723060

research-article2017
IRE0010.1177/0047117817723060International RelationsPashakhanlou

Article

International Relations
119
Fully integrated content The Author(s) 2017
Reprints and permissions:
analysis in international sagepub.co.uk/journalsPermissions.nav
DOI: 10.1177/0047117817723060
https://doi.org/10.1177/0047117817723060
relations journals.sagepub.com/home/ire

Arash Heydarian Pashakhanlou

Forsvarshogskolan

Abstract
Content analysis has once again come to the forefront of discussions regarding methods in
International Relations (IR). The first wave of content analysis in IR lasted from the 1940s to
the 1960s and was marked by a commitment to quantitative and manual analyses. The second
wave of content analysis appeared around the third millennium and continues to pervade the
discipline also proceeds in a predominantly quantitative manner but emphasizes computer-
assisted analysis rather than manual analysis. Critics and advocates of the method alike have,
highlighted numerous shortcomings with these approaches. In order to address these limitations,
the present investigation argues for a fully integrated content analysis that has the potential to
ameliorate the identified weaknesses that have hitherto plagued the method. It accomplishes
this task by combining all facets of the method: quantitative, qualitative, manual, and computer-
assisted content analyses within a single research project.

Keywords
computer assisted, content analysis, context, international relations, interpretation, manual,
method, objectivity, qualitative, quantitative, reliability, replication, validity

Introduction
Systematic explorations of communication were conducted long before content analysis
became a formal method in the social sciences. Indeed, the origin of content analysis is
usually traced back to word-frequency analyses of the Bible throughout the centuries.1
The emphasis on religious scriptures was eventually replaced by an increasing focus on
the analysis of newspapers in the nineteenth century.2 Despite its popularity, the term
content analysis did not enter the English language until 1940.3 The following year,

Corresponding author:
Arash Heydarian Pashakhanlou, Forsvarshogskolan, Drottning Kristinas vg 37, 115 93 Stockholm, Sweden.
Email: [email protected]
2 International Relations 00(0)

content analysis was the central topic in a conference on mass communication held in
Chicago.4 The first major introduction to the method appeared 7years later, and in 1952,
Bernard Berelson published Content Analysis in Communication Research which became
the leading monograph on the method.5
Over time, content analysis became increasingly used in other disciplines outside of
communication studies and entered a wide variety of fields, including International
Relations (IR).6 As the method gained popularity across different disciplines, it started to
move beyond its initial preoccupation with counting the frequency of words, concepts,
or themes to study the interrelation between issues instead.7 Nowadays, content analysis
is used in a wide variety of ways and is becoming increasingly prevalent, mainly due to
the spread and advances in computer and Internet technology.8
In this article, the emphasis is on content analysis in IR.9 Although this method has been
prominent in the study of international politics, there is no historical overview and up-to-
date analysis of its development. The present investigation addresses this gap in the exist-
ing body of work by outlining the two waves of content analysis in IR and highlighting
their limitations. In doing so, it argues that the first wave of content analysis spanned from
the 1940s to the 1960s and mainly proceeded manually and quantitatively. This approach
suffered from numerous shortcomings as leading IR scholars at the time pointed out. The
second wave of content analysis in IR emerged in the new millennium and has been char-
acterized by its quantitative and computer-centric approach. Even its advocates have, how-
ever, acknowledged its weakness in dealing with interpretation, context, and validity.
This is why the present investigation also provides a roadmap of how the identified
deficiencies of earlier generations of this method can be ameliorated. It does so by argu-
ing for a fully integrated content analysis that combines quantitative, qualitative, manual,
and computer-assisted content analyses within a single research project. Since this
approach makes use of all aspects of the method, it is better suited for handling more
complex issues that cannot be adequately addressed through the partial content analyses
that have hitherto dominated the discipline. Indeed, fully integrated content analysis has
the potential to address more demanding research questions in a more robust and system-
atic manner with greater reliability and validity than its predecessors.
These arguments are advanced throughout the remainder of this article which are
divided into six sections. The first section provides an overview of the different strands
of content analysis. In the following section, the first wave of content analysis in IR is
introduced along with the criticism that it was subjected to. The ensuing section unfolds
in a similar fashion with regard to the second wave of content analysis. A guideline of
how the identified weaknesses by the critics and proponents of content analysis can be
ameliorated by fully integrated content analysis is the topic of the fourth section. This
is followed by a discussion of the research projects and questions that could benefit
from adopting fully integrated content analysis. A practical example of how this
approach could have been utilized with great benefits in a study deemed inappropriate
for content analysis is also considered here. The sixth section illustrates how fully inte-
grated content analysis has been employed in a research project in IR. This article is
brought to a closure by a brief summary of the main points advanced in the preceding
pages and their implications. At this point, it is, however, incumbent to take a closer
look at the method of content analysis.
Pashakhanlou 3

Overview of content analysis

Content analysis is a research method that systematically analyzes the content of com-
munication. As such, any material that conveys a message can be the subject of content
analysis. This includes all forms of oral and visual communication such as articles, web-
sites, diaries, speeches, letters, interviews, images, videos and beyond. Content analysis
can be applied to study any aspect of these communications, whether manifest or latent,10
and shed light on them in line with the set research questions.11 The quality of content
analysis is usually determined by its reliability and validity. Reliability is the ability of a
measure to yield consistent results each time it is used.Validity is concerned with the
extent the instruments capture what they are designed to capture.12 The aim is to produce
the most robust findings possible, even though it is impossible to exclude all potential
sources of error.
Three main factors make content analysis a distinct method of communication. First,
content analysis is predominantly descriptive. It is primarily concerned with what is said
rather than how or why it is said. In other words, content analysis mainly focuses on
what questions, even if it can address how or why questions as well.13 Second, con-
tent analysis systematically reduces the communication.14 This is accomplished by
restricting the investigation to the theme(s), concept(s), or word(s) of interest with the
aim of condensing the communication to more relevant and manageable pieces of data
that can be explored more thoroughly.15 Third, content analysis settles for one interpreta-
tion when several are possible.16
There is a growing tendency to distinguish between quantitative and qualitative con-
tent analysis in the literature, even though it is recognized that their differences should
not be overstated.17 Nevertheless, the main line of division lies in the fact that the former
privileges manifest meaning and the latter latent meaning.18 Since manifest meaning is
easier to pin down, it can typically be established by analyzing a small segment of the
material. Determining latent meaning is more demanding and may require a wider exam-
ination of the communication or additional background information.19 As a result, quali-
tative content analysis makes more inferences to context, authors, and recipients than its
quantitative counterpart. Interpretation is central to qualitative content analysis and
peripheral to its quantitative branch.20 Due to the priority given to manifest meaning, it
is usually easier to gain high reliability with quantitative content analysis, whereas latent
meaning affords an advantage in attaining validity qualitatively.21 Since quantitative con-
tent analysis tends to focus on manifest meaning, it can address what questions, whereas
the emphasis on latent meaning makes qualitative content analysis more suited for deal-
ing with how and why questions as well.
Both strands of content analysis can be performed either manually and/or with the
help of the computer. Before the advent of computer technology, all content analyses
were conducted manually. Despite the technological advancement, a fully automated
computer content analysis is not possible. Human involvement is still essential. In gen-
eral, manual content analysis is particularly useful when dealing with latent meaning
since scholars generally have the interpretative edge over computers, for the time being.
In other words, manual content analysis is particularly well suited for qualitative inquires
and capable of dealing with how and why questions as well as what questions.22
4 International Relations 00(0)

