Digital Journalism
ISSN: (Print) (Online) Journal homepage: https://www.tandfonline.com/loi/rdij20
Chinese Digital Platform: “We Write What the
Algorithm Wants”
Emeka Umejei
To cite this article: Emeka Umejei (2022) Chinese Digital Platform: “We Write What the Algorithm
Wants”, Digital Journalism, 10:10, 1875-1892, DOI: 10.1080/21670811.2022.2151026
To link to this article:  https://doi.org/10.1080/21670811.2022.2151026
View supplementary material 
Published online: 29 Nov 2022.
Submit your article to this journal 
Article views: 194
View related articles 
View Crossmark data
Citing articles: 1 View citing articles 
Full Terms & Conditions of access and use can be found at
https://www.tandfonline.com/action/journalInformation?journalCode=rdij20
DIGITAL JOURNALISM
2022, VOL. 10, NO. 10, 1875–1892
https://doi.org/10.1080/21670811.2022.2151026
ARTICLE
Chinese Digital Platform: “We Write What the
Algorithm Wants”
Emeka Umejei
Department of Communication Studies, University of Ghana, Accra, Ghana
ABSTRACT KEYWORDS
This study examines Nigerian journalists’ perception of the affor- Audience metrics; digital
dances of recommendation algorithms in news production and intermediaries; journalistic
distribution on Opera News Hub. The study uses a triangulation autonomy; metricisation of
of methods involving content analysis, a semi-structured inter- gatekeeping; Nigerian
view, and a focus group discussion (FGD) with content creators, journalists; Opera NewsHub; soft news and
who have been trained in traditional media practices in Lagos, hard news
Nigeria. The findings indicate news stories published on the Hub
contravene the platform’s publishing guidelines. The Hub’s recom-
mender system is underpinned by “metricisation” of gatekeeping,
which affords soft news over hard news. Hence, content creators
of hard news compromise their journalistic autonomy and take
on a new professional identity in which they engage in a con-
stant struggle to produce sensational content that will attract
audience engagement on Opera News Hub. However, there are
indications that Opera News Hub may enhance media sustainabil-
ity and the welfare of journalists in Nigeria.
Introduction
In 2019, Opera News Hub, a Chinese content-sharing platform that uses recommenda-
tion algorithms for gatekeeping and distribution of media content, was launched in
Nigeria. The platform uses recommendation algorithms that help content creators
“target audience and helps users find the content that matches their interests”
(Operahub 2019). Opera News Hub is arguably the first digital platform in sub-Saharan
Africa that uses recommendation algorithms to review stories before they are pub-
lished on a digital platform. The Hub uses an automated gatekeeping process in which
the recommendation algorithm reviews content before it is published:
Every post must first go through the Opera News Review System before it can reach
millions of Opera News users. Our Review System ensures that we provide all users with
high-quality content and a premium experience. Each post published on Opera News is
first examined by machines and human editors against our content standards
(Operahub 2019).
CONTACT Emeka Umejei mosieds@gmail.com
Supplemental data for this article is available online at https://doi.org/10.1080/21670811.2022.2151026.

 2022 Informa UK Limited, trading as Taylor & Francis Group
1876 E. UMEJEI
As a way to create content that appeals to the Nigerian audience, Opera recruited
senior editors and journalists, who had been trained in traditional media practices as
content creators for Nigerian audiences. The journalists come from diverse journalistic
backgrounds which include society and entertainment, politics, investigative reporting,
and business. This study focuses on this category of content creators because it seeks
to illuminate the perception of African journalists who are experiencing, first-hand, the
role of algorithms in news distribution and production. Studies on the role of algo-
rithms in news production and distribution have been largely carried out in countries
in the Global North where there is an advanced form of democracy. While it has been
suggested that digital media will play key roles in sub-Saharan Africa, their evolution
remains debatable, if not unpredictable (Paterson 2013; Mabweazara 2018). Studies
focusing on the role of algorithms in journalism in sub-Saharan Africa have been lim-
ited to the influence of audience metrics in news production and distribution. Bunce
(2015, 2019), for example, has examined how audience metrics influence the selection
and development of news stories by foreign correspondents in Africa, as well as how
audience metrics can be used to monitor and discipline journalists. Moyo, Mare, and
Matsilele (2019) interrogate the influence of web analytics on editorial decisions and
advertising negotiations in newsrooms in three African countries. The present study is
a novel attempt to bridge this gap in the academic literature on the role of algorithms
in news production and distribution on a digital platform in a transitional democracy
such as Nigeria, where journalists are still grappling with the intrusion of digital inter-
mediaries on traditional norms of journalism.
This article begins with a review of the literature on the role of algorithms in news
production and distribution. The section that follows provides an insight into the
Nigerian media landscape as well as the business model of Opera News Hub. The last
section discusses the findings of this study in terms of the themes that emerge from
the analysis of data.
Role of Algorithms in News Production and Distribution
In traditional media, news production and distribution are vertically integrated. The
advent of the Internet upset this arrangement between production and distribution.
The introduction of digital platforms such as aggregators, search engines and social
media platforms further upended production and distribution. These digital platforms
now act as intermediaries that “filter or suppress the news during its journey from
news producer to news consumer” (Napoli 2019, 59).
Selection of News
Studies focusing on gatekeeping have evolved over time from White’s (1950) concep-
tualization of gatekeeping as an individual judgment of news selection to internal and
external factors that influence the selection of news (Shoemaker and Reese 1996).
These forces – markets, audiences, advertisers, financial markets, sources, public rela-
tions, governments, interest groups, other media and news consultants – are central
to gatekeeping theory (Shoemaker and Vos 2009, 76). The rise of digital media with
audiences functioning as both consumers and producers of media content has raised
DIGITAL JOURNALISM 1877
questions about the continued relevance of gatekeeping theory because the values
informing news selection in journalistic gatekeeping are largely different from those of
non-journalistic gatekeepers (Wallace 2018). In this regard, scholars have argued that
gatekeeping is in “transit” (Heinderyckx and Vos 2016) and requires rethinking (Vos
2015). To account for gatekeeping in digital spaces, Barzilai-Nahon (2008, 1493) sug-
gests that a “theory of network gatekeeping” that accounts for the interactions
between multiple actors is needed. In this networked environment, audiences are
involved in “gatewatching” in which they have access to multiple sources of informa-
tion and no longer depend on journalistic sources of information (Bruns 2005). The
function of gatekeeping is taken over by people outside the newsroom, a situation
akin to Singer’s (2014) concept of secondary gatekeeping in which audiences play
important roles in the gatekeeping process. Shoemaker (2020) proposes that gate-
keeping theory should instead consider the entire web of gatekeepers or the system
composed of elements (gatekeepers), interactions (relationships among them) and a
goal or function. For his part, Wallace (2018, 280) proposes a digital gatekeeping
model that involves two stages: the gatekeeper’s selection process and the gatekeep-
ing mechanism. The model identifies journalists, individual amateurs, strategic profes-
sionals, and algorithms as “gatekeepers that differ in access, selection criteria and the
framing of information, and the publication choice” (Wallace 2018, 288). Furthermore,
publication spaces are conceived as platforms where gatekeepers operate within these
platforms, either applying centralized or decentralized gatekeeping mechanisms.
