Content uploaded by Jukka Huhtamäki
Author content
All content in this area was uploaded by Jukka Huhtamäki on Oct 11, 2016
Content may be subject to copyright.
Media and Communication, 2016, Volume 4, Issue 4, Pages 97-108 97
Media and Communication (ISSN: 2183-2439)
2016, Volume 4, Issue 4, Pages 97-108
doi: 10.17645/mac.v4i4.593
Article
#JeSuisCharlie: Towards a Multi-Method Study of Hybrid Media Events
Johanna Sumiala 1, Minttu Tikka 1,*, Jukka Huhtamäki 2 and Katja Valaskivi 1
1 School of Communication, Media and Theatre, University of Tampere, 33014 Tampere, Finland; E-Mails:
johanna.sumiala@helsinki.fi (J.S.), minttu.tikka@uta.fi (M.T.), katja.valaskivi@uta.fi (K.V.)
2 Intelligent Information Systems Laboratory, Tampere University of Technology, 33720 Tampere, Finland; E-Mail:
jukka.huhtamaki@tut.fi
* Corresponding author
Submitted: 9 February 2016 | Accepted: 6 May 2016 | Published: 10 October 2016
Abstract
This article suggests a new methodological model for the study of hybrid media events with global appeal. This model,
developed in the project on the 2015 Charlie Hebdo attacks in Paris, was created specifically for researching digital
media—and in particular, Twitter. The article is structured as follows. Firstly, the methodological scope is discussed
against the theoretical context, e.g. the theory of media events. In the theoretical discussion, special emphasis is given
to i) disruptive, upsetting, or disintegrative media events and hybrid media events and ii) the conditions of today’s
heterogeneous and globalised media communication landscape. Secondly, the article introduces a multi-method
approach developed for the analysis of hybrid media events. In this model, computational social science—namely,
automated content analysis (ACA) and social network analytics (SNA)—are combined with a qualitative approach—
specifically, digital ethnography. The article outlines three key phases for research in which the interplay between
quantitative and qualitative approaches is played out. In the first phase, preliminary digital ethnography is applied to
provide the outline of the event. In the second phase, quantitative social network analytics are applied to construct the
digital field for research. In this phase, it is necessary to map a) what is circulating on the websites and b) where this
circulation takes place. The third and final phase applies a qualitative approach and digital ethnography to provide a
more nuanced, in-depth interpretation of what (substance/content) is circulating and how this material connects with
the ‘where’ in the digital landscape, hence constituting links and connections in the hybrid media landscape. In
conclusion, the article reflects on how this multi-method approach contributes to understanding the workings of
today’s hybrid media events: how they create and maintain symbolic battles over certain imagined constructs of social
imaginaries of solidarity, belonging, contestation, and exclusion, a topic of core value for the theory of media events.
Keywords
automated content analysis; Charlie Hebdo; digital ethnography; hybrid media event; social network analytics; Twitter
Issue
This article is part of the issue “Successes and Failures in Studying Social Media: Issues of Methods and Ethics”, edited
by Epp Lauk and Niina Sormanen (University of Jyväskylä, Finland).
© 2016 by the authors; licensee Cogitatio (Lisbon, Portugal). This article is licensed under a Creative Commons Attribu-
tion 4.0 International License (CC BY).
1. Introduction: Charlie Hebdo 2015, a Hybrid Media
Event
On Wednesday, 7 January 2015, at 11:30 a.m., French–
Algerian brothers Saïd and Chérif Kouachi attacked the
headquarters of Charlie Hebdo. Eleven people were
killed in the rampage. After the attack, the Kouachi
brothers returned to their car and exchanged fire with
the police officers blocking their escape route. A few
minutes later, they executed an injured police officer
named Ahmed Merabet at point-blank range. The per-
petrators escaped from the building, and the shooting
of the police officer was filmed from a nearby apart-
ment. The event instantly exploded into a transnational
media event, and the amateur video material that was
filmed began to circulate rapidly. Newsrooms all over
Media and Communication, 2016, Volume 4, Issue 4, Pages 97-108 98
the world followed the massive security operation as
the Kouachi brothers hijacked another car and fled
north out of Paris. In the evening, tens of thousands of
people took to the streets around Europe to show their
solidarity with those killed by the gunmen. The next
day, 8 January 2015, the attackers continued their es-
cape, and thousands of security personnel were de-
ployed to comb the area approximately 90 kilometres
from Paris, where the two men were last seen. Mean-
while, in Paris, reports emerged that a policewoman
had been shot and killed; however, the link with the
Charlie Hebdo attack was not immediately apparent. As
night fell, the Eiffel Tower’s lights were switched off in
memory of the victims. On Friday, 9 January 2015, the
police located the attackers in the Dammartin-en-
Goële area. The brothers were chased to an industrial
complex 35 kilometres from Paris, where they seized a
printworks and took a hostage. In east Paris, at around
12:30 p.m., a third gunman named Amedy Coulibaly
seized a Jewish supermarket, killed four people, and
took hostages. It emerged that Coulibaly was responsi-
ble for the killing of the Parisian policewoman, Clarissa
Jean-Philippe, the day before. In his phone call to the
French TV station BFM-TV, Coulibaly stated that his at-
tack was synchronised with the attacks of the Kouachi
brothers, and that they belonged to the same group of
terrorists. He also threatened to kill his hostages unless
the Kouachi brothers were allowed to go free. After
several hours of this hostage situation, police special
forces stormed the market and killed Coulibaly. The
Kouachi brothers were killed by the special forces on
the same day.
Over the course of these three days, new updates
constantly appeared on websites, on YouTube, and on
news broadcasts. Social media websites were inundated
with comments, links, and images connected to the
event, and these were shared and commented on by
both journalists and ordinary citizens. The course of the
events, as presented by professional journalists and in-
ternational and national media houses, was intermixed
with memes and comments that citizens from different
countries shared via social media. In addition, various
strategic and spontaneous (both political and religious)
interest groups made use of the situation and competed
for attention, tailoring and recycling details about the
events with content aimed at different audiences.
One of the prominent features of the Charlie Hebdo
event was the use of the slogan Je Suis Charlie (“I Am
Charlie”), which became a symbol of solidarity and
freedom of expression. The volume of communication
around the event is well illustrated by the fact that the
hashtag #JeSuisCharlie was—at least at the time of the
event—the most popular tweet in the history of Twit-
ter. The tag was tweeted 6,500 times per minute at its
height and was featured in 3.4 million tweets in one 24-
hour period (Whitehead, 2015). In addition to #JeSuis-
Charlie, there were many other expressions articulated
and shared via Twitter. The slogan Je ne suis pas Char-
lie (“I am not Charlie”) came to represent myriads of
opinions opposed to or critical of the mass Je Suis Char-
lie declaration. Another perspective was highlighted by
the slogan Je Suis Ahmed (“I am Ahmed”), which re-
ferred to the French police officer Ahmed Merabet,
who was Muslim and who was shot on the street by
terrorists shouting, “Allahu Akbar” and “We have
avenged the prophet”. The slogan Je Suis Ahmed
brought forth the perspective of French Muslims, who
opposed the association between Islam and terrorism,
as victims of the terror attack. The slogans were used
on social media, in the news, and in demonstrations,
and they were also circulated in images and caricature
drawings emblematic of the case of Charlie Hebdo.
