Content uploaded by Erlis Cela
Author content
All content in this area was uploaded by Erlis Cela on May 24, 2024
Content may be subject to copyright.
Full Terms & Conditions of access and use can be found at
https://www.tandfonline.com/action/journalInformation?journalCode=witp20
Journal of Information Technology & Politics
ISSN: (Print) (Online) Journal homepage: www.tandfonline.com/journals/witp20
The audience logic in election news reporting on
Facebook: what drives audience engagement in
transitional democracies of Albania and Kosovo?
Lindita Camaj, Erlis Çela & Gjylie Rexha
To cite this article: Lindita Camaj, Erlis Çela & Gjylie Rexha (2024) The audience logic in election
news reporting on Facebook: what drives audience engagement in transitional democracies
of Albania and Kosovo?, Journal of Information Technology & Politics, 21:1, 38-53, DOI:
10.1080/19331681.2023.2227847
To link to this article: https://doi.org/10.1080/19331681.2023.2227847
View supplementary material
Published online: 27 Jun 2023.
Submit your article to this journal
Article views: 305
View related articles
View Crossmark data
Citing articles: 1 View citing articles
The audience logic in election news reporting on Facebook: what drives audience
engagement in transitional democracies of Albania and Kosovo?
Lindita Camaj , Erlis Çela , and Gjylie Rexha
ABSTRACT
This study provides insights on how journalists in the Western Balkans conceptualize and practice
audience engagement during electoral campaigns. Taking a holistic approach, we rst explore
audience demand and news supply of strategic and negative election news on Facebook, then turn
to news editors to explore what type of audience logic drives their reporting. Our data conrm
previous ndings about audience demand for strategic news but contradict the predominance of
audience negativity bias in the context of Southeast Europe. These ndings support generic trends
in how social media audiences engage with political information, but also emphasize the impor-
tance of the socio-political context as a determinant of audience engagement with online news.
Interview data identied an alignment between journalists’ imagined readership preferences with
the reality, yet reporting patterns on Facebook do not entirely follow engagement trends.
Together, these ndings suggest that journalism culture developed in this region is more nuanced
than previously dened, while news editors embrace new technologies to serve the commercial
needs and audience strengthening logics in parallel during electoral campaigns.
KEYWORDS
Audience engagement;
strategic frame; horserace;
Facebook; election news;
social media; audience logic
A long-debated question in literature explores if
strategic game framing and negative reporting of
electoral campaigns is detrimental or beneficial to
democracy (Cushion & Thomas, 2018). Issue fram-
ing of political messages is favored because it
increases voter knowledge and participation,
while others associate strategy game frames with
political malaise (Cappella & Jamieson, 1997). Yet,
the predominance of strategic game framing in
electoral coverage has become a standard feature
of election news reporting across countries (Aelst,
2018) partially explained by the audience logic –
journalists’ presumption of public demand for stra-
tegic news (Cushion & Thomas, 2018). In this
study, we explore if audiences in two transitional
democracies of Southeast Europe indeed exhibit
a preference for strategic game frame and negative
election news and investigate what type the audi-
ence logic guides election news production on
social media.
Specifically, this paper analyzes news media use
of Facebook as a platform for campaign news
reporting and user interaction during the 2019
parliamentary elections in Kosovo and 2021 parlia-
mentary elections in Albania. The content analysis
in this article examines three trends in election
coverage on Facebook (Esser & Strömbäck,
2012): 1) media depoliticization, (2) personaliza-
tion, and (3) negativity, and ties it to audience
engagement via comments, shares, and reactions.
Guided by the concept of audience logic (Landerer,
2013), in-depth interviews examine how journalists
conceptualize audience engagement with campaign
news on social media.
This study contributes to the literature in mean-
ingful ways, theoretically and practically. With very
few exceptions (Valenzuela, Piña, & Ramírez,
2017), most research has explored the “audience
turn in journalism” either from the news produc-
tion focus or from the audience’s perspective
(Swart, Groot Kormelink, Costera Meijer, &
Broersma, 2022). This study builds on recent
research that examines the interaction between
news supply and demand on social media. We
combine content analysis and news user data with
in-depth interviews with news producers in order
to explore the prevalence of “news gap” between
what journalists and consumers of political news
deem newsworthy (de León & Vermeer, 2022) in
a non-Western context.
CONTACT Gjylie Rexha gjylije.rexha@ubt-uni.net Media and Communication, UBT - Higher Education Institution, University of Tirana, Lagjia Kalabria,
Prishtina, 10000, Albania
Supplemental data for this article can be accessed online at https://doi.org/10.1080/19331681.2023.2227847
JOURNAL OF INFORMATION TECHNOLOGY & POLITICS
2024, VOL. 21, NO. 1, 38–53
https://doi.org/10.1080/19331681.2023.2227847
© 2023 Taylor & Francis
There is an increasing trend in journalists’ use of
social media for news construction and news dis-
semination, offering new potentials to create more
diverse journalism (Humayun & Ferrucci, 2022).
Overall audience engagement is perceived as
a means to improve audience’s trust in news
media, deliver economic benefits, and strengthen
journalism’s democratic function in society (Singer
et al., 2011). On the audience side, social media
engagement is an indication of the high level of
audience involvement with election news (Oeldorf-
Hirsch & Sundar, 2015) that might lead to different
patterns in voter learning, opinion formation, and
participation. Against this backdrop, the current
study aims to identify ways in which journalists
can engage audiences with quality information
during electoral campaigns that contributes to an
informed and active voter. It is important to under-
stand how digital media transformations in
Southeast Europe have impacted audience concep-
tualization and engagement in journalism, espe-
cially if we consider the important role that the
media have played in the Balkan’s turbulent past
and the more recent democratization develop-
ments (Kovacevic & Perisin, 2018).
Theoretical framework
The democratic value of audience engagement
The democratic relevance of audience engagement
with election information is especially important if
we consider them as acts of civic engagement
(Dalton, 2017). Audience engagement with elec-
tion news on social media can be understood as
a continuum. Facebook affordances facilitate three
main ways users can engage with news: liking,
commenting and sharing, each of which carries
different levels of participation (Larsson, 2015).
Using the like/reaction button in relation to news
posts is considered a light form of engagement,
since it does not take much social capital and effort
for this type of engagement. News commenting, on
the other hand, is consider a superior form of
engagement given that citizen discussions about
politics are perceived as vital for deliberative
democracy and can contribute to learning about
issues and candidate positions during elections
(Carpini, Cook, & Jacobs, 2004). Similarly, news
sharing is a more involved practice of news engage-
ment as it indicates news endorsement and can lead
to high visibility of news posts (Kalsnes & Larsson,
2018). Popularity cues such as sharing and liking
can influence people’s attention to quality news
(Dvir-Gvirsman, 2019) and can influence other
people’s engagement with the same quality news
(Muddiman & Stroud, 2017).
This study is situated within the fast-growing
literature that investigates how social media users
engage with quality news while exploring how spe-
cific news content dimensions might determine
different forms of engagement (Trilling,
Tolochko, & Burscher, 2017). Most of this line of
research is based in Western contexts, which limits
our understanding of election news making and
consuming across media systems, political cultures
and levels of voter engagement (Cushion &
Thomas, 2018). Comparative research emphasizes
that country context is a significant predictor of
news audiences’ social media engagement (Ferrer-
Conill et al., 2021; Salgado & Bobba, 2019) and
journalists adoption of the audience-centric
approach in news production (Andresen, Hoxha,
& Godole, 2017).
This study is situated in the emerging democra-
cies of Albania and Kosovo, in Southeast Europe.
While these countries are similar in their level of
development, language and culture, historical lega-
cies, democratization, market openness, and media
system (Peruško, Vozab, & Čuvalo, 2021), they also
provide a specific context for exploring news
reporting and audience engagement with election
news. The post-socialist transformation of the
media in Central and Eastern Europe (CEE)
resulted in systems that are different from
Western democracies in terms of political paralle-
lism, media markets and journalism culture (Hallin
& Mancini, 2004; Peruško, Čuvalo, & Vozab, 2020).
