Showing papers in "Communication Methods and Measures in 2021"
TL;DR: An exhaustive comparison of sentiment analysis methods using a validation set of Dutch economic headlines to compare the performance of manual annotation, crowd coding, numerous dictionaries and machine learning using both traditional and deep learning algorithms concludes that the best performance is still attained with trained human or crowd coding.
Abstract: Sentiment is central to many studies of communication science, from negativity and polarization in political communication to analyzing product reviews and social media comments in other sub-fields...
100 citations
TL;DR: A semisupervised document scaling technique, called Latent Semantic Scaling (LSS), which can locate documents on various pre-defined dimensions by combining user-provided seed words and latent semantic analysis (word embedding).
Abstract: Many social scientists recognize that quantitative text analysis is a useful research methodology, but its application is still concentrated in documents written in European languages, especially E...
38 citations
TL;DR: A large number of the questioned people believe that online tracking should be considered as a source of income for the advertising industry, not a substitute for other forms of media consumption.
Abstract: With the increasing importance of online information environments, researchers have started investigating direct measures of online media use, such as online tracking. Most existing studies using t...
23 citations
TL;DR: In this paper, the authors proposed a method to estimate the amount of time spent using digital technology using smartphone features, such as the iOS Screen Time application, which allow researchers to obtain more objective metrics.
Abstract: People are generally poor reporters of time spent using digital technology. Advancing smartphone features, such as the iOS Screen Time application, allow researchers to obtain more objective measur...
18 citations
TL;DR: In this paper, psychological reactance, commonly operationalized as a latent construct comprised of anger and negative cognitions, represents an aversive state following a threatened or eliminated freedom, and is defined as a state following the elimination of freedom.
Abstract: Psychological reactance, routinely operationalized as a latent construct comprised of anger and negative cognitions, represents an aversive state following a threatened or eliminated freedom. The c...
18 citations
TL;DR: In this paper, the authors develop algorithm literacy to empower users to make algorithmic decision-making decisions in big data analytics. But they do not address the problem of data privacy.
Abstract: In light of the widespread use of big data analytics, internet users are increasingly confronted with algorithmic decision-making. Developing algorithm literacy is thus crucial to empower users to ...
15 citations
TL;DR: The manual coding of nonverbal cues by trained human coders, such as counting the occurrence of a specific behavior or subjective rating about a speaker, is still a major method in the study of non-verbal cues.
Abstract: The manual coding of nonverbal cues by trained human coders, such as counting the occurrence of a specific behavior or subjective rating about a speaker, is still a major method in the study of non...
13 citations
TL;DR: In this paper, a three-wave panel survey among 336 pupils enables us to disentangle which effect the friendship network has on individual media use and which effect media use has on the development and maintenance of friendship ties.
Abstract: Social interactions regarding media content are omnipresent and highlight that media use is closely linked to social life. However, surprisingly little is known about the related social dynamics and their consequences for media use and social relations. This study addresses this topic based on co-orientation theory and investigates the social dynamics of TV and YouTube use among adolescents. A three-wave panel survey among 336 pupils enables us to disentangle which effect the friendship network has on individual media use (RQ1) and which effect media use has on the development and maintenance of friendship ties (RQ2). Based on a multi-level approach of dynamic social network analysis and applying stochastic actor-oriented modeling, we find empirical evidence for processes of social influence regarding TV programs and YouTube channels. No support is found for the hypothesized social selection processes. Along with these empirical insights, the research design demonstrates how the micro-level of individual behavior can be linked with the meso-level of social groups. Furthermore, the insights can be used as a basis to understand larger-scale phenomena emerging on the macro-level of societies and media markets such as audience fragmentation and long-tail audience distributions across media contents.
13 citations
TL;DR: A first step toward instrument validation is indicated, indicating that both methods for the topic modeling of multilingual document collections yield valid, comparable results, while method-specific differences remain.
Abstract: The goal of this paper is to evaluate two methods for the topic modeling of multilingual document collections: (1) machine translation (MT), and (2) the coding of semantic concepts using a multilin...
13 citations
TL;DR: In communication research, topic modeling is primarily used for discovering systematic patterns in monolingual text corpora and an overview of recently presented stratified models is provided.
Abstract: In communication research, topic modeling is primarily used for discovering systematic patterns in monolingual text corpora. To advance the usage, we provide an overview of recently presented strat...
12 citations
TL;DR: This paper proposes a multi-agent model of signaling games with three parameters – memory length, the complexity of communication problems, and population size – as potential constraints imposed on the collective learning process to suggest that finite memory leads to suboptimal signaling systems.
Abstract: This study investigates the mechanism by which individuals learn to associate signals with meanings in a way that is agreeable to everyone, and thereby, to collectively produce common and stable si...
TL;DR: Agent-based models present an ideal tool for interrogating the dynamics of communication and exchange and allow individual aspects of human interaction to be isolated and controlled in a controllable manner.
Abstract: Agent-based models present an ideal tool for interrogating the dynamics of communication and exchange. Such models allow individual aspects of human interaction to be isolated and controlled in a w...
TL;DR: EITM combines the validity of external domain knowledge captured through expert surveys with probabilistic topic models to help researchers identify subsets of documents that cover initially unknown domain-specific topics that belong to a researcher-defined master topic.