This brief illustration of the media workings in the
Charlie Hebdo attacks is given here to demonstrate the
hybrid nature of communication around the events
and how an event can be transformed into a hybrid
media event (Vaccari, Chadwick, & O’Loughlin, 2015).
The term hybrid refers to a complex intermedia dy-
namic between mainstream news media and social
media, as well as the complex circulations between
messages and actors and the recombination of media
on a variety of media platforms (Chadwick, 2013;
Kraidy, 2002). Vaccari, Chadwick and O’Loughlin (2015,
p. 1044) describe hybrid media events as “media
events whose significance for media professionals, pol-
iticians, and non-elites is being reconfigured by the
growth of social media”. When thinking about the
Charlie Hebdo attacks as a hybrid media event, we may
approach it as a constellation of fluid social intensifica-
tions that are most typically created in a complex net-
work of Internet-based and mobile communication
technologies. The Charlie Hebdo attacks comprise ele-
ments of ceremonial mass media communication, but
these also converge with contemporary forms of ver-
nacular mass self-communication (cf. Castells, 2009),
occasionally also thought of as a form of citizen jour-
nalism (cf. Allan & Thorsen, 2009). The element of
“liveness” in the Charlie Hebdo attacks as a hybrid me-
dia event is intensified in the real-time circulation of
texts and images and the dispersion of the event in
several locations simultaneously. The level of connec-
tivity between the official and viral narratives of the
event may vary greatly, depending on the nature of a
message in circulation. Hence, the concept of the
“whole world” watching Paris needs to be analysed as
an experience that is scattered onto a multiplicity of
screens. While people may be sharing Charlie Hebdo as
a collective spectacle—to use Kellner’s (2003) termi-
nology—they are connected to it in different ways.
That is, they use different communication media to fol-
low the event, associate with different—and even con-
flicted—narratives circulating on the event, and feel
connected with different groups and identities involved
in the event. Consequently, a multiplicity of shared ex-
Media and Communication, 2016, Volume 4, Issue 4, Pages 97-108 99
periences is created in this hybrid media event. Thus,
the question of power embedded in social integration
as underlined in the classic theory of media events by
Dayan and Katz (1992) needs to be addressed on sev-
eral levels, including a variety of hybrid constellations
of sociality (cf. Sumiala & Korpiola, 2016).
In the following sections, we suggest a multi-
method approach for the empirical study of hybrid
media events, using the Charlie Hebdo attacks as a case
study. To meet this goal, we first provide a brief outline
for our theoretical framework—the theory of media
event—which is necessary to contextualise the meth-
odological model. Secondly, we introduce a multi-
method approach developed for the analysis of hybrid
media events. In this approach, computational social
science—or more specifically, a combination of auto-
mated content analysis (ACA) (Boumans & Trilling,
2016) and computational social network analytics
(SNA) (Huhtamäki, Russell, Rubens, & Still, 2015)—is
used in concert with a qualitative approach—
specifically, digital ethnography. The article outlines
three key phases for research in which the interplay
between quantitative and qualitative approaches is
played out. In the first phase a preliminary digital eth-
nography is applied to provide an initial sketch of the
event. In the second phase, quantitative social network
analytics is applied to construct the digital field for re-
search. In this phase, it is necessary to map a) what is
circulating on the websites and b) where this circula-
tion takes place. In the third and final phase, a combi-
nation of the qualitative approach and digital
ethnography is applied to provide a more nuanced, in-
depth interpretation of what (substance/content) is
circulating and how this material connects with the
‘where’ in the digital landscape, hence constituting
links and connections in the hybrid media landscape. In
conclusion, the article reflects on how this multi-
method approach contributes to the understanding of
the workings of today’s hybrid media events—how
they create and maintain symbolic battles over certain
social imaginaries of solidarity, belonging, contestation,
and exclusion. This is a topic of core value for the theo-
ry of media events.
2 Theoretical Framework: Re-Thinking Media Events
Since the birth of the modern mass media, many soci-
ologists, cultural theorists, and communication schol-
ars have examined the interplay between modern
society and mass-media saturated gatherings (Bennet
& Segerberg, 2012; Boorstin, 1973; Debord, 1967;
Kellner, 2003; Rojek, 2013; Shils & Young, 1956). A key
focal point in creating this tradition of thought in media
studies is Media events: The live broadcasting of histo-
ry, published by Daniel Dayan and Elihu Katz (1992).
According to Dayan and Katz, a media event is a special
genre that is powerful enough to interrupt everyday
media flow, bring the viewer into touch with society’s
central values, and invite the audience to participate in
the event (Dayan & Katz, 1992, pp. 5-9). In their lexi-
con, media events have their own grammar, their own
meaning structure (story form or script), and their own
practices, which are characterised by live broadcasting:
the interruption of daily media rhythms and routines,
the scripting and advance preparation of the event, a
huge audience (the “whole world” is watching), social
and normative expectations attached to viewing
(“must see”), the ceremonial tone of media narration,
and the intention to connect people.
As the story forms, media events can be divided in-
to “conquests”, “contests”, and “coronations”. Accord-
ing to Dayan and Katz (1992), these scripts constitute
(i) the main narrative possibilities within the genre, (ii)
the distribution of roles, (iii) and the ways in which
these roles are enacted. In many cases, the three story
forms are closely intertwined, and historical events
correspond to and resonate with each other at differ-
ent levels. One event may have certain features of
each form; the form of an event may also change,
transforming into another story form as the event de-
velops. It is also important to acknowledge that all
these scripts are embedded in deeper meaning struc-
tures in any given culture (Dayan & Katz, 1992, pp. 28-
29). The common denominator for Dayan and Katz’s
(1992) original work is the ceremoniality associated
with media performance. The authors indicate that the
significance of media events is in their ability to reach a
larger audience than any event that requires physical
presence. The audience itself is well aware of this, as
they follow the unfolding media event in different loca-
tions, which may be private, semi-public, or public.
Since its publication in the 1990s, the media events
theory has stimulated vigorous scholarly debate, with
its value believed to be in its theoretical and methodo-
logical innovation (Cottle, 2006; Couldry, 2003; Dayan,
2010; Fiske, 1994; Hepp & Couldry, 2010; Hepp &
Krotz, 2008; Katz & Liebes, 2007; Kyriakidou, 2008;
Liebes, 1998; Nossek, 2008; Roel, 2009; Rothenbuhler,
1998; Scannell, 1995, 2001; Sumiala, 2013). The main
criticisms of Dayan and Katz’s approach have ad-
dressed (i) the assumed ceremonial and integrative
functions of media events, (ii) the attempt to exclude
any disruptive or traumatic events from the focus of
their theory, and (iii) the strong focus on television and
broadcasting, which may result in inadequate study of
global web-based media events.