However, “this new gatekeeping model is outside the control of journalism just as
traditional gatekeeping was mainly outside the control of audiences” (Russell. 2019,
632). For instance, Tandoc (2014,571) observes that the “metricisation” of gatekeeping
could influence the selection and de-selection of stories by editors. Similarly, Vos and
Russell (2019, 2343) suggest that Silicon Valley’s “algorithmic gatekeeping power influ-
ences decisions by journalists or news organizations about what stories to pursue”.
According to Wallace (2018, 280), algorithms select and process information based
on “predefined lines of code”. They afford quantity over quality and “where humans
may apply an individual sense of quality, algorithmic selection struggles to recognize
normative factors”. In comparison to human beings, algorithm gatekeepers are more
efficient and reliable and can perform better at scale (Nechushtai and Lewis 2019,
298). Carlson (2018, 1775), however, explains that “algorithm judgment presents a fun-
damental challenge to news judgment based on the twin beliefs that human subject-
ivity is inherently suspect and in need of replacement, while algorithms are inherently
objective and in need of implementation”. For their part, Nechushtai and Lewis (2019,
304) point out that gatekeeper algorithms should not be evaluated against the norma-
tive standards of journalistic gatekeeping because there are no agreed-upon standards
for gatekeeping in journalism. Similarly, Wallace (2018, 303) contends that “judging an
algorithm by journalistic norms is just as problematic as ignoring the difference
between journalists and the individual amateurs”. However, Napoli (2015) counters
that gatekeeper algorithms should be infused with professional codes of conduct that
advance the public interest. While these studies have made a useful contribution to
the “algorithmification” of gatekeeping, they have been limited to the Global North,
with a limited focus on sub-Saharan Africa.
1878 E. UMEJEI
News Distribution
Digital platforms such as Google, Facebook, and Twitter use algorithms in news pro-
duction and distribution, a process that has been described as the “algorithmification
of gatekeeping” (Heinderyckx 2015, 257). These algorithms “assign relevance, value,
and prominence of knowledge to blend a cocktail of news, information, and entertain-
ment tailored to individual users” (Hermida 2020, 476). On digital platforms,
“algorithms select hashtags for Twitter trending topics, highlight certain headlines
atop Google search results, and determine which items appear prominently in
Facebook News Feeds” (Nechushtai and Lewis 2019, 299). One of the ways in which
US journalists resist being subjected to algorithmic gatekeeping is to cooperate with
“Facebook’s algorithm in terms of how they frame the story, tweaking headlines to be
more engaging to readers, and being deliberate about the photos and videos they
post alongside the stories” (Peterson-Salahuddin and Diakopoulos 2020, 34). In the
same vein, Bandy and Diakopoulos (2020) conclude that the algorithmic curation of
content on Apple’s news privileges soft news over hard news. In order to satisfy the
“workings of digital platforms, news producers create content that is more emotive
and shareable” (Wilding et al. 2018, 2).
Nechushtai and Lewis (2019, 300), however, point out that there are two interre-
lated issues critical to the role of algorithms as gatekeepers of distributed news,
namely, news personalization and news diversity. The former could result in a
“narrower range of news content thereby isolating individuals from a broader set of
information that might challenge their beliefs by giving personalized recommenda-
tions based on past history”. Bodo
 et al. (2019), Thurman et al. (2019), Mo€ller et al.
(2018) and Trielli and Diakopoulos (2020) have studied news recommendation systems
through diverse lenses. Bodo
 et al. (2019, 206) show that quality news organizations
that pursue reader loyalty and trust have a strong incentive to implement personaliza-
tion algorithms. For their part, Mo€ller et al. (2018, 959) conclude that “basing recom-
mendations on user histories can substantially increase topic diversity within a
recommendation set”. In this regard, Trielli and Diakopoulos (2020) show that the
Google search engine partly neutralizes differentiation by providing common results
of people from different ideological backgrounds. These studies are focused on the
algorithmic gatekeeping of distributed news on digital platforms. There is, however,
limited academic research on algorithmic gatekeeping of media content produced by
journalists before publication on a digital platform. This is the unique example of
Opera News Hub, which uses recommendation algorithms to select news produced by
content creators for publication on a digital platform. Opera’s digital platform has
been introduced at a time when the affordances of digital technology are reshaping
the processes of news production, consumption, and distribution in Nigeria.
Obalanlege (2015) and Apuke (2016) point out that the affordances of digital technol-
ogy have transformed the way journalism is practised in Nigeria, especially in terms of
sourcing, newsgathering techniques, reporting methods and the agenda-setting role
of the media. Uwalaka and Watkins (2018) argue that social media has emerged as a
“fifth estate” of the realm that compels the mainstream media in Nigeria to cover
social protests. Similarly, Akinfemisoye (2014) explains that the affordances of digital
technology have served as a link between audiences and mainstream journalists in
DIGITAL JOURNALISM 1879
the coverage of key events such as the 2011 general elections, the Occupy Nigeria
protests of 2012, and the EndSARS protests of the 2020. The affordances of digital
technology are reshaping the work routines of mainstream journalists such that
“journalists now not only have to attend to their various news beats but also have to
monitor online media sites as both competitors and news-lead-providers, in a bid to
produce newsworthy content as well as report what is happening off- and online”
(Akinfemisoye 2014, 70). However, the affordances of digital technology have not suc-
ceeded in “deprofessionalising” journalists. Instead, “mainstream journalism in Nigeria
maintains the dominant discourse by articulating and appropriating content from
social media sources for subtle economic motives” (Akinfemisoye 2014, 62). This brings
us to the research questions of this study:
RQ1: How do content creators perceive the role of recommendation algorithms in news
production and distribution on Opera News Hub?
RQ2: How do content creators respond to algorithmic selection of news on Opera
News Hub?
