Applying LDA Topic Modeling in Communication Research: Toward a Valid and Reliable Methodology

ABSTRACT

Latent Dirichlet allocation (LDA) topic models are increasingly being used in communication research. Yet, questions regarding reliability and validity of the approach have received little attention thus far. In applying LDA to textual data, researchers need to tackle at least four major challenges that affect these criteria: (a) appropriate pre-processing of the text collection; (b) adequate selection of model parameters, including the number of topics to be generated; (c) evaluation of the model’s reliability; and (d) the process of validly interpreting the resulting topics. We review the research literature dealing with these questions and propose a methodology that approaches these challenges. Our overall goal is to make LDA topic modeling more accessible to communication researchers and to ensure compliance with disciplinary standards. Consequently, we develop a brief hands-on user guide for applying LDA topic modeling. We demonstrate the value of our approach with empirical data from an ongoing research project.     

Citizen-Centered Journalism and Contested Boundaries

This article explores the existence of a “citizen-centered journalism” that sees citizens as crucial participants in the construction of news and as co-creators of their own worlds. Through qualitative case studies of three news organizations, the article examines the motivation for using a citizen-centered approach, the news routines that are required to do so, the categories of content produced, and the perceived impact of this approach. The results suggest these news organizations are working in partnership with communities and striving to give a voice to historically marginalized communities. The journalists, however, see citizen participation as complementary to professional journalistic routines that favor verifiable information, rather than assigning inherent value to it for its own sake.