Media in Southeast Europe countries have remark-
ably low levels of autonomy, while media in
Albania and Kosovo face similar problems related
to ownership concentration, political polarization,
and clientelistic ties (Kovacevic & Perisin, 2018).
These factors are determinant for journalism pro-
fessional roles including media-related practices of
audiences (Peruško, Vozab, & Čuvalo, 2021, p. 1).
More specifically, clientelistic relationships
observed in these countries (Camaj, 2016) are less
JOURNAL OF INFORMATION TECHNOLOGY & POLITICS 39
likely to encourage audience-centric news report-
ing and more likely to focus on reporting that
benefits media clients or owners, such as political
attacks that discredit political opponents.
However, this region has also undergone heavy
media commercialization (Harro-Loit & Saks,
2006) and digitalization that might impact journal-
ists’ attitudes toward their audiences. Facebook has
more than 50% penetration in this region (Internet
World Stats), while social media are the second
most consumed news source in both countries,
with more than half of the population in Albania
(53%) and two-thirds of the population in Kosovo
(65%) using these platforms as news gateways on
a daily basis (SEENPM, 2021). In this context,
journalists working for commercial media can’t
afford to ignore their audiences. In addition, pre-
vious research suggests that journalists in the
Western Balkans consider their role as facilitators
of democratic transitions (Andresen, Hoxha, &
Godole, 2017), and this context might present
new opportunities for digital news media to engage
with their audiences during elections – a critical
time in political transitions.
However, such developments need to be under-
stood within broader concept of “digital medias-
capes” that consider the important implications of
the macro factors for journalism performance
(Peruško, Čuvalo, & Vozab, 2020). Media systems
in CEE are not monolithic as countries within this
region took divergent development, economic and
political trajectories during the formative years
(Peruško, Vozab, & Čuvalo, 2021). The history of
pluralism in Kosovo has developed in different
circumstances from Albania. During the post-civil
war period, Kosovo was under the international
protectorate that dictated strict rules of campaign-
ing and campaign coverage by the media which still
affect current campaigns to this day. In Albania,
the political system is dominated by two major
parties which also have a strong influence on the
media. Therefore, the political and media system in
Albania is more polarized than in Kosovo. Hence,
we are more likely to observe a heavier focus on
political attacks and negativity in news reporting
and audience engagement with this content in
Albania than in Kosovo. Yet, corruption has been
one of the main issues that has dominated both
campaigns (Uberti, 2020); hence, we might expect
heavy engagement with news posts that emphasize
corruptive issues and candidate character. But,
given the disputed political status and the volatility
of political parties in Kosovo, we are more likely to
find an issue-focused electorate.
Audience engagement with election campaign news
Research in political communication has identified
core characteristics of election reporting that
emphasize three coverage dimensions documented
across cultures: : (1) depoliticization, (2) persona-
lization, and (3) negativity (Esser & Strömbäck,
2012). Comparative research suggests a low level
of substantial issue coverage and high emphasis on
strategic game and negative coverage has become
a standard feature of election news reporting across
countries (Aelst, 2018; Cushion & Thomas, 2018),
with a tendency to find such coverage more often
in commercialized media systems (Strömbäck &
Dimitrova, 2006), in competitive party systems
and in countries with high political polarization
(Banducci & Hanretty, 2014).
Media depoliticization. It is conceptualized as
news coverage of elections that marginalizes “the
core of politics – the substance, issues, ideologies,
and linkages between real-world problems and
proposed solutions” (Esser & Strömbäck, 2012,
p. 318). Instead, election coverage predominantly
concentrates on the process of political campaigns
(Cappella & Jamieson, 1997) adopting a strategic
game reporting style. Strategy framing refers to
election reporting that focuses on candidates’
motives and strategies used for winning elections,
and game framing refers to news stories that
emphasize who is winning or losing elections and
often involve a language or war or games (Aalberg,
Strömbäck, & de Vreese, 2012, p. 172). In this
study, we operationalize two aspects of depolariza-
tion: issue coverage and strategy game framing.
Many studies have documented the association
of the strategy game frame with the political
malaise – voter cynicism, decrease substance-
based knowledge and trust in the political process
(see Zoizner, 2021 for an overview). However,
some scholars argue that such coverage of politics
has the potential to engage audiences with election
news which can lead to higher interest in elections
(Iyengar, Norpoth, & Hahn, 2004; Trussler &
40 L. CAMAJ ET AL.
Soroka, 2014). Yet, the evidence on audience’s
appetite for strategic game framed news in online
spaces is inconsistent (Gonçalves, Pereira, & Torres
da Silva, 2020; Kalsnes & Larsson, 2018). Offering
a more nuanced perspective, recent studies suggest
that strategic game framed news gets more expo-
sure and clicks, yet issue-oriented news is more
likely to generate comments and shares (Ørmen,
2019; Stroud & Muddiman, 2019). This literature
suggests that content characteristics are linked to
different types of engagement measures (Salgado &
Bobba, 2019; Stroud & Muddiman, 2019). In other
words, not all forms of news engagement are linked
to all news characteristics. Thus, we pose the fol-
lowing question:
RQ1: How do issue focused and strategic game
framed posts affect user engagement (likes, com-
ments and shares) with election news on media’s
Facebook pages?
Personalization
We are also interested to examine the conse-
quences of personalization of election coverage
for audience engagement. In this study, we follow
Esser and Strömbäck (2012) in differentiating “per-
sonalization” as a distinct category that emphasizes
news coverage focus on individual politicians, and
can include news on candidate personality traits,
their performance, and personal lives at the
expense of parties, ideologies, or policies. The rise
of “candidate-centered politics” (Van Aelst,
Sheafer, & Stanyer, 2012) is especially dominant
on social media, where candidates adopt a more
personalized communication style when discussing
their professional, emotional, and private lives
(Bene, 2017; Metz, Kruikemeier, & Lecheler,
2020). While these studies have explored audience
interactivity with personalized information on can-
didates’ social media pages, very little is known
about how digital news audiences interact with
personalized election information. The only study
in this regard suggests that emphasis on candidates
alone might lead to decreased selection of election
news in digital spaces (Jang & Oh, 2016) but does
not illuminate how this might translate into likes,
comments, and shares. Hence, we pose the follow-
ing research question:
RQ2: How do personalized posts affect user
engagement (likes, comments and shares) with elec-
tion news on media’s Facebook pages?
Negativity
Previous research emphasizes that politics in general
and election news more specifically have a bias
toward negativity, confrontation and conflict (Esser
& Strömbäck, 2012). There are two aspects to nega-
tive news coverage: the media-initiated negativity –
initiated by journalists via news framing and media-
disseminated negativity – triggered by news sources
(Lengauer, Esser, & Berganza, 2012). In this study,
we focus on media-disseminated negativity triggered
by political candidates, to avoid collinearity with
other dimensions of elections reporting given that
some aspects of media-initiated negativity, such as
conflict and drama, are captured in strategic game
framing. Second, during election reporting we are
less likely to encounter some aspects of media-
initiated negativity, such as negative tone toward
politicians and parties, given journalists tendency
to avoid bias accusations.
There is a strong theoretical conceptualization
for audience negativity bias in news selection that
has been linked to evolutionary processes
(Shoemaker, 1996) that predisposes audiences to
be more responsive to negative than positive
messages, cognitively, emotionally, and physi-
cally. The most compelling empirical evidence
comes from a cross-national study from 17 coun-
tries that offers a comprehensive support for the
negativity bias hypothesis, as people were more
aroused by and were more attentive to negative
news (Soroka, Fournier, & Nir, 2019). Yet, other
studies doubt the inherent audience negativity
bias, offering evidence that news that is framed
as conflict is shared less on social media
(Valenzuela, Piña, & Ramírez, 2017), while news
coverage of politicians behaving with incivility
discouraged interactions with online news
(Muddiman, Pond-Cobb, & Matson, 2020).