In other words, many argue that Dayan and Katz’s
initial account of media events assumes too straight-
forward a relationship between media coverage and
audience endorsement, thereby obscuring the ideolog-
ical construction of social order, as well as the chal-
lenges and disruptive potential that are implicit in
many media events (Cottle, 2006; Couldry, 2003; Fiske,
1994; Kellner, 2003; Kyriakdou, 2008; Rothenbuhler,
Media and Communication, 2016, Volume 4, Issue 4, Pages 97-108 100
2010; Scanell, 1995, 2001). In addition, given the glob-
alisation of communication through the Internet and
social networking websites, critics have called for a re-
contextualisation of the explicit focus on TV and broad-
casting in the media events theory.
Hepp and Couldry (2010, p. 9) argue that in theoris-
ing media events today we should not perceive them
as placed at a defined locality, but rather as disembed-
ded, or even ubiquitous, communicative practices.
Drawing from the work of Hepp and Couldry, we postu-
late that today’s media events should be understood as
multi-sited, multi-temporal, multi-actor and multi-
voiced phenomena articulated by a simultaneous con-
nectivity of a variety of communication processes.
These media events may be simultaneously structured
around relatively centralised power structures, such as
national and global mainstream media—for example
the BBC or CNN—and multi-centred power structures,
such as social networking sites (Hepp & Couldry, 2010,
p. 9). Hepp and Couldry (2010, p. 12) offer a new work-
ing definition for contemporary media events to better
grasp their fluid nature:
“Media events are certain situated, thickened, cen-
tring performances of mediated communication
that are focused on a specific thematic core, cross
different media products and reach a wide and di-
verse multiplicity of audiences and participants.”
Dayan and Katz have responded to the criticism of
their original theory of media events and have re-
adjusted their ideas in different public forums. Katz and
Liebes (2007, 2010) suggest that the focus of analysis
should be shifted from conquests, contests, and coro-
nations to disaster, terror, and war. According to Katz
and Liebes (2007, p. 157):
“We believe that cynicism, disenchantment, and
segregation are undermining attention to ceremo-
nial events, while the mobility and ubiquity of tele-
vision technology, together with the downgrading
of scheduled programming, provide ready access to
disruption. If ceremonial events may be character-
ized as ‘co-productions’ of broadcasters and estab-
lishments, then disruptive events may be
characterized as ‘co-productions’ of broadcasters
and anti-establishment agencies, i.e. the perpetra-
tors of disruption.”
Furthermore, Katz and Liebes suggest that mara-
thons of terror, natural disaster, and war—media dis-
asters—should be distinguished from media events as
a separate genre. These mediatized disasters of differ-
ent kinds have become far removed from the ceremo-
nial roots of the original media events (Cottle, 2006;
Liebes, 1997; Liebes & Blonheim, 2005). Daniel Dayan
(2010) has written extensively about the changing na-
ture of media events. For him, the “macabre accou-
trements to televised ordeals, punishments, and tor-
tures” and the emphasis on “stigmatization and
shaming” in today’s mediatized public events have
caused media events to lose their potential to reduce
conflict; instead, they ‘foster divides, and install and
perpetuate schisms’ (Dayan, 2010, pp. 26-27). As a re-
sult, media events tend to lose their distinct character
and instead migrate towards other genres: new media
events are no longer clearly differentiated entities, but
exist on a continuum. Dayan (2010, p. 27) suggests this
‘banalization of the format’ produces what he calls
“almost” media events. Dayan reminds us that the
pragmatics of media events have changed as messages
have become multiple, audiences selective, and social
networks ubiquitous. Dayan (2010, p. 27) summarises
the difference between original and current media
events in the following manner:
“Interpersonal networks and diffusion processes
are active before and after the event, mobilizing at-
tention to the event and fostering intensive herme-
neutic attempts to identify its meaning. But during
the liminal moments we described in 1992, totality
and simultaneity were unbound; organizers and
broadcasters resonated together; competing chan-
nels merged into one; viewers gathered at the same
time and in every place. All eyes were fixed on the
ceremonial centre, through which each nuclear cell
was connected to all the rest.”
Dayan leaves the reader in a state of scepticism. For
him, in today’s “contested territory of media events”,
disenchantment and the loss of the “we”—the most
critical functions of media events—are the most likely
consequences. Although it is reasonable to ask wheth-
er this “we” ever existed, it is nevertheless inevitable
that the dimensions of media events have changed
with the changing media environment and the con-
temporary multiplicity of the media.
The concept of hybrid media events is one attempt
to respond to the criticism offered by Hepp and Could-
ry (2010) and the response offered by Dayan (2010)
and Katz and Liebes (2010). The idea of hybrid media
events acknowledges the situated nature of transna-
tionally or even globally mediated communication of a
certain thematic core (here, the killings and related
public reactions), while underscoring the fluidity of the
movement in the circulation of the related posts,
memes, images, news, and reports. The concept of the
hybrid media event highlights the complex intermedia
dynamics between the different media platforms
(namely, mainstream news media and social media) in
communicating those solidarities, belongings, and con-
troversies associated with the event.
One of the key challenges for the study of contem-
porary media events is a methodological one. As the
Media and Communication, 2016, Volume 4, Issue 4, Pages 97-108 101
media landscape changes and media events become
more transnational and global, the right methodologi-
cal tools need to be developed to better grasp these
changing conditions. During the 1980s and 1990s in the
empirical study of media events, the methodological
focus was mainly on qualitative research. The empirical
analysis focused on the study of national broadcast
media, such as the BBC, or the national press, and the
focus was on observation, textual analysis, and inter-
views related to the production, representation, and
reception of media events (cf. Couldry, 2003; Couldry,
Hepp, & Krotz, 2010; Eide, Kunelius, & Phillips, 2008).
While dividing the empirical focus between the
production, representation, and reception of media
events has proven a useful strategy for understanding
national media events, this approach lacks the tools to
analyse those communicative processes that go beyond
the national frame and take place between and betwixt
production, representation, and reception of media
events. In these new conditions, messages, tweets,
posts, memes, images, and symbols circulate and travel
from one context to another. The categories between
production, representation, and reception become
blurred. It takes only one click to transform the person
receiving a message into the one who produces it. As a
result, new methodological approaches and tools need
to be developed to capture these processes of commu-
nication that are crucial for today’s hybrid media events.
This suggests a new type of methodological dialogue be-
tween qualitative and quantitative approaches.
Here, the quantitative methods that make it possi-
ble to deal with a large amount of data circulating on a
variety of media platforms are combined with more in-
depth qualitative methods, such as digital ethnogra-
phy, that enable researchers to go deeper into the data
and trace pieces of meaning associated with symbolic
battles carried out in the process of communicating
about the events. In the following section, we intro-
duce our methodological model for the study of hybrid
media events with global appeal. This model, devel-
oped in the project on the 2015 Charlie Hebdo attacks
in Paris, was created primarily for researching digital
media—specifically Twitter.
