Human Communication Research, 2024, 50, 154–161

https://doi.org/10.1093/hcr/hqad037
Advance access publication 13 October 2023
Original Research

The role of theory in researching and understanding

human communication
1 2,
Timothy R. Levine and David M. Markowitz *
1
Department of Communication Studies, University of Alabama Birmingham, Birmingham, AL 35233, USA
2
Department of Communication, Michigan State University, East Lansing, MI 48824, USA
*Corresponding author: David M. Markowitz. Email: [email protected]

Abstract

Downloaded from https://academic.oup.com/hcr/article/50/2/154/7311752 by guest on 30 April 2024

Communication is a theory-driven discipline, but does it always need to be? This article raises questions related to the role of theory in communi-
cation science, with the goal of providing a thoughtful discussion about what theory is, why theory is (or is not) important, the role of exploration
in theory development, what constitutes a theoretical contribution, and the current state of theory in the field. We describe communication
researchers’ interest with theory by assessing the number of articles in the past decade of research that mention theory (nearly 80% of papers
have attended to theory in some way). This article concludes with a forward-looking view of how scholars might think about theory in their work,
why exploratory research should be valued more and not considered as conflicting with theory, and how conceptual clarity related to theoretical
interests and contributions are imperative for human communication research.
Keywords: theory, communication theory, validity, human communication, theoretical contribution.

Theory looms large in the practice of human communication perspective. Unlike similar essays on communication theory,
scholarship. College-level textbooks on various communica- our goal is not to be prescriptive about how to do theory
tion topics describe relevant theories to students enrolled in (e.g., how to create or even define theory), nor are we
communication classes. In thesis and dissertation defenses, attempting to be pollyannish about the many accepted virtues
students are often asked, “What theory are you using?,” im- of theory. Instead, a conversation about communication the-
plying that they must apply at least one theory to ground their ory is advanced by asking questions that are hard to answer,
research and contribute to the field. In the peer-review process interrogating some often-implicit presumptions about the role
of academic journals, a perceived failure to be sufficiently the- of theory in communication science, and raising some less ob-
oretical can be grounds for rejection. Few, if any, modern vious implications of theory. This article will succeed if it
communication scholars would embrace the labels of being prompts deeper thought and discussion on the topic of com-
“dust-bowl” or atheoretical. Surely, no serious communica- munication theory across many areas of the field. We envision
tion scholar is genuinely and categorically against theory. this being useful for early career communication scholars who
Theory is undeniably a desirable “warm-fuzzy good thing” in are uncertain about what theory really is and why it matters,
modern academic culture (Mook, 1983). Without it, the bed- and a thoughtful commentary for seasoned communication
rock of communication and other social sciences is shaky and researchers who may wrestle with theory development as they
uncertain. No academic discipline could be built from a move forward in their established research programs.
purely empirical foundation. Having theory—understanding
the value it provides, what we gain by building and extending
theory, and the contributions that a scholar can make to the- “What theory are you using?”
ory—is more complicated and nuanced, however. We interro- In our experience, graduate students (and undergrads, junior
gate these complications in this article. faculty, visiting scholars, etc.) are frequently asked “what the-
This article addresses a broad set of questions about the ory are you using?” when trying to position one’s work
past, present, and future role of theory in communication within communication science at large. This question raises
scholarship: What is theory? Do all scholarly contributions re- many meta-theoretical issues relevant to this article. First, the
quire theory? What does it mean to make a theoretical contri- question implies that theory is a prerequisite for scholarship.
bution? What does theory do for us? And, if we continue to Later in this article, we will argue that this perspective is un-
accept theory as a scholarly imperative, what is the current fortunately limiting, and that there is a need for exploratory
state of theory in the field and where might communication and pre-theoretical inquiry and data. Further, a relevant the-
theory go in the future? To address these questions, we follow ory is not always available for each research interest. A better
and draw on important commentaries such as those of initial question, in our opinion, asks if there is a relevant the-
Chaffee and Berger (1987), Slater and Gleason (2012), and ory to be tested, extended, or used. When no relevant theory
DeAndrea and Holbert (2017), but we provide our own exists, the lack of a theory should not preclude research nor

Received: 20 June 2023. Accepted: 8 September 2023

C The Author(s) 2023. Published by Oxford University Press on behalf of International Communication Association.
V
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which
permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited.
Human Communication Research, (2024), Vol. 50, No. 2 155

publication. Advancing theory is undeniably valuable. Not all contributing to theory. Of course, deeply theoretical work
scholarly contributions, however, explicitly advance theory in mentions theory, but so too do articles that engage in a prac-
ways that are recognized at the time they are written. tice that might be called theoretical name dropping. Our con-
When a relevant theory or theories are available, follow-up cern is that if work must invoke theory to pass the peer-
questions might address how the theory is being engaged (cf. review process, then exploratory and pre-theoretical work
Slater & Gleason, 2012). Is the theory being directly tested in might mention a theory or otherwise invoke theory-related
a way that is informative about the merits of the theory? How words to appear theoretical and thereby get published. In
is the theory being advanced? Is a boundary condition being line with this concern, DeAndrea and Holbert (2017) found
explored or the scope being extended? Merely using a theory that the use of words in articles specifically related to
to inform a topic can provide valuable direction and insights, theory evaluation were infrequent. In a sample of articles
but using theory is less likely to push the theory and its related from four top journals between 2013 and 2015 (Journal
propositions forward. Making a theoretical contribution of Communication, Human Communication Research,
involves actually advancing the theory itself (Slater & Communication Research, Communication Monographs),
Gleason, 2012). just less than one-third of them included at least one theory
evaluation word, and at least half of those were evaluating