Computer-aided content analysis can provide improved objectivity, reliability, and

replicability as machines do not suffer from individual idiosyncrasies and human errors
rooted in fatigue, loss of concentration, or the like. Computers can also process huge
amounts of data with exceptional speed, and their ability to do so is steadily increasing,
far exceeding that of human beings.23 Computer-assisted content analysis is thus essen-
tial when dealing with Big Data as human beings cannot cope with massive datasets
without the aid of computers.24
Concretely, computer-assisted content analysis can be used to swiftly generate abso-
lute or relative frequency counts of words or other symbols in the communication, mak-
ing it ideal for quantitative investigations and what questions. With explicit instructions,
it can however also differentiate the senses of homographs,25 capture the context as well
as the content of a text passage, analyze phrases or themes, and make evaluations, among
other things. This makes it capable of addressing how and why questions as well,
albeit usually to a lesser extent. Since the computer does exactly what it is told to do and
produces consistent results every time the data are examined,26 reliability essentially
becomes a nonissue when conducting computer-assisted content analysis.27
These are the characteristics of content analysis in general. How this method has been
employed in IR is another story that will be told over the next two sections. The follow-
ing segment is the first step in this endeavor.

The first wave of content analysis in IR

Content analysis entered the field of IR in the 1940s, and the first wave spanned over the
next two decades as well. Ole Holsti, Abraham Kaplan, Harold Lasswell, Robert North,
and Dinna Zinne, among others, were responsible for these developments.28 These scholars
approached content analysis in entirely or predominantly quantitative terms. Eventually,
these efforts came under fierce criticism that led to the demise of simplistic quantitative
content analysis within the discipline. As an example, Kaplan and Lasswell attempted to
quantify the works of Aristotle. In doing so, they focused on 300 sentences in his body of
work and found 25 references to political philosophy29 and 75 statements regarding politi-
cal science.30 From this numerical analysis, Hans Morgenthau claims that Kaplan and
Lasswell deduced that Aristotle was primarily a political scientist.31
As such, Morgenthau criticized their efforts. Indeed, he suggested that it is question-
able whether the 300 selected sentences are representative of Aristotles entire catalogue.
Furthermore, to reduce Aristotles body of work to 300 sentences and thereafter engage
in a mechanic numerical exercise where references to political philosophy and political
science are counted without actually engaging in an in-depth qualitative interpretation of
his work is futile. As a result, Morgenthau dismissed this approach and considered it self-
defeating and claimed that it contains the seed of its own destruction by virtue of its own
absurdity.32
Additionally, Robert Jervis argued that many significant issues in IR can only be
addressed by detailed analyses that quantitative content analysis is unequipped to deal
with. Using frequency as an indicator of importance in quantitative content analysis,
Jervis also considered problematic. This is because there may be various reasons to why
a word, concept, or theme appears more often than others that have nothing to do with its
Pashakhanlou 5

significance. Simply put, one cannot equate frequency with importance. These kinds of
inferences are unjustifiable and may generate fundamentally flawed and invalid results.33
With these points in mind, it is perhaps not hard to realize why Hedley Bull main-
tained that quantitative content analysis is as remote from the substance of interna-
tional politics as the inmates of a Victorian nunnery were from the study of sex.34
Nevertheless, Alexander George sought to salvage the method from the devastating
criticism leveled against it by leading figures in the field as early as 1959. He did so by
proposing an alternative type of content analysis that sought to ameliorate these short-
comings. George referred to this approach as non-frequency content analysis as it was
both non-quantitative and non-statistical in form.35 Georges non-frequency content
analysis however failed to capture the attention of a field under the sway of quantitative
content analysis.36

The second wave of content analysis in IR

Although content analysis continued to be used after the 1960s, elaborate discussions
regarding the method itself became increasingly rare in IR.37 This changed with the
advent of the new millennium that sparked the second wave of content analysis in IR.
Contributors to the second wave includes Margaret Hermann, Johan Eriksson, Giampiero
Giacomello, and Andrew Bennett, among others.38 The renewed interest in content anal-
ysis in the discipline has undoubtedly been rooted in advances in information and com-
puter technology. In the first instance, its advocates argue that the vast amount of
communication almost instantly accessible on the Web has markedly improved the ease
and time required for data collection in content analysis. Combined with the progress in
computer power and technology, Big Data can also more easily and accurately be ana-
lyzed by this method than ever before.39 In short, second wave content analysis in IR is
committed to computer-assisted content analysis.
To be sure, computer technology was employed in the 1940s1960s as well. The use
of computer-aided content analysis during this period was however far more limited as
one would expect, and Internet had not yet been invented by that time.40 As such, manual
quantitative content analysis was the main mode of operation during this period. In con-
trast, it is hard to overemphasize how crucial the technological development has been to
second wave content analysts in IR. Bennett even claims that he had previously ruled out
this method due to the lack of digital texts, machine reading, and underdeveloped algo-
rithms for computer-aided content analysis. Due to vast improvements in these areas,
Bennett professes that he has come to revaluate his former skepticism towards content
analysis and become a proponent of the method instead.41
Nevertheless, the second wave content analysis that Bennett advocatesshares the pre-
vious generations emphasis on quantitative methods.42 This is so even though second
wave content analysts themselves frequently acknowledge its limitations in dealing with
interpretation, context, and validity.43 In an attempt to ameliorate these weaknesses, they
typically argue that computer-aided quantitative content analysis should be comple-
mented with qualitative methods such as argumentation analysis, narrative analysis, and
especially discourse analysis.44 In doing so, they overlook the possibilities of combining
quantitative content analysis with its qualitative counterpart.
6 International Relations 00(0)

This omission is hardly surprising, considering that content analysis has traditionally
been understood in quantitative terms in the Anglo-American world.45 Indeed, the domi-
nance of quantitative content analysis has been so prevalent that many scholars are not
even aware of the existence of qualitative content analysis in this part of the globe.46
Even those who have considered this approach have often given it different names such
as Georges non-frequency content analysis discussed previously or used other labels
such as thematic coding or qualitative media analysis.47 To be sure, there has been a
growing awareness of qualitative content analysis in the English literature in more recent
years, but it has not received widespread recognition.48 Since IR has been dominated by
the Anglo-American world where qualitative content analysis has been marginalized, it
is not surprising to find that qualitative content analysis has not made its way to main-
stream IR either.
The situation in Germany has been radically different. There qualitative content anal-
ysis has flourished ever since, at least, the 1970s. The development has even led to the
creation of different types of qualitative content analysis such as strict, flexible, and
summative qualitative content analysis, to mention a few.49 This is presumably why it
is typically scholars of German descent that have shown the greatest awareness of quali-
tative content analysis in IR, even though they are not the only ones to do so.50 None of
these scholars have, however, built on this insight and introduced qualitative content
analysis to a wider audience in the field.
This is unfortunate, since the neglect of this approach impoverishes inquiries in IR.
For instance, Keren Yarhi-Milo explicitly dismisses content analysis for her research
project on the basis that this method cannot provide satisfactory contextualized under-
standing of the inference process.51 She is seemingly unaware that qualitative content
analysis can be useful for this kind of investigation. Moreover, the second wave scholars
concerns with quantitative content analysis limitations in interpretation, contextualiza-
tion, and validity in, which they seek to ameliorate by combining it with other methods,
particularly discourse analysis, can at times be better solved with the addition of qualita-
tive content analysis.
As we have seen, qualitative content analysis tends to be stronger than its quantitative
counterpart in these capacities and can therefore supplement it to enhance the interpreta-
tion, contextualization, and validity of the study. To be sure, not even qualitative content
analysis is as comprehensive as discourse analysis in these areas. With that said, not all
research projects require such a depth. The added interpretation, contextualization, and
validity provided by qualitative content analysis may be sufficient for many investiga-
tions, especially those who have opted for a quantitative version of the method in the first
place.
Specifically, qualitative content analysis is generally preferable to discourse analysis
when the project is descriptive, aimed at reducing the data to study it systematically
rather than opening it up to analyze it holistically, seeks to arrive at a single interpretation
rather than providing multiple readings, focused on what is in the material rather than
silences, directed at what rather than how questions,52 and the objective is not to
uncover how language contributes to the construction of social reality.53 In these cases,
the combination of quantitative and qualitative content analysis is preferable to that of
quantitative content analysis and discourse analysis.
Pashakhanlou 7

