Using Topic Models to Study Journalist-Audience Convergence and Divergence: The Case of Human Trafficking Coverage

• In Event: 464 - Section on Communication and Information Technologies Paper Session. Open Topic on Communication, Media, and/or Information Technologies

Abstract

Before the rise of the Internet the public agenda regarding human trafficking was generally set by legacy news outlets (Ghanem 1997; Wallinger 2010). Today online news and social media allow audiences to play a more active role (Hermida 2011), yet despite the accessibility of online news and availability of sophisticated methods for analyzing news content, little is known about the relations between news coverage of human trafficking, audiences’ reactions to and interpretations of this coverage, and the public discussions that result from online interactions between journalists and their readers. In this paper we examine journalists’ and commenters’ topic choices in coverage and discussion of human trafficking in the British newspaper The Guardian from 2009-2014. We first use Latent Semantic Analysis (LSA) to identify ten topics in our corpus of articles and readers’ comments, and then quantitatively investigate the topic preferences to identify convergence and divergence on the topics discussed by journalists and readers respectively. The goal of our study is thus to determine whether, and if so how, the subject of human trafficking is approached differently by journalists and their audiences.

Authors

• Maria Eirini Papadouka, University of North Texas

• Nicholas Evangelopoulos, University of North Texas

• Gabe Ignatow, University of North Texas