The Nigerian Media Landscape
The relationship between the Nigerian media and politics has its origins in the history
of the Nigerian press. The early newspapers that championed the quest for Nigeria’s
independence from colonial power were owned by politicians who also deployed the
press to advance partisan interests and objectives. This has since become the norm in
Nigeria where the “politics-journalism alliance” has become entrenched with conse-
quences for ethics (Kolawole 2018). The result has been situations in which allegations
of corruption against top government functionaries are treated with levity (Jibo and
Okoosi-Simbine 2003, 193) and highly placed public officials are deliberately shielded
from embarrassing questions by the media in return for favors (Jibo and Okoosi-
Simbine 2003, 187). The ethnic contestations that characterize Nigerian politics are
also visible in the Nigerian press, manifested in the form of regional parallelism in the
coverage of corrupt practices illustrated by the two major centers of newspaper pro-
duction in the country – the Lagos-Ibadan axis and the Abuja-Kaduna axis (Adebanwi
2002; Yushu’a 2010). According to Yushu’a (2010, 363), regional parallelism reflects the
influence of regional, ethnic, sectional, political and religious considerations in the
practice of journalism. This, he argues, is shaped by location, elites, political and busi-
ness interests as well as economic imbalances in media ownership. For instance, when
the Nigerian press exposes corruption, it is usually divisive in its coverage to the
extent that “when a national issue enters the public domain for debate, the Nigerian
media often, though not all the time, take a North-versus-South position on it” (Jibo
and Okoosi-Simbine 2003, 83). The media in post-independence Nigeria has thus
become “logically partisan in its reportage, agitation and agenda-setting [so] that it
was easier to associate the role of the press with objective or positive nationalism in
the context of the struggle against colonial rule” (Oyovbaire 2001, 4).
One of the biggest problems besetting the media in Nigeria is the failure to pay
salaries to journalists promptly, and it is possible for journalists to be owed for
upwards of one year (Arogundade 2010). These journalists continue to work, hoping
1880 E. UMEJEI
to thrive through unethical and corrupt practices (Jibo and Okoosi-Simbine 2003;
Ibelema 2002). The situation has been complicated by the COVID-19 pandemic which
resulted in the collapse of advertising revenues and a mass sacking of journalists
(Krippahl 2020).
The challenges confronting journalism in Nigeria have helped foreign media organiza-
tions gain a foothold in the digital media space in the country. Foreign ownership is not
alien to the Nigerian media; the Daily Times, Nigeria’s most prominent newspaper was
acquired by the London-based Daily Mirror Group in 1947 and was subsequently nation-
alized in 1974 (Chick 1996). However, in the age of digital media, foreign ownership of
digital media platforms is an emerging trend. Preceding the arrival of the Opera News
Hub in 2019 was legit.ng, owned by Gibraltar-based Genesis Media Limited, which
started out as naija.ng in 2012, and became naij.ng before its transmutation into legit.ng.
It also owns the online news portal yen.com.gh (Ghana), tuco.co.ke (Kenya) and briefly.-
co.za (South Africa). Swiss media company, Ringier AG also started Pulse.ng, a digital
media news platform in 2012 and has expanded to Ghana(pulse.com.gh),
Senegal(pulse.sn), Kenya(pulselive.co.ke), Cote d’Ivoire(pulse.ci), and Uganda(pulse.ug)
(Ringier 2022).
Opera Business Model
Opera was known for its Opera Mini, a simple data compression browser that was
designed for low-income countries in Africa. In 2016, a China-based tech conglomerate
led by Qihoo-360 purchased a quarter of Opera’s software from its Norwegian owners
for $600m. The buyers purchased Opera’s browser business, its privacy and performance
apps, its tech licensing and its brand name (Dent 2016). Since its purchase by the
Chinese conglomerate, Opera has evolved in many ways, emerging as an octopus in the
tech ecosystem in sub-Saharan Africa with interests in file sharing, news, digital pay-
ments, ride-hailing, food delivery, and news production and distribution (Williams 2020).
Opera News Hub is an app available on Android and iOS and it is estimated to
have been downloaded more than 100 times on Android across Africa as of April
2021(Olowogboyega 2021). The Hub operates a personalized news feed in which users
could choose from diverse sources across several news categories such as sports, polit-
ics, entertainment, relationship, lifestyle, technology, etc. Users could also choose to
access their news feed in different languages and have access to 50 news articles.
Content creators on Opera News Hub are mandated to open an account on the plat-
form with either a Google Gmail account or through Facebook account. After the account
has been confirmed, content creators are asked to provide account information, choose a
username, a content category that they would want to write on, upload a profile pic and
write a little about themselves in the biography section. Also, content creators are man-
dated to have an OPay account, or they won’t be able to receive payment. Once an
account has been successfully created on the platform, content creators are directed to
the Opera News dashboard through which they interact with the platform. Herein, con-
tent creators can upload and receive feedback on their stories (Olabimji 2020).
The gatekeeping on Opera News Hub involves a news review system comprising
algorithmic and human components. The recommendation algorithms select content
DIGITAL JOURNALISM 1881
produced by content providers for publication on the Hub as well as distribute news
to the Hub’s audiences. Additionally, the Hub has an aggregator component that
aggregates news from other online sources. News stories that are published on the
platform are expected to meet the following criteria such as originality, professional-
ism, clear logic, good layout, High-definition, and good quality graphics. On the other
hand, news stories are rejected if they: undermine national unity, violate editorial and
writing best practices, misrepresent the country’s political and legal systems, exposed
sensitive state information, incite hatred or discrimination, promote violence and ter-
rorism, pornographic content, poorly written headlines, spelling errors, incomplete or
confusing headline, clickbait, Fake news, fabrication or rumors, outdated content, pla-
giarism, advertisement and poorly edited content (Operahub 2022).
At the outset, Opera relied solely on generating income from Google AdSense but
has since readjusted its business model to placing direct advertising on the platform.
When Opera started in 2019, it recruited senior editors, journalists, and social media
influencers to produce content suitable for Nigerian audiences. This category of con-
tent creators had a fixed contract and earned extra income from the performance of
their stories on the Hub; if their stories garnered about 100,000 clicks, they could earn
as much as N3600 ($8) and if it attracted comments on the platform, content creators
also earned money. However, in recent times, things have changed dramatically. There
are no longer fixed contracts for content creators and all content creators generate
income from the performance of their stories. A more stringent performance model is
also in place, in terms of which content creators no longer earn on the basis of clicks
or comments but on how long audiences spend on their content:
In April 2020, the main factors of payment strategy were ‘content clicks’, ‘content quality’,
and ‘reading time’. ‘Content quality’, ‘reading time’ will be the most important
considerations in the calculation. It means if your article has high quality and a long-
staying time, it will receive a higher traffic fee (Operahub 2020).
The change in engagement measurement from clicks to time spent resulted in the
mass resignation of journalists and influencers because “when we started you could
almost predict through the traffic generated by your stories and the interactions on
your dashboard what you will earn but with the change in engagement measurement,
it was impossible for anyone to be sure what was going on” (FGD).
