Communication Science Theory and Research

An Advanced Introduction

Marina Krcmar, David R. Ewoldsen, Ascan Koerner

• Publisher: Routledge
• Year: 2016
• ISBN:
  9781136288982
• Edition: 1
• Topic: Filología
• Subtopic: Estudios de comunicación

This volume provides a graduate-level introduction to communication science, including theory and scholarship for masters and PhD students as well as practicing scholars. The work defines communication, reviews its history, and provides a broad look at how communication research is conducted. It also includes chapters reviewing the most frequently addressed topics in communication science. This book presents an overview of theory in general and of communication theory in particular, while offering a broad look at topics in communication that promote understanding of the key issues in communication science for students and scholars new to communication research. The book takes a predominantly "communication science" approach but also situates this approach in the broader field of communication, and addresses how communication science is related to and different from such approaches as critical and cultural studies and rhetoric. As an overview of communication science that will serve as a reference work for scholars as well as a text for the introduction to communication graduate studies course, this volume is an essential resource for understanding and conducting scholarship in the communication discipline.

PART I
Communication Theory

1
AN INTRODUCTION TO THEORY

In communication, indeed in any discipline, theory is often held up as the zenith of research. Theory-driven research is not simply preferred, it is necessary to meaningful scholarship. Merely collecting data, without a theory to guide the process, is likely to lead to findings that are, at best, without meaning, and at worst, accidental, coincidental, or spurious. One could argue that along with solid research design, theory is one of the best protections against Type I error (i.e., false positives in research). In other words, with the guidance and direction of theory, one is less likely to “cry wolf” with any data-analytic findings.

But what is theory? Pavitt (2000, p. 111) defines theory as a “formal system of concepts and relationships tying these concepts together, with the functions of explaining, predicting, and allowing potential control over real-world phenomena.” In other words, in order to have a theory, one must have an observed phenomenon, concepts used to label individual aspects of that observed phenomenon, and linkages between the concepts that describe and explain how they relate to one another. Ideally, that explanation would be specific, explicit, and causal. In the end, such a theory would have the benefit of allowing some degree of control over the phenomenon. For example, a solid theory of persuasion would not only explain that there were effects of, say, an anti-smoking campaign, but would describe how the effects occurred and the mechanism or processes that caused the effects; individual concepts would exist as important elements of the theoretical processes and, ultimately, that theory could provide a model for the design of future successful anti-smoking campaigns.

But such a description of theory only skims the surface. Our understanding of it is limited by our understanding of such terms as concepts, phenomenon, causality, models, and, as we continue, variables, operationalizations, propositions, and linkages. Thus, in the remainder of the chapter, we will discuss in greater detail what theory is and the terminology used to discuss it, what theory is not, and some steps in the development of theory.

Phenomenon as Starting Place

Often, in communication research or in research in any one of the natural or social scientific disciplines, theory begins with an observed phenomenon. In the social sciences, that phenomenon is most typically a social behavior that is enacted by a human being. The mere observation of that phenomenon, however, does not constitute a theory (Berger et al., 2010). Instead, the observation of a phenomenon should be taken as the starting place, the first step in the creation and drawing of a theory, if in fact drawing is the term that should be used (see below).

For example, in the mass media effects literature, several meta-analyses (Anderson & Bushman, 2001; Paik & Comstock, 1994) have been conducted that demonstrate a relationship between exposure to media violence and aggressive behavior. These meta-analyses have been taken as evidence for the link between the two variables, and in fact, some of the research has been able to test for causality through experimental design. However, simply noting a statistical relationship between the two, or even noting a causal relationship between the two by means of experimental design, does not even begin to approach theory. Rather, at this stage in the process, we have mere observation. No causal explanation, indeed, no explanation at all is offered through observation alone.

An important point to be made here is that no study, no matter how well designed, crafted, conducted, and analyzed, may be considered theoretical. Rather, research and data most often live in the service of theories. Importantly, this is not always the case, as will be discussed presently, but for the most part, data are servile to theory. Certainly data can have practical purposes in and of themselves, but a good study starts with theory. The data then are used, literally used, to test and then support or not support the theory. Data collection should be guided by theory and data analysis used to test it. A well-designed study, therefore, should be viewed as a test of a theoretical phenomenon.