There are thus clear advantages with combining quantitative and qualitative content
analysis in the study of international politics for research projects that would benefit
from their respective strengths. These rewards can be maximized if the examination is
conducted both manually and with the help of the computer as they also tend to comple-
ment one anothers strengths and weaknesses, as has been demonstrated. An inquiry that
utilizes all of these approaches in liaison with one another qualifies as a fully integrated
content analysis. In essence, this approach combines the insights of the first and second
wave content analyses with their emphasis on quantitative, manual, and computer-aided
content analyses with the qualitative dimension of the method that has been particularly
evident in Germany in a single study.

Fully integrated content analysis in IR

A fully integrated content analysis based on these premises has the potential to weaken
the criticism leveled against the method by detractors and proponents of the first and the
second wave content analyses. Their objections of the method are after all against partial
content analysis, as we have seen. As such, these issues can be ameliorated by a fully
integrated content analysis that makes use of the methods full capacity. In turn, this
means fully integrated content analysis is suitable for all research projects that fall within
the perimeters of this method but would suffer from the identified weaknesses by a par-
tial usage of the approach.
The full use of content analysis within a single research project can be advantageous
since the different aspects of the method can offset each others weaknesses and collec-
tively provide the foundation for robust and systematic studies of international politics.
Quantitative content analysis with its emphasis on manifest meaning can produce more
objective, reliable, and replicable results. Quantification of the data can also provide
precise accounts of the information that can be presented in an accessible way in tables
or graphs. For instance, it is far more precise to say that the US defense budget for 2015
was US$560 billion than to claim that it was huge. Nevertheless, there are numerous
significant issues in IR that cannot be adequately addressed through such measures
alone, as has been highlighted. In these cases, the addition of qualitative content analysis
can help to unpack latent meaning and enhance the validity, contextualization, and inter-
pretation of the communication.
Computer-assisted content analysis can enhance the quality of research even fur-
ther through its unrivaled reliability. Modern computers with fast Internet connection
can locate and process enormous quantities of communication far quicker and more
accurately than would have been possible without their aid. Unlike human beings,
they do not get fatigued and make mistakes as their attention starts to wander. With
the rapid development of information and computer technology, the benefits of com-
puter-aided content analysis will continue to grow. For the time being, there are,
however, still major limitations with the type of content analysis that computer soft-
ware can yield. They are extremely powerful in performing mechanical exercises
such as providing word frequencies but far less capable of providing demanding inter-
pretation and contextualization. This is why a fully computerized content analysis is
currently a chimera.54
8 International Relations 00(0)

For the foreseeable future, manual content analysis will therefore be essential, espe-
cially in the qualitative dimension of studies. Scholars are still far superior to computers
in performing these tasks. This is particularly evident when the communication is fraught
with interpretational difficulties in the form of irony, sarcasm, humor, and so on. It will
take time before Artificial Intelligence (AI) systems can match or surpass the interpreta-
tive skills of learned academics.55

The applicability of fully integrated content analysis in IR

The discussions above demonstrate the potential advantages of fully integrated content
analysis. What remains to be done is to outline which type of research projects and ques-
tions that can be addressed by the use of this method. The remainder of this section is
dedicated to this task. At this point, it should nonetheless already be evident that fully
integrated content analysis has the ability to address more complex and demanding
research questions than first or second wave content analysis, since it is more holistic and
have a broader range of tools at its disposal. As any other method, it does however have
certain limitations as well and is not well suited for all types of tasks.
First, as any other form of content analysis, it can only aid projects that seek to uncover
communication, but since it combines both quantitative and qualitative measures, it is
suitable for the analysis of both manifest and latent meaning. Second, since content anal-
ysis aims at dissecting communication, it is particularly strong at explaining what is said.
However, as fully integrated content analysis also includes qualitative measures for
interpretative inquiries, it allows for a greater exploration of how and why questions
as well. Third, just like other forms of content analysis, fully integrated content analysis
offers one interpretation of the communication rather than different readings of the mate-
rial. As such, it seeks to provide a clear answer to the what, how, or why question
that the research project seeks to address.
At this point, one might wonder how fully integrated content analysis can be applied
in practice. How can it aid IR scholars to shed light on the phenomena of their interest?
Due to space limitation, not every conceivable research project where this method might
be appropriated can be discussed. To highlight the practical usefulness of fully integrated
content analysis, let us, therefore, return to the study of Yarhi-Milo briefly mentioned
above. Her study is selected for two major reasons. First, her recent study on threat per-
ception in IR makes a significant contribution to a central issue in the study of interna-
tional politics. Second and more importantly, her inquiry represents a hard case for fully
integrated content analysis.
As we have seen, Yarhi-Milo herself explicitly contends that content analysis is not
suitable for her study. As she puts it:

an alternative way to analyze the documents would be to perform content analysis on archival
data and public statements. Yet such an analysis would be potentially misleading because I am
interested in capturing how observers discussed different indicators in association with
assessments of intentions, not the overall threat posed by the adversary. Such a contextualized
understanding of the interference process might be extremely hard to grasp using the content
analysis method.56
Pashakhanlou 9