Methodology
The purpose of this study was to analyze the transformation happening in journalism
in sub-Saharan Africa through the perception of content creators, who experience,
first-hand, the role of algorithms in news production and distribution on Opera News
Hub. A total of thirteen semi-structured interviews and one focus group discussion
(FGD) were conducted in January 2020 and December 2021. First, semi-structured
qualitative interviews were applied to generate responses from the interviewees’
worldview rather than validating the interviewer’s predetermined position (Lindlof
1995, 185). Second, the FGD has the advantage of allowing “researchers to explore
group dynamics, the lifeblood of social activity, as well as to explore the constitutive
power of discourse in people’s lives” (Kamberelis and Dimitriadis 2013, 6). Additionally,
1882 E. UMEJEI
a representative of Opera News was also interviewed in Lagos, Nigeria to gain a work-
ing knowledge of the operation of Opera News Hub.
The first set of interviews comprised twelve participants, and the second involved a
former editor of a national newspaper in Nigeria and an FGD of four participants (all
male). The interviewed comprised seven content creators of hard news and six soft
news while the FGD consisted of three content creators of hard news and one soft
news. The first set of participants was interviewed individually at an agreed destin-
ation in Lagos, Nigeria. The interview varied between 22min and 36min and the inter-
viewees agreed to be anonymous. Eleven were males and one was female. The
interviews were conducted via person-to-person sessions, except for two of the inter-
views that were conducted with a content creator, who was a resident outside Lagos,
and another who could not keep an appointment with me. The second interview and
FGD were conducted virtually and lasted for 66min and 70min, respectively.
Participants in the FGD were paid five thousand Naira (about $12) to purchase data.
The interviews were recorded, transcribed and analyzed manually. Subsequently, draw-
ing on inductive thematic analysis, the interview transcripts were evaluated for recur-
rence, repetition and forcefulness. In this sense, I was observing convergence and
divergence among participants (see Rakow, 2011, 423). The thirteen journalists
selected for this study were drawn from four major beats which included politics, soci-
ety, and entertainment, investigative, and business reporting.
A representative sample comprising fifteen articles was collected from Opera News
Hub between November 9 and 11, 2022. The essence of the thematic content analysis
was to determine the relationship between the Hub’s publishing guidelines and news
stories published on the platform (Dovbysh, Wijermars, and Makhortykh 2022, 7).
Interview Design
This study employed purposive sampling and used snowballing techniques to identify
content creators. In this study, the population consisted of content creators, who had
been trained in traditional media practices in Lagos, Nigeria. The interview guide con-
sisted of questions relating to the experiences and perceptions of the role of recom-
mendation algorithms in news distribution and production on the Opera News Hub.
The goal of the interview was to use the original accounts of content creators, whose
understanding of gatekeeping is informed by traditional media norms, to understand
how they perceive the affordances of recommendation algorithms in news distribution
and production before publication on the Opera News Hub. I asked a set of questions
with the aim of comprehending participants’ perceptions of the role of algorithms in
the selection and distribution of news on the Opera News Hub (see Appendix 1).
Content Analysis
The study employed purposive sampling of content produced by verified content pro-
ducers across the dominant news categories on the Hub such as politics, entertain-
ment, metro, relationship, lifestyle, etc. Also, a codebook was developed for coding
the thematic analysis of the fifteen articles that were collected from Opera News Hub
DIGITAL JOURNALISM 1883
Table 1. Codebook for qualitative content analysis.
Themes Manifestation in content
Misinformation Unverifiable source, omitting facts, false information, misleading information
Plagiarism Lacking originality, no acknowledgment of original source
Sensationalism Sensuous, lacks depth, clickbait, appealing to emotions, intentionally controversial
(see Table 1). Additionally, an independent coder was employed who verified and vali-
dated the themes that emerged from the thematic analysis of data.
Findings
Distribution of News
Participants acknowledged that the Hub’s recommender system is efficient in distributing
personalized news. They mentioned that an algorithm recommendation system is a viable
approach to media enterprise in the country “because it will help you to understand your
target audience and provide them with the kind of content that they require, unlike the
conventional media, where you have to write for everybody and most times nobody
reads it” (Interview with P7). One participant said that the recommendation algorithm
helped journalists to “understand the mood, the street, the market and know the kind of
content that is trending in the country” (Interview with P8). Participants also pointed out
that Opera’s payment policy, whereby content creators were paid on the basis of the per-
formance of their stories, was better than what obtains in traditional media, where salaries
are owed in arrears and the welfare of journalists is not prioritized. According to one par-
ticipant, instead of paying journalists N50,000 (about $100) per month in the traditional
media, it would be better to pay salaries according to the performance of stories:
Why not tell journalists to stay home and send stories to the media organisation; if the
story is good for inside pages, we pay N5,000 ($12) per story and if your stories make the
headline, you earn N20,000 ($40) per story. By so doing, you don’t have to pay salaries
every month but based on the performance of their stories. If Nigerian media
organisations adopt this approach many journalists will work harder and pursue stories
that are exclusive (Interview with P1).
Another participant said that what differentiates Opera news from the conventional
media in Nigeria is that “they take care of their workers; provide adequate gadgets
and [a] conducive environment to work” (Interview with P9). However, participants
also agreed that recommendation algorithms limit the diversity of content on the Hub
because it largely privileges soft news over hard news. The analysis of this section sug-
gests that participants believed that recommendation algorithms enhance distribution
of personalized news and suggested they should be adopted by traditional media
organizations in Nigeria.
Selection of News
The Opera News Review System on the Hub involves two stages. First, the recommenda-
tion algorithm on Opera News Hub has access to all content produced by creators
through the dashboard and selects news stories for publication on the Hub. The
1884 E. UMEJEI
selection of media content is conducted solely by the algorithm, after which human edi-
tors are expected to check through the stories for sensational and clickbaiting content.
Any sensational or clickbaiting story is removed. However, this does not seem to help in
any way as participants complained that, even when their stories were rejected, they
found that all they had to do was to tweak the headline a bit and those stories were
accepted. A senior editor explained that the human components of the gatekeeping
process were not editors, but people who were recruited to manage issues arising from
algorithmic selection of news for publication on the Hub. He explained that their role
was to guide content creators on how to respond to feedback from the algorithm gate-
keeper on the Hub. One participant explained that the role of the recommendation
algorithm was to point out issues in content produced by content creators:
The algorithm points out issues of grammar and misspelling and it will return the script
and highlight the reason the content was not approved. Sometimes it might be a quote
that it considers too long and says it is fake news, or put a code on it. At other times, the
picture you used in your story could be flagged and when you change it, the story will
be approved for publication (Interview with P14).
Another participant explained that it was frustrating for him to rely on algorithms
for the gatekeeping of his content on Opera News Hub:
For me, it was a bit frustrating, especially for someone who was coming from decades of
experience in the newsroom, where you have strict gatekeeping, where you have to go
through various processes before copies are passed; to rely on an algorithm that I felt
was a bit lax in terms of getting all the right mix and variables (FGD).
