What is Digital Journalism Studies?

What is Digital Journalism Studies? delves into the technologies, platforms, and
audience relations that constitute digital journalism studies’ central objects of
study, outlining its principal theories, the research methods being developed, its
normative underpinnings, and possible futures for the academic field.
The book argues that digital journalism studies is much more than the study
of journalism produced, distributed, and consumed with the aid of digital
technologies. Rather, the scholarly field of digital journalism studies is built on
questions that disrupt much of what previously was taken for granted concerning
media, journalism, and public spheres, asking questions like: What is a news
organisation? To what degree has news become separated from journalism? What
roles do platform companies and emerging technologies play in the production,
distribution, and consumption of news and journalism? The book reviews the
research into these questions and argues that digital journalism studies constitutes
a cross-disciplinary field that does not focus on journalism solely from the
traditions of journalism studies, but is open to research from and conversations
with related fields.
This is a timely overview of an increasingly prominent field of media studies
that will be of particular interest to academics, researchers, and students of
journalism and communication.

Steen Steensen is Professor of Journalism and former (2016–2020) Head of the

Department of Journalism and Media Studies at Oslo Metropolitan University.
He currently leads the international research project Source Criticism and Mediated
Disinformation (2020–2024). He is associate editor of Journalism Practice and has a
background as a journalist.

Oscar Westlund (PhD) is Professor in the Department of Journalism and Media

Studies at Oslo Metropolitan University, where he leads the OsloMet Digital
Journalism Research Group. He holds secondary appointments at Volda University
College and the University of Gothenburg. He is the editor-in-chief of Digital
Journalism. He leads The Epistemologies of Digital News Production research project
funded by the Swedish Foundation for Humanities and Social Sciences.
Disruptions: Studies in Digital Journalism
Series editor: Bob Franklin

Disruptions refers to the radical changes provoked by the affordances

of digital technologies that occur at a pace and on a scale that disrupts
settled understandings and traditional ways of creating value, interacting
and communicating both socially and professionally. The consequences
for digital journalism involve far reaching changes to business models,
professional practices, roles, ethics, products and even challenges to
the accepted definitions and understandings of journalism. For Digital
Journalism Studies, the field of academic inquiry which explores and
examines digital journalism, disruption results in paradigmatic and
tectonic shifts in scholarly concerns. It prompts reconsideration of
research methods, theoretical analyses and responses (oppositional and
consensual) to such changes, which have been described as being akin to
‘a moment of mind-blowing uncertainty’.
Routledge’s new book series, Disruptions: Studies in Digital Journalism,
seeks to capture, examine and analyse these moments of exciting and
explosive professional and scholarly innovation which characterize
developments in the day-to-day practice of journalism in an age of digital
media, and which are articulated in the newly emerging academic discipline
of Digital Journalism Studies.

User Comments and Moderation in Digital Journalism

Thomas B. Ksiazek and Nina Springer

Smartphones and the News

Andrew Duffy

What is Digital Journalism Studies?

Steen Steensen and Oscar Westlund

For more information, please visit: www.routledge.com/Disruptions/

book-series/DISRUPTDIGJOUR
What is Digital
Journalism Studies?

Steen Steensen and

Oscar Westlund
First published 2021
by Routledge
2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN
and by Routledge
52 Vanderbilt Avenue, New York, NY 10017
Routledge is an imprint of the Taylor & Francis Group, an informa business
© 2021 Steen Steensen and Oscar Westlund
The right of Steen Steensen and Oscar Westlund to be identified as
authors of this work has been asserted by them in accordance with
sections 77 and 78 of the Copyright, Designs and Patents Act 1988.
The Open Access version of this book, available at www.taylorfrancis.com,
has been made available under a Creative Commons Attribution-Non
Commercial-No Derivatives 4.0 license.
Trademark notice: Product or corporate names may be trademarks
or registered trademarks, and are used only for identification and
explanation without intent to infringe.
British Library Cataloguing-in-Publication Data
A catalogue record for this book is available from the British Library
Library of Congress Cataloging-in-Publication Data
Names: Steensen, Steen, (Professor of journalism) author. | Westlund,
Oscar, author.
Title: What is digital journalism studies? / Steen Steensen, Oscar
Westlund.
Description: London ; New York : Routledge, 2020. | Includes
bibliographical references and index.
Identifiers: LCCN 2020019445 | ISBN 9780367200909 (hardcover) |
ISBN 9780429259555 (ebook)
Subjects: LCSH: Online journalism—Research. | Online journalism—
Research—Methodology. | Digital media—Research.
Classification: LCC PN4784.O62 S73 2020 | DDC 070.1—dc23
LC record available at https://lccn.loc.gov/2020019445
ISBN: 978-0-367-20090-9 (hbk)
ISBN: 978-0-429-25955-5 (ebk)
Typeset in Bembo
by Apex CoVantage, LLC
Visit the eResources: www.routledge.com/9780367200909
Contents

List of figuresviii
List of tablesix
Forewordx

1 The introduction: the premises and principles

of digital journalism studies 1
1.1 Four structural premises for digital journalism studies  4
1.2 The separation of news from journalism  6
1.3 What does Digital Journalism studies look like?  8
1.3.1 The interdisciplinarity of digital journalism studies  9
1.3.2 Digital journalism studies and global diversity  11
1.4 Outline of the book  12

2 The definitions: current debates and a framework

for assessing digital journalism studies 15
2.1 Digital journalism studies: definitions and debates  15
2.2 An analytical framework: society, sector, and scholarship  19
2.2.1 Issue (in)visibility 21
2.2.2 Pro-innovation bias 22
2.2.3 Path dependency 23
2.2.4 Addressability 24
2.3 Turning to thematic clusters in Digital Journalism 25

3 The technologies: unpacking the dominant

object of study in Digital Journalism Studies 27
3.1 Data journalism 29
3.2 Analytics and metrics  30
vi  Contents
3.3 Algorithms and automation  34
3.4 Concluding discussion 38

4 The platforms: distributions and devices

in digital journalism 40
4.1 Digital journalism and platforms  41
4.1.1 Building platform presence  42
4.1.2 Platform counterbalancing 48
4.2 Digital journalism and digital devices  51
4.3 Discussions and conclusions  53

5 The theories: how digital journalism is understood 55

5.1 What is theory and why does it matter?  57
5.2 The multitude of theories in digital journalism studies  58
5.2.1 Digital journalism as a social system  59
5.2.2 Digital journalism as a socio-technical practice  62
5.2.3 Digital journalism as a democratic force  64
5.2.4 Digital journalism as post-industrial business
endeavour 66
5.2.5 Digital journalism as cultural production
and discourse  67
5.3 The theoretical blind spots of digital journalism
studies 68

6 The assumptions: the underlying normativity

of digital journalism studies 72
6.1 The normative future-predictions of digital journalism
studies 74
6.1.1 Digital journalism studies and the discourse
of crisis  76
6.1.2 Digital journalism studies and the discourse
of technological optimism  79
6.2 Digital journalism studies and the discourse
of innovation  81
6.2.1 The newness bias  82
6.2.2 The problems with change and how to deal
with them  84
6.3 Concluding remarks 85
 Contents vii
7 The methodologies: how digital journalism
is researched 87
7.1 Methods in Digital Journalism 89
7.2 Numbers, metrics, and computational methods  91
7.2.1 Advancing content analysis in digital journalism
studies 92
7.2.2 Computational methods and analysis of information
networks 93
7.2.3 Problems with big data computational methods  95
7.3 Digital ethnography 97
7.4 Audience research 99
7.5 Concluding remarks 100

8 The futures: deconstructions of and directions

for digital journalism studies 102
8.1 Digital journalism studies for or about the sector  104
8.2 Key takeaways: the formative formations of the field  106
8.3 Directions for digital journalism studies for the 2020s  109

Selected references117
Index120
Figures

1.1 Share of the most dominant disciplinary perspectives in

abstracts of articles published in Digital Journalism from
issue 1, volume 1 (2013) to issue 4, volume 7 (2019).
Every second abstract is analysed (N = 172). 10
1.2 First authors of all articles published in the journals
Digital Journalism, Journalism Studies, Journalism: TP&C,
Journalism & Mass Communication Quarterly, and
Journalism Practice during the years 2013–2019.
Definitions of the regions are based on the UN
Regional groups. 13
6.1 Google Scholar search on the search terms “journalism”
and “crisis” (excluding “crisis journalism”) and how the
result compares to a similar Google Scholar search on
just the word “journalism”. 76
6.2 Google Scholar search on the terms “journalism” and
“innovation” and how the result compares to a similar
Google Scholar search on just the word “journalism”. 82
Tables

1.1 2018 citation metrics and ranking within the discipline

of communication from SJR (SCImago Journal
Ranking), Scopus, and Google Scholar. The table
displays the five top journalism journals. 8
2.1 Definitions of digital journalism and digital journalism
studies discussed in a special issue of Digital Journalism.
The table was originally published in Eldridge II et al.
(2019, p. 392). 17
2.2 Thematic clusters of keywords used in the 343 original
articles published in Digital Journalism from volume 1,
issue 1 (2013) to volume 7, issue 4 (2019) that contained
keywords.26
7.1 Unique and clustered keywords in articles published
in Digital Journalism 2013–2019 belonging to the
Methodology thematic cluster. 89
Foreword

This book is intended for researchers, PhD students, and possibly also
post-graduate students interested in the emerging field of digital journal-
ism studies. The book would not have materialised without the aid of
many people, to whom we would like to extend our warmest gratitude.
First, we would like to thank series editor Bob Franklin for reaching out
to us with the idea for this book. Without his encouragement and enthu-
siasm the book would not have been written. Then we would like to
thank our employer, Oslo Metropolitan University, not only for allowing
us to spend time on this book, but also for granting funding for making
this book Open Access. We are truly excited about the fact that this book
can be accessed by anyone from everywhere without any costs other than
those related to having internet access and a screen to read on. We would
also like to thank the publisher, Routledge, for making this opportunity
available at a reasonable cost, and for all the work put into the production
of the book.
In the final stages of developing this book we have approached a hand-
ful of exceptionally qualified peers for feedback on one or several chap-
ters. Each chapter has benefited substantially from constructive feedback
on both bigger and smaller issues. In alphabetical order we would like to
extend our most sincere appreciation and thanks to Laura Ahva, Sherwin
Chua, Mark Deuze, Scott Eldridge II, Tine U. Figenschou, Alfred Her-
mida, Kristy Hess, Avery Holton, Karoline A. Ihlebæk, Maria Konow
Lund, Merja Myllylathi, Ragnhild K. Olsen, Chris Peters, Jane B. Singer,
Helle Sjøvaag, and Edson Tandoc Jr. We will forever be grateful for your
collegial support.
The book is written as a cooperative exercise between the two of us.
Even though all eight chapters are coauthored, we have divided the work
so that Steensen had the main responsibility for chapters 1, 5, 6, and 7
while Westlund did the heavy lifting in chapters 2, 3, 4, and 8. However,
all chapters have been revised by both authors in many rounds, so the
 Foreword xi
book is really the result of what we have experienced as a fruitful coop-
eration. Our final acknowledgement therefore goes to ourselves: Steen
would like to thank Oscar and Oscar would like to thank Steen. We have
enjoyed the experience of working with each other and integrating our
explicit knowledge about digital journalism and digital journalism studies
in coauthoring this book. It’s been a challenge, but it has been fun.
Oslo,
18-March 2020
1 The introduction
The premises and principles of digital
journalism studies

On 11 April 11 2018, Mark Zuckerberg, the Facebook CEO, sat before

the US Congress for a hearing following the Cambridge Analytica scan-
dal. He had already survived 10 hours of questioning the previous day.
The session chair, Republican congressman Greg Walden, leaned for-
ward in his large, black leather chair, his stare alternating between his
paperwork and Zuckerberg, who sat behind a long but modest desk,
several feet below him. Walden said:

Welcome, Mr.  Zuckerberg, to the Energy and Commerce Com-

mittee in the House. We’ve called you here today for two reasons.
One is to examine the alarming reports regarding breaches of trust
between your company, one of the biggest and most powerful in the
world, and its users. And the second reason is to widen our lens to
larger questions about the fundamental relationship tech companies
have with their users.

Walden then laid out in more detail the background for these two con-
cerns, before focusing on the questions he wished Zuckerberg to answer:

There are critical unanswered questions surrounding Facebook’s

business model and the entire digital ecosystem regarding online pri-
vacy and consumer protection. What exactly is Facebook? Social
platform? Data company? Advertising company? A media company?
A  common carrier in the information age? All of the above? Or
something else?

Zuckerberg was not allowed to answer, yet. He sat there quietly behind
his desk, occasionally sipping water out of a white paper cup, while look-
ing at Walden like a school boy paying attention to his teacher. It was not
until a couple of hours later, following a series of questions from other
2  The introduction
congress members, that Walden returned to the questions regarding what
kind of company Facebook actually is and asked Zuckerberg a direct
question: “Is Facebook a media company?”
Zuckerberg did not take his eyes off Walden and answered, with a
steady voice:

Thank you, Mr. Chairman. I consider us to be a technology company,

because the primary thing that we do is have engineers who write
code and build products and services for other people. There are cer-
tainly other things that we do, too. We – we do pay to help produce
content. We build enterprise software, although I don’t consider us an
enterprise software company. We build planes to help connect people,
and I don’t consider ourselves to be an aerospace company. But, over-
all, when people ask us if we’re a media company, what – what I hear
is, “Do we have a responsibility for the content that people share on
Facebook?” And I believe the answer to that question is yes.1

This answer – in fact the whole Facebook hearing, the scandal that
led up to it, and the line of questions regarding what kind of company
Facebook is in reality – is important for anyone who wants to understand
the contemporary media landscape and the information ecosystems that
make up the public spheres not only in the US, but almost everywhere.
Consequently, Walden’s questions and Zuckerberg’s answer are important
when trying to understand the nature of digital journalism studies. This
field of research – digital journalism studies – has become an important
area of study within communications during the last decade because it
addresses core questions related to the economy, technology, sociology,
culture, language, psychology, and philosophy of what journalism is. It
comes at a time when older demarcations – like those between different
institutions and companies, between audiences and professionals, prac-
tices and perceptions, production and consumption, technologies and
humans, physical and virtual, private and public, facts and fictions, truth
and lies, and many more – no longer seem valid.
The significance of Facebook and other global platforms and tech com-
panies unknown to the world before the turn of the millennium cannot
be overestimated. They constitute a major reason why digital journalism
studies is heavily influenced by what Ahva and Steensen (2017) label a
“discourse of deconstruction”, in which it has become essential to ask
fundamental questions concerning what journalism is. Let us offer a few
examples of how this discourse of deconstruction has been articulated
during the formative years of digital journalism studies as a research field.
Anderson (2013) argued that the classical newsroom is no longer the
 The introduction 3
epicenter of newswork and that bloggers, citizen journalists, and social
networks are, alongside journalists, important actors in the new “news
ecosystem”. Peters and Broersma (2013) argued that the problems facing
journalism are far more structural than previously suggested, requiring a
fundamental rethink about what journalism is. Carlson and Lewis (2015)
argued that journalism’s demarcations towards other professions and busi-
nesses are deconstructed, as are previously established internal boundaries
between for instance different journalistic genres, and groups of journal-
ists. And Boczkowski (2011, p. 162) argued for a need to shift “the stance
of theoretical work from tributary to primary” in studies focusing on
journalism in digital times.
In this book we interrogate the nature of digital journalism studies. We
probe the roots from which the field has grown, the technologies, plat-
forms, devices, and audience relations that constitute central objects of
study, the theories from which research embarks, the (sometimes) innova-
tive research methods being developed, and the normative underpinnings
and possible futures of the field. It is our early contention that digital
journalism studies is much more than simply the study of journalism
produced, distributed, and/or consumed with the aid of digital technolo-
gies. Digital journalism is not defined by its relation to technology alone;
such a definition “short-circuits a comprehensive picture of journalism”,
as Zelizer argues (2019, p. 343). The scholarly field of digital journal-
ism studies is built on questions that disrupt everything previously taken
for granted concerning media, journalism, and public spheres: What is
a media company? Who is responsible for what is published in a public
sphere? What is the difference between those who produce, those who
distribute, and those who consume media content, including journal-
ism? And indeed who is a journalist and what is journalism in this com-
plex media and information ecosystem of the 21st century? In search for
answers to such questions, digital journalism studies also moves beyond
journalism studies and constitutes a cross-disciplinary field that does not
focus on journalism only from the traditions of journalism studies, but is
open to research from, and conversations with, related fields.
In this introduction, we first look at four structural premises for why
questions such as those posed in the previous paragraph are relevant
today, and why they matter for digital journalism studies. These struc-
tural premises are related to the economy, audience relations, and the net-
worked distribution and consumption mechanisms of digital journalism.
We then argue that a fundamental development for digital journalism
studies is the way in which news has become separated from journalism
since the 1990s. The chapter outlines some empirical characteristics of
what digital journalism studies looks like today, as it is presented in the
4  The introduction
most important arena through which the field materialises, namely the
journal Digital Journalism. Finally, we present the outline of the book.

1.1 Four structural premises for digital

journalism studies
The 2018 Facebook hearing offers an interesting way to begin explor-
ing the topics introduced briefly above not only because it was such
an exceptional example of how older and familiar categories of  – and
demarcations between  – different types of companies seem no longer
valid, but also because of the scandal leading up to it, the Cambridge
Analytica scandal. This revealed the disruptive changes around how
information flows in our digital age – changes that have severe conse-
quences for journalism.
The scandal revealed that Facebook had provided access to personal
data from 87 million Facebook users to the Cambridge Analytica politi-
cal consulting and data analytics firm. It also highlighted the enormous
potential for how user data can be exploited for both commercial and
political gains without users’ knowledge or consent along with the ensu-
ing privacy protection issues (Isaak & Hanna, 2018). The scandal was a
demonstration of the consequences of what Manovich (2018) has labelled
the media analytics stage of modern technological media. It has become
evident that the real value of global platform companies like Google,
Amazon, and Facebook, as well as Asian platforms such as WeChat and
Weibo, lies in their sophisticated methods for harvesting, analysing, and
capitalising from tremendous amounts of big data on user behaviour.
These methods empower the platform companies with knowledge and
insights advertisers are willing to pay for, but also with a wider control
over cultural and social networks (Taplin, 2017). The implications of this
for journalism have been:

1 A massive shift and crisis in revenue models because advertisers have

migrated to platform companies (see for instance Kaye  & Quinn,
2010), while news publishers nowadays typically get most of their
revenue from their readers.
2 An increased emphasis on user data and audience analytics and met-
rics in journalism (Belair-Gagnon  & Holton, 2018; Cherubini  &
Nielsen, 2016; Ferrer-Conill & Tandoc, 2018).
3 Shifting patterns of distribution in which companies non-proprietary
to institutions of journalism have gained dominance (see for instance
Kalsnes & Larsson, 2018; WAN-IFRA, 2019; Westlund & Ekström,
2018).
 The introduction 5
These three implications are important structural premises for digital
journalism studies as an academic field. Moreover, the Cambridge Ana-
lytica scandal highlighted another aspect that has dominated much of
recent debates in public, industry, and academic discourses on journalism
and news; namely problems related to disinformation, “fake news”, and
trust in the media. Cambridge Analytica used the Facebook data and
other data to target US citizens with bespoke political propaganda during
the 2016 presidential election campaign and in other elections around
the world, including the UK Brexit vote. Reports following the scandal
revealed that the company had included disinformation and other forms of
information manipulation in their propaganda campaigns, and a tsunami
of revelations of similar disinformation campaigns followed (Posetti  &
Matthews, 2018). This has become a severe problem for journalism, not
only because fake news is difficult to disentangle from real news, but
also because in another dimension of fake news discourse, the term is
used to discredit what is often legitimate news (Egelhofer  & Lecheler,
2019). This dimension is seen in President Trump’s “fake news”/“fake
media” rhetoric towards legacy news institutions  – a rhetoric adopted
by other state leaders and politicians around the world (such as in Brazil
and Nicaragua), in addition to activists and interest groups, most notably
those belonging to the political far right. In sum, the two dimensions of
fake news hurt journalism because “the media’s dependence on social
media, analytics and metrics, sensationalism, novelty over newsworthi-
ness, and clickbait makes them vulnerable to such media manipulation”
(Marwick & Lewis, 2017, p. 1). In other words: the three implications
for journalism based on the structural developments in the digital media
and information landscape highlighted above – the disruptive changes in
the media economy; the emphasis on audience analytics and metrics; and
changing distribution patterns – create a fourth implication:

4 Journalism has become more vulnerable to manipulation, disinfor-

mation, and a consequent lack of public trust.

One response by news publishers has been an increased emphasis on

institutionalising practices of fact checking and information verifica-
tion (Graves, 2018), which in turn has created increased interest in both
industry and the academy in questions of epistemology: how journal-
ists produce knowledge claims, how they deal with uncertainty, what
counts as truthful information, and how all this is affected by the devel-
opments in digital media and information technology (Amazeen, 2015;
Ekström  & Westlund, 2019a; Eldridge II  & Bødker, 2019; Steensen,
2019). Moreover, increased distrust in legacy news institutions has given
6  The introduction
rise to new branches of alternative media and news outlets, especially
from the political far right, with different epistemologies (Figenschou &
Ihlebæk, 2019; Holt, Ustad Figenschou,  & Frischlich, 2019; Nygaard,
2019). The so-called five W’s (who, what, where, when, and why) have
recurrently been applied for thoughtful analyses about digital journalism
studies (Tandoc, 2019b; Waisbord, 2019), and there have been ongo-
ing efforts into the study and debate of key issues such as: what is news,
who is a journalist, who are peripheral actors, and what is their role and
power in practice (Ahva, 2019; Chua & Duffy, 2019; Eldridge, 2019).
Such studies and debates are not merely academic exercises but can have
a fundamental impact on who gets to produce and distribute news, and
whether media policy enforces functions for support or disabling. More
specifically, authorities can take charge over definitions concerning who
is a journalist, and who produces misinformation (Belair-Gagnon, Hol-
ton, & Westlund, 2019), while platform companies have avoided defin-
ing themselves as publishers and thereby are not responsible for editorial
content published and distributed on their platforms (Gillespie, 2018).
These four premises, together with the confusion concerning which
companies play which roles related to the production, distribution, con-
sumption, and technological facilitation of news, form the structural
backbone of digital journalism studies. They inform investigations into
the whos, whats, whens, and wheres of contemporary journalism and
they call into question previously established knowledge on what jour-
nalism is, who counts as a journalist, and what role journalism plays in
societies and for the people.

1.2 The separation of news from journalism

The four premises discussed above would not have materialised without
one key change in modern media landscapes and public spheres: the ways
in which news has become increasingly separated from journalism. When
the two authors of this book grew up in Norway (Steensen) and Sweden
(Westlund) during the 1970 and 80s, news was inseparably tied to jour-
nalism. News was delivered in national newspapers that landed on our
doorsteps every morning, in local newspapers delivered by paper boys
and girls every afternoon, and, most importantly, through the evening
news broadcast by the national public broadcasters NRK (Norway) and
SVT and SR (Sweden). Accessing the news was routine. It was delivered
in fixed and recognisable formats at specific times and places and it was
produced, distributed, and consumed in ritual manners (Carey, 1992).
Journalism is still very much bound by ritual, especially in how it cov-
ers events in the world and constructs and upholds social norms and
 The introduction 7
cultural values in a given society (Peters, 2019b). But news is no longer
tied to journalism in the same way. News has become dislocated from the
proprietary platforms of news companies (Ekström & Westlund, 2019b)
and news rituals have expanded way beyond the production, distribution,
and consumption of journalism. This separation of news from journalism
began with the popularisation of the World Wide Web during the 1990s
and what Manovich (2018) calls “the Web as global content creation and
distribution network” stage in the development of modern technological
media. With the web, journalistic institutions lost their almost monopo-
listic position as providers of news to mass audiences, since everyone
could now set up a web page, create content, and distribute it to a public
audience. Governments, public bodies, political parties, politicians, pri-
vate enterprises, NGOs, and other kinds of institutions could set up their
own news services through the web and bypass journalists; so could pri-
vate individuals. Some individuals were very successful, like the former
telemarketer Matt Drudge who in 1996 started publishing the Drudge
Report, which became a highly influential news provider and political
commentary website in the US (Leetaru, 2009).
The separation of news from journalism escalated when the blog for-
mat became popular in the early 2000s. Blogs allowed individuals with
limited tech savviness to set up news services with little effort and cost,
and marked the beginning of the social media platforms stage of modern
technological media (Manovich, 2018), in which discourses of participa-
tion (Singer et  al., 2011), user-generated content creation and utilisa-
tion (Ornebring, 2008; Thurman, 2008), and citizen reporting (Allan &
Thorsen, 2009) became popular in both journalism and journalism stud-
ies. This created a situation in which the boundaries between those who
produce and those who consume news became blurred and coalesced
in “produsage” (Bruns, 2010). Combined with the massive industrial
changes in the media landscape and economy globally (exemplified with
the rise of Facebook and Google), technological innovation, distribution,
and social interaction became the new kings (Albarran, 2016) who pro-
vided a forceful push towards separating news from journalism.
Today, news is something that you find in formats and on platforms of
your own choosing. News is more often than not deprived of edited con-
texts and fixed genres and formats, and reaches you in mash-ups contain-
ing journalistic news, public relations news, advertisements, news from
politicians, celebrities, sports idols, and artists, personal news from your
friends and family, professional news from your colleagues and profes-
sional associations, and perhaps also fake news from bots. These news
mash-ups, which typically reach you in social media feeds, are person-
alised interfaces with an abundance of information floating around in
8  The introduction
bits and pieces in a gigantic, digital network. Journalism is one among
these sources that both build on each other and are increasingly difficult
to separate from one another for the end-user. News used to be fixed
in time, space, culture, materiality, and patterns of production, distribu-
tion, and consumption. Now news is networked (C. W. Anderson, 2013;
Domingo, Masip, & Costera Meijer, 2015; A. Russell, 2013). It exists in
information “ecosystems” (Picard, 2014) with strong or weak connec-
tions to journalism, connections that might be difficult to detect.

1.3 What does Digital Journalism studies look like?

Throughout the book we will assess the development of the field through
a systematic review of articles published in journals, most notably the
journal which has most shaped the field, Digital Journalism. This journal
was launched in 2013 to be a “critical forum for the scholarly discus-
sion, analysis and responses to the wide-ranging implications of digital
technologies for the practice and study of journalism” (Franklin, 2013,
p. 1). Digital Journalism quickly became a highly influential journal, not
only within journalism studies, but also within the broader discipline of
communication. Table 1.1 displays citation metrics and rankings within
the discipline of communication of the five most influential journalism
journals internationally: Digital Journalism, Journalism – Theory, Practice &
Criticism, Journalism Studies, Journalism Practice, and Journalism & Mass Com-
munication Quarterly. Even though Digital Journalism is the youngest of
these journals, it became the highest-ranked journalism journal by quite
a large margin in 2018. This journal is therefore important to assess when
analysing the nature of digital journalism studies and its development.
Steensen and colleagues (2019) have previously conducted a content
analysis of Digital Journalism in order to assess what digital journalism studies,
as portrayed in this journal, looks like, and also of other journalism journals

Table 1.1 2018 citation metrics and ranking within the discipline of communication
from SJR (SCImago Journal Ranking), Scopus, and Google Scholar. The
table displays the five top journalism journals.

Journal SJR Google citations Scopus

Rank Impact factor Rank H5 Index Rank CiteScore

Digital Journalism 9 2,67 5 44 5 4,55

Journalism TP&C 19 1,62 9 39 19 2,98
Journalism Studies 20 1,55 10 38 26 2,74
Journalism Practice 27 1,36 11 36 32 2,53
Journalism & Mass Comm. Q 29 1,32 17 32 25 2,74
 The introduction 9
to determine degrees of interdisciplinarity and theoretical perspectives used
in journalism studies in general (Ahva & Steensen, 2020; Steensen & Ahva,
2015). We build further on these analyses and present findings throughout
this book, predominantly based on the analysis of keywords used to tag the
articles published in Digital Journalism from the first issue in 2013 to issue
4, 2019, abstracts of the same articles, the references cited in them, and
the nationality of the articles’ authors. For those who are interested in the
methodological procedures behind the analysis, we have added an online
appendix where these procedures are laid out and discussed.
We will briefly discuss two aspects of this analysis in this introduc-
tory chapter: the interdisciplinarity of digital journalism studies, and the
degrees to which the field is globally diverse.

1.3.1 The interdisciplinarity of digital journalism studies

In an introductory essay to a special issue discussing definitions of both
digital journalism and its study, the editorial team of Digital Journalism
argue that viewing digital journalism studies as a sub-field of journal-
ism studies “limits its value and potential to scholarship not just within
media studies and communication, but its wider interdisciplinary reach”
(Eldridge II, Hess, Tandoc, & Westlund, 2019, p. 393). The interdiscipli-
narity of digital journalism studies is in other words a key characteristic
of the field, according to the editorial team. However, journalism studies
is also reckoned to be an interdisciplinary field and the question therefore
becomes to what degree the two fields differ in their interdisciplinarity.
Journalism studies is a young academic field rooted in the social sci-
ences and the humanities. It is traditionally marked by approaches and
perspectives from sociology, political science, cultural studies, language
studies, and history (Zelizer, 2004). In a longitudinal analysis of disci-
plinary perspectives found in abstracts of articles published in the jour-
nals Journalism Studies and Journalism  – Theory, Practice  & Criticism from
2000 to 2013, Steensen and Ahva (2015) found that sociology was the
main source of influence in journalism studies and that this discipline
had become increasingly dominant. Political science perspectives, which
dominated the field in the early 2000s, was the second most common
discipline, while cultural studies, language studies, and history played
minor parts. In addition, fields and disciplines like business and adminis-
tration, economics, law, and philosophy were present, while technologi-
cal perspectives were on the rise.
The question then is whether digital journalism studies is marked by
the same disciplinary patterns as journalism studies, or if it has different
sources of influence. The discussion in section 1.1 of the premises that
form the backbone of the field suggests that digital journalism studies
10  The introduction
draws on a wider range of perspectives than journalism studies, given
digital journalism’s influences from and orientation towards practices,
professions, institutions, technologies, and cultures beyond journalism,
and given the field’s emphasis on change as a formative concept (Peters &
Carlson, 2019), a point we will discuss more thoroughly in chapter  6,
section 6.2.2. However, recent reviews reveal that there is a discrepancy
between this expected level of interdisciplinarity in digital journalism
studies and the actual research being conducted within the field. Boc-
zkowski and Mitchelstein (2017) argue that digital journalism studies is
marked by two limitations: 1) the ability to connect empirical findings
from digital journalism studies across other domains of digital culture,
and 2) a lack of conceptual exchanges with other fields and disciplines.
Steensen and colleagues (2019) found similar tendencies and argued that
digital journalism studies could benefit from more inclusion of perspec-
tives from the humanities, and of theoretical and not only methodologi-
cal perspectives from information science and computer science.
Figure 1.1 displays the disciplinary perspectives that dominate abstracts
of articles published in Digital Journalism (see online appendix for details
on methodology). One third of the abstracts analysed draw primarily
on sociological frameworks, making this the most common disciplinary

Perspectives in Digital Journalism abstracts 2013-2019

Law, 1%
History, 2% Philosophy, 1%
Other, 2%
Culture, 3%
Language, 6% Sociology, 32%

Business, 10%

Political science, 17%

Technology, 26%

Figure 1.1 Share of the most dominant disciplinary perspectives in abstracts of articles

published in Digital Journalism from issue 1, volume 1 (2013) to issue 4,
volume 7 (2019). Every second abstract is analysed (N = 172). See online
appendix for details on methodology.
 The introduction 11
perspective. This is quite similar to the dominance of sociology in journal-
ism studies in general, as is the share of articles based primarily on a politi-
cal science framework (Ahva & Steensen, 2020; Steensen & Ahva, 2015).
The difference between articles published in Digital Journalism and
in other journalism journals is that technological perspectives are much
more common in Digital Journalism, where they dominate every fourth
article. Another difference is that disciplines from the humanities, such as
cultural studies, language studies, history studies, and philosophy, are less
common in articles published in Digital Journalism than in other journal-
ism journals.
We will return to these interdisciplinary characteristics of digital jour-
nalism studies and what they mean in several chapters throughout this
book, especially in chapters 5 and 7. For now we will conclude that there
are both similarities and differences in the ways in which digital journal-
ism studies and journalism studies are interdisciplinary.