Despite the data-in-service-of-theory argument, the phenomenon, or casual observation, often comes first. A researcher may notice, observe, or use logical analysis in identifying a phenomenon or starting place. For example, I may observe the phenomenon of aggression and wonder about its antecedents. Or, I may observe someone, a child perhaps, acting aggressively after watching a violent cartoon. Note that the co-occurrence of these two phenomena (watching the cartoon and aggression) is not a theory. It is merely an extended phenomenon. Thus, we begin with an observed phenomenon. From that starting point, we may begin to craft theoretical explanations or we may test the observation through systematic research. The process, including casual observation of the phenomenon, theoretical explanation, and systematic research, occurs in an iterative process, a process that is likely to take years. Thus, the phenomenon is often the place where theory begins. At times, theory may emerge without casual observation. For example, careful reading of many validated theories may lead a researcher to see connections between them, thus allowing for a theoretical evolution or new idea in absence of any observed phenomenon. But for the purpose of this discussion, we suggest that observing a phenomenon is a starting place.

Implicit vs. Explicit Theory

We argue above that the observation of a phenomenon is the starting place of theory. However, this is not precisely accurate because it assumes that observation is somehow without antecedent, itself. It assumes that nothing, not presuppositions, preconceptions, or biases, guided the initial observation of the phenomenon. Much recent research in perception indicates that this notion of unbiased observation is false. In fact, we might argue that a phenomenon is a type of starting place, but so is theory itself.

Berger et al. (2010) distinguish between “intuitive or implicit theories, on one hand, and formally explicated theories on the other” (p. 11). They refer to these as System 1 theories and System 2 theories, and liken them to the dual processing approach to information processing (Stanovich, 2002). In this view, cognitive processing can at times be implicit, where our perceptions, interpretations, understanding, and even memory and recall occur at the preconscious level (Gigerenzer, 2007). This, they refer to as System 1. Alternately, cognitive processing can be explicit and formal. Such is the case when we read, draw conclusions, and perhaps summarize what we have read in a systematic, consciously aware way. This, they refer to as System 2.

Theory, too, can be formulated this way. After all, we are all naïve and constant theoreticians. We observe a phenomenon; perhaps at a children’s birthday party we notice the co-occurrence of wild behavior and the consumption of birthday cake. We then draw naïve conclusions about the phenomenon. We determine that sugar, found in abundance in cake, causes hyperactivity in children. We then use our theory to explain the behavior and predict future outcomes such as hyperactivity in children after any sugar consumption. We may even use our theory to guide future actions such as limiting children’s sugar intake. This is an example of a System 1 theory. Interestingly, all systematic tests of this theory about sugar and hyperactivity in children have shown no support for it whatsoever (Howard-Jones, 2014).

However, there are also systematic theories, crafted, drawn, and tested by researchers. These are conscious and careful explanations of phenomena. They are guided by formal rules of logic and inference. They are mapped out and tested, either in portion or in full through the highly meticulous and methodical process of data collection and analysis. Despite the distinction drawn between these two types of theory by Berger et al. (2010), they note that there is likely an undeniable connection between System 1 and System 2 theories. In fact, it is likely that our selection of research topics, our pursuit of certain avenues of inquiry, indeed the very attention to particular phenomena over others and particular co-occurrences over others are guided by System 1 (Popper, 1959). As we continue to formalize our thinking, it is likely that we engage System 2 to a greater degree. However, it is impossible to disengage the implicit system (Kahneman & Frederick, 2002). Thus, all theory is guided by both implicit and explicit systems. However, as social scientists, we attempt to engage System 2 and in fact, we value its processes and outcomes as true theory, whereas System 1 processes and outcomes are often viewed as less important and something to be guarded against.

Therefore, we begin with observation of a phenomenon that is guided by System 1 theorizing. We continue through the process of explicit theory building, and throughout, utilize the process of systematic research to refine our theories. Once our theories are refined, we return again to systematic research in order to further test the refined theories. In the end, theory building is not a process that can be considered linear and completely systematic. Rather, it is both iterative and at times holistic. However, as we approach it in this chapter, we are forced by the linearity of writing to proceed as if it were systematic. Thus, in the next sections, we’ll discuss theory building by invoking the idea of concepts that need to be identified and then winnowed to particular variables. These variables are connected to one another through theoretical linkages made up of propositions, extended propositions, proposed mechanisms, and links, thus resulting in theory. That theory can often be visually displayed by means of a conceptual model. Ultimately, however, it is not sufficient to craft abstract theory, or even to represent it with a model. Theory must be tested. Thus, we move to the testing phase by operationalizing variables for measurement’s sake. These operationalized variables are also connected to one another in the prediction and testing phase through the process of operational linkages.