Thus, we pose the following research question:
RQ3: How do negative posts affect user engage-
ment (likes, comments, shares) with election news
on media’s Facebook pages?
JOURNAL OF INFORMATION TECHNOLOGY & POLITICS 41
The audience logic in election news coverage
In this study, we are also interested to explore
the extent to which audience logic drives election
news reporting in the context of Southeast
Europe. The predominance of the strategic
game frame and negativity found in electoral
coverage across countries has been largely
explained by the media logic – professional
orientations and news values that prioritize
drama, negativity, and personal focus (Aalberg,
Strömbäck, & de Vreese, 2012), the commercial
objectives of news media, and the political logic
(Cushion & Thomas, 2018; Strömbäck & Esser,
2014). The role of audiences has been largely
ignored, although recent research emphasizes
the need for the shift toward the “audience
logic” in digital journalism at large (Blassnig &
Esser, 2021) and election reporting specifically
(Cushion & Thomas, 2018).
Literature distinguishes three different sub-
logics that might explain the audience turn in digi-
tal journalism (see table 1). The “Audience logic” in
election reporting style has been traditionally
defined within the economically inspired theoreti-
cal model (Landerer, 2013), namely the market
sub-logic. Within this approach, the news making
process is driven by the commercial goals, and the
audiences are perceived as consumers. Thus, news
media are more likely to focus on what attracts the
largest audiences, in order to maximize the distri-
bution of their product (Esser, 2013). The market/
commercial sub-logic, in turn, determines certain
characteristics of news content, with emphasis on
confrontation and dramatization aspects of news
associated with the horserace frame, depolariza-
tion, personalization (Esser, 2013), conflict and
negativity (Karidi, 2018). Strategic game frame of
electoral coverage has been traditionally explained
by journalists’ assumption that there is a consumer
demand for it, given that the conflict and the drama
associated with horserace frame attracts audiences’
attention making election news more exciting
(Cushion & Thomas, 2018).
Yet, another perspective defines the audience
logic within the normative sub-logic (Landerer,
2013) that prioritize social or public responsibility
over market considerations. This “public logic,” as
Landerer (2013, p. 245) refers to it, aims to advance
the public interest agenda by emphasizing issues of
public interest. News reporting is driven by what
citizens should know in order to make informed
decisions in a democracy. While it is hard to define
and empirically capture what is in public’s interest,
in the context of election coverage, journalists
emphasize the importance and concern with sub-
stance and policy issues. As Cushion and Thomas
(2018) claim “an lection news agenda – or logic –
driven by issues could also better represent the
public’s voice and policy preferences during
a campaign” (p.102). A major difference between
the two logics is that “normative logic is supply-
driven, market logic is demand-driven” (Landerer,
2013, p. 249).
Yet, a third perspective, the network media logic
(Klinger & Svensson, 2015), takes into considera-
tion the technical affordances of digital technolo-
gies and communication platforms that have been
integrated into digital journalism as determinants
of audience involvement in the news process. Social
media enable interactive and personalized commu-
nication, as users willingly share news for free set-
ting the news agenda for their own networks. In
this media ecology, peer recommendation system
and endorsements are more important factors than
a selection from professional newsmakers.
Consequently, the revolution in audience analytics
provides the ability to quantify audience behaviors
on social media and identify what topics drive
more audience attention (Lee & Tandoc, 2017).
Within the networked media logic, journalists
have to consider and balance the shareworthiness
of the stories along with their newsworthiness
(Trilling, Tolochko, & Burscher, 2017). As
Table 1. Three sub logics of audience logic in election news production.
Audience Logic
Market sub-logic Normative sub-logic Networked Media sub-logic
Audience as consumer Audience as citizens Empowered audiences
Attract largest audience Advance public interest Balance shareworthiness with newsworthiness
Focus on confrontation and dramatization Focus on issues Equal distribution of strategic game and issue frame
42 L. CAMAJ ET AL.
Blassing & Esser (2022, p.51) emphasize, “while the
market-oriented commercial logic remains
a constant force on political news coverage,
a connection-strengthening audience logic has
developed in parallel” (see Table 1).
The attitude that higher engagement equates
higher ratings and thus potentially leads to lower
quality of news has created tensions between jour-
nalists and management related to catering inter-
esting news at the expense of important news
(Belair-Gagnon & Holton, 2018). Yet, proponents
of engaged journalism challenge the traditional
journalistic notion that journalists know better
than audiences themselves what information they
need (Lawrence, Radcliffe, & Schmidt, 2018).
Although the new technological affordances pro-
vide grounds for a theorized audience empower-
ment in the new media ecosystem, it is not clear
how journalists’ perception of their audience has
changed, if at all. In this paper we explore to what
degree have news media in Southeast Europe
incorporated the audience logic and how this
logic is shaping their coverage of elections. We
are guided by the following questions:
RQ4a: How do journalists in Albania and Kosovo
perceive consumer demand for strategy game
frames, personalized and negative news?
RQ4b: How does the audience logic impact news
reporting of elections on Facebook?
Methods
In this study, we focus on 12 major news organiza-
tions in Albania and Kosovo. Our list of most
popular news outlets in digital spaces comes from
data provided by Amazon’s Alexa. Alexa’s top web-
site rankings are based on estimated daily time on
site per visitor, estimated daily unique pageview
per visitor, the percentage of traffic from search,
and the total number of sites linked to the specific
site. The list was cross-checked with data from local
providers on audience reach of individual news
outlets. Our sample of news organizations includes
a mix of mainstream (national news broadcasters,
and daily newspapers) and digital native news
outlets, and represents an important part of
Albanian news media outlets on the Internet (the
list and profile of news organizations is provided in
Table 1 in Appendix 1).
Content analysis
Sample
The content data for this study were collected using
the CrowdTangle software, an open-source web
tool that compiles engagement data for specified
Facebook accounts by accessing the Facebook API.
First, we collected all Facebook posts from 12
major news organizations posted during the last
10 days of the electoral campaign in respective
countries. In Kosovo, Facebook posts were col-
lected from September 25
th
to October 4
th
(election day was on October 6
th
, 2019), while in
Albania, Facebook posts were collected from
April 12 to 22 (election day was on April 25,
2020). From this sample, we selected only
Facebook posts that specifically focused on election
campaign coverage excluding the duplicates. The
final media sample includes 6, 749 election-related
news posts from news organizations’ Facebook
pages: 6 outlets from Albania (Top Channel =
619; TV Klan = 393; ABC News = 507; Gazeta
Panorama = 675; Balkan Web = 366; and Syri =
1024) and 6 news outlets form Kosovo (Gazeta
Express = 449; Telegrafi = 209; Koha.com = 464;
Klan Kosova = 1050; Gazeta Blic = 764, and RTK
= 229).
Coding categories
Based on previous work, we developed a codebook
which was used to manually code Facebook posts.
The coding was done by one of the authors and
a graduate student who were extensively trained to
use the codebook with four different training ses-
sions using different subsamples of n = 30 posts for
each. After establishing satisfying intercoder relia-
bility on a random sample of 10% of the Facebook
posts (Cohen’s Kappa ranged from .71 to 1.0.), they
each proceeded to code independently.
We created two different binary categories to
capture the concept of “depoliticization”: we
coded if Facebook posts mentioned issues (Kappa
= .90), and if they discussed strategic game aspect
of the election campaign (for example: strategies
JOURNAL OF INFORMATION TECHNOLOGY & POLITICS 43
used to win voters over, who is ahead in the race,
etc.; Kappa = .75). Strategy versus issue coverage
was coded as separate dimensions in a way that
the same article could be coded as both.