1.3.2 Digital journalism studies and global diversity

The Facebook hearings in 2018 illustrate a key dimension of the modern
media and technology landscape: it is inherently global. Facebook and other
platform companies within the media and technology industries know few
national boundaries, with notable exceptions like China, in which Face-
book and Google are banned. Nonetheless, the big platform companies
have global outreach, as do the infrastructure that facilitates their existence,
the internet and the World Wide Web. The Cambridge Analytica scandal
was also of global proportions, since this company had not only interfered
in the 2016 presidential elections in the US, but allegedly in more than 200
elections worldwide, including in Argentina, Nigeria, Kenya, India, and
the Czech Republic (Posetti & Matthews, 2018, p. 14).
Digital journalism, which is facilitated by the same globalised infra-
structure everywhere, is almost by default a global phenomenon. Bob
Franklin, the founding editor of Digital Journalism, proclaimed that a core
commitment of the journal’s editorial policy would be to appreciate a
multitude of geographical contexts, which would imply seeking out

studies which explore developments in digital journalism in those

regions of the globe which typically do not enjoy the same access
to the debating chamber constituted by western-based journals that
is enjoyed by scholars and journalists in the developed global north.
(Franklin, 2013, p. 3)

Digital Journalism has in recent years taken steps towards global diversi-
ties in terms of who has been invited to join the quite large and diverse
12  The introduction
editorial board, where the journal encourages submissions from, and
where it would like to engage both academics and other audiences. Con-
cerning the latter point, the 2019 appointment of three international
engagement editors in the US, Chile, and Singapore is a clear sign of the
journal’s ambition towards global outreach and diversity. However, the
journal has not (yet) managed to live up to this commitment in terms of
where authors come from, at least not compared to the other journalism
journals, which have a more globally diverse set of authors. Figure 1.2
displays an overview of the parts of the world first authors of articles pub-
lished in the five top-ranked journalism journals represent. Ninety per-
cent of the Digital Journalism first authors are based in North America or
Western Europe, making it the most Western-centric of the five journals.
Diversity is a topic we will explore in several chapters in this book: in
chapter 2, where we introduce an analytical framework for understand-
ing the relationship between digital journalism and its object of inquiry
(see section  2.2); in chapters  3 and 4, where we unpack the diversity
of the objects of study in digital journalism studies; and in chapters  5
and 6, where we discuss theoretical and methodological diversity. These
discussions will undoubtedly reveal that digital journalism studies is a
very diverse field and that its premises and founding principles assume a
global perspective and research agenda. That said, by way of published
articles in these five journals, the field remains dominated by scholars
based in North America and Western Europe. In this context it is worth
remembering that there are numerous additional journals producing large
amounts of research associated with specific geographical regions, such as
African Journalism Studies, Asian Journal of Communication, Brazilian Journal-
ism Research, and Chinese Journal of Communication, to mention some key
examples of journals from the beginning of the alphabet.

1.4 Outline of the book

This book comprises eight chapters, which in their survey of the historical
origins of digital journalism studies to its possible futures, explore in more
detail the topics raised briefly in this introduction. In chapter  2, “The
Definitions: Current Debates and a Framework for Assessing Digital Jour-
nalism Studies”, we argue that the origins of digital journalism studies lie
in the research that has explored the historic relationship between journal-
ism and technology. The chapter revisits this relationship, before moving
towards the current debates on how to define both digital journalism and
digital journalism studies. The chapter also introduces a framework to ana-
lyse the relationship between the academic field and its object of inquiry, a
framework consisting of three dimensions: society, sector, and scholarship.
 Where do authors in journalism journals come from?

Journalism Practice 42% 44% 6% 4%

Journalism & MCQ 74% 12% 1% 11%

Joirhalism TP&C 39% 45% 8% 6%

Journalism Studies 31% 49% 8% 8%

Digital Journalism 38% 52% 5% 3%

0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%

North-America Western Europa and others Oceanea Asia and the Pacific
Africa Latin America and Caribbean Eastern Europa

Figure 1.2 First authors of all articles published in the journals Digital Journalism, Journalism Studies, Journalism: TP&C, Journalism  & Mass
Communication Quarterly, and Journalism Practice during the years 2013–2019. Definitions of the regions are based on the UN
Regional groups.
Note: Data from 2013 and 2014 for Journalism Practice was not available. N = 1989 (no. of articles). See online appendix for details on methodology.
The introduction 13
14  The introduction
In chapter 3, “The Technologies: Unpacking the Dominant Object
of Study in Digital Journalism Studies”, we discuss the research on
technological aspects of digital journalism studies, specifically related
to the topics of data journalism, analytics, and metrics, as well as algo-
rithms and automation. Chapter 4, “The Platforms: Distributions and
Devices in Digital Journalism”, discusses the role of a diverse set of plat-
forms (most notably social media platforms) and how they have been
researched in the field, in addition to looking at how devices such as
tablets, smart phones, drones, and others have played a significant role
in the research field.
Chapter 5, “The Theories: How Digital Journalism is Understood”,
considers the role of theory in digital journalism studies. It builds further
on the meta-analysis of the journal Digital Journalism briefly presented
in this chapter, as this meta-research includes analysis of what role the-
ory plays in articles published in this journal and what kinds of theories
are adopted and developed by the research. Chapter 6, “The Assump-
tions: The Underlying Normativity of Digital Journalism Studies”, will
unmask and discuss the normative underpinnings of digital journalism
studies and argue that hidden normativity is a problem related to three
discourses that dominate much of digital journalism studies, namely the
discourses of crisis, technological optimism, and innovation. Chapter 7,
“The Methodologies: How Digital Journalism is Researched”, explores
the methodologies of digital journalism studies which increasingly derive
from information science and computer science as new kinds of data
become available for researchers.
Based on the discussions in the previous chapters, the last chapter,
chapter  8, “The Futures: Deconstructions of and Directions for Digi-
tal Journalism Studies”, provides a road map for the directions digital
journalism studies might take. We argue that digital journalism studies is
not best served by an agreement on what road to follow, but that several
directions must be taken simultaneously since the future of journalism in
our digital age, and the future of digital societies in general, are impossi-
ble to predict. However, the chapter also argues that there are some blind
spots left behind by digital journalism studies that need to be addressed
and some normative assumptions that need to be scrutinised.

Note
1 Transcript of the hearings found at www.washingtonpost.com/news/the-switch/
wp/2018/04/11/transcript-of-zuckerbergs-appearance-before-house-commit
tee/ (accessed 3 October 2019). Video available at www.cnbc.com/2018/04/11/
mark-zuckerberg-facebook-is-a-technology-company-not-media-company.html
(accessed 3 October 2019).
2 The definitions
Current debates and a framework for
assessing digital journalism studies

In recent years a great number of journalism studies scholars have devel-

oped research agendas which are increasingly oriented towards digital
journalism. They have been joined by scholars from many different fields
including, but not limited to, computer science, political communica-
tion, media management, mobile media, and communication. This chap-
ter discusses the parallel emergence of digital journalism studies as a field,
and digital journalism as a practice. Here we discuss the roots and current
developments of digital journalism studies, as well as important debates,
approaches, and definitions of the field. The chapter presents an analytical
framework that, going forward, allows us to further our understanding of
how research about digital journalism corresponds with developments in
digital journalism. Core questions for this chapter are: Where does digital
journalism studies come from? How can it be defined? And how does it
interplay with changes in society and the journalism sector?

2.1 Digital journalism studies: definitions

and debates
Both digital journalism and digital journalism studies are contested and
widely discussed concepts, which also represent “moving targets” that
change over time. Scholars disagree about what digital journalism and
digital journalism studies “are”, and it is a daunting task to do justice
to the different and diverse positions and nuances. We will nevertheless
attempt to outline some key aspects and arguments. Digital journalism
studies has been influenced by many fields but has largely emerged from
journalism studies, which in turn can be placed within the larger disci-
pline of communication (Carlson, Robinson, Lewis, & Berkowitz, 2018).
Importantly, though, scholars have proposed to approach digital journal-
ism studies as a field of its own, drawing not only on its well-established
links to disciplines such as communication, political communication,
16  The definitions
sociology, and economics, but also on fields more focused on “digital”
aspects, such as computer science and information science (cf. Eldridge II
et al., 2019). We stress that this obviously means that scholars can apply
and develop other approaches to their epistemological processes of pro-
ducing scientific knowledge. By approaching digital journalism studies
as a field of its own, it can evolve more openly without abiding to the
ways-of-doing largely established within journalism studies, and which
may create certain expectations and path-dependencies.
The most recent and systematic effort towards advancing this discus-
sion is to be found in a special issue in Digital Journalism published in
2019 titled “Defining Digital Journalism (Studies)”. The Digital Journal-
ism editorial team invited a number of scholars from different parts of the
world who could offer different viewpoints, arguments, and discussion
around the shape of the field. Table 2.1 presents us with the concise defi-
nitions offered by each of these contributors. The special issue starts with
a definition of digital journalism based on an empirical review and analy-
sis of the field, continues with five conceptual articles, and ends with a
synthesis by the members of the editorial team, discussing nuances in the
different approaches and definitions (Eldridge II et al., 2019). One key
aspect concerns the role of digital technologies in relation to journalism.
As Zelizer (2019) highlights, journalism is a cultural practice and scholars
should not overemphasise the role played by digital technology. Other
contributors offer arguments about how to think of digital technology
in ways that go beyond seeing them as tools and systems, but rather as
embedded in a broader set of socio-technical dynamics. Contributing
scholars highlight that the digital is transcendental (Robinson, Lewis, &
Carlson, 2019), transforming and/or expanding journalism (Steensen
et al., 2019; Waisbord, 2019), and playing an important role in bringing
forward new rules and processes (Duffy & Ang, 2019).
Both authors of this book have also been involved in forwarding defi-
nitions to this debate. The first author (Steensen) was lead author on a
review article for the above-mentioned special issue of Digital Journalism,
offering a -multidimensional assessment of all articles published in the
journal from 2013 until mid-2018, and thus providing an empirically
based definition that stresses mutual dependence with digital technol-
ogy and a symbiotic relationship with audiences (Steensen et al., 2019).
The article calls for greater awareness of the different kinds of knowledge
that digital journalism studies scholars produce. This chapter attempts to
address that void. Moreover, the second author (Westlund) coauthored
the definition forwarded by the Digital Journalism editorial team, propos-
ing their normative approach to digital journalism studies for the journal.
They introduced the “The Digital Journalism Studies Compass” (the DJS
Table 2.1 Definitions of digital journalism and digital journalism studies discussed in a special issue of Digital Journalism. The table was originally
published in Eldridge II et al. (2019, p. 392).

Steen Steensen, Anna M. Grøndahl Larsen, Ynqve Benestad Hågvar and Birgitte Kjos Fonn, 2019, 338
Digital journalism is the transforming social practice of selecting, interpreting, editing and distributing factual information of perceived public
interest to various kinds of audiences in specific, but changing genres and formats. As such, digital journalism both shapes and is shaped by
new technologies and platforms, and it is marked by an increasingly symbiotic relationship with the audiences. The actors engaged in this
social practice are bound by the structures of social institutions publicly recognized as journalistic Institutions
Sue Robinson, Seth Andrew Duffy and Silvio Waisbord, Jean Burgess and Edward Barbie Zelizer, 2019,
C. Lewis and Matt Ang Peng Hwa, 2019, 352 Hurcombe, 2019, 360 349
Carlson, 2019, 2019, 382 Digital journalism Those practices of Digital journalism thus
369-370 Digital journalism as is the networked newsgathering, reporting, takes its meaning from
Research that involves the way in which production, textual production and both practice and
newswork employing journalism embodies distribution and ancillary communication rhetoric. Its practice as
digital technologies the philosophies, consumption that reflect, respond to, newsmaking embodies
in some manner, such norms, practices, of news and and shape the social, a set of expectations,
as news websites, values and attitudes information. It is cultural and economic practices, capabilities
social platforms, of digitisation as characterized by logics of the constantly and limitations relative
mobile devices, data they relate to society. network settings changing digital media to those associated with
analytics, algorithms, These include the and practices environment. To study pre-digital and non-
etc.; Research that efficiency of control, that expand the digital journalism is to digital forms, reflecting
acknowledges how storage, retrieval, opportunities and study the transformative a difference of degree
digital dynamics of accessibility and spaces for news and isomorphic impacts of rather than kind. Its
journalism interact transmission of data; digital media technologies rhetoric heralds the hopes
with and alter formerly indusivity, interactivity and business models on and anxieties associated
discrete boundaries . . . and collaboration in the practice, product and with sustaining the
and the authority and the propagation of business of journalism, as journalistic enterprise as
forces that go along information and well as the ways that worthwhile. With the

(Continued)
The definitions 17
Table 2.1 (Continued)

with these changes and opinion; flexibility journalistic discourses, digital comprising the
configurations; Research and innovation in practices and logics in figure to journalism's
that interrogates the presenting news stories; turn shape the cultures ground, digital journalism
resulting practical and and state, institutional and technologies of those constitutes the most
cultural transformations and individual digital media platforms recent of many conduits
18  The definitions

occurring around ownership of data through which journalism over time that have
news and other acts of and its implications is practiced, and its allowed us to imagine
journalism as they relate for privacy and products are shared and optimum links between
to broader issues . . . transparency consumed journalism and its publics
Scott Eldridge, Kristy Hess, Edson Tandoc, and Oscar Westlund, 2019, 394.
Digital Journalism Studies should strive to be an academic field which critically explores, documents, and explains the interplay of digitization
and journalism, continuity and change.
Digital Journalism Studies should further strive to focus, conceptualize, and theorize tensions, configurations, power imbalances, and the debates
these continue to raise for digital journalism and its futures.
 The definitions 19
Compass) as a visual and metaphorical tool that can help guide scholars in
navigating the interrelationships between “digital” and “journalism” on
the one end, and “continuity” and “change” on the other. Their empha-
sis on “continuity” seeks to make sure scholarship builds on existing
knowledge, theories, and concepts, while “change” opens for cutting-
edge scholarship that pushes these boundaries (Eldridge II et al., 2019).

2.2 An analytical framework: society, sector,

and scholarship
This discussion of debates in the field extends into the nature of the
relationship between researchers and their objects of study. Therefore, we
have developed an analytical framework that helps visualise and explain
developments in digital journalism vis-à-vis digital journalism studies.
This framework encompasses three core dimensions: A) society, B) sector,
and C) scholarship.
With society we refer to how the world changes at a global, national,
and local level, including but not limited to political, economic, social,
and technological factors. While this dimension is very important, this
book does not aspire to discuss such changes in close detail. Our purpose
with introducing the society dimension into the analytical framework
has to do with its relevance for discussions of the two other dimensions.
The second dimension, sector, encompasses journalism as a phenomenon
and institution, as a market and industry, as well as a profession, practice,
service, and product. The society (A) and sector (B) dimensions represent
what theory of science would refer to as ontology. This concept refers to
reality and existence, essentially what the world is, and what can be said
to exist. From our perspective, the journalism sector clearly exists and
continuously changes in relation to society. The conditions surrounding
the “reality” of the journalism sector have been measured, studied, dis-
cussed, and approached in multiple ways. There are of course “facts” in
the world, such as when news publisher company X makes redundant Y
number of journalists on a specific date. We tend to see such occurrences
as facts based on widely agreed upon principles for the calendar system
and mathematics. However, things become less straightforward when we
turn to questions of the antecedents to why these journalists were sacked.
Was it because of poor leadership? Or because of publishers’ loss of reve-
nue to platform companies? Or was it perhaps a mix of these factors, and
many more? Members of the journalism sector attempt to resolve such
questions as they navigate these challenges, and so do researchers in fields
such as digital journalism studies. Importantly, we subscribe to the posi-
tion that scholars socially construct accounts of reality, whereby assessing
20  The definitions
these factors we are better able to gain a picture of the “reality” associated
with journalism. Thus, we engage in epistemological processes of pro-
ducing scientific knowledge about complex and continuously evolving
changes in society and the journalism sector as they intertwine. The schol-
arship (C) dimension thus has to do with the epistemologies with which
scholars produce knowledge.
This analytical framework helps to illuminate the interrelationship
between the journalism sector and scholarship on digital journalism studies
and can guide analyses and discussion of whether the journalism sector
and digital journalism studies scholars have focused on similar or dissimi-
lar questions. The components of the analytical framework are brought
into discussions of research in thematic clusters in chapter 3 and 4, and
has inspired our conclusions and directions in chapter 8.
We ask you to imagine a timeline with different milestones that have
had major, perhaps even disruptive, effects on journalism. For exam-
ple, imagine for a moment how (A) developments in the telecom sector
have influenced society and (B) the journalism sector, as well as specific
news publishers, and (C) subsequently also generated numerous studies
in digital journalism studies. Diffusion of mobile telephony substantially
improved journalistic fieldwork and the possibilities for getting in touch
with sources. The launch of the Apple mobile ecosystem with native apps
in 2007, and thereafter the Android mobile operating systems, spurred
substantial shifts when it comes to how the journalism sector publishes
news, and how citizens access the news (Westlund, 2013). Mobile com-
munication is a taken-for-granted part of society (Ling, 2012), and
mobile devices offer ubiquitous access to citizens, presenting publishers
with both opportunities and challenges. Indeed, cross-cultural surveys
show that most people use smartphones as their main gateway for news
and that people use social media platforms more generally to access news
(Newman, Fletcher, Kalogeropoulos, & Nielsen, 2019). Notwithstanding
this, research into mobile journalism and mobile news remains relatively
limited. Studies into newsrooms and newswork have typically focused on
“online” in general, largely overlooking approaches and practices related
to “mobile” devices and aspects (see review in Westlund & Quinn, 2018).
This area of research, discussed further in chapter  4, remains relatively
fertile ground although news publishers have experimented with mobile
news more or less as long as they have with social media.
To return to our framework of society (A), sector (B), and scholarship (C),
we wish to highlight that although mobile media and communication
have gained significance in society and the journalism sector (A+B), rela-
tively few publications have focused on, or even considered, such aspects
and developments. Ultimately, we argue that A+B disconnects with C
 The definitions 21
when it comes to how stakeholders in each camp have focused their
attention. This discussion of mobile technologies and communication, as
but one example of research in the field, shows how such disconnections
influence the knowledge production in, and thereby epistemologies of,
digital journalism studies.
Our contribution in this chapter, however, does not sit within an effort
towards offering an empirical, systematic, and comprehensive analy-
sis of the interplay between society, sector, and scholarship. Instead, we
turn now towards unpacking the interrelationship between these three
dimensions by introducing four key mechanisms: 1) Issue (in)visibility,
2) Pro-innovation bias, 3) Path dependency, and 4) Addressability. The
first focuses on issue (in)visibility in the journalism sector (B), whereas
the second (pro-innovation bias) and third (path dependency) apply to
both the journalism sector (B) and to digital journalism studies (C). The
fourth mechanism, addressability, mainly applies to digital journalism
studies. We will now introduce each of these four dimensions and con-
tinue by building on our example of mobile news.

2.2.1 Issue (in)visibility
The nature of visibility and invisibility concerning what happens in the
journalism sector (B) varies substantially. Representatives from the jour-
nalism sector may deliberately draw a lot of attention to certain innova-
tions by the news industry. For example, industry members have taken
pride in building their social media presence, increasing audience engage-
ment, and developing and launching mobile applications. They have com-
municated about such developments quite broadly and publicly, including
in industry press and trade magazines, public talks, and in the news itself
where they pitch such developments towards their audiences. By contrast,
other issues have been largely invisible, like how the journalism sector
addresses challenges relating to digital safety, including safe communica-
tion with sources, online harassment, hacks, and surveillance. Exceptions
include handbook chapters covering such issues (Franklin & Eldridge II,
2017), and Digital Journalism has also published a special issue focusing on
surveillance (Wahl-Jorgensen, Hintz, Dencik, & Bennett, 2017).
Moreover, we posit that issue (in)visibility in the journalism sector
(B) influences the extent to which different stakeholders learn from and
mimic each other (labelled: isomorphism and herd behaviour), as well as
the extent to which digital journalism scholars (C) conduct research into
specific areas. The mechanism is relational: high visibility in the sector
(B+) likely results in more digital journalism scholarship (C+), whereas
invisibility (B-) reduces the chances of such issues being researched (C-).
22  The definitions
Turning back to the example of mobile news we argue that while
mobile applications were central objects of inquiry during the forma-
tive years of smartphones and tablets with touchscreen, they have since
been appropriated and normalised. Invisibility for journalists, as a conse-
quence, increases when their news organisations incorporate mobile into
their cross-media approaches, using content management systems (CMS)
that are designed to effortlessly publish across desktop, tablet, and smart-
phone sites and apps (e.g.,  Erdal, Vaage Øie, Oppegaard,  & Westlund,
2019; Westlund, 2014). As a result, publishing across platforms can be
something journalists do not need to think about in their daily practice,
nor something they have wide awareness of or talk about. In extension
of this, practice-oriented researchers studying routines among journal-
ists in newsrooms may well not see concrete practices associated with
mobile devices taking place (Westlund & Ekström, 2020), and thus not
highlighted in studies of appropriation and normalisation of technologies
into newsrooms (Coddington, 2014; Djerf-Pierre, Ghersetti, & Hedman,
2016). We should add that many industry associations have devoted pan-
els to mobile in their conferences and workshops, and in their media
innovation work, albeit its role and significance does not necessarily sur-
face in everyday newswork. Ultimately, mobile news in the journalism
sector may have been largely invisible to digital journalism scholars con-
ducting ethnographic research, utilising interviews, mapping affordances,
analysing content on websites, and so forth.

2.2.2 Pro-innovation bias
Innovation as a concept refers to both the development and the implemen-
tation of something “new”, which can be processes, products, services,
and other things (Storsul & Krumsvik, 2013b). There is a pro-innovation
bias associated with all three aspects of our framework that rests on fun-
damental drivers of capitalism where, in order to maintain a competi-
tive advantage, companies and nations have incentives to continuously
develop their products, services, and so forth. Such drivers are continu-
ously changing society and market sectors. We see this in the ways news
publishers began experimenting with and innovating for the World Wide
Web a quarter of a century ago, developing different approaches to, and
practices for, online journalism. We also see this in the way media man-
agers and journalists in the sector, as well as scholars in digital journalism
studies and beyond, have repeatedly discussed that legacy news media
in the Western world (especially those formerly known as newspapers)
essentially “need to” innovate (Aitamurto & Lewis, 2013; Pavlik, 2013;
Storsul & Krumsvik, 2013a). Scholars from diverse fields have focused on
 The definitions 23
“innovation” as a concept and object of inquiry, even as they also found
a great deal of heterogeneity and uncertainty surrounding the bounds of
this concept (Bleyen, Lindmark, Ranaivoson, & Ballon, 2014).
However, common denominators have emerged, and these involve
developing something new, possibly by combining different parts previ-
ously held apart (Storsul & Krumsvik, 2013b). While companies in differ-
ent sectors often strive towards innovation, scholars have shown it may
well not be a solution, despite tremendous investments (Seelos & Mair,
2012). Innovation has been problematised by several researchers with
a footing in digital journalism studies, arguing (in similar terms) that-
“innovation, as a general concept, suggests creativeness and success in a
competitive environment, and is popularly held as a holy grail, something
for which to strive and claim as a source of pride” (Westlund & Lewis,
2014, p. 14). A qualitative study with 39 representatives from news pub-
lishers across numerous countries suggests that members of the journal-
ism sector have been overly obsessed with innovation related to “bright,
shiny things”, and focused less on developing long-term strategies for
sustainable innovation (Posetti, 2018).
Scholars have repeatedly advanced rhetorical and normative assump-
tions around how digital technology may either save or kill the role of
journalism and news, and with them, news organisations (see also chap-
ter  6). Essentially the pro-innovation bias mechanism has to do with
practitioners in the journalism sector (B) focusing on innovation and
emerging technologies that they envision may help them overcome con-
temporary challenges. In addition, it also sees influential scholars taking
the lead in approaching “trending” and “innovative” objects of inquiry
such as social media platforms, and many others have followed suit, result-
ing in tremendous amounts of research publications (see also chapter 4).

2.2.3 Path dependency
What journalists and news organisations do is inexorably linked to their
culture and institutionalised routines: essentially the history of how they
do things. This results in path dependency. In other words, history con-
strains the actions taken by having carved out a path, limiting the ways in
which journalists and institutions approach emerging opportunities and
challenges. Such patterns are found across the journalism sector, with news
publishers displaying herd behaviours as they engage in vicarious learning
where they follow the moves by peers, including imitating industry lead-
ers. Studies have demonstrated how seemingly predestined approaches to
emerging technologies taking shape in one news organisation materialise
into something others in the sector take notice of and imitate, and when
24  The definitions
doing so they are further reinforcing specific paths of development in the
journalism sector (Boczkowski, 2010; Westlund, 2012).
Path dependency also surfaces when it comes to how scholars develop
and maintain their research agendas. Imagine yourself as a young per-
son embarking in academia to pursue a PhD, getting further training
in theories and methods in the discipline. You read and digest massive
amounts of research to develop solid research reviews that lead to iden-
tifying important research questions, which you study empirically. You
develop expertise in that area, and with those methods and theories. If
you succeed in academia, you specialise further via various opportuni-
ties, including post-docs, tenure-track positions, research projects, and so
forth. We dare say few scholars renew themselves substantially, by which
we mean develop expertise across multiple theories, multiple methods,
and multiple objects of inquiry. Some are making steps towards renewal,
whereas some essentially build up their track record by repeatedly apply-
ing the same theories and methods for the study of changing patterns
(for example by conducting annual surveys or content analyses). These
scholars are adding to a reinforcement of scholarly path dependencies
influencing the routes embarked by others.

2.2.4 Addressability
The final mechanism, associated with (in)visibility, focuses on what we
refer to as addressability. We can address this through the same A (soci-
ety), B (sector), C (scholarship) framework. Essentially, this has to do with
the epistemological challenges that arise when trying to develop research
(C) that addresses objects of inquiry in society (A) or the sector (B).
While different theories can be used for developing research into a spe-
cific object of inquiry, some theories have become more widely used
than others by repeatedly being applied to the study of changing con-
ditions in the journalism sector. There is thus a body of literature that
demonstrates the addressability of such theories, which other scholars can
then build on. However, sometimes well-established theories which were
developed in a mass media era are criticised for not harmonising well
with conditions and patterns in the contemporary mediascape. There
are also theories and traditions found in the humanities, cultural stud-
ies, feminist critiques, postcolonial perspectives, and so forth that have
relevance, but which few scholars have pushed forward in this context.
Nevertheless, digital journalism studies is an interdisciplinary field break-
ing new terrains, advancing original research into areas that have never
been addressed before. In doing so, it may not always be self-evident how
to approach these areas theoretically, and while there is a magnitude of
 The definitions 25
theories (see chapter 5), there are also studies published without theoreti-
cal frameworks.
Another aspect of addressability has to do with research designs and
methods and the epistemic knowledge claims scholars are making based
on their empirical studies. In digital journalism studies and beyond there
has been much research about “social media” over the past decade, many
of which focus on Twitter (partly because the journalism sector has
focused relatively much on Twitter as a platform). The Twitter API and
the data which can be accessed through it have made Twitter a more
accessible platform compared to other social media platforms (though
GDPR regulations have resulted in reduced access, particularly in the
European Union). As a result, while scholars can take advantage of sound
methodologies that allow for the study of Twitter, they must simultane-
ously remain careful when making knowledge claims and avoid transfer-
ring explanations born of analysis of Twitter data onto a wider range of
social media use. We conclude that addressability in terms of easily acces-
sible data strongly influences patterns of research publications. We return
to this problem in chapter 7 (section 7.2.3).

2.3 Turning to thematic clusters in Digital Journalism

Taking as a point of departure the interdisciplinary nature of digital jour-
nalism studies, and the empirical analysis of disciplinary perspectives pre-
sented in the first chapter  (section  1.3.1), the next two chapters will
present an analysis of research published in Digital Journalism from its
inaugural issue in 2013 until issue 4 in 2019. We build further on research
previously undertaken by Steensen et al. (2019) and have analysed and
sthematised the keywords from a total of 343 original articles published
in Digital Journalism (see online appendix for details on methodology).
Table  2.2 charts the contours of the digital journalism studies field
across a wide range of distinct objects of study, grouping distinct keywords
together into so-called thematic clusters. In total there are eleven thematic
clusters, representing 65 percent of all keywords used in articles. Three of
these thematic clusters are most dominant; technology (17 percent), platform
(13 percent), and audience (10 percent). Further, while scholars commonly
use theory or method among their keywords, these keywords offer lit-
tle insight into more specific contributions. Thus, there are only nine
thematic clusters building on more specific objects of inquiry, and the
remaining six account for only two or three percent each.
Performing assessments of keywords is useful for establishing an over-
view of patterns in research publications but can quite naturally only
take us so far. There are clearly limitations here compared to the more
26  The definitions
Table 2.2 Thematic clusters of keywords used in the 343 original articles published in
Digital Journalism from volume 1, issue 1 (2013) to volume 7, issue 4 (2019)
that contained keywords.

Thematic cluster  No. of keywords Percent of all keywords

Technology  378 17%

Platform  302 13%
Audience  232 10%
Methodology  109 5%
Theory  103 5%
Business  71 3%
Region  73 3%
Genre  63 3%
Philosophy/epistemology  68 3%
Visual  36 2%
Professionalism  43 2%
Sum 1478 65%
Sum other keywords 779 35%
Sum total 2257 100%

time-demanding assessment of reading and assessing full-length arti-

cles. Chapters 3 and 4 will therefore assess and discuss more closely the
important patterns and findings from articles associated with the two
most dominant thematic clusters; technology and platforms. Each of these
chapters will link the assessments to discussions of research into audiences.
This is the third most dominant thematic cluster, and is closely interre-
lated to the two other.
Importantly, we do not claim to offer a systematic review of all litera-
ture in the field since digital journalism scholars produce more than 1000
journal articles per year.
3 The technologies
Unpacking the dominant object of study
in Digital Journalism Studies

Imagine for a moment a legacy news media organisation and how their
social actors continuously are trying to make sense of digital technolo-
gies, advancing their production and distribution of news. Imagine how
this organisation and their journalists, who are used to traditional news
reporting techniques, approach the opportunities and challenges at hand
when it comes to data journalism. What explicit and tacit knowledge do
they need, and how should they sorganise data journalism? Should they
focus on developing sspecialised teams, or enhancing generalists’ knowl-
edge about data journalism? What networks of sources (Ettema & Glasser,
1985) can they rely on in terms of datasets, and what truth claims can
they make? How does the organisation approach and make use of audi-
ence analytics – the technological systems tracing-patterns of behaviour
from digital platforms – and generate metrics from them that can then be
acted upon? Will analytics and metrics help the journalists to understand
better what news material engages their readers, facilitating conversion
into subscriptions? Moreover, can the legacy news organisation’s pub-
lisher appropriate technologies for automated news distribution and for
automated personalisation, and so forth?
These are the sorts of questions that digital journalism studies scholars
have addressed throughout the 2010s as part of what we refer to as the
technology thematic cluster comprising 378 different keywords associ-
ated with the articles published in Digital Journalism from issue 1, 2013, to
issue 4, 2019. This chapter focuses on a great number of these keywords,
joining up this focus with a brief discussion of research associated with
the third largest thematic cluster, Audiences. The technology thematic
cluster encompasses a wide array of keywords that have been dealt with
by researchers over the years. We can proceed alphabetically. Starting
with A, we find emerging research into the role of ad-tech in journal-
ism (Braun  & Eklund, 2019), studies into how analytics (and metrics)
have been appropriated and is being used in newsrooms (Carlson, 2018a;
28  The technologies
Zamith, 2018), from the perspective of audience-oriented editors (Ferrer-
Conill & Tandoc, 2018), as well as analytics companies (Belair-Gagnon &
Holton, 2018). We then find studies focusing on the role of algorithms for
news (Thurman, Moeller, Helberger, & Trilling, 2019; Wallace, 2018),
the emergence of artificial intelligence in journalism (Broussard, 2015; Stray,
2019), accountability, and, relatedly, transparency (Broersma  & Harbers,
2018; Diakopoulos, 2015). Down the alphabet, we find this thematic
cluster also comprises studies looking into the closely related area of bots
(De Maeyer & Trudel, 2018; Lokot & Diakopoulos, 2016), civic tech and
datafication (Baack, 2018), design (Petre, 2018), digital and web archives
(Severson, 2018; Weber & Napoli, 2018), digital surveillance (Thurman,
2018; Waters, 2018), drone journalism (Adams, 2019; Holton, Lawson, &
Love, 2015), hacks, hackers, and technologists (Baack, 2018; Lewis & Usher,
2014; Lewis & Westlund, 2015a), machine learning (Broersma & Harbers,
2018; Watanabe, 2018), and many more.
The intersection of journalism and data is a common denominator
across several keywords and studies published since the inception of the
journal Digital Journalism. In its inaugural issue we find one article offer-
ing rich discussions on the strategies for doing research with Twitter
data (Vis, 2013), another focusing on developing methods for automat-
ing content analysis of news content (Flaounas et al., 2013), and a third
positing a model of journalism incorporating automated journalism, and
the study of “the human actors and technological actants performing the
work, vis-à-vis the degree to which content and services are platform-
agnostic or coupled with specific affordances and logics” (Westlund,
2013, p. 19). In 2013, Digital Journalism’s inaugural year, the journal also
published its first study of analytics and metrics – a case study of how
Al Jazeera English engaged in activities for tracking and analysing their
audience through technological actants (Usher, 2013). That article shows
how Al Jazeera English’s organisational culture shaped the ways in which
journalists use analytics and metrics in practice, and how they understand
them. In 2014, a pioneering study of data journalism at seven legacy
media companies in Sweden identified time and proper training as the
main resource constraints for actors to effectively carry out data journal-
ism (Appelgren & Nygren, 2014).
Within the technology thematic cluster, there are three key themes of
research: 1) Data journalism, 2) Analytics and metrics, and 3) Algorithms and
automation. These three themes show that this interdisciplinary field has
turned towards studying the evolving and complex interplay between
digital technology and journalism, using different theories, methods,
and ways of thinking about these relationships. Next, we highlight some
emerging patterns of research in each of these areas in recent years.
 The technologies 29
3.1 Data journalism
Data journalism works within an epistemological tradition where journal-
ists turn to data as a source for reporting about certain phenomena. It is
often seen as a “pure” way of accessing information, but while raw data
may give the impression of objectivity, this is an oxymoron (Gitelman,
2013). Many scholars argue that any type of data has its biases and limi-
tations (e.g., Carlson, 2019; S. C. Lewis & Westlund, 2015b; Steensen,
2019). Studies in the field have found that data journalists themselves
take on a role of translating technical and abstract knowledge so that their
(lay) audience can understand what stories data tell (Boyles  & Meyer,
2016). In 2015, Digital Journalism published a special issue focusing on the
intersection of data and journalism (Lewis, 2015) that has had a formative
impact on much subsequent work. As of early February  2020 the five
most cited articles from that issue were: “Algorithmic Accountability”
(Diakopoulos, 2015) with 141 CrossRef citations, “Clarifying Journal-
ism’s Quantitative Turn” with 126 (Coddington, 2015), “The Robotic
Reporter” with 85 (Carlson, 2015), “Big Data and Journalism” with 72
(S. C. Lewis & Westlund, 2015b), and “Data-driven Revelation” with 50
(Parasie, 2015).
That special issue features several articles examining how social actors
approach data journalism, related for instance to the epistemological con-
cerns of such work (Lewis & Westlund, 2015b), the tensions relating to
historical developments regarding data and journalism (Anderson, 2015),
and investigative journalism (Parasie, 2015). It also offers insights into the
formative approaches to data journalism in Belgium (De Maeyer, Libert,
Domingo, Heinderyckx, & Le Cam, 2015).
Scholars and practitioners have envisioned data journalism as ena-
bling new and improved journalistic investigations and reporting prac-
tices. There are indeed multiple significant and successful data journalism
endeavours that reflect this, such as the reporting on the Panama Papers
(Carson & Farhall, 2018). Nevertheless, it remains a challenge for jour-
nalists and news organisations trying to integrate data journalism into
everyday routines of news reporting, since data journalism requires dif-
ferent sorts of expertise and work flows. Studies continue to show that
the relative proportion of data journalists is small and journalists often
struggle to access relevant and reliable data to use in their reporting
(Appelgren, Lindén, & van Dalen, 2019; Porlezza & Splendore, 2019). In
many places, journalists struggle to get hold of data as authorities restrict
access, including the ever-present risk of imprisonment or even murder
of journalists scrutinising such regimes (Lewis & Nashmi, 2019). Several
recent studies show that data journalists at prominent media, including
30  The technologies
but not limited to The New York Times and The Washington Post, often use
small data sets and seldom carry out advanced data analysis in their eve-
ryday news reporting (Anderson  & Borges-Rey, 2019; Zamith, 2019).
This highlights not only limitations in terms of resources and expertise
in newsrooms, but also an adaptation to the envisioned competence of
the readership (Anderson & Borges-Rey, 2019) and their interest in par-
ticipation (Palomo, Teruel, & Blanco-Castilla, 2019). Importantly, it has
not only been data journalists participating in producing data journal-
ism, but actors external to the field as well. Studies have found that civic
technologists in several continents have important skillsets, which can
enable data-based journalism, but also a sense that journalists are unable
to do what they should do with data (Cheruiyot, Baack, & Ferrer-Conill,
2019; Cheruiyot & Ferrer-Conill, 2018). Finally, researchers have shown
that while attention is now paid to this practice, data journalism is by
no means a new journalistic practice, but one that has developed well
over time (Anderson, 2018), and one that has emerged in dialogue with
technologists (Hermida & Young, 2019; Usher, 2016). Some argue this
sub-field has started to mature (Appelgren et al., 2019), and we also find
important advancements in data journalism scholarship from the global
south, in the form of edited books with contributions by a diverse set of
scholars (Mutsvairo, Bebawi, & Borges-Rey, 2019) as well as by scholars
and practitioners (Krøvel & Thowsen, 2019).
To sum up briefly, we can see that between 2013 and 2019 there was
a burst of scholarly interest in different forms of data journalism. There
was also a sense of optimism involved, including in the journalism sector,
which envisaged an important appropriation of data journalism in news-
rooms. While data journalism scholarship has evolved, and expanded
across geographical terrains, a significant body of work has found that, in
practice, data journalism requires substantial expertise, access to datasets,
and much more. While data journalism can help to enrich journalism, it
is likely not a major component in everyday news reporting.