3. Studying Hybrid Media Events on Twitter
Twitter is a micro-blogging website created in 2006
that enables users to send up to 140 character mes-
sages, commonly called “tweets”. According to Twit-
ter’s own statistics from December 2015, the service
has 320 million monthly active users (Twitter, 2016).
Although the user growth has stalled in 2016—as Twit-
ter is having its tenth birthday—it is still among the
most popular social networking sites, along with Face-
book and Instagram (Statista, 2016). On Twitter, mes-
sages are public by default, although the service also
offers a feature called direct message (DM), which is
private. On Twitter, the model of social relationships is
directed and non-reciprocal, meaning that users can
subscribe to other users’ tweets in order to follow
them. However, those they follow don’t have to follow
them back. When a user follows other users, the
tweets of those followed will be visible on the user’s
main Twitter homepage, constituting a “tweet time-
line” that appears in reverse chronological order. The
characteristic practices for Twitter communication al-
low individual tweets to be liked and retweeted, which
can increase the visibility and popularity of a single
tweet. The retweet practice can also push a single
tweet into a circulation that crosses the borders of dif-
ferent media platforms. Users can make a reference to
other user with the @ symbol. With the prefix @ fol-
lowed by a username, users can mention or reply to
other users. An important feature is the hashtag—a
word or phrase prefixed with the # symbol. Hashtags
provide means for labelling tweets under certain top-
ics, which gives structure to the communication on
Twitter and enables users to find the information that
interests them. Additionally, Twitter allows users to
post images, videos, and hyperlinks.
As Twitter communication is limited to short mes-
sages that can be enriched with other communicative
elements, such as images, videos, and hyperlinks, it is
suitable for fast information sharing. Due to its public
nature, it is popular among journalists, authorities, and
organisations, as well as ordinary people. It is a promi-
nent platform in the context and construction of dif-
ferent types of media events, varying from sports and
politics to crises and disasters. A recent report on Twit-
ter states that typical content on Twitter is twofold: ei-
ther conversational, with thousands of people
engaging with a particular topic for an extended period
of time, or breaking news stories that drive large spikes
in traffic over shorter periods of time (Parse.ly, 2016).
Studies focusing on Twitter during political elections
and sport events such as the Olympics give emphasis to
idea of the audiences as co-producers of a media
event, in addition to the traditional mass media (cf.
Girginova, 2015; Kreiss, Meadows, & Remensperger,
2014). In the field of crisis communications, Twitter has
been at the centre of many discussions. From the Arab
Spring to the 2011 London riots, Twitter has been iden-
tified as a prominent platform for citizen communica-
tion in several revolutions, protests, and movements,
as it connects people and bypasses the gatekeepers,
whether they be the authorities or journalists (cf. Ben-
net & Segerberg, 2012; Procter, Vis, & Voss, 2013).
From the journalistic viewpoint of crisis reporting, the
2010 Haiti earthquake has been called the first “Twitter
disaster”. This title underlines the fact that during the
first 24 hours of the Haiti earthquake, news organisa-
tions were depending on social media, and especially
the rapid and easily accessible flow of information pro-
vided by Twitter (Bruno, 2011). In times of crisis, ordi-
Media and Communication, 2016, Volume 4, Issue 4, Pages 97-108 102
nary people can actively produce information, and they
can also link and share published news stories from
mainstream news media (Utz, Schultz, & Glocka, 2013;
Parse.ly, 2016). In this context, Twitter has been per-
ceived as a symbol of change in the media landscape:
“If we allow ourselves to paraphrase the CNN effect
of the 1990s, this changeover in the media land-
scape could be called the Twitter effect. As was true
for the CNN effect, which was caused by more than
just the CNN organization, the Twitter effect must
also be considered as a symbol of a much broader
phenomenon, concerning several online tools ori-
ented to the publication of user-generated, real-
time content (Twitter, Facebook, YouTube, etc.).”
(Bruno, 2011, p. 8)
For our research, Twitter offers a fruitful context for
the study of a hybrid media event. First of all, in con-
trast to Facebook, Twitter provides an Application Pro-
gramming Interface (API) that allows access to the
majority of the data published through the service (cf.
Vis, 2013). Secondly, although we fully acknowledge
that Twitter is only one platform in the hybrid media
system, we state that it played a key role as a promi-
nent platform during the unfolding of the Charlie Heb-
do attacks, through which information, images, videos,
and links about the incident were circulated. Thirdly,
Twitter has become a key platform for breaking news,
and therefore events that draw attention tend to sur-
face on the platform. Finally, Twitter offers rich data
that also sheds light on other forms of media. Several
media organisations, politicians, and authorities use
Twitter, and the content and actors from other media
platforms are also present through a hypermedia chain
(cf. Kraidy & Mourad, 2010). To give an example, a
tweet that contains an image taken of the TV screen
showing the news is a common convention that con-
structs a chain of different media.
4. Towards a Multi-Method Model
4.1. Automated Content Analysis and Computational
Social Network Analytics
In our multi-method model, we combine computation-
al social science—more specifically, automated content
analysis (ACA) (Boumans & Trilling, 2016) in concert
with computational social network analytics (SNA)—
with a qualitative approach—particularly, digital eth-
nography. The computational approach allows for ana-
lysing both what is being said and by whom. Moreover,
the individual actors can be connected to each other
through their interactions for richer context to content,
and this allows, for example, the identification of den-
sifications in interaction between actors. More specifi-
cally, methods of automated content analysis allow us
to identify the content that is circulating in the context
of the hybrid media event under investigation. Social
network analytics give us the means to investigate the
overall structure between the actors that discuss and
share content related to the event.
Here, the computational approach is used primarily
to support digital ethnographic investigations. In terms
of content analysis, the computational approach allows
us to identify the key topics that are discussed in the
data collected on the event. Four main approaches exist
for automated content analysis: counting and dictionary,
unsupervised learning, semi-supervised learning, and
supervised learning (Boumans & Trilling, 2016; Laak-
sonen, Nelimarkka, Tuokko, Marttila, & Kekkonen,
2015). In its simplest form, automated content analysis
is implemented by counting the number of times indi-
vidual keywords or, in Twitter’s case, hashtags and
usernames, are included in the data. Unsupervised
learning allows, for example, the creation of content-
based clusters from the data to identify topics and
their combinations or, in other words, to “identify po-
tentially significant fragments” (Procter et al., 2013). In
supervised learning, part of the data is categorised
manually, and this learning data is used to teach an al-
gorithm to categorise the rest of the material according
to its category. Examples of approaches for automated
content analysis include keyword extraction, topic
modelling, natural language processing (NLP), and enti-
ty recognition (Boumans & Trilling, 2016; Finkel, Gren-
ager, & Manning, 2005).
Compared to the situation that Procter et al. (2013)
faced when they started mining tweets and found that
there was very little existing infrastructure to support
them, the availability of tools supporting analysis has
improved over the last few years. Online services and
social media analysis platforms, including Pulsar and
others, provide investigators with dashboards that are
able to manage millions of tweets. Using such envi-
ronments for research is, however, far from trivial.