These responses indicate the ways in which digital intermediaries could transform
traditional norms of journalism in sub-Saharan Africa, where traditional norms of jour-
nalism are still entrenched in the media landscape. Another participant noted that the
Hub’s audiences do not like stories that are deep and intellectually challenging to
write. He said that sometimes if you write a story that is intellectually engaging, it
might seem as if you wasted your time and energy because it might attract fewer
than 300 clicks on the digital platform. He, however, noted that:
When you want to do a story, for instance, to analyse the budget and compare it with
previous health budgets in the last four years, it might not be appreciated by many
people on the Hub. It is very frustrating because I really don’t write sensational stories, I
write serious things, and serious stories don’t get clicks on Opera News Hub (Interview
with P2).
Participants said that the recommender system on the Opera Hub affords soft
news over hard news because those are the kind of news stories that attract audi-
ence engagement on the platform. This tends to reinforce the argument that the
sections on Apple News curated by algorithms consist mainly of soft news, while
those of human editors are hard news and serious stories (Bandy and Diakopoulos
2020). The consensus of participants is that the Hub’s recommender algorithm is so
limited in the selection of media content for publication on the Hub that “you will
be wondering what gate is being kept” (Interview with P14). There are indications
that the ranking criteria of the Hub’s recommending algorithm are underpinned by
audience metrics.
DIGITAL JOURNALISM 1885
Platform Power and Professional Journalism
Participants described the Hub’s recommender system as disregarding individual pro-
fessional differences, suggesting that the “metricisation” of gatekeeping holds both
content creators of soft and hard news to the same content standards. This means
that content creators of soft and hard news negotiate their normative understanding
of journalism against the ranking criteria of the Hub’s algorithm, which is underpinned
by audience metrics. For content creators of soft news, their normative understanding
of journalism is biased toward the ranking criteria of the Hub’s recommender algo-
rithm, but this is not so for content creators of hard news (Bandy and Diakopoulos
2020). In this sense, content creators of hard news are confronted with the option of
compromising their journalistic autonomy or holding on to the traditional values of
hard news. They tend to adapt their normative understanding of journalism to con-
form to the ranking of the Hub’s recommender algorithm by producing sensational
content that will attract audience engagement on the Hub. One participant explained
that it is important for him to sensationalize his report because that was the only way
it would be selected by the Hub’s algorithm:
You try to be a bit sensational so that your story can generate clicks on the platform
because the algorithm is not interested in whether you are professionally sound. So,
sometimes the body of the story may not really align with the headline but because the
headline is what attracts clicks, you must find a way to ensure that you call attention to
it (FGD).
Another participant said that he learned the hard way when his stories could not
generate reasonable income from the Hub. He decided to readjust his normative
understanding of journalism so that it could be selected by the Hub’s recommender
algorithm. He noted that even though he specializes in business reporting if he wants
his story to be published on Opera News Hub, he has to readjust it to suit the kind of
stories that the algorithm likes because the “algorithm now determines what we write;
our training and our experience no longer count”.
So, we can still write about health issues or budget analysis but we have to use a
celebrity in the headline. Perhaps you could interview celebrities to complain about a
health [issue] or the budget; it will garner clicks on the Hub. So, we can write about
serious business issues but use celebrities to headline them; however, if you use activists
or medical doctors to talk about heath or budget, you won’t get clicks on the Hub
(Interview with P2).
Another investigative reporter added that he used to write investigative stories until
he found out the kind of stories that generate audience engagement on the Hub:
I know what the algorithm on Opera News Hub wants; it wants sensationalism and
human-interest stories. I used to write investigative stories until I realised that nobody
was reading them. I came to know that the algorithm on the Hub wants those trending
stories with screaming headlines. Even though the hard news stories that I wrote had
some clicks, not as many as the ones I did on human interest stories (interview with P4).
One participant explained that he could not write serious stories if he wanted to
maximize his presence on the Hub. He said that the monetization of clicks on the Hub
“puts pressure on you”. He noted that his refusal to write “what the algorithm wants”
had cost him income from the Hub:
1886 E. UMEJEI
Unfortunately, I can say that 99% of the stories that I have written are things that I can
defend. I did not adapt to the demand of the algorithm, but that has also cost me money
because I didn’t earn as much as I would have earned because of the poor performance
of these stories on the Hub (P6).
However, content creators were said to have devised a means of gaming the Hub’s
algorithms to generate clicks and comments on stories that have been published on
the Hub. Some content creators were said to have created multiple accounts on vari-
ous social media platforms to generate clicks for their content:
A couple of contributors were suspended from writing over these issues. There was one
content creator that was banned from contributing because he created 82 accounts on
different social media platforms that he used to generate traffic for his own stories
(Interview with P14).
There are indications that there is an economic logic that influences the selection
of content by the Hub’s recommender algorithm that derives from audience engage-
ment with content on the Hub. This reinforces the view that audience engagement is
“imbued with the corporate impulses of the companies that create these algorithms”
(Peterson-Salahuddin and Diakopoulos 2020, 30).
However, the way content creators of soft news respond to the recommender sys-
tem on the Hub differs from those of content creators of hard news. One participant,
who writes entertainment for the Hub, said that the recommender algorithm on the
Hub is only interested in trending and entertainment stories, an approach which he
considers laudable:
The algorithm is doing a good job by helping us to keep a tab on trending stories in
society. That is the story that the audience wants to read because they are interested in
what is happening in the lives of our celebrities and society in general. I think I like that
selection pattern because it does not waste time with stories that are bland (Interview
with P5).
Another participant, who writes society content for Opera News Hub, said that his
background in society reporting was profitable for his content on the Hub:
Writing for Opera News Hub, I realised that what particularly generates clicks is the
headlines and, coming from my background in ‘soft-sell’ journalism, I already knew that
when you give a captivating headline people will click on it even if they don’t read it. So,
that worked for me very well because I already have the background (FGD).
The findings of this section reinforce the argument that algorithm selection differs
significantly from the human selection because, while “humans may apply an individ-
ual sense of quality, algorithmic selection struggles to recognize normative factors”
(Wallace 2018, 282). The findings also justify the argument that “algorithm judgment
presents a fundamental challenge to news judgment based on the twin beliefs that
human subjectivity is inherently suspect and in need of replacement, while algorithms
are inherently objective and in need of implementation” (Carlson 2018, 1775).
The Credibility of Information on Opera News Hub
Results from the content analysis suggest that there is a dissonance between news sto-
ries published on the platform and the Hub’s publishing guidelines. Among the fifteen
DIGITAL JOURNALISM 1887
Table 2. Analysis of content produced by content creators on Opera Hub.
