Annie Waldherr

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.

Big Data, Big Noise The Challenge of Finding Issue Networks on the Web

In this article, we focus on noise in the sense of irrelevant information in a data set as a specific methodological challenge of web research in the era of big data. We empirically evaluate several methods for filtering hyperlink networks in order to reconstruct networks that contain only webpages that deal with a particular issue. The test corpus of webpages was collected from hyperlink networks on the issue of food safety in the United States and Germany. We applied three filtering strategies and evaluated their performance to exclude irrelevant content from the networks: keyword filtering, automated document classification with a machine-learning algorithm, and extraction of core networks with network-analytical measures. Keyword filtering and automated classification of webpages were the most effective methods for reducing noise, whereas extracting a core network did not yield satisfying results for this case.

Exploring Issues in a Networked Public Sphere: Combining Hyperlink Network Analysis and Topic Modeling

We propose a methodological approach to analyze the content of hyperlink networks which represent networked public spheres on the Internet. Using the case of the food safety movement in the United States, we demonstrate how to generate a hyperlink network with the web crawling tool Issue Crawler and merge it with the results of a probabilistic topic model of the network’s content. Combining hyperlink networks and content analysis allows us to interpret such a network in its entirety and with regard to the mobilizing potentials of specific sub-issues of the movement. We focus on two specific sub-issues in the food safety network, genetically modified food and food control, in order to trace the involved websites and their interlinking structures, respectively.

Towards an integrative approach to communication styles: The interpersonal circumplex and the Five-Factor theory of personality as frames of reference

Abstract This article reviews existing approaches to defining and distinguishing communication styles and proposes a common frame of reference for future research. The literature review yields two schools of thought: the behavior-centered perspective and the personality-oriented perspective. Although these lines of research differ in their ways of defining communication styles, they show considerable similarities with respect to their classification. Many researchers build their taxonomies on two key dimensions: assertiveness and responsiveness. We propose embedding communication styles into the Five-Factor Theory (FFT) and defining them as characteristic adaptations of personality. We also suggest the Interpersonal Circumplex (IPC) as a reference model for distinguishing communication styles as it is able to integrate substantial dimensions and facets of existing taxonomies in a parsimonious way.

The Mass Media as Actors in Innovation Systems

This chapter aims at a conceptual clarification of mass media’s influence on high-tech policy decisions. It deals with two major questions: (1) Do the mass media matter for innovation policy and thus also for high-tech policy? (2) Are the media interested in high-tech policy as a central field in innovation policy? The questions are answered theoretically in the following steps: First, it is argued that the mass media play an important role as actors in innovation systems and therefore are also relevant for high-tech policy decisions. The argument focuses on three different perspectives: processes, functions and structures in innovation systems. Second, factors influencing the activity level of media discourse are drawn from agenda-building and news selection theories. News values of innovation policy and high technologies are identified as important preconditions for the activity level of media discourse. The chapter concludes with proposing a theoretical framework integrating the interplaying factors.

Simulation gesellschaftlicher Medienwirkungsprozesse am Beispiel der Schweigespirale

Der Beitrag stellt mit der agentenbasierten Modellierung (ABM) eine Methode zur Diskussion, mit der sich dynamische Medienwirkungsprozesse auf mehreren Ebenen modellieren und simulieren lassen. Dazu wird das Mikro-Makro-Problem in der Medienwirkungsforschung genauer erläutert und aus Sicht der Komplexitätstheorie interpretiert. Die Methode der Computersimulation sozialer Prozesse, speziell mittels ABM, wird erläutert. Schließlich wird die ABM am Beispiel der Schweigespirale vorgestellt, um ihre Eignung für die Untersuchung dynamischer, gesellschaftlicher Medienwirkungsprozesse zu demonstrieren. Hierzu werden die Annahmen der Schweigespirale nach Noelle-Neumann in einem Computermodell formalisiert und in ihrer Dynamik simuliert. Nach der Darstellung zentraler Simulationsergebnisse werden abschließend Chancen und Grenzen der Simulationsmethode für die Medienwirkungsforschung diskutiert.

Homophily and prestige: An assessment of their relative strength to explain link formation in the online climate change debate

Abstract Previous work has shown that hyperlinks reflect actors’ strategic choices; these dyadic relationships depend on the actors’ exogenous attributes (e.g., homophily) and the network’s endogenous features (e.g., prestige distribution among actors). We combine these factors as explanatory variables in different exponential random graph models (ERGMs) to assess the relative strength of prestige and homophily for the actors’ link formation. We analyze the climate change discourse in a hyperlink network formed by US civil society actors from November 2014 and test how relevant the different factors are, including variables such as actor type, country, position, and topic. We find that both prestige and various aspects of homophily influence link formation online. With regard to the importance of the different factors, positional homophily stands out, followed by prestige and other homophily effects.

Modelling the Role of Social Media at Street Protests.

Occupy, the Gezi park movement, the Maidan protests, or the recent solidarity marches for Charlie Hebdo—since the uprisings of the Arab Spring, we could observe many examples of on-site protests on big squares and streets being accompanied by waves of collective action in social media. We present the design stage of an agent-based model that will allow us to explore the following questions: What role does social media play in street protests? How does social media usage influence the dynamics of collective action during street protests? Do social media affect the speed, scale, fluctuation, duration of the protest at large, and in which way? Do they impact specific crowd patterns, e.g., the development of groups within groups? The model builds on and integrates existing models of social media, protests, and crowd behavior to simulate the dynamics of street protests in an urban setting. Our central aim is to compare scenarios with intense, moderate, and no social media usage by the protesters.