Moreover, the overarching research question that Yarhi-Milo seeks to address is: how
do political leaders and intelligence organizations make inferences about adversary
intentions?57 In other words, Yarhi-Milo is concerned with a how question which is not
the primary strength of content analysis, as has been mentioned. Also, her investigation
is concerned with the assessment of intentions. Intent is a notoriously difficult phenom-
enon to study due to the other minds problem.58 If fully integrated content analysis can
be helpful in such a challenging case, it can be assumed that it would be useful in less
demanding cases as well.59 These are the reasons why Yarhi-Milos inquiry on threat
assessment is considered here.
In order to carry out her investigation, Yarhi-Milo conducts three case studies: British
assessment of Nazi Germanys intention prior to the Second World War (19341939) and
the Carter (19761980) and Reagan administrations (19851988) evaluations of Soviet
motives.60 These cases are used to test four competing theses: the capabilities thesis, the
strategic military doctrine thesis, the behavior thesis, and the selection attention thesis.
According to the capabilities thesis, assessments of adversaries are based on costly
changes in the quantity and type of the adversaries military capabilities compared to that
of the Self. The military doctrine thesis, regards the perceived orientation of the adver-
sarys military doctrine responsible for such estimations. The behavioral signal thesis
maintains that these assessments are a reflection of costly actions undertaken by the
adversary. Finally, the selective attentive thesis claims that decision makers will make
their assessments of adversarys intention on basis of vivid or salient information, usu-
ally pertaining to their military capabilities, that they subjectively deem credible.61 Yarhi-
Milo tests these propositions against 30,000 archival documents and intelligence reports
as well as interviews with former decision makers and intelligence officials. In her inves-
tigation, Yarhi-Milo finds that the selective attentive thesis has a considerably greater
explanatory power than rivaling theses.62
There are numerous ways in which fully integrated content analysis could have
enhanced Yarhi-Milos inquiry. First, her analysis is purportedly based on over 30,000
communications. Assumingly, not all the information in the approximately 30,000 com-
munications that Yarhi-Milo analyzes is directly relevant to her research question.
Content analysis would have been ideal in this case as it condenses the communication
to more relevant and smaller pieces of data. Specifically, the relevant concepts of adver-
sary, assessment, and intentions along with the four theses that focus on capabilities,
military doctrine, behavior, and information could be searched for in digitized versions
of these communications with computers. Doing so would speed up the inquiry consider-
ably, make it far less labor intense and provide a far greater reliability than reading all of
the reports in their entirety and assessing them manually.
Once the relevant passages have been identified, the inquiry can continue in two
major ways. Either the analyst can write computer codes that will tell the machine how
to categorize the data and determine which theses they support. If this proves too com-
plex and require unreasonable amount of coding, the data categorization can be done
entirely manually. Either way, manual qualitative content analysis of the material is
essential here for the validity of the research project. Automated content analysis is not
sufficiently sophisticated to complete this interpretatively demanding task successfully.
10 International Relations 00(0)

At this point, the reliability of the study can be enhanced by having other coders cat-
egorizing the data and establish whether there is an agreement between their respective
interpretations. The intersubjective intercoder reliability of content analysis can help to
overcome personal biases and prejudices.63 In studies where the author clearly favors a
particular position this is especially useful. For instance, Yarhi-Milo evidently champi-
ons her selective attentive thesis over the three existing competing propositions that she
evaluates.64 Had she gone through this process and presented the results in a transparent
manner as content analysis suggests, her findings would have been more reliable.
Once it has been established which observations that match the four tested theses,
quantitative content analysis can be appropriated to enhance the accuracy of the inquiry.
Indeed, quantitative content analysis can provide precise numerical information of the
quantity of observations that is consistent with any particular thesis. Subsequently, it can
be established exactly how the different theses rank not only against each other in gen-
eral but also in relation to each of the examined case studies. As a result, if one would,
for instance, find that the selective attentive thesis outperforms the rivaling hypotheses
across all of the three cases with a broad margin, our faith in this explanation would have
been far greater than if it only does marginally better than opposing accounts across
some or all of the cases.
Since Yarhi-Milo makes no use of these measures, she can however not provide such
precise, robust, and systemic evidence for the superiority of the selective attention thesis.
Instead, Yarhi-Milo can only argue that the support for the other examined theses is weak
compared to the selective attentive thesis without being able to back up this assertion
with clear, precise, numerical data.65 To paraphrase John Ikenberry, she is, therefore,
unable to provide a definitive answer to her research question.66
In sum, Yarhi-Milo is right to suggest that first and second wave content analyses are
insufficient to shed light on her demanding research question due to their limited scope.
Nevertheless, the fully integrated content analysis outlined in this article would not only
have been appropriate for her challenging research project but could have also enhanced
it in important ways. If fully integrated content analysis can be beneficial in such a tough
case, it can assumingly be useful for other important research projects in IR as well.

An example of fully integrated content analysis in IR

By now, the utility of fully integrated content analysis has been demonstrated, the
research projects and questions that this method can address have been specified, and it
has been illustrated that this approach can even enhance studies deemed inappropriate
for content analysis. The only thing that has not been done is to highlight how fully inte-
grated content analysis has been incorporated in practice. The remainder of this section
provides this last missing piece of the puzzle.
To my knowledge, so far, there is only one publication that combines quantitative, quali-
tative, computer-assisted, and manual content analyses within a single research project.
That is my monograph, Realism and Fear in International Relations: Morgenthau, Waltz
and Mearsheimer Reconsidered.67 As the title suggests, this publication seeks to uncover
the role of fear in the aforementioned scholars writings on international politics.68 It does
so at the conceptual, theoretical, empirical, and logical levels.
Pashakhanlou 11

Content analysis is suitable for this study for four reasons. First, the units of analysis
are written documents, mainly disseminated in the form of books and journal articles.
This makes them susceptible for methods such as content analysis that can unpack these
forms of communication. Second, the fact that content analysis only considers the data
that are contained in the text and does not extrapolate further than what is said makes it
suitable for the descriptive objectives of this inquiry. As a result, it enables Morgenthau,
Waltz, and Mearsheimer to speak for themselves rather than me reading various things
into their works.
Third, content analysis emphasis on key words, concepts, and themes makes it pos-
sible to focus the inquiry on fear in Morgenthaus, Waltzs, and Mearsheimers works
while ignoring the irrelevant bits. This time and resource saving procedure, enables an
examination of 400 of their publications, making this the most systematic and rigorous
study of these scholars works to date. Fourth, the fact that content analysis settles for a
single interpretation is necessary for this examination. Otherwise, it would be impossible
to pinpoint the conceptual, theoretical, empirical, and logical dimensions of fear in the
trajectories of Morgenthau, Waltz, and Mearsheimer.
These points explain why content analysis is appropriate for this research project but
how does fully integrated content analysis accentuate these efforts? To answer this ques-
tion, the various procedures that make up this examination must also be considered.
First, this investigation entails the detection of fear in the texts of Morgenthau, Waltz,
and Mearsheimer. Since the English language is rich and fear can be expressed in numer-
ous ways, it is not sufficient to merely locate the explicit use of this word and its deriva-
tives across these scholars publications.69 For this reason, the Merriam Webster thesaurus
is used70 along with the IR, psychology, and neuroscience literature to identify terms that
are synonymous and closely related to fear.71 To enhance the validity of this research
project, these words are tested against Morgenthaus, Waltzs, and Mearsheimers most
influential works, in order to determine which of these expressions that these scholars
themselves use synonymously or interchangeably with fear. These terms are subse-
quently also used to locate references to this emotion in all of their texts considered in
this evaluation.
Even in this initial phase of the investigation, almost all aspects of content analysis are
employed, except for its quantitative element. Computer is used to detect all references
to fear and closely related concepts in the digitized works of the examined realist schol-
ars. With the computer, this task can be achieved far quicker and more reliably than
would have been possible with the unaided human eye, especially as the examined works
span thousands of pages. Consequently, all relevant passages are read closely through
manual qualitative content analysis to determine which of these words Morgenthau,
Waltz, and Mearsheimer actually use interchangeably or synonymously with fear. By
doing so, the validity of the inquiry is enhanced.
Having determined which other words to look for in Morgenthaus, Waltzs, and
Mearsheimers writings besides fear and its derivatives, the actual objective of identify-
ing the conceptual, theoretical, empirical, and logical elements of fear in the aforemen-
tioned scholars publications is pursued. In practice, this objective is accomplished by
drawing on all aspects of content analysis. Again, computer software is used to detect all
instances of fear, in its various manifestations, across the 400 publications that are
12 International Relations 00(0)

