International Journal of E-Politics

Volume 10 • Issue 1 • January-June 2019

The Popularization of Political

Communication:
A New Definition, Its Drivers on Facebook, Its
Properties Under the Current Hybrid Media System
Diego Ceccobelli, Scuola Normale Superiore, Pisa, Italy

ABSTRACT

This article presents and adopts a new definition of the popularization of political communication,
which is defined as a strategic communicative action through which political actors try to create new
connections with those citizens who do not still know, follow and support them and to emotionally
strengthen the political bond with their current sympathizers. Second, a comparative analysis of the
Facebook pages of the main political leaders of 31 countries shows that the popularization of political
communication is a relevant phenomenon on Facebook, while a qualitative comparative analysis (QCA)
indicates that the presence of a presidential system, a high digitalization of the media system, and a
high level of trust in political institutions are three sufficient conditions for a “pop” communication
on Facebook. Finally, the article identifies and discusses its main properties and development under
the current hybrid media system.

Keywords
Facebook, Political Leaders, Popularization Of Political Communication, QCA, Social Media

INTRODUCTION

Scrolling on the Facebook pages of political actors exposes audiences to a wide array of communicative
stances. Politicians communicate what and how they want on their own digital profiles, including
textual content about a specific policy issue, harsh attacks on their opponents in videos, the use of a
populist style of communication with stylish info-graphs, and so forth. Sometimes, they also publish
pictures while campaigning with a film star, live videos buying fresh food in a supermarket or a
birthday wish to a parent in a concise but passionate text status. Respectively, the literature denotes
these three latter communicative choices as empirical examples of celebrity politics, lifestyle politics
and intimate politics. While analyses on how and why political leaders employ policy issues (Cotta
and Ceccobelli, 2017; Larsson, 2015), negative campaigning (Gainous & Wagner, 2014; Auter & Fine,
2016) and populist communication (Engesser et al., 2017; De Vreese et al., 2018) on social media
abound in the literature, less attention has been devoted to the latter phenomena, particularly from a
comparative perspective. The aim of this article is threefold: first, it provides a causal, ontological,
and realistic reasoning for analysing celebrity, lifestyle and intimate content under the popularization
of political communication’s theoretical lenses; second, through a qualitative comparative analysis
(QCA), it reveals the sufficiency conditions for the main political leaders of 31 countries’ strategic
choice of popularizing their political communication on Facebook; finally, it identifies and discusses
the main properties and implications of this strategic communicative action under the current hybrid
media system.

DOI: 10.4018/IJEP.2019010102

Copyright © 2019, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.


12
 International Journal of E-Politics
Volume 10 • Issue 1 • January-June 2019

THE POPULARIZATION OF POLITICAL COMMUNICATION

In a time characterized by the so-called “populist zeitgeist” (Mudde, 2004), the most recent literature
on leaders’ communication style has devoted less attention to other concepts that have animated
the scientific debate on this topic for many years. Before the populist zeitgeist hypothesis reached
enormous popularity within academic research, concepts such as the personalization of politics
(Karvonen, 2010; Rahat & Sheafer, 2007), celebrity politics (Street, 2004; West & Orman, 2003;
Wheeler, 2013), lifestyle politics (Bennett, 1998; Mancini, 2011) and intimate politics (Langer, 2010;
Stanyer, 2013) were among the main concepts adopted by media scholars to describe and explain the
changes occurring in the political communication of contemporary political actors. All these concepts
have been and are still employed as theoretical lenses denoting the different strategies used by political
actors to communicate proficiently with voters due to a continuously transforming media and a political
environment characterized by individualization (Inglehart, 1997), party-de-alignment (Dalton et al.,
2000), commercialization (Hallin & Mancini, 2004) media abundance and audience fragmentation
(Iyengar & Hahn, 2009; Jomini Stroud, 2011; Mancini, 2013; Prior, 2007). The transformations
denoted by these concepts are straightforward: political leaders currently sell more than parties (the
personalization of politics); they must appear like ordinary citizens and in touch with the daily lives
of their voters (lifestyle politics); they have to be perceived as extraordinary individuals, deserving
of the office they aspire to hold (celebrity politics); and finally, political leaders must make their
private lives public because voters need to trust them, and disclosing aspects of their private lives
make them appear authentic (intimate politics).
These different concepts trace different phenomena, but they all share a crucial element: the
causal reasoning determining why political leaders need to communicate their proximity with voters,
their extraordinariness and their authenticity. There are two main drivers pushing political leaders
towards these strategic moves: the need to keep their current supporters on their electoral side and
to enlarge the pool of their sympathizers to transform them into future loyal voters. This causal red
line connecting these three concepts can be identified by the notion of popularization (Street, 2015),
and in this article, it is empirically used according to the following new definition:

The popularization of political communication is a strategic communicative action through which

political actors try to create new connections with those citizens who still do not know, follow or
support them – particularly with the ones who are less interested in the news, in the actors and in
the dynamics related to the sphere of politics – and to emotionally strengthen the bond with their
current sympathizers.

So defined, the popularization of political communication does not need to be considered a

new phenomenon (during previous centuries, political actors needed to create new connections with
those citizens who still do not know, follow or support them and to emotionally strengthen the bond
with their current sympathizers). Politicians always needed and tried to create connections with new
potential voters and keep a strong bond with their more loyal ones. However, the current hybrid
media system, which is so deeply affected by digital media and their own logics (Chadwick, 2013),
might make popularization both more required and employed by political actors. Citizens consume
political news in a high-choice media environment (Prior, 2007) that is ever more dominated by
private digital companies (Kreiss & McGregor, 2017) that, through their secret algorithms, seem
to push them within echo chambers (Hannak et al., 2013; Pariser, 2011)1. Here, through selective
exposure media practices, they are exposed to the information they like the most, being a specific
political side, or no political sides at all, if they use these new digital platforms as entertaining media
outlets only (Iyengar, 1987; Wheeler, 2009). It is within this current media environment that political
actors must manage how to make up different communicative strategies, such as what the literature
has discussed using the ideas of intimate, celebrity and lifestyle politics, considered here as three

13
International Journal of E-Politics
Volume 10 • Issue 1 • January-June 2019

ways of popularizing communication and, thus, the main strategies through which they aim to keep
the pool of their supporters faithful and gain new voters.
The definition of popularization adopted here hence entails “a causal, ontological, and realistic
view” (Goertz, 2005, p. 5) of this concept. It considers popularization as a response, as an adaptation
to the abovementioned changes that have occurred in the relationship between the political sphere,
the media sphere and citizens. Popularization refers to the way in which political actors attempt to
adapt themselves to the structural dynamics of the media environment that surrounds them to avoid
being (potentially) eliminated or sub-represented in the media diets of heterogeneous electorates.
Although the concept of popularization is not widespread or used by the political communication
literature – with a small exception being several Italian publications on this topic (see, for example,
Mazzoleni and Sfardini, 2009) – its heuristic value is deduced by examining its etymological root. The
term “popularization” is a derivative of the verb “to popularize”, which has two different meanings:
i) to make a scientific or academic subject accessible to the general public by presenting it in an
understandable form; ii) to cause something to become generally liked.2 These two meanings of the
verb “to popularize” perfectly denote what is assumed and expressed by the proposed definition.
The popularization of political communication is first a strategy for trying to divulge (indeed, to
popularize), “to make accessible to the general public”, the political actor who uses this strategic
communicative action. This aim is the first goal of popularization: the strategy that political actors
employ to create new connections with those citizens who still do not know, follow or support them.
However, this is only the first goal and, for the purposes of the political actors, reaching this goal –
despite being a necessary step to reach the second one – is not enough.
The second goal is the most crucial one, but it is also the most difficult to observe, measure
and evaluate. When the second goal is achieved, popularization “causes (something) to be liked”. If
used properly, popularization also has a positive effect on the image of the political actor who uses
it. Thanks to popularization, the political actor is well known and continuously covered in the media
outlets consumed daily by a significant majority of the citizens, whether a social media platform, a
TV show or a gossip magazine. When political actors can trigger the second goal of popularization,
this strategic communicative action becomes the leverage through which they keep and gain political.
Although popularization can thus be achieved through older media as well, its pervasiveness and
implementation is exponentially growing and might have changed simply due to the abovementioned
recent developments occurring within the media sphere. For this reason, this article focuses on
Facebook, with the aim of identifying whether the most widespread social media platform worldwide
triggers popularization and the factors explaining its use by the main political leaders of Western and
Latin American countries.