Downloaded from https://academic.oup.com/hcr/article/50/2/154/7311752 by guest on 30 April 2024

Theory is ubiquitous in published method rather than theory.
communication scholarship Deciding what is truly theory and what is not requires de-
fining theory. This is where our conversation becomes more
We opened by opining that communication scholarship is complex.
consumed with theory and that contributing to theory is a pri-
ority of the discipline. To descriptively demonstrate this ex-
tent of this interest, we evaluated over 10,000 full text What is theory anyway?
communication articles across 26 journals that were pub- Despite the ubiquity of researchers’ focus on theory in pub-
lished in the last decade (see Markowitz et al., 2021), in lished communication scholarship (Figure 1), any thoughtful
search of how often they focus on five key theory-related discussion of theory is fraught from the start due to unavoid-
terms (theory, theories, theoretical, theoretical contribution, able definitional ambiguity. There is no one definition of com-
theoretical contributions). The top panel of Figure 1 repre- munication theory, nor can there be, nor should there be. As
sents the percentage of papers within each year that men- Miller and Nicholson (1976) rightfully suggested, definitions
tioned at least one of the five terms related to theory. On
are not by their nature things that are correct or incorrect.
average, 79.1% (8,320 of 10,517) of articles mentioned one
Although undeniably circular, words mean what people mean
theory-related term, with a relatively stable distribution over
by them, and people use words differently. This is especially
time (see the bottom panel of Figure 1 for a breakdown by
true of theory. No one scholar, nor do a collection of scholars,
journal). Of the 60,727 times that one of the five theory-
become the definitive authority or arbitrator on what theory
related terms appeared in the sample, over half were related
is and what it is not.
to the word theory alone (55.1%; 33,462/60,727). A thematic
Definitional diversity in conceptually constituting theory is
review of these cases revealed that theory is used in a variety
intellectually rich. In modern intellectual thought, different
of ways by communication researchers. The word is attached
specialties and perspectives are welcomed and valued. The
to specific social scientific theories (e.g., Construal Level
Theory, Social Identity Theory), the term is used abstractly to alternatives to diversity in theory definitions are hegemony
feign the appearance of theory (e.g., “message processing the- and the demise of academic freedom. Treasured intellectual
ory,” “organizational communication theory”), the word the- diversity, however, comes at the cost of potential misunder-
ory serves as sign-posts in academic papers (e.g., “in the standing stemming from people using the same word to mean
theory section above”), and finally, the term attempts to mark so many different things (e.g., see Bem, 2003). Valuing intel-
one’s contributions (e.g., “has several implications for theory lectual diversity also requires us to abandon rigidity in pre-
and research on selective exposure”).1 Together, the ubiquity scribing fixed rules of doing communication theory and
of theory and theory-related terms, we believe, stems from research.
and reflects expectations for publishing norms that value the- Although there can be no one universal definition of theory,
ory in the field of communication. we agree with DeAndrea and Holbert (2017) that regardless
Our findings align with those of Slater and Gleason (2012) of approaches, scholars should strive for greater clarity in
who examined articles in three elite communication journals what they mean by theory and their theoretical contribution.
in 2008–2009. They report that a sizable number of articles Diversity in definitions should be expected, but clarity and the
advanced theories in at least one of several ways. They thoughtful explication of one’s approach to theory should
reported, for example, that 54% of the articles they examined also be expected in any social science.
in Journal of Communication, Human Communication The lack of a shared definition for theory puts communica-
Research, and Communication Research addressed boundary tion scholars in a Catch-22. Communication scholars value
conditions, 40% expanded a theories range of application, theory and appreciate approaching theory with rigor and clar-
22% advanced a mechanism, and 12% revised a theory. ity. Communication scholars also value intellectual diversity,
Alternatively, theoretical contributions, such as creating a and valuing diverse perspectives and approaches prevents
new theory, comparing theories, and synthesizing across theo- scholars from imposing their own views of theory on other
ries, were less common occurring in only 8% of articles scholars. We find that an “anything goes” approach to theory
combined. intellectually troubling, but we are equally disturbed by im-
We note, however, that mentioning at least one theory- posing views and perspectives on other scholars. While we do
related term does not mean that the work counts as not see an easy way out of this conundrum, we see much
156 Researching and understanding human communication