considered. Without its aid, such a vast amount of literature could not have been ana-
lyzed by a single researcher within a reasonable time frame in a reliable manner.
Once fear has been detected in the texts, the relevant passages are read closely. This is
where manual qualitative content analysis comes to use. These measures provide interpre-
tation and contextualization of these texts and give meaning to the instances of fear located
by the computer. Without manual qualitative content analysis, the conceptual, theoretical,
empirical, and logical aspects of fear could not have been unraveled in an efficient and
accurate manner. As such, these tools were essential for carrying out this research project.
Quantitative content analysis complements these inquiries by providing an absolute
word frequency of the references to fear in the various strands of Morgenthaus, Waltzs,
and Mearsheimers works. This gives a clear numerical overview of the occurrences of
fear in relation to the topics discussed by these scholars. In a previous version of this
monograph, the total number of empirical cases that these scholars considered along with
the number of these cases that their respective theories succeed and fail to account for
was also identified. This procedure enabled the attainment of precise figures regarding
the explanatory power of the examined realist theories in relation to fear. Such a level of
accuracy could have only been attained through this quantification.
Altogether, this practical example illustrates how fully integrated content analysis can
inform the study of international politics and enhance the inquiry. Indeed, it demon-
strates that fully integrated content analysis can enable systemic and rigorous analysis
with reliable and valid results. What remains to be done is to illustrate that this approach
can even address why questions. The fact that content analysis can deal with what
questions is evident. The ability of fully integrated content analysis to address a how
question was also highlighted in the examination of Yarhi-Milos study. Let us, therefore,
consider this methods ability to tackle a why question. This will also be demonstrated
with reference to my study on realism and fear.
One of the issues that this monograph addresses is why Morgenthau, Waltz, and
Mearsheimer afford fear the role that they do in their theoretical frameworks. This ques-
tion is addressed through a combination of manual and computer-assisted qualitative
content analysis. This is evident if we consider the theoretical role of fear in Waltzs
defensive realism. He identifies fear as the central indicator of the common interests that
brings about alliance formation.72 To uncover why he might assign fear this specific role
in his neorealism, computer-assisted content analysis is first used to detect all instances
in which Waltzs discusses fear in relation to alliances. These passages are subsequently
dissected through manual qualitative content analysis.
After a close reading of the relevant passages, it is evident that Waltz approvingly cites
a passage from Thucydides History of the Peloponnesian War in his monograph, Man,
the State and War: A Theoretical Analysis. In this quotation, the Greek historian maintains
that it is folly to make alliances unless they are rooted in the interest of all members.
Indeed, mutual fear is the only solid basis of alliance.73 This finding thus suggests that
Waltzs appropriation of fear in relation to alliances is influenced by his reading of
Thucydides. This interpretation is subsequently strengthened by the fact that further man-
ual and computer-assisted qualitative content analyses of Waltzs works reveal that he
considers fear as the driver of alliance formation in numerous empirical cases in interna-
tional politics.74 Waltz thus evidently finds Thucydides position on this issue attractive
Pashakhanlou 13

and empirically sound. As such, this could explain why Waltz uses fear as the primary
indicator of common interest to account for alliance formation in his theory.
A fully integrated content analysis can thus evidently answer why questions as well.
Although this answer is not definitive, it is by no means shallow but rather well evi-
denced and robust as the example of Waltz above demonstrates. That said, the tasks set
for this article have been accomplished, and the time has come to summarize the main
points. This is the subject of the next section.

Conclusion
This article has provided an in-depth discussion regarding content analysis, a primarily
descriptive method that systematically reduces oral and/or visual communication to
arrive at an interpretation of the material. These inquiries can be pursued through quan-
titative and qualitative measures, manually and/or with computer support. Quantitative
and computer-aided content analysis are particularly suitable for exploring manifest
meaning and can provide more objective, replicable and precise findings. Qualitative and
manual content analyses strengths lie in untangling latent meaning, and their ability to
deal with interpretation, contextualization, and nuances in the communication. In short,
quantitative and computer-assisted content analyses have an edge in reliability, whereas
qualitative and manual content analyses are primed for validity.
During the first wave of content analysis in IR, from the 1940s to the 1960s, the method
was understood and employed in exclusively or principally quantitative and manual
terms. The second wave that arose in the twenty-first century has also been quantitative
but emphasized computer-aided content analysis instead. Both critics and proponents of
the method have therefore come to criticize content analysis for its supposed deficiencies
in interpretation, contextualization, and validation. As has been demonstrated, these
objections are rooted in a partial understanding of the method and ignored its qualitative
component.
A fully integrated content analysis utilizes all aspects of the method, that is, quantita-
tive, qualitative, manual, and computer-aided content analyses within a single research
project can ameliorate the identified shortcomings. The widely overlooked qualitative
element of the method can, for instance, enhance the interpretation, contextualization,
and validity of scholarly investigations as sought by second wave scholars. As such, fully
integrated content analysis can be used in research projects that are too complex for par-
tial content analysis and seek to address a what, how, or why question and arrive at
one interpretation of the assessed communication.75 Indeed, as the study of Yarhi-Milo
and my own monograph demonstrate, this approach can even provide robust answers to
how and why questions where partial content analysis is generally considered weak or
incompetent. In suitable studies, fully integrated content analysis has the potential to
maximize the benefits of the method and enable systematic and rigorous studies of inter-
national politics with high levels of reliability and validity.

Acknowledgements
The author would like to thank Jon Dean and the anonymous reviewers for their helpful
comments.
14 International Relations 00(0)

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship,
and/or publication of this article.

Funding
The author(s) received no financial support for the research, authorship, and/or publication of this
article.