The Pop Attribute

Together with the definition of popularization, which aims to answering the what is question, it is
fundamental to identify the predicate associated with this concept in order to apply it properly in
empirical research (Sartori, 1970). As the predicate “democratic” implies a quantitative approach
oriented towards the measurement of the concept of democracy, the predicate “pop” does the same
in the case of popularization. Investigating the degree of “pop” of the political communication of
a political actor involves placing his/her political communication on a continuum between two
poles whereby the second pole stands for the opposite of “pop”. In the case of popularization and
its pop predicate, the opposite of “pop”, that is, “not-pop”, is institutional. More precisely, political
communication is supposed to be institutional when a political actor centres its strategic communicative
actions exclusively around politics, polity and policy issues, without hybridizing them with any of
the components of popularization. From a theoretical and analytical point of view, using the notion
of popularization to establish how pop the political communication of political actors is entails an
investigation of the extent to which, as stated by the new definition proposed, these actors are trying
to create new connections with those citizens who still do not know, follow or support them and to
emotionally strengthen the bond with their current sympathizers.
14
 International Journal of E-Politics
Volume 10 • Issue 1 • January-June 2019

RESEARCH QUESTIONS

The advent of social media gave political actors a new arena for spreading their political thoughts.
Unlike “older media”, social media provides political actors total control of their communication and
allows them to build a direct and unmediated relationship with citizens (Chadwick, 2006). Moreover,
although hybrid dynamics between older and newer media exist, entrenching different media logics
(Chadwick, 2013), it seems increasingly true that communicating in a proficient and engaging way
on social media represents a necessary condition for achieving political goals. Nevertheless, because
empirical analyses are missing, particularly from a comparative perspective, it remains unclear
whether the spread of social media such as Twitter, Facebook and Instagram is causing a process of
homogenization or, vice versa, a process of differentiation in the political communication of political
actors (Filimonov et al., 2016; Kalsnes, 2016; Samuel-Azran et al., 2015).
That being said, there is no doubt that popularization is broadly exploitable on a Facebook page.
Each political leader can freely decide to publish a photo with the members of her/his family, with a
TV star, or of him/herself buying food in a supermarket, that is, three different forms of popularization.
The first research question investigated here is therefore the following:

RQ1 – Does popularization represent a relevant phenomenon on the Facebook pages of the main
political leaders of contemporary liberal democracies?

A first aim of this paper is therefore to use a comparative perspective to examine whether
popularization represents a strategic communicative action that concerns not only older media such
as television, radio and the gossip press, as already identified by empirical research on this topic
(see, for example, Chenu, 2011; Mazzoni & Ciaglia, 2013), but also social media such as Facebook
(Enli & Skogerbø, 2013; McGregor et al., 2016). The hypothesis that directs the research conducted
in this paper is the following: Facebook is a social media in which political actors undergo a strong
incentive to popularize their political communication. As the use and centrality of Facebook (as
an instrument of political communication) increase, the incentive for political leaders to employ
popularization increases. Facebook is thus considered a digital platform in which political leaders
see in popularization a natural and recurring strategic communicative action that addresses their
political and electoral needs.
The article will then try to answer an additional research question:

RQ2 – What are the main necessary and/or sufficient conditions for the presence of a pop political
communication on Facebook?

A second aim of this paper is therefore to provide a first general explanation across cases to
identify the necessary and/or sufficient conditions that imply a pop communication deployed on
Facebook by the main leaders of contemporary liberal democracies. Answering this second research
question is particularly urgent because case studies on political leaders’ Facebook pages (Borah,
2016; Larsson, 2015; Mazzoleni & Bracciale, 2018) tend to exceed cross-country research (Larsson
& Kalsnes, 2014; Zulianello et al., 2018), with the automatic consequence of not knowing whether
their specific findings can be extended to other cases as well.

METHODOLOGY

A comparative analysis was designed to answer these two research questions. The notion of
popularization has been analysed on the Facebook pages of the main political leaders of 31 Western
and Latin American countries from 01/09/2012 to 31/10/2014 (Table 1).

15
International Journal of E-Politics
Volume 10 • Issue 1 • January-June 2019

The comparative analysis has been developed by following the diverse system design ideal type,
namely, the criterion that “has as its primary objective the achievement of maximum variance along
relevant dimensions” (Gerring, 2007, p. 97). In particular, the aim is to best control for the effect
of a specific variable: the presence of a presidential form of government. For this reason, together
with twenty Western countries, Mexico and ten South American countries have been included in
the research.
In this set of countries, 121 political leaders have been analysed (see Appendix 1). To decide
which leaders to include in the dataset, the following criteria have been considered: i) each head of
government3; ii) at least one executive and one opposition leader for each country; iii) for the countries
with a parliamentary system, all the leaders of the parties that, in the elections connected with the
period under consideration, have overcome a 15% threshold (with respect to the lower house); iv) for
the countries with a presidential system, all the candidates that, in the presidential elections connected
with the period under consideration, have overcome the 15% threshold.4
The Facebook pages were analysed only in the actual period in which each single political actor
included in the dataset was effectively the leader of her/his party. Therefore, not all the pages were
analysed for the 26 months under consideration.
A Facebook application called Netvizz was used to conduct this analysis. An automatic download
of all the posts published on the Facebook pages of the leaders under consideration was possible
thanks to Netvizz. Specifically, 97,898 posts were discovered. Among these 97,898 posts, 24,1805
posts were manually coded and analysed.
The coding process concerned all the leaders with a Facebook page who had posted at least 50
posts. Eighty-four out of 121 cases matched this criterion. Therefore, the content analysis concerns
only a cluster of the cases included in this comparative analysis. To determine intercoder reliability,
another researcher coded 523 posts. Reliability for the content analysis of the popularization on
Facebook is acceptable (Krippendorff’s alpha = 0.79).

Table 1. Countries under examination

Western Europe South America North America Oceania

Italy Brazil United States Australia
Spain Argentina Canada New Zealand
France Venezuela Mexico
Portugal Colombia
Germany Peru
Austria Bolivia
Belgium Chile
Netherlands Paraguay
United Kingdom Uruguay
Ireland Ecuador
Sweden
Finland
Norway
Denmark
Switzerland
Greece

16
 International Journal of E-Politics
Volume 10 • Issue 1 • January-June 2019

The posts were coded by applying the following operational definition:

The popularization of political communication can be realized mainly through three types of strategies:
i) by appealing to people and vicissitudes related to the private life of the political actor; ii) by using
and interacting with the media environments, the rhetoric and the actors belonging to the world of
television, music, sport and movie celebrities; or iii) by reproducing and interpreting the lifestyles
and the daily lives of citizens or the dominant customs and practices of the referring popular culture.

Therefore, popularization was codified as present when it could be traced to one of the
abovementioned three types of popularization, which refers to what the literature considers as empirical
references of intimate, celebrity and lifestyle politics. To best codify the pages under consideration,
each of these types of popularization have been divided into different sub-dimensions:

1. Intimate Politics
a. Popularization of the leader (reminiscences about the youth, hobbies, body of the leader,
etc.)
b. Popularization of the family of the leader (parents, spouse, children, etc.)
2. Celebrity politics
a. Popularization associated with actors, movie stars and other aspects of the entertainment
industry (sport, cinema, TV, etc.)
b. Popularization as presence in the pop media (entertainment TV shows, entertainment radio
programs, gossip press, etc.)
3. Lifestyle politics
a. Popularization as pop practices and languages (slang, gestures, conversational styles, etc.)
b. Popularization as admixture with the pop culture (traditional food, customs and traditions,
games, etc.)
c. Popularization as admixture with daily life (school, work, weather forecast, etc.)6

Intimate, celebrity and lifestyle politics are not (completely) mutually exclusive. In some specific
circumstances, these three types of popularization could overlap. This is, for instance, the case of
a hypothetical political actor who publishes on his or her Facebook page a selfie (lifestyle politics)
with a sport celebrity (celebrity politics) and a member of her/his family (intimate politics). In this
hypothetical case, this political actor is simply using the three types of popularization in a unique
communicative way.
As in the case of populism, on which empirical study is based on the identification of three main
indicators (people centrism, anti-elitism, exclusion of non-elite out-groups) combined to form an index
of populism (Bracciale & Martella, 2017; Zulianello et al., 2018), the analysis of popularization on
Facebook does not concentrate on intimate, celebrity and lifestyle politics separately. Although it is
of unquestionable scientific interest to understand whether intimate, lifestyle and celebrity contents
spread homogeneously (or not) across countries, ideological positioning, gender, and so forth, both
space limitations and the primary aim of this article – that is, focusing on popularization, not on its
main indicators – postponed this fine-grained line of inquiry to future research.

RESEARCH HYPOTHESES

A QCA (Ragin, 2000; Schneider & Wagemann, 2013) was conducted to shed light on why
contemporary leaders develop pop communication on Facebook. This methodological choice follows
one of the main hypotheses that have guided the development of this research project: the presence of