To capture “personalization” in news coverage,
we coded if news posts discussed candidate char-
acter traits (for example: leadership, knowledge,
morality, ect., Kappa = .78). Negativity in
Facebook news posts was operationalized with
a measure that captures candidate attacks as fol-
lows: we identified posts that quoted or para-
phrased one of the candidates, and then we coded
for the function of the candidate message: acclaims
(discuss their policy achievements, plans, or their
personal character), defense (defend their actions),
or attacks (attack their political opponents’ perfor-
mance, policy or character) (Kappa =.73). This
measurement was transformed into
a dichotomous variable denoting candidate attacks.
In addition, this study coded for the following
control variables that have been identified as
important determinants of social media news
engagement: type of news genre (news, analysis,
editorial, interview, or sponsored programing)
(Kappa =.80); the name of candidates (Kappa = 1)
and the name of the party mentioned (Kappa = 1).
In addition, we adopted the coding scheme from
the Comparative Agendas Project (Baumgartner,
Green-Pedersen, & Jones, 2006) in order to code
for 10 most mentioned topics in Facebook posts
(Kappa =.73). To account for the different inter-
cepts between countries, we inserted a dummy
variable with Kosovo serving as the reference
category.
Audience engagement data included three cate-
gories: reactions (inclusive of all kinds: like, love,
wow, haha, sad, and angry), comments, and shares.
These data were collected by CrowdTangle and
calculated by simply counting how many of each
a post has. Engagements are measured at the per-
post level, which is the primary unit of analysis.
In-depth Interviews
For the qualitative part of this study, we conducted
semi-structured in-depth interviews with social
media editors from the media sample (N = 10).
We identified journalists who had strategic and
operational responsibility to manage social media
accounts of each news outlet. Out of twelve editors
contacted, two did not respond to our request for
the interview. Thus, our final sample includes
social media editors from five news media in
Albania (Top Channel, ABC News, Syri TV,
Panorama, and BalkanWeb) and five news media
in Kosovo (RTK, Koha, Klan Kosova, Express,
Telegrafi). Interviews, conducted in person or
over the Zoom, took place within a couple of
months from election day. The interviews, which
lasted between 45 minutes and an hour, were con-
ducted in Albanian language by the authors, all
native speakers, and included questions about
motivations and strategies news organizations
used to engage their audience during elections
(see interview guide in appendix 2).
Data analysis and results
User engagement with election news on facebook
Tables 2 and 3 report descriptive analysis at the
country level in terms of patterns of media cover-
age of election news on Facebook and audience
engagement with election news. The preliminary
examination of the data suggests that, across both
countries, media coverage had a similar focus
between strategic game frame and issue coverage,
with a slightly higher emphasis on strategic game
framing (58.1%) than issue coverage (42.9%).
Much less focus was given to candidate traits or
personality (9.1%), while less than 15% of the
whole sample focused on candidate attacks.
However, significant differences were observed
between the two countries. Kosovo news media
Table 2. The share of election news coverage.
Albania Kosovo Total Chi Square
Issue coverage 36.9% 49.2% 42.9% 113.809***
Strategic game frame 52.1% 64.2% 58.1% 113.769***
Candidate Traits 7.6% 10.8% 9.1% 21.810***
Negativity 18.1% 5.6% 12.2% 247.456***
N 3165 3584 6749
Percentages do not round up to 100, given that social media posts could be coded for multiple frames.
44 L. CAMAJ ET AL.
were more likely to emphasize issues and focus on
strategic game framing and candidate personalities,
while media in Albania were significantly more
likely to focus on candidate attacks (see Table 2).
When comparing audience engagement with elec-
tion news across two countries, data on Table 3
suggest that overall Facebook posts of Albanian
news media received higher audience reactions
and comments, yet Kosovo media audiences gen-
erated more than twice the number of audience
shares.
The first three research questions (RQ1, RQ2,
and RQ3) asked how different patterns of election
news coverage affect audiences’ engagement on
Facebook in the context of Albania and Kosovo
elections. Given the hierarchical structure of the
data – where Facebook posts are nested within
news organizations – we estimate multilevel mod-
els that allow the intercepts to vary across news
organizations (Gelman & Hill, 2006). Further, we
use a negative binomial regression model to esti-
mate our predicted relationships. The standard
deviations of our dependent variables are higher
than the means, indicating negative binomial
regressions as the most appropriate approach
(Gardner, Mulvey, & Shaw, 1995). Table 4 presents
a mixed effects negative binomial regression model
where the dependent variables are the rate of
Facebook reactions, comments, and shares.
The results suggest that overall online news
audiences were less likely to engage with
Facebook posts that focus on issues compared to
the posts that do not focus on issues. Issue coverage
Table 3. Descriptive Statistics of Facebook Engagement.
Reactions Comments Shares
Albania
N 3,584 3,584 3,584
Mean 91.08 27.11 2.05
Std Dev. 342.469 105.483 11.666
Min 0 0 0
Max 8,918 4,749 293
Kosovo
N 3165 3165 3165
Mean 68.36 56.01 3.52
Std Dev. 171.977 139.545 18.523
Min 0 0 0
Max 3,214 2,500 734
Table 4. Multilevel Negative binomial regression with random intercepts predicting user engagement on Facebook.
Reactions Comments Shares
b SE b SE b SE
Election Coverage
Issue coverage −.385*** .072 −.507*** .090 −.375** .127
Strategy game .126** .044 −.112* .055 .121 .077
Candidate Traits .246*** .062 .283*** .078 .007 .110
Negativity −.163** .057 −.128** .072 −.242** .103
Control Variables
Story type
(reference: news story)
Analysis −.239*** .066 .351*** .082 .142 .116
Editorial .098 .452 .090 .569 .049 .773
Interviews −.074 .047 .081 .059 .109 .083
Sponsored programing 1.425*** .119 1.477*** .148 1.535*** .202
Story Topics
(reference government operations)
Defense .267 .182 −.137 .228 −.371 .323
Law & Crime −.026 .092 −.009 .115 −.295 .164
International Affairs −.179 .096 −.439*** .120 −.729*** .171
Economy −.144 .115 −.085 .143 −.261 .206
Civil Rights −.694*** .168 −1.077** .211 −.859** .310
Social Wealth fare 1.188*** .273 1.123*** .340 .666 .464
Health .394** .156 −.105 .195 −.455 .285
Education −.673*** .132 −.868*** .165 −1.103*** .243
Immigration .647** .246 .562 .307 .740 .418
Campaign dates .041*** .006 .030*** .008 .076*** .011
Albania (reference Kosovo) −22.346*** 2.77 −16.962*** 4.69 −43.445*** 6.64
Constant 2.013 3.135 5.523
Random Effect variance .502* .253 .473* .240
Log-likelihood 40402.853 34156.387 46358.034
Dispersion parameter
N (FB posts) 6,747 6,748 6,748
Cell entries are mixed effects coefficients where Facebook posts are level 1 and news organizations are level 2. If no reference category is mentioned, variables
were coded as present or not present. Additional control variables include 1) names of specific candidates mentioned in the Facebook post and 2) names of
specific parties mentioned in the Facebook post.
***p < .001. **p < .01. *p < .05.
JOURNAL OF INFORMATION TECHNOLOGY & POLITICS 45
was associated with a decrease in reactions
(b=−.385, p < .001), comments (b=−.507, p
< .001), and shares (b=−.375, p < .01). Yet, some
issues were more likely to generate engagement
than others. For example, Facebook posts that
focused on social welfare, health, and immigration
were significantly more likely to generate news user
engagement than Facebook posts that focused on
government operations.
Facebook posts that focused on strategic
game frame generated a more nuanced engage-
ment reflection. Posts that emphasized candi-
date strategies to win elections or the game
aspect of the election race were associated with
an increased number of user reactions (b = .246,
p < .001), yet they were also associated with
a decrease in the number of comments per
post (b=−.112, p < .05).
Addressing RQ2, our data suggest that posts that
emphasized candidate traits were associated with
an increase in the number of reactions (b = .246,
p < .001) and comments (b = .283, p < .001).