Notes
1. Karin Dovring, Quantitative Semantics in 18th Century Sweden, The Public Opinion
Quarterly, 18(4), 1954, pp. 38994; Klaus Krippendorff, Content Analysis: An Introduction
to Its Methodology (London: SAGE, 2004); Udny Yule, The Statistical Study of Literary
Vocabulary (Cambridge: Cambridge University Press, 2014).
2. John Gilmer Speed, Do Newspapers Now Give the News? Forum, 15, 1893, pp. 70511.
3. Douglas Waples, Bernard Berelson and Franklyn Bradshaw, What Reading Does to People:
A Summary of Evidence on the Social Effects of Reading and a Statement of Problems for
Research (Chicago, IL: University of Chicago Press, 1940).
4. Douglas Waples, Print, Radio, and Film in a Democracy: Ten Papers on the Administration
of Mass Communications in the Public Interest (Chicago, IL: University of Chicago Press,
1942).
5. Bernard Berelson, Content Analysis in Communication Research (New York: Free Press,
1952); Bernard Berelson and Paul Felix Lazarsfeld, The Analysis of Communication Content
(Chicago, IL: University of Chicago Press, 1948).
6. Jonathan Gottschall and David Sloan Wilson, The Literary Animal: Evolution and the
Nature of Narrative (Evanston, IL: Northwestern University Press, 2005); Irena Makaryk,
Encyclopedia of Contemporary Literary Theory: Approaches, Scholars, Terms (Toronto, ON,
Canada: University of Toronto Press, 1993); Mark West (ed.), Theory, Method, and Practice
in Computer Content Analysis (Westport, CT: Greenwood Publishing Group, 2001).
7. Margrit Schreier, Qualitative Content Analysis in Practice (London: SAGE, 2012), p. 13.
8. Klaus Krippendorff, Content Analysis: An Introduction to Its Methodology (London: SAGE,
2013), p. 17.
9. Gary King, Justin Grimmer, and Phil Schrodt have, along with their collaborators, made
important contributions to content analysis in political science. None of them have, however,
focused on the subfield of IR. Moreover, they unanimously advocate computer-assisted con-
tent analysis, and their works are, therefore, equivalent to second wave content analysts in IR.
In contrast, this investigation champions fully integrated content analysis and is the first publi-
cation to do so. In sum, there is little overlap between this study and that of the aforementioned
scholars or that of any other contributor to content analysis for that matter. Daniel Hopkins and
Gary King, A Method of Automated Nonparametric Content Analysis for Social Science,
American Journal of Political Science, 54(1), 2010, pp. 22947; Justin Grimmer and Brandon
Stewart, Text as Data: The Promise and Pitfalls of Automatic Content Analysis Methods for
Political Texts, Political Analysis, 21(3), 2013, pp. 26797; Philip Schrodt, Event Data in
Foreign Policy Analysis, in Laura Neack, Patrick Haney and Jeanne Hey (eds) Foreign Policy
Analysis: Continuity and Change (New York: Prentice Hall, 1994), pp. 14566, available
at: http://www.parusanalytics.com/eventdata/papers.dir/Haney.pdf; Schrodt, Event Data in
Foreign.
10. Manifest meaning is obvious and literal, whereas latent meaning is present but hidden and not
readily apparent.
Pashakhanlou 15

11. Stephen Croucher and Daniel Cronn-Mills, Understanding Communication Research

Methods: A Theoretical and Practical Approach (New York: Routledge, 2015), p. 206.
12. Duane Monette, Thomas Sullivan and Cornell DeJong, Applied Social Research: A Tool for
the Human Services (Belmont, CA: Cengage Learning, 2014), pp. 20708; Philipp Mayring,
Qualitative Content Analysis. Theoretical Foundation, Basic Procedures and Software
Solution (Klagenfurt: Institute of Psychology and Center for Evaluation and Research, 2014),
pp. 10710; There is also a tendency to distinguish between different types and aspects of
validity and reliability in the literature. Their applicability depends on the research project
at hand. See, for example, Klaus Krippendorff, Content Analysis: An Introduction to Its
Methodology (Beverly Hills, CA: SAGE, 1980), p. 158.
13. Ole Holsti, Content Analysis for the Social Sciences and Humanities (Reading, MA: Addison-
Wesley, 1969), p. 59.
14. Lisa Webley, Qualitative Approaches to Empirical Legal Research, in Peter Cane and
Herbert Kritzer (eds) The Oxford Handbook of Empirical Legal Research (Oxford: Oxford
University Press, 2010), p. 941.
15. Schreier, Qualitative Content, pp. 30, 48; Gilbert Shapiro and John Markoff, A Matter of
Definition, in Carl Roberts (ed.) Text Analysis for the Social Sciences: Methods for Drawing
Statistical Inferences from Texts and Transcripts (Mahwah, NJ: Lawrence Erlbaum, 1997),
p. 14; Robert Philip Weber, Basic Content Analysis (London: SAGE, 1990), p. 5.
16. Karl Erik Rosengren, Advances in Content Analysis (London: SAGE, 1981), p. 28; Donald
Treadwell, Introducing Communication Research: Paths of Inquiry (London: SAGE, 2014),
p. 225.
17. Holsti, Content Analysis; Krippendorff, Content Analysis, 2004; Schreier, Qualitative
Content, pp. 145.
18. Siegfried Kracauer, The Challenge of Qualitative Content Analysis, The Public Opinion
Quarterly, 16(4), 1952, pp. 63142.
19. For instance, one must be aware of the fact that MAD is a satirical magazine to make valid
inferences regarding its publications on the War on Terror.
20. Norbert Groeben and Ruth Rustemeyer, On the Integration of Quantitative and Qualitative
Methodological Paradigms (Based on the Example of Content Analysis), in Ingwer Borg and
Peter Mohler (eds) Trends and Perspectives in Empirical Social Research (Berlin: Walter de
Gruyter, 1994), pp. 3156.
21. Kimberly Neuendorf, The Content Analysis Guidebook (Thousand Oaks, CA: SAGE,
2002).
22. Rebecca Morris, Computerized Content Analysis in Management Research: A Demonstration
of Advantages & Limitations, Journal of Management, 20(4), 1994, pp. 90331.
23. Bu Zhong and Yong Zhou, Under the Weather: The Weather Effects on U.S. Newspaper
Coverage of the 2008 Beijing Olympics, in Kim Bissell and Stephen Perry (eds) The
Olympics, Media and Society (New York: Routledge, 2013), p. 84.
24. Seth Lewis, Rodrigo Zamith and Alfred Hermida, Content Analysis in an Era of Big Data:
A Hybrid Approach to Computational and Manual Methods, Journal of Broadcasting &
Electronic Media, 57(1), 2013, pp. 3452.
25. That is, words with different meanings that are spelled the same. For instance, mine can
refer to a hole in the ground, the process of extraction or a possessive pronoun.
26. Assuming that the computer and the software are fully functional.
27. Donald Diefenbach, Historical Foundations of Computer-Assisted Content Analysis, in
Mark West (ed.) Theory, Method, and Practice in Computer Content Analysis (Westport, CT:
Greenwood Publishing Group, 2001), pp. 145; David Stewart and Prem Shamdasani, Focus
Groups: Theory and Practice (London: SAGE, 2015), p. 133.
16 International Relations 00(0)