17
International Journal of E-Politics
Volume 10 • Issue 1 • January-June 2019

a presidential system is not a necessary condition for pop communication on Facebook. Thus, under
specific circumstances, even parliamentarian leaders will produce pop communication on Facebook.
Three systemic and two agency conditions have been included in the QCA conducted in this
paper: i) the presence of a presidential system; ii) a high digitalization of the media system; iii) a
high level of trust in political institutions; iv) governing leaders; and v) conservative leaders. The first
three conditions refer to systemic and contextual variables. With a presidential system of government,
candidates – not parties – play the major role in the political arena. It pushes political leaders into
both emotionally strengthening the bond with the their current sympathizers and building a personal
connection with the part of the electorate who still do not know, follow or support them. More than
in parliamentary democracies, popularization is expected to be achieved in presidential regimes.
A high digitalization of the media system should prompt pop political communication as well:
as the number of internet users OR the number of internet users with a Facebook account increases,
the chances for a political leader to communicate with the citizens who still do not know, follow or
support them, particularly with those who are less interested in the news, actors and dynamics related
to the sphere of politics, increase.
When citizens trust political institutions, leaders should feel more at ease in developing pop
communication on Facebook. When this occurs, they should be more likely to accept a communication
strongly pivoted around intimate, celebrity and lifestyle content. Conversely, in a country with
low levels of trust in political institutions, citizens may react negatively if leaders opt for pop
communication on social media, where they can immediately express their dissatisfaction about
political elites. Therefore, a high level of trust in political institutions may allow a political actor
to play with his/her political communication. This factor acts over the communicative freedom of
the politician in the sense that a high level of trust in political institutions gives her/him a greater
degree of freedom over how to build her/his political communication on Facebook within the pop-
institutional dichotomy.
In terms of agency dimensions, both governing and oppositional leaders have strong incentives
to produce pop communication on Facebook. The former need to develop an image that represents
them as political actors in touch with ordinary citizens – and not as bureaucrats detached from the
daily lives of their community. This hypothesis fits more in presidential systems, where the head of
government corresponds with the head of state, so a strong emphasis on the institutional dimension
is automatically part of his/her position. On the other hand, opposition leaders may see in pop
communication a strategy for gaining the consensus they lacked in the election campaign they lost,
with the aim of increasing their chances of winning the next one. This fourth condition does not
allow us to hypothesize a strong discriminant effect over popularization, which is true even from
a conjunctural perspective. Therefore, its inclusion in this QCA entails a theory-building purpose.
The critical theory developed by members of the Frankfurt School had a strong influence on the
ideological apparatus of left-wing parties in Western Europe. Books such as Dialect of Enlightenment
(Horkheimer & Adorno, 1944) presented a strong critique of any process of commercialization
affecting the political sphere. Popularization has been conceptualized as a strategy employed by
political actors to create new connections with those citizens who still do not know, follow or support
them and to emotionally strengthen the bond with their current sympathizers, and as such, it is part of
the process of commercialization highlighted and criticized by the Frankfurt School. Consequently, as
a leader becomes more closely positioned to the left pole of the left–right ideological dimension, the
incentives to not develop pop communication on Facebook may strengthen. In contrast, conservative
leaders should not have any ideological roots that affect their decision to employ popularization on
Facebook.

18
 International Journal of E-Politics
Volume 10 • Issue 1 • January-June 2019

FINDINGS

Levels of popularization have been estimated in percentage terms (Table 2). As an example, the 50.5%
value associated with the case of the former prime minister of Norway, Jens Stoltenberg, means that
among the 281 posts codified in his Facebook page, more than half include pop content.
In total, 21.1% of the codified posts are pop. The median value confirms these data: for half of
the leaders studied, more than 19.6% of published posts were pop posts. The percentage of pop posts
ranges from 50.5% for Jens Stoltenberg to 0.8% for Guglielmo Epifani (Italy). While some leaders
popularize their political communication on Facebook in a relevant way, others share very few pop
posts. As confirmed by the standard deviation value (SD = 12.1), the set of the main leaders of
Western and Latin American countries is associated with a relevant variance for the popularization
on Facebook.
As for the analytical aim of this comparative research, the first step to conduct a QCA consists
of the transformation of raw data (see Appendix 2) in set membership scores through a calibration
process (see Appendix 3), which finally results in a data matrix. The analysis of necessity based
on the data matrix does not show any necessary conditions. Conversely, the analysis of sufficiency
based on a truth table deduced by the data matrix indicates the presence of three different sufficient
conditions (Table 3): i) the presence of a presidential system; ii) a high digitalization of the media
system; and iii) a high level of trust in political institutions. In an explicit form, the solution term of
this QCA is as follows:

PRESIDENTIAL (P) + DIGITALIZATION (D) + TRUST (T) -> POP

Two out of five conditions are not present in the solution term. More precisely, the two conditions
that express the agency dimension of the cases are neither sufficient conditions nor INUS conditions.
Only systemic variables individually play a role in implying pop political communication on Facebook.
The parameters of fit of the solution term indicate that this QCA can explain almost all the cases
taken into consideration (coverage score of 0.91), even though they do not do so perfectly (consistency
score of 0.82). Only two cases prove to be deviant in terms of coverage: Beppe Grillo (Italy) and Jean-
Marc Ayrault (France). These are the only cases located in the upper-left quadrant of the XY plot of
the solution term (Figure 1): Grillo and Ayrault should be less pop on Facebook than they effectively
are. Regarding the consistency score, the true logical contradictory cases highlighted in Table 3 are
located in the lower right quadrant. These are the cases that strongly dispute the sufficiency argument:
these ten cases should be more pop on Facebook than they effectively are.

DISCUSSION

This paper has provided and empirically applied a new definition of the notion of popularization of
political communication. This concept has been essentially defined as a communicative action through
which political actors try to establish new connections with those citizens who still do not know, follow
or support them and to emotionally strengthen the bond with their current sympathizers. Through a
definition of meaning that considers intimate, celebrity and lifestyle politics as the main indicators
of popularization, political communication on the Facebook pages of the main political leaders of
Western and Latin American countries has been empirically analysed from a comparative perspective.
First, the findings show that popularization appears to be a relevant phenomenon on Facebook:
21.1% of the posts were codified as pop. One may argue that considering this percentage as relevant
entails an ambiguous interpretation of the data. One rarely talks about relevance when a social
phenomenon is observed with this percentage level. However, even an apparently limited percentage
of pop posts have the capacity to alter the quality of political communication developed on Facebook.
When a cup is filled with coffee, a teaspoon of milk is enough to alter the flavour and the colour of

19
International Journal of E-Politics
Volume 10 • Issue 1 • January-June 2019

Table 2. Percentage of pop posts per leader

Percentage of pop Number of coded Percentage of pop Number of coded

Leader Leader
posts posts posts posts
Stoltenberg 50.5 281 Rutte 19.3 187
Schmidt 49.0 100 Ramirez 19.0 210
Solberg 47.2 352 Campos 18.6 156
Fujimori 46.7 107 Mendez 18.0 100
Rudd 42.6 244 Alegre 17.5 200
DiRupo 40.1 419 Lasso 17.4 465
Massa 38.8 397 Steinbruck 17.3 346
Gillard 38.7 331 Rousseff 17.2 507
Store 38.4 86 Grillo 17.0 465
Key 35.5 465 Maduro 16.3 202
Humala 34.0 100 Medina 16.3 467
Jensen 34.0 321 Merkel 16.2 370
Abbott 33.5 313 Obrador 16.0 250
Santos 33.1 465 Toledo 15.4 104
Ominami 32.3 291 Urpilainen 14.0 179
Trudeau 32.2 419 Silva 13.5 465
Gutierrez 32.0 200 Renzi 13.0 100
Obama 31.6 465 Lopez 12.8 313
Franco 31.0 174 Miliband 12.7 363
Cunliffe 30.0 260 Boehner 12.4 105
Cameron 29.8 322 Gilmore 12.1 107
Neves 29.5 414 Bordaberry 11.7 463
Rasmussen 28.3 223 Clegg 11.0 155
Matthei 26.5 234 Zualaga 10.8 268
Shearer 25.7 171 Kenny 10.6 161
Kuczynski 25.6 395 Madero 10.3 397
Nieto 25.6 395 Sanchez 8.0 100
Cartes 25.2 416 Gysi 7.5 159
Ayrault 25.0 100 Fillon 7.4 81
Mulcair 24.9 338 Samaras 6.0 84
Shorten 24.6 167 Monti 5.8 224
Romney 24.5 102 Gabriel 5.7 476
Capriles 24.1 540 Rubalcaba 5.5 253
Correa 23.7 535 Berlusconi 5.5 454
Harper 23.4 488 Heber 5.1 117
Lacalle 23.0 200 Bersani 5.0 381
Faymann 22.2 464 LePen 3.6 422
Strache 21.3 465 Rajoy 3.5 427
DeKirchner 21.0 395 Letta 2.3 257
Bachelet 20.5 361 Tsipras 2.3 395
Rinne 20.4 93 Cope 1.9 215
Lofven 20.0 300 Epifani 0.8 127
Total 21.1
Mean 20.8
Median 19.6
SD 12.9

the coffee. If the amount of milk reaches 20% of the volume of the cup, the flavour and colour of the
coffee will be significantly altered. In the interaction between milk and coffee, it is not necessary that
the former reach 51% of the volume of the cup to determine a qualitative difference in the flavour