S/N Headline News category Date of publication Classification
1 Hello Ladies – Male Organ Entertainment 10 November Misinformation
Padlock Invented
2 The Story Mysterious Woman Who Metro 9 November Misinformation
Lives Both In Land and Inside Ocean
3 South African Ladies Love Nigerian Entertainment 11 November Misinformation
Men Regardless Of Their Money
And Looks
4 You Think Adolf Hitler Was Terrible? Metro 10 November Plagiarism
There Was A Monster With More
Terror Than Hitler
5 Australopithecus: The Creature That Metro 9 November Plagiarism
Walked Like Humans, & Could Also
Climb Trees Like Apes
6 Things Men Do In Bed That Women Metro 11 November Misinformation
Don’t Like
7 Lady Who Jumped Into Lagos Lagoon Metro 11 November Crude plagiarism
On Thursday Identified As DSS Staff
8 For Men: two Positive Effects of Relationship 10 November Misinformation
Drinking a Glass of Water
Immediately After Sex
9 Meet the first man who give birth to Relationship 10 November Plagiarism
a child
10 Luckiest People in the World. Metro 9 November Plagiarism
11 Stunning photos of curvy African lifestyle 9 November Soft porn/nudity/
American women sensationalism
12 How Birth Order Affect Your Business 10 November Misinformation
Business Sucess
13 Man in this African tribe are forced to Politics 10 November Misinformation
have at least five wives, and get a
house for marrying one more
14 Today’s Headlines: APC; Tinubu Will Politics 10 November Plagiarism
Win 2023-Buhari, Jonathan Drafted
To Reconcile Wike, Atiku-Suswam
15 Video of cybercrime expert Hushpuppi Entertainment 9 November Sensationalism
with a friend surfaces the Internet
articles analyzed, three dominant themes emerged: misinformation, plagiarism, and sen-
sationalism (see Table 2). Among these, seven articles were repleted with misleading
and false information, six were plagiarized stories from other sources including media
organizations in Nigeria, and two were sensational content with some containing explicit
soft porn. For instance, the news story, “Lady who plunged into Lagos Lagoon identified
as DSS staff” was plagiarized from Channels Television and no effort was made to
rewrite it to differentiate it from the original version that was published by Channels TV.
Also, the version published on Opera News Hub has the same headline, the logo of
Channels TV as well as the byline of the Channels TV reporter, Bola Frazier (see Table 2).
Another plagiarized news story, “Meet the first man who give birth to a child” that was
published on the Hub is an example of crude plagiarism. The story was originally pub-
lished in the UK’s Guardian.com on 16 November 2020 (Booth 2020). Two years after, it
was plagiarized by a verified content creator and published on Opera News Hub without
any rewriting or acknowledgment of the source of the content. In another news story,
“Man in this African tribe are forced to have at least five wives and get a house for mar-
rying one more”, the content creator described the tribe that practices this tradition as
the Simba people of Namibia but there is no Simba tribe in Namibia. It is either the con-
tent creator was referring to the Ovahima (Himba) or Jimba people of Namibia and
1888 E. UMEJEI
there is no such practice where a man is rewarded with a house for marrying more than
five wives (Fihlani 2017). These stories published across news categories on Opera News
Hub starkly contravene the publishing guidelines of Opera News Hub and thus confirm
that gatekeeping of media content produced by content creators is underpinned by
audience metrics.
Overall, the findings of this section reinforce the argument that “when algorithmic
or personal selection criteria replace journalistic criteria, misinformation can spread
through networks” (see, for instance, Wallace 2018, 307). Additionally, the findings sug-
gest that information sourced with Opera News Hub may not be healthy for public
consumption because the platform is replete with misinformation, plagiarism, clickbait,
and poorly written news (see Table 2). This raises a concern for Nigeria’s fragile dem-
ocracy because “information is a public good and good information is necessary for
the functioning of a strong democratic state” (see, for instance, WPFD, 2021).
Discussion and Conclusion
This study examined the transformation occurring in journalism in sub-Saharan Africa
through the perception of content creators who are experiencing, first-hand, the role
of algorithms in news production and distribution on Opera News Hub. The findings
of this study suggest that the recommender system on Opera News Hub is effective in
the distribution of personalized news but limited in the gatekeeping of media content
produced by content producers. Participants were unanimous that introducing news
personalization into the Nigerian media landscape could become a game-changer that
would enhance media business and ameliorate the welfare of Nigerian journalists.
However, limited diversity remains a challenge to news personalization because audi-
ences are fed the same diet of news, even when they may want to read different con-
tent (see Nechushtai and Lewis 2019).
Content creators who have been trained in traditional media practices perceive
gatekeeping as comprising factors that are internal or external to media organiza-
tions. However, in the case of the Opera News Hub, the “metricisation” of gatekeep-
ing influences the selection and rejection of news stories produced by content
creators (see, for instance, Tandoc 2014, 571). This indicates that the ranking of the
recommender algorithm on Opera News Hub is underpinned by audience metrics.
The influence of audience metrics on the selection and rejection of stories for publi-
cation on the Hub compels journalists to adapt their professional autonomy to the
demands of the recommender algorithm (Tandoc 2014; Vos and Russell 2019; Bunce
2015, 2019; Moyo, Mare, and Matsilele 2019). However, there are differences in the
way content creators of soft news and hard news respond to the “metricisation” of
gatekeeping on Opera News Hub. Content creators of business, politics and investi-
gative stories negotiate their normative understanding of journalism against the
Hub’s algorithm’s news judgment, which is informed by quantification (see Carlson
2018; Wallace 2018). By contrast, society and entertainment journalists are biased
toward soft news, which strikes a chord with the ranking criteria of the Hub’s recom-
mender algorithm (see Bandy and Diakopoulos 2020). This tends to challenge the
journalistic orientation of content creators, who produce hard news for the Hub’s
DIGITAL JOURNALISM 1889
audiences. Hence, this category of content creators is confronted with the option of
adapting their normative understanding of journalism to conform to the demands of
the Hub’s recommender algorithm and earning an income from the Hub or holding
onto their traditional norms of journalism without deriving an income from the Hub;
most content creators tend to choose the first option. Consequently, they comprom-
ise their journalistic autonomy and take on a new identity in which there is a con-
stant struggle to provide sensational content that will attract audience engagement
on the Hub. This suggests that social media intermediation tends to diminish the
journalistic autonomy and authority of content creators of hard news on Opera
News Hub (see, for instance, Napoli 2019, 70). Hence, there are indications that
digital intermediaries are more likely to cause a transformation of norms and values
of journalism in sub-Saharan Africa than in the Global North because the affordances
of digital technology are yet to “de-professionalise” traditional journalists in Africa
(see, for instance, Akinfemisoye 2014).