Downloaded from https://academic.oup.com/hcr/article/50/2/154/7311752 by guest on 30 April 2024

Figure 1. The rate of mentioning theory in communication science articles.

value in acknowledging that it exists and being thoughtful answer “why” can count as theory. Sometimes, although we
about how we balance conflicting values. personally think this goes too far, simply adding the word the-
When scholars define theory, perhaps the most notable di- ory to a topic or phenomenon (e.g., media theory, aggression
mension of variability for the word theory is narrowness- theory, language theory) can pass as theory or at least provide
breadth. At the wide end of this continuum, theory is synony- the appearance of theory.
mous with being minimally conceptional. Explicating a con- We prefer a much narrower use of the term, where theory
struct with a conceptual definition could constitute theory can be considered a set of logically coherent and inter-related
under some of the more expansive uses of the term (cf. Slater propositions or conjectures that (a) provide a unifying explan-
& Gleason, 2012). Similarly, at the broad end of the contin- atory mechanism and (b) can be used to derive testable and
uum, theory can be synonymous with explanation. Efforts to falsifiable predictions. In this relatively narrow view, thinking
Human Communication Research, (2024), Vol. 50, No. 2 157

conceptually or just explaining is not enough. Specifying a about a topic. We also answer “no” in a narrower sense of
path model or set of mediated links might or might not count theory. In explaining why not, we acknowledge that it would
as theory. We note that some scholars may use the word the- also always be better if we had at least one good theory than
ory even more restrictively, desiring to limit the term to for- if we did not. Nevertheless, a well-articulated and relevant
mal axiomatic theories, and considering work to count as theory is not currently available for every conceivable topic or
theoretical only if it conforms to a strict hypothetico- hypothesis worthy of investigation. It is not hard to imagine
deductive depiction (or caricature) of science. Slater and useful and enlightening scholarship that is neither formally
Gleason (2012) provide a definition similar to ours. They see engaged in theory building nor explicitly testing an existing
“the primary role of theory in communication science as the theory. Simply put, if one is interested in a question or phe-
provision of explanation, of proposing causal processes, the nomenon where suitable theory is currently lacking, this
explanation of ‘how’ and ‘under what circumstances’ in ways ought not preclude research. Consequently, it follows that not
that result in empirically testable and falsifiable predictions” all valuable scholarship requires theory in the narrow sense.
(p. 216).
The point here, however, is not to advocate for particular defi-
nition of theory but instead to argue that the efforts to impose a The chicken or the egg: theory or data first?

Downloaded from https://academic.oup.com/hcr/article/50/2/154/7311752 by guest on 30 April 2024

universal definition of theory is fundamentally misguided. Which comes first, theory or data? The answer is it can be ei-
Appreciation of theory requires recognition of the diversity of ther, or the two can work together in an iterative, interactive,
approaches to theory and a willingness to be respectful of and abductive process (Rozeboom, 2016). Different disci-
approaches other than one’s own. The topic of theory needs to plines and specialties put a different emphasis on the primacy
be approached with a cognizance of its diversity. Arguments of theory in empirical research. Communication, on the one
about whose definition of theory is best will often be counter- hand, sometimes views a strict hypothetico-deductive dogma
productive. More constructive arguments will provide cogent as the ideal for formal theory testing, presupposing an exist-
reasons why an approach to defining theory is best for the ing formal theory from which to derive hypotheses.
intellectual endeavor to which it is applied. Computer science, on the other hand, is typically less strict in
its placement or appreciation of theory in the research pro-
Is theory really necessary? cess. Quite often, computer scientists will obtain data, analyze
them, and then identify the theory or theories that fit the find-
Now that we have embraced the ambiguity inherent in defin-
ings as a final step. A communication scholar may scoff at
ing theory, we next contemplate the necessity of theory. If we
this research process, though norms are powerful drivers of
cannot provide a consensus definition of theory, then how
behavior (Cialdini, 2006), and conventions related to theory
can we demand or test it? Do (or would) we even know the-
are to be appreciated and scrutinized within the context of a
ory when we see it?
discipline, specialty and even sub-specialty.
In our view, the necessity of theory varies according to the
Building theory can be a purely logical process, but we are
breadth of the term’s use. Being minimally conceptual is prob-
likely to develop more and better communication research if
ably a prerequisite for making a scholarly contribution. After
all, understanding what one is studying is typically either a relevant data from exploratory research is available.
prerequisite for, or a desired outcome of, advancing knowl- Exploratory research, we contend, is not synonymous with
edge. If scholarly activities — such as concept explication, cre- being atheoretical. We tentatively define exploratory research
ating a new measure, description, observation, and as research guided by curiosity and seeking to document a
hypothesis generation — indeed count (see Slater & Gleason, finding or set of findings rather engaging in hypothetico-
2012), then requiring theory seems constructive. deductive hypothesis testing or focusing on explanation. We
One can imagine empirically documenting an effect or phe- further note that not all hypothesis tests are theoretical. The
nomenon whose explanation is not yet understood. While this logic behind hypotheses often takes the form of “others have
might not count as a theoretical advancement under most found this, therefore we will too.” Such research falls in be-
uses of the term, it might nevertheless make a valuable contri- tween more purely exploratory work and explicit theory test-
bution to knowledge. If nothing else, we typically need to ing where hypotheses follow from theory.
know what needs explaining before we go about explaining it We contend that exploration is symbiotic with and often
(Rozin, 2001). Thus, disregarding the contribution of scholar- contributes to theory because it can highlight relationships
ship that is not “full-on” theory in some narrow sense is coun- that were unanticipated by theory, offering new hypotheses
terproductive to the advancement of knowledge. for future research. Even purely descriptive research can pro-
Park et al. (2005) provide an instructive example. Their vide an understanding of the phenomena of interest, thereby
first study simply documents the existence of a strong finding. providing a solid empirical foundation for conceptual con-
Unlike in the United States, Korean spam emails often contain struction. The placement of theory in the research process is
an apology. What follows are five experiments testing various not specifically a statement about the work’s value or rigor; it
explanations before settling on a normative account. The likely emphasizes the goals and norms of a particular research
work is not grounded in a specific theory, but it is clearly a community. Consistent with our views on defining theory, we
systematic effort to document and explain a communication encourage our colleagues to be ecumenical in approaching
phenomenon. What if, however, they had packaged their theory-data time ordering.
studies as a series of articles rather than in one. Would this
make the work any more or less theoretical?
One of the more controversial, meta-scientific questions in Will any theory do?
communication science is: Must we have theory? We answer An even more difficult question asks if all theories are equal.
“yes” in the broadest sense, as it helps to clarify our thinking If some theories are indeed better than others, then what
158 Researching and understanding human communication