The third research question (RQ3) asked how
audiences in Albania and Kosovo interact with
negative election news. The results suggest that
posts that emphasized candidate attacks had
a consistent negative association with all three
modes of engagement, as they decreased the num-
ber of reactions (b=−.163, p < .01), comments (b=
−.128, p < .01), and shares (b=−.242, p < .01).
Analyzing the set of control variables, two
relevant findings appeared: first, Kosovo media
audiences were significantly more engaged with
election news on Facebook than Albanian media
audiences across all three measures of engage-
ment. Second, audience engagement was signifi-
cantly correlated with news story type. Compared
to Facebook posts that focused on news stories,
posts that provide news analysis were less likely
to generate reactions and comments. However,
Facebook posts that contained sponsored pro-
graming from live electoral events were more
likely to generate higher number of reactions,
comments, and shares than Facebook posts that
reported campaign events. This last finding is not
surprising, given that Facebook algorithms prior-
itize live video based on audiences’ likelihood to
interact more with video content (Salgado &
Bobba, 2019).
The audience logic and election coverage on
Facebook
The in-depth interviews conducted for this study
focused on three main topics: (1) journalists’ con-
ceptualization of their social media audiences; (2)
the decision-making process when posting election
news on Facebook, and (3) the motivations for
pursuing audience engagement. The analytical pro-
cess was inductive, following a multilevel analytical
coding scheme that allows for key themes and
subthemes to emerge from the data (Corbin &
Strauss, 2014). The data were synthesized into the
following categories: market sub-logic approach,
normative sub-logic, and networked media sub-
logic approach. The quotes from the interviews
incorporated in this analysis have been translated
from Albanian language by the authors.
Market sub-logic
Interview-based data suggest that media in Albania
and Kosovo perceive social media audiences pre-
dominantly as consumers of information. The pre-
vailing attitude among social media editors
interviewed for this study is that Facebook enables
media to promote their brand and product with the
intention to generate web traffic from Facebook
audiences’ clicks. From the editors’ point of view,
Facebook has the function of a virtual marketplace
where news outlets compete for the attention of the
audience, as summarized in the following quote:
Facebook is the market, while we are market players to
sell our products. In the big Facebook market, we try to
encourage people to buy our products, in this case with
the news (a male web editor from Kosovo).
In line with the audience-focused approach, editors
acknowledge that they are oriented by audience
interaction data on social media to understand
the trend of news content consumption. Audience
engagement data on Facebook is analyzed periodi-
cally. One of the interviewees admits that he
receives monthly reports from the information
technology department, which contain recommen-
dations for news categories that need to be stimu-
lated. High audience engagement with a specific
story on Facebook is perceived as a measure that
the story will provoke more interest, as summar-
ized in the following quote by an editor at a news
46 L. CAMAJ ET AL.
outlet in Albania: “If a post provokes negative reac-
tions, we think twice whether this topic should be
addressed further or not. If we see that the audience
is interested in it, we follow the story.”
In both countries, election news generated an
increased audience interest and engagement on
Facebook. The majority of the editors concur with
the perception that social media audiences were
mostly interested in news that focuses on polls, or
the race aspect of elections, who’s wining and los-
ing, and they mostly engaged with news that
focuses on politicians’ personalities and drama,
such as accusations of corruption or similar scan-
dals. Therefore, many editors acknowledge their
election reporting on Facebook is determined to
a great degree by these trends. Here’s how a social
media editor from a television network in Albania
explains their election reporting strategy:
“At the beginning we prepared our readers by continu-
ally posting about the most contested races by county
and then we built a poll for our social media sites. For
example, the poll would ask ‘Who will win Elbasan?’and
we would list all the leading candidates . . . our format
was such that we placed the photo of the candidate from
PS (Socialist Party) and the photo from PD (Democratic
Party), with their names above their pictures and asked
‘Who will win Elbasan?’ or ‘Who will win Tirana?’ We
adopted this strategy after the candidate lists were
released. This strategy helped us increase audience
interactivity on social media.”
When discussing election stories of interest, several
Kosovo editors emphasized the case of Ramush
Haradinaj, a political candidate who would sing
folk songs or give bombastic statements during
media interviews as a spectacle that generated
much media coverage and audience reactions on
Facebook. This is how a male social media editor
from Kosovo summarized: “These spectacles had
immense engagements. For example, a statement
from Ramush Haradinaj saying “I’m not as stupid
as I look” generated many laughs and its’ effect was
ten to fifteen times higher than a random news story
on Facebook.”
Normative sub-logic. Although the “audience as
consumers” approach prevailed, the analysis of
the interviews also identified a group of editors
that tend to adopt the normative perspective on
audience conceptualization. While not denying the
commercial aspect of social media, for some editors
in Albania and Kosovo informing and educating
audiences takes on a missionary nature. Below is
how a social media editor summarizes their
approach:
We believe that the audience can be cultivated, edu-
cated. If we give the audience poor material, they will
grow with that material. I think information is ulti-
mately similar to food. If we give people, the right
food they will grow up healthy, (a female editor at
a television station in Albanian).
For this group of editors, the quality of news is
primary, while they claim to avoiding falling prey of
social media trends. As one editor from Kosovo puts
it “we don’t measure (success) by likes, but we measure
it by audience trust and loyalty.” Several editors
emphasized two criteria that drove their election
reporting: fairness and balance, and public interest
manifested in political programs. The example below
from a social media editor working for an online
publication in Kosovo summarizes this approach:
We set several parameters. We decided that if we post
a story about a political party, we should post about
other political parties as well, and not post two or three
posts about the same party, so we show that we do not
favor only one party. The second criteria was to deter-
mine if that information was relevant for the whole of
Kosovo or only for one city or village . . . it’s very
important to post if a politician discusses health insur-
ance for all Kosovo, instead of posting about what
a candidate said in a village about their election victory.
Yet, editors acknowledged tensions that emerge
between their goal to focus on public interest and
the necessity to follow social media trends when
deciding what to post. Taking a paternalistic
approach, they don’t see much value in social
media interactions and acknowledged Facebook
posts that focused on party programs and policies
received very little audience interactions. The pre-
vailing view among this group of editors is that
Facebook audiences tend to negatively influence
the quality of news, and they do not necessarily
reflect their whole audience. A few mentioned the
high volume of engagement from partisan “trolls”
in Albania, who were artificially trying to influence
users’ perception about support for their candi-
dates. As the editor of one television news station
in Albania claims, the aggressive comments of the
JOURNAL OF INFORMATION TECHNOLOGY & POLITICS 47
audience on social media “do not really show the
reality of what people think.”
Network media sub-logic
Another smaller group of editors is mostly con-
cerned with the affordances of social media that,
in addition to content distribution and the finan-
cial prospect, allow also for two-way communica-
tion with their audiences. Here’s how a social
media editor for a multimedia news organization
in Kosovo elaborates their approach:
Currently, not only here but in the whole world, social
media facilitate information distribution. This is one of
the main reasons. The other reason is communication
with our readers . . . . and of course we cannot sidetrack
the financial aspect of this. But, as I said earlier, our
readers are loyal and we don’t have much oscillation in
our readership, one day 10K and another 1K, this
doesn’t happen to us because they (audiences) are con-
stant, (social media editor from Kosovo)
They conceptualize their audiences’ interests as
a mix of demand for quality as well as light news.
At the same time, they acknowledge the need for
balancing what audiences want and what they
need, as emphasized in the following quote by
a female social media editor from a television sta-
tion in Albania: “There is a strong relationship
between those that give and those that receive the
information. I strive to orient them, but the audience
also orients me.”
Several editors emphasized that they pay atten-
tion to the comment section to gage public inter-
est in issues they post on their Facebook pages.