Opera News Hub shares similarities with Silicon Valley corporations (Google,
Facebook, and Twitter) in that they thrive on the “metricisation” of gatekeeping of
media content; thus, forcing journalists to orient media content toward audience met-
rics (see, Tandoc 2014; Vos and Russell 2019). However, the difference is that there is
limited public scrutiny of the operations of Opera News Hub from civil society, investi-
gative journalists, and public advocacy. This is not the same with Silicon Valley corpo-
rations that are often heckled by public advocacy activists (see Dovbysh, Wijermars,
and Makhortykh 2022, 16).
Opera News Hub has expanded to six African countries: Nigeria, Egypt, South
Africa, Kenya, Ghana, and Cote d’Ivoire. The expansion of Opera News Hub across
Africa depicts the rising influence of Chinese digital media in the African media space.
The magnitude of misinformation disseminated on the platform suggests that the
platform could easily lend itself to foreign influence operations in the Nigerian media
space (see Table 2). However, Opera’s business model holds a bright spot for media
sustainability and journalists in Nigeria and Africa. There are indications that the grow-
ing influence of Opera News Hub could influence better welfare packages for journal-
ists and the sustainability of media enterprises in Nigeria.
The primary focus of this study is on the perception of content creators, who are
working in a sociotechnical system for the first time. The aim of this study is to pro-
vide a starting point for debates on the ways in which digital intermediation could
transform traditional norms and values of journalism in sub-Saharan Africa. It is antici-
pated that researchers could extend this research to emerging digital intermediaries in
sub-Saharan Africa.
Acknowledgements
I would like to thank Saumya Bhadani, Drs Rotem Medzini, Edwin Tallam, and Professors Philip
Napoli, Regina Lawrence, and Judith Moller, who made useful suggestions to earlier drafts of
this article. I am also grateful to the Social Sciences Research Council (SSRC) that funded a work-
shop where this work was first presented. I would also like to thank Solomon Elusoji, who at
short notice, accepted to serve as an independent coder for the thematic content analysis.
1890 E. UMEJEI
Furthermore, I thank the editors of this special issue and the blind reviewers for their useful sug-
gestions to earlier versions of this article.
Disclosure statement
No conflict of interest has been reported by the author.
References
Adebanwi, W. 2002. “The Nigerian Press and the National Question.” In The National Question in
Nigeria: Comparative Perspectives, edited by A. Momoh & S. Adejumobi, pp. 201–216. England:
Ashgate.
Akinfemisoye, M. O. 2014. “Negotiating Convergence: “Alternative” Journalism and Institutional
Practices of Nigerian Journalists.” Digital Journalism 2 (1): 62–76.
Apuke, O. 2016. “The Impact of Social Media on Conventional Journalism Practice in Nigeria: A
Study of Journalists’ in Jalingo Metropolis.” Canadian Social Science 12 (12): 59–65.
Arogundade, L. 2010, April 15. Opening Remarks. Media Roundtable on the Working and
Welfare Situation of Journalists in Nigeria. International Press Centre, Nigeria. Retrieved from
https://library.fes.de/pdf-files/bueros/nigeria/08021.pdf
Bandy, J, and N. Diakopoulos. 2020. “Auditing News Curation Systems: A Case Study Examining
Algorithmic and Editorial Logic in Apple News.” Proceedings of the International AAAI
Conference on Web and Social Media, 14, 36–47.
Barzilai-Nahon, K. 2008. “Toward a Theory of Network Gatekeeping: A Framework for Exploring
Information Control.” Journal of the American Society for Information Science and Technology
59 (9): 1493–1512.
Bodo
, B., N. Helberger, S. Eskens, and J. Mo€ller. 2019. “Interested in Diversity: The Role of User
Attitudes, Algorithmic Feedback Loops, and Policy in News Personalization.” Digital Journalism
7 (2): 206–229.
Booth, R. 2020, November 16. Trans man loses UK legal battle to register as his child’s father.
Guardian.co.uk. Retrieved from https://www.theguardian.com/society/2020/nov/16/trans-man-
loses-uk-legal-battle-to-register-as-his-childs-father
Bruns, Axel. 2005. Gatewatching: Collaborative Online News Production: Collaborative Online News
Production. New York: Peter Lang Publishing Inc.
Bunce, M. 2015. “Africa in the Click Stream: Audience Metrics and Foreign Correspondents in
Africa.” African Journalism Studies 36 (4): 12–29.
Bunce, M. 2019. “Management and Resistance in the Digital Newsroom.” Journalism 20 (7):
890–905.
Carlson, M. 2018. “Automating Judgment? Algorithmic Judgment, News Knowledge, and
Journalistic Professionalism.” New Media & Society 20 (5): 1755–1772.
Chick, J. 1996. “Cecil King, the Press, and Politics in West Africa.” The Journal of Modern African
Studies 34 (3): 375–393.
Dent, S. 2016, July 18. Opera browser sold to a Chinese consortium for $600 million. Engadget.
Retrieved from https://www.engadget.com/2016-07-18-opera-browser-sold-to-a-chinese-consor-
tium-for-600-million.html?guce_referrer=aHR0cHM6Ly9lbi53aWtpcGVkaWEub3JnLw&guce_referrer_
sig=AQAAANTyYQCNsLDagV6WWQH_WhtL0eg4N5wARVzhZpa_1m8BIgHIXE-v4UugxPvkJCeW12jW-
phbgZMJ3aXIwjl41wg4BzClynmMXh8Iw37oRBH5j4ZBGWHk-2h7c46wvEDw9zvnVcbynjKAavywh
9irAhEIkTSoBmxFTIYjdyTd8S8Q
Dovbysh, O., M. Wijermars, and M. Makhortykh. 2022. “How to Reach Nirvana: Yandex, News
Personalisation, and the Future of Russian Journalistic Media.” Digital Journalism (2022): 1–20.
Fihlani, P. 2017, 31 August. Namibia’s Himba people caught between traditions and modernity.
bbc.com. Retrieved from https://www.bbc.com/news/world-africa-40247310
DIGITAL JOURNALISM 1891
Heinderyckx, F, and T. P. Vos. 2016. “Reformed Gatekeeping.” CM: Communication and Media 11
(38): 29–46.
Heinderyckx, F. 2015. “Gatekeeping Theory Redux.” In Gatekeeping in Transition, edited by T, P.
Vos, & F, Heinderyckx, 253–267. New York: Routledge.
Hermida, A. 2020. “Post-Publication Gatekeeping: The Interplay of Publics, Platforms,
Paraphernalia, and Practices in the Circulation of News.” Journalism & Mass Communication
Quarterly 97 (2): 469–491.
Ibelema, M. 2002. “The Nigerian Press and June 12: Pressure and Performance during a Political
Crisis.” Journalism & Communication Monographs 4 (4): 162–209.