20
 International Journal of E-Politics
Volume 10 • Issue 1 • January-June 2019

Table 3. Solution term

raw unique true logical

consistency consistency cases
coverage coverage contradictory cases
Capriles, Lacalle, Lasso, CartesO, Alegre, Medina,
Correa, Maduro, Franco, CartesG, Fujimori, Massa,
Bordaberry, Le Pen,
Neves, Matthei, Romney, Ramirez, Mendez, Gutierrez,
PRESIDENTIAL + 0.80 0.46 0.10 Lopez, Zuluaga,
Humala, Obama, Nieto, De Kirchner, Rousseff,
Madero
Ominami, Kuczynski, Bachelet, Campos, Obrador,
Toledo, Silva, Santos
Cameron, Capriles, Lacalle, Lasso, Correa, Maduro,
Fujimori, Massa, Neves, Matthei, Romney, Ramirez,
Mendez, Gutierrez, Humala, Obama, Nieto, De Kirchner,
Bordaberry, Clegg,
Rousseff, Ominami, Kuczynski, Bachelet, Campos,
Gysi, GabrielO,
DIGITALIZATION Obrador, Toledo, Silva, Santos, SolbergG, Key, JensenG,
0.89 0.79 0.06 Lopez, Miliband,
+ AbbottG, Harper, Rutte, Merkel, SolbergO, AbbottO,
GabrielG, Zuluaga,
JensenO, Rasmussen, StoltenbergG, Schmidt, Rudd,
Madero
Di Rupo, Gillard, Rinne, StoltenbergO, Store, Trudeau,
Cunliffe, Shearer, Mulcair, Shorten, Lofven, Steinbruck,
Urpilainen
Capriles, Lacalle, Lasso, Correa, Maduro, Faymann,
Gutierrez, SolbergG, Key, JensenG, AbbottG, Harper,
Rutte, Merkel, SolbergO, AbbottO, JensenO, Rasmussen, Bordaberry, Gysi,
TRUST 0.87 0.55 0.01
StoltenbergG, Schmidt, Rudd, Di Rupo, Gillard, Rinne, GabrielO, GabrielG
StoltenbergO, Store, Trudeau, Cunliffe, Shearer, Mulcair,
Shorten, Lofven, Steinbruck, Strache, Urpilainen
Solution 0.82 0.91

Figure 1. XY plot of the solution term

of the latter. As for the milk and coffee, a percentage of 21.1% of pop posts significantly alters the
quality of political communication because, as noted, popularization is a strategic communicative
action that is strongly detached from the type of communication commonly and normatively associated
with the sphere of politics: that is to say, an institutional type of communication undiluted with any
elements of celebrity, intimate or lifestyle politics.
Regarding the analytical aim of the research, a QCA indicates that none of the conditions related
to agency dimensions play a discriminant role in the occurrence of popularization on Facebook. It does
not make any difference whether a leader is governing or not, nor does his/her ideological positioning
in the left-right dimension. Only systemic factors matter. Thus, leaders of the same country develop
the same qualitative attitude about popularization. With few exceptions, Figure 1 indicates that it

21
International Journal of E-Politics
Volume 10 • Issue 1 • January-June 2019

is substantially Italian, Spanish, Greek and Irish leaders who do not opt for pop communication on
Facebook. These are the countries where the conditions highlighted by the QCA do not appear: the
presence of a presidential system, the presence of a high digitalization of the media system, and the
presence of a high level of trust in political institutions do not occur simultaneously only in Italy,
Spain, Greece and Ireland.
These findings might suggest that popularization is destined to occur on Facebook. The hypothesis
is that, after a relevant increase in the media system’s digitalization, the likelihood of establishing
new connections with those citizens who still do not know, follow or support them and the need to
emotionally strengthen the bond with their current sympathizers will bring even the main political
leaders of countries such as Italy, Spain, Greece or Ireland to undertake pop communication on
Facebook. As we are not likely to see a presidential turn or an increase of the level of trust in political
institutions in countries such as Italy, Spain, Greece and Ireland, in the long term, it would be more
plausible to hypothesize a significant increase in the digitalization of the media system in these four
countries, which will push political leaders towards popularization on Facebook.
This comparative research has thus indicated that both parliamentary and presidential leaders
develop pop communication on Facebook, showing that the need to establish new connections
with those citizens who still do not know, follow or support them and to emotionally strengthen the
bond with their current sympathizers does not characterize only the political context surrounding
presidential government. Moreover, the leaders who employ pop communication the most are those
associated with a parliamentary form of government, such as Jens Stoltenberg (Norway) or Helle-
Thorning Schmidt (Denmark). As indicated by Table 2, leaders in Norway, Denmark, Australia and
New Zealand popularize their communication relevantly on Facebook precisely because, as the QCA
denotes, they operate in countries with a high digitalization of the media system and a high level of
trust in political institutions. At the same time, all the main Italian, Spanish, Greek and Irish leaders
have published a lower amount of pop posts not because they operate in a country with a parliamentary
form of government but because none of these countries is associated with the presence of a high
digitalization of the media system or a high level of trust in political institutions. By considering the
causal mechanisms described in the research hypotheses section, this particular configuration has not
pushed and advised these politicians to adopt pop communication on Facebook.
Another important remark concerns the explicative payload of these findings when considering
the entire media system. In particular, the analysis of Facebook on Facebook has to be considered only
a first necessary step – not a sufficient one – to fully understand how the popularization developed
by political leaders on Facebook is affecting political communication dynamics. The hybrid nature
of the contemporary media system implies the need for cross-platform and cross-media analyses.
Pop contents might be highly newsworthy for, and thus covered by, mainstream media, amplifying
the communicative payload of a Facebook post beyond this specific social media environment. As
stated by the Norwegian Labour party’s head of communication:

“My impression and experience is that media often publishes tweets or Facebook updates when it comes
to sporting events (example the prime minister congratulating someone winning gold in the Olympics),
and tragic events (condolences and similar). They also sometimes pick up humorous updates. But more
political updates not so much. One example from 2013 is Støre playing hide and seek with children,
where he honours the Norwegian Trekking Association’s work for public health: https://www.facebook.
com/jonasgahrstore/photos/a.475877444237.257221.282256964237/10151823368574238 Which
led to this in one of Norway’s largest newspapers: http://www.vg.no/nyheter/innenriks/valget-2013/
se-jonas-leker-gjemsel/a/10112676/”.7

Beyond the presence of a presidential system, a high digitalization of the media system, and a
high level of trust in political institution, even the hybrid nature of the contemporary media system

22
 International Journal of E-Politics
Volume 10 • Issue 1 • January-June 2019

might play a crucial role in developing a communicative environment in which every political leader
opts for pop communication on Facebook.
Although this article has focused on a digital media platform such as Facebook, it is thus worth
pointing out that the effects and use of popularization involve other media as well. More precisely,
if popularization is a strategic communicative action through which political actors try to establish
new connections with those citizens who still do not know, follow or support them and to emotionally
strengthen the bond with their current sympathizers, it could follow that, for instance, the picture of
a totally naked politician on the first page of a celebrity magazine is not an empirical reference of
popularization. Specific non-consensual flows of information about and imagery of politicians would
seem to drop out of the conceptualization of popularization proposed (Stanyer, 2013). However, this
process simply leads to a first property of popularization. In fact, popularization can be direct and
indirect. On one hand, popularization is direct when directly produced by a political actor, as the one
taken into consideration in this comparative research on Facebook. On the other hand, popularization is
indirect each time a media content satisfies the goal of inserting a specific political actor into the media
diet of those citizens who still do not know, follow or support them, without his/her permission/control
on the content published. In this case, popularization is still produced but simply in an indirect way.
This first property of popularization is strictly connected with a second one: popularization
is also media-related. A politician can decide to be pop on Facebook and Instagram but not pop
at all on Twitter due to the affordances of these different social media that would thus shape his/
her communicative behaviour. At the same time, s/he can be pop on television and not pop at all on
social media due to the absence of a high level of trust in political institutions. As the QCA conducted
here reveals, this situation might represent a factor that would advice political actors not to employ
popularization in a media environment in which dissatisfied citizens might immediately unleash their
anger on them if they see their public officials not behaving as they believe they should do so in harsh
political and economic times. Matteo Renzi, former Italian prime minister, embodies a paradigmatic
example of this second property of popularization. Empirical analyses on his communication found
that he extensively used popularization on television and in gossip magazines (Mazzoni and Ciaglia,
2014), not on Facebook. Therefore, when popularization is studied, the analytical field must be clearly
delimitated. As it analysed popularization only on Facebook, all the empirical evidence found in this
article must be circumscribed to the social media under consideration.
Third, as it is a strategic communicative action, such popularization is a temporary trait that
each single political actor can decide to have at a specific T0 time and then abandon at a second T1
time. A political actor can choose to adopt pop communication only in a specific moment of her/his
political career. Each political actor can “play” with her/his political communication by being more
pop or more institutional under specific circumstances based on her/his political and electoral goals,
such as during an election campaign where the pressing need to collect as many votes as possible
could guide even a leader who is not comfortable with popularization to use it – see Mario Monti’s
choice during the 2013 Italian elections to popularize his political communication in gossip magazines
(Mazzoni & Ciaglia, 2014, p. 99) – although empirical analyses on this topic show that the presence
of an election campaign tend not to alter the communicative style of contemporary political leaders
on Facebook (Ceccobelli, 2018).

CONCLUSION

Having designed a comparative analysis, a rigorous and scientific investigation of the causal
mechanisms that lead a political leader to produce a pop communication on Facebook is beyond the
scope of this paper. For this reason, the design of different case studies that can identify the causal
mechanisms that explain the choice of developing pop communication both on Facebook and on
other media outlets is particularly needed and welcomed.