makes them so? Are there instances when no applicable the- questionable research practices such as p-hacking, and publi-
ory is preferred to a misapplied or unreliable theory? cation bias all combine to make the empirical merit of any
At the risk of diverting from our previous, more ecumenical claim murky at best (Dienlin et al., 2021; Lewis, 2020;
perspective, we will tentatively take the position that some Markowitz et al., 2021). Accumulating more data over time
theories are indeed preferable to other theories—at least for often further muddies the water as mixed findings pile up and
certain applications—and along certain criteria of evaluation. multiple citations can legitimately be provided in support of
For example, DeAndrea and Holbert (2017) expanded on incompatible empirical claims. Not even meta-analysis is im-
Chaffee and Berger’s (1987) list of criteria for evaluating the- mune. As prior work shows (Levine & Weber, 2020), regard-
ory. Their refined list includes explanatory power, predictive less of the topic, findings in communication are
power, parsimony, falsifiability, logical consistency, heuristic heterogeneous, and the heterogeneity is seldom resolved by
value, and organizing power. Building on this work, we fur- moderator analysis. In this way, meta-analytic results often
ther cautiously propose that theory can do more harm than document rather than resolve conclusions of mixed support
good when it is misapplied, used haphazardly, or thrown at for theoretical predictions. The net result is that a theory’s
data to see if it sticks. If the goal of scholarship is the pursuit supporters and critics can both provide plenty of citations for
of knowledge and understanding, it seems possible that cer- why the theory is well supported and clearly falsified.

Downloaded from https://academic.oup.com/hcr/article/50/2/154/7311752 by guest on 30 April 2024