3.2 Analytics and metrics

Analytics refer to the technological infrastructures, systems, and tools
for gathering and analysing metrics about audiences and their behaviours
(Zamith, 2018). Such metrics, which trace data such as page impressions,
time spent on pages and sites, completion of article reads, and so on, are
being used to guide editorial processes and decisions such as what types
of leads journalists prioritise (Chua & Duffy, 2019), and how online edi-
tors and algorithms prioritise the exposure different news articles receive
(Tandoc, 2019a; Zamith, 2018). Throughout the 2010s there has been a
 The technologies 31
surge in how journalists and newsrooms gravitate towards using analyt-
ics for looking at different sorts of editorially oriented metrics. Analytics
companies constantly experiment with their products and services to fit
the changing demands of news publishers and influence news production
practices (Belair-Gagnon & Holton, 2018).
Some would argue that in the past, reporters mainly relied on their
gut feeling to make decisions about what stories to pursue and publish
(Schultz, 2007). Nowadays there is a wealth of relevant data that can inform
news publishers’ decision-making, but one should not overlook the many
ways reporters of the past were informed about their readers’ news con-
sumption. Some news publishers have a long tradition of measuring their
audiences by for instance conducting focus group interviews and commis-
sioning surveys and opinion polls, much as researchers do. For example,
in Sweden the national newspaper association sponsored the Newspaper
Research Programme at the University of Gothenburg from 1979 to 2011,
funding studies into news consumption and attitudes. This contributed sig-
nificantly to the establishment of a research institute that conducts annual
cross-sectional surveys, similarly to how the Reuters Institute for the Study
of Journalism has later built the comparative survey project with much
funding coming from Google, which has renewed sponsorship into the
2020s. By comparison, and telling for the publisher and platform indus-
tries, funding from the national newspaper association ended in 2011 in
part because of the worsening financial situation for newspaper companies,
and in part because of the growth in options for analysing audiences.
At the same time, news publishers trying to find ways of securing
readers and revenue were also subscribing to services from industry data
providers like ComScore and Kantar, which gave them data reflecting
the behaviours and attitudes of their audiences. These companies also
began to offer analytics that could fetch and put on display data about the
behavioural digital footprints people were leaving behind as they used the
web to navigate to and from news articles. For instance, when analysing
mobile media throughout 2008–2010, Swedish journalists, technologists,
and business people combined Kantar data with services like Google
Analytics to analyse trace data about page impressions, time spent, and
audience engagement in comment fields, alongside metrics from Apple
about the number of app installs. This allowed them to evaluate how their
native mobile news app was performing (Westlund, 2011). Throughout
the 2010s, on top of using industry data from panels and such for their
business intelligence, the journalism sector has also increased its emphasis
on the use of analytics and metrics. Companies like Kantar have devel-
oped services where they combine different data, such as survey data and
metrics from websites.
32  The technologies
Research into how newsrooms approached analytics and metrics
began to emerge early in the 2010s. At the beginning of the decade,
pioneering research about developments in the US were published by
scholars based either in the US (C. W. Anderson, 2011a, 2011c; Lee,
Lewis, & Powers, 2014) or Singapore (Tandoc, 2014). There was also
early contributions into this area coming from Sweden (Karlsson  &
Clerwall, 2013). While, as noted earlier, Digital Journalism published a
pioneering study into how Al Jazeera uses analytics and metrics in its
first year (Usher, 2013), it was another two years before a second article
on this topic was published in the journal. In this article, Tandoc and
Thomas (2015) raised three concerns: 1) viewing the audience as disag-
gregated segments, 2) failing to distinguish between interest and public
interest, and 3) arguing against journalistic autonomy and romanticising
the audience. There is clearly a strong link to business activities when
it comes to developing technological actants with the purpose of meas-
uring and analysing audiences. In their study on so-called engineering
technologies, Slaček Brlek, Smrke, and Vobič (2017) discuss that news-
rooms had an inferior role compared to the business people in defining
and implementing goals for these technologies. To continue, in Dwyer
and Martin’s (2017) article on news sharing and social media analytics,
the authors discuss in a critical fashion how news media have become
dependent on such analytics and how this can influence news diversity.
Another article draws on Bourdieu’s field theory to position journalism
as a field in which audience analytics is a trend and driving force with
implications for journalism (Q. Wang, 2018).
Belair-Gagnon and Holton (2018) offer insight into how representa-
tives for web analytics companies see their role and function in news
production. They conclude that web analytics companies do not take
responsibility as journalists, and largely work towards continuous devel-
opments of their analytics through experimentation. Moreover, based on
substantial ethnographic work at an analytics start-up, Petre (2018) finds
that they engineer and design their analytics to match editorial routines
and judgments, while turning down the prevalence of managerial influ-
ence. Another ethnographic study of a company specialised in global
audience engagement services finds that there is no agreed-upon standard
for audience engagement and therefore the company struggles to quan-
tify its value proposition (Nelson, 2018). These articles were part of a
2018 special issue in Digital Journalism focusing on what the guest editor
refers to as measurable journalisms, consisting of eight diverse dimensions
(Carlson, 2018a, p. 409).
This special issue also included studies reporting on how audience-­
oriented editors engage with the metrics produced by analytics in different
 The technologies 33
activities geared towards stimulating audience engagement (Ferrer-Conill &
Tandoc, 2018). A year later a cross-cultural study into the use of analytics
and metrics in Zimbabwean, Kenyan, and South African newsrooms was
published, exemplifying the changing approaches to how news organi-
sations analyse and interact with their audiences. All in all, the authors
argue that these newsrooms engage in so-called analytics-driven journalism
(Moyo, Mare, & Matsilele, 2019).
Studies are also signaling how journalists and audience-oriented edi-
tors often have to work with metrics produced by third-party analytics
companies, such as Google (Analytics), Facebook (CrowdTangle), Chart-
beat, etc. Some of these offer unique metrics, and there are different
advantages and disadvantages to them. Journalists, editors, and others in
the newsroom are naturally interested in knowing as much as possible as
part of their newswork, and thus would want to be able to combine ana-
lytics that gather metrics from their proprietary platforms with analytics
like CrowdTangle to study audience engagement on Facebook. Powers
(2018) discusses this, examining how journalists across a diverse set of news
organisations define, measure, and discuss the potential impact of metrics
on their journalistic work. Some turn to measures focusing on whether
the published news has entered public discourse, public policy, or public
awareness, while others simply rely on the metrics coming from audience
analytics companies. In another case study, Blanchett Neheli (2018) found
the newsroom is very much oriented to traffic-based metrics, which may
negatively affect their ability to maintain journalistic standards. A group of
American scholars selected a couple of common metrics and surveyed US
newsworkers about how useful they found these to be. They concluded
that the metrics were most useful for enacting the newsworkers’ consumer
role orientation (Belair-Gagnon, Zamith, & Holton, 2020).
Analytics and metrics have certainly gained prominence in many con-
temporary newsrooms and have changed newswork and routines for
many journalists. It is an important area of research in digital journal-
ism studies, and there are many more publications in other journals also
addressing it, not to mention a recent book which has charted how this
line of research is maturing, showing the significance of work in this
field in a holistic way for the first time (Tandoc, 2019a). Throughout
the 2010s there were multiple qualitative studies raising questions con-
cerning whether analytics and metrics influence editorial judgment. By
the end of the last decade, the answer in many cases was “yes”. In this
context we should recall that such questions may well generate responses
guided by normativity and role perceptions, and reflect scenarios where
journalists do not want to admit to how their news judgment might be
influenced by metrics about their audiences’ behaviours and needs. We
34  The technologies
should also bear in mind that most studies are conducted in a Western
context, making the comparative study of analytics-driven journalism in
African countries all the more important for breaking from that limited
scope (Moyo et al., 2019). It is worth noting that, when looking beyond
Digital Journalism, we find additional journal publications into analytics
in the global south. One article focused on how correspondent in sub-
Saharan Africa, mainly Nairobi and Lagos, diverge substantially in their
approaches to analytics and metrics (Bunce, 2015), and another revealed
how metricsare used by web and traffic managers in Kenyan newsrooms.
Also several studies into analytics and metrics have been carried out in
Singapore (Duffy, Tandoc,  & Ling, 2018), South Korea (Yang, 2016),
the Philippines (David, Tandoc, & Katigbak, 2019), as well as mainland
China (Zhu et al., 2019).
Going into the 2020s, the journalism sector will likely work more
with their news organisations’ proprietary platforms, and the analytics and
metrics associated with them. With many news publishers now getting
most of their revenues from readers rather than from advertisers, there
must be a shift: metrics data does not derive its main importance in rela-
tion to advertisers and reach, but rather in providing insights that journal-
ists, editors, and other news workers can use when they make editorial
judgments, personalise news distribution, and so forth. While much is
known about how analytics and metrics shape news production and con-
tent, less is known about how individual journalists are affected and navi-
gate this. Ultimately, scholars should also advance more research into all
eight dimensions of measurable journalism, including the material aspects
of analytics (and the innovations taking place in this realm) and how this
potentially assists the economic aspects, but especially reader revenue.

3.3 Algorithms and automation

Algorithms are everywhere and they have lots of power. However, and
as Bucher (2017a) argues, they do not have instincts and are not only
technical, but also cultural, economic, social, and political. From 2013
to 2015 there were only a handful of articles in Digital Journalism analys-
ing or discussing the role of algorithms and automation  per se. These
included a model of journalism focusing on human-machine interac-
tion referenced above (Westlund, 2013), and an article presenting the
testing of a prototype software system that uses artificial intelligence to
assist reporters in processing large amounts of data and detecting leads for
investigative journalism (Broussard, 2015). The remaining articles were
published in a special issue focusing on journalism and big data (Lewis &
Westlund, 2015). One article in this issue advanced understandings of
 The technologies 35
automated journalism through a study of such practices at Narrative Sci-
ence, a tech company using artificial intelligence (AI) to convert data
into narratives that are easy to understand (Carlson, 2015), while another
analysed how computational journalism influenced the creation and dis-
semination of crime news at The Los Angeles Times and their pioneer-
ing “Homicide Report” project (Young & Hermida, 2015). The special
issue also featured an article focusing on algorithmic accountability, and
the power structures, biases, and influences associated with employing
algorithms in journalism (Diakopoulos, 2015). To date (early 2020), this
is the most cited article of all time in Digital Journalism. In 2016, Diako-
poulos contributed further to this area of research in a coauthored article
focusing on news bots in social media – those accounts that are not man-
aged by social actors but technological actants – which are automated to
participate in news distribution (Lokot & Diakopoulos, 2016).
As research in this area developed, a focus on efficacy and ethics
became salient. Scholars have studied the distinct nature of technology
by assessing more specifically technological actants. One article looked at
how these issues addressed the case of technological actants developed to
have a “nose for news”. In this article, the authors critically examined the
SocialSensor application developed in a EU project, finding that while
it indeed enabled sourcing and verification through social media, it also
reflected certain biases, such as towards using men as sources (Thurman
et al., 2016). Further, in a mapping of current qualifications of techno-
logical actants for algorithmic selection and production of news texts
through natural language generation, Dörr (2016) showed that while
there are few companies on the market, the technologies as they stand
are “good enough” to be used. However, another study finds that jour-
nalists with personal experiences of working with software for automated
journalism raised concerns about sourcing and capacity to identify news
stories (Thurman, Dörr, & Kunert, 2017). Both of these studies discuss
how the social actors interviewed, in journalism and at software compa-
nies for automation respectively, expect automated journalism to expand
further. As a counter-narrative to the dominant discourse around auto-
mation, Lindén (2017) argued that we should ask, “Why are there still so
many jobs in journalism after decades of newsroom automation?”
Scholars of digital journalism studies have continued to gear their
efforts towards this area of algorithms and automation. It is an expanding
field marked by more heterogeneity as emerging developments are stud-
ied. Digital Journalism has published empirical studies on how social actors
assess changing structures pushing automation forward, but also on the
importance of journalists’ agency, their attitudes and skillsets (Wu, Tan-
doc, & Salmon, 2019). Other studies focus on attribution regimes and
36  The technologies
bylines in automated journalism (Montal & Reich, 2017), and the chal-
lenges for adopting and communicating algorithmic transparency (Bodó,
2019). One experimental study found that people do not expect human-
written news to be more credible, and prefer automated news when it
comes to credibility (Haim  & Graefe, 2017) whereas another experi-
mental study found that audiences assess news produced by technologi-
cal actants as less credible than news produced by (human) journalists
(Waddell, 2018). A different line of research deals with how algorithms
and code are utilised to prioritise the exposure and distribution of news
for different platforms and channels (Weber & Kosterich, 2018), and also
algorithmic selection guided by individual news consumption routines.
A  cross-cultural survey study (26 countries) found that citizens favour
algorithmic selection compared to editorial curation by social actors
(Thurman, Moeller, et al., 2019).
The research into algorithms and automation in news was drawn
together in 2019, when a special issue on this theme was guest edited by
Thurman, Lewis, and Kunert (2019). The editors discuss how the articles
in the special issue advance earlier research focusing on algorithms and
automation in news, and how this can help to develop digital journalism
by taking a broader approach to understanding the technologies involved,
discussing their diverse uses as well as the challenges involved for practice
and values in utilising AI. This issue highlights that there are certainly
great challenges for using AI in investigative journalism, restricted by
factors such as the affordances of technological actants, costs, access to
data, accuracy standards, and so forth (Stray, 2019). It also foregrounds
a challenge concerning the tension between automated and human-­
produced journalism, perhaps most notably when it comes to ideals
such as autonomy, objectivity, and public service (Milosavljević & Vobič,
2019). Objectivity, or more specifically an idea of a so-called mechanical
objectivity, that comes with the discourse of automation, however, also
warrants critique. Carlson (2019) argues this discourse risks replicating an
argument that technology can offer more objective representations than
humans can.
Further in this issue, we gain insights into how the BBC has set in
motion a series of news bots operating on their website as well as third-
party platforms not owned or controlled by them, with the goal of
reaching audiences they struggle to reach via their proprietary platforms
( Jones & Jones, 2019). A study of ABC’s news bot, in the same issue, finds
that both journalists and the public express an appreciation for the forms
of news delivery it enables, and discusses concerns relating to control over
data and how ABC depends on third-party platforms (Ford & Hutchin-
son, 2019). This highlights differing priorities for different companies
 The technologies 37
involved in these processes. As Bodó (2019) reports, news publishers
work with personalisation with goals such as showing relevant news con-
tent and selling subscriptions, whereas platform companies such as Face-
book strive towards substantial engagement on their platform so they can
monetise attention through advertisements. Extending this, Helberger
(2019) offers an interdisciplinary conceptual framework for news rec-
ommenders tailored to different democratic models. On a related note,
other articles have also made theoretical advances in the journal, bringing
the field of human-machine communication into dialogue with digital
journalism studies for the study of technological actants as active message
sources and not only mediators of communication (Lewis, Guzman, &
Schmidt, 2019).
Going into the 2020s, there are many reasons to suggest algorithms and
automation will become increasingly intertwined with the production
and distribution of news. The accuracy of automated journalism depends
on the quality of the data it builds on, as well as the level of complexity
inherent in the issues and events that technological actants are to create
news and make truth claims about. The type of automated journalism
evident in the 2010s involved mainly reporting on topics and events that
are relatively easy areas for journalists and technologists to develop auto-
mated journalism for, such as fiscal reports and sports journalism. They
typically develop predefined algorithms based on news values, which are
matched with the dataset, generating news stories fed to news distribu-
tion platforms via an interconnected content management system.
There are massive investments being made in AI. However, while there
is currently a great deal of buzz about the potential of AI, we must once
again be cautious and critical about this future, rather than falling into
traps of thinking that this is the next technology that will save journal-
ism. Most investments in AI, for instance, can be traced to Asia and more
specifically to China. A power concentration in AI capacity can become
problematic if newsrooms around the world become overly dependent
on AI services provided by companies in specific countries. In extension
of this, the Chinese authorities could gain influence over significant data
and automated news flows taking place across the world.
Editorial-facing innovation in AI in the news industries will depend,
however, on the leaps taken by the industry as a whole. Appropriating AI
into news reporting raises key questions of accountability and libel (S. C.
Lewis, Sanders, & Carmody, 2019; S. C. Lewis & Westlund, 2016). News
publishers are responsible for the news they publish, and it can become
very problematic if they use AI technology to report misinformation,
especially if the ownership of such non-proprietary technology is associ-
ated with political or economic interests.
38  The technologies
Already today news publishers and journalists could take advantage of
AI by making use of advanced tools for analysing open-source Earth data
provided by Copernicus (EU funded) and NASA. This data is accessible,
overcoming a main problem in data journalism, although the analysis of
data still requires certain expertise (Appelgren et  al., 2019), while the
combination of data and images/videos allows for major truth claims,
with an implied mechanical objectivity (Carlson, 2019) because they rep-
resent visual representations of reality, and a sense of seeing things how
they are. Obviously, those with ill intents can easily manipulate images
and videos. Nevertheless, recent research has revealed how reporters suc-
cessfully have used satellite images in investigative journalism (Seo, 2020).
Moving towards using such online repositories of satellite images, big
data, and visually oriented user interfaces can help journalists advance
reporting on complex and urgently important matters such as climate
change, and possibly offer credible reporting that even climate change
deniers will embrace. Clearly, this also open doors for avenues digital
journalism studies scholars have hardly yet explored.

3.4 Concluding discussion
This chapter has dealt with three specific areas of research that are cen-
tral components of the technology thematic cluster in Digital Journalism:
data journalism, analytics and metrics, and algorithms, automation, and
news. Our assessment demonstrates how research into journalism and
technology has mainly focused on how journalism changes in relation to
data and algorithms, with a significant amount of research into how data
(analytics and metrics) drives journalism, how data (datasets, visualisations
etc.) becomes part of journalistic practice, and also how algorithms are
used for producing and distributing news.
We would like to point to some patterns from our assessment in light
of the so-called 4 A’s: social actors, technological actants, audiences, and
activities (S. C. Lewis & Westlund, 2015a). We find that relatively few
studies have focused on the technologies per se, and the agency inscribed
into the technological actant (exceptions include Diakopoulos, 2015 and
Helberger, 2019). Scholars have typically studied either how social actors
approach emerging technologies, such as journalists appropriating ana-
lytics in their daily practice, or they have studied audience attitudes to
automated journalism or personalised news recommenders. While it is
very challenging and time-consuming to adopt holistic approaches to
the study of actors, actants, and audiences, the stream of more focused
studies on only one of the A’s means that different studies must be con-
joined to attain a broader view. More holistic approaches are possible
 The technologies 39
when journals commission special issues. This is the case with some of
the research areas discussed in this chapter, such as the Digital Journal-
ism special issues into analytics and metrics (Carlson, 2018a) and data
journalism (Appelgren et al., 2019) as well as algorithms and automation
(Thurman, Lewis, et al., 2019).
Let us end this chapter  by discussing the intersection of technology
and audiences. The research we have reviewed and discussed would not
by tradition fall into a classification of audience research. However, an
underlying driving force for the research developed on analytics and met-
rics has to do with how analytics as a form of technological actant can
be developed and used for assessing behavioural patterns among audi-
ences and, in turn, how they are diffused into routines of news work.
Researchers can easily build on this line of research to study the analyt-
ics infrastructures itself, as well as turning their attention to the metrics
themselves. As for research into data journalism, the focus generally lies
with social actors and their newswork. As with the potential opportuni-
ties in analytics research, some scholars have surfaced audience-oriented
questions relating to their interest in participation (Palomo et al., 2019)
and their competences (Anderson  & Borges-Rey, 2019). Similarly, the
research into automation and algorithms has raised important questions
about how audiences may be approached through news recommender
systems depending on different democratic models (Helberger, 2019).
Associated with such developments, we find audience-oriented research
studying attitudes to news personalisation (Bodó, Helberger, Eskens, &
Möller, 2019), and the role of humans and algorithms in selecting the
news that is exposed (Thurman, Moeller, et al., 2019). In advancing this
argument, Guzman (2019, p. 1187) makes a call for researchers to further
mobilise efforts for the study of audiences in relation to technologies of
automation such as news recommenders and chatbots. In this context, we
envision further research into what Bucher (2017b) refers to as algorith-
mic imaginary among citizens would be worthwhile to advance knowl-
edge into how citizens imagine algorithms operate with news among
publishers and platforms. As we move into the 2020s, digital journalism
studies will continue to advance research into technology, and will likely
include audience-oriented research into algorithms, news recommend-
ers, chatbots, and algorithms, among other topics.
4 The platforms
Distributions and devices
in digital journalism

For much of the 20th century, legacy news media reached mass audiences
via printed newspapers or radio or television broadcasting. Legacy news
media companies have, by tradition, owned and controlled their means
of distribution. The printing press has constituted a backbone in news-
paper companies, and frequency licences and news broadcasting studios
have been central to broadcasting companies. With the World Wide Web
news publishers extended their news distribution by way of setting up
proprietary news sites, and eventually also turning to other devices such
as smartphones and tablets. However, around 2007 and 2008 news pub-
lishers started losing control over how news was distributed, and became
increasingly dependent on non-proprietary platforms. Such platforms
were gaining significance for the ways people were accessing the news.
There was a massive orientation towards creating native mobile appli-
cations for smartphones. News publishers essentially made their news
accessible for mobile devices (Westlund, 2013) and the public has since
increasingly moved towards mobile news consumption (Nelson  & Lei,
2018; Newman et al., 2019; Westlund & Färdigh, 2015).
For those news publishers who did move to mobile apps for the iPhone,
they had to contemplate the requirements defined by Apple, including
sharing 30  percent of all revenues with them. Using Apple devices as
platforms also meant that news companies would feed Apple with met-
rics about news consumption. News publishers joined each other in herd
behaviour under mantras such as being innovative, and that they were
developing a presence in the mobile “ecosystems” where their (younger)
users were. With mobile ecosystems we refer to the mobile interfaces
established and largely facilitated by Apple and Android, enabling actors
to develop native mobile applications deemed to fit with user-­friendliness
and usability for smartphones (Goggin, 2009, 2020; Gómez-Barroso
et al., 2010). News publishers mobilised efforts for developing their brand
and their content within such mobile ecosystems. At the same time,
news publishers embarked on a journey where they became increasingly
 The platforms 41
dependent on third parties for their distribution, data, revenues, and so
on. We would like to clarify that this was not entirely new. The so-called
“walled gardens” for mobile devices (sites/portals for mobile devices con-
trolled by telecom operators), such as by DoCoMo in Japan and Telia
in Sweden, by the turn of the millennium were indeed precursors with
similarities. However, with the iPhone and the App Store, this gained
widespread momentum (Westlund, 2011, 2012). Such dependence on
third parties certainly did not stop at mobile ecosystems but extended
to platform companies, most notably Facebook, Google, and Twitter in
Western democracies. From a survey with Nordic news media managers,
we learn that the opportunities the managers saw as most important were
mobile news (73 percent), social media (68 percent), and tablets (65 per-
cent) (Stone, Nel, & Wilberg, 2010).
Throughout the 2010s there has been enormous activity in the jour-
nalism sector associated with the developments of mobile ecosystems and
global platforms. The activities of companies like Apple, Facebook,  and
Google have a lot of visibility and are closely followed by the media
and other companies, and they oftentimes gain a prominent place at vari-
ous work places as well as in the everyday lives of citizens. Returning to
our analytical framework introduced in chapter 2, section 2.2, we see that
not only is there issue visibility and innovation bias at play, but also practi-
cal addressability in terms of research designs. Correspondingly, a wealth
of research has emerged in digital journalism studies focusing on platforms
and digital devices. Much of this research has been marked by optimism
and how news publishers, journalists, and users can explore the emerging
opportunities, albeit that there are also some studies focusing on challenges
and problems arising concerning dependence, loss of revenues, and privacy.
This chapter focuses on two main areas of research into the platforms
thematic cluster: 1) Digital journalism and platforms, and 2) Digital jour-
nalism and digital devices. The first area aligns with the emerging and
increasingly common approach referring to platforms as digital interme-
diaries between different stakeholders in communication, entertainment,
news distribution, and so forth. The second area focuses on how digital
devices such as desktops and smartphones (and potentially also tablets,
smart watches, smart TVs, screens in cars, and others-), are developed
and appropriated in the context of the production, distribution, and con-
sumption of news. We link the research discussed to research on audience,
the third biggest thematic cluster.

4.1 Digital journalism and platforms

Journalism, and digital journalism, has evolved in parallel and also some-
times in tandem with the developments of the World Wide Web and
42  The platforms
platforms (Burgess  & Hurcombe, 2019). There are many digital plat-
forms, including but not limited to Facebook, WhatsApp, Instagram,
Twitter, WeChat, Telegram, Snapchat, Google, and Amazon. In this
review of research on platforms, we refer to a platforms as “a digital
infrastructure with affordances offering diverse kinds of information and
communication, as well as opportunities to produce, publish and engage
with content” (Ekström & Westlund, 2019b, p. 259).
This section features two sub-sections, each referring to different facets
in digital journalism and digital journalism studies: 1) building platform
presence, and 2) platform counterbalancing. Building platform presence is a facet
marked by a rather optimistic approach to platforms (especially social
media platforms) in both industry and research, oftentimes normatively
building on the assumption that news publishers and journalists should
build a platform presence by improving their expertise, by normalising
different social media into their routines, by developing social media pol-
icies, by producing unique content for social media or customising con-
tent they already have, by striving towards expanding eyeballs via social
media traffic, or by appropriating analytics tools provided by platform
companies (e.g.,  Google, Facebook, CrowdTangle, etc.). The list goes
on. The second facet, platform counterbalancing, is associated with publish-
ers seeking to balance their relationship with, and dependence on, plat-
forms that are non-proprietary to them. Such efforts surface in different
ways, for example when publishers are more cautious about giving news
content away for free, or even producing unique news content for plat-
form companies. We borrow the concept of platform counterbalancing
from Chua and Westlund (2019), who posited it based on their longitu-
dinal study of two Singaporean news publishers shifting their approach
to platform companies over time. Please note that we present these facets
as distinct from each other (in practice and in research) for the sake of
simplicity and clarity. These facets are not in a binary relationship, and
thus a publisher can simultaneously work on building platform presence
in some ways, and engage in platform counterbalancing on other ends.

4.1.1 Building platform presence

Let us rewind to the end of the first decade of the 2000s, when You-
Tube was a new phenomenon showing tremendous growth, when Apple
launched the iPhone and App Store, when Facebook rolled out and got
more international traction for its platform, and when Twitter spurred a
rapid growth in micro-blogging. All of these platform innovations were
receiving a good deal of attention from the public, from the media, and
also from companies. News publishers were, metaphorically speaking, on
a “mobile media train” about to leave the “platform”, a train which they
 The platforms 43
needed to be quick to catch if they were to maintain their relevance in
the emerging age of mobile media and news. News publishers have been
keen to build a presence on Apple and Android platforms for mobile
news, for strategic reasons of reaching out to specific segments as well as
to gain symbolic recognition (Westlund, 2011, 2014). Similar logics seem
to have been in play when it comes to how journalists and news pub-
lishers have approached social media platforms. They either needed to
join the social media bandwagon, or be viewed as a laggard that failed to
keep up with contemporary trends and developments. Such mechanisms
of pro-innovation and pro-platforms have spurred tremendous growth,
as they collectively have built a self-fulfilling prophecy in which social
media presence has been taken for granted to be desirable. Publishers
have approached platforms with ambitions towards maintaining a broad
user base, and attracting more traffic, which potentially generates eyeballs
that drive advertising revenues. Publishers have thus allowed third-party
actors, who own and control commercial platforms non-proprietary to
them, to host their news content and facilitate participation around it.
Several reviews and books have witnessed how digital journalism schol-
ars have been eager to follow in the footsteps of the journalism sector and
how publishers have built platform presence. Studies have been developed
into how news publishers and journalists have been exploring, appropriat-
ing, and normalising social media platforms into their news work (Eldridge
II & Franklin, 2019; Franklin & Eldridge II, 2017; Steensen et al., 2019;
Witschge, Anderson, Domingo., & Hermida, 2016b). Recent literature
reviews attest to the intersection of journalism and social media being
a burgeoning area of research, encompassing thousands of journal arti-
cles over the past decade (Lewis & Molyneux, 2018; Segado-Boj, 2020).
A 2017 report from the Tow Center for Digital Journalism discussed how
silicon valley has reengineered the news industry, essentially fostering a so-
called platform press that operates much in tandem with platform compa-
nies (Bell & Owen, 2017). Many have studied how news publishers and
journalists have reconfigured what they do, recalibrating their expertise
and routines in order to develop their social media presence. This extends
not only to commercial news media, but also to public service broadcast-
ers such as SVT, NRK, and DR in Scandinavia, and the BBC in the UK.
In her case study of social media at the BBC, Belair-Gagnon (2015) shows
how social media surface as emerging media, which the newsroom and
its journalists experimented with and gradually normalised, demonstrat-
ing that management outlines formal expectations for their journalists to
cross-publish and promote their news materials on social media. Hermida
(2010) discusses how Twitter and micro-blogging have given birth to a
new form of journalism that gives precedence to instant dissemination of
brief fragments of information. He posits ambient journalism as a concept
44  The platforms
referring to how such journalism, via social media, enables citizens to
develop a sort of awareness system.
Much of digital journalism studies scholarship on platforms has been
marked by three important characteristics: 1) it has predominantly
focused on social media platforms, and less on search engine platforms
or mobile platforms provided by Android and Apple; 2) scholars have
mainly studied social media platforms used in Western contexts; and
3) there are particularly many studies of Twitter. Scholars have oftentimes
conducted research focusing on Twitter because it is a more public plat-
form for which data is relatively easily accessible, for example by using
Twittonomy to access and analyse patterns, compared to platforms such
as Facebook that has become increasingly unavailable after the previously
mentioned Cambridge Analytica case (Burgess & Hurcombe, 2019; Ven-
turini  & Rogers, 2019). Moreover, following the implementation of
GDPR regulation there are substantial restrictions to API data resulting
in less data being available for analytics companies as well as researchers
(Bruns et al., 2018). Some of the research articles that have been pub-
lished in Digital Journalism have used data that would not be possible to
access with their current data policies (Bechmann & Nielbo, 2018).
In the following, we will look more in detail at the research published
in Digital Journalism related to specific platforms: Twitter, Facebook,
WhatsApp, Google, and more.