23
International Journal of E-Politics
Volume 10 • Issue 1 • January-June 2019

This need is particularly relevant for the deviant cases in coverage (upper-left quadrant in Figure
1) identified with the QCA presented herein, such as Beppe Grillo. The political communication of
the populist leader Beppe Grillo strongly diverges from that developed by other Italian cases. Because
another new relevant Italian populist leader, Matteo Salvini, seemed to follow Beppe Grillo’s path
toward the employment of popularization on social media (Bracciale & Martella, 2018), we may be
able to hypothesize that populist leaders feel more comfortable with this type of communication.
However, the inclusion of the populism condition in a second QCA does not indicate any effect of
populism on popularization, as the cases of Silvio Berlusconi (Italy), Marine Le Pen (France) and
Alexis Tsipras (Greece) – three populist leaders who do not employ almost any popularization on
Facebook – testify. This comparative research thus confirms that populism and popularization, although
associated by an apparently common etymological root, represent two very different phenomena, and
therefore, they have to be treated and analysed accordingly in order to avoid conceptual stretching
with them (e.g., applying the indicator of lifestyle politics – political leaders behaving and portrayed
like ordinary citizens – for the analysis of populism).
Finally, other research should devote specific attention to the effects of popularization on citizens’
engagement, particularly on social media. Do political actors succeed in establishing new connections
with those citizens who still do not know, follow or support them and emotionally strengthening
the political bond with their current sympathizers when employing popularization on Facebook?
Answering this research question would allow us to understand whether the choice of publishing
pop content accomplishes political actors’ goals in making use of it or whether it represents simply
a shiny but ineffective communicative diversion.

24
 International Journal of E-Politics
Volume 10 • Issue 1 • January-June 2019

REFERENCES

Auter, Z. J., & Fine, J. A. (2016). Negative Campaigning in the Social Media Age: Attack Advertising on
Facebook. Political Behavior, 38(4), 999–1020. doi:10.1007/s11109-016-9346-8
Bennett, W. L. (1998). The uncivic culture: Communication, identity, and the rise of lifestyle politics. PS, Political
Science & Politics, 31(4), 740–761. doi:10.2307/420711
Borah, P. (2016). Political Facebook use: Campaign strategies used in 2008 and 2012 presidential elections.
Journal of Information Technology & Politics, 13(4), 326–338. doi:10.1080/19331681.2016.1163519
Bracciale, R., & Martella, A. (2017). Define the populist political communication style: The case of Italian
political leaders on Twitter. Information Communication and Society, 20(9), 1310–1329. doi:10.1080/136911
8X.2017.1328522
Ceccobelli, D. (2018). Not every day is election day: A comparative analysis of eighteen election campaigns on
Facebook. Journal of Information Technology & Politics, 15(2), 122–141. doi:10.1080/19331681.2018.1449701
Ceccobelli, D., & Cotta, B. (2016). Leaders’ ‘Green’ Posts. The Environmental Issues Shared by Politicians on
Facebook. European Policy Analysis, 2(2), 68–93.
Chadwick, A. (2006). Internet Politics: States, Citizens, and New Communication Technologies. New York:
Oxford University Press.
Chadwick, A. (2013). The hybrid media system: Politics and power. New York, NY: Oxford University Press.
doi:10.1093/acprof:oso/9780199759477.001.0001
Chenu, A. (2011). From Paths of Glory to Celebrity Boulevards: Sociology of Paris Match. Revue Francaise
de Sociologie, 51(5), 69–116. doi:10.3917/rfs.515.0069
Dalton, R. J., McAllister, I., & Wattenberg, M. P. (2000). The consequences of partisan dealignment. In R. J.
Dalton & M. P. Wattenberg (Eds.), Parties without partisans: Political change in advanced industrial democracies
(pp. 37–63). Oxford, UK: Oxford University Press.
De Vreese, C. H., Esser, F., Aalberg, T., Reinemann, C., & Stanyer, J. (2018). Populism as an expression of
political communication content and style: A new perspective. The International Journal of Press/Politics, 23(4),
423–438. doi:10.1177/1940161218790035 PMID:30886670
Engesser, S., Fawzi, N., & Larsson, A. O. (2017). Populist online communication: Introduction to the special
issue. Information Communication and Society, 20(9), 1279–1292. doi:10.1080/1369118X.2017.1328525
Enli, G. S., & Skogerbø, E. (2013). Personalized Campaigns in Party-Centred Politics. Information Communication
and Society, 16(5), 757–774. doi:10.1080/1369118X.2013.782330
Filimonov, K., Russmann, U. & Svensson, J. (2016). Picturing the Party: Instagram and Party Campaigning in
the 2014 Swedish Elections. Social Media + Society, 2(3), 1–11.
Flaxman, S., Goel, S., & Rao, J. M. (2016). Filter Bubbles, Echo Chambers, and online News Consumption.
Public Opinion Quarterly, 80(1), 298–320. doi:10.1093/poq/nfw006
Gainous, J., & Wagner, K. M. (2014). Tweeting to power: The social media revolution in American politics.
New York: Oxford University Press.
Gerring, J. (2007). Case study research: principles and practices. Cambridge: Cambridge University Press.
Goertz, G. (2005). Social Science Concepts. A User’s Guide. Princeton: Princeton University Press.
Haim, M., Graefe, A., & Brosius, H.-B. (2017). Burst of the filter bubble? Effects of personalization on the
diversity of Google News. Digital Journalism, 6(3), 330–343. doi:10.1080/21670811.2017.1338145
Hallin, D. C., & Mancini, P. (2004). Comparing media systems: Three models of media and politics. Cambridge,
UK: Cambridge University Press. doi:10.1017/CBO9780511790867

25
International Journal of E-Politics
Volume 10 • Issue 1 • January-June 2019

Hannak, A., Sapiezynski, P., Kakhki, A. M., Krishnamurthy, B., Lazer, D., Mislove, A., & Wilson, C. (2013).
Measuring Personalization of Web Search. Proceedings of the 22nd International Conference on World Wide Web
(pp. 527-538). International World Wide Web Conferences Steering Committee. doi:10.1145/2488388.2488435
Inglehart, R. (1997). Modernization and Postmodernization. Cultural, Economic, and Political Change in 43
Societies. Princeton, NJ: Princeton University Press.
Iyengar, S. (1987). Television news and citizens’ explanations of national affairs. The American Political Science
Review, 81(3), 815–831. doi:10.2307/1962678
Iyengar, S., & Hahn, K. S. (2009). Red media, blue media: Evidence of ideological selectivity in media use.
Journal of Communication, 59(1), 19–39. doi:10.1111/j.1460-2466.2008.01402.x
Jomini Stroud, N. (2011). Niche news: The politics of news choice. New York, NY: Oxford University Press.
doi:10.1093/acprof:oso/9780199755509.001.0001
Kalsnes, B. (2016). The Social Media Paradox Explained: Comparing Political Parties’ Facebook Strategy Versus
Practice. Social Media + Society, 2(2), 1–11.
Karvonen, L. (2010). The Personalisation of politics: a study of parliamentary democracies. Colchester: ECPR
Press.
Kreiss, D., & McGregor, S. C. (2017). Technology firms shape political communication: The work of Microsoft,
Facebook, Twitter, and Google with campaigns during the 2016 U.S. presidential cycle. Political Communication,
35(2), 155–177. doi:10.1080/10584609.2017.1364814
Langer, A. I. (2010). Politicization of the Private Persona: Exceptional Leaders or the New Rule? The Case
of the United Kingdom and the Blair Effect. The International Journal of Press/Politics, 15(1), 60–76.
doi:10.1177/1940161209351003
Larsson, A. O. (2015). Pandering, protesting, engaging. Norwegian party leaders on Facebook during the 2013
‘Short campaign’. Information Communication and Society, 18(4), 459–473. doi:10.1080/1369118X.2014.967269
Larsson, A. O., & Kalsnes, B. (2014). Of course we are on Facebook—” Social media adoption in Swedish and
Norwegian parliaments. European Journal of Communication, 12(2), 193–218.
Mancini, P. (2011). Between Commodification and Lifestyle Politics: Does Silvio Berlusconi Provide a New
Model of Politics for the Twenty-first Century? Oxford: University of Oxford Reuters Institute for the Study of
Journalism.
Mancini, P. (2013). Media fragmentation, party system, and democracy. The International Journal of Press/
Politics, 18(1), 43–60. doi:10.1177/1940161212458200
Mazzoleni, G. & Bracciale, R. (2018). Socially mediated populism: the communicative strategies of political
leaders on Facebook. Palgrave Communications, 4(50).
Mazzoleni, G., & Sfardini, A. (2009). Politica Pop. Da ‘Porta a Porta’ a ‘L’isola dei famosi. Bologna: Il Mulino.
Mazzoni, M., & Ciaglia, A. (2013). How Italian politics goes popular: Evidence from an empirical
analysis of gossip magazines and TV shows. International Journal of Cultural Studies, 17(4), 381–398.
doi:10.1177/1367877913496199
Mazzoni, M., & Ciaglia, A. (2014). An incomplete transition? How Italian politicians manage the celebritisation
of politics. Celebrity Studies, 5(1-2), 93–106. doi:10.1080/19392397.2013.797656
McGregor, S. C., Lawrence, R. G., & Cardona, A. (2016). Personalization, gender, and social media: Gubernatorial
candidates’ social media strategies. Information Communication and Society, 20(2), 264–283. doi:10.1080/13
69118X.2016.1167228
Pariser, E. (2011). The Filter Bubble: What the Internet is Hiding from You. London: Penguin UK.
Prior, M. (2007). Post-broadcast democracy: How media choice increases inequality in political involvement
and polarizes elections. Cambridge: Cambridge University Press. doi:10.1017/CBO9781139878425
Rahat, G., & Sheafer, T. (2007). The personalization(s) of politics: Israel, 1949–2003. Political Communication,
24(1), 65–80. doi:10.1080/10584600601128739