During election, there was a heightened interest
from the audience to communicate with news
media in different forms, sometimes commenting
on news stories, other times sending information
about political parties’ activities in their own
regions. Comments are perceived as an opportu-
nity for the audience to engage in the networked
public sphere where they can debate and express
opinions about governance dissatisfaction or
political preferences and some editors perceive
comments as contributing to the quality of
news. Some editors emphasized that their audi-
ences are educated hence they demand quality
news related to party platforms and programs.
Yet, others also warned against perils of social
media algorithms that favor content with certain
type of engagement. Here’s how a social media
editor from Albania explains
There is the principle that a news post that has many
comments is more likely to also be shared on Facebook.
Facebook favors news posts with high interactivity, and
this was used as a public relations strategy by political
parties. They created press offices and call centers that
dealt only with this issue in order to increase the share
of certain news posts (male editor from an online media
in Albania).
This group of social media editors follows closely
the changes to the Facebook algorithms in order to
adapt as quickly without risking losing their audi-
ence. Consequently, editors try to find a balance
between share-worthiness and news-worthiness for
the news they post on Facebook. In such
a situation, they need to make news selection
according to the platforms and the profile of the
audience that engages in these platforms. Overall,
most editors interviewed for this study agreed that
they are more likely to publish on Facebook news
that generates more engagement, while stories they
perceive as important and of interest to the public
are published on the website of these media and
Facebook serves as a gateway.
Discussions and conclusions
This study provides insights on how journalists in
the Western Balkans conceptualize and practice
audience engagement during electoral campaigns.
Taking a holistic approach, we first explore audi-
ence demand and news supply of strategic and
negative election news on Facebook, then turn to
news editors to explore what type of audience logic
drives their reporting.
One important takeaway lesson from this study
is that strategic game framing does not erode poli-
tical participation as previously feared. Similar to
the Western context, our data confirm Albanian
audiences’ interest in strategic news (Gonçalves,
Pereira, & Torres da Silva, 2020; Kalsnes &
Larsson, 2018), as we found that strategic game
framing was associated with higher levels of
engagement with election news on Facebook,
which might translate in increasing interests and
participation in elections (Iyengar, Norpoth, &
Hahn, 2004; Trussler & Soroka, 2014). Yet, our
data also confirm previous studies that election
48 L. CAMAJ ET AL.
news characteristics are not linked to all three
forms of engagement in the same way (Salgado &
Bobba, 2019). Strategic game framing of election
news was associated with higher number of user
reactions, yet it was likely to decrease user com-
ments. These findings point out to two explana-
tions. First, they relate to general patterns in digital
news engagement that can be explained by audi-
ences’ predispositions. Supporting the “social
media gap” hypothesis (Bright, 2016), our data
suggest that in private audience might be clicking,
reading and liking strategic game news, yet they are
less likely to share or comment on it publicly given
that this type of behavior has a higher social cur-
rency (Kalsnes & Larsson, 2018).
Second, the socio-political context is another
crucial factor that determines engagement with
online news. Political parties in both countries
have mobilized their voter base on social media
with the aim to create an opinion climate that
favors them as frontrunners, potentially increasing
the number of Facebook engagements with strate-
gic game reporting. Moreover, we found that
Facebook posts that emphasized candidate charac-
teristics were more likely to generate higher user
reactions and comments. These results suggest that
news personalization has the potential to engage
voters in news discussion that can contribute to
voter learning and participation (Carpini, Cook,
& Jacobs, 2004). Given that corruption has been
one of the major problems that both Albania and
Kosovo have faced in their transitional path
(Uberti, 2020), it is not surprising to find that
news audiences are interested to discuss candidate
characteristics on social media. However, voter
mobilization was more pronounced in Kosovo,
where the challenger party Vetvendosje had
a younger social media savvy base of supporters.
This might explain our results, suggesting that
Facebook audiences in Kosovo reacted, commen-
ted, and shared election news more than users in
Albania.
In addition, our data contradict theoretical pre-
dictions that emphasize the predominance of nega-
tivity bias in the context of Southeast Europe. We
found that news coverage that focuses on candidate
attacks was limited across both countries, although
we found it to be more prevalent in Albania than in
Kosovo. Rather than being an editorial decision, it
could be that there were more candidate attacks in
Albania due to the higher levels of political polar-
ization than in Kosovo. Hence, these cross-country
differences suggest that the extent of audience
engagement with negative campaigning might
depend on systemic factors, given that political
polarization and lower media independence are
important factors that determine news consump-
tions (Tóth, Mihelj, Štětka, & Kondor, 2022).
However, at the same time editorial decisions on
electoral reporting might be determined by audi-
ence analytics. Our data suggest that news coverage
of politicians’ attacks might discourage user inter-
actions with online news. Both countries, Albania
and Kosovo, are undergoing political transitions in
addition to facing economic hardship, and in the
case of Kosovo, disputed territorial integrity.
Hence, it is not surprising that voters might be
more drawn to engage with politicians who present
solutions to the problems that face their nations,
instead of engaging with politicians who accuse
their opponents of failure (Muddiman, Pond-
Cobb, & Matson, 2020). This, in turn, might have
important implications for editorial decisions.
Indeed, we identified an alignment between
journalists’ imagined readership preferences with
the reality (Robinson, 2019). In line with audience
behavior, social media editors assume that their
audiences prefer news that focuses on strategy
game frames and politicians’ performance and
morality. Supporting previous research (Bullard,
2015), this study found that most news media in
Albania and Kosovo are guided by the market sub-
logic in their use of social media to build their
brand name and use Facebook as a gateway to
their websites. Given the popularity of strategic
game election news on social media (Ørmen,
2019), it is not surprising that news producers pay
attention to audiences’ social media engagement.
These findings are not surprising when we consider
high levels of media commercialization (Harro-
Loit & Saks, 2006) and instrumentalization
(Örnebring, 2012) in this region where news
media are often used as tools for economic benefit
of their owners rather than subscribe to audience
focused outcomes.
Yet, the market logic does not entirely drive their
news coverage of elections. Our interviews indicate
that a smaller group of social media editors adopt
JOURNAL OF INFORMATION TECHNOLOGY & POLITICS 49
the normative sub-logic taking a paternalistic role
in educating their social media audiences while
disregarding the value of audience engagement
data. Hence, it is not surprising that a good amount
of election news in both countries is issue focused,
although such election reporting doesn’t necessa-
rily drive social media engagement. An even smal-
ler minority of social media editors interviewed for
this study are concerned with the affordances of
social media that go beyond commercial nature
and open opportunities for two-way communica-
tions with their audiences. They are not hesitant to
promote content they deem valuable on Facebook,
however also do not shy away from following audi-
ences’ preferences. Together these data show edi-
tors’ need to comply with user preferences and
Facebook’s News Feed algorithm, while also adher-
ing to traditional journalistic standards. Together,
these findings suggest that the journalism culture
developed in this region is more nuanced than
previously defined (Örnebring, 2012; Roudakova,
2008), while news editors embrace new technolo-
gies to serve the commercial and audience
strengthening logics in parallel (Blassing & Esser,
2021).
Like other research done in similar context, this
study is limited in its scope and operationalization.
These study results are limited to Facebook, which,
while it is one of the most important platforms in
the region studied here, might not be as widely
used in other contexts. This study is also limited
to examining audience behavior in the form of
reactions, comments, and shares; nevertheless, as
previous research has shown, news selection mea-
sured via clicks constitutes another important
dimension of news engagement that is more pri-
vate and thus might be driven by different report-
ing patterns. Finally, it is possible that, as some
interviewed editors mentioned, some of the
engagement with election news is not organic but
rather driven by political actors and organizations
who engage their partisans to generate engagement
to enhance positive attributes of their preferred
candidates. This is a valid concern that needs
further examination.
Despite these limitations, this study helps
understand the nature of consumer demand for
election news as a central facet to understanding
the nature of media content production. It shows
that election reporting could be more sensitive to
the democratic needs of citizens without repla-
cing strategic game coverage with policy discus-
sions. As Cushion and Thomas (2018, p. 102)
argue, it can involve “slicing together issues and
horserace reporting in ways that serve a public
logic by exciting and informing the audience.”