tain frames, stances, models, and understandings might be
counterproductive or misguided.
From our perspective, a first consideration regarding the What is a theoretical contribution?
utility of theory is one of relevance. Does the sphere of appli- A common concern in academic research publishing is to ar-
cation fall within the boundary conditions of the theory, or ticulate how one’s work makes a substantial theorical contri-
does the application involve interrogating the boundary condi- bution (DeAndrea & Holbert, 2017; Slater & Gleason,
tions of the theory? If the answer to both questions is no, then 2012). Articles in flagship, high-impact communication jour-
the application is probably ill-advised on the grounds of rele- nals are often rejected if theoretical contributions are not sub-
vance. Irrelevant theory distracts from empirical contributions. stantial and clearly expressed. For example, the Journal of
This is the theoretical equivalent of a red herring argument. Communication suggests “Submissions are expected to pre-
The second test is more difficult and involves a cost–benefit sent arguments that are theoretically sophisticated, conceptu-
analysis of gains and losses in knowledge, insight, and under- ally meaningful, and methodologically sound” (Journal of
standing. Consistent with commentators such as Levine and Communication, 2022). The words sophisticated and mean-
McCornack (2014), we envision evaluative dimensions, such ingful in their instructions for authors are subjective and elu-
as clarity, coherence, and verisimilitude, in assessing the sive. Considering the subjective aspect of such appraisal, what
scholarly value added by a theory. The more that a theory does it mean to make a theoretical contribution?
clarifies rather than clouds our understanding, the more valu- Given that we have argued so far that defining theory is
able it is. Theory can bring order to otherwise unruly facts, find- misguided, that formal theory is not necessary for all research
ings, and ideas, or it can lead to logical inconsistencies, the endeavors, and that unequivocally establishing the empirical
latter obviously being less desirable. The insights offered by the- merit of theory is nearly impossible, one might expect an ar-
ories can align with known facts and findings or it can be con- gument dismissing the very idea of theoretical contribution. A
tradicted by data and evidence (Levine & McCornack, 2014). close read of what preceded, however, conveys several ways
In practice, assessing the alignment of theory with data is to make a theoretical contribution (cf. Slater & Gleason,
an especially thorny issue in quantitative, social scientific 2012). Clarifying a conceptualization, providing an explana-
communication research. Not all scholarship strives to be em- tion, making or testing a prediction, testing theoretical bound-
pirical nor scientific, and theory-data alignment might not be ary conditions, articulating a unifying framework integrating
the point in many scholarly endeavors. But when it is, theory- two or more seemly unrelated facts, and identifying a modera-
data alignment quickly becomes deeply problematic in the ac- tor that resolves previously unexplained heterogeneity can all
tual practice of communication scholarship, particularly count as theoretical contributions if done in a way to be con-
when inferential statistics, and especially p-values, are in- ceptually coherent. One can seek to create new theory, pit
volved (Denworth, 2019). existing theories against each other, or reconcile apparently
One issue concerns “undead theories” (Ferguson & Heene, conflicting theories. In our view, all such outcomes can offer
2012). It is not unusual in the social sciences for theoretical new knowledge that conceptually builds on an existing foun-
predictions to be soundly falsified, yet, nevertheless, applied dation of empirical findings.
despite their documented empirical deficiencies. Such theories Critically, a theoretical contribution is different from dis-
are functionally sets of counterfactual conjectures that are covering a new, statistically significant finding. Moving from
passed off as good science. We anticipate that the reader will empirical findings to theoretical contribution involves answer-
have their favorite undead theory, but we also anticipate that ing questions related to the mechanism underlying the finding.
one scholar’s undead theory is another’s source of wisdom. How does the finding fit within the larger nomological net-
Both can be true, which we appreciate, and will explain. work of findings in the domain (Cronbach & Meehl, 1955)?
While the replication crisis in the social sciences has become What are the limits of the finding? How robust is it? How far
an increasingly recognized issue (Open Science Collaboration, can it be generalized? What are its moderators and antece-
2015), it has also long been recognized that modern social sci- dents? These questions, among many others, may help to po-
ence practices ensure that almost any hypothesis will receive sition a finding better as a theoretical contribution instead of
mixed support regardless of its validity or verisimilitude an empirical one-off result.
(Meehl, 1978). Essentially, the fact that the nil-null hypothesis The most basic types of theoretical contributions are con-
is never literally true regardless of the soundness of the theory ceptualizing or explicating a new construct, reconceptualizing
(Meehl’s crud factor), sub-optimal statistical power, an existing construct, or providing a new explanation for an
Human Communication Research, (2024), Vol. 50, No. 2 159

empirically documented effect. These types of contributions Generality and external validity
might be considered theoretical building blocks for subsequent Perhaps the most underappreciated benefit of theory is that it
theoretical development. Although these types of contributions can provide satisfying answers to questions of generality in
may also be seen as just minimally theoretical, they are never- ways that data simply cannot. Theory is a better approach to
theless important because other types of theoretical contribu- achieving external validity than research design.
tions require well explicated components and explanations. We have all seen data collected on college students and
Coherent conceptual structures can lead to testable and falsifi- wondered if the findings might apply to working adults. We
able hypotheses about human communication and logically co- all likely agree that for most topics, a nationally representative
herent networks of hypotheses can lead to formal theory. sample is preferable to a sample of college students (though,
A second approach to theoretical contribution involves var- see Coppock et al., 2018). Nevertheless, we might still wonder
iations on theory creation. Arguments for the desirability of a that if the data were collected at a different point in time, or if
new theory will often take one of three forms. The first notes the questions were worded a bit differently, or perhaps pre-
the absence of a relevant theory for a given topic or purpose. sented in a different order, would the findings be the same?
If no relevant theory exists and if theory is desired, then it fol- These sorts of questions cannot be answered with data be-
lows that theory creation is needed. The second type of argu- cause we can never sample everyone everywhere over all times