26
 International Journal of E-Politics
Volume 10 • Issue 1 • January-June 2019

Samuel-Azran, T., M. Yarchi & G. Wolfsfeld (2015). Equalization versus Normalization: Facebook and the 2013
Israeli Elections. Social Media + Society, 1(2), 1–9.
Sartori, G. (1970). Concept misformation in comparative politics. The American Political Science Review, 64(4),
1033–1053. doi:10.2307/1958356
Schneider, C. Q., & Wagemann, C. (2013). Set-Theoretic Methods for the Social Sciences: A Guide to Qualitative
Comparative Analysis. New York: Cambridge University Press.
Stanyer, J. (2013). Intimate Politics: Publicity, Privacy and the Personal Lives of Politicians in Media Saturated
Democracies. Cambridge: Polity Press.
Street, J. (2004). Celebrity politicians: Popular culture and political representation. British Journal of Politics
and International Relations, 6(4), 435–452. doi:10.1111/j.1467-856X.2004.00149.x
Street, J. (2015). Popularization of politics. In G. Mazzoleni, K. G. Barnhurs, K. Ikeda, H. Wessler, & R. C. Maia
(Eds.), The International Encyclopaedia of Political Communication (pp. 1–5). Hoboken: Wiley.
West, D. M., & Orman, J. M. (2003). Celebrity politics. Upper Saddle River: Prentice Hall.
Wheeler, D. L. (2009). Working around the state: internet use and political identity in the Arab world. In A.
Chadwick & P.H. Howard (Eds.), Routledge Handbook of Internet Politics. London: Routledge.
Wheeler, M. (2013). Celebrity politics. Cambridge: Polity Press.
Zulianello, M., Albertini, A., & Ceccobelli, D. (2018). A Populist Zeitgeist? The Communication Strategies of
Western and Latin American Political Leaders on Facebook. The International Journal of Press/Politics, 23(4),
439–457. doi:10.1177/1940161218783836

ENDNOTES
1
Although the echo chamber hypothesis is still predominant within both the scientific and the general
public debate, more recent empirical analyses on this topic have begun questioning its real impact and
effects (Flaxman et al., 2016; Haim et al., 2017).
2
Source: http://www.oxforddictionaries.com/.
3
For the French case, all heads of state and prime ministers have been included. For the Swiss case, all
presidents of the federal council have been considered.
4
When needed, exceptional criteria were applied for some countries.
5
The guiding criteria applied for the sampling process are the following: for each leader, at least one post
every two days in which the leader was involved in an election campaign was analysed. Let us give some
examples. A first hypothetical leader A was included in the analysis for 400 days in which no election
campaign was conducted, and in this period, he/she published 150 posts. In this case, all the 150 posts
have been coded. A second hypothetical leader B was also included in the analysis for 400 days in which
no election campaign was conducted, but he/she published 2000 posts in this time period. In this second
case, 200 posts were coded by following a casual sampling process. A third hypothetical leader C was
included in the analysis for 400 days in which he/she was involved in an election campaign, and he/she
published 2000 posts in this time period. In this third case, 230 posts were coded. More precisely, 170
posts were coded during the 340 days in which the leader was not involved in an election campaign, plus
at least 60 further posts during the 60 days before the corresponding Election Day were coded. The choice
to double, if necessary, the number of posts coded during the 60 days before an Election Day follows the
empirical observation of the data. With the unique exception of Werner Faymann (Austria), all the leaders
included in the dataset increased the number of posts published in their Facebook pages significantly
during the election campaigns. To code only one post every two days, even during an election campaign,
would have weakened the empirical analysis too much by underestimating this specific political time
period.

6
At the following links, there are three empirical examples of popularization as Facebook posts, including
elements of celebrity (https://www.facebook.com/889307941125736/posts/1318730191516840), intimate
(https://www.facebook.com/1462957374004775/posts/1485292738437905) and lifestyle (https://www.
facebook.com/653092548048400/posts/963453483678970) politics.

27
International Journal of E-Politics
Volume 10 • Issue 1 • January-June 2019

7
. Bjorn Tore Hansen, the Norwegian Labour Party’s head of communication, was interviewed by email
on April 27, 2015. The first link refers to a photo published in the Facebook page of Jonas Gahr Store,
in which the Norwegian leader is playing hide and seek with children. The second link refers to a news
story published in one of the Norway’s largest newspapers regarding the publication of the aforementioned
Facebook post.

28
 International Journal of E-Politics
Volume 10 • Issue 1 • January-June 2019

APPENDIX 1

Table 4. Leaders included in the analysis

Posts published Posts published in

Time period under Time period under
Leader in the time period Leader time period under
consideration consideration
under consideration consideration
01/09/2012- 01/09/2012-
De Kirchner (Arg) 2,289 Kenny (Ire) 250
31/10/2014 31/10/2014
01/09/2012- 01/09/2012-
Massa (Arg) 1,030 Gilmore (Ire) 116
31/10/2014 04/07/2014
01/09/2012- 05/07/2014-
Abbott (Aus) 484 Burton (Ire) 0
31/10/2014 31/10/2014
01/09/2012- 08/12/2013-
Gillard (Aus) 406 Renzi (Ita) 133
26/06/2013 31/10/2014
26/06/2013- 01/09/2012-
Rudd (Aus) 244 Berlusconi (Ita) 680
12/10/2013 31/10/2014
13/10/2013- 01/09/2012-
Shorten (Aus) 332 Grillo (Ita) 16,439
31/10/2014 31/10/2014
02/09/2014- 01/09/2012-
Mitterlehner (Au) 0 Bersani (Ita) 597
31/10/2014 23/04/2013
01/09/2012- 01/09/2012-
Spindelegger (Au) nc Monti (Ita) 211
01/09/2014 28/04/2013
01/09/2012- 28/04/2013-
Strache (Au) 5,087 Letta (Ita) 279
31/10/2014 22/02/2014
01/09/2012- 11/05/2013-
Faymann (Au) 1,465 Epifani (Ita) 254
31/10/2014 15/12/2013
01/09/2012- 01/09/2012-
De Wever (Bel) 0 Nieto (Mex) 2,027
31/10/2014 31/10/2014
01/09/2012- 01/09/2012-
Di Rupo (Bel) 698 Obrador (Mex) 500
31/10/2014 31/10/2014
11/10/2014- 01/09/2012-
Michel (Bel) 30 Madero (Mex) 2,336
31/10/2014 29/09/2014
01/09/2012- 30/09/2014-
Morales (Bol) 0 Anaya (Mex) 4
31/10/2014 31/10/2014
01/09/2012- 01/09/2012-
Medina (Bol) 6,038 Stoltenberg (Nor) 332
31/10/2014 14/06/2014
18/05/2013- 01/09/2012-
Neves (Bra) 1,951 Jensen (Nor) 502
31/10/2014 31/10/2014
01/09/2012- 14/06/2014-
Silva (Bra) 2,523 Store (Nor) 86
31/10/2014 31/10/2014
01/09/2012- 01/09/2012-
Rousseff (Bra) 4,890 Solberg (Nor) 564
31/10/2014 31/10/2014
05/10/2013- 01/09/2012-
Campos (Bra) 1,505 Key (NZ) 1,425
13/08/2014 31/10/2014
01/09/2012- 15/09/2013-
Guerra (Bra) 0 Cunliffe (NZ) 687
17/05/2013 30/09/2014
01/09/2012- 01/09/2012-
Rae (Can) 0 Shearer (NZ) 342
14/04/2013 15/09/2013
15/04/2013- 01/10/2014-
Trudeau (Can) 872 Parker (NZ) 24
31/10/2014 31/10/2014
01/09/2012- 01/09/2012-
Harper (Can) 932 Wilders (Net) nc
31/10/2014 31/10/2014
01/09/2012- 01/09/2012-
Mulcair (Can) 547 Rutte (Net) 373
31/10/2014 31/10/2014
16/12/2013- 01/09/2012-
Melero (Chi) 0 Samsom (Net) 20
09/05/2014 31/10/2014
01/09/2012- 07/08/2013-
Walker (Chi) 0 Saguier (Par) 0
30/06/2013 31/10/2014
01/07/2013- 09//12/2012-
Bachelet (Chi) 980 Cartes (Par) 1,796
31/10/2014 31/10/2014
01/09/2012- 17/12/2012-
Ominami (Chi) 1,414 Alegre (Par) 611
17/11/2013 06/08/2013
01/09/2012- 01/09/2012-
Piñera (Chi) 10 Franco (Par) 426
15/12/2013 15/08/2013
18/08/2013- 01/09/2012-
Matthei (Chi) 707 Fujimori (Per) 183
15/12/2013 31/10/2014