Jibo, M, and A. T. Okoosi-Simbine. 2003. “The Nigerian Media: An Assessment of Its Roles in
Achieving Transparent and Accountable Government in the Fourth Republic.” Nordic Journal
of African Studies 12 (2): 180–195.
Kamberelis, G, and G. Dimitriadis. 2013. Focus Groups. London: Routledge.
Kolawole, S. 2018, April 20. Nigeria media Landscape: Overview. Retrieved from https://medi-
alandscapes.org/country/nigeria
Krippahl, C. 2020, May 14. Africa’s media hit hard by COVID-19 crisis. Deutsche Welle. Retrieved
from https://www.dw.com/en/africas-media-hit-hard-by-covid-19-crisis/a-53427253
Lindlof, T. R. 1995. Qualitative Communication and Research Methods. Thousand Oaks, California:
Sage.
Mabweazara, H. 2018. “African Digital Media Review.” African Journalism Studies 39 (4): 1–2.
Mo€ller, J., D. Trilling, N. Helberger, and B. van Es. 2018. “Do Not Blame It on the Algorithm: An
Empirical Assessment of Multiple Recommender Systems and Their Impact on Content
Diversity.” Information, Communication & Society 21 (7): 959–977.
Moyo, D., A. Mare, and T. Matsilele. 2019. “Analytics-Driven Journalism? Editorial Metrics and the
Reconfiguration of Online News Production Practices in African Newsrooms.” Digital
Journalism 7 (4): 490–506.
Napoli, P. 2019. “Algorithmic Gatekeeping and the Transformation of News Organizations.”
Social Media and the Public Interest, pp. 53–79. New York: Columbia University Press.
Napoli, P. M. 2015. “Social Media and the Public Interest: Governance of News Platforms in the
Realm of Individual and Algorithmic Gatekeepers.” Telecommunications Policy 39 (9): 751–760.
Nechushtai, E, and S. C. Lewis. 2019. “What Kind of News Gatekeepers Do we Want Machines to
Be? Filter Bubbles, Fragmentation, and the Normative Dimensions of Algorithmic
Recommendations.” Computers in Human Behavior 90: 298–307.
Obalanlege, A. 2015. “Journalism Practice and New Media in Nigeria: An Exploratory Analysis of
Journalism Culture and Practice in Nigeria.” New Media and Mass Communication 41: 67–78.
Olabimji 2020. Everything you need to know about Opera News Hub. ng.operanews.com.
Retrieved from https://ng.opera.news/ng/en/technology/581240008433949ef54c964bebb7f04d
Olowogboyega, O. 2021, April 29. More Africans are reading Opera News as it adds 16 million
users in Q1 2021. Techcabal.com. Retrieved from https://techcabal.com/2021/04/29/opera-
news-q1-result/
Operahub. 2020, April. Payment Structure FAQ. Opera News Hub. Retrieved from https://hub.
opera.com/announcement/639c4e5e23f64ba1b6015062b3387a66
Operahub. 2019, May 17. FAQ: How Can I Get More Reach? Opera News Hub. Retrieved from
https://hub.opera.com/login?redirect=%2Fhome%2Fdashboard
Operahub. 2022, November. Publishing Guidelines. Retrieved from https://hub.opera.com/
explain/guidelines
Oyovbaire, S. 2001. “The Media and Democratic Process in Nigeria.” Lecture at the National
Institute for Policy and Strategic Studies, Kuru 7.
Paterson, C. 2013. “Journalism and Social Media in the African Context.” Ecquid Novi: African
Journalism Studies 34 (1): 1–6.
Peterson-Salahuddin, C, and N. Diakopoulos. 2020. “Negotiated Autonomy: The Role of Social
Media Algorithms in Editorial Decision Making.” Media and Communication 8 (3): 27–38.
Ringier. 2022, November 12. Africa. Retrieved from https://www.ringier.com/location/africa/
1892 E. UMEJEI
Shoemaker, P. J. 2020, January. Gatekeeping and Journalism. In Oxford Research Encyclopedia
of Communication. Retrieved from 10.1093/acrefore/9780190228613.001.0001/acrefore-
9780190228613-e-819.
Shoemaker, P. J, and T. Vos. 2009. Gatekeeping Theory. New York: Routledge.
Shoemaker, P. J, and S. D. Reese. 1996. Mediating the Message: Theories of Influences on Mass
Media Content. New York: Longman.
Singer, J. B. 2014. “User-Generated Visibility: Secondary Gatekeeping in a Shared Media Space.”
New Media & Society 16 (1): 55–73.
Tandoc, E. C. Jr, 2014. “Journalism is Twerking? How Web Analytics is Changing the Process of
Gatekeeping.” New Media & Society 16 (4): 559–575.
Thurman, N., J. Moeller, N. Helberger, and D. Trilling. 2019. “My Friends, Editors, Algorithms, and
I: Examining Audience Attitudes to News Selection.” Digital Journalism 7 (4): 447–469.
Trielli, D, and N. Diakopoulos. 2020. “Partisan Search Behavior and Google Results in the 2018US
Midterm Elections.” Information, Communication & Society 25 (1): 145–161.
Uwalaka, T, and J. Watkins. 2018. “Social Media as the Fifth Estate in Nigeria: An Analysis of the
2012 Occupy Nigeria Protest.” African Journalism Studies 39 (4): 22–41.
Wallace, J. 2018. “Modelling Contemporary Gatekeeping: The Rise of Individuals, Algorithms and
Platforms in Digital News Dissemination.” Digital Journalism 6 (3): 274–293.
White, D. M. 1950. “The “Gate Keeper”: A Case Study in the Selection of News.” Journalism
Quarterly 27 (4): 383–390.
Wilding, D., P. Fray, S. Molitorisz, and E. McKewon. 2018. “The Impact of Digital Platforms on
News and Journalistic Content.” University of Technology Sydney, NSW.
Williams, A. 2020, August 5. The China-Funded Mobile Browser That’s Zeroing in on the African
Market. China-Africa Project. Retrieved from https://chinaafricaproject.com/analysis/the-china-
funded-mobile-browser-thats-zeroing-in-on-the-african-market/
World Press Freedom Day. 2021. Information as a public good. UNESCO. Retrieved from https://
en.unesco.org/commemorations/worldpressfreedomday/2021
Vos, T. P, and F. M. Russell. 2019. “Theorizing Journalism’s Institutional Relationships: An
Elaboration of Gatekeeping Theory.” Journalism Studies 20 (16): 2331–2348. ()., (), -.
Vos, T. P. 2015. “Revisiting Gatekeeping Theory during a Time of Transition.” In Gatekeeping in
Transition, 18–38. New York: Routledge.
Yushu’a, M. J. 2010. “Regional Parallelism and the Reporting of Corruption in the Nigerian Press.”
Journal of African Media Studies 2 (3): 353–369.