Transparency of data and analysis routines remains a
key issue. For ethnographic research, this limitation is
not as major, as the investigation is done first and
foremost on a qualitative basis, and therefore repre-
sentative sampling is not a major issue. It is, however,
important for the ethnographic research to understand
what, in fact, is “the field” where the research takes
place. This can be a problem when using commercial
analytics services, as, due to technical and business re-
strictions, it is not always possible to gain the necessary
information on how the data has been obtained.
The key approach into the analysis of structure that
emerges from the interaction between individual ac-
tors in the data is social network analysis (SNA). Here
we follow the insight of Yang and Leskovec (2014, p.
1892) as they maintain that, “networks provide a pow-
erful way to study complex systems of interacting ob-
jects”. SNA supports investigators in observing latent
Media and Communication, 2016, Volume 4, Issue 4, Pages 97-108 103
structures and patterns in source data and in sharing
their findings with others (Freeman, 2000). When ana-
lysing communication networks, actors are represent-
ed as network nodes and connected to each other
through interactions. Network analysis allows us to
quantify both structural properties of networks, as well
as the structural positions of individual actors. Moreo-
ver, cluster identification can be used to identify
groups of nodes that are interconnected to each other.
Network-level metrics come into play when indi-
vidual network representations are compared to each
other. Moreover, network metrics support the tem-
poral analysis of network structure. Size, connection
count, density, diameter, and average path length are
examples of metrics that can be used as indicators as
to which way a network under investigation is evolving.
In investigating hybrid media events, one can, for exam-
ple, create network representations of interactions that
are related to a particular topic (identified using auto-
mated content analysis) and use network-level metrics
to compare the properties of these topic networks.
Cluster identification is a particularly useful method
for supporting early exploration of communication da-
ta. Clusters emerge from the topology of the network
and challenge the investigators to make sense of why a
particular cluster emerges. To support the sense-
making process, the investigators can use the cluster
membership to volumes of hashtags and other topic
identifiers and therefore name or label the clusters ac-
cording to their content signature.
Node-level metrics can be used for a number of
purposes. Nodes with a high “betweenness” value, for
example, are likely to act as bridges or boundary span-
ners (Hansen, Shneiderman, & Smith, 2011, “bridge
scores for boundary spanners”) connecting the differ-
ent clusters of the overall network. Nodes with a high
“in-degree”, receive attention from the other actors.
Nodes with a high “out-degree” are active in producing
new content. Closeness centrality allows us to make a
distinction between nodes with peripheral position and
those close to the core of the network.
4.2. Digital Ethnography
Concepts such as digital ethnography, virtual ethnog-
raphy, web ethnography, netnography, mobile ethnog-
raphy, ICT ethnography, and virtual ethnography have
emerged to describe fieldwork conducted in digital en-
vironments and landscapes (Boellstorff, 2008; Hine,
2015; Kozinets, 2015; Wittel, 2000). Online access to
vast amounts of archived social interactions, along with
live access to the human beings posting, changes the
practice of ethnography. Researchers of the media are
not dealing merely with words, but with images, draw-
ings, photography, sound files, edited audiovisual
presentations, website creations, and other digital arti-
facts (Kozinets, 2015, p. 4).
A characteristic of this qualitative methodological
approach is that the researcher conducts fieldwork in
the digital environment and applies participatory ob-
servation as a means to analyse human–technology in-
teractions in the media and the social and cultural
implications this interaction has for the present day
digitalized life. In more practical terms, a digital eth-
nographer constructs his/her field by following ortrac-
ing the event, phenomenon, oractivity in question. The
fieldworker makes notes, keeps field diary, takes
screen shots, downloads material, and he or she may
also interview informants by meeting them face-to-
face or via digital communication media. It is not un-
typical that digital or online ethnography is combined
with offline ethnography (cf. Postill & Pink, 2012).
Ethnographic understanding of the digital environ-
ment and its related interactions aims at in-depth, ho-
listic, and situational understanding and knowledge of
the studied event, phenomenon, activity and people
(Hine, 2015, pp. 2-3). Considering the global, fluid, and
continuously changing nature of the digital landscape,
the issue of proximity and situational knowledge also
needs re-framing. As Hine (2015, pp. 3-4) argues:
“When we watch a fight break out on Twitter we
cannot be sure whether any of the followers of
those involved are seeing the same fight, at the
same time, and understanding it in the same way
that we do…The very notion of singular ‘situation’
as a pre-existing object breaks down when we look
closely…An ethnographer in such circumstances
must get used to a perpetual feeling of uncertainty,
of wondering what has been missed, and attempt-
ing to build interpretations of events based on
sketchy evidence.”
In digital ethnography, the researcher has to deal
with his/her limited human capacity to encompass the
whole of the situation. For this challenge, computa-
tional social science offers valuable tools to map the
digital landscape and provide a broader frame for the
communicative and social processes taking place in
that landscape. The value of ethnographic thick de-
scription and situational understanding lies in the
depth, detail and the ability to grasp more profound
layers of meaning in those actions and activities taking
place in Twitter and elsewhere in digital media. To fol-
low Hine (2015, p. 5):
“Ethnography is highly necessary for understanding
the Internet in all its depth and detail, and yet it can
be challenging to develop way of conducting eth-
nographic studies which both embrace all that me-
diated communication offers and still provide us
with robust, reliable insights into something in par-
ticular.”
Media and Communication, 2016, Volume 4, Issue 4, Pages 97-108 104
5. Three Phases
In the following, we will explain in more detail how the
computational social science methods, automated con-
tent analysis (ACA) and social network analytics (SNA),
can be combined with digital ethnography, and how
this methodological interplay contributes to developing
a new multi-method model for the study of such media
events. This method has three phases:
1) Digital ethnography provides the first outline of
the event;
2) Automated content analysis and social network
analytics construct the digital field for research;
3) Digital ethnography provides an in-depth inter-
pretation of what (substance/content) is circu-
lating and how this material connects with the
‘where’ in the digital landscape, hence consti-
tuting links and connections in the hybrid media
landscape necessary for the social meaning
making of the event.
5.1. Digital Ethnography Sketches the Event
Like traditional media events, hybrid global media
events interrupt the daily routines of the media and of
the everyday. In the case of disruptive events, not only
the mainstream news media, but also the social media
environment turns to a disaster mode and begins to
broadcast and circulate news, comments, tweets,
posts, and images on the events as they unfold. This
moment of massive media saturation and circulation of
information produces the first methodological chal-
lenge for the study of hybrid media events. This first
phase of chaotic information flow demands a digital
ethnographic scope—a perspective in which the events
are followed and structured into a timeline. In the case
of Charlie Hebdo attacks, we started our pilot study
immediately as the events unfolded. As digital ethnog-
raphers, we traced the news in the mainstream media,
such as the BBC, the New York Times, the Guardian and
Le Monde, as well as on Twitter, YouTube, and Face-
book. Our personal media streams also included na-
tional news outlets, as well as friends and family
members located in our native Finland and in different
parts of the world, reporting and commenting on the
events from different local perspectives. We identified
certain prominent messages, hashtags, posts, memes,
and images circulating in those media environments.