28. Robert North, Ole Holsti, George Zaninovich and Dina Zinnes. Content Analysis: A
Handbook with Applications for the Study of International Crisis (Evanston, IL: Northwestern
University Press, 1963); Ole Holsti, Computer Content Analysis as a Tool in International
Relations Research, in Edmund Bowles (ed.) Computers in Humanistic Research: Readings
and Perspectives (Englewood Cliffs, NJ: Prentice Hall, 1967), pp. 10817; Harold Lasswell
and Nathan Leites (eds), Language of Politics: Studies in Quantitative Semantics (New York:
George W. Stewart, 1949); Harold Lasswell and Abraham Kaplan, Power and Society: A
Framework for Political Inquiry (New Haven, CT: Yale University Press, 1950).
29. Defined as demand statements and evaluations.
30. Defined as statements of facts and empirical hypotheses.
31. Lasswell and Kaplan, Power and Society, pp. 11815.
32. Hans Morgenthau, Review of Book: Power and Society, by Harold D. Lasswell and Abraham
Kaplan, The American Political Science Review, 46(1), 1952, pp. 2334.
33. Robert Jervis, The Costs of the Scientific Study of Politics: An Examination of the Stanford
Content Analysis Studies, International Studies Quarterly, 11(4), 1967, pp. 368, 3768,
3845, 390.
34. Hedley Bull, International Theory: The Case for a Classical Approach, World Politics,
18(3), 1966, pp. 366, 374.
35. Alexander George, Quantitative and Qualitative Approaches to Content Analysis, in Pool
Ithiel de Sola (ed.) Trends in Content Analysis (Urbana, IL: University of Illinois Press, 1959),
pp. 732.
36. Morgenthau, Review of Book, p. 234.
37. Samuel Fitch, The Garrison State in America: A Content Analysis of Trends in the Expectation
of Violence, Journal of Peace Research, 22(1), 1985, pp. 3145; Robert Friedheim and Joseph
Kadane, Quantitative Content Analysis of the United Nations Seabed Debate: Methodology
and a Continental Shelf Case Study, International Organization, 24(3), 1970, pp. 479502;
Deborah Welch Larson, Problems of Content Analysis in Foreign-Policy Research: Notes
from the Study of the Origins of Cold War Belief Systems, International Studies Quarterly,
32(2), 1988, pp. 24155; Jason Satterfield, Cognitive-Affective States Predict Military and
Political Aggression and Risk Taking A Content Analysis of Churchill, Hitler, Roosevelt, and
Stalin, Journal of Conflict Resolution, 42(6), 1998, pp. 66790; Michael Young and Mark
Schafer, Is There Method in Our Madness? Ways of Assessing Cognition in International
Relations, Mershon International Studies Review, 42(1), 1998, pp. 6396.
38. Andrew Bennett, Found in Translation: Combining Discourse Analysis with Computer
Assisted Content Analysis, Millennium: Journal of International Studies, 43(3), 2015,
pp. 98497; Johan Eriksson and Giampiero Giacomello, Content Analysis in the Digital
Age: Tools, Functions, and Implications for Security, in Jrg Krger, Bertram Nickolay
and Sandro Gaycken (eds) The Secure Information Society: Ethical, Legal and Political
Challenges (London: Springer, 2013), pp. 13748; Johan Eriksson and Giampiero Giacomello,
International Relations, Cybersecurity, and Content Analysis: A Constructivist Approach, in
Maximilian Mayer, Mariana Carpes and Ruth Knoblich (eds) The Global Politics of Science
and Technology Vol. 2: Perspectives, Cases and Methods (New York: Springer, 2014), pp.
20520; Margaret Hermann, Content Analysis, in Audie Klotz and Deepa Prakash (eds)
Qualitative Methods in International Relations: A Pluralist Guide (New York: Palgrave
Macmillan, 2008), pp. 15167.
39. Bennett, Found in Translation, pp. 987, 9912, 997; Eriksson and Giacomello, Content
Analysis in the Digital Age, pp. 1378, 1447; Eriksson and Giacomello, International
Relations, pp. 2057, 210, 215; Hermann, Content Analysis, p. 151.
40. Holsti, Computer Content Analysis.
Pashakhanlou 17

41. Bennett, Found in Translation, p. 992.

42. Bennett, Found in Translation, pp. 984, 9867, 991, 997; Natalie Delia Deckard, Atta Barkindo
and David Jacobson, Religiosity and Rebellion in Nigeria: Considering Boko Haram in the
Radical Tradition, Studies in Conflict & Terrorism, 38(7), 2015, pp. 510, 515; Eriksson and
Giacomello, Content Analysis in the Digital Age, pp. 138, 145; Eriksson and Giacomello,
International Relations, pp. 2056, 211, 215; Herrera and Braumoeller, Symposium, pp.
178; Will Lowe, Content Analysis and Its Place in the (Methodological) Scheme of Things,
Qualitative Methods: Newsletter of the American Political Science Association Organized
Section on Qualitative Methods, 2(1), 2004, p. 25; Roseanne McManus, Fighting Words: The
Effectiveness of Statements of Resolve in International Conflict, Journal of Peace Research,
51(6), 2014, pp. 727, 72930; Hctor Perla, Explaining Public Support for the Use of
Military Force: The Impact of Reference Point Framing and Prospective Decision Making,
International Organization, 65(1), 2011, p. 139; Rebekah Tromble and Miriam Wouters, Are
We Talking with or past One Another? Examining Transnational Political Discourse across
WesternMuslim Divides, International Studies Quarterly, 59(2), 2015, p. 373; Jinghan
Zeng, Yuefan Xiao and Shaun Breslin, Securing Chinas Core Interests: The State of the
Debate in China, International Affairs, 91(2), 2015, p. 248.
43. Bennett, Found in Translation, pp. 984, 9867, 9901; Eriksson and Giacomello, Content
Analysis in the Digital Age, p. 147; Eriksson and Giacomello, International Relations,
p. 217; Rose McDermott, Political Psychology in International Relations (Ann Arbor,
MI: University of Michigan Press, 2004), p. 34; Oliver Jtersonke and Philipp Stucki,
Consolidating Content and Discourse Analysis: Pragmatism and Context-Sensitive Content
Analysis in the Field of International Relations (International Studies Association 48th
Annual Convention, Chicago, IL, February 28, 2007), pp. 12, 4.
44. Bennett, Found in Translation; Eriksson and Giacomello, Content Analysis in the Digital
Age, p. 138; Jtersonke and Stucki, Consolidating Content and Discourse Analysis; Lowe,
Content Analysis, p. 27.
45. See for example, Arthur Asa Berger, Content Analysis, in Arthur Asa Berger (ed.) Media
and Communication Research Methods: An Introduction to Qualitative and Quantitative
Approaches (Thousand Oaks, CA: SAGE, 2010), pp. 17385; Russell Bernard and Gery
Ryan, Content Analysis, in Russell Bernard and Gery Ryan (eds) Analyzing Qualitative
Data: Systematic Approaches (Thousand Oaks, CA: SAGE, 2010), pp. 287310; Graham
Gibbs, Analysing Qualitative Data (London: SAGE, 2007).
46. Schreier, Qualitative Content, p. vii.
47. David Altheide, Qualitative Media Analysis (London: SAGE, 1996); Richard Boyatzis,
Transforming Qualitative Information: Thematic Analysis and Code Development (Thousand
Oaks, CA: SAGE, 1998); Johnny Saldana, The Coding Manual for Qualitative Researchers
(London: SAGE, 2009).
48. Satu Elo and Helvi Kyngs, The Qualitative Content Analysis Process, Journal of Advanced
Nursing, 62(1), 2008, pp. 10715; Hsiu-Fang Hsieh and Sarah Shannon, Three Approaches
to Qualitative Content Analysis, Qualitative Health Research, 15(9), 2005, pp. 127788.
49. Philipp Mayring, Qualitative Inhaltsanalyse: Grundlagen und Techniken (Weinheim: Beltz,
2008); Holger Rust, Struktur und Bedeutung (Berlin: Spiess, 1980); Ruth Rustemeyer,
Praktisch-Methodische Schritte der Inhaltsanalyse: Eine Einfhrung am Beispiel der Analyse
von Interviewtexten (Mnster: Aschendorff, 1992).
50. Jeffrey Checkel, The Constructive Turn in International Relations Theory, World
Politics, 50(2), 1998, p. 334; Friedheim and Kadane, Quantitative Content Analysis, pp.
4823; Hermann, Content Analysis, p. 156; Christopher Lamont, Research Methods in
International Relations (Los Angeles, CA: SAGE, 2015); Carmen Wunderlich, A Rogue
18 International Relations 00(0)