Downloaded from https://academic.oup.com/hcr/article/50/2/154/7311752 by guest on 30 April 2024

ment rests on making the case that existing, relevant theory is in all possible ways. No matter how much data we have, data
deficient, and the deficiencies are both sufficiently severe and are finite, and representative sampling and inferential statis-
intractable to justify a new theory as a rival. Third, prior the-
tics do not change this uncomfortable truth.
ory can be accepted, but arguments are made that the new
Fortunately, theory provides an elegant solution.
perspective offers additional insights that would not otherwise
Theoretical claims specify what is expected under what condi-
be gained.
tions. Theory, and more precisely its boundary conditions,
Once a theory exists in the literature, it is often the goal of
provide us with statements of the extent of generality. As de-
communication research to test, extend, modify, or apply a
scribed by Mook (1983), we specify generality theoretically,
theory to improve our understanding of human communica-
then we test and validate claims of generality with data.
tion. Each of these (testing, extending, modifying, or apply-
Rather than fretting over the sampling of participants, multi-
ing) moves communication theory forward. We note,
ple message instantiations, multiple situations, and a host of
however, that at least for scientific research, testing should
other study-specific idiosyncrasies, we use theory to make
typically precede the other forms of contribution to ensure
generalizations and data to test those generalizations
theoretical adequacy prior to extension or application.
(Ewoldsen, 2022). We could ask if a theory applies to non-
WEIRD cultures, and then test core claims with a non-
Theoretical bandwidth WEIRD culture (Henrich et al., 2010; Many Labs 2, 2018).
Many discussions of theoretical contributions will involve
Theory as agenda setting
value judgments regarding theoretical bandwidth. Discussions
of theoretical bandwidth, in turn, may deal with two qualita- A second underappreciated function of theory is research
tively different issues. The first relates to how theory is de- agenda setting. Just as the media might tell us which news
fined. One might think of explicating a construct as a topics and frames are important, so too does theory tell us
narrower contribution than explaining the relationship be- what we need to study, how to study it, and what to expect. It
tween two explicated constructs. Explaining how a well- is not unusual for new researchers to struggle with topic selec-
understood effect fits with a network of documented effects is tion. Theory provides a straight-forward way to come up
broader still. with a hypothesis and an approach to testing it. This topic se-
Second, communication theories vary widely in their topi- lection approach can also be flipped. We can ask, what if a
cal scope and boundary conditions. Communication theories theory was wrong? How might we show that? A research de-
might focus on a particular topic or phenomenon, others on a sign should flow from these questions.
broader domain or function, and others still might be general Theory offers an even more important agenda setting func-
theories of communication. Further, regardless of topical tion. As Berlo (1960) famously identified when defining com-
breadth, boundary conditions can vary. Communication the- munication as process, a wide variety of forces can affect how
ories, for example, might be limited to a particular age group, communication unfolds. Regardless of the specific topic or fo-
point in time, media, or culture. cus, the potentially important considerations are numerous.
It is likely tempting to equate theoretical bandwidth and Theory tells us what is most important and what is less impor-
theoretical contribution under the likely tacit presumption tant. In other words, theory tells us what to prioritize.
that more is better. While surely there are knowledge-gain
advantages to breadth, any firm link between breadth and
contribution is qualified by all other things being equal.
The current state of communication theory
Surely contribution is more closely tied to how well a theoreti- We are more pleased than not with the current state of com-
cal goal is accomplished than to how ambitious the goal is munication theory as its progress is undeniable. There was a
(cf., DeAndrea & Holbert, 2017). time when the lack of communication theory was bemoaned
and when most theories were taken from other disciplines (see
Berger, 1991). Our perception of the current literature,
What are the benefits of theory? formed by our lived experience across decades of publishing
Rather than reviewing the extensive literature on the value of and reading communication scholarship, is that the number
theory, we focus here on two benefits of theory that we be- of communication theories and theoretical ideas have grown,
lieve are highlighted less frequently but are no less important. and the communication trade deposit with related fields has
160 Researching and understanding human communication

diminished. The latter point, of course, deserves a strict em- from theory is not an admission of a research study being athe-
pirical evaluation to test how communication and other social oretical, but instead, an admission of one’s curiosity and uncer-
sciences share ideas and theories. tainty. Thus, we envision a future where exploratory and
descriptive work is more prevalent and more appreciated.
It is also important for authors to think about and explicitly
A caveat communicate the role of theory in their research. This article
As the reader has surely noticed, we have approached this ar- has noted the many functions that theory can serve; yet, these
ticle from a particular perspective. The present authors have a functions are often assumed or implied in a manuscript when
quantitative and social scientific approach to communication they could be made explicit. Being forthcoming about the role
scholarship. A consequence of originating from this scholarly of theory in one’s research will lead to conceptual and
tradition is that our commentary is better targeted for re- contribution-related clarity. This will lead to less superficial
search publishing in outlets such as Human Communication applications of theory (e.g., theoretical name dropping) and
Research than outlets like Communication, Culture, & toward more conceptual richness. If communication research
Critique. Both are worthy outlets, but they have different ori- is to value theory—and we undoubtedly think it should—then
entations and conventions. theory should be used appropriately. Theories are built on a