To give one example, the hashtag #JeSuisCharlie was
soon announced as the most-tweeted message in the
history of Twitter, offering a simple and interesting
lead to be followed in the course of the events. This
first ethnographic phase of the analysis is best de-
scribed as suggestive, and its findings may well be chal-
lenged in the later process of quantitative and
qualitative analysis. Yet, it is a necessary stage for the
process to follow, as it is this first stage of the project
in which the chaotic information flow around the
events is given its first suggestive sketch. This phase
provided insight to what might be interesting, relevant,
and peculiar in the events as they evolve and, thus, di-
rect the analysis in the next phase. As a concrete way
of gathering data, this phase results in many field
notes, screenshots, memes, images, videos, and links,
as well as a timeline of the events.
5.2. Using Automated Content Analysis and Social
Network Analytics to Map the Field
In the next phase, social network analytics are applied
to draw a more general overview of communication
around the events with more data. In the case of the
Charlie Hebdo attacks, the media platform analysed
was Twitter. In this so-called “helicopter stage” of the
analysis, social network analytics are used to construct
the research field and give an overview of the data as
well as map certain elements considered relevant
based on the first phase of the pilot study. Prior to the
analysis, the data needed to be collected. In this case,
it was acquired through the social media analytics plat-
form Pulsar using several search words
1
. The number
of hits for #JeSuisCharlie totalled 2.3 million.
At the second stage, it is important to make a dis-
tinction between what is circulating in Twitter and
where this circulation is taking place. In the Charlie
Hebdo attacks project, we began with the hashtag
#JeSuisCharlie and identified certain key groups: actors
including ordinary media users, professional media
houses; sites such as connected media platforms,
countries and connections associated with it - both
communication and non-communication between the
different virtual communities created around this par-
ticular hashtag. As a result, this mapping can be further
expanded to identify hashtags and actors that are re-
lated to #JeSuisCharlie. This mapping helps us to em-
pirically illustrate communicative networks created
around the events—where and when they take place
and how they exist in relation to each other.
Human-in-the-loop analysis is particularly im-
portant when ethnographic and computational meth-
ods and approaches are used together. Therefore, we
point to the Ostinato Model (Huhtamäki et al., 2015)
for a structured process for data-driven visual network
analytics that allows for balancing between exploration
and automation (i.e. reproducibility) of analysis. This
way, a multidisciplinary group of investigators can de-
velop the rich description of a hybrid media event in an
iterative and incremental fashion through a process that
resembles peeling an onion and, thus, to begin to quan-
1
The list of search words applied is the following: je suis
charlie, #jesuischarlie, je ne suis pas charlie, #jenesuispas
charlie, je suis ahmed, #jesuisahmed
Media and Communication, 2016, Volume 4, Issue 4, Pages 97-108 105
titatively identify what is circulating (which hashtags)
and in which digital media landscapes this circulation
takes place. Starting with Twitter helps us to follow the
circulation of certain hashtags and actors tweeting and
re-tweeting onto new networks of communication (for
example, Facebook, Instagram, or online media sites
such as Huffington Post have been identified).
However, it must be noted that while collecting da-
ta from Twitter is relatively straightforward, given that
the investigative team has the required technological
capabilities, hybrid media introduce a major issue into
data collection. The two public APIs that Twitter offers
for developers, response-request based REST API and
real-time streaming API, only allow data collection at
the time it is published on Twitter. REST API allows the
collection of limited amounts of data dating back to a
number of days, and the streaming API operates in re-
al-time by definition. The only way to collect extensive
data on Charlie Hebdo, for example, is to acquire (buy)
the data either directly from Twitter or through a social
media listening service such as Pulsar. The data ecosys-
tem has transformed since Procter et al. (2013) con-
ducted their research regarding the 2011 London riots.
Importantly, Twitter acquired Gnip
2
in 2014 that is cur-
rently the only company through which Twitter data
can be purchased.
5.3. Applying Digital Ethnography in Tracing the Social
In the final stage of the empirical analysis, networks
mapped by using quantitative analysis and social net-
work analytics and its visual illustrations are taken into
an ethnographic reconsideration. The quantitative
analysis draws a map of the field and helps to orientate
the ethnographic immersion. After choosing an inter-
esting incident within the larger event, this case is fol-
lowed in and through different media platforms. This
phase aims to develop a holistic understanding of the
chosen research object. Thus, the fieldwork in a digital
landscape integrally involves a dense description of the
observations in the form of field notes as well as doc-
umentation and recording of data by any means avail-
able, such as screenshots and prints (cf. Sumiala &
Tikka, 2013). In order to capture the research object in
a highly complex and dynamic landscape, it is useful to
go back to the timeline of the events and re-evaluate
the first sketch of the events against the quantitative
framework and, consequently, make necessary re-
orientations. In this phase, the researcher needs to re-
evaluate the incident’s relationship with the larger
event and the key nodal points in this process. This can
be carried out by searching for facts connected to the
events and identifying certain key elements such as
time, place, and people. In the digital landscape, this
2
https://blog.twitter.com/2014/twitter-welcomes-gnip-to-the-
flock
can be challenging as hybrid media events host and en-
tice myriads of interpretations, misunderstandings,
rumours as well as intentional misinformation. After
re-locating the basic elements in the event, the re-
searcher can begin to add layers of meanings to the
event. This can be done in two overlapping ways; it is
possible to conduct ethnographic fieldwork by follow-
ing paths and trails of links, streams, and algorithmic
suggestions offered by Twitter and other social media
platforms, but it may also be useful to conduct digital
ethnography by approaching the event simultaneously
from different directions for example by making
searches in search engines. In these overlapping pro-
cesses digital ethnographer develops a more nuanced
and in-depth understanding of the event and can begin
to make interpretations of those more or less visible
and hidden representations, discourses, actors and
symbols and related communicative practices that con-
tribute to creating and maintaining different types of
social imaginaries of solidarities, belongings, and exclu-
sions embedded with the events.
6. Conclusions
In this article, we suggest a new multi-model methodo-
logical approach to the study of hybrid media events de-
veloped for the study of the Charlie Hebdo attacks. In
this new condition of hybrid media, events, messages,
tweets, posts, memes, images, and symbols spread sim-
ultaneously and are constantly on the move. Old hierar-
chies between the centre and peripheries in the media
event need to be reconsidered as hybrid media events
appear more horizontal and multi-sited, multi-temporal,
multi-actor, and multi-voiced social phenomena.
As a result, new methodological approaches and
tools need to be developed to capture these processes
of communication and better understand the workings
of today’s hybrid media events. This multi-method
model proposed in the article consists of combining
quantitative automated content analysis (ACA) and so-
cial network analytics (SNA) with qualitative digital
ethnography. The key for the model is a close interplay
between the different approaches and their careful
adaption in the different phases of the research. This
offers a unique possibility to bridge the gap between
situational, in-depth knowledge achieved by qualitative
methods in the study of media events and their under-
standing in the more global communication context.