Indeed, this study shows the reciprocity of the
relationship between demand and supply in the
context of election reporting on social media
(Trussler & Soroka, 2014). Journalists can use
these tools to identify consumer demand for
news, but also engage the same tools to lead
a shift in consumer behavior toward quality news.
Disclosure statement
No potential conflict of interest was reported by the author(s).
Notes on contributors
Lindita Camaj (Ph.D., Indiana University) is an Associate
Professor at the Jack J. Valenti School of Communication,
University of Houston. Her research addresses the role of
news media in political processes, with a focus on digital
communication, media effects, access to information and
journalism in South East Europe.
Erlis Çela (Ph.D.) is an Assistant Professor at the Department
of Communication Sciences, Beder University College in
Tirana, Albania. His research interest includes the role of
social media in journalism, disinformation, and media
literacy.
Gjylie Rexha (Ph.D., University of Tirana) is an Assistant
Professor at the Faculty of Media and Communication, UBT
Higher Education Institution, in Kosovo. Her research
addresses the role of journalists and media in political com-
munication, especially focused on radio and television in
Kosovo.
ORCID
Lindita Camaj http://orcid.org/0000-0002-5934-6159
Erlis Çela http://orcid.org/0000-0002-9202-2981
References
Aalberg, T., Strömbäck, J., & de Vreese, C. H. (2012). The
framing of politics as strategy and game: A review of con-
cepts, operationalizations and key findings. Journalism, 13
(2), 162–178. doi:10.1177/1464884911427799
50 L. CAMAJ ET AL.
Aelst, P. V. (2018). The whole world is watching: Comparing
European and United States News Coverage of the U.S.
2008 and 2016 Elections. 23.
Andresen, K., Hoxha, A., & Godole, J. (2017). New roles for
media in the Western Balkans: A study of transitional
journalism. Journalism Studies, 18(5), 614–628. doi:10.
1080/1461670X.2016.1268928
Banducci, S., & Hanretty, C. (2014). Comparative determi-
nants of horse-race coverage. European Political Science
Review, 6(4), 621–640. doi:10.1017/S1755773913000271
Baumgartner, F. R., Green-Pedersen, C., & Jones, B. D. (2006).
Comparative studies of policy agendas. Journal of European
Public Policy, 13(7), 959–974. doi:10.1080/
13501760600923805
Belair-Gagnon, V., & Holton, A. E. (2018). Boundary work,
interloper media, and analytics in newsrooms: An analysis
of the roles of web analytics companies in news production.
Digital Journalism, 6(4), 492–508. doi:10.1080/21670811.
2018.1445001
Bene, M. (2017). Go viral on the Facebook! Interactions
between candidates and followers on Facebook during the
Hungarian general election campaign of 2014. Information,
Communication & Society, 20(4), 513–529. doi:10.1080/
1369118X.2016.1198411
Blassnig, S., & Esser, F. (2021). The “Audience Logic” in
digital journalism: An exploration of shifting news logics
across media types and time. Journalism Studies, 23(1),
1–22. doi:10.1080/1461670X.2021.2000339
Blassnig, S., & Esser, F. (2022). The “Audience Logic” in
Digital Journalism: An Exploration of Shifting News
Logics Across Media Types and Time. Journalism Studies,
23(1), 48–69. doi:10.1080/1461670X.2021.2000339
Bright, J. (2016). The social news gap: How news reading and
news sharing diverge. Journal of Communication, 66(3),
343–365. doi:10.1111/jcom.12232
Bullard, S. B. (2015). Editors Use Social Media Mostly to Post
Story Links. Newspaper Research Journal, 36(2), 170–183.
doi:10.1177/0739532915587288
Camaj, L. (2016). Between a rock and a hard place:
Consequences of media clientelism for journalist–politician
power relationships in the Western Balkans. Global Media
and Communication, 12(3), 229–246.3. doi:10.1177/
1742766516675649
Cappella, J. N., & Jamieson, K. H. (1997). Spiral of Cynicism:
The press and the public good. New York: Oxford University
Press.
Carpini, M. X. D., Cook, F. L., & Jacobs, L. R. (2004). Public
deliberation, discursive participation, and citizen engage-
ment: A Review of the empirical literature. Annual Review
of Political Science, 7(1), 315–344. doi:10.1146/annurev.
polisci.7.121003.091630
Corbin, J., & Strauss, A. (2014). Basics of Qualitative Research:
Techniques and Procedures for Developing Grounded
Theory. Thousand Oaks, CA: SAGE Publications.
Cushion, S., & Thomas, R. (2018). Reporting Elections:
Rethinking the logic of campaign coverage. Medfort, MA:
Polity Press.
Dalton, R. J. (2017). The participation gap: Social status and
political inequality. Oxford University Press.
de León, E., & Vermeer, S. (2022). The news sharing gap:
Divergence in online political news publication and disse-
mination patterns across elections and countries. Digital
Journalism, 11(2), 343–362. doi:10.1080/21670811.2022.
2099920
Dvir-Gvirsman, S. (2019). I like what I see: Studying the
influence of popularity cues on attention allocation and
news selection. Information, Communication & Society, 22
(2), 286–305. doi:10.1080/1369118X.2017.1379550
Esser, F. (2013). Mediatization as a Challenge: Media logic
versus political logic. In H. Kriesi, S. Lavanex, F. Esser,
J. Matthes, M. Bühlmann, & D. Bochsler (Eds.),
Democracy in the age of globalization and mediatization
(pp. 155–176). Basingstoke: Palgrave Macmillan. doi:10.
1057/9781137299871_7.
Esser, F., & Strömbäck, J. (2012). Comparing news on national
elections. In F. Esser, & T. Hanitzsch (Eds.), Handbook of
comparative communication research (pp. 308–326).
London: Routledge.
Ferrer-Conill, R., Karlsson, M., Haim, M., Kammer, A.,
Elgesem, D., & Sjøvaag, H. (2021). Toward ‘Cultures of
Engagement’? An exploratory comparison of engagement
patterns on Facebook news posts. New Media & Society, 25
(1), 95–118. doi:10.1177/14614448211009246
Gardner, W., Mulvey, E. P., & Shaw, E. C. (1995). Regression
analyses of counts and rates: Poisson, overdispersed
Poisson, and negative binomial models. Psychological
Bulletin, 118(3), 392–404. doi:10.1037/0033-2909.118.3.392
Gelman, A., & Hill, J. (2006). Data analysis using regression
and multilevel/hierarchical models. Cambridge University
Press.
Gonçalves, J., Pereira, S., & Torres da Silva, M. (2020). How to
report on elections? The effects of game, issue and negative
coverage on reader engagement and incivility. Journalism,
23(6), 1266–1284. doi:10.1177/1464884920958367
Hallin, D., & Mancini, P. (2004). Comparing media systems:
Three models of media and politics. Cambridge University
Press.
Harro-Loit, H., & Saks, K. (2006). The diminishing border
between advertising and journalism in Estonia. Journalism
Studies, 7(2), 312–322. doi:10.1080/14616700500533635
Humayun, M. F., & Ferrucci, P. (2022). Understanding social
media in journalism practice: A typology. Digital
Journalism, 0(0), 1–24. doi:10.1080/21670811.2022.
2086594
Iyengar, S., Norpoth, H., & Hahn, K. S. (2004). Consumer
demand for election news: The horserace sells. The Journal
of Politics, 66(1), 157–175. doi:10.1046/j.1468-2508.2004.
