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.     

Placing Facebook

Facebook is challenging professional journalism. These challenges were evident in three incidents from 2016: the allegation that Facebook privileged progressive-leaning news on its trending feature; Facebook’s removal of the Pulitzer Prize-winning “Napalm Girl” photo from the pages of prominent users; and the proliferation of “fake news” during the US presidential election. Using theoretical concepts from the field of boundary work, this paper examines how The Guardian, The New York Times, Columbia Journalism Review and Poynter editorialized Facebook’s role in these three incidents to discursively construct the boundary between the value of professional journalism to democracy and Facebook’s ascendant role in facilitating essential democratic functions. Findings reveal that these publications attempted to define Facebook as a news organization (i.e., include it within the boundaries of journalism) so that they could then criticize the company for not following duties traditionally incumbent upon news organizations (i.e., place it outside the boundaries of journalism). This paper advances scholarship that focuses on both inward and outward conceptions of boundary work, further explores the complex challenge of defining who a journalist is in the face of rapidly changing technological norms, and advances scholarship in the field of media ethics that positions ethical analysis at the institutional level.     

Examining the “Blanket of Protection” Conferred by Inoculation: The Effects of Inoculation Messages on the Cross-protection of Related Attitudes

Although the effectiveness of inoculation as a strategy for promoting resistance to attitude change is fairly well established, the potential of inoculation messages to offer cross-protection for related, but untreated, attitudes warrants additional attention from scholars. The reported study tested the “blanket of protection” conferred by inoculation. Participants (N = 118) were randomly assigned to read an inoculation message addressing a target topic and subsequently had their attitudes toward three related, but untreated, topics attacked. The results offer some evidence that inoculation messages can confer cross-protection for related attitudes. Participants in the inoculation condition reported greater perceptions of threat, greater counterarguing, and less attitude change in response to attacks than participants in the control condition for two of the three untreated topics. Counterarguing in response to attacks on untreated attitudes appears to be primarily responsible for cross-protection.     

Performance differences when using 26- and 29-inch-wheel bikes in Swiss National Team cross-country mountain bikers

The purpose of this study was to analyse the effect of bike type – the 26-inch-wheel bike (26“ bike) and the 29-inch-wheel bike (29“ bike) – on performance in elite mountain bikers. Ten Swiss National Team athletes (seven males, three females) completed six trials with individual start on a simulated cross-country course with 35 min of active recovery between trials (three trials on a 26“ bike and three trials on a 29“ bike, alternate order, randomised start-bike). The course consisted of two separate sections expected to favour either the 29“ bike (section A) or the 26“ bike (section B). For each trial performance, power output, cadence and heart rate were recorded and athletes’ experiences were documented. Mean overall performance (time: 304 ± 27 s vs. 311 ± 29 s; P < 0.01) and performance in sections A (P < 0.001) and B (P < 0.05) were better when using the 29“ bike. No significant differences were observed for power output, cadence or heart rate. Athletes rated the 29“ bike as better for performance in general, passing obstacles and traction. The 29“ bike supports superior performance for elite mountain bikers, even on sections supposed to favour the 26“ bike.     

Intrinsic Versus Extrinsic Motivation: An Approach/Avoidance Reformulation

The main purpose of this article is to advance an alternative perspective on the relationship between intrinsic and extrinsic motivation, and in particular to examine critically the assertion that these processes are antagonistic such that the will to learn for its own sake is inhibited by the presence of extrinsic, tangible rewards and incentives such as school grades. The presumption of an antagonistic relationship largely depends on the theoretical perspective adopted. An alternative interpretation based on need achievement theory leads to distinctly different conclusions. Exploring this new perspective allows one to identify both the conditions under which intrinsic motives may coexist with extrinsic motives as well as to consider some of the means by which intrinsic motives and caring about learning can be stimulated in their own right in school settings.     

Memories of reward events and expectancies of reward events may work in tandem

A common assumption is that expectancies of reward events in instrumental tasks are established on the basis of Pavlovian conditioning. According to the tandem hypothesis, tested in the four runway investigations reported here employing rats, memories of reward events may serve as the conditioned stimuli eliciting expectancies. In Experiments 1–3, rats were trained under a schedule of partial reward (P), which did not produce increased resistance to extinction, and subsequently shifted to consistent reward (C). According to the tandem hypothesis, the shift to the C schedule should result in increased resistance to extinction if, as hypothesized, under the P schedule the memory of reward, S^R, came to elicit the expectancy of nonreward,_EN. This hypothesis was confirmed under a variety of conditions. It was shown that increased resistance to extinction could not be attributed to the P schedule alone, to the rats receiving two schedules, P and C, to stimuli other than S^R eliciting E_N, or to the rats forgetting reward-produced memories when expecting nonreward (Experiment 4). It was shown that the tandem hypothesis could explain the divergent findings obtained in prior studies employing a shift from P to C as well as in the present study.