Twitter
Studies on Twitter and journalism include how Twitter is used for
breaking news (Bennett, 2016; Shermak, 2018; Thurman & Walters,
2013; Verweij & Van Noort, 2014; Vis, 2013), for constructing profes-
sional identities, the branding and promotion of news (Brems, Tem-
merman, Graham,  & Broersma, 2017; Molyneux  & Holton, 2015;
Olausson, 2017; F. M. Russell, 2019), agenda setting and influence
(Kapidzic, Neuberger, Stieglitz,  & Mirbabaie, 2019; F. M. Russell,
2019), as well as how journalists tweet about politics, or use politicians’
tweets about politics (Metag & Rauchfleisch, 2017; Mourão, Diehl, &
Vasudevan, 2016).
Let us discuss in brief a few among several examples. Digital journal-
ism scholars have focused on who the journalists using Twitter are, (Djerf-
Pierre et al., 2016; Hanusch & Nölleke, 2019; Hedman, 2015; Willnat &
Weaver, 2018). There are also studies of the ways that journalists/news
media, public actors, and private actors interact with each other via Twit-
ter. Publishers maintain an important role whereas private actors have
influence mostly during crisis situations (Kapidzic et al., 2019). In another
 The platforms 45
study into the intersection of journalism and crisis situations, where the
study of hyperlinks on Twitter was combined with other methods, the
authors concluded that societal resilience was established among Norwe-
gians (Steensen & Eide, 2019). Another article analysed a random sample
of 1.8 billion tweets and found that publishers only contribute to a frac-
tion of all Twitter activity, but have a more pronounced position when
it comes to countries of conflict (Malik & Pfeffer, 2016). There are also
other studies into how news publishers use official accounts for promotion
and interactivity (F. M. Russell, 2019), as well as in live sports journalism,
for which analysis, opinion, visual content, and entertainment generated
more likes and retweets than play-by-play results (Shermak, 2018), and
how journalists use Twitter for photojournalism and more personalised
reporting connected to emotions (Pantti, 2019). In this context it is worth
paying attention to a UK study of how reporters use Twitter for person-
alised (but not personal) reporting, to brand themselves (Canter, 2015).
This territory of digital journalism studies also features scholarship
advancing how Twitter can be used in the creation of an automated Twit-
ter account that sends tweets about the writings of Franklin Ford, who
was known to think about the future of the news (De Maeyer & Tru-
del, 2018). Moreover, scholars have turned to the Twittersphere in order
to study and analyse how the public tweet about specific phenomena,
such as “data-driven journalism” (X. Zhang, 2018). In another original
contribution analysing Black Twitter, the black public sphere and media
witnessing are triangulated in a discussion of how scholars can approach
sousveillance via Twitter using mobile devices (Richardson, 2017).
Throughout the 2010s journalists have learnt how to turn to social
media for online sourcing, including but not limited to Twitter. Sourc-
ing via social media potentially opens the gateway for voicing a much
broader and heterogenic public in journalism. However, in practice this
does not necessarily mean that citizens are featured in stories, as elite
groups are overrepresented on Twitter and also because journalists largely
rely on predefined networks of sources (Ekström  & Westlund, 2019a).
A  comprehensive study of Twitter interactions in relation to the Ger-
manwings accident showed that few citizens functioned as eyewitnesses,
but they did act as watchdogs of the watchdogs when communicating
about the shortcomings in news reporting (Masip, Ruiz, & Suau, 2019).
Another study indicates practices oriented towards protection of online
sources cannot be taken for granted (Henrichsen, 2019).
Relatively few scholars have simultaneously studied patterns regarding
how journalism interrelates with both Twitter and Facebook. Exceptions
here include a study on how one major news publisher, in Germany, the
UK, and the US respectively, has increased social media sourcing from
46  The platforms
Twitter and Facebook over time (von Nordheim, Boczek, & Koppers,
2018); another study finding that social media citizen sourcing remains
relatively unusual (Vliegenthart & Boukes, 2018); and a study showing
that men consistently were cited as sources from Twitter and Facebook
twice as much (Mitchelstein, Andelsman, & Boczkowski, 2019). Other
examples of articles reporting on multiple social media platforms include
an analysis of a specific issue (the Ice Bucket Challenge) finding that
articles with emotional appeals were most likely to be shared on these
platforms (Kilgo, Lough, & Riedl, 2020), as well as an interview-based
Australian study focusing on how sport organisations have turned to
Twitter, Facebook, and other social media to communicate with their
own strategic frames, resulting in news publishers having less power in
what was once a more symbiotic relationship (Sherwood, Nicholson, &
Marjoribanks, 2017).

Facebook and WhatsApp

Concerning Facebook, digital journalism studies includes studies assess-
ing for example how news publishers have developed formats (headline,
lead, and picture) for presenting their news materials on Facebook (Håg-
var, 2019; Welbers & Opgenhaffen, 2019), practices for such distribution
(Ekström & Westlund, 2019b), and how (an earlier version of ) the Face-
book algorithm influenced news exposure (Bechmann & Nielbo, 2018).
Several studies into Facebook and journalism deal with how audiences
relate to news on the platform. Cross-cultural survey research has been
used for studies into how citizens bypass the news media to follow politi-
cians via social media (Fisher, Culloty, Lee, & Park, 2019). Multi-method
research from Sweden involving a combination of representative survey
data and interviews with teenagers shows that social media news-accessing
is explained by age, interest in news, and habits for online news con-
sumption, and also that the young take for granted that they will become
informed (Bergström & Jervelycke Belfrage, 2018). One study used net-
worked analysis and media experience for an analysis of news consump-
tion patterns in Norway, and found a significant audience overlap when
it comes to Facebook and online news from local and regional news
publishers (Olsen, 2019). There are also studies into how citizens are
sharing news articles on Facebook (Almgren, 2017), as well as engage-
ment among those stumbling across news shared on Facebook (Kümpel,
2019). A  focus group study reports how people share news with each
other in private social media groups (Facebook and WhatsApp), based
on their belonging to location-based, work-related, or leisure-oriented
communities (Swart, Peters, & Broersma, 2019).
 The platforms 47
Several recent studies have focused on how WhatsApp is used for
journalism and news, especially in non-Western countries where this
platform is more popular. A study conducted in the context of the 2017
Chilean elections showed that sharing and discussing news on WhatsApp
was equally popular across different social groups (Valenzuela, Bach-
mann,  & Bargsted, 2019), while a study of Rwandan journalists shed
light on how they use WhatsApp to generate new ideas for stories,
for online sourcing and communication with audiences, for collabo-
ration with other journalists, and for news distribution (McIntyre  &
Sobel, 2019). Another study focused on how German publishers have
approached WhatsApp when it comes to technological change (distri-
bution) and relational change (engagement), concluding that there are
various approaches, some which include pushing unique news content
for WhatsApp (Boczek & Koppers, 2020).

YouTube, Google, Instagram, and more

Digital journalism studies also features important studies that investigate
the role and/or practices related to YouTube, Instagram, Google, and other
platforms. Let us briefly discuss a few examples. By assessing the most
viewed videos on the YouTube channels run by four major news compa-
nies (the Guardian, The New York Times, Washington Post, and Wall Street
Journal), one study concluded that the news videos going viral often focus
on positive news, while having less social significance (Al-Rawi, 2019).
Another study concluded that journalistic videos on YouTube came with
calls for political, economic, and social/lifestyle actions, and that audiences
developed comments about political and social accountability (Djerf-
Pierre, Lindgren,  & Budinski, 2019). Other research into YouTube has
also focused on news video consumption (Kalogeropoulos, 2018), and on
hostility and civility in comments (Ksiazek, Peer, & Zivic, 2015).
Only a few articles in Digital Journalism have focused on Instagram.
One study has reported on how Instagram users, acting as citizen photo-
journalists, posted informative images from the Charlie Hebdo incident
(Al Nashmi, 2018). Another looked into how citizen- and professional
photojournalists use Instagram in their performative work, and how
they communicate either their professionalism or amateurish authen-
ticity via their photos and communication (Borges-Rey, 2015). One
article reported on China and emerging business models for news start-
ups, in a country where several of the platform companies so popular
in the West are forbidden, where instead the Chinese use Weibo and
WeChat. Some start-ups are content-oriented, whereas others are clearly
focusing on building a platform presence, what the author refers to as
48  The platforms
platform-oriented (S. I. Zhang, 2019). There is more research into these
platforms with relevance for digital journalism studies, but published in
other journals, such as Asian Journal of Communication (Cui & Lin, 2015).
As 2020 began, there was a call for more research into studying and
comparing social-technical characteristics and implications across diverse
platforms such as WhatsApp and WeChat (Goggin, 2020).
Now let us turn to the intersection of journalism studies and Google.
The difficulties in studying this intersection presumably makes one impor-
tant reason there are relatively few studies, and such difficulties are brought
to light in a method article assessing the role of endogenous factors (like
keywords, language settings, clicks, and geo-location) as well as exogenous
factors (experimentation and randomisation), suggesting scholars should
study real-world participants or constructed research profiles (Ørmen,
2016). Researchers have furthermore studied how personalisation (explicit
and implicit) affects the source and content diversity of Google News,
leading them to discard the filter bubble hypothesis (Haim, Graefe,  &
­Brosius, 2018). Also Puschmann (2019) has questioned the validity of the
filter bubble concept, following his study of Google News and Google
Search in conjunction with the German general elections in 2017. A third
study, drawing on data from four countries (UK, US, Germany, Spain),
analyses news repertoires among users and non-users of search engines for
news (Fletcher & Nielsen, 2018). The findings question hypotheses about
echo chambers and filter bubbles because those who use search engines
for news are also more likely to access news from more diverse sources.

4.1.2 Platform counterbalancing
This section steps away from building a platform presence and focuses on
“opportunities” with platforms. Consider for a moment that the journal-
ism sector (and other sectors) has bought into the idea that it must engage
in search engine optimisation (SEO) and social media optimisation
(SMO). By engaging in SEO and SMO, news publishers essentially cus-
tomise or adapt their content, communication, and distribution in ways
that fit with the preferences of the platform companies and their algo-
rithms. Ultimately, news publishers have provided platform companies
with content for their platforms, sometimes even content optimised for
them. This essentially means that the news, produced by commercial as
well as public service news publishers, functions as a catalyst for audience
engagement on platforms non-proprietary to the news publishers (see for
instance Westlund & Ekström, 2018). This also means that when access-
ing the news, audiences leave digital footprints that platform companies
can capitalise on. Platform companies analyse these digital footprints in
close detail as they feed their advertising infrastructures and developments
 The platforms 49
of existing and new services. Every now and then it becomes publicly
known that platform companies have collected and/or shared sensitive
data in ways not expected, as with the Cambridge Analytica case dis-
cussed in chapter 1.
In their review essay on social media research related to journalism,
Lewis and Molyneux (2018) bring forth and scrutinise what they refer
to as three faulty assumptions: 1) that social media would be a net posi-
tive; 2) that social media reflects reality; and 3) that social media matters
over and above other factors. The authors intentionally seek to advance
provocations, identify blind spots, and critically reflect on scholarship.
This review article was featured in a special issue focusing on “News and
Participation through and beyond Proprietary Platforms in an Age of
Social Media”, for which the guest editors write:

Journalism studies, more specifically, should critically assess the polit-

ical economy of platform companies in relation to the news media.
This relates to how the news media are seeking to enable vis-à-vis
disable platform companies in maintaining a dominant role for news
distribution and public participation. Many news media have strug-
gled to enable and curate positive forms of participation. After years
of giving away news content to social media platforms, as well as
enabling the public to engage with the news via non-proprietary
platforms, some news organizations have started questioning the
long-term consequences of doing so.
(Westlund & Ekström, 2018, p. 8)

In essence, throughout the 2010s news publishers have become depend-

ent on platforms non-proprietary to them for exposure and participation,
yet have attracted limited revenues in such ways (Kleis Nielsen & Ganter,
2018; Myllylahti, 2018; Westlund & Ekström, 2018), with an overall dislo-
cation of news journalism taking place in which news publishers may well
engage in epistemic practices for social media platforms (Ekström & West-
lund, 2019b). However, platform companies have deliberatively commu-
nicated that they are not publishers themselves, and thus do not produce
content or take responsibility for content like news publishers with editor-
in-chiefs do, even though there has been some movement in this direction,
as was illustrated by the answer that Facebook CEO Mark Zuckerberg
gave to the Cambridge Analytica hearing question regarding what kind
of company Facebook is (see chapter 1). Gillespie (2018) traces the emer-
gence of platform companies and formative policy structures that resulted
in limited legal requirements in how they curate news content in most
markets, and oftentimes only being pressured to moderate content only in
cases of terrorism and child pornography.
50  The platforms
Platform counterbalancing has to do with the strategically oriented
countermeasures publishers engage in to balance their dependence on
platform companies. This means taking steps away from an approach
oriented towards building a platform presence where they promote and
make news content available for free but also stimulate and curate engage-
ment (Chua & Westlund, 2019). Publishers have felt pressured to build
a platform presence, but emerging research has also shown that there is
growing concern about the long-term developments of “dealing with
digital intermediaries” (Kleis Nielsen  & Ganter, 2018). There are cer-
tainly difficulties and costs involved when attempting to convert attention
and engagement associated with consumption on platforms into reader
revenues (Cornia, Sehl, Levy,  & Nielsen, 2018). Whether people pay
attention to news content that they stumble upon or follow on a plat-
form can have some significance when it comes to creating advertising
revenues. However, news consumption and audience engagement on
platforms non-proprietary to the news media do not necessarily mean
people will pay for online news.
As we enter the 2020s, more news publishers have publicly stated that
they find their relationship with platform companies problematic, and
that they will renegotiate and reposition themselves in the years ahead.
Industry representatives report that prominent news organisations such as
Schibsted in Norway and Sweden and The Washington Post in the US have
adopted such approaches (Lindskow, 2020; Seale, 2020). In this context, it
is worth noting that the International News Media Association (INMA)
has launched the Digital Platform Initiative, which strives towards helping
publishers deal with platforms, reducing threats to their financial sustain-
ability. In a 2019 INMA report titled How to Decode the Publisher-Plat-
form Relationship, one can learn about the increasingly critical sentiments
towards platforms from surveys with news publishers (Whitehead, 2019).
To date, relatively few articles in Digital Journalism have approached
the interrelationship between publishers and platforms in a critical way.
Platform companies and their owners have enormous power in relation
to social life and news publishers. Amazon CEO Jeff Bezos, one of the
richest men in the world, purchased The Washington Post, and Face-
book cofounder Chris Hughes acquired the magazine The New Republic.
A  study focusing on the metajournalistic discourse related to Hughes
suggests that such a digitally savvy owner will be able to save the maga-
zine and its journalism (Rooney  & Creech, 2019). Among other arti-
cles we find a study of German public service broadcasters that discusses
how Facebook has become a so-called uneasy bedfellow, which news
publishers serve with news materials in order to reach out to audiences
otherwise difficult to reach (Steiner, Magin,  & Stark, 2019). More­
over, Facebook has been widely criticised for how it uses algorithms and
 The platforms 51
humans in curating news content. The company has responded by dis-
cussing practices they see as appropriate (Carlson, 2018b). Such criticisms
extend to the role Facebook has played, actively and passively, in enabling
the spread of disinformation, most notably in conjunction with the 2016
American presidential election (Bakir & McStay, 2018).
A method-oriented article focusing on the multiple meaning of news
links concludes that a major issue to consider is that news publishers have
outsourced link-sharing to social media platforms like Twitter and Face-
book (Ryfe, Mensing, & Kelley, 2016). There are obviously limitations
when it comes to the influence journalists and editors exert over how
their news content is exposed on Facebook. Using material culture anal-
ysis of  Facebook’s patents, press releases, and data from US Securities
and Exchange Commission, DeVito (2017) found nine values built into
how the Facebook News Feed operated at the time of the study: friend
relationships, explicitly expressed user interests, prior user engagement,
implicitly expressed user preferences, post age, platform priorities, page
relationships, negatively expressed preferences, and content quality.
The previous section discussed a number of articles focusing on the
intersection of journalism and social media, how citizens may stumble
upon news in social media, and how they engage with the news via com-
menting, sharing, etc. In this line of research, it is common to see concerns
raised about citizens not accessing news enough to be informed citizens,
nor being willing to pay for news, but relatively few scholars have explic-
itly and critically discussed the many problems arising with platforms
from the perspective of publishers. One notable exception is a conceptual
article by Myllylahti (2019), in which she elaborates on attention as a key
concept. Myllylahti draws on research on attention and platformisation,
including some of her own earlier work into platforms and reader revenue
(Myllylahti, 2017, 2018). She forwards an analytical framework involving
three dimensions: 1) attention as a scarce and fluid commodity, 2) atten-
tion as a unit for measurement, and 3) attention as a source of monetisa-
tion. Gaining and maintaining audience attention is a key challenge for
news publishers, but attention is a fluid commodity, and publishers must
successfully measure and monetise on it, while they are under fierce com-
petition with platform companies. In her article, Myllylahti (2019) also
discusses general developments of news consumption on and off platforms
in 37 countries and how Apple News is surfacing as increasingly important.
With this in mind, let us now focus on different digital devices.

4.2 Digital journalism and digital devices

During the first decade of the 21st century there were a considerable
number of studies exploring the convergence between print and online
52  The platforms
news, with which scholars typically studied news sites for desktop. By
the time Digital Journalism launched in 2013 the field had reached a cer-
tain degree of saturation in terms of research into such objects of study.
Scholars at the time often developed and discussed distinctions between
print media and online news via desktop. Numerous articles focusing on
emerging initiatives for “online news sites” have been published, such
as a study into live blogs in the UK (Thurman & Walters, 2013), and a
comparison of how a dozen news sites cover Israel and Palestine (Segev &
Blondheim, 2013), to mention but a few. However, few researchers have
cared to differentiate between distinct devices such as desktop or laptop
computers, smartphones, tablets, or smart watches. Notable exceptions
include a study on how the internet and mobile communication affected
print journalism in Zimbabwe) and how mobile and social media were
appropriated within a community newspaper in Mozambique (Mare,
2014). This extends to how publishers have approached mobile as part
of their cross-media news work (Rodríguez, García, Westlund, & Ulloa-
Erazo, 2016; Westlund, 2011, 2014), how they develop communities of
practice in relation to sport journalism (Hutchins & Boyle, 2017), and
also studies into so-called mobile journalists, also referred to as MoJos
(Blankenship, 2016; Kumar & Mohamed Haneef, 2018; Martyn, 2009;
Westlund & Quinn, 2018).
There is also longitudinal and multi-method research into varied
aspects of news consumption, showing that people engage in monitor-
ing, checking, snacking, scanning, watching, viewing, reading, listening,
searching, and clicking (Costera Meijer & Groot Kormelink, 2015), as
well as using mobile devices for black witnessing and storytelling via
social media such as Twitter (Richardson, 2017). Notwithstanding this,
scholars have critically discussed that researchers to a large extent have
overlooked research into mobile news (Westlund & Quinn, 2018). Does
mobile matter when it comes to how journalists engage in online sourc-
ing, their branding on social media, how algorithms and personalisation
work, participation and comment fields, willingness to pay, and so forth?
We argue it most certainly does matter, not least because mobile in many
markets is the main gateway for accessing the news.
The affordances, designs, and approaches of different devices, and the
associated sites and applications used, can differ a great deal. Especially
in cases where executives and managers have subscribed to the concep-
tion that news publishers “must” customise their approach and content
for every device and channel they use. Unsurprisingly, the news access-
ing patterns for different devices vary, which fortunately is an area that
has attracted more significant amounts of research. It is worth noting
that there is a long tradition of studies differentiating between news-
papers and radioand television organisations (e.g.,  Elvestad, Phillips,  &
 The platforms 53
Feuerstein, 2018). Similarly, an early cross-cultural study focused on news
consumption across broadcast, print, and online, where the latter referred
to news sites, and compared this with social media (Nielsen & Schrøder,
2014). Also other scholars at this time gave emphasis to news websites
(Zeller, O’Kane, Godo, & Goodrum, 2014).
Research into mobile news has mostly focused on smartphones, but
with exceptions including three case studies of digital longforms tailored
for tablet devices (Dowling  & Vogan, 2015). In the inaugural issue of
Digital Journalism a review article into mobile news found few studies of
tablets and smartphones (Westlund, 2013). In the following years there
are a handful of studies into smartphones, journalism, and news, adopt-
ing an audience approach to study for example how different platforms
and devices are being used as repertoires (Wolf & Schnauber, 2015), and
what role mobile news consumption plays (Molyneux, 2018). There are
also studies into how mobile devices are used for mobile chat applications
such as WhatsApp (Dodds, 2019), into mobility, place-based knowledge,
and so-called spatial journalism (Schmitz Weiss, 2015), as well as smart-
phones for citizen photojournalism and witnessing (Allan & Peters, 2015;
Aubert & Nicey, 2015).
Turning towards the most recent developments in the field, we find
a 2020 special issue titled “News: Mobilities and Mobiles” (Duffy, Ling,
Kim, Tandoc,  & Westlund, 2020). It features articles looking into the
challenging processes of innovation adoption in the salient case of the
mobile social media application WhatsApp (Boczek & Koppers, 2020),
the role of mobile news during extraordinary events such as floods
(Paul & Sosale, 2020), but also the role of mobile news in everyday life,
in relation to mobility (Nelson, 2020) in both intentional and incidental
ways (Mäkelä, Boedeker, & Helander, 2020; Stroud, Peacock, & Curry,
2020; Van Damme, Martens, Van Leuven, Vanden Abeele, & De Marez,
2020). Clearly mobile news consumption has gained massive signifi-
cance, although an empirical national study finds that it has not resulted
in increases in political mobilisation (Ohme, 2020). There is also a link
between mobile technologies and mobile news with journalism educa-
tion (Bui  & Moran, 2020). We conclude that much literature in digi-
tal journalism studies has focused on “online journalism”, “news sites”,
or “social media” without further specification, even when focusing on
social media platforms like Twitter, Facebook, and WhatsApp that are
mainly used with mobile devices.

4.3 Discussions and conclusions

Our assessment shows that most studies focus on how publishers have
built a presence for social media platforms, with many empirical studies
54  The platforms
based on Twitter data, but also with a growing body of studies advancing
our knowledge on platforms such as Facebook, Instagram, WhatsApp,
YouTube, and Google. The scholarship published in Digital Journalism
in the 2010s has not dealt much with social media platforms having sig-
nificance outside of Europe and North America, such as WeChat and
Telegram. Relatively few studies have assessed the publisher-platform
relationship in a critical way, but as the journalism sector shifts in this way
such scholarship may grow.
Moreover, scholars have repeatedly developed research into online
journalism without differentiating between different devices, typically
focusing only on practices, behaviours, and attitudes related to desktop.
There are most likely several explanations, including but not limited to
scholars being familiar with desktop, having largely overlooked the tran-
sition to mobile devices in developing their research designs, and better
accessibility to desktop related data compared to mobile. There is some
research into mobile news consumption, but less into how journalists
and publishers deal with considerations of mobile technology in their
practice.
In light of the 4 A’s (S. C. Lewis & Westlund, 2015a), we find that
relatively few studies have focused on the technologies and platforms per
se, and the agency inscribed into the technological actant (exceptions
include Diakopoulos, 2015; Helberger, 2019). Scholars have typically
studied either how social actors approach emerging technologies, such
as journalists appropriating Twitter or analytics in their daily practice, or
they have studied how audiences are dealing with the changing medias-
cape in terms of accessing and engaging with news via social media plat-
forms. It is obviously challenging and time-consuming to adopt holistic
approaches to the study of actors, actants, and audiences. However, the
stream of studies focusing on only one of the A’s results in more frag-
mented knowledge. Moreover, problems arise in terms of comparing
different studies. Even if we constrain ourselves to comparing findings
about journalism and platforms in the salient case of Twitter, we will
find it difficult to synthesise findings from surveys with journalists from
country X and year X, with studies of how they interact on Twitter in
country Y and year Y, and interviews with audiences in country Z from
year Z about their attitudes to journalists tweeting. Ultimately, scholars
should develop more holistic and critical approaches to the study of how
publishers approach platforms and digital devices, and the consequences
for how citizens access and engage with the news.
5 The theories
How digital journalism is understood1

Let’s say a news company you know of wants to innovate. This news
company, which we will call The Daily Times, has suffered massive
declines in both readership and ad revenues since the turn of the millen-
nium, but now it has entered into a partnership with a tech start-up to
create a new journalistic product to be distributed on Snapchat. You are
curious about this development and want to initiate a research project to
investigate it. But how do you frame it theoretically? If your background
is in sociology, you might want to research how the developments at The
Daily Times affect journalism’s position and role in society, if and how
they change what it means to be a journalist, or other aspects related to
journalism as a profession, a social institution, field, or system. If you are
more interested in political science, you perhaps would like to research
to what degree the case changes journalism’s democratic function, if it
manages to get new audiences interested in public affairs, or other aspects
related to journalism’s position in the public sphere. If your interests align
with Science and Technology Studies (STS), you are maybe interested in
analysing how technology and humans interact and who and what shapes
the innovation process and outcome.
If you have a background in language studies, you might want to
investigate how journalistic genres develop in the new Snapchat product,
if new rhetorical strategies can be detected, how the new journalism
creates meaning through linguistic, discursive, or semiotic features, or
other aspects related to the production, distribution, or consumption of
multimodal texts. If you are more interested in analysing the case from
a cultural studies perspective, you perhaps want to figure  out how the
Snapchat product affects how audiences relate to news in their everyday
lives, or if and how the case changes the journalists’ self-perception and
feelings of identity, or maybe what kinds of narratives the new initiative
creates. If you are interested in philosophy, you might want to research if
and how the journalists, when working with the new Snapchat product,
56  The theories
create knowledge and make judgments about what is true and not. If you
are a historian, you might search in the history of media and technology
to find similarities with the recent development. If you are an economist
or business and administration scholar, you might be interested in how
the new product alters the supply and demand of and for journalism and
news, or how reward systems affect the decision making of the actors
involved in the innovation process, or how organisational mechanisms in
the news company and the tech start-up affect the process.
In other words: The possibilities are almost endless. Digital journalism
studies is both a cross-disciplinary field, meaning that the same case can
be researched from a variety of different disciplines, and it is interdisci-
plinary, meaning that multiple disciplinary perspectives can be combined
in one research project about the case. This cross- and interdisciplinary
nature of digital journalism studies means that there are a substantial
number of theories that potentially can be used to explore and frame
a research project on The Daily Times case. Those that you eventually
deploy reflect where your research interests lie and, evidently, which
research question(s) you would like to find answers to.
Throughout this chapter we will use this imagined research project to
look at the many ways in which theory matters for research in general
and digital journalism studies in particular. We will look at how theo-
ries from a variety of disciplines can be utilised and/or developed to
answer a myriad of possible research questions related to this one case.
In chapters 3 and 4, we discussed the topics and objects that preoccupy
digital journalism studies. We showed how the field has been dominated
by an emphasis on technologies and platforms, and also to an increas-
ing degree on audiences. This does not mean that everything about
digital journalism studies concerns technologies, platforms, and audi-
ences, nor does it mean that the theoretical perspectives, frameworks,
and assumptions researchers interested in digital journalism make use of
and develop are about technologies, platforms, and audiences. Digital
journalism studies is much more. It is a research field for scholars in all
kinds of disciplines – and therefore it can be understood and theorised
in many different ways.
In this chapter  we will take a closer look at this role of theory in
the field, what theories are commonly used and how they contribute to
making sense of digital journalism. We will also identify some theoretical
shortcomings of digital journalism studies, but first we need to discuss
the possible ways in which theory can be understood and what attitudes
towards theory are possible to for such a research project like The Daily
Times case.
 The theories 57
5.1 What is theory and why does it matter?
When you design your research project on the The Daily Times case
one of the first things you need to think about is what kind of attitude
towards theory you have. To what degree do you base your research on
theoretical assumptions? Do you want to test a specific theory, or do you
want to develop theory? Whatever you choose, you will relate to theory
in one way or the other.
The word “theory” has many connotations. It can mean the oppo-
site of practice. Theory can also be explanatory or mean something that
can be tested, verified, or falsified. Theory can be grand or grounded,
inductive, deductive, or abductive. It can be rational, critical, pragmatic,
or normative. Theory usually means one thing to a natural scientist and
something very different to a researcher from the humanities. Social sci-
ences, in turn, can encompass the whole spectrum. Mjøset (2006) dis-
tinguishes between three different attitudes towards theory in the social
sciences:

1 The standard attitude, which implies an understanding of theory as accu-

mulated knowledge based on regularities as law-like or idealised as pos-
sible. This attitude involves, in its purest sense, derivation of hypotheses
from macro-theories and testing them on empirical material.
2 The social-philosophical attitude, which implies an understanding of
theory as something that is a result of investigations into how the
human mind organises knowledge. This attitude typically involves
generating theoretical concepts suited to frame and interpret aspects
of modernity.
3 The pragmatist-participatory attitude, which implies an understanding
of theory as knowledge of observable patterns accumulated in “local
research frontiers” consisting of previously conducted empirical
inquires of similar cases and previously developed grounded theories
related to the same topic.

Given the cross- and interdisciplinary nature of digital journalism stud-

ies, we can expect to find all three attitudes towards theory in inquiries
within the research field. As such, all three attitudes could be applied
to a research project about the The Daily Times case. For instance, you
could apply Bourdieu’s field theory and investigate to what degree this
partnership between the legacy news company and a tech start-up affects
journalism as a social field. If you formulate hypotheses based on field
theory and test those hypotheses on the empirical findings of your case
58  The theories
study, you apply the standard attitude towards theory. Or you could treat
the case as an example to illustrate aspects of network theory, for instance
related to what happens to a network when new “nodes” (the tech start-
up, Snapchat) are introduced to an existing network (the news network).
Then you apply the social-philosophical attitude. Or you could sam-
ple both theoretical and empirical knowledge from previous research on
similar cases and aim at advancing that knowledge through an analysis of
your case. Then you apply the pragmatist-participatory attitude. How-
ever, choosing the attitude towards theory is only the first step towards
a research design. The second step would be to figure out exactly what
theoretical perspective would be relevant and, consequently, which
research questions to formulate.

5.2 The multitude of theories in digital

journalism studies
In chapter 1 we argued that sociological perspectives are most common in
digital journalism studies, followed by technological and political science
perspectives (see Figure 1.1 in section 1.3.1). However, what is striking
about digital journalism studies is that both sociological and political sci-
ence perspectives are quite often combined with an emphasis on technol-
ogy, implying either that technological aspects are what is being studied
or that researchers apply social or political science theory that is sensitive
to the role of technology. Thirty-eight percent of the 172 abstracts we
analysed in Digital Journalism are marked by such technology-oriented
sociological or political sciences perspectives. A typical examples is Usher
(2013) who employs news norms theory and the theory of social con-
struction of technology to analyse how Al Jazeera’s English website used
web metrics for tracking and understanding audience behaviour.
Another typical feature of the research published in Digital Journalism
is that a majority of it adopts a pragmatist-participatory attitude towards
theory and develops middle-range or even micro-theory from empiri-
cal data. In this respect, digital journalism studies is no different from
journalism studies or even communication studies in general (Ahva  &
Steensen, 2020; Bryant & Miron, 2004; Kamhawi & Weaver, 2003). The
pragmatist-participatory attitude towards theory means that the research
does not start with assumptions or perspectives derived from grand theo-
ries, but rather builds on previously established empirical knowledge in
an inductive, grounded theory-like fashion in order to advance knowl-
edge. Half of the 172 abstracts we analysed derive from such an atti-
tude towards theory. A typical example is Johnston (2016) who analyses
the uses of social media and user-generated content by journalists in the
 The theories 59
BBC World News newsroom and the effects this has on the role of the
journalists.
In the other half of the 172 abstracts from Digital Journalism we find
69 different theories and conceptual constructs. Only 11 of those are
mentioned in more than two abstracts, implying that digital journalism
studies is much about finding new ways to conceptualise the research.
A good number of the articles published in the journal are conceptual
pieces that discuss theory or introduce new conceptualisations or theo-
retical frameworks. Some of the most influential articles published in the
journal (in terms of citation metrics) fall into this category, like Lewis and
Westlund (2015a) who argue for developing a socio-technical emphasis
for the study of institutional news production.
In the following sub-sections we take a closer look at some of the most
influential and common theories used to frame and interpret empirical
findings within digital journalism studies, not only in articles published
in the journal Digital Journalism, but in a broader sense.