00146.x
Jang, S. M., & Oh, Y. W. (2016). Getting attention online in
election coverage: Audience selectivity in the 2012 US pre-
sidential election. New Media & Society, 18(10), 2271–2286.
doi:10.1177/1461444815583491
Kalsnes, B., & Larsson, A. O. (2018). Understanding news
sharing across social media: Detailing distribution on
JOURNAL OF INFORMATION TECHNOLOGY & POLITICS 51
Facebook and Twitter. Journalism Studies, 19(11),
1669–1688. doi:10.1080/1461670X.2017.1297686
Karidi, M. (2018). News Media Logic on the Move?: In search
of commercial media logic in German news. Journalism
Studies, 19(9), 1237–1256. doi:10.1080/1461670X.2016.
1266281
Klinger, U., & Svensson, J. (2015). The emergence of network
media logic in political communication: A theoretical
approach. New Media & Society, 17(8), 1241–1257. doi:10.
1177/1461444814522952
Kovacevic, P., & Perisin, T. (2018). The potential of construc-
tive journalism ideas in a croatian context. Journalism
Practice, 12(6), 747–763. doi:10.1080/17512786.2018.
1472528
Landerer, N. (2013). Rethinking the Logics: A conceptual
framework for the mediatization of politics.
Communication Theory, 23(3), 239–258. doi:10.1111/
comt.12013
Larsson, A. O. (2015). Comparing to prepare: Suggesting ways
to study social media today—and tomorrow. Social Media
+ Society, 1(1), 2056305115578680. doi:10.1177/
2056305115578680
Lawrence, R. G., Radcliffe, D., & Schmidt, T. R. (2018).
Practicing Engagement: Participatory journalism in the
Web 2.0 era. Journalism Practice, 12(10), 1220–1240.
doi:10.1080/17512786.2017.1391712
Lee, E.-J., & Tandoc, E. C., Jr. (2017). When news meets the
audience: how audience feedback online affects news pro-
duction and consumption. Human Communication
Research, 43(4), 436–449. doi:10.1111/hcre.12123
Lengauer, G., Esser, F., & Berganza, R. (2012). Negativity in
political news: A review of concepts, operationalizations
and key findings. Journalism, 13(2), 179–202. doi:10.1177/
1464884911427800
Metz, M., Kruikemeier, S., & Lecheler, S. (2020).
Personalization of politics on Facebook: Examining the
content and effects of professional, emotional and private
self-personalization. Information, Communication &
Society, 23(10), 1481–1498. doi:10.1080/1369118X.2019.
1581244
Muddiman, A., Pond-Cobb, J., & Matson, J. E. (2020).
Negativity bias or Backlash: Interaction with civil and
uncivil online political news content. Communication
Research, 47(6), 815–837. doi:10.1177/0093650216685625
Muddiman, A., & Stroud, N. J. (2017). News values, cognitive
biases, and partisan incivility in comment sections: Uncivil
comments. Journal of Communication, 67(4), 586–609.
doi:10.1111/jcom.12312
Oeldorf-Hirsch, A., & Sundar, S. S. (2015). Posting, comment-
ing, and tagging: Effects of sharing news stories on
Facebook. Computers in Human Behavior, 44, 240–249.
doi:10.1016/j.chb.2014.11.024
Ørmen, J. (2019). From consumer demand to user engage-
ment: Comparing the popularity and virality of election
coverage on the Internet. The International Journal of
Press/politics, 24(1), 49–68. doi:10.1177/1940161218809160
Örnebring, H. (2012). Clientelism, elites, and the media in
Central and Eastern Europe. The International Journal of
Press/politics, 17(4), 497–515.
Peruško, Z., Čuvalo, A., & Vozab, D. (2020). Mediatization of
journalism: Influence of the media system and media orga-
nization on journalistic practices in European digital
mediascapes. Journalism, 21(11), 1630–1654. doi:10.1177/
1464884917743176
Peruško, Z., Vozab, D., & Čuvalo, A. (2021). Comparing post-
socialist media systems: The case of Southeast Europe.
Routledge & CRC Press. doi:10.4324/9780367226787
Robinson, J. G. (2019). The audience in the mind’s eye: How
journalists imagine their readers. doi:10.7916/d8-drvj-wj06
Roudakova, N. (2008). Media-Political Clientelism: Lessons
from Anthropology’ in Media. Culture, & Society, 30(1),
41–59. doi:10.1177/01634437070843
Salgado, S., & Bobba, G. (2019). News on events and social
media: A comparative analysis of Facebook users’ reactions.
Journalism Studies, 20(15), 2258–2276. doi:10.1080/
1461670X.2019.1586566
Shoemaker, P. J. (1996). Hardwired for News: Using biological
and cultural evolution to explain the surveillance function.
Journal of Communication, 46(3), 32–47. doi:10.1111/j.
1460-2466.1996.tb01487.x
Singer, J. B., Domingo, D., Heinonen, A., Hermida, A.,
Paulussen, S. . . . Vujnovic, M. (2011). Participatory
Journalism: Guarding open gates at online newspapers.
John Wiley & Sons. doi:10.1002/9781444340747
Soroka, S., Fournier, P., & Nir, L. (2019). Cross-national
evidence of a negativity bias in psychophysiological reac-
tions to news. Proceedings of the National Academy of
Sciences. PNAS. https://www.pnas.org/doi/abs/10.1073/
pnas.1908369116
Strömbäck, J., & Dimitrova, D. V. (2006). Political and media
systems matter: A comparison of election news coverage in
Sweden and the United States. Harvard International
Journal of Press/politics, 11(4), 131–147. doi:10.1177/
1081180X06293549
Strömbäck, J., & Esser, F. (2014). Mediatization of Politics:
Towards a theoretical framework. In F. Esser & J. Strömbäck
(Eds.), Mediatization of Politics: Understanding the transfor-
mation of western democracies (pp. 3–28). Palgrave Macmillan
UK. doi:10.1057/9781137275844_1
Stroud, N. J., & Muddiman, A. (2019). Social media engage-
ment with strategy- and issue-framed political news.
Journal of Communication, 69(5), 443–466. doi:10.1093/
joc/jqz029
Swart, J., Groot Kormelink, T., Costera Meijer, I., &
Broersma, M. (2022). Advancing a radical audience turn
in journalism. fundamental dilemmas for journalism
studies. Fundamental Dilemmas for Journalism Studies
Digital Journalism, 10(1), 8–22. doi:10.1080/21670811.
2021.2024764
Tóth, F., Mihelj, S., Štětka, V., & Kondor, K. (2022). A media
repertoires approach to selective exposure: News consump-
tion and political polarization in Eastern Europe. The
52 L. CAMAJ ET AL.
International Journal of Press/politics, 1–25. doi:10.1177/
19401612211072552
Trilling, D., Tolochko, P., & Burscher, B. (2017). From news-
worthiness to shareworthiness: How to predict news sharing
based on article characteristics. Journalism & Mass
Communication Quarterly, 94(1), 38–60. doi:10.1177/
1077699016654682
Trussler, M., & Soroka, S. (2014). Consumer demand for cynical
and negative news frames. The International Journal of Press/
politics, 19(3), 360–379. doi:10.1177/1940161214524832
Uberti, L. J. (2020). Bribes, rents and industrial firm perfor-
mance in Albania and Kosovo. Comparative Economic
Studies, 62(2), 263–302. doi:10.1057/s41294-020-00112-5
Valenzuela, S., Piña, M., & Ramírez, J. (2017). Behavioral
effects of framing on social media users: How conflict,
economic, human interest, and morality frames drive
news sharing. Journal of Communication, 67(5), 803–826.
doi:10.1111/jcom.12325
Van Aelst, P., Sheafer, T., & Stanyer, J. (2012). The persona-
lization of mediated political communication: A review of
concepts, operationalizations and key findings. Journalism,
13(2), 203–220. doi:10.1177/1464884911427802
Zoizner, A. (2021). The consequences of strategic news
coverage for democracy: A meta-analysis.
Communication Research, 48(1), 3–25. doi:10.1177/
0093650218808691
JOURNAL OF INFORMATION TECHNOLOGY & POLITICS 53