Downloaded from https://academic.oup.com/hcr/article/50/2/154/7311752 by guest on 30 April 2024

foundation of empirical evidence, collected over time allowing
researchers to draw nuanced conclusions and make subse-
Looking forward
quent predictions about human communication. Using the
Like most communication scholars, we are theory advocates. term theory to sound more scientific, rigorous, or grounded is
We use, have written our own, and made contributions to the- gratuitous and should be avoided. Consequently, we envision
ory on a range of topics relevant to human communication. a future where communication theory, in form and function,
While we ascribe to the idea that “there is nothing so practical is used more thoughtfully and transparently.
as a good theory” (Lewin, 1951, p. 169), blind allegiance to Finally, we are encouraged that all major communication
theory is ill-advised. Theories, we believe, must have testable research journals have a large focus on theory in their articles
and falsifiable components to them. We encourage our fellow (e.g., at least 50% of articles in each journal mention theory
communication scholars to “follow the data.” Moving science in some manner; Figure 1). However, the degree to which the
forward requires theoretical predictions that hold up to data published communication literature is advancing theory in
over time. consequential ways or settings is unclear (DeAndrea &
Replications play an increasingly important role in theory Holbert, 2017). We encourage scholars to be flexible with
testing, but also add a final set of complications to address. their assumptions about a theory, testing it in ways that might
Theory and evidence can misalign for several reasons, and it be unconventional and creative in the pursuit of new knowl-
is usually unclear why a test failed. Perhaps some critical as- edge. To this end, null effects are still informative (Francis,
pect of the research setting was different, producing an unex- 2012; Levine, 2013), especially if a study is adequately pow-
pected result. A moderating variable may have impacted the ered. For example, understanding what leads to null effects
results, such that the findings do not invalidate the theory, but might be helpful for the development of boundary conditions
instead provides a nuanced understanding of the conditions of theory. Null effects are difficult to publish, but communica-
that led to a particular effect and those that did not (or led to tion research can lead in their normalization in the pursuit of
the opposite effect). Theory should be a guidepost for empiri- greater theoretical precision and explication. Thus, we envi-
cal research, not gospel, upon considering the results. Of sion a future where researchers are more frank about empiri-
course, theory and data can also misalign because the theory cal support, and more precise with predictions.
is mis-specified. In practice, it can be difficult to discern valid
support from false positives and mis-specified predictions
from a methodological artifact or undetected moderators. Data availability
Nevertheless, we envision a future where replication is both Data related to Figure 1 can be retrieved from Markowitz
more prevalent and more valued. et al. (2021) or by contacting David M. Markowitz.
Communication is an eclectic discipline, and science is not
the only method for understanding communication. Further,
we as a field draw on and adopt ideas from different fields, Funding
authors publish in journals outside of communication research, The authors received no financial support for the research,
and there is no singular approach to the same research ques- authorship, and/or publication of this article.
tion. We encourage authors to continue this tolerance and flex-
ibility with exploratory and “pre-theory” work as well. As
mentioned, there are times when a good theory simply does Conflicts of interest
not fit one’s phenomenon of interest. Communication scholars The authors also report no conflicts of interest with the re-
should not be faced with a “square peg, round hole” problem search, authorship, and/or publication of this article.
just to satisfy reviewers who demand more theory. One can try
to fit a square peg into a round hole with enough force, but it
will not fit well and there may be important consequences be- Note
cause of this exercise (e.g., theory–data misalignment). 1. A systematic, qualitative review of how theory terms were used across
Exploratory work should be considered and applauded when the entire sample is beyond the scope of this commentary. We used these
we simply do not know how concepts will relate to each other. data to descriptively demonstrate how theory is prevalent in communi-
Proposing a research question instead of hypotheses derived cation research, and used as a means to achieve different ends.
Human Communication Research, (2024), Vol. 50, No. 2 161

References Levine, T. R. (2013). A defense of publishing nonsignificant (ns) results.

Communication Research Reports, 30(3), 270–274. https://doi.org/
Bem, D. (2003). Writing the empirical journal article. In J. M. Darley & 10.1080/08824096.2013.806261
M. P. Zanna (Eds.), The complete academic: A practical guide for Levine, T. R., & McCornack, S. A. (2014). Theorizing about deception.
the beginning social scientist (2nd ed.). (pp. 185–219). American Journal of Language and Social Psychology, 33(4), 431–440. https://
Psychological Association. doi.org/10.1177/0261927x14536397
Berger, C. R. (1991). Communication theories and other curios. Levine, T. R., & Weber, R. (2020). Unresolved heterogeneity in
Communication Monographs, 58(1), 101–113. https://doi.org/10. meta-analysis: Combined construct invalidity, confounding, and other
1080/03637759109376216 challenges to understanding mean effect sizes. Human Communication
Berlo, D. K. (1960). The process of communication: An introduction to Research, 46(2–3), 343–354. https://doi.org/10.1093/hcr/hqz019
theory and practice. Holt, Rinehart and Winston. Lewin, K. (1951). Field theory in social science: Selected theoretical
Chaffee, S. H., & Berger, C. R. (1987). What communication scientists papers. Harper & Brothers. https://psycnet.apa.org/record/1951-
do. In C. R. Berger & S. H. Chaffee (Eds.), Handbook of 06769-000
Communication Science (pp. 99–122). Sage. Lewis, N. A. (2020). Open communication science: A primer on why
Cialdini, R. B. (2006). Influence: The psychology of persuasion. Harper
and some recommendations for how. Communication Methods and
Business.
Measures, 14(2), 71–82. https://doi.org/10.1080/19312458.2019.
Coppock, A., Leeper, T. J., & Mullinix, K. J. (2018). Generalizability of