5.2.1 Digital journalism as a social system

Sociological perspectives imply that digital journalism is understood as
a kind of social system in which certain roles are performed and practices under-
taken. A  range of social system-related macro-theories have been used
to explain and explore the role that digital journalism plays in societies,
why it matters, what makes it different from other forms of communi-
cation and other parts of society, and how it changes. To return to our
case study of The Daily Times, the tech start-up, and Snapchat: If your
primary research interest in investigating this case is to explore how it
relates to or affects journalism as a social system, you have a range of
theoretical options. Luhmann’s theory of social systems can help to
explain digital journalism’s position in a society by how it differenti-
ates itself from other social systems (like for instance the social system
the tech start-up initially belongs to) and creates boundaries of meaning
(Loosen, 2015). Bourdieu’s field theory has been used to analyse the
connections between journalistic organisations, practices, products, and
professionals, on the one side, and other social fields, like for instance
advertising (Q. Wang, 2018) on the other, or how digital interlopers (like
for instance tech-workers at the tech start-up) challenge the bounda-
ries of the journalistic field (Eldridge II, 2017). Like field theory, new
institutionalism is a social system theory that mediates “the impact of
macro-level forces on micro-level actions” (Ryfe, 2006, p. 137). Analys-
ing digital journalism as an institution means analysing the presupposi-
tions and tacit knowledge that guide journalistic practice across digital
60  The theories
newsrooms, news organisations, and other journalistic organisations.
Related to our case, new institutionalism could be applied to analyse
how the tacit knowledge of both the journalists at The Daily Times and
the technologists and entrepreneurs at the tech start-up affect their ability
to cooperate.
Central to these theories is that they provide explanations and ques-
tions from a macro perspective for how an institution/field/system like
journalism functions and develops in digital societies through analysis
of how individual behaviour coincides with larger, cross-organisational
structures. As such, social system theories provide frameworks for analys-
ing interplays between mental structures (norms, values, ideals), material
structures (economy, technology), and agency in digital journalism. We
find ways of analysing the same interplay also in middle-range theories
like organisational theory and hierarchy of influences theory. The differ-
ence is that such theories do not aim at explaining societies on a macro
level. Organisational theory provides a framework for understanding
how various kinds of organisations (like the news company and the tech
start-up) are configured and reconfigured by internal and external struc-
tures, and by the actions of different kinds of professions and labour that
are part of the organisation. Organisational theory has been applied in
digital journalism studies to analyse, for example, how specific beats, like
science journalism (Lublinski, 2011), develop in a digitised news envi-
ronment. News production studies also take news organisations as their
starting point and analyse how agency and mental and material structures
shape how news is produced in digitised news rooms (Domingo & Pater-
son, 2011; Usher, 2014). Such studies have been important in advancing
our knowledge of how classical middle-range theories of journalism, like
theories of news values (see for instance Harcup & O’Neill, 2017) and
gatekeeping theory (see for instance Bro & Wallberg, 2015) hold up in
a digital age. Such theories could also be relevant to the The Daily Times
case, if the primary aim is to understand how specific beats, genres, or
types of journalism develop with the partnership between the news com-
pany and the tech start-up in place; or how the partnership and Snapchat
as a publication and distribution platform affect what is considered news
(news values) or who gets to decide what is newsworthy (gatekeeping).
Recognising that journalism in digital times has become increasingly
independent of news organisations and influenced by all kinds of struc-
tures and agency on macro-, meso-, and micro-levels, the hierarchy of
influence theory introduced by Shoemaker and Reese (1996) provides
a model of the levels that influence digital journalism: from the macro-
social systems, via social institutions and organisations, to the micro-
levels of routine practices and individuals. Similarly, practice theory
 The theories 61
(Bourdieu, 1977; Schatzki, 2001) has been used to analyse how activities,
materiality, and discursive reflexivity connected to journalism shape what
digital journalism is and why it develops as it does, preferably without
preconceived ideas on who the key agents are, what they produce and
within what kind of organisational framework journalism operates (Ahva,
2017). Both these theories have obvious relevance to our case, if you
want to investigate how the partnership and Snapchat influence journal-
ism as practice.
The theory of media logic (Altheide & Snow, 1979) could be equally
relevant, especially if the concern is to investigate how the organisational,
technological, and aesthetic dimensions of the companies, platforms, and
products involved work together in supporting the creation of content
and how that content is shared. Even though the theory was developed
in an age of mass media, it has proved valuable in analysis of how different
media and platforms in digital times differ in their logics and how “poly-
media channels” become more and more contextualised in everyday lives
(Thimm, Anastasiadis, & Einspänner-Pflock, 2018).
In our analysis of abstracts of articles published in Digital Journalism
between 2013 and 2019 (N  =  172) we find that many of the above-
mentioned theories are among the most used. Gatekeeping theory, field
theory, institutional theory, and hierarchy of influence theory are among
the 11 theories we found explicitly mentioned on two occasions or more
in the abstracts analysed. Sociological theories related to professional-
ism, which are the most common theories in journalism studies as a
whole (Ahva & Steensen, 2020), are also common in digital journalism
studies, but to a lesser extent and with decreasing popularity in the jour-
nal Digital Journalism, when we look at how keywords belonging to the
thematic cluster professionalism have developed. The keywords belonging
to this cluster – like values, norms, role, professional identity, autonomy, ide-
als, skills, standards, and others – were used in 3 percent of the articles
published between 2013 and 2015, but only in 1 percent of articles pub-
lished between 2017 and 2019. However, major research projects like
the Worlds of Journalism study (see Hanitzsch, Hanusch, Ramaprasad, &
De Beer, 2019) have been preoccupied with how notions of journalistic
professionalism develop in a digital age in various cultural contexts, so
it would be unfair to claim that digital journalism studies is not inter-
ested in how journalism develops as a profession in digital times. And
professionalism would be an obvious framework for the The Daily Times
case, if the main research concern prompting analysis is to understand
how the partnership with both the tech start-up and Snapchat affect the
role of journalists at The Daily Times, the norms and values they adhere
to, and so forth. That said, one reason for the decreased popularity of
62  The theories
professionalism as a theoretical framework in Digital Journalism might be
that this framework has been criticised for limiting the domain that is
seen as a valid information source about journalism and hence potentially
omitting the role of participating non-journalists in the construction of
journalism (Ahva, 2017).

5.2.2 Digital journalism as a socio-technical practice

A recurring theme in digital journalism studies is the connection between
technology, power, and change. Is technology a powerful driving force
behind all the changes that mark what digital journalism is becoming, or
is technology merely one of several things that influence how journalism
develops? Or is it perhaps the other way around: that the developments in
digital journalism are the things that shape technological developments?
Influences from -STS- have enriched digital journalism studies with
a nuanced and theoretically informed understanding of the relationship
between technology and journalism. Technological determinism,
which presupposes that technologies are blind to social and cultural diver-
sity and enforce change wherever they are introduced, used to dominate
research on digital journalism (Steensen, 2011b). But the works of schol-
ars like Bijker (1995), Bolter and Grusin (Bolter, 2001; Bolter & Grusin,
1999), Manovich (2001), Mosco (2004), and perhaps most importantly
Latour (2005) have shown that technology is not a blind determinant of
change. Technology is socially and culturally shaped. Theories like the
social construction (or shaping) of technology (SCOT) (Bijker,
Hughes, & Pinch, 1987) have been adapted to the analysis of journal-
ism in digital times, most notably by Boczkowski in his book Digitizing
the News (2004), which paved the way for understanding the interplay
between technology, materiality, and social practice related to the pro-
duction of digital journalism.
Another relevant theory, which has been much used in digital journal-
ism studies and which represents a middle ground between technological
determinisms and SCOT, is affordance theory, which originally was
an ecological theory related to human perceptions of what the environ-
ment can offer (Gibson, 1979). Applied to technology, affordance theory
emphasises how the possibilities (and restraints) of certain technologies
can be utilised in their context of use (Conole  & Dyke, 2004). One
important aspect to technological affordance theory is that the affor-
dances are not understood as objective characteristics of the technol-
ogy in question, but rather as something that is perceived to have certain
affordances in given contexts, what Nagy and Neff (2015) call “imagined
affordances”. Related to our research case, affordance theory can be used
 The theories 63
to analyse how the imagined/perceived affordances of Snapchat create
both possibilities and restraints for practicing journalism on the platform.
A technological determinist approach to the same question would imply
that Snapchat is seen as determining the journalistic practice, regardless
of the social and culture context of the practice, while a SCOT approach
would imply an assumption that Snapchat is a flexible technology that is
shaped by the social context of its application.
Both SCOT and affordance theory acknowledge that technology
enters into a relationship with social contexts and that human actors
have an effect on what kind of impact technologies have. They also have
in common that technology usually is the starting point of an inquiry
and that the relevant social context is easily identifiable. However, the
increasing uncertainty as to where journalism is to be found, who produces
it, on what technological platforms it exists, and how various groups of
professionals and amateurs participate and cooperate in its coming into
existence in digital times has led to the popularity of socio-technical the-
ories that do not take anything for granted, like actor-network theory
(ANT) (Latour, 2005). ANT is a middle-range social systems theory
with no preconceived ideas about who and what shape a social system.
It emphasises not only the mutual shaping of journalism and technology
but also juxtaposes human, technological, and material actors (or actants,
which is the most common word for human and non-human actors in
ANT) as equally important to this mutual shaping. The theory has been
praised for its non-deterministic, unbiased, and empirical orientation (see
Primo & Zago, 2015 for a discussion), its adaptability to digital journal-
ism in particular (Domingo et al., 2015), but also critiqued for being a
methodological approach and not a theory, and for its inclination to pro-
duce nothing more than dull descriptions (Benson, 2017).
Similar to ANT, other network theories like homophily, resource
dependence, and social influence theory (see Fu, 2016 for an over-
view) open up the empirical field to include potential actors, which the
researcher did not think of beforehand, while simultaneously emphasis-
ing the relations between them and the actions and work being done
through those relations. The use of the keyword “network” has in recent
years grown significantly in digital journalism studies, as have spatial
keywords related to “ecosystems” and “landscapes”. Reese (2016, p. 10)
refers to “the ecosystem shift” in theories of journalism and connects this
to the emergence of digital platforms that have made some of the classical
conceptual categorisations invalid.
Returning to the The Daily Times case, an ANT approach could be
appropriate if the main aim of the research is to understand, in descrip-
tive manners, who and what are important for how the new journalistic
64  The theories
product on Snapchat turns out. If you are more interested in the rela-
tions and power dynamics between the organisations involved (The Daily
Times, the tech start-up, Snapchat, and potentially other organisations
with relations to the news company or the tech start-up), you might want
to analyse the case for instance as an affinity network (Fu, 2016). Or, if
you feel (actor-)network theory is too unstructured to apply, you can
choose a more structured adaptation, like Lewis and Westlund’s (2015a)
4As framework, which provides a model for analysing the actors, actants,
audiences, and activities involved in producing the new Snapchat output.

5.2.3 Digital journalism as a democratic force

If your research interests lie not so much with the social or socio-technical,
but rather with journalism’s link to political systems and the public
sphere, you probably want to apply a political science perspective on the
case. Political science is the third most common disciplinary framework
in digital journalism studies, according to our journal analysis, and it
usually means that digital journalism is seen as a democratic force that shapes
public discourse.
A number of potential research questions which would require a
political science perspective can be addressed to explore the The Daily
Times case: What does the new Snapchat channel contribute to the pub-
lic sphere? Does it enhance awareness about public affairs, or is it mainly
a provider of entertainment? Does it allow new voices to be heard? Does
it reach new audiences and contribute to their interest in public affairs?
Does it address the audience as (passive) consumers, (active) citizens, or
in another way? To what degree does it allow audiences to participate?
And how does the tech start-up and the whole Snapchat affair influence
the news company’s self-perceived societal role as for instance a watch-
dog, populist mobiliser, interpreter, disseminator, fourth estate, custodian
of conscience, or other professional role conceptions (Glasser & Ettema,
1989; Hanitzsch, 2017; Mellado, 2019)?
Many of these questions would require some kind of democracy or
public sphere theory. There are long, historic ties between journalism
and democracy/public sphere theories. A  free, independent press that
facilitates a public sphere in which ideas and politics can be dissemi-
nated, debated, critiqued, and shaped has been considered a cornerstone
for democracy ever since the Age of Enlightenment, in which catch-
phrases like Thomas Jefferson’s “information is the currency of democ-
racy” began to dominate the democracy discourse (Zelizer, 2013, p. 463).
Theories of journalism and democracy are usually normative theories,
implying that they prescribe what role journalism should have in a society
 The theories 65
and what a democracy should be like. Embedded in such normative theo-
ries is the notion that journalism is a prerequisite for democracy and vice
versa; journalism and democracy are so intertwined that the one cannot
exist without the other.
Such normative theories of journalism (and democracy) have been
criticised for a number of reasons. First, they cannot explain how and
why journalism exists in semi- or non-democratic societies. Siebert,
Peterson, and Schram (1963) addressed this problem in their categorisa-
tion of how journalism functions in various political systems expressed
as the four theories of the press: the authoritarian, libertarian, social
responsibility, and Soviet-totalitarian. However, the four theories of the
press did not provide an escape from normative theory, as it was discur-
sively embedded within a libertarian logic that clearly ranked the four
categories along an axis from good to bad (Nerone, 1995). Several revi-
sions of the four theories of the press and alternative models have since
been suggested, all of which are based on some degrees of normativity
(see Christians, Glasser, McQuail, Nordenstreng, & White, 2009, chap-
ter 1 for a review). Moreover, normative theories linking journalism and
democracy tend to disregard the fact that journalism, especially in our
digital age, is not the only channel through which trustworthy informa-
tion can flow in a society and a public sphere marked by a diversity of
opinions can be established. Blogs, social media, citizen journalism, and
other information channels have democratised public speech, and Zelizer
(2013) has therefore, and for other reasons, suggested it is time to put
democracy theory to rest in journalism studies.
Nevertheless, democracy theories enable us to understand the role that
digital journalism plays as a facilitator of the public sphere and how it
covers issues that require public attention. The so-called procedural or
competitive democracy theories have long framed journalism studies
and guided researchers’ attention towards the role that journalism plays in
providing information to citizens as voters between the elections and the
ways in which politicians compete over power in the public sphere, while
participatory and deliberative democracy theories became more
prominent in the 1990s as journalism moved to digitised spaces (Ström-
bäck, 2005). These models invite us to examine and assess whether jour-
nalism enables or restricts civic agency and reasoning beyond the moment
of voting, and the role of public discourse in the formation of the political
culture (e.g.,  Ettema, 2007). As a more middle-range theory developed
within communication studies, agenda-setting theory (McCombs  &
Shaw, 1972) provides a framework for analysing how journalism shapes the
public sphere and consequently the ways in which we, as the public, under-
stand the world. Theories of second-level agenda setting (Ghanem,
66  The theories
1997) and inter-media agenda setting (Danielian & Reese, 1989) refine
agenda-setting theory in ways that make it more relevant for digital journal-
ism studies. They provide frameworks for analysing how the media discuss
issues that have already made the agenda and how certain media (like elite
newspapers) influence what other media (like social media) have on their
agenda. Returning to the The Daily Times case for a moment, one question
based on an inter-media agenda-setting framework, could be to analyse
how the Snapchat initiative affects the agenda on Snapchat in general.

5.2.4 Digital journalism as post-industrial business endeavour

The financial crises in general and the economic distress of journalism
in digital times in particular have caused an increased interest in digi-
tal journalism as business, its organisational structures, and its economic
sustainability. Economic theories like rational choice theory, which
has been used to analyse journalists as “rational actors seeking to maxi-
mize materialistic and non-materialistic rewards” (Fengler & Ruß-Mohl,
2008, p. 667); path dependency theory, which can explain why legacy
news organisations have difficulties coping with change (Koch, 2008);
and more audience-centric economic theories like uses and gratifica-
tion theory, which can be used to analyse emerging patterns of news
consumption (Diddi & LaRose, 2006), have been applied.
The same holds for organisational development theory, which has
been used, among other things, to assess the perceptions and attitudes
that top newsroom managers and journalists have about initiatives aimed
at changing newsroom cultures (Gade, 2004). As such, organisational
development theory has some similarities with innovation theories, espe-
cially diffusion of innovation theory (E. M. Rogers, 2003), which has
been used in digital journalism studies to assess how processes of innova-
tion and thereby change proceed in an organisation like a news company.
Interestingly, there is a tendency within digital journalism stud-
ies that keywords such as “media industry” and “economic theory” are
decreasing in popularity and being replaced by a variety of more flexible,
individual-focused, and business-related conceptualisations, such as “sus-
tainability” or “entrepreneurialism” (Ahva & Steensen, 2020, p. 48). This
shift is connected to a situation where the journalism industry as a clearly
demarcated branch within the media industry needs to be rethought – as
proposed by the notion of “post-industrial journalism” (C. W. Anderson,
Bell, & Shirky, 2015) and Deuze and Witschge’s (2020) work on “Beyond
Journalism”, in which the authors stress the need to theorise journal-
ism beyond legacy institutions and organisations and include the increas-
ingly entrepreneurial nature of journalism. This rethinking of journalism
 The theories 67
as industry and business thereby represents a move from organisational
enterprises to individual entrepreneurship, a move that emphasises how
individual journalists can (and should) reinvent themselves as independ-
ent entrepreneurs by starting a company outside of legacy news organ-
isations. Hence concepts and theories from management and business
studies, such as business model canvas (Singer, 2016), are applied to
address how journalists can see change and disruption as business oppor-
tunities (Briggs, 2012).
The Daily Times case, in which a tech start-up becomes involved in
journalism, can certainly be analysed with such a business model canvas
perspectives in mind. Economic theories could also easily be applied to
the case, if the primary interest is to understand why the organisations
involved, the journalists, the tech start-up workers, and the audiences
choose to act as they do. Organisational development theory could also
be relevant if the aim is to investigate how the news company facilitated
and reacted to the change brought forth by the partnership with the tech
start-up. And, obviously, innovation theory is relevant if it is the process
of innovation itself, and how such processes unfold, that is of interest.

5.2.5 Digital journalism as cultural production and discourse

Perspectives from cultural and language studies occupy a smaller part
(about 9 percent combined, see Figure 1.1 in chapter 1) of digital journal-
ism studies, as revealed in our analysis of article abstracts, than in journalism
studies in general (Ahva & Steensen, 2020; Steensen & Ahva, 2015). Tra-
ditionally, analysing journalism through the lenses of cultural theory has
implied questioning what is presupposed in journalism, unravelling how
journalists view themselves, trying to understand the diversity of journal-
ism, and connecting journalistic practices and products to questions of
power, ideology, class, ethnicity, gender, identity, and so on. However, it
seems as if the cultural analysis of digital journalism is more interested in
how journalism intersects with everyday life, with the “moment at which
media production becomes communication and culture – the moment
of the use in the circumstances of everyday life” (Hartley, 2008, p. 47).
This reflects what Costera Meijer (2020) has labelled “the audience turn”
in journalism studies, a turn which is quite visible in our analysis of arti-
cles in Digital Journalism, since “audience” is the second biggest thematic
cluster of keywords next to “platform” (see chapter 2, section 2.3). The
audience cluster comprises keywords that typically signal political science
perspectives (like “citizen”, “participation”, and “public”), but also sev-
eral that signal more cultural dimensions, like “readership”, “amateur”,
“perception”, “community”, and “reader contract”).
68  The theories
There are many potential research questions related to our Daily Times
case which would require a theoretical framework from cultural or language
studies. Some examples could be: How does the partnership affect journal-
ists’ self-perception and identity? To what degree does the Snapchat product
alter the function of journalism as a meaning-making system? What genre
conventions are utilised on Snapchat and how does this affect the journalist-
text-audience relationship? How does the new product affect journalism as
a discursive practice? And how do the news company’s ambitions with the
Snapchat product align with how it is perceived by audiences?
Critical theory has traditionally been strongly connected with the
cultural analysis of journalism, especially as related to neo-Marxism and
the Frankfurt school of thought. This implies an ambition to unmask
the social and ideological power structures embedded in journalism and
to uncover the discrepancies between journalistic self-perception and
“metajournalistic discourse” (Carlson, 2016) on the one hand, and the
actual expressions and meaning production systems of journalism on
the other. Hence, language-based traditions of studying journalism are
closely related to cultural ones. The field of semiotics, in which text
is understood as not only written language, but also as still and moving
images, body language, and so on, has been important in recognising
journalism as visual culture and the diversity through which journal-
ism produces meaning. Language studies increasingly also emphasises the
social and cultural situatedness of digital journalism texts, which requires
that the studies of text are informed by material and contextual dimen-
sions, too (Richardson, 2008, p. 2).
Discourse theory (recently discussed and developed, for example, in
Kelsey, 2015), narrative theory (e.g., J. Johnston & Graham, 2012), and
genre theories (Smith & Higgins, 2013) can be important to analyse
digital journalism as a meaning-making system. Van Dijk (2009, p. 193)
has underlined that a major dimension in discourse analytical studies of
journalism is the ideological nature of news. The approach can therefore
help in examining the expression and reproduction of ideology in digital
journalism, the axiomatic beliefs underlying the social representations
shared by a group. Significantly, van Dijk points out that the role of dis-
course in reproducing racism, nationalism, and sexism should be more
carefully studied in the future.

5.3 The theoretical blind spots of digital

journalism studies
The theories we have discussed represent the main disciplinary perspec-
tives found in digital journalism studies. However, they are not the only
 The theories 69
theories employed, nor are they the only theories that might be of value.
The most significant form of theoretical knowledge found in digital
journalism studies is perhaps the increasing body of accumulated knowl-
edge concerning what digital journalism is, where it is to be found, who
produces, distributes, and consumes it, and why (or if ) it matters. This
knowledge is accumulated through empirical investigations and concep-
tual discussions in grounded theory-inspired research designs and atti-
tudes towards theory. It is a kind of knowledge that is crucial to have and
to constantly update when your object of study is constantly changing, as
is the case with digital journalism.
There are, however, some potential problems with this kind of knowl-
edge accumulation and theory building and framing. The first problem is
related to the emphasis on change that dominates the field. We will discuss
this problem in more depth in the next chapter, in section 6.2.2, so here
we will only point to one potential reason for this emphasis on change
and the potential blind spots it creates, namely that there seems to be a
lack of historic perspectives in digital journalism studies, at least in articles
published in Digital Journalism. This becomes evident when we look at
the sources referenced in articles published in the journal. The 350 arti-
cles published between 2013 and issue 4, 2019, have a total of 14,794 ref-
erences. Fifty-nine percent of these references point to research published
after 2010, and only 13 percent point to research published before 2000.
In other words, there is a lack of connection with findings from the past
and a preoccupation with the present and the future in digital journalism
studies. No doubt, an emphasis on the present is understandable, perhaps
even logical, in a field like digital journalism studies, which to a certain
degree is determined to investigate the current changes to its object of
study due to recent technological developments. However, this does not
mean that such inquiries should only emphasise what is changing, and
only look at such changes from the perspectives of recent theories and
research. We therefore conclude that digital journalism studies should
have a stronger connection with the past in order to better understand
the present and predict the future.
A second blind spot is that digital journalism studies has a social sci-
ence bias. There are many reasons why digital journalism scholars should
view digital journalism, and other forms of journalism for that mat-
ter, predominantly as a social phenomenon. A dominance of social sci-
ence perspectives and approaches is therefore not in itself a problem.
One might even argue that without such prominence, digital journal-
ism studies would neglect the social, political, and to a certain extent
cultural ramifications of the digital on journalism. However, approaches
from the humanities are also capable of analysing journalism as a social
70  The theories
(and cultural) phenomenon. When perspectives from the humanities are
marginalised as they seem to have been with the ways in which digital
journalism studies has developed in Digital Journalism (see Figure 1.1 in
chapter 1), and when methodological approaches are increasingly geared
towards computation and big data (as we will discuss in chapter 7), cru-
cial elements of digital journalism might be overlooked. As argued by
Steensen et al. (2019, p. 336):

The future reader who consults Digital Journalism to find out how
ideas and discourses were constructed in journalistic texts in the
2010s, how journalism created meaning of and for the societies and
cultures it served, how journalism functioned as a system of knowl-
edge creation, and how such questions were connected to historic
developments, is likely to be disappointed. To provide answers to
such questions, digital journalism studies should to a greater extent
embrace the disciplinary perspectives and qualitative methodologies
of the humanities.

Even though digital journalism studies no doubt is highly cross- and

interdisciplinary, in spite of these biases, there seems to be an underdevel-
oped potential of connecting not only with fields within the humanities,
but also those fields related to technology, like computer science, infor-
matics, and information science. Boczkowski and Mitchelstein (2017)
argue that digital journalism studies is marked by an inability to connect
empirical findings across other domains of digital culture, and by a lack
of conceptual exchanges with other fields and disciplines. It seems obvi-
ous that digital journalism studies should move beyond a topical inter-
est in technology and connect with fields and disciplines like computer
science and informatics on a more theoretical level. For instance, the
field of theoretical computer science “provides concepts and languages
to capture the essence, in algorithmic and descriptive terms, of any sys-
tem from specification to efficient implementation” (Van Leeuwen, 1990
Preface). As digital journalism becomes increasingly dependent on algo-
rithmic processing (see chapter 7, section 7.2), acquiring such concepts
and languages seems crucial for digital journalism scholarship. Similarly,
theoretical understandings of information transformation across natural
and engineered systems, which is the essence of informatics as an aca-
demic field, seem important for digital journalism scholarship. Practices
of digital journalism, especially those related to investigative journalism,
are increasingly preoccupied with the analysis of massive amounts of
unstructured data, which requires both methodological and theoretical
knowledge in order to make sense. Here, digital journalism scholarship
 The theories 71
needs not only the same kind of knowledge to assess such practices of
journalism critically, but also the knowledge to experiment with how
digital journalism can make sense of such information transformations.
Some examples of the latter already exist, either from within informatics
itself, like Wiedemann et al.’s (2018) experimental research on developing
tools for the analysis of massive amounts of documents like the Panama
Papers or similar big leaks, or from interdisciplinary cooperation like
Maiden et al.’s (2018), Nyre’s (2015, 2012), and Backholm et al.’s (2018)
experimentations with new journalistic applications.

Note
1 This chapter is based on and partly reuses and further develops arguments, findings,
and phrases previously published by Ahva and Steensen (2020), Steensen and Ahva
(2015), and Steensen et al. (2019).
6 The assumptions
The underlying normativity of digital
journalism studies

In the previous chapter, we discussed the role of theory in digital jour-

nalism studies and the many theoretical perspectives through which a
research project investigating aspects of digital journalism can be framed.
We used an imagined case as an example: a news company that had part-
nered with a tech start-up to develop a new journalistic product to be
distributed on Snapchat. We showed how this one case can be analysed
based on a myriad of research questions reflecting different theoretical
perspectives in order to acquire new theoretical or empirical knowledge.
Some of these research questions were quite neutral in their quest for
new knowledge, while others were based on ideas of what journalism
should be, like: does the Snapchat product allow new voices to be heard?
Does it reach new audiences and contribute to their interest in public
affairs? These two questions presuppose that new voices should be heard
in digital journalism, that digital journalism should reach new audiences,
and that it should make people interested in public affairs. In other words:
These research questions are based on normative ideas and theories.
Traditionally, much of journalism studies has been rooted in such nor-
mative ideas and theories, especially related to the role of journalism
in societies. This includes journalism’s ability to treat its audience like
informed citizens and raise awareness and public engagement on mat-
ters of perceived importance, and, consequently, on what is good and
bad journalism (Benson, 2008). Our (normative) position is that there
is nothing wrong with normativity in research in general and in digital
journalism studies in particular. However, normativity is a problem if it is
hidden, unproblematised, and masked as apparent neutrality.
There are at least three ways in which such problematic normativ-
ity could occur in digital journalism studies. First, researchers in the
field should be aware of how digital journalism as practice challenges
certain norms concerning what journalism is and should be. Second,
digital journalism researchers should be aware of the norms potentially
 The assumptions 73
embedded in the ways in which they ask and frame research questions.
Third, the methods researchers choose are not neutral. They carry with
them certain ways of viewing the world and specific norms related to
what knowledge is and how it can be obtained.
We return to the potential problems of normativity in methodology
further in chapter 7. In this chapter we predominantly discuss the nor-
mativity related to the formulation and framing of research questions
in digital journalism studies, and the ways in which this normativity
is quite often hidden and therefore in need of more transparency and
researcher awareness. Robinson et al. (2019, p. 374) argue that “norma-
tive awareness” should be one of the commitments of digital journal-
ism scholars. Our key argument is that digital journalism scholars need
to be more reflexive about their normative presuppositions, implying
also a reflexivity towards how the normativity of other scholars might
influence their work. It is, for example, quite common to argue that
investigating developments in digital journalism are important because
journalism is important to the democratic functioning of a society. Such
normative assumptions are becoming increasingly problematic, as digital
journalism has many other social functions, as new information streams
can carry the same function, and as digital journalism exists also in non-
democratic societies, among other reasons (for more in-depth discussions
of this problem, see Peters, 2019a; Zelizer, 2013).
Kreiss and Brennen (2016) argue there are four norms that are par-
ticularly present in digital journalism studies, namely that digital jour-
nalism should be: 1) participatory, since new technologies and platforms
allow for a transformation of audiences from passive consumers to active
participants; 2) deinstitutionalised, implying that the legacy news insti-
tutions should give up power and that processes and products of jour-
nalism should be decentralised; 3) innovative, to utilise new technology
and create new business models better equipped to deal with the digital
economy; and 4) entrepreneurial, implying that journalists should be self-
starters, brand themselves, and build their own funding and audiences.
Being aware of such norms and how they affect the research is important
to the academic quality of digital journalism.
In this chapter  we first look at the ways in which the relationship
between digital technology and journalism has been, and still is, nor-
matively framed in two opposite future-predictions in digital journal-
ism studies: either in optimistic terms as a saviour of journalism and
potentially also democracy, or as being part of a discourse of crisis in
which digital technology ruins everything that is good about journal-
ism. Then we look particularly at the norms embedded in the discourse
of innovation that dominates much of digital journalism studies. In the
74  The assumptions
concluding section, we will briefly discuss some ways in which we think
it is important that digital journalism studies researchers do take a norma-
tive position.

