Login | Register

Uncovering Latent Topics in Text: Using Topic Models to Identify Discussion Themes in the Brett Kavanaugh Controversy

Title:

Uncovering Latent Topics in Text: Using Topic Models to Identify Discussion Themes in the Brett Kavanaugh Controversy

Rodier, Simon (2021) Uncovering Latent Topics in Text: Using Topic Models to Identify Discussion Themes in the Brett Kavanaugh Controversy. Masters thesis, Concordia University.

[thumbnail of Rodier_MA_S2021.pdf]
Preview
Text (application/pdf)
Rodier_MA_S2021.pdf - Accepted Version
Available under License Spectrum Terms of Access.
1MB

Abstract

Social media texts are abundant and generated at a rapid pace; often, they discuss contentious issues in ways that can be extreme, with people defending counter-intuitive points of view. This project develops an understanding of how individuals discuss contentious issues in online fora. It examines gender as a contentious issue, and also examines the controversy surrounding Brett Kavanaugh's nomination to the U.S. Supreme Court amidst sexual assault allegations in 2018 as a gendered controversy. Social media posts about the controversy were collected from reddit and analyzed with a framework using topic models, which help uncover latent topics in text. Significant discussion throughout the corpus centered on four major themes: the search for and evaluation of evidence; the importance of a Supreme Court nomination and of an investigation into the allegations; sexual assault and gendered perceptions and expectations; and finally, a discussion of the judiciary hearing into the allegations. Discussion of the results suggests that notions of power and loss aversion may be at play throughout the corpus. Additional discussion reflects on the use of topic models as a qualitative research tool, suggesting that they can be an effective way of exploring broad themes within larger corpora.

Divisions:Concordia University > Faculty of Arts and Science > Education
Item Type:Thesis (Masters)
Authors:Rodier, Simon
Institution:Concordia University
Degree Name:M.A.
Program:Educational Technology
Date:10 February 2021
Thesis Supervisor(s):Venkatesh, Vivek and Schmid, Richard
ID Code:988039
Deposited By: SIMON RODIER
Deposited On:29 Jun 2021 20:56
Last Modified:29 Jun 2021 20:56
All items in Spectrum are protected by copyright, with all rights reserved. The use of items is governed by Spectrum's terms of access.

Repository Staff Only: item control page

Downloads per month over past year

Research related to the current document (at the CORE website)
- Research related to the current document (at the CORE website)
Back to top Back to top