The theory of media events was first established to
explain the social dynamics activated as people gath-
ered together around their TV sets to watch national
rituals as live history to be performed on the screen. As
discussed earlier, later developments in this theory
have challenged the assumed social cohesion created
by these events and emphasised instead the disruptive
nature of media events. This has implied a certain con-
tested view on the issue of sociality. The hybrid charac-
Media and Communication, 2016, Volume 4, Issue 4, Pages 97-108 106
ter of contemporary global, disruptive media events
such as the Charlie Hebdo attacks makes the issue of
social integration even more complicated. The ques-
tion of establishment and anti-establishment is also
changed in the hybrid media system. A terrorist attack
is as carefully crafted and designed as a coronation, de-
spite aiming at disruption rather than integration. The
hybridity of the media environment causes a situation
where no individual actor is able to control the flows of
information, attention and effect. Despite this, hybrid
media events also represent and reproduce existing
social solidarities and antagonisms. Continuity and
change take place through the circulation of meanings.
Twitter is a particular environment for the circulation
to take place. Its specific properties contribute to a cul-
ture of circulation (Lee & LiPuma, 2002) that seems
complex and dispersed. This complexity calls for the
multi-method approach.
The question of social integration in media events is
not only a theoretical one. It is important to ask how
we should empirically study the social dynamics acti-
vated in hybrid media events. In this article we suggest
a methodological model that has potential to move
from one research scale to another. The wider scale
observations of the Charlie Hebdo attacks as a hybrid
media event suggest a multiplicity of social dynamics
were activated during the events. Hence, it suggests an
interpretation that emphasises the heterogeneity as
well as the ephemerality of those social dynamics. To
understand more profoundly what kind of meanings
and interpretations are associated with those messag-
es and actors circulating in the digital landscape, an
ethnographic perspective is necessary. In the future,
more empirical research is needed to grasp these com-
plex dynamics of social imaginaries of solidarity, be-
longing, but also exclusion. The multi-method
approach is one attempt to point to this direction.
Acknowledgements
This article is based on the research project Je Suis
Charlie—The symbolic battle and struggle over atten-
tion, conducted in the School of Communication, Me-
dia and Theatre in the University of Tampere. This two-
year-project (2015–2017) is funded by the Helsingin
Sanomat Foundation. The writers wish to thank the
anonymous reviewers for their insightful comments.
Conflicts of Interests
The authors declare no conflicts of interests.
References
Allan, S., & Thorsen, E. (Eds.) (2009). Citizen journalism:
Global perspectives. New York: Peter Lang.
Bennett, L. W., & Segerberg, A. (2012). The logic of con-
nective action. Information, Communication & Socie-
ty, 15(5), 739-768.
Boellstorff, T. (2008). Coming of age in second life. An
anthropologist explores the virtually human. Prince-
ton: Princeton University Press.
Boorstin, D. J. (1973). The image—A guide to pseudo-
events in America. New York: Atheneum.
Boumans, J. W., & Trilling, D. (2016). Taking stock of the
toolkit. Digital Journalism, 4(1), 8-23.
Bruno, N. (2011). Tweet first, verify later? How real-time
information is changing the coverage of worldwide
crisis events. Reuters Institute for the Study of Jour-
nalism. Oxford: University of Oxford.
Castells, M. (2009). Communication power. New York:
Oxford University Press.
Chadwick, A. (2013). The hybrid media system: Politics
and power. New York: Oxford University Press.
Cottle, S. (2006). Mediatized rituals: beyond manufactur-
ing consent. Media, Culture & Society, 28(3), 411-
432.
Couldry, N. (2003). Media rituals: A critical approach.
London: Routlegde.
Couldry, N., Hepp, A., & Krotz, F. (Eds.). (2010). Media
events in a global age. Routledge: Abingdon.
Dayan, D. (2010). Beyond media events. Disenchant-
ment, derailment, disruption. In N. Couldry, A. Hepp,
& F. Krotz (Eds.), Media events in a global age (pp.
23-42). London: Routledge.
Dayan, D., & Katz, E. (1992). Media events. The live
broadcasting of history. Cambridge, MA: Harvard
University Press.
Debord, G. (1967). Society of the spectacle. Buchet-
Chastel.
Eide, E., Kunelius, R., & Phillips, A. (Eds.), (2008). Trans-
national media events: The Mohammed cartoons
and the imagined clash of civilizations. Göteborg,
Sweden: Nordicom.
Finkel, J. R., Grenager, T., & Manning, C. (2005). Incorpo-
rating non-local information into information extrac-
tion systems by Gibbs sampling (pp. 363-370).
Stroudsburg, PA: Association for Computational Lin-
guistics.
Fiske, J. (1994). Media matters: Everyday culture and po-
litical change. Minneapolis: University of Minnesota
Press.
Freeman, L. C. (2000). Visualizing social networks. Jour-
nal of Social Structure, 1(1). Retrieved from http://
www.cmu.edu/joss/content/articles/volume1/Free
man.html
Girginova, K. (2015). New media, creativity, and the
Olympics: A case study into the use of #NBCFail dur-
ing the Sochi winter games. Communication & Sport.
doi:10.1177/2167479515576101
Hansen, D. L., Shneiderman, B., & Smith, M. A. (2011).
Analyzing social media networks with NodeXL: In-
sights from a connected world. Burlington, MA: Mor-
gan Kaufmann.
Media and Communication, 2016, Volume 4, Issue 4, Pages 97-108 107
Hepp, A., & Couldry, N. (2010). Introduction: Media
events in globalized media cultures. In N. Couldry, A.
Hepp, & F. Krotz (Eds.), Media events in a global age
(pp. 1-20). Abingdon: Routledge.
Hepp, A., & Krotz, F. (2008). Media events, globalization
and cultural change: An introduction to the special
issue. Communications. The Journal of European
Communication Research, 33(3), 265-272.
Hine, C. (2015). Ethnography for the internet: Embedded,
embodied and everyday. London: Bloomsbury.
Huhtamäki, J., Russell, M. G., Rubens, N., & Still, K.
(2015). Ostinato: The exploration-automation cycle
of user-centric, process-automated data-driven visu-
al network analytics. In S. A. Matei, M. G. Russell, &
E. Bertino (Eds.), Transparency in social media: Tools,
methods and algorithms for mediating online inter-
actions (pp. 197-222). Retrieved from http://link.
springer.com/chapter/10.1007/978-3-319-18552-1_
11
Katz, E., & Liebes, T. (2007). ‘No more peace!’ How disas-
ter, terror and war have upstaged media events. In-
ternational Journal of Communication, 1, 157-166.
Katz, E., & Liebes, T. (2010). 'No more peace!’ How disas-
ter, terror and war have upstaged media events. In
N. Couldry, A. Hepp, & F. Krotz (Eds.), Media events
in a global age (pp. 32-42). Abingdon: Routledge.
Kellner, D. (2003). Media spectacle. London: Routledge.
Kozinets, R. (2015). Netnography: Redefined. London:
Sage.