6.1 The normative future-predictions of digital

journalism studies
On 15 October 2009, the then editor-in-chief of the Norwegian tab-
loid newspaper Dagbladet, Anne Aasheim, stood in front of some of
the members of the New Media Network, an independent Norwegian
consortium for media companies, politicians, consultants, and research-
ers interested in new media trends and developments. Aasheim was
about to give a speech on how Dagbladet was coping with convergence
in times of crisis. She started out by saying: “Today I feel like being an
optimist”.1
She paused and looked as if she were expecting some kind of reaction
reflecting disbelief among members of the audience, who were very well
aware of how hard the economic crisis had hit Dagbladet. No Norwegian
newspaper had experienced a more dramatic drop in both circulation
and turnover than Dagbladet during the previous years; no newspaper had
been obliged to let so many newsroom staffers go. Anne Aasheim didn’t
seem to care.
“Our everyday life is all about crisis”, she continued. “However, I have
been a media executive for 20 years now and I must say, it’s more fun
today than ever before!”
Some members of the audience looked at each other with slightly
raised eyebrows. More fun today? Was she joking? Had it come to a
point where the challenges of keeping a newspaper alive were so massive
that the only way to keep one’s head above water was to laugh about
it – to treat it like a joke? Or was the position Anne Aasheim took this
grey October day in 2009 a reflection of what might be considered the
only feasible solution for a struggling newspaper: to treat the crisis like a
unique opportunity to create change?
Anne Aasheim soon revealed what she had in mind. She flipped up a
PowerPoint slide that read: “The media crisis has given the media com-
panies a new opportunity to pounce on alternative innovation and to
question established truths”. She then said:
“When the crisis becomes big enough, you no longer just mend
things. You tear everything apart, and then you reconstruct it. We are
now searching for the power to introduce disruptive innovation. It’s
going to be a cut-throat competition to have the greatest power of
innovation”.
 The assumptions 75
Then she smiled before exclaiming: “And we’re going to win that
competition!”
Three months later, Anne Aasheim resigned as editor-in-chief of Dagbladet.
By the end of the first decade of the new millennium, many news
companies around the world were in a similar position as the Norwegian
tabloid newspaper Dagbladet. And as was the case for Anne Aasheim,
Dagbladet’s editor-in-chief, many editors and news company owners were
torn between competing discourses related to how to frame and tackle
the situation. A discourse of crisis was among the most prevailing. The
news industry was in a (self-perceived) state of crisis caused by financial
disarray, new ways of distributing news that drew audiences away from
legacy news companies, and an increasing decline in the public’s trust in
news. Simultaneously, there was a discourse of technological optimism hav-
ing a strong hold on the industry. This discourse proclaimed that all the
possibilities of new technology and digital culture in general would work
in favour of journalism. And finally, a discourse of innovation, in which
finding new ways of producing, distributing, and consuming news were
seen as key to the continued success of the news industry. These three,
partly competing, discourses were often present simultaneously, as they
were for the Dagbladet editor when she gave her speech in October 2009.
And they are still dominating the industry as we have entered the third
decade of the new millennium.
What is interesting to us is how these three discourses have found their
way into digital journalism studies as a research field and how they embed
certain normative understandings of journalism and its development in
digital times. Digital journalism studies has been preoccupied with crisis,
technological optimism, and innovation in an empirical sense, implying
that much research has investigated the actual state of crisis, technologi-
cal optimism, and/or innovation. And digital journalism studies has been
preoccupied with crisis, technological optimism, and innovation in a dis-
cursive sense, implying that the perspectives and ways of seeing the world
embedded in these discourses have been adopted as presuppositions in
the research. It is in such research, when scholars take crisis, technologi-
cal optimism, or the benefits of innovation for granted – as givens that
journalism in digital times must adhere to – that digital journalism studies
becomes normative in ways that are hidden and problematic.
We will get back to the discourse of innovation in the next sub-­section.
In this sub-section we will take a closer look at the research related to the
two discourses that deal with future predictions, the discourse of crisis
and the discourse of technological optimism, and we will see how they
both empirically and discursively construct digital journalism in norma-
tive ways.
76  The assumptions
6.1.1 Digital journalism studies and the discourse of crisis
First, we will look at how the discourse of crisis has travelled from indus-
try to research. A substantial number of scholarly publications focusing
on journalism include the word “crisis”, even if the phrase “crisis jour-
nalism”, which signals research on crisis journalism as a beat and not the
crisis of journalism, is excluded. During the 20 first years of the 21 st cen-
tury, Google Scholar returns 270,130 results for the Boolean search term
“journalism” and “crisis” (excluding “crisis journalism”). A similar search
for just the term “journalism” returns 947,100 results. In other words;
29 percent of the research found through Google Scholar on journalism
includes the word “crisis”, thereby suggesting that the discourse of crisis
is prominent in journalism studies. Moreover, as illustrated in Figure 6.1,
the dominance of “crisis” in journalism research increased substantially
during the years 2008 (21 percent) to 2018 (49 percent), in the decade
following the financial crisis.
We should not put too much emphasis on such a Google Scholar
search exercise. The fact that the word “crisis” appears in a scholarly pub-
lication about journalism does not mean that the publication is marked
by a discourse of crisis. For instance, a publication could discuss “crisis”
in a critical fashion, like Chyi, Lewis & Zheng’s (2012) analysis of how

Share of scholarly journalism publicaons

which include the word "crisis"
60%

50%

40%

30%

20%

10%

0%

Figure 6.1 
Google Scholar search on the search terms “journalism” and “crisis”
(excluding “crisis journalism”) and how the result compares to a similar
Google Scholar search on just the word “journalism”.
Note: The search was conducted in January 2020.
 The assumptions 77
newspapers themselves covered the newspaper “crisis”. However, the
importance of the word crisis in scholarly publications about journal-
ism during the 2000s indicates an increased significance of the discourse
of crisis to both journalism studies and digital journalism studies. This
discourse has many dimensions. Nielsen (2016) breaks it down to an eco-
nomic crisis caused by the seemingly impossibilities of making revenue on
online outlets for the legacy news companies; a professional crisis marked
by the blurring of boundaries between journalism and other kinds of
professional work; and a crisis of confidence marked by the public’s increas-
ing distrust in news. Zelizer (2015) argues the challenges to journalism
normally framed as a crisis have many dimensions: a political dimension
(news is under threat from both the left and right side of politics); eco-
nomic (the collapse of old business models); a moral dimension (too many
scandals and violations of ethical standards in journalism); an occupa-
tional dimension (traditional norms and values of journalism no longer
hold); and a technological dimension (digital, social media make visible the
authoritative voice of journalism and its reluctance to respond to calls for
transparency).
No doubt, these dimensions all reflect real challenges that have caused
severe problems for journalism in many countries. It is a well-established
fact that journalism in many countries has suffered financially across
recent decades, predominantly because the advertisement-based business
model of the printed press is not viable in a digital economy increas-
ingly dominated by big platform companies like Google, Amazon, and
Facebook (see chapter 4, section 4.1 for a more in-depth discussion of
this). Similarly, empirical studies show that the public’s trust in news has
declined in many countries (see Newman et al., 2019 and earlier Reu-
ters Institute Digital News Reports), and that anticipations of audience
participation and transparency in journalism has increased, probably as
a result of a cultural shift reflected by digital, social media. Studies also
indicate that these anticipations have not been met by journalism (Singer
et al., 2011) and that the boundaries of journalism as both practice and
profession are blurring (see, for instance, the collection of research essays
in Carlson & Lewis, 2015).
In other words: there are no problems, nor necessarily any norma-
tive assumptions lurking in the background, with the attention digital
journalism studies pays to these challenges. In fact, one might argue that
digital journalism scholars would neglect their core responsibilities if they
did not address the concerns of the industry and the challenges facing
journalism in digital times. However, there are some potential problems
with how researchers might discursively frame these challenges. Adopt-
ing the word “crisis” for any of these challenges irrevocably frames them
78  The assumptions
with a high degree of acute seriousness, with a specific urgency that
places the here-and-now of journalism in a disruptive relationship with
both the past and the future. The discourse of crisis pushes scholars into a
position in which the past is viewed as an endurable phase with manage-
able challenges, the present is perceived as a decisive moment at which
massive changes must take place, and the future is seen as a time marked
by greater uncertainty than ever, a time that relies entirely on journal-
ism’s ability to take drastic measures here and now. According to Zelizer
(2015, p. 892), the word crisis becomes “a way of lexically editing from
the picture alternative realities in order to frame the subject of address
in simplistic, familiar, and strategically useful ways”, which in turn help
“turn murky and troublesome challenges into a controllable phenom-
enon that can be identified, articulated, managed, and ultimately gotten
rid of ”. The discourse of crisis therefore adds an alarmist attitude to the
challenges facing journalism while at the same time interpreting them in
a reductionist and simplistic manner.
This discursive construction of the challenges facing journalism in
digital times therefore has some significant normative underpinnings:
First, it pushes a skewed relationship with time, in which the significance
of the present is overestimated. Digital journalism studies in general is
marked by a bias towards the contemporary, not only because it investi-
gates predominantly the present, but also, as we pointed to in chapter 5,
section 5.3, because to an excessive extent, it relies on references to con-
temporary research. Consequently, digital journalism studies risks treat-
ing current events as both more significant and more unique than they
are, since lessons from the past are not taken into account in a satisfactory
manner.
Second, the crisis discourse creates a bias towards space (or more pre-
cisely: geography) in digital journalism studies (Zelizer, 2015). The dis-
course pushes a universal understanding of the state of journalism, which
implies that there is one crisis in journalism, in singular terms, a crisis that
knows no border or cultural diversity. This, of course, is not true. First
of all, journalism is not in crisis in all parts of the world. When many
scholars speak of the crisis in journalism, what this is often understood to
mean are the challenges that journalism has faced in predominantly West-
ern democracies during the 21st century. Imposing such Western ways of
conceptualising journalism has been a problem in journalism studies in
general, a problem, which might lead to dangerous presuppositions and
over-generalisations of findings (Esser  & Hanitzsch, 2012). A  growing
body of comparative research in recent years has shown that the differ-
ences between journalistic cultures around the global are quite large, and
that national factors much more than cross-national or even global trends
 The assumptions 79
explain variances in journalism in different cultures (Hanitzsch, 2020).
For example, whereas the printed press in most Western countries indeed
has been in a steep decline in the 21 st century, the opposite is true in a
country like India, which has seen a substantial growth in print circula-
tion and advertising at the same time as digital news media has also grown
(Aneez, Chattapadhyay, Parthasarathi, & Nielsen, 2016).
There are not only differences between journalism in Western democ-
racies and other parts of the world; there are also significant difference
between Western democracies (Nielsen, 2016). For instance, the alleged
crisis in trust in journalism varies greatly between countries like Finland,
Denmark, and Portugal, where people still (in 2019) have quite high
trust in the news media, and countries like France and Greece where
the public’s trust in news is much lower (Newman et al., 2019, sec. 1,
p. 19). In addition, regarding the economic state of the news media and
people’s willingness to pay for news, there are major differences. In Nor-
way and Sweden, digital revenues are rising significantly, while the news
industry in other countries has severe problems. Twenty-six percent of
Norwegians have an ongoing subscription to a news medium, while only
6 percent of Germans and Italians have the same (Newman et al., 2019,
sec. 2, p. 33).

6.1.2 Digital journalism studies and the discourse

of technological optimism
Similar to the discourse of crisis, the discourse of technological optimism
is often rooted in an assumed causal relationship between technology
and journalism. But where this causal relationship predicts the doomsday
of journalism within the discourse of crisis, the discourse of techno-
logical optimism envisions a future utopia in which technologies change
journalism for the better. Mosco (2004) has argued that such mythical
discursive powers have dominated the relationships between media, com-
munication, and technology throughout history. The telephone, radio,
television, and computer have all been surrounded by such revolutionary
myths, either pessimistic or optimistic. The 1990s saw several publica-
tions in which authors were profoundly optimistic on behalf of the future
of journalism in new, digital media. Boczkowski (2004) and Domingo
(2006) argued that such early future-predictions were driven by tech-
nological determinism and that research into digital (or online) journal-
ism during the first decade of its existence was partly paralysed by what
Domingo (2006) labelled “utopias of online journalism”. These utopias
were especially related to how hypertext, multimedia, and interactivity
would foster innovative approaches that would revolutionise journalism.
80  The assumptions
These three concepts – hypertext, multimedia, and interactivity – were
central to how the discourse of technological optimism affected digital
journalism studies up until 2010, and even later (Steensen, 2011b).
Even though the research into hypertextual, interactive, and multime-
dia features of digital journalism has become richer and more nuanced
over the years, the discourse of technological optimism has survived as
a normative premise for much of it, alongside, or as a counterpoint to,
the competing discourse of crisis. The problems with this discourse of
technological optimism are 1) its inclination to be driven by techno-
logical determinism; 2) that the significance of technological skills and
assets is overestimated; and 3) that it is based on the assumption that new
technology will benefit journalism. Regarding technological determin-
ism, there is no doubt that a technology-centric approach to studying
digital journalism risks adopting a causal relationship between technology
and practice, in which technology is the one factor that forces change
upon journalistic practice. Influences from sciences and technology stud-
ies and ethnographic approaches to investigate the relationships between
technology and practice have, however, provided much needed nuance
to the ways in which this relationship is understood (see chapter 5, sec-
tion 5.2.2). However, since digital journalism as practice quite often pro-
motes understandings of the relationship between technology and society
in line with technological determinism (Post  & Crone, 2015), there is
always the risk that digital journalism scholars will follow.
The second problem of overestimating the significance of technologi-
cal skills and assets implies that much research, even though it may be
based on a nuanced understanding of the relationship between technol-
ogy and practice, risks placing too much emphasis on technology. The
importance of technological skills and assets for journalists in digital times
has been highlighted by many, but some studies indicate that the per-
ceived need for skills related to digital technology is much higher than
what is actually needed in the everyday practice of digital journalism
(Himma-Kadakas  & Palmiste, 2019), and that too much emphasis on
technical skills might overshadow the important basic skills in journal-
ism, related for instance to critical thinking and accountability (Ferrucci,
2018). Moreover, in in their review of a decade of research on social
media and journalism, Lewis and Molyneux (2018) argue that much of
this research has been based on the assumption that social media matter
more than other factors for journalism, an assumption, they argue, which
is not necessarily true.
The third problem of taking for granted that new technology could
benefit journalism is also found to be a problem by Lewis and Molyneux
(2018) in their review of social media and journalism research. They
 The assumptions 81
conclude that the research suffers from the normative assumptions that
social media is a net positive for journalism. As recent developments
and research have demonstrated – and as we discussed in chapter 4, sec-
tion 4.1.2 – this is a problematic assumption. The assumed benefits of
social media for journalism, like transparency, reaching new audiences,
breaking news faster, greater variance in sources, more audience par-
ticipation and engagement, have not materialised in the ways assumed
by early research. Moreover, problems like the harassment of journalists
through social media (Chen et al., 2018), the spread of fake news and
other forms of disinformation through social media (Allcott  & Gentz-
kow, 2017), and other forms of “dark participation” (Quandt, 2018) are
indeed signs that social media are not necessarily positive for journalism.
In addition, the effect social media have had on the stream of revenue
coming from advertisements for journalism tip the scale even further to
the negative side for journalism. Some even argue that social media rep-
resent “the single biggest challenge facing journalism today” (Crilley &
Gillespie, 2019, p. 173).

6.2 Digital journalism studies and the discourse

of innovation
As was the case with the discourse of crisis, the discourse of innovation is
quite dominant in journalism studies during the first 20 years of the 21st
century if we look at publications found through Google Scholar searches.
A  search on “journalism” and “innovation” returns 179,750 results for
the years 2000–2019, which amounts to 19 percent of all journalism pub-
lications found when searching for only “journalism”. And similar to the
discourse of crisis, the discourse of innovation seems to have increased in
significance. As is visible in Figure 6.2, the popularity of “innovation” in
Google Scholar search results on journalism grew steadily from 13 per-
cent of all journalism publications in 2008 to 34 percent in 2018, before
falling slightly in popularity in 2019. In other words: one third of all
scholarly publications mentioning “journalism” found in Google Scholar
and published in 2018 also included the word “innovation”.
Innovation research tends to emphasise newness and change. Whether
it is a new idea, a new technology, a new commodity or a new combi-
nation of existing ideas, technologies, or commodities, it is the newness
and its consequences that are under scrutiny. Newness and change are in
other words integral parts of innovation as discourse. This discourse also
emphasises structural factors such as technology and economy as drivers
of change (Steensen, 2013), and it is therefore linked to the discourse
of technological optimism discussed above. Posetti (2018) describes this
82  The assumptions

Share of scholarly journalism publicaons

which include the word "innovaon"
40%

35%

30%

25%

20%

15%

10%

5%

0%

Figure 6.2 Google Scholar search on the terms “journalism” and “innovation” and how
the result compares to a similar Google Scholar search on just the word
“journalism”.
Note: The search was conducted in January 2020.

link between the innovation discourse and technological optimism as the

“Shiny Things Syndrome”, a syndrome which may distract journalism
from its core functions, according to the international journalism innova-
tion leaders Posetti interviewed for her study.

6.2.1 The newness bias

The “Shiny Things Syndrome”, or the fascination with “shiny, new
things” as Kueng (2017) calls it, is related to the newness aspect of the
discourse of innovation. Even though both Posetti’s and Kueng’s find-
ings are related to warnings from the industry itself, the syndrome also
applies to digital journalism studies. Like journalists and editors, research-
ers are drawn to new, shiny things. As we have discussed earlier in this
chapter and also elsewhere in this book (see for instance chapter 2, sec-
tion 2.2.2) new technologies and their potential impact on journalism is a
recurring theme in digital journalism studies. However, the emphasis on
newness in digital journalism studies goes beyond a fascination with new
technology. For instance, the increased popularity of digital ethnography,
both in newsrooms and beyond (see Robinson & Metzler, 2016 for an
 The assumptions 83
overview), bears with it some methodological problems long recognised
in anthropology, for example. This problem concerns an embedded bias
towards behaviourism. When conducting ethnographic research in a
newsroom, for example, the researcher is automatically drawn to activi-
ties, and especially those activities that stand out, that have something
new, not previously observed, to them. Engelman (1960) labelled this
phenomenon the “activity bias” and argued that the emphasis on “overt
activities” in ethnographic research consequently “disregards experience,
negates the obvious complexities of internal behavioural dynamics, and
reduces the behaviour system to an automaton” (1960, p.  158). Even
though ethnography as a method has developed strategies to include
experiences and nuanced understandings of the actors being observed,
the bias towards overt activities is difficult to overcome, simply because an
observation-based research project that does not find something new, and
only confirms what is already known, is not going to attract any atten-
tion. This, of course, is not only a problem with ethnographic research.
In the natural sciences, experiments are rarely published if negative and –
because of an emphasis on novelty – such studies are rarely replicated.
This bias towards things and activities that seemingly stand out, the
newness bias, affect digital journalism studies in numerous ways. The
application of Actor-Network Theory (ANT) and other socio-technical
approaches to researching digital journalism (see chapter 5, section 5.2.2)
is one example. The benefit of such approaches is that they make vis-
ible the importance of non-human actors like technology in how digi-
tal journalism is practiced and developed. However, ANT and similar
approaches can lead to an overestimation of non-human actors like tech-
nology, simply because humans and non-humans belong to different
ontologies and therefore can’t be juxtaposed (Vandenberghe, 2002).
Vandenberghe argues that such approaches misjudge the power relation
between humans and non-humans in their “fetish”-like preoccupation
with objects and artefacts: “[h]owever humans are inter-connected with
non-humans, at the end of the day, it is humans who encounter non-
humans and endow them with meaning, use or value” (2002, p.  55).
There is therefore a risk that non-human actors and actants are ascribed
too much meaning and power.
Another example of the newness bias is the tendency to overesti-
mate both the significance and the newness of the things that stand out
in the research findings. The things that stand out become bigger and
more important because we as researchers choose to focus our atten-
tion on them, not because they necessarily play a more important role
in news production. The first author of this book has been a victim
of this bias when he led a research project on online sports journalism
84  The assumptions
(Steensen, 2011a). He found that “social cohesion” was a new ideal for
sports journalists who live-blogged soccer games, but he overestimated
the significance of this ideal to the professional identity of sports journal-
ists and failed to recognise that this ideal is not new to journalism. Even
in Boczkowski’s (2004) influential ethnographic study of digital initiatives
in online newsrooms we find examples of such overestimations. One of
the initiatives Boczkowski analysed was The Houston Chronicle’s Virtual
Voyager project, which was an advanced multimedia project in the late
1990s, probably one of the most advanced and innovative multimedia
projects produced by an online newspaper at the time. And yet, after
analysing the case, Boczkowski (2004, p. 138) argued: “Regarding more
general analytical matters, the story of the Virtual Voyager allows us to go
deeper into the material dimension of online editorial work”. Such an
analytical move from the particular to the general is problematic, simply
because Virtual Voyager was an exception. It was the extreme case.

6.2.2 The problems with change and how to deal with them

Strongly connected with newness, change is the other important aspect
of the discourse of innovation. The significance of change to digital
journalism studies is expressed in the introductory chapters of the two
recently published handbooks of digital journalism studies. Eldridge
II and Franklin (2017, p.  4) argue that digital journalism studies “can
be understood through the ways it has embraced unclear definitional
boundaries around journalism as it has experienced radical change in the
past few decades”. Witschge, Anderson, Domingo, and Hermida (2016a,
p. 2) argue that digital journalism studies “need to address changing con-
texts and new practices, need to reconsider theories and develop research
strategies”. Ahva and Steensen (2017) argue that digital journalism stud-
ies has evolved from viewing change as a revolution to change as decon-
struction, implying that digital journalism studies today is preoccupied
with deconstructing previously established notions of what journalism is.
Change is indeed an important aspect of digital journalism studies in
general, perhaps even its fundamental building block. And yet, or perhaps
precisely because of this, the importance of change and innovation is
often taken for granted, for instance in statements like “[t]he only con-
stant in contemporary journalism is change, and innovation is essential to
the survival of the news industry” (Posetti, 2018, p. 8). When research-
ers put much emphasis on the things that change and treat change and
innovation as constants, there is always the risk that the things that do
not change are neglected and that descriptions of change become more
important than, for example, figuring out the deeper relations between
 The assumptions 85
journalism and society. In the words of Peters and Carlson (2019, p. 639):
“one of the dangers in placing change above solidity is the increased dif-
ficulty of moving from the surface to engage in deeper social questions”.
The discourse of innovation will emphasise newness over sameness;
change over continuity; young over old, and – in relation to media and
journalism especially  – digital over analogue. The challenge for us as
researchers who are interested in the developments in journalism is not
to forget the things that stay the same, the things that are old and the
things that are analogue. But even though we are well aware of this, it is
difficult not to get caught up in the discourse of innovation. One way
of balancing the biases of the discourse of innovation is to consider the
potential counter-discourses embedded in the perspectives of transforma-
tion and practice, argues Steensen (2013). The transformation perspective
can complement innovation research by drawing attention to historical
developments and by pointing to the importance of genres and thus the
social function of texts to developments in digital journalism. Journalism
relies to a great extent on genres, implying that journalistic communica-
tion must be recognised as such by the audience in order for the commu-
nication to be successful. The only way of achieving such communicative
success is by relying on genres, understood as recognisable text formats
with specific discursive features that occur in repeated social situations
(Miller, 1984). Genres are in other words conservative, since they rely on
previous communicative experiences and established conventions. They
do not easily change. Remembering this might help in counter-balancing
the normative implications of the innovation discourse.
The practice perspective can complement both the transformation and
innovation perspectives by stressing the importance of micro-sociological
relations to developments in digital journalism, and by pointing out that
the journalistic institutions of today “allow for a different kind of inter-
play between structure and agency, where agency may pave, or block, the
way for innovation diffusion unbound by external macro-societal factors”
(Steensen, 2013, pp. 56–57).

6.3 Concluding remarks
Normativity influences digital journalism studies in many ways. We have
chosen to focus on three discourses, which we argue influence much
of digital journalism studies in normative ways: the discourses of cri-
sis, technological optimism, and innovation. This focus allowed us to
look at some of the ways in which normativity is often hidden in digital
journalism studies in relation to topics that have great importance to the
field. We do not argue that normativity should have no place in digital
86  The assumptions
journalism studies. The point we would like to stress is that normativity is
more common than one perhaps would think. It is important to increase
the awareness about such hidden normativity in digital journalism studies
in order to see how it affects research and how it can be either made more
transparent or countered by framing the research with perspectives from
other, contrasting discourses.
What we have not discussed in this chapter, is the ways in which digital
journalism studies should be normative, perhaps to a greater extent than it
is today. Some of the major societal challenges as we enter the 2020s, like
climate change, the diffusion of disinformation, and political extremism
and polarisation, have impacts on digital journalism and thereby also on
digital journalism studies. For instance, we think that normative assump-
tions should underlie any assessment of how digital technologies are used
to spread alternative “news” realities, misinform certain publics, create
polarisation, foster distrust in research, and so forth. As digital journalism
sees the rise of alternate news sites, which fundamentally challenge tradi-
tional understandings of what journalism is for, its role in democratic soci-
eties and the line between ethical and opportunistic producers of public
affairs, normativity may – in fact – be central to how research questions
should be generated. Digital journalism studies should not take a neutral
stand regarding what is journalism and what is not. Moreover, digital
journalism studies should seek, as one of its missions, to further develop
practices of journalism suited to tackle major societal challenges, perhaps
in line with emerging practise like solutions journalism (­McIntyre, 2019)
or constructive journalism (McIntyre & Gyldensted, 2017).

Note
1 The scene opening this sub-section was originally published in a longer version in
Steensen (2010, pp. 1–3). All quotes, which originally were spoken in Norwegian,
are translated by the authors.
7 The methodologies
How digital journalism is researched

Research methods are tied to theory. The bridge between them is the
research question you ask, which on the one side is connected with
the theoretical assumptions you make, and on the other determines
which methods you can apply. If we return to the imagined research
project we introduced chapter 5, the one involving the fictional news
company The Daily Times, its Snapchat initiative, and a tech start-up,
we can identify how the many research questions that could be asked
in this case would require different methods. For instance, a research
question like to what degree does the case change what it means to be a jour-
nalist would require interviews, either structured in the form of sur-
veys, semi-structured in the form of qualitative, in-depth interviews,
or unstructured as part of ethnographic field work. If your research
question is to what degree the new product manages to get new audiences
interested in public affairs, you would have to apply some kind of audi-
ence research, like focus group interviews, experiments, analysis
of the digital footprints audiences leave behind when consuming
news, or Q-methodology to analyse the media repertoires of indi-
viduals. If you are more interested in whether the new initiative allows
for new and diverse voices to be heard in journalism, you would prob-
ably want to utilise quantitative or qualitative content analysis of the
texts produced in order to trace sources. If you want to go into more detail
and understand not only what kinds of voices are represented, but how
they are represented, you would have to conduct some kind of qualitative
text analysis, like critical discourse analysis, rhetoric analysis, frame
analysis, or similar methods of text analysis. If you are more interested in
the communicative aspects of the journalism produced and how it relates
to other forms of communication, you could perform a genre analysis.
But if you want to find out how the journalism produced on Snap-
chat impacts the information network this social medium constitutes
88  The methodologies
and is part of, you may want to apply some kind of data analytics,
network analysis, or similar, more technically oriented methods. If
your research question is broad, like how the case might affect journalism’s
position and role in society, you would probably use several of the above
mentioned methods and others in a mixed-methods approach and
you would probably like to compare this one case with other cases in a
multiple case study.
The possibilities are in other words almost endless. The interdisci-
plinary nature of digital journalism studies means that the field applies
a wide range of methods from many different disciplines and fields.
A question is, however, to what degree the field has advanced its meth-
odological approaches beyond the common methods traditionally found
in journalism studies in order to address the specific characteristics of
the digital in digital journalism. In their introduction to a special issue of
Digital Journalism on research methods, Karlsson and Sjøvaag (2016b, p. 1)
argue it has not: “While journalism theory has indeed been advanced,
the same can unfortunately not be said about methodologies used in
journalism research”.
This chapter will not provide a complete account of all methods used
in digital journalism studies – such an endeavour would require a book of
its own. Instead, we will focus on the methods that recently have become
available for researchers in the field and the ones that are important in
order to answer research questions related to the themes and topics that
shape the field: technology, platforms, and audiences (see chapters 3 and
4). In other words, the chapter will focus on 1) digital methods suited
to advance content analysis and the analysis of digital journalism in net-
worked spaces; 2) digital ethnography suited to analyse digital journalism
in and beyond the newsroom; and 3) methods suited to analyse how
audiences interact with news. First, however, we will take a look at what
our analysis of articles in the journal Digital Journalism can tell us about
commonalities in methods applied.
In it is important to note from the very start that methods are not only
tied to theory; they are also tied to the sociology of knowledge. Differ-
ent methods embed, to a certain extent, different ways of assessing what
counts as valid knowledge. Choosing a method therefore involves episte-
mological decisions. A recurring theme throughout this chapter is how
new technology and the availability of digital data, both big and small,
create not only new methodological opportunities for digital journal-
ism studies, but also some potential biases and epistemological challenges
related to the kinds of knowledge that numbers can produce and the
significance of that knowledge.
 The methodologies 89
7.1 Methods in Digital Journalism
Methodology is one of the 11 thematic clusters we have identified in our
analysis of keywords in all articles published in Digital Journalism from
2013 to issue 6, 2019 (see Table 2.2 in chapter 2). This thematic cluster
comprises 28 different keywords, which occurred on 105 occasions, as
seen in Table  7.1. Seventy-three articles included at least one of these
methodology keywords, which means that 21 percent of all articles in the
journal had a methodology-oriented keyword. Keywords belonging to
the theory cluster occurred in an equal number of articles, indicating that

Table 7.1 
Unique and clustered keywords in
articles published in Digital Journalism
2013–2019 belonging to the Method-
ology thematic cluster.

Clustered keywords Count

content analysis 20
survey 13
comparative 9
research interviews 7
methods 7
qualitative 7
case study 5
ethnography 5
topic modelling 5
experiment 4
Q methodology 4
action research 2
mixed methods 2
ethnography 2
LDA 2
regression 2
topic comparison 2
cluster analysis 1
focus groups 1
genealogical analysis 1
informed consent 1
text analysis 1
multilevel analysis 1
reproducible research 1
response distribution 1
structural equation modelling 1
text mining 1
topic detection 1
90  The methodologies
authors pay equal attention to theory and methodology. However, some
of the keywords we have classified as belonging to the theory thematic
cluster, like “discourse”, “framing”, and “ANT” (actor-network theory),
are commonly also understood as methodologies. If these where to be
included in Table 7.1 they would be placed quite high up since the key-
word “discourse” occurred in 11 articles, “ANT” in 5, and “framing” in 4.
As is visible in Table  7.1, content analysis is by far the most popu-
lar method in digital journalism studies judged by the degree to which
authors who publish in Digital Journalism signal their methods in keywords.
Surveys and interviews are also popular approaches, as are comparative
methodology, which echoes the increased popularity of comparative
research in journalism studies in general (Hanitzsch, 2020).
A few interesting observations can be made concerning the keywords
listed in Table  7.1. First, most methods are only mentioned once or
twice. This does not necessarily mean that these methods or methodo-
logical concepts are applied on only one or two occasions in the research
published in Digital Journalism, since they can have been applied without
being listed as keywords. Yet, it signals that methodological diversity and
perhaps also experimentation is part of digital journalism studies, a point
which probably is reflected by the fact that “methods” in itself is quite a
popular keyword. Second, quite a few of the keywords refer to statistical
methods: topic modelling, LDA (Latent Dirichlet Allocation, which is a
type of topic modelling), regression, multilevel analysis, structural equa-
tion modelling, and topic detection. This signals that the “quantitative
turn” in journalism (Coddington, 2015) applies also to digital journalism
studies, a point we will get back to later. However, even though such
statistical keywords might signal a turn towards computational methods,
they are not used very often, and we don’t find many examples of com-
putational methods used in digital journalism studies, thereby suggesting
that manual methods are still the norm in the field.
Third, methods traditionally identified with the humanities do not
occur very often; “text analysis” is for instance only listed once as a key-
word. This is a point recognised also in Steensen and colleagues’ (2019)
qualitative analysis of 95 articles published in Digital Journalism, in which
they found that 13 of the 95 articles applied qualitative, humanistic meth-
ods and that the authors applying methods related to qualitative text and
discourse analysis “often seemed to do so without applying the research
tools commonly associated with humanistic text analysis” (2019, p. 331).
If we look at the references to methods literature in the articles pub-
lished in Digital Journalism, we find several references to the methods’
journals Communication Methods and Measures, Field Methods, International
Journal of Social Research Methodology, Sociological Methods & Research, and
 The methodologies 91
Behavior Research Methods. The popularity of these journals is a further
indication of the dominance of social science methods and perspectives in
digital journalism studies, as we discussed in chapter 5. Among the meth-
ods’ monographs or edited volumes referenced, Krippendorf ’s (2004 and
other editions) Content Analysis is the most popular, in addition to other
content analysis literature. Yin’s (2003) Case Study Research is also cited
quite many times, while books like Social Research Methods (Bryman,
2012), Digital Methods (R. Rogers, 2013), Audience Research Methodol-
ogy (Patriarche, Bilandzic, Linaa Jensen, & Jurišić, 2014), Q Methodology:
A  Sneak Preview (Van Exel  & De Graaf, 2005), and Methods of Critical
Discourse Analysis (Wodak & Meyer, 2001) are all cited in more than one
article.
If we look at articles that predominantly discuss methodology pub-
lished in the journal Digital Journalism, we find quite a few. Sixteen articles
published in the journal aim primarily at discussing research methods,
most of them published in a special issue titled Rethinking Research Meth-
ods in an Age of Digital Journalism (issue 1, 2016). About half of these 16
articles discuss various kinds of computational methods, thus signalling
that much of the methodological development within digital journalism
studies is related to how technology can advance research designs. We
will look more closely into this in the next section.

7.2 Numbers, metrics, and computational methods

In their introduction to the special issue of Digital Journalism on research
methods in an age of digital journalism, Karlsson and Sjøvaag (2016b,
p. 1) lament the lack of methodological innovation in digital journalism
studies “despite the many methodological challenges that follow from
the characteristics of digital media and digital journalism”. They argue
that as the object of study changes, old methods of investigating it may
no longer be feasible. Even though methods like content analysis, inter-
views, and surveys have not lost their significance and still will generate
valuable insights into various aspects of digital journalism, the general
emphasis on datafication in journalism, and in societies at large (Van
Dijck, 2014), has opened new opportunities and necessities for what to
analyse and how to do it.
Concerning the what: one core characteristic of digital journalism is,
as we have discussed earlier, especially in chapter 4, section 4.1, its entan-
glement in information networks, dominated by social media platforms.
Analysing such networks, and journalism’s role in them, might require
new methods. A second characteristic, which we discussed in chapters 3
and 4, is the ways in which numbers and metrics have become pivotal
92  The methodologies
in monitoring (and capitalising on) audience behaviour in digital jour-
nalism. Analysing aspects of this development might also require new
research methods. And finally, a third characteristic is that as data analytics
and computational methods become more common in practices of digi-
tal journalism, considering how this “quantitative turn” (Coddington,
2015) affects digital journalism also becomes important.
Concerning the how to research the three whats above: an increasingly
large pool of computational methods and techniques can be used to col-
lect and analyse data on a wide range of topics related to the production,
distribution, and consumption of journalism (see overviews in Bruns,
2016; Larsson, Sjøvaag, Karlsson, Stavelin,  & Moe, 2016). No doubt,
methods developed within computer science and computational linguis-
tics can be used for instance to automate content analysis, to analyse
information networks, and to analyse news use and audience metrics.
We will return to the latter in section 7.4. Here we will first look more
closely at the developments in content analysis in digital journalism stud-
ies, before we discuss computational analysis of information networks,
and if there are any potential problems with such methods.