continued on following page

29
International Journal of E-Politics
Volume 10 • Issue 1 • January-June 2019

Table 4. Continued

Posts published Posts published in

Time period under Time period under
Leader in the time period Leader time period under
consideration consideration
under consideration consideration
10/05/2014- 01/09/2012-
Mendez (Chi) 148 Kuczynski (Per) 1,449
31/10/2014 31/10/2014
01/09/2012- 01/09/2012-
Santos (Col) 853 Toledo (Per) 207
31/10/2014 31/10/2014
26/01/2014- 01/09/2012-
Zuluaga (Col) 1,577 Humala (Per) 172
31/10/2014 31/10/2014
01/09/2012- 01/09/2012-
Lopez (Col) 1,451 Seguro (Por) nc
31/10/2014 28/09/2014
26/01/2014- 01/09/2012-
Ramírez (Col) 939 Coelho (Por) 2
31/10/2014 31/10/2014
01/09/2012- 29/09/2014-
Schmidt (Dan) 181 Costa (Por) 1
31/10/2014 31/10/2014
01/09/2012- 01/09/2012-
Rasmussen (Dan) 446 Rajoy (Spa) 794
31/10/2014 31/10/2014
01/09/2012- 01/09/2012-
Gutiérrez (Ecu) 892 Rubalcaba (Spa) 387
17/02/2013 12/07/2014
01/09/2012- 13/07/2014-
Correa (Ecu) 2,571 Sanchez (Spa) 220
31/10/2014 31/10/2014
01/09/2012- 01/09/2012-
Lasso (Ecu) 2,352 Lofven (Swe) 459
31/10/2014 31/10/2014
01/09/2012- 01/09/2012-
Sipilä (Fin) 0 Reinfeldt (Swe) 27
31/10/2014 31/10/2014
01/09/2012- 01/01/2014-
Soini (Fin) 0 Burkhalter (Swi) 0
31/10/2014 31/10/2014
01/09/2012- 01/01/2013-
Katainen (Fin) nc Maurer (Swi) 0
23/06/2014 31/12/2013
24/06/2014- 01/09/2012-
Stubb (Fin) 27 Müller (Swi) 0
31/10/2014 31/10/2014
01/09/2012- 01/09/2012-
Urpilainen (Fin) 357 Sommaruga (Swi) 0
06/06/2014 31/10/2014
07/06/2014- 01/09/2012-
Rinne (Fin) 93 Brunner (Swi) 6
31/10/2014 31/10/2014
01/09/2012- 01/09/2012-
Sarkozy (Fra) 1 Widmer (Swi) 36
18/11/2012 31/12/2012
01/09/2012- 01/09/2012-
Le Pen (Fra) 1,458 Cameron (UK) 644
31/10/2014 31/10/2014
01/09/2012- 01/09/2012-
Hollande (Fra) 18 Clegg (UK) 214
31/10/2014 31/10/2014
16/06/2014- 01/09/2012-
Fillon (Fra) 81 Miliband (UK) 541
31/10/2014 31/10/2014
19/11/2012- 01/09/2012-
Copè (Fra) 308 Mujica (Uru) 0
15/06/2014 31/10/2014
01/04/2014- 01/06/2014-
Valls (Fra) 16 Vazquez (Uru) 0
31/10/2014 31/10/2014
01/09/2012- 01/06/2014-
Ayrault (Fra) 168 Lacalle (Uru) 301
31/03/2014 31/10/2014
01/09/2012- 01/09/2012-
Merkel (Ger) 463 Bordaberry (Uru) 1,485
31/10/2014 31/10/2014
28/09/2012- 01/09/2012-
Steinbruck (Ger) 525 Heber (Uru) 274
27/09/2013 22/04/2013
01/09/2012- 01/09/2012-
Gabriel (Ger) 1,662 Obama (USA) 1,216
31/10/2014 31/10/2014
17/12/2013- 01/09/2012-
Gysi (Ger) 350 Romney (USA) 308
31/10/2014 06/11/2012
01/09/2012- 07/11/2012-
Tsipras (Gre) 1,183 Boehner (USA) 210
31/10/2014 31/10/2014
01/09/2012- 01/09/2012-
Samaras (Gre) 84 Capriles (Ven) 4,025
31/10/2014 31/10/2014
01/09/2012- 01/09/2012-
Martin (Ire) 0 Chavez (Ven) 1
31/10/2014 05/03/2013
14/02/2013-
Maduro (Ven) 1,076
31/10/2014

30
 International Journal of E-Politics
Volume 10 • Issue 1 • January-June 2019

APPENDIX 2

Table 5. Raw data

% FB penetration / internet

%internet OR

ideological positioning
% internet penetration

% trust in government

% trust in parliament
form of government

% trust in political

role of the leader

% satisfaction for
% pop posts

institutions
penetration

democracy
leader

internet
%FB/
StoltenbergG 52,7 parliamentary 96,9 61 96,9 66 55 75 65,3 executive center-left
Schmidt 49 parliamentary 90 61 90 46 61 88 65 executive center-left
SolbergG 47,7 parliamentary 96,9 61 96,9 66 55 75 65,3 executive center-right
SolbergO 46,9 parliamentary 96,9 61 96,9 66 55 75 65,3 opposition center-right
Fujimori 46,7 presidential 48 81,3 81,3 39 31 25 31,7 opposition center-right
StoltenbergO 45,7 parliamentary 96,9 61 96,9 66 55 75 65,3 opposition center-left
Rudd 42,6 parliamentary 88,8 60 88,8 45 46 72 54,3 executive center-left
Di Rupo 40,1 parliamentary 81,3 58 81,3 43 44 66 51 executive center-left
Massa 38,8 presidential 60 83,3 83,3 43 34 51 42,7 opposition center-right
Gillard 38,7 parliamentary 88,8 60 88,8 45 46 72 54,3 executive center-left
Store 38,4 parliamentary 96,9 61 96,9 66 55 75 65,3 opposition center-left
AbbottO 38,3 parliamentary 88,8 60 88,8 45 46 72 54,3 opposition center-right
Key 35,5 parliamentary 86,6 60 86,6 54 20 65 46,3 executive center-right
Jensen 35,3 parliamentary 96,9 61 96,9 66 55 75 65,3 executive right
CartesO 34 presidential 34 76,5 76,5 30 62 25 39 opposition center-right
Humala 34 presidential 48 81,3 81,3 39 31 25 31,7 executive left
Santos 33,1 presidential 48 85,4 85,4 53 38 28 39,7 executive center-right
Ominami 32,3 presidential 58 87,9 87,9 29 28 38 31,7 opposition center-left
Trudeau 32,2 parliamentary 83 63 83 60 17 55 44 opposition center-left
Gutierrez 32 presidential 55 87,3 87,3 73 60 59 64 opposition center-left
Obama 31,6 presidential 84,2 62,9 84,2 30 23 75 42,7 executive center-left
Franco 31 presidential 34 76,5 76,5 30 62 25 39 executive center-right
Jensen 30,1 parliamentary 96,9 61 96,9 66 55 75 65,3 executive right
Cunliffe 30 parliamentary 86,6 60 86,6 54 20 65 46,3 opposition center-left
Cameron 29,8 parliamentary 83,6 62 83,6 25 26 61 37,3 executive center-right
Neves 29,5 presidential 53 81,1 81,1 56 33 26 38,3 opposition center-right
AbbottG 28,5 parliamentary 88,8 60 88,8 45 46 72 54,3 executive center-right
Rasmussen 28,3 parliamentary 90 61 90 46 61 88 65 opposition center-right
Matthei 26,5 presidential 58 87,9 87,9 29 28 38 31,7 opposition center-right
Shearer 25,7 parliamentary 86,6 60 86,6 54 20 65 46,3 opposition center-left
Kuczynski 25,6 presidential 48 81,3 81,3 39 31 25 31,7 opposition center-left
Nieto 25,6 presidential 41 82,9 82,9 46 42 21 36,3 executive center-left
Ayrault 25 semi-presid. 79,6 49 79,6 17 24 52 31 executive center-left
Mulcair 24,9 parliamentary 83 63 83 60 17 55 44 opposition center-left
Shorten 24,6 parliamentary 88,8 60 88,8 45 46 72 54,3 opposition center-left
Romney 24,5 presidential 84,2 62,9 84,2 30 23 75 42,7 opposition center-right
Capriles 24,1 presidential 51 90,2 90,2 47 50 42 46,3 opposition center-right
Correa 23,7 presidential 55 87,3 87,3 73 60 59 64 executive left
Harper 23,4 parliamentary 83 63 83 60 17 55 44 executive center-right
Lacalle 23 presidential 53 81,1 81,1 61 54 82 65,7 opposition center-right
Faymann 22,2 parliamentary 79,8 44 79,8 33 35 64 44 executive center-left
Strache 21,3 parliamentary 79,8 44 79,8 33 35 64 44 opposition right
DeKirchner 21 presidential 60 83,3 83,3 43 34 51 42,7 executive center-left
Bachelet 20,5 presidential 58 87,9 87,9 29 28 38 31,7 executive center-left
Rinne 20,4 parliamentary 89,4 49 89,4 56 66 80 67,3 executive center-left
CartesG 20 presidential 34 76,5 76,5 30 62 25 39 executive center-right
Lofven 20 parliamentary 92,7 59 92,7 54 67 84 68,3 opposition center-left
Rutte 19,3 parliamentary 92,9 49 92,9 50 54 80 61,3 executive center-right

