba) Abstract: This conference paper provides an overview of recent developments in the polling and market research industry in Canada. This includes the changing nature of the industry, including the declining use of field centres and traditional landline telephone surveys. Beginning in the late 1990s, many firms were moving into e-mail and web-based methodologies. More recently, we have seen the use of mixed methodologies, including the use of Interactive Voice Response (IVR) with web-based technologies. The author will provide a comparison between polls conducted by the major firms, their methodologies, and the actual outcomes of recent provincial elections and 2015 Canadian Federal Election. Discussed also will be recent developments in the regulatory environment and the 2018 bankruptcy of the Marketing Research and Intelligence Association (MRIA).
Christopher Adams is a political scientist and rector based at St. Paul’s College. He holds a PhD from Carleton and worked for twenty years in the marketing research industry in Toronto and Winnipeg. He was the co-author of a report for the 2018 inquiry into the 2017 Calgary civic election: https://www.cbc.ca/news/canada/calgary/calgary-municipal-election-poll-mria-report-release-1.4776633.
Reassessing the Digital Divide: A New Approach to Measuring and Analyzing Online Political Participation: Meyer Levy (University of Notre Dame) Abstract: This paper attempts to answer a fundamental question about the internet's role in modern politics: is the internet primarily a space for the traditionally advantaged, or can it bring new faces into the political conversation? Although the political science literature has struggled to find a clear answer to this question, I argue that this is primarily due to the difficulties inherent to the way we typically measure online activity. In order to resolve this problem, this paper combines survey data with web scraping, allowing me to directly measure the online political engagement of survey respondents. I hypothesize that, due to the low barriers to online participation relative to those of offline participation, we should see typically under-represented groups participating at higher rates than we might expect given the canonical account of political engagement. This paper contributes to the literature by attempting to answer a fundamental yet unresolved question using a new measurement approach particularly suited to this application.