Concepts and Variables

Chaffee (1996) argues that before we can begin the process of theorizing, we need to identify a concept or concepts that will be utilized in the theory-building process. In other words, we need to identify, and label, those ideas with which we are working. For example, aggression is an idea that is frequently studied in media effects research. Scholars may investigate aggressive outcomes related to exposure to violent cartoons, or aggression as an outcome resulting from exposure to video games. In any case, aggression itself is an idea with a specific definition, and a specific label to go along with it. This clearly defined, well-articulated idea can be termed a construct, and it is one of the building blocks for a theory of social learning whereby exposure to television violence can result in increases in behavioral aggression. Through the process of theorizing, one must identify the constructs one intends to work with, explore how those constructs have been used in previous research, and subsequently decide precisely what the construct will be in the research being pursued.

However, identification of the construct is a mere starting place. Theory is typically built on constructs; that is, they act as a base. However, theory is not built with constructs. Constructs are the more abstract notion, the idea that is then more clearly demarcated by the term “variable.” A variable, of course, is something that varies within a sample. Age, for example, is a ratio-level variable that does not have a more abstract construct associated with it. Often, ratio-level variables are not more clearly specified concepts; they are merely measurements of some existing, well-defined entity such as years, hours, pounds, etc. However, often in the social sciences, what we intend to measure is more abstract. But this abstract concept must be identified and measured. Once the identification takes place, and the decision is made to utilize the concept, we identify it as a variable that will be used in our theory building. In fact, we might say that a construct becomes a variable when we intend to measure it and when we use it in the process of theorizing.

Variables and Theoretical Linkages

Theory itself, however, is made up of more than merely variables. In order for a set of variables to be a theory, they must be connected to one another through linkages, proposed processes, and proposed mechanisms. These linkages, processes, and mechanisms are labeled in a theory as proposition statements or theoretical linkages. For example, I might state that aggression increases after exposure to television violence. This may be one propositions within a theory. However, there are different kinds of propositions, which we will briefly discuss here.

First, there are both existence statements and relational statements. The first, the existence statement, poses only that something exists: aggression was present in the television clip. These are typically assumptions within a theory and not part of the theory itself. The second, the relational statement, is the basis of good theory building. In relational statements, one or more propositions are forwarded, stating that two variables are linked in such a way that they are either associatively or causally related (Reynolds, 1971).

In the former case, a statement that includes two variables is proposed, and an association between the variables is posited. Thus, a proposition might state that “as exposure to television violence increases, aggression levels increase.” In the latter case, the two variables are proposed to be causally related, with exposure to television violence causing increases in aggressive behavior. Note, however, that simply stating that two variables are causally linked does not provide anything in the way of explanation. Thus, ultimately, good theory not only provides causal propositions, but it breaks those propositions down to provide processes and mechanisms for the causal link. In other words, a given causal link is further dissected to propose a process that explains the causality. As mentioned earlier in the chapter, simply stating that something occurs, or that something is related, is really observation and not theory, per se.

Rather, in good theory building, the associative or causal link must be extended by limiting or conditional propositions. For example, when claiming that exposure to television violence is associated with increases in aggression, one might further pose the proposition that this relationship is stronger among boys, or occurs with greater frequency when the depicted violence is shown being rewarded. Thus, one proposition may claim an association between exposure to television violence and aggressive behavior. A second proposition may claim that the link is stronger for boys, that is, gender moderates the associative link between exposure and aggression. A third proposition may further claim that the link is stronger when the violent actor is shown being rewarded.

In addition to providing limiting or conditional propositions, an associative or causal link may be explained by propositions that offer mechanisms or explanations for the link. For example, various extant theorizing about the link between exposure to television violence and aggression has proposed mechanisms of increased arousal or cognitive priming. In other words, solid theory proposes not only that an association exists between two variables, but may offer propositions offering an explanation for the process. For example, exposure to media violence may increase physiological arousal, which may in turn increase aggressive behavior. These process propositions offer mediating variables between two associated variables. In any case, theory building extends “mere associations” beyond the obvious relationship between two variables and determines, by means of successive propositions, how and under what conditions the relationship occurs.

For example, sociocognitive theory (Bandura, 2002) predicts that exposure to media violence will increase aggression in viewers. This occurs because people can learn behaviors not only from direct experience but also from vicarious experience, such as media exposure. However, what makes sociocognitive theory a meaningful theory is its discussion of the mechanisms by which this imitation of violence occurs. Specifically, Bandura (2002) argues that increases in aggression resulting from exposure to media violence result from subprocesses including attention, retention, production, and motivation. These subprocesses are not prop...

Table of contents

• Cover
• Half Title
• Title Page
• Copyright Page
• Table of Contents
• Author Biographies
• Preface
• Acknowledgments
• Introduction
• Part I Communication Theory
• Part II Communication Research
• Author Index
• Subject Index