Downloaded from https://academic.oup.com/hcr/article/50/2/154/7311752 by guest on 30 April 2024

1685660
heterogeneous treatment effect estimates across samples.
Many Labs 2. (2018). Many Labs 2: Investigating variation in replicabil-
Proceedings of the National Academy of Sciences, 115(49),
ity across samples and settings. Advances in Methods and Practices
12441–12446. https://doi.org/10.1073/pnas.1808083115
in Psychological Science, 1(4), 443–490. https://doi.org/10.1177/
Cronbach, L. J., & Meehl, P. E. (1955). Construct validity in psychologi-
2515245918810
cal tests. Psychological Bulletin, 52(4), 281–302. https://doi.org/10.
Markowitz, D. M., Song, H., & Taylor, S. H. (2021). Tracing the adoption
1037/h0040957
and effects of open science in communication research. Journal of
Denworth, L. (2019). The significant problem of p-values. Scientific
Communication, 71(5), 739–763. https://doi.org/10.1093/joc/jqab030
American. https://doi.org/10.1038/scientificamerican1019-62
Meehl, P. E. (1978). Theoretical risks and tabular asterisks: Sir Karl, Sir
DeAndrea, D. C., & Holbert, L. R. (2017) Increasing clarity where it is
Ronald, and the slow progress of soft psychology. Journal of
needed most: Articulating and evaluating theoretical contributions.
Annals of the International Communication Association, 41(2), Consulting and Clinical Psychology, 46(4), 806–834. https://doi.org/
168–180. https://doi.org/10.1080/23808985.2017.1304163 10.1037/0022-006X.46.4.806
Dienlin, T., Johannes, N., Bowman, N. D., Masur, P. K., Engesser, S., Miller, G. R., & Nicholson, H. E. (1976). Communication inquiry: A
Kümpel, A. S., Lukito, J., Bier, L. M., Zhang, R., Johnson, B. K., perspective on a process. Addison-Wesley Pub. Co.
Huskey, R., Schneider, F. M., Breuer, J., Parry, D. A., Vermeulen, I., Mook, D. G. (1983). In defense of external invalidity. American
Fisher, J. T., Banks, J., Weber, R., Ellis, D. A., . . . de Vreese, C. Psychologist, 38(4), 379–387. https://doi.org/10.1037/0003-066X.
(2021). An agenda for open science in communication. Journal of 38.4.379
Communication, 71(1), 1–26. https://doi.org/10.1093/JOC/JQZ052 Park, H. S., Lee, H. E., & Song, J. A. (2005). “I am sorry to send you
Ewoldsen, D. R. (2022). A discussion of falsifiability and evaluating re- spam” Cross-cultural difference in email advertising in Korea and
search: Issues of variance accounted for and external validity. Asian the US. Human Communication Research, 31(3), 365–398, https://
Communication Research, 19(2), 5–14. https://doi.org/10.20879/ doi.org/10.1111/j.1468-2958.2005.tb00876.x
acr.2022.19.2.5 Open Science Collaboration. (2015). Estimating the reproducibility of
Ferguson, C. J., & Heene, M. (2012). A vast graveyard of undead theo- psychological science. Science, 349(6251), aac4716. https://doi.org/
ries: Publication bias and psychological science’s aversion to the null. 10.1126/science.aac4716
Perspectives on Psychological Science, 7(6), 555–561. https://doi. Rozeboom, W. W. (2016). Good science is abductive, not hypothetico-
org/10.1177/1745691612459059 deductive. In L. L. Harlow, S. A. Mulaik & J. H. Steiger (Eds.), What
Francis, G. (2012). Publication bias and the failure of replication in ex- if there were no significance tests? (pp. 335–391). Routledge.
perimental psychology. Psychonomic Bulletin & Review, 19(6), Rozin, P. (2001). Social psychology and science: Some lessons from
975–991. https://doi.org/10.3758/s13423-012-0322-y Solomon Asch. Personality and Social Psychology Review, 5(1),
Henrich, J., Heine, S. J., & Norenzayan, A. (2010). The weirdest people 2–14. https://doi.org/10.1207/S15327957PSPR0501_1
in the world? Behavioral and Brain Sciences, 33(2–3), 61–83. https:// Slater, M. D., & Gleason, L. S. (2012). Contributing to theory and
doi.org/10.1017/S0140525X0999152X knowledge in quantitative communication science. Communication
Journal of Communication. (2022). Author instructions. Oxford Methods and Measures, 6(4), 215–236. https://doi.org/10.1080/
Academic. https://academic.oup.com/joc/pages/General_Instructions 19312458.2012.732626