Gone Norm Entrepreneurial? Iran within the Nuclear Nonproliferation Regime, in Wolfgang
Wagner, Wouter Werner and Michal Onderco (eds) Deviance in International Relations:
Rogue States and International Security (London: Palgrave Macmillan, 2014), pp. 1007;
Antje Wiener, Enacting Meaning-in-Use: Qualitative Research on Norms and International
Relations, Review of International Studies, 35(1), 2009, p. 186.
51. Keren Yarhi-Milo, Knowing the Adversary: Leaders, Intelligence, and Assessment of
Intentions in International Relations (Princeton, NJ: Princeton University Press, 2014), pp.
275 n. 104.
52. As has been mentioned, qualitative content analysis can deal with how questions as
well. But since it is more descriptive and explicit and less holistic than discourse analy-
sis, it tends to be weaker than discourse analysis in addressing these types of research
questions.
53. Norman Fairclough, Analysing Discourse: Textual Analysis for Social Research (London:
Routledge, 2003); Jennifer Milliken, The Study of Discourse in International Relations:
A Critique of Research and Methods, European Journal of International Relations, 5(2),
1999, pp. 22554; Schreier, Qualitative Content, pp. 458; Stefan Titscher, Michael
Meyer, Ruth Wodak and Eva Vetter, Methods of Text and Discourse Analysis: In Search
of Meaning, trans. by Bryan Jenner (Thousand Oaks, CA: SAGE, 2000); Ruth Wodak,
Methods of Discourse Analysis (ed. Ruth Wodak and Michael Meyer) (London: SAGE,
2001), pp. 6394.
54. Grimmer and Stewart, Text as Data, pp. 26768.
55. Jim Macnamara, Media Content Analysis: Its Uses, Benefits and Best Practice Methodology,
Asia-Pacific Public Relations Journal, 6(1), 2005, p. 7.
56. Yarhi-Milo, Knowing the Adversary, 275 n. 104.
57. Yarhi-Milo, Knowing the Adversary, pp. 113.
58. The other minds problem refers to the fact that we cannot read other actors minds and can,
therefore, never be certain about their current or future intentions. Ken Booth and Nicholas
Wheeler, The Security Dilemma: Fear, Cooperation and Trust in World Politics (New York:
Palgrave Macmillan, 2008); Robert Jervis, Was the Cold War a Security Dilemma? Journal
of Cold War Studies, 3(1), 2001, pp. 3660.
59. Alexander George and Andrew Bennett, Case Studies and Theory Development in the Social
Sciences (Cambridge, MA: The MIT Press, 2005).
60. Yarhi-Milo, Knowing the Adversary, pp. 6, 35.
61. Yarhi-Milo, Knowing the Adversary, pp. 3637.
62. Keren Yarhi-Milo, In the Eye of the Beholder: How Leaders and Intelligence Communities
Assess the Intentions of Adversaries, International Security, 38(1), 2013, p. 9.
63. Krippendorff, Content Analysis, 2004; Neuendorf, The Content Analysis Guidebook.
64. Yarhi-Milo, Knowing the Adversary, Chapter 1, 11.
65. Yarhi-Milo, Knowing the Adversary, p. 12.
66. John Ikenberry, Knowing the Adversary: Leaders, Intelligence, and Assessment of Intentions
in International Relations, Foreign Affairs, 94(2), 2015, p. 181.
67. For other publications on realism, see Arash Heydarian Pashakhanlou, Comparing and
Contrasting Classical Realism and Neorealism: A Re-Examination of Hans Morgenthaus and
Kenneth Waltzs Theories of International Relations, E-IR, 2009, available at: http://www.e-
ir.info/2009/07/23/comparing-and-contrasting-classical-realism-and-neo-realism/; Arash
Heydarian Pashakhanlou, Back to the Drawing Board: A Critique of Offensive Realism,
International Relations, 27(2), 2013, pp. 20225; Arash Heydarian Pashakhanlou, Waltz,
Mearsheimer and the Post-Cold War World: The Rise of America and the Fall of Structural
Realism, International Politics, 51(3), 2014, pp. 295315.
Pashakhanlou 19

68. A justification for the focus on realism and the selected theorists is provided in the book.
Arash Heydarian Pashakhanlou, Realism and Fear in International Relations: Morgenthau,
Waltz and Mearsheimer Reconsidered (London: Palgrave Macmillan, 2016).
69. Krippendorff, Content Analysis, 2013, p. 106.
70. The use of this American-English thesaurus is particularly useful since all of the examined
theorists were mainly based in the United States at the time of their scholarly activities in IR
and wrote their main works in the discipline in American-English.
71. Jos Brosschot, William Gerin and Julian Thayer, The Perseverative Cognition Hypothesis:
A Review of Worry, Prolonged Stress-Related Physiological Activation, and Health, Journal
of Psychosomatic Research, 60(2), 2006, p. 114; Violet Cheung-Blunden and Bill Blunden,
The Emotional Construal of War: Anger, Fear, and Other Negative Emotions, Peace and
Conflict: Journal of Peace Psychology, 14(2), 2008, p. 127; Neta Crawford, The Passion
of World Politics: Propositions on Emotion and Emotional Relationships, International
Security, 24(4), 2000, pp. 116, 122; Steven Fein and James Hilton, Judging Others in the
Shadow of Suspicion, Motivation and Emotion, 18(2), 1994, pp. 1689; Agneta Fischer and
Antony Manstead, Social Functions and Emotion, in Michael Lewis, Jeannette Haviland-
Jones and Lisa Feldman Barrett (eds) Handbook of Emotions (New York: Guilford Press,
2010), pp. 45670; Jonathan Jackson, Introducing Fear of Crime to Risk Research, Risk
Analysis, 26(1), 2006, pp. 25364; Merriam-Webster, Fear, 2013, available at: http://
www.merriam-webster.com/thesaurus/fear; Brian Rathbun, Uncertain about Uncertainty:
Understanding the Multiple Meanings of a Crucial Concept in International Relations
Theory, International Studies Quarterly, 51(3), 2007, p. 533; Catherine Ross and Sung
Joon Jang, Neighborhood Disorder, Fear, and Mistrust: The Buffering Role of Social Ties
with Neighbors, American Journal of Community Psychology, 28(4), 2000, p. 406; Shiping
Tang, The Social Evolutionary Psychology of Fear (and Trust): Or Why Is International
Cooperation Difficult? 2010, 1, 1 n. 1, 66; Arne hman, Fear and Anxiety: Overlaps and
Dissociations, in Michael Lewis, Jeannette Haviland-Jones and Lisa Barrett (eds) Handbook
of Emotions (New York: Guilford Press, 2008), p. 709.
72. Pashakhanlou, Realism and Fear in International Relations, Chapter Four.
73. Kenneth Waltz, Man the State and War: A Theoretical Analysis (New York: Columbia
University Press, 2001), p. 211.
74. Arash Heydarian Pashakhanlou, Realism and Fear in International Relations: Morgenthau,
Waltz and Mearsheimer Reconsidered (London: Palgrave Macmillan, 2016), Chapter Six.
75. As has been maintained, this approach is particularly strong in dealing with what questions.

Author biography
Arash Heydarian Pashakhanlou is an Assistant Professor in War Studies at the Swedish Defence
University. His works have appeared in the journals of International Relations and International
Politics, among others. Palgrave published his most recent monograph.