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.     

Investigating the quality of interactions and public engagement around scientific papers on Twitter

This study explores science communication on Twitter by investigating a sample of tweets referring to academic papers in five different scientific fields. The specifications of science communicators on Twitter, the characteristics of those who initiate actions (by tweeting), the extent and quality of reactions (retweeting), individual and group interactions, and the distribution of tweets across types of engagement in the process of science communication (i.e., dissemination, consultation, and evaluation) were explored. A broad array of actors is involved in the communication of science on Twitter, with individual citizens and individual researchers playing an important role. In principle, this is promising for creating direct interaction, which can be difficult through more traditional mass media. The vast majority of communication activities regarding academic papers is undigested dissemination with almost no sign of debate, contestation, or collective reflection. Another general finding of this study is that bot accounts play a major role in the science communication landscape on Twitter.     

Course Design,Quality Matters Training,and Student Outcomes

ABSTRACT

North Carolina Central University (NCCU) recognized the need to address the increasing rates of Ds, Fs, and Withdrawal by students matriculating in online courses. Led by two science faculty, a faculty learning community in partnership with the NCCU Division of Extended Studies was created to assess online science course offerings and instruction. Faculty within this learning community completed the Quality Matters (QM) “Applying the Quality Matters Rubric” course. This online training course, coupled with services provided by the NCCU Division of Extended Studies, offered the faculty learning community a variety of pedagogical strategies to meet their instructional needs. All faculty participating in this learning community completed the online training course and all faculty indicated that they implemented changes in their subsequent online course offerings. This article describes the impact of that training and the application of quality course design standards in the Quality Matters Rubric on the design and student outcomes for an Introductory Biology course over four terms. As faculty learning communities are broadly utilized, it is anticipated that this article may present an effective strategy to increase the quality and quantity of online science, technology, engineering and mathematics (STEM) courses at similar institutions.     

Comparing the magnitude and direction of asymmetry during the squat,countermovement and drop jump tests in elite youth female soccer players

ABSTRACT

The aims of the present study were to provide an in-depth comparison of inter-limb asymmetry and determine how consistently asymmetry favours the same limb during different vertical jump tests. Eighteen elite female under-17 soccer players conducted unilateral squat jumps (SJ), countermovement jumps (CMJ) and drop jumps (DJ) on a portable force platform, with jump height, peak force, concentric impulse and peak power as common metrics across tests. For the magnitude of asymmetry, concentric impulse was significantly greater during the SJ test compared to CMJ (p = 0.019) and DJ (p = 0.003). No other significant differences in magnitude were present. For the direction of asymmetry, Kappa coefficients revealed fair to substantial levels of agreement between the SJ and CMJ (Kappa = 0.35 to 0.61) tests, but only slight to fair levels of agreement between the SJ and DJ (Kappa = ?0.26 to 0.18) and CMJ and DJ (Kappa = ?0.13 to 0.26) tests. These results highlight that the mean asymmetry value may be a poor indicator of true variability of between-limb differences in healthy athletes. The direction of asymmetry may provide a useful monitoring tool for practitioners in healthy athletes, when no obvious between-limb deficit exists.     

Part 1. If We Frame It,They Will Respond: Undergraduate Student Responses to the Framework for Information Literacy for Higher Education

There remains significant confusion surrounding the purpose, audience, and proposed application of the Framework for Information Literacy for Higher Education. This exploratory study addresses the question of whether undergraduate students can understand the language and concepts in this document. By presenting student responses to pretest questions recorded at the beginning of a semester-long research methods course, this article shows that students can, even on first impression, begin to make sense of the complexity and richness found in the Framework for Information Literacy for Higher Education.