Abstract: The first step in many text-as-data studies is to find documents that address a specific topic within a larger document set. Researchers often rely on simple keyword searches to do this, even though this may introduce considerable selection bias. Such bias may be even greater when researchers lack the domain knowledge required to make informed search decisions, for example, in cross-national research or research on unfamiliar social contexts. We propose expert-informed topic modeling (EITM) as a hybrid approach to tackle this problem. EITM combines the validity of external domain knowledge captured through expert surveys with probabilistic topic models to help researchers identify subsets of documents that cover initially unknown domain-specific topics, such as specific events and debates, that belong to a researcher-defined master topic. EITM is a flexible and efficient approach to the thematic selection of documents from large text corpora for further study. We benchmark and validate the method by discovering blog posts that address the public role of religion within large corpora of Australian, Swiss, and Turkish blog posts and provide researchers with a complete workflow to guide the application of EITM in their own work.
TL;DR: The present study confirms and extends existing diffusion research by showing that in-degree hierarchy influences multiple diffusion outcomes; hierarchy determines the likely targets of information requests, but its effects are contingent on structural fluidity, the extent that information-seekers may deviate from typical network behavior.
Abstract: People need information to cope with the demands of life, yet, questions remain about the way social structures facilitate the spread of new information. This study applies agent-based modeling (AB...
TL;DR: Agent-based computational models create virtual laboratories in which to formalize and simulate dynamic, multi-level theories of communication as mentioned in this paper, and allow the systematic development of thought expe...
Abstract: Agent-based computational models create virtual laboratories in which to formalize and simulate dynamic, multi-level theories of communication. They allow the systematic development of thought expe...
TL;DR: A new measure to explicate and operationalize the ways multiple modalities are interconnected is proposed and exhibited strong psychometric properties and contained three types of constructive integration.
Abstract: A distinguishing feature of contemporary relationships is the integration of technological mediation into routine social interaction. The communication interdependence perspective (CIP) explains ho...
TL;DR: This paper proposed to combine off-the-shelf dictionaries with supervised machine learning algorithms to classify texts and learn about the most important features for classification, which can inform theory development on top of enabling a valid classification.
Abstract: Dictionary-based approaches to computational text analysis have been shown to perform relatively poorly, particularly when the dictionaries rely on simple bags of words, are not specified for the domain under study, and add word scores without weighting. While machine learning approaches usually perform better, they offer little insight into (a) which of the assumptions underlying dictionary approaches (bag-of-words, domain transferability, or additivity) impedes performance most, and (b) which language features drive the algorithmic classification most strongly. To fill both gaps, we offer a systematic assumption-based error analysis, using the integrative complexity of social media comments as our case in point. We show that attacking the additivity assumption offers the strongest potential for improving dictionary performance. We also propose to combine off-the-shelf dictionaries with supervised “glass box” machine learning algorithms (as opposed to the usual “black box” machine learning approaches) to classify texts and learn about the most important features for classification. This dictionary-plus-supervised-learning approach performs similarly well as classic full-text machine learning or deep learning approaches, but yields interpretable results in addition, which can inform theory development on top of enabling a valid classification.
TL;DR: Focusing on political communication as well as media effects and selection research, a model that tackles two statistical challenges of interrelated communication dynamics by measuring the reciprocal relationship between communication processes and associated outcomes in distinct sequences and the disaggregation of intraindividual and interindividual effects is presented.
Abstract: Over the last years, analyzing the dynamic reciprocal relationship between communication processes (i.e. mediated and interpersonal communication) and associated attitudes, beliefs, or behaviors ha...
TL;DR: In this paper, the precise degree to which citizens are exposed to opposing views on the Internet has been investigated and found to be either intentional or incidental, depending on the viewpoint of the user.
Abstract: Citizen’s exposure to disagreement – whether intentional or incidental – is a central concept in communication research, yet the precise degree to which citizens are exposed to opposing views onlin...
TL;DR: In this paper, the authors propose a quadrilogy of four measures of the reliability of big data data, applicable quite generally, for testing the reliabilities of really big data.
Abstract: This paper responds to the challenge of testing the reliabilities of really big data and proposes a quadrilogy of four measures of the reliability of data, applicable quite generally. These measure...
TL;DR: The increasing adoption of brain imaging methods has greatly augmented our understanding of the neural underpinnings of communication processes as mentioned in this paper, enabling us to better understand the neural networks of the human brain.
Abstract: The increasing adoption of brain imaging methods has greatly augmented our understanding of the neural underpinnings of communication processes. Enabled by recent advancements in mathematics and co...
TL;DR: It is found that training improves crowdcoding quality, while codebooks do not, and it is shown that relying on several human coders and applying majority rule to their assessments significantly improves performance.
Abstract: In this work, we evaluate different instruction strategies to improve the quality of crowdcoding for the concept of civility. We test the effectiveness of training, codebooks, and their combination...
TL;DR: This paper outlines a procedure to conduct mock website experiments that addresses methodological shortcomings using only freely available software and that does not restrict participants to specific devices nor require additional software to be installed.
Abstract: The mock website paradigm has been established as a methodological standard for experimental selective exposure research. To date, despite the amount of empirical research utilizing mock website ex...
TL;DR: This work presents a method and open-source tool for linking event data to news articles, and proposes a novel strategy for fine-tuning parameters of the event matching algorithm that requires no manual coding.
Abstract: Digital archives that document real-world events provide new opportunities for large-scale analyses of how news coverage represents reality. We present a method and open-source tool for linking eve...
TL;DR: It is found thatfoils are less associated with true items than true items are with one another, or more practically, that foils are more associated with the underlying dimension of interest than with the true items.
Abstract: Researchers frequently measure recognition of information in health messages by presenting participants with statements that were or were not in a message and then asking them to identify which wer...