Kraidy, M. M. (2002). Hybridity in cultural globalization.
Communication Theory, 12(3), 316-339.
Kraidy, M., & Mourad, S. (2010). Hypermedia space and
global communication studies lessons from the Mid-
dle East. Retrieved from http://repository.upenn.
edu/cgi/viewcontent.cgi?article=1256&context=asc_
papers
Kreiss, D., Meadows, L., & Remensperger, J. (2014). Po-
litical performance, boundary spaces, and active
spectatorship: Media production at the 2012 Demo-
cratic National Convention. Journalism, 16(5), 677-
595.
Kyriakidou, M. (2008). Rethinking media events in the
context of a global public sphere: Exploring the audi-
ence of global disasters in Greece. Communications.
The Journal of European Communication Research,
33(3), 273-291.
Laaksonen, S.-M., Nelimarkka, M., Tuokko, M., Marttila,
M., & Kekkonen, A. (2015). Data augmented ethnog-
raphy: using big data and ethnography to explore
candidates’ digital interactions. In Proceedings of the
Empiricist’s Challenge: Asking Meaningful Questions
in the Age of Big Data Conference. Mannheim: Uni-
versity of Mannheim.
Lee, B., & LiPuma, E. (2002). Cultures of circulation: The
imaginations of modernity. Public Culture, 14(1), 191-
213.
Liebes, T. (1997). Reporting the Arab Israeli conflict: How
hegemony works. London: Routledge.
Liebes, T. (1998). Television’s disaster marathons. A dan-
ger for democratic processes? In T. Liebes & J. Curran
(Eds.), Media, ritual and identity, (pp. 71-84). Lon-
don: Routledge.
Liebes, T., & Blondheim, M. (2005). Myths to the rescue:
How live television intervenes in history. In E.W.
Rothenbuhler & M. Coman (Eds.), Media anthropol-
ogy, (pp. 188-198). California: Sage.
Nossek, H. (2008). ’News media’—Media events: Terror-
ist acts as media events. Communications. The Jour-
nal of European Communication Research, 33(3),
313-330.
Parse.ly (2016). Does Twitter matter to news. The au-
thority report. Parsely.com Retrieved from http://
learn.parsely.com/rs/314-EBB-255/images/authority
-report-10.pdf
Postill, J., & Pink, S. (2012). Social media ethnography.
The digital researcher in a messy web. Media Inter-
national Australia, 145, 123-134.
Procter, R., Vis, F., & Voss, A. (2013). Reading the riots
on Twitter: Methodological innovation for the analy-
sis of big data. International Journal of Social Re-
search Methodology, 16(3), 197-214.
Roel, P. (2009). Intense media coverage. Communica-
tions. The Journal of European Communication Re-
search, 34(1), 1-20.
Rojek, C. (2013). Event power: How global events man-
age and manipulate. London & Thousand Oaks: Sage.
Rothenbuhler, E. W. (1998). Ritual communication: From
everyday conversation to mediated ceremony. Thou-
sand Oaks: Sage.
Rothenbuhler, E. W. (2010). Media events in the age of
terrorism and Internet. The Romanian Review of
Journalism and Communication, IV(2), 34-41.
Scannell, P. (1995). Media events (review). Media, Cul-
ture & Society, 17(1), 151-157.
Scannell, P. (2001). Editorial. Media, Culture & Society,
23(6), 699-705.
Shils, E., & Young, M. (1956). The meaning of the coro-
nation. Sociological Review, 1(2), 63-82.
Statista. (2016). Leading social networks worldwide as of
April 2016, ranked by number of active users (in mil-
lions). Statista.com. Retrieved from http://www.sta
tista.com/statistics/272014/global-social-networks-
ranked-by-number-of-users
Sumiala, J. (2013). Media and ritual. Death, community
and everyday life. London: Routledge.
Sumiala, J., & Korpiola, L. (2016). Tahrir 2011: Contested
dynamics of a media event. In B. Mitu & S. Poulakid-
akos (Eds.), Media events: A critical contemporary
approach. London: Palgrave McMillan.
Sumiala, J., & Tikka, M. (2013). Broadcast yourself—
Global news! A netnography of the ‘flotilla’ news on
YouTube. Communication, Culture & Critique, 6(2),
201-352.
Twitter. (2016). Twitter usage/company facts. Twitter.
Media and Communication, 2016, Volume 4, Issue 4, Pages 97-108 108
Retrieved from https://about.twitter.com/company
Utz, S., Schultz, F., & Glocka, S. (2013). Crisis communica-
tion online: How medium, crisis type and emotions
affected public reactions in the Fukushima Daiichi
nuclear disaster. Public Relations Review, 39, 40-46.
Vaccari, C., Chadwick, A., & O’Loughlin, B. (2015). Dual
screening the political: Media events, social media,
and citizen engagement. Journal of Communication,
65(6), 1041-1061.
Vis, F. (2013). A critical reflection on big data: Consider-
ing APIs, researchers and tools as data makers. First
Monday, 18(10). doi:10.5210/fm.v18i10.4878
Whitehead, T. (2015, January 1) Paris Charlie Hebdo at-
tack: Je Suis Charlie hashtag one of most popular in
Twitter history. Telegraph. Retrieved from http://
www.telegraph.co.uk/news/worldnews/europe/fran
ce/11336879/Paris-Charlie-Hebdo-attack-Je-Suis-
Charlie-hashtag-one-of-most-popular-in-Twitter-
history.html
Wittel, A. (2000). Ethnography on the move: from field
to net to Internet. Forum Qualitative Sozialfor-
schung/Forum: Qualitative Social Research, 1(1).
Yang, J., & Leskovec, J. (2014). Overlapping communities
explain core-periphery organization of networks.
Proceedings of the IEEE, 102(12), 1892-1902.
About the Authors
Johanna Sumiala (University of Tampere) is Associate Professor (Docent) at the Department of Social
Research/Media and Communication Studies, University of Helsinki. She is an expert in the fields of
media sociology and media anthropology, virtual ethnography and visual culture. She has recently
published articles on mediatized violence and ritualized (online) communication and she is the author
of the book Media and Ritual: Death, Community and Everyday Life (Routledge, 2012).
Minttu Tikka (University of Tampere) is a PhD candidate at the Department of Social Research/Media
and Communication Studies, University of Helsinki. She has published issues of crisis and social me-
dia, YouTube and news and methods of studying internet.
Jukka Huhtamäki (Intelligent Information Systems Laboratory at Tampere University of Technology,
TUT) is a co-founder of Innovation Ecosystems Network. He is specialized in developing methods for
extracting, accessing and analyzing heterogeneous data for system-level insights on various kinds of
networked phenomena.
Katja Valaskivi (University of Tampere) is the Research Director of Tampere Research Centre for
Journalism, Media and Communication (COMET). She is a media scholar, who has published widely in
the issues of media and the nation, nation branding, media and social theory, and circulation in me-
dia society. The title of her most recent book is Cool Nations: Media and the Social Imaginary of the
Branded Country (Routledge 2016).