7.2.1 Advancing content analysis in digital journalism studies

Already in the very first issue of Digital Journalism, Flaounas and colleagues
(2013) argued that the emerging field of computational social science had
much to offer digital journalism studies, especially related to advancing
the method of content analysis. The article presents the findings from a
content analysis of 2.5 million articles in online newspapers, and a main
purpose is to demonstrate “how automated approaches can access both
semantic and stylistic properties of content, and therefore how content
analysis can be scaled to sizes that were previously unreachable” (2013,
p. 102). To achieve such a scaling, the researchers used techniques of data
mining, machine learning, and natural language processing.
Such large-scale, automated content analysis have been used to analyse
for example news formats in all articles published online by the Norwe-
gian public broadcaster NRK during one year (Sjøvaag & Stavelin, 2012);
agenda divergence in Russian and Ukrainian news in the course of the
Ukrainian crisis 2013–2014 (Koltsova & Pashakhin, 2019); the percep-
tion and the conceptualisation of the term “fake news” in news media in
20 countries over a period of eight years (Cunha, Magno, Caetano, Teix-
eira, & Almeida, 2018); and gender representation in 2.3 million articles
from more than 950 online news outlets ( Jia, Lansdall-Welfare, Sudhahar,
Carter, & Cristianini, 2016), to name but a few. Automated, large-scale
content analysis has also been used with aims that span way beyond digital
 The methodologies 93
journalism studies, like Lansdall-Welfare et al. (2017), who analyse cul-
tural and social transitions in the British society through an analysis of
14 percent of all articles published in regional British newspapers over a
time period of 150 years. Moreover, Broussard (2015) shows how auto-
mated, large-scale content analysis can aid the practice of journalism,
especially investigative reporting.
Applying automated, large-scale content analysis methods is not
easy, and it often requires interdisciplinary research teams comprising
advanced skills in computer science, computational linguistics, and/or
statistics in addition to journalism studies. Boumans and Trilling (2016)
offer a toolkit of various approaches and techniques related to automated
content analysis of digital journalism, as do Günther and Quandt (2016),
who offer a step-by-step guide on how to perform such analysis. In this
guide, the authors demonstrate the complexity of the process and warn
that techniques developed within disciplines very different from the social
sciences cannot be adopted without careful preparation, simply because
“computers do not understand texts the way human coders can, and are
only as good as the algorithms they perform” (2016, p. 86).
In addition to the problems of applying techniques from computer sci-
ence to digital journalism studies, there are also problems with applying
principles of traditional content analysis to computational and automated
content analysis. For example, Karlsson and Sjøvaag (2016a, p.  178)
argue there are some problems with using established categories of con-
tent analysis to analyse emerging forms of digital journalism, since the
method and the ways in which it traditionally has been applied in jour-
nalism studies is “grounded in space/time assumptions that resonate with
analogue media, in general, and print media, in particular”. Karlsson and
Sjøvaag therefore suggest two novel approaches to content analysis of
digital journalism: a big data approach, similar to those discussed above,
and what they label “liquid content analysis”, which allows for tracking
the life cycle of a news item, which in digital media is not fixed, like
in a printed newspaper, but appears in many different iterations and is
intertwined with other current and past news. This liquidity of digital
journalism points to the increasing networked nature of contemporary
information flows, of which journalism is part. The next section will
look specifically at methods suited to analyse such networks.

7.2.2 Computational methods and analysis

of information networks
The increased significance of platform companies and the diversification
of means for distributing information in contemporary media landscapes
94  The methodologies
have boosted an interest in the ways in which information from vari-
ous sources diffuses in societies. Phrases like “information networks”
(e.g., Guo & McCombs, 2015), “news networks” (e.g., Domingo et al.,
2015), and “news ecosystem” (e.g., C. W. Anderson, 2016) have become
an integrated part of digital journalism studies, implying that news and
journalism in digital times are increasingly seen as parts of larger infor-
mation systems involving many actors and actants both proprietary and
non-proprietary to journalistic institutions. Analysing such information
and news networks and journalism’s role in them has therefore become
an important task for digital journalism studies. Typical examples include
analysis of how news related to a specific topic or event travels across or
between various platforms. For reasons we will get back to in the next
section, much of this research has focused on the relationship between
journalism and Twitter, like Wang and Guo’s (2018) inter-media agenda
setting analysis of how the discussion about genetically modified mosqui-
toes was framed in news media and on Twitter; Malik and Pfeffer’s (2016)
analysis of the dominance of news organisations on Twitter; and Steensen
and Eide’s (2019) analysis of news flows between traditional, journalistic
media and a national Twittersphere during a terrorist attack.
A common methodological approach in such research is to analyse
hyperlinks. Hyperlinks carry meaning beyond their technical materiality
in digital journalism, as they are associated with values like interactiv-
ity, transparency, credibility, and diversity (De Maeyer & Holton, 2016).
Hyperlinks can therefore be analysed in order to assess, for instance, when,
where, and how often online content travels across different platforms,
or how various information networks are structured. The advantage of
hyperlinks is that they are unique identifiers, which stay the same across
different platforms. Analysing a selection of hyperlinks across platforms
can be done manually by using platform-spesific search tools, but if the
aim is to analyse a large selection of hyperlinks or identify all hyperlinks
on a spesific platform or in a specific network in a given time period,
then computational methods and innovative approaches are required.
An illustrative example is Sjøvaag et  al.’s (2019) study of hyperlinks
in Scandinavian online news sites. In this study, the researchers analysed
22  million hyperlinks from 658 Scandinavian news websites in order
to assess the structural properties of the Scandinavian media system.
They wrote a script that collected hyperlinks from all the websites and
stored the internal links in one place and the external links in a database.
­Seventy-nine  million external links were stored, of which 22  million
were links between the 658 news websites. These 22 million hyperlinks
where then analysed using the Gephi software, including a geolayout
plugin, and the Python package NetwrokX (Sjøvaag et al., 2019, p. 514).
In other words, the methods used required advanced skills in computers
 The methodologies 95
science, and the research team was interdisciplinary, consisting of journal-
ism scholars and computer scientists.

7.2.3 Problems with big data computational methods

Analysing big data sets with computational methods have implications
beyond the research questions asked and answers found. “Big data” does
not only contribute new empirical and analytical opportunities; it also
comes with a certain discursive baggage implying a certain epistemologi-
cal normativity related to what constitutes valuable knowledge. Embed-
ded in this discourse is often an assumption that the bigger the data, the
better the research, and consequently that the more data one can ana-
lyse, the more accurate and valuable is the knowledge produced (Boyd &
Crawford, 2012). Such assumptions not only risk devaluating research
not based on big data analysis, they also risk promoting uninteresting
research simply because the value might be perceived as lying in the pos-
sibilities of capturing big data and not in the knowledge that the data
potentially can produce. However, big data can definitely produce valu-
able knowledge, as the examples above illustrate. The point is that as with
all other research approaches, the value of the knowledge produced is
only as good as the value of the research questions asked, independent of
the size of the data analysed. Digital journalism scholars should therefore
make sure that they have interesting questions to ask before they embark
on analysis of big data.
This potential big data problem is not the only problem computational
methods might cause. For instance, analysing hyperlinks can be problem-
atic because links are quite often stand-ins for what one really wants to
investigate. As argued by Finkelstein (2008), analysing hyperlinks presup-
poses that the links are carriers of relevant content and that it is possible
to measure the authoritativeness of that content by counting the number
of links to it. Analysing hyperlinks is therefore not the same as analysing
content, and there is consequently a degree of what we can call symbolic
replaceability in the kind of hyperlink research that aims at finding answers
to questions regarding the content the links refer to. This points to one
of several problems with quantitative, automated content analysis and
other computational methods in digital journalism studies, namely that
texts, both verbal and non-verbal, which ultimately make up the content
sought to be analysed, have many qualitative aspects, which will get lost
when using quantitative methods. In the words of Karlsson and Sjøvaag
(2016a, p. 189):

When accessing news content as digitally encoded material, we must

realize that what we are studying is not news items as they appear on
96  The methodologies
the screen. Digital news objects cannot be studied in the form that
they appear, but must be broken down to enable quantification – to
again be aggregated to allow for analysis.

This does not mean that quantitative methods should be avoided when
analysing texts, it only means that researchers should be aware what they
can actually analyse with such methods. As Grimmer and Stewart (2013,
p. 269) pointed out: “All quantitative models of language are wrong –
but some are useful”. Big data analysis of texts can be useful to detect
patterns and structural characteristics of large corpuses of content, but
it is not suited to acquire “the deep knowledge and understanding that
can be achieved when researchers engage with the units of analysis on
a one-to-one basis” (Karlsson  & Sjøvaag, 2016a, p.  189). Depending
on the research question asked, combining quantitative methods with
qualitative analysis could therefore be advisable. Adding automation and
machine learning to the quantitative analysis of texts might create addi-
tional problems, because the sampling process and partly also the analysis
might become invisible to the researcher, like a black box. Acknowledg-
ing this problem, Broersma and Harbers (2018) argue that only by mak-
ing transparent the classification process embedded in machine learning
algorithms can researchers employ computational methods in a reliable
and valid way.
Another problem related to the breakdown of content to make it fea-
sible to analyse with computational methods, is that the contextual and
visual elements of the content disappear from the analysis. Images and
layout has always been central elements in journalism, and are so in digital
journalism too, but they are difficult to analyse with automated content
analysis. However, some researcher have found ways to include visual ele-
ments in automated content analysis, like Jia et al. (2016), who analysed
gender bias in news, including both words and pictures.
Restrictions on access to data can also constitute a problem for
researchers, and push their focus in directions where data can be found
instead of where the interesting questions are, much like the joke about
the man searching for something lost in a different place than where it
was lost simply because the light is better where he searches. Platform
companies like Twitter and Facebook have to a large degree commer-
cialised data access, which make it difficult for researchers to analyse the
interplay between journalism and such platforms. Third party services
like Gnip (Twitter) and CrowdTangle (Facebook and Instagram), which
have been acquired by the respective platform companies, provide some
access to the platforms’ APIs to researchers, but only in a restricted fash-
ion. Full access to all Twitter content (the “firehose” API) has become
 The methodologies 97
too costly for most researchers, while CrowdTangle only allows access to
public content on Facebook and Instagram, excluding comments. How-
ever, since Twitter has allowed access to a smaller portion of its content
for free (the “gardenhose” API) and since most Twitter content is public,
analysis of content on this platform has dominated much of the social
media-related digital journalism research way beyond what the actual
significance of the platform would suggest. Digital journalism studies
therefore suffers from a Twitter bias, which is illustrated by the fact that
“Twitter” is the most frequently mentioned social media company in the
platform thematic cluster of keywords we have identified in our analysis
of keywords in articles published in Digital Journalism. “Twitter” is in
fact the third most frequent of all keywords used in Digital Journalism. It
appears 41 times in the 362 articles published between issue 1, 2013 and
issue 4, 2019. By comparison, “Facebook” appears 19 times, even though
this social media platform has a much bigger user base and therefore is of
much higher significance to digital journalism than Twitter.

7.3 Digital ethnography
The networked nature of the production, distribution, and consump-
tion of digital journalism discussed above also poses some challenges for
ethnographic research. Participatory observation, the key method used in
ethnography, became a forceful approach in journalism studies as part of
the classical news production studies during the 1970s, a research tradi-
tion which was brought back to popularity at the beginning of the new
millennium with influential publications such as Boczkowski (2004),
Paterson and Domingo (2008), and later also Domingo and Paterson
(2011), Ryfe (2012), Anderson (2013), and Usher (2014), to name but a
few. This new wave of ethnographic research sought to understand how
the internet and digital technology affected the practices and cultures
of news production. Not only was this research pivotal in establishing
an understanding of how technology and practice mutually shape one
another in newsrooms (see chapter 5, section 5.2.2), it also made appar-
ent that the ethnographic methods of pre-internet news production stud-
ies needed revisions in order to be appropriate for studies of modern,
digital newsrooms.
We discussed one problem related to digital ethnography in chapter 6,
section  6.2.1, namely that this method is associated with an “activity
bias” (Engelmann, 1960) which in many cases will favour newness over
sameness and change over continuity. However, there are also other
problems. First of all – and following from the increasingly networked
nature of news production, distribution, and consumption discussed
98  The methodologies
above – modern, digital newsrooms are much less fixed in time and space
then their more analogue predecessors. They are scattered across multiple
places, platforms, and possibly also organisations, while digital communi-
cation technologies collapse the distance between them. In the words of
Cottle (2007, p. 9): “With journalists and editors based in different loca-
tions but all working on the same story and all able to access, transmit and
edit the same news materials clearly this poses considerable challenges to
today’s ethnographer”. A single researcher cannot be several places at the
same time. This problem therefore limits the data that one researcher can
collect. Having teams of ethnographers present at multiple sites simulta-
neously can therefore be necessary, but is rarely possible because of the
costs involved. However, the discursive practice of news (i.e., the produc-
tion, distribution, and consumption of news) is no longer limited to the
increasingly scattered newsroom. It also involves third-party actors and
platforms, like citizen reporters and social media. Capturing the most rel-
evant aspects of news production might therefore mean looking beyond
the newsroom and tracing important actors and actants elsewhere (C.
W. Anderson, 2011b; Domingo et al., 2015), as well as looking at other
actors than journalists within the newsroom (like tech developers, met-
rics analysts, and marketing personnel) (S. C. Lewis & Westlund, 2015a).
Second, doing ethnographic research about digital news production
is almost impossible without access to key software, like content man-
agement systems and communication applications. Both authors of this
book have experienced, when doing ethnographic fieldwork, the silence
of modern newsrooms, a silence reflecting the digitisation of all com-
munication in applications like Slack and other digital workspace com-
munication and workflow tools. Without access to the tools in use, it
is almost impossible to capture anything sensible about what’s going on
in the production process. Such access is as essential as “getting a news-
room identification badge that lets the researcher come and go as needed
throughout the observation period” (Robinson & Metzler, 2016, p. 455).
Other applications and technological and material artefacts might also
be important, like actively following the involved journalists and others
on social media or tracing which artefacts are important for the produc-
tion process. However, the amount of digital communication data to be
traced, captured, and included in the final analysis can be so overwhelm-
ing that it is an almost impossible task to undertake, simply because “too
much is going on in digital spaces to truly be observed” (Robinson &
Metzler, 2016, p. 456). Furthermore, capturing, storing, and analysing
data from communication applications and other software and artefacts
might involve ethical issues related to harvesting personal data that are
difficult to address properly.
 The methodologies 99
These difficulties aside, the production, distribution, and consumption
of news in digital times is so complex and fast-changing that the insights
brought forth by qualitative, ethnographic research is pivotal in order to
get a sense of how digital journalism develops. A different aspect related
to this is the ways in which audiences and news consumption has become
intertwined with the practices of news making, an aspect we will turn to
in the next section.

7.4 Audience research
A key characteristic of the digital news environment is that audiences
have a magnitude of options to access news on the platforms and times
of their own choosing in a “hybrid media system” (Chadwick, 2013),
in which legacy and emerging media are intertwined. This, combined
with the many ways in which audiences can participate in, contribute to,
and even make their own news production and distribution systems, has
spurred a wave of research interest in the ways in which audiences access,
contribute to, and understand news and journalism. The “audience turn”
(Costera Meijer, 2020) is also reflected in journalistic practice itself, as
journalists, editors, and news companies have become increasingly pre-
occupied with audience reach and engagement (e.g., Chua & Westlund,
2019; Ferrer-Conill & Tandoc, 2018; Nelson, 2018; see also chapter 3,
section 3.3).
Consequently, the methods by which to study audiences and their
interactions with news and journalism in digital times have become
diversified. Classical methods like surveys, interviews, and focus groups
have been accompanied by methods like Q methodology and a range of
digital methods to measure audience engagement and interaction with
news. Some of these methods are similar to the big data and computa-
tional network analysis methods discussed in the previous section, like
for instance using software like CrowdTangle to analyse how audiences
interact with news on Facebook and Instagram (e.g.,  Majo-Vazquez,
Mukerjee, Neyazi,  & Nielsen, 2019). Using such audience metrics for
analysis of audience behaviour can, however, be compromised by “inher-
ent reductionism” (Schrøder, 2016, p. 531) because audiences are being
“reduced to quantifiable aggregates: herds of masses rather than creative
individuals or groups” (Heikkilä & Ahva, 2015, p. 50).
However, the digital traces that audiences leave behind when interact-
ing with news and journalism can also be analysed with more qualitative
approaches, for instance those affiliated with “virtual ethnography”, in
which researchers trace digital discussion forums, comments, or other
user-generated online material in order to get a sense of for instance
100  The methodologies
how various groups of audiences discuss, make sense of, and/or interact
with news (see for instance Bird & Barber, 2007). Such virtual ethnog-
raphies can be combined with computational methods that cast a wider
net over audience interactions with news, as is illustrated by Steensen’s
(2018) analysis of the Norwegian Twitter sphere during a 2011 terror-
ist attack. Overall, mixed-method approaches to audience research are
becoming more common, according to Schrøder (2016), who mentions
Jensen and Sørensen’s (2013) combination of surveys, focus groups, and
virtual ethnographies in their analysis of Facebook users, and a Finn-
ish research project applying various methods in tracing nine different
groups of audiences’ interaction with news and journalism over a period
of one year (see Heikkilä & Ahva, 2015, where the methodology of this
study is discussed).
Another method which has become popular is analysing media or
news repertoires, a concept that reflects the patterns of media and news
use which audiences establish over time (Peters & Schrøder, 2018). The
advantage of such a methodological approach is that it takes spatio-­
temporal relations into account and acknowledges that the ways in which
audiences interact with and relate to news are rooted in both habits estab-
lished over time and socio-cultural contexts. One way of analysing news
and media repertoires is to apply Q-methodology, which aims at tapping
into audiences’ subjective experiences with a specific “discursive uni-
verse” (for instance news) through exposing them to a set of cards con-
taining statements about an aspect of the universe in question (Schrøder,
2016, p. 534). The participants then sort the cards according to which
statements they agree and disagree with.
Adopting Q-methodology to the study of news use is an example of
innovative methodological advancements in digital journalism studies.
Such creative adaptations of methods can be found in a range of audience-
centric digital journalism research, like for instance using mood boards
to make sense of young peoples’ relation to news, or using storytelling,
painting, or even poetry-writing and Lego installations in order to make
sense of how people really relate to news (Costera Meijer, 2016, p. 548).

7.5 Concluding remarks
The methods we have discussed in this chapter are not the only ones cur-
rently being tested in digital journalism studies. Not by far. A range of
research approaches with origins in various disciplines is being adopted, or
is likely to be adopted in the future. Examples include “digital forensics”
(Garfinkel, 2012), a method to detect for instance the origin, validity, and
reliability of digital content; “technography” (Kien, 2009) – ethnography
 The methodologies 101
of technology  – suited to trace the workings and doings of technol-
ogy in social contexts; and conversation analysis of audience interactions
in online news spaces (Steensen, 2014). Experimental studies are also
becoming increasingly popular, for instance to test whether audiences
can spot the difference between robot- and human-produced news (see
for instance Clerwall, 2014; Haim & Graefe, 2017; Waddell, 2018); to
test how journalists respond to new software (Lindholm, Backholm, &
Högväg, 2018); or to develop and test new technological applications
related, for instance, to location-based journalism (Nyre, 2015; Nyre
et  al., 2012). These latter studies point to a new challenge related to
studying news consumption. As news consumption has moved to mobile
devices, it is difficult to assess how the context of news use affects how
news is understood and consumed, simply because the context is con-
stantly changing.
As we have discussed with big data methods in this chapter, there is a
risk that new methods bear with them a fascination that goes way beyond
what they actually can achieve. New methods can no doubt have that
effect, just as new technologies and artefacts can have a blinding effect on
the practitioners of journalism (Posetti, 2018). Digital journalism schol-
ars should therefore remain sceptical concerning the new methods they
apply and ask what this method can achieve that other methods cannot,
and whether applying the new method will provide new, valuable knowl-
edge that otherwise would have remained unknown.
The future of digital journalism studies will undoubtedly imply exper-
iments with even more new methods, some of which will be adopted
from other disciplines, and some, which someone yet has to invent and
develop. What else the future might bring is the topic of the final chap-
ter of this book.
8 The futures
Deconstructions of and directions
for digital journalism studies

We began this book by referring to the Facebook CEO Mark Zuck-

erberg’s appearance before the US Congress in April  2018 following
the Cambridge Analytics scandal. Going full circle, let us now return to
that hearing and reflect on one sentence that the chairman, Republican
congressman Greg Walden, offered during his opening statement. After
praising Zuckerberg for his success with Facebook and characterising
him as “one of the era’s greatest entrepreneurs”, Walden paused for a
moment and said: “I think it is time to ask whether Facebook may have
moved too fast and broken too many things”.
This statement, of course, referred to the Facebook motto “move fast
and break things”, originally articulated to inspire coders to keep on cod-
ing and not worry about the mistakes they made. Facebook abandoned
the motto in 2014 after it had taken on a life of its own as a symbol of
how the company, and other tech giants, had grown incredibly fast while
disrupting entire industries and changing public spheres and people’s
everyday lives with what many perceived as minimal social and moral
responsibility (Taplin, 2017). Indeed, they did move fast. And they broke
things. The motto became an emblematic embodiment of Christensen’s
(1997) theory of disruptive innovation, while simultaneously becoming
an articulation of a discourse that has had a deep impact on both the
journalism sector and digital journalism studies scholarship.
This discourse is the discourse of deconstruction. It was originally
identified by Ahva and Steensen (2017) in their analysis of the fourth
wave of digital journalism scholarship that dominated the field halfway
through the second decade of the millennium. It is not only a discourse
dominating digital journalism studies but a discourse that serves as a typi-
cal example of how the three dimensions of the framework we intro-
duced in chapter  2 are interlinked. These three dimensions  – society,
sector, and scholarship – are important to keep in mind when looking
at the broader pictures of what digital journalism studies is. Society refers
 The futures 103
to global, national, and local changes that influence and interrelate with
the ( journalism) sector, which encompasses the role and developments of
journalism in society as an institution and phenomenon, as well as prac-
tice, service, profession, and product. Scholarship refers to epistemic prac-
tices of producing knowledge, and this book has focused exclusively on
digital journalism studies as a distinct and transformative field intersecting
with journalism studies and several other fields.
The urge both to deconstruct (in practice) and to articulate a need to
deconstruct or a sense of deconstruction (as discourse) is something which
marks both the sector and the scholarship. Within the scholarship of
digital journalism studies, the discourse of deconstruction implies that
researchers are searching to deconstruct core concepts, like “journalists”,
“journalism”, “news”, and “news company”, in order to redefine them.
As Reese noted (2016, p. 3): “[U]nlike many other more settled fields,
journalism research has been obsessed with the very definition of its core
concept – what journalism is”.
However, this urge within scholarship to deconstruct did not emerge
within a vacuum. Scholars repeatedly seek legitimacy for their studies
by making reference to the discourses of disruption and crisis which sig-
nificantly have influenced the journalism sector in recent years, as we
discussed in chapter 6, section 6.1.1. The journalism sector has geared
significant efforts for building its platform presence, closely followed and
studied by researchers. There has been a separation of news from jour-
nalism where news rituals increasingly have moved beyond the produc-
tion, distribution, and consumption of journalism, and are now exercised
by alternative news media (Holt et al., 2019), fake news sites (Robert-
son & Mourão, 2020), and platforms non-proprietary to the news media
(Ekström & Westlund, 2019b). Over the last couple of years, more and
more publishers have shifted their focus back towards their proprietary
platforms, engaging in platform counterbalancing.
The urge to deconstruct doesn’t stop at the sector. The symbolic sta-
tus of the “move fast and break things” motto, and the feeling that cer-
tain, important things indeed have moved too fast and broken too many
things, is a feeling belonging to the dimension of society at large. This
feeling is a sign of how the discourses of deconstruction, disruption,
innovation, and crisis go beyond the sector of journalism to include
many aspects of society, from people’s everyday lives to politics, culture,
and a wide range of industries. This feeling is what makes the “digital”
in digital journalism studies much more than just an emphasis on binary
code, in quite the same way as the Facebook motto transformed from
an original emphasis on practices of coding to a symbolic diagnoses of
contemporary society.
104  The futures
Realising how such discourses connect scholarship with both the sec-
tor and society also makes it apparent that the discourses dominating the
scholarship are rooted in specific societies, and thereby specific sectors
within those societies. Not all societies, and not all sectors, are marked
by a feeling of things moving too fast and breaking too many things. The
discourses of deconstruction, disruption, innovation, and crisis are pre-
dominantly discourses of the journalism sector in Western democracies.
There is a tendency within digital journalism studies to forget this and
assume that these discourses are not discourses originating from within
specific societies and their sectors, and instead view them not at all as
discourses but as universal facts. Such misconceptions should be avoided,
and the scholarship should be more aware of the normative assumptions
rooted in these discourses, as we discussed in chapter 6. It is important
to have the three dimensions, scholarship, sector, and society and how
they are connected, both discursively and in practice, in mind when for-
mulating research questions, collecting data, analysing the findings, and
drawing conclusions.
In this final chapter we offer some conclusions and directions for future
research which are all rooted in this need to scrutinise the connections
between scholarship, sector, and society. Next we discuss how the schol-
arship of digital journalism studies relates to the sector of journalism,
specifically if the scholarship should be for or about the journalism sector.
Then we briefly summarise the key takeaways from the preceding chap-
ters of this book. We conclude the book by taking the liberty of assessing
what we normatively envision is most important in the road ahead and
chart five key directions for the 2020s. These directions shape the last
section of this chapter.

8.1 Digital journalism studies for or about the sector

One normative approach in digital journalism studies involves developing
research and knowledge about the frontiers of the field. Digital journalism
studies scholars typically engage in basic science for the academic field.
Applied science has ambitions to develop knowledge about the frontiers
of the field for, and potentially with representatives of the journalism sec-
tor. For example, researchers can experiment with emerging technolo-
gies and develop newswork routines for or with newsworkers. While
there are exceptions where scholars engage in such research about jour-
nalism and news (e.g., Bygdås, Clegg, & Hagen, 2019), digital journalism
studies oftentimes travels forward in the back currents of the journal-
ism sector rather than actively participating in shaping its developments
through action research and other forms of applied science in which
 The futures 105
scholars actively contribute to the (re)construction and development of
digital journalism in the journalism sector. Moreover, researching the
innovators in the journalism sector can result in digital journalism studies
scholars indeed contributing to building important knowledge that later
adopters as well as journalism educators can get worthwhile guidance
from. We do not suggest that digital journalism studies scholars exert no
influence on the developments of the journalism sector, as such research
findings and outputs are used in educating newsworkers of the future, and
sometimes also used for policy making, managerial decision-making as
well as public outreach and broader societal informing. Numerous schol-
ars have participated as experts in public inquiries relating to the journal-
ism sector, as well as platform companies, telecom, and media regulation.
Some scholars have presented or conducted research for industry associa-
tions such as World Association of Newspapers (WAN-IFRA) and the
International News Media Association (INMA), as well as for UNESCO
and the World Economic Forum. Digital journalism studies clearly has
much relevance in such contexts, but it means that the research must
be communicated in ways and via means that reach and appeal to such
stakeholders. Examples of such include, but are not limited to, the Digital
News Report series from the Reuters Institute for the Study of Journal-
ism (and other reports in their series) in the UK, as well as reports by
the Pew Research Centre in the US and other research centres. Such
outward-facing reports, or reports written for industry associations, can
devote attention to mapping current developments, and discussion of
ongoing trends, the future, and implications for how managers and policy
makers may solve some of the key problems. We certainly agree on the
importance of such reports, which well can be associated with academic
publications too (written before or after the report).
However, journal articles differ from such reports, imposed with
requirements of academic rigor and making advancements in the field.
Digital journalism studies as a field plays an important role in raising
the level of abstraction and analysis in research, which can help to guide
scholars as well as practitioners and students seeking to explore and push
the frontiers. Altogether, we think there is room for improvement across
digital journalism studies in terms of interdisciplinary approaches, accu-
mulation of knowledge, and development of theory and concepts. This
does not mean that digital journalism studies should be mostly about
the journalism sector. First, it should be sensitive to the societies beyond
the sector and the relations between the sector and the society. Second,
experimental research, applied research, and action research with or
without sector cooperation should strive to build on and further develop
theoretical dimensions in order to reach academic rigor and value to
106  The futures
both the academic community and the sector. Third, making (theoreti-
cal) advancements to the academic field does not mean being irrelevant
to the sector. In sectors, scholarships and societies influenced by dis-
courses of deconstruction, disruption, innovation, and crisis, scholarship
is not the only dimension in which theoretical discussions are relevant.
Having a dialogue about how to deconstruct and redefine core concepts
is relevant to both scholarship and the sector.

8.2 Key takeaways: the formative formations

of the field
This book has offered a multidimensional analysis of the formation of
digital journalism studies since the turn of the new millennium. Chap-
ter 1 briefly charted the course for the entire book and highlighted how
the Facebook Cambridge Analytica scandal illuminated key questions
and debates about what a media company is. Digital journalism stud-
ies is ideally global but continues to have a European/North American
dominance. The chapter introduced our meta-analysis of publications in
the journal Digital Journalism from 2013 to 2019, which we throughout
this book have used as an empirical basis to chart the field. The publica-
tions in this journal, however, do not constitute the whole field of digital
journalism studies. There are many other publications (in other journals,
in books, reports, and so on) that have contributed greatly to its forma-
tion. Nevertheless, we find that the journal Digital Journalism has played a
significant role in shaping digital journalism studies, and that an analysis
of this journal provides valuable insights into what the field looks like and
how it has developed.
Based on this meta-analysis, we concluded in chapter  1 that digital
journalism studies is an interdisciplinary field, but with a strong footing
in sociology and communication. The field focuses on journalism and
digital media yet is marked by having to consider a form of separation
(or dislocation) of news from journalism as a key transition and premise.
The introductory chapter outlined four key premises: 1) a massive shift in
revenue streams because advertisers have largely migrated to the platform
companies, and publishers have developed subscription revenues; 2) an
increased emphasis on audience metrics and analytics in the journalism
sector; 3) shifting patterns of distribution in which platform companies
non-proprietary to institutions of journalism have gained dominance;
and 4) journalism has become more vulnerable to manipulation, disin-
formation, and a consequent lack of public trust.
Chapter  2 explored current debates on how to define both digi-
tal journalism and digital journalism studies. Digital journalism is a
 The futures 107
phenomenon and practice of selecting, interpreting, editing, and dis-
tributing news about public affairs; it is linked to digital technologies
and has a symbiotic relationship with its audiences. It is clearly inter-
related with digital journalism studies as a field. Echoing Eldridge and
colleagues (2019, p. 393), we argue it is important to see digital journal-
ism studies as an interdisciplinary field in its own right, rather than as a
sub-field of journalism studies that could reinforce a journalism-centric
approach rather than the broader interplay between news, digitisation,
and diverse actors in society. However, digital journalism studies schol-
ars have covered some objects of inquiry much more thoroughly than
others, which raises questions about what mechanisms are at play. This
chapter introduced an analytical framework to analyse the dynamic and
mutual relationship between the academic field and its object of inquiry,
a framework consisting of the dimensions society, sector, and scholarship.
The chapter  further made a contribution by linking these dimensions
to four key mechanisms that help advance knowledge into why digital
journalism studies scholars pay much more attention to some topics of
inquiry than others. These mechanisms are: 1) Issue (in)visibility, 2) Pro-
innovation bias, 3) Path dependency, and 4) Addressability.
Chapters  3 and 4 focused on the most dominant thematic clusters
in the research published in Digital Journalism  – technology and plat-
forms. Chapter 3 featured a review of the technology-oriented research
in digital journalism studies and unpacked three key areas: data journal-
ism, analytics and metrics, and algorithms and automation. Each of these
areas has been the subject of one or even two special issues in the journal
Digital Journalism, and they have become more distinct from each other
over time. Moreover, the research in these areas connects with audiences,
the third largest thematic cluster. Few digital journalism studies scholars
have studied technology (as technological actants) per se, nor are there
many that have integrated the study of actors with (technological) actants
or audiences.
Chapter  4 focused on the second most dominant thematic cluster:
platforms. We differentiated between platforms (like social media and
search engines) and digital devices (desktop, smartphones, tablets, and
others). The chapter  introduced a wealth of research into platforms,
which mostly has involved the study of social media platforms in West-
ern contexts. Even though Facebook and WhatsApp are both attracting
more users than Twitter, scholars have more often developed research
designs involving the microblogging service. Few have studied Google,
and few have studied Weibo, WeChat, Telegram, and other platforms
mostly used in Asia and the global south. Much research has been marked
by exploring how news publishers build a platform presence, and how
108  The futures
citizens access and engage with the news in such ways. Relatively few
albeit more and more scholars have approached these developments in a
more critical way in the salient case of journalism, acknowledging that
more is not necessarily better and that even platform counterbalancing
may be necessary. To continue, the chapter  showed that few scholars
have used research designs for news production where they differentiate
between desktop, smartphones, and other devices, although mobile has
quickly displaced desktop as the main gateway for accessing the news.
Chapter 5 was devoted to the role of theory in digital journalism stud-
ies. It discussed a range of different theories adopted and used and what
roles theory plays. Digital journalism studies draws upon theories from a
wide array of fields and disciplines, and scholars are willing to experiment
with new theoretical framings, especially from science and technology
studies. However, there is a degree of path-dependency related to adop-
tions of theoretical frameworks common in journalism studies, mostly
related to sociological and political science perspectives. Perspectives
from the humanities (especially cultural and language studies) are lacking,
implying that the languages, discourses, and sense-making mechanisms of
digital journalism are under-researched, at least in the research published
in Digital Journalism.
Chapter 6 focused on normativity in digital journalism. Digital jour-
nalism scholarship has been, and still is, often normatively framed within a
discourse of crises or a discourse of technological optimism. This norma-
tivity is often hidden or taken-for-granted, in statements like “journalism
is in crises” or “technology has much to offer journalism”, which build
on assumptions of external influences or opportunities affecting journal-
ism and oftentimes fail to recognise the agency of journalism and its prac-
titioners. We agree with those who argue that such assumptions should
be avoided, and that scholars should develop a greater sensitivity towards,
and transparency about, normativity (Althaus, 2012; Carlson et al., 2018).
Moreover, a pro-innovation bias dominates the field, promoted by the
discourse of change and innovation so salient in both the sector and schol-
arship. This bias can make research blind to the things that do not change,
while at the same time overestimating the things that do change. How-
ever, we do not argue that all normativity should be avoided. In fact, we
think scholars should think more carefully about in what respects the field
of digital journalism studies should be normative. We think it is about time
that scholars dare to approach questions related to digital journalism’s soci-
etal role in more normative fashions, related, for instance, to how digital
journalism can contribute to counter disinformation, political polarisa-
tion, and other processes of de-democratisation, and how digital journal-
ism can contribute to solve problems like climate change and pandemics.
 The futures 109
Chapter  7 explored the wealth of methods used in digital journal-
ism studies, which includes some methods originating from informa-
tion science and computer science. Digital journalism studies pays some
attention to advancing content analysis with computational methods.
Opportunities in how to accumulate and analyse data are explored not
only for content analysis, but also for network analysis reflecting the
increasingly networked nature of news distribution and consumption.
We welcome such advancements, but also raise concerns about the risk
of putting too much emphasis on the possibilities offered by emerging
technologies and the availability of data instead of focusing on what the
important research questions are. The chapter also discussed challenges
and advancements in audiences-centric digital journalism scholarship and
in ethnographic approaches.
Altogether, these chapters have offered multiple perspectives on the
formation of digital journalism studies throughout the first two decades
of the 21st century. In the final section of this concluding chapter we turn
to our own normative directions for how we think the field of digital
journalism studies should evolve in the 2020s.