continued on following page

31
International Journal of E-Politics
Volume 10 • Issue 1 • January-June 2019

Table 5. Continued

% FB penetration / internet

%internet OR

ideological positioning
% internet penetration

% trust in government

% trust in parliament
form of government

% trust in political

role of the leader

% satisfaction for
% pop posts

institutions
penetration

democracy
leader

internet
%FB/
Ramirez 19 presidential 48 85,4 85,4 53 38 28 39,7 opposition right
Campos 18,6 presidential 53 81,1 81,1 56 33 26 38,3 opposition center-left
Mendez 18 presidential 58 87,9 87,9 29 28 38 31,7 opposition center-right
Alegre 17,5 presidential 34 76,5 76,5 30 62 25 39 opposition center-right
Lasso 17,4 presidential 55 87,3 87,3 73 60 59 64 opposition center-right
Steinbruck 17,3 parliamentary 83 36 83 48 51 72 57 opposition center-left
Rousseff 17,2 presidential 53 81,1 81,1 56 33 26 38,3 executive center-left
Grillo 17 parliamentary 58,4 65 65 17 14 27 19,3 opposition NC
Maduro 16,3 presidential 51 90,2 90,2 47 50 42 46,3 executive left
Medina 16,3 presidential 38 76,3 76,3 55 37 38 43,3 opposition center-right
Merkel 16,2 parliamentary 83 36 83 48 51 72 57 executive center-right
Obrador 16 presidential 41 82,9 82,9 46 42 21 36,3 opposition center-left
Toledo 15,4 presidential 48 81,3 81,3 39 31 25 31,7 opposition center-left
Urpilainen 14 parliamentary 89,4 49 89,4 56 66 80 67,3 executive center-left
Silva 13,5 presidential 53 81,1 81,1 56 33 26 38,3 opposition center-left
Renzi 13 parliamentary 58,4 65 65 17 14 27 19,3 executive center-left
Lopez 12,8 presidential 48 85,4 85,4 53 38 28 39,7 opposition center-left
Miliband 12,7 parliamentary 83,6 62 83,6 25 26 61 37,3 opposition center-left
Gilmore 12,1 parliamentary 76,8 60 76,8 22 22 56 33,3 opposition center-left
Bordaberry 11,7 presidential 53 81,1 81,1 61 54 82 65,7 opposition center-right
Clegg 11 parliamentary 83,6 62 83,6 25 26 61 37,3 executive center-right
Zualaga 10,8 presidential 48 85,4 85,4 53 38 28 39,7 opposition right
Kenny 10,6 parliamentary 76,8 60 76,8 22 22 56 33,3 executive center-right
Madero 10,3 presidential 41 82,9 82,9 46 42 21 36,3 opposition center-right
Sanchez 8 parliamentary 67,2 56 67,2 10 8 29 15,7 opposition center-left
Gysi 7,5 parliamentary 83 36 83 48 51 72 57 opposition left
Fillon 7,4 semi-presid. 79,6 49 79,6 17 24 52 31 opposition center-right
GabrielG 7,1 parliamentary 83 36 83 48 51 72 57 executive center-left
BerlusconiG 6,1 parliamentary 58,4 65 65 17 14 27 19,3 executive center-right
Samaras 6 parliamentary 53 67 67 16 16 20 17,3 executive center-right
Monti 5,8 parliamentary 58,4 65 65 17 14 27 19,3 executive center-right
Rubalcaba 5,5 parliamentary 67,2 56 67,2 10 8 29 15,7 opposition center-left
BerlusconiO 5,2 parliamentary 58,4 65 65 17 14 27 19,3 opposition center-right
Bersani 5 parliamentary 58,4 65 65 17 14 27 19,3 opposition center-left
Le Pen 3,6 semi-presid. 79,6 49 79,6 17 24 52 31 opposition right
Rajoy 3,5 parliamentary 67,2 56 67,2 10 8 29 15,7 executive center-right
GabrielO 2,9 parliamentary 83 36 83 48 51 72 57 opposition center-left
Letta 2,3 parliamentary 58,4 65 65 17 14 27 19,3 executive center-left
Tsipras 2,3 parliamentary 53 67 67 16 16 20 17,3 opposition left
Cope 1,9 semi-presid. 79,6 49 79,6 17 24 52 31 opposition center-right
Epifani 0,8 parliamentary 58,4 65 65 17 14 27 19,3 opposition center-left
Notes. Sources for the Internet and Facebook penetration: itu.int; internetworldstats.com; latinobarometro.org. Sources for the trust in political institu-
tions: pewresearch.org; latinobarometro.org; europa.eu; aph.gov.au; gov.uk.

32
 International Journal of E-Politics
Volume 10 • Issue 1 • January-June 2019

APPENDIX 3

The calibration table highlights the type of calibration for each condition and the three qualitative
anchors when needed.
This comparative research could not have relied on former calibrations when a direct method of
calibration has been applied (conditions “high digitalization of the media system” and “high level of
trust in political institutions”; outcome “pop communication on Facebook”). Therefore, the principles
of calibration given by Schneider and Wagemann (2013, p. 32) were followed as rigorously as possible.

Presence of a Presidential System

Each case can belong or not belong to the set “presence of a presidential system”. On the one hand,
each leader operating in a presidential system is associated with a set membership score of 1. On the
other hand, each leader not operating in a presidential system is associated with a set membership
score of 0.
Regarding the French case, the following choice has been made: only heads of state or French
leaders who have participated as candidates in a presidential election belong to the set “presence of
a presidential system”.

High Digitalization of the Media System

The second condition was calibrated with a direct calibration method. This condition is a higher-order
construct derived from merging two indicators: i) Internet penetration and ii) the ratio of Facebook
penetration to Internet penetration. The column in Appendix 2 considered for the calibration process
is “% Internet OR % Facebook/Internet”. This column presents a higher-order construct because it
expresses the maximum value among the Internet penetration and the ratio of Facebook penetration
to Internet penetration.
The 0.5 qualitative anchor has been given a 79.9% value, which means that either an Internet
penetration or a ratio between the Facebook penetration and the Internet penetration higher than 79.9%
must be determined in order to operate in a country with a high digitalization of the media system.
The 1 qualitative anchor corresponds to 90%, while the 0 qualitative anchor corresponds to 66%.
On the one hand, at or above the 90% threshold, a case is fully in the “high digitalization of the media
system” set. On the other hand, below or equal to the 66.5% threshold, a case is fully out of the “high
digitalization of the media system” set.

High level of Trust in Political Institutions

The third condition has also been calibrated with a direct method of calibration. The column in
Appendix 2 that has been considered for the calibration process is “% trust in political institutions”. The
percentage values in this column represent the mean values of three indicators: i) trust in government;
ii) trust in parliament; and iii) satisfaction with the functioning of democracy.
The 0.5 qualitative anchor has been placed at a value of 43.4%, which means that a mean value
higher than 43.4% must be overcome in order to operate in a country with a high level of trust in
political institutions. As highlighted by the calibrations table, this percentage corresponds to the
mean of the values of a hypothetical 0.5 anchor for each of the three indicators of trust in political
institutions. In accordance with satisfaction with the functioning of democracy, a higher value was
considered. Detached from the ideological positioning of a specific government or the composition
of the parliament, satisfaction with the functioning of democracy needs to reach an absolute majority
in order to overcome the 0.5 qualitative anchor.
The qualitative anchors 1 and 0 have been positioned at 55.5% and 33.5%, following the same logic.

33
International Journal of E-Politics
Volume 10 • Issue 1 • January-June 2019

Table 6. Calibrations

TYPE OF
CONDITION 0.5 ANCHOR 1 ANCHOR 0 ANCHOR
CALIBRATION
Presence of a presidential Qualitative
system (1; 0)
High digitalization of the
Direct 79.9 90 66.5
media system
43.4 55.5 35.5
High level of trust in political
Direct [(40 + 40 + [(50 + 50 + [(33.3 + 33.3 +
institutions
50)/3] 66.6)/3] 40)/3]
Qualitative
Governing leader
(1;0)
Qualitative
Conservative leader
(1; 0.66; 0.33; 0)
Pop communication on
Direct 13.2% 23.3% 3.3%
Facebook

Governing Leader
Regarding this fourth condition, each case can belong or not belong to the “governing leader” set.
On the one hand, each leader with an executive role (a head of state, a prime minister or a simple
minister) is associated with a set membership score of 1. On the other hand, each leader without an
executive role is associated with a set membership score of 0.

Conservative Leader
For the last condition, a four-value fuzzy scale was used, which means that each case can be i) fully
in the “conservative leader” set (right-wing leader, set membership score of 1); ii) more in than
out (centre-right leader, set membership score of 0.66); ii) more out than in (centre-left leader, set
membership score of 0.33); or fully out (left-wing leader, set membership score of 0).

Pop Communication on Facebook

Communication on Facebook is considered pop when the percentage of pop posts overcomes the
13.2% threshold. Communication is fully pop on Facebook from the 23.3% threshold (qualitative
anchor of 1), while it is fully non-pop below the 3.3% threshold (qualitative anchor of 0).
A temporal criterion has been applied to identify these qualitative thresholds. The 0.5 qualitative
anchor corresponds to about 4 posts a month (4/30 = 0.133). A communication is pop when a leader
publishes approximately 1 post per week on average. It is fully pop (qualitative anchor of 1) when a
leader disseminates at least 7 pop posts a month on Facebook (7/30 = 0.233). Finally, a communication
is fully non-pop on Facebook when a leader publishes less than 1 post a month (1/30 = 0.033).

34