8.3 Directions for digital journalism studies

for the 2020s
Digital journalism studies covers a rapidly growing body of literature pub-
lished in many different scientific venues. In this book we have mostly
drawn on original articles published in Digital Journalism (from 2013 to
2019), and especially so in chapters 3 and 4. While our referencing may
appear to deliver bibliographical richness, it nevertheless represents only a
minor part of the enormous body of literature produced in the field. There
are numerous journalism-focused journals, and there are many more jour-
nals welcoming submissions focusing on journalism and/or news. The
2010s ended with two encyclopedias that each generated hundreds of
entries, and additionally we find several handbooks and book series. Ulti-
mately, scholars in the field produce great amounts of research every year.
Journalism Research News is a Finnish initiative that monitors and briefly
introduces new research about journalism. As of the end of 2019 it moni-
tored 118 journals in addition to books published by recognised publish-
ers, and found more than 1100 relevant publications to journalism studies
in 2019 alone. Turning to January 2020, Journalism Research News had
listed 110 publications, essentially meaning an average of five new publi-
cations in journalism studies every workday of the week.
As discussed, digital journalism studies is interdisciplinary. Thus, schol-
ars may have to navigate far larger research volumes. On the one hand,
110  The futures
advancements of new research and knowledge should be welcomed and
cherished, but ongoing expansions of the field also make it an increas-
ingly difficult scholarly terrain to navigate. This book has shown that
scholars have mobilised around several different thematic clusters, but
even when it comes to the most dominant ones, there is much fragmen-
tation. Scholars oftentimes define narrow boundaries for their articles,
resulting in explicit advancements made being relatively minor.
In this section we take the liberty of assessing what we normatively
envision is most important in the road ahead for digital journalism studies
as a scholarly field. We focus our approach in this narrower way with the
intention to advance the field, while we continue to acknowledge that
scholars can contribute substantially to the development of the journal-
ism sector and society. We write this as digital journalism studies scholars,
and our views can diverge from those of other scholars, but also in com-
parison to media managers and news workers in the journalism sector.
One of our normative points of departure connects with this intro-
ductory discussion, and how digital journalism studies scholars, editors,
and publishers potentially can reconsider their approaches to research and
publishing in the 2020s. A second normative point of departure recon-
nects with our analytical framework proposed in chapter 2, with which
we think it is important for digital journalism studies scholars to stay
tuned to the ongoing developments of society and the journalism sector.
Scholars should account for ongoing changes in the journalism sector
so they can develop timely and important studies while maintaining a
critical perspective. This may sound self-evident, but all too often we
are exposed to research where scholars have stood on the shoulders of
previous studies in developing their research designs, seeking to replicate,
follow up, or add new geographical dimension, while having completely
missed that the journalism sector has chosen or been forced to move on.
Next, we chart a call involving a total of five directions: two directions
for where digital journalism studies should slow down, and three direc-
tions where scholars need to step up.

1 Slowing down and improving overall research activity

  Digital journalism studies is a highly influential and productive
field, which generates a large number of publications every year.
Many scholars are embedded in neo-liberal university environments,
which mean that they are influenced by the so-called “publish or
perish” dogma, encouraging (even expecting) scholars to produce a
high number of publications. Younger scholars without tenure are
especially exposed to such expectations and are benchmarked against
each other when applying for jobs and when going up for tenure.
 The futures 111
Quality, originality, and impact are of course important indicators,
but we cannot look away from the fact that scholars are often evalu-
ated based on the number of articles they produce for high-ranking
journals. Individual scholars who do not fall into line essentially
reduce their chances of a successful career. The downside of this
may be that scholars slice and dice data for maximum output rather
than method triangulation and presenting a holistic account of their
work. This is a structural problem where publishers and the editors
of journals and book series, as well as universities must take respon-
sibility. Such actors can contribute to change in digital journalism
studies, albeit this field is just one tiny fish in the sea.
  We hereby encourage scholars to slow down their overall research
output and change their mindset from maximising the total number
of publications to maximising the depth and breadth of each of their
publications. Journal articles should clearly advance the accumula-
tion of knowledge in the field, something which can be done in
multiple ways, including but not limited to theory-, concept- or
method-development, synthesis, as well original empirical studies.
Clearly word count restrictions are an important factor here, and in
the world of online publishing journals can reconsider word count
restrictions to allow high-quality works where more exhaustive
reviews are in place, and where authors can present mixed methods
and materials in a satisfactory way rather than having to split their
studies into several publications.
  We also encourage publishers not to push for establishing more
journals and increasing the number of issues of established jour-
nals as submissions increase. There is also a tendency that publishers
are driven by market interests and a striving to achieve competitive
advantage instead of quality and originality in outputs, resulting in
many similar publications, like for instance competing encyclope-
dias and handbooks. Finally, we encourage universities to pay more
attention to quality and originality instead of quantity when assessing
academic job applications and tenure promotions.
2 Reducing data-driven research outputs
  The emphasis on API-based research (Venturini & Rogers, 2019),
associated with collecting and analysing trace data as well as so-called
big data, bring specific views on how knowledge is best acquired and
what types of knowledge are needed to find the right answers to the
questions asked by digital journalism researchers. We argue there is
a risk that the availability of new types of data is increasingly setting
the agenda for research questions and not the other way around.
This creates a situation in which advancements of knowledge are
112  The futures
created because there is available data, and not necessarily because
there are new, important, and theory-driven questions that need to
be answered. We certainly subscribe to the view that scholars in the
field should analyse existing datasets when possible from a research
point of view, and when this can advance theory and the field. How-
ever, too often scholars use industry data or cross-sectional surveys
using measures that are not linked to theories or models. Repeatedly,
scholars also carry out studies of their students, who constitute a
very specific group, and thus should be avoided unless the study asks
research questions focusing on student-specific experiences and pop-
ulations per se, and when studying causal mechanisms in that specific
group. It does little good to produce such articles with extensive
discussions of findings and a short disclaimer about limitations in the
end. Scholars need to refrain from conducting research with need for
such disclaimers altogether.
  Moreover, chapter 7 on methods showed that a substantial body of
scholarship employs rather traditional methods, and often only one
method. Clearly, there are constraints in word count to the standard
journal article format, which understandably leads scholars to priori-
tise one method per publication. This calls for more flexibility from
the perspective of journals, accommodating more word count for
multi-method submission. This would correspond well with social
scientists in digital journalism studies, who have long since argued
that one should strive for method triangulation in research designs,
something that enables richer and more diverse understandings of
blind spots in the field and of evolving phenomenon in the journal-
ism sector. At the same time, it should be noted that many schol-
ars guided by approaches from the humanities and cultural studies
resist the very idea of method triangulation and argue for the value
of deep knowledge produced by using one qualitative method. We
agree with this and argue that both deep analysis of specific aspects of
digital journalism and broad knowledge produced by triangulation
are necessary. All in all, we call for more theory-driven research in
the 2020s, hoping to counterbalance the rise of data-driven research.
We also hope that scholars from the humanities and cultural studies
will advance their positions throughout the 2020s, contributing with
theoretically informed research.
3 A more nuanced understanding of the journalism sector
  Our chapters  3 and 4 focused exclusively on objects of inquiry
and gave a glimpse into research produced across the two most domi-
nant thematic clusters of the Digital Journalism journal throughout
the 2010s: technology and platforms. The analytical framework helps
 The futures 113
surface instances where there is a disconnect between what is hap-
pening in the journalism sector and subsequent scholarship. The
international reputation of specific news publishers, such as The New
York Times, The Washington Post, and the Guardian, has resulted in a
great number of studies into these particular companies. At the same
time there is very little research into what Japanese and Indian news
publishers of comparable size and readership do, and also less research
into the local and regional news publishers that often form the back-
bone of journalism. Repetitive studies into some case studies means
others are largely overlooked.
  In this context, Deuze and Witschge (2020) stress that much
research and field formation into the study of journalism has limited
itself to the study of what institutional news publishers do, what they
publish, and how such news material is accessed, while largely over-
looking journalism by individuals and start-ups. We argue that jour-
nalism researchers should consider the porous borders of journalism
and cannot study only what has formerly been seen as its centre. We
must also study peripheral actors (see special issue by Belair-Gagnon
et al., 2019) and alternative news media (Holt et al., 2019) in addi-
tion to other actors beyond the traditional news institutions in the
Western world to understand journalism and related developments.
Despite being a field generating enormous amounts of research
there are clearly many relatively under-researched areas. Scholars
have adopted pro-innovation biases and studied how journalists
and news publishers appropriate and normalise practices with social
media platforms such as Twitter into their work, and for personal
branding. Fewer have studied and critically assessed the downsides
of such ventures, when they become highly dependent on platform
companies to function. Dependence on external parties to achieve
organisational and utilitarian goals boils down to matters of power,
and how journalists and publishers have become more dependent
on platform companies for distribution, exposure, and engagement.
Loss of advertising revenues means they have become more directly
dependent on their audiences to function.
  Moreover, the journalism sector has not communicated widely
about developments of tools and systems designed for digital safety.
Consequently, there is relatively little research into the intersection of
journalism and digital safety, a fundamental component for journal-
ists to continue their work in the 2020s. Moreover, there is a wealth
of research into journalism, news, and politics where political elec-
tions are studied extensively and repeatedly. There is less attention
paid to other realms and issues of politics, such as climate change and
114  The futures
its consequences, and there is less attention paid to genres beyond
political news. Research should pay more attention to different epis-
temologies of digital journalism (Ekström, Lewis, & Westlund, 2020;
Ekström & Westlund, 2019a; Matheson & Wahl-Jorgensen, 2020),
for instance how other genres, formats, and beats develop, like those
associated with entertainment, leisure, human interest, sports, cul-
ture, and consumerism, to name but a few.
4 Towards more diversity
  Much research in digital journalism studies and communication
altogether, is dominated by Western perspectives and studies, espe-
cially by scholars in the US and Western/Northern Europe. Towards
the end of the 2010s there was more and more debate into diversity
among (digital) journalism scholars. Scholars across the globe need
to speak to each other, bringing their very best scholarship to the
forefront. There is a need for better positioning of research, and it
does not make sense to continue traditions where the US is used as
the benchmark, directly or indirectly. Clearly there is also a need for
a more diverse composition of research published, including studies
from all across the globe. This book shows that digital journalism
studies, in the more specific context of research published in Digital
Journalism (2013–2019), has a very strong Western bias: most articles
published in the journal are authored by scholars based in the US
and Western Europe. Some of these scholars originally come from
the global south, or alternatively are based in the West and study
digital journalism in the global south. Ultimately though, Western
perspectives dominate theories, normative assumptions, methods,
and geographical areas being studied. In 2019 there were more and
more calls for, and conversations about, diversity in the field (Hess,
Eldridge, Tandoc, & Westlund, 2019; Mutsvairo, 2019; Rao, 2019;
Wright, Zamith, & Bebawi, 2019). Journalism journals have trans-
formed the composition of their editorial boards to improve the
diversity of voices and expertise across gender, geographies, meth-
ods, and so forth. The academic output is only one side of the coin.
The other side essentially requires a rise in relevant and rigorous
submissions from the global south than can succeed in peer-review
processes and stimulate growth in number of published works from
the global south.
5 Improving accumulation of knowledge
  Research reviews are seldom sharp enough to be systematic and
holistic. Instead, they are at times anecdotal, essentially helping the
author(s) in arguing for their specific study. There are always several
factors one can claim in relation to assertions, such as that no other
 The futures 115
scholar has previously studied factor X or Y, and in this country or
with that method. However, absence of research does not necessar-
ily mean that such research would be worthwhile. On the contrary,
there may well be good reasons scholars have not researched the
topic in question. As discussed, there are also herd behaviours result-
ing in many scholars following suit into the study of specific objects
of study. With rather narrow yet sweeping literature reviews, often-
times reproducing citations to a more limited number of authorita-
tive scholars in the field, contributors are trying to gain legitimacy
for their work. Some scholars are hardly ever cited despite having
produced worthwhile original work. This has to do with issues such
as Western dominance, as well as limitations in expertise about more
diverse literature as well as constraints in word count. Pointing back
to the introduction to this section, and our argument for reduc-
ing research output, authors, reviewers, and editors must become
more observant about saturation and level of contribution, doing
what they can to reduce the field from being flooded with articles
that actually accomplish very little by way of advancing research.
Research should stand on the shoulders of others, taking previous
research findings into consideration, as they seek to advance new
knowledge. As digital journalism studies enters the 2020s there is
opportunity for the scholars in the field to develop more holistic and
systematic approaches in their research reviews and research designs.

These five directions constitute our normative and critical assessment of

what the future of digital journalism studies should hold (alongside more
common held ground such as interdisciplinary research). This assessment,
and other normative and critical assessments we have made throughout
this book, are of course marked by who we are, where we stand, and
where we come from. We are two white males from the Western world,
with tenured positions as full professors at one of the largest universities
in Norway. We move in the same academic circles and attend the same
conferences. Moreover, we hold editorial roles for the journals Digital
Journalism and Journalism Practice, which both have contributed signifi-
cantly to the advancement of digital journalism studies. Our normative
values and our academic standards feed into what is published, alongside
what works are actually submitted and how reviewers assess them.
In recognition of this, we would like to end this book by saluting
the diversity of digital journalism studies and encourage it even further.
In a time when both digital journalism and digital journalism studies
are marked by a discourse of deconstruction, implying that much about
journalism previously taken for granted (like who is a journalist, where
116  The futures
journalism is to be found, what role it plays, what is truth(ful) and how
it can be told, etc.) needs to be redefined, it is paramount that a variety
of voices are heard in both the journalism sector and the scholarship sur-
rounding it. This not only goes for the process of deconstruction; it is
equally important in attempts at reconstructing what journalism should
be in digital societies. We suggest digital journalism studies scholars take
a halt, reconsider their very own normative ideas and approaches, and
sharpen the worthwhileness of their future research agendas. Digital
journalism studies should include both basic and applied science, and we
anticipate the field would benefit from reflecting on the ways in which it
can take on a more socially responsible role, for instance by learning from
responsible research and innovation (RRI) perspectives (Owen, Mac-
naghten, & Stilgoe, 2012). This essentially would entail that scholarship
(i.e., digital journalism studies) not only further develops its link to the
( journalism) sector, but also work towards a scholarly agenda that is even
more socially desirable and acceptable, ultimately advancing the role of
scholarship, and possible also the sector, for society.
Defining the field of digital journalism studies  – what it is, how it
should be researched, and what its future should hold – should not be left
to us alone. We hope this book will spark a debate and increase interest
in how the field develops (and should develop), what the important ques-
tions are, and how to answer them.
Selected references

We list below a few key references mentioned in the preceding text. For
the full bibliography of this book, please visit the online eResource at
www.routledge.com/9780367200909. 

Ahva, L., & Steensen, S. (2017). Deconstructing digital journalism studies. In S. A.

Eldridge II & B. Franklin (Eds.), The Routledge companion to digital journalism studies.
London and New York: Routledge.
Al-Rawi, A. (2019). Viral news on social media. Digital Journalism, 7(1), 63–79.
https://doi.org/10.1080/21670811.2017.1387062
Bechmann, A., & Nielbo, K. L. (2018). Are we exposed to the same “News” in the
news feed? An empirical analysis of filter bubbles as information similarity for Dan-
ish Facebook users. Digital Journalism, 6(8), 990–1002. https://doi.org/10.1080/
21670811.2018.1510741
Belair-Gagnon, V., & Holton, A. E. (2018). Boundary work, interloper media, and
analytics in newsrooms. Digital Journalism, 6(4), 492–508. https://doi.org/10.1080/
21670811.2018.1445001
Boczkowski, P. J. (2004). Digitizing the news: Innovation in online newspapers. Cam-
bridge, MA: MIT Press.
Boczkowski, P. J. (2010). News at work: Imitation in an age of information abundance. Chi-
cago: University of Chicago Press. Retrieved from http://files/5332/books.html
Burgess, J., & Hurcombe, E. (2019). Digital journalism as symptom, response, and
agent of change in the platformed media environment. Digital Journalism, 7(3),
359–367. https://doi.org/10.1080/21670811.2018.1556313
Bygdås, A. L., Clegg, S., & Hagen, A. L. (Eds.). (2019). Media management and digital
transformation. Oxon, UK and New York: Routledge.
Carlson, M. (2018a). Confronting measurable journalism. Digital Journalism, 6(4),
406–417. https://doi.org/10.1080/21670811.2018.1445003
Carlson, M., Robinson, S., Lewis, S. C.,  & Berkowitz, D. A. (2018). Journalism
studies and its core commitments: The making of a communication field. Journal of
Communication, 68(1), 6–25. https://doi.org/10.1093/joc/jqx006
Chua, S., & Westlund, O. (2019). Audience-centric engagement, collaboration cul-
ture and platform counterbalancing: A longitudinal study of ongoing sensemak-
ing of emerging technologies. Media and Communication, 7(1), 153. https://doi.
org/10.17645/mac.v7i1.1760
118  Selected references
Coddington, M. (2015). Clarifying journalism’s quantitative turn. Digital Journalism,
3(3), 331–348. https://doi.org/10.1080/21670811.2014.976400
Deuze, M., & Witschge, T. (2020). Beyond journalism. Cambridge, UK and Medford,
MA: Polity Press.
Diakopoulos, N. (2015). Algorithmic accountability. Digital Journalism, 3(3), 398–
415. https://doi.org/10.1080/21670811.2014.976411
Domingo, D., & Paterson, C. (Eds.). (2011). Making online news (volume 2). Newsroom
ethnographies in the second decade of internet journalism. New York: Peter Lang.
Ekström, M., & Westlund, O. (2019b). The dislocation of news journalism: A con-
ceptual framework for the study of epistemologies of digital journalism. Media and
Communication, 7(1), 259–270. https://doi.org/10.17645/mac.v7i1.1763
Eldridge II, S. A., & Franklin, B. (Eds.). (2019). The Routledge handbook of develop-
ments in digital journalism studies. London: Routledge. https://doi.org/10.4324/
9781315270449
Eldridge II, S. A., Hess, K., Tandoc, E. C., & Westlund, O. (2019). Navigating the
scholarly Terrain: Introducing the digital journalism studies compass. Digital Jour-
nalism, 7(3), 386–403. https://doi.org/10.1080/21670811.2019.1599724
Franklin, B.,  & Eldridge II, S. A. (Eds.). (2017). The Routledge companion to digital
journalism studies. Oxon and New York: Routledge.
Helberger, N. (2019). On the democratic role of news recommenders. Digital Journal-
ism, 1–20. https://doi.org/10.1080/21670811.2019.1623700
Hermida, A. (2010). Twittering the news. Journalism Practice, 4(3), 297–308. https://
doi.org/10.1080/17512781003640703
Lewis, S. C., & Molyneux, L. (2018). A decade of research on social media and jour-
nalism: Assumptions, blind spots, and a way forward. Media and Communication,
6(4), 11. https://doi.org/10.17645/mac.v6i4.1562
Lewis, S. C.,  & Westlund, O. (2015a). Actors, actants, audiences, and activities in
cross-media news work. Digital Journalism, 3(1), 19–37. https://doi.org/10.1080/
21670811.2014.927986
Moyo, D., Mare, A., & Matsilele, T. (2019). Analytics-driven journalism? Editorial
metrics and the reconfiguration of online news production practices in African
newsrooms”. Digital Journalism, 7(4), 490–506. https://doi.org/10.1080/21670
811.2018.1533788
Paterson, C., & Domingo, D. (Eds.). (2008). Making online news. The ethnography of
new media production. New York: Peter Lang.
Peters, C., & Broersma, M. (Eds.). (2013). Rethinking journalism. Trust and participation
in a transformed news landscape. London: Routledge.
Peters, C.,  & Carlson, M. (2019). Conceptualizing change in journalism studies:
Why change at all? Journalism, 20(5), 637–641. https://doi.org/10.1177/146488
4918760674
Singer, J. B., Hermida, A., Domingo, D., Heinonen, A., Paulussen, S., Quandt, T., . . .
Vujnovic, M. (2011). Participatory journalism. Guarding open gates at online newspapers.
Oxford, UK: Wiley-Blackwell.
Steensen, S. (2011). Online journalism and the promises of new technology. Journal-
ism Studies, 12(3), 311–327. https://doi.org/10.1080/1461670X.2010.501151
 Selected references 119
Steensen, S., Grøndahl Larsen, A. M., Hågvar, Y. B., & Fonn, B. K. (2019). What
does digital journalism studies look like? Digital Journalism, 7(3), 320–342. https://
doi.org/10.1080/21670811.2019.1581071
Storsul, T., & Krumsvik, A. H. (2013b). What is media Innovation? In T. Storsul &
A. H. Krumsvik (Eds.), Media Innovations. A Multidiciplinary Study of Change (pp. 13–26).
Gothenburg: Nordicom.
Tandoc, E. C. (2019a). Analyzing analytics. Disrupting journalism one click at a time. Lon-
don: Routledge.
Usher, N. (2016). Interactive journalism: Hackers, data, and code. Urbana: University of
Illinois Press.
Waisbord, S. (2019). The 5Ws and 1H of digital journalism. Digital Journalism, 7(3),
351–358. https://doi.org/10.1080/21670811.2018.1545592
Witschge, T., Anderson, C. W., Domingo, D., & Hermida, A. (Eds.). (2016a). The
SAGE handbook of digital journalism. London, Thousand Oaks and New Delhi: Sage.
Zamith, R. (2018). Quantified audiences in news production. Digital Journalism, 6(4),
418–435. https://doi.org/10.1080/21670811.2018.1444999
Zelizer, B. (2019a). Why journalism is about more than digital technology. Digital
Journalism, 7(3), 343–350. https://doi.org/10.1080/21670811.2019.1571932
Index

Note: Page numbers in italics indicate a figure and page numbers in bold indicate a
table on the corresponding page.

Aasheim, Anne 74 – 75 BBC 36, 43, 59

ABC 36 Behavior Research Methods 91
actor-network theory (ANT) 63 – 64, Bezos, Jeff 50
83, 90 bias: newness 82 – 84; pro-innovation
addressability 24 – 25 22 – 23, 107
affordance theory 62 – 63 big data computational methods 95 – 97
African Journalism Studies 12 blogs 7, 52, 65; micro-blogging
agenda-setting theory 65 – 66 42 – 43, 107
algorithms 28 – 30, 34 – 39 bots 7, 28, 35 – 36
Al Jazeera 32, 58 Bourdieu’s field theory 32, 57, 59
Al Jazeera English 28 Brazilian Journalism Research 12
Amazon 4, 42, 50, 77 building platform presence 42 – 48
analysis of the digital footprints 87 business model canvas 67
analytical framework 19 – 25
analytics 4 – 5, 27 – 28, 30 – 34, 38 – 39, Cambridge Analytica 1, 4 – 5, 11, 44,
106 – 107 49, 106
Android 20, 40, 43 – 44 change 10, 69; the problems with 84 – 85
ANT see actor-network theory Chinese Journal of Communication 12
API 25, 44, 96 – 97, 111 Communication Methods and Measures 90
Apple 20, 31, 40 – 44; Apple News 51 computational methods 93 – 97
artificial intelligence (AI) 28, 34 – 38 content analysis 92 – 93
Asian Journal of Communication 12, 48 content management systems (CMS)
assumptions 72 – 74, 85 – 86; discourse 22, 37
of innovation 81 – 85, 82; normative crisis: discourse of 76 – 79; Google
future-predictions 74 – 81 Scholar search on 76
attitudes 38 – 39, 56 – 58 crisis journalism 76
audience 2 – 5, 25 – 26; assumptions critical discourse analysis 87
72 – 75; audience research 99 – 100; critical theory 68
futures 106 – 107; methodologies CrowdTangle 33, 42, 96 – 97, 99
87 – 88, 91 – 92; platforms 40 – 41, cultural production and discourse 67 – 68
46 – 51, 53 – 54; technologies 27 – 33,
38 – 39; theories 55 – 56, 66 – 68 Dagbladet 74 – 75
automation 34 – 39 Daily Times, The 55 – 57, 59 – 61, 63 – 64,
autonomy 32, 36, 61 66 – 68, 97
 Index  121
data 4 – 5; futures 111 – 112; disciplinary perspectives 10, 10, 25,
methodologies 92 – 93, 98 – 99; 56, 70
platforms 44 – 46, 48 – 49; problems discourse theory 68
with big data computational methods distributions 40 – 41, 46 – 49
95 – 97; technologies 27 – 31, 34 – 39 diversity 114
data analytics 4, 88, 92
data-driven research outputs 111 – 112 ecosystems 1 – 3, 40 – 41, 63, 94
datafication 28, 91 ethnographic field work 87
data journalism 29 – 30 ethnography see digital ethnography
debates 15 – 19
deconstruction 2, 102 – 104, 106, 115 – 116 Facebook 1 – 5, 41 – 46, 49 – 54, 96 – 100,
definitions 15 – 19 102 – 103, 106 – 107
democracy 64 – 66 Field Methods 90
devices see digital devices field theory 32, 57, 59
diffusion of innovation theory 66 focus group interviews 31, 87
digital devices 40 – 41, 45, 51 – 54 four theories of the press 65
digital ethnography 97 – 99 frame analysis 87
digital journalism: as cultural production framework, analytical 19 – 25
and discourse 67 – 68; definitions framing 90
17 – 18; as a democratic force 64 – 66; future-predictions 74 – 81
and digital devices 51 – 53; as post- futures 102 – 104; digital journalism
industrial business endeavour 66 – 67; studies for or about the sector
as a social system 59 – 62; as a socio- 104 – 106; directions for Digital
technical practice 62 – 64; see also Journalism studies 109 – 116; formative
methodologies; platforms; theories formations of the field 106 – 109
Digital Journalism 8 – 12, 10, 13, 17 – 18,
27 – 28, 38 – 39; algorithms and gatekeeping theory 60 – 61
automation 34 – 38; analytics and genre analysis 87
metrics 30 – 34; citation metrics and genre theories 68
ranking 8; data journalism 29 – 30; global diversity 11 – 12
directions for 109 – 116; keywords Google 47 – 48; Google Analytics
26, 89; methods in 89 – 91; and 31; Google News 48; Google
technologies 27 – 29, 32, 34 – 35, 38; Scholar 8, 76, 76, 81 – 82, 82; Google
thematic clusters in 25 – 26, 26 Search 48
digital journalism studies 1 – 4, Guardian 47, 113
8 – 12, 72 – 74, 85 – 86; and content
analysis 92 – 93; definitions and hacks 21, 28
debates 15 – 19, 17 – 18; and the hierarchy of influence theory 60 – 61
discourse of crisis 76 – 79; and the homophily 44, 63
discourse of innovation 81 – 85; Houston Chronicle, The 84
and the discourse of technological Hughes, Chris 50
optimism 79 – 81; for or about the hyperlinks 45, 94 – 95
sector 104 – 106; four structural hypertext 79 – 80
premises for 4 – 6; and global diversity
11 – 12; interdisciplinarity of 9 – 11; ideals 36, 60 – 61
multitude of theories in 58 – 68; information networks 93 – 95
normative future-predictions 74 – 81; innovation, discourse of 81 – 85, 82
separation of news from journalism Instagram 47 – 48
6 – 8; structural premises of 4 – 6; interactivity 45, 79 – 80
theoretical blind spots of 68 – 71; see interdisciplinarity 9 – 11
also assumptions inter-media agenda setting 66, 94
122 Index
International Journal of Social Research New Republic, The 50
Methodology 90 news, separation from journalism 6 – 8;
International News Media Association see also news sites; news values
(INMA) 50, 105 news sites 40, 52 – 53, 86, 94; fake news
interviews 31, 87 sites 103
invisibility 21 – 22 news values 37, 60
iPhone 40 – 42 New York Times, The 30, 47, 113
issue (in)visibility 21 – 22, 107 normativity 72 – 74, 85 – 86; discourse of
innovation 81 – 85; normative future-
journalism: Google Scholar search on predictions 74 – 81
76, 82; separation from news 6 – 8; norms 61, 72 – 73, 77
understanding of 112 – 114; see also NRK 6, 43, 92
digital journalism; digital journalism numbers 91 – 97
studies
Journalism & Mass Communication online journalism 22, 53 – 54, 79
Quarterly 8, 8, 13 organisational development theory
Journalism Practice 8, 8, 13, 115 66 – 67
Journalism Research News 109 organisational theory 60
Journalism Studies 8, 8, 13
Journalism – Theory, Practice & Criticism participatory and deliberative
8, 8, 13 democracy theories 65
path dependency 23 – 24, 66
Kantar 31 platform counterbalancing 48 – 51
keywords 25 – 28, 26, 66 – 67, 89, 89 – 90 platforms 40 – 41, 53 – 54; building
knowledge 114 – 116 platform presence 42 – 48; and digital
devices 51 – 53; and digital journalism
landscapes 5 – 7, 63 41 – 51; platform counterbalancing
legacy news media 22, 27, 40 48 – 51
Los Angeles Times, The 35 post-industrial business endeavour 66 – 67
practice theory 60 – 61
machine learning 28, 92, 96 pragmatist-participatory attitude 57 – 58
media industry 66 procedural or competitive democracy
media logic 61 theories 65
methodologies 87 – 88, 100 – 101; audience professional identity 61, 84
research 99 – 100; digital ethnography professionalism 47, 61 – 62
97 – 99; methods in Digital Journalism pro-innovation bias 22 – 23
89 – 91; numbers, metrics, and
computational methods 91 – 97 Q-methodology 87, 91, 100
Methodology thematic cluster 89
metrics 27 – 28, 30 – 34, 38 – 39, 58 – 59, rational choice theory 66
91 – 97, 106 – 107; see also analytics repertoires 100
mixed-methods approach 88 research, slowing down and
multimedia 79 – 80, 84 improving 110 – 111
multiple case study 88 resource dependence 63
Reuters Institute 31, 105
narrative theory 68 rhetoric analysis 87
network analysis 88, 99, 109
new institutionalism 59 – 60 scholarship 19 – 25
New Media Network 74 Science and Technology Studies
newness bias 82 – 84 (STS) 55, 62, 108
 Index  123
Scopus 8 Telegram 42, 54, 107
SCOT see social construction of thematic clusters 26, 27 – 28, 89, 89 – 90;
technology in Digital Journalism 25 – 26
search engine optimisation (SEO) 48 theories 55 – 58; digital journalism as
second-level agenda setting 65 – 66 cultural production and discourse
sector 19 – 25 67 – 68; digital journalism as a
semiotics 55, 68 democratic force 64 – 66; Digital
Shiny Things Syndrome 82 Journalism as post-industrial business
SJR (SCImago Journal Ranking) 8 endeavour 66 – 67; digital journalism
skills 61, 80, 93 – 94 as a social system 59 – 62; digital
Snapchat 58 – 64, 66, 68, 72 journalism as a socio-technical
social construction of technology practice 62 – 64; multitude of 58 – 68;
(SCOT) 62 – 63 theoretical blind spots of digital
social influence theory 63 journalism studies 68 – 71; theory
social media 20 – 25, 41 – 46, 49 – 54, thematic cluster 90
65 – 66, 80 – 81, 97 – 98 transformation 70 – 71, 73, 85
social media optimisation (SMO) 48 transparency 28, 36, 73, 77, 81,
social-philosophical attitude 57 – 58 94, 108
social system, digital journalism as Twitter 25, 41 – 46, 51 – 54, 94, 96 – 97
59 – 62
society 19 – 25 uses and gratification theory 66
Sociological Methods & Research 90
socio-technical practice 62 – 64 values 36 – 37, 60 – 61, 77, 94, 115
standard attitude 57 – 58 visibility 21 – 22
standards 33, 36, 61, 77, 115
surveys 31, 46, 87, 90 – 91, 99 – 100 Walden, Greg 1 – 2, 102
SVT 6, 43 Wall Street Journal 47
Washington Post, The 30, 47, 50, 113
technological determinism 62, 80 WeChat 4, 42, 47 – 48, 54, 107
technological optimism, discourse of Weibo 4, 47, 107
79 – 81 WhatsApp 42, 44, 46 – 48, 53 – 54, 107
technologies 27 – 28, 38 – 39; algorithms World Wide Web 7, 11, 22, 40 – 42
and automation 34 – 38; analytics
and metrics 30 – 34; data journalism YouTube 47, 54
29 – 30
technologists 28, 30 – 31, 37, 60 Zuckerberg, Mark 1 – 2, 49, 102